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.
162 * JIT Interface:: Using the JIT debugging interface.
164 * GDB Bugs:: Reporting bugs in @value{GDBN}
166 * Command Line Editing:: Command Line Editing
167 * Using History Interactively:: Using History Interactively
168 * Formatting Documentation:: How to format and print @value{GDBN} documentation
169 * Installing GDB:: Installing GDB
170 * Maintenance Commands:: Maintenance Commands
171 * Remote Protocol:: GDB Remote Serial Protocol
172 * Agent Expressions:: The GDB Agent Expression Mechanism
173 * Target Descriptions:: How targets can describe themselves to
175 * Operating System Information:: Getting additional information from
177 * Copying:: GNU General Public License says
178 how you can copy and share GDB
179 * GNU Free Documentation License:: The license for this documentation
188 @unnumbered Summary of @value{GDBN}
190 The purpose of a debugger such as @value{GDBN} is to allow you to see what is
191 going on ``inside'' another program while it executes---or what another
192 program was doing at the moment it crashed.
194 @value{GDBN} can do four main kinds of things (plus other things in support of
195 these) to help you catch bugs in the act:
199 Start your program, specifying anything that might affect its behavior.
202 Make your program stop on specified conditions.
205 Examine what has happened, when your program has stopped.
208 Change things in your program, so you can experiment with correcting the
209 effects of one bug and go on to learn about another.
212 You can use @value{GDBN} to debug programs written in C and C@t{++}.
213 For more information, see @ref{Supported Languages,,Supported Languages}.
214 For more information, see @ref{C,,C and C++}.
217 Support for Modula-2 is partial. For information on Modula-2, see
218 @ref{Modula-2,,Modula-2}.
221 Debugging Pascal programs which use sets, subranges, file variables, or
222 nested functions does not currently work. @value{GDBN} does not support
223 entering expressions, printing values, or similar features using Pascal
227 @value{GDBN} can be used to debug programs written in Fortran, although
228 it may be necessary to refer to some variables with a trailing
231 @value{GDBN} can be used to debug programs written in Objective-C,
232 using either the Apple/NeXT or the GNU Objective-C runtime.
235 * Free Software:: Freely redistributable software
236 * Contributors:: Contributors to GDB
240 @unnumberedsec Free Software
242 @value{GDBN} is @dfn{free software}, protected by the @sc{gnu}
243 General Public License
244 (GPL). The GPL gives you the freedom to copy or adapt a licensed
245 program---but every person getting a copy also gets with it the
246 freedom to modify that copy (which means that they must get access to
247 the source code), and the freedom to distribute further copies.
248 Typical software companies use copyrights to limit your freedoms; the
249 Free Software Foundation uses the GPL to preserve these freedoms.
251 Fundamentally, the General Public License is a license which says that
252 you have these freedoms and that you cannot take these freedoms away
255 @unnumberedsec Free Software Needs Free Documentation
257 The biggest deficiency in the free software community today is not in
258 the software---it is the lack of good free documentation that we can
259 include with the free software. Many of our most important
260 programs do not come with free reference manuals and free introductory
261 texts. Documentation is an essential part of any software package;
262 when an important free software package does not come with a free
263 manual and a free tutorial, that is a major gap. We have many such
266 Consider Perl, for instance. The tutorial manuals that people
267 normally use are non-free. How did this come about? Because the
268 authors of those manuals published them with restrictive terms---no
269 copying, no modification, source files not available---which exclude
270 them from the free software world.
272 That wasn't the first time this sort of thing happened, and it was far
273 from the last. Many times we have heard a GNU user eagerly describe a
274 manual that he is writing, his intended contribution to the community,
275 only to learn that he had ruined everything by signing a publication
276 contract to make it non-free.
278 Free documentation, like free software, is a matter of freedom, not
279 price. The problem with the non-free manual is not that publishers
280 charge a price for printed copies---that in itself is fine. (The Free
281 Software Foundation sells printed copies of manuals, too.) The
282 problem is the restrictions on the use of the manual. Free manuals
283 are available in source code form, and give you permission to copy and
284 modify. Non-free manuals do not allow this.
286 The criteria of freedom for a free manual are roughly the same as for
287 free software. Redistribution (including the normal kinds of
288 commercial redistribution) must be permitted, so that the manual can
289 accompany every copy of the program, both on-line and on paper.
291 Permission for modification of the technical content is crucial too.
292 When people modify the software, adding or changing features, if they
293 are conscientious they will change the manual too---so they can
294 provide accurate and clear documentation for the modified program. A
295 manual that leaves you no choice but to write a new manual to document
296 a changed version of the program is not really available to our
299 Some kinds of limits on the way modification is handled are
300 acceptable. For example, requirements to preserve the original
301 author's copyright notice, the distribution terms, or the list of
302 authors, are ok. It is also no problem to require modified versions
303 to include notice that they were modified. Even entire sections that
304 may not be deleted or changed are acceptable, as long as they deal
305 with nontechnical topics (like this one). These kinds of restrictions
306 are acceptable because they don't obstruct the community's normal use
309 However, it must be possible to modify all the @emph{technical}
310 content of the manual, and then distribute the result in all the usual
311 media, through all the usual channels. Otherwise, the restrictions
312 obstruct the use of the manual, it is not free, and we need another
313 manual to replace it.
315 Please spread the word about this issue. Our community continues to
316 lose manuals to proprietary publishing. If we spread the word that
317 free software needs free reference manuals and free tutorials, perhaps
318 the next person who wants to contribute by writing documentation will
319 realize, before it is too late, that only free manuals contribute to
320 the free software community.
322 If you are writing documentation, please insist on publishing it under
323 the GNU Free Documentation License or another free documentation
324 license. Remember that this decision requires your approval---you
325 don't have to let the publisher decide. Some commercial publishers
326 will use a free license if you insist, but they will not propose the
327 option; it is up to you to raise the issue and say firmly that this is
328 what you want. If the publisher you are dealing with refuses, please
329 try other publishers. If you're not sure whether a proposed license
330 is free, write to @email{licensing@@gnu.org}.
332 You can encourage commercial publishers to sell more free, copylefted
333 manuals and tutorials by buying them, and particularly by buying
334 copies from the publishers that paid for their writing or for major
335 improvements. Meanwhile, try to avoid buying non-free documentation
336 at all. Check the distribution terms of a manual before you buy it,
337 and insist that whoever seeks your business must respect your freedom.
338 Check the history of the book, and try to reward the publishers that
339 have paid or pay the authors to work on it.
341 The Free Software Foundation maintains a list of free documentation
342 published by other publishers, at
343 @url{http://www.fsf.org/doc/other-free-books.html}.
346 @unnumberedsec Contributors to @value{GDBN}
348 Richard Stallman was the original author of @value{GDBN}, and of many
349 other @sc{gnu} programs. Many others have contributed to its
350 development. This section attempts to credit major contributors. One
351 of the virtues of free software is that everyone is free to contribute
352 to it; with regret, we cannot actually acknowledge everyone here. The
353 file @file{ChangeLog} in the @value{GDBN} distribution approximates a
354 blow-by-blow account.
356 Changes much prior to version 2.0 are lost in the mists of time.
359 @emph{Plea:} Additions to this section are particularly welcome. If you
360 or your friends (or enemies, to be evenhanded) have been unfairly
361 omitted from this list, we would like to add your names!
364 So that they may not regard their many labors as thankless, we
365 particularly thank those who shepherded @value{GDBN} through major
367 Andrew Cagney (releases 6.3, 6.2, 6.1, 6.0, 5.3, 5.2, 5.1 and 5.0);
368 Jim Blandy (release 4.18);
369 Jason Molenda (release 4.17);
370 Stan Shebs (release 4.14);
371 Fred Fish (releases 4.16, 4.15, 4.13, 4.12, 4.11, 4.10, and 4.9);
372 Stu Grossman and John Gilmore (releases 4.8, 4.7, 4.6, 4.5, and 4.4);
373 John Gilmore (releases 4.3, 4.2, 4.1, 4.0, and 3.9);
374 Jim Kingdon (releases 3.5, 3.4, and 3.3);
375 and Randy Smith (releases 3.2, 3.1, and 3.0).
377 Richard Stallman, assisted at various times by Peter TerMaat, Chris
378 Hanson, and Richard Mlynarik, handled releases through 2.8.
380 Michael Tiemann is the author of most of the @sc{gnu} C@t{++} support
381 in @value{GDBN}, with significant additional contributions from Per
382 Bothner and Daniel Berlin. James Clark wrote the @sc{gnu} C@t{++}
383 demangler. Early work on C@t{++} was by Peter TerMaat (who also did
384 much general update work leading to release 3.0).
386 @value{GDBN} uses the BFD subroutine library to examine multiple
387 object-file formats; BFD was a joint project of David V.
388 Henkel-Wallace, Rich Pixley, Steve Chamberlain, and John Gilmore.
390 David Johnson wrote the original COFF support; Pace Willison did
391 the original support for encapsulated COFF.
393 Brent Benson of Harris Computer Systems contributed DWARF 2 support.
395 Adam de Boor and Bradley Davis contributed the ISI Optimum V support.
396 Per Bothner, Noboyuki Hikichi, and Alessandro Forin contributed MIPS
398 Jean-Daniel Fekete contributed Sun 386i support.
399 Chris Hanson improved the HP9000 support.
400 Noboyuki Hikichi and Tomoyuki Hasei contributed Sony/News OS 3 support.
401 David Johnson contributed Encore Umax support.
402 Jyrki Kuoppala contributed Altos 3068 support.
403 Jeff Law contributed HP PA and SOM support.
404 Keith Packard contributed NS32K support.
405 Doug Rabson contributed Acorn Risc Machine support.
406 Bob Rusk contributed Harris Nighthawk CX-UX support.
407 Chris Smith contributed Convex support (and Fortran debugging).
408 Jonathan Stone contributed Pyramid support.
409 Michael Tiemann contributed SPARC support.
410 Tim Tucker contributed support for the Gould NP1 and Gould Powernode.
411 Pace Willison contributed Intel 386 support.
412 Jay Vosburgh contributed Symmetry support.
413 Marko Mlinar contributed OpenRISC 1000 support.
415 Andreas Schwab contributed M68K @sc{gnu}/Linux support.
417 Rich Schaefer and Peter Schauer helped with support of SunOS shared
420 Jay Fenlason and Roland McGrath ensured that @value{GDBN} and GAS agree
421 about several machine instruction sets.
423 Patrick Duval, Ted Goldstein, Vikram Koka and Glenn Engel helped develop
424 remote debugging. Intel Corporation, Wind River Systems, AMD, and ARM
425 contributed remote debugging modules for the i960, VxWorks, A29K UDI,
426 and RDI targets, respectively.
428 Brian Fox is the author of the readline libraries providing
429 command-line editing and command history.
431 Andrew Beers of SUNY Buffalo wrote the language-switching code, the
432 Modula-2 support, and contributed the Languages chapter of this manual.
434 Fred Fish wrote most of the support for Unix System Vr4.
435 He also enhanced the command-completion support to cover C@t{++} overloaded
438 Hitachi America (now Renesas America), Ltd. sponsored the support for
439 H8/300, H8/500, and Super-H processors.
441 NEC sponsored the support for the v850, Vr4xxx, and Vr5xxx processors.
443 Mitsubishi (now Renesas) sponsored the support for D10V, D30V, and M32R/D
446 Toshiba sponsored the support for the TX39 Mips processor.
448 Matsushita sponsored the support for the MN10200 and MN10300 processors.
450 Fujitsu sponsored the support for SPARClite and FR30 processors.
452 Kung Hsu, Jeff Law, and Rick Sladkey added support for hardware
455 Michael Snyder added support for tracepoints.
457 Stu Grossman wrote gdbserver.
459 Jim Kingdon, Peter Schauer, Ian Taylor, and Stu Grossman made
460 nearly innumerable bug fixes and cleanups throughout @value{GDBN}.
462 The following people at the Hewlett-Packard Company contributed
463 support for the PA-RISC 2.0 architecture, HP-UX 10.20, 10.30, and 11.0
464 (narrow mode), HP's implementation of kernel threads, HP's aC@t{++}
465 compiler, and the Text User Interface (nee Terminal User Interface):
466 Ben Krepp, Richard Title, John Bishop, Susan Macchia, Kathy Mann,
467 Satish Pai, India Paul, Steve Rehrauer, and Elena Zannoni. Kim Haase
468 provided HP-specific information in this manual.
470 DJ Delorie ported @value{GDBN} to MS-DOS, for the DJGPP project.
471 Robert Hoehne made significant contributions to the DJGPP port.
473 Cygnus Solutions has sponsored @value{GDBN} maintenance and much of its
474 development since 1991. Cygnus engineers who have worked on @value{GDBN}
475 fulltime include Mark Alexander, Jim Blandy, Per Bothner, Kevin
476 Buettner, Edith Epstein, Chris Faylor, Fred Fish, Martin Hunt, Jim
477 Ingham, John Gilmore, Stu Grossman, Kung Hsu, Jim Kingdon, John Metzler,
478 Fernando Nasser, Geoffrey Noer, Dawn Perchik, Rich Pixley, Zdenek
479 Radouch, Keith Seitz, Stan Shebs, David Taylor, and Elena Zannoni. In
480 addition, Dave Brolley, Ian Carmichael, Steve Chamberlain, Nick Clifton,
481 JT Conklin, Stan Cox, DJ Delorie, Ulrich Drepper, Frank Eigler, Doug
482 Evans, Sean Fagan, David Henkel-Wallace, Richard Henderson, Jeff
483 Holcomb, Jeff Law, Jim Lemke, Tom Lord, Bob Manson, Michael Meissner,
484 Jason Merrill, Catherine Moore, Drew Moseley, Ken Raeburn, Gavin
485 Romig-Koch, Rob Savoye, Jamie Smith, Mike Stump, Ian Taylor, Angela
486 Thomas, Michael Tiemann, Tom Tromey, Ron Unrau, Jim Wilson, and David
487 Zuhn have made contributions both large and small.
489 Andrew Cagney, Fernando Nasser, and Elena Zannoni, while working for
490 Cygnus Solutions, implemented the original @sc{gdb/mi} interface.
492 Jim Blandy added support for preprocessor macros, while working for Red
495 Andrew Cagney designed @value{GDBN}'s architecture vector. Many
496 people including Andrew Cagney, Stephane Carrez, Randolph Chung, Nick
497 Duffek, Richard Henderson, Mark Kettenis, Grace Sainsbury, Kei
498 Sakamoto, Yoshinori Sato, Michael Snyder, Andreas Schwab, Jason
499 Thorpe, Corinna Vinschen, Ulrich Weigand, and Elena Zannoni, helped
500 with the migration of old architectures to this new framework.
502 Andrew Cagney completely re-designed and re-implemented @value{GDBN}'s
503 unwinder framework, this consisting of a fresh new design featuring
504 frame IDs, independent frame sniffers, and the sentinel frame. Mark
505 Kettenis implemented the @sc{dwarf 2} unwinder, Jeff Johnston the
506 libunwind unwinder, and Andrew Cagney the dummy, sentinel, tramp, and
507 trad unwinders. The architecture-specific changes, each involving a
508 complete rewrite of the architecture's frame code, were carried out by
509 Jim Blandy, Joel Brobecker, Kevin Buettner, Andrew Cagney, Stephane
510 Carrez, Randolph Chung, Orjan Friberg, Richard Henderson, Daniel
511 Jacobowitz, Jeff Johnston, Mark Kettenis, Theodore A. Roth, Kei
512 Sakamoto, Yoshinori Sato, Michael Snyder, Corinna Vinschen, and Ulrich
515 Christian Zankel, Ross Morley, Bob Wilson, and Maxim Grigoriev from
516 Tensilica, Inc.@: contributed support for Xtensa processors. Others
517 who have worked on the Xtensa port of @value{GDBN} in the past include
518 Steve Tjiang, John Newlin, and Scott Foehner.
520 Michael Eager and staff of Xilinx, Inc., contributed support for the
521 Xilinx MicroBlaze architecture.
524 @chapter A Sample @value{GDBN} Session
526 You can use this manual at your leisure to read all about @value{GDBN}.
527 However, a handful of commands are enough to get started using the
528 debugger. This chapter illustrates those commands.
531 In this sample session, we emphasize user input like this: @b{input},
532 to make it easier to pick out from the surrounding output.
535 @c FIXME: this example may not be appropriate for some configs, where
536 @c FIXME...primary interest is in remote use.
538 One of the preliminary versions of @sc{gnu} @code{m4} (a generic macro
539 processor) exhibits the following bug: sometimes, when we change its
540 quote strings from the default, the commands used to capture one macro
541 definition within another stop working. In the following short @code{m4}
542 session, we define a macro @code{foo} which expands to @code{0000}; we
543 then use the @code{m4} built-in @code{defn} to define @code{bar} as the
544 same thing. However, when we change the open quote string to
545 @code{<QUOTE>} and the close quote string to @code{<UNQUOTE>}, the same
546 procedure fails to define a new synonym @code{baz}:
555 @b{define(bar,defn(`foo'))}
559 @b{changequote(<QUOTE>,<UNQUOTE>)}
561 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
564 m4: End of input: 0: fatal error: EOF in string
568 Let us use @value{GDBN} to try to see what is going on.
571 $ @b{@value{GDBP} m4}
572 @c FIXME: this falsifies the exact text played out, to permit smallbook
573 @c FIXME... format to come out better.
574 @value{GDBN} is free software and you are welcome to distribute copies
575 of it under certain conditions; type "show copying" to see
577 There is absolutely no warranty for @value{GDBN}; type "show warranty"
580 @value{GDBN} @value{GDBVN}, Copyright 1999 Free Software Foundation, Inc...
585 @value{GDBN} reads only enough symbol data to know where to find the
586 rest when needed; as a result, the first prompt comes up very quickly.
587 We now tell @value{GDBN} to use a narrower display width than usual, so
588 that examples fit in this manual.
591 (@value{GDBP}) @b{set width 70}
595 We need to see how the @code{m4} built-in @code{changequote} works.
596 Having looked at the source, we know the relevant subroutine is
597 @code{m4_changequote}, so we set a breakpoint there with the @value{GDBN}
598 @code{break} command.
601 (@value{GDBP}) @b{break m4_changequote}
602 Breakpoint 1 at 0x62f4: file builtin.c, line 879.
606 Using the @code{run} command, we start @code{m4} running under @value{GDBN}
607 control; as long as control does not reach the @code{m4_changequote}
608 subroutine, the program runs as usual:
611 (@value{GDBP}) @b{run}
612 Starting program: /work/Editorial/gdb/gnu/m4/m4
620 To trigger the breakpoint, we call @code{changequote}. @value{GDBN}
621 suspends execution of @code{m4}, displaying information about the
622 context where it stops.
625 @b{changequote(<QUOTE>,<UNQUOTE>)}
627 Breakpoint 1, m4_changequote (argc=3, argv=0x33c70)
629 879 if (bad_argc(TOKEN_DATA_TEXT(argv[0]),argc,1,3))
633 Now we use the command @code{n} (@code{next}) to advance execution to
634 the next line of the current function.
638 882 set_quotes((argc >= 2) ? TOKEN_DATA_TEXT(argv[1])\
643 @code{set_quotes} looks like a promising subroutine. We can go into it
644 by using the command @code{s} (@code{step}) instead of @code{next}.
645 @code{step} goes to the next line to be executed in @emph{any}
646 subroutine, so it steps into @code{set_quotes}.
650 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
652 530 if (lquote != def_lquote)
656 The display that shows the subroutine where @code{m4} is now
657 suspended (and its arguments) is called a stack frame display. It
658 shows a summary of the stack. We can use the @code{backtrace}
659 command (which can also be spelled @code{bt}), to see where we are
660 in the stack as a whole: the @code{backtrace} command displays a
661 stack frame for each active subroutine.
664 (@value{GDBP}) @b{bt}
665 #0 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
667 #1 0x6344 in m4_changequote (argc=3, argv=0x33c70)
669 #2 0x8174 in expand_macro (sym=0x33320) at macro.c:242
670 #3 0x7a88 in expand_token (obs=0x0, t=209696, td=0xf7fffa30)
672 #4 0x79dc in expand_input () at macro.c:40
673 #5 0x2930 in main (argc=0, argv=0xf7fffb20) at m4.c:195
677 We step through a few more lines to see what happens. The first two
678 times, we can use @samp{s}; the next two times we use @code{n} to avoid
679 falling into the @code{xstrdup} subroutine.
683 0x3b5c 532 if (rquote != def_rquote)
685 0x3b80 535 lquote = (lq == nil || *lq == '\0') ? \
686 def_lquote : xstrdup(lq);
688 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
691 538 len_lquote = strlen(rquote);
695 The last line displayed looks a little odd; we can examine the variables
696 @code{lquote} and @code{rquote} to see if they are in fact the new left
697 and right quotes we specified. We use the command @code{p}
698 (@code{print}) to see their values.
701 (@value{GDBP}) @b{p lquote}
702 $1 = 0x35d40 "<QUOTE>"
703 (@value{GDBP}) @b{p rquote}
704 $2 = 0x35d50 "<UNQUOTE>"
708 @code{lquote} and @code{rquote} are indeed the new left and right quotes.
709 To look at some context, we can display ten lines of source
710 surrounding the current line with the @code{l} (@code{list}) command.
716 535 lquote = (lq == nil || *lq == '\0') ? def_lquote\
718 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
721 538 len_lquote = strlen(rquote);
722 539 len_rquote = strlen(lquote);
729 Let us step past the two lines that set @code{len_lquote} and
730 @code{len_rquote}, and then examine the values of those variables.
734 539 len_rquote = strlen(lquote);
737 (@value{GDBP}) @b{p len_lquote}
739 (@value{GDBP}) @b{p len_rquote}
744 That certainly looks wrong, assuming @code{len_lquote} and
745 @code{len_rquote} are meant to be the lengths of @code{lquote} and
746 @code{rquote} respectively. We can set them to better values using
747 the @code{p} command, since it can print the value of
748 any expression---and that expression can include subroutine calls and
752 (@value{GDBP}) @b{p len_lquote=strlen(lquote)}
754 (@value{GDBP}) @b{p len_rquote=strlen(rquote)}
759 Is that enough to fix the problem of using the new quotes with the
760 @code{m4} built-in @code{defn}? We can allow @code{m4} to continue
761 executing with the @code{c} (@code{continue}) command, and then try the
762 example that caused trouble initially:
768 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
775 Success! The new quotes now work just as well as the default ones. The
776 problem seems to have been just the two typos defining the wrong
777 lengths. We allow @code{m4} exit by giving it an EOF as input:
781 Program exited normally.
785 The message @samp{Program exited normally.} is from @value{GDBN}; it
786 indicates @code{m4} has finished executing. We can end our @value{GDBN}
787 session with the @value{GDBN} @code{quit} command.
790 (@value{GDBP}) @b{quit}
794 @chapter Getting In and Out of @value{GDBN}
796 This chapter discusses how to start @value{GDBN}, and how to get out of it.
800 type @samp{@value{GDBP}} to start @value{GDBN}.
802 type @kbd{quit} or @kbd{Ctrl-d} to exit.
806 * Invoking GDB:: How to start @value{GDBN}
807 * Quitting GDB:: How to quit @value{GDBN}
808 * Shell Commands:: How to use shell commands inside @value{GDBN}
809 * Logging Output:: How to log @value{GDBN}'s output to a file
813 @section Invoking @value{GDBN}
815 Invoke @value{GDBN} by running the program @code{@value{GDBP}}. Once started,
816 @value{GDBN} reads commands from the terminal until you tell it to exit.
818 You can also run @code{@value{GDBP}} with a variety of arguments and options,
819 to specify more of your debugging environment at the outset.
821 The command-line options described here are designed
822 to cover a variety of situations; in some environments, some of these
823 options may effectively be unavailable.
825 The most usual way to start @value{GDBN} is with one argument,
826 specifying an executable program:
829 @value{GDBP} @var{program}
833 You can also start with both an executable program and a core file
837 @value{GDBP} @var{program} @var{core}
840 You can, instead, specify a process ID as a second argument, if you want
841 to debug a running process:
844 @value{GDBP} @var{program} 1234
848 would attach @value{GDBN} to process @code{1234} (unless you also have a file
849 named @file{1234}; @value{GDBN} does check for a core file first).
851 Taking advantage of the second command-line argument requires a fairly
852 complete operating system; when you use @value{GDBN} as a remote
853 debugger attached to a bare board, there may not be any notion of
854 ``process'', and there is often no way to get a core dump. @value{GDBN}
855 will warn you if it is unable to attach or to read core dumps.
857 You can optionally have @code{@value{GDBP}} pass any arguments after the
858 executable file to the inferior using @code{--args}. This option stops
861 @value{GDBP} --args gcc -O2 -c foo.c
863 This will cause @code{@value{GDBP}} to debug @code{gcc}, and to set
864 @code{gcc}'s command-line arguments (@pxref{Arguments}) to @samp{-O2 -c foo.c}.
866 You can run @code{@value{GDBP}} without printing the front material, which describes
867 @value{GDBN}'s non-warranty, by specifying @code{-silent}:
874 You can further control how @value{GDBN} starts up by using command-line
875 options. @value{GDBN} itself can remind you of the options available.
885 to display all available options and briefly describe their use
886 (@samp{@value{GDBP} -h} is a shorter equivalent).
888 All options and command line arguments you give are processed
889 in sequential order. The order makes a difference when the
890 @samp{-x} option is used.
894 * File Options:: Choosing files
895 * Mode Options:: Choosing modes
896 * Startup:: What @value{GDBN} does during startup
900 @subsection Choosing Files
902 When @value{GDBN} starts, it reads any arguments other than options as
903 specifying an executable file and core file (or process ID). This is
904 the same as if the arguments were specified by the @samp{-se} and
905 @samp{-c} (or @samp{-p}) options respectively. (@value{GDBN} reads the
906 first argument that does not have an associated option flag as
907 equivalent to the @samp{-se} option followed by that argument; and the
908 second argument that does not have an associated option flag, if any, as
909 equivalent to the @samp{-c}/@samp{-p} option followed by that argument.)
910 If the second argument begins with a decimal digit, @value{GDBN} will
911 first attempt to attach to it as a process, and if that fails, attempt
912 to open it as a corefile. If you have a corefile whose name begins with
913 a digit, you can prevent @value{GDBN} from treating it as a pid by
914 prefixing it with @file{./}, e.g.@: @file{./12345}.
916 If @value{GDBN} has not been configured to included core file support,
917 such as for most embedded targets, then it will complain about a second
918 argument and ignore it.
920 Many options have both long and short forms; both are shown in the
921 following list. @value{GDBN} also recognizes the long forms if you truncate
922 them, so long as enough of the option is present to be unambiguous.
923 (If you prefer, you can flag option arguments with @samp{--} rather
924 than @samp{-}, though we illustrate the more usual convention.)
926 @c NOTE: the @cindex entries here use double dashes ON PURPOSE. This
927 @c way, both those who look for -foo and --foo in the index, will find
931 @item -symbols @var{file}
933 @cindex @code{--symbols}
935 Read symbol table from file @var{file}.
937 @item -exec @var{file}
939 @cindex @code{--exec}
941 Use file @var{file} as the executable file to execute when appropriate,
942 and for examining pure data in conjunction with a core dump.
946 Read symbol table from file @var{file} and use it as the executable
949 @item -core @var{file}
951 @cindex @code{--core}
953 Use file @var{file} as a core dump to examine.
955 @item -pid @var{number}
956 @itemx -p @var{number}
959 Connect to process ID @var{number}, as with the @code{attach} command.
961 @item -command @var{file}
963 @cindex @code{--command}
965 Execute @value{GDBN} commands from file @var{file}. @xref{Command
966 Files,, Command files}.
968 @item -eval-command @var{command}
969 @itemx -ex @var{command}
970 @cindex @code{--eval-command}
972 Execute a single @value{GDBN} command.
974 This option may be used multiple times to call multiple commands. It may
975 also be interleaved with @samp{-command} as required.
978 @value{GDBP} -ex 'target sim' -ex 'load' \
979 -x setbreakpoints -ex 'run' a.out
982 @item -directory @var{directory}
983 @itemx -d @var{directory}
984 @cindex @code{--directory}
986 Add @var{directory} to the path to search for source and script files.
990 @cindex @code{--readnow}
992 Read each symbol file's entire symbol table immediately, rather than
993 the default, which is to read it incrementally as it is needed.
994 This makes startup slower, but makes future operations faster.
999 @subsection Choosing Modes
1001 You can run @value{GDBN} in various alternative modes---for example, in
1002 batch mode or quiet mode.
1009 Do not execute commands found in any initialization files. Normally,
1010 @value{GDBN} executes the commands in these files after all the command
1011 options and arguments have been processed. @xref{Command Files,,Command
1017 @cindex @code{--quiet}
1018 @cindex @code{--silent}
1020 ``Quiet''. Do not print the introductory and copyright messages. These
1021 messages are also suppressed in batch mode.
1024 @cindex @code{--batch}
1025 Run in batch mode. Exit with status @code{0} after processing all the
1026 command files specified with @samp{-x} (and all commands from
1027 initialization files, if not inhibited with @samp{-n}). Exit with
1028 nonzero status if an error occurs in executing the @value{GDBN} commands
1029 in the command files.
1031 Batch mode may be useful for running @value{GDBN} as a filter, for
1032 example to download and run a program on another computer; in order to
1033 make this more useful, the message
1036 Program exited normally.
1040 (which is ordinarily issued whenever a program running under
1041 @value{GDBN} control terminates) is not issued when running in batch
1045 @cindex @code{--batch-silent}
1046 Run in batch mode exactly like @samp{-batch}, but totally silently. All
1047 @value{GDBN} output to @code{stdout} is prevented (@code{stderr} is
1048 unaffected). This is much quieter than @samp{-silent} and would be useless
1049 for an interactive session.
1051 This is particularly useful when using targets that give @samp{Loading section}
1052 messages, for example.
1054 Note that targets that give their output via @value{GDBN}, as opposed to
1055 writing directly to @code{stdout}, will also be made silent.
1057 @item -return-child-result
1058 @cindex @code{--return-child-result}
1059 The return code from @value{GDBN} will be the return code from the child
1060 process (the process being debugged), with the following exceptions:
1064 @value{GDBN} exits abnormally. E.g., due to an incorrect argument or an
1065 internal error. In this case the exit code is the same as it would have been
1066 without @samp{-return-child-result}.
1068 The user quits with an explicit value. E.g., @samp{quit 1}.
1070 The child process never runs, or is not allowed to terminate, in which case
1071 the exit code will be -1.
1074 This option is useful in conjunction with @samp{-batch} or @samp{-batch-silent},
1075 when @value{GDBN} is being used as a remote program loader or simulator
1080 @cindex @code{--nowindows}
1082 ``No windows''. If @value{GDBN} comes with a graphical user interface
1083 (GUI) built in, then this option tells @value{GDBN} to only use the command-line
1084 interface. If no GUI is available, this option has no effect.
1088 @cindex @code{--windows}
1090 If @value{GDBN} includes a GUI, then this option requires it to be
1093 @item -cd @var{directory}
1095 Run @value{GDBN} using @var{directory} as its working directory,
1096 instead of the current directory.
1100 @cindex @code{--fullname}
1102 @sc{gnu} Emacs sets this option when it runs @value{GDBN} as a
1103 subprocess. It tells @value{GDBN} to output the full file name and line
1104 number in a standard, recognizable fashion each time a stack frame is
1105 displayed (which includes each time your program stops). This
1106 recognizable format looks like two @samp{\032} characters, followed by
1107 the file name, line number and character position separated by colons,
1108 and a newline. The Emacs-to-@value{GDBN} interface program uses the two
1109 @samp{\032} characters as a signal to display the source code for the
1113 @cindex @code{--epoch}
1114 The Epoch Emacs-@value{GDBN} interface sets this option when it runs
1115 @value{GDBN} as a subprocess. It tells @value{GDBN} to modify its print
1116 routines so as to allow Epoch to display values of expressions in a
1119 @item -annotate @var{level}
1120 @cindex @code{--annotate}
1121 This option sets the @dfn{annotation level} inside @value{GDBN}. Its
1122 effect is identical to using @samp{set annotate @var{level}}
1123 (@pxref{Annotations}). The annotation @var{level} controls how much
1124 information @value{GDBN} prints together with its prompt, values of
1125 expressions, source lines, and other types of output. Level 0 is the
1126 normal, level 1 is for use when @value{GDBN} is run as a subprocess of
1127 @sc{gnu} Emacs, level 3 is the maximum annotation suitable for programs
1128 that control @value{GDBN}, and level 2 has been deprecated.
1130 The annotation mechanism has largely been superseded by @sc{gdb/mi}
1134 @cindex @code{--args}
1135 Change interpretation of command line so that arguments following the
1136 executable file are passed as command line arguments to the inferior.
1137 This option stops option processing.
1139 @item -baud @var{bps}
1141 @cindex @code{--baud}
1143 Set the line speed (baud rate or bits per second) of any serial
1144 interface used by @value{GDBN} for remote debugging.
1146 @item -l @var{timeout}
1148 Set the timeout (in seconds) of any communication used by @value{GDBN}
1149 for remote debugging.
1151 @item -tty @var{device}
1152 @itemx -t @var{device}
1153 @cindex @code{--tty}
1155 Run using @var{device} for your program's standard input and output.
1156 @c FIXME: kingdon thinks there is more to -tty. Investigate.
1158 @c resolve the situation of these eventually
1160 @cindex @code{--tui}
1161 Activate the @dfn{Text User Interface} when starting. The Text User
1162 Interface manages several text windows on the terminal, showing
1163 source, assembly, registers and @value{GDBN} command outputs
1164 (@pxref{TUI, ,@value{GDBN} Text User Interface}). Alternatively, the
1165 Text User Interface can be enabled by invoking the program
1166 @samp{@value{GDBTUI}}. Do not use this option if you run @value{GDBN} from
1167 Emacs (@pxref{Emacs, ,Using @value{GDBN} under @sc{gnu} Emacs}).
1170 @c @cindex @code{--xdb}
1171 @c Run in XDB compatibility mode, allowing the use of certain XDB commands.
1172 @c For information, see the file @file{xdb_trans.html}, which is usually
1173 @c installed in the directory @code{/opt/langtools/wdb/doc} on HP-UX
1176 @item -interpreter @var{interp}
1177 @cindex @code{--interpreter}
1178 Use the interpreter @var{interp} for interface with the controlling
1179 program or device. This option is meant to be set by programs which
1180 communicate with @value{GDBN} using it as a back end.
1181 @xref{Interpreters, , Command Interpreters}.
1183 @samp{--interpreter=mi} (or @samp{--interpreter=mi2}) causes
1184 @value{GDBN} to use the @dfn{@sc{gdb/mi} interface} (@pxref{GDB/MI, ,
1185 The @sc{gdb/mi} Interface}) included since @value{GDBN} version 6.0. The
1186 previous @sc{gdb/mi} interface, included in @value{GDBN} version 5.3 and
1187 selected with @samp{--interpreter=mi1}, is deprecated. Earlier
1188 @sc{gdb/mi} interfaces are no longer supported.
1191 @cindex @code{--write}
1192 Open the executable and core files for both reading and writing. This
1193 is equivalent to the @samp{set write on} command inside @value{GDBN}
1197 @cindex @code{--statistics}
1198 This option causes @value{GDBN} to print statistics about time and
1199 memory usage after it completes each command and returns to the prompt.
1202 @cindex @code{--version}
1203 This option causes @value{GDBN} to print its version number and
1204 no-warranty blurb, and exit.
1209 @subsection What @value{GDBN} Does During Startup
1210 @cindex @value{GDBN} startup
1212 Here's the description of what @value{GDBN} does during session startup:
1216 Sets up the command interpreter as specified by the command line
1217 (@pxref{Mode Options, interpreter}).
1221 Reads the system-wide @dfn{init file} (if @option{--with-system-gdbinit} was
1222 used when building @value{GDBN}; @pxref{System-wide configuration,
1223 ,System-wide configuration and settings}) and executes all the commands in
1227 Reads the init file (if any) in your home directory@footnote{On
1228 DOS/Windows systems, the home directory is the one pointed to by the
1229 @code{HOME} environment variable.} and executes all the commands in
1233 Processes command line options and operands.
1236 Reads and executes the commands from init file (if any) in the current
1237 working directory. This is only done if the current directory is
1238 different from your home directory. Thus, you can have more than one
1239 init file, one generic in your home directory, and another, specific
1240 to the program you are debugging, in the directory where you invoke
1244 Reads command files specified by the @samp{-x} option. @xref{Command
1245 Files}, for more details about @value{GDBN} command files.
1248 Reads the command history recorded in the @dfn{history file}.
1249 @xref{Command History}, for more details about the command history and the
1250 files where @value{GDBN} records it.
1253 Init files use the same syntax as @dfn{command files} (@pxref{Command
1254 Files}) and are processed by @value{GDBN} in the same way. The init
1255 file in your home directory can set options (such as @samp{set
1256 complaints}) that affect subsequent processing of command line options
1257 and operands. Init files are not executed if you use the @samp{-nx}
1258 option (@pxref{Mode Options, ,Choosing Modes}).
1260 To display the list of init files loaded by gdb at startup, you
1261 can use @kbd{gdb --help}.
1263 @cindex init file name
1264 @cindex @file{.gdbinit}
1265 @cindex @file{gdb.ini}
1266 The @value{GDBN} init files are normally called @file{.gdbinit}.
1267 The DJGPP port of @value{GDBN} uses the name @file{gdb.ini}, due to
1268 the limitations of file names imposed by DOS filesystems. The Windows
1269 ports of @value{GDBN} use the standard name, but if they find a
1270 @file{gdb.ini} file, they warn you about that and suggest to rename
1271 the file to the standard name.
1275 @section Quitting @value{GDBN}
1276 @cindex exiting @value{GDBN}
1277 @cindex leaving @value{GDBN}
1280 @kindex quit @r{[}@var{expression}@r{]}
1281 @kindex q @r{(@code{quit})}
1282 @item quit @r{[}@var{expression}@r{]}
1284 To exit @value{GDBN}, use the @code{quit} command (abbreviated
1285 @code{q}), or type an end-of-file character (usually @kbd{Ctrl-d}). If you
1286 do not supply @var{expression}, @value{GDBN} will terminate normally;
1287 otherwise it will terminate using the result of @var{expression} as the
1292 An interrupt (often @kbd{Ctrl-c}) does not exit from @value{GDBN}, but rather
1293 terminates the action of any @value{GDBN} command that is in progress and
1294 returns to @value{GDBN} command level. It is safe to type the interrupt
1295 character at any time because @value{GDBN} does not allow it to take effect
1296 until a time when it is safe.
1298 If you have been using @value{GDBN} to control an attached process or
1299 device, you can release it with the @code{detach} command
1300 (@pxref{Attach, ,Debugging an Already-running Process}).
1302 @node Shell Commands
1303 @section Shell Commands
1305 If you need to execute occasional shell commands during your
1306 debugging session, there is no need to leave or suspend @value{GDBN}; you can
1307 just use the @code{shell} command.
1311 @cindex shell escape
1312 @item shell @var{command string}
1313 Invoke a standard shell to execute @var{command string}.
1314 If it exists, the environment variable @code{SHELL} determines which
1315 shell to run. Otherwise @value{GDBN} uses the default shell
1316 (@file{/bin/sh} on Unix systems, @file{COMMAND.COM} on MS-DOS, etc.).
1319 The utility @code{make} is often needed in development environments.
1320 You do not have to use the @code{shell} command for this purpose in
1325 @cindex calling make
1326 @item make @var{make-args}
1327 Execute the @code{make} program with the specified
1328 arguments. This is equivalent to @samp{shell make @var{make-args}}.
1331 @node Logging Output
1332 @section Logging Output
1333 @cindex logging @value{GDBN} output
1334 @cindex save @value{GDBN} output to a file
1336 You may want to save the output of @value{GDBN} commands to a file.
1337 There are several commands to control @value{GDBN}'s logging.
1341 @item set logging on
1343 @item set logging off
1345 @cindex logging file name
1346 @item set logging file @var{file}
1347 Change the name of the current logfile. The default logfile is @file{gdb.txt}.
1348 @item set logging overwrite [on|off]
1349 By default, @value{GDBN} will append to the logfile. Set @code{overwrite} if
1350 you want @code{set logging on} to overwrite the logfile instead.
1351 @item set logging redirect [on|off]
1352 By default, @value{GDBN} output will go to both the terminal and the logfile.
1353 Set @code{redirect} if you want output to go only to the log file.
1354 @kindex show logging
1356 Show the current values of the logging settings.
1360 @chapter @value{GDBN} Commands
1362 You can abbreviate a @value{GDBN} command to the first few letters of the command
1363 name, if that abbreviation is unambiguous; and you can repeat certain
1364 @value{GDBN} commands by typing just @key{RET}. You can also use the @key{TAB}
1365 key to get @value{GDBN} to fill out the rest of a word in a command (or to
1366 show you the alternatives available, if there is more than one possibility).
1369 * Command Syntax:: How to give commands to @value{GDBN}
1370 * Completion:: Command completion
1371 * Help:: How to ask @value{GDBN} for help
1374 @node Command Syntax
1375 @section Command Syntax
1377 A @value{GDBN} command is a single line of input. There is no limit on
1378 how long it can be. It starts with a command name, which is followed by
1379 arguments whose meaning depends on the command name. For example, the
1380 command @code{step} accepts an argument which is the number of times to
1381 step, as in @samp{step 5}. You can also use the @code{step} command
1382 with no arguments. Some commands do not allow any arguments.
1384 @cindex abbreviation
1385 @value{GDBN} command names may always be truncated if that abbreviation is
1386 unambiguous. Other possible command abbreviations are listed in the
1387 documentation for individual commands. In some cases, even ambiguous
1388 abbreviations are allowed; for example, @code{s} is specially defined as
1389 equivalent to @code{step} even though there are other commands whose
1390 names start with @code{s}. You can test abbreviations by using them as
1391 arguments to the @code{help} command.
1393 @cindex repeating commands
1394 @kindex RET @r{(repeat last command)}
1395 A blank line as input to @value{GDBN} (typing just @key{RET}) means to
1396 repeat the previous command. Certain commands (for example, @code{run})
1397 will not repeat this way; these are commands whose unintentional
1398 repetition might cause trouble and which you are unlikely to want to
1399 repeat. User-defined commands can disable this feature; see
1400 @ref{Define, dont-repeat}.
1402 The @code{list} and @code{x} commands, when you repeat them with
1403 @key{RET}, construct new arguments rather than repeating
1404 exactly as typed. This permits easy scanning of source or memory.
1406 @value{GDBN} can also use @key{RET} in another way: to partition lengthy
1407 output, in a way similar to the common utility @code{more}
1408 (@pxref{Screen Size,,Screen Size}). Since it is easy to press one
1409 @key{RET} too many in this situation, @value{GDBN} disables command
1410 repetition after any command that generates this sort of display.
1412 @kindex # @r{(a comment)}
1414 Any text from a @kbd{#} to the end of the line is a comment; it does
1415 nothing. This is useful mainly in command files (@pxref{Command
1416 Files,,Command Files}).
1418 @cindex repeating command sequences
1419 @kindex Ctrl-o @r{(operate-and-get-next)}
1420 The @kbd{Ctrl-o} binding is useful for repeating a complex sequence of
1421 commands. This command accepts the current line, like @key{RET}, and
1422 then fetches the next line relative to the current line from the history
1426 @section Command Completion
1429 @cindex word completion
1430 @value{GDBN} can fill in the rest of a word in a command for you, if there is
1431 only one possibility; it can also show you what the valid possibilities
1432 are for the next word in a command, at any time. This works for @value{GDBN}
1433 commands, @value{GDBN} subcommands, and the names of symbols in your program.
1435 Press the @key{TAB} key whenever you want @value{GDBN} to fill out the rest
1436 of a word. If there is only one possibility, @value{GDBN} fills in the
1437 word, and waits for you to finish the command (or press @key{RET} to
1438 enter it). For example, if you type
1440 @c FIXME "@key" does not distinguish its argument sufficiently to permit
1441 @c complete accuracy in these examples; space introduced for clarity.
1442 @c If texinfo enhancements make it unnecessary, it would be nice to
1443 @c replace " @key" by "@key" in the following...
1445 (@value{GDBP}) info bre @key{TAB}
1449 @value{GDBN} fills in the rest of the word @samp{breakpoints}, since that is
1450 the only @code{info} subcommand beginning with @samp{bre}:
1453 (@value{GDBP}) info breakpoints
1457 You can either press @key{RET} at this point, to run the @code{info
1458 breakpoints} command, or backspace and enter something else, if
1459 @samp{breakpoints} does not look like the command you expected. (If you
1460 were sure you wanted @code{info breakpoints} in the first place, you
1461 might as well just type @key{RET} immediately after @samp{info bre},
1462 to exploit command abbreviations rather than command completion).
1464 If there is more than one possibility for the next word when you press
1465 @key{TAB}, @value{GDBN} sounds a bell. You can either supply more
1466 characters and try again, or just press @key{TAB} a second time;
1467 @value{GDBN} displays all the possible completions for that word. For
1468 example, you might want to set a breakpoint on a subroutine whose name
1469 begins with @samp{make_}, but when you type @kbd{b make_@key{TAB}} @value{GDBN}
1470 just sounds the bell. Typing @key{TAB} again displays all the
1471 function names in your program that begin with those characters, for
1475 (@value{GDBP}) b make_ @key{TAB}
1476 @exdent @value{GDBN} sounds bell; press @key{TAB} again, to see:
1477 make_a_section_from_file make_environ
1478 make_abs_section make_function_type
1479 make_blockvector make_pointer_type
1480 make_cleanup make_reference_type
1481 make_command make_symbol_completion_list
1482 (@value{GDBP}) b make_
1486 After displaying the available possibilities, @value{GDBN} copies your
1487 partial input (@samp{b make_} in the example) so you can finish the
1490 If you just want to see the list of alternatives in the first place, you
1491 can press @kbd{M-?} rather than pressing @key{TAB} twice. @kbd{M-?}
1492 means @kbd{@key{META} ?}. You can type this either by holding down a
1493 key designated as the @key{META} shift on your keyboard (if there is
1494 one) while typing @kbd{?}, or as @key{ESC} followed by @kbd{?}.
1496 @cindex quotes in commands
1497 @cindex completion of quoted strings
1498 Sometimes the string you need, while logically a ``word'', may contain
1499 parentheses or other characters that @value{GDBN} normally excludes from
1500 its notion of a word. To permit word completion to work in this
1501 situation, you may enclose words in @code{'} (single quote marks) in
1502 @value{GDBN} commands.
1504 The most likely situation where you might need this is in typing the
1505 name of a C@t{++} function. This is because C@t{++} allows function
1506 overloading (multiple definitions of the same function, distinguished
1507 by argument type). For example, when you want to set a breakpoint you
1508 may need to distinguish whether you mean the version of @code{name}
1509 that takes an @code{int} parameter, @code{name(int)}, or the version
1510 that takes a @code{float} parameter, @code{name(float)}. To use the
1511 word-completion facilities in this situation, type a single quote
1512 @code{'} at the beginning of the function name. This alerts
1513 @value{GDBN} that it may need to consider more information than usual
1514 when you press @key{TAB} or @kbd{M-?} to request word completion:
1517 (@value{GDBP}) b 'bubble( @kbd{M-?}
1518 bubble(double,double) bubble(int,int)
1519 (@value{GDBP}) b 'bubble(
1522 In some cases, @value{GDBN} can tell that completing a name requires using
1523 quotes. When this happens, @value{GDBN} inserts the quote for you (while
1524 completing as much as it can) if you do not type the quote in the first
1528 (@value{GDBP}) b bub @key{TAB}
1529 @exdent @value{GDBN} alters your input line to the following, and rings a bell:
1530 (@value{GDBP}) b 'bubble(
1534 In general, @value{GDBN} can tell that a quote is needed (and inserts it) if
1535 you have not yet started typing the argument list when you ask for
1536 completion on an overloaded symbol.
1538 For more information about overloaded functions, see @ref{C Plus Plus
1539 Expressions, ,C@t{++} Expressions}. You can use the command @code{set
1540 overload-resolution off} to disable overload resolution;
1541 see @ref{Debugging C Plus Plus, ,@value{GDBN} Features for C@t{++}}.
1543 @cindex completion of structure field names
1544 @cindex structure field name completion
1545 @cindex completion of union field names
1546 @cindex union field name completion
1547 When completing in an expression which looks up a field in a
1548 structure, @value{GDBN} also tries@footnote{The completer can be
1549 confused by certain kinds of invalid expressions. Also, it only
1550 examines the static type of the expression, not the dynamic type.} to
1551 limit completions to the field names available in the type of the
1555 (@value{GDBP}) p gdb_stdout.@kbd{M-?}
1556 magic to_delete to_fputs to_put to_rewind
1557 to_data to_flush to_isatty to_read to_write
1561 This is because the @code{gdb_stdout} is a variable of the type
1562 @code{struct ui_file} that is defined in @value{GDBN} sources as
1569 ui_file_flush_ftype *to_flush;
1570 ui_file_write_ftype *to_write;
1571 ui_file_fputs_ftype *to_fputs;
1572 ui_file_read_ftype *to_read;
1573 ui_file_delete_ftype *to_delete;
1574 ui_file_isatty_ftype *to_isatty;
1575 ui_file_rewind_ftype *to_rewind;
1576 ui_file_put_ftype *to_put;
1583 @section Getting Help
1584 @cindex online documentation
1587 You can always ask @value{GDBN} itself for information on its commands,
1588 using the command @code{help}.
1591 @kindex h @r{(@code{help})}
1594 You can use @code{help} (abbreviated @code{h}) with no arguments to
1595 display a short list of named classes of commands:
1599 List of classes of commands:
1601 aliases -- Aliases of other commands
1602 breakpoints -- Making program stop at certain points
1603 data -- Examining data
1604 files -- Specifying and examining files
1605 internals -- Maintenance commands
1606 obscure -- Obscure features
1607 running -- Running the program
1608 stack -- Examining the stack
1609 status -- Status inquiries
1610 support -- Support facilities
1611 tracepoints -- Tracing of program execution without
1612 stopping the program
1613 user-defined -- User-defined commands
1615 Type "help" followed by a class name for a list of
1616 commands in that class.
1617 Type "help" followed by command name for full
1619 Command name abbreviations are allowed if unambiguous.
1622 @c the above line break eliminates huge line overfull...
1624 @item help @var{class}
1625 Using one of the general help classes as an argument, you can get a
1626 list of the individual commands in that class. For example, here is the
1627 help display for the class @code{status}:
1630 (@value{GDBP}) help status
1635 @c Line break in "show" line falsifies real output, but needed
1636 @c to fit in smallbook page size.
1637 info -- Generic command for showing things
1638 about the program being debugged
1639 show -- Generic command for showing things
1642 Type "help" followed by command name for full
1644 Command name abbreviations are allowed if unambiguous.
1648 @item help @var{command}
1649 With a command name as @code{help} argument, @value{GDBN} displays a
1650 short paragraph on how to use that command.
1653 @item apropos @var{args}
1654 The @code{apropos} command searches through all of the @value{GDBN}
1655 commands, and their documentation, for the regular expression specified in
1656 @var{args}. It prints out all matches found. For example:
1667 set symbol-reloading -- Set dynamic symbol table reloading
1668 multiple times in one run
1669 show symbol-reloading -- Show dynamic symbol table reloading
1670 multiple times in one run
1675 @item complete @var{args}
1676 The @code{complete @var{args}} command lists all the possible completions
1677 for the beginning of a command. Use @var{args} to specify the beginning of the
1678 command you want completed. For example:
1684 @noindent results in:
1695 @noindent This is intended for use by @sc{gnu} Emacs.
1698 In addition to @code{help}, you can use the @value{GDBN} commands @code{info}
1699 and @code{show} to inquire about the state of your program, or the state
1700 of @value{GDBN} itself. Each command supports many topics of inquiry; this
1701 manual introduces each of them in the appropriate context. The listings
1702 under @code{info} and under @code{show} in the Index point to
1703 all the sub-commands. @xref{Index}.
1708 @kindex i @r{(@code{info})}
1710 This command (abbreviated @code{i}) is for describing the state of your
1711 program. For example, you can show the arguments passed to a function
1712 with @code{info args}, list the registers currently in use with @code{info
1713 registers}, or list the breakpoints you have set with @code{info breakpoints}.
1714 You can get a complete list of the @code{info} sub-commands with
1715 @w{@code{help info}}.
1719 You can assign the result of an expression to an environment variable with
1720 @code{set}. For example, you can set the @value{GDBN} prompt to a $-sign with
1721 @code{set prompt $}.
1725 In contrast to @code{info}, @code{show} is for describing the state of
1726 @value{GDBN} itself.
1727 You can change most of the things you can @code{show}, by using the
1728 related command @code{set}; for example, you can control what number
1729 system is used for displays with @code{set radix}, or simply inquire
1730 which is currently in use with @code{show radix}.
1733 To display all the settable parameters and their current
1734 values, you can use @code{show} with no arguments; you may also use
1735 @code{info set}. Both commands produce the same display.
1736 @c FIXME: "info set" violates the rule that "info" is for state of
1737 @c FIXME...program. Ck w/ GNU: "info set" to be called something else,
1738 @c FIXME...or change desc of rule---eg "state of prog and debugging session"?
1742 Here are three miscellaneous @code{show} subcommands, all of which are
1743 exceptional in lacking corresponding @code{set} commands:
1746 @kindex show version
1747 @cindex @value{GDBN} version number
1749 Show what version of @value{GDBN} is running. You should include this
1750 information in @value{GDBN} bug-reports. If multiple versions of
1751 @value{GDBN} are in use at your site, you may need to determine which
1752 version of @value{GDBN} you are running; as @value{GDBN} evolves, new
1753 commands are introduced, and old ones may wither away. Also, many
1754 system vendors ship variant versions of @value{GDBN}, and there are
1755 variant versions of @value{GDBN} in @sc{gnu}/Linux distributions as well.
1756 The version number is the same as the one announced when you start
1759 @kindex show copying
1760 @kindex info copying
1761 @cindex display @value{GDBN} copyright
1764 Display information about permission for copying @value{GDBN}.
1766 @kindex show warranty
1767 @kindex info warranty
1769 @itemx info warranty
1770 Display the @sc{gnu} ``NO WARRANTY'' statement, or a warranty,
1771 if your version of @value{GDBN} comes with one.
1776 @chapter Running Programs Under @value{GDBN}
1778 When you run a program under @value{GDBN}, you must first generate
1779 debugging information when you compile it.
1781 You may start @value{GDBN} with its arguments, if any, in an environment
1782 of your choice. If you are doing native debugging, you may redirect
1783 your program's input and output, debug an already running process, or
1784 kill a child process.
1787 * Compilation:: Compiling for debugging
1788 * Starting:: Starting your program
1789 * Arguments:: Your program's arguments
1790 * Environment:: Your program's environment
1792 * Working Directory:: Your program's working directory
1793 * Input/Output:: Your program's input and output
1794 * Attach:: Debugging an already-running process
1795 * Kill Process:: Killing the child process
1797 * Inferiors and Programs:: Debugging multiple inferiors and programs
1798 * Threads:: Debugging programs with multiple threads
1799 * Forks:: Debugging forks
1800 * Checkpoint/Restart:: Setting a @emph{bookmark} to return to later
1804 @section Compiling for Debugging
1806 In order to debug a program effectively, you need to generate
1807 debugging information when you compile it. This debugging information
1808 is stored in the object file; it describes the data type of each
1809 variable or function and the correspondence between source line numbers
1810 and addresses in the executable code.
1812 To request debugging information, specify the @samp{-g} option when you run
1815 Programs that are to be shipped to your customers are compiled with
1816 optimizations, using the @samp{-O} compiler option. However, some
1817 compilers are unable to handle the @samp{-g} and @samp{-O} options
1818 together. Using those compilers, you cannot generate optimized
1819 executables containing debugging information.
1821 @value{NGCC}, the @sc{gnu} C/C@t{++} compiler, supports @samp{-g} with or
1822 without @samp{-O}, making it possible to debug optimized code. We
1823 recommend that you @emph{always} use @samp{-g} whenever you compile a
1824 program. You may think your program is correct, but there is no sense
1825 in pushing your luck. For more information, see @ref{Optimized Code}.
1827 Older versions of the @sc{gnu} C compiler permitted a variant option
1828 @w{@samp{-gg}} for debugging information. @value{GDBN} no longer supports this
1829 format; if your @sc{gnu} C compiler has this option, do not use it.
1831 @value{GDBN} knows about preprocessor macros and can show you their
1832 expansion (@pxref{Macros}). Most compilers do not include information
1833 about preprocessor macros in the debugging information if you specify
1834 the @option{-g} flag alone, because this information is rather large.
1835 Version 3.1 and later of @value{NGCC}, the @sc{gnu} C compiler,
1836 provides macro information if you specify the options
1837 @option{-gdwarf-2} and @option{-g3}; the former option requests
1838 debugging information in the Dwarf 2 format, and the latter requests
1839 ``extra information''. In the future, we hope to find more compact
1840 ways to represent macro information, so that it can be included with
1845 @section Starting your Program
1851 @kindex r @r{(@code{run})}
1854 Use the @code{run} command to start your program under @value{GDBN}.
1855 You must first specify the program name (except on VxWorks) with an
1856 argument to @value{GDBN} (@pxref{Invocation, ,Getting In and Out of
1857 @value{GDBN}}), or by using the @code{file} or @code{exec-file} command
1858 (@pxref{Files, ,Commands to Specify Files}).
1862 If you are running your program in an execution environment that
1863 supports processes, @code{run} creates an inferior process and makes
1864 that process run your program. In some environments without processes,
1865 @code{run} jumps to the start of your program. Other targets,
1866 like @samp{remote}, are always running. If you get an error
1867 message like this one:
1870 The "remote" target does not support "run".
1871 Try "help target" or "continue".
1875 then use @code{continue} to run your program. You may need @code{load}
1876 first (@pxref{load}).
1878 The execution of a program is affected by certain information it
1879 receives from its superior. @value{GDBN} provides ways to specify this
1880 information, which you must do @emph{before} starting your program. (You
1881 can change it after starting your program, but such changes only affect
1882 your program the next time you start it.) This information may be
1883 divided into four categories:
1886 @item The @emph{arguments.}
1887 Specify the arguments to give your program as the arguments of the
1888 @code{run} command. If a shell is available on your target, the shell
1889 is used to pass the arguments, so that you may use normal conventions
1890 (such as wildcard expansion or variable substitution) in describing
1892 In Unix systems, you can control which shell is used with the
1893 @code{SHELL} environment variable.
1894 @xref{Arguments, ,Your Program's Arguments}.
1896 @item The @emph{environment.}
1897 Your program normally inherits its environment from @value{GDBN}, but you can
1898 use the @value{GDBN} commands @code{set environment} and @code{unset
1899 environment} to change parts of the environment that affect
1900 your program. @xref{Environment, ,Your Program's Environment}.
1902 @item The @emph{working directory.}
1903 Your program inherits its working directory from @value{GDBN}. You can set
1904 the @value{GDBN} working directory with the @code{cd} command in @value{GDBN}.
1905 @xref{Working Directory, ,Your Program's Working Directory}.
1907 @item The @emph{standard input and output.}
1908 Your program normally uses the same device for standard input and
1909 standard output as @value{GDBN} is using. You can redirect input and output
1910 in the @code{run} command line, or you can use the @code{tty} command to
1911 set a different device for your program.
1912 @xref{Input/Output, ,Your Program's Input and Output}.
1915 @emph{Warning:} While input and output redirection work, you cannot use
1916 pipes to pass the output of the program you are debugging to another
1917 program; if you attempt this, @value{GDBN} is likely to wind up debugging the
1921 When you issue the @code{run} command, your program begins to execute
1922 immediately. @xref{Stopping, ,Stopping and Continuing}, for discussion
1923 of how to arrange for your program to stop. Once your program has
1924 stopped, you may call functions in your program, using the @code{print}
1925 or @code{call} commands. @xref{Data, ,Examining Data}.
1927 If the modification time of your symbol file has changed since the last
1928 time @value{GDBN} read its symbols, @value{GDBN} discards its symbol
1929 table, and reads it again. When it does this, @value{GDBN} tries to retain
1930 your current breakpoints.
1935 @cindex run to main procedure
1936 The name of the main procedure can vary from language to language.
1937 With C or C@t{++}, the main procedure name is always @code{main}, but
1938 other languages such as Ada do not require a specific name for their
1939 main procedure. The debugger provides a convenient way to start the
1940 execution of the program and to stop at the beginning of the main
1941 procedure, depending on the language used.
1943 The @samp{start} command does the equivalent of setting a temporary
1944 breakpoint at the beginning of the main procedure and then invoking
1945 the @samp{run} command.
1947 @cindex elaboration phase
1948 Some programs contain an @dfn{elaboration} phase where some startup code is
1949 executed before the main procedure is called. This depends on the
1950 languages used to write your program. In C@t{++}, for instance,
1951 constructors for static and global objects are executed before
1952 @code{main} is called. It is therefore possible that the debugger stops
1953 before reaching the main procedure. However, the temporary breakpoint
1954 will remain to halt execution.
1956 Specify the arguments to give to your program as arguments to the
1957 @samp{start} command. These arguments will be given verbatim to the
1958 underlying @samp{run} command. Note that the same arguments will be
1959 reused if no argument is provided during subsequent calls to
1960 @samp{start} or @samp{run}.
1962 It is sometimes necessary to debug the program during elaboration. In
1963 these cases, using the @code{start} command would stop the execution of
1964 your program too late, as the program would have already completed the
1965 elaboration phase. Under these circumstances, insert breakpoints in your
1966 elaboration code before running your program.
1968 @kindex set exec-wrapper
1969 @item set exec-wrapper @var{wrapper}
1970 @itemx show exec-wrapper
1971 @itemx unset exec-wrapper
1972 When @samp{exec-wrapper} is set, the specified wrapper is used to
1973 launch programs for debugging. @value{GDBN} starts your program
1974 with a shell command of the form @kbd{exec @var{wrapper}
1975 @var{program}}. Quoting is added to @var{program} and its
1976 arguments, but not to @var{wrapper}, so you should add quotes if
1977 appropriate for your shell. The wrapper runs until it executes
1978 your program, and then @value{GDBN} takes control.
1980 You can use any program that eventually calls @code{execve} with
1981 its arguments as a wrapper. Several standard Unix utilities do
1982 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
1983 with @code{exec "$@@"} will also work.
1985 For example, you can use @code{env} to pass an environment variable to
1986 the debugged program, without setting the variable in your shell's
1990 (@value{GDBP}) set exec-wrapper env 'LD_PRELOAD=libtest.so'
1994 This command is available when debugging locally on most targets, excluding
1995 @sc{djgpp}, Cygwin, MS Windows, and QNX Neutrino.
1997 @kindex set disable-randomization
1998 @item set disable-randomization
1999 @itemx set disable-randomization on
2000 This option (enabled by default in @value{GDBN}) will turn off the native
2001 randomization of the virtual address space of the started program. This option
2002 is useful for multiple debugging sessions to make the execution better
2003 reproducible and memory addresses reusable across debugging sessions.
2005 This feature is implemented only on @sc{gnu}/Linux. You can get the same
2009 (@value{GDBP}) set exec-wrapper setarch `uname -m` -R
2012 @item set disable-randomization off
2013 Leave the behavior of the started executable unchanged. Some bugs rear their
2014 ugly heads only when the program is loaded at certain addresses. If your bug
2015 disappears when you run the program under @value{GDBN}, that might be because
2016 @value{GDBN} by default disables the address randomization on platforms, such
2017 as @sc{gnu}/Linux, which do that for stand-alone programs. Use @kbd{set
2018 disable-randomization off} to try to reproduce such elusive bugs.
2020 The virtual address space randomization is implemented only on @sc{gnu}/Linux.
2021 It protects the programs against some kinds of security attacks. In these
2022 cases the attacker needs to know the exact location of a concrete executable
2023 code. Randomizing its location makes it impossible to inject jumps misusing
2024 a code at its expected addresses.
2026 Prelinking shared libraries provides a startup performance advantage but it
2027 makes addresses in these libraries predictable for privileged processes by
2028 having just unprivileged access at the target system. Reading the shared
2029 library binary gives enough information for assembling the malicious code
2030 misusing it. Still even a prelinked shared library can get loaded at a new
2031 random address just requiring the regular relocation process during the
2032 startup. Shared libraries not already prelinked are always loaded at
2033 a randomly chosen address.
2035 Position independent executables (PIE) contain position independent code
2036 similar to the shared libraries and therefore such executables get loaded at
2037 a randomly chosen address upon startup. PIE executables always load even
2038 already prelinked shared libraries at a random address. You can build such
2039 executable using @command{gcc -fPIE -pie}.
2041 Heap (malloc storage), stack and custom mmap areas are always placed randomly
2042 (as long as the randomization is enabled).
2044 @item show disable-randomization
2045 Show the current setting of the explicit disable of the native randomization of
2046 the virtual address space of the started program.
2051 @section Your Program's Arguments
2053 @cindex arguments (to your program)
2054 The arguments to your program can be specified by the arguments of the
2056 They are passed to a shell, which expands wildcard characters and
2057 performs redirection of I/O, and thence to your program. Your
2058 @code{SHELL} environment variable (if it exists) specifies what shell
2059 @value{GDBN} uses. If you do not define @code{SHELL}, @value{GDBN} uses
2060 the default shell (@file{/bin/sh} on Unix).
2062 On non-Unix systems, the program is usually invoked directly by
2063 @value{GDBN}, which emulates I/O redirection via the appropriate system
2064 calls, and the wildcard characters are expanded by the startup code of
2065 the program, not by the shell.
2067 @code{run} with no arguments uses the same arguments used by the previous
2068 @code{run}, or those set by the @code{set args} command.
2073 Specify the arguments to be used the next time your program is run. If
2074 @code{set args} has no arguments, @code{run} executes your program
2075 with no arguments. Once you have run your program with arguments,
2076 using @code{set args} before the next @code{run} is the only way to run
2077 it again without arguments.
2081 Show the arguments to give your program when it is started.
2085 @section Your Program's Environment
2087 @cindex environment (of your program)
2088 The @dfn{environment} consists of a set of environment variables and
2089 their values. Environment variables conventionally record such things as
2090 your user name, your home directory, your terminal type, and your search
2091 path for programs to run. Usually you set up environment variables with
2092 the shell and they are inherited by all the other programs you run. When
2093 debugging, it can be useful to try running your program with a modified
2094 environment without having to start @value{GDBN} over again.
2098 @item path @var{directory}
2099 Add @var{directory} to the front of the @code{PATH} environment variable
2100 (the search path for executables) that will be passed to your program.
2101 The value of @code{PATH} used by @value{GDBN} does not change.
2102 You may specify several directory names, separated by whitespace or by a
2103 system-dependent separator character (@samp{:} on Unix, @samp{;} on
2104 MS-DOS and MS-Windows). If @var{directory} is already in the path, it
2105 is moved to the front, so it is searched sooner.
2107 You can use the string @samp{$cwd} to refer to whatever is the current
2108 working directory at the time @value{GDBN} searches the path. If you
2109 use @samp{.} instead, it refers to the directory where you executed the
2110 @code{path} command. @value{GDBN} replaces @samp{.} in the
2111 @var{directory} argument (with the current path) before adding
2112 @var{directory} to the search path.
2113 @c 'path' is explicitly nonrepeatable, but RMS points out it is silly to
2114 @c document that, since repeating it would be a no-op.
2118 Display the list of search paths for executables (the @code{PATH}
2119 environment variable).
2121 @kindex show environment
2122 @item show environment @r{[}@var{varname}@r{]}
2123 Print the value of environment variable @var{varname} to be given to
2124 your program when it starts. If you do not supply @var{varname},
2125 print the names and values of all environment variables to be given to
2126 your program. You can abbreviate @code{environment} as @code{env}.
2128 @kindex set environment
2129 @item set environment @var{varname} @r{[}=@var{value}@r{]}
2130 Set environment variable @var{varname} to @var{value}. The value
2131 changes for your program only, not for @value{GDBN} itself. @var{value} may
2132 be any string; the values of environment variables are just strings, and
2133 any interpretation is supplied by your program itself. The @var{value}
2134 parameter is optional; if it is eliminated, the variable is set to a
2136 @c "any string" here does not include leading, trailing
2137 @c blanks. Gnu asks: does anyone care?
2139 For example, this command:
2146 tells the debugged program, when subsequently run, that its user is named
2147 @samp{foo}. (The spaces around @samp{=} are used for clarity here; they
2148 are not actually required.)
2150 @kindex unset environment
2151 @item unset environment @var{varname}
2152 Remove variable @var{varname} from the environment to be passed to your
2153 program. This is different from @samp{set env @var{varname} =};
2154 @code{unset environment} removes the variable from the environment,
2155 rather than assigning it an empty value.
2158 @emph{Warning:} On Unix systems, @value{GDBN} runs your program using
2160 by your @code{SHELL} environment variable if it exists (or
2161 @code{/bin/sh} if not). If your @code{SHELL} variable names a shell
2162 that runs an initialization file---such as @file{.cshrc} for C-shell, or
2163 @file{.bashrc} for BASH---any variables you set in that file affect
2164 your program. You may wish to move setting of environment variables to
2165 files that are only run when you sign on, such as @file{.login} or
2168 @node Working Directory
2169 @section Your Program's Working Directory
2171 @cindex working directory (of your program)
2172 Each time you start your program with @code{run}, it inherits its
2173 working directory from the current working directory of @value{GDBN}.
2174 The @value{GDBN} working directory is initially whatever it inherited
2175 from its parent process (typically the shell), but you can specify a new
2176 working directory in @value{GDBN} with the @code{cd} command.
2178 The @value{GDBN} working directory also serves as a default for the commands
2179 that specify files for @value{GDBN} to operate on. @xref{Files, ,Commands to
2184 @cindex change working directory
2185 @item cd @var{directory}
2186 Set the @value{GDBN} working directory to @var{directory}.
2190 Print the @value{GDBN} working directory.
2193 It is generally impossible to find the current working directory of
2194 the process being debugged (since a program can change its directory
2195 during its run). If you work on a system where @value{GDBN} is
2196 configured with the @file{/proc} support, you can use the @code{info
2197 proc} command (@pxref{SVR4 Process Information}) to find out the
2198 current working directory of the debuggee.
2201 @section Your Program's Input and Output
2206 By default, the program you run under @value{GDBN} does input and output to
2207 the same terminal that @value{GDBN} uses. @value{GDBN} switches the terminal
2208 to its own terminal modes to interact with you, but it records the terminal
2209 modes your program was using and switches back to them when you continue
2210 running your program.
2213 @kindex info terminal
2215 Displays information recorded by @value{GDBN} about the terminal modes your
2219 You can redirect your program's input and/or output using shell
2220 redirection with the @code{run} command. For example,
2227 starts your program, diverting its output to the file @file{outfile}.
2230 @cindex controlling terminal
2231 Another way to specify where your program should do input and output is
2232 with the @code{tty} command. This command accepts a file name as
2233 argument, and causes this file to be the default for future @code{run}
2234 commands. It also resets the controlling terminal for the child
2235 process, for future @code{run} commands. For example,
2242 directs that processes started with subsequent @code{run} commands
2243 default to do input and output on the terminal @file{/dev/ttyb} and have
2244 that as their controlling terminal.
2246 An explicit redirection in @code{run} overrides the @code{tty} command's
2247 effect on the input/output device, but not its effect on the controlling
2250 When you use the @code{tty} command or redirect input in the @code{run}
2251 command, only the input @emph{for your program} is affected. The input
2252 for @value{GDBN} still comes from your terminal. @code{tty} is an alias
2253 for @code{set inferior-tty}.
2255 @cindex inferior tty
2256 @cindex set inferior controlling terminal
2257 You can use the @code{show inferior-tty} command to tell @value{GDBN} to
2258 display the name of the terminal that will be used for future runs of your
2262 @item set inferior-tty /dev/ttyb
2263 @kindex set inferior-tty
2264 Set the tty for the program being debugged to /dev/ttyb.
2266 @item show inferior-tty
2267 @kindex show inferior-tty
2268 Show the current tty for the program being debugged.
2272 @section Debugging an Already-running Process
2277 @item attach @var{process-id}
2278 This command attaches to a running process---one that was started
2279 outside @value{GDBN}. (@code{info files} shows your active
2280 targets.) The command takes as argument a process ID. The usual way to
2281 find out the @var{process-id} of a Unix process is with the @code{ps} utility,
2282 or with the @samp{jobs -l} shell command.
2284 @code{attach} does not repeat if you press @key{RET} a second time after
2285 executing the command.
2288 To use @code{attach}, your program must be running in an environment
2289 which supports processes; for example, @code{attach} does not work for
2290 programs on bare-board targets that lack an operating system. You must
2291 also have permission to send the process a signal.
2293 When you use @code{attach}, the debugger finds the program running in
2294 the process first by looking in the current working directory, then (if
2295 the program is not found) by using the source file search path
2296 (@pxref{Source Path, ,Specifying Source Directories}). You can also use
2297 the @code{file} command to load the program. @xref{Files, ,Commands to
2300 The first thing @value{GDBN} does after arranging to debug the specified
2301 process is to stop it. You can examine and modify an attached process
2302 with all the @value{GDBN} commands that are ordinarily available when
2303 you start processes with @code{run}. You can insert breakpoints; you
2304 can step and continue; you can modify storage. If you would rather the
2305 process continue running, you may use the @code{continue} command after
2306 attaching @value{GDBN} to the process.
2311 When you have finished debugging the attached process, you can use the
2312 @code{detach} command to release it from @value{GDBN} control. Detaching
2313 the process continues its execution. After the @code{detach} command,
2314 that process and @value{GDBN} become completely independent once more, and you
2315 are ready to @code{attach} another process or start one with @code{run}.
2316 @code{detach} does not repeat if you press @key{RET} again after
2317 executing the command.
2320 If you exit @value{GDBN} while you have an attached process, you detach
2321 that process. If you use the @code{run} command, you kill that process.
2322 By default, @value{GDBN} asks for confirmation if you try to do either of these
2323 things; you can control whether or not you need to confirm by using the
2324 @code{set confirm} command (@pxref{Messages/Warnings, ,Optional Warnings and
2328 @section Killing the Child Process
2333 Kill the child process in which your program is running under @value{GDBN}.
2336 This command is useful if you wish to debug a core dump instead of a
2337 running process. @value{GDBN} ignores any core dump file while your program
2340 On some operating systems, a program cannot be executed outside @value{GDBN}
2341 while you have breakpoints set on it inside @value{GDBN}. You can use the
2342 @code{kill} command in this situation to permit running your program
2343 outside the debugger.
2345 The @code{kill} command is also useful if you wish to recompile and
2346 relink your program, since on many systems it is impossible to modify an
2347 executable file while it is running in a process. In this case, when you
2348 next type @code{run}, @value{GDBN} notices that the file has changed, and
2349 reads the symbol table again (while trying to preserve your current
2350 breakpoint settings).
2352 @node Inferiors and Programs
2353 @section Debugging Multiple Inferiors and Programs
2355 @value{GDBN} lets you run and debug multiple programs in a single
2356 session. In addition, @value{GDBN} on some systems may let you run
2357 several programs simultaneously (otherwise you have to exit from one
2358 before starting another). In the most general case, you can have
2359 multiple threads of execution in each of multiple processes, launched
2360 from multiple executables.
2363 @value{GDBN} represents the state of each program execution with an
2364 object called an @dfn{inferior}. An inferior typically corresponds to
2365 a process, but is more general and applies also to targets that do not
2366 have processes. Inferiors may be created before a process runs, and
2367 may be retained after a process exits. Inferiors have unique
2368 identifiers that are different from process ids. Usually each
2369 inferior will also have its own distinct address space, although some
2370 embedded targets may have several inferiors running in different parts
2371 of a single address space. Each inferior may in turn have multiple
2372 threads running in it.
2374 To find out what inferiors exist at any moment, use @w{@code{info
2378 @kindex info inferiors
2379 @item info inferiors
2380 Print a list of all inferiors currently being managed by @value{GDBN}.
2382 @value{GDBN} displays for each inferior (in this order):
2386 the inferior number assigned by @value{GDBN}
2389 the target system's inferior identifier
2392 the name of the executable the inferior is running.
2397 An asterisk @samp{*} preceding the @value{GDBN} inferior number
2398 indicates the current inferior.
2402 @c end table here to get a little more width for example
2405 (@value{GDBP}) info inferiors
2406 Num Description Executable
2407 2 process 2307 hello
2408 * 1 process 3401 goodbye
2411 To switch focus between inferiors, use the @code{inferior} command:
2414 @kindex inferior @var{infno}
2415 @item inferior @var{infno}
2416 Make inferior number @var{infno} the current inferior. The argument
2417 @var{infno} is the inferior number assigned by @value{GDBN}, as shown
2418 in the first field of the @samp{info inferiors} display.
2422 You can get multiple executables into a debugging session via the
2423 @code{add-inferior} and @w{@code{clone-inferior}} commands. On some
2424 systems @value{GDBN} can add inferiors to the debug session
2425 automatically by following calls to @code{fork} and @code{exec}. To
2426 remove inferiors from the debugging session use the
2427 @w{@code{remove-inferior}} command.
2430 @kindex add-inferior
2431 @item add-inferior [ -copies @var{n} ] [ -exec @var{executable} ]
2432 Adds @var{n} inferiors to be run using @var{executable} as the
2433 executable. @var{n} defaults to 1. If no executable is specified,
2434 the inferiors begins empty, with no program. You can still assign or
2435 change the program assigned to the inferior at any time by using the
2436 @code{file} command with the executable name as its argument.
2438 @kindex clone-inferior
2439 @item clone-inferior [ -copies @var{n} ] [ @var{infno} ]
2440 Adds @var{n} inferiors ready to execute the same program as inferior
2441 @var{infno}. @var{n} defaults to 1. @var{infno} defaults to the
2442 number of the current inferior. This is a convenient command when you
2443 want to run another instance of the inferior you are debugging.
2446 (@value{GDBP}) info inferiors
2447 Num Description Executable
2448 * 1 process 29964 helloworld
2449 (@value{GDBP}) clone-inferior
2452 (@value{GDBP}) info inferiors
2453 Num Description Executable
2455 * 1 process 29964 helloworld
2458 You can now simply switch focus to inferior 2 and run it.
2460 @kindex remove-inferior
2461 @item remove-inferior @var{infno}
2462 Removes the inferior @var{infno}. It is not possible to remove an
2463 inferior that is running with this command. For those, use the
2464 @code{kill} or @code{detach} command first.
2468 To quit debugging one of the running inferiors that is not the current
2469 inferior, you can either detach from it by using the @w{@code{detach
2470 inferior}} command (allowing it to run independently), or kill it
2471 using the @w{@code{kill inferior}} command:
2474 @kindex detach inferior @var{infno}
2475 @item detach inferior @var{infno}
2476 Detach from the inferior identified by @value{GDBN} inferior number
2477 @var{infno}, and remove it from the inferior list.
2479 @kindex kill inferior @var{infno}
2480 @item kill inferior @var{infno}
2481 Kill the inferior identified by @value{GDBN} inferior number
2482 @var{infno}, and remove it from the inferior list.
2485 After the successful completion of a command such as @code{detach},
2486 @code{detach inferior}, @code{kill} or @code{kill inferior}, or after
2487 a normal process exit, the inferior is still valid and listed with
2488 @code{info inferiors}, ready to be restarted.
2491 To be notified when inferiors are started or exit under @value{GDBN}'s
2492 control use @w{@code{set print inferior-events}}:
2495 @kindex set print inferior-events
2496 @cindex print messages on inferior start and exit
2497 @item set print inferior-events
2498 @itemx set print inferior-events on
2499 @itemx set print inferior-events off
2500 The @code{set print inferior-events} command allows you to enable or
2501 disable printing of messages when @value{GDBN} notices that new
2502 inferiors have started or that inferiors have exited or have been
2503 detached. By default, these messages will not be printed.
2505 @kindex show print inferior-events
2506 @item show print inferior-events
2507 Show whether messages will be printed when @value{GDBN} detects that
2508 inferiors have started, exited or have been detached.
2511 Many commands will work the same with multiple programs as with a
2512 single program: e.g., @code{print myglobal} will simply display the
2513 value of @code{myglobal} in the current inferior.
2516 Occasionaly, when debugging @value{GDBN} itself, it may be useful to
2517 get more info about the relationship of inferiors, programs, address
2518 spaces in a debug session. You can do that with the @w{@code{maint
2519 info program-spaces}} command.
2522 @kindex maint info program-spaces
2523 @item maint info program-spaces
2524 Print a list of all program spaces currently being managed by
2527 @value{GDBN} displays for each program space (in this order):
2531 the program space number assigned by @value{GDBN}
2534 the name of the executable loaded into the program space, with e.g.,
2535 the @code{file} command.
2540 An asterisk @samp{*} preceding the @value{GDBN} program space number
2541 indicates the current program space.
2543 In addition, below each program space line, @value{GDBN} prints extra
2544 information that isn't suitable to display in tabular form. For
2545 example, the list of inferiors bound to the program space.
2548 (@value{GDBP}) maint info program-spaces
2551 Bound inferiors: ID 1 (process 21561)
2555 Here we can see that no inferior is running the program @code{hello},
2556 while @code{process 21561} is running the program @code{goodbye}. On
2557 some targets, it is possible that multiple inferiors are bound to the
2558 same program space. The most common example is that of debugging both
2559 the parent and child processes of a @code{vfork} call. For example,
2562 (@value{GDBP}) maint info program-spaces
2565 Bound inferiors: ID 2 (process 18050), ID 1 (process 18045)
2568 Here, both inferior 2 and inferior 1 are running in the same program
2569 space as a result of inferior 1 having executed a @code{vfork} call.
2573 @section Debugging Programs with Multiple Threads
2575 @cindex threads of execution
2576 @cindex multiple threads
2577 @cindex switching threads
2578 In some operating systems, such as HP-UX and Solaris, a single program
2579 may have more than one @dfn{thread} of execution. The precise semantics
2580 of threads differ from one operating system to another, but in general
2581 the threads of a single program are akin to multiple processes---except
2582 that they share one address space (that is, they can all examine and
2583 modify the same variables). On the other hand, each thread has its own
2584 registers and execution stack, and perhaps private memory.
2586 @value{GDBN} provides these facilities for debugging multi-thread
2590 @item automatic notification of new threads
2591 @item @samp{thread @var{threadno}}, a command to switch among threads
2592 @item @samp{info threads}, a command to inquire about existing threads
2593 @item @samp{thread apply [@var{threadno}] [@var{all}] @var{args}},
2594 a command to apply a command to a list of threads
2595 @item thread-specific breakpoints
2596 @item @samp{set print thread-events}, which controls printing of
2597 messages on thread start and exit.
2598 @item @samp{set libthread-db-search-path @var{path}}, which lets
2599 the user specify which @code{libthread_db} to use if the default choice
2600 isn't compatible with the program.
2604 @emph{Warning:} These facilities are not yet available on every
2605 @value{GDBN} configuration where the operating system supports threads.
2606 If your @value{GDBN} does not support threads, these commands have no
2607 effect. For example, a system without thread support shows no output
2608 from @samp{info threads}, and always rejects the @code{thread} command,
2612 (@value{GDBP}) info threads
2613 (@value{GDBP}) thread 1
2614 Thread ID 1 not known. Use the "info threads" command to
2615 see the IDs of currently known threads.
2617 @c FIXME to implementors: how hard would it be to say "sorry, this GDB
2618 @c doesn't support threads"?
2621 @cindex focus of debugging
2622 @cindex current thread
2623 The @value{GDBN} thread debugging facility allows you to observe all
2624 threads while your program runs---but whenever @value{GDBN} takes
2625 control, one thread in particular is always the focus of debugging.
2626 This thread is called the @dfn{current thread}. Debugging commands show
2627 program information from the perspective of the current thread.
2629 @cindex @code{New} @var{systag} message
2630 @cindex thread identifier (system)
2631 @c FIXME-implementors!! It would be more helpful if the [New...] message
2632 @c included GDB's numeric thread handle, so you could just go to that
2633 @c thread without first checking `info threads'.
2634 Whenever @value{GDBN} detects a new thread in your program, it displays
2635 the target system's identification for the thread with a message in the
2636 form @samp{[New @var{systag}]}. @var{systag} is a thread identifier
2637 whose form varies depending on the particular system. For example, on
2638 @sc{gnu}/Linux, you might see
2641 [New Thread 46912507313328 (LWP 25582)]
2645 when @value{GDBN} notices a new thread. In contrast, on an SGI system,
2646 the @var{systag} is simply something like @samp{process 368}, with no
2649 @c FIXME!! (1) Does the [New...] message appear even for the very first
2650 @c thread of a program, or does it only appear for the
2651 @c second---i.e.@: when it becomes obvious we have a multithread
2653 @c (2) *Is* there necessarily a first thread always? Or do some
2654 @c multithread systems permit starting a program with multiple
2655 @c threads ab initio?
2657 @cindex thread number
2658 @cindex thread identifier (GDB)
2659 For debugging purposes, @value{GDBN} associates its own thread
2660 number---always a single integer---with each thread in your program.
2663 @kindex info threads
2665 Display a summary of all threads currently in your
2666 program. @value{GDBN} displays for each thread (in this order):
2670 the thread number assigned by @value{GDBN}
2673 the target system's thread identifier (@var{systag})
2676 the current stack frame summary for that thread
2680 An asterisk @samp{*} to the left of the @value{GDBN} thread number
2681 indicates the current thread.
2685 @c end table here to get a little more width for example
2688 (@value{GDBP}) info threads
2689 3 process 35 thread 27 0x34e5 in sigpause ()
2690 2 process 35 thread 23 0x34e5 in sigpause ()
2691 * 1 process 35 thread 13 main (argc=1, argv=0x7ffffff8)
2697 @cindex debugging multithreaded programs (on HP-UX)
2698 @cindex thread identifier (GDB), on HP-UX
2699 For debugging purposes, @value{GDBN} associates its own thread
2700 number---a small integer assigned in thread-creation order---with each
2701 thread in your program.
2703 @cindex @code{New} @var{systag} message, on HP-UX
2704 @cindex thread identifier (system), on HP-UX
2705 @c FIXME-implementors!! It would be more helpful if the [New...] message
2706 @c included GDB's numeric thread handle, so you could just go to that
2707 @c thread without first checking `info threads'.
2708 Whenever @value{GDBN} detects a new thread in your program, it displays
2709 both @value{GDBN}'s thread number and the target system's identification for the thread with a message in the
2710 form @samp{[New @var{systag}]}. @var{systag} is a thread identifier
2711 whose form varies depending on the particular system. For example, on
2715 [New thread 2 (system thread 26594)]
2719 when @value{GDBN} notices a new thread.
2722 @kindex info threads (HP-UX)
2724 Display a summary of all threads currently in your
2725 program. @value{GDBN} displays for each thread (in this order):
2728 @item the thread number assigned by @value{GDBN}
2730 @item the target system's thread identifier (@var{systag})
2732 @item the current stack frame summary for that thread
2736 An asterisk @samp{*} to the left of the @value{GDBN} thread number
2737 indicates the current thread.
2741 @c end table here to get a little more width for example
2744 (@value{GDBP}) info threads
2745 * 3 system thread 26607 worker (wptr=0x7b09c318 "@@") \@*
2747 2 system thread 26606 0x7b0030d8 in __ksleep () \@*
2748 from /usr/lib/libc.2
2749 1 system thread 27905 0x7b003498 in _brk () \@*
2750 from /usr/lib/libc.2
2753 On Solaris, you can display more information about user threads with a
2754 Solaris-specific command:
2757 @item maint info sol-threads
2758 @kindex maint info sol-threads
2759 @cindex thread info (Solaris)
2760 Display info on Solaris user threads.
2764 @kindex thread @var{threadno}
2765 @item thread @var{threadno}
2766 Make thread number @var{threadno} the current thread. The command
2767 argument @var{threadno} is the internal @value{GDBN} thread number, as
2768 shown in the first field of the @samp{info threads} display.
2769 @value{GDBN} responds by displaying the system identifier of the thread
2770 you selected, and its current stack frame summary:
2773 @c FIXME!! This example made up; find a @value{GDBN} w/threads and get real one
2774 (@value{GDBP}) thread 2
2775 [Switching to process 35 thread 23]
2776 0x34e5 in sigpause ()
2780 As with the @samp{[New @dots{}]} message, the form of the text after
2781 @samp{Switching to} depends on your system's conventions for identifying
2784 @kindex thread apply
2785 @cindex apply command to several threads
2786 @item thread apply [@var{threadno}] [@var{all}] @var{command}
2787 The @code{thread apply} command allows you to apply the named
2788 @var{command} to one or more threads. Specify the numbers of the
2789 threads that you want affected with the command argument
2790 @var{threadno}. It can be a single thread number, one of the numbers
2791 shown in the first field of the @samp{info threads} display; or it
2792 could be a range of thread numbers, as in @code{2-4}. To apply a
2793 command to all threads, type @kbd{thread apply all @var{command}}.
2795 @kindex set print thread-events
2796 @cindex print messages on thread start and exit
2797 @item set print thread-events
2798 @itemx set print thread-events on
2799 @itemx set print thread-events off
2800 The @code{set print thread-events} command allows you to enable or
2801 disable printing of messages when @value{GDBN} notices that new threads have
2802 started or that threads have exited. By default, these messages will
2803 be printed if detection of these events is supported by the target.
2804 Note that these messages cannot be disabled on all targets.
2806 @kindex show print thread-events
2807 @item show print thread-events
2808 Show whether messages will be printed when @value{GDBN} detects that threads
2809 have started and exited.
2812 @xref{Thread Stops,,Stopping and Starting Multi-thread Programs}, for
2813 more information about how @value{GDBN} behaves when you stop and start
2814 programs with multiple threads.
2816 @xref{Set Watchpoints,,Setting Watchpoints}, for information about
2817 watchpoints in programs with multiple threads.
2820 @kindex set libthread-db-search-path
2821 @cindex search path for @code{libthread_db}
2822 @item set libthread-db-search-path @r{[}@var{path}@r{]}
2823 If this variable is set, @var{path} is a colon-separated list of
2824 directories @value{GDBN} will use to search for @code{libthread_db}.
2825 If you omit @var{path}, @samp{libthread-db-search-path} will be reset to
2828 On @sc{gnu}/Linux and Solaris systems, @value{GDBN} uses a ``helper''
2829 @code{libthread_db} library to obtain information about threads in the
2830 inferior process. @value{GDBN} will use @samp{libthread-db-search-path}
2831 to find @code{libthread_db}. If that fails, @value{GDBN} will continue
2832 with default system shared library directories, and finally the directory
2833 from which @code{libpthread} was loaded in the inferior process.
2835 For any @code{libthread_db} library @value{GDBN} finds in above directories,
2836 @value{GDBN} attempts to initialize it with the current inferior process.
2837 If this initialization fails (which could happen because of a version
2838 mismatch between @code{libthread_db} and @code{libpthread}), @value{GDBN}
2839 will unload @code{libthread_db}, and continue with the next directory.
2840 If none of @code{libthread_db} libraries initialize successfully,
2841 @value{GDBN} will issue a warning and thread debugging will be disabled.
2843 Setting @code{libthread-db-search-path} is currently implemented
2844 only on some platforms.
2846 @kindex show libthread-db-search-path
2847 @item show libthread-db-search-path
2848 Display current libthread_db search path.
2852 @section Debugging Forks
2854 @cindex fork, debugging programs which call
2855 @cindex multiple processes
2856 @cindex processes, multiple
2857 On most systems, @value{GDBN} has no special support for debugging
2858 programs which create additional processes using the @code{fork}
2859 function. When a program forks, @value{GDBN} will continue to debug the
2860 parent process and the child process will run unimpeded. If you have
2861 set a breakpoint in any code which the child then executes, the child
2862 will get a @code{SIGTRAP} signal which (unless it catches the signal)
2863 will cause it to terminate.
2865 However, if you want to debug the child process there is a workaround
2866 which isn't too painful. Put a call to @code{sleep} in the code which
2867 the child process executes after the fork. It may be useful to sleep
2868 only if a certain environment variable is set, or a certain file exists,
2869 so that the delay need not occur when you don't want to run @value{GDBN}
2870 on the child. While the child is sleeping, use the @code{ps} program to
2871 get its process ID. Then tell @value{GDBN} (a new invocation of
2872 @value{GDBN} if you are also debugging the parent process) to attach to
2873 the child process (@pxref{Attach}). From that point on you can debug
2874 the child process just like any other process which you attached to.
2876 On some systems, @value{GDBN} provides support for debugging programs that
2877 create additional processes using the @code{fork} or @code{vfork} functions.
2878 Currently, the only platforms with this feature are HP-UX (11.x and later
2879 only?) and @sc{gnu}/Linux (kernel version 2.5.60 and later).
2881 By default, when a program forks, @value{GDBN} will continue to debug
2882 the parent process and the child process will run unimpeded.
2884 If you want to follow the child process instead of the parent process,
2885 use the command @w{@code{set follow-fork-mode}}.
2888 @kindex set follow-fork-mode
2889 @item set follow-fork-mode @var{mode}
2890 Set the debugger response to a program call of @code{fork} or
2891 @code{vfork}. A call to @code{fork} or @code{vfork} creates a new
2892 process. The @var{mode} argument can be:
2896 The original process is debugged after a fork. The child process runs
2897 unimpeded. This is the default.
2900 The new process is debugged after a fork. The parent process runs
2905 @kindex show follow-fork-mode
2906 @item show follow-fork-mode
2907 Display the current debugger response to a @code{fork} or @code{vfork} call.
2910 @cindex debugging multiple processes
2911 On Linux, if you want to debug both the parent and child processes, use the
2912 command @w{@code{set detach-on-fork}}.
2915 @kindex set detach-on-fork
2916 @item set detach-on-fork @var{mode}
2917 Tells gdb whether to detach one of the processes after a fork, or
2918 retain debugger control over them both.
2922 The child process (or parent process, depending on the value of
2923 @code{follow-fork-mode}) will be detached and allowed to run
2924 independently. This is the default.
2927 Both processes will be held under the control of @value{GDBN}.
2928 One process (child or parent, depending on the value of
2929 @code{follow-fork-mode}) is debugged as usual, while the other
2934 @kindex show detach-on-fork
2935 @item show detach-on-fork
2936 Show whether detach-on-fork mode is on/off.
2939 If you choose to set @samp{detach-on-fork} mode off, then @value{GDBN}
2940 will retain control of all forked processes (including nested forks).
2941 You can list the forked processes under the control of @value{GDBN} by
2942 using the @w{@code{info inferiors}} command, and switch from one fork
2943 to another by using the @code{inferior} command (@pxref{Inferiors and
2944 Programs, ,Debugging Multiple Inferiors and Programs}).
2946 To quit debugging one of the forked processes, you can either detach
2947 from it by using the @w{@code{detach inferior}} command (allowing it
2948 to run independently), or kill it using the @w{@code{kill inferior}}
2949 command. @xref{Inferiors and Programs, ,Debugging Multiple Inferiors
2952 If you ask to debug a child process and a @code{vfork} is followed by an
2953 @code{exec}, @value{GDBN} executes the new target up to the first
2954 breakpoint in the new target. If you have a breakpoint set on
2955 @code{main} in your original program, the breakpoint will also be set on
2956 the child process's @code{main}.
2958 On some systems, when a child process is spawned by @code{vfork}, you
2959 cannot debug the child or parent until an @code{exec} call completes.
2961 If you issue a @code{run} command to @value{GDBN} after an @code{exec}
2962 call executes, the new target restarts. To restart the parent
2963 process, use the @code{file} command with the parent executable name
2964 as its argument. By default, after an @code{exec} call executes,
2965 @value{GDBN} discards the symbols of the previous executable image.
2966 You can change this behaviour with the @w{@code{set follow-exec-mode}}
2970 @kindex set follow-exec-mode
2971 @item set follow-exec-mode @var{mode}
2973 Set debugger response to a program call of @code{exec}. An
2974 @code{exec} call replaces the program image of a process.
2976 @code{follow-exec-mode} can be:
2980 @value{GDBN} creates a new inferior and rebinds the process to this
2981 new inferior. The program the process was running before the
2982 @code{exec} call can be restarted afterwards by restarting the
2988 (@value{GDBP}) info inferiors
2990 Id Description Executable
2993 process 12020 is executing new program: prog2
2994 Program exited normally.
2995 (@value{GDBP}) info inferiors
2996 Id Description Executable
3002 @value{GDBN} keeps the process bound to the same inferior. The new
3003 executable image replaces the previous executable loaded in the
3004 inferior. Restarting the inferior after the @code{exec} call, with
3005 e.g., the @code{run} command, restarts the executable the process was
3006 running after the @code{exec} call. This is the default mode.
3011 (@value{GDBP}) info inferiors
3012 Id Description Executable
3015 process 12020 is executing new program: prog2
3016 Program exited normally.
3017 (@value{GDBP}) info inferiors
3018 Id Description Executable
3025 You can use the @code{catch} command to make @value{GDBN} stop whenever
3026 a @code{fork}, @code{vfork}, or @code{exec} call is made. @xref{Set
3027 Catchpoints, ,Setting Catchpoints}.
3029 @node Checkpoint/Restart
3030 @section Setting a @emph{Bookmark} to Return to Later
3035 @cindex snapshot of a process
3036 @cindex rewind program state
3038 On certain operating systems@footnote{Currently, only
3039 @sc{gnu}/Linux.}, @value{GDBN} is able to save a @dfn{snapshot} of a
3040 program's state, called a @dfn{checkpoint}, and come back to it
3043 Returning to a checkpoint effectively undoes everything that has
3044 happened in the program since the @code{checkpoint} was saved. This
3045 includes changes in memory, registers, and even (within some limits)
3046 system state. Effectively, it is like going back in time to the
3047 moment when the checkpoint was saved.
3049 Thus, if you're stepping thru a program and you think you're
3050 getting close to the point where things go wrong, you can save
3051 a checkpoint. Then, if you accidentally go too far and miss
3052 the critical statement, instead of having to restart your program
3053 from the beginning, you can just go back to the checkpoint and
3054 start again from there.
3056 This can be especially useful if it takes a lot of time or
3057 steps to reach the point where you think the bug occurs.
3059 To use the @code{checkpoint}/@code{restart} method of debugging:
3064 Save a snapshot of the debugged program's current execution state.
3065 The @code{checkpoint} command takes no arguments, but each checkpoint
3066 is assigned a small integer id, similar to a breakpoint id.
3068 @kindex info checkpoints
3069 @item info checkpoints
3070 List the checkpoints that have been saved in the current debugging
3071 session. For each checkpoint, the following information will be
3078 @item Source line, or label
3081 @kindex restart @var{checkpoint-id}
3082 @item restart @var{checkpoint-id}
3083 Restore the program state that was saved as checkpoint number
3084 @var{checkpoint-id}. All program variables, registers, stack frames
3085 etc.@: will be returned to the values that they had when the checkpoint
3086 was saved. In essence, gdb will ``wind back the clock'' to the point
3087 in time when the checkpoint was saved.
3089 Note that breakpoints, @value{GDBN} variables, command history etc.
3090 are not affected by restoring a checkpoint. In general, a checkpoint
3091 only restores things that reside in the program being debugged, not in
3094 @kindex delete checkpoint @var{checkpoint-id}
3095 @item delete checkpoint @var{checkpoint-id}
3096 Delete the previously-saved checkpoint identified by @var{checkpoint-id}.
3100 Returning to a previously saved checkpoint will restore the user state
3101 of the program being debugged, plus a significant subset of the system
3102 (OS) state, including file pointers. It won't ``un-write'' data from
3103 a file, but it will rewind the file pointer to the previous location,
3104 so that the previously written data can be overwritten. For files
3105 opened in read mode, the pointer will also be restored so that the
3106 previously read data can be read again.
3108 Of course, characters that have been sent to a printer (or other
3109 external device) cannot be ``snatched back'', and characters received
3110 from eg.@: a serial device can be removed from internal program buffers,
3111 but they cannot be ``pushed back'' into the serial pipeline, ready to
3112 be received again. Similarly, the actual contents of files that have
3113 been changed cannot be restored (at this time).
3115 However, within those constraints, you actually can ``rewind'' your
3116 program to a previously saved point in time, and begin debugging it
3117 again --- and you can change the course of events so as to debug a
3118 different execution path this time.
3120 @cindex checkpoints and process id
3121 Finally, there is one bit of internal program state that will be
3122 different when you return to a checkpoint --- the program's process
3123 id. Each checkpoint will have a unique process id (or @var{pid}),
3124 and each will be different from the program's original @var{pid}.
3125 If your program has saved a local copy of its process id, this could
3126 potentially pose a problem.
3128 @subsection A Non-obvious Benefit of Using Checkpoints
3130 On some systems such as @sc{gnu}/Linux, address space randomization
3131 is performed on new processes for security reasons. This makes it
3132 difficult or impossible to set a breakpoint, or watchpoint, on an
3133 absolute address if you have to restart the program, since the
3134 absolute location of a symbol will change from one execution to the
3137 A checkpoint, however, is an @emph{identical} copy of a process.
3138 Therefore if you create a checkpoint at (eg.@:) the start of main,
3139 and simply return to that checkpoint instead of restarting the
3140 process, you can avoid the effects of address randomization and
3141 your symbols will all stay in the same place.
3144 @chapter Stopping and Continuing
3146 The principal purposes of using a debugger are so that you can stop your
3147 program before it terminates; or so that, if your program runs into
3148 trouble, you can investigate and find out why.
3150 Inside @value{GDBN}, your program may stop for any of several reasons,
3151 such as a signal, a breakpoint, or reaching a new line after a
3152 @value{GDBN} command such as @code{step}. You may then examine and
3153 change variables, set new breakpoints or remove old ones, and then
3154 continue execution. Usually, the messages shown by @value{GDBN} provide
3155 ample explanation of the status of your program---but you can also
3156 explicitly request this information at any time.
3159 @kindex info program
3161 Display information about the status of your program: whether it is
3162 running or not, what process it is, and why it stopped.
3166 * Breakpoints:: Breakpoints, watchpoints, and catchpoints
3167 * Continuing and Stepping:: Resuming execution
3169 * Thread Stops:: Stopping and starting multi-thread programs
3173 @section Breakpoints, Watchpoints, and Catchpoints
3176 A @dfn{breakpoint} makes your program stop whenever a certain point in
3177 the program is reached. For each breakpoint, you can add conditions to
3178 control in finer detail whether your program stops. You can set
3179 breakpoints with the @code{break} command and its variants (@pxref{Set
3180 Breaks, ,Setting Breakpoints}), to specify the place where your program
3181 should stop by line number, function name or exact address in the
3184 On some systems, you can set breakpoints in shared libraries before
3185 the executable is run. There is a minor limitation on HP-UX systems:
3186 you must wait until the executable is run in order to set breakpoints
3187 in shared library routines that are not called directly by the program
3188 (for example, routines that are arguments in a @code{pthread_create}
3192 @cindex data breakpoints
3193 @cindex memory tracing
3194 @cindex breakpoint on memory address
3195 @cindex breakpoint on variable modification
3196 A @dfn{watchpoint} is a special breakpoint that stops your program
3197 when the value of an expression changes. The expression may be a value
3198 of a variable, or it could involve values of one or more variables
3199 combined by operators, such as @samp{a + b}. This is sometimes called
3200 @dfn{data breakpoints}. You must use a different command to set
3201 watchpoints (@pxref{Set Watchpoints, ,Setting Watchpoints}), but aside
3202 from that, you can manage a watchpoint like any other breakpoint: you
3203 enable, disable, and delete both breakpoints and watchpoints using the
3206 You can arrange to have values from your program displayed automatically
3207 whenever @value{GDBN} stops at a breakpoint. @xref{Auto Display,,
3211 @cindex breakpoint on events
3212 A @dfn{catchpoint} is another special breakpoint that stops your program
3213 when a certain kind of event occurs, such as the throwing of a C@t{++}
3214 exception or the loading of a library. As with watchpoints, you use a
3215 different command to set a catchpoint (@pxref{Set Catchpoints, ,Setting
3216 Catchpoints}), but aside from that, you can manage a catchpoint like any
3217 other breakpoint. (To stop when your program receives a signal, use the
3218 @code{handle} command; see @ref{Signals, ,Signals}.)
3220 @cindex breakpoint numbers
3221 @cindex numbers for breakpoints
3222 @value{GDBN} assigns a number to each breakpoint, watchpoint, or
3223 catchpoint when you create it; these numbers are successive integers
3224 starting with one. In many of the commands for controlling various
3225 features of breakpoints you use the breakpoint number to say which
3226 breakpoint you want to change. Each breakpoint may be @dfn{enabled} or
3227 @dfn{disabled}; if disabled, it has no effect on your program until you
3230 @cindex breakpoint ranges
3231 @cindex ranges of breakpoints
3232 Some @value{GDBN} commands accept a range of breakpoints on which to
3233 operate. A breakpoint range is either a single breakpoint number, like
3234 @samp{5}, or two such numbers, in increasing order, separated by a
3235 hyphen, like @samp{5-7}. When a breakpoint range is given to a command,
3236 all breakpoints in that range are operated on.
3239 * Set Breaks:: Setting breakpoints
3240 * Set Watchpoints:: Setting watchpoints
3241 * Set Catchpoints:: Setting catchpoints
3242 * Delete Breaks:: Deleting breakpoints
3243 * Disabling:: Disabling breakpoints
3244 * Conditions:: Break conditions
3245 * Break Commands:: Breakpoint command lists
3246 * Error in Breakpoints:: ``Cannot insert breakpoints''
3247 * Breakpoint-related Warnings:: ``Breakpoint address adjusted...''
3251 @subsection Setting Breakpoints
3253 @c FIXME LMB what does GDB do if no code on line of breakpt?
3254 @c consider in particular declaration with/without initialization.
3256 @c FIXME 2 is there stuff on this already? break at fun start, already init?
3259 @kindex b @r{(@code{break})}
3260 @vindex $bpnum@r{, convenience variable}
3261 @cindex latest breakpoint
3262 Breakpoints are set with the @code{break} command (abbreviated
3263 @code{b}). The debugger convenience variable @samp{$bpnum} records the
3264 number of the breakpoint you've set most recently; see @ref{Convenience
3265 Vars,, Convenience Variables}, for a discussion of what you can do with
3266 convenience variables.
3269 @item break @var{location}
3270 Set a breakpoint at the given @var{location}, which can specify a
3271 function name, a line number, or an address of an instruction.
3272 (@xref{Specify Location}, for a list of all the possible ways to
3273 specify a @var{location}.) The breakpoint will stop your program just
3274 before it executes any of the code in the specified @var{location}.
3276 When using source languages that permit overloading of symbols, such as
3277 C@t{++}, a function name may refer to more than one possible place to break.
3278 @xref{Ambiguous Expressions,,Ambiguous Expressions}, for a discussion of
3281 It is also possible to insert a breakpoint that will stop the program
3282 only if a specific thread (@pxref{Thread-Specific Breakpoints})
3283 or a specific task (@pxref{Ada Tasks}) hits that breakpoint.
3286 When called without any arguments, @code{break} sets a breakpoint at
3287 the next instruction to be executed in the selected stack frame
3288 (@pxref{Stack, ,Examining the Stack}). In any selected frame but the
3289 innermost, this makes your program stop as soon as control
3290 returns to that frame. This is similar to the effect of a
3291 @code{finish} command in the frame inside the selected frame---except
3292 that @code{finish} does not leave an active breakpoint. If you use
3293 @code{break} without an argument in the innermost frame, @value{GDBN} stops
3294 the next time it reaches the current location; this may be useful
3297 @value{GDBN} normally ignores breakpoints when it resumes execution, until at
3298 least one instruction has been executed. If it did not do this, you
3299 would be unable to proceed past a breakpoint without first disabling the
3300 breakpoint. This rule applies whether or not the breakpoint already
3301 existed when your program stopped.
