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 * Optimized Code:: Debugging optimized code
143 * Macros:: Preprocessor Macros
144 * Tracepoints:: Debugging remote targets non-intrusively
145 * Overlays:: Debugging programs that use overlays
147 * Languages:: Using @value{GDBN} with different languages
149 * Symbols:: Examining the symbol table
150 * Altering:: Altering execution
151 * GDB Files:: @value{GDBN} files
152 * Targets:: Specifying a debugging target
153 * Remote Debugging:: Debugging remote programs
154 * Configurations:: Configuration-specific information
155 * Controlling GDB:: Controlling @value{GDBN}
156 * Extending GDB:: Extending @value{GDBN}
157 * Interpreters:: Command Interpreters
158 * TUI:: @value{GDBN} Text User Interface
159 * Emacs:: Using @value{GDBN} under @sc{gnu} Emacs
160 * GDB/MI:: @value{GDBN}'s Machine Interface.
161 * Annotations:: @value{GDBN}'s annotation interface.
163 * GDB Bugs:: Reporting bugs in @value{GDBN}
165 * Command Line Editing:: Command Line Editing
166 * Using History Interactively:: Using History Interactively
167 * Formatting Documentation:: How to format and print @value{GDBN} documentation
168 * Installing GDB:: Installing GDB
169 * Maintenance Commands:: Maintenance Commands
170 * Remote Protocol:: GDB Remote Serial Protocol
171 * Agent Expressions:: The GDB Agent Expression Mechanism
172 * Target Descriptions:: How targets can describe themselves to
174 * Operating System Information:: Getting additional information from
176 * Copying:: GNU General Public License says
177 how you can copy and share GDB
178 * GNU Free Documentation License:: The license for this documentation
187 @unnumbered Summary of @value{GDBN}
189 The purpose of a debugger such as @value{GDBN} is to allow you to see what is
190 going on ``inside'' another program while it executes---or what another
191 program was doing at the moment it crashed.
193 @value{GDBN} can do four main kinds of things (plus other things in support of
194 these) to help you catch bugs in the act:
198 Start your program, specifying anything that might affect its behavior.
201 Make your program stop on specified conditions.
204 Examine what has happened, when your program has stopped.
207 Change things in your program, so you can experiment with correcting the
208 effects of one bug and go on to learn about another.
211 You can use @value{GDBN} to debug programs written in C and C@t{++}.
212 For more information, see @ref{Supported Languages,,Supported Languages}.
213 For more information, see @ref{C,,C and C++}.
216 Support for Modula-2 is partial. For information on Modula-2, see
217 @ref{Modula-2,,Modula-2}.
220 Debugging Pascal programs which use sets, subranges, file variables, or
221 nested functions does not currently work. @value{GDBN} does not support
222 entering expressions, printing values, or similar features using Pascal
226 @value{GDBN} can be used to debug programs written in Fortran, although
227 it may be necessary to refer to some variables with a trailing
230 @value{GDBN} can be used to debug programs written in Objective-C,
231 using either the Apple/NeXT or the GNU Objective-C runtime.
234 * Free Software:: Freely redistributable software
235 * Contributors:: Contributors to GDB
239 @unnumberedsec Free Software
241 @value{GDBN} is @dfn{free software}, protected by the @sc{gnu}
242 General Public License
243 (GPL). The GPL gives you the freedom to copy or adapt a licensed
244 program---but every person getting a copy also gets with it the
245 freedom to modify that copy (which means that they must get access to
246 the source code), and the freedom to distribute further copies.
247 Typical software companies use copyrights to limit your freedoms; the
248 Free Software Foundation uses the GPL to preserve these freedoms.
250 Fundamentally, the General Public License is a license which says that
251 you have these freedoms and that you cannot take these freedoms away
254 @unnumberedsec Free Software Needs Free Documentation
256 The biggest deficiency in the free software community today is not in
257 the software---it is the lack of good free documentation that we can
258 include with the free software. Many of our most important
259 programs do not come with free reference manuals and free introductory
260 texts. Documentation is an essential part of any software package;
261 when an important free software package does not come with a free
262 manual and a free tutorial, that is a major gap. We have many such
265 Consider Perl, for instance. The tutorial manuals that people
266 normally use are non-free. How did this come about? Because the
267 authors of those manuals published them with restrictive terms---no
268 copying, no modification, source files not available---which exclude
269 them from the free software world.
271 That wasn't the first time this sort of thing happened, and it was far
272 from the last. Many times we have heard a GNU user eagerly describe a
273 manual that he is writing, his intended contribution to the community,
274 only to learn that he had ruined everything by signing a publication
275 contract to make it non-free.
277 Free documentation, like free software, is a matter of freedom, not
278 price. The problem with the non-free manual is not that publishers
279 charge a price for printed copies---that in itself is fine. (The Free
280 Software Foundation sells printed copies of manuals, too.) The
281 problem is the restrictions on the use of the manual. Free manuals
282 are available in source code form, and give you permission to copy and
283 modify. Non-free manuals do not allow this.
285 The criteria of freedom for a free manual are roughly the same as for
286 free software. Redistribution (including the normal kinds of
287 commercial redistribution) must be permitted, so that the manual can
288 accompany every copy of the program, both on-line and on paper.
290 Permission for modification of the technical content is crucial too.
291 When people modify the software, adding or changing features, if they
292 are conscientious they will change the manual too---so they can
293 provide accurate and clear documentation for the modified program. A
294 manual that leaves you no choice but to write a new manual to document
295 a changed version of the program is not really available to our
298 Some kinds of limits on the way modification is handled are
299 acceptable. For example, requirements to preserve the original
300 author's copyright notice, the distribution terms, or the list of
301 authors, are ok. It is also no problem to require modified versions
302 to include notice that they were modified. Even entire sections that
303 may not be deleted or changed are acceptable, as long as they deal
304 with nontechnical topics (like this one). These kinds of restrictions
305 are acceptable because they don't obstruct the community's normal use
308 However, it must be possible to modify all the @emph{technical}
309 content of the manual, and then distribute the result in all the usual
310 media, through all the usual channels. Otherwise, the restrictions
311 obstruct the use of the manual, it is not free, and we need another
312 manual to replace it.
314 Please spread the word about this issue. Our community continues to
315 lose manuals to proprietary publishing. If we spread the word that
316 free software needs free reference manuals and free tutorials, perhaps
317 the next person who wants to contribute by writing documentation will
318 realize, before it is too late, that only free manuals contribute to
319 the free software community.
321 If you are writing documentation, please insist on publishing it under
322 the GNU Free Documentation License or another free documentation
323 license. Remember that this decision requires your approval---you
324 don't have to let the publisher decide. Some commercial publishers
325 will use a free license if you insist, but they will not propose the
326 option; it is up to you to raise the issue and say firmly that this is
327 what you want. If the publisher you are dealing with refuses, please
328 try other publishers. If you're not sure whether a proposed license
329 is free, write to @email{licensing@@gnu.org}.
331 You can encourage commercial publishers to sell more free, copylefted
332 manuals and tutorials by buying them, and particularly by buying
333 copies from the publishers that paid for their writing or for major
334 improvements. Meanwhile, try to avoid buying non-free documentation
335 at all. Check the distribution terms of a manual before you buy it,
336 and insist that whoever seeks your business must respect your freedom.
337 Check the history of the book, and try to reward the publishers that
338 have paid or pay the authors to work on it.
340 The Free Software Foundation maintains a list of free documentation
341 published by other publishers, at
342 @url{http://www.fsf.org/doc/other-free-books.html}.
345 @unnumberedsec Contributors to @value{GDBN}
347 Richard Stallman was the original author of @value{GDBN}, and of many
348 other @sc{gnu} programs. Many others have contributed to its
349 development. This section attempts to credit major contributors. One
350 of the virtues of free software is that everyone is free to contribute
351 to it; with regret, we cannot actually acknowledge everyone here. The
352 file @file{ChangeLog} in the @value{GDBN} distribution approximates a
353 blow-by-blow account.
355 Changes much prior to version 2.0 are lost in the mists of time.
358 @emph{Plea:} Additions to this section are particularly welcome. If you
359 or your friends (or enemies, to be evenhanded) have been unfairly
360 omitted from this list, we would like to add your names!
363 So that they may not regard their many labors as thankless, we
364 particularly thank those who shepherded @value{GDBN} through major
366 Andrew Cagney (releases 6.3, 6.2, 6.1, 6.0, 5.3, 5.2, 5.1 and 5.0);
367 Jim Blandy (release 4.18);
368 Jason Molenda (release 4.17);
369 Stan Shebs (release 4.14);
370 Fred Fish (releases 4.16, 4.15, 4.13, 4.12, 4.11, 4.10, and 4.9);
371 Stu Grossman and John Gilmore (releases 4.8, 4.7, 4.6, 4.5, and 4.4);
372 John Gilmore (releases 4.3, 4.2, 4.1, 4.0, and 3.9);
373 Jim Kingdon (releases 3.5, 3.4, and 3.3);
374 and Randy Smith (releases 3.2, 3.1, and 3.0).
376 Richard Stallman, assisted at various times by Peter TerMaat, Chris
377 Hanson, and Richard Mlynarik, handled releases through 2.8.
379 Michael Tiemann is the author of most of the @sc{gnu} C@t{++} support
380 in @value{GDBN}, with significant additional contributions from Per
381 Bothner and Daniel Berlin. James Clark wrote the @sc{gnu} C@t{++}
382 demangler. Early work on C@t{++} was by Peter TerMaat (who also did
383 much general update work leading to release 3.0).
385 @value{GDBN} uses the BFD subroutine library to examine multiple
386 object-file formats; BFD was a joint project of David V.
387 Henkel-Wallace, Rich Pixley, Steve Chamberlain, and John Gilmore.
389 David Johnson wrote the original COFF support; Pace Willison did
390 the original support for encapsulated COFF.
392 Brent Benson of Harris Computer Systems contributed DWARF 2 support.
394 Adam de Boor and Bradley Davis contributed the ISI Optimum V support.
395 Per Bothner, Noboyuki Hikichi, and Alessandro Forin contributed MIPS
397 Jean-Daniel Fekete contributed Sun 386i support.
398 Chris Hanson improved the HP9000 support.
399 Noboyuki Hikichi and Tomoyuki Hasei contributed Sony/News OS 3 support.
400 David Johnson contributed Encore Umax support.
401 Jyrki Kuoppala contributed Altos 3068 support.
402 Jeff Law contributed HP PA and SOM support.
403 Keith Packard contributed NS32K support.
404 Doug Rabson contributed Acorn Risc Machine support.
405 Bob Rusk contributed Harris Nighthawk CX-UX support.
406 Chris Smith contributed Convex support (and Fortran debugging).
407 Jonathan Stone contributed Pyramid support.
408 Michael Tiemann contributed SPARC support.
409 Tim Tucker contributed support for the Gould NP1 and Gould Powernode.
410 Pace Willison contributed Intel 386 support.
411 Jay Vosburgh contributed Symmetry support.
412 Marko Mlinar contributed OpenRISC 1000 support.
414 Andreas Schwab contributed M68K @sc{gnu}/Linux support.
416 Rich Schaefer and Peter Schauer helped with support of SunOS shared
419 Jay Fenlason and Roland McGrath ensured that @value{GDBN} and GAS agree
420 about several machine instruction sets.
422 Patrick Duval, Ted Goldstein, Vikram Koka and Glenn Engel helped develop
423 remote debugging. Intel Corporation, Wind River Systems, AMD, and ARM
424 contributed remote debugging modules for the i960, VxWorks, A29K UDI,
425 and RDI targets, respectively.
427 Brian Fox is the author of the readline libraries providing
428 command-line editing and command history.
430 Andrew Beers of SUNY Buffalo wrote the language-switching code, the
431 Modula-2 support, and contributed the Languages chapter of this manual.
433 Fred Fish wrote most of the support for Unix System Vr4.
434 He also enhanced the command-completion support to cover C@t{++} overloaded
437 Hitachi America (now Renesas America), Ltd. sponsored the support for
438 H8/300, H8/500, and Super-H processors.
440 NEC sponsored the support for the v850, Vr4xxx, and Vr5xxx processors.
442 Mitsubishi (now Renesas) sponsored the support for D10V, D30V, and M32R/D
445 Toshiba sponsored the support for the TX39 Mips processor.
447 Matsushita sponsored the support for the MN10200 and MN10300 processors.
449 Fujitsu sponsored the support for SPARClite and FR30 processors.
451 Kung Hsu, Jeff Law, and Rick Sladkey added support for hardware
454 Michael Snyder added support for tracepoints.
456 Stu Grossman wrote gdbserver.
458 Jim Kingdon, Peter Schauer, Ian Taylor, and Stu Grossman made
459 nearly innumerable bug fixes and cleanups throughout @value{GDBN}.
461 The following people at the Hewlett-Packard Company contributed
462 support for the PA-RISC 2.0 architecture, HP-UX 10.20, 10.30, and 11.0
463 (narrow mode), HP's implementation of kernel threads, HP's aC@t{++}
464 compiler, and the Text User Interface (nee Terminal User Interface):
465 Ben Krepp, Richard Title, John Bishop, Susan Macchia, Kathy Mann,
466 Satish Pai, India Paul, Steve Rehrauer, and Elena Zannoni. Kim Haase
467 provided HP-specific information in this manual.
469 DJ Delorie ported @value{GDBN} to MS-DOS, for the DJGPP project.
470 Robert Hoehne made significant contributions to the DJGPP port.
472 Cygnus Solutions has sponsored @value{GDBN} maintenance and much of its
473 development since 1991. Cygnus engineers who have worked on @value{GDBN}
474 fulltime include Mark Alexander, Jim Blandy, Per Bothner, Kevin
475 Buettner, Edith Epstein, Chris Faylor, Fred Fish, Martin Hunt, Jim
476 Ingham, John Gilmore, Stu Grossman, Kung Hsu, Jim Kingdon, John Metzler,
477 Fernando Nasser, Geoffrey Noer, Dawn Perchik, Rich Pixley, Zdenek
478 Radouch, Keith Seitz, Stan Shebs, David Taylor, and Elena Zannoni. In
479 addition, Dave Brolley, Ian Carmichael, Steve Chamberlain, Nick Clifton,
480 JT Conklin, Stan Cox, DJ Delorie, Ulrich Drepper, Frank Eigler, Doug
481 Evans, Sean Fagan, David Henkel-Wallace, Richard Henderson, Jeff
482 Holcomb, Jeff Law, Jim Lemke, Tom Lord, Bob Manson, Michael Meissner,
483 Jason Merrill, Catherine Moore, Drew Moseley, Ken Raeburn, Gavin
484 Romig-Koch, Rob Savoye, Jamie Smith, Mike Stump, Ian Taylor, Angela
485 Thomas, Michael Tiemann, Tom Tromey, Ron Unrau, Jim Wilson, and David
486 Zuhn have made contributions both large and small.
488 Andrew Cagney, Fernando Nasser, and Elena Zannoni, while working for
489 Cygnus Solutions, implemented the original @sc{gdb/mi} interface.
491 Jim Blandy added support for preprocessor macros, while working for Red
494 Andrew Cagney designed @value{GDBN}'s architecture vector. Many
495 people including Andrew Cagney, Stephane Carrez, Randolph Chung, Nick
496 Duffek, Richard Henderson, Mark Kettenis, Grace Sainsbury, Kei
497 Sakamoto, Yoshinori Sato, Michael Snyder, Andreas Schwab, Jason
498 Thorpe, Corinna Vinschen, Ulrich Weigand, and Elena Zannoni, helped
499 with the migration of old architectures to this new framework.
501 Andrew Cagney completely re-designed and re-implemented @value{GDBN}'s
502 unwinder framework, this consisting of a fresh new design featuring
503 frame IDs, independent frame sniffers, and the sentinel frame. Mark
504 Kettenis implemented the @sc{dwarf 2} unwinder, Jeff Johnston the
505 libunwind unwinder, and Andrew Cagney the dummy, sentinel, tramp, and
506 trad unwinders. The architecture-specific changes, each involving a
507 complete rewrite of the architecture's frame code, were carried out by
508 Jim Blandy, Joel Brobecker, Kevin Buettner, Andrew Cagney, Stephane
509 Carrez, Randolph Chung, Orjan Friberg, Richard Henderson, Daniel
510 Jacobowitz, Jeff Johnston, Mark Kettenis, Theodore A. Roth, Kei
511 Sakamoto, Yoshinori Sato, Michael Snyder, Corinna Vinschen, and Ulrich
514 Christian Zankel, Ross Morley, Bob Wilson, and Maxim Grigoriev from
515 Tensilica, Inc.@: contributed support for Xtensa processors. Others
516 who have worked on the Xtensa port of @value{GDBN} in the past include
517 Steve Tjiang, John Newlin, and Scott Foehner.
520 @chapter A Sample @value{GDBN} Session
522 You can use this manual at your leisure to read all about @value{GDBN}.
523 However, a handful of commands are enough to get started using the
524 debugger. This chapter illustrates those commands.
527 In this sample session, we emphasize user input like this: @b{input},
528 to make it easier to pick out from the surrounding output.
531 @c FIXME: this example may not be appropriate for some configs, where
532 @c FIXME...primary interest is in remote use.
534 One of the preliminary versions of @sc{gnu} @code{m4} (a generic macro
535 processor) exhibits the following bug: sometimes, when we change its
536 quote strings from the default, the commands used to capture one macro
537 definition within another stop working. In the following short @code{m4}
538 session, we define a macro @code{foo} which expands to @code{0000}; we
539 then use the @code{m4} built-in @code{defn} to define @code{bar} as the
540 same thing. However, when we change the open quote string to
541 @code{<QUOTE>} and the close quote string to @code{<UNQUOTE>}, the same
542 procedure fails to define a new synonym @code{baz}:
551 @b{define(bar,defn(`foo'))}
555 @b{changequote(<QUOTE>,<UNQUOTE>)}
557 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
560 m4: End of input: 0: fatal error: EOF in string
564 Let us use @value{GDBN} to try to see what is going on.
567 $ @b{@value{GDBP} m4}
568 @c FIXME: this falsifies the exact text played out, to permit smallbook
569 @c FIXME... format to come out better.
570 @value{GDBN} is free software and you are welcome to distribute copies
571 of it under certain conditions; type "show copying" to see
573 There is absolutely no warranty for @value{GDBN}; type "show warranty"
576 @value{GDBN} @value{GDBVN}, Copyright 1999 Free Software Foundation, Inc...
581 @value{GDBN} reads only enough symbol data to know where to find the
582 rest when needed; as a result, the first prompt comes up very quickly.
583 We now tell @value{GDBN} to use a narrower display width than usual, so
584 that examples fit in this manual.
587 (@value{GDBP}) @b{set width 70}
591 We need to see how the @code{m4} built-in @code{changequote} works.
592 Having looked at the source, we know the relevant subroutine is
593 @code{m4_changequote}, so we set a breakpoint there with the @value{GDBN}
594 @code{break} command.
597 (@value{GDBP}) @b{break m4_changequote}
598 Breakpoint 1 at 0x62f4: file builtin.c, line 879.
602 Using the @code{run} command, we start @code{m4} running under @value{GDBN}
603 control; as long as control does not reach the @code{m4_changequote}
604 subroutine, the program runs as usual:
607 (@value{GDBP}) @b{run}
608 Starting program: /work/Editorial/gdb/gnu/m4/m4
616 To trigger the breakpoint, we call @code{changequote}. @value{GDBN}
617 suspends execution of @code{m4}, displaying information about the
618 context where it stops.
621 @b{changequote(<QUOTE>,<UNQUOTE>)}
623 Breakpoint 1, m4_changequote (argc=3, argv=0x33c70)
625 879 if (bad_argc(TOKEN_DATA_TEXT(argv[0]),argc,1,3))
629 Now we use the command @code{n} (@code{next}) to advance execution to
630 the next line of the current function.
634 882 set_quotes((argc >= 2) ? TOKEN_DATA_TEXT(argv[1])\
639 @code{set_quotes} looks like a promising subroutine. We can go into it
640 by using the command @code{s} (@code{step}) instead of @code{next}.
641 @code{step} goes to the next line to be executed in @emph{any}
642 subroutine, so it steps into @code{set_quotes}.
646 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
648 530 if (lquote != def_lquote)
652 The display that shows the subroutine where @code{m4} is now
653 suspended (and its arguments) is called a stack frame display. It
654 shows a summary of the stack. We can use the @code{backtrace}
655 command (which can also be spelled @code{bt}), to see where we are
656 in the stack as a whole: the @code{backtrace} command displays a
657 stack frame for each active subroutine.
660 (@value{GDBP}) @b{bt}
661 #0 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
663 #1 0x6344 in m4_changequote (argc=3, argv=0x33c70)
665 #2 0x8174 in expand_macro (sym=0x33320) at macro.c:242
666 #3 0x7a88 in expand_token (obs=0x0, t=209696, td=0xf7fffa30)
668 #4 0x79dc in expand_input () at macro.c:40
669 #5 0x2930 in main (argc=0, argv=0xf7fffb20) at m4.c:195
673 We step through a few more lines to see what happens. The first two
674 times, we can use @samp{s}; the next two times we use @code{n} to avoid
675 falling into the @code{xstrdup} subroutine.
679 0x3b5c 532 if (rquote != def_rquote)
681 0x3b80 535 lquote = (lq == nil || *lq == '\0') ? \
682 def_lquote : xstrdup(lq);
684 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
687 538 len_lquote = strlen(rquote);
691 The last line displayed looks a little odd; we can examine the variables
692 @code{lquote} and @code{rquote} to see if they are in fact the new left
693 and right quotes we specified. We use the command @code{p}
694 (@code{print}) to see their values.
697 (@value{GDBP}) @b{p lquote}
698 $1 = 0x35d40 "<QUOTE>"
699 (@value{GDBP}) @b{p rquote}
700 $2 = 0x35d50 "<UNQUOTE>"
704 @code{lquote} and @code{rquote} are indeed the new left and right quotes.
705 To look at some context, we can display ten lines of source
706 surrounding the current line with the @code{l} (@code{list}) command.
712 535 lquote = (lq == nil || *lq == '\0') ? def_lquote\
714 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
717 538 len_lquote = strlen(rquote);
718 539 len_rquote = strlen(lquote);
725 Let us step past the two lines that set @code{len_lquote} and
726 @code{len_rquote}, and then examine the values of those variables.
730 539 len_rquote = strlen(lquote);
733 (@value{GDBP}) @b{p len_lquote}
735 (@value{GDBP}) @b{p len_rquote}
740 That certainly looks wrong, assuming @code{len_lquote} and
741 @code{len_rquote} are meant to be the lengths of @code{lquote} and
742 @code{rquote} respectively. We can set them to better values using
743 the @code{p} command, since it can print the value of
744 any expression---and that expression can include subroutine calls and
748 (@value{GDBP}) @b{p len_lquote=strlen(lquote)}
750 (@value{GDBP}) @b{p len_rquote=strlen(rquote)}
755 Is that enough to fix the problem of using the new quotes with the
756 @code{m4} built-in @code{defn}? We can allow @code{m4} to continue
757 executing with the @code{c} (@code{continue}) command, and then try the
758 example that caused trouble initially:
764 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
771 Success! The new quotes now work just as well as the default ones. The
772 problem seems to have been just the two typos defining the wrong
773 lengths. We allow @code{m4} exit by giving it an EOF as input:
777 Program exited normally.
781 The message @samp{Program exited normally.} is from @value{GDBN}; it
782 indicates @code{m4} has finished executing. We can end our @value{GDBN}
783 session with the @value{GDBN} @code{quit} command.
786 (@value{GDBP}) @b{quit}
790 @chapter Getting In and Out of @value{GDBN}
792 This chapter discusses how to start @value{GDBN}, and how to get out of it.
796 type @samp{@value{GDBP}} to start @value{GDBN}.
798 type @kbd{quit} or @kbd{Ctrl-d} to exit.
802 * Invoking GDB:: How to start @value{GDBN}
803 * Quitting GDB:: How to quit @value{GDBN}
804 * Shell Commands:: How to use shell commands inside @value{GDBN}
805 * Logging Output:: How to log @value{GDBN}'s output to a file
809 @section Invoking @value{GDBN}
811 Invoke @value{GDBN} by running the program @code{@value{GDBP}}. Once started,
812 @value{GDBN} reads commands from the terminal until you tell it to exit.
814 You can also run @code{@value{GDBP}} with a variety of arguments and options,
815 to specify more of your debugging environment at the outset.
817 The command-line options described here are designed
818 to cover a variety of situations; in some environments, some of these
819 options may effectively be unavailable.
821 The most usual way to start @value{GDBN} is with one argument,
822 specifying an executable program:
825 @value{GDBP} @var{program}
829 You can also start with both an executable program and a core file
833 @value{GDBP} @var{program} @var{core}
836 You can, instead, specify a process ID as a second argument, if you want
837 to debug a running process:
840 @value{GDBP} @var{program} 1234
844 would attach @value{GDBN} to process @code{1234} (unless you also have a file
845 named @file{1234}; @value{GDBN} does check for a core file first).
847 Taking advantage of the second command-line argument requires a fairly
848 complete operating system; when you use @value{GDBN} as a remote
849 debugger attached to a bare board, there may not be any notion of
850 ``process'', and there is often no way to get a core dump. @value{GDBN}
851 will warn you if it is unable to attach or to read core dumps.
853 You can optionally have @code{@value{GDBP}} pass any arguments after the
854 executable file to the inferior using @code{--args}. This option stops
857 @value{GDBP} --args gcc -O2 -c foo.c
859 This will cause @code{@value{GDBP}} to debug @code{gcc}, and to set
860 @code{gcc}'s command-line arguments (@pxref{Arguments}) to @samp{-O2 -c foo.c}.
862 You can run @code{@value{GDBP}} without printing the front material, which describes
863 @value{GDBN}'s non-warranty, by specifying @code{-silent}:
870 You can further control how @value{GDBN} starts up by using command-line
871 options. @value{GDBN} itself can remind you of the options available.
881 to display all available options and briefly describe their use
882 (@samp{@value{GDBP} -h} is a shorter equivalent).
884 All options and command line arguments you give are processed
885 in sequential order. The order makes a difference when the
886 @samp{-x} option is used.
890 * File Options:: Choosing files
891 * Mode Options:: Choosing modes
892 * Startup:: What @value{GDBN} does during startup
896 @subsection Choosing Files
898 When @value{GDBN} starts, it reads any arguments other than options as
899 specifying an executable file and core file (or process ID). This is
900 the same as if the arguments were specified by the @samp{-se} and
901 @samp{-c} (or @samp{-p}) options respectively. (@value{GDBN} reads the
902 first argument that does not have an associated option flag as
903 equivalent to the @samp{-se} option followed by that argument; and the
904 second argument that does not have an associated option flag, if any, as
905 equivalent to the @samp{-c}/@samp{-p} option followed by that argument.)
906 If the second argument begins with a decimal digit, @value{GDBN} will
907 first attempt to attach to it as a process, and if that fails, attempt
908 to open it as a corefile. If you have a corefile whose name begins with
909 a digit, you can prevent @value{GDBN} from treating it as a pid by
910 prefixing it with @file{./}, e.g.@: @file{./12345}.
912 If @value{GDBN} has not been configured to included core file support,
913 such as for most embedded targets, then it will complain about a second
914 argument and ignore it.
916 Many options have both long and short forms; both are shown in the
917 following list. @value{GDBN} also recognizes the long forms if you truncate
918 them, so long as enough of the option is present to be unambiguous.
919 (If you prefer, you can flag option arguments with @samp{--} rather
920 than @samp{-}, though we illustrate the more usual convention.)
922 @c NOTE: the @cindex entries here use double dashes ON PURPOSE. This
923 @c way, both those who look for -foo and --foo in the index, will find
927 @item -symbols @var{file}
929 @cindex @code{--symbols}
931 Read symbol table from file @var{file}.
933 @item -exec @var{file}
935 @cindex @code{--exec}
937 Use file @var{file} as the executable file to execute when appropriate,
938 and for examining pure data in conjunction with a core dump.
942 Read symbol table from file @var{file} and use it as the executable
945 @item -core @var{file}
947 @cindex @code{--core}
949 Use file @var{file} as a core dump to examine.
951 @item -pid @var{number}
952 @itemx -p @var{number}
955 Connect to process ID @var{number}, as with the @code{attach} command.
957 @item -command @var{file}
959 @cindex @code{--command}
961 Execute @value{GDBN} commands from file @var{file}. @xref{Command
962 Files,, Command files}.
964 @item -eval-command @var{command}
965 @itemx -ex @var{command}
966 @cindex @code{--eval-command}
968 Execute a single @value{GDBN} command.
970 This option may be used multiple times to call multiple commands. It may
971 also be interleaved with @samp{-command} as required.
974 @value{GDBP} -ex 'target sim' -ex 'load' \
975 -x setbreakpoints -ex 'run' a.out
978 @item -directory @var{directory}
979 @itemx -d @var{directory}
980 @cindex @code{--directory}
982 Add @var{directory} to the path to search for source and script files.
986 @cindex @code{--readnow}
988 Read each symbol file's entire symbol table immediately, rather than
989 the default, which is to read it incrementally as it is needed.
990 This makes startup slower, but makes future operations faster.
995 @subsection Choosing Modes
997 You can run @value{GDBN} in various alternative modes---for example, in
998 batch mode or quiet mode.
1005 Do not execute commands found in any initialization files. Normally,
1006 @value{GDBN} executes the commands in these files after all the command
1007 options and arguments have been processed. @xref{Command Files,,Command
1013 @cindex @code{--quiet}
1014 @cindex @code{--silent}
1016 ``Quiet''. Do not print the introductory and copyright messages. These
1017 messages are also suppressed in batch mode.
1020 @cindex @code{--batch}
1021 Run in batch mode. Exit with status @code{0} after processing all the
1022 command files specified with @samp{-x} (and all commands from
1023 initialization files, if not inhibited with @samp{-n}). Exit with
1024 nonzero status if an error occurs in executing the @value{GDBN} commands
1025 in the command files.
1027 Batch mode may be useful for running @value{GDBN} as a filter, for
1028 example to download and run a program on another computer; in order to
1029 make this more useful, the message
1032 Program exited normally.
1036 (which is ordinarily issued whenever a program running under
1037 @value{GDBN} control terminates) is not issued when running in batch
1041 @cindex @code{--batch-silent}
1042 Run in batch mode exactly like @samp{-batch}, but totally silently. All
1043 @value{GDBN} output to @code{stdout} is prevented (@code{stderr} is
1044 unaffected). This is much quieter than @samp{-silent} and would be useless
1045 for an interactive session.
1047 This is particularly useful when using targets that give @samp{Loading section}
1048 messages, for example.
1050 Note that targets that give their output via @value{GDBN}, as opposed to
1051 writing directly to @code{stdout}, will also be made silent.
1053 @item -return-child-result
1054 @cindex @code{--return-child-result}
1055 The return code from @value{GDBN} will be the return code from the child
1056 process (the process being debugged), with the following exceptions:
1060 @value{GDBN} exits abnormally. E.g., due to an incorrect argument or an
1061 internal error. In this case the exit code is the same as it would have been
1062 without @samp{-return-child-result}.
1064 The user quits with an explicit value. E.g., @samp{quit 1}.
1066 The child process never runs, or is not allowed to terminate, in which case
1067 the exit code will be -1.
1070 This option is useful in conjunction with @samp{-batch} or @samp{-batch-silent},
1071 when @value{GDBN} is being used as a remote program loader or simulator
1076 @cindex @code{--nowindows}
1078 ``No windows''. If @value{GDBN} comes with a graphical user interface
1079 (GUI) built in, then this option tells @value{GDBN} to only use the command-line
1080 interface. If no GUI is available, this option has no effect.
1084 @cindex @code{--windows}
1086 If @value{GDBN} includes a GUI, then this option requires it to be
1089 @item -cd @var{directory}
1091 Run @value{GDBN} using @var{directory} as its working directory,
1092 instead of the current directory.
1096 @cindex @code{--fullname}
1098 @sc{gnu} Emacs sets this option when it runs @value{GDBN} as a
1099 subprocess. It tells @value{GDBN} to output the full file name and line
1100 number in a standard, recognizable fashion each time a stack frame is
1101 displayed (which includes each time your program stops). This
1102 recognizable format looks like two @samp{\032} characters, followed by
1103 the file name, line number and character position separated by colons,
1104 and a newline. The Emacs-to-@value{GDBN} interface program uses the two
1105 @samp{\032} characters as a signal to display the source code for the
1109 @cindex @code{--epoch}
1110 The Epoch Emacs-@value{GDBN} interface sets this option when it runs
1111 @value{GDBN} as a subprocess. It tells @value{GDBN} to modify its print
1112 routines so as to allow Epoch to display values of expressions in a
1115 @item -annotate @var{level}
1116 @cindex @code{--annotate}
1117 This option sets the @dfn{annotation level} inside @value{GDBN}. Its
1118 effect is identical to using @samp{set annotate @var{level}}
1119 (@pxref{Annotations}). The annotation @var{level} controls how much
1120 information @value{GDBN} prints together with its prompt, values of
1121 expressions, source lines, and other types of output. Level 0 is the
1122 normal, level 1 is for use when @value{GDBN} is run as a subprocess of
1123 @sc{gnu} Emacs, level 3 is the maximum annotation suitable for programs
1124 that control @value{GDBN}, and level 2 has been deprecated.
1126 The annotation mechanism has largely been superseded by @sc{gdb/mi}
1130 @cindex @code{--args}
1131 Change interpretation of command line so that arguments following the
1132 executable file are passed as command line arguments to the inferior.
1133 This option stops option processing.
1135 @item -baud @var{bps}
1137 @cindex @code{--baud}
1139 Set the line speed (baud rate or bits per second) of any serial
1140 interface used by @value{GDBN} for remote debugging.
1142 @item -l @var{timeout}
1144 Set the timeout (in seconds) of any communication used by @value{GDBN}
1145 for remote debugging.
1147 @item -tty @var{device}
1148 @itemx -t @var{device}
1149 @cindex @code{--tty}
1151 Run using @var{device} for your program's standard input and output.
1152 @c FIXME: kingdon thinks there is more to -tty. Investigate.
1154 @c resolve the situation of these eventually
1156 @cindex @code{--tui}
1157 Activate the @dfn{Text User Interface} when starting. The Text User
1158 Interface manages several text windows on the terminal, showing
1159 source, assembly, registers and @value{GDBN} command outputs
1160 (@pxref{TUI, ,@value{GDBN} Text User Interface}). Alternatively, the
1161 Text User Interface can be enabled by invoking the program
1162 @samp{@value{GDBTUI}}. Do not use this option if you run @value{GDBN} from
1163 Emacs (@pxref{Emacs, ,Using @value{GDBN} under @sc{gnu} Emacs}).
1166 @c @cindex @code{--xdb}
1167 @c Run in XDB compatibility mode, allowing the use of certain XDB commands.
1168 @c For information, see the file @file{xdb_trans.html}, which is usually
1169 @c installed in the directory @code{/opt/langtools/wdb/doc} on HP-UX
1172 @item -interpreter @var{interp}
1173 @cindex @code{--interpreter}
1174 Use the interpreter @var{interp} for interface with the controlling
1175 program or device. This option is meant to be set by programs which
1176 communicate with @value{GDBN} using it as a back end.
1177 @xref{Interpreters, , Command Interpreters}.
1179 @samp{--interpreter=mi} (or @samp{--interpreter=mi2}) causes
1180 @value{GDBN} to use the @dfn{@sc{gdb/mi} interface} (@pxref{GDB/MI, ,
1181 The @sc{gdb/mi} Interface}) included since @value{GDBN} version 6.0. The
1182 previous @sc{gdb/mi} interface, included in @value{GDBN} version 5.3 and
1183 selected with @samp{--interpreter=mi1}, is deprecated. Earlier
1184 @sc{gdb/mi} interfaces are no longer supported.
1187 @cindex @code{--write}
1188 Open the executable and core files for both reading and writing. This
1189 is equivalent to the @samp{set write on} command inside @value{GDBN}
1193 @cindex @code{--statistics}
1194 This option causes @value{GDBN} to print statistics about time and
1195 memory usage after it completes each command and returns to the prompt.
1198 @cindex @code{--version}
1199 This option causes @value{GDBN} to print its version number and
1200 no-warranty blurb, and exit.
1205 @subsection What @value{GDBN} Does During Startup
1206 @cindex @value{GDBN} startup
1208 Here's the description of what @value{GDBN} does during session startup:
1212 Sets up the command interpreter as specified by the command line
1213 (@pxref{Mode Options, interpreter}).
1217 Reads the system-wide @dfn{init file} (if @option{--with-system-gdbinit} was
1218 used when building @value{GDBN}; @pxref{System-wide configuration,
1219 ,System-wide configuration and settings}) and executes all the commands in
1223 Reads the init file (if any) in your home directory@footnote{On
1224 DOS/Windows systems, the home directory is the one pointed to by the
1225 @code{HOME} environment variable.} and executes all the commands in
1229 Processes command line options and operands.
1232 Reads and executes the commands from init file (if any) in the current
1233 working directory. This is only done if the current directory is
1234 different from your home directory. Thus, you can have more than one
1235 init file, one generic in your home directory, and another, specific
1236 to the program you are debugging, in the directory where you invoke
1240 Reads command files specified by the @samp{-x} option. @xref{Command
1241 Files}, for more details about @value{GDBN} command files.
1244 Reads the command history recorded in the @dfn{history file}.
1245 @xref{Command History}, for more details about the command history and the
1246 files where @value{GDBN} records it.
1249 Init files use the same syntax as @dfn{command files} (@pxref{Command
1250 Files}) and are processed by @value{GDBN} in the same way. The init
1251 file in your home directory can set options (such as @samp{set
1252 complaints}) that affect subsequent processing of command line options
1253 and operands. Init files are not executed if you use the @samp{-nx}
1254 option (@pxref{Mode Options, ,Choosing Modes}).
1256 To display the list of init files loaded by gdb at startup, you
1257 can use @kbd{gdb --help}.
1259 @cindex init file name
1260 @cindex @file{.gdbinit}
1261 @cindex @file{gdb.ini}
1262 The @value{GDBN} init files are normally called @file{.gdbinit}.
1263 The DJGPP port of @value{GDBN} uses the name @file{gdb.ini}, due to
1264 the limitations of file names imposed by DOS filesystems. The Windows
1265 ports of @value{GDBN} use the standard name, but if they find a
1266 @file{gdb.ini} file, they warn you about that and suggest to rename
1267 the file to the standard name.
1271 @section Quitting @value{GDBN}
1272 @cindex exiting @value{GDBN}
1273 @cindex leaving @value{GDBN}
1276 @kindex quit @r{[}@var{expression}@r{]}
1277 @kindex q @r{(@code{quit})}
1278 @item quit @r{[}@var{expression}@r{]}
1280 To exit @value{GDBN}, use the @code{quit} command (abbreviated
1281 @code{q}), or type an end-of-file character (usually @kbd{Ctrl-d}). If you
1282 do not supply @var{expression}, @value{GDBN} will terminate normally;
1283 otherwise it will terminate using the result of @var{expression} as the
1288 An interrupt (often @kbd{Ctrl-c}) does not exit from @value{GDBN}, but rather
1289 terminates the action of any @value{GDBN} command that is in progress and
1290 returns to @value{GDBN} command level. It is safe to type the interrupt
1291 character at any time because @value{GDBN} does not allow it to take effect
1292 until a time when it is safe.
1294 If you have been using @value{GDBN} to control an attached process or
1295 device, you can release it with the @code{detach} command
1296 (@pxref{Attach, ,Debugging an Already-running Process}).
1298 @node Shell Commands
1299 @section Shell Commands
1301 If you need to execute occasional shell commands during your
1302 debugging session, there is no need to leave or suspend @value{GDBN}; you can
1303 just use the @code{shell} command.
1307 @cindex shell escape
1308 @item shell @var{command string}
1309 Invoke a standard shell to execute @var{command string}.
1310 If it exists, the environment variable @code{SHELL} determines which
1311 shell to run. Otherwise @value{GDBN} uses the default shell
1312 (@file{/bin/sh} on Unix systems, @file{COMMAND.COM} on MS-DOS, etc.).
1315 The utility @code{make} is often needed in development environments.
1316 You do not have to use the @code{shell} command for this purpose in
1321 @cindex calling make
1322 @item make @var{make-args}
1323 Execute the @code{make} program with the specified
1324 arguments. This is equivalent to @samp{shell make @var{make-args}}.
1327 @node Logging Output
1328 @section Logging Output
1329 @cindex logging @value{GDBN} output
1330 @cindex save @value{GDBN} output to a file
1332 You may want to save the output of @value{GDBN} commands to a file.
1333 There are several commands to control @value{GDBN}'s logging.
1337 @item set logging on
1339 @item set logging off
1341 @cindex logging file name
1342 @item set logging file @var{file}
1343 Change the name of the current logfile. The default logfile is @file{gdb.txt}.
1344 @item set logging overwrite [on|off]
1345 By default, @value{GDBN} will append to the logfile. Set @code{overwrite} if
1346 you want @code{set logging on} to overwrite the logfile instead.
1347 @item set logging redirect [on|off]
1348 By default, @value{GDBN} output will go to both the terminal and the logfile.
1349 Set @code{redirect} if you want output to go only to the log file.
1350 @kindex show logging
1352 Show the current values of the logging settings.
1356 @chapter @value{GDBN} Commands
1358 You can abbreviate a @value{GDBN} command to the first few letters of the command
1359 name, if that abbreviation is unambiguous; and you can repeat certain
1360 @value{GDBN} commands by typing just @key{RET}. You can also use the @key{TAB}
1361 key to get @value{GDBN} to fill out the rest of a word in a command (or to
1362 show you the alternatives available, if there is more than one possibility).
1365 * Command Syntax:: How to give commands to @value{GDBN}
1366 * Completion:: Command completion
1367 * Help:: How to ask @value{GDBN} for help
1370 @node Command Syntax
1371 @section Command Syntax
1373 A @value{GDBN} command is a single line of input. There is no limit on
1374 how long it can be. It starts with a command name, which is followed by
1375 arguments whose meaning depends on the command name. For example, the
1376 command @code{step} accepts an argument which is the number of times to
1377 step, as in @samp{step 5}. You can also use the @code{step} command
1378 with no arguments. Some commands do not allow any arguments.
1380 @cindex abbreviation
1381 @value{GDBN} command names may always be truncated if that abbreviation is
1382 unambiguous. Other possible command abbreviations are listed in the
1383 documentation for individual commands. In some cases, even ambiguous
1384 abbreviations are allowed; for example, @code{s} is specially defined as
1385 equivalent to @code{step} even though there are other commands whose
1386 names start with @code{s}. You can test abbreviations by using them as
1387 arguments to the @code{help} command.
1389 @cindex repeating commands
1390 @kindex RET @r{(repeat last command)}
1391 A blank line as input to @value{GDBN} (typing just @key{RET}) means to
1392 repeat the previous command. Certain commands (for example, @code{run})
1393 will not repeat this way; these are commands whose unintentional
1394 repetition might cause trouble and which you are unlikely to want to
1395 repeat. User-defined commands can disable this feature; see
1396 @ref{Define, dont-repeat}.
1398 The @code{list} and @code{x} commands, when you repeat them with
1399 @key{RET}, construct new arguments rather than repeating
1400 exactly as typed. This permits easy scanning of source or memory.
1402 @value{GDBN} can also use @key{RET} in another way: to partition lengthy
1403 output, in a way similar to the common utility @code{more}
1404 (@pxref{Screen Size,,Screen Size}). Since it is easy to press one
1405 @key{RET} too many in this situation, @value{GDBN} disables command
1406 repetition after any command that generates this sort of display.
1408 @kindex # @r{(a comment)}
1410 Any text from a @kbd{#} to the end of the line is a comment; it does
1411 nothing. This is useful mainly in command files (@pxref{Command
1412 Files,,Command Files}).
1414 @cindex repeating command sequences
1415 @kindex Ctrl-o @r{(operate-and-get-next)}
1416 The @kbd{Ctrl-o} binding is useful for repeating a complex sequence of
1417 commands. This command accepts the current line, like @key{RET}, and
1418 then fetches the next line relative to the current line from the history
1422 @section Command Completion
1425 @cindex word completion
1426 @value{GDBN} can fill in the rest of a word in a command for you, if there is
1427 only one possibility; it can also show you what the valid possibilities
1428 are for the next word in a command, at any time. This works for @value{GDBN}
1429 commands, @value{GDBN} subcommands, and the names of symbols in your program.
1431 Press the @key{TAB} key whenever you want @value{GDBN} to fill out the rest
1432 of a word. If there is only one possibility, @value{GDBN} fills in the
1433 word, and waits for you to finish the command (or press @key{RET} to
1434 enter it). For example, if you type
1436 @c FIXME "@key" does not distinguish its argument sufficiently to permit
1437 @c complete accuracy in these examples; space introduced for clarity.
1438 @c If texinfo enhancements make it unnecessary, it would be nice to
1439 @c replace " @key" by "@key" in the following...
1441 (@value{GDBP}) info bre @key{TAB}
1445 @value{GDBN} fills in the rest of the word @samp{breakpoints}, since that is
1446 the only @code{info} subcommand beginning with @samp{bre}:
1449 (@value{GDBP}) info breakpoints
1453 You can either press @key{RET} at this point, to run the @code{info
1454 breakpoints} command, or backspace and enter something else, if
1455 @samp{breakpoints} does not look like the command you expected. (If you
1456 were sure you wanted @code{info breakpoints} in the first place, you
1457 might as well just type @key{RET} immediately after @samp{info bre},
1458 to exploit command abbreviations rather than command completion).
1460 If there is more than one possibility for the next word when you press
1461 @key{TAB}, @value{GDBN} sounds a bell. You can either supply more
1462 characters and try again, or just press @key{TAB} a second time;
1463 @value{GDBN} displays all the possible completions for that word. For
1464 example, you might want to set a breakpoint on a subroutine whose name
1465 begins with @samp{make_}, but when you type @kbd{b make_@key{TAB}} @value{GDBN}
1466 just sounds the bell. Typing @key{TAB} again displays all the
1467 function names in your program that begin with those characters, for
1471 (@value{GDBP}) b make_ @key{TAB}
1472 @exdent @value{GDBN} sounds bell; press @key{TAB} again, to see:
1473 make_a_section_from_file make_environ
1474 make_abs_section make_function_type
1475 make_blockvector make_pointer_type
1476 make_cleanup make_reference_type
1477 make_command make_symbol_completion_list
1478 (@value{GDBP}) b make_
1482 After displaying the available possibilities, @value{GDBN} copies your
1483 partial input (@samp{b make_} in the example) so you can finish the
1486 If you just want to see the list of alternatives in the first place, you
1487 can press @kbd{M-?} rather than pressing @key{TAB} twice. @kbd{M-?}
1488 means @kbd{@key{META} ?}. You can type this either by holding down a
1489 key designated as the @key{META} shift on your keyboard (if there is
1490 one) while typing @kbd{?}, or as @key{ESC} followed by @kbd{?}.
1492 @cindex quotes in commands
1493 @cindex completion of quoted strings
1494 Sometimes the string you need, while logically a ``word'', may contain
1495 parentheses or other characters that @value{GDBN} normally excludes from
1496 its notion of a word. To permit word completion to work in this
1497 situation, you may enclose words in @code{'} (single quote marks) in
1498 @value{GDBN} commands.
1500 The most likely situation where you might need this is in typing the
1501 name of a C@t{++} function. This is because C@t{++} allows function
1502 overloading (multiple definitions of the same function, distinguished
1503 by argument type). For example, when you want to set a breakpoint you
1504 may need to distinguish whether you mean the version of @code{name}
1505 that takes an @code{int} parameter, @code{name(int)}, or the version
1506 that takes a @code{float} parameter, @code{name(float)}. To use the
1507 word-completion facilities in this situation, type a single quote
1508 @code{'} at the beginning of the function name. This alerts
1509 @value{GDBN} that it may need to consider more information than usual
1510 when you press @key{TAB} or @kbd{M-?} to request word completion:
1513 (@value{GDBP}) b 'bubble( @kbd{M-?}
1514 bubble(double,double) bubble(int,int)
1515 (@value{GDBP}) b 'bubble(
1518 In some cases, @value{GDBN} can tell that completing a name requires using
1519 quotes. When this happens, @value{GDBN} inserts the quote for you (while
1520 completing as much as it can) if you do not type the quote in the first
1524 (@value{GDBP}) b bub @key{TAB}
1525 @exdent @value{GDBN} alters your input line to the following, and rings a bell:
1526 (@value{GDBP}) b 'bubble(
1530 In general, @value{GDBN} can tell that a quote is needed (and inserts it) if
1531 you have not yet started typing the argument list when you ask for
1532 completion on an overloaded symbol.
1534 For more information about overloaded functions, see @ref{C Plus Plus
1535 Expressions, ,C@t{++} Expressions}. You can use the command @code{set
1536 overload-resolution off} to disable overload resolution;
1537 see @ref{Debugging C Plus Plus, ,@value{GDBN} Features for C@t{++}}.
1539 @cindex completion of structure field names
1540 @cindex structure field name completion
1541 @cindex completion of union field names
1542 @cindex union field name completion
1543 When completing in an expression which looks up a field in a
1544 structure, @value{GDBN} also tries@footnote{The completer can be
1545 confused by certain kinds of invalid expressions. Also, it only
1546 examines the static type of the expression, not the dynamic type.} to
1547 limit completions to the field names available in the type of the
1551 (@value{GDBP}) p gdb_stdout.@kbd{M-?}
1552 magic to_delete to_fputs to_put to_rewind
1553 to_data to_flush to_isatty to_read to_write
1557 This is because the @code{gdb_stdout} is a variable of the type
1558 @code{struct ui_file} that is defined in @value{GDBN} sources as
1565 ui_file_flush_ftype *to_flush;
1566 ui_file_write_ftype *to_write;
1567 ui_file_fputs_ftype *to_fputs;
1568 ui_file_read_ftype *to_read;
1569 ui_file_delete_ftype *to_delete;
1570 ui_file_isatty_ftype *to_isatty;
1571 ui_file_rewind_ftype *to_rewind;
1572 ui_file_put_ftype *to_put;
1579 @section Getting Help
1580 @cindex online documentation
1583 You can always ask @value{GDBN} itself for information on its commands,
1584 using the command @code{help}.
1587 @kindex h @r{(@code{help})}
1590 You can use @code{help} (abbreviated @code{h}) with no arguments to
1591 display a short list of named classes of commands:
1595 List of classes of commands:
1597 aliases -- Aliases of other commands
1598 breakpoints -- Making program stop at certain points
1599 data -- Examining data
1600 files -- Specifying and examining files
1601 internals -- Maintenance commands
1602 obscure -- Obscure features
1603 running -- Running the program
1604 stack -- Examining the stack
1605 status -- Status inquiries
1606 support -- Support facilities
1607 tracepoints -- Tracing of program execution without
1608 stopping the program
1609 user-defined -- User-defined commands
1611 Type "help" followed by a class name for a list of
1612 commands in that class.
1613 Type "help" followed by command name for full
1615 Command name abbreviations are allowed if unambiguous.
1618 @c the above line break eliminates huge line overfull...
1620 @item help @var{class}
1621 Using one of the general help classes as an argument, you can get a
1622 list of the individual commands in that class. For example, here is the
1623 help display for the class @code{status}:
1626 (@value{GDBP}) help status
1631 @c Line break in "show" line falsifies real output, but needed
1632 @c to fit in smallbook page size.
1633 info -- Generic command for showing things
1634 about the program being debugged
1635 show -- Generic command for showing things
1638 Type "help" followed by command name for full
1640 Command name abbreviations are allowed if unambiguous.
1644 @item help @var{command}
1645 With a command name as @code{help} argument, @value{GDBN} displays a
1646 short paragraph on how to use that command.
1649 @item apropos @var{args}
1650 The @code{apropos} command searches through all of the @value{GDBN}
1651 commands, and their documentation, for the regular expression specified in
1652 @var{args}. It prints out all matches found. For example:
1663 set symbol-reloading -- Set dynamic symbol table reloading
1664 multiple times in one run
1665 show symbol-reloading -- Show dynamic symbol table reloading
1666 multiple times in one run
1671 @item complete @var{args}
1672 The @code{complete @var{args}} command lists all the possible completions
1673 for the beginning of a command. Use @var{args} to specify the beginning of the
1674 command you want completed. For example:
1680 @noindent results in:
1691 @noindent This is intended for use by @sc{gnu} Emacs.
1694 In addition to @code{help}, you can use the @value{GDBN} commands @code{info}
1695 and @code{show} to inquire about the state of your program, or the state
1696 of @value{GDBN} itself. Each command supports many topics of inquiry; this
1697 manual introduces each of them in the appropriate context. The listings
1698 under @code{info} and under @code{show} in the Index point to
1699 all the sub-commands. @xref{Index}.
1704 @kindex i @r{(@code{info})}
1706 This command (abbreviated @code{i}) is for describing the state of your
1707 program. For example, you can show the arguments passed to a function
1708 with @code{info args}, list the registers currently in use with @code{info
1709 registers}, or list the breakpoints you have set with @code{info breakpoints}.
1710 You can get a complete list of the @code{info} sub-commands with
1711 @w{@code{help info}}.
1715 You can assign the result of an expression to an environment variable with
1716 @code{set}. For example, you can set the @value{GDBN} prompt to a $-sign with
1717 @code{set prompt $}.
1721 In contrast to @code{info}, @code{show} is for describing the state of
1722 @value{GDBN} itself.
1723 You can change most of the things you can @code{show}, by using the
1724 related command @code{set}; for example, you can control what number
1725 system is used for displays with @code{set radix}, or simply inquire
1726 which is currently in use with @code{show radix}.
1729 To display all the settable parameters and their current
1730 values, you can use @code{show} with no arguments; you may also use
1731 @code{info set}. Both commands produce the same display.
1732 @c FIXME: "info set" violates the rule that "info" is for state of
1733 @c FIXME...program. Ck w/ GNU: "info set" to be called something else,
1734 @c FIXME...or change desc of rule---eg "state of prog and debugging session"?
1738 Here are three miscellaneous @code{show} subcommands, all of which are
1739 exceptional in lacking corresponding @code{set} commands:
1742 @kindex show version
1743 @cindex @value{GDBN} version number
1745 Show what version of @value{GDBN} is running. You should include this
1746 information in @value{GDBN} bug-reports. If multiple versions of
1747 @value{GDBN} are in use at your site, you may need to determine which
1748 version of @value{GDBN} you are running; as @value{GDBN} evolves, new
1749 commands are introduced, and old ones may wither away. Also, many
1750 system vendors ship variant versions of @value{GDBN}, and there are
1751 variant versions of @value{GDBN} in @sc{gnu}/Linux distributions as well.
1752 The version number is the same as the one announced when you start
1755 @kindex show copying
1756 @kindex info copying
1757 @cindex display @value{GDBN} copyright
1760 Display information about permission for copying @value{GDBN}.
1762 @kindex show warranty
1763 @kindex info warranty
1765 @itemx info warranty
1766 Display the @sc{gnu} ``NO WARRANTY'' statement, or a warranty,
1767 if your version of @value{GDBN} comes with one.
1772 @chapter Running Programs Under @value{GDBN}
1774 When you run a program under @value{GDBN}, you must first generate
1775 debugging information when you compile it.
1777 You may start @value{GDBN} with its arguments, if any, in an environment
1778 of your choice. If you are doing native debugging, you may redirect
1779 your program's input and output, debug an already running process, or
1780 kill a child process.
1783 * Compilation:: Compiling for debugging
1784 * Starting:: Starting your program
1785 * Arguments:: Your program's arguments
1786 * Environment:: Your program's environment
1788 * Working Directory:: Your program's working directory
1789 * Input/Output:: Your program's input and output
1790 * Attach:: Debugging an already-running process
1791 * Kill Process:: Killing the child process
1793 * Inferiors:: Debugging multiple inferiors
1794 * Threads:: Debugging programs with multiple threads
1795 * Processes:: Debugging programs with multiple processes
1796 * Checkpoint/Restart:: Setting a @emph{bookmark} to return to later
1800 @section Compiling for Debugging
1802 In order to debug a program effectively, you need to generate
1803 debugging information when you compile it. This debugging information
1804 is stored in the object file; it describes the data type of each
1805 variable or function and the correspondence between source line numbers
1806 and addresses in the executable code.
1808 To request debugging information, specify the @samp{-g} option when you run
1811 Programs that are to be shipped to your customers are compiled with
1812 optimizations, using the @samp{-O} compiler option. However, some
1813 compilers are unable to handle the @samp{-g} and @samp{-O} options
1814 together. Using those compilers, you cannot generate optimized
1815 executables containing debugging information.
1817 @value{NGCC}, the @sc{gnu} C/C@t{++} compiler, supports @samp{-g} with or
1818 without @samp{-O}, making it possible to debug optimized code. We
1819 recommend that you @emph{always} use @samp{-g} whenever you compile a
1820 program. You may think your program is correct, but there is no sense
1821 in pushing your luck. For more information, see @ref{Optimized Code}.
1823 Older versions of the @sc{gnu} C compiler permitted a variant option
1824 @w{@samp{-gg}} for debugging information. @value{GDBN} no longer supports this
1825 format; if your @sc{gnu} C compiler has this option, do not use it.
1827 @value{GDBN} knows about preprocessor macros and can show you their
1828 expansion (@pxref{Macros}). Most compilers do not include information
1829 about preprocessor macros in the debugging information if you specify
1830 the @option{-g} flag alone, because this information is rather large.
1831 Version 3.1 and later of @value{NGCC}, the @sc{gnu} C compiler,
1832 provides macro information if you specify the options
1833 @option{-gdwarf-2} and @option{-g3}; the former option requests
1834 debugging information in the Dwarf 2 format, and the latter requests
1835 ``extra information''. In the future, we hope to find more compact
1836 ways to represent macro information, so that it can be included with
1841 @section Starting your Program
1847 @kindex r @r{(@code{run})}
1850 Use the @code{run} command to start your program under @value{GDBN}.
1851 You must first specify the program name (except on VxWorks) with an
1852 argument to @value{GDBN} (@pxref{Invocation, ,Getting In and Out of
1853 @value{GDBN}}), or by using the @code{file} or @code{exec-file} command
1854 (@pxref{Files, ,Commands to Specify Files}).
1858 If you are running your program in an execution environment that
1859 supports processes, @code{run} creates an inferior process and makes
1860 that process run your program. In some environments without processes,
1861 @code{run} jumps to the start of your program. Other targets,
1862 like @samp{remote}, are always running. If you get an error
1863 message like this one:
1866 The "remote" target does not support "run".
1867 Try "help target" or "continue".
1871 then use @code{continue} to run your program. You may need @code{load}
1872 first (@pxref{load}).
1874 The execution of a program is affected by certain information it
1875 receives from its superior. @value{GDBN} provides ways to specify this
1876 information, which you must do @emph{before} starting your program. (You
1877 can change it after starting your program, but such changes only affect
1878 your program the next time you start it.) This information may be
1879 divided into four categories:
1882 @item The @emph{arguments.}
1883 Specify the arguments to give your program as the arguments of the
1884 @code{run} command. If a shell is available on your target, the shell
1885 is used to pass the arguments, so that you may use normal conventions
1886 (such as wildcard expansion or variable substitution) in describing
1888 In Unix systems, you can control which shell is used with the
1889 @code{SHELL} environment variable.
1890 @xref{Arguments, ,Your Program's Arguments}.
1892 @item The @emph{environment.}
1893 Your program normally inherits its environment from @value{GDBN}, but you can
1894 use the @value{GDBN} commands @code{set environment} and @code{unset
1895 environment} to change parts of the environment that affect
1896 your program. @xref{Environment, ,Your Program's Environment}.
1898 @item The @emph{working directory.}
1899 Your program inherits its working directory from @value{GDBN}. You can set
1900 the @value{GDBN} working directory with the @code{cd} command in @value{GDBN}.
1901 @xref{Working Directory, ,Your Program's Working Directory}.
1903 @item The @emph{standard input and output.}
1904 Your program normally uses the same device for standard input and
1905 standard output as @value{GDBN} is using. You can redirect input and output
1906 in the @code{run} command line, or you can use the @code{tty} command to
1907 set a different device for your program.
1908 @xref{Input/Output, ,Your Program's Input and Output}.
1911 @emph{Warning:} While input and output redirection work, you cannot use
1912 pipes to pass the output of the program you are debugging to another
1913 program; if you attempt this, @value{GDBN} is likely to wind up debugging the
1917 When you issue the @code{run} command, your program begins to execute
1918 immediately. @xref{Stopping, ,Stopping and Continuing}, for discussion
1919 of how to arrange for your program to stop. Once your program has
1920 stopped, you may call functions in your program, using the @code{print}
1921 or @code{call} commands. @xref{Data, ,Examining Data}.
1923 If the modification time of your symbol file has changed since the last
1924 time @value{GDBN} read its symbols, @value{GDBN} discards its symbol
1925 table, and reads it again. When it does this, @value{GDBN} tries to retain
1926 your current breakpoints.
1931 @cindex run to main procedure
1932 The name of the main procedure can vary from language to language.
1933 With C or C@t{++}, the main procedure name is always @code{main}, but
1934 other languages such as Ada do not require a specific name for their
1935 main procedure. The debugger provides a convenient way to start the
1936 execution of the program and to stop at the beginning of the main
1937 procedure, depending on the language used.
1939 The @samp{start} command does the equivalent of setting a temporary
1940 breakpoint at the beginning of the main procedure and then invoking
1941 the @samp{run} command.
1943 @cindex elaboration phase
1944 Some programs contain an @dfn{elaboration} phase where some startup code is
1945 executed before the main procedure is called. This depends on the
1946 languages used to write your program. In C@t{++}, for instance,
1947 constructors for static and global objects are executed before
1948 @code{main} is called. It is therefore possible that the debugger stops
1949 before reaching the main procedure. However, the temporary breakpoint
1950 will remain to halt execution.
1952 Specify the arguments to give to your program as arguments to the
1953 @samp{start} command. These arguments will be given verbatim to the
1954 underlying @samp{run} command. Note that the same arguments will be
1955 reused if no argument is provided during subsequent calls to
1956 @samp{start} or @samp{run}.
1958 It is sometimes necessary to debug the program during elaboration. In
1959 these cases, using the @code{start} command would stop the execution of
1960 your program too late, as the program would have already completed the
1961 elaboration phase. Under these circumstances, insert breakpoints in your
1962 elaboration code before running your program.
1964 @kindex set exec-wrapper
1965 @item set exec-wrapper @var{wrapper}
1966 @itemx show exec-wrapper
1967 @itemx unset exec-wrapper
1968 When @samp{exec-wrapper} is set, the specified wrapper is used to
1969 launch programs for debugging. @value{GDBN} starts your program
1970 with a shell command of the form @kbd{exec @var{wrapper}
1971 @var{program}}. Quoting is added to @var{program} and its
1972 arguments, but not to @var{wrapper}, so you should add quotes if
1973 appropriate for your shell. The wrapper runs until it executes
1974 your program, and then @value{GDBN} takes control.
1976 You can use any program that eventually calls @code{execve} with
1977 its arguments as a wrapper. Several standard Unix utilities do
1978 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
1979 with @code{exec "$@@"} will also work.
1981 For example, you can use @code{env} to pass an environment variable to
1982 the debugged program, without setting the variable in your shell's
1986 (@value{GDBP}) set exec-wrapper env 'LD_PRELOAD=libtest.so'
1990 This command is available when debugging locally on most targets, excluding
1991 @sc{djgpp}, Cygwin, MS Windows, and QNX Neutrino.
1993 @kindex set disable-randomization
1994 @item set disable-randomization
1995 @itemx set disable-randomization on
1996 This option (enabled by default in @value{GDBN}) will turn off the native
1997 randomization of the virtual address space of the started program. This option
1998 is useful for multiple debugging sessions to make the execution better
1999 reproducible and memory addresses reusable across debugging sessions.
2001 This feature is implemented only on @sc{gnu}/Linux. You can get the same
2005 (@value{GDBP}) set exec-wrapper setarch `uname -m` -R
2008 @item set disable-randomization off
2009 Leave the behavior of the started executable unchanged. Some bugs rear their
2010 ugly heads only when the program is loaded at certain addresses. If your bug
2011 disappears when you run the program under @value{GDBN}, that might be because
2012 @value{GDBN} by default disables the address randomization on platforms, such
2013 as @sc{gnu}/Linux, which do that for stand-alone programs. Use @kbd{set
2014 disable-randomization off} to try to reproduce such elusive bugs.
2016 The virtual address space randomization is implemented only on @sc{gnu}/Linux.
2017 It protects the programs against some kinds of security attacks. In these
2018 cases the attacker needs to know the exact location of a concrete executable
2019 code. Randomizing its location makes it impossible to inject jumps misusing
2020 a code at its expected addresses.
2022 Prelinking shared libraries provides a startup performance advantage but it
2023 makes addresses in these libraries predictable for privileged processes by
2024 having just unprivileged access at the target system. Reading the shared
2025 library binary gives enough information for assembling the malicious code
2026 misusing it. Still even a prelinked shared library can get loaded at a new
2027 random address just requiring the regular relocation process during the
2028 startup. Shared libraries not already prelinked are always loaded at
2029 a randomly chosen address.
2031 Position independent executables (PIE) contain position independent code
2032 similar to the shared libraries and therefore such executables get loaded at
2033 a randomly chosen address upon startup. PIE executables always load even
2034 already prelinked shared libraries at a random address. You can build such
2035 executable using @command{gcc -fPIE -pie}.
2037 Heap (malloc storage), stack and custom mmap areas are always placed randomly
2038 (as long as the randomization is enabled).
2040 @item show disable-randomization
2041 Show the current setting of the explicit disable of the native randomization of
2042 the virtual address space of the started program.
2047 @section Your Program's Arguments
2049 @cindex arguments (to your program)
2050 The arguments to your program can be specified by the arguments of the
2052 They are passed to a shell, which expands wildcard characters and
2053 performs redirection of I/O, and thence to your program. Your
2054 @code{SHELL} environment variable (if it exists) specifies what shell
2055 @value{GDBN} uses. If you do not define @code{SHELL}, @value{GDBN} uses
2056 the default shell (@file{/bin/sh} on Unix).
2058 On non-Unix systems, the program is usually invoked directly by
2059 @value{GDBN}, which emulates I/O redirection via the appropriate system
2060 calls, and the wildcard characters are expanded by the startup code of
2061 the program, not by the shell.
2063 @code{run} with no arguments uses the same arguments used by the previous
2064 @code{run}, or those set by the @code{set args} command.
2069 Specify the arguments to be used the next time your program is run. If
2070 @code{set args} has no arguments, @code{run} executes your program
2071 with no arguments. Once you have run your program with arguments,
2072 using @code{set args} before the next @code{run} is the only way to run
2073 it again without arguments.
2077 Show the arguments to give your program when it is started.
2081 @section Your Program's Environment
2083 @cindex environment (of your program)
2084 The @dfn{environment} consists of a set of environment variables and
2085 their values. Environment variables conventionally record such things as
2086 your user name, your home directory, your terminal type, and your search
2087 path for programs to run. Usually you set up environment variables with
2088 the shell and they are inherited by all the other programs you run. When
2089 debugging, it can be useful to try running your program with a modified
2090 environment without having to start @value{GDBN} over again.
2094 @item path @var{directory}
2095 Add @var{directory} to the front of the @code{PATH} environment variable
2096 (the search path for executables) that will be passed to your program.
2097 The value of @code{PATH} used by @value{GDBN} does not change.
2098 You may specify several directory names, separated by whitespace or by a
2099 system-dependent separator character (@samp{:} on Unix, @samp{;} on
2100 MS-DOS and MS-Windows). If @var{directory} is already in the path, it
2101 is moved to the front, so it is searched sooner.
2103 You can use the string @samp{$cwd} to refer to whatever is the current
2104 working directory at the time @value{GDBN} searches the path. If you
2105 use @samp{.} instead, it refers to the directory where you executed the
2106 @code{path} command. @value{GDBN} replaces @samp{.} in the
2107 @var{directory} argument (with the current path) before adding
2108 @var{directory} to the search path.
2109 @c 'path' is explicitly nonrepeatable, but RMS points out it is silly to
2110 @c document that, since repeating it would be a no-op.
2114 Display the list of search paths for executables (the @code{PATH}
2115 environment variable).
2117 @kindex show environment
2118 @item show environment @r{[}@var{varname}@r{]}
2119 Print the value of environment variable @var{varname} to be given to
2120 your program when it starts. If you do not supply @var{varname},
2121 print the names and values of all environment variables to be given to
2122 your program. You can abbreviate @code{environment} as @code{env}.
2124 @kindex set environment
2125 @item set environment @var{varname} @r{[}=@var{value}@r{]}
2126 Set environment variable @var{varname} to @var{value}. The value
2127 changes for your program only, not for @value{GDBN} itself. @var{value} may
2128 be any string; the values of environment variables are just strings, and
2129 any interpretation is supplied by your program itself. The @var{value}
2130 parameter is optional; if it is eliminated, the variable is set to a
2132 @c "any string" here does not include leading, trailing
2133 @c blanks. Gnu asks: does anyone care?
2135 For example, this command:
2142 tells the debugged program, when subsequently run, that its user is named
2143 @samp{foo}. (The spaces around @samp{=} are used for clarity here; they
2144 are not actually required.)
2146 @kindex unset environment
2147 @item unset environment @var{varname}
2148 Remove variable @var{varname} from the environment to be passed to your
2149 program. This is different from @samp{set env @var{varname} =};
2150 @code{unset environment} removes the variable from the environment,
2151 rather than assigning it an empty value.
2154 @emph{Warning:} On Unix systems, @value{GDBN} runs your program using
2156 by your @code{SHELL} environment variable if it exists (or
2157 @code{/bin/sh} if not). If your @code{SHELL} variable names a shell
2158 that runs an initialization file---such as @file{.cshrc} for C-shell, or
2159 @file{.bashrc} for BASH---any variables you set in that file affect
2160 your program. You may wish to move setting of environment variables to
2161 files that are only run when you sign on, such as @file{.login} or
2164 @node Working Directory
2165 @section Your Program's Working Directory
2167 @cindex working directory (of your program)
2168 Each time you start your program with @code{run}, it inherits its
2169 working directory from the current working directory of @value{GDBN}.
2170 The @value{GDBN} working directory is initially whatever it inherited
2171 from its parent process (typically the shell), but you can specify a new
2172 working directory in @value{GDBN} with the @code{cd} command.
2174 The @value{GDBN} working directory also serves as a default for the commands
2175 that specify files for @value{GDBN} to operate on. @xref{Files, ,Commands to
2180 @cindex change working directory
2181 @item cd @var{directory}
2182 Set the @value{GDBN} working directory to @var{directory}.
2186 Print the @value{GDBN} working directory.
2189 It is generally impossible to find the current working directory of
2190 the process being debugged (since a program can change its directory
2191 during its run). If you work on a system where @value{GDBN} is
2192 configured with the @file{/proc} support, you can use the @code{info
2193 proc} command (@pxref{SVR4 Process Information}) to find out the
2194 current working directory of the debuggee.
2197 @section Your Program's Input and Output
2202 By default, the program you run under @value{GDBN} does input and output to
2203 the same terminal that @value{GDBN} uses. @value{GDBN} switches the terminal
2204 to its own terminal modes to interact with you, but it records the terminal
2205 modes your program was using and switches back to them when you continue
2206 running your program.
2209 @kindex info terminal
2211 Displays information recorded by @value{GDBN} about the terminal modes your
2215 You can redirect your program's input and/or output using shell
2216 redirection with the @code{run} command. For example,
2223 starts your program, diverting its output to the file @file{outfile}.
2226 @cindex controlling terminal
2227 Another way to specify where your program should do input and output is
2228 with the @code{tty} command. This command accepts a file name as
2229 argument, and causes this file to be the default for future @code{run}
2230 commands. It also resets the controlling terminal for the child
2231 process, for future @code{run} commands. For example,
2238 directs that processes started with subsequent @code{run} commands
2239 default to do input and output on the terminal @file{/dev/ttyb} and have
2240 that as their controlling terminal.
2242 An explicit redirection in @code{run} overrides the @code{tty} command's
2243 effect on the input/output device, but not its effect on the controlling
2246 When you use the @code{tty} command or redirect input in the @code{run}
2247 command, only the input @emph{for your program} is affected. The input
2248 for @value{GDBN} still comes from your terminal. @code{tty} is an alias
2249 for @code{set inferior-tty}.
2251 @cindex inferior tty
2252 @cindex set inferior controlling terminal
2253 You can use the @code{show inferior-tty} command to tell @value{GDBN} to
2254 display the name of the terminal that will be used for future runs of your
2258 @item set inferior-tty /dev/ttyb
2259 @kindex set inferior-tty
2260 Set the tty for the program being debugged to /dev/ttyb.
2262 @item show inferior-tty
2263 @kindex show inferior-tty
2264 Show the current tty for the program being debugged.
2268 @section Debugging an Already-running Process
2273 @item attach @var{process-id}
2274 This command attaches to a running process---one that was started
2275 outside @value{GDBN}. (@code{info files} shows your active
2276 targets.) The command takes as argument a process ID. The usual way to
2277 find out the @var{process-id} of a Unix process is with the @code{ps} utility,
2278 or with the @samp{jobs -l} shell command.
2280 @code{attach} does not repeat if you press @key{RET} a second time after
2281 executing the command.
2284 To use @code{attach}, your program must be running in an environment
2285 which supports processes; for example, @code{attach} does not work for
2286 programs on bare-board targets that lack an operating system. You must
2287 also have permission to send the process a signal.
2289 When you use @code{attach}, the debugger finds the program running in
2290 the process first by looking in the current working directory, then (if
2291 the program is not found) by using the source file search path
2292 (@pxref{Source Path, ,Specifying Source Directories}). You can also use
2293 the @code{file} command to load the program. @xref{Files, ,Commands to
2296 The first thing @value{GDBN} does after arranging to debug the specified
2297 process is to stop it. You can examine and modify an attached process
2298 with all the @value{GDBN} commands that are ordinarily available when
2299 you start processes with @code{run}. You can insert breakpoints; you
2300 can step and continue; you can modify storage. If you would rather the
2301 process continue running, you may use the @code{continue} command after
2302 attaching @value{GDBN} to the process.
2307 When you have finished debugging the attached process, you can use the
2308 @code{detach} command to release it from @value{GDBN} control. Detaching
2309 the process continues its execution. After the @code{detach} command,
2310 that process and @value{GDBN} become completely independent once more, and you
2311 are ready to @code{attach} another process or start one with @code{run}.
2312 @code{detach} does not repeat if you press @key{RET} again after
2313 executing the command.
2316 If you exit @value{GDBN} while you have an attached process, you detach
2317 that process. If you use the @code{run} command, you kill that process.
2318 By default, @value{GDBN} asks for confirmation if you try to do either of these
2319 things; you can control whether or not you need to confirm by using the
2320 @code{set confirm} command (@pxref{Messages/Warnings, ,Optional Warnings and
2324 @section Killing the Child Process
2329 Kill the child process in which your program is running under @value{GDBN}.
2332 This command is useful if you wish to debug a core dump instead of a
2333 running process. @value{GDBN} ignores any core dump file while your program
2336 On some operating systems, a program cannot be executed outside @value{GDBN}
2337 while you have breakpoints set on it inside @value{GDBN}. You can use the
2338 @code{kill} command in this situation to permit running your program
2339 outside the debugger.
2341 The @code{kill} command is also useful if you wish to recompile and
2342 relink your program, since on many systems it is impossible to modify an
2343 executable file while it is running in a process. In this case, when you
2344 next type @code{run}, @value{GDBN} notices that the file has changed, and
2345 reads the symbol table again (while trying to preserve your current
2346 breakpoint settings).
2349 @section Debugging Multiple Inferiors
2351 Some @value{GDBN} targets are able to run multiple processes created
2352 from a single executable. This can happen, for instance, with an
2353 embedded system reporting back several processes via the remote
2357 @value{GDBN} represents the state of each program execution with an
2358 object called an @dfn{inferior}. An inferior typically corresponds to
2359 a process, but is more general and applies also to targets that do not
2360 have processes. Inferiors may be created before a process runs, and
2361 may (in future) be retained after a process exits. Each run of an
2362 executable creates a new inferior, as does each attachment to an
2363 existing process. Inferiors have unique identifiers that are
2364 different from process ids, and may optionally be named as well.
2365 Usually each inferior will also have its own distinct address space,
2366 although some embedded targets may have several inferiors running in
2367 different parts of a single space.
2369 Each inferior may in turn have multiple threads running in it.
2371 To find out what inferiors exist at any moment, use @code{info inferiors}:
2374 @kindex info inferiors
2375 @item info inferiors
2376 Print a list of all inferiors currently being managed by @value{GDBN}.
2379 To switch focus between inferiors, use the @code{inferior} command:
2382 @kindex inferior @var{inferior-id}
2383 @item inferior @var{inferior-id}
2384 Make inferior number @var{inferior-id} the current inferior. The
2385 argument @var{inferior-id} is the internal inferior number assigned by
2386 @value{GDBN}, as shown in the first field of the @samp{info inferiors}
2390 To quit debugging one of the inferiors, you can either detach from it
2391 by using the @w{@code{detach inferior}} command (allowing it to run
2392 independently), or kill it using the @w{@code{kill inferior}} command:
2395 @kindex detach inferior @var{inferior-id}
2396 @item detach inferior @var{inferior-id}
2397 Detach from the inferior identified by @value{GDBN} inferior number
2398 @var{inferior-id}, and remove it from the inferior list.
2400 @kindex kill inferior @var{inferior-id}
2401 @item kill inferior @var{inferior-id}
2402 Kill the inferior identified by @value{GDBN} inferior number
2403 @var{inferior-id}, and remove it from the inferior list.
2406 To be notified when inferiors are started or exit under @value{GDBN}'s
2407 control use @w{@code{set print inferior-events}}:
2410 @kindex set print inferior-events
2411 @cindex print messages on inferior start and exit
2412 @item set print inferior-events
2413 @itemx set print inferior-events on
2414 @itemx set print inferior-events off
2415 The @code{set print inferior-events} command allows you to enable or
2416 disable printing of messages when @value{GDBN} notices that new
2417 inferiors have started or that inferiors have exited or have been
2418 detached. By default, these messages will not be printed.
2420 @kindex show print inferior-events
2421 @item show print inferior-events
2422 Show whether messages will be printed when @value{GDBN} detects that
2423 inferiors have started, exited or have been detached.
2427 @section Debugging Programs with Multiple Threads
2429 @cindex threads of execution
2430 @cindex multiple threads
2431 @cindex switching threads
2432 In some operating systems, such as HP-UX and Solaris, a single program
2433 may have more than one @dfn{thread} of execution. The precise semantics
2434 of threads differ from one operating system to another, but in general
2435 the threads of a single program are akin to multiple processes---except
2436 that they share one address space (that is, they can all examine and
2437 modify the same variables). On the other hand, each thread has its own
2438 registers and execution stack, and perhaps private memory.
2440 @value{GDBN} provides these facilities for debugging multi-thread
2444 @item automatic notification of new threads
2445 @item @samp{thread @var{threadno}}, a command to switch among threads
2446 @item @samp{info threads}, a command to inquire about existing threads
2447 @item @samp{thread apply [@var{threadno}] [@var{all}] @var{args}},
2448 a command to apply a command to a list of threads
2449 @item thread-specific breakpoints
2450 @item @samp{set print thread-events}, which controls printing of
2451 messages on thread start and exit.
2452 @item @samp{set libthread-db-search-path @var{path}}, which lets
2453 the user specify which @code{libthread_db} to use if the default choice
2454 isn't compatible with the program.
2458 @emph{Warning:} These facilities are not yet available on every
2459 @value{GDBN} configuration where the operating system supports threads.
2460 If your @value{GDBN} does not support threads, these commands have no
2461 effect. For example, a system without thread support shows no output
2462 from @samp{info threads}, and always rejects the @code{thread} command,
2466 (@value{GDBP}) info threads
2467 (@value{GDBP}) thread 1
2468 Thread ID 1 not known. Use the "info threads" command to
2469 see the IDs of currently known threads.
2471 @c FIXME to implementors: how hard would it be to say "sorry, this GDB
2472 @c doesn't support threads"?
2475 @cindex focus of debugging
2476 @cindex current thread
2477 The @value{GDBN} thread debugging facility allows you to observe all
2478 threads while your program runs---but whenever @value{GDBN} takes
2479 control, one thread in particular is always the focus of debugging.
2480 This thread is called the @dfn{current thread}. Debugging commands show
2481 program information from the perspective of the current thread.
2483 @cindex @code{New} @var{systag} message
2484 @cindex thread identifier (system)
2485 @c FIXME-implementors!! It would be more helpful if the [New...] message
2486 @c included GDB's numeric thread handle, so you could just go to that
2487 @c thread without first checking `info threads'.
2488 Whenever @value{GDBN} detects a new thread in your program, it displays
2489 the target system's identification for the thread with a message in the
2490 form @samp{[New @var{systag}]}. @var{systag} is a thread identifier
2491 whose form varies depending on the particular system. For example, on
2492 @sc{gnu}/Linux, you might see
2495 [New Thread 46912507313328 (LWP 25582)]
2499 when @value{GDBN} notices a new thread. In contrast, on an SGI system,
2500 the @var{systag} is simply something like @samp{process 368}, with no
2503 @c FIXME!! (1) Does the [New...] message appear even for the very first
2504 @c thread of a program, or does it only appear for the
2505 @c second---i.e.@: when it becomes obvious we have a multithread
2507 @c (2) *Is* there necessarily a first thread always? Or do some
2508 @c multithread systems permit starting a program with multiple
2509 @c threads ab initio?
2511 @cindex thread number
2512 @cindex thread identifier (GDB)
2513 For debugging purposes, @value{GDBN} associates its own thread
2514 number---always a single integer---with each thread in your program.
2517 @kindex info threads
2519 Display a summary of all threads currently in your
2520 program. @value{GDBN} displays for each thread (in this order):
2524 the thread number assigned by @value{GDBN}
2527 the target system's thread identifier (@var{systag})
2530 the current stack frame summary for that thread
2534 An asterisk @samp{*} to the left of the @value{GDBN} thread number
2535 indicates the current thread.
2539 @c end table here to get a little more width for example
2542 (@value{GDBP}) info threads
2543 3 process 35 thread 27 0x34e5 in sigpause ()
2544 2 process 35 thread 23 0x34e5 in sigpause ()
2545 * 1 process 35 thread 13 main (argc=1, argv=0x7ffffff8)
2551 @cindex debugging multithreaded programs (on HP-UX)
2552 @cindex thread identifier (GDB), on HP-UX
2553 For debugging purposes, @value{GDBN} associates its own thread
2554 number---a small integer assigned in thread-creation order---with each
2555 thread in your program.
2557 @cindex @code{New} @var{systag} message, on HP-UX
2558 @cindex thread identifier (system), on HP-UX
2559 @c FIXME-implementors!! It would be more helpful if the [New...] message
2560 @c included GDB's numeric thread handle, so you could just go to that
2561 @c thread without first checking `info threads'.
2562 Whenever @value{GDBN} detects a new thread in your program, it displays
2563 both @value{GDBN}'s thread number and the target system's identification for the thread with a message in the
2564 form @samp{[New @var{systag}]}. @var{systag} is a thread identifier
2565 whose form varies depending on the particular system. For example, on
2569 [New thread 2 (system thread 26594)]
2573 when @value{GDBN} notices a new thread.
2576 @kindex info threads (HP-UX)
2578 Display a summary of all threads currently in your
2579 program. @value{GDBN} displays for each thread (in this order):
2582 @item the thread number assigned by @value{GDBN}
2584 @item the target system's thread identifier (@var{systag})
2586 @item the current stack frame summary for that thread
2590 An asterisk @samp{*} to the left of the @value{GDBN} thread number
2591 indicates the current thread.
2595 @c end table here to get a little more width for example
2598 (@value{GDBP}) info threads
2599 * 3 system thread 26607 worker (wptr=0x7b09c318 "@@") \@*
2601 2 system thread 26606 0x7b0030d8 in __ksleep () \@*
2602 from /usr/lib/libc.2
2603 1 system thread 27905 0x7b003498 in _brk () \@*
2604 from /usr/lib/libc.2
2607 On Solaris, you can display more information about user threads with a
2608 Solaris-specific command:
2611 @item maint info sol-threads
2612 @kindex maint info sol-threads
2613 @cindex thread info (Solaris)
2614 Display info on Solaris user threads.
2618 @kindex thread @var{threadno}
2619 @item thread @var{threadno}
2620 Make thread number @var{threadno} the current thread. The command
2621 argument @var{threadno} is the internal @value{GDBN} thread number, as
2622 shown in the first field of the @samp{info threads} display.
2623 @value{GDBN} responds by displaying the system identifier of the thread
2624 you selected, and its current stack frame summary:
2627 @c FIXME!! This example made up; find a @value{GDBN} w/threads and get real one
2628 (@value{GDBP}) thread 2
2629 [Switching to process 35 thread 23]
2630 0x34e5 in sigpause ()
2634 As with the @samp{[New @dots{}]} message, the form of the text after
2635 @samp{Switching to} depends on your system's conventions for identifying
2638 @kindex thread apply
2639 @cindex apply command to several threads
2640 @item thread apply [@var{threadno}] [@var{all}] @var{command}
2641 The @code{thread apply} command allows you to apply the named
2642 @var{command} to one or more threads. Specify the numbers of the
2643 threads that you want affected with the command argument
2644 @var{threadno}. It can be a single thread number, one of the numbers
2645 shown in the first field of the @samp{info threads} display; or it
2646 could be a range of thread numbers, as in @code{2-4}. To apply a
2647 command to all threads, type @kbd{thread apply all @var{command}}.
2649 @kindex set print thread-events
2650 @cindex print messages on thread start and exit
2651 @item set print thread-events
2652 @itemx set print thread-events on
2653 @itemx set print thread-events off
2654 The @code{set print thread-events} command allows you to enable or
2655 disable printing of messages when @value{GDBN} notices that new threads have
2656 started or that threads have exited. By default, these messages will
2657 be printed if detection of these events is supported by the target.
2658 Note that these messages cannot be disabled on all targets.
2660 @kindex show print thread-events
2661 @item show print thread-events
2662 Show whether messages will be printed when @value{GDBN} detects that threads
2663 have started and exited.
2666 @xref{Thread Stops,,Stopping and Starting Multi-thread Programs}, for
2667 more information about how @value{GDBN} behaves when you stop and start
2668 programs with multiple threads.
2670 @xref{Set Watchpoints,,Setting Watchpoints}, for information about
2671 watchpoints in programs with multiple threads.
2674 @kindex set libthread-db-search-path
2675 @cindex search path for @code{libthread_db}
2676 @item set libthread-db-search-path @r{[}@var{path}@r{]}
2677 If this variable is set, @var{path} is a colon-separated list of
2678 directories @value{GDBN} will use to search for @code{libthread_db}.
2679 If you omit @var{path}, @samp{libthread-db-search-path} will be reset to
2682 On @sc{gnu}/Linux and Solaris systems, @value{GDBN} uses a ``helper''
2683 @code{libthread_db} library to obtain information about threads in the
2684 inferior process. @value{GDBN} will use @samp{libthread-db-search-path}
2685 to find @code{libthread_db}. If that fails, @value{GDBN} will continue
2686 with default system shared library directories, and finally the directory
2687 from which @code{libpthread} was loaded in the inferior process.
2689 For any @code{libthread_db} library @value{GDBN} finds in above directories,
2690 @value{GDBN} attempts to initialize it with the current inferior process.
2691 If this initialization fails (which could happen because of a version
2692 mismatch between @code{libthread_db} and @code{libpthread}), @value{GDBN}
2693 will unload @code{libthread_db}, and continue with the next directory.
2694 If none of @code{libthread_db} libraries initialize successfully,
2695 @value{GDBN} will issue a warning and thread debugging will be disabled.
2697 Setting @code{libthread-db-search-path} is currently implemented
2698 only on some platforms.
2700 @kindex show libthread-db-search-path
2701 @item show libthread-db-search-path
2702 Display current libthread_db search path.
2706 @section Debugging Programs with Multiple Processes
2708 @cindex fork, debugging programs which call
2709 @cindex multiple processes
2710 @cindex processes, multiple
2711 On most systems, @value{GDBN} has no special support for debugging
2712 programs which create additional processes using the @code{fork}
2713 function. When a program forks, @value{GDBN} will continue to debug the
2714 parent process and the child process will run unimpeded. If you have
2715 set a breakpoint in any code which the child then executes, the child
2716 will get a @code{SIGTRAP} signal which (unless it catches the signal)
2717 will cause it to terminate.
2719 However, if you want to debug the child process there is a workaround
2720 which isn't too painful. Put a call to @code{sleep} in the code which
2721 the child process executes after the fork. It may be useful to sleep
2722 only if a certain environment variable is set, or a certain file exists,
2723 so that the delay need not occur when you don't want to run @value{GDBN}
2724 on the child. While the child is sleeping, use the @code{ps} program to
2725 get its process ID. Then tell @value{GDBN} (a new invocation of
2726 @value{GDBN} if you are also debugging the parent process) to attach to
2727 the child process (@pxref{Attach}). From that point on you can debug
2728 the child process just like any other process which you attached to.
2730 On some systems, @value{GDBN} provides support for debugging programs that
2731 create additional processes using the @code{fork} or @code{vfork} functions.
2732 Currently, the only platforms with this feature are HP-UX (11.x and later
2733 only?) and @sc{gnu}/Linux (kernel version 2.5.60 and later).
2735 By default, when a program forks, @value{GDBN} will continue to debug
2736 the parent process and the child process will run unimpeded.
2738 If you want to follow the child process instead of the parent process,
2739 use the command @w{@code{set follow-fork-mode}}.
2742 @kindex set follow-fork-mode
2743 @item set follow-fork-mode @var{mode}
2744 Set the debugger response to a program call of @code{fork} or
2745 @code{vfork}. A call to @code{fork} or @code{vfork} creates a new
2746 process. The @var{mode} argument can be:
2750 The original process is debugged after a fork. The child process runs
2751 unimpeded. This is the default.
2754 The new process is debugged after a fork. The parent process runs
2759 @kindex show follow-fork-mode
2760 @item show follow-fork-mode
2761 Display the current debugger response to a @code{fork} or @code{vfork} call.
2764 @cindex debugging multiple processes
2765 On Linux, if you want to debug both the parent and child processes, use the
2766 command @w{@code{set detach-on-fork}}.
2769 @kindex set detach-on-fork
2770 @item set detach-on-fork @var{mode}
2771 Tells gdb whether to detach one of the processes after a fork, or
2772 retain debugger control over them both.
2776 The child process (or parent process, depending on the value of
2777 @code{follow-fork-mode}) will be detached and allowed to run
2778 independently. This is the default.
2781 Both processes will be held under the control of @value{GDBN}.
2782 One process (child or parent, depending on the value of
2783 @code{follow-fork-mode}) is debugged as usual, while the other
2788 @kindex show detach-on-fork
2789 @item show detach-on-fork
2790 Show whether detach-on-fork mode is on/off.
2793 If you choose to set @samp{detach-on-fork} mode off, then @value{GDBN}
2794 will retain control of all forked processes (including nested forks).
2795 You can list the forked processes under the control of @value{GDBN} by
2796 using the @w{@code{info inferiors}} command, and switch from one fork
2797 to another by using the @code{inferior} command (@pxref{Inferiors,
2798 ,Debugging Multiple Inferiors}).
2800 To quit debugging one of the forked processes, you can either detach
2801 from it by using the @w{@code{detach inferior}} command (allowing it
2802 to run independently), or kill it using the @w{@code{kill inferior}}
2803 command. @xref{Inferiors, ,Debugging Multiple Inferiors}.
2805 If you ask to debug a child process and a @code{vfork} is followed by an
2806 @code{exec}, @value{GDBN} executes the new target up to the first
2807 breakpoint in the new target. If you have a breakpoint set on
2808 @code{main} in your original program, the breakpoint will also be set on
2809 the child process's @code{main}.
2811 On some systems, when a child process is spawned by @code{vfork}, you
2812 cannot debug the child or parent until an @code{exec} call completes.
2814 If you issue a @code{run} command to @value{GDBN} after an @code{exec}
2815 call executes, the new target restarts. To restart the parent process,
2816 use the @code{file} command with the parent executable name as its
2819 You can use the @code{catch} command to make @value{GDBN} stop whenever
2820 a @code{fork}, @code{vfork}, or @code{exec} call is made. @xref{Set
2821 Catchpoints, ,Setting Catchpoints}.
2823 @node Checkpoint/Restart
2824 @section Setting a @emph{Bookmark} to Return to Later
2829 @cindex snapshot of a process
2830 @cindex rewind program state
2832 On certain operating systems@footnote{Currently, only
2833 @sc{gnu}/Linux.}, @value{GDBN} is able to save a @dfn{snapshot} of a
2834 program's state, called a @dfn{checkpoint}, and come back to it
2837 Returning to a checkpoint effectively undoes everything that has
2838 happened in the program since the @code{checkpoint} was saved. This
2839 includes changes in memory, registers, and even (within some limits)
2840 system state. Effectively, it is like going back in time to the
2841 moment when the checkpoint was saved.
2843 Thus, if you're stepping thru a program and you think you're
2844 getting close to the point where things go wrong, you can save
2845 a checkpoint. Then, if you accidentally go too far and miss
2846 the critical statement, instead of having to restart your program
2847 from the beginning, you can just go back to the checkpoint and
2848 start again from there.
2850 This can be especially useful if it takes a lot of time or
2851 steps to reach the point where you think the bug occurs.
2853 To use the @code{checkpoint}/@code{restart} method of debugging:
2858 Save a snapshot of the debugged program's current execution state.
2859 The @code{checkpoint} command takes no arguments, but each checkpoint
2860 is assigned a small integer id, similar to a breakpoint id.
2862 @kindex info checkpoints
2863 @item info checkpoints
2864 List the checkpoints that have been saved in the current debugging
2865 session. For each checkpoint, the following information will be
2872 @item Source line, or label
2875 @kindex restart @var{checkpoint-id}
2876 @item restart @var{checkpoint-id}
2877 Restore the program state that was saved as checkpoint number
2878 @var{checkpoint-id}. All program variables, registers, stack frames
2879 etc.@: will be returned to the values that they had when the checkpoint
2880 was saved. In essence, gdb will ``wind back the clock'' to the point
2881 in time when the checkpoint was saved.
2883 Note that breakpoints, @value{GDBN} variables, command history etc.
2884 are not affected by restoring a checkpoint. In general, a checkpoint
2885 only restores things that reside in the program being debugged, not in
2888 @kindex delete checkpoint @var{checkpoint-id}
2889 @item delete checkpoint @var{checkpoint-id}
2890 Delete the previously-saved checkpoint identified by @var{checkpoint-id}.
2894 Returning to a previously saved checkpoint will restore the user state
2895 of the program being debugged, plus a significant subset of the system
2896 (OS) state, including file pointers. It won't ``un-write'' data from
2897 a file, but it will rewind the file pointer to the previous location,
2898 so that the previously written data can be overwritten. For files
2899 opened in read mode, the pointer will also be restored so that the
2900 previously read data can be read again.
2902 Of course, characters that have been sent to a printer (or other
2903 external device) cannot be ``snatched back'', and characters received
2904 from eg.@: a serial device can be removed from internal program buffers,
2905 but they cannot be ``pushed back'' into the serial pipeline, ready to
2906 be received again. Similarly, the actual contents of files that have
2907 been changed cannot be restored (at this time).
2909 However, within those constraints, you actually can ``rewind'' your
2910 program to a previously saved point in time, and begin debugging it
2911 again --- and you can change the course of events so as to debug a
2912 different execution path this time.
2914 @cindex checkpoints and process id
2915 Finally, there is one bit of internal program state that will be
2916 different when you return to a checkpoint --- the program's process
2917 id. Each checkpoint will have a unique process id (or @var{pid}),
2918 and each will be different from the program's original @var{pid}.
2919 If your program has saved a local copy of its process id, this could
2920 potentially pose a problem.
2922 @subsection A Non-obvious Benefit of Using Checkpoints
2924 On some systems such as @sc{gnu}/Linux, address space randomization
2925 is performed on new processes for security reasons. This makes it
2926 difficult or impossible to set a breakpoint, or watchpoint, on an
2927 absolute address if you have to restart the program, since the
2928 absolute location of a symbol will change from one execution to the
2931 A checkpoint, however, is an @emph{identical} copy of a process.
2932 Therefore if you create a checkpoint at (eg.@:) the start of main,
2933 and simply return to that checkpoint instead of restarting the
2934 process, you can avoid the effects of address randomization and
2935 your symbols will all stay in the same place.
2938 @chapter Stopping and Continuing
2940 The principal purposes of using a debugger are so that you can stop your
2941 program before it terminates; or so that, if your program runs into
2942 trouble, you can investigate and find out why.
2944 Inside @value{GDBN}, your program may stop for any of several reasons,
2945 such as a signal, a breakpoint, or reaching a new line after a
2946 @value{GDBN} command such as @code{step}. You may then examine and
2947 change variables, set new breakpoints or remove old ones, and then
2948 continue execution. Usually, the messages shown by @value{GDBN} provide
2949 ample explanation of the status of your program---but you can also
2950 explicitly request this information at any time.
2953 @kindex info program
2955 Display information about the status of your program: whether it is
2956 running or not, what process it is, and why it stopped.
2960 * Breakpoints:: Breakpoints, watchpoints, and catchpoints
2961 * Continuing and Stepping:: Resuming execution
2963 * Thread Stops:: Stopping and starting multi-thread programs
2967 @section Breakpoints, Watchpoints, and Catchpoints
2970 A @dfn{breakpoint} makes your program stop whenever a certain point in
2971 the program is reached. For each breakpoint, you can add conditions to
2972 control in finer detail whether your program stops. You can set
2973 breakpoints with the @code{break} command and its variants (@pxref{Set
2974 Breaks, ,Setting Breakpoints}), to specify the place where your program
2975 should stop by line number, function name or exact address in the
2978 On some systems, you can set breakpoints in shared libraries before
2979 the executable is run. There is a minor limitation on HP-UX systems:
2980 you must wait until the executable is run in order to set breakpoints
2981 in shared library routines that are not called directly by the program
2982 (for example, routines that are arguments in a @code{pthread_create}
2986 @cindex data breakpoints
2987 @cindex memory tracing
2988 @cindex breakpoint on memory address
2989 @cindex breakpoint on variable modification
2990 A @dfn{watchpoint} is a special breakpoint that stops your program
2991 when the value of an expression changes. The expression may be a value
2992 of a variable, or it could involve values of one or more variables
2993 combined by operators, such as @samp{a + b}. This is sometimes called
2994 @dfn{data breakpoints}. You must use a different command to set
2995 watchpoints (@pxref{Set Watchpoints, ,Setting Watchpoints}), but aside
2996 from that, you can manage a watchpoint like any other breakpoint: you
2997 enable, disable, and delete both breakpoints and watchpoints using the
3000 You can arrange to have values from your program displayed automatically
3001 whenever @value{GDBN} stops at a breakpoint. @xref{Auto Display,,
3005 @cindex breakpoint on events
3006 A @dfn{catchpoint} is another special breakpoint that stops your program
3007 when a certain kind of event occurs, such as the throwing of a C@t{++}
3008 exception or the loading of a library. As with watchpoints, you use a
3009 different command to set a catchpoint (@pxref{Set Catchpoints, ,Setting
3010 Catchpoints}), but aside from that, you can manage a catchpoint like any
3011 other breakpoint. (To stop when your program receives a signal, use the
3012 @code{handle} command; see @ref{Signals, ,Signals}.)
3014 @cindex breakpoint numbers
3015 @cindex numbers for breakpoints
3016 @value{GDBN} assigns a number to each breakpoint, watchpoint, or
3017 catchpoint when you create it; these numbers are successive integers
3018 starting with one. In many of the commands for controlling various
3019 features of breakpoints you use the breakpoint number to say which
3020 breakpoint you want to change. Each breakpoint may be @dfn{enabled} or
3021 @dfn{disabled}; if disabled, it has no effect on your program until you
3024 @cindex breakpoint ranges
3025 @cindex ranges of breakpoints
3026 Some @value{GDBN} commands accept a range of breakpoints on which to
3027 operate. A breakpoint range is either a single breakpoint number, like
3028 @samp{5}, or two such numbers, in increasing order, separated by a
3029 hyphen, like @samp{5-7}. When a breakpoint range is given to a command,
3030 all breakpoints in that range are operated on.
3033 * Set Breaks:: Setting breakpoints
3034 * Set Watchpoints:: Setting watchpoints
3035 * Set Catchpoints:: Setting catchpoints
3036 * Delete Breaks:: Deleting breakpoints
3037 * Disabling:: Disabling breakpoints
3038 * Conditions:: Break conditions
3039 * Break Commands:: Breakpoint command lists
3040 * Error in Breakpoints:: ``Cannot insert breakpoints''
3041 * Breakpoint-related Warnings:: ``Breakpoint address adjusted...''
3045 @subsection Setting Breakpoints
3047 @c FIXME LMB what does GDB do if no code on line of breakpt?
3048 @c consider in particular declaration with/without initialization.
3050 @c FIXME 2 is there stuff on this already? break at fun start, already init?
3053 @kindex b @r{(@code{break})}
3054 @vindex $bpnum@r{, convenience variable}
3055 @cindex latest breakpoint
3056 Breakpoints are set with the @code{break} command (abbreviated
3057 @code{b}). The debugger convenience variable @samp{$bpnum} records the
3058 number of the breakpoint you've set most recently; see @ref{Convenience
3059 Vars,, Convenience Variables}, for a discussion of what you can do with
3060 convenience variables.
3063 @item break @var{location}
3064 Set a breakpoint at the given @var{location}, which can specify a
3065 function name, a line number, or an address of an instruction.
3066 (@xref{Specify Location}, for a list of all the possible ways to
3067 specify a @var{location}.) The breakpoint will stop your program just
3068 before it executes any of the code in the specified @var{location}.
3070 When using source languages that permit overloading of symbols, such as
3071 C@t{++}, a function name may refer to more than one possible place to break.
3072 @xref{Ambiguous Expressions,,Ambiguous Expressions}, for a discussion of
3075 It is also possible to insert a breakpoint that will stop the program
3076 only if a specific thread (@pxref{Thread-Specific Breakpoints})
3077 or a specific task (@pxref{Ada Tasks}) hits that breakpoint.
3080 When called without any arguments, @code{break} sets a breakpoint at
3081 the next instruction to be executed in the selected stack frame
3082 (@pxref{Stack, ,Examining the Stack}). In any selected frame but the
3083 innermost, this makes your program stop as soon as control
3084 returns to that frame. This is similar to the effect of a
3085 @code{finish} command in the frame inside the selected frame---except
3086 that @code{finish} does not leave an active breakpoint. If you use
3087 @code{break} without an argument in the innermost frame, @value{GDBN} stops
3088 the next time it reaches the current location; this may be useful
3091 @value{GDBN} normally ignores breakpoints when it resumes execution, until at
3092 least one instruction has been executed. If it did not do this, you
3093 would be unable to proceed past a breakpoint without first disabling the
3094 breakpoint. This rule applies whether or not the breakpoint already
3095 existed when your program stopped.
3097 @item break @dots{} if @var{cond}
3098 Set a breakpoint with condition @var{cond}; evaluate the expression
3099 @var{cond} each time the breakpoint is reached, and stop only if the
3100 value is nonzero---that is, if @var{cond} evaluates as true.
3101 @samp{@dots{}} stands for one of the possible arguments described
3102 above (or no argument) specifying where to break. @xref{Conditions,
3103 ,Break Conditions}, for more information on breakpoint conditions.
3106 @item tbreak @var{args}
3107 Set a breakpoint enabled only for one stop. @var{args} are the
3108 same as for the @code{break} command, and the breakpoint is set in the same
3109 way, but the breakpoint is automatically deleted after the first time your
3110 program stops there. @xref{Disabling, ,Disabling Breakpoints}.
3113 @cindex hardware breakpoints
3114 @item hbreak @var{args}
3115 Set a hardware-assisted breakpoint. @var{args} are the same as for the
3116 @code{break} command and the breakpoint is set in the same way, but the
3117 breakpoint requires hardware support and some target hardware may not
3118 have this support. The main purpose of this is EPROM/ROM code
3119 debugging, so you can set a breakpoint at an instruction without
3120 changing the instruction. This can be used with the new trap-generation
3121 provided by SPARClite DSU and most x86-based targets. These targets
3122 will generate traps when a program accesses some data or instruction
3123 address that is assigned to the debug registers. However the hardware
3124 breakpoint registers can take a limited number of breakpoints. For
3125 example, on the DSU, only two data breakpoints can be set at a time, and
3126 @value{GDBN} will reject this command if more than two are used. Delete
3127 or disable unused hardware breakpoints before setting new ones
3128 (@pxref{Disabling, ,Disabling Breakpoints}).
3129 @xref{Conditions, ,Break Conditions}.
3130 For remote targets, you can restrict the number of hardware
3131 breakpoints @value{GDBN} will use, see @ref{set remote
3132 hardware-breakpoint-limit}.
3135 @item thbreak @var{args}
3136 Set a hardware-assisted breakpoint enabled only for one stop. @var{args}
3137 are the same as for the @code{hbreak} command and the breakpoint is set in
3138 the same way. However, like the @code{tbreak} command,
3139 the breakpoint is automatically deleted after the
3140 first time your program stops there. Also, like the @code{hbreak}
3141 command, the breakpoint requires hardware support and some target hardware
3142 may not have this support. @xref{Disabling, ,Disabling Breakpoints}.
3143 See also @ref{Conditions, ,Break Conditions}.
3146 @cindex regular expression
3147 @cindex breakpoints in functions matching a regexp
3148 @cindex set breakpoints in many functions
3149 @item rbreak @var{regex}
3150 Set breakpoints on all functions matching the regular expression
3151 @var{regex}. This command sets an unconditional breakpoint on all
3152 matches, printing a list of all breakpoints it set. Once these
3153 breakpoints are set, they are treated just like the breakpoints set with
3154 the @code{break} command. You can delete them, disable them, or make
3155 them conditional the same way as any other breakpoint.
3157 The syntax of the regular expression is the standard one used with tools
3158 like @file{grep}. Note that this is different from the syntax used by
3159 shells, so for instance @code{foo*} matches all functions that include
3160 an @code{fo} followed by zero or more @code{o}s. There is an implicit
3161 @code{.*} leading and trailing the regular expression you supply, so to
3162 match only functions that begin with @code{foo}, use @code{^foo}.
3164 @cindex non-member C@t{++} functions, set breakpoint in
3165 When debugging C@t{++} programs, @code{rbreak} is useful for setting
3166 breakpoints on overloaded functions that are not members of any special
3169 @cindex set breakpoints on all functions
3170 The @code{rbreak} command can be used to set breakpoints in
3171 @strong{all} the functions in a program, like this:
3174 (@value{GDBP}) rbreak .
3177 @kindex info breakpoints
3178 @cindex @code{$_} and @code{info breakpoints}
3179 @item info breakpoints @r{[}@var{n}@r{]}
3180 @itemx info break @r{[}@var{n}@r{]}
3181 @itemx info watchpoints @r{[}@var{n}@r{]}
3182 Print a table of all breakpoints, watchpoints, and catchpoints set and
3183 not deleted. Optional argument @var{n} means print information only
3184 about the specified breakpoint (or watchpoint or catchpoint). For
3185 each breakpoint, following columns are printed:
3188 @item Breakpoint Numbers
3190 Breakpoint, watchpoint, or catchpoint.
3192 Whether the breakpoint is marked to be disabled or deleted when hit.
3193 @item Enabled or Disabled
3194 Enabled breakpoints are marked with @samp{y}. @samp{n} marks breakpoints
3195 that are not enabled.
3197 Where the breakpoint is in your program, as a memory address. For a
3198 pending breakpoint whose address is not yet known, this field will
3199 contain @samp{<PENDING>}. Such breakpoint won't fire until a shared
3200 library that has the symbol or line referred by breakpoint is loaded.
3201 See below for details. A breakpoint with several locations will
3202 have @samp{<MULTIPLE>} in this field---see below for details.
3204 Where the breakpoint is in the source for your program, as a file and
3205 line number. For a pending breakpoint, the original string passed to
3206 the breakpoint command will be listed as it cannot be resolved until
3207 the appropriate shared library is loaded in the future.
3211 If a breakpoint is conditional, @code{info break} shows the condition on
3212 the line following the affected breakpoint; breakpoint commands, if any,
3213 are listed after that. A pending breakpoint is allowed to have a condition
3214 specified for it. The condition is not parsed for validity until a shared
3215 library is loaded that allows the pending breakpoint to resolve to a
3219 @code{info break} with a breakpoint
3220 number @var{n} as argument lists only that breakpoint. The
3221 convenience variable @code{$_} and the default examining-address for
3222 the @code{x} command are set to the address of the last breakpoint
3223 listed (@pxref{Memory, ,Examining Memory}).
3226 @code{info break} displays a count of the number of times the breakpoint
3227 has been hit. This is especially useful in conjunction with the
3228 @code{ignore} command. You can ignore a large number of breakpoint
3229 hits, look at the breakpoint info to see how many times the breakpoint
3230 was hit, and then run again, ignoring one less than that number. This
3231 will get you quickly to the last hit of that breakpoint.
3234 @value{GDBN} allows you to set any number of breakpoints at the same place in
3235 your program. There is nothing silly or meaningless about this. When
3236 the breakpoints are conditional, this is even useful
3237 (@pxref{Conditions, ,Break Conditions}).
3239 @cindex multiple locations, breakpoints
3240 @cindex breakpoints, multiple locations
3241 It is possible that a breakpoint corresponds to several locations
3242 in your program. Examples of this situation are:
3246 For a C@t{++} constructor, the @value{NGCC} compiler generates several
3247 instances of the function body, used in different cases.
3250 For a C@t{++} template function, a given line in the function can
3251 correspond to any number of instantiations.
3254 For an inlined function, a given source line can correspond to
3255 several places where that function is inlined.
3258 In all those cases, @value{GDBN} will insert a breakpoint at all
3259 the relevant locations@footnote{
3260 As of this writing, multiple-location breakpoints work only if there's
3261 line number information for all the locations. This means that they
3262 will generally not work in system libraries, unless you have debug
3263 info with line numbers for them.}.
3265 A breakpoint with multiple locations is displayed in the breakpoint
3266 table using several rows---one header row, followed by one row for
3267 each breakpoint location. The header row has @samp{<MULTIPLE>} in the
3268 address column. The rows for individual locations contain the actual
3269 addresses for locations, and show the functions to which those
3270 locations belong. The number column for a location is of the form
3271 @var{breakpoint-number}.@var{location-number}.
3276 Num Type Disp Enb Address What
3277 1 breakpoint keep y <MULTIPLE>
3279 breakpoint already hit 1 time
3280 1.1 y 0x080486a2 in void foo<int>() at t.cc:8
3281 1.2 y 0x080486ca in void foo<double>() at t.cc:8
3284 Each location can be individually enabled or disabled by passing
3285 @var{breakpoint-number}.@var{location-number} as argument to the
3286 @code{enable} and @code{disable} commands. Note that you cannot
3287 delete the individual locations from the list, you can only delete the
3288 entire list of locations that belong to their parent breakpoint (with
3289 the @kbd{delete @var{num}} command, where @var{num} is the number of
3290 the parent breakpoint, 1 in the above example). Disabling or enabling
3291 the parent breakpoint (@pxref{Disabling}) affects all of the locations
3292 that belong to that breakpoint.
3294 @cindex pending breakpoints
3295 It's quite common to have a breakpoint inside a shared library.
3296 Shared libraries can be loaded and unloaded explicitly,
3297 and possibly repeatedly, as the program is executed. To support
3298 this use case, @value{GDBN} updates breakpoint locations whenever
3299 any shared library is loaded or unloaded. Typically, you would
3300 set a breakpoint in a shared library at the beginning of your
3301 debugging session, when the library is not loaded, and when the
3302 symbols from the library are not available. When you try to set
3303 breakpoint, @value{GDBN} will ask you if you want to set
3304 a so called @dfn{pending breakpoint}---breakpoint whose address
3305 is not yet resolved.
3307 After the program is run, whenever a new shared library is loaded,
3308 @value{GDBN} reevaluates all the breakpoints. When a newly loaded
3309 shared library contains the symbol or line referred to by some
3310 pending breakpoint, that breakpoint is resolved and becomes an
3311 ordinary breakpoint. When a library is unloaded, all breakpoints
3312 that refer to its symbols or source lines become pending again.
3314 This logic works for breakpoints with multiple locations, too. For
3315 example, if you have a breakpoint in a C@t{++} template function, and
3316 a newly loaded shared library has an instantiation of that template,
3317 a new location is added to the list of locations for the breakpoint.
3319 Except for having unresolved address, pending breakpoints do not
3320 differ from regular breakpoints. You can set conditions or commands,
3321 enable and disable them and perform other breakpoint operations.
3323 @value{GDBN} provides some additional commands for controlling what
3324 happens when the @samp{break} command cannot resolve breakpoint
3325 address specification to an address:
3327 @kindex set breakpoint pending
3328 @kindex show breakpoint pending
3330 @item set breakpoint pending auto
3331 This is the default behavior. When @value{GDBN} cannot find the breakpoint
3332 location, it queries you whether a pending breakpoint should be created.
3334 @item set breakpoint pending on
3335 This indicates that an unrecognized breakpoint location should automatically
3336 result in a pending breakpoint being created.
3338 @item set breakpoint pending off
3339 This indicates that pending breakpoints are not to be created. Any
3340 unrecognized breakpoint location results in an error. This setting does
3341 not affect any pending breakpoints previously created.
3343 @item show breakpoint pending
3344 Show the current behavior setting for creating pending breakpoints.
3347 The settings above only affect the @code{break} command and its
3348 variants. Once breakpoint is set, it will be automatically updated
3349 as shared libraries are loaded and unloaded.
3351 @cindex automatic hardware breakpoints
3352 For some targets, @value{GDBN} can automatically decide if hardware or
3353 software breakpoints should be used, depending on whether the
3354 breakpoint address is read-only or read-write. This applies to
3355 breakpoints set with the @code{break} command as well as to internal
3356 breakpoints set by commands like @code{next} and @code{finish}. For
3357 breakpoints set with @code{hbreak}, @value{GDBN} will always use hardware
3360 You can control this automatic behaviour with the following commands::
3362 @kindex set breakpoint auto-hw
3363 @kindex show breakpoint auto-hw
3365 @item set breakpoint auto-hw on
3366 This is the default behavior. When @value{GDBN} sets a breakpoint, it
3367 will try to use the target memory map to decide if software or hardware
3368 breakpoint must be used.
3370 @item set breakpoint auto-hw off
3371 This indicates @value{GDBN} should not automatically select breakpoint
3372 type. If the target provides a memory map, @value{GDBN} will warn when
3373 trying to set software breakpoint at a read-only address.
3376 @value{GDBN} normally implements breakpoints by replacing the program code
3377 at the breakpoint address with a special instruction, which, when
3378 executed, given control to the debugger. By default, the program
3379 code is so modified only when the program is resumed. As soon as
3380 the program stops, @value{GDBN} restores the original instructions. This
3381 behaviour guards against leaving breakpoints inserted in the
3382 target should gdb abrubptly disconnect. However, with slow remote
3383 targets, inserting and removing breakpoint can reduce the performance.
3384 This behavior can be controlled with the following commands::
3386 @kindex set breakpoint always-inserted
3387 @kindex show breakpoint always-inserted
3389 @item set breakpoint always-inserted off
3390 All breakpoints, including newly added by the user, are inserted in
3391 the target only when the target is resumed. All breakpoints are
3392 removed from the target when it stops.
3394 @item set breakpoint always-inserted on
3395 Causes all breakpoints to be inserted in the target at all times. If
3396 the user adds a new breakpoint, or changes an existing breakpoint, the
3397 breakpoints in the target are updated immediately. A breakpoint is
3398 removed from the target only when breakpoint itself is removed.
3400 @cindex non-stop mode, and @code{breakpoint always-inserted}
3401 @item set breakpoint always-inserted auto
3402 This is the default mode. If @value{GDBN} is controlling the inferior
3403 in non-stop mode (@pxref{Non-Stop Mode}), gdb behaves as if
3404 @code{breakpoint always-inserted} mode is on. If @value{GDBN} is
3405 controlling the inferior in all-stop mode, @value{GDBN} behaves as if
3406 @code{breakpoint always-inserted} mode is off.
3409 @cindex negative breakpoint numbers
3410 @cindex internal @value{GDBN} breakpoints
3411 @value{GDBN} itself sometimes sets breakpoints in your program for
3412 special purposes, such as proper handling of @code{longjmp} (in C
3413 programs). These internal breakpoints are assigned negative numbers,
3414 starting with @code{-1}; @samp{info breakpoints} does not display them.
3415 You can see these breakpoints with the @value{GDBN} maintenance command
3416 @samp{maint info breakpoints} (@pxref{maint info breakpoints}).
3419 @node Set Watchpoints
3420 @subsection Setting Watchpoints
3422 @cindex setting watchpoints
3423 You can use a watchpoint to stop execution whenever the value of an
3424 expression changes, without having to predict a particular place where
3425 this may happen. (This is sometimes called a @dfn{data breakpoint}.)
3426 The expression may be as simple as the value of a single variable, or
3427 as complex as many variables combined by operators. Examples include:
3431 A reference to the value of a single variable.
3434 An address cast to an appropriate data type. For example,
3435 @samp{*(int *)0x12345678} will watch a 4-byte region at the specified
3436 address (assuming an @code{int} occupies 4 bytes).
3439 An arbitrarily complex expression, such as @samp{a*b + c/d}. The
3440 expression can use any operators valid in the program's native
3441 language (@pxref{Languages}).
3444 You can set a watchpoint on an expression even if the expression can
3445 not be evaluated yet. For instance, you can set a watchpoint on
3446 @samp{*global_ptr} before @samp{global_ptr} is initialized.
3447 @value{GDBN} will stop when your program sets @samp{global_ptr} and
3448 the expression produces a valid value. If the expression becomes
3449 valid in some other way than changing a variable (e.g.@: if the memory
3450 pointed to by @samp{*global_ptr} becomes readable as the result of a
3451 @code{malloc} call), @value{GDBN} may not stop until the next time
3452 the expression changes.
3454 @cindex software watchpoints
3455 @cindex hardware watchpoints
3456 Depending on your system, watchpoints may be implemented in software or
3457 hardware. @value{GDBN} does software watchpointing by single-stepping your
3458 program and testing the variable's value each time, which is hundreds of
3459 times slower than normal execution. (But this may still be worth it, to
3460 catch errors where you have no clue what part of your program is the
3463 On some systems, such as HP-UX, PowerPC, @sc{gnu}/Linux and most other
3464 x86-based targets, @value{GDBN} includes support for hardware
3465 watchpoints, which do not slow down the running of your program.
3469 @item watch @var{expr} @r{[}thread @var{threadnum}@r{]}
3470 Set a watchpoint for an expression. @value{GDBN} will break when the
3471 expression @var{expr} is written into by the program and its value
3472 changes. The simplest (and the most popular) use of this command is
3473 to watch the value of a single variable:
3476 (@value{GDBP}) watch foo
3479 If the command includes a @code{@r{[}thread @var{threadnum}@r{]}}
3480 clause, @value{GDBN} breaks only when the thread identified by
3481 @var{threadnum} changes the value of @var{expr}. If any other threads
3482 change the value of @var{expr}, @value{GDBN} will not break. Note
3483 that watchpoints restricted to a single thread in this way only work
3484 with Hardware Watchpoints.
3487 @item rwatch @var{expr} @r{[}thread @var{threadnum}@r{]}
3488 Set a watchpoint that will break when the value of @var{expr} is read
3492 @item awatch @var{expr} @r{[}thread @var{threadnum}@r{]}
3493 Set a watchpoint that will break when @var{expr} is either read from
3494 or written into by the program.
3496 @kindex info watchpoints @r{[}@var{n}@r{]}
3497 @item info watchpoints
3498 This command prints a list of watchpoints, breakpoints, and catchpoints;
3499 it is the same as @code{info break} (@pxref{Set Breaks}).
3502 @value{GDBN} sets a @dfn{hardware watchpoint} if possible. Hardware
3503 watchpoints execute very quickly, and the debugger reports a change in
3504 value at the exact instruction where the change occurs. If @value{GDBN}
3505 cannot set a hardware watchpoint, it sets a software watchpoint, which
3506 executes more slowly and reports the change in value at the next
3507 @emph{statement}, not the instruction, after the change occurs.
3509 @cindex use only software watchpoints
3510 You can force @value{GDBN} to use only software watchpoints with the
3511 @kbd{set can-use-hw-watchpoints 0} command. With this variable set to
3512 zero, @value{GDBN} will never try to use hardware watchpoints, even if
3513 the underlying system supports them. (Note that hardware-assisted
3514 watchpoints that were set @emph{before} setting
3515 @code{can-use-hw-watchpoints} to zero will still use the hardware
3516 mechanism of watching expression values.)
3519 @item set can-use-hw-watchpoints
3520 @kindex set can-use-hw-watchpoints
3521 Set whether or not to use hardware watchpoints.
3523 @item show can-use-hw-watchpoints
3524 @kindex show can-use-hw-watchpoints
3525 Show the current mode of using hardware watchpoints.
3528 For remote targets, you can restrict the number of hardware
3529 watchpoints @value{GDBN} will use, see @ref{set remote
3530 hardware-breakpoint-limit}.
3532 When you issue the @code{watch} command, @value{GDBN} reports
3535 Hardware watchpoint @var{num}: @var{expr}
3539 if it was able to set a hardware watchpoint.
3541 Currently, the @code{awatch} and @code{rwatch} commands can only set
3542 hardware watchpoints, because accesses to data that don't change the
3543 value of the watched expression cannot be detected without examining
3544 every instruction as it is being executed, and @value{GDBN} does not do
3545 that currently. If @value{GDBN} finds that it is unable to set a
3546 hardware breakpoint with the @code{awatch} or @code{rwatch} command, it
3547 will print a message like this:
3550 Expression cannot be implemented with read/access watchpoint.
3553 Sometimes, @value{GDBN} cannot set a hardware watchpoint because the
3554 data type of the watched expression is wider than what a hardware
3555 watchpoint on the target machine can handle. For example, some systems
3556 can only watch regions that are up to 4 bytes wide; on such systems you
3557 cannot set hardware watchpoints for an expression that yields a
3558 double-precision floating-point number (which is typically 8 bytes
3559 wide). As a work-around, it might be possible to break the large region
3560 into a series of smaller ones and watch them with separate watchpoints.
3562 If you set too many hardware watchpoints, @value{GDBN} might be unable
3563 to insert all of them when you resume the execution of your program.
3564 Since the precise number of active watchpoints is unknown until such
3565 time as the program is about to be resumed, @value{GDBN} might not be
3566 able to warn you about this when you set the watchpoints, and the
3567 warning will be printed only when the program is resumed:
3570 Hardware watchpoint @var{num}: Could not insert watchpoint
3574 If this happens, delete or disable some of the watchpoints.
3576 Watching complex expressions that reference many variables can also
3577 exhaust the resources available for hardware-assisted watchpoints.
3578 That's because @value{GDBN} needs to watch every variable in the
3579 expression with separately allocated resources.
3581 If you call a function interactively using @code{print} or @code{call},
3582 any watchpoints you have set will be inactive until @value{GDBN} reaches another
3583 kind of breakpoint or the call completes.
3585 @value{GDBN} automatically deletes watchpoints that watch local
3586 (automatic) variables, or expressions that involve such variables, when
3587 they go out of scope, that is, when the execution leaves the block in
3588 which these variables were defined. In particular, when the program
3589 being debugged terminates, @emph{all} local variables go out of scope,
3590 and so only watchpoints that watch global variables remain set. If you
3591 rerun the program, you will need to set all such watchpoints again. One
3592 way of doing that would be to set a code breakpoint at the entry to the
3593 @code{main} function and when it breaks, set all the watchpoints.
3595 @cindex watchpoints and threads
3596 @cindex threads and watchpoints
3597 In multi-threaded programs, watchpoints will detect changes to the
3598 watched expression from every thread.
3601 @emph{Warning:} In multi-threaded programs, software watchpoints
3602 have only limited usefulness. If @value{GDBN} creates a software
3603 watchpoint, it can only watch the value of an expression @emph{in a
3604 single thread}. If you are confident that the expression can only
3605 change due to the current thread's activity (and if you are also
3606 confident that no other thread can become current), then you can use
3607 software watchpoints as usual. However, @value{GDBN} may not notice
3608 when a non-current thread's activity changes the expression. (Hardware
3609 watchpoints, in contrast, watch an expression in all threads.)
