1 \input texinfo @c -*-texinfo-*-
2 @c Copyright (C) 1988, 1989, 1990, 1991, 1992, 1993, 1994, 1995, 1996, 1998,
3 @c 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009
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
26 @c readline appendices use @vindex, @findex and @ftable,
27 @c annotate.texi and gdbmi use @findex.
31 @c !!set GDB manual's edition---not the same as GDB version!
32 @c This is updated by GNU Press.
35 @c !!set GDB edit command default editor
38 @c THIS MANUAL REQUIRES TEXINFO 4.0 OR LATER.
40 @c This is a dir.info fragment to support semi-automated addition of
41 @c manuals to an info tree.
42 @dircategory Software development
44 * Gdb: (gdb). The GNU debugger.
48 Copyright @copyright{} 1988, 1989, 1990, 1991, 1992, 1993, 1994, 1995, 1996,
49 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009
50 Free Software Foundation, Inc.
52 Permission is granted to copy, distribute and/or modify this document
53 under the terms of the GNU Free Documentation License, Version 1.1 or
54 any later version published by the Free Software Foundation; with the
55 Invariant Sections being ``Free Software'' and ``Free Software Needs
56 Free Documentation'', with the Front-Cover Texts being ``A GNU Manual,''
57 and with the Back-Cover Texts as in (a) below.
59 (a) The FSF's Back-Cover Text is: ``You are free to copy and modify
60 this GNU Manual. Buying copies from GNU Press supports the FSF in
61 developing GNU and promoting software freedom.''
65 This file documents the @sc{gnu} debugger @value{GDBN}.
67 This is the @value{EDITION} Edition, of @cite{Debugging with
68 @value{GDBN}: the @sc{gnu} Source-Level Debugger} for @value{GDBN}
69 @ifset VERSION_PACKAGE
70 @value{VERSION_PACKAGE}
72 Version @value{GDBVN}.
78 @title Debugging with @value{GDBN}
79 @subtitle The @sc{gnu} Source-Level Debugger
81 @subtitle @value{EDITION} Edition, for @value{GDBN} version @value{GDBVN}
82 @ifset VERSION_PACKAGE
84 @subtitle @value{VERSION_PACKAGE}
86 @author Richard Stallman, Roland Pesch, Stan Shebs, et al.
90 \hfill (Send bugs and comments on @value{GDBN} to @value{BUGURL}.)\par
91 \hfill {\it Debugging with @value{GDBN}}\par
92 \hfill \TeX{}info \texinfoversion\par
96 @vskip 0pt plus 1filll
97 Published by the Free Software Foundation @*
98 51 Franklin Street, Fifth Floor,
99 Boston, MA 02110-1301, USA@*
100 ISBN 1-882114-77-9 @*
104 This edition of the GDB manual is dedicated to the memory of Fred
105 Fish. Fred was a long-standing contributor to GDB and to Free
106 software in general. We will miss him.
111 @node Top, Summary, (dir), (dir)
113 @top Debugging with @value{GDBN}
115 This file describes @value{GDBN}, the @sc{gnu} symbolic debugger.
117 This is the @value{EDITION} Edition, for @value{GDBN}
118 @ifset VERSION_PACKAGE
119 @value{VERSION_PACKAGE}
121 Version @value{GDBVN}.
123 Copyright (C) 1988-2009 Free Software Foundation, Inc.
125 This edition of the GDB manual is dedicated to the memory of Fred
126 Fish. Fred was a long-standing contributor to GDB and to Free
127 software in general. We will miss him.
130 * Summary:: Summary of @value{GDBN}
131 * Sample Session:: A sample @value{GDBN} session
133 * Invocation:: Getting in and out of @value{GDBN}
134 * Commands:: @value{GDBN} commands
135 * Running:: Running programs under @value{GDBN}
136 * Stopping:: Stopping and continuing
137 * Reverse Execution:: Running programs backward
138 * Process Record and Replay:: Recording inferior's execution and replaying it
139 * Stack:: Examining the stack
140 * Source:: Examining source files
141 * Data:: Examining data
142 * Macros:: Preprocessor Macros
143 * Tracepoints:: Debugging remote targets non-intrusively
144 * Overlays:: Debugging programs that use overlays
146 * Languages:: Using @value{GDBN} with different languages
148 * Symbols:: Examining the symbol table
149 * Altering:: Altering execution
150 * GDB Files:: @value{GDBN} files
151 * Targets:: Specifying a debugging target
152 * Remote Debugging:: Debugging remote programs
153 * Configurations:: Configuration-specific information
154 * Controlling GDB:: Controlling @value{GDBN}
155 * Extending GDB:: Extending @value{GDBN}
156 * Interpreters:: Command Interpreters
157 * TUI:: @value{GDBN} Text User Interface
158 * Emacs:: Using @value{GDBN} under @sc{gnu} Emacs
159 * GDB/MI:: @value{GDBN}'s Machine Interface.
160 * Annotations:: @value{GDBN}'s annotation interface.
162 * GDB Bugs:: Reporting bugs in @value{GDBN}
164 * Command Line Editing:: Command Line Editing
165 * Using History Interactively:: Using History Interactively
166 * Formatting Documentation:: How to format and print @value{GDBN} documentation
167 * Installing GDB:: Installing GDB
168 * Maintenance Commands:: Maintenance Commands
169 * Remote Protocol:: GDB Remote Serial Protocol
170 * Agent Expressions:: The GDB Agent Expression Mechanism
171 * Target Descriptions:: How targets can describe themselves to
173 * Operating System Information:: Getting additional information from
175 * Copying:: GNU General Public License says
176 how you can copy and share GDB
177 * GNU Free Documentation License:: The license for this documentation
186 @unnumbered Summary of @value{GDBN}
188 The purpose of a debugger such as @value{GDBN} is to allow you to see what is
189 going on ``inside'' another program while it executes---or what another
190 program was doing at the moment it crashed.
192 @value{GDBN} can do four main kinds of things (plus other things in support of
193 these) to help you catch bugs in the act:
197 Start your program, specifying anything that might affect its behavior.
200 Make your program stop on specified conditions.
203 Examine what has happened, when your program has stopped.
206 Change things in your program, so you can experiment with correcting the
207 effects of one bug and go on to learn about another.
210 You can use @value{GDBN} to debug programs written in C and C@t{++}.
211 For more information, see @ref{Supported Languages,,Supported Languages}.
212 For more information, see @ref{C,,C and C++}.
215 Support for Modula-2 is partial. For information on Modula-2, see
216 @ref{Modula-2,,Modula-2}.
219 Debugging Pascal programs which use sets, subranges, file variables, or
220 nested functions does not currently work. @value{GDBN} does not support
221 entering expressions, printing values, or similar features using Pascal
225 @value{GDBN} can be used to debug programs written in Fortran, although
226 it may be necessary to refer to some variables with a trailing
229 @value{GDBN} can be used to debug programs written in Objective-C,
230 using either the Apple/NeXT or the GNU Objective-C runtime.
233 * Free Software:: Freely redistributable software
234 * Contributors:: Contributors to GDB
238 @unnumberedsec Free Software
240 @value{GDBN} is @dfn{free software}, protected by the @sc{gnu}
241 General Public License
242 (GPL). The GPL gives you the freedom to copy or adapt a licensed
243 program---but every person getting a copy also gets with it the
244 freedom to modify that copy (which means that they must get access to
245 the source code), and the freedom to distribute further copies.
246 Typical software companies use copyrights to limit your freedoms; the
247 Free Software Foundation uses the GPL to preserve these freedoms.
249 Fundamentally, the General Public License is a license which says that
250 you have these freedoms and that you cannot take these freedoms away
253 @unnumberedsec Free Software Needs Free Documentation
255 The biggest deficiency in the free software community today is not in
256 the software---it is the lack of good free documentation that we can
257 include with the free software. Many of our most important
258 programs do not come with free reference manuals and free introductory
259 texts. Documentation is an essential part of any software package;
260 when an important free software package does not come with a free
261 manual and a free tutorial, that is a major gap. We have many such
264 Consider Perl, for instance. The tutorial manuals that people
265 normally use are non-free. How did this come about? Because the
266 authors of those manuals published them with restrictive terms---no
267 copying, no modification, source files not available---which exclude
268 them from the free software world.
270 That wasn't the first time this sort of thing happened, and it was far
271 from the last. Many times we have heard a GNU user eagerly describe a
272 manual that he is writing, his intended contribution to the community,
273 only to learn that he had ruined everything by signing a publication
274 contract to make it non-free.
276 Free documentation, like free software, is a matter of freedom, not
277 price. The problem with the non-free manual is not that publishers
278 charge a price for printed copies---that in itself is fine. (The Free
279 Software Foundation sells printed copies of manuals, too.) The
280 problem is the restrictions on the use of the manual. Free manuals
281 are available in source code form, and give you permission to copy and
282 modify. Non-free manuals do not allow this.
284 The criteria of freedom for a free manual are roughly the same as for
285 free software. Redistribution (including the normal kinds of
286 commercial redistribution) must be permitted, so that the manual can
287 accompany every copy of the program, both on-line and on paper.
289 Permission for modification of the technical content is crucial too.
290 When people modify the software, adding or changing features, if they
291 are conscientious they will change the manual too---so they can
292 provide accurate and clear documentation for the modified program. A
293 manual that leaves you no choice but to write a new manual to document
294 a changed version of the program is not really available to our
297 Some kinds of limits on the way modification is handled are
298 acceptable. For example, requirements to preserve the original
299 author's copyright notice, the distribution terms, or the list of
300 authors, are ok. It is also no problem to require modified versions
301 to include notice that they were modified. Even entire sections that
302 may not be deleted or changed are acceptable, as long as they deal
303 with nontechnical topics (like this one). These kinds of restrictions
304 are acceptable because they don't obstruct the community's normal use
307 However, it must be possible to modify all the @emph{technical}
308 content of the manual, and then distribute the result in all the usual
309 media, through all the usual channels. Otherwise, the restrictions
310 obstruct the use of the manual, it is not free, and we need another
311 manual to replace it.
313 Please spread the word about this issue. Our community continues to
314 lose manuals to proprietary publishing. If we spread the word that
315 free software needs free reference manuals and free tutorials, perhaps
316 the next person who wants to contribute by writing documentation will
317 realize, before it is too late, that only free manuals contribute to
318 the free software community.
320 If you are writing documentation, please insist on publishing it under
321 the GNU Free Documentation License or another free documentation
322 license. Remember that this decision requires your approval---you
323 don't have to let the publisher decide. Some commercial publishers
324 will use a free license if you insist, but they will not propose the
325 option; it is up to you to raise the issue and say firmly that this is
326 what you want. If the publisher you are dealing with refuses, please
327 try other publishers. If you're not sure whether a proposed license
328 is free, write to @email{licensing@@gnu.org}.
330 You can encourage commercial publishers to sell more free, copylefted
331 manuals and tutorials by buying them, and particularly by buying
332 copies from the publishers that paid for their writing or for major
333 improvements. Meanwhile, try to avoid buying non-free documentation
334 at all. Check the distribution terms of a manual before you buy it,
335 and insist that whoever seeks your business must respect your freedom.
336 Check the history of the book, and try to reward the publishers that
337 have paid or pay the authors to work on it.
339 The Free Software Foundation maintains a list of free documentation
340 published by other publishers, at
341 @url{http://www.fsf.org/doc/other-free-books.html}.
344 @unnumberedsec Contributors to @value{GDBN}
346 Richard Stallman was the original author of @value{GDBN}, and of many
347 other @sc{gnu} programs. Many others have contributed to its
348 development. This section attempts to credit major contributors. One
349 of the virtues of free software is that everyone is free to contribute
350 to it; with regret, we cannot actually acknowledge everyone here. The
351 file @file{ChangeLog} in the @value{GDBN} distribution approximates a
352 blow-by-blow account.
354 Changes much prior to version 2.0 are lost in the mists of time.
357 @emph{Plea:} Additions to this section are particularly welcome. If you
358 or your friends (or enemies, to be evenhanded) have been unfairly
359 omitted from this list, we would like to add your names!
362 So that they may not regard their many labors as thankless, we
363 particularly thank those who shepherded @value{GDBN} through major
365 Andrew Cagney (releases 6.3, 6.2, 6.1, 6.0, 5.3, 5.2, 5.1 and 5.0);
366 Jim Blandy (release 4.18);
367 Jason Molenda (release 4.17);
368 Stan Shebs (release 4.14);
369 Fred Fish (releases 4.16, 4.15, 4.13, 4.12, 4.11, 4.10, and 4.9);
370 Stu Grossman and John Gilmore (releases 4.8, 4.7, 4.6, 4.5, and 4.4);
371 John Gilmore (releases 4.3, 4.2, 4.1, 4.0, and 3.9);
372 Jim Kingdon (releases 3.5, 3.4, and 3.3);
373 and Randy Smith (releases 3.2, 3.1, and 3.0).
375 Richard Stallman, assisted at various times by Peter TerMaat, Chris
376 Hanson, and Richard Mlynarik, handled releases through 2.8.
378 Michael Tiemann is the author of most of the @sc{gnu} C@t{++} support
379 in @value{GDBN}, with significant additional contributions from Per
380 Bothner and Daniel Berlin. James Clark wrote the @sc{gnu} C@t{++}
381 demangler. Early work on C@t{++} was by Peter TerMaat (who also did
382 much general update work leading to release 3.0).
384 @value{GDBN} uses the BFD subroutine library to examine multiple
385 object-file formats; BFD was a joint project of David V.
386 Henkel-Wallace, Rich Pixley, Steve Chamberlain, and John Gilmore.
388 David Johnson wrote the original COFF support; Pace Willison did
389 the original support for encapsulated COFF.
391 Brent Benson of Harris Computer Systems contributed DWARF 2 support.
393 Adam de Boor and Bradley Davis contributed the ISI Optimum V support.
394 Per Bothner, Noboyuki Hikichi, and Alessandro Forin contributed MIPS
396 Jean-Daniel Fekete contributed Sun 386i support.
397 Chris Hanson improved the HP9000 support.
398 Noboyuki Hikichi and Tomoyuki Hasei contributed Sony/News OS 3 support.
399 David Johnson contributed Encore Umax support.
400 Jyrki Kuoppala contributed Altos 3068 support.
401 Jeff Law contributed HP PA and SOM support.
402 Keith Packard contributed NS32K support.
403 Doug Rabson contributed Acorn Risc Machine support.
404 Bob Rusk contributed Harris Nighthawk CX-UX support.
405 Chris Smith contributed Convex support (and Fortran debugging).
406 Jonathan Stone contributed Pyramid support.
407 Michael Tiemann contributed SPARC support.
408 Tim Tucker contributed support for the Gould NP1 and Gould Powernode.
409 Pace Willison contributed Intel 386 support.
410 Jay Vosburgh contributed Symmetry support.
411 Marko Mlinar contributed OpenRISC 1000 support.
413 Andreas Schwab contributed M68K @sc{gnu}/Linux support.
415 Rich Schaefer and Peter Schauer helped with support of SunOS shared
418 Jay Fenlason and Roland McGrath ensured that @value{GDBN} and GAS agree
419 about several machine instruction sets.
421 Patrick Duval, Ted Goldstein, Vikram Koka and Glenn Engel helped develop
422 remote debugging. Intel Corporation, Wind River Systems, AMD, and ARM
423 contributed remote debugging modules for the i960, VxWorks, A29K UDI,
424 and RDI targets, respectively.
426 Brian Fox is the author of the readline libraries providing
427 command-line editing and command history.
429 Andrew Beers of SUNY Buffalo wrote the language-switching code, the
430 Modula-2 support, and contributed the Languages chapter of this manual.
432 Fred Fish wrote most of the support for Unix System Vr4.
433 He also enhanced the command-completion support to cover C@t{++} overloaded
436 Hitachi America (now Renesas America), Ltd. sponsored the support for
437 H8/300, H8/500, and Super-H processors.
439 NEC sponsored the support for the v850, Vr4xxx, and Vr5xxx processors.
441 Mitsubishi (now Renesas) sponsored the support for D10V, D30V, and M32R/D
444 Toshiba sponsored the support for the TX39 Mips processor.
446 Matsushita sponsored the support for the MN10200 and MN10300 processors.
448 Fujitsu sponsored the support for SPARClite and FR30 processors.
450 Kung Hsu, Jeff Law, and Rick Sladkey added support for hardware
453 Michael Snyder added support for tracepoints.
455 Stu Grossman wrote gdbserver.
457 Jim Kingdon, Peter Schauer, Ian Taylor, and Stu Grossman made
458 nearly innumerable bug fixes and cleanups throughout @value{GDBN}.
460 The following people at the Hewlett-Packard Company contributed
461 support for the PA-RISC 2.0 architecture, HP-UX 10.20, 10.30, and 11.0
462 (narrow mode), HP's implementation of kernel threads, HP's aC@t{++}
463 compiler, and the Text User Interface (nee Terminal User Interface):
464 Ben Krepp, Richard Title, John Bishop, Susan Macchia, Kathy Mann,
465 Satish Pai, India Paul, Steve Rehrauer, and Elena Zannoni. Kim Haase
466 provided HP-specific information in this manual.
468 DJ Delorie ported @value{GDBN} to MS-DOS, for the DJGPP project.
469 Robert Hoehne made significant contributions to the DJGPP port.
471 Cygnus Solutions has sponsored @value{GDBN} maintenance and much of its
472 development since 1991. Cygnus engineers who have worked on @value{GDBN}
473 fulltime include Mark Alexander, Jim Blandy, Per Bothner, Kevin
474 Buettner, Edith Epstein, Chris Faylor, Fred Fish, Martin Hunt, Jim
475 Ingham, John Gilmore, Stu Grossman, Kung Hsu, Jim Kingdon, John Metzler,
476 Fernando Nasser, Geoffrey Noer, Dawn Perchik, Rich Pixley, Zdenek
477 Radouch, Keith Seitz, Stan Shebs, David Taylor, and Elena Zannoni. In
478 addition, Dave Brolley, Ian Carmichael, Steve Chamberlain, Nick Clifton,
479 JT Conklin, Stan Cox, DJ Delorie, Ulrich Drepper, Frank Eigler, Doug
480 Evans, Sean Fagan, David Henkel-Wallace, Richard Henderson, Jeff
481 Holcomb, Jeff Law, Jim Lemke, Tom Lord, Bob Manson, Michael Meissner,
482 Jason Merrill, Catherine Moore, Drew Moseley, Ken Raeburn, Gavin
483 Romig-Koch, Rob Savoye, Jamie Smith, Mike Stump, Ian Taylor, Angela
484 Thomas, Michael Tiemann, Tom Tromey, Ron Unrau, Jim Wilson, and David
485 Zuhn have made contributions both large and small.
487 Andrew Cagney, Fernando Nasser, and Elena Zannoni, while working for
488 Cygnus Solutions, implemented the original @sc{gdb/mi} interface.
490 Jim Blandy added support for preprocessor macros, while working for Red
493 Andrew Cagney designed @value{GDBN}'s architecture vector. Many
494 people including Andrew Cagney, Stephane Carrez, Randolph Chung, Nick
495 Duffek, Richard Henderson, Mark Kettenis, Grace Sainsbury, Kei
496 Sakamoto, Yoshinori Sato, Michael Snyder, Andreas Schwab, Jason
497 Thorpe, Corinna Vinschen, Ulrich Weigand, and Elena Zannoni, helped
498 with the migration of old architectures to this new framework.
500 Andrew Cagney completely re-designed and re-implemented @value{GDBN}'s
501 unwinder framework, this consisting of a fresh new design featuring
502 frame IDs, independent frame sniffers, and the sentinel frame. Mark
503 Kettenis implemented the @sc{dwarf 2} unwinder, Jeff Johnston the
504 libunwind unwinder, and Andrew Cagney the dummy, sentinel, tramp, and
505 trad unwinders. The architecture-specific changes, each involving a
506 complete rewrite of the architecture's frame code, were carried out by
507 Jim Blandy, Joel Brobecker, Kevin Buettner, Andrew Cagney, Stephane
508 Carrez, Randolph Chung, Orjan Friberg, Richard Henderson, Daniel
509 Jacobowitz, Jeff Johnston, Mark Kettenis, Theodore A. Roth, Kei
510 Sakamoto, Yoshinori Sato, Michael Snyder, Corinna Vinschen, and Ulrich
513 Christian Zankel, Ross Morley, Bob Wilson, and Maxim Grigoriev from
514 Tensilica, Inc.@: contributed support for Xtensa processors. Others
515 who have worked on the Xtensa port of @value{GDBN} in the past include
516 Steve Tjiang, John Newlin, and Scott Foehner.
519 @chapter A Sample @value{GDBN} Session
521 You can use this manual at your leisure to read all about @value{GDBN}.
522 However, a handful of commands are enough to get started using the
523 debugger. This chapter illustrates those commands.
526 In this sample session, we emphasize user input like this: @b{input},
527 to make it easier to pick out from the surrounding output.
530 @c FIXME: this example may not be appropriate for some configs, where
531 @c FIXME...primary interest is in remote use.
533 One of the preliminary versions of @sc{gnu} @code{m4} (a generic macro
534 processor) exhibits the following bug: sometimes, when we change its
535 quote strings from the default, the commands used to capture one macro
536 definition within another stop working. In the following short @code{m4}
537 session, we define a macro @code{foo} which expands to @code{0000}; we
538 then use the @code{m4} built-in @code{defn} to define @code{bar} as the
539 same thing. However, when we change the open quote string to
540 @code{<QUOTE>} and the close quote string to @code{<UNQUOTE>}, the same
541 procedure fails to define a new synonym @code{baz}:
550 @b{define(bar,defn(`foo'))}
554 @b{changequote(<QUOTE>,<UNQUOTE>)}
556 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
559 m4: End of input: 0: fatal error: EOF in string
563 Let us use @value{GDBN} to try to see what is going on.
566 $ @b{@value{GDBP} m4}
567 @c FIXME: this falsifies the exact text played out, to permit smallbook
568 @c FIXME... format to come out better.
569 @value{GDBN} is free software and you are welcome to distribute copies
570 of it under certain conditions; type "show copying" to see
572 There is absolutely no warranty for @value{GDBN}; type "show warranty"
575 @value{GDBN} @value{GDBVN}, Copyright 1999 Free Software Foundation, Inc...
580 @value{GDBN} reads only enough symbol data to know where to find the
581 rest when needed; as a result, the first prompt comes up very quickly.
582 We now tell @value{GDBN} to use a narrower display width than usual, so
583 that examples fit in this manual.
586 (@value{GDBP}) @b{set width 70}
590 We need to see how the @code{m4} built-in @code{changequote} works.
591 Having looked at the source, we know the relevant subroutine is
592 @code{m4_changequote}, so we set a breakpoint there with the @value{GDBN}
593 @code{break} command.
596 (@value{GDBP}) @b{break m4_changequote}
597 Breakpoint 1 at 0x62f4: file builtin.c, line 879.
601 Using the @code{run} command, we start @code{m4} running under @value{GDBN}
602 control; as long as control does not reach the @code{m4_changequote}
603 subroutine, the program runs as usual:
606 (@value{GDBP}) @b{run}
607 Starting program: /work/Editorial/gdb/gnu/m4/m4
615 To trigger the breakpoint, we call @code{changequote}. @value{GDBN}
616 suspends execution of @code{m4}, displaying information about the
617 context where it stops.
620 @b{changequote(<QUOTE>,<UNQUOTE>)}
622 Breakpoint 1, m4_changequote (argc=3, argv=0x33c70)
624 879 if (bad_argc(TOKEN_DATA_TEXT(argv[0]),argc,1,3))
628 Now we use the command @code{n} (@code{next}) to advance execution to
629 the next line of the current function.
633 882 set_quotes((argc >= 2) ? TOKEN_DATA_TEXT(argv[1])\
638 @code{set_quotes} looks like a promising subroutine. We can go into it
639 by using the command @code{s} (@code{step}) instead of @code{next}.
640 @code{step} goes to the next line to be executed in @emph{any}
641 subroutine, so it steps into @code{set_quotes}.
645 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
647 530 if (lquote != def_lquote)
651 The display that shows the subroutine where @code{m4} is now
652 suspended (and its arguments) is called a stack frame display. It
653 shows a summary of the stack. We can use the @code{backtrace}
654 command (which can also be spelled @code{bt}), to see where we are
655 in the stack as a whole: the @code{backtrace} command displays a
656 stack frame for each active subroutine.
659 (@value{GDBP}) @b{bt}
660 #0 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
662 #1 0x6344 in m4_changequote (argc=3, argv=0x33c70)
664 #2 0x8174 in expand_macro (sym=0x33320) at macro.c:242
665 #3 0x7a88 in expand_token (obs=0x0, t=209696, td=0xf7fffa30)
667 #4 0x79dc in expand_input () at macro.c:40
668 #5 0x2930 in main (argc=0, argv=0xf7fffb20) at m4.c:195
672 We step through a few more lines to see what happens. The first two
673 times, we can use @samp{s}; the next two times we use @code{n} to avoid
674 falling into the @code{xstrdup} subroutine.
678 0x3b5c 532 if (rquote != def_rquote)
680 0x3b80 535 lquote = (lq == nil || *lq == '\0') ? \
681 def_lquote : xstrdup(lq);
683 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
686 538 len_lquote = strlen(rquote);
690 The last line displayed looks a little odd; we can examine the variables
691 @code{lquote} and @code{rquote} to see if they are in fact the new left
692 and right quotes we specified. We use the command @code{p}
693 (@code{print}) to see their values.
696 (@value{GDBP}) @b{p lquote}
697 $1 = 0x35d40 "<QUOTE>"
698 (@value{GDBP}) @b{p rquote}
699 $2 = 0x35d50 "<UNQUOTE>"
703 @code{lquote} and @code{rquote} are indeed the new left and right quotes.
704 To look at some context, we can display ten lines of source
705 surrounding the current line with the @code{l} (@code{list}) command.
711 535 lquote = (lq == nil || *lq == '\0') ? def_lquote\
713 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
716 538 len_lquote = strlen(rquote);
717 539 len_rquote = strlen(lquote);
724 Let us step past the two lines that set @code{len_lquote} and
725 @code{len_rquote}, and then examine the values of those variables.
729 539 len_rquote = strlen(lquote);
732 (@value{GDBP}) @b{p len_lquote}
734 (@value{GDBP}) @b{p len_rquote}
739 That certainly looks wrong, assuming @code{len_lquote} and
740 @code{len_rquote} are meant to be the lengths of @code{lquote} and
741 @code{rquote} respectively. We can set them to better values using
742 the @code{p} command, since it can print the value of
743 any expression---and that expression can include subroutine calls and
747 (@value{GDBP}) @b{p len_lquote=strlen(lquote)}
749 (@value{GDBP}) @b{p len_rquote=strlen(rquote)}
754 Is that enough to fix the problem of using the new quotes with the
755 @code{m4} built-in @code{defn}? We can allow @code{m4} to continue
756 executing with the @code{c} (@code{continue}) command, and then try the
757 example that caused trouble initially:
763 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
770 Success! The new quotes now work just as well as the default ones. The
771 problem seems to have been just the two typos defining the wrong
772 lengths. We allow @code{m4} exit by giving it an EOF as input:
776 Program exited normally.
780 The message @samp{Program exited normally.} is from @value{GDBN}; it
781 indicates @code{m4} has finished executing. We can end our @value{GDBN}
782 session with the @value{GDBN} @code{quit} command.
785 (@value{GDBP}) @b{quit}
789 @chapter Getting In and Out of @value{GDBN}
791 This chapter discusses how to start @value{GDBN}, and how to get out of it.
795 type @samp{@value{GDBP}} to start @value{GDBN}.
797 type @kbd{quit} or @kbd{Ctrl-d} to exit.
801 * Invoking GDB:: How to start @value{GDBN}
802 * Quitting GDB:: How to quit @value{GDBN}
803 * Shell Commands:: How to use shell commands inside @value{GDBN}
804 * Logging Output:: How to log @value{GDBN}'s output to a file
808 @section Invoking @value{GDBN}
810 Invoke @value{GDBN} by running the program @code{@value{GDBP}}. Once started,
811 @value{GDBN} reads commands from the terminal until you tell it to exit.
813 You can also run @code{@value{GDBP}} with a variety of arguments and options,
814 to specify more of your debugging environment at the outset.
816 The command-line options described here are designed
817 to cover a variety of situations; in some environments, some of these
818 options may effectively be unavailable.
820 The most usual way to start @value{GDBN} is with one argument,
821 specifying an executable program:
824 @value{GDBP} @var{program}
828 You can also start with both an executable program and a core file
832 @value{GDBP} @var{program} @var{core}
835 You can, instead, specify a process ID as a second argument, if you want
836 to debug a running process:
839 @value{GDBP} @var{program} 1234
843 would attach @value{GDBN} to process @code{1234} (unless you also have a file
844 named @file{1234}; @value{GDBN} does check for a core file first).
846 Taking advantage of the second command-line argument requires a fairly
847 complete operating system; when you use @value{GDBN} as a remote
848 debugger attached to a bare board, there may not be any notion of
849 ``process'', and there is often no way to get a core dump. @value{GDBN}
850 will warn you if it is unable to attach or to read core dumps.
852 You can optionally have @code{@value{GDBP}} pass any arguments after the
853 executable file to the inferior using @code{--args}. This option stops
856 @value{GDBP} --args gcc -O2 -c foo.c
858 This will cause @code{@value{GDBP}} to debug @code{gcc}, and to set
859 @code{gcc}'s command-line arguments (@pxref{Arguments}) to @samp{-O2 -c foo.c}.
861 You can run @code{@value{GDBP}} without printing the front material, which describes
862 @value{GDBN}'s non-warranty, by specifying @code{-silent}:
869 You can further control how @value{GDBN} starts up by using command-line
870 options. @value{GDBN} itself can remind you of the options available.
880 to display all available options and briefly describe their use
881 (@samp{@value{GDBP} -h} is a shorter equivalent).
883 All options and command line arguments you give are processed
884 in sequential order. The order makes a difference when the
885 @samp{-x} option is used.
889 * File Options:: Choosing files
890 * Mode Options:: Choosing modes
891 * Startup:: What @value{GDBN} does during startup
895 @subsection Choosing Files
897 When @value{GDBN} starts, it reads any arguments other than options as
898 specifying an executable file and core file (or process ID). This is
899 the same as if the arguments were specified by the @samp{-se} and
900 @samp{-c} (or @samp{-p}) options respectively. (@value{GDBN} reads the
901 first argument that does not have an associated option flag as
902 equivalent to the @samp{-se} option followed by that argument; and the
903 second argument that does not have an associated option flag, if any, as
904 equivalent to the @samp{-c}/@samp{-p} option followed by that argument.)
905 If the second argument begins with a decimal digit, @value{GDBN} will
906 first attempt to attach to it as a process, and if that fails, attempt
907 to open it as a corefile. If you have a corefile whose name begins with
908 a digit, you can prevent @value{GDBN} from treating it as a pid by
909 prefixing it with @file{./}, e.g.@: @file{./12345}.
911 If @value{GDBN} has not been configured to included core file support,
912 such as for most embedded targets, then it will complain about a second
913 argument and ignore it.
915 Many options have both long and short forms; both are shown in the
916 following list. @value{GDBN} also recognizes the long forms if you truncate
917 them, so long as enough of the option is present to be unambiguous.
918 (If you prefer, you can flag option arguments with @samp{--} rather
919 than @samp{-}, though we illustrate the more usual convention.)
921 @c NOTE: the @cindex entries here use double dashes ON PURPOSE. This
922 @c way, both those who look for -foo and --foo in the index, will find
926 @item -symbols @var{file}
928 @cindex @code{--symbols}
930 Read symbol table from file @var{file}.
932 @item -exec @var{file}
934 @cindex @code{--exec}
936 Use file @var{file} as the executable file to execute when appropriate,
937 and for examining pure data in conjunction with a core dump.
941 Read symbol table from file @var{file} and use it as the executable
944 @item -core @var{file}
946 @cindex @code{--core}
948 Use file @var{file} as a core dump to examine.
950 @item -pid @var{number}
951 @itemx -p @var{number}
954 Connect to process ID @var{number}, as with the @code{attach} command.
956 @item -command @var{file}
958 @cindex @code{--command}
960 Execute @value{GDBN} commands from file @var{file}. @xref{Command
961 Files,, Command files}.
963 @item -eval-command @var{command}
964 @itemx -ex @var{command}
965 @cindex @code{--eval-command}
967 Execute a single @value{GDBN} command.
969 This option may be used multiple times to call multiple commands. It may
970 also be interleaved with @samp{-command} as required.
973 @value{GDBP} -ex 'target sim' -ex 'load' \
974 -x setbreakpoints -ex 'run' a.out
977 @item -directory @var{directory}
978 @itemx -d @var{directory}
979 @cindex @code{--directory}
981 Add @var{directory} to the path to search for source and script files.
985 @cindex @code{--readnow}
987 Read each symbol file's entire symbol table immediately, rather than
988 the default, which is to read it incrementally as it is needed.
989 This makes startup slower, but makes future operations faster.
994 @subsection Choosing Modes
996 You can run @value{GDBN} in various alternative modes---for example, in
997 batch mode or quiet mode.
1004 Do not execute commands found in any initialization files. Normally,
1005 @value{GDBN} executes the commands in these files after all the command
1006 options and arguments have been processed. @xref{Command Files,,Command
1012 @cindex @code{--quiet}
1013 @cindex @code{--silent}
1015 ``Quiet''. Do not print the introductory and copyright messages. These
1016 messages are also suppressed in batch mode.
1019 @cindex @code{--batch}
1020 Run in batch mode. Exit with status @code{0} after processing all the
1021 command files specified with @samp{-x} (and all commands from
1022 initialization files, if not inhibited with @samp{-n}). Exit with
1023 nonzero status if an error occurs in executing the @value{GDBN} commands
1024 in the command files.
1026 Batch mode may be useful for running @value{GDBN} as a filter, for
1027 example to download and run a program on another computer; in order to
1028 make this more useful, the message
1031 Program exited normally.
1035 (which is ordinarily issued whenever a program running under
1036 @value{GDBN} control terminates) is not issued when running in batch
1040 @cindex @code{--batch-silent}
1041 Run in batch mode exactly like @samp{-batch}, but totally silently. All
1042 @value{GDBN} output to @code{stdout} is prevented (@code{stderr} is
1043 unaffected). This is much quieter than @samp{-silent} and would be useless
1044 for an interactive session.
1046 This is particularly useful when using targets that give @samp{Loading section}
1047 messages, for example.
1049 Note that targets that give their output via @value{GDBN}, as opposed to
1050 writing directly to @code{stdout}, will also be made silent.
1052 @item -return-child-result
1053 @cindex @code{--return-child-result}
1054 The return code from @value{GDBN} will be the return code from the child
1055 process (the process being debugged), with the following exceptions:
1059 @value{GDBN} exits abnormally. E.g., due to an incorrect argument or an
1060 internal error. In this case the exit code is the same as it would have been
1061 without @samp{-return-child-result}.
1063 The user quits with an explicit value. E.g., @samp{quit 1}.
1065 The child process never runs, or is not allowed to terminate, in which case
1066 the exit code will be -1.
1069 This option is useful in conjunction with @samp{-batch} or @samp{-batch-silent},
1070 when @value{GDBN} is being used as a remote program loader or simulator
1075 @cindex @code{--nowindows}
1077 ``No windows''. If @value{GDBN} comes with a graphical user interface
1078 (GUI) built in, then this option tells @value{GDBN} to only use the command-line
1079 interface. If no GUI is available, this option has no effect.
1083 @cindex @code{--windows}
1085 If @value{GDBN} includes a GUI, then this option requires it to be
1088 @item -cd @var{directory}
1090 Run @value{GDBN} using @var{directory} as its working directory,
1091 instead of the current directory.
1095 @cindex @code{--fullname}
1097 @sc{gnu} Emacs sets this option when it runs @value{GDBN} as a
1098 subprocess. It tells @value{GDBN} to output the full file name and line
1099 number in a standard, recognizable fashion each time a stack frame is
1100 displayed (which includes each time your program stops). This
1101 recognizable format looks like two @samp{\032} characters, followed by
1102 the file name, line number and character position separated by colons,
1103 and a newline. The Emacs-to-@value{GDBN} interface program uses the two
1104 @samp{\032} characters as a signal to display the source code for the
1108 @cindex @code{--epoch}
1109 The Epoch Emacs-@value{GDBN} interface sets this option when it runs
1110 @value{GDBN} as a subprocess. It tells @value{GDBN} to modify its print
1111 routines so as to allow Epoch to display values of expressions in a
1114 @item -annotate @var{level}
1115 @cindex @code{--annotate}
1116 This option sets the @dfn{annotation level} inside @value{GDBN}. Its
1117 effect is identical to using @samp{set annotate @var{level}}
1118 (@pxref{Annotations}). The annotation @var{level} controls how much
1119 information @value{GDBN} prints together with its prompt, values of
1120 expressions, source lines, and other types of output. Level 0 is the
1121 normal, level 1 is for use when @value{GDBN} is run as a subprocess of
1122 @sc{gnu} Emacs, level 3 is the maximum annotation suitable for programs
1123 that control @value{GDBN}, and level 2 has been deprecated.
1125 The annotation mechanism has largely been superseded by @sc{gdb/mi}
1129 @cindex @code{--args}
1130 Change interpretation of command line so that arguments following the
1131 executable file are passed as command line arguments to the inferior.
1132 This option stops option processing.
1134 @item -baud @var{bps}
1136 @cindex @code{--baud}
1138 Set the line speed (baud rate or bits per second) of any serial
1139 interface used by @value{GDBN} for remote debugging.
1141 @item -l @var{timeout}
1143 Set the timeout (in seconds) of any communication used by @value{GDBN}
1144 for remote debugging.
1146 @item -tty @var{device}
1147 @itemx -t @var{device}
1148 @cindex @code{--tty}
1150 Run using @var{device} for your program's standard input and output.
1151 @c FIXME: kingdon thinks there is more to -tty. Investigate.
1153 @c resolve the situation of these eventually
1155 @cindex @code{--tui}
1156 Activate the @dfn{Text User Interface} when starting. The Text User
1157 Interface manages several text windows on the terminal, showing
1158 source, assembly, registers and @value{GDBN} command outputs
1159 (@pxref{TUI, ,@value{GDBN} Text User Interface}). Alternatively, the
1160 Text User Interface can be enabled by invoking the program
1161 @samp{@value{GDBTUI}}. Do not use this option if you run @value{GDBN} from
1162 Emacs (@pxref{Emacs, ,Using @value{GDBN} under @sc{gnu} Emacs}).
1165 @c @cindex @code{--xdb}
1166 @c Run in XDB compatibility mode, allowing the use of certain XDB commands.
1167 @c For information, see the file @file{xdb_trans.html}, which is usually
1168 @c installed in the directory @code{/opt/langtools/wdb/doc} on HP-UX
1171 @item -interpreter @var{interp}
1172 @cindex @code{--interpreter}
1173 Use the interpreter @var{interp} for interface with the controlling
1174 program or device. This option is meant to be set by programs which
1175 communicate with @value{GDBN} using it as a back end.
1176 @xref{Interpreters, , Command Interpreters}.
1178 @samp{--interpreter=mi} (or @samp{--interpreter=mi2}) causes
1179 @value{GDBN} to use the @dfn{@sc{gdb/mi} interface} (@pxref{GDB/MI, ,
1180 The @sc{gdb/mi} Interface}) included since @value{GDBN} version 6.0. The
1181 previous @sc{gdb/mi} interface, included in @value{GDBN} version 5.3 and
1182 selected with @samp{--interpreter=mi1}, is deprecated. Earlier
1183 @sc{gdb/mi} interfaces are no longer supported.
1186 @cindex @code{--write}
1187 Open the executable and core files for both reading and writing. This
1188 is equivalent to the @samp{set write on} command inside @value{GDBN}
1192 @cindex @code{--statistics}
1193 This option causes @value{GDBN} to print statistics about time and
1194 memory usage after it completes each command and returns to the prompt.
1197 @cindex @code{--version}
1198 This option causes @value{GDBN} to print its version number and
1199 no-warranty blurb, and exit.
1204 @subsection What @value{GDBN} Does During Startup
1205 @cindex @value{GDBN} startup
1207 Here's the description of what @value{GDBN} does during session startup:
1211 Sets up the command interpreter as specified by the command line
1212 (@pxref{Mode Options, interpreter}).
1216 Reads the system-wide @dfn{init file} (if @option{--with-system-gdbinit} was
1217 used when building @value{GDBN}; @pxref{System-wide configuration,
1218 ,System-wide configuration and settings}) and executes all the commands in
1222 Reads the init file (if any) in your home directory@footnote{On
1223 DOS/Windows systems, the home directory is the one pointed to by the
1224 @code{HOME} environment variable.} and executes all the commands in
1228 Processes command line options and operands.
1231 Reads and executes the commands from init file (if any) in the current
1232 working directory. This is only done if the current directory is
1233 different from your home directory. Thus, you can have more than one
1234 init file, one generic in your home directory, and another, specific
1235 to the program you are debugging, in the directory where you invoke
1239 Reads command files specified by the @samp{-x} option. @xref{Command
1240 Files}, for more details about @value{GDBN} command files.
1243 Reads the command history recorded in the @dfn{history file}.
1244 @xref{Command History}, for more details about the command history and the
1245 files where @value{GDBN} records it.
1248 Init files use the same syntax as @dfn{command files} (@pxref{Command
1249 Files}) and are processed by @value{GDBN} in the same way. The init
1250 file in your home directory can set options (such as @samp{set
1251 complaints}) that affect subsequent processing of command line options
1252 and operands. Init files are not executed if you use the @samp{-nx}
1253 option (@pxref{Mode Options, ,Choosing Modes}).
1255 To display the list of init files loaded by gdb at startup, you
1256 can use @kbd{gdb --help}.
1258 @cindex init file name
1259 @cindex @file{.gdbinit}
1260 @cindex @file{gdb.ini}
1261 The @value{GDBN} init files are normally called @file{.gdbinit}.
1262 The DJGPP port of @value{GDBN} uses the name @file{gdb.ini}, due to
1263 the limitations of file names imposed by DOS filesystems. The Windows
1264 ports of @value{GDBN} use the standard name, but if they find a
1265 @file{gdb.ini} file, they warn you about that and suggest to rename
1266 the file to the standard name.
1270 @section Quitting @value{GDBN}
1271 @cindex exiting @value{GDBN}
1272 @cindex leaving @value{GDBN}
1275 @kindex quit @r{[}@var{expression}@r{]}
1276 @kindex q @r{(@code{quit})}
1277 @item quit @r{[}@var{expression}@r{]}
1279 To exit @value{GDBN}, use the @code{quit} command (abbreviated
1280 @code{q}), or type an end-of-file character (usually @kbd{Ctrl-d}). If you
1281 do not supply @var{expression}, @value{GDBN} will terminate normally;
1282 otherwise it will terminate using the result of @var{expression} as the
1287 An interrupt (often @kbd{Ctrl-c}) does not exit from @value{GDBN}, but rather
1288 terminates the action of any @value{GDBN} command that is in progress and
1289 returns to @value{GDBN} command level. It is safe to type the interrupt
1290 character at any time because @value{GDBN} does not allow it to take effect
1291 until a time when it is safe.
1293 If you have been using @value{GDBN} to control an attached process or
1294 device, you can release it with the @code{detach} command
1295 (@pxref{Attach, ,Debugging an Already-running Process}).
1297 @node Shell Commands
1298 @section Shell Commands
1300 If you need to execute occasional shell commands during your
1301 debugging session, there is no need to leave or suspend @value{GDBN}; you can
1302 just use the @code{shell} command.
1306 @cindex shell escape
1307 @item shell @var{command string}
1308 Invoke a standard shell to execute @var{command string}.
1309 If it exists, the environment variable @code{SHELL} determines which
1310 shell to run. Otherwise @value{GDBN} uses the default shell
1311 (@file{/bin/sh} on Unix systems, @file{COMMAND.COM} on MS-DOS, etc.).
1314 The utility @code{make} is often needed in development environments.
1315 You do not have to use the @code{shell} command for this purpose in
1320 @cindex calling make
1321 @item make @var{make-args}
1322 Execute the @code{make} program with the specified
1323 arguments. This is equivalent to @samp{shell make @var{make-args}}.
1326 @node Logging Output
1327 @section Logging Output
1328 @cindex logging @value{GDBN} output
1329 @cindex save @value{GDBN} output to a file
1331 You may want to save the output of @value{GDBN} commands to a file.
1332 There are several commands to control @value{GDBN}'s logging.
1336 @item set logging on
1338 @item set logging off
1340 @cindex logging file name
1341 @item set logging file @var{file}
1342 Change the name of the current logfile. The default logfile is @file{gdb.txt}.
1343 @item set logging overwrite [on|off]
1344 By default, @value{GDBN} will append to the logfile. Set @code{overwrite} if
1345 you want @code{set logging on} to overwrite the logfile instead.
1346 @item set logging redirect [on|off]
1347 By default, @value{GDBN} output will go to both the terminal and the logfile.
1348 Set @code{redirect} if you want output to go only to the log file.
1349 @kindex show logging
1351 Show the current values of the logging settings.
1355 @chapter @value{GDBN} Commands
1357 You can abbreviate a @value{GDBN} command to the first few letters of the command
1358 name, if that abbreviation is unambiguous; and you can repeat certain
1359 @value{GDBN} commands by typing just @key{RET}. You can also use the @key{TAB}
1360 key to get @value{GDBN} to fill out the rest of a word in a command (or to
1361 show you the alternatives available, if there is more than one possibility).
1364 * Command Syntax:: How to give commands to @value{GDBN}
1365 * Completion:: Command completion
1366 * Help:: How to ask @value{GDBN} for help
1369 @node Command Syntax
1370 @section Command Syntax
1372 A @value{GDBN} command is a single line of input. There is no limit on
1373 how long it can be. It starts with a command name, which is followed by
1374 arguments whose meaning depends on the command name. For example, the
1375 command @code{step} accepts an argument which is the number of times to
1376 step, as in @samp{step 5}. You can also use the @code{step} command
1377 with no arguments. Some commands do not allow any arguments.
1379 @cindex abbreviation
1380 @value{GDBN} command names may always be truncated if that abbreviation is
1381 unambiguous. Other possible command abbreviations are listed in the
1382 documentation for individual commands. In some cases, even ambiguous
1383 abbreviations are allowed; for example, @code{s} is specially defined as
1384 equivalent to @code{step} even though there are other commands whose
1385 names start with @code{s}. You can test abbreviations by using them as
1386 arguments to the @code{help} command.
1388 @cindex repeating commands
1389 @kindex RET @r{(repeat last command)}
1390 A blank line as input to @value{GDBN} (typing just @key{RET}) means to
1391 repeat the previous command. Certain commands (for example, @code{run})
1392 will not repeat this way; these are commands whose unintentional
1393 repetition might cause trouble and which you are unlikely to want to
1394 repeat. User-defined commands can disable this feature; see
1395 @ref{Define, dont-repeat}.
1397 The @code{list} and @code{x} commands, when you repeat them with
1398 @key{RET}, construct new arguments rather than repeating
1399 exactly as typed. This permits easy scanning of source or memory.
1401 @value{GDBN} can also use @key{RET} in another way: to partition lengthy
1402 output, in a way similar to the common utility @code{more}
1403 (@pxref{Screen Size,,Screen Size}). Since it is easy to press one
1404 @key{RET} too many in this situation, @value{GDBN} disables command
1405 repetition after any command that generates this sort of display.
1407 @kindex # @r{(a comment)}
1409 Any text from a @kbd{#} to the end of the line is a comment; it does
1410 nothing. This is useful mainly in command files (@pxref{Command
1411 Files,,Command Files}).
1413 @cindex repeating command sequences
1414 @kindex Ctrl-o @r{(operate-and-get-next)}
1415 The @kbd{Ctrl-o} binding is useful for repeating a complex sequence of
1416 commands. This command accepts the current line, like @key{RET}, and
1417 then fetches the next line relative to the current line from the history
1421 @section Command Completion
1424 @cindex word completion
1425 @value{GDBN} can fill in the rest of a word in a command for you, if there is
1426 only one possibility; it can also show you what the valid possibilities
1427 are for the next word in a command, at any time. This works for @value{GDBN}
1428 commands, @value{GDBN} subcommands, and the names of symbols in your program.
1430 Press the @key{TAB} key whenever you want @value{GDBN} to fill out the rest
1431 of a word. If there is only one possibility, @value{GDBN} fills in the
1432 word, and waits for you to finish the command (or press @key{RET} to
1433 enter it). For example, if you type
1435 @c FIXME "@key" does not distinguish its argument sufficiently to permit
1436 @c complete accuracy in these examples; space introduced for clarity.
1437 @c If texinfo enhancements make it unnecessary, it would be nice to
1438 @c replace " @key" by "@key" in the following...
1440 (@value{GDBP}) info bre @key{TAB}
1444 @value{GDBN} fills in the rest of the word @samp{breakpoints}, since that is
1445 the only @code{info} subcommand beginning with @samp{bre}:
1448 (@value{GDBP}) info breakpoints
1452 You can either press @key{RET} at this point, to run the @code{info
1453 breakpoints} command, or backspace and enter something else, if
1454 @samp{breakpoints} does not look like the command you expected. (If you
1455 were sure you wanted @code{info breakpoints} in the first place, you
1456 might as well just type @key{RET} immediately after @samp{info bre},
1457 to exploit command abbreviations rather than command completion).
1459 If there is more than one possibility for the next word when you press
1460 @key{TAB}, @value{GDBN} sounds a bell. You can either supply more
1461 characters and try again, or just press @key{TAB} a second time;
1462 @value{GDBN} displays all the possible completions for that word. For
1463 example, you might want to set a breakpoint on a subroutine whose name
1464 begins with @samp{make_}, but when you type @kbd{b make_@key{TAB}} @value{GDBN}
1465 just sounds the bell. Typing @key{TAB} again displays all the
1466 function names in your program that begin with those characters, for
1470 (@value{GDBP}) b make_ @key{TAB}
1471 @exdent @value{GDBN} sounds bell; press @key{TAB} again, to see:
1472 make_a_section_from_file make_environ
1473 make_abs_section make_function_type
1474 make_blockvector make_pointer_type
1475 make_cleanup make_reference_type
1476 make_command make_symbol_completion_list
1477 (@value{GDBP}) b make_
1481 After displaying the available possibilities, @value{GDBN} copies your
1482 partial input (@samp{b make_} in the example) so you can finish the
1485 If you just want to see the list of alternatives in the first place, you
1486 can press @kbd{M-?} rather than pressing @key{TAB} twice. @kbd{M-?}
1487 means @kbd{@key{META} ?}. You can type this either by holding down a
1488 key designated as the @key{META} shift on your keyboard (if there is
1489 one) while typing @kbd{?}, or as @key{ESC} followed by @kbd{?}.
1491 @cindex quotes in commands
1492 @cindex completion of quoted strings
1493 Sometimes the string you need, while logically a ``word'', may contain
1494 parentheses or other characters that @value{GDBN} normally excludes from
1495 its notion of a word. To permit word completion to work in this
1496 situation, you may enclose words in @code{'} (single quote marks) in
1497 @value{GDBN} commands.
1499 The most likely situation where you might need this is in typing the
1500 name of a C@t{++} function. This is because C@t{++} allows function
1501 overloading (multiple definitions of the same function, distinguished
1502 by argument type). For example, when you want to set a breakpoint you
1503 may need to distinguish whether you mean the version of @code{name}
1504 that takes an @code{int} parameter, @code{name(int)}, or the version
1505 that takes a @code{float} parameter, @code{name(float)}. To use the
1506 word-completion facilities in this situation, type a single quote
1507 @code{'} at the beginning of the function name. This alerts
1508 @value{GDBN} that it may need to consider more information than usual
1509 when you press @key{TAB} or @kbd{M-?} to request word completion:
1512 (@value{GDBP}) b 'bubble( @kbd{M-?}
1513 bubble(double,double) bubble(int,int)
1514 (@value{GDBP}) b 'bubble(
1517 In some cases, @value{GDBN} can tell that completing a name requires using
1518 quotes. When this happens, @value{GDBN} inserts the quote for you (while
1519 completing as much as it can) if you do not type the quote in the first
1523 (@value{GDBP}) b bub @key{TAB}
1524 @exdent @value{GDBN} alters your input line to the following, and rings a bell:
1525 (@value{GDBP}) b 'bubble(
1529 In general, @value{GDBN} can tell that a quote is needed (and inserts it) if
1530 you have not yet started typing the argument list when you ask for
1531 completion on an overloaded symbol.
1533 For more information about overloaded functions, see @ref{C Plus Plus
1534 Expressions, ,C@t{++} Expressions}. You can use the command @code{set
1535 overload-resolution off} to disable overload resolution;
1536 see @ref{Debugging C Plus Plus, ,@value{GDBN} Features for C@t{++}}.
1538 @cindex completion of structure field names
1539 @cindex structure field name completion
1540 @cindex completion of union field names
1541 @cindex union field name completion
1542 When completing in an expression which looks up a field in a
1543 structure, @value{GDBN} also tries@footnote{The completer can be
1544 confused by certain kinds of invalid expressions. Also, it only
1545 examines the static type of the expression, not the dynamic type.} to
1546 limit completions to the field names available in the type of the
1550 (@value{GDBP}) p gdb_stdout.@kbd{M-?}
1551 magic to_delete to_fputs to_put to_rewind
1552 to_data to_flush to_isatty to_read to_write
1556 This is because the @code{gdb_stdout} is a variable of the type
1557 @code{struct ui_file} that is defined in @value{GDBN} sources as
1564 ui_file_flush_ftype *to_flush;
1565 ui_file_write_ftype *to_write;
1566 ui_file_fputs_ftype *to_fputs;
1567 ui_file_read_ftype *to_read;
1568 ui_file_delete_ftype *to_delete;
1569 ui_file_isatty_ftype *to_isatty;
1570 ui_file_rewind_ftype *to_rewind;
1571 ui_file_put_ftype *to_put;
1578 @section Getting Help
1579 @cindex online documentation
1582 You can always ask @value{GDBN} itself for information on its commands,
1583 using the command @code{help}.
1586 @kindex h @r{(@code{help})}
1589 You can use @code{help} (abbreviated @code{h}) with no arguments to
1590 display a short list of named classes of commands:
1594 List of classes of commands:
1596 aliases -- Aliases of other commands
1597 breakpoints -- Making program stop at certain points
1598 data -- Examining data
1599 files -- Specifying and examining files
1600 internals -- Maintenance commands
1601 obscure -- Obscure features
1602 running -- Running the program
1603 stack -- Examining the stack
1604 status -- Status inquiries
1605 support -- Support facilities
1606 tracepoints -- Tracing of program execution without
1607 stopping the program
1608 user-defined -- User-defined commands
1610 Type "help" followed by a class name for a list of
1611 commands in that class.
1612 Type "help" followed by command name for full
1614 Command name abbreviations are allowed if unambiguous.
1617 @c the above line break eliminates huge line overfull...
1619 @item help @var{class}
1620 Using one of the general help classes as an argument, you can get a
1621 list of the individual commands in that class. For example, here is the
1622 help display for the class @code{status}:
1625 (@value{GDBP}) help status
1630 @c Line break in "show" line falsifies real output, but needed
1631 @c to fit in smallbook page size.
1632 info -- Generic command for showing things
1633 about the program being debugged
1634 show -- Generic command for showing things
1637 Type "help" followed by command name for full
1639 Command name abbreviations are allowed if unambiguous.
1643 @item help @var{command}
1644 With a command name as @code{help} argument, @value{GDBN} displays a
1645 short paragraph on how to use that command.
1648 @item apropos @var{args}
1649 The @code{apropos} command searches through all of the @value{GDBN}
1650 commands, and their documentation, for the regular expression specified in
1651 @var{args}. It prints out all matches found. For example:
1662 set symbol-reloading -- Set dynamic symbol table reloading
1663 multiple times in one run
1664 show symbol-reloading -- Show dynamic symbol table reloading
1665 multiple times in one run
1670 @item complete @var{args}
1671 The @code{complete @var{args}} command lists all the possible completions
1672 for the beginning of a command. Use @var{args} to specify the beginning of the
1673 command you want completed. For example:
1679 @noindent results in:
1690 @noindent This is intended for use by @sc{gnu} Emacs.
1693 In addition to @code{help}, you can use the @value{GDBN} commands @code{info}
1694 and @code{show} to inquire about the state of your program, or the state
1695 of @value{GDBN} itself. Each command supports many topics of inquiry; this
1696 manual introduces each of them in the appropriate context. The listings
1697 under @code{info} and under @code{show} in the Index point to
1698 all the sub-commands. @xref{Index}.
1703 @kindex i @r{(@code{info})}
1705 This command (abbreviated @code{i}) is for describing the state of your
1706 program. For example, you can show the arguments passed to a function
1707 with @code{info args}, list the registers currently in use with @code{info
1708 registers}, or list the breakpoints you have set with @code{info breakpoints}.
1709 You can get a complete list of the @code{info} sub-commands with
1710 @w{@code{help info}}.
1714 You can assign the result of an expression to an environment variable with
1715 @code{set}. For example, you can set the @value{GDBN} prompt to a $-sign with
1716 @code{set prompt $}.
1720 In contrast to @code{info}, @code{show} is for describing the state of
1721 @value{GDBN} itself.
1722 You can change most of the things you can @code{show}, by using the
1723 related command @code{set}; for example, you can control what number
1724 system is used for displays with @code{set radix}, or simply inquire
1725 which is currently in use with @code{show radix}.
1728 To display all the settable parameters and their current
1729 values, you can use @code{show} with no arguments; you may also use
1730 @code{info set}. Both commands produce the same display.
1731 @c FIXME: "info set" violates the rule that "info" is for state of
1732 @c FIXME...program. Ck w/ GNU: "info set" to be called something else,
1733 @c FIXME...or change desc of rule---eg "state of prog and debugging session"?
1737 Here are three miscellaneous @code{show} subcommands, all of which are
1738 exceptional in lacking corresponding @code{set} commands:
1741 @kindex show version
1742 @cindex @value{GDBN} version number
1744 Show what version of @value{GDBN} is running. You should include this
1745 information in @value{GDBN} bug-reports. If multiple versions of
1746 @value{GDBN} are in use at your site, you may need to determine which
1747 version of @value{GDBN} you are running; as @value{GDBN} evolves, new
1748 commands are introduced, and old ones may wither away. Also, many
1749 system vendors ship variant versions of @value{GDBN}, and there are
1750 variant versions of @value{GDBN} in @sc{gnu}/Linux distributions as well.
1751 The version number is the same as the one announced when you start
1754 @kindex show copying
1755 @kindex info copying
1756 @cindex display @value{GDBN} copyright
1759 Display information about permission for copying @value{GDBN}.
1761 @kindex show warranty
1762 @kindex info warranty
1764 @itemx info warranty
1765 Display the @sc{gnu} ``NO WARRANTY'' statement, or a warranty,
1766 if your version of @value{GDBN} comes with one.
1771 @chapter Running Programs Under @value{GDBN}
1773 When you run a program under @value{GDBN}, you must first generate
1774 debugging information when you compile it.
1776 You may start @value{GDBN} with its arguments, if any, in an environment
1777 of your choice. If you are doing native debugging, you may redirect
1778 your program's input and output, debug an already running process, or
1779 kill a child process.
1782 * Compilation:: Compiling for debugging
1783 * Starting:: Starting your program
1784 * Arguments:: Your program's arguments
1785 * Environment:: Your program's environment
1787 * Working Directory:: Your program's working directory
1788 * Input/Output:: Your program's input and output
1789 * Attach:: Debugging an already-running process
1790 * Kill Process:: Killing the child process
1792 * Inferiors:: Debugging multiple inferiors
1793 * Threads:: Debugging programs with multiple threads
1794 * Processes:: Debugging programs with multiple processes
1795 * Checkpoint/Restart:: Setting a @emph{bookmark} to return to later
1799 @section Compiling for Debugging
1801 In order to debug a program effectively, you need to generate
1802 debugging information when you compile it. This debugging information
1803 is stored in the object file; it describes the data type of each
1804 variable or function and the correspondence between source line numbers
1805 and addresses in the executable code.
1807 To request debugging information, specify the @samp{-g} option when you run
1810 Programs that are to be shipped to your customers are compiled with
1811 optimizations, using the @samp{-O} compiler option. However, many
1812 compilers are unable to handle the @samp{-g} and @samp{-O} options
1813 together. Using those compilers, you cannot generate optimized
1814 executables containing debugging information.
1816 @value{NGCC}, the @sc{gnu} C/C@t{++} compiler, supports @samp{-g} with or
1817 without @samp{-O}, making it possible to debug optimized code. We
1818 recommend that you @emph{always} use @samp{-g} whenever you compile a
1819 program. You may think your program is correct, but there is no sense
1820 in pushing your luck.
1822 @cindex optimized code, debugging
1823 @cindex debugging optimized code
1824 When you debug a program compiled with @samp{-g -O}, remember that the
1825 optimizer is rearranging your code; the debugger shows you what is
1826 really there. Do not be too surprised when the execution path does not
1827 exactly match your source file! An extreme example: if you define a
1828 variable, but never use it, @value{GDBN} never sees that
1829 variable---because the compiler optimizes it out of existence.
1831 Some things do not work as well with @samp{-g -O} as with just
1832 @samp{-g}, particularly on machines with instruction scheduling. If in
1833 doubt, recompile with @samp{-g} alone, and if this fixes the problem,
1834 please report it to us as a bug (including a test case!).
1835 @xref{Variables}, for more information about debugging optimized code.
1837 Older versions of the @sc{gnu} C compiler permitted a variant option
1838 @w{@samp{-gg}} for debugging information. @value{GDBN} no longer supports this
1839 format; if your @sc{gnu} C compiler has this option, do not use it.
1841 @value{GDBN} knows about preprocessor macros and can show you their
1842 expansion (@pxref{Macros}). Most compilers do not include information
1843 about preprocessor macros in the debugging information if you specify
1844 the @option{-g} flag alone, because this information is rather large.
1845 Version 3.1 and later of @value{NGCC}, the @sc{gnu} C compiler,
1846 provides macro information if you specify the options
1847 @option{-gdwarf-2} and @option{-g3}; the former option requests
1848 debugging information in the Dwarf 2 format, and the latter requests
1849 ``extra information''. In the future, we hope to find more compact
1850 ways to represent macro information, so that it can be included with
1855 @section Starting your Program
1861 @kindex r @r{(@code{run})}
1864 Use the @code{run} command to start your program under @value{GDBN}.
1865 You must first specify the program name (except on VxWorks) with an
1866 argument to @value{GDBN} (@pxref{Invocation, ,Getting In and Out of
1867 @value{GDBN}}), or by using the @code{file} or @code{exec-file} command
1868 (@pxref{Files, ,Commands to Specify Files}).
1872 If you are running your program in an execution environment that
1873 supports processes, @code{run} creates an inferior process and makes
1874 that process run your program. In some environments without processes,
1875 @code{run} jumps to the start of your program. Other targets,
1876 like @samp{remote}, are always running. If you get an error
1877 message like this one:
1880 The "remote" target does not support "run".
1881 Try "help target" or "continue".
1885 then use @code{continue} to run your program. You may need @code{load}
1886 first (@pxref{load}).
1888 The execution of a program is affected by certain information it
1889 receives from its superior. @value{GDBN} provides ways to specify this
1890 information, which you must do @emph{before} starting your program. (You
1891 can change it after starting your program, but such changes only affect
1892 your program the next time you start it.) This information may be
1893 divided into four categories:
1896 @item The @emph{arguments.}
1897 Specify the arguments to give your program as the arguments of the
1898 @code{run} command. If a shell is available on your target, the shell
1899 is used to pass the arguments, so that you may use normal conventions
1900 (such as wildcard expansion or variable substitution) in describing
1902 In Unix systems, you can control which shell is used with the
1903 @code{SHELL} environment variable.
1904 @xref{Arguments, ,Your Program's Arguments}.
1906 @item The @emph{environment.}
1907 Your program normally inherits its environment from @value{GDBN}, but you can
1908 use the @value{GDBN} commands @code{set environment} and @code{unset
1909 environment} to change parts of the environment that affect
1910 your program. @xref{Environment, ,Your Program's Environment}.
1912 @item The @emph{working directory.}
1913 Your program inherits its working directory from @value{GDBN}. You can set
1914 the @value{GDBN} working directory with the @code{cd} command in @value{GDBN}.
1915 @xref{Working Directory, ,Your Program's Working Directory}.
1917 @item The @emph{standard input and output.}
1918 Your program normally uses the same device for standard input and
1919 standard output as @value{GDBN} is using. You can redirect input and output
1920 in the @code{run} command line, or you can use the @code{tty} command to
1921 set a different device for your program.
1922 @xref{Input/Output, ,Your Program's Input and Output}.
1925 @emph{Warning:} While input and output redirection work, you cannot use
1926 pipes to pass the output of the program you are debugging to another
1927 program; if you attempt this, @value{GDBN} is likely to wind up debugging the
1931 When you issue the @code{run} command, your program begins to execute
1932 immediately. @xref{Stopping, ,Stopping and Continuing}, for discussion
1933 of how to arrange for your program to stop. Once your program has
1934 stopped, you may call functions in your program, using the @code{print}
1935 or @code{call} commands. @xref{Data, ,Examining Data}.
1937 If the modification time of your symbol file has changed since the last
1938 time @value{GDBN} read its symbols, @value{GDBN} discards its symbol
1939 table, and reads it again. When it does this, @value{GDBN} tries to retain
1940 your current breakpoints.
1945 @cindex run to main procedure
1946 The name of the main procedure can vary from language to language.
1947 With C or C@t{++}, the main procedure name is always @code{main}, but
1948 other languages such as Ada do not require a specific name for their
1949 main procedure. The debugger provides a convenient way to start the
1950 execution of the program and to stop at the beginning of the main
1951 procedure, depending on the language used.
1953 The @samp{start} command does the equivalent of setting a temporary
1954 breakpoint at the beginning of the main procedure and then invoking
1955 the @samp{run} command.
1957 @cindex elaboration phase
1958 Some programs contain an @dfn{elaboration} phase where some startup code is
1959 executed before the main procedure is called. This depends on the
1960 languages used to write your program. In C@t{++}, for instance,
1961 constructors for static and global objects are executed before
1962 @code{main} is called. It is therefore possible that the debugger stops
1963 before reaching the main procedure. However, the temporary breakpoint
1964 will remain to halt execution.
1966 Specify the arguments to give to your program as arguments to the
1967 @samp{start} command. These arguments will be given verbatim to the
1968 underlying @samp{run} command. Note that the same arguments will be
1969 reused if no argument is provided during subsequent calls to
1970 @samp{start} or @samp{run}.
1972 It is sometimes necessary to debug the program during elaboration. In
1973 these cases, using the @code{start} command would stop the execution of
1974 your program too late, as the program would have already completed the
1975 elaboration phase. Under these circumstances, insert breakpoints in your
1976 elaboration code before running your program.
1978 @kindex set exec-wrapper
1979 @item set exec-wrapper @var{wrapper}
1980 @itemx show exec-wrapper
1981 @itemx unset exec-wrapper
1982 When @samp{exec-wrapper} is set, the specified wrapper is used to
1983 launch programs for debugging. @value{GDBN} starts your program
1984 with a shell command of the form @kbd{exec @var{wrapper}
1985 @var{program}}. Quoting is added to @var{program} and its
1986 arguments, but not to @var{wrapper}, so you should add quotes if
1987 appropriate for your shell. The wrapper runs until it executes
1988 your program, and then @value{GDBN} takes control.
1990 You can use any program that eventually calls @code{execve} with
1991 its arguments as a wrapper. Several standard Unix utilities do
1992 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
1993 with @code{exec "$@@"} will also work.
1995 For example, you can use @code{env} to pass an environment variable to
1996 the debugged program, without setting the variable in your shell's
2000 (@value{GDBP}) set exec-wrapper env 'LD_PRELOAD=libtest.so'
2004 This command is available when debugging locally on most targets, excluding
2005 @sc{djgpp}, Cygwin, MS Windows, and QNX Neutrino.
2007 @kindex set disable-randomization
2008 @item set disable-randomization
2009 @itemx set disable-randomization on
2010 This option (enabled by default in @value{GDBN}) will turn off the native
2011 randomization of the virtual address space of the started program. This option
2012 is useful for multiple debugging sessions to make the execution better
2013 reproducible and memory addresses reusable across debugging sessions.
2015 This feature is implemented only on @sc{gnu}/Linux. You can get the same
2019 (@value{GDBP}) set exec-wrapper setarch `uname -m` -R
2022 @item set disable-randomization off
2023 Leave the behavior of the started executable unchanged. Some bugs rear their
2024 ugly heads only when the program is loaded at certain addresses. If your bug
2025 disappears when you run the program under @value{GDBN}, that might be because
2026 @value{GDBN} by default disables the address randomization on platforms, such
2027 as @sc{gnu}/Linux, which do that for stand-alone programs. Use @kbd{set
2028 disable-randomization off} to try to reproduce such elusive bugs.
2030 The virtual address space randomization is implemented only on @sc{gnu}/Linux.
2031 It protects the programs against some kinds of security attacks. In these
2032 cases the attacker needs to know the exact location of a concrete executable
2033 code. Randomizing its location makes it impossible to inject jumps misusing
2034 a code at its expected addresses.
2036 Prelinking shared libraries provides a startup performance advantage but it
2037 makes addresses in these libraries predictable for privileged processes by
2038 having just unprivileged access at the target system. Reading the shared
2039 library binary gives enough information for assembling the malicious code
2040 misusing it. Still even a prelinked shared library can get loaded at a new
2041 random address just requiring the regular relocation process during the
2042 startup. Shared libraries not already prelinked are always loaded at
2043 a randomly chosen address.
2045 Position independent executables (PIE) contain position independent code
2046 similar to the shared libraries and therefore such executables get loaded at
2047 a randomly chosen address upon startup. PIE executables always load even
2048 already prelinked shared libraries at a random address. You can build such
2049 executable using @command{gcc -fPIE -pie}.
2051 Heap (malloc storage), stack and custom mmap areas are always placed randomly
2052 (as long as the randomization is enabled).
2054 @item show disable-randomization
2055 Show the current setting of the explicit disable of the native randomization of
2056 the virtual address space of the started program.
2061 @section Your Program's Arguments
2063 @cindex arguments (to your program)
2064 The arguments to your program can be specified by the arguments of the
2066 They are passed to a shell, which expands wildcard characters and
2067 performs redirection of I/O, and thence to your program. Your
2068 @code{SHELL} environment variable (if it exists) specifies what shell
2069 @value{GDBN} uses. If you do not define @code{SHELL}, @value{GDBN} uses
2070 the default shell (@file{/bin/sh} on Unix).
2072 On non-Unix systems, the program is usually invoked directly by
2073 @value{GDBN}, which emulates I/O redirection via the appropriate system
2074 calls, and the wildcard characters are expanded by the startup code of
2075 the program, not by the shell.
2077 @code{run} with no arguments uses the same arguments used by the previous
2078 @code{run}, or those set by the @code{set args} command.
2083 Specify the arguments to be used the next time your program is run. If
2084 @code{set args} has no arguments, @code{run} executes your program
2085 with no arguments. Once you have run your program with arguments,
2086 using @code{set args} before the next @code{run} is the only way to run
2087 it again without arguments.
2091 Show the arguments to give your program when it is started.
2095 @section Your Program's Environment
2097 @cindex environment (of your program)
2098 The @dfn{environment} consists of a set of environment variables and
2099 their values. Environment variables conventionally record such things as
2100 your user name, your home directory, your terminal type, and your search
2101 path for programs to run. Usually you set up environment variables with
2102 the shell and they are inherited by all the other programs you run. When
2103 debugging, it can be useful to try running your program with a modified
2104 environment without having to start @value{GDBN} over again.
2108 @item path @var{directory}
2109 Add @var{directory} to the front of the @code{PATH} environment variable
2110 (the search path for executables) that will be passed to your program.
2111 The value of @code{PATH} used by @value{GDBN} does not change.
2112 You may specify several directory names, separated by whitespace or by a
2113 system-dependent separator character (@samp{:} on Unix, @samp{;} on
2114 MS-DOS and MS-Windows). If @var{directory} is already in the path, it
2115 is moved to the front, so it is searched sooner.
2117 You can use the string @samp{$cwd} to refer to whatever is the current
2118 working directory at the time @value{GDBN} searches the path. If you
2119 use @samp{.} instead, it refers to the directory where you executed the
2120 @code{path} command. @value{GDBN} replaces @samp{.} in the
2121 @var{directory} argument (with the current path) before adding
2122 @var{directory} to the search path.
2123 @c 'path' is explicitly nonrepeatable, but RMS points out it is silly to
2124 @c document that, since repeating it would be a no-op.
2128 Display the list of search paths for executables (the @code{PATH}
2129 environment variable).
2131 @kindex show environment
2132 @item show environment @r{[}@var{varname}@r{]}
2133 Print the value of environment variable @var{varname} to be given to
2134 your program when it starts. If you do not supply @var{varname},
2135 print the names and values of all environment variables to be given to
2136 your program. You can abbreviate @code{environment} as @code{env}.
2138 @kindex set environment
2139 @item set environment @var{varname} @r{[}=@var{value}@r{]}
2140 Set environment variable @var{varname} to @var{value}. The value
2141 changes for your program only, not for @value{GDBN} itself. @var{value} may
2142 be any string; the values of environment variables are just strings, and
2143 any interpretation is supplied by your program itself. The @var{value}
2144 parameter is optional; if it is eliminated, the variable is set to a
2146 @c "any string" here does not include leading, trailing
2147 @c blanks. Gnu asks: does anyone care?
2149 For example, this command:
2156 tells the debugged program, when subsequently run, that its user is named
2157 @samp{foo}. (The spaces around @samp{=} are used for clarity here; they
2158 are not actually required.)
2160 @kindex unset environment
2161 @item unset environment @var{varname}
2162 Remove variable @var{varname} from the environment to be passed to your
2163 program. This is different from @samp{set env @var{varname} =};
2164 @code{unset environment} removes the variable from the environment,
2165 rather than assigning it an empty value.
2168 @emph{Warning:} On Unix systems, @value{GDBN} runs your program using
2170 by your @code{SHELL} environment variable if it exists (or
2171 @code{/bin/sh} if not). If your @code{SHELL} variable names a shell
2172 that runs an initialization file---such as @file{.cshrc} for C-shell, or
2173 @file{.bashrc} for BASH---any variables you set in that file affect
2174 your program. You may wish to move setting of environment variables to
2175 files that are only run when you sign on, such as @file{.login} or
2178 @node Working Directory
2179 @section Your Program's Working Directory
2181 @cindex working directory (of your program)
2182 Each time you start your program with @code{run}, it inherits its
2183 working directory from the current working directory of @value{GDBN}.
2184 The @value{GDBN} working directory is initially whatever it inherited
2185 from its parent process (typically the shell), but you can specify a new
2186 working directory in @value{GDBN} with the @code{cd} command.
2188 The @value{GDBN} working directory also serves as a default for the commands
2189 that specify files for @value{GDBN} to operate on. @xref{Files, ,Commands to
2194 @cindex change working directory
2195 @item cd @var{directory}
2196 Set the @value{GDBN} working directory to @var{directory}.
2200 Print the @value{GDBN} working directory.
2203 It is generally impossible to find the current working directory of
2204 the process being debugged (since a program can change its directory
2205 during its run). If you work on a system where @value{GDBN} is
2206 configured with the @file{/proc} support, you can use the @code{info
2207 proc} command (@pxref{SVR4 Process Information}) to find out the
2208 current working directory of the debuggee.
2211 @section Your Program's Input and Output
2216 By default, the program you run under @value{GDBN} does input and output to
2217 the same terminal that @value{GDBN} uses. @value{GDBN} switches the terminal
2218 to its own terminal modes to interact with you, but it records the terminal
2219 modes your program was using and switches back to them when you continue
2220 running your program.
2223 @kindex info terminal
2225 Displays information recorded by @value{GDBN} about the terminal modes your
2229 You can redirect your program's input and/or output using shell
2230 redirection with the @code{run} command. For example,
2237 starts your program, diverting its output to the file @file{outfile}.
2240 @cindex controlling terminal
2241 Another way to specify where your program should do input and output is
2242 with the @code{tty} command. This command accepts a file name as
2243 argument, and causes this file to be the default for future @code{run}
2244 commands. It also resets the controlling terminal for the child
2245 process, for future @code{run} commands. For example,
2252 directs that processes started with subsequent @code{run} commands
2253 default to do input and output on the terminal @file{/dev/ttyb} and have
2254 that as their controlling terminal.
2256 An explicit redirection in @code{run} overrides the @code{tty} command's
2257 effect on the input/output device, but not its effect on the controlling
2260 When you use the @code{tty} command or redirect input in the @code{run}
2261 command, only the input @emph{for your program} is affected. The input
2262 for @value{GDBN} still comes from your terminal. @code{tty} is an alias
2263 for @code{set inferior-tty}.
2265 @cindex inferior tty
2266 @cindex set inferior controlling terminal
2267 You can use the @code{show inferior-tty} command to tell @value{GDBN} to
2268 display the name of the terminal that will be used for future runs of your
2272 @item set inferior-tty /dev/ttyb
2273 @kindex set inferior-tty
2274 Set the tty for the program being debugged to /dev/ttyb.
2276 @item show inferior-tty
2277 @kindex show inferior-tty
2278 Show the current tty for the program being debugged.
2282 @section Debugging an Already-running Process
2287 @item attach @var{process-id}
2288 This command attaches to a running process---one that was started
2289 outside @value{GDBN}. (@code{info files} shows your active
2290 targets.) The command takes as argument a process ID. The usual way to
2291 find out the @var{process-id} of a Unix process is with the @code{ps} utility,
2292 or with the @samp{jobs -l} shell command.
2294 @code{attach} does not repeat if you press @key{RET} a second time after
2295 executing the command.
2298 To use @code{attach}, your program must be running in an environment
2299 which supports processes; for example, @code{attach} does not work for
2300 programs on bare-board targets that lack an operating system. You must
2301 also have permission to send the process a signal.
2303 When you use @code{attach}, the debugger finds the program running in
2304 the process first by looking in the current working directory, then (if
2305 the program is not found) by using the source file search path
2306 (@pxref{Source Path, ,Specifying Source Directories}). You can also use
2307 the @code{file} command to load the program. @xref{Files, ,Commands to
2310 The first thing @value{GDBN} does after arranging to debug the specified
2311 process is to stop it. You can examine and modify an attached process
2312 with all the @value{GDBN} commands that are ordinarily available when
2313 you start processes with @code{run}. You can insert breakpoints; you
2314 can step and continue; you can modify storage. If you would rather the
2315 process continue running, you may use the @code{continue} command after
2316 attaching @value{GDBN} to the process.
2321 When you have finished debugging the attached process, you can use the
2322 @code{detach} command to release it from @value{GDBN} control. Detaching
2323 the process continues its execution. After the @code{detach} command,
2324 that process and @value{GDBN} become completely independent once more, and you
2325 are ready to @code{attach} another process or start one with @code{run}.
2326 @code{detach} does not repeat if you press @key{RET} again after
2327 executing the command.
2330 If you exit @value{GDBN} while you have an attached process, you detach
2331 that process. If you use the @code{run} command, you kill that process.
2332 By default, @value{GDBN} asks for confirmation if you try to do either of these
2333 things; you can control whether or not you need to confirm by using the
2334 @code{set confirm} command (@pxref{Messages/Warnings, ,Optional Warnings and
2338 @section Killing the Child Process
2343 Kill the child process in which your program is running under @value{GDBN}.
2346 This command is useful if you wish to debug a core dump instead of a
2347 running process. @value{GDBN} ignores any core dump file while your program
2350 On some operating systems, a program cannot be executed outside @value{GDBN}
2351 while you have breakpoints set on it inside @value{GDBN}. You can use the
2352 @code{kill} command in this situation to permit running your program
2353 outside the debugger.
2355 The @code{kill} command is also useful if you wish to recompile and
2356 relink your program, since on many systems it is impossible to modify an
2357 executable file while it is running in a process. In this case, when you
2358 next type @code{run}, @value{GDBN} notices that the file has changed, and
2359 reads the symbol table again (while trying to preserve your current
2360 breakpoint settings).
2363 @section Debugging Multiple Inferiors
2365 Some @value{GDBN} targets are able to run multiple processes created
2366 from a single executable. This can happen, for instance, with an
2367 embedded system reporting back several processes via the remote
2371 @value{GDBN} represents the state of each program execution with an
2372 object called an @dfn{inferior}. An inferior typically corresponds to
2373 a process, but is more general and applies also to targets that do not
2374 have processes. Inferiors may be created before a process runs, and
2375 may (in future) be retained after a process exits. Each run of an
2376 executable creates a new inferior, as does each attachment to an
2377 existing process. Inferiors have unique identifiers that are
2378 different from process ids, and may optionally be named as well.
2379 Usually each inferior will also have its own distinct address space,
2380 although some embedded targets may have several inferiors running in
2381 different parts of a single space.
2383 Each inferior may in turn have multiple threads running in it.
2385 To find out what inferiors exist at any moment, use @code{info inferiors}:
2388 @kindex info inferiors
2389 @item info inferiors
2390 Print a list of all inferiors currently being managed by @value{GDBN}.
2392 @kindex set print inferior-events
2393 @cindex print messages on inferior start and exit
2394 @item set print inferior-events
2395 @itemx set print inferior-events on
2396 @itemx set print inferior-events off
2397 The @code{set print inferior-events} command allows you to enable or
2398 disable printing of messages when @value{GDBN} notices that new
2399 inferiors have started or that inferiors have exited or have been
2400 detached. By default, these messages will not be printed.
2402 @kindex show print inferior-events
2403 @item show print inferior-events
2404 Show whether messages will be printed when @value{GDBN} detects that
2405 inferiors have started, exited or have been detached.
2409 @section Debugging Programs with Multiple Threads
2411 @cindex threads of execution
2412 @cindex multiple threads
2413 @cindex switching threads
2414 In some operating systems, such as HP-UX and Solaris, a single program
2415 may have more than one @dfn{thread} of execution. The precise semantics
2416 of threads differ from one operating system to another, but in general
2417 the threads of a single program are akin to multiple processes---except
2418 that they share one address space (that is, they can all examine and
2419 modify the same variables). On the other hand, each thread has its own
2420 registers and execution stack, and perhaps private memory.
2422 @value{GDBN} provides these facilities for debugging multi-thread
2426 @item automatic notification of new threads
2427 @item @samp{thread @var{threadno}}, a command to switch among threads
2428 @item @samp{info threads}, a command to inquire about existing threads
2429 @item @samp{thread apply [@var{threadno}] [@var{all}] @var{args}},
2430 a command to apply a command to a list of threads
2431 @item thread-specific breakpoints
2432 @item @samp{set print thread-events}, which controls printing of
2433 messages on thread start and exit.
2434 @item @samp{set libthread-db-search-path @var{path}}, which lets
2435 the user specify which @code{libthread_db} to use if the default choice
2436 isn't compatible with the program.
2440 @emph{Warning:} These facilities are not yet available on every
2441 @value{GDBN} configuration where the operating system supports threads.
2442 If your @value{GDBN} does not support threads, these commands have no
2443 effect. For example, a system without thread support shows no output
2444 from @samp{info threads}, and always rejects the @code{thread} command,
2448 (@value{GDBP}) info threads
2449 (@value{GDBP}) thread 1
2450 Thread ID 1 not known. Use the "info threads" command to
2451 see the IDs of currently known threads.
2453 @c FIXME to implementors: how hard would it be to say "sorry, this GDB
2454 @c doesn't support threads"?
2457 @cindex focus of debugging
2458 @cindex current thread
2459 The @value{GDBN} thread debugging facility allows you to observe all
2460 threads while your program runs---but whenever @value{GDBN} takes
2461 control, one thread in particular is always the focus of debugging.
2462 This thread is called the @dfn{current thread}. Debugging commands show
2463 program information from the perspective of the current thread.
2465 @cindex @code{New} @var{systag} message
2466 @cindex thread identifier (system)
2467 @c FIXME-implementors!! It would be more helpful if the [New...] message
2468 @c included GDB's numeric thread handle, so you could just go to that
2469 @c thread without first checking `info threads'.
2470 Whenever @value{GDBN} detects a new thread in your program, it displays
2471 the target system's identification for the thread with a message in the
2472 form @samp{[New @var{systag}]}. @var{systag} is a thread identifier
2473 whose form varies depending on the particular system. For example, on
2474 @sc{gnu}/Linux, you might see
2477 [New Thread 46912507313328 (LWP 25582)]
2481 when @value{GDBN} notices a new thread. In contrast, on an SGI system,
2482 the @var{systag} is simply something like @samp{process 368}, with no
2485 @c FIXME!! (1) Does the [New...] message appear even for the very first
2486 @c thread of a program, or does it only appear for the
2487 @c second---i.e.@: when it becomes obvious we have a multithread
2489 @c (2) *Is* there necessarily a first thread always? Or do some
2490 @c multithread systems permit starting a program with multiple
2491 @c threads ab initio?
2493 @cindex thread number
2494 @cindex thread identifier (GDB)
2495 For debugging purposes, @value{GDBN} associates its own thread
2496 number---always a single integer---with each thread in your program.
2499 @kindex info threads
2501 Display a summary of all threads currently in your
2502 program. @value{GDBN} displays for each thread (in this order):
2506 the thread number assigned by @value{GDBN}
2509 the target system's thread identifier (@var{systag})
2512 the current stack frame summary for that thread
2516 An asterisk @samp{*} to the left of the @value{GDBN} thread number
2517 indicates the current thread.
2521 @c end table here to get a little more width for example
2524 (@value{GDBP}) info threads
2525 3 process 35 thread 27 0x34e5 in sigpause ()
2526 2 process 35 thread 23 0x34e5 in sigpause ()
2527 * 1 process 35 thread 13 main (argc=1, argv=0x7ffffff8)
2533 @cindex debugging multithreaded programs (on HP-UX)
2534 @cindex thread identifier (GDB), on HP-UX
2535 For debugging purposes, @value{GDBN} associates its own thread
2536 number---a small integer assigned in thread-creation order---with each
2537 thread in your program.
2539 @cindex @code{New} @var{systag} message, on HP-UX
2540 @cindex thread identifier (system), on HP-UX
2541 @c FIXME-implementors!! It would be more helpful if the [New...] message
2542 @c included GDB's numeric thread handle, so you could just go to that
2543 @c thread without first checking `info threads'.
2544 Whenever @value{GDBN} detects a new thread in your program, it displays
2545 both @value{GDBN}'s thread number and the target system's identification for the thread with a message in the
2546 form @samp{[New @var{systag}]}. @var{systag} is a thread identifier
2547 whose form varies depending on the particular system. For example, on
2551 [New thread 2 (system thread 26594)]
2555 when @value{GDBN} notices a new thread.
2558 @kindex info threads (HP-UX)
2560 Display a summary of all threads currently in your
2561 program. @value{GDBN} displays for each thread (in this order):
2564 @item the thread number assigned by @value{GDBN}
2566 @item the target system's thread identifier (@var{systag})
2568 @item the current stack frame summary for that thread
2572 An asterisk @samp{*} to the left of the @value{GDBN} thread number
2573 indicates the current thread.
2577 @c end table here to get a little more width for example
2580 (@value{GDBP}) info threads
2581 * 3 system thread 26607 worker (wptr=0x7b09c318 "@@") \@*
2583 2 system thread 26606 0x7b0030d8 in __ksleep () \@*
2584 from /usr/lib/libc.2
2585 1 system thread 27905 0x7b003498 in _brk () \@*
2586 from /usr/lib/libc.2
2589 On Solaris, you can display more information about user threads with a
2590 Solaris-specific command:
2593 @item maint info sol-threads
2594 @kindex maint info sol-threads
2595 @cindex thread info (Solaris)
2596 Display info on Solaris user threads.
2600 @kindex thread @var{threadno}
2601 @item thread @var{threadno}
2602 Make thread number @var{threadno} the current thread. The command
2603 argument @var{threadno} is the internal @value{GDBN} thread number, as
2604 shown in the first field of the @samp{info threads} display.
2605 @value{GDBN} responds by displaying the system identifier of the thread
2606 you selected, and its current stack frame summary:
2609 @c FIXME!! This example made up; find a @value{GDBN} w/threads and get real one
2610 (@value{GDBP}) thread 2
2611 [Switching to process 35 thread 23]
2612 0x34e5 in sigpause ()
2616 As with the @samp{[New @dots{}]} message, the form of the text after
2617 @samp{Switching to} depends on your system's conventions for identifying
2620 @kindex thread apply
2621 @cindex apply command to several threads
2622 @item thread apply [@var{threadno}] [@var{all}] @var{command}
2623 The @code{thread apply} command allows you to apply the named
2624 @var{command} to one or more threads. Specify the numbers of the
2625 threads that you want affected with the command argument
2626 @var{threadno}. It can be a single thread number, one of the numbers
2627 shown in the first field of the @samp{info threads} display; or it
2628 could be a range of thread numbers, as in @code{2-4}. To apply a
2629 command to all threads, type @kbd{thread apply all @var{command}}.
2631 @kindex set print thread-events
2632 @cindex print messages on thread start and exit
2633 @item set print thread-events
2634 @itemx set print thread-events on
2635 @itemx set print thread-events off
2636 The @code{set print thread-events} command allows you to enable or
2637 disable printing of messages when @value{GDBN} notices that new threads have
2638 started or that threads have exited. By default, these messages will
2639 be printed if detection of these events is supported by the target.
2640 Note that these messages cannot be disabled on all targets.
2642 @kindex show print thread-events
2643 @item show print thread-events
2644 Show whether messages will be printed when @value{GDBN} detects that threads
2645 have started and exited.
2648 @xref{Thread Stops,,Stopping and Starting Multi-thread Programs}, for
2649 more information about how @value{GDBN} behaves when you stop and start
2650 programs with multiple threads.
2652 @xref{Set Watchpoints,,Setting Watchpoints}, for information about
2653 watchpoints in programs with multiple threads.
2656 @kindex set libthread-db-search-path
2657 @cindex search path for @code{libthread_db}
2658 @item set libthread-db-search-path @r{[}@var{path}@r{]}
2659 If this variable is set, @var{path} is a colon-separated list of
2660 directories @value{GDBN} will use to search for @code{libthread_db}.
2661 If you omit @var{path}, @samp{libthread-db-search-path} will be reset to
2664 On @sc{gnu}/Linux and Solaris systems, @value{GDBN} uses a ``helper''
2665 @code{libthread_db} library to obtain information about threads in the
2666 inferior process. @value{GDBN} will use @samp{libthread-db-search-path}
2667 to find @code{libthread_db}. If that fails, @value{GDBN} will continue
2668 with default system shared library directories, and finally the directory
2669 from which @code{libpthread} was loaded in the inferior process.
2671 For any @code{libthread_db} library @value{GDBN} finds in above directories,
2672 @value{GDBN} attempts to initialize it with the current inferior process.
2673 If this initialization fails (which could happen because of a version
2674 mismatch between @code{libthread_db} and @code{libpthread}), @value{GDBN}
2675 will unload @code{libthread_db}, and continue with the next directory.
2676 If none of @code{libthread_db} libraries initialize successfully,
2677 @value{GDBN} will issue a warning and thread debugging will be disabled.
2679 Setting @code{libthread-db-search-path} is currently implemented
2680 only on some platforms.
2682 @kindex show libthread-db-search-path
2683 @item show libthread-db-search-path
2684 Display current libthread_db search path.
2688 @section Debugging Programs with Multiple Processes
2690 @cindex fork, debugging programs which call
2691 @cindex multiple processes
2692 @cindex processes, multiple
2693 On most systems, @value{GDBN} has no special support for debugging
2694 programs which create additional processes using the @code{fork}
2695 function. When a program forks, @value{GDBN} will continue to debug the
2696 parent process and the child process will run unimpeded. If you have
2697 set a breakpoint in any code which the child then executes, the child
2698 will get a @code{SIGTRAP} signal which (unless it catches the signal)
2699 will cause it to terminate.
2701 However, if you want to debug the child process there is a workaround
2702 which isn't too painful. Put a call to @code{sleep} in the code which
2703 the child process executes after the fork. It may be useful to sleep
2704 only if a certain environment variable is set, or a certain file exists,
2705 so that the delay need not occur when you don't want to run @value{GDBN}
2706 on the child. While the child is sleeping, use the @code{ps} program to
2707 get its process ID. Then tell @value{GDBN} (a new invocation of
2708 @value{GDBN} if you are also debugging the parent process) to attach to
2709 the child process (@pxref{Attach}). From that point on you can debug
2710 the child process just like any other process which you attached to.
2712 On some systems, @value{GDBN} provides support for debugging programs that
2713 create additional processes using the @code{fork} or @code{vfork} functions.
2714 Currently, the only platforms with this feature are HP-UX (11.x and later
2715 only?) and @sc{gnu}/Linux (kernel version 2.5.60 and later).
2717 By default, when a program forks, @value{GDBN} will continue to debug
2718 the parent process and the child process will run unimpeded.
2720 If you want to follow the child process instead of the parent process,
2721 use the command @w{@code{set follow-fork-mode}}.
2724 @kindex set follow-fork-mode
2725 @item set follow-fork-mode @var{mode}
2726 Set the debugger response to a program call of @code{fork} or
2727 @code{vfork}. A call to @code{fork} or @code{vfork} creates a new
2728 process. The @var{mode} argument can be:
2732 The original process is debugged after a fork. The child process runs
2733 unimpeded. This is the default.
2736 The new process is debugged after a fork. The parent process runs
2741 @kindex show follow-fork-mode
2742 @item show follow-fork-mode
2743 Display the current debugger response to a @code{fork} or @code{vfork} call.
2746 @cindex debugging multiple processes
2747 On Linux, if you want to debug both the parent and child processes, use the
2748 command @w{@code{set detach-on-fork}}.
2751 @kindex set detach-on-fork
2752 @item set detach-on-fork @var{mode}
2753 Tells gdb whether to detach one of the processes after a fork, or
2754 retain debugger control over them both.
2758 The child process (or parent process, depending on the value of
2759 @code{follow-fork-mode}) will be detached and allowed to run
2760 independently. This is the default.
2763 Both processes will be held under the control of @value{GDBN}.
2764 One process (child or parent, depending on the value of
2765 @code{follow-fork-mode}) is debugged as usual, while the other
2770 @kindex show detach-on-fork
2771 @item show detach-on-fork
2772 Show whether detach-on-fork mode is on/off.
2775 If you choose to set @samp{detach-on-fork} mode off, then
2776 @value{GDBN} will retain control of all forked processes (including
2777 nested forks). You can list the forked processes under the control of
2778 @value{GDBN} by using the @w{@code{info forks}} command, and switch
2779 from one fork to another by using the @w{@code{fork}} command.
2784 Print a list of all forked processes under the control of @value{GDBN}.
2785 The listing will include a fork id, a process id, and the current
2786 position (program counter) of the process.
2788 @kindex fork @var{fork-id}
2789 @item fork @var{fork-id}
2790 Make fork number @var{fork-id} the current process. The argument
2791 @var{fork-id} is the internal fork number assigned by @value{GDBN},
2792 as shown in the first field of the @samp{info forks} display.
2794 @kindex process @var{process-id}
2795 @item process @var{process-id}
2796 Make process number @var{process-id} the current process. The
2797 argument @var{process-id} must be one that is listed in the output of
2802 To quit debugging one of the forked processes, you can either detach
2803 from it by using the @w{@code{detach fork}} command (allowing it to
2804 run independently), or delete (and kill) it using the
2805 @w{@code{delete fork}} command.
2808 @kindex detach fork @var{fork-id}
2809 @item detach fork @var{fork-id}
2810 Detach from the process identified by @value{GDBN} fork number
2811 @var{fork-id}, and remove it from the fork list. The process will be
2812 allowed to run independently.
2814 @kindex delete fork @var{fork-id}
2815 @item delete fork @var{fork-id}
2816 Kill the process identified by @value{GDBN} fork number @var{fork-id},
2817 and remove it from the fork list.
2821 If you ask to debug a child process and a @code{vfork} is followed by an
2822 @code{exec}, @value{GDBN} executes the new target up to the first
2823 breakpoint in the new target. If you have a breakpoint set on
2824 @code{main} in your original program, the breakpoint will also be set on
2825 the child process's @code{main}.
2827 When a child process is spawned by @code{vfork}, you cannot debug the
2828 child or parent until an @code{exec} call completes.
2830 If you issue a @code{run} command to @value{GDBN} after an @code{exec}
2831 call executes, the new target restarts. To restart the parent process,
2832 use the @code{file} command with the parent executable name as its
2835 You can use the @code{catch} command to make @value{GDBN} stop whenever
2836 a @code{fork}, @code{vfork}, or @code{exec} call is made. @xref{Set
2837 Catchpoints, ,Setting Catchpoints}.
2839 @node Checkpoint/Restart
2840 @section Setting a @emph{Bookmark} to Return to Later
2845 @cindex snapshot of a process
2846 @cindex rewind program state
2848 On certain operating systems@footnote{Currently, only
2849 @sc{gnu}/Linux.}, @value{GDBN} is able to save a @dfn{snapshot} of a
2850 program's state, called a @dfn{checkpoint}, and come back to it
2853 Returning to a checkpoint effectively undoes everything that has
2854 happened in the program since the @code{checkpoint} was saved. This
2855 includes changes in memory, registers, and even (within some limits)
2856 system state. Effectively, it is like going back in time to the
2857 moment when the checkpoint was saved.
2859 Thus, if you're stepping thru a program and you think you're
2860 getting close to the point where things go wrong, you can save
2861 a checkpoint. Then, if you accidentally go too far and miss
2862 the critical statement, instead of having to restart your program
2863 from the beginning, you can just go back to the checkpoint and
2864 start again from there.
2866 This can be especially useful if it takes a lot of time or
2867 steps to reach the point where you think the bug occurs.
2869 To use the @code{checkpoint}/@code{restart} method of debugging:
2874 Save a snapshot of the debugged program's current execution state.
2875 The @code{checkpoint} command takes no arguments, but each checkpoint
2876 is assigned a small integer id, similar to a breakpoint id.
2878 @kindex info checkpoints
2879 @item info checkpoints
2880 List the checkpoints that have been saved in the current debugging
2881 session. For each checkpoint, the following information will be
2888 @item Source line, or label
2891 @kindex restart @var{checkpoint-id}
2892 @item restart @var{checkpoint-id}
2893 Restore the program state that was saved as checkpoint number
2894 @var{checkpoint-id}. All program variables, registers, stack frames
2895 etc.@: will be returned to the values that they had when the checkpoint
2896 was saved. In essence, gdb will ``wind back the clock'' to the point
2897 in time when the checkpoint was saved.
2899 Note that breakpoints, @value{GDBN} variables, command history etc.
2900 are not affected by restoring a checkpoint. In general, a checkpoint
2901 only restores things that reside in the program being debugged, not in
2904 @kindex delete checkpoint @var{checkpoint-id}
2905 @item delete checkpoint @var{checkpoint-id}
2906 Delete the previously-saved checkpoint identified by @var{checkpoint-id}.
2910 Returning to a previously saved checkpoint will restore the user state
2911 of the program being debugged, plus a significant subset of the system
2912 (OS) state, including file pointers. It won't ``un-write'' data from
2913 a file, but it will rewind the file pointer to the previous location,
2914 so that the previously written data can be overwritten. For files
2915 opened in read mode, the pointer will also be restored so that the
2916 previously read data can be read again.
2918 Of course, characters that have been sent to a printer (or other
2919 external device) cannot be ``snatched back'', and characters received
2920 from eg.@: a serial device can be removed from internal program buffers,
2921 but they cannot be ``pushed back'' into the serial pipeline, ready to
2922 be received again. Similarly, the actual contents of files that have
2923 been changed cannot be restored (at this time).
2925 However, within those constraints, you actually can ``rewind'' your
2926 program to a previously saved point in time, and begin debugging it
2927 again --- and you can change the course of events so as to debug a
2928 different execution path this time.
2930 @cindex checkpoints and process id
2931 Finally, there is one bit of internal program state that will be
2932 different when you return to a checkpoint --- the program's process
2933 id. Each checkpoint will have a unique process id (or @var{pid}),
2934 and each will be different from the program's original @var{pid}.
2935 If your program has saved a local copy of its process id, this could
2936 potentially pose a problem.
2938 @subsection A Non-obvious Benefit of Using Checkpoints
2940 On some systems such as @sc{gnu}/Linux, address space randomization
2941 is performed on new processes for security reasons. This makes it
2942 difficult or impossible to set a breakpoint, or watchpoint, on an
2943 absolute address if you have to restart the program, since the
2944 absolute location of a symbol will change from one execution to the
2947 A checkpoint, however, is an @emph{identical} copy of a process.
2948 Therefore if you create a checkpoint at (eg.@:) the start of main,
2949 and simply return to that checkpoint instead of restarting the
2950 process, you can avoid the effects of address randomization and
2951 your symbols will all stay in the same place.
2954 @chapter Stopping and Continuing
2956 The principal purposes of using a debugger are so that you can stop your
2957 program before it terminates; or so that, if your program runs into
2958 trouble, you can investigate and find out why.
2960 Inside @value{GDBN}, your program may stop for any of several reasons,
2961 such as a signal, a breakpoint, or reaching a new line after a
2962 @value{GDBN} command such as @code{step}. You may then examine and
2963 change variables, set new breakpoints or remove old ones, and then
2964 continue execution. Usually, the messages shown by @value{GDBN} provide
2965 ample explanation of the status of your program---but you can also
2966 explicitly request this information at any time.
2969 @kindex info program
2971 Display information about the status of your program: whether it is
2972 running or not, what process it is, and why it stopped.
2976 * Breakpoints:: Breakpoints, watchpoints, and catchpoints
2977 * Continuing and Stepping:: Resuming execution
2979 * Thread Stops:: Stopping and starting multi-thread programs
2983 @section Breakpoints, Watchpoints, and Catchpoints
2986 A @dfn{breakpoint} makes your program stop whenever a certain point in
2987 the program is reached. For each breakpoint, you can add conditions to
2988 control in finer detail whether your program stops. You can set
2989 breakpoints with the @code{break} command and its variants (@pxref{Set
2990 Breaks, ,Setting Breakpoints}), to specify the place where your program
2991 should stop by line number, function name or exact address in the
2994 On some systems, you can set breakpoints in shared libraries before
2995 the executable is run. There is a minor limitation on HP-UX systems:
2996 you must wait until the executable is run in order to set breakpoints
2997 in shared library routines that are not called directly by the program
2998 (for example, routines that are arguments in a @code{pthread_create}
3002 @cindex data breakpoints
3003 @cindex memory tracing
3004 @cindex breakpoint on memory address
3005 @cindex breakpoint on variable modification
3006 A @dfn{watchpoint} is a special breakpoint that stops your program
3007 when the value of an expression changes. The expression may be a value
3008 of a variable, or it could involve values of one or more variables
3009 combined by operators, such as @samp{a + b}. This is sometimes called
3010 @dfn{data breakpoints}. You must use a different command to set
3011 watchpoints (@pxref{Set Watchpoints, ,Setting Watchpoints}), but aside
3012 from that, you can manage a watchpoint like any other breakpoint: you
3013 enable, disable, and delete both breakpoints and watchpoints using the
3016 You can arrange to have values from your program displayed automatically
3017 whenever @value{GDBN} stops at a breakpoint. @xref{Auto Display,,
3021 @cindex breakpoint on events
3022 A @dfn{catchpoint} is another special breakpoint that stops your program
3023 when a certain kind of event occurs, such as the throwing of a C@t{++}
3024 exception or the loading of a library. As with watchpoints, you use a
3025 different command to set a catchpoint (@pxref{Set Catchpoints, ,Setting
3026 Catchpoints}), but aside from that, you can manage a catchpoint like any
3027 other breakpoint. (To stop when your program receives a signal, use the
3028 @code{handle} command; see @ref{Signals, ,Signals}.)
3030 @cindex breakpoint numbers
3031 @cindex numbers for breakpoints
3032 @value{GDBN} assigns a number to each breakpoint, watchpoint, or
3033 catchpoint when you create it; these numbers are successive integers
3034 starting with one. In many of the commands for controlling various
3035 features of breakpoints you use the breakpoint number to say which
3036 breakpoint you want to change. Each breakpoint may be @dfn{enabled} or
3037 @dfn{disabled}; if disabled, it has no effect on your program until you
3040 @cindex breakpoint ranges
3041 @cindex ranges of breakpoints
3042 Some @value{GDBN} commands accept a range of breakpoints on which to
3043 operate. A breakpoint range is either a single breakpoint number, like
3044 @samp{5}, or two such numbers, in increasing order, separated by a
3045 hyphen, like @samp{5-7}. When a breakpoint range is given to a command,
3046 all breakpoints in that range are operated on.
3049 * Set Breaks:: Setting breakpoints
3050 * Set Watchpoints:: Setting watchpoints
3051 * Set Catchpoints:: Setting catchpoints
3052 * Delete Breaks:: Deleting breakpoints
3053 * Disabling:: Disabling breakpoints
3054 * Conditions:: Break conditions
3055 * Break Commands:: Breakpoint command lists
3056 * Error in Breakpoints:: ``Cannot insert breakpoints''
3057 * Breakpoint-related Warnings:: ``Breakpoint address adjusted...''
3061 @subsection Setting Breakpoints
3063 @c FIXME LMB what does GDB do if no code on line of breakpt?
3064 @c consider in particular declaration with/without initialization.
3066 @c FIXME 2 is there stuff on this already? break at fun start, already init?
3069 @kindex b @r{(@code{break})}
3070 @vindex $bpnum@r{, convenience variable}
3071 @cindex latest breakpoint
3072 Breakpoints are set with the @code{break} command (abbreviated
3073 @code{b}). The debugger convenience variable @samp{$bpnum} records the
3074 number of the breakpoint you've set most recently; see @ref{Convenience
3075 Vars,, Convenience Variables}, for a discussion of what you can do with
3076 convenience variables.
3079 @item break @var{location}
3080 Set a breakpoint at the given @var{location}, which can specify a
3081 function name, a line number, or an address of an instruction.
3082 (@xref{Specify Location}, for a list of all the possible ways to
3083 specify a @var{location}.) The breakpoint will stop your program just
3084 before it executes any of the code in the specified @var{location}.
3086 When using source languages that permit overloading of symbols, such as
3087 C@t{++}, a function name may refer to more than one possible place to break.
3088 @xref{Ambiguous Expressions,,Ambiguous Expressions}, for a discussion of
3091 It is also possible to insert a breakpoint that will stop the program
3092 only if a specific thread (@pxref{Thread-Specific Breakpoints})
3093 or a specific task (@pxref{Ada Tasks}) hits that breakpoint.
3096 When called without any arguments, @code{break} sets a breakpoint at
3097 the next instruction to be executed in the selected stack frame
3098 (@pxref{Stack, ,Examining the Stack}). In any selected frame but the
3099 innermost, this makes your program stop as soon as control
3100 returns to that frame. This is similar to the effect of a
3101 @code{finish} command in the frame inside the selected frame---except
3102 that @code{finish} does not leave an active breakpoint. If you use
3103 @code{break} without an argument in the innermost frame, @value{GDBN} stops
3104 the next time it reaches the current location; this may be useful
3107 @value{GDBN} normally ignores breakpoints when it resumes execution, until at
3108 least one instruction has been executed. If it did not do this, you
3109 would be unable to proceed past a breakpoint without first disabling the
3110 breakpoint. This rule applies whether or not the breakpoint already
3111 existed when your program stopped.
3113 @item break @dots{} if @var{cond}
3114 Set a breakpoint with condition @var{cond}; evaluate the expression
3115 @var{cond} each time the breakpoint is reached, and stop only if the
3116 value is nonzero---that is, if @var{cond} evaluates as true.
3117 @samp{@dots{}} stands for one of the possible arguments described
3118 above (or no argument) specifying where to break. @xref{Conditions,
3119 ,Break Conditions}, for more information on breakpoint conditions.
3122 @item tbreak @var{args}
3123 Set a breakpoint enabled only for one stop. @var{args} are the
3124 same as for the @code{break} command, and the breakpoint is set in the same
3125 way, but the breakpoint is automatically deleted after the first time your
3126 program stops there. @xref{Disabling, ,Disabling Breakpoints}.
3129 @cindex hardware breakpoints
3130 @item hbreak @var{args}
3131 Set a hardware-assisted breakpoint. @var{args} are the same as for the
3132 @code{break} command and the breakpoint is set in the same way, but the
3133 breakpoint requires hardware support and some target hardware may not
3134 have this support. The main purpose of this is EPROM/ROM code
3135 debugging, so you can set a breakpoint at an instruction without
3136 changing the instruction. This can be used with the new trap-generation
3137 provided by SPARClite DSU and most x86-based targets. These targets
3138 will generate traps when a program accesses some data or instruction
3139 address that is assigned to the debug registers. However the hardware
3140 breakpoint registers can take a limited number of breakpoints. For
3141 example, on the DSU, only two data breakpoints can be set at a time, and
3142 @value{GDBN} will reject this command if more than two are used. Delete
3143 or disable unused hardware breakpoints before setting new ones
3144 (@pxref{Disabling, ,Disabling Breakpoints}).
3145 @xref{Conditions, ,Break Conditions}.
3146 For remote targets, you can restrict the number of hardware
3147 breakpoints @value{GDBN} will use, see @ref{set remote
3148 hardware-breakpoint-limit}.
3151 @item thbreak @var{args}
3152 Set a hardware-assisted breakpoint enabled only for one stop. @var{args}
3153 are the same as for the @code{hbreak} command and the breakpoint is set in
3154 the same way. However, like the @code{tbreak} command,
3155 the breakpoint is automatically deleted after the
3156 first time your program stops there. Also, like the @code{hbreak}
3157 command, the breakpoint requires hardware support and some target hardware
3158 may not have this support. @xref{Disabling, ,Disabling Breakpoints}.
3159 See also @ref{Conditions, ,Break Conditions}.
3162 @cindex regular expression
3163 @cindex breakpoints in functions matching a regexp
3164 @cindex set breakpoints in many functions
3165 @item rbreak @var{regex}
3166 Set breakpoints on all functions matching the regular expression
3167 @var{regex}. This command sets an unconditional breakpoint on all
3168 matches, printing a list of all breakpoints it set. Once these
3169 breakpoints are set, they are treated just like the breakpoints set with
3170 the @code{break} command. You can delete them, disable them, or make
3171 them conditional the same way as any other breakpoint.
3173 The syntax of the regular expression is the standard one used with tools
3174 like @file{grep}. Note that this is different from the syntax used by
3175 shells, so for instance @code{foo*} matches all functions that include
3176 an @code{fo} followed by zero or more @code{o}s. There is an implicit
3177 @code{.*} leading and trailing the regular expression you supply, so to
3178 match only functions that begin with @code{foo}, use @code{^foo}.
3180 @cindex non-member C@t{++} functions, set breakpoint in
3181 When debugging C@t{++} programs, @code{rbreak} is useful for setting
3182 breakpoints on overloaded functions that are not members of any special
3185 @cindex set breakpoints on all functions
3186 The @code{rbreak} command can be used to set breakpoints in
3187 @strong{all} the functions in a program, like this:
3190 (@value{GDBP}) rbreak .
3193 @kindex info breakpoints
3194 @cindex @code{$_} and @code{info breakpoints}
3195 @item info breakpoints @r{[}@var{n}@r{]}
3196 @itemx info break @r{[}@var{n}@r{]}
3197 @itemx info watchpoints @r{[}@var{n}@r{]}
3198 Print a table of all breakpoints, watchpoints, and catchpoints set and
3199 not deleted. Optional argument @var{n} means print information only
3200 about the specified breakpoint (or watchpoint or catchpoint). For
3201 each breakpoint, following columns are printed:
3204 @item Breakpoint Numbers
3206 Breakpoint, watchpoint, or catchpoint.
3208 Whether the breakpoint is marked to be disabled or deleted when hit.
3209 @item Enabled or Disabled
3210 Enabled breakpoints are marked with @samp{y}. @samp{n} marks breakpoints
3211 that are not enabled.
3213 Where the breakpoint is in your program, as a memory address. For a
3214 pending breakpoint whose address is not yet known, this field will
3215 contain @samp{<PENDING>}. Such breakpoint won't fire until a shared
3216 library that has the symbol or line referred by breakpoint is loaded.
3217 See below for details. A breakpoint with several locations will
3218 have @samp{<MULTIPLE>} in this field---see below for details.
3220 Where the breakpoint is in the source for your program, as a file and
3221 line number. For a pending breakpoint, the original string passed to
3222 the breakpoint command will be listed as it cannot be resolved until
3223 the appropriate shared library is loaded in the future.
3227 If a breakpoint is conditional, @code{info break} shows the condition on
3228 the line following the affected breakpoint; breakpoint commands, if any,
3229 are listed after that. A pending breakpoint is allowed to have a condition
3230 specified for it. The condition is not parsed for validity until a shared
3231 library is loaded that allows the pending breakpoint to resolve to a
3235 @code{info break} with a breakpoint
3236 number @var{n} as argument lists only that breakpoint. The
3237 convenience variable @code{$_} and the default examining-address for
3238 the @code{x} command are set to the address of the last breakpoint
3239 listed (@pxref{Memory, ,Examining Memory}).
3242 @code{info break} displays a count of the number of times the breakpoint
3243 has been hit. This is especially useful in conjunction with the
3244 @code{ignore} command. You can ignore a large number of breakpoint
3245 hits, look at the breakpoint info to see how many times the breakpoint
3246 was hit, and then run again, ignoring one less than that number. This
3247 will get you quickly to the last hit of that breakpoint.
3250 @value{GDBN} allows you to set any number of breakpoints at the same place in
3251 your program. There is nothing silly or meaningless about this. When
3252 the breakpoints are conditional, this is even useful
3253 (@pxref{Conditions, ,Break Conditions}).
3255 @cindex multiple locations, breakpoints
3256 @cindex breakpoints, multiple locations
3257 It is possible that a breakpoint corresponds to several locations
3258 in your program. Examples of this situation are:
3262 For a C@t{++} constructor, the @value{NGCC} compiler generates several
3263 instances of the function body, used in different cases.
3266 For a C@t{++} template function, a given line in the function can
3267 correspond to any number of instantiations.
3270 For an inlined function, a given source line can correspond to
3271 several places where that function is inlined.
3274 In all those cases, @value{GDBN} will insert a breakpoint at all
3275 the relevant locations@footnote{
3276 As of this writing, multiple-location breakpoints work only if there's
3277 line number information for all the locations. This means that they
3278 will generally not work in system libraries, unless you have debug
3279 info with line numbers for them.}.
3281 A breakpoint with multiple locations is displayed in the breakpoint
3282 table using several rows---one header row, followed by one row for
3283 each breakpoint location. The header row has @samp{<MULTIPLE>} in the
3284 address column. The rows for individual locations contain the actual
3285 addresses for locations, and show the functions to which those
3286 locations belong. The number column for a location is of the form
3287 @var{breakpoint-number}.@var{location-number}.
3292 Num Type Disp Enb Address What
3293 1 breakpoint keep y <MULTIPLE>
3295 breakpoint already hit 1 time
3296 1.1 y 0x080486a2 in void foo<int>() at t.cc:8
3297 1.2 y 0x080486ca in void foo<double>() at t.cc:8
3300 Each location can be individually enabled or disabled by passing
3301 @var{breakpoint-number}.@var{location-number} as argument to the
3302 @code{enable} and @code{disable} commands. Note that you cannot
3303 delete the individual locations from the list, you can only delete the
3304 entire list of locations that belong to their parent breakpoint (with
3305 the @kbd{delete @var{num}} command, where @var{num} is the number of
3306 the parent breakpoint, 1 in the above example). Disabling or enabling
3307 the parent breakpoint (@pxref{Disabling}) affects all of the locations
3308 that belong to that breakpoint.
3310 @cindex pending breakpoints
3311 It's quite common to have a breakpoint inside a shared library.
3312 Shared libraries can be loaded and unloaded explicitly,
3313 and possibly repeatedly, as the program is executed. To support
3314 this use case, @value{GDBN} updates breakpoint locations whenever
3315 any shared library is loaded or unloaded. Typically, you would
3316 set a breakpoint in a shared library at the beginning of your
3317 debugging session, when the library is not loaded, and when the
3318 symbols from the library are not available. When you try to set
3319 breakpoint, @value{GDBN} will ask you if you want to set
3320 a so called @dfn{pending breakpoint}---breakpoint whose address
3321 is not yet resolved.
3323 After the program is run, whenever a new shared library is loaded,
3324 @value{GDBN} reevaluates all the breakpoints. When a newly loaded
3325 shared library contains the symbol or line referred to by some
3326 pending breakpoint, that breakpoint is resolved and becomes an
3327 ordinary breakpoint. When a library is unloaded, all breakpoints
3328 that refer to its symbols or source lines become pending again.
3330 This logic works for breakpoints with multiple locations, too. For
3331 example, if you have a breakpoint in a C@t{++} template function, and
3332 a newly loaded shared library has an instantiation of that template,
3333 a new location is added to the list of locations for the breakpoint.
3335 Except for having unresolved address, pending breakpoints do not
3336 differ from regular breakpoints. You can set conditions or commands,
3337 enable and disable them and perform other breakpoint operations.
3339 @value{GDBN} provides some additional commands for controlling what
3340 happens when the @samp{break} command cannot resolve breakpoint
3341 address specification to an address:
3343 @kindex set breakpoint pending
3344 @kindex show breakpoint pending
3346 @item set breakpoint pending auto
3347 This is the default behavior. When @value{GDBN} cannot find the breakpoint
3348 location, it queries you whether a pending breakpoint should be created.
3350 @item set breakpoint pending on
3351 This indicates that an unrecognized breakpoint location should automatically
3352 result in a pending breakpoint being created.
3354 @item set breakpoint pending off
3355 This indicates that pending breakpoints are not to be created. Any
3356 unrecognized breakpoint location results in an error. This setting does
3357 not affect any pending breakpoints previously created.
3359 @item show breakpoint pending
3360 Show the current behavior setting for creating pending breakpoints.
3363 The settings above only affect the @code{break} command and its
3364 variants. Once breakpoint is set, it will be automatically updated
3365 as shared libraries are loaded and unloaded.
3367 @cindex automatic hardware breakpoints
3368 For some targets, @value{GDBN} can automatically decide if hardware or
3369 software breakpoints should be used, depending on whether the
3370 breakpoint address is read-only or read-write. This applies to
3371 breakpoints set with the @code{break} command as well as to internal
3372 breakpoints set by commands like @code{next} and @code{finish}. For
3373 breakpoints set with @code{hbreak}, @value{GDBN} will always use hardware
3376 You can control this automatic behaviour with the following commands::
3378 @kindex set breakpoint auto-hw
3379 @kindex show breakpoint auto-hw
3381 @item set breakpoint auto-hw on
3382 This is the default behavior. When @value{GDBN} sets a breakpoint, it
3383 will try to use the target memory map to decide if software or hardware
3384 breakpoint must be used.
3386 @item set breakpoint auto-hw off
3387 This indicates @value{GDBN} should not automatically select breakpoint
3388 type. If the target provides a memory map, @value{GDBN} will warn when
3389 trying to set software breakpoint at a read-only address.
3392 @value{GDBN} normally implements breakpoints by replacing the program code
3393 at the breakpoint address with a special instruction, which, when
3394 executed, given control to the debugger. By default, the program
3395 code is so modified only when the program is resumed. As soon as
3396 the program stops, @value{GDBN} restores the original instructions. This
3397 behaviour guards against leaving breakpoints inserted in the
3398 target should gdb abrubptly disconnect. However, with slow remote
3399 targets, inserting and removing breakpoint can reduce the performance.
3400 This behavior can be controlled with the following commands::
3402 @kindex set breakpoint always-inserted
3403 @kindex show breakpoint always-inserted
3405 @item set breakpoint always-inserted off
3406 All breakpoints, including newly added by the user, are inserted in
3407 the target only when the target is resumed. All breakpoints are
3408 removed from the target when it stops.
3410 @item set breakpoint always-inserted on
3411 Causes all breakpoints to be inserted in the target at all times. If
3412 the user adds a new breakpoint, or changes an existing breakpoint, the
3413 breakpoints in the target are updated immediately. A breakpoint is
3414 removed from the target only when breakpoint itself is removed.
3416 @cindex non-stop mode, and @code{breakpoint always-inserted}
3417 @item set breakpoint always-inserted auto
3418 This is the default mode. If @value{GDBN} is controlling the inferior
3419 in non-stop mode (@pxref{Non-Stop Mode}), gdb behaves as if
3420 @code{breakpoint always-inserted} mode is on. If @value{GDBN} is
3421 controlling the inferior in all-stop mode, @value{GDBN} behaves as if
3422 @code{breakpoint always-inserted} mode is off.
3425 @cindex negative breakpoint numbers
3426 @cindex internal @value{GDBN} breakpoints
3427 @value{GDBN} itself sometimes sets breakpoints in your program for
3428 special purposes, such as proper handling of @code{longjmp} (in C
3429 programs). These internal breakpoints are assigned negative numbers,
3430 starting with @code{-1}; @samp{info breakpoints} does not display them.
3431 You can see these breakpoints with the @value{GDBN} maintenance command
3432 @samp{maint info breakpoints} (@pxref{maint info breakpoints}).
3435 @node Set Watchpoints
3436 @subsection Setting Watchpoints
3438 @cindex setting watchpoints
3439 You can use a watchpoint to stop execution whenever the value of an
3440 expression changes, without having to predict a particular place where
3441 this may happen. (This is sometimes called a @dfn{data breakpoint}.)
3442 The expression may be as simple as the value of a single variable, or
3443 as complex as many variables combined by operators. Examples include:
3447 A reference to the value of a single variable.
3450 An address cast to an appropriate data type. For example,
3451 @samp{*(int *)0x12345678} will watch a 4-byte region at the specified
3452 address (assuming an @code{int} occupies 4 bytes).
3455 An arbitrarily complex expression, such as @samp{a*b + c/d}. The
3456 expression can use any operators valid in the program's native
3457 language (@pxref{Languages}).
3460 You can set a watchpoint on an expression even if the expression can
3461 not be evaluated yet. For instance, you can set a watchpoint on
3462 @samp{*global_ptr} before @samp{global_ptr} is initialized.
3463 @value{GDBN} will stop when your program sets @samp{global_ptr} and
3464 the expression produces a valid value. If the expression becomes
3465 valid in some other way than changing a variable (e.g.@: if the memory
3466 pointed to by @samp{*global_ptr} becomes readable as the result of a
3467 @code{malloc} call), @value{GDBN} may not stop until the next time
3468 the expression changes.
3470 @cindex software watchpoints
3471 @cindex hardware watchpoints
3472 Depending on your system, watchpoints may be implemented in software or
3473 hardware. @value{GDBN} does software watchpointing by single-stepping your
3474 program and testing the variable's value each time, which is hundreds of
3475 times slower than normal execution. (But this may still be worth it, to
3476 catch errors where you have no clue what part of your program is the
3479 On some systems, such as HP-UX, PowerPC, @sc{gnu}/Linux and most other
3480 x86-based targets, @value{GDBN} includes support for hardware
3481 watchpoints, which do not slow down the running of your program.
3485 @item watch @var{expr} @r{[}thread @var{threadnum}@r{]}
3486 Set a watchpoint for an expression. @value{GDBN} will break when the
3487 expression @var{expr} is written into by the program and its value
3488 changes. The simplest (and the most popular) use of this command is
3489 to watch the value of a single variable:
3492 (@value{GDBP}) watch foo
3495 If the command includes a @code{@r{[}thread @var{threadnum}@r{]}}
3496 clause, @value{GDBN} breaks only when the thread identified by
3497 @var{threadnum} changes the value of @var{expr}. If any other threads
3498 change the value of @var{expr}, @value{GDBN} will not break. Note
3499 that watchpoints restricted to a single thread in this way only work
3500 with Hardware Watchpoints.
3503 @item rwatch @var{expr} @r{[}thread @var{threadnum}@r{]}
3504 Set a watchpoint that will break when the value of @var{expr} is read
3508 @item awatch @var{expr} @r{[}thread @var{threadnum}@r{]}
3509 Set a watchpoint that will break when @var{expr} is either read from
3510 or written into by the program.
3512 @kindex info watchpoints @r{[}@var{n}@r{]}
3513 @item info watchpoints
3514 This command prints a list of watchpoints, breakpoints, and catchpoints;
3515 it is the same as @code{info break} (@pxref{Set Breaks}).