3303 @item break @dots{} if @var{cond}
3304 Set a breakpoint with condition @var{cond}; evaluate the expression
3305 @var{cond} each time the breakpoint is reached, and stop only if the
3306 value is nonzero---that is, if @var{cond} evaluates as true.
3307 @samp{@dots{}} stands for one of the possible arguments described
3308 above (or no argument) specifying where to break. @xref{Conditions,
3309 ,Break Conditions}, for more information on breakpoint conditions.
3312 @item tbreak @var{args}
3313 Set a breakpoint enabled only for one stop. @var{args} are the
3314 same as for the @code{break} command, and the breakpoint is set in the same
3315 way, but the breakpoint is automatically deleted after the first time your
3316 program stops there. @xref{Disabling, ,Disabling Breakpoints}.
3319 @cindex hardware breakpoints
3320 @item hbreak @var{args}
3321 Set a hardware-assisted breakpoint. @var{args} are the same as for the
3322 @code{break} command and the breakpoint is set in the same way, but the
3323 breakpoint requires hardware support and some target hardware may not
3324 have this support. The main purpose of this is EPROM/ROM code
3325 debugging, so you can set a breakpoint at an instruction without
3326 changing the instruction. This can be used with the new trap-generation
3327 provided by SPARClite DSU and most x86-based targets. These targets
3328 will generate traps when a program accesses some data or instruction
3329 address that is assigned to the debug registers. However the hardware
3330 breakpoint registers can take a limited number of breakpoints. For
3331 example, on the DSU, only two data breakpoints can be set at a time, and
3332 @value{GDBN} will reject this command if more than two are used. Delete
3333 or disable unused hardware breakpoints before setting new ones
3334 (@pxref{Disabling, ,Disabling Breakpoints}).
3335 @xref{Conditions, ,Break Conditions}.
3336 For remote targets, you can restrict the number of hardware
3337 breakpoints @value{GDBN} will use, see @ref{set remote
3338 hardware-breakpoint-limit}.
3341 @item thbreak @var{args}
3342 Set a hardware-assisted breakpoint enabled only for one stop. @var{args}
3343 are the same as for the @code{hbreak} command and the breakpoint is set in
3344 the same way. However, like the @code{tbreak} command,
3345 the breakpoint is automatically deleted after the
3346 first time your program stops there. Also, like the @code{hbreak}
3347 command, the breakpoint requires hardware support and some target hardware
3348 may not have this support. @xref{Disabling, ,Disabling Breakpoints}.
3349 See also @ref{Conditions, ,Break Conditions}.
3352 @cindex regular expression
3353 @cindex breakpoints in functions matching a regexp
3354 @cindex set breakpoints in many functions
3355 @item rbreak @var{regex}
3356 Set breakpoints on all functions matching the regular expression
3357 @var{regex}. This command sets an unconditional breakpoint on all
3358 matches, printing a list of all breakpoints it set. Once these
3359 breakpoints are set, they are treated just like the breakpoints set with
3360 the @code{break} command. You can delete them, disable them, or make
3361 them conditional the same way as any other breakpoint.
3363 The syntax of the regular expression is the standard one used with tools
3364 like @file{grep}. Note that this is different from the syntax used by
3365 shells, so for instance @code{foo*} matches all functions that include
3366 an @code{fo} followed by zero or more @code{o}s. There is an implicit
3367 @code{.*} leading and trailing the regular expression you supply, so to
3368 match only functions that begin with @code{foo}, use @code{^foo}.
3370 @cindex non-member C@t{++} functions, set breakpoint in
3371 When debugging C@t{++} programs, @code{rbreak} is useful for setting
3372 breakpoints on overloaded functions that are not members of any special
3375 @cindex set breakpoints on all functions
3376 The @code{rbreak} command can be used to set breakpoints in
3377 @strong{all} the functions in a program, like this:
3380 (@value{GDBP}) rbreak .
3383 @kindex info breakpoints
3384 @cindex @code{$_} and @code{info breakpoints}
3385 @item info breakpoints @r{[}@var{n}@r{]}
3386 @itemx info break @r{[}@var{n}@r{]}
3387 @itemx info watchpoints @r{[}@var{n}@r{]}
3388 Print a table of all breakpoints, watchpoints, and catchpoints set and
3389 not deleted. Optional argument @var{n} means print information only
3390 about the specified breakpoint (or watchpoint or catchpoint). For
3391 each breakpoint, following columns are printed:
3394 @item Breakpoint Numbers
3396 Breakpoint, watchpoint, or catchpoint.
3398 Whether the breakpoint is marked to be disabled or deleted when hit.
3399 @item Enabled or Disabled
3400 Enabled breakpoints are marked with @samp{y}. @samp{n} marks breakpoints
3401 that are not enabled.
3403 Where the breakpoint is in your program, as a memory address. For a
3404 pending breakpoint whose address is not yet known, this field will
3405 contain @samp{<PENDING>}. Such breakpoint won't fire until a shared
3406 library that has the symbol or line referred by breakpoint is loaded.
3407 See below for details. A breakpoint with several locations will
3408 have @samp{<MULTIPLE>} in this field---see below for details.
3410 Where the breakpoint is in the source for your program, as a file and
3411 line number. For a pending breakpoint, the original string passed to
3412 the breakpoint command will be listed as it cannot be resolved until
3413 the appropriate shared library is loaded in the future.
3417 If a breakpoint is conditional, @code{info break} shows the condition on
3418 the line following the affected breakpoint; breakpoint commands, if any,
3419 are listed after that. A pending breakpoint is allowed to have a condition
3420 specified for it. The condition is not parsed for validity until a shared
3421 library is loaded that allows the pending breakpoint to resolve to a
3425 @code{info break} with a breakpoint
3426 number @var{n} as argument lists only that breakpoint. The
3427 convenience variable @code{$_} and the default examining-address for
3428 the @code{x} command are set to the address of the last breakpoint
3429 listed (@pxref{Memory, ,Examining Memory}).
3432 @code{info break} displays a count of the number of times the breakpoint
3433 has been hit. This is especially useful in conjunction with the
3434 @code{ignore} command. You can ignore a large number of breakpoint
3435 hits, look at the breakpoint info to see how many times the breakpoint
3436 was hit, and then run again, ignoring one less than that number. This
3437 will get you quickly to the last hit of that breakpoint.
3440 @value{GDBN} allows you to set any number of breakpoints at the same place in
3441 your program. There is nothing silly or meaningless about this. When
3442 the breakpoints are conditional, this is even useful
3443 (@pxref{Conditions, ,Break Conditions}).
3445 @cindex multiple locations, breakpoints
3446 @cindex breakpoints, multiple locations
3447 It is possible that a breakpoint corresponds to several locations
3448 in your program. Examples of this situation are:
3452 For a C@t{++} constructor, the @value{NGCC} compiler generates several
3453 instances of the function body, used in different cases.
3456 For a C@t{++} template function, a given line in the function can
3457 correspond to any number of instantiations.
3460 For an inlined function, a given source line can correspond to
3461 several places where that function is inlined.
3464 In all those cases, @value{GDBN} will insert a breakpoint at all
3465 the relevant locations@footnote{
3466 As of this writing, multiple-location breakpoints work only if there's
3467 line number information for all the locations. This means that they
3468 will generally not work in system libraries, unless you have debug
3469 info with line numbers for them.}.
3471 A breakpoint with multiple locations is displayed in the breakpoint
3472 table using several rows---one header row, followed by one row for
3473 each breakpoint location. The header row has @samp{<MULTIPLE>} in the
3474 address column. The rows for individual locations contain the actual
3475 addresses for locations, and show the functions to which those
3476 locations belong. The number column for a location is of the form
3477 @var{breakpoint-number}.@var{location-number}.
3482 Num Type Disp Enb Address What
3483 1 breakpoint keep y <MULTIPLE>
3485 breakpoint already hit 1 time
3486 1.1 y 0x080486a2 in void foo<int>() at t.cc:8
3487 1.2 y 0x080486ca in void foo<double>() at t.cc:8
3490 Each location can be individually enabled or disabled by passing
3491 @var{breakpoint-number}.@var{location-number} as argument to the
3492 @code{enable} and @code{disable} commands. Note that you cannot
3493 delete the individual locations from the list, you can only delete the
3494 entire list of locations that belong to their parent breakpoint (with
3495 the @kbd{delete @var{num}} command, where @var{num} is the number of
3496 the parent breakpoint, 1 in the above example). Disabling or enabling
3497 the parent breakpoint (@pxref{Disabling}) affects all of the locations
3498 that belong to that breakpoint.
3500 @cindex pending breakpoints
3501 It's quite common to have a breakpoint inside a shared library.
3502 Shared libraries can be loaded and unloaded explicitly,
3503 and possibly repeatedly, as the program is executed. To support
3504 this use case, @value{GDBN} updates breakpoint locations whenever
3505 any shared library is loaded or unloaded. Typically, you would
3506 set a breakpoint in a shared library at the beginning of your
3507 debugging session, when the library is not loaded, and when the
3508 symbols from the library are not available. When you try to set
3509 breakpoint, @value{GDBN} will ask you if you want to set
3510 a so called @dfn{pending breakpoint}---breakpoint whose address
3511 is not yet resolved.
3513 After the program is run, whenever a new shared library is loaded,
3514 @value{GDBN} reevaluates all the breakpoints. When a newly loaded
3515 shared library contains the symbol or line referred to by some
3516 pending breakpoint, that breakpoint is resolved and becomes an
3517 ordinary breakpoint. When a library is unloaded, all breakpoints
3518 that refer to its symbols or source lines become pending again.
3520 This logic works for breakpoints with multiple locations, too. For
3521 example, if you have a breakpoint in a C@t{++} template function, and
3522 a newly loaded shared library has an instantiation of that template,
3523 a new location is added to the list of locations for the breakpoint.
3525 Except for having unresolved address, pending breakpoints do not
3526 differ from regular breakpoints. You can set conditions or commands,
3527 enable and disable them and perform other breakpoint operations.
3529 @value{GDBN} provides some additional commands for controlling what
3530 happens when the @samp{break} command cannot resolve breakpoint
3531 address specification to an address:
3533 @kindex set breakpoint pending
3534 @kindex show breakpoint pending
3536 @item set breakpoint pending auto
3537 This is the default behavior. When @value{GDBN} cannot find the breakpoint
3538 location, it queries you whether a pending breakpoint should be created.
3540 @item set breakpoint pending on
3541 This indicates that an unrecognized breakpoint location should automatically
3542 result in a pending breakpoint being created.
3544 @item set breakpoint pending off
3545 This indicates that pending breakpoints are not to be created. Any
3546 unrecognized breakpoint location results in an error. This setting does
3547 not affect any pending breakpoints previously created.
3549 @item show breakpoint pending
3550 Show the current behavior setting for creating pending breakpoints.
3553 The settings above only affect the @code{break} command and its
3554 variants. Once breakpoint is set, it will be automatically updated
3555 as shared libraries are loaded and unloaded.
3557 @cindex automatic hardware breakpoints
3558 For some targets, @value{GDBN} can automatically decide if hardware or
3559 software breakpoints should be used, depending on whether the
3560 breakpoint address is read-only or read-write. This applies to
3561 breakpoints set with the @code{break} command as well as to internal
3562 breakpoints set by commands like @code{next} and @code{finish}. For
3563 breakpoints set with @code{hbreak}, @value{GDBN} will always use hardware
3566 You can control this automatic behaviour with the following commands::
3568 @kindex set breakpoint auto-hw
3569 @kindex show breakpoint auto-hw
3571 @item set breakpoint auto-hw on
3572 This is the default behavior. When @value{GDBN} sets a breakpoint, it
3573 will try to use the target memory map to decide if software or hardware
3574 breakpoint must be used.
3576 @item set breakpoint auto-hw off
3577 This indicates @value{GDBN} should not automatically select breakpoint
3578 type. If the target provides a memory map, @value{GDBN} will warn when
3579 trying to set software breakpoint at a read-only address.
3582 @value{GDBN} normally implements breakpoints by replacing the program code
3583 at the breakpoint address with a special instruction, which, when
3584 executed, given control to the debugger. By default, the program
3585 code is so modified only when the program is resumed. As soon as
3586 the program stops, @value{GDBN} restores the original instructions. This
3587 behaviour guards against leaving breakpoints inserted in the
3588 target should gdb abrubptly disconnect. However, with slow remote
3589 targets, inserting and removing breakpoint can reduce the performance.
3590 This behavior can be controlled with the following commands::
3592 @kindex set breakpoint always-inserted
3593 @kindex show breakpoint always-inserted
3595 @item set breakpoint always-inserted off
3596 All breakpoints, including newly added by the user, are inserted in
3597 the target only when the target is resumed. All breakpoints are
3598 removed from the target when it stops.
3600 @item set breakpoint always-inserted on
3601 Causes all breakpoints to be inserted in the target at all times. If
3602 the user adds a new breakpoint, or changes an existing breakpoint, the
3603 breakpoints in the target are updated immediately. A breakpoint is
3604 removed from the target only when breakpoint itself is removed.
3606 @cindex non-stop mode, and @code{breakpoint always-inserted}
3607 @item set breakpoint always-inserted auto
3608 This is the default mode. If @value{GDBN} is controlling the inferior
3609 in non-stop mode (@pxref{Non-Stop Mode}), gdb behaves as if
3610 @code{breakpoint always-inserted} mode is on. If @value{GDBN} is
3611 controlling the inferior in all-stop mode, @value{GDBN} behaves as if
3612 @code{breakpoint always-inserted} mode is off.
3615 @cindex negative breakpoint numbers
3616 @cindex internal @value{GDBN} breakpoints
3617 @value{GDBN} itself sometimes sets breakpoints in your program for
3618 special purposes, such as proper handling of @code{longjmp} (in C
3619 programs). These internal breakpoints are assigned negative numbers,
3620 starting with @code{-1}; @samp{info breakpoints} does not display them.
3621 You can see these breakpoints with the @value{GDBN} maintenance command
3622 @samp{maint info breakpoints} (@pxref{maint info breakpoints}).
3625 @node Set Watchpoints
3626 @subsection Setting Watchpoints
3628 @cindex setting watchpoints
3629 You can use a watchpoint to stop execution whenever the value of an
3630 expression changes, without having to predict a particular place where
3631 this may happen. (This is sometimes called a @dfn{data breakpoint}.)
3632 The expression may be as simple as the value of a single variable, or
3633 as complex as many variables combined by operators. Examples include:
3637 A reference to the value of a single variable.
3640 An address cast to an appropriate data type. For example,
3641 @samp{*(int *)0x12345678} will watch a 4-byte region at the specified
3642 address (assuming an @code{int} occupies 4 bytes).
3645 An arbitrarily complex expression, such as @samp{a*b + c/d}. The
3646 expression can use any operators valid in the program's native
3647 language (@pxref{Languages}).
3650 You can set a watchpoint on an expression even if the expression can
3651 not be evaluated yet. For instance, you can set a watchpoint on
3652 @samp{*global_ptr} before @samp{global_ptr} is initialized.
3653 @value{GDBN} will stop when your program sets @samp{global_ptr} and
3654 the expression produces a valid value. If the expression becomes
3655 valid in some other way than changing a variable (e.g.@: if the memory
3656 pointed to by @samp{*global_ptr} becomes readable as the result of a
3657 @code{malloc} call), @value{GDBN} may not stop until the next time
3658 the expression changes.
3660 @cindex software watchpoints
3661 @cindex hardware watchpoints
3662 Depending on your system, watchpoints may be implemented in software or
3663 hardware. @value{GDBN} does software watchpointing by single-stepping your
3664 program and testing the variable's value each time, which is hundreds of
3665 times slower than normal execution. (But this may still be worth it, to
3666 catch errors where you have no clue what part of your program is the
3669 On some systems, such as HP-UX, PowerPC, @sc{gnu}/Linux and most other
3670 x86-based targets, @value{GDBN} includes support for hardware
3671 watchpoints, which do not slow down the running of your program.
3675 @item watch @var{expr} @r{[}thread @var{threadnum}@r{]}
3676 Set a watchpoint for an expression. @value{GDBN} will break when the
3677 expression @var{expr} is written into by the program and its value
3678 changes. The simplest (and the most popular) use of this command is
3679 to watch the value of a single variable:
3682 (@value{GDBP}) watch foo
3685 If the command includes a @code{@r{[}thread @var{threadnum}@r{]}}
3686 clause, @value{GDBN} breaks only when the thread identified by
3687 @var{threadnum} changes the value of @var{expr}. If any other threads
3688 change the value of @var{expr}, @value{GDBN} will not break. Note
3689 that watchpoints restricted to a single thread in this way only work
3690 with Hardware Watchpoints.
3693 @item rwatch @var{expr} @r{[}thread @var{threadnum}@r{]}
3694 Set a watchpoint that will break when the value of @var{expr} is read
3698 @item awatch @var{expr} @r{[}thread @var{threadnum}@r{]}
3699 Set a watchpoint that will break when @var{expr} is either read from
3700 or written into by the program.
3702 @kindex info watchpoints @r{[}@var{n}@r{]}
3703 @item info watchpoints
3704 This command prints a list of watchpoints, breakpoints, and catchpoints;
3705 it is the same as @code{info break} (@pxref{Set Breaks}).
3708 @value{GDBN} sets a @dfn{hardware watchpoint} if possible. Hardware
3709 watchpoints execute very quickly, and the debugger reports a change in
3710 value at the exact instruction where the change occurs. If @value{GDBN}
3711 cannot set a hardware watchpoint, it sets a software watchpoint, which
3712 executes more slowly and reports the change in value at the next
3713 @emph{statement}, not the instruction, after the change occurs.
3715 @cindex use only software watchpoints
3716 You can force @value{GDBN} to use only software watchpoints with the
3717 @kbd{set can-use-hw-watchpoints 0} command. With this variable set to
3718 zero, @value{GDBN} will never try to use hardware watchpoints, even if
3719 the underlying system supports them. (Note that hardware-assisted
3720 watchpoints that were set @emph{before} setting
3721 @code{can-use-hw-watchpoints} to zero will still use the hardware
3722 mechanism of watching expression values.)
3725 @item set can-use-hw-watchpoints
3726 @kindex set can-use-hw-watchpoints
3727 Set whether or not to use hardware watchpoints.
3729 @item show can-use-hw-watchpoints
3730 @kindex show can-use-hw-watchpoints
3731 Show the current mode of using hardware watchpoints.
3734 For remote targets, you can restrict the number of hardware
3735 watchpoints @value{GDBN} will use, see @ref{set remote
3736 hardware-breakpoint-limit}.
3738 When you issue the @code{watch} command, @value{GDBN} reports
3741 Hardware watchpoint @var{num}: @var{expr}
3745 if it was able to set a hardware watchpoint.
3747 Currently, the @code{awatch} and @code{rwatch} commands can only set
3748 hardware watchpoints, because accesses to data that don't change the
3749 value of the watched expression cannot be detected without examining
3750 every instruction as it is being executed, and @value{GDBN} does not do
3751 that currently. If @value{GDBN} finds that it is unable to set a
3752 hardware breakpoint with the @code{awatch} or @code{rwatch} command, it
3753 will print a message like this:
3756 Expression cannot be implemented with read/access watchpoint.
3759 Sometimes, @value{GDBN} cannot set a hardware watchpoint because the
3760 data type of the watched expression is wider than what a hardware
3761 watchpoint on the target machine can handle. For example, some systems
3762 can only watch regions that are up to 4 bytes wide; on such systems you
3763 cannot set hardware watchpoints for an expression that yields a
3764 double-precision floating-point number (which is typically 8 bytes
3765 wide). As a work-around, it might be possible to break the large region
3766 into a series of smaller ones and watch them with separate watchpoints.
3768 If you set too many hardware watchpoints, @value{GDBN} might be unable
3769 to insert all of them when you resume the execution of your program.
3770 Since the precise number of active watchpoints is unknown until such
3771 time as the program is about to be resumed, @value{GDBN} might not be
3772 able to warn you about this when you set the watchpoints, and the
3773 warning will be printed only when the program is resumed:
3776 Hardware watchpoint @var{num}: Could not insert watchpoint
3780 If this happens, delete or disable some of the watchpoints.
3782 Watching complex expressions that reference many variables can also
3783 exhaust the resources available for hardware-assisted watchpoints.
3784 That's because @value{GDBN} needs to watch every variable in the
3785 expression with separately allocated resources.
3787 If you call a function interactively using @code{print} or @code{call},
3788 any watchpoints you have set will be inactive until @value{GDBN} reaches another
3789 kind of breakpoint or the call completes.
3791 @value{GDBN} automatically deletes watchpoints that watch local
3792 (automatic) variables, or expressions that involve such variables, when
3793 they go out of scope, that is, when the execution leaves the block in
3794 which these variables were defined. In particular, when the program
3795 being debugged terminates, @emph{all} local variables go out of scope,
3796 and so only watchpoints that watch global variables remain set. If you
3797 rerun the program, you will need to set all such watchpoints again. One
3798 way of doing that would be to set a code breakpoint at the entry to the
3799 @code{main} function and when it breaks, set all the watchpoints.
3801 @cindex watchpoints and threads
3802 @cindex threads and watchpoints
3803 In multi-threaded programs, watchpoints will detect changes to the
3804 watched expression from every thread.
3807 @emph{Warning:} In multi-threaded programs, software watchpoints
3808 have only limited usefulness. If @value{GDBN} creates a software
3809 watchpoint, it can only watch the value of an expression @emph{in a
3810 single thread}. If you are confident that the expression can only
3811 change due to the current thread's activity (and if you are also
3812 confident that no other thread can become current), then you can use
3813 software watchpoints as usual. However, @value{GDBN} may not notice
3814 when a non-current thread's activity changes the expression. (Hardware
3815 watchpoints, in contrast, watch an expression in all threads.)
3818 @xref{set remote hardware-watchpoint-limit}.
3820 @node Set Catchpoints
3821 @subsection Setting Catchpoints
3822 @cindex catchpoints, setting
3823 @cindex exception handlers
3824 @cindex event handling
3826 You can use @dfn{catchpoints} to cause the debugger to stop for certain
3827 kinds of program events, such as C@t{++} exceptions or the loading of a
3828 shared library. Use the @code{catch} command to set a catchpoint.
3832 @item catch @var{event}
3833 Stop when @var{event} occurs. @var{event} can be any of the following:
3836 @cindex stop on C@t{++} exceptions
3837 The throwing of a C@t{++} exception.
3840 The catching of a C@t{++} exception.
3843 @cindex Ada exception catching
3844 @cindex catch Ada exceptions
3845 An Ada exception being raised. If an exception name is specified
3846 at the end of the command (eg @code{catch exception Program_Error}),
3847 the debugger will stop only when this specific exception is raised.
3848 Otherwise, the debugger stops execution when any Ada exception is raised.
3850 When inserting an exception catchpoint on a user-defined exception whose
3851 name is identical to one of the exceptions defined by the language, the
3852 fully qualified name must be used as the exception name. Otherwise,
3853 @value{GDBN} will assume that it should stop on the pre-defined exception
3854 rather than the user-defined one. For instance, assuming an exception
3855 called @code{Constraint_Error} is defined in package @code{Pck}, then
3856 the command to use to catch such exceptions is @kbd{catch exception
3857 Pck.Constraint_Error}.
3859 @item exception unhandled
3860 An exception that was raised but is not handled by the program.
3863 A failed Ada assertion.
3866 @cindex break on fork/exec
3867 A call to @code{exec}. This is currently only available for HP-UX
3871 @itemx syscall @r{[}@var{name} @r{|} @var{number}@r{]} @r{...}
3872 @cindex break on a system call.
3873 A call to or return from a system call, a.k.a.@: @dfn{syscall}. A
3874 syscall is a mechanism for application programs to request a service
3875 from the operating system (OS) or one of the OS system services.
3876 @value{GDBN} can catch some or all of the syscalls issued by the
3877 debuggee, and show the related information for each syscall. If no
3878 argument is specified, calls to and returns from all system calls
3881 @var{name} can be any system call name that is valid for the
3882 underlying OS. Just what syscalls are valid depends on the OS. On
3883 GNU and Unix systems, you can find the full list of valid syscall
3884 names on @file{/usr/include/asm/unistd.h}.
3886 @c For MS-Windows, the syscall names and the corresponding numbers
3887 @c can be found, e.g., on this URL:
3888 @c http://www.metasploit.com/users/opcode/syscalls.html
3889 @c but we don't support Windows syscalls yet.
3891 Normally, @value{GDBN} knows in advance which syscalls are valid for
3892 each OS, so you can use the @value{GDBN} command-line completion
3893 facilities (@pxref{Completion,, command completion}) to list the
3896 You may also specify the system call numerically. A syscall's
3897 number is the value passed to the OS's syscall dispatcher to
3898 identify the requested service. When you specify the syscall by its
3899 name, @value{GDBN} uses its database of syscalls to convert the name
3900 into the corresponding numeric code, but using the number directly
3901 may be useful if @value{GDBN}'s database does not have the complete
3902 list of syscalls on your system (e.g., because @value{GDBN} lags
3903 behind the OS upgrades).
3905 The example below illustrates how this command works if you don't provide
3909 (@value{GDBP}) catch syscall
3910 Catchpoint 1 (syscall)
3912 Starting program: /tmp/catch-syscall
3914 Catchpoint 1 (call to syscall 'close'), \
3915 0xffffe424 in __kernel_vsyscall ()
3919 Catchpoint 1 (returned from syscall 'close'), \
3920 0xffffe424 in __kernel_vsyscall ()
3924 Here is an example of catching a system call by name:
3927 (@value{GDBP}) catch syscall chroot
3928 Catchpoint 1 (syscall 'chroot' [61])
3930 Starting program: /tmp/catch-syscall
3932 Catchpoint 1 (call to syscall 'chroot'), \
3933 0xffffe424 in __kernel_vsyscall ()
3937 Catchpoint 1 (returned from syscall 'chroot'), \
3938 0xffffe424 in __kernel_vsyscall ()
3942 An example of specifying a system call numerically. In the case
3943 below, the syscall number has a corresponding entry in the XML
3944 file, so @value{GDBN} finds its name and prints it:
3947 (@value{GDBP}) catch syscall 252
3948 Catchpoint 1 (syscall(s) 'exit_group')
3950 Starting program: /tmp/catch-syscall
3952 Catchpoint 1 (call to syscall 'exit_group'), \
3953 0xffffe424 in __kernel_vsyscall ()
3957 Program exited normally.
3961 However, there can be situations when there is no corresponding name
3962 in XML file for that syscall number. In this case, @value{GDBN} prints
3963 a warning message saying that it was not able to find the syscall name,
3964 but the catchpoint will be set anyway. See the example below:
3967 (@value{GDBP}) catch syscall 764
3968 warning: The number '764' does not represent a known syscall.
3969 Catchpoint 2 (syscall 764)
3973 If you configure @value{GDBN} using the @samp{--without-expat} option,
3974 it will not be able to display syscall names. Also, if your
3975 architecture does not have an XML file describing its system calls,
3976 you will not be able to see the syscall names. It is important to
3977 notice that these two features are used for accessing the syscall
3978 name database. In either case, you will see a warning like this:
3981 (@value{GDBP}) catch syscall
3982 warning: Could not open "syscalls/i386-linux.xml"
3983 warning: Could not load the syscall XML file 'syscalls/i386-linux.xml'.
3984 GDB will not be able to display syscall names.
3985 Catchpoint 1 (syscall)
3989 Of course, the file name will change depending on your architecture and system.
3991 Still using the example above, you can also try to catch a syscall by its
3992 number. In this case, you would see something like:
3995 (@value{GDBP}) catch syscall 252
3996 Catchpoint 1 (syscall(s) 252)
3999 Again, in this case @value{GDBN} would not be able to display syscall's names.
4002 A call to @code{fork}. This is currently only available for HP-UX
4006 A call to @code{vfork}. This is currently only available for HP-UX
4011 @item tcatch @var{event}
4012 Set a catchpoint that is enabled only for one stop. The catchpoint is
4013 automatically deleted after the first time the event is caught.
4017 Use the @code{info break} command to list the current catchpoints.
4019 There are currently some limitations to C@t{++} exception handling
4020 (@code{catch throw} and @code{catch catch}) in @value{GDBN}:
4024 If you call a function interactively, @value{GDBN} normally returns
4025 control to you when the function has finished executing. If the call
4026 raises an exception, however, the call may bypass the mechanism that
4027 returns control to you and cause your program either to abort or to
4028 simply continue running until it hits a breakpoint, catches a signal
4029 that @value{GDBN} is listening for, or exits. This is the case even if
4030 you set a catchpoint for the exception; catchpoints on exceptions are
4031 disabled within interactive calls.
4034 You cannot raise an exception interactively.
4037 You cannot install an exception handler interactively.
4040 @cindex raise exceptions
4041 Sometimes @code{catch} is not the best way to debug exception handling:
4042 if you need to know exactly where an exception is raised, it is better to
4043 stop @emph{before} the exception handler is called, since that way you
4044 can see the stack before any unwinding takes place. If you set a
4045 breakpoint in an exception handler instead, it may not be easy to find
4046 out where the exception was raised.
4048 To stop just before an exception handler is called, you need some
4049 knowledge of the implementation. In the case of @sc{gnu} C@t{++}, exceptions are
4050 raised by calling a library function named @code{__raise_exception}
4051 which has the following ANSI C interface:
4054 /* @var{addr} is where the exception identifier is stored.
4055 @var{id} is the exception identifier. */
4056 void __raise_exception (void **addr, void *id);
4060 To make the debugger catch all exceptions before any stack
4061 unwinding takes place, set a breakpoint on @code{__raise_exception}
4062 (@pxref{Breakpoints, ,Breakpoints; Watchpoints; and Exceptions}).
4064 With a conditional breakpoint (@pxref{Conditions, ,Break Conditions})
4065 that depends on the value of @var{id}, you can stop your program when
4066 a specific exception is raised. You can use multiple conditional
4067 breakpoints to stop your program when any of a number of exceptions are
4072 @subsection Deleting Breakpoints
4074 @cindex clearing breakpoints, watchpoints, catchpoints
4075 @cindex deleting breakpoints, watchpoints, catchpoints
4076 It is often necessary to eliminate a breakpoint, watchpoint, or
4077 catchpoint once it has done its job and you no longer want your program
4078 to stop there. This is called @dfn{deleting} the breakpoint. A
4079 breakpoint that has been deleted no longer exists; it is forgotten.
4081 With the @code{clear} command you can delete breakpoints according to
4082 where they are in your program. With the @code{delete} command you can
4083 delete individual breakpoints, watchpoints, or catchpoints by specifying
4084 their breakpoint numbers.
4086 It is not necessary to delete a breakpoint to proceed past it. @value{GDBN}
4087 automatically ignores breakpoints on the first instruction to be executed
4088 when you continue execution without changing the execution address.
4093 Delete any breakpoints at the next instruction to be executed in the
4094 selected stack frame (@pxref{Selection, ,Selecting a Frame}). When
4095 the innermost frame is selected, this is a good way to delete a
4096 breakpoint where your program just stopped.
4098 @item clear @var{location}
4099 Delete any breakpoints set at the specified @var{location}.
4100 @xref{Specify Location}, for the various forms of @var{location}; the
4101 most useful ones are listed below:
4104 @item clear @var{function}
4105 @itemx clear @var{filename}:@var{function}
4106 Delete any breakpoints set at entry to the named @var{function}.
4108 @item clear @var{linenum}
4109 @itemx clear @var{filename}:@var{linenum}
4110 Delete any breakpoints set at or within the code of the specified
4111 @var{linenum} of the specified @var{filename}.
4114 @cindex delete breakpoints
4116 @kindex d @r{(@code{delete})}
4117 @item delete @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
4118 Delete the breakpoints, watchpoints, or catchpoints of the breakpoint
4119 ranges specified as arguments. If no argument is specified, delete all
4120 breakpoints (@value{GDBN} asks confirmation, unless you have @code{set
4121 confirm off}). You can abbreviate this command as @code{d}.
4125 @subsection Disabling Breakpoints
4127 @cindex enable/disable a breakpoint
4128 Rather than deleting a breakpoint, watchpoint, or catchpoint, you might
4129 prefer to @dfn{disable} it. This makes the breakpoint inoperative as if
4130 it had been deleted, but remembers the information on the breakpoint so
4131 that you can @dfn{enable} it again later.
4133 You disable and enable breakpoints, watchpoints, and catchpoints with
4134 the @code{enable} and @code{disable} commands, optionally specifying one
4135 or more breakpoint numbers as arguments. Use @code{info break} or
4136 @code{info watch} to print a list of breakpoints, watchpoints, and
4137 catchpoints if you do not know which numbers to use.
4139 Disabling and enabling a breakpoint that has multiple locations
4140 affects all of its locations.
4142 A breakpoint, watchpoint, or catchpoint can have any of four different
4143 states of enablement:
4147 Enabled. The breakpoint stops your program. A breakpoint set
4148 with the @code{break} command starts out in this state.
4150 Disabled. The breakpoint has no effect on your program.
4152 Enabled once. The breakpoint stops your program, but then becomes
4155 Enabled for deletion. The breakpoint stops your program, but
4156 immediately after it does so it is deleted permanently. A breakpoint
4157 set with the @code{tbreak} command starts out in this state.
4160 You can use the following commands to enable or disable breakpoints,
4161 watchpoints, and catchpoints:
4165 @kindex dis @r{(@code{disable})}
4166 @item disable @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
4167 Disable the specified breakpoints---or all breakpoints, if none are
4168 listed. A disabled breakpoint has no effect but is not forgotten. All
4169 options such as ignore-counts, conditions and commands are remembered in
4170 case the breakpoint is enabled again later. You may abbreviate
4171 @code{disable} as @code{dis}.
4174 @item enable @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
4175 Enable the specified breakpoints (or all defined breakpoints). They
4176 become effective once again in stopping your program.
4178 @item enable @r{[}breakpoints@r{]} once @var{range}@dots{}
4179 Enable the specified breakpoints temporarily. @value{GDBN} disables any
4180 of these breakpoints immediately after stopping your program.
4182 @item enable @r{[}breakpoints@r{]} delete @var{range}@dots{}
4183 Enable the specified breakpoints to work once, then die. @value{GDBN}
4184 deletes any of these breakpoints as soon as your program stops there.
4185 Breakpoints set by the @code{tbreak} command start out in this state.
4188 @c FIXME: I think the following ``Except for [...] @code{tbreak}'' is
4189 @c confusing: tbreak is also initially enabled.
4190 Except for a breakpoint set with @code{tbreak} (@pxref{Set Breaks,
4191 ,Setting Breakpoints}), breakpoints that you set are initially enabled;
4192 subsequently, they become disabled or enabled only when you use one of
4193 the commands above. (The command @code{until} can set and delete a
4194 breakpoint of its own, but it does not change the state of your other
4195 breakpoints; see @ref{Continuing and Stepping, ,Continuing and
4199 @subsection Break Conditions
4200 @cindex conditional breakpoints
4201 @cindex breakpoint conditions
4203 @c FIXME what is scope of break condition expr? Context where wanted?
4204 @c in particular for a watchpoint?
4205 The simplest sort of breakpoint breaks every time your program reaches a
4206 specified place. You can also specify a @dfn{condition} for a
4207 breakpoint. A condition is just a Boolean expression in your
4208 programming language (@pxref{Expressions, ,Expressions}). A breakpoint with
4209 a condition evaluates the expression each time your program reaches it,
4210 and your program stops only if the condition is @emph{true}.
4212 This is the converse of using assertions for program validation; in that
4213 situation, you want to stop when the assertion is violated---that is,
4214 when the condition is false. In C, if you want to test an assertion expressed
4215 by the condition @var{assert}, you should set the condition
4216 @samp{! @var{assert}} on the appropriate breakpoint.
4218 Conditions are also accepted for watchpoints; you may not need them,
4219 since a watchpoint is inspecting the value of an expression anyhow---but
4220 it might be simpler, say, to just set a watchpoint on a variable name,
4221 and specify a condition that tests whether the new value is an interesting
4224 Break conditions can have side effects, and may even call functions in
4225 your program. This can be useful, for example, to activate functions
4226 that log program progress, or to use your own print functions to
4227 format special data structures. The effects are completely predictable
4228 unless there is another enabled breakpoint at the same address. (In
4229 that case, @value{GDBN} might see the other breakpoint first and stop your
4230 program without checking the condition of this one.) Note that
4231 breakpoint commands are usually more convenient and flexible than break
4233 purpose of performing side effects when a breakpoint is reached
4234 (@pxref{Break Commands, ,Breakpoint Command Lists}).
4236 Break conditions can be specified when a breakpoint is set, by using
4237 @samp{if} in the arguments to the @code{break} command. @xref{Set
4238 Breaks, ,Setting Breakpoints}. They can also be changed at any time
4239 with the @code{condition} command.
4241 You can also use the @code{if} keyword with the @code{watch} command.
4242 The @code{catch} command does not recognize the @code{if} keyword;
4243 @code{condition} is the only way to impose a further condition on a
4248 @item condition @var{bnum} @var{expression}
4249 Specify @var{expression} as the break condition for breakpoint,
4250 watchpoint, or catchpoint number @var{bnum}. After you set a condition,
4251 breakpoint @var{bnum} stops your program only if the value of
4252 @var{expression} is true (nonzero, in C). When you use
4253 @code{condition}, @value{GDBN} checks @var{expression} immediately for
4254 syntactic correctness, and to determine whether symbols in it have
4255 referents in the context of your breakpoint. If @var{expression} uses
4256 symbols not referenced in the context of the breakpoint, @value{GDBN}
4257 prints an error message:
4260 No symbol "foo" in current context.
4265 not actually evaluate @var{expression} at the time the @code{condition}
4266 command (or a command that sets a breakpoint with a condition, like
4267 @code{break if @dots{}}) is given, however. @xref{Expressions, ,Expressions}.
4269 @item condition @var{bnum}
4270 Remove the condition from breakpoint number @var{bnum}. It becomes
4271 an ordinary unconditional breakpoint.
4274 @cindex ignore count (of breakpoint)
4275 A special case of a breakpoint condition is to stop only when the
4276 breakpoint has been reached a certain number of times. This is so
4277 useful that there is a special way to do it, using the @dfn{ignore
4278 count} of the breakpoint. Every breakpoint has an ignore count, which
4279 is an integer. Most of the time, the ignore count is zero, and
4280 therefore has no effect. But if your program reaches a breakpoint whose
4281 ignore count is positive, then instead of stopping, it just decrements
4282 the ignore count by one and continues. As a result, if the ignore count
4283 value is @var{n}, the breakpoint does not stop the next @var{n} times
4284 your program reaches it.
4288 @item ignore @var{bnum} @var{count}
4289 Set the ignore count of breakpoint number @var{bnum} to @var{count}.
4290 The next @var{count} times the breakpoint is reached, your program's
4291 execution does not stop; other than to decrement the ignore count, @value{GDBN}
4294 To make the breakpoint stop the next time it is reached, specify
4297 When you use @code{continue} to resume execution of your program from a
4298 breakpoint, you can specify an ignore count directly as an argument to
4299 @code{continue}, rather than using @code{ignore}. @xref{Continuing and
4300 Stepping,,Continuing and Stepping}.
4302 If a breakpoint has a positive ignore count and a condition, the
4303 condition is not checked. Once the ignore count reaches zero,
4304 @value{GDBN} resumes checking the condition.
4306 You could achieve the effect of the ignore count with a condition such
4307 as @w{@samp{$foo-- <= 0}} using a debugger convenience variable that
4308 is decremented each time. @xref{Convenience Vars, ,Convenience
4312 Ignore counts apply to breakpoints, watchpoints, and catchpoints.
4315 @node Break Commands
4316 @subsection Breakpoint Command Lists
4318 @cindex breakpoint commands
4319 You can give any breakpoint (or watchpoint or catchpoint) a series of
4320 commands to execute when your program stops due to that breakpoint. For
4321 example, you might want to print the values of certain expressions, or
4322 enable other breakpoints.
4326 @kindex end@r{ (breakpoint commands)}
4327 @item commands @r{[}@var{bnum}@r{]}
4328 @itemx @dots{} @var{command-list} @dots{}
4330 Specify a list of commands for breakpoint number @var{bnum}. The commands
4331 themselves appear on the following lines. Type a line containing just
4332 @code{end} to terminate the commands.
4334 To remove all commands from a breakpoint, type @code{commands} and
4335 follow it immediately with @code{end}; that is, give no commands.
4337 With no @var{bnum} argument, @code{commands} refers to the last
4338 breakpoint, watchpoint, or catchpoint set (not to the breakpoint most
4339 recently encountered).
4342 Pressing @key{RET} as a means of repeating the last @value{GDBN} command is
4343 disabled within a @var{command-list}.
4345 You can use breakpoint commands to start your program up again. Simply
4346 use the @code{continue} command, or @code{step}, or any other command
4347 that resumes execution.
4349 Any other commands in the command list, after a command that resumes
4350 execution, are ignored. This is because any time you resume execution
4351 (even with a simple @code{next} or @code{step}), you may encounter
4352 another breakpoint---which could have its own command list, leading to
4353 ambiguities about which list to execute.
4356 If the first command you specify in a command list is @code{silent}, the
4357 usual message about stopping at a breakpoint is not printed. This may
4358 be desirable for breakpoints that are to print a specific message and
4359 then continue. If none of the remaining commands print anything, you
4360 see no sign that the breakpoint was reached. @code{silent} is
4361 meaningful only at the beginning of a breakpoint command list.
4363 The commands @code{echo}, @code{output}, and @code{printf} allow you to
4364 print precisely controlled output, and are often useful in silent
4365 breakpoints. @xref{Output, ,Commands for Controlled Output}.
4367 For example, here is how you could use breakpoint commands to print the
4368 value of @code{x} at entry to @code{foo} whenever @code{x} is positive.
4374 printf "x is %d\n",x
4379 One application for breakpoint commands is to compensate for one bug so
4380 you can test for another. Put a breakpoint just after the erroneous line
4381 of code, give it a condition to detect the case in which something
4382 erroneous has been done, and give it commands to assign correct values
4383 to any variables that need them. End with the @code{continue} command
4384 so that your program does not stop, and start with the @code{silent}
4385 command so that no output is produced. Here is an example:
4396 @c @ifclear BARETARGET
4397 @node Error in Breakpoints
4398 @subsection ``Cannot insert breakpoints''
4400 If you request too many active hardware-assisted breakpoints and
4401 watchpoints, you will see this error message:
4403 @c FIXME: the precise wording of this message may change; the relevant
4404 @c source change is not committed yet (Sep 3, 1999).
4406 Stopped; cannot insert breakpoints.
4407 You may have requested too many hardware breakpoints and watchpoints.
4411 This message is printed when you attempt to resume the program, since
4412 only then @value{GDBN} knows exactly how many hardware breakpoints and
4413 watchpoints it needs to insert.
4415 When this message is printed, you need to disable or remove some of the
4416 hardware-assisted breakpoints and watchpoints, and then continue.
4418 @node Breakpoint-related Warnings
4419 @subsection ``Breakpoint address adjusted...''
4420 @cindex breakpoint address adjusted
4422 Some processor architectures place constraints on the addresses at
4423 which breakpoints may be placed. For architectures thus constrained,
4424 @value{GDBN} will attempt to adjust the breakpoint's address to comply
4425 with the constraints dictated by the architecture.
4427 One example of such an architecture is the Fujitsu FR-V. The FR-V is
4428 a VLIW architecture in which a number of RISC-like instructions may be
4429 bundled together for parallel execution. The FR-V architecture
4430 constrains the location of a breakpoint instruction within such a
4431 bundle to the instruction with the lowest address. @value{GDBN}
4432 honors this constraint by adjusting a breakpoint's address to the
4433 first in the bundle.
4435 It is not uncommon for optimized code to have bundles which contain
4436 instructions from different source statements, thus it may happen that
4437 a breakpoint's address will be adjusted from one source statement to
4438 another. Since this adjustment may significantly alter @value{GDBN}'s
4439 breakpoint related behavior from what the user expects, a warning is
4440 printed when the breakpoint is first set and also when the breakpoint
4443 A warning like the one below is printed when setting a breakpoint
4444 that's been subject to address adjustment:
4447 warning: Breakpoint address adjusted from 0x00010414 to 0x00010410.
4450 Such warnings are printed both for user settable and @value{GDBN}'s
4451 internal breakpoints. If you see one of these warnings, you should
4452 verify that a breakpoint set at the adjusted address will have the
4453 desired affect. If not, the breakpoint in question may be removed and
4454 other breakpoints may be set which will have the desired behavior.
4455 E.g., it may be sufficient to place the breakpoint at a later
4456 instruction. A conditional breakpoint may also be useful in some
4457 cases to prevent the breakpoint from triggering too often.
4459 @value{GDBN} will also issue a warning when stopping at one of these
4460 adjusted breakpoints:
4463 warning: Breakpoint 1 address previously adjusted from 0x00010414
4467 When this warning is encountered, it may be too late to take remedial
4468 action except in cases where the breakpoint is hit earlier or more
4469 frequently than expected.
4471 @node Continuing and Stepping
4472 @section Continuing and Stepping
4476 @cindex resuming execution
4477 @dfn{Continuing} means resuming program execution until your program
4478 completes normally. In contrast, @dfn{stepping} means executing just
4479 one more ``step'' of your program, where ``step'' may mean either one
4480 line of source code, or one machine instruction (depending on what
4481 particular command you use). Either when continuing or when stepping,
4482 your program may stop even sooner, due to a breakpoint or a signal. (If
4483 it stops due to a signal, you may want to use @code{handle}, or use
4484 @samp{signal 0} to resume execution. @xref{Signals, ,Signals}.)
4488 @kindex c @r{(@code{continue})}
4489 @kindex fg @r{(resume foreground execution)}
4490 @item continue @r{[}@var{ignore-count}@r{]}
4491 @itemx c @r{[}@var{ignore-count}@r{]}
4492 @itemx fg @r{[}@var{ignore-count}@r{]}
4493 Resume program execution, at the address where your program last stopped;
4494 any breakpoints set at that address are bypassed. The optional argument
4495 @var{ignore-count} allows you to specify a further number of times to
4496 ignore a breakpoint at this location; its effect is like that of
4497 @code{ignore} (@pxref{Conditions, ,Break Conditions}).
4499 The argument @var{ignore-count} is meaningful only when your program
4500 stopped due to a breakpoint. At other times, the argument to
4501 @code{continue} is ignored.
4503 The synonyms @code{c} and @code{fg} (for @dfn{foreground}, as the
4504 debugged program is deemed to be the foreground program) are provided
4505 purely for convenience, and have exactly the same behavior as
4509 To resume execution at a different place, you can use @code{return}
4510 (@pxref{Returning, ,Returning from a Function}) to go back to the
4511 calling function; or @code{jump} (@pxref{Jumping, ,Continuing at a
4512 Different Address}) to go to an arbitrary location in your program.
4514 A typical technique for using stepping is to set a breakpoint
4515 (@pxref{Breakpoints, ,Breakpoints; Watchpoints; and Catchpoints}) at the
4516 beginning of the function or the section of your program where a problem
4517 is believed to lie, run your program until it stops at that breakpoint,
4518 and then step through the suspect area, examining the variables that are
4519 interesting, until you see the problem happen.
4523 @kindex s @r{(@code{step})}
4525 Continue running your program until control reaches a different source
4526 line, then stop it and return control to @value{GDBN}. This command is
4527 abbreviated @code{s}.
4530 @c "without debugging information" is imprecise; actually "without line
4531 @c numbers in the debugging information". (gcc -g1 has debugging info but
4532 @c not line numbers). But it seems complex to try to make that
4533 @c distinction here.
4534 @emph{Warning:} If you use the @code{step} command while control is
4535 within a function that was compiled without debugging information,
4536 execution proceeds until control reaches a function that does have
4537 debugging information. Likewise, it will not step into a function which
4538 is compiled without debugging information. To step through functions
4539 without debugging information, use the @code{stepi} command, described
4543 The @code{step} command only stops at the first instruction of a source
4544 line. This prevents the multiple stops that could otherwise occur in
4545 @code{switch} statements, @code{for} loops, etc. @code{step} continues
4546 to stop if a function that has debugging information is called within
4547 the line. In other words, @code{step} @emph{steps inside} any functions
4548 called within the line.
4550 Also, the @code{step} command only enters a function if there is line
4551 number information for the function. Otherwise it acts like the
4552 @code{next} command. This avoids problems when using @code{cc -gl}
4553 on MIPS machines. Previously, @code{step} entered subroutines if there
4554 was any debugging information about the routine.
4556 @item step @var{count}
4557 Continue running as in @code{step}, but do so @var{count} times. If a
4558 breakpoint is reached, or a signal not related to stepping occurs before
4559 @var{count} steps, stepping stops right away.
4562 @kindex n @r{(@code{next})}
4563 @item next @r{[}@var{count}@r{]}
4564 Continue to the next source line in the current (innermost) stack frame.
4565 This is similar to @code{step}, but function calls that appear within
4566 the line of code are executed without stopping. Execution stops when
4567 control reaches a different line of code at the original stack level
4568 that was executing when you gave the @code{next} command. This command
4569 is abbreviated @code{n}.
4571 An argument @var{count} is a repeat count, as for @code{step}.
4574 @c FIX ME!! Do we delete this, or is there a way it fits in with
4575 @c the following paragraph? --- Vctoria
4577 @c @code{next} within a function that lacks debugging information acts like
4578 @c @code{step}, but any function calls appearing within the code of the
4579 @c function are executed without stopping.
4581 The @code{next} command only stops at the first instruction of a
4582 source line. This prevents multiple stops that could otherwise occur in
4583 @code{switch} statements, @code{for} loops, etc.
4585 @kindex set step-mode
4587 @cindex functions without line info, and stepping
4588 @cindex stepping into functions with no line info
4589 @itemx set step-mode on
4590 The @code{set step-mode on} command causes the @code{step} command to
4591 stop at the first instruction of a function which contains no debug line
4592 information rather than stepping over it.
4594 This is useful in cases where you may be interested in inspecting the
4595 machine instructions of a function which has no symbolic info and do not
4596 want @value{GDBN} to automatically skip over this function.
4598 @item set step-mode off
4599 Causes the @code{step} command to step over any functions which contains no
4600 debug information. This is the default.
4602 @item show step-mode
4603 Show whether @value{GDBN} will stop in or step over functions without
4604 source line debug information.
4607 @kindex fin @r{(@code{finish})}
4609 Continue running until just after function in the selected stack frame
4610 returns. Print the returned value (if any). This command can be
4611 abbreviated as @code{fin}.
4613 Contrast this with the @code{return} command (@pxref{Returning,
4614 ,Returning from a Function}).
4617 @kindex u @r{(@code{until})}
4618 @cindex run until specified location
4621 Continue running until a source line past the current line, in the
4622 current stack frame, is reached. This command is used to avoid single
4623 stepping through a loop more than once. It is like the @code{next}
4624 command, except that when @code{until} encounters a jump, it
4625 automatically continues execution until the program counter is greater
4626 than the address of the jump.
4628 This means that when you reach the end of a loop after single stepping
4629 though it, @code{until} makes your program continue execution until it
4630 exits the loop. In contrast, a @code{next} command at the end of a loop
4631 simply steps back to the beginning of the loop, which forces you to step
4632 through the next iteration.
4634 @code{until} always stops your program if it attempts to exit the current
4637 @code{until} may produce somewhat counterintuitive results if the order
4638 of machine code does not match the order of the source lines. For
4639 example, in the following excerpt from a debugging session, the @code{f}
4640 (@code{frame}) command shows that execution is stopped at line
4641 @code{206}; yet when we use @code{until}, we get to line @code{195}:
4645 #0 main (argc=4, argv=0xf7fffae8) at m4.c:206
4647 (@value{GDBP}) until
4648 195 for ( ; argc > 0; NEXTARG) @{
4651 This happened because, for execution efficiency, the compiler had
4652 generated code for the loop closure test at the end, rather than the
4653 start, of the loop---even though the test in a C @code{for}-loop is
4654 written before the body of the loop. The @code{until} command appeared
4655 to step back to the beginning of the loop when it advanced to this
4656 expression; however, it has not really gone to an earlier
4657 statement---not in terms of the actual machine code.
4659 @code{until} with no argument works by means of single
4660 instruction stepping, and hence is slower than @code{until} with an
4663 @item until @var{location}
4664 @itemx u @var{location}
4665 Continue running your program until either the specified location is
4666 reached, or the current stack frame returns. @var{location} is any of
4667 the forms described in @ref{Specify Location}.
4668 This form of the command uses temporary breakpoints, and
4669 hence is quicker than @code{until} without an argument. The specified
4670 location is actually reached only if it is in the current frame. This
4671 implies that @code{until} can be used to skip over recursive function
4672 invocations. For instance in the code below, if the current location is
4673 line @code{96}, issuing @code{until 99} will execute the program up to
4674 line @code{99} in the same invocation of factorial, i.e., after the inner
4675 invocations have returned.
4678 94 int factorial (int value)
4680 96 if (value > 1) @{
4681 97 value *= factorial (value - 1);
4688 @kindex advance @var{location}
4689 @itemx advance @var{location}
4690 Continue running the program up to the given @var{location}. An argument is
4691 required, which should be of one of the forms described in
4692 @ref{Specify Location}.
4693 Execution will also stop upon exit from the current stack
4694 frame. This command is similar to @code{until}, but @code{advance} will
4695 not skip over recursive function calls, and the target location doesn't
4696 have to be in the same frame as the current one.
4700 @kindex si @r{(@code{stepi})}
4702 @itemx stepi @var{arg}
4704 Execute one machine instruction, then stop and return to the debugger.
4706 It is often useful to do @samp{display/i $pc} when stepping by machine
4707 instructions. This makes @value{GDBN} automatically display the next
4708 instruction to be executed, each time your program stops. @xref{Auto
4709 Display,, Automatic Display}.
4711 An argument is a repeat count, as in @code{step}.
4715 @kindex ni @r{(@code{nexti})}
4717 @itemx nexti @var{arg}
4719 Execute one machine instruction, but if it is a function call,
4720 proceed until the function returns.
4722 An argument is a repeat count, as in @code{next}.
4729 A signal is an asynchronous event that can happen in a program. The
4730 operating system defines the possible kinds of signals, and gives each
4731 kind a name and a number. For example, in Unix @code{SIGINT} is the
4732 signal a program gets when you type an interrupt character (often @kbd{Ctrl-c});
4733 @code{SIGSEGV} is the signal a program gets from referencing a place in
4734 memory far away from all the areas in use; @code{SIGALRM} occurs when
4735 the alarm clock timer goes off (which happens only if your program has
4736 requested an alarm).
4738 @cindex fatal signals
4739 Some signals, including @code{SIGALRM}, are a normal part of the
4740 functioning of your program. Others, such as @code{SIGSEGV}, indicate
4741 errors; these signals are @dfn{fatal} (they kill your program immediately) if the
4742 program has not specified in advance some other way to handle the signal.
4743 @code{SIGINT} does not indicate an error in your program, but it is normally
4744 fatal so it can carry out the purpose of the interrupt: to kill the program.
4746 @value{GDBN} has the ability to detect any occurrence of a signal in your
4747 program. You can tell @value{GDBN} in advance what to do for each kind of
4750 @cindex handling signals
4751 Normally, @value{GDBN} is set up to let the non-erroneous signals like
4752 @code{SIGALRM} be silently passed to your program
4753 (so as not to interfere with their role in the program's functioning)
4754 but to stop your program immediately whenever an error signal happens.
4755 You can change these settings with the @code{handle} command.
4758 @kindex info signals
4762 Print a table of all the kinds of signals and how @value{GDBN} has been told to
4763 handle each one. You can use this to see the signal numbers of all
4764 the defined types of signals.
4766 @item info signals @var{sig}
4767 Similar, but print information only about the specified signal number.
4769 @code{info handle} is an alias for @code{info signals}.
4772 @item handle @var{signal} @r{[}@var{keywords}@dots{}@r{]}
4773 Change the way @value{GDBN} handles signal @var{signal}. @var{signal}
4774 can be the number of a signal or its name (with or without the
4775 @samp{SIG} at the beginning); a list of signal numbers of the form
4776 @samp{@var{low}-@var{high}}; or the word @samp{all}, meaning all the
4777 known signals. Optional arguments @var{keywords}, described below,
4778 say what change to make.
4782 The keywords allowed by the @code{handle} command can be abbreviated.
4783 Their full names are:
4787 @value{GDBN} should not stop your program when this signal happens. It may
4788 still print a message telling you that the signal has come in.
4791 @value{GDBN} should stop your program when this signal happens. This implies
4792 the @code{print} keyword as well.
4795 @value{GDBN} should print a message when this signal happens.
4798 @value{GDBN} should not mention the occurrence of the signal at all. This
4799 implies the @code{nostop} keyword as well.
4803 @value{GDBN} should allow your program to see this signal; your program
4804 can handle the signal, or else it may terminate if the signal is fatal
4805 and not handled. @code{pass} and @code{noignore} are synonyms.
4809 @value{GDBN} should not allow your program to see this signal.
4810 @code{nopass} and @code{ignore} are synonyms.
4814 When a signal stops your program, the signal is not visible to the
4816 continue. Your program sees the signal then, if @code{pass} is in
4817 effect for the signal in question @emph{at that time}. In other words,
4818 after @value{GDBN} reports a signal, you can use the @code{handle}
4819 command with @code{pass} or @code{nopass} to control whether your
4820 program sees that signal when you continue.
4822 The default is set to @code{nostop}, @code{noprint}, @code{pass} for
4823 non-erroneous signals such as @code{SIGALRM}, @code{SIGWINCH} and
4824 @code{SIGCHLD}, and to @code{stop}, @code{print}, @code{pass} for the
4827 You can also use the @code{signal} command to prevent your program from
4828 seeing a signal, or cause it to see a signal it normally would not see,
4829 or to give it any signal at any time. For example, if your program stopped
4830 due to some sort of memory reference error, you might store correct
4831 values into the erroneous variables and continue, hoping to see more
4832 execution; but your program would probably terminate immediately as
4833 a result of the fatal signal once it saw the signal. To prevent this,
4834 you can continue with @samp{signal 0}. @xref{Signaling, ,Giving your
4837 @cindex extra signal information
4838 @anchor{extra signal information}
4840 On some targets, @value{GDBN} can inspect extra signal information
4841 associated with the intercepted signal, before it is actually
4842 delivered to the program being debugged. This information is exported
4843 by the convenience variable @code{$_siginfo}, and consists of data
4844 that is passed by the kernel to the signal handler at the time of the
4845 receipt of a signal. The data type of the information itself is
4846 target dependent. You can see the data type using the @code{ptype
4847 $_siginfo} command. On Unix systems, it typically corresponds to the
4848 standard @code{siginfo_t} type, as defined in the @file{signal.h}
4851 Here's an example, on a @sc{gnu}/Linux system, printing the stray
4852 referenced address that raised a segmentation fault.
4856 (@value{GDBP}) continue
4857 Program received signal SIGSEGV, Segmentation fault.
4858 0x0000000000400766 in main ()
4860 (@value{GDBP}) ptype $_siginfo
4867 struct @{...@} _kill;
4868 struct @{...@} _timer;
4870 struct @{...@} _sigchld;
4871 struct @{...@} _sigfault;
4872 struct @{...@} _sigpoll;
4875 (@value{GDBP}) ptype $_siginfo._sifields._sigfault
4879 (@value{GDBP}) p $_siginfo._sifields._sigfault.si_addr
4880 $1 = (void *) 0x7ffff7ff7000
4884 Depending on target support, @code{$_siginfo} may also be writable.
4887 @section Stopping and Starting Multi-thread Programs
4889 @cindex stopped threads
4890 @cindex threads, stopped
4892 @cindex continuing threads
4893 @cindex threads, continuing
4895 @value{GDBN} supports debugging programs with multiple threads
4896 (@pxref{Threads,, Debugging Programs with Multiple Threads}). There
4897 are two modes of controlling execution of your program within the
4898 debugger. In the default mode, referred to as @dfn{all-stop mode},
4899 when any thread in your program stops (for example, at a breakpoint
4900 or while being stepped), all other threads in the program are also stopped by
4901 @value{GDBN}. On some targets, @value{GDBN} also supports
4902 @dfn{non-stop mode}, in which other threads can continue to run freely while
4903 you examine the stopped thread in the debugger.
4906 * All-Stop Mode:: All threads stop when GDB takes control
4907 * Non-Stop Mode:: Other threads continue to execute
4908 * Background Execution:: Running your program asynchronously
4909 * Thread-Specific Breakpoints:: Controlling breakpoints
4910 * Interrupted System Calls:: GDB may interfere with system calls
4914 @subsection All-Stop Mode
4916 @cindex all-stop mode
4918 In all-stop mode, whenever your program stops under @value{GDBN} for any reason,
4919 @emph{all} threads of execution stop, not just the current thread. This
4920 allows you to examine the overall state of the program, including
4921 switching between threads, without worrying that things may change
4924 Conversely, whenever you restart the program, @emph{all} threads start
4925 executing. @emph{This is true even when single-stepping} with commands
4926 like @code{step} or @code{next}.
4928 In particular, @value{GDBN} cannot single-step all threads in lockstep.
4929 Since thread scheduling is up to your debugging target's operating
4930 system (not controlled by @value{GDBN}), other threads may
4931 execute more than one statement while the current thread completes a
4932 single step. Moreover, in general other threads stop in the middle of a
4933 statement, rather than at a clean statement boundary, when the program
4936 You might even find your program stopped in another thread after
4937 continuing or even single-stepping. This happens whenever some other
4938 thread runs into a breakpoint, a signal, or an exception before the
4939 first thread completes whatever you requested.
4941 @cindex automatic thread selection
4942 @cindex switching threads automatically
4943 @cindex threads, automatic switching
4944 Whenever @value{GDBN} stops your program, due to a breakpoint or a
4945 signal, it automatically selects the thread where that breakpoint or
4946 signal happened. @value{GDBN} alerts you to the context switch with a
4947 message such as @samp{[Switching to Thread @var{n}]} to identify the
4950 On some OSes, you can modify @value{GDBN}'s default behavior by
4951 locking the OS scheduler to allow only a single thread to run.
4954 @item set scheduler-locking @var{mode}
4955 @cindex scheduler locking mode
4956 @cindex lock scheduler
4957 Set the scheduler locking mode. If it is @code{off}, then there is no
4958 locking and any thread may run at any time. If @code{on}, then only the
4959 current thread may run when the inferior is resumed. The @code{step}
4960 mode optimizes for single-stepping; it prevents other threads
4961 from preempting the current thread while you are stepping, so that
4962 the focus of debugging does not change unexpectedly.
4963 Other threads only rarely (or never) get a chance to run
4964 when you step. They are more likely to run when you @samp{next} over a
4965 function call, and they are completely free to run when you use commands
4966 like @samp{continue}, @samp{until}, or @samp{finish}. However, unless another
4967 thread hits a breakpoint during its timeslice, @value{GDBN} does not change
4968 the current thread away from the thread that you are debugging.
4970 @item show scheduler-locking
4971 Display the current scheduler locking mode.
4974 @cindex resume threads of multiple processes simultaneously
4975 By default, when you issue one of the execution commands such as
4976 @code{continue}, @code{next} or @code{step}, @value{GDBN} allows only
4977 threads of the current inferior to run. For example, if @value{GDBN}
4978 is attached to two inferiors, each with two threads, the
4979 @code{continue} command resumes only the two threads of the current
4980 inferior. This is useful, for example, when you debug a program that
4981 forks and you want to hold the parent stopped (so that, for instance,
4982 it doesn't run to exit), while you debug the child. In other
4983 situations, you may not be interested in inspecting the current state
4984 of any of the processes @value{GDBN} is attached to, and you may want
4985 to resume them all until some breakpoint is hit. In the latter case,
4986 you can instruct @value{GDBN} to allow all threads of all the
4987 inferiors to run with the @w{@code{set schedule-multiple}} command.
4990 @kindex set schedule-multiple
4991 @item set schedule-multiple
4992 Set the mode for allowing threads of multiple processes to be resumed
4993 when an execution command is issued. When @code{on}, all threads of
4994 all processes are allowed to run. When @code{off}, only the threads
4995 of the current process are resumed. The default is @code{off}. The
4996 @code{scheduler-locking} mode takes precedence when set to @code{on},
4997 or while you are stepping and set to @code{step}.
4999 @item show schedule-multiple
5000 Display the current mode for resuming the execution of threads of
5005 @subsection Non-Stop Mode
5007 @cindex non-stop mode
5009 @c This section is really only a place-holder, and needs to be expanded
5010 @c with more details.
5012 For some multi-threaded targets, @value{GDBN} supports an optional
5013 mode of operation in which you can examine stopped program threads in
5014 the debugger while other threads continue to execute freely. This
5015 minimizes intrusion when debugging live systems, such as programs
5016 where some threads have real-time constraints or must continue to
5017 respond to external events. This is referred to as @dfn{non-stop} mode.
5019 In non-stop mode, when a thread stops to report a debugging event,
5020 @emph{only} that thread is stopped; @value{GDBN} does not stop other
5021 threads as well, in contrast to the all-stop mode behavior. Additionally,
5022 execution commands such as @code{continue} and @code{step} apply by default
5023 only to the current thread in non-stop mode, rather than all threads as
5024 in all-stop mode. This allows you to control threads explicitly in
5025 ways that are not possible in all-stop mode --- for example, stepping
5026 one thread while allowing others to run freely, stepping
5027 one thread while holding all others stopped, or stepping several threads
5028 independently and simultaneously.
5030 To enter non-stop mode, use this sequence of commands before you run
5031 or attach to your program:
5034 # Enable the async interface.
5037 # If using the CLI, pagination breaks non-stop.
5040 # Finally, turn it on!
5044 You can use these commands to manipulate the non-stop mode setting:
5047 @kindex set non-stop
5048 @item set non-stop on
5049 Enable selection of non-stop mode.
5050 @item set non-stop off
5051 Disable selection of non-stop mode.
5052 @kindex show non-stop
5054 Show the current non-stop enablement setting.
5057 Note these commands only reflect whether non-stop mode is enabled,
5058 not whether the currently-executing program is being run in non-stop mode.
5059 In particular, the @code{set non-stop} preference is only consulted when
5060 @value{GDBN} starts or connects to the target program, and it is generally
5061 not possible to switch modes once debugging has started. Furthermore,
5062 since not all targets support non-stop mode, even when you have enabled
5063 non-stop mode, @value{GDBN} may still fall back to all-stop operation by
5066 In non-stop mode, all execution commands apply only to the current thread
5067 by default. That is, @code{continue} only continues one thread.
5068 To continue all threads, issue @code{continue -a} or @code{c -a}.
5070 You can use @value{GDBN}'s background execution commands
5071 (@pxref{Background Execution}) to run some threads in the background
5072 while you continue to examine or step others from @value{GDBN}.
5073 The MI execution commands (@pxref{GDB/MI Program Execution}) are
5074 always executed asynchronously in non-stop mode.
5076 Suspending execution is done with the @code{interrupt} command when
5077 running in the background, or @kbd{Ctrl-c} during foreground execution.
5078 In all-stop mode, this stops the whole process;
5079 but in non-stop mode the interrupt applies only to the current thread.
5080 To stop the whole program, use @code{interrupt -a}.
5082 Other execution commands do not currently support the @code{-a} option.
5084 In non-stop mode, when a thread stops, @value{GDBN} doesn't automatically make
5085 that thread current, as it does in all-stop mode. This is because the
5086 thread stop notifications are asynchronous with respect to @value{GDBN}'s
5087 command interpreter, and it would be confusing if @value{GDBN} unexpectedly
5088 changed to a different thread just as you entered a command to operate on the
5089 previously current thread.
5091 @node Background Execution
5092 @subsection Background Execution
5094 @cindex foreground execution
5095 @cindex background execution
5096 @cindex asynchronous execution
5097 @cindex execution, foreground, background and asynchronous
5099 @value{GDBN}'s execution commands have two variants: the normal
5100 foreground (synchronous) behavior, and a background
5101 (asynchronous) behavior. In foreground execution, @value{GDBN} waits for
5102 the program to report that some thread has stopped before prompting for
5103 another command. In background execution, @value{GDBN} immediately gives
5104 a command prompt so that you can issue other commands while your program runs.
5106 You need to explicitly enable asynchronous mode before you can use
5107 background execution commands. You can use these commands to
5108 manipulate the asynchronous mode setting:
5111 @kindex set target-async
5112 @item set target-async on
5113 Enable asynchronous mode.
5114 @item set target-async off
5115 Disable asynchronous mode.
5116 @kindex show target-async
5117 @item show target-async
5118 Show the current target-async setting.
5121 If the target doesn't support async mode, @value{GDBN} issues an error
5122 message if you attempt to use the background execution commands.
5124 To specify background execution, add a @code{&} to the command. For example,
5125 the background form of the @code{continue} command is @code{continue&}, or
5126 just @code{c&}. The execution commands that accept background execution
5132 @xref{Starting, , Starting your Program}.
5136 @xref{Attach, , Debugging an Already-running Process}.
5140 @xref{Continuing and Stepping, step}.
5144 @xref{Continuing and Stepping, stepi}.
5148 @xref{Continuing and Stepping, next}.
5152 @xref{Continuing and Stepping, nexti}.
5156 @xref{Continuing and Stepping, continue}.
5160 @xref{Continuing and Stepping, finish}.
5164 @xref{Continuing and Stepping, until}.
5168 Background execution is especially useful in conjunction with non-stop
5169 mode for debugging programs with multiple threads; see @ref{Non-Stop Mode}.
5170 However, you can also use these commands in the normal all-stop mode with
5171 the restriction that you cannot issue another execution command until the
5172 previous one finishes. Examples of commands that are valid in all-stop
5173 mode while the program is running include @code{help} and @code{info break}.
5175 You can interrupt your program while it is running in the background by
5176 using the @code{interrupt} command.
5183 Suspend execution of the running program. In all-stop mode,
5184 @code{interrupt} stops the whole process, but in non-stop mode, it stops
5185 only the current thread. To stop the whole program in non-stop mode,
5186 use @code{interrupt -a}.
5189 @node Thread-Specific Breakpoints
5190 @subsection Thread-Specific Breakpoints
5192 When your program has multiple threads (@pxref{Threads,, Debugging
5193 Programs with Multiple Threads}), you can choose whether to set
5194 breakpoints on all threads, or on a particular thread.
5197 @cindex breakpoints and threads
5198 @cindex thread breakpoints
5199 @kindex break @dots{} thread @var{threadno}
5200 @item break @var{linespec} thread @var{threadno}
5201 @itemx break @var{linespec} thread @var{threadno} if @dots{}
5202 @var{linespec} specifies source lines; there are several ways of
5203 writing them (@pxref{Specify Location}), but the effect is always to
5204 specify some source line.
5206 Use the qualifier @samp{thread @var{threadno}} with a breakpoint command
5207 to specify that you only want @value{GDBN} to stop the program when a
5208 particular thread reaches this breakpoint. @var{threadno} is one of the
5209 numeric thread identifiers assigned by @value{GDBN}, shown in the first
5210 column of the @samp{info threads} display.
5212 If you do not specify @samp{thread @var{threadno}} when you set a
5213 breakpoint, the breakpoint applies to @emph{all} threads of your
5216 You can use the @code{thread} qualifier on conditional breakpoints as
5217 well; in this case, place @samp{thread @var{threadno}} before the
5218 breakpoint condition, like this:
5221 (@value{GDBP}) break frik.c:13 thread 28 if bartab > lim
5226 @node Interrupted System Calls
5227 @subsection Interrupted System Calls
5229 @cindex thread breakpoints and system calls
5230 @cindex system calls and thread breakpoints
5231 @cindex premature return from system calls
5232 There is an unfortunate side effect when using @value{GDBN} to debug
5233 multi-threaded programs. If one thread stops for a
5234 breakpoint, or for some other reason, and another thread is blocked in a
5235 system call, then the system call may return prematurely. This is a
5236 consequence of the interaction between multiple threads and the signals
5237 that @value{GDBN} uses to implement breakpoints and other events that
5240 To handle this problem, your program should check the return value of
5241 each system call and react appropriately. This is good programming
5244 For example, do not write code like this:
5250 The call to @code{sleep} will return early if a different thread stops
5251 at a breakpoint or for some other reason.
5253 Instead, write this:
5258 unslept = sleep (unslept);
5261 A system call is allowed to return early, so the system is still
5262 conforming to its specification. But @value{GDBN} does cause your
5263 multi-threaded program to behave differently than it would without
5266 Also, @value{GDBN} uses internal breakpoints in the thread library to
5267 monitor certain events such as thread creation and thread destruction.
5268 When such an event happens, a system call in another thread may return
5269 prematurely, even though your program does not appear to stop.
5272 @node Reverse Execution
5273 @chapter Running programs backward
5274 @cindex reverse execution
5275 @cindex running programs backward
5277 When you are debugging a program, it is not unusual to realize that
5278 you have gone too far, and some event of interest has already happened.
5279 If the target environment supports it, @value{GDBN} can allow you to
5280 ``rewind'' the program by running it backward.
5282 A target environment that supports reverse execution should be able
5283 to ``undo'' the changes in machine state that have taken place as the
5284 program was executing normally. Variables, registers etc.@: should
5285 revert to their previous values. Obviously this requires a great
5286 deal of sophistication on the part of the target environment; not
5287 all target environments can support reverse execution.
5289 When a program is executed in reverse, the instructions that
5290 have most recently been executed are ``un-executed'', in reverse
5291 order. The program counter runs backward, following the previous
5292 thread of execution in reverse. As each instruction is ``un-executed'',
5293 the values of memory and/or registers that were changed by that
5294 instruction are reverted to their previous states. After executing
5295 a piece of source code in reverse, all side effects of that code
5296 should be ``undone'', and all variables should be returned to their
5297 prior values@footnote{
5298 Note that some side effects are easier to undo than others. For instance,
5299 memory and registers are relatively easy, but device I/O is hard. Some
5300 targets may be able undo things like device I/O, and some may not.
5302 The contract between @value{GDBN} and the reverse executing target
5303 requires only that the target do something reasonable when
5304 @value{GDBN} tells it to execute backwards, and then report the
5305 results back to @value{GDBN}. Whatever the target reports back to
5306 @value{GDBN}, @value{GDBN} will report back to the user. @value{GDBN}
5307 assumes that the memory and registers that the target reports are in a
5308 consistant state, but @value{GDBN} accepts whatever it is given.
5311 If you are debugging in a target environment that supports
5312 reverse execution, @value{GDBN} provides the following commands.
5315 @kindex reverse-continue
5316 @kindex rc @r{(@code{reverse-continue})}
5317 @item reverse-continue @r{[}@var{ignore-count}@r{]}
5318 @itemx rc @r{[}@var{ignore-count}@r{]}
5319 Beginning at the point where your program last stopped, start executing
5320 in reverse. Reverse execution will stop for breakpoints and synchronous
5321 exceptions (signals), just like normal execution. Behavior of
5322 asynchronous signals depends on the target environment.
5324 @kindex reverse-step
5325 @kindex rs @r{(@code{step})}
5326 @item reverse-step @r{[}@var{count}@r{]}
5327 Run the program backward until control reaches the start of a
5328 different source line; then stop it, and return control to @value{GDBN}.
5330 Like the @code{step} command, @code{reverse-step} will only stop
5331 at the beginning of a source line. It ``un-executes'' the previously
5332 executed source line. If the previous source line included calls to
5333 debuggable functions, @code{reverse-step} will step (backward) into
5334 the called function, stopping at the beginning of the @emph{last}
5335 statement in the called function (typically a return statement).
5337 Also, as with the @code{step} command, if non-debuggable functions are
5338 called, @code{reverse-step} will run thru them backward without stopping.
5340 @kindex reverse-stepi
5341 @kindex rsi @r{(@code{reverse-stepi})}
5342 @item reverse-stepi @r{[}@var{count}@r{]}
5343 Reverse-execute one machine instruction. Note that the instruction
5344 to be reverse-executed is @emph{not} the one pointed to by the program
5345 counter, but the instruction executed prior to that one. For instance,
5346 if the last instruction was a jump, @code{reverse-stepi} will take you
5347 back from the destination of the jump to the jump instruction itself.
5349 @kindex reverse-next
5350 @kindex rn @r{(@code{reverse-next})}
5351 @item reverse-next @r{[}@var{count}@r{]}
5352 Run backward to the beginning of the previous line executed in
5353 the current (innermost) stack frame. If the line contains function
5354 calls, they will be ``un-executed'' without stopping. Starting from
5355 the first line of a function, @code{reverse-next} will take you back
5356 to the caller of that function, @emph{before} the function was called,
5357 just as the normal @code{next} command would take you from the last
5358 line of a function back to its return to its caller
5359 @footnote{Unles the code is too heavily optimized.}.
5361 @kindex reverse-nexti
5362 @kindex rni @r{(@code{reverse-nexti})}
5363 @item reverse-nexti @r{[}@var{count}@r{]}
5364 Like @code{nexti}, @code{reverse-nexti} executes a single instruction
5365 in reverse, except that called functions are ``un-executed'' atomically.
5366 That is, if the previously executed instruction was a return from
5367 another instruction, @code{reverse-nexti} will continue to execute
5368 in reverse until the call to that function (from the current stack
5371 @kindex reverse-finish
5372 @item reverse-finish
5373 Just as the @code{finish} command takes you to the point where the
5374 current function returns, @code{reverse-finish} takes you to the point
5375 where it was called. Instead of ending up at the end of the current
5376 function invocation, you end up at the beginning.
5378 @kindex set exec-direction
5379 @item set exec-direction
5380 Set the direction of target execution.
5381 @itemx set exec-direction reverse
5382 @cindex execute forward or backward in time
5383 @value{GDBN} will perform all execution commands in reverse, until the
5384 exec-direction mode is changed to ``forward''. Affected commands include
5385 @code{step, stepi, next, nexti, continue, and finish}. The @code{return}
5386 command cannot be used in reverse mode.
5387 @item set exec-direction forward
5388 @value{GDBN} will perform all execution commands in the normal fashion.
5389 This is the default.
5393 @node Process Record and Replay
5394 @chapter Recording Inferior's Execution and Replaying It
5395 @cindex process record and replay
5396 @cindex recording inferior's execution and replaying it
5398 On some platforms, @value{GDBN} provides a special @dfn{process record
5399 and replay} target that can record a log of the process execution, and
5400 replay it later with both forward and reverse execution commands.
5403 When this target is in use, if the execution log includes the record
5404 for the next instruction, @value{GDBN} will debug in @dfn{replay
5405 mode}. In the replay mode, the inferior does not really execute code
5406 instructions. Instead, all the events that normally happen during
5407 code execution are taken from the execution log. While code is not
5408 really executed in replay mode, the values of registers (including the
5409 program counter register) and the memory of the inferior are still
5410 changed as they normally would. Their contents are taken from the
5414 If the record for the next instruction is not in the execution log,
5415 @value{GDBN} will debug in @dfn{record mode}. In this mode, the
5416 inferior executes normally, and @value{GDBN} records the execution log
5419 The process record and replay target supports reverse execution
5420 (@pxref{Reverse Execution}), even if the platform on which the
5421 inferior runs does not. However, the reverse execution is limited in
5422 this case by the range of the instructions recorded in the execution
5423 log. In other words, reverse execution on platforms that don't
5424 support it directly can only be done in the replay mode.
5426 When debugging in the reverse direction, @value{GDBN} will work in
5427 replay mode as long as the execution log includes the record for the
5428 previous instruction; otherwise, it will work in record mode, if the
5429 platform supports reverse execution, or stop if not.
5431 For architecture environments that support process record and replay,
5432 @value{GDBN} provides the following commands:
5435 @kindex target record
5439 This command starts the process record and replay target. The process
5440 record and replay target can only debug a process that is already
5441 running. Therefore, you need first to start the process with the
5442 @kbd{run} or @kbd{start} commands, and then start the recording with
5443 the @kbd{target record} command.
5445 Both @code{record} and @code{rec} are aliases of @code{target record}.
5447 @cindex displaced stepping, and process record and replay
5448 Displaced stepping (@pxref{Maintenance Commands,, displaced stepping})
5449 will be automatically disabled when process record and replay target
5450 is started. That's because the process record and replay target
5451 doesn't support displaced stepping.
5453 @cindex non-stop mode, and process record and replay
5454 @cindex asynchronous execution, and process record and replay
5455 If the inferior is in the non-stop mode (@pxref{Non-Stop Mode}) or in
5456 the asynchronous execution mode (@pxref{Background Execution}), the
5457 process record and replay target cannot be started because it doesn't
5458 support these two modes.
5463 Stop the process record and replay target. When process record and
5464 replay target stops, the entire execution log will be deleted and the
5465 inferior will either be terminated, or will remain in its final state.
5467 When you stop the process record and replay target in record mode (at
5468 the end of the execution log), the inferior will be stopped at the
5469 next instruction that would have been recorded. In other words, if
5470 you record for a while and then stop recording, the inferior process
5471 will be left in the same state as if the recording never happened.
5473 On the other hand, if the process record and replay target is stopped
5474 while in replay mode (that is, not at the end of the execution log,
5475 but at some earlier point), the inferior process will become ``live''
5476 at that earlier state, and it will then be possible to continue the
5477 usual ``live'' debugging of the process from that state.
5479 When the inferior process exits, or @value{GDBN} detaches from it,
5480 process record and replay target will automatically stop itself.
5482 @kindex set record insn-number-max
5483 @item set record insn-number-max @var{limit}
5484 Set the limit of instructions to be recorded. Default value is 200000.
5486 If @var{limit} is a positive number, then @value{GDBN} will start
5487 deleting instructions from the log once the number of the record
5488 instructions becomes greater than @var{limit}. For every new recorded
5489 instruction, @value{GDBN} will delete the earliest recorded
5490 instruction to keep the number of recorded instructions at the limit.
5491 (Since deleting recorded instructions loses information, @value{GDBN}
5492 lets you control what happens when the limit is reached, by means of
5493 the @code{stop-at-limit} option, described below.)
5495 If @var{limit} is zero, @value{GDBN} will never delete recorded
5496 instructions from the execution log. The number of recorded
5497 instructions is unlimited in this case.
5499 @kindex show record insn-number-max
5500 @item show record insn-number-max
5501 Show the limit of instructions to be recorded.
5503 @kindex set record stop-at-limit
5504 @item set record stop-at-limit
5505 Control the behavior when the number of recorded instructions reaches
5506 the limit. If ON (the default), @value{GDBN} will stop when the limit
5507 is reached for the first time and ask you whether you want to stop the
5508 inferior or continue running it and recording the execution log. If
5509 you decide to continue recording, each new recorded instruction will
5510 cause the oldest one to be deleted.
5512 If this option is OFF, @value{GDBN} will automatically delete the
5513 oldest record to make room for each new one, without asking.
5515 @kindex show record stop-at-limit
5516 @item show record stop-at-limit
5517 Show the current setting of @code{stop-at-limit}.
5519 @kindex info record insn-number
5520 @item info record insn-number
5521 Show the current number of recorded instructions.
5523 @kindex record delete
5526 When record target runs in replay mode (``in the past''), delete the
5527 subsequent execution log and begin to record a new execution log starting
5528 from the current address. This means you will abandon the previously
5529 recorded ``future'' and begin recording a new ``future''.
5534 @chapter Examining the Stack
5536 When your program has stopped, the first thing you need to know is where it
5537 stopped and how it got there.
5540 Each time your program performs a function call, information about the call
5542 That information includes the location of the call in your program,
5543 the arguments of the call,
5544 and the local variables of the function being called.
5545 The information is saved in a block of data called a @dfn{stack frame}.
5546 The stack frames are allocated in a region of memory called the @dfn{call
5549 When your program stops, the @value{GDBN} commands for examining the
5550 stack allow you to see all of this information.
5552 @cindex selected frame
5553 One of the stack frames is @dfn{selected} by @value{GDBN} and many
5554 @value{GDBN} commands refer implicitly to the selected frame. In
5555 particular, whenever you ask @value{GDBN} for the value of a variable in
5556 your program, the value is found in the selected frame. There are
5557 special @value{GDBN} commands to select whichever frame you are
5558 interested in. @xref{Selection, ,Selecting a Frame}.
5560 When your program stops, @value{GDBN} automatically selects the
5561 currently executing frame and describes it briefly, similar to the
5562 @code{frame} command (@pxref{Frame Info, ,Information about a Frame}).
5565 * Frames:: Stack frames
5566 * Backtrace:: Backtraces
5567 * Selection:: Selecting a frame
5568 * Frame Info:: Information on a frame
5573 @section Stack Frames
5575 @cindex frame, definition
5577 The call stack is divided up into contiguous pieces called @dfn{stack
5578 frames}, or @dfn{frames} for short; each frame is the data associated
5579 with one call to one function. The frame contains the arguments given
5580 to the function, the function's local variables, and the address at
5581 which the function is executing.
5583 @cindex initial frame
5584 @cindex outermost frame
5585 @cindex innermost frame
5586 When your program is started, the stack has only one frame, that of the
5587 function @code{main}. This is called the @dfn{initial} frame or the
5588 @dfn{outermost} frame. Each time a function is called, a new frame is
5589 made. Each time a function returns, the frame for that function invocation
5590 is eliminated. If a function is recursive, there can be many frames for
5591 the same function. The frame for the function in which execution is
5592 actually occurring is called the @dfn{innermost} frame. This is the most
5593 recently created of all the stack frames that still exist.
5595 @cindex frame pointer
5596 Inside your program, stack frames are identified by their addresses. A
5597 stack frame consists of many bytes, each of which has its own address; each
5598 kind of computer has a convention for choosing one byte whose
5599 address serves as the address of the frame. Usually this address is kept
5600 in a register called the @dfn{frame pointer register}
5601 (@pxref{Registers, $fp}) while execution is going on in that frame.
5603 @cindex frame number
5604 @value{GDBN} assigns numbers to all existing stack frames, starting with
5605 zero for the innermost frame, one for the frame that called it,
5606 and so on upward. These numbers do not really exist in your program;
5607 they are assigned by @value{GDBN} to give you a way of designating stack
5608 frames in @value{GDBN} commands.
5610 @c The -fomit-frame-pointer below perennially causes hbox overflow
5611 @c underflow problems.
5612 @cindex frameless execution
5613 Some compilers provide a way to compile functions so that they operate
5614 without stack frames. (For example, the @value{NGCC} option
5616 @samp{-fomit-frame-pointer}
5618 generates functions without a frame.)
5619 This is occasionally done with heavily used library functions to save
5620 the frame setup time. @value{GDBN} has limited facilities for dealing
5621 with these function invocations. If the innermost function invocation
5622 has no stack frame, @value{GDBN} nevertheless regards it as though
5623 it had a separate frame, which is numbered zero as usual, allowing
5624 correct tracing of the function call chain. However, @value{GDBN} has
5625 no provision for frameless functions elsewhere in the stack.
5628 @kindex frame@r{, command}
5629 @cindex current stack frame
5630 @item frame @var{args}
5631 The @code{frame} command allows you to move from one stack frame to another,
5632 and to print the stack frame you select. @var{args} may be either the
5633 address of the frame or the stack frame number. Without an argument,
5634 @code{frame} prints the current stack frame.
5636 @kindex select-frame
5637 @cindex selecting frame silently
5639 The @code{select-frame} command allows you to move from one stack frame
5640 to another without printing the frame. This is the silent version of
5648 @cindex call stack traces
5649 A backtrace is a summary of how your program got where it is. It shows one
5650 line per frame, for many frames, starting with the currently executing
5651 frame (frame zero), followed by its caller (frame one), and on up the
5656 @kindex bt @r{(@code{backtrace})}
5659 Print a backtrace of the entire stack: one line per frame for all
5660 frames in the stack.
5662 You can stop the backtrace at any time by typing the system interrupt
5663 character, normally @kbd{Ctrl-c}.
5665 @item backtrace @var{n}
5667 Similar, but print only the innermost @var{n} frames.
5669 @item backtrace -@var{n}
5671 Similar, but print only the outermost @var{n} frames.
5673 @item backtrace full
5675 @itemx bt full @var{n}
5676 @itemx bt full -@var{n}
5677 Print the values of the local variables also. @var{n} specifies the
5678 number of frames to print, as described above.
5683 The names @code{where} and @code{info stack} (abbreviated @code{info s})
5684 are additional aliases for @code{backtrace}.
5686 @cindex multiple threads, backtrace
5687 In a multi-threaded program, @value{GDBN} by default shows the
5688 backtrace only for the current thread. To display the backtrace for
5689 several or all of the threads, use the command @code{thread apply}
5690 (@pxref{Threads, thread apply}). For example, if you type @kbd{thread
5691 apply all backtrace}, @value{GDBN} will display the backtrace for all
5692 the threads; this is handy when you debug a core dump of a
5693 multi-threaded program.
5695 Each line in the backtrace shows the frame number and the function name.
5696 The program counter value is also shown---unless you use @code{set
5697 print address off}. The backtrace also shows the source file name and
5698 line number, as well as the arguments to the function. The program
5699 counter value is omitted if it is at the beginning of the code for that
5702 Here is an example of a backtrace. It was made with the command
5703 @samp{bt 3}, so it shows the innermost three frames.
5707 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
5709 #1 0x6e38 in expand_macro (sym=0x2b600, data=...) at macro.c:242
5710 #2 0x6840 in expand_token (obs=0x0, t=177664, td=0xf7fffb08)
5712 (More stack frames follow...)
5717 The display for frame zero does not begin with a program counter
5718 value, indicating that your program has stopped at the beginning of the
5719 code for line @code{993} of @code{builtin.c}.
5722 The value of parameter @code{data} in frame 1 has been replaced by
5723 @code{@dots{}}. By default, @value{GDBN} prints the value of a parameter
5724 only if it is a scalar (integer, pointer, enumeration, etc). See command
5725 @kbd{set print frame-arguments} in @ref{Print Settings} for more details
5726 on how to configure the way function parameter values are printed.
5728 @cindex value optimized out, in backtrace
5729 @cindex function call arguments, optimized out
5730 If your program was compiled with optimizations, some compilers will
5731 optimize away arguments passed to functions if those arguments are
5732 never used after the call. Such optimizations generate code that
5733 passes arguments through registers, but doesn't store those arguments
5734 in the stack frame. @value{GDBN} has no way of displaying such
5735 arguments in stack frames other than the innermost one. Here's what
5736 such a backtrace might look like:
5740 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
5742 #1 0x6e38 in expand_macro (sym=<value optimized out>) at macro.c:242
5743 #2 0x6840 in expand_token (obs=0x0, t=<value optimized out>, td=0xf7fffb08)
5745 (More stack frames follow...)
5750 The values of arguments that were not saved in their stack frames are
5751 shown as @samp{<value optimized out>}.
5753 If you need to display the values of such optimized-out arguments,
5754 either deduce that from other variables whose values depend on the one
5755 you are interested in, or recompile without optimizations.
5757 @cindex backtrace beyond @code{main} function
5758 @cindex program entry point
5759 @cindex startup code, and backtrace
5760 Most programs have a standard user entry point---a place where system
5761 libraries and startup code transition into user code. For C this is
5762 @code{main}@footnote{
5763 Note that embedded programs (the so-called ``free-standing''
5764 environment) are not required to have a @code{main} function as the
5765 entry point. They could even have multiple entry points.}.
5766 When @value{GDBN} finds the entry function in a backtrace
5767 it will terminate the backtrace, to avoid tracing into highly
5768 system-specific (and generally uninteresting) code.
5770 If you need to examine the startup code, or limit the number of levels
5771 in a backtrace, you can change this behavior:
5774 @item set backtrace past-main
5775 @itemx set backtrace past-main on
5776 @kindex set backtrace
5777 Backtraces will continue past the user entry point.
5779 @item set backtrace past-main off
5780 Backtraces will stop when they encounter the user entry point. This is the
5783 @item show backtrace past-main
5784 @kindex show backtrace
5785 Display the current user entry point backtrace policy.
5787 @item set backtrace past-entry
5788 @itemx set backtrace past-entry on
5789 Backtraces will continue past the internal entry point of an application.
5790 This entry point is encoded by the linker when the application is built,
5791 and is likely before the user entry point @code{main} (or equivalent) is called.
5793 @item set backtrace past-entry off
5794 Backtraces will stop when they encounter the internal entry point of an
5795 application. This is the default.
5797 @item show backtrace past-entry
5798 Display the current internal entry point backtrace policy.
5800 @item set backtrace limit @var{n}
5801 @itemx set backtrace limit 0
5802 @cindex backtrace limit
5803 Limit the backtrace to @var{n} levels. A value of zero means
5806 @item show backtrace limit
5807 Display the current limit on backtrace levels.
5811 @section Selecting a Frame
5813 Most commands for examining the stack and other data in your program work on
5814 whichever stack frame is selected at the moment. Here are the commands for
5815 selecting a stack frame; all of them finish by printing a brief description
5816 of the stack frame just selected.
5819 @kindex frame@r{, selecting}
5820 @kindex f @r{(@code{frame})}
5823 Select frame number @var{n}. Recall that frame zero is the innermost
5824 (currently executing) frame, frame one is the frame that called the
5825 innermost one, and so on. The highest-numbered frame is the one for
5828 @item frame @var{addr}
5830 Select the frame at address @var{addr}. This is useful mainly if the
5831 chaining of stack frames has been damaged by a bug, making it
5832 impossible for @value{GDBN} to assign numbers properly to all frames. In
5833 addition, this can be useful when your program has multiple stacks and
5834 switches between them.
5836 On the SPARC architecture, @code{frame} needs two addresses to
5837 select an arbitrary frame: a frame pointer and a stack pointer.
5839 On the MIPS and Alpha architecture, it needs two addresses: a stack
5840 pointer and a program counter.
5842 On the 29k architecture, it needs three addresses: a register stack
5843 pointer, a program counter, and a memory stack pointer.
5847 Move @var{n} frames up the stack. For positive numbers @var{n}, this
5848 advances toward the outermost frame, to higher frame numbers, to frames
5849 that have existed longer. @var{n} defaults to one.
5852 @kindex do @r{(@code{down})}
5854 Move @var{n} frames down the stack. For positive numbers @var{n}, this
5855 advances toward the innermost frame, to lower frame numbers, to frames
5856 that were created more recently. @var{n} defaults to one. You may
5857 abbreviate @code{down} as @code{do}.
5860 All of these commands end by printing two lines of output describing the
5861 frame. The first line shows the frame number, the function name, the
5862 arguments, and the source file and line number of execution in that
5863 frame. The second line shows the text of that source line.
5871 #1 0x22f0 in main (argc=1, argv=0xf7fffbf4, env=0xf7fffbfc)
5873 10 read_input_file (argv[i]);
5877 After such a printout, the @code{list} command with no arguments
5878 prints ten lines centered on the point of execution in the frame.
5879 You can also edit the program at the point of execution with your favorite
5880 editing program by typing @code{edit}.
5881 @xref{List, ,Printing Source Lines},
5885 @kindex down-silently
5887 @item up-silently @var{n}
5888 @itemx down-silently @var{n}
5889 These two commands are variants of @code{up} and @code{down},
5890 respectively; they differ in that they do their work silently, without
5891 causing display of the new frame. They are intended primarily for use
5892 in @value{GDBN} command scripts, where the output might be unnecessary and
5897 @section Information About a Frame
5899 There are several other commands to print information about the selected
5905 When used without any argument, this command does not change which
5906 frame is selected, but prints a brief description of the currently
5907 selected stack frame. It can be abbreviated @code{f}. With an
5908 argument, this command is used to select a stack frame.
5909 @xref{Selection, ,Selecting a Frame}.
5912 @kindex info f @r{(@code{info frame})}
5915 This command prints a verbose description of the selected stack frame,
5920 the address of the frame
5922 the address of the next frame down (called by this frame)
5924 the address of the next frame up (caller of this frame)
5926 the language in which the source code corresponding to this frame is written
5928 the address of the frame's arguments
5930 the address of the frame's local variables
5932 the program counter saved in it (the address of execution in the caller frame)
5934 which registers were saved in the frame
5937 @noindent The verbose description is useful when
5938 something has gone wrong that has made the stack format fail to fit
5939 the usual conventions.
5941 @item info frame @var{addr}
5942 @itemx info f @var{addr}
5943 Print a verbose description of the frame at address @var{addr}, without
5944 selecting that frame. The selected frame remains unchanged by this
5945 command. This requires the same kind of address (more than one for some
5946 architectures) that you specify in the @code{frame} command.
5947 @xref{Selection, ,Selecting a Frame}.
5951 Print the arguments of the selected frame, each on a separate line.
5955 Print the local variables of the selected frame, each on a separate
5956 line. These are all variables (declared either static or automatic)
5957 accessible at the point of execution of the selected frame.
5960 @cindex catch exceptions, list active handlers
5961 @cindex exception handlers, how to list
5963 Print a list of all the exception handlers that are active in the
5964 current stack frame at the current point of execution. To see other
5965 exception handlers, visit the associated frame (using the @code{up},
5966 @code{down}, or @code{frame} commands); then type @code{info catch}.
5967 @xref{Set Catchpoints, , Setting Catchpoints}.
5973 @chapter Examining Source Files
5975 @value{GDBN} can print parts of your program's source, since the debugging
5976 information recorded in the program tells @value{GDBN} what source files were
5977 used to build it. When your program stops, @value{GDBN} spontaneously prints
5978 the line where it stopped. Likewise, when you select a stack frame
5979 (@pxref{Selection, ,Selecting a Frame}), @value{GDBN} prints the line where
5980 execution in that frame has stopped. You can print other portions of
5981 source files by explicit command.
5983 If you use @value{GDBN} through its @sc{gnu} Emacs interface, you may
5984 prefer to use Emacs facilities to view source; see @ref{Emacs, ,Using
5985 @value{GDBN} under @sc{gnu} Emacs}.
5988 * List:: Printing source lines
5989 * Specify Location:: How to specify code locations
5990 * Edit:: Editing source files
5991 * Search:: Searching source files
5992 * Source Path:: Specifying source directories
5993 * Machine Code:: Source and machine code
5997 @section Printing Source Lines
6000 @kindex l @r{(@code{list})}
6001 To print lines from a source file, use the @code{list} command
6002 (abbreviated @code{l}). By default, ten lines are printed.
6003 There are several ways to specify what part of the file you want to
6004 print; see @ref{Specify Location}, for the full list.
6006 Here are the forms of the @code{list} command most commonly used:
6009 @item list @var{linenum}
6010 Print lines centered around line number @var{linenum} in the
6011 current source file.
6013 @item list @var{function}
6014 Print lines centered around the beginning of function
6018 Print more lines. If the last lines printed were printed with a
6019 @code{list} command, this prints lines following the last lines
6020 printed; however, if the last line printed was a solitary line printed
6021 as part of displaying a stack frame (@pxref{Stack, ,Examining the
6022 Stack}), this prints lines centered around that line.
6025 Print lines just before the lines last printed.
6028 @cindex @code{list}, how many lines to display
6029 By default, @value{GDBN} prints ten source lines with any of these forms of
6030 the @code{list} command. You can change this using @code{set listsize}:
6033 @kindex set listsize
6034 @item set listsize @var{count}
6035 Make the @code{list} command display @var{count} source lines (unless
6036 the @code{list} argument explicitly specifies some other number).
6038 @kindex show listsize
6040 Display the number of lines that @code{list} prints.
6043 Repeating a @code{list} command with @key{RET} discards the argument,
6044 so it is equivalent to typing just @code{list}. This is more useful
6045 than listing the same lines again. An exception is made for an
6046 argument of @samp{-}; that argument is preserved in repetition so that
6047 each repetition moves up in the source file.
6049 In general, the @code{list} command expects you to supply zero, one or two
6050 @dfn{linespecs}. Linespecs specify source lines; there are several ways
6051 of writing them (@pxref{Specify Location}), but the effect is always
6052 to specify some source line.
6054 Here is a complete description of the possible arguments for @code{list}:
6057 @item list @var{linespec}
6058 Print lines centered around the line specified by @var{linespec}.
6060 @item list @var{first},@var{last}
6061 Print lines from @var{first} to @var{last}. Both arguments are
6062 linespecs. When a @code{list} command has two linespecs, and the
6063 source file of the second linespec is omitted, this refers to
6064 the same source file as the first linespec.
6066 @item list ,@var{last}
6067 Print lines ending with @var{last}.
6069 @item list @var{first},
6070 Print lines starting with @var{first}.
6073 Print lines just after the lines last printed.
6076 Print lines just before the lines last printed.
6079 As described in the preceding table.
6082 @node Specify Location
6083 @section Specifying a Location
6084 @cindex specifying location
6087 Several @value{GDBN} commands accept arguments that specify a location
6088 of your program's code. Since @value{GDBN} is a source-level
6089 debugger, a location usually specifies some line in the source code;
6090 for that reason, locations are also known as @dfn{linespecs}.
6092 Here are all the different ways of specifying a code location that
6093 @value{GDBN} understands:
6097 Specifies the line number @var{linenum} of the current source file.
6100 @itemx +@var{offset}
6101 Specifies the line @var{offset} lines before or after the @dfn{current
6102 line}. For the @code{list} command, the current line is the last one
6103 printed; for the breakpoint commands, this is the line at which
6104 execution stopped in the currently selected @dfn{stack frame}
6105 (@pxref{Frames, ,Frames}, for a description of stack frames.) When
6106 used as the second of the two linespecs in a @code{list} command,
6107 this specifies the line @var{offset} lines up or down from the first
6110 @item @var{filename}:@var{linenum}
6111 Specifies the line @var{linenum} in the source file @var{filename}.
6113 @item @var{function}
6114 Specifies the line that begins the body of the function @var{function}.
6115 For example, in C, this is the line with the open brace.
6117 @item @var{filename}:@var{function}
6118 Specifies the line that begins the body of the function @var{function}
6119 in the file @var{filename}. You only need the file name with a
6120 function name to avoid ambiguity when there are identically named
6121 functions in different source files.
6123 @item *@var{address}
6124 Specifies the program address @var{address}. For line-oriented
6125 commands, such as @code{list} and @code{edit}, this specifies a source
6126 line that contains @var{address}. For @code{break} and other
6127 breakpoint oriented commands, this can be used to set breakpoints in
6128 parts of your program which do not have debugging information or
6131 Here @var{address} may be any expression valid in the current working
6132 language (@pxref{Languages, working language}) that specifies a code
6133 address. In addition, as a convenience, @value{GDBN} extends the
6134 semantics of expressions used in locations to cover the situations
6135 that frequently happen during debugging. Here are the various forms
6139 @item @var{expression}
6140 Any expression valid in the current working language.
6142 @item @var{funcaddr}
6143 An address of a function or procedure derived from its name. In C,
6144 C@t{++}, Java, Objective-C, Fortran, minimal, and assembly, this is
6145 simply the function's name @var{function} (and actually a special case
6146 of a valid expression). In Pascal and Modula-2, this is
6147 @code{&@var{function}}. In Ada, this is @code{@var{function}'Address}
6148 (although the Pascal form also works).
6150 This form specifies the address of the function's first instruction,
6151 before the stack frame and arguments have been set up.
6153 @item '@var{filename}'::@var{funcaddr}
6154 Like @var{funcaddr} above, but also specifies the name of the source
6155 file explicitly. This is useful if the name of the function does not
6156 specify the function unambiguously, e.g., if there are several
6157 functions with identical names in different source files.
6164 @section Editing Source Files
6165 @cindex editing source files
6168 @kindex e @r{(@code{edit})}
6169 To edit the lines in a source file, use the @code{edit} command.
6170 The editing program of your choice
6171 is invoked with the current line set to
6172 the active line in the program.
6173 Alternatively, there are several ways to specify what part of the file you
6174 want to print if you want to see other parts of the program:
6177 @item edit @var{location}
6178 Edit the source file specified by @code{location}. Editing starts at
6179 that @var{location}, e.g., at the specified source line of the
6180 specified file. @xref{Specify Location}, for all the possible forms
6181 of the @var{location} argument; here are the forms of the @code{edit}
6182 command most commonly used:
6185 @item edit @var{number}
6186 Edit the current source file with @var{number} as the active line number.
6188 @item edit @var{function}
6189 Edit the file containing @var{function} at the beginning of its definition.
6194 @subsection Choosing your Editor
6195 You can customize @value{GDBN} to use any editor you want
6197 The only restriction is that your editor (say @code{ex}), recognizes the
6198 following command-line syntax:
6200 ex +@var{number} file
6202 The optional numeric value +@var{number} specifies the number of the line in
6203 the file where to start editing.}.
6204 By default, it is @file{@value{EDITOR}}, but you can change this
6205 by setting the environment variable @code{EDITOR} before using
6206 @value{GDBN}. For example, to configure @value{GDBN} to use the
6207 @code{vi} editor, you could use these commands with the @code{sh} shell:
6213 or in the @code{csh} shell,
6215 setenv EDITOR /usr/bin/vi
6220 @section Searching Source Files
6221 @cindex searching source files
6223 There are two commands for searching through the current source file for a
6228 @kindex forward-search
6229 @item forward-search @var{regexp}
6230 @itemx search @var{regexp}
6231 The command @samp{forward-search @var{regexp}} checks each line,
6232 starting with the one following the last line listed, for a match for
6233 @var{regexp}. It lists the line that is found. You can use the
6234 synonym @samp{search @var{regexp}} or abbreviate the command name as
6237 @kindex reverse-search
6238 @item reverse-search @var{regexp}
6239 The command @samp{reverse-search @var{regexp}} checks each line, starting
6240 with the one before the last line listed and going backward, for a match
6241 for @var{regexp}. It lists the line that is found. You can abbreviate
6242 this command as @code{rev}.
6246 @section Specifying Source Directories
6249 @cindex directories for source files
6250 Executable programs sometimes do not record the directories of the source
6251 files from which they were compiled, just the names. Even when they do,
6252 the directories could be moved between the compilation and your debugging
6253 session. @value{GDBN} has a list of directories to search for source files;
6254 this is called the @dfn{source path}. Each time @value{GDBN} wants a source file,
6255 it tries all the directories in the list, in the order they are present
6256 in the list, until it finds a file with the desired name.
6258 For example, suppose an executable references the file
6259 @file{/usr/src/foo-1.0/lib/foo.c}, and our source path is
6260 @file{/mnt/cross}. The file is first looked up literally; if this
6261 fails, @file{/mnt/cross/usr/src/foo-1.0/lib/foo.c} is tried; if this
6262 fails, @file{/mnt/cross/foo.c} is opened; if this fails, an error
6263 message is printed. @value{GDBN} does not look up the parts of the
6264 source file name, such as @file{/mnt/cross/src/foo-1.0/lib/foo.c}.
6265 Likewise, the subdirectories of the source path are not searched: if
6266 the source path is @file{/mnt/cross}, and the binary refers to
6267 @file{foo.c}, @value{GDBN} would not find it under
6268 @file{/mnt/cross/usr/src/foo-1.0/lib}.
6270 Plain file names, relative file names with leading directories, file
6271 names containing dots, etc.@: are all treated as described above; for
6272 instance, if the source path is @file{/mnt/cross}, and the source file
6273 is recorded as @file{../lib/foo.c}, @value{GDBN} would first try
6274 @file{../lib/foo.c}, then @file{/mnt/cross/../lib/foo.c}, and after
6275 that---@file{/mnt/cross/foo.c}.