3612 @xref{set remote hardware-watchpoint-limit}.
3614 @node Set Catchpoints
3615 @subsection Setting Catchpoints
3616 @cindex catchpoints, setting
3617 @cindex exception handlers
3618 @cindex event handling
3620 You can use @dfn{catchpoints} to cause the debugger to stop for certain
3621 kinds of program events, such as C@t{++} exceptions or the loading of a
3622 shared library. Use the @code{catch} command to set a catchpoint.
3626 @item catch @var{event}
3627 Stop when @var{event} occurs. @var{event} can be any of the following:
3630 @cindex stop on C@t{++} exceptions
3631 The throwing of a C@t{++} exception.
3634 The catching of a C@t{++} exception.
3637 @cindex Ada exception catching
3638 @cindex catch Ada exceptions
3639 An Ada exception being raised. If an exception name is specified
3640 at the end of the command (eg @code{catch exception Program_Error}),
3641 the debugger will stop only when this specific exception is raised.
3642 Otherwise, the debugger stops execution when any Ada exception is raised.
3644 When inserting an exception catchpoint on a user-defined exception whose
3645 name is identical to one of the exceptions defined by the language, the
3646 fully qualified name must be used as the exception name. Otherwise,
3647 @value{GDBN} will assume that it should stop on the pre-defined exception
3648 rather than the user-defined one. For instance, assuming an exception
3649 called @code{Constraint_Error} is defined in package @code{Pck}, then
3650 the command to use to catch such exceptions is @kbd{catch exception
3651 Pck.Constraint_Error}.
3653 @item exception unhandled
3654 An exception that was raised but is not handled by the program.
3657 A failed Ada assertion.
3660 @cindex break on fork/exec
3661 A call to @code{exec}. This is currently only available for HP-UX
3665 A call to @code{fork}. This is currently only available for HP-UX
3669 A call to @code{vfork}. This is currently only available for HP-UX
3674 @item tcatch @var{event}
3675 Set a catchpoint that is enabled only for one stop. The catchpoint is
3676 automatically deleted after the first time the event is caught.
3680 Use the @code{info break} command to list the current catchpoints.
3682 There are currently some limitations to C@t{++} exception handling
3683 (@code{catch throw} and @code{catch catch}) in @value{GDBN}:
3687 If you call a function interactively, @value{GDBN} normally returns
3688 control to you when the function has finished executing. If the call
3689 raises an exception, however, the call may bypass the mechanism that
3690 returns control to you and cause your program either to abort or to
3691 simply continue running until it hits a breakpoint, catches a signal
3692 that @value{GDBN} is listening for, or exits. This is the case even if
3693 you set a catchpoint for the exception; catchpoints on exceptions are
3694 disabled within interactive calls.
3697 You cannot raise an exception interactively.
3700 You cannot install an exception handler interactively.
3703 @cindex raise exceptions
3704 Sometimes @code{catch} is not the best way to debug exception handling:
3705 if you need to know exactly where an exception is raised, it is better to
3706 stop @emph{before} the exception handler is called, since that way you
3707 can see the stack before any unwinding takes place. If you set a
3708 breakpoint in an exception handler instead, it may not be easy to find
3709 out where the exception was raised.
3711 To stop just before an exception handler is called, you need some
3712 knowledge of the implementation. In the case of @sc{gnu} C@t{++}, exceptions are
3713 raised by calling a library function named @code{__raise_exception}
3714 which has the following ANSI C interface:
3717 /* @var{addr} is where the exception identifier is stored.
3718 @var{id} is the exception identifier. */
3719 void __raise_exception (void **addr, void *id);
3723 To make the debugger catch all exceptions before any stack
3724 unwinding takes place, set a breakpoint on @code{__raise_exception}
3725 (@pxref{Breakpoints, ,Breakpoints; Watchpoints; and Exceptions}).
3727 With a conditional breakpoint (@pxref{Conditions, ,Break Conditions})
3728 that depends on the value of @var{id}, you can stop your program when
3729 a specific exception is raised. You can use multiple conditional
3730 breakpoints to stop your program when any of a number of exceptions are
3735 @subsection Deleting Breakpoints
3737 @cindex clearing breakpoints, watchpoints, catchpoints
3738 @cindex deleting breakpoints, watchpoints, catchpoints
3739 It is often necessary to eliminate a breakpoint, watchpoint, or
3740 catchpoint once it has done its job and you no longer want your program
3741 to stop there. This is called @dfn{deleting} the breakpoint. A
3742 breakpoint that has been deleted no longer exists; it is forgotten.
3744 With the @code{clear} command you can delete breakpoints according to
3745 where they are in your program. With the @code{delete} command you can
3746 delete individual breakpoints, watchpoints, or catchpoints by specifying
3747 their breakpoint numbers.
3749 It is not necessary to delete a breakpoint to proceed past it. @value{GDBN}
3750 automatically ignores breakpoints on the first instruction to be executed
3751 when you continue execution without changing the execution address.
3756 Delete any breakpoints at the next instruction to be executed in the
3757 selected stack frame (@pxref{Selection, ,Selecting a Frame}). When
3758 the innermost frame is selected, this is a good way to delete a
3759 breakpoint where your program just stopped.
3761 @item clear @var{location}
3762 Delete any breakpoints set at the specified @var{location}.
3763 @xref{Specify Location}, for the various forms of @var{location}; the
3764 most useful ones are listed below:
3767 @item clear @var{function}
3768 @itemx clear @var{filename}:@var{function}
3769 Delete any breakpoints set at entry to the named @var{function}.
3771 @item clear @var{linenum}
3772 @itemx clear @var{filename}:@var{linenum}
3773 Delete any breakpoints set at or within the code of the specified
3774 @var{linenum} of the specified @var{filename}.
3777 @cindex delete breakpoints
3779 @kindex d @r{(@code{delete})}
3780 @item delete @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
3781 Delete the breakpoints, watchpoints, or catchpoints of the breakpoint
3782 ranges specified as arguments. If no argument is specified, delete all
3783 breakpoints (@value{GDBN} asks confirmation, unless you have @code{set
3784 confirm off}). You can abbreviate this command as @code{d}.
3788 @subsection Disabling Breakpoints
3790 @cindex enable/disable a breakpoint
3791 Rather than deleting a breakpoint, watchpoint, or catchpoint, you might
3792 prefer to @dfn{disable} it. This makes the breakpoint inoperative as if
3793 it had been deleted, but remembers the information on the breakpoint so
3794 that you can @dfn{enable} it again later.
3796 You disable and enable breakpoints, watchpoints, and catchpoints with
3797 the @code{enable} and @code{disable} commands, optionally specifying one
3798 or more breakpoint numbers as arguments. Use @code{info break} or
3799 @code{info watch} to print a list of breakpoints, watchpoints, and
3800 catchpoints if you do not know which numbers to use.
3802 Disabling and enabling a breakpoint that has multiple locations
3803 affects all of its locations.
3805 A breakpoint, watchpoint, or catchpoint can have any of four different
3806 states of enablement:
3810 Enabled. The breakpoint stops your program. A breakpoint set
3811 with the @code{break} command starts out in this state.
3813 Disabled. The breakpoint has no effect on your program.
3815 Enabled once. The breakpoint stops your program, but then becomes
3818 Enabled for deletion. The breakpoint stops your program, but
3819 immediately after it does so it is deleted permanently. A breakpoint
3820 set with the @code{tbreak} command starts out in this state.
3823 You can use the following commands to enable or disable breakpoints,
3824 watchpoints, and catchpoints:
3828 @kindex dis @r{(@code{disable})}
3829 @item disable @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
3830 Disable the specified breakpoints---or all breakpoints, if none are
3831 listed. A disabled breakpoint has no effect but is not forgotten. All
3832 options such as ignore-counts, conditions and commands are remembered in
3833 case the breakpoint is enabled again later. You may abbreviate
3834 @code{disable} as @code{dis}.
3837 @item enable @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
3838 Enable the specified breakpoints (or all defined breakpoints). They
3839 become effective once again in stopping your program.
3841 @item enable @r{[}breakpoints@r{]} once @var{range}@dots{}
3842 Enable the specified breakpoints temporarily. @value{GDBN} disables any
3843 of these breakpoints immediately after stopping your program.
3845 @item enable @r{[}breakpoints@r{]} delete @var{range}@dots{}
3846 Enable the specified breakpoints to work once, then die. @value{GDBN}
3847 deletes any of these breakpoints as soon as your program stops there.
3848 Breakpoints set by the @code{tbreak} command start out in this state.
3851 @c FIXME: I think the following ``Except for [...] @code{tbreak}'' is
3852 @c confusing: tbreak is also initially enabled.
3853 Except for a breakpoint set with @code{tbreak} (@pxref{Set Breaks,
3854 ,Setting Breakpoints}), breakpoints that you set are initially enabled;
3855 subsequently, they become disabled or enabled only when you use one of
3856 the commands above. (The command @code{until} can set and delete a
3857 breakpoint of its own, but it does not change the state of your other
3858 breakpoints; see @ref{Continuing and Stepping, ,Continuing and
3862 @subsection Break Conditions
3863 @cindex conditional breakpoints
3864 @cindex breakpoint conditions
3866 @c FIXME what is scope of break condition expr? Context where wanted?
3867 @c in particular for a watchpoint?
3868 The simplest sort of breakpoint breaks every time your program reaches a
3869 specified place. You can also specify a @dfn{condition} for a
3870 breakpoint. A condition is just a Boolean expression in your
3871 programming language (@pxref{Expressions, ,Expressions}). A breakpoint with
3872 a condition evaluates the expression each time your program reaches it,
3873 and your program stops only if the condition is @emph{true}.
3875 This is the converse of using assertions for program validation; in that
3876 situation, you want to stop when the assertion is violated---that is,
3877 when the condition is false. In C, if you want to test an assertion expressed
3878 by the condition @var{assert}, you should set the condition
3879 @samp{! @var{assert}} on the appropriate breakpoint.
3881 Conditions are also accepted for watchpoints; you may not need them,
3882 since a watchpoint is inspecting the value of an expression anyhow---but
3883 it might be simpler, say, to just set a watchpoint on a variable name,
3884 and specify a condition that tests whether the new value is an interesting
3887 Break conditions can have side effects, and may even call functions in
3888 your program. This can be useful, for example, to activate functions
3889 that log program progress, or to use your own print functions to
3890 format special data structures. The effects are completely predictable
3891 unless there is another enabled breakpoint at the same address. (In
3892 that case, @value{GDBN} might see the other breakpoint first and stop your
3893 program without checking the condition of this one.) Note that
3894 breakpoint commands are usually more convenient and flexible than break
3896 purpose of performing side effects when a breakpoint is reached
3897 (@pxref{Break Commands, ,Breakpoint Command Lists}).
3899 Break conditions can be specified when a breakpoint is set, by using
3900 @samp{if} in the arguments to the @code{break} command. @xref{Set
3901 Breaks, ,Setting Breakpoints}. They can also be changed at any time
3902 with the @code{condition} command.
3904 You can also use the @code{if} keyword with the @code{watch} command.
3905 The @code{catch} command does not recognize the @code{if} keyword;
3906 @code{condition} is the only way to impose a further condition on a
3911 @item condition @var{bnum} @var{expression}
3912 Specify @var{expression} as the break condition for breakpoint,
3913 watchpoint, or catchpoint number @var{bnum}. After you set a condition,
3914 breakpoint @var{bnum} stops your program only if the value of
3915 @var{expression} is true (nonzero, in C). When you use
3916 @code{condition}, @value{GDBN} checks @var{expression} immediately for
3917 syntactic correctness, and to determine whether symbols in it have
3918 referents in the context of your breakpoint. If @var{expression} uses
3919 symbols not referenced in the context of the breakpoint, @value{GDBN}
3920 prints an error message:
3923 No symbol "foo" in current context.
3928 not actually evaluate @var{expression} at the time the @code{condition}
3929 command (or a command that sets a breakpoint with a condition, like
3930 @code{break if @dots{}}) is given, however. @xref{Expressions, ,Expressions}.
3932 @item condition @var{bnum}
3933 Remove the condition from breakpoint number @var{bnum}. It becomes
3934 an ordinary unconditional breakpoint.
3937 @cindex ignore count (of breakpoint)
3938 A special case of a breakpoint condition is to stop only when the
3939 breakpoint has been reached a certain number of times. This is so
3940 useful that there is a special way to do it, using the @dfn{ignore
3941 count} of the breakpoint. Every breakpoint has an ignore count, which
3942 is an integer. Most of the time, the ignore count is zero, and
3943 therefore has no effect. But if your program reaches a breakpoint whose
3944 ignore count is positive, then instead of stopping, it just decrements
3945 the ignore count by one and continues. As a result, if the ignore count
3946 value is @var{n}, the breakpoint does not stop the next @var{n} times
3947 your program reaches it.
3951 @item ignore @var{bnum} @var{count}
3952 Set the ignore count of breakpoint number @var{bnum} to @var{count}.
3953 The next @var{count} times the breakpoint is reached, your program's
3954 execution does not stop; other than to decrement the ignore count, @value{GDBN}
3957 To make the breakpoint stop the next time it is reached, specify
3960 When you use @code{continue} to resume execution of your program from a
3961 breakpoint, you can specify an ignore count directly as an argument to
3962 @code{continue}, rather than using @code{ignore}. @xref{Continuing and
3963 Stepping,,Continuing and Stepping}.
3965 If a breakpoint has a positive ignore count and a condition, the
3966 condition is not checked. Once the ignore count reaches zero,
3967 @value{GDBN} resumes checking the condition.
3969 You could achieve the effect of the ignore count with a condition such
3970 as @w{@samp{$foo-- <= 0}} using a debugger convenience variable that
3971 is decremented each time. @xref{Convenience Vars, ,Convenience
3975 Ignore counts apply to breakpoints, watchpoints, and catchpoints.
3978 @node Break Commands
3979 @subsection Breakpoint Command Lists
3981 @cindex breakpoint commands
3982 You can give any breakpoint (or watchpoint or catchpoint) a series of
3983 commands to execute when your program stops due to that breakpoint. For
3984 example, you might want to print the values of certain expressions, or
3985 enable other breakpoints.
3989 @kindex end@r{ (breakpoint commands)}
3990 @item commands @r{[}@var{bnum}@r{]}
3991 @itemx @dots{} @var{command-list} @dots{}
3993 Specify a list of commands for breakpoint number @var{bnum}. The commands
3994 themselves appear on the following lines. Type a line containing just
3995 @code{end} to terminate the commands.
3997 To remove all commands from a breakpoint, type @code{commands} and
3998 follow it immediately with @code{end}; that is, give no commands.
4000 With no @var{bnum} argument, @code{commands} refers to the last
4001 breakpoint, watchpoint, or catchpoint set (not to the breakpoint most
4002 recently encountered).
4005 Pressing @key{RET} as a means of repeating the last @value{GDBN} command is
4006 disabled within a @var{command-list}.
4008 You can use breakpoint commands to start your program up again. Simply
4009 use the @code{continue} command, or @code{step}, or any other command
4010 that resumes execution.
4012 Any other commands in the command list, after a command that resumes
4013 execution, are ignored. This is because any time you resume execution
4014 (even with a simple @code{next} or @code{step}), you may encounter
4015 another breakpoint---which could have its own command list, leading to
4016 ambiguities about which list to execute.
4019 If the first command you specify in a command list is @code{silent}, the
4020 usual message about stopping at a breakpoint is not printed. This may
4021 be desirable for breakpoints that are to print a specific message and
4022 then continue. If none of the remaining commands print anything, you
4023 see no sign that the breakpoint was reached. @code{silent} is
4024 meaningful only at the beginning of a breakpoint command list.
4026 The commands @code{echo}, @code{output}, and @code{printf} allow you to
4027 print precisely controlled output, and are often useful in silent
4028 breakpoints. @xref{Output, ,Commands for Controlled Output}.
4030 For example, here is how you could use breakpoint commands to print the
4031 value of @code{x} at entry to @code{foo} whenever @code{x} is positive.
4037 printf "x is %d\n",x
4042 One application for breakpoint commands is to compensate for one bug so
4043 you can test for another. Put a breakpoint just after the erroneous line
4044 of code, give it a condition to detect the case in which something
4045 erroneous has been done, and give it commands to assign correct values
4046 to any variables that need them. End with the @code{continue} command
4047 so that your program does not stop, and start with the @code{silent}
4048 command so that no output is produced. Here is an example:
4059 @c @ifclear BARETARGET
4060 @node Error in Breakpoints
4061 @subsection ``Cannot insert breakpoints''
4063 If you request too many active hardware-assisted breakpoints and
4064 watchpoints, you will see this error message:
4066 @c FIXME: the precise wording of this message may change; the relevant
4067 @c source change is not committed yet (Sep 3, 1999).
4069 Stopped; cannot insert breakpoints.
4070 You may have requested too many hardware breakpoints and watchpoints.
4074 This message is printed when you attempt to resume the program, since
4075 only then @value{GDBN} knows exactly how many hardware breakpoints and
4076 watchpoints it needs to insert.
4078 When this message is printed, you need to disable or remove some of the
4079 hardware-assisted breakpoints and watchpoints, and then continue.
4081 @node Breakpoint-related Warnings
4082 @subsection ``Breakpoint address adjusted...''
4083 @cindex breakpoint address adjusted
4085 Some processor architectures place constraints on the addresses at
4086 which breakpoints may be placed. For architectures thus constrained,
4087 @value{GDBN} will attempt to adjust the breakpoint's address to comply
4088 with the constraints dictated by the architecture.
4090 One example of such an architecture is the Fujitsu FR-V. The FR-V is
4091 a VLIW architecture in which a number of RISC-like instructions may be
4092 bundled together for parallel execution. The FR-V architecture
4093 constrains the location of a breakpoint instruction within such a
4094 bundle to the instruction with the lowest address. @value{GDBN}
4095 honors this constraint by adjusting a breakpoint's address to the
4096 first in the bundle.
4098 It is not uncommon for optimized code to have bundles which contain
4099 instructions from different source statements, thus it may happen that
4100 a breakpoint's address will be adjusted from one source statement to
4101 another. Since this adjustment may significantly alter @value{GDBN}'s
4102 breakpoint related behavior from what the user expects, a warning is
4103 printed when the breakpoint is first set and also when the breakpoint
4106 A warning like the one below is printed when setting a breakpoint
4107 that's been subject to address adjustment:
4110 warning: Breakpoint address adjusted from 0x00010414 to 0x00010410.
4113 Such warnings are printed both for user settable and @value{GDBN}'s
4114 internal breakpoints. If you see one of these warnings, you should
4115 verify that a breakpoint set at the adjusted address will have the
4116 desired affect. If not, the breakpoint in question may be removed and
4117 other breakpoints may be set which will have the desired behavior.
4118 E.g., it may be sufficient to place the breakpoint at a later
4119 instruction. A conditional breakpoint may also be useful in some
4120 cases to prevent the breakpoint from triggering too often.
4122 @value{GDBN} will also issue a warning when stopping at one of these
4123 adjusted breakpoints:
4126 warning: Breakpoint 1 address previously adjusted from 0x00010414
4130 When this warning is encountered, it may be too late to take remedial
4131 action except in cases where the breakpoint is hit earlier or more
4132 frequently than expected.
4134 @node Continuing and Stepping
4135 @section Continuing and Stepping
4139 @cindex resuming execution
4140 @dfn{Continuing} means resuming program execution until your program
4141 completes normally. In contrast, @dfn{stepping} means executing just
4142 one more ``step'' of your program, where ``step'' may mean either one
4143 line of source code, or one machine instruction (depending on what
4144 particular command you use). Either when continuing or when stepping,
4145 your program may stop even sooner, due to a breakpoint or a signal. (If
4146 it stops due to a signal, you may want to use @code{handle}, or use
4147 @samp{signal 0} to resume execution. @xref{Signals, ,Signals}.)
4151 @kindex c @r{(@code{continue})}
4152 @kindex fg @r{(resume foreground execution)}
4153 @item continue @r{[}@var{ignore-count}@r{]}
4154 @itemx c @r{[}@var{ignore-count}@r{]}
4155 @itemx fg @r{[}@var{ignore-count}@r{]}
4156 Resume program execution, at the address where your program last stopped;
4157 any breakpoints set at that address are bypassed. The optional argument
4158 @var{ignore-count} allows you to specify a further number of times to
4159 ignore a breakpoint at this location; its effect is like that of
4160 @code{ignore} (@pxref{Conditions, ,Break Conditions}).
4162 The argument @var{ignore-count} is meaningful only when your program
4163 stopped due to a breakpoint. At other times, the argument to
4164 @code{continue} is ignored.
4166 The synonyms @code{c} and @code{fg} (for @dfn{foreground}, as the
4167 debugged program is deemed to be the foreground program) are provided
4168 purely for convenience, and have exactly the same behavior as
4172 To resume execution at a different place, you can use @code{return}
4173 (@pxref{Returning, ,Returning from a Function}) to go back to the
4174 calling function; or @code{jump} (@pxref{Jumping, ,Continuing at a
4175 Different Address}) to go to an arbitrary location in your program.
4177 A typical technique for using stepping is to set a breakpoint
4178 (@pxref{Breakpoints, ,Breakpoints; Watchpoints; and Catchpoints}) at the
4179 beginning of the function or the section of your program where a problem
4180 is believed to lie, run your program until it stops at that breakpoint,
4181 and then step through the suspect area, examining the variables that are
4182 interesting, until you see the problem happen.
4186 @kindex s @r{(@code{step})}
4188 Continue running your program until control reaches a different source
4189 line, then stop it and return control to @value{GDBN}. This command is
4190 abbreviated @code{s}.
4193 @c "without debugging information" is imprecise; actually "without line
4194 @c numbers in the debugging information". (gcc -g1 has debugging info but
4195 @c not line numbers). But it seems complex to try to make that
4196 @c distinction here.
4197 @emph{Warning:} If you use the @code{step} command while control is
4198 within a function that was compiled without debugging information,
4199 execution proceeds until control reaches a function that does have
4200 debugging information. Likewise, it will not step into a function which
4201 is compiled without debugging information. To step through functions
4202 without debugging information, use the @code{stepi} command, described
4206 The @code{step} command only stops at the first instruction of a source
4207 line. This prevents the multiple stops that could otherwise occur in
4208 @code{switch} statements, @code{for} loops, etc. @code{step} continues
4209 to stop if a function that has debugging information is called within
4210 the line. In other words, @code{step} @emph{steps inside} any functions
4211 called within the line.
4213 Also, the @code{step} command only enters a function if there is line
4214 number information for the function. Otherwise it acts like the
4215 @code{next} command. This avoids problems when using @code{cc -gl}
4216 on MIPS machines. Previously, @code{step} entered subroutines if there
4217 was any debugging information about the routine.
4219 @item step @var{count}
4220 Continue running as in @code{step}, but do so @var{count} times. If a
4221 breakpoint is reached, or a signal not related to stepping occurs before
4222 @var{count} steps, stepping stops right away.
4225 @kindex n @r{(@code{next})}
4226 @item next @r{[}@var{count}@r{]}
4227 Continue to the next source line in the current (innermost) stack frame.
4228 This is similar to @code{step}, but function calls that appear within
4229 the line of code are executed without stopping. Execution stops when
4230 control reaches a different line of code at the original stack level
4231 that was executing when you gave the @code{next} command. This command
4232 is abbreviated @code{n}.
4234 An argument @var{count} is a repeat count, as for @code{step}.
4237 @c FIX ME!! Do we delete this, or is there a way it fits in with
4238 @c the following paragraph? --- Vctoria
4240 @c @code{next} within a function that lacks debugging information acts like
4241 @c @code{step}, but any function calls appearing within the code of the
4242 @c function are executed without stopping.
4244 The @code{next} command only stops at the first instruction of a
4245 source line. This prevents multiple stops that could otherwise occur in
4246 @code{switch} statements, @code{for} loops, etc.
4248 @kindex set step-mode
4250 @cindex functions without line info, and stepping
4251 @cindex stepping into functions with no line info
4252 @itemx set step-mode on
4253 The @code{set step-mode on} command causes the @code{step} command to
4254 stop at the first instruction of a function which contains no debug line
4255 information rather than stepping over it.
4257 This is useful in cases where you may be interested in inspecting the
4258 machine instructions of a function which has no symbolic info and do not
4259 want @value{GDBN} to automatically skip over this function.
4261 @item set step-mode off
4262 Causes the @code{step} command to step over any functions which contains no
4263 debug information. This is the default.
4265 @item show step-mode
4266 Show whether @value{GDBN} will stop in or step over functions without
4267 source line debug information.
4270 @kindex fin @r{(@code{finish})}
4272 Continue running until just after function in the selected stack frame
4273 returns. Print the returned value (if any). This command can be
4274 abbreviated as @code{fin}.
4276 Contrast this with the @code{return} command (@pxref{Returning,
4277 ,Returning from a Function}).
4280 @kindex u @r{(@code{until})}
4281 @cindex run until specified location
4284 Continue running until a source line past the current line, in the
4285 current stack frame, is reached. This command is used to avoid single
4286 stepping through a loop more than once. It is like the @code{next}
4287 command, except that when @code{until} encounters a jump, it
4288 automatically continues execution until the program counter is greater
4289 than the address of the jump.
4291 This means that when you reach the end of a loop after single stepping
4292 though it, @code{until} makes your program continue execution until it
4293 exits the loop. In contrast, a @code{next} command at the end of a loop
4294 simply steps back to the beginning of the loop, which forces you to step
4295 through the next iteration.
4297 @code{until} always stops your program if it attempts to exit the current
4300 @code{until} may produce somewhat counterintuitive results if the order
4301 of machine code does not match the order of the source lines. For
4302 example, in the following excerpt from a debugging session, the @code{f}
4303 (@code{frame}) command shows that execution is stopped at line
4304 @code{206}; yet when we use @code{until}, we get to line @code{195}:
4308 #0 main (argc=4, argv=0xf7fffae8) at m4.c:206
4310 (@value{GDBP}) until
4311 195 for ( ; argc > 0; NEXTARG) @{
4314 This happened because, for execution efficiency, the compiler had
4315 generated code for the loop closure test at the end, rather than the
4316 start, of the loop---even though the test in a C @code{for}-loop is
4317 written before the body of the loop. The @code{until} command appeared
4318 to step back to the beginning of the loop when it advanced to this
4319 expression; however, it has not really gone to an earlier
4320 statement---not in terms of the actual machine code.
4322 @code{until} with no argument works by means of single
4323 instruction stepping, and hence is slower than @code{until} with an
4326 @item until @var{location}
4327 @itemx u @var{location}
4328 Continue running your program until either the specified location is
4329 reached, or the current stack frame returns. @var{location} is any of
4330 the forms described in @ref{Specify Location}.
4331 This form of the command uses temporary breakpoints, and
4332 hence is quicker than @code{until} without an argument. The specified
4333 location is actually reached only if it is in the current frame. This
4334 implies that @code{until} can be used to skip over recursive function
4335 invocations. For instance in the code below, if the current location is
4336 line @code{96}, issuing @code{until 99} will execute the program up to
4337 line @code{99} in the same invocation of factorial, i.e., after the inner
4338 invocations have returned.
4341 94 int factorial (int value)
4343 96 if (value > 1) @{
4344 97 value *= factorial (value - 1);
4351 @kindex advance @var{location}
4352 @itemx advance @var{location}
4353 Continue running the program up to the given @var{location}. An argument is
4354 required, which should be of one of the forms described in
4355 @ref{Specify Location}.
4356 Execution will also stop upon exit from the current stack
4357 frame. This command is similar to @code{until}, but @code{advance} will
4358 not skip over recursive function calls, and the target location doesn't
4359 have to be in the same frame as the current one.
4363 @kindex si @r{(@code{stepi})}
4365 @itemx stepi @var{arg}
4367 Execute one machine instruction, then stop and return to the debugger.
4369 It is often useful to do @samp{display/i $pc} when stepping by machine
4370 instructions. This makes @value{GDBN} automatically display the next
4371 instruction to be executed, each time your program stops. @xref{Auto
4372 Display,, Automatic Display}.
4374 An argument is a repeat count, as in @code{step}.
4378 @kindex ni @r{(@code{nexti})}
4380 @itemx nexti @var{arg}
4382 Execute one machine instruction, but if it is a function call,
4383 proceed until the function returns.
4385 An argument is a repeat count, as in @code{next}.
4392 A signal is an asynchronous event that can happen in a program. The
4393 operating system defines the possible kinds of signals, and gives each
4394 kind a name and a number. For example, in Unix @code{SIGINT} is the
4395 signal a program gets when you type an interrupt character (often @kbd{Ctrl-c});
4396 @code{SIGSEGV} is the signal a program gets from referencing a place in
4397 memory far away from all the areas in use; @code{SIGALRM} occurs when
4398 the alarm clock timer goes off (which happens only if your program has
4399 requested an alarm).
4401 @cindex fatal signals
4402 Some signals, including @code{SIGALRM}, are a normal part of the
4403 functioning of your program. Others, such as @code{SIGSEGV}, indicate
4404 errors; these signals are @dfn{fatal} (they kill your program immediately) if the
4405 program has not specified in advance some other way to handle the signal.
4406 @code{SIGINT} does not indicate an error in your program, but it is normally
4407 fatal so it can carry out the purpose of the interrupt: to kill the program.
4409 @value{GDBN} has the ability to detect any occurrence of a signal in your
4410 program. You can tell @value{GDBN} in advance what to do for each kind of
4413 @cindex handling signals
4414 Normally, @value{GDBN} is set up to let the non-erroneous signals like
4415 @code{SIGALRM} be silently passed to your program
4416 (so as not to interfere with their role in the program's functioning)
4417 but to stop your program immediately whenever an error signal happens.
4418 You can change these settings with the @code{handle} command.
4421 @kindex info signals
4425 Print a table of all the kinds of signals and how @value{GDBN} has been told to
4426 handle each one. You can use this to see the signal numbers of all
4427 the defined types of signals.
4429 @item info signals @var{sig}
4430 Similar, but print information only about the specified signal number.
4432 @code{info handle} is an alias for @code{info signals}.
4435 @item handle @var{signal} @r{[}@var{keywords}@dots{}@r{]}
4436 Change the way @value{GDBN} handles signal @var{signal}. @var{signal}
4437 can be the number of a signal or its name (with or without the
4438 @samp{SIG} at the beginning); a list of signal numbers of the form
4439 @samp{@var{low}-@var{high}}; or the word @samp{all}, meaning all the
4440 known signals. Optional arguments @var{keywords}, described below,
4441 say what change to make.
4445 The keywords allowed by the @code{handle} command can be abbreviated.
4446 Their full names are:
4450 @value{GDBN} should not stop your program when this signal happens. It may
4451 still print a message telling you that the signal has come in.
4454 @value{GDBN} should stop your program when this signal happens. This implies
4455 the @code{print} keyword as well.
4458 @value{GDBN} should print a message when this signal happens.
4461 @value{GDBN} should not mention the occurrence of the signal at all. This
4462 implies the @code{nostop} keyword as well.
4466 @value{GDBN} should allow your program to see this signal; your program
4467 can handle the signal, or else it may terminate if the signal is fatal
4468 and not handled. @code{pass} and @code{noignore} are synonyms.
4472 @value{GDBN} should not allow your program to see this signal.
4473 @code{nopass} and @code{ignore} are synonyms.
4477 When a signal stops your program, the signal is not visible to the
4479 continue. Your program sees the signal then, if @code{pass} is in
4480 effect for the signal in question @emph{at that time}. In other words,
4481 after @value{GDBN} reports a signal, you can use the @code{handle}
4482 command with @code{pass} or @code{nopass} to control whether your
4483 program sees that signal when you continue.
4485 The default is set to @code{nostop}, @code{noprint}, @code{pass} for
4486 non-erroneous signals such as @code{SIGALRM}, @code{SIGWINCH} and
4487 @code{SIGCHLD}, and to @code{stop}, @code{print}, @code{pass} for the
4490 You can also use the @code{signal} command to prevent your program from
4491 seeing a signal, or cause it to see a signal it normally would not see,
4492 or to give it any signal at any time. For example, if your program stopped
4493 due to some sort of memory reference error, you might store correct
4494 values into the erroneous variables and continue, hoping to see more
4495 execution; but your program would probably terminate immediately as
4496 a result of the fatal signal once it saw the signal. To prevent this,
4497 you can continue with @samp{signal 0}. @xref{Signaling, ,Giving your
4500 @cindex extra signal information
4501 @anchor{extra signal information}
4503 On some targets, @value{GDBN} can inspect extra signal information
4504 associated with the intercepted signal, before it is actually
4505 delivered to the program being debugged. This information is exported
4506 by the convenience variable @code{$_siginfo}, and consists of data
4507 that is passed by the kernel to the signal handler at the time of the
4508 receipt of a signal. The data type of the information itself is
4509 target dependent. You can see the data type using the @code{ptype
4510 $_siginfo} command. On Unix systems, it typically corresponds to the
4511 standard @code{siginfo_t} type, as defined in the @file{signal.h}
4514 Here's an example, on a @sc{gnu}/Linux system, printing the stray
4515 referenced address that raised a segmentation fault.
4519 (@value{GDBP}) continue
4520 Program received signal SIGSEGV, Segmentation fault.
4521 0x0000000000400766 in main ()
4523 (@value{GDBP}) ptype $_siginfo
4530 struct @{...@} _kill;
4531 struct @{...@} _timer;
4533 struct @{...@} _sigchld;
4534 struct @{...@} _sigfault;
4535 struct @{...@} _sigpoll;
4538 (@value{GDBP}) ptype $_siginfo._sifields._sigfault
4542 (@value{GDBP}) p $_siginfo._sifields._sigfault.si_addr
4543 $1 = (void *) 0x7ffff7ff7000
4547 Depending on target support, @code{$_siginfo} may also be writable.
4550 @section Stopping and Starting Multi-thread Programs
4552 @cindex stopped threads
4553 @cindex threads, stopped
4555 @cindex continuing threads
4556 @cindex threads, continuing
4558 @value{GDBN} supports debugging programs with multiple threads
4559 (@pxref{Threads,, Debugging Programs with Multiple Threads}). There
4560 are two modes of controlling execution of your program within the
4561 debugger. In the default mode, referred to as @dfn{all-stop mode},
4562 when any thread in your program stops (for example, at a breakpoint
4563 or while being stepped), all other threads in the program are also stopped by
4564 @value{GDBN}. On some targets, @value{GDBN} also supports
4565 @dfn{non-stop mode}, in which other threads can continue to run freely while
4566 you examine the stopped thread in the debugger.
4569 * All-Stop Mode:: All threads stop when GDB takes control
4570 * Non-Stop Mode:: Other threads continue to execute
4571 * Background Execution:: Running your program asynchronously
4572 * Thread-Specific Breakpoints:: Controlling breakpoints
4573 * Interrupted System Calls:: GDB may interfere with system calls
4577 @subsection All-Stop Mode
4579 @cindex all-stop mode
4581 In all-stop mode, whenever your program stops under @value{GDBN} for any reason,
4582 @emph{all} threads of execution stop, not just the current thread. This
4583 allows you to examine the overall state of the program, including
4584 switching between threads, without worrying that things may change
4587 Conversely, whenever you restart the program, @emph{all} threads start
4588 executing. @emph{This is true even when single-stepping} with commands
4589 like @code{step} or @code{next}.
4591 In particular, @value{GDBN} cannot single-step all threads in lockstep.
4592 Since thread scheduling is up to your debugging target's operating
4593 system (not controlled by @value{GDBN}), other threads may
4594 execute more than one statement while the current thread completes a
4595 single step. Moreover, in general other threads stop in the middle of a
4596 statement, rather than at a clean statement boundary, when the program
4599 You might even find your program stopped in another thread after
4600 continuing or even single-stepping. This happens whenever some other
4601 thread runs into a breakpoint, a signal, or an exception before the
4602 first thread completes whatever you requested.
4604 @cindex automatic thread selection
4605 @cindex switching threads automatically
4606 @cindex threads, automatic switching
4607 Whenever @value{GDBN} stops your program, due to a breakpoint or a
4608 signal, it automatically selects the thread where that breakpoint or
4609 signal happened. @value{GDBN} alerts you to the context switch with a
4610 message such as @samp{[Switching to Thread @var{n}]} to identify the
4613 On some OSes, you can modify @value{GDBN}'s default behavior by
4614 locking the OS scheduler to allow only a single thread to run.
4617 @item set scheduler-locking @var{mode}
4618 @cindex scheduler locking mode
4619 @cindex lock scheduler
4620 Set the scheduler locking mode. If it is @code{off}, then there is no
4621 locking and any thread may run at any time. If @code{on}, then only the
4622 current thread may run when the inferior is resumed. The @code{step}
4623 mode optimizes for single-stepping; it prevents other threads
4624 from preempting the current thread while you are stepping, so that
4625 the focus of debugging does not change unexpectedly.
4626 Other threads only rarely (or never) get a chance to run
4627 when you step. They are more likely to run when you @samp{next} over a
4628 function call, and they are completely free to run when you use commands
4629 like @samp{continue}, @samp{until}, or @samp{finish}. However, unless another
4630 thread hits a breakpoint during its timeslice, @value{GDBN} does not change
4631 the current thread away from the thread that you are debugging.
4633 @item show scheduler-locking
4634 Display the current scheduler locking mode.
4637 @cindex resume threads of multiple processes simultaneously
4638 By default, when you issue one of the execution commands such as
4639 @code{continue}, @code{next} or @code{step}, @value{GDBN} allows only
4640 threads of the current inferior to run. For example, if @value{GDBN}
4641 is attached to two inferiors, each with two threads, the
4642 @code{continue} command resumes only the two threads of the current
4643 inferior. This is useful, for example, when you debug a program that
4644 forks and you want to hold the parent stopped (so that, for instance,
4645 it doesn't run to exit), while you debug the child. In other
4646 situations, you may not be interested in inspecting the current state
4647 of any of the processes @value{GDBN} is attached to, and you may want
4648 to resume them all until some breakpoint is hit. In the latter case,
4649 you can instruct @value{GDBN} to allow all threads of all the
4650 inferiors to run with the @w{@code{set schedule-multiple}} command.
4653 @kindex set schedule-multiple
4654 @item set schedule-multiple
4655 Set the mode for allowing threads of multiple processes to be resumed
4656 when an execution command is issued. When @code{on}, all threads of
4657 all processes are allowed to run. When @code{off}, only the threads
4658 of the current process are resumed. The default is @code{off}. The
4659 @code{scheduler-locking} mode takes precedence when set to @code{on},
4660 or while you are stepping and set to @code{step}.
4662 @item show schedule-multiple
4663 Display the current mode for resuming the execution of threads of
4668 @subsection Non-Stop Mode
4670 @cindex non-stop mode
4672 @c This section is really only a place-holder, and needs to be expanded
4673 @c with more details.
4675 For some multi-threaded targets, @value{GDBN} supports an optional
4676 mode of operation in which you can examine stopped program threads in
4677 the debugger while other threads continue to execute freely. This
4678 minimizes intrusion when debugging live systems, such as programs
4679 where some threads have real-time constraints or must continue to
4680 respond to external events. This is referred to as @dfn{non-stop} mode.
4682 In non-stop mode, when a thread stops to report a debugging event,
4683 @emph{only} that thread is stopped; @value{GDBN} does not stop other
4684 threads as well, in contrast to the all-stop mode behavior. Additionally,
4685 execution commands such as @code{continue} and @code{step} apply by default
4686 only to the current thread in non-stop mode, rather than all threads as
4687 in all-stop mode. This allows you to control threads explicitly in
4688 ways that are not possible in all-stop mode --- for example, stepping
4689 one thread while allowing others to run freely, stepping
4690 one thread while holding all others stopped, or stepping several threads
4691 independently and simultaneously.
4693 To enter non-stop mode, use this sequence of commands before you run
4694 or attach to your program:
4697 # Enable the async interface.
4700 # If using the CLI, pagination breaks non-stop.
4703 # Finally, turn it on!
4707 You can use these commands to manipulate the non-stop mode setting:
4710 @kindex set non-stop
4711 @item set non-stop on
4712 Enable selection of non-stop mode.
4713 @item set non-stop off
4714 Disable selection of non-stop mode.
4715 @kindex show non-stop
4717 Show the current non-stop enablement setting.
4720 Note these commands only reflect whether non-stop mode is enabled,
4721 not whether the currently-executing program is being run in non-stop mode.
4722 In particular, the @code{set non-stop} preference is only consulted when
4723 @value{GDBN} starts or connects to the target program, and it is generally
4724 not possible to switch modes once debugging has started. Furthermore,
4725 since not all targets support non-stop mode, even when you have enabled
4726 non-stop mode, @value{GDBN} may still fall back to all-stop operation by
4729 In non-stop mode, all execution commands apply only to the current thread
4730 by default. That is, @code{continue} only continues one thread.
4731 To continue all threads, issue @code{continue -a} or @code{c -a}.
4733 You can use @value{GDBN}'s background execution commands
4734 (@pxref{Background Execution}) to run some threads in the background
4735 while you continue to examine or step others from @value{GDBN}.
4736 The MI execution commands (@pxref{GDB/MI Program Execution}) are
4737 always executed asynchronously in non-stop mode.
4739 Suspending execution is done with the @code{interrupt} command when
4740 running in the background, or @kbd{Ctrl-c} during foreground execution.
4741 In all-stop mode, this stops the whole process;
4742 but in non-stop mode the interrupt applies only to the current thread.
4743 To stop the whole program, use @code{interrupt -a}.
4745 Other execution commands do not currently support the @code{-a} option.
4747 In non-stop mode, when a thread stops, @value{GDBN} doesn't automatically make
4748 that thread current, as it does in all-stop mode. This is because the
4749 thread stop notifications are asynchronous with respect to @value{GDBN}'s
4750 command interpreter, and it would be confusing if @value{GDBN} unexpectedly
4751 changed to a different thread just as you entered a command to operate on the
4752 previously current thread.
4754 @node Background Execution
4755 @subsection Background Execution
4757 @cindex foreground execution
4758 @cindex background execution
4759 @cindex asynchronous execution
4760 @cindex execution, foreground, background and asynchronous
4762 @value{GDBN}'s execution commands have two variants: the normal
4763 foreground (synchronous) behavior, and a background
4764 (asynchronous) behavior. In foreground execution, @value{GDBN} waits for
4765 the program to report that some thread has stopped before prompting for
4766 another command. In background execution, @value{GDBN} immediately gives
4767 a command prompt so that you can issue other commands while your program runs.
4769 You need to explicitly enable asynchronous mode before you can use
4770 background execution commands. You can use these commands to
4771 manipulate the asynchronous mode setting:
4774 @kindex set target-async
4775 @item set target-async on
4776 Enable asynchronous mode.
4777 @item set target-async off
4778 Disable asynchronous mode.
4779 @kindex show target-async
4780 @item show target-async
4781 Show the current target-async setting.
4784 If the target doesn't support async mode, @value{GDBN} issues an error
4785 message if you attempt to use the background execution commands.
4787 To specify background execution, add a @code{&} to the command. For example,
4788 the background form of the @code{continue} command is @code{continue&}, or
4789 just @code{c&}. The execution commands that accept background execution
4795 @xref{Starting, , Starting your Program}.
4799 @xref{Attach, , Debugging an Already-running Process}.
4803 @xref{Continuing and Stepping, step}.
4807 @xref{Continuing and Stepping, stepi}.
4811 @xref{Continuing and Stepping, next}.
4815 @xref{Continuing and Stepping, nexti}.
4819 @xref{Continuing and Stepping, continue}.
4823 @xref{Continuing and Stepping, finish}.
4827 @xref{Continuing and Stepping, until}.
4831 Background execution is especially useful in conjunction with non-stop
4832 mode for debugging programs with multiple threads; see @ref{Non-Stop Mode}.
4833 However, you can also use these commands in the normal all-stop mode with
4834 the restriction that you cannot issue another execution command until the
4835 previous one finishes. Examples of commands that are valid in all-stop
4836 mode while the program is running include @code{help} and @code{info break}.
4838 You can interrupt your program while it is running in the background by
4839 using the @code{interrupt} command.
4846 Suspend execution of the running program. In all-stop mode,
4847 @code{interrupt} stops the whole process, but in non-stop mode, it stops
4848 only the current thread. To stop the whole program in non-stop mode,
4849 use @code{interrupt -a}.
4852 @node Thread-Specific Breakpoints
4853 @subsection Thread-Specific Breakpoints
4855 When your program has multiple threads (@pxref{Threads,, Debugging
4856 Programs with Multiple Threads}), you can choose whether to set
4857 breakpoints on all threads, or on a particular thread.
4860 @cindex breakpoints and threads
4861 @cindex thread breakpoints
4862 @kindex break @dots{} thread @var{threadno}
4863 @item break @var{linespec} thread @var{threadno}
4864 @itemx break @var{linespec} thread @var{threadno} if @dots{}
4865 @var{linespec} specifies source lines; there are several ways of
4866 writing them (@pxref{Specify Location}), but the effect is always to
4867 specify some source line.
4869 Use the qualifier @samp{thread @var{threadno}} with a breakpoint command
4870 to specify that you only want @value{GDBN} to stop the program when a
4871 particular thread reaches this breakpoint. @var{threadno} is one of the
4872 numeric thread identifiers assigned by @value{GDBN}, shown in the first
4873 column of the @samp{info threads} display.
4875 If you do not specify @samp{thread @var{threadno}} when you set a
4876 breakpoint, the breakpoint applies to @emph{all} threads of your
4879 You can use the @code{thread} qualifier on conditional breakpoints as
4880 well; in this case, place @samp{thread @var{threadno}} before the
4881 breakpoint condition, like this:
4884 (@value{GDBP}) break frik.c:13 thread 28 if bartab > lim
4889 @node Interrupted System Calls
4890 @subsection Interrupted System Calls
4892 @cindex thread breakpoints and system calls
4893 @cindex system calls and thread breakpoints
4894 @cindex premature return from system calls
4895 There is an unfortunate side effect when using @value{GDBN} to debug
4896 multi-threaded programs. If one thread stops for a
4897 breakpoint, or for some other reason, and another thread is blocked in a
4898 system call, then the system call may return prematurely. This is a
4899 consequence of the interaction between multiple threads and the signals
4900 that @value{GDBN} uses to implement breakpoints and other events that
4903 To handle this problem, your program should check the return value of
4904 each system call and react appropriately. This is good programming
4907 For example, do not write code like this:
4913 The call to @code{sleep} will return early if a different thread stops
4914 at a breakpoint or for some other reason.
4916 Instead, write this:
4921 unslept = sleep (unslept);
4924 A system call is allowed to return early, so the system is still
4925 conforming to its specification. But @value{GDBN} does cause your
4926 multi-threaded program to behave differently than it would without
4929 Also, @value{GDBN} uses internal breakpoints in the thread library to
4930 monitor certain events such as thread creation and thread destruction.
4931 When such an event happens, a system call in another thread may return
4932 prematurely, even though your program does not appear to stop.
4935 @node Reverse Execution
4936 @chapter Running programs backward
4937 @cindex reverse execution
4938 @cindex running programs backward
4940 When you are debugging a program, it is not unusual to realize that
4941 you have gone too far, and some event of interest has already happened.
4942 If the target environment supports it, @value{GDBN} can allow you to
4943 ``rewind'' the program by running it backward.
4945 A target environment that supports reverse execution should be able
4946 to ``undo'' the changes in machine state that have taken place as the
4947 program was executing normally. Variables, registers etc.@: should
4948 revert to their previous values. Obviously this requires a great
4949 deal of sophistication on the part of the target environment; not
4950 all target environments can support reverse execution.
4952 When a program is executed in reverse, the instructions that
4953 have most recently been executed are ``un-executed'', in reverse
4954 order. The program counter runs backward, following the previous
4955 thread of execution in reverse. As each instruction is ``un-executed'',
4956 the values of memory and/or registers that were changed by that
4957 instruction are reverted to their previous states. After executing
4958 a piece of source code in reverse, all side effects of that code
4959 should be ``undone'', and all variables should be returned to their
4960 prior values@footnote{
4961 Note that some side effects are easier to undo than others. For instance,
4962 memory and registers are relatively easy, but device I/O is hard. Some
4963 targets may be able undo things like device I/O, and some may not.
4965 The contract between @value{GDBN} and the reverse executing target
4966 requires only that the target do something reasonable when
4967 @value{GDBN} tells it to execute backwards, and then report the
4968 results back to @value{GDBN}. Whatever the target reports back to
4969 @value{GDBN}, @value{GDBN} will report back to the user. @value{GDBN}
4970 assumes that the memory and registers that the target reports are in a
4971 consistant state, but @value{GDBN} accepts whatever it is given.
4974 If you are debugging in a target environment that supports
4975 reverse execution, @value{GDBN} provides the following commands.
4978 @kindex reverse-continue
4979 @kindex rc @r{(@code{reverse-continue})}
4980 @item reverse-continue @r{[}@var{ignore-count}@r{]}
4981 @itemx rc @r{[}@var{ignore-count}@r{]}
4982 Beginning at the point where your program last stopped, start executing
4983 in reverse. Reverse execution will stop for breakpoints and synchronous
4984 exceptions (signals), just like normal execution. Behavior of
4985 asynchronous signals depends on the target environment.
4987 @kindex reverse-step
4988 @kindex rs @r{(@code{step})}
4989 @item reverse-step @r{[}@var{count}@r{]}
4990 Run the program backward until control reaches the start of a
4991 different source line; then stop it, and return control to @value{GDBN}.
4993 Like the @code{step} command, @code{reverse-step} will only stop
4994 at the beginning of a source line. It ``un-executes'' the previously
4995 executed source line. If the previous source line included calls to
4996 debuggable functions, @code{reverse-step} will step (backward) into
4997 the called function, stopping at the beginning of the @emph{last}
4998 statement in the called function (typically a return statement).
5000 Also, as with the @code{step} command, if non-debuggable functions are
5001 called, @code{reverse-step} will run thru them backward without stopping.
5003 @kindex reverse-stepi
5004 @kindex rsi @r{(@code{reverse-stepi})}
5005 @item reverse-stepi @r{[}@var{count}@r{]}
5006 Reverse-execute one machine instruction. Note that the instruction
5007 to be reverse-executed is @emph{not} the one pointed to by the program
5008 counter, but the instruction executed prior to that one. For instance,
5009 if the last instruction was a jump, @code{reverse-stepi} will take you
5010 back from the destination of the jump to the jump instruction itself.
5012 @kindex reverse-next
5013 @kindex rn @r{(@code{reverse-next})}
5014 @item reverse-next @r{[}@var{count}@r{]}
5015 Run backward to the beginning of the previous line executed in
5016 the current (innermost) stack frame. If the line contains function
5017 calls, they will be ``un-executed'' without stopping. Starting from
5018 the first line of a function, @code{reverse-next} will take you back
5019 to the caller of that function, @emph{before} the function was called,
5020 just as the normal @code{next} command would take you from the last
5021 line of a function back to its return to its caller
5022 @footnote{Unles the code is too heavily optimized.}.
5024 @kindex reverse-nexti
5025 @kindex rni @r{(@code{reverse-nexti})}
5026 @item reverse-nexti @r{[}@var{count}@r{]}
5027 Like @code{nexti}, @code{reverse-nexti} executes a single instruction
5028 in reverse, except that called functions are ``un-executed'' atomically.
5029 That is, if the previously executed instruction was a return from
5030 another instruction, @code{reverse-nexti} will continue to execute
5031 in reverse until the call to that function (from the current stack
5034 @kindex reverse-finish
5035 @item reverse-finish
5036 Just as the @code{finish} command takes you to the point where the
5037 current function returns, @code{reverse-finish} takes you to the point
5038 where it was called. Instead of ending up at the end of the current
5039 function invocation, you end up at the beginning.
5041 @kindex set exec-direction
5042 @item set exec-direction
5043 Set the direction of target execution.
5044 @itemx set exec-direction reverse
5045 @cindex execute forward or backward in time
5046 @value{GDBN} will perform all execution commands in reverse, until the
5047 exec-direction mode is changed to ``forward''. Affected commands include
5048 @code{step, stepi, next, nexti, continue, and finish}. The @code{return}
5049 command cannot be used in reverse mode.
5050 @item set exec-direction forward
5051 @value{GDBN} will perform all execution commands in the normal fashion.
5052 This is the default.
5056 @node Process Record and Replay
5057 @chapter Recording Inferior's Execution and Replaying It
5058 @cindex process record and replay
5059 @cindex recording inferior's execution and replaying it
5061 On some platforms, @value{GDBN} provides a special @dfn{process record
5062 and replay} target that can record a log of the process execution, and
5063 replay it later with both forward and reverse execution commands.
5066 When this target is in use, if the execution log includes the record
5067 for the next instruction, @value{GDBN} will debug in @dfn{replay
5068 mode}. In the replay mode, the inferior does not really execute code
5069 instructions. Instead, all the events that normally happen during
5070 code execution are taken from the execution log. While code is not
5071 really executed in replay mode, the values of registers (including the
5072 program counter register) and the memory of the inferior are still
5073 changed as they normally would. Their contents are taken from the
5077 If the record for the next instruction is not in the execution log,
5078 @value{GDBN} will debug in @dfn{record mode}. In this mode, the
5079 inferior executes normally, and @value{GDBN} records the execution log
5082 The process record and replay target supports reverse execution
5083 (@pxref{Reverse Execution}), even if the platform on which the
5084 inferior runs does not. However, the reverse execution is limited in
5085 this case by the range of the instructions recorded in the execution
5086 log. In other words, reverse execution on platforms that don't
5087 support it directly can only be done in the replay mode.
5089 When debugging in the reverse direction, @value{GDBN} will work in
5090 replay mode as long as the execution log includes the record for the
5091 previous instruction; otherwise, it will work in record mode, if the
5092 platform supports reverse execution, or stop if not.
5094 For architecture environments that support process record and replay,
5095 @value{GDBN} provides the following commands:
5098 @kindex target record
5102 This command starts the process record and replay target. The process
5103 record and replay target can only debug a process that is already
5104 running. Therefore, you need first to start the process with the
5105 @kbd{run} or @kbd{start} commands, and then start the recording with
5106 the @kbd{target record} command.
5108 Both @code{record} and @code{rec} are aliases of @code{target record}.
5110 @cindex displaced stepping, and process record and replay
5111 Displaced stepping (@pxref{Maintenance Commands,, displaced stepping})
5112 will be automatically disabled when process record and replay target
5113 is started. That's because the process record and replay target
5114 doesn't support displaced stepping.
5116 @cindex non-stop mode, and process record and replay
5117 @cindex asynchronous execution, and process record and replay
5118 If the inferior is in the non-stop mode (@pxref{Non-Stop Mode}) or in
5119 the asynchronous execution mode (@pxref{Background Execution}), the
5120 process record and replay target cannot be started because it doesn't
5121 support these two modes.
5126 Stop the process record and replay target. When process record and
5127 replay target stops, the entire execution log will be deleted and the
5128 inferior will either be terminated, or will remain in its final state.
5130 When you stop the process record and replay target in record mode (at
5131 the end of the execution log), the inferior will be stopped at the
5132 next instruction that would have been recorded. In other words, if
5133 you record for a while and then stop recording, the inferior process
5134 will be left in the same state as if the recording never happened.
5136 On the other hand, if the process record and replay target is stopped
5137 while in replay mode (that is, not at the end of the execution log,
5138 but at some earlier point), the inferior process will become ``live''
5139 at that earlier state, and it will then be possible to continue the
5140 usual ``live'' debugging of the process from that state.
5142 When the inferior process exits, or @value{GDBN} detaches from it,
5143 process record and replay target will automatically stop itself.
5145 @kindex set record insn-number-max
5146 @item set record insn-number-max @var{limit}
5147 Set the limit of instructions to be recorded. Default value is 200000.
5149 If @var{limit} is a positive number, then @value{GDBN} will start
5150 deleting instructions from the log once the number of the record
5151 instructions becomes greater than @var{limit}. For every new recorded
5152 instruction, @value{GDBN} will delete the earliest recorded
5153 instruction to keep the number of recorded instructions at the limit.
5154 (Since deleting recorded instructions loses information, @value{GDBN}
5155 lets you control what happens when the limit is reached, by means of
5156 the @code{stop-at-limit} option, described below.)
5158 If @var{limit} is zero, @value{GDBN} will never delete recorded
5159 instructions from the execution log. The number of recorded
5160 instructions is unlimited in this case.
5162 @kindex show record insn-number-max
5163 @item show record insn-number-max
5164 Show the limit of instructions to be recorded.
5166 @kindex set record stop-at-limit
5167 @item set record stop-at-limit
5168 Control the behavior when the number of recorded instructions reaches
5169 the limit. If ON (the default), @value{GDBN} will stop when the limit
5170 is reached for the first time and ask you whether you want to stop the
5171 inferior or continue running it and recording the execution log. If
5172 you decide to continue recording, each new recorded instruction will
5173 cause the oldest one to be deleted.
5175 If this option is OFF, @value{GDBN} will automatically delete the
5176 oldest record to make room for each new one, without asking.
5178 @kindex show record stop-at-limit
5179 @item show record stop-at-limit
5180 Show the current setting of @code{stop-at-limit}.
5182 @kindex info record insn-number
5183 @item info record insn-number
5184 Show the current number of recorded instructions.
5186 @kindex record delete
5189 When record target runs in replay mode (``in the past''), delete the
5190 subsequent execution log and begin to record a new execution log starting
5191 from the current address. This means you will abandon the previously
5192 recorded ``future'' and begin recording a new ``future''.
5197 @chapter Examining the Stack
5199 When your program has stopped, the first thing you need to know is where it
5200 stopped and how it got there.
5203 Each time your program performs a function call, information about the call
5205 That information includes the location of the call in your program,
5206 the arguments of the call,
5207 and the local variables of the function being called.
5208 The information is saved in a block of data called a @dfn{stack frame}.
5209 The stack frames are allocated in a region of memory called the @dfn{call
5212 When your program stops, the @value{GDBN} commands for examining the
5213 stack allow you to see all of this information.
5215 @cindex selected frame
5216 One of the stack frames is @dfn{selected} by @value{GDBN} and many
5217 @value{GDBN} commands refer implicitly to the selected frame. In
5218 particular, whenever you ask @value{GDBN} for the value of a variable in
5219 your program, the value is found in the selected frame. There are
5220 special @value{GDBN} commands to select whichever frame you are
5221 interested in. @xref{Selection, ,Selecting a Frame}.
5223 When your program stops, @value{GDBN} automatically selects the
5224 currently executing frame and describes it briefly, similar to the
5225 @code{frame} command (@pxref{Frame Info, ,Information about a Frame}).
5228 * Frames:: Stack frames
5229 * Backtrace:: Backtraces
5230 * Selection:: Selecting a frame
5231 * Frame Info:: Information on a frame
5236 @section Stack Frames
5238 @cindex frame, definition
5240 The call stack is divided up into contiguous pieces called @dfn{stack
5241 frames}, or @dfn{frames} for short; each frame is the data associated
5242 with one call to one function. The frame contains the arguments given
5243 to the function, the function's local variables, and the address at
5244 which the function is executing.
5246 @cindex initial frame
5247 @cindex outermost frame
5248 @cindex innermost frame
5249 When your program is started, the stack has only one frame, that of the
5250 function @code{main}. This is called the @dfn{initial} frame or the
5251 @dfn{outermost} frame. Each time a function is called, a new frame is
5252 made. Each time a function returns, the frame for that function invocation
5253 is eliminated. If a function is recursive, there can be many frames for
5254 the same function. The frame for the function in which execution is
5255 actually occurring is called the @dfn{innermost} frame. This is the most
5256 recently created of all the stack frames that still exist.
5258 @cindex frame pointer
5259 Inside your program, stack frames are identified by their addresses. A
5260 stack frame consists of many bytes, each of which has its own address; each
5261 kind of computer has a convention for choosing one byte whose
5262 address serves as the address of the frame. Usually this address is kept
5263 in a register called the @dfn{frame pointer register}
5264 (@pxref{Registers, $fp}) while execution is going on in that frame.
5266 @cindex frame number
5267 @value{GDBN} assigns numbers to all existing stack frames, starting with
5268 zero for the innermost frame, one for the frame that called it,
5269 and so on upward. These numbers do not really exist in your program;
5270 they are assigned by @value{GDBN} to give you a way of designating stack
5271 frames in @value{GDBN} commands.
5273 @c The -fomit-frame-pointer below perennially causes hbox overflow
5274 @c underflow problems.
5275 @cindex frameless execution
5276 Some compilers provide a way to compile functions so that they operate
5277 without stack frames. (For example, the @value{NGCC} option
5279 @samp{-fomit-frame-pointer}
5281 generates functions without a frame.)
5282 This is occasionally done with heavily used library functions to save
5283 the frame setup time. @value{GDBN} has limited facilities for dealing
5284 with these function invocations. If the innermost function invocation
5285 has no stack frame, @value{GDBN} nevertheless regards it as though
5286 it had a separate frame, which is numbered zero as usual, allowing
5287 correct tracing of the function call chain. However, @value{GDBN} has
5288 no provision for frameless functions elsewhere in the stack.
5291 @kindex frame@r{, command}
5292 @cindex current stack frame
5293 @item frame @var{args}
5294 The @code{frame} command allows you to move from one stack frame to another,
5295 and to print the stack frame you select. @var{args} may be either the
5296 address of the frame or the stack frame number. Without an argument,
5297 @code{frame} prints the current stack frame.
5299 @kindex select-frame
5300 @cindex selecting frame silently
5302 The @code{select-frame} command allows you to move from one stack frame
5303 to another without printing the frame. This is the silent version of
5311 @cindex call stack traces
5312 A backtrace is a summary of how your program got where it is. It shows one
5313 line per frame, for many frames, starting with the currently executing
5314 frame (frame zero), followed by its caller (frame one), and on up the
5319 @kindex bt @r{(@code{backtrace})}
5322 Print a backtrace of the entire stack: one line per frame for all
5323 frames in the stack.
5325 You can stop the backtrace at any time by typing the system interrupt
5326 character, normally @kbd{Ctrl-c}.
5328 @item backtrace @var{n}
5330 Similar, but print only the innermost @var{n} frames.
5332 @item backtrace -@var{n}
5334 Similar, but print only the outermost @var{n} frames.
5336 @item backtrace full
5338 @itemx bt full @var{n}
5339 @itemx bt full -@var{n}
5340 Print the values of the local variables also. @var{n} specifies the
5341 number of frames to print, as described above.
5346 The names @code{where} and @code{info stack} (abbreviated @code{info s})
5347 are additional aliases for @code{backtrace}.
5349 @cindex multiple threads, backtrace
5350 In a multi-threaded program, @value{GDBN} by default shows the
5351 backtrace only for the current thread. To display the backtrace for
5352 several or all of the threads, use the command @code{thread apply}
5353 (@pxref{Threads, thread apply}). For example, if you type @kbd{thread
5354 apply all backtrace}, @value{GDBN} will display the backtrace for all
5355 the threads; this is handy when you debug a core dump of a
5356 multi-threaded program.
5358 Each line in the backtrace shows the frame number and the function name.
5359 The program counter value is also shown---unless you use @code{set
5360 print address off}. The backtrace also shows the source file name and
5361 line number, as well as the arguments to the function. The program
5362 counter value is omitted if it is at the beginning of the code for that
5365 Here is an example of a backtrace. It was made with the command
5366 @samp{bt 3}, so it shows the innermost three frames.
5370 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
5372 #1 0x6e38 in expand_macro (sym=0x2b600, data=...) at macro.c:242
5373 #2 0x6840 in expand_token (obs=0x0, t=177664, td=0xf7fffb08)
5375 (More stack frames follow...)
5380 The display for frame zero does not begin with a program counter
5381 value, indicating that your program has stopped at the beginning of the
5382 code for line @code{993} of @code{builtin.c}.
5385 The value of parameter @code{data} in frame 1 has been replaced by
5386 @code{@dots{}}. By default, @value{GDBN} prints the value of a parameter
5387 only if it is a scalar (integer, pointer, enumeration, etc). See command
5388 @kbd{set print frame-arguments} in @ref{Print Settings} for more details
5389 on how to configure the way function parameter values are printed.
5391 @cindex value optimized out, in backtrace
5392 @cindex function call arguments, optimized out
5393 If your program was compiled with optimizations, some compilers will
5394 optimize away arguments passed to functions if those arguments are
5395 never used after the call. Such optimizations generate code that
5396 passes arguments through registers, but doesn't store those arguments
5397 in the stack frame. @value{GDBN} has no way of displaying such
5398 arguments in stack frames other than the innermost one. Here's what
5399 such a backtrace might look like:
5403 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
5405 #1 0x6e38 in expand_macro (sym=<value optimized out>) at macro.c:242
5406 #2 0x6840 in expand_token (obs=0x0, t=<value optimized out>, td=0xf7fffb08)
5408 (More stack frames follow...)
5413 The values of arguments that were not saved in their stack frames are
5414 shown as @samp{<value optimized out>}.
5416 If you need to display the values of such optimized-out arguments,
5417 either deduce that from other variables whose values depend on the one
5418 you are interested in, or recompile without optimizations.
5420 @cindex backtrace beyond @code{main} function
5421 @cindex program entry point
5422 @cindex startup code, and backtrace
5423 Most programs have a standard user entry point---a place where system
5424 libraries and startup code transition into user code. For C this is
5425 @code{main}@footnote{
5426 Note that embedded programs (the so-called ``free-standing''
5427 environment) are not required to have a @code{main} function as the
5428 entry point. They could even have multiple entry points.}.
5429 When @value{GDBN} finds the entry function in a backtrace
5430 it will terminate the backtrace, to avoid tracing into highly
5431 system-specific (and generally uninteresting) code.
5433 If you need to examine the startup code, or limit the number of levels
5434 in a backtrace, you can change this behavior:
5437 @item set backtrace past-main
5438 @itemx set backtrace past-main on
5439 @kindex set backtrace
5440 Backtraces will continue past the user entry point.
5442 @item set backtrace past-main off
5443 Backtraces will stop when they encounter the user entry point. This is the
5446 @item show backtrace past-main
5447 @kindex show backtrace
5448 Display the current user entry point backtrace policy.
5450 @item set backtrace past-entry
5451 @itemx set backtrace past-entry on
5452 Backtraces will continue past the internal entry point of an application.
5453 This entry point is encoded by the linker when the application is built,
5454 and is likely before the user entry point @code{main} (or equivalent) is called.
5456 @item set backtrace past-entry off
5457 Backtraces will stop when they encounter the internal entry point of an
5458 application. This is the default.
5460 @item show backtrace past-entry
5461 Display the current internal entry point backtrace policy.
5463 @item set backtrace limit @var{n}
5464 @itemx set backtrace limit 0
5465 @cindex backtrace limit
5466 Limit the backtrace to @var{n} levels. A value of zero means
5469 @item show backtrace limit
5470 Display the current limit on backtrace levels.
5474 @section Selecting a Frame
5476 Most commands for examining the stack and other data in your program work on
5477 whichever stack frame is selected at the moment. Here are the commands for
5478 selecting a stack frame; all of them finish by printing a brief description
5479 of the stack frame just selected.
5482 @kindex frame@r{, selecting}
5483 @kindex f @r{(@code{frame})}
5486 Select frame number @var{n}. Recall that frame zero is the innermost
5487 (currently executing) frame, frame one is the frame that called the
5488 innermost one, and so on. The highest-numbered frame is the one for
5491 @item frame @var{addr}
5493 Select the frame at address @var{addr}. This is useful mainly if the
5494 chaining of stack frames has been damaged by a bug, making it
5495 impossible for @value{GDBN} to assign numbers properly to all frames. In
5496 addition, this can be useful when your program has multiple stacks and
5497 switches between them.
5499 On the SPARC architecture, @code{frame} needs two addresses to
5500 select an arbitrary frame: a frame pointer and a stack pointer.
5502 On the MIPS and Alpha architecture, it needs two addresses: a stack
5503 pointer and a program counter.
5505 On the 29k architecture, it needs three addresses: a register stack
5506 pointer, a program counter, and a memory stack pointer.
5510 Move @var{n} frames up the stack. For positive numbers @var{n}, this
5511 advances toward the outermost frame, to higher frame numbers, to frames
5512 that have existed longer. @var{n} defaults to one.
5515 @kindex do @r{(@code{down})}
5517 Move @var{n} frames down the stack. For positive numbers @var{n}, this
5518 advances toward the innermost frame, to lower frame numbers, to frames
5519 that were created more recently. @var{n} defaults to one. You may
5520 abbreviate @code{down} as @code{do}.
5523 All of these commands end by printing two lines of output describing the
5524 frame. The first line shows the frame number, the function name, the
5525 arguments, and the source file and line number of execution in that
5526 frame. The second line shows the text of that source line.
5534 #1 0x22f0 in main (argc=1, argv=0xf7fffbf4, env=0xf7fffbfc)
5536 10 read_input_file (argv[i]);
5540 After such a printout, the @code{list} command with no arguments
5541 prints ten lines centered on the point of execution in the frame.
5542 You can also edit the program at the point of execution with your favorite
5543 editing program by typing @code{edit}.
5544 @xref{List, ,Printing Source Lines},
5548 @kindex down-silently
5550 @item up-silently @var{n}
5551 @itemx down-silently @var{n}
5552 These two commands are variants of @code{up} and @code{down},
5553 respectively; they differ in that they do their work silently, without
5554 causing display of the new frame. They are intended primarily for use
5555 in @value{GDBN} command scripts, where the output might be unnecessary and
5560 @section Information About a Frame
5562 There are several other commands to print information about the selected
5568 When used without any argument, this command does not change which
5569 frame is selected, but prints a brief description of the currently
5570 selected stack frame. It can be abbreviated @code{f}. With an
5571 argument, this command is used to select a stack frame.
5572 @xref{Selection, ,Selecting a Frame}.
5575 @kindex info f @r{(@code{info frame})}
5578 This command prints a verbose description of the selected stack frame,
5583 the address of the frame
5585 the address of the next frame down (called by this frame)
5587 the address of the next frame up (caller of this frame)
5589 the language in which the source code corresponding to this frame is written
5591 the address of the frame's arguments
5593 the address of the frame's local variables
5595 the program counter saved in it (the address of execution in the caller frame)
5597 which registers were saved in the frame
5600 @noindent The verbose description is useful when
5601 something has gone wrong that has made the stack format fail to fit
5602 the usual conventions.
5604 @item info frame @var{addr}
5605 @itemx info f @var{addr}
5606 Print a verbose description of the frame at address @var{addr}, without
5607 selecting that frame. The selected frame remains unchanged by this
5608 command. This requires the same kind of address (more than one for some
5609 architectures) that you specify in the @code{frame} command.
5610 @xref{Selection, ,Selecting a Frame}.
5614 Print the arguments of the selected frame, each on a separate line.
5618 Print the local variables of the selected frame, each on a separate
5619 line. These are all variables (declared either static or automatic)
5620 accessible at the point of execution of the selected frame.
5623 @cindex catch exceptions, list active handlers
5624 @cindex exception handlers, how to list
5626 Print a list of all the exception handlers that are active in the
5627 current stack frame at the current point of execution. To see other
5628 exception handlers, visit the associated frame (using the @code{up},
5629 @code{down}, or @code{frame} commands); then type @code{info catch}.
5630 @xref{Set Catchpoints, , Setting Catchpoints}.
5636 @chapter Examining Source Files
5638 @value{GDBN} can print parts of your program's source, since the debugging
5639 information recorded in the program tells @value{GDBN} what source files were
5640 used to build it. When your program stops, @value{GDBN} spontaneously prints
5641 the line where it stopped. Likewise, when you select a stack frame
5642 (@pxref{Selection, ,Selecting a Frame}), @value{GDBN} prints the line where
5643 execution in that frame has stopped. You can print other portions of
5644 source files by explicit command.
5646 If you use @value{GDBN} through its @sc{gnu} Emacs interface, you may
5647 prefer to use Emacs facilities to view source; see @ref{Emacs, ,Using
5648 @value{GDBN} under @sc{gnu} Emacs}.
5651 * List:: Printing source lines
5652 * Specify Location:: How to specify code locations
5653 * Edit:: Editing source files
5654 * Search:: Searching source files
5655 * Source Path:: Specifying source directories
5656 * Machine Code:: Source and machine code
5660 @section Printing Source Lines
5663 @kindex l @r{(@code{list})}
5664 To print lines from a source file, use the @code{list} command
5665 (abbreviated @code{l}). By default, ten lines are printed.
5666 There are several ways to specify what part of the file you want to
5667 print; see @ref{Specify Location}, for the full list.
5669 Here are the forms of the @code{list} command most commonly used:
5672 @item list @var{linenum}
5673 Print lines centered around line number @var{linenum} in the
5674 current source file.
5676 @item list @var{function}
5677 Print lines centered around the beginning of function
5681 Print more lines. If the last lines printed were printed with a
5682 @code{list} command, this prints lines following the last lines
5683 printed; however, if the last line printed was a solitary line printed
5684 as part of displaying a stack frame (@pxref{Stack, ,Examining the
5685 Stack}), this prints lines centered around that line.
5688 Print lines just before the lines last printed.
5691 @cindex @code{list}, how many lines to display
5692 By default, @value{GDBN} prints ten source lines with any of these forms of
5693 the @code{list} command. You can change this using @code{set listsize}:
5696 @kindex set listsize
5697 @item set listsize @var{count}
5698 Make the @code{list} command display @var{count} source lines (unless
5699 the @code{list} argument explicitly specifies some other number).
5701 @kindex show listsize
5703 Display the number of lines that @code{list} prints.
5706 Repeating a @code{list} command with @key{RET} discards the argument,
5707 so it is equivalent to typing just @code{list}. This is more useful
5708 than listing the same lines again. An exception is made for an
5709 argument of @samp{-}; that argument is preserved in repetition so that
5710 each repetition moves up in the source file.
5712 In general, the @code{list} command expects you to supply zero, one or two
5713 @dfn{linespecs}. Linespecs specify source lines; there are several ways
5714 of writing them (@pxref{Specify Location}), but the effect is always
5715 to specify some source line.
5717 Here is a complete description of the possible arguments for @code{list}:
5720 @item list @var{linespec}
5721 Print lines centered around the line specified by @var{linespec}.
5723 @item list @var{first},@var{last}
5724 Print lines from @var{first} to @var{last}. Both arguments are
5725 linespecs. When a @code{list} command has two linespecs, and the
5726 source file of the second linespec is omitted, this refers to
5727 the same source file as the first linespec.
5729 @item list ,@var{last}
5730 Print lines ending with @var{last}.
5732 @item list @var{first},
5733 Print lines starting with @var{first}.
5736 Print lines just after the lines last printed.
5739 Print lines just before the lines last printed.
5742 As described in the preceding table.
5745 @node Specify Location
5746 @section Specifying a Location
5747 @cindex specifying location
5750 Several @value{GDBN} commands accept arguments that specify a location
5751 of your program's code. Since @value{GDBN} is a source-level
5752 debugger, a location usually specifies some line in the source code;
5753 for that reason, locations are also known as @dfn{linespecs}.
5755 Here are all the different ways of specifying a code location that
5756 @value{GDBN} understands:
5760 Specifies the line number @var{linenum} of the current source file.
5763 @itemx +@var{offset}
5764 Specifies the line @var{offset} lines before or after the @dfn{current
5765 line}. For the @code{list} command, the current line is the last one
5766 printed; for the breakpoint commands, this is the line at which
5767 execution stopped in the currently selected @dfn{stack frame}
5768 (@pxref{Frames, ,Frames}, for a description of stack frames.) When
5769 used as the second of the two linespecs in a @code{list} command,
5770 this specifies the line @var{offset} lines up or down from the first
5773 @item @var{filename}:@var{linenum}
5774 Specifies the line @var{linenum} in the source file @var{filename}.
5776 @item @var{function}
5777 Specifies the line that begins the body of the function @var{function}.
5778 For example, in C, this is the line with the open brace.
5780 @item @var{filename}:@var{function}
5781 Specifies the line that begins the body of the function @var{function}
5782 in the file @var{filename}. You only need the file name with a
5783 function name to avoid ambiguity when there are identically named
5784 functions in different source files.
5786 @item *@var{address}
5787 Specifies the program address @var{address}. For line-oriented
5788 commands, such as @code{list} and @code{edit}, this specifies a source
5789 line that contains @var{address}. For @code{break} and other
5790 breakpoint oriented commands, this can be used to set breakpoints in
5791 parts of your program which do not have debugging information or
5794 Here @var{address} may be any expression valid in the current working
5795 language (@pxref{Languages, working language}) that specifies a code
5796 address. In addition, as a convenience, @value{GDBN} extends the
5797 semantics of expressions used in locations to cover the situations
5798 that frequently happen during debugging. Here are the various forms
5802 @item @var{expression}
5803 Any expression valid in the current working language.
5805 @item @var{funcaddr}
5806 An address of a function or procedure derived from its name. In C,
5807 C@t{++}, Java, Objective-C, Fortran, minimal, and assembly, this is
5808 simply the function's name @var{function} (and actually a special case
5809 of a valid expression). In Pascal and Modula-2, this is
5810 @code{&@var{function}}. In Ada, this is @code{@var{function}'Address}
5811 (although the Pascal form also works).
5813 This form specifies the address of the function's first instruction,
5814 before the stack frame and arguments have been set up.
5816 @item '@var{filename}'::@var{funcaddr}
5817 Like @var{funcaddr} above, but also specifies the name of the source
5818 file explicitly. This is useful if the name of the function does not
5819 specify the function unambiguously, e.g., if there are several
5820 functions with identical names in different source files.
5827 @section Editing Source Files
5828 @cindex editing source files
5831 @kindex e @r{(@code{edit})}
5832 To edit the lines in a source file, use the @code{edit} command.
5833 The editing program of your choice
5834 is invoked with the current line set to
5835 the active line in the program.
5836 Alternatively, there are several ways to specify what part of the file you
5837 want to print if you want to see other parts of the program:
5840 @item edit @var{location}
5841 Edit the source file specified by @code{location}. Editing starts at
5842 that @var{location}, e.g., at the specified source line of the
5843 specified file. @xref{Specify Location}, for all the possible forms
5844 of the @var{location} argument; here are the forms of the @code{edit}
5845 command most commonly used:
5848 @item edit @var{number}
5849 Edit the current source file with @var{number} as the active line number.
5851 @item edit @var{function}
5852 Edit the file containing @var{function} at the beginning of its definition.
5857 @subsection Choosing your Editor
5858 You can customize @value{GDBN} to use any editor you want
5860 The only restriction is that your editor (say @code{ex}), recognizes the
5861 following command-line syntax:
5863 ex +@var{number} file
5865 The optional numeric value +@var{number} specifies the number of the line in
5866 the file where to start editing.}.
5867 By default, it is @file{@value{EDITOR}}, but you can change this
5868 by setting the environment variable @code{EDITOR} before using
5869 @value{GDBN}. For example, to configure @value{GDBN} to use the
5870 @code{vi} editor, you could use these commands with the @code{sh} shell:
5876 or in the @code{csh} shell,
5878 setenv EDITOR /usr/bin/vi
5883 @section Searching Source Files
5884 @cindex searching source files
5886 There are two commands for searching through the current source file for a
5891 @kindex forward-search
5892 @item forward-search @var{regexp}
5893 @itemx search @var{regexp}
5894 The command @samp{forward-search @var{regexp}} checks each line,
5895 starting with the one following the last line listed, for a match for
5896 @var{regexp}. It lists the line that is found. You can use the
5897 synonym @samp{search @var{regexp}} or abbreviate the command name as
5900 @kindex reverse-search
5901 @item reverse-search @var{regexp}
5902 The command @samp{reverse-search @var{regexp}} checks each line, starting
5903 with the one before the last line listed and going backward, for a match
5904 for @var{regexp}. It lists the line that is found. You can abbreviate
5905 this command as @code{rev}.
5909 @section Specifying Source Directories
5912 @cindex directories for source files
5913 Executable programs sometimes do not record the directories of the source
5914 files from which they were compiled, just the names. Even when they do,
5915 the directories could be moved between the compilation and your debugging
5916 session. @value{GDBN} has a list of directories to search for source files;
5917 this is called the @dfn{source path}. Each time @value{GDBN} wants a source file,
5918 it tries all the directories in the list, in the order they are present
5919 in the list, until it finds a file with the desired name.
5921 For example, suppose an executable references the file
5922 @file{/usr/src/foo-1.0/lib/foo.c}, and our source path is
5923 @file{/mnt/cross}. The file is first looked up literally; if this
5924 fails, @file{/mnt/cross/usr/src/foo-1.0/lib/foo.c} is tried; if this
5925 fails, @file{/mnt/cross/foo.c} is opened; if this fails, an error
5926 message is printed. @value{GDBN} does not look up the parts of the
5927 source file name, such as @file{/mnt/cross/src/foo-1.0/lib/foo.c}.
5928 Likewise, the subdirectories of the source path are not searched: if
5929 the source path is @file{/mnt/cross}, and the binary refers to
5930 @file{foo.c}, @value{GDBN} would not find it under
5931 @file{/mnt/cross/usr/src/foo-1.0/lib}.
5933 Plain file names, relative file names with leading directories, file
5934 names containing dots, etc.@: are all treated as described above; for
5935 instance, if the source path is @file{/mnt/cross}, and the source file
5936 is recorded as @file{../lib/foo.c}, @value{GDBN} would first try
5937 @file{../lib/foo.c}, then @file{/mnt/cross/../lib/foo.c}, and after
5938 that---@file{/mnt/cross/foo.c}.
5940 Note that the executable search path is @emph{not} used to locate the
5943 Whenever you reset or rearrange the source path, @value{GDBN} clears out
5944 any information it has cached about where source files are found and where
5945 each line is in the file.
5949 When you start @value{GDBN}, its source path includes only @samp{cdir}
5950 and @samp{cwd}, in that order.
5951 To add other directories, use the @code{directory} command.
5953 The search path is used to find both program source files and @value{GDBN}
5954 script files (read using the @samp{-command} option and @samp{source} command).
5956 In addition to the source path, @value{GDBN} provides a set of commands
5957 that manage a list of source path substitution rules. A @dfn{substitution
5958 rule} specifies how to rewrite source directories stored in the program's
5959 debug information in case the sources were moved to a different
5960 directory between compilation and debugging. A rule is made of
5961 two strings, the first specifying what needs to be rewritten in
5962 the path, and the second specifying how it should be rewritten.
5963 In @ref{set substitute-path}, we name these two parts @var{from} and
5964 @var{to} respectively. @value{GDBN} does a simple string replacement
5965 of @var{from} with @var{to} at the start of the directory part of the
5966 source file name, and uses that result instead of the original file
5967 name to look up the sources.
5969 Using the previous example, suppose the @file{foo-1.0} tree has been
5970 moved from @file{/usr/src} to @file{/mnt/cross}, then you can tell
5971 @value{GDBN} to replace @file{/usr/src} in all source path names with
5972 @file{/mnt/cross}. The first lookup will then be
5973 @file{/mnt/cross/foo-1.0/lib/foo.c} in place of the original location
5974 of @file{/usr/src/foo-1.0/lib/foo.c}. To define a source path
5975 substitution rule, use the @code{set substitute-path} command
5976 (@pxref{set substitute-path}).
5978 To avoid unexpected substitution results, a rule is applied only if the
5979 @var{from} part of the directory name ends at a directory separator.
5980 For instance, a rule substituting @file{/usr/source} into
5981 @file{/mnt/cross} will be applied to @file{/usr/source/foo-1.0} but
5982 not to @file{/usr/sourceware/foo-2.0}. And because the substitution
5983 is applied only at the beginning of the directory name, this rule will
5984 not be applied to @file{/root/usr/source/baz.c} either.
5986 In many cases, you can achieve the same result using the @code{directory}
5987 command. However, @code{set substitute-path} can be more efficient in
5988 the case where the sources are organized in a complex tree with multiple
5989 subdirectories. With the @code{directory} command, you need to add each
5990 subdirectory of your project. If you moved the entire tree while
5991 preserving its internal organization, then @code{set substitute-path}
5992 allows you to direct the debugger to all the sources with one single
5995 @code{set substitute-path} is also more than just a shortcut command.
5996 The source path is only used if the file at the original location no
5997 longer exists. On the other hand, @code{set substitute-path} modifies
5998 the debugger behavior to look at the rewritten location instead. So, if
5999 for any reason a source file that is not relevant to your executable is
6000 located at the original location, a substitution rule is the only
6001 method available to point @value{GDBN} at the new location.
6003 @cindex @samp{--with-relocated-sources}
6004 @cindex default source path substitution
6005 You can configure a default source path substitution rule by
6006 configuring @value{GDBN} with the
6007 @samp{--with-relocated-sources=@var{dir}} option. The @var{dir}
6008 should be the name of a directory under @value{GDBN}'s configured
6009 prefix (set with @samp{--prefix} or @samp{--exec-prefix}), and
6010 directory names in debug information under @var{dir} will be adjusted
6011 automatically if the installed @value{GDBN} is moved to a new
6012 location. This is useful if @value{GDBN}, libraries or executables
6013 with debug information and corresponding source code are being moved
6017 @item directory @var{dirname} @dots{}
6018 @item dir @var{dirname} @dots{}
6019 Add directory @var{dirname} to the front of the source path. Several
6020 directory names may be given to this command, separated by @samp{:}
6021 (@samp{;} on MS-DOS and MS-Windows, where @samp{:} usually appears as
6022 part of absolute file names) or
6023 whitespace. You may specify a directory that is already in the source
6024 path; this moves it forward, so @value{GDBN} searches it sooner.
6028 @vindex $cdir@r{, convenience variable}
6029 @vindex $cwd@r{, convenience variable}
6030 @cindex compilation directory
6031 @cindex current directory
6032 @cindex working directory
6033 @cindex directory, current
6034 @cindex directory, compilation
6035 You can use the string @samp{$cdir} to refer to the compilation
6036 directory (if one is recorded), and @samp{$cwd} to refer to the current
6037 working directory. @samp{$cwd} is not the same as @samp{.}---the former
6038 tracks the current working directory as it changes during your @value{GDBN}
6039 session, while the latter is immediately expanded to the current
6040 directory at the time you add an entry to the source path.
6043 Reset the source path to its default value (@samp{$cdir:$cwd} on Unix systems). This requires confirmation.
6045 @c RET-repeat for @code{directory} is explicitly disabled, but since
6046 @c repeating it would be a no-op we do not say that. (thanks to RMS)
6048 @item show directories
6049 @kindex show directories
6050 Print the source path: show which directories it contains.
6052 @anchor{set substitute-path}
6053 @item set substitute-path @var{from} @var{to}
6054 @kindex set substitute-path
6055 Define a source path substitution rule, and add it at the end of the
6056 current list of existing substitution rules. If a rule with the same
6057 @var{from} was already defined, then the old rule is also deleted.
6059 For example, if the file @file{/foo/bar/baz.c} was moved to
6060 @file{/mnt/cross/baz.c}, then the command
6063 (@value{GDBP}) set substitute-path /usr/src /mnt/cross
6067 will tell @value{GDBN} to replace @samp{/usr/src} with
6068 @samp{/mnt/cross}, which will allow @value{GDBN} to find the file
6069 @file{baz.c} even though it was moved.
6071 In the case when more than one substitution rule have been defined,
6072 the rules are evaluated one by one in the order where they have been
6073 defined. The first one matching, if any, is selected to perform
6076 For instance, if we had entered the following commands:
6079 (@value{GDBP}) set substitute-path /usr/src/include /mnt/include
6080 (@value{GDBP}) set substitute-path /usr/src /mnt/src
6084 @value{GDBN} would then rewrite @file{/usr/src/include/defs.h} into
6085 @file{/mnt/include/defs.h} by using the first rule. However, it would
6086 use the second rule to rewrite @file{/usr/src/lib/foo.c} into
6087 @file{/mnt/src/lib/foo.c}.
6090 @item unset substitute-path [path]
6091 @kindex unset substitute-path
6092 If a path is specified, search the current list of substitution rules
6093 for a rule that would rewrite that path. Delete that rule if found.
6094 A warning is emitted by the debugger if no rule could be found.
6096 If no path is specified, then all substitution rules are deleted.
6098 @item show substitute-path [path]
6099 @kindex show substitute-path
6100 If a path is specified, then print the source path substitution rule
6101 which would rewrite that path, if any.
6103 If no path is specified, then print all existing source path substitution
6108 If your source path is cluttered with directories that are no longer of
6109 interest, @value{GDBN} may sometimes cause confusion by finding the wrong
6110 versions of source. You can correct the situation as follows:
6114 Use @code{directory} with no argument to reset the source path to its default value.
6117 Use @code{directory} with suitable arguments to reinstall the
6118 directories you want in the source path. You can add all the
6119 directories in one command.
6123 @section Source and Machine Code
6124 @cindex source line and its code address
6126 You can use the command @code{info line} to map source lines to program
6127 addresses (and vice versa), and the command @code{disassemble} to display
6128 a range of addresses as machine instructions. You can use the command
6129 @code{set disassemble-next-line} to set whether to disassemble next
6130 source line when execution stops. When run under @sc{gnu} Emacs
6131 mode, the @code{info line} command causes the arrow to point to the
6132 line specified. Also, @code{info line} prints addresses in symbolic form as
6137 @item info line @var{linespec}
6138 Print the starting and ending addresses of the compiled code for
6139 source line @var{linespec}. You can specify source lines in any of
6140 the ways documented in @ref{Specify Location}.
6143 For example, we can use @code{info line} to discover the location of
6144 the object code for the first line of function
6145 @code{m4_changequote}:
6147 @c FIXME: I think this example should also show the addresses in
6148 @c symbolic form, as they usually would be displayed.
6150 (@value{GDBP}) info line m4_changequote
6151 Line 895 of "builtin.c" starts at pc 0x634c and ends at 0x6350.
6155 @cindex code address and its source line
6156 We can also inquire (using @code{*@var{addr}} as the form for
6157 @var{linespec}) what source line covers a particular address:
6159 (@value{GDBP}) info line *0x63ff
6160 Line 926 of "builtin.c" starts at pc 0x63e4 and ends at 0x6404.
6163 @cindex @code{$_} and @code{info line}
6164 @cindex @code{x} command, default address
6165 @kindex x@r{(examine), and} info line
6166 After @code{info line}, the default address for the @code{x} command
6167 is changed to the starting address of the line, so that @samp{x/i} is
6168 sufficient to begin examining the machine code (@pxref{Memory,
6169 ,Examining Memory}). Also, this address is saved as the value of the
6170 convenience variable @code{$_} (@pxref{Convenience Vars, ,Convenience
6175 @cindex assembly instructions
6176 @cindex instructions, assembly
6177 @cindex machine instructions
6178 @cindex listing machine instructions
6180 @itemx disassemble /m
6181 @itemx disassemble /r
6182 This specialized command dumps a range of memory as machine
6183 instructions. It can also print mixed source+disassembly by specifying
6184 the @code{/m} modifier and print the raw instructions in hex as well as
6185 in symbolic form by specifying the @code{/r}.
6186 The default memory range is the function surrounding the
6187 program counter of the selected frame. A single argument to this
6188 command is a program counter value; @value{GDBN} dumps the function
6189 surrounding this value. Two arguments specify a range of addresses
6190 (first inclusive, second exclusive) to dump.
6193 The following example shows the disassembly of a range of addresses of
6194 HP PA-RISC 2.0 code:
6197 (@value{GDBP}) disas 0x32c4 0x32e4
6198 Dump of assembler code from 0x32c4 to 0x32e4:
6199 0x32c4 <main+204>: addil 0,dp
6200 0x32c8 <main+208>: ldw 0x22c(sr0,r1),r26
6201 0x32cc <main+212>: ldil 0x3000,r31
6202 0x32d0 <main+216>: ble 0x3f8(sr4,r31)
6203 0x32d4 <main+220>: ldo 0(r31),rp
6204 0x32d8 <main+224>: addil -0x800,dp
6205 0x32dc <main+228>: ldo 0x588(r1),r26
6206 0x32e0 <main+232>: ldil 0x3000,r31
6207 End of assembler dump.
6210 Here is an example showing mixed source+assembly for Intel x86:
6213 (@value{GDBP}) disas /m main
6214 Dump of assembler code for function main:
6216 0x08048330 <main+0>: push %ebp
6217 0x08048331 <main+1>: mov %esp,%ebp
6218 0x08048333 <main+3>: sub $0x8,%esp
6219 0x08048336 <main+6>: and $0xfffffff0,%esp
6220 0x08048339 <main+9>: sub $0x10,%esp
6222 6 printf ("Hello.\n");
6223 0x0804833c <main+12>: movl $0x8048440,(%esp)
6224 0x08048343 <main+19>: call 0x8048284 <puts@@plt>
6228 0x08048348 <main+24>: mov $0x0,%eax
6229 0x0804834d <main+29>: leave
6230 0x0804834e <main+30>: ret
6232 End of assembler dump.
6235 Some architectures have more than one commonly-used set of instruction
6236 mnemonics or other syntax.
6238 For programs that were dynamically linked and use shared libraries,
6239 instructions that call functions or branch to locations in the shared
6240 libraries might show a seemingly bogus location---it's actually a
6241 location of the relocation table. On some architectures, @value{GDBN}
6242 might be able to resolve these to actual function names.
6245 @kindex set disassembly-flavor
6246 @cindex Intel disassembly flavor
6247 @cindex AT&T disassembly flavor
6248 @item set disassembly-flavor @var{instruction-set}
6249 Select the instruction set to use when disassembling the
6250 program via the @code{disassemble} or @code{x/i} commands.
6252 Currently this command is only defined for the Intel x86 family. You
6253 can set @var{instruction-set} to either @code{intel} or @code{att}.
6254 The default is @code{att}, the AT&T flavor used by default by Unix
6255 assemblers for x86-based targets.
6257 @kindex show disassembly-flavor
6258 @item show disassembly-flavor
6259 Show the current setting of the disassembly flavor.
6263 @kindex set disassemble-next-line
6264 @kindex show disassemble-next-line
6265 @item set disassemble-next-line
6266 @itemx show disassemble-next-line
6267 Control whether or not @value{GDBN} will disassemble the next source
6268 line or instruction when execution stops. If ON, @value{GDBN} will
6269 display disassembly of the next source line when execution of the
6270 program being debugged stops. This is @emph{in addition} to
6271 displaying the source line itself, which @value{GDBN} always does if
6272 possible. If the next source line cannot be displayed for some reason
6273 (e.g., if @value{GDBN} cannot find the source file, or there's no line
6274 info in the debug info), @value{GDBN} will display disassembly of the
6275 next @emph{instruction} instead of showing the next source line. If
6276 AUTO, @value{GDBN} will display disassembly of next instruction only
6277 if the source line cannot be displayed. This setting causes
6278 @value{GDBN} to display some feedback when you step through a function
6279 with no line info or whose source file is unavailable. The default is
6280 OFF, which means never display the disassembly of the next line or
6286 @chapter Examining Data
6288 @cindex printing data
6289 @cindex examining data
6292 @c "inspect" is not quite a synonym if you are using Epoch, which we do not
6293 @c document because it is nonstandard... Under Epoch it displays in a
6294 @c different window or something like that.
6295 The usual way to examine data in your program is with the @code{print}
6296 command (abbreviated @code{p}), or its synonym @code{inspect}. It
6297 evaluates and prints the value of an expression of the language your
6298 program is written in (@pxref{Languages, ,Using @value{GDBN} with
6299 Different Languages}).
6302 @item print @var{expr}
6303 @itemx print /@var{f} @var{expr}
6304 @var{expr} is an expression (in the source language). By default the
6305 value of @var{expr} is printed in a format appropriate to its data type;
6306 you can choose a different format by specifying @samp{/@var{f}}, where
6307 @var{f} is a letter specifying the format; see @ref{Output Formats,,Output
6311 @itemx print /@var{f}
6312 @cindex reprint the last value
6313 If you omit @var{expr}, @value{GDBN} displays the last value again (from the
6314 @dfn{value history}; @pxref{Value History, ,Value History}). This allows you to
6315 conveniently inspect the same value in an alternative format.
6318 A more low-level way of examining data is with the @code{x} command.
6319 It examines data in memory at a specified address and prints it in a
6320 specified format. @xref{Memory, ,Examining Memory}.
6322 If you are interested in information about types, or about how the
6323 fields of a struct or a class are declared, use the @code{ptype @var{exp}}
6324 command rather than @code{print}. @xref{Symbols, ,Examining the Symbol
6328 * Expressions:: Expressions
6329 * Ambiguous Expressions:: Ambiguous Expressions
6330 * Variables:: Program variables
6331 * Arrays:: Artificial arrays
6332 * Output Formats:: Output formats
6333 * Memory:: Examining memory
6334 * Auto Display:: Automatic display
6335 * Print Settings:: Print settings
6336 * Value History:: Value history
6337 * Convenience Vars:: Convenience variables
6338 * Registers:: Registers
6339 * Floating Point Hardware:: Floating point hardware
6340 * Vector Unit:: Vector Unit
6341 * OS Information:: Auxiliary data provided by operating system
6342 * Memory Region Attributes:: Memory region attributes
6343 * Dump/Restore Files:: Copy between memory and a file
6344 * Core File Generation:: Cause a program dump its core
6345 * Character Sets:: Debugging programs that use a different
6346 character set than GDB does
6347 * Caching Remote Data:: Data caching for remote targets
6348 * Searching Memory:: Searching memory for a sequence of bytes
6352 @section Expressions
6355 @code{print} and many other @value{GDBN} commands accept an expression and
6356 compute its value. Any kind of constant, variable or operator defined
6357 by the programming language you are using is valid in an expression in
6358 @value{GDBN}. This includes conditional expressions, function calls,
6359 casts, and string constants. It also includes preprocessor macros, if
6360 you compiled your program to include this information; see
6363 @cindex arrays in expressions
6364 @value{GDBN} supports array constants in expressions input by
6365 the user. The syntax is @{@var{element}, @var{element}@dots{}@}. For example,
6366 you can use the command @code{print @{1, 2, 3@}} to create an array
6367 of three integers. If you pass an array to a function or assign it
6368 to a program variable, @value{GDBN} copies the array to memory that
6369 is @code{malloc}ed in the target program.
6371 Because C is so widespread, most of the expressions shown in examples in
6372 this manual are in C. @xref{Languages, , Using @value{GDBN} with Different
6373 Languages}, for information on how to use expressions in other
6376 In this section, we discuss operators that you can use in @value{GDBN}
6377 expressions regardless of your programming language.
6379 @cindex casts, in expressions
6380 Casts are supported in all languages, not just in C, because it is so
6381 useful to cast a number into a pointer in order to examine a structure
6382 at that address in memory.
6383 @c FIXME: casts supported---Mod2 true?
6385 @value{GDBN} supports these operators, in addition to those common
6386 to programming languages:
6390 @samp{@@} is a binary operator for treating parts of memory as arrays.
6391 @xref{Arrays, ,Artificial Arrays}, for more information.
6394 @samp{::} allows you to specify a variable in terms of the file or
6395 function where it is defined. @xref{Variables, ,Program Variables}.
6397 @cindex @{@var{type}@}
6398 @cindex type casting memory
6399 @cindex memory, viewing as typed object
6400 @cindex casts, to view memory
6401 @item @{@var{type}@} @var{addr}
6402 Refers to an object of type @var{type} stored at address @var{addr} in
6403 memory. @var{addr} may be any expression whose value is an integer or
6404 pointer (but parentheses are required around binary operators, just as in
6405 a cast). This construct is allowed regardless of what kind of data is
6406 normally supposed to reside at @var{addr}.
6409 @node Ambiguous Expressions
6410 @section Ambiguous Expressions
6411 @cindex ambiguous expressions
6413 Expressions can sometimes contain some ambiguous elements. For instance,
6414 some programming languages (notably Ada, C@t{++} and Objective-C) permit
6415 a single function name to be defined several times, for application in
6416 different contexts. This is called @dfn{overloading}. Another example
6417 involving Ada is generics. A @dfn{generic package} is similar to C@t{++}
6418 templates and is typically instantiated several times, resulting in
6419 the same function name being defined in different contexts.
6421 In some cases and depending on the language, it is possible to adjust
6422 the expression to remove the ambiguity. For instance in C@t{++}, you
6423 can specify the signature of the function you want to break on, as in
6424 @kbd{break @var{function}(@var{types})}. In Ada, using the fully
6425 qualified name of your function often makes the expression unambiguous
6428 When an ambiguity that needs to be resolved is detected, the debugger
6429 has the capability to display a menu of numbered choices for each
6430 possibility, and then waits for the selection with the prompt @samp{>}.
6431 The first option is always @samp{[0] cancel}, and typing @kbd{0 @key{RET}}
6432 aborts the current command. If the command in which the expression was
6433 used allows more than one choice to be selected, the next option in the
6434 menu is @samp{[1] all}, and typing @kbd{1 @key{RET}} selects all possible
6437 For example, the following session excerpt shows an attempt to set a
6438 breakpoint at the overloaded symbol @code{String::after}.
6439 We choose three particular definitions of that function name:
6441 @c FIXME! This is likely to change to show arg type lists, at least
6444 (@value{GDBP}) b String::after
6447 [2] file:String.cc; line number:867
6448 [3] file:String.cc; line number:860
6449 [4] file:String.cc; line number:875
6450 [5] file:String.cc; line number:853
6451 [6] file:String.cc; line number:846
6452 [7] file:String.cc; line number:735
6454 Breakpoint 1 at 0xb26c: file String.cc, line 867.
6455 Breakpoint 2 at 0xb344: file String.cc, line 875.
6456 Breakpoint 3 at 0xafcc: file String.cc, line 846.
6457 Multiple breakpoints were set.
6458 Use the "delete" command to delete unwanted
6465 @kindex set multiple-symbols
6466 @item set multiple-symbols @var{mode}
6467 @cindex multiple-symbols menu
6469 This option allows you to adjust the debugger behavior when an expression
6472 By default, @var{mode} is set to @code{all}. If the command with which
6473 the expression is used allows more than one choice, then @value{GDBN}
6474 automatically selects all possible choices. For instance, inserting
6475 a breakpoint on a function using an ambiguous name results in a breakpoint
6476 inserted on each possible match. However, if a unique choice must be made,
6477 then @value{GDBN} uses the menu to help you disambiguate the expression.
6478 For instance, printing the address of an overloaded function will result
6479 in the use of the menu.
6481 When @var{mode} is set to @code{ask}, the debugger always uses the menu
6482 when an ambiguity is detected.
6484 Finally, when @var{mode} is set to @code{cancel}, the debugger reports
6485 an error due to the ambiguity and the command is aborted.
6487 @kindex show multiple-symbols
6488 @item show multiple-symbols
6489 Show the current value of the @code{multiple-symbols} setting.
6493 @section Program Variables
6495 The most common kind of expression to use is the name of a variable
6498 Variables in expressions are understood in the selected stack frame
6499 (@pxref{Selection, ,Selecting a Frame}); they must be either:
6503 global (or file-static)
6510 visible according to the scope rules of the
6511 programming language from the point of execution in that frame
6514 @noindent This means that in the function
6529 you can examine and use the variable @code{a} whenever your program is
6530 executing within the function @code{foo}, but you can only use or
6531 examine the variable @code{b} while your program is executing inside
6532 the block where @code{b} is declared.
6534 @cindex variable name conflict
6535 There is an exception: you can refer to a variable or function whose
6536 scope is a single source file even if the current execution point is not
6537 in this file. But it is possible to have more than one such variable or
6538 function with the same name (in different source files). If that
6539 happens, referring to that name has unpredictable effects. If you wish,
6540 you can specify a static variable in a particular function or file,
6541 using the colon-colon (@code{::}) notation:
6543 @cindex colon-colon, context for variables/functions
6545 @c info cannot cope with a :: index entry, but why deprive hard copy readers?
6546 @cindex @code{::}, context for variables/functions
6549 @var{file}::@var{variable}
6550 @var{function}::@var{variable}
6554 Here @var{file} or @var{function} is the name of the context for the
6555 static @var{variable}. In the case of file names, you can use quotes to
6556 make sure @value{GDBN} parses the file name as a single word---for example,
6557 to print a global value of @code{x} defined in @file{f2.c}:
6560 (@value{GDBP}) p 'f2.c'::x
6563 @cindex C@t{++} scope resolution
6564 This use of @samp{::} is very rarely in conflict with the very similar
6565 use of the same notation in C@t{++}. @value{GDBN} also supports use of the C@t{++}
6566 scope resolution operator in @value{GDBN} expressions.
6567 @c FIXME: Um, so what happens in one of those rare cases where it's in
6570 @cindex wrong values
6571 @cindex variable values, wrong
6572 @cindex function entry/exit, wrong values of variables
6573 @cindex optimized code, wrong values of variables
6575 @emph{Warning:} Occasionally, a local variable may appear to have the
6576 wrong value at certain points in a function---just after entry to a new
6577 scope, and just before exit.
6579 You may see this problem when you are stepping by machine instructions.
6580 This is because, on most machines, it takes more than one instruction to
6581 set up a stack frame (including local variable definitions); if you are
6582 stepping by machine instructions, variables may appear to have the wrong
6583 values until the stack frame is completely built. On exit, it usually
6584 also takes more than one machine instruction to destroy a stack frame;
6585 after you begin stepping through that group of instructions, local
6586 variable definitions may be gone.
6588 This may also happen when the compiler does significant optimizations.
6589 To be sure of always seeing accurate values, turn off all optimization
6592 @cindex ``No symbol "foo" in current context''
6593 Another possible effect of compiler optimizations is to optimize
6594 unused variables out of existence, or assign variables to registers (as
6595 opposed to memory addresses). Depending on the support for such cases
6596 offered by the debug info format used by the compiler, @value{GDBN}
6597 might not be able to display values for such local variables. If that
6598 happens, @value{GDBN} will print a message like this:
6601 No symbol "foo" in current context.
6604 To solve such problems, either recompile without optimizations, or use a
6605 different debug info format, if the compiler supports several such
6606 formats. For example, @value{NGCC}, the @sc{gnu} C/C@t{++} compiler,
6607 usually supports the @option{-gstabs+} option. @option{-gstabs+}
6608 produces debug info in a format that is superior to formats such as
6609 COFF. You may be able to use DWARF 2 (@option{-gdwarf-2}), which is also
6610 an effective form for debug info. @xref{Debugging Options,,Options
6611 for Debugging Your Program or GCC, gcc.info, Using the @sc{gnu}
6612 Compiler Collection (GCC)}.
6613 @xref{C, ,C and C@t{++}}, for more information about debug info formats
6614 that are best suited to C@t{++} programs.
6616 If you ask to print an object whose contents are unknown to
6617 @value{GDBN}, e.g., because its data type is not completely specified
6618 by the debug information, @value{GDBN} will say @samp{<incomplete
6619 type>}. @xref{Symbols, incomplete type}, for more about this.
6621 Strings are identified as arrays of @code{char} values without specified
6622 signedness. Arrays of either @code{signed char} or @code{unsigned char} get
6623 printed as arrays of 1 byte sized integers. @code{-fsigned-char} or
6624 @code{-funsigned-char} @value{NGCC} options have no effect as @value{GDBN}
6625 defines literal string type @code{"char"} as @code{char} without a sign.
6630 signed char var1[] = "A";
6633 You get during debugging
6638 $2 = @{65 'A', 0 '\0'@}
6642 @section Artificial Arrays
6644 @cindex artificial array
6646 @kindex @@@r{, referencing memory as an array}
6647 It is often useful to print out several successive objects of the
6648 same type in memory; a section of an array, or an array of
6649 dynamically determined size for which only a pointer exists in the
6652 You can do this by referring to a contiguous span of memory as an
6653 @dfn{artificial array}, using the binary operator @samp{@@}. The left
6654 operand of @samp{@@} should be the first element of the desired array
6655 and be an individual object. The right operand should be the desired length
6656 of the array. The result is an array value whose elements are all of
6657 the type of the left argument. The first element is actually the left
6658 argument; the second element comes from bytes of memory immediately
6659 following those that hold the first element, and so on. Here is an
6660 example. If a program says
6663 int *array = (int *) malloc (len * sizeof (int));
6667 you can print the contents of @code{array} with
6673 The left operand of @samp{@@} must reside in memory. Array values made
6674 with @samp{@@} in this way behave just like other arrays in terms of
6675 subscripting, and are coerced to pointers when used in expressions.
6676 Artificial arrays most often appear in expressions via the value history
6677 (@pxref{Value History, ,Value History}), after printing one out.
6679 Another way to create an artificial array is to use a cast.
6680 This re-interprets a value as if it were an array.
6681 The value need not be in memory:
6683 (@value{GDBP}) p/x (short[2])0x12345678
6684 $1 = @{0x1234, 0x5678@}
6687 As a convenience, if you leave the array length out (as in
6688 @samp{(@var{type}[])@var{value}}) @value{GDBN} calculates the size to fill
6689 the value (as @samp{sizeof(@var{value})/sizeof(@var{type})}:
6691 (@value{GDBP}) p/x (short[])0x12345678
6692 $2 = @{0x1234, 0x5678@}
6695 Sometimes the artificial array mechanism is not quite enough; in
6696 moderately complex data structures, the elements of interest may not
6697 actually be adjacent---for example, if you are interested in the values
6698 of pointers in an array. One useful work-around in this situation is
6699 to use a convenience variable (@pxref{Convenience Vars, ,Convenience
6700 Variables}) as a counter in an expression that prints the first
6701 interesting value, and then repeat that expression via @key{RET}. For
6702 instance, suppose you have an array @code{dtab} of pointers to
6703 structures, and you are interested in the values of a field @code{fv}
6704 in each structure. Here is an example of what you might type:
6714 @node Output Formats
6715 @section Output Formats
6717 @cindex formatted output
6718 @cindex output formats
6719 By default, @value{GDBN} prints a value according to its data type. Sometimes
6720 this is not what you want. For example, you might want to print a number
6721 in hex, or a pointer in decimal. Or you might want to view data in memory
6722 at a certain address as a character string or as an instruction. To do
6723 these things, specify an @dfn{output format} when you print a value.
6725 The simplest use of output formats is to say how to print a value
6726 already computed. This is done by starting the arguments of the
6727 @code{print} command with a slash and a format letter. The format
6728 letters supported are:
6732 Regard the bits of the value as an integer, and print the integer in
6736 Print as integer in signed decimal.
6739 Print as integer in unsigned decimal.
6742 Print as integer in octal.
6745 Print as integer in binary. The letter @samp{t} stands for ``two''.
6746 @footnote{@samp{b} cannot be used because these format letters are also
6747 used with the @code{x} command, where @samp{b} stands for ``byte'';
6748 see @ref{Memory,,Examining Memory}.}
6751 @cindex unknown address, locating
6752 @cindex locate address
6753 Print as an address, both absolute in hexadecimal and as an offset from
6754 the nearest preceding symbol. You can use this format used to discover
6755 where (in what function) an unknown address is located:
6758 (@value{GDBP}) p/a 0x54320
6759 $3 = 0x54320 <_initialize_vx+396>
6763 The command @code{info symbol 0x54320} yields similar results.
6764 @xref{Symbols, info symbol}.
6767 Regard as an integer and print it as a character constant. This
6768 prints both the numerical value and its character representation. The
6769 character representation is replaced with the octal escape @samp{\nnn}
6770 for characters outside the 7-bit @sc{ascii} range.
6772 Without this format, @value{GDBN} displays @code{char},
6773 @w{@code{unsigned char}}, and @w{@code{signed char}} data as character
6774 constants. Single-byte members of vectors are displayed as integer
6778 Regard the bits of the value as a floating point number and print
6779 using typical floating point syntax.
6782 @cindex printing strings
6783 @cindex printing byte arrays
6784 Regard as a string, if possible. With this format, pointers to single-byte
6785 data are displayed as null-terminated strings and arrays of single-byte data
6786 are displayed as fixed-length strings. Other values are displayed in their
6789 Without this format, @value{GDBN} displays pointers to and arrays of
6790 @code{char}, @w{@code{unsigned char}}, and @w{@code{signed char}} as
6791 strings. Single-byte members of a vector are displayed as an integer
6795 @cindex raw printing
6796 Print using the @samp{raw} formatting. By default, @value{GDBN} will
6797 use a type-specific pretty-printer. The @samp{r} format bypasses any
6798 pretty-printer which might exist for the value's type.
6801 For example, to print the program counter in hex (@pxref{Registers}), type
6808 Note that no space is required before the slash; this is because command
6809 names in @value{GDBN} cannot contain a slash.
6811 To reprint the last value in the value history with a different format,
6812 you can use the @code{print} command with just a format and no
6813 expression. For example, @samp{p/x} reprints the last value in hex.
6816 @section Examining Memory
6818 You can use the command @code{x} (for ``examine'') to examine memory in
6819 any of several formats, independently of your program's data types.
6821 @cindex examining memory
6823 @kindex x @r{(examine memory)}
6824 @item x/@var{nfu} @var{addr}
6827 Use the @code{x} command to examine memory.
6830 @var{n}, @var{f}, and @var{u} are all optional parameters that specify how
6831 much memory to display and how to format it; @var{addr} is an
6832 expression giving the address where you want to start displaying memory.
6833 If you use defaults for @var{nfu}, you need not type the slash @samp{/}.
6834 Several commands set convenient defaults for @var{addr}.
6837 @item @var{n}, the repeat count
6838 The repeat count is a decimal integer; the default is 1. It specifies
6839 how much memory (counting by units @var{u}) to display.
6840 @c This really is **decimal**; unaffected by 'set radix' as of GDB
6843 @item @var{f}, the display format
6844 The display format is one of the formats used by @code{print}
6845 (@samp{x}, @samp{d}, @samp{u}, @samp{o}, @samp{t}, @samp{a}, @samp{c},
6846 @samp{f}, @samp{s}), and in addition @samp{i} (for machine instructions).
6847 The default is @samp{x} (hexadecimal) initially. The default changes
6848 each time you use either @code{x} or @code{print}.
6850 @item @var{u}, the unit size
6851 The unit size is any of
6857 Halfwords (two bytes).
6859 Words (four bytes). This is the initial default.
6861 Giant words (eight bytes).
6864 Each time you specify a unit size with @code{x}, that size becomes the
6865 default unit the next time you use @code{x}. (For the @samp{s} and
6866 @samp{i} formats, the unit size is ignored and is normally not written.)
6868 @item @var{addr}, starting display address
6869 @var{addr} is the address where you want @value{GDBN} to begin displaying
6870 memory. The expression need not have a pointer value (though it may);
6871 it is always interpreted as an integer address of a byte of memory.
6872 @xref{Expressions, ,Expressions}, for more information on expressions. The default for
6873 @var{addr} is usually just after the last address examined---but several
6874 other commands also set the default address: @code{info breakpoints} (to
6875 the address of the last breakpoint listed), @code{info line} (to the
6876 starting address of a line), and @code{print} (if you use it to display
6877 a value from memory).
6880 For example, @samp{x/3uh 0x54320} is a request to display three halfwords
6881 (@code{h}) of memory, formatted as unsigned decimal integers (@samp{u}),
6882 starting at address @code{0x54320}. @samp{x/4xw $sp} prints the four
6883 words (@samp{w}) of memory above the stack pointer (here, @samp{$sp};
6884 @pxref{Registers, ,Registers}) in hexadecimal (@samp{x}).
6886 Since the letters indicating unit sizes are all distinct from the
6887 letters specifying output formats, you do not have to remember whether
6888 unit size or format comes first; either order works. The output
6889 specifications @samp{4xw} and @samp{4wx} mean exactly the same thing.
6890 (However, the count @var{n} must come first; @samp{wx4} does not work.)
6892 Even though the unit size @var{u} is ignored for the formats @samp{s}
6893 and @samp{i}, you might still want to use a count @var{n}; for example,
6894 @samp{3i} specifies that you want to see three machine instructions,
6895 including any operands. For convenience, especially when used with
6896 the @code{display} command, the @samp{i} format also prints branch delay
6897 slot instructions, if any, beyond the count specified, which immediately
6898 follow the last instruction that is within the count. The command
6899 @code{disassemble} gives an alternative way of inspecting machine
6900 instructions; see @ref{Machine Code,,Source and Machine Code}.
6902 All the defaults for the arguments to @code{x} are designed to make it
6903 easy to continue scanning memory with minimal specifications each time
6904 you use @code{x}. For example, after you have inspected three machine
6905 instructions with @samp{x/3i @var{addr}}, you can inspect the next seven
6906 with just @samp{x/7}. If you use @key{RET} to repeat the @code{x} command,
6907 the repeat count @var{n} is used again; the other arguments default as
6908 for successive uses of @code{x}.
6910 @cindex @code{$_}, @code{$__}, and value history
6911 The addresses and contents printed by the @code{x} command are not saved
6912 in the value history because there is often too much of them and they
6913 would get in the way. Instead, @value{GDBN} makes these values available for
6914 subsequent use in expressions as values of the convenience variables
6915 @code{$_} and @code{$__}. After an @code{x} command, the last address
6916 examined is available for use in expressions in the convenience variable
6917 @code{$_}. The contents of that address, as examined, are available in
6918 the convenience variable @code{$__}.
6920 If the @code{x} command has a repeat count, the address and contents saved
6921 are from the last memory unit printed; this is not the same as the last
6922 address printed if several units were printed on the last line of output.
6924 @cindex remote memory comparison
6925 @cindex verify remote memory image
6926 When you are debugging a program running on a remote target machine
6927 (@pxref{Remote Debugging}), you may wish to verify the program's image in the
6928 remote machine's memory against the executable file you downloaded to
6929 the target. The @code{compare-sections} command is provided for such
6933 @kindex compare-sections
6934 @item compare-sections @r{[}@var{section-name}@r{]}
6935 Compare the data of a loadable section @var{section-name} in the
6936 executable file of the program being debugged with the same section in
6937 the remote machine's memory, and report any mismatches. With no
6938 arguments, compares all loadable sections. This command's
6939 availability depends on the target's support for the @code{"qCRC"}
6944 @section Automatic Display
6945 @cindex automatic display
6946 @cindex display of expressions
6948 If you find that you want to print the value of an expression frequently
6949 (to see how it changes), you might want to add it to the @dfn{automatic
6950 display list} so that @value{GDBN} prints its value each time your program stops.
6951 Each expression added to the list is given a number to identify it;
6952 to remove an expression from the list, you specify that number.
6953 The automatic display looks like this:
6957 3: bar[5] = (struct hack *) 0x3804
6961 This display shows item numbers, expressions and their current values. As with
6962 displays you request manually using @code{x} or @code{print}, you can
6963 specify the output format you prefer; in fact, @code{display} decides
6964 whether to use @code{print} or @code{x} depending your format
6965 specification---it uses @code{x} if you specify either the @samp{i}
6966 or @samp{s} format, or a unit size; otherwise it uses @code{print}.
6970 @item display @var{expr}
6971 Add the expression @var{expr} to the list of expressions to display
6972 each time your program stops. @xref{Expressions, ,Expressions}.
6974 @code{display} does not repeat if you press @key{RET} again after using it.
6976 @item display/@var{fmt} @var{expr}
6977 For @var{fmt} specifying only a display format and not a size or
6978 count, add the expression @var{expr} to the auto-display list but
6979 arrange to display it each time in the specified format @var{fmt}.
6980 @xref{Output Formats,,Output Formats}.
6982 @item display/@var{fmt} @var{addr}
6983 For @var{fmt} @samp{i} or @samp{s}, or including a unit-size or a
6984 number of units, add the expression @var{addr} as a memory address to
6985 be examined each time your program stops. Examining means in effect
6986 doing @samp{x/@var{fmt} @var{addr}}. @xref{Memory, ,Examining Memory}.
6989 For example, @samp{display/i $pc} can be helpful, to see the machine
6990 instruction about to be executed each time execution stops (@samp{$pc}
6991 is a common name for the program counter; @pxref{Registers, ,Registers}).
6994 @kindex delete display
6996 @item undisplay @var{dnums}@dots{}
6997 @itemx delete display @var{dnums}@dots{}
6998 Remove item numbers @var{dnums} from the list of expressions to display.
7000 @code{undisplay} does not repeat if you press @key{RET} after using it.
7001 (Otherwise you would just get the error @samp{No display number @dots{}}.)
7003 @kindex disable display
7004 @item disable display @var{dnums}@dots{}
7005 Disable the display of item numbers @var{dnums}. A disabled display
7006 item is not printed automatically, but is not forgotten. It may be
7007 enabled again later.
7009 @kindex enable display
7010 @item enable display @var{dnums}@dots{}
7011 Enable display of item numbers @var{dnums}. It becomes effective once
7012 again in auto display of its expression, until you specify otherwise.
7015 Display the current values of the expressions on the list, just as is
7016 done when your program stops.
7018 @kindex info display
7020 Print the list of expressions previously set up to display
7021 automatically, each one with its item number, but without showing the
7022 values. This includes disabled expressions, which are marked as such.
7023 It also includes expressions which would not be displayed right now
7024 because they refer to automatic variables not currently available.
7027 @cindex display disabled out of scope
7028 If a display expression refers to local variables, then it does not make
7029 sense outside the lexical context for which it was set up. Such an
7030 expression is disabled when execution enters a context where one of its
7031 variables is not defined. For example, if you give the command
7032 @code{display last_char} while inside a function with an argument
7033 @code{last_char}, @value{GDBN} displays this argument while your program
7034 continues to stop inside that function. When it stops elsewhere---where
7035 there is no variable @code{last_char}---the display is disabled
7036 automatically. The next time your program stops where @code{last_char}
7037 is meaningful, you can enable the display expression once again.
7039 @node Print Settings
7040 @section Print Settings
7042 @cindex format options
7043 @cindex print settings
7044 @value{GDBN} provides the following ways to control how arrays, structures,
7045 and symbols are printed.
7048 These settings are useful for debugging programs in any language:
7052 @item set print address
7053 @itemx set print address on
7054 @cindex print/don't print memory addresses
7055 @value{GDBN} prints memory addresses showing the location of stack
7056 traces, structure values, pointer values, breakpoints, and so forth,
7057 even when it also displays the contents of those addresses. The default
7058 is @code{on}. For example, this is what a stack frame display looks like with
7059 @code{set print address on}:
7064 #0 set_quotes (lq=0x34c78 "<<", rq=0x34c88 ">>")
7066 530 if (lquote != def_lquote)
7070 @item set print address off
7071 Do not print addresses when displaying their contents. For example,
7072 this is the same stack frame displayed with @code{set print address off}:
7076 (@value{GDBP}) set print addr off
7078 #0 set_quotes (lq="<<", rq=">>") at input.c:530
7079 530 if (lquote != def_lquote)
7083 You can use @samp{set print address off} to eliminate all machine
7084 dependent displays from the @value{GDBN} interface. For example, with
7085 @code{print address off}, you should get the same text for backtraces on
7086 all machines---whether or not they involve pointer arguments.
7089 @item show print address
7090 Show whether or not addresses are to be printed.
7093 When @value{GDBN} prints a symbolic address, it normally prints the
7094 closest earlier symbol plus an offset. If that symbol does not uniquely
7095 identify the address (for example, it is a name whose scope is a single
7096 source file), you may need to clarify. One way to do this is with
7097 @code{info line}, for example @samp{info line *0x4537}. Alternately,
7098 you can set @value{GDBN} to print the source file and line number when
7099 it prints a symbolic address:
7102 @item set print symbol-filename on
7103 @cindex source file and line of a symbol
7104 @cindex symbol, source file and line
7105 Tell @value{GDBN} to print the source file name and line number of a
7106 symbol in the symbolic form of an address.
7108 @item set print symbol-filename off
7109 Do not print source file name and line number of a symbol. This is the
7112 @item show print symbol-filename
7113 Show whether or not @value{GDBN} will print the source file name and
7114 line number of a symbol in the symbolic form of an address.
7117 Another situation where it is helpful to show symbol filenames and line
7118 numbers is when disassembling code; @value{GDBN} shows you the line
7119 number and source file that corresponds to each instruction.
7121 Also, you may wish to see the symbolic form only if the address being
7122 printed is reasonably close to the closest earlier symbol:
7125 @item set print max-symbolic-offset @var{max-offset}
7126 @cindex maximum value for offset of closest symbol
7127 Tell @value{GDBN} to only display the symbolic form of an address if the
7128 offset between the closest earlier symbol and the address is less than
7129 @var{max-offset}. The default is 0, which tells @value{GDBN}
7130 to always print the symbolic form of an address if any symbol precedes it.
7132 @item show print max-symbolic-offset
7133 Ask how large the maximum offset is that @value{GDBN} prints in a
7137 @cindex wild pointer, interpreting
7138 @cindex pointer, finding referent
7139 If you have a pointer and you are not sure where it points, try
7140 @samp{set print symbol-filename on}. Then you can determine the name
7141 and source file location of the variable where it points, using
7142 @samp{p/a @var{pointer}}. This interprets the address in symbolic form.