3518 @value{GDBN} sets a @dfn{hardware watchpoint} if possible. Hardware
3519 watchpoints execute very quickly, and the debugger reports a change in
3520 value at the exact instruction where the change occurs. If @value{GDBN}
3521 cannot set a hardware watchpoint, it sets a software watchpoint, which
3522 executes more slowly and reports the change in value at the next
3523 @emph{statement}, not the instruction, after the change occurs.
3525 @cindex use only software watchpoints
3526 You can force @value{GDBN} to use only software watchpoints with the
3527 @kbd{set can-use-hw-watchpoints 0} command. With this variable set to
3528 zero, @value{GDBN} will never try to use hardware watchpoints, even if
3529 the underlying system supports them. (Note that hardware-assisted
3530 watchpoints that were set @emph{before} setting
3531 @code{can-use-hw-watchpoints} to zero will still use the hardware
3532 mechanism of watching expression values.)
3535 @item set can-use-hw-watchpoints
3536 @kindex set can-use-hw-watchpoints
3537 Set whether or not to use hardware watchpoints.
3539 @item show can-use-hw-watchpoints
3540 @kindex show can-use-hw-watchpoints
3541 Show the current mode of using hardware watchpoints.
3544 For remote targets, you can restrict the number of hardware
3545 watchpoints @value{GDBN} will use, see @ref{set remote
3546 hardware-breakpoint-limit}.
3548 When you issue the @code{watch} command, @value{GDBN} reports
3551 Hardware watchpoint @var{num}: @var{expr}
3555 if it was able to set a hardware watchpoint.
3557 Currently, the @code{awatch} and @code{rwatch} commands can only set
3558 hardware watchpoints, because accesses to data that don't change the
3559 value of the watched expression cannot be detected without examining
3560 every instruction as it is being executed, and @value{GDBN} does not do
3561 that currently. If @value{GDBN} finds that it is unable to set a
3562 hardware breakpoint with the @code{awatch} or @code{rwatch} command, it
3563 will print a message like this:
3566 Expression cannot be implemented with read/access watchpoint.
3569 Sometimes, @value{GDBN} cannot set a hardware watchpoint because the
3570 data type of the watched expression is wider than what a hardware
3571 watchpoint on the target machine can handle. For example, some systems
3572 can only watch regions that are up to 4 bytes wide; on such systems you
3573 cannot set hardware watchpoints for an expression that yields a
3574 double-precision floating-point number (which is typically 8 bytes
3575 wide). As a work-around, it might be possible to break the large region
3576 into a series of smaller ones and watch them with separate watchpoints.
3578 If you set too many hardware watchpoints, @value{GDBN} might be unable
3579 to insert all of them when you resume the execution of your program.
3580 Since the precise number of active watchpoints is unknown until such
3581 time as the program is about to be resumed, @value{GDBN} might not be
3582 able to warn you about this when you set the watchpoints, and the
3583 warning will be printed only when the program is resumed:
3586 Hardware watchpoint @var{num}: Could not insert watchpoint
3590 If this happens, delete or disable some of the watchpoints.
3592 Watching complex expressions that reference many variables can also
3593 exhaust the resources available for hardware-assisted watchpoints.
3594 That's because @value{GDBN} needs to watch every variable in the
3595 expression with separately allocated resources.
3597 If you call a function interactively using @code{print} or @code{call},
3598 any watchpoints you have set will be inactive until @value{GDBN} reaches another
3599 kind of breakpoint or the call completes.
3601 @value{GDBN} automatically deletes watchpoints that watch local
3602 (automatic) variables, or expressions that involve such variables, when
3603 they go out of scope, that is, when the execution leaves the block in
3604 which these variables were defined. In particular, when the program
3605 being debugged terminates, @emph{all} local variables go out of scope,
3606 and so only watchpoints that watch global variables remain set. If you
3607 rerun the program, you will need to set all such watchpoints again. One
3608 way of doing that would be to set a code breakpoint at the entry to the
3609 @code{main} function and when it breaks, set all the watchpoints.
3611 @cindex watchpoints and threads
3612 @cindex threads and watchpoints
3613 In multi-threaded programs, watchpoints will detect changes to the
3614 watched expression from every thread.
3617 @emph{Warning:} In multi-threaded programs, software watchpoints
3618 have only limited usefulness. If @value{GDBN} creates a software
3619 watchpoint, it can only watch the value of an expression @emph{in a
3620 single thread}. If you are confident that the expression can only
3621 change due to the current thread's activity (and if you are also
3622 confident that no other thread can become current), then you can use
3623 software watchpoints as usual. However, @value{GDBN} may not notice
3624 when a non-current thread's activity changes the expression. (Hardware
3625 watchpoints, in contrast, watch an expression in all threads.)
3628 @xref{set remote hardware-watchpoint-limit}.
3630 @node Set Catchpoints
3631 @subsection Setting Catchpoints
3632 @cindex catchpoints, setting
3633 @cindex exception handlers
3634 @cindex event handling
3636 You can use @dfn{catchpoints} to cause the debugger to stop for certain
3637 kinds of program events, such as C@t{++} exceptions or the loading of a
3638 shared library. Use the @code{catch} command to set a catchpoint.
3642 @item catch @var{event}
3643 Stop when @var{event} occurs. @var{event} can be any of the following:
3646 @cindex stop on C@t{++} exceptions
3647 The throwing of a C@t{++} exception.
3650 The catching of a C@t{++} exception.
3653 @cindex Ada exception catching
3654 @cindex catch Ada exceptions
3655 An Ada exception being raised. If an exception name is specified
3656 at the end of the command (eg @code{catch exception Program_Error}),
3657 the debugger will stop only when this specific exception is raised.
3658 Otherwise, the debugger stops execution when any Ada exception is raised.
3660 When inserting an exception catchpoint on a user-defined exception whose
3661 name is identical to one of the exceptions defined by the language, the
3662 fully qualified name must be used as the exception name. Otherwise,
3663 @value{GDBN} will assume that it should stop on the pre-defined exception
3664 rather than the user-defined one. For instance, assuming an exception
3665 called @code{Constraint_Error} is defined in package @code{Pck}, then
3666 the command to use to catch such exceptions is @kbd{catch exception
3667 Pck.Constraint_Error}.
3669 @item exception unhandled
3670 An exception that was raised but is not handled by the program.
3673 A failed Ada assertion.
3676 @cindex break on fork/exec
3677 A call to @code{exec}. This is currently only available for HP-UX
3681 A call to @code{fork}. This is currently only available for HP-UX
3685 A call to @code{vfork}. This is currently only available for HP-UX
3690 @item tcatch @var{event}
3691 Set a catchpoint that is enabled only for one stop. The catchpoint is
3692 automatically deleted after the first time the event is caught.
3696 Use the @code{info break} command to list the current catchpoints.
3698 There are currently some limitations to C@t{++} exception handling
3699 (@code{catch throw} and @code{catch catch}) in @value{GDBN}:
3703 If you call a function interactively, @value{GDBN} normally returns
3704 control to you when the function has finished executing. If the call
3705 raises an exception, however, the call may bypass the mechanism that
3706 returns control to you and cause your program either to abort or to
3707 simply continue running until it hits a breakpoint, catches a signal
3708 that @value{GDBN} is listening for, or exits. This is the case even if
3709 you set a catchpoint for the exception; catchpoints on exceptions are
3710 disabled within interactive calls.
3713 You cannot raise an exception interactively.
3716 You cannot install an exception handler interactively.
3719 @cindex raise exceptions
3720 Sometimes @code{catch} is not the best way to debug exception handling:
3721 if you need to know exactly where an exception is raised, it is better to
3722 stop @emph{before} the exception handler is called, since that way you
3723 can see the stack before any unwinding takes place. If you set a
3724 breakpoint in an exception handler instead, it may not be easy to find
3725 out where the exception was raised.
3727 To stop just before an exception handler is called, you need some
3728 knowledge of the implementation. In the case of @sc{gnu} C@t{++}, exceptions are
3729 raised by calling a library function named @code{__raise_exception}
3730 which has the following ANSI C interface:
3733 /* @var{addr} is where the exception identifier is stored.
3734 @var{id} is the exception identifier. */
3735 void __raise_exception (void **addr, void *id);
3739 To make the debugger catch all exceptions before any stack
3740 unwinding takes place, set a breakpoint on @code{__raise_exception}
3741 (@pxref{Breakpoints, ,Breakpoints; Watchpoints; and Exceptions}).
3743 With a conditional breakpoint (@pxref{Conditions, ,Break Conditions})
3744 that depends on the value of @var{id}, you can stop your program when
3745 a specific exception is raised. You can use multiple conditional
3746 breakpoints to stop your program when any of a number of exceptions are
3751 @subsection Deleting Breakpoints
3753 @cindex clearing breakpoints, watchpoints, catchpoints
3754 @cindex deleting breakpoints, watchpoints, catchpoints
3755 It is often necessary to eliminate a breakpoint, watchpoint, or
3756 catchpoint once it has done its job and you no longer want your program
3757 to stop there. This is called @dfn{deleting} the breakpoint. A
3758 breakpoint that has been deleted no longer exists; it is forgotten.
3760 With the @code{clear} command you can delete breakpoints according to
3761 where they are in your program. With the @code{delete} command you can
3762 delete individual breakpoints, watchpoints, or catchpoints by specifying
3763 their breakpoint numbers.
3765 It is not necessary to delete a breakpoint to proceed past it. @value{GDBN}
3766 automatically ignores breakpoints on the first instruction to be executed
3767 when you continue execution without changing the execution address.
3772 Delete any breakpoints at the next instruction to be executed in the
3773 selected stack frame (@pxref{Selection, ,Selecting a Frame}). When
3774 the innermost frame is selected, this is a good way to delete a
3775 breakpoint where your program just stopped.
3777 @item clear @var{location}
3778 Delete any breakpoints set at the specified @var{location}.
3779 @xref{Specify Location}, for the various forms of @var{location}; the
3780 most useful ones are listed below:
3783 @item clear @var{function}
3784 @itemx clear @var{filename}:@var{function}
3785 Delete any breakpoints set at entry to the named @var{function}.
3787 @item clear @var{linenum}
3788 @itemx clear @var{filename}:@var{linenum}
3789 Delete any breakpoints set at or within the code of the specified
3790 @var{linenum} of the specified @var{filename}.
3793 @cindex delete breakpoints
3795 @kindex d @r{(@code{delete})}
3796 @item delete @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
3797 Delete the breakpoints, watchpoints, or catchpoints of the breakpoint
3798 ranges specified as arguments. If no argument is specified, delete all
3799 breakpoints (@value{GDBN} asks confirmation, unless you have @code{set
3800 confirm off}). You can abbreviate this command as @code{d}.
3804 @subsection Disabling Breakpoints
3806 @cindex enable/disable a breakpoint
3807 Rather than deleting a breakpoint, watchpoint, or catchpoint, you might
3808 prefer to @dfn{disable} it. This makes the breakpoint inoperative as if
3809 it had been deleted, but remembers the information on the breakpoint so
3810 that you can @dfn{enable} it again later.
3812 You disable and enable breakpoints, watchpoints, and catchpoints with
3813 the @code{enable} and @code{disable} commands, optionally specifying one
3814 or more breakpoint numbers as arguments. Use @code{info break} or
3815 @code{info watch} to print a list of breakpoints, watchpoints, and
3816 catchpoints if you do not know which numbers to use.
3818 Disabling and enabling a breakpoint that has multiple locations
3819 affects all of its locations.
3821 A breakpoint, watchpoint, or catchpoint can have any of four different
3822 states of enablement:
3826 Enabled. The breakpoint stops your program. A breakpoint set
3827 with the @code{break} command starts out in this state.
3829 Disabled. The breakpoint has no effect on your program.
3831 Enabled once. The breakpoint stops your program, but then becomes
3834 Enabled for deletion. The breakpoint stops your program, but
3835 immediately after it does so it is deleted permanently. A breakpoint
3836 set with the @code{tbreak} command starts out in this state.
3839 You can use the following commands to enable or disable breakpoints,
3840 watchpoints, and catchpoints:
3844 @kindex dis @r{(@code{disable})}
3845 @item disable @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
3846 Disable the specified breakpoints---or all breakpoints, if none are
3847 listed. A disabled breakpoint has no effect but is not forgotten. All
3848 options such as ignore-counts, conditions and commands are remembered in
3849 case the breakpoint is enabled again later. You may abbreviate
3850 @code{disable} as @code{dis}.
3853 @item enable @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
3854 Enable the specified breakpoints (or all defined breakpoints). They
3855 become effective once again in stopping your program.
3857 @item enable @r{[}breakpoints@r{]} once @var{range}@dots{}
3858 Enable the specified breakpoints temporarily. @value{GDBN} disables any
3859 of these breakpoints immediately after stopping your program.
3861 @item enable @r{[}breakpoints@r{]} delete @var{range}@dots{}
3862 Enable the specified breakpoints to work once, then die. @value{GDBN}
3863 deletes any of these breakpoints as soon as your program stops there.
3864 Breakpoints set by the @code{tbreak} command start out in this state.
3867 @c FIXME: I think the following ``Except for [...] @code{tbreak}'' is
3868 @c confusing: tbreak is also initially enabled.
3869 Except for a breakpoint set with @code{tbreak} (@pxref{Set Breaks,
3870 ,Setting Breakpoints}), breakpoints that you set are initially enabled;
3871 subsequently, they become disabled or enabled only when you use one of
3872 the commands above. (The command @code{until} can set and delete a
3873 breakpoint of its own, but it does not change the state of your other
3874 breakpoints; see @ref{Continuing and Stepping, ,Continuing and
3878 @subsection Break Conditions
3879 @cindex conditional breakpoints
3880 @cindex breakpoint conditions
3882 @c FIXME what is scope of break condition expr? Context where wanted?
3883 @c in particular for a watchpoint?
3884 The simplest sort of breakpoint breaks every time your program reaches a
3885 specified place. You can also specify a @dfn{condition} for a
3886 breakpoint. A condition is just a Boolean expression in your
3887 programming language (@pxref{Expressions, ,Expressions}). A breakpoint with
3888 a condition evaluates the expression each time your program reaches it,
3889 and your program stops only if the condition is @emph{true}.
3891 This is the converse of using assertions for program validation; in that
3892 situation, you want to stop when the assertion is violated---that is,
3893 when the condition is false. In C, if you want to test an assertion expressed
3894 by the condition @var{assert}, you should set the condition
3895 @samp{! @var{assert}} on the appropriate breakpoint.
3897 Conditions are also accepted for watchpoints; you may not need them,
3898 since a watchpoint is inspecting the value of an expression anyhow---but
3899 it might be simpler, say, to just set a watchpoint on a variable name,
3900 and specify a condition that tests whether the new value is an interesting
3903 Break conditions can have side effects, and may even call functions in
3904 your program. This can be useful, for example, to activate functions
3905 that log program progress, or to use your own print functions to
3906 format special data structures. The effects are completely predictable
3907 unless there is another enabled breakpoint at the same address. (In
3908 that case, @value{GDBN} might see the other breakpoint first and stop your
3909 program without checking the condition of this one.) Note that
3910 breakpoint commands are usually more convenient and flexible than break
3912 purpose of performing side effects when a breakpoint is reached
3913 (@pxref{Break Commands, ,Breakpoint Command Lists}).
3915 Break conditions can be specified when a breakpoint is set, by using
3916 @samp{if} in the arguments to the @code{break} command. @xref{Set
3917 Breaks, ,Setting Breakpoints}. They can also be changed at any time
3918 with the @code{condition} command.
3920 You can also use the @code{if} keyword with the @code{watch} command.
3921 The @code{catch} command does not recognize the @code{if} keyword;
3922 @code{condition} is the only way to impose a further condition on a
3927 @item condition @var{bnum} @var{expression}
3928 Specify @var{expression} as the break condition for breakpoint,
3929 watchpoint, or catchpoint number @var{bnum}. After you set a condition,
3930 breakpoint @var{bnum} stops your program only if the value of
3931 @var{expression} is true (nonzero, in C). When you use
3932 @code{condition}, @value{GDBN} checks @var{expression} immediately for
3933 syntactic correctness, and to determine whether symbols in it have
3934 referents in the context of your breakpoint. If @var{expression} uses
3935 symbols not referenced in the context of the breakpoint, @value{GDBN}
3936 prints an error message:
3939 No symbol "foo" in current context.
3944 not actually evaluate @var{expression} at the time the @code{condition}
3945 command (or a command that sets a breakpoint with a condition, like
3946 @code{break if @dots{}}) is given, however. @xref{Expressions, ,Expressions}.
3948 @item condition @var{bnum}
3949 Remove the condition from breakpoint number @var{bnum}. It becomes
3950 an ordinary unconditional breakpoint.
3953 @cindex ignore count (of breakpoint)
3954 A special case of a breakpoint condition is to stop only when the
3955 breakpoint has been reached a certain number of times. This is so
3956 useful that there is a special way to do it, using the @dfn{ignore
3957 count} of the breakpoint. Every breakpoint has an ignore count, which
3958 is an integer. Most of the time, the ignore count is zero, and
3959 therefore has no effect. But if your program reaches a breakpoint whose
3960 ignore count is positive, then instead of stopping, it just decrements
3961 the ignore count by one and continues. As a result, if the ignore count
3962 value is @var{n}, the breakpoint does not stop the next @var{n} times
3963 your program reaches it.
3967 @item ignore @var{bnum} @var{count}
3968 Set the ignore count of breakpoint number @var{bnum} to @var{count}.
3969 The next @var{count} times the breakpoint is reached, your program's
3970 execution does not stop; other than to decrement the ignore count, @value{GDBN}
3973 To make the breakpoint stop the next time it is reached, specify
3976 When you use @code{continue} to resume execution of your program from a
3977 breakpoint, you can specify an ignore count directly as an argument to
3978 @code{continue}, rather than using @code{ignore}. @xref{Continuing and
3979 Stepping,,Continuing and Stepping}.
3981 If a breakpoint has a positive ignore count and a condition, the
3982 condition is not checked. Once the ignore count reaches zero,
3983 @value{GDBN} resumes checking the condition.
3985 You could achieve the effect of the ignore count with a condition such
3986 as @w{@samp{$foo-- <= 0}} using a debugger convenience variable that
3987 is decremented each time. @xref{Convenience Vars, ,Convenience
3991 Ignore counts apply to breakpoints, watchpoints, and catchpoints.
3994 @node Break Commands
3995 @subsection Breakpoint Command Lists
3997 @cindex breakpoint commands
3998 You can give any breakpoint (or watchpoint or catchpoint) a series of
3999 commands to execute when your program stops due to that breakpoint. For
4000 example, you might want to print the values of certain expressions, or
4001 enable other breakpoints.
4005 @kindex end@r{ (breakpoint commands)}
4006 @item commands @r{[}@var{bnum}@r{]}
4007 @itemx @dots{} @var{command-list} @dots{}
4009 Specify a list of commands for breakpoint number @var{bnum}. The commands
4010 themselves appear on the following lines. Type a line containing just
4011 @code{end} to terminate the commands.
4013 To remove all commands from a breakpoint, type @code{commands} and
4014 follow it immediately with @code{end}; that is, give no commands.
4016 With no @var{bnum} argument, @code{commands} refers to the last
4017 breakpoint, watchpoint, or catchpoint set (not to the breakpoint most
4018 recently encountered).
4021 Pressing @key{RET} as a means of repeating the last @value{GDBN} command is
4022 disabled within a @var{command-list}.
4024 You can use breakpoint commands to start your program up again. Simply
4025 use the @code{continue} command, or @code{step}, or any other command
4026 that resumes execution.
4028 Any other commands in the command list, after a command that resumes
4029 execution, are ignored. This is because any time you resume execution
4030 (even with a simple @code{next} or @code{step}), you may encounter
4031 another breakpoint---which could have its own command list, leading to
4032 ambiguities about which list to execute.
4035 If the first command you specify in a command list is @code{silent}, the
4036 usual message about stopping at a breakpoint is not printed. This may
4037 be desirable for breakpoints that are to print a specific message and
4038 then continue. If none of the remaining commands print anything, you
4039 see no sign that the breakpoint was reached. @code{silent} is
4040 meaningful only at the beginning of a breakpoint command list.
4042 The commands @code{echo}, @code{output}, and @code{printf} allow you to
4043 print precisely controlled output, and are often useful in silent
4044 breakpoints. @xref{Output, ,Commands for Controlled Output}.
4046 For example, here is how you could use breakpoint commands to print the
4047 value of @code{x} at entry to @code{foo} whenever @code{x} is positive.
4053 printf "x is %d\n",x
4058 One application for breakpoint commands is to compensate for one bug so
4059 you can test for another. Put a breakpoint just after the erroneous line
4060 of code, give it a condition to detect the case in which something
4061 erroneous has been done, and give it commands to assign correct values
4062 to any variables that need them. End with the @code{continue} command
4063 so that your program does not stop, and start with the @code{silent}
4064 command so that no output is produced. Here is an example:
4075 @c @ifclear BARETARGET
4076 @node Error in Breakpoints
4077 @subsection ``Cannot insert breakpoints''
4079 If you request too many active hardware-assisted breakpoints and
4080 watchpoints, you will see this error message:
4082 @c FIXME: the precise wording of this message may change; the relevant
4083 @c source change is not committed yet (Sep 3, 1999).
4085 Stopped; cannot insert breakpoints.
4086 You may have requested too many hardware breakpoints and watchpoints.
4090 This message is printed when you attempt to resume the program, since
4091 only then @value{GDBN} knows exactly how many hardware breakpoints and
4092 watchpoints it needs to insert.
4094 When this message is printed, you need to disable or remove some of the
4095 hardware-assisted breakpoints and watchpoints, and then continue.
4097 @node Breakpoint-related Warnings
4098 @subsection ``Breakpoint address adjusted...''
4099 @cindex breakpoint address adjusted
4101 Some processor architectures place constraints on the addresses at
4102 which breakpoints may be placed. For architectures thus constrained,
4103 @value{GDBN} will attempt to adjust the breakpoint's address to comply
4104 with the constraints dictated by the architecture.
4106 One example of such an architecture is the Fujitsu FR-V. The FR-V is
4107 a VLIW architecture in which a number of RISC-like instructions may be
4108 bundled together for parallel execution. The FR-V architecture
4109 constrains the location of a breakpoint instruction within such a
4110 bundle to the instruction with the lowest address. @value{GDBN}
4111 honors this constraint by adjusting a breakpoint's address to the
4112 first in the bundle.
4114 It is not uncommon for optimized code to have bundles which contain
4115 instructions from different source statements, thus it may happen that
4116 a breakpoint's address will be adjusted from one source statement to
4117 another. Since this adjustment may significantly alter @value{GDBN}'s
4118 breakpoint related behavior from what the user expects, a warning is
4119 printed when the breakpoint is first set and also when the breakpoint
4122 A warning like the one below is printed when setting a breakpoint
4123 that's been subject to address adjustment:
4126 warning: Breakpoint address adjusted from 0x00010414 to 0x00010410.
4129 Such warnings are printed both for user settable and @value{GDBN}'s
4130 internal breakpoints. If you see one of these warnings, you should
4131 verify that a breakpoint set at the adjusted address will have the
4132 desired affect. If not, the breakpoint in question may be removed and
4133 other breakpoints may be set which will have the desired behavior.
4134 E.g., it may be sufficient to place the breakpoint at a later
4135 instruction. A conditional breakpoint may also be useful in some
4136 cases to prevent the breakpoint from triggering too often.
4138 @value{GDBN} will also issue a warning when stopping at one of these
4139 adjusted breakpoints:
4142 warning: Breakpoint 1 address previously adjusted from 0x00010414
4146 When this warning is encountered, it may be too late to take remedial
4147 action except in cases where the breakpoint is hit earlier or more
4148 frequently than expected.
4150 @node Continuing and Stepping
4151 @section Continuing and Stepping
4155 @cindex resuming execution
4156 @dfn{Continuing} means resuming program execution until your program
4157 completes normally. In contrast, @dfn{stepping} means executing just
4158 one more ``step'' of your program, where ``step'' may mean either one
4159 line of source code, or one machine instruction (depending on what
4160 particular command you use). Either when continuing or when stepping,
4161 your program may stop even sooner, due to a breakpoint or a signal. (If
4162 it stops due to a signal, you may want to use @code{handle}, or use
4163 @samp{signal 0} to resume execution. @xref{Signals, ,Signals}.)
4167 @kindex c @r{(@code{continue})}
4168 @kindex fg @r{(resume foreground execution)}
4169 @item continue @r{[}@var{ignore-count}@r{]}
4170 @itemx c @r{[}@var{ignore-count}@r{]}
4171 @itemx fg @r{[}@var{ignore-count}@r{]}
4172 Resume program execution, at the address where your program last stopped;
4173 any breakpoints set at that address are bypassed. The optional argument
4174 @var{ignore-count} allows you to specify a further number of times to
4175 ignore a breakpoint at this location; its effect is like that of
4176 @code{ignore} (@pxref{Conditions, ,Break Conditions}).
4178 The argument @var{ignore-count} is meaningful only when your program
4179 stopped due to a breakpoint. At other times, the argument to
4180 @code{continue} is ignored.
4182 The synonyms @code{c} and @code{fg} (for @dfn{foreground}, as the
4183 debugged program is deemed to be the foreground program) are provided
4184 purely for convenience, and have exactly the same behavior as
4188 To resume execution at a different place, you can use @code{return}
4189 (@pxref{Returning, ,Returning from a Function}) to go back to the
4190 calling function; or @code{jump} (@pxref{Jumping, ,Continuing at a
4191 Different Address}) to go to an arbitrary location in your program.
4193 A typical technique for using stepping is to set a breakpoint
4194 (@pxref{Breakpoints, ,Breakpoints; Watchpoints; and Catchpoints}) at the
4195 beginning of the function or the section of your program where a problem
4196 is believed to lie, run your program until it stops at that breakpoint,
4197 and then step through the suspect area, examining the variables that are
4198 interesting, until you see the problem happen.
4202 @kindex s @r{(@code{step})}
4204 Continue running your program until control reaches a different source
4205 line, then stop it and return control to @value{GDBN}. This command is
4206 abbreviated @code{s}.
4209 @c "without debugging information" is imprecise; actually "without line
4210 @c numbers in the debugging information". (gcc -g1 has debugging info but
4211 @c not line numbers). But it seems complex to try to make that
4212 @c distinction here.
4213 @emph{Warning:} If you use the @code{step} command while control is
4214 within a function that was compiled without debugging information,
4215 execution proceeds until control reaches a function that does have
4216 debugging information. Likewise, it will not step into a function which
4217 is compiled without debugging information. To step through functions
4218 without debugging information, use the @code{stepi} command, described
4222 The @code{step} command only stops at the first instruction of a source
4223 line. This prevents the multiple stops that could otherwise occur in
4224 @code{switch} statements, @code{for} loops, etc. @code{step} continues
4225 to stop if a function that has debugging information is called within
4226 the line. In other words, @code{step} @emph{steps inside} any functions
4227 called within the line.
4229 Also, the @code{step} command only enters a function if there is line
4230 number information for the function. Otherwise it acts like the
4231 @code{next} command. This avoids problems when using @code{cc -gl}
4232 on MIPS machines. Previously, @code{step} entered subroutines if there
4233 was any debugging information about the routine.
4235 @item step @var{count}
4236 Continue running as in @code{step}, but do so @var{count} times. If a
4237 breakpoint is reached, or a signal not related to stepping occurs before
4238 @var{count} steps, stepping stops right away.
4241 @kindex n @r{(@code{next})}
4242 @item next @r{[}@var{count}@r{]}
4243 Continue to the next source line in the current (innermost) stack frame.
4244 This is similar to @code{step}, but function calls that appear within
4245 the line of code are executed without stopping. Execution stops when
4246 control reaches a different line of code at the original stack level
4247 that was executing when you gave the @code{next} command. This command
4248 is abbreviated @code{n}.
4250 An argument @var{count} is a repeat count, as for @code{step}.
4253 @c FIX ME!! Do we delete this, or is there a way it fits in with
4254 @c the following paragraph? --- Vctoria
4256 @c @code{next} within a function that lacks debugging information acts like
4257 @c @code{step}, but any function calls appearing within the code of the
4258 @c function are executed without stopping.
4260 The @code{next} command only stops at the first instruction of a
4261 source line. This prevents multiple stops that could otherwise occur in
4262 @code{switch} statements, @code{for} loops, etc.
4264 @kindex set step-mode
4266 @cindex functions without line info, and stepping
4267 @cindex stepping into functions with no line info
4268 @itemx set step-mode on
4269 The @code{set step-mode on} command causes the @code{step} command to
4270 stop at the first instruction of a function which contains no debug line
4271 information rather than stepping over it.
4273 This is useful in cases where you may be interested in inspecting the
4274 machine instructions of a function which has no symbolic info and do not
4275 want @value{GDBN} to automatically skip over this function.
4277 @item set step-mode off
4278 Causes the @code{step} command to step over any functions which contains no
4279 debug information. This is the default.
4281 @item show step-mode
4282 Show whether @value{GDBN} will stop in or step over functions without
4283 source line debug information.
4286 @kindex fin @r{(@code{finish})}
4288 Continue running until just after function in the selected stack frame
4289 returns. Print the returned value (if any). This command can be
4290 abbreviated as @code{fin}.
4292 Contrast this with the @code{return} command (@pxref{Returning,
4293 ,Returning from a Function}).
4296 @kindex u @r{(@code{until})}
4297 @cindex run until specified location
4300 Continue running until a source line past the current line, in the
4301 current stack frame, is reached. This command is used to avoid single
4302 stepping through a loop more than once. It is like the @code{next}
4303 command, except that when @code{until} encounters a jump, it
4304 automatically continues execution until the program counter is greater
4305 than the address of the jump.
4307 This means that when you reach the end of a loop after single stepping
4308 though it, @code{until} makes your program continue execution until it
4309 exits the loop. In contrast, a @code{next} command at the end of a loop
4310 simply steps back to the beginning of the loop, which forces you to step
4311 through the next iteration.
4313 @code{until} always stops your program if it attempts to exit the current
4316 @code{until} may produce somewhat counterintuitive results if the order
4317 of machine code does not match the order of the source lines. For
4318 example, in the following excerpt from a debugging session, the @code{f}
4319 (@code{frame}) command shows that execution is stopped at line
4320 @code{206}; yet when we use @code{until}, we get to line @code{195}:
4324 #0 main (argc=4, argv=0xf7fffae8) at m4.c:206
4326 (@value{GDBP}) until
4327 195 for ( ; argc > 0; NEXTARG) @{
4330 This happened because, for execution efficiency, the compiler had
4331 generated code for the loop closure test at the end, rather than the
4332 start, of the loop---even though the test in a C @code{for}-loop is
4333 written before the body of the loop. The @code{until} command appeared
4334 to step back to the beginning of the loop when it advanced to this
4335 expression; however, it has not really gone to an earlier
4336 statement---not in terms of the actual machine code.
4338 @code{until} with no argument works by means of single
4339 instruction stepping, and hence is slower than @code{until} with an
4342 @item until @var{location}
4343 @itemx u @var{location}
4344 Continue running your program until either the specified location is
4345 reached, or the current stack frame returns. @var{location} is any of
4346 the forms described in @ref{Specify Location}.
4347 This form of the command uses temporary breakpoints, and
4348 hence is quicker than @code{until} without an argument. The specified
4349 location is actually reached only if it is in the current frame. This
4350 implies that @code{until} can be used to skip over recursive function
4351 invocations. For instance in the code below, if the current location is
4352 line @code{96}, issuing @code{until 99} will execute the program up to
4353 line @code{99} in the same invocation of factorial, i.e., after the inner
4354 invocations have returned.
4357 94 int factorial (int value)
4359 96 if (value > 1) @{
4360 97 value *= factorial (value - 1);
4367 @kindex advance @var{location}
4368 @itemx advance @var{location}
4369 Continue running the program up to the given @var{location}. An argument is
4370 required, which should be of one of the forms described in
4371 @ref{Specify Location}.
4372 Execution will also stop upon exit from the current stack
4373 frame. This command is similar to @code{until}, but @code{advance} will
4374 not skip over recursive function calls, and the target location doesn't
4375 have to be in the same frame as the current one.
4379 @kindex si @r{(@code{stepi})}
4381 @itemx stepi @var{arg}
4383 Execute one machine instruction, then stop and return to the debugger.
4385 It is often useful to do @samp{display/i $pc} when stepping by machine
4386 instructions. This makes @value{GDBN} automatically display the next
4387 instruction to be executed, each time your program stops. @xref{Auto
4388 Display,, Automatic Display}.
4390 An argument is a repeat count, as in @code{step}.
4394 @kindex ni @r{(@code{nexti})}
4396 @itemx nexti @var{arg}
4398 Execute one machine instruction, but if it is a function call,
4399 proceed until the function returns.
4401 An argument is a repeat count, as in @code{next}.
4408 A signal is an asynchronous event that can happen in a program. The
4409 operating system defines the possible kinds of signals, and gives each
4410 kind a name and a number. For example, in Unix @code{SIGINT} is the
4411 signal a program gets when you type an interrupt character (often @kbd{Ctrl-c});
4412 @code{SIGSEGV} is the signal a program gets from referencing a place in
4413 memory far away from all the areas in use; @code{SIGALRM} occurs when
4414 the alarm clock timer goes off (which happens only if your program has
4415 requested an alarm).
4417 @cindex fatal signals
4418 Some signals, including @code{SIGALRM}, are a normal part of the
4419 functioning of your program. Others, such as @code{SIGSEGV}, indicate
4420 errors; these signals are @dfn{fatal} (they kill your program immediately) if the
4421 program has not specified in advance some other way to handle the signal.
4422 @code{SIGINT} does not indicate an error in your program, but it is normally
4423 fatal so it can carry out the purpose of the interrupt: to kill the program.
4425 @value{GDBN} has the ability to detect any occurrence of a signal in your
4426 program. You can tell @value{GDBN} in advance what to do for each kind of
4429 @cindex handling signals
4430 Normally, @value{GDBN} is set up to let the non-erroneous signals like
4431 @code{SIGALRM} be silently passed to your program
4432 (so as not to interfere with their role in the program's functioning)
4433 but to stop your program immediately whenever an error signal happens.
4434 You can change these settings with the @code{handle} command.
4437 @kindex info signals
4441 Print a table of all the kinds of signals and how @value{GDBN} has been told to
4442 handle each one. You can use this to see the signal numbers of all
4443 the defined types of signals.
4445 @item info signals @var{sig}
4446 Similar, but print information only about the specified signal number.
4448 @code{info handle} is an alias for @code{info signals}.
4451 @item handle @var{signal} @r{[}@var{keywords}@dots{}@r{]}
4452 Change the way @value{GDBN} handles signal @var{signal}. @var{signal}
4453 can be the number of a signal or its name (with or without the
4454 @samp{SIG} at the beginning); a list of signal numbers of the form
4455 @samp{@var{low}-@var{high}}; or the word @samp{all}, meaning all the
4456 known signals. Optional arguments @var{keywords}, described below,
4457 say what change to make.
4461 The keywords allowed by the @code{handle} command can be abbreviated.
4462 Their full names are:
4466 @value{GDBN} should not stop your program when this signal happens. It may
4467 still print a message telling you that the signal has come in.
4470 @value{GDBN} should stop your program when this signal happens. This implies
4471 the @code{print} keyword as well.
4474 @value{GDBN} should print a message when this signal happens.
4477 @value{GDBN} should not mention the occurrence of the signal at all. This
4478 implies the @code{nostop} keyword as well.
4482 @value{GDBN} should allow your program to see this signal; your program
4483 can handle the signal, or else it may terminate if the signal is fatal
4484 and not handled. @code{pass} and @code{noignore} are synonyms.
4488 @value{GDBN} should not allow your program to see this signal.
4489 @code{nopass} and @code{ignore} are synonyms.
4493 When a signal stops your program, the signal is not visible to the
4495 continue. Your program sees the signal then, if @code{pass} is in
4496 effect for the signal in question @emph{at that time}. In other words,
4497 after @value{GDBN} reports a signal, you can use the @code{handle}
4498 command with @code{pass} or @code{nopass} to control whether your
4499 program sees that signal when you continue.
4501 The default is set to @code{nostop}, @code{noprint}, @code{pass} for
4502 non-erroneous signals such as @code{SIGALRM}, @code{SIGWINCH} and
4503 @code{SIGCHLD}, and to @code{stop}, @code{print}, @code{pass} for the
4506 You can also use the @code{signal} command to prevent your program from
4507 seeing a signal, or cause it to see a signal it normally would not see,
4508 or to give it any signal at any time. For example, if your program stopped
4509 due to some sort of memory reference error, you might store correct
4510 values into the erroneous variables and continue, hoping to see more
4511 execution; but your program would probably terminate immediately as
4512 a result of the fatal signal once it saw the signal. To prevent this,
4513 you can continue with @samp{signal 0}. @xref{Signaling, ,Giving your
4516 @cindex extra signal information
4517 @anchor{extra signal information}
4519 On some targets, @value{GDBN} can inspect extra signal information
4520 associated with the intercepted signal, before it is actually
4521 delivered to the program being debugged. This information is exported
4522 by the convenience variable @code{$_siginfo}, and consists of data
4523 that is passed by the kernel to the signal handler at the time of the
4524 receipt of a signal. The data type of the information itself is
4525 target dependent. You can see the data type using the @code{ptype
4526 $_siginfo} command. On Unix systems, it typically corresponds to the
4527 standard @code{siginfo_t} type, as defined in the @file{signal.h}
4530 Here's an example, on a @sc{gnu}/Linux system, printing the stray
4531 referenced address that raised a segmentation fault.
4535 (@value{GDBP}) continue
4536 Program received signal SIGSEGV, Segmentation fault.
4537 0x0000000000400766 in main ()
4539 (@value{GDBP}) ptype $_siginfo
4546 struct @{...@} _kill;
4547 struct @{...@} _timer;
4549 struct @{...@} _sigchld;
4550 struct @{...@} _sigfault;
4551 struct @{...@} _sigpoll;
4554 (@value{GDBP}) ptype $_siginfo._sifields._sigfault
4558 (@value{GDBP}) p $_siginfo._sifields._sigfault.si_addr
4559 $1 = (void *) 0x7ffff7ff7000
4563 Depending on target support, @code{$_siginfo} may also be writable.
4566 @section Stopping and Starting Multi-thread Programs
4568 @cindex stopped threads
4569 @cindex threads, stopped
4571 @cindex continuing threads
4572 @cindex threads, continuing
4574 @value{GDBN} supports debugging programs with multiple threads
4575 (@pxref{Threads,, Debugging Programs with Multiple Threads}). There
4576 are two modes of controlling execution of your program within the
4577 debugger. In the default mode, referred to as @dfn{all-stop mode},
4578 when any thread in your program stops (for example, at a breakpoint
4579 or while being stepped), all other threads in the program are also stopped by
4580 @value{GDBN}. On some targets, @value{GDBN} also supports
4581 @dfn{non-stop mode}, in which other threads can continue to run freely while
4582 you examine the stopped thread in the debugger.
4585 * All-Stop Mode:: All threads stop when GDB takes control
4586 * Non-Stop Mode:: Other threads continue to execute
4587 * Background Execution:: Running your program asynchronously
4588 * Thread-Specific Breakpoints:: Controlling breakpoints
4589 * Interrupted System Calls:: GDB may interfere with system calls
4593 @subsection All-Stop Mode
4595 @cindex all-stop mode
4597 In all-stop mode, whenever your program stops under @value{GDBN} for any reason,
4598 @emph{all} threads of execution stop, not just the current thread. This
4599 allows you to examine the overall state of the program, including
4600 switching between threads, without worrying that things may change
4603 Conversely, whenever you restart the program, @emph{all} threads start
4604 executing. @emph{This is true even when single-stepping} with commands
4605 like @code{step} or @code{next}.
4607 In particular, @value{GDBN} cannot single-step all threads in lockstep.
4608 Since thread scheduling is up to your debugging target's operating
4609 system (not controlled by @value{GDBN}), other threads may
4610 execute more than one statement while the current thread completes a
4611 single step. Moreover, in general other threads stop in the middle of a
4612 statement, rather than at a clean statement boundary, when the program
4615 You might even find your program stopped in another thread after
4616 continuing or even single-stepping. This happens whenever some other
4617 thread runs into a breakpoint, a signal, or an exception before the
4618 first thread completes whatever you requested.
4620 @cindex automatic thread selection
4621 @cindex switching threads automatically
4622 @cindex threads, automatic switching
4623 Whenever @value{GDBN} stops your program, due to a breakpoint or a
4624 signal, it automatically selects the thread where that breakpoint or
4625 signal happened. @value{GDBN} alerts you to the context switch with a
4626 message such as @samp{[Switching to Thread @var{n}]} to identify the
4629 On some OSes, you can modify @value{GDBN}'s default behavior by
4630 locking the OS scheduler to allow only a single thread to run.
4633 @item set scheduler-locking @var{mode}
4634 @cindex scheduler locking mode
4635 @cindex lock scheduler
4636 Set the scheduler locking mode. If it is @code{off}, then there is no
4637 locking and any thread may run at any time. If @code{on}, then only the
4638 current thread may run when the inferior is resumed. The @code{step}
4639 mode optimizes for single-stepping; it prevents other threads
4640 from preempting the current thread while you are stepping, so that
4641 the focus of debugging does not change unexpectedly.
4642 Other threads only rarely (or never) get a chance to run
4643 when you step. They are more likely to run when you @samp{next} over a
4644 function call, and they are completely free to run when you use commands
4645 like @samp{continue}, @samp{until}, or @samp{finish}. However, unless another
4646 thread hits a breakpoint during its timeslice, @value{GDBN} does not change
4647 the current thread away from the thread that you are debugging.
4649 @item show scheduler-locking
4650 Display the current scheduler locking mode.
4653 @cindex resume threads of multiple processes simultaneously
4654 By default, when you issue one of the execution commands such as
4655 @code{continue}, @code{next} or @code{step}, @value{GDBN} allows only
4656 threads of the current inferior to run. For example, if @value{GDBN}
4657 is attached to two inferiors, each with two threads, the
4658 @code{continue} command resumes only the two threads of the current
4659 inferior. This is useful, for example, when you debug a program that
4660 forks and you want to hold the parent stopped (so that, for instance,
4661 it doesn't run to exit), while you debug the child. In other
4662 situations, you may not be interested in inspecting the current state
4663 of any of the processes @value{GDBN} is attached to, and you may want
4664 to resume them all until some breakpoint is hit. In the latter case,
4665 you can instruct @value{GDBN} to allow all threads of all the
4666 inferiors to run with the @w{@code{set schedule-multiple}} command.
4669 @kindex set schedule-multiple
4670 @item set schedule-multiple
4671 Set the mode for allowing threads of multiple processes to be resumed
4672 when an execution command is issued. When @code{on}, all threads of
4673 all processes are allowed to run. When @code{off}, only the threads
4674 of the current process are resumed. The default is @code{off}. The
4675 @code{scheduler-locking} mode takes precedence when set to @code{on},
4676 or while you are stepping and set to @code{step}.
4678 @item show schedule-multiple
4679 Display the current mode for resuming the execution of threads of
4684 @subsection Non-Stop Mode
4686 @cindex non-stop mode
4688 @c This section is really only a place-holder, and needs to be expanded
4689 @c with more details.
4691 For some multi-threaded targets, @value{GDBN} supports an optional
4692 mode of operation in which you can examine stopped program threads in
4693 the debugger while other threads continue to execute freely. This
4694 minimizes intrusion when debugging live systems, such as programs
4695 where some threads have real-time constraints or must continue to
4696 respond to external events. This is referred to as @dfn{non-stop} mode.
4698 In non-stop mode, when a thread stops to report a debugging event,
4699 @emph{only} that thread is stopped; @value{GDBN} does not stop other
4700 threads as well, in contrast to the all-stop mode behavior. Additionally,
4701 execution commands such as @code{continue} and @code{step} apply by default
4702 only to the current thread in non-stop mode, rather than all threads as
4703 in all-stop mode. This allows you to control threads explicitly in
4704 ways that are not possible in all-stop mode --- for example, stepping
4705 one thread while allowing others to run freely, stepping
4706 one thread while holding all others stopped, or stepping several threads
4707 independently and simultaneously.
4709 To enter non-stop mode, use this sequence of commands before you run
4710 or attach to your program:
4713 # Enable the async interface.
4716 # If using the CLI, pagination breaks non-stop.
4719 # Finally, turn it on!
4723 You can use these commands to manipulate the non-stop mode setting:
4726 @kindex set non-stop
4727 @item set non-stop on
4728 Enable selection of non-stop mode.
4729 @item set non-stop off
4730 Disable selection of non-stop mode.
4731 @kindex show non-stop
4733 Show the current non-stop enablement setting.
4736 Note these commands only reflect whether non-stop mode is enabled,
4737 not whether the currently-executing program is being run in non-stop mode.
4738 In particular, the @code{set non-stop} preference is only consulted when
4739 @value{GDBN} starts or connects to the target program, and it is generally
4740 not possible to switch modes once debugging has started. Furthermore,
4741 since not all targets support non-stop mode, even when you have enabled
4742 non-stop mode, @value{GDBN} may still fall back to all-stop operation by
4745 In non-stop mode, all execution commands apply only to the current thread
4746 by default. That is, @code{continue} only continues one thread.
4747 To continue all threads, issue @code{continue -a} or @code{c -a}.
4749 You can use @value{GDBN}'s background execution commands
4750 (@pxref{Background Execution}) to run some threads in the background
4751 while you continue to examine or step others from @value{GDBN}.
4752 The MI execution commands (@pxref{GDB/MI Program Execution}) are
4753 always executed asynchronously in non-stop mode.
4755 Suspending execution is done with the @code{interrupt} command when
4756 running in the background, or @kbd{Ctrl-c} during foreground execution.
4757 In all-stop mode, this stops the whole process;
4758 but in non-stop mode the interrupt applies only to the current thread.
4759 To stop the whole program, use @code{interrupt -a}.
4761 Other execution commands do not currently support the @code{-a} option.
4763 In non-stop mode, when a thread stops, @value{GDBN} doesn't automatically make
4764 that thread current, as it does in all-stop mode. This is because the
4765 thread stop notifications are asynchronous with respect to @value{GDBN}'s
4766 command interpreter, and it would be confusing if @value{GDBN} unexpectedly
4767 changed to a different thread just as you entered a command to operate on the
4768 previously current thread.
4770 @node Background Execution
4771 @subsection Background Execution
4773 @cindex foreground execution
4774 @cindex background execution
4775 @cindex asynchronous execution
4776 @cindex execution, foreground, background and asynchronous
4778 @value{GDBN}'s execution commands have two variants: the normal
4779 foreground (synchronous) behavior, and a background
4780 (asynchronous) behavior. In foreground execution, @value{GDBN} waits for
4781 the program to report that some thread has stopped before prompting for
4782 another command. In background execution, @value{GDBN} immediately gives
4783 a command prompt so that you can issue other commands while your program runs.
4785 You need to explicitly enable asynchronous mode before you can use
4786 background execution commands. You can use these commands to
4787 manipulate the asynchronous mode setting:
4790 @kindex set target-async
4791 @item set target-async on
4792 Enable asynchronous mode.
4793 @item set target-async off
4794 Disable asynchronous mode.
4795 @kindex show target-async
4796 @item show target-async
4797 Show the current target-async setting.
4800 If the target doesn't support async mode, @value{GDBN} issues an error
4801 message if you attempt to use the background execution commands.
4803 To specify background execution, add a @code{&} to the command. For example,
4804 the background form of the @code{continue} command is @code{continue&}, or
4805 just @code{c&}. The execution commands that accept background execution
4811 @xref{Starting, , Starting your Program}.
4815 @xref{Attach, , Debugging an Already-running Process}.
4819 @xref{Continuing and Stepping, step}.
4823 @xref{Continuing and Stepping, stepi}.
4827 @xref{Continuing and Stepping, next}.
4831 @xref{Continuing and Stepping, nexti}.
4835 @xref{Continuing and Stepping, continue}.
4839 @xref{Continuing and Stepping, finish}.
4843 @xref{Continuing and Stepping, until}.
4847 Background execution is especially useful in conjunction with non-stop
4848 mode for debugging programs with multiple threads; see @ref{Non-Stop Mode}.
4849 However, you can also use these commands in the normal all-stop mode with
4850 the restriction that you cannot issue another execution command until the
4851 previous one finishes. Examples of commands that are valid in all-stop
4852 mode while the program is running include @code{help} and @code{info break}.
4854 You can interrupt your program while it is running in the background by
4855 using the @code{interrupt} command.
4862 Suspend execution of the running program. In all-stop mode,
4863 @code{interrupt} stops the whole process, but in non-stop mode, it stops
4864 only the current thread. To stop the whole program in non-stop mode,
4865 use @code{interrupt -a}.
4868 @node Thread-Specific Breakpoints
4869 @subsection Thread-Specific Breakpoints
4871 When your program has multiple threads (@pxref{Threads,, Debugging
4872 Programs with Multiple Threads}), you can choose whether to set
4873 breakpoints on all threads, or on a particular thread.
4876 @cindex breakpoints and threads
4877 @cindex thread breakpoints
4878 @kindex break @dots{} thread @var{threadno}
4879 @item break @var{linespec} thread @var{threadno}
4880 @itemx break @var{linespec} thread @var{threadno} if @dots{}
4881 @var{linespec} specifies source lines; there are several ways of
4882 writing them (@pxref{Specify Location}), but the effect is always to
4883 specify some source line.
4885 Use the qualifier @samp{thread @var{threadno}} with a breakpoint command
4886 to specify that you only want @value{GDBN} to stop the program when a
4887 particular thread reaches this breakpoint. @var{threadno} is one of the
4888 numeric thread identifiers assigned by @value{GDBN}, shown in the first
4889 column of the @samp{info threads} display.
4891 If you do not specify @samp{thread @var{threadno}} when you set a
4892 breakpoint, the breakpoint applies to @emph{all} threads of your
4895 You can use the @code{thread} qualifier on conditional breakpoints as
4896 well; in this case, place @samp{thread @var{threadno}} before the
4897 breakpoint condition, like this:
4900 (@value{GDBP}) break frik.c:13 thread 28 if bartab > lim
4905 @node Interrupted System Calls
4906 @subsection Interrupted System Calls
4908 @cindex thread breakpoints and system calls
4909 @cindex system calls and thread breakpoints
4910 @cindex premature return from system calls
4911 There is an unfortunate side effect when using @value{GDBN} to debug
4912 multi-threaded programs. If one thread stops for a
4913 breakpoint, or for some other reason, and another thread is blocked in a
4914 system call, then the system call may return prematurely. This is a
4915 consequence of the interaction between multiple threads and the signals
4916 that @value{GDBN} uses to implement breakpoints and other events that
4919 To handle this problem, your program should check the return value of
4920 each system call and react appropriately. This is good programming
4923 For example, do not write code like this:
4929 The call to @code{sleep} will return early if a different thread stops
4930 at a breakpoint or for some other reason.
4932 Instead, write this:
4937 unslept = sleep (unslept);
4940 A system call is allowed to return early, so the system is still
4941 conforming to its specification. But @value{GDBN} does cause your
4942 multi-threaded program to behave differently than it would without
4945 Also, @value{GDBN} uses internal breakpoints in the thread library to
4946 monitor certain events such as thread creation and thread destruction.
4947 When such an event happens, a system call in another thread may return
4948 prematurely, even though your program does not appear to stop.
4951 @node Reverse Execution
4952 @chapter Running programs backward
4953 @cindex reverse execution
4954 @cindex running programs backward
4956 When you are debugging a program, it is not unusual to realize that
4957 you have gone too far, and some event of interest has already happened.
4958 If the target environment supports it, @value{GDBN} can allow you to
4959 ``rewind'' the program by running it backward.
4961 A target environment that supports reverse execution should be able
4962 to ``undo'' the changes in machine state that have taken place as the
4963 program was executing normally. Variables, registers etc.@: should
4964 revert to their previous values. Obviously this requires a great
4965 deal of sophistication on the part of the target environment; not
4966 all target environments can support reverse execution.
4968 When a program is executed in reverse, the instructions that
4969 have most recently been executed are ``un-executed'', in reverse
4970 order. The program counter runs backward, following the previous
4971 thread of execution in reverse. As each instruction is ``un-executed'',
4972 the values of memory and/or registers that were changed by that
4973 instruction are reverted to their previous states. After executing
4974 a piece of source code in reverse, all side effects of that code
4975 should be ``undone'', and all variables should be returned to their
4976 prior values@footnote{
4977 Note that some side effects are easier to undo than others. For instance,
4978 memory and registers are relatively easy, but device I/O is hard. Some
4979 targets may be able undo things like device I/O, and some may not.
4981 The contract between @value{GDBN} and the reverse executing target
4982 requires only that the target do something reasonable when
4983 @value{GDBN} tells it to execute backwards, and then report the
4984 results back to @value{GDBN}. Whatever the target reports back to
4985 @value{GDBN}, @value{GDBN} will report back to the user. @value{GDBN}
4986 assumes that the memory and registers that the target reports are in a
4987 consistant state, but @value{GDBN} accepts whatever it is given.
4990 If you are debugging in a target environment that supports
4991 reverse execution, @value{GDBN} provides the following commands.
4994 @kindex reverse-continue
4995 @kindex rc @r{(@code{reverse-continue})}
4996 @item reverse-continue @r{[}@var{ignore-count}@r{]}
4997 @itemx rc @r{[}@var{ignore-count}@r{]}
4998 Beginning at the point where your program last stopped, start executing
4999 in reverse. Reverse execution will stop for breakpoints and synchronous
5000 exceptions (signals), just like normal execution. Behavior of
5001 asynchronous signals depends on the target environment.
5003 @kindex reverse-step
5004 @kindex rs @r{(@code{step})}
5005 @item reverse-step @r{[}@var{count}@r{]}
5006 Run the program backward until control reaches the start of a
5007 different source line; then stop it, and return control to @value{GDBN}.
5009 Like the @code{step} command, @code{reverse-step} will only stop
5010 at the beginning of a source line. It ``un-executes'' the previously
5011 executed source line. If the previous source line included calls to
5012 debuggable functions, @code{reverse-step} will step (backward) into
5013 the called function, stopping at the beginning of the @emph{last}
5014 statement in the called function (typically a return statement).
5016 Also, as with the @code{step} command, if non-debuggable functions are
5017 called, @code{reverse-step} will run thru them backward without stopping.
5019 @kindex reverse-stepi
5020 @kindex rsi @r{(@code{reverse-stepi})}
5021 @item reverse-stepi @r{[}@var{count}@r{]}
5022 Reverse-execute one machine instruction. Note that the instruction
5023 to be reverse-executed is @emph{not} the one pointed to by the program
5024 counter, but the instruction executed prior to that one. For instance,
5025 if the last instruction was a jump, @code{reverse-stepi} will take you
5026 back from the destination of the jump to the jump instruction itself.
5028 @kindex reverse-next
5029 @kindex rn @r{(@code{reverse-next})}
5030 @item reverse-next @r{[}@var{count}@r{]}
5031 Run backward to the beginning of the previous line executed in
5032 the current (innermost) stack frame. If the line contains function
5033 calls, they will be ``un-executed'' without stopping. Starting from
5034 the first line of a function, @code{reverse-next} will take you back
5035 to the caller of that function, @emph{before} the function was called,
5036 just as the normal @code{next} command would take you from the last
5037 line of a function back to its return to its caller
5038 @footnote{Unles the code is too heavily optimized.}.
5040 @kindex reverse-nexti
5041 @kindex rni @r{(@code{reverse-nexti})}
5042 @item reverse-nexti @r{[}@var{count}@r{]}
5043 Like @code{nexti}, @code{reverse-nexti} executes a single instruction
5044 in reverse, except that called functions are ``un-executed'' atomically.
5045 That is, if the previously executed instruction was a return from
5046 another instruction, @code{reverse-nexti} will continue to execute
5047 in reverse until the call to that function (from the current stack
5050 @kindex reverse-finish
5051 @item reverse-finish
5052 Just as the @code{finish} command takes you to the point where the
5053 current function returns, @code{reverse-finish} takes you to the point
5054 where it was called. Instead of ending up at the end of the current
5055 function invocation, you end up at the beginning.
5057 @kindex set exec-direction
5058 @item set exec-direction
5059 Set the direction of target execution.
5060 @itemx set exec-direction reverse
5061 @cindex execute forward or backward in time
5062 @value{GDBN} will perform all execution commands in reverse, until the
5063 exec-direction mode is changed to ``forward''. Affected commands include
5064 @code{step, stepi, next, nexti, continue, and finish}. The @code{return}
5065 command cannot be used in reverse mode.
5066 @item set exec-direction forward
5067 @value{GDBN} will perform all execution commands in the normal fashion.
5068 This is the default.
5072 @node Process Record and Replay
5073 @chapter Recording Inferior's Execution and Replaying It
5074 @cindex process record and replay
5075 @cindex recording inferior's execution and replaying it
5077 On some platforms, @value{GDBN} provides a special @dfn{process record
5078 and replay} target that can record a log of the process execution, and
5079 replay it later with both forward and reverse execution commands.
5082 When this target is in use, if the execution log includes the record
5083 for the next instruction, @value{GDBN} will debug in @dfn{replay
5084 mode}. In the replay mode, the inferior does not really execute code
5085 instructions. Instead, all the events that normally happen during
5086 code execution are taken from the execution log. While code is not
5087 really executed in replay mode, the values of registers (including the
5088 program counter register) and the memory of the inferior are still
5089 changed as they normally would. Their contents are taken from the
5093 If the record for the next instruction is not in the execution log,
5094 @value{GDBN} will debug in @dfn{record mode}. In this mode, the
5095 inferior executes normally, and @value{GDBN} records the execution log
5098 The process record and replay target supports reverse execution
5099 (@pxref{Reverse Execution}), even if the platform on which the
5100 inferior runs does not. However, the reverse execution is limited in
5101 this case by the range of the instructions recorded in the execution
5102 log. In other words, reverse execution on platforms that don't
5103 support it directly can only be done in the replay mode.
5105 When debugging in the reverse direction, @value{GDBN} will work in
5106 replay mode as long as the execution log includes the record for the
5107 previous instruction; otherwise, it will work in record mode, if the
5108 platform supports reverse execution, or stop if not.
5110 For architecture environments that support process record and replay,
5111 @value{GDBN} provides the following commands:
5114 @kindex target record
5118 This command starts the process record and replay target. The process
5119 record and replay target can only debug a process that is already
5120 running. Therefore, you need first to start the process with the
5121 @kbd{run} or @kbd{start} commands, and then start the recording with
5122 the @kbd{target record} command.
5124 Both @code{record} and @code{rec} are aliases of @code{target record}.
5126 @cindex displaced stepping, and process record and replay
5127 Displaced stepping (@pxref{Maintenance Commands,, displaced stepping})
5128 will be automatically disabled when process record and replay target
5129 is started. That's because the process record and replay target
5130 doesn't support displaced stepping.
5132 @cindex non-stop mode, and process record and replay
5133 @cindex asynchronous execution, and process record and replay
5134 If the inferior is in the non-stop mode (@pxref{Non-Stop Mode}) or in
5135 the asynchronous execution mode (@pxref{Background Execution}), the
5136 process record and replay target cannot be started because it doesn't
5137 support these two modes.
5142 Stop the process record and replay target. When process record and
5143 replay target stops, the entire execution log will be deleted and the
5144 inferior will either be terminated, or will remain in its final state.
5146 When you stop the process record and replay target in record mode (at
5147 the end of the execution log), the inferior will be stopped at the
5148 next instruction that would have been recorded. In other words, if
5149 you record for a while and then stop recording, the inferior process
5150 will be left in the same state as if the recording never happened.
5152 On the other hand, if the process record and replay target is stopped
5153 while in replay mode (that is, not at the end of the execution log,
5154 but at some earlier point), the inferior process will become ``live''
5155 at that earlier state, and it will then be possible to continue the
5156 usual ``live'' debugging of the process from that state.
5158 When the inferior process exits, or @value{GDBN} detaches from it,
5159 process record and replay target will automatically stop itself.
5161 @kindex set record insn-number-max
5162 @item set record insn-number-max @var{limit}
5163 Set the limit of instructions to be recorded. Default value is 200000.
5165 If @var{limit} is a positive number, then @value{GDBN} will start
5166 deleting instructions from the log once the number of the record
5167 instructions becomes greater than @var{limit}. For every new recorded
5168 instruction, @value{GDBN} will delete the earliest recorded
5169 instruction to keep the number of recorded instructions at the limit.
5170 (Since deleting recorded instructions loses information, @value{GDBN}
5171 lets you control what happens when the limit is reached, by means of
5172 the @code{stop-at-limit} option, described below.)
5174 If @var{limit} is zero, @value{GDBN} will never delete recorded
5175 instructions from the execution log. The number of recorded
5176 instructions is unlimited in this case.
5178 @kindex show record insn-number-max
5179 @item show record insn-number-max
5180 Show the limit of instructions to be recorded.
5182 @kindex set record stop-at-limit
5183 @item set record stop-at-limit
5184 Control the behavior when the number of recorded instructions reaches
5185 the limit. If ON (the default), @value{GDBN} will stop when the limit
5186 is reached for the first time and ask you whether you want to stop the
5187 inferior or continue running it and recording the execution log. If
5188 you decide to continue recording, each new recorded instruction will
5189 cause the oldest one to be deleted.
5191 If this option is OFF, @value{GDBN} will automatically delete the
5192 oldest record to make room for each new one, without asking.
5194 @kindex show record stop-at-limit
5195 @item show record stop-at-limit
5196 Show the current setting of @code{stop-at-limit}.
5198 @kindex info record insn-number
5199 @item info record insn-number
5200 Show the current number of recorded instructions.
5202 @kindex record delete
5205 When record target runs in replay mode (``in the past''), delete the
5206 subsequent execution log and begin to record a new execution log starting
5207 from the current address. This means you will abandon the previously
5208 recorded ``future'' and begin recording a new ``future''.
5213 @chapter Examining the Stack
5215 When your program has stopped, the first thing you need to know is where it
5216 stopped and how it got there.
5219 Each time your program performs a function call, information about the call
5221 That information includes the location of the call in your program,
5222 the arguments of the call,
5223 and the local variables of the function being called.
5224 The information is saved in a block of data called a @dfn{stack frame}.
5225 The stack frames are allocated in a region of memory called the @dfn{call
5228 When your program stops, the @value{GDBN} commands for examining the
5229 stack allow you to see all of this information.
5231 @cindex selected frame
5232 One of the stack frames is @dfn{selected} by @value{GDBN} and many
5233 @value{GDBN} commands refer implicitly to the selected frame. In
5234 particular, whenever you ask @value{GDBN} for the value of a variable in
5235 your program, the value is found in the selected frame. There are
5236 special @value{GDBN} commands to select whichever frame you are
5237 interested in. @xref{Selection, ,Selecting a Frame}.
5239 When your program stops, @value{GDBN} automatically selects the
5240 currently executing frame and describes it briefly, similar to the
5241 @code{frame} command (@pxref{Frame Info, ,Information about a Frame}).
5244 * Frames:: Stack frames
5245 * Backtrace:: Backtraces
5246 * Selection:: Selecting a frame
5247 * Frame Info:: Information on a frame
5252 @section Stack Frames
5254 @cindex frame, definition
5256 The call stack is divided up into contiguous pieces called @dfn{stack
5257 frames}, or @dfn{frames} for short; each frame is the data associated
5258 with one call to one function. The frame contains the arguments given
5259 to the function, the function's local variables, and the address at
5260 which the function is executing.
5262 @cindex initial frame
5263 @cindex outermost frame
5264 @cindex innermost frame
5265 When your program is started, the stack has only one frame, that of the
5266 function @code{main}. This is called the @dfn{initial} frame or the
5267 @dfn{outermost} frame. Each time a function is called, a new frame is
5268 made. Each time a function returns, the frame for that function invocation
5269 is eliminated. If a function is recursive, there can be many frames for
5270 the same function. The frame for the function in which execution is
5271 actually occurring is called the @dfn{innermost} frame. This is the most
5272 recently created of all the stack frames that still exist.
5274 @cindex frame pointer
5275 Inside your program, stack frames are identified by their addresses. A
5276 stack frame consists of many bytes, each of which has its own address; each
5277 kind of computer has a convention for choosing one byte whose
5278 address serves as the address of the frame. Usually this address is kept
5279 in a register called the @dfn{frame pointer register}
5280 (@pxref{Registers, $fp}) while execution is going on in that frame.
5282 @cindex frame number
5283 @value{GDBN} assigns numbers to all existing stack frames, starting with
5284 zero for the innermost frame, one for the frame that called it,
5285 and so on upward. These numbers do not really exist in your program;
5286 they are assigned by @value{GDBN} to give you a way of designating stack
5287 frames in @value{GDBN} commands.
5289 @c The -fomit-frame-pointer below perennially causes hbox overflow
5290 @c underflow problems.
5291 @cindex frameless execution
5292 Some compilers provide a way to compile functions so that they operate
5293 without stack frames. (For example, the @value{NGCC} option
5295 @samp{-fomit-frame-pointer}
5297 generates functions without a frame.)
5298 This is occasionally done with heavily used library functions to save
5299 the frame setup time. @value{GDBN} has limited facilities for dealing
5300 with these function invocations. If the innermost function invocation
5301 has no stack frame, @value{GDBN} nevertheless regards it as though
5302 it had a separate frame, which is numbered zero as usual, allowing
5303 correct tracing of the function call chain. However, @value{GDBN} has
5304 no provision for frameless functions elsewhere in the stack.
5307 @kindex frame@r{, command}
5308 @cindex current stack frame
5309 @item frame @var{args}
5310 The @code{frame} command allows you to move from one stack frame to another,
5311 and to print the stack frame you select. @var{args} may be either the
5312 address of the frame or the stack frame number. Without an argument,
5313 @code{frame} prints the current stack frame.
5315 @kindex select-frame
5316 @cindex selecting frame silently
5318 The @code{select-frame} command allows you to move from one stack frame
5319 to another without printing the frame. This is the silent version of
5327 @cindex call stack traces
5328 A backtrace is a summary of how your program got where it is. It shows one
5329 line per frame, for many frames, starting with the currently executing
5330 frame (frame zero), followed by its caller (frame one), and on up the
5335 @kindex bt @r{(@code{backtrace})}
5338 Print a backtrace of the entire stack: one line per frame for all
5339 frames in the stack.
5341 You can stop the backtrace at any time by typing the system interrupt
5342 character, normally @kbd{Ctrl-c}.
5344 @item backtrace @var{n}
5346 Similar, but print only the innermost @var{n} frames.
5348 @item backtrace -@var{n}
5350 Similar, but print only the outermost @var{n} frames.
5352 @item backtrace full
5354 @itemx bt full @var{n}
5355 @itemx bt full -@var{n}
5356 Print the values of the local variables also. @var{n} specifies the
5357 number of frames to print, as described above.
5362 The names @code{where} and @code{info stack} (abbreviated @code{info s})
5363 are additional aliases for @code{backtrace}.
5365 @cindex multiple threads, backtrace
5366 In a multi-threaded program, @value{GDBN} by default shows the
5367 backtrace only for the current thread. To display the backtrace for
5368 several or all of the threads, use the command @code{thread apply}
5369 (@pxref{Threads, thread apply}). For example, if you type @kbd{thread
5370 apply all backtrace}, @value{GDBN} will display the backtrace for all
5371 the threads; this is handy when you debug a core dump of a
5372 multi-threaded program.
5374 Each line in the backtrace shows the frame number and the function name.
5375 The program counter value is also shown---unless you use @code{set
5376 print address off}. The backtrace also shows the source file name and
5377 line number, as well as the arguments to the function. The program
5378 counter value is omitted if it is at the beginning of the code for that
5381 Here is an example of a backtrace. It was made with the command
5382 @samp{bt 3}, so it shows the innermost three frames.
5386 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
5388 #1 0x6e38 in expand_macro (sym=0x2b600, data=...) at macro.c:242
5389 #2 0x6840 in expand_token (obs=0x0, t=177664, td=0xf7fffb08)
5391 (More stack frames follow...)
5396 The display for frame zero does not begin with a program counter
5397 value, indicating that your program has stopped at the beginning of the
5398 code for line @code{993} of @code{builtin.c}.
5401 The value of parameter @code{data} in frame 1 has been replaced by
5402 @code{@dots{}}. By default, @value{GDBN} prints the value of a parameter
5403 only if it is a scalar (integer, pointer, enumeration, etc). See command
5404 @kbd{set print frame-arguments} in @ref{Print Settings} for more details
5405 on how to configure the way function parameter values are printed.
5407 @cindex value optimized out, in backtrace
5408 @cindex function call arguments, optimized out
5409 If your program was compiled with optimizations, some compilers will
5410 optimize away arguments passed to functions if those arguments are
5411 never used after the call. Such optimizations generate code that
5412 passes arguments through registers, but doesn't store those arguments
5413 in the stack frame. @value{GDBN} has no way of displaying such
5414 arguments in stack frames other than the innermost one. Here's what
5415 such a backtrace might look like:
5419 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
5421 #1 0x6e38 in expand_macro (sym=<value optimized out>) at macro.c:242
5422 #2 0x6840 in expand_token (obs=0x0, t=<value optimized out>, td=0xf7fffb08)
5424 (More stack frames follow...)
5429 The values of arguments that were not saved in their stack frames are
5430 shown as @samp{<value optimized out>}.
5432 If you need to display the values of such optimized-out arguments,
5433 either deduce that from other variables whose values depend on the one
5434 you are interested in, or recompile without optimizations.
5436 @cindex backtrace beyond @code{main} function
5437 @cindex program entry point
5438 @cindex startup code, and backtrace
5439 Most programs have a standard user entry point---a place where system
5440 libraries and startup code transition into user code. For C this is
5441 @code{main}@footnote{
5442 Note that embedded programs (the so-called ``free-standing''
5443 environment) are not required to have a @code{main} function as the
5444 entry point. They could even have multiple entry points.}.
5445 When @value{GDBN} finds the entry function in a backtrace
5446 it will terminate the backtrace, to avoid tracing into highly
5447 system-specific (and generally uninteresting) code.
5449 If you need to examine the startup code, or limit the number of levels
5450 in a backtrace, you can change this behavior:
5453 @item set backtrace past-main
5454 @itemx set backtrace past-main on
5455 @kindex set backtrace
5456 Backtraces will continue past the user entry point.
5458 @item set backtrace past-main off
5459 Backtraces will stop when they encounter the user entry point. This is the
5462 @item show backtrace past-main
5463 @kindex show backtrace
5464 Display the current user entry point backtrace policy.
5466 @item set backtrace past-entry
5467 @itemx set backtrace past-entry on
5468 Backtraces will continue past the internal entry point of an application.
5469 This entry point is encoded by the linker when the application is built,
5470 and is likely before the user entry point @code{main} (or equivalent) is called.
5472 @item set backtrace past-entry off
5473 Backtraces will stop when they encounter the internal entry point of an
5474 application. This is the default.
5476 @item show backtrace past-entry
5477 Display the current internal entry point backtrace policy.
5479 @item set backtrace limit @var{n}
5480 @itemx set backtrace limit 0
5481 @cindex backtrace limit
5482 Limit the backtrace to @var{n} levels. A value of zero means
5485 @item show backtrace limit
5486 Display the current limit on backtrace levels.
5490 @section Selecting a Frame
5492 Most commands for examining the stack and other data in your program work on
5493 whichever stack frame is selected at the moment. Here are the commands for
5494 selecting a stack frame; all of them finish by printing a brief description
5495 of the stack frame just selected.
5498 @kindex frame@r{, selecting}
5499 @kindex f @r{(@code{frame})}
5502 Select frame number @var{n}. Recall that frame zero is the innermost
5503 (currently executing) frame, frame one is the frame that called the
5504 innermost one, and so on. The highest-numbered frame is the one for
5507 @item frame @var{addr}
5509 Select the frame at address @var{addr}. This is useful mainly if the
5510 chaining of stack frames has been damaged by a bug, making it
5511 impossible for @value{GDBN} to assign numbers properly to all frames. In
5512 addition, this can be useful when your program has multiple stacks and
5513 switches between them.
5515 On the SPARC architecture, @code{frame} needs two addresses to
5516 select an arbitrary frame: a frame pointer and a stack pointer.
5518 On the MIPS and Alpha architecture, it needs two addresses: a stack
5519 pointer and a program counter.
5521 On the 29k architecture, it needs three addresses: a register stack
5522 pointer, a program counter, and a memory stack pointer.
5526 Move @var{n} frames up the stack. For positive numbers @var{n}, this
5527 advances toward the outermost frame, to higher frame numbers, to frames
5528 that have existed longer. @var{n} defaults to one.
5531 @kindex do @r{(@code{down})}
5533 Move @var{n} frames down the stack. For positive numbers @var{n}, this
5534 advances toward the innermost frame, to lower frame numbers, to frames
5535 that were created more recently. @var{n} defaults to one. You may
5536 abbreviate @code{down} as @code{do}.
5539 All of these commands end by printing two lines of output describing the
5540 frame. The first line shows the frame number, the function name, the
5541 arguments, and the source file and line number of execution in that
5542 frame. The second line shows the text of that source line.
5550 #1 0x22f0 in main (argc=1, argv=0xf7fffbf4, env=0xf7fffbfc)
5552 10 read_input_file (argv[i]);
5556 After such a printout, the @code{list} command with no arguments
5557 prints ten lines centered on the point of execution in the frame.
5558 You can also edit the program at the point of execution with your favorite
5559 editing program by typing @code{edit}.
5560 @xref{List, ,Printing Source Lines},
5564 @kindex down-silently
5566 @item up-silently @var{n}
5567 @itemx down-silently @var{n}
5568 These two commands are variants of @code{up} and @code{down},
5569 respectively; they differ in that they do their work silently, without
5570 causing display of the new frame. They are intended primarily for use
5571 in @value{GDBN} command scripts, where the output might be unnecessary and
5576 @section Information About a Frame
5578 There are several other commands to print information about the selected
5584 When used without any argument, this command does not change which
5585 frame is selected, but prints a brief description of the currently
5586 selected stack frame. It can be abbreviated @code{f}. With an
5587 argument, this command is used to select a stack frame.
5588 @xref{Selection, ,Selecting a Frame}.
5591 @kindex info f @r{(@code{info frame})}
5594 This command prints a verbose description of the selected stack frame,
5599 the address of the frame
5601 the address of the next frame down (called by this frame)
5603 the address of the next frame up (caller of this frame)
5605 the language in which the source code corresponding to this frame is written
5607 the address of the frame's arguments
5609 the address of the frame's local variables
5611 the program counter saved in it (the address of execution in the caller frame)
5613 which registers were saved in the frame
5616 @noindent The verbose description is useful when
5617 something has gone wrong that has made the stack format fail to fit
5618 the usual conventions.
5620 @item info frame @var{addr}
5621 @itemx info f @var{addr}
5622 Print a verbose description of the frame at address @var{addr}, without
5623 selecting that frame. The selected frame remains unchanged by this
5624 command. This requires the same kind of address (more than one for some
5625 architectures) that you specify in the @code{frame} command.
5626 @xref{Selection, ,Selecting a Frame}.
5630 Print the arguments of the selected frame, each on a separate line.
5634 Print the local variables of the selected frame, each on a separate
5635 line. These are all variables (declared either static or automatic)
5636 accessible at the point of execution of the selected frame.
5639 @cindex catch exceptions, list active handlers
5640 @cindex exception handlers, how to list
5642 Print a list of all the exception handlers that are active in the
5643 current stack frame at the current point of execution. To see other
5644 exception handlers, visit the associated frame (using the @code{up},
5645 @code{down}, or @code{frame} commands); then type @code{info catch}.
5646 @xref{Set Catchpoints, , Setting Catchpoints}.
5652 @chapter Examining Source Files
5654 @value{GDBN} can print parts of your program's source, since the debugging
5655 information recorded in the program tells @value{GDBN} what source files were
5656 used to build it. When your program stops, @value{GDBN} spontaneously prints
5657 the line where it stopped. Likewise, when you select a stack frame
5658 (@pxref{Selection, ,Selecting a Frame}), @value{GDBN} prints the line where
5659 execution in that frame has stopped. You can print other portions of
5660 source files by explicit command.
5662 If you use @value{GDBN} through its @sc{gnu} Emacs interface, you may
5663 prefer to use Emacs facilities to view source; see @ref{Emacs, ,Using
5664 @value{GDBN} under @sc{gnu} Emacs}.
5667 * List:: Printing source lines
5668 * Specify Location:: How to specify code locations
5669 * Edit:: Editing source files
5670 * Search:: Searching source files
5671 * Source Path:: Specifying source directories
5672 * Machine Code:: Source and machine code
5676 @section Printing Source Lines
5679 @kindex l @r{(@code{list})}
5680 To print lines from a source file, use the @code{list} command
5681 (abbreviated @code{l}). By default, ten lines are printed.
5682 There are several ways to specify what part of the file you want to
5683 print; see @ref{Specify Location}, for the full list.
5685 Here are the forms of the @code{list} command most commonly used:
5688 @item list @var{linenum}
5689 Print lines centered around line number @var{linenum} in the
5690 current source file.
5692 @item list @var{function}
5693 Print lines centered around the beginning of function
5697 Print more lines. If the last lines printed were printed with a
5698 @code{list} command, this prints lines following the last lines
5699 printed; however, if the last line printed was a solitary line printed
5700 as part of displaying a stack frame (@pxref{Stack, ,Examining the
5701 Stack}), this prints lines centered around that line.
5704 Print lines just before the lines last printed.
5707 @cindex @code{list}, how many lines to display
5708 By default, @value{GDBN} prints ten source lines with any of these forms of
5709 the @code{list} command. You can change this using @code{set listsize}:
5712 @kindex set listsize
5713 @item set listsize @var{count}
5714 Make the @code{list} command display @var{count} source lines (unless
5715 the @code{list} argument explicitly specifies some other number).
5717 @kindex show listsize
5719 Display the number of lines that @code{list} prints.
5722 Repeating a @code{list} command with @key{RET} discards the argument,
5723 so it is equivalent to typing just @code{list}. This is more useful
5724 than listing the same lines again. An exception is made for an
5725 argument of @samp{-}; that argument is preserved in repetition so that
5726 each repetition moves up in the source file.
5728 In general, the @code{list} command expects you to supply zero, one or two
5729 @dfn{linespecs}. Linespecs specify source lines; there are several ways
5730 of writing them (@pxref{Specify Location}), but the effect is always
5731 to specify some source line.
5733 Here is a complete description of the possible arguments for @code{list}:
5736 @item list @var{linespec}
5737 Print lines centered around the line specified by @var{linespec}.
5739 @item list @var{first},@var{last}
5740 Print lines from @var{first} to @var{last}. Both arguments are
5741 linespecs. When a @code{list} command has two linespecs, and the
5742 source file of the second linespec is omitted, this refers to
5743 the same source file as the first linespec.
5745 @item list ,@var{last}
5746 Print lines ending with @var{last}.
5748 @item list @var{first},
5749 Print lines starting with @var{first}.
5752 Print lines just after the lines last printed.
5755 Print lines just before the lines last printed.
5758 As described in the preceding table.
5761 @node Specify Location
5762 @section Specifying a Location
5763 @cindex specifying location
5766 Several @value{GDBN} commands accept arguments that specify a location
5767 of your program's code. Since @value{GDBN} is a source-level
5768 debugger, a location usually specifies some line in the source code;
5769 for that reason, locations are also known as @dfn{linespecs}.
5771 Here are all the different ways of specifying a code location that
5772 @value{GDBN} understands:
5776 Specifies the line number @var{linenum} of the current source file.
5779 @itemx +@var{offset}
5780 Specifies the line @var{offset} lines before or after the @dfn{current
5781 line}. For the @code{list} command, the current line is the last one
5782 printed; for the breakpoint commands, this is the line at which
5783 execution stopped in the currently selected @dfn{stack frame}
5784 (@pxref{Frames, ,Frames}, for a description of stack frames.) When
5785 used as the second of the two linespecs in a @code{list} command,
5786 this specifies the line @var{offset} lines up or down from the first
5789 @item @var{filename}:@var{linenum}
5790 Specifies the line @var{linenum} in the source file @var{filename}.
5792 @item @var{function}
5793 Specifies the line that begins the body of the function @var{function}.
5794 For example, in C, this is the line with the open brace.
5796 @item @var{filename}:@var{function}
5797 Specifies the line that begins the body of the function @var{function}
5798 in the file @var{filename}. You only need the file name with a
5799 function name to avoid ambiguity when there are identically named
5800 functions in different source files.
5802 @item *@var{address}
5803 Specifies the program address @var{address}. For line-oriented
5804 commands, such as @code{list} and @code{edit}, this specifies a source
5805 line that contains @var{address}. For @code{break} and other
5806 breakpoint oriented commands, this can be used to set breakpoints in
5807 parts of your program which do not have debugging information or
5810 Here @var{address} may be any expression valid in the current working
5811 language (@pxref{Languages, working language}) that specifies a code
5812 address. In addition, as a convenience, @value{GDBN} extends the
5813 semantics of expressions used in locations to cover the situations
5814 that frequently happen during debugging. Here are the various forms
5818 @item @var{expression}
5819 Any expression valid in the current working language.
5821 @item @var{funcaddr}
5822 An address of a function or procedure derived from its name. In C,
5823 C@t{++}, Java, Objective-C, Fortran, minimal, and assembly, this is
5824 simply the function's name @var{function} (and actually a special case
5825 of a valid expression). In Pascal and Modula-2, this is
5826 @code{&@var{function}}. In Ada, this is @code{@var{function}'Address}
5827 (although the Pascal form also works).
5829 This form specifies the address of the function's first instruction,
5830 before the stack frame and arguments have been set up.
5832 @item '@var{filename}'::@var{funcaddr}
5833 Like @var{funcaddr} above, but also specifies the name of the source
5834 file explicitly. This is useful if the name of the function does not
5835 specify the function unambiguously, e.g., if there are several
5836 functions with identical names in different source files.
5843 @section Editing Source Files
5844 @cindex editing source files
5847 @kindex e @r{(@code{edit})}
5848 To edit the lines in a source file, use the @code{edit} command.
5849 The editing program of your choice
5850 is invoked with the current line set to
5851 the active line in the program.
5852 Alternatively, there are several ways to specify what part of the file you
5853 want to print if you want to see other parts of the program:
5856 @item edit @var{location}
5857 Edit the source file specified by @code{location}. Editing starts at
5858 that @var{location}, e.g., at the specified source line of the
5859 specified file. @xref{Specify Location}, for all the possible forms
5860 of the @var{location} argument; here are the forms of the @code{edit}
5861 command most commonly used:
5864 @item edit @var{number}
5865 Edit the current source file with @var{number} as the active line number.
5867 @item edit @var{function}
5868 Edit the file containing @var{function} at the beginning of its definition.
5873 @subsection Choosing your Editor
5874 You can customize @value{GDBN} to use any editor you want
5876 The only restriction is that your editor (say @code{ex}), recognizes the
5877 following command-line syntax:
5879 ex +@var{number} file
5881 The optional numeric value +@var{number} specifies the number of the line in
5882 the file where to start editing.}.
5883 By default, it is @file{@value{EDITOR}}, but you can change this
5884 by setting the environment variable @code{EDITOR} before using
5885 @value{GDBN}. For example, to configure @value{GDBN} to use the
5886 @code{vi} editor, you could use these commands with the @code{sh} shell:
5892 or in the @code{csh} shell,
5894 setenv EDITOR /usr/bin/vi
5899 @section Searching Source Files
5900 @cindex searching source files
5902 There are two commands for searching through the current source file for a
5907 @kindex forward-search
5908 @item forward-search @var{regexp}
5909 @itemx search @var{regexp}
5910 The command @samp{forward-search @var{regexp}} checks each line,
5911 starting with the one following the last line listed, for a match for
5912 @var{regexp}. It lists the line that is found. You can use the
5913 synonym @samp{search @var{regexp}} or abbreviate the command name as
5916 @kindex reverse-search
5917 @item reverse-search @var{regexp}
5918 The command @samp{reverse-search @var{regexp}} checks each line, starting
5919 with the one before the last line listed and going backward, for a match
5920 for @var{regexp}. It lists the line that is found. You can abbreviate
5921 this command as @code{rev}.
5925 @section Specifying Source Directories
5928 @cindex directories for source files
5929 Executable programs sometimes do not record the directories of the source
5930 files from which they were compiled, just the names. Even when they do,
5931 the directories could be moved between the compilation and your debugging
5932 session. @value{GDBN} has a list of directories to search for source files;
5933 this is called the @dfn{source path}. Each time @value{GDBN} wants a source file,
5934 it tries all the directories in the list, in the order they are present
5935 in the list, until it finds a file with the desired name.
5937 For example, suppose an executable references the file
5938 @file{/usr/src/foo-1.0/lib/foo.c}, and our source path is
5939 @file{/mnt/cross}. The file is first looked up literally; if this
5940 fails, @file{/mnt/cross/usr/src/foo-1.0/lib/foo.c} is tried; if this
5941 fails, @file{/mnt/cross/foo.c} is opened; if this fails, an error
5942 message is printed. @value{GDBN} does not look up the parts of the
5943 source file name, such as @file{/mnt/cross/src/foo-1.0/lib/foo.c}.
5944 Likewise, the subdirectories of the source path are not searched: if
5945 the source path is @file{/mnt/cross}, and the binary refers to
5946 @file{foo.c}, @value{GDBN} would not find it under
5947 @file{/mnt/cross/usr/src/foo-1.0/lib}.
5949 Plain file names, relative file names with leading directories, file
5950 names containing dots, etc.@: are all treated as described above; for
5951 instance, if the source path is @file{/mnt/cross}, and the source file
5952 is recorded as @file{../lib/foo.c}, @value{GDBN} would first try
5953 @file{../lib/foo.c}, then @file{/mnt/cross/../lib/foo.c}, and after
5954 that---@file{/mnt/cross/foo.c}.
5956 Note that the executable search path is @emph{not} used to locate the
5959 Whenever you reset or rearrange the source path, @value{GDBN} clears out
5960 any information it has cached about where source files are found and where
5961 each line is in the file.
5965 When you start @value{GDBN}, its source path includes only @samp{cdir}
5966 and @samp{cwd}, in that order.
5967 To add other directories, use the @code{directory} command.
5969 The search path is used to find both program source files and @value{GDBN}
5970 script files (read using the @samp{-command} option and @samp{source} command).
5972 In addition to the source path, @value{GDBN} provides a set of commands
5973 that manage a list of source path substitution rules. A @dfn{substitution
5974 rule} specifies how to rewrite source directories stored in the program's
5975 debug information in case the sources were moved to a different
5976 directory between compilation and debugging. A rule is made of
5977 two strings, the first specifying what needs to be rewritten in
5978 the path, and the second specifying how it should be rewritten.
5979 In @ref{set substitute-path}, we name these two parts @var{from} and
5980 @var{to} respectively. @value{GDBN} does a simple string replacement
5981 of @var{from} with @var{to} at the start of the directory part of the
5982 source file name, and uses that result instead of the original file
5983 name to look up the sources.
5985 Using the previous example, suppose the @file{foo-1.0} tree has been
5986 moved from @file{/usr/src} to @file{/mnt/cross}, then you can tell
5987 @value{GDBN} to replace @file{/usr/src} in all source path names with
5988 @file{/mnt/cross}. The first lookup will then be
5989 @file{/mnt/cross/foo-1.0/lib/foo.c} in place of the original location
5990 of @file{/usr/src/foo-1.0/lib/foo.c}. To define a source path
5991 substitution rule, use the @code{set substitute-path} command
5992 (@pxref{set substitute-path}).
5994 To avoid unexpected substitution results, a rule is applied only if the
5995 @var{from} part of the directory name ends at a directory separator.
5996 For instance, a rule substituting @file{/usr/source} into
5997 @file{/mnt/cross} will be applied to @file{/usr/source/foo-1.0} but
5998 not to @file{/usr/sourceware/foo-2.0}. And because the substitution
5999 is applied only at the beginning of the directory name, this rule will
6000 not be applied to @file{/root/usr/source/baz.c} either.
6002 In many cases, you can achieve the same result using the @code{directory}
6003 command. However, @code{set substitute-path} can be more efficient in
6004 the case where the sources are organized in a complex tree with multiple
6005 subdirectories. With the @code{directory} command, you need to add each
6006 subdirectory of your project. If you moved the entire tree while
6007 preserving its internal organization, then @code{set substitute-path}
6008 allows you to direct the debugger to all the sources with one single
6011 @code{set substitute-path} is also more than just a shortcut command.
6012 The source path is only used if the file at the original location no
6013 longer exists. On the other hand, @code{set substitute-path} modifies
6014 the debugger behavior to look at the rewritten location instead. So, if
6015 for any reason a source file that is not relevant to your executable is
6016 located at the original location, a substitution rule is the only
6017 method available to point @value{GDBN} at the new location.
6019 @cindex @samp{--with-relocated-sources}
6020 @cindex default source path substitution
6021 You can configure a default source path substitution rule by
6022 configuring @value{GDBN} with the
6023 @samp{--with-relocated-sources=@var{dir}} option. The @var{dir}
6024 should be the name of a directory under @value{GDBN}'s configured
6025 prefix (set with @samp{--prefix} or @samp{--exec-prefix}), and
6026 directory names in debug information under @var{dir} will be adjusted
6027 automatically if the installed @value{GDBN} is moved to a new
6028 location. This is useful if @value{GDBN}, libraries or executables
6029 with debug information and corresponding source code are being moved
6033 @item directory @var{dirname} @dots{}
6034 @item dir @var{dirname} @dots{}
6035 Add directory @var{dirname} to the front of the source path. Several
6036 directory names may be given to this command, separated by @samp{:}
6037 (@samp{;} on MS-DOS and MS-Windows, where @samp{:} usually appears as
6038 part of absolute file names) or
6039 whitespace. You may specify a directory that is already in the source
6040 path; this moves it forward, so @value{GDBN} searches it sooner.
6044 @vindex $cdir@r{, convenience variable}
6045 @vindex $cwd@r{, convenience variable}
6046 @cindex compilation directory
6047 @cindex current directory
6048 @cindex working directory
6049 @cindex directory, current
6050 @cindex directory, compilation
6051 You can use the string @samp{$cdir} to refer to the compilation
6052 directory (if one is recorded), and @samp{$cwd} to refer to the current
6053 working directory. @samp{$cwd} is not the same as @samp{.}---the former
6054 tracks the current working directory as it changes during your @value{GDBN}
6055 session, while the latter is immediately expanded to the current
6056 directory at the time you add an entry to the source path.
6059 Reset the source path to its default value (@samp{$cdir:$cwd} on Unix systems). This requires confirmation.
6061 @c RET-repeat for @code{directory} is explicitly disabled, but since
6062 @c repeating it would be a no-op we do not say that. (thanks to RMS)
6064 @item show directories
6065 @kindex show directories
6066 Print the source path: show which directories it contains.
6068 @anchor{set substitute-path}
6069 @item set substitute-path @var{from} @var{to}
6070 @kindex set substitute-path
6071 Define a source path substitution rule, and add it at the end of the
6072 current list of existing substitution rules. If a rule with the same
6073 @var{from} was already defined, then the old rule is also deleted.
6075 For example, if the file @file{/foo/bar/baz.c} was moved to
6076 @file{/mnt/cross/baz.c}, then the command
6079 (@value{GDBP}) set substitute-path /usr/src /mnt/cross
6083 will tell @value{GDBN} to replace @samp{/usr/src} with
6084 @samp{/mnt/cross}, which will allow @value{GDBN} to find the file
6085 @file{baz.c} even though it was moved.
6087 In the case when more than one substitution rule have been defined,
6088 the rules are evaluated one by one in the order where they have been
6089 defined. The first one matching, if any, is selected to perform
6092 For instance, if we had entered the following commands:
6095 (@value{GDBP}) set substitute-path /usr/src/include /mnt/include
6096 (@value{GDBP}) set substitute-path /usr/src /mnt/src
6100 @value{GDBN} would then rewrite @file{/usr/src/include/defs.h} into
6101 @file{/mnt/include/defs.h} by using the first rule. However, it would
6102 use the second rule to rewrite @file{/usr/src/lib/foo.c} into
6103 @file{/mnt/src/lib/foo.c}.
6106 @item unset substitute-path [path]
6107 @kindex unset substitute-path
6108 If a path is specified, search the current list of substitution rules
6109 for a rule that would rewrite that path. Delete that rule if found.
6110 A warning is emitted by the debugger if no rule could be found.
6112 If no path is specified, then all substitution rules are deleted.
6114 @item show substitute-path [path]
6115 @kindex show substitute-path
6116 If a path is specified, then print the source path substitution rule
6117 which would rewrite that path, if any.
6119 If no path is specified, then print all existing source path substitution
6124 If your source path is cluttered with directories that are no longer of
6125 interest, @value{GDBN} may sometimes cause confusion by finding the wrong
6126 versions of source. You can correct the situation as follows:
6130 Use @code{directory} with no argument to reset the source path to its default value.
6133 Use @code{directory} with suitable arguments to reinstall the
6134 directories you want in the source path. You can add all the
6135 directories in one command.
6139 @section Source and Machine Code
6140 @cindex source line and its code address
6142 You can use the command @code{info line} to map source lines to program
6143 addresses (and vice versa), and the command @code{disassemble} to display
6144 a range of addresses as machine instructions. You can use the command
6145 @code{set disassemble-next-line} to set whether to disassemble next
6146 source line when execution stops. When run under @sc{gnu} Emacs
6147 mode, the @code{info line} command causes the arrow to point to the
6148 line specified. Also, @code{info line} prints addresses in symbolic form as
6153 @item info line @var{linespec}
6154 Print the starting and ending addresses of the compiled code for
6155 source line @var{linespec}. You can specify source lines in any of
6156 the ways documented in @ref{Specify Location}.
6159 For example, we can use @code{info line} to discover the location of
6160 the object code for the first line of function
6161 @code{m4_changequote}:
6163 @c FIXME: I think this example should also show the addresses in
6164 @c symbolic form, as they usually would be displayed.
6166 (@value{GDBP}) info line m4_changequote
6167 Line 895 of "builtin.c" starts at pc 0x634c and ends at 0x6350.
6171 @cindex code address and its source line
6172 We can also inquire (using @code{*@var{addr}} as the form for
6173 @var{linespec}) what source line covers a particular address:
6175 (@value{GDBP}) info line *0x63ff
6176 Line 926 of "builtin.c" starts at pc 0x63e4 and ends at 0x6404.
6179 @cindex @code{$_} and @code{info line}
6180 @cindex @code{x} command, default address
6181 @kindex x@r{(examine), and} info line
6182 After @code{info line}, the default address for the @code{x} command
6183 is changed to the starting address of the line, so that @samp{x/i} is
6184 sufficient to begin examining the machine code (@pxref{Memory,
6185 ,Examining Memory}). Also, this address is saved as the value of the
6186 convenience variable @code{$_} (@pxref{Convenience Vars, ,Convenience
6191 @cindex assembly instructions
6192 @cindex instructions, assembly
6193 @cindex machine instructions
6194 @cindex listing machine instructions
6196 @itemx disassemble /m
6197 This specialized command dumps a range of memory as machine
6198 instructions. It can also print mixed source+disassembly by specifying
6199 the @code{/m} modifier.
6200 The default memory range is the function surrounding the
6201 program counter of the selected frame. A single argument to this
6202 command is a program counter value; @value{GDBN} dumps the function
6203 surrounding this value. Two arguments specify a range of addresses
6204 (first inclusive, second exclusive) to dump.
6207 The following example shows the disassembly of a range of addresses of
6208 HP PA-RISC 2.0 code:
6211 (@value{GDBP}) disas 0x32c4 0x32e4
6212 Dump of assembler code from 0x32c4 to 0x32e4:
6213 0x32c4 <main+204>: addil 0,dp
6214 0x32c8 <main+208>: ldw 0x22c(sr0,r1),r26
6215 0x32cc <main+212>: ldil 0x3000,r31
6216 0x32d0 <main+216>: ble 0x3f8(sr4,r31)
6217 0x32d4 <main+220>: ldo 0(r31),rp
6218 0x32d8 <main+224>: addil -0x800,dp
6219 0x32dc <main+228>: ldo 0x588(r1),r26
6220 0x32e0 <main+232>: ldil 0x3000,r31
6221 End of assembler dump.
6224 Here is an example showing mixed source+assembly for Intel x86:
6227 (@value{GDBP}) disas /m main
6228 Dump of assembler code for function main:
6230 0x08048330 <main+0>: push %ebp
6231 0x08048331 <main+1>: mov %esp,%ebp
6232 0x08048333 <main+3>: sub $0x8,%esp
6233 0x08048336 <main+6>: and $0xfffffff0,%esp
6234 0x08048339 <main+9>: sub $0x10,%esp
6236 6 printf ("Hello.\n");
6237 0x0804833c <main+12>: movl $0x8048440,(%esp)
6238 0x08048343 <main+19>: call 0x8048284 <puts@@plt>
6242 0x08048348 <main+24>: mov $0x0,%eax
6243 0x0804834d <main+29>: leave
6244 0x0804834e <main+30>: ret
6246 End of assembler dump.
6249 Some architectures have more than one commonly-used set of instruction
6250 mnemonics or other syntax.
6252 For programs that were dynamically linked and use shared libraries,
6253 instructions that call functions or branch to locations in the shared
6254 libraries might show a seemingly bogus location---it's actually a
6255 location of the relocation table. On some architectures, @value{GDBN}
6256 might be able to resolve these to actual function names.
6259 @kindex set disassembly-flavor
6260 @cindex Intel disassembly flavor
6261 @cindex AT&T disassembly flavor
6262 @item set disassembly-flavor @var{instruction-set}
6263 Select the instruction set to use when disassembling the
6264 program via the @code{disassemble} or @code{x/i} commands.
6266 Currently this command is only defined for the Intel x86 family. You
6267 can set @var{instruction-set} to either @code{intel} or @code{att}.
6268 The default is @code{att}, the AT&T flavor used by default by Unix
6269 assemblers for x86-based targets.
6271 @kindex show disassembly-flavor
6272 @item show disassembly-flavor
6273 Show the current setting of the disassembly flavor.
6277 @kindex set disassemble-next-line
6278 @kindex show disassemble-next-line
6279 @item set disassemble-next-line
6280 @itemx show disassemble-next-line
6281 Control whether or not @value{GDBN} will disassemble the next source
6282 line or instruction when execution stops. If ON, @value{GDBN} will
6283 display disassembly of the next source line when execution of the
6284 program being debugged stops. This is @emph{in addition} to
6285 displaying the source line itself, which @value{GDBN} always does if
6286 possible. If the next source line cannot be displayed for some reason
6287 (e.g., if @value{GDBN} cannot find the source file, or there's no line
6288 info in the debug info), @value{GDBN} will display disassembly of the
6289 next @emph{instruction} instead of showing the next source line. If
6290 AUTO, @value{GDBN} will display disassembly of next instruction only
6291 if the source line cannot be displayed. This setting causes
6292 @value{GDBN} to display some feedback when you step through a function
6293 with no line info or whose source file is unavailable. The default is
6294 OFF, which means never display the disassembly of the next line or
6300 @chapter Examining Data
6302 @cindex printing data
6303 @cindex examining data
6306 @c "inspect" is not quite a synonym if you are using Epoch, which we do not
6307 @c document because it is nonstandard... Under Epoch it displays in a
6308 @c different window or something like that.
6309 The usual way to examine data in your program is with the @code{print}
6310 command (abbreviated @code{p}), or its synonym @code{inspect}. It
6311 evaluates and prints the value of an expression of the language your
6312 program is written in (@pxref{Languages, ,Using @value{GDBN} with
6313 Different Languages}).
6316 @item print @var{expr}
6317 @itemx print /@var{f} @var{expr}
6318 @var{expr} is an expression (in the source language). By default the
6319 value of @var{expr} is printed in a format appropriate to its data type;
6320 you can choose a different format by specifying @samp{/@var{f}}, where
6321 @var{f} is a letter specifying the format; see @ref{Output Formats,,Output
6325 @itemx print /@var{f}
6326 @cindex reprint the last value
6327 If you omit @var{expr}, @value{GDBN} displays the last value again (from the
6328 @dfn{value history}; @pxref{Value History, ,Value History}). This allows you to
6329 conveniently inspect the same value in an alternative format.
6332 A more low-level way of examining data is with the @code{x} command.
6333 It examines data in memory at a specified address and prints it in a
6334 specified format. @xref{Memory, ,Examining Memory}.
6336 If you are interested in information about types, or about how the
6337 fields of a struct or a class are declared, use the @code{ptype @var{exp}}
6338 command rather than @code{print}. @xref{Symbols, ,Examining the Symbol
6342 * Expressions:: Expressions
6343 * Ambiguous Expressions:: Ambiguous Expressions
6344 * Variables:: Program variables
6345 * Arrays:: Artificial arrays
6346 * Output Formats:: Output formats
6347 * Memory:: Examining memory
6348 * Auto Display:: Automatic display
6349 * Print Settings:: Print settings
6350 * Value History:: Value history
6351 * Convenience Vars:: Convenience variables
6352 * Registers:: Registers
6353 * Floating Point Hardware:: Floating point hardware
6354 * Vector Unit:: Vector Unit
6355 * OS Information:: Auxiliary data provided by operating system
6356 * Memory Region Attributes:: Memory region attributes
6357 * Dump/Restore Files:: Copy between memory and a file
6358 * Core File Generation:: Cause a program dump its core
6359 * Character Sets:: Debugging programs that use a different
6360 character set than GDB does
6361 * Caching Remote Data:: Data caching for remote targets
6362 * Searching Memory:: Searching memory for a sequence of bytes
6366 @section Expressions
6369 @code{print} and many other @value{GDBN} commands accept an expression and
6370 compute its value. Any kind of constant, variable or operator defined
6371 by the programming language you are using is valid in an expression in
6372 @value{GDBN}. This includes conditional expressions, function calls,
6373 casts, and string constants. It also includes preprocessor macros, if
6374 you compiled your program to include this information; see
6377 @cindex arrays in expressions
6378 @value{GDBN} supports array constants in expressions input by
6379 the user. The syntax is @{@var{element}, @var{element}@dots{}@}. For example,
6380 you can use the command @code{print @{1, 2, 3@}} to create an array
6381 of three integers. If you pass an array to a function or assign it
6382 to a program variable, @value{GDBN} copies the array to memory that
6383 is @code{malloc}ed in the target program.
6385 Because C is so widespread, most of the expressions shown in examples in
6386 this manual are in C. @xref{Languages, , Using @value{GDBN} with Different
6387 Languages}, for information on how to use expressions in other
6390 In this section, we discuss operators that you can use in @value{GDBN}
6391 expressions regardless of your programming language.
6393 @cindex casts, in expressions
6394 Casts are supported in all languages, not just in C, because it is so
6395 useful to cast a number into a pointer in order to examine a structure
6396 at that address in memory.
6397 @c FIXME: casts supported---Mod2 true?
6399 @value{GDBN} supports these operators, in addition to those common
6400 to programming languages:
6404 @samp{@@} is a binary operator for treating parts of memory as arrays.
6405 @xref{Arrays, ,Artificial Arrays}, for more information.
6408 @samp{::} allows you to specify a variable in terms of the file or
6409 function where it is defined. @xref{Variables, ,Program Variables}.
6411 @cindex @{@var{type}@}
6412 @cindex type casting memory
6413 @cindex memory, viewing as typed object
6414 @cindex casts, to view memory
6415 @item @{@var{type}@} @var{addr}
6416 Refers to an object of type @var{type} stored at address @var{addr} in
6417 memory. @var{addr} may be any expression whose value is an integer or
6418 pointer (but parentheses are required around binary operators, just as in
6419 a cast). This construct is allowed regardless of what kind of data is
6420 normally supposed to reside at @var{addr}.
6423 @node Ambiguous Expressions
6424 @section Ambiguous Expressions
6425 @cindex ambiguous expressions
6427 Expressions can sometimes contain some ambiguous elements. For instance,
6428 some programming languages (notably Ada, C@t{++} and Objective-C) permit
6429 a single function name to be defined several times, for application in
6430 different contexts. This is called @dfn{overloading}. Another example
6431 involving Ada is generics. A @dfn{generic package} is similar to C@t{++}
6432 templates and is typically instantiated several times, resulting in
6433 the same function name being defined in different contexts.
6435 In some cases and depending on the language, it is possible to adjust
6436 the expression to remove the ambiguity. For instance in C@t{++}, you
6437 can specify the signature of the function you want to break on, as in
6438 @kbd{break @var{function}(@var{types})}. In Ada, using the fully
6439 qualified name of your function often makes the expression unambiguous
6442 When an ambiguity that needs to be resolved is detected, the debugger
6443 has the capability to display a menu of numbered choices for each
6444 possibility, and then waits for the selection with the prompt @samp{>}.
6445 The first option is always @samp{[0] cancel}, and typing @kbd{0 @key{RET}}
6446 aborts the current command. If the command in which the expression was
6447 used allows more than one choice to be selected, the next option in the
6448 menu is @samp{[1] all}, and typing @kbd{1 @key{RET}} selects all possible
6451 For example, the following session excerpt shows an attempt to set a
6452 breakpoint at the overloaded symbol @code{String::after}.
6453 We choose three particular definitions of that function name:
6455 @c FIXME! This is likely to change to show arg type lists, at least
6458 (@value{GDBP}) b String::after
6461 [2] file:String.cc; line number:867
6462 [3] file:String.cc; line number:860
6463 [4] file:String.cc; line number:875
6464 [5] file:String.cc; line number:853
6465 [6] file:String.cc; line number:846
6466 [7] file:String.cc; line number:735
6468 Breakpoint 1 at 0xb26c: file String.cc, line 867.
6469 Breakpoint 2 at 0xb344: file String.cc, line 875.
6470 Breakpoint 3 at 0xafcc: file String.cc, line 846.
6471 Multiple breakpoints were set.
6472 Use the "delete" command to delete unwanted
6479 @kindex set multiple-symbols
6480 @item set multiple-symbols @var{mode}
6481 @cindex multiple-symbols menu
6483 This option allows you to adjust the debugger behavior when an expression
6486 By default, @var{mode} is set to @code{all}. If the command with which
6487 the expression is used allows more than one choice, then @value{GDBN}
6488 automatically selects all possible choices. For instance, inserting
6489 a breakpoint on a function using an ambiguous name results in a breakpoint
6490 inserted on each possible match. However, if a unique choice must be made,
6491 then @value{GDBN} uses the menu to help you disambiguate the expression.
6492 For instance, printing the address of an overloaded function will result
6493 in the use of the menu.
6495 When @var{mode} is set to @code{ask}, the debugger always uses the menu
6496 when an ambiguity is detected.
6498 Finally, when @var{mode} is set to @code{cancel}, the debugger reports
6499 an error due to the ambiguity and the command is aborted.
6501 @kindex show multiple-symbols
6502 @item show multiple-symbols
6503 Show the current value of the @code{multiple-symbols} setting.
6507 @section Program Variables
6509 The most common kind of expression to use is the name of a variable
6512 Variables in expressions are understood in the selected stack frame
6513 (@pxref{Selection, ,Selecting a Frame}); they must be either:
6517 global (or file-static)
6524 visible according to the scope rules of the
6525 programming language from the point of execution in that frame
6528 @noindent This means that in the function
6543 you can examine and use the variable @code{a} whenever your program is
6544 executing within the function @code{foo}, but you can only use or
6545 examine the variable @code{b} while your program is executing inside
6546 the block where @code{b} is declared.
6548 @cindex variable name conflict
6549 There is an exception: you can refer to a variable or function whose
6550 scope is a single source file even if the current execution point is not
6551 in this file. But it is possible to have more than one such variable or
6552 function with the same name (in different source files). If that
6553 happens, referring to that name has unpredictable effects. If you wish,
6554 you can specify a static variable in a particular function or file,
6555 using the colon-colon (@code{::}) notation:
6557 @cindex colon-colon, context for variables/functions
6559 @c info cannot cope with a :: index entry, but why deprive hard copy readers?
6560 @cindex @code{::}, context for variables/functions
6563 @var{file}::@var{variable}
6564 @var{function}::@var{variable}
6568 Here @var{file} or @var{function} is the name of the context for the
6569 static @var{variable}. In the case of file names, you can use quotes to
6570 make sure @value{GDBN} parses the file name as a single word---for example,
6571 to print a global value of @code{x} defined in @file{f2.c}:
6574 (@value{GDBP}) p 'f2.c'::x
6577 @cindex C@t{++} scope resolution
6578 This use of @samp{::} is very rarely in conflict with the very similar
6579 use of the same notation in C@t{++}. @value{GDBN} also supports use of the C@t{++}
6580 scope resolution operator in @value{GDBN} expressions.
6581 @c FIXME: Um, so what happens in one of those rare cases where it's in
6584 @cindex wrong values
6585 @cindex variable values, wrong
6586 @cindex function entry/exit, wrong values of variables
6587 @cindex optimized code, wrong values of variables
6589 @emph{Warning:} Occasionally, a local variable may appear to have the
6590 wrong value at certain points in a function---just after entry to a new
6591 scope, and just before exit.
6593 You may see this problem when you are stepping by machine instructions.
6594 This is because, on most machines, it takes more than one instruction to
6595 set up a stack frame (including local variable definitions); if you are
6596 stepping by machine instructions, variables may appear to have the wrong
6597 values until the stack frame is completely built. On exit, it usually
6598 also takes more than one machine instruction to destroy a stack frame;
6599 after you begin stepping through that group of instructions, local
6600 variable definitions may be gone.
6602 This may also happen when the compiler does significant optimizations.
6603 To be sure of always seeing accurate values, turn off all optimization
6606 @cindex ``No symbol "foo" in current context''
6607 Another possible effect of compiler optimizations is to optimize
6608 unused variables out of existence, or assign variables to registers (as
6609 opposed to memory addresses). Depending on the support for such cases
6610 offered by the debug info format used by the compiler, @value{GDBN}
6611 might not be able to display values for such local variables. If that
6612 happens, @value{GDBN} will print a message like this:
6615 No symbol "foo" in current context.
6618 To solve such problems, either recompile without optimizations, or use a
6619 different debug info format, if the compiler supports several such
6620 formats. For example, @value{NGCC}, the @sc{gnu} C/C@t{++} compiler,
6621 usually supports the @option{-gstabs+} option. @option{-gstabs+}
6622 produces debug info in a format that is superior to formats such as
6623 COFF. You may be able to use DWARF 2 (@option{-gdwarf-2}), which is also
6624 an effective form for debug info. @xref{Debugging Options,,Options
6625 for Debugging Your Program or GCC, gcc.info, Using the @sc{gnu}
6626 Compiler Collection (GCC)}.
6627 @xref{C, ,C and C@t{++}}, for more information about debug info formats
6628 that are best suited to C@t{++} programs.
6630 If you ask to print an object whose contents are unknown to
6631 @value{GDBN}, e.g., because its data type is not completely specified
6632 by the debug information, @value{GDBN} will say @samp{<incomplete
6633 type>}. @xref{Symbols, incomplete type}, for more about this.
6635 Strings are identified as arrays of @code{char} values without specified
6636 signedness. Arrays of either @code{signed char} or @code{unsigned char} get
6637 printed as arrays of 1 byte sized integers. @code{-fsigned-char} or
6638 @code{-funsigned-char} @value{NGCC} options have no effect as @value{GDBN}
6639 defines literal string type @code{"char"} as @code{char} without a sign.
6644 signed char var1[] = "A";
6647 You get during debugging
6652 $2 = @{65 'A', 0 '\0'@}
6656 @section Artificial Arrays
6658 @cindex artificial array
6660 @kindex @@@r{, referencing memory as an array}
6661 It is often useful to print out several successive objects of the
6662 same type in memory; a section of an array, or an array of
6663 dynamically determined size for which only a pointer exists in the
6666 You can do this by referring to a contiguous span of memory as an
6667 @dfn{artificial array}, using the binary operator @samp{@@}. The left
6668 operand of @samp{@@} should be the first element of the desired array
6669 and be an individual object. The right operand should be the desired length
6670 of the array. The result is an array value whose elements are all of
6671 the type of the left argument. The first element is actually the left
6672 argument; the second element comes from bytes of memory immediately
6673 following those that hold the first element, and so on. Here is an
6674 example. If a program says
6677 int *array = (int *) malloc (len * sizeof (int));
6681 you can print the contents of @code{array} with
6687 The left operand of @samp{@@} must reside in memory. Array values made
6688 with @samp{@@} in this way behave just like other arrays in terms of
6689 subscripting, and are coerced to pointers when used in expressions.
6690 Artificial arrays most often appear in expressions via the value history
6691 (@pxref{Value History, ,Value History}), after printing one out.
6693 Another way to create an artificial array is to use a cast.
6694 This re-interprets a value as if it were an array.
6695 The value need not be in memory:
6697 (@value{GDBP}) p/x (short[2])0x12345678
6698 $1 = @{0x1234, 0x5678@}
6701 As a convenience, if you leave the array length out (as in
6702 @samp{(@var{type}[])@var{value}}) @value{GDBN} calculates the size to fill
6703 the value (as @samp{sizeof(@var{value})/sizeof(@var{type})}:
6705 (@value{GDBP}) p/x (short[])0x12345678
6706 $2 = @{0x1234, 0x5678@}
6709 Sometimes the artificial array mechanism is not quite enough; in
6710 moderately complex data structures, the elements of interest may not
6711 actually be adjacent---for example, if you are interested in the values
6712 of pointers in an array. One useful work-around in this situation is
6713 to use a convenience variable (@pxref{Convenience Vars, ,Convenience
6714 Variables}) as a counter in an expression that prints the first
6715 interesting value, and then repeat that expression via @key{RET}. For
6716 instance, suppose you have an array @code{dtab} of pointers to
6717 structures, and you are interested in the values of a field @code{fv}
6718 in each structure. Here is an example of what you might type:
6728 @node Output Formats
6729 @section Output Formats
6731 @cindex formatted output
6732 @cindex output formats
6733 By default, @value{GDBN} prints a value according to its data type. Sometimes
6734 this is not what you want. For example, you might want to print a number
6735 in hex, or a pointer in decimal. Or you might want to view data in memory
6736 at a certain address as a character string or as an instruction. To do
6737 these things, specify an @dfn{output format} when you print a value.
6739 The simplest use of output formats is to say how to print a value
6740 already computed. This is done by starting the arguments of the
6741 @code{print} command with a slash and a format letter. The format
6742 letters supported are:
6746 Regard the bits of the value as an integer, and print the integer in
6750 Print as integer in signed decimal.
6753 Print as integer in unsigned decimal.
6756 Print as integer in octal.
6759 Print as integer in binary. The letter @samp{t} stands for ``two''.
6760 @footnote{@samp{b} cannot be used because these format letters are also
6761 used with the @code{x} command, where @samp{b} stands for ``byte'';
6762 see @ref{Memory,,Examining Memory}.}
6765 @cindex unknown address, locating
6766 @cindex locate address
6767 Print as an address, both absolute in hexadecimal and as an offset from
6768 the nearest preceding symbol. You can use this format used to discover
6769 where (in what function) an unknown address is located:
6772 (@value{GDBP}) p/a 0x54320
6773 $3 = 0x54320 <_initialize_vx+396>
6777 The command @code{info symbol 0x54320} yields similar results.
6778 @xref{Symbols, info symbol}.
6781 Regard as an integer and print it as a character constant. This
6782 prints both the numerical value and its character representation. The
6783 character representation is replaced with the octal escape @samp{\nnn}
6784 for characters outside the 7-bit @sc{ascii} range.
6786 Without this format, @value{GDBN} displays @code{char},
6787 @w{@code{unsigned char}}, and @w{@code{signed char}} data as character
6788 constants. Single-byte members of vectors are displayed as integer
6792 Regard the bits of the value as a floating point number and print
6793 using typical floating point syntax.
6796 @cindex printing strings
6797 @cindex printing byte arrays
6798 Regard as a string, if possible. With this format, pointers to single-byte
6799 data are displayed as null-terminated strings and arrays of single-byte data
6800 are displayed as fixed-length strings. Other values are displayed in their
6803 Without this format, @value{GDBN} displays pointers to and arrays of
6804 @code{char}, @w{@code{unsigned char}}, and @w{@code{signed char}} as
6805 strings. Single-byte members of a vector are displayed as an integer
6809 @cindex raw printing
6810 Print using the @samp{raw} formatting. By default, @value{GDBN} will
6811 use a type-specific pretty-printer. The @samp{r} format bypasses any
6812 pretty-printer which might exist for the value's type.
6815 For example, to print the program counter in hex (@pxref{Registers}), type
6822 Note that no space is required before the slash; this is because command
6823 names in @value{GDBN} cannot contain a slash.
6825 To reprint the last value in the value history with a different format,
6826 you can use the @code{print} command with just a format and no
6827 expression. For example, @samp{p/x} reprints the last value in hex.
6830 @section Examining Memory
6832 You can use the command @code{x} (for ``examine'') to examine memory in
6833 any of several formats, independently of your program's data types.
6835 @cindex examining memory
6837 @kindex x @r{(examine memory)}
6838 @item x/@var{nfu} @var{addr}
6841 Use the @code{x} command to examine memory.
6844 @var{n}, @var{f}, and @var{u} are all optional parameters that specify how
6845 much memory to display and how to format it; @var{addr} is an
6846 expression giving the address where you want to start displaying memory.
6847 If you use defaults for @var{nfu}, you need not type the slash @samp{/}.
6848 Several commands set convenient defaults for @var{addr}.
6851 @item @var{n}, the repeat count
6852 The repeat count is a decimal integer; the default is 1. It specifies
6853 how much memory (counting by units @var{u}) to display.
6854 @c This really is **decimal**; unaffected by 'set radix' as of GDB
6857 @item @var{f}, the display format
6858 The display format is one of the formats used by @code{print}
6859 (@samp{x}, @samp{d}, @samp{u}, @samp{o}, @samp{t}, @samp{a}, @samp{c},
6860 @samp{f}, @samp{s}), and in addition @samp{i} (for machine instructions).
6861 The default is @samp{x} (hexadecimal) initially. The default changes
6862 each time you use either @code{x} or @code{print}.
6864 @item @var{u}, the unit size
6865 The unit size is any of
6871 Halfwords (two bytes).
6873 Words (four bytes). This is the initial default.
6875 Giant words (eight bytes).
6878 Each time you specify a unit size with @code{x}, that size becomes the
6879 default unit the next time you use @code{x}. (For the @samp{s} and
6880 @samp{i} formats, the unit size is ignored and is normally not written.)
6882 @item @var{addr}, starting display address
6883 @var{addr} is the address where you want @value{GDBN} to begin displaying
6884 memory. The expression need not have a pointer value (though it may);
6885 it is always interpreted as an integer address of a byte of memory.
6886 @xref{Expressions, ,Expressions}, for more information on expressions. The default for
6887 @var{addr} is usually just after the last address examined---but several
6888 other commands also set the default address: @code{info breakpoints} (to
6889 the address of the last breakpoint listed), @code{info line} (to the
6890 starting address of a line), and @code{print} (if you use it to display
6891 a value from memory).
6894 For example, @samp{x/3uh 0x54320} is a request to display three halfwords
6895 (@code{h}) of memory, formatted as unsigned decimal integers (@samp{u}),
6896 starting at address @code{0x54320}. @samp{x/4xw $sp} prints the four
6897 words (@samp{w}) of memory above the stack pointer (here, @samp{$sp};
6898 @pxref{Registers, ,Registers}) in hexadecimal (@samp{x}).
6900 Since the letters indicating unit sizes are all distinct from the
6901 letters specifying output formats, you do not have to remember whether
6902 unit size or format comes first; either order works. The output
6903 specifications @samp{4xw} and @samp{4wx} mean exactly the same thing.
6904 (However, the count @var{n} must come first; @samp{wx4} does not work.)
6906 Even though the unit size @var{u} is ignored for the formats @samp{s}
6907 and @samp{i}, you might still want to use a count @var{n}; for example,
6908 @samp{3i} specifies that you want to see three machine instructions,
6909 including any operands. For convenience, especially when used with
6910 the @code{display} command, the @samp{i} format also prints branch delay
6911 slot instructions, if any, beyond the count specified, which immediately
6912 follow the last instruction that is within the count. The command
6913 @code{disassemble} gives an alternative way of inspecting machine
6914 instructions; see @ref{Machine Code,,Source and Machine Code}.
6916 All the defaults for the arguments to @code{x} are designed to make it
6917 easy to continue scanning memory with minimal specifications each time
6918 you use @code{x}. For example, after you have inspected three machine
6919 instructions with @samp{x/3i @var{addr}}, you can inspect the next seven
6920 with just @samp{x/7}. If you use @key{RET} to repeat the @code{x} command,
6921 the repeat count @var{n} is used again; the other arguments default as
6922 for successive uses of @code{x}.
6924 @cindex @code{$_}, @code{$__}, and value history
6925 The addresses and contents printed by the @code{x} command are not saved
6926 in the value history because there is often too much of them and they
6927 would get in the way. Instead, @value{GDBN} makes these values available for
6928 subsequent use in expressions as values of the convenience variables
6929 @code{$_} and @code{$__}. After an @code{x} command, the last address
6930 examined is available for use in expressions in the convenience variable
6931 @code{$_}. The contents of that address, as examined, are available in
6932 the convenience variable @code{$__}.
6934 If the @code{x} command has a repeat count, the address and contents saved
6935 are from the last memory unit printed; this is not the same as the last
6936 address printed if several units were printed on the last line of output.
6938 @cindex remote memory comparison
6939 @cindex verify remote memory image
6940 When you are debugging a program running on a remote target machine
6941 (@pxref{Remote Debugging}), you may wish to verify the program's image in the
6942 remote machine's memory against the executable file you downloaded to
6943 the target. The @code{compare-sections} command is provided for such
6947 @kindex compare-sections
6948 @item compare-sections @r{[}@var{section-name}@r{]}
6949 Compare the data of a loadable section @var{section-name} in the
6950 executable file of the program being debugged with the same section in
6951 the remote machine's memory, and report any mismatches. With no
6952 arguments, compares all loadable sections. This command's
6953 availability depends on the target's support for the @code{"qCRC"}
6958 @section Automatic Display
6959 @cindex automatic display
6960 @cindex display of expressions
6962 If you find that you want to print the value of an expression frequently
6963 (to see how it changes), you might want to add it to the @dfn{automatic
6964 display list} so that @value{GDBN} prints its value each time your program stops.
6965 Each expression added to the list is given a number to identify it;
6966 to remove an expression from the list, you specify that number.
6967 The automatic display looks like this:
6971 3: bar[5] = (struct hack *) 0x3804
6975 This display shows item numbers, expressions and their current values. As with
6976 displays you request manually using @code{x} or @code{print}, you can
6977 specify the output format you prefer; in fact, @code{display} decides
6978 whether to use @code{print} or @code{x} depending your format
6979 specification---it uses @code{x} if you specify either the @samp{i}
6980 or @samp{s} format, or a unit size; otherwise it uses @code{print}.
6984 @item display @var{expr}
6985 Add the expression @var{expr} to the list of expressions to display
6986 each time your program stops. @xref{Expressions, ,Expressions}.
6988 @code{display} does not repeat if you press @key{RET} again after using it.
6990 @item display/@var{fmt} @var{expr}
6991 For @var{fmt} specifying only a display format and not a size or
6992 count, add the expression @var{expr} to the auto-display list but
6993 arrange to display it each time in the specified format @var{fmt}.
6994 @xref{Output Formats,,Output Formats}.
6996 @item display/@var{fmt} @var{addr}
6997 For @var{fmt} @samp{i} or @samp{s}, or including a unit-size or a
6998 number of units, add the expression @var{addr} as a memory address to
6999 be examined each time your program stops. Examining means in effect
7000 doing @samp{x/@var{fmt} @var{addr}}. @xref{Memory, ,Examining Memory}.
7003 For example, @samp{display/i $pc} can be helpful, to see the machine
7004 instruction about to be executed each time execution stops (@samp{$pc}
7005 is a common name for the program counter; @pxref{Registers, ,Registers}).
7008 @kindex delete display
7010 @item undisplay @var{dnums}@dots{}
7011 @itemx delete display @var{dnums}@dots{}
7012 Remove item numbers @var{dnums} from the list of expressions to display.
7014 @code{undisplay} does not repeat if you press @key{RET} after using it.
7015 (Otherwise you would just get the error @samp{No display number @dots{}}.)
7017 @kindex disable display
7018 @item disable display @var{dnums}@dots{}
7019 Disable the display of item numbers @var{dnums}. A disabled display
7020 item is not printed automatically, but is not forgotten. It may be
7021 enabled again later.
7023 @kindex enable display
7024 @item enable display @var{dnums}@dots{}
7025 Enable display of item numbers @var{dnums}. It becomes effective once
7026 again in auto display of its expression, until you specify otherwise.