6277 Note that the executable search path is @emph{not} used to locate the
6280 Whenever you reset or rearrange the source path, @value{GDBN} clears out
6281 any information it has cached about where source files are found and where
6282 each line is in the file.
6286 When you start @value{GDBN}, its source path includes only @samp{cdir}
6287 and @samp{cwd}, in that order.
6288 To add other directories, use the @code{directory} command.
6290 The search path is used to find both program source files and @value{GDBN}
6291 script files (read using the @samp{-command} option and @samp{source} command).
6293 In addition to the source path, @value{GDBN} provides a set of commands
6294 that manage a list of source path substitution rules. A @dfn{substitution
6295 rule} specifies how to rewrite source directories stored in the program's
6296 debug information in case the sources were moved to a different
6297 directory between compilation and debugging. A rule is made of
6298 two strings, the first specifying what needs to be rewritten in
6299 the path, and the second specifying how it should be rewritten.
6300 In @ref{set substitute-path}, we name these two parts @var{from} and
6301 @var{to} respectively. @value{GDBN} does a simple string replacement
6302 of @var{from} with @var{to} at the start of the directory part of the
6303 source file name, and uses that result instead of the original file
6304 name to look up the sources.
6306 Using the previous example, suppose the @file{foo-1.0} tree has been
6307 moved from @file{/usr/src} to @file{/mnt/cross}, then you can tell
6308 @value{GDBN} to replace @file{/usr/src} in all source path names with
6309 @file{/mnt/cross}. The first lookup will then be
6310 @file{/mnt/cross/foo-1.0/lib/foo.c} in place of the original location
6311 of @file{/usr/src/foo-1.0/lib/foo.c}. To define a source path
6312 substitution rule, use the @code{set substitute-path} command
6313 (@pxref{set substitute-path}).
6315 To avoid unexpected substitution results, a rule is applied only if the
6316 @var{from} part of the directory name ends at a directory separator.
6317 For instance, a rule substituting @file{/usr/source} into
6318 @file{/mnt/cross} will be applied to @file{/usr/source/foo-1.0} but
6319 not to @file{/usr/sourceware/foo-2.0}. And because the substitution
6320 is applied only at the beginning of the directory name, this rule will
6321 not be applied to @file{/root/usr/source/baz.c} either.
6323 In many cases, you can achieve the same result using the @code{directory}
6324 command. However, @code{set substitute-path} can be more efficient in
6325 the case where the sources are organized in a complex tree with multiple
6326 subdirectories. With the @code{directory} command, you need to add each
6327 subdirectory of your project. If you moved the entire tree while
6328 preserving its internal organization, then @code{set substitute-path}
6329 allows you to direct the debugger to all the sources with one single
6332 @code{set substitute-path} is also more than just a shortcut command.
6333 The source path is only used if the file at the original location no
6334 longer exists. On the other hand, @code{set substitute-path} modifies
6335 the debugger behavior to look at the rewritten location instead. So, if
6336 for any reason a source file that is not relevant to your executable is
6337 located at the original location, a substitution rule is the only
6338 method available to point @value{GDBN} at the new location.
6340 @cindex @samp{--with-relocated-sources}
6341 @cindex default source path substitution
6342 You can configure a default source path substitution rule by
6343 configuring @value{GDBN} with the
6344 @samp{--with-relocated-sources=@var{dir}} option. The @var{dir}
6345 should be the name of a directory under @value{GDBN}'s configured
6346 prefix (set with @samp{--prefix} or @samp{--exec-prefix}), and
6347 directory names in debug information under @var{dir} will be adjusted
6348 automatically if the installed @value{GDBN} is moved to a new
6349 location. This is useful if @value{GDBN}, libraries or executables
6350 with debug information and corresponding source code are being moved
6354 @item directory @var{dirname} @dots{}
6355 @item dir @var{dirname} @dots{}
6356 Add directory @var{dirname} to the front of the source path. Several
6357 directory names may be given to this command, separated by @samp{:}
6358 (@samp{;} on MS-DOS and MS-Windows, where @samp{:} usually appears as
6359 part of absolute file names) or
6360 whitespace. You may specify a directory that is already in the source
6361 path; this moves it forward, so @value{GDBN} searches it sooner.
6365 @vindex $cdir@r{, convenience variable}
6366 @vindex $cwd@r{, convenience variable}
6367 @cindex compilation directory
6368 @cindex current directory
6369 @cindex working directory
6370 @cindex directory, current
6371 @cindex directory, compilation
6372 You can use the string @samp{$cdir} to refer to the compilation
6373 directory (if one is recorded), and @samp{$cwd} to refer to the current
6374 working directory. @samp{$cwd} is not the same as @samp{.}---the former
6375 tracks the current working directory as it changes during your @value{GDBN}
6376 session, while the latter is immediately expanded to the current
6377 directory at the time you add an entry to the source path.
6380 Reset the source path to its default value (@samp{$cdir:$cwd} on Unix systems). This requires confirmation.
6382 @c RET-repeat for @code{directory} is explicitly disabled, but since
6383 @c repeating it would be a no-op we do not say that. (thanks to RMS)
6385 @item show directories
6386 @kindex show directories
6387 Print the source path: show which directories it contains.
6389 @anchor{set substitute-path}
6390 @item set substitute-path @var{from} @var{to}
6391 @kindex set substitute-path
6392 Define a source path substitution rule, and add it at the end of the
6393 current list of existing substitution rules. If a rule with the same
6394 @var{from} was already defined, then the old rule is also deleted.
6396 For example, if the file @file{/foo/bar/baz.c} was moved to
6397 @file{/mnt/cross/baz.c}, then the command
6400 (@value{GDBP}) set substitute-path /usr/src /mnt/cross
6404 will tell @value{GDBN} to replace @samp{/usr/src} with
6405 @samp{/mnt/cross}, which will allow @value{GDBN} to find the file
6406 @file{baz.c} even though it was moved.
6408 In the case when more than one substitution rule have been defined,
6409 the rules are evaluated one by one in the order where they have been
6410 defined. The first one matching, if any, is selected to perform
6413 For instance, if we had entered the following commands:
6416 (@value{GDBP}) set substitute-path /usr/src/include /mnt/include
6417 (@value{GDBP}) set substitute-path /usr/src /mnt/src
6421 @value{GDBN} would then rewrite @file{/usr/src/include/defs.h} into
6422 @file{/mnt/include/defs.h} by using the first rule. However, it would
6423 use the second rule to rewrite @file{/usr/src/lib/foo.c} into
6424 @file{/mnt/src/lib/foo.c}.
6427 @item unset substitute-path [path]
6428 @kindex unset substitute-path
6429 If a path is specified, search the current list of substitution rules
6430 for a rule that would rewrite that path. Delete that rule if found.
6431 A warning is emitted by the debugger if no rule could be found.
6433 If no path is specified, then all substitution rules are deleted.
6435 @item show substitute-path [path]
6436 @kindex show substitute-path
6437 If a path is specified, then print the source path substitution rule
6438 which would rewrite that path, if any.
6440 If no path is specified, then print all existing source path substitution
6445 If your source path is cluttered with directories that are no longer of
6446 interest, @value{GDBN} may sometimes cause confusion by finding the wrong
6447 versions of source. You can correct the situation as follows:
6451 Use @code{directory} with no argument to reset the source path to its default value.
6454 Use @code{directory} with suitable arguments to reinstall the
6455 directories you want in the source path. You can add all the
6456 directories in one command.
6460 @section Source and Machine Code
6461 @cindex source line and its code address
6463 You can use the command @code{info line} to map source lines to program
6464 addresses (and vice versa), and the command @code{disassemble} to display
6465 a range of addresses as machine instructions. You can use the command
6466 @code{set disassemble-next-line} to set whether to disassemble next
6467 source line when execution stops. When run under @sc{gnu} Emacs
6468 mode, the @code{info line} command causes the arrow to point to the
6469 line specified. Also, @code{info line} prints addresses in symbolic form as
6474 @item info line @var{linespec}
6475 Print the starting and ending addresses of the compiled code for
6476 source line @var{linespec}. You can specify source lines in any of
6477 the ways documented in @ref{Specify Location}.
6480 For example, we can use @code{info line} to discover the location of
6481 the object code for the first line of function
6482 @code{m4_changequote}:
6484 @c FIXME: I think this example should also show the addresses in
6485 @c symbolic form, as they usually would be displayed.
6487 (@value{GDBP}) info line m4_changequote
6488 Line 895 of "builtin.c" starts at pc 0x634c and ends at 0x6350.
6492 @cindex code address and its source line
6493 We can also inquire (using @code{*@var{addr}} as the form for
6494 @var{linespec}) what source line covers a particular address:
6496 (@value{GDBP}) info line *0x63ff
6497 Line 926 of "builtin.c" starts at pc 0x63e4 and ends at 0x6404.
6500 @cindex @code{$_} and @code{info line}
6501 @cindex @code{x} command, default address
6502 @kindex x@r{(examine), and} info line
6503 After @code{info line}, the default address for the @code{x} command
6504 is changed to the starting address of the line, so that @samp{x/i} is
6505 sufficient to begin examining the machine code (@pxref{Memory,
6506 ,Examining Memory}). Also, this address is saved as the value of the
6507 convenience variable @code{$_} (@pxref{Convenience Vars, ,Convenience
6512 @cindex assembly instructions
6513 @cindex instructions, assembly
6514 @cindex machine instructions
6515 @cindex listing machine instructions
6517 @itemx disassemble /m
6518 @itemx disassemble /r
6519 This specialized command dumps a range of memory as machine
6520 instructions. It can also print mixed source+disassembly by specifying
6521 the @code{/m} modifier and print the raw instructions in hex as well as
6522 in symbolic form by specifying the @code{/r}.
6523 The default memory range is the function surrounding the
6524 program counter of the selected frame. A single argument to this
6525 command is a program counter value; @value{GDBN} dumps the function
6526 surrounding this value. Two arguments specify a range of addresses
6527 (first inclusive, second exclusive) to dump.
6530 The following example shows the disassembly of a range of addresses of
6531 HP PA-RISC 2.0 code:
6534 (@value{GDBP}) disas 0x32c4 0x32e4
6535 Dump of assembler code from 0x32c4 to 0x32e4:
6536 0x32c4 <main+204>: addil 0,dp
6537 0x32c8 <main+208>: ldw 0x22c(sr0,r1),r26
6538 0x32cc <main+212>: ldil 0x3000,r31
6539 0x32d0 <main+216>: ble 0x3f8(sr4,r31)
6540 0x32d4 <main+220>: ldo 0(r31),rp
6541 0x32d8 <main+224>: addil -0x800,dp
6542 0x32dc <main+228>: ldo 0x588(r1),r26
6543 0x32e0 <main+232>: ldil 0x3000,r31
6544 End of assembler dump.
6547 Here is an example showing mixed source+assembly for Intel x86:
6550 (@value{GDBP}) disas /m main
6551 Dump of assembler code for function main:
6553 0x08048330 <main+0>: push %ebp
6554 0x08048331 <main+1>: mov %esp,%ebp
6555 0x08048333 <main+3>: sub $0x8,%esp
6556 0x08048336 <main+6>: and $0xfffffff0,%esp
6557 0x08048339 <main+9>: sub $0x10,%esp
6559 6 printf ("Hello.\n");
6560 0x0804833c <main+12>: movl $0x8048440,(%esp)
6561 0x08048343 <main+19>: call 0x8048284 <puts@@plt>
6565 0x08048348 <main+24>: mov $0x0,%eax
6566 0x0804834d <main+29>: leave
6567 0x0804834e <main+30>: ret
6569 End of assembler dump.
6572 Some architectures have more than one commonly-used set of instruction
6573 mnemonics or other syntax.
6575 For programs that were dynamically linked and use shared libraries,
6576 instructions that call functions or branch to locations in the shared
6577 libraries might show a seemingly bogus location---it's actually a
6578 location of the relocation table. On some architectures, @value{GDBN}
6579 might be able to resolve these to actual function names.
6582 @kindex set disassembly-flavor
6583 @cindex Intel disassembly flavor
6584 @cindex AT&T disassembly flavor
6585 @item set disassembly-flavor @var{instruction-set}
6586 Select the instruction set to use when disassembling the
6587 program via the @code{disassemble} or @code{x/i} commands.
6589 Currently this command is only defined for the Intel x86 family. You
6590 can set @var{instruction-set} to either @code{intel} or @code{att}.
6591 The default is @code{att}, the AT&T flavor used by default by Unix
6592 assemblers for x86-based targets.
6594 @kindex show disassembly-flavor
6595 @item show disassembly-flavor
6596 Show the current setting of the disassembly flavor.
6600 @kindex set disassemble-next-line
6601 @kindex show disassemble-next-line
6602 @item set disassemble-next-line
6603 @itemx show disassemble-next-line
6604 Control whether or not @value{GDBN} will disassemble the next source
6605 line or instruction when execution stops. If ON, @value{GDBN} will
6606 display disassembly of the next source line when execution of the
6607 program being debugged stops. This is @emph{in addition} to
6608 displaying the source line itself, which @value{GDBN} always does if
6609 possible. If the next source line cannot be displayed for some reason
6610 (e.g., if @value{GDBN} cannot find the source file, or there's no line
6611 info in the debug info), @value{GDBN} will display disassembly of the
6612 next @emph{instruction} instead of showing the next source line. If
6613 AUTO, @value{GDBN} will display disassembly of next instruction only
6614 if the source line cannot be displayed. This setting causes
6615 @value{GDBN} to display some feedback when you step through a function
6616 with no line info or whose source file is unavailable. The default is
6617 OFF, which means never display the disassembly of the next line or
6623 @chapter Examining Data
6625 @cindex printing data
6626 @cindex examining data
6629 @c "inspect" is not quite a synonym if you are using Epoch, which we do not
6630 @c document because it is nonstandard... Under Epoch it displays in a
6631 @c different window or something like that.
6632 The usual way to examine data in your program is with the @code{print}
6633 command (abbreviated @code{p}), or its synonym @code{inspect}. It
6634 evaluates and prints the value of an expression of the language your
6635 program is written in (@pxref{Languages, ,Using @value{GDBN} with
6636 Different Languages}).
6639 @item print @var{expr}
6640 @itemx print /@var{f} @var{expr}
6641 @var{expr} is an expression (in the source language). By default the
6642 value of @var{expr} is printed in a format appropriate to its data type;
6643 you can choose a different format by specifying @samp{/@var{f}}, where
6644 @var{f} is a letter specifying the format; see @ref{Output Formats,,Output
6648 @itemx print /@var{f}
6649 @cindex reprint the last value
6650 If you omit @var{expr}, @value{GDBN} displays the last value again (from the
6651 @dfn{value history}; @pxref{Value History, ,Value History}). This allows you to
6652 conveniently inspect the same value in an alternative format.
6655 A more low-level way of examining data is with the @code{x} command.
6656 It examines data in memory at a specified address and prints it in a
6657 specified format. @xref{Memory, ,Examining Memory}.
6659 If you are interested in information about types, or about how the
6660 fields of a struct or a class are declared, use the @code{ptype @var{exp}}
6661 command rather than @code{print}. @xref{Symbols, ,Examining the Symbol
6665 * Expressions:: Expressions
6666 * Ambiguous Expressions:: Ambiguous Expressions
6667 * Variables:: Program variables
6668 * Arrays:: Artificial arrays
6669 * Output Formats:: Output formats
6670 * Memory:: Examining memory
6671 * Auto Display:: Automatic display
6672 * Print Settings:: Print settings
6673 * Value History:: Value history
6674 * Convenience Vars:: Convenience variables
6675 * Registers:: Registers
6676 * Floating Point Hardware:: Floating point hardware
6677 * Vector Unit:: Vector Unit
6678 * OS Information:: Auxiliary data provided by operating system
6679 * Memory Region Attributes:: Memory region attributes
6680 * Dump/Restore Files:: Copy between memory and a file
6681 * Core File Generation:: Cause a program dump its core
6682 * Character Sets:: Debugging programs that use a different
6683 character set than GDB does
6684 * Caching Remote Data:: Data caching for remote targets
6685 * Searching Memory:: Searching memory for a sequence of bytes
6689 @section Expressions
6692 @code{print} and many other @value{GDBN} commands accept an expression and
6693 compute its value. Any kind of constant, variable or operator defined
6694 by the programming language you are using is valid in an expression in
6695 @value{GDBN}. This includes conditional expressions, function calls,
6696 casts, and string constants. It also includes preprocessor macros, if
6697 you compiled your program to include this information; see
6700 @cindex arrays in expressions
6701 @value{GDBN} supports array constants in expressions input by
6702 the user. The syntax is @{@var{element}, @var{element}@dots{}@}. For example,
6703 you can use the command @code{print @{1, 2, 3@}} to create an array
6704 of three integers. If you pass an array to a function or assign it
6705 to a program variable, @value{GDBN} copies the array to memory that
6706 is @code{malloc}ed in the target program.
6708 Because C is so widespread, most of the expressions shown in examples in
6709 this manual are in C. @xref{Languages, , Using @value{GDBN} with Different
6710 Languages}, for information on how to use expressions in other
6713 In this section, we discuss operators that you can use in @value{GDBN}
6714 expressions regardless of your programming language.
6716 @cindex casts, in expressions
6717 Casts are supported in all languages, not just in C, because it is so
6718 useful to cast a number into a pointer in order to examine a structure
6719 at that address in memory.
6720 @c FIXME: casts supported---Mod2 true?
6722 @value{GDBN} supports these operators, in addition to those common
6723 to programming languages:
6727 @samp{@@} is a binary operator for treating parts of memory as arrays.
6728 @xref{Arrays, ,Artificial Arrays}, for more information.
6731 @samp{::} allows you to specify a variable in terms of the file or
6732 function where it is defined. @xref{Variables, ,Program Variables}.
6734 @cindex @{@var{type}@}
6735 @cindex type casting memory
6736 @cindex memory, viewing as typed object
6737 @cindex casts, to view memory
6738 @item @{@var{type}@} @var{addr}
6739 Refers to an object of type @var{type} stored at address @var{addr} in
6740 memory. @var{addr} may be any expression whose value is an integer or
6741 pointer (but parentheses are required around binary operators, just as in
6742 a cast). This construct is allowed regardless of what kind of data is
6743 normally supposed to reside at @var{addr}.
6746 @node Ambiguous Expressions
6747 @section Ambiguous Expressions
6748 @cindex ambiguous expressions
6750 Expressions can sometimes contain some ambiguous elements. For instance,
6751 some programming languages (notably Ada, C@t{++} and Objective-C) permit
6752 a single function name to be defined several times, for application in
6753 different contexts. This is called @dfn{overloading}. Another example
6754 involving Ada is generics. A @dfn{generic package} is similar to C@t{++}
6755 templates and is typically instantiated several times, resulting in
6756 the same function name being defined in different contexts.
6758 In some cases and depending on the language, it is possible to adjust
6759 the expression to remove the ambiguity. For instance in C@t{++}, you
6760 can specify the signature of the function you want to break on, as in
6761 @kbd{break @var{function}(@var{types})}. In Ada, using the fully
6762 qualified name of your function often makes the expression unambiguous
6765 When an ambiguity that needs to be resolved is detected, the debugger
6766 has the capability to display a menu of numbered choices for each
6767 possibility, and then waits for the selection with the prompt @samp{>}.
6768 The first option is always @samp{[0] cancel}, and typing @kbd{0 @key{RET}}
6769 aborts the current command. If the command in which the expression was
6770 used allows more than one choice to be selected, the next option in the
6771 menu is @samp{[1] all}, and typing @kbd{1 @key{RET}} selects all possible
6774 For example, the following session excerpt shows an attempt to set a
6775 breakpoint at the overloaded symbol @code{String::after}.
6776 We choose three particular definitions of that function name:
6778 @c FIXME! This is likely to change to show arg type lists, at least
6781 (@value{GDBP}) b String::after
6784 [2] file:String.cc; line number:867
6785 [3] file:String.cc; line number:860
6786 [4] file:String.cc; line number:875
6787 [5] file:String.cc; line number:853
6788 [6] file:String.cc; line number:846
6789 [7] file:String.cc; line number:735
6791 Breakpoint 1 at 0xb26c: file String.cc, line 867.
6792 Breakpoint 2 at 0xb344: file String.cc, line 875.
6793 Breakpoint 3 at 0xafcc: file String.cc, line 846.
6794 Multiple breakpoints were set.
6795 Use the "delete" command to delete unwanted
6802 @kindex set multiple-symbols
6803 @item set multiple-symbols @var{mode}
6804 @cindex multiple-symbols menu
6806 This option allows you to adjust the debugger behavior when an expression
6809 By default, @var{mode} is set to @code{all}. If the command with which
6810 the expression is used allows more than one choice, then @value{GDBN}
6811 automatically selects all possible choices. For instance, inserting
6812 a breakpoint on a function using an ambiguous name results in a breakpoint
6813 inserted on each possible match. However, if a unique choice must be made,
6814 then @value{GDBN} uses the menu to help you disambiguate the expression.
6815 For instance, printing the address of an overloaded function will result
6816 in the use of the menu.
6818 When @var{mode} is set to @code{ask}, the debugger always uses the menu
6819 when an ambiguity is detected.
6821 Finally, when @var{mode} is set to @code{cancel}, the debugger reports
6822 an error due to the ambiguity and the command is aborted.
6824 @kindex show multiple-symbols
6825 @item show multiple-symbols
6826 Show the current value of the @code{multiple-symbols} setting.
6830 @section Program Variables
6832 The most common kind of expression to use is the name of a variable
6835 Variables in expressions are understood in the selected stack frame
6836 (@pxref{Selection, ,Selecting a Frame}); they must be either:
6840 global (or file-static)
6847 visible according to the scope rules of the
6848 programming language from the point of execution in that frame
6851 @noindent This means that in the function
6866 you can examine and use the variable @code{a} whenever your program is
6867 executing within the function @code{foo}, but you can only use or
6868 examine the variable @code{b} while your program is executing inside
6869 the block where @code{b} is declared.
6871 @cindex variable name conflict
6872 There is an exception: you can refer to a variable or function whose
6873 scope is a single source file even if the current execution point is not
6874 in this file. But it is possible to have more than one such variable or
6875 function with the same name (in different source files). If that
6876 happens, referring to that name has unpredictable effects. If you wish,
6877 you can specify a static variable in a particular function or file,
6878 using the colon-colon (@code{::}) notation:
6880 @cindex colon-colon, context for variables/functions
6882 @c info cannot cope with a :: index entry, but why deprive hard copy readers?
6883 @cindex @code{::}, context for variables/functions
6886 @var{file}::@var{variable}
6887 @var{function}::@var{variable}
6891 Here @var{file} or @var{function} is the name of the context for the
6892 static @var{variable}. In the case of file names, you can use quotes to
6893 make sure @value{GDBN} parses the file name as a single word---for example,
6894 to print a global value of @code{x} defined in @file{f2.c}:
6897 (@value{GDBP}) p 'f2.c'::x
6900 @cindex C@t{++} scope resolution
6901 This use of @samp{::} is very rarely in conflict with the very similar
6902 use of the same notation in C@t{++}. @value{GDBN} also supports use of the C@t{++}
6903 scope resolution operator in @value{GDBN} expressions.
6904 @c FIXME: Um, so what happens in one of those rare cases where it's in
6907 @cindex wrong values
6908 @cindex variable values, wrong
6909 @cindex function entry/exit, wrong values of variables
6910 @cindex optimized code, wrong values of variables
6912 @emph{Warning:} Occasionally, a local variable may appear to have the
6913 wrong value at certain points in a function---just after entry to a new
6914 scope, and just before exit.
6916 You may see this problem when you are stepping by machine instructions.
6917 This is because, on most machines, it takes more than one instruction to
6918 set up a stack frame (including local variable definitions); if you are
6919 stepping by machine instructions, variables may appear to have the wrong
6920 values until the stack frame is completely built. On exit, it usually
6921 also takes more than one machine instruction to destroy a stack frame;
6922 after you begin stepping through that group of instructions, local
6923 variable definitions may be gone.
6925 This may also happen when the compiler does significant optimizations.
6926 To be sure of always seeing accurate values, turn off all optimization
6929 @cindex ``No symbol "foo" in current context''
6930 Another possible effect of compiler optimizations is to optimize
6931 unused variables out of existence, or assign variables to registers (as
6932 opposed to memory addresses). Depending on the support for such cases
6933 offered by the debug info format used by the compiler, @value{GDBN}
6934 might not be able to display values for such local variables. If that
6935 happens, @value{GDBN} will print a message like this:
6938 No symbol "foo" in current context.
6941 To solve such problems, either recompile without optimizations, or use a
6942 different debug info format, if the compiler supports several such
6943 formats. For example, @value{NGCC}, the @sc{gnu} C/C@t{++} compiler,
6944 usually supports the @option{-gstabs+} option. @option{-gstabs+}
6945 produces debug info in a format that is superior to formats such as
6946 COFF. You may be able to use DWARF 2 (@option{-gdwarf-2}), which is also
6947 an effective form for debug info. @xref{Debugging Options,,Options
6948 for Debugging Your Program or GCC, gcc.info, Using the @sc{gnu}
6949 Compiler Collection (GCC)}.
6950 @xref{C, ,C and C@t{++}}, for more information about debug info formats
6951 that are best suited to C@t{++} programs.
6953 If you ask to print an object whose contents are unknown to
6954 @value{GDBN}, e.g., because its data type is not completely specified
6955 by the debug information, @value{GDBN} will say @samp{<incomplete
6956 type>}. @xref{Symbols, incomplete type}, for more about this.
6958 Strings are identified as arrays of @code{char} values without specified
6959 signedness. Arrays of either @code{signed char} or @code{unsigned char} get
6960 printed as arrays of 1 byte sized integers. @code{-fsigned-char} or
6961 @code{-funsigned-char} @value{NGCC} options have no effect as @value{GDBN}
6962 defines literal string type @code{"char"} as @code{char} without a sign.
6967 signed char var1[] = "A";
6970 You get during debugging
6975 $2 = @{65 'A', 0 '\0'@}
6979 @section Artificial Arrays
6981 @cindex artificial array
6983 @kindex @@@r{, referencing memory as an array}
6984 It is often useful to print out several successive objects of the
6985 same type in memory; a section of an array, or an array of
6986 dynamically determined size for which only a pointer exists in the
6989 You can do this by referring to a contiguous span of memory as an
6990 @dfn{artificial array}, using the binary operator @samp{@@}. The left
6991 operand of @samp{@@} should be the first element of the desired array
6992 and be an individual object. The right operand should be the desired length
6993 of the array. The result is an array value whose elements are all of
6994 the type of the left argument. The first element is actually the left
6995 argument; the second element comes from bytes of memory immediately
6996 following those that hold the first element, and so on. Here is an
6997 example. If a program says
7000 int *array = (int *) malloc (len * sizeof (int));
7004 you can print the contents of @code{array} with
7010 The left operand of @samp{@@} must reside in memory. Array values made
7011 with @samp{@@} in this way behave just like other arrays in terms of
7012 subscripting, and are coerced to pointers when used in expressions.
7013 Artificial arrays most often appear in expressions via the value history
7014 (@pxref{Value History, ,Value History}), after printing one out.
7016 Another way to create an artificial array is to use a cast.
7017 This re-interprets a value as if it were an array.
7018 The value need not be in memory:
7020 (@value{GDBP}) p/x (short[2])0x12345678
7021 $1 = @{0x1234, 0x5678@}
7024 As a convenience, if you leave the array length out (as in
7025 @samp{(@var{type}[])@var{value}}) @value{GDBN} calculates the size to fill
7026 the value (as @samp{sizeof(@var{value})/sizeof(@var{type})}:
7028 (@value{GDBP}) p/x (short[])0x12345678
7029 $2 = @{0x1234, 0x5678@}
7032 Sometimes the artificial array mechanism is not quite enough; in
7033 moderately complex data structures, the elements of interest may not
7034 actually be adjacent---for example, if you are interested in the values
7035 of pointers in an array. One useful work-around in this situation is
7036 to use a convenience variable (@pxref{Convenience Vars, ,Convenience
7037 Variables}) as a counter in an expression that prints the first
7038 interesting value, and then repeat that expression via @key{RET}. For
7039 instance, suppose you have an array @code{dtab} of pointers to
7040 structures, and you are interested in the values of a field @code{fv}
7041 in each structure. Here is an example of what you might type:
7051 @node Output Formats
7052 @section Output Formats
7054 @cindex formatted output
7055 @cindex output formats
7056 By default, @value{GDBN} prints a value according to its data type. Sometimes
7057 this is not what you want. For example, you might want to print a number
7058 in hex, or a pointer in decimal. Or you might want to view data in memory
7059 at a certain address as a character string or as an instruction. To do
7060 these things, specify an @dfn{output format} when you print a value.
7062 The simplest use of output formats is to say how to print a value
7063 already computed. This is done by starting the arguments of the
7064 @code{print} command with a slash and a format letter. The format
7065 letters supported are:
7069 Regard the bits of the value as an integer, and print the integer in
7073 Print as integer in signed decimal.
7076 Print as integer in unsigned decimal.
7079 Print as integer in octal.
7082 Print as integer in binary. The letter @samp{t} stands for ``two''.
7083 @footnote{@samp{b} cannot be used because these format letters are also
7084 used with the @code{x} command, where @samp{b} stands for ``byte'';
7085 see @ref{Memory,,Examining Memory}.}
7088 @cindex unknown address, locating
7089 @cindex locate address
7090 Print as an address, both absolute in hexadecimal and as an offset from
7091 the nearest preceding symbol. You can use this format used to discover
7092 where (in what function) an unknown address is located:
7095 (@value{GDBP}) p/a 0x54320
7096 $3 = 0x54320 <_initialize_vx+396>
7100 The command @code{info symbol 0x54320} yields similar results.
7101 @xref{Symbols, info symbol}.
7104 Regard as an integer and print it as a character constant. This
7105 prints both the numerical value and its character representation. The
7106 character representation is replaced with the octal escape @samp{\nnn}
7107 for characters outside the 7-bit @sc{ascii} range.
7109 Without this format, @value{GDBN} displays @code{char},
7110 @w{@code{unsigned char}}, and @w{@code{signed char}} data as character
7111 constants. Single-byte members of vectors are displayed as integer
7115 Regard the bits of the value as a floating point number and print
7116 using typical floating point syntax.
7119 @cindex printing strings
7120 @cindex printing byte arrays
7121 Regard as a string, if possible. With this format, pointers to single-byte
7122 data are displayed as null-terminated strings and arrays of single-byte data
7123 are displayed as fixed-length strings. Other values are displayed in their
7126 Without this format, @value{GDBN} displays pointers to and arrays of
7127 @code{char}, @w{@code{unsigned char}}, and @w{@code{signed char}} as
7128 strings. Single-byte members of a vector are displayed as an integer
7132 @cindex raw printing
7133 Print using the @samp{raw} formatting. By default, @value{GDBN} will
7134 use a type-specific pretty-printer. The @samp{r} format bypasses any
7135 pretty-printer which might exist for the value's type.
7138 For example, to print the program counter in hex (@pxref{Registers}), type
7145 Note that no space is required before the slash; this is because command
7146 names in @value{GDBN} cannot contain a slash.
7148 To reprint the last value in the value history with a different format,
7149 you can use the @code{print} command with just a format and no
7150 expression. For example, @samp{p/x} reprints the last value in hex.
7153 @section Examining Memory
7155 You can use the command @code{x} (for ``examine'') to examine memory in
7156 any of several formats, independently of your program's data types.
7158 @cindex examining memory
7160 @kindex x @r{(examine memory)}
7161 @item x/@var{nfu} @var{addr}
7164 Use the @code{x} command to examine memory.
7167 @var{n}, @var{f}, and @var{u} are all optional parameters that specify how
7168 much memory to display and how to format it; @var{addr} is an
7169 expression giving the address where you want to start displaying memory.
7170 If you use defaults for @var{nfu}, you need not type the slash @samp{/}.
7171 Several commands set convenient defaults for @var{addr}.
7174 @item @var{n}, the repeat count
7175 The repeat count is a decimal integer; the default is 1. It specifies
7176 how much memory (counting by units @var{u}) to display.
7177 @c This really is **decimal**; unaffected by 'set radix' as of GDB
7180 @item @var{f}, the display format
7181 The display format is one of the formats used by @code{print}
7182 (@samp{x}, @samp{d}, @samp{u}, @samp{o}, @samp{t}, @samp{a}, @samp{c},
7183 @samp{f}, @samp{s}), and in addition @samp{i} (for machine instructions).
7184 The default is @samp{x} (hexadecimal) initially. The default changes
7185 each time you use either @code{x} or @code{print}.
7187 @item @var{u}, the unit size
7188 The unit size is any of
7194 Halfwords (two bytes).
7196 Words (four bytes). This is the initial default.
7198 Giant words (eight bytes).
7201 Each time you specify a unit size with @code{x}, that size becomes the
7202 default unit the next time you use @code{x}. (For the @samp{s} and
7203 @samp{i} formats, the unit size is ignored and is normally not written.)
7205 @item @var{addr}, starting display address
7206 @var{addr} is the address where you want @value{GDBN} to begin displaying
7207 memory. The expression need not have a pointer value (though it may);
7208 it is always interpreted as an integer address of a byte of memory.
7209 @xref{Expressions, ,Expressions}, for more information on expressions. The default for
7210 @var{addr} is usually just after the last address examined---but several
7211 other commands also set the default address: @code{info breakpoints} (to
7212 the address of the last breakpoint listed), @code{info line} (to the
7213 starting address of a line), and @code{print} (if you use it to display
7214 a value from memory).
7217 For example, @samp{x/3uh 0x54320} is a request to display three halfwords
7218 (@code{h}) of memory, formatted as unsigned decimal integers (@samp{u}),
7219 starting at address @code{0x54320}. @samp{x/4xw $sp} prints the four
7220 words (@samp{w}) of memory above the stack pointer (here, @samp{$sp};
7221 @pxref{Registers, ,Registers}) in hexadecimal (@samp{x}).
7223 Since the letters indicating unit sizes are all distinct from the
7224 letters specifying output formats, you do not have to remember whether
7225 unit size or format comes first; either order works. The output
7226 specifications @samp{4xw} and @samp{4wx} mean exactly the same thing.
7227 (However, the count @var{n} must come first; @samp{wx4} does not work.)
7229 Even though the unit size @var{u} is ignored for the formats @samp{s}
7230 and @samp{i}, you might still want to use a count @var{n}; for example,
7231 @samp{3i} specifies that you want to see three machine instructions,
7232 including any operands. For convenience, especially when used with
7233 the @code{display} command, the @samp{i} format also prints branch delay
7234 slot instructions, if any, beyond the count specified, which immediately
7235 follow the last instruction that is within the count. The command
7236 @code{disassemble} gives an alternative way of inspecting machine
7237 instructions; see @ref{Machine Code,,Source and Machine Code}.
7239 All the defaults for the arguments to @code{x} are designed to make it
7240 easy to continue scanning memory with minimal specifications each time
7241 you use @code{x}. For example, after you have inspected three machine
7242 instructions with @samp{x/3i @var{addr}}, you can inspect the next seven
7243 with just @samp{x/7}. If you use @key{RET} to repeat the @code{x} command,
7244 the repeat count @var{n} is used again; the other arguments default as
7245 for successive uses of @code{x}.
7247 @cindex @code{$_}, @code{$__}, and value history
7248 The addresses and contents printed by the @code{x} command are not saved
7249 in the value history because there is often too much of them and they
7250 would get in the way. Instead, @value{GDBN} makes these values available for
7251 subsequent use in expressions as values of the convenience variables
7252 @code{$_} and @code{$__}. After an @code{x} command, the last address
7253 examined is available for use in expressions in the convenience variable
7254 @code{$_}. The contents of that address, as examined, are available in
7255 the convenience variable @code{$__}.
7257 If the @code{x} command has a repeat count, the address and contents saved
7258 are from the last memory unit printed; this is not the same as the last
7259 address printed if several units were printed on the last line of output.
7261 @cindex remote memory comparison
7262 @cindex verify remote memory image
7263 When you are debugging a program running on a remote target machine
7264 (@pxref{Remote Debugging}), you may wish to verify the program's image in the
7265 remote machine's memory against the executable file you downloaded to
7266 the target. The @code{compare-sections} command is provided for such
7270 @kindex compare-sections
7271 @item compare-sections @r{[}@var{section-name}@r{]}
7272 Compare the data of a loadable section @var{section-name} in the
7273 executable file of the program being debugged with the same section in
7274 the remote machine's memory, and report any mismatches. With no
7275 arguments, compares all loadable sections. This command's
7276 availability depends on the target's support for the @code{"qCRC"}
7281 @section Automatic Display
7282 @cindex automatic display
7283 @cindex display of expressions
7285 If you find that you want to print the value of an expression frequently
7286 (to see how it changes), you might want to add it to the @dfn{automatic
7287 display list} so that @value{GDBN} prints its value each time your program stops.
7288 Each expression added to the list is given a number to identify it;
7289 to remove an expression from the list, you specify that number.
7290 The automatic display looks like this:
7294 3: bar[5] = (struct hack *) 0x3804
7298 This display shows item numbers, expressions and their current values. As with
7299 displays you request manually using @code{x} or @code{print}, you can
7300 specify the output format you prefer; in fact, @code{display} decides
7301 whether to use @code{print} or @code{x} depending your format
7302 specification---it uses @code{x} if you specify either the @samp{i}
7303 or @samp{s} format, or a unit size; otherwise it uses @code{print}.
7307 @item display @var{expr}
7308 Add the expression @var{expr} to the list of expressions to display
7309 each time your program stops. @xref{Expressions, ,Expressions}.
7311 @code{display} does not repeat if you press @key{RET} again after using it.
7313 @item display/@var{fmt} @var{expr}
7314 For @var{fmt} specifying only a display format and not a size or
7315 count, add the expression @var{expr} to the auto-display list but
7316 arrange to display it each time in the specified format @var{fmt}.
7317 @xref{Output Formats,,Output Formats}.
7319 @item display/@var{fmt} @var{addr}
7320 For @var{fmt} @samp{i} or @samp{s}, or including a unit-size or a
7321 number of units, add the expression @var{addr} as a memory address to
7322 be examined each time your program stops. Examining means in effect
7323 doing @samp{x/@var{fmt} @var{addr}}. @xref{Memory, ,Examining Memory}.
7326 For example, @samp{display/i $pc} can be helpful, to see the machine
7327 instruction about to be executed each time execution stops (@samp{$pc}
7328 is a common name for the program counter; @pxref{Registers, ,Registers}).
7331 @kindex delete display
7333 @item undisplay @var{dnums}@dots{}
7334 @itemx delete display @var{dnums}@dots{}
7335 Remove item numbers @var{dnums} from the list of expressions to display.
7337 @code{undisplay} does not repeat if you press @key{RET} after using it.
7338 (Otherwise you would just get the error @samp{No display number @dots{}}.)
7340 @kindex disable display
7341 @item disable display @var{dnums}@dots{}
7342 Disable the display of item numbers @var{dnums}. A disabled display
7343 item is not printed automatically, but is not forgotten. It may be
7344 enabled again later.
7346 @kindex enable display
7347 @item enable display @var{dnums}@dots{}
7348 Enable display of item numbers @var{dnums}. It becomes effective once
7349 again in auto display of its expression, until you specify otherwise.
7352 Display the current values of the expressions on the list, just as is
7353 done when your program stops.
7355 @kindex info display
7357 Print the list of expressions previously set up to display
7358 automatically, each one with its item number, but without showing the
7359 values. This includes disabled expressions, which are marked as such.
7360 It also includes expressions which would not be displayed right now
7361 because they refer to automatic variables not currently available.
7364 @cindex display disabled out of scope
7365 If a display expression refers to local variables, then it does not make
7366 sense outside the lexical context for which it was set up. Such an
7367 expression is disabled when execution enters a context where one of its
7368 variables is not defined. For example, if you give the command
7369 @code{display last_char} while inside a function with an argument
7370 @code{last_char}, @value{GDBN} displays this argument while your program
7371 continues to stop inside that function. When it stops elsewhere---where
7372 there is no variable @code{last_char}---the display is disabled
7373 automatically. The next time your program stops where @code{last_char}
7374 is meaningful, you can enable the display expression once again.
7376 @node Print Settings
7377 @section Print Settings
7379 @cindex format options
7380 @cindex print settings
7381 @value{GDBN} provides the following ways to control how arrays, structures,
7382 and symbols are printed.
7385 These settings are useful for debugging programs in any language:
7389 @item set print address
7390 @itemx set print address on
7391 @cindex print/don't print memory addresses
7392 @value{GDBN} prints memory addresses showing the location of stack
7393 traces, structure values, pointer values, breakpoints, and so forth,
7394 even when it also displays the contents of those addresses. The default
7395 is @code{on}. For example, this is what a stack frame display looks like with
7396 @code{set print address on}:
7401 #0 set_quotes (lq=0x34c78 "<<", rq=0x34c88 ">>")
7403 530 if (lquote != def_lquote)
7407 @item set print address off
7408 Do not print addresses when displaying their contents. For example,
7409 this is the same stack frame displayed with @code{set print address off}:
7413 (@value{GDBP}) set print addr off
7415 #0 set_quotes (lq="<<", rq=">>") at input.c:530
7416 530 if (lquote != def_lquote)
7420 You can use @samp{set print address off} to eliminate all machine
7421 dependent displays from the @value{GDBN} interface. For example, with
7422 @code{print address off}, you should get the same text for backtraces on
7423 all machines---whether or not they involve pointer arguments.
7426 @item show print address
7427 Show whether or not addresses are to be printed.
7430 When @value{GDBN} prints a symbolic address, it normally prints the
7431 closest earlier symbol plus an offset. If that symbol does not uniquely
7432 identify the address (for example, it is a name whose scope is a single
7433 source file), you may need to clarify. One way to do this is with
7434 @code{info line}, for example @samp{info line *0x4537}. Alternately,
7435 you can set @value{GDBN} to print the source file and line number when
7436 it prints a symbolic address:
7439 @item set print symbol-filename on
7440 @cindex source file and line of a symbol
7441 @cindex symbol, source file and line
7442 Tell @value{GDBN} to print the source file name and line number of a
7443 symbol in the symbolic form of an address.
7445 @item set print symbol-filename off
7446 Do not print source file name and line number of a symbol. This is the
7449 @item show print symbol-filename
7450 Show whether or not @value{GDBN} will print the source file name and
7451 line number of a symbol in the symbolic form of an address.
7454 Another situation where it is helpful to show symbol filenames and line
7455 numbers is when disassembling code; @value{GDBN} shows you the line
7456 number and source file that corresponds to each instruction.
7458 Also, you may wish to see the symbolic form only if the address being
7459 printed is reasonably close to the closest earlier symbol:
7462 @item set print max-symbolic-offset @var{max-offset}
7463 @cindex maximum value for offset of closest symbol
7464 Tell @value{GDBN} to only display the symbolic form of an address if the
7465 offset between the closest earlier symbol and the address is less than
7466 @var{max-offset}. The default is 0, which tells @value{GDBN}
7467 to always print the symbolic form of an address if any symbol precedes it.
7469 @item show print max-symbolic-offset
7470 Ask how large the maximum offset is that @value{GDBN} prints in a
7474 @cindex wild pointer, interpreting
7475 @cindex pointer, finding referent
7476 If you have a pointer and you are not sure where it points, try
7477 @samp{set print symbol-filename on}. Then you can determine the name
7478 and source file location of the variable where it points, using
7479 @samp{p/a @var{pointer}}. This interprets the address in symbolic form.
7480 For example, here @value{GDBN} shows that a variable @code{ptt} points
7481 at another variable @code{t}, defined in @file{hi2.c}:
7484 (@value{GDBP}) set print symbol-filename on
7485 (@value{GDBP}) p/a ptt
7486 $4 = 0xe008 <t in hi2.c>
7490 @emph{Warning:} For pointers that point to a local variable, @samp{p/a}
7491 does not show the symbol name and filename of the referent, even with
7492 the appropriate @code{set print} options turned on.
7495 Other settings control how different kinds of objects are printed:
7498 @item set print array
7499 @itemx set print array on
7500 @cindex pretty print arrays
7501 Pretty print arrays. This format is more convenient to read,
7502 but uses more space. The default is off.
7504 @item set print array off
7505 Return to compressed format for arrays.
7507 @item show print array
7508 Show whether compressed or pretty format is selected for displaying
7511 @cindex print array indexes
7512 @item set print array-indexes
7513 @itemx set print array-indexes on
7514 Print the index of each element when displaying arrays. May be more
7515 convenient to locate a given element in the array or quickly find the
7516 index of a given element in that printed array. The default is off.
7518 @item set print array-indexes off
7519 Stop printing element indexes when displaying arrays.
7521 @item show print array-indexes
7522 Show whether the index of each element is printed when displaying
7525 @item set print elements @var{number-of-elements}
7526 @cindex number of array elements to print
7527 @cindex limit on number of printed array elements
7528 Set a limit on how many elements of an array @value{GDBN} will print.
7529 If @value{GDBN} is printing a large array, it stops printing after it has
7530 printed the number of elements set by the @code{set print elements} command.
7531 This limit also applies to the display of strings.
7532 When @value{GDBN} starts, this limit is set to 200.
7533 Setting @var{number-of-elements} to zero means that the printing is unlimited.
7535 @item show print elements
7536 Display the number of elements of a large array that @value{GDBN} will print.
7537 If the number is 0, then the printing is unlimited.
7539 @item set print frame-arguments @var{value}
7540 @kindex set print frame-arguments
7541 @cindex printing frame argument values
7542 @cindex print all frame argument values
7543 @cindex print frame argument values for scalars only
7544 @cindex do not print frame argument values
7545 This command allows to control how the values of arguments are printed
7546 when the debugger prints a frame (@pxref{Frames}). The possible
7551 The values of all arguments are printed.
7554 Print the value of an argument only if it is a scalar. The value of more
7555 complex arguments such as arrays, structures, unions, etc, is replaced
7556 by @code{@dots{}}. This is the default. Here is an example where
7557 only scalar arguments are shown:
7560 #1 0x08048361 in call_me (i=3, s=@dots{}, ss=0xbf8d508c, u=@dots{}, e=green)
7565 None of the argument values are printed. Instead, the value of each argument
7566 is replaced by @code{@dots{}}. In this case, the example above now becomes:
7569 #1 0x08048361 in call_me (i=@dots{}, s=@dots{}, ss=@dots{}, u=@dots{}, e=@dots{})
7574 By default, only scalar arguments are printed. This command can be used
7575 to configure the debugger to print the value of all arguments, regardless
7576 of their type. However, it is often advantageous to not print the value
7577 of more complex parameters. For instance, it reduces the amount of
7578 information printed in each frame, making the backtrace more readable.
7579 Also, it improves performance when displaying Ada frames, because
7580 the computation of large arguments can sometimes be CPU-intensive,
7581 especially in large applications. Setting @code{print frame-arguments}
7582 to @code{scalars} (the default) or @code{none} avoids this computation,
7583 thus speeding up the display of each Ada frame.
7585 @item show print frame-arguments
7586 Show how the value of arguments should be displayed when printing a frame.
7588 @item set print repeats
7589 @cindex repeated array elements
7590 Set the threshold for suppressing display of repeated array
7591 elements. When the number of consecutive identical elements of an
7592 array exceeds the threshold, @value{GDBN} prints the string
7593 @code{"<repeats @var{n} times>"}, where @var{n} is the number of
7594 identical repetitions, instead of displaying the identical elements
7595 themselves. Setting the threshold to zero will cause all elements to
7596 be individually printed. The default threshold is 10.
7598 @item show print repeats
7599 Display the current threshold for printing repeated identical
7602 @item set print null-stop
7603 @cindex @sc{null} elements in arrays
7604 Cause @value{GDBN} to stop printing the characters of an array when the first
7605 @sc{null} is encountered. This is useful when large arrays actually
7606 contain only short strings.
7609 @item show print null-stop
7610 Show whether @value{GDBN} stops printing an array on the first
7611 @sc{null} character.
7613 @item set print pretty on
7614 @cindex print structures in indented form
7615 @cindex indentation in structure display
7616 Cause @value{GDBN} to print structures in an indented format with one member
7617 per line, like this:
7632 @item set print pretty off
7633 Cause @value{GDBN} to print structures in a compact format, like this:
7637 $1 = @{next = 0x0, flags = @{sweet = 1, sour = 1@}, \
7638 meat = 0x54 "Pork"@}
7643 This is the default format.
7645 @item show print pretty
7646 Show which format @value{GDBN} is using to print structures.
7648 @item set print sevenbit-strings on
7649 @cindex eight-bit characters in strings
7650 @cindex octal escapes in strings
7651 Print using only seven-bit characters; if this option is set,
7652 @value{GDBN} displays any eight-bit characters (in strings or
7653 character values) using the notation @code{\}@var{nnn}. This setting is
7654 best if you are working in English (@sc{ascii}) and you use the
7655 high-order bit of characters as a marker or ``meta'' bit.
7657 @item set print sevenbit-strings off
7658 Print full eight-bit characters. This allows the use of more
7659 international character sets, and is the default.
7661 @item show print sevenbit-strings
7662 Show whether or not @value{GDBN} is printing only seven-bit characters.
7664 @item set print union on
7665 @cindex unions in structures, printing
7666 Tell @value{GDBN} to print unions which are contained in structures
7667 and other unions. This is the default setting.
7669 @item set print union off
7670 Tell @value{GDBN} not to print unions which are contained in
7671 structures and other unions. @value{GDBN} will print @code{"@{...@}"}
7674 @item show print union
7675 Ask @value{GDBN} whether or not it will print unions which are contained in
7676 structures and other unions.
7678 For example, given the declarations
7681 typedef enum @{Tree, Bug@} Species;
7682 typedef enum @{Big_tree, Acorn, Seedling@} Tree_forms;
7683 typedef enum @{Caterpillar, Cocoon, Butterfly@}
7694 struct thing foo = @{Tree, @{Acorn@}@};
7698 with @code{set print union on} in effect @samp{p foo} would print
7701 $1 = @{it = Tree, form = @{tree = Acorn, bug = Cocoon@}@}
7705 and with @code{set print union off} in effect it would print
7708 $1 = @{it = Tree, form = @{...@}@}
7712 @code{set print union} affects programs written in C-like languages
7718 These settings are of interest when debugging C@t{++} programs:
7721 @cindex demangling C@t{++} names
7722 @item set print demangle
7723 @itemx set print demangle on
7724 Print C@t{++} names in their source form rather than in the encoded
7725 (``mangled'') form passed to the assembler and linker for type-safe
7726 linkage. The default is on.
7728 @item show print demangle
7729 Show whether C@t{++} names are printed in mangled or demangled form.
7731 @item set print asm-demangle
7732 @itemx set print asm-demangle on
7733 Print C@t{++} names in their source form rather than their mangled form, even
7734 in assembler code printouts such as instruction disassemblies.
7737 @item show print asm-demangle
7738 Show whether C@t{++} names in assembly listings are printed in mangled
7741 @cindex C@t{++} symbol decoding style
7742 @cindex symbol decoding style, C@t{++}
7743 @kindex set demangle-style
7744 @item set demangle-style @var{style}
7745 Choose among several encoding schemes used by different compilers to
7746 represent C@t{++} names. The choices for @var{style} are currently:
7750 Allow @value{GDBN} to choose a decoding style by inspecting your program.
7753 Decode based on the @sc{gnu} C@t{++} compiler (@code{g++}) encoding algorithm.
7754 This is the default.
7757 Decode based on the HP ANSI C@t{++} (@code{aCC}) encoding algorithm.
7760 Decode based on the Lucid C@t{++} compiler (@code{lcc}) encoding algorithm.
7763 Decode using the algorithm in the @cite{C@t{++} Annotated Reference Manual}.
7764 @strong{Warning:} this setting alone is not sufficient to allow
7765 debugging @code{cfront}-generated executables. @value{GDBN} would
7766 require further enhancement to permit that.
7769 If you omit @var{style}, you will see a list of possible formats.
7771 @item show demangle-style
7772 Display the encoding style currently in use for decoding C@t{++} symbols.
7774 @item set print object
7775 @itemx set print object on
7776 @cindex derived type of an object, printing
7777 @cindex display derived types
7778 When displaying a pointer to an object, identify the @emph{actual}
7779 (derived) type of the object rather than the @emph{declared} type, using
7780 the virtual function table.
7782 @item set print object off
7783 Display only the declared type of objects, without reference to the
7784 virtual function table. This is the default setting.
7786 @item show print object
7787 Show whether actual, or declared, object types are displayed.
7789 @item set print static-members
7790 @itemx set print static-members on
7791 @cindex static members of C@t{++} objects
7792 Print static members when displaying a C@t{++} object. The default is on.
7794 @item set print static-members off
7795 Do not print static members when displaying a C@t{++} object.
7797 @item show print static-members
7798 Show whether C@t{++} static members are printed or not.
7800 @item set print pascal_static-members
7801 @itemx set print pascal_static-members on
7802 @cindex static members of Pascal objects
7803 @cindex Pascal objects, static members display
7804 Print static members when displaying a Pascal object. The default is on.
7806 @item set print pascal_static-members off
7807 Do not print static members when displaying a Pascal object.
7809 @item show print pascal_static-members
7810 Show whether Pascal static members are printed or not.
7812 @c These don't work with HP ANSI C++ yet.
7813 @item set print vtbl
7814 @itemx set print vtbl on
7815 @cindex pretty print C@t{++} virtual function tables
7816 @cindex virtual functions (C@t{++}) display
7817 @cindex VTBL display
7818 Pretty print C@t{++} virtual function tables. The default is off.
7819 (The @code{vtbl} commands do not work on programs compiled with the HP
7820 ANSI C@t{++} compiler (@code{aCC}).)
7822 @item set print vtbl off
7823 Do not pretty print C@t{++} virtual function tables.
7825 @item show print vtbl
7826 Show whether C@t{++} virtual function tables are pretty printed, or not.
7830 @section Value History
7832 @cindex value history
7833 @cindex history of values printed by @value{GDBN}
7834 Values printed by the @code{print} command are saved in the @value{GDBN}
7835 @dfn{value history}. This allows you to refer to them in other expressions.
7836 Values are kept until the symbol table is re-read or discarded
7837 (for example with the @code{file} or @code{symbol-file} commands).
7838 When the symbol table changes, the value history is discarded,
7839 since the values may contain pointers back to the types defined in the
7844 @cindex history number
7845 The values printed are given @dfn{history numbers} by which you can
7846 refer to them. These are successive integers starting with one.
7847 @code{print} shows you the history number assigned to a value by
7848 printing @samp{$@var{num} = } before the value; here @var{num} is the
7851 To refer to any previous value, use @samp{$} followed by the value's
7852 history number. The way @code{print} labels its output is designed to
7853 remind you of this. Just @code{$} refers to the most recent value in
7854 the history, and @code{$$} refers to the value before that.
7855 @code{$$@var{n}} refers to the @var{n}th value from the end; @code{$$2}
7856 is the value just prior to @code{$$}, @code{$$1} is equivalent to
7857 @code{$$}, and @code{$$0} is equivalent to @code{$}.
7859 For example, suppose you have just printed a pointer to a structure and
7860 want to see the contents of the structure. It suffices to type
7866 If you have a chain of structures where the component @code{next} points
7867 to the next one, you can print the contents of the next one with this:
7874 You can print successive links in the chain by repeating this
7875 command---which you can do by just typing @key{RET}.
7877 Note that the history records values, not expressions. If the value of
7878 @code{x} is 4 and you type these commands:
7886 then the value recorded in the value history by the @code{print} command
7887 remains 4 even though the value of @code{x} has changed.
7892 Print the last ten values in the value history, with their item numbers.
7893 This is like @samp{p@ $$9} repeated ten times, except that @code{show
7894 values} does not change the history.
7896 @item show values @var{n}
7897 Print ten history values centered on history item number @var{n}.
7900 Print ten history values just after the values last printed. If no more
7901 values are available, @code{show values +} produces no display.
7904 Pressing @key{RET} to repeat @code{show values @var{n}} has exactly the
7905 same effect as @samp{show values +}.
7907 @node Convenience Vars
7908 @section Convenience Variables
7910 @cindex convenience variables
7911 @cindex user-defined variables
7912 @value{GDBN} provides @dfn{convenience variables} that you can use within
7913 @value{GDBN} to hold on to a value and refer to it later. These variables
7914 exist entirely within @value{GDBN}; they are not part of your program, and
7915 setting a convenience variable has no direct effect on further execution
7916 of your program. That is why you can use them freely.
7918 Convenience variables are prefixed with @samp{$}. Any name preceded by
7919 @samp{$} can be used for a convenience variable, unless it is one of
7920 the predefined machine-specific register names (@pxref{Registers, ,Registers}).
7921 (Value history references, in contrast, are @emph{numbers} preceded
7922 by @samp{$}. @xref{Value History, ,Value History}.)
7924 You can save a value in a convenience variable with an assignment
7925 expression, just as you would set a variable in your program.
7929 set $foo = *object_ptr
7933 would save in @code{$foo} the value contained in the object pointed to by
7936 Using a convenience variable for the first time creates it, but its
7937 value is @code{void} until you assign a new value. You can alter the
7938 value with another assignment at any time.
7940 Convenience variables have no fixed types. You can assign a convenience
7941 variable any type of value, including structures and arrays, even if
7942 that variable already has a value of a different type. The convenience
7943 variable, when used as an expression, has the type of its current value.
7946 @kindex show convenience
7947 @cindex show all user variables
7948 @item show convenience
7949 Print a list of convenience variables used so far, and their values.
7950 Abbreviated @code{show conv}.
7952 @kindex init-if-undefined
7953 @cindex convenience variables, initializing
7954 @item init-if-undefined $@var{variable} = @var{expression}
7955 Set a convenience variable if it has not already been set. This is useful
7956 for user-defined commands that keep some state. It is similar, in concept,
7957 to using local static variables with initializers in C (except that
7958 convenience variables are global). It can also be used to allow users to
7959 override default values used in a command script.
7961 If the variable is already defined then the expression is not evaluated so
7962 any side-effects do not occur.
7965 One of the ways to use a convenience variable is as a counter to be
7966 incremented or a pointer to be advanced. For example, to print
7967 a field from successive elements of an array of structures:
7971 print bar[$i++]->contents
7975 Repeat that command by typing @key{RET}.
7977 Some convenience variables are created automatically by @value{GDBN} and given
7978 values likely to be useful.
7981 @vindex $_@r{, convenience variable}
7983 The variable @code{$_} is automatically set by the @code{x} command to
7984 the last address examined (@pxref{Memory, ,Examining Memory}). Other
7985 commands which provide a default address for @code{x} to examine also
7986 set @code{$_} to that address; these commands include @code{info line}
7987 and @code{info breakpoint}. The type of @code{$_} is @code{void *}
7988 except when set by the @code{x} command, in which case it is a pointer
7989 to the type of @code{$__}.
7991 @vindex $__@r{, convenience variable}
7993 The variable @code{$__} is automatically set by the @code{x} command
7994 to the value found in the last address examined. Its type is chosen
7995 to match the format in which the data was printed.
7998 @vindex $_exitcode@r{, convenience variable}
7999 The variable @code{$_exitcode} is automatically set to the exit code when
8000 the program being debugged terminates.
8003 @vindex $_siginfo@r{, convenience variable}
8004 The variable @code{$_siginfo} contains extra signal information
8005 (@pxref{extra signal information}). Note that @code{$_siginfo}
8006 could be empty, if the application has not yet received any signals.
8007 For example, it will be empty before you execute the @code{run} command.
8010 On HP-UX systems, if you refer to a function or variable name that
8011 begins with a dollar sign, @value{GDBN} searches for a user or system
8012 name first, before it searches for a convenience variable.
8014 @cindex convenience functions
8015 @value{GDBN} also supplies some @dfn{convenience functions}. These
8016 have a syntax similar to convenience variables. A convenience
8017 function can be used in an expression just like an ordinary function;
8018 however, a convenience function is implemented internally to
8023 @kindex help function
8024 @cindex show all convenience functions
8025 Print a list of all convenience functions.
8032 You can refer to machine register contents, in expressions, as variables
8033 with names starting with @samp{$}. The names of registers are different
8034 for each machine; use @code{info registers} to see the names used on
8038 @kindex info registers
8039 @item info registers
8040 Print the names and values of all registers except floating-point
8041 and vector registers (in the selected stack frame).
8043 @kindex info all-registers
8044 @cindex floating point registers
8045 @item info all-registers
8046 Print the names and values of all registers, including floating-point
8047 and vector registers (in the selected stack frame).
8049 @item info registers @var{regname} @dots{}
8050 Print the @dfn{relativized} value of each specified register @var{regname}.
8051 As discussed in detail below, register values are normally relative to
8052 the selected stack frame. @var{regname} may be any register name valid on
8053 the machine you are using, with or without the initial @samp{$}.
8056 @cindex stack pointer register
8057 @cindex program counter register
8058 @cindex process status register
8059 @cindex frame pointer register
8060 @cindex standard registers
8061 @value{GDBN} has four ``standard'' register names that are available (in
8062 expressions) on most machines---whenever they do not conflict with an
8063 architecture's canonical mnemonics for registers. The register names
8064 @code{$pc} and @code{$sp} are used for the program counter register and
8065 the stack pointer. @code{$fp} is used for a register that contains a
8066 pointer to the current stack frame, and @code{$ps} is used for a
8067 register that contains the processor status. For example,
8068 you could print the program counter in hex with
8075 or print the instruction to be executed next with
8082 or add four to the stack pointer@footnote{This is a way of removing
8083 one word from the stack, on machines where stacks grow downward in
8084 memory (most machines, nowadays). This assumes that the innermost
8085 stack frame is selected; setting @code{$sp} is not allowed when other
8086 stack frames are selected. To pop entire frames off the stack,
8087 regardless of machine architecture, use @code{return};
8088 see @ref{Returning, ,Returning from a Function}.} with
8094 Whenever possible, these four standard register names are available on
8095 your machine even though the machine has different canonical mnemonics,
8096 so long as there is no conflict. The @code{info registers} command
8097 shows the canonical names. For example, on the SPARC, @code{info
8098 registers} displays the processor status register as @code{$psr} but you
8099 can also refer to it as @code{$ps}; and on x86-based machines @code{$ps}
8100 is an alias for the @sc{eflags} register.
8102 @value{GDBN} always considers the contents of an ordinary register as an
8103 integer when the register is examined in this way. Some machines have
8104 special registers which can hold nothing but floating point; these
8105 registers are considered to have floating point values. There is no way
8106 to refer to the contents of an ordinary register as floating point value
8107 (although you can @emph{print} it as a floating point value with
8108 @samp{print/f $@var{regname}}).
8110 Some registers have distinct ``raw'' and ``virtual'' data formats. This
8111 means that the data format in which the register contents are saved by
8112 the operating system is not the same one that your program normally
8113 sees. For example, the registers of the 68881 floating point
8114 coprocessor are always saved in ``extended'' (raw) format, but all C
8115 programs expect to work with ``double'' (virtual) format. In such
8116 cases, @value{GDBN} normally works with the virtual format only (the format
8117 that makes sense for your program), but the @code{info registers} command
8118 prints the data in both formats.
8120 @cindex SSE registers (x86)
8121 @cindex MMX registers (x86)
8122 Some machines have special registers whose contents can be interpreted
8123 in several different ways. For example, modern x86-based machines
8124 have SSE and MMX registers that can hold several values packed
8125 together in several different formats. @value{GDBN} refers to such
8126 registers in @code{struct} notation:
8129 (@value{GDBP}) print $xmm1
8131 v4_float = @{0, 3.43859137e-038, 1.54142831e-044, 1.821688e-044@},
8132 v2_double = @{9.92129282474342e-303, 2.7585945287983262e-313@},
8133 v16_int8 = "\000\000\000\000\3706;\001\v\000\000\000\r\000\000",
8134 v8_int16 = @{0, 0, 14072, 315, 11, 0, 13, 0@},
8135 v4_int32 = @{0, 20657912, 11, 13@},
8136 v2_int64 = @{88725056443645952, 55834574859@},
8137 uint128 = 0x0000000d0000000b013b36f800000000
8142 To set values of such registers, you need to tell @value{GDBN} which
8143 view of the register you wish to change, as if you were assigning
8144 value to a @code{struct} member:
8147 (@value{GDBP}) set $xmm1.uint128 = 0x000000000000000000000000FFFFFFFF
8150 Normally, register values are relative to the selected stack frame
8151 (@pxref{Selection, ,Selecting a Frame}). This means that you get the
8152 value that the register would contain if all stack frames farther in
8153 were exited and their saved registers restored. In order to see the
8154 true contents of hardware registers, you must select the innermost
8155 frame (with @samp{frame 0}).
8157 However, @value{GDBN} must deduce where registers are saved, from the machine
8158 code generated by your compiler. If some registers are not saved, or if
8159 @value{GDBN} is unable to locate the saved registers, the selected stack
8160 frame makes no difference.
8162 @node Floating Point Hardware
8163 @section Floating Point Hardware
8164 @cindex floating point
8166 Depending on the configuration, @value{GDBN} may be able to give
8167 you more information about the status of the floating point hardware.
8172 Display hardware-dependent information about the floating
8173 point unit. The exact contents and layout vary depending on the
8174 floating point chip. Currently, @samp{info float} is supported on
8175 the ARM and x86 machines.
8179 @section Vector Unit
8182 Depending on the configuration, @value{GDBN} may be able to give you
8183 more information about the status of the vector unit.
8188 Display information about the vector unit. The exact contents and
8189 layout vary depending on the hardware.
8192 @node OS Information
8193 @section Operating System Auxiliary Information
8194 @cindex OS information
8196 @value{GDBN} provides interfaces to useful OS facilities that can help
8197 you debug your program.
8199 @cindex @code{ptrace} system call
8200 @cindex @code{struct user} contents
8201 When @value{GDBN} runs on a @dfn{Posix system} (such as GNU or Unix
8202 machines), it interfaces with the inferior via the @code{ptrace}
8203 system call. The operating system creates a special sata structure,
8204 called @code{struct user}, for this interface. You can use the
8205 command @code{info udot} to display the contents of this data
8211 Display the contents of the @code{struct user} maintained by the OS
8212 kernel for the program being debugged. @value{GDBN} displays the
8213 contents of @code{struct user} as a list of hex numbers, similar to
8214 the @code{examine} command.
8217 @cindex auxiliary vector
8218 @cindex vector, auxiliary
8219 Some operating systems supply an @dfn{auxiliary vector} to programs at
8220 startup. This is akin to the arguments and environment that you
8221 specify for a program, but contains a system-dependent variety of
8222 binary values that tell system libraries important details about the
8223 hardware, operating system, and process. Each value's purpose is
8224 identified by an integer tag; the meanings are well-known but system-specific.
8225 Depending on the configuration and operating system facilities,
8226 @value{GDBN} may be able to show you this information. For remote
8227 targets, this functionality may further depend on the remote stub's
8228 support of the @samp{qXfer:auxv:read} packet, see
8229 @ref{qXfer auxiliary vector read}.
8234 Display the auxiliary vector of the inferior, which can be either a
8235 live process or a core dump file. @value{GDBN} prints each tag value
8236 numerically, and also shows names and text descriptions for recognized
8237 tags. Some values in the vector are numbers, some bit masks, and some
8238 pointers to strings or other data. @value{GDBN} displays each value in the
8239 most appropriate form for a recognized tag, and in hexadecimal for
8240 an unrecognized tag.
8243 On some targets, @value{GDBN} can access operating-system-specific information
8244 and display it to user, without interpretation. For remote targets,
8245 this functionality depends on the remote stub's support of the
8246 @samp{qXfer:osdata:read} packet, see @ref{qXfer osdata read}.
8249 @kindex info os processes
8250 @item info os processes
8251 Display the list of processes on the target. For each process,
8252 @value{GDBN} prints the process identifier, the name of the user, and
8253 the command corresponding to the process.
8256 @node Memory Region Attributes
8257 @section Memory Region Attributes
8258 @cindex memory region attributes
8260 @dfn{Memory region attributes} allow you to describe special handling
8261 required by regions of your target's memory. @value{GDBN} uses
8262 attributes to determine whether to allow certain types of memory
8263 accesses; whether to use specific width accesses; and whether to cache
8264 target memory. By default the description of memory regions is
8265 fetched from the target (if the current target supports this), but the
8266 user can override the fetched regions.
8268 Defined memory regions can be individually enabled and disabled. When a
8269 memory region is disabled, @value{GDBN} uses the default attributes when
8270 accessing memory in that region. Similarly, if no memory regions have
8271 been defined, @value{GDBN} uses the default attributes when accessing
8274 When a memory region is defined, it is given a number to identify it;
8275 to enable, disable, or remove a memory region, you specify that number.
8279 @item mem @var{lower} @var{upper} @var{attributes}@dots{}
8280 Define a memory region bounded by @var{lower} and @var{upper} with
8281 attributes @var{attributes}@dots{}, and add it to the list of regions
8282 monitored by @value{GDBN}. Note that @var{upper} == 0 is a special
8283 case: it is treated as the target's maximum memory address.
8284 (0xffff on 16 bit targets, 0xffffffff on 32 bit targets, etc.)
8287 Discard any user changes to the memory regions and use target-supplied
8288 regions, if available, or no regions if the target does not support.
8291 @item delete mem @var{nums}@dots{}
8292 Remove memory regions @var{nums}@dots{} from the list of regions
8293 monitored by @value{GDBN}.
8296 @item disable mem @var{nums}@dots{}
8297 Disable monitoring of memory regions @var{nums}@dots{}.
8298 A disabled memory region is not forgotten.
8299 It may be enabled again later.
8302 @item enable mem @var{nums}@dots{}
8303 Enable monitoring of memory regions @var{nums}@dots{}.
8307 Print a table of all defined memory regions, with the following columns
8311 @item Memory Region Number
8312 @item Enabled or Disabled.
8313 Enabled memory regions are marked with @samp{y}.
8314 Disabled memory regions are marked with @samp{n}.
8317 The address defining the inclusive lower bound of the memory region.
8320 The address defining the exclusive upper bound of the memory region.
8323 The list of attributes set for this memory region.
8328 @subsection Attributes
8330 @subsubsection Memory Access Mode
8331 The access mode attributes set whether @value{GDBN} may make read or
8332 write accesses to a memory region.
8334 While these attributes prevent @value{GDBN} from performing invalid
8335 memory accesses, they do nothing to prevent the target system, I/O DMA,
8336 etc.@: from accessing memory.
8340 Memory is read only.
8342 Memory is write only.
8344 Memory is read/write. This is the default.
8347 @subsubsection Memory Access Size
8348 The access size attribute tells @value{GDBN} to use specific sized
8349 accesses in the memory region. Often memory mapped device registers
8350 require specific sized accesses. If no access size attribute is
8351 specified, @value{GDBN} may use accesses of any size.
8355 Use 8 bit memory accesses.
8357 Use 16 bit memory accesses.
8359 Use 32 bit memory accesses.
8361 Use 64 bit memory accesses.
8364 @c @subsubsection Hardware/Software Breakpoints
8365 @c The hardware/software breakpoint attributes set whether @value{GDBN}
8366 @c will use hardware or software breakpoints for the internal breakpoints
8367 @c used by the step, next, finish, until, etc. commands.
8371 @c Always use hardware breakpoints
8372 @c @item swbreak (default)
8375 @subsubsection Data Cache
8376 The data cache attributes set whether @value{GDBN} will cache target
8377 memory. While this generally improves performance by reducing debug
8378 protocol overhead, it can lead to incorrect results because @value{GDBN}
8379 does not know about volatile variables or memory mapped device
8384 Enable @value{GDBN} to cache target memory.
8386 Disable @value{GDBN} from caching target memory. This is the default.
8389 @subsection Memory Access Checking
8390 @value{GDBN} can be instructed to refuse accesses to memory that is
8391 not explicitly described. This can be useful if accessing such
8392 regions has undesired effects for a specific target, or to provide
8393 better error checking. The following commands control this behaviour.
8396 @kindex set mem inaccessible-by-default
8397 @item set mem inaccessible-by-default [on|off]
8398 If @code{on} is specified, make @value{GDBN} treat memory not
8399 explicitly described by the memory ranges as non-existent and refuse accesses
8400 to such memory. The checks are only performed if there's at least one
8401 memory range defined. If @code{off} is specified, make @value{GDBN}
8402 treat the memory not explicitly described by the memory ranges as RAM.
8403 The default value is @code{on}.
8404 @kindex show mem inaccessible-by-default
8405 @item show mem inaccessible-by-default
8406 Show the current handling of accesses to unknown memory.
8410 @c @subsubsection Memory Write Verification
8411 @c The memory write verification attributes set whether @value{GDBN}
8412 @c will re-reads data after each write to verify the write was successful.
8416 @c @item noverify (default)
8419 @node Dump/Restore Files
8420 @section Copy Between Memory and a File
8421 @cindex dump/restore files
8422 @cindex append data to a file
8423 @cindex dump data to a file
8424 @cindex restore data from a file
8426 You can use the commands @code{dump}, @code{append}, and
8427 @code{restore} to copy data between target memory and a file. The
8428 @code{dump} and @code{append} commands write data to a file, and the
8429 @code{restore} command reads data from a file back into the inferior's
8430 memory. Files may be in binary, Motorola S-record, Intel hex, or
8431 Tektronix Hex format; however, @value{GDBN} can only append to binary
8437 @item dump @r{[}@var{format}@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
8438 @itemx dump @r{[}@var{format}@r{]} value @var{filename} @var{expr}
8439 Dump the contents of memory from @var{start_addr} to @var{end_addr},
8440 or the value of @var{expr}, to @var{filename} in the given format.
8442 The @var{format} parameter may be any one of:
8449 Motorola S-record format.
8451 Tektronix Hex format.
8454 @value{GDBN} uses the same definitions of these formats as the
8455 @sc{gnu} binary utilities, like @samp{objdump} and @samp{objcopy}. If
8456 @var{format} is omitted, @value{GDBN} dumps the data in raw binary
8460 @item append @r{[}binary@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
8461 @itemx append @r{[}binary@r{]} value @var{filename} @var{expr}
8462 Append the contents of memory from @var{start_addr} to @var{end_addr},
8463 or the value of @var{expr}, to the file @var{filename}, in raw binary form.
8464 (@value{GDBN} can only append data to files in raw binary form.)
8467 @item restore @var{filename} @r{[}binary@r{]} @var{bias} @var{start} @var{end}
8468 Restore the contents of file @var{filename} into memory. The
8469 @code{restore} command can automatically recognize any known @sc{bfd}
8470 file format, except for raw binary. To restore a raw binary file you
8471 must specify the optional keyword @code{binary} after the filename.
8473 If @var{bias} is non-zero, its value will be added to the addresses
8474 contained in the file. Binary files always start at address zero, so
8475 they will be restored at address @var{bias}. Other bfd files have
8476 a built-in location; they will be restored at offset @var{bias}
8479 If @var{start} and/or @var{end} are non-zero, then only data between
8480 file offset @var{start} and file offset @var{end} will be restored.
8481 These offsets are relative to the addresses in the file, before
8482 the @var{bias} argument is applied.
8486 @node Core File Generation
8487 @section How to Produce a Core File from Your Program
8488 @cindex dump core from inferior
8490 A @dfn{core file} or @dfn{core dump} is a file that records the memory
8491 image of a running process and its process status (register values
8492 etc.). Its primary use is post-mortem debugging of a program that
8493 crashed while it ran outside a debugger. A program that crashes
8494 automatically produces a core file, unless this feature is disabled by
8495 the user. @xref{Files}, for information on invoking @value{GDBN} in
8496 the post-mortem debugging mode.
8498 Occasionally, you may wish to produce a core file of the program you
8499 are debugging in order to preserve a snapshot of its state.
8500 @value{GDBN} has a special command for that.
8504 @kindex generate-core-file
8505 @item generate-core-file [@var{file}]
8506 @itemx gcore [@var{file}]
8507 Produce a core dump of the inferior process. The optional argument
8508 @var{file} specifies the file name where to put the core dump. If not
8509 specified, the file name defaults to @file{core.@var{pid}}, where
8510 @var{pid} is the inferior process ID.
8512 Note that this command is implemented only for some systems (as of
8513 this writing, @sc{gnu}/Linux, FreeBSD, Solaris, Unixware, and S390).
8516 @node Character Sets
8517 @section Character Sets
8518 @cindex character sets
8520 @cindex translating between character sets
8521 @cindex host character set
8522 @cindex target character set
8524 If the program you are debugging uses a different character set to
8525 represent characters and strings than the one @value{GDBN} uses itself,
8526 @value{GDBN} can automatically translate between the character sets for
8527 you. The character set @value{GDBN} uses we call the @dfn{host
8528 character set}; the one the inferior program uses we call the
8529 @dfn{target character set}.
8531 For example, if you are running @value{GDBN} on a @sc{gnu}/Linux system, which
8532 uses the ISO Latin 1 character set, but you are using @value{GDBN}'s
8533 remote protocol (@pxref{Remote Debugging}) to debug a program
8534 running on an IBM mainframe, which uses the @sc{ebcdic} character set,
8535 then the host character set is Latin-1, and the target character set is
8536 @sc{ebcdic}. If you give @value{GDBN} the command @code{set
8537 target-charset EBCDIC-US}, then @value{GDBN} translates between
8538 @sc{ebcdic} and Latin 1 as you print character or string values, or use
8539 character and string literals in expressions.
8541 @value{GDBN} has no way to automatically recognize which character set
8542 the inferior program uses; you must tell it, using the @code{set
8543 target-charset} command, described below.
8545 Here are the commands for controlling @value{GDBN}'s character set
8549 @item set target-charset @var{charset}
8550 @kindex set target-charset
8551 Set the current target character set to @var{charset}. To display the
8552 list of supported target character sets, type
8553 @kbd{@w{set target-charset @key{TAB}@key{TAB}}}.
8555 @item set host-charset @var{charset}
8556 @kindex set host-charset
8557 Set the current host character set to @var{charset}.
8559 By default, @value{GDBN} uses a host character set appropriate to the
8560 system it is running on; you can override that default using the
8561 @code{set host-charset} command. On some systems, @value{GDBN} cannot
8562 automatically determine the appropriate host character set. In this
8563 case, @value{GDBN} uses @samp{UTF-8}.
8565 @value{GDBN} can only use certain character sets as its host character
8566 set. If you type @kbd{@w{set target-charset @key{TAB}@key{TAB}}},
8567 @value{GDBN} will list the host character sets it supports.
8569 @item set charset @var{charset}
8571 Set the current host and target character sets to @var{charset}. As
8572 above, if you type @kbd{@w{set charset @key{TAB}@key{TAB}}},
8573 @value{GDBN} will list the names of the character sets that can be used
8574 for both host and target.
8577 @kindex show charset
8578 Show the names of the current host and target character sets.
8580 @item show host-charset
8581 @kindex show host-charset
8582 Show the name of the current host character set.
8584 @item show target-charset
8585 @kindex show target-charset
8586 Show the name of the current target character set.
8588 @item set target-wide-charset @var{charset}
8589 @kindex set target-wide-charset
8590 Set the current target's wide character set to @var{charset}. This is
8591 the character set used by the target's @code{wchar_t} type. To
8592 display the list of supported wide character sets, type
8593 @kbd{@w{set target-wide-charset @key{TAB}@key{TAB}}}.
8595 @item show target-wide-charset
8596 @kindex show target-wide-charset
8597 Show the name of the current target's wide character set.
8600 Here is an example of @value{GDBN}'s character set support in action.
8601 Assume that the following source code has been placed in the file
8602 @file{charset-test.c}:
8608 = @{72, 101, 108, 108, 111, 44, 32, 119,
8609 111, 114, 108, 100, 33, 10, 0@};
8610 char ibm1047_hello[]
8611 = @{200, 133, 147, 147, 150, 107, 64, 166,
8612 150, 153, 147, 132, 90, 37, 0@};
8616 printf ("Hello, world!\n");
8620 In this program, @code{ascii_hello} and @code{ibm1047_hello} are arrays
8621 containing the string @samp{Hello, world!} followed by a newline,
8622 encoded in the @sc{ascii} and @sc{ibm1047} character sets.
8624 We compile the program, and invoke the debugger on it:
8627 $ gcc -g charset-test.c -o charset-test
8628 $ gdb -nw charset-test
8629 GNU gdb 2001-12-19-cvs
8630 Copyright 2001 Free Software Foundation, Inc.
8635 We can use the @code{show charset} command to see what character sets
8636 @value{GDBN} is currently using to interpret and display characters and
8640 (@value{GDBP}) show charset
8641 The current host and target character set is `ISO-8859-1'.
8645 For the sake of printing this manual, let's use @sc{ascii} as our
8646 initial character set:
8648 (@value{GDBP}) set charset ASCII
8649 (@value{GDBP}) show charset
8650 The current host and target character set is `ASCII'.
8654 Let's assume that @sc{ascii} is indeed the correct character set for our
8655 host system --- in other words, let's assume that if @value{GDBN} prints
8656 characters using the @sc{ascii} character set, our terminal will display
8657 them properly. Since our current target character set is also
8658 @sc{ascii}, the contents of @code{ascii_hello} print legibly:
8661 (@value{GDBP}) print ascii_hello
8662 $1 = 0x401698 "Hello, world!\n"
8663 (@value{GDBP}) print ascii_hello[0]
8668 @value{GDBN} uses the target character set for character and string
8669 literals you use in expressions:
8672 (@value{GDBP}) print '+'
8677 The @sc{ascii} character set uses the number 43 to encode the @samp{+}
8680 @value{GDBN} relies on the user to tell it which character set the
8681 target program uses. If we print @code{ibm1047_hello} while our target
8682 character set is still @sc{ascii}, we get jibberish:
8685 (@value{GDBP}) print ibm1047_hello
8686 $4 = 0x4016a8 "\310\205\223\223\226k@@\246\226\231\223\204Z%"
8687 (@value{GDBP}) print ibm1047_hello[0]
8692 If we invoke the @code{set target-charset} followed by @key{TAB}@key{TAB},
8693 @value{GDBN} tells us the character sets it supports:
8696 (@value{GDBP}) set target-charset
8697 ASCII EBCDIC-US IBM1047 ISO-8859-1
8698 (@value{GDBP}) set target-charset
8701 We can select @sc{ibm1047} as our target character set, and examine the
8702 program's strings again. Now the @sc{ascii} string is wrong, but
8703 @value{GDBN} translates the contents of @code{ibm1047_hello} from the
8704 target character set, @sc{ibm1047}, to the host character set,
8705 @sc{ascii}, and they display correctly:
8708 (@value{GDBP}) set target-charset IBM1047
8709 (@value{GDBP}) show charset
8710 The current host character set is `ASCII'.
8711 The current target character set is `IBM1047'.
8712 (@value{GDBP}) print ascii_hello
8713 $6 = 0x401698 "\110\145%%?\054\040\167?\162%\144\041\012"
8714 (@value{GDBP}) print ascii_hello[0]
8716 (@value{GDBP}) print ibm1047_hello
8717 $8 = 0x4016a8 "Hello, world!\n"
8718 (@value{GDBP}) print ibm1047_hello[0]
8723 As above, @value{GDBN} uses the target character set for character and
8724 string literals you use in expressions:
8727 (@value{GDBP}) print '+'
8732 The @sc{ibm1047} character set uses the number 78 to encode the @samp{+}
8735 @node Caching Remote Data
8736 @section Caching Data of Remote Targets
8737 @cindex caching data of remote targets
8739 @value{GDBN} caches data exchanged between the debugger and a
8740 remote target (@pxref{Remote Debugging}). Such caching generally improves
8741 performance, because it reduces the overhead of the remote protocol by
8742 bundling memory reads and writes into large chunks. Unfortunately, simply
8743 caching everything would lead to incorrect results, since @value{GDBN}
8744 does not necessarily know anything about volatile values, memory-mapped I/O
8745 addresses, etc. Furthermore, in non-stop mode (@pxref{Non-Stop Mode})
8746 memory can be changed @emph{while} a gdb command is executing.
8747 Therefore, by default, @value{GDBN} only caches data
8748 known to be on the stack@footnote{In non-stop mode, it is moderately
8749 rare for a running thread to modify the stack of a stopped thread
8750 in a way that would interfere with a backtrace, and caching of
8751 stack reads provides a significant speed up of remote backtraces.}.
8752 Other regions of memory can be explicitly marked as
8753 cacheable; see @pxref{Memory Region Attributes}.
8756 @kindex set remotecache
8757 @item set remotecache on
8758 @itemx set remotecache off
8759 This option no longer does anything; it exists for compatibility
8762 @kindex show remotecache
8763 @item show remotecache
8764 Show the current state of the obsolete remotecache flag.
8766 @kindex set stack-cache
8767 @item set stack-cache on
8768 @itemx set stack-cache off
8769 Enable or disable caching of stack accesses. When @code{ON}, use
8770 caching. By default, this option is @code{ON}.
8772 @kindex show stack-cache
8773 @item show stack-cache
8774 Show the current state of data caching for memory accesses.
8777 @item info dcache @r{[}line@r{]}
8778 Print the information about the data cache performance. The
8779 information displayed includes the dcache width and depth, and for
8780 each cache line, its number, address, and how many times it was
8781 referenced. This command is useful for debugging the data cache
8784 If a line number is specified, the contents of that line will be
8788 @node Searching Memory
8789 @section Search Memory
8790 @cindex searching memory
8792 Memory can be searched for a particular sequence of bytes with the
8793 @code{find} command.
8797 @item find @r{[}/@var{sn}@r{]} @var{start_addr}, +@var{len}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
8798 @itemx find @r{[}/@var{sn}@r{]} @var{start_addr}, @var{end_addr}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
8799 Search memory for the sequence of bytes specified by @var{val1}, @var{val2},
8800 etc. The search begins at address @var{start_addr} and continues for either
8801 @var{len} bytes or through to @var{end_addr} inclusive.
8804 @var{s} and @var{n} are optional parameters.
8805 They may be specified in either order, apart or together.
8808 @item @var{s}, search query size
8809 The size of each search query value.
8815 halfwords (two bytes)
8819 giant words (eight bytes)
8822 All values are interpreted in the current language.
8823 This means, for example, that if the current source language is C/C@t{++}
8824 then searching for the string ``hello'' includes the trailing '\0'.
8826 If the value size is not specified, it is taken from the
8827 value's type in the current language.
8828 This is useful when one wants to specify the search
8829 pattern as a mixture of types.
8830 Note that this means, for example, that in the case of C-like languages
8831 a search for an untyped 0x42 will search for @samp{(int) 0x42}
8832 which is typically four bytes.
8834 @item @var{n}, maximum number of finds
8835 The maximum number of matches to print. The default is to print all finds.
8838 You can use strings as search values. Quote them with double-quotes
8840 The string value is copied into the search pattern byte by byte,
8841 regardless of the endianness of the target and the size specification.
8843 The address of each match found is printed as well as a count of the
8844 number of matches found.
8846 The address of the last value found is stored in convenience variable
8848 A count of the number of matches is stored in @samp{$numfound}.
8850 For example, if stopped at the @code{printf} in this function:
8856 static char hello[] = "hello-hello";
8857 static struct @{ char c; short s; int i; @}
8858 __attribute__ ((packed)) mixed
8859 = @{ 'c', 0x1234, 0x87654321 @};
8860 printf ("%s\n", hello);
8865 you get during debugging:
8868 (gdb) find &hello[0], +sizeof(hello), "hello"
8869 0x804956d <hello.1620+6>
8871 (gdb) find &hello[0], +sizeof(hello), 'h', 'e', 'l', 'l', 'o'
8872 0x8049567 <hello.1620>
8873 0x804956d <hello.1620+6>
8875 (gdb) find /b1 &hello[0], +sizeof(hello), 'h', 0x65, 'l'
8876 0x8049567 <hello.1620>
8878 (gdb) find &mixed, +sizeof(mixed), (char) 'c', (short) 0x1234, (int) 0x87654321
8879 0x8049560 <mixed.1625>
8881 (gdb) print $numfound
8884 $2 = (void *) 0x8049560
8887 @node Optimized Code
8888 @chapter Debugging Optimized Code
8889 @cindex optimized code, debugging
8890 @cindex debugging optimized code
8892 Almost all compilers support optimization. With optimization
8893 disabled, the compiler generates assembly code that corresponds
8894 directly to your source code, in a simplistic way. As the compiler
8895 applies more powerful optimizations, the generated assembly code
8896 diverges from your original source code. With help from debugging
8897 information generated by the compiler, @value{GDBN} can map from
8898 the running program back to constructs from your original source.
8900 @value{GDBN} is more accurate with optimization disabled. If you
8901 can recompile without optimization, it is easier to follow the
8902 progress of your program during debugging. But, there are many cases
8903 where you may need to debug an optimized version.
8905 When you debug a program compiled with @samp{-g -O}, remember that the
8906 optimizer has rearranged your code; the debugger shows you what is
8907 really there. Do not be too surprised when the execution path does not
8908 exactly match your source file! An extreme example: if you define a
8909 variable, but never use it, @value{GDBN} never sees that
8910 variable---because the compiler optimizes it out of existence.
8912 Some things do not work as well with @samp{-g -O} as with just
8913 @samp{-g}, particularly on machines with instruction scheduling. If in
8914 doubt, recompile with @samp{-g} alone, and if this fixes the problem,
8915 please report it to us as a bug (including a test case!).
8916 @xref{Variables}, for more information about debugging optimized code.
8919 * Inline Functions:: How @value{GDBN} presents inlining
8922 @node Inline Functions
8923 @section Inline Functions
8924 @cindex inline functions, debugging
8926 @dfn{Inlining} is an optimization that inserts a copy of the function
8927 body directly at each call site, instead of jumping to a shared
8928 routine. @value{GDBN} displays inlined functions just like
8929 non-inlined functions. They appear in backtraces. You can view their
8930 arguments and local variables, step into them with @code{step}, skip
8931 them with @code{next}, and escape from them with @code{finish}.
8932 You can check whether a function was inlined by using the
8933 @code{info frame} command.
8935 For @value{GDBN} to support inlined functions, the compiler must
8936 record information about inlining in the debug information ---
8937 @value{NGCC} using the @sc{dwarf 2} format does this, and several
8938 other compilers do also. @value{GDBN} only supports inlined functions
8939 when using @sc{dwarf 2}. Versions of @value{NGCC} before 4.1
8940 do not emit two required attributes (@samp{DW_AT_call_file} and
8941 @samp{DW_AT_call_line}); @value{GDBN} does not display inlined
8942 function calls with earlier versions of @value{NGCC}. It instead
8943 displays the arguments and local variables of inlined functions as
8944 local variables in the caller.
8946 The body of an inlined function is directly included at its call site;
8947 unlike a non-inlined function, there are no instructions devoted to
8948 the call. @value{GDBN} still pretends that the call site and the
8949 start of the inlined function are different instructions. Stepping to
8950 the call site shows the call site, and then stepping again shows
8951 the first line of the inlined function, even though no additional
8952 instructions are executed.
8954 This makes source-level debugging much clearer; you can see both the
8955 context of the call and then the effect of the call. Only stepping by
8956 a single instruction using @code{stepi} or @code{nexti} does not do
8957 this; single instruction steps always show the inlined body.
8959 There are some ways that @value{GDBN} does not pretend that inlined
8960 function calls are the same as normal calls:
8964 You cannot set breakpoints on inlined functions. @value{GDBN}
8965 either reports that there is no symbol with that name, or else sets the
8966 breakpoint only on non-inlined copies of the function. This limitation
8967 will be removed in a future version of @value{GDBN}; until then,
8968 set a breakpoint by line number on the first line of the inlined
8972 Setting breakpoints at the call site of an inlined function may not
8973 work, because the call site does not contain any code. @value{GDBN}
8974 may incorrectly move the breakpoint to the next line of the enclosing
8975 function, after the call. This limitation will be removed in a future
8976 version of @value{GDBN}; until then, set a breakpoint on an earlier line
8977 or inside the inlined function instead.
8980 @value{GDBN} cannot locate the return value of inlined calls after
8981 using the @code{finish} command. This is a limitation of compiler-generated
8982 debugging information; after @code{finish}, you can step to the next line
8983 and print a variable where your program stored the return value.
8989 @chapter C Preprocessor Macros
8991 Some languages, such as C and C@t{++}, provide a way to define and invoke
8992 ``preprocessor macros'' which expand into strings of tokens.
8993 @value{GDBN} can evaluate expressions containing macro invocations, show
8994 the result of macro expansion, and show a macro's definition, including
8995 where it was defined.
8997 You may need to compile your program specially to provide @value{GDBN}
8998 with information about preprocessor macros. Most compilers do not
8999 include macros in their debugging information, even when you compile
9000 with the @option{-g} flag. @xref{Compilation}.
9002 A program may define a macro at one point, remove that definition later,
9003 and then provide a different definition after that. Thus, at different
9004 points in the program, a macro may have different definitions, or have
9005 no definition at all. If there is a current stack frame, @value{GDBN}
9006 uses the macros in scope at that frame's source code line. Otherwise,
9007 @value{GDBN} uses the macros in scope at the current listing location;
9010 Whenever @value{GDBN} evaluates an expression, it always expands any
9011 macro invocations present in the expression. @value{GDBN} also provides
9012 the following commands for working with macros explicitly.
9016 @kindex macro expand
9017 @cindex macro expansion, showing the results of preprocessor
9018 @cindex preprocessor macro expansion, showing the results of
9019 @cindex expanding preprocessor macros
9020 @item macro expand @var{expression}
9021 @itemx macro exp @var{expression}
9022 Show the results of expanding all preprocessor macro invocations in
9023 @var{expression}. Since @value{GDBN} simply expands macros, but does
9024 not parse the result, @var{expression} need not be a valid expression;
9025 it can be any string of tokens.
9028 @item macro expand-once @var{expression}
9029 @itemx macro exp1 @var{expression}
9030 @cindex expand macro once
9031 @i{(This command is not yet implemented.)} Show the results of
9032 expanding those preprocessor macro invocations that appear explicitly in
9033 @var{expression}. Macro invocations appearing in that expansion are
9034 left unchanged. This command allows you to see the effect of a
9035 particular macro more clearly, without being confused by further
9036 expansions. Since @value{GDBN} simply expands macros, but does not
9037 parse the result, @var{expression} need not be a valid expression; it
9038 can be any string of tokens.
9041 @cindex macro definition, showing
9042 @cindex definition, showing a macro's
9043 @item info macro @var{macro}
9044 Show the definition of the macro named @var{macro}, and describe the
9045 source location or compiler command-line where that definition was established.
9047 @kindex macro define
9048 @cindex user-defined macros
9049 @cindex defining macros interactively
9050 @cindex macros, user-defined
9051 @item macro define @var{macro} @var{replacement-list}
9052 @itemx macro define @var{macro}(@var{arglist}) @var{replacement-list}
9053 Introduce a definition for a preprocessor macro named @var{macro},
9054 invocations of which are replaced by the tokens given in
9055 @var{replacement-list}. The first form of this command defines an
9056 ``object-like'' macro, which takes no arguments; the second form
9057 defines a ``function-like'' macro, which takes the arguments given in
9060 A definition introduced by this command is in scope in every
9061 expression evaluated in @value{GDBN}, until it is removed with the
9062 @code{macro undef} command, described below. The definition overrides
9063 all definitions for @var{macro} present in the program being debugged,
9064 as well as any previous user-supplied definition.
9067 @item macro undef @var{macro}
9068 Remove any user-supplied definition for the macro named @var{macro}.
9069 This command only affects definitions provided with the @code{macro
9070 define} command, described above; it cannot remove definitions present
9071 in the program being debugged.
9075 List all the macros defined using the @code{macro define} command.
9078 @cindex macros, example of debugging with
9079 Here is a transcript showing the above commands in action. First, we
9080 show our source files:
9088 #define ADD(x) (M + x)
9093 printf ("Hello, world!\n");
9095 printf ("We're so creative.\n");
9097 printf ("Goodbye, world!\n");
9104 Now, we compile the program using the @sc{gnu} C compiler, @value{NGCC}.
9105 We pass the @option{-gdwarf-2} and @option{-g3} flags to ensure the
9106 compiler includes information about preprocessor macros in the debugging
9110 $ gcc -gdwarf-2 -g3 sample.c -o sample
9114 Now, we start @value{GDBN} on our sample program:
9118 GNU gdb 2002-05-06-cvs
9119 Copyright 2002 Free Software Foundation, Inc.
9120 GDB is free software, @dots{}
9124 We can expand macros and examine their definitions, even when the
9125 program is not running. @value{GDBN} uses the current listing position
9126 to decide which macro definitions are in scope:
9129 (@value{GDBP}) list main
9132 5 #define ADD(x) (M + x)
9137 10 printf ("Hello, world!\n");
9139 12 printf ("We're so creative.\n");
9140 (@value{GDBP}) info macro ADD
9141 Defined at /home/jimb/gdb/macros/play/sample.c:5
9142 #define ADD(x) (M + x)
9143 (@value{GDBP}) info macro Q
9144 Defined at /home/jimb/gdb/macros/play/sample.h:1
9145 included at /home/jimb/gdb/macros/play/sample.c:2
9147 (@value{GDBP}) macro expand ADD(1)
9148 expands to: (42 + 1)
9149 (@value{GDBP}) macro expand-once ADD(1)
9150 expands to: once (M + 1)
9154 In the example above, note that @code{macro expand-once} expands only
9155 the macro invocation explicit in the original text --- the invocation of
9156 @code{ADD} --- but does not expand the invocation of the macro @code{M},
9157 which was introduced by @code{ADD}.
9159 Once the program is running, @value{GDBN} uses the macro definitions in
9160 force at the source line of the current stack frame:
9163 (@value{GDBP}) break main
9164 Breakpoint 1 at 0x8048370: file sample.c, line 10.
9166 Starting program: /home/jimb/gdb/macros/play/sample
9168 Breakpoint 1, main () at sample.c:10
9169 10 printf ("Hello, world!\n");
9173 At line 10, the definition of the macro @code{N} at line 9 is in force:
9176 (@value{GDBP}) info macro N
9177 Defined at /home/jimb/gdb/macros/play/sample.c:9
9179 (@value{GDBP}) macro expand N Q M
9181 (@value{GDBP}) print N Q M
9186 As we step over directives that remove @code{N}'s definition, and then
9187 give it a new definition, @value{GDBN} finds the definition (or lack
9188 thereof) in force at each point:
9193 12 printf ("We're so creative.\n");
9194 (@value{GDBP}) info macro N
9195 The symbol `N' has no definition as a C/C++ preprocessor macro
9196 at /home/jimb/gdb/macros/play/sample.c:12
9199 14 printf ("Goodbye, world!\n");
9200 (@value{GDBP}) info macro N
9201 Defined at /home/jimb/gdb/macros/play/sample.c:13
9203 (@value{GDBP}) macro expand N Q M
9204 expands to: 1729 < 42
9205 (@value{GDBP}) print N Q M
9210 In addition to source files, macros can be defined on the compilation command
9211 line using the @option{-D@var{name}=@var{value}} syntax. For macros defined in
9212 such a way, @value{GDBN} displays the location of their definition as line zero
9213 of the source file submitted to the compiler.
9216 (@value{GDBP}) info macro __STDC__
9217 Defined at /home/jimb/gdb/macros/play/sample.c:0
9224 @chapter Tracepoints
9225 @c This chapter is based on the documentation written by Michael
9226 @c Snyder, David Taylor, Jim Blandy, and Elena Zannoni.
9229 In some applications, it is not feasible for the debugger to interrupt
9230 the program's execution long enough for the developer to learn
9231 anything helpful about its behavior. If the program's correctness
9232 depends on its real-time behavior, delays introduced by a debugger
9233 might cause the program to change its behavior drastically, or perhaps
9234 fail, even when the code itself is correct. It is useful to be able
9235 to observe the program's behavior without interrupting it.
9237 Using @value{GDBN}'s @code{trace} and @code{collect} commands, you can
9238 specify locations in the program, called @dfn{tracepoints}, and
9239 arbitrary expressions to evaluate when those tracepoints are reached.
9240 Later, using the @code{tfind} command, you can examine the values
9241 those expressions had when the program hit the tracepoints. The
9242 expressions may also denote objects in memory---structures or arrays,
9243 for example---whose values @value{GDBN} should record; while visiting
9244 a particular tracepoint, you may inspect those objects as if they were
9245 in memory at that moment. However, because @value{GDBN} records these
9246 values without interacting with you, it can do so quickly and
9247 unobtrusively, hopefully not disturbing the program's behavior.
9249 The tracepoint facility is currently available only for remote
9250 targets. @xref{Targets}. In addition, your remote target must know
9251 how to collect trace data. This functionality is implemented in the
9252 remote stub; however, none of the stubs distributed with @value{GDBN}
9253 support tracepoints as of this writing. The format of the remote
9254 packets used to implement tracepoints are described in @ref{Tracepoint
9257 This chapter describes the tracepoint commands and features.
9261 * Analyze Collected Data::
9262 * Tracepoint Variables::
9265 @node Set Tracepoints
9266 @section Commands to Set Tracepoints
9268 Before running such a @dfn{trace experiment}, an arbitrary number of
9269 tracepoints can be set. A tracepoint is actually a special type of
9270 breakpoint (@pxref{Set Breaks}), so you can manipulate it using
9271 standard breakpoint commands. For instance, as with breakpoints,
9272 tracepoint numbers are successive integers starting from one, and many
9273 of the commands associated with tracepoints take the tracepoint number
9274 as their argument, to identify which tracepoint to work on.
9276 For each tracepoint, you can specify, in advance, some arbitrary set
9277 of data that you want the target to collect in the trace buffer when
9278 it hits that tracepoint. The collected data can include registers,
9279 local variables, or global data. Later, you can use @value{GDBN}
9280 commands to examine the values these data had at the time the
9283 Tracepoints do not support every breakpoint feature. Conditional
9284 expressions and ignore counts on tracepoints have no effect, and
9285 tracepoints cannot run @value{GDBN} commands when they are
9286 hit. Tracepoints may not be thread-specific either.
9288 This section describes commands to set tracepoints and associated
9289 conditions and actions.
9292 * Create and Delete Tracepoints::
9293 * Enable and Disable Tracepoints::
9294 * Tracepoint Passcounts::
9295 * Tracepoint Conditions::
9296 * Tracepoint Actions::
9297 * Listing Tracepoints::
9298 * Starting and Stopping Trace Experiments::
9301 @node Create and Delete Tracepoints
9302 @subsection Create and Delete Tracepoints
9305 @cindex set tracepoint
9307 @item trace @var{location}
9308 The @code{trace} command is very similar to the @code{break} command.
9309 Its argument @var{location} can be a source line, a function name, or
9310 an address in the target program. @xref{Specify Location}. The
9311 @code{trace} command defines a tracepoint, which is a point in the
9312 target program where the debugger will briefly stop, collect some
9313 data, and then allow the program to continue. Setting a tracepoint or
9314 changing its actions doesn't take effect until the next @code{tstart}
9315 command, and once a trace experiment is running, further changes will
9316 not have any effect until the next trace experiment starts.
9318 Here are some examples of using the @code{trace} command:
9321 (@value{GDBP}) @b{trace foo.c:121} // a source file and line number
9323 (@value{GDBP}) @b{trace +2} // 2 lines forward
9325 (@value{GDBP}) @b{trace my_function} // first source line of function
9327 (@value{GDBP}) @b{trace *my_function} // EXACT start address of function
9329 (@value{GDBP}) @b{trace *0x2117c4} // an address
9333 You can abbreviate @code{trace} as @code{tr}.
9335 @item trace @var{location} if @var{cond}
9336 Set a tracepoint with condition @var{cond}; evaluate the expression
9337 @var{cond} each time the tracepoint is reached, and collect data only
9338 if the value is nonzero---that is, if @var{cond} evaluates as true.
9339 @xref{Tracepoint Conditions, ,Tracepoint Conditions}, for more
9340 information on tracepoint conditions.
9343 @cindex last tracepoint number
9344 @cindex recent tracepoint number
9345 @cindex tracepoint number
9346 The convenience variable @code{$tpnum} records the tracepoint number
9347 of the most recently set tracepoint.
9349 @kindex delete tracepoint
9350 @cindex tracepoint deletion
9351 @item delete tracepoint @r{[}@var{num}@r{]}
9352 Permanently delete one or more tracepoints. With no argument, the
9353 default is to delete all tracepoints. Note that the regular
9354 @code{delete} command can remove tracepoints also.
9359 (@value{GDBP}) @b{delete trace 1 2 3} // remove three tracepoints
9361 (@value{GDBP}) @b{delete trace} // remove all tracepoints
9365 You can abbreviate this command as @code{del tr}.
9368 @node Enable and Disable Tracepoints
9369 @subsection Enable and Disable Tracepoints
9371 These commands are deprecated; they are equivalent to plain @code{disable} and @code{enable}.
9374 @kindex disable tracepoint
9375 @item disable tracepoint @r{[}@var{num}@r{]}
9376 Disable tracepoint @var{num}, or all tracepoints if no argument
9377 @var{num} is given. A disabled tracepoint will have no effect during
9378 the next trace experiment, but it is not forgotten. You can re-enable
9379 a disabled tracepoint using the @code{enable tracepoint} command.
9381 @kindex enable tracepoint
9382 @item enable tracepoint @r{[}@var{num}@r{]}
9383 Enable tracepoint @var{num}, or all tracepoints. The enabled
9384 tracepoints will become effective the next time a trace experiment is
9388 @node Tracepoint Passcounts
9389 @subsection Tracepoint Passcounts
9393 @cindex tracepoint pass count
9394 @item passcount @r{[}@var{n} @r{[}@var{num}@r{]]}
9395 Set the @dfn{passcount} of a tracepoint. The passcount is a way to
9396 automatically stop a trace experiment. If a tracepoint's passcount is
9397 @var{n}, then the trace experiment will be automatically stopped on
9398 the @var{n}'th time that tracepoint is hit. If the tracepoint number
9399 @var{num} is not specified, the @code{passcount} command sets the
9400 passcount of the most recently defined tracepoint. If no passcount is
9401 given, the trace experiment will run until stopped explicitly by the
9407 (@value{GDBP}) @b{passcount 5 2} // Stop on the 5th execution of
9408 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// tracepoint 2}
9410 (@value{GDBP}) @b{passcount 12} // Stop on the 12th execution of the
9411 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// most recently defined tracepoint.}
9412 (@value{GDBP}) @b{trace foo}
9413 (@value{GDBP}) @b{pass 3}
9414 (@value{GDBP}) @b{trace bar}
9415 (@value{GDBP}) @b{pass 2}
9416 (@value{GDBP}) @b{trace baz}
9417 (@value{GDBP}) @b{pass 1} // Stop tracing when foo has been
9418 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// executed 3 times OR when bar has}
9419 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// been executed 2 times}
9420 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// OR when baz has been executed 1 time.}
9424 @node Tracepoint Conditions
9425 @subsection Tracepoint Conditions
9426 @cindex conditional tracepoints
9427 @cindex tracepoint conditions
9429 The simplest sort of tracepoint collects data every time your program
9430 reaches a specified place. You can also specify a @dfn{condition} for
9431 a tracepoint. A condition is just a Boolean expression in your
9432 programming language (@pxref{Expressions, ,Expressions}). A
9433 tracepoint with a condition evaluates the expression each time your
9434 program reaches it, and data collection happens only if the condition
9437 Tracepoint conditions can be specified when a tracepoint is set, by
9438 using @samp{if} in the arguments to the @code{trace} command.
9439 @xref{Create and Delete Tracepoints, ,Setting Tracepoints}. They can
9440 also be set or changed at any time with the @code{condition} command,
9441 just as with breakpoints.
9443 Unlike breakpoint conditions, @value{GDBN} does not actually evaluate
9444 the conditional expression itself. Instead, @value{GDBN} encodes the
9445 expression into an agent expression (@pxref{Agent Expressions}
9446 suitable for execution on the target, independently of @value{GDBN}.