7143 For example, here @value{GDBN} shows that a variable @code{ptt} points
7144 at another variable @code{t}, defined in @file{hi2.c}:
7147 (@value{GDBP}) set print symbol-filename on
7148 (@value{GDBP}) p/a ptt
7149 $4 = 0xe008 <t in hi2.c>
7153 @emph{Warning:} For pointers that point to a local variable, @samp{p/a}
7154 does not show the symbol name and filename of the referent, even with
7155 the appropriate @code{set print} options turned on.
7158 Other settings control how different kinds of objects are printed:
7161 @item set print array
7162 @itemx set print array on
7163 @cindex pretty print arrays
7164 Pretty print arrays. This format is more convenient to read,
7165 but uses more space. The default is off.
7167 @item set print array off
7168 Return to compressed format for arrays.
7170 @item show print array
7171 Show whether compressed or pretty format is selected for displaying
7174 @cindex print array indexes
7175 @item set print array-indexes
7176 @itemx set print array-indexes on
7177 Print the index of each element when displaying arrays. May be more
7178 convenient to locate a given element in the array or quickly find the
7179 index of a given element in that printed array. The default is off.
7181 @item set print array-indexes off
7182 Stop printing element indexes when displaying arrays.
7184 @item show print array-indexes
7185 Show whether the index of each element is printed when displaying
7188 @item set print elements @var{number-of-elements}
7189 @cindex number of array elements to print
7190 @cindex limit on number of printed array elements
7191 Set a limit on how many elements of an array @value{GDBN} will print.
7192 If @value{GDBN} is printing a large array, it stops printing after it has
7193 printed the number of elements set by the @code{set print elements} command.
7194 This limit also applies to the display of strings.
7195 When @value{GDBN} starts, this limit is set to 200.
7196 Setting @var{number-of-elements} to zero means that the printing is unlimited.
7198 @item show print elements
7199 Display the number of elements of a large array that @value{GDBN} will print.
7200 If the number is 0, then the printing is unlimited.
7202 @item set print frame-arguments @var{value}
7203 @kindex set print frame-arguments
7204 @cindex printing frame argument values
7205 @cindex print all frame argument values
7206 @cindex print frame argument values for scalars only
7207 @cindex do not print frame argument values
7208 This command allows to control how the values of arguments are printed
7209 when the debugger prints a frame (@pxref{Frames}). The possible
7214 The values of all arguments are printed.
7217 Print the value of an argument only if it is a scalar. The value of more
7218 complex arguments such as arrays, structures, unions, etc, is replaced
7219 by @code{@dots{}}. This is the default. Here is an example where
7220 only scalar arguments are shown:
7223 #1 0x08048361 in call_me (i=3, s=@dots{}, ss=0xbf8d508c, u=@dots{}, e=green)
7228 None of the argument values are printed. Instead, the value of each argument
7229 is replaced by @code{@dots{}}. In this case, the example above now becomes:
7232 #1 0x08048361 in call_me (i=@dots{}, s=@dots{}, ss=@dots{}, u=@dots{}, e=@dots{})
7237 By default, only scalar arguments are printed. This command can be used
7238 to configure the debugger to print the value of all arguments, regardless
7239 of their type. However, it is often advantageous to not print the value
7240 of more complex parameters. For instance, it reduces the amount of
7241 information printed in each frame, making the backtrace more readable.
7242 Also, it improves performance when displaying Ada frames, because
7243 the computation of large arguments can sometimes be CPU-intensive,
7244 especially in large applications. Setting @code{print frame-arguments}
7245 to @code{scalars} (the default) or @code{none} avoids this computation,
7246 thus speeding up the display of each Ada frame.
7248 @item show print frame-arguments
7249 Show how the value of arguments should be displayed when printing a frame.
7251 @item set print repeats
7252 @cindex repeated array elements
7253 Set the threshold for suppressing display of repeated array
7254 elements. When the number of consecutive identical elements of an
7255 array exceeds the threshold, @value{GDBN} prints the string
7256 @code{"<repeats @var{n} times>"}, where @var{n} is the number of
7257 identical repetitions, instead of displaying the identical elements
7258 themselves. Setting the threshold to zero will cause all elements to
7259 be individually printed. The default threshold is 10.
7261 @item show print repeats
7262 Display the current threshold for printing repeated identical
7265 @item set print null-stop
7266 @cindex @sc{null} elements in arrays
7267 Cause @value{GDBN} to stop printing the characters of an array when the first
7268 @sc{null} is encountered. This is useful when large arrays actually
7269 contain only short strings.
7272 @item show print null-stop
7273 Show whether @value{GDBN} stops printing an array on the first
7274 @sc{null} character.
7276 @item set print pretty on
7277 @cindex print structures in indented form
7278 @cindex indentation in structure display
7279 Cause @value{GDBN} to print structures in an indented format with one member
7280 per line, like this:
7295 @item set print pretty off
7296 Cause @value{GDBN} to print structures in a compact format, like this:
7300 $1 = @{next = 0x0, flags = @{sweet = 1, sour = 1@}, \
7301 meat = 0x54 "Pork"@}
7306 This is the default format.
7308 @item show print pretty
7309 Show which format @value{GDBN} is using to print structures.
7311 @item set print sevenbit-strings on
7312 @cindex eight-bit characters in strings
7313 @cindex octal escapes in strings
7314 Print using only seven-bit characters; if this option is set,
7315 @value{GDBN} displays any eight-bit characters (in strings or
7316 character values) using the notation @code{\}@var{nnn}. This setting is
7317 best if you are working in English (@sc{ascii}) and you use the
7318 high-order bit of characters as a marker or ``meta'' bit.
7320 @item set print sevenbit-strings off
7321 Print full eight-bit characters. This allows the use of more
7322 international character sets, and is the default.
7324 @item show print sevenbit-strings
7325 Show whether or not @value{GDBN} is printing only seven-bit characters.
7327 @item set print union on
7328 @cindex unions in structures, printing
7329 Tell @value{GDBN} to print unions which are contained in structures
7330 and other unions. This is the default setting.
7332 @item set print union off
7333 Tell @value{GDBN} not to print unions which are contained in
7334 structures and other unions. @value{GDBN} will print @code{"@{...@}"}
7337 @item show print union
7338 Ask @value{GDBN} whether or not it will print unions which are contained in
7339 structures and other unions.
7341 For example, given the declarations
7344 typedef enum @{Tree, Bug@} Species;
7345 typedef enum @{Big_tree, Acorn, Seedling@} Tree_forms;
7346 typedef enum @{Caterpillar, Cocoon, Butterfly@}
7357 struct thing foo = @{Tree, @{Acorn@}@};
7361 with @code{set print union on} in effect @samp{p foo} would print
7364 $1 = @{it = Tree, form = @{tree = Acorn, bug = Cocoon@}@}
7368 and with @code{set print union off} in effect it would print
7371 $1 = @{it = Tree, form = @{...@}@}
7375 @code{set print union} affects programs written in C-like languages
7381 These settings are of interest when debugging C@t{++} programs:
7384 @cindex demangling C@t{++} names
7385 @item set print demangle
7386 @itemx set print demangle on
7387 Print C@t{++} names in their source form rather than in the encoded
7388 (``mangled'') form passed to the assembler and linker for type-safe
7389 linkage. The default is on.
7391 @item show print demangle
7392 Show whether C@t{++} names are printed in mangled or demangled form.
7394 @item set print asm-demangle
7395 @itemx set print asm-demangle on
7396 Print C@t{++} names in their source form rather than their mangled form, even
7397 in assembler code printouts such as instruction disassemblies.
7400 @item show print asm-demangle
7401 Show whether C@t{++} names in assembly listings are printed in mangled
7404 @cindex C@t{++} symbol decoding style
7405 @cindex symbol decoding style, C@t{++}
7406 @kindex set demangle-style
7407 @item set demangle-style @var{style}
7408 Choose among several encoding schemes used by different compilers to
7409 represent C@t{++} names. The choices for @var{style} are currently:
7413 Allow @value{GDBN} to choose a decoding style by inspecting your program.
7416 Decode based on the @sc{gnu} C@t{++} compiler (@code{g++}) encoding algorithm.
7417 This is the default.
7420 Decode based on the HP ANSI C@t{++} (@code{aCC}) encoding algorithm.
7423 Decode based on the Lucid C@t{++} compiler (@code{lcc}) encoding algorithm.
7426 Decode using the algorithm in the @cite{C@t{++} Annotated Reference Manual}.
7427 @strong{Warning:} this setting alone is not sufficient to allow
7428 debugging @code{cfront}-generated executables. @value{GDBN} would
7429 require further enhancement to permit that.
7432 If you omit @var{style}, you will see a list of possible formats.
7434 @item show demangle-style
7435 Display the encoding style currently in use for decoding C@t{++} symbols.
7437 @item set print object
7438 @itemx set print object on
7439 @cindex derived type of an object, printing
7440 @cindex display derived types
7441 When displaying a pointer to an object, identify the @emph{actual}
7442 (derived) type of the object rather than the @emph{declared} type, using
7443 the virtual function table.
7445 @item set print object off
7446 Display only the declared type of objects, without reference to the
7447 virtual function table. This is the default setting.
7449 @item show print object
7450 Show whether actual, or declared, object types are displayed.
7452 @item set print static-members
7453 @itemx set print static-members on
7454 @cindex static members of C@t{++} objects
7455 Print static members when displaying a C@t{++} object. The default is on.
7457 @item set print static-members off
7458 Do not print static members when displaying a C@t{++} object.
7460 @item show print static-members
7461 Show whether C@t{++} static members are printed or not.
7463 @item set print pascal_static-members
7464 @itemx set print pascal_static-members on
7465 @cindex static members of Pascal objects
7466 @cindex Pascal objects, static members display
7467 Print static members when displaying a Pascal object. The default is on.
7469 @item set print pascal_static-members off
7470 Do not print static members when displaying a Pascal object.
7472 @item show print pascal_static-members
7473 Show whether Pascal static members are printed or not.
7475 @c These don't work with HP ANSI C++ yet.
7476 @item set print vtbl
7477 @itemx set print vtbl on
7478 @cindex pretty print C@t{++} virtual function tables
7479 @cindex virtual functions (C@t{++}) display
7480 @cindex VTBL display
7481 Pretty print C@t{++} virtual function tables. The default is off.
7482 (The @code{vtbl} commands do not work on programs compiled with the HP
7483 ANSI C@t{++} compiler (@code{aCC}).)
7485 @item set print vtbl off
7486 Do not pretty print C@t{++} virtual function tables.
7488 @item show print vtbl
7489 Show whether C@t{++} virtual function tables are pretty printed, or not.
7493 @section Value History
7495 @cindex value history
7496 @cindex history of values printed by @value{GDBN}
7497 Values printed by the @code{print} command are saved in the @value{GDBN}
7498 @dfn{value history}. This allows you to refer to them in other expressions.
7499 Values are kept until the symbol table is re-read or discarded
7500 (for example with the @code{file} or @code{symbol-file} commands).
7501 When the symbol table changes, the value history is discarded,
7502 since the values may contain pointers back to the types defined in the
7507 @cindex history number
7508 The values printed are given @dfn{history numbers} by which you can
7509 refer to them. These are successive integers starting with one.
7510 @code{print} shows you the history number assigned to a value by
7511 printing @samp{$@var{num} = } before the value; here @var{num} is the
7514 To refer to any previous value, use @samp{$} followed by the value's
7515 history number. The way @code{print} labels its output is designed to
7516 remind you of this. Just @code{$} refers to the most recent value in
7517 the history, and @code{$$} refers to the value before that.
7518 @code{$$@var{n}} refers to the @var{n}th value from the end; @code{$$2}
7519 is the value just prior to @code{$$}, @code{$$1} is equivalent to
7520 @code{$$}, and @code{$$0} is equivalent to @code{$}.
7522 For example, suppose you have just printed a pointer to a structure and
7523 want to see the contents of the structure. It suffices to type
7529 If you have a chain of structures where the component @code{next} points
7530 to the next one, you can print the contents of the next one with this:
7537 You can print successive links in the chain by repeating this
7538 command---which you can do by just typing @key{RET}.
7540 Note that the history records values, not expressions. If the value of
7541 @code{x} is 4 and you type these commands:
7549 then the value recorded in the value history by the @code{print} command
7550 remains 4 even though the value of @code{x} has changed.
7555 Print the last ten values in the value history, with their item numbers.
7556 This is like @samp{p@ $$9} repeated ten times, except that @code{show
7557 values} does not change the history.
7559 @item show values @var{n}
7560 Print ten history values centered on history item number @var{n}.
7563 Print ten history values just after the values last printed. If no more
7564 values are available, @code{show values +} produces no display.
7567 Pressing @key{RET} to repeat @code{show values @var{n}} has exactly the
7568 same effect as @samp{show values +}.
7570 @node Convenience Vars
7571 @section Convenience Variables
7573 @cindex convenience variables
7574 @cindex user-defined variables
7575 @value{GDBN} provides @dfn{convenience variables} that you can use within
7576 @value{GDBN} to hold on to a value and refer to it later. These variables
7577 exist entirely within @value{GDBN}; they are not part of your program, and
7578 setting a convenience variable has no direct effect on further execution
7579 of your program. That is why you can use them freely.
7581 Convenience variables are prefixed with @samp{$}. Any name preceded by
7582 @samp{$} can be used for a convenience variable, unless it is one of
7583 the predefined machine-specific register names (@pxref{Registers, ,Registers}).
7584 (Value history references, in contrast, are @emph{numbers} preceded
7585 by @samp{$}. @xref{Value History, ,Value History}.)
7587 You can save a value in a convenience variable with an assignment
7588 expression, just as you would set a variable in your program.
7592 set $foo = *object_ptr
7596 would save in @code{$foo} the value contained in the object pointed to by
7599 Using a convenience variable for the first time creates it, but its
7600 value is @code{void} until you assign a new value. You can alter the
7601 value with another assignment at any time.
7603 Convenience variables have no fixed types. You can assign a convenience
7604 variable any type of value, including structures and arrays, even if
7605 that variable already has a value of a different type. The convenience
7606 variable, when used as an expression, has the type of its current value.
7609 @kindex show convenience
7610 @cindex show all user variables
7611 @item show convenience
7612 Print a list of convenience variables used so far, and their values.
7613 Abbreviated @code{show conv}.
7615 @kindex init-if-undefined
7616 @cindex convenience variables, initializing
7617 @item init-if-undefined $@var{variable} = @var{expression}
7618 Set a convenience variable if it has not already been set. This is useful
7619 for user-defined commands that keep some state. It is similar, in concept,
7620 to using local static variables with initializers in C (except that
7621 convenience variables are global). It can also be used to allow users to
7622 override default values used in a command script.
7624 If the variable is already defined then the expression is not evaluated so
7625 any side-effects do not occur.
7628 One of the ways to use a convenience variable is as a counter to be
7629 incremented or a pointer to be advanced. For example, to print
7630 a field from successive elements of an array of structures:
7634 print bar[$i++]->contents
7638 Repeat that command by typing @key{RET}.
7640 Some convenience variables are created automatically by @value{GDBN} and given
7641 values likely to be useful.
7644 @vindex $_@r{, convenience variable}
7646 The variable @code{$_} is automatically set by the @code{x} command to
7647 the last address examined (@pxref{Memory, ,Examining Memory}). Other
7648 commands which provide a default address for @code{x} to examine also
7649 set @code{$_} to that address; these commands include @code{info line}
7650 and @code{info breakpoint}. The type of @code{$_} is @code{void *}
7651 except when set by the @code{x} command, in which case it is a pointer
7652 to the type of @code{$__}.
7654 @vindex $__@r{, convenience variable}
7656 The variable @code{$__} is automatically set by the @code{x} command
7657 to the value found in the last address examined. Its type is chosen
7658 to match the format in which the data was printed.
7661 @vindex $_exitcode@r{, convenience variable}
7662 The variable @code{$_exitcode} is automatically set to the exit code when
7663 the program being debugged terminates.
7666 @vindex $_siginfo@r{, convenience variable}
7667 The variable @code{$_siginfo} is bound to extra signal information
7668 inspection (@pxref{extra signal information}).
7671 On HP-UX systems, if you refer to a function or variable name that
7672 begins with a dollar sign, @value{GDBN} searches for a user or system
7673 name first, before it searches for a convenience variable.
7675 @cindex convenience functions
7676 @value{GDBN} also supplies some @dfn{convenience functions}. These
7677 have a syntax similar to convenience variables. A convenience
7678 function can be used in an expression just like an ordinary function;
7679 however, a convenience function is implemented internally to
7684 @kindex help function
7685 @cindex show all convenience functions
7686 Print a list of all convenience functions.
7693 You can refer to machine register contents, in expressions, as variables
7694 with names starting with @samp{$}. The names of registers are different
7695 for each machine; use @code{info registers} to see the names used on
7699 @kindex info registers
7700 @item info registers
7701 Print the names and values of all registers except floating-point
7702 and vector registers (in the selected stack frame).
7704 @kindex info all-registers
7705 @cindex floating point registers
7706 @item info all-registers
7707 Print the names and values of all registers, including floating-point
7708 and vector registers (in the selected stack frame).
7710 @item info registers @var{regname} @dots{}
7711 Print the @dfn{relativized} value of each specified register @var{regname}.
7712 As discussed in detail below, register values are normally relative to
7713 the selected stack frame. @var{regname} may be any register name valid on
7714 the machine you are using, with or without the initial @samp{$}.
7717 @cindex stack pointer register
7718 @cindex program counter register
7719 @cindex process status register
7720 @cindex frame pointer register
7721 @cindex standard registers
7722 @value{GDBN} has four ``standard'' register names that are available (in
7723 expressions) on most machines---whenever they do not conflict with an
7724 architecture's canonical mnemonics for registers. The register names
7725 @code{$pc} and @code{$sp} are used for the program counter register and
7726 the stack pointer. @code{$fp} is used for a register that contains a
7727 pointer to the current stack frame, and @code{$ps} is used for a
7728 register that contains the processor status. For example,
7729 you could print the program counter in hex with
7736 or print the instruction to be executed next with
7743 or add four to the stack pointer@footnote{This is a way of removing
7744 one word from the stack, on machines where stacks grow downward in
7745 memory (most machines, nowadays). This assumes that the innermost
7746 stack frame is selected; setting @code{$sp} is not allowed when other
7747 stack frames are selected. To pop entire frames off the stack,
7748 regardless of machine architecture, use @code{return};
7749 see @ref{Returning, ,Returning from a Function}.} with
7755 Whenever possible, these four standard register names are available on
7756 your machine even though the machine has different canonical mnemonics,
7757 so long as there is no conflict. The @code{info registers} command
7758 shows the canonical names. For example, on the SPARC, @code{info
7759 registers} displays the processor status register as @code{$psr} but you
7760 can also refer to it as @code{$ps}; and on x86-based machines @code{$ps}
7761 is an alias for the @sc{eflags} register.
7763 @value{GDBN} always considers the contents of an ordinary register as an
7764 integer when the register is examined in this way. Some machines have
7765 special registers which can hold nothing but floating point; these
7766 registers are considered to have floating point values. There is no way
7767 to refer to the contents of an ordinary register as floating point value
7768 (although you can @emph{print} it as a floating point value with
7769 @samp{print/f $@var{regname}}).
7771 Some registers have distinct ``raw'' and ``virtual'' data formats. This
7772 means that the data format in which the register contents are saved by
7773 the operating system is not the same one that your program normally
7774 sees. For example, the registers of the 68881 floating point
7775 coprocessor are always saved in ``extended'' (raw) format, but all C
7776 programs expect to work with ``double'' (virtual) format. In such
7777 cases, @value{GDBN} normally works with the virtual format only (the format
7778 that makes sense for your program), but the @code{info registers} command
7779 prints the data in both formats.
7781 @cindex SSE registers (x86)
7782 @cindex MMX registers (x86)
7783 Some machines have special registers whose contents can be interpreted
7784 in several different ways. For example, modern x86-based machines
7785 have SSE and MMX registers that can hold several values packed
7786 together in several different formats. @value{GDBN} refers to such
7787 registers in @code{struct} notation:
7790 (@value{GDBP}) print $xmm1
7792 v4_float = @{0, 3.43859137e-038, 1.54142831e-044, 1.821688e-044@},
7793 v2_double = @{9.92129282474342e-303, 2.7585945287983262e-313@},
7794 v16_int8 = "\000\000\000\000\3706;\001\v\000\000\000\r\000\000",
7795 v8_int16 = @{0, 0, 14072, 315, 11, 0, 13, 0@},
7796 v4_int32 = @{0, 20657912, 11, 13@},
7797 v2_int64 = @{88725056443645952, 55834574859@},
7798 uint128 = 0x0000000d0000000b013b36f800000000
7803 To set values of such registers, you need to tell @value{GDBN} which
7804 view of the register you wish to change, as if you were assigning
7805 value to a @code{struct} member:
7808 (@value{GDBP}) set $xmm1.uint128 = 0x000000000000000000000000FFFFFFFF
7811 Normally, register values are relative to the selected stack frame
7812 (@pxref{Selection, ,Selecting a Frame}). This means that you get the
7813 value that the register would contain if all stack frames farther in
7814 were exited and their saved registers restored. In order to see the
7815 true contents of hardware registers, you must select the innermost
7816 frame (with @samp{frame 0}).
7818 However, @value{GDBN} must deduce where registers are saved, from the machine
7819 code generated by your compiler. If some registers are not saved, or if
7820 @value{GDBN} is unable to locate the saved registers, the selected stack
7821 frame makes no difference.
7823 @node Floating Point Hardware
7824 @section Floating Point Hardware
7825 @cindex floating point
7827 Depending on the configuration, @value{GDBN} may be able to give
7828 you more information about the status of the floating point hardware.
7833 Display hardware-dependent information about the floating
7834 point unit. The exact contents and layout vary depending on the
7835 floating point chip. Currently, @samp{info float} is supported on
7836 the ARM and x86 machines.
7840 @section Vector Unit
7843 Depending on the configuration, @value{GDBN} may be able to give you
7844 more information about the status of the vector unit.
7849 Display information about the vector unit. The exact contents and
7850 layout vary depending on the hardware.
7853 @node OS Information
7854 @section Operating System Auxiliary Information
7855 @cindex OS information
7857 @value{GDBN} provides interfaces to useful OS facilities that can help
7858 you debug your program.
7860 @cindex @code{ptrace} system call
7861 @cindex @code{struct user} contents
7862 When @value{GDBN} runs on a @dfn{Posix system} (such as GNU or Unix
7863 machines), it interfaces with the inferior via the @code{ptrace}
7864 system call. The operating system creates a special sata structure,
7865 called @code{struct user}, for this interface. You can use the
7866 command @code{info udot} to display the contents of this data
7872 Display the contents of the @code{struct user} maintained by the OS
7873 kernel for the program being debugged. @value{GDBN} displays the
7874 contents of @code{struct user} as a list of hex numbers, similar to
7875 the @code{examine} command.
7878 @cindex auxiliary vector
7879 @cindex vector, auxiliary
7880 Some operating systems supply an @dfn{auxiliary vector} to programs at
7881 startup. This is akin to the arguments and environment that you
7882 specify for a program, but contains a system-dependent variety of
7883 binary values that tell system libraries important details about the
7884 hardware, operating system, and process. Each value's purpose is
7885 identified by an integer tag; the meanings are well-known but system-specific.
7886 Depending on the configuration and operating system facilities,
7887 @value{GDBN} may be able to show you this information. For remote
7888 targets, this functionality may further depend on the remote stub's
7889 support of the @samp{qXfer:auxv:read} packet, see
7890 @ref{qXfer auxiliary vector read}.
7895 Display the auxiliary vector of the inferior, which can be either a
7896 live process or a core dump file. @value{GDBN} prints each tag value
7897 numerically, and also shows names and text descriptions for recognized
7898 tags. Some values in the vector are numbers, some bit masks, and some
7899 pointers to strings or other data. @value{GDBN} displays each value in the
7900 most appropriate form for a recognized tag, and in hexadecimal for
7901 an unrecognized tag.
7904 On some targets, @value{GDBN} can access operating-system-specific information
7905 and display it to user, without interpretation. For remote targets,
7906 this functionality depends on the remote stub's support of the
7907 @samp{qXfer:osdata:read} packet, see @ref{qXfer osdata read}.
7910 @kindex info os processes
7911 @item info os processes
7912 Display the list of processes on the target. For each process,
7913 @value{GDBN} prints the process identifier, the name of the user, and
7914 the command corresponding to the process.
7917 @node Memory Region Attributes
7918 @section Memory Region Attributes
7919 @cindex memory region attributes
7921 @dfn{Memory region attributes} allow you to describe special handling
7922 required by regions of your target's memory. @value{GDBN} uses
7923 attributes to determine whether to allow certain types of memory
7924 accesses; whether to use specific width accesses; and whether to cache
7925 target memory. By default the description of memory regions is
7926 fetched from the target (if the current target supports this), but the
7927 user can override the fetched regions.
7929 Defined memory regions can be individually enabled and disabled. When a
7930 memory region is disabled, @value{GDBN} uses the default attributes when
7931 accessing memory in that region. Similarly, if no memory regions have
7932 been defined, @value{GDBN} uses the default attributes when accessing
7935 When a memory region is defined, it is given a number to identify it;
7936 to enable, disable, or remove a memory region, you specify that number.
7940 @item mem @var{lower} @var{upper} @var{attributes}@dots{}
7941 Define a memory region bounded by @var{lower} and @var{upper} with
7942 attributes @var{attributes}@dots{}, and add it to the list of regions
7943 monitored by @value{GDBN}. Note that @var{upper} == 0 is a special
7944 case: it is treated as the target's maximum memory address.
7945 (0xffff on 16 bit targets, 0xffffffff on 32 bit targets, etc.)
7948 Discard any user changes to the memory regions and use target-supplied
7949 regions, if available, or no regions if the target does not support.
7952 @item delete mem @var{nums}@dots{}
7953 Remove memory regions @var{nums}@dots{} from the list of regions
7954 monitored by @value{GDBN}.
7957 @item disable mem @var{nums}@dots{}
7958 Disable monitoring of memory regions @var{nums}@dots{}.
7959 A disabled memory region is not forgotten.
7960 It may be enabled again later.
7963 @item enable mem @var{nums}@dots{}
7964 Enable monitoring of memory regions @var{nums}@dots{}.
7968 Print a table of all defined memory regions, with the following columns
7972 @item Memory Region Number
7973 @item Enabled or Disabled.
7974 Enabled memory regions are marked with @samp{y}.
7975 Disabled memory regions are marked with @samp{n}.
7978 The address defining the inclusive lower bound of the memory region.
7981 The address defining the exclusive upper bound of the memory region.
7984 The list of attributes set for this memory region.
7989 @subsection Attributes
7991 @subsubsection Memory Access Mode
7992 The access mode attributes set whether @value{GDBN} may make read or
7993 write accesses to a memory region.
7995 While these attributes prevent @value{GDBN} from performing invalid
7996 memory accesses, they do nothing to prevent the target system, I/O DMA,
7997 etc.@: from accessing memory.
8001 Memory is read only.
8003 Memory is write only.
8005 Memory is read/write. This is the default.
8008 @subsubsection Memory Access Size
8009 The access size attribute tells @value{GDBN} to use specific sized
8010 accesses in the memory region. Often memory mapped device registers
8011 require specific sized accesses. If no access size attribute is
8012 specified, @value{GDBN} may use accesses of any size.
8016 Use 8 bit memory accesses.
8018 Use 16 bit memory accesses.
8020 Use 32 bit memory accesses.
8022 Use 64 bit memory accesses.
8025 @c @subsubsection Hardware/Software Breakpoints
8026 @c The hardware/software breakpoint attributes set whether @value{GDBN}
8027 @c will use hardware or software breakpoints for the internal breakpoints
8028 @c used by the step, next, finish, until, etc. commands.
8032 @c Always use hardware breakpoints
8033 @c @item swbreak (default)
8036 @subsubsection Data Cache
8037 The data cache attributes set whether @value{GDBN} will cache target
8038 memory. While this generally improves performance by reducing debug
8039 protocol overhead, it can lead to incorrect results because @value{GDBN}
8040 does not know about volatile variables or memory mapped device
8045 Enable @value{GDBN} to cache target memory.
8047 Disable @value{GDBN} from caching target memory. This is the default.
8050 @subsection Memory Access Checking
8051 @value{GDBN} can be instructed to refuse accesses to memory that is
8052 not explicitly described. This can be useful if accessing such
8053 regions has undesired effects for a specific target, or to provide
8054 better error checking. The following commands control this behaviour.
8057 @kindex set mem inaccessible-by-default
8058 @item set mem inaccessible-by-default [on|off]
8059 If @code{on} is specified, make @value{GDBN} treat memory not
8060 explicitly described by the memory ranges as non-existent and refuse accesses
8061 to such memory. The checks are only performed if there's at least one
8062 memory range defined. If @code{off} is specified, make @value{GDBN}
8063 treat the memory not explicitly described by the memory ranges as RAM.
8064 The default value is @code{on}.
8065 @kindex show mem inaccessible-by-default
8066 @item show mem inaccessible-by-default
8067 Show the current handling of accesses to unknown memory.
8071 @c @subsubsection Memory Write Verification
8072 @c The memory write verification attributes set whether @value{GDBN}
8073 @c will re-reads data after each write to verify the write was successful.
8077 @c @item noverify (default)
8080 @node Dump/Restore Files
8081 @section Copy Between Memory and a File
8082 @cindex dump/restore files
8083 @cindex append data to a file
8084 @cindex dump data to a file
8085 @cindex restore data from a file
8087 You can use the commands @code{dump}, @code{append}, and
8088 @code{restore} to copy data between target memory and a file. The
8089 @code{dump} and @code{append} commands write data to a file, and the
8090 @code{restore} command reads data from a file back into the inferior's
8091 memory. Files may be in binary, Motorola S-record, Intel hex, or
8092 Tektronix Hex format; however, @value{GDBN} can only append to binary
8098 @item dump @r{[}@var{format}@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
8099 @itemx dump @r{[}@var{format}@r{]} value @var{filename} @var{expr}
8100 Dump the contents of memory from @var{start_addr} to @var{end_addr},
8101 or the value of @var{expr}, to @var{filename} in the given format.
8103 The @var{format} parameter may be any one of:
8110 Motorola S-record format.
8112 Tektronix Hex format.
8115 @value{GDBN} uses the same definitions of these formats as the
8116 @sc{gnu} binary utilities, like @samp{objdump} and @samp{objcopy}. If
8117 @var{format} is omitted, @value{GDBN} dumps the data in raw binary
8121 @item append @r{[}binary@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
8122 @itemx append @r{[}binary@r{]} value @var{filename} @var{expr}
8123 Append the contents of memory from @var{start_addr} to @var{end_addr},
8124 or the value of @var{expr}, to the file @var{filename}, in raw binary form.
8125 (@value{GDBN} can only append data to files in raw binary form.)
8128 @item restore @var{filename} @r{[}binary@r{]} @var{bias} @var{start} @var{end}
8129 Restore the contents of file @var{filename} into memory. The
8130 @code{restore} command can automatically recognize any known @sc{bfd}
8131 file format, except for raw binary. To restore a raw binary file you
8132 must specify the optional keyword @code{binary} after the filename.
8134 If @var{bias} is non-zero, its value will be added to the addresses
8135 contained in the file. Binary files always start at address zero, so
8136 they will be restored at address @var{bias}. Other bfd files have
8137 a built-in location; they will be restored at offset @var{bias}
8140 If @var{start} and/or @var{end} are non-zero, then only data between
8141 file offset @var{start} and file offset @var{end} will be restored.
8142 These offsets are relative to the addresses in the file, before
8143 the @var{bias} argument is applied.
8147 @node Core File Generation
8148 @section How to Produce a Core File from Your Program
8149 @cindex dump core from inferior
8151 A @dfn{core file} or @dfn{core dump} is a file that records the memory
8152 image of a running process and its process status (register values
8153 etc.). Its primary use is post-mortem debugging of a program that
8154 crashed while it ran outside a debugger. A program that crashes
8155 automatically produces a core file, unless this feature is disabled by
8156 the user. @xref{Files}, for information on invoking @value{GDBN} in
8157 the post-mortem debugging mode.
8159 Occasionally, you may wish to produce a core file of the program you
8160 are debugging in order to preserve a snapshot of its state.
8161 @value{GDBN} has a special command for that.
8165 @kindex generate-core-file
8166 @item generate-core-file [@var{file}]
8167 @itemx gcore [@var{file}]
8168 Produce a core dump of the inferior process. The optional argument
8169 @var{file} specifies the file name where to put the core dump. If not
8170 specified, the file name defaults to @file{core.@var{pid}}, where
8171 @var{pid} is the inferior process ID.
8173 Note that this command is implemented only for some systems (as of
8174 this writing, @sc{gnu}/Linux, FreeBSD, Solaris, Unixware, and S390).
8177 @node Character Sets
8178 @section Character Sets
8179 @cindex character sets
8181 @cindex translating between character sets
8182 @cindex host character set
8183 @cindex target character set
8185 If the program you are debugging uses a different character set to
8186 represent characters and strings than the one @value{GDBN} uses itself,
8187 @value{GDBN} can automatically translate between the character sets for
8188 you. The character set @value{GDBN} uses we call the @dfn{host
8189 character set}; the one the inferior program uses we call the
8190 @dfn{target character set}.
8192 For example, if you are running @value{GDBN} on a @sc{gnu}/Linux system, which
8193 uses the ISO Latin 1 character set, but you are using @value{GDBN}'s
8194 remote protocol (@pxref{Remote Debugging}) to debug a program
8195 running on an IBM mainframe, which uses the @sc{ebcdic} character set,
8196 then the host character set is Latin-1, and the target character set is
8197 @sc{ebcdic}. If you give @value{GDBN} the command @code{set
8198 target-charset EBCDIC-US}, then @value{GDBN} translates between
8199 @sc{ebcdic} and Latin 1 as you print character or string values, or use
8200 character and string literals in expressions.
8202 @value{GDBN} has no way to automatically recognize which character set
8203 the inferior program uses; you must tell it, using the @code{set
8204 target-charset} command, described below.
8206 Here are the commands for controlling @value{GDBN}'s character set
8210 @item set target-charset @var{charset}
8211 @kindex set target-charset
8212 Set the current target character set to @var{charset}. To display the
8213 list of supported target character sets, type
8214 @kbd{@w{set target-charset @key{TAB}@key{TAB}}}.
8216 @item set host-charset @var{charset}
8217 @kindex set host-charset
8218 Set the current host character set to @var{charset}.
8220 By default, @value{GDBN} uses a host character set appropriate to the
8221 system it is running on; you can override that default using the
8222 @code{set host-charset} command. On some systems, @value{GDBN} cannot
8223 automatically determine the appropriate host character set. In this
8224 case, @value{GDBN} uses @samp{UTF-8}.
8226 @value{GDBN} can only use certain character sets as its host character
8227 set. If you type @kbd{@w{set target-charset @key{TAB}@key{TAB}}},
8228 @value{GDBN} will list the host character sets it supports.
8230 @item set charset @var{charset}
8232 Set the current host and target character sets to @var{charset}. As
8233 above, if you type @kbd{@w{set charset @key{TAB}@key{TAB}}},
8234 @value{GDBN} will list the names of the character sets that can be used
8235 for both host and target.
8238 @kindex show charset
8239 Show the names of the current host and target character sets.
8241 @item show host-charset
8242 @kindex show host-charset
8243 Show the name of the current host character set.
8245 @item show target-charset
8246 @kindex show target-charset
8247 Show the name of the current target character set.
8249 @item set target-wide-charset @var{charset}
8250 @kindex set target-wide-charset
8251 Set the current target's wide character set to @var{charset}. This is
8252 the character set used by the target's @code{wchar_t} type. To
8253 display the list of supported wide character sets, type
8254 @kbd{@w{set target-wide-charset @key{TAB}@key{TAB}}}.
8256 @item show target-wide-charset
8257 @kindex show target-wide-charset
8258 Show the name of the current target's wide character set.
8261 Here is an example of @value{GDBN}'s character set support in action.
8262 Assume that the following source code has been placed in the file
8263 @file{charset-test.c}:
8269 = @{72, 101, 108, 108, 111, 44, 32, 119,
8270 111, 114, 108, 100, 33, 10, 0@};
8271 char ibm1047_hello[]
8272 = @{200, 133, 147, 147, 150, 107, 64, 166,
8273 150, 153, 147, 132, 90, 37, 0@};
8277 printf ("Hello, world!\n");
8281 In this program, @code{ascii_hello} and @code{ibm1047_hello} are arrays
8282 containing the string @samp{Hello, world!} followed by a newline,
8283 encoded in the @sc{ascii} and @sc{ibm1047} character sets.
8285 We compile the program, and invoke the debugger on it:
8288 $ gcc -g charset-test.c -o charset-test
8289 $ gdb -nw charset-test
8290 GNU gdb 2001-12-19-cvs
8291 Copyright 2001 Free Software Foundation, Inc.
8296 We can use the @code{show charset} command to see what character sets
8297 @value{GDBN} is currently using to interpret and display characters and
8301 (@value{GDBP}) show charset
8302 The current host and target character set is `ISO-8859-1'.
8306 For the sake of printing this manual, let's use @sc{ascii} as our
8307 initial character set:
8309 (@value{GDBP}) set charset ASCII
8310 (@value{GDBP}) show charset
8311 The current host and target character set is `ASCII'.
8315 Let's assume that @sc{ascii} is indeed the correct character set for our
8316 host system --- in other words, let's assume that if @value{GDBN} prints
8317 characters using the @sc{ascii} character set, our terminal will display
8318 them properly. Since our current target character set is also
8319 @sc{ascii}, the contents of @code{ascii_hello} print legibly:
8322 (@value{GDBP}) print ascii_hello
8323 $1 = 0x401698 "Hello, world!\n"
8324 (@value{GDBP}) print ascii_hello[0]
8329 @value{GDBN} uses the target character set for character and string
8330 literals you use in expressions:
8333 (@value{GDBP}) print '+'
8338 The @sc{ascii} character set uses the number 43 to encode the @samp{+}
8341 @value{GDBN} relies on the user to tell it which character set the
8342 target program uses. If we print @code{ibm1047_hello} while our target
8343 character set is still @sc{ascii}, we get jibberish:
8346 (@value{GDBP}) print ibm1047_hello
8347 $4 = 0x4016a8 "\310\205\223\223\226k@@\246\226\231\223\204Z%"
8348 (@value{GDBP}) print ibm1047_hello[0]
8353 If we invoke the @code{set target-charset} followed by @key{TAB}@key{TAB},
8354 @value{GDBN} tells us the character sets it supports:
8357 (@value{GDBP}) set target-charset
8358 ASCII EBCDIC-US IBM1047 ISO-8859-1
8359 (@value{GDBP}) set target-charset
8362 We can select @sc{ibm1047} as our target character set, and examine the
8363 program's strings again. Now the @sc{ascii} string is wrong, but
8364 @value{GDBN} translates the contents of @code{ibm1047_hello} from the
8365 target character set, @sc{ibm1047}, to the host character set,
8366 @sc{ascii}, and they display correctly:
8369 (@value{GDBP}) set target-charset IBM1047
8370 (@value{GDBP}) show charset
8371 The current host character set is `ASCII'.
8372 The current target character set is `IBM1047'.
8373 (@value{GDBP}) print ascii_hello
8374 $6 = 0x401698 "\110\145%%?\054\040\167?\162%\144\041\012"
8375 (@value{GDBP}) print ascii_hello[0]
8377 (@value{GDBP}) print ibm1047_hello
8378 $8 = 0x4016a8 "Hello, world!\n"
8379 (@value{GDBP}) print ibm1047_hello[0]
8384 As above, @value{GDBN} uses the target character set for character and
8385 string literals you use in expressions:
8388 (@value{GDBP}) print '+'
8393 The @sc{ibm1047} character set uses the number 78 to encode the @samp{+}
8396 @node Caching Remote Data
8397 @section Caching Data of Remote Targets
8398 @cindex caching data of remote targets
8400 @value{GDBN} can cache data exchanged between the debugger and a
8401 remote target (@pxref{Remote Debugging}). Such caching generally improves
8402 performance, because it reduces the overhead of the remote protocol by
8403 bundling memory reads and writes into large chunks. Unfortunately,
8404 @value{GDBN} does not currently know anything about volatile
8405 registers, and thus data caching will produce incorrect results when
8406 volatile registers are in use.
8409 @kindex set remotecache
8410 @item set remotecache on
8411 @itemx set remotecache off
8412 Set caching state for remote targets. When @code{ON}, use data
8413 caching. By default, this option is @code{OFF}.
8415 @kindex show remotecache
8416 @item show remotecache
8417 Show the current state of data caching for remote targets.
8421 Print the information about the data cache performance. The
8422 information displayed includes: the dcache width and depth; and for
8423 each cache line, how many times it was referenced, and its data and
8424 state (invalid, dirty, valid). This command is useful for debugging
8425 the data cache operation.
8428 @node Searching Memory
8429 @section Search Memory
8430 @cindex searching memory
8432 Memory can be searched for a particular sequence of bytes with the
8433 @code{find} command.
8437 @item find @r{[}/@var{sn}@r{]} @var{start_addr}, +@var{len}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
8438 @itemx find @r{[}/@var{sn}@r{]} @var{start_addr}, @var{end_addr}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
8439 Search memory for the sequence of bytes specified by @var{val1}, @var{val2},
8440 etc. The search begins at address @var{start_addr} and continues for either
8441 @var{len} bytes or through to @var{end_addr} inclusive.
8444 @var{s} and @var{n} are optional parameters.
8445 They may be specified in either order, apart or together.
8448 @item @var{s}, search query size
8449 The size of each search query value.
8455 halfwords (two bytes)
8459 giant words (eight bytes)
8462 All values are interpreted in the current language.
8463 This means, for example, that if the current source language is C/C@t{++}
8464 then searching for the string ``hello'' includes the trailing '\0'.
8466 If the value size is not specified, it is taken from the
8467 value's type in the current language.
8468 This is useful when one wants to specify the search
8469 pattern as a mixture of types.
8470 Note that this means, for example, that in the case of C-like languages
8471 a search for an untyped 0x42 will search for @samp{(int) 0x42}
8472 which is typically four bytes.
8474 @item @var{n}, maximum number of finds
8475 The maximum number of matches to print. The default is to print all finds.
8478 You can use strings as search values. Quote them with double-quotes
8480 The string value is copied into the search pattern byte by byte,
8481 regardless of the endianness of the target and the size specification.
8483 The address of each match found is printed as well as a count of the
8484 number of matches found.
8486 The address of the last value found is stored in convenience variable
8488 A count of the number of matches is stored in @samp{$numfound}.
8490 For example, if stopped at the @code{printf} in this function:
8496 static char hello[] = "hello-hello";
8497 static struct @{ char c; short s; int i; @}
8498 __attribute__ ((packed)) mixed
8499 = @{ 'c', 0x1234, 0x87654321 @};
8500 printf ("%s\n", hello);
8505 you get during debugging:
8508 (gdb) find &hello[0], +sizeof(hello), "hello"
8509 0x804956d <hello.1620+6>
8511 (gdb) find &hello[0], +sizeof(hello), 'h', 'e', 'l', 'l', 'o'
8512 0x8049567 <hello.1620>
8513 0x804956d <hello.1620+6>
8515 (gdb) find /b1 &hello[0], +sizeof(hello), 'h', 0x65, 'l'
8516 0x8049567 <hello.1620>
8518 (gdb) find &mixed, +sizeof(mixed), (char) 'c', (short) 0x1234, (int) 0x87654321
8519 0x8049560 <mixed.1625>
8521 (gdb) print $numfound
8524 $2 = (void *) 0x8049560
8527 @node Optimized Code
8528 @chapter Debugging Optimized Code
8529 @cindex optimized code, debugging
8530 @cindex debugging optimized code
8532 Almost all compilers support optimization. With optimization
8533 disabled, the compiler generates assembly code that corresponds
8534 directly to your source code, in a simplistic way. As the compiler
8535 applies more powerful optimizations, the generated assembly code
8536 diverges from your original source code. With help from debugging
8537 information generated by the compiler, @value{GDBN} can map from
8538 the running program back to constructs from your original source.
8540 @value{GDBN} is more accurate with optimization disabled. If you
8541 can recompile without optimization, it is easier to follow the
8542 progress of your program during debugging. But, there are many cases
8543 where you may need to debug an optimized version.
8545 When you debug a program compiled with @samp{-g -O}, remember that the
8546 optimizer has rearranged your code; the debugger shows you what is
8547 really there. Do not be too surprised when the execution path does not
8548 exactly match your source file! An extreme example: if you define a
8549 variable, but never use it, @value{GDBN} never sees that
8550 variable---because the compiler optimizes it out of existence.
8552 Some things do not work as well with @samp{-g -O} as with just
8553 @samp{-g}, particularly on machines with instruction scheduling. If in
8554 doubt, recompile with @samp{-g} alone, and if this fixes the problem,
8555 please report it to us as a bug (including a test case!).
8556 @xref{Variables}, for more information about debugging optimized code.
8559 * Inline Functions:: How @value{GDBN} presents inlining
8562 @node Inline Functions
8563 @section Inline Functions
8564 @cindex inline functions, debugging
8566 @dfn{Inlining} is an optimization that inserts a copy of the function
8567 body directly at each call site, instead of jumping to a shared
8568 routine. @value{GDBN} displays inlined functions just like
8569 non-inlined functions. They appear in backtraces. You can view their
8570 arguments and local variables, step into them with @code{step}, skip
8571 them with @code{next}, and escape from them with @code{finish}.
8572 You can check whether a function was inlined by using the
8573 @code{info frame} command.
8575 For @value{GDBN} to support inlined functions, the compiler must
8576 record information about inlining in the debug information ---
8577 @value{NGCC} using the @sc{dwarf 2} format does this, and several
8578 other compilers do also. @value{GDBN} only supports inlined functions
8579 when using @sc{dwarf 2}. Versions of @value{NGCC} before 4.1
8580 do not emit two required attributes (@samp{DW_AT_call_file} and
8581 @samp{DW_AT_call_line}); @value{GDBN} does not display inlined
8582 function calls with earlier versions of @value{NGCC}. It instead
8583 displays the arguments and local variables of inlined functions as
8584 local variables in the caller.
8586 The body of an inlined function is directly included at its call site;
8587 unlike a non-inlined function, there are no instructions devoted to
8588 the call. @value{GDBN} still pretends that the call site and the
8589 start of the inlined function are different instructions. Stepping to
8590 the call site shows the call site, and then stepping again shows
8591 the first line of the inlined function, even though no additional
8592 instructions are executed.
8594 This makes source-level debugging much clearer; you can see both the
8595 context of the call and then the effect of the call. Only stepping by
8596 a single instruction using @code{stepi} or @code{nexti} does not do
8597 this; single instruction steps always show the inlined body.
8599 There are some ways that @value{GDBN} does not pretend that inlined
8600 function calls are the same as normal calls:
8604 You cannot set breakpoints on inlined functions. @value{GDBN}
8605 either reports that there is no symbol with that name, or else sets the
8606 breakpoint only on non-inlined copies of the function. This limitation
8607 will be removed in a future version of @value{GDBN}; until then,
8608 set a breakpoint by line number on the first line of the inlined
8612 Setting breakpoints at the call site of an inlined function may not
8613 work, because the call site does not contain any code. @value{GDBN}
8614 may incorrectly move the breakpoint to the next line of the enclosing
8615 function, after the call. This limitation will be removed in a future
8616 version of @value{GDBN}; until then, set a breakpoint on an earlier line
8617 or inside the inlined function instead.
8620 @value{GDBN} cannot locate the return value of inlined calls after
8621 using the @code{finish} command. This is a limitation of compiler-generated
8622 debugging information; after @code{finish}, you can step to the next line
8623 and print a variable where your program stored the return value.
8629 @chapter C Preprocessor Macros
8631 Some languages, such as C and C@t{++}, provide a way to define and invoke
8632 ``preprocessor macros'' which expand into strings of tokens.
8633 @value{GDBN} can evaluate expressions containing macro invocations, show
8634 the result of macro expansion, and show a macro's definition, including
8635 where it was defined.
8637 You may need to compile your program specially to provide @value{GDBN}
8638 with information about preprocessor macros. Most compilers do not
8639 include macros in their debugging information, even when you compile
8640 with the @option{-g} flag. @xref{Compilation}.
8642 A program may define a macro at one point, remove that definition later,
8643 and then provide a different definition after that. Thus, at different
8644 points in the program, a macro may have different definitions, or have
8645 no definition at all. If there is a current stack frame, @value{GDBN}
8646 uses the macros in scope at that frame's source code line. Otherwise,
8647 @value{GDBN} uses the macros in scope at the current listing location;
8650 Whenever @value{GDBN} evaluates an expression, it always expands any
8651 macro invocations present in the expression. @value{GDBN} also provides
8652 the following commands for working with macros explicitly.
8656 @kindex macro expand
8657 @cindex macro expansion, showing the results of preprocessor
8658 @cindex preprocessor macro expansion, showing the results of
8659 @cindex expanding preprocessor macros
8660 @item macro expand @var{expression}
8661 @itemx macro exp @var{expression}
8662 Show the results of expanding all preprocessor macro invocations in
8663 @var{expression}. Since @value{GDBN} simply expands macros, but does
8664 not parse the result, @var{expression} need not be a valid expression;
8665 it can be any string of tokens.
8668 @item macro expand-once @var{expression}
8669 @itemx macro exp1 @var{expression}
8670 @cindex expand macro once
8671 @i{(This command is not yet implemented.)} Show the results of
8672 expanding those preprocessor macro invocations that appear explicitly in
8673 @var{expression}. Macro invocations appearing in that expansion are
8674 left unchanged. This command allows you to see the effect of a
8675 particular macro more clearly, without being confused by further
8676 expansions. Since @value{GDBN} simply expands macros, but does not
8677 parse the result, @var{expression} need not be a valid expression; it
8678 can be any string of tokens.
8681 @cindex macro definition, showing
8682 @cindex definition, showing a macro's
8683 @item info macro @var{macro}
8684 Show the definition of the macro named @var{macro}, and describe the
8685 source location or compiler command-line where that definition was established.
8687 @kindex macro define
8688 @cindex user-defined macros
8689 @cindex defining macros interactively
8690 @cindex macros, user-defined
8691 @item macro define @var{macro} @var{replacement-list}
8692 @itemx macro define @var{macro}(@var{arglist}) @var{replacement-list}
8693 Introduce a definition for a preprocessor macro named @var{macro},
8694 invocations of which are replaced by the tokens given in
8695 @var{replacement-list}. The first form of this command defines an
8696 ``object-like'' macro, which takes no arguments; the second form
8697 defines a ``function-like'' macro, which takes the arguments given in
8700 A definition introduced by this command is in scope in every
8701 expression evaluated in @value{GDBN}, until it is removed with the
8702 @code{macro undef} command, described below. The definition overrides
8703 all definitions for @var{macro} present in the program being debugged,
8704 as well as any previous user-supplied definition.
8707 @item macro undef @var{macro}
8708 Remove any user-supplied definition for the macro named @var{macro}.
8709 This command only affects definitions provided with the @code{macro
8710 define} command, described above; it cannot remove definitions present
8711 in the program being debugged.
8715 List all the macros defined using the @code{macro define} command.
8718 @cindex macros, example of debugging with
8719 Here is a transcript showing the above commands in action. First, we
8720 show our source files:
8728 #define ADD(x) (M + x)
8733 printf ("Hello, world!\n");
8735 printf ("We're so creative.\n");
8737 printf ("Goodbye, world!\n");
8744 Now, we compile the program using the @sc{gnu} C compiler, @value{NGCC}.
8745 We pass the @option{-gdwarf-2} and @option{-g3} flags to ensure the
8746 compiler includes information about preprocessor macros in the debugging
8750 $ gcc -gdwarf-2 -g3 sample.c -o sample
8754 Now, we start @value{GDBN} on our sample program:
8758 GNU gdb 2002-05-06-cvs
8759 Copyright 2002 Free Software Foundation, Inc.
8760 GDB is free software, @dots{}
8764 We can expand macros and examine their definitions, even when the
8765 program is not running. @value{GDBN} uses the current listing position
8766 to decide which macro definitions are in scope:
8769 (@value{GDBP}) list main
8772 5 #define ADD(x) (M + x)
8777 10 printf ("Hello, world!\n");
8779 12 printf ("We're so creative.\n");
8780 (@value{GDBP}) info macro ADD
8781 Defined at /home/jimb/gdb/macros/play/sample.c:5
8782 #define ADD(x) (M + x)
8783 (@value{GDBP}) info macro Q
8784 Defined at /home/jimb/gdb/macros/play/sample.h:1
8785 included at /home/jimb/gdb/macros/play/sample.c:2
8787 (@value{GDBP}) macro expand ADD(1)
8788 expands to: (42 + 1)
8789 (@value{GDBP}) macro expand-once ADD(1)
8790 expands to: once (M + 1)
8794 In the example above, note that @code{macro expand-once} expands only
8795 the macro invocation explicit in the original text --- the invocation of
8796 @code{ADD} --- but does not expand the invocation of the macro @code{M},
8797 which was introduced by @code{ADD}.
8799 Once the program is running, @value{GDBN} uses the macro definitions in
8800 force at the source line of the current stack frame:
8803 (@value{GDBP}) break main
8804 Breakpoint 1 at 0x8048370: file sample.c, line 10.
8806 Starting program: /home/jimb/gdb/macros/play/sample
8808 Breakpoint 1, main () at sample.c:10
8809 10 printf ("Hello, world!\n");
8813 At line 10, the definition of the macro @code{N} at line 9 is in force:
8816 (@value{GDBP}) info macro N
8817 Defined at /home/jimb/gdb/macros/play/sample.c:9
8819 (@value{GDBP}) macro expand N Q M
8821 (@value{GDBP}) print N Q M
8826 As we step over directives that remove @code{N}'s definition, and then
8827 give it a new definition, @value{GDBN} finds the definition (or lack
8828 thereof) in force at each point:
8833 12 printf ("We're so creative.\n");
8834 (@value{GDBP}) info macro N
8835 The symbol `N' has no definition as a C/C++ preprocessor macro
8836 at /home/jimb/gdb/macros/play/sample.c:12
8839 14 printf ("Goodbye, world!\n");
8840 (@value{GDBP}) info macro N
8841 Defined at /home/jimb/gdb/macros/play/sample.c:13
8843 (@value{GDBP}) macro expand N Q M
8844 expands to: 1729 < 42
8845 (@value{GDBP}) print N Q M
8850 In addition to source files, macros can be defined on the compilation command
8851 line using the @option{-D@var{name}=@var{value}} syntax. For macros defined in
8852 such a way, @value{GDBN} displays the location of their definition as line zero
8853 of the source file submitted to the compiler.
8856 (@value{GDBP}) info macro __STDC__
8857 Defined at /home/jimb/gdb/macros/play/sample.c:0
8864 @chapter Tracepoints
8865 @c This chapter is based on the documentation written by Michael
8866 @c Snyder, David Taylor, Jim Blandy, and Elena Zannoni.
8869 In some applications, it is not feasible for the debugger to interrupt
8870 the program's execution long enough for the developer to learn
8871 anything helpful about its behavior. If the program's correctness
8872 depends on its real-time behavior, delays introduced by a debugger
8873 might cause the program to change its behavior drastically, or perhaps
8874 fail, even when the code itself is correct. It is useful to be able
8875 to observe the program's behavior without interrupting it.
8877 Using @value{GDBN}'s @code{trace} and @code{collect} commands, you can
8878 specify locations in the program, called @dfn{tracepoints}, and
8879 arbitrary expressions to evaluate when those tracepoints are reached.
8880 Later, using the @code{tfind} command, you can examine the values
8881 those expressions had when the program hit the tracepoints. The
8882 expressions may also denote objects in memory---structures or arrays,
8883 for example---whose values @value{GDBN} should record; while visiting
8884 a particular tracepoint, you may inspect those objects as if they were
8885 in memory at that moment. However, because @value{GDBN} records these
8886 values without interacting with you, it can do so quickly and
8887 unobtrusively, hopefully not disturbing the program's behavior.
8889 The tracepoint facility is currently available only for remote
8890 targets. @xref{Targets}. In addition, your remote target must know
8891 how to collect trace data. This functionality is implemented in the
8892 remote stub; however, none of the stubs distributed with @value{GDBN}
8893 support tracepoints as of this writing. The format of the remote
8894 packets used to implement tracepoints are described in @ref{Tracepoint
8897 This chapter describes the tracepoint commands and features.
8901 * Analyze Collected Data::
8902 * Tracepoint Variables::
8905 @node Set Tracepoints
8906 @section Commands to Set Tracepoints
8908 Before running such a @dfn{trace experiment}, an arbitrary number of
8909 tracepoints can be set. A tracepoint is actually a special type of
8910 breakpoint (@pxref{Set Breaks}), so you can manipulate it using
8911 standard breakpoint commands. For instance, as with breakpoints,
8912 tracepoint numbers are successive integers starting from one, and many
8913 of the commands associated with tracepoints take the tracepoint number
8914 as their argument, to identify which tracepoint to work on.
8916 For each tracepoint, you can specify, in advance, some arbitrary set
8917 of data that you want the target to collect in the trace buffer when
8918 it hits that tracepoint. The collected data can include registers,
8919 local variables, or global data. Later, you can use @value{GDBN}
8920 commands to examine the values these data had at the time the
8923 Tracepoints do not support every breakpoint feature. Conditional
8924 expressions and ignore counts on tracepoints have no effect, and
8925 tracepoints cannot run @value{GDBN} commands when they are
8926 hit. Tracepoints may not be thread-specific either.
8928 This section describes commands to set tracepoints and associated
8929 conditions and actions.
8932 * Create and Delete Tracepoints::
8933 * Enable and Disable Tracepoints::
8934 * Tracepoint Passcounts::
8935 * Tracepoint Actions::
8936 * Listing Tracepoints::
8937 * Starting and Stopping Trace Experiments::
8940 @node Create and Delete Tracepoints
8941 @subsection Create and Delete Tracepoints
8944 @cindex set tracepoint
8946 @item trace @var{location}
8947 The @code{trace} command is very similar to the @code{break} command.
8948 Its argument @var{location} can be a source line, a function name, or
8949 an address in the target program. @xref{Specify Location}. The
8950 @code{trace} command defines a tracepoint, which is a point in the
8951 target program where the debugger will briefly stop, collect some
8952 data, and then allow the program to continue. Setting a tracepoint or
8953 changing its actions doesn't take effect until the next @code{tstart}
8954 command, and once a trace experiment is running, further changes will
8955 not have any effect until the next trace experiment starts.
8957 Here are some examples of using the @code{trace} command:
8960 (@value{GDBP}) @b{trace foo.c:121} // a source file and line number
8962 (@value{GDBP}) @b{trace +2} // 2 lines forward
8964 (@value{GDBP}) @b{trace my_function} // first source line of function
8966 (@value{GDBP}) @b{trace *my_function} // EXACT start address of function
8968 (@value{GDBP}) @b{trace *0x2117c4} // an address
8972 You can abbreviate @code{trace} as @code{tr}.
8975 @cindex last tracepoint number
8976 @cindex recent tracepoint number
8977 @cindex tracepoint number
8978 The convenience variable @code{$tpnum} records the tracepoint number
8979 of the most recently set tracepoint.
8981 @kindex delete tracepoint
8982 @cindex tracepoint deletion
8983 @item delete tracepoint @r{[}@var{num}@r{]}
8984 Permanently delete one or more tracepoints. With no argument, the
8985 default is to delete all tracepoints. Note that the regular
8986 @code{delete} command can remove tracepoints also.
8991 (@value{GDBP}) @b{delete trace 1 2 3} // remove three tracepoints
8993 (@value{GDBP}) @b{delete trace} // remove all tracepoints
8997 You can abbreviate this command as @code{del tr}.
9000 @node Enable and Disable Tracepoints
9001 @subsection Enable and Disable Tracepoints
9003 These commands are deprecated; they are equivalent to plain @code{disable} and @code{enable}.
9006 @kindex disable tracepoint
9007 @item disable tracepoint @r{[}@var{num}@r{]}
9008 Disable tracepoint @var{num}, or all tracepoints if no argument
9009 @var{num} is given. A disabled tracepoint will have no effect during
9010 the next trace experiment, but it is not forgotten. You can re-enable
9011 a disabled tracepoint using the @code{enable tracepoint} command.
9013 @kindex enable tracepoint
9014 @item enable tracepoint @r{[}@var{num}@r{]}
9015 Enable tracepoint @var{num}, or all tracepoints. The enabled
9016 tracepoints will become effective the next time a trace experiment is
9020 @node Tracepoint Passcounts
9021 @subsection Tracepoint Passcounts
9025 @cindex tracepoint pass count
9026 @item passcount @r{[}@var{n} @r{[}@var{num}@r{]]}
9027 Set the @dfn{passcount} of a tracepoint. The passcount is a way to
9028 automatically stop a trace experiment. If a tracepoint's passcount is
9029 @var{n}, then the trace experiment will be automatically stopped on
9030 the @var{n}'th time that tracepoint is hit. If the tracepoint number
9031 @var{num} is not specified, the @code{passcount} command sets the
9032 passcount of the most recently defined tracepoint. If no passcount is
9033 given, the trace experiment will run until stopped explicitly by the
9039 (@value{GDBP}) @b{passcount 5 2} // Stop on the 5th execution of
9040 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// tracepoint 2}
9042 (@value{GDBP}) @b{passcount 12} // Stop on the 12th execution of the
9043 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// most recently defined tracepoint.}
9044 (@value{GDBP}) @b{trace foo}
9045 (@value{GDBP}) @b{pass 3}
9046 (@value{GDBP}) @b{trace bar}
9047 (@value{GDBP}) @b{pass 2}
9048 (@value{GDBP}) @b{trace baz}
9049 (@value{GDBP}) @b{pass 1} // Stop tracing when foo has been
9050 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// executed 3 times OR when bar has}
9051 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// been executed 2 times}
9052 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// OR when baz has been executed 1 time.}
9056 @node Tracepoint Actions
9057 @subsection Tracepoint Action Lists
9061 @cindex tracepoint actions
9062 @item actions @r{[}@var{num}@r{]}
9063 This command will prompt for a list of actions to be taken when the
9064 tracepoint is hit. If the tracepoint number @var{num} is not
9065 specified, this command sets the actions for the one that was most
9066 recently defined (so that you can define a tracepoint and then say
9067 @code{actions} without bothering about its number). You specify the
9068 actions themselves on the following lines, one action at a time, and
9069 terminate the actions list with a line containing just @code{end}. So
9070 far, the only defined actions are @code{collect} and
9071 @code{while-stepping}.
9073 @cindex remove actions from a tracepoint
9074 To remove all actions from a tracepoint, type @samp{actions @var{num}}
9075 and follow it immediately with @samp{end}.
9078 (@value{GDBP}) @b{collect @var{data}} // collect some data
9080 (@value{GDBP}) @b{while-stepping 5} // single-step 5 times, collect data
9082 (@value{GDBP}) @b{end} // signals the end of actions.
9085 In the following example, the action list begins with @code{collect}
9086 commands indicating the things to be collected when the tracepoint is
9087 hit. Then, in order to single-step and collect additional data
9088 following the tracepoint, a @code{while-stepping} command is used,
9089 followed by the list of things to be collected while stepping. The
9090 @code{while-stepping} command is terminated by its own separate
9091 @code{end} command. Lastly, the action list is terminated by an
9095 (@value{GDBP}) @b{trace foo}
9096 (@value{GDBP}) @b{actions}
9097 Enter actions for tracepoint 1, one per line:
9106 @kindex collect @r{(tracepoints)}
9107 @item collect @var{expr1}, @var{expr2}, @dots{}
9108 Collect values of the given expressions when the tracepoint is hit.
9109 This command accepts a comma-separated list of any valid expressions.
9110 In addition to global, static, or local variables, the following
9111 special arguments are supported:
9115 collect all registers
9118 collect all function arguments
9121 collect all local variables.
9124 You can give several consecutive @code{collect} commands, each one
9125 with a single argument, or one @code{collect} command with several
9126 arguments separated by commas: the effect is the same.
9128 The command @code{info scope} (@pxref{Symbols, info scope}) is
9129 particularly useful for figuring out what data to collect.
9131 @kindex while-stepping @r{(tracepoints)}
9132 @item while-stepping @var{n}
9133 Perform @var{n} single-step traces after the tracepoint, collecting
9134 new data at each step. The @code{while-stepping} command is
9135 followed by the list of what to collect while stepping (followed by
9136 its own @code{end} command):
9140 > collect $regs, myglobal
9146 You may abbreviate @code{while-stepping} as @code{ws} or
9150 @node Listing Tracepoints
9151 @subsection Listing Tracepoints
9154 @kindex info tracepoints
9156 @cindex information about tracepoints
9157 @item info tracepoints @r{[}@var{num}@r{]}
9158 Display information about the tracepoint @var{num}. If you don't
9159 specify a tracepoint number, displays information about all the
9160 tracepoints defined so far. The format is similar to that used for
9161 @code{info breakpoints}; in fact, @code{info tracepoints} is the same
9162 command, simply restricting itself to tracepoints.
9164 A tracepoint's listing may include additional information specific to
9169 its passcount as given by the @code{passcount @var{n}} command
9171 its step count as given by the @code{while-stepping @var{n}} command
9173 its action list as given by the @code{actions} command. The actions
9174 are prefixed with an @samp{A} so as to distinguish them from commands.
9178 (@value{GDBP}) @b{info trace}
9179 Num Type Disp Enb Address What
9180 1 tracepoint keep y 0x0804ab57 in foo() at main.cxx:7
9184 A collect globfoo, $regs
9192 This command can be abbreviated @code{info tp}.
9195 @node Starting and Stopping Trace Experiments
9196 @subsection Starting and Stopping Trace Experiments
9200 @cindex start a new trace experiment
9201 @cindex collected data discarded
9203 This command takes no arguments. It starts the trace experiment, and
9204 begins collecting data. This has the side effect of discarding all
9205 the data collected in the trace buffer during the previous trace
9209 @cindex stop a running trace experiment
9211 This command takes no arguments. It ends the trace experiment, and
9212 stops collecting data.
9214 @strong{Note}: a trace experiment and data collection may stop
9215 automatically if any tracepoint's passcount is reached
9216 (@pxref{Tracepoint Passcounts}), or if the trace buffer becomes full.
9219 @cindex status of trace data collection
9220 @cindex trace experiment, status of
9222 This command displays the status of the current trace data
9226 Here is an example of the commands we described so far:
9229 (@value{GDBP}) @b{trace gdb_c_test}
9230 (@value{GDBP}) @b{actions}
9231 Enter actions for tracepoint #1, one per line.
9232 > collect $regs,$locals,$args
9237 (@value{GDBP}) @b{tstart}
9238 [time passes @dots{}]
9239 (@value{GDBP}) @b{tstop}
9243 @node Analyze Collected Data
9244 @section Using the Collected Data
9246 After the tracepoint experiment ends, you use @value{GDBN} commands
9247 for examining the trace data. The basic idea is that each tracepoint
9248 collects a trace @dfn{snapshot} every time it is hit and another
9249 snapshot every time it single-steps. All these snapshots are
9250 consecutively numbered from zero and go into a buffer, and you can
9251 examine them later. The way you examine them is to @dfn{focus} on a
9252 specific trace snapshot. When the remote stub is focused on a trace
9253 snapshot, it will respond to all @value{GDBN} requests for memory and
9254 registers by reading from the buffer which belongs to that snapshot,
9255 rather than from @emph{real} memory or registers of the program being
9256 debugged. This means that @strong{all} @value{GDBN} commands
9257 (@code{print}, @code{info registers}, @code{backtrace}, etc.) will
9258 behave as if we were currently debugging the program state as it was
9259 when the tracepoint occurred. Any requests for data that are not in
9260 the buffer will fail.
9263 * tfind:: How to select a trace snapshot
9264 * tdump:: How to display all data for a snapshot
9265 * save-tracepoints:: How to save tracepoints for a future run
9269 @subsection @code{tfind @var{n}}
9272 @cindex select trace snapshot
9273 @cindex find trace snapshot
9274 The basic command for selecting a trace snapshot from the buffer is
9275 @code{tfind @var{n}}, which finds trace snapshot number @var{n},
9276 counting from zero. If no argument @var{n} is given, the next
9277 snapshot is selected.
9279 Here are the various forms of using the @code{tfind} command.
9283 Find the first snapshot in the buffer. This is a synonym for
9284 @code{tfind 0} (since 0 is the number of the first snapshot).
9287 Stop debugging trace snapshots, resume @emph{live} debugging.
9290 Same as @samp{tfind none}.
9293 No argument means find the next trace snapshot.
9296 Find the previous trace snapshot before the current one. This permits
9297 retracing earlier steps.
9299 @item tfind tracepoint @var{num}
9300 Find the next snapshot associated with tracepoint @var{num}. Search
9301 proceeds forward from the last examined trace snapshot. If no
9302 argument @var{num} is given, it means find the next snapshot collected
9303 for the same tracepoint as the current snapshot.
9305 @item tfind pc @var{addr}
9306 Find the next snapshot associated with the value @var{addr} of the
9307 program counter. Search proceeds forward from the last examined trace
9308 snapshot. If no argument @var{addr} is given, it means find the next
9309 snapshot with the same value of PC as the current snapshot.
9311 @item tfind outside @var{addr1}, @var{addr2}
9312 Find the next snapshot whose PC is outside the given range of
9315 @item tfind range @var{addr1}, @var{addr2}
9316 Find the next snapshot whose PC is between @var{addr1} and
9317 @var{addr2}. @c FIXME: Is the range inclusive or exclusive?
9319 @item tfind line @r{[}@var{file}:@r{]}@var{n}
9320 Find the next snapshot associated with the source line @var{n}. If
9321 the optional argument @var{file} is given, refer to line @var{n} in
9322 that source file. Search proceeds forward from the last examined
9323 trace snapshot. If no argument @var{n} is given, it means find the
9324 next line other than the one currently being examined; thus saying
9325 @code{tfind line} repeatedly can appear to have the same effect as
9326 stepping from line to line in a @emph{live} debugging session.
9329 The default arguments for the @code{tfind} commands are specifically
9330 designed to make it easy to scan through the trace buffer. For
9331 instance, @code{tfind} with no argument selects the next trace
9332 snapshot, and @code{tfind -} with no argument selects the previous
9333 trace snapshot. So, by giving one @code{tfind} command, and then
9334 simply hitting @key{RET} repeatedly you can examine all the trace
9335 snapshots in order. Or, by saying @code{tfind -} and then hitting
9336 @key{RET} repeatedly you can examine the snapshots in reverse order.
9337 The @code{tfind line} command with no argument selects the snapshot
9338 for the next source line executed. The @code{tfind pc} command with
9339 no argument selects the next snapshot with the same program counter
9340 (PC) as the current frame. The @code{tfind tracepoint} command with
9341 no argument selects the next trace snapshot collected by the same
9342 tracepoint as the current one.
9344 In addition to letting you scan through the trace buffer manually,
9345 these commands make it easy to construct @value{GDBN} scripts that
9346 scan through the trace buffer and print out whatever collected data
9347 you are interested in. Thus, if we want to examine the PC, FP, and SP
9348 registers from each trace frame in the buffer, we can say this:
9351 (@value{GDBP}) @b{tfind start}
9352 (@value{GDBP}) @b{while ($trace_frame != -1)}
9353 > printf "Frame %d, PC = %08X, SP = %08X, FP = %08X\n", \
9354 $trace_frame, $pc, $sp, $fp
9358 Frame 0, PC = 0020DC64, SP = 0030BF3C, FP = 0030BF44
9359 Frame 1, PC = 0020DC6C, SP = 0030BF38, FP = 0030BF44
9360 Frame 2, PC = 0020DC70, SP = 0030BF34, FP = 0030BF44
9361 Frame 3, PC = 0020DC74, SP = 0030BF30, FP = 0030BF44
9362 Frame 4, PC = 0020DC78, SP = 0030BF2C, FP = 0030BF44
9363 Frame 5, PC = 0020DC7C, SP = 0030BF28, FP = 0030BF44
9364 Frame 6, PC = 0020DC80, SP = 0030BF24, FP = 0030BF44
9365 Frame 7, PC = 0020DC84, SP = 0030BF20, FP = 0030BF44
9366 Frame 8, PC = 0020DC88, SP = 0030BF1C, FP = 0030BF44
9367 Frame 9, PC = 0020DC8E, SP = 0030BF18, FP = 0030BF44
9368 Frame 10, PC = 00203F6C, SP = 0030BE3C, FP = 0030BF14
9371 Or, if we want to examine the variable @code{X} at each source line in
9375 (@value{GDBP}) @b{tfind start}
9376 (@value{GDBP}) @b{while ($trace_frame != -1)}
9377 > printf "Frame %d, X == %d\n", $trace_frame, X
9387 @subsection @code{tdump}
9389 @cindex dump all data collected at tracepoint
9390 @cindex tracepoint data, display
9392 This command takes no arguments. It prints all the data collected at
9393 the current trace snapshot.
9396 (@value{GDBP}) @b{trace 444}
9397 (@value{GDBP}) @b{actions}
9398 Enter actions for tracepoint #2, one per line:
9399 > collect $regs, $locals, $args, gdb_long_test
9402 (@value{GDBP}) @b{tstart}
9404 (@value{GDBP}) @b{tfind line 444}
9405 #0 gdb_test (p1=0x11, p2=0x22, p3=0x33, p4=0x44, p5=0x55, p6=0x66)
9407 444 printp( "%s: arguments = 0x%X 0x%X 0x%X 0x%X 0x%X 0x%X\n", )
9409 (@value{GDBP}) @b{tdump}
9410 Data collected at tracepoint 2, trace frame 1:
9411 d0 0xc4aa0085 -995491707
9415 d4 0x71aea3d 119204413
9420 a1 0x3000668 50333288
9423 a4 0x3000698 50333336
9425 fp 0x30bf3c 0x30bf3c
9426 sp 0x30bf34 0x30bf34
9428 pc 0x20b2c8 0x20b2c8
9432 p = 0x20e5b4 "gdb-test"
9439 gdb_long_test = 17 '\021'
9444 @node save-tracepoints
9445 @subsection @code{save-tracepoints @var{filename}}
9446 @kindex save-tracepoints
9447 @cindex save tracepoints for future sessions
9449 This command saves all current tracepoint definitions together with
9450 their actions and passcounts, into a file @file{@var{filename}}
9451 suitable for use in a later debugging session. To read the saved
9452 tracepoint definitions, use the @code{source} command (@pxref{Command
9455 @node Tracepoint Variables
9456 @section Convenience Variables for Tracepoints
9457 @cindex tracepoint variables
9458 @cindex convenience variables for tracepoints
9461 @vindex $trace_frame
9462 @item (int) $trace_frame
9463 The current trace snapshot (a.k.a.@: @dfn{frame}) number, or -1 if no
9464 snapshot is selected.
9467 @item (int) $tracepoint
9468 The tracepoint for the current trace snapshot.
9471 @item (int) $trace_line
9472 The line number for the current trace snapshot.
9475 @item (char []) $trace_file
9476 The source file for the current trace snapshot.
9479 @item (char []) $trace_func
9480 The name of the function containing @code{$tracepoint}.
9483 Note: @code{$trace_file} is not suitable for use in @code{printf},
9484 use @code{output} instead.
9486 Here's a simple example of using these convenience variables for
9487 stepping through all the trace snapshots and printing some of their
9491 (@value{GDBP}) @b{tfind start}
9493 (@value{GDBP}) @b{while $trace_frame != -1}
9494 > output $trace_file
9495 > printf ", line %d (tracepoint #%d)\n", $trace_line, $tracepoint
9501 @chapter Debugging Programs That Use Overlays
9504 If your program is too large to fit completely in your target system's
9505 memory, you can sometimes use @dfn{overlays} to work around this
9506 problem. @value{GDBN} provides some support for debugging programs that
9510 * How Overlays Work:: A general explanation of overlays.
9511 * Overlay Commands:: Managing overlays in @value{GDBN}.
9512 * Automatic Overlay Debugging:: @value{GDBN} can find out which overlays are
9513 mapped by asking the inferior.
9514 * Overlay Sample Program:: A sample program using overlays.
9517 @node How Overlays Work
9518 @section How Overlays Work
9519 @cindex mapped overlays
9520 @cindex unmapped overlays
9521 @cindex load address, overlay's
9522 @cindex mapped address
9523 @cindex overlay area
9525 Suppose you have a computer whose instruction address space is only 64
9526 kilobytes long, but which has much more memory which can be accessed by
9527 other means: special instructions, segment registers, or memory
9528 management hardware, for example. Suppose further that you want to
9529 adapt a program which is larger than 64 kilobytes to run on this system.
9531 One solution is to identify modules of your program which are relatively
9532 independent, and need not call each other directly; call these modules
9533 @dfn{overlays}. Separate the overlays from the main program, and place
9534 their machine code in the larger memory. Place your main program in
9535 instruction memory, but leave at least enough space there to hold the
9536 largest overlay as well.
9538 Now, to call a function located in an overlay, you must first copy that
9539 overlay's machine code from the large memory into the space set aside
9540 for it in the instruction memory, and then jump to its entry point
9543 @c NB: In the below the mapped area's size is greater or equal to the
9544 @c size of all overlays. This is intentional to remind the developer
9545 @c that overlays don't necessarily need to be the same size.
9549 Data Instruction Larger
9550 Address Space Address Space Address Space
9551 +-----------+ +-----------+ +-----------+
9553 +-----------+ +-----------+ +-----------+<-- overlay 1
9554 | program | | main | .----| overlay 1 | load address
9555 | variables | | program | | +-----------+
9556 | and heap | | | | | |
9557 +-----------+ | | | +-----------+<-- overlay 2
9558 | | +-----------+ | | | load address
9559 +-----------+ | | | .-| overlay 2 |
9561 mapped --->+-----------+ | | +-----------+
9563 | overlay | <-' | | |
9564 | area | <---' +-----------+<-- overlay 3
9565 | | <---. | | load address
9566 +-----------+ `--| overlay 3 |
9573 @anchor{A code overlay}A code overlay
9577 The diagram (@pxref{A code overlay}) shows a system with separate data
9578 and instruction address spaces. To map an overlay, the program copies
9579 its code from the larger address space to the instruction address space.
9580 Since the overlays shown here all use the same mapped address, only one
9581 may be mapped at a time. For a system with a single address space for
9582 data and instructions, the diagram would be similar, except that the
9583 program variables and heap would share an address space with the main
9584 program and the overlay area.
9586 An overlay loaded into instruction memory and ready for use is called a
9587 @dfn{mapped} overlay; its @dfn{mapped address} is its address in the
9588 instruction memory. An overlay not present (or only partially present)
9589 in instruction memory is called @dfn{unmapped}; its @dfn{load address}
9590 is its address in the larger memory. The mapped address is also called
9591 the @dfn{virtual memory address}, or @dfn{VMA}; the load address is also
9592 called the @dfn{load memory address}, or @dfn{LMA}.
9594 Unfortunately, overlays are not a completely transparent way to adapt a
9595 program to limited instruction memory. They introduce a new set of
9596 global constraints you must keep in mind as you design your program:
9601 Before calling or returning to a function in an overlay, your program
9602 must make sure that overlay is actually mapped. Otherwise, the call or
9603 return will transfer control to the right address, but in the wrong
9604 overlay, and your program will probably crash.
9607 If the process of mapping an overlay is expensive on your system, you
9608 will need to choose your overlays carefully to minimize their effect on
9609 your program's performance.
9612 The executable file you load onto your system must contain each
9613 overlay's instructions, appearing at the overlay's load address, not its
9614 mapped address. However, each overlay's instructions must be relocated
9615 and its symbols defined as if the overlay were at its mapped address.
9616 You can use GNU linker scripts to specify different load and relocation
9617 addresses for pieces of your program; see @ref{Overlay Description,,,
9618 ld.info, Using ld: the GNU linker}.
9621 The procedure for loading executable files onto your system must be able
9622 to load their contents into the larger address space as well as the
9623 instruction and data spaces.
9627 The overlay system described above is rather simple, and could be
9628 improved in many ways:
9633 If your system has suitable bank switch registers or memory management
9634 hardware, you could use those facilities to make an overlay's load area
9635 contents simply appear at their mapped address in instruction space.
9636 This would probably be faster than copying the overlay to its mapped
9637 area in the usual way.
9640 If your overlays are small enough, you could set aside more than one
9641 overlay area, and have more than one overlay mapped at a time.
9644 You can use overlays to manage data, as well as instructions. In
9645 general, data overlays are even less transparent to your design than
9646 code overlays: whereas code overlays only require care when you call or
9647 return to functions, data overlays require care every time you access
9648 the data. Also, if you change the contents of a data overlay, you
9649 must copy its contents back out to its load address before you can copy a
9650 different data overlay into the same mapped area.
9655 @node Overlay Commands
9656 @section Overlay Commands
9658 To use @value{GDBN}'s overlay support, each overlay in your program must
9659 correspond to a separate section of the executable file. The section's
9660 virtual memory address and load memory address must be the overlay's
9661 mapped and load addresses. Identifying overlays with sections allows
9662 @value{GDBN} to determine the appropriate address of a function or
9663 variable, depending on whether the overlay is mapped or not.
9665 @value{GDBN}'s overlay commands all start with the word @code{overlay};
9666 you can abbreviate this as @code{ov} or @code{ovly}. The commands are:
9671 Disable @value{GDBN}'s overlay support. When overlay support is
9672 disabled, @value{GDBN} assumes that all functions and variables are
9673 always present at their mapped addresses. By default, @value{GDBN}'s
9674 overlay support is disabled.
9676 @item overlay manual
9677 @cindex manual overlay debugging
9678 Enable @dfn{manual} overlay debugging. In this mode, @value{GDBN}
9679 relies on you to tell it which overlays are mapped, and which are not,
9680 using the @code{overlay map-overlay} and @code{overlay unmap-overlay}
9681 commands described below.
9683 @item overlay map-overlay @var{overlay}
9684 @itemx overlay map @var{overlay}
9685 @cindex map an overlay
9686 Tell @value{GDBN} that @var{overlay} is now mapped; @var{overlay} must
9687 be the name of the object file section containing the overlay. When an
9688 overlay is mapped, @value{GDBN} assumes it can find the overlay's
9689 functions and variables at their mapped addresses. @value{GDBN} assumes
9690 that any other overlays whose mapped ranges overlap that of
9691 @var{overlay} are now unmapped.
9693 @item overlay unmap-overlay @var{overlay}
9694 @itemx overlay unmap @var{overlay}
9695 @cindex unmap an overlay
9696 Tell @value{GDBN} that @var{overlay} is no longer mapped; @var{overlay}
9697 must be the name of the object file section containing the overlay.
9698 When an overlay is unmapped, @value{GDBN} assumes it can find the
9699 overlay's functions and variables at their load addresses.
9702 Enable @dfn{automatic} overlay debugging. In this mode, @value{GDBN}
9703 consults a data structure the overlay manager maintains in the inferior
9704 to see which overlays are mapped. For details, see @ref{Automatic
9707 @item overlay load-target
9709 @cindex reloading the overlay table
9710 Re-read the overlay table from the inferior. Normally, @value{GDBN}
9711 re-reads the table @value{GDBN} automatically each time the inferior
9712 stops, so this command should only be necessary if you have changed the
9713 overlay mapping yourself using @value{GDBN}. This command is only
9714 useful when using automatic overlay debugging.
9716 @item overlay list-overlays
9718 @cindex listing mapped overlays
9719 Display a list of the overlays currently mapped, along with their mapped
9720 addresses, load addresses, and sizes.
9724 Normally, when @value{GDBN} prints a code address, it includes the name
9725 of the function the address falls in:
9728 (@value{GDBP}) print main
9729 $3 = @{int ()@} 0x11a0 <main>
9732 When overlay debugging is enabled, @value{GDBN} recognizes code in
9733 unmapped overlays, and prints the names of unmapped functions with
9734 asterisks around them. For example, if @code{foo} is a function in an
9735 unmapped overlay, @value{GDBN} prints it this way:
9738 (@value{GDBP}) overlay list
9739 No sections are mapped.
9740 (@value{GDBP}) print foo
9741 $5 = @{int (int)@} 0x100000 <*foo*>
9744 When @code{foo}'s overlay is mapped, @value{GDBN} prints the function's
9748 (@value{GDBP}) overlay list
9749 Section .ov.foo.text, loaded at 0x100000 - 0x100034,
9750 mapped at 0x1016 - 0x104a
9751 (@value{GDBP}) print foo
9752 $6 = @{int (int)@} 0x1016 <foo>
9755 When overlay debugging is enabled, @value{GDBN} can find the correct
9756 address for functions and variables in an overlay, whether or not the
9757 overlay is mapped. This allows most @value{GDBN} commands, like
9758 @code{break} and @code{disassemble}, to work normally, even on unmapped
9759 code. However, @value{GDBN}'s breakpoint support has some limitations:
9763 @cindex breakpoints in overlays
9764 @cindex overlays, setting breakpoints in
9765 You can set breakpoints in functions in unmapped overlays, as long as
9766 @value{GDBN} can write to the overlay at its load address.
9768 @value{GDBN} can not set hardware or simulator-based breakpoints in
9769 unmapped overlays. However, if you set a breakpoint at the end of your
9770 overlay manager (and tell @value{GDBN} which overlays are now mapped, if
9771 you are using manual overlay management), @value{GDBN} will re-set its
9772 breakpoints properly.
9776 @node Automatic Overlay Debugging
9777 @section Automatic Overlay Debugging
9778 @cindex automatic overlay debugging
9780 @value{GDBN} can automatically track which overlays are mapped and which
9781 are not, given some simple co-operation from the overlay manager in the
9782 inferior. If you enable automatic overlay debugging with the
9783 @code{overlay auto} command (@pxref{Overlay Commands}), @value{GDBN}
9784 looks in the inferior's memory for certain variables describing the
9785 current state of the overlays.
9787 Here are the variables your overlay manager must define to support
9788 @value{GDBN}'s automatic overlay debugging:
9792 @item @code{_ovly_table}:
9793 This variable must be an array of the following structures:
9798 /* The overlay's mapped address. */
9801 /* The size of the overlay, in bytes. */
9804 /* The overlay's load address. */
9807 /* Non-zero if the overlay is currently mapped;
9809 unsigned long mapped;
9813 @item @code{_novlys}:
9814 This variable must be a four-byte signed integer, holding the total
9815 number of elements in @code{_ovly_table}.
9819 To decide whether a particular overlay is mapped or not, @value{GDBN}
9820 looks for an entry in @w{@code{_ovly_table}} whose @code{vma} and
9821 @code{lma} members equal the VMA and LMA of the overlay's section in the
9822 executable file. When @value{GDBN} finds a matching entry, it consults
9823 the entry's @code{mapped} member to determine whether the overlay is
9826 In addition, your overlay manager may define a function called
9827 @code{_ovly_debug_event}. If this function is defined, @value{GDBN}
9828 will silently set a breakpoint there. If the overlay manager then
9829 calls this function whenever it has changed the overlay table, this
9830 will enable @value{GDBN} to accurately keep track of which overlays
9831 are in program memory, and update any breakpoints that may be set
9832 in overlays. This will allow breakpoints to work even if the
9833 overlays are kept in ROM or other non-writable memory while they
9834 are not being executed.
9836 @node Overlay Sample Program
9837 @section Overlay Sample Program
9838 @cindex overlay example program
9840 When linking a program which uses overlays, you must place the overlays
9841 at their load addresses, while relocating them to run at their mapped
9842 addresses. To do this, you must write a linker script (@pxref{Overlay
9843 Description,,, ld.info, Using ld: the GNU linker}). Unfortunately,
9844 since linker scripts are specific to a particular host system, target
9845 architecture, and target memory layout, this manual cannot provide
9846 portable sample code demonstrating @value{GDBN}'s overlay support.
9848 However, the @value{GDBN} source distribution does contain an overlaid
9849 program, with linker scripts for a few systems, as part of its test
9850 suite. The program consists of the following files from
9851 @file{gdb/testsuite/gdb.base}:
9855 The main program file.
9857 A simple overlay manager, used by @file{overlays.c}.
9862 Overlay modules, loaded and used by @file{overlays.c}.
9865 Linker scripts for linking the test program on the @code{d10v-elf}
9866 and @code{m32r-elf} targets.
9869 You can build the test program using the @code{d10v-elf} GCC
9870 cross-compiler like this:
9873 $ d10v-elf-gcc -g -c overlays.c
9874 $ d10v-elf-gcc -g -c ovlymgr.c
9875 $ d10v-elf-gcc -g -c foo.c
9876 $ d10v-elf-gcc -g -c bar.c
9877 $ d10v-elf-gcc -g -c baz.c
9878 $ d10v-elf-gcc -g -c grbx.c
9879 $ d10v-elf-gcc -g overlays.o ovlymgr.o foo.o bar.o \
9880 baz.o grbx.o -Wl,-Td10v.ld -o overlays
9883 The build process is identical for any other architecture, except that
9884 you must substitute the appropriate compiler and linker script for the
9885 target system for @code{d10v-elf-gcc} and @code{d10v.ld}.
9889 @chapter Using @value{GDBN} with Different Languages
9892 Although programming languages generally have common aspects, they are
9893 rarely expressed in the same manner. For instance, in ANSI C,
9894 dereferencing a pointer @code{p} is accomplished by @code{*p}, but in
9895 Modula-2, it is accomplished by @code{p^}. Values can also be
9896 represented (and displayed) differently. Hex numbers in C appear as
9897 @samp{0x1ae}, while in Modula-2 they appear as @samp{1AEH}.
9899 @cindex working language
9900 Language-specific information is built into @value{GDBN} for some languages,
9901 allowing you to express operations like the above in your program's
9902 native language, and allowing @value{GDBN} to output values in a manner
9903 consistent with the syntax of your program's native language. The
9904 language you use to build expressions is called the @dfn{working
9908 * Setting:: Switching between source languages
9909 * Show:: Displaying the language
9910 * Checks:: Type and range checks
9911 * Supported Languages:: Supported languages
9912 * Unsupported Languages:: Unsupported languages
9916 @section Switching Between Source Languages
9918 There are two ways to control the working language---either have @value{GDBN}
9919 set it automatically, or select it manually yourself. You can use the
9920 @code{set language} command for either purpose. On startup, @value{GDBN}
9921 defaults to setting the language automatically. The working language is
9922 used to determine how expressions you type are interpreted, how values
9925 In addition to the working language, every source file that
9926 @value{GDBN} knows about has its own working language. For some object
9927 file formats, the compiler might indicate which language a particular
9928 source file is in. However, most of the time @value{GDBN} infers the
9929 language from the name of the file. The language of a source file
9930 controls whether C@t{++} names are demangled---this way @code{backtrace} can
9931 show each frame appropriately for its own language. There is no way to
9932 set the language of a source file from within @value{GDBN}, but you can
9933 set the language associated with a filename extension. @xref{Show, ,
9934 Displaying the Language}.
9936 This is most commonly a problem when you use a program, such
9937 as @code{cfront} or @code{f2c}, that generates C but is written in
9938 another language. In that case, make the
9939 program use @code{#line} directives in its C output; that way
9940 @value{GDBN} will know the correct language of the source code of the original
9941 program, and will display that source code, not the generated C code.
9944 * Filenames:: Filename extensions and languages.
9945 * Manually:: Setting the working language manually
9946 * Automatically:: Having @value{GDBN} infer the source language
9950 @subsection List of Filename Extensions and Languages
9952 If a source file name ends in one of the following extensions, then
9953 @value{GDBN} infers that its language is the one indicated.
9974 Objective-C source file
9981 Modula-2 source file
9985 Assembler source file. This actually behaves almost like C, but
9986 @value{GDBN} does not skip over function prologues when stepping.
9989 In addition, you may set the language associated with a filename
9990 extension. @xref{Show, , Displaying the Language}.
9993 @subsection Setting the Working Language
9995 If you allow @value{GDBN} to set the language automatically,
9996 expressions are interpreted the same way in your debugging session and
9999 @kindex set language
10000 If you wish, you may set the language manually. To do this, issue the
10001 command @samp{set language @var{lang}}, where @var{lang} is the name of
10002 a language, such as
10003 @code{c} or @code{modula-2}.
10004 For a list of the supported languages, type @samp{set language}.
10006 Setting the language manually prevents @value{GDBN} from updating the working
10007 language automatically. This can lead to confusion if you try
10008 to debug a program when the working language is not the same as the
10009 source language, when an expression is acceptable to both
10010 languages---but means different things. For instance, if the current
10011 source file were written in C, and @value{GDBN} was parsing Modula-2, a
10019 might not have the effect you intended. In C, this means to add
10020 @code{b} and @code{c} and place the result in @code{a}. The result
10021 printed would be the value of @code{a}. In Modula-2, this means to compare
10022 @code{a} to the result of @code{b+c}, yielding a @code{BOOLEAN} value.
10024 @node Automatically
10025 @subsection Having @value{GDBN} Infer the Source Language
10027 To have @value{GDBN} set the working language automatically, use
10028 @samp{set language local} or @samp{set language auto}. @value{GDBN}
10029 then infers the working language. That is, when your program stops in a
10030 frame (usually by encountering a breakpoint), @value{GDBN} sets the
10031 working language to the language recorded for the function in that
10032 frame. If the language for a frame is unknown (that is, if the function
10033 or block corresponding to the frame was defined in a source file that
10034 does not have a recognized extension), the current working language is
10035 not changed, and @value{GDBN} issues a warning.
10037 This may not seem necessary for most programs, which are written
10038 entirely in one source language. However, program modules and libraries
10039 written in one source language can be used by a main program written in
10040 a different source language. Using @samp{set language auto} in this
10041 case frees you from having to set the working language manually.
10044 @section Displaying the Language
10046 The following commands help you find out which language is the
10047 working language, and also what language source files were written in.
10050 @item show language
10051 @kindex show language
10052 Display the current working language. This is the
10053 language you can use with commands such as @code{print} to
10054 build and compute expressions that may involve variables in your program.
10057 @kindex info frame@r{, show the source language}
10058 Display the source language for this frame. This language becomes the
10059 working language if you use an identifier from this frame.
10060 @xref{Frame Info, ,Information about a Frame}, to identify the other
10061 information listed here.
10064 @kindex info source@r{, show the source language}
10065 Display the source language of this source file.
10066 @xref{Symbols, ,Examining the Symbol Table}, to identify the other
10067 information listed here.
10070 In unusual circumstances, you may have source files with extensions
10071 not in the standard list. You can then set the extension associated
10072 with a language explicitly:
10075 @item set extension-language @var{ext} @var{language}
10076 @kindex set extension-language
10077 Tell @value{GDBN} that source files with extension @var{ext} are to be
10078 assumed as written in the source language @var{language}.
10080 @item info extensions
10081 @kindex info extensions
10082 List all the filename extensions and the associated languages.
10086 @section Type and Range Checking
10089 @emph{Warning:} In this release, the @value{GDBN} commands for type and range
10090 checking are included, but they do not yet have any effect. This
10091 section documents the intended facilities.
10093 @c FIXME remove warning when type/range code added
10095 Some languages are designed to guard you against making seemingly common
10096 errors through a series of compile- and run-time checks. These include
10097 checking the type of arguments to functions and operators, and making
10098 sure mathematical overflows are caught at run time. Checks such as
10099 these help to ensure a program's correctness once it has been compiled
10100 by eliminating type mismatches, and providing active checks for range
10101 errors when your program is running.
10103 @value{GDBN} can check for conditions like the above if you wish.
10104 Although @value{GDBN} does not check the statements in your program,
10105 it can check expressions entered directly into @value{GDBN} for
10106 evaluation via the @code{print} command, for example. As with the
10107 working language, @value{GDBN} can also decide whether or not to check
10108 automatically based on your program's source language.
10109 @xref{Supported Languages, ,Supported Languages}, for the default
10110 settings of supported languages.
10113 * Type Checking:: An overview of type checking
10114 * Range Checking:: An overview of range checking
10117 @cindex type checking
10118 @cindex checks, type
10119 @node Type Checking
10120 @subsection An Overview of Type Checking
10122 Some languages, such as Modula-2, are strongly typed, meaning that the
10123 arguments to operators and functions have to be of the correct type,
10124 otherwise an error occurs. These checks prevent type mismatch
10125 errors from ever causing any run-time problems. For example,
10133 The second example fails because the @code{CARDINAL} 1 is not
10134 type-compatible with the @code{REAL} 2.3.
10136 For the expressions you use in @value{GDBN} commands, you can tell the
10137 @value{GDBN} type checker to skip checking;
10138 to treat any mismatches as errors and abandon the expression;
10139 or to only issue warnings when type mismatches occur,
10140 but evaluate the expression anyway. When you choose the last of
10141 these, @value{GDBN} evaluates expressions like the second example above, but
10142 also issues a warning.
10144 Even if you turn type checking off, there may be other reasons
10145 related to type that prevent @value{GDBN} from evaluating an expression.
10146 For instance, @value{GDBN} does not know how to add an @code{int} and
10147 a @code{struct foo}. These particular type errors have nothing to do
10148 with the language in use, and usually arise from expressions, such as
10149 the one described above, which make little sense to evaluate anyway.
10151 Each language defines to what degree it is strict about type. For
10152 instance, both Modula-2 and C require the arguments to arithmetical
10153 operators to be numbers. In C, enumerated types and pointers can be
10154 represented as numbers, so that they are valid arguments to mathematical
10155 operators. @xref{Supported Languages, ,Supported Languages}, for further
10156 details on specific languages.
10158 @value{GDBN} provides some additional commands for controlling the type checker:
10160 @kindex set check type
10161 @kindex show check type
10163 @item set check type auto
10164 Set type checking on or off based on the current working language.
10165 @xref{Supported Languages, ,Supported Languages}, for the default settings for
10168 @item set check type on
10169 @itemx set check type off
10170 Set type checking on or off, overriding the default setting for the
10171 current working language. Issue a warning if the setting does not
10172 match the language default. If any type mismatches occur in
10173 evaluating an expression while type checking is on, @value{GDBN} prints a
10174 message and aborts evaluation of the expression.
10176 @item set check type warn
10177 Cause the type checker to issue warnings, but to always attempt to
10178 evaluate the expression. Evaluating the expression may still
10179 be impossible for other reasons. For example, @value{GDBN} cannot add
10180 numbers and structures.
10183 Show the current setting of the type checker, and whether or not @value{GDBN}
10184 is setting it automatically.
10187 @cindex range checking
10188 @cindex checks, range
10189 @node Range Checking
10190 @subsection An Overview of Range Checking
10192 In some languages (such as Modula-2), it is an error to exceed the
10193 bounds of a type; this is enforced with run-time checks. Such range
10194 checking is meant to ensure program correctness by making sure
10195 computations do not overflow, or indices on an array element access do
10196 not exceed the bounds of the array.
10198 For expressions you use in @value{GDBN} commands, you can tell
10199 @value{GDBN} to treat range errors in one of three ways: ignore them,
10200 always treat them as errors and abandon the expression, or issue
10201 warnings but evaluate the expression anyway.
10203 A range error can result from numerical overflow, from exceeding an
10204 array index bound, or when you type a constant that is not a member
10205 of any type. Some languages, however, do not treat overflows as an
10206 error. In many implementations of C, mathematical overflow causes the
10207 result to ``wrap around'' to lower values---for example, if @var{m} is
10208 the largest integer value, and @var{s} is the smallest, then
10211 @var{m} + 1 @result{} @var{s}
10214 This, too, is specific to individual languages, and in some cases
10215 specific to individual compilers or machines. @xref{Supported Languages, ,
10216 Supported Languages}, for further details on specific languages.
10218 @value{GDBN} provides some additional commands for controlling the range checker:
10220 @kindex set check range
10221 @kindex show check range
10223 @item set check range auto
10224 Set range checking on or off based on the current working language.
10225 @xref{Supported Languages, ,Supported Languages}, for the default settings for
10228 @item set check range on
10229 @itemx set check range off
10230 Set range checking on or off, overriding the default setting for the
10231 current working language. A warning is issued if the setting does not
10232 match the language default. If a range error occurs and range checking is on,
10233 then a message is printed and evaluation of the expression is aborted.
10235 @item set check range warn
10236 Output messages when the @value{GDBN} range checker detects a range error,
10237 but attempt to evaluate the expression anyway. Evaluating the
10238 expression may still be impossible for other reasons, such as accessing
10239 memory that the process does not own (a typical example from many Unix
10243 Show the current setting of the range checker, and whether or not it is
10244 being set automatically by @value{GDBN}.
10247 @node Supported Languages
10248 @section Supported Languages
10250 @value{GDBN} supports C, C@t{++}, Objective-C, Fortran, Java, Pascal,
10251 assembly, Modula-2, and Ada.
10252 @c This is false ...
10253 Some @value{GDBN} features may be used in expressions regardless of the
10254 language you use: the @value{GDBN} @code{@@} and @code{::} operators,
10255 and the @samp{@{type@}addr} construct (@pxref{Expressions,
10256 ,Expressions}) can be used with the constructs of any supported
10259 The following sections detail to what degree each source language is
10260 supported by @value{GDBN}. These sections are not meant to be language
10261 tutorials or references, but serve only as a reference guide to what the
10262 @value{GDBN} expression parser accepts, and what input and output
10263 formats should look like for different languages. There are many good
10264 books written on each of these languages; please look to these for a
10265 language reference or tutorial.
10268 * C:: C and C@t{++}
10269 * Objective-C:: Objective-C
10270 * Fortran:: Fortran
10272 * Modula-2:: Modula-2
10277 @subsection C and C@t{++}
10279 @cindex C and C@t{++}
10280 @cindex expressions in C or C@t{++}
10282 Since C and C@t{++} are so closely related, many features of @value{GDBN} apply
10283 to both languages. Whenever this is the case, we discuss those languages
10287 @cindex @code{g++}, @sc{gnu} C@t{++} compiler
10288 @cindex @sc{gnu} C@t{++}
10289 The C@t{++} debugging facilities are jointly implemented by the C@t{++}
10290 compiler and @value{GDBN}. Therefore, to debug your C@t{++} code
10291 effectively, you must compile your C@t{++} programs with a supported
10292 C@t{++} compiler, such as @sc{gnu} @code{g++}, or the HP ANSI C@t{++}
10293 compiler (@code{aCC}).
10295 For best results when using @sc{gnu} C@t{++}, use the DWARF 2 debugging
10296 format; if it doesn't work on your system, try the stabs+ debugging
10297 format. You can select those formats explicitly with the @code{g++}
10298 command-line options @option{-gdwarf-2} and @option{-gstabs+}.
10299 @xref{Debugging Options,,Options for Debugging Your Program or GCC,
10300 gcc.info, Using the @sc{gnu} Compiler Collection (GCC)}.
10303 * C Operators:: C and C@t{++} operators
10304 * C Constants:: C and C@t{++} constants
10305 * C Plus Plus Expressions:: C@t{++} expressions
10306 * C Defaults:: Default settings for C and C@t{++}
10307 * C Checks:: C and C@t{++} type and range checks
10308 * Debugging C:: @value{GDBN} and C
10309 * Debugging C Plus Plus:: @value{GDBN} features for C@t{++}
10310 * Decimal Floating Point:: Numbers in Decimal Floating Point format
10314 @subsubsection C and C@t{++} Operators
10316 @cindex C and C@t{++} operators
10318 Operators must be defined on values of specific types. For instance,
10319 @code{+} is defined on numbers, but not on structures. Operators are
10320 often defined on groups of types.
10322 For the purposes of C and C@t{++}, the following definitions hold:
10327 @emph{Integral types} include @code{int} with any of its storage-class
10328 specifiers; @code{char}; @code{enum}; and, for C@t{++}, @code{bool}.
10331 @emph{Floating-point types} include @code{float}, @code{double}, and
10332 @code{long double} (if supported by the target platform).
10335 @emph{Pointer types} include all types defined as @code{(@var{type} *)}.
10338 @emph{Scalar types} include all of the above.
10343 The following operators are supported. They are listed here
10344 in order of increasing precedence:
10348 The comma or sequencing operator. Expressions in a comma-separated list
10349 are evaluated from left to right, with the result of the entire
10350 expression being the last expression evaluated.
10353 Assignment. The value of an assignment expression is the value
10354 assigned. Defined on scalar types.
10357 Used in an expression of the form @w{@code{@var{a} @var{op}= @var{b}}},
10358 and translated to @w{@code{@var{a} = @var{a op b}}}.
10359 @w{@code{@var{op}=}} and @code{=} have the same precedence.
10360 @var{op} is any one of the operators @code{|}, @code{^}, @code{&},
10361 @code{<<}, @code{>>}, @code{+}, @code{-}, @code{*}, @code{/}, @code{%}.
10364 The ternary operator. @code{@var{a} ? @var{b} : @var{c}} can be thought
10365 of as: if @var{a} then @var{b} else @var{c}. @var{a} should be of an
10369 Logical @sc{or}. Defined on integral types.
10372 Logical @sc{and}. Defined on integral types.
10375 Bitwise @sc{or}. Defined on integral types.
10378 Bitwise exclusive-@sc{or}. Defined on integral types.
10381 Bitwise @sc{and}. Defined on integral types.
10384 Equality and inequality. Defined on scalar types. The value of these
10385 expressions is 0 for false and non-zero for true.
10387 @item <@r{, }>@r{, }<=@r{, }>=
10388 Less than, greater than, less than or equal, greater than or equal.
10389 Defined on scalar types. The value of these expressions is 0 for false
10390 and non-zero for true.
10393 left shift, and right shift. Defined on integral types.
10396 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
10399 Addition and subtraction. Defined on integral types, floating-point types and
10402 @item *@r{, }/@r{, }%
10403 Multiplication, division, and modulus. Multiplication and division are
10404 defined on integral and floating-point types. Modulus is defined on
10408 Increment and decrement. When appearing before a variable, the
10409 operation is performed before the variable is used in an expression;
10410 when appearing after it, the variable's value is used before the
10411 operation takes place.
10414 Pointer dereferencing. Defined on pointer types. Same precedence as
10418 Address operator. Defined on variables. Same precedence as @code{++}.
10420 For debugging C@t{++}, @value{GDBN} implements a use of @samp{&} beyond what is
10421 allowed in the C@t{++} language itself: you can use @samp{&(&@var{ref})}
10422 to examine the address
10423 where a C@t{++} reference variable (declared with @samp{&@var{ref}}) is
10427 Negative. Defined on integral and floating-point types. Same
10428 precedence as @code{++}.
10431 Logical negation. Defined on integral types. Same precedence as
10435 Bitwise complement operator. Defined on integral types. Same precedence as
10440 Structure member, and pointer-to-structure member. For convenience,
10441 @value{GDBN} regards the two as equivalent, choosing whether to dereference a
10442 pointer based on the stored type information.
10443 Defined on @code{struct} and @code{union} data.
10446 Dereferences of pointers to members.
10449 Array indexing. @code{@var{a}[@var{i}]} is defined as
10450 @code{*(@var{a}+@var{i})}. Same precedence as @code{->}.
10453 Function parameter list. Same precedence as @code{->}.
10456 C@t{++} scope resolution operator. Defined on @code{struct}, @code{union},
10457 and @code{class} types.
10460 Doubled colons also represent the @value{GDBN} scope operator
10461 (@pxref{Expressions, ,Expressions}). Same precedence as @code{::},
10465 If an operator is redefined in the user code, @value{GDBN} usually
10466 attempts to invoke the redefined version instead of using the operator's
10467 predefined meaning.
10470 @subsubsection C and C@t{++} Constants
10472 @cindex C and C@t{++} constants
10474 @value{GDBN} allows you to express the constants of C and C@t{++} in the
10479 Integer constants are a sequence of digits. Octal constants are
10480 specified by a leading @samp{0} (i.e.@: zero), and hexadecimal constants
10481 by a leading @samp{0x} or @samp{0X}. Constants may also end with a letter
10482 @samp{l}, specifying that the constant should be treated as a
10486 Floating point constants are a sequence of digits, followed by a decimal
10487 point, followed by a sequence of digits, and optionally followed by an
10488 exponent. An exponent is of the form:
10489 @samp{@w{e@r{[[}+@r{]|}-@r{]}@var{nnn}}}, where @var{nnn} is another
10490 sequence of digits. The @samp{+} is optional for positive exponents.
10491 A floating-point constant may also end with a letter @samp{f} or
10492 @samp{F}, specifying that the constant should be treated as being of
10493 the @code{float} (as opposed to the default @code{double}) type; or with
10494 a letter @samp{l} or @samp{L}, which specifies a @code{long double}
10498 Enumerated constants consist of enumerated identifiers, or their
10499 integral equivalents.
10502 Character constants are a single character surrounded by single quotes
10503 (@code{'}), or a number---the ordinal value of the corresponding character
10504 (usually its @sc{ascii} value). Within quotes, the single character may
10505 be represented by a letter or by @dfn{escape sequences}, which are of
10506 the form @samp{\@var{nnn}}, where @var{nnn} is the octal representation
10507 of the character's ordinal value; or of the form @samp{\@var{x}}, where
10508 @samp{@var{x}} is a predefined special character---for example,
10509 @samp{\n} for newline.
10512 String constants are a sequence of character constants surrounded by
10513 double quotes (@code{"}). Any valid character constant (as described
10514 above) may appear. Double quotes within the string must be preceded by
10515 a backslash, so for instance @samp{"a\"b'c"} is a string of five
10519 Pointer constants are an integral value. You can also write pointers
10520 to constants using the C operator @samp{&}.
10523 Array constants are comma-separated lists surrounded by braces @samp{@{}
10524 and @samp{@}}; for example, @samp{@{1,2,3@}} is a three-element array of
10525 integers, @samp{@{@{1,2@}, @{3,4@}, @{5,6@}@}} is a three-by-two array,
10526 and @samp{@{&"hi", &"there", &"fred"@}} is a three-element array of pointers.
10529 @node C Plus Plus Expressions
10530 @subsubsection C@t{++} Expressions
10532 @cindex expressions in C@t{++}
10533 @value{GDBN} expression handling can interpret most C@t{++} expressions.
10535 @cindex debugging C@t{++} programs
10536 @cindex C@t{++} compilers
10537 @cindex debug formats and C@t{++}
10538 @cindex @value{NGCC} and C@t{++}
10540 @emph{Warning:} @value{GDBN} can only debug C@t{++} code if you use the
10541 proper compiler and the proper debug format. Currently, @value{GDBN}
10542 works best when debugging C@t{++} code that is compiled with
10543 @value{NGCC} 2.95.3 or with @value{NGCC} 3.1 or newer, using the options
10544 @option{-gdwarf-2} or @option{-gstabs+}. DWARF 2 is preferred over
10545 stabs+. Most configurations of @value{NGCC} emit either DWARF 2 or
10546 stabs+ as their default debug format, so you usually don't need to
10547 specify a debug format explicitly. Other compilers and/or debug formats
10548 are likely to work badly or not at all when using @value{GDBN} to debug
10554 @cindex member functions
10556 Member function calls are allowed; you can use expressions like
10559 count = aml->GetOriginal(x, y)
10562 @vindex this@r{, inside C@t{++} member functions}
10563 @cindex namespace in C@t{++}
10565 While a member function is active (in the selected stack frame), your
10566 expressions have the same namespace available as the member function;
10567 that is, @value{GDBN} allows implicit references to the class instance
10568 pointer @code{this} following the same rules as C@t{++}.
10570 @cindex call overloaded functions
10571 @cindex overloaded functions, calling
10572 @cindex type conversions in C@t{++}
10574 You can call overloaded functions; @value{GDBN} resolves the function
10575 call to the right definition, with some restrictions. @value{GDBN} does not
10576 perform overload resolution involving user-defined type conversions,
10577 calls to constructors, or instantiations of templates that do not exist
10578 in the program. It also cannot handle ellipsis argument lists or
10581 It does perform integral conversions and promotions, floating-point
10582 promotions, arithmetic conversions, pointer conversions, conversions of
10583 class objects to base classes, and standard conversions such as those of
10584 functions or arrays to pointers; it requires an exact match on the
10585 number of function arguments.
10587 Overload resolution is always performed, unless you have specified
10588 @code{set overload-resolution off}. @xref{Debugging C Plus Plus,
10589 ,@value{GDBN} Features for C@t{++}}.
10591 You must specify @code{set overload-resolution off} in order to use an
10592 explicit function signature to call an overloaded function, as in
10594 p 'foo(char,int)'('x', 13)
10597 The @value{GDBN} command-completion facility can simplify this;
10598 see @ref{Completion, ,Command Completion}.
10600 @cindex reference declarations
10602 @value{GDBN} understands variables declared as C@t{++} references; you can use
10603 them in expressions just as you do in C@t{++} source---they are automatically
10606 In the parameter list shown when @value{GDBN} displays a frame, the values of
10607 reference variables are not displayed (unlike other variables); this
10608 avoids clutter, since references are often used for large structures.
10609 The @emph{address} of a reference variable is always shown, unless
10610 you have specified @samp{set print address off}.
10613 @value{GDBN} supports the C@t{++} name resolution operator @code{::}---your
10614 expressions can use it just as expressions in your program do. Since
10615 one scope may be defined in another, you can use @code{::} repeatedly if
10616 necessary, for example in an expression like
10617 @samp{@var{scope1}::@var{scope2}::@var{name}}. @value{GDBN} also allows
10618 resolving name scope by reference to source files, in both C and C@t{++}
10619 debugging (@pxref{Variables, ,Program Variables}).
10622 In addition, when used with HP's C@t{++} compiler, @value{GDBN} supports
10623 calling virtual functions correctly, printing out virtual bases of
10624 objects, calling functions in a base subobject, casting objects, and
10625 invoking user-defined operators.
10628 @subsubsection C and C@t{++} Defaults
10630 @cindex C and C@t{++} defaults
10632 If you allow @value{GDBN} to set type and range checking automatically, they
10633 both default to @code{off} whenever the working language changes to
10634 C or C@t{++}. This happens regardless of whether you or @value{GDBN}
10635 selects the working language.
10637 If you allow @value{GDBN} to set the language automatically, it
10638 recognizes source files whose names end with @file{.c}, @file{.C}, or
10639 @file{.cc}, etc, and when @value{GDBN} enters code compiled from one of
10640 these files, it sets the working language to C or C@t{++}.
10641 @xref{Automatically, ,Having @value{GDBN} Infer the Source Language},
10642 for further details.
10644 @c Type checking is (a) primarily motivated by Modula-2, and (b)
10645 @c unimplemented. If (b) changes, it might make sense to let this node
10646 @c appear even if Mod-2 does not, but meanwhile ignore it. roland 16jul93.
10649 @subsubsection C and C@t{++} Type and Range Checks
10651 @cindex C and C@t{++} checks
10653 By default, when @value{GDBN} parses C or C@t{++} expressions, type checking
10654 is not used. However, if you turn type checking on, @value{GDBN}
10655 considers two variables type equivalent if:
10659 The two variables are structured and have the same structure, union, or
10663 The two variables have the same type name, or types that have been
10664 declared equivalent through @code{typedef}.
10667 @c leaving this out because neither J Gilmore nor R Pesch understand it.
10670 The two @code{struct}, @code{union}, or @code{enum} variables are
10671 declared in the same declaration. (Note: this may not be true for all C
10676 Range checking, if turned on, is done on mathematical operations. Array
10677 indices are not checked, since they are often used to index a pointer
10678 that is not itself an array.
10681 @subsubsection @value{GDBN} and C
10683 The @code{set print union} and @code{show print union} commands apply to
10684 the @code{union} type. When set to @samp{on}, any @code{union} that is
10685 inside a @code{struct} or @code{class} is also printed. Otherwise, it
10686 appears as @samp{@{...@}}.
10688 The @code{@@} operator aids in the debugging of dynamic arrays, formed
10689 with pointers and a memory allocation function. @xref{Expressions,
10692 @node Debugging C Plus Plus
10693 @subsubsection @value{GDBN} Features for C@t{++}
10695 @cindex commands for C@t{++}
10697 Some @value{GDBN} commands are particularly useful with C@t{++}, and some are
10698 designed specifically for use with C@t{++}. Here is a summary:
10701 @cindex break in overloaded functions
10702 @item @r{breakpoint menus}
10703 When you want a breakpoint in a function whose name is overloaded,
10704 @value{GDBN} has the capability to display a menu of possible breakpoint
10705 locations to help you specify which function definition you want.
10706 @xref{Ambiguous Expressions,,Ambiguous Expressions}.
10708 @cindex overloading in C@t{++}
10709 @item rbreak @var{regex}
10710 Setting breakpoints using regular expressions is helpful for setting
10711 breakpoints on overloaded functions that are not members of any special
10713 @xref{Set Breaks, ,Setting Breakpoints}.
10715 @cindex C@t{++} exception handling
10718 Debug C@t{++} exception handling using these commands. @xref{Set
10719 Catchpoints, , Setting Catchpoints}.
10721 @cindex inheritance
10722 @item ptype @var{typename}
10723 Print inheritance relationships as well as other information for type
10725 @xref{Symbols, ,Examining the Symbol Table}.
10727 @cindex C@t{++} symbol display
10728 @item set print demangle
10729 @itemx show print demangle
10730 @itemx set print asm-demangle
10731 @itemx show print asm-demangle
10732 Control whether C@t{++} symbols display in their source form, both when
10733 displaying code as C@t{++} source and when displaying disassemblies.
10734 @xref{Print Settings, ,Print Settings}.
10736 @item set print object
10737 @itemx show print object
10738 Choose whether to print derived (actual) or declared types of objects.
10739 @xref{Print Settings, ,Print Settings}.
10741 @item set print vtbl
10742 @itemx show print vtbl
10743 Control the format for printing virtual function tables.
10744 @xref{Print Settings, ,Print Settings}.
10745 (The @code{vtbl} commands do not work on programs compiled with the HP
10746 ANSI C@t{++} compiler (@code{aCC}).)
10748 @kindex set overload-resolution
10749 @cindex overloaded functions, overload resolution
10750 @item set overload-resolution on
10751 Enable overload resolution for C@t{++} expression evaluation. The default
10752 is on. For overloaded functions, @value{GDBN} evaluates the arguments
10753 and searches for a function whose signature matches the argument types,
10754 using the standard C@t{++} conversion rules (see @ref{C Plus Plus
10755 Expressions, ,C@t{++} Expressions}, for details).
10756 If it cannot find a match, it emits a message.
10758 @item set overload-resolution off
10759 Disable overload resolution for C@t{++} expression evaluation. For
10760 overloaded functions that are not class member functions, @value{GDBN}
10761 chooses the first function of the specified name that it finds in the
10762 symbol table, whether or not its arguments are of the correct type. For
10763 overloaded functions that are class member functions, @value{GDBN}
10764 searches for a function whose signature @emph{exactly} matches the
10767 @kindex show overload-resolution
10768 @item show overload-resolution
10769 Show the current setting of overload resolution.
10771 @item @r{Overloaded symbol names}
10772 You can specify a particular definition of an overloaded symbol, using
10773 the same notation that is used to declare such symbols in C@t{++}: type
10774 @code{@var{symbol}(@var{types})} rather than just @var{symbol}. You can
10775 also use the @value{GDBN} command-line word completion facilities to list the
10776 available choices, or to finish the type list for you.
10777 @xref{Completion,, Command Completion}, for details on how to do this.
10780 @node Decimal Floating Point
10781 @subsubsection Decimal Floating Point format
10782 @cindex decimal floating point format
10784 @value{GDBN} can examine, set and perform computations with numbers in
10785 decimal floating point format, which in the C language correspond to the
10786 @code{_Decimal32}, @code{_Decimal64} and @code{_Decimal128} types as
10787 specified by the extension to support decimal floating-point arithmetic.
10789 There are two encodings in use, depending on the architecture: BID (Binary
10790 Integer Decimal) for x86 and x86-64, and DPD (Densely Packed Decimal) for
10791 PowerPC. @value{GDBN} will use the appropriate encoding for the configured
10794 Because of a limitation in @file{libdecnumber}, the library used by @value{GDBN}
10795 to manipulate decimal floating point numbers, it is not possible to convert
10796 (using a cast, for example) integers wider than 32-bit to decimal float.
10798 In addition, in order to imitate @value{GDBN}'s behaviour with binary floating
10799 point computations, error checking in decimal float operations ignores
10800 underflow, overflow and divide by zero exceptions.
10802 In the PowerPC architecture, @value{GDBN} provides a set of pseudo-registers
10803 to inspect @code{_Decimal128} values stored in floating point registers. See
10804 @ref{PowerPC,,PowerPC} for more details.
10807 @subsection Objective-C
10809 @cindex Objective-C
10810 This section provides information about some commands and command
10811 options that are useful for debugging Objective-C code. See also
10812 @ref{Symbols, info classes}, and @ref{Symbols, info selectors}, for a
10813 few more commands specific to Objective-C support.
10816 * Method Names in Commands::
10817 * The Print Command with Objective-C::
10820 @node Method Names in Commands
10821 @subsubsection Method Names in Commands
10823 The following commands have been extended to accept Objective-C method
10824 names as line specifications:
10826 @kindex clear@r{, and Objective-C}
10827 @kindex break@r{, and Objective-C}
10828 @kindex info line@r{, and Objective-C}
10829 @kindex jump@r{, and Objective-C}
10830 @kindex list@r{, and Objective-C}
10834 @item @code{info line}
10839 A fully qualified Objective-C method name is specified as
10842 -[@var{Class} @var{methodName}]
10845 where the minus sign is used to indicate an instance method and a
10846 plus sign (not shown) is used to indicate a class method. The class
10847 name @var{Class} and method name @var{methodName} are enclosed in
10848 brackets, similar to the way messages are specified in Objective-C
10849 source code. For example, to set a breakpoint at the @code{create}
10850 instance method of class @code{Fruit} in the program currently being
10854 break -[Fruit create]
10857 To list ten program lines around the @code{initialize} class method,
10861 list +[NSText initialize]
10864 In the current version of @value{GDBN}, the plus or minus sign is
10865 required. In future versions of @value{GDBN}, the plus or minus
10866 sign will be optional, but you can use it to narrow the search. It
10867 is also possible to specify just a method name:
10873 You must specify the complete method name, including any colons. If
10874 your program's source files contain more than one @code{create} method,
10875 you'll be presented with a numbered list of classes that implement that
10876 method. Indicate your choice by number, or type @samp{0} to exit if
10879 As another example, to clear a breakpoint established at the
10880 @code{makeKeyAndOrderFront:} method of the @code{NSWindow} class, enter:
10883 clear -[NSWindow makeKeyAndOrderFront:]
10886 @node The Print Command with Objective-C
10887 @subsubsection The Print Command With Objective-C
10888 @cindex Objective-C, print objects
10889 @kindex print-object
10890 @kindex po @r{(@code{print-object})}
10892 The print command has also been extended to accept methods. For example:
10895 print -[@var{object} hash]
10898 @cindex print an Objective-C object description
10899 @cindex @code{_NSPrintForDebugger}, and printing Objective-C objects
10901 will tell @value{GDBN} to send the @code{hash} message to @var{object}
10902 and print the result. Also, an additional command has been added,
10903 @code{print-object} or @code{po} for short, which is meant to print
10904 the description of an object. However, this command may only work
10905 with certain Objective-C libraries that have a particular hook
10906 function, @code{_NSPrintForDebugger}, defined.
10909 @subsection Fortran
10910 @cindex Fortran-specific support in @value{GDBN}
10912 @value{GDBN} can be used to debug programs written in Fortran, but it
10913 currently supports only the features of Fortran 77 language.
10915 @cindex trailing underscore, in Fortran symbols
10916 Some Fortran compilers (@sc{gnu} Fortran 77 and Fortran 95 compilers
10917 among them) append an underscore to the names of variables and
10918 functions. When you debug programs compiled by those compilers, you
10919 will need to refer to variables and functions with a trailing
10923 * Fortran Operators:: Fortran operators and expressions
10924 * Fortran Defaults:: Default settings for Fortran
10925 * Special Fortran Commands:: Special @value{GDBN} commands for Fortran
10928 @node Fortran Operators
10929 @subsubsection Fortran Operators and Expressions
10931 @cindex Fortran operators and expressions
10933 Operators must be defined on values of specific types. For instance,
10934 @code{+} is defined on numbers, but not on characters or other non-
10935 arithmetic types. Operators are often defined on groups of types.
10939 The exponentiation operator. It raises the first operand to the power
10943 The range operator. Normally used in the form of array(low:high) to
10944 represent a section of array.
10947 The access component operator. Normally used to access elements in derived
10948 types. Also suitable for unions. As unions aren't part of regular Fortran,
10949 this can only happen when accessing a register that uses a gdbarch-defined
10953 @node Fortran Defaults
10954 @subsubsection Fortran Defaults
10956 @cindex Fortran Defaults
10958 Fortran symbols are usually case-insensitive, so @value{GDBN} by
10959 default uses case-insensitive matches for Fortran symbols. You can
10960 change that with the @samp{set case-insensitive} command, see
10961 @ref{Symbols}, for the details.