7029 Display the current values of the expressions on the list, just as is
7030 done when your program stops.
7032 @kindex info display
7034 Print the list of expressions previously set up to display
7035 automatically, each one with its item number, but without showing the
7036 values. This includes disabled expressions, which are marked as such.
7037 It also includes expressions which would not be displayed right now
7038 because they refer to automatic variables not currently available.
7041 @cindex display disabled out of scope
7042 If a display expression refers to local variables, then it does not make
7043 sense outside the lexical context for which it was set up. Such an
7044 expression is disabled when execution enters a context where one of its
7045 variables is not defined. For example, if you give the command
7046 @code{display last_char} while inside a function with an argument
7047 @code{last_char}, @value{GDBN} displays this argument while your program
7048 continues to stop inside that function. When it stops elsewhere---where
7049 there is no variable @code{last_char}---the display is disabled
7050 automatically. The next time your program stops where @code{last_char}
7051 is meaningful, you can enable the display expression once again.
7053 @node Print Settings
7054 @section Print Settings
7056 @cindex format options
7057 @cindex print settings
7058 @value{GDBN} provides the following ways to control how arrays, structures,
7059 and symbols are printed.
7062 These settings are useful for debugging programs in any language:
7066 @item set print address
7067 @itemx set print address on
7068 @cindex print/don't print memory addresses
7069 @value{GDBN} prints memory addresses showing the location of stack
7070 traces, structure values, pointer values, breakpoints, and so forth,
7071 even when it also displays the contents of those addresses. The default
7072 is @code{on}. For example, this is what a stack frame display looks like with
7073 @code{set print address on}:
7078 #0 set_quotes (lq=0x34c78 "<<", rq=0x34c88 ">>")
7080 530 if (lquote != def_lquote)
7084 @item set print address off
7085 Do not print addresses when displaying their contents. For example,
7086 this is the same stack frame displayed with @code{set print address off}:
7090 (@value{GDBP}) set print addr off
7092 #0 set_quotes (lq="<<", rq=">>") at input.c:530
7093 530 if (lquote != def_lquote)
7097 You can use @samp{set print address off} to eliminate all machine
7098 dependent displays from the @value{GDBN} interface. For example, with
7099 @code{print address off}, you should get the same text for backtraces on
7100 all machines---whether or not they involve pointer arguments.
7103 @item show print address
7104 Show whether or not addresses are to be printed.
7107 When @value{GDBN} prints a symbolic address, it normally prints the
7108 closest earlier symbol plus an offset. If that symbol does not uniquely
7109 identify the address (for example, it is a name whose scope is a single
7110 source file), you may need to clarify. One way to do this is with
7111 @code{info line}, for example @samp{info line *0x4537}. Alternately,
7112 you can set @value{GDBN} to print the source file and line number when
7113 it prints a symbolic address:
7116 @item set print symbol-filename on
7117 @cindex source file and line of a symbol
7118 @cindex symbol, source file and line
7119 Tell @value{GDBN} to print the source file name and line number of a
7120 symbol in the symbolic form of an address.
7122 @item set print symbol-filename off
7123 Do not print source file name and line number of a symbol. This is the
7126 @item show print symbol-filename
7127 Show whether or not @value{GDBN} will print the source file name and
7128 line number of a symbol in the symbolic form of an address.
7131 Another situation where it is helpful to show symbol filenames and line
7132 numbers is when disassembling code; @value{GDBN} shows you the line
7133 number and source file that corresponds to each instruction.
7135 Also, you may wish to see the symbolic form only if the address being
7136 printed is reasonably close to the closest earlier symbol:
7139 @item set print max-symbolic-offset @var{max-offset}
7140 @cindex maximum value for offset of closest symbol
7141 Tell @value{GDBN} to only display the symbolic form of an address if the
7142 offset between the closest earlier symbol and the address is less than
7143 @var{max-offset}. The default is 0, which tells @value{GDBN}
7144 to always print the symbolic form of an address if any symbol precedes it.
7146 @item show print max-symbolic-offset
7147 Ask how large the maximum offset is that @value{GDBN} prints in a
7151 @cindex wild pointer, interpreting
7152 @cindex pointer, finding referent
7153 If you have a pointer and you are not sure where it points, try
7154 @samp{set print symbol-filename on}. Then you can determine the name
7155 and source file location of the variable where it points, using
7156 @samp{p/a @var{pointer}}. This interprets the address in symbolic form.
7157 For example, here @value{GDBN} shows that a variable @code{ptt} points
7158 at another variable @code{t}, defined in @file{hi2.c}:
7161 (@value{GDBP}) set print symbol-filename on
7162 (@value{GDBP}) p/a ptt
7163 $4 = 0xe008 <t in hi2.c>
7167 @emph{Warning:} For pointers that point to a local variable, @samp{p/a}
7168 does not show the symbol name and filename of the referent, even with
7169 the appropriate @code{set print} options turned on.
7172 Other settings control how different kinds of objects are printed:
7175 @item set print array
7176 @itemx set print array on
7177 @cindex pretty print arrays
7178 Pretty print arrays. This format is more convenient to read,
7179 but uses more space. The default is off.
7181 @item set print array off
7182 Return to compressed format for arrays.
7184 @item show print array
7185 Show whether compressed or pretty format is selected for displaying
7188 @cindex print array indexes
7189 @item set print array-indexes
7190 @itemx set print array-indexes on
7191 Print the index of each element when displaying arrays. May be more
7192 convenient to locate a given element in the array or quickly find the
7193 index of a given element in that printed array. The default is off.
7195 @item set print array-indexes off
7196 Stop printing element indexes when displaying arrays.
7198 @item show print array-indexes
7199 Show whether the index of each element is printed when displaying
7202 @item set print elements @var{number-of-elements}
7203 @cindex number of array elements to print
7204 @cindex limit on number of printed array elements
7205 Set a limit on how many elements of an array @value{GDBN} will print.
7206 If @value{GDBN} is printing a large array, it stops printing after it has
7207 printed the number of elements set by the @code{set print elements} command.
7208 This limit also applies to the display of strings.
7209 When @value{GDBN} starts, this limit is set to 200.
7210 Setting @var{number-of-elements} to zero means that the printing is unlimited.
7212 @item show print elements
7213 Display the number of elements of a large array that @value{GDBN} will print.
7214 If the number is 0, then the printing is unlimited.
7216 @item set print frame-arguments @var{value}
7217 @kindex set print frame-arguments
7218 @cindex printing frame argument values
7219 @cindex print all frame argument values
7220 @cindex print frame argument values for scalars only
7221 @cindex do not print frame argument values
7222 This command allows to control how the values of arguments are printed
7223 when the debugger prints a frame (@pxref{Frames}). The possible
7228 The values of all arguments are printed.
7231 Print the value of an argument only if it is a scalar. The value of more
7232 complex arguments such as arrays, structures, unions, etc, is replaced
7233 by @code{@dots{}}. This is the default. Here is an example where
7234 only scalar arguments are shown:
7237 #1 0x08048361 in call_me (i=3, s=@dots{}, ss=0xbf8d508c, u=@dots{}, e=green)
7242 None of the argument values are printed. Instead, the value of each argument
7243 is replaced by @code{@dots{}}. In this case, the example above now becomes:
7246 #1 0x08048361 in call_me (i=@dots{}, s=@dots{}, ss=@dots{}, u=@dots{}, e=@dots{})
7251 By default, only scalar arguments are printed. This command can be used
7252 to configure the debugger to print the value of all arguments, regardless
7253 of their type. However, it is often advantageous to not print the value
7254 of more complex parameters. For instance, it reduces the amount of
7255 information printed in each frame, making the backtrace more readable.
7256 Also, it improves performance when displaying Ada frames, because
7257 the computation of large arguments can sometimes be CPU-intensive,
7258 especially in large applications. Setting @code{print frame-arguments}
7259 to @code{scalars} (the default) or @code{none} avoids this computation,
7260 thus speeding up the display of each Ada frame.
7262 @item show print frame-arguments
7263 Show how the value of arguments should be displayed when printing a frame.
7265 @item set print repeats
7266 @cindex repeated array elements
7267 Set the threshold for suppressing display of repeated array
7268 elements. When the number of consecutive identical elements of an
7269 array exceeds the threshold, @value{GDBN} prints the string
7270 @code{"<repeats @var{n} times>"}, where @var{n} is the number of
7271 identical repetitions, instead of displaying the identical elements
7272 themselves. Setting the threshold to zero will cause all elements to
7273 be individually printed. The default threshold is 10.
7275 @item show print repeats
7276 Display the current threshold for printing repeated identical
7279 @item set print null-stop
7280 @cindex @sc{null} elements in arrays
7281 Cause @value{GDBN} to stop printing the characters of an array when the first
7282 @sc{null} is encountered. This is useful when large arrays actually
7283 contain only short strings.
7286 @item show print null-stop
7287 Show whether @value{GDBN} stops printing an array on the first
7288 @sc{null} character.
7290 @item set print pretty on
7291 @cindex print structures in indented form
7292 @cindex indentation in structure display
7293 Cause @value{GDBN} to print structures in an indented format with one member
7294 per line, like this:
7309 @item set print pretty off
7310 Cause @value{GDBN} to print structures in a compact format, like this:
7314 $1 = @{next = 0x0, flags = @{sweet = 1, sour = 1@}, \
7315 meat = 0x54 "Pork"@}
7320 This is the default format.
7322 @item show print pretty
7323 Show which format @value{GDBN} is using to print structures.
7325 @item set print sevenbit-strings on
7326 @cindex eight-bit characters in strings
7327 @cindex octal escapes in strings
7328 Print using only seven-bit characters; if this option is set,
7329 @value{GDBN} displays any eight-bit characters (in strings or
7330 character values) using the notation @code{\}@var{nnn}. This setting is
7331 best if you are working in English (@sc{ascii}) and you use the
7332 high-order bit of characters as a marker or ``meta'' bit.
7334 @item set print sevenbit-strings off
7335 Print full eight-bit characters. This allows the use of more
7336 international character sets, and is the default.
7338 @item show print sevenbit-strings
7339 Show whether or not @value{GDBN} is printing only seven-bit characters.
7341 @item set print union on
7342 @cindex unions in structures, printing
7343 Tell @value{GDBN} to print unions which are contained in structures
7344 and other unions. This is the default setting.
7346 @item set print union off
7347 Tell @value{GDBN} not to print unions which are contained in
7348 structures and other unions. @value{GDBN} will print @code{"@{...@}"}
7351 @item show print union
7352 Ask @value{GDBN} whether or not it will print unions which are contained in
7353 structures and other unions.
7355 For example, given the declarations
7358 typedef enum @{Tree, Bug@} Species;
7359 typedef enum @{Big_tree, Acorn, Seedling@} Tree_forms;
7360 typedef enum @{Caterpillar, Cocoon, Butterfly@}
7371 struct thing foo = @{Tree, @{Acorn@}@};
7375 with @code{set print union on} in effect @samp{p foo} would print
7378 $1 = @{it = Tree, form = @{tree = Acorn, bug = Cocoon@}@}
7382 and with @code{set print union off} in effect it would print
7385 $1 = @{it = Tree, form = @{...@}@}
7389 @code{set print union} affects programs written in C-like languages
7395 These settings are of interest when debugging C@t{++} programs:
7398 @cindex demangling C@t{++} names
7399 @item set print demangle
7400 @itemx set print demangle on
7401 Print C@t{++} names in their source form rather than in the encoded
7402 (``mangled'') form passed to the assembler and linker for type-safe
7403 linkage. The default is on.
7405 @item show print demangle
7406 Show whether C@t{++} names are printed in mangled or demangled form.
7408 @item set print asm-demangle
7409 @itemx set print asm-demangle on
7410 Print C@t{++} names in their source form rather than their mangled form, even
7411 in assembler code printouts such as instruction disassemblies.
7414 @item show print asm-demangle
7415 Show whether C@t{++} names in assembly listings are printed in mangled
7418 @cindex C@t{++} symbol decoding style
7419 @cindex symbol decoding style, C@t{++}
7420 @kindex set demangle-style
7421 @item set demangle-style @var{style}
7422 Choose among several encoding schemes used by different compilers to
7423 represent C@t{++} names. The choices for @var{style} are currently:
7427 Allow @value{GDBN} to choose a decoding style by inspecting your program.
7430 Decode based on the @sc{gnu} C@t{++} compiler (@code{g++}) encoding algorithm.
7431 This is the default.
7434 Decode based on the HP ANSI C@t{++} (@code{aCC}) encoding algorithm.
7437 Decode based on the Lucid C@t{++} compiler (@code{lcc}) encoding algorithm.
7440 Decode using the algorithm in the @cite{C@t{++} Annotated Reference Manual}.
7441 @strong{Warning:} this setting alone is not sufficient to allow
7442 debugging @code{cfront}-generated executables. @value{GDBN} would
7443 require further enhancement to permit that.
7446 If you omit @var{style}, you will see a list of possible formats.
7448 @item show demangle-style
7449 Display the encoding style currently in use for decoding C@t{++} symbols.
7451 @item set print object
7452 @itemx set print object on
7453 @cindex derived type of an object, printing
7454 @cindex display derived types
7455 When displaying a pointer to an object, identify the @emph{actual}
7456 (derived) type of the object rather than the @emph{declared} type, using
7457 the virtual function table.
7459 @item set print object off
7460 Display only the declared type of objects, without reference to the
7461 virtual function table. This is the default setting.
7463 @item show print object
7464 Show whether actual, or declared, object types are displayed.
7466 @item set print static-members
7467 @itemx set print static-members on
7468 @cindex static members of C@t{++} objects
7469 Print static members when displaying a C@t{++} object. The default is on.
7471 @item set print static-members off
7472 Do not print static members when displaying a C@t{++} object.
7474 @item show print static-members
7475 Show whether C@t{++} static members are printed or not.
7477 @item set print pascal_static-members
7478 @itemx set print pascal_static-members on
7479 @cindex static members of Pascal objects
7480 @cindex Pascal objects, static members display
7481 Print static members when displaying a Pascal object. The default is on.
7483 @item set print pascal_static-members off
7484 Do not print static members when displaying a Pascal object.
7486 @item show print pascal_static-members
7487 Show whether Pascal static members are printed or not.
7489 @c These don't work with HP ANSI C++ yet.
7490 @item set print vtbl
7491 @itemx set print vtbl on
7492 @cindex pretty print C@t{++} virtual function tables
7493 @cindex virtual functions (C@t{++}) display
7494 @cindex VTBL display
7495 Pretty print C@t{++} virtual function tables. The default is off.
7496 (The @code{vtbl} commands do not work on programs compiled with the HP
7497 ANSI C@t{++} compiler (@code{aCC}).)
7499 @item set print vtbl off
7500 Do not pretty print C@t{++} virtual function tables.
7502 @item show print vtbl
7503 Show whether C@t{++} virtual function tables are pretty printed, or not.
7507 @section Value History
7509 @cindex value history
7510 @cindex history of values printed by @value{GDBN}
7511 Values printed by the @code{print} command are saved in the @value{GDBN}
7512 @dfn{value history}. This allows you to refer to them in other expressions.
7513 Values are kept until the symbol table is re-read or discarded
7514 (for example with the @code{file} or @code{symbol-file} commands).
7515 When the symbol table changes, the value history is discarded,
7516 since the values may contain pointers back to the types defined in the
7521 @cindex history number
7522 The values printed are given @dfn{history numbers} by which you can
7523 refer to them. These are successive integers starting with one.
7524 @code{print} shows you the history number assigned to a value by
7525 printing @samp{$@var{num} = } before the value; here @var{num} is the
7528 To refer to any previous value, use @samp{$} followed by the value's
7529 history number. The way @code{print} labels its output is designed to
7530 remind you of this. Just @code{$} refers to the most recent value in
7531 the history, and @code{$$} refers to the value before that.
7532 @code{$$@var{n}} refers to the @var{n}th value from the end; @code{$$2}
7533 is the value just prior to @code{$$}, @code{$$1} is equivalent to
7534 @code{$$}, and @code{$$0} is equivalent to @code{$}.
7536 For example, suppose you have just printed a pointer to a structure and
7537 want to see the contents of the structure. It suffices to type
7543 If you have a chain of structures where the component @code{next} points
7544 to the next one, you can print the contents of the next one with this:
7551 You can print successive links in the chain by repeating this
7552 command---which you can do by just typing @key{RET}.
7554 Note that the history records values, not expressions. If the value of
7555 @code{x} is 4 and you type these commands:
7563 then the value recorded in the value history by the @code{print} command
7564 remains 4 even though the value of @code{x} has changed.
7569 Print the last ten values in the value history, with their item numbers.
7570 This is like @samp{p@ $$9} repeated ten times, except that @code{show
7571 values} does not change the history.
7573 @item show values @var{n}
7574 Print ten history values centered on history item number @var{n}.
7577 Print ten history values just after the values last printed. If no more
7578 values are available, @code{show values +} produces no display.
7581 Pressing @key{RET} to repeat @code{show values @var{n}} has exactly the
7582 same effect as @samp{show values +}.
7584 @node Convenience Vars
7585 @section Convenience Variables
7587 @cindex convenience variables
7588 @cindex user-defined variables
7589 @value{GDBN} provides @dfn{convenience variables} that you can use within
7590 @value{GDBN} to hold on to a value and refer to it later. These variables
7591 exist entirely within @value{GDBN}; they are not part of your program, and
7592 setting a convenience variable has no direct effect on further execution
7593 of your program. That is why you can use them freely.
7595 Convenience variables are prefixed with @samp{$}. Any name preceded by
7596 @samp{$} can be used for a convenience variable, unless it is one of
7597 the predefined machine-specific register names (@pxref{Registers, ,Registers}).
7598 (Value history references, in contrast, are @emph{numbers} preceded
7599 by @samp{$}. @xref{Value History, ,Value History}.)
7601 You can save a value in a convenience variable with an assignment
7602 expression, just as you would set a variable in your program.
7606 set $foo = *object_ptr
7610 would save in @code{$foo} the value contained in the object pointed to by
7613 Using a convenience variable for the first time creates it, but its
7614 value is @code{void} until you assign a new value. You can alter the
7615 value with another assignment at any time.
7617 Convenience variables have no fixed types. You can assign a convenience
7618 variable any type of value, including structures and arrays, even if
7619 that variable already has a value of a different type. The convenience
7620 variable, when used as an expression, has the type of its current value.
7623 @kindex show convenience
7624 @cindex show all user variables
7625 @item show convenience
7626 Print a list of convenience variables used so far, and their values.
7627 Abbreviated @code{show conv}.
7629 @kindex init-if-undefined
7630 @cindex convenience variables, initializing
7631 @item init-if-undefined $@var{variable} = @var{expression}
7632 Set a convenience variable if it has not already been set. This is useful
7633 for user-defined commands that keep some state. It is similar, in concept,
7634 to using local static variables with initializers in C (except that
7635 convenience variables are global). It can also be used to allow users to
7636 override default values used in a command script.
7638 If the variable is already defined then the expression is not evaluated so
7639 any side-effects do not occur.
7642 One of the ways to use a convenience variable is as a counter to be
7643 incremented or a pointer to be advanced. For example, to print
7644 a field from successive elements of an array of structures:
7648 print bar[$i++]->contents
7652 Repeat that command by typing @key{RET}.
7654 Some convenience variables are created automatically by @value{GDBN} and given
7655 values likely to be useful.
7658 @vindex $_@r{, convenience variable}
7660 The variable @code{$_} is automatically set by the @code{x} command to
7661 the last address examined (@pxref{Memory, ,Examining Memory}). Other
7662 commands which provide a default address for @code{x} to examine also
7663 set @code{$_} to that address; these commands include @code{info line}
7664 and @code{info breakpoint}. The type of @code{$_} is @code{void *}
7665 except when set by the @code{x} command, in which case it is a pointer
7666 to the type of @code{$__}.
7668 @vindex $__@r{, convenience variable}
7670 The variable @code{$__} is automatically set by the @code{x} command
7671 to the value found in the last address examined. Its type is chosen
7672 to match the format in which the data was printed.
7675 @vindex $_exitcode@r{, convenience variable}
7676 The variable @code{$_exitcode} is automatically set to the exit code when
7677 the program being debugged terminates.
7680 @vindex $_siginfo@r{, convenience variable}
7681 The variable @code{$_siginfo} is bound to extra signal information
7682 inspection (@pxref{extra signal information}).
7685 On HP-UX systems, if you refer to a function or variable name that
7686 begins with a dollar sign, @value{GDBN} searches for a user or system
7687 name first, before it searches for a convenience variable.
7689 @cindex convenience functions
7690 @value{GDBN} also supplies some @dfn{convenience functions}. These
7691 have a syntax similar to convenience variables. A convenience
7692 function can be used in an expression just like an ordinary function;
7693 however, a convenience function is implemented internally to
7698 @kindex help function
7699 @cindex show all convenience functions
7700 Print a list of all convenience functions.
7707 You can refer to machine register contents, in expressions, as variables
7708 with names starting with @samp{$}. The names of registers are different
7709 for each machine; use @code{info registers} to see the names used on
7713 @kindex info registers
7714 @item info registers
7715 Print the names and values of all registers except floating-point
7716 and vector registers (in the selected stack frame).
7718 @kindex info all-registers
7719 @cindex floating point registers
7720 @item info all-registers
7721 Print the names and values of all registers, including floating-point
7722 and vector registers (in the selected stack frame).
7724 @item info registers @var{regname} @dots{}
7725 Print the @dfn{relativized} value of each specified register @var{regname}.
7726 As discussed in detail below, register values are normally relative to
7727 the selected stack frame. @var{regname} may be any register name valid on
7728 the machine you are using, with or without the initial @samp{$}.
7731 @cindex stack pointer register
7732 @cindex program counter register
7733 @cindex process status register
7734 @cindex frame pointer register
7735 @cindex standard registers
7736 @value{GDBN} has four ``standard'' register names that are available (in
7737 expressions) on most machines---whenever they do not conflict with an
7738 architecture's canonical mnemonics for registers. The register names
7739 @code{$pc} and @code{$sp} are used for the program counter register and
7740 the stack pointer. @code{$fp} is used for a register that contains a
7741 pointer to the current stack frame, and @code{$ps} is used for a
7742 register that contains the processor status. For example,
7743 you could print the program counter in hex with
7750 or print the instruction to be executed next with
7757 or add four to the stack pointer@footnote{This is a way of removing
7758 one word from the stack, on machines where stacks grow downward in
7759 memory (most machines, nowadays). This assumes that the innermost
7760 stack frame is selected; setting @code{$sp} is not allowed when other
7761 stack frames are selected. To pop entire frames off the stack,
7762 regardless of machine architecture, use @code{return};
7763 see @ref{Returning, ,Returning from a Function}.} with
7769 Whenever possible, these four standard register names are available on
7770 your machine even though the machine has different canonical mnemonics,
7771 so long as there is no conflict. The @code{info registers} command
7772 shows the canonical names. For example, on the SPARC, @code{info
7773 registers} displays the processor status register as @code{$psr} but you
7774 can also refer to it as @code{$ps}; and on x86-based machines @code{$ps}
7775 is an alias for the @sc{eflags} register.
7777 @value{GDBN} always considers the contents of an ordinary register as an
7778 integer when the register is examined in this way. Some machines have
7779 special registers which can hold nothing but floating point; these
7780 registers are considered to have floating point values. There is no way
7781 to refer to the contents of an ordinary register as floating point value
7782 (although you can @emph{print} it as a floating point value with
7783 @samp{print/f $@var{regname}}).
7785 Some registers have distinct ``raw'' and ``virtual'' data formats. This
7786 means that the data format in which the register contents are saved by
7787 the operating system is not the same one that your program normally
7788 sees. For example, the registers of the 68881 floating point
7789 coprocessor are always saved in ``extended'' (raw) format, but all C
7790 programs expect to work with ``double'' (virtual) format. In such
7791 cases, @value{GDBN} normally works with the virtual format only (the format
7792 that makes sense for your program), but the @code{info registers} command
7793 prints the data in both formats.
7795 @cindex SSE registers (x86)
7796 @cindex MMX registers (x86)
7797 Some machines have special registers whose contents can be interpreted
7798 in several different ways. For example, modern x86-based machines
7799 have SSE and MMX registers that can hold several values packed
7800 together in several different formats. @value{GDBN} refers to such
7801 registers in @code{struct} notation:
7804 (@value{GDBP}) print $xmm1
7806 v4_float = @{0, 3.43859137e-038, 1.54142831e-044, 1.821688e-044@},
7807 v2_double = @{9.92129282474342e-303, 2.7585945287983262e-313@},
7808 v16_int8 = "\000\000\000\000\3706;\001\v\000\000\000\r\000\000",
7809 v8_int16 = @{0, 0, 14072, 315, 11, 0, 13, 0@},
7810 v4_int32 = @{0, 20657912, 11, 13@},
7811 v2_int64 = @{88725056443645952, 55834574859@},
7812 uint128 = 0x0000000d0000000b013b36f800000000
7817 To set values of such registers, you need to tell @value{GDBN} which
7818 view of the register you wish to change, as if you were assigning
7819 value to a @code{struct} member:
7822 (@value{GDBP}) set $xmm1.uint128 = 0x000000000000000000000000FFFFFFFF
7825 Normally, register values are relative to the selected stack frame
7826 (@pxref{Selection, ,Selecting a Frame}). This means that you get the
7827 value that the register would contain if all stack frames farther in
7828 were exited and their saved registers restored. In order to see the
7829 true contents of hardware registers, you must select the innermost
7830 frame (with @samp{frame 0}).
7832 However, @value{GDBN} must deduce where registers are saved, from the machine
7833 code generated by your compiler. If some registers are not saved, or if
7834 @value{GDBN} is unable to locate the saved registers, the selected stack
7835 frame makes no difference.
7837 @node Floating Point Hardware
7838 @section Floating Point Hardware
7839 @cindex floating point
7841 Depending on the configuration, @value{GDBN} may be able to give
7842 you more information about the status of the floating point hardware.
7847 Display hardware-dependent information about the floating
7848 point unit. The exact contents and layout vary depending on the
7849 floating point chip. Currently, @samp{info float} is supported on
7850 the ARM and x86 machines.
7854 @section Vector Unit
7857 Depending on the configuration, @value{GDBN} may be able to give you
7858 more information about the status of the vector unit.
7863 Display information about the vector unit. The exact contents and
7864 layout vary depending on the hardware.
7867 @node OS Information
7868 @section Operating System Auxiliary Information
7869 @cindex OS information
7871 @value{GDBN} provides interfaces to useful OS facilities that can help
7872 you debug your program.
7874 @cindex @code{ptrace} system call
7875 @cindex @code{struct user} contents
7876 When @value{GDBN} runs on a @dfn{Posix system} (such as GNU or Unix
7877 machines), it interfaces with the inferior via the @code{ptrace}
7878 system call. The operating system creates a special sata structure,
7879 called @code{struct user}, for this interface. You can use the
7880 command @code{info udot} to display the contents of this data
7886 Display the contents of the @code{struct user} maintained by the OS
7887 kernel for the program being debugged. @value{GDBN} displays the
7888 contents of @code{struct user} as a list of hex numbers, similar to
7889 the @code{examine} command.
7892 @cindex auxiliary vector
7893 @cindex vector, auxiliary
7894 Some operating systems supply an @dfn{auxiliary vector} to programs at
7895 startup. This is akin to the arguments and environment that you
7896 specify for a program, but contains a system-dependent variety of
7897 binary values that tell system libraries important details about the
7898 hardware, operating system, and process. Each value's purpose is
7899 identified by an integer tag; the meanings are well-known but system-specific.
7900 Depending on the configuration and operating system facilities,
7901 @value{GDBN} may be able to show you this information. For remote
7902 targets, this functionality may further depend on the remote stub's
7903 support of the @samp{qXfer:auxv:read} packet, see
7904 @ref{qXfer auxiliary vector read}.
7909 Display the auxiliary vector of the inferior, which can be either a
7910 live process or a core dump file. @value{GDBN} prints each tag value
7911 numerically, and also shows names and text descriptions for recognized
7912 tags. Some values in the vector are numbers, some bit masks, and some
7913 pointers to strings or other data. @value{GDBN} displays each value in the
7914 most appropriate form for a recognized tag, and in hexadecimal for
7915 an unrecognized tag.
7918 On some targets, @value{GDBN} can access operating-system-specific information
7919 and display it to user, without interpretation. For remote targets,
7920 this functionality depends on the remote stub's support of the
7921 @samp{qXfer:osdata:read} packet, see @ref{qXfer osdata read}.
7924 @kindex info os processes
7925 @item info os processes
7926 Display the list of processes on the target. For each process,
7927 @value{GDBN} prints the process identifier, the name of the user, and
7928 the command corresponding to the process.
7931 @node Memory Region Attributes
7932 @section Memory Region Attributes
7933 @cindex memory region attributes
7935 @dfn{Memory region attributes} allow you to describe special handling
7936 required by regions of your target's memory. @value{GDBN} uses
7937 attributes to determine whether to allow certain types of memory
7938 accesses; whether to use specific width accesses; and whether to cache
7939 target memory. By default the description of memory regions is
7940 fetched from the target (if the current target supports this), but the
7941 user can override the fetched regions.
7943 Defined memory regions can be individually enabled and disabled. When a
7944 memory region is disabled, @value{GDBN} uses the default attributes when
7945 accessing memory in that region. Similarly, if no memory regions have
7946 been defined, @value{GDBN} uses the default attributes when accessing
7949 When a memory region is defined, it is given a number to identify it;
7950 to enable, disable, or remove a memory region, you specify that number.
7954 @item mem @var{lower} @var{upper} @var{attributes}@dots{}
7955 Define a memory region bounded by @var{lower} and @var{upper} with
7956 attributes @var{attributes}@dots{}, and add it to the list of regions
7957 monitored by @value{GDBN}. Note that @var{upper} == 0 is a special
7958 case: it is treated as the target's maximum memory address.
7959 (0xffff on 16 bit targets, 0xffffffff on 32 bit targets, etc.)
7962 Discard any user changes to the memory regions and use target-supplied
7963 regions, if available, or no regions if the target does not support.
7966 @item delete mem @var{nums}@dots{}
7967 Remove memory regions @var{nums}@dots{} from the list of regions
7968 monitored by @value{GDBN}.
7971 @item disable mem @var{nums}@dots{}
7972 Disable monitoring of memory regions @var{nums}@dots{}.
7973 A disabled memory region is not forgotten.
7974 It may be enabled again later.
7977 @item enable mem @var{nums}@dots{}
7978 Enable monitoring of memory regions @var{nums}@dots{}.
7982 Print a table of all defined memory regions, with the following columns
7986 @item Memory Region Number
7987 @item Enabled or Disabled.
7988 Enabled memory regions are marked with @samp{y}.
7989 Disabled memory regions are marked with @samp{n}.
7992 The address defining the inclusive lower bound of the memory region.
7995 The address defining the exclusive upper bound of the memory region.
7998 The list of attributes set for this memory region.
8003 @subsection Attributes
8005 @subsubsection Memory Access Mode
8006 The access mode attributes set whether @value{GDBN} may make read or
8007 write accesses to a memory region.
8009 While these attributes prevent @value{GDBN} from performing invalid
8010 memory accesses, they do nothing to prevent the target system, I/O DMA,
8011 etc.@: from accessing memory.
8015 Memory is read only.
8017 Memory is write only.
8019 Memory is read/write. This is the default.
8022 @subsubsection Memory Access Size
8023 The access size attribute tells @value{GDBN} to use specific sized
8024 accesses in the memory region. Often memory mapped device registers
8025 require specific sized accesses. If no access size attribute is
8026 specified, @value{GDBN} may use accesses of any size.
8030 Use 8 bit memory accesses.
8032 Use 16 bit memory accesses.
8034 Use 32 bit memory accesses.
8036 Use 64 bit memory accesses.
8039 @c @subsubsection Hardware/Software Breakpoints
8040 @c The hardware/software breakpoint attributes set whether @value{GDBN}
8041 @c will use hardware or software breakpoints for the internal breakpoints
8042 @c used by the step, next, finish, until, etc. commands.
8046 @c Always use hardware breakpoints
8047 @c @item swbreak (default)
8050 @subsubsection Data Cache
8051 The data cache attributes set whether @value{GDBN} will cache target
8052 memory. While this generally improves performance by reducing debug
8053 protocol overhead, it can lead to incorrect results because @value{GDBN}
8054 does not know about volatile variables or memory mapped device
8059 Enable @value{GDBN} to cache target memory.
8061 Disable @value{GDBN} from caching target memory. This is the default.
8064 @subsection Memory Access Checking
8065 @value{GDBN} can be instructed to refuse accesses to memory that is
8066 not explicitly described. This can be useful if accessing such
8067 regions has undesired effects for a specific target, or to provide
8068 better error checking. The following commands control this behaviour.
8071 @kindex set mem inaccessible-by-default
8072 @item set mem inaccessible-by-default [on|off]
8073 If @code{on} is specified, make @value{GDBN} treat memory not
8074 explicitly described by the memory ranges as non-existent and refuse accesses
8075 to such memory. The checks are only performed if there's at least one
8076 memory range defined. If @code{off} is specified, make @value{GDBN}
8077 treat the memory not explicitly described by the memory ranges as RAM.
8078 The default value is @code{on}.
8079 @kindex show mem inaccessible-by-default
8080 @item show mem inaccessible-by-default
8081 Show the current handling of accesses to unknown memory.
8085 @c @subsubsection Memory Write Verification
8086 @c The memory write verification attributes set whether @value{GDBN}
8087 @c will re-reads data after each write to verify the write was successful.
8091 @c @item noverify (default)
8094 @node Dump/Restore Files
8095 @section Copy Between Memory and a File
8096 @cindex dump/restore files
8097 @cindex append data to a file
8098 @cindex dump data to a file
8099 @cindex restore data from a file
8101 You can use the commands @code{dump}, @code{append}, and
8102 @code{restore} to copy data between target memory and a file. The
8103 @code{dump} and @code{append} commands write data to a file, and the
8104 @code{restore} command reads data from a file back into the inferior's
8105 memory. Files may be in binary, Motorola S-record, Intel hex, or
8106 Tektronix Hex format; however, @value{GDBN} can only append to binary
8112 @item dump @r{[}@var{format}@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
8113 @itemx dump @r{[}@var{format}@r{]} value @var{filename} @var{expr}
8114 Dump the contents of memory from @var{start_addr} to @var{end_addr},
8115 or the value of @var{expr}, to @var{filename} in the given format.
8117 The @var{format} parameter may be any one of:
8124 Motorola S-record format.
8126 Tektronix Hex format.
8129 @value{GDBN} uses the same definitions of these formats as the
8130 @sc{gnu} binary utilities, like @samp{objdump} and @samp{objcopy}. If
8131 @var{format} is omitted, @value{GDBN} dumps the data in raw binary
8135 @item append @r{[}binary@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
8136 @itemx append @r{[}binary@r{]} value @var{filename} @var{expr}
8137 Append the contents of memory from @var{start_addr} to @var{end_addr},
8138 or the value of @var{expr}, to the file @var{filename}, in raw binary form.
8139 (@value{GDBN} can only append data to files in raw binary form.)
8142 @item restore @var{filename} @r{[}binary@r{]} @var{bias} @var{start} @var{end}
8143 Restore the contents of file @var{filename} into memory. The
8144 @code{restore} command can automatically recognize any known @sc{bfd}
8145 file format, except for raw binary. To restore a raw binary file you
8146 must specify the optional keyword @code{binary} after the filename.
8148 If @var{bias} is non-zero, its value will be added to the addresses
8149 contained in the file. Binary files always start at address zero, so
8150 they will be restored at address @var{bias}. Other bfd files have
8151 a built-in location; they will be restored at offset @var{bias}
8154 If @var{start} and/or @var{end} are non-zero, then only data between
8155 file offset @var{start} and file offset @var{end} will be restored.
8156 These offsets are relative to the addresses in the file, before
8157 the @var{bias} argument is applied.
8161 @node Core File Generation
8162 @section How to Produce a Core File from Your Program
8163 @cindex dump core from inferior
8165 A @dfn{core file} or @dfn{core dump} is a file that records the memory
8166 image of a running process and its process status (register values
8167 etc.). Its primary use is post-mortem debugging of a program that
8168 crashed while it ran outside a debugger. A program that crashes
8169 automatically produces a core file, unless this feature is disabled by
8170 the user. @xref{Files}, for information on invoking @value{GDBN} in
8171 the post-mortem debugging mode.
8173 Occasionally, you may wish to produce a core file of the program you
8174 are debugging in order to preserve a snapshot of its state.
8175 @value{GDBN} has a special command for that.
8179 @kindex generate-core-file
8180 @item generate-core-file [@var{file}]
8181 @itemx gcore [@var{file}]
8182 Produce a core dump of the inferior process. The optional argument
8183 @var{file} specifies the file name where to put the core dump. If not
8184 specified, the file name defaults to @file{core.@var{pid}}, where
8185 @var{pid} is the inferior process ID.
8187 Note that this command is implemented only for some systems (as of
8188 this writing, @sc{gnu}/Linux, FreeBSD, Solaris, Unixware, and S390).
8191 @node Character Sets
8192 @section Character Sets
8193 @cindex character sets
8195 @cindex translating between character sets
8196 @cindex host character set
8197 @cindex target character set
8199 If the program you are debugging uses a different character set to
8200 represent characters and strings than the one @value{GDBN} uses itself,
8201 @value{GDBN} can automatically translate between the character sets for
8202 you. The character set @value{GDBN} uses we call the @dfn{host
8203 character set}; the one the inferior program uses we call the
8204 @dfn{target character set}.
8206 For example, if you are running @value{GDBN} on a @sc{gnu}/Linux system, which
8207 uses the ISO Latin 1 character set, but you are using @value{GDBN}'s
8208 remote protocol (@pxref{Remote Debugging}) to debug a program
8209 running on an IBM mainframe, which uses the @sc{ebcdic} character set,
8210 then the host character set is Latin-1, and the target character set is
8211 @sc{ebcdic}. If you give @value{GDBN} the command @code{set
8212 target-charset EBCDIC-US}, then @value{GDBN} translates between
8213 @sc{ebcdic} and Latin 1 as you print character or string values, or use
8214 character and string literals in expressions.
8216 @value{GDBN} has no way to automatically recognize which character set
8217 the inferior program uses; you must tell it, using the @code{set
8218 target-charset} command, described below.
8220 Here are the commands for controlling @value{GDBN}'s character set
8224 @item set target-charset @var{charset}
8225 @kindex set target-charset
8226 Set the current target character set to @var{charset}. To display the
8227 list of supported target character sets, type
8228 @kbd{@w{set target-charset @key{TAB}@key{TAB}}}.
8230 @item set host-charset @var{charset}
8231 @kindex set host-charset
8232 Set the current host character set to @var{charset}.
8234 By default, @value{GDBN} uses a host character set appropriate to the
8235 system it is running on; you can override that default using the
8236 @code{set host-charset} command. On some systems, @value{GDBN} cannot
8237 automatically determine the appropriate host character set. In this
8238 case, @value{GDBN} uses @samp{UTF-8}.
8240 @value{GDBN} can only use certain character sets as its host character
8241 set. If you type @kbd{@w{set target-charset @key{TAB}@key{TAB}}},
8242 @value{GDBN} will list the host character sets it supports.
8244 @item set charset @var{charset}
8246 Set the current host and target character sets to @var{charset}. As
8247 above, if you type @kbd{@w{set charset @key{TAB}@key{TAB}}},
8248 @value{GDBN} will list the names of the character sets that can be used
8249 for both host and target.
8252 @kindex show charset
8253 Show the names of the current host and target character sets.
8255 @item show host-charset
8256 @kindex show host-charset
8257 Show the name of the current host character set.
8259 @item show target-charset
8260 @kindex show target-charset
8261 Show the name of the current target character set.
8263 @item set target-wide-charset @var{charset}
8264 @kindex set target-wide-charset
8265 Set the current target's wide character set to @var{charset}. This is
8266 the character set used by the target's @code{wchar_t} type. To
8267 display the list of supported wide character sets, type
8268 @kbd{@w{set target-wide-charset @key{TAB}@key{TAB}}}.
8270 @item show target-wide-charset
8271 @kindex show target-wide-charset
8272 Show the name of the current target's wide character set.
8275 Here is an example of @value{GDBN}'s character set support in action.
8276 Assume that the following source code has been placed in the file
8277 @file{charset-test.c}:
8283 = @{72, 101, 108, 108, 111, 44, 32, 119,
8284 111, 114, 108, 100, 33, 10, 0@};
8285 char ibm1047_hello[]
8286 = @{200, 133, 147, 147, 150, 107, 64, 166,
8287 150, 153, 147, 132, 90, 37, 0@};
8291 printf ("Hello, world!\n");
8295 In this program, @code{ascii_hello} and @code{ibm1047_hello} are arrays
8296 containing the string @samp{Hello, world!} followed by a newline,
8297 encoded in the @sc{ascii} and @sc{ibm1047} character sets.
8299 We compile the program, and invoke the debugger on it:
8302 $ gcc -g charset-test.c -o charset-test
8303 $ gdb -nw charset-test
8304 GNU gdb 2001-12-19-cvs
8305 Copyright 2001 Free Software Foundation, Inc.
8310 We can use the @code{show charset} command to see what character sets
8311 @value{GDBN} is currently using to interpret and display characters and
8315 (@value{GDBP}) show charset
8316 The current host and target character set is `ISO-8859-1'.
8320 For the sake of printing this manual, let's use @sc{ascii} as our
8321 initial character set:
8323 (@value{GDBP}) set charset ASCII
8324 (@value{GDBP}) show charset
8325 The current host and target character set is `ASCII'.
8329 Let's assume that @sc{ascii} is indeed the correct character set for our
8330 host system --- in other words, let's assume that if @value{GDBN} prints
8331 characters using the @sc{ascii} character set, our terminal will display
8332 them properly. Since our current target character set is also
8333 @sc{ascii}, the contents of @code{ascii_hello} print legibly:
8336 (@value{GDBP}) print ascii_hello
8337 $1 = 0x401698 "Hello, world!\n"
8338 (@value{GDBP}) print ascii_hello[0]
8343 @value{GDBN} uses the target character set for character and string
8344 literals you use in expressions:
8347 (@value{GDBP}) print '+'
8352 The @sc{ascii} character set uses the number 43 to encode the @samp{+}
8355 @value{GDBN} relies on the user to tell it which character set the
8356 target program uses. If we print @code{ibm1047_hello} while our target
8357 character set is still @sc{ascii}, we get jibberish:
8360 (@value{GDBP}) print ibm1047_hello
8361 $4 = 0x4016a8 "\310\205\223\223\226k@@\246\226\231\223\204Z%"
8362 (@value{GDBP}) print ibm1047_hello[0]
8367 If we invoke the @code{set target-charset} followed by @key{TAB}@key{TAB},
8368 @value{GDBN} tells us the character sets it supports:
8371 (@value{GDBP}) set target-charset
8372 ASCII EBCDIC-US IBM1047 ISO-8859-1
8373 (@value{GDBP}) set target-charset
8376 We can select @sc{ibm1047} as our target character set, and examine the
8377 program's strings again. Now the @sc{ascii} string is wrong, but
8378 @value{GDBN} translates the contents of @code{ibm1047_hello} from the
8379 target character set, @sc{ibm1047}, to the host character set,
8380 @sc{ascii}, and they display correctly:
8383 (@value{GDBP}) set target-charset IBM1047
8384 (@value{GDBP}) show charset
8385 The current host character set is `ASCII'.
8386 The current target character set is `IBM1047'.
8387 (@value{GDBP}) print ascii_hello
8388 $6 = 0x401698 "\110\145%%?\054\040\167?\162%\144\041\012"
8389 (@value{GDBP}) print ascii_hello[0]
8391 (@value{GDBP}) print ibm1047_hello
8392 $8 = 0x4016a8 "Hello, world!\n"
8393 (@value{GDBP}) print ibm1047_hello[0]
8398 As above, @value{GDBN} uses the target character set for character and
8399 string literals you use in expressions:
8402 (@value{GDBP}) print '+'
8407 The @sc{ibm1047} character set uses the number 78 to encode the @samp{+}
8410 @node Caching Remote Data
8411 @section Caching Data of Remote Targets
8412 @cindex caching data of remote targets
8414 @value{GDBN} can cache data exchanged between the debugger and a
8415 remote target (@pxref{Remote Debugging}). Such caching generally improves
8416 performance, because it reduces the overhead of the remote protocol by
8417 bundling memory reads and writes into large chunks. Unfortunately,
8418 @value{GDBN} does not currently know anything about volatile
8419 registers, and thus data caching will produce incorrect results when
8420 volatile registers are in use.
8423 @kindex set remotecache
8424 @item set remotecache on
8425 @itemx set remotecache off
8426 Set caching state for remote targets. When @code{ON}, use data
8427 caching. By default, this option is @code{OFF}.
8429 @kindex show remotecache
8430 @item show remotecache
8431 Show the current state of data caching for remote targets.
8435 Print the information about the data cache performance. The
8436 information displayed includes: the dcache width and depth; and for
8437 each cache line, how many times it was referenced, and its data and
8438 state (invalid, dirty, valid). This command is useful for debugging
8439 the data cache operation.
8442 @node Searching Memory
8443 @section Search Memory
8444 @cindex searching memory
8446 Memory can be searched for a particular sequence of bytes with the
8447 @code{find} command.
8451 @item find @r{[}/@var{sn}@r{]} @var{start_addr}, +@var{len}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
8452 @itemx find @r{[}/@var{sn}@r{]} @var{start_addr}, @var{end_addr}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
8453 Search memory for the sequence of bytes specified by @var{val1}, @var{val2},
8454 etc. The search begins at address @var{start_addr} and continues for either
8455 @var{len} bytes or through to @var{end_addr} inclusive.
8458 @var{s} and @var{n} are optional parameters.
8459 They may be specified in either order, apart or together.
8462 @item @var{s}, search query size
8463 The size of each search query value.
8469 halfwords (two bytes)
8473 giant words (eight bytes)
8476 All values are interpreted in the current language.
8477 This means, for example, that if the current source language is C/C@t{++}
8478 then searching for the string ``hello'' includes the trailing '\0'.
8480 If the value size is not specified, it is taken from the
8481 value's type in the current language.
8482 This is useful when one wants to specify the search
8483 pattern as a mixture of types.
8484 Note that this means, for example, that in the case of C-like languages
8485 a search for an untyped 0x42 will search for @samp{(int) 0x42}
8486 which is typically four bytes.
8488 @item @var{n}, maximum number of finds
8489 The maximum number of matches to print. The default is to print all finds.
8492 You can use strings as search values. Quote them with double-quotes
8494 The string value is copied into the search pattern byte by byte,
8495 regardless of the endianness of the target and the size specification.
8497 The address of each match found is printed as well as a count of the
8498 number of matches found.
8500 The address of the last value found is stored in convenience variable
8502 A count of the number of matches is stored in @samp{$numfound}.
8504 For example, if stopped at the @code{printf} in this function:
8510 static char hello[] = "hello-hello";
8511 static struct @{ char c; short s; int i; @}
8512 __attribute__ ((packed)) mixed
8513 = @{ 'c', 0x1234, 0x87654321 @};
8514 printf ("%s\n", hello);
8519 you get during debugging:
8522 (gdb) find &hello[0], +sizeof(hello), "hello"
8523 0x804956d <hello.1620+6>
8525 (gdb) find &hello[0], +sizeof(hello), 'h', 'e', 'l', 'l', 'o'
8526 0x8049567 <hello.1620>
8527 0x804956d <hello.1620+6>
8529 (gdb) find /b1 &hello[0], +sizeof(hello), 'h', 0x65, 'l'
8530 0x8049567 <hello.1620>
8532 (gdb) find &mixed, +sizeof(mixed), (char) 'c', (short) 0x1234, (int) 0x87654321
8533 0x8049560 <mixed.1625>
8535 (gdb) print $numfound
8538 $2 = (void *) 0x8049560
8542 @chapter C Preprocessor Macros
8544 Some languages, such as C and C@t{++}, provide a way to define and invoke
8545 ``preprocessor macros'' which expand into strings of tokens.
8546 @value{GDBN} can evaluate expressions containing macro invocations, show
8547 the result of macro expansion, and show a macro's definition, including
8548 where it was defined.
8550 You may need to compile your program specially to provide @value{GDBN}
8551 with information about preprocessor macros. Most compilers do not
8552 include macros in their debugging information, even when you compile
8553 with the @option{-g} flag. @xref{Compilation}.
8555 A program may define a macro at one point, remove that definition later,
8556 and then provide a different definition after that. Thus, at different
8557 points in the program, a macro may have different definitions, or have
8558 no definition at all. If there is a current stack frame, @value{GDBN}
8559 uses the macros in scope at that frame's source code line. Otherwise,
8560 @value{GDBN} uses the macros in scope at the current listing location;
8563 Whenever @value{GDBN} evaluates an expression, it always expands any
8564 macro invocations present in the expression. @value{GDBN} also provides
8565 the following commands for working with macros explicitly.
8569 @kindex macro expand
8570 @cindex macro expansion, showing the results of preprocessor
8571 @cindex preprocessor macro expansion, showing the results of
8572 @cindex expanding preprocessor macros
8573 @item macro expand @var{expression}
8574 @itemx macro exp @var{expression}
8575 Show the results of expanding all preprocessor macro invocations in
8576 @var{expression}. Since @value{GDBN} simply expands macros, but does
8577 not parse the result, @var{expression} need not be a valid expression;
8578 it can be any string of tokens.
8581 @item macro expand-once @var{expression}
8582 @itemx macro exp1 @var{expression}
8583 @cindex expand macro once
8584 @i{(This command is not yet implemented.)} Show the results of
8585 expanding those preprocessor macro invocations that appear explicitly in
8586 @var{expression}. Macro invocations appearing in that expansion are
8587 left unchanged. This command allows you to see the effect of a
8588 particular macro more clearly, without being confused by further
8589 expansions. Since @value{GDBN} simply expands macros, but does not
8590 parse the result, @var{expression} need not be a valid expression; it
8591 can be any string of tokens.
8594 @cindex macro definition, showing
8595 @cindex definition, showing a macro's
8596 @item info macro @var{macro}
8597 Show the definition of the macro named @var{macro}, and describe the
8598 source location or compiler command-line where that definition was established.
8600 @kindex macro define
8601 @cindex user-defined macros
8602 @cindex defining macros interactively
8603 @cindex macros, user-defined
8604 @item macro define @var{macro} @var{replacement-list}
8605 @itemx macro define @var{macro}(@var{arglist}) @var{replacement-list}
8606 Introduce a definition for a preprocessor macro named @var{macro},
8607 invocations of which are replaced by the tokens given in
8608 @var{replacement-list}. The first form of this command defines an
8609 ``object-like'' macro, which takes no arguments; the second form
8610 defines a ``function-like'' macro, which takes the arguments given in
8613 A definition introduced by this command is in scope in every
8614 expression evaluated in @value{GDBN}, until it is removed with the
8615 @code{macro undef} command, described below. The definition overrides
8616 all definitions for @var{macro} present in the program being debugged,
8617 as well as any previous user-supplied definition.
8620 @item macro undef @var{macro}
8621 Remove any user-supplied definition for the macro named @var{macro}.
8622 This command only affects definitions provided with the @code{macro
8623 define} command, described above; it cannot remove definitions present
8624 in the program being debugged.
8628 List all the macros defined using the @code{macro define} command.
8631 @cindex macros, example of debugging with
8632 Here is a transcript showing the above commands in action. First, we
8633 show our source files:
8641 #define ADD(x) (M + x)
8646 printf ("Hello, world!\n");
8648 printf ("We're so creative.\n");
8650 printf ("Goodbye, world!\n");
8657 Now, we compile the program using the @sc{gnu} C compiler, @value{NGCC}.
8658 We pass the @option{-gdwarf-2} and @option{-g3} flags to ensure the
8659 compiler includes information about preprocessor macros in the debugging
8663 $ gcc -gdwarf-2 -g3 sample.c -o sample
8667 Now, we start @value{GDBN} on our sample program:
8671 GNU gdb 2002-05-06-cvs
8672 Copyright 2002 Free Software Foundation, Inc.
8673 GDB is free software, @dots{}
8677 We can expand macros and examine their definitions, even when the
8678 program is not running. @value{GDBN} uses the current listing position
8679 to decide which macro definitions are in scope:
8682 (@value{GDBP}) list main
8685 5 #define ADD(x) (M + x)
8690 10 printf ("Hello, world!\n");
8692 12 printf ("We're so creative.\n");
8693 (@value{GDBP}) info macro ADD
8694 Defined at /home/jimb/gdb/macros/play/sample.c:5
8695 #define ADD(x) (M + x)
8696 (@value{GDBP}) info macro Q
8697 Defined at /home/jimb/gdb/macros/play/sample.h:1
8698 included at /home/jimb/gdb/macros/play/sample.c:2
8700 (@value{GDBP}) macro expand ADD(1)
8701 expands to: (42 + 1)
8702 (@value{GDBP}) macro expand-once ADD(1)
8703 expands to: once (M + 1)
8707 In the example above, note that @code{macro expand-once} expands only
8708 the macro invocation explicit in the original text --- the invocation of
8709 @code{ADD} --- but does not expand the invocation of the macro @code{M},
8710 which was introduced by @code{ADD}.
8712 Once the program is running, @value{GDBN} uses the macro definitions in
8713 force at the source line of the current stack frame:
8716 (@value{GDBP}) break main
8717 Breakpoint 1 at 0x8048370: file sample.c, line 10.
8719 Starting program: /home/jimb/gdb/macros/play/sample
8721 Breakpoint 1, main () at sample.c:10
8722 10 printf ("Hello, world!\n");
8726 At line 10, the definition of the macro @code{N} at line 9 is in force:
8729 (@value{GDBP}) info macro N
8730 Defined at /home/jimb/gdb/macros/play/sample.c:9
8732 (@value{GDBP}) macro expand N Q M
8734 (@value{GDBP}) print N Q M
8739 As we step over directives that remove @code{N}'s definition, and then
8740 give it a new definition, @value{GDBN} finds the definition (or lack
8741 thereof) in force at each point:
8746 12 printf ("We're so creative.\n");
8747 (@value{GDBP}) info macro N
8748 The symbol `N' has no definition as a C/C++ preprocessor macro
8749 at /home/jimb/gdb/macros/play/sample.c:12
8752 14 printf ("Goodbye, world!\n");
8753 (@value{GDBP}) info macro N
8754 Defined at /home/jimb/gdb/macros/play/sample.c:13
8756 (@value{GDBP}) macro expand N Q M
8757 expands to: 1729 < 42
8758 (@value{GDBP}) print N Q M
8763 In addition to source files, macros can be defined on the compilation command
8764 line using the @option{-D@var{name}=@var{value}} syntax. For macros defined in
8765 such a way, @value{GDBN} displays the location of their definition as line zero
8766 of the source file submitted to the compiler.
8769 (@value{GDBP}) info macro __STDC__
8770 Defined at /home/jimb/gdb/macros/play/sample.c:0
8777 @chapter Tracepoints
8778 @c This chapter is based on the documentation written by Michael
8779 @c Snyder, David Taylor, Jim Blandy, and Elena Zannoni.
8782 In some applications, it is not feasible for the debugger to interrupt
8783 the program's execution long enough for the developer to learn
8784 anything helpful about its behavior. If the program's correctness
8785 depends on its real-time behavior, delays introduced by a debugger
8786 might cause the program to change its behavior drastically, or perhaps
8787 fail, even when the code itself is correct. It is useful to be able
8788 to observe the program's behavior without interrupting it.
8790 Using @value{GDBN}'s @code{trace} and @code{collect} commands, you can
8791 specify locations in the program, called @dfn{tracepoints}, and
8792 arbitrary expressions to evaluate when those tracepoints are reached.
8793 Later, using the @code{tfind} command, you can examine the values
8794 those expressions had when the program hit the tracepoints. The
8795 expressions may also denote objects in memory---structures or arrays,
8796 for example---whose values @value{GDBN} should record; while visiting
8797 a particular tracepoint, you may inspect those objects as if they were
8798 in memory at that moment. However, because @value{GDBN} records these
8799 values without interacting with you, it can do so quickly and
8800 unobtrusively, hopefully not disturbing the program's behavior.
8802 The tracepoint facility is currently available only for remote
8803 targets. @xref{Targets}. In addition, your remote target must know
8804 how to collect trace data. This functionality is implemented in the
8805 remote stub; however, none of the stubs distributed with @value{GDBN}
8806 support tracepoints as of this writing. The format of the remote
8807 packets used to implement tracepoints are described in @ref{Tracepoint
8810 This chapter describes the tracepoint commands and features.
8814 * Analyze Collected Data::
8815 * Tracepoint Variables::
8818 @node Set Tracepoints
8819 @section Commands to Set Tracepoints
8821 Before running such a @dfn{trace experiment}, an arbitrary number of
8822 tracepoints can be set. A tracepoint is actually a special type of
8823 breakpoint (@pxref{Set Breaks}), so you can manipulate it using
8824 standard breakpoint commands. For instance, as with breakpoints,
8825 tracepoint numbers are successive integers starting from one, and many
8826 of the commands associated with tracepoints take the tracepoint number
8827 as their argument, to identify which tracepoint to work on.
8829 For each tracepoint, you can specify, in advance, some arbitrary set
8830 of data that you want the target to collect in the trace buffer when
8831 it hits that tracepoint. The collected data can include registers,
8832 local variables, or global data. Later, you can use @value{GDBN}
8833 commands to examine the values these data had at the time the
8836 Tracepoints do not support every breakpoint feature. Conditional
8837 expressions and ignore counts on tracepoints have no effect, and
8838 tracepoints cannot run @value{GDBN} commands when they are
8839 hit. Tracepoints may not be thread-specific either.
8841 This section describes commands to set tracepoints and associated
8842 conditions and actions.
8845 * Create and Delete Tracepoints::
8846 * Enable and Disable Tracepoints::
8847 * Tracepoint Passcounts::
8848 * Tracepoint Actions::
8849 * Listing Tracepoints::
8850 * Starting and Stopping Trace Experiments::
8853 @node Create and Delete Tracepoints
8854 @subsection Create and Delete Tracepoints
8857 @cindex set tracepoint
8859 @item trace @var{location}
8860 The @code{trace} command is very similar to the @code{break} command.
8861 Its argument @var{location} can be a source line, a function name, or
8862 an address in the target program. @xref{Specify Location}. The
8863 @code{trace} command defines a tracepoint, which is a point in the
8864 target program where the debugger will briefly stop, collect some
8865 data, and then allow the program to continue. Setting a tracepoint or
8866 changing its actions doesn't take effect until the next @code{tstart}
8867 command, and once a trace experiment is running, further changes will
8868 not have any effect until the next trace experiment starts.
8870 Here are some examples of using the @code{trace} command:
8873 (@value{GDBP}) @b{trace foo.c:121} // a source file and line number
8875 (@value{GDBP}) @b{trace +2} // 2 lines forward
8877 (@value{GDBP}) @b{trace my_function} // first source line of function
8879 (@value{GDBP}) @b{trace *my_function} // EXACT start address of function
8881 (@value{GDBP}) @b{trace *0x2117c4} // an address
8885 You can abbreviate @code{trace} as @code{tr}.
8888 @cindex last tracepoint number
8889 @cindex recent tracepoint number
8890 @cindex tracepoint number
8891 The convenience variable @code{$tpnum} records the tracepoint number
8892 of the most recently set tracepoint.
8894 @kindex delete tracepoint
8895 @cindex tracepoint deletion
8896 @item delete tracepoint @r{[}@var{num}@r{]}
8897 Permanently delete one or more tracepoints. With no argument, the
8898 default is to delete all tracepoints. Note that the regular
8899 @code{delete} command can remove tracepoints also.
8904 (@value{GDBP}) @b{delete trace 1 2 3} // remove three tracepoints
8906 (@value{GDBP}) @b{delete trace} // remove all tracepoints
8910 You can abbreviate this command as @code{del tr}.
8913 @node Enable and Disable Tracepoints
8914 @subsection Enable and Disable Tracepoints
8916 These commands are deprecated; they are equivalent to plain @code{disable} and @code{enable}.
8919 @kindex disable tracepoint
8920 @item disable tracepoint @r{[}@var{num}@r{]}
8921 Disable tracepoint @var{num}, or all tracepoints if no argument
8922 @var{num} is given. A disabled tracepoint will have no effect during
8923 the next trace experiment, but it is not forgotten. You can re-enable
8924 a disabled tracepoint using the @code{enable tracepoint} command.
8926 @kindex enable tracepoint
8927 @item enable tracepoint @r{[}@var{num}@r{]}
8928 Enable tracepoint @var{num}, or all tracepoints. The enabled
8929 tracepoints will become effective the next time a trace experiment is
8933 @node Tracepoint Passcounts
8934 @subsection Tracepoint Passcounts
8938 @cindex tracepoint pass count
8939 @item passcount @r{[}@var{n} @r{[}@var{num}@r{]]}
8940 Set the @dfn{passcount} of a tracepoint. The passcount is a way to
8941 automatically stop a trace experiment. If a tracepoint's passcount is
8942 @var{n}, then the trace experiment will be automatically stopped on
8943 the @var{n}'th time that tracepoint is hit. If the tracepoint number
8944 @var{num} is not specified, the @code{passcount} command sets the
8945 passcount of the most recently defined tracepoint. If no passcount is
8946 given, the trace experiment will run until stopped explicitly by the
8952 (@value{GDBP}) @b{passcount 5 2} // Stop on the 5th execution of
8953 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// tracepoint 2}
8955 (@value{GDBP}) @b{passcount 12} // Stop on the 12th execution of the
8956 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// most recently defined tracepoint.}
8957 (@value{GDBP}) @b{trace foo}
8958 (@value{GDBP}) @b{pass 3}
8959 (@value{GDBP}) @b{trace bar}
8960 (@value{GDBP}) @b{pass 2}
8961 (@value{GDBP}) @b{trace baz}
8962 (@value{GDBP}) @b{pass 1} // Stop tracing when foo has been
8963 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// executed 3 times OR when bar has}
8964 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// been executed 2 times}
8965 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// OR when baz has been executed 1 time.}
8969 @node Tracepoint Actions
8970 @subsection Tracepoint Action Lists
8974 @cindex tracepoint actions
8975 @item actions @r{[}@var{num}@r{]}
8976 This command will prompt for a list of actions to be taken when the
8977 tracepoint is hit. If the tracepoint number @var{num} is not
8978 specified, this command sets the actions for the one that was most
8979 recently defined (so that you can define a tracepoint and then say
8980 @code{actions} without bothering about its number). You specify the
8981 actions themselves on the following lines, one action at a time, and
8982 terminate the actions list with a line containing just @code{end}. So
8983 far, the only defined actions are @code{collect} and
8984 @code{while-stepping}.
8986 @cindex remove actions from a tracepoint
8987 To remove all actions from a tracepoint, type @samp{actions @var{num}}
8988 and follow it immediately with @samp{end}.
8991 (@value{GDBP}) @b{collect @var{data}} // collect some data
8993 (@value{GDBP}) @b{while-stepping 5} // single-step 5 times, collect data
8995 (@value{GDBP}) @b{end} // signals the end of actions.
8998 In the following example, the action list begins with @code{collect}
8999 commands indicating the things to be collected when the tracepoint is
9000 hit. Then, in order to single-step and collect additional data
9001 following the tracepoint, a @code{while-stepping} command is used,
9002 followed by the list of things to be collected while stepping. The
9003 @code{while-stepping} command is terminated by its own separate
9004 @code{end} command. Lastly, the action list is terminated by an
9008 (@value{GDBP}) @b{trace foo}
9009 (@value{GDBP}) @b{actions}
9010 Enter actions for tracepoint 1, one per line:
9019 @kindex collect @r{(tracepoints)}
9020 @item collect @var{expr1}, @var{expr2}, @dots{}
9021 Collect values of the given expressions when the tracepoint is hit.
9022 This command accepts a comma-separated list of any valid expressions.
9023 In addition to global, static, or local variables, the following
9024 special arguments are supported:
9028 collect all registers
9031 collect all function arguments
9034 collect all local variables.
9037 You can give several consecutive @code{collect} commands, each one
9038 with a single argument, or one @code{collect} command with several
9039 arguments separated by commas: the effect is the same.
9041 The command @code{info scope} (@pxref{Symbols, info scope}) is
9042 particularly useful for figuring out what data to collect.
9044 @kindex while-stepping @r{(tracepoints)}
9045 @item while-stepping @var{n}
9046 Perform @var{n} single-step traces after the tracepoint, collecting
9047 new data at each step. The @code{while-stepping} command is
9048 followed by the list of what to collect while stepping (followed by
9049 its own @code{end} command):
9053 > collect $regs, myglobal
9059 You may abbreviate @code{while-stepping} as @code{ws} or
9063 @node Listing Tracepoints
9064 @subsection Listing Tracepoints
9067 @kindex info tracepoints
9069 @cindex information about tracepoints
9070 @item info tracepoints @r{[}@var{num}@r{]}
9071 Display information about the tracepoint @var{num}. If you don't
9072 specify a tracepoint number, displays information about all the
9073 tracepoints defined so far. The format is similar to that used for
9074 @code{info breakpoints}; in fact, @code{info tracepoints} is the same
9075 command, simply restricting itself to tracepoints.
9077 A tracepoint's listing may include additional information specific to
9082 its passcount as given by the @code{passcount @var{n}} command
9084 its step count as given by the @code{while-stepping @var{n}} command
9086 its action list as given by the @code{actions} command. The actions
9087 are prefixed with an @samp{A} so as to distinguish them from commands.
9091 (@value{GDBP}) @b{info trace}
9092 Num Type Disp Enb Address What
9093 1 tracepoint keep y 0x0804ab57 in foo() at main.cxx:7
9097 A collect globfoo, $regs
9105 This command can be abbreviated @code{info tp}.
9108 @node Starting and Stopping Trace Experiments
9109 @subsection Starting and Stopping Trace Experiments
9113 @cindex start a new trace experiment
9114 @cindex collected data discarded
9116 This command takes no arguments. It starts the trace experiment, and
9117 begins collecting data. This has the side effect of discarding all
9118 the data collected in the trace buffer during the previous trace
9122 @cindex stop a running trace experiment
9124 This command takes no arguments. It ends the trace experiment, and
9125 stops collecting data.
9127 @strong{Note}: a trace experiment and data collection may stop
9128 automatically if any tracepoint's passcount is reached
9129 (@pxref{Tracepoint Passcounts}), or if the trace buffer becomes full.
9132 @cindex status of trace data collection
9133 @cindex trace experiment, status of
9135 This command displays the status of the current trace data
9139 Here is an example of the commands we described so far:
9142 (@value{GDBP}) @b{trace gdb_c_test}
9143 (@value{GDBP}) @b{actions}
9144 Enter actions for tracepoint #1, one per line.
9145 > collect $regs,$locals,$args
9150 (@value{GDBP}) @b{tstart}
9151 [time passes @dots{}]
9152 (@value{GDBP}) @b{tstop}
9156 @node Analyze Collected Data
9157 @section Using the Collected Data
9159 After the tracepoint experiment ends, you use @value{GDBN} commands
9160 for examining the trace data. The basic idea is that each tracepoint
9161 collects a trace @dfn{snapshot} every time it is hit and another
9162 snapshot every time it single-steps. All these snapshots are
9163 consecutively numbered from zero and go into a buffer, and you can
9164 examine them later. The way you examine them is to @dfn{focus} on a
9165 specific trace snapshot. When the remote stub is focused on a trace
9166 snapshot, it will respond to all @value{GDBN} requests for memory and
9167 registers by reading from the buffer which belongs to that snapshot,
9168 rather than from @emph{real} memory or registers of the program being
9169 debugged. This means that @strong{all} @value{GDBN} commands
9170 (@code{print}, @code{info registers}, @code{backtrace}, etc.) will
9171 behave as if we were currently debugging the program state as it was
9172 when the tracepoint occurred. Any requests for data that are not in
9173 the buffer will fail.
9176 * tfind:: How to select a trace snapshot
9177 * tdump:: How to display all data for a snapshot
9178 * save-tracepoints:: How to save tracepoints for a future run
9182 @subsection @code{tfind @var{n}}
9185 @cindex select trace snapshot
9186 @cindex find trace snapshot
9187 The basic command for selecting a trace snapshot from the buffer is
9188 @code{tfind @var{n}}, which finds trace snapshot number @var{n},
9189 counting from zero. If no argument @var{n} is given, the next
9190 snapshot is selected.
9192 Here are the various forms of using the @code{tfind} command.
9196 Find the first snapshot in the buffer. This is a synonym for
9197 @code{tfind 0} (since 0 is the number of the first snapshot).
9200 Stop debugging trace snapshots, resume @emph{live} debugging.
9203 Same as @samp{tfind none}.
9206 No argument means find the next trace snapshot.
9209 Find the previous trace snapshot before the current one. This permits
9210 retracing earlier steps.
9212 @item tfind tracepoint @var{num}
9213 Find the next snapshot associated with tracepoint @var{num}. Search
9214 proceeds forward from the last examined trace snapshot. If no
9215 argument @var{num} is given, it means find the next snapshot collected
9216 for the same tracepoint as the current snapshot.
9218 @item tfind pc @var{addr}
9219 Find the next snapshot associated with the value @var{addr} of the
9220 program counter. Search proceeds forward from the last examined trace
9221 snapshot. If no argument @var{addr} is given, it means find the next
9222 snapshot with the same value of PC as the current snapshot.
9224 @item tfind outside @var{addr1}, @var{addr2}
9225 Find the next snapshot whose PC is outside the given range of
9228 @item tfind range @var{addr1}, @var{addr2}
9229 Find the next snapshot whose PC is between @var{addr1} and
9230 @var{addr2}. @c FIXME: Is the range inclusive or exclusive?
9232 @item tfind line @r{[}@var{file}:@r{]}@var{n}
9233 Find the next snapshot associated with the source line @var{n}. If
9234 the optional argument @var{file} is given, refer to line @var{n} in
9235 that source file. Search proceeds forward from the last examined
9236 trace snapshot. If no argument @var{n} is given, it means find the
9237 next line other than the one currently being examined; thus saying
9238 @code{tfind line} repeatedly can appear to have the same effect as
9239 stepping from line to line in a @emph{live} debugging session.
9242 The default arguments for the @code{tfind} commands are specifically
9243 designed to make it easy to scan through the trace buffer. For
9244 instance, @code{tfind} with no argument selects the next trace
9245 snapshot, and @code{tfind -} with no argument selects the previous
9246 trace snapshot. So, by giving one @code{tfind} command, and then
9247 simply hitting @key{RET} repeatedly you can examine all the trace
9248 snapshots in order. Or, by saying @code{tfind -} and then hitting
9249 @key{RET} repeatedly you can examine the snapshots in reverse order.
9250 The @code{tfind line} command with no argument selects the snapshot
9251 for the next source line executed. The @code{tfind pc} command with
9252 no argument selects the next snapshot with the same program counter
9253 (PC) as the current frame. The @code{tfind tracepoint} command with
9254 no argument selects the next trace snapshot collected by the same
9255 tracepoint as the current one.
9257 In addition to letting you scan through the trace buffer manually,
9258 these commands make it easy to construct @value{GDBN} scripts that
9259 scan through the trace buffer and print out whatever collected data
9260 you are interested in. Thus, if we want to examine the PC, FP, and SP
9261 registers from each trace frame in the buffer, we can say this:
9264 (@value{GDBP}) @b{tfind start}
9265 (@value{GDBP}) @b{while ($trace_frame != -1)}
9266 > printf "Frame %d, PC = %08X, SP = %08X, FP = %08X\n", \
9267 $trace_frame, $pc, $sp, $fp
9271 Frame 0, PC = 0020DC64, SP = 0030BF3C, FP = 0030BF44
9272 Frame 1, PC = 0020DC6C, SP = 0030BF38, FP = 0030BF44
9273 Frame 2, PC = 0020DC70, SP = 0030BF34, FP = 0030BF44
9274 Frame 3, PC = 0020DC74, SP = 0030BF30, FP = 0030BF44
9275 Frame 4, PC = 0020DC78, SP = 0030BF2C, FP = 0030BF44
9276 Frame 5, PC = 0020DC7C, SP = 0030BF28, FP = 0030BF44
9277 Frame 6, PC = 0020DC80, SP = 0030BF24, FP = 0030BF44
9278 Frame 7, PC = 0020DC84, SP = 0030BF20, FP = 0030BF44
9279 Frame 8, PC = 0020DC88, SP = 0030BF1C, FP = 0030BF44
9280 Frame 9, PC = 0020DC8E, SP = 0030BF18, FP = 0030BF44
9281 Frame 10, PC = 00203F6C, SP = 0030BE3C, FP = 0030BF14
9284 Or, if we want to examine the variable @code{X} at each source line in
9288 (@value{GDBP}) @b{tfind start}
9289 (@value{GDBP}) @b{while ($trace_frame != -1)}
9290 > printf "Frame %d, X == %d\n", $trace_frame, X
9300 @subsection @code{tdump}
9302 @cindex dump all data collected at tracepoint
9303 @cindex tracepoint data, display
9305 This command takes no arguments. It prints all the data collected at
9306 the current trace snapshot.
9309 (@value{GDBP}) @b{trace 444}
9310 (@value{GDBP}) @b{actions}
9311 Enter actions for tracepoint #2, one per line:
9312 > collect $regs, $locals, $args, gdb_long_test
9315 (@value{GDBP}) @b{tstart}
9317 (@value{GDBP}) @b{tfind line 444}
9318 #0 gdb_test (p1=0x11, p2=0x22, p3=0x33, p4=0x44, p5=0x55, p6=0x66)
9320 444 printp( "%s: arguments = 0x%X 0x%X 0x%X 0x%X 0x%X 0x%X\n", )
9322 (@value{GDBP}) @b{tdump}
9323 Data collected at tracepoint 2, trace frame 1:
9324 d0 0xc4aa0085 -995491707
9328 d4 0x71aea3d 119204413
9333 a1 0x3000668 50333288
9336 a4 0x3000698 50333336
9338 fp 0x30bf3c 0x30bf3c
9339 sp 0x30bf34 0x30bf34
9341 pc 0x20b2c8 0x20b2c8
9345 p = 0x20e5b4 "gdb-test"
9352 gdb_long_test = 17 '\021'
9357 @node save-tracepoints
9358 @subsection @code{save-tracepoints @var{filename}}
9359 @kindex save-tracepoints
9360 @cindex save tracepoints for future sessions
9362 This command saves all current tracepoint definitions together with
9363 their actions and passcounts, into a file @file{@var{filename}}
9364 suitable for use in a later debugging session. To read the saved
9365 tracepoint definitions, use the @code{source} command (@pxref{Command
9368 @node Tracepoint Variables
9369 @section Convenience Variables for Tracepoints
9370 @cindex tracepoint variables
9371 @cindex convenience variables for tracepoints
9374 @vindex $trace_frame
9375 @item (int) $trace_frame
9376 The current trace snapshot (a.k.a.@: @dfn{frame}) number, or -1 if no
9377 snapshot is selected.
9380 @item (int) $tracepoint
9381 The tracepoint for the current trace snapshot.
9384 @item (int) $trace_line
9385 The line number for the current trace snapshot.
9388 @item (char []) $trace_file
9389 The source file for the current trace snapshot.
9392 @item (char []) $trace_func
9393 The name of the function containing @code{$tracepoint}.
9396 Note: @code{$trace_file} is not suitable for use in @code{printf},
9397 use @code{output} instead.
9399 Here's a simple example of using these convenience variables for
9400 stepping through all the trace snapshots and printing some of their
9404 (@value{GDBP}) @b{tfind start}
9406 (@value{GDBP}) @b{while $trace_frame != -1}
9407 > output $trace_file
9408 > printf ", line %d (tracepoint #%d)\n", $trace_line, $tracepoint
9414 @chapter Debugging Programs That Use Overlays
9417 If your program is too large to fit completely in your target system's
9418 memory, you can sometimes use @dfn{overlays} to work around this
9419 problem. @value{GDBN} provides some support for debugging programs that
9423 * How Overlays Work:: A general explanation of overlays.
9424 * Overlay Commands:: Managing overlays in @value{GDBN}.
9425 * Automatic Overlay Debugging:: @value{GDBN} can find out which overlays are
9426 mapped by asking the inferior.
9427 * Overlay Sample Program:: A sample program using overlays.
9430 @node How Overlays Work
9431 @section How Overlays Work
9432 @cindex mapped overlays
9433 @cindex unmapped overlays
9434 @cindex load address, overlay's
9435 @cindex mapped address
9436 @cindex overlay area
9438 Suppose you have a computer whose instruction address space is only 64
9439 kilobytes long, but which has much more memory which can be accessed by
9440 other means: special instructions, segment registers, or memory
9441 management hardware, for example. Suppose further that you want to
9442 adapt a program which is larger than 64 kilobytes to run on this system.
9444 One solution is to identify modules of your program which are relatively
9445 independent, and need not call each other directly; call these modules
9446 @dfn{overlays}. Separate the overlays from the main program, and place
9447 their machine code in the larger memory. Place your main program in
9448 instruction memory, but leave at least enough space there to hold the
9449 largest overlay as well.
9451 Now, to call a function located in an overlay, you must first copy that
9452 overlay's machine code from the large memory into the space set aside
9453 for it in the instruction memory, and then jump to its entry point
9456 @c NB: In the below the mapped area's size is greater or equal to the
9457 @c size of all overlays. This is intentional to remind the developer
9458 @c that overlays don't necessarily need to be the same size.
9462 Data Instruction Larger
9463 Address Space Address Space Address Space
9464 +-----------+ +-----------+ +-----------+
9466 +-----------+ +-----------+ +-----------+<-- overlay 1
9467 | program | | main | .----| overlay 1 | load address
9468 | variables | | program | | +-----------+
9469 | and heap | | | | | |
9470 +-----------+ | | | +-----------+<-- overlay 2
9471 | | +-----------+ | | | load address
9472 +-----------+ | | | .-| overlay 2 |
9474 mapped --->+-----------+ | | +-----------+
9476 | overlay | <-' | | |
9477 | area | <---' +-----------+<-- overlay 3
9478 | | <---. | | load address
9479 +-----------+ `--| overlay 3 |
9486 @anchor{A code overlay}A code overlay
9490 The diagram (@pxref{A code overlay}) shows a system with separate data
9491 and instruction address spaces. To map an overlay, the program copies
9492 its code from the larger address space to the instruction address space.
9493 Since the overlays shown here all use the same mapped address, only one
9494 may be mapped at a time. For a system with a single address space for
9495 data and instructions, the diagram would be similar, except that the
9496 program variables and heap would share an address space with the main
9497 program and the overlay area.
9499 An overlay loaded into instruction memory and ready for use is called a
9500 @dfn{mapped} overlay; its @dfn{mapped address} is its address in the
9501 instruction memory. An overlay not present (or only partially present)
9502 in instruction memory is called @dfn{unmapped}; its @dfn{load address}
9503 is its address in the larger memory. The mapped address is also called
9504 the @dfn{virtual memory address}, or @dfn{VMA}; the load address is also
9505 called the @dfn{load memory address}, or @dfn{LMA}.
9507 Unfortunately, overlays are not a completely transparent way to adapt a
9508 program to limited instruction memory. They introduce a new set of
9509 global constraints you must keep in mind as you design your program:
9514 Before calling or returning to a function in an overlay, your program
9515 must make sure that overlay is actually mapped. Otherwise, the call or
9516 return will transfer control to the right address, but in the wrong
9517 overlay, and your program will probably crash.
9520 If the process of mapping an overlay is expensive on your system, you
9521 will need to choose your overlays carefully to minimize their effect on
9522 your program's performance.
9525 The executable file you load onto your system must contain each
9526 overlay's instructions, appearing at the overlay's load address, not its
9527 mapped address. However, each overlay's instructions must be relocated
9528 and its symbols defined as if the overlay were at its mapped address.
9529 You can use GNU linker scripts to specify different load and relocation
9530 addresses for pieces of your program; see @ref{Overlay Description,,,
9531 ld.info, Using ld: the GNU linker}.
9534 The procedure for loading executable files onto your system must be able
9535 to load their contents into the larger address space as well as the
9536 instruction and data spaces.
9540 The overlay system described above is rather simple, and could be
9541 improved in many ways:
9546 If your system has suitable bank switch registers or memory management
9547 hardware, you could use those facilities to make an overlay's load area
9548 contents simply appear at their mapped address in instruction space.
9549 This would probably be faster than copying the overlay to its mapped
9550 area in the usual way.
9553 If your overlays are small enough, you could set aside more than one
9554 overlay area, and have more than one overlay mapped at a time.
9557 You can use overlays to manage data, as well as instructions. In
9558 general, data overlays are even less transparent to your design than
9559 code overlays: whereas code overlays only require care when you call or
9560 return to functions, data overlays require care every time you access
9561 the data. Also, if you change the contents of a data overlay, you
9562 must copy its contents back out to its load address before you can copy a
9563 different data overlay into the same mapped area.
9568 @node Overlay Commands
9569 @section Overlay Commands
9571 To use @value{GDBN}'s overlay support, each overlay in your program must
9572 correspond to a separate section of the executable file. The section's
9573 virtual memory address and load memory address must be the overlay's
9574 mapped and load addresses. Identifying overlays with sections allows
9575 @value{GDBN} to determine the appropriate address of a function or
9576 variable, depending on whether the overlay is mapped or not.
9578 @value{GDBN}'s overlay commands all start with the word @code{overlay};
9579 you can abbreviate this as @code{ov} or @code{ovly}. The commands are:
9584 Disable @value{GDBN}'s overlay support. When overlay support is
9585 disabled, @value{GDBN} assumes that all functions and variables are
9586 always present at their mapped addresses. By default, @value{GDBN}'s
9587 overlay support is disabled.
9589 @item overlay manual
9590 @cindex manual overlay debugging
9591 Enable @dfn{manual} overlay debugging. In this mode, @value{GDBN}
9592 relies on you to tell it which overlays are mapped, and which are not,
9593 using the @code{overlay map-overlay} and @code{overlay unmap-overlay}
9594 commands described below.
9596 @item overlay map-overlay @var{overlay}
9597 @itemx overlay map @var{overlay}
9598 @cindex map an overlay
9599 Tell @value{GDBN} that @var{overlay} is now mapped; @var{overlay} must
9600 be the name of the object file section containing the overlay. When an
9601 overlay is mapped, @value{GDBN} assumes it can find the overlay's
9602 functions and variables at their mapped addresses. @value{GDBN} assumes
9603 that any other overlays whose mapped ranges overlap that of
9604 @var{overlay} are now unmapped.
9606 @item overlay unmap-overlay @var{overlay}
9607 @itemx overlay unmap @var{overlay}
9608 @cindex unmap an overlay
9609 Tell @value{GDBN} that @var{overlay} is no longer mapped; @var{overlay}
9610 must be the name of the object file section containing the overlay.
9611 When an overlay is unmapped, @value{GDBN} assumes it can find the
9612 overlay's functions and variables at their load addresses.
9615 Enable @dfn{automatic} overlay debugging. In this mode, @value{GDBN}
9616 consults a data structure the overlay manager maintains in the inferior
9617 to see which overlays are mapped. For details, see @ref{Automatic
9620 @item overlay load-target
9622 @cindex reloading the overlay table
9623 Re-read the overlay table from the inferior. Normally, @value{GDBN}
9624 re-reads the table @value{GDBN} automatically each time the inferior
9625 stops, so this command should only be necessary if you have changed the
9626 overlay mapping yourself using @value{GDBN}. This command is only
9627 useful when using automatic overlay debugging.
9629 @item overlay list-overlays
9631 @cindex listing mapped overlays
9632 Display a list of the overlays currently mapped, along with their mapped
9633 addresses, load addresses, and sizes.
9637 Normally, when @value{GDBN} prints a code address, it includes the name
9638 of the function the address falls in:
9641 (@value{GDBP}) print main
9642 $3 = @{int ()@} 0x11a0 <main>
9645 When overlay debugging is enabled, @value{GDBN} recognizes code in
9646 unmapped overlays, and prints the names of unmapped functions with
9647 asterisks around them. For example, if @code{foo} is a function in an
9648 unmapped overlay, @value{GDBN} prints it this way:
9651 (@value{GDBP}) overlay list
9652 No sections are mapped.
9653 (@value{GDBP}) print foo
9654 $5 = @{int (int)@} 0x100000 <*foo*>
9657 When @code{foo}'s overlay is mapped, @value{GDBN} prints the function's
9661 (@value{GDBP}) overlay list
9662 Section .ov.foo.text, loaded at 0x100000 - 0x100034,
9663 mapped at 0x1016 - 0x104a
9664 (@value{GDBP}) print foo
9665 $6 = @{int (int)@} 0x1016 <foo>
9668 When overlay debugging is enabled, @value{GDBN} can find the correct
9669 address for functions and variables in an overlay, whether or not the
9670 overlay is mapped. This allows most @value{GDBN} commands, like
9671 @code{break} and @code{disassemble}, to work normally, even on unmapped
9672 code. However, @value{GDBN}'s breakpoint support has some limitations:
9676 @cindex breakpoints in overlays
9677 @cindex overlays, setting breakpoints in
9678 You can set breakpoints in functions in unmapped overlays, as long as
9679 @value{GDBN} can write to the overlay at its load address.
9681 @value{GDBN} can not set hardware or simulator-based breakpoints in
9682 unmapped overlays. However, if you set a breakpoint at the end of your
9683 overlay manager (and tell @value{GDBN} which overlays are now mapped, if
9684 you are using manual overlay management), @value{GDBN} will re-set its
9685 breakpoints properly.
9689 @node Automatic Overlay Debugging
9690 @section Automatic Overlay Debugging
9691 @cindex automatic overlay debugging
9693 @value{GDBN} can automatically track which overlays are mapped and which
9694 are not, given some simple co-operation from the overlay manager in the
9695 inferior. If you enable automatic overlay debugging with the
9696 @code{overlay auto} command (@pxref{Overlay Commands}), @value{GDBN}
9697 looks in the inferior's memory for certain variables describing the
9698 current state of the overlays.
9700 Here are the variables your overlay manager must define to support
9701 @value{GDBN}'s automatic overlay debugging:
9705 @item @code{_ovly_table}:
9706 This variable must be an array of the following structures:
9711 /* The overlay's mapped address. */
9714 /* The size of the overlay, in bytes. */
9717 /* The overlay's load address. */
9720 /* Non-zero if the overlay is currently mapped;
9722 unsigned long mapped;
9726 @item @code{_novlys}:
9727 This variable must be a four-byte signed integer, holding the total
9728 number of elements in @code{_ovly_table}.
9732 To decide whether a particular overlay is mapped or not, @value{GDBN}
9733 looks for an entry in @w{@code{_ovly_table}} whose @code{vma} and
9734 @code{lma} members equal the VMA and LMA of the overlay's section in the
9735 executable file. When @value{GDBN} finds a matching entry, it consults
9736 the entry's @code{mapped} member to determine whether the overlay is
9739 In addition, your overlay manager may define a function called
9740 @code{_ovly_debug_event}. If this function is defined, @value{GDBN}
9741 will silently set a breakpoint there. If the overlay manager then
9742 calls this function whenever it has changed the overlay table, this
9743 will enable @value{GDBN} to accurately keep track of which overlays
9744 are in program memory, and update any breakpoints that may be set
9745 in overlays. This will allow breakpoints to work even if the
9746 overlays are kept in ROM or other non-writable memory while they
9747 are not being executed.
9749 @node Overlay Sample Program
9750 @section Overlay Sample Program
9751 @cindex overlay example program
9753 When linking a program which uses overlays, you must place the overlays
9754 at their load addresses, while relocating them to run at their mapped
9755 addresses. To do this, you must write a linker script (@pxref{Overlay
9756 Description,,, ld.info, Using ld: the GNU linker}). Unfortunately,
9757 since linker scripts are specific to a particular host system, target
9758 architecture, and target memory layout, this manual cannot provide
9759 portable sample code demonstrating @value{GDBN}'s overlay support.
9761 However, the @value{GDBN} source distribution does contain an overlaid
9762 program, with linker scripts for a few systems, as part of its test
9763 suite. The program consists of the following files from
9764 @file{gdb/testsuite/gdb.base}:
9768 The main program file.
9770 A simple overlay manager, used by @file{overlays.c}.
9775 Overlay modules, loaded and used by @file{overlays.c}.
9778 Linker scripts for linking the test program on the @code{d10v-elf}
9779 and @code{m32r-elf} targets.
9782 You can build the test program using the @code{d10v-elf} GCC
9783 cross-compiler like this:
9786 $ d10v-elf-gcc -g -c overlays.c
9787 $ d10v-elf-gcc -g -c ovlymgr.c
9788 $ d10v-elf-gcc -g -c foo.c
9789 $ d10v-elf-gcc -g -c bar.c
9790 $ d10v-elf-gcc -g -c baz.c
9791 $ d10v-elf-gcc -g -c grbx.c
9792 $ d10v-elf-gcc -g overlays.o ovlymgr.o foo.o bar.o \
9793 baz.o grbx.o -Wl,-Td10v.ld -o overlays
9796 The build process is identical for any other architecture, except that
9797 you must substitute the appropriate compiler and linker script for the
9798 target system for @code{d10v-elf-gcc} and @code{d10v.ld}.
9802 @chapter Using @value{GDBN} with Different Languages
9805 Although programming languages generally have common aspects, they are
9806 rarely expressed in the same manner. For instance, in ANSI C,
9807 dereferencing a pointer @code{p} is accomplished by @code{*p}, but in
9808 Modula-2, it is accomplished by @code{p^}. Values can also be
9809 represented (and displayed) differently. Hex numbers in C appear as
9810 @samp{0x1ae}, while in Modula-2 they appear as @samp{1AEH}.
9812 @cindex working language
9813 Language-specific information is built into @value{GDBN} for some languages,
9814 allowing you to express operations like the above in your program's
9815 native language, and allowing @value{GDBN} to output values in a manner
9816 consistent with the syntax of your program's native language. The
9817 language you use to build expressions is called the @dfn{working
9821 * Setting:: Switching between source languages
9822 * Show:: Displaying the language
9823 * Checks:: Type and range checks
9824 * Supported Languages:: Supported languages
9825 * Unsupported Languages:: Unsupported languages
9829 @section Switching Between Source Languages
9831 There are two ways to control the working language---either have @value{GDBN}
9832 set it automatically, or select it manually yourself. You can use the
9833 @code{set language} command for either purpose. On startup, @value{GDBN}
9834 defaults to setting the language automatically. The working language is
9835 used to determine how expressions you type are interpreted, how values
9838 In addition to the working language, every source file that
9839 @value{GDBN} knows about has its own working language. For some object
9840 file formats, the compiler might indicate which language a particular
9841 source file is in. However, most of the time @value{GDBN} infers the
9842 language from the name of the file. The language of a source file
9843 controls whether C@t{++} names are demangled---this way @code{backtrace} can
9844 show each frame appropriately for its own language. There is no way to
9845 set the language of a source file from within @value{GDBN}, but you can
9846 set the language associated with a filename extension. @xref{Show, ,
9847 Displaying the Language}.
9849 This is most commonly a problem when you use a program, such
9850 as @code{cfront} or @code{f2c}, that generates C but is written in
9851 another language. In that case, make the
9852 program use @code{#line} directives in its C output; that way
9853 @value{GDBN} will know the correct language of the source code of the original
9854 program, and will display that source code, not the generated C code.
9857 * Filenames:: Filename extensions and languages.
9858 * Manually:: Setting the working language manually
9859 * Automatically:: Having @value{GDBN} infer the source language
9863 @subsection List of Filename Extensions and Languages
9865 If a source file name ends in one of the following extensions, then
9866 @value{GDBN} infers that its language is the one indicated.
9887 Objective-C source file
9894 Modula-2 source file
9898 Assembler source file. This actually behaves almost like C, but
9899 @value{GDBN} does not skip over function prologues when stepping.
9902 In addition, you may set the language associated with a filename
9903 extension. @xref{Show, , Displaying the Language}.
9906 @subsection Setting the Working Language
9908 If you allow @value{GDBN} to set the language automatically,
9909 expressions are interpreted the same way in your debugging session and
9912 @kindex set language
9913 If you wish, you may set the language manually. To do this, issue the
9914 command @samp{set language @var{lang}}, where @var{lang} is the name of
9916 @code{c} or @code{modula-2}.
9917 For a list of the supported languages, type @samp{set language}.
9919 Setting the language manually prevents @value{GDBN} from updating the working
9920 language automatically. This can lead to confusion if you try
9921 to debug a program when the working language is not the same as the
9922 source language, when an expression is acceptable to both
9923 languages---but means different things. For instance, if the current
9924 source file were written in C, and @value{GDBN} was parsing Modula-2, a
9932 might not have the effect you intended. In C, this means to add
9933 @code{b} and @code{c} and place the result in @code{a}. The result
9934 printed would be the value of @code{a}. In Modula-2, this means to compare
9935 @code{a} to the result of @code{b+c}, yielding a @code{BOOLEAN} value.
9938 @subsection Having @value{GDBN} Infer the Source Language
9940 To have @value{GDBN} set the working language automatically, use
9941 @samp{set language local} or @samp{set language auto}. @value{GDBN}
9942 then infers the working language. That is, when your program stops in a
9943 frame (usually by encountering a breakpoint), @value{GDBN} sets the
9944 working language to the language recorded for the function in that
9945 frame. If the language for a frame is unknown (that is, if the function
9946 or block corresponding to the frame was defined in a source file that
9947 does not have a recognized extension), the current working language is
9948 not changed, and @value{GDBN} issues a warning.
9950 This may not seem necessary for most programs, which are written
9951 entirely in one source language. However, program modules and libraries
9952 written in one source language can be used by a main program written in
9953 a different source language. Using @samp{set language auto} in this
9954 case frees you from having to set the working language manually.
9957 @section Displaying the Language
9959 The following commands help you find out which language is the
9960 working language, and also what language source files were written in.
9964 @kindex show language
9965 Display the current working language. This is the
9966 language you can use with commands such as @code{print} to
9967 build and compute expressions that may involve variables in your program.
9970 @kindex info frame@r{, show the source language}
9971 Display the source language for this frame. This language becomes the
9972 working language if you use an identifier from this frame.
9973 @xref{Frame Info, ,Information about a Frame}, to identify the other
9974 information listed here.
9977 @kindex info source@r{, show the source language}
9978 Display the source language of this source file.
9979 @xref{Symbols, ,Examining the Symbol Table}, to identify the other
9980 information listed here.
9983 In unusual circumstances, you may have source files with extensions
9984 not in the standard list. You can then set the extension associated
9985 with a language explicitly:
9988 @item set extension-language @var{ext} @var{language}
9989 @kindex set extension-language
9990 Tell @value{GDBN} that source files with extension @var{ext} are to be
9991 assumed as written in the source language @var{language}.
9993 @item info extensions
9994 @kindex info extensions
9995 List all the filename extensions and the associated languages.
9999 @section Type and Range Checking
10002 @emph{Warning:} In this release, the @value{GDBN} commands for type and range
10003 checking are included, but they do not yet have any effect. This
10004 section documents the intended facilities.
10006 @c FIXME remove warning when type/range code added
10008 Some languages are designed to guard you against making seemingly common
10009 errors through a series of compile- and run-time checks. These include
10010 checking the type of arguments to functions and operators, and making
10011 sure mathematical overflows are caught at run time. Checks such as
10012 these help to ensure a program's correctness once it has been compiled
10013 by eliminating type mismatches, and providing active checks for range
10014 errors when your program is running.
10016 @value{GDBN} can check for conditions like the above if you wish.
10017 Although @value{GDBN} does not check the statements in your program,
10018 it can check expressions entered directly into @value{GDBN} for
10019 evaluation via the @code{print} command, for example. As with the
10020 working language, @value{GDBN} can also decide whether or not to check
10021 automatically based on your program's source language.
10022 @xref{Supported Languages, ,Supported Languages}, for the default
10023 settings of supported languages.
10026 * Type Checking:: An overview of type checking
10027 * Range Checking:: An overview of range checking
10030 @cindex type checking
10031 @cindex checks, type
10032 @node Type Checking
10033 @subsection An Overview of Type Checking
10035 Some languages, such as Modula-2, are strongly typed, meaning that the
10036 arguments to operators and functions have to be of the correct type,
10037 otherwise an error occurs. These checks prevent type mismatch
10038 errors from ever causing any run-time problems. For example,
10046 The second example fails because the @code{CARDINAL} 1 is not
10047 type-compatible with the @code{REAL} 2.3.
10049 For the expressions you use in @value{GDBN} commands, you can tell the
10050 @value{GDBN} type checker to skip checking;
10051 to treat any mismatches as errors and abandon the expression;
10052 or to only issue warnings when type mismatches occur,
10053 but evaluate the expression anyway. When you choose the last of
10054 these, @value{GDBN} evaluates expressions like the second example above, but
10055 also issues a warning.
10057 Even if you turn type checking off, there may be other reasons
10058 related to type that prevent @value{GDBN} from evaluating an expression.
10059 For instance, @value{GDBN} does not know how to add an @code{int} and
10060 a @code{struct foo}. These particular type errors have nothing to do
10061 with the language in use, and usually arise from expressions, such as
10062 the one described above, which make little sense to evaluate anyway.
10064 Each language defines to what degree it is strict about type. For
10065 instance, both Modula-2 and C require the arguments to arithmetical
10066 operators to be numbers. In C, enumerated types and pointers can be
10067 represented as numbers, so that they are valid arguments to mathematical
10068 operators. @xref{Supported Languages, ,Supported Languages}, for further
10069 details on specific languages.
10071 @value{GDBN} provides some additional commands for controlling the type checker:
10073 @kindex set check type
10074 @kindex show check type
10076 @item set check type auto
10077 Set type checking on or off based on the current working language.
10078 @xref{Supported Languages, ,Supported Languages}, for the default settings for
10081 @item set check type on
10082 @itemx set check type off
10083 Set type checking on or off, overriding the default setting for the
10084 current working language. Issue a warning if the setting does not
10085 match the language default. If any type mismatches occur in
10086 evaluating an expression while type checking is on, @value{GDBN} prints a
10087 message and aborts evaluation of the expression.
10089 @item set check type warn
10090 Cause the type checker to issue warnings, but to always attempt to
10091 evaluate the expression. Evaluating the expression may still
10092 be impossible for other reasons. For example, @value{GDBN} cannot add
10093 numbers and structures.
10096 Show the current setting of the type checker, and whether or not @value{GDBN}
10097 is setting it automatically.
10100 @cindex range checking
10101 @cindex checks, range
10102 @node Range Checking
10103 @subsection An Overview of Range Checking
10105 In some languages (such as Modula-2), it is an error to exceed the
10106 bounds of a type; this is enforced with run-time checks. Such range
10107 checking is meant to ensure program correctness by making sure
10108 computations do not overflow, or indices on an array element access do
10109 not exceed the bounds of the array.
10111 For expressions you use in @value{GDBN} commands, you can tell
10112 @value{GDBN} to treat range errors in one of three ways: ignore them,
10113 always treat them as errors and abandon the expression, or issue
10114 warnings but evaluate the expression anyway.
10116 A range error can result from numerical overflow, from exceeding an
10117 array index bound, or when you type a constant that is not a member
10118 of any type. Some languages, however, do not treat overflows as an
10119 error. In many implementations of C, mathematical overflow causes the
10120 result to ``wrap around'' to lower values---for example, if @var{m} is
10121 the largest integer value, and @var{s} is the smallest, then
10124 @var{m} + 1 @result{} @var{s}
10127 This, too, is specific to individual languages, and in some cases
10128 specific to individual compilers or machines. @xref{Supported Languages, ,
10129 Supported Languages}, for further details on specific languages.
10131 @value{GDBN} provides some additional commands for controlling the range checker:
10133 @kindex set check range
10134 @kindex show check range
10136 @item set check range auto
10137 Set range checking on or off based on the current working language.
10138 @xref{Supported Languages, ,Supported Languages}, for the default settings for
10141 @item set check range on
10142 @itemx set check range off
10143 Set range checking on or off, overriding the default setting for the
10144 current working language. A warning is issued if the setting does not
10145 match the language default. If a range error occurs and range checking is on,
10146 then a message is printed and evaluation of the expression is aborted.
10148 @item set check range warn
10149 Output messages when the @value{GDBN} range checker detects a range error,
10150 but attempt to evaluate the expression anyway. Evaluating the
10151 expression may still be impossible for other reasons, such as accessing
10152 memory that the process does not own (a typical example from many Unix
10156 Show the current setting of the range checker, and whether or not it is
10157 being set automatically by @value{GDBN}.
10160 @node Supported Languages
10161 @section Supported Languages
10163 @value{GDBN} supports C, C@t{++}, Objective-C, Fortran, Java, Pascal,
10164 assembly, Modula-2, and Ada.
10165 @c This is false ...
10166 Some @value{GDBN} features may be used in expressions regardless of the
10167 language you use: the @value{GDBN} @code{@@} and @code{::} operators,
10168 and the @samp{@{type@}addr} construct (@pxref{Expressions,
10169 ,Expressions}) can be used with the constructs of any supported
10172 The following sections detail to what degree each source language is
10173 supported by @value{GDBN}. These sections are not meant to be language
10174 tutorials or references, but serve only as a reference guide to what the
10175 @value{GDBN} expression parser accepts, and what input and output
10176 formats should look like for different languages. There are many good
10177 books written on each of these languages; please look to these for a
10178 language reference or tutorial.
10181 * C:: C and C@t{++}
10182 * Objective-C:: Objective-C
10183 * Fortran:: Fortran
10185 * Modula-2:: Modula-2
10190 @subsection C and C@t{++}
10192 @cindex C and C@t{++}
10193 @cindex expressions in C or C@t{++}
10195 Since C and C@t{++} are so closely related, many features of @value{GDBN} apply
10196 to both languages. Whenever this is the case, we discuss those languages
10200 @cindex @code{g++}, @sc{gnu} C@t{++} compiler
10201 @cindex @sc{gnu} C@t{++}
10202 The C@t{++} debugging facilities are jointly implemented by the C@t{++}
10203 compiler and @value{GDBN}. Therefore, to debug your C@t{++} code
10204 effectively, you must compile your C@t{++} programs with a supported
10205 C@t{++} compiler, such as @sc{gnu} @code{g++}, or the HP ANSI C@t{++}
10206 compiler (@code{aCC}).
10208 For best results when using @sc{gnu} C@t{++}, use the DWARF 2 debugging
10209 format; if it doesn't work on your system, try the stabs+ debugging
10210 format. You can select those formats explicitly with the @code{g++}
10211 command-line options @option{-gdwarf-2} and @option{-gstabs+}.
10212 @xref{Debugging Options,,Options for Debugging Your Program or GCC,
10213 gcc.info, Using the @sc{gnu} Compiler Collection (GCC)}.
10216 * C Operators:: C and C@t{++} operators
10217 * C Constants:: C and C@t{++} constants
10218 * C Plus Plus Expressions:: C@t{++} expressions
10219 * C Defaults:: Default settings for C and C@t{++}
10220 * C Checks:: C and C@t{++} type and range checks
10221 * Debugging C:: @value{GDBN} and C
10222 * Debugging C Plus Plus:: @value{GDBN} features for C@t{++}
10223 * Decimal Floating Point:: Numbers in Decimal Floating Point format
10227 @subsubsection C and C@t{++} Operators
10229 @cindex C and C@t{++} operators
10231 Operators must be defined on values of specific types. For instance,
10232 @code{+} is defined on numbers, but not on structures. Operators are
10233 often defined on groups of types.
10235 For the purposes of C and C@t{++}, the following definitions hold:
10240 @emph{Integral types} include @code{int} with any of its storage-class
10241 specifiers; @code{char}; @code{enum}; and, for C@t{++}, @code{bool}.
10244 @emph{Floating-point types} include @code{float}, @code{double}, and
10245 @code{long double} (if supported by the target platform).
10248 @emph{Pointer types} include all types defined as @code{(@var{type} *)}.
10251 @emph{Scalar types} include all of the above.
10256 The following operators are supported. They are listed here
10257 in order of increasing precedence:
10261 The comma or sequencing operator. Expressions in a comma-separated list
10262 are evaluated from left to right, with the result of the entire
10263 expression being the last expression evaluated.
10266 Assignment. The value of an assignment expression is the value
10267 assigned. Defined on scalar types.
10270 Used in an expression of the form @w{@code{@var{a} @var{op}= @var{b}}},
10271 and translated to @w{@code{@var{a} = @var{a op b}}}.
10272 @w{@code{@var{op}=}} and @code{=} have the same precedence.
10273 @var{op} is any one of the operators @code{|}, @code{^}, @code{&},
10274 @code{<<}, @code{>>}, @code{+}, @code{-}, @code{*}, @code{/}, @code{%}.
10277 The ternary operator. @code{@var{a} ? @var{b} : @var{c}} can be thought
10278 of as: if @var{a} then @var{b} else @var{c}. @var{a} should be of an
10282 Logical @sc{or}. Defined on integral types.
10285 Logical @sc{and}. Defined on integral types.
10288 Bitwise @sc{or}. Defined on integral types.
10291 Bitwise exclusive-@sc{or}. Defined on integral types.
10294 Bitwise @sc{and}. Defined on integral types.
10297 Equality and inequality. Defined on scalar types. The value of these
10298 expressions is 0 for false and non-zero for true.
10300 @item <@r{, }>@r{, }<=@r{, }>=
10301 Less than, greater than, less than or equal, greater than or equal.
10302 Defined on scalar types. The value of these expressions is 0 for false
10303 and non-zero for true.
10306 left shift, and right shift. Defined on integral types.
10309 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
10312 Addition and subtraction. Defined on integral types, floating-point types and
10315 @item *@r{, }/@r{, }%
10316 Multiplication, division, and modulus. Multiplication and division are
10317 defined on integral and floating-point types. Modulus is defined on
10321 Increment and decrement. When appearing before a variable, the
10322 operation is performed before the variable is used in an expression;
10323 when appearing after it, the variable's value is used before the
10324 operation takes place.
10327 Pointer dereferencing. Defined on pointer types. Same precedence as
10331 Address operator. Defined on variables. Same precedence as @code{++}.
10333 For debugging C@t{++}, @value{GDBN} implements a use of @samp{&} beyond what is
10334 allowed in the C@t{++} language itself: you can use @samp{&(&@var{ref})}
10335 to examine the address
10336 where a C@t{++} reference variable (declared with @samp{&@var{ref}}) is
10340 Negative. Defined on integral and floating-point types. Same
10341 precedence as @code{++}.
10344 Logical negation. Defined on integral types. Same precedence as
10348 Bitwise complement operator. Defined on integral types. Same precedence as
10353 Structure member, and pointer-to-structure member. For convenience,
10354 @value{GDBN} regards the two as equivalent, choosing whether to dereference a
10355 pointer based on the stored type information.
10356 Defined on @code{struct} and @code{union} data.
10359 Dereferences of pointers to members.
10362 Array indexing. @code{@var{a}[@var{i}]} is defined as
10363 @code{*(@var{a}+@var{i})}. Same precedence as @code{->}.
10366 Function parameter list. Same precedence as @code{->}.
10369 C@t{++} scope resolution operator. Defined on @code{struct}, @code{union},
10370 and @code{class} types.
10373 Doubled colons also represent the @value{GDBN} scope operator
10374 (@pxref{Expressions, ,Expressions}). Same precedence as @code{::},
10378 If an operator is redefined in the user code, @value{GDBN} usually
10379 attempts to invoke the redefined version instead of using the operator's
10380 predefined meaning.
10383 @subsubsection C and C@t{++} Constants
10385 @cindex C and C@t{++} constants
10387 @value{GDBN} allows you to express the constants of C and C@t{++} in the
10392 Integer constants are a sequence of digits. Octal constants are
10393 specified by a leading @samp{0} (i.e.@: zero), and hexadecimal constants
10394 by a leading @samp{0x} or @samp{0X}. Constants may also end with a letter
10395 @samp{l}, specifying that the constant should be treated as a
10399 Floating point constants are a sequence of digits, followed by a decimal
10400 point, followed by a sequence of digits, and optionally followed by an
10401 exponent. An exponent is of the form:
10402 @samp{@w{e@r{[[}+@r{]|}-@r{]}@var{nnn}}}, where @var{nnn} is another
10403 sequence of digits. The @samp{+} is optional for positive exponents.
10404 A floating-point constant may also end with a letter @samp{f} or
10405 @samp{F}, specifying that the constant should be treated as being of
10406 the @code{float} (as opposed to the default @code{double}) type; or with
10407 a letter @samp{l} or @samp{L}, which specifies a @code{long double}
10411 Enumerated constants consist of enumerated identifiers, or their
10412 integral equivalents.
10415 Character constants are a single character surrounded by single quotes
10416 (@code{'}), or a number---the ordinal value of the corresponding character
10417 (usually its @sc{ascii} value). Within quotes, the single character may
10418 be represented by a letter or by @dfn{escape sequences}, which are of
10419 the form @samp{\@var{nnn}}, where @var{nnn} is the octal representation
10420 of the character's ordinal value; or of the form @samp{\@var{x}}, where
10421 @samp{@var{x}} is a predefined special character---for example,
10422 @samp{\n} for newline.
10425 String constants are a sequence of character constants surrounded by
10426 double quotes (@code{"}). Any valid character constant (as described
10427 above) may appear. Double quotes within the string must be preceded by
10428 a backslash, so for instance @samp{"a\"b'c"} is a string of five
10432 Pointer constants are an integral value. You can also write pointers
10433 to constants using the C operator @samp{&}.
10436 Array constants are comma-separated lists surrounded by braces @samp{@{}
10437 and @samp{@}}; for example, @samp{@{1,2,3@}} is a three-element array of
10438 integers, @samp{@{@{1,2@}, @{3,4@}, @{5,6@}@}} is a three-by-two array,
10439 and @samp{@{&"hi", &"there", &"fred"@}} is a three-element array of pointers.
10442 @node C Plus Plus Expressions
10443 @subsubsection C@t{++} Expressions
10445 @cindex expressions in C@t{++}
10446 @value{GDBN} expression handling can interpret most C@t{++} expressions.
10448 @cindex debugging C@t{++} programs
10449 @cindex C@t{++} compilers
10450 @cindex debug formats and C@t{++}
10451 @cindex @value{NGCC} and C@t{++}
10453 @emph{Warning:} @value{GDBN} can only debug C@t{++} code if you use the
10454 proper compiler and the proper debug format. Currently, @value{GDBN}
10455 works best when debugging C@t{++} code that is compiled with
10456 @value{NGCC} 2.95.3 or with @value{NGCC} 3.1 or newer, using the options
10457 @option{-gdwarf-2} or @option{-gstabs+}. DWARF 2 is preferred over
10458 stabs+. Most configurations of @value{NGCC} emit either DWARF 2 or
10459 stabs+ as their default debug format, so you usually don't need to
10460 specify a debug format explicitly. Other compilers and/or debug formats
10461 are likely to work badly or not at all when using @value{GDBN} to debug
10467 @cindex member functions
10469 Member function calls are allowed; you can use expressions like
10472 count = aml->GetOriginal(x, y)
10475 @vindex this@r{, inside C@t{++} member functions}
10476 @cindex namespace in C@t{++}
10478 While a member function is active (in the selected stack frame), your
10479 expressions have the same namespace available as the member function;
10480 that is, @value{GDBN} allows implicit references to the class instance
10481 pointer @code{this} following the same rules as C@t{++}.
10483 @cindex call overloaded functions
10484 @cindex overloaded functions, calling
10485 @cindex type conversions in C@t{++}
10487 You can call overloaded functions; @value{GDBN} resolves the function
10488 call to the right definition, with some restrictions. @value{GDBN} does not
10489 perform overload resolution involving user-defined type conversions,
10490 calls to constructors, or instantiations of templates that do not exist
10491 in the program. It also cannot handle ellipsis argument lists or
10494 It does perform integral conversions and promotions, floating-point
10495 promotions, arithmetic conversions, pointer conversions, conversions of
10496 class objects to base classes, and standard conversions such as those of
10497 functions or arrays to pointers; it requires an exact match on the
10498 number of function arguments.
10500 Overload resolution is always performed, unless you have specified
10501 @code{set overload-resolution off}. @xref{Debugging C Plus Plus,
10502 ,@value{GDBN} Features for C@t{++}}.
10504 You must specify @code{set overload-resolution off} in order to use an
10505 explicit function signature to call an overloaded function, as in
10507 p 'foo(char,int)'('x', 13)
10510 The @value{GDBN} command-completion facility can simplify this;
10511 see @ref{Completion, ,Command Completion}.
10513 @cindex reference declarations
10515 @value{GDBN} understands variables declared as C@t{++} references; you can use
10516 them in expressions just as you do in C@t{++} source---they are automatically
10519 In the parameter list shown when @value{GDBN} displays a frame, the values of
10520 reference variables are not displayed (unlike other variables); this
10521 avoids clutter, since references are often used for large structures.
10522 The @emph{address} of a reference variable is always shown, unless
10523 you have specified @samp{set print address off}.
10526 @value{GDBN} supports the C@t{++} name resolution operator @code{::}---your
10527 expressions can use it just as expressions in your program do. Since
10528 one scope may be defined in another, you can use @code{::} repeatedly if
10529 necessary, for example in an expression like
10530 @samp{@var{scope1}::@var{scope2}::@var{name}}. @value{GDBN} also allows
10531 resolving name scope by reference to source files, in both C and C@t{++}
10532 debugging (@pxref{Variables, ,Program Variables}).
10535 In addition, when used with HP's C@t{++} compiler, @value{GDBN} supports
10536 calling virtual functions correctly, printing out virtual bases of
10537 objects, calling functions in a base subobject, casting objects, and
10538 invoking user-defined operators.
10541 @subsubsection C and C@t{++} Defaults
10543 @cindex C and C@t{++} defaults
10545 If you allow @value{GDBN} to set type and range checking automatically, they
10546 both default to @code{off} whenever the working language changes to
10547 C or C@t{++}. This happens regardless of whether you or @value{GDBN}
10548 selects the working language.
10550 If you allow @value{GDBN} to set the language automatically, it
10551 recognizes source files whose names end with @file{.c}, @file{.C}, or
10552 @file{.cc}, etc, and when @value{GDBN} enters code compiled from one of
10553 these files, it sets the working language to C or C@t{++}.
10554 @xref{Automatically, ,Having @value{GDBN} Infer the Source Language},
10555 for further details.
10557 @c Type checking is (a) primarily motivated by Modula-2, and (b)
10558 @c unimplemented. If (b) changes, it might make sense to let this node
10559 @c appear even if Mod-2 does not, but meanwhile ignore it. roland 16jul93.
10562 @subsubsection C and C@t{++} Type and Range Checks
10564 @cindex C and C@t{++} checks
10566 By default, when @value{GDBN} parses C or C@t{++} expressions, type checking
10567 is not used. However, if you turn type checking on, @value{GDBN}
10568 considers two variables type equivalent if:
10572 The two variables are structured and have the same structure, union, or
10576 The two variables have the same type name, or types that have been
10577 declared equivalent through @code{typedef}.
10580 @c leaving this out because neither J Gilmore nor R Pesch understand it.
10583 The two @code{struct}, @code{union}, or @code{enum} variables are
10584 declared in the same declaration. (Note: this may not be true for all C
10589 Range checking, if turned on, is done on mathematical operations. Array
10590 indices are not checked, since they are often used to index a pointer
10591 that is not itself an array.
10594 @subsubsection @value{GDBN} and C
10596 The @code{set print union} and @code{show print union} commands apply to
10597 the @code{union} type. When set to @samp{on}, any @code{union} that is
10598 inside a @code{struct} or @code{class} is also printed. Otherwise, it
10599 appears as @samp{@{...@}}.
10601 The @code{@@} operator aids in the debugging of dynamic arrays, formed
10602 with pointers and a memory allocation function. @xref{Expressions,
10605 @node Debugging C Plus Plus
10606 @subsubsection @value{GDBN} Features for C@t{++}
10608 @cindex commands for C@t{++}
10610 Some @value{GDBN} commands are particularly useful with C@t{++}, and some are
10611 designed specifically for use with C@t{++}. Here is a summary:
10614 @cindex break in overloaded functions
10615 @item @r{breakpoint menus}
10616 When you want a breakpoint in a function whose name is overloaded,
10617 @value{GDBN} has the capability to display a menu of possible breakpoint
10618 locations to help you specify which function definition you want.
10619 @xref{Ambiguous Expressions,,Ambiguous Expressions}.
10621 @cindex overloading in C@t{++}
10622 @item rbreak @var{regex}
10623 Setting breakpoints using regular expressions is helpful for setting
10624 breakpoints on overloaded functions that are not members of any special
10626 @xref{Set Breaks, ,Setting Breakpoints}.
10628 @cindex C@t{++} exception handling
10631 Debug C@t{++} exception handling using these commands. @xref{Set
10632 Catchpoints, , Setting Catchpoints}.
10634 @cindex inheritance
10635 @item ptype @var{typename}
10636 Print inheritance relationships as well as other information for type
10638 @xref{Symbols, ,Examining the Symbol Table}.
10640 @cindex C@t{++} symbol display
10641 @item set print demangle
10642 @itemx show print demangle
10643 @itemx set print asm-demangle
10644 @itemx show print asm-demangle
10645 Control whether C@t{++} symbols display in their source form, both when
10646 displaying code as C@t{++} source and when displaying disassemblies.
10647 @xref{Print Settings, ,Print Settings}.
10649 @item set print object
10650 @itemx show print object
10651 Choose whether to print derived (actual) or declared types of objects.
10652 @xref{Print Settings, ,Print Settings}.
10654 @item set print vtbl
10655 @itemx show print vtbl
10656 Control the format for printing virtual function tables.
10657 @xref{Print Settings, ,Print Settings}.
10658 (The @code{vtbl} commands do not work on programs compiled with the HP
10659 ANSI C@t{++} compiler (@code{aCC}).)
10661 @kindex set overload-resolution
10662 @cindex overloaded functions, overload resolution
10663 @item set overload-resolution on
10664 Enable overload resolution for C@t{++} expression evaluation. The default
10665 is on. For overloaded functions, @value{GDBN} evaluates the arguments
10666 and searches for a function whose signature matches the argument types,
10667 using the standard C@t{++} conversion rules (see @ref{C Plus Plus
10668 Expressions, ,C@t{++} Expressions}, for details).
10669 If it cannot find a match, it emits a message.
10671 @item set overload-resolution off
10672 Disable overload resolution for C@t{++} expression evaluation. For
10673 overloaded functions that are not class member functions, @value{GDBN}
10674 chooses the first function of the specified name that it finds in the
10675 symbol table, whether or not its arguments are of the correct type. For
10676 overloaded functions that are class member functions, @value{GDBN}
10677 searches for a function whose signature @emph{exactly} matches the
10680 @kindex show overload-resolution
10681 @item show overload-resolution
10682 Show the current setting of overload resolution.
10684 @item @r{Overloaded symbol names}
10685 You can specify a particular definition of an overloaded symbol, using
10686 the same notation that is used to declare such symbols in C@t{++}: type
10687 @code{@var{symbol}(@var{types})} rather than just @var{symbol}. You can
10688 also use the @value{GDBN} command-line word completion facilities to list the
10689 available choices, or to finish the type list for you.
10690 @xref{Completion,, Command Completion}, for details on how to do this.
10693 @node Decimal Floating Point
10694 @subsubsection Decimal Floating Point format
10695 @cindex decimal floating point format
10697 @value{GDBN} can examine, set and perform computations with numbers in
10698 decimal floating point format, which in the C language correspond to the
10699 @code{_Decimal32}, @code{_Decimal64} and @code{_Decimal128} types as
10700 specified by the extension to support decimal floating-point arithmetic.
10702 There are two encodings in use, depending on the architecture: BID (Binary
10703 Integer Decimal) for x86 and x86-64, and DPD (Densely Packed Decimal) for
10704 PowerPC. @value{GDBN} will use the appropriate encoding for the configured
10707 Because of a limitation in @file{libdecnumber}, the library used by @value{GDBN}
10708 to manipulate decimal floating point numbers, it is not possible to convert
10709 (using a cast, for example) integers wider than 32-bit to decimal float.
10711 In addition, in order to imitate @value{GDBN}'s behaviour with binary floating
10712 point computations, error checking in decimal float operations ignores
10713 underflow, overflow and divide by zero exceptions.
10715 In the PowerPC architecture, @value{GDBN} provides a set of pseudo-registers
10716 to inspect @code{_Decimal128} values stored in floating point registers. See
10717 @ref{PowerPC,,PowerPC} for more details.
10720 @subsection Objective-C
10722 @cindex Objective-C
10723 This section provides information about some commands and command
10724 options that are useful for debugging Objective-C code. See also
10725 @ref{Symbols, info classes}, and @ref{Symbols, info selectors}, for a
10726 few more commands specific to Objective-C support.
10729 * Method Names in Commands::
10730 * The Print Command with Objective-C::
10733 @node Method Names in Commands
10734 @subsubsection Method Names in Commands
10736 The following commands have been extended to accept Objective-C method
10737 names as line specifications:
10739 @kindex clear@r{, and Objective-C}
10740 @kindex break@r{, and Objective-C}
10741 @kindex info line@r{, and Objective-C}
10742 @kindex jump@r{, and Objective-C}
10743 @kindex list@r{, and Objective-C}
10747 @item @code{info line}
10752 A fully qualified Objective-C method name is specified as
10755 -[@var{Class} @var{methodName}]
10758 where the minus sign is used to indicate an instance method and a
10759 plus sign (not shown) is used to indicate a class method. The class
10760 name @var{Class} and method name @var{methodName} are enclosed in
10761 brackets, similar to the way messages are specified in Objective-C
10762 source code. For example, to set a breakpoint at the @code{create}
10763 instance method of class @code{Fruit} in the program currently being
10767 break -[Fruit create]
10770 To list ten program lines around the @code{initialize} class method,
10774 list +[NSText initialize]
10777 In the current version of @value{GDBN}, the plus or minus sign is
10778 required. In future versions of @value{GDBN}, the plus or minus
10779 sign will be optional, but you can use it to narrow the search. It
10780 is also possible to specify just a method name:
10786 You must specify the complete method name, including any colons. If
10787 your program's source files contain more than one @code{create} method,
10788 you'll be presented with a numbered list of classes that implement that
10789 method. Indicate your choice by number, or type @samp{0} to exit if
10792 As another example, to clear a breakpoint established at the
10793 @code{makeKeyAndOrderFront:} method of the @code{NSWindow} class, enter:
10796 clear -[NSWindow makeKeyAndOrderFront:]
10799 @node The Print Command with Objective-C
10800 @subsubsection The Print Command With Objective-C
10801 @cindex Objective-C, print objects
10802 @kindex print-object
10803 @kindex po @r{(@code{print-object})}
10805 The print command has also been extended to accept methods. For example:
10808 print -[@var{object} hash]
10811 @cindex print an Objective-C object description
10812 @cindex @code{_NSPrintForDebugger}, and printing Objective-C objects
10814 will tell @value{GDBN} to send the @code{hash} message to @var{object}
10815 and print the result. Also, an additional command has been added,
10816 @code{print-object} or @code{po} for short, which is meant to print
10817 the description of an object. However, this command may only work
10818 with certain Objective-C libraries that have a particular hook
10819 function, @code{_NSPrintForDebugger}, defined.
10822 @subsection Fortran
10823 @cindex Fortran-specific support in @value{GDBN}
10825 @value{GDBN} can be used to debug programs written in Fortran, but it
10826 currently supports only the features of Fortran 77 language.
10828 @cindex trailing underscore, in Fortran symbols
10829 Some Fortran compilers (@sc{gnu} Fortran 77 and Fortran 95 compilers
10830 among them) append an underscore to the names of variables and
10831 functions. When you debug programs compiled by those compilers, you
10832 will need to refer to variables and functions with a trailing
10836 * Fortran Operators:: Fortran operators and expressions
10837 * Fortran Defaults:: Default settings for Fortran
10838 * Special Fortran Commands:: Special @value{GDBN} commands for Fortran
10841 @node Fortran Operators
10842 @subsubsection Fortran Operators and Expressions
10844 @cindex Fortran operators and expressions
10846 Operators must be defined on values of specific types. For instance,
10847 @code{+} is defined on numbers, but not on characters or other non-
10848 arithmetic types. Operators are often defined on groups of types.
10852 The exponentiation operator. It raises the first operand to the power
10856 The range operator. Normally used in the form of array(low:high) to
10857 represent a section of array.