9447 Global variables become raw memory locations, locals become stack
9448 accesses, and so forth.
9450 For instance, suppose you have a function that is usually called
9451 frequently, but should not be called after an error has occurred. You
9452 could use the following tracepoint command to collect data about calls
9453 of that function that happen while the error code is propagating
9454 through the program; an unconditional tracepoint could end up
9455 collecting thousands of useless trace frames that you would have to
9459 (@value{GDBP}) @kbd{trace normal_operation if errcode > 0}
9462 @node Tracepoint Actions
9463 @subsection Tracepoint Action Lists
9467 @cindex tracepoint actions
9468 @item actions @r{[}@var{num}@r{]}
9469 This command will prompt for a list of actions to be taken when the
9470 tracepoint is hit. If the tracepoint number @var{num} is not
9471 specified, this command sets the actions for the one that was most
9472 recently defined (so that you can define a tracepoint and then say
9473 @code{actions} without bothering about its number). You specify the
9474 actions themselves on the following lines, one action at a time, and
9475 terminate the actions list with a line containing just @code{end}. So
9476 far, the only defined actions are @code{collect} and
9477 @code{while-stepping}.
9479 @cindex remove actions from a tracepoint
9480 To remove all actions from a tracepoint, type @samp{actions @var{num}}
9481 and follow it immediately with @samp{end}.
9484 (@value{GDBP}) @b{collect @var{data}} // collect some data
9486 (@value{GDBP}) @b{while-stepping 5} // single-step 5 times, collect data
9488 (@value{GDBP}) @b{end} // signals the end of actions.
9491 In the following example, the action list begins with @code{collect}
9492 commands indicating the things to be collected when the tracepoint is
9493 hit. Then, in order to single-step and collect additional data
9494 following the tracepoint, a @code{while-stepping} command is used,
9495 followed by the list of things to be collected while stepping. The
9496 @code{while-stepping} command is terminated by its own separate
9497 @code{end} command. Lastly, the action list is terminated by an
9501 (@value{GDBP}) @b{trace foo}
9502 (@value{GDBP}) @b{actions}
9503 Enter actions for tracepoint 1, one per line:
9512 @kindex collect @r{(tracepoints)}
9513 @item collect @var{expr1}, @var{expr2}, @dots{}
9514 Collect values of the given expressions when the tracepoint is hit.
9515 This command accepts a comma-separated list of any valid expressions.
9516 In addition to global, static, or local variables, the following
9517 special arguments are supported:
9521 collect all registers
9524 collect all function arguments
9527 collect all local variables.
9530 You can give several consecutive @code{collect} commands, each one
9531 with a single argument, or one @code{collect} command with several
9532 arguments separated by commas: the effect is the same.
9534 The command @code{info scope} (@pxref{Symbols, info scope}) is
9535 particularly useful for figuring out what data to collect.
9537 @kindex while-stepping @r{(tracepoints)}
9538 @item while-stepping @var{n}
9539 Perform @var{n} single-step traces after the tracepoint, collecting
9540 new data at each step. The @code{while-stepping} command is
9541 followed by the list of what to collect while stepping (followed by
9542 its own @code{end} command):
9546 > collect $regs, myglobal
9552 You may abbreviate @code{while-stepping} as @code{ws} or
9556 @node Listing Tracepoints
9557 @subsection Listing Tracepoints
9560 @kindex info tracepoints
9562 @cindex information about tracepoints
9563 @item info tracepoints @r{[}@var{num}@r{]}
9564 Display information about the tracepoint @var{num}. If you don't
9565 specify a tracepoint number, displays information about all the
9566 tracepoints defined so far. The format is similar to that used for
9567 @code{info breakpoints}; in fact, @code{info tracepoints} is the same
9568 command, simply restricting itself to tracepoints.
9570 A tracepoint's listing may include additional information specific to
9575 its passcount as given by the @code{passcount @var{n}} command
9577 its step count as given by the @code{while-stepping @var{n}} command
9579 its action list as given by the @code{actions} command. The actions
9580 are prefixed with an @samp{A} so as to distinguish them from commands.
9584 (@value{GDBP}) @b{info trace}
9585 Num Type Disp Enb Address What
9586 1 tracepoint keep y 0x0804ab57 in foo() at main.cxx:7
9590 A collect globfoo, $regs
9598 This command can be abbreviated @code{info tp}.
9601 @node Starting and Stopping Trace Experiments
9602 @subsection Starting and Stopping Trace Experiments
9606 @cindex start a new trace experiment
9607 @cindex collected data discarded
9609 This command takes no arguments. It starts the trace experiment, and
9610 begins collecting data. This has the side effect of discarding all
9611 the data collected in the trace buffer during the previous trace
9615 @cindex stop a running trace experiment
9617 This command takes no arguments. It ends the trace experiment, and
9618 stops collecting data.
9620 @strong{Note}: a trace experiment and data collection may stop
9621 automatically if any tracepoint's passcount is reached
9622 (@pxref{Tracepoint Passcounts}), or if the trace buffer becomes full.
9625 @cindex status of trace data collection
9626 @cindex trace experiment, status of
9628 This command displays the status of the current trace data
9632 Here is an example of the commands we described so far:
9635 (@value{GDBP}) @b{trace gdb_c_test}
9636 (@value{GDBP}) @b{actions}
9637 Enter actions for tracepoint #1, one per line.
9638 > collect $regs,$locals,$args
9643 (@value{GDBP}) @b{tstart}
9644 [time passes @dots{}]
9645 (@value{GDBP}) @b{tstop}
9649 @node Analyze Collected Data
9650 @section Using the Collected Data
9652 After the tracepoint experiment ends, you use @value{GDBN} commands
9653 for examining the trace data. The basic idea is that each tracepoint
9654 collects a trace @dfn{snapshot} every time it is hit and another
9655 snapshot every time it single-steps. All these snapshots are
9656 consecutively numbered from zero and go into a buffer, and you can
9657 examine them later. The way you examine them is to @dfn{focus} on a
9658 specific trace snapshot. When the remote stub is focused on a trace
9659 snapshot, it will respond to all @value{GDBN} requests for memory and
9660 registers by reading from the buffer which belongs to that snapshot,
9661 rather than from @emph{real} memory or registers of the program being
9662 debugged. This means that @strong{all} @value{GDBN} commands
9663 (@code{print}, @code{info registers}, @code{backtrace}, etc.) will
9664 behave as if we were currently debugging the program state as it was
9665 when the tracepoint occurred. Any requests for data that are not in
9666 the buffer will fail.
9669 * tfind:: How to select a trace snapshot
9670 * tdump:: How to display all data for a snapshot
9671 * save-tracepoints:: How to save tracepoints for a future run
9675 @subsection @code{tfind @var{n}}
9678 @cindex select trace snapshot
9679 @cindex find trace snapshot
9680 The basic command for selecting a trace snapshot from the buffer is
9681 @code{tfind @var{n}}, which finds trace snapshot number @var{n},
9682 counting from zero. If no argument @var{n} is given, the next
9683 snapshot is selected.
9685 Here are the various forms of using the @code{tfind} command.
9689 Find the first snapshot in the buffer. This is a synonym for
9690 @code{tfind 0} (since 0 is the number of the first snapshot).
9693 Stop debugging trace snapshots, resume @emph{live} debugging.
9696 Same as @samp{tfind none}.
9699 No argument means find the next trace snapshot.
9702 Find the previous trace snapshot before the current one. This permits
9703 retracing earlier steps.
9705 @item tfind tracepoint @var{num}
9706 Find the next snapshot associated with tracepoint @var{num}. Search
9707 proceeds forward from the last examined trace snapshot. If no
9708 argument @var{num} is given, it means find the next snapshot collected
9709 for the same tracepoint as the current snapshot.
9711 @item tfind pc @var{addr}
9712 Find the next snapshot associated with the value @var{addr} of the
9713 program counter. Search proceeds forward from the last examined trace
9714 snapshot. If no argument @var{addr} is given, it means find the next
9715 snapshot with the same value of PC as the current snapshot.
9717 @item tfind outside @var{addr1}, @var{addr2}
9718 Find the next snapshot whose PC is outside the given range of
9721 @item tfind range @var{addr1}, @var{addr2}
9722 Find the next snapshot whose PC is between @var{addr1} and
9723 @var{addr2}. @c FIXME: Is the range inclusive or exclusive?
9725 @item tfind line @r{[}@var{file}:@r{]}@var{n}
9726 Find the next snapshot associated with the source line @var{n}. If
9727 the optional argument @var{file} is given, refer to line @var{n} in
9728 that source file. Search proceeds forward from the last examined
9729 trace snapshot. If no argument @var{n} is given, it means find the
9730 next line other than the one currently being examined; thus saying
9731 @code{tfind line} repeatedly can appear to have the same effect as
9732 stepping from line to line in a @emph{live} debugging session.
9735 The default arguments for the @code{tfind} commands are specifically
9736 designed to make it easy to scan through the trace buffer. For
9737 instance, @code{tfind} with no argument selects the next trace
9738 snapshot, and @code{tfind -} with no argument selects the previous
9739 trace snapshot. So, by giving one @code{tfind} command, and then
9740 simply hitting @key{RET} repeatedly you can examine all the trace
9741 snapshots in order. Or, by saying @code{tfind -} and then hitting
9742 @key{RET} repeatedly you can examine the snapshots in reverse order.
9743 The @code{tfind line} command with no argument selects the snapshot
9744 for the next source line executed. The @code{tfind pc} command with
9745 no argument selects the next snapshot with the same program counter
9746 (PC) as the current frame. The @code{tfind tracepoint} command with
9747 no argument selects the next trace snapshot collected by the same
9748 tracepoint as the current one.
9750 In addition to letting you scan through the trace buffer manually,
9751 these commands make it easy to construct @value{GDBN} scripts that
9752 scan through the trace buffer and print out whatever collected data
9753 you are interested in. Thus, if we want to examine the PC, FP, and SP
9754 registers from each trace frame in the buffer, we can say this:
9757 (@value{GDBP}) @b{tfind start}
9758 (@value{GDBP}) @b{while ($trace_frame != -1)}
9759 > printf "Frame %d, PC = %08X, SP = %08X, FP = %08X\n", \
9760 $trace_frame, $pc, $sp, $fp
9764 Frame 0, PC = 0020DC64, SP = 0030BF3C, FP = 0030BF44
9765 Frame 1, PC = 0020DC6C, SP = 0030BF38, FP = 0030BF44
9766 Frame 2, PC = 0020DC70, SP = 0030BF34, FP = 0030BF44
9767 Frame 3, PC = 0020DC74, SP = 0030BF30, FP = 0030BF44
9768 Frame 4, PC = 0020DC78, SP = 0030BF2C, FP = 0030BF44
9769 Frame 5, PC = 0020DC7C, SP = 0030BF28, FP = 0030BF44
9770 Frame 6, PC = 0020DC80, SP = 0030BF24, FP = 0030BF44
9771 Frame 7, PC = 0020DC84, SP = 0030BF20, FP = 0030BF44
9772 Frame 8, PC = 0020DC88, SP = 0030BF1C, FP = 0030BF44
9773 Frame 9, PC = 0020DC8E, SP = 0030BF18, FP = 0030BF44
9774 Frame 10, PC = 00203F6C, SP = 0030BE3C, FP = 0030BF14
9777 Or, if we want to examine the variable @code{X} at each source line in
9781 (@value{GDBP}) @b{tfind start}
9782 (@value{GDBP}) @b{while ($trace_frame != -1)}
9783 > printf "Frame %d, X == %d\n", $trace_frame, X
9793 @subsection @code{tdump}
9795 @cindex dump all data collected at tracepoint
9796 @cindex tracepoint data, display
9798 This command takes no arguments. It prints all the data collected at
9799 the current trace snapshot.
9802 (@value{GDBP}) @b{trace 444}
9803 (@value{GDBP}) @b{actions}
9804 Enter actions for tracepoint #2, one per line:
9805 > collect $regs, $locals, $args, gdb_long_test
9808 (@value{GDBP}) @b{tstart}
9810 (@value{GDBP}) @b{tfind line 444}
9811 #0 gdb_test (p1=0x11, p2=0x22, p3=0x33, p4=0x44, p5=0x55, p6=0x66)
9813 444 printp( "%s: arguments = 0x%X 0x%X 0x%X 0x%X 0x%X 0x%X\n", )
9815 (@value{GDBP}) @b{tdump}
9816 Data collected at tracepoint 2, trace frame 1:
9817 d0 0xc4aa0085 -995491707
9821 d4 0x71aea3d 119204413
9826 a1 0x3000668 50333288
9829 a4 0x3000698 50333336
9831 fp 0x30bf3c 0x30bf3c
9832 sp 0x30bf34 0x30bf34
9834 pc 0x20b2c8 0x20b2c8
9838 p = 0x20e5b4 "gdb-test"
9845 gdb_long_test = 17 '\021'
9850 @node save-tracepoints
9851 @subsection @code{save-tracepoints @var{filename}}
9852 @kindex save-tracepoints
9853 @cindex save tracepoints for future sessions
9855 This command saves all current tracepoint definitions together with
9856 their actions and passcounts, into a file @file{@var{filename}}
9857 suitable for use in a later debugging session. To read the saved
9858 tracepoint definitions, use the @code{source} command (@pxref{Command
9861 @node Tracepoint Variables
9862 @section Convenience Variables for Tracepoints
9863 @cindex tracepoint variables
9864 @cindex convenience variables for tracepoints
9867 @vindex $trace_frame
9868 @item (int) $trace_frame
9869 The current trace snapshot (a.k.a.@: @dfn{frame}) number, or -1 if no
9870 snapshot is selected.
9873 @item (int) $tracepoint
9874 The tracepoint for the current trace snapshot.
9877 @item (int) $trace_line
9878 The line number for the current trace snapshot.
9881 @item (char []) $trace_file
9882 The source file for the current trace snapshot.
9885 @item (char []) $trace_func
9886 The name of the function containing @code{$tracepoint}.
9889 Note: @code{$trace_file} is not suitable for use in @code{printf},
9890 use @code{output} instead.
9892 Here's a simple example of using these convenience variables for
9893 stepping through all the trace snapshots and printing some of their
9897 (@value{GDBP}) @b{tfind start}
9899 (@value{GDBP}) @b{while $trace_frame != -1}
9900 > output $trace_file
9901 > printf ", line %d (tracepoint #%d)\n", $trace_line, $tracepoint
9907 @chapter Debugging Programs That Use Overlays
9910 If your program is too large to fit completely in your target system's
9911 memory, you can sometimes use @dfn{overlays} to work around this
9912 problem. @value{GDBN} provides some support for debugging programs that
9916 * How Overlays Work:: A general explanation of overlays.
9917 * Overlay Commands:: Managing overlays in @value{GDBN}.
9918 * Automatic Overlay Debugging:: @value{GDBN} can find out which overlays are
9919 mapped by asking the inferior.
9920 * Overlay Sample Program:: A sample program using overlays.
9923 @node How Overlays Work
9924 @section How Overlays Work
9925 @cindex mapped overlays
9926 @cindex unmapped overlays
9927 @cindex load address, overlay's
9928 @cindex mapped address
9929 @cindex overlay area
9931 Suppose you have a computer whose instruction address space is only 64
9932 kilobytes long, but which has much more memory which can be accessed by
9933 other means: special instructions, segment registers, or memory
9934 management hardware, for example. Suppose further that you want to
9935 adapt a program which is larger than 64 kilobytes to run on this system.
9937 One solution is to identify modules of your program which are relatively
9938 independent, and need not call each other directly; call these modules
9939 @dfn{overlays}. Separate the overlays from the main program, and place
9940 their machine code in the larger memory. Place your main program in
9941 instruction memory, but leave at least enough space there to hold the
9942 largest overlay as well.
9944 Now, to call a function located in an overlay, you must first copy that
9945 overlay's machine code from the large memory into the space set aside
9946 for it in the instruction memory, and then jump to its entry point
9949 @c NB: In the below the mapped area's size is greater or equal to the
9950 @c size of all overlays. This is intentional to remind the developer
9951 @c that overlays don't necessarily need to be the same size.
9955 Data Instruction Larger
9956 Address Space Address Space Address Space
9957 +-----------+ +-----------+ +-----------+
9959 +-----------+ +-----------+ +-----------+<-- overlay 1
9960 | program | | main | .----| overlay 1 | load address
9961 | variables | | program | | +-----------+
9962 | and heap | | | | | |
9963 +-----------+ | | | +-----------+<-- overlay 2
9964 | | +-----------+ | | | load address
9965 +-----------+ | | | .-| overlay 2 |
9967 mapped --->+-----------+ | | +-----------+
9969 | overlay | <-' | | |
9970 | area | <---' +-----------+<-- overlay 3
9971 | | <---. | | load address
9972 +-----------+ `--| overlay 3 |
9979 @anchor{A code overlay}A code overlay
9983 The diagram (@pxref{A code overlay}) shows a system with separate data
9984 and instruction address spaces. To map an overlay, the program copies
9985 its code from the larger address space to the instruction address space.
9986 Since the overlays shown here all use the same mapped address, only one
9987 may be mapped at a time. For a system with a single address space for
9988 data and instructions, the diagram would be similar, except that the
9989 program variables and heap would share an address space with the main
9990 program and the overlay area.
9992 An overlay loaded into instruction memory and ready for use is called a
9993 @dfn{mapped} overlay; its @dfn{mapped address} is its address in the
9994 instruction memory. An overlay not present (or only partially present)
9995 in instruction memory is called @dfn{unmapped}; its @dfn{load address}
9996 is its address in the larger memory. The mapped address is also called
9997 the @dfn{virtual memory address}, or @dfn{VMA}; the load address is also
9998 called the @dfn{load memory address}, or @dfn{LMA}.
10000 Unfortunately, overlays are not a completely transparent way to adapt a
10001 program to limited instruction memory. They introduce a new set of
10002 global constraints you must keep in mind as you design your program:
10007 Before calling or returning to a function in an overlay, your program
10008 must make sure that overlay is actually mapped. Otherwise, the call or
10009 return will transfer control to the right address, but in the wrong
10010 overlay, and your program will probably crash.
10013 If the process of mapping an overlay is expensive on your system, you
10014 will need to choose your overlays carefully to minimize their effect on
10015 your program's performance.
10018 The executable file you load onto your system must contain each
10019 overlay's instructions, appearing at the overlay's load address, not its
10020 mapped address. However, each overlay's instructions must be relocated
10021 and its symbols defined as if the overlay were at its mapped address.
10022 You can use GNU linker scripts to specify different load and relocation
10023 addresses for pieces of your program; see @ref{Overlay Description,,,
10024 ld.info, Using ld: the GNU linker}.
10027 The procedure for loading executable files onto your system must be able
10028 to load their contents into the larger address space as well as the
10029 instruction and data spaces.
10033 The overlay system described above is rather simple, and could be
10034 improved in many ways:
10039 If your system has suitable bank switch registers or memory management
10040 hardware, you could use those facilities to make an overlay's load area
10041 contents simply appear at their mapped address in instruction space.
10042 This would probably be faster than copying the overlay to its mapped
10043 area in the usual way.
10046 If your overlays are small enough, you could set aside more than one
10047 overlay area, and have more than one overlay mapped at a time.
10050 You can use overlays to manage data, as well as instructions. In
10051 general, data overlays are even less transparent to your design than
10052 code overlays: whereas code overlays only require care when you call or
10053 return to functions, data overlays require care every time you access
10054 the data. Also, if you change the contents of a data overlay, you
10055 must copy its contents back out to its load address before you can copy a
10056 different data overlay into the same mapped area.
10061 @node Overlay Commands
10062 @section Overlay Commands
10064 To use @value{GDBN}'s overlay support, each overlay in your program must
10065 correspond to a separate section of the executable file. The section's
10066 virtual memory address and load memory address must be the overlay's
10067 mapped and load addresses. Identifying overlays with sections allows
10068 @value{GDBN} to determine the appropriate address of a function or
10069 variable, depending on whether the overlay is mapped or not.
10071 @value{GDBN}'s overlay commands all start with the word @code{overlay};
10072 you can abbreviate this as @code{ov} or @code{ovly}. The commands are:
10077 Disable @value{GDBN}'s overlay support. When overlay support is
10078 disabled, @value{GDBN} assumes that all functions and variables are
10079 always present at their mapped addresses. By default, @value{GDBN}'s
10080 overlay support is disabled.
10082 @item overlay manual
10083 @cindex manual overlay debugging
10084 Enable @dfn{manual} overlay debugging. In this mode, @value{GDBN}
10085 relies on you to tell it which overlays are mapped, and which are not,
10086 using the @code{overlay map-overlay} and @code{overlay unmap-overlay}
10087 commands described below.
10089 @item overlay map-overlay @var{overlay}
10090 @itemx overlay map @var{overlay}
10091 @cindex map an overlay
10092 Tell @value{GDBN} that @var{overlay} is now mapped; @var{overlay} must
10093 be the name of the object file section containing the overlay. When an
10094 overlay is mapped, @value{GDBN} assumes it can find the overlay's
10095 functions and variables at their mapped addresses. @value{GDBN} assumes
10096 that any other overlays whose mapped ranges overlap that of
10097 @var{overlay} are now unmapped.
10099 @item overlay unmap-overlay @var{overlay}
10100 @itemx overlay unmap @var{overlay}
10101 @cindex unmap an overlay
10102 Tell @value{GDBN} that @var{overlay} is no longer mapped; @var{overlay}
10103 must be the name of the object file section containing the overlay.
10104 When an overlay is unmapped, @value{GDBN} assumes it can find the
10105 overlay's functions and variables at their load addresses.
10108 Enable @dfn{automatic} overlay debugging. In this mode, @value{GDBN}
10109 consults a data structure the overlay manager maintains in the inferior
10110 to see which overlays are mapped. For details, see @ref{Automatic
10111 Overlay Debugging}.
10113 @item overlay load-target
10114 @itemx overlay load
10115 @cindex reloading the overlay table
10116 Re-read the overlay table from the inferior. Normally, @value{GDBN}
10117 re-reads the table @value{GDBN} automatically each time the inferior
10118 stops, so this command should only be necessary if you have changed the
10119 overlay mapping yourself using @value{GDBN}. This command is only
10120 useful when using automatic overlay debugging.
10122 @item overlay list-overlays
10123 @itemx overlay list
10124 @cindex listing mapped overlays
10125 Display a list of the overlays currently mapped, along with their mapped
10126 addresses, load addresses, and sizes.
10130 Normally, when @value{GDBN} prints a code address, it includes the name
10131 of the function the address falls in:
10134 (@value{GDBP}) print main
10135 $3 = @{int ()@} 0x11a0 <main>
10138 When overlay debugging is enabled, @value{GDBN} recognizes code in
10139 unmapped overlays, and prints the names of unmapped functions with
10140 asterisks around them. For example, if @code{foo} is a function in an
10141 unmapped overlay, @value{GDBN} prints it this way:
10144 (@value{GDBP}) overlay list
10145 No sections are mapped.
10146 (@value{GDBP}) print foo
10147 $5 = @{int (int)@} 0x100000 <*foo*>
10150 When @code{foo}'s overlay is mapped, @value{GDBN} prints the function's
10154 (@value{GDBP}) overlay list
10155 Section .ov.foo.text, loaded at 0x100000 - 0x100034,
10156 mapped at 0x1016 - 0x104a
10157 (@value{GDBP}) print foo
10158 $6 = @{int (int)@} 0x1016 <foo>
10161 When overlay debugging is enabled, @value{GDBN} can find the correct
10162 address for functions and variables in an overlay, whether or not the
10163 overlay is mapped. This allows most @value{GDBN} commands, like
10164 @code{break} and @code{disassemble}, to work normally, even on unmapped
10165 code. However, @value{GDBN}'s breakpoint support has some limitations:
10169 @cindex breakpoints in overlays
10170 @cindex overlays, setting breakpoints in
10171 You can set breakpoints in functions in unmapped overlays, as long as
10172 @value{GDBN} can write to the overlay at its load address.
10174 @value{GDBN} can not set hardware or simulator-based breakpoints in
10175 unmapped overlays. However, if you set a breakpoint at the end of your
10176 overlay manager (and tell @value{GDBN} which overlays are now mapped, if
10177 you are using manual overlay management), @value{GDBN} will re-set its
10178 breakpoints properly.
10182 @node Automatic Overlay Debugging
10183 @section Automatic Overlay Debugging
10184 @cindex automatic overlay debugging
10186 @value{GDBN} can automatically track which overlays are mapped and which
10187 are not, given some simple co-operation from the overlay manager in the
10188 inferior. If you enable automatic overlay debugging with the
10189 @code{overlay auto} command (@pxref{Overlay Commands}), @value{GDBN}
10190 looks in the inferior's memory for certain variables describing the
10191 current state of the overlays.
10193 Here are the variables your overlay manager must define to support
10194 @value{GDBN}'s automatic overlay debugging:
10198 @item @code{_ovly_table}:
10199 This variable must be an array of the following structures:
10204 /* The overlay's mapped address. */
10207 /* The size of the overlay, in bytes. */
10208 unsigned long size;
10210 /* The overlay's load address. */
10213 /* Non-zero if the overlay is currently mapped;
10215 unsigned long mapped;
10219 @item @code{_novlys}:
10220 This variable must be a four-byte signed integer, holding the total
10221 number of elements in @code{_ovly_table}.
10225 To decide whether a particular overlay is mapped or not, @value{GDBN}
10226 looks for an entry in @w{@code{_ovly_table}} whose @code{vma} and
10227 @code{lma} members equal the VMA and LMA of the overlay's section in the
10228 executable file. When @value{GDBN} finds a matching entry, it consults
10229 the entry's @code{mapped} member to determine whether the overlay is
10232 In addition, your overlay manager may define a function called
10233 @code{_ovly_debug_event}. If this function is defined, @value{GDBN}
10234 will silently set a breakpoint there. If the overlay manager then
10235 calls this function whenever it has changed the overlay table, this
10236 will enable @value{GDBN} to accurately keep track of which overlays
10237 are in program memory, and update any breakpoints that may be set
10238 in overlays. This will allow breakpoints to work even if the
10239 overlays are kept in ROM or other non-writable memory while they
10240 are not being executed.
10242 @node Overlay Sample Program
10243 @section Overlay Sample Program
10244 @cindex overlay example program
10246 When linking a program which uses overlays, you must place the overlays
10247 at their load addresses, while relocating them to run at their mapped
10248 addresses. To do this, you must write a linker script (@pxref{Overlay
10249 Description,,, ld.info, Using ld: the GNU linker}). Unfortunately,
10250 since linker scripts are specific to a particular host system, target
10251 architecture, and target memory layout, this manual cannot provide
10252 portable sample code demonstrating @value{GDBN}'s overlay support.
10254 However, the @value{GDBN} source distribution does contain an overlaid
10255 program, with linker scripts for a few systems, as part of its test
10256 suite. The program consists of the following files from
10257 @file{gdb/testsuite/gdb.base}:
10261 The main program file.
10263 A simple overlay manager, used by @file{overlays.c}.
10268 Overlay modules, loaded and used by @file{overlays.c}.
10271 Linker scripts for linking the test program on the @code{d10v-elf}
10272 and @code{m32r-elf} targets.
10275 You can build the test program using the @code{d10v-elf} GCC
10276 cross-compiler like this:
10279 $ d10v-elf-gcc -g -c overlays.c
10280 $ d10v-elf-gcc -g -c ovlymgr.c
10281 $ d10v-elf-gcc -g -c foo.c
10282 $ d10v-elf-gcc -g -c bar.c
10283 $ d10v-elf-gcc -g -c baz.c
10284 $ d10v-elf-gcc -g -c grbx.c
10285 $ d10v-elf-gcc -g overlays.o ovlymgr.o foo.o bar.o \
10286 baz.o grbx.o -Wl,-Td10v.ld -o overlays
10289 The build process is identical for any other architecture, except that
10290 you must substitute the appropriate compiler and linker script for the
10291 target system for @code{d10v-elf-gcc} and @code{d10v.ld}.
10295 @chapter Using @value{GDBN} with Different Languages
10298 Although programming languages generally have common aspects, they are
10299 rarely expressed in the same manner. For instance, in ANSI C,
10300 dereferencing a pointer @code{p} is accomplished by @code{*p}, but in
10301 Modula-2, it is accomplished by @code{p^}. Values can also be
10302 represented (and displayed) differently. Hex numbers in C appear as
10303 @samp{0x1ae}, while in Modula-2 they appear as @samp{1AEH}.
10305 @cindex working language
10306 Language-specific information is built into @value{GDBN} for some languages,
10307 allowing you to express operations like the above in your program's
10308 native language, and allowing @value{GDBN} to output values in a manner
10309 consistent with the syntax of your program's native language. The
10310 language you use to build expressions is called the @dfn{working
10314 * Setting:: Switching between source languages
10315 * Show:: Displaying the language
10316 * Checks:: Type and range checks
10317 * Supported Languages:: Supported languages
10318 * Unsupported Languages:: Unsupported languages
10322 @section Switching Between Source Languages
10324 There are two ways to control the working language---either have @value{GDBN}
10325 set it automatically, or select it manually yourself. You can use the
10326 @code{set language} command for either purpose. On startup, @value{GDBN}
10327 defaults to setting the language automatically. The working language is
10328 used to determine how expressions you type are interpreted, how values
10331 In addition to the working language, every source file that
10332 @value{GDBN} knows about has its own working language. For some object
10333 file formats, the compiler might indicate which language a particular
10334 source file is in. However, most of the time @value{GDBN} infers the
10335 language from the name of the file. The language of a source file
10336 controls whether C@t{++} names are demangled---this way @code{backtrace} can
10337 show each frame appropriately for its own language. There is no way to
10338 set the language of a source file from within @value{GDBN}, but you can
10339 set the language associated with a filename extension. @xref{Show, ,
10340 Displaying the Language}.
10342 This is most commonly a problem when you use a program, such
10343 as @code{cfront} or @code{f2c}, that generates C but is written in
10344 another language. In that case, make the
10345 program use @code{#line} directives in its C output; that way
10346 @value{GDBN} will know the correct language of the source code of the original
10347 program, and will display that source code, not the generated C code.
10350 * Filenames:: Filename extensions and languages.
10351 * Manually:: Setting the working language manually
10352 * Automatically:: Having @value{GDBN} infer the source language
10356 @subsection List of Filename Extensions and Languages
10358 If a source file name ends in one of the following extensions, then
10359 @value{GDBN} infers that its language is the one indicated.
10377 C@t{++} source file
10380 Objective-C source file
10384 Fortran source file
10387 Modula-2 source file
10391 Assembler source file. This actually behaves almost like C, but
10392 @value{GDBN} does not skip over function prologues when stepping.
10395 In addition, you may set the language associated with a filename
10396 extension. @xref{Show, , Displaying the Language}.
10399 @subsection Setting the Working Language
10401 If you allow @value{GDBN} to set the language automatically,
10402 expressions are interpreted the same way in your debugging session and
10405 @kindex set language
10406 If you wish, you may set the language manually. To do this, issue the
10407 command @samp{set language @var{lang}}, where @var{lang} is the name of
10408 a language, such as
10409 @code{c} or @code{modula-2}.
10410 For a list of the supported languages, type @samp{set language}.
10412 Setting the language manually prevents @value{GDBN} from updating the working
10413 language automatically. This can lead to confusion if you try
10414 to debug a program when the working language is not the same as the
10415 source language, when an expression is acceptable to both
10416 languages---but means different things. For instance, if the current
10417 source file were written in C, and @value{GDBN} was parsing Modula-2, a
10425 might not have the effect you intended. In C, this means to add
10426 @code{b} and @code{c} and place the result in @code{a}. The result
10427 printed would be the value of @code{a}. In Modula-2, this means to compare
10428 @code{a} to the result of @code{b+c}, yielding a @code{BOOLEAN} value.
10430 @node Automatically
10431 @subsection Having @value{GDBN} Infer the Source Language
10433 To have @value{GDBN} set the working language automatically, use
10434 @samp{set language local} or @samp{set language auto}. @value{GDBN}
10435 then infers the working language. That is, when your program stops in a
10436 frame (usually by encountering a breakpoint), @value{GDBN} sets the
10437 working language to the language recorded for the function in that
10438 frame. If the language for a frame is unknown (that is, if the function
10439 or block corresponding to the frame was defined in a source file that
10440 does not have a recognized extension), the current working language is
10441 not changed, and @value{GDBN} issues a warning.
10443 This may not seem necessary for most programs, which are written
10444 entirely in one source language. However, program modules and libraries
10445 written in one source language can be used by a main program written in
10446 a different source language. Using @samp{set language auto} in this
10447 case frees you from having to set the working language manually.
10450 @section Displaying the Language
10452 The following commands help you find out which language is the
10453 working language, and also what language source files were written in.
10456 @item show language
10457 @kindex show language
10458 Display the current working language. This is the
10459 language you can use with commands such as @code{print} to
10460 build and compute expressions that may involve variables in your program.
10463 @kindex info frame@r{, show the source language}
10464 Display the source language for this frame. This language becomes the
10465 working language if you use an identifier from this frame.
10466 @xref{Frame Info, ,Information about a Frame}, to identify the other
10467 information listed here.
10470 @kindex info source@r{, show the source language}
10471 Display the source language of this source file.
10472 @xref{Symbols, ,Examining the Symbol Table}, to identify the other
10473 information listed here.
10476 In unusual circumstances, you may have source files with extensions
10477 not in the standard list. You can then set the extension associated
10478 with a language explicitly:
10481 @item set extension-language @var{ext} @var{language}
10482 @kindex set extension-language
10483 Tell @value{GDBN} that source files with extension @var{ext} are to be
10484 assumed as written in the source language @var{language}.
10486 @item info extensions
10487 @kindex info extensions
10488 List all the filename extensions and the associated languages.
10492 @section Type and Range Checking
10495 @emph{Warning:} In this release, the @value{GDBN} commands for type and range
10496 checking are included, but they do not yet have any effect. This
10497 section documents the intended facilities.
10499 @c FIXME remove warning when type/range code added
10501 Some languages are designed to guard you against making seemingly common
10502 errors through a series of compile- and run-time checks. These include
10503 checking the type of arguments to functions and operators, and making
10504 sure mathematical overflows are caught at run time. Checks such as
10505 these help to ensure a program's correctness once it has been compiled
10506 by eliminating type mismatches, and providing active checks for range
10507 errors when your program is running.
10509 @value{GDBN} can check for conditions like the above if you wish.
10510 Although @value{GDBN} does not check the statements in your program,
10511 it can check expressions entered directly into @value{GDBN} for
10512 evaluation via the @code{print} command, for example. As with the
10513 working language, @value{GDBN} can also decide whether or not to check
10514 automatically based on your program's source language.
10515 @xref{Supported Languages, ,Supported Languages}, for the default
10516 settings of supported languages.
10519 * Type Checking:: An overview of type checking
10520 * Range Checking:: An overview of range checking
10523 @cindex type checking
10524 @cindex checks, type
10525 @node Type Checking
10526 @subsection An Overview of Type Checking
10528 Some languages, such as Modula-2, are strongly typed, meaning that the
10529 arguments to operators and functions have to be of the correct type,
10530 otherwise an error occurs. These checks prevent type mismatch
10531 errors from ever causing any run-time problems. For example,
10539 The second example fails because the @code{CARDINAL} 1 is not
10540 type-compatible with the @code{REAL} 2.3.
10542 For the expressions you use in @value{GDBN} commands, you can tell the
10543 @value{GDBN} type checker to skip checking;
10544 to treat any mismatches as errors and abandon the expression;
10545 or to only issue warnings when type mismatches occur,
10546 but evaluate the expression anyway. When you choose the last of
10547 these, @value{GDBN} evaluates expressions like the second example above, but
10548 also issues a warning.
10550 Even if you turn type checking off, there may be other reasons
10551 related to type that prevent @value{GDBN} from evaluating an expression.
10552 For instance, @value{GDBN} does not know how to add an @code{int} and
10553 a @code{struct foo}. These particular type errors have nothing to do
10554 with the language in use, and usually arise from expressions, such as
10555 the one described above, which make little sense to evaluate anyway.
10557 Each language defines to what degree it is strict about type. For
10558 instance, both Modula-2 and C require the arguments to arithmetical
10559 operators to be numbers. In C, enumerated types and pointers can be
10560 represented as numbers, so that they are valid arguments to mathematical
10561 operators. @xref{Supported Languages, ,Supported Languages}, for further
10562 details on specific languages.
10564 @value{GDBN} provides some additional commands for controlling the type checker:
10566 @kindex set check type
10567 @kindex show check type
10569 @item set check type auto
10570 Set type checking on or off based on the current working language.
10571 @xref{Supported Languages, ,Supported Languages}, for the default settings for
10574 @item set check type on
10575 @itemx set check type off
10576 Set type checking on or off, overriding the default setting for the
10577 current working language. Issue a warning if the setting does not
10578 match the language default. If any type mismatches occur in
10579 evaluating an expression while type checking is on, @value{GDBN} prints a
10580 message and aborts evaluation of the expression.
10582 @item set check type warn
10583 Cause the type checker to issue warnings, but to always attempt to
10584 evaluate the expression. Evaluating the expression may still
10585 be impossible for other reasons. For example, @value{GDBN} cannot add
10586 numbers and structures.
10589 Show the current setting of the type checker, and whether or not @value{GDBN}
10590 is setting it automatically.
10593 @cindex range checking
10594 @cindex checks, range
10595 @node Range Checking
10596 @subsection An Overview of Range Checking
10598 In some languages (such as Modula-2), it is an error to exceed the
10599 bounds of a type; this is enforced with run-time checks. Such range
10600 checking is meant to ensure program correctness by making sure
10601 computations do not overflow, or indices on an array element access do
10602 not exceed the bounds of the array.
10604 For expressions you use in @value{GDBN} commands, you can tell
10605 @value{GDBN} to treat range errors in one of three ways: ignore them,
10606 always treat them as errors and abandon the expression, or issue
10607 warnings but evaluate the expression anyway.
10609 A range error can result from numerical overflow, from exceeding an
10610 array index bound, or when you type a constant that is not a member
10611 of any type. Some languages, however, do not treat overflows as an
10612 error. In many implementations of C, mathematical overflow causes the
10613 result to ``wrap around'' to lower values---for example, if @var{m} is
10614 the largest integer value, and @var{s} is the smallest, then
10617 @var{m} + 1 @result{} @var{s}
10620 This, too, is specific to individual languages, and in some cases
10621 specific to individual compilers or machines. @xref{Supported Languages, ,
10622 Supported Languages}, for further details on specific languages.
10624 @value{GDBN} provides some additional commands for controlling the range checker:
10626 @kindex set check range
10627 @kindex show check range
10629 @item set check range auto
10630 Set range checking on or off based on the current working language.
10631 @xref{Supported Languages, ,Supported Languages}, for the default settings for
10634 @item set check range on
10635 @itemx set check range off
10636 Set range checking on or off, overriding the default setting for the
10637 current working language. A warning is issued if the setting does not
10638 match the language default. If a range error occurs and range checking is on,
10639 then a message is printed and evaluation of the expression is aborted.
10641 @item set check range warn
10642 Output messages when the @value{GDBN} range checker detects a range error,
10643 but attempt to evaluate the expression anyway. Evaluating the
10644 expression may still be impossible for other reasons, such as accessing
10645 memory that the process does not own (a typical example from many Unix
10649 Show the current setting of the range checker, and whether or not it is
10650 being set automatically by @value{GDBN}.
10653 @node Supported Languages
10654 @section Supported Languages
10656 @value{GDBN} supports C, C@t{++}, Objective-C, Fortran, Java, Pascal,
10657 assembly, Modula-2, and Ada.
10658 @c This is false ...
10659 Some @value{GDBN} features may be used in expressions regardless of the
10660 language you use: the @value{GDBN} @code{@@} and @code{::} operators,
10661 and the @samp{@{type@}addr} construct (@pxref{Expressions,
10662 ,Expressions}) can be used with the constructs of any supported
10665 The following sections detail to what degree each source language is
10666 supported by @value{GDBN}. These sections are not meant to be language
10667 tutorials or references, but serve only as a reference guide to what the
10668 @value{GDBN} expression parser accepts, and what input and output
10669 formats should look like for different languages. There are many good
10670 books written on each of these languages; please look to these for a
10671 language reference or tutorial.
10674 * C:: C and C@t{++}
10675 * Objective-C:: Objective-C
10676 * Fortran:: Fortran
10678 * Modula-2:: Modula-2
10683 @subsection C and C@t{++}
10685 @cindex C and C@t{++}
10686 @cindex expressions in C or C@t{++}
10688 Since C and C@t{++} are so closely related, many features of @value{GDBN} apply
10689 to both languages. Whenever this is the case, we discuss those languages
10693 @cindex @code{g++}, @sc{gnu} C@t{++} compiler
10694 @cindex @sc{gnu} C@t{++}
10695 The C@t{++} debugging facilities are jointly implemented by the C@t{++}
10696 compiler and @value{GDBN}. Therefore, to debug your C@t{++} code
10697 effectively, you must compile your C@t{++} programs with a supported
10698 C@t{++} compiler, such as @sc{gnu} @code{g++}, or the HP ANSI C@t{++}
10699 compiler (@code{aCC}).
10701 For best results when using @sc{gnu} C@t{++}, use the DWARF 2 debugging
10702 format; if it doesn't work on your system, try the stabs+ debugging
10703 format. You can select those formats explicitly with the @code{g++}
10704 command-line options @option{-gdwarf-2} and @option{-gstabs+}.
10705 @xref{Debugging Options,,Options for Debugging Your Program or GCC,
10706 gcc.info, Using the @sc{gnu} Compiler Collection (GCC)}.
10709 * C Operators:: C and C@t{++} operators
10710 * C Constants:: C and C@t{++} constants
10711 * C Plus Plus Expressions:: C@t{++} expressions
10712 * C Defaults:: Default settings for C and C@t{++}
10713 * C Checks:: C and C@t{++} type and range checks
10714 * Debugging C:: @value{GDBN} and C
10715 * Debugging C Plus Plus:: @value{GDBN} features for C@t{++}
10716 * Decimal Floating Point:: Numbers in Decimal Floating Point format
10720 @subsubsection C and C@t{++} Operators
10722 @cindex C and C@t{++} operators
10724 Operators must be defined on values of specific types. For instance,
10725 @code{+} is defined on numbers, but not on structures. Operators are
10726 often defined on groups of types.
10728 For the purposes of C and C@t{++}, the following definitions hold:
10733 @emph{Integral types} include @code{int} with any of its storage-class
10734 specifiers; @code{char}; @code{enum}; and, for C@t{++}, @code{bool}.
10737 @emph{Floating-point types} include @code{float}, @code{double}, and
10738 @code{long double} (if supported by the target platform).
10741 @emph{Pointer types} include all types defined as @code{(@var{type} *)}.
10744 @emph{Scalar types} include all of the above.
10749 The following operators are supported. They are listed here
10750 in order of increasing precedence:
10754 The comma or sequencing operator. Expressions in a comma-separated list
10755 are evaluated from left to right, with the result of the entire
10756 expression being the last expression evaluated.
10759 Assignment. The value of an assignment expression is the value
10760 assigned. Defined on scalar types.
10763 Used in an expression of the form @w{@code{@var{a} @var{op}= @var{b}}},
10764 and translated to @w{@code{@var{a} = @var{a op b}}}.
10765 @w{@code{@var{op}=}} and @code{=} have the same precedence.
10766 @var{op} is any one of the operators @code{|}, @code{^}, @code{&},
10767 @code{<<}, @code{>>}, @code{+}, @code{-}, @code{*}, @code{/}, @code{%}.
10770 The ternary operator. @code{@var{a} ? @var{b} : @var{c}} can be thought
10771 of as: if @var{a} then @var{b} else @var{c}. @var{a} should be of an
10775 Logical @sc{or}. Defined on integral types.
10778 Logical @sc{and}. Defined on integral types.
10781 Bitwise @sc{or}. Defined on integral types.
10784 Bitwise exclusive-@sc{or}. Defined on integral types.
10787 Bitwise @sc{and}. Defined on integral types.
10790 Equality and inequality. Defined on scalar types. The value of these
10791 expressions is 0 for false and non-zero for true.
10793 @item <@r{, }>@r{, }<=@r{, }>=
10794 Less than, greater than, less than or equal, greater than or equal.
10795 Defined on scalar types. The value of these expressions is 0 for false
10796 and non-zero for true.
10799 left shift, and right shift. Defined on integral types.
10802 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
10805 Addition and subtraction. Defined on integral types, floating-point types and
10808 @item *@r{, }/@r{, }%
10809 Multiplication, division, and modulus. Multiplication and division are
10810 defined on integral and floating-point types. Modulus is defined on
10814 Increment and decrement. When appearing before a variable, the
10815 operation is performed before the variable is used in an expression;
10816 when appearing after it, the variable's value is used before the
10817 operation takes place.
10820 Pointer dereferencing. Defined on pointer types. Same precedence as
10824 Address operator. Defined on variables. Same precedence as @code{++}.
10826 For debugging C@t{++}, @value{GDBN} implements a use of @samp{&} beyond what is
10827 allowed in the C@t{++} language itself: you can use @samp{&(&@var{ref})}
10828 to examine the address
10829 where a C@t{++} reference variable (declared with @samp{&@var{ref}}) is
10833 Negative. Defined on integral and floating-point types. Same
10834 precedence as @code{++}.
10837 Logical negation. Defined on integral types. Same precedence as
10841 Bitwise complement operator. Defined on integral types. Same precedence as
10846 Structure member, and pointer-to-structure member. For convenience,
10847 @value{GDBN} regards the two as equivalent, choosing whether to dereference a
10848 pointer based on the stored type information.
10849 Defined on @code{struct} and @code{union} data.
10852 Dereferences of pointers to members.
10855 Array indexing. @code{@var{a}[@var{i}]} is defined as
10856 @code{*(@var{a}+@var{i})}. Same precedence as @code{->}.
10859 Function parameter list. Same precedence as @code{->}.
10862 C@t{++} scope resolution operator. Defined on @code{struct}, @code{union},
10863 and @code{class} types.
10866 Doubled colons also represent the @value{GDBN} scope operator
10867 (@pxref{Expressions, ,Expressions}). Same precedence as @code{::},
10871 If an operator is redefined in the user code, @value{GDBN} usually
10872 attempts to invoke the redefined version instead of using the operator's
10873 predefined meaning.
10876 @subsubsection C and C@t{++} Constants
10878 @cindex C and C@t{++} constants
10880 @value{GDBN} allows you to express the constants of C and C@t{++} in the
10885 Integer constants are a sequence of digits. Octal constants are
10886 specified by a leading @samp{0} (i.e.@: zero), and hexadecimal constants
10887 by a leading @samp{0x} or @samp{0X}. Constants may also end with a letter
10888 @samp{l}, specifying that the constant should be treated as a
10892 Floating point constants are a sequence of digits, followed by a decimal
10893 point, followed by a sequence of digits, and optionally followed by an
10894 exponent. An exponent is of the form:
10895 @samp{@w{e@r{[[}+@r{]|}-@r{]}@var{nnn}}}, where @var{nnn} is another
10896 sequence of digits. The @samp{+} is optional for positive exponents.
10897 A floating-point constant may also end with a letter @samp{f} or
10898 @samp{F}, specifying that the constant should be treated as being of
10899 the @code{float} (as opposed to the default @code{double}) type; or with
10900 a letter @samp{l} or @samp{L}, which specifies a @code{long double}
10904 Enumerated constants consist of enumerated identifiers, or their
10905 integral equivalents.
10908 Character constants are a single character surrounded by single quotes
10909 (@code{'}), or a number---the ordinal value of the corresponding character
10910 (usually its @sc{ascii} value). Within quotes, the single character may
10911 be represented by a letter or by @dfn{escape sequences}, which are of
10912 the form @samp{\@var{nnn}}, where @var{nnn} is the octal representation
10913 of the character's ordinal value; or of the form @samp{\@var{x}}, where
10914 @samp{@var{x}} is a predefined special character---for example,
10915 @samp{\n} for newline.
10918 String constants are a sequence of character constants surrounded by
10919 double quotes (@code{"}). Any valid character constant (as described
10920 above) may appear. Double quotes within the string must be preceded by
10921 a backslash, so for instance @samp{"a\"b'c"} is a string of five
10925 Pointer constants are an integral value. You can also write pointers
10926 to constants using the C operator @samp{&}.
10929 Array constants are comma-separated lists surrounded by braces @samp{@{}
10930 and @samp{@}}; for example, @samp{@{1,2,3@}} is a three-element array of
10931 integers, @samp{@{@{1,2@}, @{3,4@}, @{5,6@}@}} is a three-by-two array,
10932 and @samp{@{&"hi", &"there", &"fred"@}} is a three-element array of pointers.
10935 @node C Plus Plus Expressions
10936 @subsubsection C@t{++} Expressions
10938 @cindex expressions in C@t{++}
10939 @value{GDBN} expression handling can interpret most C@t{++} expressions.
10941 @cindex debugging C@t{++} programs
10942 @cindex C@t{++} compilers
10943 @cindex debug formats and C@t{++}
10944 @cindex @value{NGCC} and C@t{++}
10946 @emph{Warning:} @value{GDBN} can only debug C@t{++} code if you use the
10947 proper compiler and the proper debug format. Currently, @value{GDBN}
10948 works best when debugging C@t{++} code that is compiled with
10949 @value{NGCC} 2.95.3 or with @value{NGCC} 3.1 or newer, using the options
10950 @option{-gdwarf-2} or @option{-gstabs+}. DWARF 2 is preferred over
10951 stabs+. Most configurations of @value{NGCC} emit either DWARF 2 or
10952 stabs+ as their default debug format, so you usually don't need to
10953 specify a debug format explicitly. Other compilers and/or debug formats
10954 are likely to work badly or not at all when using @value{GDBN} to debug
10960 @cindex member functions
10962 Member function calls are allowed; you can use expressions like
10965 count = aml->GetOriginal(x, y)
10968 @vindex this@r{, inside C@t{++} member functions}
10969 @cindex namespace in C@t{++}
10971 While a member function is active (in the selected stack frame), your
10972 expressions have the same namespace available as the member function;
10973 that is, @value{GDBN} allows implicit references to the class instance
10974 pointer @code{this} following the same rules as C@t{++}.
10976 @cindex call overloaded functions
10977 @cindex overloaded functions, calling
10978 @cindex type conversions in C@t{++}
10980 You can call overloaded functions; @value{GDBN} resolves the function
10981 call to the right definition, with some restrictions. @value{GDBN} does not
10982 perform overload resolution involving user-defined type conversions,
10983 calls to constructors, or instantiations of templates that do not exist
10984 in the program. It also cannot handle ellipsis argument lists or
10987 It does perform integral conversions and promotions, floating-point
10988 promotions, arithmetic conversions, pointer conversions, conversions of
10989 class objects to base classes, and standard conversions such as those of
10990 functions or arrays to pointers; it requires an exact match on the
10991 number of function arguments.
10993 Overload resolution is always performed, unless you have specified
10994 @code{set overload-resolution off}. @xref{Debugging C Plus Plus,
10995 ,@value{GDBN} Features for C@t{++}}.
10997 You must specify @code{set overload-resolution off} in order to use an
10998 explicit function signature to call an overloaded function, as in
11000 p 'foo(char,int)'('x', 13)
11003 The @value{GDBN} command-completion facility can simplify this;
11004 see @ref{Completion, ,Command Completion}.
11006 @cindex reference declarations
11008 @value{GDBN} understands variables declared as C@t{++} references; you can use
11009 them in expressions just as you do in C@t{++} source---they are automatically
11012 In the parameter list shown when @value{GDBN} displays a frame, the values of
11013 reference variables are not displayed (unlike other variables); this
11014 avoids clutter, since references are often used for large structures.
11015 The @emph{address} of a reference variable is always shown, unless
11016 you have specified @samp{set print address off}.
11019 @value{GDBN} supports the C@t{++} name resolution operator @code{::}---your
11020 expressions can use it just as expressions in your program do. Since
11021 one scope may be defined in another, you can use @code{::} repeatedly if
11022 necessary, for example in an expression like
11023 @samp{@var{scope1}::@var{scope2}::@var{name}}. @value{GDBN} also allows
11024 resolving name scope by reference to source files, in both C and C@t{++}
11025 debugging (@pxref{Variables, ,Program Variables}).
11028 In addition, when used with HP's C@t{++} compiler, @value{GDBN} supports
11029 calling virtual functions correctly, printing out virtual bases of
11030 objects, calling functions in a base subobject, casting objects, and
11031 invoking user-defined operators.
11034 @subsubsection C and C@t{++} Defaults
11036 @cindex C and C@t{++} defaults
11038 If you allow @value{GDBN} to set type and range checking automatically, they
11039 both default to @code{off} whenever the working language changes to
11040 C or C@t{++}. This happens regardless of whether you or @value{GDBN}
11041 selects the working language.
11043 If you allow @value{GDBN} to set the language automatically, it
11044 recognizes source files whose names end with @file{.c}, @file{.C}, or
11045 @file{.cc}, etc, and when @value{GDBN} enters code compiled from one of
11046 these files, it sets the working language to C or C@t{++}.
11047 @xref{Automatically, ,Having @value{GDBN} Infer the Source Language},
11048 for further details.
11050 @c Type checking is (a) primarily motivated by Modula-2, and (b)
11051 @c unimplemented. If (b) changes, it might make sense to let this node
11052 @c appear even if Mod-2 does not, but meanwhile ignore it. roland 16jul93.
11055 @subsubsection C and C@t{++} Type and Range Checks
11057 @cindex C and C@t{++} checks
11059 By default, when @value{GDBN} parses C or C@t{++} expressions, type checking
11060 is not used. However, if you turn type checking on, @value{GDBN}
11061 considers two variables type equivalent if:
11065 The two variables are structured and have the same structure, union, or
11069 The two variables have the same type name, or types that have been
11070 declared equivalent through @code{typedef}.
11073 @c leaving this out because neither J Gilmore nor R Pesch understand it.
11076 The two @code{struct}, @code{union}, or @code{enum} variables are
11077 declared in the same declaration. (Note: this may not be true for all C
11082 Range checking, if turned on, is done on mathematical operations. Array
11083 indices are not checked, since they are often used to index a pointer
11084 that is not itself an array.
11087 @subsubsection @value{GDBN} and C
11089 The @code{set print union} and @code{show print union} commands apply to
11090 the @code{union} type. When set to @samp{on}, any @code{union} that is
11091 inside a @code{struct} or @code{class} is also printed. Otherwise, it
11092 appears as @samp{@{...@}}.
11094 The @code{@@} operator aids in the debugging of dynamic arrays, formed
11095 with pointers and a memory allocation function. @xref{Expressions,
11098 @node Debugging C Plus Plus
11099 @subsubsection @value{GDBN} Features for C@t{++}
11101 @cindex commands for C@t{++}
11103 Some @value{GDBN} commands are particularly useful with C@t{++}, and some are
11104 designed specifically for use with C@t{++}. Here is a summary:
11107 @cindex break in overloaded functions
11108 @item @r{breakpoint menus}
11109 When you want a breakpoint in a function whose name is overloaded,
11110 @value{GDBN} has the capability to display a menu of possible breakpoint
11111 locations to help you specify which function definition you want.
11112 @xref{Ambiguous Expressions,,Ambiguous Expressions}.
11114 @cindex overloading in C@t{++}
11115 @item rbreak @var{regex}
11116 Setting breakpoints using regular expressions is helpful for setting
11117 breakpoints on overloaded functions that are not members of any special
11119 @xref{Set Breaks, ,Setting Breakpoints}.
11121 @cindex C@t{++} exception handling
11124 Debug C@t{++} exception handling using these commands. @xref{Set
11125 Catchpoints, , Setting Catchpoints}.
11127 @cindex inheritance
11128 @item ptype @var{typename}
11129 Print inheritance relationships as well as other information for type
11131 @xref{Symbols, ,Examining the Symbol Table}.
11133 @cindex C@t{++} symbol display
11134 @item set print demangle
11135 @itemx show print demangle
11136 @itemx set print asm-demangle
11137 @itemx show print asm-demangle
11138 Control whether C@t{++} symbols display in their source form, both when
11139 displaying code as C@t{++} source and when displaying disassemblies.
11140 @xref{Print Settings, ,Print Settings}.
11142 @item set print object
11143 @itemx show print object
11144 Choose whether to print derived (actual) or declared types of objects.
11145 @xref{Print Settings, ,Print Settings}.
11147 @item set print vtbl
11148 @itemx show print vtbl
11149 Control the format for printing virtual function tables.
11150 @xref{Print Settings, ,Print Settings}.
11151 (The @code{vtbl} commands do not work on programs compiled with the HP
11152 ANSI C@t{++} compiler (@code{aCC}).)
11154 @kindex set overload-resolution
11155 @cindex overloaded functions, overload resolution
11156 @item set overload-resolution on
11157 Enable overload resolution for C@t{++} expression evaluation. The default
11158 is on. For overloaded functions, @value{GDBN} evaluates the arguments
11159 and searches for a function whose signature matches the argument types,
11160 using the standard C@t{++} conversion rules (see @ref{C Plus Plus
11161 Expressions, ,C@t{++} Expressions}, for details).
11162 If it cannot find a match, it emits a message.
11164 @item set overload-resolution off
11165 Disable overload resolution for C@t{++} expression evaluation. For
11166 overloaded functions that are not class member functions, @value{GDBN}
11167 chooses the first function of the specified name that it finds in the
11168 symbol table, whether or not its arguments are of the correct type. For
11169 overloaded functions that are class member functions, @value{GDBN}
11170 searches for a function whose signature @emph{exactly} matches the
11173 @kindex show overload-resolution
11174 @item show overload-resolution
11175 Show the current setting of overload resolution.
11177 @item @r{Overloaded symbol names}
11178 You can specify a particular definition of an overloaded symbol, using
11179 the same notation that is used to declare such symbols in C@t{++}: type
11180 @code{@var{symbol}(@var{types})} rather than just @var{symbol}. You can
11181 also use the @value{GDBN} command-line word completion facilities to list the
11182 available choices, or to finish the type list for you.
11183 @xref{Completion,, Command Completion}, for details on how to do this.
11186 @node Decimal Floating Point
11187 @subsubsection Decimal Floating Point format
11188 @cindex decimal floating point format
11190 @value{GDBN} can examine, set and perform computations with numbers in
11191 decimal floating point format, which in the C language correspond to the
11192 @code{_Decimal32}, @code{_Decimal64} and @code{_Decimal128} types as
11193 specified by the extension to support decimal floating-point arithmetic.
11195 There are two encodings in use, depending on the architecture: BID (Binary
11196 Integer Decimal) for x86 and x86-64, and DPD (Densely Packed Decimal) for
11197 PowerPC. @value{GDBN} will use the appropriate encoding for the configured
11200 Because of a limitation in @file{libdecnumber}, the library used by @value{GDBN}
11201 to manipulate decimal floating point numbers, it is not possible to convert
11202 (using a cast, for example) integers wider than 32-bit to decimal float.
11204 In addition, in order to imitate @value{GDBN}'s behaviour with binary floating
11205 point computations, error checking in decimal float operations ignores
11206 underflow, overflow and divide by zero exceptions.
11208 In the PowerPC architecture, @value{GDBN} provides a set of pseudo-registers
11209 to inspect @code{_Decimal128} values stored in floating point registers.
11210 See @ref{PowerPC,,PowerPC} for more details.
11213 @subsection Objective-C
11215 @cindex Objective-C
11216 This section provides information about some commands and command
11217 options that are useful for debugging Objective-C code. See also
11218 @ref{Symbols, info classes}, and @ref{Symbols, info selectors}, for a
11219 few more commands specific to Objective-C support.
11222 * Method Names in Commands::
11223 * The Print Command with Objective-C::
11226 @node Method Names in Commands
11227 @subsubsection Method Names in Commands
11229 The following commands have been extended to accept Objective-C method
11230 names as line specifications:
11232 @kindex clear@r{, and Objective-C}
11233 @kindex break@r{, and Objective-C}
11234 @kindex info line@r{, and Objective-C}
11235 @kindex jump@r{, and Objective-C}
11236 @kindex list@r{, and Objective-C}
11240 @item @code{info line}
11245 A fully qualified Objective-C method name is specified as
11248 -[@var{Class} @var{methodName}]
11251 where the minus sign is used to indicate an instance method and a
11252 plus sign (not shown) is used to indicate a class method. The class
11253 name @var{Class} and method name @var{methodName} are enclosed in
11254 brackets, similar to the way messages are specified in Objective-C
11255 source code. For example, to set a breakpoint at the @code{create}
11256 instance method of class @code{Fruit} in the program currently being
11260 break -[Fruit create]
11263 To list ten program lines around the @code{initialize} class method,
11267 list +[NSText initialize]
11270 In the current version of @value{GDBN}, the plus or minus sign is
11271 required. In future versions of @value{GDBN}, the plus or minus
11272 sign will be optional, but you can use it to narrow the search. It
11273 is also possible to specify just a method name:
11279 You must specify the complete method name, including any colons. If
11280 your program's source files contain more than one @code{create} method,
11281 you'll be presented with a numbered list of classes that implement that
11282 method. Indicate your choice by number, or type @samp{0} to exit if
11285 As another example, to clear a breakpoint established at the
11286 @code{makeKeyAndOrderFront:} method of the @code{NSWindow} class, enter:
11289 clear -[NSWindow makeKeyAndOrderFront:]
11292 @node The Print Command with Objective-C
11293 @subsubsection The Print Command With Objective-C
11294 @cindex Objective-C, print objects
11295 @kindex print-object
11296 @kindex po @r{(@code{print-object})}
11298 The print command has also been extended to accept methods. For example:
11301 print -[@var{object} hash]
11304 @cindex print an Objective-C object description
11305 @cindex @code{_NSPrintForDebugger}, and printing Objective-C objects
11307 will tell @value{GDBN} to send the @code{hash} message to @var{object}
11308 and print the result. Also, an additional command has been added,
11309 @code{print-object} or @code{po} for short, which is meant to print
11310 the description of an object. However, this command may only work
11311 with certain Objective-C libraries that have a particular hook
11312 function, @code{_NSPrintForDebugger}, defined.
11315 @subsection Fortran
11316 @cindex Fortran-specific support in @value{GDBN}
11318 @value{GDBN} can be used to debug programs written in Fortran, but it
11319 currently supports only the features of Fortran 77 language.
11321 @cindex trailing underscore, in Fortran symbols
11322 Some Fortran compilers (@sc{gnu} Fortran 77 and Fortran 95 compilers
11323 among them) append an underscore to the names of variables and
11324 functions. When you debug programs compiled by those compilers, you
11325 will need to refer to variables and functions with a trailing
11329 * Fortran Operators:: Fortran operators and expressions
11330 * Fortran Defaults:: Default settings for Fortran
11331 * Special Fortran Commands:: Special @value{GDBN} commands for Fortran
11334 @node Fortran Operators
11335 @subsubsection Fortran Operators and Expressions
11337 @cindex Fortran operators and expressions
11339 Operators must be defined on values of specific types. For instance,
11340 @code{+} is defined on numbers, but not on characters or other non-
11341 arithmetic types. Operators are often defined on groups of types.
11345 The exponentiation operator. It raises the first operand to the power
11349 The range operator. Normally used in the form of array(low:high) to
11350 represent a section of array.
11353 The access component operator. Normally used to access elements in derived
11354 types. Also suitable for unions. As unions aren't part of regular Fortran,
11355 this can only happen when accessing a register that uses a gdbarch-defined
11359 @node Fortran Defaults
11360 @subsubsection Fortran Defaults
11362 @cindex Fortran Defaults
11364 Fortran symbols are usually case-insensitive, so @value{GDBN} by
11365 default uses case-insensitive matches for Fortran symbols. You can
11366 change that with the @samp{set case-insensitive} command, see
11367 @ref{Symbols}, for the details.
11369 @node Special Fortran Commands
11370 @subsubsection Special Fortran Commands
11372 @cindex Special Fortran commands
11374 @value{GDBN} has some commands to support Fortran-specific features,
11375 such as displaying common blocks.
11378 @cindex @code{COMMON} blocks, Fortran
11379 @kindex info common
11380 @item info common @r{[}@var{common-name}@r{]}
11381 This command prints the values contained in the Fortran @code{COMMON}
11382 block whose name is @var{common-name}. With no argument, the names of
11383 all @code{COMMON} blocks visible at the current program location are
11390 @cindex Pascal support in @value{GDBN}, limitations
11391 Debugging Pascal programs which use sets, subranges, file variables, or
11392 nested functions does not currently work. @value{GDBN} does not support
11393 entering expressions, printing values, or similar features using Pascal
11396 The Pascal-specific command @code{set print pascal_static-members}
11397 controls whether static members of Pascal objects are displayed.
11398 @xref{Print Settings, pascal_static-members}.
11401 @subsection Modula-2
11403 @cindex Modula-2, @value{GDBN} support
11405 The extensions made to @value{GDBN} to support Modula-2 only support
11406 output from the @sc{gnu} Modula-2 compiler (which is currently being
11407 developed). Other Modula-2 compilers are not currently supported, and
11408 attempting to debug executables produced by them is most likely
11409 to give an error as @value{GDBN} reads in the executable's symbol
11412 @cindex expressions in Modula-2
11414 * M2 Operators:: Built-in operators
11415 * Built-In Func/Proc:: Built-in functions and procedures
11416 * M2 Constants:: Modula-2 constants
11417 * M2 Types:: Modula-2 types
11418 * M2 Defaults:: Default settings for Modula-2
11419 * Deviations:: Deviations from standard Modula-2
11420 * M2 Checks:: Modula-2 type and range checks
11421 * M2 Scope:: The scope operators @code{::} and @code{.}
11422 * GDB/M2:: @value{GDBN} and Modula-2
11426 @subsubsection Operators
11427 @cindex Modula-2 operators
11429 Operators must be defined on values of specific types. For instance,
11430 @code{+} is defined on numbers, but not on structures. Operators are
11431 often defined on groups of types. For the purposes of Modula-2, the
11432 following definitions hold:
11437 @emph{Integral types} consist of @code{INTEGER}, @code{CARDINAL}, and
11441 @emph{Character types} consist of @code{CHAR} and its subranges.
11444 @emph{Floating-point types} consist of @code{REAL}.
11447 @emph{Pointer types} consist of anything declared as @code{POINTER TO
11451 @emph{Scalar types} consist of all of the above.
11454 @emph{Set types} consist of @code{SET} and @code{BITSET} types.
11457 @emph{Boolean types} consist of @code{BOOLEAN}.
11461 The following operators are supported, and appear in order of
11462 increasing precedence:
11466 Function argument or array index separator.
11469 Assignment. The value of @var{var} @code{:=} @var{value} is
11473 Less than, greater than on integral, floating-point, or enumerated
11477 Less than or equal to, greater than or equal to
11478 on integral, floating-point and enumerated types, or set inclusion on
11479 set types. Same precedence as @code{<}.
11481 @item =@r{, }<>@r{, }#
11482 Equality and two ways of expressing inequality, valid on scalar types.
11483 Same precedence as @code{<}. In @value{GDBN} scripts, only @code{<>} is
11484 available for inequality, since @code{#} conflicts with the script
11488 Set membership. Defined on set types and the types of their members.
11489 Same precedence as @code{<}.
11492 Boolean disjunction. Defined on boolean types.
11495 Boolean conjunction. Defined on boolean types.
11498 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
11501 Addition and subtraction on integral and floating-point types, or union
11502 and difference on set types.
11505 Multiplication on integral and floating-point types, or set intersection
11509 Division on floating-point types, or symmetric set difference on set
11510 types. Same precedence as @code{*}.
11513 Integer division and remainder. Defined on integral types. Same
11514 precedence as @code{*}.
11517 Negative. Defined on @code{INTEGER} and @code{REAL} data.
11520 Pointer dereferencing. Defined on pointer types.
11523 Boolean negation. Defined on boolean types. Same precedence as
11527 @code{RECORD} field selector. Defined on @code{RECORD} data. Same
11528 precedence as @code{^}.
11531 Array indexing. Defined on @code{ARRAY} data. Same precedence as @code{^}.
11534 Procedure argument list. Defined on @code{PROCEDURE} objects. Same precedence
11538 @value{GDBN} and Modula-2 scope operators.
11542 @emph{Warning:} Set expressions and their operations are not yet supported, so @value{GDBN}
11543 treats the use of the operator @code{IN}, or the use of operators
11544 @code{+}, @code{-}, @code{*}, @code{/}, @code{=}, , @code{<>}, @code{#},
11545 @code{<=}, and @code{>=} on sets as an error.
11549 @node Built-In Func/Proc
11550 @subsubsection Built-in Functions and Procedures
11551 @cindex Modula-2 built-ins
11553 Modula-2 also makes available several built-in procedures and functions.
11554 In describing these, the following metavariables are used:
11559 represents an @code{ARRAY} variable.
11562 represents a @code{CHAR} constant or variable.
11565 represents a variable or constant of integral type.
11568 represents an identifier that belongs to a set. Generally used in the
11569 same function with the metavariable @var{s}. The type of @var{s} should
11570 be @code{SET OF @var{mtype}} (where @var{mtype} is the type of @var{m}).
11573 represents a variable or constant of integral or floating-point type.
11576 represents a variable or constant of floating-point type.
11582 represents a variable.
11585 represents a variable or constant of one of many types. See the
11586 explanation of the function for details.
11589 All Modula-2 built-in procedures also return a result, described below.
11593 Returns the absolute value of @var{n}.
11596 If @var{c} is a lower case letter, it returns its upper case
11597 equivalent, otherwise it returns its argument.
11600 Returns the character whose ordinal value is @var{i}.
11603 Decrements the value in the variable @var{v} by one. Returns the new value.
11605 @item DEC(@var{v},@var{i})
11606 Decrements the value in the variable @var{v} by @var{i}. Returns the
11609 @item EXCL(@var{m},@var{s})
11610 Removes the element @var{m} from the set @var{s}. Returns the new
11613 @item FLOAT(@var{i})
11614 Returns the floating point equivalent of the integer @var{i}.
11616 @item HIGH(@var{a})
11617 Returns the index of the last member of @var{a}.
11620 Increments the value in the variable @var{v} by one. Returns the new value.
11622 @item INC(@var{v},@var{i})
11623 Increments the value in the variable @var{v} by @var{i}. Returns the
11626 @item INCL(@var{m},@var{s})
11627 Adds the element @var{m} to the set @var{s} if it is not already
11628 there. Returns the new set.
11631 Returns the maximum value of the type @var{t}.
11634 Returns the minimum value of the type @var{t}.
11637 Returns boolean TRUE if @var{i} is an odd number.
11640 Returns the ordinal value of its argument. For example, the ordinal
11641 value of a character is its @sc{ascii} value (on machines supporting the
11642 @sc{ascii} character set). @var{x} must be of an ordered type, which include
11643 integral, character and enumerated types.
11645 @item SIZE(@var{x})
11646 Returns the size of its argument. @var{x} can be a variable or a type.
11648 @item TRUNC(@var{r})
11649 Returns the integral part of @var{r}.
11651 @item TSIZE(@var{x})
11652 Returns the size of its argument. @var{x} can be a variable or a type.
11654 @item VAL(@var{t},@var{i})
11655 Returns the member of the type @var{t} whose ordinal value is @var{i}.
11659 @emph{Warning:} Sets and their operations are not yet supported, so
11660 @value{GDBN} treats the use of procedures @code{INCL} and @code{EXCL} as
11664 @cindex Modula-2 constants
11666 @subsubsection Constants
11668 @value{GDBN} allows you to express the constants of Modula-2 in the following
11674 Integer constants are simply a sequence of digits. When used in an
11675 expression, a constant is interpreted to be type-compatible with the
11676 rest of the expression. Hexadecimal integers are specified by a
11677 trailing @samp{H}, and octal integers by a trailing @samp{B}.
11680 Floating point constants appear as a sequence of digits, followed by a
11681 decimal point and another sequence of digits. An optional exponent can
11682 then be specified, in the form @samp{E@r{[}+@r{|}-@r{]}@var{nnn}}, where
11683 @samp{@r{[}+@r{|}-@r{]}@var{nnn}} is the desired exponent. All of the
11684 digits of the floating point constant must be valid decimal (base 10)
11688 Character constants consist of a single character enclosed by a pair of
11689 like quotes, either single (@code{'}) or double (@code{"}). They may
11690 also be expressed by their ordinal value (their @sc{ascii} value, usually)
11691 followed by a @samp{C}.
11694 String constants consist of a sequence of characters enclosed by a
11695 pair of like quotes, either single (@code{'}) or double (@code{"}).
11696 Escape sequences in the style of C are also allowed. @xref{C
11697 Constants, ,C and C@t{++} Constants}, for a brief explanation of escape
11701 Enumerated constants consist of an enumerated identifier.
11704 Boolean constants consist of the identifiers @code{TRUE} and
11708 Pointer constants consist of integral values only.
11711 Set constants are not yet supported.
11715 @subsubsection Modula-2 Types
11716 @cindex Modula-2 types
11718 Currently @value{GDBN} can print the following data types in Modula-2
11719 syntax: array types, record types, set types, pointer types, procedure
11720 types, enumerated types, subrange types and base types. You can also
11721 print the contents of variables declared using these type.
11722 This section gives a number of simple source code examples together with
11723 sample @value{GDBN} sessions.
11725 The first example contains the following section of code:
11734 and you can request @value{GDBN} to interrogate the type and value of
11735 @code{r} and @code{s}.
11738 (@value{GDBP}) print s
11740 (@value{GDBP}) ptype s
11742 (@value{GDBP}) print r
11744 (@value{GDBP}) ptype r
11749 Likewise if your source code declares @code{s} as:
11753 s: SET ['A'..'Z'] ;
11757 then you may query the type of @code{s} by:
11760 (@value{GDBP}) ptype s
11761 type = SET ['A'..'Z']
11765 Note that at present you cannot interactively manipulate set
11766 expressions using the debugger.
11768 The following example shows how you might declare an array in Modula-2
11769 and how you can interact with @value{GDBN} to print its type and contents:
11773 s: ARRAY [-10..10] OF CHAR ;
11777 (@value{GDBP}) ptype s
11778 ARRAY [-10..10] OF CHAR
11781 Note that the array handling is not yet complete and although the type
11782 is printed correctly, expression handling still assumes that all
11783 arrays have a lower bound of zero and not @code{-10} as in the example
11786 Here are some more type related Modula-2 examples:
11790 colour = (blue, red, yellow, green) ;
11791 t = [blue..yellow] ;
11799 The @value{GDBN} interaction shows how you can query the data type
11800 and value of a variable.
11803 (@value{GDBP}) print s
11805 (@value{GDBP}) ptype t
11806 type = [blue..yellow]
11810 In this example a Modula-2 array is declared and its contents
11811 displayed. Observe that the contents are written in the same way as
11812 their @code{C} counterparts.
11816 s: ARRAY [1..5] OF CARDINAL ;
11822 (@value{GDBP}) print s
11823 $1 = @{1, 0, 0, 0, 0@}
11824 (@value{GDBP}) ptype s
11825 type = ARRAY [1..5] OF CARDINAL
11828 The Modula-2 language interface to @value{GDBN} also understands
11829 pointer types as shown in this example:
11833 s: POINTER TO ARRAY [1..5] OF CARDINAL ;
11840 and you can request that @value{GDBN} describes the type of @code{s}.
11843 (@value{GDBP}) ptype s
11844 type = POINTER TO ARRAY [1..5] OF CARDINAL
11847 @value{GDBN} handles compound types as we can see in this example.
11848 Here we combine array types, record types, pointer types and subrange
11859 myarray = ARRAY myrange OF CARDINAL ;
11860 myrange = [-2..2] ;
11862 s: POINTER TO ARRAY myrange OF foo ;
11866 and you can ask @value{GDBN} to describe the type of @code{s} as shown
11870 (@value{GDBP}) ptype s
11871 type = POINTER TO ARRAY [-2..2] OF foo = RECORD
11874 f3 : ARRAY [-2..2] OF CARDINAL;
11879 @subsubsection Modula-2 Defaults
11880 @cindex Modula-2 defaults
11882 If type and range checking are set automatically by @value{GDBN}, they
11883 both default to @code{on} whenever the working language changes to
11884 Modula-2. This happens regardless of whether you or @value{GDBN}
11885 selected the working language.
11887 If you allow @value{GDBN} to set the language automatically, then entering
11888 code compiled from a file whose name ends with @file{.mod} sets the
11889 working language to Modula-2. @xref{Automatically, ,Having @value{GDBN}
11890 Infer the Source Language}, for further details.
11893 @subsubsection Deviations from Standard Modula-2
11894 @cindex Modula-2, deviations from
11896 A few changes have been made to make Modula-2 programs easier to debug.
11897 This is done primarily via loosening its type strictness:
11901 Unlike in standard Modula-2, pointer constants can be formed by
11902 integers. This allows you to modify pointer variables during
11903 debugging. (In standard Modula-2, the actual address contained in a
11904 pointer variable is hidden from you; it can only be modified
11905 through direct assignment to another pointer variable or expression that
11906 returned a pointer.)
11909 C escape sequences can be used in strings and characters to represent
11910 non-printable characters. @value{GDBN} prints out strings with these
11911 escape sequences embedded. Single non-printable characters are
11912 printed using the @samp{CHR(@var{nnn})} format.
11915 The assignment operator (@code{:=}) returns the value of its right-hand
11919 All built-in procedures both modify @emph{and} return their argument.
11923 @subsubsection Modula-2 Type and Range Checks
11924 @cindex Modula-2 checks
11927 @emph{Warning:} in this release, @value{GDBN} does not yet perform type or
11930 @c FIXME remove warning when type/range checks added
11932 @value{GDBN} considers two Modula-2 variables type equivalent if:
11936 They are of types that have been declared equivalent via a @code{TYPE
11937 @var{t1} = @var{t2}} statement
11940 They have been declared on the same line. (Note: This is true of the
11941 @sc{gnu} Modula-2 compiler, but it may not be true of other compilers.)
11944 As long as type checking is enabled, any attempt to combine variables
11945 whose types are not equivalent is an error.
11947 Range checking is done on all mathematical operations, assignment, array
11948 index bounds, and all built-in functions and procedures.
11951 @subsubsection The Scope Operators @code{::} and @code{.}
11953 @cindex @code{.}, Modula-2 scope operator
11954 @cindex colon, doubled as scope operator
11956 @vindex colon-colon@r{, in Modula-2}
11957 @c Info cannot handle :: but TeX can.
11960 @vindex ::@r{, in Modula-2}
11963 There are a few subtle differences between the Modula-2 scope operator
11964 (@code{.}) and the @value{GDBN} scope operator (@code{::}). The two have
11969 @var{module} . @var{id}
11970 @var{scope} :: @var{id}
11974 where @var{scope} is the name of a module or a procedure,
11975 @var{module} the name of a module, and @var{id} is any declared
11976 identifier within your program, except another module.
11978 Using the @code{::} operator makes @value{GDBN} search the scope
11979 specified by @var{scope} for the identifier @var{id}. If it is not
11980 found in the specified scope, then @value{GDBN} searches all scopes
11981 enclosing the one specified by @var{scope}.
11983 Using the @code{.} operator makes @value{GDBN} search the current scope for
11984 the identifier specified by @var{id} that was imported from the
11985 definition module specified by @var{module}. With this operator, it is
11986 an error if the identifier @var{id} was not imported from definition
11987 module @var{module}, or if @var{id} is not an identifier in
11991 @subsubsection @value{GDBN} and Modula-2
11993 Some @value{GDBN} commands have little use when debugging Modula-2 programs.
11994 Five subcommands of @code{set print} and @code{show print} apply
11995 specifically to C and C@t{++}: @samp{vtbl}, @samp{demangle},
11996 @samp{asm-demangle}, @samp{object}, and @samp{union}. The first four
11997 apply to C@t{++}, and the last to the C @code{union} type, which has no direct
11998 analogue in Modula-2.
12000 The @code{@@} operator (@pxref{Expressions, ,Expressions}), while available
12001 with any language, is not useful with Modula-2. Its
12002 intent is to aid the debugging of @dfn{dynamic arrays}, which cannot be
12003 created in Modula-2 as they can in C or C@t{++}. However, because an
12004 address can be specified by an integral constant, the construct
12005 @samp{@{@var{type}@}@var{adrexp}} is still useful.
12007 @cindex @code{#} in Modula-2
12008 In @value{GDBN} scripts, the Modula-2 inequality operator @code{#} is
12009 interpreted as the beginning of a comment. Use @code{<>} instead.
12015 The extensions made to @value{GDBN} for Ada only support
12016 output from the @sc{gnu} Ada (GNAT) compiler.
12017 Other Ada compilers are not currently supported, and
12018 attempting to debug executables produced by them is most likely
12022 @cindex expressions in Ada
12024 * Ada Mode Intro:: General remarks on the Ada syntax
12025 and semantics supported by Ada mode
12027 * Omissions from Ada:: Restrictions on the Ada expression syntax.
12028 * Additions to Ada:: Extensions of the Ada expression syntax.
12029 * Stopping Before Main Program:: Debugging the program during elaboration.
12030 * Ada Tasks:: Listing and setting breakpoints in tasks.
12031 * Ada Tasks and Core Files:: Tasking Support when Debugging Core Files
12032 * Ada Glitches:: Known peculiarities of Ada mode.
12035 @node Ada Mode Intro
12036 @subsubsection Introduction
12037 @cindex Ada mode, general
12039 The Ada mode of @value{GDBN} supports a fairly large subset of Ada expression
12040 syntax, with some extensions.
12041 The philosophy behind the design of this subset is
12045 That @value{GDBN} should provide basic literals and access to operations for
12046 arithmetic, dereferencing, field selection, indexing, and subprogram calls,
12047 leaving more sophisticated computations to subprograms written into the
12048 program (which therefore may be called from @value{GDBN}).
12051 That type safety and strict adherence to Ada language restrictions
12052 are not particularly important to the @value{GDBN} user.
12055 That brevity is important to the @value{GDBN} user.
12058 Thus, for brevity, the debugger acts as if all names declared in
12059 user-written packages are directly visible, even if they are not visible
12060 according to Ada rules, thus making it unnecessary to fully qualify most
12061 names with their packages, regardless of context. Where this causes
12062 ambiguity, @value{GDBN} asks the user's intent.
12064 The debugger will start in Ada mode if it detects an Ada main program.
12065 As for other languages, it will enter Ada mode when stopped in a program that
12066 was translated from an Ada source file.
12068 While in Ada mode, you may use `@t{--}' for comments. This is useful
12069 mostly for documenting command files. The standard @value{GDBN} comment
12070 (@samp{#}) still works at the beginning of a line in Ada mode, but not in the
12071 middle (to allow based literals).
12073 The debugger supports limited overloading. Given a subprogram call in which
12074 the function symbol has multiple definitions, it will use the number of
12075 actual parameters and some information about their types to attempt to narrow
12076 the set of definitions. It also makes very limited use of context, preferring
12077 procedures to functions in the context of the @code{call} command, and
12078 functions to procedures elsewhere.
12080 @node Omissions from Ada
12081 @subsubsection Omissions from Ada
12082 @cindex Ada, omissions from
12084 Here are the notable omissions from the subset:
12088 Only a subset of the attributes are supported:
12092 @t{'First}, @t{'Last}, and @t{'Length}
12093 on array objects (not on types and subtypes).
12096 @t{'Min} and @t{'Max}.
12099 @t{'Pos} and @t{'Val}.
12105 @t{'Range} on array objects (not subtypes), but only as the right
12106 operand of the membership (@code{in}) operator.
12109 @t{'Access}, @t{'Unchecked_Access}, and
12110 @t{'Unrestricted_Access} (a GNAT extension).
12118 @code{Characters.Latin_1} are not available and
12119 concatenation is not implemented. Thus, escape characters in strings are
12120 not currently available.
12123 Equality tests (@samp{=} and @samp{/=}) on arrays test for bitwise
12124 equality of representations. They will generally work correctly
12125 for strings and arrays whose elements have integer or enumeration types.
12126 They may not work correctly for arrays whose element
12127 types have user-defined equality, for arrays of real values
12128 (in particular, IEEE-conformant floating point, because of negative
12129 zeroes and NaNs), and for arrays whose elements contain unused bits with
12130 indeterminate values.
12133 The other component-by-component array operations (@code{and}, @code{or},
12134 @code{xor}, @code{not}, and relational tests other than equality)
12135 are not implemented.
12138 @cindex array aggregates (Ada)
12139 @cindex record aggregates (Ada)
12140 @cindex aggregates (Ada)
12141 There is limited support for array and record aggregates. They are
12142 permitted only on the right sides of assignments, as in these examples:
12145 (@value{GDBP}) set An_Array := (1, 2, 3, 4, 5, 6)
12146 (@value{GDBP}) set An_Array := (1, others => 0)
12147 (@value{GDBP}) set An_Array := (0|4 => 1, 1..3 => 2, 5 => 6)
12148 (@value{GDBP}) set A_2D_Array := ((1, 2, 3), (4, 5, 6), (7, 8, 9))
12149 (@value{GDBP}) set A_Record := (1, "Peter", True);
12150 (@value{GDBP}) set A_Record := (Name => "Peter", Id => 1, Alive => True)
12154 discriminant's value by assigning an aggregate has an
12155 undefined effect if that discriminant is used within the record.
12156 However, you can first modify discriminants by directly assigning to
12157 them (which normally would not be allowed in Ada), and then performing an
12158 aggregate assignment. For example, given a variable @code{A_Rec}
12159 declared to have a type such as:
12162 type Rec (Len : Small_Integer := 0) is record
12164 Vals : IntArray (1 .. Len);
12168 you can assign a value with a different size of @code{Vals} with two
12172 (@value{GDBP}) set A_Rec.Len := 4
12173 (@value{GDBP}) set A_Rec := (Id => 42, Vals => (1, 2, 3, 4))
12176 As this example also illustrates, @value{GDBN} is very loose about the usual
12177 rules concerning aggregates. You may leave out some of the
12178 components of an array or record aggregate (such as the @code{Len}
12179 component in the assignment to @code{A_Rec} above); they will retain their
12180 original values upon assignment. You may freely use dynamic values as
12181 indices in component associations. You may even use overlapping or
12182 redundant component associations, although which component values are
12183 assigned in such cases is not defined.
12186 Calls to dispatching subprograms are not implemented.
12189 The overloading algorithm is much more limited (i.e., less selective)
12190 than that of real Ada. It makes only limited use of the context in
12191 which a subexpression appears to resolve its meaning, and it is much
12192 looser in its rules for allowing type matches. As a result, some
12193 function calls will be ambiguous, and the user will be asked to choose
12194 the proper resolution.
12197 The @code{new} operator is not implemented.
12200 Entry calls are not implemented.
12203 Aside from printing, arithmetic operations on the native VAX floating-point
12204 formats are not supported.
12207 It is not possible to slice a packed array.
12210 The names @code{True} and @code{False}, when not part of a qualified name,
12211 are interpreted as if implicitly prefixed by @code{Standard}, regardless of
12213 Should your program
12214 redefine these names in a package or procedure (at best a dubious practice),
12215 you will have to use fully qualified names to access their new definitions.
12218 @node Additions to Ada
12219 @subsubsection Additions to Ada
12220 @cindex Ada, deviations from
12222 As it does for other languages, @value{GDBN} makes certain generic
12223 extensions to Ada (@pxref{Expressions}):
12227 If the expression @var{E} is a variable residing in memory (typically
12228 a local variable or array element) and @var{N} is a positive integer,
12229 then @code{@var{E}@@@var{N}} displays the values of @var{E} and the
12230 @var{N}-1 adjacent variables following it in memory as an array. In
12231 Ada, this operator is generally not necessary, since its prime use is
12232 in displaying parts of an array, and slicing will usually do this in
12233 Ada. However, there are occasional uses when debugging programs in
12234 which certain debugging information has been optimized away.
12237 @code{@var{B}::@var{var}} means ``the variable named @var{var} that
12238 appears in function or file @var{B}.'' When @var{B} is a file name,
12239 you must typically surround it in single quotes.
12242 The expression @code{@{@var{type}@} @var{addr}} means ``the variable of type
12243 @var{type} that appears at address @var{addr}.''
12246 A name starting with @samp{$} is a convenience variable
12247 (@pxref{Convenience Vars}) or a machine register (@pxref{Registers}).
12250 In addition, @value{GDBN} provides a few other shortcuts and outright
12251 additions specific to Ada:
12255 The assignment statement is allowed as an expression, returning
12256 its right-hand operand as its value. Thus, you may enter
12259 (@value{GDBP}) set x := y + 3
12260 (@value{GDBP}) print A(tmp := y + 1)
12264 The semicolon is allowed as an ``operator,'' returning as its value
12265 the value of its right-hand operand.
12266 This allows, for example,
12267 complex conditional breaks:
12270 (@value{GDBP}) break f
12271 (@value{GDBP}) condition 1 (report(i); k += 1; A(k) > 100)
12275 Rather than use catenation and symbolic character names to introduce special
12276 characters into strings, one may instead use a special bracket notation,
12277 which is also used to print strings. A sequence of characters of the form
12278 @samp{["@var{XX}"]} within a string or character literal denotes the
12279 (single) character whose numeric encoding is @var{XX} in hexadecimal. The
12280 sequence of characters @samp{["""]} also denotes a single quotation mark
12281 in strings. For example,
12283 "One line.["0a"]Next line.["0a"]"
12286 contains an ASCII newline character (@code{Ada.Characters.Latin_1.LF})
12290 The subtype used as a prefix for the attributes @t{'Pos}, @t{'Min}, and
12291 @t{'Max} is optional (and is ignored in any case). For example, it is valid
12295 (@value{GDBP}) print 'max(x, y)
12299 When printing arrays, @value{GDBN} uses positional notation when the
12300 array has a lower bound of 1, and uses a modified named notation otherwise.
12301 For example, a one-dimensional array of three integers with a lower bound
12302 of 3 might print as
12309 That is, in contrast to valid Ada, only the first component has a @code{=>}
12313 You may abbreviate attributes in expressions with any unique,
12314 multi-character subsequence of
12315 their names (an exact match gets preference).
12316 For example, you may use @t{a'len}, @t{a'gth}, or @t{a'lh}
12317 in place of @t{a'length}.
12320 @cindex quoting Ada internal identifiers
12321 Since Ada is case-insensitive, the debugger normally maps identifiers you type
12322 to lower case. The GNAT compiler uses upper-case characters for
12323 some of its internal identifiers, which are normally of no interest to users.
12324 For the rare occasions when you actually have to look at them,
12325 enclose them in angle brackets to avoid the lower-case mapping.
12328 (@value{GDBP}) print <JMPBUF_SAVE>[0]
12332 Printing an object of class-wide type or dereferencing an
12333 access-to-class-wide value will display all the components of the object's
12334 specific type (as indicated by its run-time tag). Likewise, component
12335 selection on such a value will operate on the specific type of the
12340 @node Stopping Before Main Program
12341 @subsubsection Stopping at the Very Beginning
12343 @cindex breakpointing Ada elaboration code
12344 It is sometimes necessary to debug the program during elaboration, and
12345 before reaching the main procedure.
12346 As defined in the Ada Reference
12347 Manual, the elaboration code is invoked from a procedure called
12348 @code{adainit}. To run your program up to the beginning of
12349 elaboration, simply use the following two commands:
12350 @code{tbreak adainit} and @code{run}.
12353 @subsubsection Extensions for Ada Tasks
12354 @cindex Ada, tasking
12356 Support for Ada tasks is analogous to that for threads (@pxref{Threads}).
12357 @value{GDBN} provides the following task-related commands:
12362 This command shows a list of current Ada tasks, as in the following example:
12369 (@value{GDBP}) info tasks
12370 ID TID P-ID Pri State Name
12371 1 8088000 0 15 Child Activation Wait main_task
12372 2 80a4000 1 15 Accept Statement b
12373 3 809a800 1 15 Child Activation Wait a
12374 * 4 80ae800 3 15 Runnable c
12379 In this listing, the asterisk before the last task indicates it to be the
12380 task currently being inspected.
12384 Represents @value{GDBN}'s internal task number.
12390 The parent's task ID (@value{GDBN}'s internal task number).
12393 The base priority of the task.
12396 Current state of the task.
12400 The task has been created but has not been activated. It cannot be
12404 The task is not blocked for any reason known to Ada. (It may be waiting
12405 for a mutex, though.) It is conceptually "executing" in normal mode.
12408 The task is terminated, in the sense of ARM 9.3 (5). Any dependents
12409 that were waiting on terminate alternatives have been awakened and have
12410 terminated themselves.
12412 @item Child Activation Wait
12413 The task is waiting for created tasks to complete activation.
12415 @item Accept Statement
12416 The task is waiting on an accept or selective wait statement.
12418 @item Waiting on entry call
12419 The task is waiting on an entry call.
12421 @item Async Select Wait
12422 The task is waiting to start the abortable part of an asynchronous
12426 The task is waiting on a select statement with only a delay
12429 @item Child Termination Wait
12430 The task is sleeping having completed a master within itself, and is
12431 waiting for the tasks dependent on that master to become terminated or
12432 waiting on a terminate Phase.
12434 @item Wait Child in Term Alt
12435 The task is sleeping waiting for tasks on terminate alternatives to
12436 finish terminating.
12438 @item Accepting RV with @var{taskno}
12439 The task is accepting a rendez-vous with the task @var{taskno}.
12443 Name of the task in the program.
12447 @kindex info task @var{taskno}
12448 @item info task @var{taskno}
12449 This command shows detailled informations on the specified task, as in
12450 the following example:
12455 (@value{GDBP}) info tasks
12456 ID TID P-ID Pri State Name
12457 1 8077880 0 15 Child Activation Wait main_task
12458 * 2 807c468 1 15 Runnable task_1
12459 (@value{GDBP}) info task 2
12460 Ada Task: 0x807c468
12463 Parent: 1 (main_task)
12469 @kindex task@r{ (Ada)}
12470 @cindex current Ada task ID
12471 This command prints the ID of the current task.
12477 (@value{GDBP}) info tasks
12478 ID TID P-ID Pri State Name
12479 1 8077870 0 15 Child Activation Wait main_task
12480 * 2 807c458 1 15 Runnable t
12481 (@value{GDBP}) task
12482 [Current task is 2]
12485 @item task @var{taskno}
12486 @cindex Ada task switching
12487 This command is like the @code{thread @var{threadno}}
12488 command (@pxref{Threads}). It switches the context of debugging
12489 from the current task to the given task.
12495 (@value{GDBP}) info tasks
12496 ID TID P-ID Pri State Name
12497 1 8077870 0 15 Child Activation Wait main_task
12498 * 2 807c458 1 15 Runnable t
12499 (@value{GDBP}) task 1
12500 [Switching to task 1]
12501 #0 0x8067726 in pthread_cond_wait ()
12503 #0 0x8067726 in pthread_cond_wait ()
12504 #1 0x8056714 in system.os_interface.pthread_cond_wait ()
12505 #2 0x805cb63 in system.task_primitives.operations.sleep ()
12506 #3 0x806153e in system.tasking.stages.activate_tasks ()
12507 #4 0x804aacc in un () at un.adb:5
12510 @item break @var{linespec} task @var{taskno}
12511 @itemx break @var{linespec} task @var{taskno} if @dots{}
12512 @cindex breakpoints and tasks, in Ada
12513 @cindex task breakpoints, in Ada
12514 @kindex break @dots{} task @var{taskno}@r{ (Ada)}
12515 These commands are like the @code{break @dots{} thread @dots{}}
12516 command (@pxref{Thread Stops}).
12517 @var{linespec} specifies source lines, as described
12518 in @ref{Specify Location}.
12520 Use the qualifier @samp{task @var{taskno}} with a breakpoint command
12521 to specify that you only want @value{GDBN} to stop the program when a
12522 particular Ada task reaches this breakpoint. @var{taskno} is one of the
12523 numeric task identifiers assigned by @value{GDBN}, shown in the first
12524 column of the @samp{info tasks} display.
12526 If you do not specify @samp{task @var{taskno}} when you set a
12527 breakpoint, the breakpoint applies to @emph{all} tasks of your
12530 You can use the @code{task} qualifier on conditional breakpoints as
12531 well; in this case, place @samp{task @var{taskno}} before the
12532 breakpoint condition (before the @code{if}).
12540 (@value{GDBP}) info tasks
12541 ID TID P-ID Pri State Name
12542 1 140022020 0 15 Child Activation Wait main_task
12543 2 140045060 1 15 Accept/Select Wait t2
12544 3 140044840 1 15 Runnable t1
12545 * 4 140056040 1 15 Runnable t3
12546 (@value{GDBP}) b 15 task 2
12547 Breakpoint 5 at 0x120044cb0: file test_task_debug.adb, line 15.
12548 (@value{GDBP}) cont
12553 Breakpoint 5, test_task_debug () at test_task_debug.adb:15
12555 (@value{GDBP}) info tasks
12556 ID TID P-ID Pri State Name
12557 1 140022020 0 15 Child Activation Wait main_task
12558 * 2 140045060 1 15 Runnable t2
12559 3 140044840 1 15 Runnable t1
12560 4 140056040 1 15 Delay Sleep t3
12564 @node Ada Tasks and Core Files
12565 @subsubsection Tasking Support when Debugging Core Files
12566 @cindex Ada tasking and core file debugging
12568 When inspecting a core file, as opposed to debugging a live program,
12569 tasking support may be limited or even unavailable, depending on
12570 the platform being used.
12571 For instance, on x86-linux, the list of tasks is available, but task
12572 switching is not supported. On Tru64, however, task switching will work
12575 On certain platforms, including Tru64, the debugger needs to perform some
12576 memory writes in order to provide Ada tasking support. When inspecting
12577 a core file, this means that the core file must be opened with read-write
12578 privileges, using the command @samp{"set write on"} (@pxref{Patching}).
12579 Under these circumstances, you should make a backup copy of the core
12580 file before inspecting it with @value{GDBN}.
12583 @subsubsection Known Peculiarities of Ada Mode
12584 @cindex Ada, problems
12586 Besides the omissions listed previously (@pxref{Omissions from Ada}),
12587 we know of several problems with and limitations of Ada mode in
12589 some of which will be fixed with planned future releases of the debugger
12590 and the GNU Ada compiler.
12594 Currently, the debugger
12595 has insufficient information to determine whether certain pointers represent
12596 pointers to objects or the objects themselves.
12597 Thus, the user may have to tack an extra @code{.all} after an expression
12598 to get it printed properly.
12601 Static constants that the compiler chooses not to materialize as objects in
12602 storage are invisible to the debugger.
12605 Named parameter associations in function argument lists are ignored (the
12606 argument lists are treated as positional).
12609 Many useful library packages are currently invisible to the debugger.
12612 Fixed-point arithmetic, conversions, input, and output is carried out using
12613 floating-point arithmetic, and may give results that only approximate those on
12617 The GNAT compiler never generates the prefix @code{Standard} for any of
12618 the standard symbols defined by the Ada language. @value{GDBN} knows about
12619 this: it will strip the prefix from names when you use it, and will never
12620 look for a name you have so qualified among local symbols, nor match against
12621 symbols in other packages or subprograms. If you have
12622 defined entities anywhere in your program other than parameters and
12623 local variables whose simple names match names in @code{Standard},
12624 GNAT's lack of qualification here can cause confusion. When this happens,
12625 you can usually resolve the confusion
12626 by qualifying the problematic names with package
12627 @code{Standard} explicitly.
12630 @node Unsupported Languages
12631 @section Unsupported Languages
12633 @cindex unsupported languages
12634 @cindex minimal language
12635 In addition to the other fully-supported programming languages,
12636 @value{GDBN} also provides a pseudo-language, called @code{minimal}.
12637 It does not represent a real programming language, but provides a set
12638 of capabilities close to what the C or assembly languages provide.
12639 This should allow most simple operations to be performed while debugging
12640 an application that uses a language currently not supported by @value{GDBN}.
12642 If the language is set to @code{auto}, @value{GDBN} will automatically
12643 select this language if the current frame corresponds to an unsupported
12647 @chapter Examining the Symbol Table
12649 The commands described in this chapter allow you to inquire about the
12650 symbols (names of variables, functions and types) defined in your
12651 program. This information is inherent in the text of your program and
12652 does not change as your program executes. @value{GDBN} finds it in your
12653 program's symbol table, in the file indicated when you started @value{GDBN}
12654 (@pxref{File Options, ,Choosing Files}), or by one of the
12655 file-management commands (@pxref{Files, ,Commands to Specify Files}).
12657 @cindex symbol names
12658 @cindex names of symbols
12659 @cindex quoting names
12660 Occasionally, you may need to refer to symbols that contain unusual
12661 characters, which @value{GDBN} ordinarily treats as word delimiters. The
12662 most frequent case is in referring to static variables in other
12663 source files (@pxref{Variables,,Program Variables}). File names
12664 are recorded in object files as debugging symbols, but @value{GDBN} would
12665 ordinarily parse a typical file name, like @file{foo.c}, as the three words
12666 @samp{foo} @samp{.} @samp{c}. To allow @value{GDBN} to recognize
12667 @samp{foo.c} as a single symbol, enclose it in single quotes; for example,
12674 looks up the value of @code{x} in the scope of the file @file{foo.c}.
12677 @cindex case-insensitive symbol names
12678 @cindex case sensitivity in symbol names
12679 @kindex set case-sensitive
12680 @item set case-sensitive on
12681 @itemx set case-sensitive off
12682 @itemx set case-sensitive auto
12683 Normally, when @value{GDBN} looks up symbols, it matches their names
12684 with case sensitivity determined by the current source language.
12685 Occasionally, you may wish to control that. The command @code{set
12686 case-sensitive} lets you do that by specifying @code{on} for
12687 case-sensitive matches or @code{off} for case-insensitive ones. If
12688 you specify @code{auto}, case sensitivity is reset to the default
12689 suitable for the source language. The default is case-sensitive
12690 matches for all languages except for Fortran, for which the default is
12691 case-insensitive matches.
12693 @kindex show case-sensitive
12694 @item show case-sensitive
12695 This command shows the current setting of case sensitivity for symbols
12698 @kindex info address
12699 @cindex address of a symbol
12700 @item info address @var{symbol}
12701 Describe where the data for @var{symbol} is stored. For a register
12702 variable, this says which register it is kept in. For a non-register
12703 local variable, this prints the stack-frame offset at which the variable
12706 Note the contrast with @samp{print &@var{symbol}}, which does not work
12707 at all for a register variable, and for a stack local variable prints
12708 the exact address of the current instantiation of the variable.
12710 @kindex info symbol
12711 @cindex symbol from address
12712 @cindex closest symbol and offset for an address
12713 @item info symbol @var{addr}
12714 Print the name of a symbol which is stored at the address @var{addr}.
12715 If no symbol is stored exactly at @var{addr}, @value{GDBN} prints the
12716 nearest symbol and an offset from it:
12719 (@value{GDBP}) info symbol 0x54320
12720 _initialize_vx + 396 in section .text
12724 This is the opposite of the @code{info address} command. You can use
12725 it to find out the name of a variable or a function given its address.
12727 For dynamically linked executables, the name of executable or shared
12728 library containing the symbol is also printed:
12731 (@value{GDBP}) info symbol 0x400225
12732 _start + 5 in section .text of /tmp/a.out
12733 (@value{GDBP}) info symbol 0x2aaaac2811cf
12734 __read_nocancel + 6 in section .text of /usr/lib64/libc.so.6
12738 @item whatis [@var{arg}]
12739 Print the data type of @var{arg}, which can be either an expression or
12740 a data type. With no argument, print the data type of @code{$}, the
12741 last value in the value history. If @var{arg} is an expression, it is
12742 not actually evaluated, and any side-effecting operations (such as
12743 assignments or function calls) inside it do not take place. If
12744 @var{arg} is a type name, it may be the name of a type or typedef, or
12745 for C code it may have the form @samp{class @var{class-name}},
12746 @samp{struct @var{struct-tag}}, @samp{union @var{union-tag}} or
12747 @samp{enum @var{enum-tag}}.
12748 @xref{Expressions, ,Expressions}.
12751 @item ptype [@var{arg}]
12752 @code{ptype} accepts the same arguments as @code{whatis}, but prints a
12753 detailed description of the type, instead of just the name of the type.
12754 @xref{Expressions, ,Expressions}.
12756 For example, for this variable declaration:
12759 struct complex @{double real; double imag;@} v;
12763 the two commands give this output:
12767 (@value{GDBP}) whatis v
12768 type = struct complex
12769 (@value{GDBP}) ptype v
12770 type = struct complex @{
12778 As with @code{whatis}, using @code{ptype} without an argument refers to
12779 the type of @code{$}, the last value in the value history.
12781 @cindex incomplete type
12782 Sometimes, programs use opaque data types or incomplete specifications
12783 of complex data structure. If the debug information included in the
12784 program does not allow @value{GDBN} to display a full declaration of
12785 the data type, it will say @samp{<incomplete type>}. For example,
12786 given these declarations:
12790 struct foo *fooptr;
12794 but no definition for @code{struct foo} itself, @value{GDBN} will say:
12797 (@value{GDBP}) ptype foo
12798 $1 = <incomplete type>
12802 ``Incomplete type'' is C terminology for data types that are not
12803 completely specified.
12806 @item info types @var{regexp}
12808 Print a brief description of all types whose names match the regular
12809 expression @var{regexp} (or all types in your program, if you supply
12810 no argument). Each complete typename is matched as though it were a
12811 complete line; thus, @samp{i type value} gives information on all
12812 types in your program whose names include the string @code{value}, but
12813 @samp{i type ^value$} gives information only on types whose complete
12814 name is @code{value}.
12816 This command differs from @code{ptype} in two ways: first, like
12817 @code{whatis}, it does not print a detailed description; second, it
12818 lists all source files where a type is defined.
12821 @cindex local variables
12822 @item info scope @var{location}
12823 List all the variables local to a particular scope. This command
12824 accepts a @var{location} argument---a function name, a source line, or
12825 an address preceded by a @samp{*}, and prints all the variables local
12826 to the scope defined by that location. (@xref{Specify Location}, for
12827 details about supported forms of @var{location}.) For example:
12830 (@value{GDBP}) @b{info scope command_line_handler}
12831 Scope for command_line_handler:
12832 Symbol rl is an argument at stack/frame offset 8, length 4.
12833 Symbol linebuffer is in static storage at address 0x150a18, length 4.
12834 Symbol linelength is in static storage at address 0x150a1c, length 4.
12835 Symbol p is a local variable in register $esi, length 4.
12836 Symbol p1 is a local variable in register $ebx, length 4.
12837 Symbol nline is a local variable in register $edx, length 4.
12838 Symbol repeat is a local variable at frame offset -8, length 4.
12842 This command is especially useful for determining what data to collect
12843 during a @dfn{trace experiment}, see @ref{Tracepoint Actions,
12846 @kindex info source
12848 Show information about the current source file---that is, the source file for
12849 the function containing the current point of execution:
12852 the name of the source file, and the directory containing it,
12854 the directory it was compiled in,
12856 its length, in lines,
12858 which programming language it is written in,
12860 whether the executable includes debugging information for that file, and
12861 if so, what format the information is in (e.g., STABS, Dwarf 2, etc.), and
12863 whether the debugging information includes information about
12864 preprocessor macros.
12868 @kindex info sources
12870 Print the names of all source files in your program for which there is
12871 debugging information, organized into two lists: files whose symbols
12872 have already been read, and files whose symbols will be read when needed.
12874 @kindex info functions
12875 @item info functions
12876 Print the names and data types of all defined functions.
12878 @item info functions @var{regexp}
12879 Print the names and data types of all defined functions
12880 whose names contain a match for regular expression @var{regexp}.