10963 @node Special Fortran Commands
10964 @subsubsection Special Fortran Commands
10966 @cindex Special Fortran commands
10968 @value{GDBN} has some commands to support Fortran-specific features,
10969 such as displaying common blocks.
10972 @cindex @code{COMMON} blocks, Fortran
10973 @kindex info common
10974 @item info common @r{[}@var{common-name}@r{]}
10975 This command prints the values contained in the Fortran @code{COMMON}
10976 block whose name is @var{common-name}. With no argument, the names of
10977 all @code{COMMON} blocks visible at the current program location are
10984 @cindex Pascal support in @value{GDBN}, limitations
10985 Debugging Pascal programs which use sets, subranges, file variables, or
10986 nested functions does not currently work. @value{GDBN} does not support
10987 entering expressions, printing values, or similar features using Pascal
10990 The Pascal-specific command @code{set print pascal_static-members}
10991 controls whether static members of Pascal objects are displayed.
10992 @xref{Print Settings, pascal_static-members}.
10995 @subsection Modula-2
10997 @cindex Modula-2, @value{GDBN} support
10999 The extensions made to @value{GDBN} to support Modula-2 only support
11000 output from the @sc{gnu} Modula-2 compiler (which is currently being
11001 developed). Other Modula-2 compilers are not currently supported, and
11002 attempting to debug executables produced by them is most likely
11003 to give an error as @value{GDBN} reads in the executable's symbol
11006 @cindex expressions in Modula-2
11008 * M2 Operators:: Built-in operators
11009 * Built-In Func/Proc:: Built-in functions and procedures
11010 * M2 Constants:: Modula-2 constants
11011 * M2 Types:: Modula-2 types
11012 * M2 Defaults:: Default settings for Modula-2
11013 * Deviations:: Deviations from standard Modula-2
11014 * M2 Checks:: Modula-2 type and range checks
11015 * M2 Scope:: The scope operators @code{::} and @code{.}
11016 * GDB/M2:: @value{GDBN} and Modula-2
11020 @subsubsection Operators
11021 @cindex Modula-2 operators
11023 Operators must be defined on values of specific types. For instance,
11024 @code{+} is defined on numbers, but not on structures. Operators are
11025 often defined on groups of types. For the purposes of Modula-2, the
11026 following definitions hold:
11031 @emph{Integral types} consist of @code{INTEGER}, @code{CARDINAL}, and
11035 @emph{Character types} consist of @code{CHAR} and its subranges.
11038 @emph{Floating-point types} consist of @code{REAL}.
11041 @emph{Pointer types} consist of anything declared as @code{POINTER TO
11045 @emph{Scalar types} consist of all of the above.
11048 @emph{Set types} consist of @code{SET} and @code{BITSET} types.
11051 @emph{Boolean types} consist of @code{BOOLEAN}.
11055 The following operators are supported, and appear in order of
11056 increasing precedence:
11060 Function argument or array index separator.
11063 Assignment. The value of @var{var} @code{:=} @var{value} is
11067 Less than, greater than on integral, floating-point, or enumerated
11071 Less than or equal to, greater than or equal to
11072 on integral, floating-point and enumerated types, or set inclusion on
11073 set types. Same precedence as @code{<}.
11075 @item =@r{, }<>@r{, }#
11076 Equality and two ways of expressing inequality, valid on scalar types.
11077 Same precedence as @code{<}. In @value{GDBN} scripts, only @code{<>} is
11078 available for inequality, since @code{#} conflicts with the script
11082 Set membership. Defined on set types and the types of their members.
11083 Same precedence as @code{<}.
11086 Boolean disjunction. Defined on boolean types.
11089 Boolean conjunction. Defined on boolean types.
11092 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
11095 Addition and subtraction on integral and floating-point types, or union
11096 and difference on set types.
11099 Multiplication on integral and floating-point types, or set intersection
11103 Division on floating-point types, or symmetric set difference on set
11104 types. Same precedence as @code{*}.
11107 Integer division and remainder. Defined on integral types. Same
11108 precedence as @code{*}.
11111 Negative. Defined on @code{INTEGER} and @code{REAL} data.
11114 Pointer dereferencing. Defined on pointer types.
11117 Boolean negation. Defined on boolean types. Same precedence as
11121 @code{RECORD} field selector. Defined on @code{RECORD} data. Same
11122 precedence as @code{^}.
11125 Array indexing. Defined on @code{ARRAY} data. Same precedence as @code{^}.
11128 Procedure argument list. Defined on @code{PROCEDURE} objects. Same precedence
11132 @value{GDBN} and Modula-2 scope operators.
11136 @emph{Warning:} Set expressions and their operations are not yet supported, so @value{GDBN}
11137 treats the use of the operator @code{IN}, or the use of operators
11138 @code{+}, @code{-}, @code{*}, @code{/}, @code{=}, , @code{<>}, @code{#},
11139 @code{<=}, and @code{>=} on sets as an error.
11143 @node Built-In Func/Proc
11144 @subsubsection Built-in Functions and Procedures
11145 @cindex Modula-2 built-ins
11147 Modula-2 also makes available several built-in procedures and functions.
11148 In describing these, the following metavariables are used:
11153 represents an @code{ARRAY} variable.
11156 represents a @code{CHAR} constant or variable.
11159 represents a variable or constant of integral type.
11162 represents an identifier that belongs to a set. Generally used in the
11163 same function with the metavariable @var{s}. The type of @var{s} should
11164 be @code{SET OF @var{mtype}} (where @var{mtype} is the type of @var{m}).
11167 represents a variable or constant of integral or floating-point type.
11170 represents a variable or constant of floating-point type.
11176 represents a variable.
11179 represents a variable or constant of one of many types. See the
11180 explanation of the function for details.
11183 All Modula-2 built-in procedures also return a result, described below.
11187 Returns the absolute value of @var{n}.
11190 If @var{c} is a lower case letter, it returns its upper case
11191 equivalent, otherwise it returns its argument.
11194 Returns the character whose ordinal value is @var{i}.
11197 Decrements the value in the variable @var{v} by one. Returns the new value.
11199 @item DEC(@var{v},@var{i})
11200 Decrements the value in the variable @var{v} by @var{i}. Returns the
11203 @item EXCL(@var{m},@var{s})
11204 Removes the element @var{m} from the set @var{s}. Returns the new
11207 @item FLOAT(@var{i})
11208 Returns the floating point equivalent of the integer @var{i}.
11210 @item HIGH(@var{a})
11211 Returns the index of the last member of @var{a}.
11214 Increments the value in the variable @var{v} by one. Returns the new value.
11216 @item INC(@var{v},@var{i})
11217 Increments the value in the variable @var{v} by @var{i}. Returns the
11220 @item INCL(@var{m},@var{s})
11221 Adds the element @var{m} to the set @var{s} if it is not already
11222 there. Returns the new set.
11225 Returns the maximum value of the type @var{t}.
11228 Returns the minimum value of the type @var{t}.
11231 Returns boolean TRUE if @var{i} is an odd number.
11234 Returns the ordinal value of its argument. For example, the ordinal
11235 value of a character is its @sc{ascii} value (on machines supporting the
11236 @sc{ascii} character set). @var{x} must be of an ordered type, which include
11237 integral, character and enumerated types.
11239 @item SIZE(@var{x})
11240 Returns the size of its argument. @var{x} can be a variable or a type.
11242 @item TRUNC(@var{r})
11243 Returns the integral part of @var{r}.
11245 @item TSIZE(@var{x})
11246 Returns the size of its argument. @var{x} can be a variable or a type.
11248 @item VAL(@var{t},@var{i})
11249 Returns the member of the type @var{t} whose ordinal value is @var{i}.
11253 @emph{Warning:} Sets and their operations are not yet supported, so
11254 @value{GDBN} treats the use of procedures @code{INCL} and @code{EXCL} as
11258 @cindex Modula-2 constants
11260 @subsubsection Constants
11262 @value{GDBN} allows you to express the constants of Modula-2 in the following
11268 Integer constants are simply a sequence of digits. When used in an
11269 expression, a constant is interpreted to be type-compatible with the
11270 rest of the expression. Hexadecimal integers are specified by a
11271 trailing @samp{H}, and octal integers by a trailing @samp{B}.
11274 Floating point constants appear as a sequence of digits, followed by a
11275 decimal point and another sequence of digits. An optional exponent can
11276 then be specified, in the form @samp{E@r{[}+@r{|}-@r{]}@var{nnn}}, where
11277 @samp{@r{[}+@r{|}-@r{]}@var{nnn}} is the desired exponent. All of the
11278 digits of the floating point constant must be valid decimal (base 10)
11282 Character constants consist of a single character enclosed by a pair of
11283 like quotes, either single (@code{'}) or double (@code{"}). They may
11284 also be expressed by their ordinal value (their @sc{ascii} value, usually)
11285 followed by a @samp{C}.
11288 String constants consist of a sequence of characters enclosed by a
11289 pair of like quotes, either single (@code{'}) or double (@code{"}).
11290 Escape sequences in the style of C are also allowed. @xref{C
11291 Constants, ,C and C@t{++} Constants}, for a brief explanation of escape
11295 Enumerated constants consist of an enumerated identifier.
11298 Boolean constants consist of the identifiers @code{TRUE} and
11302 Pointer constants consist of integral values only.
11305 Set constants are not yet supported.
11309 @subsubsection Modula-2 Types
11310 @cindex Modula-2 types
11312 Currently @value{GDBN} can print the following data types in Modula-2
11313 syntax: array types, record types, set types, pointer types, procedure
11314 types, enumerated types, subrange types and base types. You can also
11315 print the contents of variables declared using these type.
11316 This section gives a number of simple source code examples together with
11317 sample @value{GDBN} sessions.
11319 The first example contains the following section of code:
11328 and you can request @value{GDBN} to interrogate the type and value of
11329 @code{r} and @code{s}.
11332 (@value{GDBP}) print s
11334 (@value{GDBP}) ptype s
11336 (@value{GDBP}) print r
11338 (@value{GDBP}) ptype r
11343 Likewise if your source code declares @code{s} as:
11347 s: SET ['A'..'Z'] ;
11351 then you may query the type of @code{s} by:
11354 (@value{GDBP}) ptype s
11355 type = SET ['A'..'Z']
11359 Note that at present you cannot interactively manipulate set
11360 expressions using the debugger.
11362 The following example shows how you might declare an array in Modula-2
11363 and how you can interact with @value{GDBN} to print its type and contents:
11367 s: ARRAY [-10..10] OF CHAR ;
11371 (@value{GDBP}) ptype s
11372 ARRAY [-10..10] OF CHAR
11375 Note that the array handling is not yet complete and although the type
11376 is printed correctly, expression handling still assumes that all
11377 arrays have a lower bound of zero and not @code{-10} as in the example
11380 Here are some more type related Modula-2 examples:
11384 colour = (blue, red, yellow, green) ;
11385 t = [blue..yellow] ;
11393 The @value{GDBN} interaction shows how you can query the data type
11394 and value of a variable.
11397 (@value{GDBP}) print s
11399 (@value{GDBP}) ptype t
11400 type = [blue..yellow]
11404 In this example a Modula-2 array is declared and its contents
11405 displayed. Observe that the contents are written in the same way as
11406 their @code{C} counterparts.
11410 s: ARRAY [1..5] OF CARDINAL ;
11416 (@value{GDBP}) print s
11417 $1 = @{1, 0, 0, 0, 0@}
11418 (@value{GDBP}) ptype s
11419 type = ARRAY [1..5] OF CARDINAL
11422 The Modula-2 language interface to @value{GDBN} also understands
11423 pointer types as shown in this example:
11427 s: POINTER TO ARRAY [1..5] OF CARDINAL ;
11434 and you can request that @value{GDBN} describes the type of @code{s}.
11437 (@value{GDBP}) ptype s
11438 type = POINTER TO ARRAY [1..5] OF CARDINAL
11441 @value{GDBN} handles compound types as we can see in this example.
11442 Here we combine array types, record types, pointer types and subrange
11453 myarray = ARRAY myrange OF CARDINAL ;
11454 myrange = [-2..2] ;
11456 s: POINTER TO ARRAY myrange OF foo ;
11460 and you can ask @value{GDBN} to describe the type of @code{s} as shown
11464 (@value{GDBP}) ptype s
11465 type = POINTER TO ARRAY [-2..2] OF foo = RECORD
11468 f3 : ARRAY [-2..2] OF CARDINAL;
11473 @subsubsection Modula-2 Defaults
11474 @cindex Modula-2 defaults
11476 If type and range checking are set automatically by @value{GDBN}, they
11477 both default to @code{on} whenever the working language changes to
11478 Modula-2. This happens regardless of whether you or @value{GDBN}
11479 selected the working language.
11481 If you allow @value{GDBN} to set the language automatically, then entering
11482 code compiled from a file whose name ends with @file{.mod} sets the
11483 working language to Modula-2. @xref{Automatically, ,Having @value{GDBN}
11484 Infer the Source Language}, for further details.
11487 @subsubsection Deviations from Standard Modula-2
11488 @cindex Modula-2, deviations from
11490 A few changes have been made to make Modula-2 programs easier to debug.
11491 This is done primarily via loosening its type strictness:
11495 Unlike in standard Modula-2, pointer constants can be formed by
11496 integers. This allows you to modify pointer variables during
11497 debugging. (In standard Modula-2, the actual address contained in a
11498 pointer variable is hidden from you; it can only be modified
11499 through direct assignment to another pointer variable or expression that
11500 returned a pointer.)
11503 C escape sequences can be used in strings and characters to represent
11504 non-printable characters. @value{GDBN} prints out strings with these
11505 escape sequences embedded. Single non-printable characters are
11506 printed using the @samp{CHR(@var{nnn})} format.
11509 The assignment operator (@code{:=}) returns the value of its right-hand
11513 All built-in procedures both modify @emph{and} return their argument.
11517 @subsubsection Modula-2 Type and Range Checks
11518 @cindex Modula-2 checks
11521 @emph{Warning:} in this release, @value{GDBN} does not yet perform type or
11524 @c FIXME remove warning when type/range checks added
11526 @value{GDBN} considers two Modula-2 variables type equivalent if:
11530 They are of types that have been declared equivalent via a @code{TYPE
11531 @var{t1} = @var{t2}} statement
11534 They have been declared on the same line. (Note: This is true of the
11535 @sc{gnu} Modula-2 compiler, but it may not be true of other compilers.)
11538 As long as type checking is enabled, any attempt to combine variables
11539 whose types are not equivalent is an error.
11541 Range checking is done on all mathematical operations, assignment, array
11542 index bounds, and all built-in functions and procedures.
11545 @subsubsection The Scope Operators @code{::} and @code{.}
11547 @cindex @code{.}, Modula-2 scope operator
11548 @cindex colon, doubled as scope operator
11550 @vindex colon-colon@r{, in Modula-2}
11551 @c Info cannot handle :: but TeX can.
11554 @vindex ::@r{, in Modula-2}
11557 There are a few subtle differences between the Modula-2 scope operator
11558 (@code{.}) and the @value{GDBN} scope operator (@code{::}). The two have
11563 @var{module} . @var{id}
11564 @var{scope} :: @var{id}
11568 where @var{scope} is the name of a module or a procedure,
11569 @var{module} the name of a module, and @var{id} is any declared
11570 identifier within your program, except another module.
11572 Using the @code{::} operator makes @value{GDBN} search the scope
11573 specified by @var{scope} for the identifier @var{id}. If it is not
11574 found in the specified scope, then @value{GDBN} searches all scopes
11575 enclosing the one specified by @var{scope}.
11577 Using the @code{.} operator makes @value{GDBN} search the current scope for
11578 the identifier specified by @var{id} that was imported from the
11579 definition module specified by @var{module}. With this operator, it is
11580 an error if the identifier @var{id} was not imported from definition
11581 module @var{module}, or if @var{id} is not an identifier in
11585 @subsubsection @value{GDBN} and Modula-2
11587 Some @value{GDBN} commands have little use when debugging Modula-2 programs.
11588 Five subcommands of @code{set print} and @code{show print} apply
11589 specifically to C and C@t{++}: @samp{vtbl}, @samp{demangle},
11590 @samp{asm-demangle}, @samp{object}, and @samp{union}. The first four
11591 apply to C@t{++}, and the last to the C @code{union} type, which has no direct
11592 analogue in Modula-2.
11594 The @code{@@} operator (@pxref{Expressions, ,Expressions}), while available
11595 with any language, is not useful with Modula-2. Its
11596 intent is to aid the debugging of @dfn{dynamic arrays}, which cannot be
11597 created in Modula-2 as they can in C or C@t{++}. However, because an
11598 address can be specified by an integral constant, the construct
11599 @samp{@{@var{type}@}@var{adrexp}} is still useful.
11601 @cindex @code{#} in Modula-2
11602 In @value{GDBN} scripts, the Modula-2 inequality operator @code{#} is
11603 interpreted as the beginning of a comment. Use @code{<>} instead.
11609 The extensions made to @value{GDBN} for Ada only support
11610 output from the @sc{gnu} Ada (GNAT) compiler.
11611 Other Ada compilers are not currently supported, and
11612 attempting to debug executables produced by them is most likely
11616 @cindex expressions in Ada
11618 * Ada Mode Intro:: General remarks on the Ada syntax
11619 and semantics supported by Ada mode
11621 * Omissions from Ada:: Restrictions on the Ada expression syntax.
11622 * Additions to Ada:: Extensions of the Ada expression syntax.
11623 * Stopping Before Main Program:: Debugging the program during elaboration.
11624 * Ada Tasks:: Listing and setting breakpoints in tasks.
11625 * Ada Tasks and Core Files:: Tasking Support when Debugging Core Files
11626 * Ada Glitches:: Known peculiarities of Ada mode.
11629 @node Ada Mode Intro
11630 @subsubsection Introduction
11631 @cindex Ada mode, general
11633 The Ada mode of @value{GDBN} supports a fairly large subset of Ada expression
11634 syntax, with some extensions.
11635 The philosophy behind the design of this subset is
11639 That @value{GDBN} should provide basic literals and access to operations for
11640 arithmetic, dereferencing, field selection, indexing, and subprogram calls,
11641 leaving more sophisticated computations to subprograms written into the
11642 program (which therefore may be called from @value{GDBN}).
11645 That type safety and strict adherence to Ada language restrictions
11646 are not particularly important to the @value{GDBN} user.
11649 That brevity is important to the @value{GDBN} user.
11652 Thus, for brevity, the debugger acts as if all names declared in
11653 user-written packages are directly visible, even if they are not visible
11654 according to Ada rules, thus making it unnecessary to fully qualify most
11655 names with their packages, regardless of context. Where this causes
11656 ambiguity, @value{GDBN} asks the user's intent.
11658 The debugger will start in Ada mode if it detects an Ada main program.
11659 As for other languages, it will enter Ada mode when stopped in a program that
11660 was translated from an Ada source file.
11662 While in Ada mode, you may use `@t{--}' for comments. This is useful
11663 mostly for documenting command files. The standard @value{GDBN} comment
11664 (@samp{#}) still works at the beginning of a line in Ada mode, but not in the
11665 middle (to allow based literals).
11667 The debugger supports limited overloading. Given a subprogram call in which
11668 the function symbol has multiple definitions, it will use the number of
11669 actual parameters and some information about their types to attempt to narrow
11670 the set of definitions. It also makes very limited use of context, preferring
11671 procedures to functions in the context of the @code{call} command, and
11672 functions to procedures elsewhere.
11674 @node Omissions from Ada
11675 @subsubsection Omissions from Ada
11676 @cindex Ada, omissions from
11678 Here are the notable omissions from the subset:
11682 Only a subset of the attributes are supported:
11686 @t{'First}, @t{'Last}, and @t{'Length}
11687 on array objects (not on types and subtypes).
11690 @t{'Min} and @t{'Max}.
11693 @t{'Pos} and @t{'Val}.
11699 @t{'Range} on array objects (not subtypes), but only as the right
11700 operand of the membership (@code{in}) operator.
11703 @t{'Access}, @t{'Unchecked_Access}, and
11704 @t{'Unrestricted_Access} (a GNAT extension).
11712 @code{Characters.Latin_1} are not available and
11713 concatenation is not implemented. Thus, escape characters in strings are
11714 not currently available.
11717 Equality tests (@samp{=} and @samp{/=}) on arrays test for bitwise
11718 equality of representations. They will generally work correctly
11719 for strings and arrays whose elements have integer or enumeration types.
11720 They may not work correctly for arrays whose element
11721 types have user-defined equality, for arrays of real values
11722 (in particular, IEEE-conformant floating point, because of negative
11723 zeroes and NaNs), and for arrays whose elements contain unused bits with
11724 indeterminate values.
11727 The other component-by-component array operations (@code{and}, @code{or},
11728 @code{xor}, @code{not}, and relational tests other than equality)
11729 are not implemented.
11732 @cindex array aggregates (Ada)
11733 @cindex record aggregates (Ada)
11734 @cindex aggregates (Ada)
11735 There is limited support for array and record aggregates. They are
11736 permitted only on the right sides of assignments, as in these examples:
11739 (@value{GDBP}) set An_Array := (1, 2, 3, 4, 5, 6)
11740 (@value{GDBP}) set An_Array := (1, others => 0)
11741 (@value{GDBP}) set An_Array := (0|4 => 1, 1..3 => 2, 5 => 6)
11742 (@value{GDBP}) set A_2D_Array := ((1, 2, 3), (4, 5, 6), (7, 8, 9))
11743 (@value{GDBP}) set A_Record := (1, "Peter", True);
11744 (@value{GDBP}) set A_Record := (Name => "Peter", Id => 1, Alive => True)
11748 discriminant's value by assigning an aggregate has an
11749 undefined effect if that discriminant is used within the record.
11750 However, you can first modify discriminants by directly assigning to
11751 them (which normally would not be allowed in Ada), and then performing an
11752 aggregate assignment. For example, given a variable @code{A_Rec}
11753 declared to have a type such as:
11756 type Rec (Len : Small_Integer := 0) is record
11758 Vals : IntArray (1 .. Len);
11762 you can assign a value with a different size of @code{Vals} with two
11766 (@value{GDBP}) set A_Rec.Len := 4
11767 (@value{GDBP}) set A_Rec := (Id => 42, Vals => (1, 2, 3, 4))
11770 As this example also illustrates, @value{GDBN} is very loose about the usual
11771 rules concerning aggregates. You may leave out some of the
11772 components of an array or record aggregate (such as the @code{Len}
11773 component in the assignment to @code{A_Rec} above); they will retain their
11774 original values upon assignment. You may freely use dynamic values as
11775 indices in component associations. You may even use overlapping or
11776 redundant component associations, although which component values are
11777 assigned in such cases is not defined.
11780 Calls to dispatching subprograms are not implemented.
11783 The overloading algorithm is much more limited (i.e., less selective)
11784 than that of real Ada. It makes only limited use of the context in
11785 which a subexpression appears to resolve its meaning, and it is much
11786 looser in its rules for allowing type matches. As a result, some
11787 function calls will be ambiguous, and the user will be asked to choose
11788 the proper resolution.
11791 The @code{new} operator is not implemented.
11794 Entry calls are not implemented.
11797 Aside from printing, arithmetic operations on the native VAX floating-point
11798 formats are not supported.
11801 It is not possible to slice a packed array.
11804 The names @code{True} and @code{False}, when not part of a qualified name,
11805 are interpreted as if implicitly prefixed by @code{Standard}, regardless of
11807 Should your program
11808 redefine these names in a package or procedure (at best a dubious practice),
11809 you will have to use fully qualified names to access their new definitions.
11812 @node Additions to Ada
11813 @subsubsection Additions to Ada
11814 @cindex Ada, deviations from
11816 As it does for other languages, @value{GDBN} makes certain generic
11817 extensions to Ada (@pxref{Expressions}):
11821 If the expression @var{E} is a variable residing in memory (typically
11822 a local variable or array element) and @var{N} is a positive integer,
11823 then @code{@var{E}@@@var{N}} displays the values of @var{E} and the
11824 @var{N}-1 adjacent variables following it in memory as an array. In
11825 Ada, this operator is generally not necessary, since its prime use is
11826 in displaying parts of an array, and slicing will usually do this in
11827 Ada. However, there are occasional uses when debugging programs in
11828 which certain debugging information has been optimized away.
11831 @code{@var{B}::@var{var}} means ``the variable named @var{var} that
11832 appears in function or file @var{B}.'' When @var{B} is a file name,
11833 you must typically surround it in single quotes.
11836 The expression @code{@{@var{type}@} @var{addr}} means ``the variable of type
11837 @var{type} that appears at address @var{addr}.''
11840 A name starting with @samp{$} is a convenience variable
11841 (@pxref{Convenience Vars}) or a machine register (@pxref{Registers}).
11844 In addition, @value{GDBN} provides a few other shortcuts and outright
11845 additions specific to Ada:
11849 The assignment statement is allowed as an expression, returning
11850 its right-hand operand as its value. Thus, you may enter
11853 (@value{GDBP}) set x := y + 3
11854 (@value{GDBP}) print A(tmp := y + 1)
11858 The semicolon is allowed as an ``operator,'' returning as its value
11859 the value of its right-hand operand.
11860 This allows, for example,
11861 complex conditional breaks:
11864 (@value{GDBP}) break f
11865 (@value{GDBP}) condition 1 (report(i); k += 1; A(k) > 100)
11869 Rather than use catenation and symbolic character names to introduce special
11870 characters into strings, one may instead use a special bracket notation,
11871 which is also used to print strings. A sequence of characters of the form
11872 @samp{["@var{XX}"]} within a string or character literal denotes the
11873 (single) character whose numeric encoding is @var{XX} in hexadecimal. The
11874 sequence of characters @samp{["""]} also denotes a single quotation mark
11875 in strings. For example,
11877 "One line.["0a"]Next line.["0a"]"
11880 contains an ASCII newline character (@code{Ada.Characters.Latin_1.LF})
11884 The subtype used as a prefix for the attributes @t{'Pos}, @t{'Min}, and
11885 @t{'Max} is optional (and is ignored in any case). For example, it is valid
11889 (@value{GDBP}) print 'max(x, y)
11893 When printing arrays, @value{GDBN} uses positional notation when the
11894 array has a lower bound of 1, and uses a modified named notation otherwise.
11895 For example, a one-dimensional array of three integers with a lower bound
11896 of 3 might print as
11903 That is, in contrast to valid Ada, only the first component has a @code{=>}
11907 You may abbreviate attributes in expressions with any unique,
11908 multi-character subsequence of
11909 their names (an exact match gets preference).
11910 For example, you may use @t{a'len}, @t{a'gth}, or @t{a'lh}
11911 in place of @t{a'length}.
11914 @cindex quoting Ada internal identifiers
11915 Since Ada is case-insensitive, the debugger normally maps identifiers you type
11916 to lower case. The GNAT compiler uses upper-case characters for
11917 some of its internal identifiers, which are normally of no interest to users.
11918 For the rare occasions when you actually have to look at them,
11919 enclose them in angle brackets to avoid the lower-case mapping.
11922 (@value{GDBP}) print <JMPBUF_SAVE>[0]
11926 Printing an object of class-wide type or dereferencing an
11927 access-to-class-wide value will display all the components of the object's
11928 specific type (as indicated by its run-time tag). Likewise, component
11929 selection on such a value will operate on the specific type of the
11934 @node Stopping Before Main Program
11935 @subsubsection Stopping at the Very Beginning
11937 @cindex breakpointing Ada elaboration code
11938 It is sometimes necessary to debug the program during elaboration, and
11939 before reaching the main procedure.
11940 As defined in the Ada Reference
11941 Manual, the elaboration code is invoked from a procedure called
11942 @code{adainit}. To run your program up to the beginning of
11943 elaboration, simply use the following two commands:
11944 @code{tbreak adainit} and @code{run}.
11947 @subsubsection Extensions for Ada Tasks
11948 @cindex Ada, tasking
11950 Support for Ada tasks is analogous to that for threads (@pxref{Threads}).
11951 @value{GDBN} provides the following task-related commands:
11956 This command shows a list of current Ada tasks, as in the following example:
11963 (@value{GDBP}) info tasks
11964 ID TID P-ID Pri State Name
11965 1 8088000 0 15 Child Activation Wait main_task
11966 2 80a4000 1 15 Accept Statement b
11967 3 809a800 1 15 Child Activation Wait a
11968 * 4 80ae800 3 15 Runnable c
11973 In this listing, the asterisk before the last task indicates it to be the
11974 task currently being inspected.
11978 Represents @value{GDBN}'s internal task number.
11984 The parent's task ID (@value{GDBN}'s internal task number).
11987 The base priority of the task.
11990 Current state of the task.
11994 The task has been created but has not been activated. It cannot be
11998 The task is not blocked for any reason known to Ada. (It may be waiting
11999 for a mutex, though.) It is conceptually "executing" in normal mode.
12002 The task is terminated, in the sense of ARM 9.3 (5). Any dependents
12003 that were waiting on terminate alternatives have been awakened and have
12004 terminated themselves.
12006 @item Child Activation Wait
12007 The task is waiting for created tasks to complete activation.
12009 @item Accept Statement
12010 The task is waiting on an accept or selective wait statement.
12012 @item Waiting on entry call
12013 The task is waiting on an entry call.
12015 @item Async Select Wait
12016 The task is waiting to start the abortable part of an asynchronous
12020 The task is waiting on a select statement with only a delay
12023 @item Child Termination Wait
12024 The task is sleeping having completed a master within itself, and is
12025 waiting for the tasks dependent on that master to become terminated or
12026 waiting on a terminate Phase.
12028 @item Wait Child in Term Alt
12029 The task is sleeping waiting for tasks on terminate alternatives to
12030 finish terminating.
12032 @item Accepting RV with @var{taskno}
12033 The task is accepting a rendez-vous with the task @var{taskno}.
12037 Name of the task in the program.
12041 @kindex info task @var{taskno}
12042 @item info task @var{taskno}
12043 This command shows detailled informations on the specified task, as in
12044 the following example:
12049 (@value{GDBP}) info tasks
12050 ID TID P-ID Pri State Name
12051 1 8077880 0 15 Child Activation Wait main_task
12052 * 2 807c468 1 15 Runnable task_1
12053 (@value{GDBP}) info task 2
12054 Ada Task: 0x807c468
12057 Parent: 1 (main_task)
12063 @kindex task@r{ (Ada)}
12064 @cindex current Ada task ID
12065 This command prints the ID of the current task.
12071 (@value{GDBP}) info tasks
12072 ID TID P-ID Pri State Name
12073 1 8077870 0 15 Child Activation Wait main_task
12074 * 2 807c458 1 15 Runnable t
12075 (@value{GDBP}) task
12076 [Current task is 2]
12079 @item task @var{taskno}
12080 @cindex Ada task switching
12081 This command is like the @code{thread @var{threadno}}
12082 command (@pxref{Threads}). It switches the context of debugging
12083 from the current task to the given task.
12089 (@value{GDBP}) info tasks
12090 ID TID P-ID Pri State Name
12091 1 8077870 0 15 Child Activation Wait main_task
12092 * 2 807c458 1 15 Runnable t
12093 (@value{GDBP}) task 1
12094 [Switching to task 1]
12095 #0 0x8067726 in pthread_cond_wait ()
12097 #0 0x8067726 in pthread_cond_wait ()
12098 #1 0x8056714 in system.os_interface.pthread_cond_wait ()
12099 #2 0x805cb63 in system.task_primitives.operations.sleep ()
12100 #3 0x806153e in system.tasking.stages.activate_tasks ()
12101 #4 0x804aacc in un () at un.adb:5
12104 @item break @var{linespec} task @var{taskno}
12105 @itemx break @var{linespec} task @var{taskno} if @dots{}
12106 @cindex breakpoints and tasks, in Ada
12107 @cindex task breakpoints, in Ada
12108 @kindex break @dots{} task @var{taskno}@r{ (Ada)}
12109 These commands are like the @code{break @dots{} thread @dots{}}
12110 command (@pxref{Thread Stops}).
12111 @var{linespec} specifies source lines, as described
12112 in @ref{Specify Location}.
12114 Use the qualifier @samp{task @var{taskno}} with a breakpoint command
12115 to specify that you only want @value{GDBN} to stop the program when a
12116 particular Ada task reaches this breakpoint. @var{taskno} is one of the
12117 numeric task identifiers assigned by @value{GDBN}, shown in the first
12118 column of the @samp{info tasks} display.
12120 If you do not specify @samp{task @var{taskno}} when you set a
12121 breakpoint, the breakpoint applies to @emph{all} tasks of your
12124 You can use the @code{task} qualifier on conditional breakpoints as
12125 well; in this case, place @samp{task @var{taskno}} before the
12126 breakpoint condition (before the @code{if}).
12134 (@value{GDBP}) info tasks
12135 ID TID P-ID Pri State Name
12136 1 140022020 0 15 Child Activation Wait main_task
12137 2 140045060 1 15 Accept/Select Wait t2
12138 3 140044840 1 15 Runnable t1
12139 * 4 140056040 1 15 Runnable t3
12140 (@value{GDBP}) b 15 task 2
12141 Breakpoint 5 at 0x120044cb0: file test_task_debug.adb, line 15.
12142 (@value{GDBP}) cont
12147 Breakpoint 5, test_task_debug () at test_task_debug.adb:15
12149 (@value{GDBP}) info tasks
12150 ID TID P-ID Pri State Name
12151 1 140022020 0 15 Child Activation Wait main_task
12152 * 2 140045060 1 15 Runnable t2
12153 3 140044840 1 15 Runnable t1
12154 4 140056040 1 15 Delay Sleep t3
12158 @node Ada Tasks and Core Files
12159 @subsubsection Tasking Support when Debugging Core Files
12160 @cindex Ada tasking and core file debugging
12162 When inspecting a core file, as opposed to debugging a live program,
12163 tasking support may be limited or even unavailable, depending on
12164 the platform being used.
12165 For instance, on x86-linux, the list of tasks is available, but task
12166 switching is not supported. On Tru64, however, task switching will work
12169 On certain platforms, including Tru64, the debugger needs to perform some
12170 memory writes in order to provide Ada tasking support. When inspecting
12171 a core file, this means that the core file must be opened with read-write
12172 privileges, using the command @samp{"set write on"} (@pxref{Patching}).
12173 Under these circumstances, you should make a backup copy of the core
12174 file before inspecting it with @value{GDBN}.
12177 @subsubsection Known Peculiarities of Ada Mode
12178 @cindex Ada, problems
12180 Besides the omissions listed previously (@pxref{Omissions from Ada}),
12181 we know of several problems with and limitations of Ada mode in
12183 some of which will be fixed with planned future releases of the debugger
12184 and the GNU Ada compiler.
12188 Currently, the debugger
12189 has insufficient information to determine whether certain pointers represent
12190 pointers to objects or the objects themselves.
12191 Thus, the user may have to tack an extra @code{.all} after an expression
12192 to get it printed properly.
12195 Static constants that the compiler chooses not to materialize as objects in
12196 storage are invisible to the debugger.
12199 Named parameter associations in function argument lists are ignored (the
12200 argument lists are treated as positional).
12203 Many useful library packages are currently invisible to the debugger.
12206 Fixed-point arithmetic, conversions, input, and output is carried out using
12207 floating-point arithmetic, and may give results that only approximate those on
12211 The GNAT compiler never generates the prefix @code{Standard} for any of
12212 the standard symbols defined by the Ada language. @value{GDBN} knows about
12213 this: it will strip the prefix from names when you use it, and will never
12214 look for a name you have so qualified among local symbols, nor match against
12215 symbols in other packages or subprograms. If you have
12216 defined entities anywhere in your program other than parameters and
12217 local variables whose simple names match names in @code{Standard},
12218 GNAT's lack of qualification here can cause confusion. When this happens,
12219 you can usually resolve the confusion
12220 by qualifying the problematic names with package
12221 @code{Standard} explicitly.
12224 @node Unsupported Languages
12225 @section Unsupported Languages
12227 @cindex unsupported languages
12228 @cindex minimal language
12229 In addition to the other fully-supported programming languages,
12230 @value{GDBN} also provides a pseudo-language, called @code{minimal}.
12231 It does not represent a real programming language, but provides a set
12232 of capabilities close to what the C or assembly languages provide.
12233 This should allow most simple operations to be performed while debugging
12234 an application that uses a language currently not supported by @value{GDBN}.
12236 If the language is set to @code{auto}, @value{GDBN} will automatically
12237 select this language if the current frame corresponds to an unsupported
12241 @chapter Examining the Symbol Table
12243 The commands described in this chapter allow you to inquire about the
12244 symbols (names of variables, functions and types) defined in your
12245 program. This information is inherent in the text of your program and
12246 does not change as your program executes. @value{GDBN} finds it in your
12247 program's symbol table, in the file indicated when you started @value{GDBN}
12248 (@pxref{File Options, ,Choosing Files}), or by one of the
12249 file-management commands (@pxref{Files, ,Commands to Specify Files}).
12251 @cindex symbol names
12252 @cindex names of symbols
12253 @cindex quoting names
12254 Occasionally, you may need to refer to symbols that contain unusual
12255 characters, which @value{GDBN} ordinarily treats as word delimiters. The
12256 most frequent case is in referring to static variables in other
12257 source files (@pxref{Variables,,Program Variables}). File names
12258 are recorded in object files as debugging symbols, but @value{GDBN} would
12259 ordinarily parse a typical file name, like @file{foo.c}, as the three words
12260 @samp{foo} @samp{.} @samp{c}. To allow @value{GDBN} to recognize
12261 @samp{foo.c} as a single symbol, enclose it in single quotes; for example,
12268 looks up the value of @code{x} in the scope of the file @file{foo.c}.
12271 @cindex case-insensitive symbol names
12272 @cindex case sensitivity in symbol names
12273 @kindex set case-sensitive
12274 @item set case-sensitive on
12275 @itemx set case-sensitive off
12276 @itemx set case-sensitive auto
12277 Normally, when @value{GDBN} looks up symbols, it matches their names
12278 with case sensitivity determined by the current source language.
12279 Occasionally, you may wish to control that. The command @code{set
12280 case-sensitive} lets you do that by specifying @code{on} for
12281 case-sensitive matches or @code{off} for case-insensitive ones. If
12282 you specify @code{auto}, case sensitivity is reset to the default
12283 suitable for the source language. The default is case-sensitive
12284 matches for all languages except for Fortran, for which the default is
12285 case-insensitive matches.
12287 @kindex show case-sensitive
12288 @item show case-sensitive
12289 This command shows the current setting of case sensitivity for symbols
12292 @kindex info address
12293 @cindex address of a symbol
12294 @item info address @var{symbol}
12295 Describe where the data for @var{symbol} is stored. For a register
12296 variable, this says which register it is kept in. For a non-register
12297 local variable, this prints the stack-frame offset at which the variable
12300 Note the contrast with @samp{print &@var{symbol}}, which does not work
12301 at all for a register variable, and for a stack local variable prints
12302 the exact address of the current instantiation of the variable.
12304 @kindex info symbol
12305 @cindex symbol from address
12306 @cindex closest symbol and offset for an address
12307 @item info symbol @var{addr}
12308 Print the name of a symbol which is stored at the address @var{addr}.
12309 If no symbol is stored exactly at @var{addr}, @value{GDBN} prints the
12310 nearest symbol and an offset from it:
12313 (@value{GDBP}) info symbol 0x54320
12314 _initialize_vx + 396 in section .text
12318 This is the opposite of the @code{info address} command. You can use
12319 it to find out the name of a variable or a function given its address.
12321 For dynamically linked executables, the name of executable or shared
12322 library containing the symbol is also printed:
12325 (@value{GDBP}) info symbol 0x400225
12326 _start + 5 in section .text of /tmp/a.out
12327 (@value{GDBP}) info symbol 0x2aaaac2811cf
12328 __read_nocancel + 6 in section .text of /usr/lib64/libc.so.6
12332 @item whatis [@var{arg}]
12333 Print the data type of @var{arg}, which can be either an expression or
12334 a data type. With no argument, print the data type of @code{$}, the
12335 last value in the value history. If @var{arg} is an expression, it is
12336 not actually evaluated, and any side-effecting operations (such as
12337 assignments or function calls) inside it do not take place. If
12338 @var{arg} is a type name, it may be the name of a type or typedef, or
12339 for C code it may have the form @samp{class @var{class-name}},
12340 @samp{struct @var{struct-tag}}, @samp{union @var{union-tag}} or
12341 @samp{enum @var{enum-tag}}.
12342 @xref{Expressions, ,Expressions}.
12345 @item ptype [@var{arg}]
12346 @code{ptype} accepts the same arguments as @code{whatis}, but prints a
12347 detailed description of the type, instead of just the name of the type.
12348 @xref{Expressions, ,Expressions}.
12350 For example, for this variable declaration:
12353 struct complex @{double real; double imag;@} v;
12357 the two commands give this output:
12361 (@value{GDBP}) whatis v
12362 type = struct complex
12363 (@value{GDBP}) ptype v
12364 type = struct complex @{
12372 As with @code{whatis}, using @code{ptype} without an argument refers to
12373 the type of @code{$}, the last value in the value history.
12375 @cindex incomplete type
12376 Sometimes, programs use opaque data types or incomplete specifications
12377 of complex data structure. If the debug information included in the
12378 program does not allow @value{GDBN} to display a full declaration of
12379 the data type, it will say @samp{<incomplete type>}. For example,
12380 given these declarations:
12384 struct foo *fooptr;
12388 but no definition for @code{struct foo} itself, @value{GDBN} will say:
12391 (@value{GDBP}) ptype foo
12392 $1 = <incomplete type>
12396 ``Incomplete type'' is C terminology for data types that are not
12397 completely specified.
12400 @item info types @var{regexp}
12402 Print a brief description of all types whose names match the regular
12403 expression @var{regexp} (or all types in your program, if you supply
12404 no argument). Each complete typename is matched as though it were a
12405 complete line; thus, @samp{i type value} gives information on all
12406 types in your program whose names include the string @code{value}, but
12407 @samp{i type ^value$} gives information only on types whose complete
12408 name is @code{value}.
12410 This command differs from @code{ptype} in two ways: first, like
12411 @code{whatis}, it does not print a detailed description; second, it
12412 lists all source files where a type is defined.
12415 @cindex local variables
12416 @item info scope @var{location}
12417 List all the variables local to a particular scope. This command
12418 accepts a @var{location} argument---a function name, a source line, or
12419 an address preceded by a @samp{*}, and prints all the variables local
12420 to the scope defined by that location. (@xref{Specify Location}, for
12421 details about supported forms of @var{location}.) For example:
12424 (@value{GDBP}) @b{info scope command_line_handler}
12425 Scope for command_line_handler:
12426 Symbol rl is an argument at stack/frame offset 8, length 4.
12427 Symbol linebuffer is in static storage at address 0x150a18, length 4.
12428 Symbol linelength is in static storage at address 0x150a1c, length 4.
12429 Symbol p is a local variable in register $esi, length 4.
12430 Symbol p1 is a local variable in register $ebx, length 4.
12431 Symbol nline is a local variable in register $edx, length 4.
12432 Symbol repeat is a local variable at frame offset -8, length 4.
12436 This command is especially useful for determining what data to collect
12437 during a @dfn{trace experiment}, see @ref{Tracepoint Actions,
12440 @kindex info source
12442 Show information about the current source file---that is, the source file for
12443 the function containing the current point of execution:
12446 the name of the source file, and the directory containing it,
12448 the directory it was compiled in,
12450 its length, in lines,
12452 which programming language it is written in,
12454 whether the executable includes debugging information for that file, and
12455 if so, what format the information is in (e.g., STABS, Dwarf 2, etc.), and
12457 whether the debugging information includes information about
12458 preprocessor macros.
12462 @kindex info sources
12464 Print the names of all source files in your program for which there is
12465 debugging information, organized into two lists: files whose symbols
12466 have already been read, and files whose symbols will be read when needed.
12468 @kindex info functions
12469 @item info functions
12470 Print the names and data types of all defined functions.
12472 @item info functions @var{regexp}
12473 Print the names and data types of all defined functions
12474 whose names contain a match for regular expression @var{regexp}.
12475 Thus, @samp{info fun step} finds all functions whose names
12476 include @code{step}; @samp{info fun ^step} finds those whose names
12477 start with @code{step}. If a function name contains characters
12478 that conflict with the regular expression language (e.g.@:
12479 @samp{operator*()}), they may be quoted with a backslash.
12481 @kindex info variables
12482 @item info variables
12483 Print the names and data types of all variables that are declared
12484 outside of functions (i.e.@: excluding local variables).
12486 @item info variables @var{regexp}
12487 Print the names and data types of all variables (except for local
12488 variables) whose names contain a match for regular expression
12491 @kindex info classes
12492 @cindex Objective-C, classes and selectors
12494 @itemx info classes @var{regexp}
12495 Display all Objective-C classes in your program, or
12496 (with the @var{regexp} argument) all those matching a particular regular
12499 @kindex info selectors
12500 @item info selectors
12501 @itemx info selectors @var{regexp}
12502 Display all Objective-C selectors in your program, or
12503 (with the @var{regexp} argument) all those matching a particular regular
12507 This was never implemented.
12508 @kindex info methods
12510 @itemx info methods @var{regexp}
12511 The @code{info methods} command permits the user to examine all defined
12512 methods within C@t{++} program, or (with the @var{regexp} argument) a
12513 specific set of methods found in the various C@t{++} classes. Many
12514 C@t{++} classes provide a large number of methods. Thus, the output
12515 from the @code{ptype} command can be overwhelming and hard to use. The
12516 @code{info-methods} command filters the methods, printing only those
12517 which match the regular-expression @var{regexp}.
12520 @cindex reloading symbols
12521 Some systems allow individual object files that make up your program to
12522 be replaced without stopping and restarting your program. For example,
12523 in VxWorks you can simply recompile a defective object file and keep on
12524 running. If you are running on one of these systems, you can allow
12525 @value{GDBN} to reload the symbols for automatically relinked modules:
12528 @kindex set symbol-reloading
12529 @item set symbol-reloading on
12530 Replace symbol definitions for the corresponding source file when an
12531 object file with a particular name is seen again.
12533 @item set symbol-reloading off
12534 Do not replace symbol definitions when encountering object files of the
12535 same name more than once. This is the default state; if you are not
12536 running on a system that permits automatic relinking of modules, you
12537 should leave @code{symbol-reloading} off, since otherwise @value{GDBN}
12538 may discard symbols when linking large programs, that may contain
12539 several modules (from different directories or libraries) with the same
12542 @kindex show symbol-reloading
12543 @item show symbol-reloading
12544 Show the current @code{on} or @code{off} setting.
12547 @cindex opaque data types
12548 @kindex set opaque-type-resolution
12549 @item set opaque-type-resolution on
12550 Tell @value{GDBN} to resolve opaque types. An opaque type is a type
12551 declared as a pointer to a @code{struct}, @code{class}, or
12552 @code{union}---for example, @code{struct MyType *}---that is used in one
12553 source file although the full declaration of @code{struct MyType} is in
12554 another source file. The default is on.
12556 A change in the setting of this subcommand will not take effect until
12557 the next time symbols for a file are loaded.
12559 @item set opaque-type-resolution off
12560 Tell @value{GDBN} not to resolve opaque types. In this case, the type
12561 is printed as follows:
12563 @{<no data fields>@}
12566 @kindex show opaque-type-resolution
12567 @item show opaque-type-resolution
12568 Show whether opaque types are resolved or not.
12570 @kindex set print symbol-loading
12571 @cindex print messages when symbols are loaded
12572 @item set print symbol-loading
12573 @itemx set print symbol-loading on
12574 @itemx set print symbol-loading off
12575 The @code{set print symbol-loading} command allows you to enable or
12576 disable printing of messages when @value{GDBN} loads symbols.
12577 By default, these messages will be printed, and normally this is what
12578 you want. Disabling these messages is useful when debugging applications
12579 with lots of shared libraries where the quantity of output can be more
12580 annoying than useful.
12582 @kindex show print symbol-loading
12583 @item show print symbol-loading
12584 Show whether messages will be printed when @value{GDBN} loads symbols.
12586 @kindex maint print symbols
12587 @cindex symbol dump
12588 @kindex maint print psymbols
12589 @cindex partial symbol dump
12590 @item maint print symbols @var{filename}
12591 @itemx maint print psymbols @var{filename}
12592 @itemx maint print msymbols @var{filename}
12593 Write a dump of debugging symbol data into the file @var{filename}.
12594 These commands are used to debug the @value{GDBN} symbol-reading code. Only
12595 symbols with debugging data are included. If you use @samp{maint print
12596 symbols}, @value{GDBN} includes all the symbols for which it has already
12597 collected full details: that is, @var{filename} reflects symbols for
12598 only those files whose symbols @value{GDBN} has read. You can use the
12599 command @code{info sources} to find out which files these are. If you
12600 use @samp{maint print psymbols} instead, the dump shows information about
12601 symbols that @value{GDBN} only knows partially---that is, symbols defined in
12602 files that @value{GDBN} has skimmed, but not yet read completely. Finally,
12603 @samp{maint print msymbols} dumps just the minimal symbol information
12604 required for each object file from which @value{GDBN} has read some symbols.
12605 @xref{Files, ,Commands to Specify Files}, for a discussion of how
12606 @value{GDBN} reads symbols (in the description of @code{symbol-file}).
12608 @kindex maint info symtabs
12609 @kindex maint info psymtabs
12610 @cindex listing @value{GDBN}'s internal symbol tables
12611 @cindex symbol tables, listing @value{GDBN}'s internal
12612 @cindex full symbol tables, listing @value{GDBN}'s internal
12613 @cindex partial symbol tables, listing @value{GDBN}'s internal
12614 @item maint info symtabs @r{[} @var{regexp} @r{]}
12615 @itemx maint info psymtabs @r{[} @var{regexp} @r{]}
12617 List the @code{struct symtab} or @code{struct partial_symtab}
12618 structures whose names match @var{regexp}. If @var{regexp} is not
12619 given, list them all. The output includes expressions which you can
12620 copy into a @value{GDBN} debugging this one to examine a particular
12621 structure in more detail. For example:
12624 (@value{GDBP}) maint info psymtabs dwarf2read
12625 @{ objfile /home/gnu/build/gdb/gdb
12626 ((struct objfile *) 0x82e69d0)
12627 @{ psymtab /home/gnu/src/gdb/dwarf2read.c
12628 ((struct partial_symtab *) 0x8474b10)
12631 text addresses 0x814d3c8 -- 0x8158074
12632 globals (* (struct partial_symbol **) 0x8507a08 @@ 9)
12633 statics (* (struct partial_symbol **) 0x40e95b78 @@ 2882)
12634 dependencies (none)
12637 (@value{GDBP}) maint info symtabs
12641 We see that there is one partial symbol table whose filename contains
12642 the string @samp{dwarf2read}, belonging to the @samp{gdb} executable;
12643 and we see that @value{GDBN} has not read in any symtabs yet at all.
12644 If we set a breakpoint on a function, that will cause @value{GDBN} to
12645 read the symtab for the compilation unit containing that function:
12648 (@value{GDBP}) break dwarf2_psymtab_to_symtab
12649 Breakpoint 1 at 0x814e5da: file /home/gnu/src/gdb/dwarf2read.c,
12651 (@value{GDBP}) maint info symtabs
12652 @{ objfile /home/gnu/build/gdb/gdb
12653 ((struct objfile *) 0x82e69d0)
12654 @{ symtab /home/gnu/src/gdb/dwarf2read.c
12655 ((struct symtab *) 0x86c1f38)
12658 blockvector ((struct blockvector *) 0x86c1bd0) (primary)
12659 linetable ((struct linetable *) 0x8370fa0)
12660 debugformat DWARF 2
12669 @chapter Altering Execution
12671 Once you think you have found an error in your program, you might want to
12672 find out for certain whether correcting the apparent error would lead to
12673 correct results in the rest of the run. You can find the answer by
12674 experiment, using the @value{GDBN} features for altering execution of the
12677 For example, you can store new values into variables or memory
12678 locations, give your program a signal, restart it at a different
12679 address, or even return prematurely from a function.
12682 * Assignment:: Assignment to variables
12683 * Jumping:: Continuing at a different address
12684 * Signaling:: Giving your program a signal
12685 * Returning:: Returning from a function
12686 * Calling:: Calling your program's functions
12687 * Patching:: Patching your program
12691 @section Assignment to Variables
12694 @cindex setting variables
12695 To alter the value of a variable, evaluate an assignment expression.
12696 @xref{Expressions, ,Expressions}. For example,
12703 stores the value 4 into the variable @code{x}, and then prints the
12704 value of the assignment expression (which is 4).
12705 @xref{Languages, ,Using @value{GDBN} with Different Languages}, for more
12706 information on operators in supported languages.
12708 @kindex set variable
12709 @cindex variables, setting
12710 If you are not interested in seeing the value of the assignment, use the
12711 @code{set} command instead of the @code{print} command. @code{set} is
12712 really the same as @code{print} except that the expression's value is
12713 not printed and is not put in the value history (@pxref{Value History,
12714 ,Value History}). The expression is evaluated only for its effects.
12716 If the beginning of the argument string of the @code{set} command
12717 appears identical to a @code{set} subcommand, use the @code{set
12718 variable} command instead of just @code{set}. This command is identical
12719 to @code{set} except for its lack of subcommands. For example, if your
12720 program has a variable @code{width}, you get an error if you try to set
12721 a new value with just @samp{set width=13}, because @value{GDBN} has the
12722 command @code{set width}:
12725 (@value{GDBP}) whatis width
12727 (@value{GDBP}) p width
12729 (@value{GDBP}) set width=47
12730 Invalid syntax in expression.
12734 The invalid expression, of course, is @samp{=47}. In
12735 order to actually set the program's variable @code{width}, use
12738 (@value{GDBP}) set var width=47
12741 Because the @code{set} command has many subcommands that can conflict
12742 with the names of program variables, it is a good idea to use the
12743 @code{set variable} command instead of just @code{set}. For example, if
12744 your program has a variable @code{g}, you run into problems if you try
12745 to set a new value with just @samp{set g=4}, because @value{GDBN} has
12746 the command @code{set gnutarget}, abbreviated @code{set g}:
12750 (@value{GDBP}) whatis g
12754 (@value{GDBP}) set g=4
12758 The program being debugged has been started already.
12759 Start it from the beginning? (y or n) y
12760 Starting program: /home/smith/cc_progs/a.out
12761 "/home/smith/cc_progs/a.out": can't open to read symbols:
12762 Invalid bfd target.
12763 (@value{GDBP}) show g
12764 The current BFD target is "=4".
12769 The program variable @code{g} did not change, and you silently set the
12770 @code{gnutarget} to an invalid value. In order to set the variable
12774 (@value{GDBP}) set var g=4
12777 @value{GDBN} allows more implicit conversions in assignments than C; you can
12778 freely store an integer value into a pointer variable or vice versa,
12779 and you can convert any structure to any other structure that is the
12780 same length or shorter.
12781 @comment FIXME: how do structs align/pad in these conversions?
12782 @comment /doc@cygnus.com 18dec1990
12784 To store values into arbitrary places in memory, use the @samp{@{@dots{}@}}
12785 construct to generate a value of specified type at a specified address
12786 (@pxref{Expressions, ,Expressions}). For example, @code{@{int@}0x83040} refers
12787 to memory location @code{0x83040} as an integer (which implies a certain size
12788 and representation in memory), and
12791 set @{int@}0x83040 = 4
12795 stores the value 4 into that memory location.
12798 @section Continuing at a Different Address
12800 Ordinarily, when you continue your program, you do so at the place where
12801 it stopped, with the @code{continue} command. You can instead continue at
12802 an address of your own choosing, with the following commands:
12806 @item jump @var{linespec}
12807 @itemx jump @var{location}
12808 Resume execution at line @var{linespec} or at address given by
12809 @var{location}. Execution stops again immediately if there is a
12810 breakpoint there. @xref{Specify Location}, for a description of the
12811 different forms of @var{linespec} and @var{location}. It is common
12812 practice to use the @code{tbreak} command in conjunction with
12813 @code{jump}. @xref{Set Breaks, ,Setting Breakpoints}.
12815 The @code{jump} command does not change the current stack frame, or
12816 the stack pointer, or the contents of any memory location or any
12817 register other than the program counter. If line @var{linespec} is in
12818 a different function from the one currently executing, the results may
12819 be bizarre if the two functions expect different patterns of arguments or
12820 of local variables. For this reason, the @code{jump} command requests
12821 confirmation if the specified line is not in the function currently
12822 executing. However, even bizarre results are predictable if you are
12823 well acquainted with the machine-language code of your program.
12826 @c Doesn't work on HP-UX; have to set $pcoqh and $pcoqt.
12827 On many systems, you can get much the same effect as the @code{jump}
12828 command by storing a new value into the register @code{$pc}. The
12829 difference is that this does not start your program running; it only
12830 changes the address of where it @emph{will} run when you continue. For
12838 makes the next @code{continue} command or stepping command execute at
12839 address @code{0x485}, rather than at the address where your program stopped.
12840 @xref{Continuing and Stepping, ,Continuing and Stepping}.
12842 The most common occasion to use the @code{jump} command is to back
12843 up---perhaps with more breakpoints set---over a portion of a program
12844 that has already executed, in order to examine its execution in more
12849 @section Giving your Program a Signal
12850 @cindex deliver a signal to a program
12854 @item signal @var{signal}
12855 Resume execution where your program stopped, but immediately give it the
12856 signal @var{signal}. @var{signal} can be the name or the number of a
12857 signal. For example, on many systems @code{signal 2} and @code{signal
12858 SIGINT} are both ways of sending an interrupt signal.
12860 Alternatively, if @var{signal} is zero, continue execution without
12861 giving a signal. This is useful when your program stopped on account of
12862 a signal and would ordinary see the signal when resumed with the
12863 @code{continue} command; @samp{signal 0} causes it to resume without a
12866 @code{signal} does not repeat when you press @key{RET} a second time
12867 after executing the command.
12871 Invoking the @code{signal} command is not the same as invoking the
12872 @code{kill} utility from the shell. Sending a signal with @code{kill}
12873 causes @value{GDBN} to decide what to do with the signal depending on
12874 the signal handling tables (@pxref{Signals}). The @code{signal} command
12875 passes the signal directly to your program.
12879 @section Returning from a Function
12882 @cindex returning from a function
12885 @itemx return @var{expression}
12886 You can cancel execution of a function call with the @code{return}
12887 command. If you give an
12888 @var{expression} argument, its value is used as the function's return
12892 When you use @code{return}, @value{GDBN} discards the selected stack frame
12893 (and all frames within it). You can think of this as making the
12894 discarded frame return prematurely. If you wish to specify a value to
12895 be returned, give that value as the argument to @code{return}.
12897 This pops the selected stack frame (@pxref{Selection, ,Selecting a
12898 Frame}), and any other frames inside of it, leaving its caller as the
12899 innermost remaining frame. That frame becomes selected. The
12900 specified value is stored in the registers used for returning values
12903 The @code{return} command does not resume execution; it leaves the
12904 program stopped in the state that would exist if the function had just
12905 returned. In contrast, the @code{finish} command (@pxref{Continuing
12906 and Stepping, ,Continuing and Stepping}) resumes execution until the
12907 selected stack frame returns naturally.
12909 @value{GDBN} needs to know how the @var{expression} argument should be set for
12910 the inferior. The concrete registers assignment depends on the OS ABI and the
12911 type being returned by the selected stack frame. For example it is common for
12912 OS ABI to return floating point values in FPU registers while integer values in
12913 CPU registers. Still some ABIs return even floating point values in CPU
12914 registers. Larger integer widths (such as @code{long long int}) also have
12915 specific placement rules. @value{GDBN} already knows the OS ABI from its
12916 current target so it needs to find out also the type being returned to make the
12917 assignment into the right register(s).
12919 Normally, the selected stack frame has debug info. @value{GDBN} will always
12920 use the debug info instead of the implicit type of @var{expression} when the
12921 debug info is available. For example, if you type @kbd{return -1}, and the
12922 function in the current stack frame is declared to return a @code{long long
12923 int}, @value{GDBN} transparently converts the implicit @code{int} value of -1
12924 into a @code{long long int}:
12927 Breakpoint 1, func () at gdb.base/return-nodebug.c:29
12929 (@value{GDBP}) return -1
12930 Make func return now? (y or n) y
12931 #0 0x004004f6 in main () at gdb.base/return-nodebug.c:43
12932 43 printf ("result=%lld\n", func ());
12936 However, if the selected stack frame does not have a debug info, e.g., if the
12937 function was compiled without debug info, @value{GDBN} has to find out the type
12938 to return from user. Specifying a different type by mistake may set the value
12939 in different inferior registers than the caller code expects. For example,
12940 typing @kbd{return -1} with its implicit type @code{int} would set only a part
12941 of a @code{long long int} result for a debug info less function (on 32-bit
12942 architectures). Therefore the user is required to specify the return type by
12943 an appropriate cast explicitly:
12946 Breakpoint 2, 0x0040050b in func ()
12947 (@value{GDBP}) return -1
12948 Return value type not available for selected stack frame.
12949 Please use an explicit cast of the value to return.
12950 (@value{GDBP}) return (long long int) -1
12951 Make selected stack frame return now? (y or n) y
12952 #0 0x00400526 in main ()
12957 @section Calling Program Functions
12960 @cindex calling functions
12961 @cindex inferior functions, calling
12962 @item print @var{expr}
12963 Evaluate the expression @var{expr} and display the resulting value.
12964 @var{expr} may include calls to functions in the program being
12968 @item call @var{expr}
12969 Evaluate the expression @var{expr} without displaying @code{void}
12972 You can use this variant of the @code{print} command if you want to
12973 execute a function from your program that does not return anything
12974 (a.k.a.@: @dfn{a void function}), but without cluttering the output
12975 with @code{void} returned values that @value{GDBN} will otherwise
12976 print. If the result is not void, it is printed and saved in the
12980 It is possible for the function you call via the @code{print} or
12981 @code{call} command to generate a signal (e.g., if there's a bug in
12982 the function, or if you passed it incorrect arguments). What happens
12983 in that case is controlled by the @code{set unwindonsignal} command.
12985 Similarly, with a C@t{++} program it is possible for the function you
12986 call via the @code{print} or @code{call} command to generate an
12987 exception that is not handled due to the constraints of the dummy
12988 frame. In this case, any exception that is raised in the frame, but has
12989 an out-of-frame exception handler will not be found. GDB builds a
12990 dummy-frame for the inferior function call, and the unwinder cannot
12991 seek for exception handlers outside of this dummy-frame. What happens
12992 in that case is controlled by the
12993 @code{set unwind-on-terminating-exception} command.
12996 @item set unwindonsignal
12997 @kindex set unwindonsignal
12998 @cindex unwind stack in called functions
12999 @cindex call dummy stack unwinding
13000 Set unwinding of the stack if a signal is received while in a function
13001 that @value{GDBN} called in the program being debugged. If set to on,
13002 @value{GDBN} unwinds the stack it created for the call and restores
13003 the context to what it was before the call. If set to off (the
13004 default), @value{GDBN} stops in the frame where the signal was
13007 @item show unwindonsignal
13008 @kindex show unwindonsignal
13009 Show the current setting of stack unwinding in the functions called by
13012 @item set unwind-on-terminating-exception
13013 @kindex set unwind-on-terminating-exception
13014 @cindex unwind stack in called functions with unhandled exceptions
13015 @cindex call dummy stack unwinding on unhandled exception.
13016 Set unwinding of the stack if a C@t{++} exception is raised, but left
13017 unhandled while in a function that @value{GDBN} called in the program being
13018 debugged. If set to on (the default), @value{GDBN} unwinds the stack
13019 it created for the call and restores the context to what it was before
13020 the call. If set to off, @value{GDBN} the exception is delivered to
13021 the default C@t{++} exception handler and the inferior terminated.
13023 @item show unwind-on-terminating-exception
13024 @kindex show unwind-on-terminating-exception
13025 Show the current setting of stack unwinding in the functions called by
13030 @cindex weak alias functions
13031 Sometimes, a function you wish to call is actually a @dfn{weak alias}
13032 for another function. In such case, @value{GDBN} might not pick up
13033 the type information, including the types of the function arguments,
13034 which causes @value{GDBN} to call the inferior function incorrectly.
13035 As a result, the called function will function erroneously and may
13036 even crash. A solution to that is to use the name of the aliased
13040 @section Patching Programs
13042 @cindex patching binaries
13043 @cindex writing into executables
13044 @cindex writing into corefiles
13046 By default, @value{GDBN} opens the file containing your program's
13047 executable code (or the corefile) read-only. This prevents accidental
13048 alterations to machine code; but it also prevents you from intentionally
13049 patching your program's binary.
13051 If you'd like to be able to patch the binary, you can specify that
13052 explicitly with the @code{set write} command. For example, you might
13053 want to turn on internal debugging flags, or even to make emergency
13059 @itemx set write off
13060 If you specify @samp{set write on}, @value{GDBN} opens executable and
13061 core files for both reading and writing; if you specify @kbd{set write
13062 off} (the default), @value{GDBN} opens them read-only.
13064 If you have already loaded a file, you must load it again (using the
13065 @code{exec-file} or @code{core-file} command) after changing @code{set
13066 write}, for your new setting to take effect.
13070 Display whether executable files and core files are opened for writing
13071 as well as reading.
13075 @chapter @value{GDBN} Files
13077 @value{GDBN} needs to know the file name of the program to be debugged,
13078 both in order to read its symbol table and in order to start your
13079 program. To debug a core dump of a previous run, you must also tell
13080 @value{GDBN} the name of the core dump file.
13083 * Files:: Commands to specify files
13084 * Separate Debug Files:: Debugging information in separate files
13085 * Symbol Errors:: Errors reading symbol files
13086 * Data Files:: GDB data files
13090 @section Commands to Specify Files
13092 @cindex symbol table
13093 @cindex core dump file
13095 You may want to specify executable and core dump file names. The usual
13096 way to do this is at start-up time, using the arguments to
13097 @value{GDBN}'s start-up commands (@pxref{Invocation, , Getting In and
13098 Out of @value{GDBN}}).
13100 Occasionally it is necessary to change to a different file during a
13101 @value{GDBN} session. Or you may run @value{GDBN} and forget to
13102 specify a file you want to use. Or you are debugging a remote target
13103 via @code{gdbserver} (@pxref{Server, file, Using the @code{gdbserver}
13104 Program}). In these situations the @value{GDBN} commands to specify
13105 new files are useful.
13108 @cindex executable file
13110 @item file @var{filename}
13111 Use @var{filename} as the program to be debugged. It is read for its
13112 symbols and for the contents of pure memory. It is also the program
13113 executed when you use the @code{run} command. If you do not specify a
13114 directory and the file is not found in the @value{GDBN} working directory,
13115 @value{GDBN} uses the environment variable @code{PATH} as a list of
13116 directories to search, just as the shell does when looking for a program
13117 to run. You can change the value of this variable, for both @value{GDBN}
13118 and your program, using the @code{path} command.
13120 @cindex unlinked object files
13121 @cindex patching object files
13122 You can load unlinked object @file{.o} files into @value{GDBN} using
13123 the @code{file} command. You will not be able to ``run'' an object
13124 file, but you can disassemble functions and inspect variables. Also,
13125 if the underlying BFD functionality supports it, you could use
13126 @kbd{gdb -write} to patch object files using this technique. Note
13127 that @value{GDBN} can neither interpret nor modify relocations in this
13128 case, so branches and some initialized variables will appear to go to
13129 the wrong place. But this feature is still handy from time to time.
13132 @code{file} with no argument makes @value{GDBN} discard any information it
13133 has on both executable file and the symbol table.
13136 @item exec-file @r{[} @var{filename} @r{]}
13137 Specify that the program to be run (but not the symbol table) is found
13138 in @var{filename}. @value{GDBN} searches the environment variable @code{PATH}
13139 if necessary to locate your program. Omitting @var{filename} means to
13140 discard information on the executable file.
13142 @kindex symbol-file
13143 @item symbol-file @r{[} @var{filename} @r{]}
13144 Read symbol table information from file @var{filename}. @code{PATH} is
13145 searched when necessary. Use the @code{file} command to get both symbol
13146 table and program to run from the same file.
13148 @code{symbol-file} with no argument clears out @value{GDBN} information on your
13149 program's symbol table.
13151 The @code{symbol-file} command causes @value{GDBN} to forget the contents of
13152 some breakpoints and auto-display expressions. This is because they may
13153 contain pointers to the internal data recording symbols and data types,
13154 which are part of the old symbol table data being discarded inside
13157 @code{symbol-file} does not repeat if you press @key{RET} again after
13160 When @value{GDBN} is configured for a particular environment, it
13161 understands debugging information in whatever format is the standard
13162 generated for that environment; you may use either a @sc{gnu} compiler, or
13163 other compilers that adhere to the local conventions.
13164 Best results are usually obtained from @sc{gnu} compilers; for example,
13165 using @code{@value{NGCC}} you can generate debugging information for
13168 For most kinds of object files, with the exception of old SVR3 systems
13169 using COFF, the @code{symbol-file} command does not normally read the
13170 symbol table in full right away. Instead, it scans the symbol table
13171 quickly to find which source files and which symbols are present. The
13172 details are read later, one source file at a time, as they are needed.
13174 The purpose of this two-stage reading strategy is to make @value{GDBN}
13175 start up faster. For the most part, it is invisible except for
13176 occasional pauses while the symbol table details for a particular source
13177 file are being read. (The @code{set verbose} command can turn these
13178 pauses into messages if desired. @xref{Messages/Warnings, ,Optional
13179 Warnings and Messages}.)
13181 We have not implemented the two-stage strategy for COFF yet. When the
13182 symbol table is stored in COFF format, @code{symbol-file} reads the
13183 symbol table data in full right away. Note that ``stabs-in-COFF''
13184 still does the two-stage strategy, since the debug info is actually
13188 @cindex reading symbols immediately
13189 @cindex symbols, reading immediately
13190 @item symbol-file @var{filename} @r{[} -readnow @r{]}
13191 @itemx file @var{filename} @r{[} -readnow @r{]}
13192 You can override the @value{GDBN} two-stage strategy for reading symbol
13193 tables by using the @samp{-readnow} option with any of the commands that
13194 load symbol table information, if you want to be sure @value{GDBN} has the
13195 entire symbol table available.
13197 @c FIXME: for now no mention of directories, since this seems to be in
13198 @c flux. 13mar1992 status is that in theory GDB would look either in
13199 @c current dir or in same dir as myprog; but issues like competing
13200 @c GDB's, or clutter in system dirs, mean that in practice right now
13201 @c only current dir is used. FFish says maybe a special GDB hierarchy
13202 @c (eg rooted in val of env var GDBSYMS) could exist for mappable symbol
13206 @item core-file @r{[}@var{filename}@r{]}
13208 Specify the whereabouts of a core dump file to be used as the ``contents
13209 of memory''. Traditionally, core files contain only some parts of the
13210 address space of the process that generated them; @value{GDBN} can access the
13211 executable file itself for other parts.
13213 @code{core-file} with no argument specifies that no core file is
13216 Note that the core file is ignored when your program is actually running
13217 under @value{GDBN}. So, if you have been running your program and you
13218 wish to debug a core file instead, you must kill the subprocess in which
13219 the program is running. To do this, use the @code{kill} command
13220 (@pxref{Kill Process, ,Killing the Child Process}).
13222 @kindex add-symbol-file
13223 @cindex dynamic linking
13224 @item add-symbol-file @var{filename} @var{address}
13225 @itemx add-symbol-file @var{filename} @var{address} @r{[} -readnow @r{]}
13226 @itemx add-symbol-file @var{filename} @r{-s}@var{section} @var{address} @dots{}
13227 The @code{add-symbol-file} command reads additional symbol table
13228 information from the file @var{filename}. You would use this command
13229 when @var{filename} has been dynamically loaded (by some other means)
13230 into the program that is running. @var{address} should be the memory
13231 address at which the file has been loaded; @value{GDBN} cannot figure
13232 this out for itself. You can additionally specify an arbitrary number
13233 of @samp{@r{-s}@var{section} @var{address}} pairs, to give an explicit
13234 section name and base address for that section. You can specify any
13235 @var{address} as an expression.
13237 The symbol table of the file @var{filename} is added to the symbol table
13238 originally read with the @code{symbol-file} command. You can use the
13239 @code{add-symbol-file} command any number of times; the new symbol data
13240 thus read keeps adding to the old. To discard all old symbol data
13241 instead, use the @code{symbol-file} command without any arguments.
13243 @cindex relocatable object files, reading symbols from
13244 @cindex object files, relocatable, reading symbols from
13245 @cindex reading symbols from relocatable object files
13246 @cindex symbols, reading from relocatable object files
13247 @cindex @file{.o} files, reading symbols from
13248 Although @var{filename} is typically a shared library file, an
13249 executable file, or some other object file which has been fully
13250 relocated for loading into a process, you can also load symbolic
13251 information from relocatable @file{.o} files, as long as:
13255 the file's symbolic information refers only to linker symbols defined in
13256 that file, not to symbols defined by other object files,
13258 every section the file's symbolic information refers to has actually
13259 been loaded into the inferior, as it appears in the file, and
13261 you can determine the address at which every section was loaded, and
13262 provide these to the @code{add-symbol-file} command.
13266 Some embedded operating systems, like Sun Chorus and VxWorks, can load
13267 relocatable files into an already running program; such systems
13268 typically make the requirements above easy to meet. However, it's
13269 important to recognize that many native systems use complex link
13270 procedures (@code{.linkonce} section factoring and C@t{++} constructor table
13271 assembly, for example) that make the requirements difficult to meet. In
13272 general, one cannot assume that using @code{add-symbol-file} to read a
13273 relocatable object file's symbolic information will have the same effect
13274 as linking the relocatable object file into the program in the normal
13277 @code{add-symbol-file} does not repeat if you press @key{RET} after using it.
13279 @kindex add-symbol-file-from-memory
13280 @cindex @code{syscall DSO}
13281 @cindex load symbols from memory
13282 @item add-symbol-file-from-memory @var{address}
13283 Load symbols from the given @var{address} in a dynamically loaded
13284 object file whose image is mapped directly into the inferior's memory.
13285 For example, the Linux kernel maps a @code{syscall DSO} into each
13286 process's address space; this DSO provides kernel-specific code for
13287 some system calls. The argument can be any expression whose
13288 evaluation yields the address of the file's shared object file header.
13289 For this command to work, you must have used @code{symbol-file} or
13290 @code{exec-file} commands in advance.
13292 @kindex add-shared-symbol-files
13294 @item add-shared-symbol-files @var{library-file}
13295 @itemx assf @var{library-file}
13296 The @code{add-shared-symbol-files} command can currently be used only
13297 in the Cygwin build of @value{GDBN} on MS-Windows OS, where it is an
13298 alias for the @code{dll-symbols} command (@pxref{Cygwin Native}).
13299 @value{GDBN} automatically looks for shared libraries, however if
13300 @value{GDBN} does not find yours, you can invoke
13301 @code{add-shared-symbol-files}. It takes one argument: the shared
13302 library's file name. @code{assf} is a shorthand alias for
13303 @code{add-shared-symbol-files}.
13306 @item section @var{section} @var{addr}
13307 The @code{section} command changes the base address of the named
13308 @var{section} of the exec file to @var{addr}. This can be used if the
13309 exec file does not contain section addresses, (such as in the
13310 @code{a.out} format), or when the addresses specified in the file
13311 itself are wrong. Each section must be changed separately. The
13312 @code{info files} command, described below, lists all the sections and
13316 @kindex info target
13319 @code{info files} and @code{info target} are synonymous; both print the
13320 current target (@pxref{Targets, ,Specifying a Debugging Target}),
13321 including the names of the executable and core dump files currently in
13322 use by @value{GDBN}, and the files from which symbols were loaded. The
13323 command @code{help target} lists all possible targets rather than
13326 @kindex maint info sections
13327 @item maint info sections
13328 Another command that can give you extra information about program sections
13329 is @code{maint info sections}. In addition to the section information
13330 displayed by @code{info files}, this command displays the flags and file
13331 offset of each section in the executable and core dump files. In addition,
13332 @code{maint info sections} provides the following command options (which
13333 may be arbitrarily combined):
13337 Display sections for all loaded object files, including shared libraries.
13338 @item @var{sections}
13339 Display info only for named @var{sections}.
13340 @item @var{section-flags}
13341 Display info only for sections for which @var{section-flags} are true.
13342 The section flags that @value{GDBN} currently knows about are:
13345 Section will have space allocated in the process when loaded.
13346 Set for all sections except those containing debug information.
13348 Section will be loaded from the file into the child process memory.
13349 Set for pre-initialized code and data, clear for @code{.bss} sections.
13351 Section needs to be relocated before loading.
13353 Section cannot be modified by the child process.
13355 Section contains executable code only.
13357 Section contains data only (no executable code).
13359 Section will reside in ROM.
13361 Section contains data for constructor/destructor lists.
13363 Section is not empty.
13365 An instruction to the linker to not output the section.
13366 @item COFF_SHARED_LIBRARY
13367 A notification to the linker that the section contains
13368 COFF shared library information.
13370 Section contains common symbols.
13373 @kindex set trust-readonly-sections
13374 @cindex read-only sections
13375 @item set trust-readonly-sections on
13376 Tell @value{GDBN} that readonly sections in your object file
13377 really are read-only (i.e.@: that their contents will not change).
13378 In that case, @value{GDBN} can fetch values from these sections
13379 out of the object file, rather than from the target program.
13380 For some targets (notably embedded ones), this can be a significant
13381 enhancement to debugging performance.
13383 The default is off.
13385 @item set trust-readonly-sections off
13386 Tell @value{GDBN} not to trust readonly sections. This means that
13387 the contents of the section might change while the program is running,
13388 and must therefore be fetched from the target when needed.
13390 @item show trust-readonly-sections
13391 Show the current setting of trusting readonly sections.
13394 All file-specifying commands allow both absolute and relative file names
13395 as arguments. @value{GDBN} always converts the file name to an absolute file
13396 name and remembers it that way.
13398 @cindex shared libraries
13399 @anchor{Shared Libraries}
13400 @value{GDBN} supports @sc{gnu}/Linux, MS-Windows, HP-UX, SunOS, SVr4, Irix,
13401 and IBM RS/6000 AIX shared libraries.
13403 On MS-Windows @value{GDBN} must be linked with the Expat library to support
13404 shared libraries. @xref{Expat}.
13406 @value{GDBN} automatically loads symbol definitions from shared libraries
13407 when you use the @code{run} command, or when you examine a core file.
13408 (Before you issue the @code{run} command, @value{GDBN} does not understand
13409 references to a function in a shared library, however---unless you are
13410 debugging a core file).
13412 On HP-UX, if the program loads a library explicitly, @value{GDBN}
13413 automatically loads the symbols at the time of the @code{shl_load} call.
13415 @c FIXME: some @value{GDBN} release may permit some refs to undef
13416 @c FIXME...symbols---eg in a break cmd---assuming they are from a shared
13417 @c FIXME...lib; check this from time to time when updating manual
13419 There are times, however, when you may wish to not automatically load
13420 symbol definitions from shared libraries, such as when they are
13421 particularly large or there are many of them.
13423 To control the automatic loading of shared library symbols, use the
13427 @kindex set auto-solib-add
13428 @item set auto-solib-add @var{mode}
13429 If @var{mode} is @code{on}, symbols from all shared object libraries
13430 will be loaded automatically when the inferior begins execution, you
13431 attach to an independently started inferior, or when the dynamic linker
13432 informs @value{GDBN} that a new library has been loaded. If @var{mode}
13433 is @code{off}, symbols must be loaded manually, using the
13434 @code{sharedlibrary} command. The default value is @code{on}.
13436 @cindex memory used for symbol tables
13437 If your program uses lots of shared libraries with debug info that
13438 takes large amounts of memory, you can decrease the @value{GDBN}
13439 memory footprint by preventing it from automatically loading the
13440 symbols from shared libraries. To that end, type @kbd{set
13441 auto-solib-add off} before running the inferior, then load each
13442 library whose debug symbols you do need with @kbd{sharedlibrary
13443 @var{regexp}}, where @var{regexp} is a regular expression that matches
13444 the libraries whose symbols you want to be loaded.
13446 @kindex show auto-solib-add
13447 @item show auto-solib-add
13448 Display the current autoloading mode.
13451 @cindex load shared library
13452 To explicitly load shared library symbols, use the @code{sharedlibrary}
13456 @kindex info sharedlibrary
13459 @itemx info sharedlibrary
13460 Print the names of the shared libraries which are currently loaded.
13462 @kindex sharedlibrary
13464 @item sharedlibrary @var{regex}
13465 @itemx share @var{regex}
13466 Load shared object library symbols for files matching a
13467 Unix regular expression.
13468 As with files loaded automatically, it only loads shared libraries
13469 required by your program for a core file or after typing @code{run}. If
13470 @var{regex} is omitted all shared libraries required by your program are
13473 @item nosharedlibrary
13474 @kindex nosharedlibrary
13475 @cindex unload symbols from shared libraries
13476 Unload all shared object library symbols. This discards all symbols
13477 that have been loaded from all shared libraries. Symbols from shared
13478 libraries that were loaded by explicit user requests are not
13482 Sometimes you may wish that @value{GDBN} stops and gives you control
13483 when any of shared library events happen. Use the @code{set
13484 stop-on-solib-events} command for this:
13487 @item set stop-on-solib-events
13488 @kindex set stop-on-solib-events
13489 This command controls whether @value{GDBN} should give you control
13490 when the dynamic linker notifies it about some shared library event.
13491 The most common event of interest is loading or unloading of a new
13494 @item show stop-on-solib-events
13495 @kindex show stop-on-solib-events
13496 Show whether @value{GDBN} stops and gives you control when shared
13497 library events happen.
13500 Shared libraries are also supported in many cross or remote debugging
13501 configurations. @value{GDBN} needs to have access to the target's libraries;
13502 this can be accomplished either by providing copies of the libraries
13503 on the host system, or by asking @value{GDBN} to automatically retrieve the
13504 libraries from the target. If copies of the target libraries are
13505 provided, they need to be the same as the target libraries, although the
13506 copies on the target can be stripped as long as the copies on the host are
13509 @cindex where to look for shared libraries
13510 For remote debugging, you need to tell @value{GDBN} where the target
13511 libraries are, so that it can load the correct copies---otherwise, it
13512 may try to load the host's libraries. @value{GDBN} has two variables
13513 to specify the search directories for target libraries.
13516 @cindex prefix for shared library file names
13517 @cindex system root, alternate
13518 @kindex set solib-absolute-prefix
13519 @kindex set sysroot
13520 @item set sysroot @var{path}
13521 Use @var{path} as the system root for the program being debugged. Any
13522 absolute shared library paths will be prefixed with @var{path}; many
13523 runtime loaders store the absolute paths to the shared library in the
13524 target program's memory. If you use @code{set sysroot} to find shared
13525 libraries, they need to be laid out in the same way that they are on
13526 the target, with e.g.@: a @file{/lib} and @file{/usr/lib} hierarchy
13529 If @var{path} starts with the sequence @file{remote:}, @value{GDBN} will
13530 retrieve the target libraries from the remote system. This is only
13531 supported when using a remote target that supports the @code{remote get}
13532 command (@pxref{File Transfer,,Sending files to a remote system}).
13533 The part of @var{path} following the initial @file{remote:}
13534 (if present) is used as system root prefix on the remote file system.
13535 @footnote{If you want to specify a local system root using a directory
13536 that happens to be named @file{remote:}, you need to use some equivalent
13537 variant of the name like @file{./remote:}.}
13539 The @code{set solib-absolute-prefix} command is an alias for @code{set
13542 @cindex default system root
13543 @cindex @samp{--with-sysroot}
13544 You can set the default system root by using the configure-time
13545 @samp{--with-sysroot} option. If the system root is inside
13546 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
13547 @samp{--exec-prefix}), then the default system root will be updated
13548 automatically if the installed @value{GDBN} is moved to a new
13551 @kindex show sysroot
13553 Display the current shared library prefix.
13555 @kindex set solib-search-path
13556 @item set solib-search-path @var{path}
13557 If this variable is set, @var{path} is a colon-separated list of
13558 directories to search for shared libraries. @samp{solib-search-path}
13559 is used after @samp{sysroot} fails to locate the library, or if the
13560 path to the library is relative instead of absolute. If you want to
13561 use @samp{solib-search-path} instead of @samp{sysroot}, be sure to set
13562 @samp{sysroot} to a nonexistent directory to prevent @value{GDBN} from
13563 finding your host's libraries. @samp{sysroot} is preferred; setting
13564 it to a nonexistent directory may interfere with automatic loading
13565 of shared library symbols.
13567 @kindex show solib-search-path
13568 @item show solib-search-path
13569 Display the current shared library search path.
13573 @node Separate Debug Files
13574 @section Debugging Information in Separate Files
13575 @cindex separate debugging information files
13576 @cindex debugging information in separate files
13577 @cindex @file{.debug} subdirectories
13578 @cindex debugging information directory, global
13579 @cindex global debugging information directory
13580 @cindex build ID, and separate debugging files
13581 @cindex @file{.build-id} directory
13583 @value{GDBN} allows you to put a program's debugging information in a
13584 file separate from the executable itself, in a way that allows
13585 @value{GDBN} to find and load the debugging information automatically.
13586 Since debugging information can be very large---sometimes larger
13587 than the executable code itself---some systems distribute debugging
13588 information for their executables in separate files, which users can
13589 install only when they need to debug a problem.
13591 @value{GDBN} supports two ways of specifying the separate debug info
13596 The executable contains a @dfn{debug link} that specifies the name of
13597 the separate debug info file. The separate debug file's name is
13598 usually @file{@var{executable}.debug}, where @var{executable} is the
13599 name of the corresponding executable file without leading directories
13600 (e.g., @file{ls.debug} for @file{/usr/bin/ls}). In addition, the
13601 debug link specifies a CRC32 checksum for the debug file, which
13602 @value{GDBN} uses to validate that the executable and the debug file
13603 came from the same build.
13606 The executable contains a @dfn{build ID}, a unique bit string that is
13607 also present in the corresponding debug info file. (This is supported
13608 only on some operating systems, notably those which use the ELF format
13609 for binary files and the @sc{gnu} Binutils.) For more details about
13610 this feature, see the description of the @option{--build-id}
13611 command-line option in @ref{Options, , Command Line Options, ld.info,
13612 The GNU Linker}. The debug info file's name is not specified
13613 explicitly by the build ID, but can be computed from the build ID, see
13617 Depending on the way the debug info file is specified, @value{GDBN}
13618 uses two different methods of looking for the debug file:
13622 For the ``debug link'' method, @value{GDBN} looks up the named file in
13623 the directory of the executable file, then in a subdirectory of that
13624 directory named @file{.debug}, and finally under the global debug
13625 directory, in a subdirectory whose name is identical to the leading
13626 directories of the executable's absolute file name.
13629 For the ``build ID'' method, @value{GDBN} looks in the
13630 @file{.build-id} subdirectory of the global debug directory for a file
13631 named @file{@var{nn}/@var{nnnnnnnn}.debug}, where @var{nn} are the
13632 first 2 hex characters of the build ID bit string, and @var{nnnnnnnn}
13633 are the rest of the bit string. (Real build ID strings are 32 or more
13634 hex characters, not 10.)
13637 So, for example, suppose you ask @value{GDBN} to debug
13638 @file{/usr/bin/ls}, which has a debug link that specifies the
13639 file @file{ls.debug}, and a build ID whose value in hex is
13640 @code{abcdef1234}. If the global debug directory is
13641 @file{/usr/lib/debug}, then @value{GDBN} will look for the following
13642 debug information files, in the indicated order:
13646 @file{/usr/lib/debug/.build-id/ab/cdef1234.debug}
13648 @file{/usr/bin/ls.debug}
13650 @file{/usr/bin/.debug/ls.debug}
13652 @file{/usr/lib/debug/usr/bin/ls.debug}.
13655 You can set the global debugging info directory's name, and view the
13656 name @value{GDBN} is currently using.
13660 @kindex set debug-file-directory
13661 @item set debug-file-directory @var{directory}
13662 Set the directory which @value{GDBN} searches for separate debugging
13663 information files to @var{directory}.
13665 @kindex show debug-file-directory
13666 @item show debug-file-directory
13667 Show the directory @value{GDBN} searches for separate debugging
13672 @cindex @code{.gnu_debuglink} sections
13673 @cindex debug link sections
13674 A debug link is a special section of the executable file named
13675 @code{.gnu_debuglink}. The section must contain:
13679 A filename, with any leading directory components removed, followed by
13682 zero to three bytes of padding, as needed to reach the next four-byte
13683 boundary within the section, and
13685 a four-byte CRC checksum, stored in the same endianness used for the
13686 executable file itself. The checksum is computed on the debugging
13687 information file's full contents by the function given below, passing
13688 zero as the @var{crc} argument.
13691 Any executable file format can carry a debug link, as long as it can
13692 contain a section named @code{.gnu_debuglink} with the contents
13695 @cindex @code{.note.gnu.build-id} sections
13696 @cindex build ID sections
13697 The build ID is a special section in the executable file (and in other
13698 ELF binary files that @value{GDBN} may consider). This section is
13699 often named @code{.note.gnu.build-id}, but that name is not mandatory.
13700 It contains unique identification for the built files---the ID remains
13701 the same across multiple builds of the same build tree. The default
13702 algorithm SHA1 produces 160 bits (40 hexadecimal characters) of the
13703 content for the build ID string. The same section with an identical
13704 value is present in the original built binary with symbols, in its
13705 stripped variant, and in the separate debugging information file.
13707 The debugging information file itself should be an ordinary
13708 executable, containing a full set of linker symbols, sections, and
13709 debugging information. The sections of the debugging information file
13710 should have the same names, addresses, and sizes as the original file,
13711 but they need not contain any data---much like a @code{.bss} section
13712 in an ordinary executable.
13714 The @sc{gnu} binary utilities (Binutils) package includes the
13715 @samp{objcopy} utility that can produce
13716 the separated executable / debugging information file pairs using the
13717 following commands:
13720 @kbd{objcopy --only-keep-debug foo foo.debug}
13725 These commands remove the debugging
13726 information from the executable file @file{foo} and place it in the file
13727 @file{foo.debug}. You can use the first, second or both methods to link the
13732 The debug link method needs the following additional command to also leave
13733 behind a debug link in @file{foo}:
13736 @kbd{objcopy --add-gnu-debuglink=foo.debug foo}
13739 Ulrich Drepper's @file{elfutils} package, starting with version 0.53, contains
13740 a version of the @code{strip} command such that the command @kbd{strip foo -f
13741 foo.debug} has the same functionality as the two @code{objcopy} commands and
13742 the @code{ln -s} command above, together.
13745 Build ID gets embedded into the main executable using @code{ld --build-id} or
13746 the @value{NGCC} counterpart @code{gcc -Wl,--build-id}. Build ID support plus
13747 compatibility fixes for debug files separation are present in @sc{gnu} binary
13748 utilities (Binutils) package since version 2.18.
13753 Since there are many different ways to compute CRC's for the debug
13754 link (different polynomials, reversals, byte ordering, etc.), the
13755 simplest way to describe the CRC used in @code{.gnu_debuglink}
13756 sections is to give the complete code for a function that computes it:
13758 @kindex gnu_debuglink_crc32
13761 gnu_debuglink_crc32 (unsigned long crc,
13762 unsigned char *buf, size_t len)
13764 static const unsigned long crc32_table[256] =
13766 0x00000000, 0x77073096, 0xee0e612c, 0x990951ba, 0x076dc419,
13767 0x706af48f, 0xe963a535, 0x9e6495a3, 0x0edb8832, 0x79dcb8a4,
13768 0xe0d5e91e, 0x97d2d988, 0x09b64c2b, 0x7eb17cbd, 0xe7b82d07,
13769 0x90bf1d91, 0x1db71064, 0x6ab020f2, 0xf3b97148, 0x84be41de,
13770 0x1adad47d, 0x6ddde4eb, 0xf4d4b551, 0x83d385c7, 0x136c9856,
13771 0x646ba8c0, 0xfd62f97a, 0x8a65c9ec, 0x14015c4f, 0x63066cd9,
13772 0xfa0f3d63, 0x8d080df5, 0x3b6e20c8, 0x4c69105e, 0xd56041e4,
13773 0xa2677172, 0x3c03e4d1, 0x4b04d447, 0xd20d85fd, 0xa50ab56b,
13774 0x35b5a8fa, 0x42b2986c, 0xdbbbc9d6, 0xacbcf940, 0x32d86ce3,
13775 0x45df5c75, 0xdcd60dcf, 0xabd13d59, 0x26d930ac, 0x51de003a,
13776 0xc8d75180, 0xbfd06116, 0x21b4f4b5, 0x56b3c423, 0xcfba9599,
13777 0xb8bda50f, 0x2802b89e, 0x5f058808, 0xc60cd9b2, 0xb10be924,
13778 0x2f6f7c87, 0x58684c11, 0xc1611dab, 0xb6662d3d, 0x76dc4190,
13779 0x01db7106, 0x98d220bc, 0xefd5102a, 0x71b18589, 0x06b6b51f,
13780 0x9fbfe4a5, 0xe8b8d433, 0x7807c9a2, 0x0f00f934, 0x9609a88e,
13781 0xe10e9818, 0x7f6a0dbb, 0x086d3d2d, 0x91646c97, 0xe6635c01,
13782 0x6b6b51f4, 0x1c6c6162, 0x856530d8, 0xf262004e, 0x6c0695ed,
13783 0x1b01a57b, 0x8208f4c1, 0xf50fc457, 0x65b0d9c6, 0x12b7e950,
13784 0x8bbeb8ea, 0xfcb9887c, 0x62dd1ddf, 0x15da2d49, 0x8cd37cf3,
13785 0xfbd44c65, 0x4db26158, 0x3ab551ce, 0xa3bc0074, 0xd4bb30e2,
13786 0x4adfa541, 0x3dd895d7, 0xa4d1c46d, 0xd3d6f4fb, 0x4369e96a,
13787 0x346ed9fc, 0xad678846, 0xda60b8d0, 0x44042d73, 0x33031de5,
13788 0xaa0a4c5f, 0xdd0d7cc9, 0x5005713c, 0x270241aa, 0xbe0b1010,
13789 0xc90c2086, 0x5768b525, 0x206f85b3, 0xb966d409, 0xce61e49f,
13790 0x5edef90e, 0x29d9c998, 0xb0d09822, 0xc7d7a8b4, 0x59b33d17,
13791 0x2eb40d81, 0xb7bd5c3b, 0xc0ba6cad, 0xedb88320, 0x9abfb3b6,
13792 0x03b6e20c, 0x74b1d29a, 0xead54739, 0x9dd277af, 0x04db2615,
13793 0x73dc1683, 0xe3630b12, 0x94643b84, 0x0d6d6a3e, 0x7a6a5aa8,
13794 0xe40ecf0b, 0x9309ff9d, 0x0a00ae27, 0x7d079eb1, 0xf00f9344,
13795 0x8708a3d2, 0x1e01f268, 0x6906c2fe, 0xf762575d, 0x806567cb,
13796 0x196c3671, 0x6e6b06e7, 0xfed41b76, 0x89d32be0, 0x10da7a5a,
13797 0x67dd4acc, 0xf9b9df6f, 0x8ebeeff9, 0x17b7be43, 0x60b08ed5,
13798 0xd6d6a3e8, 0xa1d1937e, 0x38d8c2c4, 0x4fdff252, 0xd1bb67f1,
13799 0xa6bc5767, 0x3fb506dd, 0x48b2364b, 0xd80d2bda, 0xaf0a1b4c,
13800 0x36034af6, 0x41047a60, 0xdf60efc3, 0xa867df55, 0x316e8eef,
13801 0x4669be79, 0xcb61b38c, 0xbc66831a, 0x256fd2a0, 0x5268e236,
13802 0xcc0c7795, 0xbb0b4703, 0x220216b9, 0x5505262f, 0xc5ba3bbe,
13803 0xb2bd0b28, 0x2bb45a92, 0x5cb36a04, 0xc2d7ffa7, 0xb5d0cf31,
13804 0x2cd99e8b, 0x5bdeae1d, 0x9b64c2b0, 0xec63f226, 0x756aa39c,
13805 0x026d930a, 0x9c0906a9, 0xeb0e363f, 0x72076785, 0x05005713,
13806 0x95bf4a82, 0xe2b87a14, 0x7bb12bae, 0x0cb61b38, 0x92d28e9b,
13807 0xe5d5be0d, 0x7cdcefb7, 0x0bdbdf21, 0x86d3d2d4, 0xf1d4e242,
13808 0x68ddb3f8, 0x1fda836e, 0x81be16cd, 0xf6b9265b, 0x6fb077e1,
13809 0x18b74777, 0x88085ae6, 0xff0f6a70, 0x66063bca, 0x11010b5c,
13810 0x8f659eff, 0xf862ae69, 0x616bffd3, 0x166ccf45, 0xa00ae278,
13811 0xd70dd2ee, 0x4e048354, 0x3903b3c2, 0xa7672661, 0xd06016f7,
13812 0x4969474d, 0x3e6e77db, 0xaed16a4a, 0xd9d65adc, 0x40df0b66,
13813 0x37d83bf0, 0xa9bcae53, 0xdebb9ec5, 0x47b2cf7f, 0x30b5ffe9,
13814 0xbdbdf21c, 0xcabac28a, 0x53b39330, 0x24b4a3a6, 0xbad03605,
13815 0xcdd70693, 0x54de5729, 0x23d967bf, 0xb3667a2e, 0xc4614ab8,
13816 0x5d681b02, 0x2a6f2b94, 0xb40bbe37, 0xc30c8ea1, 0x5a05df1b,
13819 unsigned char *end;
13821 crc = ~crc & 0xffffffff;
13822 for (end = buf + len; buf < end; ++buf)
13823 crc = crc32_table[(crc ^ *buf) & 0xff] ^ (crc >> 8);
13824 return ~crc & 0xffffffff;
13829 This computation does not apply to the ``build ID'' method.
13832 @node Symbol Errors
13833 @section Errors Reading Symbol Files
13835 While reading a symbol file, @value{GDBN} occasionally encounters problems,
13836 such as symbol types it does not recognize, or known bugs in compiler
13837 output. By default, @value{GDBN} does not notify you of such problems, since
13838 they are relatively common and primarily of interest to people
13839 debugging compilers. If you are interested in seeing information
13840 about ill-constructed symbol tables, you can either ask @value{GDBN} to print
13841 only one message about each such type of problem, no matter how many
13842 times the problem occurs; or you can ask @value{GDBN} to print more messages,
13843 to see how many times the problems occur, with the @code{set
13844 complaints} command (@pxref{Messages/Warnings, ,Optional Warnings and
13847 The messages currently printed, and their meanings, include:
13850 @item inner block not inside outer block in @var{symbol}
13852 The symbol information shows where symbol scopes begin and end
13853 (such as at the start of a function or a block of statements). This
13854 error indicates that an inner scope block is not fully contained
13855 in its outer scope blocks.
13857 @value{GDBN} circumvents the problem by treating the inner block as if it had
13858 the same scope as the outer block. In the error message, @var{symbol}
13859 may be shown as ``@code{(don't know)}'' if the outer block is not a
13862 @item block at @var{address} out of order
13864 The symbol information for symbol scope blocks should occur in
13865 order of increasing addresses. This error indicates that it does not
13868 @value{GDBN} does not circumvent this problem, and has trouble
13869 locating symbols in the source file whose symbols it is reading. (You
13870 can often determine what source file is affected by specifying
13871 @code{set verbose on}. @xref{Messages/Warnings, ,Optional Warnings and
13874 @item bad block start address patched
13876 The symbol information for a symbol scope block has a start address
13877 smaller than the address of the preceding source line. This is known
13878 to occur in the SunOS 4.1.1 (and earlier) C compiler.
13880 @value{GDBN} circumvents the problem by treating the symbol scope block as
13881 starting on the previous source line.
13883 @item bad string table offset in symbol @var{n}
13886 Symbol number @var{n} contains a pointer into the string table which is
13887 larger than the size of the string table.
13889 @value{GDBN} circumvents the problem by considering the symbol to have the
13890 name @code{foo}, which may cause other problems if many symbols end up
13893 @item unknown symbol type @code{0x@var{nn}}
13895 The symbol information contains new data types that @value{GDBN} does
13896 not yet know how to read. @code{0x@var{nn}} is the symbol type of the
13897 uncomprehended information, in hexadecimal.
13899 @value{GDBN} circumvents the error by ignoring this symbol information.
13900 This usually allows you to debug your program, though certain symbols
13901 are not accessible. If you encounter such a problem and feel like
13902 debugging it, you can debug @code{@value{GDBP}} with itself, breakpoint
13903 on @code{complain}, then go up to the function @code{read_dbx_symtab}
13904 and examine @code{*bufp} to see the symbol.
13906 @item stub type has NULL name
13908 @value{GDBN} could not find the full definition for a struct or class.
13910 @item const/volatile indicator missing (ok if using g++ v1.x), got@dots{}
13911 The symbol information for a C@t{++} member function is missing some
13912 information that recent versions of the compiler should have output for
13915 @item info mismatch between compiler and debugger
13917 @value{GDBN} could not parse a type specification output by the compiler.
13922 @section GDB Data Files
13924 @cindex prefix for data files
13925 @value{GDBN} will sometimes read an auxiliary data file. These files
13926 are kept in a directory known as the @dfn{data directory}.
13928 You can set the data directory's name, and view the name @value{GDBN}
13929 is currently using.
13932 @kindex set data-directory
13933 @item set data-directory @var{directory}
13934 Set the directory which @value{GDBN} searches for auxiliary data files
13935 to @var{directory}.
13937 @kindex show data-directory
13938 @item show data-directory
13939 Show the directory @value{GDBN} searches for auxiliary data files.
13942 @cindex default data directory
13943 @cindex @samp{--with-gdb-datadir}
13944 You can set the default data directory by using the configure-time
13945 @samp{--with-gdb-datadir} option. If the data directory is inside
13946 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
13947 @samp{--exec-prefix}), then the default data directory will be updated
13948 automatically if the installed @value{GDBN} is moved to a new
13952 @chapter Specifying a Debugging Target
13954 @cindex debugging target
13955 A @dfn{target} is the execution environment occupied by your program.
13957 Often, @value{GDBN} runs in the same host environment as your program;
13958 in that case, the debugging target is specified as a side effect when
13959 you use the @code{file} or @code{core} commands. When you need more
13960 flexibility---for example, running @value{GDBN} on a physically separate
13961 host, or controlling a standalone system over a serial port or a
13962 realtime system over a TCP/IP connection---you can use the @code{target}
13963 command to specify one of the target types configured for @value{GDBN}
13964 (@pxref{Target Commands, ,Commands for Managing Targets}).
13966 @cindex target architecture
13967 It is possible to build @value{GDBN} for several different @dfn{target
13968 architectures}. When @value{GDBN} is built like that, you can choose
13969 one of the available architectures with the @kbd{set architecture}
13973 @kindex set architecture
13974 @kindex show architecture
13975 @item set architecture @var{arch}
13976 This command sets the current target architecture to @var{arch}. The
13977 value of @var{arch} can be @code{"auto"}, in addition to one of the
13978 supported architectures.
13980 @item show architecture
13981 Show the current target architecture.
13983 @item set processor
13985 @kindex set processor
13986 @kindex show processor
13987 These are alias commands for, respectively, @code{set architecture}
13988 and @code{show architecture}.
13992 * Active Targets:: Active targets
13993 * Target Commands:: Commands for managing targets
13994 * Byte Order:: Choosing target byte order
13997 @node Active Targets
13998 @section Active Targets
14000 @cindex stacking targets
14001 @cindex active targets
14002 @cindex multiple targets
14004 There are three classes of targets: processes, core files, and
14005 executable files. @value{GDBN} can work concurrently on up to three
14006 active targets, one in each class. This allows you to (for example)
14007 start a process and inspect its activity without abandoning your work on
14010 For example, if you execute @samp{gdb a.out}, then the executable file
14011 @code{a.out} is the only active target. If you designate a core file as
14012 well---presumably from a prior run that crashed and coredumped---then
14013 @value{GDBN} has two active targets and uses them in tandem, looking
14014 first in the corefile target, then in the executable file, to satisfy
14015 requests for memory addresses. (Typically, these two classes of target
14016 are complementary, since core files contain only a program's
14017 read-write memory---variables and so on---plus machine status, while
14018 executable files contain only the program text and initialized data.)
14020 When you type @code{run}, your executable file becomes an active process
14021 target as well. When a process target is active, all @value{GDBN}
14022 commands requesting memory addresses refer to that target; addresses in
14023 an active core file or executable file target are obscured while the
14024 process target is active.
14026 Use the @code{core-file} and @code{exec-file} commands to select a new
14027 core file or executable target (@pxref{Files, ,Commands to Specify
14028 Files}). To specify as a target a process that is already running, use
14029 the @code{attach} command (@pxref{Attach, ,Debugging an Already-running
14032 @node Target Commands
14033 @section Commands for Managing Targets
14036 @item target @var{type} @var{parameters}
14037 Connects the @value{GDBN} host environment to a target machine or
14038 process. A target is typically a protocol for talking to debugging
14039 facilities. You use the argument @var{type} to specify the type or
14040 protocol of the target machine.
14042 Further @var{parameters} are interpreted by the target protocol, but
14043 typically include things like device names or host names to connect
14044 with, process numbers, and baud rates.
14046 The @code{target} command does not repeat if you press @key{RET} again
14047 after executing the command.
14049 @kindex help target
14051 Displays the names of all targets available. To display targets
14052 currently selected, use either @code{info target} or @code{info files}
14053 (@pxref{Files, ,Commands to Specify Files}).
14055 @item help target @var{name}
14056 Describe a particular target, including any parameters necessary to
14059 @kindex set gnutarget
14060 @item set gnutarget @var{args}
14061 @value{GDBN} uses its own library BFD to read your files. @value{GDBN}
14062 knows whether it is reading an @dfn{executable},
14063 a @dfn{core}, or a @dfn{.o} file; however, you can specify the file format
14064 with the @code{set gnutarget} command. Unlike most @code{target} commands,
14065 with @code{gnutarget} the @code{target} refers to a program, not a machine.
14068 @emph{Warning:} To specify a file format with @code{set gnutarget},
14069 you must know the actual BFD name.
14073 @xref{Files, , Commands to Specify Files}.
14075 @kindex show gnutarget
14076 @item show gnutarget
14077 Use the @code{show gnutarget} command to display what file format
14078 @code{gnutarget} is set to read. If you have not set @code{gnutarget},
14079 @value{GDBN} will determine the file format for each file automatically,
14080 and @code{show gnutarget} displays @samp{The current BDF target is "auto"}.
14083 @cindex common targets
14084 Here are some common targets (available, or not, depending on the GDB
14089 @item target exec @var{program}
14090 @cindex executable file target
14091 An executable file. @samp{target exec @var{program}} is the same as
14092 @samp{exec-file @var{program}}.
14094 @item target core @var{filename}
14095 @cindex core dump file target
14096 A core dump file. @samp{target core @var{filename}} is the same as
14097 @samp{core-file @var{filename}}.
14099 @item target remote @var{medium}
14100 @cindex remote target
14101 A remote system connected to @value{GDBN} via a serial line or network
14102 connection. This command tells @value{GDBN} to use its own remote
14103 protocol over @var{medium} for debugging. @xref{Remote Debugging}.
14105 For example, if you have a board connected to @file{/dev/ttya} on the
14106 machine running @value{GDBN}, you could say:
14109 target remote /dev/ttya
14112 @code{target remote} supports the @code{load} command. This is only
14113 useful if you have some other way of getting the stub to the target
14114 system, and you can put it somewhere in memory where it won't get
14115 clobbered by the download.
14118 @cindex built-in simulator target
14119 Builtin CPU simulator. @value{GDBN} includes simulators for most architectures.
14127 works; however, you cannot assume that a specific memory map, device
14128 drivers, or even basic I/O is available, although some simulators do
14129 provide these. For info about any processor-specific simulator details,
14130 see the appropriate section in @ref{Embedded Processors, ,Embedded
14135 Some configurations may include these targets as well:
14139 @item target nrom @var{dev}
14140 @cindex NetROM ROM emulator target
14141 NetROM ROM emulator. This target only supports downloading.
14145 Different targets are available on different configurations of @value{GDBN};
14146 your configuration may have more or fewer targets.
14148 Many remote targets require you to download the executable's code once
14149 you've successfully established a connection. You may wish to control
14150 various aspects of this process.
14155 @kindex set hash@r{, for remote monitors}
14156 @cindex hash mark while downloading
14157 This command controls whether a hash mark @samp{#} is displayed while
14158 downloading a file to the remote monitor. If on, a hash mark is
14159 displayed after each S-record is successfully downloaded to the
14163 @kindex show hash@r{, for remote monitors}
14164 Show the current status of displaying the hash mark.
14166 @item set debug monitor
14167 @kindex set debug monitor
14168 @cindex display remote monitor communications
14169 Enable or disable display of communications messages between
14170 @value{GDBN} and the remote monitor.
14172 @item show debug monitor
14173 @kindex show debug monitor
14174 Show the current status of displaying communications between
14175 @value{GDBN} and the remote monitor.
14180 @kindex load @var{filename}
14181 @item load @var{filename}
14183 Depending on what remote debugging facilities are configured into
14184 @value{GDBN}, the @code{load} command may be available. Where it exists, it
14185 is meant to make @var{filename} (an executable) available for debugging
14186 on the remote system---by downloading, or dynamic linking, for example.
14187 @code{load} also records the @var{filename} symbol table in @value{GDBN}, like
14188 the @code{add-symbol-file} command.
14190 If your @value{GDBN} does not have a @code{load} command, attempting to
14191 execute it gets the error message ``@code{You can't do that when your
14192 target is @dots{}}''
14194 The file is loaded at whatever address is specified in the executable.
14195 For some object file formats, you can specify the load address when you
14196 link the program; for other formats, like a.out, the object file format
14197 specifies a fixed address.
14198 @c FIXME! This would be a good place for an xref to the GNU linker doc.
14200 Depending on the remote side capabilities, @value{GDBN} may be able to
14201 load programs into flash memory.
14203 @code{load} does not repeat if you press @key{RET} again after using it.
14207 @section Choosing Target Byte Order
14209 @cindex choosing target byte order
14210 @cindex target byte order
14212 Some types of processors, such as the MIPS, PowerPC, and Renesas SH,
14213 offer the ability to run either big-endian or little-endian byte
14214 orders. Usually the executable or symbol will include a bit to
14215 designate the endian-ness, and you will not need to worry about
14216 which to use. However, you may still find it useful to adjust
14217 @value{GDBN}'s idea of processor endian-ness manually.
14221 @item set endian big
14222 Instruct @value{GDBN} to assume the target is big-endian.
14224 @item set endian little
14225 Instruct @value{GDBN} to assume the target is little-endian.
14227 @item set endian auto
14228 Instruct @value{GDBN} to use the byte order associated with the
14232 Display @value{GDBN}'s current idea of the target byte order.
14236 Note that these commands merely adjust interpretation of symbolic
14237 data on the host, and that they have absolutely no effect on the
14241 @node Remote Debugging
14242 @chapter Debugging Remote Programs
14243 @cindex remote debugging
14245 If you are trying to debug a program running on a machine that cannot run
14246 @value{GDBN} in the usual way, it is often useful to use remote debugging.
14247 For example, you might use remote debugging on an operating system kernel,
14248 or on a small system which does not have a general purpose operating system
14249 powerful enough to run a full-featured debugger.
14251 Some configurations of @value{GDBN} have special serial or TCP/IP interfaces
14252 to make this work with particular debugging targets. In addition,
14253 @value{GDBN} comes with a generic serial protocol (specific to @value{GDBN},
14254 but not specific to any particular target system) which you can use if you
14255 write the remote stubs---the code that runs on the remote system to
14256 communicate with @value{GDBN}.
14258 Other remote targets may be available in your
14259 configuration of @value{GDBN}; use @code{help target} to list them.
14262 * Connecting:: Connecting to a remote target
14263 * File Transfer:: Sending files to a remote system
14264 * Server:: Using the gdbserver program
14265 * Remote Configuration:: Remote configuration
14266 * Remote Stub:: Implementing a remote stub
14270 @section Connecting to a Remote Target
14272 On the @value{GDBN} host machine, you will need an unstripped copy of
14273 your program, since @value{GDBN} needs symbol and debugging information.
14274 Start up @value{GDBN} as usual, using the name of the local copy of your
14275 program as the first argument.
14277 @cindex @code{target remote}
14278 @value{GDBN} can communicate with the target over a serial line, or
14279 over an @acronym{IP} network using @acronym{TCP} or @acronym{UDP}. In
14280 each case, @value{GDBN} uses the same protocol for debugging your
14281 program; only the medium carrying the debugging packets varies. The
14282 @code{target remote} command establishes a connection to the target.
14283 Its arguments indicate which medium to use:
14287 @item target remote @var{serial-device}
14288 @cindex serial line, @code{target remote}
14289 Use @var{serial-device} to communicate with the target. For example,
14290 to use a serial line connected to the device named @file{/dev/ttyb}:
14293 target remote /dev/ttyb
14296 If you're using a serial line, you may want to give @value{GDBN} the
14297 @w{@samp{--baud}} option, or use the @code{set remotebaud} command
14298 (@pxref{Remote Configuration, set remotebaud}) before the
14299 @code{target} command.
14301 @item target remote @code{@var{host}:@var{port}}
14302 @itemx target remote @code{tcp:@var{host}:@var{port}}
14303 @cindex @acronym{TCP} port, @code{target remote}
14304 Debug using a @acronym{TCP} connection to @var{port} on @var{host}.
14305 The @var{host} may be either a host name or a numeric @acronym{IP}
14306 address; @var{port} must be a decimal number. The @var{host} could be
14307 the target machine itself, if it is directly connected to the net, or
14308 it might be a terminal server which in turn has a serial line to the
14311 For example, to connect to port 2828 on a terminal server named
14315 target remote manyfarms:2828
14318 If your remote target is actually running on the same machine as your
14319 debugger session (e.g.@: a simulator for your target running on the
14320 same host), you can omit the hostname. For example, to connect to
14321 port 1234 on your local machine:
14324 target remote :1234
14328 Note that the colon is still required here.
14330 @item target remote @code{udp:@var{host}:@var{port}}
14331 @cindex @acronym{UDP} port, @code{target remote}
14332 Debug using @acronym{UDP} packets to @var{port} on @var{host}. For example, to
14333 connect to @acronym{UDP} port 2828 on a terminal server named @code{manyfarms}:
14336 target remote udp:manyfarms:2828
14339 When using a @acronym{UDP} connection for remote debugging, you should
14340 keep in mind that the `U' stands for ``Unreliable''. @acronym{UDP}
14341 can silently drop packets on busy or unreliable networks, which will
14342 cause havoc with your debugging session.
14344 @item target remote | @var{command}
14345 @cindex pipe, @code{target remote} to
14346 Run @var{command} in the background and communicate with it using a
14347 pipe. The @var{command} is a shell command, to be parsed and expanded
14348 by the system's command shell, @code{/bin/sh}; it should expect remote
14349 protocol packets on its standard input, and send replies on its
14350 standard output. You could use this to run a stand-alone simulator
14351 that speaks the remote debugging protocol, to make net connections
14352 using programs like @code{ssh}, or for other similar tricks.
14354 If @var{command} closes its standard output (perhaps by exiting),
14355 @value{GDBN} will try to send it a @code{SIGTERM} signal. (If the
14356 program has already exited, this will have no effect.)
14360 Once the connection has been established, you can use all the usual
14361 commands to examine and change data. The remote program is already
14362 running; you can use @kbd{step} and @kbd{continue}, and you do not
14363 need to use @kbd{run}.
14365 @cindex interrupting remote programs
14366 @cindex remote programs, interrupting
14367 Whenever @value{GDBN} is waiting for the remote program, if you type the
14368 interrupt character (often @kbd{Ctrl-c}), @value{GDBN} attempts to stop the
14369 program. This may or may not succeed, depending in part on the hardware
14370 and the serial drivers the remote system uses. If you type the
14371 interrupt character once again, @value{GDBN} displays this prompt:
14374 Interrupted while waiting for the program.
14375 Give up (and stop debugging it)? (y or n)
14378 If you type @kbd{y}, @value{GDBN} abandons the remote debugging session.
14379 (If you decide you want to try again later, you can use @samp{target
14380 remote} again to connect once more.) If you type @kbd{n}, @value{GDBN}
14381 goes back to waiting.
14384 @kindex detach (remote)
14386 When you have finished debugging the remote program, you can use the
14387 @code{detach} command to release it from @value{GDBN} control.
14388 Detaching from the target normally resumes its execution, but the results
14389 will depend on your particular remote stub. After the @code{detach}
14390 command, @value{GDBN} is free to connect to another target.
14394 The @code{disconnect} command behaves like @code{detach}, except that
14395 the target is generally not resumed. It will wait for @value{GDBN}
14396 (this instance or another one) to connect and continue debugging. After
14397 the @code{disconnect} command, @value{GDBN} is again free to connect to
14400 @cindex send command to remote monitor
14401 @cindex extend @value{GDBN} for remote targets
14402 @cindex add new commands for external monitor
14404 @item monitor @var{cmd}
14405 This command allows you to send arbitrary commands directly to the
14406 remote monitor. Since @value{GDBN} doesn't care about the commands it
14407 sends like this, this command is the way to extend @value{GDBN}---you
14408 can add new commands that only the external monitor will understand
14412 @node File Transfer
14413 @section Sending files to a remote system
14414 @cindex remote target, file transfer
14415 @cindex file transfer
14416 @cindex sending files to remote systems
14418 Some remote targets offer the ability to transfer files over the same
14419 connection used to communicate with @value{GDBN}. This is convenient
14420 for targets accessible through other means, e.g.@: @sc{gnu}/Linux systems
14421 running @code{gdbserver} over a network interface. For other targets,
14422 e.g.@: embedded devices with only a single serial port, this may be
14423 the only way to upload or download files.
14425 Not all remote targets support these commands.
14429 @item remote put @var{hostfile} @var{targetfile}
14430 Copy file @var{hostfile} from the host system (the machine running
14431 @value{GDBN}) to @var{targetfile} on the target system.
14434 @item remote get @var{targetfile} @var{hostfile}
14435 Copy file @var{targetfile} from the target system to @var{hostfile}
14436 on the host system.
14438 @kindex remote delete
14439 @item remote delete @var{targetfile}
14440 Delete @var{targetfile} from the target system.
14445 @section Using the @code{gdbserver} Program
14448 @cindex remote connection without stubs
14449 @code{gdbserver} is a control program for Unix-like systems, which
14450 allows you to connect your program with a remote @value{GDBN} via
14451 @code{target remote}---but without linking in the usual debugging stub.
14453 @code{gdbserver} is not a complete replacement for the debugging stubs,
14454 because it requires essentially the same operating-system facilities
14455 that @value{GDBN} itself does. In fact, a system that can run
14456 @code{gdbserver} to connect to a remote @value{GDBN} could also run
14457 @value{GDBN} locally! @code{gdbserver} is sometimes useful nevertheless,
14458 because it is a much smaller program than @value{GDBN} itself. It is
14459 also easier to port than all of @value{GDBN}, so you may be able to get
14460 started more quickly on a new system by using @code{gdbserver}.
14461 Finally, if you develop code for real-time systems, you may find that
14462 the tradeoffs involved in real-time operation make it more convenient to
14463 do as much development work as possible on another system, for example
14464 by cross-compiling. You can use @code{gdbserver} to make a similar
14465 choice for debugging.
14467 @value{GDBN} and @code{gdbserver} communicate via either a serial line
14468 or a TCP connection, using the standard @value{GDBN} remote serial
14472 @emph{Warning:} @code{gdbserver} does not have any built-in security.
14473 Do not run @code{gdbserver} connected to any public network; a
14474 @value{GDBN} connection to @code{gdbserver} provides access to the
14475 target system with the same privileges as the user running
14479 @subsection Running @code{gdbserver}
14480 @cindex arguments, to @code{gdbserver}
14482 Run @code{gdbserver} on the target system. You need a copy of the
14483 program you want to debug, including any libraries it requires.
14484 @code{gdbserver} does not need your program's symbol table, so you can
14485 strip the program if necessary to save space. @value{GDBN} on the host
14486 system does all the symbol handling.
14488 To use the server, you must tell it how to communicate with @value{GDBN};
14489 the name of your program; and the arguments for your program. The usual
14493 target> gdbserver @var{comm} @var{program} [ @var{args} @dots{} ]
14496 @var{comm} is either a device name (to use a serial line) or a TCP
14497 hostname and portnumber. For example, to debug Emacs with the argument
14498 @samp{foo.txt} and communicate with @value{GDBN} over the serial port
14502 target> gdbserver /dev/com1 emacs foo.txt
14505 @code{gdbserver} waits passively for the host @value{GDBN} to communicate
14508 To use a TCP connection instead of a serial line:
14511 target> gdbserver host:2345 emacs foo.txt
14514 The only difference from the previous example is the first argument,
14515 specifying that you are communicating with the host @value{GDBN} via
14516 TCP. The @samp{host:2345} argument means that @code{gdbserver} is to
14517 expect a TCP connection from machine @samp{host} to local TCP port 2345.
14518 (Currently, the @samp{host} part is ignored.) You can choose any number
14519 you want for the port number as long as it does not conflict with any
14520 TCP ports already in use on the target system (for example, @code{23} is
14521 reserved for @code{telnet}).@footnote{If you choose a port number that
14522 conflicts with another service, @code{gdbserver} prints an error message
14523 and exits.} You must use the same port number with the host @value{GDBN}
14524 @code{target remote} command.
14526 @subsubsection Attaching to a Running Program
14528 On some targets, @code{gdbserver} can also attach to running programs.
14529 This is accomplished via the @code{--attach} argument. The syntax is:
14532 target> gdbserver --attach @var{comm} @var{pid}
14535 @var{pid} is the process ID of a currently running process. It isn't necessary
14536 to point @code{gdbserver} at a binary for the running process.
14539 @cindex attach to a program by name
14540 You can debug processes by name instead of process ID if your target has the
14541 @code{pidof} utility:
14544 target> gdbserver --attach @var{comm} `pidof @var{program}`
14547 In case more than one copy of @var{program} is running, or @var{program}
14548 has multiple threads, most versions of @code{pidof} support the
14549 @code{-s} option to only return the first process ID.
14551 @subsubsection Multi-Process Mode for @code{gdbserver}
14552 @cindex gdbserver, multiple processes
14553 @cindex multiple processes with gdbserver
14555 When you connect to @code{gdbserver} using @code{target remote},
14556 @code{gdbserver} debugs the specified program only once. When the
14557 program exits, or you detach from it, @value{GDBN} closes the connection
14558 and @code{gdbserver} exits.
14560 If you connect using @kbd{target extended-remote}, @code{gdbserver}
14561 enters multi-process mode. When the debugged program exits, or you
14562 detach from it, @value{GDBN} stays connected to @code{gdbserver} even
14563 though no program is running. The @code{run} and @code{attach}
14564 commands instruct @code{gdbserver} to run or attach to a new program.
14565 The @code{run} command uses @code{set remote exec-file} (@pxref{set
14566 remote exec-file}) to select the program to run. Command line
14567 arguments are supported, except for wildcard expansion and I/O
14568 redirection (@pxref{Arguments}).
14570 To start @code{gdbserver} without supplying an initial command to run
14571 or process ID to attach, use the @option{--multi} command line option.
14572 Then you can connect using @kbd{target extended-remote} and start
14573 the program you want to debug.
14575 @code{gdbserver} does not automatically exit in multi-process mode.
14576 You can terminate it by using @code{monitor exit}
14577 (@pxref{Monitor Commands for gdbserver}).
14579 @subsubsection Other Command-Line Arguments for @code{gdbserver}
14581 The @option{--debug} option tells @code{gdbserver} to display extra
14582 status information about the debugging process. The
14583 @option{--remote-debug} option tells @code{gdbserver} to display
14584 remote protocol debug output. These options are intended for
14585 @code{gdbserver} development and for bug reports to the developers.
14587 The @option{--wrapper} option specifies a wrapper to launch programs
14588 for debugging. The option should be followed by the name of the
14589 wrapper, then any command-line arguments to pass to the wrapper, then
14590 @kbd{--} indicating the end of the wrapper arguments.
14592 @code{gdbserver} runs the specified wrapper program with a combined
14593 command line including the wrapper arguments, then the name of the
14594 program to debug, then any arguments to the program. The wrapper
14595 runs until it executes your program, and then @value{GDBN} gains control.
14597 You can use any program that eventually calls @code{execve} with
14598 its arguments as a wrapper. Several standard Unix utilities do
14599 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
14600 with @code{exec "$@@"} will also work.