10860 The access component operator. Normally used to access elements in derived
10861 types. Also suitable for unions. As unions aren't part of regular Fortran,
10862 this can only happen when accessing a register that uses a gdbarch-defined
10866 @node Fortran Defaults
10867 @subsubsection Fortran Defaults
10869 @cindex Fortran Defaults
10871 Fortran symbols are usually case-insensitive, so @value{GDBN} by
10872 default uses case-insensitive matches for Fortran symbols. You can
10873 change that with the @samp{set case-insensitive} command, see
10874 @ref{Symbols}, for the details.
10876 @node Special Fortran Commands
10877 @subsubsection Special Fortran Commands
10879 @cindex Special Fortran commands
10881 @value{GDBN} has some commands to support Fortran-specific features,
10882 such as displaying common blocks.
10885 @cindex @code{COMMON} blocks, Fortran
10886 @kindex info common
10887 @item info common @r{[}@var{common-name}@r{]}
10888 This command prints the values contained in the Fortran @code{COMMON}
10889 block whose name is @var{common-name}. With no argument, the names of
10890 all @code{COMMON} blocks visible at the current program location are
10897 @cindex Pascal support in @value{GDBN}, limitations
10898 Debugging Pascal programs which use sets, subranges, file variables, or
10899 nested functions does not currently work. @value{GDBN} does not support
10900 entering expressions, printing values, or similar features using Pascal
10903 The Pascal-specific command @code{set print pascal_static-members}
10904 controls whether static members of Pascal objects are displayed.
10905 @xref{Print Settings, pascal_static-members}.
10908 @subsection Modula-2
10910 @cindex Modula-2, @value{GDBN} support
10912 The extensions made to @value{GDBN} to support Modula-2 only support
10913 output from the @sc{gnu} Modula-2 compiler (which is currently being
10914 developed). Other Modula-2 compilers are not currently supported, and
10915 attempting to debug executables produced by them is most likely
10916 to give an error as @value{GDBN} reads in the executable's symbol
10919 @cindex expressions in Modula-2
10921 * M2 Operators:: Built-in operators
10922 * Built-In Func/Proc:: Built-in functions and procedures
10923 * M2 Constants:: Modula-2 constants
10924 * M2 Types:: Modula-2 types
10925 * M2 Defaults:: Default settings for Modula-2
10926 * Deviations:: Deviations from standard Modula-2
10927 * M2 Checks:: Modula-2 type and range checks
10928 * M2 Scope:: The scope operators @code{::} and @code{.}
10929 * GDB/M2:: @value{GDBN} and Modula-2
10933 @subsubsection Operators
10934 @cindex Modula-2 operators
10936 Operators must be defined on values of specific types. For instance,
10937 @code{+} is defined on numbers, but not on structures. Operators are
10938 often defined on groups of types. For the purposes of Modula-2, the
10939 following definitions hold:
10944 @emph{Integral types} consist of @code{INTEGER}, @code{CARDINAL}, and
10948 @emph{Character types} consist of @code{CHAR} and its subranges.
10951 @emph{Floating-point types} consist of @code{REAL}.
10954 @emph{Pointer types} consist of anything declared as @code{POINTER TO
10958 @emph{Scalar types} consist of all of the above.
10961 @emph{Set types} consist of @code{SET} and @code{BITSET} types.
10964 @emph{Boolean types} consist of @code{BOOLEAN}.
10968 The following operators are supported, and appear in order of
10969 increasing precedence:
10973 Function argument or array index separator.
10976 Assignment. The value of @var{var} @code{:=} @var{value} is
10980 Less than, greater than on integral, floating-point, or enumerated
10984 Less than or equal to, greater than or equal to
10985 on integral, floating-point and enumerated types, or set inclusion on
10986 set types. Same precedence as @code{<}.
10988 @item =@r{, }<>@r{, }#
10989 Equality and two ways of expressing inequality, valid on scalar types.
10990 Same precedence as @code{<}. In @value{GDBN} scripts, only @code{<>} is
10991 available for inequality, since @code{#} conflicts with the script
10995 Set membership. Defined on set types and the types of their members.
10996 Same precedence as @code{<}.
10999 Boolean disjunction. Defined on boolean types.
11002 Boolean conjunction. Defined on boolean types.
11005 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
11008 Addition and subtraction on integral and floating-point types, or union
11009 and difference on set types.
11012 Multiplication on integral and floating-point types, or set intersection
11016 Division on floating-point types, or symmetric set difference on set
11017 types. Same precedence as @code{*}.
11020 Integer division and remainder. Defined on integral types. Same
11021 precedence as @code{*}.
11024 Negative. Defined on @code{INTEGER} and @code{REAL} data.
11027 Pointer dereferencing. Defined on pointer types.
11030 Boolean negation. Defined on boolean types. Same precedence as
11034 @code{RECORD} field selector. Defined on @code{RECORD} data. Same
11035 precedence as @code{^}.
11038 Array indexing. Defined on @code{ARRAY} data. Same precedence as @code{^}.
11041 Procedure argument list. Defined on @code{PROCEDURE} objects. Same precedence
11045 @value{GDBN} and Modula-2 scope operators.
11049 @emph{Warning:} Set expressions and their operations are not yet supported, so @value{GDBN}
11050 treats the use of the operator @code{IN}, or the use of operators
11051 @code{+}, @code{-}, @code{*}, @code{/}, @code{=}, , @code{<>}, @code{#},
11052 @code{<=}, and @code{>=} on sets as an error.
11056 @node Built-In Func/Proc
11057 @subsubsection Built-in Functions and Procedures
11058 @cindex Modula-2 built-ins
11060 Modula-2 also makes available several built-in procedures and functions.
11061 In describing these, the following metavariables are used:
11066 represents an @code{ARRAY} variable.
11069 represents a @code{CHAR} constant or variable.
11072 represents a variable or constant of integral type.
11075 represents an identifier that belongs to a set. Generally used in the
11076 same function with the metavariable @var{s}. The type of @var{s} should
11077 be @code{SET OF @var{mtype}} (where @var{mtype} is the type of @var{m}).
11080 represents a variable or constant of integral or floating-point type.
11083 represents a variable or constant of floating-point type.
11089 represents a variable.
11092 represents a variable or constant of one of many types. See the
11093 explanation of the function for details.
11096 All Modula-2 built-in procedures also return a result, described below.
11100 Returns the absolute value of @var{n}.
11103 If @var{c} is a lower case letter, it returns its upper case
11104 equivalent, otherwise it returns its argument.
11107 Returns the character whose ordinal value is @var{i}.
11110 Decrements the value in the variable @var{v} by one. Returns the new value.
11112 @item DEC(@var{v},@var{i})
11113 Decrements the value in the variable @var{v} by @var{i}. Returns the
11116 @item EXCL(@var{m},@var{s})
11117 Removes the element @var{m} from the set @var{s}. Returns the new
11120 @item FLOAT(@var{i})
11121 Returns the floating point equivalent of the integer @var{i}.
11123 @item HIGH(@var{a})
11124 Returns the index of the last member of @var{a}.
11127 Increments the value in the variable @var{v} by one. Returns the new value.
11129 @item INC(@var{v},@var{i})
11130 Increments the value in the variable @var{v} by @var{i}. Returns the
11133 @item INCL(@var{m},@var{s})
11134 Adds the element @var{m} to the set @var{s} if it is not already
11135 there. Returns the new set.
11138 Returns the maximum value of the type @var{t}.
11141 Returns the minimum value of the type @var{t}.
11144 Returns boolean TRUE if @var{i} is an odd number.
11147 Returns the ordinal value of its argument. For example, the ordinal
11148 value of a character is its @sc{ascii} value (on machines supporting the
11149 @sc{ascii} character set). @var{x} must be of an ordered type, which include
11150 integral, character and enumerated types.
11152 @item SIZE(@var{x})
11153 Returns the size of its argument. @var{x} can be a variable or a type.
11155 @item TRUNC(@var{r})
11156 Returns the integral part of @var{r}.
11158 @item TSIZE(@var{x})
11159 Returns the size of its argument. @var{x} can be a variable or a type.
11161 @item VAL(@var{t},@var{i})
11162 Returns the member of the type @var{t} whose ordinal value is @var{i}.
11166 @emph{Warning:} Sets and their operations are not yet supported, so
11167 @value{GDBN} treats the use of procedures @code{INCL} and @code{EXCL} as
11171 @cindex Modula-2 constants
11173 @subsubsection Constants
11175 @value{GDBN} allows you to express the constants of Modula-2 in the following
11181 Integer constants are simply a sequence of digits. When used in an
11182 expression, a constant is interpreted to be type-compatible with the
11183 rest of the expression. Hexadecimal integers are specified by a
11184 trailing @samp{H}, and octal integers by a trailing @samp{B}.
11187 Floating point constants appear as a sequence of digits, followed by a
11188 decimal point and another sequence of digits. An optional exponent can
11189 then be specified, in the form @samp{E@r{[}+@r{|}-@r{]}@var{nnn}}, where
11190 @samp{@r{[}+@r{|}-@r{]}@var{nnn}} is the desired exponent. All of the
11191 digits of the floating point constant must be valid decimal (base 10)
11195 Character constants consist of a single character enclosed by a pair of
11196 like quotes, either single (@code{'}) or double (@code{"}). They may
11197 also be expressed by their ordinal value (their @sc{ascii} value, usually)
11198 followed by a @samp{C}.
11201 String constants consist of a sequence of characters enclosed by a
11202 pair of like quotes, either single (@code{'}) or double (@code{"}).
11203 Escape sequences in the style of C are also allowed. @xref{C
11204 Constants, ,C and C@t{++} Constants}, for a brief explanation of escape
11208 Enumerated constants consist of an enumerated identifier.
11211 Boolean constants consist of the identifiers @code{TRUE} and
11215 Pointer constants consist of integral values only.
11218 Set constants are not yet supported.
11222 @subsubsection Modula-2 Types
11223 @cindex Modula-2 types
11225 Currently @value{GDBN} can print the following data types in Modula-2
11226 syntax: array types, record types, set types, pointer types, procedure
11227 types, enumerated types, subrange types and base types. You can also
11228 print the contents of variables declared using these type.
11229 This section gives a number of simple source code examples together with
11230 sample @value{GDBN} sessions.
11232 The first example contains the following section of code:
11241 and you can request @value{GDBN} to interrogate the type and value of
11242 @code{r} and @code{s}.
11245 (@value{GDBP}) print s
11247 (@value{GDBP}) ptype s
11249 (@value{GDBP}) print r
11251 (@value{GDBP}) ptype r
11256 Likewise if your source code declares @code{s} as:
11260 s: SET ['A'..'Z'] ;
11264 then you may query the type of @code{s} by:
11267 (@value{GDBP}) ptype s
11268 type = SET ['A'..'Z']
11272 Note that at present you cannot interactively manipulate set
11273 expressions using the debugger.
11275 The following example shows how you might declare an array in Modula-2
11276 and how you can interact with @value{GDBN} to print its type and contents:
11280 s: ARRAY [-10..10] OF CHAR ;
11284 (@value{GDBP}) ptype s
11285 ARRAY [-10..10] OF CHAR
11288 Note that the array handling is not yet complete and although the type
11289 is printed correctly, expression handling still assumes that all
11290 arrays have a lower bound of zero and not @code{-10} as in the example
11293 Here are some more type related Modula-2 examples:
11297 colour = (blue, red, yellow, green) ;
11298 t = [blue..yellow] ;
11306 The @value{GDBN} interaction shows how you can query the data type
11307 and value of a variable.
11310 (@value{GDBP}) print s
11312 (@value{GDBP}) ptype t
11313 type = [blue..yellow]
11317 In this example a Modula-2 array is declared and its contents
11318 displayed. Observe that the contents are written in the same way as
11319 their @code{C} counterparts.
11323 s: ARRAY [1..5] OF CARDINAL ;
11329 (@value{GDBP}) print s
11330 $1 = @{1, 0, 0, 0, 0@}
11331 (@value{GDBP}) ptype s
11332 type = ARRAY [1..5] OF CARDINAL
11335 The Modula-2 language interface to @value{GDBN} also understands
11336 pointer types as shown in this example:
11340 s: POINTER TO ARRAY [1..5] OF CARDINAL ;
11347 and you can request that @value{GDBN} describes the type of @code{s}.
11350 (@value{GDBP}) ptype s
11351 type = POINTER TO ARRAY [1..5] OF CARDINAL
11354 @value{GDBN} handles compound types as we can see in this example.
11355 Here we combine array types, record types, pointer types and subrange
11366 myarray = ARRAY myrange OF CARDINAL ;
11367 myrange = [-2..2] ;
11369 s: POINTER TO ARRAY myrange OF foo ;
11373 and you can ask @value{GDBN} to describe the type of @code{s} as shown
11377 (@value{GDBP}) ptype s
11378 type = POINTER TO ARRAY [-2..2] OF foo = RECORD
11381 f3 : ARRAY [-2..2] OF CARDINAL;
11386 @subsubsection Modula-2 Defaults
11387 @cindex Modula-2 defaults
11389 If type and range checking are set automatically by @value{GDBN}, they
11390 both default to @code{on} whenever the working language changes to
11391 Modula-2. This happens regardless of whether you or @value{GDBN}
11392 selected the working language.
11394 If you allow @value{GDBN} to set the language automatically, then entering
11395 code compiled from a file whose name ends with @file{.mod} sets the
11396 working language to Modula-2. @xref{Automatically, ,Having @value{GDBN}
11397 Infer the Source Language}, for further details.
11400 @subsubsection Deviations from Standard Modula-2
11401 @cindex Modula-2, deviations from
11403 A few changes have been made to make Modula-2 programs easier to debug.
11404 This is done primarily via loosening its type strictness:
11408 Unlike in standard Modula-2, pointer constants can be formed by
11409 integers. This allows you to modify pointer variables during
11410 debugging. (In standard Modula-2, the actual address contained in a
11411 pointer variable is hidden from you; it can only be modified
11412 through direct assignment to another pointer variable or expression that
11413 returned a pointer.)
11416 C escape sequences can be used in strings and characters to represent
11417 non-printable characters. @value{GDBN} prints out strings with these
11418 escape sequences embedded. Single non-printable characters are
11419 printed using the @samp{CHR(@var{nnn})} format.
11422 The assignment operator (@code{:=}) returns the value of its right-hand
11426 All built-in procedures both modify @emph{and} return their argument.
11430 @subsubsection Modula-2 Type and Range Checks
11431 @cindex Modula-2 checks
11434 @emph{Warning:} in this release, @value{GDBN} does not yet perform type or
11437 @c FIXME remove warning when type/range checks added
11439 @value{GDBN} considers two Modula-2 variables type equivalent if:
11443 They are of types that have been declared equivalent via a @code{TYPE
11444 @var{t1} = @var{t2}} statement
11447 They have been declared on the same line. (Note: This is true of the
11448 @sc{gnu} Modula-2 compiler, but it may not be true of other compilers.)
11451 As long as type checking is enabled, any attempt to combine variables
11452 whose types are not equivalent is an error.
11454 Range checking is done on all mathematical operations, assignment, array
11455 index bounds, and all built-in functions and procedures.
11458 @subsubsection The Scope Operators @code{::} and @code{.}
11460 @cindex @code{.}, Modula-2 scope operator
11461 @cindex colon, doubled as scope operator
11463 @vindex colon-colon@r{, in Modula-2}
11464 @c Info cannot handle :: but TeX can.
11467 @vindex ::@r{, in Modula-2}
11470 There are a few subtle differences between the Modula-2 scope operator
11471 (@code{.}) and the @value{GDBN} scope operator (@code{::}). The two have
11476 @var{module} . @var{id}
11477 @var{scope} :: @var{id}
11481 where @var{scope} is the name of a module or a procedure,
11482 @var{module} the name of a module, and @var{id} is any declared
11483 identifier within your program, except another module.
11485 Using the @code{::} operator makes @value{GDBN} search the scope
11486 specified by @var{scope} for the identifier @var{id}. If it is not
11487 found in the specified scope, then @value{GDBN} searches all scopes
11488 enclosing the one specified by @var{scope}.
11490 Using the @code{.} operator makes @value{GDBN} search the current scope for
11491 the identifier specified by @var{id} that was imported from the
11492 definition module specified by @var{module}. With this operator, it is
11493 an error if the identifier @var{id} was not imported from definition
11494 module @var{module}, or if @var{id} is not an identifier in
11498 @subsubsection @value{GDBN} and Modula-2
11500 Some @value{GDBN} commands have little use when debugging Modula-2 programs.
11501 Five subcommands of @code{set print} and @code{show print} apply
11502 specifically to C and C@t{++}: @samp{vtbl}, @samp{demangle},
11503 @samp{asm-demangle}, @samp{object}, and @samp{union}. The first four
11504 apply to C@t{++}, and the last to the C @code{union} type, which has no direct
11505 analogue in Modula-2.
11507 The @code{@@} operator (@pxref{Expressions, ,Expressions}), while available
11508 with any language, is not useful with Modula-2. Its
11509 intent is to aid the debugging of @dfn{dynamic arrays}, which cannot be
11510 created in Modula-2 as they can in C or C@t{++}. However, because an
11511 address can be specified by an integral constant, the construct
11512 @samp{@{@var{type}@}@var{adrexp}} is still useful.
11514 @cindex @code{#} in Modula-2
11515 In @value{GDBN} scripts, the Modula-2 inequality operator @code{#} is
11516 interpreted as the beginning of a comment. Use @code{<>} instead.
11522 The extensions made to @value{GDBN} for Ada only support
11523 output from the @sc{gnu} Ada (GNAT) compiler.
11524 Other Ada compilers are not currently supported, and
11525 attempting to debug executables produced by them is most likely
11529 @cindex expressions in Ada
11531 * Ada Mode Intro:: General remarks on the Ada syntax
11532 and semantics supported by Ada mode
11534 * Omissions from Ada:: Restrictions on the Ada expression syntax.
11535 * Additions to Ada:: Extensions of the Ada expression syntax.
11536 * Stopping Before Main Program:: Debugging the program during elaboration.
11537 * Ada Tasks:: Listing and setting breakpoints in tasks.
11538 * Ada Tasks and Core Files:: Tasking Support when Debugging Core Files
11539 * Ada Glitches:: Known peculiarities of Ada mode.
11542 @node Ada Mode Intro
11543 @subsubsection Introduction
11544 @cindex Ada mode, general
11546 The Ada mode of @value{GDBN} supports a fairly large subset of Ada expression
11547 syntax, with some extensions.
11548 The philosophy behind the design of this subset is
11552 That @value{GDBN} should provide basic literals and access to operations for
11553 arithmetic, dereferencing, field selection, indexing, and subprogram calls,
11554 leaving more sophisticated computations to subprograms written into the
11555 program (which therefore may be called from @value{GDBN}).
11558 That type safety and strict adherence to Ada language restrictions
11559 are not particularly important to the @value{GDBN} user.
11562 That brevity is important to the @value{GDBN} user.
11565 Thus, for brevity, the debugger acts as if all names declared in
11566 user-written packages are directly visible, even if they are not visible
11567 according to Ada rules, thus making it unnecessary to fully qualify most
11568 names with their packages, regardless of context. Where this causes
11569 ambiguity, @value{GDBN} asks the user's intent.
11571 The debugger will start in Ada mode if it detects an Ada main program.
11572 As for other languages, it will enter Ada mode when stopped in a program that
11573 was translated from an Ada source file.
11575 While in Ada mode, you may use `@t{--}' for comments. This is useful
11576 mostly for documenting command files. The standard @value{GDBN} comment
11577 (@samp{#}) still works at the beginning of a line in Ada mode, but not in the
11578 middle (to allow based literals).
11580 The debugger supports limited overloading. Given a subprogram call in which
11581 the function symbol has multiple definitions, it will use the number of
11582 actual parameters and some information about their types to attempt to narrow
11583 the set of definitions. It also makes very limited use of context, preferring
11584 procedures to functions in the context of the @code{call} command, and
11585 functions to procedures elsewhere.
11587 @node Omissions from Ada
11588 @subsubsection Omissions from Ada
11589 @cindex Ada, omissions from
11591 Here are the notable omissions from the subset:
11595 Only a subset of the attributes are supported:
11599 @t{'First}, @t{'Last}, and @t{'Length}
11600 on array objects (not on types and subtypes).
11603 @t{'Min} and @t{'Max}.
11606 @t{'Pos} and @t{'Val}.
11612 @t{'Range} on array objects (not subtypes), but only as the right
11613 operand of the membership (@code{in}) operator.
11616 @t{'Access}, @t{'Unchecked_Access}, and
11617 @t{'Unrestricted_Access} (a GNAT extension).
11625 @code{Characters.Latin_1} are not available and
11626 concatenation is not implemented. Thus, escape characters in strings are
11627 not currently available.
11630 Equality tests (@samp{=} and @samp{/=}) on arrays test for bitwise
11631 equality of representations. They will generally work correctly
11632 for strings and arrays whose elements have integer or enumeration types.
11633 They may not work correctly for arrays whose element
11634 types have user-defined equality, for arrays of real values
11635 (in particular, IEEE-conformant floating point, because of negative
11636 zeroes and NaNs), and for arrays whose elements contain unused bits with
11637 indeterminate values.
11640 The other component-by-component array operations (@code{and}, @code{or},
11641 @code{xor}, @code{not}, and relational tests other than equality)
11642 are not implemented.
11645 @cindex array aggregates (Ada)
11646 @cindex record aggregates (Ada)
11647 @cindex aggregates (Ada)
11648 There is limited support for array and record aggregates. They are
11649 permitted only on the right sides of assignments, as in these examples:
11652 (@value{GDBP}) set An_Array := (1, 2, 3, 4, 5, 6)
11653 (@value{GDBP}) set An_Array := (1, others => 0)
11654 (@value{GDBP}) set An_Array := (0|4 => 1, 1..3 => 2, 5 => 6)
11655 (@value{GDBP}) set A_2D_Array := ((1, 2, 3), (4, 5, 6), (7, 8, 9))
11656 (@value{GDBP}) set A_Record := (1, "Peter", True);
11657 (@value{GDBP}) set A_Record := (Name => "Peter", Id => 1, Alive => True)
11661 discriminant's value by assigning an aggregate has an
11662 undefined effect if that discriminant is used within the record.
11663 However, you can first modify discriminants by directly assigning to
11664 them (which normally would not be allowed in Ada), and then performing an
11665 aggregate assignment. For example, given a variable @code{A_Rec}
11666 declared to have a type such as:
11669 type Rec (Len : Small_Integer := 0) is record
11671 Vals : IntArray (1 .. Len);
11675 you can assign a value with a different size of @code{Vals} with two
11679 (@value{GDBP}) set A_Rec.Len := 4
11680 (@value{GDBP}) set A_Rec := (Id => 42, Vals => (1, 2, 3, 4))
11683 As this example also illustrates, @value{GDBN} is very loose about the usual
11684 rules concerning aggregates. You may leave out some of the
11685 components of an array or record aggregate (such as the @code{Len}
11686 component in the assignment to @code{A_Rec} above); they will retain their
11687 original values upon assignment. You may freely use dynamic values as
11688 indices in component associations. You may even use overlapping or
11689 redundant component associations, although which component values are
11690 assigned in such cases is not defined.
11693 Calls to dispatching subprograms are not implemented.
11696 The overloading algorithm is much more limited (i.e., less selective)
11697 than that of real Ada. It makes only limited use of the context in
11698 which a subexpression appears to resolve its meaning, and it is much
11699 looser in its rules for allowing type matches. As a result, some
11700 function calls will be ambiguous, and the user will be asked to choose
11701 the proper resolution.
11704 The @code{new} operator is not implemented.
11707 Entry calls are not implemented.
11710 Aside from printing, arithmetic operations on the native VAX floating-point
11711 formats are not supported.
11714 It is not possible to slice a packed array.
11717 The names @code{True} and @code{False}, when not part of a qualified name,
11718 are interpreted as if implicitly prefixed by @code{Standard}, regardless of
11720 Should your program
11721 redefine these names in a package or procedure (at best a dubious practice),
11722 you will have to use fully qualified names to access their new definitions.
11725 @node Additions to Ada
11726 @subsubsection Additions to Ada
11727 @cindex Ada, deviations from
11729 As it does for other languages, @value{GDBN} makes certain generic
11730 extensions to Ada (@pxref{Expressions}):
11734 If the expression @var{E} is a variable residing in memory (typically
11735 a local variable or array element) and @var{N} is a positive integer,
11736 then @code{@var{E}@@@var{N}} displays the values of @var{E} and the
11737 @var{N}-1 adjacent variables following it in memory as an array. In
11738 Ada, this operator is generally not necessary, since its prime use is
11739 in displaying parts of an array, and slicing will usually do this in
11740 Ada. However, there are occasional uses when debugging programs in
11741 which certain debugging information has been optimized away.
11744 @code{@var{B}::@var{var}} means ``the variable named @var{var} that
11745 appears in function or file @var{B}.'' When @var{B} is a file name,
11746 you must typically surround it in single quotes.
11749 The expression @code{@{@var{type}@} @var{addr}} means ``the variable of type
11750 @var{type} that appears at address @var{addr}.''
11753 A name starting with @samp{$} is a convenience variable
11754 (@pxref{Convenience Vars}) or a machine register (@pxref{Registers}).
11757 In addition, @value{GDBN} provides a few other shortcuts and outright
11758 additions specific to Ada:
11762 The assignment statement is allowed as an expression, returning
11763 its right-hand operand as its value. Thus, you may enter
11766 (@value{GDBP}) set x := y + 3
11767 (@value{GDBP}) print A(tmp := y + 1)
11771 The semicolon is allowed as an ``operator,'' returning as its value
11772 the value of its right-hand operand.
11773 This allows, for example,
11774 complex conditional breaks:
11777 (@value{GDBP}) break f
11778 (@value{GDBP}) condition 1 (report(i); k += 1; A(k) > 100)
11782 Rather than use catenation and symbolic character names to introduce special
11783 characters into strings, one may instead use a special bracket notation,
11784 which is also used to print strings. A sequence of characters of the form
11785 @samp{["@var{XX}"]} within a string or character literal denotes the
11786 (single) character whose numeric encoding is @var{XX} in hexadecimal. The
11787 sequence of characters @samp{["""]} also denotes a single quotation mark
11788 in strings. For example,
11790 "One line.["0a"]Next line.["0a"]"
11793 contains an ASCII newline character (@code{Ada.Characters.Latin_1.LF})
11797 The subtype used as a prefix for the attributes @t{'Pos}, @t{'Min}, and
11798 @t{'Max} is optional (and is ignored in any case). For example, it is valid
11802 (@value{GDBP}) print 'max(x, y)
11806 When printing arrays, @value{GDBN} uses positional notation when the
11807 array has a lower bound of 1, and uses a modified named notation otherwise.
11808 For example, a one-dimensional array of three integers with a lower bound
11809 of 3 might print as
11816 That is, in contrast to valid Ada, only the first component has a @code{=>}
11820 You may abbreviate attributes in expressions with any unique,
11821 multi-character subsequence of
11822 their names (an exact match gets preference).
11823 For example, you may use @t{a'len}, @t{a'gth}, or @t{a'lh}
11824 in place of @t{a'length}.
11827 @cindex quoting Ada internal identifiers
11828 Since Ada is case-insensitive, the debugger normally maps identifiers you type
11829 to lower case. The GNAT compiler uses upper-case characters for
11830 some of its internal identifiers, which are normally of no interest to users.
11831 For the rare occasions when you actually have to look at them,
11832 enclose them in angle brackets to avoid the lower-case mapping.
11835 (@value{GDBP}) print <JMPBUF_SAVE>[0]
11839 Printing an object of class-wide type or dereferencing an
11840 access-to-class-wide value will display all the components of the object's
11841 specific type (as indicated by its run-time tag). Likewise, component
11842 selection on such a value will operate on the specific type of the
11847 @node Stopping Before Main Program
11848 @subsubsection Stopping at the Very Beginning
11850 @cindex breakpointing Ada elaboration code
11851 It is sometimes necessary to debug the program during elaboration, and
11852 before reaching the main procedure.
11853 As defined in the Ada Reference
11854 Manual, the elaboration code is invoked from a procedure called
11855 @code{adainit}. To run your program up to the beginning of
11856 elaboration, simply use the following two commands:
11857 @code{tbreak adainit} and @code{run}.
11860 @subsubsection Extensions for Ada Tasks
11861 @cindex Ada, tasking
11863 Support for Ada tasks is analogous to that for threads (@pxref{Threads}).
11864 @value{GDBN} provides the following task-related commands:
11869 This command shows a list of current Ada tasks, as in the following example:
11876 (@value{GDBP}) info tasks
11877 ID TID P-ID Pri State Name
11878 1 8088000 0 15 Child Activation Wait main_task
11879 2 80a4000 1 15 Accept Statement b
11880 3 809a800 1 15 Child Activation Wait a
11881 * 4 80ae800 3 15 Runnable c
11886 In this listing, the asterisk before the last task indicates it to be the
11887 task currently being inspected.
11891 Represents @value{GDBN}'s internal task number.
11897 The parent's task ID (@value{GDBN}'s internal task number).
11900 The base priority of the task.
11903 Current state of the task.
11907 The task has been created but has not been activated. It cannot be
11911 The task is not blocked for any reason known to Ada. (It may be waiting
11912 for a mutex, though.) It is conceptually "executing" in normal mode.
11915 The task is terminated, in the sense of ARM 9.3 (5). Any dependents
11916 that were waiting on terminate alternatives have been awakened and have
11917 terminated themselves.
11919 @item Child Activation Wait
11920 The task is waiting for created tasks to complete activation.
11922 @item Accept Statement
11923 The task is waiting on an accept or selective wait statement.
11925 @item Waiting on entry call
11926 The task is waiting on an entry call.
11928 @item Async Select Wait
11929 The task is waiting to start the abortable part of an asynchronous
11933 The task is waiting on a select statement with only a delay
11936 @item Child Termination Wait
11937 The task is sleeping having completed a master within itself, and is
11938 waiting for the tasks dependent on that master to become terminated or
11939 waiting on a terminate Phase.
11941 @item Wait Child in Term Alt
11942 The task is sleeping waiting for tasks on terminate alternatives to
11943 finish terminating.
11945 @item Accepting RV with @var{taskno}
11946 The task is accepting a rendez-vous with the task @var{taskno}.
11950 Name of the task in the program.
11954 @kindex info task @var{taskno}
11955 @item info task @var{taskno}
11956 This command shows detailled informations on the specified task, as in
11957 the following example:
11962 (@value{GDBP}) info tasks
11963 ID TID P-ID Pri State Name
11964 1 8077880 0 15 Child Activation Wait main_task
11965 * 2 807c468 1 15 Runnable task_1
11966 (@value{GDBP}) info task 2
11967 Ada Task: 0x807c468
11970 Parent: 1 (main_task)
11976 @kindex task@r{ (Ada)}
11977 @cindex current Ada task ID
11978 This command prints the ID of the current task.
11984 (@value{GDBP}) info tasks
11985 ID TID P-ID Pri State Name
11986 1 8077870 0 15 Child Activation Wait main_task
11987 * 2 807c458 1 15 Runnable t
11988 (@value{GDBP}) task
11989 [Current task is 2]
11992 @item task @var{taskno}
11993 @cindex Ada task switching
11994 This command is like the @code{thread @var{threadno}}
11995 command (@pxref{Threads}). It switches the context of debugging
11996 from the current task to the given task.
12002 (@value{GDBP}) info tasks
12003 ID TID P-ID Pri State Name
12004 1 8077870 0 15 Child Activation Wait main_task
12005 * 2 807c458 1 15 Runnable t
12006 (@value{GDBP}) task 1
12007 [Switching to task 1]
12008 #0 0x8067726 in pthread_cond_wait ()
12010 #0 0x8067726 in pthread_cond_wait ()
12011 #1 0x8056714 in system.os_interface.pthread_cond_wait ()
12012 #2 0x805cb63 in system.task_primitives.operations.sleep ()
12013 #3 0x806153e in system.tasking.stages.activate_tasks ()
12014 #4 0x804aacc in un () at un.adb:5
12017 @item break @var{linespec} task @var{taskno}
12018 @itemx break @var{linespec} task @var{taskno} if @dots{}
12019 @cindex breakpoints and tasks, in Ada
12020 @cindex task breakpoints, in Ada
12021 @kindex break @dots{} task @var{taskno}@r{ (Ada)}
12022 These commands are like the @code{break @dots{} thread @dots{}}
12023 command (@pxref{Thread Stops}).
12024 @var{linespec} specifies source lines, as described
12025 in @ref{Specify Location}.
12027 Use the qualifier @samp{task @var{taskno}} with a breakpoint command
12028 to specify that you only want @value{GDBN} to stop the program when a
12029 particular Ada task reaches this breakpoint. @var{taskno} is one of the
12030 numeric task identifiers assigned by @value{GDBN}, shown in the first
12031 column of the @samp{info tasks} display.
12033 If you do not specify @samp{task @var{taskno}} when you set a
12034 breakpoint, the breakpoint applies to @emph{all} tasks of your
12037 You can use the @code{task} qualifier on conditional breakpoints as
12038 well; in this case, place @samp{task @var{taskno}} before the
12039 breakpoint condition (before the @code{if}).
12047 (@value{GDBP}) info tasks
12048 ID TID P-ID Pri State Name
12049 1 140022020 0 15 Child Activation Wait main_task
12050 2 140045060 1 15 Accept/Select Wait t2
12051 3 140044840 1 15 Runnable t1
12052 * 4 140056040 1 15 Runnable t3
12053 (@value{GDBP}) b 15 task 2
12054 Breakpoint 5 at 0x120044cb0: file test_task_debug.adb, line 15.
12055 (@value{GDBP}) cont
12060 Breakpoint 5, test_task_debug () at test_task_debug.adb:15
12062 (@value{GDBP}) info tasks
12063 ID TID P-ID Pri State Name
12064 1 140022020 0 15 Child Activation Wait main_task
12065 * 2 140045060 1 15 Runnable t2
12066 3 140044840 1 15 Runnable t1
12067 4 140056040 1 15 Delay Sleep t3
12071 @node Ada Tasks and Core Files
12072 @subsubsection Tasking Support when Debugging Core Files
12073 @cindex Ada tasking and core file debugging
12075 When inspecting a core file, as opposed to debugging a live program,
12076 tasking support may be limited or even unavailable, depending on
12077 the platform being used.
12078 For instance, on x86-linux, the list of tasks is available, but task
12079 switching is not supported. On Tru64, however, task switching will work
12082 On certain platforms, including Tru64, the debugger needs to perform some
12083 memory writes in order to provide Ada tasking support. When inspecting
12084 a core file, this means that the core file must be opened with read-write
12085 privileges, using the command @samp{"set write on"} (@pxref{Patching}).
12086 Under these circumstances, you should make a backup copy of the core
12087 file before inspecting it with @value{GDBN}.
12090 @subsubsection Known Peculiarities of Ada Mode
12091 @cindex Ada, problems
12093 Besides the omissions listed previously (@pxref{Omissions from Ada}),
12094 we know of several problems with and limitations of Ada mode in
12096 some of which will be fixed with planned future releases of the debugger
12097 and the GNU Ada compiler.
12101 Currently, the debugger
12102 has insufficient information to determine whether certain pointers represent
12103 pointers to objects or the objects themselves.
12104 Thus, the user may have to tack an extra @code{.all} after an expression
12105 to get it printed properly.
12108 Static constants that the compiler chooses not to materialize as objects in
12109 storage are invisible to the debugger.
12112 Named parameter associations in function argument lists are ignored (the
12113 argument lists are treated as positional).
12116 Many useful library packages are currently invisible to the debugger.
12119 Fixed-point arithmetic, conversions, input, and output is carried out using
12120 floating-point arithmetic, and may give results that only approximate those on
12124 The GNAT compiler never generates the prefix @code{Standard} for any of
12125 the standard symbols defined by the Ada language. @value{GDBN} knows about
12126 this: it will strip the prefix from names when you use it, and will never
12127 look for a name you have so qualified among local symbols, nor match against
12128 symbols in other packages or subprograms. If you have
12129 defined entities anywhere in your program other than parameters and
12130 local variables whose simple names match names in @code{Standard},
12131 GNAT's lack of qualification here can cause confusion. When this happens,
12132 you can usually resolve the confusion
12133 by qualifying the problematic names with package
12134 @code{Standard} explicitly.
12137 @node Unsupported Languages
12138 @section Unsupported Languages
12140 @cindex unsupported languages
12141 @cindex minimal language
12142 In addition to the other fully-supported programming languages,
12143 @value{GDBN} also provides a pseudo-language, called @code{minimal}.
12144 It does not represent a real programming language, but provides a set
12145 of capabilities close to what the C or assembly languages provide.
12146 This should allow most simple operations to be performed while debugging
12147 an application that uses a language currently not supported by @value{GDBN}.
12149 If the language is set to @code{auto}, @value{GDBN} will automatically
12150 select this language if the current frame corresponds to an unsupported
12154 @chapter Examining the Symbol Table
12156 The commands described in this chapter allow you to inquire about the
12157 symbols (names of variables, functions and types) defined in your
12158 program. This information is inherent in the text of your program and
12159 does not change as your program executes. @value{GDBN} finds it in your
12160 program's symbol table, in the file indicated when you started @value{GDBN}
12161 (@pxref{File Options, ,Choosing Files}), or by one of the
12162 file-management commands (@pxref{Files, ,Commands to Specify Files}).
12164 @cindex symbol names
12165 @cindex names of symbols
12166 @cindex quoting names
12167 Occasionally, you may need to refer to symbols that contain unusual
12168 characters, which @value{GDBN} ordinarily treats as word delimiters. The
12169 most frequent case is in referring to static variables in other
12170 source files (@pxref{Variables,,Program Variables}). File names
12171 are recorded in object files as debugging symbols, but @value{GDBN} would
12172 ordinarily parse a typical file name, like @file{foo.c}, as the three words
12173 @samp{foo} @samp{.} @samp{c}. To allow @value{GDBN} to recognize
12174 @samp{foo.c} as a single symbol, enclose it in single quotes; for example,
12181 looks up the value of @code{x} in the scope of the file @file{foo.c}.
12184 @cindex case-insensitive symbol names
12185 @cindex case sensitivity in symbol names
12186 @kindex set case-sensitive
12187 @item set case-sensitive on
12188 @itemx set case-sensitive off
12189 @itemx set case-sensitive auto
12190 Normally, when @value{GDBN} looks up symbols, it matches their names
12191 with case sensitivity determined by the current source language.
12192 Occasionally, you may wish to control that. The command @code{set
12193 case-sensitive} lets you do that by specifying @code{on} for
12194 case-sensitive matches or @code{off} for case-insensitive ones. If
12195 you specify @code{auto}, case sensitivity is reset to the default
12196 suitable for the source language. The default is case-sensitive
12197 matches for all languages except for Fortran, for which the default is
12198 case-insensitive matches.
12200 @kindex show case-sensitive
12201 @item show case-sensitive
12202 This command shows the current setting of case sensitivity for symbols
12205 @kindex info address
12206 @cindex address of a symbol
12207 @item info address @var{symbol}
12208 Describe where the data for @var{symbol} is stored. For a register
12209 variable, this says which register it is kept in. For a non-register
12210 local variable, this prints the stack-frame offset at which the variable
12213 Note the contrast with @samp{print &@var{symbol}}, which does not work
12214 at all for a register variable, and for a stack local variable prints
12215 the exact address of the current instantiation of the variable.
12217 @kindex info symbol
12218 @cindex symbol from address
12219 @cindex closest symbol and offset for an address
12220 @item info symbol @var{addr}
12221 Print the name of a symbol which is stored at the address @var{addr}.
12222 If no symbol is stored exactly at @var{addr}, @value{GDBN} prints the
12223 nearest symbol and an offset from it:
12226 (@value{GDBP}) info symbol 0x54320
12227 _initialize_vx + 396 in section .text
12231 This is the opposite of the @code{info address} command. You can use
12232 it to find out the name of a variable or a function given its address.
12234 For dynamically linked executables, the name of executable or shared
12235 library containing the symbol is also printed:
12238 (@value{GDBP}) info symbol 0x400225
12239 _start + 5 in section .text of /tmp/a.out
12240 (@value{GDBP}) info symbol 0x2aaaac2811cf
12241 __read_nocancel + 6 in section .text of /usr/lib64/libc.so.6
12245 @item whatis [@var{arg}]
12246 Print the data type of @var{arg}, which can be either an expression or
12247 a data type. With no argument, print the data type of @code{$}, the
12248 last value in the value history. If @var{arg} is an expression, it is
12249 not actually evaluated, and any side-effecting operations (such as
12250 assignments or function calls) inside it do not take place. If
12251 @var{arg} is a type name, it may be the name of a type or typedef, or
12252 for C code it may have the form @samp{class @var{class-name}},
12253 @samp{struct @var{struct-tag}}, @samp{union @var{union-tag}} or
12254 @samp{enum @var{enum-tag}}.
12255 @xref{Expressions, ,Expressions}.
12258 @item ptype [@var{arg}]
12259 @code{ptype} accepts the same arguments as @code{whatis}, but prints a
12260 detailed description of the type, instead of just the name of the type.
12261 @xref{Expressions, ,Expressions}.
12263 For example, for this variable declaration:
12266 struct complex @{double real; double imag;@} v;
12270 the two commands give this output:
12274 (@value{GDBP}) whatis v
12275 type = struct complex
12276 (@value{GDBP}) ptype v
12277 type = struct complex @{
12285 As with @code{whatis}, using @code{ptype} without an argument refers to
12286 the type of @code{$}, the last value in the value history.
12288 @cindex incomplete type
12289 Sometimes, programs use opaque data types or incomplete specifications
12290 of complex data structure. If the debug information included in the
12291 program does not allow @value{GDBN} to display a full declaration of
12292 the data type, it will say @samp{<incomplete type>}. For example,
12293 given these declarations:
12297 struct foo *fooptr;
12301 but no definition for @code{struct foo} itself, @value{GDBN} will say:
12304 (@value{GDBP}) ptype foo
12305 $1 = <incomplete type>
12309 ``Incomplete type'' is C terminology for data types that are not
12310 completely specified.
12313 @item info types @var{regexp}
12315 Print a brief description of all types whose names match the regular
12316 expression @var{regexp} (or all types in your program, if you supply
12317 no argument). Each complete typename is matched as though it were a
12318 complete line; thus, @samp{i type value} gives information on all
12319 types in your program whose names include the string @code{value}, but
12320 @samp{i type ^value$} gives information only on types whose complete
12321 name is @code{value}.
12323 This command differs from @code{ptype} in two ways: first, like
12324 @code{whatis}, it does not print a detailed description; second, it
12325 lists all source files where a type is defined.
12328 @cindex local variables
12329 @item info scope @var{location}
12330 List all the variables local to a particular scope. This command
12331 accepts a @var{location} argument---a function name, a source line, or
12332 an address preceded by a @samp{*}, and prints all the variables local
12333 to the scope defined by that location. (@xref{Specify Location}, for
12334 details about supported forms of @var{location}.) For example:
12337 (@value{GDBP}) @b{info scope command_line_handler}
12338 Scope for command_line_handler:
12339 Symbol rl is an argument at stack/frame offset 8, length 4.
12340 Symbol linebuffer is in static storage at address 0x150a18, length 4.
12341 Symbol linelength is in static storage at address 0x150a1c, length 4.
12342 Symbol p is a local variable in register $esi, length 4.
12343 Symbol p1 is a local variable in register $ebx, length 4.
12344 Symbol nline is a local variable in register $edx, length 4.
12345 Symbol repeat is a local variable at frame offset -8, length 4.
12349 This command is especially useful for determining what data to collect
12350 during a @dfn{trace experiment}, see @ref{Tracepoint Actions,
12353 @kindex info source
12355 Show information about the current source file---that is, the source file for
12356 the function containing the current point of execution:
12359 the name of the source file, and the directory containing it,
12361 the directory it was compiled in,
12363 its length, in lines,
12365 which programming language it is written in,
12367 whether the executable includes debugging information for that file, and
12368 if so, what format the information is in (e.g., STABS, Dwarf 2, etc.), and
12370 whether the debugging information includes information about
12371 preprocessor macros.
12375 @kindex info sources
12377 Print the names of all source files in your program for which there is
12378 debugging information, organized into two lists: files whose symbols
12379 have already been read, and files whose symbols will be read when needed.
12381 @kindex info functions
12382 @item info functions
12383 Print the names and data types of all defined functions.
12385 @item info functions @var{regexp}
12386 Print the names and data types of all defined functions
12387 whose names contain a match for regular expression @var{regexp}.
12388 Thus, @samp{info fun step} finds all functions whose names
12389 include @code{step}; @samp{info fun ^step} finds those whose names
12390 start with @code{step}. If a function name contains characters
12391 that conflict with the regular expression language (e.g.@:
12392 @samp{operator*()}), they may be quoted with a backslash.
12394 @kindex info variables
12395 @item info variables
12396 Print the names and data types of all variables that are declared
12397 outside of functions (i.e.@: excluding local variables).
12399 @item info variables @var{regexp}
12400 Print the names and data types of all variables (except for local
12401 variables) whose names contain a match for regular expression
12404 @kindex info classes
12405 @cindex Objective-C, classes and selectors
12407 @itemx info classes @var{regexp}
12408 Display all Objective-C classes in your program, or
12409 (with the @var{regexp} argument) all those matching a particular regular
12412 @kindex info selectors
12413 @item info selectors
12414 @itemx info selectors @var{regexp}
12415 Display all Objective-C selectors in your program, or
12416 (with the @var{regexp} argument) all those matching a particular regular
12420 This was never implemented.
12421 @kindex info methods
12423 @itemx info methods @var{regexp}
12424 The @code{info methods} command permits the user to examine all defined
12425 methods within C@t{++} program, or (with the @var{regexp} argument) a
12426 specific set of methods found in the various C@t{++} classes. Many
12427 C@t{++} classes provide a large number of methods. Thus, the output
12428 from the @code{ptype} command can be overwhelming and hard to use. The
12429 @code{info-methods} command filters the methods, printing only those
12430 which match the regular-expression @var{regexp}.
12433 @cindex reloading symbols
12434 Some systems allow individual object files that make up your program to
12435 be replaced without stopping and restarting your program. For example,
12436 in VxWorks you can simply recompile a defective object file and keep on
12437 running. If you are running on one of these systems, you can allow
12438 @value{GDBN} to reload the symbols for automatically relinked modules:
12441 @kindex set symbol-reloading
12442 @item set symbol-reloading on
12443 Replace symbol definitions for the corresponding source file when an
12444 object file with a particular name is seen again.
12446 @item set symbol-reloading off
12447 Do not replace symbol definitions when encountering object files of the
12448 same name more than once. This is the default state; if you are not
12449 running on a system that permits automatic relinking of modules, you
12450 should leave @code{symbol-reloading} off, since otherwise @value{GDBN}
12451 may discard symbols when linking large programs, that may contain
12452 several modules (from different directories or libraries) with the same
12455 @kindex show symbol-reloading
12456 @item show symbol-reloading
12457 Show the current @code{on} or @code{off} setting.
12460 @cindex opaque data types
12461 @kindex set opaque-type-resolution
12462 @item set opaque-type-resolution on
12463 Tell @value{GDBN} to resolve opaque types. An opaque type is a type
12464 declared as a pointer to a @code{struct}, @code{class}, or
12465 @code{union}---for example, @code{struct MyType *}---that is used in one
12466 source file although the full declaration of @code{struct MyType} is in
12467 another source file. The default is on.
12469 A change in the setting of this subcommand will not take effect until
12470 the next time symbols for a file are loaded.
12472 @item set opaque-type-resolution off
12473 Tell @value{GDBN} not to resolve opaque types. In this case, the type
12474 is printed as follows:
12476 @{<no data fields>@}
12479 @kindex show opaque-type-resolution
12480 @item show opaque-type-resolution
12481 Show whether opaque types are resolved or not.
12483 @kindex set print symbol-loading
12484 @cindex print messages when symbols are loaded
12485 @item set print symbol-loading
12486 @itemx set print symbol-loading on
12487 @itemx set print symbol-loading off
12488 The @code{set print symbol-loading} command allows you to enable or
12489 disable printing of messages when @value{GDBN} loads symbols.
12490 By default, these messages will be printed, and normally this is what
12491 you want. Disabling these messages is useful when debugging applications
12492 with lots of shared libraries where the quantity of output can be more
12493 annoying than useful.
12495 @kindex show print symbol-loading
12496 @item show print symbol-loading
12497 Show whether messages will be printed when @value{GDBN} loads symbols.
12499 @kindex maint print symbols
12500 @cindex symbol dump
12501 @kindex maint print psymbols
12502 @cindex partial symbol dump
12503 @item maint print symbols @var{filename}
12504 @itemx maint print psymbols @var{filename}
12505 @itemx maint print msymbols @var{filename}
12506 Write a dump of debugging symbol data into the file @var{filename}.
12507 These commands are used to debug the @value{GDBN} symbol-reading code. Only
12508 symbols with debugging data are included. If you use @samp{maint print
12509 symbols}, @value{GDBN} includes all the symbols for which it has already
12510 collected full details: that is, @var{filename} reflects symbols for
12511 only those files whose symbols @value{GDBN} has read. You can use the
12512 command @code{info sources} to find out which files these are. If you
12513 use @samp{maint print psymbols} instead, the dump shows information about
12514 symbols that @value{GDBN} only knows partially---that is, symbols defined in
12515 files that @value{GDBN} has skimmed, but not yet read completely. Finally,
12516 @samp{maint print msymbols} dumps just the minimal symbol information
12517 required for each object file from which @value{GDBN} has read some symbols.
12518 @xref{Files, ,Commands to Specify Files}, for a discussion of how
12519 @value{GDBN} reads symbols (in the description of @code{symbol-file}).
12521 @kindex maint info symtabs
12522 @kindex maint info psymtabs
12523 @cindex listing @value{GDBN}'s internal symbol tables
12524 @cindex symbol tables, listing @value{GDBN}'s internal
12525 @cindex full symbol tables, listing @value{GDBN}'s internal
12526 @cindex partial symbol tables, listing @value{GDBN}'s internal
12527 @item maint info symtabs @r{[} @var{regexp} @r{]}
12528 @itemx maint info psymtabs @r{[} @var{regexp} @r{]}
12530 List the @code{struct symtab} or @code{struct partial_symtab}
12531 structures whose names match @var{regexp}. If @var{regexp} is not
12532 given, list them all. The output includes expressions which you can
12533 copy into a @value{GDBN} debugging this one to examine a particular
12534 structure in more detail. For example:
12537 (@value{GDBP}) maint info psymtabs dwarf2read
12538 @{ objfile /home/gnu/build/gdb/gdb
12539 ((struct objfile *) 0x82e69d0)
12540 @{ psymtab /home/gnu/src/gdb/dwarf2read.c
12541 ((struct partial_symtab *) 0x8474b10)
12544 text addresses 0x814d3c8 -- 0x8158074
12545 globals (* (struct partial_symbol **) 0x8507a08 @@ 9)
12546 statics (* (struct partial_symbol **) 0x40e95b78 @@ 2882)
12547 dependencies (none)
12550 (@value{GDBP}) maint info symtabs
12554 We see that there is one partial symbol table whose filename contains
12555 the string @samp{dwarf2read}, belonging to the @samp{gdb} executable;
12556 and we see that @value{GDBN} has not read in any symtabs yet at all.
12557 If we set a breakpoint on a function, that will cause @value{GDBN} to
12558 read the symtab for the compilation unit containing that function:
12561 (@value{GDBP}) break dwarf2_psymtab_to_symtab
12562 Breakpoint 1 at 0x814e5da: file /home/gnu/src/gdb/dwarf2read.c,
12564 (@value{GDBP}) maint info symtabs
12565 @{ objfile /home/gnu/build/gdb/gdb
12566 ((struct objfile *) 0x82e69d0)
12567 @{ symtab /home/gnu/src/gdb/dwarf2read.c
12568 ((struct symtab *) 0x86c1f38)
12571 blockvector ((struct blockvector *) 0x86c1bd0) (primary)
12572 linetable ((struct linetable *) 0x8370fa0)
12573 debugformat DWARF 2
12582 @chapter Altering Execution
12584 Once you think you have found an error in your program, you might want to
12585 find out for certain whether correcting the apparent error would lead to
12586 correct results in the rest of the run. You can find the answer by
12587 experiment, using the @value{GDBN} features for altering execution of the
12590 For example, you can store new values into variables or memory
12591 locations, give your program a signal, restart it at a different
12592 address, or even return prematurely from a function.
12595 * Assignment:: Assignment to variables
12596 * Jumping:: Continuing at a different address
12597 * Signaling:: Giving your program a signal
12598 * Returning:: Returning from a function
12599 * Calling:: Calling your program's functions
12600 * Patching:: Patching your program
12604 @section Assignment to Variables
12607 @cindex setting variables
12608 To alter the value of a variable, evaluate an assignment expression.
12609 @xref{Expressions, ,Expressions}. For example,
12616 stores the value 4 into the variable @code{x}, and then prints the
12617 value of the assignment expression (which is 4).
12618 @xref{Languages, ,Using @value{GDBN} with Different Languages}, for more
12619 information on operators in supported languages.
12621 @kindex set variable
12622 @cindex variables, setting
12623 If you are not interested in seeing the value of the assignment, use the
12624 @code{set} command instead of the @code{print} command. @code{set} is
12625 really the same as @code{print} except that the expression's value is
12626 not printed and is not put in the value history (@pxref{Value History,
12627 ,Value History}). The expression is evaluated only for its effects.
12629 If the beginning of the argument string of the @code{set} command
12630 appears identical to a @code{set} subcommand, use the @code{set
12631 variable} command instead of just @code{set}. This command is identical
12632 to @code{set} except for its lack of subcommands. For example, if your
12633 program has a variable @code{width}, you get an error if you try to set
12634 a new value with just @samp{set width=13}, because @value{GDBN} has the
12635 command @code{set width}:
12638 (@value{GDBP}) whatis width
12640 (@value{GDBP}) p width
12642 (@value{GDBP}) set width=47
12643 Invalid syntax in expression.
12647 The invalid expression, of course, is @samp{=47}. In
12648 order to actually set the program's variable @code{width}, use
12651 (@value{GDBP}) set var width=47
12654 Because the @code{set} command has many subcommands that can conflict
12655 with the names of program variables, it is a good idea to use the
12656 @code{set variable} command instead of just @code{set}. For example, if
12657 your program has a variable @code{g}, you run into problems if you try
12658 to set a new value with just @samp{set g=4}, because @value{GDBN} has
12659 the command @code{set gnutarget}, abbreviated @code{set g}:
12663 (@value{GDBP}) whatis g
12667 (@value{GDBP}) set g=4
12671 The program being debugged has been started already.
12672 Start it from the beginning? (y or n) y
12673 Starting program: /home/smith/cc_progs/a.out
12674 "/home/smith/cc_progs/a.out": can't open to read symbols:
12675 Invalid bfd target.
12676 (@value{GDBP}) show g
12677 The current BFD target is "=4".
12682 The program variable @code{g} did not change, and you silently set the
12683 @code{gnutarget} to an invalid value. In order to set the variable
12687 (@value{GDBP}) set var g=4
12690 @value{GDBN} allows more implicit conversions in assignments than C; you can
12691 freely store an integer value into a pointer variable or vice versa,
12692 and you can convert any structure to any other structure that is the
12693 same length or shorter.
12694 @comment FIXME: how do structs align/pad in these conversions?
12695 @comment /doc@cygnus.com 18dec1990
12697 To store values into arbitrary places in memory, use the @samp{@{@dots{}@}}
12698 construct to generate a value of specified type at a specified address
12699 (@pxref{Expressions, ,Expressions}). For example, @code{@{int@}0x83040} refers
12700 to memory location @code{0x83040} as an integer (which implies a certain size
12701 and representation in memory), and
12704 set @{int@}0x83040 = 4
12708 stores the value 4 into that memory location.
12711 @section Continuing at a Different Address
12713 Ordinarily, when you continue your program, you do so at the place where
12714 it stopped, with the @code{continue} command. You can instead continue at
12715 an address of your own choosing, with the following commands:
12719 @item jump @var{linespec}
12720 @itemx jump @var{location}
12721 Resume execution at line @var{linespec} or at address given by
12722 @var{location}. Execution stops again immediately if there is a
12723 breakpoint there. @xref{Specify Location}, for a description of the
12724 different forms of @var{linespec} and @var{location}. It is common
12725 practice to use the @code{tbreak} command in conjunction with
12726 @code{jump}. @xref{Set Breaks, ,Setting Breakpoints}.
12728 The @code{jump} command does not change the current stack frame, or
12729 the stack pointer, or the contents of any memory location or any
12730 register other than the program counter. If line @var{linespec} is in
12731 a different function from the one currently executing, the results may
12732 be bizarre if the two functions expect different patterns of arguments or
12733 of local variables. For this reason, the @code{jump} command requests
12734 confirmation if the specified line is not in the function currently
12735 executing. However, even bizarre results are predictable if you are
12736 well acquainted with the machine-language code of your program.
12739 @c Doesn't work on HP-UX; have to set $pcoqh and $pcoqt.
12740 On many systems, you can get much the same effect as the @code{jump}
12741 command by storing a new value into the register @code{$pc}. The
12742 difference is that this does not start your program running; it only
12743 changes the address of where it @emph{will} run when you continue. For
12751 makes the next @code{continue} command or stepping command execute at
12752 address @code{0x485}, rather than at the address where your program stopped.
12753 @xref{Continuing and Stepping, ,Continuing and Stepping}.
12755 The most common occasion to use the @code{jump} command is to back
12756 up---perhaps with more breakpoints set---over a portion of a program
12757 that has already executed, in order to examine its execution in more
12762 @section Giving your Program a Signal
12763 @cindex deliver a signal to a program
12767 @item signal @var{signal}
12768 Resume execution where your program stopped, but immediately give it the
12769 signal @var{signal}. @var{signal} can be the name or the number of a
12770 signal. For example, on many systems @code{signal 2} and @code{signal
12771 SIGINT} are both ways of sending an interrupt signal.
12773 Alternatively, if @var{signal} is zero, continue execution without
12774 giving a signal. This is useful when your program stopped on account of
12775 a signal and would ordinary see the signal when resumed with the
12776 @code{continue} command; @samp{signal 0} causes it to resume without a
12779 @code{signal} does not repeat when you press @key{RET} a second time
12780 after executing the command.
12784 Invoking the @code{signal} command is not the same as invoking the
12785 @code{kill} utility from the shell. Sending a signal with @code{kill}
12786 causes @value{GDBN} to decide what to do with the signal depending on
12787 the signal handling tables (@pxref{Signals}). The @code{signal} command
12788 passes the signal directly to your program.
12792 @section Returning from a Function
12795 @cindex returning from a function
12798 @itemx return @var{expression}
12799 You can cancel execution of a function call with the @code{return}
12800 command. If you give an
12801 @var{expression} argument, its value is used as the function's return
12805 When you use @code{return}, @value{GDBN} discards the selected stack frame
12806 (and all frames within it). You can think of this as making the
12807 discarded frame return prematurely. If you wish to specify a value to
12808 be returned, give that value as the argument to @code{return}.
12810 This pops the selected stack frame (@pxref{Selection, ,Selecting a
12811 Frame}), and any other frames inside of it, leaving its caller as the
12812 innermost remaining frame. That frame becomes selected. The
12813 specified value is stored in the registers used for returning values
12816 The @code{return} command does not resume execution; it leaves the
12817 program stopped in the state that would exist if the function had just
12818 returned. In contrast, the @code{finish} command (@pxref{Continuing
12819 and Stepping, ,Continuing and Stepping}) resumes execution until the
12820 selected stack frame returns naturally.
12822 @value{GDBN} needs to know how the @var{expression} argument should be set for
12823 the inferior. The concrete registers assignment depends on the OS ABI and the
12824 type being returned by the selected stack frame. For example it is common for
12825 OS ABI to return floating point values in FPU registers while integer values in
12826 CPU registers. Still some ABIs return even floating point values in CPU
12827 registers. Larger integer widths (such as @code{long long int}) also have
12828 specific placement rules. @value{GDBN} already knows the OS ABI from its
12829 current target so it needs to find out also the type being returned to make the
12830 assignment into the right register(s).
12832 Normally, the selected stack frame has debug info. @value{GDBN} will always
12833 use the debug info instead of the implicit type of @var{expression} when the
12834 debug info is available. For example, if you type @kbd{return -1}, and the
12835 function in the current stack frame is declared to return a @code{long long
12836 int}, @value{GDBN} transparently converts the implicit @code{int} value of -1
12837 into a @code{long long int}:
12840 Breakpoint 1, func () at gdb.base/return-nodebug.c:29
12842 (@value{GDBP}) return -1
12843 Make func return now? (y or n) y
12844 #0 0x004004f6 in main () at gdb.base/return-nodebug.c:43
12845 43 printf ("result=%lld\n", func ());
12849 However, if the selected stack frame does not have a debug info, e.g., if the
12850 function was compiled without debug info, @value{GDBN} has to find out the type
12851 to return from user. Specifying a different type by mistake may set the value
12852 in different inferior registers than the caller code expects. For example,
12853 typing @kbd{return -1} with its implicit type @code{int} would set only a part
12854 of a @code{long long int} result for a debug info less function (on 32-bit
12855 architectures). Therefore the user is required to specify the return type by
12856 an appropriate cast explicitly:
12859 Breakpoint 2, 0x0040050b in func ()
12860 (@value{GDBP}) return -1
12861 Return value type not available for selected stack frame.
12862 Please use an explicit cast of the value to return.
12863 (@value{GDBP}) return (long long int) -1
12864 Make selected stack frame return now? (y or n) y
12865 #0 0x00400526 in main ()
12870 @section Calling Program Functions
12873 @cindex calling functions
12874 @cindex inferior functions, calling
12875 @item print @var{expr}
12876 Evaluate the expression @var{expr} and display the resulting value.
12877 @var{expr} may include calls to functions in the program being
12881 @item call @var{expr}
12882 Evaluate the expression @var{expr} without displaying @code{void}
12885 You can use this variant of the @code{print} command if you want to
12886 execute a function from your program that does not return anything
12887 (a.k.a.@: @dfn{a void function}), but without cluttering the output
12888 with @code{void} returned values that @value{GDBN} will otherwise
12889 print. If the result is not void, it is printed and saved in the
12893 It is possible for the function you call via the @code{print} or
12894 @code{call} command to generate a signal (e.g., if there's a bug in
12895 the function, or if you passed it incorrect arguments). What happens
12896 in that case is controlled by the @code{set unwindonsignal} command.
12898 Similarly, with a C@t{++} program it is possible for the function you
12899 call via the @code{print} or @code{call} command to generate an
12900 exception that is not handled due to the constraints of the dummy
12901 frame. In this case, any exception that is raised in the frame, but has
12902 an out-of-frame exception handler will not be found. GDB builds a
12903 dummy-frame for the inferior function call, and the unwinder cannot
12904 seek for exception handlers outside of this dummy-frame. What happens
12905 in that case is controlled by the
12906 @code{set unwind-on-terminating-exception} command.
12909 @item set unwindonsignal
12910 @kindex set unwindonsignal
12911 @cindex unwind stack in called functions
12912 @cindex call dummy stack unwinding
12913 Set unwinding of the stack if a signal is received while in a function
12914 that @value{GDBN} called in the program being debugged. If set to on,
12915 @value{GDBN} unwinds the stack it created for the call and restores
12916 the context to what it was before the call. If set to off (the
12917 default), @value{GDBN} stops in the frame where the signal was
12920 @item show unwindonsignal
12921 @kindex show unwindonsignal
12922 Show the current setting of stack unwinding in the functions called by
12925 @item set unwind-on-terminating-exception
12926 @kindex set unwind-on-terminating-exception
12927 @cindex unwind stack in called functions with unhandled exceptions
12928 @cindex call dummy stack unwinding on unhandled exception.
12929 Set unwinding of the stack if a C@t{++} exception is raised, but left
12930 unhandled while in a function that @value{GDBN} called in the program being
12931 debugged. If set to on (the default), @value{GDBN} unwinds the stack
12932 it created for the call and restores the context to what it was before
12933 the call. If set to off, @value{GDBN} the exception is delivered to
12934 the default C@t{++} exception handler and the inferior terminated.
12936 @item show unwind-on-terminating-exception
12937 @kindex show unwind-on-terminating-exception
12938 Show the current setting of stack unwinding in the functions called by
12943 @cindex weak alias functions
12944 Sometimes, a function you wish to call is actually a @dfn{weak alias}
12945 for another function. In such case, @value{GDBN} might not pick up
12946 the type information, including the types of the function arguments,
12947 which causes @value{GDBN} to call the inferior function incorrectly.
12948 As a result, the called function will function erroneously and may
12949 even crash. A solution to that is to use the name of the aliased
12953 @section Patching Programs
12955 @cindex patching binaries
12956 @cindex writing into executables
12957 @cindex writing into corefiles
12959 By default, @value{GDBN} opens the file containing your program's
12960 executable code (or the corefile) read-only. This prevents accidental
12961 alterations to machine code; but it also prevents you from intentionally
12962 patching your program's binary.
12964 If you'd like to be able to patch the binary, you can specify that
12965 explicitly with the @code{set write} command. For example, you might
12966 want to turn on internal debugging flags, or even to make emergency
12972 @itemx set write off
12973 If you specify @samp{set write on}, @value{GDBN} opens executable and
12974 core files for both reading and writing; if you specify @kbd{set write
12975 off} (the default), @value{GDBN} opens them read-only.
12977 If you have already loaded a file, you must load it again (using the
12978 @code{exec-file} or @code{core-file} command) after changing @code{set
12979 write}, for your new setting to take effect.
12983 Display whether executable files and core files are opened for writing
12984 as well as reading.
12988 @chapter @value{GDBN} Files
12990 @value{GDBN} needs to know the file name of the program to be debugged,
12991 both in order to read its symbol table and in order to start your
12992 program. To debug a core dump of a previous run, you must also tell
12993 @value{GDBN} the name of the core dump file.
12996 * Files:: Commands to specify files
12997 * Separate Debug Files:: Debugging information in separate files
12998 * Symbol Errors:: Errors reading symbol files
12999 * Data Files:: GDB data files
13003 @section Commands to Specify Files
13005 @cindex symbol table
13006 @cindex core dump file
13008 You may want to specify executable and core dump file names. The usual
13009 way to do this is at start-up time, using the arguments to
13010 @value{GDBN}'s start-up commands (@pxref{Invocation, , Getting In and
13011 Out of @value{GDBN}}).
13013 Occasionally it is necessary to change to a different file during a
13014 @value{GDBN} session. Or you may run @value{GDBN} and forget to
13015 specify a file you want to use. Or you are debugging a remote target
13016 via @code{gdbserver} (@pxref{Server, file, Using the @code{gdbserver}
13017 Program}). In these situations the @value{GDBN} commands to specify
13018 new files are useful.
13021 @cindex executable file
13023 @item file @var{filename}
13024 Use @var{filename} as the program to be debugged. It is read for its
13025 symbols and for the contents of pure memory. It is also the program
13026 executed when you use the @code{run} command. If you do not specify a
13027 directory and the file is not found in the @value{GDBN} working directory,
13028 @value{GDBN} uses the environment variable @code{PATH} as a list of
13029 directories to search, just as the shell does when looking for a program
13030 to run. You can change the value of this variable, for both @value{GDBN}
13031 and your program, using the @code{path} command.
13033 @cindex unlinked object files
13034 @cindex patching object files
13035 You can load unlinked object @file{.o} files into @value{GDBN} using
13036 the @code{file} command. You will not be able to ``run'' an object
13037 file, but you can disassemble functions and inspect variables. Also,
13038 if the underlying BFD functionality supports it, you could use
13039 @kbd{gdb -write} to patch object files using this technique. Note
13040 that @value{GDBN} can neither interpret nor modify relocations in this
13041 case, so branches and some initialized variables will appear to go to
13042 the wrong place. But this feature is still handy from time to time.
13045 @code{file} with no argument makes @value{GDBN} discard any information it
13046 has on both executable file and the symbol table.
13049 @item exec-file @r{[} @var{filename} @r{]}
13050 Specify that the program to be run (but not the symbol table) is found
13051 in @var{filename}. @value{GDBN} searches the environment variable @code{PATH}
13052 if necessary to locate your program. Omitting @var{filename} means to
13053 discard information on the executable file.
13055 @kindex symbol-file
13056 @item symbol-file @r{[} @var{filename} @r{]}
13057 Read symbol table information from file @var{filename}. @code{PATH} is
13058 searched when necessary. Use the @code{file} command to get both symbol
13059 table and program to run from the same file.
13061 @code{symbol-file} with no argument clears out @value{GDBN} information on your
13062 program's symbol table.
13064 The @code{symbol-file} command causes @value{GDBN} to forget the contents of
13065 some breakpoints and auto-display expressions. This is because they may
13066 contain pointers to the internal data recording symbols and data types,
13067 which are part of the old symbol table data being discarded inside
13070 @code{symbol-file} does not repeat if you press @key{RET} again after
13073 When @value{GDBN} is configured for a particular environment, it
13074 understands debugging information in whatever format is the standard
13075 generated for that environment; you may use either a @sc{gnu} compiler, or
13076 other compilers that adhere to the local conventions.
13077 Best results are usually obtained from @sc{gnu} compilers; for example,
13078 using @code{@value{NGCC}} you can generate debugging information for
13081 For most kinds of object files, with the exception of old SVR3 systems
13082 using COFF, the @code{symbol-file} command does not normally read the
13083 symbol table in full right away. Instead, it scans the symbol table
13084 quickly to find which source files and which symbols are present. The
13085 details are read later, one source file at a time, as they are needed.
13087 The purpose of this two-stage reading strategy is to make @value{GDBN}
13088 start up faster. For the most part, it is invisible except for
13089 occasional pauses while the symbol table details for a particular source
13090 file are being read. (The @code{set verbose} command can turn these
13091 pauses into messages if desired. @xref{Messages/Warnings, ,Optional
13092 Warnings and Messages}.)
13094 We have not implemented the two-stage strategy for COFF yet. When the
13095 symbol table is stored in COFF format, @code{symbol-file} reads the
13096 symbol table data in full right away. Note that ``stabs-in-COFF''
13097 still does the two-stage strategy, since the debug info is actually
13101 @cindex reading symbols immediately
13102 @cindex symbols, reading immediately
13103 @item symbol-file @var{filename} @r{[} -readnow @r{]}
13104 @itemx file @var{filename} @r{[} -readnow @r{]}
13105 You can override the @value{GDBN} two-stage strategy for reading symbol
13106 tables by using the @samp{-readnow} option with any of the commands that
13107 load symbol table information, if you want to be sure @value{GDBN} has the
13108 entire symbol table available.
13110 @c FIXME: for now no mention of directories, since this seems to be in
13111 @c flux. 13mar1992 status is that in theory GDB would look either in
13112 @c current dir or in same dir as myprog; but issues like competing
13113 @c GDB's, or clutter in system dirs, mean that in practice right now
13114 @c only current dir is used. FFish says maybe a special GDB hierarchy
13115 @c (eg rooted in val of env var GDBSYMS) could exist for mappable symbol
13119 @item core-file @r{[}@var{filename}@r{]}
13121 Specify the whereabouts of a core dump file to be used as the ``contents
13122 of memory''. Traditionally, core files contain only some parts of the
13123 address space of the process that generated them; @value{GDBN} can access the
13124 executable file itself for other parts.
13126 @code{core-file} with no argument specifies that no core file is
13129 Note that the core file is ignored when your program is actually running
13130 under @value{GDBN}. So, if you have been running your program and you
13131 wish to debug a core file instead, you must kill the subprocess in which
13132 the program is running. To do this, use the @code{kill} command
13133 (@pxref{Kill Process, ,Killing the Child Process}).
13135 @kindex add-symbol-file
13136 @cindex dynamic linking
13137 @item add-symbol-file @var{filename} @var{address}
13138 @itemx add-symbol-file @var{filename} @var{address} @r{[} -readnow @r{]}
13139 @itemx add-symbol-file @var{filename} @r{-s}@var{section} @var{address} @dots{}
13140 The @code{add-symbol-file} command reads additional symbol table
13141 information from the file @var{filename}. You would use this command
13142 when @var{filename} has been dynamically loaded (by some other means)
13143 into the program that is running. @var{address} should be the memory
13144 address at which the file has been loaded; @value{GDBN} cannot figure
13145 this out for itself. You can additionally specify an arbitrary number
13146 of @samp{@r{-s}@var{section} @var{address}} pairs, to give an explicit
13147 section name and base address for that section. You can specify any
13148 @var{address} as an expression.
13150 The symbol table of the file @var{filename} is added to the symbol table
13151 originally read with the @code{symbol-file} command. You can use the
13152 @code{add-symbol-file} command any number of times; the new symbol data
13153 thus read keeps adding to the old. To discard all old symbol data
13154 instead, use the @code{symbol-file} command without any arguments.
13156 @cindex relocatable object files, reading symbols from
13157 @cindex object files, relocatable, reading symbols from
13158 @cindex reading symbols from relocatable object files
13159 @cindex symbols, reading from relocatable object files
13160 @cindex @file{.o} files, reading symbols from
13161 Although @var{filename} is typically a shared library file, an
13162 executable file, or some other object file which has been fully
13163 relocated for loading into a process, you can also load symbolic
13164 information from relocatable @file{.o} files, as long as:
13168 the file's symbolic information refers only to linker symbols defined in
13169 that file, not to symbols defined by other object files,
13171 every section the file's symbolic information refers to has actually
13172 been loaded into the inferior, as it appears in the file, and
13174 you can determine the address at which every section was loaded, and
13175 provide these to the @code{add-symbol-file} command.
13179 Some embedded operating systems, like Sun Chorus and VxWorks, can load
13180 relocatable files into an already running program; such systems
13181 typically make the requirements above easy to meet. However, it's
13182 important to recognize that many native systems use complex link
13183 procedures (@code{.linkonce} section factoring and C@t{++} constructor table
13184 assembly, for example) that make the requirements difficult to meet. In
13185 general, one cannot assume that using @code{add-symbol-file} to read a
13186 relocatable object file's symbolic information will have the same effect
13187 as linking the relocatable object file into the program in the normal
13190 @code{add-symbol-file} does not repeat if you press @key{RET} after using it.
13192 @kindex add-symbol-file-from-memory
13193 @cindex @code{syscall DSO}
13194 @cindex load symbols from memory
13195 @item add-symbol-file-from-memory @var{address}
13196 Load symbols from the given @var{address} in a dynamically loaded
13197 object file whose image is mapped directly into the inferior's memory.
13198 For example, the Linux kernel maps a @code{syscall DSO} into each
13199 process's address space; this DSO provides kernel-specific code for
13200 some system calls. The argument can be any expression whose
13201 evaluation yields the address of the file's shared object file header.
13202 For this command to work, you must have used @code{symbol-file} or
13203 @code{exec-file} commands in advance.
13205 @kindex add-shared-symbol-files
13207 @item add-shared-symbol-files @var{library-file}
13208 @itemx assf @var{library-file}
13209 The @code{add-shared-symbol-files} command can currently be used only
13210 in the Cygwin build of @value{GDBN} on MS-Windows OS, where it is an
13211 alias for the @code{dll-symbols} command (@pxref{Cygwin Native}).
13212 @value{GDBN} automatically looks for shared libraries, however if
13213 @value{GDBN} does not find yours, you can invoke
13214 @code{add-shared-symbol-files}. It takes one argument: the shared
13215 library's file name. @code{assf} is a shorthand alias for
13216 @code{add-shared-symbol-files}.
13219 @item section @var{section} @var{addr}
13220 The @code{section} command changes the base address of the named
13221 @var{section} of the exec file to @var{addr}. This can be used if the
13222 exec file does not contain section addresses, (such as in the
13223 @code{a.out} format), or when the addresses specified in the file
13224 itself are wrong. Each section must be changed separately. The
13225 @code{info files} command, described below, lists all the sections and
13229 @kindex info target
13232 @code{info files} and @code{info target} are synonymous; both print the
13233 current target (@pxref{Targets, ,Specifying a Debugging Target}),
13234 including the names of the executable and core dump files currently in
13235 use by @value{GDBN}, and the files from which symbols were loaded. The
13236 command @code{help target} lists all possible targets rather than
13239 @kindex maint info sections
13240 @item maint info sections
13241 Another command that can give you extra information about program sections
13242 is @code{maint info sections}. In addition to the section information
13243 displayed by @code{info files}, this command displays the flags and file
13244 offset of each section in the executable and core dump files. In addition,
13245 @code{maint info sections} provides the following command options (which
13246 may be arbitrarily combined):
13250 Display sections for all loaded object files, including shared libraries.
13251 @item @var{sections}
13252 Display info only for named @var{sections}.
13253 @item @var{section-flags}
13254 Display info only for sections for which @var{section-flags} are true.
13255 The section flags that @value{GDBN} currently knows about are:
13258 Section will have space allocated in the process when loaded.
13259 Set for all sections except those containing debug information.
13261 Section will be loaded from the file into the child process memory.
13262 Set for pre-initialized code and data, clear for @code{.bss} sections.
13264 Section needs to be relocated before loading.
13266 Section cannot be modified by the child process.
13268 Section contains executable code only.
13270 Section contains data only (no executable code).
13272 Section will reside in ROM.
13274 Section contains data for constructor/destructor lists.
13276 Section is not empty.
13278 An instruction to the linker to not output the section.
13279 @item COFF_SHARED_LIBRARY
13280 A notification to the linker that the section contains
13281 COFF shared library information.
13283 Section contains common symbols.
13286 @kindex set trust-readonly-sections
13287 @cindex read-only sections
13288 @item set trust-readonly-sections on
13289 Tell @value{GDBN} that readonly sections in your object file
13290 really are read-only (i.e.@: that their contents will not change).
13291 In that case, @value{GDBN} can fetch values from these sections
13292 out of the object file, rather than from the target program.
13293 For some targets (notably embedded ones), this can be a significant
13294 enhancement to debugging performance.
13296 The default is off.
13298 @item set trust-readonly-sections off
13299 Tell @value{GDBN} not to trust readonly sections. This means that
13300 the contents of the section might change while the program is running,
13301 and must therefore be fetched from the target when needed.
13303 @item show trust-readonly-sections
13304 Show the current setting of trusting readonly sections.
13307 All file-specifying commands allow both absolute and relative file names
13308 as arguments. @value{GDBN} always converts the file name to an absolute file
13309 name and remembers it that way.
13311 @cindex shared libraries
13312 @anchor{Shared Libraries}
13313 @value{GDBN} supports @sc{gnu}/Linux, MS-Windows, HP-UX, SunOS, SVr4, Irix,
13314 and IBM RS/6000 AIX shared libraries.
13316 On MS-Windows @value{GDBN} must be linked with the Expat library to support
13317 shared libraries. @xref{Expat}.
13319 @value{GDBN} automatically loads symbol definitions from shared libraries
13320 when you use the @code{run} command, or when you examine a core file.
13321 (Before you issue the @code{run} command, @value{GDBN} does not understand
13322 references to a function in a shared library, however---unless you are
13323 debugging a core file).
13325 On HP-UX, if the program loads a library explicitly, @value{GDBN}
13326 automatically loads the symbols at the time of the @code{shl_load} call.
13328 @c FIXME: some @value{GDBN} release may permit some refs to undef
13329 @c FIXME...symbols---eg in a break cmd---assuming they are from a shared
13330 @c FIXME...lib; check this from time to time when updating manual
13332 There are times, however, when you may wish to not automatically load
13333 symbol definitions from shared libraries, such as when they are
13334 particularly large or there are many of them.
13336 To control the automatic loading of shared library symbols, use the
13340 @kindex set auto-solib-add
13341 @item set auto-solib-add @var{mode}
13342 If @var{mode} is @code{on}, symbols from all shared object libraries
13343 will be loaded automatically when the inferior begins execution, you
13344 attach to an independently started inferior, or when the dynamic linker
13345 informs @value{GDBN} that a new library has been loaded. If @var{mode}
13346 is @code{off}, symbols must be loaded manually, using the
13347 @code{sharedlibrary} command. The default value is @code{on}.
13349 @cindex memory used for symbol tables
13350 If your program uses lots of shared libraries with debug info that
13351 takes large amounts of memory, you can decrease the @value{GDBN}
13352 memory footprint by preventing it from automatically loading the
13353 symbols from shared libraries. To that end, type @kbd{set
13354 auto-solib-add off} before running the inferior, then load each
13355 library whose debug symbols you do need with @kbd{sharedlibrary
13356 @var{regexp}}, where @var{regexp} is a regular expression that matches
13357 the libraries whose symbols you want to be loaded.
13359 @kindex show auto-solib-add
13360 @item show auto-solib-add
13361 Display the current autoloading mode.
13364 @cindex load shared library
13365 To explicitly load shared library symbols, use the @code{sharedlibrary}
13369 @kindex info sharedlibrary
13372 @itemx info sharedlibrary
13373 Print the names of the shared libraries which are currently loaded.
13375 @kindex sharedlibrary
13377 @item sharedlibrary @var{regex}
13378 @itemx share @var{regex}
13379 Load shared object library symbols for files matching a
13380 Unix regular expression.
13381 As with files loaded automatically, it only loads shared libraries
13382 required by your program for a core file or after typing @code{run}. If
13383 @var{regex} is omitted all shared libraries required by your program are
13386 @item nosharedlibrary
13387 @kindex nosharedlibrary
13388 @cindex unload symbols from shared libraries
13389 Unload all shared object library symbols. This discards all symbols
13390 that have been loaded from all shared libraries. Symbols from shared
13391 libraries that were loaded by explicit user requests are not
13395 Sometimes you may wish that @value{GDBN} stops and gives you control
13396 when any of shared library events happen. Use the @code{set
13397 stop-on-solib-events} command for this:
13400 @item set stop-on-solib-events
13401 @kindex set stop-on-solib-events
13402 This command controls whether @value{GDBN} should give you control
13403 when the dynamic linker notifies it about some shared library event.
13404 The most common event of interest is loading or unloading of a new
13407 @item show stop-on-solib-events
13408 @kindex show stop-on-solib-events
13409 Show whether @value{GDBN} stops and gives you control when shared
13410 library events happen.
13413 Shared libraries are also supported in many cross or remote debugging
13414 configurations. @value{GDBN} needs to have access to the target's libraries;
13415 this can be accomplished either by providing copies of the libraries
13416 on the host system, or by asking @value{GDBN} to automatically retrieve the
13417 libraries from the target. If copies of the target libraries are
13418 provided, they need to be the same as the target libraries, although the
13419 copies on the target can be stripped as long as the copies on the host are
13422 @cindex where to look for shared libraries
13423 For remote debugging, you need to tell @value{GDBN} where the target
13424 libraries are, so that it can load the correct copies---otherwise, it
13425 may try to load the host's libraries. @value{GDBN} has two variables
13426 to specify the search directories for target libraries.
13429 @cindex prefix for shared library file names
13430 @cindex system root, alternate
13431 @kindex set solib-absolute-prefix
13432 @kindex set sysroot
13433 @item set sysroot @var{path}
13434 Use @var{path} as the system root for the program being debugged. Any
13435 absolute shared library paths will be prefixed with @var{path}; many
13436 runtime loaders store the absolute paths to the shared library in the
13437 target program's memory. If you use @code{set sysroot} to find shared
13438 libraries, they need to be laid out in the same way that they are on
13439 the target, with e.g.@: a @file{/lib} and @file{/usr/lib} hierarchy
13442 If @var{path} starts with the sequence @file{remote:}, @value{GDBN} will
13443 retrieve the target libraries from the remote system. This is only
13444 supported when using a remote target that supports the @code{remote get}
13445 command (@pxref{File Transfer,,Sending files to a remote system}).
13446 The part of @var{path} following the initial @file{remote:}
13447 (if present) is used as system root prefix on the remote file system.
13448 @footnote{If you want to specify a local system root using a directory
13449 that happens to be named @file{remote:}, you need to use some equivalent
13450 variant of the name like @file{./remote:}.}
13452 The @code{set solib-absolute-prefix} command is an alias for @code{set
13455 @cindex default system root
13456 @cindex @samp{--with-sysroot}
13457 You can set the default system root by using the configure-time
13458 @samp{--with-sysroot} option. If the system root is inside
13459 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
13460 @samp{--exec-prefix}), then the default system root will be updated
13461 automatically if the installed @value{GDBN} is moved to a new
13464 @kindex show sysroot
13466 Display the current shared library prefix.
13468 @kindex set solib-search-path
13469 @item set solib-search-path @var{path}
13470 If this variable is set, @var{path} is a colon-separated list of
13471 directories to search for shared libraries. @samp{solib-search-path}
13472 is used after @samp{sysroot} fails to locate the library, or if the
13473 path to the library is relative instead of absolute. If you want to
13474 use @samp{solib-search-path} instead of @samp{sysroot}, be sure to set
13475 @samp{sysroot} to a nonexistent directory to prevent @value{GDBN} from
13476 finding your host's libraries. @samp{sysroot} is preferred; setting
13477 it to a nonexistent directory may interfere with automatic loading
13478 of shared library symbols.
13480 @kindex show solib-search-path
13481 @item show solib-search-path
13482 Display the current shared library search path.
13486 @node Separate Debug Files
13487 @section Debugging Information in Separate Files
13488 @cindex separate debugging information files
13489 @cindex debugging information in separate files
13490 @cindex @file{.debug} subdirectories
13491 @cindex debugging information directory, global
13492 @cindex global debugging information directory
13493 @cindex build ID, and separate debugging files
13494 @cindex @file{.build-id} directory
13496 @value{GDBN} allows you to put a program's debugging information in a
13497 file separate from the executable itself, in a way that allows
13498 @value{GDBN} to find and load the debugging information automatically.
13499 Since debugging information can be very large---sometimes larger
13500 than the executable code itself---some systems distribute debugging
13501 information for their executables in separate files, which users can
13502 install only when they need to debug a problem.
13504 @value{GDBN} supports two ways of specifying the separate debug info
13509 The executable contains a @dfn{debug link} that specifies the name of
13510 the separate debug info file. The separate debug file's name is
13511 usually @file{@var{executable}.debug}, where @var{executable} is the
13512 name of the corresponding executable file without leading directories
13513 (e.g., @file{ls.debug} for @file{/usr/bin/ls}). In addition, the
13514 debug link specifies a CRC32 checksum for the debug file, which
13515 @value{GDBN} uses to validate that the executable and the debug file
13516 came from the same build.
13519 The executable contains a @dfn{build ID}, a unique bit string that is
13520 also present in the corresponding debug info file. (This is supported
13521 only on some operating systems, notably those which use the ELF format
13522 for binary files and the @sc{gnu} Binutils.) For more details about
13523 this feature, see the description of the @option{--build-id}
13524 command-line option in @ref{Options, , Command Line Options, ld.info,
13525 The GNU Linker}. The debug info file's name is not specified
13526 explicitly by the build ID, but can be computed from the build ID, see
13530 Depending on the way the debug info file is specified, @value{GDBN}
13531 uses two different methods of looking for the debug file:
13535 For the ``debug link'' method, @value{GDBN} looks up the named file in
13536 the directory of the executable file, then in a subdirectory of that
13537 directory named @file{.debug}, and finally under the global debug
13538 directory, in a subdirectory whose name is identical to the leading
13539 directories of the executable's absolute file name.
13542 For the ``build ID'' method, @value{GDBN} looks in the
13543 @file{.build-id} subdirectory of the global debug directory for a file
13544 named @file{@var{nn}/@var{nnnnnnnn}.debug}, where @var{nn} are the
13545 first 2 hex characters of the build ID bit string, and @var{nnnnnnnn}
13546 are the rest of the bit string. (Real build ID strings are 32 or more
13547 hex characters, not 10.)
13550 So, for example, suppose you ask @value{GDBN} to debug
13551 @file{/usr/bin/ls}, which has a debug link that specifies the
13552 file @file{ls.debug}, and a build ID whose value in hex is
13553 @code{abcdef1234}. If the global debug directory is
13554 @file{/usr/lib/debug}, then @value{GDBN} will look for the following
13555 debug information files, in the indicated order:
13559 @file{/usr/lib/debug/.build-id/ab/cdef1234.debug}
13561 @file{/usr/bin/ls.debug}
13563 @file{/usr/bin/.debug/ls.debug}
13565 @file{/usr/lib/debug/usr/bin/ls.debug}.
13568 You can set the global debugging info directory's name, and view the
13569 name @value{GDBN} is currently using.
13573 @kindex set debug-file-directory
13574 @item set debug-file-directory @var{directory}
13575 Set the directory which @value{GDBN} searches for separate debugging
13576 information files to @var{directory}.
13578 @kindex show debug-file-directory
13579 @item show debug-file-directory
13580 Show the directory @value{GDBN} searches for separate debugging
13585 @cindex @code{.gnu_debuglink} sections
13586 @cindex debug link sections
13587 A debug link is a special section of the executable file named
13588 @code{.gnu_debuglink}. The section must contain:
13592 A filename, with any leading directory components removed, followed by
13595 zero to three bytes of padding, as needed to reach the next four-byte
13596 boundary within the section, and
13598 a four-byte CRC checksum, stored in the same endianness used for the
13599 executable file itself. The checksum is computed on the debugging
13600 information file's full contents by the function given below, passing
13601 zero as the @var{crc} argument.
13604 Any executable file format can carry a debug link, as long as it can
13605 contain a section named @code{.gnu_debuglink} with the contents
13608 @cindex @code{.note.gnu.build-id} sections
13609 @cindex build ID sections
13610 The build ID is a special section in the executable file (and in other
13611 ELF binary files that @value{GDBN} may consider). This section is
13612 often named @code{.note.gnu.build-id}, but that name is not mandatory.
13613 It contains unique identification for the built files---the ID remains
13614 the same across multiple builds of the same build tree. The default
13615 algorithm SHA1 produces 160 bits (40 hexadecimal characters) of the
13616 content for the build ID string. The same section with an identical
13617 value is present in the original built binary with symbols, in its
13618 stripped variant, and in the separate debugging information file.
13620 The debugging information file itself should be an ordinary
13621 executable, containing a full set of linker symbols, sections, and
13622 debugging information. The sections of the debugging information file
13623 should have the same names, addresses, and sizes as the original file,
13624 but they need not contain any data---much like a @code{.bss} section
13625 in an ordinary executable.
13627 The @sc{gnu} binary utilities (Binutils) package includes the
13628 @samp{objcopy} utility that can produce
13629 the separated executable / debugging information file pairs using the
13630 following commands:
13633 @kbd{objcopy --only-keep-debug foo foo.debug}
13638 These commands remove the debugging
13639 information from the executable file @file{foo} and place it in the file
13640 @file{foo.debug}. You can use the first, second or both methods to link the
13645 The debug link method needs the following additional command to also leave
13646 behind a debug link in @file{foo}:
13649 @kbd{objcopy --add-gnu-debuglink=foo.debug foo}
13652 Ulrich Drepper's @file{elfutils} package, starting with version 0.53, contains
13653 a version of the @code{strip} command such that the command @kbd{strip foo -f
13654 foo.debug} has the same functionality as the two @code{objcopy} commands and
13655 the @code{ln -s} command above, together.
13658 Build ID gets embedded into the main executable using @code{ld --build-id} or
13659 the @value{NGCC} counterpart @code{gcc -Wl,--build-id}. Build ID support plus
13660 compatibility fixes for debug files separation are present in @sc{gnu} binary
13661 utilities (Binutils) package since version 2.18.
13666 Since there are many different ways to compute CRC's for the debug
13667 link (different polynomials, reversals, byte ordering, etc.), the
13668 simplest way to describe the CRC used in @code{.gnu_debuglink}
13669 sections is to give the complete code for a function that computes it:
13671 @kindex gnu_debuglink_crc32
13674 gnu_debuglink_crc32 (unsigned long crc,
13675 unsigned char *buf, size_t len)
13677 static const unsigned long crc32_table[256] =
13679 0x00000000, 0x77073096, 0xee0e612c, 0x990951ba, 0x076dc419,
13680 0x706af48f, 0xe963a535, 0x9e6495a3, 0x0edb8832, 0x79dcb8a4,
13681 0xe0d5e91e, 0x97d2d988, 0x09b64c2b, 0x7eb17cbd, 0xe7b82d07,
13682 0x90bf1d91, 0x1db71064, 0x6ab020f2, 0xf3b97148, 0x84be41de,
13683 0x1adad47d, 0x6ddde4eb, 0xf4d4b551, 0x83d385c7, 0x136c9856,
13684 0x646ba8c0, 0xfd62f97a, 0x8a65c9ec, 0x14015c4f, 0x63066cd9,
13685 0xfa0f3d63, 0x8d080df5, 0x3b6e20c8, 0x4c69105e, 0xd56041e4,
13686 0xa2677172, 0x3c03e4d1, 0x4b04d447, 0xd20d85fd, 0xa50ab56b,
13687 0x35b5a8fa, 0x42b2986c, 0xdbbbc9d6, 0xacbcf940, 0x32d86ce3,
13688 0x45df5c75, 0xdcd60dcf, 0xabd13d59, 0x26d930ac, 0x51de003a,
13689 0xc8d75180, 0xbfd06116, 0x21b4f4b5, 0x56b3c423, 0xcfba9599,
13690 0xb8bda50f, 0x2802b89e, 0x5f058808, 0xc60cd9b2, 0xb10be924,
13691 0x2f6f7c87, 0x58684c11, 0xc1611dab, 0xb6662d3d, 0x76dc4190,
13692 0x01db7106, 0x98d220bc, 0xefd5102a, 0x71b18589, 0x06b6b51f,
13693 0x9fbfe4a5, 0xe8b8d433, 0x7807c9a2, 0x0f00f934, 0x9609a88e,
13694 0xe10e9818, 0x7f6a0dbb, 0x086d3d2d, 0x91646c97, 0xe6635c01,
13695 0x6b6b51f4, 0x1c6c6162, 0x856530d8, 0xf262004e, 0x6c0695ed,
13696 0x1b01a57b, 0x8208f4c1, 0xf50fc457, 0x65b0d9c6, 0x12b7e950,
13697 0x8bbeb8ea, 0xfcb9887c, 0x62dd1ddf, 0x15da2d49, 0x8cd37cf3,
13698 0xfbd44c65, 0x4db26158, 0x3ab551ce, 0xa3bc0074, 0xd4bb30e2,
13699 0x4adfa541, 0x3dd895d7, 0xa4d1c46d, 0xd3d6f4fb, 0x4369e96a,
13700 0x346ed9fc, 0xad678846, 0xda60b8d0, 0x44042d73, 0x33031de5,
13701 0xaa0a4c5f, 0xdd0d7cc9, 0x5005713c, 0x270241aa, 0xbe0b1010,
13702 0xc90c2086, 0x5768b525, 0x206f85b3, 0xb966d409, 0xce61e49f,
13703 0x5edef90e, 0x29d9c998, 0xb0d09822, 0xc7d7a8b4, 0x59b33d17,
13704 0x2eb40d81, 0xb7bd5c3b, 0xc0ba6cad, 0xedb88320, 0x9abfb3b6,
13705 0x03b6e20c, 0x74b1d29a, 0xead54739, 0x9dd277af, 0x04db2615,
13706 0x73dc1683, 0xe3630b12, 0x94643b84, 0x0d6d6a3e, 0x7a6a5aa8,
13707 0xe40ecf0b, 0x9309ff9d, 0x0a00ae27, 0x7d079eb1, 0xf00f9344,
13708 0x8708a3d2, 0x1e01f268, 0x6906c2fe, 0xf762575d, 0x806567cb,
13709 0x196c3671, 0x6e6b06e7, 0xfed41b76, 0x89d32be0, 0x10da7a5a,
13710 0x67dd4acc, 0xf9b9df6f, 0x8ebeeff9, 0x17b7be43, 0x60b08ed5,
13711 0xd6d6a3e8, 0xa1d1937e, 0x38d8c2c4, 0x4fdff252, 0xd1bb67f1,
13712 0xa6bc5767, 0x3fb506dd, 0x48b2364b, 0xd80d2bda, 0xaf0a1b4c,
13713 0x36034af6, 0x41047a60, 0xdf60efc3, 0xa867df55, 0x316e8eef,
13714 0x4669be79, 0xcb61b38c, 0xbc66831a, 0x256fd2a0, 0x5268e236,
13715 0xcc0c7795, 0xbb0b4703, 0x220216b9, 0x5505262f, 0xc5ba3bbe,
13716 0xb2bd0b28, 0x2bb45a92, 0x5cb36a04, 0xc2d7ffa7, 0xb5d0cf31,
13717 0x2cd99e8b, 0x5bdeae1d, 0x9b64c2b0, 0xec63f226, 0x756aa39c,
13718 0x026d930a, 0x9c0906a9, 0xeb0e363f, 0x72076785, 0x05005713,
13719 0x95bf4a82, 0xe2b87a14, 0x7bb12bae, 0x0cb61b38, 0x92d28e9b,
13720 0xe5d5be0d, 0x7cdcefb7, 0x0bdbdf21, 0x86d3d2d4, 0xf1d4e242,
13721 0x68ddb3f8, 0x1fda836e, 0x81be16cd, 0xf6b9265b, 0x6fb077e1,
13722 0x18b74777, 0x88085ae6, 0xff0f6a70, 0x66063bca, 0x11010b5c,
13723 0x8f659eff, 0xf862ae69, 0x616bffd3, 0x166ccf45, 0xa00ae278,
13724 0xd70dd2ee, 0x4e048354, 0x3903b3c2, 0xa7672661, 0xd06016f7,
13725 0x4969474d, 0x3e6e77db, 0xaed16a4a, 0xd9d65adc, 0x40df0b66,
13726 0x37d83bf0, 0xa9bcae53, 0xdebb9ec5, 0x47b2cf7f, 0x30b5ffe9,
13727 0xbdbdf21c, 0xcabac28a, 0x53b39330, 0x24b4a3a6, 0xbad03605,
13728 0xcdd70693, 0x54de5729, 0x23d967bf, 0xb3667a2e, 0xc4614ab8,
13729 0x5d681b02, 0x2a6f2b94, 0xb40bbe37, 0xc30c8ea1, 0x5a05df1b,
13732 unsigned char *end;
13734 crc = ~crc & 0xffffffff;
13735 for (end = buf + len; buf < end; ++buf)
13736 crc = crc32_table[(crc ^ *buf) & 0xff] ^ (crc >> 8);
13737 return ~crc & 0xffffffff;
13742 This computation does not apply to the ``build ID'' method.
13745 @node Symbol Errors
13746 @section Errors Reading Symbol Files
13748 While reading a symbol file, @value{GDBN} occasionally encounters problems,
13749 such as symbol types it does not recognize, or known bugs in compiler
13750 output. By default, @value{GDBN} does not notify you of such problems, since
13751 they are relatively common and primarily of interest to people
13752 debugging compilers. If you are interested in seeing information
13753 about ill-constructed symbol tables, you can either ask @value{GDBN} to print
13754 only one message about each such type of problem, no matter how many
13755 times the problem occurs; or you can ask @value{GDBN} to print more messages,
13756 to see how many times the problems occur, with the @code{set
13757 complaints} command (@pxref{Messages/Warnings, ,Optional Warnings and
13760 The messages currently printed, and their meanings, include:
13763 @item inner block not inside outer block in @var{symbol}
13765 The symbol information shows where symbol scopes begin and end
13766 (such as at the start of a function or a block of statements). This
13767 error indicates that an inner scope block is not fully contained
13768 in its outer scope blocks.
13770 @value{GDBN} circumvents the problem by treating the inner block as if it had
13771 the same scope as the outer block. In the error message, @var{symbol}
13772 may be shown as ``@code{(don't know)}'' if the outer block is not a
13775 @item block at @var{address} out of order
13777 The symbol information for symbol scope blocks should occur in
13778 order of increasing addresses. This error indicates that it does not
13781 @value{GDBN} does not circumvent this problem, and has trouble
13782 locating symbols in the source file whose symbols it is reading. (You
13783 can often determine what source file is affected by specifying
13784 @code{set verbose on}. @xref{Messages/Warnings, ,Optional Warnings and
13787 @item bad block start address patched
13789 The symbol information for a symbol scope block has a start address
13790 smaller than the address of the preceding source line. This is known
13791 to occur in the SunOS 4.1.1 (and earlier) C compiler.
13793 @value{GDBN} circumvents the problem by treating the symbol scope block as
13794 starting on the previous source line.
13796 @item bad string table offset in symbol @var{n}
13799 Symbol number @var{n} contains a pointer into the string table which is
13800 larger than the size of the string table.
13802 @value{GDBN} circumvents the problem by considering the symbol to have the
13803 name @code{foo}, which may cause other problems if many symbols end up
13806 @item unknown symbol type @code{0x@var{nn}}
13808 The symbol information contains new data types that @value{GDBN} does
13809 not yet know how to read. @code{0x@var{nn}} is the symbol type of the
13810 uncomprehended information, in hexadecimal.
13812 @value{GDBN} circumvents the error by ignoring this symbol information.
13813 This usually allows you to debug your program, though certain symbols
13814 are not accessible. If you encounter such a problem and feel like
13815 debugging it, you can debug @code{@value{GDBP}} with itself, breakpoint
13816 on @code{complain}, then go up to the function @code{read_dbx_symtab}
13817 and examine @code{*bufp} to see the symbol.
13819 @item stub type has NULL name
13821 @value{GDBN} could not find the full definition for a struct or class.
13823 @item const/volatile indicator missing (ok if using g++ v1.x), got@dots{}
13824 The symbol information for a C@t{++} member function is missing some
13825 information that recent versions of the compiler should have output for
13828 @item info mismatch between compiler and debugger
13830 @value{GDBN} could not parse a type specification output by the compiler.
13835 @section GDB Data Files
13837 @cindex prefix for data files
13838 @value{GDBN} will sometimes read an auxiliary data file. These files
13839 are kept in a directory known as the @dfn{data directory}.
13841 You can set the data directory's name, and view the name @value{GDBN}
13842 is currently using.
13845 @kindex set data-directory
13846 @item set data-directory @var{directory}
13847 Set the directory which @value{GDBN} searches for auxiliary data files
13848 to @var{directory}.
13850 @kindex show data-directory
13851 @item show data-directory
13852 Show the directory @value{GDBN} searches for auxiliary data files.
13855 @cindex default data directory
13856 @cindex @samp{--with-gdb-datadir}
13857 You can set the default data directory by using the configure-time
13858 @samp{--with-gdb-datadir} option. If the data directory is inside
13859 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
13860 @samp{--exec-prefix}), then the default data directory will be updated
13861 automatically if the installed @value{GDBN} is moved to a new
13865 @chapter Specifying a Debugging Target
13867 @cindex debugging target
13868 A @dfn{target} is the execution environment occupied by your program.
13870 Often, @value{GDBN} runs in the same host environment as your program;
13871 in that case, the debugging target is specified as a side effect when
13872 you use the @code{file} or @code{core} commands. When you need more
13873 flexibility---for example, running @value{GDBN} on a physically separate
13874 host, or controlling a standalone system over a serial port or a
13875 realtime system over a TCP/IP connection---you can use the @code{target}
13876 command to specify one of the target types configured for @value{GDBN}
13877 (@pxref{Target Commands, ,Commands for Managing Targets}).
13879 @cindex target architecture
13880 It is possible to build @value{GDBN} for several different @dfn{target
13881 architectures}. When @value{GDBN} is built like that, you can choose
13882 one of the available architectures with the @kbd{set architecture}
13886 @kindex set architecture
13887 @kindex show architecture
13888 @item set architecture @var{arch}
13889 This command sets the current target architecture to @var{arch}. The
13890 value of @var{arch} can be @code{"auto"}, in addition to one of the
13891 supported architectures.
13893 @item show architecture
13894 Show the current target architecture.
13896 @item set processor
13898 @kindex set processor
13899 @kindex show processor
13900 These are alias commands for, respectively, @code{set architecture}
13901 and @code{show architecture}.
13905 * Active Targets:: Active targets
13906 * Target Commands:: Commands for managing targets
13907 * Byte Order:: Choosing target byte order
13910 @node Active Targets
13911 @section Active Targets
13913 @cindex stacking targets
13914 @cindex active targets
13915 @cindex multiple targets
13917 There are three classes of targets: processes, core files, and
13918 executable files. @value{GDBN} can work concurrently on up to three
13919 active targets, one in each class. This allows you to (for example)
13920 start a process and inspect its activity without abandoning your work on
13923 For example, if you execute @samp{gdb a.out}, then the executable file
13924 @code{a.out} is the only active target. If you designate a core file as
13925 well---presumably from a prior run that crashed and coredumped---then
13926 @value{GDBN} has two active targets and uses them in tandem, looking
13927 first in the corefile target, then in the executable file, to satisfy
13928 requests for memory addresses. (Typically, these two classes of target
13929 are complementary, since core files contain only a program's
13930 read-write memory---variables and so on---plus machine status, while
13931 executable files contain only the program text and initialized data.)
13933 When you type @code{run}, your executable file becomes an active process
13934 target as well. When a process target is active, all @value{GDBN}
13935 commands requesting memory addresses refer to that target; addresses in
13936 an active core file or executable file target are obscured while the
13937 process target is active.
13939 Use the @code{core-file} and @code{exec-file} commands to select a new
13940 core file or executable target (@pxref{Files, ,Commands to Specify
13941 Files}). To specify as a target a process that is already running, use
13942 the @code{attach} command (@pxref{Attach, ,Debugging an Already-running
13945 @node Target Commands
13946 @section Commands for Managing Targets
13949 @item target @var{type} @var{parameters}
13950 Connects the @value{GDBN} host environment to a target machine or
13951 process. A target is typically a protocol for talking to debugging
13952 facilities. You use the argument @var{type} to specify the type or
13953 protocol of the target machine.
13955 Further @var{parameters} are interpreted by the target protocol, but
13956 typically include things like device names or host names to connect
13957 with, process numbers, and baud rates.
13959 The @code{target} command does not repeat if you press @key{RET} again
13960 after executing the command.
13962 @kindex help target
13964 Displays the names of all targets available. To display targets
13965 currently selected, use either @code{info target} or @code{info files}
13966 (@pxref{Files, ,Commands to Specify Files}).
13968 @item help target @var{name}
13969 Describe a particular target, including any parameters necessary to
13972 @kindex set gnutarget
13973 @item set gnutarget @var{args}
13974 @value{GDBN} uses its own library BFD to read your files. @value{GDBN}
13975 knows whether it is reading an @dfn{executable},
13976 a @dfn{core}, or a @dfn{.o} file; however, you can specify the file format
13977 with the @code{set gnutarget} command. Unlike most @code{target} commands,
13978 with @code{gnutarget} the @code{target} refers to a program, not a machine.
13981 @emph{Warning:} To specify a file format with @code{set gnutarget},
13982 you must know the actual BFD name.
13986 @xref{Files, , Commands to Specify Files}.
13988 @kindex show gnutarget
13989 @item show gnutarget
13990 Use the @code{show gnutarget} command to display what file format
13991 @code{gnutarget} is set to read. If you have not set @code{gnutarget},
13992 @value{GDBN} will determine the file format for each file automatically,
13993 and @code{show gnutarget} displays @samp{The current BDF target is "auto"}.
13996 @cindex common targets
13997 Here are some common targets (available, or not, depending on the GDB
14002 @item target exec @var{program}
14003 @cindex executable file target
14004 An executable file. @samp{target exec @var{program}} is the same as
14005 @samp{exec-file @var{program}}.
14007 @item target core @var{filename}
14008 @cindex core dump file target
14009 A core dump file. @samp{target core @var{filename}} is the same as
14010 @samp{core-file @var{filename}}.
14012 @item target remote @var{medium}
14013 @cindex remote target
14014 A remote system connected to @value{GDBN} via a serial line or network
14015 connection. This command tells @value{GDBN} to use its own remote
14016 protocol over @var{medium} for debugging. @xref{Remote Debugging}.
14018 For example, if you have a board connected to @file{/dev/ttya} on the
14019 machine running @value{GDBN}, you could say:
14022 target remote /dev/ttya
14025 @code{target remote} supports the @code{load} command. This is only
14026 useful if you have some other way of getting the stub to the target
14027 system, and you can put it somewhere in memory where it won't get
14028 clobbered by the download.
14031 @cindex built-in simulator target
14032 Builtin CPU simulator. @value{GDBN} includes simulators for most architectures.
14040 works; however, you cannot assume that a specific memory map, device
14041 drivers, or even basic I/O is available, although some simulators do
14042 provide these. For info about any processor-specific simulator details,
14043 see the appropriate section in @ref{Embedded Processors, ,Embedded
14048 Some configurations may include these targets as well:
14052 @item target nrom @var{dev}
14053 @cindex NetROM ROM emulator target
14054 NetROM ROM emulator. This target only supports downloading.
14058 Different targets are available on different configurations of @value{GDBN};
14059 your configuration may have more or fewer targets.
14061 Many remote targets require you to download the executable's code once
14062 you've successfully established a connection. You may wish to control
14063 various aspects of this process.
14068 @kindex set hash@r{, for remote monitors}
14069 @cindex hash mark while downloading
14070 This command controls whether a hash mark @samp{#} is displayed while
14071 downloading a file to the remote monitor. If on, a hash mark is
14072 displayed after each S-record is successfully downloaded to the
14076 @kindex show hash@r{, for remote monitors}
14077 Show the current status of displaying the hash mark.
14079 @item set debug monitor
14080 @kindex set debug monitor
14081 @cindex display remote monitor communications
14082 Enable or disable display of communications messages between
14083 @value{GDBN} and the remote monitor.
14085 @item show debug monitor
14086 @kindex show debug monitor
14087 Show the current status of displaying communications between
14088 @value{GDBN} and the remote monitor.
14093 @kindex load @var{filename}
14094 @item load @var{filename}
14096 Depending on what remote debugging facilities are configured into
14097 @value{GDBN}, the @code{load} command may be available. Where it exists, it
14098 is meant to make @var{filename} (an executable) available for debugging
14099 on the remote system---by downloading, or dynamic linking, for example.
14100 @code{load} also records the @var{filename} symbol table in @value{GDBN}, like
14101 the @code{add-symbol-file} command.
14103 If your @value{GDBN} does not have a @code{load} command, attempting to
14104 execute it gets the error message ``@code{You can't do that when your
14105 target is @dots{}}''
14107 The file is loaded at whatever address is specified in the executable.
14108 For some object file formats, you can specify the load address when you
14109 link the program; for other formats, like a.out, the object file format
14110 specifies a fixed address.
14111 @c FIXME! This would be a good place for an xref to the GNU linker doc.
14113 Depending on the remote side capabilities, @value{GDBN} may be able to
14114 load programs into flash memory.
14116 @code{load} does not repeat if you press @key{RET} again after using it.
14120 @section Choosing Target Byte Order
14122 @cindex choosing target byte order
14123 @cindex target byte order
14125 Some types of processors, such as the MIPS, PowerPC, and Renesas SH,
14126 offer the ability to run either big-endian or little-endian byte
14127 orders. Usually the executable or symbol will include a bit to
14128 designate the endian-ness, and you will not need to worry about
14129 which to use. However, you may still find it useful to adjust
14130 @value{GDBN}'s idea of processor endian-ness manually.
14134 @item set endian big
14135 Instruct @value{GDBN} to assume the target is big-endian.
14137 @item set endian little
14138 Instruct @value{GDBN} to assume the target is little-endian.
14140 @item set endian auto
14141 Instruct @value{GDBN} to use the byte order associated with the
14145 Display @value{GDBN}'s current idea of the target byte order.
14149 Note that these commands merely adjust interpretation of symbolic
14150 data on the host, and that they have absolutely no effect on the
14154 @node Remote Debugging
14155 @chapter Debugging Remote Programs
14156 @cindex remote debugging
14158 If you are trying to debug a program running on a machine that cannot run
14159 @value{GDBN} in the usual way, it is often useful to use remote debugging.
14160 For example, you might use remote debugging on an operating system kernel,
14161 or on a small system which does not have a general purpose operating system
14162 powerful enough to run a full-featured debugger.
14164 Some configurations of @value{GDBN} have special serial or TCP/IP interfaces
14165 to make this work with particular debugging targets. In addition,
14166 @value{GDBN} comes with a generic serial protocol (specific to @value{GDBN},
14167 but not specific to any particular target system) which you can use if you
14168 write the remote stubs---the code that runs on the remote system to
14169 communicate with @value{GDBN}.
14171 Other remote targets may be available in your
14172 configuration of @value{GDBN}; use @code{help target} to list them.
14175 * Connecting:: Connecting to a remote target
14176 * File Transfer:: Sending files to a remote system
14177 * Server:: Using the gdbserver program
14178 * Remote Configuration:: Remote configuration
14179 * Remote Stub:: Implementing a remote stub
14183 @section Connecting to a Remote Target
14185 On the @value{GDBN} host machine, you will need an unstripped copy of
14186 your program, since @value{GDBN} needs symbol and debugging information.
14187 Start up @value{GDBN} as usual, using the name of the local copy of your
14188 program as the first argument.
14190 @cindex @code{target remote}
14191 @value{GDBN} can communicate with the target over a serial line, or
14192 over an @acronym{IP} network using @acronym{TCP} or @acronym{UDP}. In
14193 each case, @value{GDBN} uses the same protocol for debugging your
14194 program; only the medium carrying the debugging packets varies. The
14195 @code{target remote} command establishes a connection to the target.
14196 Its arguments indicate which medium to use:
14200 @item target remote @var{serial-device}
14201 @cindex serial line, @code{target remote}
14202 Use @var{serial-device} to communicate with the target. For example,
14203 to use a serial line connected to the device named @file{/dev/ttyb}:
14206 target remote /dev/ttyb
14209 If you're using a serial line, you may want to give @value{GDBN} the
14210 @w{@samp{--baud}} option, or use the @code{set remotebaud} command
14211 (@pxref{Remote Configuration, set remotebaud}) before the
14212 @code{target} command.
14214 @item target remote @code{@var{host}:@var{port}}
14215 @itemx target remote @code{tcp:@var{host}:@var{port}}
14216 @cindex @acronym{TCP} port, @code{target remote}
14217 Debug using a @acronym{TCP} connection to @var{port} on @var{host}.
14218 The @var{host} may be either a host name or a numeric @acronym{IP}
14219 address; @var{port} must be a decimal number. The @var{host} could be
14220 the target machine itself, if it is directly connected to the net, or
14221 it might be a terminal server which in turn has a serial line to the
14224 For example, to connect to port 2828 on a terminal server named
14228 target remote manyfarms:2828
14231 If your remote target is actually running on the same machine as your
14232 debugger session (e.g.@: a simulator for your target running on the
14233 same host), you can omit the hostname. For example, to connect to
14234 port 1234 on your local machine:
14237 target remote :1234
14241 Note that the colon is still required here.
14243 @item target remote @code{udp:@var{host}:@var{port}}
14244 @cindex @acronym{UDP} port, @code{target remote}
14245 Debug using @acronym{UDP} packets to @var{port} on @var{host}. For example, to
14246 connect to @acronym{UDP} port 2828 on a terminal server named @code{manyfarms}:
14249 target remote udp:manyfarms:2828
14252 When using a @acronym{UDP} connection for remote debugging, you should
14253 keep in mind that the `U' stands for ``Unreliable''. @acronym{UDP}
14254 can silently drop packets on busy or unreliable networks, which will
14255 cause havoc with your debugging session.
14257 @item target remote | @var{command}
14258 @cindex pipe, @code{target remote} to
14259 Run @var{command} in the background and communicate with it using a
14260 pipe. The @var{command} is a shell command, to be parsed and expanded
14261 by the system's command shell, @code{/bin/sh}; it should expect remote
14262 protocol packets on its standard input, and send replies on its
14263 standard output. You could use this to run a stand-alone simulator
14264 that speaks the remote debugging protocol, to make net connections
14265 using programs like @code{ssh}, or for other similar tricks.
14267 If @var{command} closes its standard output (perhaps by exiting),
14268 @value{GDBN} will try to send it a @code{SIGTERM} signal. (If the
14269 program has already exited, this will have no effect.)
14273 Once the connection has been established, you can use all the usual
14274 commands to examine and change data. The remote program is already
14275 running; you can use @kbd{step} and @kbd{continue}, and you do not
14276 need to use @kbd{run}.
14278 @cindex interrupting remote programs
14279 @cindex remote programs, interrupting
14280 Whenever @value{GDBN} is waiting for the remote program, if you type the
14281 interrupt character (often @kbd{Ctrl-c}), @value{GDBN} attempts to stop the
14282 program. This may or may not succeed, depending in part on the hardware
14283 and the serial drivers the remote system uses. If you type the
14284 interrupt character once again, @value{GDBN} displays this prompt:
14287 Interrupted while waiting for the program.
14288 Give up (and stop debugging it)? (y or n)
14291 If you type @kbd{y}, @value{GDBN} abandons the remote debugging session.
14292 (If you decide you want to try again later, you can use @samp{target
14293 remote} again to connect once more.) If you type @kbd{n}, @value{GDBN}
14294 goes back to waiting.
14297 @kindex detach (remote)
14299 When you have finished debugging the remote program, you can use the
14300 @code{detach} command to release it from @value{GDBN} control.
14301 Detaching from the target normally resumes its execution, but the results
14302 will depend on your particular remote stub. After the @code{detach}
14303 command, @value{GDBN} is free to connect to another target.
14307 The @code{disconnect} command behaves like @code{detach}, except that
14308 the target is generally not resumed. It will wait for @value{GDBN}
14309 (this instance or another one) to connect and continue debugging. After
14310 the @code{disconnect} command, @value{GDBN} is again free to connect to
14313 @cindex send command to remote monitor
14314 @cindex extend @value{GDBN} for remote targets
14315 @cindex add new commands for external monitor
14317 @item monitor @var{cmd}
14318 This command allows you to send arbitrary commands directly to the
14319 remote monitor. Since @value{GDBN} doesn't care about the commands it
14320 sends like this, this command is the way to extend @value{GDBN}---you
14321 can add new commands that only the external monitor will understand
14325 @node File Transfer
14326 @section Sending files to a remote system
14327 @cindex remote target, file transfer
14328 @cindex file transfer
14329 @cindex sending files to remote systems
14331 Some remote targets offer the ability to transfer files over the same
14332 connection used to communicate with @value{GDBN}. This is convenient
14333 for targets accessible through other means, e.g.@: @sc{gnu}/Linux systems
14334 running @code{gdbserver} over a network interface. For other targets,
14335 e.g.@: embedded devices with only a single serial port, this may be
14336 the only way to upload or download files.
14338 Not all remote targets support these commands.
14342 @item remote put @var{hostfile} @var{targetfile}
14343 Copy file @var{hostfile} from the host system (the machine running
14344 @value{GDBN}) to @var{targetfile} on the target system.
14347 @item remote get @var{targetfile} @var{hostfile}
14348 Copy file @var{targetfile} from the target system to @var{hostfile}
14349 on the host system.
14351 @kindex remote delete
14352 @item remote delete @var{targetfile}
14353 Delete @var{targetfile} from the target system.
14358 @section Using the @code{gdbserver} Program
14361 @cindex remote connection without stubs
14362 @code{gdbserver} is a control program for Unix-like systems, which
14363 allows you to connect your program with a remote @value{GDBN} via
14364 @code{target remote}---but without linking in the usual debugging stub.
14366 @code{gdbserver} is not a complete replacement for the debugging stubs,
14367 because it requires essentially the same operating-system facilities
14368 that @value{GDBN} itself does. In fact, a system that can run
14369 @code{gdbserver} to connect to a remote @value{GDBN} could also run
14370 @value{GDBN} locally! @code{gdbserver} is sometimes useful nevertheless,
14371 because it is a much smaller program than @value{GDBN} itself. It is
14372 also easier to port than all of @value{GDBN}, so you may be able to get
14373 started more quickly on a new system by using @code{gdbserver}.
14374 Finally, if you develop code for real-time systems, you may find that
14375 the tradeoffs involved in real-time operation make it more convenient to
14376 do as much development work as possible on another system, for example
14377 by cross-compiling. You can use @code{gdbserver} to make a similar
14378 choice for debugging.
14380 @value{GDBN} and @code{gdbserver} communicate via either a serial line
14381 or a TCP connection, using the standard @value{GDBN} remote serial
14385 @emph{Warning:} @code{gdbserver} does not have any built-in security.
14386 Do not run @code{gdbserver} connected to any public network; a
14387 @value{GDBN} connection to @code{gdbserver} provides access to the
14388 target system with the same privileges as the user running
14392 @subsection Running @code{gdbserver}
14393 @cindex arguments, to @code{gdbserver}
14395 Run @code{gdbserver} on the target system. You need a copy of the
14396 program you want to debug, including any libraries it requires.
14397 @code{gdbserver} does not need your program's symbol table, so you can
14398 strip the program if necessary to save space. @value{GDBN} on the host
14399 system does all the symbol handling.
14401 To use the server, you must tell it how to communicate with @value{GDBN};
14402 the name of your program; and the arguments for your program. The usual
14406 target> gdbserver @var{comm} @var{program} [ @var{args} @dots{} ]
14409 @var{comm} is either a device name (to use a serial line) or a TCP
14410 hostname and portnumber. For example, to debug Emacs with the argument
14411 @samp{foo.txt} and communicate with @value{GDBN} over the serial port
14415 target> gdbserver /dev/com1 emacs foo.txt
14418 @code{gdbserver} waits passively for the host @value{GDBN} to communicate
14421 To use a TCP connection instead of a serial line:
14424 target> gdbserver host:2345 emacs foo.txt
14427 The only difference from the previous example is the first argument,
14428 specifying that you are communicating with the host @value{GDBN} via
14429 TCP. The @samp{host:2345} argument means that @code{gdbserver} is to
14430 expect a TCP connection from machine @samp{host} to local TCP port 2345.
14431 (Currently, the @samp{host} part is ignored.) You can choose any number
14432 you want for the port number as long as it does not conflict with any
14433 TCP ports already in use on the target system (for example, @code{23} is
14434 reserved for @code{telnet}).@footnote{If you choose a port number that
14435 conflicts with another service, @code{gdbserver} prints an error message
14436 and exits.} You must use the same port number with the host @value{GDBN}
14437 @code{target remote} command.
14439 @subsubsection Attaching to a Running Program
14441 On some targets, @code{gdbserver} can also attach to running programs.
14442 This is accomplished via the @code{--attach} argument. The syntax is:
14445 target> gdbserver --attach @var{comm} @var{pid}
14448 @var{pid} is the process ID of a currently running process. It isn't necessary
14449 to point @code{gdbserver} at a binary for the running process.
14452 @cindex attach to a program by name
14453 You can debug processes by name instead of process ID if your target has the
14454 @code{pidof} utility:
14457 target> gdbserver --attach @var{comm} `pidof @var{program}`
14460 In case more than one copy of @var{program} is running, or @var{program}
14461 has multiple threads, most versions of @code{pidof} support the
14462 @code{-s} option to only return the first process ID.
14464 @subsubsection Multi-Process Mode for @code{gdbserver}
14465 @cindex gdbserver, multiple processes
14466 @cindex multiple processes with gdbserver
14468 When you connect to @code{gdbserver} using @code{target remote},
14469 @code{gdbserver} debugs the specified program only once. When the
14470 program exits, or you detach from it, @value{GDBN} closes the connection
14471 and @code{gdbserver} exits.
14473 If you connect using @kbd{target extended-remote}, @code{gdbserver}
14474 enters multi-process mode. When the debugged program exits, or you
14475 detach from it, @value{GDBN} stays connected to @code{gdbserver} even
14476 though no program is running. The @code{run} and @code{attach}
14477 commands instruct @code{gdbserver} to run or attach to a new program.
14478 The @code{run} command uses @code{set remote exec-file} (@pxref{set
14479 remote exec-file}) to select the program to run. Command line
14480 arguments are supported, except for wildcard expansion and I/O
14481 redirection (@pxref{Arguments}).
14483 To start @code{gdbserver} without supplying an initial command to run
14484 or process ID to attach, use the @option{--multi} command line option.
14485 Then you can connect using @kbd{target extended-remote} and start
14486 the program you want to debug.
14488 @code{gdbserver} does not automatically exit in multi-process mode.
14489 You can terminate it by using @code{monitor exit}
14490 (@pxref{Monitor Commands for gdbserver}).
14492 @subsubsection Other Command-Line Arguments for @code{gdbserver}
14494 The @option{--debug} option tells @code{gdbserver} to display extra
14495 status information about the debugging process. The
14496 @option{--remote-debug} option tells @code{gdbserver} to display
14497 remote protocol debug output. These options are intended for
14498 @code{gdbserver} development and for bug reports to the developers.
14500 The @option{--wrapper} option specifies a wrapper to launch programs
14501 for debugging. The option should be followed by the name of the
14502 wrapper, then any command-line arguments to pass to the wrapper, then
14503 @kbd{--} indicating the end of the wrapper arguments.