12881 Thus, @samp{info fun step} finds all functions whose names
12882 include @code{step}; @samp{info fun ^step} finds those whose names
12883 start with @code{step}. If a function name contains characters
12884 that conflict with the regular expression language (e.g.@:
12885 @samp{operator*()}), they may be quoted with a backslash.
12887 @kindex info variables
12888 @item info variables
12889 Print the names and data types of all variables that are declared
12890 outside of functions (i.e.@: excluding local variables).
12892 @item info variables @var{regexp}
12893 Print the names and data types of all variables (except for local
12894 variables) whose names contain a match for regular expression
12897 @kindex info classes
12898 @cindex Objective-C, classes and selectors
12900 @itemx info classes @var{regexp}
12901 Display all Objective-C classes in your program, or
12902 (with the @var{regexp} argument) all those matching a particular regular
12905 @kindex info selectors
12906 @item info selectors
12907 @itemx info selectors @var{regexp}
12908 Display all Objective-C selectors in your program, or
12909 (with the @var{regexp} argument) all those matching a particular regular
12913 This was never implemented.
12914 @kindex info methods
12916 @itemx info methods @var{regexp}
12917 The @code{info methods} command permits the user to examine all defined
12918 methods within C@t{++} program, or (with the @var{regexp} argument) a
12919 specific set of methods found in the various C@t{++} classes. Many
12920 C@t{++} classes provide a large number of methods. Thus, the output
12921 from the @code{ptype} command can be overwhelming and hard to use. The
12922 @code{info-methods} command filters the methods, printing only those
12923 which match the regular-expression @var{regexp}.
12926 @cindex reloading symbols
12927 Some systems allow individual object files that make up your program to
12928 be replaced without stopping and restarting your program. For example,
12929 in VxWorks you can simply recompile a defective object file and keep on
12930 running. If you are running on one of these systems, you can allow
12931 @value{GDBN} to reload the symbols for automatically relinked modules:
12934 @kindex set symbol-reloading
12935 @item set symbol-reloading on
12936 Replace symbol definitions for the corresponding source file when an
12937 object file with a particular name is seen again.
12939 @item set symbol-reloading off
12940 Do not replace symbol definitions when encountering object files of the
12941 same name more than once. This is the default state; if you are not
12942 running on a system that permits automatic relinking of modules, you
12943 should leave @code{symbol-reloading} off, since otherwise @value{GDBN}
12944 may discard symbols when linking large programs, that may contain
12945 several modules (from different directories or libraries) with the same
12948 @kindex show symbol-reloading
12949 @item show symbol-reloading
12950 Show the current @code{on} or @code{off} setting.
12953 @cindex opaque data types
12954 @kindex set opaque-type-resolution
12955 @item set opaque-type-resolution on
12956 Tell @value{GDBN} to resolve opaque types. An opaque type is a type
12957 declared as a pointer to a @code{struct}, @code{class}, or
12958 @code{union}---for example, @code{struct MyType *}---that is used in one
12959 source file although the full declaration of @code{struct MyType} is in
12960 another source file. The default is on.
12962 A change in the setting of this subcommand will not take effect until
12963 the next time symbols for a file are loaded.
12965 @item set opaque-type-resolution off
12966 Tell @value{GDBN} not to resolve opaque types. In this case, the type
12967 is printed as follows:
12969 @{<no data fields>@}
12972 @kindex show opaque-type-resolution
12973 @item show opaque-type-resolution
12974 Show whether opaque types are resolved or not.
12976 @kindex maint print symbols
12977 @cindex symbol dump
12978 @kindex maint print psymbols
12979 @cindex partial symbol dump
12980 @item maint print symbols @var{filename}
12981 @itemx maint print psymbols @var{filename}
12982 @itemx maint print msymbols @var{filename}
12983 Write a dump of debugging symbol data into the file @var{filename}.
12984 These commands are used to debug the @value{GDBN} symbol-reading code. Only
12985 symbols with debugging data are included. If you use @samp{maint print
12986 symbols}, @value{GDBN} includes all the symbols for which it has already
12987 collected full details: that is, @var{filename} reflects symbols for
12988 only those files whose symbols @value{GDBN} has read. You can use the
12989 command @code{info sources} to find out which files these are. If you
12990 use @samp{maint print psymbols} instead, the dump shows information about
12991 symbols that @value{GDBN} only knows partially---that is, symbols defined in
12992 files that @value{GDBN} has skimmed, but not yet read completely. Finally,
12993 @samp{maint print msymbols} dumps just the minimal symbol information
12994 required for each object file from which @value{GDBN} has read some symbols.
12995 @xref{Files, ,Commands to Specify Files}, for a discussion of how
12996 @value{GDBN} reads symbols (in the description of @code{symbol-file}).
12998 @kindex maint info symtabs
12999 @kindex maint info psymtabs
13000 @cindex listing @value{GDBN}'s internal symbol tables
13001 @cindex symbol tables, listing @value{GDBN}'s internal
13002 @cindex full symbol tables, listing @value{GDBN}'s internal
13003 @cindex partial symbol tables, listing @value{GDBN}'s internal
13004 @item maint info symtabs @r{[} @var{regexp} @r{]}
13005 @itemx maint info psymtabs @r{[} @var{regexp} @r{]}
13007 List the @code{struct symtab} or @code{struct partial_symtab}
13008 structures whose names match @var{regexp}. If @var{regexp} is not
13009 given, list them all. The output includes expressions which you can
13010 copy into a @value{GDBN} debugging this one to examine a particular
13011 structure in more detail. For example:
13014 (@value{GDBP}) maint info psymtabs dwarf2read
13015 @{ objfile /home/gnu/build/gdb/gdb
13016 ((struct objfile *) 0x82e69d0)
13017 @{ psymtab /home/gnu/src/gdb/dwarf2read.c
13018 ((struct partial_symtab *) 0x8474b10)
13021 text addresses 0x814d3c8 -- 0x8158074
13022 globals (* (struct partial_symbol **) 0x8507a08 @@ 9)
13023 statics (* (struct partial_symbol **) 0x40e95b78 @@ 2882)
13024 dependencies (none)
13027 (@value{GDBP}) maint info symtabs
13031 We see that there is one partial symbol table whose filename contains
13032 the string @samp{dwarf2read}, belonging to the @samp{gdb} executable;
13033 and we see that @value{GDBN} has not read in any symtabs yet at all.
13034 If we set a breakpoint on a function, that will cause @value{GDBN} to
13035 read the symtab for the compilation unit containing that function:
13038 (@value{GDBP}) break dwarf2_psymtab_to_symtab
13039 Breakpoint 1 at 0x814e5da: file /home/gnu/src/gdb/dwarf2read.c,
13041 (@value{GDBP}) maint info symtabs
13042 @{ objfile /home/gnu/build/gdb/gdb
13043 ((struct objfile *) 0x82e69d0)
13044 @{ symtab /home/gnu/src/gdb/dwarf2read.c
13045 ((struct symtab *) 0x86c1f38)
13048 blockvector ((struct blockvector *) 0x86c1bd0) (primary)
13049 linetable ((struct linetable *) 0x8370fa0)
13050 debugformat DWARF 2
13059 @chapter Altering Execution
13061 Once you think you have found an error in your program, you might want to
13062 find out for certain whether correcting the apparent error would lead to
13063 correct results in the rest of the run. You can find the answer by
13064 experiment, using the @value{GDBN} features for altering execution of the
13067 For example, you can store new values into variables or memory
13068 locations, give your program a signal, restart it at a different
13069 address, or even return prematurely from a function.
13072 * Assignment:: Assignment to variables
13073 * Jumping:: Continuing at a different address
13074 * Signaling:: Giving your program a signal
13075 * Returning:: Returning from a function
13076 * Calling:: Calling your program's functions
13077 * Patching:: Patching your program
13081 @section Assignment to Variables
13084 @cindex setting variables
13085 To alter the value of a variable, evaluate an assignment expression.
13086 @xref{Expressions, ,Expressions}. For example,
13093 stores the value 4 into the variable @code{x}, and then prints the
13094 value of the assignment expression (which is 4).
13095 @xref{Languages, ,Using @value{GDBN} with Different Languages}, for more
13096 information on operators in supported languages.
13098 @kindex set variable
13099 @cindex variables, setting
13100 If you are not interested in seeing the value of the assignment, use the
13101 @code{set} command instead of the @code{print} command. @code{set} is
13102 really the same as @code{print} except that the expression's value is
13103 not printed and is not put in the value history (@pxref{Value History,
13104 ,Value History}). The expression is evaluated only for its effects.
13106 If the beginning of the argument string of the @code{set} command
13107 appears identical to a @code{set} subcommand, use the @code{set
13108 variable} command instead of just @code{set}. This command is identical
13109 to @code{set} except for its lack of subcommands. For example, if your
13110 program has a variable @code{width}, you get an error if you try to set
13111 a new value with just @samp{set width=13}, because @value{GDBN} has the
13112 command @code{set width}:
13115 (@value{GDBP}) whatis width
13117 (@value{GDBP}) p width
13119 (@value{GDBP}) set width=47
13120 Invalid syntax in expression.
13124 The invalid expression, of course, is @samp{=47}. In
13125 order to actually set the program's variable @code{width}, use
13128 (@value{GDBP}) set var width=47
13131 Because the @code{set} command has many subcommands that can conflict
13132 with the names of program variables, it is a good idea to use the
13133 @code{set variable} command instead of just @code{set}. For example, if
13134 your program has a variable @code{g}, you run into problems if you try
13135 to set a new value with just @samp{set g=4}, because @value{GDBN} has
13136 the command @code{set gnutarget}, abbreviated @code{set g}:
13140 (@value{GDBP}) whatis g
13144 (@value{GDBP}) set g=4
13148 The program being debugged has been started already.
13149 Start it from the beginning? (y or n) y
13150 Starting program: /home/smith/cc_progs/a.out
13151 "/home/smith/cc_progs/a.out": can't open to read symbols:
13152 Invalid bfd target.
13153 (@value{GDBP}) show g
13154 The current BFD target is "=4".
13159 The program variable @code{g} did not change, and you silently set the
13160 @code{gnutarget} to an invalid value. In order to set the variable
13164 (@value{GDBP}) set var g=4
13167 @value{GDBN} allows more implicit conversions in assignments than C; you can
13168 freely store an integer value into a pointer variable or vice versa,
13169 and you can convert any structure to any other structure that is the
13170 same length or shorter.
13171 @comment FIXME: how do structs align/pad in these conversions?
13172 @comment /doc@cygnus.com 18dec1990
13174 To store values into arbitrary places in memory, use the @samp{@{@dots{}@}}
13175 construct to generate a value of specified type at a specified address
13176 (@pxref{Expressions, ,Expressions}). For example, @code{@{int@}0x83040} refers
13177 to memory location @code{0x83040} as an integer (which implies a certain size
13178 and representation in memory), and
13181 set @{int@}0x83040 = 4
13185 stores the value 4 into that memory location.
13188 @section Continuing at a Different Address
13190 Ordinarily, when you continue your program, you do so at the place where
13191 it stopped, with the @code{continue} command. You can instead continue at
13192 an address of your own choosing, with the following commands:
13196 @item jump @var{linespec}
13197 @itemx jump @var{location}
13198 Resume execution at line @var{linespec} or at address given by
13199 @var{location}. Execution stops again immediately if there is a
13200 breakpoint there. @xref{Specify Location}, for a description of the
13201 different forms of @var{linespec} and @var{location}. It is common
13202 practice to use the @code{tbreak} command in conjunction with
13203 @code{jump}. @xref{Set Breaks, ,Setting Breakpoints}.
13205 The @code{jump} command does not change the current stack frame, or
13206 the stack pointer, or the contents of any memory location or any
13207 register other than the program counter. If line @var{linespec} is in
13208 a different function from the one currently executing, the results may
13209 be bizarre if the two functions expect different patterns of arguments or
13210 of local variables. For this reason, the @code{jump} command requests
13211 confirmation if the specified line is not in the function currently
13212 executing. However, even bizarre results are predictable if you are
13213 well acquainted with the machine-language code of your program.
13216 @c Doesn't work on HP-UX; have to set $pcoqh and $pcoqt.
13217 On many systems, you can get much the same effect as the @code{jump}
13218 command by storing a new value into the register @code{$pc}. The
13219 difference is that this does not start your program running; it only
13220 changes the address of where it @emph{will} run when you continue. For
13228 makes the next @code{continue} command or stepping command execute at
13229 address @code{0x485}, rather than at the address where your program stopped.
13230 @xref{Continuing and Stepping, ,Continuing and Stepping}.
13232 The most common occasion to use the @code{jump} command is to back
13233 up---perhaps with more breakpoints set---over a portion of a program
13234 that has already executed, in order to examine its execution in more
13239 @section Giving your Program a Signal
13240 @cindex deliver a signal to a program
13244 @item signal @var{signal}
13245 Resume execution where your program stopped, but immediately give it the
13246 signal @var{signal}. @var{signal} can be the name or the number of a
13247 signal. For example, on many systems @code{signal 2} and @code{signal
13248 SIGINT} are both ways of sending an interrupt signal.
13250 Alternatively, if @var{signal} is zero, continue execution without
13251 giving a signal. This is useful when your program stopped on account of
13252 a signal and would ordinary see the signal when resumed with the
13253 @code{continue} command; @samp{signal 0} causes it to resume without a
13256 @code{signal} does not repeat when you press @key{RET} a second time
13257 after executing the command.
13261 Invoking the @code{signal} command is not the same as invoking the
13262 @code{kill} utility from the shell. Sending a signal with @code{kill}
13263 causes @value{GDBN} to decide what to do with the signal depending on
13264 the signal handling tables (@pxref{Signals}). The @code{signal} command
13265 passes the signal directly to your program.
13269 @section Returning from a Function
13272 @cindex returning from a function
13275 @itemx return @var{expression}
13276 You can cancel execution of a function call with the @code{return}
13277 command. If you give an
13278 @var{expression} argument, its value is used as the function's return
13282 When you use @code{return}, @value{GDBN} discards the selected stack frame
13283 (and all frames within it). You can think of this as making the
13284 discarded frame return prematurely. If you wish to specify a value to
13285 be returned, give that value as the argument to @code{return}.
13287 This pops the selected stack frame (@pxref{Selection, ,Selecting a
13288 Frame}), and any other frames inside of it, leaving its caller as the
13289 innermost remaining frame. That frame becomes selected. The
13290 specified value is stored in the registers used for returning values
13293 The @code{return} command does not resume execution; it leaves the
13294 program stopped in the state that would exist if the function had just
13295 returned. In contrast, the @code{finish} command (@pxref{Continuing
13296 and Stepping, ,Continuing and Stepping}) resumes execution until the
13297 selected stack frame returns naturally.
13299 @value{GDBN} needs to know how the @var{expression} argument should be set for
13300 the inferior. The concrete registers assignment depends on the OS ABI and the
13301 type being returned by the selected stack frame. For example it is common for
13302 OS ABI to return floating point values in FPU registers while integer values in
13303 CPU registers. Still some ABIs return even floating point values in CPU
13304 registers. Larger integer widths (such as @code{long long int}) also have
13305 specific placement rules. @value{GDBN} already knows the OS ABI from its
13306 current target so it needs to find out also the type being returned to make the
13307 assignment into the right register(s).
13309 Normally, the selected stack frame has debug info. @value{GDBN} will always
13310 use the debug info instead of the implicit type of @var{expression} when the
13311 debug info is available. For example, if you type @kbd{return -1}, and the
13312 function in the current stack frame is declared to return a @code{long long
13313 int}, @value{GDBN} transparently converts the implicit @code{int} value of -1
13314 into a @code{long long int}:
13317 Breakpoint 1, func () at gdb.base/return-nodebug.c:29
13319 (@value{GDBP}) return -1
13320 Make func return now? (y or n) y
13321 #0 0x004004f6 in main () at gdb.base/return-nodebug.c:43
13322 43 printf ("result=%lld\n", func ());
13326 However, if the selected stack frame does not have a debug info, e.g., if the
13327 function was compiled without debug info, @value{GDBN} has to find out the type
13328 to return from user. Specifying a different type by mistake may set the value
13329 in different inferior registers than the caller code expects. For example,
13330 typing @kbd{return -1} with its implicit type @code{int} would set only a part
13331 of a @code{long long int} result for a debug info less function (on 32-bit
13332 architectures). Therefore the user is required to specify the return type by
13333 an appropriate cast explicitly:
13336 Breakpoint 2, 0x0040050b in func ()
13337 (@value{GDBP}) return -1
13338 Return value type not available for selected stack frame.
13339 Please use an explicit cast of the value to return.
13340 (@value{GDBP}) return (long long int) -1
13341 Make selected stack frame return now? (y or n) y
13342 #0 0x00400526 in main ()
13347 @section Calling Program Functions
13350 @cindex calling functions
13351 @cindex inferior functions, calling
13352 @item print @var{expr}
13353 Evaluate the expression @var{expr} and display the resulting value.
13354 @var{expr} may include calls to functions in the program being
13358 @item call @var{expr}
13359 Evaluate the expression @var{expr} without displaying @code{void}
13362 You can use this variant of the @code{print} command if you want to
13363 execute a function from your program that does not return anything
13364 (a.k.a.@: @dfn{a void function}), but without cluttering the output
13365 with @code{void} returned values that @value{GDBN} will otherwise
13366 print. If the result is not void, it is printed and saved in the
13370 It is possible for the function you call via the @code{print} or
13371 @code{call} command to generate a signal (e.g., if there's a bug in
13372 the function, or if you passed it incorrect arguments). What happens
13373 in that case is controlled by the @code{set unwindonsignal} command.
13375 Similarly, with a C@t{++} program it is possible for the function you
13376 call via the @code{print} or @code{call} command to generate an
13377 exception that is not handled due to the constraints of the dummy
13378 frame. In this case, any exception that is raised in the frame, but has
13379 an out-of-frame exception handler will not be found. GDB builds a
13380 dummy-frame for the inferior function call, and the unwinder cannot
13381 seek for exception handlers outside of this dummy-frame. What happens
13382 in that case is controlled by the
13383 @code{set unwind-on-terminating-exception} command.
13386 @item set unwindonsignal
13387 @kindex set unwindonsignal
13388 @cindex unwind stack in called functions
13389 @cindex call dummy stack unwinding
13390 Set unwinding of the stack if a signal is received while in a function
13391 that @value{GDBN} called in the program being debugged. If set to on,
13392 @value{GDBN} unwinds the stack it created for the call and restores
13393 the context to what it was before the call. If set to off (the
13394 default), @value{GDBN} stops in the frame where the signal was
13397 @item show unwindonsignal
13398 @kindex show unwindonsignal
13399 Show the current setting of stack unwinding in the functions called by
13402 @item set unwind-on-terminating-exception
13403 @kindex set unwind-on-terminating-exception
13404 @cindex unwind stack in called functions with unhandled exceptions
13405 @cindex call dummy stack unwinding on unhandled exception.
13406 Set unwinding of the stack if a C@t{++} exception is raised, but left
13407 unhandled while in a function that @value{GDBN} called in the program being
13408 debugged. If set to on (the default), @value{GDBN} unwinds the stack
13409 it created for the call and restores the context to what it was before
13410 the call. If set to off, @value{GDBN} the exception is delivered to
13411 the default C@t{++} exception handler and the inferior terminated.
13413 @item show unwind-on-terminating-exception
13414 @kindex show unwind-on-terminating-exception
13415 Show the current setting of stack unwinding in the functions called by
13420 @cindex weak alias functions
13421 Sometimes, a function you wish to call is actually a @dfn{weak alias}
13422 for another function. In such case, @value{GDBN} might not pick up
13423 the type information, including the types of the function arguments,
13424 which causes @value{GDBN} to call the inferior function incorrectly.
13425 As a result, the called function will function erroneously and may
13426 even crash. A solution to that is to use the name of the aliased
13430 @section Patching Programs
13432 @cindex patching binaries
13433 @cindex writing into executables
13434 @cindex writing into corefiles
13436 By default, @value{GDBN} opens the file containing your program's
13437 executable code (or the corefile) read-only. This prevents accidental
13438 alterations to machine code; but it also prevents you from intentionally
13439 patching your program's binary.
13441 If you'd like to be able to patch the binary, you can specify that
13442 explicitly with the @code{set write} command. For example, you might
13443 want to turn on internal debugging flags, or even to make emergency
13449 @itemx set write off
13450 If you specify @samp{set write on}, @value{GDBN} opens executable and
13451 core files for both reading and writing; if you specify @kbd{set write
13452 off} (the default), @value{GDBN} opens them read-only.
13454 If you have already loaded a file, you must load it again (using the
13455 @code{exec-file} or @code{core-file} command) after changing @code{set
13456 write}, for your new setting to take effect.
13460 Display whether executable files and core files are opened for writing
13461 as well as reading.
13465 @chapter @value{GDBN} Files
13467 @value{GDBN} needs to know the file name of the program to be debugged,
13468 both in order to read its symbol table and in order to start your
13469 program. To debug a core dump of a previous run, you must also tell
13470 @value{GDBN} the name of the core dump file.
13473 * Files:: Commands to specify files
13474 * Separate Debug Files:: Debugging information in separate files
13475 * Symbol Errors:: Errors reading symbol files
13476 * Data Files:: GDB data files
13480 @section Commands to Specify Files
13482 @cindex symbol table
13483 @cindex core dump file
13485 You may want to specify executable and core dump file names. The usual
13486 way to do this is at start-up time, using the arguments to
13487 @value{GDBN}'s start-up commands (@pxref{Invocation, , Getting In and
13488 Out of @value{GDBN}}).
13490 Occasionally it is necessary to change to a different file during a
13491 @value{GDBN} session. Or you may run @value{GDBN} and forget to
13492 specify a file you want to use. Or you are debugging a remote target
13493 via @code{gdbserver} (@pxref{Server, file, Using the @code{gdbserver}
13494 Program}). In these situations the @value{GDBN} commands to specify
13495 new files are useful.
13498 @cindex executable file
13500 @item file @var{filename}
13501 Use @var{filename} as the program to be debugged. It is read for its
13502 symbols and for the contents of pure memory. It is also the program
13503 executed when you use the @code{run} command. If you do not specify a
13504 directory and the file is not found in the @value{GDBN} working directory,
13505 @value{GDBN} uses the environment variable @code{PATH} as a list of
13506 directories to search, just as the shell does when looking for a program
13507 to run. You can change the value of this variable, for both @value{GDBN}
13508 and your program, using the @code{path} command.
13510 @cindex unlinked object files
13511 @cindex patching object files
13512 You can load unlinked object @file{.o} files into @value{GDBN} using
13513 the @code{file} command. You will not be able to ``run'' an object
13514 file, but you can disassemble functions and inspect variables. Also,
13515 if the underlying BFD functionality supports it, you could use
13516 @kbd{gdb -write} to patch object files using this technique. Note
13517 that @value{GDBN} can neither interpret nor modify relocations in this
13518 case, so branches and some initialized variables will appear to go to
13519 the wrong place. But this feature is still handy from time to time.
13522 @code{file} with no argument makes @value{GDBN} discard any information it
13523 has on both executable file and the symbol table.
13526 @item exec-file @r{[} @var{filename} @r{]}
13527 Specify that the program to be run (but not the symbol table) is found
13528 in @var{filename}. @value{GDBN} searches the environment variable @code{PATH}
13529 if necessary to locate your program. Omitting @var{filename} means to
13530 discard information on the executable file.
13532 @kindex symbol-file
13533 @item symbol-file @r{[} @var{filename} @r{]}
13534 Read symbol table information from file @var{filename}. @code{PATH} is
13535 searched when necessary. Use the @code{file} command to get both symbol
13536 table and program to run from the same file.
13538 @code{symbol-file} with no argument clears out @value{GDBN} information on your
13539 program's symbol table.
13541 The @code{symbol-file} command causes @value{GDBN} to forget the contents of
13542 some breakpoints and auto-display expressions. This is because they may
13543 contain pointers to the internal data recording symbols and data types,
13544 which are part of the old symbol table data being discarded inside
13547 @code{symbol-file} does not repeat if you press @key{RET} again after
13550 When @value{GDBN} is configured for a particular environment, it
13551 understands debugging information in whatever format is the standard
13552 generated for that environment; you may use either a @sc{gnu} compiler, or
13553 other compilers that adhere to the local conventions.
13554 Best results are usually obtained from @sc{gnu} compilers; for example,
13555 using @code{@value{NGCC}} you can generate debugging information for
13558 For most kinds of object files, with the exception of old SVR3 systems
13559 using COFF, the @code{symbol-file} command does not normally read the
13560 symbol table in full right away. Instead, it scans the symbol table
13561 quickly to find which source files and which symbols are present. The
13562 details are read later, one source file at a time, as they are needed.
13564 The purpose of this two-stage reading strategy is to make @value{GDBN}
13565 start up faster. For the most part, it is invisible except for
13566 occasional pauses while the symbol table details for a particular source
13567 file are being read. (The @code{set verbose} command can turn these
13568 pauses into messages if desired. @xref{Messages/Warnings, ,Optional
13569 Warnings and Messages}.)
13571 We have not implemented the two-stage strategy for COFF yet. When the
13572 symbol table is stored in COFF format, @code{symbol-file} reads the
13573 symbol table data in full right away. Note that ``stabs-in-COFF''
13574 still does the two-stage strategy, since the debug info is actually
13578 @cindex reading symbols immediately
13579 @cindex symbols, reading immediately
13580 @item symbol-file @var{filename} @r{[} -readnow @r{]}
13581 @itemx file @var{filename} @r{[} -readnow @r{]}
13582 You can override the @value{GDBN} two-stage strategy for reading symbol
13583 tables by using the @samp{-readnow} option with any of the commands that
13584 load symbol table information, if you want to be sure @value{GDBN} has the
13585 entire symbol table available.
13587 @c FIXME: for now no mention of directories, since this seems to be in
13588 @c flux. 13mar1992 status is that in theory GDB would look either in
13589 @c current dir or in same dir as myprog; but issues like competing
13590 @c GDB's, or clutter in system dirs, mean that in practice right now
13591 @c only current dir is used. FFish says maybe a special GDB hierarchy
13592 @c (eg rooted in val of env var GDBSYMS) could exist for mappable symbol
13596 @item core-file @r{[}@var{filename}@r{]}
13598 Specify the whereabouts of a core dump file to be used as the ``contents
13599 of memory''. Traditionally, core files contain only some parts of the
13600 address space of the process that generated them; @value{GDBN} can access the
13601 executable file itself for other parts.
13603 @code{core-file} with no argument specifies that no core file is
13606 Note that the core file is ignored when your program is actually running
13607 under @value{GDBN}. So, if you have been running your program and you
13608 wish to debug a core file instead, you must kill the subprocess in which
13609 the program is running. To do this, use the @code{kill} command
13610 (@pxref{Kill Process, ,Killing the Child Process}).
13612 @kindex add-symbol-file
13613 @cindex dynamic linking
13614 @item add-symbol-file @var{filename} @var{address}
13615 @itemx add-symbol-file @var{filename} @var{address} @r{[} -readnow @r{]}
13616 @itemx add-symbol-file @var{filename} @r{-s}@var{section} @var{address} @dots{}
13617 The @code{add-symbol-file} command reads additional symbol table
13618 information from the file @var{filename}. You would use this command
13619 when @var{filename} has been dynamically loaded (by some other means)
13620 into the program that is running. @var{address} should be the memory
13621 address at which the file has been loaded; @value{GDBN} cannot figure
13622 this out for itself. You can additionally specify an arbitrary number
13623 of @samp{@r{-s}@var{section} @var{address}} pairs, to give an explicit
13624 section name and base address for that section. You can specify any
13625 @var{address} as an expression.
13627 The symbol table of the file @var{filename} is added to the symbol table
13628 originally read with the @code{symbol-file} command. You can use the
13629 @code{add-symbol-file} command any number of times; the new symbol data
13630 thus read keeps adding to the old. To discard all old symbol data
13631 instead, use the @code{symbol-file} command without any arguments.
13633 @cindex relocatable object files, reading symbols from
13634 @cindex object files, relocatable, reading symbols from
13635 @cindex reading symbols from relocatable object files
13636 @cindex symbols, reading from relocatable object files
13637 @cindex @file{.o} files, reading symbols from
13638 Although @var{filename} is typically a shared library file, an
13639 executable file, or some other object file which has been fully
13640 relocated for loading into a process, you can also load symbolic
13641 information from relocatable @file{.o} files, as long as:
13645 the file's symbolic information refers only to linker symbols defined in
13646 that file, not to symbols defined by other object files,
13648 every section the file's symbolic information refers to has actually
13649 been loaded into the inferior, as it appears in the file, and
13651 you can determine the address at which every section was loaded, and
13652 provide these to the @code{add-symbol-file} command.
13656 Some embedded operating systems, like Sun Chorus and VxWorks, can load
13657 relocatable files into an already running program; such systems
13658 typically make the requirements above easy to meet. However, it's
13659 important to recognize that many native systems use complex link
13660 procedures (@code{.linkonce} section factoring and C@t{++} constructor table
13661 assembly, for example) that make the requirements difficult to meet. In
13662 general, one cannot assume that using @code{add-symbol-file} to read a
13663 relocatable object file's symbolic information will have the same effect
13664 as linking the relocatable object file into the program in the normal
13667 @code{add-symbol-file} does not repeat if you press @key{RET} after using it.
13669 @kindex add-symbol-file-from-memory
13670 @cindex @code{syscall DSO}
13671 @cindex load symbols from memory
13672 @item add-symbol-file-from-memory @var{address}
13673 Load symbols from the given @var{address} in a dynamically loaded
13674 object file whose image is mapped directly into the inferior's memory.
13675 For example, the Linux kernel maps a @code{syscall DSO} into each
13676 process's address space; this DSO provides kernel-specific code for
13677 some system calls. The argument can be any expression whose
13678 evaluation yields the address of the file's shared object file header.
13679 For this command to work, you must have used @code{symbol-file} or
13680 @code{exec-file} commands in advance.
13682 @kindex add-shared-symbol-files
13684 @item add-shared-symbol-files @var{library-file}
13685 @itemx assf @var{library-file}
13686 The @code{add-shared-symbol-files} command can currently be used only
13687 in the Cygwin build of @value{GDBN} on MS-Windows OS, where it is an
13688 alias for the @code{dll-symbols} command (@pxref{Cygwin Native}).
13689 @value{GDBN} automatically looks for shared libraries, however if
13690 @value{GDBN} does not find yours, you can invoke
13691 @code{add-shared-symbol-files}. It takes one argument: the shared
13692 library's file name. @code{assf} is a shorthand alias for
13693 @code{add-shared-symbol-files}.
13696 @item section @var{section} @var{addr}
13697 The @code{section} command changes the base address of the named
13698 @var{section} of the exec file to @var{addr}. This can be used if the
13699 exec file does not contain section addresses, (such as in the
13700 @code{a.out} format), or when the addresses specified in the file
13701 itself are wrong. Each section must be changed separately. The
13702 @code{info files} command, described below, lists all the sections and
13706 @kindex info target
13709 @code{info files} and @code{info target} are synonymous; both print the
13710 current target (@pxref{Targets, ,Specifying a Debugging Target}),
13711 including the names of the executable and core dump files currently in
13712 use by @value{GDBN}, and the files from which symbols were loaded. The
13713 command @code{help target} lists all possible targets rather than
13716 @kindex maint info sections
13717 @item maint info sections
13718 Another command that can give you extra information about program sections
13719 is @code{maint info sections}. In addition to the section information
13720 displayed by @code{info files}, this command displays the flags and file
13721 offset of each section in the executable and core dump files. In addition,
13722 @code{maint info sections} provides the following command options (which
13723 may be arbitrarily combined):
13727 Display sections for all loaded object files, including shared libraries.
13728 @item @var{sections}
13729 Display info only for named @var{sections}.
13730 @item @var{section-flags}
13731 Display info only for sections for which @var{section-flags} are true.
13732 The section flags that @value{GDBN} currently knows about are:
13735 Section will have space allocated in the process when loaded.
13736 Set for all sections except those containing debug information.
13738 Section will be loaded from the file into the child process memory.
13739 Set for pre-initialized code and data, clear for @code{.bss} sections.
13741 Section needs to be relocated before loading.
13743 Section cannot be modified by the child process.
13745 Section contains executable code only.
13747 Section contains data only (no executable code).
13749 Section will reside in ROM.
13751 Section contains data for constructor/destructor lists.
13753 Section is not empty.
13755 An instruction to the linker to not output the section.
13756 @item COFF_SHARED_LIBRARY
13757 A notification to the linker that the section contains
13758 COFF shared library information.
13760 Section contains common symbols.
13763 @kindex set trust-readonly-sections
13764 @cindex read-only sections
13765 @item set trust-readonly-sections on
13766 Tell @value{GDBN} that readonly sections in your object file
13767 really are read-only (i.e.@: that their contents will not change).
13768 In that case, @value{GDBN} can fetch values from these sections
13769 out of the object file, rather than from the target program.
13770 For some targets (notably embedded ones), this can be a significant
13771 enhancement to debugging performance.
13773 The default is off.
13775 @item set trust-readonly-sections off
13776 Tell @value{GDBN} not to trust readonly sections. This means that
13777 the contents of the section might change while the program is running,
13778 and must therefore be fetched from the target when needed.
13780 @item show trust-readonly-sections
13781 Show the current setting of trusting readonly sections.
13784 All file-specifying commands allow both absolute and relative file names
13785 as arguments. @value{GDBN} always converts the file name to an absolute file
13786 name and remembers it that way.
13788 @cindex shared libraries
13789 @anchor{Shared Libraries}
13790 @value{GDBN} supports @sc{gnu}/Linux, MS-Windows, HP-UX, SunOS, SVr4, Irix,
13791 and IBM RS/6000 AIX shared libraries.
13793 On MS-Windows @value{GDBN} must be linked with the Expat library to support
13794 shared libraries. @xref{Expat}.
13796 @value{GDBN} automatically loads symbol definitions from shared libraries
13797 when you use the @code{run} command, or when you examine a core file.
13798 (Before you issue the @code{run} command, @value{GDBN} does not understand
13799 references to a function in a shared library, however---unless you are
13800 debugging a core file).
13802 On HP-UX, if the program loads a library explicitly, @value{GDBN}
13803 automatically loads the symbols at the time of the @code{shl_load} call.
13805 @c FIXME: some @value{GDBN} release may permit some refs to undef
13806 @c FIXME...symbols---eg in a break cmd---assuming they are from a shared
13807 @c FIXME...lib; check this from time to time when updating manual
13809 There are times, however, when you may wish to not automatically load
13810 symbol definitions from shared libraries, such as when they are
13811 particularly large or there are many of them.
13813 To control the automatic loading of shared library symbols, use the
13817 @kindex set auto-solib-add
13818 @item set auto-solib-add @var{mode}
13819 If @var{mode} is @code{on}, symbols from all shared object libraries
13820 will be loaded automatically when the inferior begins execution, you
13821 attach to an independently started inferior, or when the dynamic linker
13822 informs @value{GDBN} that a new library has been loaded. If @var{mode}
13823 is @code{off}, symbols must be loaded manually, using the
13824 @code{sharedlibrary} command. The default value is @code{on}.
13826 @cindex memory used for symbol tables
13827 If your program uses lots of shared libraries with debug info that
13828 takes large amounts of memory, you can decrease the @value{GDBN}
13829 memory footprint by preventing it from automatically loading the
13830 symbols from shared libraries. To that end, type @kbd{set
13831 auto-solib-add off} before running the inferior, then load each
13832 library whose debug symbols you do need with @kbd{sharedlibrary
13833 @var{regexp}}, where @var{regexp} is a regular expression that matches
13834 the libraries whose symbols you want to be loaded.
13836 @kindex show auto-solib-add
13837 @item show auto-solib-add
13838 Display the current autoloading mode.
13841 @cindex load shared library
13842 To explicitly load shared library symbols, use the @code{sharedlibrary}
13846 @kindex info sharedlibrary
13848 @item info share @var{regex}
13849 @itemx info sharedlibrary @var{regex}
13850 Print the names of the shared libraries which are currently loaded
13851 that match @var{regex}. If @var{regex} is omitted then print
13852 all shared libraries that are loaded.
13854 @kindex sharedlibrary
13856 @item sharedlibrary @var{regex}
13857 @itemx share @var{regex}
13858 Load shared object library symbols for files matching a
13859 Unix regular expression.
13860 As with files loaded automatically, it only loads shared libraries
13861 required by your program for a core file or after typing @code{run}. If
13862 @var{regex} is omitted all shared libraries required by your program are
13865 @item nosharedlibrary
13866 @kindex nosharedlibrary
13867 @cindex unload symbols from shared libraries
13868 Unload all shared object library symbols. This discards all symbols
13869 that have been loaded from all shared libraries. Symbols from shared
13870 libraries that were loaded by explicit user requests are not
13874 Sometimes you may wish that @value{GDBN} stops and gives you control
13875 when any of shared library events happen. Use the @code{set
13876 stop-on-solib-events} command for this:
13879 @item set stop-on-solib-events
13880 @kindex set stop-on-solib-events
13881 This command controls whether @value{GDBN} should give you control
13882 when the dynamic linker notifies it about some shared library event.
13883 The most common event of interest is loading or unloading of a new
13886 @item show stop-on-solib-events
13887 @kindex show stop-on-solib-events
13888 Show whether @value{GDBN} stops and gives you control when shared
13889 library events happen.
13892 Shared libraries are also supported in many cross or remote debugging
13893 configurations. @value{GDBN} needs to have access to the target's libraries;
13894 this can be accomplished either by providing copies of the libraries
13895 on the host system, or by asking @value{GDBN} to automatically retrieve the
13896 libraries from the target. If copies of the target libraries are
13897 provided, they need to be the same as the target libraries, although the
13898 copies on the target can be stripped as long as the copies on the host are
13901 @cindex where to look for shared libraries
13902 For remote debugging, you need to tell @value{GDBN} where the target
13903 libraries are, so that it can load the correct copies---otherwise, it
13904 may try to load the host's libraries. @value{GDBN} has two variables
13905 to specify the search directories for target libraries.
13908 @cindex prefix for shared library file names
13909 @cindex system root, alternate
13910 @kindex set solib-absolute-prefix
13911 @kindex set sysroot
13912 @item set sysroot @var{path}
13913 Use @var{path} as the system root for the program being debugged. Any
13914 absolute shared library paths will be prefixed with @var{path}; many
13915 runtime loaders store the absolute paths to the shared library in the
13916 target program's memory. If you use @code{set sysroot} to find shared
13917 libraries, they need to be laid out in the same way that they are on
13918 the target, with e.g.@: a @file{/lib} and @file{/usr/lib} hierarchy
13921 If @var{path} starts with the sequence @file{remote:}, @value{GDBN} will
13922 retrieve the target libraries from the remote system. This is only
13923 supported when using a remote target that supports the @code{remote get}
13924 command (@pxref{File Transfer,,Sending files to a remote system}).
13925 The part of @var{path} following the initial @file{remote:}
13926 (if present) is used as system root prefix on the remote file system.
13927 @footnote{If you want to specify a local system root using a directory
13928 that happens to be named @file{remote:}, you need to use some equivalent
13929 variant of the name like @file{./remote:}.}
13931 The @code{set solib-absolute-prefix} command is an alias for @code{set
13934 @cindex default system root
13935 @cindex @samp{--with-sysroot}
13936 You can set the default system root by using the configure-time
13937 @samp{--with-sysroot} option. If the system root is inside
13938 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
13939 @samp{--exec-prefix}), then the default system root will be updated
13940 automatically if the installed @value{GDBN} is moved to a new
13943 @kindex show sysroot
13945 Display the current shared library prefix.
13947 @kindex set solib-search-path
13948 @item set solib-search-path @var{path}
13949 If this variable is set, @var{path} is a colon-separated list of
13950 directories to search for shared libraries. @samp{solib-search-path}
13951 is used after @samp{sysroot} fails to locate the library, or if the
13952 path to the library is relative instead of absolute. If you want to
13953 use @samp{solib-search-path} instead of @samp{sysroot}, be sure to set
13954 @samp{sysroot} to a nonexistent directory to prevent @value{GDBN} from
13955 finding your host's libraries. @samp{sysroot} is preferred; setting
13956 it to a nonexistent directory may interfere with automatic loading
13957 of shared library symbols.
13959 @kindex show solib-search-path
13960 @item show solib-search-path
13961 Display the current shared library search path.
13965 @node Separate Debug Files
13966 @section Debugging Information in Separate Files
13967 @cindex separate debugging information files
13968 @cindex debugging information in separate files
13969 @cindex @file{.debug} subdirectories
13970 @cindex debugging information directory, global
13971 @cindex global debugging information directory
13972 @cindex build ID, and separate debugging files
13973 @cindex @file{.build-id} directory
13975 @value{GDBN} allows you to put a program's debugging information in a
13976 file separate from the executable itself, in a way that allows
13977 @value{GDBN} to find and load the debugging information automatically.
13978 Since debugging information can be very large---sometimes larger
13979 than the executable code itself---some systems distribute debugging
13980 information for their executables in separate files, which users can
13981 install only when they need to debug a problem.
13983 @value{GDBN} supports two ways of specifying the separate debug info
13988 The executable contains a @dfn{debug link} that specifies the name of
13989 the separate debug info file. The separate debug file's name is
13990 usually @file{@var{executable}.debug}, where @var{executable} is the
13991 name of the corresponding executable file without leading directories
13992 (e.g., @file{ls.debug} for @file{/usr/bin/ls}). In addition, the
13993 debug link specifies a 32-bit @dfn{Cyclic Redundancy Check} (CRC)
13994 checksum for the debug file, which @value{GDBN} uses to validate that
13995 the executable and the debug file came from the same build.
13998 The executable contains a @dfn{build ID}, a unique bit string that is
13999 also present in the corresponding debug info file. (This is supported
14000 only on some operating systems, notably those which use the ELF format
14001 for binary files and the @sc{gnu} Binutils.) For more details about
14002 this feature, see the description of the @option{--build-id}
14003 command-line option in @ref{Options, , Command Line Options, ld.info,
14004 The GNU Linker}. The debug info file's name is not specified
14005 explicitly by the build ID, but can be computed from the build ID, see
14009 Depending on the way the debug info file is specified, @value{GDBN}
14010 uses two different methods of looking for the debug file:
14014 For the ``debug link'' method, @value{GDBN} looks up the named file in
14015 the directory of the executable file, then in a subdirectory of that
14016 directory named @file{.debug}, and finally under the global debug
14017 directory, in a subdirectory whose name is identical to the leading
14018 directories of the executable's absolute file name.
14021 For the ``build ID'' method, @value{GDBN} looks in the
14022 @file{.build-id} subdirectory of the global debug directory for a file
14023 named @file{@var{nn}/@var{nnnnnnnn}.debug}, where @var{nn} are the
14024 first 2 hex characters of the build ID bit string, and @var{nnnnnnnn}
14025 are the rest of the bit string. (Real build ID strings are 32 or more
14026 hex characters, not 10.)
14029 So, for example, suppose you ask @value{GDBN} to debug
14030 @file{/usr/bin/ls}, which has a debug link that specifies the
14031 file @file{ls.debug}, and a build ID whose value in hex is
14032 @code{abcdef1234}. If the global debug directory is
14033 @file{/usr/lib/debug}, then @value{GDBN} will look for the following
14034 debug information files, in the indicated order:
14038 @file{/usr/lib/debug/.build-id/ab/cdef1234.debug}
14040 @file{/usr/bin/ls.debug}
14042 @file{/usr/bin/.debug/ls.debug}
14044 @file{/usr/lib/debug/usr/bin/ls.debug}.
14047 You can set the global debugging info directory's name, and view the
14048 name @value{GDBN} is currently using.
14052 @kindex set debug-file-directory
14053 @item set debug-file-directory @var{directory}
14054 Set the directory which @value{GDBN} searches for separate debugging
14055 information files to @var{directory}.
14057 @kindex show debug-file-directory
14058 @item show debug-file-directory
14059 Show the directory @value{GDBN} searches for separate debugging
14064 @cindex @code{.gnu_debuglink} sections
14065 @cindex debug link sections
14066 A debug link is a special section of the executable file named
14067 @code{.gnu_debuglink}. The section must contain:
14071 A filename, with any leading directory components removed, followed by
14074 zero to three bytes of padding, as needed to reach the next four-byte
14075 boundary within the section, and
14077 a four-byte CRC checksum, stored in the same endianness used for the
14078 executable file itself. The checksum is computed on the debugging
14079 information file's full contents by the function given below, passing
14080 zero as the @var{crc} argument.
14083 Any executable file format can carry a debug link, as long as it can
14084 contain a section named @code{.gnu_debuglink} with the contents
14087 @cindex @code{.note.gnu.build-id} sections
14088 @cindex build ID sections
14089 The build ID is a special section in the executable file (and in other
14090 ELF binary files that @value{GDBN} may consider). This section is
14091 often named @code{.note.gnu.build-id}, but that name is not mandatory.
14092 It contains unique identification for the built files---the ID remains
14093 the same across multiple builds of the same build tree. The default
14094 algorithm SHA1 produces 160 bits (40 hexadecimal characters) of the
14095 content for the build ID string. The same section with an identical
14096 value is present in the original built binary with symbols, in its
14097 stripped variant, and in the separate debugging information file.
14099 The debugging information file itself should be an ordinary
14100 executable, containing a full set of linker symbols, sections, and
14101 debugging information. The sections of the debugging information file
14102 should have the same names, addresses, and sizes as the original file,
14103 but they need not contain any data---much like a @code{.bss} section
14104 in an ordinary executable.
14106 The @sc{gnu} binary utilities (Binutils) package includes the
14107 @samp{objcopy} utility that can produce
14108 the separated executable / debugging information file pairs using the
14109 following commands:
14112 @kbd{objcopy --only-keep-debug foo foo.debug}
14117 These commands remove the debugging
14118 information from the executable file @file{foo} and place it in the file
14119 @file{foo.debug}. You can use the first, second or both methods to link the
14124 The debug link method needs the following additional command to also leave
14125 behind a debug link in @file{foo}:
14128 @kbd{objcopy --add-gnu-debuglink=foo.debug foo}
14131 Ulrich Drepper's @file{elfutils} package, starting with version 0.53, contains
14132 a version of the @code{strip} command such that the command @kbd{strip foo -f
14133 foo.debug} has the same functionality as the two @code{objcopy} commands and
14134 the @code{ln -s} command above, together.
14137 Build ID gets embedded into the main executable using @code{ld --build-id} or
14138 the @value{NGCC} counterpart @code{gcc -Wl,--build-id}. Build ID support plus
14139 compatibility fixes for debug files separation are present in @sc{gnu} binary
14140 utilities (Binutils) package since version 2.18.
14145 @cindex CRC algorithm definition
14146 The CRC used in @code{.gnu_debuglink} is the CRC-32 defined in
14147 IEEE 802.3 using the polynomial:
14149 @c TexInfo requires naked braces for multi-digit exponents for Tex
14150 @c output, but this causes HTML output to barf. HTML has to be set using
14151 @c raw commands. So we end up having to specify this equation in 2
14156 <em>x</em><sup>32</sup> + <em>x</em><sup>26</sup> + <em>x</em><sup>23</sup> + <em>x</em><sup>22</sup> + <em>x</em><sup>16</sup> + <em>x</em><sup>12</sup> + <em>x</em><sup>11</sup>
14157 + <em>x</em><sup>10</sup> + <em>x</em><sup>8</sup> + <em>x</em><sup>7</sup> + <em>x</em><sup>5</sup> + <em>x</em><sup>4</sup> + <em>x</em><sup>2</sup> + <em>x</em> + 1
14163 @math{x^{32} + x^{26} + x^{23} + x^{22} + x^{16} + x^{12} + x^{11}}
14164 @math{+ x^{10} + x^8 + x^7 + x^5 + x^4 + x^2 + x + 1}
14168 The function is computed byte at a time, taking the least
14169 significant bit of each byte first. The initial pattern
14170 @code{0xffffffff} is used, to ensure leading zeros affect the CRC and
14171 the final result is inverted to ensure trailing zeros also affect the
14174 @emph{Note:} This is the same CRC polynomial as used in handling the
14175 @dfn{Remote Serial Protocol} @code{qCRC} packet (@pxref{Remote Protocol,
14176 , @value{GDBN} Remote Serial Protocol}). However in the
14177 case of the Remote Serial Protocol, the CRC is computed @emph{most}
14178 significant bit first, and the result is not inverted, so trailing
14179 zeros have no effect on the CRC value.
14181 To complete the description, we show below the code of the function
14182 which produces the CRC used in @code{.gnu_debuglink}. Inverting the
14183 initially supplied @code{crc} argument means that an initial call to
14184 this function passing in zero will start computing the CRC using
14187 @kindex gnu_debuglink_crc32
14190 gnu_debuglink_crc32 (unsigned long crc,
14191 unsigned char *buf, size_t len)
14193 static const unsigned long crc32_table[256] =
14195 0x00000000, 0x77073096, 0xee0e612c, 0x990951ba, 0x076dc419,
14196 0x706af48f, 0xe963a535, 0x9e6495a3, 0x0edb8832, 0x79dcb8a4,
14197 0xe0d5e91e, 0x97d2d988, 0x09b64c2b, 0x7eb17cbd, 0xe7b82d07,
14198 0x90bf1d91, 0x1db71064, 0x6ab020f2, 0xf3b97148, 0x84be41de,
14199 0x1adad47d, 0x6ddde4eb, 0xf4d4b551, 0x83d385c7, 0x136c9856,
14200 0x646ba8c0, 0xfd62f97a, 0x8a65c9ec, 0x14015c4f, 0x63066cd9,
14201 0xfa0f3d63, 0x8d080df5, 0x3b6e20c8, 0x4c69105e, 0xd56041e4,
14202 0xa2677172, 0x3c03e4d1, 0x4b04d447, 0xd20d85fd, 0xa50ab56b,
14203 0x35b5a8fa, 0x42b2986c, 0xdbbbc9d6, 0xacbcf940, 0x32d86ce3,
14204 0x45df5c75, 0xdcd60dcf, 0xabd13d59, 0x26d930ac, 0x51de003a,
14205 0xc8d75180, 0xbfd06116, 0x21b4f4b5, 0x56b3c423, 0xcfba9599,
14206 0xb8bda50f, 0x2802b89e, 0x5f058808, 0xc60cd9b2, 0xb10be924,
14207 0x2f6f7c87, 0x58684c11, 0xc1611dab, 0xb6662d3d, 0x76dc4190,
14208 0x01db7106, 0x98d220bc, 0xefd5102a, 0x71b18589, 0x06b6b51f,
14209 0x9fbfe4a5, 0xe8b8d433, 0x7807c9a2, 0x0f00f934, 0x9609a88e,
14210 0xe10e9818, 0x7f6a0dbb, 0x086d3d2d, 0x91646c97, 0xe6635c01,
14211 0x6b6b51f4, 0x1c6c6162, 0x856530d8, 0xf262004e, 0x6c0695ed,
14212 0x1b01a57b, 0x8208f4c1, 0xf50fc457, 0x65b0d9c6, 0x12b7e950,
14213 0x8bbeb8ea, 0xfcb9887c, 0x62dd1ddf, 0x15da2d49, 0x8cd37cf3,
14214 0xfbd44c65, 0x4db26158, 0x3ab551ce, 0xa3bc0074, 0xd4bb30e2,
14215 0x4adfa541, 0x3dd895d7, 0xa4d1c46d, 0xd3d6f4fb, 0x4369e96a,
14216 0x346ed9fc, 0xad678846, 0xda60b8d0, 0x44042d73, 0x33031de5,
14217 0xaa0a4c5f, 0xdd0d7cc9, 0x5005713c, 0x270241aa, 0xbe0b1010,
14218 0xc90c2086, 0x5768b525, 0x206f85b3, 0xb966d409, 0xce61e49f,
14219 0x5edef90e, 0x29d9c998, 0xb0d09822, 0xc7d7a8b4, 0x59b33d17,
14220 0x2eb40d81, 0xb7bd5c3b, 0xc0ba6cad, 0xedb88320, 0x9abfb3b6,
14221 0x03b6e20c, 0x74b1d29a, 0xead54739, 0x9dd277af, 0x04db2615,
14222 0x73dc1683, 0xe3630b12, 0x94643b84, 0x0d6d6a3e, 0x7a6a5aa8,
14223 0xe40ecf0b, 0x9309ff9d, 0x0a00ae27, 0x7d079eb1, 0xf00f9344,
14224 0x8708a3d2, 0x1e01f268, 0x6906c2fe, 0xf762575d, 0x806567cb,
14225 0x196c3671, 0x6e6b06e7, 0xfed41b76, 0x89d32be0, 0x10da7a5a,
14226 0x67dd4acc, 0xf9b9df6f, 0x8ebeeff9, 0x17b7be43, 0x60b08ed5,
14227 0xd6d6a3e8, 0xa1d1937e, 0x38d8c2c4, 0x4fdff252, 0xd1bb67f1,
14228 0xa6bc5767, 0x3fb506dd, 0x48b2364b, 0xd80d2bda, 0xaf0a1b4c,
14229 0x36034af6, 0x41047a60, 0xdf60efc3, 0xa867df55, 0x316e8eef,
14230 0x4669be79, 0xcb61b38c, 0xbc66831a, 0x256fd2a0, 0x5268e236,
14231 0xcc0c7795, 0xbb0b4703, 0x220216b9, 0x5505262f, 0xc5ba3bbe,
14232 0xb2bd0b28, 0x2bb45a92, 0x5cb36a04, 0xc2d7ffa7, 0xb5d0cf31,
14233 0x2cd99e8b, 0x5bdeae1d, 0x9b64c2b0, 0xec63f226, 0x756aa39c,
14234 0x026d930a, 0x9c0906a9, 0xeb0e363f, 0x72076785, 0x05005713,
14235 0x95bf4a82, 0xe2b87a14, 0x7bb12bae, 0x0cb61b38, 0x92d28e9b,
14236 0xe5d5be0d, 0x7cdcefb7, 0x0bdbdf21, 0x86d3d2d4, 0xf1d4e242,
14237 0x68ddb3f8, 0x1fda836e, 0x81be16cd, 0xf6b9265b, 0x6fb077e1,
14238 0x18b74777, 0x88085ae6, 0xff0f6a70, 0x66063bca, 0x11010b5c,
14239 0x8f659eff, 0xf862ae69, 0x616bffd3, 0x166ccf45, 0xa00ae278,
14240 0xd70dd2ee, 0x4e048354, 0x3903b3c2, 0xa7672661, 0xd06016f7,
14241 0x4969474d, 0x3e6e77db, 0xaed16a4a, 0xd9d65adc, 0x40df0b66,
14242 0x37d83bf0, 0xa9bcae53, 0xdebb9ec5, 0x47b2cf7f, 0x30b5ffe9,
14243 0xbdbdf21c, 0xcabac28a, 0x53b39330, 0x24b4a3a6, 0xbad03605,
14244 0xcdd70693, 0x54de5729, 0x23d967bf, 0xb3667a2e, 0xc4614ab8,
14245 0x5d681b02, 0x2a6f2b94, 0xb40bbe37, 0xc30c8ea1, 0x5a05df1b,
14248 unsigned char *end;
14250 crc = ~crc & 0xffffffff;
14251 for (end = buf + len; buf < end; ++buf)
14252 crc = crc32_table[(crc ^ *buf) & 0xff] ^ (crc >> 8);
14253 return ~crc & 0xffffffff;
14258 This computation does not apply to the ``build ID'' method.
14261 @node Symbol Errors
14262 @section Errors Reading Symbol Files
14264 While reading a symbol file, @value{GDBN} occasionally encounters problems,
14265 such as symbol types it does not recognize, or known bugs in compiler
14266 output. By default, @value{GDBN} does not notify you of such problems, since
14267 they are relatively common and primarily of interest to people
14268 debugging compilers. If you are interested in seeing information
14269 about ill-constructed symbol tables, you can either ask @value{GDBN} to print
14270 only one message about each such type of problem, no matter how many
14271 times the problem occurs; or you can ask @value{GDBN} to print more messages,
14272 to see how many times the problems occur, with the @code{set
14273 complaints} command (@pxref{Messages/Warnings, ,Optional Warnings and
14276 The messages currently printed, and their meanings, include:
14279 @item inner block not inside outer block in @var{symbol}
14281 The symbol information shows where symbol scopes begin and end
14282 (such as at the start of a function or a block of statements). This
14283 error indicates that an inner scope block is not fully contained
14284 in its outer scope blocks.
14286 @value{GDBN} circumvents the problem by treating the inner block as if it had
14287 the same scope as the outer block. In the error message, @var{symbol}
14288 may be shown as ``@code{(don't know)}'' if the outer block is not a
14291 @item block at @var{address} out of order
14293 The symbol information for symbol scope blocks should occur in
14294 order of increasing addresses. This error indicates that it does not
14297 @value{GDBN} does not circumvent this problem, and has trouble
14298 locating symbols in the source file whose symbols it is reading. (You
14299 can often determine what source file is affected by specifying
14300 @code{set verbose on}. @xref{Messages/Warnings, ,Optional Warnings and
14303 @item bad block start address patched
14305 The symbol information for a symbol scope block has a start address
14306 smaller than the address of the preceding source line. This is known
14307 to occur in the SunOS 4.1.1 (and earlier) C compiler.
14309 @value{GDBN} circumvents the problem by treating the symbol scope block as
14310 starting on the previous source line.
14312 @item bad string table offset in symbol @var{n}
14315 Symbol number @var{n} contains a pointer into the string table which is
14316 larger than the size of the string table.
14318 @value{GDBN} circumvents the problem by considering the symbol to have the
14319 name @code{foo}, which may cause other problems if many symbols end up
14322 @item unknown symbol type @code{0x@var{nn}}
14324 The symbol information contains new data types that @value{GDBN} does
14325 not yet know how to read. @code{0x@var{nn}} is the symbol type of the
14326 uncomprehended information, in hexadecimal.
14328 @value{GDBN} circumvents the error by ignoring this symbol information.
14329 This usually allows you to debug your program, though certain symbols
14330 are not accessible. If you encounter such a problem and feel like
14331 debugging it, you can debug @code{@value{GDBP}} with itself, breakpoint
14332 on @code{complain}, then go up to the function @code{read_dbx_symtab}
14333 and examine @code{*bufp} to see the symbol.
14335 @item stub type has NULL name
14337 @value{GDBN} could not find the full definition for a struct or class.
14339 @item const/volatile indicator missing (ok if using g++ v1.x), got@dots{}
14340 The symbol information for a C@t{++} member function is missing some
14341 information that recent versions of the compiler should have output for
14344 @item info mismatch between compiler and debugger
14346 @value{GDBN} could not parse a type specification output by the compiler.
14351 @section GDB Data Files
14353 @cindex prefix for data files
14354 @value{GDBN} will sometimes read an auxiliary data file. These files
14355 are kept in a directory known as the @dfn{data directory}.
14357 You can set the data directory's name, and view the name @value{GDBN}
14358 is currently using.
14361 @kindex set data-directory
14362 @item set data-directory @var{directory}
14363 Set the directory which @value{GDBN} searches for auxiliary data files
14364 to @var{directory}.
14366 @kindex show data-directory
14367 @item show data-directory
14368 Show the directory @value{GDBN} searches for auxiliary data files.
14371 @cindex default data directory
14372 @cindex @samp{--with-gdb-datadir}
14373 You can set the default data directory by using the configure-time
14374 @samp{--with-gdb-datadir} option. If the data directory is inside
14375 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
14376 @samp{--exec-prefix}), then the default data directory will be updated
14377 automatically if the installed @value{GDBN} is moved to a new
14381 @chapter Specifying a Debugging Target
14383 @cindex debugging target
14384 A @dfn{target} is the execution environment occupied by your program.
14386 Often, @value{GDBN} runs in the same host environment as your program;
14387 in that case, the debugging target is specified as a side effect when
14388 you use the @code{file} or @code{core} commands. When you need more
14389 flexibility---for example, running @value{GDBN} on a physically separate
14390 host, or controlling a standalone system over a serial port or a
14391 realtime system over a TCP/IP connection---you can use the @code{target}
14392 command to specify one of the target types configured for @value{GDBN}
14393 (@pxref{Target Commands, ,Commands for Managing Targets}).
14395 @cindex target architecture
14396 It is possible to build @value{GDBN} for several different @dfn{target
14397 architectures}. When @value{GDBN} is built like that, you can choose
14398 one of the available architectures with the @kbd{set architecture}
14402 @kindex set architecture
14403 @kindex show architecture
14404 @item set architecture @var{arch}
14405 This command sets the current target architecture to @var{arch}. The
14406 value of @var{arch} can be @code{"auto"}, in addition to one of the
14407 supported architectures.
14409 @item show architecture
14410 Show the current target architecture.
14412 @item set processor
14414 @kindex set processor
14415 @kindex show processor
14416 These are alias commands for, respectively, @code{set architecture}
14417 and @code{show architecture}.
14421 * Active Targets:: Active targets
14422 * Target Commands:: Commands for managing targets
14423 * Byte Order:: Choosing target byte order
14426 @node Active Targets
14427 @section Active Targets
14429 @cindex stacking targets
14430 @cindex active targets
14431 @cindex multiple targets
14433 There are three classes of targets: processes, core files, and
14434 executable files. @value{GDBN} can work concurrently on up to three
14435 active targets, one in each class. This allows you to (for example)
14436 start a process and inspect its activity without abandoning your work on
14439 For example, if you execute @samp{gdb a.out}, then the executable file
14440 @code{a.out} is the only active target. If you designate a core file as
14441 well---presumably from a prior run that crashed and coredumped---then
14442 @value{GDBN} has two active targets and uses them in tandem, looking
14443 first in the corefile target, then in the executable file, to satisfy
14444 requests for memory addresses. (Typically, these two classes of target
14445 are complementary, since core files contain only a program's
14446 read-write memory---variables and so on---plus machine status, while
14447 executable files contain only the program text and initialized data.)
14449 When you type @code{run}, your executable file becomes an active process
14450 target as well. When a process target is active, all @value{GDBN}
14451 commands requesting memory addresses refer to that target; addresses in
14452 an active core file or executable file target are obscured while the
14453 process target is active.
14455 Use the @code{core-file} and @code{exec-file} commands to select a new
14456 core file or executable target (@pxref{Files, ,Commands to Specify
14457 Files}). To specify as a target a process that is already running, use
14458 the @code{attach} command (@pxref{Attach, ,Debugging an Already-running
14461 @node Target Commands
14462 @section Commands for Managing Targets
14465 @item target @var{type} @var{parameters}
14466 Connects the @value{GDBN} host environment to a target machine or
14467 process. A target is typically a protocol for talking to debugging
14468 facilities. You use the argument @var{type} to specify the type or
14469 protocol of the target machine.
14471 Further @var{parameters} are interpreted by the target protocol, but
14472 typically include things like device names or host names to connect
14473 with, process numbers, and baud rates.
14475 The @code{target} command does not repeat if you press @key{RET} again
14476 after executing the command.
14478 @kindex help target
14480 Displays the names of all targets available. To display targets
14481 currently selected, use either @code{info target} or @code{info files}
14482 (@pxref{Files, ,Commands to Specify Files}).
14484 @item help target @var{name}
14485 Describe a particular target, including any parameters necessary to
14488 @kindex set gnutarget
14489 @item set gnutarget @var{args}
14490 @value{GDBN} uses its own library BFD to read your files. @value{GDBN}
14491 knows whether it is reading an @dfn{executable},
14492 a @dfn{core}, or a @dfn{.o} file; however, you can specify the file format
14493 with the @code{set gnutarget} command. Unlike most @code{target} commands,
14494 with @code{gnutarget} the @code{target} refers to a program, not a machine.
14497 @emph{Warning:} To specify a file format with @code{set gnutarget},
14498 you must know the actual BFD name.
14502 @xref{Files, , Commands to Specify Files}.
14504 @kindex show gnutarget
14505 @item show gnutarget
14506 Use the @code{show gnutarget} command to display what file format
14507 @code{gnutarget} is set to read. If you have not set @code{gnutarget},
14508 @value{GDBN} will determine the file format for each file automatically,
14509 and @code{show gnutarget} displays @samp{The current BDF target is "auto"}.
14512 @cindex common targets
14513 Here are some common targets (available, or not, depending on the GDB
14518 @item target exec @var{program}
14519 @cindex executable file target
14520 An executable file. @samp{target exec @var{program}} is the same as
14521 @samp{exec-file @var{program}}.
14523 @item target core @var{filename}
14524 @cindex core dump file target
14525 A core dump file. @samp{target core @var{filename}} is the same as
14526 @samp{core-file @var{filename}}.
14528 @item target remote @var{medium}
14529 @cindex remote target
14530 A remote system connected to @value{GDBN} via a serial line or network
14531 connection. This command tells @value{GDBN} to use its own remote
14532 protocol over @var{medium} for debugging. @xref{Remote Debugging}.
14534 For example, if you have a board connected to @file{/dev/ttya} on the
14535 machine running @value{GDBN}, you could say:
14538 target remote /dev/ttya
14541 @code{target remote} supports the @code{load} command. This is only
14542 useful if you have some other way of getting the stub to the target
14543 system, and you can put it somewhere in memory where it won't get
14544 clobbered by the download.
14547 @cindex built-in simulator target
14548 Builtin CPU simulator. @value{GDBN} includes simulators for most architectures.
14556 works; however, you cannot assume that a specific memory map, device
14557 drivers, or even basic I/O is available, although some simulators do
14558 provide these. For info about any processor-specific simulator details,
14559 see the appropriate section in @ref{Embedded Processors, ,Embedded
14564 Some configurations may include these targets as well:
14568 @item target nrom @var{dev}
14569 @cindex NetROM ROM emulator target
14570 NetROM ROM emulator. This target only supports downloading.
14574 Different targets are available on different configurations of @value{GDBN};
14575 your configuration may have more or fewer targets.
14577 Many remote targets require you to download the executable's code once
14578 you've successfully established a connection. You may wish to control
14579 various aspects of this process.
14584 @kindex set hash@r{, for remote monitors}
14585 @cindex hash mark while downloading
14586 This command controls whether a hash mark @samp{#} is displayed while
14587 downloading a file to the remote monitor. If on, a hash mark is
14588 displayed after each S-record is successfully downloaded to the
14592 @kindex show hash@r{, for remote monitors}
14593 Show the current status of displaying the hash mark.
14595 @item set debug monitor
14596 @kindex set debug monitor
14597 @cindex display remote monitor communications
14598 Enable or disable display of communications messages between
14599 @value{GDBN} and the remote monitor.
14601 @item show debug monitor
14602 @kindex show debug monitor
14603 Show the current status of displaying communications between
14604 @value{GDBN} and the remote monitor.
14609 @kindex load @var{filename}
14610 @item load @var{filename}
14612 Depending on what remote debugging facilities are configured into
14613 @value{GDBN}, the @code{load} command may be available. Where it exists, it
14614 is meant to make @var{filename} (an executable) available for debugging
14615 on the remote system---by downloading, or dynamic linking, for example.
14616 @code{load} also records the @var{filename} symbol table in @value{GDBN}, like
14617 the @code{add-symbol-file} command.
14619 If your @value{GDBN} does not have a @code{load} command, attempting to
14620 execute it gets the error message ``@code{You can't do that when your
14621 target is @dots{}}''
14623 The file is loaded at whatever address is specified in the executable.
14624 For some object file formats, you can specify the load address when you
14625 link the program; for other formats, like a.out, the object file format
14626 specifies a fixed address.
14627 @c FIXME! This would be a good place for an xref to the GNU linker doc.
14629 Depending on the remote side capabilities, @value{GDBN} may be able to
14630 load programs into flash memory.
14632 @code{load} does not repeat if you press @key{RET} again after using it.
14636 @section Choosing Target Byte Order
14638 @cindex choosing target byte order
14639 @cindex target byte order
14641 Some types of processors, such as the MIPS, PowerPC, and Renesas SH,
14642 offer the ability to run either big-endian or little-endian byte
14643 orders. Usually the executable or symbol will include a bit to
14644 designate the endian-ness, and you will not need to worry about
14645 which to use. However, you may still find it useful to adjust
14646 @value{GDBN}'s idea of processor endian-ness manually.
14650 @item set endian big
14651 Instruct @value{GDBN} to assume the target is big-endian.
14653 @item set endian little
14654 Instruct @value{GDBN} to assume the target is little-endian.
14656 @item set endian auto
14657 Instruct @value{GDBN} to use the byte order associated with the
14661 Display @value{GDBN}'s current idea of the target byte order.
14665 Note that these commands merely adjust interpretation of symbolic
14666 data on the host, and that they have absolutely no effect on the
14670 @node Remote Debugging
14671 @chapter Debugging Remote Programs
14672 @cindex remote debugging
14674 If you are trying to debug a program running on a machine that cannot run
14675 @value{GDBN} in the usual way, it is often useful to use remote debugging.
14676 For example, you might use remote debugging on an operating system kernel,
14677 or on a small system which does not have a general purpose operating system
14678 powerful enough to run a full-featured debugger.
14680 Some configurations of @value{GDBN} have special serial or TCP/IP interfaces
14681 to make this work with particular debugging targets. In addition,
14682 @value{GDBN} comes with a generic serial protocol (specific to @value{GDBN},
14683 but not specific to any particular target system) which you can use if you
14684 write the remote stubs---the code that runs on the remote system to
14685 communicate with @value{GDBN}.
14687 Other remote targets may be available in your
14688 configuration of @value{GDBN}; use @code{help target} to list them.
14691 * Connecting:: Connecting to a remote target
14692 * File Transfer:: Sending files to a remote system
14693 * Server:: Using the gdbserver program
14694 * Remote Configuration:: Remote configuration
14695 * Remote Stub:: Implementing a remote stub
14699 @section Connecting to a Remote Target
14701 On the @value{GDBN} host machine, you will need an unstripped copy of
14702 your program, since @value{GDBN} needs symbol and debugging information.
14703 Start up @value{GDBN} as usual, using the name of the local copy of your
14704 program as the first argument.
14706 @cindex @code{target remote}
14707 @value{GDBN} can communicate with the target over a serial line, or
14708 over an @acronym{IP} network using @acronym{TCP} or @acronym{UDP}. In
14709 each case, @value{GDBN} uses the same protocol for debugging your
14710 program; only the medium carrying the debugging packets varies. The
14711 @code{target remote} command establishes a connection to the target.
14712 Its arguments indicate which medium to use:
14716 @item target remote @var{serial-device}
14717 @cindex serial line, @code{target remote}
14718 Use @var{serial-device} to communicate with the target. For example,
14719 to use a serial line connected to the device named @file{/dev/ttyb}:
14722 target remote /dev/ttyb
14725 If you're using a serial line, you may want to give @value{GDBN} the
14726 @w{@samp{--baud}} option, or use the @code{set remotebaud} command
14727 (@pxref{Remote Configuration, set remotebaud}) before the
14728 @code{target} command.
14730 @item target remote @code{@var{host}:@var{port}}
14731 @itemx target remote @code{tcp:@var{host}:@var{port}}
14732 @cindex @acronym{TCP} port, @code{target remote}
14733 Debug using a @acronym{TCP} connection to @var{port} on @var{host}.
14734 The @var{host} may be either a host name or a numeric @acronym{IP}
14735 address; @var{port} must be a decimal number. The @var{host} could be
14736 the target machine itself, if it is directly connected to the net, or
14737 it might be a terminal server which in turn has a serial line to the
14740 For example, to connect to port 2828 on a terminal server named
14744 target remote manyfarms:2828
14747 If your remote target is actually running on the same machine as your
14748 debugger session (e.g.@: a simulator for your target running on the
14749 same host), you can omit the hostname. For example, to connect to
14750 port 1234 on your local machine:
14753 target remote :1234
14757 Note that the colon is still required here.
14759 @item target remote @code{udp:@var{host}:@var{port}}
14760 @cindex @acronym{UDP} port, @code{target remote}
14761 Debug using @acronym{UDP} packets to @var{port} on @var{host}. For example, to
14762 connect to @acronym{UDP} port 2828 on a terminal server named @code{manyfarms}:
14765 target remote udp:manyfarms:2828
14768 When using a @acronym{UDP} connection for remote debugging, you should
14769 keep in mind that the `U' stands for ``Unreliable''. @acronym{UDP}
14770 can silently drop packets on busy or unreliable networks, which will
14771 cause havoc with your debugging session.
14773 @item target remote | @var{command}
14774 @cindex pipe, @code{target remote} to
14775 Run @var{command} in the background and communicate with it using a
14776 pipe. The @var{command} is a shell command, to be parsed and expanded
14777 by the system's command shell, @code{/bin/sh}; it should expect remote
14778 protocol packets on its standard input, and send replies on its
14779 standard output. You could use this to run a stand-alone simulator
14780 that speaks the remote debugging protocol, to make net connections
14781 using programs like @code{ssh}, or for other similar tricks.
14783 If @var{command} closes its standard output (perhaps by exiting),
14784 @value{GDBN} will try to send it a @code{SIGTERM} signal. (If the
14785 program has already exited, this will have no effect.)
14789 Once the connection has been established, you can use all the usual
14790 commands to examine and change data. The remote program is already
14791 running; you can use @kbd{step} and @kbd{continue}, and you do not
14792 need to use @kbd{run}.
14794 @cindex interrupting remote programs
14795 @cindex remote programs, interrupting
14796 Whenever @value{GDBN} is waiting for the remote program, if you type the
14797 interrupt character (often @kbd{Ctrl-c}), @value{GDBN} attempts to stop the
14798 program. This may or may not succeed, depending in part on the hardware
14799 and the serial drivers the remote system uses. If you type the
14800 interrupt character once again, @value{GDBN} displays this prompt:
14803 Interrupted while waiting for the program.
14804 Give up (and stop debugging it)? (y or n)
14807 If you type @kbd{y}, @value{GDBN} abandons the remote debugging session.
14808 (If you decide you want to try again later, you can use @samp{target
14809 remote} again to connect once more.) If you type @kbd{n}, @value{GDBN}
14810 goes back to waiting.
14813 @kindex detach (remote)
14815 When you have finished debugging the remote program, you can use the
14816 @code{detach} command to release it from @value{GDBN} control.
14817 Detaching from the target normally resumes its execution, but the results
14818 will depend on your particular remote stub. After the @code{detach}
14819 command, @value{GDBN} is free to connect to another target.
14823 The @code{disconnect} command behaves like @code{detach}, except that
14824 the target is generally not resumed. It will wait for @value{GDBN}
14825 (this instance or another one) to connect and continue debugging. After
14826 the @code{disconnect} command, @value{GDBN} is again free to connect to
14829 @cindex send command to remote monitor
14830 @cindex extend @value{GDBN} for remote targets
14831 @cindex add new commands for external monitor
14833 @item monitor @var{cmd}
14834 This command allows you to send arbitrary commands directly to the
14835 remote monitor. Since @value{GDBN} doesn't care about the commands it
14836 sends like this, this command is the way to extend @value{GDBN}---you
14837 can add new commands that only the external monitor will understand
14841 @node File Transfer
14842 @section Sending files to a remote system
14843 @cindex remote target, file transfer
14844 @cindex file transfer
14845 @cindex sending files to remote systems
14847 Some remote targets offer the ability to transfer files over the same
14848 connection used to communicate with @value{GDBN}. This is convenient
14849 for targets accessible through other means, e.g.@: @sc{gnu}/Linux systems
14850 running @code{gdbserver} over a network interface. For other targets,
14851 e.g.@: embedded devices with only a single serial port, this may be
14852 the only way to upload or download files.
14854 Not all remote targets support these commands.
14858 @item remote put @var{hostfile} @var{targetfile}
14859 Copy file @var{hostfile} from the host system (the machine running
14860 @value{GDBN}) to @var{targetfile} on the target system.
14863 @item remote get @var{targetfile} @var{hostfile}
14864 Copy file @var{targetfile} from the target system to @var{hostfile}
14865 on the host system.
14867 @kindex remote delete
14868 @item remote delete @var{targetfile}
14869 Delete @var{targetfile} from the target system.
14874 @section Using the @code{gdbserver} Program
14877 @cindex remote connection without stubs
14878 @code{gdbserver} is a control program for Unix-like systems, which
14879 allows you to connect your program with a remote @value{GDBN} via
14880 @code{target remote}---but without linking in the usual debugging stub.
14882 @code{gdbserver} is not a complete replacement for the debugging stubs,
14883 because it requires essentially the same operating-system facilities
14884 that @value{GDBN} itself does. In fact, a system that can run
14885 @code{gdbserver} to connect to a remote @value{GDBN} could also run
14886 @value{GDBN} locally! @code{gdbserver} is sometimes useful nevertheless,
14887 because it is a much smaller program than @value{GDBN} itself. It is
14888 also easier to port than all of @value{GDBN}, so you may be able to get
14889 started more quickly on a new system by using @code{gdbserver}.
14890 Finally, if you develop code for real-time systems, you may find that
14891 the tradeoffs involved in real-time operation make it more convenient to
14892 do as much development work as possible on another system, for example
14893 by cross-compiling. You can use @code{gdbserver} to make a similar
14894 choice for debugging.
14896 @value{GDBN} and @code{gdbserver} communicate via either a serial line
14897 or a TCP connection, using the standard @value{GDBN} remote serial
14901 @emph{Warning:} @code{gdbserver} does not have any built-in security.
14902 Do not run @code{gdbserver} connected to any public network; a
14903 @value{GDBN} connection to @code{gdbserver} provides access to the
14904 target system with the same privileges as the user running
14908 @subsection Running @code{gdbserver}
14909 @cindex arguments, to @code{gdbserver}
14911 Run @code{gdbserver} on the target system. You need a copy of the
14912 program you want to debug, including any libraries it requires.
14913 @code{gdbserver} does not need your program's symbol table, so you can
14914 strip the program if necessary to save space. @value{GDBN} on the host
14915 system does all the symbol handling.
14917 To use the server, you must tell it how to communicate with @value{GDBN};
14918 the name of your program; and the arguments for your program. The usual
14922 target> gdbserver @var{comm} @var{program} [ @var{args} @dots{} ]
14925 @var{comm} is either a device name (to use a serial line) or a TCP
14926 hostname and portnumber. For example, to debug Emacs with the argument
14927 @samp{foo.txt} and communicate with @value{GDBN} over the serial port
14931 target> gdbserver /dev/com1 emacs foo.txt
14934 @code{gdbserver} waits passively for the host @value{GDBN} to communicate
14937 To use a TCP connection instead of a serial line:
14940 target> gdbserver host:2345 emacs foo.txt
14943 The only difference from the previous example is the first argument,
14944 specifying that you are communicating with the host @value{GDBN} via
14945 TCP. The @samp{host:2345} argument means that @code{gdbserver} is to
14946 expect a TCP connection from machine @samp{host} to local TCP port 2345.
14947 (Currently, the @samp{host} part is ignored.) You can choose any number
14948 you want for the port number as long as it does not conflict with any
14949 TCP ports already in use on the target system (for example, @code{23} is
14950 reserved for @code{telnet}).@footnote{If you choose a port number that
14951 conflicts with another service, @code{gdbserver} prints an error message
14952 and exits.} You must use the same port number with the host @value{GDBN}
14953 @code{target remote} command.
14955 @subsubsection Attaching to a Running Program
14957 On some targets, @code{gdbserver} can also attach to running programs.
14958 This is accomplished via the @code{--attach} argument. The syntax is:
14961 target> gdbserver --attach @var{comm} @var{pid}
14964 @var{pid} is the process ID of a currently running process. It isn't necessary
14965 to point @code{gdbserver} at a binary for the running process.
14968 @cindex attach to a program by name
14969 You can debug processes by name instead of process ID if your target has the
14970 @code{pidof} utility:
14973 target> gdbserver --attach @var{comm} `pidof @var{program}`
14976 In case more than one copy of @var{program} is running, or @var{program}
14977 has multiple threads, most versions of @code{pidof} support the
14978 @code{-s} option to only return the first process ID.
14980 @subsubsection Multi-Process Mode for @code{gdbserver}
14981 @cindex gdbserver, multiple processes
14982 @cindex multiple processes with gdbserver
14984 When you connect to @code{gdbserver} using @code{target remote},
14985 @code{gdbserver} debugs the specified program only once. When the
14986 program exits, or you detach from it, @value{GDBN} closes the connection
14987 and @code{gdbserver} exits.
14989 If you connect using @kbd{target extended-remote}, @code{gdbserver}
14990 enters multi-process mode. When the debugged program exits, or you
14991 detach from it, @value{GDBN} stays connected to @code{gdbserver} even
14992 though no program is running. The @code{run} and @code{attach}
14993 commands instruct @code{gdbserver} to run or attach to a new program.
14994 The @code{run} command uses @code{set remote exec-file} (@pxref{set
14995 remote exec-file}) to select the program to run. Command line
14996 arguments are supported, except for wildcard expansion and I/O
14997 redirection (@pxref{Arguments}).
14999 To start @code{gdbserver} without supplying an initial command to run
15000 or process ID to attach, use the @option{--multi} command line option.
15001 Then you can connect using @kbd{target extended-remote} and start
15002 the program you want to debug.
15004 @code{gdbserver} does not automatically exit in multi-process mode.
15005 You can terminate it by using @code{monitor exit}
15006 (@pxref{Monitor Commands for gdbserver}).
15008 @subsubsection Other Command-Line Arguments for @code{gdbserver}
15010 The @option{--debug} option tells @code{gdbserver} to display extra
15011 status information about the debugging process. The
15012 @option{--remote-debug} option tells @code{gdbserver} to display
15013 remote protocol debug output. These options are intended for
15014 @code{gdbserver} development and for bug reports to the developers.
15016 The @option{--wrapper} option specifies a wrapper to launch programs
15017 for debugging. The option should be followed by the name of the
15018 wrapper, then any command-line arguments to pass to the wrapper, then
15019 @kbd{--} indicating the end of the wrapper arguments.
15021 @code{gdbserver} runs the specified wrapper program with a combined
15022 command line including the wrapper arguments, then the name of the
15023 program to debug, then any arguments to the program. The wrapper
15024 runs until it executes your program, and then @value{GDBN} gains control.
15026 You can use any program that eventually calls @code{execve} with
15027 its arguments as a wrapper. Several standard Unix utilities do
15028 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
15029 with @code{exec "$@@"} will also work.
15031 For example, you can use @code{env} to pass an environment variable to
15032 the debugged program, without setting the variable in @code{gdbserver}'s
15036 $ gdbserver --wrapper env LD_PRELOAD=libtest.so -- :2222 ./testprog
15039 @subsection Connecting to @code{gdbserver}
15041 Run @value{GDBN} on the host system.
15043 First make sure you have the necessary symbol files. Load symbols for
15044 your application using the @code{file} command before you connect. Use
15045 @code{set sysroot} to locate target libraries (unless your @value{GDBN}
15046 was compiled with the correct sysroot using @code{--with-sysroot}).
15048 The symbol file and target libraries must exactly match the executable
15049 and libraries on the target, with one exception: the files on the host
15050 system should not be stripped, even if the files on the target system
15051 are. Mismatched or missing files will lead to confusing results
15052 during debugging. On @sc{gnu}/Linux targets, mismatched or missing
15053 files may also prevent @code{gdbserver} from debugging multi-threaded
15056 Connect to your target (@pxref{Connecting,,Connecting to a Remote Target}).
15057 For TCP connections, you must start up @code{gdbserver} prior to using
15058 the @code{target remote} command. Otherwise you may get an error whose
15059 text depends on the host system, but which usually looks something like
15060 @samp{Connection refused}. Don't use the @code{load}
15061 command in @value{GDBN} when using @code{gdbserver}, since the program is
15062 already on the target.
15064 @subsection Monitor Commands for @code{gdbserver}
15065 @cindex monitor commands, for @code{gdbserver}
15066 @anchor{Monitor Commands for gdbserver}
15068 During a @value{GDBN} session using @code{gdbserver}, you can use the
15069 @code{monitor} command to send special requests to @code{gdbserver}.
15070 Here are the available commands.
15074 List the available monitor commands.
15076 @item monitor set debug 0
15077 @itemx monitor set debug 1
15078 Disable or enable general debugging messages.
15080 @item monitor set remote-debug 0
15081 @itemx monitor set remote-debug 1
15082 Disable or enable specific debugging messages associated with the remote
15083 protocol (@pxref{Remote Protocol}).
15085 @item monitor set libthread-db-search-path [PATH]
15086 @cindex gdbserver, search path for @code{libthread_db}
15087 When this command is issued, @var{path} is a colon-separated list of
15088 directories to search for @code{libthread_db} (@pxref{Threads,,set
15089 libthread-db-search-path}). If you omit @var{path},
15090 @samp{libthread-db-search-path} will be reset to an empty list.
15093 Tell gdbserver to exit immediately. This command should be followed by
15094 @code{disconnect} to close the debugging session. @code{gdbserver} will
15095 detach from any attached processes and kill any processes it created.
15096 Use @code{monitor exit} to terminate @code{gdbserver} at the end
15097 of a multi-process mode debug session.
15101 @node Remote Configuration
15102 @section Remote Configuration
15105 @kindex show remote
15106 This section documents the configuration options available when
15107 debugging remote programs. For the options related to the File I/O
15108 extensions of the remote protocol, see @ref{system,
15109 system-call-allowed}.
15112 @item set remoteaddresssize @var{bits}
15113 @cindex address size for remote targets
15114 @cindex bits in remote address
15115 Set the maximum size of address in a memory packet to the specified
15116 number of bits. @value{GDBN} will mask off the address bits above
15117 that number, when it passes addresses to the remote target. The
15118 default value is the number of bits in the target's address.
15120 @item show remoteaddresssize
15121 Show the current value of remote address size in bits.
15123 @item set remotebaud @var{n}
15124 @cindex baud rate for remote targets
15125 Set the baud rate for the remote serial I/O to @var{n} baud. The
15126 value is used to set the speed of the serial port used for debugging
15129 @item show remotebaud
15130 Show the current speed of the remote connection.
15132 @item set remotebreak
15133 @cindex interrupt remote programs
15134 @cindex BREAK signal instead of Ctrl-C
15135 @anchor{set remotebreak}
15136 If set to on, @value{GDBN} sends a @code{BREAK} signal to the remote
15137 when you type @kbd{Ctrl-c} to interrupt the program running
15138 on the remote. If set to off, @value{GDBN} sends the @samp{Ctrl-C}
15139 character instead. The default is off, since most remote systems
15140 expect to see @samp{Ctrl-C} as the interrupt signal.
15142 @item show remotebreak
15143 Show whether @value{GDBN} sends @code{BREAK} or @samp{Ctrl-C} to
15144 interrupt the remote program.
15146 @item set remoteflow on
15147 @itemx set remoteflow off
15148 @kindex set remoteflow
15149 Enable or disable hardware flow control (@code{RTS}/@code{CTS})
15150 on the serial port used to communicate to the remote target.
15152 @item show remoteflow
15153 @kindex show remoteflow
15154 Show the current setting of hardware flow control.
15156 @item set remotelogbase @var{base}
15157 Set the base (a.k.a.@: radix) of logging serial protocol
15158 communications to @var{base}. Supported values of @var{base} are:
15159 @code{ascii}, @code{octal}, and @code{hex}. The default is
15162 @item show remotelogbase
15163 Show the current setting of the radix for logging remote serial
15166 @item set remotelogfile @var{file}
15167 @cindex record serial communications on file
15168 Record remote serial communications on the named @var{file}. The
15169 default is not to record at all.
15171 @item show remotelogfile.
15172 Show the current setting of the file name on which to record the
15173 serial communications.
15175 @item set remotetimeout @var{num}
15176 @cindex timeout for serial communications
15177 @cindex remote timeout
15178 Set the timeout limit to wait for the remote target to respond to
15179 @var{num} seconds. The default is 2 seconds.
15181 @item show remotetimeout
15182 Show the current number of seconds to wait for the remote target
15185 @cindex limit hardware breakpoints and watchpoints
15186 @cindex remote target, limit break- and watchpoints
15187 @anchor{set remote hardware-watchpoint-limit}
15188 @anchor{set remote hardware-breakpoint-limit}
15189 @item set remote hardware-watchpoint-limit @var{limit}
15190 @itemx set remote hardware-breakpoint-limit @var{limit}
15191 Restrict @value{GDBN} to using @var{limit} remote hardware breakpoint or
15192 watchpoints. A limit of -1, the default, is treated as unlimited.
15194 @item set remote exec-file @var{filename}
15195 @itemx show remote exec-file
15196 @anchor{set remote exec-file}
15197 @cindex executable file, for remote target
15198 Select the file used for @code{run} with @code{target
15199 extended-remote}. This should be set to a filename valid on the
15200 target system. If it is not set, the target will use a default
15201 filename (e.g.@: the last program run).
15205 @item set tcp auto-retry on
15206 @cindex auto-retry, for remote TCP target
15207 Enable auto-retry for remote TCP connections. This is useful if the remote
15208 debugging agent is launched in parallel with @value{GDBN}; there is a race
15209 condition because the agent may not become ready to accept the connection
15210 before @value{GDBN} attempts to connect. When auto-retry is
15211 enabled, if the initial attempt to connect fails, @value{GDBN} reattempts
15212 to establish the connection using the timeout specified by
15213 @code{set tcp connect-timeout}.
15215 @item set tcp auto-retry off
15216 Do not auto-retry failed TCP connections.
15218 @item show tcp auto-retry
15219 Show the current auto-retry setting.
15221 @item set tcp connect-timeout @var{seconds}
15222 @cindex connection timeout, for remote TCP target
15223 @cindex timeout, for remote target connection
15224 Set the timeout for establishing a TCP connection to the remote target to
15225 @var{seconds}. The timeout affects both polling to retry failed connections
15226 (enabled by @code{set tcp auto-retry on}) and waiting for connections
15227 that are merely slow to complete, and represents an approximate cumulative
15230 @item show tcp connect-timeout
15231 Show the current connection timeout setting.
15234 @cindex remote packets, enabling and disabling
15235 The @value{GDBN} remote protocol autodetects the packets supported by
15236 your debugging stub. If you need to override the autodetection, you
15237 can use these commands to enable or disable individual packets. Each
15238 packet can be set to @samp{on} (the remote target supports this
15239 packet), @samp{off} (the remote target does not support this packet),
15240 or @samp{auto} (detect remote target support for this packet). They
15241 all default to @samp{auto}. For more information about each packet,
15242 see @ref{Remote Protocol}.
15244 During normal use, you should not have to use any of these commands.
15245 If you do, that may be a bug in your remote debugging stub, or a bug
15246 in @value{GDBN}. You may want to report the problem to the
15247 @value{GDBN} developers.
15249 For each packet @var{name}, the command to enable or disable the
15250 packet is @code{set remote @var{name}-packet}. The available settings
15253 @multitable @columnfractions 0.28 0.32 0.25
15256 @tab Related Features
15258 @item @code{fetch-register}
15260 @tab @code{info registers}
15262 @item @code{set-register}
15266 @item @code{binary-download}
15268 @tab @code{load}, @code{set}
15270 @item @code{read-aux-vector}
15271 @tab @code{qXfer:auxv:read}
15272 @tab @code{info auxv}
15274 @item @code{symbol-lookup}
15275 @tab @code{qSymbol}
15276 @tab Detecting multiple threads
15278 @item @code{attach}
15279 @tab @code{vAttach}
15282 @item @code{verbose-resume}
15284 @tab Stepping or resuming multiple threads
15290 @item @code{software-breakpoint}
15294 @item @code{hardware-breakpoint}
15298 @item @code{write-watchpoint}
15302 @item @code{read-watchpoint}
15306 @item @code{access-watchpoint}
15310 @item @code{target-features}
15311 @tab @code{qXfer:features:read}
15312 @tab @code{set architecture}
15314 @item @code{library-info}
15315 @tab @code{qXfer:libraries:read}
15316 @tab @code{info sharedlibrary}
15318 @item @code{memory-map}
15319 @tab @code{qXfer:memory-map:read}
15320 @tab @code{info mem}
15322 @item @code{read-spu-object}
15323 @tab @code{qXfer:spu:read}
15324 @tab @code{info spu}
15326 @item @code{write-spu-object}
15327 @tab @code{qXfer:spu:write}
15328 @tab @code{info spu}
15330 @item @code{read-siginfo-object}
15331 @tab @code{qXfer:siginfo:read}
15332 @tab @code{print $_siginfo}
15334 @item @code{write-siginfo-object}
15335 @tab @code{qXfer:siginfo:write}
15336 @tab @code{set $_siginfo}
15338 @item @code{get-thread-local-@*storage-address}
15339 @tab @code{qGetTLSAddr}
15340 @tab Displaying @code{__thread} variables
15342 @item @code{search-memory}
15343 @tab @code{qSearch:memory}
15346 @item @code{supported-packets}
15347 @tab @code{qSupported}
15348 @tab Remote communications parameters
15350 @item @code{pass-signals}
15351 @tab @code{QPassSignals}
15352 @tab @code{handle @var{signal}}
15354 @item @code{hostio-close-packet}
15355 @tab @code{vFile:close}
15356 @tab @code{remote get}, @code{remote put}
15358 @item @code{hostio-open-packet}
15359 @tab @code{vFile:open}
15360 @tab @code{remote get}, @code{remote put}
15362 @item @code{hostio-pread-packet}
15363 @tab @code{vFile:pread}
15364 @tab @code{remote get}, @code{remote put}
15366 @item @code{hostio-pwrite-packet}
15367 @tab @code{vFile:pwrite}
15368 @tab @code{remote get}, @code{remote put}
15370 @item @code{hostio-unlink-packet}
15371 @tab @code{vFile:unlink}
15372 @tab @code{remote delete}
15374 @item @code{noack-packet}
15375 @tab @code{QStartNoAckMode}
15376 @tab Packet acknowledgment
15378 @item @code{osdata}
15379 @tab @code{qXfer:osdata:read}
15380 @tab @code{info os}
15382 @item @code{query-attached}
15383 @tab @code{qAttached}
15384 @tab Querying remote process attach state.
15388 @section Implementing a Remote Stub
15390 @cindex debugging stub, example
15391 @cindex remote stub, example
15392 @cindex stub example, remote debugging
15393 The stub files provided with @value{GDBN} implement the target side of the
15394 communication protocol, and the @value{GDBN} side is implemented in the
15395 @value{GDBN} source file @file{remote.c}. Normally, you can simply allow
15396 these subroutines to communicate, and ignore the details. (If you're
15397 implementing your own stub file, you can still ignore the details: start
15398 with one of the existing stub files. @file{sparc-stub.c} is the best
15399 organized, and therefore the easiest to read.)
15401 @cindex remote serial debugging, overview
15402 To debug a program running on another machine (the debugging
15403 @dfn{target} machine), you must first arrange for all the usual
15404 prerequisites for the program to run by itself. For example, for a C
15409 A startup routine to set up the C runtime environment; these usually
15410 have a name like @file{crt0}. The startup routine may be supplied by
15411 your hardware supplier, or you may have to write your own.
15414 A C subroutine library to support your program's
15415 subroutine calls, notably managing input and output.
15418 A way of getting your program to the other machine---for example, a
15419 download program. These are often supplied by the hardware
15420 manufacturer, but you may have to write your own from hardware
15424 The next step is to arrange for your program to use a serial port to
15425 communicate with the machine where @value{GDBN} is running (the @dfn{host}
15426 machine). In general terms, the scheme looks like this:
15430 @value{GDBN} already understands how to use this protocol; when everything
15431 else is set up, you can simply use the @samp{target remote} command
15432 (@pxref{Targets,,Specifying a Debugging Target}).
15434 @item On the target,
15435 you must link with your program a few special-purpose subroutines that
15436 implement the @value{GDBN} remote serial protocol. The file containing these
15437 subroutines is called a @dfn{debugging stub}.
15439 On certain remote targets, you can use an auxiliary program
15440 @code{gdbserver} instead of linking a stub into your program.
15441 @xref{Server,,Using the @code{gdbserver} Program}, for details.
15444 The debugging stub is specific to the architecture of the remote
15445 machine; for example, use @file{sparc-stub.c} to debug programs on
15448 @cindex remote serial stub list
15449 These working remote stubs are distributed with @value{GDBN}:
15454 @cindex @file{i386-stub.c}
15457 For Intel 386 and compatible architectures.
15460 @cindex @file{m68k-stub.c}
15461 @cindex Motorola 680x0
15463 For Motorola 680x0 architectures.
15466 @cindex @file{sh-stub.c}
15469 For Renesas SH architectures.
15472 @cindex @file{sparc-stub.c}
15474 For @sc{sparc} architectures.
15476 @item sparcl-stub.c
15477 @cindex @file{sparcl-stub.c}
15480 For Fujitsu @sc{sparclite} architectures.
15484 The @file{README} file in the @value{GDBN} distribution may list other
15485 recently added stubs.
15488 * Stub Contents:: What the stub can do for you
15489 * Bootstrapping:: What you must do for the stub
15490 * Debug Session:: Putting it all together
15493 @node Stub Contents
15494 @subsection What the Stub Can Do for You
15496 @cindex remote serial stub
15497 The debugging stub for your architecture supplies these three
15501 @item set_debug_traps
15502 @findex set_debug_traps
15503 @cindex remote serial stub, initialization
15504 This routine arranges for @code{handle_exception} to run when your
15505 program stops. You must call this subroutine explicitly near the
15506 beginning of your program.
15508 @item handle_exception
15509 @findex handle_exception
15510 @cindex remote serial stub, main routine
15511 This is the central workhorse, but your program never calls it
15512 explicitly---the setup code arranges for @code{handle_exception} to
15513 run when a trap is triggered.
15515 @code{handle_exception} takes control when your program stops during
15516 execution (for example, on a breakpoint), and mediates communications
15517 with @value{GDBN} on the host machine. This is where the communications
15518 protocol is implemented; @code{handle_exception} acts as the @value{GDBN}
15519 representative on the target machine. It begins by sending summary
15520 information on the state of your program, then continues to execute,
15521 retrieving and transmitting any information @value{GDBN} needs, until you
15522 execute a @value{GDBN} command that makes your program resume; at that point,
15523 @code{handle_exception} returns control to your own code on the target
15527 @cindex @code{breakpoint} subroutine, remote
15528 Use this auxiliary subroutine to make your program contain a
15529 breakpoint. Depending on the particular situation, this may be the only
15530 way for @value{GDBN} to get control. For instance, if your target
15531 machine has some sort of interrupt button, you won't need to call this;
15532 pressing the interrupt button transfers control to
15533 @code{handle_exception}---in effect, to @value{GDBN}. On some machines,
15534 simply receiving characters on the serial port may also trigger a trap;
15535 again, in that situation, you don't need to call @code{breakpoint} from
15536 your own program---simply running @samp{target remote} from the host
15537 @value{GDBN} session gets control.
15539 Call @code{breakpoint} if none of these is true, or if you simply want
15540 to make certain your program stops at a predetermined point for the
15541 start of your debugging session.
15544 @node Bootstrapping
15545 @subsection What You Must Do for the Stub
15547 @cindex remote stub, support routines
15548 The debugging stubs that come with @value{GDBN} are set up for a particular
15549 chip architecture, but they have no information about the rest of your
15550 debugging target machine.
15552 First of all you need to tell the stub how to communicate with the
15556 @item int getDebugChar()
15557 @findex getDebugChar
15558 Write this subroutine to read a single character from the serial port.
15559 It may be identical to @code{getchar} for your target system; a
15560 different name is used to allow you to distinguish the two if you wish.
15562 @item void putDebugChar(int)
15563 @findex putDebugChar
15564 Write this subroutine to write a single character to the serial port.
15565 It may be identical to @code{putchar} for your target system; a
15566 different name is used to allow you to distinguish the two if you wish.
15569 @cindex control C, and remote debugging
15570 @cindex interrupting remote targets
15571 If you want @value{GDBN} to be able to stop your program while it is
15572 running, you need to use an interrupt-driven serial driver, and arrange
15573 for it to stop when it receives a @code{^C} (@samp{\003}, the control-C
15574 character). That is the character which @value{GDBN} uses to tell the
15575 remote system to stop.
15577 Getting the debugging target to return the proper status to @value{GDBN}
15578 probably requires changes to the standard stub; one quick and dirty way
15579 is to just execute a breakpoint instruction (the ``dirty'' part is that
15580 @value{GDBN} reports a @code{SIGTRAP} instead of a @code{SIGINT}).
15582 Other routines you need to supply are:
15585 @item void exceptionHandler (int @var{exception_number}, void *@var{exception_address})
15586 @findex exceptionHandler
15587 Write this function to install @var{exception_address} in the exception
15588 handling tables. You need to do this because the stub does not have any
15589 way of knowing what the exception handling tables on your target system
15590 are like (for example, the processor's table might be in @sc{rom},
15591 containing entries which point to a table in @sc{ram}).
15592 @var{exception_number} is the exception number which should be changed;
15593 its meaning is architecture-dependent (for example, different numbers
15594 might represent divide by zero, misaligned access, etc). When this
15595 exception occurs, control should be transferred directly to
15596 @var{exception_address}, and the processor state (stack, registers,
15597 and so on) should be just as it is when a processor exception occurs. So if
15598 you want to use a jump instruction to reach @var{exception_address}, it
15599 should be a simple jump, not a jump to subroutine.
15601 For the 386, @var{exception_address} should be installed as an interrupt
15602 gate so that interrupts are masked while the handler runs. The gate
15603 should be at privilege level 0 (the most privileged level). The
15604 @sc{sparc} and 68k stubs are able to mask interrupts themselves without
15605 help from @code{exceptionHandler}.
15607 @item void flush_i_cache()
15608 @findex flush_i_cache
15609 On @sc{sparc} and @sc{sparclite} only, write this subroutine to flush the
15610 instruction cache, if any, on your target machine. If there is no
15611 instruction cache, this subroutine may be a no-op.
15613 On target machines that have instruction caches, @value{GDBN} requires this
15614 function to make certain that the state of your program is stable.
15618 You must also make sure this library routine is available:
15621 @item void *memset(void *, int, int)
15623 This is the standard library function @code{memset} that sets an area of
15624 memory to a known value. If you have one of the free versions of
15625 @code{libc.a}, @code{memset} can be found there; otherwise, you must
15626 either obtain it from your hardware manufacturer, or write your own.
15629 If you do not use the GNU C compiler, you may need other standard
15630 library subroutines as well; this varies from one stub to another,
15631 but in general the stubs are likely to use any of the common library
15632 subroutines which @code{@value{NGCC}} generates as inline code.
15635 @node Debug Session
15636 @subsection Putting it All Together
15638 @cindex remote serial debugging summary
15639 In summary, when your program is ready to debug, you must follow these
15644 Make sure you have defined the supporting low-level routines
15645 (@pxref{Bootstrapping,,What You Must Do for the Stub}):
15647 @code{getDebugChar}, @code{putDebugChar},
15648 @code{flush_i_cache}, @code{memset}, @code{exceptionHandler}.
15652 Insert these lines near the top of your program:
15660 For the 680x0 stub only, you need to provide a variable called
15661 @code{exceptionHook}. Normally you just use:
15664 void (*exceptionHook)() = 0;
15668 but if before calling @code{set_debug_traps}, you set it to point to a
15669 function in your program, that function is called when
15670 @code{@value{GDBN}} continues after stopping on a trap (for example, bus
15671 error). The function indicated by @code{exceptionHook} is called with
15672 one parameter: an @code{int} which is the exception number.
15675 Compile and link together: your program, the @value{GDBN} debugging stub for
15676 your target architecture, and the supporting subroutines.
15679 Make sure you have a serial connection between your target machine and
15680 the @value{GDBN} host, and identify the serial port on the host.
15683 @c The "remote" target now provides a `load' command, so we should
15684 @c document that. FIXME.
15685 Download your program to your target machine (or get it there by
15686 whatever means the manufacturer provides), and start it.
15689 Start @value{GDBN} on the host, and connect to the target
15690 (@pxref{Connecting,,Connecting to a Remote Target}).
15694 @node Configurations
15695 @chapter Configuration-Specific Information
15697 While nearly all @value{GDBN} commands are available for all native and
15698 cross versions of the debugger, there are some exceptions. This chapter
15699 describes things that are only available in certain configurations.
15701 There are three major categories of configurations: native
15702 configurations, where the host and target are the same, embedded
15703 operating system configurations, which are usually the same for several
15704 different processor architectures, and bare embedded processors, which
15705 are quite different from each other.
15710 * Embedded Processors::
15717 This section describes details specific to particular native
15722 * BSD libkvm Interface:: Debugging BSD kernel memory images
15723 * SVR4 Process Information:: SVR4 process information
15724 * DJGPP Native:: Features specific to the DJGPP port
15725 * Cygwin Native:: Features specific to the Cygwin port
15726 * Hurd Native:: Features specific to @sc{gnu} Hurd
15727 * Neutrino:: Features specific to QNX Neutrino
15728 * Darwin:: Features specific to Darwin
15734 On HP-UX systems, if you refer to a function or variable name that
15735 begins with a dollar sign, @value{GDBN} searches for a user or system
15736 name first, before it searches for a convenience variable.
15739 @node BSD libkvm Interface
15740 @subsection BSD libkvm Interface
15743 @cindex kernel memory image
15744 @cindex kernel crash dump
15746 BSD-derived systems (FreeBSD/NetBSD/OpenBSD) have a kernel memory
15747 interface that provides a uniform interface for accessing kernel virtual
15748 memory images, including live systems and crash dumps. @value{GDBN}
15749 uses this interface to allow you to debug live kernels and kernel crash
15750 dumps on many native BSD configurations. This is implemented as a
15751 special @code{kvm} debugging target. For debugging a live system, load
15752 the currently running kernel into @value{GDBN} and connect to the
15756 (@value{GDBP}) @b{target kvm}
15759 For debugging crash dumps, provide the file name of the crash dump as an
15763 (@value{GDBP}) @b{target kvm /var/crash/bsd.0}
15766 Once connected to the @code{kvm} target, the following commands are
15772 Set current context from the @dfn{Process Control Block} (PCB) address.
15775 Set current context from proc address. This command isn't available on
15776 modern FreeBSD systems.
15779 @node SVR4 Process Information
15780 @subsection SVR4 Process Information
15782 @cindex examine process image
15783 @cindex process info via @file{/proc}
15785 Many versions of SVR4 and compatible systems provide a facility called
15786 @samp{/proc} that can be used to examine the image of a running
15787 process using file-system subroutines. If @value{GDBN} is configured
15788 for an operating system with this facility, the command @code{info
15789 proc} is available to report information about the process running
15790 your program, or about any process running on your system. @code{info
15791 proc} works only on SVR4 systems that include the @code{procfs} code.
15792 This includes, as of this writing, @sc{gnu}/Linux, OSF/1 (Digital
15793 Unix), Solaris, Irix, and Unixware, but not HP-UX, for example.
15799 @itemx info proc @var{process-id}
15800 Summarize available information about any running process. If a
15801 process ID is specified by @var{process-id}, display information about
15802 that process; otherwise display information about the program being
15803 debugged. The summary includes the debugged process ID, the command
15804 line used to invoke it, its current working directory, and its
15805 executable file's absolute file name.
15807 On some systems, @var{process-id} can be of the form
15808 @samp{[@var{pid}]/@var{tid}} which specifies a certain thread ID
15809 within a process. If the optional @var{pid} part is missing, it means
15810 a thread from the process being debugged (the leading @samp{/} still
15811 needs to be present, or else @value{GDBN} will interpret the number as
15812 a process ID rather than a thread ID).
15814 @item info proc mappings
15815 @cindex memory address space mappings
15816 Report the memory address space ranges accessible in the program, with
15817 information on whether the process has read, write, or execute access
15818 rights to each range. On @sc{gnu}/Linux systems, each memory range
15819 includes the object file which is mapped to that range, instead of the
15820 memory access rights to that range.