14602 For example, you can use @code{env} to pass an environment variable to
14603 the debugged program, without setting the variable in @code{gdbserver}'s
14607 $ gdbserver --wrapper env LD_PRELOAD=libtest.so -- :2222 ./testprog
14610 @subsection Connecting to @code{gdbserver}
14612 Run @value{GDBN} on the host system.
14614 First make sure you have the necessary symbol files. Load symbols for
14615 your application using the @code{file} command before you connect. Use
14616 @code{set sysroot} to locate target libraries (unless your @value{GDBN}
14617 was compiled with the correct sysroot using @code{--with-sysroot}).
14619 The symbol file and target libraries must exactly match the executable
14620 and libraries on the target, with one exception: the files on the host
14621 system should not be stripped, even if the files on the target system
14622 are. Mismatched or missing files will lead to confusing results
14623 during debugging. On @sc{gnu}/Linux targets, mismatched or missing
14624 files may also prevent @code{gdbserver} from debugging multi-threaded
14627 Connect to your target (@pxref{Connecting,,Connecting to a Remote Target}).
14628 For TCP connections, you must start up @code{gdbserver} prior to using
14629 the @code{target remote} command. Otherwise you may get an error whose
14630 text depends on the host system, but which usually looks something like
14631 @samp{Connection refused}. Don't use the @code{load}
14632 command in @value{GDBN} when using @code{gdbserver}, since the program is
14633 already on the target.
14635 @subsection Monitor Commands for @code{gdbserver}
14636 @cindex monitor commands, for @code{gdbserver}
14637 @anchor{Monitor Commands for gdbserver}
14639 During a @value{GDBN} session using @code{gdbserver}, you can use the
14640 @code{monitor} command to send special requests to @code{gdbserver}.
14641 Here are the available commands.
14645 List the available monitor commands.
14647 @item monitor set debug 0
14648 @itemx monitor set debug 1
14649 Disable or enable general debugging messages.
14651 @item monitor set remote-debug 0
14652 @itemx monitor set remote-debug 1
14653 Disable or enable specific debugging messages associated with the remote
14654 protocol (@pxref{Remote Protocol}).
14657 Tell gdbserver to exit immediately. This command should be followed by
14658 @code{disconnect} to close the debugging session. @code{gdbserver} will
14659 detach from any attached processes and kill any processes it created.
14660 Use @code{monitor exit} to terminate @code{gdbserver} at the end
14661 of a multi-process mode debug session.
14665 @node Remote Configuration
14666 @section Remote Configuration
14669 @kindex show remote
14670 This section documents the configuration options available when
14671 debugging remote programs. For the options related to the File I/O
14672 extensions of the remote protocol, see @ref{system,
14673 system-call-allowed}.
14676 @item set remoteaddresssize @var{bits}
14677 @cindex address size for remote targets
14678 @cindex bits in remote address
14679 Set the maximum size of address in a memory packet to the specified
14680 number of bits. @value{GDBN} will mask off the address bits above
14681 that number, when it passes addresses to the remote target. The
14682 default value is the number of bits in the target's address.
14684 @item show remoteaddresssize
14685 Show the current value of remote address size in bits.
14687 @item set remotebaud @var{n}
14688 @cindex baud rate for remote targets
14689 Set the baud rate for the remote serial I/O to @var{n} baud. The
14690 value is used to set the speed of the serial port used for debugging
14693 @item show remotebaud
14694 Show the current speed of the remote connection.
14696 @item set remotebreak
14697 @cindex interrupt remote programs
14698 @cindex BREAK signal instead of Ctrl-C
14699 @anchor{set remotebreak}
14700 If set to on, @value{GDBN} sends a @code{BREAK} signal to the remote
14701 when you type @kbd{Ctrl-c} to interrupt the program running
14702 on the remote. If set to off, @value{GDBN} sends the @samp{Ctrl-C}
14703 character instead. The default is off, since most remote systems
14704 expect to see @samp{Ctrl-C} as the interrupt signal.
14706 @item show remotebreak
14707 Show whether @value{GDBN} sends @code{BREAK} or @samp{Ctrl-C} to
14708 interrupt the remote program.
14710 @item set remoteflow on
14711 @itemx set remoteflow off
14712 @kindex set remoteflow
14713 Enable or disable hardware flow control (@code{RTS}/@code{CTS})
14714 on the serial port used to communicate to the remote target.
14716 @item show remoteflow
14717 @kindex show remoteflow
14718 Show the current setting of hardware flow control.
14720 @item set remotelogbase @var{base}
14721 Set the base (a.k.a.@: radix) of logging serial protocol
14722 communications to @var{base}. Supported values of @var{base} are:
14723 @code{ascii}, @code{octal}, and @code{hex}. The default is
14726 @item show remotelogbase
14727 Show the current setting of the radix for logging remote serial
14730 @item set remotelogfile @var{file}
14731 @cindex record serial communications on file
14732 Record remote serial communications on the named @var{file}. The
14733 default is not to record at all.
14735 @item show remotelogfile.
14736 Show the current setting of the file name on which to record the
14737 serial communications.
14739 @item set remotetimeout @var{num}
14740 @cindex timeout for serial communications
14741 @cindex remote timeout
14742 Set the timeout limit to wait for the remote target to respond to
14743 @var{num} seconds. The default is 2 seconds.
14745 @item show remotetimeout
14746 Show the current number of seconds to wait for the remote target
14749 @cindex limit hardware breakpoints and watchpoints
14750 @cindex remote target, limit break- and watchpoints
14751 @anchor{set remote hardware-watchpoint-limit}
14752 @anchor{set remote hardware-breakpoint-limit}
14753 @item set remote hardware-watchpoint-limit @var{limit}
14754 @itemx set remote hardware-breakpoint-limit @var{limit}
14755 Restrict @value{GDBN} to using @var{limit} remote hardware breakpoint or
14756 watchpoints. A limit of -1, the default, is treated as unlimited.
14758 @item set remote exec-file @var{filename}
14759 @itemx show remote exec-file
14760 @anchor{set remote exec-file}
14761 @cindex executable file, for remote target
14762 Select the file used for @code{run} with @code{target
14763 extended-remote}. This should be set to a filename valid on the
14764 target system. If it is not set, the target will use a default
14765 filename (e.g.@: the last program run).
14769 @item set tcp auto-retry on
14770 @cindex auto-retry, for remote TCP target
14771 Enable auto-retry for remote TCP connections. This is useful if the remote
14772 debugging agent is launched in parallel with @value{GDBN}; there is a race
14773 condition because the agent may not become ready to accept the connection
14774 before @value{GDBN} attempts to connect. When auto-retry is
14775 enabled, if the initial attempt to connect fails, @value{GDBN} reattempts
14776 to establish the connection using the timeout specified by
14777 @code{set tcp connect-timeout}.
14779 @item set tcp auto-retry off
14780 Do not auto-retry failed TCP connections.
14782 @item show tcp auto-retry
14783 Show the current auto-retry setting.
14785 @item set tcp connect-timeout @var{seconds}
14786 @cindex connection timeout, for remote TCP target
14787 @cindex timeout, for remote target connection
14788 Set the timeout for establishing a TCP connection to the remote target to
14789 @var{seconds}. The timeout affects both polling to retry failed connections
14790 (enabled by @code{set tcp auto-retry on}) and waiting for connections
14791 that are merely slow to complete, and represents an approximate cumulative
14794 @item show tcp connect-timeout
14795 Show the current connection timeout setting.
14798 @cindex remote packets, enabling and disabling
14799 The @value{GDBN} remote protocol autodetects the packets supported by
14800 your debugging stub. If you need to override the autodetection, you
14801 can use these commands to enable or disable individual packets. Each
14802 packet can be set to @samp{on} (the remote target supports this
14803 packet), @samp{off} (the remote target does not support this packet),
14804 or @samp{auto} (detect remote target support for this packet). They
14805 all default to @samp{auto}. For more information about each packet,
14806 see @ref{Remote Protocol}.
14808 During normal use, you should not have to use any of these commands.
14809 If you do, that may be a bug in your remote debugging stub, or a bug
14810 in @value{GDBN}. You may want to report the problem to the
14811 @value{GDBN} developers.
14813 For each packet @var{name}, the command to enable or disable the
14814 packet is @code{set remote @var{name}-packet}. The available settings
14817 @multitable @columnfractions 0.28 0.32 0.25
14820 @tab Related Features
14822 @item @code{fetch-register}
14824 @tab @code{info registers}
14826 @item @code{set-register}
14830 @item @code{binary-download}
14832 @tab @code{load}, @code{set}
14834 @item @code{read-aux-vector}
14835 @tab @code{qXfer:auxv:read}
14836 @tab @code{info auxv}
14838 @item @code{symbol-lookup}
14839 @tab @code{qSymbol}
14840 @tab Detecting multiple threads
14842 @item @code{attach}
14843 @tab @code{vAttach}
14846 @item @code{verbose-resume}
14848 @tab Stepping or resuming multiple threads
14854 @item @code{software-breakpoint}
14858 @item @code{hardware-breakpoint}
14862 @item @code{write-watchpoint}
14866 @item @code{read-watchpoint}
14870 @item @code{access-watchpoint}
14874 @item @code{target-features}
14875 @tab @code{qXfer:features:read}
14876 @tab @code{set architecture}
14878 @item @code{library-info}
14879 @tab @code{qXfer:libraries:read}
14880 @tab @code{info sharedlibrary}
14882 @item @code{memory-map}
14883 @tab @code{qXfer:memory-map:read}
14884 @tab @code{info mem}
14886 @item @code{read-spu-object}
14887 @tab @code{qXfer:spu:read}
14888 @tab @code{info spu}
14890 @item @code{write-spu-object}
14891 @tab @code{qXfer:spu:write}
14892 @tab @code{info spu}
14894 @item @code{read-siginfo-object}
14895 @tab @code{qXfer:siginfo:read}
14896 @tab @code{print $_siginfo}
14898 @item @code{write-siginfo-object}
14899 @tab @code{qXfer:siginfo:write}
14900 @tab @code{set $_siginfo}
14902 @item @code{get-thread-local-@*storage-address}
14903 @tab @code{qGetTLSAddr}
14904 @tab Displaying @code{__thread} variables
14906 @item @code{search-memory}
14907 @tab @code{qSearch:memory}
14910 @item @code{supported-packets}
14911 @tab @code{qSupported}
14912 @tab Remote communications parameters
14914 @item @code{pass-signals}
14915 @tab @code{QPassSignals}
14916 @tab @code{handle @var{signal}}
14918 @item @code{hostio-close-packet}
14919 @tab @code{vFile:close}
14920 @tab @code{remote get}, @code{remote put}
14922 @item @code{hostio-open-packet}
14923 @tab @code{vFile:open}
14924 @tab @code{remote get}, @code{remote put}
14926 @item @code{hostio-pread-packet}
14927 @tab @code{vFile:pread}
14928 @tab @code{remote get}, @code{remote put}
14930 @item @code{hostio-pwrite-packet}
14931 @tab @code{vFile:pwrite}
14932 @tab @code{remote get}, @code{remote put}
14934 @item @code{hostio-unlink-packet}
14935 @tab @code{vFile:unlink}
14936 @tab @code{remote delete}
14938 @item @code{noack-packet}
14939 @tab @code{QStartNoAckMode}
14940 @tab Packet acknowledgment
14942 @item @code{osdata}
14943 @tab @code{qXfer:osdata:read}
14944 @tab @code{info os}
14946 @item @code{query-attached}
14947 @tab @code{qAttached}
14948 @tab Querying remote process attach state.
14952 @section Implementing a Remote Stub
14954 @cindex debugging stub, example
14955 @cindex remote stub, example
14956 @cindex stub example, remote debugging
14957 The stub files provided with @value{GDBN} implement the target side of the
14958 communication protocol, and the @value{GDBN} side is implemented in the
14959 @value{GDBN} source file @file{remote.c}. Normally, you can simply allow
14960 these subroutines to communicate, and ignore the details. (If you're
14961 implementing your own stub file, you can still ignore the details: start
14962 with one of the existing stub files. @file{sparc-stub.c} is the best
14963 organized, and therefore the easiest to read.)
14965 @cindex remote serial debugging, overview
14966 To debug a program running on another machine (the debugging
14967 @dfn{target} machine), you must first arrange for all the usual
14968 prerequisites for the program to run by itself. For example, for a C
14973 A startup routine to set up the C runtime environment; these usually
14974 have a name like @file{crt0}. The startup routine may be supplied by
14975 your hardware supplier, or you may have to write your own.
14978 A C subroutine library to support your program's
14979 subroutine calls, notably managing input and output.
14982 A way of getting your program to the other machine---for example, a
14983 download program. These are often supplied by the hardware
14984 manufacturer, but you may have to write your own from hardware
14988 The next step is to arrange for your program to use a serial port to
14989 communicate with the machine where @value{GDBN} is running (the @dfn{host}
14990 machine). In general terms, the scheme looks like this:
14994 @value{GDBN} already understands how to use this protocol; when everything
14995 else is set up, you can simply use the @samp{target remote} command
14996 (@pxref{Targets,,Specifying a Debugging Target}).
14998 @item On the target,
14999 you must link with your program a few special-purpose subroutines that
15000 implement the @value{GDBN} remote serial protocol. The file containing these
15001 subroutines is called a @dfn{debugging stub}.
15003 On certain remote targets, you can use an auxiliary program
15004 @code{gdbserver} instead of linking a stub into your program.
15005 @xref{Server,,Using the @code{gdbserver} Program}, for details.
15008 The debugging stub is specific to the architecture of the remote
15009 machine; for example, use @file{sparc-stub.c} to debug programs on
15012 @cindex remote serial stub list
15013 These working remote stubs are distributed with @value{GDBN}:
15018 @cindex @file{i386-stub.c}
15021 For Intel 386 and compatible architectures.
15024 @cindex @file{m68k-stub.c}
15025 @cindex Motorola 680x0
15027 For Motorola 680x0 architectures.
15030 @cindex @file{sh-stub.c}
15033 For Renesas SH architectures.
15036 @cindex @file{sparc-stub.c}
15038 For @sc{sparc} architectures.
15040 @item sparcl-stub.c
15041 @cindex @file{sparcl-stub.c}
15044 For Fujitsu @sc{sparclite} architectures.
15048 The @file{README} file in the @value{GDBN} distribution may list other
15049 recently added stubs.
15052 * Stub Contents:: What the stub can do for you
15053 * Bootstrapping:: What you must do for the stub
15054 * Debug Session:: Putting it all together
15057 @node Stub Contents
15058 @subsection What the Stub Can Do for You
15060 @cindex remote serial stub
15061 The debugging stub for your architecture supplies these three
15065 @item set_debug_traps
15066 @findex set_debug_traps
15067 @cindex remote serial stub, initialization
15068 This routine arranges for @code{handle_exception} to run when your
15069 program stops. You must call this subroutine explicitly near the
15070 beginning of your program.
15072 @item handle_exception
15073 @findex handle_exception
15074 @cindex remote serial stub, main routine
15075 This is the central workhorse, but your program never calls it
15076 explicitly---the setup code arranges for @code{handle_exception} to
15077 run when a trap is triggered.
15079 @code{handle_exception} takes control when your program stops during
15080 execution (for example, on a breakpoint), and mediates communications
15081 with @value{GDBN} on the host machine. This is where the communications
15082 protocol is implemented; @code{handle_exception} acts as the @value{GDBN}
15083 representative on the target machine. It begins by sending summary
15084 information on the state of your program, then continues to execute,
15085 retrieving and transmitting any information @value{GDBN} needs, until you
15086 execute a @value{GDBN} command that makes your program resume; at that point,
15087 @code{handle_exception} returns control to your own code on the target
15091 @cindex @code{breakpoint} subroutine, remote
15092 Use this auxiliary subroutine to make your program contain a
15093 breakpoint. Depending on the particular situation, this may be the only
15094 way for @value{GDBN} to get control. For instance, if your target
15095 machine has some sort of interrupt button, you won't need to call this;
15096 pressing the interrupt button transfers control to
15097 @code{handle_exception}---in effect, to @value{GDBN}. On some machines,
15098 simply receiving characters on the serial port may also trigger a trap;
15099 again, in that situation, you don't need to call @code{breakpoint} from
15100 your own program---simply running @samp{target remote} from the host
15101 @value{GDBN} session gets control.
15103 Call @code{breakpoint} if none of these is true, or if you simply want
15104 to make certain your program stops at a predetermined point for the
15105 start of your debugging session.
15108 @node Bootstrapping
15109 @subsection What You Must Do for the Stub
15111 @cindex remote stub, support routines
15112 The debugging stubs that come with @value{GDBN} are set up for a particular
15113 chip architecture, but they have no information about the rest of your
15114 debugging target machine.
15116 First of all you need to tell the stub how to communicate with the
15120 @item int getDebugChar()
15121 @findex getDebugChar
15122 Write this subroutine to read a single character from the serial port.
15123 It may be identical to @code{getchar} for your target system; a
15124 different name is used to allow you to distinguish the two if you wish.
15126 @item void putDebugChar(int)
15127 @findex putDebugChar
15128 Write this subroutine to write a single character to the serial port.
15129 It may be identical to @code{putchar} for your target system; a
15130 different name is used to allow you to distinguish the two if you wish.
15133 @cindex control C, and remote debugging
15134 @cindex interrupting remote targets
15135 If you want @value{GDBN} to be able to stop your program while it is
15136 running, you need to use an interrupt-driven serial driver, and arrange
15137 for it to stop when it receives a @code{^C} (@samp{\003}, the control-C
15138 character). That is the character which @value{GDBN} uses to tell the
15139 remote system to stop.
15141 Getting the debugging target to return the proper status to @value{GDBN}
15142 probably requires changes to the standard stub; one quick and dirty way
15143 is to just execute a breakpoint instruction (the ``dirty'' part is that
15144 @value{GDBN} reports a @code{SIGTRAP} instead of a @code{SIGINT}).
15146 Other routines you need to supply are:
15149 @item void exceptionHandler (int @var{exception_number}, void *@var{exception_address})
15150 @findex exceptionHandler
15151 Write this function to install @var{exception_address} in the exception
15152 handling tables. You need to do this because the stub does not have any
15153 way of knowing what the exception handling tables on your target system
15154 are like (for example, the processor's table might be in @sc{rom},
15155 containing entries which point to a table in @sc{ram}).
15156 @var{exception_number} is the exception number which should be changed;
15157 its meaning is architecture-dependent (for example, different numbers
15158 might represent divide by zero, misaligned access, etc). When this
15159 exception occurs, control should be transferred directly to
15160 @var{exception_address}, and the processor state (stack, registers,
15161 and so on) should be just as it is when a processor exception occurs. So if
15162 you want to use a jump instruction to reach @var{exception_address}, it
15163 should be a simple jump, not a jump to subroutine.
15165 For the 386, @var{exception_address} should be installed as an interrupt
15166 gate so that interrupts are masked while the handler runs. The gate
15167 should be at privilege level 0 (the most privileged level). The
15168 @sc{sparc} and 68k stubs are able to mask interrupts themselves without
15169 help from @code{exceptionHandler}.
15171 @item void flush_i_cache()
15172 @findex flush_i_cache
15173 On @sc{sparc} and @sc{sparclite} only, write this subroutine to flush the
15174 instruction cache, if any, on your target machine. If there is no
15175 instruction cache, this subroutine may be a no-op.
15177 On target machines that have instruction caches, @value{GDBN} requires this
15178 function to make certain that the state of your program is stable.
15182 You must also make sure this library routine is available:
15185 @item void *memset(void *, int, int)
15187 This is the standard library function @code{memset} that sets an area of
15188 memory to a known value. If you have one of the free versions of
15189 @code{libc.a}, @code{memset} can be found there; otherwise, you must
15190 either obtain it from your hardware manufacturer, or write your own.
15193 If you do not use the GNU C compiler, you may need other standard
15194 library subroutines as well; this varies from one stub to another,
15195 but in general the stubs are likely to use any of the common library
15196 subroutines which @code{@value{NGCC}} generates as inline code.
15199 @node Debug Session
15200 @subsection Putting it All Together
15202 @cindex remote serial debugging summary
15203 In summary, when your program is ready to debug, you must follow these
15208 Make sure you have defined the supporting low-level routines
15209 (@pxref{Bootstrapping,,What You Must Do for the Stub}):
15211 @code{getDebugChar}, @code{putDebugChar},
15212 @code{flush_i_cache}, @code{memset}, @code{exceptionHandler}.
15216 Insert these lines near the top of your program:
15224 For the 680x0 stub only, you need to provide a variable called
15225 @code{exceptionHook}. Normally you just use:
15228 void (*exceptionHook)() = 0;
15232 but if before calling @code{set_debug_traps}, you set it to point to a
15233 function in your program, that function is called when
15234 @code{@value{GDBN}} continues after stopping on a trap (for example, bus
15235 error). The function indicated by @code{exceptionHook} is called with
15236 one parameter: an @code{int} which is the exception number.
15239 Compile and link together: your program, the @value{GDBN} debugging stub for
15240 your target architecture, and the supporting subroutines.
15243 Make sure you have a serial connection between your target machine and
15244 the @value{GDBN} host, and identify the serial port on the host.
15247 @c The "remote" target now provides a `load' command, so we should
15248 @c document that. FIXME.
15249 Download your program to your target machine (or get it there by
15250 whatever means the manufacturer provides), and start it.
15253 Start @value{GDBN} on the host, and connect to the target
15254 (@pxref{Connecting,,Connecting to a Remote Target}).
15258 @node Configurations
15259 @chapter Configuration-Specific Information
15261 While nearly all @value{GDBN} commands are available for all native and
15262 cross versions of the debugger, there are some exceptions. This chapter
15263 describes things that are only available in certain configurations.
15265 There are three major categories of configurations: native
15266 configurations, where the host and target are the same, embedded
15267 operating system configurations, which are usually the same for several
15268 different processor architectures, and bare embedded processors, which
15269 are quite different from each other.
15274 * Embedded Processors::
15281 This section describes details specific to particular native
15286 * BSD libkvm Interface:: Debugging BSD kernel memory images
15287 * SVR4 Process Information:: SVR4 process information
15288 * DJGPP Native:: Features specific to the DJGPP port
15289 * Cygwin Native:: Features specific to the Cygwin port
15290 * Hurd Native:: Features specific to @sc{gnu} Hurd
15291 * Neutrino:: Features specific to QNX Neutrino
15292 * Darwin:: Features specific to Darwin
15298 On HP-UX systems, if you refer to a function or variable name that
15299 begins with a dollar sign, @value{GDBN} searches for a user or system
15300 name first, before it searches for a convenience variable.
15303 @node BSD libkvm Interface
15304 @subsection BSD libkvm Interface
15307 @cindex kernel memory image
15308 @cindex kernel crash dump
15310 BSD-derived systems (FreeBSD/NetBSD/OpenBSD) have a kernel memory
15311 interface that provides a uniform interface for accessing kernel virtual
15312 memory images, including live systems and crash dumps. @value{GDBN}
15313 uses this interface to allow you to debug live kernels and kernel crash
15314 dumps on many native BSD configurations. This is implemented as a
15315 special @code{kvm} debugging target. For debugging a live system, load
15316 the currently running kernel into @value{GDBN} and connect to the
15320 (@value{GDBP}) @b{target kvm}
15323 For debugging crash dumps, provide the file name of the crash dump as an
15327 (@value{GDBP}) @b{target kvm /var/crash/bsd.0}
15330 Once connected to the @code{kvm} target, the following commands are
15336 Set current context from the @dfn{Process Control Block} (PCB) address.
15339 Set current context from proc address. This command isn't available on
15340 modern FreeBSD systems.
15343 @node SVR4 Process Information
15344 @subsection SVR4 Process Information
15346 @cindex examine process image
15347 @cindex process info via @file{/proc}
15349 Many versions of SVR4 and compatible systems provide a facility called
15350 @samp{/proc} that can be used to examine the image of a running
15351 process using file-system subroutines. If @value{GDBN} is configured
15352 for an operating system with this facility, the command @code{info
15353 proc} is available to report information about the process running
15354 your program, or about any process running on your system. @code{info
15355 proc} works only on SVR4 systems that include the @code{procfs} code.
15356 This includes, as of this writing, @sc{gnu}/Linux, OSF/1 (Digital
15357 Unix), Solaris, Irix, and Unixware, but not HP-UX, for example.
15363 @itemx info proc @var{process-id}
15364 Summarize available information about any running process. If a
15365 process ID is specified by @var{process-id}, display information about
15366 that process; otherwise display information about the program being
15367 debugged. The summary includes the debugged process ID, the command
15368 line used to invoke it, its current working directory, and its
15369 executable file's absolute file name.
15371 On some systems, @var{process-id} can be of the form
15372 @samp{[@var{pid}]/@var{tid}} which specifies a certain thread ID
15373 within a process. If the optional @var{pid} part is missing, it means
15374 a thread from the process being debugged (the leading @samp{/} still
15375 needs to be present, or else @value{GDBN} will interpret the number as
15376 a process ID rather than a thread ID).
15378 @item info proc mappings
15379 @cindex memory address space mappings
15380 Report the memory address space ranges accessible in the program, with
15381 information on whether the process has read, write, or execute access
15382 rights to each range. On @sc{gnu}/Linux systems, each memory range
15383 includes the object file which is mapped to that range, instead of the
15384 memory access rights to that range.
15386 @item info proc stat
15387 @itemx info proc status
15388 @cindex process detailed status information
15389 These subcommands are specific to @sc{gnu}/Linux systems. They show
15390 the process-related information, including the user ID and group ID;
15391 how many threads are there in the process; its virtual memory usage;
15392 the signals that are pending, blocked, and ignored; its TTY; its
15393 consumption of system and user time; its stack size; its @samp{nice}
15394 value; etc. For more information, see the @samp{proc} man page
15395 (type @kbd{man 5 proc} from your shell prompt).
15397 @item info proc all
15398 Show all the information about the process described under all of the
15399 above @code{info proc} subcommands.
15402 @comment These sub-options of 'info proc' were not included when
15403 @comment procfs.c was re-written. Keep their descriptions around
15404 @comment against the day when someone finds the time to put them back in.
15405 @kindex info proc times
15406 @item info proc times
15407 Starting time, user CPU time, and system CPU time for your program and
15410 @kindex info proc id
15412 Report on the process IDs related to your program: its own process ID,
15413 the ID of its parent, the process group ID, and the session ID.
15416 @item set procfs-trace
15417 @kindex set procfs-trace
15418 @cindex @code{procfs} API calls
15419 This command enables and disables tracing of @code{procfs} API calls.
15421 @item show procfs-trace
15422 @kindex show procfs-trace
15423 Show the current state of @code{procfs} API call tracing.
15425 @item set procfs-file @var{file}
15426 @kindex set procfs-file
15427 Tell @value{GDBN} to write @code{procfs} API trace to the named
15428 @var{file}. @value{GDBN} appends the trace info to the previous
15429 contents of the file. The default is to display the trace on the
15432 @item show procfs-file
15433 @kindex show procfs-file
15434 Show the file to which @code{procfs} API trace is written.
15436 @item proc-trace-entry
15437 @itemx proc-trace-exit
15438 @itemx proc-untrace-entry
15439 @itemx proc-untrace-exit
15440 @kindex proc-trace-entry
15441 @kindex proc-trace-exit
15442 @kindex proc-untrace-entry
15443 @kindex proc-untrace-exit
15444 These commands enable and disable tracing of entries into and exits
15445 from the @code{syscall} interface.
15448 @kindex info pidlist
15449 @cindex process list, QNX Neutrino
15450 For QNX Neutrino only, this command displays the list of all the
15451 processes and all the threads within each process.
15454 @kindex info meminfo
15455 @cindex mapinfo list, QNX Neutrino
15456 For QNX Neutrino only, this command displays the list of all mapinfos.
15460 @subsection Features for Debugging @sc{djgpp} Programs
15461 @cindex @sc{djgpp} debugging
15462 @cindex native @sc{djgpp} debugging
15463 @cindex MS-DOS-specific commands
15466 @sc{djgpp} is a port of the @sc{gnu} development tools to MS-DOS and
15467 MS-Windows. @sc{djgpp} programs are 32-bit protected-mode programs
15468 that use the @dfn{DPMI} (DOS Protected-Mode Interface) API to run on
15469 top of real-mode DOS systems and their emulations.
15471 @value{GDBN} supports native debugging of @sc{djgpp} programs, and
15472 defines a few commands specific to the @sc{djgpp} port. This
15473 subsection describes those commands.
15478 This is a prefix of @sc{djgpp}-specific commands which print
15479 information about the target system and important OS structures.
15482 @cindex MS-DOS system info
15483 @cindex free memory information (MS-DOS)
15484 @item info dos sysinfo
15485 This command displays assorted information about the underlying
15486 platform: the CPU type and features, the OS version and flavor, the
15487 DPMI version, and the available conventional and DPMI memory.
15492 @cindex segment descriptor tables
15493 @cindex descriptor tables display
15495 @itemx info dos ldt
15496 @itemx info dos idt
15497 These 3 commands display entries from, respectively, Global, Local,
15498 and Interrupt Descriptor Tables (GDT, LDT, and IDT). The descriptor
15499 tables are data structures which store a descriptor for each segment
15500 that is currently in use. The segment's selector is an index into a
15501 descriptor table; the table entry for that index holds the
15502 descriptor's base address and limit, and its attributes and access
15505 A typical @sc{djgpp} program uses 3 segments: a code segment, a data
15506 segment (used for both data and the stack), and a DOS segment (which
15507 allows access to DOS/BIOS data structures and absolute addresses in
15508 conventional memory). However, the DPMI host will usually define
15509 additional segments in order to support the DPMI environment.
15511 @cindex garbled pointers
15512 These commands allow to display entries from the descriptor tables.
15513 Without an argument, all entries from the specified table are
15514 displayed. An argument, which should be an integer expression, means
15515 display a single entry whose index is given by the argument. For
15516 example, here's a convenient way to display information about the
15517 debugged program's data segment:
15520 @exdent @code{(@value{GDBP}) info dos ldt $ds}
15521 @exdent @code{0x13f: base=0x11970000 limit=0x0009ffff 32-Bit Data (Read/Write, Exp-up)}
15525 This comes in handy when you want to see whether a pointer is outside
15526 the data segment's limit (i.e.@: @dfn{garbled}).
15528 @cindex page tables display (MS-DOS)
15530 @itemx info dos pte
15531 These two commands display entries from, respectively, the Page
15532 Directory and the Page Tables. Page Directories and Page Tables are
15533 data structures which control how virtual memory addresses are mapped
15534 into physical addresses. A Page Table includes an entry for every
15535 page of memory that is mapped into the program's address space; there
15536 may be several Page Tables, each one holding up to 4096 entries. A
15537 Page Directory has up to 4096 entries, one each for every Page Table
15538 that is currently in use.
15540 Without an argument, @kbd{info dos pde} displays the entire Page
15541 Directory, and @kbd{info dos pte} displays all the entries in all of
15542 the Page Tables. An argument, an integer expression, given to the
15543 @kbd{info dos pde} command means display only that entry from the Page
15544 Directory table. An argument given to the @kbd{info dos pte} command
15545 means display entries from a single Page Table, the one pointed to by
15546 the specified entry in the Page Directory.
15548 @cindex direct memory access (DMA) on MS-DOS
15549 These commands are useful when your program uses @dfn{DMA} (Direct
15550 Memory Access), which needs physical addresses to program the DMA
15553 These commands are supported only with some DPMI servers.
15555 @cindex physical address from linear address
15556 @item info dos address-pte @var{addr}
15557 This command displays the Page Table entry for a specified linear
15558 address. The argument @var{addr} is a linear address which should
15559 already have the appropriate segment's base address added to it,
15560 because this command accepts addresses which may belong to @emph{any}
15561 segment. For example, here's how to display the Page Table entry for
15562 the page where a variable @code{i} is stored:
15565 @exdent @code{(@value{GDBP}) info dos address-pte __djgpp_base_address + (char *)&i}
15566 @exdent @code{Page Table entry for address 0x11a00d30:}
15567 @exdent @code{Base=0x02698000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0xd30}
15571 This says that @code{i} is stored at offset @code{0xd30} from the page
15572 whose physical base address is @code{0x02698000}, and shows all the
15573 attributes of that page.
15575 Note that you must cast the addresses of variables to a @code{char *},
15576 since otherwise the value of @code{__djgpp_base_address}, the base
15577 address of all variables and functions in a @sc{djgpp} program, will
15578 be added using the rules of C pointer arithmetics: if @code{i} is
15579 declared an @code{int}, @value{GDBN} will add 4 times the value of
15580 @code{__djgpp_base_address} to the address of @code{i}.
15582 Here's another example, it displays the Page Table entry for the
15586 @exdent @code{(@value{GDBP}) info dos address-pte *((unsigned *)&_go32_info_block + 3)}
15587 @exdent @code{Page Table entry for address 0x29110:}
15588 @exdent @code{Base=0x00029000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0x110}
15592 (The @code{+ 3} offset is because the transfer buffer's address is the
15593 3rd member of the @code{_go32_info_block} structure.) The output
15594 clearly shows that this DPMI server maps the addresses in conventional
15595 memory 1:1, i.e.@: the physical (@code{0x00029000} + @code{0x110}) and
15596 linear (@code{0x29110}) addresses are identical.
15598 This command is supported only with some DPMI servers.
15601 @cindex DOS serial data link, remote debugging
15602 In addition to native debugging, the DJGPP port supports remote
15603 debugging via a serial data link. The following commands are specific
15604 to remote serial debugging in the DJGPP port of @value{GDBN}.
15607 @kindex set com1base
15608 @kindex set com1irq
15609 @kindex set com2base
15610 @kindex set com2irq
15611 @kindex set com3base
15612 @kindex set com3irq
15613 @kindex set com4base
15614 @kindex set com4irq
15615 @item set com1base @var{addr}
15616 This command sets the base I/O port address of the @file{COM1} serial
15619 @item set com1irq @var{irq}
15620 This command sets the @dfn{Interrupt Request} (@code{IRQ}) line to use
15621 for the @file{COM1} serial port.
15623 There are similar commands @samp{set com2base}, @samp{set com3irq},
15624 etc.@: for setting the port address and the @code{IRQ} lines for the
15627 @kindex show com1base
15628 @kindex show com1irq
15629 @kindex show com2base
15630 @kindex show com2irq
15631 @kindex show com3base
15632 @kindex show com3irq
15633 @kindex show com4base
15634 @kindex show com4irq
15635 The related commands @samp{show com1base}, @samp{show com1irq} etc.@:
15636 display the current settings of the base address and the @code{IRQ}
15637 lines used by the COM ports.
15640 @kindex info serial
15641 @cindex DOS serial port status
15642 This command prints the status of the 4 DOS serial ports. For each
15643 port, it prints whether it's active or not, its I/O base address and
15644 IRQ number, whether it uses a 16550-style FIFO, its baudrate, and the
15645 counts of various errors encountered so far.
15649 @node Cygwin Native
15650 @subsection Features for Debugging MS Windows PE Executables
15651 @cindex MS Windows debugging
15652 @cindex native Cygwin debugging
15653 @cindex Cygwin-specific commands
15655 @value{GDBN} supports native debugging of MS Windows programs, including
15656 DLLs with and without symbolic debugging information. There are various
15657 additional Cygwin-specific commands, described in this section.
15658 Working with DLLs that have no debugging symbols is described in
15659 @ref{Non-debug DLL Symbols}.
15664 This is a prefix of MS Windows-specific commands which print
15665 information about the target system and important OS structures.
15667 @item info w32 selector
15668 This command displays information returned by
15669 the Win32 API @code{GetThreadSelectorEntry} function.
15670 It takes an optional argument that is evaluated to
15671 a long value to give the information about this given selector.
15672 Without argument, this command displays information
15673 about the six segment registers.
15677 This is a Cygwin-specific alias of @code{info shared}.
15679 @kindex dll-symbols
15681 This command loads symbols from a dll similarly to
15682 add-sym command but without the need to specify a base address.
15684 @kindex set cygwin-exceptions
15685 @cindex debugging the Cygwin DLL
15686 @cindex Cygwin DLL, debugging
15687 @item set cygwin-exceptions @var{mode}
15688 If @var{mode} is @code{on}, @value{GDBN} will break on exceptions that
15689 happen inside the Cygwin DLL. If @var{mode} is @code{off},
15690 @value{GDBN} will delay recognition of exceptions, and may ignore some
15691 exceptions which seem to be caused by internal Cygwin DLL
15692 ``bookkeeping''. This option is meant primarily for debugging the
15693 Cygwin DLL itself; the default value is @code{off} to avoid annoying
15694 @value{GDBN} users with false @code{SIGSEGV} signals.
15696 @kindex show cygwin-exceptions
15697 @item show cygwin-exceptions
15698 Displays whether @value{GDBN} will break on exceptions that happen
15699 inside the Cygwin DLL itself.
15701 @kindex set new-console
15702 @item set new-console @var{mode}
15703 If @var{mode} is @code{on} the debuggee will
15704 be started in a new console on next start.
15705 If @var{mode} is @code{off}i, the debuggee will
15706 be started in the same console as the debugger.
15708 @kindex show new-console
15709 @item show new-console
15710 Displays whether a new console is used
15711 when the debuggee is started.
15713 @kindex set new-group
15714 @item set new-group @var{mode}
15715 This boolean value controls whether the debuggee should
15716 start a new group or stay in the same group as the debugger.
15717 This affects the way the Windows OS handles
15720 @kindex show new-group
15721 @item show new-group
15722 Displays current value of new-group boolean.
15724 @kindex set debugevents
15725 @item set debugevents
15726 This boolean value adds debug output concerning kernel events related
15727 to the debuggee seen by the debugger. This includes events that
15728 signal thread and process creation and exit, DLL loading and
15729 unloading, console interrupts, and debugging messages produced by the
15730 Windows @code{OutputDebugString} API call.
15732 @kindex set debugexec
15733 @item set debugexec
15734 This boolean value adds debug output concerning execute events
15735 (such as resume thread) seen by the debugger.
15737 @kindex set debugexceptions
15738 @item set debugexceptions
15739 This boolean value adds debug output concerning exceptions in the
15740 debuggee seen by the debugger.
15742 @kindex set debugmemory
15743 @item set debugmemory
15744 This boolean value adds debug output concerning debuggee memory reads
15745 and writes by the debugger.
15749 This boolean values specifies whether the debuggee is called
15750 via a shell or directly (default value is on).
15754 Displays if the debuggee will be started with a shell.
15759 * Non-debug DLL Symbols:: Support for DLLs without debugging symbols
15762 @node Non-debug DLL Symbols
15763 @subsubsection Support for DLLs without Debugging Symbols
15764 @cindex DLLs with no debugging symbols
15765 @cindex Minimal symbols and DLLs
15767 Very often on windows, some of the DLLs that your program relies on do
15768 not include symbolic debugging information (for example,
15769 @file{kernel32.dll}). When @value{GDBN} doesn't recognize any debugging
15770 symbols in a DLL, it relies on the minimal amount of symbolic
15771 information contained in the DLL's export table. This section
15772 describes working with such symbols, known internally to @value{GDBN} as
15773 ``minimal symbols''.
15775 Note that before the debugged program has started execution, no DLLs
15776 will have been loaded. The easiest way around this problem is simply to
15777 start the program --- either by setting a breakpoint or letting the
15778 program run once to completion. It is also possible to force
15779 @value{GDBN} to load a particular DLL before starting the executable ---
15780 see the shared library information in @ref{Files}, or the
15781 @code{dll-symbols} command in @ref{Cygwin Native}. Currently,
15782 explicitly loading symbols from a DLL with no debugging information will
15783 cause the symbol names to be duplicated in @value{GDBN}'s lookup table,
15784 which may adversely affect symbol lookup performance.
15786 @subsubsection DLL Name Prefixes
15788 In keeping with the naming conventions used by the Microsoft debugging
15789 tools, DLL export symbols are made available with a prefix based on the
15790 DLL name, for instance @code{KERNEL32!CreateFileA}. The plain name is
15791 also entered into the symbol table, so @code{CreateFileA} is often
15792 sufficient. In some cases there will be name clashes within a program
15793 (particularly if the executable itself includes full debugging symbols)
15794 necessitating the use of the fully qualified name when referring to the
15795 contents of the DLL. Use single-quotes around the name to avoid the
15796 exclamation mark (``!'') being interpreted as a language operator.
15798 Note that the internal name of the DLL may be all upper-case, even
15799 though the file name of the DLL is lower-case, or vice-versa. Since
15800 symbols within @value{GDBN} are @emph{case-sensitive} this may cause
15801 some confusion. If in doubt, try the @code{info functions} and
15802 @code{info variables} commands or even @code{maint print msymbols}
15803 (@pxref{Symbols}). Here's an example:
15806 (@value{GDBP}) info function CreateFileA
15807 All functions matching regular expression "CreateFileA":
15809 Non-debugging symbols:
15810 0x77e885f4 CreateFileA
15811 0x77e885f4 KERNEL32!CreateFileA
15815 (@value{GDBP}) info function !
15816 All functions matching regular expression "!":
15818 Non-debugging symbols:
15819 0x6100114c cygwin1!__assert
15820 0x61004034 cygwin1!_dll_crt0@@0
15821 0x61004240 cygwin1!dll_crt0(per_process *)
15825 @subsubsection Working with Minimal Symbols
15827 Symbols extracted from a DLL's export table do not contain very much
15828 type information. All that @value{GDBN} can do is guess whether a symbol
15829 refers to a function or variable depending on the linker section that
15830 contains the symbol. Also note that the actual contents of the memory
15831 contained in a DLL are not available unless the program is running. This
15832 means that you cannot examine the contents of a variable or disassemble
15833 a function within a DLL without a running program.
15835 Variables are generally treated as pointers and dereferenced
15836 automatically. For this reason, it is often necessary to prefix a
15837 variable name with the address-of operator (``&'') and provide explicit
15838 type information in the command. Here's an example of the type of
15842 (@value{GDBP}) print 'cygwin1!__argv'
15847 (@value{GDBP}) x 'cygwin1!__argv'
15848 0x10021610: "\230y\""
15851 And two possible solutions:
15854 (@value{GDBP}) print ((char **)'cygwin1!__argv')[0]
15855 $2 = 0x22fd98 "/cygdrive/c/mydirectory/myprogram"
15859 (@value{GDBP}) x/2x &'cygwin1!__argv'
15860 0x610c0aa8 <cygwin1!__argv>: 0x10021608 0x00000000
15861 (@value{GDBP}) x/x 0x10021608
15862 0x10021608: 0x0022fd98
15863 (@value{GDBP}) x/s 0x0022fd98
15864 0x22fd98: "/cygdrive/c/mydirectory/myprogram"
15867 Setting a break point within a DLL is possible even before the program
15868 starts execution. However, under these circumstances, @value{GDBN} can't
15869 examine the initial instructions of the function in order to skip the
15870 function's frame set-up code. You can work around this by using ``*&''
15871 to set the breakpoint at a raw memory address:
15874 (@value{GDBP}) break *&'python22!PyOS_Readline'
15875 Breakpoint 1 at 0x1e04eff0
15878 The author of these extensions is not entirely convinced that setting a
15879 break point within a shared DLL like @file{kernel32.dll} is completely
15883 @subsection Commands Specific to @sc{gnu} Hurd Systems
15884 @cindex @sc{gnu} Hurd debugging
15886 This subsection describes @value{GDBN} commands specific to the
15887 @sc{gnu} Hurd native debugging.
15892 @kindex set signals@r{, Hurd command}
15893 @kindex set sigs@r{, Hurd command}
15894 This command toggles the state of inferior signal interception by
15895 @value{GDBN}. Mach exceptions, such as breakpoint traps, are not
15896 affected by this command. @code{sigs} is a shorthand alias for
15901 @kindex show signals@r{, Hurd command}
15902 @kindex show sigs@r{, Hurd command}
15903 Show the current state of intercepting inferior's signals.
15905 @item set signal-thread
15906 @itemx set sigthread
15907 @kindex set signal-thread
15908 @kindex set sigthread
15909 This command tells @value{GDBN} which thread is the @code{libc} signal
15910 thread. That thread is run when a signal is delivered to a running
15911 process. @code{set sigthread} is the shorthand alias of @code{set
15914 @item show signal-thread
15915 @itemx show sigthread
15916 @kindex show signal-thread
15917 @kindex show sigthread
15918 These two commands show which thread will run when the inferior is
15919 delivered a signal.
15922 @kindex set stopped@r{, Hurd command}
15923 This commands tells @value{GDBN} that the inferior process is stopped,
15924 as with the @code{SIGSTOP} signal. The stopped process can be
15925 continued by delivering a signal to it.
15928 @kindex show stopped@r{, Hurd command}
15929 This command shows whether @value{GDBN} thinks the debuggee is
15932 @item set exceptions
15933 @kindex set exceptions@r{, Hurd command}
15934 Use this command to turn off trapping of exceptions in the inferior.
15935 When exception trapping is off, neither breakpoints nor
15936 single-stepping will work. To restore the default, set exception
15939 @item show exceptions
15940 @kindex show exceptions@r{, Hurd command}
15941 Show the current state of trapping exceptions in the inferior.
15943 @item set task pause
15944 @kindex set task@r{, Hurd commands}
15945 @cindex task attributes (@sc{gnu} Hurd)
15946 @cindex pause current task (@sc{gnu} Hurd)
15947 This command toggles task suspension when @value{GDBN} has control.
15948 Setting it to on takes effect immediately, and the task is suspended
15949 whenever @value{GDBN} gets control. Setting it to off will take
15950 effect the next time the inferior is continued. If this option is set
15951 to off, you can use @code{set thread default pause on} or @code{set
15952 thread pause on} (see below) to pause individual threads.
15954 @item show task pause
15955 @kindex show task@r{, Hurd commands}
15956 Show the current state of task suspension.
15958 @item set task detach-suspend-count
15959 @cindex task suspend count
15960 @cindex detach from task, @sc{gnu} Hurd
15961 This command sets the suspend count the task will be left with when
15962 @value{GDBN} detaches from it.
15964 @item show task detach-suspend-count
15965 Show the suspend count the task will be left with when detaching.
15967 @item set task exception-port
15968 @itemx set task excp
15969 @cindex task exception port, @sc{gnu} Hurd
15970 This command sets the task exception port to which @value{GDBN} will
15971 forward exceptions. The argument should be the value of the @dfn{send
15972 rights} of the task. @code{set task excp} is a shorthand alias.
15974 @item set noninvasive
15975 @cindex noninvasive task options
15976 This command switches @value{GDBN} to a mode that is the least
15977 invasive as far as interfering with the inferior is concerned. This
15978 is the same as using @code{set task pause}, @code{set exceptions}, and
15979 @code{set signals} to values opposite to the defaults.
15981 @item info send-rights
15982 @itemx info receive-rights
15983 @itemx info port-rights
15984 @itemx info port-sets
15985 @itemx info dead-names
15988 @cindex send rights, @sc{gnu} Hurd
15989 @cindex receive rights, @sc{gnu} Hurd
15990 @cindex port rights, @sc{gnu} Hurd
15991 @cindex port sets, @sc{gnu} Hurd
15992 @cindex dead names, @sc{gnu} Hurd
15993 These commands display information about, respectively, send rights,
15994 receive rights, port rights, port sets, and dead names of a task.
15995 There are also shorthand aliases: @code{info ports} for @code{info
15996 port-rights} and @code{info psets} for @code{info port-sets}.
15998 @item set thread pause
15999 @kindex set thread@r{, Hurd command}
16000 @cindex thread properties, @sc{gnu} Hurd
16001 @cindex pause current thread (@sc{gnu} Hurd)
16002 This command toggles current thread suspension when @value{GDBN} has
16003 control. Setting it to on takes effect immediately, and the current
16004 thread is suspended whenever @value{GDBN} gets control. Setting it to
16005 off will take effect the next time the inferior is continued.
16006 Normally, this command has no effect, since when @value{GDBN} has
16007 control, the whole task is suspended. However, if you used @code{set
16008 task pause off} (see above), this command comes in handy to suspend
16009 only the current thread.
16011 @item show thread pause
16012 @kindex show thread@r{, Hurd command}
16013 This command shows the state of current thread suspension.
16015 @item set thread run
16016 This command sets whether the current thread is allowed to run.
16018 @item show thread run
16019 Show whether the current thread is allowed to run.
16021 @item set thread detach-suspend-count
16022 @cindex thread suspend count, @sc{gnu} Hurd
16023 @cindex detach from thread, @sc{gnu} Hurd
16024 This command sets the suspend count @value{GDBN} will leave on a
16025 thread when detaching. This number is relative to the suspend count
16026 found by @value{GDBN} when it notices the thread; use @code{set thread
16027 takeover-suspend-count} to force it to an absolute value.
16029 @item show thread detach-suspend-count
16030 Show the suspend count @value{GDBN} will leave on the thread when
16033 @item set thread exception-port
16034 @itemx set thread excp
16035 Set the thread exception port to which to forward exceptions. This
16036 overrides the port set by @code{set task exception-port} (see above).
16037 @code{set thread excp} is the shorthand alias.
16039 @item set thread takeover-suspend-count
16040 Normally, @value{GDBN}'s thread suspend counts are relative to the
16041 value @value{GDBN} finds when it notices each thread. This command
16042 changes the suspend counts to be absolute instead.
16044 @item set thread default
16045 @itemx show thread default
16046 @cindex thread default settings, @sc{gnu} Hurd
16047 Each of the above @code{set thread} commands has a @code{set thread
16048 default} counterpart (e.g., @code{set thread default pause}, @code{set
16049 thread default exception-port}, etc.). The @code{thread default}
16050 variety of commands sets the default thread properties for all
16051 threads; you can then change the properties of individual threads with
16052 the non-default commands.
16057 @subsection QNX Neutrino
16058 @cindex QNX Neutrino
16060 @value{GDBN} provides the following commands specific to the QNX
16064 @item set debug nto-debug
16065 @kindex set debug nto-debug
16066 When set to on, enables debugging messages specific to the QNX
16069 @item show debug nto-debug
16070 @kindex show debug nto-debug
16071 Show the current state of QNX Neutrino messages.
16078 @value{GDBN} provides the following commands specific to the Darwin target:
16081 @item set debug darwin @var{num}
16082 @kindex set debug darwin
16083 When set to a non zero value, enables debugging messages specific to
16084 the Darwin support. Higher values produce more verbose output.
16086 @item show debug darwin
16087 @kindex show debug darwin
16088 Show the current state of Darwin messages.
16090 @item set debug mach-o @var{num}
16091 @kindex set debug mach-o
16092 When set to a non zero value, enables debugging messages while
16093 @value{GDBN} is reading Darwin object files. (@dfn{Mach-O} is the
16094 file format used on Darwin for object and executable files.) Higher
16095 values produce more verbose output. This is a command to diagnose
16096 problems internal to @value{GDBN} and should not be needed in normal
16099 @item show debug mach-o
16100 @kindex show debug mach-o
16101 Show the current state of Mach-O file messages.
16103 @item set mach-exceptions on
16104 @itemx set mach-exceptions off
16105 @kindex set mach-exceptions
16106 On Darwin, faults are first reported as a Mach exception and are then
16107 mapped to a Posix signal. Use this command to turn on trapping of
16108 Mach exceptions in the inferior. This might be sometimes useful to
16109 better understand the cause of a fault. The default is off.
16111 @item show mach-exceptions
16112 @kindex show mach-exceptions
16113 Show the current state of exceptions trapping.
16118 @section Embedded Operating Systems
16120 This section describes configurations involving the debugging of
16121 embedded operating systems that are available for several different
16125 * VxWorks:: Using @value{GDBN} with VxWorks
16128 @value{GDBN} includes the ability to debug programs running on
16129 various real-time operating systems.
16132 @subsection Using @value{GDBN} with VxWorks
16138 @kindex target vxworks
16139 @item target vxworks @var{machinename}
16140 A VxWorks system, attached via TCP/IP. The argument @var{machinename}
16141 is the target system's machine name or IP address.
16145 On VxWorks, @code{load} links @var{filename} dynamically on the
16146 current target system as well as adding its symbols in @value{GDBN}.
16148 @value{GDBN} enables developers to spawn and debug tasks running on networked
16149 VxWorks targets from a Unix host. Already-running tasks spawned from
16150 the VxWorks shell can also be debugged. @value{GDBN} uses code that runs on
16151 both the Unix host and on the VxWorks target. The program
16152 @code{@value{GDBP}} is installed and executed on the Unix host. (It may be
16153 installed with the name @code{vxgdb}, to distinguish it from a
16154 @value{GDBN} for debugging programs on the host itself.)
16157 @item VxWorks-timeout @var{args}
16158 @kindex vxworks-timeout
16159 All VxWorks-based targets now support the option @code{vxworks-timeout}.
16160 This option is set by the user, and @var{args} represents the number of
16161 seconds @value{GDBN} waits for responses to rpc's. You might use this if
16162 your VxWorks target is a slow software simulator or is on the far side
16163 of a thin network line.
16166 The following information on connecting to VxWorks was current when
16167 this manual was produced; newer releases of VxWorks may use revised
16170 @findex INCLUDE_RDB
16171 To use @value{GDBN} with VxWorks, you must rebuild your VxWorks kernel
16172 to include the remote debugging interface routines in the VxWorks
16173 library @file{rdb.a}. To do this, define @code{INCLUDE_RDB} in the
16174 VxWorks configuration file @file{configAll.h} and rebuild your VxWorks
16175 kernel. The resulting kernel contains @file{rdb.a}, and spawns the
16176 source debugging task @code{tRdbTask} when VxWorks is booted. For more
16177 information on configuring and remaking VxWorks, see the manufacturer's
16179 @c VxWorks, see the @cite{VxWorks Programmer's Guide}.
16181 Once you have included @file{rdb.a} in your VxWorks system image and set
16182 your Unix execution search path to find @value{GDBN}, you are ready to
16183 run @value{GDBN}. From your Unix host, run @code{@value{GDBP}} (or
16184 @code{vxgdb}, depending on your installation).
16186 @value{GDBN} comes up showing the prompt:
16193 * VxWorks Connection:: Connecting to VxWorks
16194 * VxWorks Download:: VxWorks download
16195 * VxWorks Attach:: Running tasks
16198 @node VxWorks Connection
16199 @subsubsection Connecting to VxWorks
16201 The @value{GDBN} command @code{target} lets you connect to a VxWorks target on the
16202 network. To connect to a target whose host name is ``@code{tt}'', type:
16205 (vxgdb) target vxworks tt
16209 @value{GDBN} displays messages like these:
16212 Attaching remote machine across net...
16217 @value{GDBN} then attempts to read the symbol tables of any object modules
16218 loaded into the VxWorks target since it was last booted. @value{GDBN} locates
16219 these files by searching the directories listed in the command search
16220 path (@pxref{Environment, ,Your Program's Environment}); if it fails
16221 to find an object file, it displays a message such as:
16224 prog.o: No such file or directory.
16227 When this happens, add the appropriate directory to the search path with
16228 the @value{GDBN} command @code{path}, and execute the @code{target}
16231 @node VxWorks Download
16232 @subsubsection VxWorks Download
16234 @cindex download to VxWorks
16235 If you have connected to the VxWorks target and you want to debug an
16236 object that has not yet been loaded, you can use the @value{GDBN}
16237 @code{load} command to download a file from Unix to VxWorks
16238 incrementally. The object file given as an argument to the @code{load}
16239 command is actually opened twice: first by the VxWorks target in order
16240 to download the code, then by @value{GDBN} in order to read the symbol
16241 table. This can lead to problems if the current working directories on
16242 the two systems differ. If both systems have NFS mounted the same
16243 filesystems, you can avoid these problems by using absolute paths.
16244 Otherwise, it is simplest to set the working directory on both systems
16245 to the directory in which the object file resides, and then to reference
16246 the file by its name, without any path. For instance, a program
16247 @file{prog.o} may reside in @file{@var{vxpath}/vw/demo/rdb} in VxWorks
16248 and in @file{@var{hostpath}/vw/demo/rdb} on the host. To load this
16249 program, type this on VxWorks:
16252 -> cd "@var{vxpath}/vw/demo/rdb"
16256 Then, in @value{GDBN}, type:
16259 (vxgdb) cd @var{hostpath}/vw/demo/rdb
16260 (vxgdb) load prog.o
16263 @value{GDBN} displays a response similar to this:
16266 Reading symbol data from wherever/vw/demo/rdb/prog.o... done.
16269 You can also use the @code{load} command to reload an object module
16270 after editing and recompiling the corresponding source file. Note that
16271 this makes @value{GDBN} delete all currently-defined breakpoints,
16272 auto-displays, and convenience variables, and to clear the value
16273 history. (This is necessary in order to preserve the integrity of
16274 debugger's data structures that reference the target system's symbol
16277 @node VxWorks Attach
16278 @subsubsection Running Tasks
16280 @cindex running VxWorks tasks
16281 You can also attach to an existing task using the @code{attach} command as
16285 (vxgdb) attach @var{task}
16289 where @var{task} is the VxWorks hexadecimal task ID. The task can be running
16290 or suspended when you attach to it. Running tasks are suspended at
16291 the time of attachment.
16293 @node Embedded Processors
16294 @section Embedded Processors
16296 This section goes into details specific to particular embedded
16299 @cindex send command to simulator
16300 Whenever a specific embedded processor has a simulator, @value{GDBN}
16301 allows to send an arbitrary command to the simulator.
16304 @item sim @var{command}
16305 @kindex sim@r{, a command}
16306 Send an arbitrary @var{command} string to the simulator. Consult the
16307 documentation for the specific simulator in use for information about
16308 acceptable commands.
16314 * M32R/D:: Renesas M32R/D
16315 * M68K:: Motorola M68K
16316 * MIPS Embedded:: MIPS Embedded
16317 * OpenRISC 1000:: OpenRisc 1000
16318 * PA:: HP PA Embedded
16319 * PowerPC Embedded:: PowerPC Embedded
16320 * Sparclet:: Tsqware Sparclet
16321 * Sparclite:: Fujitsu Sparclite
16322 * Z8000:: Zilog Z8000
16325 * Super-H:: Renesas Super-H
16334 @item target rdi @var{dev}
16335 ARM Angel monitor, via RDI library interface to ADP protocol. You may
16336 use this target to communicate with both boards running the Angel
16337 monitor, or with the EmbeddedICE JTAG debug device.
16340 @item target rdp @var{dev}
16345 @value{GDBN} provides the following ARM-specific commands:
16348 @item set arm disassembler
16350 This commands selects from a list of disassembly styles. The
16351 @code{"std"} style is the standard style.
16353 @item show arm disassembler
16355 Show the current disassembly style.
16357 @item set arm apcs32
16358 @cindex ARM 32-bit mode
16359 This command toggles ARM operation mode between 32-bit and 26-bit.
16361 @item show arm apcs32
16362 Display the current usage of the ARM 32-bit mode.
16364 @item set arm fpu @var{fputype}
16365 This command sets the ARM floating-point unit (FPU) type. The
16366 argument @var{fputype} can be one of these:
16370 Determine the FPU type by querying the OS ABI.
16372 Software FPU, with mixed-endian doubles on little-endian ARM
16375 GCC-compiled FPA co-processor.
16377 Software FPU with pure-endian doubles.
16383 Show the current type of the FPU.
16386 This command forces @value{GDBN} to use the specified ABI.
16389 Show the currently used ABI.
16391 @item set arm fallback-mode (arm|thumb|auto)
16392 @value{GDBN} uses the symbol table, when available, to determine
16393 whether instructions are ARM or Thumb. This command controls
16394 @value{GDBN}'s default behavior when the symbol table is not
16395 available. The default is @samp{auto}, which causes @value{GDBN} to
16396 use the current execution mode (from the @code{T} bit in the @code{CPSR}
16399 @item show arm fallback-mode
16400 Show the current fallback instruction mode.
16402 @item set arm force-mode (arm|thumb|auto)
16403 This command overrides use of the symbol table to determine whether
16404 instructions are ARM or Thumb. The default is @samp{auto}, which
16405 causes @value{GDBN} to use the symbol table and then the setting
16406 of @samp{set arm fallback-mode}.
16408 @item show arm force-mode
16409 Show the current forced instruction mode.
16411 @item set debug arm
16412 Toggle whether to display ARM-specific debugging messages from the ARM
16413 target support subsystem.
16415 @item show debug arm
16416 Show whether ARM-specific debugging messages are enabled.
16419 The following commands are available when an ARM target is debugged
16420 using the RDI interface:
16423 @item rdilogfile @r{[}@var{file}@r{]}
16425 @cindex ADP (Angel Debugger Protocol) logging
16426 Set the filename for the ADP (Angel Debugger Protocol) packet log.
16427 With an argument, sets the log file to the specified @var{file}. With
16428 no argument, show the current log file name. The default log file is
16431 @item rdilogenable @r{[}@var{arg}@r{]}
16432 @kindex rdilogenable
16433 Control logging of ADP packets. With an argument of 1 or @code{"yes"}
16434 enables logging, with an argument 0 or @code{"no"} disables it. With
16435 no arguments displays the current setting. When logging is enabled,
16436 ADP packets exchanged between @value{GDBN} and the RDI target device
16437 are logged to a file.
16439 @item set rdiromatzero
16440 @kindex set rdiromatzero
16441 @cindex ROM at zero address, RDI
16442 Tell @value{GDBN} whether the target has ROM at address 0. If on,
16443 vector catching is disabled, so that zero address can be used. If off
16444 (the default), vector catching is enabled. For this command to take
16445 effect, it needs to be invoked prior to the @code{target rdi} command.
16447 @item show rdiromatzero
16448 @kindex show rdiromatzero
16449 Show the current setting of ROM at zero address.
16451 @item set rdiheartbeat
16452 @kindex set rdiheartbeat
16453 @cindex RDI heartbeat
16454 Enable or disable RDI heartbeat packets. It is not recommended to
16455 turn on this option, since it confuses ARM and EPI JTAG interface, as
16456 well as the Angel monitor.
16458 @item show rdiheartbeat
16459 @kindex show rdiheartbeat
16460 Show the setting of RDI heartbeat packets.
16465 @subsection Renesas M32R/D and M32R/SDI
16468 @kindex target m32r
16469 @item target m32r @var{dev}
16470 Renesas M32R/D ROM monitor.
16472 @kindex target m32rsdi
16473 @item target m32rsdi @var{dev}
16474 Renesas M32R SDI server, connected via parallel port to the board.
16477 The following @value{GDBN} commands are specific to the M32R monitor:
16480 @item set download-path @var{path}
16481 @kindex set download-path
16482 @cindex find downloadable @sc{srec} files (M32R)
16483 Set the default path for finding downloadable @sc{srec} files.
16485 @item show download-path
16486 @kindex show download-path
16487 Show the default path for downloadable @sc{srec} files.
16489 @item set board-address @var{addr}
16490 @kindex set board-address
16491 @cindex M32-EVA target board address
16492 Set the IP address for the M32R-EVA target board.
16494 @item show board-address
16495 @kindex show board-address
16496 Show the current IP address of the target board.
16498 @item set server-address @var{addr}
16499 @kindex set server-address
16500 @cindex download server address (M32R)
16501 Set the IP address for the download server, which is the @value{GDBN}'s
16504 @item show server-address
16505 @kindex show server-address
16506 Display the IP address of the download server.
16508 @item upload @r{[}@var{file}@r{]}
16509 @kindex upload@r{, M32R}
16510 Upload the specified @sc{srec} @var{file} via the monitor's Ethernet
16511 upload capability. If no @var{file} argument is given, the current
16512 executable file is uploaded.
16514 @item tload @r{[}@var{file}@r{]}
16515 @kindex tload@r{, M32R}
16516 Test the @code{upload} command.
16519 The following commands are available for M32R/SDI:
16524 @cindex reset SDI connection, M32R
16525 This command resets the SDI connection.
16529 This command shows the SDI connection status.
16532 @kindex debug_chaos
16533 @cindex M32R/Chaos debugging
16534 Instructs the remote that M32R/Chaos debugging is to be used.
16536 @item use_debug_dma
16537 @kindex use_debug_dma
16538 Instructs the remote to use the DEBUG_DMA method of accessing memory.
16541 @kindex use_mon_code
16542 Instructs the remote to use the MON_CODE method of accessing memory.
16545 @kindex use_ib_break
16546 Instructs the remote to set breakpoints by IB break.
16548 @item use_dbt_break
16549 @kindex use_dbt_break
16550 Instructs the remote to set breakpoints by DBT.
16556 The Motorola m68k configuration includes ColdFire support, and a
16557 target command for the following ROM monitor.
16561 @kindex target dbug
16562 @item target dbug @var{dev}
16563 dBUG ROM monitor for Motorola ColdFire.
16567 @node MIPS Embedded
16568 @subsection MIPS Embedded
16570 @cindex MIPS boards
16571 @value{GDBN} can use the MIPS remote debugging protocol to talk to a
16572 MIPS board attached to a serial line. This is available when
16573 you configure @value{GDBN} with @samp{--target=mips-idt-ecoff}.
16576 Use these @value{GDBN} commands to specify the connection to your target board:
16579 @item target mips @var{port}
16580 @kindex target mips @var{port}
16581 To run a program on the board, start up @code{@value{GDBP}} with the
16582 name of your program as the argument. To connect to the board, use the
16583 command @samp{target mips @var{port}}, where @var{port} is the name of
16584 the serial port connected to the board. If the program has not already
16585 been downloaded to the board, you may use the @code{load} command to
16586 download it. You can then use all the usual @value{GDBN} commands.
16588 For example, this sequence connects to the target board through a serial
16589 port, and loads and runs a program called @var{prog} through the
16593 host$ @value{GDBP} @var{prog}
16594 @value{GDBN} is free software and @dots{}
16595 (@value{GDBP}) target mips /dev/ttyb
16596 (@value{GDBP}) load @var{prog}
16600 @item target mips @var{hostname}:@var{portnumber}
16601 On some @value{GDBN} host configurations, you can specify a TCP
16602 connection (for instance, to a serial line managed by a terminal
16603 concentrator) instead of a serial port, using the syntax
16604 @samp{@var{hostname}:@var{portnumber}}.
16606 @item target pmon @var{port}
16607 @kindex target pmon @var{port}
16610 @item target ddb @var{port}
16611 @kindex target ddb @var{port}
16612 NEC's DDB variant of PMON for Vr4300.
16614 @item target lsi @var{port}
16615 @kindex target lsi @var{port}
16616 LSI variant of PMON.
16618 @kindex target r3900
16619 @item target r3900 @var{dev}
16620 Densan DVE-R3900 ROM monitor for Toshiba R3900 Mips.
16622 @kindex target array
16623 @item target array @var{dev}
16624 Array Tech LSI33K RAID controller board.
16630 @value{GDBN} also supports these special commands for MIPS targets:
16633 @item set mipsfpu double
16634 @itemx set mipsfpu single
16635 @itemx set mipsfpu none
16636 @itemx set mipsfpu auto
16637 @itemx show mipsfpu
16638 @kindex set mipsfpu
16639 @kindex show mipsfpu
16640 @cindex MIPS remote floating point
16641 @cindex floating point, MIPS remote
16642 If your target board does not support the MIPS floating point
16643 coprocessor, you should use the command @samp{set mipsfpu none} (if you
16644 need this, you may wish to put the command in your @value{GDBN} init
16645 file). This tells @value{GDBN} how to find the return value of
16646 functions which return floating point values. It also allows
16647 @value{GDBN} to avoid saving the floating point registers when calling
16648 functions on the board. If you are using a floating point coprocessor
16649 with only single precision floating point support, as on the @sc{r4650}
16650 processor, use the command @samp{set mipsfpu single}. The default
16651 double precision floating point coprocessor may be selected using
16652 @samp{set mipsfpu double}.
16654 In previous versions the only choices were double precision or no
16655 floating point, so @samp{set mipsfpu on} will select double precision
16656 and @samp{set mipsfpu off} will select no floating point.
16658 As usual, you can inquire about the @code{mipsfpu} variable with
16659 @samp{show mipsfpu}.
16661 @item set timeout @var{seconds}
16662 @itemx set retransmit-timeout @var{seconds}
16663 @itemx show timeout
16664 @itemx show retransmit-timeout
16665 @cindex @code{timeout}, MIPS protocol
16666 @cindex @code{retransmit-timeout}, MIPS protocol
16667 @kindex set timeout
16668 @kindex show timeout
16669 @kindex set retransmit-timeout
16670 @kindex show retransmit-timeout
16671 You can control the timeout used while waiting for a packet, in the MIPS
16672 remote protocol, with the @code{set timeout @var{seconds}} command. The
16673 default is 5 seconds. Similarly, you can control the timeout used while
16674 waiting for an acknowledgment of a packet with the @code{set
16675 retransmit-timeout @var{seconds}} command. The default is 3 seconds.
16676 You can inspect both values with @code{show timeout} and @code{show
16677 retransmit-timeout}. (These commands are @emph{only} available when
16678 @value{GDBN} is configured for @samp{--target=mips-idt-ecoff}.)
16680 The timeout set by @code{set timeout} does not apply when @value{GDBN}
16681 is waiting for your program to stop. In that case, @value{GDBN} waits
16682 forever because it has no way of knowing how long the program is going
16683 to run before stopping.
16685 @item set syn-garbage-limit @var{num}
16686 @kindex set syn-garbage-limit@r{, MIPS remote}
16687 @cindex synchronize with remote MIPS target
16688 Limit the maximum number of characters @value{GDBN} should ignore when
16689 it tries to synchronize with the remote target. The default is 10
16690 characters. Setting the limit to -1 means there's no limit.
16692 @item show syn-garbage-limit
16693 @kindex show syn-garbage-limit@r{, MIPS remote}
16694 Show the current limit on the number of characters to ignore when
16695 trying to synchronize with the remote system.
16697 @item set monitor-prompt @var{prompt}
16698 @kindex set monitor-prompt@r{, MIPS remote}
16699 @cindex remote monitor prompt
16700 Tell @value{GDBN} to expect the specified @var{prompt} string from the
16701 remote monitor. The default depends on the target:
16711 @item show monitor-prompt
16712 @kindex show monitor-prompt@r{, MIPS remote}
16713 Show the current strings @value{GDBN} expects as the prompt from the
16716 @item set monitor-warnings
16717 @kindex set monitor-warnings@r{, MIPS remote}
16718 Enable or disable monitor warnings about hardware breakpoints. This
16719 has effect only for the @code{lsi} target. When on, @value{GDBN} will
16720 display warning messages whose codes are returned by the @code{lsi}
16721 PMON monitor for breakpoint commands.
16723 @item show monitor-warnings
16724 @kindex show monitor-warnings@r{, MIPS remote}
16725 Show the current setting of printing monitor warnings.
16727 @item pmon @var{command}
16728 @kindex pmon@r{, MIPS remote}
16729 @cindex send PMON command
16730 This command allows sending an arbitrary @var{command} string to the
16731 monitor. The monitor must be in debug mode for this to work.
16734 @node OpenRISC 1000
16735 @subsection OpenRISC 1000
16736 @cindex OpenRISC 1000
16738 @cindex or1k boards
16739 See OR1k Architecture document (@uref{www.opencores.org}) for more information
16740 about platform and commands.
16744 @kindex target jtag
16745 @item target jtag jtag://@var{host}:@var{port}
16747 Connects to remote JTAG server.
16748 JTAG remote server can be either an or1ksim or JTAG server,
16749 connected via parallel port to the board.
16751 Example: @code{target jtag jtag://localhost:9999}
16754 @item or1ksim @var{command}
16755 If connected to @code{or1ksim} OpenRISC 1000 Architectural
16756 Simulator, proprietary commands can be executed.
16758 @kindex info or1k spr
16759 @item info or1k spr
16760 Displays spr groups.
16762 @item info or1k spr @var{group}
16763 @itemx info or1k spr @var{groupno}
16764 Displays register names in selected group.
16766 @item info or1k spr @var{group} @var{register}
16767 @itemx info or1k spr @var{register}
16768 @itemx info or1k spr @var{groupno} @var{registerno}
16769 @itemx info or1k spr @var{registerno}
16770 Shows information about specified spr register.
16773 @item spr @var{group} @var{register} @var{value}
16774 @itemx spr @var{register @var{value}}
16775 @itemx spr @var{groupno} @var{registerno @var{value}}
16776 @itemx spr @var{registerno @var{value}}
16777 Writes @var{value} to specified spr register.
16780 Some implementations of OpenRISC 1000 Architecture also have hardware trace.
16781 It is very similar to @value{GDBN} trace, except it does not interfere with normal
16782 program execution and is thus much faster. Hardware breakpoints/watchpoint
16783 triggers can be set using:
16786 Load effective address/data
16788 Store effective address/data
16790 Access effective address ($SEA or $LEA) or data ($SDATA/$LDATA)
16795 When triggered, it can capture low level data, like: @code{PC}, @code{LSEA},
16796 @code{LDATA}, @code{SDATA}, @code{READSPR}, @code{WRITESPR}, @code{INSTR}.
16798 @code{htrace} commands:
16799 @cindex OpenRISC 1000 htrace
16802 @item hwatch @var{conditional}
16803 Set hardware watchpoint on combination of Load/Store Effective Address(es)
16804 or Data. For example:
16806 @code{hwatch ($LEA == my_var) && ($LDATA < 50) || ($SEA == my_var) && ($SDATA >= 50)}
16808 @code{hwatch ($LEA == my_var) && ($LDATA < 50) || ($SEA == my_var) && ($SDATA >= 50)}
16812 Display information about current HW trace configuration.
16814 @item htrace trigger @var{conditional}
16815 Set starting criteria for HW trace.
16817 @item htrace qualifier @var{conditional}
16818 Set acquisition qualifier for HW trace.
16820 @item htrace stop @var{conditional}
16821 Set HW trace stopping criteria.
16823 @item htrace record [@var{data}]*
16824 Selects the data to be recorded, when qualifier is met and HW trace was
16827 @item htrace enable
16828 @itemx htrace disable
16829 Enables/disables the HW trace.
16831 @item htrace rewind [@var{filename}]
16832 Clears currently recorded trace data.
16834 If filename is specified, new trace file is made and any newly collected data
16835 will be written there.
16837 @item htrace print [@var{start} [@var{len}]]
16838 Prints trace buffer, using current record configuration.
16840 @item htrace mode continuous
16841 Set continuous trace mode.
16843 @item htrace mode suspend
16844 Set suspend trace mode.
16848 @node PowerPC Embedded
16849 @subsection PowerPC Embedded
16851 @value{GDBN} provides the following PowerPC-specific commands:
16854 @kindex set powerpc
16855 @item set powerpc soft-float
16856 @itemx show powerpc soft-float
16857 Force @value{GDBN} to use (or not use) a software floating point calling
16858 convention. By default, @value{GDBN} selects the calling convention based
16859 on the selected architecture and the provided executable file.
16861 @item set powerpc vector-abi
16862 @itemx show powerpc vector-abi
16863 Force @value{GDBN} to use the specified calling convention for vector
16864 arguments and return values. The valid options are @samp{auto};
16865 @samp{generic}, to avoid vector registers even if they are present;
16866 @samp{altivec}, to use AltiVec registers; and @samp{spe} to use SPE
16867 registers. By default, @value{GDBN} selects the calling convention
16868 based on the selected architecture and the provided executable file.
16870 @kindex target dink32
16871 @item target dink32 @var{dev}
16872 DINK32 ROM monitor.
16874 @kindex target ppcbug
16875 @item target ppcbug @var{dev}
16876 @kindex target ppcbug1
16877 @item target ppcbug1 @var{dev}
16878 PPCBUG ROM monitor for PowerPC.
16881 @item target sds @var{dev}
16882 SDS monitor, running on a PowerPC board (such as Motorola's ADS).
16885 @cindex SDS protocol
16886 The following commands specific to the SDS protocol are supported
16890 @item set sdstimeout @var{nsec}
16891 @kindex set sdstimeout
16892 Set the timeout for SDS protocol reads to be @var{nsec} seconds. The
16893 default is 2 seconds.
16895 @item show sdstimeout
16896 @kindex show sdstimeout
16897 Show the current value of the SDS timeout.
16899 @item sds @var{command}
16900 @kindex sds@r{, a command}
16901 Send the specified @var{command} string to the SDS monitor.
16906 @subsection HP PA Embedded
16910 @kindex target op50n
16911 @item target op50n @var{dev}
16912 OP50N monitor, running on an OKI HPPA board.
16914 @kindex target w89k
16915 @item target w89k @var{dev}
16916 W89K monitor, running on a Winbond HPPA board.
16921 @subsection Tsqware Sparclet
16925 @value{GDBN} enables developers to debug tasks running on
16926 Sparclet targets from a Unix host.
16927 @value{GDBN} uses code that runs on
16928 both the Unix host and on the Sparclet target. The program
16929 @code{@value{GDBP}} is installed and executed on the Unix host.
16932 @item remotetimeout @var{args}
16933 @kindex remotetimeout
16934 @value{GDBN} supports the option @code{remotetimeout}.
16935 This option is set by the user, and @var{args} represents the number of
16936 seconds @value{GDBN} waits for responses.
16939 @cindex compiling, on Sparclet
16940 When compiling for debugging, include the options @samp{-g} to get debug
16941 information and @samp{-Ttext} to relocate the program to where you wish to
16942 load it on the target. You may also want to add the options @samp{-n} or
16943 @samp{-N} in order to reduce the size of the sections. Example:
16946 sparclet-aout-gcc prog.c -Ttext 0x12010000 -g -o prog -N
16949 You can use @code{objdump} to verify that the addresses are what you intended:
16952 sparclet-aout-objdump --headers --syms prog
16955 @cindex running, on Sparclet
16957 your Unix execution search path to find @value{GDBN}, you are ready to
16958 run @value{GDBN}. From your Unix host, run @code{@value{GDBP}}
16959 (or @code{sparclet-aout-gdb}, depending on your installation).
16961 @value{GDBN} comes up showing the prompt:
16968 * Sparclet File:: Setting the file to debug
16969 * Sparclet Connection:: Connecting to Sparclet
16970 * Sparclet Download:: Sparclet download
16971 * Sparclet Execution:: Running and debugging
16974 @node Sparclet File
16975 @subsubsection Setting File to Debug
16977 The @value{GDBN} command @code{file} lets you choose with program to debug.
16980 (gdbslet) file prog
16984 @value{GDBN} then attempts to read the symbol table of @file{prog}.
16985 @value{GDBN} locates
16986 the file by searching the directories listed in the command search
16988 If the file was compiled with debug information (option @samp{-g}), source
16989 files will be searched as well.
16990 @value{GDBN} locates
16991 the source files by searching the directories listed in the directory search
16992 path (@pxref{Environment, ,Your Program's Environment}).
16994 to find a file, it displays a message such as:
16997 prog: No such file or directory.
17000 When this happens, add the appropriate directories to the search paths with
17001 the @value{GDBN} commands @code{path} and @code{dir}, and execute the
17002 @code{target} command again.
17004 @node Sparclet Connection
17005 @subsubsection Connecting to Sparclet
17007 The @value{GDBN} command @code{target} lets you connect to a Sparclet target.
17008 To connect to a target on serial port ``@code{ttya}'', type:
17011 (gdbslet) target sparclet /dev/ttya
17012 Remote target sparclet connected to /dev/ttya
17013 main () at ../prog.c:3
17017 @value{GDBN} displays messages like these:
17023 @node Sparclet Download
17024 @subsubsection Sparclet Download
17026 @cindex download to Sparclet
17027 Once connected to the Sparclet target,
17028 you can use the @value{GDBN}
17029 @code{load} command to download the file from the host to the target.
17030 The file name and load offset should be given as arguments to the @code{load}
17032 Since the file format is aout, the program must be loaded to the starting
17033 address. You can use @code{objdump} to find out what this value is. The load
17034 offset is an offset which is added to the VMA (virtual memory address)
17035 of each of the file's sections.
17036 For instance, if the program
17037 @file{prog} was linked to text address 0x1201000, with data at 0x12010160
17038 and bss at 0x12010170, in @value{GDBN}, type:
17041 (gdbslet) load prog 0x12010000
17042 Loading section .text, size 0xdb0 vma 0x12010000
17045 If the code is loaded at a different address then what the program was linked
17046 to, you may need to use the @code{section} and @code{add-symbol-file} commands
17047 to tell @value{GDBN} where to map the symbol table.
17049 @node Sparclet Execution
17050 @subsubsection Running and Debugging
17052 @cindex running and debugging Sparclet programs
17053 You can now begin debugging the task using @value{GDBN}'s execution control
17054 commands, @code{b}, @code{step}, @code{run}, etc. See the @value{GDBN}
17055 manual for the list of commands.
17059 Breakpoint 1 at 0x12010000: file prog.c, line 3.
17061 Starting program: prog
17062 Breakpoint 1, main (argc=1, argv=0xeffff21c) at prog.c:3
17063 3 char *symarg = 0;
17065 4 char *execarg = "hello!";
17070 @subsection Fujitsu Sparclite
17074 @kindex target sparclite
17075 @item target sparclite @var{dev}
17076 Fujitsu sparclite boards, used only for the purpose of loading.
17077 You must use an additional command to debug the program.
17078 For example: target remote @var{dev} using @value{GDBN} standard
17084 @subsection Zilog Z8000
17087 @cindex simulator, Z8000
17088 @cindex Zilog Z8000 simulator
17090 When configured for debugging Zilog Z8000 targets, @value{GDBN} includes
17093 For the Z8000 family, @samp{target sim} simulates either the Z8002 (the
17094 unsegmented variant of the Z8000 architecture) or the Z8001 (the
17095 segmented variant). The simulator recognizes which architecture is
17096 appropriate by inspecting the object code.
17099 @item target sim @var{args}
17101 @kindex target sim@r{, with Z8000}
17102 Debug programs on a simulated CPU. If the simulator supports setup
17103 options, specify them via @var{args}.
17107 After specifying this target, you can debug programs for the simulated
17108 CPU in the same style as programs for your host computer; use the
17109 @code{file} command to load a new program image, the @code{run} command
17110 to run your program, and so on.
17112 As well as making available all the usual machine registers
17113 (@pxref{Registers, ,Registers}), the Z8000 simulator provides three
17114 additional items of information as specially named registers:
17119 Counts clock-ticks in the simulator.
17122 Counts instructions run in the simulator.
17125 Execution time in 60ths of a second.
17129 You can refer to these values in @value{GDBN} expressions with the usual
17130 conventions; for example, @w{@samp{b fputc if $cycles>5000}} sets a
17131 conditional breakpoint that suspends only after at least 5000
17132 simulated clock ticks.
17135 @subsection Atmel AVR
17138 When configured for debugging the Atmel AVR, @value{GDBN} supports the
17139 following AVR-specific commands:
17142 @item info io_registers
17143 @kindex info io_registers@r{, AVR}
17144 @cindex I/O registers (Atmel AVR)
17145 This command displays information about the AVR I/O registers. For
17146 each register, @value{GDBN} prints its number and value.
17153 When configured for debugging CRIS, @value{GDBN} provides the
17154 following CRIS-specific commands:
17157 @item set cris-version @var{ver}
17158 @cindex CRIS version
17159 Set the current CRIS version to @var{ver}, either @samp{10} or @samp{32}.
17160 The CRIS version affects register names and sizes. This command is useful in
17161 case autodetection of the CRIS version fails.
17163 @item show cris-version
17164 Show the current CRIS version.
17166 @item set cris-dwarf2-cfi
17167 @cindex DWARF-2 CFI and CRIS
17168 Set the usage of DWARF-2 CFI for CRIS debugging. The default is @samp{on}.
17169 Change to @samp{off} when using @code{gcc-cris} whose version is below
17172 @item show cris-dwarf2-cfi
17173 Show the current state of using DWARF-2 CFI.
17175 @item set cris-mode @var{mode}
17177 Set the current CRIS mode to @var{mode}. It should only be changed when
17178 debugging in guru mode, in which case it should be set to
17179 @samp{guru} (the default is @samp{normal}).
17181 @item show cris-mode
17182 Show the current CRIS mode.
17186 @subsection Renesas Super-H
17189 For the Renesas Super-H processor, @value{GDBN} provides these
17194 @kindex regs@r{, Super-H}
17195 Show the values of all Super-H registers.
17197 @item set sh calling-convention @var{convention}
17198 @kindex set sh calling-convention
17199 Set the calling-convention used when calling functions from @value{GDBN}.
17200 Allowed values are @samp{gcc}, which is the default setting, and @samp{renesas}.
17201 With the @samp{gcc} setting, functions are called using the @value{NGCC} calling
17202 convention. If the DWARF-2 information of the called function specifies
17203 that the function follows the Renesas calling convention, the function
17204 is called using the Renesas calling convention. If the calling convention
17205 is set to @samp{renesas}, the Renesas calling convention is always used,
17206 regardless of the DWARF-2 information. This can be used to override the
17207 default of @samp{gcc} if debug information is missing, or the compiler
17208 does not emit the DWARF-2 calling convention entry for a function.
17210 @item show sh calling-convention
17211 @kindex show sh calling-convention
17212 Show the current calling convention setting.
17217 @node Architectures
17218 @section Architectures
17220 This section describes characteristics of architectures that affect
17221 all uses of @value{GDBN} with the architecture, both native and cross.
17228 * HPPA:: HP PA architecture
17229 * SPU:: Cell Broadband Engine SPU architecture
17234 @subsection x86 Architecture-specific Issues
17237 @item set struct-convention @var{mode}
17238 @kindex set struct-convention
17239 @cindex struct return convention
17240 @cindex struct/union returned in registers
17241 Set the convention used by the inferior to return @code{struct}s and
17242 @code{union}s from functions to @var{mode}. Possible values of
17243 @var{mode} are @code{"pcc"}, @code{"reg"}, and @code{"default"} (the
17244 default). @code{"default"} or @code{"pcc"} means that @code{struct}s
17245 are returned on the stack, while @code{"reg"} means that a
17246 @code{struct} or a @code{union} whose size is 1, 2, 4, or 8 bytes will
17247 be returned in a register.
17249 @item show struct-convention
17250 @kindex show struct-convention
17251 Show the current setting of the convention to return @code{struct}s
17260 @kindex set rstack_high_address
17261 @cindex AMD 29K register stack
17262 @cindex register stack, AMD29K
17263 @item set rstack_high_address @var{address}
17264 On AMD 29000 family processors, registers are saved in a separate
17265 @dfn{register stack}. There is no way for @value{GDBN} to determine the
17266 extent of this stack. Normally, @value{GDBN} just assumes that the
17267 stack is ``large enough''. This may result in @value{GDBN} referencing
17268 memory locations that do not exist. If necessary, you can get around
17269 this problem by specifying the ending address of the register stack with
17270 the @code{set rstack_high_address} command. The argument should be an
17271 address, which you probably want to precede with @samp{0x} to specify in
17274 @kindex show rstack_high_address
17275 @item show rstack_high_address
17276 Display the current limit of the register stack, on AMD 29000 family
17284 See the following section.
17289 @cindex stack on Alpha
17290 @cindex stack on MIPS
17291 @cindex Alpha stack
17293 Alpha- and MIPS-based computers use an unusual stack frame, which
17294 sometimes requires @value{GDBN} to search backward in the object code to
17295 find the beginning of a function.
17297 @cindex response time, MIPS debugging
17298 To improve response time (especially for embedded applications, where
17299 @value{GDBN} may be restricted to a slow serial line for this search)
17300 you may want to limit the size of this search, using one of these
17304 @cindex @code{heuristic-fence-post} (Alpha, MIPS)
17305 @item set heuristic-fence-post @var{limit}
17306 Restrict @value{GDBN} to examining at most @var{limit} bytes in its
17307 search for the beginning of a function. A value of @var{0} (the
17308 default) means there is no limit. However, except for @var{0}, the
17309 larger the limit the more bytes @code{heuristic-fence-post} must search
17310 and therefore the longer it takes to run. You should only need to use
17311 this command when debugging a stripped executable.
17313 @item show heuristic-fence-post
17314 Display the current limit.
17318 These commands are available @emph{only} when @value{GDBN} is configured
17319 for debugging programs on Alpha or MIPS processors.
17321 Several MIPS-specific commands are available when debugging MIPS
17325 @item set mips abi @var{arg}
17326 @kindex set mips abi
17327 @cindex set ABI for MIPS
17328 Tell @value{GDBN} which MIPS ABI is used by the inferior. Possible
17329 values of @var{arg} are:
17333 The default ABI associated with the current binary (this is the
17344 @item show mips abi
17345 @kindex show mips abi
17346 Show the MIPS ABI used by @value{GDBN} to debug the inferior.
17349 @itemx show mipsfpu
17350 @xref{MIPS Embedded, set mipsfpu}.
17352 @item set mips mask-address @var{arg}
17353 @kindex set mips mask-address
17354 @cindex MIPS addresses, masking
17355 This command determines whether the most-significant 32 bits of 64-bit
17356 MIPS addresses are masked off. The argument @var{arg} can be
17357 @samp{on}, @samp{off}, or @samp{auto}. The latter is the default
17358 setting, which lets @value{GDBN} determine the correct value.
17360 @item show mips mask-address
17361 @kindex show mips mask-address
17362 Show whether the upper 32 bits of MIPS addresses are masked off or
17365 @item set remote-mips64-transfers-32bit-regs
17366 @kindex set remote-mips64-transfers-32bit-regs
17367 This command controls compatibility with 64-bit MIPS targets that
17368 transfer data in 32-bit quantities. If you have an old MIPS 64 target
17369 that transfers 32 bits for some registers, like @sc{sr} and @sc{fsr},
17370 and 64 bits for other registers, set this option to @samp{on}.
17372 @item show remote-mips64-transfers-32bit-regs
17373 @kindex show remote-mips64-transfers-32bit-regs
17374 Show the current setting of compatibility with older MIPS 64 targets.
17376 @item set debug mips
17377 @kindex set debug mips
17378 This command turns on and off debugging messages for the MIPS-specific
17379 target code in @value{GDBN}.
17381 @item show debug mips
17382 @kindex show debug mips
17383 Show the current setting of MIPS debugging messages.
17389 @cindex HPPA support
17391 When @value{GDBN} is debugging the HP PA architecture, it provides the
17392 following special commands:
17395 @item set debug hppa
17396 @kindex set debug hppa
17397 This command determines whether HPPA architecture-specific debugging
17398 messages are to be displayed.
17400 @item show debug hppa
17401 Show whether HPPA debugging messages are displayed.
17403 @item maint print unwind @var{address}
17404 @kindex maint print unwind@r{, HPPA}
17405 This command displays the contents of the unwind table entry at the
17406 given @var{address}.
17412 @subsection Cell Broadband Engine SPU architecture
17413 @cindex Cell Broadband Engine
17416 When @value{GDBN} is debugging the Cell Broadband Engine SPU architecture,
17417 it provides the following special commands:
17420 @item info spu event
17422 Display SPU event facility status. Shows current event mask
17423 and pending event status.
17425 @item info spu signal
17426 Display SPU signal notification facility status. Shows pending
17427 signal-control word and signal notification mode of both signal
17428 notification channels.
17430 @item info spu mailbox
17431 Display SPU mailbox facility status. Shows all pending entries,
17432 in order of processing, in each of the SPU Write Outbound,
17433 SPU Write Outbound Interrupt, and SPU Read Inbound mailboxes.
17436 Display MFC DMA status. Shows all pending commands in the MFC
17437 DMA queue. For each entry, opcode, tag, class IDs, effective
17438 and local store addresses and transfer size are shown.
17440 @item info spu proxydma
17441 Display MFC Proxy-DMA status. Shows all pending commands in the MFC
17442 Proxy-DMA queue. For each entry, opcode, tag, class IDs, effective
17443 and local store addresses and transfer size are shown.
17448 @subsection PowerPC
17449 @cindex PowerPC architecture
17451 When @value{GDBN} is debugging the PowerPC architecture, it provides a set of
17452 pseudo-registers to enable inspection of 128-bit wide Decimal Floating Point
17453 numbers stored in the floating point registers. These values must be stored
17454 in two consecutive registers, always starting at an even register like
17455 @code{f0} or @code{f2}.
17457 The pseudo-registers go from @code{$dl0} through @code{$dl15}, and are formed
17458 by joining the even/odd register pairs @code{f0} and @code{f1} for @code{$dl0},
17459 @code{f2} and @code{f3} for @code{$dl1} and so on.
17461 For POWER7 processors, @value{GDBN} provides a set of pseudo-registers, the 64-bit
17462 wide Extended Floating Point Registers (@samp{f32} through @samp{f63}).
17465 @node Controlling GDB
17466 @chapter Controlling @value{GDBN}
17468 You can alter the way @value{GDBN} interacts with you by using the
17469 @code{set} command. For commands controlling how @value{GDBN} displays
17470 data, see @ref{Print Settings, ,Print Settings}. Other settings are
17475 * Editing:: Command editing
17476 * Command History:: Command history
17477 * Screen Size:: Screen size
17478 * Numbers:: Numbers
17479 * ABI:: Configuring the current ABI
17480 * Messages/Warnings:: Optional warnings and messages
17481 * Debugging Output:: Optional messages about internal happenings
17489 @value{GDBN} indicates its readiness to read a command by printing a string
17490 called the @dfn{prompt}. This string is normally @samp{(@value{GDBP})}. You
17491 can change the prompt string with the @code{set prompt} command. For
17492 instance, when debugging @value{GDBN} with @value{GDBN}, it is useful to change
17493 the prompt in one of the @value{GDBN} sessions so that you can always tell
17494 which one you are talking to.
17496 @emph{Note:} @code{set prompt} does not add a space for you after the
17497 prompt you set. This allows you to set a prompt which ends in a space
17498 or a prompt that does not.
17502 @item set prompt @var{newprompt}
17503 Directs @value{GDBN} to use @var{newprompt} as its prompt string henceforth.
17505 @kindex show prompt
17507 Prints a line of the form: @samp{Gdb's prompt is: @var{your-prompt}}
17511 @section Command Editing
17513 @cindex command line editing
17515 @value{GDBN} reads its input commands via the @dfn{Readline} interface. This
17516 @sc{gnu} library provides consistent behavior for programs which provide a
17517 command line interface to the user. Advantages are @sc{gnu} Emacs-style
17518 or @dfn{vi}-style inline editing of commands, @code{csh}-like history
17519 substitution, and a storage and recall of command history across
17520 debugging sessions.
17522 You may control the behavior of command line editing in @value{GDBN} with the
17523 command @code{set}.
17526 @kindex set editing
17529 @itemx set editing on
17530 Enable command line editing (enabled by default).
17532 @item set editing off
17533 Disable command line editing.
17535 @kindex show editing
17537 Show whether command line editing is enabled.
17540 @xref{Command Line Editing}, for more details about the Readline
17541 interface. Users unfamiliar with @sc{gnu} Emacs or @code{vi} are
17542 encouraged to read that chapter.
17544 @node Command History
17545 @section Command History
17546 @cindex command history
17548 @value{GDBN} can keep track of the commands you type during your
17549 debugging sessions, so that you can be certain of precisely what
17550 happened. Use these commands to manage the @value{GDBN} command
17553 @value{GDBN} uses the @sc{gnu} History library, a part of the Readline
17554 package, to provide the history facility. @xref{Using History
17555 Interactively}, for the detailed description of the History library.
17557 To issue a command to @value{GDBN} without affecting certain aspects of
17558 the state which is seen by users, prefix it with @samp{server }
17559 (@pxref{Server Prefix}). This
17560 means that this command will not affect the command history, nor will it
17561 affect @value{GDBN}'s notion of which command to repeat if @key{RET} is
17562 pressed on a line by itself.
17564 @cindex @code{server}, command prefix
17565 The server prefix does not affect the recording of values into the value
17566 history; to print a value without recording it into the value history,
17567 use the @code{output} command instead of the @code{print} command.
17569 Here is the description of @value{GDBN} commands related to command
17573 @cindex history substitution
17574 @cindex history file
17575 @kindex set history filename
17576 @cindex @env{GDBHISTFILE}, environment variable
17577 @item set history filename @var{fname}
17578 Set the name of the @value{GDBN} command history file to @var{fname}.
17579 This is the file where @value{GDBN} reads an initial command history
17580 list, and where it writes the command history from this session when it
17581 exits. You can access this list through history expansion or through
17582 the history command editing characters listed below. This file defaults
17583 to the value of the environment variable @code{GDBHISTFILE}, or to
17584 @file{./.gdb_history} (@file{./_gdb_history} on MS-DOS) if this variable
17587 @cindex save command history
17588 @kindex set history save
17589 @item set history save
17590 @itemx set history save on
17591 Record command history in a file, whose name may be specified with the
17592 @code{set history filename} command. By default, this option is disabled.
17594 @item set history save off
17595 Stop recording command history in a file.
17597 @cindex history size
17598 @kindex set history size
17599 @cindex @env{HISTSIZE}, environment variable
17600 @item set history size @var{size}
17601 Set the number of commands which @value{GDBN} keeps in its history list.
17602 This defaults to the value of the environment variable
17603 @code{HISTSIZE}, or to 256 if this variable is not set.
17606 History expansion assigns special meaning to the character @kbd{!}.
17607 @xref{Event Designators}, for more details.
17609 @cindex history expansion, turn on/off
17610 Since @kbd{!} is also the logical not operator in C, history expansion
17611 is off by default. If you decide to enable history expansion with the
17612 @code{set history expansion on} command, you may sometimes need to
17613 follow @kbd{!} (when it is used as logical not, in an expression) with
17614 a space or a tab to prevent it from being expanded. The readline
17615 history facilities do not attempt substitution on the strings
17616 @kbd{!=} and @kbd{!(}, even when history expansion is enabled.
17618 The commands to control history expansion are:
17621 @item set history expansion on
17622 @itemx set history expansion
17623 @kindex set history expansion
17624 Enable history expansion. History expansion is off by default.
17626 @item set history expansion off
17627 Disable history expansion.
17630 @kindex show history
17632 @itemx show history filename
17633 @itemx show history save
17634 @itemx show history size
17635 @itemx show history expansion
17636 These commands display the state of the @value{GDBN} history parameters.
17637 @code{show history} by itself displays all four states.
17642 @kindex show commands
17643 @cindex show last commands
17644 @cindex display command history
17645 @item show commands
17646 Display the last ten commands in the command history.
17648 @item show commands @var{n}
17649 Print ten commands centered on command number @var{n}.
17651 @item show commands +
17652 Print ten commands just after the commands last printed.
17656 @section Screen Size
17657 @cindex size of screen
17658 @cindex pauses in output
17660 Certain commands to @value{GDBN} may produce large amounts of
17661 information output to the screen. To help you read all of it,
17662 @value{GDBN} pauses and asks you for input at the end of each page of
17663 output. Type @key{RET} when you want to continue the output, or @kbd{q}
17664 to discard the remaining output. Also, the screen width setting
17665 determines when to wrap lines of output. Depending on what is being
17666 printed, @value{GDBN} tries to break the line at a readable place,
17667 rather than simply letting it overflow onto the following line.
17669 Normally @value{GDBN} knows the size of the screen from the terminal
17670 driver software. For example, on Unix @value{GDBN} uses the termcap data base
17671 together with the value of the @code{TERM} environment variable and the
17672 @code{stty rows} and @code{stty cols} settings. If this is not correct,
17673 you can override it with the @code{set height} and @code{set
17680 @kindex show height
17681 @item set height @var{lpp}
17683 @itemx set width @var{cpl}
17685 These @code{set} commands specify a screen height of @var{lpp} lines and
17686 a screen width of @var{cpl} characters. The associated @code{show}
17687 commands display the current settings.
17689 If you specify a height of zero lines, @value{GDBN} does not pause during
17690 output no matter how long the output is. This is useful if output is to a
17691 file or to an editor buffer.
17693 Likewise, you can specify @samp{set width 0} to prevent @value{GDBN}
17694 from wrapping its output.
17696 @item set pagination on
17697 @itemx set pagination off
17698 @kindex set pagination
17699 Turn the output pagination on or off; the default is on. Turning
17700 pagination off is the alternative to @code{set height 0}.
17702 @item show pagination
17703 @kindex show pagination
17704 Show the current pagination mode.
17709 @cindex number representation
17710 @cindex entering numbers
17712 You can always enter numbers in octal, decimal, or hexadecimal in
17713 @value{GDBN} by the usual conventions: octal numbers begin with
17714 @samp{0}, decimal numbers end with @samp{.}, and hexadecimal numbers
17715 begin with @samp{0x}. Numbers that neither begin with @samp{0} or
17716 @samp{0x}, nor end with a @samp{.} are, by default, entered in base
17717 10; likewise, the default display for numbers---when no particular
17718 format is specified---is base 10. You can change the default base for
17719 both input and output with the commands described below.
17722 @kindex set input-radix
17723 @item set input-radix @var{base}
17724 Set the default base for numeric input. Supported choices
17725 for @var{base} are decimal 8, 10, or 16. @var{base} must itself be
17726 specified either unambiguously or using the current input radix; for
17730 set input-radix 012
17731 set input-radix 10.
17732 set input-radix 0xa
17736 sets the input base to decimal. On the other hand, @samp{set input-radix 10}
17737 leaves the input radix unchanged, no matter what it was, since
17738 @samp{10}, being without any leading or trailing signs of its base, is
17739 interpreted in the current radix. Thus, if the current radix is 16,
17740 @samp{10} is interpreted in hex, i.e.@: as 16 decimal, which doesn't
17743 @kindex set output-radix
17744 @item set output-radix @var{base}
17745 Set the default base for numeric display. Supported choices
17746 for @var{base} are decimal 8, 10, or 16. @var{base} must itself be
17747 specified either unambiguously or using the current input radix.
17749 @kindex show input-radix
17750 @item show input-radix
17751 Display the current default base for numeric input.
17753 @kindex show output-radix
17754 @item show output-radix
17755 Display the current default base for numeric display.
17757 @item set radix @r{[}@var{base}@r{]}
17761 These commands set and show the default base for both input and output
17762 of numbers. @code{set radix} sets the radix of input and output to
17763 the same base; without an argument, it resets the radix back to its
17764 default value of 10.
17769 @section Configuring the Current ABI
17771 @value{GDBN} can determine the @dfn{ABI} (Application Binary Interface) of your
17772 application automatically. However, sometimes you need to override its
17773 conclusions. Use these commands to manage @value{GDBN}'s view of the
17780 One @value{GDBN} configuration can debug binaries for multiple operating
17781 system targets, either via remote debugging or native emulation.
17782 @value{GDBN} will autodetect the @dfn{OS ABI} (Operating System ABI) in use,
17783 but you can override its conclusion using the @code{set osabi} command.
17784 One example where this is useful is in debugging of binaries which use
17785 an alternate C library (e.g.@: @sc{uClibc} for @sc{gnu}/Linux) which does
17786 not have the same identifying marks that the standard C library for your
17791 Show the OS ABI currently in use.
17794 With no argument, show the list of registered available OS ABI's.
17796 @item set osabi @var{abi}
17797 Set the current OS ABI to @var{abi}.
17800 @cindex float promotion
17802 Generally, the way that an argument of type @code{float} is passed to a
17803 function depends on whether the function is prototyped. For a prototyped
17804 (i.e.@: ANSI/ISO style) function, @code{float} arguments are passed unchanged,
17805 according to the architecture's convention for @code{float}. For unprototyped
17806 (i.e.@: K&R style) functions, @code{float} arguments are first promoted to type
17807 @code{double} and then passed.
17809 Unfortunately, some forms of debug information do not reliably indicate whether
17810 a function is prototyped. If @value{GDBN} calls a function that is not marked
17811 as prototyped, it consults @kbd{set coerce-float-to-double}.
17814 @kindex set coerce-float-to-double
17815 @item set coerce-float-to-double
17816 @itemx set coerce-float-to-double on
17817 Arguments of type @code{float} will be promoted to @code{double} when passed
17818 to an unprototyped function. This is the default setting.
17820 @item set coerce-float-to-double off
17821 Arguments of type @code{float} will be passed directly to unprototyped
17824 @kindex show coerce-float-to-double
17825 @item show coerce-float-to-double
17826 Show the current setting of promoting @code{float} to @code{double}.
17830 @kindex show cp-abi
17831 @value{GDBN} needs to know the ABI used for your program's C@t{++}
17832 objects. The correct C@t{++} ABI depends on which C@t{++} compiler was
17833 used to build your application. @value{GDBN} only fully supports
17834 programs with a single C@t{++} ABI; if your program contains code using
17835 multiple C@t{++} ABI's or if @value{GDBN} can not identify your
17836 program's ABI correctly, you can tell @value{GDBN} which ABI to use.
17837 Currently supported ABI's include ``gnu-v2'', for @code{g++} versions
17838 before 3.0, ``gnu-v3'', for @code{g++} versions 3.0 and later, and
17839 ``hpaCC'' for the HP ANSI C@t{++} compiler. Other C@t{++} compilers may
17840 use the ``gnu-v2'' or ``gnu-v3'' ABI's as well. The default setting is
17845 Show the C@t{++} ABI currently in use.
17848 With no argument, show the list of supported C@t{++} ABI's.
17850 @item set cp-abi @var{abi}
17851 @itemx set cp-abi auto
17852 Set the current C@t{++} ABI to @var{abi}, or return to automatic detection.
17855 @node Messages/Warnings
17856 @section Optional Warnings and Messages
17858 @cindex verbose operation
17859 @cindex optional warnings
17860 By default, @value{GDBN} is silent about its inner workings. If you are
17861 running on a slow machine, you may want to use the @code{set verbose}
17862 command. This makes @value{GDBN} tell you when it does a lengthy
17863 internal operation, so you will not think it has crashed.
17865 Currently, the messages controlled by @code{set verbose} are those
17866 which announce that the symbol table for a source file is being read;
17867 see @code{symbol-file} in @ref{Files, ,Commands to Specify Files}.
17870 @kindex set verbose
17871 @item set verbose on
17872 Enables @value{GDBN} output of certain informational messages.
17874 @item set verbose off
17875 Disables @value{GDBN} output of certain informational messages.
17877 @kindex show verbose
17879 Displays whether @code{set verbose} is on or off.
17882 By default, if @value{GDBN} encounters bugs in the symbol table of an
17883 object file, it is silent; but if you are debugging a compiler, you may
17884 find this information useful (@pxref{Symbol Errors, ,Errors Reading
17889 @kindex set complaints
17890 @item set complaints @var{limit}
17891 Permits @value{GDBN} to output @var{limit} complaints about each type of
17892 unusual symbols before becoming silent about the problem. Set
17893 @var{limit} to zero to suppress all complaints; set it to a large number
17894 to prevent complaints from being suppressed.
17896 @kindex show complaints
17897 @item show complaints
17898 Displays how many symbol complaints @value{GDBN} is permitted to produce.
17902 By default, @value{GDBN} is cautious, and asks what sometimes seems to be a
17903 lot of stupid questions to confirm certain commands. For example, if
17904 you try to run a program which is already running:
17908 The program being debugged has been started already.
17909 Start it from the beginning? (y or n)
17912 If you are willing to unflinchingly face the consequences of your own
17913 commands, you can disable this ``feature'':
17917 @kindex set confirm
17919 @cindex confirmation
17920 @cindex stupid questions
17921 @item set confirm off
17922 Disables confirmation requests.
17924 @item set confirm on
17925 Enables confirmation requests (the default).
17927 @kindex show confirm
17929 Displays state of confirmation requests.
17933 @cindex command tracing
17934 If you need to debug user-defined commands or sourced files you may find it
17935 useful to enable @dfn{command tracing}. In this mode each command will be
17936 printed as it is executed, prefixed with one or more @samp{+} symbols, the
17937 quantity denoting the call depth of each command.
17940 @kindex set trace-commands
17941 @cindex command scripts, debugging
17942 @item set trace-commands on
17943 Enable command tracing.
17944 @item set trace-commands off
17945 Disable command tracing.
17946 @item show trace-commands
17947 Display the current state of command tracing.
17950 @node Debugging Output
17951 @section Optional Messages about Internal Happenings
17952 @cindex optional debugging messages
17954 @value{GDBN} has commands that enable optional debugging messages from
17955 various @value{GDBN} subsystems; normally these commands are of
17956 interest to @value{GDBN} maintainers, or when reporting a bug. This
17957 section documents those commands.
17960 @kindex set exec-done-display
17961 @item set exec-done-display
17962 Turns on or off the notification of asynchronous commands'
17963 completion. When on, @value{GDBN} will print a message when an
17964 asynchronous command finishes its execution. The default is off.
17965 @kindex show exec-done-display
17966 @item show exec-done-display
17967 Displays the current setting of asynchronous command completion
17970 @cindex gdbarch debugging info
17971 @cindex architecture debugging info
17972 @item set debug arch
17973 Turns on or off display of gdbarch debugging info. The default is off
17975 @item show debug arch
17976 Displays the current state of displaying gdbarch debugging info.
17977 @item set debug aix-thread
17978 @cindex AIX threads
17979 Display debugging messages about inner workings of the AIX thread
17981 @item show debug aix-thread
17982 Show the current state of AIX thread debugging info display.
17983 @item set debug dwarf2-die
17984 @cindex DWARF2 DIEs
17985 Dump DWARF2 DIEs after they are read in.
17986 The value is the number of nesting levels to print.
17987 A value of zero turns off the display.
17988 @item show debug dwarf2-die
17989 Show the current state of DWARF2 DIE debugging.
17990 @item set debug displaced
17991 @cindex displaced stepping debugging info
17992 Turns on or off display of @value{GDBN} debugging info for the
17993 displaced stepping support. The default is off.
17994 @item show debug displaced
17995 Displays the current state of displaying @value{GDBN} debugging info
17996 related to displaced stepping.
17997 @item set debug event
17998 @cindex event debugging info
17999 Turns on or off display of @value{GDBN} event debugging info. The
18001 @item show debug event
18002 Displays the current state of displaying @value{GDBN} event debugging
18004 @item set debug expression
18005 @cindex expression debugging info
18006 Turns on or off display of debugging info about @value{GDBN}
18007 expression parsing. The default is off.
18008 @item show debug expression
18009 Displays the current state of displaying debugging info about
18010 @value{GDBN} expression parsing.
18011 @item set debug frame
18012 @cindex frame debugging info
18013 Turns on or off display of @value{GDBN} frame debugging info. The
18015 @item show debug frame
18016 Displays the current state of displaying @value{GDBN} frame debugging
18018 @item set debug gnu-nat
18019 @cindex @sc{gnu}/Hurd debug messages
18020 Turns on or off debugging messages from the @sc{gnu}/Hurd debug support.
18021 @item show debug gnu-nat
18022 Show the current state of @sc{gnu}/Hurd debugging messages.
18023 @item set debug infrun
18024 @cindex inferior debugging info
18025 Turns on or off display of @value{GDBN} debugging info for running the inferior.
18026 The default is off. @file{infrun.c} contains GDB's runtime state machine used
18027 for implementing operations such as single-stepping the inferior.
18028 @item show debug infrun
18029 Displays the current state of @value{GDBN} inferior debugging.
18030 @item set debug lin-lwp
18031 @cindex @sc{gnu}/Linux LWP debug messages
18032 @cindex Linux lightweight processes
18033 Turns on or off debugging messages from the Linux LWP debug support.
18034 @item show debug lin-lwp
18035 Show the current state of Linux LWP debugging messages.
18036 @item set debug lin-lwp-async
18037 @cindex @sc{gnu}/Linux LWP async debug messages
18038 @cindex Linux lightweight processes
18039 Turns on or off debugging messages from the Linux LWP async debug support.
18040 @item show debug lin-lwp-async
18041 Show the current state of Linux LWP async debugging messages.
18042 @item set debug observer
18043 @cindex observer debugging info
18044 Turns on or off display of @value{GDBN} observer debugging. This
18045 includes info such as the notification of observable events.
18046 @item show debug observer
18047 Displays the current state of observer debugging.
18048 @item set debug overload
18049 @cindex C@t{++} overload debugging info
18050 Turns on or off display of @value{GDBN} C@t{++} overload debugging
18051 info. This includes info such as ranking of functions, etc. The default
18053 @item show debug overload
18054 Displays the current state of displaying @value{GDBN} C@t{++} overload
18056 @cindex packets, reporting on stdout
18057 @cindex serial connections, debugging
18058 @cindex debug remote protocol
18059 @cindex remote protocol debugging
18060 @cindex display remote packets
18061 @item set debug remote
18062 Turns on or off display of reports on all packets sent back and forth across
18063 the serial line to the remote machine. The info is printed on the
18064 @value{GDBN} standard output stream. The default is off.
18065 @item show debug remote
18066 Displays the state of display of remote packets.
18067 @item set debug serial
18068 Turns on or off display of @value{GDBN} serial debugging info. The
18070 @item show debug serial
18071 Displays the current state of displaying @value{GDBN} serial debugging
18073 @item set debug solib-frv
18074 @cindex FR-V shared-library debugging
18075 Turns on or off debugging messages for FR-V shared-library code.
18076 @item show debug solib-frv
18077 Display the current state of FR-V shared-library code debugging
18079 @item set debug target
18080 @cindex target debugging info
18081 Turns on or off display of @value{GDBN} target debugging info. This info
18082 includes what is going on at the target level of GDB, as it happens. The
18083 default is 0. Set it to 1 to track events, and to 2 to also track the
18084 value of large memory transfers. Changes to this flag do not take effect
18085 until the next time you connect to a target or use the @code{run} command.
18086 @item show debug target
18087 Displays the current state of displaying @value{GDBN} target debugging
18089 @item set debug timestamp
18090 @cindex timestampping debugging info
18091 Turns on or off display of timestamps with @value{GDBN} debugging info.
18092 When enabled, seconds and microseconds are displayed before each debugging
18094 @item show debug timestamp
18095 Displays the current state of displaying timestamps with @value{GDBN}
18097 @item set debugvarobj
18098 @cindex variable object debugging info
18099 Turns on or off display of @value{GDBN} variable object debugging
18100 info. The default is off.
18101 @item show debugvarobj
18102 Displays the current state of displaying @value{GDBN} variable object
18104 @item set debug xml
18105 @cindex XML parser debugging
18106 Turns on or off debugging messages for built-in XML parsers.
18107 @item show debug xml
18108 Displays the current state of XML debugging messages.
18111 @node Extending GDB
18112 @chapter Extending @value{GDBN}
18113 @cindex extending GDB
18115 @value{GDBN} provides two mechanisms for extension. The first is based
18116 on composition of @value{GDBN} commands, and the second is based on the
18117 Python scripting language.
18120 * Sequences:: Canned Sequences of Commands
18121 * Python:: Scripting @value{GDBN} using Python
18125 @section Canned Sequences of Commands
18127 Aside from breakpoint commands (@pxref{Break Commands, ,Breakpoint
18128 Command Lists}), @value{GDBN} provides two ways to store sequences of
18129 commands for execution as a unit: user-defined commands and command
18133 * Define:: How to define your own commands
18134 * Hooks:: Hooks for user-defined commands
18135 * Command Files:: How to write scripts of commands to be stored in a file
18136 * Output:: Commands for controlled output
18140 @subsection User-defined Commands
18142 @cindex user-defined command
18143 @cindex arguments, to user-defined commands
18144 A @dfn{user-defined command} is a sequence of @value{GDBN} commands to
18145 which you assign a new name as a command. This is done with the
18146 @code{define} command. User commands may accept up to 10 arguments
18147 separated by whitespace. Arguments are accessed within the user command
18148 via @code{$arg0@dots{}$arg9}. A trivial example:
18152 print $arg0 + $arg1 + $arg2
18157 To execute the command use:
18164 This defines the command @code{adder}, which prints the sum of
18165 its three arguments. Note the arguments are text substitutions, so they may
18166 reference variables, use complex expressions, or even perform inferior
18169 @cindex argument count in user-defined commands
18170 @cindex how many arguments (user-defined commands)
18171 In addition, @code{$argc} may be used to find out how many arguments have
18172 been passed. This expands to a number in the range 0@dots{}10.
18177 print $arg0 + $arg1
18180 print $arg0 + $arg1 + $arg2
18188 @item define @var{commandname}
18189 Define a command named @var{commandname}. If there is already a command
18190 by that name, you are asked to confirm that you want to redefine it.
18191 @var{commandname} may be a bare command name consisting of letters,
18192 numbers, dashes, and underscores. It may also start with any predefined
18193 prefix command. For example, @samp{define target my-target} creates
18194 a user-defined @samp{target my-target} command.
18196 The definition of the command is made up of other @value{GDBN} command lines,
18197 which are given following the @code{define} command. The end of these
18198 commands is marked by a line containing @code{end}.
18201 @kindex end@r{ (user-defined commands)}
18202 @item document @var{commandname}
18203 Document the user-defined command @var{commandname}, so that it can be
18204 accessed by @code{help}. The command @var{commandname} must already be
18205 defined. This command reads lines of documentation just as @code{define}
18206 reads the lines of the command definition, ending with @code{end}.
18207 After the @code{document} command is finished, @code{help} on command
18208 @var{commandname} displays the documentation you have written.
18210 You may use the @code{document} command again to change the
18211 documentation of a command. Redefining the command with @code{define}
18212 does not change the documentation.
18214 @kindex dont-repeat
18215 @cindex don't repeat command
18217 Used inside a user-defined command, this tells @value{GDBN} that this
18218 command should not be repeated when the user hits @key{RET}
18219 (@pxref{Command Syntax, repeat last command}).
18221 @kindex help user-defined
18222 @item help user-defined
18223 List all user-defined commands, with the first line of the documentation
18228 @itemx show user @var{commandname}
18229 Display the @value{GDBN} commands used to define @var{commandname} (but
18230 not its documentation). If no @var{commandname} is given, display the
18231 definitions for all user-defined commands.
18233 @cindex infinite recursion in user-defined commands
18234 @kindex show max-user-call-depth
18235 @kindex set max-user-call-depth
18236 @item show max-user-call-depth
18237 @itemx set max-user-call-depth
18238 The value of @code{max-user-call-depth} controls how many recursion
18239 levels are allowed in user-defined commands before @value{GDBN} suspects an
18240 infinite recursion and aborts the command.
18243 In addition to the above commands, user-defined commands frequently
18244 use control flow commands, described in @ref{Command Files}.
18246 When user-defined commands are executed, the
18247 commands of the definition are not printed. An error in any command
18248 stops execution of the user-defined command.
18250 If used interactively, commands that would ask for confirmation proceed
18251 without asking when used inside a user-defined command. Many @value{GDBN}
18252 commands that normally print messages to say what they are doing omit the
18253 messages when used in a user-defined command.
18256 @subsection User-defined Command Hooks
18257 @cindex command hooks
18258 @cindex hooks, for commands
18259 @cindex hooks, pre-command
18262 You may define @dfn{hooks}, which are a special kind of user-defined
18263 command. Whenever you run the command @samp{foo}, if the user-defined
18264 command @samp{hook-foo} exists, it is executed (with no arguments)
18265 before that command.
18267 @cindex hooks, post-command
18269 A hook may also be defined which is run after the command you executed.
18270 Whenever you run the command @samp{foo}, if the user-defined command
18271 @samp{hookpost-foo} exists, it is executed (with no arguments) after
18272 that command. Post-execution hooks may exist simultaneously with
18273 pre-execution hooks, for the same command.
18275 It is valid for a hook to call the command which it hooks. If this
18276 occurs, the hook is not re-executed, thereby avoiding infinite recursion.
18278 @c It would be nice if hookpost could be passed a parameter indicating
18279 @c if the command it hooks executed properly or not. FIXME!
18281 @kindex stop@r{, a pseudo-command}
18282 In addition, a pseudo-command, @samp{stop} exists. Defining
18283 (@samp{hook-stop}) makes the associated commands execute every time
18284 execution stops in your program: before breakpoint commands are run,
18285 displays are printed, or the stack frame is printed.
18287 For example, to ignore @code{SIGALRM} signals while
18288 single-stepping, but treat them normally during normal execution,
18293 handle SIGALRM nopass
18297 handle SIGALRM pass
18300 define hook-continue
18301 handle SIGALRM pass
18305 As a further example, to hook at the beginning and end of the @code{echo}
18306 command, and to add extra text to the beginning and end of the message,
18314 define hookpost-echo
18318 (@value{GDBP}) echo Hello World
18319 <<<---Hello World--->>>
18324 You can define a hook for any single-word command in @value{GDBN}, but
18325 not for command aliases; you should define a hook for the basic command
18326 name, e.g.@: @code{backtrace} rather than @code{bt}.
18327 @c FIXME! So how does Joe User discover whether a command is an alias
18329 You can hook a multi-word command by adding @code{hook-} or
18330 @code{hookpost-} to the last word of the command, e.g.@:
18331 @samp{define target hook-remote} to add a hook to @samp{target remote}.
18333 If an error occurs during the execution of your hook, execution of
18334 @value{GDBN} commands stops and @value{GDBN} issues a prompt
18335 (before the command that you actually typed had a chance to run).
18337 If you try to define a hook which does not match any known command, you
18338 get a warning from the @code{define} command.
18340 @node Command Files
18341 @subsection Command Files
18343 @cindex command files
18344 @cindex scripting commands
18345 A command file for @value{GDBN} is a text file made of lines that are
18346 @value{GDBN} commands. Comments (lines starting with @kbd{#}) may
18347 also be included. An empty line in a command file does nothing; it
18348 does not mean to repeat the last command, as it would from the
18351 You can request the execution of a command file with the @code{source}
18356 @cindex execute commands from a file
18357 @item source [@code{-v}] @var{filename}
18358 Execute the command file @var{filename}.
18361 The lines in a command file are generally executed sequentially,
18362 unless the order of execution is changed by one of the
18363 @emph{flow-control commands} described below. The commands are not
18364 printed as they are executed. An error in any command terminates
18365 execution of the command file and control is returned to the console.
18367 @value{GDBN} searches for @var{filename} in the current directory and then
18368 on the search path (specified with the @samp{directory} command).
18370 If @code{-v}, for verbose mode, is given then @value{GDBN} displays
18371 each command as it is executed. The option must be given before
18372 @var{filename}, and is interpreted as part of the filename anywhere else.
18374 Commands that would ask for confirmation if used interactively proceed
18375 without asking when used in a command file. Many @value{GDBN} commands that
18376 normally print messages to say what they are doing omit the messages
18377 when called from command files.
18379 @value{GDBN} also accepts command input from standard input. In this
18380 mode, normal output goes to standard output and error output goes to
18381 standard error. Errors in a command file supplied on standard input do
18382 not terminate execution of the command file---execution continues with
18386 gdb < cmds > log 2>&1
18389 (The syntax above will vary depending on the shell used.) This example
18390 will execute commands from the file @file{cmds}. All output and errors
18391 would be directed to @file{log}.
18393 Since commands stored on command files tend to be more general than
18394 commands typed interactively, they frequently need to deal with
18395 complicated situations, such as different or unexpected values of
18396 variables and symbols, changes in how the program being debugged is
18397 built, etc. @value{GDBN} provides a set of flow-control commands to
18398 deal with these complexities. Using these commands, you can write
18399 complex scripts that loop over data structures, execute commands
18400 conditionally, etc.
18407 This command allows to include in your script conditionally executed
18408 commands. The @code{if} command takes a single argument, which is an
18409 expression to evaluate. It is followed by a series of commands that
18410 are executed only if the expression is true (its value is nonzero).
18411 There can then optionally be an @code{else} line, followed by a series
18412 of commands that are only executed if the expression was false. The
18413 end of the list is marked by a line containing @code{end}.
18417 This command allows to write loops. Its syntax is similar to
18418 @code{if}: the command takes a single argument, which is an expression
18419 to evaluate, and must be followed by the commands to execute, one per
18420 line, terminated by an @code{end}. These commands are called the
18421 @dfn{body} of the loop. The commands in the body of @code{while} are
18422 executed repeatedly as long as the expression evaluates to true.
18426 This command exits the @code{while} loop in whose body it is included.
18427 Execution of the script continues after that @code{while}s @code{end}
18430 @kindex loop_continue
18431 @item loop_continue
18432 This command skips the execution of the rest of the body of commands
18433 in the @code{while} loop in whose body it is included. Execution
18434 branches to the beginning of the @code{while} loop, where it evaluates
18435 the controlling expression.
18437 @kindex end@r{ (if/else/while commands)}
18439 Terminate the block of commands that are the body of @code{if},
18440 @code{else}, or @code{while} flow-control commands.
18445 @subsection Commands for Controlled Output
18447 During the execution of a command file or a user-defined command, normal
18448 @value{GDBN} output is suppressed; the only output that appears is what is
18449 explicitly printed by the commands in the definition. This section
18450 describes three commands useful for generating exactly the output you
18455 @item echo @var{text}
18456 @c I do not consider backslash-space a standard C escape sequence
18457 @c because it is not in ANSI.
18458 Print @var{text}. Nonprinting characters can be included in
18459 @var{text} using C escape sequences, such as @samp{\n} to print a
18460 newline. @strong{No newline is printed unless you specify one.}
18461 In addition to the standard C escape sequences, a backslash followed
18462 by a space stands for a space. This is useful for displaying a
18463 string with spaces at the beginning or the end, since leading and
18464 trailing spaces are otherwise trimmed from all arguments.