14505 @code{gdbserver} runs the specified wrapper program with a combined
14506 command line including the wrapper arguments, then the name of the
14507 program to debug, then any arguments to the program. The wrapper
14508 runs until it executes your program, and then @value{GDBN} gains control.
14510 You can use any program that eventually calls @code{execve} with
14511 its arguments as a wrapper. Several standard Unix utilities do
14512 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
14513 with @code{exec "$@@"} will also work.
14515 For example, you can use @code{env} to pass an environment variable to
14516 the debugged program, without setting the variable in @code{gdbserver}'s
14520 $ gdbserver --wrapper env LD_PRELOAD=libtest.so -- :2222 ./testprog
14523 @subsection Connecting to @code{gdbserver}
14525 Run @value{GDBN} on the host system.
14527 First make sure you have the necessary symbol files. Load symbols for
14528 your application using the @code{file} command before you connect. Use
14529 @code{set sysroot} to locate target libraries (unless your @value{GDBN}
14530 was compiled with the correct sysroot using @code{--with-sysroot}).
14532 The symbol file and target libraries must exactly match the executable
14533 and libraries on the target, with one exception: the files on the host
14534 system should not be stripped, even if the files on the target system
14535 are. Mismatched or missing files will lead to confusing results
14536 during debugging. On @sc{gnu}/Linux targets, mismatched or missing
14537 files may also prevent @code{gdbserver} from debugging multi-threaded
14540 Connect to your target (@pxref{Connecting,,Connecting to a Remote Target}).
14541 For TCP connections, you must start up @code{gdbserver} prior to using
14542 the @code{target remote} command. Otherwise you may get an error whose
14543 text depends on the host system, but which usually looks something like
14544 @samp{Connection refused}. Don't use the @code{load}
14545 command in @value{GDBN} when using @code{gdbserver}, since the program is
14546 already on the target.
14548 @subsection Monitor Commands for @code{gdbserver}
14549 @cindex monitor commands, for @code{gdbserver}
14550 @anchor{Monitor Commands for gdbserver}
14552 During a @value{GDBN} session using @code{gdbserver}, you can use the
14553 @code{monitor} command to send special requests to @code{gdbserver}.
14554 Here are the available commands.
14558 List the available monitor commands.
14560 @item monitor set debug 0
14561 @itemx monitor set debug 1
14562 Disable or enable general debugging messages.
14564 @item monitor set remote-debug 0
14565 @itemx monitor set remote-debug 1
14566 Disable or enable specific debugging messages associated with the remote
14567 protocol (@pxref{Remote Protocol}).
14570 Tell gdbserver to exit immediately. This command should be followed by
14571 @code{disconnect} to close the debugging session. @code{gdbserver} will
14572 detach from any attached processes and kill any processes it created.
14573 Use @code{monitor exit} to terminate @code{gdbserver} at the end
14574 of a multi-process mode debug session.
14578 @node Remote Configuration
14579 @section Remote Configuration
14582 @kindex show remote
14583 This section documents the configuration options available when
14584 debugging remote programs. For the options related to the File I/O
14585 extensions of the remote protocol, see @ref{system,
14586 system-call-allowed}.
14589 @item set remoteaddresssize @var{bits}
14590 @cindex address size for remote targets
14591 @cindex bits in remote address
14592 Set the maximum size of address in a memory packet to the specified
14593 number of bits. @value{GDBN} will mask off the address bits above
14594 that number, when it passes addresses to the remote target. The
14595 default value is the number of bits in the target's address.
14597 @item show remoteaddresssize
14598 Show the current value of remote address size in bits.
14600 @item set remotebaud @var{n}
14601 @cindex baud rate for remote targets
14602 Set the baud rate for the remote serial I/O to @var{n} baud. The
14603 value is used to set the speed of the serial port used for debugging
14606 @item show remotebaud
14607 Show the current speed of the remote connection.
14609 @item set remotebreak
14610 @cindex interrupt remote programs
14611 @cindex BREAK signal instead of Ctrl-C
14612 @anchor{set remotebreak}
14613 If set to on, @value{GDBN} sends a @code{BREAK} signal to the remote
14614 when you type @kbd{Ctrl-c} to interrupt the program running
14615 on the remote. If set to off, @value{GDBN} sends the @samp{Ctrl-C}
14616 character instead. The default is off, since most remote systems
14617 expect to see @samp{Ctrl-C} as the interrupt signal.
14619 @item show remotebreak
14620 Show whether @value{GDBN} sends @code{BREAK} or @samp{Ctrl-C} to
14621 interrupt the remote program.
14623 @item set remoteflow on
14624 @itemx set remoteflow off
14625 @kindex set remoteflow
14626 Enable or disable hardware flow control (@code{RTS}/@code{CTS})
14627 on the serial port used to communicate to the remote target.
14629 @item show remoteflow
14630 @kindex show remoteflow
14631 Show the current setting of hardware flow control.
14633 @item set remotelogbase @var{base}
14634 Set the base (a.k.a.@: radix) of logging serial protocol
14635 communications to @var{base}. Supported values of @var{base} are:
14636 @code{ascii}, @code{octal}, and @code{hex}. The default is
14639 @item show remotelogbase
14640 Show the current setting of the radix for logging remote serial
14643 @item set remotelogfile @var{file}
14644 @cindex record serial communications on file
14645 Record remote serial communications on the named @var{file}. The
14646 default is not to record at all.
14648 @item show remotelogfile.
14649 Show the current setting of the file name on which to record the
14650 serial communications.
14652 @item set remotetimeout @var{num}
14653 @cindex timeout for serial communications
14654 @cindex remote timeout
14655 Set the timeout limit to wait for the remote target to respond to
14656 @var{num} seconds. The default is 2 seconds.
14658 @item show remotetimeout
14659 Show the current number of seconds to wait for the remote target
14662 @cindex limit hardware breakpoints and watchpoints
14663 @cindex remote target, limit break- and watchpoints
14664 @anchor{set remote hardware-watchpoint-limit}
14665 @anchor{set remote hardware-breakpoint-limit}
14666 @item set remote hardware-watchpoint-limit @var{limit}
14667 @itemx set remote hardware-breakpoint-limit @var{limit}
14668 Restrict @value{GDBN} to using @var{limit} remote hardware breakpoint or
14669 watchpoints. A limit of -1, the default, is treated as unlimited.
14671 @item set remote exec-file @var{filename}
14672 @itemx show remote exec-file
14673 @anchor{set remote exec-file}
14674 @cindex executable file, for remote target
14675 Select the file used for @code{run} with @code{target
14676 extended-remote}. This should be set to a filename valid on the
14677 target system. If it is not set, the target will use a default
14678 filename (e.g.@: the last program run).
14682 @item set tcp auto-retry on
14683 @cindex auto-retry, for remote TCP target
14684 Enable auto-retry for remote TCP connections. This is useful if the remote
14685 debugging agent is launched in parallel with @value{GDBN}; there is a race
14686 condition because the agent may not become ready to accept the connection
14687 before @value{GDBN} attempts to connect. When auto-retry is
14688 enabled, if the initial attempt to connect fails, @value{GDBN} reattempts
14689 to establish the connection using the timeout specified by
14690 @code{set tcp connect-timeout}.
14692 @item set tcp auto-retry off
14693 Do not auto-retry failed TCP connections.
14695 @item show tcp auto-retry
14696 Show the current auto-retry setting.
14698 @item set tcp connect-timeout @var{seconds}
14699 @cindex connection timeout, for remote TCP target
14700 @cindex timeout, for remote target connection
14701 Set the timeout for establishing a TCP connection to the remote target to
14702 @var{seconds}. The timeout affects both polling to retry failed connections
14703 (enabled by @code{set tcp auto-retry on}) and waiting for connections
14704 that are merely slow to complete, and represents an approximate cumulative
14707 @item show tcp connect-timeout
14708 Show the current connection timeout setting.
14711 @cindex remote packets, enabling and disabling
14712 The @value{GDBN} remote protocol autodetects the packets supported by
14713 your debugging stub. If you need to override the autodetection, you
14714 can use these commands to enable or disable individual packets. Each
14715 packet can be set to @samp{on} (the remote target supports this
14716 packet), @samp{off} (the remote target does not support this packet),
14717 or @samp{auto} (detect remote target support for this packet). They
14718 all default to @samp{auto}. For more information about each packet,
14719 see @ref{Remote Protocol}.
14721 During normal use, you should not have to use any of these commands.
14722 If you do, that may be a bug in your remote debugging stub, or a bug
14723 in @value{GDBN}. You may want to report the problem to the
14724 @value{GDBN} developers.
14726 For each packet @var{name}, the command to enable or disable the
14727 packet is @code{set remote @var{name}-packet}. The available settings
14730 @multitable @columnfractions 0.28 0.32 0.25
14733 @tab Related Features
14735 @item @code{fetch-register}
14737 @tab @code{info registers}
14739 @item @code{set-register}
14743 @item @code{binary-download}
14745 @tab @code{load}, @code{set}
14747 @item @code{read-aux-vector}
14748 @tab @code{qXfer:auxv:read}
14749 @tab @code{info auxv}
14751 @item @code{symbol-lookup}
14752 @tab @code{qSymbol}
14753 @tab Detecting multiple threads
14755 @item @code{attach}
14756 @tab @code{vAttach}
14759 @item @code{verbose-resume}
14761 @tab Stepping or resuming multiple threads
14767 @item @code{software-breakpoint}
14771 @item @code{hardware-breakpoint}
14775 @item @code{write-watchpoint}
14779 @item @code{read-watchpoint}
14783 @item @code{access-watchpoint}
14787 @item @code{target-features}
14788 @tab @code{qXfer:features:read}
14789 @tab @code{set architecture}
14791 @item @code{library-info}
14792 @tab @code{qXfer:libraries:read}
14793 @tab @code{info sharedlibrary}
14795 @item @code{memory-map}
14796 @tab @code{qXfer:memory-map:read}
14797 @tab @code{info mem}
14799 @item @code{read-spu-object}
14800 @tab @code{qXfer:spu:read}
14801 @tab @code{info spu}
14803 @item @code{write-spu-object}
14804 @tab @code{qXfer:spu:write}
14805 @tab @code{info spu}
14807 @item @code{read-siginfo-object}
14808 @tab @code{qXfer:siginfo:read}
14809 @tab @code{print $_siginfo}
14811 @item @code{write-siginfo-object}
14812 @tab @code{qXfer:siginfo:write}
14813 @tab @code{set $_siginfo}
14815 @item @code{get-thread-local-@*storage-address}
14816 @tab @code{qGetTLSAddr}
14817 @tab Displaying @code{__thread} variables
14819 @item @code{search-memory}
14820 @tab @code{qSearch:memory}
14823 @item @code{supported-packets}
14824 @tab @code{qSupported}
14825 @tab Remote communications parameters
14827 @item @code{pass-signals}
14828 @tab @code{QPassSignals}
14829 @tab @code{handle @var{signal}}
14831 @item @code{hostio-close-packet}
14832 @tab @code{vFile:close}
14833 @tab @code{remote get}, @code{remote put}
14835 @item @code{hostio-open-packet}
14836 @tab @code{vFile:open}
14837 @tab @code{remote get}, @code{remote put}
14839 @item @code{hostio-pread-packet}
14840 @tab @code{vFile:pread}
14841 @tab @code{remote get}, @code{remote put}
14843 @item @code{hostio-pwrite-packet}
14844 @tab @code{vFile:pwrite}
14845 @tab @code{remote get}, @code{remote put}
14847 @item @code{hostio-unlink-packet}
14848 @tab @code{vFile:unlink}
14849 @tab @code{remote delete}
14851 @item @code{noack-packet}
14852 @tab @code{QStartNoAckMode}
14853 @tab Packet acknowledgment
14855 @item @code{osdata}
14856 @tab @code{qXfer:osdata:read}
14857 @tab @code{info os}
14859 @item @code{query-attached}
14860 @tab @code{qAttached}
14861 @tab Querying remote process attach state.
14865 @section Implementing a Remote Stub
14867 @cindex debugging stub, example
14868 @cindex remote stub, example
14869 @cindex stub example, remote debugging
14870 The stub files provided with @value{GDBN} implement the target side of the
14871 communication protocol, and the @value{GDBN} side is implemented in the
14872 @value{GDBN} source file @file{remote.c}. Normally, you can simply allow
14873 these subroutines to communicate, and ignore the details. (If you're
14874 implementing your own stub file, you can still ignore the details: start
14875 with one of the existing stub files. @file{sparc-stub.c} is the best
14876 organized, and therefore the easiest to read.)
14878 @cindex remote serial debugging, overview
14879 To debug a program running on another machine (the debugging
14880 @dfn{target} machine), you must first arrange for all the usual
14881 prerequisites for the program to run by itself. For example, for a C
14886 A startup routine to set up the C runtime environment; these usually
14887 have a name like @file{crt0}. The startup routine may be supplied by
14888 your hardware supplier, or you may have to write your own.
14891 A C subroutine library to support your program's
14892 subroutine calls, notably managing input and output.
14895 A way of getting your program to the other machine---for example, a
14896 download program. These are often supplied by the hardware
14897 manufacturer, but you may have to write your own from hardware
14901 The next step is to arrange for your program to use a serial port to
14902 communicate with the machine where @value{GDBN} is running (the @dfn{host}
14903 machine). In general terms, the scheme looks like this:
14907 @value{GDBN} already understands how to use this protocol; when everything
14908 else is set up, you can simply use the @samp{target remote} command
14909 (@pxref{Targets,,Specifying a Debugging Target}).
14911 @item On the target,
14912 you must link with your program a few special-purpose subroutines that
14913 implement the @value{GDBN} remote serial protocol. The file containing these
14914 subroutines is called a @dfn{debugging stub}.
14916 On certain remote targets, you can use an auxiliary program
14917 @code{gdbserver} instead of linking a stub into your program.
14918 @xref{Server,,Using the @code{gdbserver} Program}, for details.
14921 The debugging stub is specific to the architecture of the remote
14922 machine; for example, use @file{sparc-stub.c} to debug programs on
14925 @cindex remote serial stub list
14926 These working remote stubs are distributed with @value{GDBN}:
14931 @cindex @file{i386-stub.c}
14934 For Intel 386 and compatible architectures.
14937 @cindex @file{m68k-stub.c}
14938 @cindex Motorola 680x0
14940 For Motorola 680x0 architectures.
14943 @cindex @file{sh-stub.c}
14946 For Renesas SH architectures.
14949 @cindex @file{sparc-stub.c}
14951 For @sc{sparc} architectures.
14953 @item sparcl-stub.c
14954 @cindex @file{sparcl-stub.c}
14957 For Fujitsu @sc{sparclite} architectures.
14961 The @file{README} file in the @value{GDBN} distribution may list other
14962 recently added stubs.
14965 * Stub Contents:: What the stub can do for you
14966 * Bootstrapping:: What you must do for the stub
14967 * Debug Session:: Putting it all together
14970 @node Stub Contents
14971 @subsection What the Stub Can Do for You
14973 @cindex remote serial stub
14974 The debugging stub for your architecture supplies these three
14978 @item set_debug_traps
14979 @findex set_debug_traps
14980 @cindex remote serial stub, initialization
14981 This routine arranges for @code{handle_exception} to run when your
14982 program stops. You must call this subroutine explicitly near the
14983 beginning of your program.
14985 @item handle_exception
14986 @findex handle_exception
14987 @cindex remote serial stub, main routine
14988 This is the central workhorse, but your program never calls it
14989 explicitly---the setup code arranges for @code{handle_exception} to
14990 run when a trap is triggered.
14992 @code{handle_exception} takes control when your program stops during
14993 execution (for example, on a breakpoint), and mediates communications
14994 with @value{GDBN} on the host machine. This is where the communications
14995 protocol is implemented; @code{handle_exception} acts as the @value{GDBN}
14996 representative on the target machine. It begins by sending summary
14997 information on the state of your program, then continues to execute,
14998 retrieving and transmitting any information @value{GDBN} needs, until you
14999 execute a @value{GDBN} command that makes your program resume; at that point,
15000 @code{handle_exception} returns control to your own code on the target
15004 @cindex @code{breakpoint} subroutine, remote
15005 Use this auxiliary subroutine to make your program contain a
15006 breakpoint. Depending on the particular situation, this may be the only
15007 way for @value{GDBN} to get control. For instance, if your target
15008 machine has some sort of interrupt button, you won't need to call this;
15009 pressing the interrupt button transfers control to
15010 @code{handle_exception}---in effect, to @value{GDBN}. On some machines,
15011 simply receiving characters on the serial port may also trigger a trap;
15012 again, in that situation, you don't need to call @code{breakpoint} from
15013 your own program---simply running @samp{target remote} from the host
15014 @value{GDBN} session gets control.
15016 Call @code{breakpoint} if none of these is true, or if you simply want
15017 to make certain your program stops at a predetermined point for the
15018 start of your debugging session.
15021 @node Bootstrapping
15022 @subsection What You Must Do for the Stub
15024 @cindex remote stub, support routines
15025 The debugging stubs that come with @value{GDBN} are set up for a particular
15026 chip architecture, but they have no information about the rest of your
15027 debugging target machine.
15029 First of all you need to tell the stub how to communicate with the
15033 @item int getDebugChar()
15034 @findex getDebugChar
15035 Write this subroutine to read a single character from the serial port.
15036 It may be identical to @code{getchar} for your target system; a
15037 different name is used to allow you to distinguish the two if you wish.
15039 @item void putDebugChar(int)
15040 @findex putDebugChar
15041 Write this subroutine to write a single character to the serial port.
15042 It may be identical to @code{putchar} for your target system; a
15043 different name is used to allow you to distinguish the two if you wish.
15046 @cindex control C, and remote debugging
15047 @cindex interrupting remote targets
15048 If you want @value{GDBN} to be able to stop your program while it is
15049 running, you need to use an interrupt-driven serial driver, and arrange
15050 for it to stop when it receives a @code{^C} (@samp{\003}, the control-C
15051 character). That is the character which @value{GDBN} uses to tell the
15052 remote system to stop.
15054 Getting the debugging target to return the proper status to @value{GDBN}
15055 probably requires changes to the standard stub; one quick and dirty way
15056 is to just execute a breakpoint instruction (the ``dirty'' part is that
15057 @value{GDBN} reports a @code{SIGTRAP} instead of a @code{SIGINT}).
15059 Other routines you need to supply are:
15062 @item void exceptionHandler (int @var{exception_number}, void *@var{exception_address})
15063 @findex exceptionHandler
15064 Write this function to install @var{exception_address} in the exception
15065 handling tables. You need to do this because the stub does not have any
15066 way of knowing what the exception handling tables on your target system
15067 are like (for example, the processor's table might be in @sc{rom},
15068 containing entries which point to a table in @sc{ram}).
15069 @var{exception_number} is the exception number which should be changed;
15070 its meaning is architecture-dependent (for example, different numbers
15071 might represent divide by zero, misaligned access, etc). When this
15072 exception occurs, control should be transferred directly to
15073 @var{exception_address}, and the processor state (stack, registers,
15074 and so on) should be just as it is when a processor exception occurs. So if
15075 you want to use a jump instruction to reach @var{exception_address}, it
15076 should be a simple jump, not a jump to subroutine.
15078 For the 386, @var{exception_address} should be installed as an interrupt
15079 gate so that interrupts are masked while the handler runs. The gate
15080 should be at privilege level 0 (the most privileged level). The
15081 @sc{sparc} and 68k stubs are able to mask interrupts themselves without
15082 help from @code{exceptionHandler}.
15084 @item void flush_i_cache()
15085 @findex flush_i_cache
15086 On @sc{sparc} and @sc{sparclite} only, write this subroutine to flush the
15087 instruction cache, if any, on your target machine. If there is no
15088 instruction cache, this subroutine may be a no-op.
15090 On target machines that have instruction caches, @value{GDBN} requires this
15091 function to make certain that the state of your program is stable.
15095 You must also make sure this library routine is available:
15098 @item void *memset(void *, int, int)
15100 This is the standard library function @code{memset} that sets an area of
15101 memory to a known value. If you have one of the free versions of
15102 @code{libc.a}, @code{memset} can be found there; otherwise, you must
15103 either obtain it from your hardware manufacturer, or write your own.
15106 If you do not use the GNU C compiler, you may need other standard
15107 library subroutines as well; this varies from one stub to another,
15108 but in general the stubs are likely to use any of the common library
15109 subroutines which @code{@value{NGCC}} generates as inline code.
15112 @node Debug Session
15113 @subsection Putting it All Together
15115 @cindex remote serial debugging summary
15116 In summary, when your program is ready to debug, you must follow these
15121 Make sure you have defined the supporting low-level routines
15122 (@pxref{Bootstrapping,,What You Must Do for the Stub}):
15124 @code{getDebugChar}, @code{putDebugChar},
15125 @code{flush_i_cache}, @code{memset}, @code{exceptionHandler}.
15129 Insert these lines near the top of your program:
15137 For the 680x0 stub only, you need to provide a variable called
15138 @code{exceptionHook}. Normally you just use:
15141 void (*exceptionHook)() = 0;
15145 but if before calling @code{set_debug_traps}, you set it to point to a
15146 function in your program, that function is called when
15147 @code{@value{GDBN}} continues after stopping on a trap (for example, bus
15148 error). The function indicated by @code{exceptionHook} is called with
15149 one parameter: an @code{int} which is the exception number.
15152 Compile and link together: your program, the @value{GDBN} debugging stub for
15153 your target architecture, and the supporting subroutines.
15156 Make sure you have a serial connection between your target machine and
15157 the @value{GDBN} host, and identify the serial port on the host.
15160 @c The "remote" target now provides a `load' command, so we should
15161 @c document that. FIXME.
15162 Download your program to your target machine (or get it there by
15163 whatever means the manufacturer provides), and start it.
15166 Start @value{GDBN} on the host, and connect to the target
15167 (@pxref{Connecting,,Connecting to a Remote Target}).
15171 @node Configurations
15172 @chapter Configuration-Specific Information
15174 While nearly all @value{GDBN} commands are available for all native and
15175 cross versions of the debugger, there are some exceptions. This chapter
15176 describes things that are only available in certain configurations.
15178 There are three major categories of configurations: native
15179 configurations, where the host and target are the same, embedded
15180 operating system configurations, which are usually the same for several
15181 different processor architectures, and bare embedded processors, which
15182 are quite different from each other.
15187 * Embedded Processors::
15194 This section describes details specific to particular native
15199 * BSD libkvm Interface:: Debugging BSD kernel memory images
15200 * SVR4 Process Information:: SVR4 process information
15201 * DJGPP Native:: Features specific to the DJGPP port
15202 * Cygwin Native:: Features specific to the Cygwin port
15203 * Hurd Native:: Features specific to @sc{gnu} Hurd
15204 * Neutrino:: Features specific to QNX Neutrino
15205 * Darwin:: Features specific to Darwin
15211 On HP-UX systems, if you refer to a function or variable name that
15212 begins with a dollar sign, @value{GDBN} searches for a user or system
15213 name first, before it searches for a convenience variable.
15216 @node BSD libkvm Interface
15217 @subsection BSD libkvm Interface
15220 @cindex kernel memory image
15221 @cindex kernel crash dump
15223 BSD-derived systems (FreeBSD/NetBSD/OpenBSD) have a kernel memory
15224 interface that provides a uniform interface for accessing kernel virtual
15225 memory images, including live systems and crash dumps. @value{GDBN}
15226 uses this interface to allow you to debug live kernels and kernel crash
15227 dumps on many native BSD configurations. This is implemented as a
15228 special @code{kvm} debugging target. For debugging a live system, load
15229 the currently running kernel into @value{GDBN} and connect to the
15233 (@value{GDBP}) @b{target kvm}
15236 For debugging crash dumps, provide the file name of the crash dump as an
15240 (@value{GDBP}) @b{target kvm /var/crash/bsd.0}
15243 Once connected to the @code{kvm} target, the following commands are
15249 Set current context from the @dfn{Process Control Block} (PCB) address.
15252 Set current context from proc address. This command isn't available on
15253 modern FreeBSD systems.
15256 @node SVR4 Process Information
15257 @subsection SVR4 Process Information
15259 @cindex examine process image
15260 @cindex process info via @file{/proc}
15262 Many versions of SVR4 and compatible systems provide a facility called
15263 @samp{/proc} that can be used to examine the image of a running
15264 process using file-system subroutines. If @value{GDBN} is configured
15265 for an operating system with this facility, the command @code{info
15266 proc} is available to report information about the process running
15267 your program, or about any process running on your system. @code{info
15268 proc} works only on SVR4 systems that include the @code{procfs} code.
15269 This includes, as of this writing, @sc{gnu}/Linux, OSF/1 (Digital
15270 Unix), Solaris, Irix, and Unixware, but not HP-UX, for example.
15276 @itemx info proc @var{process-id}
15277 Summarize available information about any running process. If a
15278 process ID is specified by @var{process-id}, display information about
15279 that process; otherwise display information about the program being
15280 debugged. The summary includes the debugged process ID, the command
15281 line used to invoke it, its current working directory, and its
15282 executable file's absolute file name.
15284 On some systems, @var{process-id} can be of the form
15285 @samp{[@var{pid}]/@var{tid}} which specifies a certain thread ID
15286 within a process. If the optional @var{pid} part is missing, it means
15287 a thread from the process being debugged (the leading @samp{/} still
15288 needs to be present, or else @value{GDBN} will interpret the number as
15289 a process ID rather than a thread ID).
15291 @item info proc mappings
15292 @cindex memory address space mappings
15293 Report the memory address space ranges accessible in the program, with
15294 information on whether the process has read, write, or execute access
15295 rights to each range. On @sc{gnu}/Linux systems, each memory range
15296 includes the object file which is mapped to that range, instead of the
15297 memory access rights to that range.
15299 @item info proc stat
15300 @itemx info proc status
15301 @cindex process detailed status information
15302 These subcommands are specific to @sc{gnu}/Linux systems. They show
15303 the process-related information, including the user ID and group ID;
15304 how many threads are there in the process; its virtual memory usage;
15305 the signals that are pending, blocked, and ignored; its TTY; its
15306 consumption of system and user time; its stack size; its @samp{nice}
15307 value; etc. For more information, see the @samp{proc} man page
15308 (type @kbd{man 5 proc} from your shell prompt).
15310 @item info proc all
15311 Show all the information about the process described under all of the
15312 above @code{info proc} subcommands.
15315 @comment These sub-options of 'info proc' were not included when
15316 @comment procfs.c was re-written. Keep their descriptions around
15317 @comment against the day when someone finds the time to put them back in.
15318 @kindex info proc times
15319 @item info proc times
15320 Starting time, user CPU time, and system CPU time for your program and
15323 @kindex info proc id
15325 Report on the process IDs related to your program: its own process ID,
15326 the ID of its parent, the process group ID, and the session ID.
15329 @item set procfs-trace
15330 @kindex set procfs-trace
15331 @cindex @code{procfs} API calls
15332 This command enables and disables tracing of @code{procfs} API calls.
15334 @item show procfs-trace
15335 @kindex show procfs-trace
15336 Show the current state of @code{procfs} API call tracing.
15338 @item set procfs-file @var{file}
15339 @kindex set procfs-file
15340 Tell @value{GDBN} to write @code{procfs} API trace to the named
15341 @var{file}. @value{GDBN} appends the trace info to the previous
15342 contents of the file. The default is to display the trace on the
15345 @item show procfs-file
15346 @kindex show procfs-file
15347 Show the file to which @code{procfs} API trace is written.
15349 @item proc-trace-entry
15350 @itemx proc-trace-exit
15351 @itemx proc-untrace-entry
15352 @itemx proc-untrace-exit
15353 @kindex proc-trace-entry
15354 @kindex proc-trace-exit
15355 @kindex proc-untrace-entry
15356 @kindex proc-untrace-exit
15357 These commands enable and disable tracing of entries into and exits
15358 from the @code{syscall} interface.
15361 @kindex info pidlist
15362 @cindex process list, QNX Neutrino
15363 For QNX Neutrino only, this command displays the list of all the
15364 processes and all the threads within each process.
15367 @kindex info meminfo
15368 @cindex mapinfo list, QNX Neutrino
15369 For QNX Neutrino only, this command displays the list of all mapinfos.
15373 @subsection Features for Debugging @sc{djgpp} Programs
15374 @cindex @sc{djgpp} debugging
15375 @cindex native @sc{djgpp} debugging
15376 @cindex MS-DOS-specific commands
15379 @sc{djgpp} is a port of the @sc{gnu} development tools to MS-DOS and
15380 MS-Windows. @sc{djgpp} programs are 32-bit protected-mode programs
15381 that use the @dfn{DPMI} (DOS Protected-Mode Interface) API to run on
15382 top of real-mode DOS systems and their emulations.
15384 @value{GDBN} supports native debugging of @sc{djgpp} programs, and
15385 defines a few commands specific to the @sc{djgpp} port. This
15386 subsection describes those commands.
15391 This is a prefix of @sc{djgpp}-specific commands which print
15392 information about the target system and important OS structures.
15395 @cindex MS-DOS system info
15396 @cindex free memory information (MS-DOS)
15397 @item info dos sysinfo
15398 This command displays assorted information about the underlying
15399 platform: the CPU type and features, the OS version and flavor, the
15400 DPMI version, and the available conventional and DPMI memory.
15405 @cindex segment descriptor tables
15406 @cindex descriptor tables display
15408 @itemx info dos ldt
15409 @itemx info dos idt
15410 These 3 commands display entries from, respectively, Global, Local,
15411 and Interrupt Descriptor Tables (GDT, LDT, and IDT). The descriptor
15412 tables are data structures which store a descriptor for each segment
15413 that is currently in use. The segment's selector is an index into a
15414 descriptor table; the table entry for that index holds the
15415 descriptor's base address and limit, and its attributes and access
15418 A typical @sc{djgpp} program uses 3 segments: a code segment, a data
15419 segment (used for both data and the stack), and a DOS segment (which
15420 allows access to DOS/BIOS data structures and absolute addresses in
15421 conventional memory). However, the DPMI host will usually define
15422 additional segments in order to support the DPMI environment.
15424 @cindex garbled pointers
15425 These commands allow to display entries from the descriptor tables.
15426 Without an argument, all entries from the specified table are
15427 displayed. An argument, which should be an integer expression, means
15428 display a single entry whose index is given by the argument. For
15429 example, here's a convenient way to display information about the
15430 debugged program's data segment:
15433 @exdent @code{(@value{GDBP}) info dos ldt $ds}
15434 @exdent @code{0x13f: base=0x11970000 limit=0x0009ffff 32-Bit Data (Read/Write, Exp-up)}
15438 This comes in handy when you want to see whether a pointer is outside
15439 the data segment's limit (i.e.@: @dfn{garbled}).
15441 @cindex page tables display (MS-DOS)
15443 @itemx info dos pte
15444 These two commands display entries from, respectively, the Page
15445 Directory and the Page Tables. Page Directories and Page Tables are
15446 data structures which control how virtual memory addresses are mapped
15447 into physical addresses. A Page Table includes an entry for every
15448 page of memory that is mapped into the program's address space; there
15449 may be several Page Tables, each one holding up to 4096 entries. A
15450 Page Directory has up to 4096 entries, one each for every Page Table
15451 that is currently in use.
15453 Without an argument, @kbd{info dos pde} displays the entire Page
15454 Directory, and @kbd{info dos pte} displays all the entries in all of
15455 the Page Tables. An argument, an integer expression, given to the
15456 @kbd{info dos pde} command means display only that entry from the Page
15457 Directory table. An argument given to the @kbd{info dos pte} command
15458 means display entries from a single Page Table, the one pointed to by
15459 the specified entry in the Page Directory.
15461 @cindex direct memory access (DMA) on MS-DOS
15462 These commands are useful when your program uses @dfn{DMA} (Direct
15463 Memory Access), which needs physical addresses to program the DMA
15466 These commands are supported only with some DPMI servers.
15468 @cindex physical address from linear address
15469 @item info dos address-pte @var{addr}
15470 This command displays the Page Table entry for a specified linear
15471 address. The argument @var{addr} is a linear address which should
15472 already have the appropriate segment's base address added to it,
15473 because this command accepts addresses which may belong to @emph{any}
15474 segment. For example, here's how to display the Page Table entry for
15475 the page where a variable @code{i} is stored:
15478 @exdent @code{(@value{GDBP}) info dos address-pte __djgpp_base_address + (char *)&i}
15479 @exdent @code{Page Table entry for address 0x11a00d30:}
15480 @exdent @code{Base=0x02698000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0xd30}
15484 This says that @code{i} is stored at offset @code{0xd30} from the page
15485 whose physical base address is @code{0x02698000}, and shows all the
15486 attributes of that page.
15488 Note that you must cast the addresses of variables to a @code{char *},
15489 since otherwise the value of @code{__djgpp_base_address}, the base
15490 address of all variables and functions in a @sc{djgpp} program, will
15491 be added using the rules of C pointer arithmetics: if @code{i} is
15492 declared an @code{int}, @value{GDBN} will add 4 times the value of
15493 @code{__djgpp_base_address} to the address of @code{i}.
15495 Here's another example, it displays the Page Table entry for the
15499 @exdent @code{(@value{GDBP}) info dos address-pte *((unsigned *)&_go32_info_block + 3)}
15500 @exdent @code{Page Table entry for address 0x29110:}
15501 @exdent @code{Base=0x00029000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0x110}
15505 (The @code{+ 3} offset is because the transfer buffer's address is the
15506 3rd member of the @code{_go32_info_block} structure.) The output
15507 clearly shows that this DPMI server maps the addresses in conventional
15508 memory 1:1, i.e.@: the physical (@code{0x00029000} + @code{0x110}) and
15509 linear (@code{0x29110}) addresses are identical.
15511 This command is supported only with some DPMI servers.
15514 @cindex DOS serial data link, remote debugging
15515 In addition to native debugging, the DJGPP port supports remote
15516 debugging via a serial data link. The following commands are specific
15517 to remote serial debugging in the DJGPP port of @value{GDBN}.
15520 @kindex set com1base
15521 @kindex set com1irq
15522 @kindex set com2base
15523 @kindex set com2irq
15524 @kindex set com3base
15525 @kindex set com3irq
15526 @kindex set com4base
15527 @kindex set com4irq
15528 @item set com1base @var{addr}
15529 This command sets the base I/O port address of the @file{COM1} serial
15532 @item set com1irq @var{irq}
15533 This command sets the @dfn{Interrupt Request} (@code{IRQ}) line to use
15534 for the @file{COM1} serial port.
15536 There are similar commands @samp{set com2base}, @samp{set com3irq},
15537 etc.@: for setting the port address and the @code{IRQ} lines for the
15540 @kindex show com1base
15541 @kindex show com1irq
15542 @kindex show com2base
15543 @kindex show com2irq
15544 @kindex show com3base
15545 @kindex show com3irq
15546 @kindex show com4base
15547 @kindex show com4irq
15548 The related commands @samp{show com1base}, @samp{show com1irq} etc.@:
15549 display the current settings of the base address and the @code{IRQ}
15550 lines used by the COM ports.
15553 @kindex info serial
15554 @cindex DOS serial port status
15555 This command prints the status of the 4 DOS serial ports. For each
15556 port, it prints whether it's active or not, its I/O base address and
15557 IRQ number, whether it uses a 16550-style FIFO, its baudrate, and the
15558 counts of various errors encountered so far.
15562 @node Cygwin Native
15563 @subsection Features for Debugging MS Windows PE Executables
15564 @cindex MS Windows debugging
15565 @cindex native Cygwin debugging
15566 @cindex Cygwin-specific commands
15568 @value{GDBN} supports native debugging of MS Windows programs, including
15569 DLLs with and without symbolic debugging information. There are various
15570 additional Cygwin-specific commands, described in this section.
15571 Working with DLLs that have no debugging symbols is described in
15572 @ref{Non-debug DLL Symbols}.
15577 This is a prefix of MS Windows-specific commands which print
15578 information about the target system and important OS structures.
15580 @item info w32 selector
15581 This command displays information returned by
15582 the Win32 API @code{GetThreadSelectorEntry} function.
15583 It takes an optional argument that is evaluated to
15584 a long value to give the information about this given selector.
15585 Without argument, this command displays information
15586 about the six segment registers.
15590 This is a Cygwin-specific alias of @code{info shared}.
15592 @kindex dll-symbols
15594 This command loads symbols from a dll similarly to
15595 add-sym command but without the need to specify a base address.
15597 @kindex set cygwin-exceptions
15598 @cindex debugging the Cygwin DLL
15599 @cindex Cygwin DLL, debugging
15600 @item set cygwin-exceptions @var{mode}
15601 If @var{mode} is @code{on}, @value{GDBN} will break on exceptions that
15602 happen inside the Cygwin DLL. If @var{mode} is @code{off},
15603 @value{GDBN} will delay recognition of exceptions, and may ignore some
15604 exceptions which seem to be caused by internal Cygwin DLL
15605 ``bookkeeping''. This option is meant primarily for debugging the
15606 Cygwin DLL itself; the default value is @code{off} to avoid annoying
15607 @value{GDBN} users with false @code{SIGSEGV} signals.
15609 @kindex show cygwin-exceptions
15610 @item show cygwin-exceptions
15611 Displays whether @value{GDBN} will break on exceptions that happen
15612 inside the Cygwin DLL itself.
15614 @kindex set new-console
15615 @item set new-console @var{mode}
15616 If @var{mode} is @code{on} the debuggee will
15617 be started in a new console on next start.
15618 If @var{mode} is @code{off}i, the debuggee will
15619 be started in the same console as the debugger.
15621 @kindex show new-console
15622 @item show new-console
15623 Displays whether a new console is used
15624 when the debuggee is started.
15626 @kindex set new-group
15627 @item set new-group @var{mode}
15628 This boolean value controls whether the debuggee should
15629 start a new group or stay in the same group as the debugger.
15630 This affects the way the Windows OS handles
15633 @kindex show new-group
15634 @item show new-group
15635 Displays current value of new-group boolean.
15637 @kindex set debugevents
15638 @item set debugevents
15639 This boolean value adds debug output concerning kernel events related
15640 to the debuggee seen by the debugger. This includes events that
15641 signal thread and process creation and exit, DLL loading and
15642 unloading, console interrupts, and debugging messages produced by the
15643 Windows @code{OutputDebugString} API call.
15645 @kindex set debugexec
15646 @item set debugexec
15647 This boolean value adds debug output concerning execute events
15648 (such as resume thread) seen by the debugger.
15650 @kindex set debugexceptions
15651 @item set debugexceptions
15652 This boolean value adds debug output concerning exceptions in the
15653 debuggee seen by the debugger.
15655 @kindex set debugmemory
15656 @item set debugmemory
15657 This boolean value adds debug output concerning debuggee memory reads
15658 and writes by the debugger.
15662 This boolean values specifies whether the debuggee is called
15663 via a shell or directly (default value is on).
15667 Displays if the debuggee will be started with a shell.
15672 * Non-debug DLL Symbols:: Support for DLLs without debugging symbols
15675 @node Non-debug DLL Symbols
15676 @subsubsection Support for DLLs without Debugging Symbols
15677 @cindex DLLs with no debugging symbols
15678 @cindex Minimal symbols and DLLs
15680 Very often on windows, some of the DLLs that your program relies on do
15681 not include symbolic debugging information (for example,
15682 @file{kernel32.dll}). When @value{GDBN} doesn't recognize any debugging
15683 symbols in a DLL, it relies on the minimal amount of symbolic
15684 information contained in the DLL's export table. This section
15685 describes working with such symbols, known internally to @value{GDBN} as
15686 ``minimal symbols''.
15688 Note that before the debugged program has started execution, no DLLs
15689 will have been loaded. The easiest way around this problem is simply to
15690 start the program --- either by setting a breakpoint or letting the
15691 program run once to completion. It is also possible to force
15692 @value{GDBN} to load a particular DLL before starting the executable ---
15693 see the shared library information in @ref{Files}, or the
15694 @code{dll-symbols} command in @ref{Cygwin Native}. Currently,
15695 explicitly loading symbols from a DLL with no debugging information will
15696 cause the symbol names to be duplicated in @value{GDBN}'s lookup table,
15697 which may adversely affect symbol lookup performance.
15699 @subsubsection DLL Name Prefixes
15701 In keeping with the naming conventions used by the Microsoft debugging
15702 tools, DLL export symbols are made available with a prefix based on the
15703 DLL name, for instance @code{KERNEL32!CreateFileA}. The plain name is
15704 also entered into the symbol table, so @code{CreateFileA} is often
15705 sufficient. In some cases there will be name clashes within a program
15706 (particularly if the executable itself includes full debugging symbols)
15707 necessitating the use of the fully qualified name when referring to the
15708 contents of the DLL. Use single-quotes around the name to avoid the
15709 exclamation mark (``!'') being interpreted as a language operator.
15711 Note that the internal name of the DLL may be all upper-case, even
15712 though the file name of the DLL is lower-case, or vice-versa. Since
15713 symbols within @value{GDBN} are @emph{case-sensitive} this may cause
15714 some confusion. If in doubt, try the @code{info functions} and
15715 @code{info variables} commands or even @code{maint print msymbols}
15716 (@pxref{Symbols}). Here's an example:
15719 (@value{GDBP}) info function CreateFileA
15720 All functions matching regular expression "CreateFileA":
15722 Non-debugging symbols:
15723 0x77e885f4 CreateFileA
15724 0x77e885f4 KERNEL32!CreateFileA
15728 (@value{GDBP}) info function !
15729 All functions matching regular expression "!":
15731 Non-debugging symbols:
15732 0x6100114c cygwin1!__assert
15733 0x61004034 cygwin1!_dll_crt0@@0
15734 0x61004240 cygwin1!dll_crt0(per_process *)
15738 @subsubsection Working with Minimal Symbols
15740 Symbols extracted from a DLL's export table do not contain very much
15741 type information. All that @value{GDBN} can do is guess whether a symbol
15742 refers to a function or variable depending on the linker section that
15743 contains the symbol. Also note that the actual contents of the memory
15744 contained in a DLL are not available unless the program is running. This
15745 means that you cannot examine the contents of a variable or disassemble
15746 a function within a DLL without a running program.
15748 Variables are generally treated as pointers and dereferenced
15749 automatically. For this reason, it is often necessary to prefix a
15750 variable name with the address-of operator (``&'') and provide explicit
15751 type information in the command. Here's an example of the type of
15755 (@value{GDBP}) print 'cygwin1!__argv'
15760 (@value{GDBP}) x 'cygwin1!__argv'
15761 0x10021610: "\230y\""
15764 And two possible solutions:
15767 (@value{GDBP}) print ((char **)'cygwin1!__argv')[0]
15768 $2 = 0x22fd98 "/cygdrive/c/mydirectory/myprogram"
15772 (@value{GDBP}) x/2x &'cygwin1!__argv'
15773 0x610c0aa8 <cygwin1!__argv>: 0x10021608 0x00000000
15774 (@value{GDBP}) x/x 0x10021608
15775 0x10021608: 0x0022fd98
15776 (@value{GDBP}) x/s 0x0022fd98
15777 0x22fd98: "/cygdrive/c/mydirectory/myprogram"
15780 Setting a break point within a DLL is possible even before the program
15781 starts execution. However, under these circumstances, @value{GDBN} can't
15782 examine the initial instructions of the function in order to skip the
15783 function's frame set-up code. You can work around this by using ``*&''
15784 to set the breakpoint at a raw memory address:
15787 (@value{GDBP}) break *&'python22!PyOS_Readline'
15788 Breakpoint 1 at 0x1e04eff0
15791 The author of these extensions is not entirely convinced that setting a
15792 break point within a shared DLL like @file{kernel32.dll} is completely
15796 @subsection Commands Specific to @sc{gnu} Hurd Systems
15797 @cindex @sc{gnu} Hurd debugging
15799 This subsection describes @value{GDBN} commands specific to the
15800 @sc{gnu} Hurd native debugging.
15805 @kindex set signals@r{, Hurd command}
15806 @kindex set sigs@r{, Hurd command}
15807 This command toggles the state of inferior signal interception by
15808 @value{GDBN}. Mach exceptions, such as breakpoint traps, are not
15809 affected by this command. @code{sigs} is a shorthand alias for
15814 @kindex show signals@r{, Hurd command}
15815 @kindex show sigs@r{, Hurd command}
15816 Show the current state of intercepting inferior's signals.
15818 @item set signal-thread
15819 @itemx set sigthread
15820 @kindex set signal-thread
15821 @kindex set sigthread
15822 This command tells @value{GDBN} which thread is the @code{libc} signal
15823 thread. That thread is run when a signal is delivered to a running
15824 process. @code{set sigthread} is the shorthand alias of @code{set
15827 @item show signal-thread
15828 @itemx show sigthread
15829 @kindex show signal-thread
15830 @kindex show sigthread
15831 These two commands show which thread will run when the inferior is
15832 delivered a signal.
15835 @kindex set stopped@r{, Hurd command}
15836 This commands tells @value{GDBN} that the inferior process is stopped,
15837 as with the @code{SIGSTOP} signal. The stopped process can be
15838 continued by delivering a signal to it.
15841 @kindex show stopped@r{, Hurd command}
15842 This command shows whether @value{GDBN} thinks the debuggee is
15845 @item set exceptions
15846 @kindex set exceptions@r{, Hurd command}
15847 Use this command to turn off trapping of exceptions in the inferior.
15848 When exception trapping is off, neither breakpoints nor
15849 single-stepping will work. To restore the default, set exception
15852 @item show exceptions
15853 @kindex show exceptions@r{, Hurd command}
15854 Show the current state of trapping exceptions in the inferior.
15856 @item set task pause
15857 @kindex set task@r{, Hurd commands}
15858 @cindex task attributes (@sc{gnu} Hurd)
15859 @cindex pause current task (@sc{gnu} Hurd)
15860 This command toggles task suspension when @value{GDBN} has control.
15861 Setting it to on takes effect immediately, and the task is suspended
15862 whenever @value{GDBN} gets control. Setting it to off will take
15863 effect the next time the inferior is continued. If this option is set
15864 to off, you can use @code{set thread default pause on} or @code{set
15865 thread pause on} (see below) to pause individual threads.
15867 @item show task pause
15868 @kindex show task@r{, Hurd commands}
15869 Show the current state of task suspension.
15871 @item set task detach-suspend-count
15872 @cindex task suspend count
15873 @cindex detach from task, @sc{gnu} Hurd
15874 This command sets the suspend count the task will be left with when
15875 @value{GDBN} detaches from it.
15877 @item show task detach-suspend-count
15878 Show the suspend count the task will be left with when detaching.
15880 @item set task exception-port
15881 @itemx set task excp
15882 @cindex task exception port, @sc{gnu} Hurd
15883 This command sets the task exception port to which @value{GDBN} will
15884 forward exceptions. The argument should be the value of the @dfn{send
15885 rights} of the task. @code{set task excp} is a shorthand alias.
15887 @item set noninvasive
15888 @cindex noninvasive task options
15889 This command switches @value{GDBN} to a mode that is the least
15890 invasive as far as interfering with the inferior is concerned. This
15891 is the same as using @code{set task pause}, @code{set exceptions}, and
15892 @code{set signals} to values opposite to the defaults.
15894 @item info send-rights
15895 @itemx info receive-rights
15896 @itemx info port-rights
15897 @itemx info port-sets
15898 @itemx info dead-names
15901 @cindex send rights, @sc{gnu} Hurd
15902 @cindex receive rights, @sc{gnu} Hurd
15903 @cindex port rights, @sc{gnu} Hurd
15904 @cindex port sets, @sc{gnu} Hurd
15905 @cindex dead names, @sc{gnu} Hurd
15906 These commands display information about, respectively, send rights,
15907 receive rights, port rights, port sets, and dead names of a task.
15908 There are also shorthand aliases: @code{info ports} for @code{info
15909 port-rights} and @code{info psets} for @code{info port-sets}.
15911 @item set thread pause
15912 @kindex set thread@r{, Hurd command}
15913 @cindex thread properties, @sc{gnu} Hurd
15914 @cindex pause current thread (@sc{gnu} Hurd)
15915 This command toggles current thread suspension when @value{GDBN} has
15916 control. Setting it to on takes effect immediately, and the current
15917 thread is suspended whenever @value{GDBN} gets control. Setting it to
15918 off will take effect the next time the inferior is continued.
15919 Normally, this command has no effect, since when @value{GDBN} has
15920 control, the whole task is suspended. However, if you used @code{set
15921 task pause off} (see above), this command comes in handy to suspend
15922 only the current thread.
15924 @item show thread pause
15925 @kindex show thread@r{, Hurd command}
15926 This command shows the state of current thread suspension.
15928 @item set thread run
15929 This command sets whether the current thread is allowed to run.
15931 @item show thread run
15932 Show whether the current thread is allowed to run.
15934 @item set thread detach-suspend-count
15935 @cindex thread suspend count, @sc{gnu} Hurd
15936 @cindex detach from thread, @sc{gnu} Hurd
15937 This command sets the suspend count @value{GDBN} will leave on a
15938 thread when detaching. This number is relative to the suspend count
15939 found by @value{GDBN} when it notices the thread; use @code{set thread
15940 takeover-suspend-count} to force it to an absolute value.
15942 @item show thread detach-suspend-count
15943 Show the suspend count @value{GDBN} will leave on the thread when
15946 @item set thread exception-port
15947 @itemx set thread excp
15948 Set the thread exception port to which to forward exceptions. This
15949 overrides the port set by @code{set task exception-port} (see above).
15950 @code{set thread excp} is the shorthand alias.
15952 @item set thread takeover-suspend-count
15953 Normally, @value{GDBN}'s thread suspend counts are relative to the
15954 value @value{GDBN} finds when it notices each thread. This command
15955 changes the suspend counts to be absolute instead.
15957 @item set thread default
15958 @itemx show thread default
15959 @cindex thread default settings, @sc{gnu} Hurd
15960 Each of the above @code{set thread} commands has a @code{set thread
15961 default} counterpart (e.g., @code{set thread default pause}, @code{set
15962 thread default exception-port}, etc.). The @code{thread default}
15963 variety of commands sets the default thread properties for all
15964 threads; you can then change the properties of individual threads with
15965 the non-default commands.
15970 @subsection QNX Neutrino
15971 @cindex QNX Neutrino
15973 @value{GDBN} provides the following commands specific to the QNX
15977 @item set debug nto-debug
15978 @kindex set debug nto-debug
15979 When set to on, enables debugging messages specific to the QNX
15982 @item show debug nto-debug
15983 @kindex show debug nto-debug
15984 Show the current state of QNX Neutrino messages.
15991 @value{GDBN} provides the following commands specific to the Darwin target:
15994 @item set debug darwin @var{num}
15995 @kindex set debug darwin
15996 When set to a non zero value, enables debugging messages specific to
15997 the Darwin support. Higher values produce more verbose output.
15999 @item show debug darwin
16000 @kindex show debug darwin
16001 Show the current state of Darwin messages.
16003 @item set debug mach-o @var{num}
16004 @kindex set debug mach-o
16005 When set to a non zero value, enables debugging messages while
16006 @value{GDBN} is reading Darwin object files. (@dfn{Mach-O} is the
16007 file format used on Darwin for object and executable files.) Higher
16008 values produce more verbose output. This is a command to diagnose
16009 problems internal to @value{GDBN} and should not be needed in normal
16012 @item show debug mach-o
16013 @kindex show debug mach-o
16014 Show the current state of Mach-O file messages.
16016 @item set mach-exceptions on
16017 @itemx set mach-exceptions off
16018 @kindex set mach-exceptions
16019 On Darwin, faults are first reported as a Mach exception and are then
16020 mapped to a Posix signal. Use this command to turn on trapping of
16021 Mach exceptions in the inferior. This might be sometimes useful to
16022 better understand the cause of a fault. The default is off.
16024 @item show mach-exceptions
16025 @kindex show mach-exceptions
16026 Show the current state of exceptions trapping.
16031 @section Embedded Operating Systems
16033 This section describes configurations involving the debugging of
16034 embedded operating systems that are available for several different
16038 * VxWorks:: Using @value{GDBN} with VxWorks
16041 @value{GDBN} includes the ability to debug programs running on
16042 various real-time operating systems.
16045 @subsection Using @value{GDBN} with VxWorks
16051 @kindex target vxworks
16052 @item target vxworks @var{machinename}
16053 A VxWorks system, attached via TCP/IP. The argument @var{machinename}
16054 is the target system's machine name or IP address.
16058 On VxWorks, @code{load} links @var{filename} dynamically on the
16059 current target system as well as adding its symbols in @value{GDBN}.
16061 @value{GDBN} enables developers to spawn and debug tasks running on networked
16062 VxWorks targets from a Unix host. Already-running tasks spawned from
16063 the VxWorks shell can also be debugged. @value{GDBN} uses code that runs on
16064 both the Unix host and on the VxWorks target. The program
16065 @code{@value{GDBP}} is installed and executed on the Unix host. (It may be
16066 installed with the name @code{vxgdb}, to distinguish it from a
16067 @value{GDBN} for debugging programs on the host itself.)
16070 @item VxWorks-timeout @var{args}
16071 @kindex vxworks-timeout
16072 All VxWorks-based targets now support the option @code{vxworks-timeout}.
16073 This option is set by the user, and @var{args} represents the number of
16074 seconds @value{GDBN} waits for responses to rpc's. You might use this if
16075 your VxWorks target is a slow software simulator or is on the far side
16076 of a thin network line.
16079 The following information on connecting to VxWorks was current when
16080 this manual was produced; newer releases of VxWorks may use revised
16083 @findex INCLUDE_RDB
16084 To use @value{GDBN} with VxWorks, you must rebuild your VxWorks kernel
16085 to include the remote debugging interface routines in the VxWorks
16086 library @file{rdb.a}. To do this, define @code{INCLUDE_RDB} in the
16087 VxWorks configuration file @file{configAll.h} and rebuild your VxWorks
16088 kernel. The resulting kernel contains @file{rdb.a}, and spawns the
16089 source debugging task @code{tRdbTask} when VxWorks is booted. For more
16090 information on configuring and remaking VxWorks, see the manufacturer's
16092 @c VxWorks, see the @cite{VxWorks Programmer's Guide}.
16094 Once you have included @file{rdb.a} in your VxWorks system image and set
16095 your Unix execution search path to find @value{GDBN}, you are ready to
16096 run @value{GDBN}. From your Unix host, run @code{@value{GDBP}} (or
16097 @code{vxgdb}, depending on your installation).
16099 @value{GDBN} comes up showing the prompt:
16106 * VxWorks Connection:: Connecting to VxWorks
16107 * VxWorks Download:: VxWorks download
16108 * VxWorks Attach:: Running tasks
16111 @node VxWorks Connection
16112 @subsubsection Connecting to VxWorks
16114 The @value{GDBN} command @code{target} lets you connect to a VxWorks target on the
16115 network. To connect to a target whose host name is ``@code{tt}'', type:
16118 (vxgdb) target vxworks tt
16122 @value{GDBN} displays messages like these:
16125 Attaching remote machine across net...
16130 @value{GDBN} then attempts to read the symbol tables of any object modules
16131 loaded into the VxWorks target since it was last booted. @value{GDBN} locates
16132 these files by searching the directories listed in the command search
16133 path (@pxref{Environment, ,Your Program's Environment}); if it fails
16134 to find an object file, it displays a message such as:
16137 prog.o: No such file or directory.
16140 When this happens, add the appropriate directory to the search path with
16141 the @value{GDBN} command @code{path}, and execute the @code{target}
16144 @node VxWorks Download
16145 @subsubsection VxWorks Download
16147 @cindex download to VxWorks
16148 If you have connected to the VxWorks target and you want to debug an
16149 object that has not yet been loaded, you can use the @value{GDBN}
16150 @code{load} command to download a file from Unix to VxWorks
16151 incrementally. The object file given as an argument to the @code{load}
16152 command is actually opened twice: first by the VxWorks target in order
16153 to download the code, then by @value{GDBN} in order to read the symbol
16154 table. This can lead to problems if the current working directories on
16155 the two systems differ. If both systems have NFS mounted the same
16156 filesystems, you can avoid these problems by using absolute paths.
16157 Otherwise, it is simplest to set the working directory on both systems
16158 to the directory in which the object file resides, and then to reference
16159 the file by its name, without any path. For instance, a program
16160 @file{prog.o} may reside in @file{@var{vxpath}/vw/demo/rdb} in VxWorks
16161 and in @file{@var{hostpath}/vw/demo/rdb} on the host. To load this
16162 program, type this on VxWorks:
16165 -> cd "@var{vxpath}/vw/demo/rdb"
16169 Then, in @value{GDBN}, type:
16172 (vxgdb) cd @var{hostpath}/vw/demo/rdb
16173 (vxgdb) load prog.o
16176 @value{GDBN} displays a response similar to this:
16179 Reading symbol data from wherever/vw/demo/rdb/prog.o... done.
16182 You can also use the @code{load} command to reload an object module
16183 after editing and recompiling the corresponding source file. Note that
16184 this makes @value{GDBN} delete all currently-defined breakpoints,
16185 auto-displays, and convenience variables, and to clear the value
16186 history. (This is necessary in order to preserve the integrity of
16187 debugger's data structures that reference the target system's symbol
16190 @node VxWorks Attach
16191 @subsubsection Running Tasks
16193 @cindex running VxWorks tasks
16194 You can also attach to an existing task using the @code{attach} command as
16198 (vxgdb) attach @var{task}
16202 where @var{task} is the VxWorks hexadecimal task ID. The task can be running
16203 or suspended when you attach to it. Running tasks are suspended at
16204 the time of attachment.
16206 @node Embedded Processors
16207 @section Embedded Processors
16209 This section goes into details specific to particular embedded
16212 @cindex send command to simulator
16213 Whenever a specific embedded processor has a simulator, @value{GDBN}
16214 allows to send an arbitrary command to the simulator.
16217 @item sim @var{command}
16218 @kindex sim@r{, a command}
16219 Send an arbitrary @var{command} string to the simulator. Consult the
16220 documentation for the specific simulator in use for information about
16221 acceptable commands.
16227 * M32R/D:: Renesas M32R/D
16228 * M68K:: Motorola M68K
16229 * MIPS Embedded:: MIPS Embedded
16230 * OpenRISC 1000:: OpenRisc 1000
16231 * PA:: HP PA Embedded
16232 * PowerPC Embedded:: PowerPC Embedded
16233 * Sparclet:: Tsqware Sparclet
16234 * Sparclite:: Fujitsu Sparclite
16235 * Z8000:: Zilog Z8000
16238 * Super-H:: Renesas Super-H
16247 @item target rdi @var{dev}
16248 ARM Angel monitor, via RDI library interface to ADP protocol. You may
16249 use this target to communicate with both boards running the Angel
16250 monitor, or with the EmbeddedICE JTAG debug device.
16253 @item target rdp @var{dev}
16258 @value{GDBN} provides the following ARM-specific commands:
16261 @item set arm disassembler
16263 This commands selects from a list of disassembly styles. The
16264 @code{"std"} style is the standard style.
16266 @item show arm disassembler
16268 Show the current disassembly style.
16270 @item set arm apcs32
16271 @cindex ARM 32-bit mode
16272 This command toggles ARM operation mode between 32-bit and 26-bit.
16274 @item show arm apcs32
16275 Display the current usage of the ARM 32-bit mode.
16277 @item set arm fpu @var{fputype}
16278 This command sets the ARM floating-point unit (FPU) type. The
16279 argument @var{fputype} can be one of these:
16283 Determine the FPU type by querying the OS ABI.
16285 Software FPU, with mixed-endian doubles on little-endian ARM
16288 GCC-compiled FPA co-processor.
16290 Software FPU with pure-endian doubles.
16296 Show the current type of the FPU.
16299 This command forces @value{GDBN} to use the specified ABI.
16302 Show the currently used ABI.
16304 @item set arm fallback-mode (arm|thumb|auto)
16305 @value{GDBN} uses the symbol table, when available, to determine
16306 whether instructions are ARM or Thumb. This command controls
16307 @value{GDBN}'s default behavior when the symbol table is not
16308 available. The default is @samp{auto}, which causes @value{GDBN} to
16309 use the current execution mode (from the @code{T} bit in the @code{CPSR}
16312 @item show arm fallback-mode
16313 Show the current fallback instruction mode.
16315 @item set arm force-mode (arm|thumb|auto)
16316 This command overrides use of the symbol table to determine whether
16317 instructions are ARM or Thumb. The default is @samp{auto}, which
16318 causes @value{GDBN} to use the symbol table and then the setting
16319 of @samp{set arm fallback-mode}.
16321 @item show arm force-mode
16322 Show the current forced instruction mode.
16324 @item set debug arm
16325 Toggle whether to display ARM-specific debugging messages from the ARM
16326 target support subsystem.
16328 @item show debug arm
16329 Show whether ARM-specific debugging messages are enabled.
16332 The following commands are available when an ARM target is debugged
16333 using the RDI interface:
16336 @item rdilogfile @r{[}@var{file}@r{]}
16338 @cindex ADP (Angel Debugger Protocol) logging
16339 Set the filename for the ADP (Angel Debugger Protocol) packet log.
16340 With an argument, sets the log file to the specified @var{file}. With
16341 no argument, show the current log file name. The default log file is
16344 @item rdilogenable @r{[}@var{arg}@r{]}
16345 @kindex rdilogenable
16346 Control logging of ADP packets. With an argument of 1 or @code{"yes"}
16347 enables logging, with an argument 0 or @code{"no"} disables it. With
16348 no arguments displays the current setting. When logging is enabled,
16349 ADP packets exchanged between @value{GDBN} and the RDI target device
16350 are logged to a file.
16352 @item set rdiromatzero
16353 @kindex set rdiromatzero
16354 @cindex ROM at zero address, RDI
16355 Tell @value{GDBN} whether the target has ROM at address 0. If on,
16356 vector catching is disabled, so that zero address can be used. If off
16357 (the default), vector catching is enabled. For this command to take
16358 effect, it needs to be invoked prior to the @code{target rdi} command.
16360 @item show rdiromatzero
16361 @kindex show rdiromatzero
16362 Show the current setting of ROM at zero address.
16364 @item set rdiheartbeat
16365 @kindex set rdiheartbeat
16366 @cindex RDI heartbeat
16367 Enable or disable RDI heartbeat packets. It is not recommended to
16368 turn on this option, since it confuses ARM and EPI JTAG interface, as
16369 well as the Angel monitor.
16371 @item show rdiheartbeat
16372 @kindex show rdiheartbeat
16373 Show the setting of RDI heartbeat packets.
16378 @subsection Renesas M32R/D and M32R/SDI
16381 @kindex target m32r
16382 @item target m32r @var{dev}
16383 Renesas M32R/D ROM monitor.
16385 @kindex target m32rsdi
16386 @item target m32rsdi @var{dev}
16387 Renesas M32R SDI server, connected via parallel port to the board.
16390 The following @value{GDBN} commands are specific to the M32R monitor:
16393 @item set download-path @var{path}
16394 @kindex set download-path
16395 @cindex find downloadable @sc{srec} files (M32R)
16396 Set the default path for finding downloadable @sc{srec} files.
16398 @item show download-path
16399 @kindex show download-path
16400 Show the default path for downloadable @sc{srec} files.
16402 @item set board-address @var{addr}
16403 @kindex set board-address
16404 @cindex M32-EVA target board address
16405 Set the IP address for the M32R-EVA target board.
16407 @item show board-address
16408 @kindex show board-address
16409 Show the current IP address of the target board.
16411 @item set server-address @var{addr}
16412 @kindex set server-address
16413 @cindex download server address (M32R)
16414 Set the IP address for the download server, which is the @value{GDBN}'s
16417 @item show server-address
16418 @kindex show server-address
16419 Display the IP address of the download server.
16421 @item upload @r{[}@var{file}@r{]}
16422 @kindex upload@r{, M32R}
16423 Upload the specified @sc{srec} @var{file} via the monitor's Ethernet
16424 upload capability. If no @var{file} argument is given, the current
16425 executable file is uploaded.
16427 @item tload @r{[}@var{file}@r{]}
16428 @kindex tload@r{, M32R}
16429 Test the @code{upload} command.
16432 The following commands are available for M32R/SDI:
16437 @cindex reset SDI connection, M32R
16438 This command resets the SDI connection.
16442 This command shows the SDI connection status.
16445 @kindex debug_chaos
16446 @cindex M32R/Chaos debugging
16447 Instructs the remote that M32R/Chaos debugging is to be used.
16449 @item use_debug_dma
16450 @kindex use_debug_dma
16451 Instructs the remote to use the DEBUG_DMA method of accessing memory.
16454 @kindex use_mon_code
16455 Instructs the remote to use the MON_CODE method of accessing memory.
16458 @kindex use_ib_break
16459 Instructs the remote to set breakpoints by IB break.
16461 @item use_dbt_break
16462 @kindex use_dbt_break
16463 Instructs the remote to set breakpoints by DBT.
16469 The Motorola m68k configuration includes ColdFire support, and a
16470 target command for the following ROM monitor.
16474 @kindex target dbug
16475 @item target dbug @var{dev}
16476 dBUG ROM monitor for Motorola ColdFire.
16480 @node MIPS Embedded
16481 @subsection MIPS Embedded
16483 @cindex MIPS boards
16484 @value{GDBN} can use the MIPS remote debugging protocol to talk to a
16485 MIPS board attached to a serial line. This is available when
16486 you configure @value{GDBN} with @samp{--target=mips-idt-ecoff}.
16489 Use these @value{GDBN} commands to specify the connection to your target board:
16492 @item target mips @var{port}
16493 @kindex target mips @var{port}
16494 To run a program on the board, start up @code{@value{GDBP}} with the
16495 name of your program as the argument. To connect to the board, use the
16496 command @samp{target mips @var{port}}, where @var{port} is the name of
16497 the serial port connected to the board. If the program has not already
16498 been downloaded to the board, you may use the @code{load} command to
16499 download it. You can then use all the usual @value{GDBN} commands.
16501 For example, this sequence connects to the target board through a serial
16502 port, and loads and runs a program called @var{prog} through the
16506 host$ @value{GDBP} @var{prog}
16507 @value{GDBN} is free software and @dots{}
16508 (@value{GDBP}) target mips /dev/ttyb
16509 (@value{GDBP}) load @var{prog}
16513 @item target mips @var{hostname}:@var{portnumber}
16514 On some @value{GDBN} host configurations, you can specify a TCP
16515 connection (for instance, to a serial line managed by a terminal
16516 concentrator) instead of a serial port, using the syntax
16517 @samp{@var{hostname}:@var{portnumber}}.
16519 @item target pmon @var{port}
16520 @kindex target pmon @var{port}
16523 @item target ddb @var{port}
16524 @kindex target ddb @var{port}
16525 NEC's DDB variant of PMON for Vr4300.
16527 @item target lsi @var{port}
16528 @kindex target lsi @var{port}
16529 LSI variant of PMON.
16531 @kindex target r3900
16532 @item target r3900 @var{dev}
16533 Densan DVE-R3900 ROM monitor for Toshiba R3900 Mips.
16535 @kindex target array
16536 @item target array @var{dev}
16537 Array Tech LSI33K RAID controller board.
16543 @value{GDBN} also supports these special commands for MIPS targets:
16546 @item set mipsfpu double
16547 @itemx set mipsfpu single
16548 @itemx set mipsfpu none
16549 @itemx set mipsfpu auto
16550 @itemx show mipsfpu
16551 @kindex set mipsfpu
16552 @kindex show mipsfpu
16553 @cindex MIPS remote floating point
16554 @cindex floating point, MIPS remote
16555 If your target board does not support the MIPS floating point
16556 coprocessor, you should use the command @samp{set mipsfpu none} (if you
16557 need this, you may wish to put the command in your @value{GDBN} init
16558 file). This tells @value{GDBN} how to find the return value of
16559 functions which return floating point values. It also allows
16560 @value{GDBN} to avoid saving the floating point registers when calling
16561 functions on the board. If you are using a floating point coprocessor
16562 with only single precision floating point support, as on the @sc{r4650}
16563 processor, use the command @samp{set mipsfpu single}. The default
16564 double precision floating point coprocessor may be selected using
16565 @samp{set mipsfpu double}.
16567 In previous versions the only choices were double precision or no
16568 floating point, so @samp{set mipsfpu on} will select double precision
16569 and @samp{set mipsfpu off} will select no floating point.
16571 As usual, you can inquire about the @code{mipsfpu} variable with
16572 @samp{show mipsfpu}.
16574 @item set timeout @var{seconds}
16575 @itemx set retransmit-timeout @var{seconds}
16576 @itemx show timeout
16577 @itemx show retransmit-timeout
16578 @cindex @code{timeout}, MIPS protocol
16579 @cindex @code{retransmit-timeout}, MIPS protocol
16580 @kindex set timeout
16581 @kindex show timeout
16582 @kindex set retransmit-timeout
16583 @kindex show retransmit-timeout
16584 You can control the timeout used while waiting for a packet, in the MIPS
16585 remote protocol, with the @code{set timeout @var{seconds}} command. The
16586 default is 5 seconds. Similarly, you can control the timeout used while
16587 waiting for an acknowledgment of a packet with the @code{set
16588 retransmit-timeout @var{seconds}} command. The default is 3 seconds.
16589 You can inspect both values with @code{show timeout} and @code{show
16590 retransmit-timeout}. (These commands are @emph{only} available when
16591 @value{GDBN} is configured for @samp{--target=mips-idt-ecoff}.)
16593 The timeout set by @code{set timeout} does not apply when @value{GDBN}
16594 is waiting for your program to stop. In that case, @value{GDBN} waits
16595 forever because it has no way of knowing how long the program is going
16596 to run before stopping.
16598 @item set syn-garbage-limit @var{num}
16599 @kindex set syn-garbage-limit@r{, MIPS remote}
16600 @cindex synchronize with remote MIPS target
16601 Limit the maximum number of characters @value{GDBN} should ignore when
16602 it tries to synchronize with the remote target. The default is 10
16603 characters. Setting the limit to -1 means there's no limit.
16605 @item show syn-garbage-limit
16606 @kindex show syn-garbage-limit@r{, MIPS remote}
16607 Show the current limit on the number of characters to ignore when
16608 trying to synchronize with the remote system.
16610 @item set monitor-prompt @var{prompt}
16611 @kindex set monitor-prompt@r{, MIPS remote}
16612 @cindex remote monitor prompt
16613 Tell @value{GDBN} to expect the specified @var{prompt} string from the
16614 remote monitor. The default depends on the target:
16624 @item show monitor-prompt
16625 @kindex show monitor-prompt@r{, MIPS remote}
16626 Show the current strings @value{GDBN} expects as the prompt from the
16629 @item set monitor-warnings
16630 @kindex set monitor-warnings@r{, MIPS remote}
16631 Enable or disable monitor warnings about hardware breakpoints. This
16632 has effect only for the @code{lsi} target. When on, @value{GDBN} will
16633 display warning messages whose codes are returned by the @code{lsi}
16634 PMON monitor for breakpoint commands.
16636 @item show monitor-warnings
16637 @kindex show monitor-warnings@r{, MIPS remote}
16638 Show the current setting of printing monitor warnings.
16640 @item pmon @var{command}
16641 @kindex pmon@r{, MIPS remote}
16642 @cindex send PMON command
16643 This command allows sending an arbitrary @var{command} string to the
16644 monitor. The monitor must be in debug mode for this to work.
16647 @node OpenRISC 1000
16648 @subsection OpenRISC 1000
16649 @cindex OpenRISC 1000
16651 @cindex or1k boards
16652 See OR1k Architecture document (@uref{www.opencores.org}) for more information
16653 about platform and commands.
16657 @kindex target jtag
16658 @item target jtag jtag://@var{host}:@var{port}
16660 Connects to remote JTAG server.
16661 JTAG remote server can be either an or1ksim or JTAG server,
16662 connected via parallel port to the board.
16664 Example: @code{target jtag jtag://localhost:9999}
16667 @item or1ksim @var{command}
16668 If connected to @code{or1ksim} OpenRISC 1000 Architectural
16669 Simulator, proprietary commands can be executed.
16671 @kindex info or1k spr
16672 @item info or1k spr
16673 Displays spr groups.
16675 @item info or1k spr @var{group}
16676 @itemx info or1k spr @var{groupno}
16677 Displays register names in selected group.
16679 @item info or1k spr @var{group} @var{register}
16680 @itemx info or1k spr @var{register}
16681 @itemx info or1k spr @var{groupno} @var{registerno}
16682 @itemx info or1k spr @var{registerno}
16683 Shows information about specified spr register.
16686 @item spr @var{group} @var{register} @var{value}
16687 @itemx spr @var{register @var{value}}
16688 @itemx spr @var{groupno} @var{registerno @var{value}}
16689 @itemx spr @var{registerno @var{value}}
16690 Writes @var{value} to specified spr register.
16693 Some implementations of OpenRISC 1000 Architecture also have hardware trace.
16694 It is very similar to @value{GDBN} trace, except it does not interfere with normal
16695 program execution and is thus much faster. Hardware breakpoints/watchpoint
16696 triggers can be set using:
16699 Load effective address/data
16701 Store effective address/data
16703 Access effective address ($SEA or $LEA) or data ($SDATA/$LDATA)
16708 When triggered, it can capture low level data, like: @code{PC}, @code{LSEA},
16709 @code{LDATA}, @code{SDATA}, @code{READSPR}, @code{WRITESPR}, @code{INSTR}.
16711 @code{htrace} commands:
16712 @cindex OpenRISC 1000 htrace
16715 @item hwatch @var{conditional}
16716 Set hardware watchpoint on combination of Load/Store Effective Address(es)
16717 or Data. For example:
16719 @code{hwatch ($LEA == my_var) && ($LDATA < 50) || ($SEA == my_var) && ($SDATA >= 50)}
16721 @code{hwatch ($LEA == my_var) && ($LDATA < 50) || ($SEA == my_var) && ($SDATA >= 50)}
16725 Display information about current HW trace configuration.
16727 @item htrace trigger @var{conditional}
16728 Set starting criteria for HW trace.
16730 @item htrace qualifier @var{conditional}
16731 Set acquisition qualifier for HW trace.
16733 @item htrace stop @var{conditional}
16734 Set HW trace stopping criteria.
16736 @item htrace record [@var{data}]*
16737 Selects the data to be recorded, when qualifier is met and HW trace was
16740 @item htrace enable
16741 @itemx htrace disable
16742 Enables/disables the HW trace.
16744 @item htrace rewind [@var{filename}]
16745 Clears currently recorded trace data.
16747 If filename is specified, new trace file is made and any newly collected data
16748 will be written there.
16750 @item htrace print [@var{start} [@var{len}]]
16751 Prints trace buffer, using current record configuration.
16753 @item htrace mode continuous
16754 Set continuous trace mode.
16756 @item htrace mode suspend
16757 Set suspend trace mode.
16761 @node PowerPC Embedded
16762 @subsection PowerPC Embedded
16764 @value{GDBN} provides the following PowerPC-specific commands:
16767 @kindex set powerpc
16768 @item set powerpc soft-float
16769 @itemx show powerpc soft-float
16770 Force @value{GDBN} to use (or not use) a software floating point calling
16771 convention. By default, @value{GDBN} selects the calling convention based
16772 on the selected architecture and the provided executable file.
16774 @item set powerpc vector-abi
16775 @itemx show powerpc vector-abi
16776 Force @value{GDBN} to use the specified calling convention for vector
16777 arguments and return values. The valid options are @samp{auto};
16778 @samp{generic}, to avoid vector registers even if they are present;
16779 @samp{altivec}, to use AltiVec registers; and @samp{spe} to use SPE
16780 registers. By default, @value{GDBN} selects the calling convention
16781 based on the selected architecture and the provided executable file.
16783 @kindex target dink32
16784 @item target dink32 @var{dev}
16785 DINK32 ROM monitor.
16787 @kindex target ppcbug
16788 @item target ppcbug @var{dev}
16789 @kindex target ppcbug1
16790 @item target ppcbug1 @var{dev}
16791 PPCBUG ROM monitor for PowerPC.
16794 @item target sds @var{dev}
16795 SDS monitor, running on a PowerPC board (such as Motorola's ADS).
16798 @cindex SDS protocol
16799 The following commands specific to the SDS protocol are supported
16803 @item set sdstimeout @var{nsec}
16804 @kindex set sdstimeout
16805 Set the timeout for SDS protocol reads to be @var{nsec} seconds. The
16806 default is 2 seconds.
16808 @item show sdstimeout
16809 @kindex show sdstimeout
16810 Show the current value of the SDS timeout.
16812 @item sds @var{command}
16813 @kindex sds@r{, a command}
16814 Send the specified @var{command} string to the SDS monitor.
16819 @subsection HP PA Embedded
16823 @kindex target op50n
16824 @item target op50n @var{dev}
16825 OP50N monitor, running on an OKI HPPA board.
16827 @kindex target w89k
16828 @item target w89k @var{dev}
16829 W89K monitor, running on a Winbond HPPA board.
16834 @subsection Tsqware Sparclet
16838 @value{GDBN} enables developers to debug tasks running on
16839 Sparclet targets from a Unix host.
16840 @value{GDBN} uses code that runs on
16841 both the Unix host and on the Sparclet target. The program
16842 @code{@value{GDBP}} is installed and executed on the Unix host.
16845 @item remotetimeout @var{args}
16846 @kindex remotetimeout
16847 @value{GDBN} supports the option @code{remotetimeout}.
16848 This option is set by the user, and @var{args} represents the number of
16849 seconds @value{GDBN} waits for responses.
16852 @cindex compiling, on Sparclet
16853 When compiling for debugging, include the options @samp{-g} to get debug
16854 information and @samp{-Ttext} to relocate the program to where you wish to
16855 load it on the target. You may also want to add the options @samp{-n} or
16856 @samp{-N} in order to reduce the size of the sections. Example:
16859 sparclet-aout-gcc prog.c -Ttext 0x12010000 -g -o prog -N
16862 You can use @code{objdump} to verify that the addresses are what you intended:
16865 sparclet-aout-objdump --headers --syms prog
16868 @cindex running, on Sparclet
16870 your Unix execution search path to find @value{GDBN}, you are ready to
16871 run @value{GDBN}. From your Unix host, run @code{@value{GDBP}}
16872 (or @code{sparclet-aout-gdb}, depending on your installation).
16874 @value{GDBN} comes up showing the prompt:
16881 * Sparclet File:: Setting the file to debug
16882 * Sparclet Connection:: Connecting to Sparclet
16883 * Sparclet Download:: Sparclet download
16884 * Sparclet Execution:: Running and debugging
16887 @node Sparclet File
16888 @subsubsection Setting File to Debug
16890 The @value{GDBN} command @code{file} lets you choose with program to debug.
16893 (gdbslet) file prog
16897 @value{GDBN} then attempts to read the symbol table of @file{prog}.
16898 @value{GDBN} locates
16899 the file by searching the directories listed in the command search
16901 If the file was compiled with debug information (option @samp{-g}), source
16902 files will be searched as well.
16903 @value{GDBN} locates
16904 the source files by searching the directories listed in the directory search
16905 path (@pxref{Environment, ,Your Program's Environment}).
16907 to find a file, it displays a message such as:
16910 prog: No such file or directory.
16913 When this happens, add the appropriate directories to the search paths with
16914 the @value{GDBN} commands @code{path} and @code{dir}, and execute the
16915 @code{target} command again.
16917 @node Sparclet Connection
16918 @subsubsection Connecting to Sparclet
16920 The @value{GDBN} command @code{target} lets you connect to a Sparclet target.
16921 To connect to a target on serial port ``@code{ttya}'', type:
16924 (gdbslet) target sparclet /dev/ttya
16925 Remote target sparclet connected to /dev/ttya
16926 main () at ../prog.c:3
16930 @value{GDBN} displays messages like these:
16936 @node Sparclet Download
16937 @subsubsection Sparclet Download
16939 @cindex download to Sparclet
16940 Once connected to the Sparclet target,
16941 you can use the @value{GDBN}
16942 @code{load} command to download the file from the host to the target.
16943 The file name and load offset should be given as arguments to the @code{load}
16945 Since the file format is aout, the program must be loaded to the starting
16946 address. You can use @code{objdump} to find out what this value is. The load
16947 offset is an offset which is added to the VMA (virtual memory address)
16948 of each of the file's sections.
16949 For instance, if the program
16950 @file{prog} was linked to text address 0x1201000, with data at 0x12010160
16951 and bss at 0x12010170, in @value{GDBN}, type:
16954 (gdbslet) load prog 0x12010000
16955 Loading section .text, size 0xdb0 vma 0x12010000
16958 If the code is loaded at a different address then what the program was linked
16959 to, you may need to use the @code{section} and @code{add-symbol-file} commands
16960 to tell @value{GDBN} where to map the symbol table.
16962 @node Sparclet Execution
16963 @subsubsection Running and Debugging
16965 @cindex running and debugging Sparclet programs
16966 You can now begin debugging the task using @value{GDBN}'s execution control
16967 commands, @code{b}, @code{step}, @code{run}, etc. See the @value{GDBN}
16968 manual for the list of commands.
16972 Breakpoint 1 at 0x12010000: file prog.c, line 3.
16974 Starting program: prog
16975 Breakpoint 1, main (argc=1, argv=0xeffff21c) at prog.c:3
16976 3 char *symarg = 0;
16978 4 char *execarg = "hello!";
16983 @subsection Fujitsu Sparclite
16987 @kindex target sparclite
16988 @item target sparclite @var{dev}
16989 Fujitsu sparclite boards, used only for the purpose of loading.
16990 You must use an additional command to debug the program.
16991 For example: target remote @var{dev} using @value{GDBN} standard
16997 @subsection Zilog Z8000
17000 @cindex simulator, Z8000
17001 @cindex Zilog Z8000 simulator
17003 When configured for debugging Zilog Z8000 targets, @value{GDBN} includes
17006 For the Z8000 family, @samp{target sim} simulates either the Z8002 (the
17007 unsegmented variant of the Z8000 architecture) or the Z8001 (the
17008 segmented variant). The simulator recognizes which architecture is
17009 appropriate by inspecting the object code.
17012 @item target sim @var{args}
17014 @kindex target sim@r{, with Z8000}
17015 Debug programs on a simulated CPU. If the simulator supports setup
17016 options, specify them via @var{args}.
17020 After specifying this target, you can debug programs for the simulated
17021 CPU in the same style as programs for your host computer; use the
17022 @code{file} command to load a new program image, the @code{run} command
17023 to run your program, and so on.
17025 As well as making available all the usual machine registers
17026 (@pxref{Registers, ,Registers}), the Z8000 simulator provides three
17027 additional items of information as specially named registers:
17032 Counts clock-ticks in the simulator.
17035 Counts instructions run in the simulator.
17038 Execution time in 60ths of a second.
17042 You can refer to these values in @value{GDBN} expressions with the usual
17043 conventions; for example, @w{@samp{b fputc if $cycles>5000}} sets a
17044 conditional breakpoint that suspends only after at least 5000
17045 simulated clock ticks.
17048 @subsection Atmel AVR
17051 When configured for debugging the Atmel AVR, @value{GDBN} supports the
17052 following AVR-specific commands:
17055 @item info io_registers
17056 @kindex info io_registers@r{, AVR}
17057 @cindex I/O registers (Atmel AVR)
17058 This command displays information about the AVR I/O registers. For
17059 each register, @value{GDBN} prints its number and value.
17066 When configured for debugging CRIS, @value{GDBN} provides the
17067 following CRIS-specific commands:
17070 @item set cris-version @var{ver}
17071 @cindex CRIS version
17072 Set the current CRIS version to @var{ver}, either @samp{10} or @samp{32}.
17073 The CRIS version affects register names and sizes. This command is useful in
17074 case autodetection of the CRIS version fails.
17076 @item show cris-version
17077 Show the current CRIS version.
17079 @item set cris-dwarf2-cfi
17080 @cindex DWARF-2 CFI and CRIS
17081 Set the usage of DWARF-2 CFI for CRIS debugging. The default is @samp{on}.
17082 Change to @samp{off} when using @code{gcc-cris} whose version is below
17085 @item show cris-dwarf2-cfi
17086 Show the current state of using DWARF-2 CFI.
17088 @item set cris-mode @var{mode}
17090 Set the current CRIS mode to @var{mode}. It should only be changed when
17091 debugging in guru mode, in which case it should be set to
17092 @samp{guru} (the default is @samp{normal}).
17094 @item show cris-mode
17095 Show the current CRIS mode.
17099 @subsection Renesas Super-H
17102 For the Renesas Super-H processor, @value{GDBN} provides these
17107 @kindex regs@r{, Super-H}
17108 Show the values of all Super-H registers.
17110 @item set sh calling-convention @var{convention}
17111 @kindex set sh calling-convention
17112 Set the calling-convention used when calling functions from @value{GDBN}.
17113 Allowed values are @samp{gcc}, which is the default setting, and @samp{renesas}.
17114 With the @samp{gcc} setting, functions are called using the @value{NGCC} calling
17115 convention. If the DWARF-2 information of the called function specifies
17116 that the function follows the Renesas calling convention, the function
17117 is called using the Renesas calling convention. If the calling convention
17118 is set to @samp{renesas}, the Renesas calling convention is always used,
17119 regardless of the DWARF-2 information. This can be used to override the
17120 default of @samp{gcc} if debug information is missing, or the compiler
17121 does not emit the DWARF-2 calling convention entry for a function.
17123 @item show sh calling-convention
17124 @kindex show sh calling-convention
17125 Show the current calling convention setting.
17130 @node Architectures
17131 @section Architectures
17133 This section describes characteristics of architectures that affect
17134 all uses of @value{GDBN} with the architecture, both native and cross.
17141 * HPPA:: HP PA architecture
17142 * SPU:: Cell Broadband Engine SPU architecture
17147 @subsection x86 Architecture-specific Issues
17150 @item set struct-convention @var{mode}
17151 @kindex set struct-convention
17152 @cindex struct return convention
17153 @cindex struct/union returned in registers
17154 Set the convention used by the inferior to return @code{struct}s and
17155 @code{union}s from functions to @var{mode}. Possible values of
17156 @var{mode} are @code{"pcc"}, @code{"reg"}, and @code{"default"} (the
17157 default). @code{"default"} or @code{"pcc"} means that @code{struct}s
17158 are returned on the stack, while @code{"reg"} means that a
17159 @code{struct} or a @code{union} whose size is 1, 2, 4, or 8 bytes will
17160 be returned in a register.
17162 @item show struct-convention
17163 @kindex show struct-convention
17164 Show the current setting of the convention to return @code{struct}s
17173 @kindex set rstack_high_address
17174 @cindex AMD 29K register stack
17175 @cindex register stack, AMD29K
17176 @item set rstack_high_address @var{address}
17177 On AMD 29000 family processors, registers are saved in a separate
17178 @dfn{register stack}. There is no way for @value{GDBN} to determine the
17179 extent of this stack. Normally, @value{GDBN} just assumes that the
17180 stack is ``large enough''. This may result in @value{GDBN} referencing
17181 memory locations that do not exist. If necessary, you can get around
17182 this problem by specifying the ending address of the register stack with
17183 the @code{set rstack_high_address} command. The argument should be an
17184 address, which you probably want to precede with @samp{0x} to specify in
17187 @kindex show rstack_high_address
17188 @item show rstack_high_address
17189 Display the current limit of the register stack, on AMD 29000 family
17197 See the following section.
17202 @cindex stack on Alpha
17203 @cindex stack on MIPS
17204 @cindex Alpha stack
17206 Alpha- and MIPS-based computers use an unusual stack frame, which
17207 sometimes requires @value{GDBN} to search backward in the object code to
17208 find the beginning of a function.
17210 @cindex response time, MIPS debugging
17211 To improve response time (especially for embedded applications, where
17212 @value{GDBN} may be restricted to a slow serial line for this search)
17213 you may want to limit the size of this search, using one of these
17217 @cindex @code{heuristic-fence-post} (Alpha, MIPS)
17218 @item set heuristic-fence-post @var{limit}
17219 Restrict @value{GDBN} to examining at most @var{limit} bytes in its
17220 search for the beginning of a function. A value of @var{0} (the
17221 default) means there is no limit. However, except for @var{0}, the
17222 larger the limit the more bytes @code{heuristic-fence-post} must search
17223 and therefore the longer it takes to run. You should only need to use
17224 this command when debugging a stripped executable.
17226 @item show heuristic-fence-post
17227 Display the current limit.
17231 These commands are available @emph{only} when @value{GDBN} is configured
17232 for debugging programs on Alpha or MIPS processors.
17234 Several MIPS-specific commands are available when debugging MIPS
17238 @item set mips abi @var{arg}
17239 @kindex set mips abi
17240 @cindex set ABI for MIPS
17241 Tell @value{GDBN} which MIPS ABI is used by the inferior. Possible
17242 values of @var{arg} are:
17246 The default ABI associated with the current binary (this is the
17257 @item show mips abi
17258 @kindex show mips abi
17259 Show the MIPS ABI used by @value{GDBN} to debug the inferior.
17262 @itemx show mipsfpu
17263 @xref{MIPS Embedded, set mipsfpu}.
17265 @item set mips mask-address @var{arg}
17266 @kindex set mips mask-address
17267 @cindex MIPS addresses, masking
17268 This command determines whether the most-significant 32 bits of 64-bit
17269 MIPS addresses are masked off. The argument @var{arg} can be
17270 @samp{on}, @samp{off}, or @samp{auto}. The latter is the default
17271 setting, which lets @value{GDBN} determine the correct value.
17273 @item show mips mask-address
17274 @kindex show mips mask-address
17275 Show whether the upper 32 bits of MIPS addresses are masked off or
17278 @item set remote-mips64-transfers-32bit-regs
17279 @kindex set remote-mips64-transfers-32bit-regs
17280 This command controls compatibility with 64-bit MIPS targets that
17281 transfer data in 32-bit quantities. If you have an old MIPS 64 target
17282 that transfers 32 bits for some registers, like @sc{sr} and @sc{fsr},
17283 and 64 bits for other registers, set this option to @samp{on}.
17285 @item show remote-mips64-transfers-32bit-regs
17286 @kindex show remote-mips64-transfers-32bit-regs
17287 Show the current setting of compatibility with older MIPS 64 targets.
17289 @item set debug mips
17290 @kindex set debug mips
17291 This command turns on and off debugging messages for the MIPS-specific
17292 target code in @value{GDBN}.
17294 @item show debug mips
17295 @kindex show debug mips
17296 Show the current setting of MIPS debugging messages.
17302 @cindex HPPA support
17304 When @value{GDBN} is debugging the HP PA architecture, it provides the
17305 following special commands:
17308 @item set debug hppa
17309 @kindex set debug hppa
17310 This command determines whether HPPA architecture-specific debugging
17311 messages are to be displayed.
17313 @item show debug hppa
17314 Show whether HPPA debugging messages are displayed.
17316 @item maint print unwind @var{address}
17317 @kindex maint print unwind@r{, HPPA}
17318 This command displays the contents of the unwind table entry at the
17319 given @var{address}.
17325 @subsection Cell Broadband Engine SPU architecture
17326 @cindex Cell Broadband Engine
17329 When @value{GDBN} is debugging the Cell Broadband Engine SPU architecture,
17330 it provides the following special commands:
17333 @item info spu event
17335 Display SPU event facility status. Shows current event mask
17336 and pending event status.
17338 @item info spu signal
17339 Display SPU signal notification facility status. Shows pending
17340 signal-control word and signal notification mode of both signal
17341 notification channels.
17343 @item info spu mailbox
17344 Display SPU mailbox facility status. Shows all pending entries,
17345 in order of processing, in each of the SPU Write Outbound,
17346 SPU Write Outbound Interrupt, and SPU Read Inbound mailboxes.
17349 Display MFC DMA status. Shows all pending commands in the MFC
17350 DMA queue. For each entry, opcode, tag, class IDs, effective
17351 and local store addresses and transfer size are shown.
17353 @item info spu proxydma
17354 Display MFC Proxy-DMA status. Shows all pending commands in the MFC
17355 Proxy-DMA queue. For each entry, opcode, tag, class IDs, effective
17356 and local store addresses and transfer size are shown.
17361 @subsection PowerPC
17362 @cindex PowerPC architecture
17364 When @value{GDBN} is debugging the PowerPC architecture, it provides a set of
17365 pseudo-registers to enable inspection of 128-bit wide Decimal Floating Point
17366 numbers stored in the floating point registers. These values must be stored
17367 in two consecutive registers, always starting at an even register like
17368 @code{f0} or @code{f2}.
17370 The pseudo-registers go from @code{$dl0} through @code{$dl15}, and are formed
17371 by joining the even/odd register pairs @code{f0} and @code{f1} for @code{$dl0},
17372 @code{f2} and @code{f3} for @code{$dl1} and so on.
17374 For POWER7 processors, @value{GDBN} provides a set of pseudo-registers, the 64-bit
17375 wide Extended Floating Point Registers (@samp{f32} through @samp{f63}).
17378 @node Controlling GDB
17379 @chapter Controlling @value{GDBN}
17381 You can alter the way @value{GDBN} interacts with you by using the
17382 @code{set} command. For commands controlling how @value{GDBN} displays
17383 data, see @ref{Print Settings, ,Print Settings}. Other settings are
17388 * Editing:: Command editing
17389 * Command History:: Command history
17390 * Screen Size:: Screen size
17391 * Numbers:: Numbers
17392 * ABI:: Configuring the current ABI
17393 * Messages/Warnings:: Optional warnings and messages
17394 * Debugging Output:: Optional messages about internal happenings
17402 @value{GDBN} indicates its readiness to read a command by printing a string
17403 called the @dfn{prompt}. This string is normally @samp{(@value{GDBP})}. You
17404 can change the prompt string with the @code{set prompt} command. For
17405 instance, when debugging @value{GDBN} with @value{GDBN}, it is useful to change
17406 the prompt in one of the @value{GDBN} sessions so that you can always tell
17407 which one you are talking to.
17409 @emph{Note:} @code{set prompt} does not add a space for you after the
17410 prompt you set. This allows you to set a prompt which ends in a space
17411 or a prompt that does not.
17415 @item set prompt @var{newprompt}
17416 Directs @value{GDBN} to use @var{newprompt} as its prompt string henceforth.
17418 @kindex show prompt
17420 Prints a line of the form: @samp{Gdb's prompt is: @var{your-prompt}}
17424 @section Command Editing
17426 @cindex command line editing
17428 @value{GDBN} reads its input commands via the @dfn{Readline} interface. This
17429 @sc{gnu} library provides consistent behavior for programs which provide a
17430 command line interface to the user. Advantages are @sc{gnu} Emacs-style
17431 or @dfn{vi}-style inline editing of commands, @code{csh}-like history
17432 substitution, and a storage and recall of command history across
17433 debugging sessions.
17435 You may control the behavior of command line editing in @value{GDBN} with the
17436 command @code{set}.
17439 @kindex set editing
17442 @itemx set editing on
17443 Enable command line editing (enabled by default).
17445 @item set editing off
17446 Disable command line editing.
17448 @kindex show editing
17450 Show whether command line editing is enabled.
17453 @xref{Command Line Editing}, for more details about the Readline
17454 interface. Users unfamiliar with @sc{gnu} Emacs or @code{vi} are
17455 encouraged to read that chapter.
17457 @node Command History
17458 @section Command History
17459 @cindex command history
17461 @value{GDBN} can keep track of the commands you type during your
17462 debugging sessions, so that you can be certain of precisely what
17463 happened. Use these commands to manage the @value{GDBN} command
17466 @value{GDBN} uses the @sc{gnu} History library, a part of the Readline
17467 package, to provide the history facility. @xref{Using History
17468 Interactively}, for the detailed description of the History library.
17470 To issue a command to @value{GDBN} without affecting certain aspects of
17471 the state which is seen by users, prefix it with @samp{server }
17472 (@pxref{Server Prefix}). This
17473 means that this command will not affect the command history, nor will it
17474 affect @value{GDBN}'s notion of which command to repeat if @key{RET} is
17475 pressed on a line by itself.
17477 @cindex @code{server}, command prefix
17478 The server prefix does not affect the recording of values into the value
17479 history; to print a value without recording it into the value history,
17480 use the @code{output} command instead of the @code{print} command.
17482 Here is the description of @value{GDBN} commands related to command
17486 @cindex history substitution
17487 @cindex history file
17488 @kindex set history filename
17489 @cindex @env{GDBHISTFILE}, environment variable
17490 @item set history filename @var{fname}
17491 Set the name of the @value{GDBN} command history file to @var{fname}.
17492 This is the file where @value{GDBN} reads an initial command history
17493 list, and where it writes the command history from this session when it
17494 exits. You can access this list through history expansion or through
17495 the history command editing characters listed below. This file defaults
17496 to the value of the environment variable @code{GDBHISTFILE}, or to
17497 @file{./.gdb_history} (@file{./_gdb_history} on MS-DOS) if this variable
17500 @cindex save command history
17501 @kindex set history save
17502 @item set history save
17503 @itemx set history save on
17504 Record command history in a file, whose name may be specified with the
17505 @code{set history filename} command. By default, this option is disabled.
17507 @item set history save off
17508 Stop recording command history in a file.
17510 @cindex history size
17511 @kindex set history size
17512 @cindex @env{HISTSIZE}, environment variable
17513 @item set history size @var{size}
17514 Set the number of commands which @value{GDBN} keeps in its history list.
17515 This defaults to the value of the environment variable
17516 @code{HISTSIZE}, or to 256 if this variable is not set.
17519 History expansion assigns special meaning to the character @kbd{!}.
17520 @xref{Event Designators}, for more details.
17522 @cindex history expansion, turn on/off
17523 Since @kbd{!} is also the logical not operator in C, history expansion
17524 is off by default. If you decide to enable history expansion with the
17525 @code{set history expansion on} command, you may sometimes need to
17526 follow @kbd{!} (when it is used as logical not, in an expression) with
17527 a space or a tab to prevent it from being expanded. The readline
17528 history facilities do not attempt substitution on the strings
17529 @kbd{!=} and @kbd{!(}, even when history expansion is enabled.
17531 The commands to control history expansion are:
17534 @item set history expansion on
17535 @itemx set history expansion
17536 @kindex set history expansion
17537 Enable history expansion. History expansion is off by default.
17539 @item set history expansion off
17540 Disable history expansion.
17543 @kindex show history
17545 @itemx show history filename
17546 @itemx show history save
17547 @itemx show history size
17548 @itemx show history expansion
17549 These commands display the state of the @value{GDBN} history parameters.
17550 @code{show history} by itself displays all four states.
17555 @kindex show commands
17556 @cindex show last commands
17557 @cindex display command history
17558 @item show commands
17559 Display the last ten commands in the command history.
17561 @item show commands @var{n}
17562 Print ten commands centered on command number @var{n}.
17564 @item show commands +
17565 Print ten commands just after the commands last printed.
17569 @section Screen Size
17570 @cindex size of screen
17571 @cindex pauses in output
17573 Certain commands to @value{GDBN} may produce large amounts of
17574 information output to the screen. To help you read all of it,
17575 @value{GDBN} pauses and asks you for input at the end of each page of
17576 output. Type @key{RET} when you want to continue the output, or @kbd{q}
17577 to discard the remaining output. Also, the screen width setting
17578 determines when to wrap lines of output. Depending on what is being
17579 printed, @value{GDBN} tries to break the line at a readable place,
17580 rather than simply letting it overflow onto the following line.
17582 Normally @value{GDBN} knows the size of the screen from the terminal
17583 driver software. For example, on Unix @value{GDBN} uses the termcap data base
17584 together with the value of the @code{TERM} environment variable and the
17585 @code{stty rows} and @code{stty cols} settings. If this is not correct,
17586 you can override it with the @code{set height} and @code{set
17593 @kindex show height
17594 @item set height @var{lpp}
17596 @itemx set width @var{cpl}
17598 These @code{set} commands specify a screen height of @var{lpp} lines and
17599 a screen width of @var{cpl} characters. The associated @code{show}
17600 commands display the current settings.
17602 If you specify a height of zero lines, @value{GDBN} does not pause during
17603 output no matter how long the output is. This is useful if output is to a
17604 file or to an editor buffer.
17606 Likewise, you can specify @samp{set width 0} to prevent @value{GDBN}
17607 from wrapping its output.
17609 @item set pagination on
17610 @itemx set pagination off
17611 @kindex set pagination
17612 Turn the output pagination on or off; the default is on. Turning
17613 pagination off is the alternative to @code{set height 0}.
17615 @item show pagination
17616 @kindex show pagination
17617 Show the current pagination mode.
17622 @cindex number representation
17623 @cindex entering numbers
17625 You can always enter numbers in octal, decimal, or hexadecimal in
17626 @value{GDBN} by the usual conventions: octal numbers begin with
17627 @samp{0}, decimal numbers end with @samp{.}, and hexadecimal numbers
17628 begin with @samp{0x}. Numbers that neither begin with @samp{0} or
17629 @samp{0x}, nor end with a @samp{.} are, by default, entered in base
17630 10; likewise, the default display for numbers---when no particular
17631 format is specified---is base 10. You can change the default base for
17632 both input and output with the commands described below.
17635 @kindex set input-radix
17636 @item set input-radix @var{base}
17637 Set the default base for numeric input. Supported choices
17638 for @var{base} are decimal 8, 10, or 16. @var{base} must itself be
17639 specified either unambiguously or using the current input radix; for
17643 set input-radix 012
17644 set input-radix 10.
17645 set input-radix 0xa
17649 sets the input base to decimal. On the other hand, @samp{set input-radix 10}
17650 leaves the input radix unchanged, no matter what it was, since
17651 @samp{10}, being without any leading or trailing signs of its base, is
17652 interpreted in the current radix. Thus, if the current radix is 16,
17653 @samp{10} is interpreted in hex, i.e.@: as 16 decimal, which doesn't
17656 @kindex set output-radix
17657 @item set output-radix @var{base}
17658 Set the default base for numeric display. Supported choices
17659 for @var{base} are decimal 8, 10, or 16. @var{base} must itself be
17660 specified either unambiguously or using the current input radix.
17662 @kindex show input-radix
17663 @item show input-radix
17664 Display the current default base for numeric input.
17666 @kindex show output-radix
17667 @item show output-radix
17668 Display the current default base for numeric display.
17670 @item set radix @r{[}@var{base}@r{]}
17674 These commands set and show the default base for both input and output
17675 of numbers. @code{set radix} sets the radix of input and output to
17676 the same base; without an argument, it resets the radix back to its
17677 default value of 10.
17682 @section Configuring the Current ABI
17684 @value{GDBN} can determine the @dfn{ABI} (Application Binary Interface) of your
17685 application automatically. However, sometimes you need to override its
17686 conclusions. Use these commands to manage @value{GDBN}'s view of the
17693 One @value{GDBN} configuration can debug binaries for multiple operating
17694 system targets, either via remote debugging or native emulation.
17695 @value{GDBN} will autodetect the @dfn{OS ABI} (Operating System ABI) in use,
17696 but you can override its conclusion using the @code{set osabi} command.
17697 One example where this is useful is in debugging of binaries which use
17698 an alternate C library (e.g.@: @sc{uClibc} for @sc{gnu}/Linux) which does
17699 not have the same identifying marks that the standard C library for your
17704 Show the OS ABI currently in use.
17707 With no argument, show the list of registered available OS ABI's.
17709 @item set osabi @var{abi}
17710 Set the current OS ABI to @var{abi}.
17713 @cindex float promotion
17715 Generally, the way that an argument of type @code{float} is passed to a
17716 function depends on whether the function is prototyped. For a prototyped
17717 (i.e.@: ANSI/ISO style) function, @code{float} arguments are passed unchanged,
17718 according to the architecture's convention for @code{float}. For unprototyped
17719 (i.e.@: K&R style) functions, @code{float} arguments are first promoted to type
17720 @code{double} and then passed.
17722 Unfortunately, some forms of debug information do not reliably indicate whether
17723 a function is prototyped. If @value{GDBN} calls a function that is not marked
17724 as prototyped, it consults @kbd{set coerce-float-to-double}.
17727 @kindex set coerce-float-to-double
17728 @item set coerce-float-to-double
17729 @itemx set coerce-float-to-double on
17730 Arguments of type @code{float} will be promoted to @code{double} when passed
17731 to an unprototyped function. This is the default setting.
17733 @item set coerce-float-to-double off
17734 Arguments of type @code{float} will be passed directly to unprototyped
17737 @kindex show coerce-float-to-double
17738 @item show coerce-float-to-double
17739 Show the current setting of promoting @code{float} to @code{double}.
17743 @kindex show cp-abi
17744 @value{GDBN} needs to know the ABI used for your program's C@t{++}
17745 objects. The correct C@t{++} ABI depends on which C@t{++} compiler was
17746 used to build your application. @value{GDBN} only fully supports
17747 programs with a single C@t{++} ABI; if your program contains code using
17748 multiple C@t{++} ABI's or if @value{GDBN} can not identify your
17749 program's ABI correctly, you can tell @value{GDBN} which ABI to use.
17750 Currently supported ABI's include ``gnu-v2'', for @code{g++} versions
17751 before 3.0, ``gnu-v3'', for @code{g++} versions 3.0 and later, and
17752 ``hpaCC'' for the HP ANSI C@t{++} compiler. Other C@t{++} compilers may
17753 use the ``gnu-v2'' or ``gnu-v3'' ABI's as well. The default setting is
17758 Show the C@t{++} ABI currently in use.
17761 With no argument, show the list of supported C@t{++} ABI's.
17763 @item set cp-abi @var{abi}
17764 @itemx set cp-abi auto
17765 Set the current C@t{++} ABI to @var{abi}, or return to automatic detection.
17768 @node Messages/Warnings
17769 @section Optional Warnings and Messages
17771 @cindex verbose operation
17772 @cindex optional warnings
17773 By default, @value{GDBN} is silent about its inner workings. If you are
17774 running on a slow machine, you may want to use the @code{set verbose}
17775 command. This makes @value{GDBN} tell you when it does a lengthy
17776 internal operation, so you will not think it has crashed.
17778 Currently, the messages controlled by @code{set verbose} are those
17779 which announce that the symbol table for a source file is being read;
17780 see @code{symbol-file} in @ref{Files, ,Commands to Specify Files}.
17783 @kindex set verbose
17784 @item set verbose on
17785 Enables @value{GDBN} output of certain informational messages.
17787 @item set verbose off
17788 Disables @value{GDBN} output of certain informational messages.
17790 @kindex show verbose
17792 Displays whether @code{set verbose} is on or off.
17795 By default, if @value{GDBN} encounters bugs in the symbol table of an
17796 object file, it is silent; but if you are debugging a compiler, you may
17797 find this information useful (@pxref{Symbol Errors, ,Errors Reading
17802 @kindex set complaints
17803 @item set complaints @var{limit}
17804 Permits @value{GDBN} to output @var{limit} complaints about each type of
17805 unusual symbols before becoming silent about the problem. Set
17806 @var{limit} to zero to suppress all complaints; set it to a large number
17807 to prevent complaints from being suppressed.
17809 @kindex show complaints
17810 @item show complaints
17811 Displays how many symbol complaints @value{GDBN} is permitted to produce.
17815 By default, @value{GDBN} is cautious, and asks what sometimes seems to be a
17816 lot of stupid questions to confirm certain commands. For example, if
17817 you try to run a program which is already running:
17821 The program being debugged has been started already.
17822 Start it from the beginning? (y or n)
17825 If you are willing to unflinchingly face the consequences of your own
17826 commands, you can disable this ``feature'':
17830 @kindex set confirm
17832 @cindex confirmation
17833 @cindex stupid questions
17834 @item set confirm off
17835 Disables confirmation requests.
17837 @item set confirm on
17838 Enables confirmation requests (the default).
17840 @kindex show confirm
17842 Displays state of confirmation requests.
17846 @cindex command tracing
17847 If you need to debug user-defined commands or sourced files you may find it
17848 useful to enable @dfn{command tracing}. In this mode each command will be
17849 printed as it is executed, prefixed with one or more @samp{+} symbols, the
17850 quantity denoting the call depth of each command.
17853 @kindex set trace-commands
17854 @cindex command scripts, debugging
17855 @item set trace-commands on
17856 Enable command tracing.
17857 @item set trace-commands off
17858 Disable command tracing.
17859 @item show trace-commands
17860 Display the current state of command tracing.
17863 @node Debugging Output
17864 @section Optional Messages about Internal Happenings
17865 @cindex optional debugging messages
17867 @value{GDBN} has commands that enable optional debugging messages from
17868 various @value{GDBN} subsystems; normally these commands are of
17869 interest to @value{GDBN} maintainers, or when reporting a bug. This
17870 section documents those commands.
17873 @kindex set exec-done-display
17874 @item set exec-done-display
17875 Turns on or off the notification of asynchronous commands'
17876 completion. When on, @value{GDBN} will print a message when an
17877 asynchronous command finishes its execution. The default is off.
17878 @kindex show exec-done-display
17879 @item show exec-done-display
17880 Displays the current setting of asynchronous command completion
17883 @cindex gdbarch debugging info
17884 @cindex architecture debugging info
17885 @item set debug arch
17886 Turns on or off display of gdbarch debugging info. The default is off
17888 @item show debug arch
17889 Displays the current state of displaying gdbarch debugging info.
17890 @item set debug aix-thread
17891 @cindex AIX threads
17892 Display debugging messages about inner workings of the AIX thread
17894 @item show debug aix-thread
17895 Show the current state of AIX thread debugging info display.
17896 @item set debug dwarf2-die
17897 @cindex DWARF2 DIEs
17898 Dump DWARF2 DIEs after they are read in.
17899 The value is the number of nesting levels to print.
17900 A value of zero turns off the display.
17901 @item show debug dwarf2-die
17902 Show the current state of DWARF2 DIE debugging.
17903 @item set debug displaced
17904 @cindex displaced stepping debugging info
17905 Turns on or off display of @value{GDBN} debugging info for the
17906 displaced stepping support. The default is off.
17907 @item show debug displaced
17908 Displays the current state of displaying @value{GDBN} debugging info
17909 related to displaced stepping.
17910 @item set debug event
17911 @cindex event debugging info
17912 Turns on or off display of @value{GDBN} event debugging info. The
17914 @item show debug event
17915 Displays the current state of displaying @value{GDBN} event debugging
17917 @item set debug expression
17918 @cindex expression debugging info
17919 Turns on or off display of debugging info about @value{GDBN}
17920 expression parsing. The default is off.
17921 @item show debug expression
17922 Displays the current state of displaying debugging info about
17923 @value{GDBN} expression parsing.
17924 @item set debug frame
17925 @cindex frame debugging info
17926 Turns on or off display of @value{GDBN} frame debugging info. The
17928 @item show debug frame
17929 Displays the current state of displaying @value{GDBN} frame debugging
17931 @item set debug gnu-nat
17932 @cindex @sc{gnu}/Hurd debug messages
17933 Turns on or off debugging messages from the @sc{gnu}/Hurd debug support.
17934 @item show debug gnu-nat
17935 Show the current state of @sc{gnu}/Hurd debugging messages.
17936 @item set debug infrun
17937 @cindex inferior debugging info
17938 Turns on or off display of @value{GDBN} debugging info for running the inferior.
17939 The default is off. @file{infrun.c} contains GDB's runtime state machine used
17940 for implementing operations such as single-stepping the inferior.
17941 @item show debug infrun
17942 Displays the current state of @value{GDBN} inferior debugging.
17943 @item set debug lin-lwp
17944 @cindex @sc{gnu}/Linux LWP debug messages
17945 @cindex Linux lightweight processes
17946 Turns on or off debugging messages from the Linux LWP debug support.
17947 @item show debug lin-lwp
17948 Show the current state of Linux LWP debugging messages.
17949 @item set debug lin-lwp-async
17950 @cindex @sc{gnu}/Linux LWP async debug messages
17951 @cindex Linux lightweight processes
17952 Turns on or off debugging messages from the Linux LWP async debug support.
17953 @item show debug lin-lwp-async
17954 Show the current state of Linux LWP async debugging messages.
17955 @item set debug observer
17956 @cindex observer debugging info
17957 Turns on or off display of @value{GDBN} observer debugging. This
17958 includes info such as the notification of observable events.
17959 @item show debug observer
17960 Displays the current state of observer debugging.
17961 @item set debug overload
17962 @cindex C@t{++} overload debugging info
17963 Turns on or off display of @value{GDBN} C@t{++} overload debugging
17964 info. This includes info such as ranking of functions, etc. The default
17966 @item show debug overload
17967 Displays the current state of displaying @value{GDBN} C@t{++} overload
17969 @cindex packets, reporting on stdout
17970 @cindex serial connections, debugging
17971 @cindex debug remote protocol
17972 @cindex remote protocol debugging
17973 @cindex display remote packets
17974 @item set debug remote
17975 Turns on or off display of reports on all packets sent back and forth across
17976 the serial line to the remote machine. The info is printed on the
17977 @value{GDBN} standard output stream. The default is off.
17978 @item show debug remote
17979 Displays the state of display of remote packets.
17980 @item set debug serial
17981 Turns on or off display of @value{GDBN} serial debugging info. The
17983 @item show debug serial
17984 Displays the current state of displaying @value{GDBN} serial debugging
17986 @item set debug solib-frv
17987 @cindex FR-V shared-library debugging
17988 Turns on or off debugging messages for FR-V shared-library code.
17989 @item show debug solib-frv
17990 Display the current state of FR-V shared-library code debugging
17992 @item set debug target
17993 @cindex target debugging info
17994 Turns on or off display of @value{GDBN} target debugging info. This info
17995 includes what is going on at the target level of GDB, as it happens. The
17996 default is 0. Set it to 1 to track events, and to 2 to also track the
17997 value of large memory transfers. Changes to this flag do not take effect
17998 until the next time you connect to a target or use the @code{run} command.
17999 @item show debug target
18000 Displays the current state of displaying @value{GDBN} target debugging
18002 @item set debug timestamp
18003 @cindex timestampping debugging info
18004 Turns on or off display of timestamps with @value{GDBN} debugging info.
18005 When enabled, seconds and microseconds are displayed before each debugging
18007 @item show debug timestamp
18008 Displays the current state of displaying timestamps with @value{GDBN}
18010 @item set debugvarobj
18011 @cindex variable object debugging info
18012 Turns on or off display of @value{GDBN} variable object debugging
18013 info. The default is off.
18014 @item show debugvarobj
18015 Displays the current state of displaying @value{GDBN} variable object
18017 @item set debug xml
18018 @cindex XML parser debugging
18019 Turns on or off debugging messages for built-in XML parsers.
18020 @item show debug xml
18021 Displays the current state of XML debugging messages.
18024 @node Extending GDB
18025 @chapter Extending @value{GDBN}
18026 @cindex extending GDB
18028 @value{GDBN} provides two mechanisms for extension. The first is based
18029 on composition of @value{GDBN} commands, and the second is based on the
18030 Python scripting language.
18033 * Sequences:: Canned Sequences of Commands
18034 * Python:: Scripting @value{GDBN} using Python
18038 @section Canned Sequences of Commands
18040 Aside from breakpoint commands (@pxref{Break Commands, ,Breakpoint
18041 Command Lists}), @value{GDBN} provides two ways to store sequences of
18042 commands for execution as a unit: user-defined commands and command
18046 * Define:: How to define your own commands
18047 * Hooks:: Hooks for user-defined commands
18048 * Command Files:: How to write scripts of commands to be stored in a file
18049 * Output:: Commands for controlled output
18053 @subsection User-defined Commands
18055 @cindex user-defined command
18056 @cindex arguments, to user-defined commands
18057 A @dfn{user-defined command} is a sequence of @value{GDBN} commands to
18058 which you assign a new name as a command. This is done with the
18059 @code{define} command. User commands may accept up to 10 arguments
18060 separated by whitespace. Arguments are accessed within the user command
18061 via @code{$arg0@dots{}$arg9}. A trivial example:
18065 print $arg0 + $arg1 + $arg2
18070 To execute the command use:
18077 This defines the command @code{adder}, which prints the sum of
18078 its three arguments. Note the arguments are text substitutions, so they may
18079 reference variables, use complex expressions, or even perform inferior
18082 @cindex argument count in user-defined commands
18083 @cindex how many arguments (user-defined commands)
18084 In addition, @code{$argc} may be used to find out how many arguments have
18085 been passed. This expands to a number in the range 0@dots{}10.
18090 print $arg0 + $arg1
18093 print $arg0 + $arg1 + $arg2
18101 @item define @var{commandname}
18102 Define a command named @var{commandname}. If there is already a command
18103 by that name, you are asked to confirm that you want to redefine it.
18104 @var{commandname} may be a bare command name consisting of letters,
18105 numbers, dashes, and underscores. It may also start with any predefined
18106 prefix command. For example, @samp{define target my-target} creates
18107 a user-defined @samp{target my-target} command.
18109 The definition of the command is made up of other @value{GDBN} command lines,
18110 which are given following the @code{define} command. The end of these
18111 commands is marked by a line containing @code{end}.
18114 @kindex end@r{ (user-defined commands)}
18115 @item document @var{commandname}
18116 Document the user-defined command @var{commandname}, so that it can be
18117 accessed by @code{help}. The command @var{commandname} must already be
18118 defined. This command reads lines of documentation just as @code{define}
18119 reads the lines of the command definition, ending with @code{end}.
18120 After the @code{document} command is finished, @code{help} on command
18121 @var{commandname} displays the documentation you have written.
18123 You may use the @code{document} command again to change the
18124 documentation of a command. Redefining the command with @code{define}
18125 does not change the documentation.
18127 @kindex dont-repeat
18128 @cindex don't repeat command
18130 Used inside a user-defined command, this tells @value{GDBN} that this
18131 command should not be repeated when the user hits @key{RET}
18132 (@pxref{Command Syntax, repeat last command}).
18134 @kindex help user-defined
18135 @item help user-defined
18136 List all user-defined commands, with the first line of the documentation
18141 @itemx show user @var{commandname}
18142 Display the @value{GDBN} commands used to define @var{commandname} (but
18143 not its documentation). If no @var{commandname} is given, display the
18144 definitions for all user-defined commands.
18146 @cindex infinite recursion in user-defined commands
18147 @kindex show max-user-call-depth
18148 @kindex set max-user-call-depth
18149 @item show max-user-call-depth
18150 @itemx set max-user-call-depth
18151 The value of @code{max-user-call-depth} controls how many recursion
18152 levels are allowed in user-defined commands before @value{GDBN} suspects an
18153 infinite recursion and aborts the command.
18156 In addition to the above commands, user-defined commands frequently
18157 use control flow commands, described in @ref{Command Files}.
18159 When user-defined commands are executed, the
18160 commands of the definition are not printed. An error in any command
18161 stops execution of the user-defined command.
18163 If used interactively, commands that would ask for confirmation proceed
18164 without asking when used inside a user-defined command. Many @value{GDBN}
18165 commands that normally print messages to say what they are doing omit the
18166 messages when used in a user-defined command.
18169 @subsection User-defined Command Hooks
18170 @cindex command hooks
18171 @cindex hooks, for commands
18172 @cindex hooks, pre-command
18175 You may define @dfn{hooks}, which are a special kind of user-defined
18176 command. Whenever you run the command @samp{foo}, if the user-defined
18177 command @samp{hook-foo} exists, it is executed (with no arguments)
18178 before that command.
18180 @cindex hooks, post-command
18182 A hook may also be defined which is run after the command you executed.
18183 Whenever you run the command @samp{foo}, if the user-defined command
18184 @samp{hookpost-foo} exists, it is executed (with no arguments) after
18185 that command. Post-execution hooks may exist simultaneously with
18186 pre-execution hooks, for the same command.
18188 It is valid for a hook to call the command which it hooks. If this
18189 occurs, the hook is not re-executed, thereby avoiding infinite recursion.
18191 @c It would be nice if hookpost could be passed a parameter indicating
18192 @c if the command it hooks executed properly or not. FIXME!
18194 @kindex stop@r{, a pseudo-command}
18195 In addition, a pseudo-command, @samp{stop} exists. Defining
18196 (@samp{hook-stop}) makes the associated commands execute every time
18197 execution stops in your program: before breakpoint commands are run,
18198 displays are printed, or the stack frame is printed.
18200 For example, to ignore @code{SIGALRM} signals while
18201 single-stepping, but treat them normally during normal execution,
18206 handle SIGALRM nopass
18210 handle SIGALRM pass
18213 define hook-continue
18214 handle SIGALRM pass
18218 As a further example, to hook at the beginning and end of the @code{echo}
18219 command, and to add extra text to the beginning and end of the message,
18227 define hookpost-echo
18231 (@value{GDBP}) echo Hello World
18232 <<<---Hello World--->>>
18237 You can define a hook for any single-word command in @value{GDBN}, but
18238 not for command aliases; you should define a hook for the basic command
18239 name, e.g.@: @code{backtrace} rather than @code{bt}.
18240 @c FIXME! So how does Joe User discover whether a command is an alias
18242 You can hook a multi-word command by adding @code{hook-} or
18243 @code{hookpost-} to the last word of the command, e.g.@:
18244 @samp{define target hook-remote} to add a hook to @samp{target remote}.
18246 If an error occurs during the execution of your hook, execution of
18247 @value{GDBN} commands stops and @value{GDBN} issues a prompt
18248 (before the command that you actually typed had a chance to run).
18250 If you try to define a hook which does not match any known command, you
18251 get a warning from the @code{define} command.
18253 @node Command Files
18254 @subsection Command Files
18256 @cindex command files
18257 @cindex scripting commands
18258 A command file for @value{GDBN} is a text file made of lines that are
18259 @value{GDBN} commands. Comments (lines starting with @kbd{#}) may
18260 also be included. An empty line in a command file does nothing; it
18261 does not mean to repeat the last command, as it would from the
18264 You can request the execution of a command file with the @code{source}
18269 @cindex execute commands from a file
18270 @item source [@code{-v}] @var{filename}
18271 Execute the command file @var{filename}.
18274 The lines in a command file are generally executed sequentially,
18275 unless the order of execution is changed by one of the
18276 @emph{flow-control commands} described below. The commands are not
18277 printed as they are executed. An error in any command terminates
18278 execution of the command file and control is returned to the console.
18280 @value{GDBN} searches for @var{filename} in the current directory and then
18281 on the search path (specified with the @samp{directory} command).
18283 If @code{-v}, for verbose mode, is given then @value{GDBN} displays
18284 each command as it is executed. The option must be given before
18285 @var{filename}, and is interpreted as part of the filename anywhere else.
18287 Commands that would ask for confirmation if used interactively proceed
18288 without asking when used in a command file. Many @value{GDBN} commands that
18289 normally print messages to say what they are doing omit the messages
18290 when called from command files.
18292 @value{GDBN} also accepts command input from standard input. In this
18293 mode, normal output goes to standard output and error output goes to
18294 standard error. Errors in a command file supplied on standard input do
18295 not terminate execution of the command file---execution continues with
18299 gdb < cmds > log 2>&1
18302 (The syntax above will vary depending on the shell used.) This example
18303 will execute commands from the file @file{cmds}. All output and errors
18304 would be directed to @file{log}.
18306 Since commands stored on command files tend to be more general than
18307 commands typed interactively, they frequently need to deal with
18308 complicated situations, such as different or unexpected values of
18309 variables and symbols, changes in how the program being debugged is
18310 built, etc. @value{GDBN} provides a set of flow-control commands to
18311 deal with these complexities. Using these commands, you can write
18312 complex scripts that loop over data structures, execute commands
18313 conditionally, etc.
18320 This command allows to include in your script conditionally executed
18321 commands. The @code{if} command takes a single argument, which is an
18322 expression to evaluate. It is followed by a series of commands that
18323 are executed only if the expression is true (its value is nonzero).
18324 There can then optionally be an @code{else} line, followed by a series
18325 of commands that are only executed if the expression was false. The
18326 end of the list is marked by a line containing @code{end}.
18330 This command allows to write loops. Its syntax is similar to
18331 @code{if}: the command takes a single argument, which is an expression
18332 to evaluate, and must be followed by the commands to execute, one per
18333 line, terminated by an @code{end}. These commands are called the
18334 @dfn{body} of the loop. The commands in the body of @code{while} are
18335 executed repeatedly as long as the expression evaluates to true.
18339 This command exits the @code{while} loop in whose body it is included.
18340 Execution of the script continues after that @code{while}s @code{end}
18343 @kindex loop_continue
18344 @item loop_continue
18345 This command skips the execution of the rest of the body of commands
18346 in the @code{while} loop in whose body it is included. Execution
18347 branches to the beginning of the @code{while} loop, where it evaluates
18348 the controlling expression.
18350 @kindex end@r{ (if/else/while commands)}
18352 Terminate the block of commands that are the body of @code{if},
18353 @code{else}, or @code{while} flow-control commands.
18358 @subsection Commands for Controlled Output
18360 During the execution of a command file or a user-defined command, normal
18361 @value{GDBN} output is suppressed; the only output that appears is what is
18362 explicitly printed by the commands in the definition. This section
18363 describes three commands useful for generating exactly the output you
18368 @item echo @var{text}
18369 @c I do not consider backslash-space a standard C escape sequence
18370 @c because it is not in ANSI.