15822 @item info proc stat
15823 @itemx info proc status
15824 @cindex process detailed status information
15825 These subcommands are specific to @sc{gnu}/Linux systems. They show
15826 the process-related information, including the user ID and group ID;
15827 how many threads are there in the process; its virtual memory usage;
15828 the signals that are pending, blocked, and ignored; its TTY; its
15829 consumption of system and user time; its stack size; its @samp{nice}
15830 value; etc. For more information, see the @samp{proc} man page
15831 (type @kbd{man 5 proc} from your shell prompt).
15833 @item info proc all
15834 Show all the information about the process described under all of the
15835 above @code{info proc} subcommands.
15838 @comment These sub-options of 'info proc' were not included when
15839 @comment procfs.c was re-written. Keep their descriptions around
15840 @comment against the day when someone finds the time to put them back in.
15841 @kindex info proc times
15842 @item info proc times
15843 Starting time, user CPU time, and system CPU time for your program and
15846 @kindex info proc id
15848 Report on the process IDs related to your program: its own process ID,
15849 the ID of its parent, the process group ID, and the session ID.
15852 @item set procfs-trace
15853 @kindex set procfs-trace
15854 @cindex @code{procfs} API calls
15855 This command enables and disables tracing of @code{procfs} API calls.
15857 @item show procfs-trace
15858 @kindex show procfs-trace
15859 Show the current state of @code{procfs} API call tracing.
15861 @item set procfs-file @var{file}
15862 @kindex set procfs-file
15863 Tell @value{GDBN} to write @code{procfs} API trace to the named
15864 @var{file}. @value{GDBN} appends the trace info to the previous
15865 contents of the file. The default is to display the trace on the
15868 @item show procfs-file
15869 @kindex show procfs-file
15870 Show the file to which @code{procfs} API trace is written.
15872 @item proc-trace-entry
15873 @itemx proc-trace-exit
15874 @itemx proc-untrace-entry
15875 @itemx proc-untrace-exit
15876 @kindex proc-trace-entry
15877 @kindex proc-trace-exit
15878 @kindex proc-untrace-entry
15879 @kindex proc-untrace-exit
15880 These commands enable and disable tracing of entries into and exits
15881 from the @code{syscall} interface.
15884 @kindex info pidlist
15885 @cindex process list, QNX Neutrino
15886 For QNX Neutrino only, this command displays the list of all the
15887 processes and all the threads within each process.
15890 @kindex info meminfo
15891 @cindex mapinfo list, QNX Neutrino
15892 For QNX Neutrino only, this command displays the list of all mapinfos.
15896 @subsection Features for Debugging @sc{djgpp} Programs
15897 @cindex @sc{djgpp} debugging
15898 @cindex native @sc{djgpp} debugging
15899 @cindex MS-DOS-specific commands
15902 @sc{djgpp} is a port of the @sc{gnu} development tools to MS-DOS and
15903 MS-Windows. @sc{djgpp} programs are 32-bit protected-mode programs
15904 that use the @dfn{DPMI} (DOS Protected-Mode Interface) API to run on
15905 top of real-mode DOS systems and their emulations.
15907 @value{GDBN} supports native debugging of @sc{djgpp} programs, and
15908 defines a few commands specific to the @sc{djgpp} port. This
15909 subsection describes those commands.
15914 This is a prefix of @sc{djgpp}-specific commands which print
15915 information about the target system and important OS structures.
15918 @cindex MS-DOS system info
15919 @cindex free memory information (MS-DOS)
15920 @item info dos sysinfo
15921 This command displays assorted information about the underlying
15922 platform: the CPU type and features, the OS version and flavor, the
15923 DPMI version, and the available conventional and DPMI memory.
15928 @cindex segment descriptor tables
15929 @cindex descriptor tables display
15931 @itemx info dos ldt
15932 @itemx info dos idt
15933 These 3 commands display entries from, respectively, Global, Local,
15934 and Interrupt Descriptor Tables (GDT, LDT, and IDT). The descriptor
15935 tables are data structures which store a descriptor for each segment
15936 that is currently in use. The segment's selector is an index into a
15937 descriptor table; the table entry for that index holds the
15938 descriptor's base address and limit, and its attributes and access
15941 A typical @sc{djgpp} program uses 3 segments: a code segment, a data
15942 segment (used for both data and the stack), and a DOS segment (which
15943 allows access to DOS/BIOS data structures and absolute addresses in
15944 conventional memory). However, the DPMI host will usually define
15945 additional segments in order to support the DPMI environment.
15947 @cindex garbled pointers
15948 These commands allow to display entries from the descriptor tables.
15949 Without an argument, all entries from the specified table are
15950 displayed. An argument, which should be an integer expression, means
15951 display a single entry whose index is given by the argument. For
15952 example, here's a convenient way to display information about the
15953 debugged program's data segment:
15956 @exdent @code{(@value{GDBP}) info dos ldt $ds}
15957 @exdent @code{0x13f: base=0x11970000 limit=0x0009ffff 32-Bit Data (Read/Write, Exp-up)}
15961 This comes in handy when you want to see whether a pointer is outside
15962 the data segment's limit (i.e.@: @dfn{garbled}).
15964 @cindex page tables display (MS-DOS)
15966 @itemx info dos pte
15967 These two commands display entries from, respectively, the Page
15968 Directory and the Page Tables. Page Directories and Page Tables are
15969 data structures which control how virtual memory addresses are mapped
15970 into physical addresses. A Page Table includes an entry for every
15971 page of memory that is mapped into the program's address space; there
15972 may be several Page Tables, each one holding up to 4096 entries. A
15973 Page Directory has up to 4096 entries, one each for every Page Table
15974 that is currently in use.
15976 Without an argument, @kbd{info dos pde} displays the entire Page
15977 Directory, and @kbd{info dos pte} displays all the entries in all of
15978 the Page Tables. An argument, an integer expression, given to the
15979 @kbd{info dos pde} command means display only that entry from the Page
15980 Directory table. An argument given to the @kbd{info dos pte} command
15981 means display entries from a single Page Table, the one pointed to by
15982 the specified entry in the Page Directory.
15984 @cindex direct memory access (DMA) on MS-DOS
15985 These commands are useful when your program uses @dfn{DMA} (Direct
15986 Memory Access), which needs physical addresses to program the DMA
15989 These commands are supported only with some DPMI servers.
15991 @cindex physical address from linear address
15992 @item info dos address-pte @var{addr}
15993 This command displays the Page Table entry for a specified linear
15994 address. The argument @var{addr} is a linear address which should
15995 already have the appropriate segment's base address added to it,
15996 because this command accepts addresses which may belong to @emph{any}
15997 segment. For example, here's how to display the Page Table entry for
15998 the page where a variable @code{i} is stored:
16001 @exdent @code{(@value{GDBP}) info dos address-pte __djgpp_base_address + (char *)&i}
16002 @exdent @code{Page Table entry for address 0x11a00d30:}
16003 @exdent @code{Base=0x02698000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0xd30}
16007 This says that @code{i} is stored at offset @code{0xd30} from the page
16008 whose physical base address is @code{0x02698000}, and shows all the
16009 attributes of that page.
16011 Note that you must cast the addresses of variables to a @code{char *},
16012 since otherwise the value of @code{__djgpp_base_address}, the base
16013 address of all variables and functions in a @sc{djgpp} program, will
16014 be added using the rules of C pointer arithmetics: if @code{i} is
16015 declared an @code{int}, @value{GDBN} will add 4 times the value of
16016 @code{__djgpp_base_address} to the address of @code{i}.
16018 Here's another example, it displays the Page Table entry for the
16022 @exdent @code{(@value{GDBP}) info dos address-pte *((unsigned *)&_go32_info_block + 3)}
16023 @exdent @code{Page Table entry for address 0x29110:}
16024 @exdent @code{Base=0x00029000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0x110}
16028 (The @code{+ 3} offset is because the transfer buffer's address is the
16029 3rd member of the @code{_go32_info_block} structure.) The output
16030 clearly shows that this DPMI server maps the addresses in conventional
16031 memory 1:1, i.e.@: the physical (@code{0x00029000} + @code{0x110}) and
16032 linear (@code{0x29110}) addresses are identical.
16034 This command is supported only with some DPMI servers.
16037 @cindex DOS serial data link, remote debugging
16038 In addition to native debugging, the DJGPP port supports remote
16039 debugging via a serial data link. The following commands are specific
16040 to remote serial debugging in the DJGPP port of @value{GDBN}.
16043 @kindex set com1base
16044 @kindex set com1irq
16045 @kindex set com2base
16046 @kindex set com2irq
16047 @kindex set com3base
16048 @kindex set com3irq
16049 @kindex set com4base
16050 @kindex set com4irq
16051 @item set com1base @var{addr}
16052 This command sets the base I/O port address of the @file{COM1} serial
16055 @item set com1irq @var{irq}
16056 This command sets the @dfn{Interrupt Request} (@code{IRQ}) line to use
16057 for the @file{COM1} serial port.
16059 There are similar commands @samp{set com2base}, @samp{set com3irq},
16060 etc.@: for setting the port address and the @code{IRQ} lines for the
16063 @kindex show com1base
16064 @kindex show com1irq
16065 @kindex show com2base
16066 @kindex show com2irq
16067 @kindex show com3base
16068 @kindex show com3irq
16069 @kindex show com4base
16070 @kindex show com4irq
16071 The related commands @samp{show com1base}, @samp{show com1irq} etc.@:
16072 display the current settings of the base address and the @code{IRQ}
16073 lines used by the COM ports.
16076 @kindex info serial
16077 @cindex DOS serial port status
16078 This command prints the status of the 4 DOS serial ports. For each
16079 port, it prints whether it's active or not, its I/O base address and
16080 IRQ number, whether it uses a 16550-style FIFO, its baudrate, and the
16081 counts of various errors encountered so far.
16085 @node Cygwin Native
16086 @subsection Features for Debugging MS Windows PE Executables
16087 @cindex MS Windows debugging
16088 @cindex native Cygwin debugging
16089 @cindex Cygwin-specific commands
16091 @value{GDBN} supports native debugging of MS Windows programs, including
16092 DLLs with and without symbolic debugging information.
16094 @cindex Ctrl-BREAK, MS-Windows
16095 @cindex interrupt debuggee on MS-Windows
16096 MS-Windows programs that call @code{SetConsoleMode} to switch off the
16097 special meaning of the @samp{Ctrl-C} keystroke cannot be interrupted
16098 by typing @kbd{C-c}. For this reason, @value{GDBN} on MS-Windows
16099 supports @kbd{C-@key{BREAK}} as an alternative interrupt key
16100 sequence, which can be used to interrupt the debuggee even if it
16103 There are various additional Cygwin-specific commands, described in
16104 this section. Working with DLLs that have no debugging symbols is
16105 described in @ref{Non-debug DLL Symbols}.
16110 This is a prefix of MS Windows-specific commands which print
16111 information about the target system and important OS structures.
16113 @item info w32 selector
16114 This command displays information returned by
16115 the Win32 API @code{GetThreadSelectorEntry} function.
16116 It takes an optional argument that is evaluated to
16117 a long value to give the information about this given selector.
16118 Without argument, this command displays information
16119 about the six segment registers.
16123 This is a Cygwin-specific alias of @code{info shared}.
16125 @kindex dll-symbols
16127 This command loads symbols from a dll similarly to
16128 add-sym command but without the need to specify a base address.
16130 @kindex set cygwin-exceptions
16131 @cindex debugging the Cygwin DLL
16132 @cindex Cygwin DLL, debugging
16133 @item set cygwin-exceptions @var{mode}
16134 If @var{mode} is @code{on}, @value{GDBN} will break on exceptions that
16135 happen inside the Cygwin DLL. If @var{mode} is @code{off},
16136 @value{GDBN} will delay recognition of exceptions, and may ignore some
16137 exceptions which seem to be caused by internal Cygwin DLL
16138 ``bookkeeping''. This option is meant primarily for debugging the
16139 Cygwin DLL itself; the default value is @code{off} to avoid annoying
16140 @value{GDBN} users with false @code{SIGSEGV} signals.
16142 @kindex show cygwin-exceptions
16143 @item show cygwin-exceptions
16144 Displays whether @value{GDBN} will break on exceptions that happen
16145 inside the Cygwin DLL itself.
16147 @kindex set new-console
16148 @item set new-console @var{mode}
16149 If @var{mode} is @code{on} the debuggee will
16150 be started in a new console on next start.
16151 If @var{mode} is @code{off}i, the debuggee will
16152 be started in the same console as the debugger.
16154 @kindex show new-console
16155 @item show new-console
16156 Displays whether a new console is used
16157 when the debuggee is started.
16159 @kindex set new-group
16160 @item set new-group @var{mode}
16161 This boolean value controls whether the debuggee should
16162 start a new group or stay in the same group as the debugger.
16163 This affects the way the Windows OS handles
16166 @kindex show new-group
16167 @item show new-group
16168 Displays current value of new-group boolean.
16170 @kindex set debugevents
16171 @item set debugevents
16172 This boolean value adds debug output concerning kernel events related
16173 to the debuggee seen by the debugger. This includes events that
16174 signal thread and process creation and exit, DLL loading and
16175 unloading, console interrupts, and debugging messages produced by the
16176 Windows @code{OutputDebugString} API call.
16178 @kindex set debugexec
16179 @item set debugexec
16180 This boolean value adds debug output concerning execute events
16181 (such as resume thread) seen by the debugger.
16183 @kindex set debugexceptions
16184 @item set debugexceptions
16185 This boolean value adds debug output concerning exceptions in the
16186 debuggee seen by the debugger.
16188 @kindex set debugmemory
16189 @item set debugmemory
16190 This boolean value adds debug output concerning debuggee memory reads
16191 and writes by the debugger.
16195 This boolean values specifies whether the debuggee is called
16196 via a shell or directly (default value is on).
16200 Displays if the debuggee will be started with a shell.
16205 * Non-debug DLL Symbols:: Support for DLLs without debugging symbols
16208 @node Non-debug DLL Symbols
16209 @subsubsection Support for DLLs without Debugging Symbols
16210 @cindex DLLs with no debugging symbols
16211 @cindex Minimal symbols and DLLs
16213 Very often on windows, some of the DLLs that your program relies on do
16214 not include symbolic debugging information (for example,
16215 @file{kernel32.dll}). When @value{GDBN} doesn't recognize any debugging
16216 symbols in a DLL, it relies on the minimal amount of symbolic
16217 information contained in the DLL's export table. This section
16218 describes working with such symbols, known internally to @value{GDBN} as
16219 ``minimal symbols''.
16221 Note that before the debugged program has started execution, no DLLs
16222 will have been loaded. The easiest way around this problem is simply to
16223 start the program --- either by setting a breakpoint or letting the
16224 program run once to completion. It is also possible to force
16225 @value{GDBN} to load a particular DLL before starting the executable ---
16226 see the shared library information in @ref{Files}, or the
16227 @code{dll-symbols} command in @ref{Cygwin Native}. Currently,
16228 explicitly loading symbols from a DLL with no debugging information will
16229 cause the symbol names to be duplicated in @value{GDBN}'s lookup table,
16230 which may adversely affect symbol lookup performance.
16232 @subsubsection DLL Name Prefixes
16234 In keeping with the naming conventions used by the Microsoft debugging
16235 tools, DLL export symbols are made available with a prefix based on the
16236 DLL name, for instance @code{KERNEL32!CreateFileA}. The plain name is
16237 also entered into the symbol table, so @code{CreateFileA} is often
16238 sufficient. In some cases there will be name clashes within a program
16239 (particularly if the executable itself includes full debugging symbols)
16240 necessitating the use of the fully qualified name when referring to the
16241 contents of the DLL. Use single-quotes around the name to avoid the
16242 exclamation mark (``!'') being interpreted as a language operator.
16244 Note that the internal name of the DLL may be all upper-case, even
16245 though the file name of the DLL is lower-case, or vice-versa. Since
16246 symbols within @value{GDBN} are @emph{case-sensitive} this may cause
16247 some confusion. If in doubt, try the @code{info functions} and
16248 @code{info variables} commands or even @code{maint print msymbols}
16249 (@pxref{Symbols}). Here's an example:
16252 (@value{GDBP}) info function CreateFileA
16253 All functions matching regular expression "CreateFileA":
16255 Non-debugging symbols:
16256 0x77e885f4 CreateFileA
16257 0x77e885f4 KERNEL32!CreateFileA
16261 (@value{GDBP}) info function !
16262 All functions matching regular expression "!":
16264 Non-debugging symbols:
16265 0x6100114c cygwin1!__assert
16266 0x61004034 cygwin1!_dll_crt0@@0
16267 0x61004240 cygwin1!dll_crt0(per_process *)
16271 @subsubsection Working with Minimal Symbols
16273 Symbols extracted from a DLL's export table do not contain very much
16274 type information. All that @value{GDBN} can do is guess whether a symbol
16275 refers to a function or variable depending on the linker section that
16276 contains the symbol. Also note that the actual contents of the memory
16277 contained in a DLL are not available unless the program is running. This
16278 means that you cannot examine the contents of a variable or disassemble
16279 a function within a DLL without a running program.
16281 Variables are generally treated as pointers and dereferenced
16282 automatically. For this reason, it is often necessary to prefix a
16283 variable name with the address-of operator (``&'') and provide explicit
16284 type information in the command. Here's an example of the type of
16288 (@value{GDBP}) print 'cygwin1!__argv'
16293 (@value{GDBP}) x 'cygwin1!__argv'
16294 0x10021610: "\230y\""
16297 And two possible solutions:
16300 (@value{GDBP}) print ((char **)'cygwin1!__argv')[0]
16301 $2 = 0x22fd98 "/cygdrive/c/mydirectory/myprogram"
16305 (@value{GDBP}) x/2x &'cygwin1!__argv'
16306 0x610c0aa8 <cygwin1!__argv>: 0x10021608 0x00000000
16307 (@value{GDBP}) x/x 0x10021608
16308 0x10021608: 0x0022fd98
16309 (@value{GDBP}) x/s 0x0022fd98
16310 0x22fd98: "/cygdrive/c/mydirectory/myprogram"
16313 Setting a break point within a DLL is possible even before the program
16314 starts execution. However, under these circumstances, @value{GDBN} can't
16315 examine the initial instructions of the function in order to skip the
16316 function's frame set-up code. You can work around this by using ``*&''
16317 to set the breakpoint at a raw memory address:
16320 (@value{GDBP}) break *&'python22!PyOS_Readline'
16321 Breakpoint 1 at 0x1e04eff0
16324 The author of these extensions is not entirely convinced that setting a
16325 break point within a shared DLL like @file{kernel32.dll} is completely
16329 @subsection Commands Specific to @sc{gnu} Hurd Systems
16330 @cindex @sc{gnu} Hurd debugging
16332 This subsection describes @value{GDBN} commands specific to the
16333 @sc{gnu} Hurd native debugging.
16338 @kindex set signals@r{, Hurd command}
16339 @kindex set sigs@r{, Hurd command}
16340 This command toggles the state of inferior signal interception by
16341 @value{GDBN}. Mach exceptions, such as breakpoint traps, are not
16342 affected by this command. @code{sigs} is a shorthand alias for
16347 @kindex show signals@r{, Hurd command}
16348 @kindex show sigs@r{, Hurd command}
16349 Show the current state of intercepting inferior's signals.
16351 @item set signal-thread
16352 @itemx set sigthread
16353 @kindex set signal-thread
16354 @kindex set sigthread
16355 This command tells @value{GDBN} which thread is the @code{libc} signal
16356 thread. That thread is run when a signal is delivered to a running
16357 process. @code{set sigthread} is the shorthand alias of @code{set
16360 @item show signal-thread
16361 @itemx show sigthread
16362 @kindex show signal-thread
16363 @kindex show sigthread
16364 These two commands show which thread will run when the inferior is
16365 delivered a signal.
16368 @kindex set stopped@r{, Hurd command}
16369 This commands tells @value{GDBN} that the inferior process is stopped,
16370 as with the @code{SIGSTOP} signal. The stopped process can be
16371 continued by delivering a signal to it.
16374 @kindex show stopped@r{, Hurd command}
16375 This command shows whether @value{GDBN} thinks the debuggee is
16378 @item set exceptions
16379 @kindex set exceptions@r{, Hurd command}
16380 Use this command to turn off trapping of exceptions in the inferior.
16381 When exception trapping is off, neither breakpoints nor
16382 single-stepping will work. To restore the default, set exception
16385 @item show exceptions
16386 @kindex show exceptions@r{, Hurd command}
16387 Show the current state of trapping exceptions in the inferior.
16389 @item set task pause
16390 @kindex set task@r{, Hurd commands}
16391 @cindex task attributes (@sc{gnu} Hurd)
16392 @cindex pause current task (@sc{gnu} Hurd)
16393 This command toggles task suspension when @value{GDBN} has control.
16394 Setting it to on takes effect immediately, and the task is suspended
16395 whenever @value{GDBN} gets control. Setting it to off will take
16396 effect the next time the inferior is continued. If this option is set
16397 to off, you can use @code{set thread default pause on} or @code{set
16398 thread pause on} (see below) to pause individual threads.
16400 @item show task pause
16401 @kindex show task@r{, Hurd commands}
16402 Show the current state of task suspension.
16404 @item set task detach-suspend-count
16405 @cindex task suspend count
16406 @cindex detach from task, @sc{gnu} Hurd
16407 This command sets the suspend count the task will be left with when
16408 @value{GDBN} detaches from it.
16410 @item show task detach-suspend-count
16411 Show the suspend count the task will be left with when detaching.
16413 @item set task exception-port
16414 @itemx set task excp
16415 @cindex task exception port, @sc{gnu} Hurd
16416 This command sets the task exception port to which @value{GDBN} will
16417 forward exceptions. The argument should be the value of the @dfn{send
16418 rights} of the task. @code{set task excp} is a shorthand alias.
16420 @item set noninvasive
16421 @cindex noninvasive task options
16422 This command switches @value{GDBN} to a mode that is the least
16423 invasive as far as interfering with the inferior is concerned. This
16424 is the same as using @code{set task pause}, @code{set exceptions}, and
16425 @code{set signals} to values opposite to the defaults.
16427 @item info send-rights
16428 @itemx info receive-rights
16429 @itemx info port-rights
16430 @itemx info port-sets
16431 @itemx info dead-names
16434 @cindex send rights, @sc{gnu} Hurd
16435 @cindex receive rights, @sc{gnu} Hurd
16436 @cindex port rights, @sc{gnu} Hurd
16437 @cindex port sets, @sc{gnu} Hurd
16438 @cindex dead names, @sc{gnu} Hurd
16439 These commands display information about, respectively, send rights,
16440 receive rights, port rights, port sets, and dead names of a task.
16441 There are also shorthand aliases: @code{info ports} for @code{info
16442 port-rights} and @code{info psets} for @code{info port-sets}.
16444 @item set thread pause
16445 @kindex set thread@r{, Hurd command}
16446 @cindex thread properties, @sc{gnu} Hurd
16447 @cindex pause current thread (@sc{gnu} Hurd)
16448 This command toggles current thread suspension when @value{GDBN} has
16449 control. Setting it to on takes effect immediately, and the current
16450 thread is suspended whenever @value{GDBN} gets control. Setting it to
16451 off will take effect the next time the inferior is continued.
16452 Normally, this command has no effect, since when @value{GDBN} has
16453 control, the whole task is suspended. However, if you used @code{set
16454 task pause off} (see above), this command comes in handy to suspend
16455 only the current thread.
16457 @item show thread pause
16458 @kindex show thread@r{, Hurd command}
16459 This command shows the state of current thread suspension.
16461 @item set thread run
16462 This command sets whether the current thread is allowed to run.
16464 @item show thread run
16465 Show whether the current thread is allowed to run.
16467 @item set thread detach-suspend-count
16468 @cindex thread suspend count, @sc{gnu} Hurd
16469 @cindex detach from thread, @sc{gnu} Hurd
16470 This command sets the suspend count @value{GDBN} will leave on a
16471 thread when detaching. This number is relative to the suspend count
16472 found by @value{GDBN} when it notices the thread; use @code{set thread
16473 takeover-suspend-count} to force it to an absolute value.
16475 @item show thread detach-suspend-count
16476 Show the suspend count @value{GDBN} will leave on the thread when
16479 @item set thread exception-port
16480 @itemx set thread excp
16481 Set the thread exception port to which to forward exceptions. This
16482 overrides the port set by @code{set task exception-port} (see above).
16483 @code{set thread excp} is the shorthand alias.
16485 @item set thread takeover-suspend-count
16486 Normally, @value{GDBN}'s thread suspend counts are relative to the
16487 value @value{GDBN} finds when it notices each thread. This command
16488 changes the suspend counts to be absolute instead.
16490 @item set thread default
16491 @itemx show thread default
16492 @cindex thread default settings, @sc{gnu} Hurd
16493 Each of the above @code{set thread} commands has a @code{set thread
16494 default} counterpart (e.g., @code{set thread default pause}, @code{set
16495 thread default exception-port}, etc.). The @code{thread default}
16496 variety of commands sets the default thread properties for all
16497 threads; you can then change the properties of individual threads with
16498 the non-default commands.
16503 @subsection QNX Neutrino
16504 @cindex QNX Neutrino
16506 @value{GDBN} provides the following commands specific to the QNX
16510 @item set debug nto-debug
16511 @kindex set debug nto-debug
16512 When set to on, enables debugging messages specific to the QNX
16515 @item show debug nto-debug
16516 @kindex show debug nto-debug
16517 Show the current state of QNX Neutrino messages.
16524 @value{GDBN} provides the following commands specific to the Darwin target:
16527 @item set debug darwin @var{num}
16528 @kindex set debug darwin
16529 When set to a non zero value, enables debugging messages specific to
16530 the Darwin support. Higher values produce more verbose output.
16532 @item show debug darwin
16533 @kindex show debug darwin
16534 Show the current state of Darwin messages.
16536 @item set debug mach-o @var{num}
16537 @kindex set debug mach-o
16538 When set to a non zero value, enables debugging messages while
16539 @value{GDBN} is reading Darwin object files. (@dfn{Mach-O} is the
16540 file format used on Darwin for object and executable files.) Higher
16541 values produce more verbose output. This is a command to diagnose
16542 problems internal to @value{GDBN} and should not be needed in normal
16545 @item show debug mach-o
16546 @kindex show debug mach-o
16547 Show the current state of Mach-O file messages.
16549 @item set mach-exceptions on
16550 @itemx set mach-exceptions off
16551 @kindex set mach-exceptions
16552 On Darwin, faults are first reported as a Mach exception and are then
16553 mapped to a Posix signal. Use this command to turn on trapping of
16554 Mach exceptions in the inferior. This might be sometimes useful to
16555 better understand the cause of a fault. The default is off.
16557 @item show mach-exceptions
16558 @kindex show mach-exceptions
16559 Show the current state of exceptions trapping.
16564 @section Embedded Operating Systems
16566 This section describes configurations involving the debugging of
16567 embedded operating systems that are available for several different
16571 * VxWorks:: Using @value{GDBN} with VxWorks
16574 @value{GDBN} includes the ability to debug programs running on
16575 various real-time operating systems.
16578 @subsection Using @value{GDBN} with VxWorks
16584 @kindex target vxworks
16585 @item target vxworks @var{machinename}
16586 A VxWorks system, attached via TCP/IP. The argument @var{machinename}
16587 is the target system's machine name or IP address.
16591 On VxWorks, @code{load} links @var{filename} dynamically on the
16592 current target system as well as adding its symbols in @value{GDBN}.
16594 @value{GDBN} enables developers to spawn and debug tasks running on networked
16595 VxWorks targets from a Unix host. Already-running tasks spawned from
16596 the VxWorks shell can also be debugged. @value{GDBN} uses code that runs on
16597 both the Unix host and on the VxWorks target. The program
16598 @code{@value{GDBP}} is installed and executed on the Unix host. (It may be
16599 installed with the name @code{vxgdb}, to distinguish it from a
16600 @value{GDBN} for debugging programs on the host itself.)
16603 @item VxWorks-timeout @var{args}
16604 @kindex vxworks-timeout
16605 All VxWorks-based targets now support the option @code{vxworks-timeout}.
16606 This option is set by the user, and @var{args} represents the number of
16607 seconds @value{GDBN} waits for responses to rpc's. You might use this if
16608 your VxWorks target is a slow software simulator or is on the far side
16609 of a thin network line.
16612 The following information on connecting to VxWorks was current when
16613 this manual was produced; newer releases of VxWorks may use revised
16616 @findex INCLUDE_RDB
16617 To use @value{GDBN} with VxWorks, you must rebuild your VxWorks kernel
16618 to include the remote debugging interface routines in the VxWorks
16619 library @file{rdb.a}. To do this, define @code{INCLUDE_RDB} in the
16620 VxWorks configuration file @file{configAll.h} and rebuild your VxWorks
16621 kernel. The resulting kernel contains @file{rdb.a}, and spawns the
16622 source debugging task @code{tRdbTask} when VxWorks is booted. For more
16623 information on configuring and remaking VxWorks, see the manufacturer's
16625 @c VxWorks, see the @cite{VxWorks Programmer's Guide}.
16627 Once you have included @file{rdb.a} in your VxWorks system image and set
16628 your Unix execution search path to find @value{GDBN}, you are ready to
16629 run @value{GDBN}. From your Unix host, run @code{@value{GDBP}} (or
16630 @code{vxgdb}, depending on your installation).
16632 @value{GDBN} comes up showing the prompt:
16639 * VxWorks Connection:: Connecting to VxWorks
16640 * VxWorks Download:: VxWorks download
16641 * VxWorks Attach:: Running tasks
16644 @node VxWorks Connection
16645 @subsubsection Connecting to VxWorks
16647 The @value{GDBN} command @code{target} lets you connect to a VxWorks target on the
16648 network. To connect to a target whose host name is ``@code{tt}'', type:
16651 (vxgdb) target vxworks tt
16655 @value{GDBN} displays messages like these:
16658 Attaching remote machine across net...
16663 @value{GDBN} then attempts to read the symbol tables of any object modules
16664 loaded into the VxWorks target since it was last booted. @value{GDBN} locates
16665 these files by searching the directories listed in the command search
16666 path (@pxref{Environment, ,Your Program's Environment}); if it fails
16667 to find an object file, it displays a message such as:
16670 prog.o: No such file or directory.
16673 When this happens, add the appropriate directory to the search path with
16674 the @value{GDBN} command @code{path}, and execute the @code{target}
16677 @node VxWorks Download
16678 @subsubsection VxWorks Download
16680 @cindex download to VxWorks
16681 If you have connected to the VxWorks target and you want to debug an
16682 object that has not yet been loaded, you can use the @value{GDBN}
16683 @code{load} command to download a file from Unix to VxWorks
16684 incrementally. The object file given as an argument to the @code{load}
16685 command is actually opened twice: first by the VxWorks target in order
16686 to download the code, then by @value{GDBN} in order to read the symbol
16687 table. This can lead to problems if the current working directories on
16688 the two systems differ. If both systems have NFS mounted the same
16689 filesystems, you can avoid these problems by using absolute paths.
16690 Otherwise, it is simplest to set the working directory on both systems
16691 to the directory in which the object file resides, and then to reference
16692 the file by its name, without any path. For instance, a program
16693 @file{prog.o} may reside in @file{@var{vxpath}/vw/demo/rdb} in VxWorks
16694 and in @file{@var{hostpath}/vw/demo/rdb} on the host. To load this
16695 program, type this on VxWorks:
16698 -> cd "@var{vxpath}/vw/demo/rdb"
16702 Then, in @value{GDBN}, type:
16705 (vxgdb) cd @var{hostpath}/vw/demo/rdb
16706 (vxgdb) load prog.o
16709 @value{GDBN} displays a response similar to this:
16712 Reading symbol data from wherever/vw/demo/rdb/prog.o... done.
16715 You can also use the @code{load} command to reload an object module
16716 after editing and recompiling the corresponding source file. Note that
16717 this makes @value{GDBN} delete all currently-defined breakpoints,
16718 auto-displays, and convenience variables, and to clear the value
16719 history. (This is necessary in order to preserve the integrity of
16720 debugger's data structures that reference the target system's symbol
16723 @node VxWorks Attach
16724 @subsubsection Running Tasks
16726 @cindex running VxWorks tasks
16727 You can also attach to an existing task using the @code{attach} command as
16731 (vxgdb) attach @var{task}
16735 where @var{task} is the VxWorks hexadecimal task ID. The task can be running
16736 or suspended when you attach to it. Running tasks are suspended at
16737 the time of attachment.
16739 @node Embedded Processors
16740 @section Embedded Processors
16742 This section goes into details specific to particular embedded
16745 @cindex send command to simulator
16746 Whenever a specific embedded processor has a simulator, @value{GDBN}
16747 allows to send an arbitrary command to the simulator.
16750 @item sim @var{command}
16751 @kindex sim@r{, a command}
16752 Send an arbitrary @var{command} string to the simulator. Consult the
16753 documentation for the specific simulator in use for information about
16754 acceptable commands.
16760 * M32R/D:: Renesas M32R/D
16761 * M68K:: Motorola M68K
16762 * MicroBlaze:: Xilinx MicroBlaze
16763 * MIPS Embedded:: MIPS Embedded
16764 * OpenRISC 1000:: OpenRisc 1000
16765 * PA:: HP PA Embedded
16766 * PowerPC Embedded:: PowerPC Embedded
16767 * Sparclet:: Tsqware Sparclet
16768 * Sparclite:: Fujitsu Sparclite
16769 * Z8000:: Zilog Z8000
16772 * Super-H:: Renesas Super-H
16781 @item target rdi @var{dev}
16782 ARM Angel monitor, via RDI library interface to ADP protocol. You may
16783 use this target to communicate with both boards running the Angel
16784 monitor, or with the EmbeddedICE JTAG debug device.
16787 @item target rdp @var{dev}
16792 @value{GDBN} provides the following ARM-specific commands:
16795 @item set arm disassembler
16797 This commands selects from a list of disassembly styles. The
16798 @code{"std"} style is the standard style.
16800 @item show arm disassembler
16802 Show the current disassembly style.
16804 @item set arm apcs32
16805 @cindex ARM 32-bit mode
16806 This command toggles ARM operation mode between 32-bit and 26-bit.
16808 @item show arm apcs32
16809 Display the current usage of the ARM 32-bit mode.
16811 @item set arm fpu @var{fputype}
16812 This command sets the ARM floating-point unit (FPU) type. The
16813 argument @var{fputype} can be one of these:
16817 Determine the FPU type by querying the OS ABI.
16819 Software FPU, with mixed-endian doubles on little-endian ARM
16822 GCC-compiled FPA co-processor.
16824 Software FPU with pure-endian doubles.
16830 Show the current type of the FPU.
16833 This command forces @value{GDBN} to use the specified ABI.
16836 Show the currently used ABI.
16838 @item set arm fallback-mode (arm|thumb|auto)
16839 @value{GDBN} uses the symbol table, when available, to determine
16840 whether instructions are ARM or Thumb. This command controls
16841 @value{GDBN}'s default behavior when the symbol table is not
16842 available. The default is @samp{auto}, which causes @value{GDBN} to
16843 use the current execution mode (from the @code{T} bit in the @code{CPSR}
16846 @item show arm fallback-mode
16847 Show the current fallback instruction mode.
16849 @item set arm force-mode (arm|thumb|auto)
16850 This command overrides use of the symbol table to determine whether
16851 instructions are ARM or Thumb. The default is @samp{auto}, which
16852 causes @value{GDBN} to use the symbol table and then the setting
16853 of @samp{set arm fallback-mode}.
16855 @item show arm force-mode
16856 Show the current forced instruction mode.
16858 @item set debug arm
16859 Toggle whether to display ARM-specific debugging messages from the ARM
16860 target support subsystem.
16862 @item show debug arm
16863 Show whether ARM-specific debugging messages are enabled.
16866 The following commands are available when an ARM target is debugged
16867 using the RDI interface:
16870 @item rdilogfile @r{[}@var{file}@r{]}
16872 @cindex ADP (Angel Debugger Protocol) logging
16873 Set the filename for the ADP (Angel Debugger Protocol) packet log.
16874 With an argument, sets the log file to the specified @var{file}. With
16875 no argument, show the current log file name. The default log file is
16878 @item rdilogenable @r{[}@var{arg}@r{]}
16879 @kindex rdilogenable
16880 Control logging of ADP packets. With an argument of 1 or @code{"yes"}
16881 enables logging, with an argument 0 or @code{"no"} disables it. With
16882 no arguments displays the current setting. When logging is enabled,
16883 ADP packets exchanged between @value{GDBN} and the RDI target device
16884 are logged to a file.
16886 @item set rdiromatzero
16887 @kindex set rdiromatzero
16888 @cindex ROM at zero address, RDI
16889 Tell @value{GDBN} whether the target has ROM at address 0. If on,
16890 vector catching is disabled, so that zero address can be used. If off
16891 (the default), vector catching is enabled. For this command to take
16892 effect, it needs to be invoked prior to the @code{target rdi} command.
16894 @item show rdiromatzero
16895 @kindex show rdiromatzero
16896 Show the current setting of ROM at zero address.
16898 @item set rdiheartbeat
16899 @kindex set rdiheartbeat
16900 @cindex RDI heartbeat
16901 Enable or disable RDI heartbeat packets. It is not recommended to
16902 turn on this option, since it confuses ARM and EPI JTAG interface, as
16903 well as the Angel monitor.
16905 @item show rdiheartbeat
16906 @kindex show rdiheartbeat
16907 Show the setting of RDI heartbeat packets.
16912 @subsection Renesas M32R/D and M32R/SDI
16915 @kindex target m32r
16916 @item target m32r @var{dev}
16917 Renesas M32R/D ROM monitor.
16919 @kindex target m32rsdi
16920 @item target m32rsdi @var{dev}
16921 Renesas M32R SDI server, connected via parallel port to the board.
16924 The following @value{GDBN} commands are specific to the M32R monitor:
16927 @item set download-path @var{path}
16928 @kindex set download-path
16929 @cindex find downloadable @sc{srec} files (M32R)
16930 Set the default path for finding downloadable @sc{srec} files.
16932 @item show download-path
16933 @kindex show download-path
16934 Show the default path for downloadable @sc{srec} files.
16936 @item set board-address @var{addr}
16937 @kindex set board-address
16938 @cindex M32-EVA target board address
16939 Set the IP address for the M32R-EVA target board.
16941 @item show board-address
16942 @kindex show board-address
16943 Show the current IP address of the target board.
16945 @item set server-address @var{addr}
16946 @kindex set server-address
16947 @cindex download server address (M32R)
16948 Set the IP address for the download server, which is the @value{GDBN}'s
16951 @item show server-address
16952 @kindex show server-address
16953 Display the IP address of the download server.
16955 @item upload @r{[}@var{file}@r{]}
16956 @kindex upload@r{, M32R}
16957 Upload the specified @sc{srec} @var{file} via the monitor's Ethernet
16958 upload capability. If no @var{file} argument is given, the current
16959 executable file is uploaded.
16961 @item tload @r{[}@var{file}@r{]}
16962 @kindex tload@r{, M32R}
16963 Test the @code{upload} command.
16966 The following commands are available for M32R/SDI:
16971 @cindex reset SDI connection, M32R
16972 This command resets the SDI connection.
16976 This command shows the SDI connection status.
16979 @kindex debug_chaos
16980 @cindex M32R/Chaos debugging
16981 Instructs the remote that M32R/Chaos debugging is to be used.
16983 @item use_debug_dma
16984 @kindex use_debug_dma
16985 Instructs the remote to use the DEBUG_DMA method of accessing memory.
16988 @kindex use_mon_code
16989 Instructs the remote to use the MON_CODE method of accessing memory.
16992 @kindex use_ib_break
16993 Instructs the remote to set breakpoints by IB break.
16995 @item use_dbt_break
16996 @kindex use_dbt_break
16997 Instructs the remote to set breakpoints by DBT.
17003 The Motorola m68k configuration includes ColdFire support, and a
17004 target command for the following ROM monitor.
17008 @kindex target dbug
17009 @item target dbug @var{dev}
17010 dBUG ROM monitor for Motorola ColdFire.
17015 @subsection MicroBlaze
17016 @cindex Xilinx MicroBlaze
17017 @cindex XMD, Xilinx Microprocessor Debugger
17019 The MicroBlaze is a soft-core processor supported on various Xilinx
17020 FPGAs, such as Spartan or Virtex series. Boards with these processors
17021 usually have JTAG ports which connect to a host system running the Xilinx
17022 Embedded Development Kit (EDK) or Software Development Kit (SDK).
17023 This host system is used to download the configuration bitstream to
17024 the target FPGA. The Xilinx Microprocessor Debugger (XMD) program
17025 communicates with the target board using the JTAG interface and
17026 presents a @code{gdbserver} interface to the board. By default
17027 @code{xmd} uses port @code{1234}. (While it is possible to change
17028 this default port, it requires the use of undocumented @code{xmd}
17029 commands. Contact Xilinx support if you need to do this.)
17031 Use these GDB commands to connect to the MicroBlaze target processor.
17034 @item target remote :1234
17035 Use this command to connect to the target if you are running @value{GDBN}
17036 on the same system as @code{xmd}.
17038 @item target remote @var{xmd-host}:1234
17039 Use this command to connect to the target if it is connected to @code{xmd}
17040 running on a different system named @var{xmd-host}.
17043 Use this command to download a program to the MicroBlaze target.
17045 @item set debug microblaze @var{n}
17046 Enable MicroBlaze-specific debugging messages if non-zero.
17048 @item show debug microblaze @var{n}
17049 Show MicroBlaze-specific debugging level.
17052 @node MIPS Embedded
17053 @subsection MIPS Embedded
17055 @cindex MIPS boards
17056 @value{GDBN} can use the MIPS remote debugging protocol to talk to a
17057 MIPS board attached to a serial line. This is available when
17058 you configure @value{GDBN} with @samp{--target=mips-idt-ecoff}.
17061 Use these @value{GDBN} commands to specify the connection to your target board:
17064 @item target mips @var{port}
17065 @kindex target mips @var{port}
17066 To run a program on the board, start up @code{@value{GDBP}} with the
17067 name of your program as the argument. To connect to the board, use the
17068 command @samp{target mips @var{port}}, where @var{port} is the name of
17069 the serial port connected to the board. If the program has not already
17070 been downloaded to the board, you may use the @code{load} command to
17071 download it. You can then use all the usual @value{GDBN} commands.
17073 For example, this sequence connects to the target board through a serial
17074 port, and loads and runs a program called @var{prog} through the
17078 host$ @value{GDBP} @var{prog}
17079 @value{GDBN} is free software and @dots{}
17080 (@value{GDBP}) target mips /dev/ttyb
17081 (@value{GDBP}) load @var{prog}
17085 @item target mips @var{hostname}:@var{portnumber}
17086 On some @value{GDBN} host configurations, you can specify a TCP
17087 connection (for instance, to a serial line managed by a terminal
17088 concentrator) instead of a serial port, using the syntax
17089 @samp{@var{hostname}:@var{portnumber}}.
17091 @item target pmon @var{port}
17092 @kindex target pmon @var{port}
17095 @item target ddb @var{port}
17096 @kindex target ddb @var{port}
17097 NEC's DDB variant of PMON for Vr4300.
17099 @item target lsi @var{port}
17100 @kindex target lsi @var{port}
17101 LSI variant of PMON.
17103 @kindex target r3900
17104 @item target r3900 @var{dev}
17105 Densan DVE-R3900 ROM monitor for Toshiba R3900 Mips.
17107 @kindex target array
17108 @item target array @var{dev}
17109 Array Tech LSI33K RAID controller board.
17115 @value{GDBN} also supports these special commands for MIPS targets:
17118 @item set mipsfpu double
17119 @itemx set mipsfpu single
17120 @itemx set mipsfpu none
17121 @itemx set mipsfpu auto
17122 @itemx show mipsfpu
17123 @kindex set mipsfpu
17124 @kindex show mipsfpu
17125 @cindex MIPS remote floating point
17126 @cindex floating point, MIPS remote
17127 If your target board does not support the MIPS floating point
17128 coprocessor, you should use the command @samp{set mipsfpu none} (if you
17129 need this, you may wish to put the command in your @value{GDBN} init
17130 file). This tells @value{GDBN} how to find the return value of
17131 functions which return floating point values. It also allows
17132 @value{GDBN} to avoid saving the floating point registers when calling
17133 functions on the board. If you are using a floating point coprocessor
17134 with only single precision floating point support, as on the @sc{r4650}
17135 processor, use the command @samp{set mipsfpu single}. The default
17136 double precision floating point coprocessor may be selected using
17137 @samp{set mipsfpu double}.
17139 In previous versions the only choices were double precision or no
17140 floating point, so @samp{set mipsfpu on} will select double precision
17141 and @samp{set mipsfpu off} will select no floating point.
17143 As usual, you can inquire about the @code{mipsfpu} variable with
17144 @samp{show mipsfpu}.
17146 @item set timeout @var{seconds}
17147 @itemx set retransmit-timeout @var{seconds}
17148 @itemx show timeout
17149 @itemx show retransmit-timeout
17150 @cindex @code{timeout}, MIPS protocol
17151 @cindex @code{retransmit-timeout}, MIPS protocol
17152 @kindex set timeout
17153 @kindex show timeout
17154 @kindex set retransmit-timeout
17155 @kindex show retransmit-timeout
17156 You can control the timeout used while waiting for a packet, in the MIPS
17157 remote protocol, with the @code{set timeout @var{seconds}} command. The
17158 default is 5 seconds. Similarly, you can control the timeout used while
17159 waiting for an acknowledgment of a packet with the @code{set
17160 retransmit-timeout @var{seconds}} command. The default is 3 seconds.
17161 You can inspect both values with @code{show timeout} and @code{show
17162 retransmit-timeout}. (These commands are @emph{only} available when
17163 @value{GDBN} is configured for @samp{--target=mips-idt-ecoff}.)
17165 The timeout set by @code{set timeout} does not apply when @value{GDBN}
17166 is waiting for your program to stop. In that case, @value{GDBN} waits
17167 forever because it has no way of knowing how long the program is going
17168 to run before stopping.
17170 @item set syn-garbage-limit @var{num}
17171 @kindex set syn-garbage-limit@r{, MIPS remote}
17172 @cindex synchronize with remote MIPS target
17173 Limit the maximum number of characters @value{GDBN} should ignore when
17174 it tries to synchronize with the remote target. The default is 10
17175 characters. Setting the limit to -1 means there's no limit.
17177 @item show syn-garbage-limit
17178 @kindex show syn-garbage-limit@r{, MIPS remote}
17179 Show the current limit on the number of characters to ignore when
17180 trying to synchronize with the remote system.
17182 @item set monitor-prompt @var{prompt}
17183 @kindex set monitor-prompt@r{, MIPS remote}
17184 @cindex remote monitor prompt
17185 Tell @value{GDBN} to expect the specified @var{prompt} string from the
17186 remote monitor. The default depends on the target:
17196 @item show monitor-prompt
17197 @kindex show monitor-prompt@r{, MIPS remote}
17198 Show the current strings @value{GDBN} expects as the prompt from the
17201 @item set monitor-warnings
17202 @kindex set monitor-warnings@r{, MIPS remote}
17203 Enable or disable monitor warnings about hardware breakpoints. This
17204 has effect only for the @code{lsi} target. When on, @value{GDBN} will
17205 display warning messages whose codes are returned by the @code{lsi}
17206 PMON monitor for breakpoint commands.
17208 @item show monitor-warnings
17209 @kindex show monitor-warnings@r{, MIPS remote}
17210 Show the current setting of printing monitor warnings.
17212 @item pmon @var{command}
17213 @kindex pmon@r{, MIPS remote}
17214 @cindex send PMON command
17215 This command allows sending an arbitrary @var{command} string to the
17216 monitor. The monitor must be in debug mode for this to work.
17219 @node OpenRISC 1000
17220 @subsection OpenRISC 1000
17221 @cindex OpenRISC 1000
17223 @cindex or1k boards
17224 See OR1k Architecture document (@uref{www.opencores.org}) for more information
17225 about platform and commands.
17229 @kindex target jtag
17230 @item target jtag jtag://@var{host}:@var{port}
17232 Connects to remote JTAG server.
17233 JTAG remote server can be either an or1ksim or JTAG server,
17234 connected via parallel port to the board.
17236 Example: @code{target jtag jtag://localhost:9999}
17239 @item or1ksim @var{command}
17240 If connected to @code{or1ksim} OpenRISC 1000 Architectural
17241 Simulator, proprietary commands can be executed.
17243 @kindex info or1k spr
17244 @item info or1k spr
17245 Displays spr groups.
17247 @item info or1k spr @var{group}
17248 @itemx info or1k spr @var{groupno}
17249 Displays register names in selected group.
17251 @item info or1k spr @var{group} @var{register}
17252 @itemx info or1k spr @var{register}
17253 @itemx info or1k spr @var{groupno} @var{registerno}
17254 @itemx info or1k spr @var{registerno}
17255 Shows information about specified spr register.
17258 @item spr @var{group} @var{register} @var{value}
17259 @itemx spr @var{register @var{value}}
17260 @itemx spr @var{groupno} @var{registerno @var{value}}
17261 @itemx spr @var{registerno @var{value}}
17262 Writes @var{value} to specified spr register.
17265 Some implementations of OpenRISC 1000 Architecture also have hardware trace.
17266 It is very similar to @value{GDBN} trace, except it does not interfere with normal
17267 program execution and is thus much faster. Hardware breakpoints/watchpoint
17268 triggers can be set using:
17271 Load effective address/data
17273 Store effective address/data
17275 Access effective address ($SEA or $LEA) or data ($SDATA/$LDATA)
17280 When triggered, it can capture low level data, like: @code{PC}, @code{LSEA},
17281 @code{LDATA}, @code{SDATA}, @code{READSPR}, @code{WRITESPR}, @code{INSTR}.
17283 @code{htrace} commands:
17284 @cindex OpenRISC 1000 htrace
17287 @item hwatch @var{conditional}
17288 Set hardware watchpoint on combination of Load/Store Effective Address(es)
17289 or Data. For example:
17291 @code{hwatch ($LEA == my_var) && ($LDATA < 50) || ($SEA == my_var) && ($SDATA >= 50)}
17293 @code{hwatch ($LEA == my_var) && ($LDATA < 50) || ($SEA == my_var) && ($SDATA >= 50)}
17297 Display information about current HW trace configuration.
17299 @item htrace trigger @var{conditional}
17300 Set starting criteria for HW trace.
17302 @item htrace qualifier @var{conditional}
17303 Set acquisition qualifier for HW trace.
17305 @item htrace stop @var{conditional}
17306 Set HW trace stopping criteria.
17308 @item htrace record [@var{data}]*
17309 Selects the data to be recorded, when qualifier is met and HW trace was
17312 @item htrace enable
17313 @itemx htrace disable
17314 Enables/disables the HW trace.
17316 @item htrace rewind [@var{filename}]
17317 Clears currently recorded trace data.
17319 If filename is specified, new trace file is made and any newly collected data
17320 will be written there.
17322 @item htrace print [@var{start} [@var{len}]]
17323 Prints trace buffer, using current record configuration.
17325 @item htrace mode continuous
17326 Set continuous trace mode.
17328 @item htrace mode suspend
17329 Set suspend trace mode.
17333 @node PowerPC Embedded
17334 @subsection PowerPC Embedded
17336 @value{GDBN} provides the following PowerPC-specific commands:
17339 @kindex set powerpc
17340 @item set powerpc soft-float
17341 @itemx show powerpc soft-float
17342 Force @value{GDBN} to use (or not use) a software floating point calling
17343 convention. By default, @value{GDBN} selects the calling convention based
17344 on the selected architecture and the provided executable file.
17346 @item set powerpc vector-abi
17347 @itemx show powerpc vector-abi
17348 Force @value{GDBN} to use the specified calling convention for vector
17349 arguments and return values. The valid options are @samp{auto};
17350 @samp{generic}, to avoid vector registers even if they are present;
17351 @samp{altivec}, to use AltiVec registers; and @samp{spe} to use SPE
17352 registers. By default, @value{GDBN} selects the calling convention
17353 based on the selected architecture and the provided executable file.
17355 @kindex target dink32
17356 @item target dink32 @var{dev}
17357 DINK32 ROM monitor.
17359 @kindex target ppcbug
17360 @item target ppcbug @var{dev}
17361 @kindex target ppcbug1
17362 @item target ppcbug1 @var{dev}
17363 PPCBUG ROM monitor for PowerPC.
17366 @item target sds @var{dev}
17367 SDS monitor, running on a PowerPC board (such as Motorola's ADS).
17370 @cindex SDS protocol
17371 The following commands specific to the SDS protocol are supported
17375 @item set sdstimeout @var{nsec}
17376 @kindex set sdstimeout
17377 Set the timeout for SDS protocol reads to be @var{nsec} seconds. The
17378 default is 2 seconds.
17380 @item show sdstimeout
17381 @kindex show sdstimeout
17382 Show the current value of the SDS timeout.
17384 @item sds @var{command}
17385 @kindex sds@r{, a command}
17386 Send the specified @var{command} string to the SDS monitor.
17391 @subsection HP PA Embedded
17395 @kindex target op50n
17396 @item target op50n @var{dev}
17397 OP50N monitor, running on an OKI HPPA board.
17399 @kindex target w89k
17400 @item target w89k @var{dev}
17401 W89K monitor, running on a Winbond HPPA board.
17406 @subsection Tsqware Sparclet
17410 @value{GDBN} enables developers to debug tasks running on
17411 Sparclet targets from a Unix host.
17412 @value{GDBN} uses code that runs on
17413 both the Unix host and on the Sparclet target. The program
17414 @code{@value{GDBP}} is installed and executed on the Unix host.
17417 @item remotetimeout @var{args}
17418 @kindex remotetimeout
17419 @value{GDBN} supports the option @code{remotetimeout}.
17420 This option is set by the user, and @var{args} represents the number of
17421 seconds @value{GDBN} waits for responses.
17424 @cindex compiling, on Sparclet
17425 When compiling for debugging, include the options @samp{-g} to get debug
17426 information and @samp{-Ttext} to relocate the program to where you wish to
17427 load it on the target. You may also want to add the options @samp{-n} or
17428 @samp{-N} in order to reduce the size of the sections. Example:
17431 sparclet-aout-gcc prog.c -Ttext 0x12010000 -g -o prog -N
17434 You can use @code{objdump} to verify that the addresses are what you intended:
17437 sparclet-aout-objdump --headers --syms prog
17440 @cindex running, on Sparclet
17442 your Unix execution search path to find @value{GDBN}, you are ready to
17443 run @value{GDBN}. From your Unix host, run @code{@value{GDBP}}
17444 (or @code{sparclet-aout-gdb}, depending on your installation).
17446 @value{GDBN} comes up showing the prompt:
17453 * Sparclet File:: Setting the file to debug
17454 * Sparclet Connection:: Connecting to Sparclet
17455 * Sparclet Download:: Sparclet download
17456 * Sparclet Execution:: Running and debugging
17459 @node Sparclet File
17460 @subsubsection Setting File to Debug
17462 The @value{GDBN} command @code{file} lets you choose with program to debug.
17465 (gdbslet) file prog
17469 @value{GDBN} then attempts to read the symbol table of @file{prog}.
17470 @value{GDBN} locates
17471 the file by searching the directories listed in the command search
17473 If the file was compiled with debug information (option @samp{-g}), source
17474 files will be searched as well.
17475 @value{GDBN} locates
17476 the source files by searching the directories listed in the directory search
17477 path (@pxref{Environment, ,Your Program's Environment}).
17479 to find a file, it displays a message such as:
17482 prog: No such file or directory.
17485 When this happens, add the appropriate directories to the search paths with
17486 the @value{GDBN} commands @code{path} and @code{dir}, and execute the
17487 @code{target} command again.
17489 @node Sparclet Connection
17490 @subsubsection Connecting to Sparclet
17492 The @value{GDBN} command @code{target} lets you connect to a Sparclet target.
17493 To connect to a target on serial port ``@code{ttya}'', type:
17496 (gdbslet) target sparclet /dev/ttya
17497 Remote target sparclet connected to /dev/ttya
17498 main () at ../prog.c:3
17502 @value{GDBN} displays messages like these:
17508 @node Sparclet Download
17509 @subsubsection Sparclet Download
17511 @cindex download to Sparclet
17512 Once connected to the Sparclet target,
17513 you can use the @value{GDBN}
17514 @code{load} command to download the file from the host to the target.
17515 The file name and load offset should be given as arguments to the @code{load}
17517 Since the file format is aout, the program must be loaded to the starting
17518 address. You can use @code{objdump} to find out what this value is. The load
17519 offset is an offset which is added to the VMA (virtual memory address)
17520 of each of the file's sections.
17521 For instance, if the program
17522 @file{prog} was linked to text address 0x1201000, with data at 0x12010160
17523 and bss at 0x12010170, in @value{GDBN}, type:
17526 (gdbslet) load prog 0x12010000
17527 Loading section .text, size 0xdb0 vma 0x12010000
17530 If the code is loaded at a different address then what the program was linked
17531 to, you may need to use the @code{section} and @code{add-symbol-file} commands
17532 to tell @value{GDBN} where to map the symbol table.
17534 @node Sparclet Execution
17535 @subsubsection Running and Debugging
17537 @cindex running and debugging Sparclet programs
17538 You can now begin debugging the task using @value{GDBN}'s execution control
17539 commands, @code{b}, @code{step}, @code{run}, etc. See the @value{GDBN}
17540 manual for the list of commands.
17544 Breakpoint 1 at 0x12010000: file prog.c, line 3.
17546 Starting program: prog
17547 Breakpoint 1, main (argc=1, argv=0xeffff21c) at prog.c:3
17548 3 char *symarg = 0;
17550 4 char *execarg = "hello!";
17555 @subsection Fujitsu Sparclite
17559 @kindex target sparclite
17560 @item target sparclite @var{dev}
17561 Fujitsu sparclite boards, used only for the purpose of loading.
17562 You must use an additional command to debug the program.
17563 For example: target remote @var{dev} using @value{GDBN} standard
17569 @subsection Zilog Z8000
17572 @cindex simulator, Z8000
17573 @cindex Zilog Z8000 simulator
17575 When configured for debugging Zilog Z8000 targets, @value{GDBN} includes
17578 For the Z8000 family, @samp{target sim} simulates either the Z8002 (the
17579 unsegmented variant of the Z8000 architecture) or the Z8001 (the
17580 segmented variant). The simulator recognizes which architecture is
17581 appropriate by inspecting the object code.
17584 @item target sim @var{args}
17586 @kindex target sim@r{, with Z8000}
17587 Debug programs on a simulated CPU. If the simulator supports setup
17588 options, specify them via @var{args}.
17592 After specifying this target, you can debug programs for the simulated
17593 CPU in the same style as programs for your host computer; use the
17594 @code{file} command to load a new program image, the @code{run} command
17595 to run your program, and so on.
17597 As well as making available all the usual machine registers
17598 (@pxref{Registers, ,Registers}), the Z8000 simulator provides three
17599 additional items of information as specially named registers:
17604 Counts clock-ticks in the simulator.
17607 Counts instructions run in the simulator.
17610 Execution time in 60ths of a second.
17614 You can refer to these values in @value{GDBN} expressions with the usual
17615 conventions; for example, @w{@samp{b fputc if $cycles>5000}} sets a
17616 conditional breakpoint that suspends only after at least 5000
17617 simulated clock ticks.
17620 @subsection Atmel AVR
17623 When configured for debugging the Atmel AVR, @value{GDBN} supports the
17624 following AVR-specific commands:
17627 @item info io_registers
17628 @kindex info io_registers@r{, AVR}
17629 @cindex I/O registers (Atmel AVR)
17630 This command displays information about the AVR I/O registers. For
17631 each register, @value{GDBN} prints its number and value.
17638 When configured for debugging CRIS, @value{GDBN} provides the
17639 following CRIS-specific commands:
17642 @item set cris-version @var{ver}
17643 @cindex CRIS version
17644 Set the current CRIS version to @var{ver}, either @samp{10} or @samp{32}.
17645 The CRIS version affects register names and sizes. This command is useful in
17646 case autodetection of the CRIS version fails.
17648 @item show cris-version
17649 Show the current CRIS version.
17651 @item set cris-dwarf2-cfi
17652 @cindex DWARF-2 CFI and CRIS
17653 Set the usage of DWARF-2 CFI for CRIS debugging. The default is @samp{on}.
17654 Change to @samp{off} when using @code{gcc-cris} whose version is below
17657 @item show cris-dwarf2-cfi
17658 Show the current state of using DWARF-2 CFI.
17660 @item set cris-mode @var{mode}
17662 Set the current CRIS mode to @var{mode}. It should only be changed when
17663 debugging in guru mode, in which case it should be set to
17664 @samp{guru} (the default is @samp{normal}).
17666 @item show cris-mode
17667 Show the current CRIS mode.
17671 @subsection Renesas Super-H
17674 For the Renesas Super-H processor, @value{GDBN} provides these
17679 @kindex regs@r{, Super-H}
17680 Show the values of all Super-H registers.
17682 @item set sh calling-convention @var{convention}
17683 @kindex set sh calling-convention
17684 Set the calling-convention used when calling functions from @value{GDBN}.
17685 Allowed values are @samp{gcc}, which is the default setting, and @samp{renesas}.
17686 With the @samp{gcc} setting, functions are called using the @value{NGCC} calling
17687 convention. If the DWARF-2 information of the called function specifies
17688 that the function follows the Renesas calling convention, the function
17689 is called using the Renesas calling convention. If the calling convention
17690 is set to @samp{renesas}, the Renesas calling convention is always used,
17691 regardless of the DWARF-2 information. This can be used to override the
17692 default of @samp{gcc} if debug information is missing, or the compiler
17693 does not emit the DWARF-2 calling convention entry for a function.
17695 @item show sh calling-convention
17696 @kindex show sh calling-convention
17697 Show the current calling convention setting.
17702 @node Architectures
17703 @section Architectures
17705 This section describes characteristics of architectures that affect
17706 all uses of @value{GDBN} with the architecture, both native and cross.
17713 * HPPA:: HP PA architecture
17714 * SPU:: Cell Broadband Engine SPU architecture
17719 @subsection x86 Architecture-specific Issues
17722 @item set struct-convention @var{mode}
17723 @kindex set struct-convention
17724 @cindex struct return convention
17725 @cindex struct/union returned in registers
17726 Set the convention used by the inferior to return @code{struct}s and
17727 @code{union}s from functions to @var{mode}. Possible values of
17728 @var{mode} are @code{"pcc"}, @code{"reg"}, and @code{"default"} (the
17729 default). @code{"default"} or @code{"pcc"} means that @code{struct}s
17730 are returned on the stack, while @code{"reg"} means that a
17731 @code{struct} or a @code{union} whose size is 1, 2, 4, or 8 bytes will
17732 be returned in a register.
17734 @item show struct-convention
17735 @kindex show struct-convention
17736 Show the current setting of the convention to return @code{struct}s
17745 @kindex set rstack_high_address
17746 @cindex AMD 29K register stack
17747 @cindex register stack, AMD29K
17748 @item set rstack_high_address @var{address}
17749 On AMD 29000 family processors, registers are saved in a separate
17750 @dfn{register stack}. There is no way for @value{GDBN} to determine the
17751 extent of this stack. Normally, @value{GDBN} just assumes that the
17752 stack is ``large enough''. This may result in @value{GDBN} referencing
17753 memory locations that do not exist. If necessary, you can get around
17754 this problem by specifying the ending address of the register stack with
17755 the @code{set rstack_high_address} command. The argument should be an
17756 address, which you probably want to precede with @samp{0x} to specify in
17759 @kindex show rstack_high_address
17760 @item show rstack_high_address
17761 Display the current limit of the register stack, on AMD 29000 family
17769 See the following section.
17774 @cindex stack on Alpha
17775 @cindex stack on MIPS
17776 @cindex Alpha stack
17778 Alpha- and MIPS-based computers use an unusual stack frame, which
17779 sometimes requires @value{GDBN} to search backward in the object code to
17780 find the beginning of a function.
17782 @cindex response time, MIPS debugging
17783 To improve response time (especially for embedded applications, where
17784 @value{GDBN} may be restricted to a slow serial line for this search)
17785 you may want to limit the size of this search, using one of these
17789 @cindex @code{heuristic-fence-post} (Alpha, MIPS)
17790 @item set heuristic-fence-post @var{limit}
17791 Restrict @value{GDBN} to examining at most @var{limit} bytes in its
17792 search for the beginning of a function. A value of @var{0} (the
17793 default) means there is no limit. However, except for @var{0}, the
17794 larger the limit the more bytes @code{heuristic-fence-post} must search
17795 and therefore the longer it takes to run. You should only need to use
17796 this command when debugging a stripped executable.
17798 @item show heuristic-fence-post
17799 Display the current limit.
17803 These commands are available @emph{only} when @value{GDBN} is configured
17804 for debugging programs on Alpha or MIPS processors.
17806 Several MIPS-specific commands are available when debugging MIPS
17810 @item set mips abi @var{arg}
17811 @kindex set mips abi
17812 @cindex set ABI for MIPS
17813 Tell @value{GDBN} which MIPS ABI is used by the inferior. Possible
17814 values of @var{arg} are:
17818 The default ABI associated with the current binary (this is the
17829 @item show mips abi
17830 @kindex show mips abi
17831 Show the MIPS ABI used by @value{GDBN} to debug the inferior.
17834 @itemx show mipsfpu
17835 @xref{MIPS Embedded, set mipsfpu}.
17837 @item set mips mask-address @var{arg}
17838 @kindex set mips mask-address
17839 @cindex MIPS addresses, masking
17840 This command determines whether the most-significant 32 bits of 64-bit
17841 MIPS addresses are masked off. The argument @var{arg} can be
17842 @samp{on}, @samp{off}, or @samp{auto}. The latter is the default
17843 setting, which lets @value{GDBN} determine the correct value.
17845 @item show mips mask-address
17846 @kindex show mips mask-address
17847 Show whether the upper 32 bits of MIPS addresses are masked off or
17850 @item set remote-mips64-transfers-32bit-regs
17851 @kindex set remote-mips64-transfers-32bit-regs
17852 This command controls compatibility with 64-bit MIPS targets that
17853 transfer data in 32-bit quantities. If you have an old MIPS 64 target
17854 that transfers 32 bits for some registers, like @sc{sr} and @sc{fsr},
17855 and 64 bits for other registers, set this option to @samp{on}.
17857 @item show remote-mips64-transfers-32bit-regs
17858 @kindex show remote-mips64-transfers-32bit-regs
17859 Show the current setting of compatibility with older MIPS 64 targets.
17861 @item set debug mips
17862 @kindex set debug mips
17863 This command turns on and off debugging messages for the MIPS-specific
17864 target code in @value{GDBN}.
17866 @item show debug mips
17867 @kindex show debug mips
17868 Show the current setting of MIPS debugging messages.
17874 @cindex HPPA support
17876 When @value{GDBN} is debugging the HP PA architecture, it provides the
17877 following special commands:
17880 @item set debug hppa
17881 @kindex set debug hppa
17882 This command determines whether HPPA architecture-specific debugging
17883 messages are to be displayed.
17885 @item show debug hppa
17886 Show whether HPPA debugging messages are displayed.
17888 @item maint print unwind @var{address}
17889 @kindex maint print unwind@r{, HPPA}
17890 This command displays the contents of the unwind table entry at the
17891 given @var{address}.
17897 @subsection Cell Broadband Engine SPU architecture
17898 @cindex Cell Broadband Engine
17901 When @value{GDBN} is debugging the Cell Broadband Engine SPU architecture,
17902 it provides the following special commands:
17905 @item info spu event
17907 Display SPU event facility status. Shows current event mask
17908 and pending event status.
17910 @item info spu signal
17911 Display SPU signal notification facility status. Shows pending
17912 signal-control word and signal notification mode of both signal
17913 notification channels.
17915 @item info spu mailbox
17916 Display SPU mailbox facility status. Shows all pending entries,
17917 in order of processing, in each of the SPU Write Outbound,
17918 SPU Write Outbound Interrupt, and SPU Read Inbound mailboxes.
17921 Display MFC DMA status. Shows all pending commands in the MFC
17922 DMA queue. For each entry, opcode, tag, class IDs, effective
17923 and local store addresses and transfer size are shown.
17925 @item info spu proxydma
17926 Display MFC Proxy-DMA status. Shows all pending commands in the MFC
17927 Proxy-DMA queue. For each entry, opcode, tag, class IDs, effective
17928 and local store addresses and transfer size are shown.
17932 When @value{GDBN} is debugging a combined PowerPC/SPU application
17933 on the Cell Broadband Engine, it provides in addition the following
17937 @item set spu stop-on-load @var{arg}
17939 Set whether to stop for new SPE threads. When set to @code{on}, @value{GDBN}
17940 will give control to the user when a new SPE thread enters its @code{main}
17941 function. The default is @code{off}.
17943 @item show spu stop-on-load
17945 Show whether to stop for new SPE threads.
17947 @item set spu auto-flush-cache @var{arg}
17948 Set whether to automatically flush the software-managed cache. When set to
17949 @code{on}, @value{GDBN} will automatically cause the SPE software-managed
17950 cache to be flushed whenever SPE execution stops. This provides a consistent
17951 view of PowerPC memory that is accessed via the cache. If an application
17952 does not use the software-managed cache, this option has no effect.
17954 @item show spu auto-flush-cache
17955 Show whether to automatically flush the software-managed cache.
17960 @subsection PowerPC
17961 @cindex PowerPC architecture
17963 When @value{GDBN} is debugging the PowerPC architecture, it provides a set of
17964 pseudo-registers to enable inspection of 128-bit wide Decimal Floating Point
17965 numbers stored in the floating point registers. These values must be stored
17966 in two consecutive registers, always starting at an even register like
17967 @code{f0} or @code{f2}.
17969 The pseudo-registers go from @code{$dl0} through @code{$dl15}, and are formed
17970 by joining the even/odd register pairs @code{f0} and @code{f1} for @code{$dl0},
17971 @code{f2} and @code{f3} for @code{$dl1} and so on.
17973 For POWER7 processors, @value{GDBN} provides a set of pseudo-registers, the 64-bit
17974 wide Extended Floating Point Registers (@samp{f32} through @samp{f63}).
17977 @node Controlling GDB
17978 @chapter Controlling @value{GDBN}
17980 You can alter the way @value{GDBN} interacts with you by using the
17981 @code{set} command. For commands controlling how @value{GDBN} displays
17982 data, see @ref{Print Settings, ,Print Settings}. Other settings are
17987 * Editing:: Command editing
17988 * Command History:: Command history
17989 * Screen Size:: Screen size
17990 * Numbers:: Numbers
17991 * ABI:: Configuring the current ABI
17992 * Messages/Warnings:: Optional warnings and messages
17993 * Debugging Output:: Optional messages about internal happenings
17994 * Other Misc Settings:: Other Miscellaneous Settings
18002 @value{GDBN} indicates its readiness to read a command by printing a string
18003 called the @dfn{prompt}. This string is normally @samp{(@value{GDBP})}. You
18004 can change the prompt string with the @code{set prompt} command. For
18005 instance, when debugging @value{GDBN} with @value{GDBN}, it is useful to change
18006 the prompt in one of the @value{GDBN} sessions so that you can always tell
18007 which one you are talking to.
18009 @emph{Note:} @code{set prompt} does not add a space for you after the
18010 prompt you set. This allows you to set a prompt which ends in a space
18011 or a prompt that does not.
18015 @item set prompt @var{newprompt}
18016 Directs @value{GDBN} to use @var{newprompt} as its prompt string henceforth.
18018 @kindex show prompt
18020 Prints a line of the form: @samp{Gdb's prompt is: @var{your-prompt}}
18024 @section Command Editing
18026 @cindex command line editing
18028 @value{GDBN} reads its input commands via the @dfn{Readline} interface. This
18029 @sc{gnu} library provides consistent behavior for programs which provide a
18030 command line interface to the user. Advantages are @sc{gnu} Emacs-style
18031 or @dfn{vi}-style inline editing of commands, @code{csh}-like history
18032 substitution, and a storage and recall of command history across
18033 debugging sessions.
18035 You may control the behavior of command line editing in @value{GDBN} with the
18036 command @code{set}.
18039 @kindex set editing
18042 @itemx set editing on
18043 Enable command line editing (enabled by default).
18045 @item set editing off
18046 Disable command line editing.
18048 @kindex show editing
18050 Show whether command line editing is enabled.
18053 @xref{Command Line Editing}, for more details about the Readline
18054 interface. Users unfamiliar with @sc{gnu} Emacs or @code{vi} are
18055 encouraged to read that chapter.
18057 @node Command History
18058 @section Command History
18059 @cindex command history
18061 @value{GDBN} can keep track of the commands you type during your
18062 debugging sessions, so that you can be certain of precisely what
18063 happened. Use these commands to manage the @value{GDBN} command
18066 @value{GDBN} uses the @sc{gnu} History library, a part of the Readline
18067 package, to provide the history facility. @xref{Using History
18068 Interactively}, for the detailed description of the History library.
18070 To issue a command to @value{GDBN} without affecting certain aspects of
18071 the state which is seen by users, prefix it with @samp{server }
18072 (@pxref{Server Prefix}). This
18073 means that this command will not affect the command history, nor will it
18074 affect @value{GDBN}'s notion of which command to repeat if @key{RET} is
18075 pressed on a line by itself.
18077 @cindex @code{server}, command prefix
18078 The server prefix does not affect the recording of values into the value
18079 history; to print a value without recording it into the value history,
18080 use the @code{output} command instead of the @code{print} command.
18082 Here is the description of @value{GDBN} commands related to command
18086 @cindex history substitution
18087 @cindex history file
18088 @kindex set history filename
18089 @cindex @env{GDBHISTFILE}, environment variable
18090 @item set history filename @var{fname}
18091 Set the name of the @value{GDBN} command history file to @var{fname}.
18092 This is the file where @value{GDBN} reads an initial command history
18093 list, and where it writes the command history from this session when it
18094 exits. You can access this list through history expansion or through
18095 the history command editing characters listed below. This file defaults
18096 to the value of the environment variable @code{GDBHISTFILE}, or to
18097 @file{./.gdb_history} (@file{./_gdb_history} on MS-DOS) if this variable
18100 @cindex save command history
18101 @kindex set history save
18102 @item set history save
18103 @itemx set history save on
18104 Record command history in a file, whose name may be specified with the
18105 @code{set history filename} command. By default, this option is disabled.
18107 @item set history save off
18108 Stop recording command history in a file.
18110 @cindex history size
18111 @kindex set history size
18112 @cindex @env{HISTSIZE}, environment variable
18113 @item set history size @var{size}
18114 Set the number of commands which @value{GDBN} keeps in its history list.
18115 This defaults to the value of the environment variable
18116 @code{HISTSIZE}, or to 256 if this variable is not set.
18119 History expansion assigns special meaning to the character @kbd{!}.
18120 @xref{Event Designators}, for more details.
18122 @cindex history expansion, turn on/off
18123 Since @kbd{!} is also the logical not operator in C, history expansion
18124 is off by default. If you decide to enable history expansion with the
18125 @code{set history expansion on} command, you may sometimes need to
18126 follow @kbd{!} (when it is used as logical not, in an expression) with
18127 a space or a tab to prevent it from being expanded. The readline
18128 history facilities do not attempt substitution on the strings
18129 @kbd{!=} and @kbd{!(}, even when history expansion is enabled.
18131 The commands to control history expansion are:
18134 @item set history expansion on
18135 @itemx set history expansion
18136 @kindex set history expansion
18137 Enable history expansion. History expansion is off by default.
18139 @item set history expansion off
18140 Disable history expansion.
18143 @kindex show history
18145 @itemx show history filename
18146 @itemx show history save
18147 @itemx show history size
18148 @itemx show history expansion
18149 These commands display the state of the @value{GDBN} history parameters.
18150 @code{show history} by itself displays all four states.
18155 @kindex show commands
18156 @cindex show last commands
18157 @cindex display command history
18158 @item show commands
18159 Display the last ten commands in the command history.
18161 @item show commands @var{n}
18162 Print ten commands centered on command number @var{n}.
18164 @item show commands +
18165 Print ten commands just after the commands last printed.
18169 @section Screen Size
18170 @cindex size of screen
18171 @cindex pauses in output
18173 Certain commands to @value{GDBN} may produce large amounts of
18174 information output to the screen. To help you read all of it,
18175 @value{GDBN} pauses and asks you for input at the end of each page of
18176 output. Type @key{RET} when you want to continue the output, or @kbd{q}
18177 to discard the remaining output. Also, the screen width setting
18178 determines when to wrap lines of output. Depending on what is being
18179 printed, @value{GDBN} tries to break the line at a readable place,
18180 rather than simply letting it overflow onto the following line.
18182 Normally @value{GDBN} knows the size of the screen from the terminal
18183 driver software. For example, on Unix @value{GDBN} uses the termcap data base
18184 together with the value of the @code{TERM} environment variable and the
18185 @code{stty rows} and @code{stty cols} settings. If this is not correct,
18186 you can override it with the @code{set height} and @code{set
18193 @kindex show height
18194 @item set height @var{lpp}
18196 @itemx set width @var{cpl}
18198 These @code{set} commands specify a screen height of @var{lpp} lines and
18199 a screen width of @var{cpl} characters. The associated @code{show}
18200 commands display the current settings.
18202 If you specify a height of zero lines, @value{GDBN} does not pause during
18203 output no matter how long the output is. This is useful if output is to a
18204 file or to an editor buffer.
18206 Likewise, you can specify @samp{set width 0} to prevent @value{GDBN}
18207 from wrapping its output.
18209 @item set pagination on
18210 @itemx set pagination off
18211 @kindex set pagination
18212 Turn the output pagination on or off; the default is on. Turning
18213 pagination off is the alternative to @code{set height 0}.
18215 @item show pagination
18216 @kindex show pagination
18217 Show the current pagination mode.
18222 @cindex number representation
18223 @cindex entering numbers
18225 You can always enter numbers in octal, decimal, or hexadecimal in
18226 @value{GDBN} by the usual conventions: octal numbers begin with
18227 @samp{0}, decimal numbers end with @samp{.}, and hexadecimal numbers
18228 begin with @samp{0x}. Numbers that neither begin with @samp{0} or
18229 @samp{0x}, nor end with a @samp{.} are, by default, entered in base
18230 10; likewise, the default display for numbers---when no particular
18231 format is specified---is base 10. You can change the default base for
18232 both input and output with the commands described below.
18235 @kindex set input-radix
18236 @item set input-radix @var{base}
18237 Set the default base for numeric input. Supported choices
18238 for @var{base} are decimal 8, 10, or 16. @var{base} must itself be
18239 specified either unambiguously or using the current input radix; for
18243 set input-radix 012
18244 set input-radix 10.
18245 set input-radix 0xa
18249 sets the input base to decimal. On the other hand, @samp{set input-radix 10}
18250 leaves the input radix unchanged, no matter what it was, since
18251 @samp{10}, being without any leading or trailing signs of its base, is
18252 interpreted in the current radix. Thus, if the current radix is 16,
18253 @samp{10} is interpreted in hex, i.e.@: as 16 decimal, which doesn't
18256 @kindex set output-radix
18257 @item set output-radix @var{base}
18258 Set the default base for numeric display. Supported choices
18259 for @var{base} are decimal 8, 10, or 16. @var{base} must itself be
18260 specified either unambiguously or using the current input radix.
18262 @kindex show input-radix
18263 @item show input-radix
18264 Display the current default base for numeric input.
18266 @kindex show output-radix
18267 @item show output-radix
18268 Display the current default base for numeric display.
18270 @item set radix @r{[}@var{base}@r{]}
18274 These commands set and show the default base for both input and output
18275 of numbers. @code{set radix} sets the radix of input and output to
18276 the same base; without an argument, it resets the radix back to its
18277 default value of 10.
18282 @section Configuring the Current ABI
18284 @value{GDBN} can determine the @dfn{ABI} (Application Binary Interface) of your
18285 application automatically. However, sometimes you need to override its
18286 conclusions. Use these commands to manage @value{GDBN}'s view of the
18293 One @value{GDBN} configuration can debug binaries for multiple operating
18294 system targets, either via remote debugging or native emulation.
18295 @value{GDBN} will autodetect the @dfn{OS ABI} (Operating System ABI) in use,
18296 but you can override its conclusion using the @code{set osabi} command.
18297 One example where this is useful is in debugging of binaries which use
18298 an alternate C library (e.g.@: @sc{uClibc} for @sc{gnu}/Linux) which does
18299 not have the same identifying marks that the standard C library for your
18304 Show the OS ABI currently in use.
18307 With no argument, show the list of registered available OS ABI's.
18309 @item set osabi @var{abi}
18310 Set the current OS ABI to @var{abi}.
18313 @cindex float promotion
18315 Generally, the way that an argument of type @code{float} is passed to a
18316 function depends on whether the function is prototyped. For a prototyped
18317 (i.e.@: ANSI/ISO style) function, @code{float} arguments are passed unchanged,
18318 according to the architecture's convention for @code{float}. For unprototyped
18319 (i.e.@: K&R style) functions, @code{float} arguments are first promoted to type
18320 @code{double} and then passed.
18322 Unfortunately, some forms of debug information do not reliably indicate whether
18323 a function is prototyped. If @value{GDBN} calls a function that is not marked
18324 as prototyped, it consults @kbd{set coerce-float-to-double}.
18327 @kindex set coerce-float-to-double
18328 @item set coerce-float-to-double
18329 @itemx set coerce-float-to-double on
18330 Arguments of type @code{float} will be promoted to @code{double} when passed
18331 to an unprototyped function. This is the default setting.
18333 @item set coerce-float-to-double off
18334 Arguments of type @code{float} will be passed directly to unprototyped
18337 @kindex show coerce-float-to-double
18338 @item show coerce-float-to-double
18339 Show the current setting of promoting @code{float} to @code{double}.
18343 @kindex show cp-abi
18344 @value{GDBN} needs to know the ABI used for your program's C@t{++}
18345 objects. The correct C@t{++} ABI depends on which C@t{++} compiler was
18346 used to build your application. @value{GDBN} only fully supports
18347 programs with a single C@t{++} ABI; if your program contains code using
18348 multiple C@t{++} ABI's or if @value{GDBN} can not identify your
18349 program's ABI correctly, you can tell @value{GDBN} which ABI to use.
18350 Currently supported ABI's include ``gnu-v2'', for @code{g++} versions
18351 before 3.0, ``gnu-v3'', for @code{g++} versions 3.0 and later, and
18352 ``hpaCC'' for the HP ANSI C@t{++} compiler. Other C@t{++} compilers may
18353 use the ``gnu-v2'' or ``gnu-v3'' ABI's as well. The default setting is
18358 Show the C@t{++} ABI currently in use.
18361 With no argument, show the list of supported C@t{++} ABI's.
18363 @item set cp-abi @var{abi}
18364 @itemx set cp-abi auto
18365 Set the current C@t{++} ABI to @var{abi}, or return to automatic detection.
18368 @node Messages/Warnings
18369 @section Optional Warnings and Messages
18371 @cindex verbose operation
18372 @cindex optional warnings
18373 By default, @value{GDBN} is silent about its inner workings. If you are
18374 running on a slow machine, you may want to use the @code{set verbose}
18375 command. This makes @value{GDBN} tell you when it does a lengthy
18376 internal operation, so you will not think it has crashed.
18378 Currently, the messages controlled by @code{set verbose} are those
18379 which announce that the symbol table for a source file is being read;
18380 see @code{symbol-file} in @ref{Files, ,Commands to Specify Files}.
18383 @kindex set verbose
18384 @item set verbose on
18385 Enables @value{GDBN} output of certain informational messages.
18387 @item set verbose off
18388 Disables @value{GDBN} output of certain informational messages.
18390 @kindex show verbose
18392 Displays whether @code{set verbose} is on or off.
18395 By default, if @value{GDBN} encounters bugs in the symbol table of an
18396 object file, it is silent; but if you are debugging a compiler, you may
18397 find this information useful (@pxref{Symbol Errors, ,Errors Reading
18402 @kindex set complaints
18403 @item set complaints @var{limit}
18404 Permits @value{GDBN} to output @var{limit} complaints about each type of
18405 unusual symbols before becoming silent about the problem. Set
18406 @var{limit} to zero to suppress all complaints; set it to a large number
18407 to prevent complaints from being suppressed.
18409 @kindex show complaints
18410 @item show complaints
18411 Displays how many symbol complaints @value{GDBN} is permitted to produce.
18415 @anchor{confirmation requests}
18416 By default, @value{GDBN} is cautious, and asks what sometimes seems to be a
18417 lot of stupid questions to confirm certain commands. For example, if
18418 you try to run a program which is already running:
18422 The program being debugged has been started already.
18423 Start it from the beginning? (y or n)
18426 If you are willing to unflinchingly face the consequences of your own
18427 commands, you can disable this ``feature'':
18431 @kindex set confirm
18433 @cindex confirmation
18434 @cindex stupid questions
18435 @item set confirm off
18436 Disables confirmation requests.
18438 @item set confirm on
18439 Enables confirmation requests (the default).
18441 @kindex show confirm
18443 Displays state of confirmation requests.
18447 @cindex command tracing
18448 If you need to debug user-defined commands or sourced files you may find it
18449 useful to enable @dfn{command tracing}. In this mode each command will be
18450 printed as it is executed, prefixed with one or more @samp{+} symbols, the
18451 quantity denoting the call depth of each command.
18454 @kindex set trace-commands
18455 @cindex command scripts, debugging
18456 @item set trace-commands on
18457 Enable command tracing.
18458 @item set trace-commands off
18459 Disable command tracing.
18460 @item show trace-commands
18461 Display the current state of command tracing.
18464 @node Debugging Output
18465 @section Optional Messages about Internal Happenings
18466 @cindex optional debugging messages
18468 @value{GDBN} has commands that enable optional debugging messages from
18469 various @value{GDBN} subsystems; normally these commands are of
18470 interest to @value{GDBN} maintainers, or when reporting a bug. This
18471 section documents those commands.
18474 @kindex set exec-done-display
18475 @item set exec-done-display
18476 Turns on or off the notification of asynchronous commands'
18477 completion. When on, @value{GDBN} will print a message when an
18478 asynchronous command finishes its execution. The default is off.
18479 @kindex show exec-done-display
18480 @item show exec-done-display
18481 Displays the current setting of asynchronous command completion
18484 @cindex gdbarch debugging info
18485 @cindex architecture debugging info
18486 @item set debug arch
18487 Turns on or off display of gdbarch debugging info. The default is off
18489 @item show debug arch
18490 Displays the current state of displaying gdbarch debugging info.
18491 @item set debug aix-thread
18492 @cindex AIX threads
18493 Display debugging messages about inner workings of the AIX thread
18495 @item show debug aix-thread
18496 Show the current state of AIX thread debugging info display.
18497 @item set debug dwarf2-die
18498 @cindex DWARF2 DIEs
18499 Dump DWARF2 DIEs after they are read in.
18500 The value is the number of nesting levels to print.
18501 A value of zero turns off the display.
18502 @item show debug dwarf2-die
18503 Show the current state of DWARF2 DIE debugging.
18504 @item set debug displaced
18505 @cindex displaced stepping debugging info
18506 Turns on or off display of @value{GDBN} debugging info for the
18507 displaced stepping support. The default is off.
18508 @item show debug displaced
18509 Displays the current state of displaying @value{GDBN} debugging info
18510 related to displaced stepping.
18511 @item set debug event
18512 @cindex event debugging info
18513 Turns on or off display of @value{GDBN} event debugging info. The
18515 @item show debug event
18516 Displays the current state of displaying @value{GDBN} event debugging
18518 @item set debug expression
18519 @cindex expression debugging info
18520 Turns on or off display of debugging info about @value{GDBN}
18521 expression parsing. The default is off.
18522 @item show debug expression
18523 Displays the current state of displaying debugging info about
18524 @value{GDBN} expression parsing.
18525 @item set debug frame
18526 @cindex frame debugging info
18527 Turns on or off display of @value{GDBN} frame debugging info. The
18529 @item show debug frame
18530 Displays the current state of displaying @value{GDBN} frame debugging
18532 @item set debug gnu-nat
18533 @cindex @sc{gnu}/Hurd debug messages
18534 Turns on or off debugging messages from the @sc{gnu}/Hurd debug support.
18535 @item show debug gnu-nat
18536 Show the current state of @sc{gnu}/Hurd debugging messages.
18537 @item set debug infrun
18538 @cindex inferior debugging info
18539 Turns on or off display of @value{GDBN} debugging info for running the inferior.
18540 The default is off. @file{infrun.c} contains GDB's runtime state machine used
18541 for implementing operations such as single-stepping the inferior.
18542 @item show debug infrun
18543 Displays the current state of @value{GDBN} inferior debugging.
18544 @item set debug lin-lwp
18545 @cindex @sc{gnu}/Linux LWP debug messages
18546 @cindex Linux lightweight processes
18547 Turns on or off debugging messages from the Linux LWP debug support.
18548 @item show debug lin-lwp
18549 Show the current state of Linux LWP debugging messages.
18550 @item set debug lin-lwp-async
18551 @cindex @sc{gnu}/Linux LWP async debug messages
18552 @cindex Linux lightweight processes
18553 Turns on or off debugging messages from the Linux LWP async debug support.
18554 @item show debug lin-lwp-async
18555 Show the current state of Linux LWP async debugging messages.
18556 @item set debug observer
18557 @cindex observer debugging info
18558 Turns on or off display of @value{GDBN} observer debugging. This
18559 includes info such as the notification of observable events.
18560 @item show debug observer
18561 Displays the current state of observer debugging.
18562 @item set debug overload
18563 @cindex C@t{++} overload debugging info
18564 Turns on or off display of @value{GDBN} C@t{++} overload debugging
18565 info. This includes info such as ranking of functions, etc. The default
18567 @item show debug overload
18568 Displays the current state of displaying @value{GDBN} C@t{++} overload
18570 @cindex packets, reporting on stdout
18571 @cindex serial connections, debugging
18572 @cindex debug remote protocol
18573 @cindex remote protocol debugging
18574 @cindex display remote packets
18575 @item set debug remote
18576 Turns on or off display of reports on all packets sent back and forth across
18577 the serial line to the remote machine. The info is printed on the
18578 @value{GDBN} standard output stream. The default is off.
18579 @item show debug remote
18580 Displays the state of display of remote packets.
18581 @item set debug serial
18582 Turns on or off display of @value{GDBN} serial debugging info. The
18584 @item show debug serial
18585 Displays the current state of displaying @value{GDBN} serial debugging
18587 @item set debug solib-frv
18588 @cindex FR-V shared-library debugging
18589 Turns on or off debugging messages for FR-V shared-library code.
18590 @item show debug solib-frv
18591 Display the current state of FR-V shared-library code debugging
18593 @item set debug target
18594 @cindex target debugging info
18595 Turns on or off display of @value{GDBN} target debugging info. This info
18596 includes what is going on at the target level of GDB, as it happens. The
18597 default is 0. Set it to 1 to track events, and to 2 to also track the
18598 value of large memory transfers. Changes to this flag do not take effect
18599 until the next time you connect to a target or use the @code{run} command.
18600 @item show debug target
18601 Displays the current state of displaying @value{GDBN} target debugging
18603 @item set debug timestamp
18604 @cindex timestampping debugging info
18605 Turns on or off display of timestamps with @value{GDBN} debugging info.
18606 When enabled, seconds and microseconds are displayed before each debugging
18608 @item show debug timestamp
18609 Displays the current state of displaying timestamps with @value{GDBN}
18611 @item set debugvarobj
18612 @cindex variable object debugging info
18613 Turns on or off display of @value{GDBN} variable object debugging
18614 info. The default is off.
18615 @item show debugvarobj
18616 Displays the current state of displaying @value{GDBN} variable object
18618 @item set debug xml
18619 @cindex XML parser debugging
18620 Turns on or off debugging messages for built-in XML parsers.
18621 @item show debug xml
18622 Displays the current state of XML debugging messages.
18625 @node Other Misc Settings
18626 @section Other Miscellaneous Settings
18627 @cindex miscellaneous settings
18630 @kindex set interactive-mode
18631 @item set interactive-mode
18632 If @code{on}, forces @value{GDBN} to operate interactively.
18633 If @code{off}, forces @value{GDBN} to operate non-interactively,
18634 If @code{auto} (the default), @value{GDBN} guesses which mode to use,
18635 based on whether the debugger was started in a terminal or not.
18637 In the vast majority of cases, the debugger should be able to guess
18638 correctly which mode should be used. But this setting can be useful
18639 in certain specific cases, such as running a MinGW @value{GDBN}
18640 inside a cygwin window.
18642 @kindex show interactive-mode
18643 @item show interactive-mode
18644 Displays whether the debugger is operating in interactive mode or not.
18647 @node Extending GDB
18648 @chapter Extending @value{GDBN}
18649 @cindex extending GDB
18651 @value{GDBN} provides two mechanisms for extension. The first is based
18652 on composition of @value{GDBN} commands, and the second is based on the
18653 Python scripting language.
18656 * Sequences:: Canned Sequences of Commands
18657 * Python:: Scripting @value{GDBN} using Python
18661 @section Canned Sequences of Commands
18663 Aside from breakpoint commands (@pxref{Break Commands, ,Breakpoint
18664 Command Lists}), @value{GDBN} provides two ways to store sequences of
18665 commands for execution as a unit: user-defined commands and command
18669 * Define:: How to define your own commands
18670 * Hooks:: Hooks for user-defined commands
18671 * Command Files:: How to write scripts of commands to be stored in a file
18672 * Output:: Commands for controlled output
18676 @subsection User-defined Commands
18678 @cindex user-defined command
18679 @cindex arguments, to user-defined commands
18680 A @dfn{user-defined command} is a sequence of @value{GDBN} commands to
18681 which you assign a new name as a command. This is done with the
18682 @code{define} command. User commands may accept up to 10 arguments
18683 separated by whitespace. Arguments are accessed within the user command
18684 via @code{$arg0@dots{}$arg9}. A trivial example:
18688 print $arg0 + $arg1 + $arg2
18693 To execute the command use:
18700 This defines the command @code{adder}, which prints the sum of
18701 its three arguments. Note the arguments are text substitutions, so they may
18702 reference variables, use complex expressions, or even perform inferior
18705 @cindex argument count in user-defined commands
18706 @cindex how many arguments (user-defined commands)
18707 In addition, @code{$argc} may be used to find out how many arguments have
18708 been passed. This expands to a number in the range 0@dots{}10.
18713 print $arg0 + $arg1
18716 print $arg0 + $arg1 + $arg2
18724 @item define @var{commandname}
18725 Define a command named @var{commandname}. If there is already a command
18726 by that name, you are asked to confirm that you want to redefine it.
18727 @var{commandname} may be a bare command name consisting of letters,
18728 numbers, dashes, and underscores. It may also start with any predefined
18729 prefix command. For example, @samp{define target my-target} creates
18730 a user-defined @samp{target my-target} command.
18732 The definition of the command is made up of other @value{GDBN} command lines,
18733 which are given following the @code{define} command. The end of these
18734 commands is marked by a line containing @code{end}.
18737 @kindex end@r{ (user-defined commands)}
18738 @item document @var{commandname}
18739 Document the user-defined command @var{commandname}, so that it can be
18740 accessed by @code{help}. The command @var{commandname} must already be
18741 defined. This command reads lines of documentation just as @code{define}
18742 reads the lines of the command definition, ending with @code{end}.
18743 After the @code{document} command is finished, @code{help} on command
18744 @var{commandname} displays the documentation you have written.
18746 You may use the @code{document} command again to change the
18747 documentation of a command. Redefining the command with @code{define}
18748 does not change the documentation.
18750 @kindex dont-repeat
18751 @cindex don't repeat command
18753 Used inside a user-defined command, this tells @value{GDBN} that this
18754 command should not be repeated when the user hits @key{RET}
18755 (@pxref{Command Syntax, repeat last command}).
18757 @kindex help user-defined
18758 @item help user-defined
18759 List all user-defined commands, with the first line of the documentation
18764 @itemx show user @var{commandname}
18765 Display the @value{GDBN} commands used to define @var{commandname} (but
18766 not its documentation). If no @var{commandname} is given, display the
18767 definitions for all user-defined commands.
18769 @cindex infinite recursion in user-defined commands
18770 @kindex show max-user-call-depth
18771 @kindex set max-user-call-depth
18772 @item show max-user-call-depth
18773 @itemx set max-user-call-depth
18774 The value of @code{max-user-call-depth} controls how many recursion
18775 levels are allowed in user-defined commands before @value{GDBN} suspects an
18776 infinite recursion and aborts the command.
18779 In addition to the above commands, user-defined commands frequently
18780 use control flow commands, described in @ref{Command Files}.
18782 When user-defined commands are executed, the
18783 commands of the definition are not printed. An error in any command
18784 stops execution of the user-defined command.
18786 If used interactively, commands that would ask for confirmation proceed
18787 without asking when used inside a user-defined command. Many @value{GDBN}
18788 commands that normally print messages to say what they are doing omit the
18789 messages when used in a user-defined command.
18792 @subsection User-defined Command Hooks
18793 @cindex command hooks
18794 @cindex hooks, for commands
18795 @cindex hooks, pre-command
18798 You may define @dfn{hooks}, which are a special kind of user-defined
18799 command. Whenever you run the command @samp{foo}, if the user-defined
18800 command @samp{hook-foo} exists, it is executed (with no arguments)
18801 before that command.
18803 @cindex hooks, post-command
18805 A hook may also be defined which is run after the command you executed.
18806 Whenever you run the command @samp{foo}, if the user-defined command
18807 @samp{hookpost-foo} exists, it is executed (with no arguments) after
18808 that command. Post-execution hooks may exist simultaneously with
18809 pre-execution hooks, for the same command.
18811 It is valid for a hook to call the command which it hooks. If this
18812 occurs, the hook is not re-executed, thereby avoiding infinite recursion.
18814 @c It would be nice if hookpost could be passed a parameter indicating
18815 @c if the command it hooks executed properly or not. FIXME!
18817 @kindex stop@r{, a pseudo-command}
18818 In addition, a pseudo-command, @samp{stop} exists. Defining
18819 (@samp{hook-stop}) makes the associated commands execute every time
18820 execution stops in your program: before breakpoint commands are run,
18821 displays are printed, or the stack frame is printed.
18823 For example, to ignore @code{SIGALRM} signals while
18824 single-stepping, but treat them normally during normal execution,
18829 handle SIGALRM nopass
18833 handle SIGALRM pass
18836 define hook-continue
18837 handle SIGALRM pass
18841 As a further example, to hook at the beginning and end of the @code{echo}
18842 command, and to add extra text to the beginning and end of the message,
18850 define hookpost-echo
18854 (@value{GDBP}) echo Hello World
18855 <<<---Hello World--->>>
18860 You can define a hook for any single-word command in @value{GDBN}, but
18861 not for command aliases; you should define a hook for the basic command
18862 name, e.g.@: @code{backtrace} rather than @code{bt}.
18863 @c FIXME! So how does Joe User discover whether a command is an alias
18865 You can hook a multi-word command by adding @code{hook-} or
18866 @code{hookpost-} to the last word of the command, e.g.@:
18867 @samp{define target hook-remote} to add a hook to @samp{target remote}.
18869 If an error occurs during the execution of your hook, execution of
18870 @value{GDBN} commands stops and @value{GDBN} issues a prompt
18871 (before the command that you actually typed had a chance to run).
18873 If you try to define a hook which does not match any known command, you
18874 get a warning from the @code{define} command.
18876 @node Command Files
18877 @subsection Command Files
18879 @cindex command files
18880 @cindex scripting commands
18881 A command file for @value{GDBN} is a text file made of lines that are
18882 @value{GDBN} commands. Comments (lines starting with @kbd{#}) may
18883 also be included. An empty line in a command file does nothing; it
18884 does not mean to repeat the last command, as it would from the
18887 You can request the execution of a command file with the @code{source}
18892 @cindex execute commands from a file
18893 @item source [@code{-v}] @var{filename}
18894 Execute the command file @var{filename}.
18897 The lines in a command file are generally executed sequentially,
18898 unless the order of execution is changed by one of the
18899 @emph{flow-control commands} described below. The commands are not
18900 printed as they are executed. An error in any command terminates
18901 execution of the command file and control is returned to the console.
18903 @value{GDBN} searches for @var{filename} in the current directory and then
18904 on the search path (specified with the @samp{directory} command).
18906 If @code{-v}, for verbose mode, is given then @value{GDBN} displays
18907 each command as it is executed. The option must be given before
18908 @var{filename}, and is interpreted as part of the filename anywhere else.
18910 Commands that would ask for confirmation if used interactively proceed
18911 without asking when used in a command file. Many @value{GDBN} commands that
18912 normally print messages to say what they are doing omit the messages
18913 when called from command files.
18915 @value{GDBN} also accepts command input from standard input. In this
18916 mode, normal output goes to standard output and error output goes to
18917 standard error. Errors in a command file supplied on standard input do
18918 not terminate execution of the command file---execution continues with
18922 gdb < cmds > log 2>&1
18925 (The syntax above will vary depending on the shell used.) This example
18926 will execute commands from the file @file{cmds}. All output and errors
18927 would be directed to @file{log}.
18929 Since commands stored on command files tend to be more general than
18930 commands typed interactively, they frequently need to deal with
18931 complicated situations, such as different or unexpected values of
18932 variables and symbols, changes in how the program being debugged is
18933 built, etc. @value{GDBN} provides a set of flow-control commands to
18934 deal with these complexities. Using these commands, you can write
18935 complex scripts that loop over data structures, execute commands
18936 conditionally, etc.
18943 This command allows to include in your script conditionally executed
18944 commands. The @code{if} command takes a single argument, which is an
18945 expression to evaluate. It is followed by a series of commands that
18946 are executed only if the expression is true (its value is nonzero).
18947 There can then optionally be an @code{else} line, followed by a series
18948 of commands that are only executed if the expression was false. The
18949 end of the list is marked by a line containing @code{end}.
18953 This command allows to write loops. Its syntax is similar to
18954 @code{if}: the command takes a single argument, which is an expression
18955 to evaluate, and must be followed by the commands to execute, one per
18956 line, terminated by an @code{end}. These commands are called the
18957 @dfn{body} of the loop. The commands in the body of @code{while} are
18958 executed repeatedly as long as the expression evaluates to true.
18962 This command exits the @code{while} loop in whose body it is included.
18963 Execution of the script continues after that @code{while}s @code{end}
18966 @kindex loop_continue
18967 @item loop_continue
18968 This command skips the execution of the rest of the body of commands
18969 in the @code{while} loop in whose body it is included. Execution
18970 branches to the beginning of the @code{while} loop, where it evaluates
18971 the controlling expression.
18973 @kindex end@r{ (if/else/while commands)}
18975 Terminate the block of commands that are the body of @code{if},
18976 @code{else}, or @code{while} flow-control commands.
18981 @subsection Commands for Controlled Output
18983 During the execution of a command file or a user-defined command, normal
18984 @value{GDBN} output is suppressed; the only output that appears is what is
18985 explicitly printed by the commands in the definition. This section
18986 describes three commands useful for generating exactly the output you
18991 @item echo @var{text}
18992 @c I do not consider backslash-space a standard C escape sequence
18993 @c because it is not in ANSI.
18994 Print @var{text}. Nonprinting characters can be included in
18995 @var{text} using C escape sequences, such as @samp{\n} to print a
18996 newline. @strong{No newline is printed unless you specify one.}
18997 In addition to the standard C escape sequences, a backslash followed
18998 by a space stands for a space. This is useful for displaying a
18999 string with spaces at the beginning or the end, since leading and
19000 trailing spaces are otherwise trimmed from all arguments.
19001 To print @samp{@w{ }and foo =@w{ }}, use the command
19002 @samp{echo \@w{ }and foo = \@w{ }}.
19004 A backslash at the end of @var{text} can be used, as in C, to continue
19005 the command onto subsequent lines. For example,
19008 echo This is some text\n\
19009 which is continued\n\
19010 onto several lines.\n
19013 produces the same output as
19016 echo This is some text\n
19017 echo which is continued\n
19018 echo onto several lines.\n
19022 @item output @var{expression}
19023 Print the value of @var{expression} and nothing but that value: no
19024 newlines, no @samp{$@var{nn} = }. The value is not entered in the
19025 value history either. @xref{Expressions, ,Expressions}, for more information
19028 @item output/@var{fmt} @var{expression}
19029 Print the value of @var{expression} in format @var{fmt}. You can use
19030 the same formats as for @code{print}. @xref{Output Formats,,Output
19031 Formats}, for more information.
19034 @item printf @var{template}, @var{expressions}@dots{}
19035 Print the values of one or more @var{expressions} under the control of
19036 the string @var{template}. To print several values, make
19037 @var{expressions} be a comma-separated list of individual expressions,
19038 which may be either numbers or pointers. Their values are printed as
19039 specified by @var{template}, exactly as a C program would do by
19040 executing the code below:
19043 printf (@var{template}, @var{expressions}@dots{});
19046 As in @code{C} @code{printf}, ordinary characters in @var{template}
19047 are printed verbatim, while @dfn{conversion specification} introduced
19048 by the @samp{%} character cause subsequent @var{expressions} to be
19049 evaluated, their values converted and formatted according to type and
19050 style information encoded in the conversion specifications, and then
19053 For example, you can print two values in hex like this:
19056 printf "foo, bar-foo = 0x%x, 0x%x\n", foo, bar-foo
19059 @code{printf} supports all the standard @code{C} conversion
19060 specifications, including the flags and modifiers between the @samp{%}
19061 character and the conversion letter, with the following exceptions:
19065 The argument-ordering modifiers, such as @samp{2$}, are not supported.
19068 The modifier @samp{*} is not supported for specifying precision or
19072 The @samp{'} flag (for separation of digits into groups according to
19073 @code{LC_NUMERIC'}) is not supported.
19076 The type modifiers @samp{hh}, @samp{j}, @samp{t}, and @samp{z} are not
19080 The conversion letter @samp{n} (as in @samp{%n}) is not supported.
19083 The conversion letters @samp{a} and @samp{A} are not supported.
19087 Note that the @samp{ll} type modifier is supported only if the
19088 underlying @code{C} implementation used to build @value{GDBN} supports
19089 the @code{long long int} type, and the @samp{L} type modifier is
19090 supported only if @code{long double} type is available.
19092 As in @code{C}, @code{printf} supports simple backslash-escape
19093 sequences, such as @code{\n}, @samp{\t}, @samp{\\}, @samp{\"},
19094 @samp{\a}, and @samp{\f}, that consist of backslash followed by a
19095 single character. Octal and hexadecimal escape sequences are not
19098 Additionally, @code{printf} supports conversion specifications for DFP
19099 (@dfn{Decimal Floating Point}) types using the following length modifiers
19100 together with a floating point specifier.
19105 @samp{H} for printing @code{Decimal32} types.
19108 @samp{D} for printing @code{Decimal64} types.
19111 @samp{DD} for printing @code{Decimal128} types.
19114 If the underlying @code{C} implementation used to build @value{GDBN} has
19115 support for the three length modifiers for DFP types, other modifiers
19116 such as width and precision will also be available for @value{GDBN} to use.
19118 In case there is no such @code{C} support, no additional modifiers will be
19119 available and the value will be printed in the standard way.