18465 To print @samp{@w{ }and foo =@w{ }}, use the command
18466 @samp{echo \@w{ }and foo = \@w{ }}.
18468 A backslash at the end of @var{text} can be used, as in C, to continue
18469 the command onto subsequent lines. For example,
18472 echo This is some text\n\
18473 which is continued\n\
18474 onto several lines.\n
18477 produces the same output as
18480 echo This is some text\n
18481 echo which is continued\n
18482 echo onto several lines.\n
18486 @item output @var{expression}
18487 Print the value of @var{expression} and nothing but that value: no
18488 newlines, no @samp{$@var{nn} = }. The value is not entered in the
18489 value history either. @xref{Expressions, ,Expressions}, for more information
18492 @item output/@var{fmt} @var{expression}
18493 Print the value of @var{expression} in format @var{fmt}. You can use
18494 the same formats as for @code{print}. @xref{Output Formats,,Output
18495 Formats}, for more information.
18498 @item printf @var{template}, @var{expressions}@dots{}
18499 Print the values of one or more @var{expressions} under the control of
18500 the string @var{template}. To print several values, make
18501 @var{expressions} be a comma-separated list of individual expressions,
18502 which may be either numbers or pointers. Their values are printed as
18503 specified by @var{template}, exactly as a C program would do by
18504 executing the code below:
18507 printf (@var{template}, @var{expressions}@dots{});
18510 As in @code{C} @code{printf}, ordinary characters in @var{template}
18511 are printed verbatim, while @dfn{conversion specification} introduced
18512 by the @samp{%} character cause subsequent @var{expressions} to be
18513 evaluated, their values converted and formatted according to type and
18514 style information encoded in the conversion specifications, and then
18517 For example, you can print two values in hex like this:
18520 printf "foo, bar-foo = 0x%x, 0x%x\n", foo, bar-foo
18523 @code{printf} supports all the standard @code{C} conversion
18524 specifications, including the flags and modifiers between the @samp{%}
18525 character and the conversion letter, with the following exceptions:
18529 The argument-ordering modifiers, such as @samp{2$}, are not supported.
18532 The modifier @samp{*} is not supported for specifying precision or
18536 The @samp{'} flag (for separation of digits into groups according to
18537 @code{LC_NUMERIC'}) is not supported.
18540 The type modifiers @samp{hh}, @samp{j}, @samp{t}, and @samp{z} are not
18544 The conversion letter @samp{n} (as in @samp{%n}) is not supported.
18547 The conversion letters @samp{a} and @samp{A} are not supported.
18551 Note that the @samp{ll} type modifier is supported only if the
18552 underlying @code{C} implementation used to build @value{GDBN} supports
18553 the @code{long long int} type, and the @samp{L} type modifier is
18554 supported only if @code{long double} type is available.
18556 As in @code{C}, @code{printf} supports simple backslash-escape
18557 sequences, such as @code{\n}, @samp{\t}, @samp{\\}, @samp{\"},
18558 @samp{\a}, and @samp{\f}, that consist of backslash followed by a
18559 single character. Octal and hexadecimal escape sequences are not
18562 Additionally, @code{printf} supports conversion specifications for DFP
18563 (@dfn{Decimal Floating Point}) types using the following length modifiers
18564 together with a floating point specifier.
18569 @samp{H} for printing @code{Decimal32} types.
18572 @samp{D} for printing @code{Decimal64} types.
18575 @samp{DD} for printing @code{Decimal128} types.
18578 If the underlying @code{C} implementation used to build @value{GDBN} has
18579 support for the three length modifiers for DFP types, other modifiers
18580 such as width and precision will also be available for @value{GDBN} to use.
18582 In case there is no such @code{C} support, no additional modifiers will be
18583 available and the value will be printed in the standard way.
18585 Here's an example of printing DFP types using the above conversion letters:
18587 printf "D32: %Hf - D64: %Df - D128: %DDf\n",1.2345df,1.2E10dd,1.2E1dl
18593 @section Scripting @value{GDBN} using Python
18594 @cindex python scripting
18595 @cindex scripting with python
18597 You can script @value{GDBN} using the @uref{http://www.python.org/,
18598 Python programming language}. This feature is available only if
18599 @value{GDBN} was configured using @option{--with-python}.
18602 * Python Commands:: Accessing Python from @value{GDBN}.
18603 * Python API:: Accessing @value{GDBN} from Python.
18606 @node Python Commands
18607 @subsection Python Commands
18608 @cindex python commands
18609 @cindex commands to access python
18611 @value{GDBN} provides one command for accessing the Python interpreter,
18612 and one related setting:
18616 @item python @r{[}@var{code}@r{]}
18617 The @code{python} command can be used to evaluate Python code.
18619 If given an argument, the @code{python} command will evaluate the
18620 argument as a Python command. For example:
18623 (@value{GDBP}) python print 23
18627 If you do not provide an argument to @code{python}, it will act as a
18628 multi-line command, like @code{define}. In this case, the Python
18629 script is made up of subsequent command lines, given after the
18630 @code{python} command. This command list is terminated using a line
18631 containing @code{end}. For example:
18634 (@value{GDBP}) python
18636 End with a line saying just "end".
18642 @kindex maint set python print-stack
18643 @item maint set python print-stack
18644 By default, @value{GDBN} will print a stack trace when an error occurs
18645 in a Python script. This can be controlled using @code{maint set
18646 python print-stack}: if @code{on}, the default, then Python stack
18647 printing is enabled; if @code{off}, then Python stack printing is
18652 @subsection Python API
18654 @cindex programming in python
18656 @cindex python stdout
18657 @cindex python pagination
18658 At startup, @value{GDBN} overrides Python's @code{sys.stdout} and
18659 @code{sys.stderr} to print using @value{GDBN}'s output-paging streams.
18660 A Python program which outputs to one of these streams may have its
18661 output interrupted by the user (@pxref{Screen Size}). In this
18662 situation, a Python @code{KeyboardInterrupt} exception is thrown.
18665 * Basic Python:: Basic Python Functions.
18666 * Exception Handling::
18667 * Auto-loading:: Automatically loading Python code.
18668 * Values From Inferior::
18669 * Types In Python:: Python representation of types.
18670 * Pretty Printing:: Pretty-printing values.
18671 * Selecting Pretty-Printers:: How GDB chooses a pretty-printer.
18672 * Commands In Python:: Implementing new commands in Python.
18673 * Functions In Python:: Writing new convenience functions.
18674 * Objfiles In Python:: Object files.
18675 * Frames In Python:: Acessing inferior stack frames from Python.
18679 @subsubsection Basic Python
18681 @cindex python functions
18682 @cindex python module
18684 @value{GDBN} introduces a new Python module, named @code{gdb}. All
18685 methods and classes added by @value{GDBN} are placed in this module.
18686 @value{GDBN} automatically @code{import}s the @code{gdb} module for
18687 use in all scripts evaluated by the @code{python} command.
18689 @findex gdb.execute
18690 @defun execute command [from_tty]
18691 Evaluate @var{command}, a string, as a @value{GDBN} CLI command.
18692 If a GDB exception happens while @var{command} runs, it is
18693 translated as described in @ref{Exception Handling,,Exception Handling}.
18694 If no exceptions occur, this function returns @code{None}.
18696 @var{from_tty} specifies whether @value{GDBN} ought to consider this
18697 command as having originated from the user invoking it interactively.
18698 It must be a boolean value. If omitted, it defaults to @code{False}.
18701 @findex gdb.parameter
18702 @defun parameter parameter
18703 Return the value of a @value{GDBN} parameter. @var{parameter} is a
18704 string naming the parameter to look up; @var{parameter} may contain
18705 spaces if the parameter has a multi-part name. For example,
18706 @samp{print object} is a valid parameter name.
18708 If the named parameter does not exist, this function throws a
18709 @code{RuntimeError}. Otherwise, the parameter's value is converted to
18710 a Python value of the appropriate type, and returned.
18713 @findex gdb.history
18714 @defun history number
18715 Return a value from @value{GDBN}'s value history (@pxref{Value
18716 History}). @var{number} indicates which history element to return.
18717 If @var{number} is negative, then @value{GDBN} will take its absolute value
18718 and count backward from the last element (i.e., the most recent element) to
18719 find the value to return. If @var{number} is zero, then @value{GDBN} will
18720 return the most recent element. If the element specified by @var{number}
18721 doesn't exist in the value history, a @code{RuntimeError} exception will be
18724 If no exception is raised, the return value is always an instance of
18725 @code{gdb.Value} (@pxref{Values From Inferior}).
18729 @defun write string
18730 Print a string to @value{GDBN}'s paginated standard output stream.
18731 Writing to @code{sys.stdout} or @code{sys.stderr} will automatically
18732 call this function.
18737 Flush @value{GDBN}'s paginated standard output stream. Flushing
18738 @code{sys.stdout} or @code{sys.stderr} will automatically call this
18742 @node Exception Handling
18743 @subsubsection Exception Handling
18744 @cindex python exceptions
18745 @cindex exceptions, python
18747 When executing the @code{python} command, Python exceptions
18748 uncaught within the Python code are translated to calls to
18749 @value{GDBN} error-reporting mechanism. If the command that called
18750 @code{python} does not handle the error, @value{GDBN} will
18751 terminate it and print an error message containing the Python
18752 exception name, the associated value, and the Python call stack
18753 backtrace at the point where the exception was raised. Example:
18756 (@value{GDBP}) python print foo
18757 Traceback (most recent call last):
18758 File "<string>", line 1, in <module>
18759 NameError: name 'foo' is not defined
18762 @value{GDBN} errors that happen in @value{GDBN} commands invoked by Python
18763 code are converted to Python @code{RuntimeError} exceptions. User
18764 interrupt (via @kbd{C-c} or by typing @kbd{q} at a pagination
18765 prompt) is translated to a Python @code{KeyboardInterrupt}
18766 exception. If you catch these exceptions in your Python code, your
18767 exception handler will see @code{RuntimeError} or
18768 @code{KeyboardInterrupt} as the exception type, the @value{GDBN} error
18769 message as its value, and the Python call stack backtrace at the
18770 Python statement closest to where the @value{GDBN} error occured as the
18774 @subsubsection Auto-loading
18775 @cindex auto-loading, Python
18777 When a new object file is read (for example, due to the @code{file}
18778 command, or because the inferior has loaded a shared library),
18779 @value{GDBN} will look for a file named @file{@var{objfile}-gdb.py},
18780 where @var{objfile} is the object file's real name, formed by ensuring
18781 that the file name is absolute, following all symlinks, and resolving
18782 @code{.} and @code{..} components. If this file exists and is
18783 readable, @value{GDBN} will evaluate it as a Python script.
18785 If this file does not exist, and if the parameter
18786 @code{debug-file-directory} is set (@pxref{Separate Debug Files}),
18787 then @value{GDBN} will use the file named
18788 @file{@var{debug-file-directory}/@var{real-name}}, where
18789 @var{real-name} is the object file's real name, as described above.
18791 Finally, if this file does not exist, then @value{GDBN} will look for
18792 a file named @file{@var{data-directory}/python/auto-load/@var{real-name}}, where
18793 @var{data-directory} is @value{GDBN}'s data directory (available via
18794 @code{show data-directory}, @pxref{Data Files}), and @var{real-name}
18795 is the object file's real name, as described above.
18797 When reading an auto-loaded file, @value{GDBN} sets the ``current
18798 objfile''. This is available via the @code{gdb.current_objfile}
18799 function (@pxref{Objfiles In Python}). This can be useful for
18800 registering objfile-specific pretty-printers.
18802 The auto-loading feature is useful for supplying application-specific
18803 debugging commands and scripts. You can enable or disable this
18804 feature, and view its current state.
18807 @kindex maint set python auto-load
18808 @item maint set python auto-load [yes|no]
18809 Enable or disable the Python auto-loading feature.
18811 @kindex show python auto-load
18812 @item show python auto-load
18813 Show whether Python auto-loading is enabled or disabled.
18816 @value{GDBN} does not track which files it has already auto-loaded.
18817 So, your @samp{-gdb.py} file should take care to ensure that it may be
18818 evaluated multiple times without error.
18820 @node Values From Inferior
18821 @subsubsection Values From Inferior
18822 @cindex values from inferior, with Python
18823 @cindex python, working with values from inferior
18825 @cindex @code{gdb.Value}
18826 @value{GDBN} provides values it obtains from the inferior program in
18827 an object of type @code{gdb.Value}. @value{GDBN} uses this object
18828 for its internal bookkeeping of the inferior's values, and for
18829 fetching values when necessary.
18831 Inferior values that are simple scalars can be used directly in
18832 Python expressions that are valid for the value's data type. Here's
18833 an example for an integer or floating-point value @code{some_val}:
18840 As result of this, @code{bar} will also be a @code{gdb.Value} object
18841 whose values are of the same type as those of @code{some_val}.
18843 Inferior values that are structures or instances of some class can
18844 be accessed using the Python @dfn{dictionary syntax}. For example, if
18845 @code{some_val} is a @code{gdb.Value} instance holding a structure, you
18846 can access its @code{foo} element with:
18849 bar = some_val['foo']
18852 Again, @code{bar} will also be a @code{gdb.Value} object.
18854 The following attributes are provided:
18857 @defivar Value address
18858 If this object is addressable, this read-only attribute holds a
18859 @code{gdb.Value} object representing the address. Otherwise,
18860 this attribute holds @code{None}.
18863 @cindex optimized out value in Python
18864 @defivar Value is_optimized_out
18865 This read-only boolean attribute is true if the compiler optimized out
18866 this value, thus it is not available for fetching from the inferior.
18869 @defivar Value type
18870 The type of this @code{gdb.Value}. The value of this attribute is a
18871 @code{gdb.Type} object.
18875 The following methods are provided:
18878 @defmethod Value dereference
18879 For pointer data types, this method returns a new @code{gdb.Value} object
18880 whose contents is the object pointed to by the pointer. For example, if
18881 @code{foo} is a C pointer to an @code{int}, declared in your C program as
18888 then you can use the corresponding @code{gdb.Value} to access what
18889 @code{foo} points to like this:
18892 bar = foo.dereference ()
18895 The result @code{bar} will be a @code{gdb.Value} object holding the
18896 value pointed to by @code{foo}.
18899 @defmethod Value string @r{[}encoding@r{]} @r{[}errors@r{]} @r{[}length@r{]}
18900 If this @code{gdb.Value} represents a string, then this method
18901 converts the contents to a Python string. Otherwise, this method will
18902 throw an exception.
18904 Strings are recognized in a language-specific way; whether a given
18905 @code{gdb.Value} represents a string is determined by the current
18908 For C-like languages, a value is a string if it is a pointer to or an
18909 array of characters or ints. The string is assumed to be terminated
18910 by a zero of the appropriate width. However if the optional length
18911 argument is given, the string will be converted to that given length,
18912 ignoring any embedded zeros that the string may contain.
18914 If the optional @var{encoding} argument is given, it must be a string
18915 naming the encoding of the string in the @code{gdb.Value}, such as
18916 @code{"ascii"}, @code{"iso-8859-6"} or @code{"utf-8"}. It accepts
18917 the same encodings as the corresponding argument to Python's
18918 @code{string.decode} method, and the Python codec machinery will be used
18919 to convert the string. If @var{encoding} is not given, or if
18920 @var{encoding} is the empty string, then either the @code{target-charset}
18921 (@pxref{Character Sets}) will be used, or a language-specific encoding
18922 will be used, if the current language is able to supply one.
18924 The optional @var{errors} argument is the same as the corresponding
18925 argument to Python's @code{string.decode} method.
18927 If the optional @var{length} argument is given, the string will be
18928 fetched and converted to the given length.
18932 @node Types In Python
18933 @subsubsection Types In Python
18934 @cindex types in Python
18935 @cindex Python, working with types
18938 @value{GDBN} represents types from the inferior using the class
18941 The following type-related functions are available in the @code{gdb}
18944 @findex gdb.lookup_type
18945 @defun lookup_type name [block]
18946 This function looks up a type by name. @var{name} is the name of the
18947 type to look up. It must be a string.
18949 Ordinarily, this function will return an instance of @code{gdb.Type}.
18950 If the named type cannot be found, it will throw an exception.
18953 An instance of @code{Type} has the following attributes:
18957 The type code for this type. The type code will be one of the
18958 @code{TYPE_CODE_} constants defined below.
18961 @defivar Type sizeof
18962 The size of this type, in target @code{char} units. Usually, a
18963 target's @code{char} type will be an 8-bit byte. However, on some
18964 unusual platforms, this type may have a different size.
18968 The tag name for this type. The tag name is the name after
18969 @code{struct}, @code{union}, or @code{enum} in C and C@t{++}; not all
18970 languages have this concept. If this type has no tag name, then
18971 @code{None} is returned.
18975 The following methods are provided:
18978 @defmethod Type fields
18979 For structure and union types, this method returns the fields. Range
18980 types have two fields, the minimum and maximum values. Enum types
18981 have one field per enum constant. Function and method types have one
18982 field per parameter. The base types of C@t{++} classes are also
18983 represented as fields. If the type has no fields, or does not fit
18984 into one of these categories, an empty sequence will be returned.
18986 Each field is an object, with some pre-defined attributes:
18989 This attribute is not available for @code{static} fields (as in
18990 C@t{++} or Java). For non-@code{static} fields, the value is the bit
18991 position of the field.
18994 The name of the field, or @code{None} for anonymous fields.
18997 This is @code{True} if the field is artificial, usually meaning that
18998 it was provided by the compiler and not the user. This attribute is
18999 always provided, and is @code{False} if the field is not artificial.
19002 If the field is packed, or is a bitfield, then this will have a
19003 non-zero value, which is the size of the field in bits. Otherwise,
19004 this will be zero; in this case the field's size is given by its type.
19007 The type of the field. This is usually an instance of @code{Type},
19008 but it can be @code{None} in some situations.
19012 @defmethod Type const
19013 Return a new @code{gdb.Type} object which represents a
19014 @code{const}-qualified variant of this type.
19017 @defmethod Type volatile
19018 Return a new @code{gdb.Type} object which represents a
19019 @code{volatile}-qualified variant of this type.
19022 @defmethod Type unqualified
19023 Return a new @code{gdb.Type} object which represents an unqualified
19024 variant of this type. That is, the result is neither @code{const} nor
19028 @defmethod Type reference
19029 Return a new @code{gdb.Type} object which represents a reference to this
19033 @defmethod Type strip_typedefs
19034 Return a new @code{gdb.Type} that represents the real type,
19035 after removing all layers of typedefs.
19038 @defmethod Type target
19039 Return a new @code{gdb.Type} object which represents the target type
19042 For a pointer type, the target type is the type of the pointed-to
19043 object. For an array type (meaning C-like arrays), the target type is
19044 the type of the elements of the array. For a function or method type,
19045 the target type is the type of the return value. For a complex type,
19046 the target type is the type of the elements. For a typedef, the
19047 target type is the aliased type.
19049 If the type does not have a target, this method will throw an
19053 @defmethod Type template_argument n
19054 If this @code{gdb.Type} is an instantiation of a template, this will
19055 return a new @code{gdb.Type} which represents the type of the
19056 @var{n}th template argument.
19058 If this @code{gdb.Type} is not a template type, this will throw an
19059 exception. Ordinarily, only C@t{++} code will have template types.
19061 @var{name} is searched for globally.
19066 Each type has a code, which indicates what category this type falls
19067 into. The available type categories are represented by constants
19068 defined in the @code{gdb} module:
19071 @findex TYPE_CODE_PTR
19072 @findex gdb.TYPE_CODE_PTR
19073 @item TYPE_CODE_PTR
19074 The type is a pointer.
19076 @findex TYPE_CODE_ARRAY
19077 @findex gdb.TYPE_CODE_ARRAY
19078 @item TYPE_CODE_ARRAY
19079 The type is an array.
19081 @findex TYPE_CODE_STRUCT
19082 @findex gdb.TYPE_CODE_STRUCT
19083 @item TYPE_CODE_STRUCT
19084 The type is a structure.
19086 @findex TYPE_CODE_UNION
19087 @findex gdb.TYPE_CODE_UNION
19088 @item TYPE_CODE_UNION
19089 The type is a union.
19091 @findex TYPE_CODE_ENUM
19092 @findex gdb.TYPE_CODE_ENUM
19093 @item TYPE_CODE_ENUM
19094 The type is an enum.
19096 @findex TYPE_CODE_FLAGS
19097 @findex gdb.TYPE_CODE_FLAGS
19098 @item TYPE_CODE_FLAGS
19099 A bit flags type, used for things such as status registers.
19101 @findex TYPE_CODE_FUNC
19102 @findex gdb.TYPE_CODE_FUNC
19103 @item TYPE_CODE_FUNC
19104 The type is a function.
19106 @findex TYPE_CODE_INT
19107 @findex gdb.TYPE_CODE_INT
19108 @item TYPE_CODE_INT
19109 The type is an integer type.
19111 @findex TYPE_CODE_FLT
19112 @findex gdb.TYPE_CODE_FLT
19113 @item TYPE_CODE_FLT
19114 A floating point type.
19116 @findex TYPE_CODE_VOID
19117 @findex gdb.TYPE_CODE_VOID
19118 @item TYPE_CODE_VOID
19119 The special type @code{void}.
19121 @findex TYPE_CODE_SET
19122 @findex gdb.TYPE_CODE_SET
19123 @item TYPE_CODE_SET
19126 @findex TYPE_CODE_RANGE
19127 @findex gdb.TYPE_CODE_RANGE
19128 @item TYPE_CODE_RANGE
19129 A range type, that is, an integer type with bounds.
19131 @findex TYPE_CODE_STRING
19132 @findex gdb.TYPE_CODE_STRING
19133 @item TYPE_CODE_STRING
19134 A string type. Note that this is only used for certain languages with
19135 language-defined string types; C strings are not represented this way.
19137 @findex TYPE_CODE_BITSTRING
19138 @findex gdb.TYPE_CODE_BITSTRING
19139 @item TYPE_CODE_BITSTRING
19142 @findex TYPE_CODE_ERROR
19143 @findex gdb.TYPE_CODE_ERROR
19144 @item TYPE_CODE_ERROR
19145 An unknown or erroneous type.
19147 @findex TYPE_CODE_METHOD
19148 @findex gdb.TYPE_CODE_METHOD
19149 @item TYPE_CODE_METHOD
19150 A method type, as found in C@t{++} or Java.
19152 @findex TYPE_CODE_METHODPTR
19153 @findex gdb.TYPE_CODE_METHODPTR
19154 @item TYPE_CODE_METHODPTR
19155 A pointer-to-member-function.
19157 @findex TYPE_CODE_MEMBERPTR
19158 @findex gdb.TYPE_CODE_MEMBERPTR
19159 @item TYPE_CODE_MEMBERPTR
19160 A pointer-to-member.
19162 @findex TYPE_CODE_REF
19163 @findex gdb.TYPE_CODE_REF
19164 @item TYPE_CODE_REF
19167 @findex TYPE_CODE_CHAR
19168 @findex gdb.TYPE_CODE_CHAR
19169 @item TYPE_CODE_CHAR
19172 @findex TYPE_CODE_BOOL
19173 @findex gdb.TYPE_CODE_BOOL
19174 @item TYPE_CODE_BOOL
19177 @findex TYPE_CODE_COMPLEX
19178 @findex gdb.TYPE_CODE_COMPLEX
19179 @item TYPE_CODE_COMPLEX
19180 A complex float type.
19182 @findex TYPE_CODE_TYPEDEF
19183 @findex gdb.TYPE_CODE_TYPEDEF
19184 @item TYPE_CODE_TYPEDEF
19185 A typedef to some other type.
19187 @findex TYPE_CODE_NAMESPACE
19188 @findex gdb.TYPE_CODE_NAMESPACE
19189 @item TYPE_CODE_NAMESPACE
19190 A C@t{++} namespace.
19192 @findex TYPE_CODE_DECFLOAT
19193 @findex gdb.TYPE_CODE_DECFLOAT
19194 @item TYPE_CODE_DECFLOAT
19195 A decimal floating point type.
19197 @findex TYPE_CODE_INTERNAL_FUNCTION
19198 @findex gdb.TYPE_CODE_INTERNAL_FUNCTION
19199 @item TYPE_CODE_INTERNAL_FUNCTION
19200 A function internal to @value{GDBN}. This is the type used to represent
19201 convenience functions.
19204 @node Pretty Printing
19205 @subsubsection Pretty Printing
19207 @value{GDBN} provides a mechanism to allow pretty-printing of values
19208 using Python code. The pretty-printer API allows application-specific
19209 code to greatly simplify the display of complex objects. This
19210 mechanism works for both MI and the CLI.
19212 For example, here is how a C@t{++} @code{std::string} looks without a
19216 (@value{GDBP}) print s
19218 static npos = 4294967295,
19220 <std::allocator<char>> = @{
19221 <__gnu_cxx::new_allocator<char>> = @{<No data fields>@}, <No data fields>@},
19222 members of std::basic_string<char, std::char_traits<char>, std::allocator<char> >::_Alloc_hider:
19223 _M_p = 0x804a014 "abcd"
19228 After a pretty-printer for @code{std::string} has been installed, only
19229 the contents are printed:
19232 (@value{GDBP}) print s
19236 A pretty-printer is just an object that holds a value and implements a
19237 specific interface, defined here.
19239 @defop Operation {pretty printer} children (self)
19240 @value{GDBN} will call this method on a pretty-printer to compute the
19241 children of the pretty-printer's value.
19243 This method must return an object conforming to the Python iterator
19244 protocol. Each item returned by the iterator must be a tuple holding
19245 two elements. The first element is the ``name'' of the child; the
19246 second element is the child's value. The value can be any Python
19247 object which is convertible to a @value{GDBN} value.
19249 This method is optional. If it does not exist, @value{GDBN} will act
19250 as though the value has no children.
19253 @defop Operation {pretty printer} display_hint (self)
19254 The CLI may call this method and use its result to change the
19255 formatting of a value. The result will also be supplied to an MI
19256 consumer as a @samp{displayhint} attribute of the variable being
19259 This method is optional. If it does exist, this method must return a
19262 Some display hints are predefined by @value{GDBN}:
19266 Indicate that the object being printed is ``array-like''. The CLI
19267 uses this to respect parameters such as @code{set print elements} and
19268 @code{set print array}.
19271 Indicate that the object being printed is ``map-like'', and that the
19272 children of this value can be assumed to alternate between keys and
19276 Indicate that the object being printed is ``string-like''. If the
19277 printer's @code{to_string} method returns a Python string of some
19278 kind, then @value{GDBN} will call its internal language-specific
19279 string-printing function to format the string. For the CLI this means
19280 adding quotation marks, possibly escaping some characters, respecting
19281 @code{set print elements}, and the like.
19285 @defop Operation {pretty printer} to_string (self)
19286 @value{GDBN} will call this method to display the string
19287 representation of the value passed to the object's constructor.
19289 When printing from the CLI, if the @code{to_string} method exists,
19290 then @value{GDBN} will prepend its result to the values returned by
19291 @code{children}. Exactly how this formatting is done is dependent on
19292 the display hint, and may change as more hints are added. Also,
19293 depending on the print settings (@pxref{Print Settings}), the CLI may
19294 print just the result of @code{to_string} in a stack trace, omitting
19295 the result of @code{children}.
19297 If this method returns a string, it is printed verbatim.
19299 Otherwise, if this method returns an instance of @code{gdb.Value},
19300 then @value{GDBN} prints this value. This may result in a call to
19301 another pretty-printer.
19303 If instead the method returns a Python value which is convertible to a
19304 @code{gdb.Value}, then @value{GDBN} performs the conversion and prints
19305 the resulting value. Again, this may result in a call to another
19306 pretty-printer. Python scalars (integers, floats, and booleans) and
19307 strings are convertible to @code{gdb.Value}; other types are not.
19309 If the result is not one of these types, an exception is raised.
19312 @node Selecting Pretty-Printers
19313 @subsubsection Selecting Pretty-Printers
19315 The Python list @code{gdb.pretty_printers} contains an array of
19316 functions that have been registered via addition as a pretty-printer.
19317 Each @code{gdb.Objfile} also contains a @code{pretty_printers}
19320 A function on one of these lists is passed a single @code{gdb.Value}
19321 argument and should return a pretty-printer object conforming to the
19322 interface definition above (@pxref{Pretty Printing}). If a function
19323 cannot create a pretty-printer for the value, it should return
19326 @value{GDBN} first checks the @code{pretty_printers} attribute of each
19327 @code{gdb.Objfile} and iteratively calls each function in the list for
19328 that @code{gdb.Objfile} until it receives a pretty-printer object.
19329 After these lists have been exhausted, it tries the global
19330 @code{gdb.pretty-printers} list, again calling each function until an
19331 object is returned.
19333 The order in which the objfiles are searched is not specified. For a
19334 given list, functions are always invoked from the head of the list,
19335 and iterated over sequentially until the end of the list, or a printer
19336 object is returned.
19338 Here is an example showing how a @code{std::string} printer might be
19342 class StdStringPrinter:
19343 "Print a std::string"
19345 def __init__ (self, val):
19348 def to_string (self):
19349 return self.val['_M_dataplus']['_M_p']
19351 def display_hint (self):
19355 And here is an example showing how a lookup function for the printer
19356 example above might be written.
19359 def str_lookup_function (val):
19361 lookup_tag = val.type.tag
19362 regex = re.compile ("^std::basic_string<char,.*>$")
19363 if lookup_tag == None:
19365 if regex.match (lookup_tag):
19366 return StdStringPrinter (val)
19371 The example lookup function extracts the value's type, and attempts to
19372 match it to a type that it can pretty-print. If it is a type the
19373 printer can pretty-print, it will return a printer object. If not, it
19374 returns @code{None}.
19376 We recommend that you put your core pretty-printers into a Python
19377 package. If your pretty-printers are for use with a library, we
19378 further recommend embedding a version number into the package name.
19379 This practice will enable @value{GDBN} to load multiple versions of
19380 your pretty-printers at the same time, because they will have
19383 You should write auto-loaded code (@pxref{Auto-loading}) such that it
19384 can be evaluated multiple times without changing its meaning. An
19385 ideal auto-load file will consist solely of @code{import}s of your
19386 printer modules, followed by a call to a register pretty-printers with
19387 the current objfile.
19389 Taken as a whole, this approach will scale nicely to multiple
19390 inferiors, each potentially using a different library version.
19391 Embedding a version number in the Python package name will ensure that
19392 @value{GDBN} is able to load both sets of printers simultaneously.
19393 Then, because the search for pretty-printers is done by objfile, and
19394 because your auto-loaded code took care to register your library's
19395 printers with a specific objfile, @value{GDBN} will find the correct
19396 printers for the specific version of the library used by each
19399 To continue the @code{std::string} example (@pxref{Pretty Printing}),
19400 this code might appear in @code{gdb.libstdcxx.v6}:
19403 def register_printers (objfile):
19404 objfile.pretty_printers.add (str_lookup_function)
19408 And then the corresponding contents of the auto-load file would be:
19411 import gdb.libstdcxx.v6
19412 gdb.libstdcxx.v6.register_printers (gdb.current_objfile ())
19415 @node Commands In Python
19416 @subsubsection Commands In Python
19418 @cindex commands in python
19419 @cindex python commands
19420 You can implement new @value{GDBN} CLI commands in Python. A CLI
19421 command is implemented using an instance of the @code{gdb.Command}
19422 class, most commonly using a subclass.
19424 @defmethod Command __init__ name @var{command_class} @r{[}@var{completer_class}@r{]} @r{[}@var{prefix}@r{]}
19425 The object initializer for @code{Command} registers the new command
19426 with @value{GDBN}. This initializer is normally invoked from the
19427 subclass' own @code{__init__} method.
19429 @var{name} is the name of the command. If @var{name} consists of
19430 multiple words, then the initial words are looked for as prefix
19431 commands. In this case, if one of the prefix commands does not exist,
19432 an exception is raised.
19434 There is no support for multi-line commands.
19436 @var{command_class} should be one of the @samp{COMMAND_} constants
19437 defined below. This argument tells @value{GDBN} how to categorize the
19438 new command in the help system.
19440 @var{completer_class} is an optional argument. If given, it should be
19441 one of the @samp{COMPLETE_} constants defined below. This argument
19442 tells @value{GDBN} how to perform completion for this command. If not
19443 given, @value{GDBN} will attempt to complete using the object's
19444 @code{complete} method (see below); if no such method is found, an
19445 error will occur when completion is attempted.
19447 @var{prefix} is an optional argument. If @code{True}, then the new
19448 command is a prefix command; sub-commands of this command may be
19451 The help text for the new command is taken from the Python
19452 documentation string for the command's class, if there is one. If no
19453 documentation string is provided, the default value ``This command is
19454 not documented.'' is used.
19457 @cindex don't repeat Python command
19458 @defmethod Command dont_repeat
19459 By default, a @value{GDBN} command is repeated when the user enters a
19460 blank line at the command prompt. A command can suppress this
19461 behavior by invoking the @code{dont_repeat} method. This is similar
19462 to the user command @code{dont-repeat}, see @ref{Define, dont-repeat}.
19465 @defmethod Command invoke argument from_tty
19466 This method is called by @value{GDBN} when this command is invoked.
19468 @var{argument} is a string. It is the argument to the command, after
19469 leading and trailing whitespace has been stripped.
19471 @var{from_tty} is a boolean argument. When true, this means that the
19472 command was entered by the user at the terminal; when false it means
19473 that the command came from elsewhere.
19475 If this method throws an exception, it is turned into a @value{GDBN}
19476 @code{error} call. Otherwise, the return value is ignored.
19479 @cindex completion of Python commands
19480 @defmethod Command complete text word
19481 This method is called by @value{GDBN} when the user attempts
19482 completion on this command. All forms of completion are handled by
19483 this method, that is, the @key{TAB} and @key{M-?} key bindings
19484 (@pxref{Completion}), and the @code{complete} command (@pxref{Help,
19487 The arguments @var{text} and @var{word} are both strings. @var{text}
19488 holds the complete command line up to the cursor's location.
19489 @var{word} holds the last word of the command line; this is computed
19490 using a word-breaking heuristic.
19492 The @code{complete} method can return several values:
19495 If the return value is a sequence, the contents of the sequence are
19496 used as the completions. It is up to @code{complete} to ensure that the
19497 contents actually do complete the word. A zero-length sequence is
19498 allowed, it means that there were no completions available. Only
19499 string elements of the sequence are used; other elements in the
19500 sequence are ignored.
19503 If the return value is one of the @samp{COMPLETE_} constants defined
19504 below, then the corresponding @value{GDBN}-internal completion
19505 function is invoked, and its result is used.
19508 All other results are treated as though there were no available
19513 When a new command is registered, it must be declared as a member of
19514 some general class of commands. This is used to classify top-level
19515 commands in the on-line help system; note that prefix commands are not
19516 listed under their own category but rather that of their top-level
19517 command. The available classifications are represented by constants
19518 defined in the @code{gdb} module:
19521 @findex COMMAND_NONE
19522 @findex gdb.COMMAND_NONE
19524 The command does not belong to any particular class. A command in
19525 this category will not be displayed in any of the help categories.
19527 @findex COMMAND_RUNNING
19528 @findex gdb.COMMAND_RUNNING
19529 @item COMMAND_RUNNING
19530 The command is related to running the inferior. For example,
19531 @code{start}, @code{step}, and @code{continue} are in this category.
19532 Type @kbd{help running} at the @value{GDBN} prompt to see a list of
19533 commands in this category.
19535 @findex COMMAND_DATA
19536 @findex gdb.COMMAND_DATA
19538 The command is related to data or variables. For example,
19539 @code{call}, @code{find}, and @code{print} are in this category. Type
19540 @kbd{help data} at the @value{GDBN} prompt to see a list of commands
19543 @findex COMMAND_STACK
19544 @findex gdb.COMMAND_STACK
19545 @item COMMAND_STACK
19546 The command has to do with manipulation of the stack. For example,
19547 @code{backtrace}, @code{frame}, and @code{return} are in this
19548 category. Type @kbd{help stack} at the @value{GDBN} prompt to see a
19549 list of commands in this category.
19551 @findex COMMAND_FILES
19552 @findex gdb.COMMAND_FILES
19553 @item COMMAND_FILES
19554 This class is used for file-related commands. For example,
19555 @code{file}, @code{list} and @code{section} are in this category.
19556 Type @kbd{help files} at the @value{GDBN} prompt to see a list of
19557 commands in this category.
19559 @findex COMMAND_SUPPORT
19560 @findex gdb.COMMAND_SUPPORT
19561 @item COMMAND_SUPPORT
19562 This should be used for ``support facilities'', generally meaning
19563 things that are useful to the user when interacting with @value{GDBN},
19564 but not related to the state of the inferior. For example,
19565 @code{help}, @code{make}, and @code{shell} are in this category. Type
19566 @kbd{help support} at the @value{GDBN} prompt to see a list of
19567 commands in this category.
19569 @findex COMMAND_STATUS
19570 @findex gdb.COMMAND_STATUS
19571 @item COMMAND_STATUS
19572 The command is an @samp{info}-related command, that is, related to the
19573 state of @value{GDBN} itself. For example, @code{info}, @code{macro},
19574 and @code{show} are in this category. Type @kbd{help status} at the
19575 @value{GDBN} prompt to see a list of commands in this category.
19577 @findex COMMAND_BREAKPOINTS
19578 @findex gdb.COMMAND_BREAKPOINTS
19579 @item COMMAND_BREAKPOINTS
19580 The command has to do with breakpoints. For example, @code{break},
19581 @code{clear}, and @code{delete} are in this category. Type @kbd{help
19582 breakpoints} at the @value{GDBN} prompt to see a list of commands in
19585 @findex COMMAND_TRACEPOINTS
19586 @findex gdb.COMMAND_TRACEPOINTS
19587 @item COMMAND_TRACEPOINTS
19588 The command has to do with tracepoints. For example, @code{trace},
19589 @code{actions}, and @code{tfind} are in this category. Type
19590 @kbd{help tracepoints} at the @value{GDBN} prompt to see a list of
19591 commands in this category.
19593 @findex COMMAND_OBSCURE
19594 @findex gdb.COMMAND_OBSCURE
19595 @item COMMAND_OBSCURE
19596 The command is only used in unusual circumstances, or is not of
19597 general interest to users. For example, @code{checkpoint},
19598 @code{fork}, and @code{stop} are in this category. Type @kbd{help
19599 obscure} at the @value{GDBN} prompt to see a list of commands in this
19602 @findex COMMAND_MAINTENANCE
19603 @findex gdb.COMMAND_MAINTENANCE
19604 @item COMMAND_MAINTENANCE
19605 The command is only useful to @value{GDBN} maintainers. The
19606 @code{maintenance} and @code{flushregs} commands are in this category.
19607 Type @kbd{help internals} at the @value{GDBN} prompt to see a list of
19608 commands in this category.
19611 A new command can use a predefined completion function, either by
19612 specifying it via an argument at initialization, or by returning it
19613 from the @code{complete} method. These predefined completion
19614 constants are all defined in the @code{gdb} module:
19617 @findex COMPLETE_NONE
19618 @findex gdb.COMPLETE_NONE
19619 @item COMPLETE_NONE
19620 This constant means that no completion should be done.
19622 @findex COMPLETE_FILENAME
19623 @findex gdb.COMPLETE_FILENAME
19624 @item COMPLETE_FILENAME
19625 This constant means that filename completion should be performed.
19627 @findex COMPLETE_LOCATION
19628 @findex gdb.COMPLETE_LOCATION
19629 @item COMPLETE_LOCATION
19630 This constant means that location completion should be done.
19631 @xref{Specify Location}.
19633 @findex COMPLETE_COMMAND
19634 @findex gdb.COMPLETE_COMMAND
19635 @item COMPLETE_COMMAND
19636 This constant means that completion should examine @value{GDBN}
19639 @findex COMPLETE_SYMBOL
19640 @findex gdb.COMPLETE_SYMBOL
19641 @item COMPLETE_SYMBOL
19642 This constant means that completion should be done using symbol names
19646 The following code snippet shows how a trivial CLI command can be
19647 implemented in Python:
19650 class HelloWorld (gdb.Command):
19651 """Greet the whole world."""
19653 def __init__ (self):
19654 super (HelloWorld, self).__init__ ("hello-world", gdb.COMMAND_OBSCURE)
19656 def invoke (self, arg, from_tty):
19657 print "Hello, World!"
19662 The last line instantiates the class, and is necessary to trigger the
19663 registration of the command with @value{GDBN}. Depending on how the
19664 Python code is read into @value{GDBN}, you may need to import the
19665 @code{gdb} module explicitly.
19667 @node Functions In Python
19668 @subsubsection Writing new convenience functions
19670 @cindex writing convenience functions
19671 @cindex convenience functions in python
19672 @cindex python convenience functions
19673 @tindex gdb.Function
19675 You can implement new convenience functions (@pxref{Convenience Vars})
19676 in Python. A convenience function is an instance of a subclass of the
19677 class @code{gdb.Function}.
19679 @defmethod Function __init__ name
19680 The initializer for @code{Function} registers the new function with
19681 @value{GDBN}. The argument @var{name} is the name of the function,
19682 a string. The function will be visible to the user as a convenience
19683 variable of type @code{internal function}, whose name is the same as
19684 the given @var{name}.
19686 The documentation for the new function is taken from the documentation
19687 string for the new class.
19690 @defmethod Function invoke @var{*args}
19691 When a convenience function is evaluated, its arguments are converted
19692 to instances of @code{gdb.Value}, and then the function's
19693 @code{invoke} method is called. Note that @value{GDBN} does not
19694 predetermine the arity of convenience functions. Instead, all
19695 available arguments are passed to @code{invoke}, following the
19696 standard Python calling convention. In particular, a convenience
19697 function can have default values for parameters without ill effect.
19699 The return value of this method is used as its value in the enclosing
19700 expression. If an ordinary Python value is returned, it is converted
19701 to a @code{gdb.Value} following the usual rules.
19704 The following code snippet shows how a trivial convenience function can
19705 be implemented in Python:
19708 class Greet (gdb.Function):
19709 """Return string to greet someone.
19710 Takes a name as argument."""
19712 def __init__ (self):
19713 super (Greet, self).__init__ ("greet")
19715 def invoke (self, name):
19716 return "Hello, %s!" % name.string ()
19721 The last line instantiates the class, and is necessary to trigger the
19722 registration of the function with @value{GDBN}. Depending on how the
19723 Python code is read into @value{GDBN}, you may need to import the
19724 @code{gdb} module explicitly.
19726 @node Objfiles In Python
19727 @subsubsection Objfiles In Python
19729 @cindex objfiles in python
19730 @tindex gdb.Objfile
19732 @value{GDBN} loads symbols for an inferior from various
19733 symbol-containing files (@pxref{Files}). These include the primary
19734 executable file, any shared libraries used by the inferior, and any
19735 separate debug info files (@pxref{Separate Debug Files}).
19736 @value{GDBN} calls these symbol-containing files @dfn{objfiles}.
19738 The following objfile-related functions are available in the
19741 @findex gdb.current_objfile
19742 @defun current_objfile
19743 When auto-loading a Python script (@pxref{Auto-loading}), @value{GDBN}
19744 sets the ``current objfile'' to the corresponding objfile. This
19745 function returns the current objfile. If there is no current objfile,
19746 this function returns @code{None}.
19749 @findex gdb.objfiles
19751 Return a sequence of all the objfiles current known to @value{GDBN}.
19752 @xref{Objfiles In Python}.
19755 Each objfile is represented by an instance of the @code{gdb.Objfile}
19758 @defivar Objfile filename
19759 The file name of the objfile as a string.
19762 @defivar Objfile pretty_printers
19763 The @code{pretty_printers} attribute is a list of functions. It is
19764 used to look up pretty-printers. A @code{Value} is passed to each
19765 function in order; if the function returns @code{None}, then the
19766 search continues. Otherwise, the return value should be an object
19767 which is used to format the value. @xref{Pretty Printing}, for more
19771 @node Frames In Python
19772 @subsubsection Acessing inferior stack frames from Python.
19774 @cindex frames in python
19775 When the debugged program stops, @value{GDBN} is able to analyze its call
19776 stack (@pxref{Frames,,Stack frames}). The @code{gdb.Frame} class
19777 represents a frame in the stack. A @code{gdb.Frame} object is only valid
19778 while its corresponding frame exists in the inferior's stack. If you try
19779 to use an invalid frame object, @value{GDBN} will throw a @code{RuntimeError}
19782 Two @code{gdb.Frame} objects can be compared for equality with the @code{==}
19786 (@value{GDBP}) python print gdb.newest_frame() == gdb.selected_frame ()
19790 The following frame-related functions are available in the @code{gdb} module:
19792 @findex gdb.selected_frame
19793 @defun selected_frame
19794 Return the selected frame object. (@pxref{Selection,,Selecting a Frame}).
19797 @defun frame_stop_reason_string reason
19798 Return a string explaining the reason why @value{GDBN} stopped unwinding
19799 frames, as expressed by the given @var{reason} code (an integer, see the
19800 @code{unwind_stop_reason} method further down in this section).
19803 A @code{gdb.Frame} object has the following methods:
19806 @defmethod Frame is_valid
19807 Returns true if the @code{gdb.Frame} object is valid, false if not.
19808 A frame object can become invalid if the frame it refers to doesn't
19809 exist anymore in the inferior. All @code{gdb.Frame} methods will throw
19810 an exception if it is invalid at the time the method is called.
19813 @defmethod Frame name
19814 Returns the function name of the frame, or @code{None} if it can't be
19818 @defmethod Frame type
19819 Returns the type of the frame. The value can be one of
19820 @code{gdb.NORMAL_FRAME}, @code{gdb.DUMMY_FRAME}, @code{gdb.SIGTRAMP_FRAME}
19821 or @code{gdb.SENTINEL_FRAME}.
19824 @defmethod Frame unwind_stop_reason
19825 Return an integer representing the reason why it's not possible to find
19826 more frames toward the outermost frame. Use
19827 @code{gdb.frame_stop_reason_string} to convert the value returned by this
19828 function to a string.
19831 @defmethod Frame pc
19832 Returns the frame's resume address.
19835 @defmethod Frame older
19836 Return the frame that called this frame.
19839 @defmethod Frame newer
19840 Return the frame called by this frame.
19843 @defmethod Frame read_var variable
19844 Return the value of the given variable in this frame. @var{variable} must
19850 @chapter Command Interpreters
19851 @cindex command interpreters
19853 @value{GDBN} supports multiple command interpreters, and some command
19854 infrastructure to allow users or user interface writers to switch
19855 between interpreters or run commands in other interpreters.
19857 @value{GDBN} currently supports two command interpreters, the console
19858 interpreter (sometimes called the command-line interpreter or @sc{cli})
19859 and the machine interface interpreter (or @sc{gdb/mi}). This manual
19860 describes both of these interfaces in great detail.
19862 By default, @value{GDBN} will start with the console interpreter.
19863 However, the user may choose to start @value{GDBN} with another
19864 interpreter by specifying the @option{-i} or @option{--interpreter}
19865 startup options. Defined interpreters include:
19869 @cindex console interpreter
19870 The traditional console or command-line interpreter. This is the most often
19871 used interpreter with @value{GDBN}. With no interpreter specified at runtime,
19872 @value{GDBN} will use this interpreter.
19875 @cindex mi interpreter
19876 The newest @sc{gdb/mi} interface (currently @code{mi2}). Used primarily
19877 by programs wishing to use @value{GDBN} as a backend for a debugger GUI
19878 or an IDE. For more information, see @ref{GDB/MI, ,The @sc{gdb/mi}
19882 @cindex mi2 interpreter
19883 The current @sc{gdb/mi} interface.
19886 @cindex mi1 interpreter
19887 The @sc{gdb/mi} interface included in @value{GDBN} 5.1, 5.2, and 5.3.
19891 @cindex invoke another interpreter
19892 The interpreter being used by @value{GDBN} may not be dynamically
19893 switched at runtime. Although possible, this could lead to a very
19894 precarious situation. Consider an IDE using @sc{gdb/mi}. If a user
19895 enters the command "interpreter-set console" in a console view,
19896 @value{GDBN} would switch to using the console interpreter, rendering
19897 the IDE inoperable!
19899 @kindex interpreter-exec
19900 Although you may only choose a single interpreter at startup, you may execute
19901 commands in any interpreter from the current interpreter using the appropriate
19902 command. If you are running the console interpreter, simply use the
19903 @code{interpreter-exec} command:
19906 interpreter-exec mi "-data-list-register-names"
19909 @sc{gdb/mi} has a similar command, although it is only available in versions of
19910 @value{GDBN} which support @sc{gdb/mi} version 2 (or greater).
19913 @chapter @value{GDBN} Text User Interface
19915 @cindex Text User Interface
19918 * TUI Overview:: TUI overview
19919 * TUI Keys:: TUI key bindings
19920 * TUI Single Key Mode:: TUI single key mode
19921 * TUI Commands:: TUI-specific commands
19922 * TUI Configuration:: TUI configuration variables
19925 The @value{GDBN} Text User Interface (TUI) is a terminal
19926 interface which uses the @code{curses} library to show the source
19927 file, the assembly output, the program registers and @value{GDBN}
19928 commands in separate text windows. The TUI mode is supported only
19929 on platforms where a suitable version of the @code{curses} library
19932 @pindex @value{GDBTUI}
19933 The TUI mode is enabled by default when you invoke @value{GDBN} as
19934 either @samp{@value{GDBTUI}} or @samp{@value{GDBP} -tui}.
19935 You can also switch in and out of TUI mode while @value{GDBN} runs by
19936 using various TUI commands and key bindings, such as @kbd{C-x C-a}.
19937 @xref{TUI Keys, ,TUI Key Bindings}.
19940 @section TUI Overview
19942 In TUI mode, @value{GDBN} can display several text windows:
19946 This window is the @value{GDBN} command window with the @value{GDBN}
19947 prompt and the @value{GDBN} output. The @value{GDBN} input is still
19948 managed using readline.
19951 The source window shows the source file of the program. The current
19952 line and active breakpoints are displayed in this window.
19955 The assembly window shows the disassembly output of the program.
19958 This window shows the processor registers. Registers are highlighted
19959 when their values change.
19962 The source and assembly windows show the current program position
19963 by highlighting the current line and marking it with a @samp{>} marker.
19964 Breakpoints are indicated with two markers. The first marker
19965 indicates the breakpoint type:
19969 Breakpoint which was hit at least once.
19972 Breakpoint which was never hit.
19975 Hardware breakpoint which was hit at least once.
19978 Hardware breakpoint which was never hit.
19981 The second marker indicates whether the breakpoint is enabled or not:
19985 Breakpoint is enabled.
19988 Breakpoint is disabled.
19991 The source, assembly and register windows are updated when the current
19992 thread changes, when the frame changes, or when the program counter
19995 These windows are not all visible at the same time. The command
19996 window is always visible. The others can be arranged in several
20007 source and assembly,
20010 source and registers, or
20013 assembly and registers.
20016 A status line above the command window shows the following information:
20020 Indicates the current @value{GDBN} target.
20021 (@pxref{Targets, ,Specifying a Debugging Target}).
20024 Gives the current process or thread number.
20025 When no process is being debugged, this field is set to @code{No process}.
20028 Gives the current function name for the selected frame.
20029 The name is demangled if demangling is turned on (@pxref{Print Settings}).
20030 When there is no symbol corresponding to the current program counter,
20031 the string @code{??} is displayed.
20034 Indicates the current line number for the selected frame.
20035 When the current line number is not known, the string @code{??} is displayed.
20038 Indicates the current program counter address.
20042 @section TUI Key Bindings
20043 @cindex TUI key bindings
20045 The TUI installs several key bindings in the readline keymaps
20046 (@pxref{Command Line Editing}). The following key bindings
20047 are installed for both TUI mode and the @value{GDBN} standard mode.
20056 Enter or leave the TUI mode. When leaving the TUI mode,
20057 the curses window management stops and @value{GDBN} operates using
20058 its standard mode, writing on the terminal directly. When reentering
20059 the TUI mode, control is given back to the curses windows.
20060 The screen is then refreshed.
20064 Use a TUI layout with only one window. The layout will
20065 either be @samp{source} or @samp{assembly}. When the TUI mode
20066 is not active, it will switch to the TUI mode.
20068 Think of this key binding as the Emacs @kbd{C-x 1} binding.
20072 Use a TUI layout with at least two windows. When the current
20073 layout already has two windows, the next layout with two windows is used.
20074 When a new layout is chosen, one window will always be common to the
20075 previous layout and the new one.
20077 Think of it as the Emacs @kbd{C-x 2} binding.
20081 Change the active window. The TUI associates several key bindings
20082 (like scrolling and arrow keys) with the active window. This command
20083 gives the focus to the next TUI window.
20085 Think of it as the Emacs @kbd{C-x o} binding.
20089 Switch in and out of the TUI SingleKey mode that binds single
20090 keys to @value{GDBN} commands (@pxref{TUI Single Key Mode}).
20093 The following key bindings only work in the TUI mode:
20098 Scroll the active window one page up.
20102 Scroll the active window one page down.
20106 Scroll the active window one line up.
20110 Scroll the active window one line down.
20114 Scroll the active window one column left.
20118 Scroll the active window one column right.
20122 Refresh the screen.
20125 Because the arrow keys scroll the active window in the TUI mode, they
20126 are not available for their normal use by readline unless the command
20127 window has the focus. When another window is active, you must use
20128 other readline key bindings such as @kbd{C-p}, @kbd{C-n}, @kbd{C-b}
20129 and @kbd{C-f} to control the command window.
20131 @node TUI Single Key Mode
20132 @section TUI Single Key Mode
20133 @cindex TUI single key mode
20135 The TUI also provides a @dfn{SingleKey} mode, which binds several
20136 frequently used @value{GDBN} commands to single keys. Type @kbd{C-x s} to
20137 switch into this mode, where the following key bindings are used:
20140 @kindex c @r{(SingleKey TUI key)}
20144 @kindex d @r{(SingleKey TUI key)}
20148 @kindex f @r{(SingleKey TUI key)}
20152 @kindex n @r{(SingleKey TUI key)}
20156 @kindex q @r{(SingleKey TUI key)}
20158 exit the SingleKey mode.
20160 @kindex r @r{(SingleKey TUI key)}
20164 @kindex s @r{(SingleKey TUI key)}
20168 @kindex u @r{(SingleKey TUI key)}
20172 @kindex v @r{(SingleKey TUI key)}
20176 @kindex w @r{(SingleKey TUI key)}
20181 Other keys temporarily switch to the @value{GDBN} command prompt.
20182 The key that was pressed is inserted in the editing buffer so that
20183 it is possible to type most @value{GDBN} commands without interaction
20184 with the TUI SingleKey mode. Once the command is entered the TUI
20185 SingleKey mode is restored. The only way to permanently leave
20186 this mode is by typing @kbd{q} or @kbd{C-x s}.
20190 @section TUI-specific Commands
20191 @cindex TUI commands
20193 The TUI has specific commands to control the text windows.
20194 These commands are always available, even when @value{GDBN} is not in
20195 the TUI mode. When @value{GDBN} is in the standard mode, most
20196 of these commands will automatically switch to the TUI mode.
20201 List and give the size of all displayed windows.
20205 Display the next layout.
20208 Display the previous layout.
20211 Display the source window only.
20214 Display the assembly window only.
20217 Display the source and assembly window.
20220 Display the register window together with the source or assembly window.
20224 Make the next window active for scrolling.
20227 Make the previous window active for scrolling.
20230 Make the source window active for scrolling.
20233 Make the assembly window active for scrolling.
20236 Make the register window active for scrolling.
20239 Make the command window active for scrolling.
20243 Refresh the screen. This is similar to typing @kbd{C-L}.
20245 @item tui reg float
20247 Show the floating point registers in the register window.
20249 @item tui reg general
20250 Show the general registers in the register window.
20253 Show the next register group. The list of register groups as well as
20254 their order is target specific. The predefined register groups are the
20255 following: @code{general}, @code{float}, @code{system}, @code{vector},
20256 @code{all}, @code{save}, @code{restore}.
20258 @item tui reg system
20259 Show the system registers in the register window.
20263 Update the source window and the current execution point.
20265 @item winheight @var{name} +@var{count}
20266 @itemx winheight @var{name} -@var{count}
20268 Change the height of the window @var{name} by @var{count}
20269 lines. Positive counts increase the height, while negative counts
20272 @item tabset @var{nchars}
20274 Set the width of tab stops to be @var{nchars} characters.
20277 @node TUI Configuration
20278 @section TUI Configuration Variables
20279 @cindex TUI configuration variables
20281 Several configuration variables control the appearance of TUI windows.
20284 @item set tui border-kind @var{kind}
20285 @kindex set tui border-kind
20286 Select the border appearance for the source, assembly and register windows.
20287 The possible values are the following:
20290 Use a space character to draw the border.
20293 Use @sc{ascii} characters @samp{+}, @samp{-} and @samp{|} to draw the border.
20296 Use the Alternate Character Set to draw the border. The border is
20297 drawn using character line graphics if the terminal supports them.
20300 @item set tui border-mode @var{mode}
20301 @kindex set tui border-mode
20302 @itemx set tui active-border-mode @var{mode}
20303 @kindex set tui active-border-mode
20304 Select the display attributes for the borders of the inactive windows
20305 or the active window. The @var{mode} can be one of the following:
20308 Use normal attributes to display the border.
20314 Use reverse video mode.
20317 Use half bright mode.
20319 @item half-standout
20320 Use half bright and standout mode.
20323 Use extra bright or bold mode.
20325 @item bold-standout
20326 Use extra bright or bold and standout mode.
20331 @chapter Using @value{GDBN} under @sc{gnu} Emacs
20334 @cindex @sc{gnu} Emacs
20335 A special interface allows you to use @sc{gnu} Emacs to view (and
20336 edit) the source files for the program you are debugging with
20339 To use this interface, use the command @kbd{M-x gdb} in Emacs. Give the
20340 executable file you want to debug as an argument. This command starts
20341 @value{GDBN} as a subprocess of Emacs, with input and output through a newly
20342 created Emacs buffer.
20343 @c (Do not use the @code{-tui} option to run @value{GDBN} from Emacs.)
20345 Running @value{GDBN} under Emacs can be just like running @value{GDBN} normally except for two
20350 All ``terminal'' input and output goes through an Emacs buffer, called
20353 This applies both to @value{GDBN} commands and their output, and to the input
20354 and output done by the program you are debugging.
20356 This is useful because it means that you can copy the text of previous
20357 commands and input them again; you can even use parts of the output
20360 All the facilities of Emacs' Shell mode are available for interacting
20361 with your program. In particular, you can send signals the usual
20362 way---for example, @kbd{C-c C-c} for an interrupt, @kbd{C-c C-z} for a
20366 @value{GDBN} displays source code through Emacs.
20368 Each time @value{GDBN} displays a stack frame, Emacs automatically finds the
20369 source file for that frame and puts an arrow (@samp{=>}) at the
20370 left margin of the current line. Emacs uses a separate buffer for
20371 source display, and splits the screen to show both your @value{GDBN} session
20374 Explicit @value{GDBN} @code{list} or search commands still produce output as
20375 usual, but you probably have no reason to use them from Emacs.
20378 We call this @dfn{text command mode}. Emacs 22.1, and later, also uses
20379 a graphical mode, enabled by default, which provides further buffers
20380 that can control the execution and describe the state of your program.
20381 @xref{GDB Graphical Interface,,, Emacs, The @sc{gnu} Emacs Manual}.
20383 If you specify an absolute file name when prompted for the @kbd{M-x
20384 gdb} argument, then Emacs sets your current working directory to where
20385 your program resides. If you only specify the file name, then Emacs
20386 sets your current working directory to to the directory associated
20387 with the previous buffer. In this case, @value{GDBN} may find your
20388 program by searching your environment's @code{PATH} variable, but on
20389 some operating systems it might not find the source. So, although the
20390 @value{GDBN} input and output session proceeds normally, the auxiliary
20391 buffer does not display the current source and line of execution.
20393 The initial working directory of @value{GDBN} is printed on the top
20394 line of the GUD buffer and this serves as a default for the commands
20395 that specify files for @value{GDBN} to operate on. @xref{Files,
20396 ,Commands to Specify Files}.
20398 By default, @kbd{M-x gdb} calls the program called @file{gdb}. If you
20399 need to call @value{GDBN} by a different name (for example, if you
20400 keep several configurations around, with different names) you can
20401 customize the Emacs variable @code{gud-gdb-command-name} to run the
20404 In the GUD buffer, you can use these special Emacs commands in
20405 addition to the standard Shell mode commands:
20409 Describe the features of Emacs' GUD Mode.
20412 Execute to another source line, like the @value{GDBN} @code{step} command; also
20413 update the display window to show the current file and location.
20416 Execute to next source line in this function, skipping all function
20417 calls, like the @value{GDBN} @code{next} command. Then update the display window
20418 to show the current file and location.
20421 Execute one instruction, like the @value{GDBN} @code{stepi} command; update
20422 display window accordingly.
20425 Execute until exit from the selected stack frame, like the @value{GDBN}
20426 @code{finish} command.
20429 Continue execution of your program, like the @value{GDBN} @code{continue}
20433 Go up the number of frames indicated by the numeric argument
20434 (@pxref{Arguments, , Numeric Arguments, Emacs, The @sc{gnu} Emacs Manual}),
20435 like the @value{GDBN} @code{up} command.
20438 Go down the number of frames indicated by the numeric argument, like the
20439 @value{GDBN} @code{down} command.
20442 In any source file, the Emacs command @kbd{C-x @key{SPC}} (@code{gud-break})
20443 tells @value{GDBN} to set a breakpoint on the source line point is on.
20445 In text command mode, if you type @kbd{M-x speedbar}, Emacs displays a
20446 separate frame which shows a backtrace when the GUD buffer is current.
20447 Move point to any frame in the stack and type @key{RET} to make it
20448 become the current frame and display the associated source in the
20449 source buffer. Alternatively, click @kbd{Mouse-2} to make the
20450 selected frame become the current one. In graphical mode, the
20451 speedbar displays watch expressions.
20453 If you accidentally delete the source-display buffer, an easy way to get
20454 it back is to type the command @code{f} in the @value{GDBN} buffer, to
20455 request a frame display; when you run under Emacs, this recreates
20456 the source buffer if necessary to show you the context of the current
20459 The source files displayed in Emacs are in ordinary Emacs buffers
20460 which are visiting the source files in the usual way. You can edit
20461 the files with these buffers if you wish; but keep in mind that @value{GDBN}
20462 communicates with Emacs in terms of line numbers. If you add or
20463 delete lines from the text, the line numbers that @value{GDBN} knows cease
20464 to correspond properly with the code.
20466 A more detailed description of Emacs' interaction with @value{GDBN} is
20467 given in the Emacs manual (@pxref{Debuggers,,, Emacs, The @sc{gnu}
20470 @c The following dropped because Epoch is nonstandard. Reactivate
20471 @c if/when v19 does something similar. ---doc@cygnus.com 19dec1990
20473 @kindex Emacs Epoch environment
20477 Version 18 of @sc{gnu} Emacs has a built-in window system
20478 called the @code{epoch}
20479 environment. Users of this environment can use a new command,
20480 @code{inspect} which performs identically to @code{print} except that
20481 each value is printed in its own window.
20486 @chapter The @sc{gdb/mi} Interface
20488 @unnumberedsec Function and Purpose
20490 @cindex @sc{gdb/mi}, its purpose
20491 @sc{gdb/mi} is a line based machine oriented text interface to
20492 @value{GDBN} and is activated by specifying using the
20493 @option{--interpreter} command line option (@pxref{Mode Options}). It
20494 is specifically intended to support the development of systems which
20495 use the debugger as just one small component of a larger system.
20497 This chapter is a specification of the @sc{gdb/mi} interface. It is written
20498 in the form of a reference manual.
20500 Note that @sc{gdb/mi} is still under construction, so some of the
20501 features described below are incomplete and subject to change
20502 (@pxref{GDB/MI Development and Front Ends, , @sc{gdb/mi} Development and Front Ends}).
20504 @unnumberedsec Notation and Terminology
20506 @cindex notational conventions, for @sc{gdb/mi}
20507 This chapter uses the following notation:
20511 @code{|} separates two alternatives.
20514 @code{[ @var{something} ]} indicates that @var{something} is optional:
20515 it may or may not be given.
20518 @code{( @var{group} )*} means that @var{group} inside the parentheses
20519 may repeat zero or more times.
20522 @code{( @var{group} )+} means that @var{group} inside the parentheses
20523 may repeat one or more times.
20526 @code{"@var{string}"} means a literal @var{string}.
20530 @heading Dependencies
20534 * GDB/MI General Design::
20535 * GDB/MI Command Syntax::
20536 * GDB/MI Compatibility with CLI::
20537 * GDB/MI Development and Front Ends::
20538 * GDB/MI Output Records::
20539 * GDB/MI Simple Examples::
20540 * GDB/MI Command Description Format::
20541 * GDB/MI Breakpoint Commands::
20542 * GDB/MI Program Context::
20543 * GDB/MI Thread Commands::
20544 * GDB/MI Program Execution::
20545 * GDB/MI Stack Manipulation::
20546 * GDB/MI Variable Objects::
20547 * GDB/MI Data Manipulation::
20548 * GDB/MI Tracepoint Commands::
20549 * GDB/MI Symbol Query::
20550 * GDB/MI File Commands::
20552 * GDB/MI Kod Commands::
20553 * GDB/MI Memory Overlay Commands::
20554 * GDB/MI Signal Handling Commands::
20556 * GDB/MI Target Manipulation::
20557 * GDB/MI File Transfer Commands::
20558 * GDB/MI Miscellaneous Commands::
20561 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
20562 @node GDB/MI General Design
20563 @section @sc{gdb/mi} General Design
20564 @cindex GDB/MI General Design
20566 Interaction of a @sc{GDB/MI} frontend with @value{GDBN} involves three
20567 parts---commands sent to @value{GDBN}, responses to those commands
20568 and notifications. Each command results in exactly one response,
20569 indicating either successful completion of the command, or an error.
20570 For the commands that do not resume the target, the response contains the
20571 requested information. For the commands that resume the target, the
20572 response only indicates whether the target was successfully resumed.
20573 Notifications is the mechanism for reporting changes in the state of the
20574 target, or in @value{GDBN} state, that cannot conveniently be associated with
20575 a command and reported as part of that command response.
20577 The important examples of notifications are:
20581 Exec notifications. These are used to report changes in
20582 target state---when a target is resumed, or stopped. It would not
20583 be feasible to include this information in response of resuming
20584 commands, because one resume commands can result in multiple events in
20585 different threads. Also, quite some time may pass before any event
20586 happens in the target, while a frontend needs to know whether the resuming
20587 command itself was successfully executed.
20590 Console output, and status notifications. Console output
20591 notifications are used to report output of CLI commands, as well as
20592 diagnostics for other commands. Status notifications are used to
20593 report the progress of a long-running operation. Naturally, including
20594 this information in command response would mean no output is produced
20595 until the command is finished, which is undesirable.
20598 General notifications. Commands may have various side effects on
20599 the @value{GDBN} or target state beyond their official purpose. For example,
20600 a command may change the selected thread. Although such changes can
20601 be included in command response, using notification allows for more
20602 orthogonal frontend design.
20606 There's no guarantee that whenever an MI command reports an error,
20607 @value{GDBN} or the target are in any specific state, and especially,
20608 the state is not reverted to the state before the MI command was
20609 processed. Therefore, whenever an MI command results in an error,
20610 we recommend that the frontend refreshes all the information shown in
20611 the user interface.
20615 * Context management::
20616 * Asynchronous and non-stop modes::
20620 @node Context management
20621 @subsection Context management
20623 In most cases when @value{GDBN} accesses the target, this access is
20624 done in context of a specific thread and frame (@pxref{Frames}).
20625 Often, even when accessing global data, the target requires that a thread
20626 be specified. The CLI interface maintains the selected thread and frame,
20627 and supplies them to target on each command. This is convenient,
20628 because a command line user would not want to specify that information
20629 explicitly on each command, and because user interacts with
20630 @value{GDBN} via a single terminal, so no confusion is possible as
20631 to what thread and frame are the current ones.
20633 In the case of MI, the concept of selected thread and frame is less
20634 useful. First, a frontend can easily remember this information
20635 itself. Second, a graphical frontend can have more than one window,
20636 each one used for debugging a different thread, and the frontend might
20637 want to access additional threads for internal purposes. This
20638 increases the risk that by relying on implicitly selected thread, the
20639 frontend may be operating on a wrong one. Therefore, each MI command
20640 should explicitly specify which thread and frame to operate on. To
20641 make it possible, each MI command accepts the @samp{--thread} and
20642 @samp{--frame} options, the value to each is @value{GDBN} identifier
20643 for thread and frame to operate on.
20645 Usually, each top-level window in a frontend allows the user to select
20646 a thread and a frame, and remembers the user selection for further
20647 operations. However, in some cases @value{GDBN} may suggest that the
20648 current thread be changed. For example, when stopping on a breakpoint
20649 it is reasonable to switch to the thread where breakpoint is hit. For
20650 another example, if the user issues the CLI @samp{thread} command via
20651 the frontend, it is desirable to change the frontend's selected thread to the
20652 one specified by user. @value{GDBN} communicates the suggestion to
20653 change current thread using the @samp{=thread-selected} notification.
20654 No such notification is available for the selected frame at the moment.
20656 Note that historically, MI shares the selected thread with CLI, so
20657 frontends used the @code{-thread-select} to execute commands in the
20658 right context. However, getting this to work right is cumbersome. The
20659 simplest way is for frontend to emit @code{-thread-select} command
20660 before every command. This doubles the number of commands that need
20661 to be sent. The alternative approach is to suppress @code{-thread-select}
20662 if the selected thread in @value{GDBN} is supposed to be identical to the
20663 thread the frontend wants to operate on. However, getting this
20664 optimization right can be tricky. In particular, if the frontend
20665 sends several commands to @value{GDBN}, and one of the commands changes the
20666 selected thread, then the behaviour of subsequent commands will
20667 change. So, a frontend should either wait for response from such
20668 problematic commands, or explicitly add @code{-thread-select} for
20669 all subsequent commands. No frontend is known to do this exactly
20670 right, so it is suggested to just always pass the @samp{--thread} and
20671 @samp{--frame} options.
20673 @node Asynchronous and non-stop modes
20674 @subsection Asynchronous command execution and non-stop mode
20676 On some targets, @value{GDBN} is capable of processing MI commands
20677 even while the target is running. This is called @dfn{asynchronous
20678 command execution} (@pxref{Background Execution}). The frontend may
20679 specify a preferrence for asynchronous execution using the
20680 @code{-gdb-set target-async 1} command, which should be emitted before
20681 either running the executable or attaching to the target. After the
20682 frontend has started the executable or attached to the target, it can
20683 find if asynchronous execution is enabled using the
20684 @code{-list-target-features} command.
20686 Even if @value{GDBN} can accept a command while target is running,
20687 many commands that access the target do not work when the target is
20688 running. Therefore, asynchronous command execution is most useful
20689 when combined with non-stop mode (@pxref{Non-Stop Mode}). Then,
20690 it is possible to examine the state of one thread, while other threads
20693 When a given thread is running, MI commands that try to access the
20694 target in the context of that thread may not work, or may work only on
20695 some targets. In particular, commands that try to operate on thread's
20696 stack will not work, on any target. Commands that read memory, or
20697 modify breakpoints, may work or not work, depending on the target. Note
20698 that even commands that operate on global state, such as @code{print},
20699 @code{set}, and breakpoint commands, still access the target in the
20700 context of a specific thread, so frontend should try to find a
20701 stopped thread and perform the operation on that thread (using the
20702 @samp{--thread} option).
20704 Which commands will work in the context of a running thread is
20705 highly target dependent. However, the two commands
20706 @code{-exec-interrupt}, to stop a thread, and @code{-thread-info},
20707 to find the state of a thread, will always work.
20709 @node Thread groups
20710 @subsection Thread groups
20711 @value{GDBN} may be used to debug several processes at the same time.
20712 On some platfroms, @value{GDBN} may support debugging of several
20713 hardware systems, each one having several cores with several different
20714 processes running on each core. This section describes the MI
20715 mechanism to support such debugging scenarios.
20717 The key observation is that regardless of the structure of the
20718 target, MI can have a global list of threads, because most commands that
20719 accept the @samp{--thread} option do not need to know what process that
20720 thread belongs to. Therefore, it is not necessary to introduce
20721 neither additional @samp{--process} option, nor an notion of the
20722 current process in the MI interface. The only strictly new feature
20723 that is required is the ability to find how the threads are grouped
20726 To allow the user to discover such grouping, and to support arbitrary
20727 hierarchy of machines/cores/processes, MI introduces the concept of a
20728 @dfn{thread group}. Thread group is a collection of threads and other
20729 thread groups. A thread group always has a string identifier, a type,
20730 and may have additional attributes specific to the type. A new
20731 command, @code{-list-thread-groups}, returns the list of top-level
20732 thread groups, which correspond to processes that @value{GDBN} is
20733 debugging at the moment. By passing an identifier of a thread group
20734 to the @code{-list-thread-groups} command, it is possible to obtain
20735 the members of specific thread group.
20737 To allow the user to easily discover processes, and other objects, he
20738 wishes to debug, a concept of @dfn{available thread group} is
20739 introduced. Available thread group is an thread group that
20740 @value{GDBN} is not debugging, but that can be attached to, using the
20741 @code{-target-attach} command. The list of available top-level thread
20742 groups can be obtained using @samp{-list-thread-groups --available}.
20743 In general, the content of a thread group may be only retrieved only
20744 after attaching to that thread group.
20746 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
20747 @node GDB/MI Command Syntax
20748 @section @sc{gdb/mi} Command Syntax
20751 * GDB/MI Input Syntax::
20752 * GDB/MI Output Syntax::
20755 @node GDB/MI Input Syntax
20756 @subsection @sc{gdb/mi} Input Syntax
20758 @cindex input syntax for @sc{gdb/mi}
20759 @cindex @sc{gdb/mi}, input syntax
20761 @item @var{command} @expansion{}
20762 @code{@var{cli-command} | @var{mi-command}}
20764 @item @var{cli-command} @expansion{}
20765 @code{[ @var{token} ] @var{cli-command} @var{nl}}, where
20766 @var{cli-command} is any existing @value{GDBN} CLI command.
20768 @item @var{mi-command} @expansion{}
20769 @code{[ @var{token} ] "-" @var{operation} ( " " @var{option} )*
20770 @code{[} " --" @code{]} ( " " @var{parameter} )* @var{nl}}
20772 @item @var{token} @expansion{}
20773 "any sequence of digits"
20775 @item @var{option} @expansion{}
20776 @code{"-" @var{parameter} [ " " @var{parameter} ]}
20778 @item @var{parameter} @expansion{}
20779 @code{@var{non-blank-sequence} | @var{c-string}}
20781 @item @var{operation} @expansion{}
20782 @emph{any of the operations described in this chapter}
20784 @item @var{non-blank-sequence} @expansion{}
20785 @emph{anything, provided it doesn't contain special characters such as
20786 "-", @var{nl}, """ and of course " "}
20788 @item @var{c-string} @expansion{}
20789 @code{""" @var{seven-bit-iso-c-string-content} """}
20791 @item @var{nl} @expansion{}
20800 The CLI commands are still handled by the @sc{mi} interpreter; their
20801 output is described below.
20804 The @code{@var{token}}, when present, is passed back when the command
20808 Some @sc{mi} commands accept optional arguments as part of the parameter
20809 list. Each option is identified by a leading @samp{-} (dash) and may be
20810 followed by an optional argument parameter. Options occur first in the
20811 parameter list and can be delimited from normal parameters using
20812 @samp{--} (this is useful when some parameters begin with a dash).
20819 We want easy access to the existing CLI syntax (for debugging).
20822 We want it to be easy to spot a @sc{mi} operation.
20825 @node GDB/MI Output Syntax
20826 @subsection @sc{gdb/mi} Output Syntax
20828 @cindex output syntax of @sc{gdb/mi}
20829 @cindex @sc{gdb/mi}, output syntax
20830 The output from @sc{gdb/mi} consists of zero or more out-of-band records
20831 followed, optionally, by a single result record. This result record
20832 is for the most recent command. The sequence of output records is
20833 terminated by @samp{(gdb)}.
20835 If an input command was prefixed with a @code{@var{token}} then the
20836 corresponding output for that command will also be prefixed by that same
20840 @item @var{output} @expansion{}
20841 @code{( @var{out-of-band-record} )* [ @var{result-record} ] "(gdb)" @var{nl}}
20843 @item @var{result-record} @expansion{}
20844 @code{ [ @var{token} ] "^" @var{result-class} ( "," @var{result} )* @var{nl}}
20846 @item @var{out-of-band-record} @expansion{}
20847 @code{@var{async-record} | @var{stream-record}}
20849 @item @var{async-record} @expansion{}
20850 @code{@var{exec-async-output} | @var{status-async-output} | @var{notify-async-output}}
20852 @item @var{exec-async-output} @expansion{}
20853 @code{[ @var{token} ] "*" @var{async-output}}
20855 @item @var{status-async-output} @expansion{}
20856 @code{[ @var{token} ] "+" @var{async-output}}
20858 @item @var{notify-async-output} @expansion{}
20859 @code{[ @var{token} ] "=" @var{async-output}}
20861 @item @var{async-output} @expansion{}
20862 @code{@var{async-class} ( "," @var{result} )* @var{nl}}
20864 @item @var{result-class} @expansion{}
20865 @code{"done" | "running" | "connected" | "error" | "exit"}
20867 @item @var{async-class} @expansion{}
20868 @code{"stopped" | @var{others}} (where @var{others} will be added
20869 depending on the needs---this is still in development).
20871 @item @var{result} @expansion{}
20872 @code{ @var{variable} "=" @var{value}}
20874 @item @var{variable} @expansion{}
20875 @code{ @var{string} }
20877 @item @var{value} @expansion{}
20878 @code{ @var{const} | @var{tuple} | @var{list} }
20880 @item @var{const} @expansion{}
20881 @code{@var{c-string}}
20883 @item @var{tuple} @expansion{}
20884 @code{ "@{@}" | "@{" @var{result} ( "," @var{result} )* "@}" }
20886 @item @var{list} @expansion{}
20887 @code{ "[]" | "[" @var{value} ( "," @var{value} )* "]" | "["
20888 @var{result} ( "," @var{result} )* "]" }
20890 @item @var{stream-record} @expansion{}
20891 @code{@var{console-stream-output} | @var{target-stream-output} | @var{log-stream-output}}
20893 @item @var{console-stream-output} @expansion{}
20894 @code{"~" @var{c-string}}
20896 @item @var{target-stream-output} @expansion{}
20897 @code{"@@" @var{c-string}}
20899 @item @var{log-stream-output} @expansion{}
20900 @code{"&" @var{c-string}}
20902 @item @var{nl} @expansion{}
20905 @item @var{token} @expansion{}
20906 @emph{any sequence of digits}.
20914 All output sequences end in a single line containing a period.
20917 The @code{@var{token}} is from the corresponding request. Note that
20918 for all async output, while the token is allowed by the grammar and
20919 may be output by future versions of @value{GDBN} for select async
20920 output messages, it is generally omitted. Frontends should treat
20921 all async output as reporting general changes in the state of the
20922 target and there should be no need to associate async output to any
20926 @cindex status output in @sc{gdb/mi}
20927 @var{status-async-output} contains on-going status information about the
20928 progress of a slow operation. It can be discarded. All status output is
20929 prefixed by @samp{+}.
20932 @cindex async output in @sc{gdb/mi}
20933 @var{exec-async-output} contains asynchronous state change on the target
20934 (stopped, started, disappeared). All async output is prefixed by
20938 @cindex notify output in @sc{gdb/mi}
20939 @var{notify-async-output} contains supplementary information that the
20940 client should handle (e.g., a new breakpoint information). All notify
20941 output is prefixed by @samp{=}.
20944 @cindex console output in @sc{gdb/mi}
20945 @var{console-stream-output} is output that should be displayed as is in the
20946 console. It is the textual response to a CLI command. All the console
20947 output is prefixed by @samp{~}.
20950 @cindex target output in @sc{gdb/mi}
20951 @var{target-stream-output} is the output produced by the target program.
20952 All the target output is prefixed by @samp{@@}.
20955 @cindex log output in @sc{gdb/mi}
20956 @var{log-stream-output} is output text coming from @value{GDBN}'s internals, for
20957 instance messages that should be displayed as part of an error log. All
20958 the log output is prefixed by @samp{&}.
20961 @cindex list output in @sc{gdb/mi}
20962 New @sc{gdb/mi} commands should only output @var{lists} containing
20968 @xref{GDB/MI Stream Records, , @sc{gdb/mi} Stream Records}, for more
20969 details about the various output records.
20971 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
20972 @node GDB/MI Compatibility with CLI
20973 @section @sc{gdb/mi} Compatibility with CLI
20975 @cindex compatibility, @sc{gdb/mi} and CLI
20976 @cindex @sc{gdb/mi}, compatibility with CLI
20978 For the developers convenience CLI commands can be entered directly,
20979 but there may be some unexpected behaviour. For example, commands
20980 that query the user will behave as if the user replied yes, breakpoint
20981 command lists are not executed and some CLI commands, such as
20982 @code{if}, @code{when} and @code{define}, prompt for further input with
20983 @samp{>}, which is not valid MI output.
20985 This feature may be removed at some stage in the future and it is
20986 recommended that front ends use the @code{-interpreter-exec} command
20987 (@pxref{-interpreter-exec}).
20989 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
20990 @node GDB/MI Development and Front Ends
20991 @section @sc{gdb/mi} Development and Front Ends
20992 @cindex @sc{gdb/mi} development
20994 The application which takes the MI output and presents the state of the
20995 program being debugged to the user is called a @dfn{front end}.
20997 Although @sc{gdb/mi} is still incomplete, it is currently being used
20998 by a variety of front ends to @value{GDBN}. This makes it difficult
20999 to introduce new functionality without breaking existing usage. This
21000 section tries to minimize the problems by describing how the protocol
21003 Some changes in MI need not break a carefully designed front end, and
21004 for these the MI version will remain unchanged. The following is a
21005 list of changes that may occur within one level, so front ends should
21006 parse MI output in a way that can handle them:
21010 New MI commands may be added.
21013 New fields may be added to the output of any MI command.
21016 The range of values for fields with specified values, e.g.,
21017 @code{in_scope} (@pxref{-var-update}) may be extended.
21019 @c The format of field's content e.g type prefix, may change so parse it
21020 @c at your own risk. Yes, in general?
21022 @c The order of fields may change? Shouldn't really matter but it might
21023 @c resolve inconsistencies.
21026 If the changes are likely to break front ends, the MI version level
21027 will be increased by one. This will allow the front end to parse the
21028 output according to the MI version. Apart from mi0, new versions of
21029 @value{GDBN} will not support old versions of MI and it will be the
21030 responsibility of the front end to work with the new one.
21032 @c Starting with mi3, add a new command -mi-version that prints the MI
21035 The best way to avoid unexpected changes in MI that might break your front
21036 end is to make your project known to @value{GDBN} developers and
21037 follow development on @email{gdb@@sourceware.org} and
21038 @email{gdb-patches@@sourceware.org}.
21039 @cindex mailing lists
21041 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
21042 @node GDB/MI Output Records
21043 @section @sc{gdb/mi} Output Records
21046 * GDB/MI Result Records::
21047 * GDB/MI Stream Records::
21048 * GDB/MI Async Records::
21049 * GDB/MI Frame Information::
21052 @node GDB/MI Result Records
21053 @subsection @sc{gdb/mi} Result Records
21055 @cindex result records in @sc{gdb/mi}
21056 @cindex @sc{gdb/mi}, result records
21057 In addition to a number of out-of-band notifications, the response to a
21058 @sc{gdb/mi} command includes one of the following result indications:
21062 @item "^done" [ "," @var{results} ]
21063 The synchronous operation was successful, @code{@var{results}} are the return
21068 @c Is this one correct? Should it be an out-of-band notification?
21069 The asynchronous operation was successfully started. The target is
21074 @value{GDBN} has connected to a remote target.
21076 @item "^error" "," @var{c-string}
21078 The operation failed. The @code{@var{c-string}} contains the corresponding
21083 @value{GDBN} has terminated.
21087 @node GDB/MI Stream Records
21088 @subsection @sc{gdb/mi} Stream Records
21090 @cindex @sc{gdb/mi}, stream records
21091 @cindex stream records in @sc{gdb/mi}
21092 @value{GDBN} internally maintains a number of output streams: the console, the
21093 target, and the log. The output intended for each of these streams is
21094 funneled through the @sc{gdb/mi} interface using @dfn{stream records}.
21096 Each stream record begins with a unique @dfn{prefix character} which
21097 identifies its stream (@pxref{GDB/MI Output Syntax, , @sc{gdb/mi} Output
21098 Syntax}). In addition to the prefix, each stream record contains a
21099 @code{@var{string-output}}. This is either raw text (with an implicit new
21100 line) or a quoted C string (which does not contain an implicit newline).
21103 @item "~" @var{string-output}
21104 The console output stream contains text that should be displayed in the
21105 CLI console window. It contains the textual responses to CLI commands.
21107 @item "@@" @var{string-output}
21108 The target output stream contains any textual output from the running
21109 target. This is only present when GDB's event loop is truly
21110 asynchronous, which is currently only the case for remote targets.
21112 @item "&" @var{string-output}
21113 The log stream contains debugging messages being produced by @value{GDBN}'s
21117 @node GDB/MI Async Records
21118 @subsection @sc{gdb/mi} Async Records
21120 @cindex async records in @sc{gdb/mi}
21121 @cindex @sc{gdb/mi}, async records
21122 @dfn{Async} records are used to notify the @sc{gdb/mi} client of
21123 additional changes that have occurred. Those changes can either be a
21124 consequence of @sc{gdb/mi} commands (e.g., a breakpoint modified) or a result of
21125 target activity (e.g., target stopped).
21127 The following is the list of possible async records:
21131 @item *running,thread-id="@var{thread}"
21132 The target is now running. The @var{thread} field tells which
21133 specific thread is now running, and can be @samp{all} if all threads
21134 are running. The frontend should assume that no interaction with a
21135 running thread is possible after this notification is produced.
21136 The frontend should not assume that this notification is output
21137 only once for any command. @value{GDBN} may emit this notification
21138 several times, either for different threads, because it cannot resume
21139 all threads together, or even for a single thread, if the thread must
21140 be stepped though some code before letting it run freely.
21142 @item *stopped,reason="@var{reason}",thread-id="@var{id}",stopped-threads="@var{stopped}"
21143 The target has stopped. The @var{reason} field can have one of the
21147 @item breakpoint-hit
21148 A breakpoint was reached.
21149 @item watchpoint-trigger
21150 A watchpoint was triggered.
21151 @item read-watchpoint-trigger
21152 A read watchpoint was triggered.
21153 @item access-watchpoint-trigger
21154 An access watchpoint was triggered.
21155 @item function-finished
21156 An -exec-finish or similar CLI command was accomplished.
21157 @item location-reached
21158 An -exec-until or similar CLI command was accomplished.
21159 @item watchpoint-scope
21160 A watchpoint has gone out of scope.
21161 @item end-stepping-range
21162 An -exec-next, -exec-next-instruction, -exec-step, -exec-step-instruction or
21163 similar CLI command was accomplished.
21164 @item exited-signalled
21165 The inferior exited because of a signal.
21167 The inferior exited.
21168 @item exited-normally
21169 The inferior exited normally.
21170 @item signal-received
21171 A signal was received by the inferior.
21174 The @var{id} field identifies the thread that directly caused the stop
21175 -- for example by hitting a breakpoint. Depending on whether all-stop
21176 mode is in effect (@pxref{All-Stop Mode}), @value{GDBN} may either
21177 stop all threads, or only the thread that directly triggered the stop.
21178 If all threads are stopped, the @var{stopped} field will have the
21179 value of @code{"all"}. Otherwise, the value of the @var{stopped}
21180 field will be a list of thread identifiers. Presently, this list will
21181 always include a single thread, but frontend should be prepared to see
21182 several threads in the list.
21184 @item =thread-group-created,id="@var{id}"
21185 @itemx =thread-group-exited,id="@var{id}"
21186 A thread thread group either was attached to, or has exited/detached
21187 from. The @var{id} field contains the @value{GDBN} identifier of the
21190 @item =thread-created,id="@var{id}",group-id="@var{gid}"
21191 @itemx =thread-exited,id="@var{id}",group-id="@var{gid}"
21192 A thread either was created, or has exited. The @var{id} field
21193 contains the @value{GDBN} identifier of the thread. The @var{gid}
21194 field identifies the thread group this thread belongs to.
21196 @item =thread-selected,id="@var{id}"
21197 Informs that the selected thread was changed as result of the last
21198 command. This notification is not emitted as result of @code{-thread-select}
21199 command but is emitted whenever an MI command that is not documented
21200 to change the selected thread actually changes it. In particular,
21201 invoking, directly or indirectly (via user-defined command), the CLI
21202 @code{thread} command, will generate this notification.
21204 We suggest that in response to this notification, front ends
21205 highlight the selected thread and cause subsequent commands to apply to
21208 @item =library-loaded,...
21209 Reports that a new library file was loaded by the program. This
21210 notification has 4 fields---@var{id}, @var{target-name},
21211 @var{host-name}, and @var{symbols-loaded}. The @var{id} field is an
21212 opaque identifier of the library. For remote debugging case,
21213 @var{target-name} and @var{host-name} fields give the name of the
21214 library file on the target, and on the host respectively. For native
21215 debugging, both those fields have the same value. The
21216 @var{symbols-loaded} field reports if the debug symbols for this
21217 library are loaded.
21219 @item =library-unloaded,...
21220 Reports that a library was unloaded by the program. This notification
21221 has 3 fields---@var{id}, @var{target-name} and @var{host-name} with
21222 the same meaning as for the @code{=library-loaded} notification
21226 @node GDB/MI Frame Information
21227 @subsection @sc{gdb/mi} Frame Information
21229 Response from many MI commands includes an information about stack
21230 frame. This information is a tuple that may have the following
21235 The level of the stack frame. The innermost frame has the level of
21236 zero. This field is always present.
21239 The name of the function corresponding to the frame. This field may
21240 be absent if @value{GDBN} is unable to determine the function name.
21243 The code address for the frame. This field is always present.
21246 The name of the source files that correspond to the frame's code
21247 address. This field may be absent.
21250 The source line corresponding to the frames' code address. This field
21254 The name of the binary file (either executable or shared library) the
21255 corresponds to the frame's code address. This field may be absent.
21260 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
21261 @node GDB/MI Simple Examples
21262 @section Simple Examples of @sc{gdb/mi} Interaction
21263 @cindex @sc{gdb/mi}, simple examples
21265 This subsection presents several simple examples of interaction using
21266 the @sc{gdb/mi} interface. In these examples, @samp{->} means that the
21267 following line is passed to @sc{gdb/mi} as input, while @samp{<-} means
21268 the output received from @sc{gdb/mi}.
21270 Note the line breaks shown in the examples are here only for
21271 readability, they don't appear in the real output.
21273 @subheading Setting a Breakpoint
21275 Setting a breakpoint generates synchronous output which contains detailed
21276 information of the breakpoint.
21279 -> -break-insert main
21280 <- ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
21281 enabled="y",addr="0x08048564",func="main",file="myprog.c",
21282 fullname="/home/nickrob/myprog.c",line="68",times="0"@}
21286 @subheading Program Execution
21288 Program execution generates asynchronous records and MI gives the
21289 reason that execution stopped.
21295 <- *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
21296 frame=@{addr="0x08048564",func="main",
21297 args=[@{name="argc",value="1"@},@{name="argv",value="0xbfc4d4d4"@}],
21298 file="myprog.c",fullname="/home/nickrob/myprog.c",line="68"@}
21303 <- *stopped,reason="exited-normally"
21307 @subheading Quitting @value{GDBN}
21309 Quitting @value{GDBN} just prints the result class @samp{^exit}.
21317 @subheading A Bad Command
21319 Here's what happens if you pass a non-existent command:
21323 <- ^error,msg="Undefined MI command: rubbish"
21328 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
21329 @node GDB/MI Command Description Format
21330 @section @sc{gdb/mi} Command Description Format
21332 The remaining sections describe blocks of commands. Each block of
21333 commands is laid out in a fashion similar to this section.
21335 @subheading Motivation
21337 The motivation for this collection of commands.
21339 @subheading Introduction
21341 A brief introduction to this collection of commands as a whole.
21343 @subheading Commands
21345 For each command in the block, the following is described:
21347 @subsubheading Synopsis
21350 -command @var{args}@dots{}
21353 @subsubheading Result
21355 @subsubheading @value{GDBN} Command
21357 The corresponding @value{GDBN} CLI command(s), if any.
21359 @subsubheading Example
21361 Example(s) formatted for readability. Some of the described commands have
21362 not been implemented yet and these are labeled N.A.@: (not available).
21365 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
21366 @node GDB/MI Breakpoint Commands
21367 @section @sc{gdb/mi} Breakpoint Commands
21369 @cindex breakpoint commands for @sc{gdb/mi}
21370 @cindex @sc{gdb/mi}, breakpoint commands
21371 This section documents @sc{gdb/mi} commands for manipulating
21374 @subheading The @code{-break-after} Command
21375 @findex -break-after
21377 @subsubheading Synopsis
21380 -break-after @var{number} @var{count}
21383 The breakpoint number @var{number} is not in effect until it has been
21384 hit @var{count} times. To see how this is reflected in the output of
21385 the @samp{-break-list} command, see the description of the
21386 @samp{-break-list} command below.
21388 @subsubheading @value{GDBN} Command
21390 The corresponding @value{GDBN} command is @samp{ignore}.
21392 @subsubheading Example
21397 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
21398 enabled="y",addr="0x000100d0",func="main",file="hello.c",
21399 fullname="/home/foo/hello.c",line="5",times="0"@}
21406 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
21407 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
21408 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
21409 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
21410 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
21411 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
21412 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
21413 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
21414 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
21415 line="5",times="0",ignore="3"@}]@}
21420 @subheading The @code{-break-catch} Command
21421 @findex -break-catch
21423 @subheading The @code{-break-commands} Command
21424 @findex -break-commands
21428 @subheading The @code{-break-condition} Command
21429 @findex -break-condition
21431 @subsubheading Synopsis
21434 -break-condition @var{number} @var{expr}
21437 Breakpoint @var{number} will stop the program only if the condition in
21438 @var{expr} is true. The condition becomes part of the
21439 @samp{-break-list} output (see the description of the @samp{-break-list}
21442 @subsubheading @value{GDBN} Command
21444 The corresponding @value{GDBN} command is @samp{condition}.
21446 @subsubheading Example
21450 -break-condition 1 1
21454 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
21455 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
21456 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
21457 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
21458 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
21459 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
21460 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
21461 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
21462 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
21463 line="5",cond="1",times="0",ignore="3"@}]@}
21467 @subheading The @code{-break-delete} Command
21468 @findex -break-delete
21470 @subsubheading Synopsis
21473 -break-delete ( @var{breakpoint} )+
21476 Delete the breakpoint(s) whose number(s) are specified in the argument
21477 list. This is obviously reflected in the breakpoint list.
21479 @subsubheading @value{GDBN} Command
21481 The corresponding @value{GDBN} command is @samp{delete}.
21483 @subsubheading Example
21491 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
21492 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
21493 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
21494 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
21495 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
21496 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
21497 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
21502 @subheading The @code{-break-disable} Command
21503 @findex -break-disable
21505 @subsubheading Synopsis
21508 -break-disable ( @var{breakpoint} )+
21511 Disable the named @var{breakpoint}(s). The field @samp{enabled} in the
21512 break list is now set to @samp{n} for the named @var{breakpoint}(s).
21514 @subsubheading @value{GDBN} Command
21516 The corresponding @value{GDBN} command is @samp{disable}.
21518 @subsubheading Example
21526 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
21527 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
21528 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
21529 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
21530 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
21531 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
21532 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
21533 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="n",
21534 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
21535 line="5",times="0"@}]@}
21539 @subheading The @code{-break-enable} Command
21540 @findex -break-enable
21542 @subsubheading Synopsis
21545 -break-enable ( @var{breakpoint} )+
21548 Enable (previously disabled) @var{breakpoint}(s).
21550 @subsubheading @value{GDBN} Command
21552 The corresponding @value{GDBN} command is @samp{enable}.
21554 @subsubheading Example
21562 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
21563 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
21564 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
21565 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
21566 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
21567 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
21568 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
21569 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
21570 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
21571 line="5",times="0"@}]@}
21575 @subheading The @code{-break-info} Command
21576 @findex -break-info
21578 @subsubheading Synopsis
21581 -break-info @var{breakpoint}
21585 Get information about a single breakpoint.
21587 @subsubheading @value{GDBN} Command
21589 The corresponding @value{GDBN} command is @samp{info break @var{breakpoint}}.
21591 @subsubheading Example
21594 @subheading The @code{-break-insert} Command
21595 @findex -break-insert
21597 @subsubheading Synopsis
21600 -break-insert [ -t ] [ -h ] [ -f ] [ -d ]
21601 [ -c @var{condition} ] [ -i @var{ignore-count} ]
21602 [ -p @var{thread} ] [ @var{location} ]
21606 If specified, @var{location}, can be one of:
21613 @item filename:linenum
21614 @item filename:function
21618 The possible optional parameters of this command are:
21622 Insert a temporary breakpoint.
21624 Insert a hardware breakpoint.
21625 @item -c @var{condition}
21626 Make the breakpoint conditional on @var{condition}.
21627 @item -i @var{ignore-count}
21628 Initialize the @var{ignore-count}.
21630 If @var{location} cannot be parsed (for example if it
21631 refers to unknown files or functions), create a pending
21632 breakpoint. Without this flag, @value{GDBN} will report
21633 an error, and won't create a breakpoint, if @var{location}
21636 Create a disabled breakpoint.
21639 @subsubheading Result
21641 The result is in the form:
21644 ^done,bkpt=@{number="@var{number}",type="@var{type}",disp="del"|"keep",
21645 enabled="y"|"n",addr="@var{hex}",func="@var{funcname}",file="@var{filename}",
21646 fullname="@var{full_filename}",line="@var{lineno}",[thread="@var{threadno},]
21647 times="@var{times}"@}
21651 where @var{number} is the @value{GDBN} number for this breakpoint,
21652 @var{funcname} is the name of the function where the breakpoint was
21653 inserted, @var{filename} is the name of the source file which contains
21654 this function, @var{lineno} is the source line number within that file
21655 and @var{times} the number of times that the breakpoint has been hit
21656 (always 0 for -break-insert but may be greater for -break-info or -break-list
21657 which use the same output).
21659 Note: this format is open to change.
21660 @c An out-of-band breakpoint instead of part of the result?
21662 @subsubheading @value{GDBN} Command
21664 The corresponding @value{GDBN} commands are @samp{break}, @samp{tbreak},
21665 @samp{hbreak}, @samp{thbreak}, and @samp{rbreak}.
21667 @subsubheading Example
21672 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",
21673 fullname="/home/foo/recursive2.c,line="4",times="0"@}
21675 -break-insert -t foo
21676 ^done,bkpt=@{number="2",addr="0x00010774",file="recursive2.c",
21677 fullname="/home/foo/recursive2.c,line="11",times="0"@}
21680 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
21681 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
21682 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
21683 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
21684 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
21685 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
21686 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
21687 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
21688 addr="0x0001072c", func="main",file="recursive2.c",
21689 fullname="/home/foo/recursive2.c,"line="4",times="0"@},
21690 bkpt=@{number="2",type="breakpoint",disp="del",enabled="y",
21691 addr="0x00010774",func="foo",file="recursive2.c",
21692 fullname="/home/foo/recursive2.c",line="11",times="0"@}]@}
21694 -break-insert -r foo.*
21695 ~int foo(int, int);
21696 ^done,bkpt=@{number="3",addr="0x00010774",file="recursive2.c,
21697 "fullname="/home/foo/recursive2.c",line="11",times="0"@}
21701 @subheading The @code{-break-list} Command
21702 @findex -break-list
21704 @subsubheading Synopsis
21710 Displays the list of inserted breakpoints, showing the following fields:
21714 number of the breakpoint
21716 type of the breakpoint: @samp{breakpoint} or @samp{watchpoint}
21718 should the breakpoint be deleted or disabled when it is hit: @samp{keep}
21721 is the breakpoint enabled or no: @samp{y} or @samp{n}
21723 memory location at which the breakpoint is set
21725 logical location of the breakpoint, expressed by function name, file
21728 number of times the breakpoint has been hit
21731 If there are no breakpoints or watchpoints, the @code{BreakpointTable}
21732 @code{body} field is an empty list.
21734 @subsubheading @value{GDBN} Command
21736 The corresponding @value{GDBN} command is @samp{info break}.
21738 @subsubheading Example
21743 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
21744 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
21745 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
21746 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
21747 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
21748 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
21749 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
21750 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
21751 addr="0x000100d0",func="main",file="hello.c",line="5",times="0"@},
21752 bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
21753 addr="0x00010114",func="foo",file="hello.c",fullname="/home/foo/hello.c",
21754 line="13",times="0"@}]@}
21758 Here's an example of the result when there are no breakpoints:
21763 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
21764 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
21765 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
21766 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
21767 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
21768 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
21769 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
21774 @subheading The @code{-break-watch} Command
21775 @findex -break-watch
21777 @subsubheading Synopsis
21780 -break-watch [ -a | -r ]
21783 Create a watchpoint. With the @samp{-a} option it will create an
21784 @dfn{access} watchpoint, i.e., a watchpoint that triggers either on a
21785 read from or on a write to the memory location. With the @samp{-r}
21786 option, the watchpoint created is a @dfn{read} watchpoint, i.e., it will
21787 trigger only when the memory location is accessed for reading. Without
21788 either of the options, the watchpoint created is a regular watchpoint,
21789 i.e., it will trigger when the memory location is accessed for writing.
21790 @xref{Set Watchpoints, , Setting Watchpoints}.
21792 Note that @samp{-break-list} will report a single list of watchpoints and
21793 breakpoints inserted.
21795 @subsubheading @value{GDBN} Command
21797 The corresponding @value{GDBN} commands are @samp{watch}, @samp{awatch}, and
21800 @subsubheading Example
21802 Setting a watchpoint on a variable in the @code{main} function:
21807 ^done,wpt=@{number="2",exp="x"@}
21812 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="x"@},
21813 value=@{old="-268439212",new="55"@},
21814 frame=@{func="main",args=[],file="recursive2.c",
21815 fullname="/home/foo/bar/recursive2.c",line="5"@}
21819 Setting a watchpoint on a variable local to a function. @value{GDBN} will stop
21820 the program execution twice: first for the variable changing value, then
21821 for the watchpoint going out of scope.
21826 ^done,wpt=@{number="5",exp="C"@}
21831 *stopped,reason="watchpoint-trigger",
21832 wpt=@{number="5",exp="C"@},value=@{old="-276895068",new="3"@},
21833 frame=@{func="callee4",args=[],
21834 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
21835 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13"@}
21840 *stopped,reason="watchpoint-scope",wpnum="5",
21841 frame=@{func="callee3",args=[@{name="strarg",
21842 value="0x11940 \"A string argument.\""@}],
21843 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
21844 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
21848 Listing breakpoints and watchpoints, at different points in the program
21849 execution. Note that once the watchpoint goes out of scope, it is
21855 ^done,wpt=@{number="2",exp="C"@}
21858 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
21859 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
21860 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
21861 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
21862 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
21863 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
21864 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
21865 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
21866 addr="0x00010734",func="callee4",
21867 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
21868 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c"line="8",times="1"@},
21869 bkpt=@{number="2",type="watchpoint",disp="keep",
21870 enabled="y",addr="",what="C",times="0"@}]@}
21875 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="C"@},
21876 value=@{old="-276895068",new="3"@},
21877 frame=@{func="callee4",args=[],
21878 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
21879 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13"@}
21882 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
21883 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
21884 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
21885 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
21886 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
21887 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
21888 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
21889 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
21890 addr="0x00010734",func="callee4",
21891 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
21892 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",times="1"@},
21893 bkpt=@{number="2",type="watchpoint",disp="keep",
21894 enabled="y",addr="",what="C",times="-5"@}]@}
21898 ^done,reason="watchpoint-scope",wpnum="2",
21899 frame=@{func="callee3",args=[@{name="strarg",
21900 value="0x11940 \"A string argument.\""@}],
21901 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
21902 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
21905 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
21906 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
21907 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
21908 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
21909 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
21910 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
21911 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
21912 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
21913 addr="0x00010734",func="callee4",
21914 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
21915 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",
21920 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
21921 @node GDB/MI Program Context
21922 @section @sc{gdb/mi} Program Context
21924 @subheading The @code{-exec-arguments} Command
21925 @findex -exec-arguments
21928 @subsubheading Synopsis
21931 -exec-arguments @var{args}
21934 Set the inferior program arguments, to be used in the next
21937 @subsubheading @value{GDBN} Command
21939 The corresponding @value{GDBN} command is @samp{set args}.
21941 @subsubheading Example
21945 -exec-arguments -v word
21952 @subheading The @code{-exec-show-arguments} Command
21953 @findex -exec-show-arguments
21955 @subsubheading Synopsis
21958 -exec-show-arguments
21961 Print the arguments of the program.
21963 @subsubheading @value{GDBN} Command
21965 The corresponding @value{GDBN} command is @samp{show args}.
21967 @subsubheading Example
21972 @subheading The @code{-environment-cd} Command
21973 @findex -environment-cd
21975 @subsubheading Synopsis
21978 -environment-cd @var{pathdir}
21981 Set @value{GDBN}'s working directory.
21983 @subsubheading @value{GDBN} Command
21985 The corresponding @value{GDBN} command is @samp{cd}.
21987 @subsubheading Example
21991 -environment-cd /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
21997 @subheading The @code{-environment-directory} Command
21998 @findex -environment-directory
22000 @subsubheading Synopsis
22003 -environment-directory [ -r ] [ @var{pathdir} ]+
22006 Add directories @var{pathdir} to beginning of search path for source files.
22007 If the @samp{-r} option is used, the search path is reset to the default
22008 search path. If directories @var{pathdir} are supplied in addition to the
22009 @samp{-r} option, the search path is first reset and then addition
22011 Multiple directories may be specified, separated by blanks. Specifying
22012 multiple directories in a single command
22013 results in the directories added to the beginning of the
22014 search path in the same order they were presented in the command.
22015 If blanks are needed as
22016 part of a directory name, double-quotes should be used around
22017 the name. In the command output, the path will show up separated
22018 by the system directory-separator character. The directory-separator
22019 character must not be used
22020 in any directory name.
22021 If no directories are specified, the current search path is displayed.
22023 @subsubheading @value{GDBN} Command
22025 The corresponding @value{GDBN} command is @samp{dir}.
22027 @subsubheading Example
22031 -environment-directory /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
22032 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
22034 -environment-directory ""
22035 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
22037 -environment-directory -r /home/jjohnstn/src/gdb /usr/src
22038 ^done,source-path="/home/jjohnstn/src/gdb:/usr/src:$cdir:$cwd"
22040 -environment-directory -r
22041 ^done,source-path="$cdir:$cwd"
22046 @subheading The @code{-environment-path} Command
22047 @findex -environment-path
22049 @subsubheading Synopsis
22052 -environment-path [ -r ] [ @var{pathdir} ]+
22055 Add directories @var{pathdir} to beginning of search path for object files.
22056 If the @samp{-r} option is used, the search path is reset to the original
22057 search path that existed at gdb start-up. If directories @var{pathdir} are
22058 supplied in addition to the
22059 @samp{-r} option, the search path is first reset and then addition
22061 Multiple directories may be specified, separated by blanks. Specifying
22062 multiple directories in a single command
22063 results in the directories added to the beginning of the
22064 search path in the same order they were presented in the command.
22065 If blanks are needed as
22066 part of a directory name, double-quotes should be used around
22067 the name. In the command output, the path will show up separated
22068 by the system directory-separator character. The directory-separator
22069 character must not be used
22070 in any directory name.
22071 If no directories are specified, the current path is displayed.
22074 @subsubheading @value{GDBN} Command
22076 The corresponding @value{GDBN} command is @samp{path}.
22078 @subsubheading Example
22083 ^done,path="/usr/bin"
22085 -environment-path /kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb /bin
22086 ^done,path="/kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb:/bin:/usr/bin"
22088 -environment-path -r /usr/local/bin
22089 ^done,path="/usr/local/bin:/usr/bin"
22094 @subheading The @code{-environment-pwd} Command
22095 @findex -environment-pwd
22097 @subsubheading Synopsis
22103 Show the current working directory.
22105 @subsubheading @value{GDBN} Command
22107 The corresponding @value{GDBN} command is @samp{pwd}.
22109 @subsubheading Example
22114 ^done,cwd="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb"
22118 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
22119 @node GDB/MI Thread Commands
22120 @section @sc{gdb/mi} Thread Commands
22123 @subheading The @code{-thread-info} Command
22124 @findex -thread-info
22126 @subsubheading Synopsis
22129 -thread-info [ @var{thread-id} ]
22132 Reports information about either a specific thread, if
22133 the @var{thread-id} parameter is present, or about all
22134 threads. When printing information about all threads,
22135 also reports the current thread.
22137 @subsubheading @value{GDBN} Command
22139 The @samp{info thread} command prints the same information
22142 @subsubheading Example
22147 @{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
22148 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",args=[]@},state="running"@},
22149 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
22150 frame=@{level="0",addr="0x0804891f",func="foo",args=[@{name="i",value="10"@}],
22151 file="/tmp/a.c",fullname="/tmp/a.c",line="158"@},state="running"@}],
22152 current-thread-id="1"
22156 The @samp{state} field may have the following values:
22160 The thread is stopped. Frame information is available for stopped
22164 The thread is running. There's no frame information for running
22169 @subheading The @code{-thread-list-ids} Command
22170 @findex -thread-list-ids
22172 @subsubheading Synopsis
22178 Produces a list of the currently known @value{GDBN} thread ids. At the
22179 end of the list it also prints the total number of such threads.
22181 This command is retained for historical reasons, the
22182 @code{-thread-info} command should be used instead.
22184 @subsubheading @value{GDBN} Command
22186 Part of @samp{info threads} supplies the same information.
22188 @subsubheading Example
22193 ^done,thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
22194 current-thread-id="1",number-of-threads="3"
22199 @subheading The @code{-thread-select} Command
22200 @findex -thread-select
22202 @subsubheading Synopsis
22205 -thread-select @var{threadnum}
22208 Make @var{threadnum} the current thread. It prints the number of the new
22209 current thread, and the topmost frame for that thread.
22211 This command is deprecated in favor of explicitly using the
22212 @samp{--thread} option to each command.
22214 @subsubheading @value{GDBN} Command
22216 The corresponding @value{GDBN} command is @samp{thread}.
22218 @subsubheading Example
22225 *stopped,reason="end-stepping-range",thread-id="2",line="187",
22226 file="../../../devo/gdb/testsuite/gdb.threads/linux-dp.c"
22230 thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
22231 number-of-threads="3"
22234 ^done,new-thread-id="3",
22235 frame=@{level="0",func="vprintf",
22236 args=[@{name="format",value="0x8048e9c \"%*s%c %d %c\\n\""@},
22237 @{name="arg",value="0x2"@}],file="vprintf.c",line="31"@}
22241 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
22242 @node GDB/MI Program Execution
22243 @section @sc{gdb/mi} Program Execution
22245 These are the asynchronous commands which generate the out-of-band
22246 record @samp{*stopped}. Currently @value{GDBN} only really executes
22247 asynchronously with remote targets and this interaction is mimicked in
22250 @subheading The @code{-exec-continue} Command
22251 @findex -exec-continue
22253 @subsubheading Synopsis
22256 -exec-continue [--all|--thread-group N]
22259 Resumes the execution of the inferior program until a breakpoint is
22260 encountered, or until the inferior exits. In all-stop mode
22261 (@pxref{All-Stop Mode}), may resume only one thread, or all threads,
22262 depending on the value of the @samp{scheduler-locking} variable. In
22263 non-stop mode (@pxref{Non-Stop Mode}), if the @samp{--all} is not
22264 specified, only the thread specified with the @samp{--thread} option
22265 (or current thread, if no @samp{--thread} is provided) is resumed. If
22266 @samp{--all} is specified, all threads will be resumed. The
22267 @samp{--all} option is ignored in all-stop mode. If the
22268 @samp{--thread-group} options is specified, then all threads in that
22269 thread group are resumed.
22271 @subsubheading @value{GDBN} Command
22273 The corresponding @value{GDBN} corresponding is @samp{continue}.
22275 @subsubheading Example
22282 *stopped,reason="breakpoint-hit",disp="keep",bkptno="2",frame=@{
22283 func="foo",args=[],file="hello.c",fullname="/home/foo/bar/hello.c",
22289 @subheading The @code{-exec-finish} Command
22290 @findex -exec-finish
22292 @subsubheading Synopsis
22298 Resumes the execution of the inferior program until the current
22299 function is exited. Displays the results returned by the function.
22301 @subsubheading @value{GDBN} Command
22303 The corresponding @value{GDBN} command is @samp{finish}.
22305 @subsubheading Example
22307 Function returning @code{void}.
22314 *stopped,reason="function-finished",frame=@{func="main",args=[],
22315 file="hello.c",fullname="/home/foo/bar/hello.c",line="7"@}
22319 Function returning other than @code{void}. The name of the internal
22320 @value{GDBN} variable storing the result is printed, together with the
22327 *stopped,reason="function-finished",frame=@{addr="0x000107b0",func="foo",
22328 args=[@{name="a",value="1"],@{name="b",value="9"@}@},
22329 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
22330 gdb-result-var="$1",return-value="0"
22335 @subheading The @code{-exec-interrupt} Command
22336 @findex -exec-interrupt
22338 @subsubheading Synopsis
22341 -exec-interrupt [--all|--thread-group N]
22344 Interrupts the background execution of the target. Note how the token
22345 associated with the stop message is the one for the execution command
22346 that has been interrupted. The token for the interrupt itself only
22347 appears in the @samp{^done} output. If the user is trying to
22348 interrupt a non-running program, an error message will be printed.
22350 Note that when asynchronous execution is enabled, this command is
22351 asynchronous just like other execution commands. That is, first the
22352 @samp{^done} response will be printed, and the target stop will be
22353 reported after that using the @samp{*stopped} notification.
22355 In non-stop mode, only the context thread is interrupted by default.
22356 All threads will be interrupted if the @samp{--all} option is
22357 specified. If the @samp{--thread-group} option is specified, all
22358 threads in that group will be interrupted.
22360 @subsubheading @value{GDBN} Command
22362 The corresponding @value{GDBN} command is @samp{interrupt}.
22364 @subsubheading Example
22375 111*stopped,signal-name="SIGINT",signal-meaning="Interrupt",
22376 frame=@{addr="0x00010140",func="foo",args=[],file="try.c",
22377 fullname="/home/foo/bar/try.c",line="13"@}
22382 ^error,msg="mi_cmd_exec_interrupt: Inferior not executing."
22386 @subheading The @code{-exec-jump} Command
22389 @subsubheading Synopsis
22392 -exec-jump @var{location}
22395 Resumes execution of the inferior program at the location specified by
22396 parameter. @xref{Specify Location}, for a description of the
22397 different forms of @var{location}.
22399 @subsubheading @value{GDBN} Command
22401 The corresponding @value{GDBN} command is @samp{jump}.
22403 @subsubheading Example
22406 -exec-jump foo.c:10
22407 *running,thread-id="all"
22412 @subheading The @code{-exec-next} Command
22415 @subsubheading Synopsis
22421 Resumes execution of the inferior program, stopping when the beginning
22422 of the next source line is reached.
22424 @subsubheading @value{GDBN} Command
22426 The corresponding @value{GDBN} command is @samp{next}.
22428 @subsubheading Example
22434 *stopped,reason="end-stepping-range",line="8",file="hello.c"
22439 @subheading The @code{-exec-next-instruction} Command
22440 @findex -exec-next-instruction
22442 @subsubheading Synopsis
22445 -exec-next-instruction
22448 Executes one machine instruction. If the instruction is a function
22449 call, continues until the function returns. If the program stops at an
22450 instruction in the middle of a source line, the address will be
22453 @subsubheading @value{GDBN} Command
22455 The corresponding @value{GDBN} command is @samp{nexti}.
22457 @subsubheading Example
22461 -exec-next-instruction
22465 *stopped,reason="end-stepping-range",
22466 addr="0x000100d4",line="5",file="hello.c"
22471 @subheading The @code{-exec-return} Command
22472 @findex -exec-return
22474 @subsubheading Synopsis
22480 Makes current function return immediately. Doesn't execute the inferior.
22481 Displays the new current frame.
22483 @subsubheading @value{GDBN} Command
22485 The corresponding @value{GDBN} command is @samp{return}.
22487 @subsubheading Example
22491 200-break-insert callee4
22492 200^done,bkpt=@{number="1",addr="0x00010734",
22493 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",line="8"@}
22498 000*stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
22499 frame=@{func="callee4",args=[],
22500 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
22501 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8"@}
22507 111^done,frame=@{level="0",func="callee3",
22508 args=[@{name="strarg",
22509 value="0x11940 \"A string argument.\""@}],
22510 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
22511 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
22516 @subheading The @code{-exec-run} Command
22519 @subsubheading Synopsis
22525 Starts execution of the inferior from the beginning. The inferior
22526 executes until either a breakpoint is encountered or the program
22527 exits. In the latter case the output will include an exit code, if
22528 the program has exited exceptionally.
22530 @subsubheading @value{GDBN} Command
22532 The corresponding @value{GDBN} command is @samp{run}.
22534 @subsubheading Examples
22539 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",line="4"@}
22544 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
22545 frame=@{func="main",args=[],file="recursive2.c",
22546 fullname="/home/foo/bar/recursive2.c",line="4"@}
22551 Program exited normally:
22559 *stopped,reason="exited-normally"
22564 Program exited exceptionally:
22572 *stopped,reason="exited",exit-code="01"
22576 Another way the program can terminate is if it receives a signal such as
22577 @code{SIGINT}. In this case, @sc{gdb/mi} displays this:
22581 *stopped,reason="exited-signalled",signal-name="SIGINT",
22582 signal-meaning="Interrupt"
22586 @c @subheading -exec-signal
22589 @subheading The @code{-exec-step} Command
22592 @subsubheading Synopsis
22598 Resumes execution of the inferior program, stopping when the beginning
22599 of the next source line is reached, if the next source line is not a
22600 function call. If it is, stop at the first instruction of the called
22603 @subsubheading @value{GDBN} Command
22605 The corresponding @value{GDBN} command is @samp{step}.
22607 @subsubheading Example
22609 Stepping into a function:
22615 *stopped,reason="end-stepping-range",
22616 frame=@{func="foo",args=[@{name="a",value="10"@},
22617 @{name="b",value="0"@}],file="recursive2.c",
22618 fullname="/home/foo/bar/recursive2.c",line="11"@}
22628 *stopped,reason="end-stepping-range",line="14",file="recursive2.c"
22633 @subheading The @code{-exec-step-instruction} Command
22634 @findex -exec-step-instruction
22636 @subsubheading Synopsis
22639 -exec-step-instruction
22642 Resumes the inferior which executes one machine instruction. The
22643 output, once @value{GDBN} has stopped, will vary depending on whether
22644 we have stopped in the middle of a source line or not. In the former
22645 case, the address at which the program stopped will be printed as
22648 @subsubheading @value{GDBN} Command
22650 The corresponding @value{GDBN} command is @samp{stepi}.
22652 @subsubheading Example
22656 -exec-step-instruction
22660 *stopped,reason="end-stepping-range",
22661 frame=@{func="foo",args=[],file="try.c",
22662 fullname="/home/foo/bar/try.c",line="10"@}
22664 -exec-step-instruction
22668 *stopped,reason="end-stepping-range",
22669 frame=@{addr="0x000100f4",func="foo",args=[],file="try.c",
22670 fullname="/home/foo/bar/try.c",line="10"@}
22675 @subheading The @code{-exec-until} Command
22676 @findex -exec-until
22678 @subsubheading Synopsis
22681 -exec-until [ @var{location} ]
22684 Executes the inferior until the @var{location} specified in the
22685 argument is reached. If there is no argument, the inferior executes
22686 until a source line greater than the current one is reached. The
22687 reason for stopping in this case will be @samp{location-reached}.
22689 @subsubheading @value{GDBN} Command
22691 The corresponding @value{GDBN} command is @samp{until}.
22693 @subsubheading Example
22697 -exec-until recursive2.c:6
22701 *stopped,reason="location-reached",frame=@{func="main",args=[],
22702 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="6"@}
22707 @subheading -file-clear
22708 Is this going away????
22711 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
22712 @node GDB/MI Stack Manipulation
22713 @section @sc{gdb/mi} Stack Manipulation Commands
22716 @subheading The @code{-stack-info-frame} Command
22717 @findex -stack-info-frame
22719 @subsubheading Synopsis
22725 Get info on the selected frame.
22727 @subsubheading @value{GDBN} Command
22729 The corresponding @value{GDBN} command is @samp{info frame} or @samp{frame}
22730 (without arguments).
22732 @subsubheading Example
22737 ^done,frame=@{level="1",addr="0x0001076c",func="callee3",
22738 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
22739 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17"@}
22743 @subheading The @code{-stack-info-depth} Command
22744 @findex -stack-info-depth
22746 @subsubheading Synopsis
22749 -stack-info-depth [ @var{max-depth} ]
22752 Return the depth of the stack. If the integer argument @var{max-depth}
22753 is specified, do not count beyond @var{max-depth} frames.
22755 @subsubheading @value{GDBN} Command
22757 There's no equivalent @value{GDBN} command.
22759 @subsubheading Example
22761 For a stack with frame levels 0 through 11:
22768 -stack-info-depth 4
22771 -stack-info-depth 12
22774 -stack-info-depth 11
22777 -stack-info-depth 13
22782 @subheading The @code{-stack-list-arguments} Command
22783 @findex -stack-list-arguments
22785 @subsubheading Synopsis
22788 -stack-list-arguments @var{show-values}
22789 [ @var{low-frame} @var{high-frame} ]
22792 Display a list of the arguments for the frames between @var{low-frame}
22793 and @var{high-frame} (inclusive). If @var{low-frame} and
22794 @var{high-frame} are not provided, list the arguments for the whole
22795 call stack. If the two arguments are equal, show the single frame
22796 at the corresponding level. It is an error if @var{low-frame} is
22797 larger than the actual number of frames. On the other hand,
22798 @var{high-frame} may be larger than the actual number of frames, in
22799 which case only existing frames will be returned.
22801 The @var{show-values} argument must have a value of 0 or 1. A value of
22802 0 means that only the names of the arguments are listed, a value of 1
22803 means that both names and values of the arguments are printed.
22805 @subsubheading @value{GDBN} Command
22807 @value{GDBN} does not have an equivalent command. @code{gdbtk} has a
22808 @samp{gdb_get_args} command which partially overlaps with the
22809 functionality of @samp{-stack-list-arguments}.
22811 @subsubheading Example
22818 frame=@{level="0",addr="0x00010734",func="callee4",
22819 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
22820 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8"@},
22821 frame=@{level="1",addr="0x0001076c",func="callee3",
22822 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
22823 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17"@},
22824 frame=@{level="2",addr="0x0001078c",func="callee2",
22825 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
22826 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="22"@},
22827 frame=@{level="3",addr="0x000107b4",func="callee1",
22828 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
22829 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="27"@},
22830 frame=@{level="4",addr="0x000107e0",func="main",
22831 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
22832 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="32"@}]
22834 -stack-list-arguments 0
22837 frame=@{level="0",args=[]@},
22838 frame=@{level="1",args=[name="strarg"]@},
22839 frame=@{level="2",args=[name="intarg",name="strarg"]@},
22840 frame=@{level="3",args=[name="intarg",name="strarg",name="fltarg"]@},
22841 frame=@{level="4",args=[]@}]
22843 -stack-list-arguments 1
22846 frame=@{level="0",args=[]@},
22848 args=[@{name="strarg",value="0x11940 \"A string argument.\""@}]@},
22849 frame=@{level="2",args=[
22850 @{name="intarg",value="2"@},
22851 @{name="strarg",value="0x11940 \"A string argument.\""@}]@},
22852 @{frame=@{level="3",args=[
22853 @{name="intarg",value="2"@},
22854 @{name="strarg",value="0x11940 \"A string argument.\""@},
22855 @{name="fltarg",value="3.5"@}]@},
22856 frame=@{level="4",args=[]@}]
22858 -stack-list-arguments 0 2 2
22859 ^done,stack-args=[frame=@{level="2",args=[name="intarg",name="strarg"]@}]
22861 -stack-list-arguments 1 2 2
22862 ^done,stack-args=[frame=@{level="2",
22863 args=[@{name="intarg",value="2"@},
22864 @{name="strarg",value="0x11940 \"A string argument.\""@}]@}]
22868 @c @subheading -stack-list-exception-handlers
22871 @subheading The @code{-stack-list-frames} Command
22872 @findex -stack-list-frames
22874 @subsubheading Synopsis
22877 -stack-list-frames [ @var{low-frame} @var{high-frame} ]
22880 List the frames currently on the stack. For each frame it displays the
22885 The frame number, 0 being the topmost frame, i.e., the innermost function.
22887 The @code{$pc} value for that frame.
22891 File name of the source file where the function lives.
22893 Line number corresponding to the @code{$pc}.
22896 If invoked without arguments, this command prints a backtrace for the
22897 whole stack. If given two integer arguments, it shows the frames whose
22898 levels are between the two arguments (inclusive). If the two arguments
22899 are equal, it shows the single frame at the corresponding level. It is
22900 an error if @var{low-frame} is larger than the actual number of
22901 frames. On the other hand, @var{high-frame} may be larger than the
22902 actual number of frames, in which case only existing frames will be returned.