18371 Print @var{text}. Nonprinting characters can be included in
18372 @var{text} using C escape sequences, such as @samp{\n} to print a
18373 newline. @strong{No newline is printed unless you specify one.}
18374 In addition to the standard C escape sequences, a backslash followed
18375 by a space stands for a space. This is useful for displaying a
18376 string with spaces at the beginning or the end, since leading and
18377 trailing spaces are otherwise trimmed from all arguments.
18378 To print @samp{@w{ }and foo =@w{ }}, use the command
18379 @samp{echo \@w{ }and foo = \@w{ }}.
18381 A backslash at the end of @var{text} can be used, as in C, to continue
18382 the command onto subsequent lines. For example,
18385 echo This is some text\n\
18386 which is continued\n\
18387 onto several lines.\n
18390 produces the same output as
18393 echo This is some text\n
18394 echo which is continued\n
18395 echo onto several lines.\n
18399 @item output @var{expression}
18400 Print the value of @var{expression} and nothing but that value: no
18401 newlines, no @samp{$@var{nn} = }. The value is not entered in the
18402 value history either. @xref{Expressions, ,Expressions}, for more information
18405 @item output/@var{fmt} @var{expression}
18406 Print the value of @var{expression} in format @var{fmt}. You can use
18407 the same formats as for @code{print}. @xref{Output Formats,,Output
18408 Formats}, for more information.
18411 @item printf @var{template}, @var{expressions}@dots{}
18412 Print the values of one or more @var{expressions} under the control of
18413 the string @var{template}. To print several values, make
18414 @var{expressions} be a comma-separated list of individual expressions,
18415 which may be either numbers or pointers. Their values are printed as
18416 specified by @var{template}, exactly as a C program would do by
18417 executing the code below:
18420 printf (@var{template}, @var{expressions}@dots{});
18423 As in @code{C} @code{printf}, ordinary characters in @var{template}
18424 are printed verbatim, while @dfn{conversion specification} introduced
18425 by the @samp{%} character cause subsequent @var{expressions} to be
18426 evaluated, their values converted and formatted according to type and
18427 style information encoded in the conversion specifications, and then
18430 For example, you can print two values in hex like this:
18433 printf "foo, bar-foo = 0x%x, 0x%x\n", foo, bar-foo
18436 @code{printf} supports all the standard @code{C} conversion
18437 specifications, including the flags and modifiers between the @samp{%}
18438 character and the conversion letter, with the following exceptions:
18442 The argument-ordering modifiers, such as @samp{2$}, are not supported.
18445 The modifier @samp{*} is not supported for specifying precision or
18449 The @samp{'} flag (for separation of digits into groups according to
18450 @code{LC_NUMERIC'}) is not supported.
18453 The type modifiers @samp{hh}, @samp{j}, @samp{t}, and @samp{z} are not
18457 The conversion letter @samp{n} (as in @samp{%n}) is not supported.
18460 The conversion letters @samp{a} and @samp{A} are not supported.
18464 Note that the @samp{ll} type modifier is supported only if the
18465 underlying @code{C} implementation used to build @value{GDBN} supports
18466 the @code{long long int} type, and the @samp{L} type modifier is
18467 supported only if @code{long double} type is available.
18469 As in @code{C}, @code{printf} supports simple backslash-escape
18470 sequences, such as @code{\n}, @samp{\t}, @samp{\\}, @samp{\"},
18471 @samp{\a}, and @samp{\f}, that consist of backslash followed by a
18472 single character. Octal and hexadecimal escape sequences are not
18475 Additionally, @code{printf} supports conversion specifications for DFP
18476 (@dfn{Decimal Floating Point}) types using the following length modifiers
18477 together with a floating point specifier.
18482 @samp{H} for printing @code{Decimal32} types.
18485 @samp{D} for printing @code{Decimal64} types.
18488 @samp{DD} for printing @code{Decimal128} types.
18491 If the underlying @code{C} implementation used to build @value{GDBN} has
18492 support for the three length modifiers for DFP types, other modifiers
18493 such as width and precision will also be available for @value{GDBN} to use.
18495 In case there is no such @code{C} support, no additional modifiers will be
18496 available and the value will be printed in the standard way.
18498 Here's an example of printing DFP types using the above conversion letters:
18500 printf "D32: %Hf - D64: %Df - D128: %DDf\n",1.2345df,1.2E10dd,1.2E1dl
18506 @section Scripting @value{GDBN} using Python
18507 @cindex python scripting
18508 @cindex scripting with python
18510 You can script @value{GDBN} using the @uref{http://www.python.org/,
18511 Python programming language}. This feature is available only if
18512 @value{GDBN} was configured using @option{--with-python}.
18515 * Python Commands:: Accessing Python from @value{GDBN}.
18516 * Python API:: Accessing @value{GDBN} from Python.
18519 @node Python Commands
18520 @subsection Python Commands
18521 @cindex python commands
18522 @cindex commands to access python
18524 @value{GDBN} provides one command for accessing the Python interpreter,
18525 and one related setting:
18529 @item python @r{[}@var{code}@r{]}
18530 The @code{python} command can be used to evaluate Python code.
18532 If given an argument, the @code{python} command will evaluate the
18533 argument as a Python command. For example:
18536 (@value{GDBP}) python print 23
18540 If you do not provide an argument to @code{python}, it will act as a
18541 multi-line command, like @code{define}. In this case, the Python
18542 script is made up of subsequent command lines, given after the
18543 @code{python} command. This command list is terminated using a line
18544 containing @code{end}. For example:
18547 (@value{GDBP}) python
18549 End with a line saying just "end".
18555 @kindex maint set python print-stack
18556 @item maint set python print-stack
18557 By default, @value{GDBN} will print a stack trace when an error occurs
18558 in a Python script. This can be controlled using @code{maint set
18559 python print-stack}: if @code{on}, the default, then Python stack
18560 printing is enabled; if @code{off}, then Python stack printing is
18565 @subsection Python API
18567 @cindex programming in python
18569 @cindex python stdout
18570 @cindex python pagination
18571 At startup, @value{GDBN} overrides Python's @code{sys.stdout} and
18572 @code{sys.stderr} to print using @value{GDBN}'s output-paging streams.
18573 A Python program which outputs to one of these streams may have its
18574 output interrupted by the user (@pxref{Screen Size}). In this
18575 situation, a Python @code{KeyboardInterrupt} exception is thrown.
18578 * Basic Python:: Basic Python Functions.
18579 * Exception Handling::
18580 * Auto-loading:: Automatically loading Python code.
18581 * Values From Inferior::
18582 * Types In Python:: Python representation of types.
18583 * Pretty Printing:: Pretty-printing values.
18584 * Selecting Pretty-Printers:: How GDB chooses a pretty-printer.
18585 * Commands In Python:: Implementing new commands in Python.
18586 * Functions In Python:: Writing new convenience functions.
18587 * Objfiles In Python:: Object files.
18588 * Frames In Python:: Acessing inferior stack frames from Python.
18592 @subsubsection Basic Python
18594 @cindex python functions
18595 @cindex python module
18597 @value{GDBN} introduces a new Python module, named @code{gdb}. All
18598 methods and classes added by @value{GDBN} are placed in this module.
18599 @value{GDBN} automatically @code{import}s the @code{gdb} module for
18600 use in all scripts evaluated by the @code{python} command.
18602 @findex gdb.execute
18603 @defun execute command [from_tty]
18604 Evaluate @var{command}, a string, as a @value{GDBN} CLI command.
18605 If a GDB exception happens while @var{command} runs, it is
18606 translated as described in @ref{Exception Handling,,Exception Handling}.
18607 If no exceptions occur, this function returns @code{None}.
18609 @var{from_tty} specifies whether @value{GDBN} ought to consider this
18610 command as having originated from the user invoking it interactively.
18611 It must be a boolean value. If omitted, it defaults to @code{False}.
18614 @findex gdb.parameter
18615 @defun parameter parameter
18616 Return the value of a @value{GDBN} parameter. @var{parameter} is a
18617 string naming the parameter to look up; @var{parameter} may contain
18618 spaces if the parameter has a multi-part name. For example,
18619 @samp{print object} is a valid parameter name.
18621 If the named parameter does not exist, this function throws a
18622 @code{RuntimeError}. Otherwise, the parameter's value is converted to
18623 a Python value of the appropriate type, and returned.
18626 @findex gdb.history
18627 @defun history number
18628 Return a value from @value{GDBN}'s value history (@pxref{Value
18629 History}). @var{number} indicates which history element to return.
18630 If @var{number} is negative, then @value{GDBN} will take its absolute value
18631 and count backward from the last element (i.e., the most recent element) to
18632 find the value to return. If @var{number} is zero, then @value{GDBN} will
18633 return the most recent element. If the element specified by @var{number}
18634 doesn't exist in the value history, a @code{RuntimeError} exception will be
18637 If no exception is raised, the return value is always an instance of
18638 @code{gdb.Value} (@pxref{Values From Inferior}).
18642 @defun write string
18643 Print a string to @value{GDBN}'s paginated standard output stream.
18644 Writing to @code{sys.stdout} or @code{sys.stderr} will automatically
18645 call this function.
18650 Flush @value{GDBN}'s paginated standard output stream. Flushing
18651 @code{sys.stdout} or @code{sys.stderr} will automatically call this
18655 @node Exception Handling
18656 @subsubsection Exception Handling
18657 @cindex python exceptions
18658 @cindex exceptions, python
18660 When executing the @code{python} command, Python exceptions
18661 uncaught within the Python code are translated to calls to
18662 @value{GDBN} error-reporting mechanism. If the command that called
18663 @code{python} does not handle the error, @value{GDBN} will
18664 terminate it and print an error message containing the Python
18665 exception name, the associated value, and the Python call stack
18666 backtrace at the point where the exception was raised. Example:
18669 (@value{GDBP}) python print foo
18670 Traceback (most recent call last):
18671 File "<string>", line 1, in <module>
18672 NameError: name 'foo' is not defined
18675 @value{GDBN} errors that happen in @value{GDBN} commands invoked by Python
18676 code are converted to Python @code{RuntimeError} exceptions. User
18677 interrupt (via @kbd{C-c} or by typing @kbd{q} at a pagination
18678 prompt) is translated to a Python @code{KeyboardInterrupt}
18679 exception. If you catch these exceptions in your Python code, your
18680 exception handler will see @code{RuntimeError} or
18681 @code{KeyboardInterrupt} as the exception type, the @value{GDBN} error
18682 message as its value, and the Python call stack backtrace at the
18683 Python statement closest to where the @value{GDBN} error occured as the
18687 @subsubsection Auto-loading
18688 @cindex auto-loading, Python
18690 When a new object file is read (for example, due to the @code{file}
18691 command, or because the inferior has loaded a shared library),
18692 @value{GDBN} will look for a file named @file{@var{objfile}-gdb.py},
18693 where @var{objfile} is the object file's real name, formed by ensuring
18694 that the file name is absolute, following all symlinks, and resolving
18695 @code{.} and @code{..} components. If this file exists and is
18696 readable, @value{GDBN} will evaluate it as a Python script.
18698 If this file does not exist, and if the parameter
18699 @code{debug-file-directory} is set (@pxref{Separate Debug Files}),
18700 then @value{GDBN} will use the file named
18701 @file{@var{debug-file-directory}/@var{real-name}}, where
18702 @var{real-name} is the object file's real name, as described above.
18704 Finally, if this file does not exist, then @value{GDBN} will look for
18705 a file named @file{@var{data-directory}/python/auto-load/@var{real-name}}, where
18706 @var{data-directory} is @value{GDBN}'s data directory (available via
18707 @code{show data-directory}, @pxref{Data Files}), and @var{real-name}
18708 is the object file's real name, as described above.
18710 When reading an auto-loaded file, @value{GDBN} sets the ``current
18711 objfile''. This is available via the @code{gdb.current_objfile}
18712 function (@pxref{Objfiles In Python}). This can be useful for
18713 registering objfile-specific pretty-printers.
18715 The auto-loading feature is useful for supplying application-specific
18716 debugging commands and scripts. You can enable or disable this
18717 feature, and view its current state.
18720 @kindex maint set python auto-load
18721 @item maint set python auto-load [yes|no]
18722 Enable or disable the Python auto-loading feature.
18724 @kindex show python auto-load
18725 @item show python auto-load
18726 Show whether Python auto-loading is enabled or disabled.
18729 @value{GDBN} does not track which files it has already auto-loaded.
18730 So, your @samp{-gdb.py} file should take care to ensure that it may be
18731 evaluated multiple times without error.
18733 @node Values From Inferior
18734 @subsubsection Values From Inferior
18735 @cindex values from inferior, with Python
18736 @cindex python, working with values from inferior
18738 @cindex @code{gdb.Value}
18739 @value{GDBN} provides values it obtains from the inferior program in
18740 an object of type @code{gdb.Value}. @value{GDBN} uses this object
18741 for its internal bookkeeping of the inferior's values, and for
18742 fetching values when necessary.
18744 Inferior values that are simple scalars can be used directly in
18745 Python expressions that are valid for the value's data type. Here's
18746 an example for an integer or floating-point value @code{some_val}:
18753 As result of this, @code{bar} will also be a @code{gdb.Value} object
18754 whose values are of the same type as those of @code{some_val}.
18756 Inferior values that are structures or instances of some class can
18757 be accessed using the Python @dfn{dictionary syntax}. For example, if
18758 @code{some_val} is a @code{gdb.Value} instance holding a structure, you
18759 can access its @code{foo} element with:
18762 bar = some_val['foo']
18765 Again, @code{bar} will also be a @code{gdb.Value} object.
18767 The following attributes are provided:
18770 @defivar Value address
18771 If this object is addressable, this read-only attribute holds a
18772 @code{gdb.Value} object representing the address. Otherwise,
18773 this attribute holds @code{None}.
18776 @cindex optimized out value in Python
18777 @defivar Value is_optimized_out
18778 This read-only boolean attribute is true if the compiler optimized out
18779 this value, thus it is not available for fetching from the inferior.
18782 @defivar Value type
18783 The type of this @code{gdb.Value}. The value of this attribute is a
18784 @code{gdb.Type} object.
18788 The following methods are provided:
18791 @defmethod Value dereference
18792 For pointer data types, this method returns a new @code{gdb.Value} object
18793 whose contents is the object pointed to by the pointer. For example, if
18794 @code{foo} is a C pointer to an @code{int}, declared in your C program as
18801 then you can use the corresponding @code{gdb.Value} to access what
18802 @code{foo} points to like this:
18805 bar = foo.dereference ()
18808 The result @code{bar} will be a @code{gdb.Value} object holding the
18809 value pointed to by @code{foo}.
18812 @defmethod Value string @r{[}encoding@r{]} @r{[}errors@r{]}
18813 If this @code{gdb.Value} represents a string, then this method
18814 converts the contents to a Python string. Otherwise, this method will
18815 throw an exception.
18817 Strings are recognized in a language-specific way; whether a given
18818 @code{gdb.Value} represents a string is determined by the current
18821 For C-like languages, a value is a string if it is a pointer to or an
18822 array of characters or ints. The string is assumed to be terminated
18823 by a zero of the appropriate width.
18825 If the optional @var{encoding} argument is given, it must be a string
18826 naming the encoding of the string in the @code{gdb.Value}, such as
18827 @code{"ascii"}, @code{"iso-8859-6"} or @code{"utf-8"}. It accepts
18828 the same encodings as the corresponding argument to Python's
18829 @code{string.decode} method, and the Python codec machinery will be used
18830 to convert the string. If @var{encoding} is not given, or if
18831 @var{encoding} is the empty string, then either the @code{target-charset}
18832 (@pxref{Character Sets}) will be used, or a language-specific encoding
18833 will be used, if the current language is able to supply one.
18835 The optional @var{errors} argument is the same as the corresponding
18836 argument to Python's @code{string.decode} method.
18840 @node Types In Python
18841 @subsubsection Types In Python
18842 @cindex types in Python
18843 @cindex Python, working with types
18846 @value{GDBN} represents types from the inferior using the class
18849 The following type-related functions are available in the @code{gdb}
18852 @findex gdb.lookup_type
18853 @defun lookup_type name [block]
18854 This function looks up a type by name. @var{name} is the name of the
18855 type to look up. It must be a string.
18857 Ordinarily, this function will return an instance of @code{gdb.Type}.
18858 If the named type cannot be found, it will throw an exception.
18861 An instance of @code{Type} has the following attributes:
18865 The type code for this type. The type code will be one of the
18866 @code{TYPE_CODE_} constants defined below.
18869 @defivar Type sizeof
18870 The size of this type, in target @code{char} units. Usually, a
18871 target's @code{char} type will be an 8-bit byte. However, on some
18872 unusual platforms, this type may have a different size.
18876 The tag name for this type. The tag name is the name after
18877 @code{struct}, @code{union}, or @code{enum} in C and C@t{++}; not all
18878 languages have this concept. If this type has no tag name, then
18879 @code{None} is returned.
18883 The following methods are provided:
18886 @defmethod Type fields
18887 For structure and union types, this method returns the fields. Range
18888 types have two fields, the minimum and maximum values. Enum types
18889 have one field per enum constant. Function and method types have one
18890 field per parameter. The base types of C@t{++} classes are also
18891 represented as fields. If the type has no fields, or does not fit
18892 into one of these categories, an empty sequence will be returned.
18894 Each field is an object, with some pre-defined attributes:
18897 This attribute is not available for @code{static} fields (as in
18898 C@t{++} or Java). For non-@code{static} fields, the value is the bit
18899 position of the field.
18902 The name of the field, or @code{None} for anonymous fields.
18905 This is @code{True} if the field is artificial, usually meaning that
18906 it was provided by the compiler and not the user. This attribute is
18907 always provided, and is @code{False} if the field is not artificial.
18910 If the field is packed, or is a bitfield, then this will have a
18911 non-zero value, which is the size of the field in bits. Otherwise,
18912 this will be zero; in this case the field's size is given by its type.
18915 The type of the field. This is usually an instance of @code{Type},
18916 but it can be @code{None} in some situations.
18920 @defmethod Type const
18921 Return a new @code{gdb.Type} object which represents a
18922 @code{const}-qualified variant of this type.
18925 @defmethod Type volatile
18926 Return a new @code{gdb.Type} object which represents a
18927 @code{volatile}-qualified variant of this type.
18930 @defmethod Type unqualified
18931 Return a new @code{gdb.Type} object which represents an unqualified
18932 variant of this type. That is, the result is neither @code{const} nor
18936 @defmethod Type reference
18937 Return a new @code{gdb.Type} object which represents a reference to this
18941 @defmethod Type strip_typedefs
18942 Return a new @code{gdb.Type} that represents the real type,
18943 after removing all layers of typedefs.
18946 @defmethod Type target
18947 Return a new @code{gdb.Type} object which represents the target type
18950 For a pointer type, the target type is the type of the pointed-to
18951 object. For an array type (meaning C-like arrays), the target type is
18952 the type of the elements of the array. For a function or method type,
18953 the target type is the type of the return value. For a complex type,
18954 the target type is the type of the elements. For a typedef, the
18955 target type is the aliased type.
18957 If the type does not have a target, this method will throw an
18961 @defmethod Type template_argument n
18962 If this @code{gdb.Type} is an instantiation of a template, this will
18963 return a new @code{gdb.Type} which represents the type of the
18964 @var{n}th template argument.
18966 If this @code{gdb.Type} is not a template type, this will throw an
18967 exception. Ordinarily, only C@t{++} code will have template types.
18969 @var{name} is searched for globally.
18974 Each type has a code, which indicates what category this type falls
18975 into. The available type categories are represented by constants
18976 defined in the @code{gdb} module:
18979 @findex TYPE_CODE_PTR
18980 @findex gdb.TYPE_CODE_PTR
18981 @item TYPE_CODE_PTR
18982 The type is a pointer.
18984 @findex TYPE_CODE_ARRAY
18985 @findex gdb.TYPE_CODE_ARRAY
18986 @item TYPE_CODE_ARRAY
18987 The type is an array.
18989 @findex TYPE_CODE_STRUCT
18990 @findex gdb.TYPE_CODE_STRUCT
18991 @item TYPE_CODE_STRUCT
18992 The type is a structure.
18994 @findex TYPE_CODE_UNION
18995 @findex gdb.TYPE_CODE_UNION
18996 @item TYPE_CODE_UNION
18997 The type is a union.
18999 @findex TYPE_CODE_ENUM
19000 @findex gdb.TYPE_CODE_ENUM
19001 @item TYPE_CODE_ENUM
19002 The type is an enum.
19004 @findex TYPE_CODE_FLAGS
19005 @findex gdb.TYPE_CODE_FLAGS
19006 @item TYPE_CODE_FLAGS
19007 A bit flags type, used for things such as status registers.
19009 @findex TYPE_CODE_FUNC
19010 @findex gdb.TYPE_CODE_FUNC
19011 @item TYPE_CODE_FUNC
19012 The type is a function.
19014 @findex TYPE_CODE_INT
19015 @findex gdb.TYPE_CODE_INT
19016 @item TYPE_CODE_INT
19017 The type is an integer type.
19019 @findex TYPE_CODE_FLT
19020 @findex gdb.TYPE_CODE_FLT
19021 @item TYPE_CODE_FLT
19022 A floating point type.
19024 @findex TYPE_CODE_VOID
19025 @findex gdb.TYPE_CODE_VOID
19026 @item TYPE_CODE_VOID
19027 The special type @code{void}.
19029 @findex TYPE_CODE_SET
19030 @findex gdb.TYPE_CODE_SET
19031 @item TYPE_CODE_SET
19034 @findex TYPE_CODE_RANGE
19035 @findex gdb.TYPE_CODE_RANGE
19036 @item TYPE_CODE_RANGE
19037 A range type, that is, an integer type with bounds.
19039 @findex TYPE_CODE_STRING
19040 @findex gdb.TYPE_CODE_STRING
19041 @item TYPE_CODE_STRING
19042 A string type. Note that this is only used for certain languages with
19043 language-defined string types; C strings are not represented this way.
19045 @findex TYPE_CODE_BITSTRING
19046 @findex gdb.TYPE_CODE_BITSTRING
19047 @item TYPE_CODE_BITSTRING
19050 @findex TYPE_CODE_ERROR
19051 @findex gdb.TYPE_CODE_ERROR
19052 @item TYPE_CODE_ERROR
19053 An unknown or erroneous type.
19055 @findex TYPE_CODE_METHOD
19056 @findex gdb.TYPE_CODE_METHOD
19057 @item TYPE_CODE_METHOD
19058 A method type, as found in C@t{++} or Java.
19060 @findex TYPE_CODE_METHODPTR
19061 @findex gdb.TYPE_CODE_METHODPTR
19062 @item TYPE_CODE_METHODPTR
19063 A pointer-to-member-function.
19065 @findex TYPE_CODE_MEMBERPTR
19066 @findex gdb.TYPE_CODE_MEMBERPTR
19067 @item TYPE_CODE_MEMBERPTR
19068 A pointer-to-member.
19070 @findex TYPE_CODE_REF
19071 @findex gdb.TYPE_CODE_REF
19072 @item TYPE_CODE_REF
19075 @findex TYPE_CODE_CHAR
19076 @findex gdb.TYPE_CODE_CHAR
19077 @item TYPE_CODE_CHAR
19080 @findex TYPE_CODE_BOOL
19081 @findex gdb.TYPE_CODE_BOOL
19082 @item TYPE_CODE_BOOL
19085 @findex TYPE_CODE_COMPLEX
19086 @findex gdb.TYPE_CODE_COMPLEX
19087 @item TYPE_CODE_COMPLEX
19088 A complex float type.
19090 @findex TYPE_CODE_TYPEDEF
19091 @findex gdb.TYPE_CODE_TYPEDEF
19092 @item TYPE_CODE_TYPEDEF
19093 A typedef to some other type.
19095 @findex TYPE_CODE_NAMESPACE
19096 @findex gdb.TYPE_CODE_NAMESPACE
19097 @item TYPE_CODE_NAMESPACE
19098 A C@t{++} namespace.
19100 @findex TYPE_CODE_DECFLOAT
19101 @findex gdb.TYPE_CODE_DECFLOAT
19102 @item TYPE_CODE_DECFLOAT
19103 A decimal floating point type.
19105 @findex TYPE_CODE_INTERNAL_FUNCTION
19106 @findex gdb.TYPE_CODE_INTERNAL_FUNCTION
19107 @item TYPE_CODE_INTERNAL_FUNCTION
19108 A function internal to @value{GDBN}. This is the type used to represent
19109 convenience functions.
19112 @node Pretty Printing
19113 @subsubsection Pretty Printing
19115 @value{GDBN} provides a mechanism to allow pretty-printing of values
19116 using Python code. The pretty-printer API allows application-specific
19117 code to greatly simplify the display of complex objects. This
19118 mechanism works for both MI and the CLI.
19120 For example, here is how a C@t{++} @code{std::string} looks without a
19124 (@value{GDBP}) print s
19126 static npos = 4294967295,
19128 <std::allocator<char>> = @{
19129 <__gnu_cxx::new_allocator<char>> = @{<No data fields>@}, <No data fields>@},
19130 members of std::basic_string<char, std::char_traits<char>, std::allocator<char> >::_Alloc_hider:
19131 _M_p = 0x804a014 "abcd"
19136 After a pretty-printer for @code{std::string} has been installed, only
19137 the contents are printed:
19140 (@value{GDBP}) print s
19144 A pretty-printer is just an object that holds a value and implements a
19145 specific interface, defined here.
19147 @defop Operation {pretty printer} children (self)
19148 @value{GDBN} will call this method on a pretty-printer to compute the
19149 children of the pretty-printer's value.
19151 This method must return an object conforming to the Python iterator
19152 protocol. Each item returned by the iterator must be a tuple holding
19153 two elements. The first element is the ``name'' of the child; the
19154 second element is the child's value. The value can be any Python
19155 object which is convertible to a @value{GDBN} value.
19157 This method is optional. If it does not exist, @value{GDBN} will act
19158 as though the value has no children.
19161 @defop Operation {pretty printer} display_hint (self)
19162 The CLI may call this method and use its result to change the
19163 formatting of a value. The result will also be supplied to an MI
19164 consumer as a @samp{displayhint} attribute of the variable being
19167 This method is optional. If it does exist, this method must return a
19170 Some display hints are predefined by @value{GDBN}:
19174 Indicate that the object being printed is ``array-like''. The CLI
19175 uses this to respect parameters such as @code{set print elements} and
19176 @code{set print array}.
19179 Indicate that the object being printed is ``map-like'', and that the
19180 children of this value can be assumed to alternate between keys and
19184 Indicate that the object being printed is ``string-like''. If the
19185 printer's @code{to_string} method returns a Python string of some
19186 kind, then @value{GDBN} will call its internal language-specific
19187 string-printing function to format the string. For the CLI this means
19188 adding quotation marks, possibly escaping some characters, respecting
19189 @code{set print elements}, and the like.
19193 @defop Operation {pretty printer} to_string (self)
19194 @value{GDBN} will call this method to display the string
19195 representation of the value passed to the object's constructor.
19197 When printing from the CLI, if the @code{to_string} method exists,
19198 then @value{GDBN} will prepend its result to the values returned by
19199 @code{children}. Exactly how this formatting is done is dependent on
19200 the display hint, and may change as more hints are added. Also,
19201 depending on the print settings (@pxref{Print Settings}), the CLI may
19202 print just the result of @code{to_string} in a stack trace, omitting
19203 the result of @code{children}.
19205 If this method returns a string, it is printed verbatim.
19207 Otherwise, if this method returns an instance of @code{gdb.Value},
19208 then @value{GDBN} prints this value. This may result in a call to
19209 another pretty-printer.
19211 If instead the method returns a Python value which is convertible to a
19212 @code{gdb.Value}, then @value{GDBN} performs the conversion and prints
19213 the resulting value. Again, this may result in a call to another
19214 pretty-printer. Python scalars (integers, floats, and booleans) and
19215 strings are convertible to @code{gdb.Value}; other types are not.
19217 If the result is not one of these types, an exception is raised.
19220 @node Selecting Pretty-Printers
19221 @subsubsection Selecting Pretty-Printers
19223 The Python list @code{gdb.pretty_printers} contains an array of
19224 functions that have been registered via addition as a pretty-printer.
19225 Each @code{gdb.Objfile} also contains a @code{pretty_printers}
19228 A function on one of these lists is passed a single @code{gdb.Value}
19229 argument and should return a pretty-printer object conforming to the
19230 interface definition above (@pxref{Pretty Printing}). If a function
19231 cannot create a pretty-printer for the value, it should return
19234 @value{GDBN} first checks the @code{pretty_printers} attribute of each
19235 @code{gdb.Objfile} and iteratively calls each function in the list for
19236 that @code{gdb.Objfile} until it receives a pretty-printer object.
19237 After these lists have been exhausted, it tries the global
19238 @code{gdb.pretty-printers} list, again calling each function until an
19239 object is returned.
19241 The order in which the objfiles are searched is not specified. For a
19242 given list, functions are always invoked from the head of the list,
19243 and iterated over sequentially until the end of the list, or a printer
19244 object is returned.
19246 Here is an example showing how a @code{std::string} printer might be
19250 class StdStringPrinter:
19251 "Print a std::string"
19253 def __init__ (self, val):
19256 def to_string (self):
19257 return self.val['_M_dataplus']['_M_p']
19259 def display_hint (self):
19263 And here is an example showing how a lookup function for the printer
19264 example above might be written.
19267 def str_lookup_function (val):
19269 lookup_tag = val.type.tag
19270 regex = re.compile ("^std::basic_string<char,.*>$")
19271 if lookup_tag == None:
19273 if regex.match (lookup_tag):
19274 return StdStringPrinter (val)
19279 The example lookup function extracts the value's type, and attempts to
19280 match it to a type that it can pretty-print. If it is a type the
19281 printer can pretty-print, it will return a printer object. If not, it
19282 returns @code{None}.
19284 We recommend that you put your core pretty-printers into a Python
19285 package. If your pretty-printers are for use with a library, we
19286 further recommend embedding a version number into the package name.
19287 This practice will enable @value{GDBN} to load multiple versions of
19288 your pretty-printers at the same time, because they will have
19291 You should write auto-loaded code (@pxref{Auto-loading}) such that it
19292 can be evaluated multiple times without changing its meaning. An
19293 ideal auto-load file will consist solely of @code{import}s of your
19294 printer modules, followed by a call to a register pretty-printers with
19295 the current objfile.
19297 Taken as a whole, this approach will scale nicely to multiple
19298 inferiors, each potentially using a different library version.
19299 Embedding a version number in the Python package name will ensure that
19300 @value{GDBN} is able to load both sets of printers simultaneously.
19301 Then, because the search for pretty-printers is done by objfile, and
19302 because your auto-loaded code took care to register your library's
19303 printers with a specific objfile, @value{GDBN} will find the correct
19304 printers for the specific version of the library used by each
19307 To continue the @code{std::string} example (@pxref{Pretty Printing}),
19308 this code might appear in @code{gdb.libstdcxx.v6}:
19311 def register_printers (objfile):
19312 objfile.pretty_printers.add (str_lookup_function)
19316 And then the corresponding contents of the auto-load file would be:
19319 import gdb.libstdcxx.v6
19320 gdb.libstdcxx.v6.register_printers (gdb.current_objfile ())
19323 @node Commands In Python
19324 @subsubsection Commands In Python
19326 @cindex commands in python
19327 @cindex python commands
19328 You can implement new @value{GDBN} CLI commands in Python. A CLI
19329 command is implemented using an instance of the @code{gdb.Command}
19330 class, most commonly using a subclass.
19332 @defmethod Command __init__ name @var{command_class} @r{[}@var{completer_class}@r{]} @r{[}@var{prefix}@r{]}
19333 The object initializer for @code{Command} registers the new command
19334 with @value{GDBN}. This initializer is normally invoked from the
19335 subclass' own @code{__init__} method.
19337 @var{name} is the name of the command. If @var{name} consists of
19338 multiple words, then the initial words are looked for as prefix
19339 commands. In this case, if one of the prefix commands does not exist,
19340 an exception is raised.
19342 There is no support for multi-line commands.
19344 @var{command_class} should be one of the @samp{COMMAND_} constants
19345 defined below. This argument tells @value{GDBN} how to categorize the
19346 new command in the help system.
19348 @var{completer_class} is an optional argument. If given, it should be
19349 one of the @samp{COMPLETE_} constants defined below. This argument
19350 tells @value{GDBN} how to perform completion for this command. If not
19351 given, @value{GDBN} will attempt to complete using the object's
19352 @code{complete} method (see below); if no such method is found, an
19353 error will occur when completion is attempted.
19355 @var{prefix} is an optional argument. If @code{True}, then the new
19356 command is a prefix command; sub-commands of this command may be
19359 The help text for the new command is taken from the Python
19360 documentation string for the command's class, if there is one. If no
19361 documentation string is provided, the default value ``This command is
19362 not documented.'' is used.
19365 @cindex don't repeat Python command
19366 @defmethod Command dont_repeat
19367 By default, a @value{GDBN} command is repeated when the user enters a
19368 blank line at the command prompt. A command can suppress this
19369 behavior by invoking the @code{dont_repeat} method. This is similar
19370 to the user command @code{dont-repeat}, see @ref{Define, dont-repeat}.
19373 @defmethod Command invoke argument from_tty
19374 This method is called by @value{GDBN} when this command is invoked.
19376 @var{argument} is a string. It is the argument to the command, after
19377 leading and trailing whitespace has been stripped.
19379 @var{from_tty} is a boolean argument. When true, this means that the
19380 command was entered by the user at the terminal; when false it means
19381 that the command came from elsewhere.
19383 If this method throws an exception, it is turned into a @value{GDBN}
19384 @code{error} call. Otherwise, the return value is ignored.
19387 @cindex completion of Python commands
19388 @defmethod Command complete text word
19389 This method is called by @value{GDBN} when the user attempts
19390 completion on this command. All forms of completion are handled by
19391 this method, that is, the @key{TAB} and @key{M-?} key bindings
19392 (@pxref{Completion}), and the @code{complete} command (@pxref{Help,
19395 The arguments @var{text} and @var{word} are both strings. @var{text}
19396 holds the complete command line up to the cursor's location.
19397 @var{word} holds the last word of the command line; this is computed
19398 using a word-breaking heuristic.
19400 The @code{complete} method can return several values:
19403 If the return value is a sequence, the contents of the sequence are
19404 used as the completions. It is up to @code{complete} to ensure that the
19405 contents actually do complete the word. A zero-length sequence is
19406 allowed, it means that there were no completions available. Only
19407 string elements of the sequence are used; other elements in the
19408 sequence are ignored.
19411 If the return value is one of the @samp{COMPLETE_} constants defined
19412 below, then the corresponding @value{GDBN}-internal completion
19413 function is invoked, and its result is used.
19416 All other results are treated as though there were no available
19421 When a new command is registered, it must be declared as a member of
19422 some general class of commands. This is used to classify top-level
19423 commands in the on-line help system; note that prefix commands are not
19424 listed under their own category but rather that of their top-level
19425 command. The available classifications are represented by constants
19426 defined in the @code{gdb} module:
19429 @findex COMMAND_NONE
19430 @findex gdb.COMMAND_NONE
19432 The command does not belong to any particular class. A command in
19433 this category will not be displayed in any of the help categories.
19435 @findex COMMAND_RUNNING
19436 @findex gdb.COMMAND_RUNNING
19437 @item COMMAND_RUNNING
19438 The command is related to running the inferior. For example,
19439 @code{start}, @code{step}, and @code{continue} are in this category.
19440 Type @kbd{help running} at the @value{GDBN} prompt to see a list of
19441 commands in this category.
19443 @findex COMMAND_DATA
19444 @findex gdb.COMMAND_DATA
19446 The command is related to data or variables. For example,
19447 @code{call}, @code{find}, and @code{print} are in this category. Type
19448 @kbd{help data} at the @value{GDBN} prompt to see a list of commands
19451 @findex COMMAND_STACK
19452 @findex gdb.COMMAND_STACK
19453 @item COMMAND_STACK
19454 The command has to do with manipulation of the stack. For example,
19455 @code{backtrace}, @code{frame}, and @code{return} are in this
19456 category. Type @kbd{help stack} at the @value{GDBN} prompt to see a
19457 list of commands in this category.
19459 @findex COMMAND_FILES
19460 @findex gdb.COMMAND_FILES
19461 @item COMMAND_FILES
19462 This class is used for file-related commands. For example,
19463 @code{file}, @code{list} and @code{section} are in this category.
19464 Type @kbd{help files} at the @value{GDBN} prompt to see a list of
19465 commands in this category.
19467 @findex COMMAND_SUPPORT
19468 @findex gdb.COMMAND_SUPPORT
19469 @item COMMAND_SUPPORT
19470 This should be used for ``support facilities'', generally meaning
19471 things that are useful to the user when interacting with @value{GDBN},
19472 but not related to the state of the inferior. For example,
19473 @code{help}, @code{make}, and @code{shell} are in this category. Type
19474 @kbd{help support} at the @value{GDBN} prompt to see a list of
19475 commands in this category.
19477 @findex COMMAND_STATUS
19478 @findex gdb.COMMAND_STATUS
19479 @item COMMAND_STATUS
19480 The command is an @samp{info}-related command, that is, related to the
19481 state of @value{GDBN} itself. For example, @code{info}, @code{macro},
19482 and @code{show} are in this category. Type @kbd{help status} at the
19483 @value{GDBN} prompt to see a list of commands in this category.
19485 @findex COMMAND_BREAKPOINTS
19486 @findex gdb.COMMAND_BREAKPOINTS
19487 @item COMMAND_BREAKPOINTS
19488 The command has to do with breakpoints. For example, @code{break},
19489 @code{clear}, and @code{delete} are in this category. Type @kbd{help
19490 breakpoints} at the @value{GDBN} prompt to see a list of commands in
19493 @findex COMMAND_TRACEPOINTS
19494 @findex gdb.COMMAND_TRACEPOINTS
19495 @item COMMAND_TRACEPOINTS
19496 The command has to do with tracepoints. For example, @code{trace},
19497 @code{actions}, and @code{tfind} are in this category. Type
19498 @kbd{help tracepoints} at the @value{GDBN} prompt to see a list of
19499 commands in this category.
19501 @findex COMMAND_OBSCURE
19502 @findex gdb.COMMAND_OBSCURE
19503 @item COMMAND_OBSCURE
19504 The command is only used in unusual circumstances, or is not of
19505 general interest to users. For example, @code{checkpoint},
19506 @code{fork}, and @code{stop} are in this category. Type @kbd{help
19507 obscure} at the @value{GDBN} prompt to see a list of commands in this
19510 @findex COMMAND_MAINTENANCE
19511 @findex gdb.COMMAND_MAINTENANCE
19512 @item COMMAND_MAINTENANCE
19513 The command is only useful to @value{GDBN} maintainers. The
19514 @code{maintenance} and @code{flushregs} commands are in this category.
19515 Type @kbd{help internals} at the @value{GDBN} prompt to see a list of
19516 commands in this category.
19519 A new command can use a predefined completion function, either by
19520 specifying it via an argument at initialization, or by returning it
19521 from the @code{complete} method. These predefined completion
19522 constants are all defined in the @code{gdb} module:
19525 @findex COMPLETE_NONE
19526 @findex gdb.COMPLETE_NONE
19527 @item COMPLETE_NONE
19528 This constant means that no completion should be done.
19530 @findex COMPLETE_FILENAME
19531 @findex gdb.COMPLETE_FILENAME
19532 @item COMPLETE_FILENAME
19533 This constant means that filename completion should be performed.
19535 @findex COMPLETE_LOCATION
19536 @findex gdb.COMPLETE_LOCATION
19537 @item COMPLETE_LOCATION
19538 This constant means that location completion should be done.
19539 @xref{Specify Location}.
19541 @findex COMPLETE_COMMAND
19542 @findex gdb.COMPLETE_COMMAND
19543 @item COMPLETE_COMMAND
19544 This constant means that completion should examine @value{GDBN}
19547 @findex COMPLETE_SYMBOL
19548 @findex gdb.COMPLETE_SYMBOL
19549 @item COMPLETE_SYMBOL
19550 This constant means that completion should be done using symbol names
19554 The following code snippet shows how a trivial CLI command can be
19555 implemented in Python:
19558 class HelloWorld (gdb.Command):
19559 """Greet the whole world."""
19561 def __init__ (self):
19562 super (HelloWorld, self).__init__ ("hello-world", gdb.COMMAND_OBSCURE)
19564 def invoke (self, arg, from_tty):
19565 print "Hello, World!"
19570 The last line instantiates the class, and is necessary to trigger the
19571 registration of the command with @value{GDBN}. Depending on how the
19572 Python code is read into @value{GDBN}, you may need to import the
19573 @code{gdb} module explicitly.
19575 @node Functions In Python
19576 @subsubsection Writing new convenience functions
19578 @cindex writing convenience functions
19579 @cindex convenience functions in python
19580 @cindex python convenience functions
19581 @tindex gdb.Function
19583 You can implement new convenience functions (@pxref{Convenience Vars})
19584 in Python. A convenience function is an instance of a subclass of the
19585 class @code{gdb.Function}.
19587 @defmethod Function __init__ name
19588 The initializer for @code{Function} registers the new function with
19589 @value{GDBN}. The argument @var{name} is the name of the function,
19590 a string. The function will be visible to the user as a convenience
19591 variable of type @code{internal function}, whose name is the same as
19592 the given @var{name}.
19594 The documentation for the new function is taken from the documentation
19595 string for the new class.
19598 @defmethod Function invoke @var{*args}
19599 When a convenience function is evaluated, its arguments are converted
19600 to instances of @code{gdb.Value}, and then the function's
19601 @code{invoke} method is called. Note that @value{GDBN} does not
19602 predetermine the arity of convenience functions. Instead, all
19603 available arguments are passed to @code{invoke}, following the
19604 standard Python calling convention. In particular, a convenience
19605 function can have default values for parameters without ill effect.
19607 The return value of this method is used as its value in the enclosing
19608 expression. If an ordinary Python value is returned, it is converted
19609 to a @code{gdb.Value} following the usual rules.
19612 The following code snippet shows how a trivial convenience function can
19613 be implemented in Python:
19616 class Greet (gdb.Function):
19617 """Return string to greet someone.
19618 Takes a name as argument."""
19620 def __init__ (self):
19621 super (Greet, self).__init__ ("greet")
19623 def invoke (self, name):
19624 return "Hello, %s!" % name.string ()
19629 The last line instantiates the class, and is necessary to trigger the
19630 registration of the function with @value{GDBN}. Depending on how the
19631 Python code is read into @value{GDBN}, you may need to import the
19632 @code{gdb} module explicitly.
19634 @node Objfiles In Python
19635 @subsubsection Objfiles In Python
19637 @cindex objfiles in python
19638 @tindex gdb.Objfile
19640 @value{GDBN} loads symbols for an inferior from various
19641 symbol-containing files (@pxref{Files}). These include the primary
19642 executable file, any shared libraries used by the inferior, and any
19643 separate debug info files (@pxref{Separate Debug Files}).
19644 @value{GDBN} calls these symbol-containing files @dfn{objfiles}.
19646 The following objfile-related functions are available in the
19649 @findex gdb.current_objfile
19650 @defun current_objfile
19651 When auto-loading a Python script (@pxref{Auto-loading}), @value{GDBN}
19652 sets the ``current objfile'' to the corresponding objfile. This
19653 function returns the current objfile. If there is no current objfile,
19654 this function returns @code{None}.
19657 @findex gdb.objfiles
19659 Return a sequence of all the objfiles current known to @value{GDBN}.
19660 @xref{Objfiles In Python}.
19663 Each objfile is represented by an instance of the @code{gdb.Objfile}
19666 @defivar Objfile filename
19667 The file name of the objfile as a string.
19670 @defivar Objfile pretty_printers
19671 The @code{pretty_printers} attribute is a list of functions. It is
19672 used to look up pretty-printers. A @code{Value} is passed to each
19673 function in order; if the function returns @code{None}, then the
19674 search continues. Otherwise, the return value should be an object
19675 which is used to format the value. @xref{Pretty Printing}, for more
19679 @node Frames In Python
19680 @subsubsection Acessing inferior stack frames from Python.
19682 @cindex frames in python
19683 When the debugged program stops, @value{GDBN} is able to analyze its call
19684 stack (@pxref{Frames,,Stack frames}). The @code{gdb.Frame} class
19685 represents a frame in the stack. A @code{gdb.Frame} object is only valid
19686 while its corresponding frame exists in the inferior's stack. If you try
19687 to use an invalid frame object, @value{GDBN} will throw a @code{RuntimeError}
19690 Two @code{gdb.Frame} objects can be compared for equality with the @code{==}
19694 (@value{GDBP}) python print gdb.newest_frame() == gdb.selected_frame ()
19698 The following frame-related functions are available in the @code{gdb} module:
19700 @findex gdb.selected_frame
19701 @defun selected_frame
19702 Return the selected frame object. (@pxref{Selection,,Selecting a Frame}).
19705 @defun frame_stop_reason_string reason
19706 Return a string explaining the reason why @value{GDBN} stopped unwinding
19707 frames, as expressed by the given @var{reason} code (an integer, see the
19708 @code{unwind_stop_reason} method further down in this section).
19711 A @code{gdb.Frame} object has the following methods:
19714 @defmethod Frame is_valid
19715 Returns true if the @code{gdb.Frame} object is valid, false if not.
19716 A frame object can become invalid if the frame it refers to doesn't
19717 exist anymore in the inferior. All @code{gdb.Frame} methods will throw
19718 an exception if it is invalid at the time the method is called.
19721 @defmethod Frame name
19722 Returns the function name of the frame, or @code{None} if it can't be
19726 @defmethod Frame type
19727 Returns the type of the frame. The value can be one of
19728 @code{gdb.NORMAL_FRAME}, @code{gdb.DUMMY_FRAME}, @code{gdb.SIGTRAMP_FRAME}
19729 or @code{gdb.SENTINEL_FRAME}.
19732 @defmethod Frame unwind_stop_reason
19733 Return an integer representing the reason why it's not possible to find
19734 more frames toward the outermost frame. Use
19735 @code{gdb.frame_stop_reason_string} to convert the value returned by this
19736 function to a string.
19739 @defmethod Frame pc
19740 Returns the frame's resume address.
19743 @defmethod Frame older
19744 Return the frame that called this frame.
19747 @defmethod Frame newer
19748 Return the frame called by this frame.
19751 @defmethod Frame read_var variable
19752 Return the value of the given variable in this frame. @var{variable} must
19758 @chapter Command Interpreters
19759 @cindex command interpreters
19761 @value{GDBN} supports multiple command interpreters, and some command
19762 infrastructure to allow users or user interface writers to switch
19763 between interpreters or run commands in other interpreters.
19765 @value{GDBN} currently supports two command interpreters, the console
19766 interpreter (sometimes called the command-line interpreter or @sc{cli})
19767 and the machine interface interpreter (or @sc{gdb/mi}). This manual
19768 describes both of these interfaces in great detail.
19770 By default, @value{GDBN} will start with the console interpreter.
19771 However, the user may choose to start @value{GDBN} with another
19772 interpreter by specifying the @option{-i} or @option{--interpreter}
19773 startup options. Defined interpreters include:
19777 @cindex console interpreter
19778 The traditional console or command-line interpreter. This is the most often
19779 used interpreter with @value{GDBN}. With no interpreter specified at runtime,
19780 @value{GDBN} will use this interpreter.
19783 @cindex mi interpreter
19784 The newest @sc{gdb/mi} interface (currently @code{mi2}). Used primarily
19785 by programs wishing to use @value{GDBN} as a backend for a debugger GUI
19786 or an IDE. For more information, see @ref{GDB/MI, ,The @sc{gdb/mi}
19790 @cindex mi2 interpreter
19791 The current @sc{gdb/mi} interface.
19794 @cindex mi1 interpreter
19795 The @sc{gdb/mi} interface included in @value{GDBN} 5.1, 5.2, and 5.3.
19799 @cindex invoke another interpreter
19800 The interpreter being used by @value{GDBN} may not be dynamically
19801 switched at runtime. Although possible, this could lead to a very
19802 precarious situation. Consider an IDE using @sc{gdb/mi}. If a user
19803 enters the command "interpreter-set console" in a console view,
19804 @value{GDBN} would switch to using the console interpreter, rendering
19805 the IDE inoperable!
19807 @kindex interpreter-exec
19808 Although you may only choose a single interpreter at startup, you may execute
19809 commands in any interpreter from the current interpreter using the appropriate
19810 command. If you are running the console interpreter, simply use the
19811 @code{interpreter-exec} command:
19814 interpreter-exec mi "-data-list-register-names"
19817 @sc{gdb/mi} has a similar command, although it is only available in versions of
19818 @value{GDBN} which support @sc{gdb/mi} version 2 (or greater).
19821 @chapter @value{GDBN} Text User Interface
19823 @cindex Text User Interface
19826 * TUI Overview:: TUI overview
19827 * TUI Keys:: TUI key bindings
19828 * TUI Single Key Mode:: TUI single key mode
19829 * TUI Commands:: TUI-specific commands
19830 * TUI Configuration:: TUI configuration variables
19833 The @value{GDBN} Text User Interface (TUI) is a terminal
19834 interface which uses the @code{curses} library to show the source
19835 file, the assembly output, the program registers and @value{GDBN}
19836 commands in separate text windows. The TUI mode is supported only
19837 on platforms where a suitable version of the @code{curses} library
19840 @pindex @value{GDBTUI}
19841 The TUI mode is enabled by default when you invoke @value{GDBN} as
19842 either @samp{@value{GDBTUI}} or @samp{@value{GDBP} -tui}.
19843 You can also switch in and out of TUI mode while @value{GDBN} runs by
19844 using various TUI commands and key bindings, such as @kbd{C-x C-a}.
19845 @xref{TUI Keys, ,TUI Key Bindings}.
19848 @section TUI Overview
19850 In TUI mode, @value{GDBN} can display several text windows:
19854 This window is the @value{GDBN} command window with the @value{GDBN}
19855 prompt and the @value{GDBN} output. The @value{GDBN} input is still
19856 managed using readline.
19859 The source window shows the source file of the program. The current
19860 line and active breakpoints are displayed in this window.
19863 The assembly window shows the disassembly output of the program.
19866 This window shows the processor registers. Registers are highlighted
19867 when their values change.
19870 The source and assembly windows show the current program position
19871 by highlighting the current line and marking it with a @samp{>} marker.
19872 Breakpoints are indicated with two markers. The first marker
19873 indicates the breakpoint type:
19877 Breakpoint which was hit at least once.
19880 Breakpoint which was never hit.
19883 Hardware breakpoint which was hit at least once.
19886 Hardware breakpoint which was never hit.
19889 The second marker indicates whether the breakpoint is enabled or not:
19893 Breakpoint is enabled.
19896 Breakpoint is disabled.
19899 The source, assembly and register windows are updated when the current
19900 thread changes, when the frame changes, or when the program counter
19903 These windows are not all visible at the same time. The command
19904 window is always visible. The others can be arranged in several
19915 source and assembly,
19918 source and registers, or
19921 assembly and registers.
19924 A status line above the command window shows the following information:
19928 Indicates the current @value{GDBN} target.
19929 (@pxref{Targets, ,Specifying a Debugging Target}).
19932 Gives the current process or thread number.
19933 When no process is being debugged, this field is set to @code{No process}.
19936 Gives the current function name for the selected frame.
19937 The name is demangled if demangling is turned on (@pxref{Print Settings}).
19938 When there is no symbol corresponding to the current program counter,
19939 the string @code{??} is displayed.
19942 Indicates the current line number for the selected frame.
19943 When the current line number is not known, the string @code{??} is displayed.
19946 Indicates the current program counter address.
19950 @section TUI Key Bindings
19951 @cindex TUI key bindings
19953 The TUI installs several key bindings in the readline keymaps
19954 (@pxref{Command Line Editing}). The following key bindings
19955 are installed for both TUI mode and the @value{GDBN} standard mode.
19964 Enter or leave the TUI mode. When leaving the TUI mode,
19965 the curses window management stops and @value{GDBN} operates using
19966 its standard mode, writing on the terminal directly. When reentering
19967 the TUI mode, control is given back to the curses windows.
19968 The screen is then refreshed.
19972 Use a TUI layout with only one window. The layout will
19973 either be @samp{source} or @samp{assembly}. When the TUI mode
19974 is not active, it will switch to the TUI mode.
19976 Think of this key binding as the Emacs @kbd{C-x 1} binding.
19980 Use a TUI layout with at least two windows. When the current
19981 layout already has two windows, the next layout with two windows is used.
19982 When a new layout is chosen, one window will always be common to the
19983 previous layout and the new one.
19985 Think of it as the Emacs @kbd{C-x 2} binding.
19989 Change the active window. The TUI associates several key bindings
19990 (like scrolling and arrow keys) with the active window. This command
19991 gives the focus to the next TUI window.
19993 Think of it as the Emacs @kbd{C-x o} binding.
19997 Switch in and out of the TUI SingleKey mode that binds single
19998 keys to @value{GDBN} commands (@pxref{TUI Single Key Mode}).
20001 The following key bindings only work in the TUI mode:
20006 Scroll the active window one page up.
20010 Scroll the active window one page down.
20014 Scroll the active window one line up.
20018 Scroll the active window one line down.
20022 Scroll the active window one column left.
20026 Scroll the active window one column right.
20030 Refresh the screen.
20033 Because the arrow keys scroll the active window in the TUI mode, they
20034 are not available for their normal use by readline unless the command
20035 window has the focus. When another window is active, you must use
20036 other readline key bindings such as @kbd{C-p}, @kbd{C-n}, @kbd{C-b}
20037 and @kbd{C-f} to control the command window.
20039 @node TUI Single Key Mode
20040 @section TUI Single Key Mode
20041 @cindex TUI single key mode
20043 The TUI also provides a @dfn{SingleKey} mode, which binds several
20044 frequently used @value{GDBN} commands to single keys. Type @kbd{C-x s} to
20045 switch into this mode, where the following key bindings are used:
20048 @kindex c @r{(SingleKey TUI key)}
20052 @kindex d @r{(SingleKey TUI key)}
20056 @kindex f @r{(SingleKey TUI key)}
20060 @kindex n @r{(SingleKey TUI key)}
20064 @kindex q @r{(SingleKey TUI key)}
20066 exit the SingleKey mode.
20068 @kindex r @r{(SingleKey TUI key)}
20072 @kindex s @r{(SingleKey TUI key)}
20076 @kindex u @r{(SingleKey TUI key)}
20080 @kindex v @r{(SingleKey TUI key)}
20084 @kindex w @r{(SingleKey TUI key)}
20089 Other keys temporarily switch to the @value{GDBN} command prompt.
20090 The key that was pressed is inserted in the editing buffer so that
20091 it is possible to type most @value{GDBN} commands without interaction
20092 with the TUI SingleKey mode. Once the command is entered the TUI
20093 SingleKey mode is restored. The only way to permanently leave
20094 this mode is by typing @kbd{q} or @kbd{C-x s}.
20098 @section TUI-specific Commands
20099 @cindex TUI commands
20101 The TUI has specific commands to control the text windows.
20102 These commands are always available, even when @value{GDBN} is not in
20103 the TUI mode. When @value{GDBN} is in the standard mode, most
20104 of these commands will automatically switch to the TUI mode.
20109 List and give the size of all displayed windows.
20113 Display the next layout.
20116 Display the previous layout.
20119 Display the source window only.
20122 Display the assembly window only.
20125 Display the source and assembly window.
20128 Display the register window together with the source or assembly window.
20132 Make the next window active for scrolling.
20135 Make the previous window active for scrolling.
20138 Make the source window active for scrolling.
20141 Make the assembly window active for scrolling.
20144 Make the register window active for scrolling.
20147 Make the command window active for scrolling.
20151 Refresh the screen. This is similar to typing @kbd{C-L}.
20153 @item tui reg float
20155 Show the floating point registers in the register window.
20157 @item tui reg general
20158 Show the general registers in the register window.
20161 Show the next register group. The list of register groups as well as
20162 their order is target specific. The predefined register groups are the
20163 following: @code{general}, @code{float}, @code{system}, @code{vector},
20164 @code{all}, @code{save}, @code{restore}.
20166 @item tui reg system
20167 Show the system registers in the register window.
20171 Update the source window and the current execution point.
20173 @item winheight @var{name} +@var{count}
20174 @itemx winheight @var{name} -@var{count}
20176 Change the height of the window @var{name} by @var{count}
20177 lines. Positive counts increase the height, while negative counts
20180 @item tabset @var{nchars}
20182 Set the width of tab stops to be @var{nchars} characters.
20185 @node TUI Configuration
20186 @section TUI Configuration Variables
20187 @cindex TUI configuration variables
20189 Several configuration variables control the appearance of TUI windows.
20192 @item set tui border-kind @var{kind}
20193 @kindex set tui border-kind
20194 Select the border appearance for the source, assembly and register windows.
20195 The possible values are the following:
20198 Use a space character to draw the border.
20201 Use @sc{ascii} characters @samp{+}, @samp{-} and @samp{|} to draw the border.
20204 Use the Alternate Character Set to draw the border. The border is
20205 drawn using character line graphics if the terminal supports them.
20208 @item set tui border-mode @var{mode}
20209 @kindex set tui border-mode
20210 @itemx set tui active-border-mode @var{mode}
20211 @kindex set tui active-border-mode
20212 Select the display attributes for the borders of the inactive windows
20213 or the active window. The @var{mode} can be one of the following:
20216 Use normal attributes to display the border.
20222 Use reverse video mode.
20225 Use half bright mode.
20227 @item half-standout
20228 Use half bright and standout mode.
20231 Use extra bright or bold mode.
20233 @item bold-standout
20234 Use extra bright or bold and standout mode.
20239 @chapter Using @value{GDBN} under @sc{gnu} Emacs
20242 @cindex @sc{gnu} Emacs
20243 A special interface allows you to use @sc{gnu} Emacs to view (and
20244 edit) the source files for the program you are debugging with
20247 To use this interface, use the command @kbd{M-x gdb} in Emacs. Give the
20248 executable file you want to debug as an argument. This command starts
20249 @value{GDBN} as a subprocess of Emacs, with input and output through a newly
20250 created Emacs buffer.
20251 @c (Do not use the @code{-tui} option to run @value{GDBN} from Emacs.)
20253 Running @value{GDBN} under Emacs can be just like running @value{GDBN} normally except for two
20258 All ``terminal'' input and output goes through an Emacs buffer, called
20261 This applies both to @value{GDBN} commands and their output, and to the input
20262 and output done by the program you are debugging.
20264 This is useful because it means that you can copy the text of previous
20265 commands and input them again; you can even use parts of the output
20268 All the facilities of Emacs' Shell mode are available for interacting
20269 with your program. In particular, you can send signals the usual
20270 way---for example, @kbd{C-c C-c} for an interrupt, @kbd{C-c C-z} for a
20274 @value{GDBN} displays source code through Emacs.
20276 Each time @value{GDBN} displays a stack frame, Emacs automatically finds the
20277 source file for that frame and puts an arrow (@samp{=>}) at the
20278 left margin of the current line. Emacs uses a separate buffer for
20279 source display, and splits the screen to show both your @value{GDBN} session
20282 Explicit @value{GDBN} @code{list} or search commands still produce output as
20283 usual, but you probably have no reason to use them from Emacs.
20286 We call this @dfn{text command mode}. Emacs 22.1, and later, also uses
20287 a graphical mode, enabled by default, which provides further buffers
20288 that can control the execution and describe the state of your program.
20289 @xref{GDB Graphical Interface,,, Emacs, The @sc{gnu} Emacs Manual}.
20291 If you specify an absolute file name when prompted for the @kbd{M-x
20292 gdb} argument, then Emacs sets your current working directory to where
20293 your program resides. If you only specify the file name, then Emacs
20294 sets your current working directory to to the directory associated
20295 with the previous buffer. In this case, @value{GDBN} may find your
20296 program by searching your environment's @code{PATH} variable, but on
20297 some operating systems it might not find the source. So, although the
20298 @value{GDBN} input and output session proceeds normally, the auxiliary
20299 buffer does not display the current source and line of execution.
20301 The initial working directory of @value{GDBN} is printed on the top
20302 line of the GUD buffer and this serves as a default for the commands
20303 that specify files for @value{GDBN} to operate on. @xref{Files,
20304 ,Commands to Specify Files}.
20306 By default, @kbd{M-x gdb} calls the program called @file{gdb}. If you
20307 need to call @value{GDBN} by a different name (for example, if you
20308 keep several configurations around, with different names) you can
20309 customize the Emacs variable @code{gud-gdb-command-name} to run the
20312 In the GUD buffer, you can use these special Emacs commands in
20313 addition to the standard Shell mode commands:
20317 Describe the features of Emacs' GUD Mode.
20320 Execute to another source line, like the @value{GDBN} @code{step} command; also
20321 update the display window to show the current file and location.
20324 Execute to next source line in this function, skipping all function
20325 calls, like the @value{GDBN} @code{next} command. Then update the display window
20326 to show the current file and location.
20329 Execute one instruction, like the @value{GDBN} @code{stepi} command; update
20330 display window accordingly.
20333 Execute until exit from the selected stack frame, like the @value{GDBN}
20334 @code{finish} command.
20337 Continue execution of your program, like the @value{GDBN} @code{continue}
20341 Go up the number of frames indicated by the numeric argument
20342 (@pxref{Arguments, , Numeric Arguments, Emacs, The @sc{gnu} Emacs Manual}),
20343 like the @value{GDBN} @code{up} command.
20346 Go down the number of frames indicated by the numeric argument, like the
20347 @value{GDBN} @code{down} command.
20350 In any source file, the Emacs command @kbd{C-x @key{SPC}} (@code{gud-break})
20351 tells @value{GDBN} to set a breakpoint on the source line point is on.
20353 In text command mode, if you type @kbd{M-x speedbar}, Emacs displays a
20354 separate frame which shows a backtrace when the GUD buffer is current.
20355 Move point to any frame in the stack and type @key{RET} to make it
20356 become the current frame and display the associated source in the
20357 source buffer. Alternatively, click @kbd{Mouse-2} to make the
20358 selected frame become the current one. In graphical mode, the
20359 speedbar displays watch expressions.
20361 If you accidentally delete the source-display buffer, an easy way to get
20362 it back is to type the command @code{f} in the @value{GDBN} buffer, to
20363 request a frame display; when you run under Emacs, this recreates
20364 the source buffer if necessary to show you the context of the current
20367 The source files displayed in Emacs are in ordinary Emacs buffers
20368 which are visiting the source files in the usual way. You can edit
20369 the files with these buffers if you wish; but keep in mind that @value{GDBN}
20370 communicates with Emacs in terms of line numbers. If you add or
20371 delete lines from the text, the line numbers that @value{GDBN} knows cease
20372 to correspond properly with the code.
20374 A more detailed description of Emacs' interaction with @value{GDBN} is
20375 given in the Emacs manual (@pxref{Debuggers,,, Emacs, The @sc{gnu}
20378 @c The following dropped because Epoch is nonstandard. Reactivate
20379 @c if/when v19 does something similar. ---doc@cygnus.com 19dec1990
20381 @kindex Emacs Epoch environment
20385 Version 18 of @sc{gnu} Emacs has a built-in window system
20386 called the @code{epoch}
20387 environment. Users of this environment can use a new command,
20388 @code{inspect} which performs identically to @code{print} except that
20389 each value is printed in its own window.
20394 @chapter The @sc{gdb/mi} Interface
20396 @unnumberedsec Function and Purpose
20398 @cindex @sc{gdb/mi}, its purpose
20399 @sc{gdb/mi} is a line based machine oriented text interface to
20400 @value{GDBN} and is activated by specifying using the
20401 @option{--interpreter} command line option (@pxref{Mode Options}). It
20402 is specifically intended to support the development of systems which
20403 use the debugger as just one small component of a larger system.
20405 This chapter is a specification of the @sc{gdb/mi} interface. It is written
20406 in the form of a reference manual.
20408 Note that @sc{gdb/mi} is still under construction, so some of the
20409 features described below are incomplete and subject to change
20410 (@pxref{GDB/MI Development and Front Ends, , @sc{gdb/mi} Development and Front Ends}).
20412 @unnumberedsec Notation and Terminology
20414 @cindex notational conventions, for @sc{gdb/mi}
20415 This chapter uses the following notation:
20419 @code{|} separates two alternatives.
20422 @code{[ @var{something} ]} indicates that @var{something} is optional:
20423 it may or may not be given.
20426 @code{( @var{group} )*} means that @var{group} inside the parentheses
20427 may repeat zero or more times.
20430 @code{( @var{group} )+} means that @var{group} inside the parentheses
20431 may repeat one or more times.
20434 @code{"@var{string}"} means a literal @var{string}.
20438 @heading Dependencies
20442 * GDB/MI General Design::
20443 * GDB/MI Command Syntax::
20444 * GDB/MI Compatibility with CLI::
20445 * GDB/MI Development and Front Ends::
20446 * GDB/MI Output Records::
20447 * GDB/MI Simple Examples::
20448 * GDB/MI Command Description Format::
20449 * GDB/MI Breakpoint Commands::
20450 * GDB/MI Program Context::
20451 * GDB/MI Thread Commands::
20452 * GDB/MI Program Execution::
20453 * GDB/MI Stack Manipulation::
20454 * GDB/MI Variable Objects::
20455 * GDB/MI Data Manipulation::
20456 * GDB/MI Tracepoint Commands::
20457 * GDB/MI Symbol Query::
20458 * GDB/MI File Commands::
20460 * GDB/MI Kod Commands::
20461 * GDB/MI Memory Overlay Commands::
20462 * GDB/MI Signal Handling Commands::
20464 * GDB/MI Target Manipulation::
20465 * GDB/MI File Transfer Commands::
20466 * GDB/MI Miscellaneous Commands::
20469 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
20470 @node GDB/MI General Design
20471 @section @sc{gdb/mi} General Design
20472 @cindex GDB/MI General Design
20474 Interaction of a @sc{GDB/MI} frontend with @value{GDBN} involves three
20475 parts---commands sent to @value{GDBN}, responses to those commands
20476 and notifications. Each command results in exactly one response,
20477 indicating either successful completion of the command, or an error.
20478 For the commands that do not resume the target, the response contains the
20479 requested information. For the commands that resume the target, the
20480 response only indicates whether the target was successfully resumed.
20481 Notifications is the mechanism for reporting changes in the state of the
20482 target, or in @value{GDBN} state, that cannot conveniently be associated with
20483 a command and reported as part of that command response.
20485 The important examples of notifications are:
20489 Exec notifications. These are used to report changes in
20490 target state---when a target is resumed, or stopped. It would not
20491 be feasible to include this information in response of resuming
20492 commands, because one resume commands can result in multiple events in
20493 different threads. Also, quite some time may pass before any event
20494 happens in the target, while a frontend needs to know whether the resuming
20495 command itself was successfully executed.
20498 Console output, and status notifications. Console output
20499 notifications are used to report output of CLI commands, as well as
20500 diagnostics for other commands. Status notifications are used to
20501 report the progress of a long-running operation. Naturally, including
20502 this information in command response would mean no output is produced
20503 until the command is finished, which is undesirable.
20506 General notifications. Commands may have various side effects on
20507 the @value{GDBN} or target state beyond their official purpose. For example,
20508 a command may change the selected thread. Although such changes can
20509 be included in command response, using notification allows for more
20510 orthogonal frontend design.
20514 There's no guarantee that whenever an MI command reports an error,
20515 @value{GDBN} or the target are in any specific state, and especially,
20516 the state is not reverted to the state before the MI command was
20517 processed. Therefore, whenever an MI command results in an error,
20518 we recommend that the frontend refreshes all the information shown in
20519 the user interface.
20523 * Context management::
20524 * Asynchronous and non-stop modes::
20528 @node Context management
20529 @subsection Context management
20531 In most cases when @value{GDBN} accesses the target, this access is
20532 done in context of a specific thread and frame (@pxref{Frames}).
20533 Often, even when accessing global data, the target requires that a thread
20534 be specified. The CLI interface maintains the selected thread and frame,
20535 and supplies them to target on each command. This is convenient,
20536 because a command line user would not want to specify that information
20537 explicitly on each command, and because user interacts with
20538 @value{GDBN} via a single terminal, so no confusion is possible as
20539 to what thread and frame are the current ones.
20541 In the case of MI, the concept of selected thread and frame is less
20542 useful. First, a frontend can easily remember this information
20543 itself. Second, a graphical frontend can have more than one window,
20544 each one used for debugging a different thread, and the frontend might
20545 want to access additional threads for internal purposes. This
20546 increases the risk that by relying on implicitly selected thread, the
20547 frontend may be operating on a wrong one. Therefore, each MI command
20548 should explicitly specify which thread and frame to operate on. To
20549 make it possible, each MI command accepts the @samp{--thread} and
20550 @samp{--frame} options, the value to each is @value{GDBN} identifier
20551 for thread and frame to operate on.
20553 Usually, each top-level window in a frontend allows the user to select
20554 a thread and a frame, and remembers the user selection for further
20555 operations. However, in some cases @value{GDBN} may suggest that the
20556 current thread be changed. For example, when stopping on a breakpoint
20557 it is reasonable to switch to the thread where breakpoint is hit. For
20558 another example, if the user issues the CLI @samp{thread} command via
20559 the frontend, it is desirable to change the frontend's selected thread to the
20560 one specified by user. @value{GDBN} communicates the suggestion to
20561 change current thread using the @samp{=thread-selected} notification.
20562 No such notification is available for the selected frame at the moment.
20564 Note that historically, MI shares the selected thread with CLI, so
20565 frontends used the @code{-thread-select} to execute commands in the
20566 right context. However, getting this to work right is cumbersome. The
20567 simplest way is for frontend to emit @code{-thread-select} command
20568 before every command. This doubles the number of commands that need
20569 to be sent. The alternative approach is to suppress @code{-thread-select}
20570 if the selected thread in @value{GDBN} is supposed to be identical to the
20571 thread the frontend wants to operate on. However, getting this
20572 optimization right can be tricky. In particular, if the frontend
20573 sends several commands to @value{GDBN}, and one of the commands changes the
20574 selected thread, then the behaviour of subsequent commands will
20575 change. So, a frontend should either wait for response from such
20576 problematic commands, or explicitly add @code{-thread-select} for
20577 all subsequent commands. No frontend is known to do this exactly
20578 right, so it is suggested to just always pass the @samp{--thread} and
20579 @samp{--frame} options.
20581 @node Asynchronous and non-stop modes
20582 @subsection Asynchronous command execution and non-stop mode
20584 On some targets, @value{GDBN} is capable of processing MI commands
20585 even while the target is running. This is called @dfn{asynchronous
20586 command execution} (@pxref{Background Execution}). The frontend may
20587 specify a preferrence for asynchronous execution using the
20588 @code{-gdb-set target-async 1} command, which should be emitted before
20589 either running the executable or attaching to the target. After the
20590 frontend has started the executable or attached to the target, it can
20591 find if asynchronous execution is enabled using the
20592 @code{-list-target-features} command.
20594 Even if @value{GDBN} can accept a command while target is running,
20595 many commands that access the target do not work when the target is
20596 running. Therefore, asynchronous command execution is most useful
20597 when combined with non-stop mode (@pxref{Non-Stop Mode}). Then,
20598 it is possible to examine the state of one thread, while other threads
20601 When a given thread is running, MI commands that try to access the
20602 target in the context of that thread may not work, or may work only on
20603 some targets. In particular, commands that try to operate on thread's
20604 stack will not work, on any target. Commands that read memory, or
20605 modify breakpoints, may work or not work, depending on the target. Note
20606 that even commands that operate on global state, such as @code{print},
20607 @code{set}, and breakpoint commands, still access the target in the
20608 context of a specific thread, so frontend should try to find a
20609 stopped thread and perform the operation on that thread (using the
20610 @samp{--thread} option).
20612 Which commands will work in the context of a running thread is
20613 highly target dependent. However, the two commands
20614 @code{-exec-interrupt}, to stop a thread, and @code{-thread-info},
20615 to find the state of a thread, will always work.
20617 @node Thread groups
20618 @subsection Thread groups
20619 @value{GDBN} may be used to debug several processes at the same time.
20620 On some platfroms, @value{GDBN} may support debugging of several
20621 hardware systems, each one having several cores with several different
20622 processes running on each core. This section describes the MI
20623 mechanism to support such debugging scenarios.
20625 The key observation is that regardless of the structure of the
20626 target, MI can have a global list of threads, because most commands that
20627 accept the @samp{--thread} option do not need to know what process that
20628 thread belongs to. Therefore, it is not necessary to introduce
20629 neither additional @samp{--process} option, nor an notion of the
20630 current process in the MI interface. The only strictly new feature
20631 that is required is the ability to find how the threads are grouped
20634 To allow the user to discover such grouping, and to support arbitrary
20635 hierarchy of machines/cores/processes, MI introduces the concept of a
20636 @dfn{thread group}. Thread group is a collection of threads and other
20637 thread groups. A thread group always has a string identifier, a type,
20638 and may have additional attributes specific to the type. A new
20639 command, @code{-list-thread-groups}, returns the list of top-level
20640 thread groups, which correspond to processes that @value{GDBN} is
20641 debugging at the moment. By passing an identifier of a thread group
20642 to the @code{-list-thread-groups} command, it is possible to obtain
20643 the members of specific thread group.
20645 To allow the user to easily discover processes, and other objects, he
20646 wishes to debug, a concept of @dfn{available thread group} is
20647 introduced. Available thread group is an thread group that
20648 @value{GDBN} is not debugging, but that can be attached to, using the
20649 @code{-target-attach} command. The list of available top-level thread
20650 groups can be obtained using @samp{-list-thread-groups --available}.
20651 In general, the content of a thread group may be only retrieved only
20652 after attaching to that thread group.
20654 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
20655 @node GDB/MI Command Syntax
20656 @section @sc{gdb/mi} Command Syntax
20659 * GDB/MI Input Syntax::
20660 * GDB/MI Output Syntax::
20663 @node GDB/MI Input Syntax
20664 @subsection @sc{gdb/mi} Input Syntax
20666 @cindex input syntax for @sc{gdb/mi}
20667 @cindex @sc{gdb/mi}, input syntax
20669 @item @var{command} @expansion{}
20670 @code{@var{cli-command} | @var{mi-command}}
20672 @item @var{cli-command} @expansion{}
20673 @code{[ @var{token} ] @var{cli-command} @var{nl}}, where
20674 @var{cli-command} is any existing @value{GDBN} CLI command.
20676 @item @var{mi-command} @expansion{}
20677 @code{[ @var{token} ] "-" @var{operation} ( " " @var{option} )*
20678 @code{[} " --" @code{]} ( " " @var{parameter} )* @var{nl}}
20680 @item @var{token} @expansion{}
20681 "any sequence of digits"
20683 @item @var{option} @expansion{}
20684 @code{"-" @var{parameter} [ " " @var{parameter} ]}
20686 @item @var{parameter} @expansion{}
20687 @code{@var{non-blank-sequence} | @var{c-string}}
20689 @item @var{operation} @expansion{}
20690 @emph{any of the operations described in this chapter}
20692 @item @var{non-blank-sequence} @expansion{}
20693 @emph{anything, provided it doesn't contain special characters such as
20694 "-", @var{nl}, """ and of course " "}
20696 @item @var{c-string} @expansion{}
20697 @code{""" @var{seven-bit-iso-c-string-content} """}
20699 @item @var{nl} @expansion{}
20708 The CLI commands are still handled by the @sc{mi} interpreter; their
20709 output is described below.
20712 The @code{@var{token}}, when present, is passed back when the command
20716 Some @sc{mi} commands accept optional arguments as part of the parameter
20717 list. Each option is identified by a leading @samp{-} (dash) and may be
20718 followed by an optional argument parameter. Options occur first in the
20719 parameter list and can be delimited from normal parameters using
20720 @samp{--} (this is useful when some parameters begin with a dash).
20727 We want easy access to the existing CLI syntax (for debugging).
20730 We want it to be easy to spot a @sc{mi} operation.
20733 @node GDB/MI Output Syntax
20734 @subsection @sc{gdb/mi} Output Syntax
20736 @cindex output syntax of @sc{gdb/mi}
20737 @cindex @sc{gdb/mi}, output syntax
20738 The output from @sc{gdb/mi} consists of zero or more out-of-band records
20739 followed, optionally, by a single result record. This result record
20740 is for the most recent command. The sequence of output records is
20741 terminated by @samp{(gdb)}.
20743 If an input command was prefixed with a @code{@var{token}} then the
20744 corresponding output for that command will also be prefixed by that same
20748 @item @var{output} @expansion{}
20749 @code{( @var{out-of-band-record} )* [ @var{result-record} ] "(gdb)" @var{nl}}
20751 @item @var{result-record} @expansion{}
20752 @code{ [ @var{token} ] "^" @var{result-class} ( "," @var{result} )* @var{nl}}
20754 @item @var{out-of-band-record} @expansion{}
20755 @code{@var{async-record} | @var{stream-record}}
20757 @item @var{async-record} @expansion{}
20758 @code{@var{exec-async-output} | @var{status-async-output} | @var{notify-async-output}}
20760 @item @var{exec-async-output} @expansion{}
20761 @code{[ @var{token} ] "*" @var{async-output}}
20763 @item @var{status-async-output} @expansion{}
20764 @code{[ @var{token} ] "+" @var{async-output}}
20766 @item @var{notify-async-output} @expansion{}
20767 @code{[ @var{token} ] "=" @var{async-output}}
20769 @item @var{async-output} @expansion{}
20770 @code{@var{async-class} ( "," @var{result} )* @var{nl}}
20772 @item @var{result-class} @expansion{}
20773 @code{"done" | "running" | "connected" | "error" | "exit"}
20775 @item @var{async-class} @expansion{}
20776 @code{"stopped" | @var{others}} (where @var{others} will be added
20777 depending on the needs---this is still in development).
20779 @item @var{result} @expansion{}
20780 @code{ @var{variable} "=" @var{value}}
20782 @item @var{variable} @expansion{}
20783 @code{ @var{string} }
20785 @item @var{value} @expansion{}
20786 @code{ @var{const} | @var{tuple} | @var{list} }
20788 @item @var{const} @expansion{}
20789 @code{@var{c-string}}
20791 @item @var{tuple} @expansion{}
20792 @code{ "@{@}" | "@{" @var{result} ( "," @var{result} )* "@}" }
20794 @item @var{list} @expansion{}
20795 @code{ "[]" | "[" @var{value} ( "," @var{value} )* "]" | "["
20796 @var{result} ( "," @var{result} )* "]" }
20798 @item @var{stream-record} @expansion{}
20799 @code{@var{console-stream-output} | @var{target-stream-output} | @var{log-stream-output}}
20801 @item @var{console-stream-output} @expansion{}
20802 @code{"~" @var{c-string}}
20804 @item @var{target-stream-output} @expansion{}
20805 @code{"@@" @var{c-string}}
20807 @item @var{log-stream-output} @expansion{}
20808 @code{"&" @var{c-string}}
20810 @item @var{nl} @expansion{}
20813 @item @var{token} @expansion{}
20814 @emph{any sequence of digits}.
20822 All output sequences end in a single line containing a period.
20825 The @code{@var{token}} is from the corresponding request. Note that
20826 for all async output, while the token is allowed by the grammar and
20827 may be output by future versions of @value{GDBN} for select async
20828 output messages, it is generally omitted. Frontends should treat
20829 all async output as reporting general changes in the state of the
20830 target and there should be no need to associate async output to any
20834 @cindex status output in @sc{gdb/mi}
20835 @var{status-async-output} contains on-going status information about the
20836 progress of a slow operation. It can be discarded. All status output is
20837 prefixed by @samp{+}.
20840 @cindex async output in @sc{gdb/mi}
20841 @var{exec-async-output} contains asynchronous state change on the target
20842 (stopped, started, disappeared). All async output is prefixed by
20846 @cindex notify output in @sc{gdb/mi}
20847 @var{notify-async-output} contains supplementary information that the
20848 client should handle (e.g., a new breakpoint information). All notify
20849 output is prefixed by @samp{=}.
20852 @cindex console output in @sc{gdb/mi}
20853 @var{console-stream-output} is output that should be displayed as is in the
20854 console. It is the textual response to a CLI command. All the console
20855 output is prefixed by @samp{~}.
20858 @cindex target output in @sc{gdb/mi}
20859 @var{target-stream-output} is the output produced by the target program.
20860 All the target output is prefixed by @samp{@@}.
20863 @cindex log output in @sc{gdb/mi}
20864 @var{log-stream-output} is output text coming from @value{GDBN}'s internals, for
20865 instance messages that should be displayed as part of an error log. All
20866 the log output is prefixed by @samp{&}.
20869 @cindex list output in @sc{gdb/mi}
20870 New @sc{gdb/mi} commands should only output @var{lists} containing
20876 @xref{GDB/MI Stream Records, , @sc{gdb/mi} Stream Records}, for more
20877 details about the various output records.
20879 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
20880 @node GDB/MI Compatibility with CLI
20881 @section @sc{gdb/mi} Compatibility with CLI
20883 @cindex compatibility, @sc{gdb/mi} and CLI
20884 @cindex @sc{gdb/mi}, compatibility with CLI
20886 For the developers convenience CLI commands can be entered directly,
20887 but there may be some unexpected behaviour. For example, commands
20888 that query the user will behave as if the user replied yes, breakpoint
20889 command lists are not executed and some CLI commands, such as
20890 @code{if}, @code{when} and @code{define}, prompt for further input with
20891 @samp{>}, which is not valid MI output.
20893 This feature may be removed at some stage in the future and it is
20894 recommended that front ends use the @code{-interpreter-exec} command
20895 (@pxref{-interpreter-exec}).
20897 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
20898 @node GDB/MI Development and Front Ends
20899 @section @sc{gdb/mi} Development and Front Ends
20900 @cindex @sc{gdb/mi} development
20902 The application which takes the MI output and presents the state of the
20903 program being debugged to the user is called a @dfn{front end}.
20905 Although @sc{gdb/mi} is still incomplete, it is currently being used
20906 by a variety of front ends to @value{GDBN}. This makes it difficult
20907 to introduce new functionality without breaking existing usage. This
20908 section tries to minimize the problems by describing how the protocol
20911 Some changes in MI need not break a carefully designed front end, and
20912 for these the MI version will remain unchanged. The following is a
20913 list of changes that may occur within one level, so front ends should
20914 parse MI output in a way that can handle them:
20918 New MI commands may be added.
20921 New fields may be added to the output of any MI command.
20924 The range of values for fields with specified values, e.g.,
20925 @code{in_scope} (@pxref{-var-update}) may be extended.
20927 @c The format of field's content e.g type prefix, may change so parse it
20928 @c at your own risk. Yes, in general?
20930 @c The order of fields may change? Shouldn't really matter but it might
20931 @c resolve inconsistencies.
20934 If the changes are likely to break front ends, the MI version level
20935 will be increased by one. This will allow the front end to parse the
20936 output according to the MI version. Apart from mi0, new versions of
20937 @value{GDBN} will not support old versions of MI and it will be the
20938 responsibility of the front end to work with the new one.
20940 @c Starting with mi3, add a new command -mi-version that prints the MI
20943 The best way to avoid unexpected changes in MI that might break your front
20944 end is to make your project known to @value{GDBN} developers and
20945 follow development on @email{gdb@@sourceware.org} and
20946 @email{gdb-patches@@sourceware.org}.
20947 @cindex mailing lists
20949 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
20950 @node GDB/MI Output Records
20951 @section @sc{gdb/mi} Output Records
20954 * GDB/MI Result Records::
20955 * GDB/MI Stream Records::
20956 * GDB/MI Async Records::
20957 * GDB/MI Frame Information::
20960 @node GDB/MI Result Records
20961 @subsection @sc{gdb/mi} Result Records
20963 @cindex result records in @sc{gdb/mi}
20964 @cindex @sc{gdb/mi}, result records
20965 In addition to a number of out-of-band notifications, the response to a
20966 @sc{gdb/mi} command includes one of the following result indications:
20970 @item "^done" [ "," @var{results} ]
20971 The synchronous operation was successful, @code{@var{results}} are the return
20976 @c Is this one correct? Should it be an out-of-band notification?
20977 The asynchronous operation was successfully started. The target is
20982 @value{GDBN} has connected to a remote target.
20984 @item "^error" "," @var{c-string}
20986 The operation failed. The @code{@var{c-string}} contains the corresponding
20991 @value{GDBN} has terminated.
20995 @node GDB/MI Stream Records
20996 @subsection @sc{gdb/mi} Stream Records
20998 @cindex @sc{gdb/mi}, stream records
20999 @cindex stream records in @sc{gdb/mi}
21000 @value{GDBN} internally maintains a number of output streams: the console, the
21001 target, and the log. The output intended for each of these streams is
21002 funneled through the @sc{gdb/mi} interface using @dfn{stream records}.
21004 Each stream record begins with a unique @dfn{prefix character} which
21005 identifies its stream (@pxref{GDB/MI Output Syntax, , @sc{gdb/mi} Output
21006 Syntax}). In addition to the prefix, each stream record contains a
21007 @code{@var{string-output}}. This is either raw text (with an implicit new
21008 line) or a quoted C string (which does not contain an implicit newline).
21011 @item "~" @var{string-output}
21012 The console output stream contains text that should be displayed in the
21013 CLI console window. It contains the textual responses to CLI commands.
21015 @item "@@" @var{string-output}
21016 The target output stream contains any textual output from the running
21017 target. This is only present when GDB's event loop is truly
21018 asynchronous, which is currently only the case for remote targets.
21020 @item "&" @var{string-output}
21021 The log stream contains debugging messages being produced by @value{GDBN}'s
21025 @node GDB/MI Async Records
21026 @subsection @sc{gdb/mi} Async Records
21028 @cindex async records in @sc{gdb/mi}
21029 @cindex @sc{gdb/mi}, async records
21030 @dfn{Async} records are used to notify the @sc{gdb/mi} client of
21031 additional changes that have occurred. Those changes can either be a
21032 consequence of @sc{gdb/mi} commands (e.g., a breakpoint modified) or a result of
21033 target activity (e.g., target stopped).
21035 The following is the list of possible async records:
21039 @item *running,thread-id="@var{thread}"
21040 The target is now running. The @var{thread} field tells which
21041 specific thread is now running, and can be @samp{all} if all threads
21042 are running. The frontend should assume that no interaction with a
21043 running thread is possible after this notification is produced.
21044 The frontend should not assume that this notification is output
21045 only once for any command. @value{GDBN} may emit this notification
21046 several times, either for different threads, because it cannot resume
21047 all threads together, or even for a single thread, if the thread must
21048 be stepped though some code before letting it run freely.
21050 @item *stopped,reason="@var{reason}",thread-id="@var{id}",stopped-threads="@var{stopped}"
21051 The target has stopped. The @var{reason} field can have one of the
21055 @item breakpoint-hit
21056 A breakpoint was reached.
21057 @item watchpoint-trigger
21058 A watchpoint was triggered.
21059 @item read-watchpoint-trigger
21060 A read watchpoint was triggered.
21061 @item access-watchpoint-trigger
21062 An access watchpoint was triggered.
21063 @item function-finished
21064 An -exec-finish or similar CLI command was accomplished.
21065 @item location-reached
21066 An -exec-until or similar CLI command was accomplished.
21067 @item watchpoint-scope
21068 A watchpoint has gone out of scope.
21069 @item end-stepping-range
21070 An -exec-next, -exec-next-instruction, -exec-step, -exec-step-instruction or
21071 similar CLI command was accomplished.
21072 @item exited-signalled
21073 The inferior exited because of a signal.
21075 The inferior exited.
21076 @item exited-normally
21077 The inferior exited normally.
21078 @item signal-received
21079 A signal was received by the inferior.
21082 The @var{id} field identifies the thread that directly caused the stop
21083 -- for example by hitting a breakpoint. Depending on whether all-stop
21084 mode is in effect (@pxref{All-Stop Mode}), @value{GDBN} may either
21085 stop all threads, or only the thread that directly triggered the stop.
21086 If all threads are stopped, the @var{stopped} field will have the
21087 value of @code{"all"}. Otherwise, the value of the @var{stopped}
21088 field will be a list of thread identifiers. Presently, this list will
21089 always include a single thread, but frontend should be prepared to see
21090 several threads in the list.
21092 @item =thread-group-created,id="@var{id}"
21093 @itemx =thread-group-exited,id="@var{id}"
21094 A thread thread group either was attached to, or has exited/detached
21095 from. The @var{id} field contains the @value{GDBN} identifier of the
21098 @item =thread-created,id="@var{id}",group-id="@var{gid}"
21099 @itemx =thread-exited,id="@var{id}",group-id="@var{gid}"
21100 A thread either was created, or has exited. The @var{id} field
21101 contains the @value{GDBN} identifier of the thread. The @var{gid}
21102 field identifies the thread group this thread belongs to.
21104 @item =thread-selected,id="@var{id}"
21105 Informs that the selected thread was changed as result of the last
21106 command. This notification is not emitted as result of @code{-thread-select}
21107 command but is emitted whenever an MI command that is not documented
21108 to change the selected thread actually changes it. In particular,
21109 invoking, directly or indirectly (via user-defined command), the CLI
21110 @code{thread} command, will generate this notification.
21112 We suggest that in response to this notification, front ends
21113 highlight the selected thread and cause subsequent commands to apply to
21116 @item =library-loaded,...
21117 Reports that a new library file was loaded by the program. This
21118 notification has 4 fields---@var{id}, @var{target-name},
21119 @var{host-name}, and @var{symbols-loaded}. The @var{id} field is an
21120 opaque identifier of the library. For remote debugging case,
21121 @var{target-name} and @var{host-name} fields give the name of the
21122 library file on the target, and on the host respectively. For native
21123 debugging, both those fields have the same value. The
21124 @var{symbols-loaded} field reports if the debug symbols for this
21125 library are loaded.
21127 @item =library-unloaded,...
21128 Reports that a library was unloaded by the program. This notification
21129 has 3 fields---@var{id}, @var{target-name} and @var{host-name} with
21130 the same meaning as for the @code{=library-loaded} notification
21134 @node GDB/MI Frame Information
21135 @subsection @sc{gdb/mi} Frame Information
21137 Response from many MI commands includes an information about stack
21138 frame. This information is a tuple that may have the following
21143 The level of the stack frame. The innermost frame has the level of
21144 zero. This field is always present.
21147 The name of the function corresponding to the frame. This field may
21148 be absent if @value{GDBN} is unable to determine the function name.
21151 The code address for the frame. This field is always present.
21154 The name of the source files that correspond to the frame's code
21155 address. This field may be absent.
21158 The source line corresponding to the frames' code address. This field
21162 The name of the binary file (either executable or shared library) the
21163 corresponds to the frame's code address. This field may be absent.
21168 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
21169 @node GDB/MI Simple Examples
21170 @section Simple Examples of @sc{gdb/mi} Interaction
21171 @cindex @sc{gdb/mi}, simple examples
21173 This subsection presents several simple examples of interaction using
21174 the @sc{gdb/mi} interface. In these examples, @samp{->} means that the
21175 following line is passed to @sc{gdb/mi} as input, while @samp{<-} means
21176 the output received from @sc{gdb/mi}.
21178 Note the line breaks shown in the examples are here only for
21179 readability, they don't appear in the real output.
21181 @subheading Setting a Breakpoint
21183 Setting a breakpoint generates synchronous output which contains detailed
21184 information of the breakpoint.
21187 -> -break-insert main
21188 <- ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
21189 enabled="y",addr="0x08048564",func="main",file="myprog.c",
21190 fullname="/home/nickrob/myprog.c",line="68",times="0"@}
21194 @subheading Program Execution
21196 Program execution generates asynchronous records and MI gives the
21197 reason that execution stopped.
21203 <- *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
21204 frame=@{addr="0x08048564",func="main",
21205 args=[@{name="argc",value="1"@},@{name="argv",value="0xbfc4d4d4"@}],
21206 file="myprog.c",fullname="/home/nickrob/myprog.c",line="68"@}
21211 <- *stopped,reason="exited-normally"
21215 @subheading Quitting @value{GDBN}
21217 Quitting @value{GDBN} just prints the result class @samp{^exit}.
21225 @subheading A Bad Command
21227 Here's what happens if you pass a non-existent command:
21231 <- ^error,msg="Undefined MI command: rubbish"
21236 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
21237 @node GDB/MI Command Description Format
21238 @section @sc{gdb/mi} Command Description Format
21240 The remaining sections describe blocks of commands. Each block of
21241 commands is laid out in a fashion similar to this section.
21243 @subheading Motivation
21245 The motivation for this collection of commands.
21247 @subheading Introduction
21249 A brief introduction to this collection of commands as a whole.
21251 @subheading Commands
21253 For each command in the block, the following is described:
21255 @subsubheading Synopsis
21258 -command @var{args}@dots{}
21261 @subsubheading Result
21263 @subsubheading @value{GDBN} Command
21265 The corresponding @value{GDBN} CLI command(s), if any.
21267 @subsubheading Example
21269 Example(s) formatted for readability. Some of the described commands have
21270 not been implemented yet and these are labeled N.A.@: (not available).
21273 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
21274 @node GDB/MI Breakpoint Commands
21275 @section @sc{gdb/mi} Breakpoint Commands
21277 @cindex breakpoint commands for @sc{gdb/mi}
21278 @cindex @sc{gdb/mi}, breakpoint commands
21279 This section documents @sc{gdb/mi} commands for manipulating
21282 @subheading The @code{-break-after} Command
21283 @findex -break-after
21285 @subsubheading Synopsis
21288 -break-after @var{number} @var{count}
21291 The breakpoint number @var{number} is not in effect until it has been
21292 hit @var{count} times. To see how this is reflected in the output of
21293 the @samp{-break-list} command, see the description of the
21294 @samp{-break-list} command below.
21296 @subsubheading @value{GDBN} Command
21298 The corresponding @value{GDBN} command is @samp{ignore}.
21300 @subsubheading Example
21305 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
21306 enabled="y",addr="0x000100d0",func="main",file="hello.c",
21307 fullname="/home/foo/hello.c",line="5",times="0"@}
21314 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
21315 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
21316 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
21317 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
21318 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
21319 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
21320 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
21321 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
21322 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
21323 line="5",times="0",ignore="3"@}]@}
21328 @subheading The @code{-break-catch} Command
21329 @findex -break-catch
21331 @subheading The @code{-break-commands} Command
21332 @findex -break-commands
21336 @subheading The @code{-break-condition} Command
21337 @findex -break-condition
21339 @subsubheading Synopsis
21342 -break-condition @var{number} @var{expr}
21345 Breakpoint @var{number} will stop the program only if the condition in
21346 @var{expr} is true. The condition becomes part of the
21347 @samp{-break-list} output (see the description of the @samp{-break-list}
21350 @subsubheading @value{GDBN} Command
21352 The corresponding @value{GDBN} command is @samp{condition}.
21354 @subsubheading Example
21358 -break-condition 1 1
21362 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
21363 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
21364 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
21365 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
21366 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
21367 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
21368 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
21369 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
21370 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
21371 line="5",cond="1",times="0",ignore="3"@}]@}
21375 @subheading The @code{-break-delete} Command
21376 @findex -break-delete
21378 @subsubheading Synopsis
21381 -break-delete ( @var{breakpoint} )+
21384 Delete the breakpoint(s) whose number(s) are specified in the argument
21385 list. This is obviously reflected in the breakpoint list.
21387 @subsubheading @value{GDBN} Command
21389 The corresponding @value{GDBN} command is @samp{delete}.
21391 @subsubheading Example
21399 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
21400 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
21401 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
21402 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
21403 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
21404 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
21405 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
21410 @subheading The @code{-break-disable} Command
21411 @findex -break-disable
21413 @subsubheading Synopsis
21416 -break-disable ( @var{breakpoint} )+
21419 Disable the named @var{breakpoint}(s). The field @samp{enabled} in the
21420 break list is now set to @samp{n} for the named @var{breakpoint}(s).
21422 @subsubheading @value{GDBN} Command
21424 The corresponding @value{GDBN} command is @samp{disable}.
21426 @subsubheading Example
21434 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
21435 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
21436 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
21437 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
21438 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
21439 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
21440 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
21441 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="n",
21442 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
21443 line="5",times="0"@}]@}
21447 @subheading The @code{-break-enable} Command
21448 @findex -break-enable
21450 @subsubheading Synopsis
21453 -break-enable ( @var{breakpoint} )+
21456 Enable (previously disabled) @var{breakpoint}(s).
21458 @subsubheading @value{GDBN} Command
21460 The corresponding @value{GDBN} command is @samp{enable}.
21462 @subsubheading Example
21470 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
21471 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
21472 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
21473 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
21474 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
21475 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
21476 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
21477 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
21478 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
21479 line="5",times="0"@}]@}
21483 @subheading The @code{-break-info} Command
21484 @findex -break-info
21486 @subsubheading Synopsis
21489 -break-info @var{breakpoint}
21493 Get information about a single breakpoint.
21495 @subsubheading @value{GDBN} Command
21497 The corresponding @value{GDBN} command is @samp{info break @var{breakpoint}}.
21499 @subsubheading Example
21502 @subheading The @code{-break-insert} Command
21503 @findex -break-insert
21505 @subsubheading Synopsis
21508 -break-insert [ -t ] [ -h ] [ -f ] [ -d ]
21509 [ -c @var{condition} ] [ -i @var{ignore-count} ]
21510 [ -p @var{thread} ] [ @var{location} ]
21514 If specified, @var{location}, can be one of:
21521 @item filename:linenum
21522 @item filename:function
21526 The possible optional parameters of this command are:
21530 Insert a temporary breakpoint.
21532 Insert a hardware breakpoint.
21533 @item -c @var{condition}
21534 Make the breakpoint conditional on @var{condition}.
21535 @item -i @var{ignore-count}
21536 Initialize the @var{ignore-count}.
21538 If @var{location} cannot be parsed (for example if it
21539 refers to unknown files or functions), create a pending
21540 breakpoint. Without this flag, @value{GDBN} will report
21541 an error, and won't create a breakpoint, if @var{location}
21544 Create a disabled breakpoint.
21547 @subsubheading Result
21549 The result is in the form:
21552 ^done,bkpt=@{number="@var{number}",type="@var{type}",disp="del"|"keep",
21553 enabled="y"|"n",addr="@var{hex}",func="@var{funcname}",file="@var{filename}",
21554 fullname="@var{full_filename}",line="@var{lineno}",[thread="@var{threadno},]
21555 times="@var{times}"@}
21559 where @var{number} is the @value{GDBN} number for this breakpoint,
21560 @var{funcname} is the name of the function where the breakpoint was
21561 inserted, @var{filename} is the name of the source file which contains
21562 this function, @var{lineno} is the source line number within that file
21563 and @var{times} the number of times that the breakpoint has been hit
21564 (always 0 for -break-insert but may be greater for -break-info or -break-list
21565 which use the same output).
21567 Note: this format is open to change.
21568 @c An out-of-band breakpoint instead of part of the result?
21570 @subsubheading @value{GDBN} Command
21572 The corresponding @value{GDBN} commands are @samp{break}, @samp{tbreak},
21573 @samp{hbreak}, @samp{thbreak}, and @samp{rbreak}.
21575 @subsubheading Example
21580 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",
21581 fullname="/home/foo/recursive2.c,line="4",times="0"@}
21583 -break-insert -t foo
21584 ^done,bkpt=@{number="2",addr="0x00010774",file="recursive2.c",
21585 fullname="/home/foo/recursive2.c,line="11",times="0"@}
21588 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
21589 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
21590 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
21591 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
21592 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
21593 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
21594 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
21595 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
21596 addr="0x0001072c", func="main",file="recursive2.c",
21597 fullname="/home/foo/recursive2.c,"line="4",times="0"@},
21598 bkpt=@{number="2",type="breakpoint",disp="del",enabled="y",
21599 addr="0x00010774",func="foo",file="recursive2.c",
21600 fullname="/home/foo/recursive2.c",line="11",times="0"@}]@}
21602 -break-insert -r foo.*
21603 ~int foo(int, int);
21604 ^done,bkpt=@{number="3",addr="0x00010774",file="recursive2.c,
21605 "fullname="/home/foo/recursive2.c",line="11",times="0"@}
21609 @subheading The @code{-break-list} Command
21610 @findex -break-list
21612 @subsubheading Synopsis
21618 Displays the list of inserted breakpoints, showing the following fields:
21622 number of the breakpoint
21624 type of the breakpoint: @samp{breakpoint} or @samp{watchpoint}
21626 should the breakpoint be deleted or disabled when it is hit: @samp{keep}
21629 is the breakpoint enabled or no: @samp{y} or @samp{n}
21631 memory location at which the breakpoint is set
21633 logical location of the breakpoint, expressed by function name, file
21636 number of times the breakpoint has been hit
21639 If there are no breakpoints or watchpoints, the @code{BreakpointTable}
21640 @code{body} field is an empty list.
21642 @subsubheading @value{GDBN} Command
21644 The corresponding @value{GDBN} command is @samp{info break}.
21646 @subsubheading Example
21651 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
21652 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
21653 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
21654 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
21655 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
21656 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
21657 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
21658 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
21659 addr="0x000100d0",func="main",file="hello.c",line="5",times="0"@},
21660 bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
21661 addr="0x00010114",func="foo",file="hello.c",fullname="/home/foo/hello.c",
21662 line="13",times="0"@}]@}
21666 Here's an example of the result when there are no breakpoints:
21671 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
21672 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
21673 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
21674 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
21675 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
21676 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
21677 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
21682 @subheading The @code{-break-watch} Command
21683 @findex -break-watch
21685 @subsubheading Synopsis
21688 -break-watch [ -a | -r ]
21691 Create a watchpoint. With the @samp{-a} option it will create an
21692 @dfn{access} watchpoint, i.e., a watchpoint that triggers either on a
21693 read from or on a write to the memory location. With the @samp{-r}
21694 option, the watchpoint created is a @dfn{read} watchpoint, i.e., it will
21695 trigger only when the memory location is accessed for reading. Without
21696 either of the options, the watchpoint created is a regular watchpoint,
21697 i.e., it will trigger when the memory location is accessed for writing.
21698 @xref{Set Watchpoints, , Setting Watchpoints}.
21700 Note that @samp{-break-list} will report a single list of watchpoints and
21701 breakpoints inserted.
21703 @subsubheading @value{GDBN} Command
21705 The corresponding @value{GDBN} commands are @samp{watch}, @samp{awatch}, and
21708 @subsubheading Example
21710 Setting a watchpoint on a variable in the @code{main} function:
21715 ^done,wpt=@{number="2",exp="x"@}
21720 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="x"@},
21721 value=@{old="-268439212",new="55"@},
21722 frame=@{func="main",args=[],file="recursive2.c",
21723 fullname="/home/foo/bar/recursive2.c",line="5"@}
21727 Setting a watchpoint on a variable local to a function. @value{GDBN} will stop
21728 the program execution twice: first for the variable changing value, then
21729 for the watchpoint going out of scope.
21734 ^done,wpt=@{number="5",exp="C"@}
21739 *stopped,reason="watchpoint-trigger",
21740 wpt=@{number="5",exp="C"@},value=@{old="-276895068",new="3"@},
21741 frame=@{func="callee4",args=[],
21742 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
21743 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13"@}
21748 *stopped,reason="watchpoint-scope",wpnum="5",
21749 frame=@{func="callee3",args=[@{name="strarg",
21750 value="0x11940 \"A string argument.\""@}],
21751 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
21752 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
21756 Listing breakpoints and watchpoints, at different points in the program
21757 execution. Note that once the watchpoint goes out of scope, it is
21763 ^done,wpt=@{number="2",exp="C"@}
21766 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
21767 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
21768 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
21769 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
21770 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
21771 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
21772 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
21773 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
21774 addr="0x00010734",func="callee4",
21775 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
21776 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c"line="8",times="1"@},
21777 bkpt=@{number="2",type="watchpoint",disp="keep",
21778 enabled="y",addr="",what="C",times="0"@}]@}
21783 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="C"@},
21784 value=@{old="-276895068",new="3"@},
21785 frame=@{func="callee4",args=[],
21786 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
21787 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13"@}
21790 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
21791 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
21792 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
21793 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
21794 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
21795 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
21796 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
21797 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
21798 addr="0x00010734",func="callee4",
21799 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
21800 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",times="1"@},
21801 bkpt=@{number="2",type="watchpoint",disp="keep",
21802 enabled="y",addr="",what="C",times="-5"@}]@}
21806 ^done,reason="watchpoint-scope",wpnum="2",
21807 frame=@{func="callee3",args=[@{name="strarg",
21808 value="0x11940 \"A string argument.\""@}],
21809 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
21810 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
21813 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
21814 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
21815 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
21816 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
21817 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
21818 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
21819 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
21820 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
21821 addr="0x00010734",func="callee4",
21822 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
21823 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",
21828 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
21829 @node GDB/MI Program Context
21830 @section @sc{gdb/mi} Program Context
21832 @subheading The @code{-exec-arguments} Command
21833 @findex -exec-arguments
21836 @subsubheading Synopsis
21839 -exec-arguments @var{args}
21842 Set the inferior program arguments, to be used in the next
21845 @subsubheading @value{GDBN} Command
21847 The corresponding @value{GDBN} command is @samp{set args}.
21849 @subsubheading Example
21853 -exec-arguments -v word
21859 @subheading The @code{-exec-show-arguments} Command
21860 @findex -exec-show-arguments
21862 @subsubheading Synopsis
21865 -exec-show-arguments
21868 Print the arguments of the program.
21870 @subsubheading @value{GDBN} Command
21872 The corresponding @value{GDBN} command is @samp{show args}.
21874 @subsubheading Example
21878 @subheading The @code{-environment-cd} Command
21879 @findex -environment-cd
21881 @subsubheading Synopsis
21884 -environment-cd @var{pathdir}
21887 Set @value{GDBN}'s working directory.
21889 @subsubheading @value{GDBN} Command
21891 The corresponding @value{GDBN} command is @samp{cd}.
21893 @subsubheading Example
21897 -environment-cd /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
21903 @subheading The @code{-environment-directory} Command
21904 @findex -environment-directory
21906 @subsubheading Synopsis
21909 -environment-directory [ -r ] [ @var{pathdir} ]+
21912 Add directories @var{pathdir} to beginning of search path for source files.
21913 If the @samp{-r} option is used, the search path is reset to the default
21914 search path. If directories @var{pathdir} are supplied in addition to the
21915 @samp{-r} option, the search path is first reset and then addition
21917 Multiple directories may be specified, separated by blanks. Specifying
21918 multiple directories in a single command
21919 results in the directories added to the beginning of the
21920 search path in the same order they were presented in the command.
21921 If blanks are needed as
21922 part of a directory name, double-quotes should be used around
21923 the name. In the command output, the path will show up separated
21924 by the system directory-separator character. The directory-separator
21925 character must not be used
21926 in any directory name.
21927 If no directories are specified, the current search path is displayed.
21929 @subsubheading @value{GDBN} Command
21931 The corresponding @value{GDBN} command is @samp{dir}.
21933 @subsubheading Example
21937 -environment-directory /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
21938 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
21940 -environment-directory ""
21941 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
21943 -environment-directory -r /home/jjohnstn/src/gdb /usr/src
21944 ^done,source-path="/home/jjohnstn/src/gdb:/usr/src:$cdir:$cwd"
21946 -environment-directory -r
21947 ^done,source-path="$cdir:$cwd"
21952 @subheading The @code{-environment-path} Command
21953 @findex -environment-path
21955 @subsubheading Synopsis
21958 -environment-path [ -r ] [ @var{pathdir} ]+
21961 Add directories @var{pathdir} to beginning of search path for object files.
21962 If the @samp{-r} option is used, the search path is reset to the original
21963 search path that existed at gdb start-up. If directories @var{pathdir} are
21964 supplied in addition to the
21965 @samp{-r} option, the search path is first reset and then addition
21967 Multiple directories may be specified, separated by blanks. Specifying
21968 multiple directories in a single command
21969 results in the directories added to the beginning of the
21970 search path in the same order they were presented in the command.
21971 If blanks are needed as
21972 part of a directory name, double-quotes should be used around
21973 the name. In the command output, the path will show up separated
21974 by the system directory-separator character. The directory-separator
21975 character must not be used
21976 in any directory name.
21977 If no directories are specified, the current path is displayed.
21980 @subsubheading @value{GDBN} Command
21982 The corresponding @value{GDBN} command is @samp{path}.
21984 @subsubheading Example
21989 ^done,path="/usr/bin"
21991 -environment-path /kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb /bin
21992 ^done,path="/kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb:/bin:/usr/bin"
21994 -environment-path -r /usr/local/bin
21995 ^done,path="/usr/local/bin:/usr/bin"
22000 @subheading The @code{-environment-pwd} Command
22001 @findex -environment-pwd
22003 @subsubheading Synopsis
22009 Show the current working directory.
22011 @subsubheading @value{GDBN} Command
22013 The corresponding @value{GDBN} command is @samp{pwd}.
22015 @subsubheading Example
22020 ^done,cwd="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb"
22024 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
22025 @node GDB/MI Thread Commands
22026 @section @sc{gdb/mi} Thread Commands
22029 @subheading The @code{-thread-info} Command
22030 @findex -thread-info
22032 @subsubheading Synopsis
22035 -thread-info [ @var{thread-id} ]
22038 Reports information about either a specific thread, if
22039 the @var{thread-id} parameter is present, or about all
22040 threads. When printing information about all threads,
22041 also reports the current thread.
22043 @subsubheading @value{GDBN} Command
22045 The @samp{info thread} command prints the same information
22048 @subsubheading Example
22053 @{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
22054 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",args=[]@},state="running"@},
22055 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
22056 frame=@{level="0",addr="0x0804891f",func="foo",args=[@{name="i",value="10"@}],
22057 file="/tmp/a.c",fullname="/tmp/a.c",line="158"@},state="running"@}],
22058 current-thread-id="1"
22062 The @samp{state} field may have the following values:
22066 The thread is stopped. Frame information is available for stopped
22070 The thread is running. There's no frame information for running
22075 @subheading The @code{-thread-list-ids} Command
22076 @findex -thread-list-ids
22078 @subsubheading Synopsis
22084 Produces a list of the currently known @value{GDBN} thread ids. At the
22085 end of the list it also prints the total number of such threads.
22087 This command is retained for historical reasons, the
22088 @code{-thread-info} command should be used instead.
22090 @subsubheading @value{GDBN} Command
22092 Part of @samp{info threads} supplies the same information.
22094 @subsubheading Example
22099 ^done,thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
22100 current-thread-id="1",number-of-threads="3"
22105 @subheading The @code{-thread-select} Command
22106 @findex -thread-select
22108 @subsubheading Synopsis
22111 -thread-select @var{threadnum}
22114 Make @var{threadnum} the current thread. It prints the number of the new
22115 current thread, and the topmost frame for that thread.
22117 This command is deprecated in favor of explicitly using the
22118 @samp{--thread} option to each command.
22120 @subsubheading @value{GDBN} Command
22122 The corresponding @value{GDBN} command is @samp{thread}.
22124 @subsubheading Example
22131 *stopped,reason="end-stepping-range",thread-id="2",line="187",
22132 file="../../../devo/gdb/testsuite/gdb.threads/linux-dp.c"
22136 thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
22137 number-of-threads="3"
22140 ^done,new-thread-id="3",
22141 frame=@{level="0",func="vprintf",
22142 args=[@{name="format",value="0x8048e9c \"%*s%c %d %c\\n\""@},
22143 @{name="arg",value="0x2"@}],file="vprintf.c",line="31"@}
22147 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
22148 @node GDB/MI Program Execution
22149 @section @sc{gdb/mi} Program Execution
22151 These are the asynchronous commands which generate the out-of-band
22152 record @samp{*stopped}. Currently @value{GDBN} only really executes
22153 asynchronously with remote targets and this interaction is mimicked in
22156 @subheading The @code{-exec-continue} Command
22157 @findex -exec-continue
22159 @subsubheading Synopsis
22162 -exec-continue [--all|--thread-group N]
22165 Resumes the execution of the inferior program until a breakpoint is
22166 encountered, or until the inferior exits. In all-stop mode
22167 (@pxref{All-Stop Mode}), may resume only one thread, or all threads,
22168 depending on the value of the @samp{scheduler-locking} variable. In
22169 non-stop mode (@pxref{Non-Stop Mode}), if the @samp{--all} is not
22170 specified, only the thread specified with the @samp{--thread} option
22171 (or current thread, if no @samp{--thread} is provided) is resumed. If
22172 @samp{--all} is specified, all threads will be resumed. The
22173 @samp{--all} option is ignored in all-stop mode. If the
22174 @samp{--thread-group} options is specified, then all threads in that
22175 thread group are resumed.
22177 @subsubheading @value{GDBN} Command
22179 The corresponding @value{GDBN} corresponding is @samp{continue}.
22181 @subsubheading Example
22188 *stopped,reason="breakpoint-hit",disp="keep",bkptno="2",frame=@{
22189 func="foo",args=[],file="hello.c",fullname="/home/foo/bar/hello.c",
22195 @subheading The @code{-exec-finish} Command
22196 @findex -exec-finish
22198 @subsubheading Synopsis
22204 Resumes the execution of the inferior program until the current
22205 function is exited. Displays the results returned by the function.
22207 @subsubheading @value{GDBN} Command
22209 The corresponding @value{GDBN} command is @samp{finish}.
22211 @subsubheading Example
22213 Function returning @code{void}.
22220 *stopped,reason="function-finished",frame=@{func="main",args=[],
22221 file="hello.c",fullname="/home/foo/bar/hello.c",line="7"@}
22225 Function returning other than @code{void}. The name of the internal
22226 @value{GDBN} variable storing the result is printed, together with the
22233 *stopped,reason="function-finished",frame=@{addr="0x000107b0",func="foo",
22234 args=[@{name="a",value="1"],@{name="b",value="9"@}@},
22235 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
22236 gdb-result-var="$1",return-value="0"
22241 @subheading The @code{-exec-interrupt} Command
22242 @findex -exec-interrupt
22244 @subsubheading Synopsis
22247 -exec-interrupt [--all|--thread-group N]
22250 Interrupts the background execution of the target. Note how the token
22251 associated with the stop message is the one for the execution command
22252 that has been interrupted. The token for the interrupt itself only
22253 appears in the @samp{^done} output. If the user is trying to
22254 interrupt a non-running program, an error message will be printed.
22256 Note that when asynchronous execution is enabled, this command is
22257 asynchronous just like other execution commands. That is, first the
22258 @samp{^done} response will be printed, and the target stop will be
22259 reported after that using the @samp{*stopped} notification.
22261 In non-stop mode, only the context thread is interrupted by default.
22262 All threads will be interrupted if the @samp{--all} option is
22263 specified. If the @samp{--thread-group} option is specified, all
22264 threads in that group will be interrupted.
22266 @subsubheading @value{GDBN} Command
22268 The corresponding @value{GDBN} command is @samp{interrupt}.
22270 @subsubheading Example
22281 111*stopped,signal-name="SIGINT",signal-meaning="Interrupt",
22282 frame=@{addr="0x00010140",func="foo",args=[],file="try.c",
22283 fullname="/home/foo/bar/try.c",line="13"@}
22288 ^error,msg="mi_cmd_exec_interrupt: Inferior not executing."
22292 @subheading The @code{-exec-jump} Command
22295 @subsubheading Synopsis
22298 -exec-jump @var{location}
22301 Resumes execution of the inferior program at the location specified by
22302 parameter. @xref{Specify Location}, for a description of the
22303 different forms of @var{location}.
22305 @subsubheading @value{GDBN} Command
22307 The corresponding @value{GDBN} command is @samp{jump}.
22309 @subsubheading Example
22312 -exec-jump foo.c:10
22313 *running,thread-id="all"
22318 @subheading The @code{-exec-next} Command
22321 @subsubheading Synopsis
22327 Resumes execution of the inferior program, stopping when the beginning
22328 of the next source line is reached.
22330 @subsubheading @value{GDBN} Command
22332 The corresponding @value{GDBN} command is @samp{next}.
22334 @subsubheading Example
22340 *stopped,reason="end-stepping-range",line="8",file="hello.c"
22345 @subheading The @code{-exec-next-instruction} Command
22346 @findex -exec-next-instruction
22348 @subsubheading Synopsis
22351 -exec-next-instruction
22354 Executes one machine instruction. If the instruction is a function
22355 call, continues until the function returns. If the program stops at an
22356 instruction in the middle of a source line, the address will be
22359 @subsubheading @value{GDBN} Command
22361 The corresponding @value{GDBN} command is @samp{nexti}.
22363 @subsubheading Example
22367 -exec-next-instruction
22371 *stopped,reason="end-stepping-range",
22372 addr="0x000100d4",line="5",file="hello.c"
22377 @subheading The @code{-exec-return} Command
22378 @findex -exec-return
22380 @subsubheading Synopsis
22386 Makes current function return immediately. Doesn't execute the inferior.
22387 Displays the new current frame.
22389 @subsubheading @value{GDBN} Command
22391 The corresponding @value{GDBN} command is @samp{return}.
22393 @subsubheading Example
22397 200-break-insert callee4
22398 200^done,bkpt=@{number="1",addr="0x00010734",
22399 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",line="8"@}
22404 000*stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
22405 frame=@{func="callee4",args=[],
22406 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
22407 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8"@}
22413 111^done,frame=@{level="0",func="callee3",
22414 args=[@{name="strarg",
22415 value="0x11940 \"A string argument.\""@}],
22416 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
22417 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
22422 @subheading The @code{-exec-run} Command
22425 @subsubheading Synopsis
22431 Starts execution of the inferior from the beginning. The inferior
22432 executes until either a breakpoint is encountered or the program
22433 exits. In the latter case the output will include an exit code, if
22434 the program has exited exceptionally.
22436 @subsubheading @value{GDBN} Command
22438 The corresponding @value{GDBN} command is @samp{run}.
22440 @subsubheading Examples
22445 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",line="4"@}
22450 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
22451 frame=@{func="main",args=[],file="recursive2.c",
22452 fullname="/home/foo/bar/recursive2.c",line="4"@}
22457 Program exited normally:
22465 *stopped,reason="exited-normally"
22470 Program exited exceptionally:
22478 *stopped,reason="exited",exit-code="01"
22482 Another way the program can terminate is if it receives a signal such as
22483 @code{SIGINT}. In this case, @sc{gdb/mi} displays this:
22487 *stopped,reason="exited-signalled",signal-name="SIGINT",
22488 signal-meaning="Interrupt"
22492 @c @subheading -exec-signal
22495 @subheading The @code{-exec-step} Command
22498 @subsubheading Synopsis
22504 Resumes execution of the inferior program, stopping when the beginning
22505 of the next source line is reached, if the next source line is not a
22506 function call. If it is, stop at the first instruction of the called
22509 @subsubheading @value{GDBN} Command
22511 The corresponding @value{GDBN} command is @samp{step}.
22513 @subsubheading Example
22515 Stepping into a function:
22521 *stopped,reason="end-stepping-range",
22522 frame=@{func="foo",args=[@{name="a",value="10"@},
22523 @{name="b",value="0"@}],file="recursive2.c",
22524 fullname="/home/foo/bar/recursive2.c",line="11"@}
22534 *stopped,reason="end-stepping-range",line="14",file="recursive2.c"
22539 @subheading The @code{-exec-step-instruction} Command
22540 @findex -exec-step-instruction
22542 @subsubheading Synopsis
22545 -exec-step-instruction
22548 Resumes the inferior which executes one machine instruction. The
22549 output, once @value{GDBN} has stopped, will vary depending on whether
22550 we have stopped in the middle of a source line or not. In the former
22551 case, the address at which the program stopped will be printed as
22554 @subsubheading @value{GDBN} Command
22556 The corresponding @value{GDBN} command is @samp{stepi}.
22558 @subsubheading Example
22562 -exec-step-instruction
22566 *stopped,reason="end-stepping-range",
22567 frame=@{func="foo",args=[],file="try.c",
22568 fullname="/home/foo/bar/try.c",line="10"@}
22570 -exec-step-instruction
22574 *stopped,reason="end-stepping-range",
22575 frame=@{addr="0x000100f4",func="foo",args=[],file="try.c",
22576 fullname="/home/foo/bar/try.c",line="10"@}
22581 @subheading The @code{-exec-until} Command
22582 @findex -exec-until
22584 @subsubheading Synopsis
22587 -exec-until [ @var{location} ]
22590 Executes the inferior until the @var{location} specified in the
22591 argument is reached. If there is no argument, the inferior executes
22592 until a source line greater than the current one is reached. The
22593 reason for stopping in this case will be @samp{location-reached}.
22595 @subsubheading @value{GDBN} Command
22597 The corresponding @value{GDBN} command is @samp{until}.
22599 @subsubheading Example
22603 -exec-until recursive2.c:6
22607 *stopped,reason="location-reached",frame=@{func="main",args=[],
22608 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="6"@}
22613 @subheading -file-clear
22614 Is this going away????
22617 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
22618 @node GDB/MI Stack Manipulation
22619 @section @sc{gdb/mi} Stack Manipulation Commands
22622 @subheading The @code{-stack-info-frame} Command
22623 @findex -stack-info-frame
22625 @subsubheading Synopsis
22631 Get info on the selected frame.
22633 @subsubheading @value{GDBN} Command
22635 The corresponding @value{GDBN} command is @samp{info frame} or @samp{frame}
22636 (without arguments).
22638 @subsubheading Example
22643 ^done,frame=@{level="1",addr="0x0001076c",func="callee3",
22644 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
22645 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17"@}
22649 @subheading The @code{-stack-info-depth} Command
22650 @findex -stack-info-depth
22652 @subsubheading Synopsis
22655 -stack-info-depth [ @var{max-depth} ]
22658 Return the depth of the stack. If the integer argument @var{max-depth}
22659 is specified, do not count beyond @var{max-depth} frames.
22661 @subsubheading @value{GDBN} Command
22663 There's no equivalent @value{GDBN} command.
22665 @subsubheading Example
22667 For a stack with frame levels 0 through 11:
22674 -stack-info-depth 4
22677 -stack-info-depth 12
22680 -stack-info-depth 11
22683 -stack-info-depth 13
22688 @subheading The @code{-stack-list-arguments} Command
22689 @findex -stack-list-arguments
22691 @subsubheading Synopsis
22694 -stack-list-arguments @var{show-values}
22695 [ @var{low-frame} @var{high-frame} ]
22698 Display a list of the arguments for the frames between @var{low-frame}
22699 and @var{high-frame} (inclusive). If @var{low-frame} and
22700 @var{high-frame} are not provided, list the arguments for the whole
22701 call stack. If the two arguments are equal, show the single frame
22702 at the corresponding level. It is an error if @var{low-frame} is
22703 larger than the actual number of frames. On the other hand,
22704 @var{high-frame} may be larger than the actual number of frames, in
22705 which case only existing frames will be returned.
22707 The @var{show-values} argument must have a value of 0 or 1. A value of
22708 0 means that only the names of the arguments are listed, a value of 1
22709 means that both names and values of the arguments are printed.
22711 @subsubheading @value{GDBN} Command
22713 @value{GDBN} does not have an equivalent command. @code{gdbtk} has a
22714 @samp{gdb_get_args} command which partially overlaps with the
22715 functionality of @samp{-stack-list-arguments}.