19121 Here's an example of printing DFP types using the above conversion letters:
19123 printf "D32: %Hf - D64: %Df - D128: %DDf\n",1.2345df,1.2E10dd,1.2E1dl
19129 @section Scripting @value{GDBN} using Python
19130 @cindex python scripting
19131 @cindex scripting with python
19133 You can script @value{GDBN} using the @uref{http://www.python.org/,
19134 Python programming language}. This feature is available only if
19135 @value{GDBN} was configured using @option{--with-python}.
19138 * Python Commands:: Accessing Python from @value{GDBN}.
19139 * Python API:: Accessing @value{GDBN} from Python.
19142 @node Python Commands
19143 @subsection Python Commands
19144 @cindex python commands
19145 @cindex commands to access python
19147 @value{GDBN} provides one command for accessing the Python interpreter,
19148 and one related setting:
19152 @item python @r{[}@var{code}@r{]}
19153 The @code{python} command can be used to evaluate Python code.
19155 If given an argument, the @code{python} command will evaluate the
19156 argument as a Python command. For example:
19159 (@value{GDBP}) python print 23
19163 If you do not provide an argument to @code{python}, it will act as a
19164 multi-line command, like @code{define}. In this case, the Python
19165 script is made up of subsequent command lines, given after the
19166 @code{python} command. This command list is terminated using a line
19167 containing @code{end}. For example:
19170 (@value{GDBP}) python
19172 End with a line saying just "end".
19178 @kindex maint set python print-stack
19179 @item maint set python print-stack
19180 By default, @value{GDBN} will print a stack trace when an error occurs
19181 in a Python script. This can be controlled using @code{maint set
19182 python print-stack}: if @code{on}, the default, then Python stack
19183 printing is enabled; if @code{off}, then Python stack printing is
19188 @subsection Python API
19190 @cindex programming in python
19192 @cindex python stdout
19193 @cindex python pagination
19194 At startup, @value{GDBN} overrides Python's @code{sys.stdout} and
19195 @code{sys.stderr} to print using @value{GDBN}'s output-paging streams.
19196 A Python program which outputs to one of these streams may have its
19197 output interrupted by the user (@pxref{Screen Size}). In this
19198 situation, a Python @code{KeyboardInterrupt} exception is thrown.
19201 * Basic Python:: Basic Python Functions.
19202 * Exception Handling::
19203 * Auto-loading:: Automatically loading Python code.
19204 * Values From Inferior::
19205 * Types In Python:: Python representation of types.
19206 * Pretty Printing:: Pretty-printing values.
19207 * Selecting Pretty-Printers:: How GDB chooses a pretty-printer.
19208 * Commands In Python:: Implementing new commands in Python.
19209 * Functions In Python:: Writing new convenience functions.
19210 * Objfiles In Python:: Object files.
19211 * Frames In Python:: Acessing inferior stack frames from Python.
19215 @subsubsection Basic Python
19217 @cindex python functions
19218 @cindex python module
19220 @value{GDBN} introduces a new Python module, named @code{gdb}. All
19221 methods and classes added by @value{GDBN} are placed in this module.
19222 @value{GDBN} automatically @code{import}s the @code{gdb} module for
19223 use in all scripts evaluated by the @code{python} command.
19225 @findex gdb.execute
19226 @defun execute command [from_tty]
19227 Evaluate @var{command}, a string, as a @value{GDBN} CLI command.
19228 If a GDB exception happens while @var{command} runs, it is
19229 translated as described in @ref{Exception Handling,,Exception Handling}.
19230 If no exceptions occur, this function returns @code{None}.
19232 @var{from_tty} specifies whether @value{GDBN} ought to consider this
19233 command as having originated from the user invoking it interactively.
19234 It must be a boolean value. If omitted, it defaults to @code{False}.
19237 @findex gdb.parameter
19238 @defun parameter parameter
19239 Return the value of a @value{GDBN} parameter. @var{parameter} is a
19240 string naming the parameter to look up; @var{parameter} may contain
19241 spaces if the parameter has a multi-part name. For example,
19242 @samp{print object} is a valid parameter name.
19244 If the named parameter does not exist, this function throws a
19245 @code{RuntimeError}. Otherwise, the parameter's value is converted to
19246 a Python value of the appropriate type, and returned.
19249 @findex gdb.history
19250 @defun history number
19251 Return a value from @value{GDBN}'s value history (@pxref{Value
19252 History}). @var{number} indicates which history element to return.
19253 If @var{number} is negative, then @value{GDBN} will take its absolute value
19254 and count backward from the last element (i.e., the most recent element) to
19255 find the value to return. If @var{number} is zero, then @value{GDBN} will
19256 return the most recent element. If the element specified by @var{number}
19257 doesn't exist in the value history, a @code{RuntimeError} exception will be
19260 If no exception is raised, the return value is always an instance of
19261 @code{gdb.Value} (@pxref{Values From Inferior}).
19265 @defun write string
19266 Print a string to @value{GDBN}'s paginated standard output stream.
19267 Writing to @code{sys.stdout} or @code{sys.stderr} will automatically
19268 call this function.
19273 Flush @value{GDBN}'s paginated standard output stream. Flushing
19274 @code{sys.stdout} or @code{sys.stderr} will automatically call this
19278 @node Exception Handling
19279 @subsubsection Exception Handling
19280 @cindex python exceptions
19281 @cindex exceptions, python
19283 When executing the @code{python} command, Python exceptions
19284 uncaught within the Python code are translated to calls to
19285 @value{GDBN} error-reporting mechanism. If the command that called
19286 @code{python} does not handle the error, @value{GDBN} will
19287 terminate it and print an error message containing the Python
19288 exception name, the associated value, and the Python call stack
19289 backtrace at the point where the exception was raised. Example:
19292 (@value{GDBP}) python print foo
19293 Traceback (most recent call last):
19294 File "<string>", line 1, in <module>
19295 NameError: name 'foo' is not defined
19298 @value{GDBN} errors that happen in @value{GDBN} commands invoked by Python
19299 code are converted to Python @code{RuntimeError} exceptions. User
19300 interrupt (via @kbd{C-c} or by typing @kbd{q} at a pagination
19301 prompt) is translated to a Python @code{KeyboardInterrupt}
19302 exception. If you catch these exceptions in your Python code, your
19303 exception handler will see @code{RuntimeError} or
19304 @code{KeyboardInterrupt} as the exception type, the @value{GDBN} error
19305 message as its value, and the Python call stack backtrace at the
19306 Python statement closest to where the @value{GDBN} error occured as the
19310 @subsubsection Auto-loading
19311 @cindex auto-loading, Python
19313 When a new object file is read (for example, due to the @code{file}
19314 command, or because the inferior has loaded a shared library),
19315 @value{GDBN} will look for a file named @file{@var{objfile}-gdb.py},
19316 where @var{objfile} is the object file's real name, formed by ensuring
19317 that the file name is absolute, following all symlinks, and resolving
19318 @code{.} and @code{..} components. If this file exists and is
19319 readable, @value{GDBN} will evaluate it as a Python script.
19321 If this file does not exist, and if the parameter
19322 @code{debug-file-directory} is set (@pxref{Separate Debug Files}),
19323 then @value{GDBN} will use the file named
19324 @file{@var{debug-file-directory}/@var{real-name}}, where
19325 @var{real-name} is the object file's real name, as described above.
19327 Finally, if this file does not exist, then @value{GDBN} will look for
19328 a file named @file{@var{data-directory}/python/auto-load/@var{real-name}}, where
19329 @var{data-directory} is @value{GDBN}'s data directory (available via
19330 @code{show data-directory}, @pxref{Data Files}), and @var{real-name}
19331 is the object file's real name, as described above.
19333 When reading an auto-loaded file, @value{GDBN} sets the ``current
19334 objfile''. This is available via the @code{gdb.current_objfile}
19335 function (@pxref{Objfiles In Python}). This can be useful for
19336 registering objfile-specific pretty-printers.
19338 The auto-loading feature is useful for supplying application-specific
19339 debugging commands and scripts. You can enable or disable this
19340 feature, and view its current state.
19343 @kindex maint set python auto-load
19344 @item maint set python auto-load [yes|no]
19345 Enable or disable the Python auto-loading feature.
19347 @kindex show python auto-load
19348 @item show python auto-load
19349 Show whether Python auto-loading is enabled or disabled.
19352 @value{GDBN} does not track which files it has already auto-loaded.
19353 So, your @samp{-gdb.py} file should take care to ensure that it may be
19354 evaluated multiple times without error.
19356 @node Values From Inferior
19357 @subsubsection Values From Inferior
19358 @cindex values from inferior, with Python
19359 @cindex python, working with values from inferior
19361 @cindex @code{gdb.Value}
19362 @value{GDBN} provides values it obtains from the inferior program in
19363 an object of type @code{gdb.Value}. @value{GDBN} uses this object
19364 for its internal bookkeeping of the inferior's values, and for
19365 fetching values when necessary.
19367 Inferior values that are simple scalars can be used directly in
19368 Python expressions that are valid for the value's data type. Here's
19369 an example for an integer or floating-point value @code{some_val}:
19376 As result of this, @code{bar} will also be a @code{gdb.Value} object
19377 whose values are of the same type as those of @code{some_val}.
19379 Inferior values that are structures or instances of some class can
19380 be accessed using the Python @dfn{dictionary syntax}. For example, if
19381 @code{some_val} is a @code{gdb.Value} instance holding a structure, you
19382 can access its @code{foo} element with:
19385 bar = some_val['foo']
19388 Again, @code{bar} will also be a @code{gdb.Value} object.
19390 The following attributes are provided:
19393 @defivar Value address
19394 If this object is addressable, this read-only attribute holds a
19395 @code{gdb.Value} object representing the address. Otherwise,
19396 this attribute holds @code{None}.
19399 @cindex optimized out value in Python
19400 @defivar Value is_optimized_out
19401 This read-only boolean attribute is true if the compiler optimized out
19402 this value, thus it is not available for fetching from the inferior.
19405 @defivar Value type
19406 The type of this @code{gdb.Value}. The value of this attribute is a
19407 @code{gdb.Type} object.
19411 The following methods are provided:
19414 @defmethod Value dereference
19415 For pointer data types, this method returns a new @code{gdb.Value} object
19416 whose contents is the object pointed to by the pointer. For example, if
19417 @code{foo} is a C pointer to an @code{int}, declared in your C program as
19424 then you can use the corresponding @code{gdb.Value} to access what
19425 @code{foo} points to like this:
19428 bar = foo.dereference ()
19431 The result @code{bar} will be a @code{gdb.Value} object holding the
19432 value pointed to by @code{foo}.
19435 @defmethod Value string @r{[}encoding@r{]} @r{[}errors@r{]} @r{[}length@r{]}
19436 If this @code{gdb.Value} represents a string, then this method
19437 converts the contents to a Python string. Otherwise, this method will
19438 throw an exception.
19440 Strings are recognized in a language-specific way; whether a given
19441 @code{gdb.Value} represents a string is determined by the current
19444 For C-like languages, a value is a string if it is a pointer to or an
19445 array of characters or ints. The string is assumed to be terminated
19446 by a zero of the appropriate width. However if the optional length
19447 argument is given, the string will be converted to that given length,
19448 ignoring any embedded zeros that the string may contain.
19450 If the optional @var{encoding} argument is given, it must be a string
19451 naming the encoding of the string in the @code{gdb.Value}, such as
19452 @code{"ascii"}, @code{"iso-8859-6"} or @code{"utf-8"}. It accepts
19453 the same encodings as the corresponding argument to Python's
19454 @code{string.decode} method, and the Python codec machinery will be used
19455 to convert the string. If @var{encoding} is not given, or if
19456 @var{encoding} is the empty string, then either the @code{target-charset}
19457 (@pxref{Character Sets}) will be used, or a language-specific encoding
19458 will be used, if the current language is able to supply one.
19460 The optional @var{errors} argument is the same as the corresponding
19461 argument to Python's @code{string.decode} method.
19463 If the optional @var{length} argument is given, the string will be
19464 fetched and converted to the given length.
19468 @node Types In Python
19469 @subsubsection Types In Python
19470 @cindex types in Python
19471 @cindex Python, working with types
19474 @value{GDBN} represents types from the inferior using the class
19477 The following type-related functions are available in the @code{gdb}
19480 @findex gdb.lookup_type
19481 @defun lookup_type name [block]
19482 This function looks up a type by name. @var{name} is the name of the
19483 type to look up. It must be a string.
19485 Ordinarily, this function will return an instance of @code{gdb.Type}.
19486 If the named type cannot be found, it will throw an exception.
19489 An instance of @code{Type} has the following attributes:
19493 The type code for this type. The type code will be one of the
19494 @code{TYPE_CODE_} constants defined below.
19497 @defivar Type sizeof
19498 The size of this type, in target @code{char} units. Usually, a
19499 target's @code{char} type will be an 8-bit byte. However, on some
19500 unusual platforms, this type may have a different size.
19504 The tag name for this type. The tag name is the name after
19505 @code{struct}, @code{union}, or @code{enum} in C and C@t{++}; not all
19506 languages have this concept. If this type has no tag name, then
19507 @code{None} is returned.
19511 The following methods are provided:
19514 @defmethod Type fields
19515 For structure and union types, this method returns the fields. Range
19516 types have two fields, the minimum and maximum values. Enum types
19517 have one field per enum constant. Function and method types have one
19518 field per parameter. The base types of C@t{++} classes are also
19519 represented as fields. If the type has no fields, or does not fit
19520 into one of these categories, an empty sequence will be returned.
19522 Each field is an object, with some pre-defined attributes:
19525 This attribute is not available for @code{static} fields (as in
19526 C@t{++} or Java). For non-@code{static} fields, the value is the bit
19527 position of the field.
19530 The name of the field, or @code{None} for anonymous fields.
19533 This is @code{True} if the field is artificial, usually meaning that
19534 it was provided by the compiler and not the user. This attribute is
19535 always provided, and is @code{False} if the field is not artificial.
19538 If the field is packed, or is a bitfield, then this will have a
19539 non-zero value, which is the size of the field in bits. Otherwise,
19540 this will be zero; in this case the field's size is given by its type.
19543 The type of the field. This is usually an instance of @code{Type},
19544 but it can be @code{None} in some situations.
19548 @defmethod Type const
19549 Return a new @code{gdb.Type} object which represents a
19550 @code{const}-qualified variant of this type.
19553 @defmethod Type volatile
19554 Return a new @code{gdb.Type} object which represents a
19555 @code{volatile}-qualified variant of this type.
19558 @defmethod Type unqualified
19559 Return a new @code{gdb.Type} object which represents an unqualified
19560 variant of this type. That is, the result is neither @code{const} nor
19564 @defmethod Type reference
19565 Return a new @code{gdb.Type} object which represents a reference to this
19569 @defmethod Type strip_typedefs
19570 Return a new @code{gdb.Type} that represents the real type,
19571 after removing all layers of typedefs.
19574 @defmethod Type target
19575 Return a new @code{gdb.Type} object which represents the target type
19578 For a pointer type, the target type is the type of the pointed-to
19579 object. For an array type (meaning C-like arrays), the target type is
19580 the type of the elements of the array. For a function or method type,
19581 the target type is the type of the return value. For a complex type,
19582 the target type is the type of the elements. For a typedef, the
19583 target type is the aliased type.
19585 If the type does not have a target, this method will throw an
19589 @defmethod Type template_argument n
19590 If this @code{gdb.Type} is an instantiation of a template, this will
19591 return a new @code{gdb.Type} which represents the type of the
19592 @var{n}th template argument.
19594 If this @code{gdb.Type} is not a template type, this will throw an
19595 exception. Ordinarily, only C@t{++} code will have template types.
19597 @var{name} is searched for globally.
19602 Each type has a code, which indicates what category this type falls
19603 into. The available type categories are represented by constants
19604 defined in the @code{gdb} module:
19607 @findex TYPE_CODE_PTR
19608 @findex gdb.TYPE_CODE_PTR
19609 @item TYPE_CODE_PTR
19610 The type is a pointer.
19612 @findex TYPE_CODE_ARRAY
19613 @findex gdb.TYPE_CODE_ARRAY
19614 @item TYPE_CODE_ARRAY
19615 The type is an array.
19617 @findex TYPE_CODE_STRUCT
19618 @findex gdb.TYPE_CODE_STRUCT
19619 @item TYPE_CODE_STRUCT
19620 The type is a structure.
19622 @findex TYPE_CODE_UNION
19623 @findex gdb.TYPE_CODE_UNION
19624 @item TYPE_CODE_UNION
19625 The type is a union.
19627 @findex TYPE_CODE_ENUM
19628 @findex gdb.TYPE_CODE_ENUM
19629 @item TYPE_CODE_ENUM
19630 The type is an enum.
19632 @findex TYPE_CODE_FLAGS
19633 @findex gdb.TYPE_CODE_FLAGS
19634 @item TYPE_CODE_FLAGS
19635 A bit flags type, used for things such as status registers.
19637 @findex TYPE_CODE_FUNC
19638 @findex gdb.TYPE_CODE_FUNC
19639 @item TYPE_CODE_FUNC
19640 The type is a function.
19642 @findex TYPE_CODE_INT
19643 @findex gdb.TYPE_CODE_INT
19644 @item TYPE_CODE_INT
19645 The type is an integer type.
19647 @findex TYPE_CODE_FLT
19648 @findex gdb.TYPE_CODE_FLT
19649 @item TYPE_CODE_FLT
19650 A floating point type.
19652 @findex TYPE_CODE_VOID
19653 @findex gdb.TYPE_CODE_VOID
19654 @item TYPE_CODE_VOID
19655 The special type @code{void}.
19657 @findex TYPE_CODE_SET
19658 @findex gdb.TYPE_CODE_SET
19659 @item TYPE_CODE_SET
19662 @findex TYPE_CODE_RANGE
19663 @findex gdb.TYPE_CODE_RANGE
19664 @item TYPE_CODE_RANGE
19665 A range type, that is, an integer type with bounds.
19667 @findex TYPE_CODE_STRING
19668 @findex gdb.TYPE_CODE_STRING
19669 @item TYPE_CODE_STRING
19670 A string type. Note that this is only used for certain languages with
19671 language-defined string types; C strings are not represented this way.
19673 @findex TYPE_CODE_BITSTRING
19674 @findex gdb.TYPE_CODE_BITSTRING
19675 @item TYPE_CODE_BITSTRING
19678 @findex TYPE_CODE_ERROR
19679 @findex gdb.TYPE_CODE_ERROR
19680 @item TYPE_CODE_ERROR
19681 An unknown or erroneous type.
19683 @findex TYPE_CODE_METHOD
19684 @findex gdb.TYPE_CODE_METHOD
19685 @item TYPE_CODE_METHOD
19686 A method type, as found in C@t{++} or Java.
19688 @findex TYPE_CODE_METHODPTR
19689 @findex gdb.TYPE_CODE_METHODPTR
19690 @item TYPE_CODE_METHODPTR
19691 A pointer-to-member-function.
19693 @findex TYPE_CODE_MEMBERPTR
19694 @findex gdb.TYPE_CODE_MEMBERPTR
19695 @item TYPE_CODE_MEMBERPTR
19696 A pointer-to-member.
19698 @findex TYPE_CODE_REF
19699 @findex gdb.TYPE_CODE_REF
19700 @item TYPE_CODE_REF
19703 @findex TYPE_CODE_CHAR
19704 @findex gdb.TYPE_CODE_CHAR
19705 @item TYPE_CODE_CHAR
19708 @findex TYPE_CODE_BOOL
19709 @findex gdb.TYPE_CODE_BOOL
19710 @item TYPE_CODE_BOOL
19713 @findex TYPE_CODE_COMPLEX
19714 @findex gdb.TYPE_CODE_COMPLEX
19715 @item TYPE_CODE_COMPLEX
19716 A complex float type.
19718 @findex TYPE_CODE_TYPEDEF
19719 @findex gdb.TYPE_CODE_TYPEDEF
19720 @item TYPE_CODE_TYPEDEF
19721 A typedef to some other type.
19723 @findex TYPE_CODE_NAMESPACE
19724 @findex gdb.TYPE_CODE_NAMESPACE
19725 @item TYPE_CODE_NAMESPACE
19726 A C@t{++} namespace.
19728 @findex TYPE_CODE_DECFLOAT
19729 @findex gdb.TYPE_CODE_DECFLOAT
19730 @item TYPE_CODE_DECFLOAT
19731 A decimal floating point type.
19733 @findex TYPE_CODE_INTERNAL_FUNCTION
19734 @findex gdb.TYPE_CODE_INTERNAL_FUNCTION
19735 @item TYPE_CODE_INTERNAL_FUNCTION
19736 A function internal to @value{GDBN}. This is the type used to represent
19737 convenience functions.
19740 @node Pretty Printing
19741 @subsubsection Pretty Printing
19743 @value{GDBN} provides a mechanism to allow pretty-printing of values
19744 using Python code. The pretty-printer API allows application-specific
19745 code to greatly simplify the display of complex objects. This
19746 mechanism works for both MI and the CLI.
19748 For example, here is how a C@t{++} @code{std::string} looks without a
19752 (@value{GDBP}) print s
19754 static npos = 4294967295,
19756 <std::allocator<char>> = @{
19757 <__gnu_cxx::new_allocator<char>> = @{<No data fields>@}, <No data fields>@},
19758 members of std::basic_string<char, std::char_traits<char>, std::allocator<char> >::_Alloc_hider:
19759 _M_p = 0x804a014 "abcd"
19764 After a pretty-printer for @code{std::string} has been installed, only
19765 the contents are printed:
19768 (@value{GDBP}) print s
19772 A pretty-printer is just an object that holds a value and implements a
19773 specific interface, defined here.
19775 @defop Operation {pretty printer} children (self)
19776 @value{GDBN} will call this method on a pretty-printer to compute the
19777 children of the pretty-printer's value.
19779 This method must return an object conforming to the Python iterator
19780 protocol. Each item returned by the iterator must be a tuple holding
19781 two elements. The first element is the ``name'' of the child; the
19782 second element is the child's value. The value can be any Python
19783 object which is convertible to a @value{GDBN} value.
19785 This method is optional. If it does not exist, @value{GDBN} will act
19786 as though the value has no children.
19789 @defop Operation {pretty printer} display_hint (self)
19790 The CLI may call this method and use its result to change the
19791 formatting of a value. The result will also be supplied to an MI
19792 consumer as a @samp{displayhint} attribute of the variable being
19795 This method is optional. If it does exist, this method must return a
19798 Some display hints are predefined by @value{GDBN}:
19802 Indicate that the object being printed is ``array-like''. The CLI
19803 uses this to respect parameters such as @code{set print elements} and
19804 @code{set print array}.
19807 Indicate that the object being printed is ``map-like'', and that the
19808 children of this value can be assumed to alternate between keys and
19812 Indicate that the object being printed is ``string-like''. If the
19813 printer's @code{to_string} method returns a Python string of some
19814 kind, then @value{GDBN} will call its internal language-specific
19815 string-printing function to format the string. For the CLI this means
19816 adding quotation marks, possibly escaping some characters, respecting
19817 @code{set print elements}, and the like.
19821 @defop Operation {pretty printer} to_string (self)
19822 @value{GDBN} will call this method to display the string
19823 representation of the value passed to the object's constructor.
19825 When printing from the CLI, if the @code{to_string} method exists,
19826 then @value{GDBN} will prepend its result to the values returned by
19827 @code{children}. Exactly how this formatting is done is dependent on
19828 the display hint, and may change as more hints are added. Also,
19829 depending on the print settings (@pxref{Print Settings}), the CLI may
19830 print just the result of @code{to_string} in a stack trace, omitting
19831 the result of @code{children}.
19833 If this method returns a string, it is printed verbatim.
19835 Otherwise, if this method returns an instance of @code{gdb.Value},
19836 then @value{GDBN} prints this value. This may result in a call to
19837 another pretty-printer.
19839 If instead the method returns a Python value which is convertible to a
19840 @code{gdb.Value}, then @value{GDBN} performs the conversion and prints
19841 the resulting value. Again, this may result in a call to another
19842 pretty-printer. Python scalars (integers, floats, and booleans) and
19843 strings are convertible to @code{gdb.Value}; other types are not.
19845 If the result is not one of these types, an exception is raised.
19848 @node Selecting Pretty-Printers
19849 @subsubsection Selecting Pretty-Printers
19851 The Python list @code{gdb.pretty_printers} contains an array of
19852 functions that have been registered via addition as a pretty-printer.
19853 Each @code{gdb.Objfile} also contains a @code{pretty_printers}
19856 A function on one of these lists is passed a single @code{gdb.Value}
19857 argument and should return a pretty-printer object conforming to the
19858 interface definition above (@pxref{Pretty Printing}). If a function
19859 cannot create a pretty-printer for the value, it should return
19862 @value{GDBN} first checks the @code{pretty_printers} attribute of each
19863 @code{gdb.Objfile} and iteratively calls each function in the list for
19864 that @code{gdb.Objfile} until it receives a pretty-printer object.
19865 After these lists have been exhausted, it tries the global
19866 @code{gdb.pretty-printers} list, again calling each function until an
19867 object is returned.
19869 The order in which the objfiles are searched is not specified. For a
19870 given list, functions are always invoked from the head of the list,
19871 and iterated over sequentially until the end of the list, or a printer
19872 object is returned.
19874 Here is an example showing how a @code{std::string} printer might be
19878 class StdStringPrinter:
19879 "Print a std::string"
19881 def __init__ (self, val):
19884 def to_string (self):
19885 return self.val['_M_dataplus']['_M_p']
19887 def display_hint (self):
19891 And here is an example showing how a lookup function for the printer
19892 example above might be written.
19895 def str_lookup_function (val):
19897 lookup_tag = val.type.tag
19898 regex = re.compile ("^std::basic_string<char,.*>$")
19899 if lookup_tag == None:
19901 if regex.match (lookup_tag):
19902 return StdStringPrinter (val)
19907 The example lookup function extracts the value's type, and attempts to
19908 match it to a type that it can pretty-print. If it is a type the
19909 printer can pretty-print, it will return a printer object. If not, it
19910 returns @code{None}.
19912 We recommend that you put your core pretty-printers into a Python
19913 package. If your pretty-printers are for use with a library, we
19914 further recommend embedding a version number into the package name.
19915 This practice will enable @value{GDBN} to load multiple versions of
19916 your pretty-printers at the same time, because they will have
19919 You should write auto-loaded code (@pxref{Auto-loading}) such that it
19920 can be evaluated multiple times without changing its meaning. An
19921 ideal auto-load file will consist solely of @code{import}s of your
19922 printer modules, followed by a call to a register pretty-printers with
19923 the current objfile.
19925 Taken as a whole, this approach will scale nicely to multiple
19926 inferiors, each potentially using a different library version.
19927 Embedding a version number in the Python package name will ensure that
19928 @value{GDBN} is able to load both sets of printers simultaneously.
19929 Then, because the search for pretty-printers is done by objfile, and
19930 because your auto-loaded code took care to register your library's
19931 printers with a specific objfile, @value{GDBN} will find the correct
19932 printers for the specific version of the library used by each
19935 To continue the @code{std::string} example (@pxref{Pretty Printing}),
19936 this code might appear in @code{gdb.libstdcxx.v6}:
19939 def register_printers (objfile):
19940 objfile.pretty_printers.add (str_lookup_function)
19944 And then the corresponding contents of the auto-load file would be:
19947 import gdb.libstdcxx.v6
19948 gdb.libstdcxx.v6.register_printers (gdb.current_objfile ())
19951 @node Commands In Python
19952 @subsubsection Commands In Python
19954 @cindex commands in python
19955 @cindex python commands
19956 You can implement new @value{GDBN} CLI commands in Python. A CLI
19957 command is implemented using an instance of the @code{gdb.Command}
19958 class, most commonly using a subclass.
19960 @defmethod Command __init__ name @var{command_class} @r{[}@var{completer_class}@r{]} @r{[}@var{prefix}@r{]}
19961 The object initializer for @code{Command} registers the new command
19962 with @value{GDBN}. This initializer is normally invoked from the
19963 subclass' own @code{__init__} method.
19965 @var{name} is the name of the command. If @var{name} consists of
19966 multiple words, then the initial words are looked for as prefix
19967 commands. In this case, if one of the prefix commands does not exist,
19968 an exception is raised.
19970 There is no support for multi-line commands.
19972 @var{command_class} should be one of the @samp{COMMAND_} constants
19973 defined below. This argument tells @value{GDBN} how to categorize the
19974 new command in the help system.
19976 @var{completer_class} is an optional argument. If given, it should be
19977 one of the @samp{COMPLETE_} constants defined below. This argument
19978 tells @value{GDBN} how to perform completion for this command. If not
19979 given, @value{GDBN} will attempt to complete using the object's
19980 @code{complete} method (see below); if no such method is found, an
19981 error will occur when completion is attempted.
19983 @var{prefix} is an optional argument. If @code{True}, then the new
19984 command is a prefix command; sub-commands of this command may be
19987 The help text for the new command is taken from the Python
19988 documentation string for the command's class, if there is one. If no
19989 documentation string is provided, the default value ``This command is
19990 not documented.'' is used.
19993 @cindex don't repeat Python command
19994 @defmethod Command dont_repeat
19995 By default, a @value{GDBN} command is repeated when the user enters a
19996 blank line at the command prompt. A command can suppress this
19997 behavior by invoking the @code{dont_repeat} method. This is similar
19998 to the user command @code{dont-repeat}, see @ref{Define, dont-repeat}.
20001 @defmethod Command invoke argument from_tty
20002 This method is called by @value{GDBN} when this command is invoked.
20004 @var{argument} is a string. It is the argument to the command, after
20005 leading and trailing whitespace has been stripped.
20007 @var{from_tty} is a boolean argument. When true, this means that the
20008 command was entered by the user at the terminal; when false it means
20009 that the command came from elsewhere.
20011 If this method throws an exception, it is turned into a @value{GDBN}
20012 @code{error} call. Otherwise, the return value is ignored.
20015 @cindex completion of Python commands
20016 @defmethod Command complete text word
20017 This method is called by @value{GDBN} when the user attempts
20018 completion on this command. All forms of completion are handled by
20019 this method, that is, the @key{TAB} and @key{M-?} key bindings
20020 (@pxref{Completion}), and the @code{complete} command (@pxref{Help,
20023 The arguments @var{text} and @var{word} are both strings. @var{text}
20024 holds the complete command line up to the cursor's location.
20025 @var{word} holds the last word of the command line; this is computed
20026 using a word-breaking heuristic.
20028 The @code{complete} method can return several values:
20031 If the return value is a sequence, the contents of the sequence are
20032 used as the completions. It is up to @code{complete} to ensure that the
20033 contents actually do complete the word. A zero-length sequence is
20034 allowed, it means that there were no completions available. Only
20035 string elements of the sequence are used; other elements in the
20036 sequence are ignored.
20039 If the return value is one of the @samp{COMPLETE_} constants defined
20040 below, then the corresponding @value{GDBN}-internal completion
20041 function is invoked, and its result is used.
20044 All other results are treated as though there were no available
20049 When a new command is registered, it must be declared as a member of
20050 some general class of commands. This is used to classify top-level
20051 commands in the on-line help system; note that prefix commands are not
20052 listed under their own category but rather that of their top-level
20053 command. The available classifications are represented by constants
20054 defined in the @code{gdb} module:
20057 @findex COMMAND_NONE
20058 @findex gdb.COMMAND_NONE
20060 The command does not belong to any particular class. A command in
20061 this category will not be displayed in any of the help categories.
20063 @findex COMMAND_RUNNING
20064 @findex gdb.COMMAND_RUNNING
20065 @item COMMAND_RUNNING
20066 The command is related to running the inferior. For example,
20067 @code{start}, @code{step}, and @code{continue} are in this category.
20068 Type @kbd{help running} at the @value{GDBN} prompt to see a list of
20069 commands in this category.
20071 @findex COMMAND_DATA
20072 @findex gdb.COMMAND_DATA
20074 The command is related to data or variables. For example,
20075 @code{call}, @code{find}, and @code{print} are in this category. Type
20076 @kbd{help data} at the @value{GDBN} prompt to see a list of commands
20079 @findex COMMAND_STACK
20080 @findex gdb.COMMAND_STACK
20081 @item COMMAND_STACK
20082 The command has to do with manipulation of the stack. For example,
20083 @code{backtrace}, @code{frame}, and @code{return} are in this
20084 category. Type @kbd{help stack} at the @value{GDBN} prompt to see a
20085 list of commands in this category.
20087 @findex COMMAND_FILES
20088 @findex gdb.COMMAND_FILES
20089 @item COMMAND_FILES
20090 This class is used for file-related commands. For example,
20091 @code{file}, @code{list} and @code{section} are in this category.
20092 Type @kbd{help files} at the @value{GDBN} prompt to see a list of
20093 commands in this category.
20095 @findex COMMAND_SUPPORT
20096 @findex gdb.COMMAND_SUPPORT
20097 @item COMMAND_SUPPORT
20098 This should be used for ``support facilities'', generally meaning
20099 things that are useful to the user when interacting with @value{GDBN},
20100 but not related to the state of the inferior. For example,
20101 @code{help}, @code{make}, and @code{shell} are in this category. Type
20102 @kbd{help support} at the @value{GDBN} prompt to see a list of
20103 commands in this category.
20105 @findex COMMAND_STATUS
20106 @findex gdb.COMMAND_STATUS
20107 @item COMMAND_STATUS
20108 The command is an @samp{info}-related command, that is, related to the
20109 state of @value{GDBN} itself. For example, @code{info}, @code{macro},
20110 and @code{show} are in this category. Type @kbd{help status} at the
20111 @value{GDBN} prompt to see a list of commands in this category.
20113 @findex COMMAND_BREAKPOINTS
20114 @findex gdb.COMMAND_BREAKPOINTS
20115 @item COMMAND_BREAKPOINTS
20116 The command has to do with breakpoints. For example, @code{break},
20117 @code{clear}, and @code{delete} are in this category. Type @kbd{help
20118 breakpoints} at the @value{GDBN} prompt to see a list of commands in
20121 @findex COMMAND_TRACEPOINTS
20122 @findex gdb.COMMAND_TRACEPOINTS
20123 @item COMMAND_TRACEPOINTS
20124 The command has to do with tracepoints. For example, @code{trace},
20125 @code{actions}, and @code{tfind} are in this category. Type
20126 @kbd{help tracepoints} at the @value{GDBN} prompt to see a list of
20127 commands in this category.
20129 @findex COMMAND_OBSCURE
20130 @findex gdb.COMMAND_OBSCURE
20131 @item COMMAND_OBSCURE
20132 The command is only used in unusual circumstances, or is not of
20133 general interest to users. For example, @code{checkpoint},
20134 @code{fork}, and @code{stop} are in this category. Type @kbd{help
20135 obscure} at the @value{GDBN} prompt to see a list of commands in this
20138 @findex COMMAND_MAINTENANCE
20139 @findex gdb.COMMAND_MAINTENANCE
20140 @item COMMAND_MAINTENANCE
20141 The command is only useful to @value{GDBN} maintainers. The
20142 @code{maintenance} and @code{flushregs} commands are in this category.
20143 Type @kbd{help internals} at the @value{GDBN} prompt to see a list of
20144 commands in this category.
20147 A new command can use a predefined completion function, either by
20148 specifying it via an argument at initialization, or by returning it
20149 from the @code{complete} method. These predefined completion
20150 constants are all defined in the @code{gdb} module:
20153 @findex COMPLETE_NONE
20154 @findex gdb.COMPLETE_NONE
20155 @item COMPLETE_NONE
20156 This constant means that no completion should be done.
20158 @findex COMPLETE_FILENAME
20159 @findex gdb.COMPLETE_FILENAME
20160 @item COMPLETE_FILENAME
20161 This constant means that filename completion should be performed.
20163 @findex COMPLETE_LOCATION
20164 @findex gdb.COMPLETE_LOCATION
20165 @item COMPLETE_LOCATION
20166 This constant means that location completion should be done.
20167 @xref{Specify Location}.
20169 @findex COMPLETE_COMMAND
20170 @findex gdb.COMPLETE_COMMAND
20171 @item COMPLETE_COMMAND
20172 This constant means that completion should examine @value{GDBN}
20175 @findex COMPLETE_SYMBOL
20176 @findex gdb.COMPLETE_SYMBOL
20177 @item COMPLETE_SYMBOL
20178 This constant means that completion should be done using symbol names
20182 The following code snippet shows how a trivial CLI command can be
20183 implemented in Python:
20186 class HelloWorld (gdb.Command):
20187 """Greet the whole world."""
20189 def __init__ (self):
20190 super (HelloWorld, self).__init__ ("hello-world", gdb.COMMAND_OBSCURE)
20192 def invoke (self, arg, from_tty):
20193 print "Hello, World!"
20198 The last line instantiates the class, and is necessary to trigger the
20199 registration of the command with @value{GDBN}. Depending on how the
20200 Python code is read into @value{GDBN}, you may need to import the
20201 @code{gdb} module explicitly.
20203 @node Functions In Python
20204 @subsubsection Writing new convenience functions
20206 @cindex writing convenience functions
20207 @cindex convenience functions in python
20208 @cindex python convenience functions
20209 @tindex gdb.Function
20211 You can implement new convenience functions (@pxref{Convenience Vars})
20212 in Python. A convenience function is an instance of a subclass of the
20213 class @code{gdb.Function}.
20215 @defmethod Function __init__ name
20216 The initializer for @code{Function} registers the new function with
20217 @value{GDBN}. The argument @var{name} is the name of the function,
20218 a string. The function will be visible to the user as a convenience
20219 variable of type @code{internal function}, whose name is the same as
20220 the given @var{name}.
20222 The documentation for the new function is taken from the documentation
20223 string for the new class.
20226 @defmethod Function invoke @var{*args}
20227 When a convenience function is evaluated, its arguments are converted
20228 to instances of @code{gdb.Value}, and then the function's
20229 @code{invoke} method is called. Note that @value{GDBN} does not
20230 predetermine the arity of convenience functions. Instead, all
20231 available arguments are passed to @code{invoke}, following the
20232 standard Python calling convention. In particular, a convenience
20233 function can have default values for parameters without ill effect.
20235 The return value of this method is used as its value in the enclosing
20236 expression. If an ordinary Python value is returned, it is converted
20237 to a @code{gdb.Value} following the usual rules.
20240 The following code snippet shows how a trivial convenience function can
20241 be implemented in Python:
20244 class Greet (gdb.Function):
20245 """Return string to greet someone.
20246 Takes a name as argument."""
20248 def __init__ (self):
20249 super (Greet, self).__init__ ("greet")
20251 def invoke (self, name):
20252 return "Hello, %s!" % name.string ()
20257 The last line instantiates the class, and is necessary to trigger the
20258 registration of the function with @value{GDBN}. Depending on how the
20259 Python code is read into @value{GDBN}, you may need to import the
20260 @code{gdb} module explicitly.
20262 @node Objfiles In Python
20263 @subsubsection Objfiles In Python
20265 @cindex objfiles in python
20266 @tindex gdb.Objfile
20268 @value{GDBN} loads symbols for an inferior from various
20269 symbol-containing files (@pxref{Files}). These include the primary
20270 executable file, any shared libraries used by the inferior, and any
20271 separate debug info files (@pxref{Separate Debug Files}).
20272 @value{GDBN} calls these symbol-containing files @dfn{objfiles}.
20274 The following objfile-related functions are available in the
20277 @findex gdb.current_objfile
20278 @defun current_objfile
20279 When auto-loading a Python script (@pxref{Auto-loading}), @value{GDBN}
20280 sets the ``current objfile'' to the corresponding objfile. This
20281 function returns the current objfile. If there is no current objfile,
20282 this function returns @code{None}.
20285 @findex gdb.objfiles
20287 Return a sequence of all the objfiles current known to @value{GDBN}.
20288 @xref{Objfiles In Python}.
20291 Each objfile is represented by an instance of the @code{gdb.Objfile}
20294 @defivar Objfile filename
20295 The file name of the objfile as a string.
20298 @defivar Objfile pretty_printers
20299 The @code{pretty_printers} attribute is a list of functions. It is
20300 used to look up pretty-printers. A @code{Value} is passed to each
20301 function in order; if the function returns @code{None}, then the
20302 search continues. Otherwise, the return value should be an object
20303 which is used to format the value. @xref{Pretty Printing}, for more
20307 @node Frames In Python
20308 @subsubsection Acessing inferior stack frames from Python.
20310 @cindex frames in python
20311 When the debugged program stops, @value{GDBN} is able to analyze its call
20312 stack (@pxref{Frames,,Stack frames}). The @code{gdb.Frame} class
20313 represents a frame in the stack. A @code{gdb.Frame} object is only valid
20314 while its corresponding frame exists in the inferior's stack. If you try
20315 to use an invalid frame object, @value{GDBN} will throw a @code{RuntimeError}
20318 Two @code{gdb.Frame} objects can be compared for equality with the @code{==}
20322 (@value{GDBP}) python print gdb.newest_frame() == gdb.selected_frame ()
20326 The following frame-related functions are available in the @code{gdb} module:
20328 @findex gdb.selected_frame
20329 @defun selected_frame
20330 Return the selected frame object. (@pxref{Selection,,Selecting a Frame}).
20333 @defun frame_stop_reason_string reason
20334 Return a string explaining the reason why @value{GDBN} stopped unwinding
20335 frames, as expressed by the given @var{reason} code (an integer, see the
20336 @code{unwind_stop_reason} method further down in this section).
20339 A @code{gdb.Frame} object has the following methods:
20342 @defmethod Frame is_valid
20343 Returns true if the @code{gdb.Frame} object is valid, false if not.
20344 A frame object can become invalid if the frame it refers to doesn't
20345 exist anymore in the inferior. All @code{gdb.Frame} methods will throw
20346 an exception if it is invalid at the time the method is called.
20349 @defmethod Frame name
20350 Returns the function name of the frame, or @code{None} if it can't be
20354 @defmethod Frame type
20355 Returns the type of the frame. The value can be one of
20356 @code{gdb.NORMAL_FRAME}, @code{gdb.DUMMY_FRAME}, @code{gdb.SIGTRAMP_FRAME}
20357 or @code{gdb.SENTINEL_FRAME}.
20360 @defmethod Frame unwind_stop_reason
20361 Return an integer representing the reason why it's not possible to find
20362 more frames toward the outermost frame. Use
20363 @code{gdb.frame_stop_reason_string} to convert the value returned by this
20364 function to a string.
20367 @defmethod Frame pc
20368 Returns the frame's resume address.
20371 @defmethod Frame older
20372 Return the frame that called this frame.
20375 @defmethod Frame newer
20376 Return the frame called by this frame.
20379 @defmethod Frame read_var variable
20380 Return the value of the given variable in this frame. @var{variable} must
20386 @chapter Command Interpreters
20387 @cindex command interpreters
20389 @value{GDBN} supports multiple command interpreters, and some command
20390 infrastructure to allow users or user interface writers to switch
20391 between interpreters or run commands in other interpreters.
20393 @value{GDBN} currently supports two command interpreters, the console
20394 interpreter (sometimes called the command-line interpreter or @sc{cli})
20395 and the machine interface interpreter (or @sc{gdb/mi}). This manual
20396 describes both of these interfaces in great detail.
20398 By default, @value{GDBN} will start with the console interpreter.
20399 However, the user may choose to start @value{GDBN} with another
20400 interpreter by specifying the @option{-i} or @option{--interpreter}
20401 startup options. Defined interpreters include:
20405 @cindex console interpreter
20406 The traditional console or command-line interpreter. This is the most often
20407 used interpreter with @value{GDBN}. With no interpreter specified at runtime,
20408 @value{GDBN} will use this interpreter.
20411 @cindex mi interpreter
20412 The newest @sc{gdb/mi} interface (currently @code{mi2}). Used primarily
20413 by programs wishing to use @value{GDBN} as a backend for a debugger GUI
20414 or an IDE. For more information, see @ref{GDB/MI, ,The @sc{gdb/mi}
20418 @cindex mi2 interpreter
20419 The current @sc{gdb/mi} interface.
20422 @cindex mi1 interpreter
20423 The @sc{gdb/mi} interface included in @value{GDBN} 5.1, 5.2, and 5.3.
20427 @cindex invoke another interpreter
20428 The interpreter being used by @value{GDBN} may not be dynamically
20429 switched at runtime. Although possible, this could lead to a very
20430 precarious situation. Consider an IDE using @sc{gdb/mi}. If a user
20431 enters the command "interpreter-set console" in a console view,
20432 @value{GDBN} would switch to using the console interpreter, rendering
20433 the IDE inoperable!
20435 @kindex interpreter-exec
20436 Although you may only choose a single interpreter at startup, you may execute
20437 commands in any interpreter from the current interpreter using the appropriate
20438 command. If you are running the console interpreter, simply use the
20439 @code{interpreter-exec} command:
20442 interpreter-exec mi "-data-list-register-names"
20445 @sc{gdb/mi} has a similar command, although it is only available in versions of
20446 @value{GDBN} which support @sc{gdb/mi} version 2 (or greater).
20449 @chapter @value{GDBN} Text User Interface
20451 @cindex Text User Interface
20454 * TUI Overview:: TUI overview
20455 * TUI Keys:: TUI key bindings
20456 * TUI Single Key Mode:: TUI single key mode
20457 * TUI Commands:: TUI-specific commands
20458 * TUI Configuration:: TUI configuration variables
20461 The @value{GDBN} Text User Interface (TUI) is a terminal
20462 interface which uses the @code{curses} library to show the source
20463 file, the assembly output, the program registers and @value{GDBN}
20464 commands in separate text windows. The TUI mode is supported only
20465 on platforms where a suitable version of the @code{curses} library
20468 @pindex @value{GDBTUI}
20469 The TUI mode is enabled by default when you invoke @value{GDBN} as
20470 either @samp{@value{GDBTUI}} or @samp{@value{GDBP} -tui}.
20471 You can also switch in and out of TUI mode while @value{GDBN} runs by
20472 using various TUI commands and key bindings, such as @kbd{C-x C-a}.
20473 @xref{TUI Keys, ,TUI Key Bindings}.
20476 @section TUI Overview
20478 In TUI mode, @value{GDBN} can display several text windows:
20482 This window is the @value{GDBN} command window with the @value{GDBN}
20483 prompt and the @value{GDBN} output. The @value{GDBN} input is still
20484 managed using readline.
20487 The source window shows the source file of the program. The current
20488 line and active breakpoints are displayed in this window.
20491 The assembly window shows the disassembly output of the program.
20494 This window shows the processor registers. Registers are highlighted
20495 when their values change.
20498 The source and assembly windows show the current program position
20499 by highlighting the current line and marking it with a @samp{>} marker.
20500 Breakpoints are indicated with two markers. The first marker
20501 indicates the breakpoint type:
20505 Breakpoint which was hit at least once.
20508 Breakpoint which was never hit.
20511 Hardware breakpoint which was hit at least once.
20514 Hardware breakpoint which was never hit.
20517 The second marker indicates whether the breakpoint is enabled or not:
20521 Breakpoint is enabled.
20524 Breakpoint is disabled.
20527 The source, assembly and register windows are updated when the current
20528 thread changes, when the frame changes, or when the program counter
20531 These windows are not all visible at the same time. The command
20532 window is always visible. The others can be arranged in several
20543 source and assembly,
20546 source and registers, or
20549 assembly and registers.
20552 A status line above the command window shows the following information:
20556 Indicates the current @value{GDBN} target.
20557 (@pxref{Targets, ,Specifying a Debugging Target}).
20560 Gives the current process or thread number.
20561 When no process is being debugged, this field is set to @code{No process}.
20564 Gives the current function name for the selected frame.
20565 The name is demangled if demangling is turned on (@pxref{Print Settings}).
20566 When there is no symbol corresponding to the current program counter,
20567 the string @code{??} is displayed.
20570 Indicates the current line number for the selected frame.
20571 When the current line number is not known, the string @code{??} is displayed.
20574 Indicates the current program counter address.
20578 @section TUI Key Bindings
20579 @cindex TUI key bindings
20581 The TUI installs several key bindings in the readline keymaps
20582 (@pxref{Command Line Editing}). The following key bindings
20583 are installed for both TUI mode and the @value{GDBN} standard mode.
20592 Enter or leave the TUI mode. When leaving the TUI mode,
20593 the curses window management stops and @value{GDBN} operates using
20594 its standard mode, writing on the terminal directly. When reentering
20595 the TUI mode, control is given back to the curses windows.
20596 The screen is then refreshed.
20600 Use a TUI layout with only one window. The layout will
20601 either be @samp{source} or @samp{assembly}. When the TUI mode
20602 is not active, it will switch to the TUI mode.
20604 Think of this key binding as the Emacs @kbd{C-x 1} binding.
20608 Use a TUI layout with at least two windows. When the current
20609 layout already has two windows, the next layout with two windows is used.
20610 When a new layout is chosen, one window will always be common to the
20611 previous layout and the new one.
20613 Think of it as the Emacs @kbd{C-x 2} binding.
20617 Change the active window. The TUI associates several key bindings
20618 (like scrolling and arrow keys) with the active window. This command
20619 gives the focus to the next TUI window.
20621 Think of it as the Emacs @kbd{C-x o} binding.
20625 Switch in and out of the TUI SingleKey mode that binds single
20626 keys to @value{GDBN} commands (@pxref{TUI Single Key Mode}).
20629 The following key bindings only work in the TUI mode:
20634 Scroll the active window one page up.
20638 Scroll the active window one page down.
20642 Scroll the active window one line up.
20646 Scroll the active window one line down.
20650 Scroll the active window one column left.
20654 Scroll the active window one column right.
20658 Refresh the screen.
20661 Because the arrow keys scroll the active window in the TUI mode, they
20662 are not available for their normal use by readline unless the command
20663 window has the focus. When another window is active, you must use
20664 other readline key bindings such as @kbd{C-p}, @kbd{C-n}, @kbd{C-b}
20665 and @kbd{C-f} to control the command window.
20667 @node TUI Single Key Mode
20668 @section TUI Single Key Mode
20669 @cindex TUI single key mode
20671 The TUI also provides a @dfn{SingleKey} mode, which binds several
20672 frequently used @value{GDBN} commands to single keys. Type @kbd{C-x s} to
20673 switch into this mode, where the following key bindings are used:
20676 @kindex c @r{(SingleKey TUI key)}
20680 @kindex d @r{(SingleKey TUI key)}
20684 @kindex f @r{(SingleKey TUI key)}
20688 @kindex n @r{(SingleKey TUI key)}
20692 @kindex q @r{(SingleKey TUI key)}
20694 exit the SingleKey mode.
20696 @kindex r @r{(SingleKey TUI key)}
20700 @kindex s @r{(SingleKey TUI key)}
20704 @kindex u @r{(SingleKey TUI key)}
20708 @kindex v @r{(SingleKey TUI key)}
20712 @kindex w @r{(SingleKey TUI key)}
20717 Other keys temporarily switch to the @value{GDBN} command prompt.
20718 The key that was pressed is inserted in the editing buffer so that
20719 it is possible to type most @value{GDBN} commands without interaction
20720 with the TUI SingleKey mode. Once the command is entered the TUI
20721 SingleKey mode is restored. The only way to permanently leave
20722 this mode is by typing @kbd{q} or @kbd{C-x s}.
20726 @section TUI-specific Commands
20727 @cindex TUI commands
20729 The TUI has specific commands to control the text windows.
20730 These commands are always available, even when @value{GDBN} is not in
20731 the TUI mode. When @value{GDBN} is in the standard mode, most
20732 of these commands will automatically switch to the TUI mode.
20737 List and give the size of all displayed windows.
20741 Display the next layout.
20744 Display the previous layout.
20747 Display the source window only.
20750 Display the assembly window only.
20753 Display the source and assembly window.
20756 Display the register window together with the source or assembly window.
20760 Make the next window active for scrolling.
20763 Make the previous window active for scrolling.
20766 Make the source window active for scrolling.
20769 Make the assembly window active for scrolling.
20772 Make the register window active for scrolling.
20775 Make the command window active for scrolling.
20779 Refresh the screen. This is similar to typing @kbd{C-L}.
20781 @item tui reg float
20783 Show the floating point registers in the register window.
20785 @item tui reg general
20786 Show the general registers in the register window.
20789 Show the next register group. The list of register groups as well as
20790 their order is target specific. The predefined register groups are the
20791 following: @code{general}, @code{float}, @code{system}, @code{vector},
20792 @code{all}, @code{save}, @code{restore}.
20794 @item tui reg system
20795 Show the system registers in the register window.
20799 Update the source window and the current execution point.
20801 @item winheight @var{name} +@var{count}
20802 @itemx winheight @var{name} -@var{count}
20804 Change the height of the window @var{name} by @var{count}
20805 lines. Positive counts increase the height, while negative counts
20808 @item tabset @var{nchars}
20810 Set the width of tab stops to be @var{nchars} characters.
20813 @node TUI Configuration
20814 @section TUI Configuration Variables
20815 @cindex TUI configuration variables
20817 Several configuration variables control the appearance of TUI windows.
20820 @item set tui border-kind @var{kind}
20821 @kindex set tui border-kind
20822 Select the border appearance for the source, assembly and register windows.
20823 The possible values are the following:
20826 Use a space character to draw the border.
20829 Use @sc{ascii} characters @samp{+}, @samp{-} and @samp{|} to draw the border.
20832 Use the Alternate Character Set to draw the border. The border is
20833 drawn using character line graphics if the terminal supports them.
20836 @item set tui border-mode @var{mode}
20837 @kindex set tui border-mode
20838 @itemx set tui active-border-mode @var{mode}
20839 @kindex set tui active-border-mode
20840 Select the display attributes for the borders of the inactive windows
20841 or the active window. The @var{mode} can be one of the following:
20844 Use normal attributes to display the border.
20850 Use reverse video mode.
20853 Use half bright mode.
20855 @item half-standout
20856 Use half bright and standout mode.
20859 Use extra bright or bold mode.
20861 @item bold-standout
20862 Use extra bright or bold and standout mode.
20867 @chapter Using @value{GDBN} under @sc{gnu} Emacs
20870 @cindex @sc{gnu} Emacs
20871 A special interface allows you to use @sc{gnu} Emacs to view (and
20872 edit) the source files for the program you are debugging with
20875 To use this interface, use the command @kbd{M-x gdb} in Emacs. Give the
20876 executable file you want to debug as an argument. This command starts
20877 @value{GDBN} as a subprocess of Emacs, with input and output through a newly
20878 created Emacs buffer.
20879 @c (Do not use the @code{-tui} option to run @value{GDBN} from Emacs.)
20881 Running @value{GDBN} under Emacs can be just like running @value{GDBN} normally except for two
20886 All ``terminal'' input and output goes through an Emacs buffer, called
20889 This applies both to @value{GDBN} commands and their output, and to the input
20890 and output done by the program you are debugging.
20892 This is useful because it means that you can copy the text of previous
20893 commands and input them again; you can even use parts of the output
20896 All the facilities of Emacs' Shell mode are available for interacting
20897 with your program. In particular, you can send signals the usual
20898 way---for example, @kbd{C-c C-c} for an interrupt, @kbd{C-c C-z} for a
20902 @value{GDBN} displays source code through Emacs.
20904 Each time @value{GDBN} displays a stack frame, Emacs automatically finds the
20905 source file for that frame and puts an arrow (@samp{=>}) at the
20906 left margin of the current line. Emacs uses a separate buffer for
20907 source display, and splits the screen to show both your @value{GDBN} session
20910 Explicit @value{GDBN} @code{list} or search commands still produce output as
20911 usual, but you probably have no reason to use them from Emacs.
20914 We call this @dfn{text command mode}. Emacs 22.1, and later, also uses
20915 a graphical mode, enabled by default, which provides further buffers
20916 that can control the execution and describe the state of your program.
20917 @xref{GDB Graphical Interface,,, Emacs, The @sc{gnu} Emacs Manual}.
20919 If you specify an absolute file name when prompted for the @kbd{M-x
20920 gdb} argument, then Emacs sets your current working directory to where
20921 your program resides. If you only specify the file name, then Emacs
20922 sets your current working directory to to the directory associated
20923 with the previous buffer. In this case, @value{GDBN} may find your
20924 program by searching your environment's @code{PATH} variable, but on
20925 some operating systems it might not find the source. So, although the
20926 @value{GDBN} input and output session proceeds normally, the auxiliary
20927 buffer does not display the current source and line of execution.
20929 The initial working directory of @value{GDBN} is printed on the top
20930 line of the GUD buffer and this serves as a default for the commands
20931 that specify files for @value{GDBN} to operate on. @xref{Files,
20932 ,Commands to Specify Files}.
20934 By default, @kbd{M-x gdb} calls the program called @file{gdb}. If you
20935 need to call @value{GDBN} by a different name (for example, if you
20936 keep several configurations around, with different names) you can
20937 customize the Emacs variable @code{gud-gdb-command-name} to run the
20940 In the GUD buffer, you can use these special Emacs commands in
20941 addition to the standard Shell mode commands:
20945 Describe the features of Emacs' GUD Mode.
20948 Execute to another source line, like the @value{GDBN} @code{step} command; also
20949 update the display window to show the current file and location.
20952 Execute to next source line in this function, skipping all function
20953 calls, like the @value{GDBN} @code{next} command. Then update the display window
20954 to show the current file and location.
20957 Execute one instruction, like the @value{GDBN} @code{stepi} command; update
20958 display window accordingly.
20961 Execute until exit from the selected stack frame, like the @value{GDBN}
20962 @code{finish} command.
20965 Continue execution of your program, like the @value{GDBN} @code{continue}
20969 Go up the number of frames indicated by the numeric argument
20970 (@pxref{Arguments, , Numeric Arguments, Emacs, The @sc{gnu} Emacs Manual}),
20971 like the @value{GDBN} @code{up} command.
20974 Go down the number of frames indicated by the numeric argument, like the
20975 @value{GDBN} @code{down} command.
20978 In any source file, the Emacs command @kbd{C-x @key{SPC}} (@code{gud-break})
20979 tells @value{GDBN} to set a breakpoint on the source line point is on.
20981 In text command mode, if you type @kbd{M-x speedbar}, Emacs displays a
20982 separate frame which shows a backtrace when the GUD buffer is current.
20983 Move point to any frame in the stack and type @key{RET} to make it
20984 become the current frame and display the associated source in the
20985 source buffer. Alternatively, click @kbd{Mouse-2} to make the
20986 selected frame become the current one. In graphical mode, the
20987 speedbar displays watch expressions.
20989 If you accidentally delete the source-display buffer, an easy way to get
20990 it back is to type the command @code{f} in the @value{GDBN} buffer, to
20991 request a frame display; when you run under Emacs, this recreates
20992 the source buffer if necessary to show you the context of the current
20995 The source files displayed in Emacs are in ordinary Emacs buffers
20996 which are visiting the source files in the usual way. You can edit
20997 the files with these buffers if you wish; but keep in mind that @value{GDBN}
20998 communicates with Emacs in terms of line numbers. If you add or
20999 delete lines from the text, the line numbers that @value{GDBN} knows cease
21000 to correspond properly with the code.
21002 A more detailed description of Emacs' interaction with @value{GDBN} is
21003 given in the Emacs manual (@pxref{Debuggers,,, Emacs, The @sc{gnu}
21006 @c The following dropped because Epoch is nonstandard. Reactivate
21007 @c if/when v19 does something similar. ---doc@cygnus.com 19dec1990
21009 @kindex Emacs Epoch environment
21013 Version 18 of @sc{gnu} Emacs has a built-in window system
21014 called the @code{epoch}
21015 environment. Users of this environment can use a new command,
21016 @code{inspect} which performs identically to @code{print} except that
21017 each value is printed in its own window.
21022 @chapter The @sc{gdb/mi} Interface
21024 @unnumberedsec Function and Purpose
21026 @cindex @sc{gdb/mi}, its purpose
21027 @sc{gdb/mi} is a line based machine oriented text interface to
21028 @value{GDBN} and is activated by specifying using the
21029 @option{--interpreter} command line option (@pxref{Mode Options}). It
21030 is specifically intended to support the development of systems which
21031 use the debugger as just one small component of a larger system.
21033 This chapter is a specification of the @sc{gdb/mi} interface. It is written
21034 in the form of a reference manual.
21036 Note that @sc{gdb/mi} is still under construction, so some of the
21037 features described below are incomplete and subject to change
21038 (@pxref{GDB/MI Development and Front Ends, , @sc{gdb/mi} Development and Front Ends}).
21040 @unnumberedsec Notation and Terminology
21042 @cindex notational conventions, for @sc{gdb/mi}
21043 This chapter uses the following notation:
21047 @code{|} separates two alternatives.
21050 @code{[ @var{something} ]} indicates that @var{something} is optional:
21051 it may or may not be given.
21054 @code{( @var{group} )*} means that @var{group} inside the parentheses
21055 may repeat zero or more times.
21058 @code{( @var{group} )+} means that @var{group} inside the parentheses
21059 may repeat one or more times.
21062 @code{"@var{string}"} means a literal @var{string}.
21066 @heading Dependencies
21070 * GDB/MI General Design::
21071 * GDB/MI Command Syntax::
21072 * GDB/MI Compatibility with CLI::
21073 * GDB/MI Development and Front Ends::
21074 * GDB/MI Output Records::
21075 * GDB/MI Simple Examples::
21076 * GDB/MI Command Description Format::
21077 * GDB/MI Breakpoint Commands::
21078 * GDB/MI Program Context::
21079 * GDB/MI Thread Commands::
21080 * GDB/MI Program Execution::
21081 * GDB/MI Stack Manipulation::
21082 * GDB/MI Variable Objects::
21083 * GDB/MI Data Manipulation::
21084 * GDB/MI Tracepoint Commands::
21085 * GDB/MI Symbol Query::
21086 * GDB/MI File Commands::
21088 * GDB/MI Kod Commands::
21089 * GDB/MI Memory Overlay Commands::
21090 * GDB/MI Signal Handling Commands::
21092 * GDB/MI Target Manipulation::
21093 * GDB/MI File Transfer Commands::
21094 * GDB/MI Miscellaneous Commands::
21097 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
21098 @node GDB/MI General Design
21099 @section @sc{gdb/mi} General Design
21100 @cindex GDB/MI General Design
21102 Interaction of a @sc{GDB/MI} frontend with @value{GDBN} involves three
21103 parts---commands sent to @value{GDBN}, responses to those commands
21104 and notifications. Each command results in exactly one response,
21105 indicating either successful completion of the command, or an error.
21106 For the commands that do not resume the target, the response contains the
21107 requested information. For the commands that resume the target, the
21108 response only indicates whether the target was successfully resumed.
21109 Notifications is the mechanism for reporting changes in the state of the
21110 target, or in @value{GDBN} state, that cannot conveniently be associated with
21111 a command and reported as part of that command response.
21113 The important examples of notifications are:
21117 Exec notifications. These are used to report changes in
21118 target state---when a target is resumed, or stopped. It would not
21119 be feasible to include this information in response of resuming
21120 commands, because one resume commands can result in multiple events in
21121 different threads. Also, quite some time may pass before any event
21122 happens in the target, while a frontend needs to know whether the resuming
21123 command itself was successfully executed.
21126 Console output, and status notifications. Console output
21127 notifications are used to report output of CLI commands, as well as
21128 diagnostics for other commands. Status notifications are used to
21129 report the progress of a long-running operation. Naturally, including
21130 this information in command response would mean no output is produced
21131 until the command is finished, which is undesirable.
21134 General notifications. Commands may have various side effects on
21135 the @value{GDBN} or target state beyond their official purpose. For example,
21136 a command may change the selected thread. Although such changes can
21137 be included in command response, using notification allows for more
21138 orthogonal frontend design.
21142 There's no guarantee that whenever an MI command reports an error,
21143 @value{GDBN} or the target are in any specific state, and especially,
21144 the state is not reverted to the state before the MI command was
21145 processed. Therefore, whenever an MI command results in an error,
21146 we recommend that the frontend refreshes all the information shown in
21147 the user interface.
21151 * Context management::
21152 * Asynchronous and non-stop modes::
21156 @node Context management
21157 @subsection Context management
21159 In most cases when @value{GDBN} accesses the target, this access is
21160 done in context of a specific thread and frame (@pxref{Frames}).
21161 Often, even when accessing global data, the target requires that a thread
21162 be specified. The CLI interface maintains the selected thread and frame,
21163 and supplies them to target on each command. This is convenient,
21164 because a command line user would not want to specify that information
21165 explicitly on each command, and because user interacts with
21166 @value{GDBN} via a single terminal, so no confusion is possible as
21167 to what thread and frame are the current ones.
21169 In the case of MI, the concept of selected thread and frame is less
21170 useful. First, a frontend can easily remember this information
21171 itself. Second, a graphical frontend can have more than one window,
21172 each one used for debugging a different thread, and the frontend might
21173 want to access additional threads for internal purposes. This
21174 increases the risk that by relying on implicitly selected thread, the
21175 frontend may be operating on a wrong one. Therefore, each MI command
21176 should explicitly specify which thread and frame to operate on. To
21177 make it possible, each MI command accepts the @samp{--thread} and
21178 @samp{--frame} options, the value to each is @value{GDBN} identifier
21179 for thread and frame to operate on.
21181 Usually, each top-level window in a frontend allows the user to select
21182 a thread and a frame, and remembers the user selection for further
21183 operations. However, in some cases @value{GDBN} may suggest that the
21184 current thread be changed. For example, when stopping on a breakpoint
21185 it is reasonable to switch to the thread where breakpoint is hit. For
21186 another example, if the user issues the CLI @samp{thread} command via
21187 the frontend, it is desirable to change the frontend's selected thread to the
21188 one specified by user. @value{GDBN} communicates the suggestion to
21189 change current thread using the @samp{=thread-selected} notification.
21190 No such notification is available for the selected frame at the moment.
21192 Note that historically, MI shares the selected thread with CLI, so
21193 frontends used the @code{-thread-select} to execute commands in the
21194 right context. However, getting this to work right is cumbersome. The
21195 simplest way is for frontend to emit @code{-thread-select} command
21196 before every command. This doubles the number of commands that need
21197 to be sent. The alternative approach is to suppress @code{-thread-select}
21198 if the selected thread in @value{GDBN} is supposed to be identical to the
21199 thread the frontend wants to operate on. However, getting this
21200 optimization right can be tricky. In particular, if the frontend
21201 sends several commands to @value{GDBN}, and one of the commands changes the
21202 selected thread, then the behaviour of subsequent commands will
21203 change. So, a frontend should either wait for response from such
21204 problematic commands, or explicitly add @code{-thread-select} for
21205 all subsequent commands. No frontend is known to do this exactly
21206 right, so it is suggested to just always pass the @samp{--thread} and
21207 @samp{--frame} options.
21209 @node Asynchronous and non-stop modes
21210 @subsection Asynchronous command execution and non-stop mode
21212 On some targets, @value{GDBN} is capable of processing MI commands
21213 even while the target is running. This is called @dfn{asynchronous
21214 command execution} (@pxref{Background Execution}). The frontend may
21215 specify a preferrence for asynchronous execution using the
21216 @code{-gdb-set target-async 1} command, which should be emitted before
21217 either running the executable or attaching to the target. After the
21218 frontend has started the executable or attached to the target, it can
21219 find if asynchronous execution is enabled using the
21220 @code{-list-target-features} command.
21222 Even if @value{GDBN} can accept a command while target is running,
21223 many commands that access the target do not work when the target is
21224 running. Therefore, asynchronous command execution is most useful
21225 when combined with non-stop mode (@pxref{Non-Stop Mode}). Then,
21226 it is possible to examine the state of one thread, while other threads
21229 When a given thread is running, MI commands that try to access the
21230 target in the context of that thread may not work, or may work only on
21231 some targets. In particular, commands that try to operate on thread's
21232 stack will not work, on any target. Commands that read memory, or
21233 modify breakpoints, may work or not work, depending on the target. Note
21234 that even commands that operate on global state, such as @code{print},
21235 @code{set}, and breakpoint commands, still access the target in the
21236 context of a specific thread, so frontend should try to find a
21237 stopped thread and perform the operation on that thread (using the
21238 @samp{--thread} option).
21240 Which commands will work in the context of a running thread is
21241 highly target dependent. However, the two commands
21242 @code{-exec-interrupt}, to stop a thread, and @code{-thread-info},
21243 to find the state of a thread, will always work.
21245 @node Thread groups
21246 @subsection Thread groups
21247 @value{GDBN} may be used to debug several processes at the same time.
21248 On some platfroms, @value{GDBN} may support debugging of several
21249 hardware systems, each one having several cores with several different
21250 processes running on each core. This section describes the MI
21251 mechanism to support such debugging scenarios.
21253 The key observation is that regardless of the structure of the
21254 target, MI can have a global list of threads, because most commands that
21255 accept the @samp{--thread} option do not need to know what process that
21256 thread belongs to. Therefore, it is not necessary to introduce
21257 neither additional @samp{--process} option, nor an notion of the
21258 current process in the MI interface. The only strictly new feature
21259 that is required is the ability to find how the threads are grouped
21262 To allow the user to discover such grouping, and to support arbitrary
21263 hierarchy of machines/cores/processes, MI introduces the concept of a
21264 @dfn{thread group}. Thread group is a collection of threads and other
21265 thread groups. A thread group always has a string identifier, a type,
21266 and may have additional attributes specific to the type. A new
21267 command, @code{-list-thread-groups}, returns the list of top-level
21268 thread groups, which correspond to processes that @value{GDBN} is
21269 debugging at the moment. By passing an identifier of a thread group
21270 to the @code{-list-thread-groups} command, it is possible to obtain
21271 the members of specific thread group.
21273 To allow the user to easily discover processes, and other objects, he
21274 wishes to debug, a concept of @dfn{available thread group} is
21275 introduced. Available thread group is an thread group that
21276 @value{GDBN} is not debugging, but that can be attached to, using the
21277 @code{-target-attach} command. The list of available top-level thread
21278 groups can be obtained using @samp{-list-thread-groups --available}.
21279 In general, the content of a thread group may be only retrieved only
21280 after attaching to that thread group.
21282 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
21283 @node GDB/MI Command Syntax
21284 @section @sc{gdb/mi} Command Syntax
21287 * GDB/MI Input Syntax::
21288 * GDB/MI Output Syntax::
21291 @node GDB/MI Input Syntax
21292 @subsection @sc{gdb/mi} Input Syntax
21294 @cindex input syntax for @sc{gdb/mi}
21295 @cindex @sc{gdb/mi}, input syntax
21297 @item @var{command} @expansion{}
21298 @code{@var{cli-command} | @var{mi-command}}
21300 @item @var{cli-command} @expansion{}
21301 @code{[ @var{token} ] @var{cli-command} @var{nl}}, where
21302 @var{cli-command} is any existing @value{GDBN} CLI command.
21304 @item @var{mi-command} @expansion{}
21305 @code{[ @var{token} ] "-" @var{operation} ( " " @var{option} )*
21306 @code{[} " --" @code{]} ( " " @var{parameter} )* @var{nl}}
21308 @item @var{token} @expansion{}
21309 "any sequence of digits"
21311 @item @var{option} @expansion{}
21312 @code{"-" @var{parameter} [ " " @var{parameter} ]}
21314 @item @var{parameter} @expansion{}
21315 @code{@var{non-blank-sequence} | @var{c-string}}
21317 @item @var{operation} @expansion{}
21318 @emph{any of the operations described in this chapter}
21320 @item @var{non-blank-sequence} @expansion{}
21321 @emph{anything, provided it doesn't contain special characters such as
21322 "-", @var{nl}, """ and of course " "}
21324 @item @var{c-string} @expansion{}
21325 @code{""" @var{seven-bit-iso-c-string-content} """}
21327 @item @var{nl} @expansion{}
21336 The CLI commands are still handled by the @sc{mi} interpreter; their
21337 output is described below.
21340 The @code{@var{token}}, when present, is passed back when the command
21344 Some @sc{mi} commands accept optional arguments as part of the parameter
21345 list. Each option is identified by a leading @samp{-} (dash) and may be
21346 followed by an optional argument parameter. Options occur first in the
21347 parameter list and can be delimited from normal parameters using
21348 @samp{--} (this is useful when some parameters begin with a dash).
21355 We want easy access to the existing CLI syntax (for debugging).
21358 We want it to be easy to spot a @sc{mi} operation.
21361 @node GDB/MI Output Syntax
21362 @subsection @sc{gdb/mi} Output Syntax
21364 @cindex output syntax of @sc{gdb/mi}
21365 @cindex @sc{gdb/mi}, output syntax
21366 The output from @sc{gdb/mi} consists of zero or more out-of-band records
21367 followed, optionally, by a single result record. This result record
21368 is for the most recent command. The sequence of output records is
21369 terminated by @samp{(gdb)}.
21371 If an input command was prefixed with a @code{@var{token}} then the
21372 corresponding output for that command will also be prefixed by that same
21376 @item @var{output} @expansion{}
21377 @code{( @var{out-of-band-record} )* [ @var{result-record} ] "(gdb)" @var{nl}}
21379 @item @var{result-record} @expansion{}
21380 @code{ [ @var{token} ] "^" @var{result-class} ( "," @var{result} )* @var{nl}}
21382 @item @var{out-of-band-record} @expansion{}
21383 @code{@var{async-record} | @var{stream-record}}
21385 @item @var{async-record} @expansion{}
21386 @code{@var{exec-async-output} | @var{status-async-output} | @var{notify-async-output}}
21388 @item @var{exec-async-output} @expansion{}
21389 @code{[ @var{token} ] "*" @var{async-output}}
21391 @item @var{status-async-output} @expansion{}
21392 @code{[ @var{token} ] "+" @var{async-output}}
21394 @item @var{notify-async-output} @expansion{}
21395 @code{[ @var{token} ] "=" @var{async-output}}
21397 @item @var{async-output} @expansion{}
21398 @code{@var{async-class} ( "," @var{result} )* @var{nl}}
21400 @item @var{result-class} @expansion{}
21401 @code{"done" | "running" | "connected" | "error" | "exit"}
21403 @item @var{async-class} @expansion{}
21404 @code{"stopped" | @var{others}} (where @var{others} will be added
21405 depending on the needs---this is still in development).
21407 @item @var{result} @expansion{}
21408 @code{ @var{variable} "=" @var{value}}
21410 @item @var{variable} @expansion{}
21411 @code{ @var{string} }
21413 @item @var{value} @expansion{}
21414 @code{ @var{const} | @var{tuple} | @var{list} }
21416 @item @var{const} @expansion{}
21417 @code{@var{c-string}}
21419 @item @var{tuple} @expansion{}
21420 @code{ "@{@}" | "@{" @var{result} ( "," @var{result} )* "@}" }
21422 @item @var{list} @expansion{}
21423 @code{ "[]" | "[" @var{value} ( "," @var{value} )* "]" | "["
21424 @var{result} ( "," @var{result} )* "]" }
21426 @item @var{stream-record} @expansion{}
21427 @code{@var{console-stream-output} | @var{target-stream-output} | @var{log-stream-output}}
21429 @item @var{console-stream-output} @expansion{}
21430 @code{"~" @var{c-string}}
21432 @item @var{target-stream-output} @expansion{}
21433 @code{"@@" @var{c-string}}
21435 @item @var{log-stream-output} @expansion{}
21436 @code{"&" @var{c-string}}
21438 @item @var{nl} @expansion{}
21441 @item @var{token} @expansion{}
21442 @emph{any sequence of digits}.
21450 All output sequences end in a single line containing a period.
21453 The @code{@var{token}} is from the corresponding request. Note that
21454 for all async output, while the token is allowed by the grammar and
21455 may be output by future versions of @value{GDBN} for select async
21456 output messages, it is generally omitted. Frontends should treat
21457 all async output as reporting general changes in the state of the
21458 target and there should be no need to associate async output to any
21462 @cindex status output in @sc{gdb/mi}
21463 @var{status-async-output} contains on-going status information about the
21464 progress of a slow operation. It can be discarded. All status output is
21465 prefixed by @samp{+}.
21468 @cindex async output in @sc{gdb/mi}
21469 @var{exec-async-output} contains asynchronous state change on the target
21470 (stopped, started, disappeared). All async output is prefixed by
21474 @cindex notify output in @sc{gdb/mi}
21475 @var{notify-async-output} contains supplementary information that the
21476 client should handle (e.g., a new breakpoint information). All notify
21477 output is prefixed by @samp{=}.
21480 @cindex console output in @sc{gdb/mi}
21481 @var{console-stream-output} is output that should be displayed as is in the
21482 console. It is the textual response to a CLI command. All the console
21483 output is prefixed by @samp{~}.
21486 @cindex target output in @sc{gdb/mi}
21487 @var{target-stream-output} is the output produced by the target program.
21488 All the target output is prefixed by @samp{@@}.
21491 @cindex log output in @sc{gdb/mi}
21492 @var{log-stream-output} is output text coming from @value{GDBN}'s internals, for
21493 instance messages that should be displayed as part of an error log. All
21494 the log output is prefixed by @samp{&}.
21497 @cindex list output in @sc{gdb/mi}
21498 New @sc{gdb/mi} commands should only output @var{lists} containing
21504 @xref{GDB/MI Stream Records, , @sc{gdb/mi} Stream Records}, for more
21505 details about the various output records.
21507 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
21508 @node GDB/MI Compatibility with CLI
21509 @section @sc{gdb/mi} Compatibility with CLI
21511 @cindex compatibility, @sc{gdb/mi} and CLI
21512 @cindex @sc{gdb/mi}, compatibility with CLI
21514 For the developers convenience CLI commands can be entered directly,
21515 but there may be some unexpected behaviour. For example, commands
21516 that query the user will behave as if the user replied yes, breakpoint
21517 command lists are not executed and some CLI commands, such as
21518 @code{if}, @code{when} and @code{define}, prompt for further input with
21519 @samp{>}, which is not valid MI output.
21521 This feature may be removed at some stage in the future and it is
21522 recommended that front ends use the @code{-interpreter-exec} command
21523 (@pxref{-interpreter-exec}).
21525 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
21526 @node GDB/MI Development and Front Ends
21527 @section @sc{gdb/mi} Development and Front Ends
21528 @cindex @sc{gdb/mi} development
21530 The application which takes the MI output and presents the state of the
21531 program being debugged to the user is called a @dfn{front end}.
21533 Although @sc{gdb/mi} is still incomplete, it is currently being used
21534 by a variety of front ends to @value{GDBN}. This makes it difficult
21535 to introduce new functionality without breaking existing usage. This
21536 section tries to minimize the problems by describing how the protocol
21539 Some changes in MI need not break a carefully designed front end, and
21540 for these the MI version will remain unchanged. The following is a
21541 list of changes that may occur within one level, so front ends should
21542 parse MI output in a way that can handle them:
21546 New MI commands may be added.
21549 New fields may be added to the output of any MI command.
21552 The range of values for fields with specified values, e.g.,
21553 @code{in_scope} (@pxref{-var-update}) may be extended.
21555 @c The format of field's content e.g type prefix, may change so parse it
21556 @c at your own risk. Yes, in general?
21558 @c The order of fields may change? Shouldn't really matter but it might
21559 @c resolve inconsistencies.
21562 If the changes are likely to break front ends, the MI version level
21563 will be increased by one. This will allow the front end to parse the
21564 output according to the MI version. Apart from mi0, new versions of
21565 @value{GDBN} will not support old versions of MI and it will be the
21566 responsibility of the front end to work with the new one.
21568 @c Starting with mi3, add a new command -mi-version that prints the MI
21571 The best way to avoid unexpected changes in MI that might break your front
21572 end is to make your project known to @value{GDBN} developers and
21573 follow development on @email{gdb@@sourceware.org} and
21574 @email{gdb-patches@@sourceware.org}.
21575 @cindex mailing lists
21577 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
21578 @node GDB/MI Output Records
21579 @section @sc{gdb/mi} Output Records
21582 * GDB/MI Result Records::
21583 * GDB/MI Stream Records::
21584 * GDB/MI Async Records::
21585 * GDB/MI Frame Information::
21588 @node GDB/MI Result Records
21589 @subsection @sc{gdb/mi} Result Records
21591 @cindex result records in @sc{gdb/mi}
21592 @cindex @sc{gdb/mi}, result records
21593 In addition to a number of out-of-band notifications, the response to a
21594 @sc{gdb/mi} command includes one of the following result indications:
21598 @item "^done" [ "," @var{results} ]
21599 The synchronous operation was successful, @code{@var{results}} are the return
21604 @c Is this one correct? Should it be an out-of-band notification?
21605 The asynchronous operation was successfully started. The target is
21610 @value{GDBN} has connected to a remote target.
21612 @item "^error" "," @var{c-string}
21614 The operation failed. The @code{@var{c-string}} contains the corresponding
21619 @value{GDBN} has terminated.
21623 @node GDB/MI Stream Records
21624 @subsection @sc{gdb/mi} Stream Records
21626 @cindex @sc{gdb/mi}, stream records
21627 @cindex stream records in @sc{gdb/mi}
21628 @value{GDBN} internally maintains a number of output streams: the console, the
21629 target, and the log. The output intended for each of these streams is
21630 funneled through the @sc{gdb/mi} interface using @dfn{stream records}.
21632 Each stream record begins with a unique @dfn{prefix character} which
21633 identifies its stream (@pxref{GDB/MI Output Syntax, , @sc{gdb/mi} Output
21634 Syntax}). In addition to the prefix, each stream record contains a
21635 @code{@var{string-output}}. This is either raw text (with an implicit new
21636 line) or a quoted C string (which does not contain an implicit newline).
21639 @item "~" @var{string-output}
21640 The console output stream contains text that should be displayed in the
21641 CLI console window. It contains the textual responses to CLI commands.
21643 @item "@@" @var{string-output}
21644 The target output stream contains any textual output from the running
21645 target. This is only present when GDB's event loop is truly
21646 asynchronous, which is currently only the case for remote targets.
21648 @item "&" @var{string-output}
21649 The log stream contains debugging messages being produced by @value{GDBN}'s
21653 @node GDB/MI Async Records
21654 @subsection @sc{gdb/mi} Async Records
21656 @cindex async records in @sc{gdb/mi}
21657 @cindex @sc{gdb/mi}, async records
21658 @dfn{Async} records are used to notify the @sc{gdb/mi} client of
21659 additional changes that have occurred. Those changes can either be a
21660 consequence of @sc{gdb/mi} commands (e.g., a breakpoint modified) or a result of
21661 target activity (e.g., target stopped).
21663 The following is the list of possible async records:
21667 @item *running,thread-id="@var{thread}"
21668 The target is now running. The @var{thread} field tells which
21669 specific thread is now running, and can be @samp{all} if all threads
21670 are running. The frontend should assume that no interaction with a
21671 running thread is possible after this notification is produced.
21672 The frontend should not assume that this notification is output
21673 only once for any command. @value{GDBN} may emit this notification
21674 several times, either for different threads, because it cannot resume
21675 all threads together, or even for a single thread, if the thread must
21676 be stepped though some code before letting it run freely.
21678 @item *stopped,reason="@var{reason}",thread-id="@var{id}",stopped-threads="@var{stopped}"
21679 The target has stopped. The @var{reason} field can have one of the
21683 @item breakpoint-hit
21684 A breakpoint was reached.
21685 @item watchpoint-trigger
21686 A watchpoint was triggered.
21687 @item read-watchpoint-trigger
21688 A read watchpoint was triggered.
21689 @item access-watchpoint-trigger
21690 An access watchpoint was triggered.
21691 @item function-finished
21692 An -exec-finish or similar CLI command was accomplished.
21693 @item location-reached
21694 An -exec-until or similar CLI command was accomplished.
21695 @item watchpoint-scope
21696 A watchpoint has gone out of scope.
21697 @item end-stepping-range
21698 An -exec-next, -exec-next-instruction, -exec-step, -exec-step-instruction or
21699 similar CLI command was accomplished.
21700 @item exited-signalled
21701 The inferior exited because of a signal.
21703 The inferior exited.
21704 @item exited-normally
21705 The inferior exited normally.
21706 @item signal-received
21707 A signal was received by the inferior.
21710 The @var{id} field identifies the thread that directly caused the stop
21711 -- for example by hitting a breakpoint. Depending on whether all-stop
21712 mode is in effect (@pxref{All-Stop Mode}), @value{GDBN} may either
21713 stop all threads, or only the thread that directly triggered the stop.
21714 If all threads are stopped, the @var{stopped} field will have the
21715 value of @code{"all"}. Otherwise, the value of the @var{stopped}
21716 field will be a list of thread identifiers. Presently, this list will
21717 always include a single thread, but frontend should be prepared to see
21718 several threads in the list.
21720 @item =thread-group-created,id="@var{id}"
21721 @itemx =thread-group-exited,id="@var{id}"
21722 A thread thread group either was attached to, or has exited/detached
21723 from. The @var{id} field contains the @value{GDBN} identifier of the
21726 @item =thread-created,id="@var{id}",group-id="@var{gid}"
21727 @itemx =thread-exited,id="@var{id}",group-id="@var{gid}"
21728 A thread either was created, or has exited. The @var{id} field
21729 contains the @value{GDBN} identifier of the thread. The @var{gid}
21730 field identifies the thread group this thread belongs to.
21732 @item =thread-selected,id="@var{id}"
21733 Informs that the selected thread was changed as result of the last
21734 command. This notification is not emitted as result of @code{-thread-select}
21735 command but is emitted whenever an MI command that is not documented
21736 to change the selected thread actually changes it. In particular,
21737 invoking, directly or indirectly (via user-defined command), the CLI
21738 @code{thread} command, will generate this notification.
21740 We suggest that in response to this notification, front ends
21741 highlight the selected thread and cause subsequent commands to apply to
21744 @item =library-loaded,...
21745 Reports that a new library file was loaded by the program. This
21746 notification has 4 fields---@var{id}, @var{target-name},
21747 @var{host-name}, and @var{symbols-loaded}. The @var{id} field is an
21748 opaque identifier of the library. For remote debugging case,
21749 @var{target-name} and @var{host-name} fields give the name of the
21750 library file on the target, and on the host respectively. For native
21751 debugging, both those fields have the same value. The
21752 @var{symbols-loaded} field reports if the debug symbols for this
21753 library are loaded.
21755 @item =library-unloaded,...
21756 Reports that a library was unloaded by the program. This notification
21757 has 3 fields---@var{id}, @var{target-name} and @var{host-name} with
21758 the same meaning as for the @code{=library-loaded} notification
21762 @node GDB/MI Frame Information
21763 @subsection @sc{gdb/mi} Frame Information
21765 Response from many MI commands includes an information about stack
21766 frame. This information is a tuple that may have the following
21771 The level of the stack frame. The innermost frame has the level of
21772 zero. This field is always present.
21775 The name of the function corresponding to the frame. This field may
21776 be absent if @value{GDBN} is unable to determine the function name.
21779 The code address for the frame. This field is always present.
21782 The name of the source files that correspond to the frame's code
21783 address. This field may be absent.
21786 The source line corresponding to the frames' code address. This field
21790 The name of the binary file (either executable or shared library) the
21791 corresponds to the frame's code address. This field may be absent.
21796 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
21797 @node GDB/MI Simple Examples
21798 @section Simple Examples of @sc{gdb/mi} Interaction
21799 @cindex @sc{gdb/mi}, simple examples
21801 This subsection presents several simple examples of interaction using
21802 the @sc{gdb/mi} interface. In these examples, @samp{->} means that the
21803 following line is passed to @sc{gdb/mi} as input, while @samp{<-} means
21804 the output received from @sc{gdb/mi}.
21806 Note the line breaks shown in the examples are here only for
21807 readability, they don't appear in the real output.
21809 @subheading Setting a Breakpoint
21811 Setting a breakpoint generates synchronous output which contains detailed
21812 information of the breakpoint.
21815 -> -break-insert main
21816 <- ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
21817 enabled="y",addr="0x08048564",func="main",file="myprog.c",
21818 fullname="/home/nickrob/myprog.c",line="68",times="0"@}
21822 @subheading Program Execution
21824 Program execution generates asynchronous records and MI gives the
21825 reason that execution stopped.
21831 <- *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
21832 frame=@{addr="0x08048564",func="main",
21833 args=[@{name="argc",value="1"@},@{name="argv",value="0xbfc4d4d4"@}],
21834 file="myprog.c",fullname="/home/nickrob/myprog.c",line="68"@}
21839 <- *stopped,reason="exited-normally"
21843 @subheading Quitting @value{GDBN}
21845 Quitting @value{GDBN} just prints the result class @samp{^exit}.
21853 @subheading A Bad Command
21855 Here's what happens if you pass a non-existent command:
21859 <- ^error,msg="Undefined MI command: rubbish"
21864 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
21865 @node GDB/MI Command Description Format
21866 @section @sc{gdb/mi} Command Description Format
21868 The remaining sections describe blocks of commands. Each block of
21869 commands is laid out in a fashion similar to this section.
21871 @subheading Motivation
21873 The motivation for this collection of commands.
21875 @subheading Introduction
21877 A brief introduction to this collection of commands as a whole.
21879 @subheading Commands
21881 For each command in the block, the following is described:
21883 @subsubheading Synopsis
21886 -command @var{args}@dots{}
21889 @subsubheading Result
21891 @subsubheading @value{GDBN} Command
21893 The corresponding @value{GDBN} CLI command(s), if any.
21895 @subsubheading Example
21897 Example(s) formatted for readability. Some of the described commands have
21898 not been implemented yet and these are labeled N.A.@: (not available).
21901 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
21902 @node GDB/MI Breakpoint Commands
21903 @section @sc{gdb/mi} Breakpoint Commands
21905 @cindex breakpoint commands for @sc{gdb/mi}
21906 @cindex @sc{gdb/mi}, breakpoint commands
21907 This section documents @sc{gdb/mi} commands for manipulating
21910 @subheading The @code{-break-after} Command
21911 @findex -break-after
21913 @subsubheading Synopsis
21916 -break-after @var{number} @var{count}
21919 The breakpoint number @var{number} is not in effect until it has been
21920 hit @var{count} times. To see how this is reflected in the output of
21921 the @samp{-break-list} command, see the description of the
21922 @samp{-break-list} command below.
21924 @subsubheading @value{GDBN} Command
21926 The corresponding @value{GDBN} command is @samp{ignore}.
21928 @subsubheading Example
21933 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
21934 enabled="y",addr="0x000100d0",func="main",file="hello.c",
21935 fullname="/home/foo/hello.c",line="5",times="0"@}
21942 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
21943 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
21944 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
21945 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
21946 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
21947 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
21948 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
21949 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
21950 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
21951 line="5",times="0",ignore="3"@}]@}
21956 @subheading The @code{-break-catch} Command
21957 @findex -break-catch
21960 @subheading The @code{-break-commands} Command
21961 @findex -break-commands
21963 @subsubheading Synopsis
21966 -break-commands @var{number} [ @var{command1} ... @var{commandN} ]
21969 Specifies the CLI commands that should be executed when breakpoint
21970 @var{number} is hit. The parameters @var{command1} to @var{commandN}
21971 are the commands. If no command is specified, any previously-set
21972 commands are cleared. @xref{Break Commands}. Typical use of this
21973 functionality is tracing a program, that is, printing of values of
21974 some variables whenever breakpoint is hit and then continuing.
21976 @subsubheading @value{GDBN} Command
21978 The corresponding @value{GDBN} command is @samp{commands}.
21980 @subsubheading Example
21985 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
21986 enabled="y",addr="0x000100d0",func="main",file="hello.c",
21987 fullname="/home/foo/hello.c",line="5",times="0"@}
21989 -break-commands 1 "print v" "continue"
21994 @subheading The @code{-break-condition} Command
21995 @findex -break-condition
21997 @subsubheading Synopsis
22000 -break-condition @var{number} @var{expr}
22003 Breakpoint @var{number} will stop the program only if the condition in
22004 @var{expr} is true. The condition becomes part of the
22005 @samp{-break-list} output (see the description of the @samp{-break-list}
22008 @subsubheading @value{GDBN} Command
22010 The corresponding @value{GDBN} command is @samp{condition}.
22012 @subsubheading Example
22016 -break-condition 1 1
22020 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
22021 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
22022 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
22023 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
22024 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
22025 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
22026 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
22027 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
22028 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
22029 line="5",cond="1",times="0",ignore="3"@}]@}
22033 @subheading The @code{-break-delete} Command
22034 @findex -break-delete
22036 @subsubheading Synopsis
22039 -break-delete ( @var{breakpoint} )+
22042 Delete the breakpoint(s) whose number(s) are specified in the argument
22043 list. This is obviously reflected in the breakpoint list.
22045 @subsubheading @value{GDBN} Command
22047 The corresponding @value{GDBN} command is @samp{delete}.
22049 @subsubheading Example
22057 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
22058 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
22059 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
22060 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
22061 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
22062 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
22063 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
22068 @subheading The @code{-break-disable} Command
22069 @findex -break-disable
22071 @subsubheading Synopsis
22074 -break-disable ( @var{breakpoint} )+
22077 Disable the named @var{breakpoint}(s). The field @samp{enabled} in the
22078 break list is now set to @samp{n} for the named @var{breakpoint}(s).
22080 @subsubheading @value{GDBN} Command
22082 The corresponding @value{GDBN} command is @samp{disable}.
22084 @subsubheading Example
22092 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
22093 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
22094 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
22095 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
22096 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
22097 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
22098 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
22099 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="n",
22100 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
22101 line="5",times="0"@}]@}
22105 @subheading The @code{-break-enable} Command
22106 @findex -break-enable
22108 @subsubheading Synopsis
22111 -break-enable ( @var{breakpoint} )+
22114 Enable (previously disabled) @var{breakpoint}(s).
22116 @subsubheading @value{GDBN} Command
22118 The corresponding @value{GDBN} command is @samp{enable}.
22120 @subsubheading Example
22128 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
22129 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
22130 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
22131 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
22132 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
22133 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
22134 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
22135 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
22136 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
22137 line="5",times="0"@}]@}
22141 @subheading The @code{-break-info} Command
22142 @findex -break-info
22144 @subsubheading Synopsis
22147 -break-info @var{breakpoint}
22151 Get information about a single breakpoint.
22153 @subsubheading @value{GDBN} Command
22155 The corresponding @value{GDBN} command is @samp{info break @var{breakpoint}}.
22157 @subsubheading Example
22160 @subheading The @code{-break-insert} Command
22161 @findex -break-insert
22163 @subsubheading Synopsis
22166 -break-insert [ -t ] [ -h ] [ -f ] [ -d ]
22167 [ -c @var{condition} ] [ -i @var{ignore-count} ]
22168 [ -p @var{thread} ] [ @var{location} ]
22172 If specified, @var{location}, can be one of:
22179 @item filename:linenum
22180 @item filename:function
22184 The possible optional parameters of this command are:
22188 Insert a temporary breakpoint.
22190 Insert a hardware breakpoint.
22191 @item -c @var{condition}
22192 Make the breakpoint conditional on @var{condition}.
22193 @item -i @var{ignore-count}
22194 Initialize the @var{ignore-count}.
22196 If @var{location} cannot be parsed (for example if it
22197 refers to unknown files or functions), create a pending
22198 breakpoint. Without this flag, @value{GDBN} will report
22199 an error, and won't create a breakpoint, if @var{location}
22202 Create a disabled breakpoint.
22205 @subsubheading Result
22207 The result is in the form:
22210 ^done,bkpt=@{number="@var{number}",type="@var{type}",disp="del"|"keep",
22211 enabled="y"|"n",addr="@var{hex}",func="@var{funcname}",file="@var{filename}",
22212 fullname="@var{full_filename}",line="@var{lineno}",[thread="@var{threadno},]
22213 times="@var{times}"@}
22217 where @var{number} is the @value{GDBN} number for this breakpoint,
22218 @var{funcname} is the name of the function where the breakpoint was
22219 inserted, @var{filename} is the name of the source file which contains
22220 this function, @var{lineno} is the source line number within that file
22221 and @var{times} the number of times that the breakpoint has been hit
22222 (always 0 for -break-insert but may be greater for -break-info or -break-list
22223 which use the same output).
22225 Note: this format is open to change.
22226 @c An out-of-band breakpoint instead of part of the result?
22228 @subsubheading @value{GDBN} Command
22230 The corresponding @value{GDBN} commands are @samp{break}, @samp{tbreak},
22231 @samp{hbreak}, @samp{thbreak}, and @samp{rbreak}.
22233 @subsubheading Example
22238 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",
22239 fullname="/home/foo/recursive2.c,line="4",times="0"@}
22241 -break-insert -t foo
22242 ^done,bkpt=@{number="2",addr="0x00010774",file="recursive2.c",
22243 fullname="/home/foo/recursive2.c,line="11",times="0"@}
22246 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
22247 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
22248 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
22249 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
22250 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
22251 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
22252 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
22253 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
22254 addr="0x0001072c", func="main",file="recursive2.c",
22255 fullname="/home/foo/recursive2.c,"line="4",times="0"@},
22256 bkpt=@{number="2",type="breakpoint",disp="del",enabled="y",
22257 addr="0x00010774",func="foo",file="recursive2.c",
22258 fullname="/home/foo/recursive2.c",line="11",times="0"@}]@}
22260 -break-insert -r foo.*
22261 ~int foo(int, int);
22262 ^done,bkpt=@{number="3",addr="0x00010774",file="recursive2.c,
22263 "fullname="/home/foo/recursive2.c",line="11",times="0"@}
22267 @subheading The @code{-break-list} Command
22268 @findex -break-list
22270 @subsubheading Synopsis
22276 Displays the list of inserted breakpoints, showing the following fields:
22280 number of the breakpoint
22282 type of the breakpoint: @samp{breakpoint} or @samp{watchpoint}
22284 should the breakpoint be deleted or disabled when it is hit: @samp{keep}
22287 is the breakpoint enabled or no: @samp{y} or @samp{n}
22289 memory location at which the breakpoint is set
22291 logical location of the breakpoint, expressed by function name, file
22294 number of times the breakpoint has been hit
22297 If there are no breakpoints or watchpoints, the @code{BreakpointTable}
22298 @code{body} field is an empty list.
22300 @subsubheading @value{GDBN} Command
22302 The corresponding @value{GDBN} command is @samp{info break}.
22304 @subsubheading Example
22309 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
22310 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
22311 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
22312 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
22313 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
22314 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
22315 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
22316 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
22317 addr="0x000100d0",func="main",file="hello.c",line="5",times="0"@},
22318 bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
22319 addr="0x00010114",func="foo",file="hello.c",fullname="/home/foo/hello.c",
22320 line="13",times="0"@}]@}
22324 Here's an example of the result when there are no breakpoints:
22329 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
22330 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
22331 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
22332 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
22333 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
22334 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
22335 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
22340 @subheading The @code{-break-watch} Command
22341 @findex -break-watch
22343 @subsubheading Synopsis
22346 -break-watch [ -a | -r ]
22349 Create a watchpoint. With the @samp{-a} option it will create an
22350 @dfn{access} watchpoint, i.e., a watchpoint that triggers either on a
22351 read from or on a write to the memory location. With the @samp{-r}
22352 option, the watchpoint created is a @dfn{read} watchpoint, i.e., it will
22353 trigger only when the memory location is accessed for reading. Without
22354 either of the options, the watchpoint created is a regular watchpoint,
22355 i.e., it will trigger when the memory location is accessed for writing.
22356 @xref{Set Watchpoints, , Setting Watchpoints}.
22358 Note that @samp{-break-list} will report a single list of watchpoints and
22359 breakpoints inserted.
22361 @subsubheading @value{GDBN} Command
22363 The corresponding @value{GDBN} commands are @samp{watch}, @samp{awatch}, and
22366 @subsubheading Example
22368 Setting a watchpoint on a variable in the @code{main} function:
22373 ^done,wpt=@{number="2",exp="x"@}
22378 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="x"@},
22379 value=@{old="-268439212",new="55"@},
22380 frame=@{func="main",args=[],file="recursive2.c",
22381 fullname="/home/foo/bar/recursive2.c",line="5"@}
22385 Setting a watchpoint on a variable local to a function. @value{GDBN} will stop
22386 the program execution twice: first for the variable changing value, then
22387 for the watchpoint going out of scope.
22392 ^done,wpt=@{number="5",exp="C"@}
22397 *stopped,reason="watchpoint-trigger",
22398 wpt=@{number="5",exp="C"@},value=@{old="-276895068",new="3"@},
22399 frame=@{func="callee4",args=[],
22400 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
22401 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13"@}
22406 *stopped,reason="watchpoint-scope",wpnum="5",
22407 frame=@{func="callee3",args=[@{name="strarg",
22408 value="0x11940 \"A string argument.\""@}],
22409 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
22410 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
22414 Listing breakpoints and watchpoints, at different points in the program
22415 execution. Note that once the watchpoint goes out of scope, it is
22421 ^done,wpt=@{number="2",exp="C"@}
22424 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
22425 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
22426 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
22427 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
22428 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
22429 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
22430 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
22431 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
22432 addr="0x00010734",func="callee4",
22433 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
22434 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c"line="8",times="1"@},
22435 bkpt=@{number="2",type="watchpoint",disp="keep",
22436 enabled="y",addr="",what="C",times="0"@}]@}
22441 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="C"@},
22442 value=@{old="-276895068",new="3"@},
22443 frame=@{func="callee4",args=[],
22444 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
22445 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13"@}
22448 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
22449 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
22450 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
22451 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
22452 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
22453 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
22454 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
22455 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
22456 addr="0x00010734",func="callee4",
22457 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
22458 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",times="1"@},
22459 bkpt=@{number="2",type="watchpoint",disp="keep",
22460 enabled="y",addr="",what="C",times="-5"@}]@}
22464 ^done,reason="watchpoint-scope",wpnum="2",
22465 frame=@{func="callee3",args=[@{name="strarg",
22466 value="0x11940 \"A string argument.\""@}],
22467 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
22468 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
22471 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
22472 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
22473 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
22474 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
22475 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
22476 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
22477 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
22478 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
22479 addr="0x00010734",func="callee4",
22480 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
22481 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",
22486 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
22487 @node GDB/MI Program Context
22488 @section @sc{gdb/mi} Program Context
22490 @subheading The @code{-exec-arguments} Command
22491 @findex -exec-arguments
22494 @subsubheading Synopsis
22497 -exec-arguments @var{args}
22500 Set the inferior program arguments, to be used in the next
22503 @subsubheading @value{GDBN} Command
22505 The corresponding @value{GDBN} command is @samp{set args}.
22507 @subsubheading Example
22511 -exec-arguments -v word
22518 @subheading The @code{-exec-show-arguments} Command
22519 @findex -exec-show-arguments
22521 @subsubheading Synopsis
22524 -exec-show-arguments
22527 Print the arguments of the program.
22529 @subsubheading @value{GDBN} Command
22531 The corresponding @value{GDBN} command is @samp{show args}.
22533 @subsubheading Example
22538 @subheading The @code{-environment-cd} Command
22539 @findex -environment-cd
22541 @subsubheading Synopsis
22544 -environment-cd @var{pathdir}
22547 Set @value{GDBN}'s working directory.
22549 @subsubheading @value{GDBN} Command
22551 The corresponding @value{GDBN} command is @samp{cd}.
22553 @subsubheading Example
22557 -environment-cd /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
22563 @subheading The @code{-environment-directory} Command
22564 @findex -environment-directory
22566 @subsubheading Synopsis
22569 -environment-directory [ -r ] [ @var{pathdir} ]+
22572 Add directories @var{pathdir} to beginning of search path for source files.
22573 If the @samp{-r} option is used, the search path is reset to the default
22574 search path. If directories @var{pathdir} are supplied in addition to the
22575 @samp{-r} option, the search path is first reset and then addition
22577 Multiple directories may be specified, separated by blanks. Specifying
22578 multiple directories in a single command
22579 results in the directories added to the beginning of the
22580 search path in the same order they were presented in the command.
22581 If blanks are needed as
22582 part of a directory name, double-quotes should be used around
22583 the name. In the command output, the path will show up separated
22584 by the system directory-separator character. The directory-separator
22585 character must not be used
22586 in any directory name.
22587 If no directories are specified, the current search path is displayed.
22589 @subsubheading @value{GDBN} Command
22591 The corresponding @value{GDBN} command is @samp{dir}.
22593 @subsubheading Example
22597 -environment-directory /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
22598 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
22600 -environment-directory ""
22601 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
22603 -environment-directory -r /home/jjohnstn/src/gdb /usr/src
22604 ^done,source-path="/home/jjohnstn/src/gdb:/usr/src:$cdir:$cwd"
22606 -environment-directory -r
22607 ^done,source-path="$cdir:$cwd"
22612 @subheading The @code{-environment-path} Command
22613 @findex -environment-path
22615 @subsubheading Synopsis
22618 -environment-path [ -r ] [ @var{pathdir} ]+
22621 Add directories @var{pathdir} to beginning of search path for object files.
22622 If the @samp{-r} option is used, the search path is reset to the original
22623 search path that existed at gdb start-up. If directories @var{pathdir} are
22624 supplied in addition to the
22625 @samp{-r} option, the search path is first reset and then addition
22627 Multiple directories may be specified, separated by blanks. Specifying
22628 multiple directories in a single command
22629 results in the directories added to the beginning of the
22630 search path in the same order they were presented in the command.
22631 If blanks are needed as
22632 part of a directory name, double-quotes should be used around
22633 the name. In the command output, the path will show up separated
22634 by the system directory-separator character. The directory-separator
22635 character must not be used
22636 in any directory name.
22637 If no directories are specified, the current path is displayed.
22640 @subsubheading @value{GDBN} Command
22642 The corresponding @value{GDBN} command is @samp{path}.
22644 @subsubheading Example
22649 ^done,path="/usr/bin"
22651 -environment-path /kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb /bin
22652 ^done,path="/kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb:/bin:/usr/bin"
22654 -environment-path -r /usr/local/bin
22655 ^done,path="/usr/local/bin:/usr/bin"
22660 @subheading The @code{-environment-pwd} Command
22661 @findex -environment-pwd
22663 @subsubheading Synopsis
22669 Show the current working directory.
22671 @subsubheading @value{GDBN} Command
22673 The corresponding @value{GDBN} command is @samp{pwd}.
22675 @subsubheading Example
22680 ^done,cwd="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb"
22684 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
22685 @node GDB/MI Thread Commands
22686 @section @sc{gdb/mi} Thread Commands
22689 @subheading The @code{-thread-info} Command
22690 @findex -thread-info
22692 @subsubheading Synopsis
22695 -thread-info [ @var{thread-id} ]
22698 Reports information about either a specific thread, if
22699 the @var{thread-id} parameter is present, or about all
22700 threads. When printing information about all threads,
22701 also reports the current thread.
22703 @subsubheading @value{GDBN} Command
22705 The @samp{info thread} command prints the same information
22708 @subsubheading Example
22713 @{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
22714 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",args=[]@},state="running"@},
22715 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
22716 frame=@{level="0",addr="0x0804891f",func="foo",args=[@{name="i",value="10"@}],
22717 file="/tmp/a.c",fullname="/tmp/a.c",line="158"@},state="running"@}],
22718 current-thread-id="1"
22722 The @samp{state} field may have the following values:
22726 The thread is stopped. Frame information is available for stopped
22730 The thread is running. There's no frame information for running
22735 @subheading The @code{-thread-list-ids} Command
22736 @findex -thread-list-ids
22738 @subsubheading Synopsis
22744 Produces a list of the currently known @value{GDBN} thread ids. At the
22745 end of the list it also prints the total number of such threads.
22747 This command is retained for historical reasons, the
22748 @code{-thread-info} command should be used instead.
22750 @subsubheading @value{GDBN} Command
22752 Part of @samp{info threads} supplies the same information.
22754 @subsubheading Example
22759 ^done,thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
22760 current-thread-id="1",number-of-threads="3"
22765 @subheading The @code{-thread-select} Command
22766 @findex -thread-select
22768 @subsubheading Synopsis
22771 -thread-select @var{threadnum}
22774 Make @var{threadnum} the current thread. It prints the number of the new
22775 current thread, and the topmost frame for that thread.