22904 @subsubheading @value{GDBN} Command
22906 The corresponding @value{GDBN} commands are @samp{backtrace} and @samp{where}.
22908 @subsubheading Example
22910 Full stack backtrace:
22916 [frame=@{level="0",addr="0x0001076c",func="foo",
22917 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="11"@},
22918 frame=@{level="1",addr="0x000107a4",func="foo",
22919 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
22920 frame=@{level="2",addr="0x000107a4",func="foo",
22921 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
22922 frame=@{level="3",addr="0x000107a4",func="foo",
22923 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
22924 frame=@{level="4",addr="0x000107a4",func="foo",
22925 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
22926 frame=@{level="5",addr="0x000107a4",func="foo",
22927 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
22928 frame=@{level="6",addr="0x000107a4",func="foo",
22929 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
22930 frame=@{level="7",addr="0x000107a4",func="foo",
22931 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
22932 frame=@{level="8",addr="0x000107a4",func="foo",
22933 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
22934 frame=@{level="9",addr="0x000107a4",func="foo",
22935 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
22936 frame=@{level="10",addr="0x000107a4",func="foo",
22937 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
22938 frame=@{level="11",addr="0x00010738",func="main",
22939 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="4"@}]
22943 Show frames between @var{low_frame} and @var{high_frame}:
22947 -stack-list-frames 3 5
22949 [frame=@{level="3",addr="0x000107a4",func="foo",
22950 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
22951 frame=@{level="4",addr="0x000107a4",func="foo",
22952 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
22953 frame=@{level="5",addr="0x000107a4",func="foo",
22954 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@}]
22958 Show a single frame:
22962 -stack-list-frames 3 3
22964 [frame=@{level="3",addr="0x000107a4",func="foo",
22965 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@}]
22970 @subheading The @code{-stack-list-locals} Command
22971 @findex -stack-list-locals
22973 @subsubheading Synopsis
22976 -stack-list-locals @var{print-values}
22979 Display the local variable names for the selected frame. If
22980 @var{print-values} is 0 or @code{--no-values}, print only the names of
22981 the variables; if it is 1 or @code{--all-values}, print also their
22982 values; and if it is 2 or @code{--simple-values}, print the name,
22983 type and value for simple data types and the name and type for arrays,
22984 structures and unions. In this last case, a frontend can immediately
22985 display the value of simple data types and create variable objects for
22986 other data types when the user wishes to explore their values in
22989 @subsubheading @value{GDBN} Command
22991 @samp{info locals} in @value{GDBN}, @samp{gdb_get_locals} in @code{gdbtk}.
22993 @subsubheading Example
22997 -stack-list-locals 0
22998 ^done,locals=[name="A",name="B",name="C"]
23000 -stack-list-locals --all-values
23001 ^done,locals=[@{name="A",value="1"@},@{name="B",value="2"@},
23002 @{name="C",value="@{1, 2, 3@}"@}]
23003 -stack-list-locals --simple-values
23004 ^done,locals=[@{name="A",type="int",value="1"@},
23005 @{name="B",type="int",value="2"@},@{name="C",type="int [3]"@}]
23010 @subheading The @code{-stack-select-frame} Command
23011 @findex -stack-select-frame
23013 @subsubheading Synopsis
23016 -stack-select-frame @var{framenum}
23019 Change the selected frame. Select a different frame @var{framenum} on
23022 This command in deprecated in favor of passing the @samp{--frame}
23023 option to every command.
23025 @subsubheading @value{GDBN} Command
23027 The corresponding @value{GDBN} commands are @samp{frame}, @samp{up},
23028 @samp{down}, @samp{select-frame}, @samp{up-silent}, and @samp{down-silent}.
23030 @subsubheading Example
23034 -stack-select-frame 2
23039 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
23040 @node GDB/MI Variable Objects
23041 @section @sc{gdb/mi} Variable Objects
23045 @subheading Motivation for Variable Objects in @sc{gdb/mi}
23047 For the implementation of a variable debugger window (locals, watched
23048 expressions, etc.), we are proposing the adaptation of the existing code
23049 used by @code{Insight}.
23051 The two main reasons for that are:
23055 It has been proven in practice (it is already on its second generation).
23058 It will shorten development time (needless to say how important it is
23062 The original interface was designed to be used by Tcl code, so it was
23063 slightly changed so it could be used through @sc{gdb/mi}. This section
23064 describes the @sc{gdb/mi} operations that will be available and gives some
23065 hints about their use.
23067 @emph{Note}: In addition to the set of operations described here, we
23068 expect the @sc{gui} implementation of a variable window to require, at
23069 least, the following operations:
23072 @item @code{-gdb-show} @code{output-radix}
23073 @item @code{-stack-list-arguments}
23074 @item @code{-stack-list-locals}
23075 @item @code{-stack-select-frame}
23080 @subheading Introduction to Variable Objects
23082 @cindex variable objects in @sc{gdb/mi}
23084 Variable objects are "object-oriented" MI interface for examining and
23085 changing values of expressions. Unlike some other MI interfaces that
23086 work with expressions, variable objects are specifically designed for
23087 simple and efficient presentation in the frontend. A variable object
23088 is identified by string name. When a variable object is created, the
23089 frontend specifies the expression for that variable object. The
23090 expression can be a simple variable, or it can be an arbitrary complex
23091 expression, and can even involve CPU registers. After creating a
23092 variable object, the frontend can invoke other variable object
23093 operations---for example to obtain or change the value of a variable
23094 object, or to change display format.
23096 Variable objects have hierarchical tree structure. Any variable object
23097 that corresponds to a composite type, such as structure in C, has
23098 a number of child variable objects, for example corresponding to each
23099 element of a structure. A child variable object can itself have
23100 children, recursively. Recursion ends when we reach
23101 leaf variable objects, which always have built-in types. Child variable
23102 objects are created only by explicit request, so if a frontend
23103 is not interested in the children of a particular variable object, no
23104 child will be created.
23106 For a leaf variable object it is possible to obtain its value as a
23107 string, or set the value from a string. String value can be also
23108 obtained for a non-leaf variable object, but it's generally a string
23109 that only indicates the type of the object, and does not list its
23110 contents. Assignment to a non-leaf variable object is not allowed.
23112 A frontend does not need to read the values of all variable objects each time
23113 the program stops. Instead, MI provides an update command that lists all
23114 variable objects whose values has changed since the last update
23115 operation. This considerably reduces the amount of data that must
23116 be transferred to the frontend. As noted above, children variable
23117 objects are created on demand, and only leaf variable objects have a
23118 real value. As result, gdb will read target memory only for leaf
23119 variables that frontend has created.
23121 The automatic update is not always desirable. For example, a frontend
23122 might want to keep a value of some expression for future reference,
23123 and never update it. For another example, fetching memory is
23124 relatively slow for embedded targets, so a frontend might want
23125 to disable automatic update for the variables that are either not
23126 visible on the screen, or ``closed''. This is possible using so
23127 called ``frozen variable objects''. Such variable objects are never
23128 implicitly updated.
23130 Variable objects can be either @dfn{fixed} or @dfn{floating}. For the
23131 fixed variable object, the expression is parsed when the variable
23132 object is created, including associating identifiers to specific
23133 variables. The meaning of expression never changes. For a floating
23134 variable object the values of variables whose names appear in the
23135 expressions are re-evaluated every time in the context of the current
23136 frame. Consider this example:
23141 struct work_state state;
23148 If a fixed variable object for the @code{state} variable is created in
23149 this function, and we enter the recursive call, the the variable
23150 object will report the value of @code{state} in the top-level
23151 @code{do_work} invocation. On the other hand, a floating variable
23152 object will report the value of @code{state} in the current frame.
23154 If an expression specified when creating a fixed variable object
23155 refers to a local variable, the variable object becomes bound to the
23156 thread and frame in which the variable object is created. When such
23157 variable object is updated, @value{GDBN} makes sure that the
23158 thread/frame combination the variable object is bound to still exists,
23159 and re-evaluates the variable object in context of that thread/frame.
23161 The following is the complete set of @sc{gdb/mi} operations defined to
23162 access this functionality:
23164 @multitable @columnfractions .4 .6
23165 @item @strong{Operation}
23166 @tab @strong{Description}
23168 @item @code{-var-create}
23169 @tab create a variable object
23170 @item @code{-var-delete}
23171 @tab delete the variable object and/or its children
23172 @item @code{-var-set-format}
23173 @tab set the display format of this variable
23174 @item @code{-var-show-format}
23175 @tab show the display format of this variable
23176 @item @code{-var-info-num-children}
23177 @tab tells how many children this object has
23178 @item @code{-var-list-children}
23179 @tab return a list of the object's children
23180 @item @code{-var-info-type}
23181 @tab show the type of this variable object
23182 @item @code{-var-info-expression}
23183 @tab print parent-relative expression that this variable object represents
23184 @item @code{-var-info-path-expression}
23185 @tab print full expression that this variable object represents
23186 @item @code{-var-show-attributes}
23187 @tab is this variable editable? does it exist here?
23188 @item @code{-var-evaluate-expression}
23189 @tab get the value of this variable
23190 @item @code{-var-assign}
23191 @tab set the value of this variable
23192 @item @code{-var-update}
23193 @tab update the variable and its children
23194 @item @code{-var-set-frozen}
23195 @tab set frozeness attribute
23198 In the next subsection we describe each operation in detail and suggest
23199 how it can be used.
23201 @subheading Description And Use of Operations on Variable Objects
23203 @subheading The @code{-var-create} Command
23204 @findex -var-create
23206 @subsubheading Synopsis
23209 -var-create @{@var{name} | "-"@}
23210 @{@var{frame-addr} | "*" | "@@"@} @var{expression}
23213 This operation creates a variable object, which allows the monitoring of
23214 a variable, the result of an expression, a memory cell or a CPU
23217 The @var{name} parameter is the string by which the object can be
23218 referenced. It must be unique. If @samp{-} is specified, the varobj
23219 system will generate a string ``varNNNNNN'' automatically. It will be
23220 unique provided that one does not specify @var{name} of that format.
23221 The command fails if a duplicate name is found.
23223 The frame under which the expression should be evaluated can be
23224 specified by @var{frame-addr}. A @samp{*} indicates that the current
23225 frame should be used. A @samp{@@} indicates that a floating variable
23226 object must be created.
23228 @var{expression} is any expression valid on the current language set (must not
23229 begin with a @samp{*}), or one of the following:
23233 @samp{*@var{addr}}, where @var{addr} is the address of a memory cell
23236 @samp{*@var{addr}-@var{addr}} --- a memory address range (TBD)
23239 @samp{$@var{regname}} --- a CPU register name
23242 @subsubheading Result
23244 This operation returns the name, number of children and the type of the
23245 object created. Type is returned as a string as the ones generated by
23246 the @value{GDBN} CLI. If a fixed variable object is bound to a
23247 specific thread, the thread is is also printed:
23250 name="@var{name}",numchild="@var{N}",type="@var{type}",thread-id="@var{M}"
23254 @subheading The @code{-var-delete} Command
23255 @findex -var-delete
23257 @subsubheading Synopsis
23260 -var-delete [ -c ] @var{name}
23263 Deletes a previously created variable object and all of its children.
23264 With the @samp{-c} option, just deletes the children.
23266 Returns an error if the object @var{name} is not found.
23269 @subheading The @code{-var-set-format} Command
23270 @findex -var-set-format
23272 @subsubheading Synopsis
23275 -var-set-format @var{name} @var{format-spec}
23278 Sets the output format for the value of the object @var{name} to be
23281 @anchor{-var-set-format}
23282 The syntax for the @var{format-spec} is as follows:
23285 @var{format-spec} @expansion{}
23286 @{binary | decimal | hexadecimal | octal | natural@}
23289 The natural format is the default format choosen automatically
23290 based on the variable type (like decimal for an @code{int}, hex
23291 for pointers, etc.).
23293 For a variable with children, the format is set only on the
23294 variable itself, and the children are not affected.
23296 @subheading The @code{-var-show-format} Command
23297 @findex -var-show-format
23299 @subsubheading Synopsis
23302 -var-show-format @var{name}
23305 Returns the format used to display the value of the object @var{name}.
23308 @var{format} @expansion{}
23313 @subheading The @code{-var-info-num-children} Command
23314 @findex -var-info-num-children
23316 @subsubheading Synopsis
23319 -var-info-num-children @var{name}
23322 Returns the number of children of a variable object @var{name}:
23329 @subheading The @code{-var-list-children} Command
23330 @findex -var-list-children
23332 @subsubheading Synopsis
23335 -var-list-children [@var{print-values}] @var{name}
23337 @anchor{-var-list-children}
23339 Return a list of the children of the specified variable object and
23340 create variable objects for them, if they do not already exist. With
23341 a single argument or if @var{print-values} has a value for of 0 or
23342 @code{--no-values}, print only the names of the variables; if
23343 @var{print-values} is 1 or @code{--all-values}, also print their
23344 values; and if it is 2 or @code{--simple-values} print the name and
23345 value for simple data types and just the name for arrays, structures
23348 For each child the following results are returned:
23353 Name of the variable object created for this child.
23356 The expression to be shown to the user by the front end to designate this child.
23357 For example this may be the name of a structure member.
23359 For C/C@t{++} structures there are several pseudo children returned to
23360 designate access qualifiers. For these pseudo children @var{exp} is
23361 @samp{public}, @samp{private}, or @samp{protected}. In this case the
23362 type and value are not present.
23365 Number of children this child has.
23368 The type of the child.
23371 If values were requested, this is the value.
23374 If this variable object is associated with a thread, this is the thread id.
23375 Otherwise this result is not present.
23378 If the variable object is frozen, this variable will be present with a value of 1.
23381 @subsubheading Example
23385 -var-list-children n
23386 ^done,numchild=@var{n},children=[child=@{name=@var{name},exp=@var{exp},
23387 numchild=@var{n},type=@var{type}@},@r{(repeats N times)}]
23389 -var-list-children --all-values n
23390 ^done,numchild=@var{n},children=[child=@{name=@var{name},exp=@var{exp},
23391 numchild=@var{n},value=@var{value},type=@var{type}@},@r{(repeats N times)}]
23395 @subheading The @code{-var-info-type} Command
23396 @findex -var-info-type
23398 @subsubheading Synopsis
23401 -var-info-type @var{name}
23404 Returns the type of the specified variable @var{name}. The type is
23405 returned as a string in the same format as it is output by the
23409 type=@var{typename}
23413 @subheading The @code{-var-info-expression} Command
23414 @findex -var-info-expression
23416 @subsubheading Synopsis
23419 -var-info-expression @var{name}
23422 Returns a string that is suitable for presenting this
23423 variable object in user interface. The string is generally
23424 not valid expression in the current language, and cannot be evaluated.
23426 For example, if @code{a} is an array, and variable object
23427 @code{A} was created for @code{a}, then we'll get this output:
23430 (gdb) -var-info-expression A.1
23431 ^done,lang="C",exp="1"
23435 Here, the values of @code{lang} can be @code{@{"C" | "C++" | "Java"@}}.
23437 Note that the output of the @code{-var-list-children} command also
23438 includes those expressions, so the @code{-var-info-expression} command
23441 @subheading The @code{-var-info-path-expression} Command
23442 @findex -var-info-path-expression
23444 @subsubheading Synopsis
23447 -var-info-path-expression @var{name}
23450 Returns an expression that can be evaluated in the current
23451 context and will yield the same value that a variable object has.
23452 Compare this with the @code{-var-info-expression} command, which
23453 result can be used only for UI presentation. Typical use of
23454 the @code{-var-info-path-expression} command is creating a
23455 watchpoint from a variable object.
23457 For example, suppose @code{C} is a C@t{++} class, derived from class
23458 @code{Base}, and that the @code{Base} class has a member called
23459 @code{m_size}. Assume a variable @code{c} is has the type of
23460 @code{C} and a variable object @code{C} was created for variable
23461 @code{c}. Then, we'll get this output:
23463 (gdb) -var-info-path-expression C.Base.public.m_size
23464 ^done,path_expr=((Base)c).m_size)
23467 @subheading The @code{-var-show-attributes} Command
23468 @findex -var-show-attributes
23470 @subsubheading Synopsis
23473 -var-show-attributes @var{name}
23476 List attributes of the specified variable object @var{name}:
23479 status=@var{attr} [ ( ,@var{attr} )* ]
23483 where @var{attr} is @code{@{ @{ editable | noneditable @} | TBD @}}.
23485 @subheading The @code{-var-evaluate-expression} Command
23486 @findex -var-evaluate-expression
23488 @subsubheading Synopsis
23491 -var-evaluate-expression [-f @var{format-spec}] @var{name}
23494 Evaluates the expression that is represented by the specified variable
23495 object and returns its value as a string. The format of the string
23496 can be specified with the @samp{-f} option. The possible values of
23497 this option are the same as for @code{-var-set-format}
23498 (@pxref{-var-set-format}). If the @samp{-f} option is not specified,
23499 the current display format will be used. The current display format
23500 can be changed using the @code{-var-set-format} command.
23506 Note that one must invoke @code{-var-list-children} for a variable
23507 before the value of a child variable can be evaluated.
23509 @subheading The @code{-var-assign} Command
23510 @findex -var-assign
23512 @subsubheading Synopsis
23515 -var-assign @var{name} @var{expression}
23518 Assigns the value of @var{expression} to the variable object specified
23519 by @var{name}. The object must be @samp{editable}. If the variable's
23520 value is altered by the assign, the variable will show up in any
23521 subsequent @code{-var-update} list.
23523 @subsubheading Example
23531 ^done,changelist=[@{name="var1",in_scope="true",type_changed="false"@}]
23535 @subheading The @code{-var-update} Command
23536 @findex -var-update
23538 @subsubheading Synopsis
23541 -var-update [@var{print-values}] @{@var{name} | "*"@}
23544 Reevaluate the expressions corresponding to the variable object
23545 @var{name} and all its direct and indirect children, and return the
23546 list of variable objects whose values have changed; @var{name} must
23547 be a root variable object. Here, ``changed'' means that the result of
23548 @code{-var-evaluate-expression} before and after the
23549 @code{-var-update} is different. If @samp{*} is used as the variable
23550 object names, all existing variable objects are updated, except
23551 for frozen ones (@pxref{-var-set-frozen}). The option
23552 @var{print-values} determines whether both names and values, or just
23553 names are printed. The possible values of this option are the same
23554 as for @code{-var-list-children} (@pxref{-var-list-children}). It is
23555 recommended to use the @samp{--all-values} option, to reduce the
23556 number of MI commands needed on each program stop.
23558 With the @samp{*} parameter, if a variable object is bound to a
23559 currently running thread, it will not be updated, without any
23562 @subsubheading Example
23569 -var-update --all-values var1
23570 ^done,changelist=[@{name="var1",value="3",in_scope="true",
23571 type_changed="false"@}]
23575 @anchor{-var-update}
23576 The field in_scope may take three values:
23580 The variable object's current value is valid.
23583 The variable object does not currently hold a valid value but it may
23584 hold one in the future if its associated expression comes back into
23588 The variable object no longer holds a valid value.
23589 This can occur when the executable file being debugged has changed,
23590 either through recompilation or by using the @value{GDBN} @code{file}
23591 command. The front end should normally choose to delete these variable
23595 In the future new values may be added to this list so the front should
23596 be prepared for this possibility. @xref{GDB/MI Development and Front Ends, ,@sc{GDB/MI} Development and Front Ends}.
23598 @subheading The @code{-var-set-frozen} Command
23599 @findex -var-set-frozen
23600 @anchor{-var-set-frozen}
23602 @subsubheading Synopsis
23605 -var-set-frozen @var{name} @var{flag}
23608 Set the frozenness flag on the variable object @var{name}. The
23609 @var{flag} parameter should be either @samp{1} to make the variable
23610 frozen or @samp{0} to make it unfrozen. If a variable object is
23611 frozen, then neither itself, nor any of its children, are
23612 implicitly updated by @code{-var-update} of
23613 a parent variable or by @code{-var-update *}. Only
23614 @code{-var-update} of the variable itself will update its value and
23615 values of its children. After a variable object is unfrozen, it is
23616 implicitly updated by all subsequent @code{-var-update} operations.
23617 Unfreezing a variable does not update it, only subsequent
23618 @code{-var-update} does.
23620 @subsubheading Example
23624 -var-set-frozen V 1
23629 @subheading The @code{-var-set-visualizer} command
23630 @findex -var-set-visualizer
23631 @anchor{-var-set-visualizer}
23633 @subsubheading Synopsis
23636 -var-set-visualizer @var{name} @var{visualizer}
23639 Set a visualizer for the variable object @var{name}.
23641 @var{visualizer} is the visualizer to use. The special value
23642 @samp{None} means to disable any visualizer in use.
23644 If not @samp{None}, @var{visualizer} must be a Python expression.
23645 This expression must evaluate to a callable object which accepts a
23646 single argument. @value{GDBN} will call this object with the value of
23647 the varobj @var{name} as an argument (this is done so that the same
23648 Python pretty-printing code can be used for both the CLI and MI).
23649 When called, this object must return an object which conforms to the
23650 pretty-printing interface (@pxref{Pretty Printing}).
23652 The pre-defined function @code{gdb.default_visualizer} may be used to
23653 select a visualizer by following the built-in process
23654 (@pxref{Selecting Pretty-Printers}). This is done automatically when
23655 a varobj is created, and so ordinarily is not needed.
23657 This feature is only available if Python support is enabled. The MI
23658 command @code{-list-features} (@pxref{GDB/MI Miscellaneous Commands})
23659 can be used to check this.
23661 @subsubheading Example
23663 Resetting the visualizer:
23667 -var-set-visualizer V None
23671 Reselecting the default (type-based) visualizer:
23675 -var-set-visualizer V gdb.default_visualizer
23679 Suppose @code{SomeClass} is a visualizer class. A lambda expression
23680 can be used to instantiate this class for a varobj:
23684 -var-set-visualizer V "lambda val: SomeClass()"
23688 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
23689 @node GDB/MI Data Manipulation
23690 @section @sc{gdb/mi} Data Manipulation
23692 @cindex data manipulation, in @sc{gdb/mi}
23693 @cindex @sc{gdb/mi}, data manipulation
23694 This section describes the @sc{gdb/mi} commands that manipulate data:
23695 examine memory and registers, evaluate expressions, etc.
23697 @c REMOVED FROM THE INTERFACE.
23698 @c @subheading -data-assign
23699 @c Change the value of a program variable. Plenty of side effects.
23700 @c @subsubheading GDB Command
23702 @c @subsubheading Example
23705 @subheading The @code{-data-disassemble} Command
23706 @findex -data-disassemble
23708 @subsubheading Synopsis
23712 [ -s @var{start-addr} -e @var{end-addr} ]
23713 | [ -f @var{filename} -l @var{linenum} [ -n @var{lines} ] ]
23721 @item @var{start-addr}
23722 is the beginning address (or @code{$pc})
23723 @item @var{end-addr}
23725 @item @var{filename}
23726 is the name of the file to disassemble
23727 @item @var{linenum}
23728 is the line number to disassemble around
23730 is the number of disassembly lines to be produced. If it is -1,
23731 the whole function will be disassembled, in case no @var{end-addr} is
23732 specified. If @var{end-addr} is specified as a non-zero value, and
23733 @var{lines} is lower than the number of disassembly lines between
23734 @var{start-addr} and @var{end-addr}, only @var{lines} lines are
23735 displayed; if @var{lines} is higher than the number of lines between
23736 @var{start-addr} and @var{end-addr}, only the lines up to @var{end-addr}
23739 is either 0 (meaning only disassembly) or 1 (meaning mixed source and
23743 @subsubheading Result
23745 The output for each instruction is composed of four fields:
23754 Note that whatever included in the instruction field, is not manipulated
23755 directly by @sc{gdb/mi}, i.e., it is not possible to adjust its format.
23757 @subsubheading @value{GDBN} Command
23759 There's no direct mapping from this command to the CLI.
23761 @subsubheading Example
23763 Disassemble from the current value of @code{$pc} to @code{$pc + 20}:
23767 -data-disassemble -s $pc -e "$pc + 20" -- 0
23770 @{address="0x000107c0",func-name="main",offset="4",
23771 inst="mov 2, %o0"@},
23772 @{address="0x000107c4",func-name="main",offset="8",
23773 inst="sethi %hi(0x11800), %o2"@},
23774 @{address="0x000107c8",func-name="main",offset="12",
23775 inst="or %o2, 0x140, %o1\t! 0x11940 <_lib_version+8>"@},
23776 @{address="0x000107cc",func-name="main",offset="16",
23777 inst="sethi %hi(0x11800), %o2"@},
23778 @{address="0x000107d0",func-name="main",offset="20",
23779 inst="or %o2, 0x168, %o4\t! 0x11968 <_lib_version+48>"@}]
23783 Disassemble the whole @code{main} function. Line 32 is part of
23787 -data-disassemble -f basics.c -l 32 -- 0
23789 @{address="0x000107bc",func-name="main",offset="0",
23790 inst="save %sp, -112, %sp"@},
23791 @{address="0x000107c0",func-name="main",offset="4",
23792 inst="mov 2, %o0"@},
23793 @{address="0x000107c4",func-name="main",offset="8",
23794 inst="sethi %hi(0x11800), %o2"@},
23796 @{address="0x0001081c",func-name="main",offset="96",inst="ret "@},
23797 @{address="0x00010820",func-name="main",offset="100",inst="restore "@}]
23801 Disassemble 3 instructions from the start of @code{main}:
23805 -data-disassemble -f basics.c -l 32 -n 3 -- 0
23807 @{address="0x000107bc",func-name="main",offset="0",
23808 inst="save %sp, -112, %sp"@},
23809 @{address="0x000107c0",func-name="main",offset="4",
23810 inst="mov 2, %o0"@},
23811 @{address="0x000107c4",func-name="main",offset="8",
23812 inst="sethi %hi(0x11800), %o2"@}]
23816 Disassemble 3 instructions from the start of @code{main} in mixed mode:
23820 -data-disassemble -f basics.c -l 32 -n 3 -- 1
23822 src_and_asm_line=@{line="31",
23823 file="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb/ \
23824 testsuite/gdb.mi/basics.c",line_asm_insn=[
23825 @{address="0x000107bc",func-name="main",offset="0",
23826 inst="save %sp, -112, %sp"@}]@},
23827 src_and_asm_line=@{line="32",
23828 file="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb/ \
23829 testsuite/gdb.mi/basics.c",line_asm_insn=[
23830 @{address="0x000107c0",func-name="main",offset="4",
23831 inst="mov 2, %o0"@},
23832 @{address="0x000107c4",func-name="main",offset="8",
23833 inst="sethi %hi(0x11800), %o2"@}]@}]
23838 @subheading The @code{-data-evaluate-expression} Command
23839 @findex -data-evaluate-expression
23841 @subsubheading Synopsis
23844 -data-evaluate-expression @var{expr}
23847 Evaluate @var{expr} as an expression. The expression could contain an
23848 inferior function call. The function call will execute synchronously.
23849 If the expression contains spaces, it must be enclosed in double quotes.
23851 @subsubheading @value{GDBN} Command
23853 The corresponding @value{GDBN} commands are @samp{print}, @samp{output}, and
23854 @samp{call}. In @code{gdbtk} only, there's a corresponding
23855 @samp{gdb_eval} command.
23857 @subsubheading Example
23859 In the following example, the numbers that precede the commands are the
23860 @dfn{tokens} described in @ref{GDB/MI Command Syntax, ,@sc{gdb/mi}
23861 Command Syntax}. Notice how @sc{gdb/mi} returns the same tokens in its
23865 211-data-evaluate-expression A
23868 311-data-evaluate-expression &A
23869 311^done,value="0xefffeb7c"
23871 411-data-evaluate-expression A+3
23874 511-data-evaluate-expression "A + 3"
23880 @subheading The @code{-data-list-changed-registers} Command
23881 @findex -data-list-changed-registers
23883 @subsubheading Synopsis
23886 -data-list-changed-registers
23889 Display a list of the registers that have changed.
23891 @subsubheading @value{GDBN} Command
23893 @value{GDBN} doesn't have a direct analog for this command; @code{gdbtk}
23894 has the corresponding command @samp{gdb_changed_register_list}.
23896 @subsubheading Example
23898 On a PPC MBX board:
23906 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",frame=@{
23907 func="main",args=[],file="try.c",fullname="/home/foo/bar/try.c",
23910 -data-list-changed-registers
23911 ^done,changed-registers=["0","1","2","4","5","6","7","8","9",
23912 "10","11","13","14","15","16","17","18","19","20","21","22","23",
23913 "24","25","26","27","28","30","31","64","65","66","67","69"]
23918 @subheading The @code{-data-list-register-names} Command
23919 @findex -data-list-register-names
23921 @subsubheading Synopsis
23924 -data-list-register-names [ ( @var{regno} )+ ]
23927 Show a list of register names for the current target. If no arguments
23928 are given, it shows a list of the names of all the registers. If
23929 integer numbers are given as arguments, it will print a list of the
23930 names of the registers corresponding to the arguments. To ensure
23931 consistency between a register name and its number, the output list may
23932 include empty register names.
23934 @subsubheading @value{GDBN} Command
23936 @value{GDBN} does not have a command which corresponds to
23937 @samp{-data-list-register-names}. In @code{gdbtk} there is a
23938 corresponding command @samp{gdb_regnames}.
23940 @subsubheading Example
23942 For the PPC MBX board:
23945 -data-list-register-names
23946 ^done,register-names=["r0","r1","r2","r3","r4","r5","r6","r7",
23947 "r8","r9","r10","r11","r12","r13","r14","r15","r16","r17","r18",
23948 "r19","r20","r21","r22","r23","r24","r25","r26","r27","r28","r29",
23949 "r30","r31","f0","f1","f2","f3","f4","f5","f6","f7","f8","f9",
23950 "f10","f11","f12","f13","f14","f15","f16","f17","f18","f19","f20",
23951 "f21","f22","f23","f24","f25","f26","f27","f28","f29","f30","f31",
23952 "", "pc","ps","cr","lr","ctr","xer"]
23954 -data-list-register-names 1 2 3
23955 ^done,register-names=["r1","r2","r3"]
23959 @subheading The @code{-data-list-register-values} Command
23960 @findex -data-list-register-values
23962 @subsubheading Synopsis
23965 -data-list-register-values @var{fmt} [ ( @var{regno} )*]
23968 Display the registers' contents. @var{fmt} is the format according to
23969 which the registers' contents are to be returned, followed by an optional
23970 list of numbers specifying the registers to display. A missing list of
23971 numbers indicates that the contents of all the registers must be returned.
23973 Allowed formats for @var{fmt} are:
23990 @subsubheading @value{GDBN} Command
23992 The corresponding @value{GDBN} commands are @samp{info reg}, @samp{info
23993 all-reg}, and (in @code{gdbtk}) @samp{gdb_fetch_registers}.
23995 @subsubheading Example
23997 For a PPC MBX board (note: line breaks are for readability only, they
23998 don't appear in the actual output):
24002 -data-list-register-values r 64 65
24003 ^done,register-values=[@{number="64",value="0xfe00a300"@},
24004 @{number="65",value="0x00029002"@}]
24006 -data-list-register-values x
24007 ^done,register-values=[@{number="0",value="0xfe0043c8"@},
24008 @{number="1",value="0x3fff88"@},@{number="2",value="0xfffffffe"@},
24009 @{number="3",value="0x0"@},@{number="4",value="0xa"@},
24010 @{number="5",value="0x3fff68"@},@{number="6",value="0x3fff58"@},
24011 @{number="7",value="0xfe011e98"@},@{number="8",value="0x2"@},
24012 @{number="9",value="0xfa202820"@},@{number="10",value="0xfa202808"@},
24013 @{number="11",value="0x1"@},@{number="12",value="0x0"@},
24014 @{number="13",value="0x4544"@},@{number="14",value="0xffdfffff"@},
24015 @{number="15",value="0xffffffff"@},@{number="16",value="0xfffffeff"@},
24016 @{number="17",value="0xefffffed"@},@{number="18",value="0xfffffffe"@},
24017 @{number="19",value="0xffffffff"@},@{number="20",value="0xffffffff"@},
24018 @{number="21",value="0xffffffff"@},@{number="22",value="0xfffffff7"@},
24019 @{number="23",value="0xffffffff"@},@{number="24",value="0xffffffff"@},
24020 @{number="25",value="0xffffffff"@},@{number="26",value="0xfffffffb"@},
24021 @{number="27",value="0xffffffff"@},@{number="28",value="0xf7bfffff"@},
24022 @{number="29",value="0x0"@},@{number="30",value="0xfe010000"@},
24023 @{number="31",value="0x0"@},@{number="32",value="0x0"@},
24024 @{number="33",value="0x0"@},@{number="34",value="0x0"@},
24025 @{number="35",value="0x0"@},@{number="36",value="0x0"@},
24026 @{number="37",value="0x0"@},@{number="38",value="0x0"@},
24027 @{number="39",value="0x0"@},@{number="40",value="0x0"@},
24028 @{number="41",value="0x0"@},@{number="42",value="0x0"@},
24029 @{number="43",value="0x0"@},@{number="44",value="0x0"@},
24030 @{number="45",value="0x0"@},@{number="46",value="0x0"@},
24031 @{number="47",value="0x0"@},@{number="48",value="0x0"@},
24032 @{number="49",value="0x0"@},@{number="50",value="0x0"@},
24033 @{number="51",value="0x0"@},@{number="52",value="0x0"@},
24034 @{number="53",value="0x0"@},@{number="54",value="0x0"@},
24035 @{number="55",value="0x0"@},@{number="56",value="0x0"@},
24036 @{number="57",value="0x0"@},@{number="58",value="0x0"@},
24037 @{number="59",value="0x0"@},@{number="60",value="0x0"@},
24038 @{number="61",value="0x0"@},@{number="62",value="0x0"@},
24039 @{number="63",value="0x0"@},@{number="64",value="0xfe00a300"@},
24040 @{number="65",value="0x29002"@},@{number="66",value="0x202f04b5"@},
24041 @{number="67",value="0xfe0043b0"@},@{number="68",value="0xfe00b3e4"@},
24042 @{number="69",value="0x20002b03"@}]
24047 @subheading The @code{-data-read-memory} Command
24048 @findex -data-read-memory
24050 @subsubheading Synopsis
24053 -data-read-memory [ -o @var{byte-offset} ]
24054 @var{address} @var{word-format} @var{word-size}
24055 @var{nr-rows} @var{nr-cols} [ @var{aschar} ]
24062 @item @var{address}
24063 An expression specifying the address of the first memory word to be
24064 read. Complex expressions containing embedded white space should be
24065 quoted using the C convention.
24067 @item @var{word-format}
24068 The format to be used to print the memory words. The notation is the
24069 same as for @value{GDBN}'s @code{print} command (@pxref{Output Formats,
24072 @item @var{word-size}
24073 The size of each memory word in bytes.
24075 @item @var{nr-rows}
24076 The number of rows in the output table.
24078 @item @var{nr-cols}
24079 The number of columns in the output table.
24082 If present, indicates that each row should include an @sc{ascii} dump. The
24083 value of @var{aschar} is used as a padding character when a byte is not a
24084 member of the printable @sc{ascii} character set (printable @sc{ascii}
24085 characters are those whose code is between 32 and 126, inclusively).
24087 @item @var{byte-offset}
24088 An offset to add to the @var{address} before fetching memory.
24091 This command displays memory contents as a table of @var{nr-rows} by
24092 @var{nr-cols} words, each word being @var{word-size} bytes. In total,
24093 @code{@var{nr-rows} * @var{nr-cols} * @var{word-size}} bytes are read
24094 (returned as @samp{total-bytes}). Should less than the requested number
24095 of bytes be returned by the target, the missing words are identified
24096 using @samp{N/A}. The number of bytes read from the target is returned
24097 in @samp{nr-bytes} and the starting address used to read memory in
24100 The address of the next/previous row or page is available in
24101 @samp{next-row} and @samp{prev-row}, @samp{next-page} and
24104 @subsubheading @value{GDBN} Command
24106 The corresponding @value{GDBN} command is @samp{x}. @code{gdbtk} has
24107 @samp{gdb_get_mem} memory read command.
24109 @subsubheading Example
24111 Read six bytes of memory starting at @code{bytes+6} but then offset by
24112 @code{-6} bytes. Format as three rows of two columns. One byte per
24113 word. Display each word in hex.
24117 9-data-read-memory -o -6 -- bytes+6 x 1 3 2
24118 9^done,addr="0x00001390",nr-bytes="6",total-bytes="6",
24119 next-row="0x00001396",prev-row="0x0000138e",next-page="0x00001396",
24120 prev-page="0x0000138a",memory=[
24121 @{addr="0x00001390",data=["0x00","0x01"]@},
24122 @{addr="0x00001392",data=["0x02","0x03"]@},
24123 @{addr="0x00001394",data=["0x04","0x05"]@}]
24127 Read two bytes of memory starting at address @code{shorts + 64} and
24128 display as a single word formatted in decimal.
24132 5-data-read-memory shorts+64 d 2 1 1
24133 5^done,addr="0x00001510",nr-bytes="2",total-bytes="2",
24134 next-row="0x00001512",prev-row="0x0000150e",
24135 next-page="0x00001512",prev-page="0x0000150e",memory=[
24136 @{addr="0x00001510",data=["128"]@}]
24140 Read thirty two bytes of memory starting at @code{bytes+16} and format
24141 as eight rows of four columns. Include a string encoding with @samp{x}
24142 used as the non-printable character.
24146 4-data-read-memory bytes+16 x 1 8 4 x
24147 4^done,addr="0x000013a0",nr-bytes="32",total-bytes="32",
24148 next-row="0x000013c0",prev-row="0x0000139c",
24149 next-page="0x000013c0",prev-page="0x00001380",memory=[
24150 @{addr="0x000013a0",data=["0x10","0x11","0x12","0x13"],ascii="xxxx"@},
24151 @{addr="0x000013a4",data=["0x14","0x15","0x16","0x17"],ascii="xxxx"@},
24152 @{addr="0x000013a8",data=["0x18","0x19","0x1a","0x1b"],ascii="xxxx"@},
24153 @{addr="0x000013ac",data=["0x1c","0x1d","0x1e","0x1f"],ascii="xxxx"@},
24154 @{addr="0x000013b0",data=["0x20","0x21","0x22","0x23"],ascii=" !\"#"@},
24155 @{addr="0x000013b4",data=["0x24","0x25","0x26","0x27"],ascii="$%&'"@},
24156 @{addr="0x000013b8",data=["0x28","0x29","0x2a","0x2b"],ascii="()*+"@},
24157 @{addr="0x000013bc",data=["0x2c","0x2d","0x2e","0x2f"],ascii=",-./"@}]
24161 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
24162 @node GDB/MI Tracepoint Commands
24163 @section @sc{gdb/mi} Tracepoint Commands
24165 The tracepoint commands are not yet implemented.
24167 @c @subheading -trace-actions
24169 @c @subheading -trace-delete
24171 @c @subheading -trace-disable
24173 @c @subheading -trace-dump
24175 @c @subheading -trace-enable
24177 @c @subheading -trace-exists
24179 @c @subheading -trace-find
24181 @c @subheading -trace-frame-number
24183 @c @subheading -trace-info
24185 @c @subheading -trace-insert
24187 @c @subheading -trace-list
24189 @c @subheading -trace-pass-count
24191 @c @subheading -trace-save
24193 @c @subheading -trace-start
24195 @c @subheading -trace-stop
24198 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
24199 @node GDB/MI Symbol Query
24200 @section @sc{gdb/mi} Symbol Query Commands
24204 @subheading The @code{-symbol-info-address} Command
24205 @findex -symbol-info-address
24207 @subsubheading Synopsis
24210 -symbol-info-address @var{symbol}
24213 Describe where @var{symbol} is stored.
24215 @subsubheading @value{GDBN} Command
24217 The corresponding @value{GDBN} command is @samp{info address}.
24219 @subsubheading Example
24223 @subheading The @code{-symbol-info-file} Command
24224 @findex -symbol-info-file
24226 @subsubheading Synopsis
24232 Show the file for the symbol.
24234 @subsubheading @value{GDBN} Command
24236 There's no equivalent @value{GDBN} command. @code{gdbtk} has
24237 @samp{gdb_find_file}.
24239 @subsubheading Example
24243 @subheading The @code{-symbol-info-function} Command
24244 @findex -symbol-info-function
24246 @subsubheading Synopsis
24249 -symbol-info-function
24252 Show which function the symbol lives in.
24254 @subsubheading @value{GDBN} Command
24256 @samp{gdb_get_function} in @code{gdbtk}.
24258 @subsubheading Example
24262 @subheading The @code{-symbol-info-line} Command
24263 @findex -symbol-info-line
24265 @subsubheading Synopsis
24271 Show the core addresses of the code for a source line.
24273 @subsubheading @value{GDBN} Command
24275 The corresponding @value{GDBN} command is @samp{info line}.
24276 @code{gdbtk} has the @samp{gdb_get_line} and @samp{gdb_get_file} commands.
24278 @subsubheading Example
24282 @subheading The @code{-symbol-info-symbol} Command
24283 @findex -symbol-info-symbol
24285 @subsubheading Synopsis
24288 -symbol-info-symbol @var{addr}
24291 Describe what symbol is at location @var{addr}.
24293 @subsubheading @value{GDBN} Command
24295 The corresponding @value{GDBN} command is @samp{info symbol}.
24297 @subsubheading Example
24301 @subheading The @code{-symbol-list-functions} Command
24302 @findex -symbol-list-functions
24304 @subsubheading Synopsis
24307 -symbol-list-functions
24310 List the functions in the executable.
24312 @subsubheading @value{GDBN} Command
24314 @samp{info functions} in @value{GDBN}, @samp{gdb_listfunc} and
24315 @samp{gdb_search} in @code{gdbtk}.
24317 @subsubheading Example
24322 @subheading The @code{-symbol-list-lines} Command
24323 @findex -symbol-list-lines
24325 @subsubheading Synopsis
24328 -symbol-list-lines @var{filename}
24331 Print the list of lines that contain code and their associated program
24332 addresses for the given source filename. The entries are sorted in
24333 ascending PC order.
24335 @subsubheading @value{GDBN} Command
24337 There is no corresponding @value{GDBN} command.
24339 @subsubheading Example
24342 -symbol-list-lines basics.c
24343 ^done,lines=[@{pc="0x08048554",line="7"@},@{pc="0x0804855a",line="8"@}]
24349 @subheading The @code{-symbol-list-types} Command
24350 @findex -symbol-list-types
24352 @subsubheading Synopsis
24358 List all the type names.
24360 @subsubheading @value{GDBN} Command
24362 The corresponding commands are @samp{info types} in @value{GDBN},
24363 @samp{gdb_search} in @code{gdbtk}.
24365 @subsubheading Example
24369 @subheading The @code{-symbol-list-variables} Command
24370 @findex -symbol-list-variables
24372 @subsubheading Synopsis
24375 -symbol-list-variables
24378 List all the global and static variable names.
24380 @subsubheading @value{GDBN} Command
24382 @samp{info variables} in @value{GDBN}, @samp{gdb_search} in @code{gdbtk}.
24384 @subsubheading Example
24388 @subheading The @code{-symbol-locate} Command
24389 @findex -symbol-locate
24391 @subsubheading Synopsis
24397 @subsubheading @value{GDBN} Command
24399 @samp{gdb_loc} in @code{gdbtk}.
24401 @subsubheading Example
24405 @subheading The @code{-symbol-type} Command
24406 @findex -symbol-type
24408 @subsubheading Synopsis
24411 -symbol-type @var{variable}
24414 Show type of @var{variable}.
24416 @subsubheading @value{GDBN} Command
24418 The corresponding @value{GDBN} command is @samp{ptype}, @code{gdbtk} has
24419 @samp{gdb_obj_variable}.
24421 @subsubheading Example
24426 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
24427 @node GDB/MI File Commands
24428 @section @sc{gdb/mi} File Commands
24430 This section describes the GDB/MI commands to specify executable file names
24431 and to read in and obtain symbol table information.
24433 @subheading The @code{-file-exec-and-symbols} Command
24434 @findex -file-exec-and-symbols
24436 @subsubheading Synopsis
24439 -file-exec-and-symbols @var{file}
24442 Specify the executable file to be debugged. This file is the one from
24443 which the symbol table is also read. If no file is specified, the
24444 command clears the executable and symbol information. If breakpoints
24445 are set when using this command with no arguments, @value{GDBN} will produce
24446 error messages. Otherwise, no output is produced, except a completion
24449 @subsubheading @value{GDBN} Command
24451 The corresponding @value{GDBN} command is @samp{file}.
24453 @subsubheading Example
24457 -file-exec-and-symbols /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
24463 @subheading The @code{-file-exec-file} Command
24464 @findex -file-exec-file
24466 @subsubheading Synopsis
24469 -file-exec-file @var{file}
24472 Specify the executable file to be debugged. Unlike
24473 @samp{-file-exec-and-symbols}, the symbol table is @emph{not} read
24474 from this file. If used without argument, @value{GDBN} clears the information
24475 about the executable file. No output is produced, except a completion
24478 @subsubheading @value{GDBN} Command
24480 The corresponding @value{GDBN} command is @samp{exec-file}.
24482 @subsubheading Example
24486 -file-exec-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
24493 @subheading The @code{-file-list-exec-sections} Command
24494 @findex -file-list-exec-sections
24496 @subsubheading Synopsis
24499 -file-list-exec-sections
24502 List the sections of the current executable file.
24504 @subsubheading @value{GDBN} Command
24506 The @value{GDBN} command @samp{info file} shows, among the rest, the same
24507 information as this command. @code{gdbtk} has a corresponding command
24508 @samp{gdb_load_info}.
24510 @subsubheading Example
24515 @subheading The @code{-file-list-exec-source-file} Command
24516 @findex -file-list-exec-source-file
24518 @subsubheading Synopsis
24521 -file-list-exec-source-file
24524 List the line number, the current source file, and the absolute path
24525 to the current source file for the current executable. The macro
24526 information field has a value of @samp{1} or @samp{0} depending on
24527 whether or not the file includes preprocessor macro information.
24529 @subsubheading @value{GDBN} Command
24531 The @value{GDBN} equivalent is @samp{info source}
24533 @subsubheading Example
24537 123-file-list-exec-source-file
24538 123^done,line="1",file="foo.c",fullname="/home/bar/foo.c,macro-info="1"
24543 @subheading The @code{-file-list-exec-source-files} Command
24544 @findex -file-list-exec-source-files
24546 @subsubheading Synopsis
24549 -file-list-exec-source-files
24552 List the source files for the current executable.
24554 It will always output the filename, but only when @value{GDBN} can find
24555 the absolute file name of a source file, will it output the fullname.
24557 @subsubheading @value{GDBN} Command
24559 The @value{GDBN} equivalent is @samp{info sources}.
24560 @code{gdbtk} has an analogous command @samp{gdb_listfiles}.
24562 @subsubheading Example
24565 -file-list-exec-source-files
24567 @{file=foo.c,fullname=/home/foo.c@},
24568 @{file=/home/bar.c,fullname=/home/bar.c@},
24569 @{file=gdb_could_not_find_fullpath.c@}]
24574 @subheading The @code{-file-list-shared-libraries} Command
24575 @findex -file-list-shared-libraries
24577 @subsubheading Synopsis
24580 -file-list-shared-libraries
24583 List the shared libraries in the program.
24585 @subsubheading @value{GDBN} Command
24587 The corresponding @value{GDBN} command is @samp{info shared}.
24589 @subsubheading Example
24593 @subheading The @code{-file-list-symbol-files} Command
24594 @findex -file-list-symbol-files
24596 @subsubheading Synopsis
24599 -file-list-symbol-files
24604 @subsubheading @value{GDBN} Command
24606 The corresponding @value{GDBN} command is @samp{info file} (part of it).
24608 @subsubheading Example
24613 @subheading The @code{-file-symbol-file} Command
24614 @findex -file-symbol-file
24616 @subsubheading Synopsis
24619 -file-symbol-file @var{file}
24622 Read symbol table info from the specified @var{file} argument. When
24623 used without arguments, clears @value{GDBN}'s symbol table info. No output is
24624 produced, except for a completion notification.
24626 @subsubheading @value{GDBN} Command
24628 The corresponding @value{GDBN} command is @samp{symbol-file}.
24630 @subsubheading Example
24634 -file-symbol-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
24640 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
24641 @node GDB/MI Memory Overlay Commands
24642 @section @sc{gdb/mi} Memory Overlay Commands
24644 The memory overlay commands are not implemented.
24646 @c @subheading -overlay-auto
24648 @c @subheading -overlay-list-mapping-state
24650 @c @subheading -overlay-list-overlays
24652 @c @subheading -overlay-map
24654 @c @subheading -overlay-off
24656 @c @subheading -overlay-on
24658 @c @subheading -overlay-unmap
24660 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
24661 @node GDB/MI Signal Handling Commands
24662 @section @sc{gdb/mi} Signal Handling Commands
24664 Signal handling commands are not implemented.
24666 @c @subheading -signal-handle
24668 @c @subheading -signal-list-handle-actions
24670 @c @subheading -signal-list-signal-types
24674 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
24675 @node GDB/MI Target Manipulation
24676 @section @sc{gdb/mi} Target Manipulation Commands
24679 @subheading The @code{-target-attach} Command
24680 @findex -target-attach
24682 @subsubheading Synopsis
24685 -target-attach @var{pid} | @var{gid} | @var{file}
24688 Attach to a process @var{pid} or a file @var{file} outside of
24689 @value{GDBN}, or a thread group @var{gid}. If attaching to a thread
24690 group, the id previously returned by
24691 @samp{-list-thread-groups --available} must be used.
24693 @subsubheading @value{GDBN} Command
24695 The corresponding @value{GDBN} command is @samp{attach}.
24697 @subsubheading Example
24701 =thread-created,id="1"
24702 *stopped,thread-id="1",frame=@{addr="0xb7f7e410",func="bar",args=[]@}
24708 @subheading The @code{-target-compare-sections} Command
24709 @findex -target-compare-sections
24711 @subsubheading Synopsis
24714 -target-compare-sections [ @var{section} ]
24717 Compare data of section @var{section} on target to the exec file.
24718 Without the argument, all sections are compared.
24720 @subsubheading @value{GDBN} Command
24722 The @value{GDBN} equivalent is @samp{compare-sections}.
24724 @subsubheading Example
24729 @subheading The @code{-target-detach} Command
24730 @findex -target-detach
24732 @subsubheading Synopsis
24735 -target-detach [ @var{pid} | @var{gid} ]
24738 Detach from the remote target which normally resumes its execution.
24739 If either @var{pid} or @var{gid} is specified, detaches from either
24740 the specified process, or specified thread group. There's no output.
24742 @subsubheading @value{GDBN} Command
24744 The corresponding @value{GDBN} command is @samp{detach}.
24746 @subsubheading Example
24756 @subheading The @code{-target-disconnect} Command
24757 @findex -target-disconnect
24759 @subsubheading Synopsis
24765 Disconnect from the remote target. There's no output and the target is
24766 generally not resumed.
24768 @subsubheading @value{GDBN} Command
24770 The corresponding @value{GDBN} command is @samp{disconnect}.
24772 @subsubheading Example
24782 @subheading The @code{-target-download} Command
24783 @findex -target-download
24785 @subsubheading Synopsis
24791 Loads the executable onto the remote target.
24792 It prints out an update message every half second, which includes the fields:
24796 The name of the section.
24798 The size of what has been sent so far for that section.
24800 The size of the section.
24802 The total size of what was sent so far (the current and the previous sections).
24804 The size of the overall executable to download.
24808 Each message is sent as status record (@pxref{GDB/MI Output Syntax, ,
24809 @sc{gdb/mi} Output Syntax}).
24811 In addition, it prints the name and size of the sections, as they are
24812 downloaded. These messages include the following fields:
24816 The name of the section.
24818 The size of the section.
24820 The size of the overall executable to download.
24824 At the end, a summary is printed.
24826 @subsubheading @value{GDBN} Command
24828 The corresponding @value{GDBN} command is @samp{load}.
24830 @subsubheading Example
24832 Note: each status message appears on a single line. Here the messages
24833 have been broken down so that they can fit onto a page.
24838 +download,@{section=".text",section-size="6668",total-size="9880"@}
24839 +download,@{section=".text",section-sent="512",section-size="6668",
24840 total-sent="512",total-size="9880"@}
24841 +download,@{section=".text",section-sent="1024",section-size="6668",
24842 total-sent="1024",total-size="9880"@}
24843 +download,@{section=".text",section-sent="1536",section-size="6668",
24844 total-sent="1536",total-size="9880"@}
24845 +download,@{section=".text",section-sent="2048",section-size="6668",
24846 total-sent="2048",total-size="9880"@}
24847 +download,@{section=".text",section-sent="2560",section-size="6668",
24848 total-sent="2560",total-size="9880"@}
24849 +download,@{section=".text",section-sent="3072",section-size="6668",
24850 total-sent="3072",total-size="9880"@}
24851 +download,@{section=".text",section-sent="3584",section-size="6668",
24852 total-sent="3584",total-size="9880"@}
24853 +download,@{section=".text",section-sent="4096",section-size="6668",
24854 total-sent="4096",total-size="9880"@}
24855 +download,@{section=".text",section-sent="4608",section-size="6668",
24856 total-sent="4608",total-size="9880"@}
24857 +download,@{section=".text",section-sent="5120",section-size="6668",
24858 total-sent="5120",total-size="9880"@}
24859 +download,@{section=".text",section-sent="5632",section-size="6668",
24860 total-sent="5632",total-size="9880"@}
24861 +download,@{section=".text",section-sent="6144",section-size="6668",
24862 total-sent="6144",total-size="9880"@}
24863 +download,@{section=".text",section-sent="6656",section-size="6668",
24864 total-sent="6656",total-size="9880"@}
24865 +download,@{section=".init",section-size="28",total-size="9880"@}
24866 +download,@{section=".fini",section-size="28",total-size="9880"@}
24867 +download,@{section=".data",section-size="3156",total-size="9880"@}
24868 +download,@{section=".data",section-sent="512",section-size="3156",
24869 total-sent="7236",total-size="9880"@}
24870 +download,@{section=".data",section-sent="1024",section-size="3156",
24871 total-sent="7748",total-size="9880"@}
24872 +download,@{section=".data",section-sent="1536",section-size="3156",
24873 total-sent="8260",total-size="9880"@}
24874 +download,@{section=".data",section-sent="2048",section-size="3156",
24875 total-sent="8772",total-size="9880"@}
24876 +download,@{section=".data",section-sent="2560",section-size="3156",
24877 total-sent="9284",total-size="9880"@}
24878 +download,@{section=".data",section-sent="3072",section-size="3156",
24879 total-sent="9796",total-size="9880"@}
24880 ^done,address="0x10004",load-size="9880",transfer-rate="6586",
24887 @subheading The @code{-target-exec-status} Command
24888 @findex -target-exec-status
24890 @subsubheading Synopsis
24893 -target-exec-status
24896 Provide information on the state of the target (whether it is running or
24897 not, for instance).
24899 @subsubheading @value{GDBN} Command
24901 There's no equivalent @value{GDBN} command.
24903 @subsubheading Example
24907 @subheading The @code{-target-list-available-targets} Command
24908 @findex -target-list-available-targets
24910 @subsubheading Synopsis
24913 -target-list-available-targets
24916 List the possible targets to connect to.
24918 @subsubheading @value{GDBN} Command
24920 The corresponding @value{GDBN} command is @samp{help target}.
24922 @subsubheading Example
24926 @subheading The @code{-target-list-current-targets} Command
24927 @findex -target-list-current-targets
24929 @subsubheading Synopsis
24932 -target-list-current-targets
24935 Describe the current target.
24937 @subsubheading @value{GDBN} Command
24939 The corresponding information is printed by @samp{info file} (among
24942 @subsubheading Example
24946 @subheading The @code{-target-list-parameters} Command
24947 @findex -target-list-parameters
24949 @subsubheading Synopsis
24952 -target-list-parameters
24958 @subsubheading @value{GDBN} Command
24962 @subsubheading Example
24966 @subheading The @code{-target-select} Command
24967 @findex -target-select
24969 @subsubheading Synopsis
24972 -target-select @var{type} @var{parameters @dots{}}
24975 Connect @value{GDBN} to the remote target. This command takes two args:
24979 The type of target, for instance @samp{remote}, etc.
24980 @item @var{parameters}
24981 Device names, host names and the like. @xref{Target Commands, ,
24982 Commands for Managing Targets}, for more details.
24985 The output is a connection notification, followed by the address at
24986 which the target program is, in the following form:
24989 ^connected,addr="@var{address}",func="@var{function name}",
24990 args=[@var{arg list}]
24993 @subsubheading @value{GDBN} Command
24995 The corresponding @value{GDBN} command is @samp{target}.
24997 @subsubheading Example
25001 -target-select remote /dev/ttya
25002 ^connected,addr="0xfe00a300",func="??",args=[]
25006 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
25007 @node GDB/MI File Transfer Commands
25008 @section @sc{gdb/mi} File Transfer Commands
25011 @subheading The @code{-target-file-put} Command
25012 @findex -target-file-put
25014 @subsubheading Synopsis
25017 -target-file-put @var{hostfile} @var{targetfile}
25020 Copy file @var{hostfile} from the host system (the machine running
25021 @value{GDBN}) to @var{targetfile} on the target system.
25023 @subsubheading @value{GDBN} Command
25025 The corresponding @value{GDBN} command is @samp{remote put}.
25027 @subsubheading Example
25031 -target-file-put localfile remotefile
25037 @subheading The @code{-target-file-get} Command
25038 @findex -target-file-get
25040 @subsubheading Synopsis
25043 -target-file-get @var{targetfile} @var{hostfile}
25046 Copy file @var{targetfile} from the target system to @var{hostfile}
25047 on the host system.
25049 @subsubheading @value{GDBN} Command
25051 The corresponding @value{GDBN} command is @samp{remote get}.
25053 @subsubheading Example
25057 -target-file-get remotefile localfile
25063 @subheading The @code{-target-file-delete} Command
25064 @findex -target-file-delete
25066 @subsubheading Synopsis
25069 -target-file-delete @var{targetfile}
25072 Delete @var{targetfile} from the target system.
25074 @subsubheading @value{GDBN} Command
25076 The corresponding @value{GDBN} command is @samp{remote delete}.
25078 @subsubheading Example
25082 -target-file-delete remotefile
25088 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
25089 @node GDB/MI Miscellaneous Commands
25090 @section Miscellaneous @sc{gdb/mi} Commands
25092 @c @subheading -gdb-complete
25094 @subheading The @code{-gdb-exit} Command
25097 @subsubheading Synopsis
25103 Exit @value{GDBN} immediately.
25105 @subsubheading @value{GDBN} Command
25107 Approximately corresponds to @samp{quit}.
25109 @subsubheading Example
25119 @subheading The @code{-exec-abort} Command
25120 @findex -exec-abort
25122 @subsubheading Synopsis
25128 Kill the inferior running program.
25130 @subsubheading @value{GDBN} Command
25132 The corresponding @value{GDBN} command is @samp{kill}.
25134 @subsubheading Example
25139 @subheading The @code{-gdb-set} Command
25142 @subsubheading Synopsis
25148 Set an internal @value{GDBN} variable.
25149 @c IS THIS A DOLLAR VARIABLE? OR SOMETHING LIKE ANNOTATE ?????
25151 @subsubheading @value{GDBN} Command
25153 The corresponding @value{GDBN} command is @samp{set}.
25155 @subsubheading Example
25165 @subheading The @code{-gdb-show} Command
25168 @subsubheading Synopsis
25174 Show the current value of a @value{GDBN} variable.
25176 @subsubheading @value{GDBN} Command
25178 The corresponding @value{GDBN} command is @samp{show}.
25180 @subsubheading Example
25189 @c @subheading -gdb-source
25192 @subheading The @code{-gdb-version} Command
25193 @findex -gdb-version
25195 @subsubheading Synopsis
25201 Show version information for @value{GDBN}. Used mostly in testing.
25203 @subsubheading @value{GDBN} Command
25205 The @value{GDBN} equivalent is @samp{show version}. @value{GDBN} by
25206 default shows this information when you start an interactive session.
25208 @subsubheading Example
25210 @c This example modifies the actual output from GDB to avoid overfull
25216 ~Copyright 2000 Free Software Foundation, Inc.
25217 ~GDB is free software, covered by the GNU General Public License, and
25218 ~you are welcome to change it and/or distribute copies of it under
25219 ~ certain conditions.
25220 ~Type "show copying" to see the conditions.
25221 ~There is absolutely no warranty for GDB. Type "show warranty" for
25223 ~This GDB was configured as
25224 "--host=sparc-sun-solaris2.5.1 --target=ppc-eabi".
25229 @subheading The @code{-list-features} Command
25230 @findex -list-features
25232 Returns a list of particular features of the MI protocol that
25233 this version of gdb implements. A feature can be a command,
25234 or a new field in an output of some command, or even an
25235 important bugfix. While a frontend can sometimes detect presence
25236 of a feature at runtime, it is easier to perform detection at debugger
25239 The command returns a list of strings, with each string naming an
25240 available feature. Each returned string is just a name, it does not
25241 have any internal structure. The list of possible feature names
25247 (gdb) -list-features
25248 ^done,result=["feature1","feature2"]
25251 The current list of features is:
25254 @item frozen-varobjs
25255 Indicates presence of the @code{-var-set-frozen} command, as well
25256 as possible presense of the @code{frozen} field in the output
25257 of @code{-varobj-create}.
25258 @item pending-breakpoints
25259 Indicates presence of the @option{-f} option to the @code{-break-insert} command.
25261 Indicates presence of Python scripting support, Python-based
25262 pretty-printing commands, and possible presence of the
25263 @samp{display_hint} field in the output of @code{-var-list-children}
25265 Indicates presence of the @code{-thread-info} command.
25269 @subheading The @code{-list-target-features} Command
25270 @findex -list-target-features
25272 Returns a list of particular features that are supported by the
25273 target. Those features affect the permitted MI commands, but
25274 unlike the features reported by the @code{-list-features} command, the
25275 features depend on which target GDB is using at the moment. Whenever
25276 a target can change, due to commands such as @code{-target-select},
25277 @code{-target-attach} or @code{-exec-run}, the list of target features
25278 may change, and the frontend should obtain it again.
25282 (gdb) -list-features
25283 ^done,result=["async"]
25286 The current list of features is:
25290 Indicates that the target is capable of asynchronous command
25291 execution, which means that @value{GDBN} will accept further commands
25292 while the target is running.
25296 @subheading The @code{-list-thread-groups} Command
25297 @findex -list-thread-groups
25299 @subheading Synopsis
25302 -list-thread-groups [ --available ] [ @var{group} ]
25305 When used without the @var{group} parameter, lists top-level thread
25306 groups that are being debugged. When used with the @var{group}
25307 parameter, the children of the specified group are listed. The
25308 children can be either threads, or other groups. At present,
25309 @value{GDBN} will not report both threads and groups as children at
25310 the same time, but it may change in future.
25312 With the @samp{--available} option, instead of reporting groups that
25313 are been debugged, GDB will report all thread groups available on the
25314 target. Using the @samp{--available} option together with @var{group}
25317 @subheading Example
25321 -list-thread-groups
25322 ^done,groups=[@{id="17",type="process",pid="yyy",num_children="2"@}]
25323 -list-thread-groups 17
25324 ^done,threads=[@{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
25325 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",args=[]@},state="running"@},
25326 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
25327 frame=@{level="0",addr="0x0804891f",func="foo",args=[@{name="i",value="10"@}],
25328 file="/tmp/a.c",fullname="/tmp/a.c",line="158"@},state="running"@}]]
25331 @subheading The @code{-interpreter-exec} Command
25332 @findex -interpreter-exec
25334 @subheading Synopsis
25337 -interpreter-exec @var{interpreter} @var{command}
25339 @anchor{-interpreter-exec}
25341 Execute the specified @var{command} in the given @var{interpreter}.
25343 @subheading @value{GDBN} Command
25345 The corresponding @value{GDBN} command is @samp{interpreter-exec}.
25347 @subheading Example
25351 -interpreter-exec console "break main"
25352 &"During symbol reading, couldn't parse type; debugger out of date?.\n"
25353 &"During symbol reading, bad structure-type format.\n"
25354 ~"Breakpoint 1 at 0x8074fc6: file ../../src/gdb/main.c, line 743.\n"
25359 @subheading The @code{-inferior-tty-set} Command
25360 @findex -inferior-tty-set
25362 @subheading Synopsis
25365 -inferior-tty-set /dev/pts/1
25368 Set terminal for future runs of the program being debugged.
25370 @subheading @value{GDBN} Command
25372 The corresponding @value{GDBN} command is @samp{set inferior-tty} /dev/pts/1.
25374 @subheading Example
25378 -inferior-tty-set /dev/pts/1
25383 @subheading The @code{-inferior-tty-show} Command
25384 @findex -inferior-tty-show
25386 @subheading Synopsis
25392 Show terminal for future runs of program being debugged.
25394 @subheading @value{GDBN} Command
25396 The corresponding @value{GDBN} command is @samp{show inferior-tty}.
25398 @subheading Example
25402 -inferior-tty-set /dev/pts/1
25406 ^done,inferior_tty_terminal="/dev/pts/1"
25410 @subheading The @code{-enable-timings} Command
25411 @findex -enable-timings
25413 @subheading Synopsis
25416 -enable-timings [yes | no]
25419 Toggle the printing of the wallclock, user and system times for an MI
25420 command as a field in its output. This command is to help frontend
25421 developers optimize the performance of their code. No argument is
25422 equivalent to @samp{yes}.
25424 @subheading @value{GDBN} Command
25428 @subheading Example
25436 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
25437 addr="0x080484ed",func="main",file="myprog.c",
25438 fullname="/home/nickrob/myprog.c",line="73",times="0"@},
25439 time=@{wallclock="0.05185",user="0.00800",system="0.00000"@}
25447 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
25448 frame=@{addr="0x080484ed",func="main",args=[@{name="argc",value="1"@},
25449 @{name="argv",value="0xbfb60364"@}],file="myprog.c",
25450 fullname="/home/nickrob/myprog.c",line="73"@}
25455 @chapter @value{GDBN} Annotations
25457 This chapter describes annotations in @value{GDBN}. Annotations were
25458 designed to interface @value{GDBN} to graphical user interfaces or other
25459 similar programs which want to interact with @value{GDBN} at a
25460 relatively high level.
25462 The annotation mechanism has largely been superseded by @sc{gdb/mi}
25466 This is Edition @value{EDITION}, @value{DATE}.
25470 * Annotations Overview:: What annotations are; the general syntax.
25471 * Server Prefix:: Issuing a command without affecting user state.
25472 * Prompting:: Annotations marking @value{GDBN}'s need for input.
25473 * Errors:: Annotations for error messages.
25474 * Invalidation:: Some annotations describe things now invalid.
25475 * Annotations for Running::
25476 Whether the program is running, how it stopped, etc.
25477 * Source Annotations:: Annotations describing source code.
25480 @node Annotations Overview
25481 @section What is an Annotation?
25482 @cindex annotations
25484 Annotations start with a newline character, two @samp{control-z}
25485 characters, and the name of the annotation. If there is no additional
25486 information associated with this annotation, the name of the annotation
25487 is followed immediately by a newline. If there is additional
25488 information, the name of the annotation is followed by a space, the
25489 additional information, and a newline. The additional information
25490 cannot contain newline characters.
25492 Any output not beginning with a newline and two @samp{control-z}
25493 characters denotes literal output from @value{GDBN}. Currently there is
25494 no need for @value{GDBN} to output a newline followed by two
25495 @samp{control-z} characters, but if there was such a need, the
25496 annotations could be extended with an @samp{escape} annotation which
25497 means those three characters as output.
25499 The annotation @var{level}, which is specified using the
25500 @option{--annotate} command line option (@pxref{Mode Options}), controls
25501 how much information @value{GDBN} prints together with its prompt,
25502 values of expressions, source lines, and other types of output. Level 0
25503 is for no annotations, level 1 is for use when @value{GDBN} is run as a
25504 subprocess of @sc{gnu} Emacs, level 3 is the maximum annotation suitable
25505 for programs that control @value{GDBN}, and level 2 annotations have
25506 been made obsolete (@pxref{Limitations, , Limitations of the Annotation
25507 Interface, annotate, GDB's Obsolete Annotations}).
25510 @kindex set annotate
25511 @item set annotate @var{level}
25512 The @value{GDBN} command @code{set annotate} sets the level of
25513 annotations to the specified @var{level}.
25515 @item show annotate
25516 @kindex show annotate
25517 Show the current annotation level.
25520 This chapter describes level 3 annotations.
25522 A simple example of starting up @value{GDBN} with annotations is:
25525 $ @kbd{gdb --annotate=3}
25527 Copyright 2003 Free Software Foundation, Inc.
25528 GDB is free software, covered by the GNU General Public License,
25529 and you are welcome to change it and/or distribute copies of it
25530 under certain conditions.
25531 Type "show copying" to see the conditions.
25532 There is absolutely no warranty for GDB. Type "show warranty"
25534 This GDB was configured as "i386-pc-linux-gnu"
25545 Here @samp{quit} is input to @value{GDBN}; the rest is output from
25546 @value{GDBN}. The three lines beginning @samp{^Z^Z} (where @samp{^Z}
25547 denotes a @samp{control-z} character) are annotations; the rest is
25548 output from @value{GDBN}.
25550 @node Server Prefix
25551 @section The Server Prefix
25552 @cindex server prefix
25554 If you prefix a command with @samp{server } then it will not affect
25555 the command history, nor will it affect @value{GDBN}'s notion of which
25556 command to repeat if @key{RET} is pressed on a line by itself. This
25557 means that commands can be run behind a user's back by a front-end in
25558 a transparent manner.
25560 The server prefix does not affect the recording of values into the value
25561 history; to print a value without recording it into the value history,
25562 use the @code{output} command instead of the @code{print} command.
25565 @section Annotation for @value{GDBN} Input
25567 @cindex annotations for prompts
25568 When @value{GDBN} prompts for input, it annotates this fact so it is possible
25569 to know when to send output, when the output from a given command is
25572 Different kinds of input each have a different @dfn{input type}. Each
25573 input type has three annotations: a @code{pre-} annotation, which
25574 denotes the beginning of any prompt which is being output, a plain
25575 annotation, which denotes the end of the prompt, and then a @code{post-}
25576 annotation which denotes the end of any echo which may (or may not) be
25577 associated with the input. For example, the @code{prompt} input type
25578 features the following annotations:
25586 The input types are
25589 @findex pre-prompt annotation
25590 @findex prompt annotation
25591 @findex post-prompt annotation
25593 When @value{GDBN} is prompting for a command (the main @value{GDBN} prompt).
25595 @findex pre-commands annotation
25596 @findex commands annotation
25597 @findex post-commands annotation
25599 When @value{GDBN} prompts for a set of commands, like in the @code{commands}
25600 command. The annotations are repeated for each command which is input.
25602 @findex pre-overload-choice annotation
25603 @findex overload-choice annotation
25604 @findex post-overload-choice annotation
25605 @item overload-choice
25606 When @value{GDBN} wants the user to select between various overloaded functions.
25608 @findex pre-query annotation
25609 @findex query annotation
25610 @findex post-query annotation
25612 When @value{GDBN} wants the user to confirm a potentially dangerous operation.
25614 @findex pre-prompt-for-continue annotation
25615 @findex prompt-for-continue annotation
25616 @findex post-prompt-for-continue annotation
25617 @item prompt-for-continue
25618 When @value{GDBN} is asking the user to press return to continue. Note: Don't
25619 expect this to work well; instead use @code{set height 0} to disable
25620 prompting. This is because the counting of lines is buggy in the
25621 presence of annotations.
25626 @cindex annotations for errors, warnings and interrupts
25628 @findex quit annotation
25633 This annotation occurs right before @value{GDBN} responds to an interrupt.
25635 @findex error annotation
25640 This annotation occurs right before @value{GDBN} responds to an error.
25642 Quit and error annotations indicate that any annotations which @value{GDBN} was
25643 in the middle of may end abruptly. For example, if a
25644 @code{value-history-begin} annotation is followed by a @code{error}, one
25645 cannot expect to receive the matching @code{value-history-end}. One
25646 cannot expect not to receive it either, however; an error annotation
25647 does not necessarily mean that @value{GDBN} is immediately returning all the way
25650 @findex error-begin annotation
25651 A quit or error annotation may be preceded by
25657 Any output between that and the quit or error annotation is the error
25660 Warning messages are not yet annotated.
25661 @c If we want to change that, need to fix warning(), type_error(),
25662 @c range_error(), and possibly other places.
25665 @section Invalidation Notices
25667 @cindex annotations for invalidation messages
25668 The following annotations say that certain pieces of state may have
25672 @findex frames-invalid annotation
25673 @item ^Z^Zframes-invalid
25675 The frames (for example, output from the @code{backtrace} command) may
25678 @findex breakpoints-invalid annotation
25679 @item ^Z^Zbreakpoints-invalid
25681 The breakpoints may have changed. For example, the user just added or
25682 deleted a breakpoint.
25685 @node Annotations for Running
25686 @section Running the Program
25687 @cindex annotations for running programs
25689 @findex starting annotation
25690 @findex stopping annotation
25691 When the program starts executing due to a @value{GDBN} command such as
25692 @code{step} or @code{continue},
25698 is output. When the program stops,
25704 is output. Before the @code{stopped} annotation, a variety of
25705 annotations describe how the program stopped.
25708 @findex exited annotation
25709 @item ^Z^Zexited @var{exit-status}
25710 The program exited, and @var{exit-status} is the exit status (zero for
25711 successful exit, otherwise nonzero).
25713 @findex signalled annotation
25714 @findex signal-name annotation
25715 @findex signal-name-end annotation
25716 @findex signal-string annotation
25717 @findex signal-string-end annotation
25718 @item ^Z^Zsignalled
25719 The program exited with a signal. After the @code{^Z^Zsignalled}, the
25720 annotation continues:
25726 ^Z^Zsignal-name-end
25730 ^Z^Zsignal-string-end
25735 where @var{name} is the name of the signal, such as @code{SIGILL} or
25736 @code{SIGSEGV}, and @var{string} is the explanation of the signal, such
25737 as @code{Illegal Instruction} or @code{Segmentation fault}.
25738 @var{intro-text}, @var{middle-text}, and @var{end-text} are for the
25739 user's benefit and have no particular format.
25741 @findex signal annotation
25743 The syntax of this annotation is just like @code{signalled}, but @value{GDBN} is
25744 just saying that the program received the signal, not that it was
25745 terminated with it.
25747 @findex breakpoint annotation
25748 @item ^Z^Zbreakpoint @var{number}
25749 The program hit breakpoint number @var{number}.
25751 @findex watchpoint annotation
25752 @item ^Z^Zwatchpoint @var{number}
25753 The program hit watchpoint number @var{number}.
25756 @node Source Annotations
25757 @section Displaying Source
25758 @cindex annotations for source display
25760 @findex source annotation
25761 The following annotation is used instead of displaying source code:
25764 ^Z^Zsource @var{filename}:@var{line}:@var{character}:@var{middle}:@var{addr}
25767 where @var{filename} is an absolute file name indicating which source
25768 file, @var{line} is the line number within that file (where 1 is the
25769 first line in the file), @var{character} is the character position
25770 within the file (where 0 is the first character in the file) (for most
25771 debug formats this will necessarily point to the beginning of a line),
25772 @var{middle} is @samp{middle} if @var{addr} is in the middle of the
25773 line, or @samp{beg} if @var{addr} is at the beginning of the line, and
25774 @var{addr} is the address in the target program associated with the
25775 source which is being displayed. @var{addr} is in the form @samp{0x}
25776 followed by one or more lowercase hex digits (note that this does not
25777 depend on the language).
25780 @chapter Reporting Bugs in @value{GDBN}
25781 @cindex bugs in @value{GDBN}
25782 @cindex reporting bugs in @value{GDBN}
25784 Your bug reports play an essential role in making @value{GDBN} reliable.
25786 Reporting a bug may help you by bringing a solution to your problem, or it
25787 may not. But in any case the principal function of a bug report is to help
25788 the entire community by making the next version of @value{GDBN} work better. Bug
25789 reports are your contribution to the maintenance of @value{GDBN}.
25791 In order for a bug report to serve its purpose, you must include the
25792 information that enables us to fix the bug.
25795 * Bug Criteria:: Have you found a bug?
25796 * Bug Reporting:: How to report bugs
25800 @section Have You Found a Bug?
25801 @cindex bug criteria
25803 If you are not sure whether you have found a bug, here are some guidelines:
25806 @cindex fatal signal
25807 @cindex debugger crash
25808 @cindex crash of debugger
25810 If the debugger gets a fatal signal, for any input whatever, that is a
25811 @value{GDBN} bug. Reliable debuggers never crash.
25813 @cindex error on valid input
25815 If @value{GDBN} produces an error message for valid input, that is a
25816 bug. (Note that if you're cross debugging, the problem may also be
25817 somewhere in the connection to the target.)
25819 @cindex invalid input
25821 If @value{GDBN} does not produce an error message for invalid input,
25822 that is a bug. However, you should note that your idea of
25823 ``invalid input'' might be our idea of ``an extension'' or ``support
25824 for traditional practice''.
25827 If you are an experienced user of debugging tools, your suggestions
25828 for improvement of @value{GDBN} are welcome in any case.
25831 @node Bug Reporting
25832 @section How to Report Bugs
25833 @cindex bug reports
25834 @cindex @value{GDBN} bugs, reporting
25836 A number of companies and individuals offer support for @sc{gnu} products.
25837 If you obtained @value{GDBN} from a support organization, we recommend you
25838 contact that organization first.
25840 You can find contact information for many support companies and
25841 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
25843 @c should add a web page ref...
25846 @ifset BUGURL_DEFAULT
25847 In any event, we also recommend that you submit bug reports for
25848 @value{GDBN}. The preferred method is to submit them directly using
25849 @uref{http://www.gnu.org/software/gdb/bugs/, @value{GDBN}'s Bugs web
25850 page}. Alternatively, the @email{bug-gdb@@gnu.org, e-mail gateway} can
25853 @strong{Do not send bug reports to @samp{info-gdb}, or to
25854 @samp{help-gdb}, or to any newsgroups.} Most users of @value{GDBN} do
25855 not want to receive bug reports. Those that do have arranged to receive
25858 The mailing list @samp{bug-gdb} has a newsgroup @samp{gnu.gdb.bug} which
25859 serves as a repeater. The mailing list and the newsgroup carry exactly
25860 the same messages. Often people think of posting bug reports to the
25861 newsgroup instead of mailing them. This appears to work, but it has one
25862 problem which can be crucial: a newsgroup posting often lacks a mail
25863 path back to the sender. Thus, if we need to ask for more information,
25864 we may be unable to reach you. For this reason, it is better to send
25865 bug reports to the mailing list.
25867 @ifclear BUGURL_DEFAULT
25868 In any event, we also recommend that you submit bug reports for
25869 @value{GDBN} to @value{BUGURL}.
25873 The fundamental principle of reporting bugs usefully is this:
25874 @strong{report all the facts}. If you are not sure whether to state a
25875 fact or leave it out, state it!
25877 Often people omit facts because they think they know what causes the
25878 problem and assume that some details do not matter. Thus, you might
25879 assume that the name of the variable you use in an example does not matter.
25880 Well, probably it does not, but one cannot be sure. Perhaps the bug is a
25881 stray memory reference which happens to fetch from the location where that
25882 name is stored in memory; perhaps, if the name were different, the contents
25883 of that location would fool the debugger into doing the right thing despite
25884 the bug. Play it safe and give a specific, complete example. That is the
25885 easiest thing for you to do, and the most helpful.
25887 Keep in mind that the purpose of a bug report is to enable us to fix the
25888 bug. It may be that the bug has been reported previously, but neither
25889 you nor we can know that unless your bug report is complete and
25892 Sometimes people give a few sketchy facts and ask, ``Does this ring a
25893 bell?'' Those bug reports are useless, and we urge everyone to
25894 @emph{refuse to respond to them} except to chide the sender to report
25897 To enable us to fix the bug, you should include all these things:
25901 The version of @value{GDBN}. @value{GDBN} announces it if you start
25902 with no arguments; you can also print it at any time using @code{show
25905 Without this, we will not know whether there is any point in looking for
25906 the bug in the current version of @value{GDBN}.
25909 The type of machine you are using, and the operating system name and
25913 What compiler (and its version) was used to compile @value{GDBN}---e.g.@:
25914 ``@value{GCC}--2.8.1''.
25917 What compiler (and its version) was used to compile the program you are
25918 debugging---e.g.@: ``@value{GCC}--2.8.1'', or ``HP92453-01 A.10.32.03 HP
25919 C Compiler''. For @value{NGCC}, you can say @kbd{@value{GCC} --version}
25920 to get this information; for other compilers, see the documentation for
25924 The command arguments you gave the compiler to compile your example and
25925 observe the bug. For example, did you use @samp{-O}? To guarantee
25926 you will not omit something important, list them all. A copy of the
25927 Makefile (or the output from make) is sufficient.
25929 If we were to try to guess the arguments, we would probably guess wrong
25930 and then we might not encounter the bug.
25933 A complete input script, and all necessary source files, that will
25937 A description of what behavior you observe that you believe is
25938 incorrect. For example, ``It gets a fatal signal.''
25940 Of course, if the bug is that @value{GDBN} gets a fatal signal, then we
25941 will certainly notice it. But if the bug is incorrect output, we might
25942 not notice unless it is glaringly wrong. You might as well not give us
25943 a chance to make a mistake.
25945 Even if the problem you experience is a fatal signal, you should still
25946 say so explicitly. Suppose something strange is going on, such as, your
25947 copy of @value{GDBN} is out of synch, or you have encountered a bug in
25948 the C library on your system. (This has happened!) Your copy might
25949 crash and ours would not. If you told us to expect a crash, then when
25950 ours fails to crash, we would know that the bug was not happening for
25951 us. If you had not told us to expect a crash, then we would not be able
25952 to draw any conclusion from our observations.
25955 @cindex recording a session script
25956 To collect all this information, you can use a session recording program
25957 such as @command{script}, which is available on many Unix systems.
25958 Just run your @value{GDBN} session inside @command{script} and then
25959 include the @file{typescript} file with your bug report.
25961 Another way to record a @value{GDBN} session is to run @value{GDBN}
25962 inside Emacs and then save the entire buffer to a file.
25965 If you wish to suggest changes to the @value{GDBN} source, send us context
25966 diffs. If you even discuss something in the @value{GDBN} source, refer to
25967 it by context, not by line number.
25969 The line numbers in our development sources will not match those in your
25970 sources. Your line numbers would convey no useful information to us.
25974 Here are some things that are not necessary:
25978 A description of the envelope of the bug.
25980 Often people who encounter a bug spend a lot of time investigating
25981 which changes to the input file will make the bug go away and which
25982 changes will not affect it.
25984 This is often time consuming and not very useful, because the way we
25985 will find the bug is by running a single example under the debugger
25986 with breakpoints, not by pure deduction from a series of examples.
25987 We recommend that you save your time for something else.
25989 Of course, if you can find a simpler example to report @emph{instead}
25990 of the original one, that is a convenience for us. Errors in the
25991 output will be easier to spot, running under the debugger will take
25992 less time, and so on.
25994 However, simplification is not vital; if you do not want to do this,
25995 report the bug anyway and send us the entire test case you used.
25998 A patch for the bug.
26000 A patch for the bug does help us if it is a good one. But do not omit
26001 the necessary information, such as the test case, on the assumption that
26002 a patch is all we need. We might see problems with your patch and decide
26003 to fix the problem another way, or we might not understand it at all.
26005 Sometimes with a program as complicated as @value{GDBN} it is very hard to
26006 construct an example that will make the program follow a certain path
26007 through the code. If you do not send us the example, we will not be able
26008 to construct one, so we will not be able to verify that the bug is fixed.
26010 And if we cannot understand what bug you are trying to fix, or why your
26011 patch should be an improvement, we will not install it. A test case will
26012 help us to understand.
26015 A guess about what the bug is or what it depends on.
26017 Such guesses are usually wrong. Even we cannot guess right about such
26018 things without first using the debugger to find the facts.
26021 @c The readline documentation is distributed with the readline code
26022 @c and consists of the two following files:
26024 @c inc-hist.texinfo
26025 @c Use -I with makeinfo to point to the appropriate directory,
26026 @c environment var TEXINPUTS with TeX.
26027 @include rluser.texi
26028 @include inc-hist.texinfo
26031 @node Formatting Documentation
26032 @appendix Formatting Documentation
26034 @cindex @value{GDBN} reference card
26035 @cindex reference card
26036 The @value{GDBN} 4 release includes an already-formatted reference card, ready
26037 for printing with PostScript or Ghostscript, in the @file{gdb}
26038 subdirectory of the main source directory@footnote{In
26039 @file{gdb-@value{GDBVN}/gdb/refcard.ps} of the version @value{GDBVN}
26040 release.}. If you can use PostScript or Ghostscript with your printer,
26041 you can print the reference card immediately with @file{refcard.ps}.
26043 The release also includes the source for the reference card. You
26044 can format it, using @TeX{}, by typing:
26050 The @value{GDBN} reference card is designed to print in @dfn{landscape}
26051 mode on US ``letter'' size paper;
26052 that is, on a sheet 11 inches wide by 8.5 inches
26053 high. You will need to specify this form of printing as an option to
26054 your @sc{dvi} output program.
26056 @cindex documentation
26058 All the documentation for @value{GDBN} comes as part of the machine-readable
26059 distribution. The documentation is written in Texinfo format, which is
26060 a documentation system that uses a single source file to produce both
26061 on-line information and a printed manual. You can use one of the Info
26062 formatting commands to create the on-line version of the documentation
26063 and @TeX{} (or @code{texi2roff}) to typeset the printed version.
26065 @value{GDBN} includes an already formatted copy of the on-line Info
26066 version of this manual in the @file{gdb} subdirectory. The main Info
26067 file is @file{gdb-@value{GDBVN}/gdb/gdb.info}, and it refers to
26068 subordinate files matching @samp{gdb.info*} in the same directory. If
26069 necessary, you can print out these files, or read them with any editor;
26070 but they are easier to read using the @code{info} subsystem in @sc{gnu}
26071 Emacs or the standalone @code{info} program, available as part of the
26072 @sc{gnu} Texinfo distribution.
26074 If you want to format these Info files yourself, you need one of the
26075 Info formatting programs, such as @code{texinfo-format-buffer} or
26078 If you have @code{makeinfo} installed, and are in the top level
26079 @value{GDBN} source directory (@file{gdb-@value{GDBVN}}, in the case of
26080 version @value{GDBVN}), you can make the Info file by typing:
26087 If you want to typeset and print copies of this manual, you need @TeX{},
26088 a program to print its @sc{dvi} output files, and @file{texinfo.tex}, the
26089 Texinfo definitions file.
26091 @TeX{} is a typesetting program; it does not print files directly, but
26092 produces output files called @sc{dvi} files. To print a typeset
26093 document, you need a program to print @sc{dvi} files. If your system
26094 has @TeX{} installed, chances are it has such a program. The precise
26095 command to use depends on your system; @kbd{lpr -d} is common; another
26096 (for PostScript devices) is @kbd{dvips}. The @sc{dvi} print command may
26097 require a file name without any extension or a @samp{.dvi} extension.
26099 @TeX{} also requires a macro definitions file called
26100 @file{texinfo.tex}. This file tells @TeX{} how to typeset a document
26101 written in Texinfo format. On its own, @TeX{} cannot either read or
26102 typeset a Texinfo file. @file{texinfo.tex} is distributed with GDB
26103 and is located in the @file{gdb-@var{version-number}/texinfo}
26106 If you have @TeX{} and a @sc{dvi} printer program installed, you can
26107 typeset and print this manual. First switch to the @file{gdb}
26108 subdirectory of the main source directory (for example, to
26109 @file{gdb-@value{GDBVN}/gdb}) and type:
26115 Then give @file{gdb.dvi} to your @sc{dvi} printing program.
26117 @node Installing GDB
26118 @appendix Installing @value{GDBN}
26119 @cindex installation
26122 * Requirements:: Requirements for building @value{GDBN}
26123 * Running Configure:: Invoking the @value{GDBN} @file{configure} script
26124 * Separate Objdir:: Compiling @value{GDBN} in another directory
26125 * Config Names:: Specifying names for hosts and targets
26126 * Configure Options:: Summary of options for configure
26127 * System-wide configuration:: Having a system-wide init file
26131 @section Requirements for Building @value{GDBN}
26132 @cindex building @value{GDBN}, requirements for
26134 Building @value{GDBN} requires various tools and packages to be available.
26135 Other packages will be used only if they are found.
26137 @heading Tools/Packages Necessary for Building @value{GDBN}
26139 @item ISO C90 compiler
26140 @value{GDBN} is written in ISO C90. It should be buildable with any
26141 working C90 compiler, e.g.@: GCC.
26145 @heading Tools/Packages Optional for Building @value{GDBN}
26149 @value{GDBN} can use the Expat XML parsing library. This library may be
26150 included with your operating system distribution; if it is not, you
26151 can get the latest version from @url{http://expat.sourceforge.net}.
26152 The @file{configure} script will search for this library in several
26153 standard locations; if it is installed in an unusual path, you can
26154 use the @option{--with-libexpat-prefix} option to specify its location.
26160 Remote protocol memory maps (@pxref{Memory Map Format})
26162 Target descriptions (@pxref{Target Descriptions})
26164 Remote shared library lists (@pxref{Library List Format})
26166 MS-Windows shared libraries (@pxref{Shared Libraries})
26170 @cindex compressed debug sections
26171 @value{GDBN} will use the @samp{zlib} library, if available, to read
26172 compressed debug sections. Some linkers, such as GNU gold, are capable
26173 of producing binaries with compressed debug sections. If @value{GDBN}
26174 is compiled with @samp{zlib}, it will be able to read the debug
26175 information in such binaries.
26177 The @samp{zlib} library is likely included with your operating system
26178 distribution; if it is not, you can get the latest version from
26179 @url{http://zlib.net}.
26182 @value{GDBN}'s features related to character sets (@pxref{Character
26183 Sets}) require a functioning @code{iconv} implementation. If you are
26184 on a GNU system, then this is provided by the GNU C Library. Some
26185 other systems also provide a working @code{iconv}.
26187 On systems with @code{iconv}, you can install GNU Libiconv. If you
26188 have previously installed Libiconv, you can use the
26189 @option{--with-libiconv-prefix} option to configure.
26191 @value{GDBN}'s top-level @file{configure} and @file{Makefile} will
26192 arrange to build Libiconv if a directory named @file{libiconv} appears
26193 in the top-most source directory. If Libiconv is built this way, and
26194 if the operating system does not provide a suitable @code{iconv}
26195 implementation, then the just-built library will automatically be used
26196 by @value{GDBN}. One easy way to set this up is to download GNU
26197 Libiconv, unpack it, and then rename the directory holding the
26198 Libiconv source code to @samp{libiconv}.
26201 @node Running Configure
26202 @section Invoking the @value{GDBN} @file{configure} Script
26203 @cindex configuring @value{GDBN}
26204 @value{GDBN} comes with a @file{configure} script that automates the process
26205 of preparing @value{GDBN} for installation; you can then use @code{make} to
26206 build the @code{gdb} program.
26208 @c irrelevant in info file; it's as current as the code it lives with.
26209 @footnote{If you have a more recent version of @value{GDBN} than @value{GDBVN},
26210 look at the @file{README} file in the sources; we may have improved the
26211 installation procedures since publishing this manual.}
26214 The @value{GDBN} distribution includes all the source code you need for
26215 @value{GDBN} in a single directory, whose name is usually composed by
26216 appending the version number to @samp{gdb}.
26218 For example, the @value{GDBN} version @value{GDBVN} distribution is in the
26219 @file{gdb-@value{GDBVN}} directory. That directory contains:
26222 @item gdb-@value{GDBVN}/configure @r{(and supporting files)}
26223 script for configuring @value{GDBN} and all its supporting libraries
26225 @item gdb-@value{GDBVN}/gdb
26226 the source specific to @value{GDBN} itself
26228 @item gdb-@value{GDBVN}/bfd
26229 source for the Binary File Descriptor library
26231 @item gdb-@value{GDBVN}/include
26232 @sc{gnu} include files
26234 @item gdb-@value{GDBVN}/libiberty
26235 source for the @samp{-liberty} free software library
26237 @item gdb-@value{GDBVN}/opcodes
26238 source for the library of opcode tables and disassemblers
26240 @item gdb-@value{GDBVN}/readline
26241 source for the @sc{gnu} command-line interface
26243 @item gdb-@value{GDBVN}/glob
26244 source for the @sc{gnu} filename pattern-matching subroutine
26246 @item gdb-@value{GDBVN}/mmalloc
26247 source for the @sc{gnu} memory-mapped malloc package
26250 The simplest way to configure and build @value{GDBN} is to run @file{configure}
26251 from the @file{gdb-@var{version-number}} source directory, which in
26252 this example is the @file{gdb-@value{GDBVN}} directory.
26254 First switch to the @file{gdb-@var{version-number}} source directory
26255 if you are not already in it; then run @file{configure}. Pass the
26256 identifier for the platform on which @value{GDBN} will run as an
26262 cd gdb-@value{GDBVN}
26263 ./configure @var{host}
26268 where @var{host} is an identifier such as @samp{sun4} or
26269 @samp{decstation}, that identifies the platform where @value{GDBN} will run.
26270 (You can often leave off @var{host}; @file{configure} tries to guess the
26271 correct value by examining your system.)
26273 Running @samp{configure @var{host}} and then running @code{make} builds the
26274 @file{bfd}, @file{readline}, @file{mmalloc}, and @file{libiberty}
26275 libraries, then @code{gdb} itself. The configured source files, and the
26276 binaries, are left in the corresponding source directories.
26279 @file{configure} is a Bourne-shell (@code{/bin/sh}) script; if your
26280 system does not recognize this automatically when you run a different
26281 shell, you may need to run @code{sh} on it explicitly:
26284 sh configure @var{host}
26287 If you run @file{configure} from a directory that contains source
26288 directories for multiple libraries or programs, such as the
26289 @file{gdb-@value{GDBVN}} source directory for version @value{GDBVN},
26291 creates configuration files for every directory level underneath (unless
26292 you tell it not to, with the @samp{--norecursion} option).
26294 You should run the @file{configure} script from the top directory in the
26295 source tree, the @file{gdb-@var{version-number}} directory. If you run
26296 @file{configure} from one of the subdirectories, you will configure only
26297 that subdirectory. That is usually not what you want. In particular,
26298 if you run the first @file{configure} from the @file{gdb} subdirectory
26299 of the @file{gdb-@var{version-number}} directory, you will omit the
26300 configuration of @file{bfd}, @file{readline}, and other sibling
26301 directories of the @file{gdb} subdirectory. This leads to build errors
26302 about missing include files such as @file{bfd/bfd.h}.
26304 You can install @code{@value{GDBP}} anywhere; it has no hardwired paths.
26305 However, you should make sure that the shell on your path (named by
26306 the @samp{SHELL} environment variable) is publicly readable. Remember
26307 that @value{GDBN} uses the shell to start your program---some systems refuse to
26308 let @value{GDBN} debug child processes whose programs are not readable.
26310 @node Separate Objdir
26311 @section Compiling @value{GDBN} in Another Directory
26313 If you want to run @value{GDBN} versions for several host or target machines,
26314 you need a different @code{gdb} compiled for each combination of
26315 host and target. @file{configure} is designed to make this easy by
26316 allowing you to generate each configuration in a separate subdirectory,
26317 rather than in the source directory. If your @code{make} program
26318 handles the @samp{VPATH} feature (@sc{gnu} @code{make} does), running
26319 @code{make} in each of these directories builds the @code{gdb}
26320 program specified there.
26322 To build @code{gdb} in a separate directory, run @file{configure}
26323 with the @samp{--srcdir} option to specify where to find the source.
26324 (You also need to specify a path to find @file{configure}
26325 itself from your working directory. If the path to @file{configure}
26326 would be the same as the argument to @samp{--srcdir}, you can leave out
26327 the @samp{--srcdir} option; it is assumed.)
26329 For example, with version @value{GDBVN}, you can build @value{GDBN} in a
26330 separate directory for a Sun 4 like this:
26334 cd gdb-@value{GDBVN}
26337 ../gdb-@value{GDBVN}/configure sun4
26342 When @file{configure} builds a configuration using a remote source
26343 directory, it creates a tree for the binaries with the same structure
26344 (and using the same names) as the tree under the source directory. In
26345 the example, you'd find the Sun 4 library @file{libiberty.a} in the
26346 directory @file{gdb-sun4/libiberty}, and @value{GDBN} itself in
26347 @file{gdb-sun4/gdb}.
26349 Make sure that your path to the @file{configure} script has just one
26350 instance of @file{gdb} in it. If your path to @file{configure} looks
26351 like @file{../gdb-@value{GDBVN}/gdb/configure}, you are configuring only
26352 one subdirectory of @value{GDBN}, not the whole package. This leads to
26353 build errors about missing include files such as @file{bfd/bfd.h}.
26355 One popular reason to build several @value{GDBN} configurations in separate
26356 directories is to configure @value{GDBN} for cross-compiling (where
26357 @value{GDBN} runs on one machine---the @dfn{host}---while debugging
26358 programs that run on another machine---the @dfn{target}).
26359 You specify a cross-debugging target by
26360 giving the @samp{--target=@var{target}} option to @file{configure}.
26362 When you run @code{make} to build a program or library, you must run
26363 it in a configured directory---whatever directory you were in when you
26364 called @file{configure} (or one of its subdirectories).
26366 The @code{Makefile} that @file{configure} generates in each source
26367 directory also runs recursively. If you type @code{make} in a source
26368 directory such as @file{gdb-@value{GDBVN}} (or in a separate configured
26369 directory configured with @samp{--srcdir=@var{dirname}/gdb-@value{GDBVN}}), you
26370 will build all the required libraries, and then build GDB.
26372 When you have multiple hosts or targets configured in separate
26373 directories, you can run @code{make} on them in parallel (for example,
26374 if they are NFS-mounted on each of the hosts); they will not interfere
26378 @section Specifying Names for Hosts and Targets
26380 The specifications used for hosts and targets in the @file{configure}
26381 script are based on a three-part naming scheme, but some short predefined
26382 aliases are also supported. The full naming scheme encodes three pieces
26383 of information in the following pattern:
26386 @var{architecture}-@var{vendor}-@var{os}
26389 For example, you can use the alias @code{sun4} as a @var{host} argument,
26390 or as the value for @var{target} in a @code{--target=@var{target}}
26391 option. The equivalent full name is @samp{sparc-sun-sunos4}.
26393 The @file{configure} script accompanying @value{GDBN} does not provide
26394 any query facility to list all supported host and target names or
26395 aliases. @file{configure} calls the Bourne shell script
26396 @code{config.sub} to map abbreviations to full names; you can read the
26397 script, if you wish, or you can use it to test your guesses on
26398 abbreviations---for example:
26401 % sh config.sub i386-linux
26403 % sh config.sub alpha-linux
26404 alpha-unknown-linux-gnu
26405 % sh config.sub hp9k700
26407 % sh config.sub sun4
26408 sparc-sun-sunos4.1.1
26409 % sh config.sub sun3
26410 m68k-sun-sunos4.1.1
26411 % sh config.sub i986v
26412 Invalid configuration `i986v': machine `i986v' not recognized
26416 @code{config.sub} is also distributed in the @value{GDBN} source
26417 directory (@file{gdb-@value{GDBVN}}, for version @value{GDBVN}).
26419 @node Configure Options
26420 @section @file{configure} Options
26422 Here is a summary of the @file{configure} options and arguments that
26423 are most often useful for building @value{GDBN}. @file{configure} also has
26424 several other options not listed here. @inforef{What Configure
26425 Does,,configure.info}, for a full explanation of @file{configure}.
26428 configure @r{[}--help@r{]}
26429 @r{[}--prefix=@var{dir}@r{]}
26430 @r{[}--exec-prefix=@var{dir}@r{]}
26431 @r{[}--srcdir=@var{dirname}@r{]}
26432 @r{[}--norecursion@r{]} @r{[}--rm@r{]}
26433 @r{[}--target=@var{target}@r{]}
26438 You may introduce options with a single @samp{-} rather than
26439 @samp{--} if you prefer; but you may abbreviate option names if you use
26444 Display a quick summary of how to invoke @file{configure}.
26446 @item --prefix=@var{dir}
26447 Configure the source to install programs and files under directory
26450 @item --exec-prefix=@var{dir}
26451 Configure the source to install programs under directory
26454 @c avoid splitting the warning from the explanation:
26456 @item --srcdir=@var{dirname}
26457 @strong{Warning: using this option requires @sc{gnu} @code{make}, or another
26458 @code{make} that implements the @code{VPATH} feature.}@*
26459 Use this option to make configurations in directories separate from the
26460 @value{GDBN} source directories. Among other things, you can use this to
26461 build (or maintain) several configurations simultaneously, in separate
26462 directories. @file{configure} writes configuration-specific files in
26463 the current directory, but arranges for them to use the source in the
26464 directory @var{dirname}. @file{configure} creates directories under
26465 the working directory in parallel to the source directories below
26468 @item --norecursion
26469 Configure only the directory level where @file{configure} is executed; do not
26470 propagate configuration to subdirectories.
26472 @item --target=@var{target}
26473 Configure @value{GDBN} for cross-debugging programs running on the specified
26474 @var{target}. Without this option, @value{GDBN} is configured to debug
26475 programs that run on the same machine (@var{host}) as @value{GDBN} itself.
26477 There is no convenient way to generate a list of all available targets.
26479 @item @var{host} @dots{}
26480 Configure @value{GDBN} to run on the specified @var{host}.
26482 There is no convenient way to generate a list of all available hosts.
26485 There are many other options available as well, but they are generally
26486 needed for special purposes only.
26488 @node System-wide configuration
26489 @section System-wide configuration and settings
26490 @cindex system-wide init file
26492 @value{GDBN} can be configured to have a system-wide init file;
26493 this file will be read and executed at startup (@pxref{Startup, , What
26494 @value{GDBN} does during startup}).
26496 Here is the corresponding configure option:
26499 @item --with-system-gdbinit=@var{file}
26500 Specify that the default location of the system-wide init file is
26504 If @value{GDBN} has been configured with the option @option{--prefix=$prefix},
26505 it may be subject to relocation. Two possible cases:
26509 If the default location of this init file contains @file{$prefix},
26510 it will be subject to relocation. Suppose that the configure options
26511 are @option{--prefix=$prefix --with-system-gdbinit=$prefix/etc/gdbinit};
26512 if @value{GDBN} is moved from @file{$prefix} to @file{$install}, the system
26513 init file is looked for as @file{$install/etc/gdbinit} instead of
26514 @file{$prefix/etc/gdbinit}.
26517 By contrast, if the default location does not contain the prefix,
26518 it will not be relocated. E.g.@: if @value{GDBN} has been configured with
26519 @option{--prefix=/usr/local --with-system-gdbinit=/usr/share/gdb/gdbinit},
26520 then @value{GDBN} will always look for @file{/usr/share/gdb/gdbinit},
26521 wherever @value{GDBN} is installed.
26524 @node Maintenance Commands
26525 @appendix Maintenance Commands
26526 @cindex maintenance commands
26527 @cindex internal commands
26529 In addition to commands intended for @value{GDBN} users, @value{GDBN}
26530 includes a number of commands intended for @value{GDBN} developers,
26531 that are not documented elsewhere in this manual. These commands are
26532 provided here for reference. (For commands that turn on debugging
26533 messages, see @ref{Debugging Output}.)
26536 @kindex maint agent
26537 @item maint agent @var{expression}
26538 Translate the given @var{expression} into remote agent bytecodes.
26539 This command is useful for debugging the Agent Expression mechanism
26540 (@pxref{Agent Expressions}).
26542 @kindex maint info breakpoints
26543 @item @anchor{maint info breakpoints}maint info breakpoints
26544 Using the same format as @samp{info breakpoints}, display both the
26545 breakpoints you've set explicitly, and those @value{GDBN} is using for
26546 internal purposes. Internal breakpoints are shown with negative
26547 breakpoint numbers. The type column identifies what kind of breakpoint
26552 Normal, explicitly set breakpoint.
26555 Normal, explicitly set watchpoint.
26558 Internal breakpoint, used to handle correctly stepping through
26559 @code{longjmp} calls.
26561 @item longjmp resume
26562 Internal breakpoint at the target of a @code{longjmp}.
26565 Temporary internal breakpoint used by the @value{GDBN} @code{until} command.
26568 Temporary internal breakpoint used by the @value{GDBN} @code{finish} command.
26571 Shared library events.
26575 @kindex set displaced-stepping
26576 @kindex show displaced-stepping
26577 @cindex displaced stepping support
26578 @cindex out-of-line single-stepping
26579 @item set displaced-stepping
26580 @itemx show displaced-stepping
26581 Control whether or not @value{GDBN} will do @dfn{displaced stepping}
26582 if the target supports it. Displaced stepping is a way to single-step
26583 over breakpoints without removing them from the inferior, by executing
26584 an out-of-line copy of the instruction that was originally at the
26585 breakpoint location. It is also known as out-of-line single-stepping.
26588 @item set displaced-stepping on
26589 If the target architecture supports it, @value{GDBN} will use
26590 displaced stepping to step over breakpoints.
26592 @item set displaced-stepping off
26593 @value{GDBN} will not use displaced stepping to step over breakpoints,
26594 even if such is supported by the target architecture.
26596 @cindex non-stop mode, and @samp{set displaced-stepping}
26597 @item set displaced-stepping auto
26598 This is the default mode. @value{GDBN} will use displaced stepping
26599 only if non-stop mode is active (@pxref{Non-Stop Mode}) and the target
26600 architecture supports displaced stepping.
26603 @kindex maint check-symtabs
26604 @item maint check-symtabs
26605 Check the consistency of psymtabs and symtabs.
26607 @kindex maint cplus first_component
26608 @item maint cplus first_component @var{name}
26609 Print the first C@t{++} class/namespace component of @var{name}.
26611 @kindex maint cplus namespace
26612 @item maint cplus namespace
26613 Print the list of possible C@t{++} namespaces.
26615 @kindex maint demangle
26616 @item maint demangle @var{name}
26617 Demangle a C@t{++} or Objective-C mangled @var{name}.
26619 @kindex maint deprecate
26620 @kindex maint undeprecate
26621 @cindex deprecated commands
26622 @item maint deprecate @var{command} @r{[}@var{replacement}@r{]}
26623 @itemx maint undeprecate @var{command}
26624 Deprecate or undeprecate the named @var{command}. Deprecated commands
26625 cause @value{GDBN} to issue a warning when you use them. The optional
26626 argument @var{replacement} says which newer command should be used in
26627 favor of the deprecated one; if it is given, @value{GDBN} will mention
26628 the replacement as part of the warning.
26630 @kindex maint dump-me
26631 @item maint dump-me
26632 @cindex @code{SIGQUIT} signal, dump core of @value{GDBN}
26633 Cause a fatal signal in the debugger and force it to dump its core.
26634 This is supported only on systems which support aborting a program
26635 with the @code{SIGQUIT} signal.
26637 @kindex maint internal-error
26638 @kindex maint internal-warning
26639 @item maint internal-error @r{[}@var{message-text}@r{]}
26640 @itemx maint internal-warning @r{[}@var{message-text}@r{]}
26641 Cause @value{GDBN} to call the internal function @code{internal_error}
26642 or @code{internal_warning} and hence behave as though an internal error
26643 or internal warning has been detected. In addition to reporting the
26644 internal problem, these functions give the user the opportunity to
26645 either quit @value{GDBN} or create a core file of the current
26646 @value{GDBN} session.
26648 These commands take an optional parameter @var{message-text} that is
26649 used as the text of the error or warning message.
26651 Here's an example of using @code{internal-error}:
26654 (@value{GDBP}) @kbd{maint internal-error testing, 1, 2}
26655 @dots{}/maint.c:121: internal-error: testing, 1, 2
26656 A problem internal to GDB has been detected. Further
26657 debugging may prove unreliable.
26658 Quit this debugging session? (y or n) @kbd{n}
26659 Create a core file? (y or n) @kbd{n}
26663 @cindex @value{GDBN} internal error
26664 @cindex internal errors, control of @value{GDBN} behavior
26666 @kindex maint set internal-error
26667 @kindex maint show internal-error
26668 @kindex maint set internal-warning
26669 @kindex maint show internal-warning
26670 @item maint set internal-error @var{action} [ask|yes|no]
26671 @itemx maint show internal-error @var{action}
26672 @itemx maint set internal-warning @var{action} [ask|yes|no]
26673 @itemx maint show internal-warning @var{action}
26674 When @value{GDBN} reports an internal problem (error or warning) it
26675 gives the user the opportunity to both quit @value{GDBN} and create a
26676 core file of the current @value{GDBN} session. These commands let you
26677 override the default behaviour for each particular @var{action},
26678 described in the table below.
26682 You can specify that @value{GDBN} should always (yes) or never (no)
26683 quit. The default is to ask the user what to do.
26686 You can specify that @value{GDBN} should always (yes) or never (no)
26687 create a core file. The default is to ask the user what to do.
26690 @kindex maint packet
26691 @item maint packet @var{text}
26692 If @value{GDBN} is talking to an inferior via the serial protocol,
26693 then this command sends the string @var{text} to the inferior, and
26694 displays the response packet. @value{GDBN} supplies the initial
26695 @samp{$} character, the terminating @samp{#} character, and the
26698 @kindex maint print architecture
26699 @item maint print architecture @r{[}@var{file}@r{]}
26700 Print the entire architecture configuration. The optional argument
26701 @var{file} names the file where the output goes.
26703 @kindex maint print c-tdesc
26704 @item maint print c-tdesc
26705 Print the current target description (@pxref{Target Descriptions}) as
26706 a C source file. The created source file can be used in @value{GDBN}
26707 when an XML parser is not available to parse the description.
26709 @kindex maint print dummy-frames
26710 @item maint print dummy-frames
26711 Prints the contents of @value{GDBN}'s internal dummy-frame stack.
26714 (@value{GDBP}) @kbd{b add}
26716 (@value{GDBP}) @kbd{print add(2,3)}
26717 Breakpoint 2, add (a=2, b=3) at @dots{}
26719 The program being debugged stopped while in a function called from GDB.
26721 (@value{GDBP}) @kbd{maint print dummy-frames}
26722 0x1a57c80: pc=0x01014068 fp=0x0200bddc sp=0x0200bdd6
26723 top=0x0200bdd4 id=@{stack=0x200bddc,code=0x101405c@}
26724 call_lo=0x01014000 call_hi=0x01014001
26728 Takes an optional file parameter.
26730 @kindex maint print registers
26731 @kindex maint print raw-registers
26732 @kindex maint print cooked-registers
26733 @kindex maint print register-groups
26734 @item maint print registers @r{[}@var{file}@r{]}
26735 @itemx maint print raw-registers @r{[}@var{file}@r{]}
26736 @itemx maint print cooked-registers @r{[}@var{file}@r{]}
26737 @itemx maint print register-groups @r{[}@var{file}@r{]}
26738 Print @value{GDBN}'s internal register data structures.
26740 The command @code{maint print raw-registers} includes the contents of
26741 the raw register cache; the command @code{maint print cooked-registers}
26742 includes the (cooked) value of all registers; and the command
26743 @code{maint print register-groups} includes the groups that each
26744 register is a member of. @xref{Registers,, Registers, gdbint,
26745 @value{GDBN} Internals}.
26747 These commands take an optional parameter, a file name to which to
26748 write the information.
26750 @kindex maint print reggroups
26751 @item maint print reggroups @r{[}@var{file}@r{]}
26752 Print @value{GDBN}'s internal register group data structures. The
26753 optional argument @var{file} tells to what file to write the
26756 The register groups info looks like this:
26759 (@value{GDBP}) @kbd{maint print reggroups}
26772 This command forces @value{GDBN} to flush its internal register cache.
26774 @kindex maint print objfiles
26775 @cindex info for known object files
26776 @item maint print objfiles
26777 Print a dump of all known object files. For each object file, this
26778 command prints its name, address in memory, and all of its psymtabs
26781 @kindex maint print statistics
26782 @cindex bcache statistics
26783 @item maint print statistics
26784 This command prints, for each object file in the program, various data
26785 about that object file followed by the byte cache (@dfn{bcache})
26786 statistics for the object file. The objfile data includes the number
26787 of minimal, partial, full, and stabs symbols, the number of types
26788 defined by the objfile, the number of as yet unexpanded psym tables,
26789 the number of line tables and string tables, and the amount of memory
26790 used by the various tables. The bcache statistics include the counts,
26791 sizes, and counts of duplicates of all and unique objects, max,
26792 average, and median entry size, total memory used and its overhead and
26793 savings, and various measures of the hash table size and chain
26796 @kindex maint print target-stack
26797 @cindex target stack description
26798 @item maint print target-stack
26799 A @dfn{target} is an interface between the debugger and a particular
26800 kind of file or process. Targets can be stacked in @dfn{strata},
26801 so that more than one target can potentially respond to a request.
26802 In particular, memory accesses will walk down the stack of targets
26803 until they find a target that is interested in handling that particular
26806 This command prints a short description of each layer that was pushed on
26807 the @dfn{target stack}, starting from the top layer down to the bottom one.
26809 @kindex maint print type
26810 @cindex type chain of a data type
26811 @item maint print type @var{expr}
26812 Print the type chain for a type specified by @var{expr}. The argument
26813 can be either a type name or a symbol. If it is a symbol, the type of
26814 that symbol is described. The type chain produced by this command is
26815 a recursive definition of the data type as stored in @value{GDBN}'s
26816 data structures, including its flags and contained types.
26818 @kindex maint set dwarf2 max-cache-age
26819 @kindex maint show dwarf2 max-cache-age
26820 @item maint set dwarf2 max-cache-age
26821 @itemx maint show dwarf2 max-cache-age
26822 Control the DWARF 2 compilation unit cache.
26824 @cindex DWARF 2 compilation units cache
26825 In object files with inter-compilation-unit references, such as those
26826 produced by the GCC option @samp{-feliminate-dwarf2-dups}, the DWARF 2
26827 reader needs to frequently refer to previously read compilation units.
26828 This setting controls how long a compilation unit will remain in the
26829 cache if it is not referenced. A higher limit means that cached
26830 compilation units will be stored in memory longer, and more total
26831 memory will be used. Setting it to zero disables caching, which will
26832 slow down @value{GDBN} startup, but reduce memory consumption.
26834 @kindex maint set profile
26835 @kindex maint show profile
26836 @cindex profiling GDB
26837 @item maint set profile
26838 @itemx maint show profile
26839 Control profiling of @value{GDBN}.
26841 Profiling will be disabled until you use the @samp{maint set profile}
26842 command to enable it. When you enable profiling, the system will begin
26843 collecting timing and execution count data; when you disable profiling or
26844 exit @value{GDBN}, the results will be written to a log file. Remember that
26845 if you use profiling, @value{GDBN} will overwrite the profiling log file
26846 (often called @file{gmon.out}). If you have a record of important profiling
26847 data in a @file{gmon.out} file, be sure to move it to a safe location.
26849 Configuring with @samp{--enable-profiling} arranges for @value{GDBN} to be
26850 compiled with the @samp{-pg} compiler option.
26852 @kindex maint set show-debug-regs
26853 @kindex maint show show-debug-regs
26854 @cindex hardware debug registers
26855 @item maint set show-debug-regs
26856 @itemx maint show show-debug-regs
26857 Control whether to show variables that mirror the hardware debug
26858 registers. Use @code{ON} to enable, @code{OFF} to disable. If
26859 enabled, the debug registers values are shown when @value{GDBN} inserts or
26860 removes a hardware breakpoint or watchpoint, and when the inferior
26861 triggers a hardware-assisted breakpoint or watchpoint.
26863 @kindex maint space
26864 @cindex memory used by commands
26866 Control whether to display memory usage for each command. If set to a
26867 nonzero value, @value{GDBN} will display how much memory each command
26868 took, following the command's own output. This can also be requested
26869 by invoking @value{GDBN} with the @option{--statistics} command-line
26870 switch (@pxref{Mode Options}).
26873 @cindex time of command execution
26875 Control whether to display the execution time for each command. If
26876 set to a nonzero value, @value{GDBN} will display how much time it
26877 took to execute each command, following the command's own output.
26878 The time is not printed for the commands that run the target, since
26879 there's no mechanism currently to compute how much time was spend
26880 by @value{GDBN} and how much time was spend by the program been debugged.
26881 it's not possibly currently
26882 This can also be requested by invoking @value{GDBN} with the
26883 @option{--statistics} command-line switch (@pxref{Mode Options}).
26885 @kindex maint translate-address
26886 @item maint translate-address @r{[}@var{section}@r{]} @var{addr}
26887 Find the symbol stored at the location specified by the address
26888 @var{addr} and an optional section name @var{section}. If found,
26889 @value{GDBN} prints the name of the closest symbol and an offset from
26890 the symbol's location to the specified address. This is similar to
26891 the @code{info address} command (@pxref{Symbols}), except that this
26892 command also allows to find symbols in other sections.
26894 If section was not specified, the section in which the symbol was found
26895 is also printed. For dynamically linked executables, the name of
26896 executable or shared library containing the symbol is printed as well.
26900 The following command is useful for non-interactive invocations of
26901 @value{GDBN}, such as in the test suite.
26904 @item set watchdog @var{nsec}
26905 @kindex set watchdog
26906 @cindex watchdog timer
26907 @cindex timeout for commands
26908 Set the maximum number of seconds @value{GDBN} will wait for the
26909 target operation to finish. If this time expires, @value{GDBN}
26910 reports and error and the command is aborted.
26912 @item show watchdog
26913 Show the current setting of the target wait timeout.
26916 @node Remote Protocol
26917 @appendix @value{GDBN} Remote Serial Protocol
26922 * Stop Reply Packets::
26923 * General Query Packets::
26924 * Register Packet Format::
26925 * Tracepoint Packets::
26926 * Host I/O Packets::
26928 * Notification Packets::
26929 * Remote Non-Stop::
26930 * Packet Acknowledgment::
26932 * File-I/O Remote Protocol Extension::
26933 * Library List Format::
26934 * Memory Map Format::
26940 There may be occasions when you need to know something about the
26941 protocol---for example, if there is only one serial port to your target
26942 machine, you might want your program to do something special if it
26943 recognizes a packet meant for @value{GDBN}.
26945 In the examples below, @samp{->} and @samp{<-} are used to indicate
26946 transmitted and received data, respectively.
26948 @cindex protocol, @value{GDBN} remote serial
26949 @cindex serial protocol, @value{GDBN} remote
26950 @cindex remote serial protocol
26951 All @value{GDBN} commands and responses (other than acknowledgments
26952 and notifications, see @ref{Notification Packets}) are sent as a
26953 @var{packet}. A @var{packet} is introduced with the character
26954 @samp{$}, the actual @var{packet-data}, and the terminating character
26955 @samp{#} followed by a two-digit @var{checksum}:
26958 @code{$}@var{packet-data}@code{#}@var{checksum}
26962 @cindex checksum, for @value{GDBN} remote
26964 The two-digit @var{checksum} is computed as the modulo 256 sum of all
26965 characters between the leading @samp{$} and the trailing @samp{#} (an
26966 eight bit unsigned checksum).
26968 Implementors should note that prior to @value{GDBN} 5.0 the protocol
26969 specification also included an optional two-digit @var{sequence-id}:
26972 @code{$}@var{sequence-id}@code{:}@var{packet-data}@code{#}@var{checksum}
26975 @cindex sequence-id, for @value{GDBN} remote
26977 That @var{sequence-id} was appended to the acknowledgment. @value{GDBN}
26978 has never output @var{sequence-id}s. Stubs that handle packets added
26979 since @value{GDBN} 5.0 must not accept @var{sequence-id}.
26981 When either the host or the target machine receives a packet, the first
26982 response expected is an acknowledgment: either @samp{+} (to indicate
26983 the package was received correctly) or @samp{-} (to request
26987 -> @code{$}@var{packet-data}@code{#}@var{checksum}
26992 The @samp{+}/@samp{-} acknowledgments can be disabled
26993 once a connection is established.
26994 @xref{Packet Acknowledgment}, for details.
26996 The host (@value{GDBN}) sends @var{command}s, and the target (the
26997 debugging stub incorporated in your program) sends a @var{response}. In
26998 the case of step and continue @var{command}s, the response is only sent
26999 when the operation has completed, and the target has again stopped all
27000 threads in all attached processes. This is the default all-stop mode
27001 behavior, but the remote protocol also supports @value{GDBN}'s non-stop
27002 execution mode; see @ref{Remote Non-Stop}, for details.
27004 @var{packet-data} consists of a sequence of characters with the
27005 exception of @samp{#} and @samp{$} (see @samp{X} packet for additional
27008 @cindex remote protocol, field separator
27009 Fields within the packet should be separated using @samp{,} @samp{;} or
27010 @samp{:}. Except where otherwise noted all numbers are represented in
27011 @sc{hex} with leading zeros suppressed.
27013 Implementors should note that prior to @value{GDBN} 5.0, the character
27014 @samp{:} could not appear as the third character in a packet (as it
27015 would potentially conflict with the @var{sequence-id}).
27017 @cindex remote protocol, binary data
27018 @anchor{Binary Data}
27019 Binary data in most packets is encoded either as two hexadecimal
27020 digits per byte of binary data. This allowed the traditional remote
27021 protocol to work over connections which were only seven-bit clean.
27022 Some packets designed more recently assume an eight-bit clean
27023 connection, and use a more efficient encoding to send and receive
27026 The binary data representation uses @code{7d} (@sc{ascii} @samp{@}})
27027 as an escape character. Any escaped byte is transmitted as the escape
27028 character followed by the original character XORed with @code{0x20}.
27029 For example, the byte @code{0x7d} would be transmitted as the two
27030 bytes @code{0x7d 0x5d}. The bytes @code{0x23} (@sc{ascii} @samp{#}),
27031 @code{0x24} (@sc{ascii} @samp{$}), and @code{0x7d} (@sc{ascii}
27032 @samp{@}}) must always be escaped. Responses sent by the stub
27033 must also escape @code{0x2a} (@sc{ascii} @samp{*}), so that it
27034 is not interpreted as the start of a run-length encoded sequence
27037 Response @var{data} can be run-length encoded to save space.
27038 Run-length encoding replaces runs of identical characters with one
27039 instance of the repeated character, followed by a @samp{*} and a
27040 repeat count. The repeat count is itself sent encoded, to avoid
27041 binary characters in @var{data}: a value of @var{n} is sent as
27042 @code{@var{n}+29}. For a repeat count greater or equal to 3, this
27043 produces a printable @sc{ascii} character, e.g.@: a space (@sc{ascii}
27044 code 32) for a repeat count of 3. (This is because run-length
27045 encoding starts to win for counts 3 or more.) Thus, for example,
27046 @samp{0* } is a run-length encoding of ``0000'': the space character
27047 after @samp{*} means repeat the leading @code{0} @w{@code{32 - 29 =
27050 The printable characters @samp{#} and @samp{$} or with a numeric value
27051 greater than 126 must not be used. Runs of six repeats (@samp{#}) or
27052 seven repeats (@samp{$}) can be expanded using a repeat count of only
27053 five (@samp{"}). For example, @samp{00000000} can be encoded as
27056 The error response returned for some packets includes a two character
27057 error number. That number is not well defined.
27059 @cindex empty response, for unsupported packets
27060 For any @var{command} not supported by the stub, an empty response
27061 (@samp{$#00}) should be returned. That way it is possible to extend the
27062 protocol. A newer @value{GDBN} can tell if a packet is supported based
27065 A stub is required to support the @samp{g}, @samp{G}, @samp{m}, @samp{M},
27066 @samp{c}, and @samp{s} @var{command}s. All other @var{command}s are
27072 The following table provides a complete list of all currently defined
27073 @var{command}s and their corresponding response @var{data}.