22717 @subsubheading Example
22724 frame=@{level="0",addr="0x00010734",func="callee4",
22725 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
22726 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8"@},
22727 frame=@{level="1",addr="0x0001076c",func="callee3",
22728 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
22729 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17"@},
22730 frame=@{level="2",addr="0x0001078c",func="callee2",
22731 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
22732 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="22"@},
22733 frame=@{level="3",addr="0x000107b4",func="callee1",
22734 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
22735 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="27"@},
22736 frame=@{level="4",addr="0x000107e0",func="main",
22737 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
22738 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="32"@}]
22740 -stack-list-arguments 0
22743 frame=@{level="0",args=[]@},
22744 frame=@{level="1",args=[name="strarg"]@},
22745 frame=@{level="2",args=[name="intarg",name="strarg"]@},
22746 frame=@{level="3",args=[name="intarg",name="strarg",name="fltarg"]@},
22747 frame=@{level="4",args=[]@}]
22749 -stack-list-arguments 1
22752 frame=@{level="0",args=[]@},
22754 args=[@{name="strarg",value="0x11940 \"A string argument.\""@}]@},
22755 frame=@{level="2",args=[
22756 @{name="intarg",value="2"@},
22757 @{name="strarg",value="0x11940 \"A string argument.\""@}]@},
22758 @{frame=@{level="3",args=[
22759 @{name="intarg",value="2"@},
22760 @{name="strarg",value="0x11940 \"A string argument.\""@},
22761 @{name="fltarg",value="3.5"@}]@},
22762 frame=@{level="4",args=[]@}]
22764 -stack-list-arguments 0 2 2
22765 ^done,stack-args=[frame=@{level="2",args=[name="intarg",name="strarg"]@}]
22767 -stack-list-arguments 1 2 2
22768 ^done,stack-args=[frame=@{level="2",
22769 args=[@{name="intarg",value="2"@},
22770 @{name="strarg",value="0x11940 \"A string argument.\""@}]@}]
22774 @c @subheading -stack-list-exception-handlers
22777 @subheading The @code{-stack-list-frames} Command
22778 @findex -stack-list-frames
22780 @subsubheading Synopsis
22783 -stack-list-frames [ @var{low-frame} @var{high-frame} ]
22786 List the frames currently on the stack. For each frame it displays the
22791 The frame number, 0 being the topmost frame, i.e., the innermost function.
22793 The @code{$pc} value for that frame.
22797 File name of the source file where the function lives.
22799 Line number corresponding to the @code{$pc}.
22802 If invoked without arguments, this command prints a backtrace for the
22803 whole stack. If given two integer arguments, it shows the frames whose
22804 levels are between the two arguments (inclusive). If the two arguments
22805 are equal, it shows the single frame at the corresponding level. It is
22806 an error if @var{low-frame} is larger than the actual number of
22807 frames. On the other hand, @var{high-frame} may be larger than the
22808 actual number of frames, in which case only existing frames will be returned.
22810 @subsubheading @value{GDBN} Command
22812 The corresponding @value{GDBN} commands are @samp{backtrace} and @samp{where}.
22814 @subsubheading Example
22816 Full stack backtrace:
22822 [frame=@{level="0",addr="0x0001076c",func="foo",
22823 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="11"@},
22824 frame=@{level="1",addr="0x000107a4",func="foo",
22825 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
22826 frame=@{level="2",addr="0x000107a4",func="foo",
22827 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
22828 frame=@{level="3",addr="0x000107a4",func="foo",
22829 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
22830 frame=@{level="4",addr="0x000107a4",func="foo",
22831 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
22832 frame=@{level="5",addr="0x000107a4",func="foo",
22833 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
22834 frame=@{level="6",addr="0x000107a4",func="foo",
22835 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
22836 frame=@{level="7",addr="0x000107a4",func="foo",
22837 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
22838 frame=@{level="8",addr="0x000107a4",func="foo",
22839 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
22840 frame=@{level="9",addr="0x000107a4",func="foo",
22841 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
22842 frame=@{level="10",addr="0x000107a4",func="foo",
22843 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
22844 frame=@{level="11",addr="0x00010738",func="main",
22845 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="4"@}]
22849 Show frames between @var{low_frame} and @var{high_frame}:
22853 -stack-list-frames 3 5
22855 [frame=@{level="3",addr="0x000107a4",func="foo",
22856 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
22857 frame=@{level="4",addr="0x000107a4",func="foo",
22858 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
22859 frame=@{level="5",addr="0x000107a4",func="foo",
22860 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@}]
22864 Show a single frame:
22868 -stack-list-frames 3 3
22870 [frame=@{level="3",addr="0x000107a4",func="foo",
22871 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@}]
22876 @subheading The @code{-stack-list-locals} Command
22877 @findex -stack-list-locals
22879 @subsubheading Synopsis
22882 -stack-list-locals @var{print-values}
22885 Display the local variable names for the selected frame. If
22886 @var{print-values} is 0 or @code{--no-values}, print only the names of
22887 the variables; if it is 1 or @code{--all-values}, print also their
22888 values; and if it is 2 or @code{--simple-values}, print the name,
22889 type and value for simple data types and the name and type for arrays,
22890 structures and unions. In this last case, a frontend can immediately
22891 display the value of simple data types and create variable objects for
22892 other data types when the user wishes to explore their values in
22895 @subsubheading @value{GDBN} Command
22897 @samp{info locals} in @value{GDBN}, @samp{gdb_get_locals} in @code{gdbtk}.
22899 @subsubheading Example
22903 -stack-list-locals 0
22904 ^done,locals=[name="A",name="B",name="C"]
22906 -stack-list-locals --all-values
22907 ^done,locals=[@{name="A",value="1"@},@{name="B",value="2"@},
22908 @{name="C",value="@{1, 2, 3@}"@}]
22909 -stack-list-locals --simple-values
22910 ^done,locals=[@{name="A",type="int",value="1"@},
22911 @{name="B",type="int",value="2"@},@{name="C",type="int [3]"@}]
22916 @subheading The @code{-stack-select-frame} Command
22917 @findex -stack-select-frame
22919 @subsubheading Synopsis
22922 -stack-select-frame @var{framenum}
22925 Change the selected frame. Select a different frame @var{framenum} on
22928 This command in deprecated in favor of passing the @samp{--frame}
22929 option to every command.
22931 @subsubheading @value{GDBN} Command
22933 The corresponding @value{GDBN} commands are @samp{frame}, @samp{up},
22934 @samp{down}, @samp{select-frame}, @samp{up-silent}, and @samp{down-silent}.
22936 @subsubheading Example
22940 -stack-select-frame 2
22945 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
22946 @node GDB/MI Variable Objects
22947 @section @sc{gdb/mi} Variable Objects
22951 @subheading Motivation for Variable Objects in @sc{gdb/mi}
22953 For the implementation of a variable debugger window (locals, watched
22954 expressions, etc.), we are proposing the adaptation of the existing code
22955 used by @code{Insight}.
22957 The two main reasons for that are:
22961 It has been proven in practice (it is already on its second generation).
22964 It will shorten development time (needless to say how important it is
22968 The original interface was designed to be used by Tcl code, so it was
22969 slightly changed so it could be used through @sc{gdb/mi}. This section
22970 describes the @sc{gdb/mi} operations that will be available and gives some
22971 hints about their use.
22973 @emph{Note}: In addition to the set of operations described here, we
22974 expect the @sc{gui} implementation of a variable window to require, at
22975 least, the following operations:
22978 @item @code{-gdb-show} @code{output-radix}
22979 @item @code{-stack-list-arguments}
22980 @item @code{-stack-list-locals}
22981 @item @code{-stack-select-frame}
22986 @subheading Introduction to Variable Objects
22988 @cindex variable objects in @sc{gdb/mi}
22990 Variable objects are "object-oriented" MI interface for examining and
22991 changing values of expressions. Unlike some other MI interfaces that
22992 work with expressions, variable objects are specifically designed for
22993 simple and efficient presentation in the frontend. A variable object
22994 is identified by string name. When a variable object is created, the
22995 frontend specifies the expression for that variable object. The
22996 expression can be a simple variable, or it can be an arbitrary complex
22997 expression, and can even involve CPU registers. After creating a
22998 variable object, the frontend can invoke other variable object
22999 operations---for example to obtain or change the value of a variable
23000 object, or to change display format.
23002 Variable objects have hierarchical tree structure. Any variable object
23003 that corresponds to a composite type, such as structure in C, has
23004 a number of child variable objects, for example corresponding to each
23005 element of a structure. A child variable object can itself have
23006 children, recursively. Recursion ends when we reach
23007 leaf variable objects, which always have built-in types. Child variable
23008 objects are created only by explicit request, so if a frontend
23009 is not interested in the children of a particular variable object, no
23010 child will be created.
23012 For a leaf variable object it is possible to obtain its value as a
23013 string, or set the value from a string. String value can be also
23014 obtained for a non-leaf variable object, but it's generally a string
23015 that only indicates the type of the object, and does not list its
23016 contents. Assignment to a non-leaf variable object is not allowed.
23018 A frontend does not need to read the values of all variable objects each time
23019 the program stops. Instead, MI provides an update command that lists all
23020 variable objects whose values has changed since the last update
23021 operation. This considerably reduces the amount of data that must
23022 be transferred to the frontend. As noted above, children variable
23023 objects are created on demand, and only leaf variable objects have a
23024 real value. As result, gdb will read target memory only for leaf
23025 variables that frontend has created.
23027 The automatic update is not always desirable. For example, a frontend
23028 might want to keep a value of some expression for future reference,
23029 and never update it. For another example, fetching memory is
23030 relatively slow for embedded targets, so a frontend might want
23031 to disable automatic update for the variables that are either not
23032 visible on the screen, or ``closed''. This is possible using so
23033 called ``frozen variable objects''. Such variable objects are never
23034 implicitly updated.
23036 Variable objects can be either @dfn{fixed} or @dfn{floating}. For the
23037 fixed variable object, the expression is parsed when the variable
23038 object is created, including associating identifiers to specific
23039 variables. The meaning of expression never changes. For a floating
23040 variable object the values of variables whose names appear in the
23041 expressions are re-evaluated every time in the context of the current
23042 frame. Consider this example:
23047 struct work_state state;
23054 If a fixed variable object for the @code{state} variable is created in
23055 this function, and we enter the recursive call, the the variable
23056 object will report the value of @code{state} in the top-level
23057 @code{do_work} invocation. On the other hand, a floating variable
23058 object will report the value of @code{state} in the current frame.
23060 If an expression specified when creating a fixed variable object
23061 refers to a local variable, the variable object becomes bound to the
23062 thread and frame in which the variable object is created. When such
23063 variable object is updated, @value{GDBN} makes sure that the
23064 thread/frame combination the variable object is bound to still exists,
23065 and re-evaluates the variable object in context of that thread/frame.
23067 The following is the complete set of @sc{gdb/mi} operations defined to
23068 access this functionality:
23070 @multitable @columnfractions .4 .6
23071 @item @strong{Operation}
23072 @tab @strong{Description}
23074 @item @code{-var-create}
23075 @tab create a variable object
23076 @item @code{-var-delete}
23077 @tab delete the variable object and/or its children
23078 @item @code{-var-set-format}
23079 @tab set the display format of this variable
23080 @item @code{-var-show-format}
23081 @tab show the display format of this variable
23082 @item @code{-var-info-num-children}
23083 @tab tells how many children this object has
23084 @item @code{-var-list-children}
23085 @tab return a list of the object's children
23086 @item @code{-var-info-type}
23087 @tab show the type of this variable object
23088 @item @code{-var-info-expression}
23089 @tab print parent-relative expression that this variable object represents
23090 @item @code{-var-info-path-expression}
23091 @tab print full expression that this variable object represents
23092 @item @code{-var-show-attributes}
23093 @tab is this variable editable? does it exist here?
23094 @item @code{-var-evaluate-expression}
23095 @tab get the value of this variable
23096 @item @code{-var-assign}
23097 @tab set the value of this variable
23098 @item @code{-var-update}
23099 @tab update the variable and its children
23100 @item @code{-var-set-frozen}
23101 @tab set frozeness attribute
23104 In the next subsection we describe each operation in detail and suggest
23105 how it can be used.
23107 @subheading Description And Use of Operations on Variable Objects
23109 @subheading The @code{-var-create} Command
23110 @findex -var-create
23112 @subsubheading Synopsis
23115 -var-create @{@var{name} | "-"@}
23116 @{@var{frame-addr} | "*" | "@@"@} @var{expression}
23119 This operation creates a variable object, which allows the monitoring of
23120 a variable, the result of an expression, a memory cell or a CPU
23123 The @var{name} parameter is the string by which the object can be
23124 referenced. It must be unique. If @samp{-} is specified, the varobj
23125 system will generate a string ``varNNNNNN'' automatically. It will be
23126 unique provided that one does not specify @var{name} of that format.
23127 The command fails if a duplicate name is found.
23129 The frame under which the expression should be evaluated can be
23130 specified by @var{frame-addr}. A @samp{*} indicates that the current
23131 frame should be used. A @samp{@@} indicates that a floating variable
23132 object must be created.
23134 @var{expression} is any expression valid on the current language set (must not
23135 begin with a @samp{*}), or one of the following:
23139 @samp{*@var{addr}}, where @var{addr} is the address of a memory cell
23142 @samp{*@var{addr}-@var{addr}} --- a memory address range (TBD)
23145 @samp{$@var{regname}} --- a CPU register name
23148 @subsubheading Result
23150 This operation returns the name, number of children and the type of the
23151 object created. Type is returned as a string as the ones generated by
23152 the @value{GDBN} CLI. If a fixed variable object is bound to a
23153 specific thread, the thread is is also printed:
23156 name="@var{name}",numchild="@var{N}",type="@var{type}",thread-id="@var{M}"
23160 @subheading The @code{-var-delete} Command
23161 @findex -var-delete
23163 @subsubheading Synopsis
23166 -var-delete [ -c ] @var{name}
23169 Deletes a previously created variable object and all of its children.
23170 With the @samp{-c} option, just deletes the children.
23172 Returns an error if the object @var{name} is not found.
23175 @subheading The @code{-var-set-format} Command
23176 @findex -var-set-format
23178 @subsubheading Synopsis
23181 -var-set-format @var{name} @var{format-spec}
23184 Sets the output format for the value of the object @var{name} to be
23187 @anchor{-var-set-format}
23188 The syntax for the @var{format-spec} is as follows:
23191 @var{format-spec} @expansion{}
23192 @{binary | decimal | hexadecimal | octal | natural@}
23195 The natural format is the default format choosen automatically
23196 based on the variable type (like decimal for an @code{int}, hex
23197 for pointers, etc.).
23199 For a variable with children, the format is set only on the
23200 variable itself, and the children are not affected.
23202 @subheading The @code{-var-show-format} Command
23203 @findex -var-show-format
23205 @subsubheading Synopsis
23208 -var-show-format @var{name}
23211 Returns the format used to display the value of the object @var{name}.
23214 @var{format} @expansion{}
23219 @subheading The @code{-var-info-num-children} Command
23220 @findex -var-info-num-children
23222 @subsubheading Synopsis
23225 -var-info-num-children @var{name}
23228 Returns the number of children of a variable object @var{name}:
23235 @subheading The @code{-var-list-children} Command
23236 @findex -var-list-children
23238 @subsubheading Synopsis
23241 -var-list-children [@var{print-values}] @var{name}
23243 @anchor{-var-list-children}
23245 Return a list of the children of the specified variable object and
23246 create variable objects for them, if they do not already exist. With
23247 a single argument or if @var{print-values} has a value for of 0 or
23248 @code{--no-values}, print only the names of the variables; if
23249 @var{print-values} is 1 or @code{--all-values}, also print their
23250 values; and if it is 2 or @code{--simple-values} print the name and
23251 value for simple data types and just the name for arrays, structures
23254 @subsubheading Example
23258 -var-list-children n
23259 ^done,numchild=@var{n},children=[@{name=@var{name},
23260 numchild=@var{n},type=@var{type}@},@r{(repeats N times)}]
23262 -var-list-children --all-values n
23263 ^done,numchild=@var{n},children=[@{name=@var{name},
23264 numchild=@var{n},value=@var{value},type=@var{type}@},@r{(repeats N times)}]
23268 @subheading The @code{-var-info-type} Command
23269 @findex -var-info-type
23271 @subsubheading Synopsis
23274 -var-info-type @var{name}
23277 Returns the type of the specified variable @var{name}. The type is
23278 returned as a string in the same format as it is output by the
23282 type=@var{typename}
23286 @subheading The @code{-var-info-expression} Command
23287 @findex -var-info-expression
23289 @subsubheading Synopsis
23292 -var-info-expression @var{name}
23295 Returns a string that is suitable for presenting this
23296 variable object in user interface. The string is generally
23297 not valid expression in the current language, and cannot be evaluated.
23299 For example, if @code{a} is an array, and variable object
23300 @code{A} was created for @code{a}, then we'll get this output:
23303 (gdb) -var-info-expression A.1
23304 ^done,lang="C",exp="1"
23308 Here, the values of @code{lang} can be @code{@{"C" | "C++" | "Java"@}}.
23310 Note that the output of the @code{-var-list-children} command also
23311 includes those expressions, so the @code{-var-info-expression} command
23314 @subheading The @code{-var-info-path-expression} Command
23315 @findex -var-info-path-expression
23317 @subsubheading Synopsis
23320 -var-info-path-expression @var{name}
23323 Returns an expression that can be evaluated in the current
23324 context and will yield the same value that a variable object has.
23325 Compare this with the @code{-var-info-expression} command, which
23326 result can be used only for UI presentation. Typical use of
23327 the @code{-var-info-path-expression} command is creating a
23328 watchpoint from a variable object.
23330 For example, suppose @code{C} is a C@t{++} class, derived from class
23331 @code{Base}, and that the @code{Base} class has a member called
23332 @code{m_size}. Assume a variable @code{c} is has the type of
23333 @code{C} and a variable object @code{C} was created for variable
23334 @code{c}. Then, we'll get this output:
23336 (gdb) -var-info-path-expression C.Base.public.m_size
23337 ^done,path_expr=((Base)c).m_size)
23340 @subheading The @code{-var-show-attributes} Command
23341 @findex -var-show-attributes
23343 @subsubheading Synopsis
23346 -var-show-attributes @var{name}
23349 List attributes of the specified variable object @var{name}:
23352 status=@var{attr} [ ( ,@var{attr} )* ]
23356 where @var{attr} is @code{@{ @{ editable | noneditable @} | TBD @}}.
23358 @subheading The @code{-var-evaluate-expression} Command
23359 @findex -var-evaluate-expression
23361 @subsubheading Synopsis
23364 -var-evaluate-expression [-f @var{format-spec}] @var{name}
23367 Evaluates the expression that is represented by the specified variable
23368 object and returns its value as a string. The format of the string
23369 can be specified with the @samp{-f} option. The possible values of
23370 this option are the same as for @code{-var-set-format}
23371 (@pxref{-var-set-format}). If the @samp{-f} option is not specified,
23372 the current display format will be used. The current display format
23373 can be changed using the @code{-var-set-format} command.
23379 Note that one must invoke @code{-var-list-children} for a variable
23380 before the value of a child variable can be evaluated.
23382 @subheading The @code{-var-assign} Command
23383 @findex -var-assign
23385 @subsubheading Synopsis
23388 -var-assign @var{name} @var{expression}
23391 Assigns the value of @var{expression} to the variable object specified
23392 by @var{name}. The object must be @samp{editable}. If the variable's
23393 value is altered by the assign, the variable will show up in any
23394 subsequent @code{-var-update} list.
23396 @subsubheading Example
23404 ^done,changelist=[@{name="var1",in_scope="true",type_changed="false"@}]
23408 @subheading The @code{-var-update} Command
23409 @findex -var-update
23411 @subsubheading Synopsis
23414 -var-update [@var{print-values}] @{@var{name} | "*"@}
23417 Reevaluate the expressions corresponding to the variable object
23418 @var{name} and all its direct and indirect children, and return the
23419 list of variable objects whose values have changed; @var{name} must
23420 be a root variable object. Here, ``changed'' means that the result of
23421 @code{-var-evaluate-expression} before and after the
23422 @code{-var-update} is different. If @samp{*} is used as the variable
23423 object names, all existing variable objects are updated, except
23424 for frozen ones (@pxref{-var-set-frozen}). The option
23425 @var{print-values} determines whether both names and values, or just
23426 names are printed. The possible values of this option are the same
23427 as for @code{-var-list-children} (@pxref{-var-list-children}). It is
23428 recommended to use the @samp{--all-values} option, to reduce the
23429 number of MI commands needed on each program stop.
23431 With the @samp{*} parameter, if a variable object is bound to a
23432 currently running thread, it will not be updated, without any
23435 @subsubheading Example
23442 -var-update --all-values var1
23443 ^done,changelist=[@{name="var1",value="3",in_scope="true",
23444 type_changed="false"@}]
23448 @anchor{-var-update}
23449 The field in_scope may take three values:
23453 The variable object's current value is valid.
23456 The variable object does not currently hold a valid value but it may
23457 hold one in the future if its associated expression comes back into
23461 The variable object no longer holds a valid value.
23462 This can occur when the executable file being debugged has changed,
23463 either through recompilation or by using the @value{GDBN} @code{file}
23464 command. The front end should normally choose to delete these variable
23468 In the future new values may be added to this list so the front should
23469 be prepared for this possibility. @xref{GDB/MI Development and Front Ends, ,@sc{GDB/MI} Development and Front Ends}.
23471 @subheading The @code{-var-set-frozen} Command
23472 @findex -var-set-frozen
23473 @anchor{-var-set-frozen}
23475 @subsubheading Synopsis
23478 -var-set-frozen @var{name} @var{flag}
23481 Set the frozenness flag on the variable object @var{name}. The
23482 @var{flag} parameter should be either @samp{1} to make the variable
23483 frozen or @samp{0} to make it unfrozen. If a variable object is
23484 frozen, then neither itself, nor any of its children, are
23485 implicitly updated by @code{-var-update} of
23486 a parent variable or by @code{-var-update *}. Only
23487 @code{-var-update} of the variable itself will update its value and
23488 values of its children. After a variable object is unfrozen, it is
23489 implicitly updated by all subsequent @code{-var-update} operations.
23490 Unfreezing a variable does not update it, only subsequent
23491 @code{-var-update} does.
23493 @subsubheading Example
23497 -var-set-frozen V 1
23502 @subheading The @code{-var-set-visualizer} command
23503 @findex -var-set-visualizer
23504 @anchor{-var-set-visualizer}
23506 @subsubheading Synopsis
23509 -var-set-visualizer @var{name} @var{visualizer}
23512 Set a visualizer for the variable object @var{name}.
23514 @var{visualizer} is the visualizer to use. The special value
23515 @samp{None} means to disable any visualizer in use.
23517 If not @samp{None}, @var{visualizer} must be a Python expression.
23518 This expression must evaluate to a callable object which accepts a
23519 single argument. @value{GDBN} will call this object with the value of
23520 the varobj @var{name} as an argument (this is done so that the same
23521 Python pretty-printing code can be used for both the CLI and MI).
23522 When called, this object must return an object which conforms to the
23523 pretty-printing interface (@pxref{Pretty Printing}).
23525 The pre-defined function @code{gdb.default_visualizer} may be used to
23526 select a visualizer by following the built-in process
23527 (@pxref{Selecting Pretty-Printers}). This is done automatically when
23528 a varobj is created, and so ordinarily is not needed.
23530 This feature is only available if Python support is enabled. The MI
23531 command @code{-list-features} (@pxref{GDB/MI Miscellaneous Commands})
23532 can be used to check this.
23534 @subsubheading Example
23536 Resetting the visualizer:
23540 -var-set-visualizer V None
23544 Reselecting the default (type-based) visualizer:
23548 -var-set-visualizer V gdb.default_visualizer
23552 Suppose @code{SomeClass} is a visualizer class. A lambda expression
23553 can be used to instantiate this class for a varobj:
23557 -var-set-visualizer V "lambda val: SomeClass()"
23561 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
23562 @node GDB/MI Data Manipulation
23563 @section @sc{gdb/mi} Data Manipulation
23565 @cindex data manipulation, in @sc{gdb/mi}
23566 @cindex @sc{gdb/mi}, data manipulation
23567 This section describes the @sc{gdb/mi} commands that manipulate data:
23568 examine memory and registers, evaluate expressions, etc.
23570 @c REMOVED FROM THE INTERFACE.
23571 @c @subheading -data-assign
23572 @c Change the value of a program variable. Plenty of side effects.
23573 @c @subsubheading GDB Command
23575 @c @subsubheading Example
23578 @subheading The @code{-data-disassemble} Command
23579 @findex -data-disassemble
23581 @subsubheading Synopsis
23585 [ -s @var{start-addr} -e @var{end-addr} ]
23586 | [ -f @var{filename} -l @var{linenum} [ -n @var{lines} ] ]
23594 @item @var{start-addr}
23595 is the beginning address (or @code{$pc})
23596 @item @var{end-addr}
23598 @item @var{filename}
23599 is the name of the file to disassemble
23600 @item @var{linenum}
23601 is the line number to disassemble around
23603 is the number of disassembly lines to be produced. If it is -1,
23604 the whole function will be disassembled, in case no @var{end-addr} is
23605 specified. If @var{end-addr} is specified as a non-zero value, and
23606 @var{lines} is lower than the number of disassembly lines between
23607 @var{start-addr} and @var{end-addr}, only @var{lines} lines are
23608 displayed; if @var{lines} is higher than the number of lines between
23609 @var{start-addr} and @var{end-addr}, only the lines up to @var{end-addr}
23612 is either 0 (meaning only disassembly) or 1 (meaning mixed source and
23616 @subsubheading Result
23618 The output for each instruction is composed of four fields:
23627 Note that whatever included in the instruction field, is not manipulated
23628 directly by @sc{gdb/mi}, i.e., it is not possible to adjust its format.
23630 @subsubheading @value{GDBN} Command
23632 There's no direct mapping from this command to the CLI.
23634 @subsubheading Example
23636 Disassemble from the current value of @code{$pc} to @code{$pc + 20}:
23640 -data-disassemble -s $pc -e "$pc + 20" -- 0
23643 @{address="0x000107c0",func-name="main",offset="4",
23644 inst="mov 2, %o0"@},
23645 @{address="0x000107c4",func-name="main",offset="8",
23646 inst="sethi %hi(0x11800), %o2"@},
23647 @{address="0x000107c8",func-name="main",offset="12",
23648 inst="or %o2, 0x140, %o1\t! 0x11940 <_lib_version+8>"@},
23649 @{address="0x000107cc",func-name="main",offset="16",
23650 inst="sethi %hi(0x11800), %o2"@},
23651 @{address="0x000107d0",func-name="main",offset="20",
23652 inst="or %o2, 0x168, %o4\t! 0x11968 <_lib_version+48>"@}]
23656 Disassemble the whole @code{main} function. Line 32 is part of
23660 -data-disassemble -f basics.c -l 32 -- 0
23662 @{address="0x000107bc",func-name="main",offset="0",
23663 inst="save %sp, -112, %sp"@},
23664 @{address="0x000107c0",func-name="main",offset="4",
23665 inst="mov 2, %o0"@},
23666 @{address="0x000107c4",func-name="main",offset="8",
23667 inst="sethi %hi(0x11800), %o2"@},
23669 @{address="0x0001081c",func-name="main",offset="96",inst="ret "@},
23670 @{address="0x00010820",func-name="main",offset="100",inst="restore "@}]
23674 Disassemble 3 instructions from the start of @code{main}:
23678 -data-disassemble -f basics.c -l 32 -n 3 -- 0
23680 @{address="0x000107bc",func-name="main",offset="0",
23681 inst="save %sp, -112, %sp"@},
23682 @{address="0x000107c0",func-name="main",offset="4",
23683 inst="mov 2, %o0"@},
23684 @{address="0x000107c4",func-name="main",offset="8",
23685 inst="sethi %hi(0x11800), %o2"@}]
23689 Disassemble 3 instructions from the start of @code{main} in mixed mode:
23693 -data-disassemble -f basics.c -l 32 -n 3 -- 1
23695 src_and_asm_line=@{line="31",
23696 file="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb/ \
23697 testsuite/gdb.mi/basics.c",line_asm_insn=[
23698 @{address="0x000107bc",func-name="main",offset="0",
23699 inst="save %sp, -112, %sp"@}]@},
23700 src_and_asm_line=@{line="32",
23701 file="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb/ \
23702 testsuite/gdb.mi/basics.c",line_asm_insn=[
23703 @{address="0x000107c0",func-name="main",offset="4",
23704 inst="mov 2, %o0"@},
23705 @{address="0x000107c4",func-name="main",offset="8",
23706 inst="sethi %hi(0x11800), %o2"@}]@}]
23711 @subheading The @code{-data-evaluate-expression} Command
23712 @findex -data-evaluate-expression
23714 @subsubheading Synopsis
23717 -data-evaluate-expression @var{expr}
23720 Evaluate @var{expr} as an expression. The expression could contain an
23721 inferior function call. The function call will execute synchronously.
23722 If the expression contains spaces, it must be enclosed in double quotes.
23724 @subsubheading @value{GDBN} Command
23726 The corresponding @value{GDBN} commands are @samp{print}, @samp{output}, and
23727 @samp{call}. In @code{gdbtk} only, there's a corresponding
23728 @samp{gdb_eval} command.
23730 @subsubheading Example
23732 In the following example, the numbers that precede the commands are the
23733 @dfn{tokens} described in @ref{GDB/MI Command Syntax, ,@sc{gdb/mi}
23734 Command Syntax}. Notice how @sc{gdb/mi} returns the same tokens in its
23738 211-data-evaluate-expression A
23741 311-data-evaluate-expression &A
23742 311^done,value="0xefffeb7c"
23744 411-data-evaluate-expression A+3
23747 511-data-evaluate-expression "A + 3"
23753 @subheading The @code{-data-list-changed-registers} Command
23754 @findex -data-list-changed-registers
23756 @subsubheading Synopsis
23759 -data-list-changed-registers
23762 Display a list of the registers that have changed.
23764 @subsubheading @value{GDBN} Command
23766 @value{GDBN} doesn't have a direct analog for this command; @code{gdbtk}
23767 has the corresponding command @samp{gdb_changed_register_list}.
23769 @subsubheading Example
23771 On a PPC MBX board:
23779 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",frame=@{
23780 func="main",args=[],file="try.c",fullname="/home/foo/bar/try.c",
23783 -data-list-changed-registers
23784 ^done,changed-registers=["0","1","2","4","5","6","7","8","9",
23785 "10","11","13","14","15","16","17","18","19","20","21","22","23",
23786 "24","25","26","27","28","30","31","64","65","66","67","69"]
23791 @subheading The @code{-data-list-register-names} Command
23792 @findex -data-list-register-names
23794 @subsubheading Synopsis
23797 -data-list-register-names [ ( @var{regno} )+ ]
23800 Show a list of register names for the current target. If no arguments
23801 are given, it shows a list of the names of all the registers. If
23802 integer numbers are given as arguments, it will print a list of the
23803 names of the registers corresponding to the arguments. To ensure
23804 consistency between a register name and its number, the output list may
23805 include empty register names.
23807 @subsubheading @value{GDBN} Command
23809 @value{GDBN} does not have a command which corresponds to
23810 @samp{-data-list-register-names}. In @code{gdbtk} there is a
23811 corresponding command @samp{gdb_regnames}.
23813 @subsubheading Example
23815 For the PPC MBX board:
23818 -data-list-register-names
23819 ^done,register-names=["r0","r1","r2","r3","r4","r5","r6","r7",
23820 "r8","r9","r10","r11","r12","r13","r14","r15","r16","r17","r18",
23821 "r19","r20","r21","r22","r23","r24","r25","r26","r27","r28","r29",
23822 "r30","r31","f0","f1","f2","f3","f4","f5","f6","f7","f8","f9",
23823 "f10","f11","f12","f13","f14","f15","f16","f17","f18","f19","f20",
23824 "f21","f22","f23","f24","f25","f26","f27","f28","f29","f30","f31",
23825 "", "pc","ps","cr","lr","ctr","xer"]
23827 -data-list-register-names 1 2 3
23828 ^done,register-names=["r1","r2","r3"]
23832 @subheading The @code{-data-list-register-values} Command
23833 @findex -data-list-register-values
23835 @subsubheading Synopsis
23838 -data-list-register-values @var{fmt} [ ( @var{regno} )*]
23841 Display the registers' contents. @var{fmt} is the format according to
23842 which the registers' contents are to be returned, followed by an optional
23843 list of numbers specifying the registers to display. A missing list of
23844 numbers indicates that the contents of all the registers must be returned.
23846 Allowed formats for @var{fmt} are:
23863 @subsubheading @value{GDBN} Command
23865 The corresponding @value{GDBN} commands are @samp{info reg}, @samp{info
23866 all-reg}, and (in @code{gdbtk}) @samp{gdb_fetch_registers}.
23868 @subsubheading Example
23870 For a PPC MBX board (note: line breaks are for readability only, they
23871 don't appear in the actual output):
23875 -data-list-register-values r 64 65
23876 ^done,register-values=[@{number="64",value="0xfe00a300"@},
23877 @{number="65",value="0x00029002"@}]
23879 -data-list-register-values x
23880 ^done,register-values=[@{number="0",value="0xfe0043c8"@},
23881 @{number="1",value="0x3fff88"@},@{number="2",value="0xfffffffe"@},
23882 @{number="3",value="0x0"@},@{number="4",value="0xa"@},
23883 @{number="5",value="0x3fff68"@},@{number="6",value="0x3fff58"@},
23884 @{number="7",value="0xfe011e98"@},@{number="8",value="0x2"@},
23885 @{number="9",value="0xfa202820"@},@{number="10",value="0xfa202808"@},
23886 @{number="11",value="0x1"@},@{number="12",value="0x0"@},
23887 @{number="13",value="0x4544"@},@{number="14",value="0xffdfffff"@},
23888 @{number="15",value="0xffffffff"@},@{number="16",value="0xfffffeff"@},
23889 @{number="17",value="0xefffffed"@},@{number="18",value="0xfffffffe"@},
23890 @{number="19",value="0xffffffff"@},@{number="20",value="0xffffffff"@},
23891 @{number="21",value="0xffffffff"@},@{number="22",value="0xfffffff7"@},
23892 @{number="23",value="0xffffffff"@},@{number="24",value="0xffffffff"@},
23893 @{number="25",value="0xffffffff"@},@{number="26",value="0xfffffffb"@},
23894 @{number="27",value="0xffffffff"@},@{number="28",value="0xf7bfffff"@},
23895 @{number="29",value="0x0"@},@{number="30",value="0xfe010000"@},
23896 @{number="31",value="0x0"@},@{number="32",value="0x0"@},
23897 @{number="33",value="0x0"@},@{number="34",value="0x0"@},
23898 @{number="35",value="0x0"@},@{number="36",value="0x0"@},
23899 @{number="37",value="0x0"@},@{number="38",value="0x0"@},
23900 @{number="39",value="0x0"@},@{number="40",value="0x0"@},
23901 @{number="41",value="0x0"@},@{number="42",value="0x0"@},
23902 @{number="43",value="0x0"@},@{number="44",value="0x0"@},
23903 @{number="45",value="0x0"@},@{number="46",value="0x0"@},
23904 @{number="47",value="0x0"@},@{number="48",value="0x0"@},
23905 @{number="49",value="0x0"@},@{number="50",value="0x0"@},
23906 @{number="51",value="0x0"@},@{number="52",value="0x0"@},
23907 @{number="53",value="0x0"@},@{number="54",value="0x0"@},
23908 @{number="55",value="0x0"@},@{number="56",value="0x0"@},
23909 @{number="57",value="0x0"@},@{number="58",value="0x0"@},
23910 @{number="59",value="0x0"@},@{number="60",value="0x0"@},
23911 @{number="61",value="0x0"@},@{number="62",value="0x0"@},
23912 @{number="63",value="0x0"@},@{number="64",value="0xfe00a300"@},
23913 @{number="65",value="0x29002"@},@{number="66",value="0x202f04b5"@},
23914 @{number="67",value="0xfe0043b0"@},@{number="68",value="0xfe00b3e4"@},
23915 @{number="69",value="0x20002b03"@}]
23920 @subheading The @code{-data-read-memory} Command
23921 @findex -data-read-memory
23923 @subsubheading Synopsis
23926 -data-read-memory [ -o @var{byte-offset} ]
23927 @var{address} @var{word-format} @var{word-size}
23928 @var{nr-rows} @var{nr-cols} [ @var{aschar} ]
23935 @item @var{address}
23936 An expression specifying the address of the first memory word to be
23937 read. Complex expressions containing embedded white space should be
23938 quoted using the C convention.
23940 @item @var{word-format}
23941 The format to be used to print the memory words. The notation is the
23942 same as for @value{GDBN}'s @code{print} command (@pxref{Output Formats,
23945 @item @var{word-size}
23946 The size of each memory word in bytes.
23948 @item @var{nr-rows}
23949 The number of rows in the output table.
23951 @item @var{nr-cols}
23952 The number of columns in the output table.
23955 If present, indicates that each row should include an @sc{ascii} dump. The
23956 value of @var{aschar} is used as a padding character when a byte is not a
23957 member of the printable @sc{ascii} character set (printable @sc{ascii}
23958 characters are those whose code is between 32 and 126, inclusively).
23960 @item @var{byte-offset}
23961 An offset to add to the @var{address} before fetching memory.
23964 This command displays memory contents as a table of @var{nr-rows} by
23965 @var{nr-cols} words, each word being @var{word-size} bytes. In total,
23966 @code{@var{nr-rows} * @var{nr-cols} * @var{word-size}} bytes are read
23967 (returned as @samp{total-bytes}). Should less than the requested number
23968 of bytes be returned by the target, the missing words are identified
23969 using @samp{N/A}. The number of bytes read from the target is returned
23970 in @samp{nr-bytes} and the starting address used to read memory in
23973 The address of the next/previous row or page is available in
23974 @samp{next-row} and @samp{prev-row}, @samp{next-page} and
23977 @subsubheading @value{GDBN} Command
23979 The corresponding @value{GDBN} command is @samp{x}. @code{gdbtk} has
23980 @samp{gdb_get_mem} memory read command.
23982 @subsubheading Example
23984 Read six bytes of memory starting at @code{bytes+6} but then offset by
23985 @code{-6} bytes. Format as three rows of two columns. One byte per
23986 word. Display each word in hex.
23990 9-data-read-memory -o -6 -- bytes+6 x 1 3 2
23991 9^done,addr="0x00001390",nr-bytes="6",total-bytes="6",
23992 next-row="0x00001396",prev-row="0x0000138e",next-page="0x00001396",
23993 prev-page="0x0000138a",memory=[
23994 @{addr="0x00001390",data=["0x00","0x01"]@},
23995 @{addr="0x00001392",data=["0x02","0x03"]@},
23996 @{addr="0x00001394",data=["0x04","0x05"]@}]
24000 Read two bytes of memory starting at address @code{shorts + 64} and
24001 display as a single word formatted in decimal.
24005 5-data-read-memory shorts+64 d 2 1 1
24006 5^done,addr="0x00001510",nr-bytes="2",total-bytes="2",
24007 next-row="0x00001512",prev-row="0x0000150e",
24008 next-page="0x00001512",prev-page="0x0000150e",memory=[
24009 @{addr="0x00001510",data=["128"]@}]
24013 Read thirty two bytes of memory starting at @code{bytes+16} and format
24014 as eight rows of four columns. Include a string encoding with @samp{x}
24015 used as the non-printable character.
24019 4-data-read-memory bytes+16 x 1 8 4 x
24020 4^done,addr="0x000013a0",nr-bytes="32",total-bytes="32",
24021 next-row="0x000013c0",prev-row="0x0000139c",
24022 next-page="0x000013c0",prev-page="0x00001380",memory=[
24023 @{addr="0x000013a0",data=["0x10","0x11","0x12","0x13"],ascii="xxxx"@},
24024 @{addr="0x000013a4",data=["0x14","0x15","0x16","0x17"],ascii="xxxx"@},
24025 @{addr="0x000013a8",data=["0x18","0x19","0x1a","0x1b"],ascii="xxxx"@},
24026 @{addr="0x000013ac",data=["0x1c","0x1d","0x1e","0x1f"],ascii="xxxx"@},
24027 @{addr="0x000013b0",data=["0x20","0x21","0x22","0x23"],ascii=" !\"#"@},
24028 @{addr="0x000013b4",data=["0x24","0x25","0x26","0x27"],ascii="$%&'"@},
24029 @{addr="0x000013b8",data=["0x28","0x29","0x2a","0x2b"],ascii="()*+"@},
24030 @{addr="0x000013bc",data=["0x2c","0x2d","0x2e","0x2f"],ascii=",-./"@}]
24034 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
24035 @node GDB/MI Tracepoint Commands
24036 @section @sc{gdb/mi} Tracepoint Commands
24038 The tracepoint commands are not yet implemented.
24040 @c @subheading -trace-actions
24042 @c @subheading -trace-delete
24044 @c @subheading -trace-disable
24046 @c @subheading -trace-dump
24048 @c @subheading -trace-enable
24050 @c @subheading -trace-exists
24052 @c @subheading -trace-find
24054 @c @subheading -trace-frame-number
24056 @c @subheading -trace-info
24058 @c @subheading -trace-insert
24060 @c @subheading -trace-list
24062 @c @subheading -trace-pass-count
24064 @c @subheading -trace-save
24066 @c @subheading -trace-start
24068 @c @subheading -trace-stop
24071 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
24072 @node GDB/MI Symbol Query
24073 @section @sc{gdb/mi} Symbol Query Commands
24076 @subheading The @code{-symbol-info-address} Command
24077 @findex -symbol-info-address
24079 @subsubheading Synopsis
24082 -symbol-info-address @var{symbol}
24085 Describe where @var{symbol} is stored.
24087 @subsubheading @value{GDBN} Command
24089 The corresponding @value{GDBN} command is @samp{info address}.
24091 @subsubheading Example
24095 @subheading The @code{-symbol-info-file} Command
24096 @findex -symbol-info-file
24098 @subsubheading Synopsis
24104 Show the file for the symbol.
24106 @subsubheading @value{GDBN} Command
24108 There's no equivalent @value{GDBN} command. @code{gdbtk} has
24109 @samp{gdb_find_file}.
24111 @subsubheading Example
24115 @subheading The @code{-symbol-info-function} Command
24116 @findex -symbol-info-function
24118 @subsubheading Synopsis
24121 -symbol-info-function
24124 Show which function the symbol lives in.
24126 @subsubheading @value{GDBN} Command
24128 @samp{gdb_get_function} in @code{gdbtk}.
24130 @subsubheading Example
24134 @subheading The @code{-symbol-info-line} Command
24135 @findex -symbol-info-line
24137 @subsubheading Synopsis
24143 Show the core addresses of the code for a source line.
24145 @subsubheading @value{GDBN} Command
24147 The corresponding @value{GDBN} command is @samp{info line}.
24148 @code{gdbtk} has the @samp{gdb_get_line} and @samp{gdb_get_file} commands.
24150 @subsubheading Example
24154 @subheading The @code{-symbol-info-symbol} Command
24155 @findex -symbol-info-symbol
24157 @subsubheading Synopsis
24160 -symbol-info-symbol @var{addr}
24163 Describe what symbol is at location @var{addr}.
24165 @subsubheading @value{GDBN} Command
24167 The corresponding @value{GDBN} command is @samp{info symbol}.
24169 @subsubheading Example
24173 @subheading The @code{-symbol-list-functions} Command
24174 @findex -symbol-list-functions
24176 @subsubheading Synopsis
24179 -symbol-list-functions
24182 List the functions in the executable.
24184 @subsubheading @value{GDBN} Command
24186 @samp{info functions} in @value{GDBN}, @samp{gdb_listfunc} and
24187 @samp{gdb_search} in @code{gdbtk}.
24189 @subsubheading Example
24193 @subheading The @code{-symbol-list-lines} Command
24194 @findex -symbol-list-lines
24196 @subsubheading Synopsis
24199 -symbol-list-lines @var{filename}
24202 Print the list of lines that contain code and their associated program
24203 addresses for the given source filename. The entries are sorted in
24204 ascending PC order.
24206 @subsubheading @value{GDBN} Command
24208 There is no corresponding @value{GDBN} command.
24210 @subsubheading Example
24213 -symbol-list-lines basics.c
24214 ^done,lines=[@{pc="0x08048554",line="7"@},@{pc="0x0804855a",line="8"@}]
24219 @subheading The @code{-symbol-list-types} Command
24220 @findex -symbol-list-types
24222 @subsubheading Synopsis
24228 List all the type names.
24230 @subsubheading @value{GDBN} Command
24232 The corresponding commands are @samp{info types} in @value{GDBN},
24233 @samp{gdb_search} in @code{gdbtk}.
24235 @subsubheading Example
24239 @subheading The @code{-symbol-list-variables} Command
24240 @findex -symbol-list-variables
24242 @subsubheading Synopsis
24245 -symbol-list-variables
24248 List all the global and static variable names.
24250 @subsubheading @value{GDBN} Command
24252 @samp{info variables} in @value{GDBN}, @samp{gdb_search} in @code{gdbtk}.
24254 @subsubheading Example
24258 @subheading The @code{-symbol-locate} Command
24259 @findex -symbol-locate
24261 @subsubheading Synopsis
24267 @subsubheading @value{GDBN} Command
24269 @samp{gdb_loc} in @code{gdbtk}.
24271 @subsubheading Example
24275 @subheading The @code{-symbol-type} Command
24276 @findex -symbol-type
24278 @subsubheading Synopsis
24281 -symbol-type @var{variable}
24284 Show type of @var{variable}.
24286 @subsubheading @value{GDBN} Command
24288 The corresponding @value{GDBN} command is @samp{ptype}, @code{gdbtk} has
24289 @samp{gdb_obj_variable}.
24291 @subsubheading Example
24295 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
24296 @node GDB/MI File Commands
24297 @section @sc{gdb/mi} File Commands
24299 This section describes the GDB/MI commands to specify executable file names
24300 and to read in and obtain symbol table information.
24302 @subheading The @code{-file-exec-and-symbols} Command
24303 @findex -file-exec-and-symbols
24305 @subsubheading Synopsis
24308 -file-exec-and-symbols @var{file}
24311 Specify the executable file to be debugged. This file is the one from
24312 which the symbol table is also read. If no file is specified, the
24313 command clears the executable and symbol information. If breakpoints
24314 are set when using this command with no arguments, @value{GDBN} will produce
24315 error messages. Otherwise, no output is produced, except a completion
24318 @subsubheading @value{GDBN} Command
24320 The corresponding @value{GDBN} command is @samp{file}.
24322 @subsubheading Example
24326 -file-exec-and-symbols /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
24332 @subheading The @code{-file-exec-file} Command
24333 @findex -file-exec-file
24335 @subsubheading Synopsis
24338 -file-exec-file @var{file}
24341 Specify the executable file to be debugged. Unlike
24342 @samp{-file-exec-and-symbols}, the symbol table is @emph{not} read
24343 from this file. If used without argument, @value{GDBN} clears the information
24344 about the executable file. No output is produced, except a completion
24347 @subsubheading @value{GDBN} Command
24349 The corresponding @value{GDBN} command is @samp{exec-file}.
24351 @subsubheading Example
24355 -file-exec-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
24361 @subheading The @code{-file-list-exec-sections} Command
24362 @findex -file-list-exec-sections
24364 @subsubheading Synopsis
24367 -file-list-exec-sections
24370 List the sections of the current executable file.
24372 @subsubheading @value{GDBN} Command
24374 The @value{GDBN} command @samp{info file} shows, among the rest, the same
24375 information as this command. @code{gdbtk} has a corresponding command
24376 @samp{gdb_load_info}.
24378 @subsubheading Example
24382 @subheading The @code{-file-list-exec-source-file} Command
24383 @findex -file-list-exec-source-file
24385 @subsubheading Synopsis
24388 -file-list-exec-source-file
24391 List the line number, the current source file, and the absolute path
24392 to the current source file for the current executable. The macro
24393 information field has a value of @samp{1} or @samp{0} depending on
24394 whether or not the file includes preprocessor macro information.
24396 @subsubheading @value{GDBN} Command
24398 The @value{GDBN} equivalent is @samp{info source}
24400 @subsubheading Example
24404 123-file-list-exec-source-file
24405 123^done,line="1",file="foo.c",fullname="/home/bar/foo.c,macro-info="1"
24410 @subheading The @code{-file-list-exec-source-files} Command
24411 @findex -file-list-exec-source-files
24413 @subsubheading Synopsis
24416 -file-list-exec-source-files
24419 List the source files for the current executable.
24421 It will always output the filename, but only when @value{GDBN} can find
24422 the absolute file name of a source file, will it output the fullname.
24424 @subsubheading @value{GDBN} Command
24426 The @value{GDBN} equivalent is @samp{info sources}.
24427 @code{gdbtk} has an analogous command @samp{gdb_listfiles}.
24429 @subsubheading Example
24432 -file-list-exec-source-files
24434 @{file=foo.c,fullname=/home/foo.c@},
24435 @{file=/home/bar.c,fullname=/home/bar.c@},
24436 @{file=gdb_could_not_find_fullpath.c@}]
24440 @subheading The @code{-file-list-shared-libraries} Command
24441 @findex -file-list-shared-libraries
24443 @subsubheading Synopsis
24446 -file-list-shared-libraries
24449 List the shared libraries in the program.
24451 @subsubheading @value{GDBN} Command
24453 The corresponding @value{GDBN} command is @samp{info shared}.
24455 @subsubheading Example
24459 @subheading The @code{-file-list-symbol-files} Command
24460 @findex -file-list-symbol-files
24462 @subsubheading Synopsis
24465 -file-list-symbol-files
24470 @subsubheading @value{GDBN} Command
24472 The corresponding @value{GDBN} command is @samp{info file} (part of it).
24474 @subsubheading Example
24478 @subheading The @code{-file-symbol-file} Command
24479 @findex -file-symbol-file
24481 @subsubheading Synopsis
24484 -file-symbol-file @var{file}
24487 Read symbol table info from the specified @var{file} argument. When
24488 used without arguments, clears @value{GDBN}'s symbol table info. No output is
24489 produced, except for a completion notification.
24491 @subsubheading @value{GDBN} Command
24493 The corresponding @value{GDBN} command is @samp{symbol-file}.
24495 @subsubheading Example
24499 -file-symbol-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
24505 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
24506 @node GDB/MI Memory Overlay Commands
24507 @section @sc{gdb/mi} Memory Overlay Commands
24509 The memory overlay commands are not implemented.
24511 @c @subheading -overlay-auto
24513 @c @subheading -overlay-list-mapping-state
24515 @c @subheading -overlay-list-overlays
24517 @c @subheading -overlay-map
24519 @c @subheading -overlay-off
24521 @c @subheading -overlay-on
24523 @c @subheading -overlay-unmap
24525 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
24526 @node GDB/MI Signal Handling Commands
24527 @section @sc{gdb/mi} Signal Handling Commands
24529 Signal handling commands are not implemented.
24531 @c @subheading -signal-handle
24533 @c @subheading -signal-list-handle-actions
24535 @c @subheading -signal-list-signal-types
24539 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
24540 @node GDB/MI Target Manipulation
24541 @section @sc{gdb/mi} Target Manipulation Commands
24544 @subheading The @code{-target-attach} Command
24545 @findex -target-attach
24547 @subsubheading Synopsis
24550 -target-attach @var{pid} | @var{gid} | @var{file}
24553 Attach to a process @var{pid} or a file @var{file} outside of
24554 @value{GDBN}, or a thread group @var{gid}. If attaching to a thread
24555 group, the id previously returned by
24556 @samp{-list-thread-groups --available} must be used.
24558 @subsubheading @value{GDBN} Command
24560 The corresponding @value{GDBN} command is @samp{attach}.
24562 @subsubheading Example
24566 =thread-created,id="1"
24567 *stopped,thread-id="1",frame=@{addr="0xb7f7e410",func="bar",args=[]@}
24572 @subheading The @code{-target-compare-sections} Command
24573 @findex -target-compare-sections
24575 @subsubheading Synopsis
24578 -target-compare-sections [ @var{section} ]
24581 Compare data of section @var{section} on target to the exec file.
24582 Without the argument, all sections are compared.
24584 @subsubheading @value{GDBN} Command
24586 The @value{GDBN} equivalent is @samp{compare-sections}.
24588 @subsubheading Example
24592 @subheading The @code{-target-detach} Command
24593 @findex -target-detach
24595 @subsubheading Synopsis
24598 -target-detach [ @var{pid} | @var{gid} ]
24601 Detach from the remote target which normally resumes its execution.
24602 If either @var{pid} or @var{gid} is specified, detaches from either
24603 the specified process, or specified thread group. There's no output.
24605 @subsubheading @value{GDBN} Command
24607 The corresponding @value{GDBN} command is @samp{detach}.
24609 @subsubheading Example
24619 @subheading The @code{-target-disconnect} Command
24620 @findex -target-disconnect
24622 @subsubheading Synopsis
24628 Disconnect from the remote target. There's no output and the target is
24629 generally not resumed.
24631 @subsubheading @value{GDBN} Command
24633 The corresponding @value{GDBN} command is @samp{disconnect}.
24635 @subsubheading Example
24645 @subheading The @code{-target-download} Command
24646 @findex -target-download
24648 @subsubheading Synopsis
24654 Loads the executable onto the remote target.
24655 It prints out an update message every half second, which includes the fields:
24659 The name of the section.
24661 The size of what has been sent so far for that section.
24663 The size of the section.
24665 The total size of what was sent so far (the current and the previous sections).
24667 The size of the overall executable to download.
24671 Each message is sent as status record (@pxref{GDB/MI Output Syntax, ,
24672 @sc{gdb/mi} Output Syntax}).
24674 In addition, it prints the name and size of the sections, as they are
24675 downloaded. These messages include the following fields:
24679 The name of the section.
24681 The size of the section.
24683 The size of the overall executable to download.
24687 At the end, a summary is printed.
24689 @subsubheading @value{GDBN} Command
24691 The corresponding @value{GDBN} command is @samp{load}.
24693 @subsubheading Example
24695 Note: each status message appears on a single line. Here the messages
24696 have been broken down so that they can fit onto a page.
24701 +download,@{section=".text",section-size="6668",total-size="9880"@}
24702 +download,@{section=".text",section-sent="512",section-size="6668",
24703 total-sent="512",total-size="9880"@}
24704 +download,@{section=".text",section-sent="1024",section-size="6668",
24705 total-sent="1024",total-size="9880"@}
24706 +download,@{section=".text",section-sent="1536",section-size="6668",
24707 total-sent="1536",total-size="9880"@}
24708 +download,@{section=".text",section-sent="2048",section-size="6668",
24709 total-sent="2048",total-size="9880"@}
24710 +download,@{section=".text",section-sent="2560",section-size="6668",
24711 total-sent="2560",total-size="9880"@}
24712 +download,@{section=".text",section-sent="3072",section-size="6668",
24713 total-sent="3072",total-size="9880"@}
24714 +download,@{section=".text",section-sent="3584",section-size="6668",
24715 total-sent="3584",total-size="9880"@}
24716 +download,@{section=".text",section-sent="4096",section-size="6668",
24717 total-sent="4096",total-size="9880"@}
24718 +download,@{section=".text",section-sent="4608",section-size="6668",
24719 total-sent="4608",total-size="9880"@}
24720 +download,@{section=".text",section-sent="5120",section-size="6668",
24721 total-sent="5120",total-size="9880"@}
24722 +download,@{section=".text",section-sent="5632",section-size="6668",
24723 total-sent="5632",total-size="9880"@}
24724 +download,@{section=".text",section-sent="6144",section-size="6668",
24725 total-sent="6144",total-size="9880"@}
24726 +download,@{section=".text",section-sent="6656",section-size="6668",
24727 total-sent="6656",total-size="9880"@}
24728 +download,@{section=".init",section-size="28",total-size="9880"@}
24729 +download,@{section=".fini",section-size="28",total-size="9880"@}
24730 +download,@{section=".data",section-size="3156",total-size="9880"@}
24731 +download,@{section=".data",section-sent="512",section-size="3156",
24732 total-sent="7236",total-size="9880"@}
24733 +download,@{section=".data",section-sent="1024",section-size="3156",
24734 total-sent="7748",total-size="9880"@}
24735 +download,@{section=".data",section-sent="1536",section-size="3156",
24736 total-sent="8260",total-size="9880"@}
24737 +download,@{section=".data",section-sent="2048",section-size="3156",
24738 total-sent="8772",total-size="9880"@}
24739 +download,@{section=".data",section-sent="2560",section-size="3156",
24740 total-sent="9284",total-size="9880"@}
24741 +download,@{section=".data",section-sent="3072",section-size="3156",
24742 total-sent="9796",total-size="9880"@}
24743 ^done,address="0x10004",load-size="9880",transfer-rate="6586",
24749 @subheading The @code{-target-exec-status} Command
24750 @findex -target-exec-status
24752 @subsubheading Synopsis
24755 -target-exec-status
24758 Provide information on the state of the target (whether it is running or
24759 not, for instance).
24761 @subsubheading @value{GDBN} Command
24763 There's no equivalent @value{GDBN} command.
24765 @subsubheading Example
24769 @subheading The @code{-target-list-available-targets} Command
24770 @findex -target-list-available-targets
24772 @subsubheading Synopsis
24775 -target-list-available-targets
24778 List the possible targets to connect to.
24780 @subsubheading @value{GDBN} Command
24782 The corresponding @value{GDBN} command is @samp{help target}.
24784 @subsubheading Example
24788 @subheading The @code{-target-list-current-targets} Command
24789 @findex -target-list-current-targets
24791 @subsubheading Synopsis
24794 -target-list-current-targets
24797 Describe the current target.
24799 @subsubheading @value{GDBN} Command
24801 The corresponding information is printed by @samp{info file} (among
24804 @subsubheading Example
24808 @subheading The @code{-target-list-parameters} Command
24809 @findex -target-list-parameters
24811 @subsubheading Synopsis
24814 -target-list-parameters
24819 @subsubheading @value{GDBN} Command
24823 @subsubheading Example
24827 @subheading The @code{-target-select} Command
24828 @findex -target-select
24830 @subsubheading Synopsis
24833 -target-select @var{type} @var{parameters @dots{}}
24836 Connect @value{GDBN} to the remote target. This command takes two args:
24840 The type of target, for instance @samp{remote}, etc.
24841 @item @var{parameters}
24842 Device names, host names and the like. @xref{Target Commands, ,
24843 Commands for Managing Targets}, for more details.
24846 The output is a connection notification, followed by the address at
24847 which the target program is, in the following form:
24850 ^connected,addr="@var{address}",func="@var{function name}",
24851 args=[@var{arg list}]
24854 @subsubheading @value{GDBN} Command
24856 The corresponding @value{GDBN} command is @samp{target}.
24858 @subsubheading Example
24862 -target-select remote /dev/ttya
24863 ^connected,addr="0xfe00a300",func="??",args=[]
24867 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
24868 @node GDB/MI File Transfer Commands
24869 @section @sc{gdb/mi} File Transfer Commands
24872 @subheading The @code{-target-file-put} Command
24873 @findex -target-file-put
24875 @subsubheading Synopsis
24878 -target-file-put @var{hostfile} @var{targetfile}
24881 Copy file @var{hostfile} from the host system (the machine running
24882 @value{GDBN}) to @var{targetfile} on the target system.
24884 @subsubheading @value{GDBN} Command
24886 The corresponding @value{GDBN} command is @samp{remote put}.
24888 @subsubheading Example
24892 -target-file-put localfile remotefile
24898 @subheading The @code{-target-file-get} Command
24899 @findex -target-file-get
24901 @subsubheading Synopsis
24904 -target-file-get @var{targetfile} @var{hostfile}
24907 Copy file @var{targetfile} from the target system to @var{hostfile}
24908 on the host system.
24910 @subsubheading @value{GDBN} Command
24912 The corresponding @value{GDBN} command is @samp{remote get}.
24914 @subsubheading Example
24918 -target-file-get remotefile localfile
24924 @subheading The @code{-target-file-delete} Command
24925 @findex -target-file-delete
24927 @subsubheading Synopsis
24930 -target-file-delete @var{targetfile}
24933 Delete @var{targetfile} from the target system.
24935 @subsubheading @value{GDBN} Command
24937 The corresponding @value{GDBN} command is @samp{remote delete}.
24939 @subsubheading Example
24943 -target-file-delete remotefile
24949 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
24950 @node GDB/MI Miscellaneous Commands
24951 @section Miscellaneous @sc{gdb/mi} Commands
24953 @c @subheading -gdb-complete
24955 @subheading The @code{-gdb-exit} Command
24958 @subsubheading Synopsis
24964 Exit @value{GDBN} immediately.
24966 @subsubheading @value{GDBN} Command
24968 Approximately corresponds to @samp{quit}.
24970 @subsubheading Example
24979 @subheading The @code{-exec-abort} Command
24980 @findex -exec-abort
24982 @subsubheading Synopsis
24988 Kill the inferior running program.
24990 @subsubheading @value{GDBN} Command
24992 The corresponding @value{GDBN} command is @samp{kill}.
24994 @subsubheading Example
24998 @subheading The @code{-gdb-set} Command
25001 @subsubheading Synopsis
25007 Set an internal @value{GDBN} variable.
25008 @c IS THIS A DOLLAR VARIABLE? OR SOMETHING LIKE ANNOTATE ?????
25010 @subsubheading @value{GDBN} Command
25012 The corresponding @value{GDBN} command is @samp{set}.
25014 @subsubheading Example
25024 @subheading The @code{-gdb-show} Command
25027 @subsubheading Synopsis
25033 Show the current value of a @value{GDBN} variable.
25035 @subsubheading @value{GDBN} Command
25037 The corresponding @value{GDBN} command is @samp{show}.
25039 @subsubheading Example
25048 @c @subheading -gdb-source
25051 @subheading The @code{-gdb-version} Command
25052 @findex -gdb-version
25054 @subsubheading Synopsis
25060 Show version information for @value{GDBN}. Used mostly in testing.
25062 @subsubheading @value{GDBN} Command
25064 The @value{GDBN} equivalent is @samp{show version}. @value{GDBN} by
25065 default shows this information when you start an interactive session.
25067 @subsubheading Example
25069 @c This example modifies the actual output from GDB to avoid overfull
25075 ~Copyright 2000 Free Software Foundation, Inc.
25076 ~GDB is free software, covered by the GNU General Public License, and
25077 ~you are welcome to change it and/or distribute copies of it under
25078 ~ certain conditions.
25079 ~Type "show copying" to see the conditions.
25080 ~There is absolutely no warranty for GDB. Type "show warranty" for
25082 ~This GDB was configured as
25083 "--host=sparc-sun-solaris2.5.1 --target=ppc-eabi".
25088 @subheading The @code{-list-features} Command
25089 @findex -list-features
25091 Returns a list of particular features of the MI protocol that
25092 this version of gdb implements. A feature can be a command,
25093 or a new field in an output of some command, or even an
25094 important bugfix. While a frontend can sometimes detect presence
25095 of a feature at runtime, it is easier to perform detection at debugger
25098 The command returns a list of strings, with each string naming an
25099 available feature. Each returned string is just a name, it does not
25100 have any internal structure. The list of possible feature names
25106 (gdb) -list-features
25107 ^done,result=["feature1","feature2"]
25110 The current list of features is:
25113 @item frozen-varobjs
25114 Indicates presence of the @code{-var-set-frozen} command, as well
25115 as possible presense of the @code{frozen} field in the output
25116 of @code{-varobj-create}.
25117 @item pending-breakpoints
25118 Indicates presence of the @option{-f} option to the @code{-break-insert} command.
25120 Indicates presence of Python scripting support, Python-based
25121 pretty-printing commands, and possible presence of the
25122 @samp{display_hint} field in the output of @code{-var-list-children}
25124 Indicates presence of the @code{-thread-info} command.
25128 @subheading The @code{-list-target-features} Command
25129 @findex -list-target-features
25131 Returns a list of particular features that are supported by the
25132 target. Those features affect the permitted MI commands, but
25133 unlike the features reported by the @code{-list-features} command, the
25134 features depend on which target GDB is using at the moment. Whenever
25135 a target can change, due to commands such as @code{-target-select},
25136 @code{-target-attach} or @code{-exec-run}, the list of target features
25137 may change, and the frontend should obtain it again.
25141 (gdb) -list-features
25142 ^done,result=["async"]
25145 The current list of features is:
25149 Indicates that the target is capable of asynchronous command
25150 execution, which means that @value{GDBN} will accept further commands
25151 while the target is running.
25155 @subheading The @code{-list-thread-groups} Command
25156 @findex -list-thread-groups
25158 @subheading Synopsis
25161 -list-thread-groups [ --available ] [ @var{group} ]
25164 When used without the @var{group} parameter, lists top-level thread
25165 groups that are being debugged. When used with the @var{group}
25166 parameter, the children of the specified group are listed. The
25167 children can be either threads, or other groups. At present,
25168 @value{GDBN} will not report both threads and groups as children at
25169 the same time, but it may change in future.
25171 With the @samp{--available} option, instead of reporting groups that
25172 are been debugged, GDB will report all thread groups available on the
25173 target. Using the @samp{--available} option together with @var{group}
25176 @subheading Example
25180 -list-thread-groups
25181 ^done,groups=[@{id="17",type="process",pid="yyy",num_children="2"@}]
25182 -list-thread-groups 17
25183 ^done,threads=[@{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
25184 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",args=[]@},state="running"@},
25185 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
25186 frame=@{level="0",addr="0x0804891f",func="foo",args=[@{name="i",value="10"@}],
25187 file="/tmp/a.c",fullname="/tmp/a.c",line="158"@},state="running"@}]]
25190 @subheading The @code{-interpreter-exec} Command
25191 @findex -interpreter-exec
25193 @subheading Synopsis
25196 -interpreter-exec @var{interpreter} @var{command}
25198 @anchor{-interpreter-exec}
25200 Execute the specified @var{command} in the given @var{interpreter}.
25202 @subheading @value{GDBN} Command
25204 The corresponding @value{GDBN} command is @samp{interpreter-exec}.
25206 @subheading Example
25210 -interpreter-exec console "break main"
25211 &"During symbol reading, couldn't parse type; debugger out of date?.\n"
25212 &"During symbol reading, bad structure-type format.\n"
25213 ~"Breakpoint 1 at 0x8074fc6: file ../../src/gdb/main.c, line 743.\n"
25218 @subheading The @code{-inferior-tty-set} Command
25219 @findex -inferior-tty-set
25221 @subheading Synopsis
25224 -inferior-tty-set /dev/pts/1
25227 Set terminal for future runs of the program being debugged.
25229 @subheading @value{GDBN} Command
25231 The corresponding @value{GDBN} command is @samp{set inferior-tty} /dev/pts/1.
25233 @subheading Example
25237 -inferior-tty-set /dev/pts/1
25242 @subheading The @code{-inferior-tty-show} Command
25243 @findex -inferior-tty-show
25245 @subheading Synopsis
25251 Show terminal for future runs of program being debugged.
25253 @subheading @value{GDBN} Command
25255 The corresponding @value{GDBN} command is @samp{show inferior-tty}.
25257 @subheading Example
25261 -inferior-tty-set /dev/pts/1
25265 ^done,inferior_tty_terminal="/dev/pts/1"
25269 @subheading The @code{-enable-timings} Command
25270 @findex -enable-timings
25272 @subheading Synopsis
25275 -enable-timings [yes | no]
25278 Toggle the printing of the wallclock, user and system times for an MI
25279 command as a field in its output. This command is to help frontend
25280 developers optimize the performance of their code. No argument is
25281 equivalent to @samp{yes}.
25283 @subheading @value{GDBN} Command
25287 @subheading Example
25295 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
25296 addr="0x080484ed",func="main",file="myprog.c",
25297 fullname="/home/nickrob/myprog.c",line="73",times="0"@},
25298 time=@{wallclock="0.05185",user="0.00800",system="0.00000"@}
25306 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
25307 frame=@{addr="0x080484ed",func="main",args=[@{name="argc",value="1"@},
25308 @{name="argv",value="0xbfb60364"@}],file="myprog.c",
25309 fullname="/home/nickrob/myprog.c",line="73"@}
25314 @chapter @value{GDBN} Annotations
25316 This chapter describes annotations in @value{GDBN}. Annotations were
25317 designed to interface @value{GDBN} to graphical user interfaces or other
25318 similar programs which want to interact with @value{GDBN} at a
25319 relatively high level.
25321 The annotation mechanism has largely been superseded by @sc{gdb/mi}
25325 This is Edition @value{EDITION}, @value{DATE}.
25329 * Annotations Overview:: What annotations are; the general syntax.
25330 * Server Prefix:: Issuing a command without affecting user state.
25331 * Prompting:: Annotations marking @value{GDBN}'s need for input.
25332 * Errors:: Annotations for error messages.
25333 * Invalidation:: Some annotations describe things now invalid.
25334 * Annotations for Running::
25335 Whether the program is running, how it stopped, etc.
25336 * Source Annotations:: Annotations describing source code.
25339 @node Annotations Overview
25340 @section What is an Annotation?
25341 @cindex annotations
25343 Annotations start with a newline character, two @samp{control-z}
25344 characters, and the name of the annotation. If there is no additional
25345 information associated with this annotation, the name of the annotation
25346 is followed immediately by a newline. If there is additional
25347 information, the name of the annotation is followed by a space, the
25348 additional information, and a newline. The additional information
25349 cannot contain newline characters.
25351 Any output not beginning with a newline and two @samp{control-z}
25352 characters denotes literal output from @value{GDBN}. Currently there is
25353 no need for @value{GDBN} to output a newline followed by two
25354 @samp{control-z} characters, but if there was such a need, the
25355 annotations could be extended with an @samp{escape} annotation which
25356 means those three characters as output.
25358 The annotation @var{level}, which is specified using the
25359 @option{--annotate} command line option (@pxref{Mode Options}), controls
25360 how much information @value{GDBN} prints together with its prompt,
25361 values of expressions, source lines, and other types of output. Level 0
25362 is for no annotations, level 1 is for use when @value{GDBN} is run as a
25363 subprocess of @sc{gnu} Emacs, level 3 is the maximum annotation suitable
25364 for programs that control @value{GDBN}, and level 2 annotations have
25365 been made obsolete (@pxref{Limitations, , Limitations of the Annotation
25366 Interface, annotate, GDB's Obsolete Annotations}).
25369 @kindex set annotate
25370 @item set annotate @var{level}
25371 The @value{GDBN} command @code{set annotate} sets the level of
25372 annotations to the specified @var{level}.
25374 @item show annotate
25375 @kindex show annotate
25376 Show the current annotation level.
25379 This chapter describes level 3 annotations.
25381 A simple example of starting up @value{GDBN} with annotations is:
25384 $ @kbd{gdb --annotate=3}
25386 Copyright 2003 Free Software Foundation, Inc.
25387 GDB is free software, covered by the GNU General Public License,
25388 and you are welcome to change it and/or distribute copies of it
25389 under certain conditions.
25390 Type "show copying" to see the conditions.
25391 There is absolutely no warranty for GDB. Type "show warranty"
25393 This GDB was configured as "i386-pc-linux-gnu"
25404 Here @samp{quit} is input to @value{GDBN}; the rest is output from
25405 @value{GDBN}. The three lines beginning @samp{^Z^Z} (where @samp{^Z}
25406 denotes a @samp{control-z} character) are annotations; the rest is
25407 output from @value{GDBN}.
25409 @node Server Prefix
25410 @section The Server Prefix
25411 @cindex server prefix
25413 If you prefix a command with @samp{server } then it will not affect
25414 the command history, nor will it affect @value{GDBN}'s notion of which
25415 command to repeat if @key{RET} is pressed on a line by itself. This
25416 means that commands can be run behind a user's back by a front-end in
25417 a transparent manner.
25419 The server prefix does not affect the recording of values into the value
25420 history; to print a value without recording it into the value history,
25421 use the @code{output} command instead of the @code{print} command.
25424 @section Annotation for @value{GDBN} Input
25426 @cindex annotations for prompts
25427 When @value{GDBN} prompts for input, it annotates this fact so it is possible
25428 to know when to send output, when the output from a given command is
25431 Different kinds of input each have a different @dfn{input type}. Each
25432 input type has three annotations: a @code{pre-} annotation, which
25433 denotes the beginning of any prompt which is being output, a plain
25434 annotation, which denotes the end of the prompt, and then a @code{post-}
25435 annotation which denotes the end of any echo which may (or may not) be
25436 associated with the input. For example, the @code{prompt} input type
25437 features the following annotations:
25445 The input types are
25448 @findex pre-prompt annotation
25449 @findex prompt annotation
25450 @findex post-prompt annotation
25452 When @value{GDBN} is prompting for a command (the main @value{GDBN} prompt).
25454 @findex pre-commands annotation
25455 @findex commands annotation
25456 @findex post-commands annotation
25458 When @value{GDBN} prompts for a set of commands, like in the @code{commands}
25459 command. The annotations are repeated for each command which is input.
25461 @findex pre-overload-choice annotation
25462 @findex overload-choice annotation
25463 @findex post-overload-choice annotation
25464 @item overload-choice
25465 When @value{GDBN} wants the user to select between various overloaded functions.
25467 @findex pre-query annotation
25468 @findex query annotation
25469 @findex post-query annotation
25471 When @value{GDBN} wants the user to confirm a potentially dangerous operation.
25473 @findex pre-prompt-for-continue annotation
25474 @findex prompt-for-continue annotation
25475 @findex post-prompt-for-continue annotation
25476 @item prompt-for-continue
25477 When @value{GDBN} is asking the user to press return to continue. Note: Don't
25478 expect this to work well; instead use @code{set height 0} to disable
25479 prompting. This is because the counting of lines is buggy in the
25480 presence of annotations.
25485 @cindex annotations for errors, warnings and interrupts
25487 @findex quit annotation
25492 This annotation occurs right before @value{GDBN} responds to an interrupt.
25494 @findex error annotation
25499 This annotation occurs right before @value{GDBN} responds to an error.
25501 Quit and error annotations indicate that any annotations which @value{GDBN} was
25502 in the middle of may end abruptly. For example, if a
25503 @code{value-history-begin} annotation is followed by a @code{error}, one
25504 cannot expect to receive the matching @code{value-history-end}. One
25505 cannot expect not to receive it either, however; an error annotation
25506 does not necessarily mean that @value{GDBN} is immediately returning all the way
25509 @findex error-begin annotation
25510 A quit or error annotation may be preceded by
25516 Any output between that and the quit or error annotation is the error
25519 Warning messages are not yet annotated.
25520 @c If we want to change that, need to fix warning(), type_error(),
25521 @c range_error(), and possibly other places.
25524 @section Invalidation Notices
25526 @cindex annotations for invalidation messages
25527 The following annotations say that certain pieces of state may have
25531 @findex frames-invalid annotation
25532 @item ^Z^Zframes-invalid
25534 The frames (for example, output from the @code{backtrace} command) may
25537 @findex breakpoints-invalid annotation
25538 @item ^Z^Zbreakpoints-invalid
25540 The breakpoints may have changed. For example, the user just added or
25541 deleted a breakpoint.
25544 @node Annotations for Running
25545 @section Running the Program
25546 @cindex annotations for running programs
25548 @findex starting annotation
25549 @findex stopping annotation
25550 When the program starts executing due to a @value{GDBN} command such as
25551 @code{step} or @code{continue},
25557 is output. When the program stops,
25563 is output. Before the @code{stopped} annotation, a variety of
25564 annotations describe how the program stopped.
25567 @findex exited annotation
25568 @item ^Z^Zexited @var{exit-status}
25569 The program exited, and @var{exit-status} is the exit status (zero for
25570 successful exit, otherwise nonzero).
25572 @findex signalled annotation
25573 @findex signal-name annotation
25574 @findex signal-name-end annotation
25575 @findex signal-string annotation
25576 @findex signal-string-end annotation
25577 @item ^Z^Zsignalled
25578 The program exited with a signal. After the @code{^Z^Zsignalled}, the
25579 annotation continues:
25585 ^Z^Zsignal-name-end
25589 ^Z^Zsignal-string-end
25594 where @var{name} is the name of the signal, such as @code{SIGILL} or
25595 @code{SIGSEGV}, and @var{string} is the explanation of the signal, such
25596 as @code{Illegal Instruction} or @code{Segmentation fault}.
25597 @var{intro-text}, @var{middle-text}, and @var{end-text} are for the
25598 user's benefit and have no particular format.
25600 @findex signal annotation
25602 The syntax of this annotation is just like @code{signalled}, but @value{GDBN} is
25603 just saying that the program received the signal, not that it was
25604 terminated with it.
25606 @findex breakpoint annotation
25607 @item ^Z^Zbreakpoint @var{number}
25608 The program hit breakpoint number @var{number}.
25610 @findex watchpoint annotation
25611 @item ^Z^Zwatchpoint @var{number}
25612 The program hit watchpoint number @var{number}.
25615 @node Source Annotations
25616 @section Displaying Source
25617 @cindex annotations for source display
25619 @findex source annotation
25620 The following annotation is used instead of displaying source code:
25623 ^Z^Zsource @var{filename}:@var{line}:@var{character}:@var{middle}:@var{addr}
25626 where @var{filename} is an absolute file name indicating which source
25627 file, @var{line} is the line number within that file (where 1 is the
25628 first line in the file), @var{character} is the character position
25629 within the file (where 0 is the first character in the file) (for most
25630 debug formats this will necessarily point to the beginning of a line),
25631 @var{middle} is @samp{middle} if @var{addr} is in the middle of the
25632 line, or @samp{beg} if @var{addr} is at the beginning of the line, and
25633 @var{addr} is the address in the target program associated with the
25634 source which is being displayed. @var{addr} is in the form @samp{0x}
25635 followed by one or more lowercase hex digits (note that this does not
25636 depend on the language).
25639 @chapter Reporting Bugs in @value{GDBN}
25640 @cindex bugs in @value{GDBN}
25641 @cindex reporting bugs in @value{GDBN}
25643 Your bug reports play an essential role in making @value{GDBN} reliable.
25645 Reporting a bug may help you by bringing a solution to your problem, or it
25646 may not. But in any case the principal function of a bug report is to help
25647 the entire community by making the next version of @value{GDBN} work better. Bug
25648 reports are your contribution to the maintenance of @value{GDBN}.
25650 In order for a bug report to serve its purpose, you must include the
25651 information that enables us to fix the bug.
25654 * Bug Criteria:: Have you found a bug?
25655 * Bug Reporting:: How to report bugs
25659 @section Have You Found a Bug?
25660 @cindex bug criteria
25662 If you are not sure whether you have found a bug, here are some guidelines:
25665 @cindex fatal signal
25666 @cindex debugger crash
25667 @cindex crash of debugger
25669 If the debugger gets a fatal signal, for any input whatever, that is a
25670 @value{GDBN} bug. Reliable debuggers never crash.
25672 @cindex error on valid input
25674 If @value{GDBN} produces an error message for valid input, that is a
25675 bug. (Note that if you're cross debugging, the problem may also be
25676 somewhere in the connection to the target.)
25678 @cindex invalid input
25680 If @value{GDBN} does not produce an error message for invalid input,
25681 that is a bug. However, you should note that your idea of
25682 ``invalid input'' might be our idea of ``an extension'' or ``support
25683 for traditional practice''.
25686 If you are an experienced user of debugging tools, your suggestions
25687 for improvement of @value{GDBN} are welcome in any case.
25690 @node Bug Reporting
25691 @section How to Report Bugs
25692 @cindex bug reports
25693 @cindex @value{GDBN} bugs, reporting
25695 A number of companies and individuals offer support for @sc{gnu} products.
25696 If you obtained @value{GDBN} from a support organization, we recommend you
25697 contact that organization first.
25699 You can find contact information for many support companies and
25700 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
25702 @c should add a web page ref...
25705 @ifset BUGURL_DEFAULT
25706 In any event, we also recommend that you submit bug reports for
25707 @value{GDBN}. The preferred method is to submit them directly using
25708 @uref{http://www.gnu.org/software/gdb/bugs/, @value{GDBN}'s Bugs web
25709 page}. Alternatively, the @email{bug-gdb@@gnu.org, e-mail gateway} can
25712 @strong{Do not send bug reports to @samp{info-gdb}, or to
25713 @samp{help-gdb}, or to any newsgroups.} Most users of @value{GDBN} do
25714 not want to receive bug reports. Those that do have arranged to receive
25717 The mailing list @samp{bug-gdb} has a newsgroup @samp{gnu.gdb.bug} which
25718 serves as a repeater. The mailing list and the newsgroup carry exactly
25719 the same messages. Often people think of posting bug reports to the
25720 newsgroup instead of mailing them. This appears to work, but it has one
25721 problem which can be crucial: a newsgroup posting often lacks a mail
25722 path back to the sender. Thus, if we need to ask for more information,
25723 we may be unable to reach you. For this reason, it is better to send
25724 bug reports to the mailing list.
25726 @ifclear BUGURL_DEFAULT
25727 In any event, we also recommend that you submit bug reports for
25728 @value{GDBN} to @value{BUGURL}.
25732 The fundamental principle of reporting bugs usefully is this:
25733 @strong{report all the facts}. If you are not sure whether to state a
25734 fact or leave it out, state it!
25736 Often people omit facts because they think they know what causes the
25737 problem and assume that some details do not matter. Thus, you might
25738 assume that the name of the variable you use in an example does not matter.
25739 Well, probably it does not, but one cannot be sure. Perhaps the bug is a
25740 stray memory reference which happens to fetch from the location where that
25741 name is stored in memory; perhaps, if the name were different, the contents
25742 of that location would fool the debugger into doing the right thing despite
25743 the bug. Play it safe and give a specific, complete example. That is the
25744 easiest thing for you to do, and the most helpful.
25746 Keep in mind that the purpose of a bug report is to enable us to fix the
25747 bug. It may be that the bug has been reported previously, but neither
25748 you nor we can know that unless your bug report is complete and
25751 Sometimes people give a few sketchy facts and ask, ``Does this ring a
25752 bell?'' Those bug reports are useless, and we urge everyone to
25753 @emph{refuse to respond to them} except to chide the sender to report
25756 To enable us to fix the bug, you should include all these things:
25760 The version of @value{GDBN}. @value{GDBN} announces it if you start
25761 with no arguments; you can also print it at any time using @code{show
25764 Without this, we will not know whether there is any point in looking for
25765 the bug in the current version of @value{GDBN}.
25768 The type of machine you are using, and the operating system name and
25772 What compiler (and its version) was used to compile @value{GDBN}---e.g.@:
25773 ``@value{GCC}--2.8.1''.
25776 What compiler (and its version) was used to compile the program you are
25777 debugging---e.g.@: ``@value{GCC}--2.8.1'', or ``HP92453-01 A.10.32.03 HP
25778 C Compiler''. For @value{NGCC}, you can say @kbd{@value{GCC} --version}
25779 to get this information; for other compilers, see the documentation for
25783 The command arguments you gave the compiler to compile your example and
25784 observe the bug. For example, did you use @samp{-O}? To guarantee
25785 you will not omit something important, list them all. A copy of the
25786 Makefile (or the output from make) is sufficient.
25788 If we were to try to guess the arguments, we would probably guess wrong
25789 and then we might not encounter the bug.
25792 A complete input script, and all necessary source files, that will
25796 A description of what behavior you observe that you believe is
25797 incorrect. For example, ``It gets a fatal signal.''
25799 Of course, if the bug is that @value{GDBN} gets a fatal signal, then we
25800 will certainly notice it. But if the bug is incorrect output, we might
25801 not notice unless it is glaringly wrong. You might as well not give us
25802 a chance to make a mistake.
25804 Even if the problem you experience is a fatal signal, you should still
25805 say so explicitly. Suppose something strange is going on, such as, your
25806 copy of @value{GDBN} is out of synch, or you have encountered a bug in
25807 the C library on your system. (This has happened!) Your copy might
25808 crash and ours would not. If you told us to expect a crash, then when
25809 ours fails to crash, we would know that the bug was not happening for
25810 us. If you had not told us to expect a crash, then we would not be able
25811 to draw any conclusion from our observations.
25814 @cindex recording a session script
25815 To collect all this information, you can use a session recording program
25816 such as @command{script}, which is available on many Unix systems.
25817 Just run your @value{GDBN} session inside @command{script} and then
25818 include the @file{typescript} file with your bug report.
25820 Another way to record a @value{GDBN} session is to run @value{GDBN}
25821 inside Emacs and then save the entire buffer to a file.
25824 If you wish to suggest changes to the @value{GDBN} source, send us context
25825 diffs. If you even discuss something in the @value{GDBN} source, refer to
25826 it by context, not by line number.
25828 The line numbers in our development sources will not match those in your
25829 sources. Your line numbers would convey no useful information to us.
25833 Here are some things that are not necessary:
25837 A description of the envelope of the bug.
25839 Often people who encounter a bug spend a lot of time investigating
25840 which changes to the input file will make the bug go away and which
25841 changes will not affect it.
25843 This is often time consuming and not very useful, because the way we
25844 will find the bug is by running a single example under the debugger
25845 with breakpoints, not by pure deduction from a series of examples.
25846 We recommend that you save your time for something else.
25848 Of course, if you can find a simpler example to report @emph{instead}
25849 of the original one, that is a convenience for us. Errors in the
25850 output will be easier to spot, running under the debugger will take
25851 less time, and so on.
25853 However, simplification is not vital; if you do not want to do this,
25854 report the bug anyway and send us the entire test case you used.
25857 A patch for the bug.
25859 A patch for the bug does help us if it is a good one. But do not omit
25860 the necessary information, such as the test case, on the assumption that
25861 a patch is all we need. We might see problems with your patch and decide
25862 to fix the problem another way, or we might not understand it at all.
25864 Sometimes with a program as complicated as @value{GDBN} it is very hard to
25865 construct an example that will make the program follow a certain path
25866 through the code. If you do not send us the example, we will not be able
25867 to construct one, so we will not be able to verify that the bug is fixed.
25869 And if we cannot understand what bug you are trying to fix, or why your
25870 patch should be an improvement, we will not install it. A test case will
25871 help us to understand.
25874 A guess about what the bug is or what it depends on.
25876 Such guesses are usually wrong. Even we cannot guess right about such
25877 things without first using the debugger to find the facts.
25880 @c The readline documentation is distributed with the readline code
25881 @c and consists of the two following files:
25883 @c inc-hist.texinfo
25884 @c Use -I with makeinfo to point to the appropriate directory,
25885 @c environment var TEXINPUTS with TeX.
25886 @include rluser.texi
25887 @include inc-hist.texinfo
25890 @node Formatting Documentation
25891 @appendix Formatting Documentation
25893 @cindex @value{GDBN} reference card
25894 @cindex reference card
25895 The @value{GDBN} 4 release includes an already-formatted reference card, ready
25896 for printing with PostScript or Ghostscript, in the @file{gdb}
25897 subdirectory of the main source directory@footnote{In
25898 @file{gdb-@value{GDBVN}/gdb/refcard.ps} of the version @value{GDBVN}
25899 release.}. If you can use PostScript or Ghostscript with your printer,
25900 you can print the reference card immediately with @file{refcard.ps}.
25902 The release also includes the source for the reference card. You
25903 can format it, using @TeX{}, by typing:
25909 The @value{GDBN} reference card is designed to print in @dfn{landscape}
25910 mode on US ``letter'' size paper;
25911 that is, on a sheet 11 inches wide by 8.5 inches
25912 high. You will need to specify this form of printing as an option to
25913 your @sc{dvi} output program.
25915 @cindex documentation
25917 All the documentation for @value{GDBN} comes as part of the machine-readable
25918 distribution. The documentation is written in Texinfo format, which is
25919 a documentation system that uses a single source file to produce both
25920 on-line information and a printed manual. You can use one of the Info
25921 formatting commands to create the on-line version of the documentation
25922 and @TeX{} (or @code{texi2roff}) to typeset the printed version.
25924 @value{GDBN} includes an already formatted copy of the on-line Info
25925 version of this manual in the @file{gdb} subdirectory. The main Info
25926 file is @file{gdb-@value{GDBVN}/gdb/gdb.info}, and it refers to
25927 subordinate files matching @samp{gdb.info*} in the same directory. If
25928 necessary, you can print out these files, or read them with any editor;
25929 but they are easier to read using the @code{info} subsystem in @sc{gnu}
25930 Emacs or the standalone @code{info} program, available as part of the
25931 @sc{gnu} Texinfo distribution.
25933 If you want to format these Info files yourself, you need one of the
25934 Info formatting programs, such as @code{texinfo-format-buffer} or
25937 If you have @code{makeinfo} installed, and are in the top level
25938 @value{GDBN} source directory (@file{gdb-@value{GDBVN}}, in the case of
25939 version @value{GDBVN}), you can make the Info file by typing:
25946 If you want to typeset and print copies of this manual, you need @TeX{},
25947 a program to print its @sc{dvi} output files, and @file{texinfo.tex}, the
25948 Texinfo definitions file.
25950 @TeX{} is a typesetting program; it does not print files directly, but
25951 produces output files called @sc{dvi} files. To print a typeset
25952 document, you need a program to print @sc{dvi} files. If your system
25953 has @TeX{} installed, chances are it has such a program. The precise
25954 command to use depends on your system; @kbd{lpr -d} is common; another
25955 (for PostScript devices) is @kbd{dvips}. The @sc{dvi} print command may
25956 require a file name without any extension or a @samp{.dvi} extension.
25958 @TeX{} also requires a macro definitions file called
25959 @file{texinfo.tex}. This file tells @TeX{} how to typeset a document
25960 written in Texinfo format. On its own, @TeX{} cannot either read or
25961 typeset a Texinfo file. @file{texinfo.tex} is distributed with GDB
25962 and is located in the @file{gdb-@var{version-number}/texinfo}
25965 If you have @TeX{} and a @sc{dvi} printer program installed, you can
25966 typeset and print this manual. First switch to the @file{gdb}
25967 subdirectory of the main source directory (for example, to
25968 @file{gdb-@value{GDBVN}/gdb}) and type:
25974 Then give @file{gdb.dvi} to your @sc{dvi} printing program.
25976 @node Installing GDB
25977 @appendix Installing @value{GDBN}
25978 @cindex installation
25981 * Requirements:: Requirements for building @value{GDBN}
25982 * Running Configure:: Invoking the @value{GDBN} @file{configure} script
25983 * Separate Objdir:: Compiling @value{GDBN} in another directory
25984 * Config Names:: Specifying names for hosts and targets
25985 * Configure Options:: Summary of options for configure
25986 * System-wide configuration:: Having a system-wide init file
25990 @section Requirements for Building @value{GDBN}
25991 @cindex building @value{GDBN}, requirements for
25993 Building @value{GDBN} requires various tools and packages to be available.
25994 Other packages will be used only if they are found.
25996 @heading Tools/Packages Necessary for Building @value{GDBN}
25998 @item ISO C90 compiler
25999 @value{GDBN} is written in ISO C90. It should be buildable with any
26000 working C90 compiler, e.g.@: GCC.
26004 @heading Tools/Packages Optional for Building @value{GDBN}
26008 @value{GDBN} can use the Expat XML parsing library. This library may be
26009 included with your operating system distribution; if it is not, you
26010 can get the latest version from @url{http://expat.sourceforge.net}.
26011 The @file{configure} script will search for this library in several
26012 standard locations; if it is installed in an unusual path, you can
26013 use the @option{--with-libexpat-prefix} option to specify its location.
26019 Remote protocol memory maps (@pxref{Memory Map Format})
26021 Target descriptions (@pxref{Target Descriptions})
26023 Remote shared library lists (@pxref{Library List Format})
26025 MS-Windows shared libraries (@pxref{Shared Libraries})
26029 @cindex compressed debug sections
26030 @value{GDBN} will use the @samp{zlib} library, if available, to read
26031 compressed debug sections. Some linkers, such as GNU gold, are capable
26032 of producing binaries with compressed debug sections. If @value{GDBN}
26033 is compiled with @samp{zlib}, it will be able to read the debug
26034 information in such binaries.
26036 The @samp{zlib} library is likely included with your operating system
26037 distribution; if it is not, you can get the latest version from
26038 @url{http://zlib.net}.
26041 @value{GDBN}'s features related to character sets (@pxref{Character
26042 Sets}) require a functioning @code{iconv} implementation. If you are
26043 on a GNU system, then this is provided by the GNU C Library. Some
26044 other systems also provide a working @code{iconv}.
26046 On systems with @code{iconv}, you can install GNU Libiconv. If you
26047 have previously installed Libiconv, you can use the
26048 @option{--with-libiconv-prefix} option to configure.
26050 @value{GDBN}'s top-level @file{configure} and @file{Makefile} will
26051 arrange to build Libiconv if a directory named @file{libiconv} appears
26052 in the top-most source directory. If Libiconv is built this way, and
26053 if the operating system does not provide a suitable @code{iconv}
26054 implementation, then the just-built library will automatically be used
26055 by @value{GDBN}. One easy way to set this up is to download GNU
26056 Libiconv, unpack it, and then rename the directory holding the
26057 Libiconv source code to @samp{libiconv}.
26060 @node Running Configure
26061 @section Invoking the @value{GDBN} @file{configure} Script
26062 @cindex configuring @value{GDBN}
26063 @value{GDBN} comes with a @file{configure} script that automates the process
26064 of preparing @value{GDBN} for installation; you can then use @code{make} to
26065 build the @code{gdb} program.
26067 @c irrelevant in info file; it's as current as the code it lives with.
26068 @footnote{If you have a more recent version of @value{GDBN} than @value{GDBVN},
26069 look at the @file{README} file in the sources; we may have improved the
26070 installation procedures since publishing this manual.}
26073 The @value{GDBN} distribution includes all the source code you need for
26074 @value{GDBN} in a single directory, whose name is usually composed by
26075 appending the version number to @samp{gdb}.
26077 For example, the @value{GDBN} version @value{GDBVN} distribution is in the
26078 @file{gdb-@value{GDBVN}} directory. That directory contains:
26081 @item gdb-@value{GDBVN}/configure @r{(and supporting files)}
26082 script for configuring @value{GDBN} and all its supporting libraries
26084 @item gdb-@value{GDBVN}/gdb
26085 the source specific to @value{GDBN} itself
26087 @item gdb-@value{GDBVN}/bfd
26088 source for the Binary File Descriptor library
26090 @item gdb-@value{GDBVN}/include
26091 @sc{gnu} include files
26093 @item gdb-@value{GDBVN}/libiberty
26094 source for the @samp{-liberty} free software library
26096 @item gdb-@value{GDBVN}/opcodes
26097 source for the library of opcode tables and disassemblers
26099 @item gdb-@value{GDBVN}/readline
26100 source for the @sc{gnu} command-line interface
26102 @item gdb-@value{GDBVN}/glob
26103 source for the @sc{gnu} filename pattern-matching subroutine
26105 @item gdb-@value{GDBVN}/mmalloc
26106 source for the @sc{gnu} memory-mapped malloc package
26109 The simplest way to configure and build @value{GDBN} is to run @file{configure}
26110 from the @file{gdb-@var{version-number}} source directory, which in
26111 this example is the @file{gdb-@value{GDBVN}} directory.
26113 First switch to the @file{gdb-@var{version-number}} source directory
26114 if you are not already in it; then run @file{configure}. Pass the
26115 identifier for the platform on which @value{GDBN} will run as an
26121 cd gdb-@value{GDBVN}
26122 ./configure @var{host}
26127 where @var{host} is an identifier such as @samp{sun4} or
26128 @samp{decstation}, that identifies the platform where @value{GDBN} will run.
26129 (You can often leave off @var{host}; @file{configure} tries to guess the
26130 correct value by examining your system.)
26132 Running @samp{configure @var{host}} and then running @code{make} builds the
26133 @file{bfd}, @file{readline}, @file{mmalloc}, and @file{libiberty}
26134 libraries, then @code{gdb} itself. The configured source files, and the
26135 binaries, are left in the corresponding source directories.
26138 @file{configure} is a Bourne-shell (@code{/bin/sh}) script; if your
26139 system does not recognize this automatically when you run a different
26140 shell, you may need to run @code{sh} on it explicitly:
26143 sh configure @var{host}
26146 If you run @file{configure} from a directory that contains source
26147 directories for multiple libraries or programs, such as the
26148 @file{gdb-@value{GDBVN}} source directory for version @value{GDBVN},
26150 creates configuration files for every directory level underneath (unless
26151 you tell it not to, with the @samp{--norecursion} option).
26153 You should run the @file{configure} script from the top directory in the
26154 source tree, the @file{gdb-@var{version-number}} directory. If you run
26155 @file{configure} from one of the subdirectories, you will configure only
26156 that subdirectory. That is usually not what you want. In particular,
26157 if you run the first @file{configure} from the @file{gdb} subdirectory
26158 of the @file{gdb-@var{version-number}} directory, you will omit the
26159 configuration of @file{bfd}, @file{readline}, and other sibling
26160 directories of the @file{gdb} subdirectory. This leads to build errors
26161 about missing include files such as @file{bfd/bfd.h}.
26163 You can install @code{@value{GDBP}} anywhere; it has no hardwired paths.
26164 However, you should make sure that the shell on your path (named by
26165 the @samp{SHELL} environment variable) is publicly readable. Remember
26166 that @value{GDBN} uses the shell to start your program---some systems refuse to
26167 let @value{GDBN} debug child processes whose programs are not readable.
26169 @node Separate Objdir
26170 @section Compiling @value{GDBN} in Another Directory
26172 If you want to run @value{GDBN} versions for several host or target machines,
26173 you need a different @code{gdb} compiled for each combination of
26174 host and target. @file{configure} is designed to make this easy by
26175 allowing you to generate each configuration in a separate subdirectory,
26176 rather than in the source directory. If your @code{make} program
26177 handles the @samp{VPATH} feature (@sc{gnu} @code{make} does), running
26178 @code{make} in each of these directories builds the @code{gdb}
26179 program specified there.
26181 To build @code{gdb} in a separate directory, run @file{configure}
26182 with the @samp{--srcdir} option to specify where to find the source.
26183 (You also need to specify a path to find @file{configure}
26184 itself from your working directory. If the path to @file{configure}
26185 would be the same as the argument to @samp{--srcdir}, you can leave out
26186 the @samp{--srcdir} option; it is assumed.)
26188 For example, with version @value{GDBVN}, you can build @value{GDBN} in a
26189 separate directory for a Sun 4 like this:
26193 cd gdb-@value{GDBVN}
26196 ../gdb-@value{GDBVN}/configure sun4
26201 When @file{configure} builds a configuration using a remote source
26202 directory, it creates a tree for the binaries with the same structure
26203 (and using the same names) as the tree under the source directory. In
26204 the example, you'd find the Sun 4 library @file{libiberty.a} in the
26205 directory @file{gdb-sun4/libiberty}, and @value{GDBN} itself in
26206 @file{gdb-sun4/gdb}.
26208 Make sure that your path to the @file{configure} script has just one
26209 instance of @file{gdb} in it. If your path to @file{configure} looks
26210 like @file{../gdb-@value{GDBVN}/gdb/configure}, you are configuring only
26211 one subdirectory of @value{GDBN}, not the whole package. This leads to
26212 build errors about missing include files such as @file{bfd/bfd.h}.
26214 One popular reason to build several @value{GDBN} configurations in separate
26215 directories is to configure @value{GDBN} for cross-compiling (where
26216 @value{GDBN} runs on one machine---the @dfn{host}---while debugging
26217 programs that run on another machine---the @dfn{target}).
26218 You specify a cross-debugging target by
26219 giving the @samp{--target=@var{target}} option to @file{configure}.
26221 When you run @code{make} to build a program or library, you must run
26222 it in a configured directory---whatever directory you were in when you
26223 called @file{configure} (or one of its subdirectories).
26225 The @code{Makefile} that @file{configure} generates in each source
26226 directory also runs recursively. If you type @code{make} in a source
26227 directory such as @file{gdb-@value{GDBVN}} (or in a separate configured
26228 directory configured with @samp{--srcdir=@var{dirname}/gdb-@value{GDBVN}}), you
26229 will build all the required libraries, and then build GDB.
26231 When you have multiple hosts or targets configured in separate
26232 directories, you can run @code{make} on them in parallel (for example,
26233 if they are NFS-mounted on each of the hosts); they will not interfere
26237 @section Specifying Names for Hosts and Targets
26239 The specifications used for hosts and targets in the @file{configure}
26240 script are based on a three-part naming scheme, but some short predefined
26241 aliases are also supported. The full naming scheme encodes three pieces
26242 of information in the following pattern:
26245 @var{architecture}-@var{vendor}-@var{os}
26248 For example, you can use the alias @code{sun4} as a @var{host} argument,
26249 or as the value for @var{target} in a @code{--target=@var{target}}
26250 option. The equivalent full name is @samp{sparc-sun-sunos4}.
26252 The @file{configure} script accompanying @value{GDBN} does not provide
26253 any query facility to list all supported host and target names or
26254 aliases. @file{configure} calls the Bourne shell script
26255 @code{config.sub} to map abbreviations to full names; you can read the
26256 script, if you wish, or you can use it to test your guesses on
26257 abbreviations---for example:
26260 % sh config.sub i386-linux
26262 % sh config.sub alpha-linux
26263 alpha-unknown-linux-gnu
26264 % sh config.sub hp9k700
26266 % sh config.sub sun4
26267 sparc-sun-sunos4.1.1
26268 % sh config.sub sun3
26269 m68k-sun-sunos4.1.1
26270 % sh config.sub i986v
26271 Invalid configuration `i986v': machine `i986v' not recognized
26275 @code{config.sub} is also distributed in the @value{GDBN} source
26276 directory (@file{gdb-@value{GDBVN}}, for version @value{GDBVN}).
26278 @node Configure Options
26279 @section @file{configure} Options
26281 Here is a summary of the @file{configure} options and arguments that
26282 are most often useful for building @value{GDBN}. @file{configure} also has
26283 several other options not listed here. @inforef{What Configure
26284 Does,,configure.info}, for a full explanation of @file{configure}.
26287 configure @r{[}--help@r{]}
26288 @r{[}--prefix=@var{dir}@r{]}
26289 @r{[}--exec-prefix=@var{dir}@r{]}
26290 @r{[}--srcdir=@var{dirname}@r{]}
26291 @r{[}--norecursion@r{]} @r{[}--rm@r{]}
26292 @r{[}--target=@var{target}@r{]}
26297 You may introduce options with a single @samp{-} rather than
26298 @samp{--} if you prefer; but you may abbreviate option names if you use
26303 Display a quick summary of how to invoke @file{configure}.
26305 @item --prefix=@var{dir}
26306 Configure the source to install programs and files under directory
26309 @item --exec-prefix=@var{dir}
26310 Configure the source to install programs under directory
26313 @c avoid splitting the warning from the explanation:
26315 @item --srcdir=@var{dirname}
26316 @strong{Warning: using this option requires @sc{gnu} @code{make}, or another
26317 @code{make} that implements the @code{VPATH} feature.}@*
26318 Use this option to make configurations in directories separate from the
26319 @value{GDBN} source directories. Among other things, you can use this to
26320 build (or maintain) several configurations simultaneously, in separate
26321 directories. @file{configure} writes configuration-specific files in
26322 the current directory, but arranges for them to use the source in the
26323 directory @var{dirname}. @file{configure} creates directories under
26324 the working directory in parallel to the source directories below
26327 @item --norecursion
26328 Configure only the directory level where @file{configure} is executed; do not
26329 propagate configuration to subdirectories.
26331 @item --target=@var{target}
26332 Configure @value{GDBN} for cross-debugging programs running on the specified
26333 @var{target}. Without this option, @value{GDBN} is configured to debug
26334 programs that run on the same machine (@var{host}) as @value{GDBN} itself.
26336 There is no convenient way to generate a list of all available targets.
26338 @item @var{host} @dots{}
26339 Configure @value{GDBN} to run on the specified @var{host}.
26341 There is no convenient way to generate a list of all available hosts.
26344 There are many other options available as well, but they are generally
26345 needed for special purposes only.
26347 @node System-wide configuration
26348 @section System-wide configuration and settings
26349 @cindex system-wide init file
26351 @value{GDBN} can be configured to have a system-wide init file;
26352 this file will be read and executed at startup (@pxref{Startup, , What
26353 @value{GDBN} does during startup}).
26355 Here is the corresponding configure option:
26358 @item --with-system-gdbinit=@var{file}
26359 Specify that the default location of the system-wide init file is
26363 If @value{GDBN} has been configured with the option @option{--prefix=$prefix},
26364 it may be subject to relocation. Two possible cases:
26368 If the default location of this init file contains @file{$prefix},
26369 it will be subject to relocation. Suppose that the configure options
26370 are @option{--prefix=$prefix --with-system-gdbinit=$prefix/etc/gdbinit};
26371 if @value{GDBN} is moved from @file{$prefix} to @file{$install}, the system
26372 init file is looked for as @file{$install/etc/gdbinit} instead of
26373 @file{$prefix/etc/gdbinit}.
26376 By contrast, if the default location does not contain the prefix,
26377 it will not be relocated. E.g.@: if @value{GDBN} has been configured with
26378 @option{--prefix=/usr/local --with-system-gdbinit=/usr/share/gdb/gdbinit},
26379 then @value{GDBN} will always look for @file{/usr/share/gdb/gdbinit},
26380 wherever @value{GDBN} is installed.
26383 @node Maintenance Commands
26384 @appendix Maintenance Commands
26385 @cindex maintenance commands
26386 @cindex internal commands
26388 In addition to commands intended for @value{GDBN} users, @value{GDBN}
26389 includes a number of commands intended for @value{GDBN} developers,
26390 that are not documented elsewhere in this manual. These commands are
26391 provided here for reference. (For commands that turn on debugging
26392 messages, see @ref{Debugging Output}.)
26395 @kindex maint agent
26396 @item maint agent @var{expression}
26397 Translate the given @var{expression} into remote agent bytecodes.
26398 This command is useful for debugging the Agent Expression mechanism
26399 (@pxref{Agent Expressions}).
26401 @kindex maint info breakpoints
26402 @item @anchor{maint info breakpoints}maint info breakpoints
26403 Using the same format as @samp{info breakpoints}, display both the
26404 breakpoints you've set explicitly, and those @value{GDBN} is using for
26405 internal purposes. Internal breakpoints are shown with negative
26406 breakpoint numbers. The type column identifies what kind of breakpoint
26411 Normal, explicitly set breakpoint.
26414 Normal, explicitly set watchpoint.
26417 Internal breakpoint, used to handle correctly stepping through
26418 @code{longjmp} calls.
26420 @item longjmp resume
26421 Internal breakpoint at the target of a @code{longjmp}.
26424 Temporary internal breakpoint used by the @value{GDBN} @code{until} command.
26427 Temporary internal breakpoint used by the @value{GDBN} @code{finish} command.
26430 Shared library events.
26434 @kindex set displaced-stepping
26435 @kindex show displaced-stepping
26436 @cindex displaced stepping support
26437 @cindex out-of-line single-stepping
26438 @item set displaced-stepping
26439 @itemx show displaced-stepping
26440 Control whether or not @value{GDBN} will do @dfn{displaced stepping}
26441 if the target supports it. Displaced stepping is a way to single-step
26442 over breakpoints without removing them from the inferior, by executing
26443 an out-of-line copy of the instruction that was originally at the
26444 breakpoint location. It is also known as out-of-line single-stepping.
26447 @item set displaced-stepping on
26448 If the target architecture supports it, @value{GDBN} will use
26449 displaced stepping to step over breakpoints.
26451 @item set displaced-stepping off
26452 @value{GDBN} will not use displaced stepping to step over breakpoints,
26453 even if such is supported by the target architecture.
26455 @cindex non-stop mode, and @samp{set displaced-stepping}
26456 @item set displaced-stepping auto
26457 This is the default mode. @value{GDBN} will use displaced stepping
26458 only if non-stop mode is active (@pxref{Non-Stop Mode}) and the target
26459 architecture supports displaced stepping.
26462 @kindex maint check-symtabs
26463 @item maint check-symtabs
26464 Check the consistency of psymtabs and symtabs.
26466 @kindex maint cplus first_component
26467 @item maint cplus first_component @var{name}
26468 Print the first C@t{++} class/namespace component of @var{name}.
26470 @kindex maint cplus namespace
26471 @item maint cplus namespace
26472 Print the list of possible C@t{++} namespaces.
26474 @kindex maint demangle
26475 @item maint demangle @var{name}
26476 Demangle a C@t{++} or Objective-C mangled @var{name}.
26478 @kindex maint deprecate
26479 @kindex maint undeprecate
26480 @cindex deprecated commands
26481 @item maint deprecate @var{command} @r{[}@var{replacement}@r{]}
26482 @itemx maint undeprecate @var{command}
26483 Deprecate or undeprecate the named @var{command}. Deprecated commands
26484 cause @value{GDBN} to issue a warning when you use them. The optional
26485 argument @var{replacement} says which newer command should be used in
26486 favor of the deprecated one; if it is given, @value{GDBN} will mention
26487 the replacement as part of the warning.
26489 @kindex maint dump-me
26490 @item maint dump-me
26491 @cindex @code{SIGQUIT} signal, dump core of @value{GDBN}
26492 Cause a fatal signal in the debugger and force it to dump its core.
26493 This is supported only on systems which support aborting a program
26494 with the @code{SIGQUIT} signal.
26496 @kindex maint internal-error
26497 @kindex maint internal-warning
26498 @item maint internal-error @r{[}@var{message-text}@r{]}
26499 @itemx maint internal-warning @r{[}@var{message-text}@r{]}
26500 Cause @value{GDBN} to call the internal function @code{internal_error}
26501 or @code{internal_warning} and hence behave as though an internal error
26502 or internal warning has been detected. In addition to reporting the
26503 internal problem, these functions give the user the opportunity to
26504 either quit @value{GDBN} or create a core file of the current
26505 @value{GDBN} session.
26507 These commands take an optional parameter @var{message-text} that is
26508 used as the text of the error or warning message.
26510 Here's an example of using @code{internal-error}:
26513 (@value{GDBP}) @kbd{maint internal-error testing, 1, 2}
26514 @dots{}/maint.c:121: internal-error: testing, 1, 2
26515 A problem internal to GDB has been detected. Further
26516 debugging may prove unreliable.
26517 Quit this debugging session? (y or n) @kbd{n}
26518 Create a core file? (y or n) @kbd{n}
26522 @cindex @value{GDBN} internal error
26523 @cindex internal errors, control of @value{GDBN} behavior
26525 @kindex maint set internal-error
26526 @kindex maint show internal-error
26527 @kindex maint set internal-warning
26528 @kindex maint show internal-warning
26529 @item maint set internal-error @var{action} [ask|yes|no]
26530 @itemx maint show internal-error @var{action}
26531 @itemx maint set internal-warning @var{action} [ask|yes|no]
26532 @itemx maint show internal-warning @var{action}
26533 When @value{GDBN} reports an internal problem (error or warning) it
26534 gives the user the opportunity to both quit @value{GDBN} and create a
26535 core file of the current @value{GDBN} session. These commands let you
26536 override the default behaviour for each particular @var{action},
26537 described in the table below.
26541 You can specify that @value{GDBN} should always (yes) or never (no)
26542 quit. The default is to ask the user what to do.
26545 You can specify that @value{GDBN} should always (yes) or never (no)
26546 create a core file. The default is to ask the user what to do.
26549 @kindex maint packet
26550 @item maint packet @var{text}
26551 If @value{GDBN} is talking to an inferior via the serial protocol,
26552 then this command sends the string @var{text} to the inferior, and
26553 displays the response packet. @value{GDBN} supplies the initial
26554 @samp{$} character, the terminating @samp{#} character, and the
26557 @kindex maint print architecture
26558 @item maint print architecture @r{[}@var{file}@r{]}
26559 Print the entire architecture configuration. The optional argument
26560 @var{file} names the file where the output goes.
26562 @kindex maint print c-tdesc
26563 @item maint print c-tdesc
26564 Print the current target description (@pxref{Target Descriptions}) as
26565 a C source file. The created source file can be used in @value{GDBN}
26566 when an XML parser is not available to parse the description.
26568 @kindex maint print dummy-frames
26569 @item maint print dummy-frames
26570 Prints the contents of @value{GDBN}'s internal dummy-frame stack.
26573 (@value{GDBP}) @kbd{b add}
26575 (@value{GDBP}) @kbd{print add(2,3)}
26576 Breakpoint 2, add (a=2, b=3) at @dots{}
26578 The program being debugged stopped while in a function called from GDB.
26580 (@value{GDBP}) @kbd{maint print dummy-frames}
26581 0x1a57c80: pc=0x01014068 fp=0x0200bddc sp=0x0200bdd6
26582 top=0x0200bdd4 id=@{stack=0x200bddc,code=0x101405c@}
26583 call_lo=0x01014000 call_hi=0x01014001
26587 Takes an optional file parameter.
26589 @kindex maint print registers
26590 @kindex maint print raw-registers
26591 @kindex maint print cooked-registers
26592 @kindex maint print register-groups
26593 @item maint print registers @r{[}@var{file}@r{]}
26594 @itemx maint print raw-registers @r{[}@var{file}@r{]}
26595 @itemx maint print cooked-registers @r{[}@var{file}@r{]}
26596 @itemx maint print register-groups @r{[}@var{file}@r{]}
26597 Print @value{GDBN}'s internal register data structures.
26599 The command @code{maint print raw-registers} includes the contents of
26600 the raw register cache; the command @code{maint print cooked-registers}
26601 includes the (cooked) value of all registers; and the command
26602 @code{maint print register-groups} includes the groups that each
26603 register is a member of. @xref{Registers,, Registers, gdbint,
26604 @value{GDBN} Internals}.
26606 These commands take an optional parameter, a file name to which to
26607 write the information.
26609 @kindex maint print reggroups
26610 @item maint print reggroups @r{[}@var{file}@r{]}
26611 Print @value{GDBN}'s internal register group data structures. The
26612 optional argument @var{file} tells to what file to write the
26615 The register groups info looks like this:
26618 (@value{GDBP}) @kbd{maint print reggroups}
26631 This command forces @value{GDBN} to flush its internal register cache.
26633 @kindex maint print objfiles
26634 @cindex info for known object files
26635 @item maint print objfiles
26636 Print a dump of all known object files. For each object file, this
26637 command prints its name, address in memory, and all of its psymtabs
26640 @kindex maint print statistics
26641 @cindex bcache statistics
26642 @item maint print statistics
26643 This command prints, for each object file in the program, various data
26644 about that object file followed by the byte cache (@dfn{bcache})
26645 statistics for the object file. The objfile data includes the number
26646 of minimal, partial, full, and stabs symbols, the number of types
26647 defined by the objfile, the number of as yet unexpanded psym tables,
26648 the number of line tables and string tables, and the amount of memory
26649 used by the various tables. The bcache statistics include the counts,
26650 sizes, and counts of duplicates of all and unique objects, max,
26651 average, and median entry size, total memory used and its overhead and
26652 savings, and various measures of the hash table size and chain
26655 @kindex maint print target-stack
26656 @cindex target stack description
26657 @item maint print target-stack
26658 A @dfn{target} is an interface between the debugger and a particular
26659 kind of file or process. Targets can be stacked in @dfn{strata},
26660 so that more than one target can potentially respond to a request.
26661 In particular, memory accesses will walk down the stack of targets
26662 until they find a target that is interested in handling that particular
26665 This command prints a short description of each layer that was pushed on
26666 the @dfn{target stack}, starting from the top layer down to the bottom one.
26668 @kindex maint print type
26669 @cindex type chain of a data type
26670 @item maint print type @var{expr}
26671 Print the type chain for a type specified by @var{expr}. The argument
26672 can be either a type name or a symbol. If it is a symbol, the type of
26673 that symbol is described. The type chain produced by this command is
26674 a recursive definition of the data type as stored in @value{GDBN}'s
26675 data structures, including its flags and contained types.
26677 @kindex maint set dwarf2 max-cache-age
26678 @kindex maint show dwarf2 max-cache-age
26679 @item maint set dwarf2 max-cache-age
26680 @itemx maint show dwarf2 max-cache-age
26681 Control the DWARF 2 compilation unit cache.
26683 @cindex DWARF 2 compilation units cache
26684 In object files with inter-compilation-unit references, such as those
26685 produced by the GCC option @samp{-feliminate-dwarf2-dups}, the DWARF 2
26686 reader needs to frequently refer to previously read compilation units.
26687 This setting controls how long a compilation unit will remain in the
26688 cache if it is not referenced. A higher limit means that cached
26689 compilation units will be stored in memory longer, and more total
26690 memory will be used. Setting it to zero disables caching, which will
26691 slow down @value{GDBN} startup, but reduce memory consumption.
26693 @kindex maint set profile
26694 @kindex maint show profile
26695 @cindex profiling GDB
26696 @item maint set profile
26697 @itemx maint show profile
26698 Control profiling of @value{GDBN}.
26700 Profiling will be disabled until you use the @samp{maint set profile}
26701 command to enable it. When you enable profiling, the system will begin
26702 collecting timing and execution count data; when you disable profiling or
26703 exit @value{GDBN}, the results will be written to a log file. Remember that
26704 if you use profiling, @value{GDBN} will overwrite the profiling log file
26705 (often called @file{gmon.out}). If you have a record of important profiling
26706 data in a @file{gmon.out} file, be sure to move it to a safe location.
26708 Configuring with @samp{--enable-profiling} arranges for @value{GDBN} to be
26709 compiled with the @samp{-pg} compiler option.
26711 @kindex maint set show-debug-regs
26712 @kindex maint show show-debug-regs
26713 @cindex hardware debug registers
26714 @item maint set show-debug-regs
26715 @itemx maint show show-debug-regs
26716 Control whether to show variables that mirror the hardware debug
26717 registers. Use @code{ON} to enable, @code{OFF} to disable. If
26718 enabled, the debug registers values are shown when @value{GDBN} inserts or
26719 removes a hardware breakpoint or watchpoint, and when the inferior
26720 triggers a hardware-assisted breakpoint or watchpoint.
26722 @kindex maint space
26723 @cindex memory used by commands
26725 Control whether to display memory usage for each command. If set to a
26726 nonzero value, @value{GDBN} will display how much memory each command
26727 took, following the command's own output. This can also be requested
26728 by invoking @value{GDBN} with the @option{--statistics} command-line
26729 switch (@pxref{Mode Options}).
26732 @cindex time of command execution
26734 Control whether to display the execution time for each command. If
26735 set to a nonzero value, @value{GDBN} will display how much time it
26736 took to execute each command, following the command's own output.
26737 The time is not printed for the commands that run the target, since
26738 there's no mechanism currently to compute how much time was spend
26739 by @value{GDBN} and how much time was spend by the program been debugged.
26740 it's not possibly currently
26741 This can also be requested by invoking @value{GDBN} with the
26742 @option{--statistics} command-line switch (@pxref{Mode Options}).
26744 @kindex maint translate-address
26745 @item maint translate-address @r{[}@var{section}@r{]} @var{addr}
26746 Find the symbol stored at the location specified by the address
26747 @var{addr} and an optional section name @var{section}. If found,
26748 @value{GDBN} prints the name of the closest symbol and an offset from
26749 the symbol's location to the specified address. This is similar to
26750 the @code{info address} command (@pxref{Symbols}), except that this
26751 command also allows to find symbols in other sections.
26753 If section was not specified, the section in which the symbol was found
26754 is also printed. For dynamically linked executables, the name of
26755 executable or shared library containing the symbol is printed as well.
26759 The following command is useful for non-interactive invocations of
26760 @value{GDBN}, such as in the test suite.
26763 @item set watchdog @var{nsec}
26764 @kindex set watchdog
26765 @cindex watchdog timer
26766 @cindex timeout for commands
26767 Set the maximum number of seconds @value{GDBN} will wait for the
26768 target operation to finish. If this time expires, @value{GDBN}
26769 reports and error and the command is aborted.
26771 @item show watchdog
26772 Show the current setting of the target wait timeout.
26775 @node Remote Protocol
26776 @appendix @value{GDBN} Remote Serial Protocol
26781 * Stop Reply Packets::
26782 * General Query Packets::
26783 * Register Packet Format::
26784 * Tracepoint Packets::
26785 * Host I/O Packets::
26787 * Notification Packets::
26788 * Remote Non-Stop::
26789 * Packet Acknowledgment::
26791 * File-I/O Remote Protocol Extension::
26792 * Library List Format::
26793 * Memory Map Format::
26799 There may be occasions when you need to know something about the
26800 protocol---for example, if there is only one serial port to your target
26801 machine, you might want your program to do something special if it
26802 recognizes a packet meant for @value{GDBN}.
26804 In the examples below, @samp{->} and @samp{<-} are used to indicate
26805 transmitted and received data, respectively.
26807 @cindex protocol, @value{GDBN} remote serial
26808 @cindex serial protocol, @value{GDBN} remote
26809 @cindex remote serial protocol
26810 All @value{GDBN} commands and responses (other than acknowledgments
26811 and notifications, see @ref{Notification Packets}) are sent as a
26812 @var{packet}. A @var{packet} is introduced with the character
26813 @samp{$}, the actual @var{packet-data}, and the terminating character
26814 @samp{#} followed by a two-digit @var{checksum}:
26817 @code{$}@var{packet-data}@code{#}@var{checksum}
26821 @cindex checksum, for @value{GDBN} remote
26823 The two-digit @var{checksum} is computed as the modulo 256 sum of all
26824 characters between the leading @samp{$} and the trailing @samp{#} (an
26825 eight bit unsigned checksum).
26827 Implementors should note that prior to @value{GDBN} 5.0 the protocol
26828 specification also included an optional two-digit @var{sequence-id}:
26831 @code{$}@var{sequence-id}@code{:}@var{packet-data}@code{#}@var{checksum}
26834 @cindex sequence-id, for @value{GDBN} remote
26836 That @var{sequence-id} was appended to the acknowledgment. @value{GDBN}
26837 has never output @var{sequence-id}s. Stubs that handle packets added
26838 since @value{GDBN} 5.0 must not accept @var{sequence-id}.
26840 When either the host or the target machine receives a packet, the first
26841 response expected is an acknowledgment: either @samp{+} (to indicate
26842 the package was received correctly) or @samp{-} (to request
26846 -> @code{$}@var{packet-data}@code{#}@var{checksum}
26851 The @samp{+}/@samp{-} acknowledgments can be disabled
26852 once a connection is established.
26853 @xref{Packet Acknowledgment}, for details.
26855 The host (@value{GDBN}) sends @var{command}s, and the target (the
26856 debugging stub incorporated in your program) sends a @var{response}. In
26857 the case of step and continue @var{command}s, the response is only sent
26858 when the operation has completed, and the target has again stopped all
26859 threads in all attached processes. This is the default all-stop mode
26860 behavior, but the remote protocol also supports @value{GDBN}'s non-stop
26861 execution mode; see @ref{Remote Non-Stop}, for details.
26863 @var{packet-data} consists of a sequence of characters with the
26864 exception of @samp{#} and @samp{$} (see @samp{X} packet for additional
26867 @cindex remote protocol, field separator
26868 Fields within the packet should be separated using @samp{,} @samp{;} or
26869 @samp{:}. Except where otherwise noted all numbers are represented in
26870 @sc{hex} with leading zeros suppressed.
26872 Implementors should note that prior to @value{GDBN} 5.0, the character
26873 @samp{:} could not appear as the third character in a packet (as it
26874 would potentially conflict with the @var{sequence-id}).
26876 @cindex remote protocol, binary data
26877 @anchor{Binary Data}
26878 Binary data in most packets is encoded either as two hexadecimal
26879 digits per byte of binary data. This allowed the traditional remote
26880 protocol to work over connections which were only seven-bit clean.
26881 Some packets designed more recently assume an eight-bit clean
26882 connection, and use a more efficient encoding to send and receive
26885 The binary data representation uses @code{7d} (@sc{ascii} @samp{@}})
26886 as an escape character. Any escaped byte is transmitted as the escape
26887 character followed by the original character XORed with @code{0x20}.
26888 For example, the byte @code{0x7d} would be transmitted as the two
26889 bytes @code{0x7d 0x5d}. The bytes @code{0x23} (@sc{ascii} @samp{#}),
26890 @code{0x24} (@sc{ascii} @samp{$}), and @code{0x7d} (@sc{ascii}
26891 @samp{@}}) must always be escaped. Responses sent by the stub
26892 must also escape @code{0x2a} (@sc{ascii} @samp{*}), so that it
26893 is not interpreted as the start of a run-length encoded sequence
26896 Response @var{data} can be run-length encoded to save space.
26897 Run-length encoding replaces runs of identical characters with one
26898 instance of the repeated character, followed by a @samp{*} and a
26899 repeat count. The repeat count is itself sent encoded, to avoid
26900 binary characters in @var{data}: a value of @var{n} is sent as
26901 @code{@var{n}+29}. For a repeat count greater or equal to 3, this
26902 produces a printable @sc{ascii} character, e.g.@: a space (@sc{ascii}
26903 code 32) for a repeat count of 3. (This is because run-length
26904 encoding starts to win for counts 3 or more.) Thus, for example,
26905 @samp{0* } is a run-length encoding of ``0000'': the space character
26906 after @samp{*} means repeat the leading @code{0} @w{@code{32 - 29 =
26909 The printable characters @samp{#} and @samp{$} or with a numeric value
26910 greater than 126 must not be used. Runs of six repeats (@samp{#}) or
26911 seven repeats (@samp{$}) can be expanded using a repeat count of only
26912 five (@samp{"}). For example, @samp{00000000} can be encoded as
26915 The error response returned for some packets includes a two character
26916 error number. That number is not well defined.
26918 @cindex empty response, for unsupported packets
26919 For any @var{command} not supported by the stub, an empty response
26920 (@samp{$#00}) should be returned. That way it is possible to extend the
26921 protocol. A newer @value{GDBN} can tell if a packet is supported based
26924 A stub is required to support the @samp{g}, @samp{G}, @samp{m}, @samp{M},
26925 @samp{c}, and @samp{s} @var{command}s. All other @var{command}s are
26931 The following table provides a complete list of all currently defined
26932 @var{command}s and their corresponding response @var{data}.