27074 @xref{File-I/O Remote Protocol Extension}, for details about the File
27075 I/O extension of the remote protocol.
27077 Each packet's description has a template showing the packet's overall
27078 syntax, followed by an explanation of the packet's meaning. We
27079 include spaces in some of the templates for clarity; these are not
27080 part of the packet's syntax. No @value{GDBN} packet uses spaces to
27081 separate its components. For example, a template like @samp{foo
27082 @var{bar} @var{baz}} describes a packet beginning with the three ASCII
27083 bytes @samp{foo}, followed by a @var{bar}, followed directly by a
27084 @var{baz}. @value{GDBN} does not transmit a space character between the
27085 @samp{foo} and the @var{bar}, or between the @var{bar} and the
27088 @cindex @var{thread-id}, in remote protocol
27089 @anchor{thread-id syntax}
27090 Several packets and replies include a @var{thread-id} field to identify
27091 a thread. Normally these are positive numbers with a target-specific
27092 interpretation, formatted as big-endian hex strings. A @var{thread-id}
27093 can also be a literal @samp{-1} to indicate all threads, or @samp{0} to
27096 In addition, the remote protocol supports a multiprocess feature in
27097 which the @var{thread-id} syntax is extended to optionally include both
27098 process and thread ID fields, as @samp{p@var{pid}.@var{tid}}.
27099 The @var{pid} (process) and @var{tid} (thread) components each have the
27100 format described above: a positive number with target-specific
27101 interpretation formatted as a big-endian hex string, literal @samp{-1}
27102 to indicate all processes or threads (respectively), or @samp{0} to
27103 indicate an arbitrary process or thread. Specifying just a process, as
27104 @samp{p@var{pid}}, is equivalent to @samp{p@var{pid}.-1}. It is an
27105 error to specify all processes but a specific thread, such as
27106 @samp{p-1.@var{tid}}. Note that the @samp{p} prefix is @emph{not} used
27107 for those packets and replies explicitly documented to include a process
27108 ID, rather than a @var{thread-id}.
27110 The multiprocess @var{thread-id} syntax extensions are only used if both
27111 @value{GDBN} and the stub report support for the @samp{multiprocess}
27112 feature using @samp{qSupported}. @xref{multiprocess extensions}, for
27115 Note that all packet forms beginning with an upper- or lower-case
27116 letter, other than those described here, are reserved for future use.
27118 Here are the packet descriptions.
27123 @cindex @samp{!} packet
27124 @anchor{extended mode}
27125 Enable extended mode. In extended mode, the remote server is made
27126 persistent. The @samp{R} packet is used to restart the program being
27132 The remote target both supports and has enabled extended mode.
27136 @cindex @samp{?} packet
27137 Indicate the reason the target halted. The reply is the same as for
27138 step and continue. This packet has a special interpretation when the
27139 target is in non-stop mode; see @ref{Remote Non-Stop}.
27142 @xref{Stop Reply Packets}, for the reply specifications.
27144 @item A @var{arglen},@var{argnum},@var{arg},@dots{}
27145 @cindex @samp{A} packet
27146 Initialized @code{argv[]} array passed into program. @var{arglen}
27147 specifies the number of bytes in the hex encoded byte stream
27148 @var{arg}. See @code{gdbserver} for more details.
27153 The arguments were set.
27159 @cindex @samp{b} packet
27160 (Don't use this packet; its behavior is not well-defined.)
27161 Change the serial line speed to @var{baud}.
27163 JTC: @emph{When does the transport layer state change? When it's
27164 received, or after the ACK is transmitted. In either case, there are
27165 problems if the command or the acknowledgment packet is dropped.}
27167 Stan: @emph{If people really wanted to add something like this, and get
27168 it working for the first time, they ought to modify ser-unix.c to send
27169 some kind of out-of-band message to a specially-setup stub and have the
27170 switch happen "in between" packets, so that from remote protocol's point
27171 of view, nothing actually happened.}
27173 @item B @var{addr},@var{mode}
27174 @cindex @samp{B} packet
27175 Set (@var{mode} is @samp{S}) or clear (@var{mode} is @samp{C}) a
27176 breakpoint at @var{addr}.
27178 Don't use this packet. Use the @samp{Z} and @samp{z} packets instead
27179 (@pxref{insert breakpoint or watchpoint packet}).
27182 @cindex @samp{bc} packet
27183 Backward continue. Execute the target system in reverse. No parameter.
27184 @xref{Reverse Execution}, for more information.
27187 @xref{Stop Reply Packets}, for the reply specifications.
27190 @cindex @samp{bs} packet
27191 Backward single step. Execute one instruction in reverse. No parameter.
27192 @xref{Reverse Execution}, for more information.
27195 @xref{Stop Reply Packets}, for the reply specifications.
27197 @item c @r{[}@var{addr}@r{]}
27198 @cindex @samp{c} packet
27199 Continue. @var{addr} is address to resume. If @var{addr} is omitted,
27200 resume at current address.
27203 @xref{Stop Reply Packets}, for the reply specifications.
27205 @item C @var{sig}@r{[};@var{addr}@r{]}
27206 @cindex @samp{C} packet
27207 Continue with signal @var{sig} (hex signal number). If
27208 @samp{;@var{addr}} is omitted, resume at same address.
27211 @xref{Stop Reply Packets}, for the reply specifications.
27214 @cindex @samp{d} packet
27217 Don't use this packet; instead, define a general set packet
27218 (@pxref{General Query Packets}).
27222 @cindex @samp{D} packet
27223 The first form of the packet is used to detach @value{GDBN} from the
27224 remote system. It is sent to the remote target
27225 before @value{GDBN} disconnects via the @code{detach} command.
27227 The second form, including a process ID, is used when multiprocess
27228 protocol extensions are enabled (@pxref{multiprocess extensions}), to
27229 detach only a specific process. The @var{pid} is specified as a
27230 big-endian hex string.
27240 @item F @var{RC},@var{EE},@var{CF};@var{XX}
27241 @cindex @samp{F} packet
27242 A reply from @value{GDBN} to an @samp{F} packet sent by the target.
27243 This is part of the File-I/O protocol extension. @xref{File-I/O
27244 Remote Protocol Extension}, for the specification.
27247 @anchor{read registers packet}
27248 @cindex @samp{g} packet
27249 Read general registers.
27253 @item @var{XX@dots{}}
27254 Each byte of register data is described by two hex digits. The bytes
27255 with the register are transmitted in target byte order. The size of
27256 each register and their position within the @samp{g} packet are
27257 determined by the @value{GDBN} internal gdbarch functions
27258 @code{DEPRECATED_REGISTER_RAW_SIZE} and @code{gdbarch_register_name}. The
27259 specification of several standard @samp{g} packets is specified below.
27264 @item G @var{XX@dots{}}
27265 @cindex @samp{G} packet
27266 Write general registers. @xref{read registers packet}, for a
27267 description of the @var{XX@dots{}} data.
27277 @item H @var{c} @var{thread-id}
27278 @cindex @samp{H} packet
27279 Set thread for subsequent operations (@samp{m}, @samp{M}, @samp{g},
27280 @samp{G}, et.al.). @var{c} depends on the operation to be performed: it
27281 should be @samp{c} for step and continue operations, @samp{g} for other
27282 operations. The thread designator @var{thread-id} has the format and
27283 interpretation described in @ref{thread-id syntax}.
27294 @c 'H': How restrictive (or permissive) is the thread model. If a
27295 @c thread is selected and stopped, are other threads allowed
27296 @c to continue to execute? As I mentioned above, I think the
27297 @c semantics of each command when a thread is selected must be
27298 @c described. For example:
27300 @c 'g': If the stub supports threads and a specific thread is
27301 @c selected, returns the register block from that thread;
27302 @c otherwise returns current registers.
27304 @c 'G' If the stub supports threads and a specific thread is
27305 @c selected, sets the registers of the register block of
27306 @c that thread; otherwise sets current registers.
27308 @item i @r{[}@var{addr}@r{[},@var{nnn}@r{]]}
27309 @anchor{cycle step packet}
27310 @cindex @samp{i} packet
27311 Step the remote target by a single clock cycle. If @samp{,@var{nnn}} is
27312 present, cycle step @var{nnn} cycles. If @var{addr} is present, cycle
27313 step starting at that address.
27316 @cindex @samp{I} packet
27317 Signal, then cycle step. @xref{step with signal packet}. @xref{cycle
27321 @cindex @samp{k} packet
27324 FIXME: @emph{There is no description of how to operate when a specific
27325 thread context has been selected (i.e.@: does 'k' kill only that
27328 @item m @var{addr},@var{length}
27329 @cindex @samp{m} packet
27330 Read @var{length} bytes of memory starting at address @var{addr}.
27331 Note that @var{addr} may not be aligned to any particular boundary.
27333 The stub need not use any particular size or alignment when gathering
27334 data from memory for the response; even if @var{addr} is word-aligned
27335 and @var{length} is a multiple of the word size, the stub is free to
27336 use byte accesses, or not. For this reason, this packet may not be
27337 suitable for accessing memory-mapped I/O devices.
27338 @cindex alignment of remote memory accesses
27339 @cindex size of remote memory accesses
27340 @cindex memory, alignment and size of remote accesses
27344 @item @var{XX@dots{}}
27345 Memory contents; each byte is transmitted as a two-digit hexadecimal
27346 number. The reply may contain fewer bytes than requested if the
27347 server was able to read only part of the region of memory.
27352 @item M @var{addr},@var{length}:@var{XX@dots{}}
27353 @cindex @samp{M} packet
27354 Write @var{length} bytes of memory starting at address @var{addr}.
27355 @var{XX@dots{}} is the data; each byte is transmitted as a two-digit
27356 hexadecimal number.
27363 for an error (this includes the case where only part of the data was
27368 @cindex @samp{p} packet
27369 Read the value of register @var{n}; @var{n} is in hex.
27370 @xref{read registers packet}, for a description of how the returned
27371 register value is encoded.
27375 @item @var{XX@dots{}}
27376 the register's value
27380 Indicating an unrecognized @var{query}.
27383 @item P @var{n@dots{}}=@var{r@dots{}}
27384 @anchor{write register packet}
27385 @cindex @samp{P} packet
27386 Write register @var{n@dots{}} with value @var{r@dots{}}. The register
27387 number @var{n} is in hexadecimal, and @var{r@dots{}} contains two hex
27388 digits for each byte in the register (target byte order).
27398 @item q @var{name} @var{params}@dots{}
27399 @itemx Q @var{name} @var{params}@dots{}
27400 @cindex @samp{q} packet
27401 @cindex @samp{Q} packet
27402 General query (@samp{q}) and set (@samp{Q}). These packets are
27403 described fully in @ref{General Query Packets}.
27406 @cindex @samp{r} packet
27407 Reset the entire system.
27409 Don't use this packet; use the @samp{R} packet instead.
27412 @cindex @samp{R} packet
27413 Restart the program being debugged. @var{XX}, while needed, is ignored.
27414 This packet is only available in extended mode (@pxref{extended mode}).
27416 The @samp{R} packet has no reply.
27418 @item s @r{[}@var{addr}@r{]}
27419 @cindex @samp{s} packet
27420 Single step. @var{addr} is the address at which to resume. If
27421 @var{addr} is omitted, resume at same address.
27424 @xref{Stop Reply Packets}, for the reply specifications.
27426 @item S @var{sig}@r{[};@var{addr}@r{]}
27427 @anchor{step with signal packet}
27428 @cindex @samp{S} packet
27429 Step with signal. This is analogous to the @samp{C} packet, but
27430 requests a single-step, rather than a normal resumption of execution.
27433 @xref{Stop Reply Packets}, for the reply specifications.
27435 @item t @var{addr}:@var{PP},@var{MM}
27436 @cindex @samp{t} packet
27437 Search backwards starting at address @var{addr} for a match with pattern
27438 @var{PP} and mask @var{MM}. @var{PP} and @var{MM} are 4 bytes.
27439 @var{addr} must be at least 3 digits.
27441 @item T @var{thread-id}
27442 @cindex @samp{T} packet
27443 Find out if the thread @var{thread-id} is alive. @xref{thread-id syntax}.
27448 thread is still alive
27454 Packets starting with @samp{v} are identified by a multi-letter name,
27455 up to the first @samp{;} or @samp{?} (or the end of the packet).
27457 @item vAttach;@var{pid}
27458 @cindex @samp{vAttach} packet
27459 Attach to a new process with the specified process ID @var{pid}.
27460 The process ID is a
27461 hexadecimal integer identifying the process. In all-stop mode, all
27462 threads in the attached process are stopped; in non-stop mode, it may be
27463 attached without being stopped if that is supported by the target.
27465 @c In non-stop mode, on a successful vAttach, the stub should set the
27466 @c current thread to a thread of the newly-attached process. After
27467 @c attaching, GDB queries for the attached process's thread ID with qC.
27468 @c Also note that, from a user perspective, whether or not the
27469 @c target is stopped on attach in non-stop mode depends on whether you
27470 @c use the foreground or background version of the attach command, not
27471 @c on what vAttach does; GDB does the right thing with respect to either
27472 @c stopping or restarting threads.
27474 This packet is only available in extended mode (@pxref{extended mode}).
27480 @item @r{Any stop packet}
27481 for success in all-stop mode (@pxref{Stop Reply Packets})
27483 for success in non-stop mode (@pxref{Remote Non-Stop})
27486 @item vCont@r{[};@var{action}@r{[}:@var{thread-id}@r{]]}@dots{}
27487 @cindex @samp{vCont} packet
27488 Resume the inferior, specifying different actions for each thread.
27489 If an action is specified with no @var{thread-id}, then it is applied to any
27490 threads that don't have a specific action specified; if no default action is
27491 specified then other threads should remain stopped in all-stop mode and
27492 in their current state in non-stop mode.
27493 Specifying multiple
27494 default actions is an error; specifying no actions is also an error.
27495 Thread IDs are specified using the syntax described in @ref{thread-id syntax}.
27497 Currently supported actions are:
27503 Continue with signal @var{sig}. The signal @var{sig} should be two hex digits.
27507 Step with signal @var{sig}. The signal @var{sig} should be two hex digits.
27511 Stop with signal @var{sig}. The signal @var{sig} should be two hex digits.
27514 The optional argument @var{addr} normally associated with the
27515 @samp{c}, @samp{C}, @samp{s}, and @samp{S} packets is
27516 not supported in @samp{vCont}.
27518 The @samp{t} and @samp{T} actions are only relevant in non-stop mode
27519 (@pxref{Remote Non-Stop}) and may be ignored by the stub otherwise.
27520 A stop reply should be generated for any affected thread not already stopped.
27521 When a thread is stopped by means of a @samp{t} action,
27522 the corresponding stop reply should indicate that the thread has stopped with
27523 signal @samp{0}, regardless of whether the target uses some other signal
27524 as an implementation detail.
27527 @xref{Stop Reply Packets}, for the reply specifications.
27530 @cindex @samp{vCont?} packet
27531 Request a list of actions supported by the @samp{vCont} packet.
27535 @item vCont@r{[};@var{action}@dots{}@r{]}
27536 The @samp{vCont} packet is supported. Each @var{action} is a supported
27537 command in the @samp{vCont} packet.
27539 The @samp{vCont} packet is not supported.
27542 @item vFile:@var{operation}:@var{parameter}@dots{}
27543 @cindex @samp{vFile} packet
27544 Perform a file operation on the target system. For details,
27545 see @ref{Host I/O Packets}.
27547 @item vFlashErase:@var{addr},@var{length}
27548 @cindex @samp{vFlashErase} packet
27549 Direct the stub to erase @var{length} bytes of flash starting at
27550 @var{addr}. The region may enclose any number of flash blocks, but
27551 its start and end must fall on block boundaries, as indicated by the
27552 flash block size appearing in the memory map (@pxref{Memory Map
27553 Format}). @value{GDBN} groups flash memory programming operations
27554 together, and sends a @samp{vFlashDone} request after each group; the
27555 stub is allowed to delay erase operation until the @samp{vFlashDone}
27556 packet is received.
27558 The stub must support @samp{vCont} if it reports support for
27559 multiprocess extensions (@pxref{multiprocess extensions}). Note that in
27560 this case @samp{vCont} actions can be specified to apply to all threads
27561 in a process by using the @samp{p@var{pid}.-1} form of the
27572 @item vFlashWrite:@var{addr}:@var{XX@dots{}}
27573 @cindex @samp{vFlashWrite} packet
27574 Direct the stub to write data to flash address @var{addr}. The data
27575 is passed in binary form using the same encoding as for the @samp{X}
27576 packet (@pxref{Binary Data}). The memory ranges specified by
27577 @samp{vFlashWrite} packets preceding a @samp{vFlashDone} packet must
27578 not overlap, and must appear in order of increasing addresses
27579 (although @samp{vFlashErase} packets for higher addresses may already
27580 have been received; the ordering is guaranteed only between
27581 @samp{vFlashWrite} packets). If a packet writes to an address that was
27582 neither erased by a preceding @samp{vFlashErase} packet nor by some other
27583 target-specific method, the results are unpredictable.
27591 for vFlashWrite addressing non-flash memory
27597 @cindex @samp{vFlashDone} packet
27598 Indicate to the stub that flash programming operation is finished.
27599 The stub is permitted to delay or batch the effects of a group of
27600 @samp{vFlashErase} and @samp{vFlashWrite} packets until a
27601 @samp{vFlashDone} packet is received. The contents of the affected
27602 regions of flash memory are unpredictable until the @samp{vFlashDone}
27603 request is completed.
27605 @item vKill;@var{pid}
27606 @cindex @samp{vKill} packet
27607 Kill the process with the specified process ID. @var{pid} is a
27608 hexadecimal integer identifying the process. This packet is used in
27609 preference to @samp{k} when multiprocess protocol extensions are
27610 supported; see @ref{multiprocess extensions}.
27620 @item vRun;@var{filename}@r{[};@var{argument}@r{]}@dots{}
27621 @cindex @samp{vRun} packet
27622 Run the program @var{filename}, passing it each @var{argument} on its
27623 command line. The file and arguments are hex-encoded strings. If
27624 @var{filename} is an empty string, the stub may use a default program
27625 (e.g.@: the last program run). The program is created in the stopped
27628 @c FIXME: What about non-stop mode?
27630 This packet is only available in extended mode (@pxref{extended mode}).
27636 @item @r{Any stop packet}
27637 for success (@pxref{Stop Reply Packets})
27641 @anchor{vStopped packet}
27642 @cindex @samp{vStopped} packet
27644 In non-stop mode (@pxref{Remote Non-Stop}), acknowledge a previous stop
27645 reply and prompt for the stub to report another one.
27649 @item @r{Any stop packet}
27650 if there is another unreported stop event (@pxref{Stop Reply Packets})
27652 if there are no unreported stop events
27655 @item X @var{addr},@var{length}:@var{XX@dots{}}
27657 @cindex @samp{X} packet
27658 Write data to memory, where the data is transmitted in binary.
27659 @var{addr} is address, @var{length} is number of bytes,
27660 @samp{@var{XX}@dots{}} is binary data (@pxref{Binary Data}).
27670 @item z @var{type},@var{addr},@var{length}
27671 @itemx Z @var{type},@var{addr},@var{length}
27672 @anchor{insert breakpoint or watchpoint packet}
27673 @cindex @samp{z} packet
27674 @cindex @samp{Z} packets
27675 Insert (@samp{Z}) or remove (@samp{z}) a @var{type} breakpoint or
27676 watchpoint starting at address @var{address} and covering the next
27677 @var{length} bytes.
27679 Each breakpoint and watchpoint packet @var{type} is documented
27682 @emph{Implementation notes: A remote target shall return an empty string
27683 for an unrecognized breakpoint or watchpoint packet @var{type}. A
27684 remote target shall support either both or neither of a given
27685 @samp{Z@var{type}@dots{}} and @samp{z@var{type}@dots{}} packet pair. To
27686 avoid potential problems with duplicate packets, the operations should
27687 be implemented in an idempotent way.}
27689 @item z0,@var{addr},@var{length}
27690 @itemx Z0,@var{addr},@var{length}
27691 @cindex @samp{z0} packet
27692 @cindex @samp{Z0} packet
27693 Insert (@samp{Z0}) or remove (@samp{z0}) a memory breakpoint at address
27694 @var{addr} of size @var{length}.
27696 A memory breakpoint is implemented by replacing the instruction at
27697 @var{addr} with a software breakpoint or trap instruction. The
27698 @var{length} is used by targets that indicates the size of the
27699 breakpoint (in bytes) that should be inserted (e.g., the @sc{arm} and
27700 @sc{mips} can insert either a 2 or 4 byte breakpoint).
27702 @emph{Implementation note: It is possible for a target to copy or move
27703 code that contains memory breakpoints (e.g., when implementing
27704 overlays). The behavior of this packet, in the presence of such a
27705 target, is not defined.}
27717 @item z1,@var{addr},@var{length}
27718 @itemx Z1,@var{addr},@var{length}
27719 @cindex @samp{z1} packet
27720 @cindex @samp{Z1} packet
27721 Insert (@samp{Z1}) or remove (@samp{z1}) a hardware breakpoint at
27722 address @var{addr} of size @var{length}.
27724 A hardware breakpoint is implemented using a mechanism that is not
27725 dependant on being able to modify the target's memory.
27727 @emph{Implementation note: A hardware breakpoint is not affected by code
27740 @item z2,@var{addr},@var{length}
27741 @itemx Z2,@var{addr},@var{length}
27742 @cindex @samp{z2} packet
27743 @cindex @samp{Z2} packet
27744 Insert (@samp{Z2}) or remove (@samp{z2}) a write watchpoint.
27756 @item z3,@var{addr},@var{length}
27757 @itemx Z3,@var{addr},@var{length}
27758 @cindex @samp{z3} packet
27759 @cindex @samp{Z3} packet
27760 Insert (@samp{Z3}) or remove (@samp{z3}) a read watchpoint.
27772 @item z4,@var{addr},@var{length}
27773 @itemx Z4,@var{addr},@var{length}
27774 @cindex @samp{z4} packet
27775 @cindex @samp{Z4} packet
27776 Insert (@samp{Z4}) or remove (@samp{z4}) an access watchpoint.
27790 @node Stop Reply Packets
27791 @section Stop Reply Packets
27792 @cindex stop reply packets
27794 The @samp{C}, @samp{c}, @samp{S}, @samp{s}, @samp{vCont},
27795 @samp{vAttach}, @samp{vRun}, @samp{vStopped}, and @samp{?} packets can
27796 receive any of the below as a reply. Except for @samp{?}
27797 and @samp{vStopped}, that reply is only returned
27798 when the target halts. In the below the exact meaning of @dfn{signal
27799 number} is defined by the header @file{include/gdb/signals.h} in the
27800 @value{GDBN} source code.
27802 As in the description of request packets, we include spaces in the
27803 reply templates for clarity; these are not part of the reply packet's
27804 syntax. No @value{GDBN} stop reply packet uses spaces to separate its
27810 The program received signal number @var{AA} (a two-digit hexadecimal
27811 number). This is equivalent to a @samp{T} response with no
27812 @var{n}:@var{r} pairs.
27814 @item T @var{AA} @var{n1}:@var{r1};@var{n2}:@var{r2};@dots{}
27815 @cindex @samp{T} packet reply
27816 The program received signal number @var{AA} (a two-digit hexadecimal
27817 number). This is equivalent to an @samp{S} response, except that the
27818 @samp{@var{n}:@var{r}} pairs can carry values of important registers
27819 and other information directly in the stop reply packet, reducing
27820 round-trip latency. Single-step and breakpoint traps are reported
27821 this way. Each @samp{@var{n}:@var{r}} pair is interpreted as follows:
27825 If @var{n} is a hexadecimal number, it is a register number, and the
27826 corresponding @var{r} gives that register's value. @var{r} is a
27827 series of bytes in target byte order, with each byte given by a
27828 two-digit hex number.
27831 If @var{n} is @samp{thread}, then @var{r} is the @var{thread-id} of
27832 the stopped thread, as specified in @ref{thread-id syntax}.
27835 If @var{n} is a recognized @dfn{stop reason}, it describes a more
27836 specific event that stopped the target. The currently defined stop
27837 reasons are listed below. @var{aa} should be @samp{05}, the trap
27838 signal. At most one stop reason should be present.
27841 Otherwise, @value{GDBN} should ignore this @samp{@var{n}:@var{r}} pair
27842 and go on to the next; this allows us to extend the protocol in the
27846 The currently defined stop reasons are:
27852 The packet indicates a watchpoint hit, and @var{r} is the data address, in
27855 @cindex shared library events, remote reply
27857 The packet indicates that the loaded libraries have changed.
27858 @value{GDBN} should use @samp{qXfer:libraries:read} to fetch a new
27859 list of loaded libraries. @var{r} is ignored.
27861 @cindex replay log events, remote reply
27863 The packet indicates that the target cannot continue replaying
27864 logged execution events, because it has reached the end (or the
27865 beginning when executing backward) of the log. The value of @var{r}
27866 will be either @samp{begin} or @samp{end}. @xref{Reverse Execution},
27867 for more information.
27873 @itemx W @var{AA} ; process:@var{pid}
27874 The process exited, and @var{AA} is the exit status. This is only
27875 applicable to certain targets.
27877 The second form of the response, including the process ID of the exited
27878 process, can be used only when @value{GDBN} has reported support for
27879 multiprocess protocol extensions; see @ref{multiprocess extensions}.
27880 The @var{pid} is formatted as a big-endian hex string.
27883 @itemx X @var{AA} ; process:@var{pid}
27884 The process terminated with signal @var{AA}.
27886 The second form of the response, including the process ID of the
27887 terminated process, can be used only when @value{GDBN} has reported
27888 support for multiprocess protocol extensions; see @ref{multiprocess
27889 extensions}. The @var{pid} is formatted as a big-endian hex string.
27891 @item O @var{XX}@dots{}
27892 @samp{@var{XX}@dots{}} is hex encoding of @sc{ascii} data, to be
27893 written as the program's console output. This can happen at any time
27894 while the program is running and the debugger should continue to wait
27895 for @samp{W}, @samp{T}, etc. This reply is not permitted in non-stop mode.
27897 @item F @var{call-id},@var{parameter}@dots{}
27898 @var{call-id} is the identifier which says which host system call should
27899 be called. This is just the name of the function. Translation into the
27900 correct system call is only applicable as it's defined in @value{GDBN}.
27901 @xref{File-I/O Remote Protocol Extension}, for a list of implemented
27904 @samp{@var{parameter}@dots{}} is a list of parameters as defined for
27905 this very system call.
27907 The target replies with this packet when it expects @value{GDBN} to
27908 call a host system call on behalf of the target. @value{GDBN} replies
27909 with an appropriate @samp{F} packet and keeps up waiting for the next
27910 reply packet from the target. The latest @samp{C}, @samp{c}, @samp{S}
27911 or @samp{s} action is expected to be continued. @xref{File-I/O Remote
27912 Protocol Extension}, for more details.
27916 @node General Query Packets
27917 @section General Query Packets
27918 @cindex remote query requests
27920 Packets starting with @samp{q} are @dfn{general query packets};
27921 packets starting with @samp{Q} are @dfn{general set packets}. General
27922 query and set packets are a semi-unified form for retrieving and
27923 sending information to and from the stub.
27925 The initial letter of a query or set packet is followed by a name
27926 indicating what sort of thing the packet applies to. For example,
27927 @value{GDBN} may use a @samp{qSymbol} packet to exchange symbol
27928 definitions with the stub. These packet names follow some
27933 The name must not contain commas, colons or semicolons.
27935 Most @value{GDBN} query and set packets have a leading upper case
27938 The names of custom vendor packets should use a company prefix, in
27939 lower case, followed by a period. For example, packets designed at
27940 the Acme Corporation might begin with @samp{qacme.foo} (for querying
27941 foos) or @samp{Qacme.bar} (for setting bars).
27944 The name of a query or set packet should be separated from any
27945 parameters by a @samp{:}; the parameters themselves should be
27946 separated by @samp{,} or @samp{;}. Stubs must be careful to match the
27947 full packet name, and check for a separator or the end of the packet,
27948 in case two packet names share a common prefix. New packets should not begin
27949 with @samp{qC}, @samp{qP}, or @samp{qL}@footnote{The @samp{qP} and @samp{qL}
27950 packets predate these conventions, and have arguments without any terminator
27951 for the packet name; we suspect they are in widespread use in places that
27952 are difficult to upgrade. The @samp{qC} packet has no arguments, but some
27953 existing stubs (e.g.@: RedBoot) are known to not check for the end of the
27956 Like the descriptions of the other packets, each description here
27957 has a template showing the packet's overall syntax, followed by an
27958 explanation of the packet's meaning. We include spaces in some of the
27959 templates for clarity; these are not part of the packet's syntax. No
27960 @value{GDBN} packet uses spaces to separate its components.
27962 Here are the currently defined query and set packets:
27967 @cindex current thread, remote request
27968 @cindex @samp{qC} packet
27969 Return the current thread ID.
27973 @item QC @var{thread-id}
27974 Where @var{thread-id} is a thread ID as documented in
27975 @ref{thread-id syntax}.
27976 @item @r{(anything else)}
27977 Any other reply implies the old thread ID.
27980 @item qCRC:@var{addr},@var{length}
27981 @cindex CRC of memory block, remote request
27982 @cindex @samp{qCRC} packet
27983 Compute the CRC checksum of a block of memory.
27987 An error (such as memory fault)
27988 @item C @var{crc32}
27989 The specified memory region's checksum is @var{crc32}.
27993 @itemx qsThreadInfo
27994 @cindex list active threads, remote request
27995 @cindex @samp{qfThreadInfo} packet
27996 @cindex @samp{qsThreadInfo} packet
27997 Obtain a list of all active thread IDs from the target (OS). Since there
27998 may be too many active threads to fit into one reply packet, this query
27999 works iteratively: it may require more than one query/reply sequence to
28000 obtain the entire list of threads. The first query of the sequence will
28001 be the @samp{qfThreadInfo} query; subsequent queries in the
28002 sequence will be the @samp{qsThreadInfo} query.
28004 NOTE: This packet replaces the @samp{qL} query (see below).
28008 @item m @var{thread-id}
28010 @item m @var{thread-id},@var{thread-id}@dots{}
28011 a comma-separated list of thread IDs
28013 (lower case letter @samp{L}) denotes end of list.
28016 In response to each query, the target will reply with a list of one or
28017 more thread IDs, separated by commas.
28018 @value{GDBN} will respond to each reply with a request for more thread
28019 ids (using the @samp{qs} form of the query), until the target responds
28020 with @samp{l} (lower-case el, for @dfn{last}).
28021 Refer to @ref{thread-id syntax}, for the format of the @var{thread-id}
28024 @item qGetTLSAddr:@var{thread-id},@var{offset},@var{lm}
28025 @cindex get thread-local storage address, remote request
28026 @cindex @samp{qGetTLSAddr} packet
28027 Fetch the address associated with thread local storage specified
28028 by @var{thread-id}, @var{offset}, and @var{lm}.
28030 @var{thread-id} is the thread ID associated with the
28031 thread for which to fetch the TLS address. @xref{thread-id syntax}.
28033 @var{offset} is the (big endian, hex encoded) offset associated with the
28034 thread local variable. (This offset is obtained from the debug
28035 information associated with the variable.)
28037 @var{lm} is the (big endian, hex encoded) OS/ABI-specific encoding of the
28038 the load module associated with the thread local storage. For example,
28039 a @sc{gnu}/Linux system will pass the link map address of the shared
28040 object associated with the thread local storage under consideration.
28041 Other operating environments may choose to represent the load module
28042 differently, so the precise meaning of this parameter will vary.
28046 @item @var{XX}@dots{}
28047 Hex encoded (big endian) bytes representing the address of the thread
28048 local storage requested.
28051 An error occurred. @var{nn} are hex digits.
28054 An empty reply indicates that @samp{qGetTLSAddr} is not supported by the stub.
28057 @item qL @var{startflag} @var{threadcount} @var{nextthread}
28058 Obtain thread information from RTOS. Where: @var{startflag} (one hex
28059 digit) is one to indicate the first query and zero to indicate a
28060 subsequent query; @var{threadcount} (two hex digits) is the maximum
28061 number of threads the response packet can contain; and @var{nextthread}
28062 (eight hex digits), for subsequent queries (@var{startflag} is zero), is
28063 returned in the response as @var{argthread}.
28065 Don't use this packet; use the @samp{qfThreadInfo} query instead (see above).
28069 @item qM @var{count} @var{done} @var{argthread} @var{thread}@dots{}
28070 Where: @var{count} (two hex digits) is the number of threads being
28071 returned; @var{done} (one hex digit) is zero to indicate more threads
28072 and one indicates no further threads; @var{argthreadid} (eight hex
28073 digits) is @var{nextthread} from the request packet; @var{thread}@dots{}
28074 is a sequence of thread IDs from the target. @var{threadid} (eight hex
28075 digits). See @code{remote.c:parse_threadlist_response()}.
28079 @cindex section offsets, remote request
28080 @cindex @samp{qOffsets} packet
28081 Get section offsets that the target used when relocating the downloaded
28086 @item Text=@var{xxx};Data=@var{yyy}@r{[};Bss=@var{zzz}@r{]}
28087 Relocate the @code{Text} section by @var{xxx} from its original address.
28088 Relocate the @code{Data} section by @var{yyy} from its original address.
28089 If the object file format provides segment information (e.g.@: @sc{elf}
28090 @samp{PT_LOAD} program headers), @value{GDBN} will relocate entire
28091 segments by the supplied offsets.
28093 @emph{Note: while a @code{Bss} offset may be included in the response,
28094 @value{GDBN} ignores this and instead applies the @code{Data} offset
28095 to the @code{Bss} section.}
28097 @item TextSeg=@var{xxx}@r{[};DataSeg=@var{yyy}@r{]}
28098 Relocate the first segment of the object file, which conventionally
28099 contains program code, to a starting address of @var{xxx}. If
28100 @samp{DataSeg} is specified, relocate the second segment, which
28101 conventionally contains modifiable data, to a starting address of
28102 @var{yyy}. @value{GDBN} will report an error if the object file
28103 does not contain segment information, or does not contain at least
28104 as many segments as mentioned in the reply. Extra segments are
28105 kept at fixed offsets relative to the last relocated segment.
28108 @item qP @var{mode} @var{thread-id}
28109 @cindex thread information, remote request
28110 @cindex @samp{qP} packet
28111 Returns information on @var{thread-id}. Where: @var{mode} is a hex
28112 encoded 32 bit mode; @var{thread-id} is a thread ID
28113 (@pxref{thread-id syntax}).
28115 Don't use this packet; use the @samp{qThreadExtraInfo} query instead
28118 Reply: see @code{remote.c:remote_unpack_thread_info_response()}.
28122 @cindex non-stop mode, remote request
28123 @cindex @samp{QNonStop} packet
28125 Enter non-stop (@samp{QNonStop:1}) or all-stop (@samp{QNonStop:0}) mode.
28126 @xref{Remote Non-Stop}, for more information.
28131 The request succeeded.
28134 An error occurred. @var{nn} are hex digits.
28137 An empty reply indicates that @samp{QNonStop} is not supported by
28141 This packet is not probed by default; the remote stub must request it,
28142 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
28143 Use of this packet is controlled by the @code{set non-stop} command;
28144 @pxref{Non-Stop Mode}.
28146 @item QPassSignals: @var{signal} @r{[};@var{signal}@r{]}@dots{}
28147 @cindex pass signals to inferior, remote request
28148 @cindex @samp{QPassSignals} packet
28149 @anchor{QPassSignals}
28150 Each listed @var{signal} should be passed directly to the inferior process.
28151 Signals are numbered identically to continue packets and stop replies
28152 (@pxref{Stop Reply Packets}). Each @var{signal} list item should be
28153 strictly greater than the previous item. These signals do not need to stop
28154 the inferior, or be reported to @value{GDBN}. All other signals should be
28155 reported to @value{GDBN}. Multiple @samp{QPassSignals} packets do not
28156 combine; any earlier @samp{QPassSignals} list is completely replaced by the
28157 new list. This packet improves performance when using @samp{handle
28158 @var{signal} nostop noprint pass}.
28163 The request succeeded.
28166 An error occurred. @var{nn} are hex digits.
28169 An empty reply indicates that @samp{QPassSignals} is not supported by
28173 Use of this packet is controlled by the @code{set remote pass-signals}
28174 command (@pxref{Remote Configuration, set remote pass-signals}).
28175 This packet is not probed by default; the remote stub must request it,
28176 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
28178 @item qRcmd,@var{command}
28179 @cindex execute remote command, remote request
28180 @cindex @samp{qRcmd} packet
28181 @var{command} (hex encoded) is passed to the local interpreter for
28182 execution. Invalid commands should be reported using the output
28183 string. Before the final result packet, the target may also respond
28184 with a number of intermediate @samp{O@var{output}} console output
28185 packets. @emph{Implementors should note that providing access to a
28186 stubs's interpreter may have security implications}.
28191 A command response with no output.
28193 A command response with the hex encoded output string @var{OUTPUT}.
28195 Indicate a badly formed request.
28197 An empty reply indicates that @samp{qRcmd} is not recognized.
28200 (Note that the @code{qRcmd} packet's name is separated from the
28201 command by a @samp{,}, not a @samp{:}, contrary to the naming
28202 conventions above. Please don't use this packet as a model for new
28205 @item qSearch:memory:@var{address};@var{length};@var{search-pattern}
28206 @cindex searching memory, in remote debugging
28207 @cindex @samp{qSearch:memory} packet
28208 @anchor{qSearch memory}
28209 Search @var{length} bytes at @var{address} for @var{search-pattern}.
28210 @var{address} and @var{length} are encoded in hex.
28211 @var{search-pattern} is a sequence of bytes, hex encoded.
28216 The pattern was not found.
28218 The pattern was found at @var{address}.
28220 A badly formed request or an error was encountered while searching memory.
28222 An empty reply indicates that @samp{qSearch:memory} is not recognized.
28225 @item QStartNoAckMode
28226 @cindex @samp{QStartNoAckMode} packet
28227 @anchor{QStartNoAckMode}
28228 Request that the remote stub disable the normal @samp{+}/@samp{-}
28229 protocol acknowledgments (@pxref{Packet Acknowledgment}).
28234 The stub has switched to no-acknowledgment mode.
28235 @value{GDBN} acknowledges this reponse,
28236 but neither the stub nor @value{GDBN} shall send or expect further
28237 @samp{+}/@samp{-} acknowledgments in the current connection.
28239 An empty reply indicates that the stub does not support no-acknowledgment mode.
28242 @item qSupported @r{[}:@var{gdbfeature} @r{[};@var{gdbfeature}@r{]}@dots{} @r{]}
28243 @cindex supported packets, remote query
28244 @cindex features of the remote protocol
28245 @cindex @samp{qSupported} packet
28246 @anchor{qSupported}
28247 Tell the remote stub about features supported by @value{GDBN}, and
28248 query the stub for features it supports. This packet allows
28249 @value{GDBN} and the remote stub to take advantage of each others'
28250 features. @samp{qSupported} also consolidates multiple feature probes
28251 at startup, to improve @value{GDBN} performance---a single larger
28252 packet performs better than multiple smaller probe packets on
28253 high-latency links. Some features may enable behavior which must not
28254 be on by default, e.g.@: because it would confuse older clients or
28255 stubs. Other features may describe packets which could be
28256 automatically probed for, but are not. These features must be
28257 reported before @value{GDBN} will use them. This ``default
28258 unsupported'' behavior is not appropriate for all packets, but it
28259 helps to keep the initial connection time under control with new
28260 versions of @value{GDBN} which support increasing numbers of packets.
28264 @item @var{stubfeature} @r{[};@var{stubfeature}@r{]}@dots{}
28265 The stub supports or does not support each returned @var{stubfeature},
28266 depending on the form of each @var{stubfeature} (see below for the
28269 An empty reply indicates that @samp{qSupported} is not recognized,
28270 or that no features needed to be reported to @value{GDBN}.
28273 The allowed forms for each feature (either a @var{gdbfeature} in the
28274 @samp{qSupported} packet, or a @var{stubfeature} in the response)
28278 @item @var{name}=@var{value}
28279 The remote protocol feature @var{name} is supported, and associated
28280 with the specified @var{value}. The format of @var{value} depends
28281 on the feature, but it must not include a semicolon.
28283 The remote protocol feature @var{name} is supported, and does not
28284 need an associated value.
28286 The remote protocol feature @var{name} is not supported.
28288 The remote protocol feature @var{name} may be supported, and
28289 @value{GDBN} should auto-detect support in some other way when it is
28290 needed. This form will not be used for @var{gdbfeature} notifications,
28291 but may be used for @var{stubfeature} responses.
28294 Whenever the stub receives a @samp{qSupported} request, the
28295 supplied set of @value{GDBN} features should override any previous
28296 request. This allows @value{GDBN} to put the stub in a known
28297 state, even if the stub had previously been communicating with
28298 a different version of @value{GDBN}.
28300 The following values of @var{gdbfeature} (for the packet sent by @value{GDBN})
28305 This feature indicates whether @value{GDBN} supports multiprocess
28306 extensions to the remote protocol. @value{GDBN} does not use such
28307 extensions unless the stub also reports that it supports them by
28308 including @samp{multiprocess+} in its @samp{qSupported} reply.
28309 @xref{multiprocess extensions}, for details.
28312 Stubs should ignore any unknown values for
28313 @var{gdbfeature}. Any @value{GDBN} which sends a @samp{qSupported}
28314 packet supports receiving packets of unlimited length (earlier
28315 versions of @value{GDBN} may reject overly long responses). Additional values
28316 for @var{gdbfeature} may be defined in the future to let the stub take
28317 advantage of new features in @value{GDBN}, e.g.@: incompatible
28318 improvements in the remote protocol---the @samp{multiprocess} feature is
28319 an example of such a feature. The stub's reply should be independent
28320 of the @var{gdbfeature} entries sent by @value{GDBN}; first @value{GDBN}
28321 describes all the features it supports, and then the stub replies with
28322 all the features it supports.
28324 Similarly, @value{GDBN} will silently ignore unrecognized stub feature
28325 responses, as long as each response uses one of the standard forms.
28327 Some features are flags. A stub which supports a flag feature
28328 should respond with a @samp{+} form response. Other features
28329 require values, and the stub should respond with an @samp{=}
28332 Each feature has a default value, which @value{GDBN} will use if
28333 @samp{qSupported} is not available or if the feature is not mentioned
28334 in the @samp{qSupported} response. The default values are fixed; a
28335 stub is free to omit any feature responses that match the defaults.
28337 Not all features can be probed, but for those which can, the probing
28338 mechanism is useful: in some cases, a stub's internal
28339 architecture may not allow the protocol layer to know some information
28340 about the underlying target in advance. This is especially common in
28341 stubs which may be configured for multiple targets.
28343 These are the currently defined stub features and their properties:
28345 @multitable @columnfractions 0.35 0.2 0.12 0.2
28346 @c NOTE: The first row should be @headitem, but we do not yet require
28347 @c a new enough version of Texinfo (4.7) to use @headitem.
28349 @tab Value Required
28353 @item @samp{PacketSize}
28358 @item @samp{qXfer:auxv:read}
28363 @item @samp{qXfer:features:read}
28368 @item @samp{qXfer:libraries:read}
28373 @item @samp{qXfer:memory-map:read}
28378 @item @samp{qXfer:spu:read}
28383 @item @samp{qXfer:spu:write}
28388 @item @samp{qXfer:siginfo:read}
28393 @item @samp{qXfer:siginfo:write}
28398 @item @samp{QNonStop}
28403 @item @samp{QPassSignals}
28408 @item @samp{QStartNoAckMode}
28413 @item @samp{multiprocess}
28420 These are the currently defined stub features, in more detail:
28423 @cindex packet size, remote protocol
28424 @item PacketSize=@var{bytes}
28425 The remote stub can accept packets up to at least @var{bytes} in
28426 length. @value{GDBN} will send packets up to this size for bulk
28427 transfers, and will never send larger packets. This is a limit on the
28428 data characters in the packet, including the frame and checksum.
28429 There is no trailing NUL byte in a remote protocol packet; if the stub
28430 stores packets in a NUL-terminated format, it should allow an extra
28431 byte in its buffer for the NUL. If this stub feature is not supported,
28432 @value{GDBN} guesses based on the size of the @samp{g} packet response.
28434 @item qXfer:auxv:read
28435 The remote stub understands the @samp{qXfer:auxv:read} packet
28436 (@pxref{qXfer auxiliary vector read}).
28438 @item qXfer:features:read
28439 The remote stub understands the @samp{qXfer:features:read} packet
28440 (@pxref{qXfer target description read}).
28442 @item qXfer:libraries:read
28443 The remote stub understands the @samp{qXfer:libraries:read} packet
28444 (@pxref{qXfer library list read}).
28446 @item qXfer:memory-map:read
28447 The remote stub understands the @samp{qXfer:memory-map:read} packet
28448 (@pxref{qXfer memory map read}).
28450 @item qXfer:spu:read
28451 The remote stub understands the @samp{qXfer:spu:read} packet
28452 (@pxref{qXfer spu read}).
28454 @item qXfer:spu:write
28455 The remote stub understands the @samp{qXfer:spu:write} packet
28456 (@pxref{qXfer spu write}).
28458 @item qXfer:siginfo:read
28459 The remote stub understands the @samp{qXfer:siginfo:read} packet
28460 (@pxref{qXfer siginfo read}).
28462 @item qXfer:siginfo:write
28463 The remote stub understands the @samp{qXfer:siginfo:write} packet
28464 (@pxref{qXfer siginfo write}).
28467 The remote stub understands the @samp{QNonStop} packet
28468 (@pxref{QNonStop}).
28471 The remote stub understands the @samp{QPassSignals} packet
28472 (@pxref{QPassSignals}).
28474 @item QStartNoAckMode
28475 The remote stub understands the @samp{QStartNoAckMode} packet and
28476 prefers to operate in no-acknowledgment mode. @xref{Packet Acknowledgment}.
28479 @anchor{multiprocess extensions}
28480 @cindex multiprocess extensions, in remote protocol
28481 The remote stub understands the multiprocess extensions to the remote
28482 protocol syntax. The multiprocess extensions affect the syntax of
28483 thread IDs in both packets and replies (@pxref{thread-id syntax}), and
28484 add process IDs to the @samp{D} packet and @samp{W} and @samp{X}
28485 replies. Note that reporting this feature indicates support for the
28486 syntactic extensions only, not that the stub necessarily supports
28487 debugging of more than one process at a time. The stub must not use
28488 multiprocess extensions in packet replies unless @value{GDBN} has also
28489 indicated it supports them in its @samp{qSupported} request.
28491 @item qXfer:osdata:read
28492 The remote stub understands the @samp{qXfer:osdata:read} packet
28493 ((@pxref{qXfer osdata read}).
28498 @cindex symbol lookup, remote request
28499 @cindex @samp{qSymbol} packet
28500 Notify the target that @value{GDBN} is prepared to serve symbol lookup
28501 requests. Accept requests from the target for the values of symbols.
28506 The target does not need to look up any (more) symbols.
28507 @item qSymbol:@var{sym_name}
28508 The target requests the value of symbol @var{sym_name} (hex encoded).
28509 @value{GDBN} may provide the value by using the
28510 @samp{qSymbol:@var{sym_value}:@var{sym_name}} message, described
28514 @item qSymbol:@var{sym_value}:@var{sym_name}
28515 Set the value of @var{sym_name} to @var{sym_value}.
28517 @var{sym_name} (hex encoded) is the name of a symbol whose value the
28518 target has previously requested.
28520 @var{sym_value} (hex) is the value for symbol @var{sym_name}. If
28521 @value{GDBN} cannot supply a value for @var{sym_name}, then this field
28527 The target does not need to look up any (more) symbols.
28528 @item qSymbol:@var{sym_name}
28529 The target requests the value of a new symbol @var{sym_name} (hex
28530 encoded). @value{GDBN} will continue to supply the values of symbols
28531 (if available), until the target ceases to request them.
28536 @xref{Tracepoint Packets}.
28538 @item qThreadExtraInfo,@var{thread-id}
28539 @cindex thread attributes info, remote request
28540 @cindex @samp{qThreadExtraInfo} packet
28541 Obtain a printable string description of a thread's attributes from
28542 the target OS. @var{thread-id} is a thread ID;
28543 see @ref{thread-id syntax}. This
28544 string may contain anything that the target OS thinks is interesting
28545 for @value{GDBN} to tell the user about the thread. The string is
28546 displayed in @value{GDBN}'s @code{info threads} display. Some
28547 examples of possible thread extra info strings are @samp{Runnable}, or
28548 @samp{Blocked on Mutex}.
28552 @item @var{XX}@dots{}
28553 Where @samp{@var{XX}@dots{}} is a hex encoding of @sc{ascii} data,
28554 comprising the printable string containing the extra information about
28555 the thread's attributes.
28558 (Note that the @code{qThreadExtraInfo} packet's name is separated from
28559 the command by a @samp{,}, not a @samp{:}, contrary to the naming
28560 conventions above. Please don't use this packet as a model for new
28568 @xref{Tracepoint Packets}.
28570 @item qXfer:@var{object}:read:@var{annex}:@var{offset},@var{length}
28571 @cindex read special object, remote request
28572 @cindex @samp{qXfer} packet
28573 @anchor{qXfer read}
28574 Read uninterpreted bytes from the target's special data area
28575 identified by the keyword @var{object}. Request @var{length} bytes
28576 starting at @var{offset} bytes into the data. The content and
28577 encoding of @var{annex} is specific to @var{object}; it can supply
28578 additional details about what data to access.
28580 Here are the specific requests of this form defined so far. All
28581 @samp{qXfer:@var{object}:read:@dots{}} requests use the same reply
28582 formats, listed below.
28585 @item qXfer:auxv:read::@var{offset},@var{length}
28586 @anchor{qXfer auxiliary vector read}
28587 Access the target's @dfn{auxiliary vector}. @xref{OS Information,
28588 auxiliary vector}. Note @var{annex} must be empty.
28590 This packet is not probed by default; the remote stub must request it,
28591 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
28593 @item qXfer:features:read:@var{annex}:@var{offset},@var{length}
28594 @anchor{qXfer target description read}
28595 Access the @dfn{target description}. @xref{Target Descriptions}. The
28596 annex specifies which XML document to access. The main description is
28597 always loaded from the @samp{target.xml} annex.
28599 This packet is not probed by default; the remote stub must request it,
28600 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
28602 @item qXfer:libraries:read:@var{annex}:@var{offset},@var{length}
28603 @anchor{qXfer library list read}
28604 Access the target's list of loaded libraries. @xref{Library List Format}.
28605 The annex part of the generic @samp{qXfer} packet must be empty
28606 (@pxref{qXfer read}).
28608 Targets which maintain a list of libraries in the program's memory do
28609 not need to implement this packet; it is designed for platforms where
28610 the operating system manages the list of loaded libraries.
28612 This packet is not probed by default; the remote stub must request it,
28613 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
28615 @item qXfer:memory-map:read::@var{offset},@var{length}
28616 @anchor{qXfer memory map read}
28617 Access the target's @dfn{memory-map}. @xref{Memory Map Format}. The
28618 annex part of the generic @samp{qXfer} packet must be empty
28619 (@pxref{qXfer read}).
28621 This packet is not probed by default; the remote stub must request it,
28622 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
28624 @item qXfer:siginfo:read::@var{offset},@var{length}
28625 @anchor{qXfer siginfo read}
28626 Read contents of the extra signal information on the target
28627 system. The annex part of the generic @samp{qXfer} packet must be
28628 empty (@pxref{qXfer read}).
28630 This packet is not probed by default; the remote stub must request it,
28631 by supplying an appropriate @samp{qSupported} response
28632 (@pxref{qSupported}).
28634 @item qXfer:spu:read:@var{annex}:@var{offset},@var{length}
28635 @anchor{qXfer spu read}
28636 Read contents of an @code{spufs} file on the target system. The
28637 annex specifies which file to read; it must be of the form
28638 @file{@var{id}/@var{name}}, where @var{id} specifies an SPU context ID
28639 in the target process, and @var{name} identifes the @code{spufs} file
28640 in that context to be accessed.
28642 This packet is not probed by default; the remote stub must request it,
28643 by supplying an appropriate @samp{qSupported} response
28644 (@pxref{qSupported}).
28646 @item qXfer:osdata:read::@var{offset},@var{length}
28647 @anchor{qXfer osdata read}
28648 Access the target's @dfn{operating system information}.
28649 @xref{Operating System Information}.
28656 Data @var{data} (@pxref{Binary Data}) has been read from the
28657 target. There may be more data at a higher address (although
28658 it is permitted to return @samp{m} even for the last valid
28659 block of data, as long as at least one byte of data was read).
28660 @var{data} may have fewer bytes than the @var{length} in the
28664 Data @var{data} (@pxref{Binary Data}) has been read from the target.
28665 There is no more data to be read. @var{data} may have fewer bytes
28666 than the @var{length} in the request.
28669 The @var{offset} in the request is at the end of the data.
28670 There is no more data to be read.
28673 The request was malformed, or @var{annex} was invalid.
28676 The offset was invalid, or there was an error encountered reading the data.
28677 @var{nn} is a hex-encoded @code{errno} value.
28680 An empty reply indicates the @var{object} string was not recognized by
28681 the stub, or that the object does not support reading.
28684 @item qXfer:@var{object}:write:@var{annex}:@var{offset}:@var{data}@dots{}
28685 @cindex write data into object, remote request
28686 @anchor{qXfer write}
28687 Write uninterpreted bytes into the target's special data area
28688 identified by the keyword @var{object}, starting at @var{offset} bytes
28689 into the data. @var{data}@dots{} is the binary-encoded data
28690 (@pxref{Binary Data}) to be written. The content and encoding of @var{annex}
28691 is specific to @var{object}; it can supply additional details about what data
28694 Here are the specific requests of this form defined so far. All
28695 @samp{qXfer:@var{object}:write:@dots{}} requests use the same reply
28696 formats, listed below.
28699 @item qXfer:siginfo:write::@var{offset}:@var{data}@dots{}
28700 @anchor{qXfer siginfo write}
28701 Write @var{data} to the extra signal information on the target system.
28702 The annex part of the generic @samp{qXfer} packet must be
28703 empty (@pxref{qXfer write}).
28705 This packet is not probed by default; the remote stub must request it,
28706 by supplying an appropriate @samp{qSupported} response
28707 (@pxref{qSupported}).
28709 @item qXfer:spu:write:@var{annex}:@var{offset}:@var{data}@dots{}
28710 @anchor{qXfer spu write}
28711 Write @var{data} to an @code{spufs} file on the target system. The
28712 annex specifies which file to write; it must be of the form
28713 @file{@var{id}/@var{name}}, where @var{id} specifies an SPU context ID
28714 in the target process, and @var{name} identifes the @code{spufs} file
28715 in that context to be accessed.
28717 This packet is not probed by default; the remote stub must request it,
28718 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
28724 @var{nn} (hex encoded) is the number of bytes written.
28725 This may be fewer bytes than supplied in the request.
28728 The request was malformed, or @var{annex} was invalid.
28731 The offset was invalid, or there was an error encountered writing the data.
28732 @var{nn} is a hex-encoded @code{errno} value.
28735 An empty reply indicates the @var{object} string was not
28736 recognized by the stub, or that the object does not support writing.
28739 @item qXfer:@var{object}:@var{operation}:@dots{}
28740 Requests of this form may be added in the future. When a stub does
28741 not recognize the @var{object} keyword, or its support for
28742 @var{object} does not recognize the @var{operation} keyword, the stub
28743 must respond with an empty packet.
28745 @item qAttached:@var{pid}
28746 @cindex query attached, remote request
28747 @cindex @samp{qAttached} packet
28748 Return an indication of whether the remote server attached to an
28749 existing process or created a new process. When the multiprocess
28750 protocol extensions are supported (@pxref{multiprocess extensions}),
28751 @var{pid} is an integer in hexadecimal format identifying the target
28752 process. Otherwise, @value{GDBN} will omit the @var{pid} field and
28753 the query packet will be simplified as @samp{qAttached}.
28755 This query is used, for example, to know whether the remote process
28756 should be detached or killed when a @value{GDBN} session is ended with
28757 the @code{quit} command.
28762 The remote server attached to an existing process.
28764 The remote server created a new process.
28766 A badly formed request or an error was encountered.
28771 @node Register Packet Format
28772 @section Register Packet Format
28774 The following @code{g}/@code{G} packets have previously been defined.
28775 In the below, some thirty-two bit registers are transferred as
28776 sixty-four bits. Those registers should be zero/sign extended (which?)
28777 to fill the space allocated. Register bytes are transferred in target
28778 byte order. The two nibbles within a register byte are transferred
28779 most-significant - least-significant.
28785 All registers are transferred as thirty-two bit quantities in the order:
28786 32 general-purpose; sr; lo; hi; bad; cause; pc; 32 floating-point
28787 registers; fsr; fir; fp.
28791 All registers are transferred as sixty-four bit quantities (including
28792 thirty-two bit registers such as @code{sr}). The ordering is the same
28797 @node Tracepoint Packets
28798 @section Tracepoint Packets
28799 @cindex tracepoint packets
28800 @cindex packets, tracepoint
28802 Here we describe the packets @value{GDBN} uses to implement
28803 tracepoints (@pxref{Tracepoints}).
28807 @item QTDP:@var{n}:@var{addr}:@var{ena}:@var{step}:@var{pass}@r{[}-@r{]}
28808 Create a new tracepoint, number @var{n}, at @var{addr}. If @var{ena}
28809 is @samp{E}, then the tracepoint is enabled; if it is @samp{D}, then
28810 the tracepoint is disabled. @var{step} is the tracepoint's step
28811 count, and @var{pass} is its pass count. If the trailing @samp{-} is
28812 present, further @samp{QTDP} packets will follow to specify this
28813 tracepoint's actions.
28818 The packet was understood and carried out.
28820 The packet was not recognized.
28823 @item QTDP:-@var{n}:@var{addr}:@r{[}S@r{]}@var{action}@dots{}@r{[}-@r{]}
28824 Define actions to be taken when a tracepoint is hit. @var{n} and
28825 @var{addr} must be the same as in the initial @samp{QTDP} packet for
28826 this tracepoint. This packet may only be sent immediately after
28827 another @samp{QTDP} packet that ended with a @samp{-}. If the
28828 trailing @samp{-} is present, further @samp{QTDP} packets will follow,
28829 specifying more actions for this tracepoint.
28831 In the series of action packets for a given tracepoint, at most one
28832 can have an @samp{S} before its first @var{action}. If such a packet
28833 is sent, it and the following packets define ``while-stepping''
28834 actions. Any prior packets define ordinary actions --- that is, those
28835 taken when the tracepoint is first hit. If no action packet has an
28836 @samp{S}, then all the packets in the series specify ordinary
28837 tracepoint actions.
28839 The @samp{@var{action}@dots{}} portion of the packet is a series of
28840 actions, concatenated without separators. Each action has one of the
28846 Collect the registers whose bits are set in @var{mask}. @var{mask} is
28847 a hexadecimal number whose @var{i}'th bit is set if register number
28848 @var{i} should be collected. (The least significant bit is numbered
28849 zero.) Note that @var{mask} may be any number of digits long; it may
28850 not fit in a 32-bit word.
28852 @item M @var{basereg},@var{offset},@var{len}
28853 Collect @var{len} bytes of memory starting at the address in register
28854 number @var{basereg}, plus @var{offset}. If @var{basereg} is
28855 @samp{-1}, then the range has a fixed address: @var{offset} is the
28856 address of the lowest byte to collect. The @var{basereg},
28857 @var{offset}, and @var{len} parameters are all unsigned hexadecimal
28858 values (the @samp{-1} value for @var{basereg} is a special case).
28860 @item X @var{len},@var{expr}
28861 Evaluate @var{expr}, whose length is @var{len}, and collect memory as
28862 it directs. @var{expr} is an agent expression, as described in
28863 @ref{Agent Expressions}. Each byte of the expression is encoded as a
28864 two-digit hex number in the packet; @var{len} is the number of bytes
28865 in the expression (and thus one-half the number of hex digits in the
28870 Any number of actions may be packed together in a single @samp{QTDP}
28871 packet, as long as the packet does not exceed the maximum packet
28872 length (400 bytes, for many stubs). There may be only one @samp{R}
28873 action per tracepoint, and it must precede any @samp{M} or @samp{X}
28874 actions. Any registers referred to by @samp{M} and @samp{X} actions
28875 must be collected by a preceding @samp{R} action. (The
28876 ``while-stepping'' actions are treated as if they were attached to a
28877 separate tracepoint, as far as these restrictions are concerned.)
28882 The packet was understood and carried out.
28884 The packet was not recognized.
28887 @item QTFrame:@var{n}
28888 Select the @var{n}'th tracepoint frame from the buffer, and use the
28889 register and memory contents recorded there to answer subsequent
28890 request packets from @value{GDBN}.
28892 A successful reply from the stub indicates that the stub has found the
28893 requested frame. The response is a series of parts, concatenated
28894 without separators, describing the frame we selected. Each part has
28895 one of the following forms:
28899 The selected frame is number @var{n} in the trace frame buffer;
28900 @var{f} is a hexadecimal number. If @var{f} is @samp{-1}, then there
28901 was no frame matching the criteria in the request packet.
28904 The selected trace frame records a hit of tracepoint number @var{t};
28905 @var{t} is a hexadecimal number.
28909 @item QTFrame:pc:@var{addr}
28910 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
28911 currently selected frame whose PC is @var{addr};
28912 @var{addr} is a hexadecimal number.
28914 @item QTFrame:tdp:@var{t}
28915 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
28916 currently selected frame that is a hit of tracepoint @var{t}; @var{t}
28917 is a hexadecimal number.
28919 @item QTFrame:range:@var{start}:@var{end}
28920 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
28921 currently selected frame whose PC is between @var{start} (inclusive)
28922 and @var{end} (exclusive); @var{start} and @var{end} are hexadecimal
28925 @item QTFrame:outside:@var{start}:@var{end}
28926 Like @samp{QTFrame:range:@var{start}:@var{end}}, but select the first
28927 frame @emph{outside} the given range of addresses.
28930 Begin the tracepoint experiment. Begin collecting data from tracepoint
28931 hits in the trace frame buffer.
28934 End the tracepoint experiment. Stop collecting trace frames.
28937 Clear the table of tracepoints, and empty the trace frame buffer.
28939 @item QTro:@var{start1},@var{end1}:@var{start2},@var{end2}:@dots{}
28940 Establish the given ranges of memory as ``transparent''. The stub
28941 will answer requests for these ranges from memory's current contents,
28942 if they were not collected as part of the tracepoint hit.
28944 @value{GDBN} uses this to mark read-only regions of memory, like those
28945 containing program code. Since these areas never change, they should
28946 still have the same contents they did when the tracepoint was hit, so
28947 there's no reason for the stub to refuse to provide their contents.
28950 Ask the stub if there is a trace experiment running right now.
28955 There is no trace experiment running.
28957 There is a trace experiment running.
28963 @node Host I/O Packets
28964 @section Host I/O Packets
28965 @cindex Host I/O, remote protocol
28966 @cindex file transfer, remote protocol
28968 The @dfn{Host I/O} packets allow @value{GDBN} to perform I/O
28969 operations on the far side of a remote link. For example, Host I/O is
28970 used to upload and download files to a remote target with its own
28971 filesystem. Host I/O uses the same constant values and data structure
28972 layout as the target-initiated File-I/O protocol. However, the
28973 Host I/O packets are structured differently. The target-initiated
28974 protocol relies on target memory to store parameters and buffers.
28975 Host I/O requests are initiated by @value{GDBN}, and the
28976 target's memory is not involved. @xref{File-I/O Remote Protocol
28977 Extension}, for more details on the target-initiated protocol.
28979 The Host I/O request packets all encode a single operation along with
28980 its arguments. They have this format:
28984 @item vFile:@var{operation}: @var{parameter}@dots{}
28985 @var{operation} is the name of the particular request; the target
28986 should compare the entire packet name up to the second colon when checking
28987 for a supported operation. The format of @var{parameter} depends on
28988 the operation. Numbers are always passed in hexadecimal. Negative
28989 numbers have an explicit minus sign (i.e.@: two's complement is not
28990 used). Strings (e.g.@: filenames) are encoded as a series of
28991 hexadecimal bytes. The last argument to a system call may be a
28992 buffer of escaped binary data (@pxref{Binary Data}).
28996 The valid responses to Host I/O packets are:
29000 @item F @var{result} [, @var{errno}] [; @var{attachment}]
29001 @var{result} is the integer value returned by this operation, usually
29002 non-negative for success and -1 for errors. If an error has occured,
29003 @var{errno} will be included in the result. @var{errno} will have a
29004 value defined by the File-I/O protocol (@pxref{Errno Values}). For
29005 operations which return data, @var{attachment} supplies the data as a
29006 binary buffer. Binary buffers in response packets are escaped in the
29007 normal way (@pxref{Binary Data}). See the individual packet
29008 documentation for the interpretation of @var{result} and
29012 An empty response indicates that this operation is not recognized.
29016 These are the supported Host I/O operations:
29019 @item vFile:open: @var{pathname}, @var{flags}, @var{mode}
29020 Open a file at @var{pathname} and return a file descriptor for it, or
29021 return -1 if an error occurs. @var{pathname} is a string,
29022 @var{flags} is an integer indicating a mask of open flags
29023 (@pxref{Open Flags}), and @var{mode} is an integer indicating a mask
29024 of mode bits to use if the file is created (@pxref{mode_t Values}).
29025 @xref{open}, for details of the open flags and mode values.
29027 @item vFile:close: @var{fd}
29028 Close the open file corresponding to @var{fd} and return 0, or
29029 -1 if an error occurs.
29031 @item vFile:pread: @var{fd}, @var{count}, @var{offset}
29032 Read data from the open file corresponding to @var{fd}. Up to
29033 @var{count} bytes will be read from the file, starting at @var{offset}
29034 relative to the start of the file. The target may read fewer bytes;
29035 common reasons include packet size limits and an end-of-file
29036 condition. The number of bytes read is returned. Zero should only be
29037 returned for a successful read at the end of the file, or if
29038 @var{count} was zero.
29040 The data read should be returned as a binary attachment on success.
29041 If zero bytes were read, the response should include an empty binary
29042 attachment (i.e.@: a trailing semicolon). The return value is the
29043 number of target bytes read; the binary attachment may be longer if
29044 some characters were escaped.
29046 @item vFile:pwrite: @var{fd}, @var{offset}, @var{data}
29047 Write @var{data} (a binary buffer) to the open file corresponding
29048 to @var{fd}. Start the write at @var{offset} from the start of the
29049 file. Unlike many @code{write} system calls, there is no
29050 separate @var{count} argument; the length of @var{data} in the
29051 packet is used. @samp{vFile:write} returns the number of bytes written,
29052 which may be shorter than the length of @var{data}, or -1 if an
29055 @item vFile:unlink: @var{pathname}
29056 Delete the file at @var{pathname} on the target. Return 0,
29057 or -1 if an error occurs. @var{pathname} is a string.
29062 @section Interrupts
29063 @cindex interrupts (remote protocol)
29065 When a program on the remote target is running, @value{GDBN} may
29066 attempt to interrupt it by sending a @samp{Ctrl-C} or a @code{BREAK},
29067 control of which is specified via @value{GDBN}'s @samp{remotebreak}
29068 setting (@pxref{set remotebreak}).
29070 The precise meaning of @code{BREAK} is defined by the transport
29071 mechanism and may, in fact, be undefined. @value{GDBN} does not
29072 currently define a @code{BREAK} mechanism for any of the network
29073 interfaces except for TCP, in which case @value{GDBN} sends the
29074 @code{telnet} BREAK sequence.
29076 @samp{Ctrl-C}, on the other hand, is defined and implemented for all
29077 transport mechanisms. It is represented by sending the single byte
29078 @code{0x03} without any of the usual packet overhead described in
29079 the Overview section (@pxref{Overview}). When a @code{0x03} byte is
29080 transmitted as part of a packet, it is considered to be packet data
29081 and does @emph{not} represent an interrupt. E.g., an @samp{X} packet
29082 (@pxref{X packet}), used for binary downloads, may include an unescaped
29083 @code{0x03} as part of its packet.
29085 Stubs are not required to recognize these interrupt mechanisms and the
29086 precise meaning associated with receipt of the interrupt is
29087 implementation defined. If the target supports debugging of multiple
29088 threads and/or processes, it should attempt to interrupt all
29089 currently-executing threads and processes.
29090 If the stub is successful at interrupting the
29091 running program, it should send one of the stop
29092 reply packets (@pxref{Stop Reply Packets}) to @value{GDBN} as a result
29093 of successfully stopping the program in all-stop mode, and a stop reply
29094 for each stopped thread in non-stop mode.
29095 Interrupts received while the
29096 program is stopped are discarded.
29098 @node Notification Packets
29099 @section Notification Packets
29100 @cindex notification packets
29101 @cindex packets, notification
29103 The @value{GDBN} remote serial protocol includes @dfn{notifications},
29104 packets that require no acknowledgment. Both the GDB and the stub
29105 may send notifications (although the only notifications defined at
29106 present are sent by the stub). Notifications carry information
29107 without incurring the round-trip latency of an acknowledgment, and so
29108 are useful for low-impact communications where occasional packet loss
29111 A notification packet has the form @samp{% @var{data} #
29112 @var{checksum}}, where @var{data} is the content of the notification,
29113 and @var{checksum} is a checksum of @var{data}, computed and formatted
29114 as for ordinary @value{GDBN} packets. A notification's @var{data}
29115 never contains @samp{$}, @samp{%} or @samp{#} characters. Upon
29116 receiving a notification, the recipient sends no @samp{+} or @samp{-}
29117 to acknowledge the notification's receipt or to report its corruption.
29119 Every notification's @var{data} begins with a name, which contains no
29120 colon characters, followed by a colon character.
29122 Recipients should silently ignore corrupted notifications and
29123 notifications they do not understand. Recipients should restart
29124 timeout periods on receipt of a well-formed notification, whether or
29125 not they understand it.
29127 Senders should only send the notifications described here when this
29128 protocol description specifies that they are permitted. In the
29129 future, we may extend the protocol to permit existing notifications in
29130 new contexts; this rule helps older senders avoid confusing newer
29133 (Older versions of @value{GDBN} ignore bytes received until they see
29134 the @samp{$} byte that begins an ordinary packet, so new stubs may
29135 transmit notifications without fear of confusing older clients. There
29136 are no notifications defined for @value{GDBN} to send at the moment, but we
29137 assume that most older stubs would ignore them, as well.)
29139 The following notification packets from the stub to @value{GDBN} are
29143 @item Stop: @var{reply}
29144 Report an asynchronous stop event in non-stop mode.
29145 The @var{reply} has the form of a stop reply, as
29146 described in @ref{Stop Reply Packets}. Refer to @ref{Remote Non-Stop},
29147 for information on how these notifications are acknowledged by
29151 @node Remote Non-Stop
29152 @section Remote Protocol Support for Non-Stop Mode
29154 @value{GDBN}'s remote protocol supports non-stop debugging of
29155 multi-threaded programs, as described in @ref{Non-Stop Mode}. If the stub
29156 supports non-stop mode, it should report that to @value{GDBN} by including
29157 @samp{QNonStop+} in its @samp{qSupported} response (@pxref{qSupported}).
29159 @value{GDBN} typically sends a @samp{QNonStop} packet only when
29160 establishing a new connection with the stub. Entering non-stop mode
29161 does not alter the state of any currently-running threads, but targets
29162 must stop all threads in any already-attached processes when entering
29163 all-stop mode. @value{GDBN} uses the @samp{?} packet as necessary to
29164 probe the target state after a mode change.
29166 In non-stop mode, when an attached process encounters an event that
29167 would otherwise be reported with a stop reply, it uses the
29168 asynchronous notification mechanism (@pxref{Notification Packets}) to
29169 inform @value{GDBN}. In contrast to all-stop mode, where all threads
29170 in all processes are stopped when a stop reply is sent, in non-stop
29171 mode only the thread reporting the stop event is stopped. That is,
29172 when reporting a @samp{S} or @samp{T} response to indicate completion
29173 of a step operation, hitting a breakpoint, or a fault, only the
29174 affected thread is stopped; any other still-running threads continue
29175 to run. When reporting a @samp{W} or @samp{X} response, all running
29176 threads belonging to other attached processes continue to run.
29178 Only one stop reply notification at a time may be pending; if
29179 additional stop events occur before @value{GDBN} has acknowledged the
29180 previous notification, they must be queued by the stub for later
29181 synchronous transmission in response to @samp{vStopped} packets from
29182 @value{GDBN}. Because the notification mechanism is unreliable,
29183 the stub is permitted to resend a stop reply notification
29184 if it believes @value{GDBN} may not have received it. @value{GDBN}
29185 ignores additional stop reply notifications received before it has
29186 finished processing a previous notification and the stub has completed
29187 sending any queued stop events.
29189 Otherwise, @value{GDBN} must be prepared to receive a stop reply
29190 notification at any time. Specifically, they may appear when
29191 @value{GDBN} is not otherwise reading input from the stub, or when
29192 @value{GDBN} is expecting to read a normal synchronous response or a
29193 @samp{+}/@samp{-} acknowledgment to a packet it has sent.
29194 Notification packets are distinct from any other communication from
29195 the stub so there is no ambiguity.
29197 After receiving a stop reply notification, @value{GDBN} shall
29198 acknowledge it by sending a @samp{vStopped} packet (@pxref{vStopped packet})
29199 as a regular, synchronous request to the stub. Such acknowledgment
29200 is not required to happen immediately, as @value{GDBN} is permitted to
29201 send other, unrelated packets to the stub first, which the stub should
29204 Upon receiving a @samp{vStopped} packet, if the stub has other queued
29205 stop events to report to @value{GDBN}, it shall respond by sending a
29206 normal stop reply response. @value{GDBN} shall then send another
29207 @samp{vStopped} packet to solicit further responses; again, it is
29208 permitted to send other, unrelated packets as well which the stub
29209 should process normally.
29211 If the stub receives a @samp{vStopped} packet and there are no
29212 additional stop events to report, the stub shall return an @samp{OK}
29213 response. At this point, if further stop events occur, the stub shall
29214 send a new stop reply notification, @value{GDBN} shall accept the
29215 notification, and the process shall be repeated.
29217 In non-stop mode, the target shall respond to the @samp{?} packet as
29218 follows. First, any incomplete stop reply notification/@samp{vStopped}
29219 sequence in progress is abandoned. The target must begin a new
29220 sequence reporting stop events for all stopped threads, whether or not
29221 it has previously reported those events to @value{GDBN}. The first
29222 stop reply is sent as a synchronous reply to the @samp{?} packet, and
29223 subsequent stop replies are sent as responses to @samp{vStopped} packets
29224 using the mechanism described above. The target must not send
29225 asynchronous stop reply notifications until the sequence is complete.
29226 If all threads are running when the target receives the @samp{?} packet,
29227 or if the target is not attached to any process, it shall respond
29230 @node Packet Acknowledgment
29231 @section Packet Acknowledgment
29233 @cindex acknowledgment, for @value{GDBN} remote
29234 @cindex packet acknowledgment, for @value{GDBN} remote
29235 By default, when either the host or the target machine receives a packet,
29236 the first response expected is an acknowledgment: either @samp{+} (to indicate
29237 the package was received correctly) or @samp{-} (to request retransmission).
29238 This mechanism allows the @value{GDBN} remote protocol to operate over
29239 unreliable transport mechanisms, such as a serial line.
29241 In cases where the transport mechanism is itself reliable (such as a pipe or
29242 TCP connection), the @samp{+}/@samp{-} acknowledgments are redundant.
29243 It may be desirable to disable them in that case to reduce communication
29244 overhead, or for other reasons. This can be accomplished by means of the
29245 @samp{QStartNoAckMode} packet; @pxref{QStartNoAckMode}.
29247 When in no-acknowledgment mode, neither the stub nor @value{GDBN} shall send or
29248 expect @samp{+}/@samp{-} protocol acknowledgments. The packet
29249 and response format still includes the normal checksum, as described in
29250 @ref{Overview}, but the checksum may be ignored by the receiver.
29252 If the stub supports @samp{QStartNoAckMode} and prefers to operate in
29253 no-acknowledgment mode, it should report that to @value{GDBN}
29254 by including @samp{QStartNoAckMode+} in its response to @samp{qSupported};
29255 @pxref{qSupported}.
29256 If @value{GDBN} also supports @samp{QStartNoAckMode} and it has not been
29257 disabled via the @code{set remote noack-packet off} command
29258 (@pxref{Remote Configuration}),
29259 @value{GDBN} may then send a @samp{QStartNoAckMode} packet to the stub.
29260 Only then may the stub actually turn off packet acknowledgments.
29261 @value{GDBN} sends a final @samp{+} acknowledgment of the stub's @samp{OK}
29262 response, which can be safely ignored by the stub.
29264 Note that @code{set remote noack-packet} command only affects negotiation
29265 between @value{GDBN} and the stub when subsequent connections are made;
29266 it does not affect the protocol acknowledgment state for any current
29268 Since @samp{+}/@samp{-} acknowledgments are enabled by default when a
29269 new connection is established,
29270 there is also no protocol request to re-enable the acknowledgments
29271 for the current connection, once disabled.
29276 Example sequence of a target being re-started. Notice how the restart
29277 does not get any direct output:
29282 @emph{target restarts}
29285 <- @code{T001:1234123412341234}
29289 Example sequence of a target being stepped by a single instruction:
29292 -> @code{G1445@dots{}}
29297 <- @code{T001:1234123412341234}
29301 <- @code{1455@dots{}}
29305 @node File-I/O Remote Protocol Extension
29306 @section File-I/O Remote Protocol Extension
29307 @cindex File-I/O remote protocol extension
29310 * File-I/O Overview::
29311 * Protocol Basics::
29312 * The F Request Packet::
29313 * The F Reply Packet::
29314 * The Ctrl-C Message::
29316 * List of Supported Calls::
29317 * Protocol-specific Representation of Datatypes::
29319 * File-I/O Examples::
29322 @node File-I/O Overview
29323 @subsection File-I/O Overview
29324 @cindex file-i/o overview
29326 The @dfn{File I/O remote protocol extension} (short: File-I/O) allows the
29327 target to use the host's file system and console I/O to perform various
29328 system calls. System calls on the target system are translated into a
29329 remote protocol packet to the host system, which then performs the needed
29330 actions and returns a response packet to the target system.
29331 This simulates file system operations even on targets that lack file systems.
29333 The protocol is defined to be independent of both the host and target systems.
29334 It uses its own internal representation of datatypes and values. Both
29335 @value{GDBN} and the target's @value{GDBN} stub are responsible for
29336 translating the system-dependent value representations into the internal
29337 protocol representations when data is transmitted.
29339 The communication is synchronous. A system call is possible only when
29340 @value{GDBN} is waiting for a response from the @samp{C}, @samp{c}, @samp{S}
29341 or @samp{s} packets. While @value{GDBN} handles the request for a system call,
29342 the target is stopped to allow deterministic access to the target's
29343 memory. Therefore File-I/O is not interruptible by target signals. On
29344 the other hand, it is possible to interrupt File-I/O by a user interrupt
29345 (@samp{Ctrl-C}) within @value{GDBN}.
29347 The target's request to perform a host system call does not finish
29348 the latest @samp{C}, @samp{c}, @samp{S} or @samp{s} action. That means,
29349 after finishing the system call, the target returns to continuing the
29350 previous activity (continue, step). No additional continue or step
29351 request from @value{GDBN} is required.
29354 (@value{GDBP}) continue
29355 <- target requests 'system call X'
29356 target is stopped, @value{GDBN} executes system call
29357 -> @value{GDBN} returns result
29358 ... target continues, @value{GDBN} returns to wait for the target
29359 <- target hits breakpoint and sends a Txx packet
29362 The protocol only supports I/O on the console and to regular files on
29363 the host file system. Character or block special devices, pipes,
29364 named pipes, sockets or any other communication method on the host
29365 system are not supported by this protocol.
29367 File I/O is not supported in non-stop mode.
29369 @node Protocol Basics
29370 @subsection Protocol Basics
29371 @cindex protocol basics, file-i/o
29373 The File-I/O protocol uses the @code{F} packet as the request as well
29374 as reply packet. Since a File-I/O system call can only occur when
29375 @value{GDBN} is waiting for a response from the continuing or stepping target,
29376 the File-I/O request is a reply that @value{GDBN} has to expect as a result
29377 of a previous @samp{C}, @samp{c}, @samp{S} or @samp{s} packet.
29378 This @code{F} packet contains all information needed to allow @value{GDBN}
29379 to call the appropriate host system call:
29383 A unique identifier for the requested system call.
29386 All parameters to the system call. Pointers are given as addresses
29387 in the target memory address space. Pointers to strings are given as
29388 pointer/length pair. Numerical values are given as they are.
29389 Numerical control flags are given in a protocol-specific representation.
29393 At this point, @value{GDBN} has to perform the following actions.
29397 If the parameters include pointer values to data needed as input to a
29398 system call, @value{GDBN} requests this data from the target with a
29399 standard @code{m} packet request. This additional communication has to be
29400 expected by the target implementation and is handled as any other @code{m}
29404 @value{GDBN} translates all value from protocol representation to host
29405 representation as needed. Datatypes are coerced into the host types.
29408 @value{GDBN} calls the system call.
29411 It then coerces datatypes back to protocol representation.
29414 If the system call is expected to return data in buffer space specified
29415 by pointer parameters to the call, the data is transmitted to the
29416 target using a @code{M} or @code{X} packet. This packet has to be expected
29417 by the target implementation and is handled as any other @code{M} or @code{X}
29422 Eventually @value{GDBN} replies with another @code{F} packet which contains all
29423 necessary information for the target to continue. This at least contains
29430 @code{errno}, if has been changed by the system call.
29437 After having done the needed type and value coercion, the target continues
29438 the latest continue or step action.
29440 @node The F Request Packet
29441 @subsection The @code{F} Request Packet
29442 @cindex file-i/o request packet
29443 @cindex @code{F} request packet
29445 The @code{F} request packet has the following format:
29448 @item F@var{call-id},@var{parameter@dots{}}
29450 @var{call-id} is the identifier to indicate the host system call to be called.
29451 This is just the name of the function.
29453 @var{parameter@dots{}} are the parameters to the system call.
29454 Parameters are hexadecimal integer values, either the actual values in case
29455 of scalar datatypes, pointers to target buffer space in case of compound
29456 datatypes and unspecified memory areas, or pointer/length pairs in case
29457 of string parameters. These are appended to the @var{call-id} as a
29458 comma-delimited list. All values are transmitted in ASCII
29459 string representation, pointer/length pairs separated by a slash.
29465 @node The F Reply Packet
29466 @subsection The @code{F} Reply Packet
29467 @cindex file-i/o reply packet
29468 @cindex @code{F} reply packet
29470 The @code{F} reply packet has the following format:
29474 @item F@var{retcode},@var{errno},@var{Ctrl-C flag};@var{call-specific attachment}
29476 @var{retcode} is the return code of the system call as hexadecimal value.
29478 @var{errno} is the @code{errno} set by the call, in protocol-specific
29480 This parameter can be omitted if the call was successful.
29482 @var{Ctrl-C flag} is only sent if the user requested a break. In this
29483 case, @var{errno} must be sent as well, even if the call was successful.
29484 The @var{Ctrl-C flag} itself consists of the character @samp{C}:
29491 or, if the call was interrupted before the host call has been performed:
29498 assuming 4 is the protocol-specific representation of @code{EINTR}.
29503 @node The Ctrl-C Message
29504 @subsection The @samp{Ctrl-C} Message
29505 @cindex ctrl-c message, in file-i/o protocol
29507 If the @samp{Ctrl-C} flag is set in the @value{GDBN}
29508 reply packet (@pxref{The F Reply Packet}),
29509 the target should behave as if it had
29510 gotten a break message. The meaning for the target is ``system call
29511 interrupted by @code{SIGINT}''. Consequentially, the target should actually stop
29512 (as with a break message) and return to @value{GDBN} with a @code{T02}
29515 It's important for the target to know in which
29516 state the system call was interrupted. There are two possible cases:
29520 The system call hasn't been performed on the host yet.
29523 The system call on the host has been finished.
29527 These two states can be distinguished by the target by the value of the
29528 returned @code{errno}. If it's the protocol representation of @code{EINTR}, the system
29529 call hasn't been performed. This is equivalent to the @code{EINTR} handling
29530 on POSIX systems. In any other case, the target may presume that the
29531 system call has been finished --- successfully or not --- and should behave
29532 as if the break message arrived right after the system call.
29534 @value{GDBN} must behave reliably. If the system call has not been called
29535 yet, @value{GDBN} may send the @code{F} reply immediately, setting @code{EINTR} as
29536 @code{errno} in the packet. If the system call on the host has been finished
29537 before the user requests a break, the full action must be finished by
29538 @value{GDBN}. This requires sending @code{M} or @code{X} packets as necessary.
29539 The @code{F} packet may only be sent when either nothing has happened
29540 or the full action has been completed.
29543 @subsection Console I/O
29544 @cindex console i/o as part of file-i/o
29546 By default and if not explicitly closed by the target system, the file
29547 descriptors 0, 1 and 2 are connected to the @value{GDBN} console. Output
29548 on the @value{GDBN} console is handled as any other file output operation
29549 (@code{write(1, @dots{})} or @code{write(2, @dots{})}). Console input is handled
29550 by @value{GDBN} so that after the target read request from file descriptor
29551 0 all following typing is buffered until either one of the following
29556 The user types @kbd{Ctrl-c}. The behaviour is as explained above, and the
29558 system call is treated as finished.
29561 The user presses @key{RET}. This is treated as end of input with a trailing
29565 The user types @kbd{Ctrl-d}. This is treated as end of input. No trailing
29566 character (neither newline nor @samp{Ctrl-D}) is appended to the input.
29570 If the user has typed more characters than fit in the buffer given to
29571 the @code{read} call, the trailing characters are buffered in @value{GDBN} until
29572 either another @code{read(0, @dots{})} is requested by the target, or debugging
29573 is stopped at the user's request.
29576 @node List of Supported Calls
29577 @subsection List of Supported Calls
29578 @cindex list of supported file-i/o calls
29595 @unnumberedsubsubsec open
29596 @cindex open, file-i/o system call
29601 int open(const char *pathname, int flags);
29602 int open(const char *pathname, int flags, mode_t mode);
29606 @samp{Fopen,@var{pathptr}/@var{len},@var{flags},@var{mode}}
29609 @var{flags} is the bitwise @code{OR} of the following values:
29613 If the file does not exist it will be created. The host
29614 rules apply as far as file ownership and time stamps
29618 When used with @code{O_CREAT}, if the file already exists it is
29619 an error and open() fails.
29622 If the file already exists and the open mode allows
29623 writing (@code{O_RDWR} or @code{O_WRONLY} is given) it will be
29624 truncated to zero length.
29627 The file is opened in append mode.
29630 The file is opened for reading only.
29633 The file is opened for writing only.
29636 The file is opened for reading and writing.
29640 Other bits are silently ignored.
29644 @var{mode} is the bitwise @code{OR} of the following values:
29648 User has read permission.
29651 User has write permission.
29654 Group has read permission.
29657 Group has write permission.
29660 Others have read permission.
29663 Others have write permission.
29667 Other bits are silently ignored.
29670 @item Return value:
29671 @code{open} returns the new file descriptor or -1 if an error
29678 @var{pathname} already exists and @code{O_CREAT} and @code{O_EXCL} were used.
29681 @var{pathname} refers to a directory.
29684 The requested access is not allowed.
29687 @var{pathname} was too long.
29690 A directory component in @var{pathname} does not exist.
29693 @var{pathname} refers to a device, pipe, named pipe or socket.
29696 @var{pathname} refers to a file on a read-only filesystem and
29697 write access was requested.
29700 @var{pathname} is an invalid pointer value.
29703 No space on device to create the file.
29706 The process already has the maximum number of files open.
29709 The limit on the total number of files open on the system
29713 The call was interrupted by the user.
29719 @unnumberedsubsubsec close
29720 @cindex close, file-i/o system call
29729 @samp{Fclose,@var{fd}}
29731 @item Return value:
29732 @code{close} returns zero on success, or -1 if an error occurred.
29738 @var{fd} isn't a valid open file descriptor.
29741 The call was interrupted by the user.
29747 @unnumberedsubsubsec read
29748 @cindex read, file-i/o system call
29753 int read(int fd, void *buf, unsigned int count);
29757 @samp{Fread,@var{fd},@var{bufptr},@var{count}}
29759 @item Return value:
29760 On success, the number of bytes read is returned.
29761 Zero indicates end of file. If count is zero, read
29762 returns zero as well. On error, -1 is returned.
29768 @var{fd} is not a valid file descriptor or is not open for
29772 @var{bufptr} is an invalid pointer value.
29775 The call was interrupted by the user.
29781 @unnumberedsubsubsec write
29782 @cindex write, file-i/o system call
29787 int write(int fd, const void *buf, unsigned int count);
29791 @samp{Fwrite,@var{fd},@var{bufptr},@var{count}}
29793 @item Return value:
29794 On success, the number of bytes written are returned.
29795 Zero indicates nothing was written. On error, -1
29802 @var{fd} is not a valid file descriptor or is not open for
29806 @var{bufptr} is an invalid pointer value.
29809 An attempt was made to write a file that exceeds the
29810 host-specific maximum file size allowed.
29813 No space on device to write the data.
29816 The call was interrupted by the user.
29822 @unnumberedsubsubsec lseek
29823 @cindex lseek, file-i/o system call
29828 long lseek (int fd, long offset, int flag);
29832 @samp{Flseek,@var{fd},@var{offset},@var{flag}}
29834 @var{flag} is one of:
29838 The offset is set to @var{offset} bytes.
29841 The offset is set to its current location plus @var{offset}
29845 The offset is set to the size of the file plus @var{offset}
29849 @item Return value:
29850 On success, the resulting unsigned offset in bytes from
29851 the beginning of the file is returned. Otherwise, a
29852 value of -1 is returned.
29858 @var{fd} is not a valid open file descriptor.
29861 @var{fd} is associated with the @value{GDBN} console.
29864 @var{flag} is not a proper value.
29867 The call was interrupted by the user.
29873 @unnumberedsubsubsec rename
29874 @cindex rename, file-i/o system call
29879 int rename(const char *oldpath, const char *newpath);
29883 @samp{Frename,@var{oldpathptr}/@var{len},@var{newpathptr}/@var{len}}
29885 @item Return value:
29886 On success, zero is returned. On error, -1 is returned.
29892 @var{newpath} is an existing directory, but @var{oldpath} is not a
29896 @var{newpath} is a non-empty directory.
29899 @var{oldpath} or @var{newpath} is a directory that is in use by some
29903 An attempt was made to make a directory a subdirectory
29907 A component used as a directory in @var{oldpath} or new
29908 path is not a directory. Or @var{oldpath} is a directory
29909 and @var{newpath} exists but is not a directory.
29912 @var{oldpathptr} or @var{newpathptr} are invalid pointer values.
29915 No access to the file or the path of the file.
29919 @var{oldpath} or @var{newpath} was too long.
29922 A directory component in @var{oldpath} or @var{newpath} does not exist.
29925 The file is on a read-only filesystem.
29928 The device containing the file has no room for the new
29932 The call was interrupted by the user.
29938 @unnumberedsubsubsec unlink
29939 @cindex unlink, file-i/o system call
29944 int unlink(const char *pathname);
29948 @samp{Funlink,@var{pathnameptr}/@var{len}}
29950 @item Return value:
29951 On success, zero is returned. On error, -1 is returned.
29957 No access to the file or the path of the file.
29960 The system does not allow unlinking of directories.
29963 The file @var{pathname} cannot be unlinked because it's
29964 being used by another process.
29967 @var{pathnameptr} is an invalid pointer value.
29970 @var{pathname} was too long.
29973 A directory component in @var{pathname} does not exist.
29976 A component of the path is not a directory.
29979 The file is on a read-only filesystem.
29982 The call was interrupted by the user.
29988 @unnumberedsubsubsec stat/fstat
29989 @cindex fstat, file-i/o system call
29990 @cindex stat, file-i/o system call
29995 int stat(const char *pathname, struct stat *buf);
29996 int fstat(int fd, struct stat *buf);
30000 @samp{Fstat,@var{pathnameptr}/@var{len},@var{bufptr}}@*
30001 @samp{Ffstat,@var{fd},@var{bufptr}}
30003 @item Return value:
30004 On success, zero is returned. On error, -1 is returned.
30010 @var{fd} is not a valid open file.
30013 A directory component in @var{pathname} does not exist or the
30014 path is an empty string.
30017 A component of the path is not a directory.
30020 @var{pathnameptr} is an invalid pointer value.
30023 No access to the file or the path of the file.
30026 @var{pathname} was too long.
30029 The call was interrupted by the user.
30035 @unnumberedsubsubsec gettimeofday
30036 @cindex gettimeofday, file-i/o system call
30041 int gettimeofday(struct timeval *tv, void *tz);
30045 @samp{Fgettimeofday,@var{tvptr},@var{tzptr}}
30047 @item Return value:
30048 On success, 0 is returned, -1 otherwise.
30054 @var{tz} is a non-NULL pointer.
30057 @var{tvptr} and/or @var{tzptr} is an invalid pointer value.
30063 @unnumberedsubsubsec isatty
30064 @cindex isatty, file-i/o system call
30069 int isatty(int fd);
30073 @samp{Fisatty,@var{fd}}
30075 @item Return value:
30076 Returns 1 if @var{fd} refers to the @value{GDBN} console, 0 otherwise.
30082 The call was interrupted by the user.
30087 Note that the @code{isatty} call is treated as a special case: it returns
30088 1 to the target if the file descriptor is attached
30089 to the @value{GDBN} console, 0 otherwise. Implementing through system calls
30090 would require implementing @code{ioctl} and would be more complex than
30095 @unnumberedsubsubsec system
30096 @cindex system, file-i/o system call
30101 int system(const char *command);
30105 @samp{Fsystem,@var{commandptr}/@var{len}}
30107 @item Return value:
30108 If @var{len} is zero, the return value indicates whether a shell is
30109 available. A zero return value indicates a shell is not available.
30110 For non-zero @var{len}, the value returned is -1 on error and the
30111 return status of the command otherwise. Only the exit status of the
30112 command is returned, which is extracted from the host's @code{system}
30113 return value by calling @code{WEXITSTATUS(retval)}. In case
30114 @file{/bin/sh} could not be executed, 127 is returned.
30120 The call was interrupted by the user.
30125 @value{GDBN} takes over the full task of calling the necessary host calls
30126 to perform the @code{system} call. The return value of @code{system} on
30127 the host is simplified before it's returned
30128 to the target. Any termination signal information from the child process
30129 is discarded, and the return value consists
30130 entirely of the exit status of the called command.
30132 Due to security concerns, the @code{system} call is by default refused
30133 by @value{GDBN}. The user has to allow this call explicitly with the
30134 @code{set remote system-call-allowed 1} command.
30137 @item set remote system-call-allowed
30138 @kindex set remote system-call-allowed
30139 Control whether to allow the @code{system} calls in the File I/O
30140 protocol for the remote target. The default is zero (disabled).
30142 @item show remote system-call-allowed
30143 @kindex show remote system-call-allowed
30144 Show whether the @code{system} calls are allowed in the File I/O
30148 @node Protocol-specific Representation of Datatypes
30149 @subsection Protocol-specific Representation of Datatypes
30150 @cindex protocol-specific representation of datatypes, in file-i/o protocol
30153 * Integral Datatypes::
30155 * Memory Transfer::
30160 @node Integral Datatypes
30161 @unnumberedsubsubsec Integral Datatypes
30162 @cindex integral datatypes, in file-i/o protocol
30164 The integral datatypes used in the system calls are @code{int},
30165 @code{unsigned int}, @code{long}, @code{unsigned long},
30166 @code{mode_t}, and @code{time_t}.
30168 @code{int}, @code{unsigned int}, @code{mode_t} and @code{time_t} are
30169 implemented as 32 bit values in this protocol.
30171 @code{long} and @code{unsigned long} are implemented as 64 bit types.
30173 @xref{Limits}, for corresponding MIN and MAX values (similar to those
30174 in @file{limits.h}) to allow range checking on host and target.
30176 @code{time_t} datatypes are defined as seconds since the Epoch.
30178 All integral datatypes transferred as part of a memory read or write of a
30179 structured datatype e.g.@: a @code{struct stat} have to be given in big endian
30182 @node Pointer Values
30183 @unnumberedsubsubsec Pointer Values
30184 @cindex pointer values, in file-i/o protocol
30186 Pointers to target data are transmitted as they are. An exception
30187 is made for pointers to buffers for which the length isn't
30188 transmitted as part of the function call, namely strings. Strings
30189 are transmitted as a pointer/length pair, both as hex values, e.g.@:
30196 which is a pointer to data of length 18 bytes at position 0x1aaf.
30197 The length is defined as the full string length in bytes, including
30198 the trailing null byte. For example, the string @code{"hello world"}
30199 at address 0x123456 is transmitted as
30205 @node Memory Transfer
30206 @unnumberedsubsubsec Memory Transfer
30207 @cindex memory transfer, in file-i/o protocol
30209 Structured data which is transferred using a memory read or write (for
30210 example, a @code{struct stat}) is expected to be in a protocol-specific format
30211 with all scalar multibyte datatypes being big endian. Translation to
30212 this representation needs to be done both by the target before the @code{F}
30213 packet is sent, and by @value{GDBN} before
30214 it transfers memory to the target. Transferred pointers to structured
30215 data should point to the already-coerced data at any time.
30219 @unnumberedsubsubsec struct stat
30220 @cindex struct stat, in file-i/o protocol
30222 The buffer of type @code{struct stat} used by the target and @value{GDBN}
30223 is defined as follows:
30227 unsigned int st_dev; /* device */
30228 unsigned int st_ino; /* inode */
30229 mode_t st_mode; /* protection */
30230 unsigned int st_nlink; /* number of hard links */
30231 unsigned int st_uid; /* user ID of owner */
30232 unsigned int st_gid; /* group ID of owner */
30233 unsigned int st_rdev; /* device type (if inode device) */
30234 unsigned long st_size; /* total size, in bytes */
30235 unsigned long st_blksize; /* blocksize for filesystem I/O */
30236 unsigned long st_blocks; /* number of blocks allocated */
30237 time_t st_atime; /* time of last access */
30238 time_t st_mtime; /* time of last modification */
30239 time_t st_ctime; /* time of last change */
30243 The integral datatypes conform to the definitions given in the
30244 appropriate section (see @ref{Integral Datatypes}, for details) so this
30245 structure is of size 64 bytes.
30247 The values of several fields have a restricted meaning and/or
30253 A value of 0 represents a file, 1 the console.
30256 No valid meaning for the target. Transmitted unchanged.
30259 Valid mode bits are described in @ref{Constants}. Any other
30260 bits have currently no meaning for the target.
30265 No valid meaning for the target. Transmitted unchanged.
30270 These values have a host and file system dependent
30271 accuracy. Especially on Windows hosts, the file system may not
30272 support exact timing values.
30275 The target gets a @code{struct stat} of the above representation and is
30276 responsible for coercing it to the target representation before
30279 Note that due to size differences between the host, target, and protocol
30280 representations of @code{struct stat} members, these members could eventually
30281 get truncated on the target.
30283 @node struct timeval
30284 @unnumberedsubsubsec struct timeval
30285 @cindex struct timeval, in file-i/o protocol
30287 The buffer of type @code{struct timeval} used by the File-I/O protocol
30288 is defined as follows:
30292 time_t tv_sec; /* second */
30293 long tv_usec; /* microsecond */
30297 The integral datatypes conform to the definitions given in the
30298 appropriate section (see @ref{Integral Datatypes}, for details) so this
30299 structure is of size 8 bytes.
30302 @subsection Constants
30303 @cindex constants, in file-i/o protocol
30305 The following values are used for the constants inside of the
30306 protocol. @value{GDBN} and target are responsible for translating these
30307 values before and after the call as needed.
30318 @unnumberedsubsubsec Open Flags
30319 @cindex open flags, in file-i/o protocol
30321 All values are given in hexadecimal representation.
30333 @node mode_t Values
30334 @unnumberedsubsubsec mode_t Values
30335 @cindex mode_t values, in file-i/o protocol
30337 All values are given in octal representation.
30354 @unnumberedsubsubsec Errno Values
30355 @cindex errno values, in file-i/o protocol
30357 All values are given in decimal representation.
30382 @code{EUNKNOWN} is used as a fallback error value if a host system returns
30383 any error value not in the list of supported error numbers.
30386 @unnumberedsubsubsec Lseek Flags
30387 @cindex lseek flags, in file-i/o protocol
30396 @unnumberedsubsubsec Limits
30397 @cindex limits, in file-i/o protocol
30399 All values are given in decimal representation.
30402 INT_MIN -2147483648
30404 UINT_MAX 4294967295
30405 LONG_MIN -9223372036854775808
30406 LONG_MAX 9223372036854775807
30407 ULONG_MAX 18446744073709551615
30410 @node File-I/O Examples
30411 @subsection File-I/O Examples
30412 @cindex file-i/o examples
30414 Example sequence of a write call, file descriptor 3, buffer is at target
30415 address 0x1234, 6 bytes should be written:
30418 <- @code{Fwrite,3,1234,6}
30419 @emph{request memory read from target}
30422 @emph{return "6 bytes written"}
30426 Example sequence of a read call, file descriptor 3, buffer is at target
30427 address 0x1234, 6 bytes should be read:
30430 <- @code{Fread,3,1234,6}
30431 @emph{request memory write to target}
30432 -> @code{X1234,6:XXXXXX}
30433 @emph{return "6 bytes read"}
30437 Example sequence of a read call, call fails on the host due to invalid
30438 file descriptor (@code{EBADF}):
30441 <- @code{Fread,3,1234,6}
30445 Example sequence of a read call, user presses @kbd{Ctrl-c} before syscall on
30449 <- @code{Fread,3,1234,6}
30454 Example sequence of a read call, user presses @kbd{Ctrl-c} after syscall on
30458 <- @code{Fread,3,1234,6}
30459 -> @code{X1234,6:XXXXXX}
30463 @node Library List Format
30464 @section Library List Format
30465 @cindex library list format, remote protocol
30467 On some platforms, a dynamic loader (e.g.@: @file{ld.so}) runs in the
30468 same process as your application to manage libraries. In this case,
30469 @value{GDBN} can use the loader's symbol table and normal memory
30470 operations to maintain a list of shared libraries. On other
30471 platforms, the operating system manages loaded libraries.
30472 @value{GDBN} can not retrieve the list of currently loaded libraries
30473 through memory operations, so it uses the @samp{qXfer:libraries:read}
30474 packet (@pxref{qXfer library list read}) instead. The remote stub
30475 queries the target's operating system and reports which libraries
30478 The @samp{qXfer:libraries:read} packet returns an XML document which
30479 lists loaded libraries and their offsets. Each library has an
30480 associated name and one or more segment or section base addresses,
30481 which report where the library was loaded in memory.
30483 For the common case of libraries that are fully linked binaries, the
30484 library should have a list of segments. If the target supports
30485 dynamic linking of a relocatable object file, its library XML element
30486 should instead include a list of allocated sections. The segment or
30487 section bases are start addresses, not relocation offsets; they do not
30488 depend on the library's link-time base addresses.
30490 @value{GDBN} must be linked with the Expat library to support XML
30491 library lists. @xref{Expat}.
30493 A simple memory map, with one loaded library relocated by a single
30494 offset, looks like this:
30498 <library name="/lib/libc.so.6">
30499 <segment address="0x10000000"/>
30504 Another simple memory map, with one loaded library with three
30505 allocated sections (.text, .data, .bss), looks like this:
30509 <library name="sharedlib.o">
30510 <section address="0x10000000"/>
30511 <section address="0x20000000"/>
30512 <section address="0x30000000"/>
30517 The format of a library list is described by this DTD:
30520 <!-- library-list: Root element with versioning -->
30521 <!ELEMENT library-list (library)*>
30522 <!ATTLIST library-list version CDATA #FIXED "1.0">
30523 <!ELEMENT library (segment*, section*)>
30524 <!ATTLIST library name CDATA #REQUIRED>
30525 <!ELEMENT segment EMPTY>
30526 <!ATTLIST segment address CDATA #REQUIRED>
30527 <!ELEMENT section EMPTY>
30528 <!ATTLIST section address CDATA #REQUIRED>
30531 In addition, segments and section descriptors cannot be mixed within a
30532 single library element, and you must supply at least one segment or
30533 section for each library.
30535 @node Memory Map Format
30536 @section Memory Map Format
30537 @cindex memory map format
30539 To be able to write into flash memory, @value{GDBN} needs to obtain a
30540 memory map from the target. This section describes the format of the
30543 The memory map is obtained using the @samp{qXfer:memory-map:read}
30544 (@pxref{qXfer memory map read}) packet and is an XML document that
30545 lists memory regions.
30547 @value{GDBN} must be linked with the Expat library to support XML
30548 memory maps. @xref{Expat}.
30550 The top-level structure of the document is shown below:
30553 <?xml version="1.0"?>
30554 <!DOCTYPE memory-map
30555 PUBLIC "+//IDN gnu.org//DTD GDB Memory Map V1.0//EN"
30556 "http://sourceware.org/gdb/gdb-memory-map.dtd">
30562 Each region can be either:
30567 A region of RAM starting at @var{addr} and extending for @var{length}
30571 <memory type="ram" start="@var{addr}" length="@var{length}"/>
30576 A region of read-only memory:
30579 <memory type="rom" start="@var{addr}" length="@var{length}"/>
30584 A region of flash memory, with erasure blocks @var{blocksize}
30588 <memory type="flash" start="@var{addr}" length="@var{length}">
30589 <property name="blocksize">@var{blocksize}</property>
30595 Regions must not overlap. @value{GDBN} assumes that areas of memory not covered
30596 by the memory map are RAM, and uses the ordinary @samp{M} and @samp{X}
30597 packets to write to addresses in such ranges.
30599 The formal DTD for memory map format is given below:
30602 <!-- ................................................... -->
30603 <!-- Memory Map XML DTD ................................ -->
30604 <!-- File: memory-map.dtd .............................. -->
30605 <!-- .................................... .............. -->
30606 <!-- memory-map.dtd -->
30607 <!-- memory-map: Root element with versioning -->
30608 <!ELEMENT memory-map (memory | property)>
30609 <!ATTLIST memory-map version CDATA #FIXED "1.0.0">
30610 <!ELEMENT memory (property)>
30611 <!-- memory: Specifies a memory region,
30612 and its type, or device. -->
30613 <!ATTLIST memory type CDATA #REQUIRED
30614 start CDATA #REQUIRED
30615 length CDATA #REQUIRED
30616 device CDATA #IMPLIED>
30617 <!-- property: Generic attribute tag -->
30618 <!ELEMENT property (#PCDATA | property)*>
30619 <!ATTLIST property name CDATA #REQUIRED>
30622 @include agentexpr.texi
30624 @node Target Descriptions
30625 @appendix Target Descriptions
30626 @cindex target descriptions
30628 @strong{Warning:} target descriptions are still under active development,
30629 and the contents and format may change between @value{GDBN} releases.
30630 The format is expected to stabilize in the future.
30632 One of the challenges of using @value{GDBN} to debug embedded systems
30633 is that there are so many minor variants of each processor
30634 architecture in use. It is common practice for vendors to start with
30635 a standard processor core --- ARM, PowerPC, or MIPS, for example ---
30636 and then make changes to adapt it to a particular market niche. Some
30637 architectures have hundreds of variants, available from dozens of
30638 vendors. This leads to a number of problems:
30642 With so many different customized processors, it is difficult for
30643 the @value{GDBN} maintainers to keep up with the changes.
30645 Since individual variants may have short lifetimes or limited
30646 audiences, it may not be worthwhile to carry information about every
30647 variant in the @value{GDBN} source tree.
30649 When @value{GDBN} does support the architecture of the embedded system
30650 at hand, the task of finding the correct architecture name to give the
30651 @command{set architecture} command can be error-prone.
30654 To address these problems, the @value{GDBN} remote protocol allows a
30655 target system to not only identify itself to @value{GDBN}, but to
30656 actually describe its own features. This lets @value{GDBN} support
30657 processor variants it has never seen before --- to the extent that the
30658 descriptions are accurate, and that @value{GDBN} understands them.
30660 @value{GDBN} must be linked with the Expat library to support XML
30661 target descriptions. @xref{Expat}.
30664 * Retrieving Descriptions:: How descriptions are fetched from a target.
30665 * Target Description Format:: The contents of a target description.
30666 * Predefined Target Types:: Standard types available for target
30668 * Standard Target Features:: Features @value{GDBN} knows about.
30671 @node Retrieving Descriptions
30672 @section Retrieving Descriptions
30674 Target descriptions can be read from the target automatically, or
30675 specified by the user manually. The default behavior is to read the
30676 description from the target. @value{GDBN} retrieves it via the remote
30677 protocol using @samp{qXfer} requests (@pxref{General Query Packets,
30678 qXfer}). The @var{annex} in the @samp{qXfer} packet will be
30679 @samp{target.xml}. The contents of the @samp{target.xml} annex are an
30680 XML document, of the form described in @ref{Target Description
30683 Alternatively, you can specify a file to read for the target description.
30684 If a file is set, the target will not be queried. The commands to
30685 specify a file are:
30688 @cindex set tdesc filename
30689 @item set tdesc filename @var{path}
30690 Read the target description from @var{path}.
30692 @cindex unset tdesc filename
30693 @item unset tdesc filename
30694 Do not read the XML target description from a file. @value{GDBN}
30695 will use the description supplied by the current target.
30697 @cindex show tdesc filename
30698 @item show tdesc filename
30699 Show the filename to read for a target description, if any.
30703 @node Target Description Format
30704 @section Target Description Format
30705 @cindex target descriptions, XML format
30707 A target description annex is an @uref{http://www.w3.org/XML/, XML}
30708 document which complies with the Document Type Definition provided in
30709 the @value{GDBN} sources in @file{gdb/features/gdb-target.dtd}. This
30710 means you can use generally available tools like @command{xmllint} to
30711 check that your feature descriptions are well-formed and valid.
30712 However, to help people unfamiliar with XML write descriptions for
30713 their targets, we also describe the grammar here.
30715 Target descriptions can identify the architecture of the remote target
30716 and (for some architectures) provide information about custom register
30717 sets. @value{GDBN} can use this information to autoconfigure for your
30718 target, or to warn you if you connect to an unsupported target.
30720 Here is a simple target description:
30723 <target version="1.0">
30724 <architecture>i386:x86-64</architecture>
30729 This minimal description only says that the target uses
30730 the x86-64 architecture.
30732 A target description has the following overall form, with [ ] marking
30733 optional elements and @dots{} marking repeatable elements. The elements
30734 are explained further below.
30737 <?xml version="1.0"?>
30738 <!DOCTYPE target SYSTEM "gdb-target.dtd">
30739 <target version="1.0">
30740 @r{[}@var{architecture}@r{]}
30741 @r{[}@var{feature}@dots{}@r{]}
30746 The description is generally insensitive to whitespace and line
30747 breaks, under the usual common-sense rules. The XML version
30748 declaration and document type declaration can generally be omitted
30749 (@value{GDBN} does not require them), but specifying them may be
30750 useful for XML validation tools. The @samp{version} attribute for
30751 @samp{<target>} may also be omitted, but we recommend
30752 including it; if future versions of @value{GDBN} use an incompatible
30753 revision of @file{gdb-target.dtd}, they will detect and report
30754 the version mismatch.
30756 @subsection Inclusion
30757 @cindex target descriptions, inclusion
30760 @cindex <xi:include>
30763 It can sometimes be valuable to split a target description up into
30764 several different annexes, either for organizational purposes, or to
30765 share files between different possible target descriptions. You can
30766 divide a description into multiple files by replacing any element of
30767 the target description with an inclusion directive of the form:
30770 <xi:include href="@var{document}"/>
30774 When @value{GDBN} encounters an element of this form, it will retrieve
30775 the named XML @var{document}, and replace the inclusion directive with
30776 the contents of that document. If the current description was read
30777 using @samp{qXfer}, then so will be the included document;
30778 @var{document} will be interpreted as the name of an annex. If the
30779 current description was read from a file, @value{GDBN} will look for
30780 @var{document} as a file in the same directory where it found the
30781 original description.
30783 @subsection Architecture
30784 @cindex <architecture>
30786 An @samp{<architecture>} element has this form:
30789 <architecture>@var{arch}</architecture>
30792 @var{arch} is an architecture name from the same selection
30793 accepted by @code{set architecture} (@pxref{Targets, ,Specifying a
30794 Debugging Target}).
30796 @subsection Features
30799 Each @samp{<feature>} describes some logical portion of the target
30800 system. Features are currently used to describe available CPU
30801 registers and the types of their contents. A @samp{<feature>} element
30805 <feature name="@var{name}">
30806 @r{[}@var{type}@dots{}@r{]}
30812 Each feature's name should be unique within the description. The name
30813 of a feature does not matter unless @value{GDBN} has some special
30814 knowledge of the contents of that feature; if it does, the feature
30815 should have its standard name. @xref{Standard Target Features}.
30819 Any register's value is a collection of bits which @value{GDBN} must
30820 interpret. The default interpretation is a two's complement integer,
30821 but other types can be requested by name in the register description.
30822 Some predefined types are provided by @value{GDBN} (@pxref{Predefined
30823 Target Types}), and the description can define additional composite types.
30825 Each type element must have an @samp{id} attribute, which gives
30826 a unique (within the containing @samp{<feature>}) name to the type.
30827 Types must be defined before they are used.
30830 Some targets offer vector registers, which can be treated as arrays
30831 of scalar elements. These types are written as @samp{<vector>} elements,
30832 specifying the array element type, @var{type}, and the number of elements,
30836 <vector id="@var{id}" type="@var{type}" count="@var{count}"/>
30840 If a register's value is usefully viewed in multiple ways, define it
30841 with a union type containing the useful representations. The
30842 @samp{<union>} element contains one or more @samp{<field>} elements,
30843 each of which has a @var{name} and a @var{type}:
30846 <union id="@var{id}">
30847 <field name="@var{name}" type="@var{type}"/>
30852 @subsection Registers
30855 Each register is represented as an element with this form:
30858 <reg name="@var{name}"
30859 bitsize="@var{size}"
30860 @r{[}regnum="@var{num}"@r{]}
30861 @r{[}save-restore="@var{save-restore}"@r{]}
30862 @r{[}type="@var{type}"@r{]}
30863 @r{[}group="@var{group}"@r{]}/>
30867 The components are as follows:
30872 The register's name; it must be unique within the target description.
30875 The register's size, in bits.
30878 The register's number. If omitted, a register's number is one greater
30879 than that of the previous register (either in the current feature or in
30880 a preceeding feature); the first register in the target description
30881 defaults to zero. This register number is used to read or write
30882 the register; e.g.@: it is used in the remote @code{p} and @code{P}
30883 packets, and registers appear in the @code{g} and @code{G} packets
30884 in order of increasing register number.
30887 Whether the register should be preserved across inferior function
30888 calls; this must be either @code{yes} or @code{no}. The default is
30889 @code{yes}, which is appropriate for most registers except for
30890 some system control registers; this is not related to the target's
30894 The type of the register. @var{type} may be a predefined type, a type
30895 defined in the current feature, or one of the special types @code{int}
30896 and @code{float}. @code{int} is an integer type of the correct size
30897 for @var{bitsize}, and @code{float} is a floating point type (in the
30898 architecture's normal floating point format) of the correct size for
30899 @var{bitsize}. The default is @code{int}.
30902 The register group to which this register belongs. @var{group} must
30903 be either @code{general}, @code{float}, or @code{vector}. If no
30904 @var{group} is specified, @value{GDBN} will not display the register
30905 in @code{info registers}.
30909 @node Predefined Target Types
30910 @section Predefined Target Types
30911 @cindex target descriptions, predefined types
30913 Type definitions in the self-description can build up composite types
30914 from basic building blocks, but can not define fundamental types. Instead,
30915 standard identifiers are provided by @value{GDBN} for the fundamental
30916 types. The currently supported types are:
30925 Signed integer types holding the specified number of bits.
30932 Unsigned integer types holding the specified number of bits.
30936 Pointers to unspecified code and data. The program counter and
30937 any dedicated return address register may be marked as code
30938 pointers; printing a code pointer converts it into a symbolic
30939 address. The stack pointer and any dedicated address registers
30940 may be marked as data pointers.
30943 Single precision IEEE floating point.
30946 Double precision IEEE floating point.
30949 The 12-byte extended precision format used by ARM FPA registers.
30953 @node Standard Target Features
30954 @section Standard Target Features
30955 @cindex target descriptions, standard features
30957 A target description must contain either no registers or all the
30958 target's registers. If the description contains no registers, then
30959 @value{GDBN} will assume a default register layout, selected based on
30960 the architecture. If the description contains any registers, the
30961 default layout will not be used; the standard registers must be
30962 described in the target description, in such a way that @value{GDBN}
30963 can recognize them.
30965 This is accomplished by giving specific names to feature elements
30966 which contain standard registers. @value{GDBN} will look for features
30967 with those names and verify that they contain the expected registers;
30968 if any known feature is missing required registers, or if any required
30969 feature is missing, @value{GDBN} will reject the target
30970 description. You can add additional registers to any of the
30971 standard features --- @value{GDBN} will display them just as if
30972 they were added to an unrecognized feature.
30974 This section lists the known features and their expected contents.
30975 Sample XML documents for these features are included in the
30976 @value{GDBN} source tree, in the directory @file{gdb/features}.
30978 Names recognized by @value{GDBN} should include the name of the
30979 company or organization which selected the name, and the overall
30980 architecture to which the feature applies; so e.g.@: the feature
30981 containing ARM core registers is named @samp{org.gnu.gdb.arm.core}.
30983 The names of registers are not case sensitive for the purpose
30984 of recognizing standard features, but @value{GDBN} will only display
30985 registers using the capitalization used in the description.
30991 * PowerPC Features::
30996 @subsection ARM Features
30997 @cindex target descriptions, ARM features
30999 The @samp{org.gnu.gdb.arm.core} feature is required for ARM targets.
31000 It should contain registers @samp{r0} through @samp{r13}, @samp{sp},
31001 @samp{lr}, @samp{pc}, and @samp{cpsr}.
31003 The @samp{org.gnu.gdb.arm.fpa} feature is optional. If present, it
31004 should contain registers @samp{f0} through @samp{f7} and @samp{fps}.
31006 The @samp{org.gnu.gdb.xscale.iwmmxt} feature is optional. If present,
31007 it should contain at least registers @samp{wR0} through @samp{wR15} and
31008 @samp{wCGR0} through @samp{wCGR3}. The @samp{wCID}, @samp{wCon},
31009 @samp{wCSSF}, and @samp{wCASF} registers are optional.
31011 @node MIPS Features
31012 @subsection MIPS Features
31013 @cindex target descriptions, MIPS features
31015 The @samp{org.gnu.gdb.mips.cpu} feature is required for MIPS targets.
31016 It should contain registers @samp{r0} through @samp{r31}, @samp{lo},
31017 @samp{hi}, and @samp{pc}. They may be 32-bit or 64-bit depending
31020 The @samp{org.gnu.gdb.mips.cp0} feature is also required. It should
31021 contain at least the @samp{status}, @samp{badvaddr}, and @samp{cause}
31022 registers. They may be 32-bit or 64-bit depending on the target.
31024 The @samp{org.gnu.gdb.mips.fpu} feature is currently required, though
31025 it may be optional in a future version of @value{GDBN}. It should
31026 contain registers @samp{f0} through @samp{f31}, @samp{fcsr}, and
31027 @samp{fir}. They may be 32-bit or 64-bit depending on the target.
31029 The @samp{org.gnu.gdb.mips.linux} feature is optional. It should
31030 contain a single register, @samp{restart}, which is used by the
31031 Linux kernel to control restartable syscalls.
31033 @node M68K Features
31034 @subsection M68K Features
31035 @cindex target descriptions, M68K features
31038 @item @samp{org.gnu.gdb.m68k.core}
31039 @itemx @samp{org.gnu.gdb.coldfire.core}
31040 @itemx @samp{org.gnu.gdb.fido.core}
31041 One of those features must be always present.
31042 The feature that is present determines which flavor of m68k is
31043 used. The feature that is present should contain registers
31044 @samp{d0} through @samp{d7}, @samp{a0} through @samp{a5}, @samp{fp},
31045 @samp{sp}, @samp{ps} and @samp{pc}.
31047 @item @samp{org.gnu.gdb.coldfire.fp}
31048 This feature is optional. If present, it should contain registers
31049 @samp{fp0} through @samp{fp7}, @samp{fpcontrol}, @samp{fpstatus} and
31053 @node PowerPC Features
31054 @subsection PowerPC Features
31055 @cindex target descriptions, PowerPC features
31057 The @samp{org.gnu.gdb.power.core} feature is required for PowerPC
31058 targets. It should contain registers @samp{r0} through @samp{r31},
31059 @samp{pc}, @samp{msr}, @samp{cr}, @samp{lr}, @samp{ctr}, and
31060 @samp{xer}. They may be 32-bit or 64-bit depending on the target.
31062 The @samp{org.gnu.gdb.power.fpu} feature is optional. It should
31063 contain registers @samp{f0} through @samp{f31} and @samp{fpscr}.
31065 The @samp{org.gnu.gdb.power.altivec} feature is optional. It should
31066 contain registers @samp{vr0} through @samp{vr31}, @samp{vscr},
31069 The @samp{org.gnu.gdb.power.vsx} feature is optional. It should
31070 contain registers @samp{vs0h} through @samp{vs31h}. @value{GDBN}
31071 will combine these registers with the floating point registers
31072 (@samp{f0} through @samp{f31}) and the altivec registers (@samp{vr0}
31073 through @samp{vr31}) to present the 128-bit wide registers @samp{vs0}
31074 through @samp{vs63}, the set of vector registers for POWER7.
31076 The @samp{org.gnu.gdb.power.spe} feature is optional. It should
31077 contain registers @samp{ev0h} through @samp{ev31h}, @samp{acc}, and
31078 @samp{spefscr}. SPE targets should provide 32-bit registers in
31079 @samp{org.gnu.gdb.power.core} and provide the upper halves in
31080 @samp{ev0h} through @samp{ev31h}. @value{GDBN} will combine
31081 these to present registers @samp{ev0} through @samp{ev31} to the
31084 @node Operating System Information
31085 @appendix Operating System Information
31086 @cindex operating system information
31092 Users of @value{GDBN} often wish to obtain information about the state of
31093 the operating system running on the target---for example the list of
31094 processes, or the list of open files. This section describes the
31095 mechanism that makes it possible. This mechanism is similar to the
31096 target features mechanism (@pxref{Target Descriptions}), but focuses
31097 on a different aspect of target.
31099 Operating system information is retrived from the target via the
31100 remote protocol, using @samp{qXfer} requests (@pxref{qXfer osdata
31101 read}). The object name in the request should be @samp{osdata}, and
31102 the @var{annex} identifies the data to be fetched.
31105 @appendixsection Process list
31106 @cindex operating system information, process list
31108 When requesting the process list, the @var{annex} field in the
31109 @samp{qXfer} request should be @samp{processes}. The returned data is
31110 an XML document. The formal syntax of this document is defined in
31111 @file{gdb/features/osdata.dtd}.
31113 An example document is:
31116 <?xml version="1.0"?>
31117 <!DOCTYPE target SYSTEM "osdata.dtd">
31118 <osdata type="processes">
31120 <column name="pid">1</column>
31121 <column name="user">root</column>
31122 <column name="command">/sbin/init</column>
31127 Each item should include a column whose name is @samp{pid}. The value
31128 of that column should identify the process on the target. The
31129 @samp{user} and @samp{command} columns are optional, and will be
31130 displayed by @value{GDBN}. Target may provide additional columns,
31131 which @value{GDBN} currently ignores.
31145 % I think something like @colophon should be in texinfo. In the
31147 \long\def\colophon{\hbox to0pt{}\vfill
31148 \centerline{The body of this manual is set in}
31149 \centerline{\fontname\tenrm,}
31150 \centerline{with headings in {\bf\fontname\tenbf}}
31151 \centerline{and examples in {\tt\fontname\tentt}.}
31152 \centerline{{\it\fontname\tenit\/},}
31153 \centerline{{\bf\fontname\tenbf}, and}
31154 \centerline{{\sl\fontname\tensl\/}}
31155 \centerline{are used for emphasis.}\vfill}
31157 % Blame: doc@cygnus.com, 1991.