26933 @xref{File-I/O Remote Protocol Extension}, for details about the File
26934 I/O extension of the remote protocol.
26936 Each packet's description has a template showing the packet's overall
26937 syntax, followed by an explanation of the packet's meaning. We
26938 include spaces in some of the templates for clarity; these are not
26939 part of the packet's syntax. No @value{GDBN} packet uses spaces to
26940 separate its components. For example, a template like @samp{foo
26941 @var{bar} @var{baz}} describes a packet beginning with the three ASCII
26942 bytes @samp{foo}, followed by a @var{bar}, followed directly by a
26943 @var{baz}. @value{GDBN} does not transmit a space character between the
26944 @samp{foo} and the @var{bar}, or between the @var{bar} and the
26947 @cindex @var{thread-id}, in remote protocol
26948 @anchor{thread-id syntax}
26949 Several packets and replies include a @var{thread-id} field to identify
26950 a thread. Normally these are positive numbers with a target-specific
26951 interpretation, formatted as big-endian hex strings. A @var{thread-id}
26952 can also be a literal @samp{-1} to indicate all threads, or @samp{0} to
26955 In addition, the remote protocol supports a multiprocess feature in
26956 which the @var{thread-id} syntax is extended to optionally include both
26957 process and thread ID fields, as @samp{p@var{pid}.@var{tid}}.
26958 The @var{pid} (process) and @var{tid} (thread) components each have the
26959 format described above: a positive number with target-specific
26960 interpretation formatted as a big-endian hex string, literal @samp{-1}
26961 to indicate all processes or threads (respectively), or @samp{0} to
26962 indicate an arbitrary process or thread. Specifying just a process, as
26963 @samp{p@var{pid}}, is equivalent to @samp{p@var{pid}.-1}. It is an
26964 error to specify all processes but a specific thread, such as
26965 @samp{p-1.@var{tid}}. Note that the @samp{p} prefix is @emph{not} used
26966 for those packets and replies explicitly documented to include a process
26967 ID, rather than a @var{thread-id}.
26969 The multiprocess @var{thread-id} syntax extensions are only used if both
26970 @value{GDBN} and the stub report support for the @samp{multiprocess}
26971 feature using @samp{qSupported}. @xref{multiprocess extensions}, for
26974 Note that all packet forms beginning with an upper- or lower-case
26975 letter, other than those described here, are reserved for future use.
26977 Here are the packet descriptions.
26982 @cindex @samp{!} packet
26983 @anchor{extended mode}
26984 Enable extended mode. In extended mode, the remote server is made
26985 persistent. The @samp{R} packet is used to restart the program being
26991 The remote target both supports and has enabled extended mode.
26995 @cindex @samp{?} packet
26996 Indicate the reason the target halted. The reply is the same as for
26997 step and continue. This packet has a special interpretation when the
26998 target is in non-stop mode; see @ref{Remote Non-Stop}.
27001 @xref{Stop Reply Packets}, for the reply specifications.
27003 @item A @var{arglen},@var{argnum},@var{arg},@dots{}
27004 @cindex @samp{A} packet
27005 Initialized @code{argv[]} array passed into program. @var{arglen}
27006 specifies the number of bytes in the hex encoded byte stream
27007 @var{arg}. See @code{gdbserver} for more details.
27012 The arguments were set.
27018 @cindex @samp{b} packet
27019 (Don't use this packet; its behavior is not well-defined.)
27020 Change the serial line speed to @var{baud}.
27022 JTC: @emph{When does the transport layer state change? When it's
27023 received, or after the ACK is transmitted. In either case, there are
27024 problems if the command or the acknowledgment packet is dropped.}
27026 Stan: @emph{If people really wanted to add something like this, and get
27027 it working for the first time, they ought to modify ser-unix.c to send
27028 some kind of out-of-band message to a specially-setup stub and have the
27029 switch happen "in between" packets, so that from remote protocol's point
27030 of view, nothing actually happened.}
27032 @item B @var{addr},@var{mode}
27033 @cindex @samp{B} packet
27034 Set (@var{mode} is @samp{S}) or clear (@var{mode} is @samp{C}) a
27035 breakpoint at @var{addr}.
27037 Don't use this packet. Use the @samp{Z} and @samp{z} packets instead
27038 (@pxref{insert breakpoint or watchpoint packet}).
27041 @cindex @samp{bc} packet
27042 Backward continue. Execute the target system in reverse. No parameter.
27043 @xref{Reverse Execution}, for more information.
27046 @xref{Stop Reply Packets}, for the reply specifications.
27049 @cindex @samp{bs} packet
27050 Backward single step. Execute one instruction in reverse. No parameter.
27051 @xref{Reverse Execution}, for more information.
27054 @xref{Stop Reply Packets}, for the reply specifications.
27056 @item c @r{[}@var{addr}@r{]}
27057 @cindex @samp{c} packet
27058 Continue. @var{addr} is address to resume. If @var{addr} is omitted,
27059 resume at current address.
27062 @xref{Stop Reply Packets}, for the reply specifications.
27064 @item C @var{sig}@r{[};@var{addr}@r{]}
27065 @cindex @samp{C} packet
27066 Continue with signal @var{sig} (hex signal number). If
27067 @samp{;@var{addr}} is omitted, resume at same address.
27070 @xref{Stop Reply Packets}, for the reply specifications.
27073 @cindex @samp{d} packet
27076 Don't use this packet; instead, define a general set packet
27077 (@pxref{General Query Packets}).
27081 @cindex @samp{D} packet
27082 The first form of the packet is used to detach @value{GDBN} from the
27083 remote system. It is sent to the remote target
27084 before @value{GDBN} disconnects via the @code{detach} command.
27086 The second form, including a process ID, is used when multiprocess
27087 protocol extensions are enabled (@pxref{multiprocess extensions}), to
27088 detach only a specific process. The @var{pid} is specified as a
27089 big-endian hex string.
27099 @item F @var{RC},@var{EE},@var{CF};@var{XX}
27100 @cindex @samp{F} packet
27101 A reply from @value{GDBN} to an @samp{F} packet sent by the target.
27102 This is part of the File-I/O protocol extension. @xref{File-I/O
27103 Remote Protocol Extension}, for the specification.
27106 @anchor{read registers packet}
27107 @cindex @samp{g} packet
27108 Read general registers.
27112 @item @var{XX@dots{}}
27113 Each byte of register data is described by two hex digits. The bytes
27114 with the register are transmitted in target byte order. The size of
27115 each register and their position within the @samp{g} packet are
27116 determined by the @value{GDBN} internal gdbarch functions
27117 @code{DEPRECATED_REGISTER_RAW_SIZE} and @code{gdbarch_register_name}. The
27118 specification of several standard @samp{g} packets is specified below.
27123 @item G @var{XX@dots{}}
27124 @cindex @samp{G} packet
27125 Write general registers. @xref{read registers packet}, for a
27126 description of the @var{XX@dots{}} data.
27136 @item H @var{c} @var{thread-id}
27137 @cindex @samp{H} packet
27138 Set thread for subsequent operations (@samp{m}, @samp{M}, @samp{g},
27139 @samp{G}, et.al.). @var{c} depends on the operation to be performed: it
27140 should be @samp{c} for step and continue operations, @samp{g} for other
27141 operations. The thread designator @var{thread-id} has the format and
27142 interpretation described in @ref{thread-id syntax}.
27153 @c 'H': How restrictive (or permissive) is the thread model. If a
27154 @c thread is selected and stopped, are other threads allowed
27155 @c to continue to execute? As I mentioned above, I think the
27156 @c semantics of each command when a thread is selected must be
27157 @c described. For example:
27159 @c 'g': If the stub supports threads and a specific thread is
27160 @c selected, returns the register block from that thread;
27161 @c otherwise returns current registers.
27163 @c 'G' If the stub supports threads and a specific thread is
27164 @c selected, sets the registers of the register block of
27165 @c that thread; otherwise sets current registers.
27167 @item i @r{[}@var{addr}@r{[},@var{nnn}@r{]]}
27168 @anchor{cycle step packet}
27169 @cindex @samp{i} packet
27170 Step the remote target by a single clock cycle. If @samp{,@var{nnn}} is
27171 present, cycle step @var{nnn} cycles. If @var{addr} is present, cycle
27172 step starting at that address.
27175 @cindex @samp{I} packet
27176 Signal, then cycle step. @xref{step with signal packet}. @xref{cycle
27180 @cindex @samp{k} packet
27183 FIXME: @emph{There is no description of how to operate when a specific
27184 thread context has been selected (i.e.@: does 'k' kill only that
27187 @item m @var{addr},@var{length}
27188 @cindex @samp{m} packet
27189 Read @var{length} bytes of memory starting at address @var{addr}.
27190 Note that @var{addr} may not be aligned to any particular boundary.
27192 The stub need not use any particular size or alignment when gathering
27193 data from memory for the response; even if @var{addr} is word-aligned
27194 and @var{length} is a multiple of the word size, the stub is free to
27195 use byte accesses, or not. For this reason, this packet may not be
27196 suitable for accessing memory-mapped I/O devices.
27197 @cindex alignment of remote memory accesses
27198 @cindex size of remote memory accesses
27199 @cindex memory, alignment and size of remote accesses
27203 @item @var{XX@dots{}}
27204 Memory contents; each byte is transmitted as a two-digit hexadecimal
27205 number. The reply may contain fewer bytes than requested if the
27206 server was able to read only part of the region of memory.
27211 @item M @var{addr},@var{length}:@var{XX@dots{}}
27212 @cindex @samp{M} packet
27213 Write @var{length} bytes of memory starting at address @var{addr}.
27214 @var{XX@dots{}} is the data; each byte is transmitted as a two-digit
27215 hexadecimal number.
27222 for an error (this includes the case where only part of the data was
27227 @cindex @samp{p} packet
27228 Read the value of register @var{n}; @var{n} is in hex.
27229 @xref{read registers packet}, for a description of how the returned
27230 register value is encoded.
27234 @item @var{XX@dots{}}
27235 the register's value
27239 Indicating an unrecognized @var{query}.
27242 @item P @var{n@dots{}}=@var{r@dots{}}
27243 @anchor{write register packet}
27244 @cindex @samp{P} packet
27245 Write register @var{n@dots{}} with value @var{r@dots{}}. The register
27246 number @var{n} is in hexadecimal, and @var{r@dots{}} contains two hex
27247 digits for each byte in the register (target byte order).
27257 @item q @var{name} @var{params}@dots{}
27258 @itemx Q @var{name} @var{params}@dots{}
27259 @cindex @samp{q} packet
27260 @cindex @samp{Q} packet
27261 General query (@samp{q}) and set (@samp{Q}). These packets are
27262 described fully in @ref{General Query Packets}.
27265 @cindex @samp{r} packet
27266 Reset the entire system.
27268 Don't use this packet; use the @samp{R} packet instead.
27271 @cindex @samp{R} packet
27272 Restart the program being debugged. @var{XX}, while needed, is ignored.
27273 This packet is only available in extended mode (@pxref{extended mode}).
27275 The @samp{R} packet has no reply.
27277 @item s @r{[}@var{addr}@r{]}
27278 @cindex @samp{s} packet
27279 Single step. @var{addr} is the address at which to resume. If
27280 @var{addr} is omitted, resume at same address.
27283 @xref{Stop Reply Packets}, for the reply specifications.
27285 @item S @var{sig}@r{[};@var{addr}@r{]}
27286 @anchor{step with signal packet}
27287 @cindex @samp{S} packet
27288 Step with signal. This is analogous to the @samp{C} packet, but
27289 requests a single-step, rather than a normal resumption of execution.
27292 @xref{Stop Reply Packets}, for the reply specifications.
27294 @item t @var{addr}:@var{PP},@var{MM}
27295 @cindex @samp{t} packet
27296 Search backwards starting at address @var{addr} for a match with pattern
27297 @var{PP} and mask @var{MM}. @var{PP} and @var{MM} are 4 bytes.
27298 @var{addr} must be at least 3 digits.
27300 @item T @var{thread-id}
27301 @cindex @samp{T} packet
27302 Find out if the thread @var{thread-id} is alive. @xref{thread-id syntax}.
27307 thread is still alive
27313 Packets starting with @samp{v} are identified by a multi-letter name,
27314 up to the first @samp{;} or @samp{?} (or the end of the packet).
27316 @item vAttach;@var{pid}
27317 @cindex @samp{vAttach} packet
27318 Attach to a new process with the specified process ID @var{pid}.
27319 The process ID is a
27320 hexadecimal integer identifying the process. In all-stop mode, all
27321 threads in the attached process are stopped; in non-stop mode, it may be
27322 attached without being stopped if that is supported by the target.
27324 @c In non-stop mode, on a successful vAttach, the stub should set the
27325 @c current thread to a thread of the newly-attached process. After
27326 @c attaching, GDB queries for the attached process's thread ID with qC.
27327 @c Also note that, from a user perspective, whether or not the
27328 @c target is stopped on attach in non-stop mode depends on whether you
27329 @c use the foreground or background version of the attach command, not
27330 @c on what vAttach does; GDB does the right thing with respect to either
27331 @c stopping or restarting threads.
27333 This packet is only available in extended mode (@pxref{extended mode}).
27339 @item @r{Any stop packet}
27340 for success in all-stop mode (@pxref{Stop Reply Packets})
27342 for success in non-stop mode (@pxref{Remote Non-Stop})
27345 @item vCont@r{[};@var{action}@r{[}:@var{thread-id}@r{]]}@dots{}
27346 @cindex @samp{vCont} packet
27347 Resume the inferior, specifying different actions for each thread.
27348 If an action is specified with no @var{thread-id}, then it is applied to any
27349 threads that don't have a specific action specified; if no default action is
27350 specified then other threads should remain stopped in all-stop mode and
27351 in their current state in non-stop mode.
27352 Specifying multiple
27353 default actions is an error; specifying no actions is also an error.
27354 Thread IDs are specified using the syntax described in @ref{thread-id syntax}.
27356 Currently supported actions are:
27362 Continue with signal @var{sig}. The signal @var{sig} should be two hex digits.
27366 Step with signal @var{sig}. The signal @var{sig} should be two hex digits.
27370 Stop with signal @var{sig}. The signal @var{sig} should be two hex digits.
27373 The optional argument @var{addr} normally associated with the
27374 @samp{c}, @samp{C}, @samp{s}, and @samp{S} packets is
27375 not supported in @samp{vCont}.
27377 The @samp{t} and @samp{T} actions are only relevant in non-stop mode
27378 (@pxref{Remote Non-Stop}) and may be ignored by the stub otherwise.
27379 A stop reply should be generated for any affected thread not already stopped.
27380 When a thread is stopped by means of a @samp{t} action,
27381 the corresponding stop reply should indicate that the thread has stopped with
27382 signal @samp{0}, regardless of whether the target uses some other signal
27383 as an implementation detail.
27386 @xref{Stop Reply Packets}, for the reply specifications.
27389 @cindex @samp{vCont?} packet
27390 Request a list of actions supported by the @samp{vCont} packet.
27394 @item vCont@r{[};@var{action}@dots{}@r{]}
27395 The @samp{vCont} packet is supported. Each @var{action} is a supported
27396 command in the @samp{vCont} packet.
27398 The @samp{vCont} packet is not supported.
27401 @item vFile:@var{operation}:@var{parameter}@dots{}
27402 @cindex @samp{vFile} packet
27403 Perform a file operation on the target system. For details,
27404 see @ref{Host I/O Packets}.
27406 @item vFlashErase:@var{addr},@var{length}
27407 @cindex @samp{vFlashErase} packet
27408 Direct the stub to erase @var{length} bytes of flash starting at
27409 @var{addr}. The region may enclose any number of flash blocks, but
27410 its start and end must fall on block boundaries, as indicated by the
27411 flash block size appearing in the memory map (@pxref{Memory Map
27412 Format}). @value{GDBN} groups flash memory programming operations
27413 together, and sends a @samp{vFlashDone} request after each group; the
27414 stub is allowed to delay erase operation until the @samp{vFlashDone}
27415 packet is received.
27417 The stub must support @samp{vCont} if it reports support for
27418 multiprocess extensions (@pxref{multiprocess extensions}). Note that in
27419 this case @samp{vCont} actions can be specified to apply to all threads
27420 in a process by using the @samp{p@var{pid}.-1} form of the
27431 @item vFlashWrite:@var{addr}:@var{XX@dots{}}
27432 @cindex @samp{vFlashWrite} packet
27433 Direct the stub to write data to flash address @var{addr}. The data
27434 is passed in binary form using the same encoding as for the @samp{X}
27435 packet (@pxref{Binary Data}). The memory ranges specified by
27436 @samp{vFlashWrite} packets preceding a @samp{vFlashDone} packet must
27437 not overlap, and must appear in order of increasing addresses
27438 (although @samp{vFlashErase} packets for higher addresses may already
27439 have been received; the ordering is guaranteed only between
27440 @samp{vFlashWrite} packets). If a packet writes to an address that was
27441 neither erased by a preceding @samp{vFlashErase} packet nor by some other
27442 target-specific method, the results are unpredictable.
27450 for vFlashWrite addressing non-flash memory
27456 @cindex @samp{vFlashDone} packet
27457 Indicate to the stub that flash programming operation is finished.
27458 The stub is permitted to delay or batch the effects of a group of
27459 @samp{vFlashErase} and @samp{vFlashWrite} packets until a
27460 @samp{vFlashDone} packet is received. The contents of the affected
27461 regions of flash memory are unpredictable until the @samp{vFlashDone}
27462 request is completed.
27464 @item vKill;@var{pid}
27465 @cindex @samp{vKill} packet
27466 Kill the process with the specified process ID. @var{pid} is a
27467 hexadecimal integer identifying the process. This packet is used in
27468 preference to @samp{k} when multiprocess protocol extensions are
27469 supported; see @ref{multiprocess extensions}.
27479 @item vRun;@var{filename}@r{[};@var{argument}@r{]}@dots{}
27480 @cindex @samp{vRun} packet
27481 Run the program @var{filename}, passing it each @var{argument} on its
27482 command line. The file and arguments are hex-encoded strings. If
27483 @var{filename} is an empty string, the stub may use a default program
27484 (e.g.@: the last program run). The program is created in the stopped
27487 @c FIXME: What about non-stop mode?
27489 This packet is only available in extended mode (@pxref{extended mode}).
27495 @item @r{Any stop packet}
27496 for success (@pxref{Stop Reply Packets})
27500 @anchor{vStopped packet}
27501 @cindex @samp{vStopped} packet
27503 In non-stop mode (@pxref{Remote Non-Stop}), acknowledge a previous stop
27504 reply and prompt for the stub to report another one.
27508 @item @r{Any stop packet}
27509 if there is another unreported stop event (@pxref{Stop Reply Packets})
27511 if there are no unreported stop events
27514 @item X @var{addr},@var{length}:@var{XX@dots{}}
27516 @cindex @samp{X} packet
27517 Write data to memory, where the data is transmitted in binary.
27518 @var{addr} is address, @var{length} is number of bytes,
27519 @samp{@var{XX}@dots{}} is binary data (@pxref{Binary Data}).
27529 @item z @var{type},@var{addr},@var{length}
27530 @itemx Z @var{type},@var{addr},@var{length}
27531 @anchor{insert breakpoint or watchpoint packet}
27532 @cindex @samp{z} packet
27533 @cindex @samp{Z} packets
27534 Insert (@samp{Z}) or remove (@samp{z}) a @var{type} breakpoint or
27535 watchpoint starting at address @var{address} and covering the next
27536 @var{length} bytes.
27538 Each breakpoint and watchpoint packet @var{type} is documented
27541 @emph{Implementation notes: A remote target shall return an empty string
27542 for an unrecognized breakpoint or watchpoint packet @var{type}. A
27543 remote target shall support either both or neither of a given
27544 @samp{Z@var{type}@dots{}} and @samp{z@var{type}@dots{}} packet pair. To
27545 avoid potential problems with duplicate packets, the operations should
27546 be implemented in an idempotent way.}
27548 @item z0,@var{addr},@var{length}
27549 @itemx Z0,@var{addr},@var{length}
27550 @cindex @samp{z0} packet
27551 @cindex @samp{Z0} packet
27552 Insert (@samp{Z0}) or remove (@samp{z0}) a memory breakpoint at address
27553 @var{addr} of size @var{length}.
27555 A memory breakpoint is implemented by replacing the instruction at
27556 @var{addr} with a software breakpoint or trap instruction. The
27557 @var{length} is used by targets that indicates the size of the
27558 breakpoint (in bytes) that should be inserted (e.g., the @sc{arm} and
27559 @sc{mips} can insert either a 2 or 4 byte breakpoint).
27561 @emph{Implementation note: It is possible for a target to copy or move
27562 code that contains memory breakpoints (e.g., when implementing
27563 overlays). The behavior of this packet, in the presence of such a
27564 target, is not defined.}
27576 @item z1,@var{addr},@var{length}
27577 @itemx Z1,@var{addr},@var{length}
27578 @cindex @samp{z1} packet
27579 @cindex @samp{Z1} packet
27580 Insert (@samp{Z1}) or remove (@samp{z1}) a hardware breakpoint at
27581 address @var{addr} of size @var{length}.
27583 A hardware breakpoint is implemented using a mechanism that is not
27584 dependant on being able to modify the target's memory.
27586 @emph{Implementation note: A hardware breakpoint is not affected by code
27599 @item z2,@var{addr},@var{length}
27600 @itemx Z2,@var{addr},@var{length}
27601 @cindex @samp{z2} packet
27602 @cindex @samp{Z2} packet
27603 Insert (@samp{Z2}) or remove (@samp{z2}) a write watchpoint.
27615 @item z3,@var{addr},@var{length}
27616 @itemx Z3,@var{addr},@var{length}
27617 @cindex @samp{z3} packet
27618 @cindex @samp{Z3} packet
27619 Insert (@samp{Z3}) or remove (@samp{z3}) a read watchpoint.
27631 @item z4,@var{addr},@var{length}
27632 @itemx Z4,@var{addr},@var{length}
27633 @cindex @samp{z4} packet
27634 @cindex @samp{Z4} packet
27635 Insert (@samp{Z4}) or remove (@samp{z4}) an access watchpoint.
27649 @node Stop Reply Packets
27650 @section Stop Reply Packets
27651 @cindex stop reply packets
27653 The @samp{C}, @samp{c}, @samp{S}, @samp{s}, @samp{vCont},
27654 @samp{vAttach}, @samp{vRun}, @samp{vStopped}, and @samp{?} packets can
27655 receive any of the below as a reply. Except for @samp{?}
27656 and @samp{vStopped}, that reply is only returned
27657 when the target halts. In the below the exact meaning of @dfn{signal
27658 number} is defined by the header @file{include/gdb/signals.h} in the
27659 @value{GDBN} source code.
27661 As in the description of request packets, we include spaces in the
27662 reply templates for clarity; these are not part of the reply packet's
27663 syntax. No @value{GDBN} stop reply packet uses spaces to separate its
27669 The program received signal number @var{AA} (a two-digit hexadecimal
27670 number). This is equivalent to a @samp{T} response with no
27671 @var{n}:@var{r} pairs.
27673 @item T @var{AA} @var{n1}:@var{r1};@var{n2}:@var{r2};@dots{}
27674 @cindex @samp{T} packet reply
27675 The program received signal number @var{AA} (a two-digit hexadecimal
27676 number). This is equivalent to an @samp{S} response, except that the
27677 @samp{@var{n}:@var{r}} pairs can carry values of important registers
27678 and other information directly in the stop reply packet, reducing
27679 round-trip latency. Single-step and breakpoint traps are reported
27680 this way. Each @samp{@var{n}:@var{r}} pair is interpreted as follows:
27684 If @var{n} is a hexadecimal number, it is a register number, and the
27685 corresponding @var{r} gives that register's value. @var{r} is a
27686 series of bytes in target byte order, with each byte given by a
27687 two-digit hex number.
27690 If @var{n} is @samp{thread}, then @var{r} is the @var{thread-id} of
27691 the stopped thread, as specified in @ref{thread-id syntax}.
27694 If @var{n} is a recognized @dfn{stop reason}, it describes a more
27695 specific event that stopped the target. The currently defined stop
27696 reasons are listed below. @var{aa} should be @samp{05}, the trap
27697 signal. At most one stop reason should be present.
27700 Otherwise, @value{GDBN} should ignore this @samp{@var{n}:@var{r}} pair
27701 and go on to the next; this allows us to extend the protocol in the
27705 The currently defined stop reasons are:
27711 The packet indicates a watchpoint hit, and @var{r} is the data address, in
27714 @cindex shared library events, remote reply
27716 The packet indicates that the loaded libraries have changed.
27717 @value{GDBN} should use @samp{qXfer:libraries:read} to fetch a new
27718 list of loaded libraries. @var{r} is ignored.
27720 @cindex replay log events, remote reply
27722 The packet indicates that the target cannot continue replaying
27723 logged execution events, because it has reached the end (or the
27724 beginning when executing backward) of the log. The value of @var{r}
27725 will be either @samp{begin} or @samp{end}. @xref{Reverse Execution},
27726 for more information.
27732 @itemx W @var{AA} ; process:@var{pid}
27733 The process exited, and @var{AA} is the exit status. This is only
27734 applicable to certain targets.
27736 The second form of the response, including the process ID of the exited
27737 process, can be used only when @value{GDBN} has reported support for
27738 multiprocess protocol extensions; see @ref{multiprocess extensions}.
27739 The @var{pid} is formatted as a big-endian hex string.
27742 @itemx X @var{AA} ; process:@var{pid}
27743 The process terminated with signal @var{AA}.
27745 The second form of the response, including the process ID of the
27746 terminated process, can be used only when @value{GDBN} has reported
27747 support for multiprocess protocol extensions; see @ref{multiprocess
27748 extensions}. The @var{pid} is formatted as a big-endian hex string.
27750 @item O @var{XX}@dots{}
27751 @samp{@var{XX}@dots{}} is hex encoding of @sc{ascii} data, to be
27752 written as the program's console output. This can happen at any time
27753 while the program is running and the debugger should continue to wait
27754 for @samp{W}, @samp{T}, etc. This reply is not permitted in non-stop mode.
27756 @item F @var{call-id},@var{parameter}@dots{}
27757 @var{call-id} is the identifier which says which host system call should
27758 be called. This is just the name of the function. Translation into the
27759 correct system call is only applicable as it's defined in @value{GDBN}.
27760 @xref{File-I/O Remote Protocol Extension}, for a list of implemented
27763 @samp{@var{parameter}@dots{}} is a list of parameters as defined for
27764 this very system call.
27766 The target replies with this packet when it expects @value{GDBN} to
27767 call a host system call on behalf of the target. @value{GDBN} replies
27768 with an appropriate @samp{F} packet and keeps up waiting for the next
27769 reply packet from the target. The latest @samp{C}, @samp{c}, @samp{S}
27770 or @samp{s} action is expected to be continued. @xref{File-I/O Remote
27771 Protocol Extension}, for more details.
27775 @node General Query Packets
27776 @section General Query Packets
27777 @cindex remote query requests
27779 Packets starting with @samp{q} are @dfn{general query packets};
27780 packets starting with @samp{Q} are @dfn{general set packets}. General
27781 query and set packets are a semi-unified form for retrieving and
27782 sending information to and from the stub.
27784 The initial letter of a query or set packet is followed by a name
27785 indicating what sort of thing the packet applies to. For example,
27786 @value{GDBN} may use a @samp{qSymbol} packet to exchange symbol
27787 definitions with the stub. These packet names follow some
27792 The name must not contain commas, colons or semicolons.
27794 Most @value{GDBN} query and set packets have a leading upper case
27797 The names of custom vendor packets should use a company prefix, in
27798 lower case, followed by a period. For example, packets designed at
27799 the Acme Corporation might begin with @samp{qacme.foo} (for querying
27800 foos) or @samp{Qacme.bar} (for setting bars).
27803 The name of a query or set packet should be separated from any
27804 parameters by a @samp{:}; the parameters themselves should be
27805 separated by @samp{,} or @samp{;}. Stubs must be careful to match the
27806 full packet name, and check for a separator or the end of the packet,
27807 in case two packet names share a common prefix. New packets should not begin
27808 with @samp{qC}, @samp{qP}, or @samp{qL}@footnote{The @samp{qP} and @samp{qL}
27809 packets predate these conventions, and have arguments without any terminator
27810 for the packet name; we suspect they are in widespread use in places that
27811 are difficult to upgrade. The @samp{qC} packet has no arguments, but some
27812 existing stubs (e.g.@: RedBoot) are known to not check for the end of the
27815 Like the descriptions of the other packets, each description here
27816 has a template showing the packet's overall syntax, followed by an
27817 explanation of the packet's meaning. We include spaces in some of the
27818 templates for clarity; these are not part of the packet's syntax. No
27819 @value{GDBN} packet uses spaces to separate its components.
27821 Here are the currently defined query and set packets:
27826 @cindex current thread, remote request
27827 @cindex @samp{qC} packet
27828 Return the current thread ID.
27832 @item QC @var{thread-id}
27833 Where @var{thread-id} is a thread ID as documented in
27834 @ref{thread-id syntax}.
27835 @item @r{(anything else)}
27836 Any other reply implies the old thread ID.
27839 @item qCRC:@var{addr},@var{length}
27840 @cindex CRC of memory block, remote request
27841 @cindex @samp{qCRC} packet
27842 Compute the CRC checksum of a block of memory.
27846 An error (such as memory fault)
27847 @item C @var{crc32}
27848 The specified memory region's checksum is @var{crc32}.
27852 @itemx qsThreadInfo
27853 @cindex list active threads, remote request
27854 @cindex @samp{qfThreadInfo} packet
27855 @cindex @samp{qsThreadInfo} packet
27856 Obtain a list of all active thread IDs from the target (OS). Since there
27857 may be too many active threads to fit into one reply packet, this query
27858 works iteratively: it may require more than one query/reply sequence to
27859 obtain the entire list of threads. The first query of the sequence will
27860 be the @samp{qfThreadInfo} query; subsequent queries in the
27861 sequence will be the @samp{qsThreadInfo} query.
27863 NOTE: This packet replaces the @samp{qL} query (see below).
27867 @item m @var{thread-id}
27869 @item m @var{thread-id},@var{thread-id}@dots{}
27870 a comma-separated list of thread IDs
27872 (lower case letter @samp{L}) denotes end of list.
27875 In response to each query, the target will reply with a list of one or
27876 more thread IDs, separated by commas.
27877 @value{GDBN} will respond to each reply with a request for more thread
27878 ids (using the @samp{qs} form of the query), until the target responds
27879 with @samp{l} (lower-case el, for @dfn{last}).
27880 Refer to @ref{thread-id syntax}, for the format of the @var{thread-id}
27883 @item qGetTLSAddr:@var{thread-id},@var{offset},@var{lm}
27884 @cindex get thread-local storage address, remote request
27885 @cindex @samp{qGetTLSAddr} packet
27886 Fetch the address associated with thread local storage specified
27887 by @var{thread-id}, @var{offset}, and @var{lm}.
27889 @var{thread-id} is the thread ID associated with the
27890 thread for which to fetch the TLS address. @xref{thread-id syntax}.
27892 @var{offset} is the (big endian, hex encoded) offset associated with the
27893 thread local variable. (This offset is obtained from the debug
27894 information associated with the variable.)
27896 @var{lm} is the (big endian, hex encoded) OS/ABI-specific encoding of the
27897 the load module associated with the thread local storage. For example,
27898 a @sc{gnu}/Linux system will pass the link map address of the shared
27899 object associated with the thread local storage under consideration.
27900 Other operating environments may choose to represent the load module
27901 differently, so the precise meaning of this parameter will vary.
27905 @item @var{XX}@dots{}
27906 Hex encoded (big endian) bytes representing the address of the thread
27907 local storage requested.
27910 An error occurred. @var{nn} are hex digits.
27913 An empty reply indicates that @samp{qGetTLSAddr} is not supported by the stub.
27916 @item qL @var{startflag} @var{threadcount} @var{nextthread}
27917 Obtain thread information from RTOS. Where: @var{startflag} (one hex
27918 digit) is one to indicate the first query and zero to indicate a
27919 subsequent query; @var{threadcount} (two hex digits) is the maximum
27920 number of threads the response packet can contain; and @var{nextthread}
27921 (eight hex digits), for subsequent queries (@var{startflag} is zero), is
27922 returned in the response as @var{argthread}.
27924 Don't use this packet; use the @samp{qfThreadInfo} query instead (see above).
27928 @item qM @var{count} @var{done} @var{argthread} @var{thread}@dots{}
27929 Where: @var{count} (two hex digits) is the number of threads being
27930 returned; @var{done} (one hex digit) is zero to indicate more threads
27931 and one indicates no further threads; @var{argthreadid} (eight hex
27932 digits) is @var{nextthread} from the request packet; @var{thread}@dots{}
27933 is a sequence of thread IDs from the target. @var{threadid} (eight hex
27934 digits). See @code{remote.c:parse_threadlist_response()}.
27938 @cindex section offsets, remote request
27939 @cindex @samp{qOffsets} packet
27940 Get section offsets that the target used when relocating the downloaded
27945 @item Text=@var{xxx};Data=@var{yyy}@r{[};Bss=@var{zzz}@r{]}
27946 Relocate the @code{Text} section by @var{xxx} from its original address.
27947 Relocate the @code{Data} section by @var{yyy} from its original address.
27948 If the object file format provides segment information (e.g.@: @sc{elf}
27949 @samp{PT_LOAD} program headers), @value{GDBN} will relocate entire
27950 segments by the supplied offsets.
27952 @emph{Note: while a @code{Bss} offset may be included in the response,
27953 @value{GDBN} ignores this and instead applies the @code{Data} offset
27954 to the @code{Bss} section.}
27956 @item TextSeg=@var{xxx}@r{[};DataSeg=@var{yyy}@r{]}
27957 Relocate the first segment of the object file, which conventionally
27958 contains program code, to a starting address of @var{xxx}. If
27959 @samp{DataSeg} is specified, relocate the second segment, which
27960 conventionally contains modifiable data, to a starting address of
27961 @var{yyy}. @value{GDBN} will report an error if the object file
27962 does not contain segment information, or does not contain at least
27963 as many segments as mentioned in the reply. Extra segments are
27964 kept at fixed offsets relative to the last relocated segment.
27967 @item qP @var{mode} @var{thread-id}
27968 @cindex thread information, remote request
27969 @cindex @samp{qP} packet
27970 Returns information on @var{thread-id}. Where: @var{mode} is a hex
27971 encoded 32 bit mode; @var{thread-id} is a thread ID
27972 (@pxref{thread-id syntax}).
27974 Don't use this packet; use the @samp{qThreadExtraInfo} query instead
27977 Reply: see @code{remote.c:remote_unpack_thread_info_response()}.
27981 @cindex non-stop mode, remote request
27982 @cindex @samp{QNonStop} packet
27984 Enter non-stop (@samp{QNonStop:1}) or all-stop (@samp{QNonStop:0}) mode.
27985 @xref{Remote Non-Stop}, for more information.
27990 The request succeeded.
27993 An error occurred. @var{nn} are hex digits.
27996 An empty reply indicates that @samp{QNonStop} is not supported by
28000 This packet is not probed by default; the remote stub must request it,
28001 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
28002 Use of this packet is controlled by the @code{set non-stop} command;
28003 @pxref{Non-Stop Mode}.
28005 @item QPassSignals: @var{signal} @r{[};@var{signal}@r{]}@dots{}
28006 @cindex pass signals to inferior, remote request
28007 @cindex @samp{QPassSignals} packet
28008 @anchor{QPassSignals}
28009 Each listed @var{signal} should be passed directly to the inferior process.
28010 Signals are numbered identically to continue packets and stop replies
28011 (@pxref{Stop Reply Packets}). Each @var{signal} list item should be
28012 strictly greater than the previous item. These signals do not need to stop
28013 the inferior, or be reported to @value{GDBN}. All other signals should be
28014 reported to @value{GDBN}. Multiple @samp{QPassSignals} packets do not
28015 combine; any earlier @samp{QPassSignals} list is completely replaced by the
28016 new list. This packet improves performance when using @samp{handle
28017 @var{signal} nostop noprint pass}.
28022 The request succeeded.
28025 An error occurred. @var{nn} are hex digits.
28028 An empty reply indicates that @samp{QPassSignals} is not supported by
28032 Use of this packet is controlled by the @code{set remote pass-signals}
28033 command (@pxref{Remote Configuration, set remote pass-signals}).
28034 This packet is not probed by default; the remote stub must request it,
28035 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
28037 @item qRcmd,@var{command}
28038 @cindex execute remote command, remote request
28039 @cindex @samp{qRcmd} packet
28040 @var{command} (hex encoded) is passed to the local interpreter for
28041 execution. Invalid commands should be reported using the output
28042 string. Before the final result packet, the target may also respond
28043 with a number of intermediate @samp{O@var{output}} console output
28044 packets. @emph{Implementors should note that providing access to a
28045 stubs's interpreter may have security implications}.
28050 A command response with no output.
28052 A command response with the hex encoded output string @var{OUTPUT}.
28054 Indicate a badly formed request.
28056 An empty reply indicates that @samp{qRcmd} is not recognized.
28059 (Note that the @code{qRcmd} packet's name is separated from the
28060 command by a @samp{,}, not a @samp{:}, contrary to the naming
28061 conventions above. Please don't use this packet as a model for new
28064 @item qSearch:memory:@var{address};@var{length};@var{search-pattern}
28065 @cindex searching memory, in remote debugging
28066 @cindex @samp{qSearch:memory} packet
28067 @anchor{qSearch memory}
28068 Search @var{length} bytes at @var{address} for @var{search-pattern}.
28069 @var{address} and @var{length} are encoded in hex.
28070 @var{search-pattern} is a sequence of bytes, hex encoded.
28075 The pattern was not found.
28077 The pattern was found at @var{address}.
28079 A badly formed request or an error was encountered while searching memory.
28081 An empty reply indicates that @samp{qSearch:memory} is not recognized.
28084 @item QStartNoAckMode
28085 @cindex @samp{QStartNoAckMode} packet
28086 @anchor{QStartNoAckMode}
28087 Request that the remote stub disable the normal @samp{+}/@samp{-}
28088 protocol acknowledgments (@pxref{Packet Acknowledgment}).
28093 The stub has switched to no-acknowledgment mode.
28094 @value{GDBN} acknowledges this reponse,
28095 but neither the stub nor @value{GDBN} shall send or expect further
28096 @samp{+}/@samp{-} acknowledgments in the current connection.
28098 An empty reply indicates that the stub does not support no-acknowledgment mode.
28101 @item qSupported @r{[}:@var{gdbfeature} @r{[};@var{gdbfeature}@r{]}@dots{} @r{]}
28102 @cindex supported packets, remote query
28103 @cindex features of the remote protocol
28104 @cindex @samp{qSupported} packet
28105 @anchor{qSupported}
28106 Tell the remote stub about features supported by @value{GDBN}, and
28107 query the stub for features it supports. This packet allows
28108 @value{GDBN} and the remote stub to take advantage of each others'
28109 features. @samp{qSupported} also consolidates multiple feature probes
28110 at startup, to improve @value{GDBN} performance---a single larger
28111 packet performs better than multiple smaller probe packets on
28112 high-latency links. Some features may enable behavior which must not
28113 be on by default, e.g.@: because it would confuse older clients or
28114 stubs. Other features may describe packets which could be
28115 automatically probed for, but are not. These features must be
28116 reported before @value{GDBN} will use them. This ``default
28117 unsupported'' behavior is not appropriate for all packets, but it
28118 helps to keep the initial connection time under control with new
28119 versions of @value{GDBN} which support increasing numbers of packets.
28123 @item @var{stubfeature} @r{[};@var{stubfeature}@r{]}@dots{}
28124 The stub supports or does not support each returned @var{stubfeature},
28125 depending on the form of each @var{stubfeature} (see below for the
28128 An empty reply indicates that @samp{qSupported} is not recognized,
28129 or that no features needed to be reported to @value{GDBN}.
28132 The allowed forms for each feature (either a @var{gdbfeature} in the
28133 @samp{qSupported} packet, or a @var{stubfeature} in the response)
28137 @item @var{name}=@var{value}
28138 The remote protocol feature @var{name} is supported, and associated
28139 with the specified @var{value}. The format of @var{value} depends
28140 on the feature, but it must not include a semicolon.
28142 The remote protocol feature @var{name} is supported, and does not
28143 need an associated value.
28145 The remote protocol feature @var{name} is not supported.
28147 The remote protocol feature @var{name} may be supported, and
28148 @value{GDBN} should auto-detect support in some other way when it is
28149 needed. This form will not be used for @var{gdbfeature} notifications,
28150 but may be used for @var{stubfeature} responses.
28153 Whenever the stub receives a @samp{qSupported} request, the
28154 supplied set of @value{GDBN} features should override any previous
28155 request. This allows @value{GDBN} to put the stub in a known
28156 state, even if the stub had previously been communicating with
28157 a different version of @value{GDBN}.
28159 The following values of @var{gdbfeature} (for the packet sent by @value{GDBN})
28164 This feature indicates whether @value{GDBN} supports multiprocess
28165 extensions to the remote protocol. @value{GDBN} does not use such
28166 extensions unless the stub also reports that it supports them by
28167 including @samp{multiprocess+} in its @samp{qSupported} reply.
28168 @xref{multiprocess extensions}, for details.
28171 Stubs should ignore any unknown values for
28172 @var{gdbfeature}. Any @value{GDBN} which sends a @samp{qSupported}
28173 packet supports receiving packets of unlimited length (earlier
28174 versions of @value{GDBN} may reject overly long responses). Additional values
28175 for @var{gdbfeature} may be defined in the future to let the stub take
28176 advantage of new features in @value{GDBN}, e.g.@: incompatible
28177 improvements in the remote protocol---the @samp{multiprocess} feature is
28178 an example of such a feature. The stub's reply should be independent
28179 of the @var{gdbfeature} entries sent by @value{GDBN}; first @value{GDBN}
28180 describes all the features it supports, and then the stub replies with
28181 all the features it supports.
28183 Similarly, @value{GDBN} will silently ignore unrecognized stub feature
28184 responses, as long as each response uses one of the standard forms.
28186 Some features are flags. A stub which supports a flag feature
28187 should respond with a @samp{+} form response. Other features
28188 require values, and the stub should respond with an @samp{=}
28191 Each feature has a default value, which @value{GDBN} will use if
28192 @samp{qSupported} is not available or if the feature is not mentioned
28193 in the @samp{qSupported} response. The default values are fixed; a
28194 stub is free to omit any feature responses that match the defaults.
28196 Not all features can be probed, but for those which can, the probing
28197 mechanism is useful: in some cases, a stub's internal
28198 architecture may not allow the protocol layer to know some information
28199 about the underlying target in advance. This is especially common in
28200 stubs which may be configured for multiple targets.
28202 These are the currently defined stub features and their properties:
28204 @multitable @columnfractions 0.35 0.2 0.12 0.2
28205 @c NOTE: The first row should be @headitem, but we do not yet require
28206 @c a new enough version of Texinfo (4.7) to use @headitem.
28208 @tab Value Required
28212 @item @samp{PacketSize}
28217 @item @samp{qXfer:auxv:read}
28222 @item @samp{qXfer:features:read}
28227 @item @samp{qXfer:libraries:read}
28232 @item @samp{qXfer:memory-map:read}
28237 @item @samp{qXfer:spu:read}
28242 @item @samp{qXfer:spu:write}
28247 @item @samp{qXfer:siginfo:read}
28252 @item @samp{qXfer:siginfo:write}
28257 @item @samp{QNonStop}
28262 @item @samp{QPassSignals}
28267 @item @samp{QStartNoAckMode}
28272 @item @samp{multiprocess}
28279 These are the currently defined stub features, in more detail:
28282 @cindex packet size, remote protocol
28283 @item PacketSize=@var{bytes}
28284 The remote stub can accept packets up to at least @var{bytes} in
28285 length. @value{GDBN} will send packets up to this size for bulk
28286 transfers, and will never send larger packets. This is a limit on the
28287 data characters in the packet, including the frame and checksum.
28288 There is no trailing NUL byte in a remote protocol packet; if the stub
28289 stores packets in a NUL-terminated format, it should allow an extra
28290 byte in its buffer for the NUL. If this stub feature is not supported,
28291 @value{GDBN} guesses based on the size of the @samp{g} packet response.
28293 @item qXfer:auxv:read
28294 The remote stub understands the @samp{qXfer:auxv:read} packet
28295 (@pxref{qXfer auxiliary vector read}).
28297 @item qXfer:features:read
28298 The remote stub understands the @samp{qXfer:features:read} packet
28299 (@pxref{qXfer target description read}).
28301 @item qXfer:libraries:read
28302 The remote stub understands the @samp{qXfer:libraries:read} packet
28303 (@pxref{qXfer library list read}).
28305 @item qXfer:memory-map:read
28306 The remote stub understands the @samp{qXfer:memory-map:read} packet
28307 (@pxref{qXfer memory map read}).
28309 @item qXfer:spu:read
28310 The remote stub understands the @samp{qXfer:spu:read} packet
28311 (@pxref{qXfer spu read}).
28313 @item qXfer:spu:write
28314 The remote stub understands the @samp{qXfer:spu:write} packet
28315 (@pxref{qXfer spu write}).
28317 @item qXfer:siginfo:read
28318 The remote stub understands the @samp{qXfer:siginfo:read} packet
28319 (@pxref{qXfer siginfo read}).
28321 @item qXfer:siginfo:write
28322 The remote stub understands the @samp{qXfer:siginfo:write} packet
28323 (@pxref{qXfer siginfo write}).
28326 The remote stub understands the @samp{QNonStop} packet
28327 (@pxref{QNonStop}).
28330 The remote stub understands the @samp{QPassSignals} packet
28331 (@pxref{QPassSignals}).
28333 @item QStartNoAckMode
28334 The remote stub understands the @samp{QStartNoAckMode} packet and
28335 prefers to operate in no-acknowledgment mode. @xref{Packet Acknowledgment}.
28338 @anchor{multiprocess extensions}
28339 @cindex multiprocess extensions, in remote protocol
28340 The remote stub understands the multiprocess extensions to the remote
28341 protocol syntax. The multiprocess extensions affect the syntax of
28342 thread IDs in both packets and replies (@pxref{thread-id syntax}), and
28343 add process IDs to the @samp{D} packet and @samp{W} and @samp{X}
28344 replies. Note that reporting this feature indicates support for the
28345 syntactic extensions only, not that the stub necessarily supports
28346 debugging of more than one process at a time. The stub must not use
28347 multiprocess extensions in packet replies unless @value{GDBN} has also
28348 indicated it supports them in its @samp{qSupported} request.
28350 @item qXfer:osdata:read
28351 The remote stub understands the @samp{qXfer:osdata:read} packet
28352 ((@pxref{qXfer osdata read}).
28357 @cindex symbol lookup, remote request
28358 @cindex @samp{qSymbol} packet
28359 Notify the target that @value{GDBN} is prepared to serve symbol lookup
28360 requests. Accept requests from the target for the values of symbols.
28365 The target does not need to look up any (more) symbols.
28366 @item qSymbol:@var{sym_name}
28367 The target requests the value of symbol @var{sym_name} (hex encoded).
28368 @value{GDBN} may provide the value by using the
28369 @samp{qSymbol:@var{sym_value}:@var{sym_name}} message, described
28373 @item qSymbol:@var{sym_value}:@var{sym_name}
28374 Set the value of @var{sym_name} to @var{sym_value}.
28376 @var{sym_name} (hex encoded) is the name of a symbol whose value the
28377 target has previously requested.
28379 @var{sym_value} (hex) is the value for symbol @var{sym_name}. If
28380 @value{GDBN} cannot supply a value for @var{sym_name}, then this field
28386 The target does not need to look up any (more) symbols.
28387 @item qSymbol:@var{sym_name}
28388 The target requests the value of a new symbol @var{sym_name} (hex
28389 encoded). @value{GDBN} will continue to supply the values of symbols
28390 (if available), until the target ceases to request them.
28395 @xref{Tracepoint Packets}.
28397 @item qThreadExtraInfo,@var{thread-id}
28398 @cindex thread attributes info, remote request
28399 @cindex @samp{qThreadExtraInfo} packet
28400 Obtain a printable string description of a thread's attributes from
28401 the target OS. @var{thread-id} is a thread ID;
28402 see @ref{thread-id syntax}. This
28403 string may contain anything that the target OS thinks is interesting
28404 for @value{GDBN} to tell the user about the thread. The string is
28405 displayed in @value{GDBN}'s @code{info threads} display. Some
28406 examples of possible thread extra info strings are @samp{Runnable}, or
28407 @samp{Blocked on Mutex}.
28411 @item @var{XX}@dots{}
28412 Where @samp{@var{XX}@dots{}} is a hex encoding of @sc{ascii} data,
28413 comprising the printable string containing the extra information about
28414 the thread's attributes.
28417 (Note that the @code{qThreadExtraInfo} packet's name is separated from
28418 the command by a @samp{,}, not a @samp{:}, contrary to the naming
28419 conventions above. Please don't use this packet as a model for new
28427 @xref{Tracepoint Packets}.
28429 @item qXfer:@var{object}:read:@var{annex}:@var{offset},@var{length}
28430 @cindex read special object, remote request
28431 @cindex @samp{qXfer} packet
28432 @anchor{qXfer read}
28433 Read uninterpreted bytes from the target's special data area
28434 identified by the keyword @var{object}. Request @var{length} bytes
28435 starting at @var{offset} bytes into the data. The content and
28436 encoding of @var{annex} is specific to @var{object}; it can supply
28437 additional details about what data to access.
28439 Here are the specific requests of this form defined so far. All
28440 @samp{qXfer:@var{object}:read:@dots{}} requests use the same reply
28441 formats, listed below.
28444 @item qXfer:auxv:read::@var{offset},@var{length}
28445 @anchor{qXfer auxiliary vector read}
28446 Access the target's @dfn{auxiliary vector}. @xref{OS Information,
28447 auxiliary vector}. Note @var{annex} must be empty.
28449 This packet is not probed by default; the remote stub must request it,
28450 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
28452 @item qXfer:features:read:@var{annex}:@var{offset},@var{length}
28453 @anchor{qXfer target description read}
28454 Access the @dfn{target description}. @xref{Target Descriptions}. The
28455 annex specifies which XML document to access. The main description is
28456 always loaded from the @samp{target.xml} annex.
28458 This packet is not probed by default; the remote stub must request it,
28459 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
28461 @item qXfer:libraries:read:@var{annex}:@var{offset},@var{length}
28462 @anchor{qXfer library list read}
28463 Access the target's list of loaded libraries. @xref{Library List Format}.
28464 The annex part of the generic @samp{qXfer} packet must be empty
28465 (@pxref{qXfer read}).
28467 Targets which maintain a list of libraries in the program's memory do
28468 not need to implement this packet; it is designed for platforms where
28469 the operating system manages the list of loaded libraries.
28471 This packet is not probed by default; the remote stub must request it,
28472 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
28474 @item qXfer:memory-map:read::@var{offset},@var{length}
28475 @anchor{qXfer memory map read}
28476 Access the target's @dfn{memory-map}. @xref{Memory Map Format}. The
28477 annex part of the generic @samp{qXfer} packet must be empty
28478 (@pxref{qXfer read}).
28480 This packet is not probed by default; the remote stub must request it,
28481 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
28483 @item qXfer:siginfo:read::@var{offset},@var{length}
28484 @anchor{qXfer siginfo read}
28485 Read contents of the extra signal information on the target
28486 system. The annex part of the generic @samp{qXfer} packet must be
28487 empty (@pxref{qXfer read}).
28489 This packet is not probed by default; the remote stub must request it,
28490 by supplying an appropriate @samp{qSupported} response
28491 (@pxref{qSupported}).
28493 @item qXfer:spu:read:@var{annex}:@var{offset},@var{length}
28494 @anchor{qXfer spu read}
28495 Read contents of an @code{spufs} file on the target system. The
28496 annex specifies which file to read; it must be of the form
28497 @file{@var{id}/@var{name}}, where @var{id} specifies an SPU context ID
28498 in the target process, and @var{name} identifes the @code{spufs} file
28499 in that context to be accessed.
28501 This packet is not probed by default; the remote stub must request it,
28502 by supplying an appropriate @samp{qSupported} response
28503 (@pxref{qSupported}).
28505 @item qXfer:osdata:read::@var{offset},@var{length}
28506 @anchor{qXfer osdata read}
28507 Access the target's @dfn{operating system information}.
28508 @xref{Operating System Information}.
28515 Data @var{data} (@pxref{Binary Data}) has been read from the
28516 target. There may be more data at a higher address (although
28517 it is permitted to return @samp{m} even for the last valid
28518 block of data, as long as at least one byte of data was read).
28519 @var{data} may have fewer bytes than the @var{length} in the
28523 Data @var{data} (@pxref{Binary Data}) has been read from the target.
28524 There is no more data to be read. @var{data} may have fewer bytes
28525 than the @var{length} in the request.
28528 The @var{offset} in the request is at the end of the data.
28529 There is no more data to be read.
28532 The request was malformed, or @var{annex} was invalid.
28535 The offset was invalid, or there was an error encountered reading the data.
28536 @var{nn} is a hex-encoded @code{errno} value.
28539 An empty reply indicates the @var{object} string was not recognized by
28540 the stub, or that the object does not support reading.
28543 @item qXfer:@var{object}:write:@var{annex}:@var{offset}:@var{data}@dots{}
28544 @cindex write data into object, remote request
28545 @anchor{qXfer write}
28546 Write uninterpreted bytes into the target's special data area
28547 identified by the keyword @var{object}, starting at @var{offset} bytes
28548 into the data. @var{data}@dots{} is the binary-encoded data
28549 (@pxref{Binary Data}) to be written. The content and encoding of @var{annex}
28550 is specific to @var{object}; it can supply additional details about what data
28553 Here are the specific requests of this form defined so far. All
28554 @samp{qXfer:@var{object}:write:@dots{}} requests use the same reply
28555 formats, listed below.
28558 @item qXfer:siginfo:write::@var{offset}:@var{data}@dots{}
28559 @anchor{qXfer siginfo write}
28560 Write @var{data} to the extra signal information on the target system.
28561 The annex part of the generic @samp{qXfer} packet must be
28562 empty (@pxref{qXfer write}).
28564 This packet is not probed by default; the remote stub must request it,
28565 by supplying an appropriate @samp{qSupported} response
28566 (@pxref{qSupported}).
28568 @item qXfer:spu:write:@var{annex}:@var{offset}:@var{data}@dots{}
28569 @anchor{qXfer spu write}
28570 Write @var{data} to an @code{spufs} file on the target system. The
28571 annex specifies which file to write; it must be of the form
28572 @file{@var{id}/@var{name}}, where @var{id} specifies an SPU context ID
28573 in the target process, and @var{name} identifes the @code{spufs} file
28574 in that context to be accessed.
28576 This packet is not probed by default; the remote stub must request it,
28577 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
28583 @var{nn} (hex encoded) is the number of bytes written.
28584 This may be fewer bytes than supplied in the request.
28587 The request was malformed, or @var{annex} was invalid.
28590 The offset was invalid, or there was an error encountered writing the data.
28591 @var{nn} is a hex-encoded @code{errno} value.
28594 An empty reply indicates the @var{object} string was not
28595 recognized by the stub, or that the object does not support writing.
28598 @item qXfer:@var{object}:@var{operation}:@dots{}
28599 Requests of this form may be added in the future. When a stub does
28600 not recognize the @var{object} keyword, or its support for
28601 @var{object} does not recognize the @var{operation} keyword, the stub
28602 must respond with an empty packet.
28604 @item qAttached:@var{pid}
28605 @cindex query attached, remote request
28606 @cindex @samp{qAttached} packet
28607 Return an indication of whether the remote server attached to an
28608 existing process or created a new process. When the multiprocess
28609 protocol extensions are supported (@pxref{multiprocess extensions}),
28610 @var{pid} is an integer in hexadecimal format identifying the target
28611 process. Otherwise, @value{GDBN} will omit the @var{pid} field and
28612 the query packet will be simplified as @samp{qAttached}.
28614 This query is used, for example, to know whether the remote process
28615 should be detached or killed when a @value{GDBN} session is ended with
28616 the @code{quit} command.
28621 The remote server attached to an existing process.
28623 The remote server created a new process.
28625 A badly formed request or an error was encountered.
28630 @node Register Packet Format
28631 @section Register Packet Format
28633 The following @code{g}/@code{G} packets have previously been defined.
28634 In the below, some thirty-two bit registers are transferred as
28635 sixty-four bits. Those registers should be zero/sign extended (which?)
28636 to fill the space allocated. Register bytes are transferred in target
28637 byte order. The two nibbles within a register byte are transferred
28638 most-significant - least-significant.
28644 All registers are transferred as thirty-two bit quantities in the order:
28645 32 general-purpose; sr; lo; hi; bad; cause; pc; 32 floating-point
28646 registers; fsr; fir; fp.
28650 All registers are transferred as sixty-four bit quantities (including
28651 thirty-two bit registers such as @code{sr}). The ordering is the same
28656 @node Tracepoint Packets
28657 @section Tracepoint Packets
28658 @cindex tracepoint packets
28659 @cindex packets, tracepoint
28661 Here we describe the packets @value{GDBN} uses to implement
28662 tracepoints (@pxref{Tracepoints}).
28666 @item QTDP:@var{n}:@var{addr}:@var{ena}:@var{step}:@var{pass}@r{[}-@r{]}
28667 Create a new tracepoint, number @var{n}, at @var{addr}. If @var{ena}
28668 is @samp{E}, then the tracepoint is enabled; if it is @samp{D}, then
28669 the tracepoint is disabled. @var{step} is the tracepoint's step
28670 count, and @var{pass} is its pass count. If the trailing @samp{-} is
28671 present, further @samp{QTDP} packets will follow to specify this
28672 tracepoint's actions.
28677 The packet was understood and carried out.
28679 The packet was not recognized.
28682 @item QTDP:-@var{n}:@var{addr}:@r{[}S@r{]}@var{action}@dots{}@r{[}-@r{]}
28683 Define actions to be taken when a tracepoint is hit. @var{n} and
28684 @var{addr} must be the same as in the initial @samp{QTDP} packet for
28685 this tracepoint. This packet may only be sent immediately after
28686 another @samp{QTDP} packet that ended with a @samp{-}. If the
28687 trailing @samp{-} is present, further @samp{QTDP} packets will follow,
28688 specifying more actions for this tracepoint.
28690 In the series of action packets for a given tracepoint, at most one
28691 can have an @samp{S} before its first @var{action}. If such a packet
28692 is sent, it and the following packets define ``while-stepping''
28693 actions. Any prior packets define ordinary actions --- that is, those
28694 taken when the tracepoint is first hit. If no action packet has an
28695 @samp{S}, then all the packets in the series specify ordinary
28696 tracepoint actions.
28698 The @samp{@var{action}@dots{}} portion of the packet is a series of
28699 actions, concatenated without separators. Each action has one of the
28705 Collect the registers whose bits are set in @var{mask}. @var{mask} is
28706 a hexadecimal number whose @var{i}'th bit is set if register number
28707 @var{i} should be collected. (The least significant bit is numbered
28708 zero.) Note that @var{mask} may be any number of digits long; it may
28709 not fit in a 32-bit word.
28711 @item M @var{basereg},@var{offset},@var{len}
28712 Collect @var{len} bytes of memory starting at the address in register
28713 number @var{basereg}, plus @var{offset}. If @var{basereg} is
28714 @samp{-1}, then the range has a fixed address: @var{offset} is the
28715 address of the lowest byte to collect. The @var{basereg},
28716 @var{offset}, and @var{len} parameters are all unsigned hexadecimal
28717 values (the @samp{-1} value for @var{basereg} is a special case).
28719 @item X @var{len},@var{expr}
28720 Evaluate @var{expr}, whose length is @var{len}, and collect memory as
28721 it directs. @var{expr} is an agent expression, as described in
28722 @ref{Agent Expressions}. Each byte of the expression is encoded as a
28723 two-digit hex number in the packet; @var{len} is the number of bytes
28724 in the expression (and thus one-half the number of hex digits in the
28729 Any number of actions may be packed together in a single @samp{QTDP}
28730 packet, as long as the packet does not exceed the maximum packet
28731 length (400 bytes, for many stubs). There may be only one @samp{R}
28732 action per tracepoint, and it must precede any @samp{M} or @samp{X}
28733 actions. Any registers referred to by @samp{M} and @samp{X} actions
28734 must be collected by a preceding @samp{R} action. (The
28735 ``while-stepping'' actions are treated as if they were attached to a
28736 separate tracepoint, as far as these restrictions are concerned.)
28741 The packet was understood and carried out.
28743 The packet was not recognized.
28746 @item QTFrame:@var{n}
28747 Select the @var{n}'th tracepoint frame from the buffer, and use the
28748 register and memory contents recorded there to answer subsequent
28749 request packets from @value{GDBN}.
28751 A successful reply from the stub indicates that the stub has found the
28752 requested frame. The response is a series of parts, concatenated
28753 without separators, describing the frame we selected. Each part has
28754 one of the following forms:
28758 The selected frame is number @var{n} in the trace frame buffer;
28759 @var{f} is a hexadecimal number. If @var{f} is @samp{-1}, then there
28760 was no frame matching the criteria in the request packet.
28763 The selected trace frame records a hit of tracepoint number @var{t};
28764 @var{t} is a hexadecimal number.
28768 @item QTFrame:pc:@var{addr}
28769 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
28770 currently selected frame whose PC is @var{addr};
28771 @var{addr} is a hexadecimal number.
28773 @item QTFrame:tdp:@var{t}
28774 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
28775 currently selected frame that is a hit of tracepoint @var{t}; @var{t}
28776 is a hexadecimal number.
28778 @item QTFrame:range:@var{start}:@var{end}
28779 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
28780 currently selected frame whose PC is between @var{start} (inclusive)
28781 and @var{end} (exclusive); @var{start} and @var{end} are hexadecimal
28784 @item QTFrame:outside:@var{start}:@var{end}
28785 Like @samp{QTFrame:range:@var{start}:@var{end}}, but select the first
28786 frame @emph{outside} the given range of addresses.
28789 Begin the tracepoint experiment. Begin collecting data from tracepoint
28790 hits in the trace frame buffer.
28793 End the tracepoint experiment. Stop collecting trace frames.
28796 Clear the table of tracepoints, and empty the trace frame buffer.
28798 @item QTro:@var{start1},@var{end1}:@var{start2},@var{end2}:@dots{}
28799 Establish the given ranges of memory as ``transparent''. The stub
28800 will answer requests for these ranges from memory's current contents,
28801 if they were not collected as part of the tracepoint hit.
28803 @value{GDBN} uses this to mark read-only regions of memory, like those
28804 containing program code. Since these areas never change, they should
28805 still have the same contents they did when the tracepoint was hit, so
28806 there's no reason for the stub to refuse to provide their contents.
28809 Ask the stub if there is a trace experiment running right now.
28814 There is no trace experiment running.
28816 There is a trace experiment running.
28822 @node Host I/O Packets
28823 @section Host I/O Packets
28824 @cindex Host I/O, remote protocol
28825 @cindex file transfer, remote protocol
28827 The @dfn{Host I/O} packets allow @value{GDBN} to perform I/O
28828 operations on the far side of a remote link. For example, Host I/O is
28829 used to upload and download files to a remote target with its own
28830 filesystem. Host I/O uses the same constant values and data structure
28831 layout as the target-initiated File-I/O protocol. However, the
28832 Host I/O packets are structured differently. The target-initiated
28833 protocol relies on target memory to store parameters and buffers.
28834 Host I/O requests are initiated by @value{GDBN}, and the
28835 target's memory is not involved. @xref{File-I/O Remote Protocol
28836 Extension}, for more details on the target-initiated protocol.
28838 The Host I/O request packets all encode a single operation along with
28839 its arguments. They have this format:
28843 @item vFile:@var{operation}: @var{parameter}@dots{}
28844 @var{operation} is the name of the particular request; the target
28845 should compare the entire packet name up to the second colon when checking
28846 for a supported operation. The format of @var{parameter} depends on
28847 the operation. Numbers are always passed in hexadecimal. Negative
28848 numbers have an explicit minus sign (i.e.@: two's complement is not
28849 used). Strings (e.g.@: filenames) are encoded as a series of
28850 hexadecimal bytes. The last argument to a system call may be a
28851 buffer of escaped binary data (@pxref{Binary Data}).
28855 The valid responses to Host I/O packets are:
28859 @item F @var{result} [, @var{errno}] [; @var{attachment}]
28860 @var{result} is the integer value returned by this operation, usually
28861 non-negative for success and -1 for errors. If an error has occured,
28862 @var{errno} will be included in the result. @var{errno} will have a
28863 value defined by the File-I/O protocol (@pxref{Errno Values}). For
28864 operations which return data, @var{attachment} supplies the data as a
28865 binary buffer. Binary buffers in response packets are escaped in the
28866 normal way (@pxref{Binary Data}). See the individual packet
28867 documentation for the interpretation of @var{result} and
28871 An empty response indicates that this operation is not recognized.
28875 These are the supported Host I/O operations:
28878 @item vFile:open: @var{pathname}, @var{flags}, @var{mode}
28879 Open a file at @var{pathname} and return a file descriptor for it, or
28880 return -1 if an error occurs. @var{pathname} is a string,
28881 @var{flags} is an integer indicating a mask of open flags
28882 (@pxref{Open Flags}), and @var{mode} is an integer indicating a mask
28883 of mode bits to use if the file is created (@pxref{mode_t Values}).
28884 @xref{open}, for details of the open flags and mode values.
28886 @item vFile:close: @var{fd}
28887 Close the open file corresponding to @var{fd} and return 0, or
28888 -1 if an error occurs.
28890 @item vFile:pread: @var{fd}, @var{count}, @var{offset}
28891 Read data from the open file corresponding to @var{fd}. Up to
28892 @var{count} bytes will be read from the file, starting at @var{offset}
28893 relative to the start of the file. The target may read fewer bytes;
28894 common reasons include packet size limits and an end-of-file
28895 condition. The number of bytes read is returned. Zero should only be
28896 returned for a successful read at the end of the file, or if
28897 @var{count} was zero.
28899 The data read should be returned as a binary attachment on success.
28900 If zero bytes were read, the response should include an empty binary
28901 attachment (i.e.@: a trailing semicolon). The return value is the
28902 number of target bytes read; the binary attachment may be longer if
28903 some characters were escaped.
28905 @item vFile:pwrite: @var{fd}, @var{offset}, @var{data}
28906 Write @var{data} (a binary buffer) to the open file corresponding
28907 to @var{fd}. Start the write at @var{offset} from the start of the
28908 file. Unlike many @code{write} system calls, there is no
28909 separate @var{count} argument; the length of @var{data} in the
28910 packet is used. @samp{vFile:write} returns the number of bytes written,
28911 which may be shorter than the length of @var{data}, or -1 if an
28914 @item vFile:unlink: @var{pathname}
28915 Delete the file at @var{pathname} on the target. Return 0,
28916 or -1 if an error occurs. @var{pathname} is a string.
28921 @section Interrupts
28922 @cindex interrupts (remote protocol)
28924 When a program on the remote target is running, @value{GDBN} may
28925 attempt to interrupt it by sending a @samp{Ctrl-C} or a @code{BREAK},
28926 control of which is specified via @value{GDBN}'s @samp{remotebreak}
28927 setting (@pxref{set remotebreak}).
28929 The precise meaning of @code{BREAK} is defined by the transport
28930 mechanism and may, in fact, be undefined. @value{GDBN} does not
28931 currently define a @code{BREAK} mechanism for any of the network
28932 interfaces except for TCP, in which case @value{GDBN} sends the
28933 @code{telnet} BREAK sequence.
28935 @samp{Ctrl-C}, on the other hand, is defined and implemented for all
28936 transport mechanisms. It is represented by sending the single byte
28937 @code{0x03} without any of the usual packet overhead described in
28938 the Overview section (@pxref{Overview}). When a @code{0x03} byte is
28939 transmitted as part of a packet, it is considered to be packet data
28940 and does @emph{not} represent an interrupt. E.g., an @samp{X} packet
28941 (@pxref{X packet}), used for binary downloads, may include an unescaped
28942 @code{0x03} as part of its packet.
28944 Stubs are not required to recognize these interrupt mechanisms and the
28945 precise meaning associated with receipt of the interrupt is
28946 implementation defined. If the target supports debugging of multiple
28947 threads and/or processes, it should attempt to interrupt all
28948 currently-executing threads and processes.
28949 If the stub is successful at interrupting the
28950 running program, it should send one of the stop
28951 reply packets (@pxref{Stop Reply Packets}) to @value{GDBN} as a result
28952 of successfully stopping the program in all-stop mode, and a stop reply
28953 for each stopped thread in non-stop mode.
28954 Interrupts received while the
28955 program is stopped are discarded.
28957 @node Notification Packets
28958 @section Notification Packets
28959 @cindex notification packets
28960 @cindex packets, notification
28962 The @value{GDBN} remote serial protocol includes @dfn{notifications},
28963 packets that require no acknowledgment. Both the GDB and the stub
28964 may send notifications (although the only notifications defined at
28965 present are sent by the stub). Notifications carry information
28966 without incurring the round-trip latency of an acknowledgment, and so
28967 are useful for low-impact communications where occasional packet loss
28970 A notification packet has the form @samp{% @var{data} #
28971 @var{checksum}}, where @var{data} is the content of the notification,
28972 and @var{checksum} is a checksum of @var{data}, computed and formatted
28973 as for ordinary @value{GDBN} packets. A notification's @var{data}
28974 never contains @samp{$}, @samp{%} or @samp{#} characters. Upon
28975 receiving a notification, the recipient sends no @samp{+} or @samp{-}
28976 to acknowledge the notification's receipt or to report its corruption.
28978 Every notification's @var{data} begins with a name, which contains no
28979 colon characters, followed by a colon character.
28981 Recipients should silently ignore corrupted notifications and
28982 notifications they do not understand. Recipients should restart
28983 timeout periods on receipt of a well-formed notification, whether or
28984 not they understand it.
28986 Senders should only send the notifications described here when this
28987 protocol description specifies that they are permitted. In the
28988 future, we may extend the protocol to permit existing notifications in
28989 new contexts; this rule helps older senders avoid confusing newer
28992 (Older versions of @value{GDBN} ignore bytes received until they see
28993 the @samp{$} byte that begins an ordinary packet, so new stubs may
28994 transmit notifications without fear of confusing older clients. There
28995 are no notifications defined for @value{GDBN} to send at the moment, but we
28996 assume that most older stubs would ignore them, as well.)
28998 The following notification packets from the stub to @value{GDBN} are
29002 @item Stop: @var{reply}
29003 Report an asynchronous stop event in non-stop mode.
29004 The @var{reply} has the form of a stop reply, as
29005 described in @ref{Stop Reply Packets}. Refer to @ref{Remote Non-Stop},
29006 for information on how these notifications are acknowledged by
29010 @node Remote Non-Stop
29011 @section Remote Protocol Support for Non-Stop Mode
29013 @value{GDBN}'s remote protocol supports non-stop debugging of
29014 multi-threaded programs, as described in @ref{Non-Stop Mode}. If the stub
29015 supports non-stop mode, it should report that to @value{GDBN} by including
29016 @samp{QNonStop+} in its @samp{qSupported} response (@pxref{qSupported}).
29018 @value{GDBN} typically sends a @samp{QNonStop} packet only when
29019 establishing a new connection with the stub. Entering non-stop mode
29020 does not alter the state of any currently-running threads, but targets
29021 must stop all threads in any already-attached processes when entering
29022 all-stop mode. @value{GDBN} uses the @samp{?} packet as necessary to
29023 probe the target state after a mode change.
29025 In non-stop mode, when an attached process encounters an event that
29026 would otherwise be reported with a stop reply, it uses the
29027 asynchronous notification mechanism (@pxref{Notification Packets}) to
29028 inform @value{GDBN}. In contrast to all-stop mode, where all threads
29029 in all processes are stopped when a stop reply is sent, in non-stop
29030 mode only the thread reporting the stop event is stopped. That is,
29031 when reporting a @samp{S} or @samp{T} response to indicate completion
29032 of a step operation, hitting a breakpoint, or a fault, only the
29033 affected thread is stopped; any other still-running threads continue
29034 to run. When reporting a @samp{W} or @samp{X} response, all running
29035 threads belonging to other attached processes continue to run.
29037 Only one stop reply notification at a time may be pending; if
29038 additional stop events occur before @value{GDBN} has acknowledged the
29039 previous notification, they must be queued by the stub for later
29040 synchronous transmission in response to @samp{vStopped} packets from
29041 @value{GDBN}. Because the notification mechanism is unreliable,
29042 the stub is permitted to resend a stop reply notification
29043 if it believes @value{GDBN} may not have received it. @value{GDBN}
29044 ignores additional stop reply notifications received before it has
29045 finished processing a previous notification and the stub has completed
29046 sending any queued stop events.
29048 Otherwise, @value{GDBN} must be prepared to receive a stop reply
29049 notification at any time. Specifically, they may appear when
29050 @value{GDBN} is not otherwise reading input from the stub, or when
29051 @value{GDBN} is expecting to read a normal synchronous response or a
29052 @samp{+}/@samp{-} acknowledgment to a packet it has sent.
29053 Notification packets are distinct from any other communication from
29054 the stub so there is no ambiguity.
29056 After receiving a stop reply notification, @value{GDBN} shall
29057 acknowledge it by sending a @samp{vStopped} packet (@pxref{vStopped packet})
29058 as a regular, synchronous request to the stub. Such acknowledgment
29059 is not required to happen immediately, as @value{GDBN} is permitted to
29060 send other, unrelated packets to the stub first, which the stub should
29063 Upon receiving a @samp{vStopped} packet, if the stub has other queued
29064 stop events to report to @value{GDBN}, it shall respond by sending a
29065 normal stop reply response. @value{GDBN} shall then send another
29066 @samp{vStopped} packet to solicit further responses; again, it is
29067 permitted to send other, unrelated packets as well which the stub
29068 should process normally.
29070 If the stub receives a @samp{vStopped} packet and there are no
29071 additional stop events to report, the stub shall return an @samp{OK}
29072 response. At this point, if further stop events occur, the stub shall
29073 send a new stop reply notification, @value{GDBN} shall accept the
29074 notification, and the process shall be repeated.
29076 In non-stop mode, the target shall respond to the @samp{?} packet as
29077 follows. First, any incomplete stop reply notification/@samp{vStopped}
29078 sequence in progress is abandoned. The target must begin a new
29079 sequence reporting stop events for all stopped threads, whether or not
29080 it has previously reported those events to @value{GDBN}. The first
29081 stop reply is sent as a synchronous reply to the @samp{?} packet, and
29082 subsequent stop replies are sent as responses to @samp{vStopped} packets
29083 using the mechanism described above. The target must not send
29084 asynchronous stop reply notifications until the sequence is complete.
29085 If all threads are running when the target receives the @samp{?} packet,
29086 or if the target is not attached to any process, it shall respond
29089 @node Packet Acknowledgment
29090 @section Packet Acknowledgment
29092 @cindex acknowledgment, for @value{GDBN} remote
29093 @cindex packet acknowledgment, for @value{GDBN} remote
29094 By default, when either the host or the target machine receives a packet,
29095 the first response expected is an acknowledgment: either @samp{+} (to indicate
29096 the package was received correctly) or @samp{-} (to request retransmission).
29097 This mechanism allows the @value{GDBN} remote protocol to operate over
29098 unreliable transport mechanisms, such as a serial line.
29100 In cases where the transport mechanism is itself reliable (such as a pipe or
29101 TCP connection), the @samp{+}/@samp{-} acknowledgments are redundant.
29102 It may be desirable to disable them in that case to reduce communication
29103 overhead, or for other reasons. This can be accomplished by means of the
29104 @samp{QStartNoAckMode} packet; @pxref{QStartNoAckMode}.
29106 When in no-acknowledgment mode, neither the stub nor @value{GDBN} shall send or
29107 expect @samp{+}/@samp{-} protocol acknowledgments. The packet
29108 and response format still includes the normal checksum, as described in
29109 @ref{Overview}, but the checksum may be ignored by the receiver.
29111 If the stub supports @samp{QStartNoAckMode} and prefers to operate in
29112 no-acknowledgment mode, it should report that to @value{GDBN}
29113 by including @samp{QStartNoAckMode+} in its response to @samp{qSupported};
29114 @pxref{qSupported}.
29115 If @value{GDBN} also supports @samp{QStartNoAckMode} and it has not been
29116 disabled via the @code{set remote noack-packet off} command
29117 (@pxref{Remote Configuration}),
29118 @value{GDBN} may then send a @samp{QStartNoAckMode} packet to the stub.
29119 Only then may the stub actually turn off packet acknowledgments.
29120 @value{GDBN} sends a final @samp{+} acknowledgment of the stub's @samp{OK}
29121 response, which can be safely ignored by the stub.
29123 Note that @code{set remote noack-packet} command only affects negotiation
29124 between @value{GDBN} and the stub when subsequent connections are made;
29125 it does not affect the protocol acknowledgment state for any current
29127 Since @samp{+}/@samp{-} acknowledgments are enabled by default when a
29128 new connection is established,
29129 there is also no protocol request to re-enable the acknowledgments
29130 for the current connection, once disabled.
29135 Example sequence of a target being re-started. Notice how the restart
29136 does not get any direct output:
29141 @emph{target restarts}
29144 <- @code{T001:1234123412341234}
29148 Example sequence of a target being stepped by a single instruction:
29151 -> @code{G1445@dots{}}
29156 <- @code{T001:1234123412341234}
29160 <- @code{1455@dots{}}
29164 @node File-I/O Remote Protocol Extension
29165 @section File-I/O Remote Protocol Extension
29166 @cindex File-I/O remote protocol extension
29169 * File-I/O Overview::
29170 * Protocol Basics::
29171 * The F Request Packet::
29172 * The F Reply Packet::
29173 * The Ctrl-C Message::
29175 * List of Supported Calls::
29176 * Protocol-specific Representation of Datatypes::
29178 * File-I/O Examples::
29181 @node File-I/O Overview
29182 @subsection File-I/O Overview
29183 @cindex file-i/o overview
29185 The @dfn{File I/O remote protocol extension} (short: File-I/O) allows the
29186 target to use the host's file system and console I/O to perform various
29187 system calls. System calls on the target system are translated into a
29188 remote protocol packet to the host system, which then performs the needed
29189 actions and returns a response packet to the target system.
29190 This simulates file system operations even on targets that lack file systems.
29192 The protocol is defined to be independent of both the host and target systems.
29193 It uses its own internal representation of datatypes and values. Both
29194 @value{GDBN} and the target's @value{GDBN} stub are responsible for
29195 translating the system-dependent value representations into the internal
29196 protocol representations when data is transmitted.
29198 The communication is synchronous. A system call is possible only when
29199 @value{GDBN} is waiting for a response from the @samp{C}, @samp{c}, @samp{S}
29200 or @samp{s} packets. While @value{GDBN} handles the request for a system call,
29201 the target is stopped to allow deterministic access to the target's
29202 memory. Therefore File-I/O is not interruptible by target signals. On
29203 the other hand, it is possible to interrupt File-I/O by a user interrupt
29204 (@samp{Ctrl-C}) within @value{GDBN}.
29206 The target's request to perform a host system call does not finish
29207 the latest @samp{C}, @samp{c}, @samp{S} or @samp{s} action. That means,
29208 after finishing the system call, the target returns to continuing the
29209 previous activity (continue, step). No additional continue or step
29210 request from @value{GDBN} is required.
29213 (@value{GDBP}) continue
29214 <- target requests 'system call X'
29215 target is stopped, @value{GDBN} executes system call
29216 -> @value{GDBN} returns result
29217 ... target continues, @value{GDBN} returns to wait for the target
29218 <- target hits breakpoint and sends a Txx packet
29221 The protocol only supports I/O on the console and to regular files on
29222 the host file system. Character or block special devices, pipes,
29223 named pipes, sockets or any other communication method on the host
29224 system are not supported by this protocol.
29226 File I/O is not supported in non-stop mode.
29228 @node Protocol Basics
29229 @subsection Protocol Basics
29230 @cindex protocol basics, file-i/o
29232 The File-I/O protocol uses the @code{F} packet as the request as well
29233 as reply packet. Since a File-I/O system call can only occur when
29234 @value{GDBN} is waiting for a response from the continuing or stepping target,
29235 the File-I/O request is a reply that @value{GDBN} has to expect as a result
29236 of a previous @samp{C}, @samp{c}, @samp{S} or @samp{s} packet.
29237 This @code{F} packet contains all information needed to allow @value{GDBN}
29238 to call the appropriate host system call:
29242 A unique identifier for the requested system call.
29245 All parameters to the system call. Pointers are given as addresses
29246 in the target memory address space. Pointers to strings are given as
29247 pointer/length pair. Numerical values are given as they are.
29248 Numerical control flags are given in a protocol-specific representation.
29252 At this point, @value{GDBN} has to perform the following actions.
29256 If the parameters include pointer values to data needed as input to a
29257 system call, @value{GDBN} requests this data from the target with a
29258 standard @code{m} packet request. This additional communication has to be
29259 expected by the target implementation and is handled as any other @code{m}
29263 @value{GDBN} translates all value from protocol representation to host
29264 representation as needed. Datatypes are coerced into the host types.
29267 @value{GDBN} calls the system call.
29270 It then coerces datatypes back to protocol representation.
29273 If the system call is expected to return data in buffer space specified
29274 by pointer parameters to the call, the data is transmitted to the
29275 target using a @code{M} or @code{X} packet. This packet has to be expected
29276 by the target implementation and is handled as any other @code{M} or @code{X}
29281 Eventually @value{GDBN} replies with another @code{F} packet which contains all
29282 necessary information for the target to continue. This at least contains
29289 @code{errno}, if has been changed by the system call.
29296 After having done the needed type and value coercion, the target continues
29297 the latest continue or step action.
29299 @node The F Request Packet
29300 @subsection The @code{F} Request Packet
29301 @cindex file-i/o request packet
29302 @cindex @code{F} request packet
29304 The @code{F} request packet has the following format:
29307 @item F@var{call-id},@var{parameter@dots{}}
29309 @var{call-id} is the identifier to indicate the host system call to be called.
29310 This is just the name of the function.
29312 @var{parameter@dots{}} are the parameters to the system call.
29313 Parameters are hexadecimal integer values, either the actual values in case
29314 of scalar datatypes, pointers to target buffer space in case of compound
29315 datatypes and unspecified memory areas, or pointer/length pairs in case
29316 of string parameters. These are appended to the @var{call-id} as a
29317 comma-delimited list. All values are transmitted in ASCII
29318 string representation, pointer/length pairs separated by a slash.
29324 @node The F Reply Packet
29325 @subsection The @code{F} Reply Packet
29326 @cindex file-i/o reply packet
29327 @cindex @code{F} reply packet
29329 The @code{F} reply packet has the following format:
29333 @item F@var{retcode},@var{errno},@var{Ctrl-C flag};@var{call-specific attachment}
29335 @var{retcode} is the return code of the system call as hexadecimal value.
29337 @var{errno} is the @code{errno} set by the call, in protocol-specific
29339 This parameter can be omitted if the call was successful.
29341 @var{Ctrl-C flag} is only sent if the user requested a break. In this
29342 case, @var{errno} must be sent as well, even if the call was successful.
29343 The @var{Ctrl-C flag} itself consists of the character @samp{C}:
29350 or, if the call was interrupted before the host call has been performed:
29357 assuming 4 is the protocol-specific representation of @code{EINTR}.
29362 @node The Ctrl-C Message
29363 @subsection The @samp{Ctrl-C} Message
29364 @cindex ctrl-c message, in file-i/o protocol
29366 If the @samp{Ctrl-C} flag is set in the @value{GDBN}
29367 reply packet (@pxref{The F Reply Packet}),
29368 the target should behave as if it had
29369 gotten a break message. The meaning for the target is ``system call
29370 interrupted by @code{SIGINT}''. Consequentially, the target should actually stop
29371 (as with a break message) and return to @value{GDBN} with a @code{T02}
29374 It's important for the target to know in which
29375 state the system call was interrupted. There are two possible cases:
29379 The system call hasn't been performed on the host yet.
29382 The system call on the host has been finished.
29386 These two states can be distinguished by the target by the value of the
29387 returned @code{errno}. If it's the protocol representation of @code{EINTR}, the system
29388 call hasn't been performed. This is equivalent to the @code{EINTR} handling
29389 on POSIX systems. In any other case, the target may presume that the
29390 system call has been finished --- successfully or not --- and should behave
29391 as if the break message arrived right after the system call.
29393 @value{GDBN} must behave reliably. If the system call has not been called
29394 yet, @value{GDBN} may send the @code{F} reply immediately, setting @code{EINTR} as
29395 @code{errno} in the packet. If the system call on the host has been finished
29396 before the user requests a break, the full action must be finished by
29397 @value{GDBN}. This requires sending @code{M} or @code{X} packets as necessary.
29398 The @code{F} packet may only be sent when either nothing has happened
29399 or the full action has been completed.
29402 @subsection Console I/O
29403 @cindex console i/o as part of file-i/o
29405 By default and if not explicitly closed by the target system, the file
29406 descriptors 0, 1 and 2 are connected to the @value{GDBN} console. Output
29407 on the @value{GDBN} console is handled as any other file output operation
29408 (@code{write(1, @dots{})} or @code{write(2, @dots{})}). Console input is handled
29409 by @value{GDBN} so that after the target read request from file descriptor
29410 0 all following typing is buffered until either one of the following
29415 The user types @kbd{Ctrl-c}. The behaviour is as explained above, and the
29417 system call is treated as finished.
29420 The user presses @key{RET}. This is treated as end of input with a trailing
29424 The user types @kbd{Ctrl-d}. This is treated as end of input. No trailing
29425 character (neither newline nor @samp{Ctrl-D}) is appended to the input.
29429 If the user has typed more characters than fit in the buffer given to
29430 the @code{read} call, the trailing characters are buffered in @value{GDBN} until
29431 either another @code{read(0, @dots{})} is requested by the target, or debugging
29432 is stopped at the user's request.
29435 @node List of Supported Calls
29436 @subsection List of Supported Calls
29437 @cindex list of supported file-i/o calls
29454 @unnumberedsubsubsec open
29455 @cindex open, file-i/o system call
29460 int open(const char *pathname, int flags);
29461 int open(const char *pathname, int flags, mode_t mode);
29465 @samp{Fopen,@var{pathptr}/@var{len},@var{flags},@var{mode}}
29468 @var{flags} is the bitwise @code{OR} of the following values:
29472 If the file does not exist it will be created. The host
29473 rules apply as far as file ownership and time stamps
29477 When used with @code{O_CREAT}, if the file already exists it is
29478 an error and open() fails.
29481 If the file already exists and the open mode allows
29482 writing (@code{O_RDWR} or @code{O_WRONLY} is given) it will be
29483 truncated to zero length.
29486 The file is opened in append mode.
29489 The file is opened for reading only.
29492 The file is opened for writing only.
29495 The file is opened for reading and writing.
29499 Other bits are silently ignored.
29503 @var{mode} is the bitwise @code{OR} of the following values:
29507 User has read permission.
29510 User has write permission.
29513 Group has read permission.
29516 Group has write permission.
29519 Others have read permission.
29522 Others have write permission.
29526 Other bits are silently ignored.
29529 @item Return value:
29530 @code{open} returns the new file descriptor or -1 if an error
29537 @var{pathname} already exists and @code{O_CREAT} and @code{O_EXCL} were used.
29540 @var{pathname} refers to a directory.
29543 The requested access is not allowed.
29546 @var{pathname} was too long.
29549 A directory component in @var{pathname} does not exist.
29552 @var{pathname} refers to a device, pipe, named pipe or socket.
29555 @var{pathname} refers to a file on a read-only filesystem and
29556 write access was requested.
29559 @var{pathname} is an invalid pointer value.
29562 No space on device to create the file.
29565 The process already has the maximum number of files open.
29568 The limit on the total number of files open on the system
29572 The call was interrupted by the user.
29578 @unnumberedsubsubsec close
29579 @cindex close, file-i/o system call
29588 @samp{Fclose,@var{fd}}
29590 @item Return value:
29591 @code{close} returns zero on success, or -1 if an error occurred.
29597 @var{fd} isn't a valid open file descriptor.
29600 The call was interrupted by the user.
29606 @unnumberedsubsubsec read
29607 @cindex read, file-i/o system call
29612 int read(int fd, void *buf, unsigned int count);
29616 @samp{Fread,@var{fd},@var{bufptr},@var{count}}
29618 @item Return value:
29619 On success, the number of bytes read is returned.
29620 Zero indicates end of file. If count is zero, read
29621 returns zero as well. On error, -1 is returned.
29627 @var{fd} is not a valid file descriptor or is not open for
29631 @var{bufptr} is an invalid pointer value.
29634 The call was interrupted by the user.
29640 @unnumberedsubsubsec write
29641 @cindex write, file-i/o system call
29646 int write(int fd, const void *buf, unsigned int count);
29650 @samp{Fwrite,@var{fd},@var{bufptr},@var{count}}
29652 @item Return value:
29653 On success, the number of bytes written are returned.
29654 Zero indicates nothing was written. On error, -1
29661 @var{fd} is not a valid file descriptor or is not open for
29665 @var{bufptr} is an invalid pointer value.
29668 An attempt was made to write a file that exceeds the
29669 host-specific maximum file size allowed.
29672 No space on device to write the data.
29675 The call was interrupted by the user.
29681 @unnumberedsubsubsec lseek
29682 @cindex lseek, file-i/o system call
29687 long lseek (int fd, long offset, int flag);
29691 @samp{Flseek,@var{fd},@var{offset},@var{flag}}
29693 @var{flag} is one of:
29697 The offset is set to @var{offset} bytes.
29700 The offset is set to its current location plus @var{offset}
29704 The offset is set to the size of the file plus @var{offset}
29708 @item Return value:
29709 On success, the resulting unsigned offset in bytes from
29710 the beginning of the file is returned. Otherwise, a
29711 value of -1 is returned.
29717 @var{fd} is not a valid open file descriptor.
29720 @var{fd} is associated with the @value{GDBN} console.
29723 @var{flag} is not a proper value.
29726 The call was interrupted by the user.
29732 @unnumberedsubsubsec rename
29733 @cindex rename, file-i/o system call
29738 int rename(const char *oldpath, const char *newpath);
29742 @samp{Frename,@var{oldpathptr}/@var{len},@var{newpathptr}/@var{len}}
29744 @item Return value:
29745 On success, zero is returned. On error, -1 is returned.
29751 @var{newpath} is an existing directory, but @var{oldpath} is not a
29755 @var{newpath} is a non-empty directory.
29758 @var{oldpath} or @var{newpath} is a directory that is in use by some
29762 An attempt was made to make a directory a subdirectory
29766 A component used as a directory in @var{oldpath} or new
29767 path is not a directory. Or @var{oldpath} is a directory
29768 and @var{newpath} exists but is not a directory.
29771 @var{oldpathptr} or @var{newpathptr} are invalid pointer values.
29774 No access to the file or the path of the file.
29778 @var{oldpath} or @var{newpath} was too long.
29781 A directory component in @var{oldpath} or @var{newpath} does not exist.
29784 The file is on a read-only filesystem.
29787 The device containing the file has no room for the new
29791 The call was interrupted by the user.
29797 @unnumberedsubsubsec unlink
29798 @cindex unlink, file-i/o system call
29803 int unlink(const char *pathname);
29807 @samp{Funlink,@var{pathnameptr}/@var{len}}
29809 @item Return value:
29810 On success, zero is returned. On error, -1 is returned.
29816 No access to the file or the path of the file.
29819 The system does not allow unlinking of directories.
29822 The file @var{pathname} cannot be unlinked because it's
29823 being used by another process.
29826 @var{pathnameptr} is an invalid pointer value.
29829 @var{pathname} was too long.
29832 A directory component in @var{pathname} does not exist.
29835 A component of the path is not a directory.
29838 The file is on a read-only filesystem.
29841 The call was interrupted by the user.
29847 @unnumberedsubsubsec stat/fstat
29848 @cindex fstat, file-i/o system call
29849 @cindex stat, file-i/o system call
29854 int stat(const char *pathname, struct stat *buf);
29855 int fstat(int fd, struct stat *buf);
29859 @samp{Fstat,@var{pathnameptr}/@var{len},@var{bufptr}}@*
29860 @samp{Ffstat,@var{fd},@var{bufptr}}
29862 @item Return value:
29863 On success, zero is returned. On error, -1 is returned.
29869 @var{fd} is not a valid open file.
29872 A directory component in @var{pathname} does not exist or the
29873 path is an empty string.
29876 A component of the path is not a directory.
29879 @var{pathnameptr} is an invalid pointer value.
29882 No access to the file or the path of the file.
29885 @var{pathname} was too long.
29888 The call was interrupted by the user.
29894 @unnumberedsubsubsec gettimeofday
29895 @cindex gettimeofday, file-i/o system call
29900 int gettimeofday(struct timeval *tv, void *tz);
29904 @samp{Fgettimeofday,@var{tvptr},@var{tzptr}}
29906 @item Return value:
29907 On success, 0 is returned, -1 otherwise.
29913 @var{tz} is a non-NULL pointer.
29916 @var{tvptr} and/or @var{tzptr} is an invalid pointer value.
29922 @unnumberedsubsubsec isatty
29923 @cindex isatty, file-i/o system call
29928 int isatty(int fd);
29932 @samp{Fisatty,@var{fd}}
29934 @item Return value:
29935 Returns 1 if @var{fd} refers to the @value{GDBN} console, 0 otherwise.
29941 The call was interrupted by the user.
29946 Note that the @code{isatty} call is treated as a special case: it returns
29947 1 to the target if the file descriptor is attached
29948 to the @value{GDBN} console, 0 otherwise. Implementing through system calls
29949 would require implementing @code{ioctl} and would be more complex than
29954 @unnumberedsubsubsec system
29955 @cindex system, file-i/o system call
29960 int system(const char *command);
29964 @samp{Fsystem,@var{commandptr}/@var{len}}
29966 @item Return value:
29967 If @var{len} is zero, the return value indicates whether a shell is
29968 available. A zero return value indicates a shell is not available.
29969 For non-zero @var{len}, the value returned is -1 on error and the
29970 return status of the command otherwise. Only the exit status of the
29971 command is returned, which is extracted from the host's @code{system}
29972 return value by calling @code{WEXITSTATUS(retval)}. In case
29973 @file{/bin/sh} could not be executed, 127 is returned.
29979 The call was interrupted by the user.
29984 @value{GDBN} takes over the full task of calling the necessary host calls
29985 to perform the @code{system} call. The return value of @code{system} on
29986 the host is simplified before it's returned
29987 to the target. Any termination signal information from the child process
29988 is discarded, and the return value consists
29989 entirely of the exit status of the called command.
29991 Due to security concerns, the @code{system} call is by default refused
29992 by @value{GDBN}. The user has to allow this call explicitly with the
29993 @code{set remote system-call-allowed 1} command.
29996 @item set remote system-call-allowed
29997 @kindex set remote system-call-allowed
29998 Control whether to allow the @code{system} calls in the File I/O
29999 protocol for the remote target. The default is zero (disabled).
30001 @item show remote system-call-allowed
30002 @kindex show remote system-call-allowed
30003 Show whether the @code{system} calls are allowed in the File I/O
30007 @node Protocol-specific Representation of Datatypes
30008 @subsection Protocol-specific Representation of Datatypes
30009 @cindex protocol-specific representation of datatypes, in file-i/o protocol
30012 * Integral Datatypes::
30014 * Memory Transfer::
30019 @node Integral Datatypes
30020 @unnumberedsubsubsec Integral Datatypes
30021 @cindex integral datatypes, in file-i/o protocol
30023 The integral datatypes used in the system calls are @code{int},
30024 @code{unsigned int}, @code{long}, @code{unsigned long},
30025 @code{mode_t}, and @code{time_t}.
30027 @code{int}, @code{unsigned int}, @code{mode_t} and @code{time_t} are
30028 implemented as 32 bit values in this protocol.
30030 @code{long} and @code{unsigned long} are implemented as 64 bit types.
30032 @xref{Limits}, for corresponding MIN and MAX values (similar to those
30033 in @file{limits.h}) to allow range checking on host and target.
30035 @code{time_t} datatypes are defined as seconds since the Epoch.
30037 All integral datatypes transferred as part of a memory read or write of a
30038 structured datatype e.g.@: a @code{struct stat} have to be given in big endian
30041 @node Pointer Values
30042 @unnumberedsubsubsec Pointer Values
30043 @cindex pointer values, in file-i/o protocol
30045 Pointers to target data are transmitted as they are. An exception
30046 is made for pointers to buffers for which the length isn't
30047 transmitted as part of the function call, namely strings. Strings
30048 are transmitted as a pointer/length pair, both as hex values, e.g.@:
30055 which is a pointer to data of length 18 bytes at position 0x1aaf.
30056 The length is defined as the full string length in bytes, including
30057 the trailing null byte. For example, the string @code{"hello world"}
30058 at address 0x123456 is transmitted as
30064 @node Memory Transfer
30065 @unnumberedsubsubsec Memory Transfer
30066 @cindex memory transfer, in file-i/o protocol
30068 Structured data which is transferred using a memory read or write (for
30069 example, a @code{struct stat}) is expected to be in a protocol-specific format
30070 with all scalar multibyte datatypes being big endian. Translation to
30071 this representation needs to be done both by the target before the @code{F}
30072 packet is sent, and by @value{GDBN} before
30073 it transfers memory to the target. Transferred pointers to structured
30074 data should point to the already-coerced data at any time.
30078 @unnumberedsubsubsec struct stat
30079 @cindex struct stat, in file-i/o protocol
30081 The buffer of type @code{struct stat} used by the target and @value{GDBN}
30082 is defined as follows:
30086 unsigned int st_dev; /* device */
30087 unsigned int st_ino; /* inode */
30088 mode_t st_mode; /* protection */
30089 unsigned int st_nlink; /* number of hard links */
30090 unsigned int st_uid; /* user ID of owner */
30091 unsigned int st_gid; /* group ID of owner */
30092 unsigned int st_rdev; /* device type (if inode device) */
30093 unsigned long st_size; /* total size, in bytes */
30094 unsigned long st_blksize; /* blocksize for filesystem I/O */
30095 unsigned long st_blocks; /* number of blocks allocated */
30096 time_t st_atime; /* time of last access */
30097 time_t st_mtime; /* time of last modification */
30098 time_t st_ctime; /* time of last change */
30102 The integral datatypes conform to the definitions given in the
30103 appropriate section (see @ref{Integral Datatypes}, for details) so this
30104 structure is of size 64 bytes.
30106 The values of several fields have a restricted meaning and/or
30112 A value of 0 represents a file, 1 the console.
30115 No valid meaning for the target. Transmitted unchanged.
30118 Valid mode bits are described in @ref{Constants}. Any other
30119 bits have currently no meaning for the target.
30124 No valid meaning for the target. Transmitted unchanged.
30129 These values have a host and file system dependent
30130 accuracy. Especially on Windows hosts, the file system may not
30131 support exact timing values.
30134 The target gets a @code{struct stat} of the above representation and is
30135 responsible for coercing it to the target representation before
30138 Note that due to size differences between the host, target, and protocol
30139 representations of @code{struct stat} members, these members could eventually
30140 get truncated on the target.
30142 @node struct timeval
30143 @unnumberedsubsubsec struct timeval
30144 @cindex struct timeval, in file-i/o protocol
30146 The buffer of type @code{struct timeval} used by the File-I/O protocol
30147 is defined as follows:
30151 time_t tv_sec; /* second */
30152 long tv_usec; /* microsecond */
30156 The integral datatypes conform to the definitions given in the
30157 appropriate section (see @ref{Integral Datatypes}, for details) so this
30158 structure is of size 8 bytes.
30161 @subsection Constants
30162 @cindex constants, in file-i/o protocol
30164 The following values are used for the constants inside of the
30165 protocol. @value{GDBN} and target are responsible for translating these
30166 values before and after the call as needed.
30177 @unnumberedsubsubsec Open Flags
30178 @cindex open flags, in file-i/o protocol
30180 All values are given in hexadecimal representation.
30192 @node mode_t Values
30193 @unnumberedsubsubsec mode_t Values
30194 @cindex mode_t values, in file-i/o protocol
30196 All values are given in octal representation.
30213 @unnumberedsubsubsec Errno Values
30214 @cindex errno values, in file-i/o protocol
30216 All values are given in decimal representation.
30241 @code{EUNKNOWN} is used as a fallback error value if a host system returns
30242 any error value not in the list of supported error numbers.
30245 @unnumberedsubsubsec Lseek Flags
30246 @cindex lseek flags, in file-i/o protocol
30255 @unnumberedsubsubsec Limits
30256 @cindex limits, in file-i/o protocol
30258 All values are given in decimal representation.
30261 INT_MIN -2147483648
30263 UINT_MAX 4294967295
30264 LONG_MIN -9223372036854775808
30265 LONG_MAX 9223372036854775807
30266 ULONG_MAX 18446744073709551615
30269 @node File-I/O Examples
30270 @subsection File-I/O Examples
30271 @cindex file-i/o examples
30273 Example sequence of a write call, file descriptor 3, buffer is at target
30274 address 0x1234, 6 bytes should be written:
30277 <- @code{Fwrite,3,1234,6}
30278 @emph{request memory read from target}
30281 @emph{return "6 bytes written"}
30285 Example sequence of a read call, file descriptor 3, buffer is at target
30286 address 0x1234, 6 bytes should be read:
30289 <- @code{Fread,3,1234,6}
30290 @emph{request memory write to target}
30291 -> @code{X1234,6:XXXXXX}
30292 @emph{return "6 bytes read"}
30296 Example sequence of a read call, call fails on the host due to invalid
30297 file descriptor (@code{EBADF}):
30300 <- @code{Fread,3,1234,6}
30304 Example sequence of a read call, user presses @kbd{Ctrl-c} before syscall on
30308 <- @code{Fread,3,1234,6}
30313 Example sequence of a read call, user presses @kbd{Ctrl-c} after syscall on
30317 <- @code{Fread,3,1234,6}
30318 -> @code{X1234,6:XXXXXX}
30322 @node Library List Format
30323 @section Library List Format
30324 @cindex library list format, remote protocol
30326 On some platforms, a dynamic loader (e.g.@: @file{ld.so}) runs in the
30327 same process as your application to manage libraries. In this case,
30328 @value{GDBN} can use the loader's symbol table and normal memory
30329 operations to maintain a list of shared libraries. On other
30330 platforms, the operating system manages loaded libraries.
30331 @value{GDBN} can not retrieve the list of currently loaded libraries
30332 through memory operations, so it uses the @samp{qXfer:libraries:read}
30333 packet (@pxref{qXfer library list read}) instead. The remote stub
30334 queries the target's operating system and reports which libraries
30337 The @samp{qXfer:libraries:read} packet returns an XML document which
30338 lists loaded libraries and their offsets. Each library has an
30339 associated name and one or more segment or section base addresses,
30340 which report where the library was loaded in memory.
30342 For the common case of libraries that are fully linked binaries, the
30343 library should have a list of segments. If the target supports
30344 dynamic linking of a relocatable object file, its library XML element
30345 should instead include a list of allocated sections. The segment or
30346 section bases are start addresses, not relocation offsets; they do not
30347 depend on the library's link-time base addresses.
30349 @value{GDBN} must be linked with the Expat library to support XML
30350 library lists. @xref{Expat}.
30352 A simple memory map, with one loaded library relocated by a single
30353 offset, looks like this:
30357 <library name="/lib/libc.so.6">
30358 <segment address="0x10000000"/>
30363 Another simple memory map, with one loaded library with three
30364 allocated sections (.text, .data, .bss), looks like this:
30368 <library name="sharedlib.o">
30369 <section address="0x10000000"/>
30370 <section address="0x20000000"/>
30371 <section address="0x30000000"/>
30376 The format of a library list is described by this DTD:
30379 <!-- library-list: Root element with versioning -->
30380 <!ELEMENT library-list (library)*>
30381 <!ATTLIST library-list version CDATA #FIXED "1.0">
30382 <!ELEMENT library (segment*, section*)>
30383 <!ATTLIST library name CDATA #REQUIRED>
30384 <!ELEMENT segment EMPTY>
30385 <!ATTLIST segment address CDATA #REQUIRED>
30386 <!ELEMENT section EMPTY>
30387 <!ATTLIST section address CDATA #REQUIRED>
30390 In addition, segments and section descriptors cannot be mixed within a
30391 single library element, and you must supply at least one segment or
30392 section for each library.
30394 @node Memory Map Format
30395 @section Memory Map Format
30396 @cindex memory map format
30398 To be able to write into flash memory, @value{GDBN} needs to obtain a
30399 memory map from the target. This section describes the format of the
30402 The memory map is obtained using the @samp{qXfer:memory-map:read}
30403 (@pxref{qXfer memory map read}) packet and is an XML document that
30404 lists memory regions.
30406 @value{GDBN} must be linked with the Expat library to support XML
30407 memory maps. @xref{Expat}.
30409 The top-level structure of the document is shown below:
30412 <?xml version="1.0"?>
30413 <!DOCTYPE memory-map
30414 PUBLIC "+//IDN gnu.org//DTD GDB Memory Map V1.0//EN"
30415 "http://sourceware.org/gdb/gdb-memory-map.dtd">
30421 Each region can be either:
30426 A region of RAM starting at @var{addr} and extending for @var{length}
30430 <memory type="ram" start="@var{addr}" length="@var{length}"/>
30435 A region of read-only memory:
30438 <memory type="rom" start="@var{addr}" length="@var{length}"/>
30443 A region of flash memory, with erasure blocks @var{blocksize}
30447 <memory type="flash" start="@var{addr}" length="@var{length}">
30448 <property name="blocksize">@var{blocksize}</property>
30454 Regions must not overlap. @value{GDBN} assumes that areas of memory not covered
30455 by the memory map are RAM, and uses the ordinary @samp{M} and @samp{X}
30456 packets to write to addresses in such ranges.
30458 The formal DTD for memory map format is given below:
30461 <!-- ................................................... -->
30462 <!-- Memory Map XML DTD ................................ -->
30463 <!-- File: memory-map.dtd .............................. -->
30464 <!-- .................................... .............. -->
30465 <!-- memory-map.dtd -->
30466 <!-- memory-map: Root element with versioning -->
30467 <!ELEMENT memory-map (memory | property)>
30468 <!ATTLIST memory-map version CDATA #FIXED "1.0.0">
30469 <!ELEMENT memory (property)>
30470 <!-- memory: Specifies a memory region,
30471 and its type, or device. -->
30472 <!ATTLIST memory type CDATA #REQUIRED
30473 start CDATA #REQUIRED
30474 length CDATA #REQUIRED
30475 device CDATA #IMPLIED>
30476 <!-- property: Generic attribute tag -->
30477 <!ELEMENT property (#PCDATA | property)*>
30478 <!ATTLIST property name CDATA #REQUIRED>
30481 @include agentexpr.texi
30483 @node Target Descriptions
30484 @appendix Target Descriptions
30485 @cindex target descriptions
30487 @strong{Warning:} target descriptions are still under active development,
30488 and the contents and format may change between @value{GDBN} releases.
30489 The format is expected to stabilize in the future.
30491 One of the challenges of using @value{GDBN} to debug embedded systems
30492 is that there are so many minor variants of each processor
30493 architecture in use. It is common practice for vendors to start with
30494 a standard processor core --- ARM, PowerPC, or MIPS, for example ---
30495 and then make changes to adapt it to a particular market niche. Some
30496 architectures have hundreds of variants, available from dozens of
30497 vendors. This leads to a number of problems:
30501 With so many different customized processors, it is difficult for
30502 the @value{GDBN} maintainers to keep up with the changes.
30504 Since individual variants may have short lifetimes or limited
30505 audiences, it may not be worthwhile to carry information about every
30506 variant in the @value{GDBN} source tree.
30508 When @value{GDBN} does support the architecture of the embedded system
30509 at hand, the task of finding the correct architecture name to give the
30510 @command{set architecture} command can be error-prone.
30513 To address these problems, the @value{GDBN} remote protocol allows a
30514 target system to not only identify itself to @value{GDBN}, but to
30515 actually describe its own features. This lets @value{GDBN} support
30516 processor variants it has never seen before --- to the extent that the
30517 descriptions are accurate, and that @value{GDBN} understands them.
30519 @value{GDBN} must be linked with the Expat library to support XML
30520 target descriptions. @xref{Expat}.
30523 * Retrieving Descriptions:: How descriptions are fetched from a target.
30524 * Target Description Format:: The contents of a target description.
30525 * Predefined Target Types:: Standard types available for target
30527 * Standard Target Features:: Features @value{GDBN} knows about.
30530 @node Retrieving Descriptions
30531 @section Retrieving Descriptions
30533 Target descriptions can be read from the target automatically, or
30534 specified by the user manually. The default behavior is to read the
30535 description from the target. @value{GDBN} retrieves it via the remote
30536 protocol using @samp{qXfer} requests (@pxref{General Query Packets,
30537 qXfer}). The @var{annex} in the @samp{qXfer} packet will be
30538 @samp{target.xml}. The contents of the @samp{target.xml} annex are an
30539 XML document, of the form described in @ref{Target Description
30542 Alternatively, you can specify a file to read for the target description.
30543 If a file is set, the target will not be queried. The commands to
30544 specify a file are:
30547 @cindex set tdesc filename
30548 @item set tdesc filename @var{path}
30549 Read the target description from @var{path}.
30551 @cindex unset tdesc filename
30552 @item unset tdesc filename
30553 Do not read the XML target description from a file. @value{GDBN}
30554 will use the description supplied by the current target.
30556 @cindex show tdesc filename
30557 @item show tdesc filename
30558 Show the filename to read for a target description, if any.
30562 @node Target Description Format
30563 @section Target Description Format
30564 @cindex target descriptions, XML format
30566 A target description annex is an @uref{http://www.w3.org/XML/, XML}
30567 document which complies with the Document Type Definition provided in
30568 the @value{GDBN} sources in @file{gdb/features/gdb-target.dtd}. This
30569 means you can use generally available tools like @command{xmllint} to
30570 check that your feature descriptions are well-formed and valid.
30571 However, to help people unfamiliar with XML write descriptions for
30572 their targets, we also describe the grammar here.
30574 Target descriptions can identify the architecture of the remote target
30575 and (for some architectures) provide information about custom register
30576 sets. @value{GDBN} can use this information to autoconfigure for your
30577 target, or to warn you if you connect to an unsupported target.
30579 Here is a simple target description:
30582 <target version="1.0">
30583 <architecture>i386:x86-64</architecture>
30588 This minimal description only says that the target uses
30589 the x86-64 architecture.
30591 A target description has the following overall form, with [ ] marking
30592 optional elements and @dots{} marking repeatable elements. The elements
30593 are explained further below.
30596 <?xml version="1.0"?>
30597 <!DOCTYPE target SYSTEM "gdb-target.dtd">
30598 <target version="1.0">
30599 @r{[}@var{architecture}@r{]}
30600 @r{[}@var{feature}@dots{}@r{]}
30605 The description is generally insensitive to whitespace and line
30606 breaks, under the usual common-sense rules. The XML version
30607 declaration and document type declaration can generally be omitted
30608 (@value{GDBN} does not require them), but specifying them may be
30609 useful for XML validation tools. The @samp{version} attribute for
30610 @samp{<target>} may also be omitted, but we recommend
30611 including it; if future versions of @value{GDBN} use an incompatible
30612 revision of @file{gdb-target.dtd}, they will detect and report
30613 the version mismatch.
30615 @subsection Inclusion
30616 @cindex target descriptions, inclusion
30619 @cindex <xi:include>
30622 It can sometimes be valuable to split a target description up into
30623 several different annexes, either for organizational purposes, or to
30624 share files between different possible target descriptions. You can
30625 divide a description into multiple files by replacing any element of
30626 the target description with an inclusion directive of the form:
30629 <xi:include href="@var{document}"/>
30633 When @value{GDBN} encounters an element of this form, it will retrieve
30634 the named XML @var{document}, and replace the inclusion directive with
30635 the contents of that document. If the current description was read
30636 using @samp{qXfer}, then so will be the included document;
30637 @var{document} will be interpreted as the name of an annex. If the
30638 current description was read from a file, @value{GDBN} will look for
30639 @var{document} as a file in the same directory where it found the
30640 original description.
30642 @subsection Architecture
30643 @cindex <architecture>
30645 An @samp{<architecture>} element has this form:
30648 <architecture>@var{arch}</architecture>
30651 @var{arch} is an architecture name from the same selection
30652 accepted by @code{set architecture} (@pxref{Targets, ,Specifying a
30653 Debugging Target}).
30655 @subsection Features
30658 Each @samp{<feature>} describes some logical portion of the target
30659 system. Features are currently used to describe available CPU
30660 registers and the types of their contents. A @samp{<feature>} element
30664 <feature name="@var{name}">
30665 @r{[}@var{type}@dots{}@r{]}
30671 Each feature's name should be unique within the description. The name
30672 of a feature does not matter unless @value{GDBN} has some special
30673 knowledge of the contents of that feature; if it does, the feature
30674 should have its standard name. @xref{Standard Target Features}.
30678 Any register's value is a collection of bits which @value{GDBN} must
30679 interpret. The default interpretation is a two's complement integer,
30680 but other types can be requested by name in the register description.
30681 Some predefined types are provided by @value{GDBN} (@pxref{Predefined
30682 Target Types}), and the description can define additional composite types.
30684 Each type element must have an @samp{id} attribute, which gives
30685 a unique (within the containing @samp{<feature>}) name to the type.
30686 Types must be defined before they are used.
30689 Some targets offer vector registers, which can be treated as arrays
30690 of scalar elements. These types are written as @samp{<vector>} elements,
30691 specifying the array element type, @var{type}, and the number of elements,
30695 <vector id="@var{id}" type="@var{type}" count="@var{count}"/>
30699 If a register's value is usefully viewed in multiple ways, define it
30700 with a union type containing the useful representations. The
30701 @samp{<union>} element contains one or more @samp{<field>} elements,
30702 each of which has a @var{name} and a @var{type}:
30705 <union id="@var{id}">
30706 <field name="@var{name}" type="@var{type}"/>
30711 @subsection Registers
30714 Each register is represented as an element with this form:
30717 <reg name="@var{name}"
30718 bitsize="@var{size}"
30719 @r{[}regnum="@var{num}"@r{]}
30720 @r{[}save-restore="@var{save-restore}"@r{]}
30721 @r{[}type="@var{type}"@r{]}
30722 @r{[}group="@var{group}"@r{]}/>
30726 The components are as follows:
30731 The register's name; it must be unique within the target description.
30734 The register's size, in bits.
30737 The register's number. If omitted, a register's number is one greater
30738 than that of the previous register (either in the current feature or in
30739 a preceeding feature); the first register in the target description
30740 defaults to zero. This register number is used to read or write
30741 the register; e.g.@: it is used in the remote @code{p} and @code{P}
30742 packets, and registers appear in the @code{g} and @code{G} packets
30743 in order of increasing register number.
30746 Whether the register should be preserved across inferior function
30747 calls; this must be either @code{yes} or @code{no}. The default is
30748 @code{yes}, which is appropriate for most registers except for
30749 some system control registers; this is not related to the target's
30753 The type of the register. @var{type} may be a predefined type, a type
30754 defined in the current feature, or one of the special types @code{int}
30755 and @code{float}. @code{int} is an integer type of the correct size
30756 for @var{bitsize}, and @code{float} is a floating point type (in the
30757 architecture's normal floating point format) of the correct size for
30758 @var{bitsize}. The default is @code{int}.
30761 The register group to which this register belongs. @var{group} must
30762 be either @code{general}, @code{float}, or @code{vector}. If no
30763 @var{group} is specified, @value{GDBN} will not display the register
30764 in @code{info registers}.
30768 @node Predefined Target Types
30769 @section Predefined Target Types
30770 @cindex target descriptions, predefined types
30772 Type definitions in the self-description can build up composite types
30773 from basic building blocks, but can not define fundamental types. Instead,
30774 standard identifiers are provided by @value{GDBN} for the fundamental
30775 types. The currently supported types are:
30784 Signed integer types holding the specified number of bits.
30791 Unsigned integer types holding the specified number of bits.
30795 Pointers to unspecified code and data. The program counter and
30796 any dedicated return address register may be marked as code
30797 pointers; printing a code pointer converts it into a symbolic
30798 address. The stack pointer and any dedicated address registers
30799 may be marked as data pointers.
30802 Single precision IEEE floating point.
30805 Double precision IEEE floating point.
30808 The 12-byte extended precision format used by ARM FPA registers.
30812 @node Standard Target Features
30813 @section Standard Target Features
30814 @cindex target descriptions, standard features
30816 A target description must contain either no registers or all the
30817 target's registers. If the description contains no registers, then
30818 @value{GDBN} will assume a default register layout, selected based on
30819 the architecture. If the description contains any registers, the
30820 default layout will not be used; the standard registers must be
30821 described in the target description, in such a way that @value{GDBN}
30822 can recognize them.
30824 This is accomplished by giving specific names to feature elements
30825 which contain standard registers. @value{GDBN} will look for features
30826 with those names and verify that they contain the expected registers;
30827 if any known feature is missing required registers, or if any required
30828 feature is missing, @value{GDBN} will reject the target
30829 description. You can add additional registers to any of the
30830 standard features --- @value{GDBN} will display them just as if
30831 they were added to an unrecognized feature.
30833 This section lists the known features and their expected contents.
30834 Sample XML documents for these features are included in the
30835 @value{GDBN} source tree, in the directory @file{gdb/features}.
30837 Names recognized by @value{GDBN} should include the name of the
30838 company or organization which selected the name, and the overall
30839 architecture to which the feature applies; so e.g.@: the feature
30840 containing ARM core registers is named @samp{org.gnu.gdb.arm.core}.
30842 The names of registers are not case sensitive for the purpose
30843 of recognizing standard features, but @value{GDBN} will only display
30844 registers using the capitalization used in the description.
30850 * PowerPC Features::
30855 @subsection ARM Features
30856 @cindex target descriptions, ARM features
30858 The @samp{org.gnu.gdb.arm.core} feature is required for ARM targets.
30859 It should contain registers @samp{r0} through @samp{r13}, @samp{sp},
30860 @samp{lr}, @samp{pc}, and @samp{cpsr}.
30862 The @samp{org.gnu.gdb.arm.fpa} feature is optional. If present, it
30863 should contain registers @samp{f0} through @samp{f7} and @samp{fps}.
30865 The @samp{org.gnu.gdb.xscale.iwmmxt} feature is optional. If present,
30866 it should contain at least registers @samp{wR0} through @samp{wR15} and
30867 @samp{wCGR0} through @samp{wCGR3}. The @samp{wCID}, @samp{wCon},
30868 @samp{wCSSF}, and @samp{wCASF} registers are optional.
30870 @node MIPS Features
30871 @subsection MIPS Features
30872 @cindex target descriptions, MIPS features
30874 The @samp{org.gnu.gdb.mips.cpu} feature is required for MIPS targets.
30875 It should contain registers @samp{r0} through @samp{r31}, @samp{lo},
30876 @samp{hi}, and @samp{pc}. They may be 32-bit or 64-bit depending
30879 The @samp{org.gnu.gdb.mips.cp0} feature is also required. It should
30880 contain at least the @samp{status}, @samp{badvaddr}, and @samp{cause}
30881 registers. They may be 32-bit or 64-bit depending on the target.
30883 The @samp{org.gnu.gdb.mips.fpu} feature is currently required, though
30884 it may be optional in a future version of @value{GDBN}. It should
30885 contain registers @samp{f0} through @samp{f31}, @samp{fcsr}, and
30886 @samp{fir}. They may be 32-bit or 64-bit depending on the target.
30888 The @samp{org.gnu.gdb.mips.linux} feature is optional. It should
30889 contain a single register, @samp{restart}, which is used by the
30890 Linux kernel to control restartable syscalls.
30892 @node M68K Features
30893 @subsection M68K Features
30894 @cindex target descriptions, M68K features
30897 @item @samp{org.gnu.gdb.m68k.core}
30898 @itemx @samp{org.gnu.gdb.coldfire.core}
30899 @itemx @samp{org.gnu.gdb.fido.core}
30900 One of those features must be always present.
30901 The feature that is present determines which flavor of m68k is
30902 used. The feature that is present should contain registers
30903 @samp{d0} through @samp{d7}, @samp{a0} through @samp{a5}, @samp{fp},
30904 @samp{sp}, @samp{ps} and @samp{pc}.
30906 @item @samp{org.gnu.gdb.coldfire.fp}
30907 This feature is optional. If present, it should contain registers
30908 @samp{fp0} through @samp{fp7}, @samp{fpcontrol}, @samp{fpstatus} and
30912 @node PowerPC Features
30913 @subsection PowerPC Features
30914 @cindex target descriptions, PowerPC features
30916 The @samp{org.gnu.gdb.power.core} feature is required for PowerPC
30917 targets. It should contain registers @samp{r0} through @samp{r31},
30918 @samp{pc}, @samp{msr}, @samp{cr}, @samp{lr}, @samp{ctr}, and
30919 @samp{xer}. They may be 32-bit or 64-bit depending on the target.
30921 The @samp{org.gnu.gdb.power.fpu} feature is optional. It should
30922 contain registers @samp{f0} through @samp{f31} and @samp{fpscr}.
30924 The @samp{org.gnu.gdb.power.altivec} feature is optional. It should
30925 contain registers @samp{vr0} through @samp{vr31}, @samp{vscr},
30928 The @samp{org.gnu.gdb.power.vsx} feature is optional. It should
30929 contain registers @samp{vs0h} through @samp{vs31h}. @value{GDBN}
30930 will combine these registers with the floating point registers
30931 (@samp{f0} through @samp{f31}) and the altivec registers (@samp{vr0}
30932 through @samp{vr31}) to present the 128-bit wide registers @samp{vs0}
30933 through @samp{vs63}, the set of vector registers for POWER7.
30935 The @samp{org.gnu.gdb.power.spe} feature is optional. It should
30936 contain registers @samp{ev0h} through @samp{ev31h}, @samp{acc}, and
30937 @samp{spefscr}. SPE targets should provide 32-bit registers in
30938 @samp{org.gnu.gdb.power.core} and provide the upper halves in
30939 @samp{ev0h} through @samp{ev31h}. @value{GDBN} will combine
30940 these to present registers @samp{ev0} through @samp{ev31} to the
30943 @node Operating System Information
30944 @appendix Operating System Information
30945 @cindex operating system information
30951 Users of @value{GDBN} often wish to obtain information about the state of
30952 the operating system running on the target---for example the list of
30953 processes, or the list of open files. This section describes the
30954 mechanism that makes it possible. This mechanism is similar to the
30955 target features mechanism (@pxref{Target Descriptions}), but focuses
30956 on a different aspect of target.
30958 Operating system information is retrived from the target via the
30959 remote protocol, using @samp{qXfer} requests (@pxref{qXfer osdata
30960 read}). The object name in the request should be @samp{osdata}, and
30961 the @var{annex} identifies the data to be fetched.
30964 @appendixsection Process list
30965 @cindex operating system information, process list
30967 When requesting the process list, the @var{annex} field in the
30968 @samp{qXfer} request should be @samp{processes}. The returned data is
30969 an XML document. The formal syntax of this document is defined in
30970 @file{gdb/features/osdata.dtd}.
30972 An example document is:
30975 <?xml version="1.0"?>
30976 <!DOCTYPE target SYSTEM "osdata.dtd">
30977 <osdata type="processes">
30979 <column name="pid">1</column>
30980 <column name="user">root</column>
30981 <column name="command">/sbin/init</column>
30986 Each item should include a column whose name is @samp{pid}. The value
30987 of that column should identify the process on the target. The
30988 @samp{user} and @samp{command} columns are optional, and will be
30989 displayed by @value{GDBN}. Target may provide additional columns,
30990 which @value{GDBN} currently ignores.
31004 % I think something like @colophon should be in texinfo. In the
31006 \long\def\colophon{\hbox to0pt{}\vfill
31007 \centerline{The body of this manual is set in}
31008 \centerline{\fontname\tenrm,}
31009 \centerline{with headings in {\bf\fontname\tenbf}}
31010 \centerline{and examples in {\tt\fontname\tentt}.}
31011 \centerline{{\it\fontname\tenit\/},}
31012 \centerline{{\bf\fontname\tenbf}, and}
31013 \centerline{{\sl\fontname\tensl\/}}
31014 \centerline{are used for emphasis.}\vfill}
31016 % Blame: doc@cygnus.com, 1991.