22777 This command is deprecated in favor of explicitly using the
22778 @samp{--thread} option to each command.
22780 @subsubheading @value{GDBN} Command
22782 The corresponding @value{GDBN} command is @samp{thread}.
22784 @subsubheading Example
22791 *stopped,reason="end-stepping-range",thread-id="2",line="187",
22792 file="../../../devo/gdb/testsuite/gdb.threads/linux-dp.c"
22796 thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
22797 number-of-threads="3"
22800 ^done,new-thread-id="3",
22801 frame=@{level="0",func="vprintf",
22802 args=[@{name="format",value="0x8048e9c \"%*s%c %d %c\\n\""@},
22803 @{name="arg",value="0x2"@}],file="vprintf.c",line="31"@}
22807 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
22808 @node GDB/MI Program Execution
22809 @section @sc{gdb/mi} Program Execution
22811 These are the asynchronous commands which generate the out-of-band
22812 record @samp{*stopped}. Currently @value{GDBN} only really executes
22813 asynchronously with remote targets and this interaction is mimicked in
22816 @subheading The @code{-exec-continue} Command
22817 @findex -exec-continue
22819 @subsubheading Synopsis
22822 -exec-continue [--all|--thread-group N]
22825 Resumes the execution of the inferior program until a breakpoint is
22826 encountered, or until the inferior exits. In all-stop mode
22827 (@pxref{All-Stop Mode}), may resume only one thread, or all threads,
22828 depending on the value of the @samp{scheduler-locking} variable. In
22829 non-stop mode (@pxref{Non-Stop Mode}), if the @samp{--all} is not
22830 specified, only the thread specified with the @samp{--thread} option
22831 (or current thread, if no @samp{--thread} is provided) is resumed. If
22832 @samp{--all} is specified, all threads will be resumed. The
22833 @samp{--all} option is ignored in all-stop mode. If the
22834 @samp{--thread-group} options is specified, then all threads in that
22835 thread group are resumed.
22837 @subsubheading @value{GDBN} Command
22839 The corresponding @value{GDBN} corresponding is @samp{continue}.
22841 @subsubheading Example
22848 *stopped,reason="breakpoint-hit",disp="keep",bkptno="2",frame=@{
22849 func="foo",args=[],file="hello.c",fullname="/home/foo/bar/hello.c",
22855 @subheading The @code{-exec-finish} Command
22856 @findex -exec-finish
22858 @subsubheading Synopsis
22864 Resumes the execution of the inferior program until the current
22865 function is exited. Displays the results returned by the function.
22867 @subsubheading @value{GDBN} Command
22869 The corresponding @value{GDBN} command is @samp{finish}.
22871 @subsubheading Example
22873 Function returning @code{void}.
22880 *stopped,reason="function-finished",frame=@{func="main",args=[],
22881 file="hello.c",fullname="/home/foo/bar/hello.c",line="7"@}
22885 Function returning other than @code{void}. The name of the internal
22886 @value{GDBN} variable storing the result is printed, together with the
22893 *stopped,reason="function-finished",frame=@{addr="0x000107b0",func="foo",
22894 args=[@{name="a",value="1"],@{name="b",value="9"@}@},
22895 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
22896 gdb-result-var="$1",return-value="0"
22901 @subheading The @code{-exec-interrupt} Command
22902 @findex -exec-interrupt
22904 @subsubheading Synopsis
22907 -exec-interrupt [--all|--thread-group N]
22910 Interrupts the background execution of the target. Note how the token
22911 associated with the stop message is the one for the execution command
22912 that has been interrupted. The token for the interrupt itself only
22913 appears in the @samp{^done} output. If the user is trying to
22914 interrupt a non-running program, an error message will be printed.
22916 Note that when asynchronous execution is enabled, this command is
22917 asynchronous just like other execution commands. That is, first the
22918 @samp{^done} response will be printed, and the target stop will be
22919 reported after that using the @samp{*stopped} notification.
22921 In non-stop mode, only the context thread is interrupted by default.
22922 All threads will be interrupted if the @samp{--all} option is
22923 specified. If the @samp{--thread-group} option is specified, all
22924 threads in that group will be interrupted.
22926 @subsubheading @value{GDBN} Command
22928 The corresponding @value{GDBN} command is @samp{interrupt}.
22930 @subsubheading Example
22941 111*stopped,signal-name="SIGINT",signal-meaning="Interrupt",
22942 frame=@{addr="0x00010140",func="foo",args=[],file="try.c",
22943 fullname="/home/foo/bar/try.c",line="13"@}
22948 ^error,msg="mi_cmd_exec_interrupt: Inferior not executing."
22952 @subheading The @code{-exec-jump} Command
22955 @subsubheading Synopsis
22958 -exec-jump @var{location}
22961 Resumes execution of the inferior program at the location specified by
22962 parameter. @xref{Specify Location}, for a description of the
22963 different forms of @var{location}.
22965 @subsubheading @value{GDBN} Command
22967 The corresponding @value{GDBN} command is @samp{jump}.
22969 @subsubheading Example
22972 -exec-jump foo.c:10
22973 *running,thread-id="all"
22978 @subheading The @code{-exec-next} Command
22981 @subsubheading Synopsis
22987 Resumes execution of the inferior program, stopping when the beginning
22988 of the next source line is reached.
22990 @subsubheading @value{GDBN} Command
22992 The corresponding @value{GDBN} command is @samp{next}.
22994 @subsubheading Example
23000 *stopped,reason="end-stepping-range",line="8",file="hello.c"
23005 @subheading The @code{-exec-next-instruction} Command
23006 @findex -exec-next-instruction
23008 @subsubheading Synopsis
23011 -exec-next-instruction
23014 Executes one machine instruction. If the instruction is a function
23015 call, continues until the function returns. If the program stops at an
23016 instruction in the middle of a source line, the address will be
23019 @subsubheading @value{GDBN} Command
23021 The corresponding @value{GDBN} command is @samp{nexti}.
23023 @subsubheading Example
23027 -exec-next-instruction
23031 *stopped,reason="end-stepping-range",
23032 addr="0x000100d4",line="5",file="hello.c"
23037 @subheading The @code{-exec-return} Command
23038 @findex -exec-return
23040 @subsubheading Synopsis
23046 Makes current function return immediately. Doesn't execute the inferior.
23047 Displays the new current frame.
23049 @subsubheading @value{GDBN} Command
23051 The corresponding @value{GDBN} command is @samp{return}.
23053 @subsubheading Example
23057 200-break-insert callee4
23058 200^done,bkpt=@{number="1",addr="0x00010734",
23059 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",line="8"@}
23064 000*stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
23065 frame=@{func="callee4",args=[],
23066 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
23067 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8"@}
23073 111^done,frame=@{level="0",func="callee3",
23074 args=[@{name="strarg",
23075 value="0x11940 \"A string argument.\""@}],
23076 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
23077 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
23082 @subheading The @code{-exec-run} Command
23085 @subsubheading Synopsis
23091 Starts execution of the inferior from the beginning. The inferior
23092 executes until either a breakpoint is encountered or the program
23093 exits. In the latter case the output will include an exit code, if
23094 the program has exited exceptionally.
23096 @subsubheading @value{GDBN} Command
23098 The corresponding @value{GDBN} command is @samp{run}.
23100 @subsubheading Examples
23105 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",line="4"@}
23110 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
23111 frame=@{func="main",args=[],file="recursive2.c",
23112 fullname="/home/foo/bar/recursive2.c",line="4"@}
23117 Program exited normally:
23125 *stopped,reason="exited-normally"
23130 Program exited exceptionally:
23138 *stopped,reason="exited",exit-code="01"
23142 Another way the program can terminate is if it receives a signal such as
23143 @code{SIGINT}. In this case, @sc{gdb/mi} displays this:
23147 *stopped,reason="exited-signalled",signal-name="SIGINT",
23148 signal-meaning="Interrupt"
23152 @c @subheading -exec-signal
23155 @subheading The @code{-exec-step} Command
23158 @subsubheading Synopsis
23164 Resumes execution of the inferior program, stopping when the beginning
23165 of the next source line is reached, if the next source line is not a
23166 function call. If it is, stop at the first instruction of the called
23169 @subsubheading @value{GDBN} Command
23171 The corresponding @value{GDBN} command is @samp{step}.
23173 @subsubheading Example
23175 Stepping into a function:
23181 *stopped,reason="end-stepping-range",
23182 frame=@{func="foo",args=[@{name="a",value="10"@},
23183 @{name="b",value="0"@}],file="recursive2.c",
23184 fullname="/home/foo/bar/recursive2.c",line="11"@}
23194 *stopped,reason="end-stepping-range",line="14",file="recursive2.c"
23199 @subheading The @code{-exec-step-instruction} Command
23200 @findex -exec-step-instruction
23202 @subsubheading Synopsis
23205 -exec-step-instruction
23208 Resumes the inferior which executes one machine instruction. The
23209 output, once @value{GDBN} has stopped, will vary depending on whether
23210 we have stopped in the middle of a source line or not. In the former
23211 case, the address at which the program stopped will be printed as
23214 @subsubheading @value{GDBN} Command
23216 The corresponding @value{GDBN} command is @samp{stepi}.
23218 @subsubheading Example
23222 -exec-step-instruction
23226 *stopped,reason="end-stepping-range",
23227 frame=@{func="foo",args=[],file="try.c",
23228 fullname="/home/foo/bar/try.c",line="10"@}
23230 -exec-step-instruction
23234 *stopped,reason="end-stepping-range",
23235 frame=@{addr="0x000100f4",func="foo",args=[],file="try.c",
23236 fullname="/home/foo/bar/try.c",line="10"@}
23241 @subheading The @code{-exec-until} Command
23242 @findex -exec-until
23244 @subsubheading Synopsis
23247 -exec-until [ @var{location} ]
23250 Executes the inferior until the @var{location} specified in the
23251 argument is reached. If there is no argument, the inferior executes
23252 until a source line greater than the current one is reached. The
23253 reason for stopping in this case will be @samp{location-reached}.
23255 @subsubheading @value{GDBN} Command
23257 The corresponding @value{GDBN} command is @samp{until}.
23259 @subsubheading Example
23263 -exec-until recursive2.c:6
23267 *stopped,reason="location-reached",frame=@{func="main",args=[],
23268 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="6"@}
23273 @subheading -file-clear
23274 Is this going away????
23277 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
23278 @node GDB/MI Stack Manipulation
23279 @section @sc{gdb/mi} Stack Manipulation Commands
23282 @subheading The @code{-stack-info-frame} Command
23283 @findex -stack-info-frame
23285 @subsubheading Synopsis
23291 Get info on the selected frame.
23293 @subsubheading @value{GDBN} Command
23295 The corresponding @value{GDBN} command is @samp{info frame} or @samp{frame}
23296 (without arguments).
23298 @subsubheading Example
23303 ^done,frame=@{level="1",addr="0x0001076c",func="callee3",
23304 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
23305 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17"@}
23309 @subheading The @code{-stack-info-depth} Command
23310 @findex -stack-info-depth
23312 @subsubheading Synopsis
23315 -stack-info-depth [ @var{max-depth} ]
23318 Return the depth of the stack. If the integer argument @var{max-depth}
23319 is specified, do not count beyond @var{max-depth} frames.
23321 @subsubheading @value{GDBN} Command
23323 There's no equivalent @value{GDBN} command.
23325 @subsubheading Example
23327 For a stack with frame levels 0 through 11:
23334 -stack-info-depth 4
23337 -stack-info-depth 12
23340 -stack-info-depth 11
23343 -stack-info-depth 13
23348 @subheading The @code{-stack-list-arguments} Command
23349 @findex -stack-list-arguments
23351 @subsubheading Synopsis
23354 -stack-list-arguments @var{show-values}
23355 [ @var{low-frame} @var{high-frame} ]
23358 Display a list of the arguments for the frames between @var{low-frame}
23359 and @var{high-frame} (inclusive). If @var{low-frame} and
23360 @var{high-frame} are not provided, list the arguments for the whole
23361 call stack. If the two arguments are equal, show the single frame
23362 at the corresponding level. It is an error if @var{low-frame} is
23363 larger than the actual number of frames. On the other hand,
23364 @var{high-frame} may be larger than the actual number of frames, in
23365 which case only existing frames will be returned.
23367 The @var{show-values} argument must have a value of 0 or 1. A value of
23368 0 means that only the names of the arguments are listed, a value of 1
23369 means that both names and values of the arguments are printed.
23371 Use of this command to obtain arguments in a single frame is
23372 deprecated in favor of the @samp{-stack-list-variables} command.
23374 @subsubheading @value{GDBN} Command
23376 @value{GDBN} does not have an equivalent command. @code{gdbtk} has a
23377 @samp{gdb_get_args} command which partially overlaps with the
23378 functionality of @samp{-stack-list-arguments}.
23380 @subsubheading Example
23387 frame=@{level="0",addr="0x00010734",func="callee4",
23388 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
23389 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8"@},
23390 frame=@{level="1",addr="0x0001076c",func="callee3",
23391 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
23392 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17"@},
23393 frame=@{level="2",addr="0x0001078c",func="callee2",
23394 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
23395 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="22"@},
23396 frame=@{level="3",addr="0x000107b4",func="callee1",
23397 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
23398 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="27"@},
23399 frame=@{level="4",addr="0x000107e0",func="main",
23400 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
23401 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="32"@}]
23403 -stack-list-arguments 0
23406 frame=@{level="0",args=[]@},
23407 frame=@{level="1",args=[name="strarg"]@},
23408 frame=@{level="2",args=[name="intarg",name="strarg"]@},
23409 frame=@{level="3",args=[name="intarg",name="strarg",name="fltarg"]@},
23410 frame=@{level="4",args=[]@}]
23412 -stack-list-arguments 1
23415 frame=@{level="0",args=[]@},
23417 args=[@{name="strarg",value="0x11940 \"A string argument.\""@}]@},
23418 frame=@{level="2",args=[
23419 @{name="intarg",value="2"@},
23420 @{name="strarg",value="0x11940 \"A string argument.\""@}]@},
23421 @{frame=@{level="3",args=[
23422 @{name="intarg",value="2"@},
23423 @{name="strarg",value="0x11940 \"A string argument.\""@},
23424 @{name="fltarg",value="3.5"@}]@},
23425 frame=@{level="4",args=[]@}]
23427 -stack-list-arguments 0 2 2
23428 ^done,stack-args=[frame=@{level="2",args=[name="intarg",name="strarg"]@}]
23430 -stack-list-arguments 1 2 2
23431 ^done,stack-args=[frame=@{level="2",
23432 args=[@{name="intarg",value="2"@},
23433 @{name="strarg",value="0x11940 \"A string argument.\""@}]@}]
23437 @c @subheading -stack-list-exception-handlers
23440 @subheading The @code{-stack-list-frames} Command
23441 @findex -stack-list-frames
23443 @subsubheading Synopsis
23446 -stack-list-frames [ @var{low-frame} @var{high-frame} ]
23449 List the frames currently on the stack. For each frame it displays the
23454 The frame number, 0 being the topmost frame, i.e., the innermost function.
23456 The @code{$pc} value for that frame.
23460 File name of the source file where the function lives.
23462 Line number corresponding to the @code{$pc}.
23465 If invoked without arguments, this command prints a backtrace for the
23466 whole stack. If given two integer arguments, it shows the frames whose
23467 levels are between the two arguments (inclusive). If the two arguments
23468 are equal, it shows the single frame at the corresponding level. It is
23469 an error if @var{low-frame} is larger than the actual number of
23470 frames. On the other hand, @var{high-frame} may be larger than the
23471 actual number of frames, in which case only existing frames will be returned.
23473 @subsubheading @value{GDBN} Command
23475 The corresponding @value{GDBN} commands are @samp{backtrace} and @samp{where}.
23477 @subsubheading Example
23479 Full stack backtrace:
23485 [frame=@{level="0",addr="0x0001076c",func="foo",
23486 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="11"@},
23487 frame=@{level="1",addr="0x000107a4",func="foo",
23488 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
23489 frame=@{level="2",addr="0x000107a4",func="foo",
23490 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
23491 frame=@{level="3",addr="0x000107a4",func="foo",
23492 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
23493 frame=@{level="4",addr="0x000107a4",func="foo",
23494 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
23495 frame=@{level="5",addr="0x000107a4",func="foo",
23496 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
23497 frame=@{level="6",addr="0x000107a4",func="foo",
23498 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
23499 frame=@{level="7",addr="0x000107a4",func="foo",
23500 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
23501 frame=@{level="8",addr="0x000107a4",func="foo",
23502 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
23503 frame=@{level="9",addr="0x000107a4",func="foo",
23504 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
23505 frame=@{level="10",addr="0x000107a4",func="foo",
23506 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
23507 frame=@{level="11",addr="0x00010738",func="main",
23508 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="4"@}]
23512 Show frames between @var{low_frame} and @var{high_frame}:
23516 -stack-list-frames 3 5
23518 [frame=@{level="3",addr="0x000107a4",func="foo",
23519 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
23520 frame=@{level="4",addr="0x000107a4",func="foo",
23521 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
23522 frame=@{level="5",addr="0x000107a4",func="foo",
23523 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@}]
23527 Show a single frame:
23531 -stack-list-frames 3 3
23533 [frame=@{level="3",addr="0x000107a4",func="foo",
23534 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@}]
23539 @subheading The @code{-stack-list-locals} Command
23540 @findex -stack-list-locals
23542 @subsubheading Synopsis
23545 -stack-list-locals @var{print-values}
23548 Display the local variable names for the selected frame. If
23549 @var{print-values} is 0 or @code{--no-values}, print only the names of
23550 the variables; if it is 1 or @code{--all-values}, print also their
23551 values; and if it is 2 or @code{--simple-values}, print the name,
23552 type and value for simple data types and the name and type for arrays,
23553 structures and unions. In this last case, a frontend can immediately
23554 display the value of simple data types and create variable objects for
23555 other data types when the user wishes to explore their values in
23558 This command is deprecated in favor of the
23559 @samp{-stack-list-variables} command.
23561 @subsubheading @value{GDBN} Command
23563 @samp{info locals} in @value{GDBN}, @samp{gdb_get_locals} in @code{gdbtk}.
23565 @subsubheading Example
23569 -stack-list-locals 0
23570 ^done,locals=[name="A",name="B",name="C"]
23572 -stack-list-locals --all-values
23573 ^done,locals=[@{name="A",value="1"@},@{name="B",value="2"@},
23574 @{name="C",value="@{1, 2, 3@}"@}]
23575 -stack-list-locals --simple-values
23576 ^done,locals=[@{name="A",type="int",value="1"@},
23577 @{name="B",type="int",value="2"@},@{name="C",type="int [3]"@}]
23581 @subheading The @code{-stack-list-variables} Command
23582 @findex -stack-list-variables
23584 @subsubheading Synopsis
23587 -stack-list-variables @var{print-values}
23590 Display the names of local variables and function arguments for the selected frame. If
23591 @var{print-values} is 0 or @code{--no-values}, print only the names of
23592 the variables; if it is 1 or @code{--all-values}, print also their
23593 values; and if it is 2 or @code{--simple-values}, print the name,
23594 type and value for simple data types and the name and type for arrays,
23595 structures and unions.
23597 @subsubheading Example
23601 -stack-list-variables --thread 1 --frame 0 --all-values
23602 ^done,variables=[@{name="x",value="11"@},@{name="s",value="@{a = 1, b = 2@}"@}]
23607 @subheading The @code{-stack-select-frame} Command
23608 @findex -stack-select-frame
23610 @subsubheading Synopsis
23613 -stack-select-frame @var{framenum}
23616 Change the selected frame. Select a different frame @var{framenum} on
23619 This command in deprecated in favor of passing the @samp{--frame}
23620 option to every command.
23622 @subsubheading @value{GDBN} Command
23624 The corresponding @value{GDBN} commands are @samp{frame}, @samp{up},
23625 @samp{down}, @samp{select-frame}, @samp{up-silent}, and @samp{down-silent}.
23627 @subsubheading Example
23631 -stack-select-frame 2
23636 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
23637 @node GDB/MI Variable Objects
23638 @section @sc{gdb/mi} Variable Objects
23642 @subheading Motivation for Variable Objects in @sc{gdb/mi}
23644 For the implementation of a variable debugger window (locals, watched
23645 expressions, etc.), we are proposing the adaptation of the existing code
23646 used by @code{Insight}.
23648 The two main reasons for that are:
23652 It has been proven in practice (it is already on its second generation).
23655 It will shorten development time (needless to say how important it is
23659 The original interface was designed to be used by Tcl code, so it was
23660 slightly changed so it could be used through @sc{gdb/mi}. This section
23661 describes the @sc{gdb/mi} operations that will be available and gives some
23662 hints about their use.
23664 @emph{Note}: In addition to the set of operations described here, we
23665 expect the @sc{gui} implementation of a variable window to require, at
23666 least, the following operations:
23669 @item @code{-gdb-show} @code{output-radix}
23670 @item @code{-stack-list-arguments}
23671 @item @code{-stack-list-locals}
23672 @item @code{-stack-select-frame}
23677 @subheading Introduction to Variable Objects
23679 @cindex variable objects in @sc{gdb/mi}
23681 Variable objects are "object-oriented" MI interface for examining and
23682 changing values of expressions. Unlike some other MI interfaces that
23683 work with expressions, variable objects are specifically designed for
23684 simple and efficient presentation in the frontend. A variable object
23685 is identified by string name. When a variable object is created, the
23686 frontend specifies the expression for that variable object. The
23687 expression can be a simple variable, or it can be an arbitrary complex
23688 expression, and can even involve CPU registers. After creating a
23689 variable object, the frontend can invoke other variable object
23690 operations---for example to obtain or change the value of a variable
23691 object, or to change display format.
23693 Variable objects have hierarchical tree structure. Any variable object
23694 that corresponds to a composite type, such as structure in C, has
23695 a number of child variable objects, for example corresponding to each
23696 element of a structure. A child variable object can itself have
23697 children, recursively. Recursion ends when we reach
23698 leaf variable objects, which always have built-in types. Child variable
23699 objects are created only by explicit request, so if a frontend
23700 is not interested in the children of a particular variable object, no
23701 child will be created.
23703 For a leaf variable object it is possible to obtain its value as a
23704 string, or set the value from a string. String value can be also
23705 obtained for a non-leaf variable object, but it's generally a string
23706 that only indicates the type of the object, and does not list its
23707 contents. Assignment to a non-leaf variable object is not allowed.
23709 A frontend does not need to read the values of all variable objects each time
23710 the program stops. Instead, MI provides an update command that lists all
23711 variable objects whose values has changed since the last update
23712 operation. This considerably reduces the amount of data that must
23713 be transferred to the frontend. As noted above, children variable
23714 objects are created on demand, and only leaf variable objects have a
23715 real value. As result, gdb will read target memory only for leaf
23716 variables that frontend has created.
23718 The automatic update is not always desirable. For example, a frontend
23719 might want to keep a value of some expression for future reference,
23720 and never update it. For another example, fetching memory is
23721 relatively slow for embedded targets, so a frontend might want
23722 to disable automatic update for the variables that are either not
23723 visible on the screen, or ``closed''. This is possible using so
23724 called ``frozen variable objects''. Such variable objects are never
23725 implicitly updated.
23727 Variable objects can be either @dfn{fixed} or @dfn{floating}. For the
23728 fixed variable object, the expression is parsed when the variable
23729 object is created, including associating identifiers to specific
23730 variables. The meaning of expression never changes. For a floating
23731 variable object the values of variables whose names appear in the
23732 expressions are re-evaluated every time in the context of the current
23733 frame. Consider this example:
23738 struct work_state state;
23745 If a fixed variable object for the @code{state} variable is created in
23746 this function, and we enter the recursive call, the the variable
23747 object will report the value of @code{state} in the top-level
23748 @code{do_work} invocation. On the other hand, a floating variable
23749 object will report the value of @code{state} in the current frame.
23751 If an expression specified when creating a fixed variable object
23752 refers to a local variable, the variable object becomes bound to the
23753 thread and frame in which the variable object is created. When such
23754 variable object is updated, @value{GDBN} makes sure that the
23755 thread/frame combination the variable object is bound to still exists,
23756 and re-evaluates the variable object in context of that thread/frame.
23758 The following is the complete set of @sc{gdb/mi} operations defined to
23759 access this functionality:
23761 @multitable @columnfractions .4 .6
23762 @item @strong{Operation}
23763 @tab @strong{Description}
23765 @item @code{-enable-pretty-printing}
23766 @tab enable Python-based pretty-printing
23767 @item @code{-var-create}
23768 @tab create a variable object
23769 @item @code{-var-delete}
23770 @tab delete the variable object and/or its children
23771 @item @code{-var-set-format}
23772 @tab set the display format of this variable
23773 @item @code{-var-show-format}
23774 @tab show the display format of this variable
23775 @item @code{-var-info-num-children}
23776 @tab tells how many children this object has
23777 @item @code{-var-list-children}
23778 @tab return a list of the object's children
23779 @item @code{-var-info-type}
23780 @tab show the type of this variable object
23781 @item @code{-var-info-expression}
23782 @tab print parent-relative expression that this variable object represents
23783 @item @code{-var-info-path-expression}
23784 @tab print full expression that this variable object represents
23785 @item @code{-var-show-attributes}
23786 @tab is this variable editable? does it exist here?
23787 @item @code{-var-evaluate-expression}
23788 @tab get the value of this variable
23789 @item @code{-var-assign}
23790 @tab set the value of this variable
23791 @item @code{-var-update}
23792 @tab update the variable and its children
23793 @item @code{-var-set-frozen}
23794 @tab set frozeness attribute
23795 @item @code{-var-set-update-range}
23796 @tab set range of children to display on update
23799 In the next subsection we describe each operation in detail and suggest
23800 how it can be used.
23802 @subheading Description And Use of Operations on Variable Objects
23804 @subheading The @code{-enable-pretty-printing} Command
23805 @findex -enable-pretty-printing
23808 -enable-pretty-printing
23811 @value{GDBN} allows Python-based visualizers to affect the output of the
23812 MI variable object commands. However, because there was no way to
23813 implement this in a fully backward-compatible way, a front end must
23814 request that this functionality be enabled.
23816 Once enabled, this feature cannot be disabled.
23818 Note that if Python support has not been compiled into @value{GDBN},
23819 this command will still succeed (and do nothing).
23821 This feature is currently (as of @value{GDBN} 7.0) experimental, and
23822 may work differently in future versions of @value{GDBN}.
23824 @subheading The @code{-var-create} Command
23825 @findex -var-create
23827 @subsubheading Synopsis
23830 -var-create @{@var{name} | "-"@}
23831 @{@var{frame-addr} | "*" | "@@"@} @var{expression}
23834 This operation creates a variable object, which allows the monitoring of
23835 a variable, the result of an expression, a memory cell or a CPU
23838 The @var{name} parameter is the string by which the object can be
23839 referenced. It must be unique. If @samp{-} is specified, the varobj
23840 system will generate a string ``varNNNNNN'' automatically. It will be
23841 unique provided that one does not specify @var{name} of that format.
23842 The command fails if a duplicate name is found.
23844 The frame under which the expression should be evaluated can be
23845 specified by @var{frame-addr}. A @samp{*} indicates that the current
23846 frame should be used. A @samp{@@} indicates that a floating variable
23847 object must be created.
23849 @var{expression} is any expression valid on the current language set (must not
23850 begin with a @samp{*}), or one of the following:
23854 @samp{*@var{addr}}, where @var{addr} is the address of a memory cell
23857 @samp{*@var{addr}-@var{addr}} --- a memory address range (TBD)
23860 @samp{$@var{regname}} --- a CPU register name
23863 @cindex dynamic varobj
23864 A varobj's contents may be provided by a Python-based pretty-printer. In this
23865 case the varobj is known as a @dfn{dynamic varobj}. Dynamic varobjs
23866 have slightly different semantics in some cases. If the
23867 @code{-enable-pretty-printing} command is not sent, then @value{GDBN}
23868 will never create a dynamic varobj. This ensures backward
23869 compatibility for existing clients.
23871 @subsubheading Result
23873 This operation returns attributes of the newly-created varobj. These
23878 The name of the varobj.
23881 The number of children of the varobj. This number is not necessarily
23882 reliable for a dynamic varobj. Instead, you must examine the
23883 @samp{has_more} attribute.
23886 The varobj's scalar value. For a varobj whose type is some sort of
23887 aggregate (e.g., a @code{struct}), or for a dynamic varobj, this value
23888 will not be interesting.
23891 The varobj's type. This is a string representation of the type, as
23892 would be printed by the @value{GDBN} CLI.
23895 If a variable object is bound to a specific thread, then this is the
23896 thread's identifier.
23899 For a dynamic varobj, this indicates whether there appear to be any
23900 children available. For a non-dynamic varobj, this will be 0.
23903 This attribute will be present and have the value @samp{1} if the
23904 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
23905 then this attribute will not be present.
23908 A dynamic varobj can supply a display hint to the front end. The
23909 value comes directly from the Python pretty-printer object's
23910 @code{display_hint} method. @xref{Pretty Printing}.
23913 Typical output will look like this:
23916 name="@var{name}",numchild="@var{N}",type="@var{type}",thread-id="@var{M}",
23917 has_more="@var{has_more}"
23921 @subheading The @code{-var-delete} Command
23922 @findex -var-delete
23924 @subsubheading Synopsis
23927 -var-delete [ -c ] @var{name}
23930 Deletes a previously created variable object and all of its children.
23931 With the @samp{-c} option, just deletes the children.
23933 Returns an error if the object @var{name} is not found.
23936 @subheading The @code{-var-set-format} Command
23937 @findex -var-set-format
23939 @subsubheading Synopsis
23942 -var-set-format @var{name} @var{format-spec}
23945 Sets the output format for the value of the object @var{name} to be
23948 @anchor{-var-set-format}
23949 The syntax for the @var{format-spec} is as follows:
23952 @var{format-spec} @expansion{}
23953 @{binary | decimal | hexadecimal | octal | natural@}
23956 The natural format is the default format choosen automatically
23957 based on the variable type (like decimal for an @code{int}, hex
23958 for pointers, etc.).
23960 For a variable with children, the format is set only on the
23961 variable itself, and the children are not affected.
23963 @subheading The @code{-var-show-format} Command
23964 @findex -var-show-format
23966 @subsubheading Synopsis
23969 -var-show-format @var{name}
23972 Returns the format used to display the value of the object @var{name}.
23975 @var{format} @expansion{}
23980 @subheading The @code{-var-info-num-children} Command
23981 @findex -var-info-num-children
23983 @subsubheading Synopsis
23986 -var-info-num-children @var{name}
23989 Returns the number of children of a variable object @var{name}:
23995 Note that this number is not completely reliable for a dynamic varobj.
23996 It will return the current number of children, but more children may
24000 @subheading The @code{-var-list-children} Command
24001 @findex -var-list-children
24003 @subsubheading Synopsis
24006 -var-list-children [@var{print-values}] @var{name} [@var{from} @var{to}]
24008 @anchor{-var-list-children}
24010 Return a list of the children of the specified variable object and
24011 create variable objects for them, if they do not already exist. With
24012 a single argument or if @var{print-values} has a value for of 0 or
24013 @code{--no-values}, print only the names of the variables; if
24014 @var{print-values} is 1 or @code{--all-values}, also print their
24015 values; and if it is 2 or @code{--simple-values} print the name and
24016 value for simple data types and just the name for arrays, structures
24019 @var{from} and @var{to}, if specified, indicate the range of children
24020 to report. If @var{from} or @var{to} is less than zero, the range is
24021 reset and all children will be reported. Otherwise, children starting
24022 at @var{from} (zero-based) and up to and excluding @var{to} will be
24025 If a child range is requested, it will only affect the current call to
24026 @code{-var-list-children}, but not future calls to @code{-var-update}.
24027 For this, you must instead use @code{-var-set-update-range}. The
24028 intent of this approach is to enable a front end to implement any
24029 update approach it likes; for example, scrolling a view may cause the
24030 front end to request more children with @code{-var-list-children}, and
24031 then the front end could call @code{-var-set-update-range} with a
24032 different range to ensure that future updates are restricted to just
24035 For each child the following results are returned:
24040 Name of the variable object created for this child.
24043 The expression to be shown to the user by the front end to designate this child.
24044 For example this may be the name of a structure member.
24046 For a dynamic varobj, this value cannot be used to form an
24047 expression. There is no way to do this at all with a dynamic varobj.
24049 For C/C@t{++} structures there are several pseudo children returned to
24050 designate access qualifiers. For these pseudo children @var{exp} is
24051 @samp{public}, @samp{private}, or @samp{protected}. In this case the
24052 type and value are not present.
24054 A dynamic varobj will not report the access qualifying
24055 pseudo-children, regardless of the language. This information is not
24056 available at all with a dynamic varobj.
24059 Number of children this child has. For a dynamic varobj, this will be
24063 The type of the child.
24066 If values were requested, this is the value.
24069 If this variable object is associated with a thread, this is the thread id.
24070 Otherwise this result is not present.
24073 If the variable object is frozen, this variable will be present with a value of 1.
24076 The result may have its own attributes:
24080 A dynamic varobj can supply a display hint to the front end. The
24081 value comes directly from the Python pretty-printer object's
24082 @code{display_hint} method. @xref{Pretty Printing}.
24085 This is an integer attribute which is nonzero if there are children
24086 remaining after the end of the selected range.
24089 @subsubheading Example
24093 -var-list-children n
24094 ^done,numchild=@var{n},children=[child=@{name=@var{name},exp=@var{exp},
24095 numchild=@var{n},type=@var{type}@},@r{(repeats N times)}]
24097 -var-list-children --all-values n
24098 ^done,numchild=@var{n},children=[child=@{name=@var{name},exp=@var{exp},
24099 numchild=@var{n},value=@var{value},type=@var{type}@},@r{(repeats N times)}]
24103 @subheading The @code{-var-info-type} Command
24104 @findex -var-info-type
24106 @subsubheading Synopsis
24109 -var-info-type @var{name}
24112 Returns the type of the specified variable @var{name}. The type is
24113 returned as a string in the same format as it is output by the
24117 type=@var{typename}
24121 @subheading The @code{-var-info-expression} Command
24122 @findex -var-info-expression
24124 @subsubheading Synopsis
24127 -var-info-expression @var{name}
24130 Returns a string that is suitable for presenting this
24131 variable object in user interface. The string is generally
24132 not valid expression in the current language, and cannot be evaluated.
24134 For example, if @code{a} is an array, and variable object
24135 @code{A} was created for @code{a}, then we'll get this output:
24138 (gdb) -var-info-expression A.1
24139 ^done,lang="C",exp="1"
24143 Here, the values of @code{lang} can be @code{@{"C" | "C++" | "Java"@}}.
24145 Note that the output of the @code{-var-list-children} command also
24146 includes those expressions, so the @code{-var-info-expression} command
24149 @subheading The @code{-var-info-path-expression} Command
24150 @findex -var-info-path-expression
24152 @subsubheading Synopsis
24155 -var-info-path-expression @var{name}
24158 Returns an expression that can be evaluated in the current
24159 context and will yield the same value that a variable object has.
24160 Compare this with the @code{-var-info-expression} command, which
24161 result can be used only for UI presentation. Typical use of
24162 the @code{-var-info-path-expression} command is creating a
24163 watchpoint from a variable object.
24165 This command is currently not valid for children of a dynamic varobj,
24166 and will give an error when invoked on one.
24168 For example, suppose @code{C} is a C@t{++} class, derived from class
24169 @code{Base}, and that the @code{Base} class has a member called
24170 @code{m_size}. Assume a variable @code{c} is has the type of
24171 @code{C} and a variable object @code{C} was created for variable
24172 @code{c}. Then, we'll get this output:
24174 (gdb) -var-info-path-expression C.Base.public.m_size
24175 ^done,path_expr=((Base)c).m_size)
24178 @subheading The @code{-var-show-attributes} Command
24179 @findex -var-show-attributes
24181 @subsubheading Synopsis
24184 -var-show-attributes @var{name}
24187 List attributes of the specified variable object @var{name}:
24190 status=@var{attr} [ ( ,@var{attr} )* ]
24194 where @var{attr} is @code{@{ @{ editable | noneditable @} | TBD @}}.
24196 @subheading The @code{-var-evaluate-expression} Command
24197 @findex -var-evaluate-expression
24199 @subsubheading Synopsis
24202 -var-evaluate-expression [-f @var{format-spec}] @var{name}
24205 Evaluates the expression that is represented by the specified variable
24206 object and returns its value as a string. The format of the string
24207 can be specified with the @samp{-f} option. The possible values of
24208 this option are the same as for @code{-var-set-format}
24209 (@pxref{-var-set-format}). If the @samp{-f} option is not specified,
24210 the current display format will be used. The current display format
24211 can be changed using the @code{-var-set-format} command.
24217 Note that one must invoke @code{-var-list-children} for a variable
24218 before the value of a child variable can be evaluated.
24220 @subheading The @code{-var-assign} Command
24221 @findex -var-assign
24223 @subsubheading Synopsis
24226 -var-assign @var{name} @var{expression}
24229 Assigns the value of @var{expression} to the variable object specified
24230 by @var{name}. The object must be @samp{editable}. If the variable's
24231 value is altered by the assign, the variable will show up in any
24232 subsequent @code{-var-update} list.
24234 @subsubheading Example
24242 ^done,changelist=[@{name="var1",in_scope="true",type_changed="false"@}]
24246 @subheading The @code{-var-update} Command
24247 @findex -var-update
24249 @subsubheading Synopsis
24252 -var-update [@var{print-values}] @{@var{name} | "*"@}
24255 Reevaluate the expressions corresponding to the variable object
24256 @var{name} and all its direct and indirect children, and return the
24257 list of variable objects whose values have changed; @var{name} must
24258 be a root variable object. Here, ``changed'' means that the result of
24259 @code{-var-evaluate-expression} before and after the
24260 @code{-var-update} is different. If @samp{*} is used as the variable
24261 object names, all existing variable objects are updated, except
24262 for frozen ones (@pxref{-var-set-frozen}). The option
24263 @var{print-values} determines whether both names and values, or just
24264 names are printed. The possible values of this option are the same
24265 as for @code{-var-list-children} (@pxref{-var-list-children}). It is
24266 recommended to use the @samp{--all-values} option, to reduce the
24267 number of MI commands needed on each program stop.
24269 With the @samp{*} parameter, if a variable object is bound to a
24270 currently running thread, it will not be updated, without any
24273 If @code{-var-set-update-range} was previously used on a varobj, then
24274 only the selected range of children will be reported.
24276 @code{-var-update} reports all the changed varobjs in a tuple named
24279 Each item in the change list is itself a tuple holding:
24283 The name of the varobj.
24286 If values were requested for this update, then this field will be
24287 present and will hold the value of the varobj.
24290 @anchor{-var-update}
24291 This field is a string which may take one of three values:
24295 The variable object's current value is valid.
24298 The variable object does not currently hold a valid value but it may
24299 hold one in the future if its associated expression comes back into
24303 The variable object no longer holds a valid value.
24304 This can occur when the executable file being debugged has changed,
24305 either through recompilation or by using the @value{GDBN} @code{file}
24306 command. The front end should normally choose to delete these variable
24310 In the future new values may be added to this list so the front should
24311 be prepared for this possibility. @xref{GDB/MI Development and Front Ends, ,@sc{GDB/MI} Development and Front Ends}.
24314 This is only present if the varobj is still valid. If the type
24315 changed, then this will be the string @samp{true}; otherwise it will
24319 If the varobj's type changed, then this field will be present and will
24322 @item new_num_children
24323 For a dynamic varobj, if the number of children changed, or if the
24324 type changed, this will be the new number of children.
24326 The @samp{numchild} field in other varobj responses is generally not
24327 valid for a dynamic varobj -- it will show the number of children that
24328 @value{GDBN} knows about, but because dynamic varobjs lazily
24329 instantiate their children, this will not reflect the number of
24330 children which may be available.
24332 The @samp{new_num_children} attribute only reports changes to the
24333 number of children known by @value{GDBN}. This is the only way to
24334 detect whether an update has removed children (which necessarily can
24335 only happen at the end of the update range).
24338 The display hint, if any.
24341 This is an integer value, which will be 1 if there are more children
24342 available outside the varobj's update range.
24345 This attribute will be present and have the value @samp{1} if the
24346 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
24347 then this attribute will not be present.
24350 If new children were added to a dynamic varobj within the selected
24351 update range (as set by @code{-var-set-update-range}), then they will
24352 be listed in this attribute.
24355 @subsubheading Example
24362 -var-update --all-values var1
24363 ^done,changelist=[@{name="var1",value="3",in_scope="true",
24364 type_changed="false"@}]
24368 @subheading The @code{-var-set-frozen} Command
24369 @findex -var-set-frozen
24370 @anchor{-var-set-frozen}
24372 @subsubheading Synopsis
24375 -var-set-frozen @var{name} @var{flag}
24378 Set the frozenness flag on the variable object @var{name}. The
24379 @var{flag} parameter should be either @samp{1} to make the variable
24380 frozen or @samp{0} to make it unfrozen. If a variable object is
24381 frozen, then neither itself, nor any of its children, are
24382 implicitly updated by @code{-var-update} of
24383 a parent variable or by @code{-var-update *}. Only
24384 @code{-var-update} of the variable itself will update its value and
24385 values of its children. After a variable object is unfrozen, it is
24386 implicitly updated by all subsequent @code{-var-update} operations.
24387 Unfreezing a variable does not update it, only subsequent
24388 @code{-var-update} does.
24390 @subsubheading Example
24394 -var-set-frozen V 1
24399 @subheading The @code{-var-set-update-range} command
24400 @findex -var-set-update-range
24401 @anchor{-var-set-update-range}
24403 @subsubheading Synopsis
24406 -var-set-update-range @var{name} @var{from} @var{to}
24409 Set the range of children to be returned by future invocations of
24410 @code{-var-update}.
24412 @var{from} and @var{to} indicate the range of children to report. If
24413 @var{from} or @var{to} is less than zero, the range is reset and all
24414 children will be reported. Otherwise, children starting at @var{from}
24415 (zero-based) and up to and excluding @var{to} will be reported.
24417 @subsubheading Example
24421 -var-set-update-range V 1 2
24425 @subheading The @code{-var-set-visualizer} command
24426 @findex -var-set-visualizer
24427 @anchor{-var-set-visualizer}
24429 @subsubheading Synopsis
24432 -var-set-visualizer @var{name} @var{visualizer}
24435 Set a visualizer for the variable object @var{name}.
24437 @var{visualizer} is the visualizer to use. The special value
24438 @samp{None} means to disable any visualizer in use.
24440 If not @samp{None}, @var{visualizer} must be a Python expression.
24441 This expression must evaluate to a callable object which accepts a
24442 single argument. @value{GDBN} will call this object with the value of
24443 the varobj @var{name} as an argument (this is done so that the same
24444 Python pretty-printing code can be used for both the CLI and MI).
24445 When called, this object must return an object which conforms to the
24446 pretty-printing interface (@pxref{Pretty Printing}).
24448 The pre-defined function @code{gdb.default_visualizer} may be used to
24449 select a visualizer by following the built-in process
24450 (@pxref{Selecting Pretty-Printers}). This is done automatically when
24451 a varobj is created, and so ordinarily is not needed.
24453 This feature is only available if Python support is enabled. The MI
24454 command @code{-list-features} (@pxref{GDB/MI Miscellaneous Commands})
24455 can be used to check this.
24457 @subsubheading Example
24459 Resetting the visualizer:
24463 -var-set-visualizer V None
24467 Reselecting the default (type-based) visualizer:
24471 -var-set-visualizer V gdb.default_visualizer
24475 Suppose @code{SomeClass} is a visualizer class. A lambda expression
24476 can be used to instantiate this class for a varobj:
24480 -var-set-visualizer V "lambda val: SomeClass()"
24484 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
24485 @node GDB/MI Data Manipulation
24486 @section @sc{gdb/mi} Data Manipulation
24488 @cindex data manipulation, in @sc{gdb/mi}
24489 @cindex @sc{gdb/mi}, data manipulation
24490 This section describes the @sc{gdb/mi} commands that manipulate data:
24491 examine memory and registers, evaluate expressions, etc.
24493 @c REMOVED FROM THE INTERFACE.
24494 @c @subheading -data-assign
24495 @c Change the value of a program variable. Plenty of side effects.
24496 @c @subsubheading GDB Command
24498 @c @subsubheading Example
24501 @subheading The @code{-data-disassemble} Command
24502 @findex -data-disassemble
24504 @subsubheading Synopsis
24508 [ -s @var{start-addr} -e @var{end-addr} ]
24509 | [ -f @var{filename} -l @var{linenum} [ -n @var{lines} ] ]
24517 @item @var{start-addr}
24518 is the beginning address (or @code{$pc})
24519 @item @var{end-addr}
24521 @item @var{filename}
24522 is the name of the file to disassemble
24523 @item @var{linenum}
24524 is the line number to disassemble around
24526 is the number of disassembly lines to be produced. If it is -1,
24527 the whole function will be disassembled, in case no @var{end-addr} is
24528 specified. If @var{end-addr} is specified as a non-zero value, and
24529 @var{lines} is lower than the number of disassembly lines between
24530 @var{start-addr} and @var{end-addr}, only @var{lines} lines are
24531 displayed; if @var{lines} is higher than the number of lines between
24532 @var{start-addr} and @var{end-addr}, only the lines up to @var{end-addr}
24535 is either 0 (meaning only disassembly) or 1 (meaning mixed source and
24539 @subsubheading Result
24541 The output for each instruction is composed of four fields:
24550 Note that whatever included in the instruction field, is not manipulated
24551 directly by @sc{gdb/mi}, i.e., it is not possible to adjust its format.
24553 @subsubheading @value{GDBN} Command
24555 There's no direct mapping from this command to the CLI.
24557 @subsubheading Example
24559 Disassemble from the current value of @code{$pc} to @code{$pc + 20}:
24563 -data-disassemble -s $pc -e "$pc + 20" -- 0
24566 @{address="0x000107c0",func-name="main",offset="4",
24567 inst="mov 2, %o0"@},
24568 @{address="0x000107c4",func-name="main",offset="8",
24569 inst="sethi %hi(0x11800), %o2"@},
24570 @{address="0x000107c8",func-name="main",offset="12",
24571 inst="or %o2, 0x140, %o1\t! 0x11940 <_lib_version+8>"@},
24572 @{address="0x000107cc",func-name="main",offset="16",
24573 inst="sethi %hi(0x11800), %o2"@},
24574 @{address="0x000107d0",func-name="main",offset="20",
24575 inst="or %o2, 0x168, %o4\t! 0x11968 <_lib_version+48>"@}]
24579 Disassemble the whole @code{main} function. Line 32 is part of
24583 -data-disassemble -f basics.c -l 32 -- 0
24585 @{address="0x000107bc",func-name="main",offset="0",
24586 inst="save %sp, -112, %sp"@},
24587 @{address="0x000107c0",func-name="main",offset="4",
24588 inst="mov 2, %o0"@},
24589 @{address="0x000107c4",func-name="main",offset="8",
24590 inst="sethi %hi(0x11800), %o2"@},
24592 @{address="0x0001081c",func-name="main",offset="96",inst="ret "@},
24593 @{address="0x00010820",func-name="main",offset="100",inst="restore "@}]
24597 Disassemble 3 instructions from the start of @code{main}:
24601 -data-disassemble -f basics.c -l 32 -n 3 -- 0
24603 @{address="0x000107bc",func-name="main",offset="0",
24604 inst="save %sp, -112, %sp"@},
24605 @{address="0x000107c0",func-name="main",offset="4",
24606 inst="mov 2, %o0"@},
24607 @{address="0x000107c4",func-name="main",offset="8",
24608 inst="sethi %hi(0x11800), %o2"@}]
24612 Disassemble 3 instructions from the start of @code{main} in mixed mode:
24616 -data-disassemble -f basics.c -l 32 -n 3 -- 1
24618 src_and_asm_line=@{line="31",
24619 file="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb/ \
24620 testsuite/gdb.mi/basics.c",line_asm_insn=[
24621 @{address="0x000107bc",func-name="main",offset="0",
24622 inst="save %sp, -112, %sp"@}]@},
24623 src_and_asm_line=@{line="32",
24624 file="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb/ \
24625 testsuite/gdb.mi/basics.c",line_asm_insn=[
24626 @{address="0x000107c0",func-name="main",offset="4",
24627 inst="mov 2, %o0"@},
24628 @{address="0x000107c4",func-name="main",offset="8",
24629 inst="sethi %hi(0x11800), %o2"@}]@}]
24634 @subheading The @code{-data-evaluate-expression} Command
24635 @findex -data-evaluate-expression
24637 @subsubheading Synopsis
24640 -data-evaluate-expression @var{expr}
24643 Evaluate @var{expr} as an expression. The expression could contain an
24644 inferior function call. The function call will execute synchronously.
24645 If the expression contains spaces, it must be enclosed in double quotes.
24647 @subsubheading @value{GDBN} Command
24649 The corresponding @value{GDBN} commands are @samp{print}, @samp{output}, and
24650 @samp{call}. In @code{gdbtk} only, there's a corresponding
24651 @samp{gdb_eval} command.
24653 @subsubheading Example
24655 In the following example, the numbers that precede the commands are the
24656 @dfn{tokens} described in @ref{GDB/MI Command Syntax, ,@sc{gdb/mi}
24657 Command Syntax}. Notice how @sc{gdb/mi} returns the same tokens in its
24661 211-data-evaluate-expression A
24664 311-data-evaluate-expression &A
24665 311^done,value="0xefffeb7c"
24667 411-data-evaluate-expression A+3
24670 511-data-evaluate-expression "A + 3"
24676 @subheading The @code{-data-list-changed-registers} Command
24677 @findex -data-list-changed-registers
24679 @subsubheading Synopsis
24682 -data-list-changed-registers
24685 Display a list of the registers that have changed.
24687 @subsubheading @value{GDBN} Command
24689 @value{GDBN} doesn't have a direct analog for this command; @code{gdbtk}
24690 has the corresponding command @samp{gdb_changed_register_list}.
24692 @subsubheading Example
24694 On a PPC MBX board:
24702 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",frame=@{
24703 func="main",args=[],file="try.c",fullname="/home/foo/bar/try.c",
24706 -data-list-changed-registers
24707 ^done,changed-registers=["0","1","2","4","5","6","7","8","9",
24708 "10","11","13","14","15","16","17","18","19","20","21","22","23",
24709 "24","25","26","27","28","30","31","64","65","66","67","69"]
24714 @subheading The @code{-data-list-register-names} Command
24715 @findex -data-list-register-names
24717 @subsubheading Synopsis
24720 -data-list-register-names [ ( @var{regno} )+ ]
24723 Show a list of register names for the current target. If no arguments
24724 are given, it shows a list of the names of all the registers. If
24725 integer numbers are given as arguments, it will print a list of the
24726 names of the registers corresponding to the arguments. To ensure
24727 consistency between a register name and its number, the output list may
24728 include empty register names.
24730 @subsubheading @value{GDBN} Command
24732 @value{GDBN} does not have a command which corresponds to
24733 @samp{-data-list-register-names}. In @code{gdbtk} there is a
24734 corresponding command @samp{gdb_regnames}.
24736 @subsubheading Example
24738 For the PPC MBX board:
24741 -data-list-register-names
24742 ^done,register-names=["r0","r1","r2","r3","r4","r5","r6","r7",
24743 "r8","r9","r10","r11","r12","r13","r14","r15","r16","r17","r18",
24744 "r19","r20","r21","r22","r23","r24","r25","r26","r27","r28","r29",
24745 "r30","r31","f0","f1","f2","f3","f4","f5","f6","f7","f8","f9",
24746 "f10","f11","f12","f13","f14","f15","f16","f17","f18","f19","f20",
24747 "f21","f22","f23","f24","f25","f26","f27","f28","f29","f30","f31",
24748 "", "pc","ps","cr","lr","ctr","xer"]
24750 -data-list-register-names 1 2 3
24751 ^done,register-names=["r1","r2","r3"]
24755 @subheading The @code{-data-list-register-values} Command
24756 @findex -data-list-register-values
24758 @subsubheading Synopsis
24761 -data-list-register-values @var{fmt} [ ( @var{regno} )*]
24764 Display the registers' contents. @var{fmt} is the format according to
24765 which the registers' contents are to be returned, followed by an optional
24766 list of numbers specifying the registers to display. A missing list of
24767 numbers indicates that the contents of all the registers must be returned.
24769 Allowed formats for @var{fmt} are:
24786 @subsubheading @value{GDBN} Command
24788 The corresponding @value{GDBN} commands are @samp{info reg}, @samp{info
24789 all-reg}, and (in @code{gdbtk}) @samp{gdb_fetch_registers}.
24791 @subsubheading Example
24793 For a PPC MBX board (note: line breaks are for readability only, they
24794 don't appear in the actual output):
24798 -data-list-register-values r 64 65
24799 ^done,register-values=[@{number="64",value="0xfe00a300"@},
24800 @{number="65",value="0x00029002"@}]
24802 -data-list-register-values x
24803 ^done,register-values=[@{number="0",value="0xfe0043c8"@},
24804 @{number="1",value="0x3fff88"@},@{number="2",value="0xfffffffe"@},
24805 @{number="3",value="0x0"@},@{number="4",value="0xa"@},
24806 @{number="5",value="0x3fff68"@},@{number="6",value="0x3fff58"@},
24807 @{number="7",value="0xfe011e98"@},@{number="8",value="0x2"@},
24808 @{number="9",value="0xfa202820"@},@{number="10",value="0xfa202808"@},
24809 @{number="11",value="0x1"@},@{number="12",value="0x0"@},
24810 @{number="13",value="0x4544"@},@{number="14",value="0xffdfffff"@},
24811 @{number="15",value="0xffffffff"@},@{number="16",value="0xfffffeff"@},
24812 @{number="17",value="0xefffffed"@},@{number="18",value="0xfffffffe"@},
24813 @{number="19",value="0xffffffff"@},@{number="20",value="0xffffffff"@},
24814 @{number="21",value="0xffffffff"@},@{number="22",value="0xfffffff7"@},
24815 @{number="23",value="0xffffffff"@},@{number="24",value="0xffffffff"@},
24816 @{number="25",value="0xffffffff"@},@{number="26",value="0xfffffffb"@},
24817 @{number="27",value="0xffffffff"@},@{number="28",value="0xf7bfffff"@},
24818 @{number="29",value="0x0"@},@{number="30",value="0xfe010000"@},
24819 @{number="31",value="0x0"@},@{number="32",value="0x0"@},
24820 @{number="33",value="0x0"@},@{number="34",value="0x0"@},
24821 @{number="35",value="0x0"@},@{number="36",value="0x0"@},
24822 @{number="37",value="0x0"@},@{number="38",value="0x0"@},
24823 @{number="39",value="0x0"@},@{number="40",value="0x0"@},
24824 @{number="41",value="0x0"@},@{number="42",value="0x0"@},
24825 @{number="43",value="0x0"@},@{number="44",value="0x0"@},
24826 @{number="45",value="0x0"@},@{number="46",value="0x0"@},
24827 @{number="47",value="0x0"@},@{number="48",value="0x0"@},
24828 @{number="49",value="0x0"@},@{number="50",value="0x0"@},
24829 @{number="51",value="0x0"@},@{number="52",value="0x0"@},
24830 @{number="53",value="0x0"@},@{number="54",value="0x0"@},
24831 @{number="55",value="0x0"@},@{number="56",value="0x0"@},
24832 @{number="57",value="0x0"@},@{number="58",value="0x0"@},
24833 @{number="59",value="0x0"@},@{number="60",value="0x0"@},
24834 @{number="61",value="0x0"@},@{number="62",value="0x0"@},
24835 @{number="63",value="0x0"@},@{number="64",value="0xfe00a300"@},
24836 @{number="65",value="0x29002"@},@{number="66",value="0x202f04b5"@},
24837 @{number="67",value="0xfe0043b0"@},@{number="68",value="0xfe00b3e4"@},
24838 @{number="69",value="0x20002b03"@}]
24843 @subheading The @code{-data-read-memory} Command
24844 @findex -data-read-memory
24846 @subsubheading Synopsis
24849 -data-read-memory [ -o @var{byte-offset} ]
24850 @var{address} @var{word-format} @var{word-size}
24851 @var{nr-rows} @var{nr-cols} [ @var{aschar} ]
24858 @item @var{address}
24859 An expression specifying the address of the first memory word to be
24860 read. Complex expressions containing embedded white space should be
24861 quoted using the C convention.
24863 @item @var{word-format}
24864 The format to be used to print the memory words. The notation is the
24865 same as for @value{GDBN}'s @code{print} command (@pxref{Output Formats,
24868 @item @var{word-size}
24869 The size of each memory word in bytes.
24871 @item @var{nr-rows}
24872 The number of rows in the output table.
24874 @item @var{nr-cols}
24875 The number of columns in the output table.
24878 If present, indicates that each row should include an @sc{ascii} dump. The
24879 value of @var{aschar} is used as a padding character when a byte is not a
24880 member of the printable @sc{ascii} character set (printable @sc{ascii}
24881 characters are those whose code is between 32 and 126, inclusively).
24883 @item @var{byte-offset}
24884 An offset to add to the @var{address} before fetching memory.
24887 This command displays memory contents as a table of @var{nr-rows} by
24888 @var{nr-cols} words, each word being @var{word-size} bytes. In total,
24889 @code{@var{nr-rows} * @var{nr-cols} * @var{word-size}} bytes are read
24890 (returned as @samp{total-bytes}). Should less than the requested number
24891 of bytes be returned by the target, the missing words are identified
24892 using @samp{N/A}. The number of bytes read from the target is returned
24893 in @samp{nr-bytes} and the starting address used to read memory in
24896 The address of the next/previous row or page is available in
24897 @samp{next-row} and @samp{prev-row}, @samp{next-page} and
24900 @subsubheading @value{GDBN} Command
24902 The corresponding @value{GDBN} command is @samp{x}. @code{gdbtk} has
24903 @samp{gdb_get_mem} memory read command.
24905 @subsubheading Example
24907 Read six bytes of memory starting at @code{bytes+6} but then offset by
24908 @code{-6} bytes. Format as three rows of two columns. One byte per
24909 word. Display each word in hex.
24913 9-data-read-memory -o -6 -- bytes+6 x 1 3 2
24914 9^done,addr="0x00001390",nr-bytes="6",total-bytes="6",
24915 next-row="0x00001396",prev-row="0x0000138e",next-page="0x00001396",
24916 prev-page="0x0000138a",memory=[
24917 @{addr="0x00001390",data=["0x00","0x01"]@},
24918 @{addr="0x00001392",data=["0x02","0x03"]@},
24919 @{addr="0x00001394",data=["0x04","0x05"]@}]
24923 Read two bytes of memory starting at address @code{shorts + 64} and
24924 display as a single word formatted in decimal.
24928 5-data-read-memory shorts+64 d 2 1 1
24929 5^done,addr="0x00001510",nr-bytes="2",total-bytes="2",
24930 next-row="0x00001512",prev-row="0x0000150e",
24931 next-page="0x00001512",prev-page="0x0000150e",memory=[
24932 @{addr="0x00001510",data=["128"]@}]
24936 Read thirty two bytes of memory starting at @code{bytes+16} and format
24937 as eight rows of four columns. Include a string encoding with @samp{x}
24938 used as the non-printable character.
24942 4-data-read-memory bytes+16 x 1 8 4 x
24943 4^done,addr="0x000013a0",nr-bytes="32",total-bytes="32",
24944 next-row="0x000013c0",prev-row="0x0000139c",
24945 next-page="0x000013c0",prev-page="0x00001380",memory=[
24946 @{addr="0x000013a0",data=["0x10","0x11","0x12","0x13"],ascii="xxxx"@},
24947 @{addr="0x000013a4",data=["0x14","0x15","0x16","0x17"],ascii="xxxx"@},
24948 @{addr="0x000013a8",data=["0x18","0x19","0x1a","0x1b"],ascii="xxxx"@},
24949 @{addr="0x000013ac",data=["0x1c","0x1d","0x1e","0x1f"],ascii="xxxx"@},
24950 @{addr="0x000013b0",data=["0x20","0x21","0x22","0x23"],ascii=" !\"#"@},
24951 @{addr="0x000013b4",data=["0x24","0x25","0x26","0x27"],ascii="$%&'"@},
24952 @{addr="0x000013b8",data=["0x28","0x29","0x2a","0x2b"],ascii="()*+"@},
24953 @{addr="0x000013bc",data=["0x2c","0x2d","0x2e","0x2f"],ascii=",-./"@}]
24957 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
24958 @node GDB/MI Tracepoint Commands
24959 @section @sc{gdb/mi} Tracepoint Commands
24961 The tracepoint commands are not yet implemented.
24963 @c @subheading -trace-actions
24965 @c @subheading -trace-delete
24967 @c @subheading -trace-disable
24969 @c @subheading -trace-dump
24971 @c @subheading -trace-enable
24973 @c @subheading -trace-exists
24975 @c @subheading -trace-find
24977 @c @subheading -trace-frame-number
24979 @c @subheading -trace-info
24981 @c @subheading -trace-insert
24983 @c @subheading -trace-list
24985 @c @subheading -trace-pass-count
24987 @c @subheading -trace-save
24989 @c @subheading -trace-start
24991 @c @subheading -trace-stop
24994 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
24995 @node GDB/MI Symbol Query
24996 @section @sc{gdb/mi} Symbol Query Commands
25000 @subheading The @code{-symbol-info-address} Command
25001 @findex -symbol-info-address
25003 @subsubheading Synopsis
25006 -symbol-info-address @var{symbol}
25009 Describe where @var{symbol} is stored.
25011 @subsubheading @value{GDBN} Command
25013 The corresponding @value{GDBN} command is @samp{info address}.
25015 @subsubheading Example
25019 @subheading The @code{-symbol-info-file} Command
25020 @findex -symbol-info-file
25022 @subsubheading Synopsis
25028 Show the file for the symbol.
25030 @subsubheading @value{GDBN} Command
25032 There's no equivalent @value{GDBN} command. @code{gdbtk} has
25033 @samp{gdb_find_file}.
25035 @subsubheading Example
25039 @subheading The @code{-symbol-info-function} Command
25040 @findex -symbol-info-function
25042 @subsubheading Synopsis
25045 -symbol-info-function
25048 Show which function the symbol lives in.
25050 @subsubheading @value{GDBN} Command
25052 @samp{gdb_get_function} in @code{gdbtk}.
25054 @subsubheading Example
25058 @subheading The @code{-symbol-info-line} Command
25059 @findex -symbol-info-line
25061 @subsubheading Synopsis
25067 Show the core addresses of the code for a source line.
25069 @subsubheading @value{GDBN} Command
25071 The corresponding @value{GDBN} command is @samp{info line}.
25072 @code{gdbtk} has the @samp{gdb_get_line} and @samp{gdb_get_file} commands.
25074 @subsubheading Example
25078 @subheading The @code{-symbol-info-symbol} Command
25079 @findex -symbol-info-symbol
25081 @subsubheading Synopsis
25084 -symbol-info-symbol @var{addr}
25087 Describe what symbol is at location @var{addr}.
25089 @subsubheading @value{GDBN} Command
25091 The corresponding @value{GDBN} command is @samp{info symbol}.
25093 @subsubheading Example
25097 @subheading The @code{-symbol-list-functions} Command
25098 @findex -symbol-list-functions
25100 @subsubheading Synopsis
25103 -symbol-list-functions
25106 List the functions in the executable.
25108 @subsubheading @value{GDBN} Command
25110 @samp{info functions} in @value{GDBN}, @samp{gdb_listfunc} and
25111 @samp{gdb_search} in @code{gdbtk}.
25113 @subsubheading Example
25118 @subheading The @code{-symbol-list-lines} Command
25119 @findex -symbol-list-lines
25121 @subsubheading Synopsis
25124 -symbol-list-lines @var{filename}
25127 Print the list of lines that contain code and their associated program
25128 addresses for the given source filename. The entries are sorted in
25129 ascending PC order.
25131 @subsubheading @value{GDBN} Command
25133 There is no corresponding @value{GDBN} command.
25135 @subsubheading Example
25138 -symbol-list-lines basics.c
25139 ^done,lines=[@{pc="0x08048554",line="7"@},@{pc="0x0804855a",line="8"@}]
25145 @subheading The @code{-symbol-list-types} Command
25146 @findex -symbol-list-types
25148 @subsubheading Synopsis
25154 List all the type names.
25156 @subsubheading @value{GDBN} Command
25158 The corresponding commands are @samp{info types} in @value{GDBN},
25159 @samp{gdb_search} in @code{gdbtk}.
25161 @subsubheading Example
25165 @subheading The @code{-symbol-list-variables} Command
25166 @findex -symbol-list-variables
25168 @subsubheading Synopsis
25171 -symbol-list-variables
25174 List all the global and static variable names.
25176 @subsubheading @value{GDBN} Command
25178 @samp{info variables} in @value{GDBN}, @samp{gdb_search} in @code{gdbtk}.
25180 @subsubheading Example
25184 @subheading The @code{-symbol-locate} Command
25185 @findex -symbol-locate
25187 @subsubheading Synopsis
25193 @subsubheading @value{GDBN} Command
25195 @samp{gdb_loc} in @code{gdbtk}.
25197 @subsubheading Example
25201 @subheading The @code{-symbol-type} Command
25202 @findex -symbol-type
25204 @subsubheading Synopsis
25207 -symbol-type @var{variable}
25210 Show type of @var{variable}.
25212 @subsubheading @value{GDBN} Command
25214 The corresponding @value{GDBN} command is @samp{ptype}, @code{gdbtk} has
25215 @samp{gdb_obj_variable}.
25217 @subsubheading Example
25222 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
25223 @node GDB/MI File Commands
25224 @section @sc{gdb/mi} File Commands
25226 This section describes the GDB/MI commands to specify executable file names
25227 and to read in and obtain symbol table information.
25229 @subheading The @code{-file-exec-and-symbols} Command
25230 @findex -file-exec-and-symbols
25232 @subsubheading Synopsis
25235 -file-exec-and-symbols @var{file}
25238 Specify the executable file to be debugged. This file is the one from
25239 which the symbol table is also read. If no file is specified, the
25240 command clears the executable and symbol information. If breakpoints
25241 are set when using this command with no arguments, @value{GDBN} will produce
25242 error messages. Otherwise, no output is produced, except a completion
25245 @subsubheading @value{GDBN} Command
25247 The corresponding @value{GDBN} command is @samp{file}.
25249 @subsubheading Example
25253 -file-exec-and-symbols /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
25259 @subheading The @code{-file-exec-file} Command
25260 @findex -file-exec-file
25262 @subsubheading Synopsis
25265 -file-exec-file @var{file}
25268 Specify the executable file to be debugged. Unlike
25269 @samp{-file-exec-and-symbols}, the symbol table is @emph{not} read
25270 from this file. If used without argument, @value{GDBN} clears the information
25271 about the executable file. No output is produced, except a completion
25274 @subsubheading @value{GDBN} Command
25276 The corresponding @value{GDBN} command is @samp{exec-file}.
25278 @subsubheading Example
25282 -file-exec-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
25289 @subheading The @code{-file-list-exec-sections} Command
25290 @findex -file-list-exec-sections
25292 @subsubheading Synopsis
25295 -file-list-exec-sections
25298 List the sections of the current executable file.
25300 @subsubheading @value{GDBN} Command
25302 The @value{GDBN} command @samp{info file} shows, among the rest, the same
25303 information as this command. @code{gdbtk} has a corresponding command
25304 @samp{gdb_load_info}.
25306 @subsubheading Example
25311 @subheading The @code{-file-list-exec-source-file} Command
25312 @findex -file-list-exec-source-file
25314 @subsubheading Synopsis
25317 -file-list-exec-source-file
25320 List the line number, the current source file, and the absolute path
25321 to the current source file for the current executable. The macro
25322 information field has a value of @samp{1} or @samp{0} depending on
25323 whether or not the file includes preprocessor macro information.
25325 @subsubheading @value{GDBN} Command
25327 The @value{GDBN} equivalent is @samp{info source}
25329 @subsubheading Example
25333 123-file-list-exec-source-file
25334 123^done,line="1",file="foo.c",fullname="/home/bar/foo.c,macro-info="1"
25339 @subheading The @code{-file-list-exec-source-files} Command
25340 @findex -file-list-exec-source-files
25342 @subsubheading Synopsis
25345 -file-list-exec-source-files
25348 List the source files for the current executable.
25350 It will always output the filename, but only when @value{GDBN} can find
25351 the absolute file name of a source file, will it output the fullname.
25353 @subsubheading @value{GDBN} Command
25355 The @value{GDBN} equivalent is @samp{info sources}.
25356 @code{gdbtk} has an analogous command @samp{gdb_listfiles}.
25358 @subsubheading Example
25361 -file-list-exec-source-files
25363 @{file=foo.c,fullname=/home/foo.c@},
25364 @{file=/home/bar.c,fullname=/home/bar.c@},
25365 @{file=gdb_could_not_find_fullpath.c@}]
25370 @subheading The @code{-file-list-shared-libraries} Command
25371 @findex -file-list-shared-libraries
25373 @subsubheading Synopsis
25376 -file-list-shared-libraries
25379 List the shared libraries in the program.
25381 @subsubheading @value{GDBN} Command
25383 The corresponding @value{GDBN} command is @samp{info shared}.
25385 @subsubheading Example
25389 @subheading The @code{-file-list-symbol-files} Command
25390 @findex -file-list-symbol-files
25392 @subsubheading Synopsis
25395 -file-list-symbol-files
25400 @subsubheading @value{GDBN} Command
25402 The corresponding @value{GDBN} command is @samp{info file} (part of it).
25404 @subsubheading Example
25409 @subheading The @code{-file-symbol-file} Command
25410 @findex -file-symbol-file
25412 @subsubheading Synopsis
25415 -file-symbol-file @var{file}
25418 Read symbol table info from the specified @var{file} argument. When
25419 used without arguments, clears @value{GDBN}'s symbol table info. No output is
25420 produced, except for a completion notification.
25422 @subsubheading @value{GDBN} Command
25424 The corresponding @value{GDBN} command is @samp{symbol-file}.
25426 @subsubheading Example
25430 -file-symbol-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
25436 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
25437 @node GDB/MI Memory Overlay Commands
25438 @section @sc{gdb/mi} Memory Overlay Commands
25440 The memory overlay commands are not implemented.
25442 @c @subheading -overlay-auto
25444 @c @subheading -overlay-list-mapping-state
25446 @c @subheading -overlay-list-overlays
25448 @c @subheading -overlay-map
25450 @c @subheading -overlay-off
25452 @c @subheading -overlay-on
25454 @c @subheading -overlay-unmap
25456 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
25457 @node GDB/MI Signal Handling Commands
25458 @section @sc{gdb/mi} Signal Handling Commands
25460 Signal handling commands are not implemented.
25462 @c @subheading -signal-handle
25464 @c @subheading -signal-list-handle-actions
25466 @c @subheading -signal-list-signal-types
25470 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
25471 @node GDB/MI Target Manipulation
25472 @section @sc{gdb/mi} Target Manipulation Commands
25475 @subheading The @code{-target-attach} Command
25476 @findex -target-attach
25478 @subsubheading Synopsis
25481 -target-attach @var{pid} | @var{gid} | @var{file}
25484 Attach to a process @var{pid} or a file @var{file} outside of
25485 @value{GDBN}, or a thread group @var{gid}. If attaching to a thread
25486 group, the id previously returned by
25487 @samp{-list-thread-groups --available} must be used.
25489 @subsubheading @value{GDBN} Command
25491 The corresponding @value{GDBN} command is @samp{attach}.
25493 @subsubheading Example
25497 =thread-created,id="1"
25498 *stopped,thread-id="1",frame=@{addr="0xb7f7e410",func="bar",args=[]@}
25504 @subheading The @code{-target-compare-sections} Command
25505 @findex -target-compare-sections
25507 @subsubheading Synopsis
25510 -target-compare-sections [ @var{section} ]
25513 Compare data of section @var{section} on target to the exec file.
25514 Without the argument, all sections are compared.
25516 @subsubheading @value{GDBN} Command
25518 The @value{GDBN} equivalent is @samp{compare-sections}.
25520 @subsubheading Example
25525 @subheading The @code{-target-detach} Command
25526 @findex -target-detach
25528 @subsubheading Synopsis
25531 -target-detach [ @var{pid} | @var{gid} ]
25534 Detach from the remote target which normally resumes its execution.
25535 If either @var{pid} or @var{gid} is specified, detaches from either
25536 the specified process, or specified thread group. There's no output.
25538 @subsubheading @value{GDBN} Command
25540 The corresponding @value{GDBN} command is @samp{detach}.
25542 @subsubheading Example
25552 @subheading The @code{-target-disconnect} Command
25553 @findex -target-disconnect
25555 @subsubheading Synopsis
25561 Disconnect from the remote target. There's no output and the target is
25562 generally not resumed.
25564 @subsubheading @value{GDBN} Command
25566 The corresponding @value{GDBN} command is @samp{disconnect}.
25568 @subsubheading Example
25578 @subheading The @code{-target-download} Command
25579 @findex -target-download
25581 @subsubheading Synopsis
25587 Loads the executable onto the remote target.
25588 It prints out an update message every half second, which includes the fields:
25592 The name of the section.
25594 The size of what has been sent so far for that section.
25596 The size of the section.
25598 The total size of what was sent so far (the current and the previous sections).
25600 The size of the overall executable to download.
25604 Each message is sent as status record (@pxref{GDB/MI Output Syntax, ,
25605 @sc{gdb/mi} Output Syntax}).
25607 In addition, it prints the name and size of the sections, as they are
25608 downloaded. These messages include the following fields:
25612 The name of the section.
25614 The size of the section.
25616 The size of the overall executable to download.
25620 At the end, a summary is printed.
25622 @subsubheading @value{GDBN} Command
25624 The corresponding @value{GDBN} command is @samp{load}.
25626 @subsubheading Example
25628 Note: each status message appears on a single line. Here the messages
25629 have been broken down so that they can fit onto a page.
25634 +download,@{section=".text",section-size="6668",total-size="9880"@}
25635 +download,@{section=".text",section-sent="512",section-size="6668",
25636 total-sent="512",total-size="9880"@}
25637 +download,@{section=".text",section-sent="1024",section-size="6668",
25638 total-sent="1024",total-size="9880"@}
25639 +download,@{section=".text",section-sent="1536",section-size="6668",
25640 total-sent="1536",total-size="9880"@}
25641 +download,@{section=".text",section-sent="2048",section-size="6668",
25642 total-sent="2048",total-size="9880"@}
25643 +download,@{section=".text",section-sent="2560",section-size="6668",
25644 total-sent="2560",total-size="9880"@}
25645 +download,@{section=".text",section-sent="3072",section-size="6668",
25646 total-sent="3072",total-size="9880"@}
25647 +download,@{section=".text",section-sent="3584",section-size="6668",
25648 total-sent="3584",total-size="9880"@}
25649 +download,@{section=".text",section-sent="4096",section-size="6668",
25650 total-sent="4096",total-size="9880"@}
25651 +download,@{section=".text",section-sent="4608",section-size="6668",
25652 total-sent="4608",total-size="9880"@}
25653 +download,@{section=".text",section-sent="5120",section-size="6668",
25654 total-sent="5120",total-size="9880"@}
25655 +download,@{section=".text",section-sent="5632",section-size="6668",
25656 total-sent="5632",total-size="9880"@}
25657 +download,@{section=".text",section-sent="6144",section-size="6668",
25658 total-sent="6144",total-size="9880"@}
25659 +download,@{section=".text",section-sent="6656",section-size="6668",
25660 total-sent="6656",total-size="9880"@}
25661 +download,@{section=".init",section-size="28",total-size="9880"@}
25662 +download,@{section=".fini",section-size="28",total-size="9880"@}
25663 +download,@{section=".data",section-size="3156",total-size="9880"@}
25664 +download,@{section=".data",section-sent="512",section-size="3156",
25665 total-sent="7236",total-size="9880"@}
25666 +download,@{section=".data",section-sent="1024",section-size="3156",
25667 total-sent="7748",total-size="9880"@}
25668 +download,@{section=".data",section-sent="1536",section-size="3156",
25669 total-sent="8260",total-size="9880"@}
25670 +download,@{section=".data",section-sent="2048",section-size="3156",
25671 total-sent="8772",total-size="9880"@}
25672 +download,@{section=".data",section-sent="2560",section-size="3156",
25673 total-sent="9284",total-size="9880"@}
25674 +download,@{section=".data",section-sent="3072",section-size="3156",
25675 total-sent="9796",total-size="9880"@}
25676 ^done,address="0x10004",load-size="9880",transfer-rate="6586",
25683 @subheading The @code{-target-exec-status} Command
25684 @findex -target-exec-status
25686 @subsubheading Synopsis
25689 -target-exec-status
25692 Provide information on the state of the target (whether it is running or
25693 not, for instance).
25695 @subsubheading @value{GDBN} Command
25697 There's no equivalent @value{GDBN} command.
25699 @subsubheading Example
25703 @subheading The @code{-target-list-available-targets} Command
25704 @findex -target-list-available-targets
25706 @subsubheading Synopsis
25709 -target-list-available-targets
25712 List the possible targets to connect to.
25714 @subsubheading @value{GDBN} Command
25716 The corresponding @value{GDBN} command is @samp{help target}.
25718 @subsubheading Example
25722 @subheading The @code{-target-list-current-targets} Command
25723 @findex -target-list-current-targets
25725 @subsubheading Synopsis
25728 -target-list-current-targets
25731 Describe the current target.
25733 @subsubheading @value{GDBN} Command
25735 The corresponding information is printed by @samp{info file} (among
25738 @subsubheading Example
25742 @subheading The @code{-target-list-parameters} Command
25743 @findex -target-list-parameters
25745 @subsubheading Synopsis
25748 -target-list-parameters
25754 @subsubheading @value{GDBN} Command
25758 @subsubheading Example
25762 @subheading The @code{-target-select} Command
25763 @findex -target-select
25765 @subsubheading Synopsis
25768 -target-select @var{type} @var{parameters @dots{}}
25771 Connect @value{GDBN} to the remote target. This command takes two args:
25775 The type of target, for instance @samp{remote}, etc.
25776 @item @var{parameters}
25777 Device names, host names and the like. @xref{Target Commands, ,
25778 Commands for Managing Targets}, for more details.
25781 The output is a connection notification, followed by the address at
25782 which the target program is, in the following form:
25785 ^connected,addr="@var{address}",func="@var{function name}",
25786 args=[@var{arg list}]
25789 @subsubheading @value{GDBN} Command
25791 The corresponding @value{GDBN} command is @samp{target}.
25793 @subsubheading Example
25797 -target-select remote /dev/ttya
25798 ^connected,addr="0xfe00a300",func="??",args=[]
25802 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
25803 @node GDB/MI File Transfer Commands
25804 @section @sc{gdb/mi} File Transfer Commands
25807 @subheading The @code{-target-file-put} Command
25808 @findex -target-file-put
25810 @subsubheading Synopsis
25813 -target-file-put @var{hostfile} @var{targetfile}
25816 Copy file @var{hostfile} from the host system (the machine running
25817 @value{GDBN}) to @var{targetfile} on the target system.
25819 @subsubheading @value{GDBN} Command
25821 The corresponding @value{GDBN} command is @samp{remote put}.
25823 @subsubheading Example
25827 -target-file-put localfile remotefile
25833 @subheading The @code{-target-file-get} Command
25834 @findex -target-file-get
25836 @subsubheading Synopsis
25839 -target-file-get @var{targetfile} @var{hostfile}
25842 Copy file @var{targetfile} from the target system to @var{hostfile}
25843 on the host system.
25845 @subsubheading @value{GDBN} Command
25847 The corresponding @value{GDBN} command is @samp{remote get}.
25849 @subsubheading Example
25853 -target-file-get remotefile localfile
25859 @subheading The @code{-target-file-delete} Command
25860 @findex -target-file-delete
25862 @subsubheading Synopsis
25865 -target-file-delete @var{targetfile}
25868 Delete @var{targetfile} from the target system.
25870 @subsubheading @value{GDBN} Command
25872 The corresponding @value{GDBN} command is @samp{remote delete}.
25874 @subsubheading Example
25878 -target-file-delete remotefile
25884 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
25885 @node GDB/MI Miscellaneous Commands
25886 @section Miscellaneous @sc{gdb/mi} Commands
25888 @c @subheading -gdb-complete
25890 @subheading The @code{-gdb-exit} Command
25893 @subsubheading Synopsis
25899 Exit @value{GDBN} immediately.
25901 @subsubheading @value{GDBN} Command
25903 Approximately corresponds to @samp{quit}.
25905 @subsubheading Example
25915 @subheading The @code{-exec-abort} Command
25916 @findex -exec-abort
25918 @subsubheading Synopsis
25924 Kill the inferior running program.
25926 @subsubheading @value{GDBN} Command
25928 The corresponding @value{GDBN} command is @samp{kill}.
25930 @subsubheading Example
25935 @subheading The @code{-gdb-set} Command
25938 @subsubheading Synopsis
25944 Set an internal @value{GDBN} variable.
25945 @c IS THIS A DOLLAR VARIABLE? OR SOMETHING LIKE ANNOTATE ?????
25947 @subsubheading @value{GDBN} Command
25949 The corresponding @value{GDBN} command is @samp{set}.
25951 @subsubheading Example
25961 @subheading The @code{-gdb-show} Command
25964 @subsubheading Synopsis
25970 Show the current value of a @value{GDBN} variable.
25972 @subsubheading @value{GDBN} Command
25974 The corresponding @value{GDBN} command is @samp{show}.
25976 @subsubheading Example
25985 @c @subheading -gdb-source
25988 @subheading The @code{-gdb-version} Command
25989 @findex -gdb-version
25991 @subsubheading Synopsis
25997 Show version information for @value{GDBN}. Used mostly in testing.
25999 @subsubheading @value{GDBN} Command
26001 The @value{GDBN} equivalent is @samp{show version}. @value{GDBN} by
26002 default shows this information when you start an interactive session.
26004 @subsubheading Example
26006 @c This example modifies the actual output from GDB to avoid overfull
26012 ~Copyright 2000 Free Software Foundation, Inc.
26013 ~GDB is free software, covered by the GNU General Public License, and
26014 ~you are welcome to change it and/or distribute copies of it under
26015 ~ certain conditions.
26016 ~Type "show copying" to see the conditions.
26017 ~There is absolutely no warranty for GDB. Type "show warranty" for
26019 ~This GDB was configured as
26020 "--host=sparc-sun-solaris2.5.1 --target=ppc-eabi".
26025 @subheading The @code{-list-features} Command
26026 @findex -list-features
26028 Returns a list of particular features of the MI protocol that
26029 this version of gdb implements. A feature can be a command,
26030 or a new field in an output of some command, or even an
26031 important bugfix. While a frontend can sometimes detect presence
26032 of a feature at runtime, it is easier to perform detection at debugger
26035 The command returns a list of strings, with each string naming an
26036 available feature. Each returned string is just a name, it does not
26037 have any internal structure. The list of possible feature names
26043 (gdb) -list-features
26044 ^done,result=["feature1","feature2"]
26047 The current list of features is:
26050 @item frozen-varobjs
26051 Indicates presence of the @code{-var-set-frozen} command, as well
26052 as possible presense of the @code{frozen} field in the output
26053 of @code{-varobj-create}.
26054 @item pending-breakpoints
26055 Indicates presence of the @option{-f} option to the @code{-break-insert} command.
26057 Indicates presence of Python scripting support, Python-based
26058 pretty-printing commands, and possible presence of the
26059 @samp{display_hint} field in the output of @code{-var-list-children}
26061 Indicates presence of the @code{-thread-info} command.
26065 @subheading The @code{-list-target-features} Command
26066 @findex -list-target-features
26068 Returns a list of particular features that are supported by the
26069 target. Those features affect the permitted MI commands, but
26070 unlike the features reported by the @code{-list-features} command, the
26071 features depend on which target GDB is using at the moment. Whenever
26072 a target can change, due to commands such as @code{-target-select},
26073 @code{-target-attach} or @code{-exec-run}, the list of target features
26074 may change, and the frontend should obtain it again.
26078 (gdb) -list-features
26079 ^done,result=["async"]
26082 The current list of features is:
26086 Indicates that the target is capable of asynchronous command
26087 execution, which means that @value{GDBN} will accept further commands
26088 while the target is running.
26092 @subheading The @code{-list-thread-groups} Command
26093 @findex -list-thread-groups
26095 @subheading Synopsis
26098 -list-thread-groups [ --available ] [ @var{group} ]
26101 When used without the @var{group} parameter, lists top-level thread
26102 groups that are being debugged. When used with the @var{group}
26103 parameter, the children of the specified group are listed. The
26104 children can be either threads, or other groups. At present,
26105 @value{GDBN} will not report both threads and groups as children at
26106 the same time, but it may change in future.
26108 With the @samp{--available} option, instead of reporting groups that
26109 are been debugged, GDB will report all thread groups available on the
26110 target. Using the @samp{--available} option together with @var{group}
26113 @subheading Example
26117 -list-thread-groups
26118 ^done,groups=[@{id="17",type="process",pid="yyy",num_children="2"@}]
26119 -list-thread-groups 17
26120 ^done,threads=[@{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
26121 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",args=[]@},state="running"@},
26122 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
26123 frame=@{level="0",addr="0x0804891f",func="foo",args=[@{name="i",value="10"@}],
26124 file="/tmp/a.c",fullname="/tmp/a.c",line="158"@},state="running"@}]]
26127 @subheading The @code{-interpreter-exec} Command
26128 @findex -interpreter-exec
26130 @subheading Synopsis
26133 -interpreter-exec @var{interpreter} @var{command}
26135 @anchor{-interpreter-exec}
26137 Execute the specified @var{command} in the given @var{interpreter}.
26139 @subheading @value{GDBN} Command
26141 The corresponding @value{GDBN} command is @samp{interpreter-exec}.
26143 @subheading Example
26147 -interpreter-exec console "break main"
26148 &"During symbol reading, couldn't parse type; debugger out of date?.\n"
26149 &"During symbol reading, bad structure-type format.\n"
26150 ~"Breakpoint 1 at 0x8074fc6: file ../../src/gdb/main.c, line 743.\n"
26155 @subheading The @code{-inferior-tty-set} Command
26156 @findex -inferior-tty-set
26158 @subheading Synopsis
26161 -inferior-tty-set /dev/pts/1
26164 Set terminal for future runs of the program being debugged.
26166 @subheading @value{GDBN} Command
26168 The corresponding @value{GDBN} command is @samp{set inferior-tty} /dev/pts/1.
26170 @subheading Example
26174 -inferior-tty-set /dev/pts/1
26179 @subheading The @code{-inferior-tty-show} Command
26180 @findex -inferior-tty-show
26182 @subheading Synopsis
26188 Show terminal for future runs of program being debugged.
26190 @subheading @value{GDBN} Command
26192 The corresponding @value{GDBN} command is @samp{show inferior-tty}.
26194 @subheading Example
26198 -inferior-tty-set /dev/pts/1
26202 ^done,inferior_tty_terminal="/dev/pts/1"
26206 @subheading The @code{-enable-timings} Command
26207 @findex -enable-timings
26209 @subheading Synopsis
26212 -enable-timings [yes | no]
26215 Toggle the printing of the wallclock, user and system times for an MI
26216 command as a field in its output. This command is to help frontend
26217 developers optimize the performance of their code. No argument is
26218 equivalent to @samp{yes}.
26220 @subheading @value{GDBN} Command
26224 @subheading Example
26232 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
26233 addr="0x080484ed",func="main",file="myprog.c",
26234 fullname="/home/nickrob/myprog.c",line="73",times="0"@},
26235 time=@{wallclock="0.05185",user="0.00800",system="0.00000"@}
26243 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
26244 frame=@{addr="0x080484ed",func="main",args=[@{name="argc",value="1"@},
26245 @{name="argv",value="0xbfb60364"@}],file="myprog.c",
26246 fullname="/home/nickrob/myprog.c",line="73"@}
26251 @chapter @value{GDBN} Annotations
26253 This chapter describes annotations in @value{GDBN}. Annotations were
26254 designed to interface @value{GDBN} to graphical user interfaces or other
26255 similar programs which want to interact with @value{GDBN} at a
26256 relatively high level.
26258 The annotation mechanism has largely been superseded by @sc{gdb/mi}
26262 This is Edition @value{EDITION}, @value{DATE}.
26266 * Annotations Overview:: What annotations are; the general syntax.
26267 * Server Prefix:: Issuing a command without affecting user state.
26268 * Prompting:: Annotations marking @value{GDBN}'s need for input.
26269 * Errors:: Annotations for error messages.
26270 * Invalidation:: Some annotations describe things now invalid.
26271 * Annotations for Running::
26272 Whether the program is running, how it stopped, etc.
26273 * Source Annotations:: Annotations describing source code.
26276 @node Annotations Overview
26277 @section What is an Annotation?
26278 @cindex annotations
26280 Annotations start with a newline character, two @samp{control-z}
26281 characters, and the name of the annotation. If there is no additional
26282 information associated with this annotation, the name of the annotation
26283 is followed immediately by a newline. If there is additional
26284 information, the name of the annotation is followed by a space, the
26285 additional information, and a newline. The additional information
26286 cannot contain newline characters.
26288 Any output not beginning with a newline and two @samp{control-z}
26289 characters denotes literal output from @value{GDBN}. Currently there is
26290 no need for @value{GDBN} to output a newline followed by two
26291 @samp{control-z} characters, but if there was such a need, the
26292 annotations could be extended with an @samp{escape} annotation which
26293 means those three characters as output.
26295 The annotation @var{level}, which is specified using the
26296 @option{--annotate} command line option (@pxref{Mode Options}), controls
26297 how much information @value{GDBN} prints together with its prompt,
26298 values of expressions, source lines, and other types of output. Level 0
26299 is for no annotations, level 1 is for use when @value{GDBN} is run as a
26300 subprocess of @sc{gnu} Emacs, level 3 is the maximum annotation suitable
26301 for programs that control @value{GDBN}, and level 2 annotations have
26302 been made obsolete (@pxref{Limitations, , Limitations of the Annotation
26303 Interface, annotate, GDB's Obsolete Annotations}).
26306 @kindex set annotate
26307 @item set annotate @var{level}
26308 The @value{GDBN} command @code{set annotate} sets the level of
26309 annotations to the specified @var{level}.
26311 @item show annotate
26312 @kindex show annotate
26313 Show the current annotation level.
26316 This chapter describes level 3 annotations.
26318 A simple example of starting up @value{GDBN} with annotations is:
26321 $ @kbd{gdb --annotate=3}
26323 Copyright 2003 Free Software Foundation, Inc.
26324 GDB is free software, covered by the GNU General Public License,
26325 and you are welcome to change it and/or distribute copies of it
26326 under certain conditions.
26327 Type "show copying" to see the conditions.
26328 There is absolutely no warranty for GDB. Type "show warranty"
26330 This GDB was configured as "i386-pc-linux-gnu"
26341 Here @samp{quit} is input to @value{GDBN}; the rest is output from
26342 @value{GDBN}. The three lines beginning @samp{^Z^Z} (where @samp{^Z}
26343 denotes a @samp{control-z} character) are annotations; the rest is
26344 output from @value{GDBN}.
26346 @node Server Prefix
26347 @section The Server Prefix
26348 @cindex server prefix
26350 If you prefix a command with @samp{server } then it will not affect
26351 the command history, nor will it affect @value{GDBN}'s notion of which
26352 command to repeat if @key{RET} is pressed on a line by itself. This
26353 means that commands can be run behind a user's back by a front-end in
26354 a transparent manner.
26356 The @code{server } prefix does not affect the recording of values into
26357 the value history; to print a value without recording it into the
26358 value history, use the @code{output} command instead of the
26359 @code{print} command.
26361 Using this prefix also disables confirmation requests
26362 (@pxref{confirmation requests}).
26365 @section Annotation for @value{GDBN} Input
26367 @cindex annotations for prompts
26368 When @value{GDBN} prompts for input, it annotates this fact so it is possible
26369 to know when to send output, when the output from a given command is
26372 Different kinds of input each have a different @dfn{input type}. Each
26373 input type has three annotations: a @code{pre-} annotation, which
26374 denotes the beginning of any prompt which is being output, a plain
26375 annotation, which denotes the end of the prompt, and then a @code{post-}
26376 annotation which denotes the end of any echo which may (or may not) be
26377 associated with the input. For example, the @code{prompt} input type
26378 features the following annotations:
26386 The input types are
26389 @findex pre-prompt annotation
26390 @findex prompt annotation
26391 @findex post-prompt annotation
26393 When @value{GDBN} is prompting for a command (the main @value{GDBN} prompt).
26395 @findex pre-commands annotation
26396 @findex commands annotation
26397 @findex post-commands annotation
26399 When @value{GDBN} prompts for a set of commands, like in the @code{commands}
26400 command. The annotations are repeated for each command which is input.
26402 @findex pre-overload-choice annotation
26403 @findex overload-choice annotation
26404 @findex post-overload-choice annotation
26405 @item overload-choice
26406 When @value{GDBN} wants the user to select between various overloaded functions.
26408 @findex pre-query annotation
26409 @findex query annotation
26410 @findex post-query annotation
26412 When @value{GDBN} wants the user to confirm a potentially dangerous operation.
26414 @findex pre-prompt-for-continue annotation
26415 @findex prompt-for-continue annotation
26416 @findex post-prompt-for-continue annotation
26417 @item prompt-for-continue
26418 When @value{GDBN} is asking the user to press return to continue. Note: Don't
26419 expect this to work well; instead use @code{set height 0} to disable
26420 prompting. This is because the counting of lines is buggy in the
26421 presence of annotations.
26426 @cindex annotations for errors, warnings and interrupts
26428 @findex quit annotation
26433 This annotation occurs right before @value{GDBN} responds to an interrupt.
26435 @findex error annotation
26440 This annotation occurs right before @value{GDBN} responds to an error.
26442 Quit and error annotations indicate that any annotations which @value{GDBN} was
26443 in the middle of may end abruptly. For example, if a
26444 @code{value-history-begin} annotation is followed by a @code{error}, one
26445 cannot expect to receive the matching @code{value-history-end}. One
26446 cannot expect not to receive it either, however; an error annotation
26447 does not necessarily mean that @value{GDBN} is immediately returning all the way
26450 @findex error-begin annotation
26451 A quit or error annotation may be preceded by
26457 Any output between that and the quit or error annotation is the error
26460 Warning messages are not yet annotated.
26461 @c If we want to change that, need to fix warning(), type_error(),
26462 @c range_error(), and possibly other places.
26465 @section Invalidation Notices
26467 @cindex annotations for invalidation messages
26468 The following annotations say that certain pieces of state may have
26472 @findex frames-invalid annotation
26473 @item ^Z^Zframes-invalid
26475 The frames (for example, output from the @code{backtrace} command) may
26478 @findex breakpoints-invalid annotation
26479 @item ^Z^Zbreakpoints-invalid
26481 The breakpoints may have changed. For example, the user just added or
26482 deleted a breakpoint.
26485 @node Annotations for Running
26486 @section Running the Program
26487 @cindex annotations for running programs
26489 @findex starting annotation
26490 @findex stopping annotation
26491 When the program starts executing due to a @value{GDBN} command such as
26492 @code{step} or @code{continue},
26498 is output. When the program stops,
26504 is output. Before the @code{stopped} annotation, a variety of
26505 annotations describe how the program stopped.
26508 @findex exited annotation
26509 @item ^Z^Zexited @var{exit-status}
26510 The program exited, and @var{exit-status} is the exit status (zero for
26511 successful exit, otherwise nonzero).
26513 @findex signalled annotation
26514 @findex signal-name annotation
26515 @findex signal-name-end annotation
26516 @findex signal-string annotation
26517 @findex signal-string-end annotation
26518 @item ^Z^Zsignalled
26519 The program exited with a signal. After the @code{^Z^Zsignalled}, the
26520 annotation continues:
26526 ^Z^Zsignal-name-end
26530 ^Z^Zsignal-string-end
26535 where @var{name} is the name of the signal, such as @code{SIGILL} or
26536 @code{SIGSEGV}, and @var{string} is the explanation of the signal, such
26537 as @code{Illegal Instruction} or @code{Segmentation fault}.
26538 @var{intro-text}, @var{middle-text}, and @var{end-text} are for the
26539 user's benefit and have no particular format.
26541 @findex signal annotation
26543 The syntax of this annotation is just like @code{signalled}, but @value{GDBN} is
26544 just saying that the program received the signal, not that it was
26545 terminated with it.
26547 @findex breakpoint annotation
26548 @item ^Z^Zbreakpoint @var{number}
26549 The program hit breakpoint number @var{number}.
26551 @findex watchpoint annotation
26552 @item ^Z^Zwatchpoint @var{number}
26553 The program hit watchpoint number @var{number}.
26556 @node Source Annotations
26557 @section Displaying Source
26558 @cindex annotations for source display
26560 @findex source annotation
26561 The following annotation is used instead of displaying source code:
26564 ^Z^Zsource @var{filename}:@var{line}:@var{character}:@var{middle}:@var{addr}
26567 where @var{filename} is an absolute file name indicating which source
26568 file, @var{line} is the line number within that file (where 1 is the
26569 first line in the file), @var{character} is the character position
26570 within the file (where 0 is the first character in the file) (for most
26571 debug formats this will necessarily point to the beginning of a line),
26572 @var{middle} is @samp{middle} if @var{addr} is in the middle of the
26573 line, or @samp{beg} if @var{addr} is at the beginning of the line, and
26574 @var{addr} is the address in the target program associated with the
26575 source which is being displayed. @var{addr} is in the form @samp{0x}
26576 followed by one or more lowercase hex digits (note that this does not
26577 depend on the language).
26579 @node JIT Interface
26580 @chapter JIT Compilation Interface
26581 @cindex just-in-time compilation
26582 @cindex JIT compilation interface
26584 This chapter documents @value{GDBN}'s @dfn{just-in-time} (JIT) compilation
26585 interface. A JIT compiler is a program or library that generates native
26586 executable code at runtime and executes it, usually in order to achieve good
26587 performance while maintaining platform independence.
26589 Programs that use JIT compilation are normally difficult to debug because
26590 portions of their code are generated at runtime, instead of being loaded from
26591 object files, which is where @value{GDBN} normally finds the program's symbols
26592 and debug information. In order to debug programs that use JIT compilation,
26593 @value{GDBN} has an interface that allows the program to register in-memory
26594 symbol files with @value{GDBN} at runtime.
26596 If you are using @value{GDBN} to debug a program that uses this interface, then
26597 it should work transparently so long as you have not stripped the binary. If
26598 you are developing a JIT compiler, then the interface is documented in the rest
26599 of this chapter. At this time, the only known client of this interface is the
26602 Broadly speaking, the JIT interface mirrors the dynamic loader interface. The
26603 JIT compiler communicates with @value{GDBN} by writing data into a global
26604 variable and calling a fuction at a well-known symbol. When @value{GDBN}
26605 attaches, it reads a linked list of symbol files from the global variable to
26606 find existing code, and puts a breakpoint in the function so that it can find
26607 out about additional code.
26610 * Declarations:: Relevant C struct declarations
26611 * Registering Code:: Steps to register code
26612 * Unregistering Code:: Steps to unregister code
26616 @section JIT Declarations
26618 These are the relevant struct declarations that a C program should include to
26619 implement the interface:
26629 struct jit_code_entry
26631 struct jit_code_entry *next_entry;
26632 struct jit_code_entry *prev_entry;
26633 const char *symfile_addr;
26634 uint64_t symfile_size;
26637 struct jit_descriptor
26640 /* This type should be jit_actions_t, but we use uint32_t
26641 to be explicit about the bitwidth. */
26642 uint32_t action_flag;
26643 struct jit_code_entry *relevant_entry;
26644 struct jit_code_entry *first_entry;
26647 /* GDB puts a breakpoint in this function. */
26648 void __attribute__((noinline)) __jit_debug_register_code() @{ @};
26650 /* Make sure to specify the version statically, because the
26651 debugger may check the version before we can set it. */
26652 struct jit_descriptor __jit_debug_descriptor = @{ 1, 0, 0, 0 @};
26655 If the JIT is multi-threaded, then it is important that the JIT synchronize any
26656 modifications to this global data properly, which can easily be done by putting
26657 a global mutex around modifications to these structures.
26659 @node Registering Code
26660 @section Registering Code
26662 To register code with @value{GDBN}, the JIT should follow this protocol:
26666 Generate an object file in memory with symbols and other desired debug
26667 information. The file must include the virtual addresses of the sections.
26670 Create a code entry for the file, which gives the start and size of the symbol
26674 Add it to the linked list in the JIT descriptor.
26677 Point the relevant_entry field of the descriptor at the entry.
26680 Set @code{action_flag} to @code{JIT_REGISTER} and call
26681 @code{__jit_debug_register_code}.
26684 When @value{GDBN} is attached and the breakpoint fires, @value{GDBN} uses the
26685 @code{relevant_entry} pointer so it doesn't have to walk the list looking for
26686 new code. However, the linked list must still be maintained in order to allow
26687 @value{GDBN} to attach to a running process and still find the symbol files.
26689 @node Unregistering Code
26690 @section Unregistering Code
26692 If code is freed, then the JIT should use the following protocol:
26696 Remove the code entry corresponding to the code from the linked list.
26699 Point the @code{relevant_entry} field of the descriptor at the code entry.
26702 Set @code{action_flag} to @code{JIT_UNREGISTER} and call
26703 @code{__jit_debug_register_code}.
26706 If the JIT frees or recompiles code without unregistering it, then @value{GDBN}
26707 and the JIT will leak the memory used for the associated symbol files.
26710 @chapter Reporting Bugs in @value{GDBN}
26711 @cindex bugs in @value{GDBN}
26712 @cindex reporting bugs in @value{GDBN}
26714 Your bug reports play an essential role in making @value{GDBN} reliable.
26716 Reporting a bug may help you by bringing a solution to your problem, or it
26717 may not. But in any case the principal function of a bug report is to help
26718 the entire community by making the next version of @value{GDBN} work better. Bug
26719 reports are your contribution to the maintenance of @value{GDBN}.
26721 In order for a bug report to serve its purpose, you must include the
26722 information that enables us to fix the bug.
26725 * Bug Criteria:: Have you found a bug?
26726 * Bug Reporting:: How to report bugs
26730 @section Have You Found a Bug?
26731 @cindex bug criteria
26733 If you are not sure whether you have found a bug, here are some guidelines:
26736 @cindex fatal signal
26737 @cindex debugger crash
26738 @cindex crash of debugger
26740 If the debugger gets a fatal signal, for any input whatever, that is a
26741 @value{GDBN} bug. Reliable debuggers never crash.
26743 @cindex error on valid input
26745 If @value{GDBN} produces an error message for valid input, that is a
26746 bug. (Note that if you're cross debugging, the problem may also be
26747 somewhere in the connection to the target.)
26749 @cindex invalid input
26751 If @value{GDBN} does not produce an error message for invalid input,
26752 that is a bug. However, you should note that your idea of
26753 ``invalid input'' might be our idea of ``an extension'' or ``support
26754 for traditional practice''.
26757 If you are an experienced user of debugging tools, your suggestions
26758 for improvement of @value{GDBN} are welcome in any case.
26761 @node Bug Reporting
26762 @section How to Report Bugs
26763 @cindex bug reports
26764 @cindex @value{GDBN} bugs, reporting
26766 A number of companies and individuals offer support for @sc{gnu} products.
26767 If you obtained @value{GDBN} from a support organization, we recommend you
26768 contact that organization first.
26770 You can find contact information for many support companies and
26771 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
26773 @c should add a web page ref...
26776 @ifset BUGURL_DEFAULT
26777 In any event, we also recommend that you submit bug reports for
26778 @value{GDBN}. The preferred method is to submit them directly using
26779 @uref{http://www.gnu.org/software/gdb/bugs/, @value{GDBN}'s Bugs web
26780 page}. Alternatively, the @email{bug-gdb@@gnu.org, e-mail gateway} can
26783 @strong{Do not send bug reports to @samp{info-gdb}, or to
26784 @samp{help-gdb}, or to any newsgroups.} Most users of @value{GDBN} do
26785 not want to receive bug reports. Those that do have arranged to receive
26788 The mailing list @samp{bug-gdb} has a newsgroup @samp{gnu.gdb.bug} which
26789 serves as a repeater. The mailing list and the newsgroup carry exactly
26790 the same messages. Often people think of posting bug reports to the
26791 newsgroup instead of mailing them. This appears to work, but it has one
26792 problem which can be crucial: a newsgroup posting often lacks a mail
26793 path back to the sender. Thus, if we need to ask for more information,
26794 we may be unable to reach you. For this reason, it is better to send
26795 bug reports to the mailing list.
26797 @ifclear BUGURL_DEFAULT
26798 In any event, we also recommend that you submit bug reports for
26799 @value{GDBN} to @value{BUGURL}.
26803 The fundamental principle of reporting bugs usefully is this:
26804 @strong{report all the facts}. If you are not sure whether to state a
26805 fact or leave it out, state it!
26807 Often people omit facts because they think they know what causes the
26808 problem and assume that some details do not matter. Thus, you might
26809 assume that the name of the variable you use in an example does not matter.
26810 Well, probably it does not, but one cannot be sure. Perhaps the bug is a
26811 stray memory reference which happens to fetch from the location where that
26812 name is stored in memory; perhaps, if the name were different, the contents
26813 of that location would fool the debugger into doing the right thing despite
26814 the bug. Play it safe and give a specific, complete example. That is the
26815 easiest thing for you to do, and the most helpful.
26817 Keep in mind that the purpose of a bug report is to enable us to fix the
26818 bug. It may be that the bug has been reported previously, but neither
26819 you nor we can know that unless your bug report is complete and
26822 Sometimes people give a few sketchy facts and ask, ``Does this ring a
26823 bell?'' Those bug reports are useless, and we urge everyone to
26824 @emph{refuse to respond to them} except to chide the sender to report
26827 To enable us to fix the bug, you should include all these things:
26831 The version of @value{GDBN}. @value{GDBN} announces it if you start
26832 with no arguments; you can also print it at any time using @code{show
26835 Without this, we will not know whether there is any point in looking for
26836 the bug in the current version of @value{GDBN}.
26839 The type of machine you are using, and the operating system name and
26843 What compiler (and its version) was used to compile @value{GDBN}---e.g.@:
26844 ``@value{GCC}--2.8.1''.
26847 What compiler (and its version) was used to compile the program you are
26848 debugging---e.g.@: ``@value{GCC}--2.8.1'', or ``HP92453-01 A.10.32.03 HP
26849 C Compiler''. For @value{NGCC}, you can say @kbd{@value{GCC} --version}
26850 to get this information; for other compilers, see the documentation for
26854 The command arguments you gave the compiler to compile your example and
26855 observe the bug. For example, did you use @samp{-O}? To guarantee
26856 you will not omit something important, list them all. A copy of the
26857 Makefile (or the output from make) is sufficient.
26859 If we were to try to guess the arguments, we would probably guess wrong
26860 and then we might not encounter the bug.
26863 A complete input script, and all necessary source files, that will
26867 A description of what behavior you observe that you believe is
26868 incorrect. For example, ``It gets a fatal signal.''
26870 Of course, if the bug is that @value{GDBN} gets a fatal signal, then we
26871 will certainly notice it. But if the bug is incorrect output, we might
26872 not notice unless it is glaringly wrong. You might as well not give us
26873 a chance to make a mistake.
26875 Even if the problem you experience is a fatal signal, you should still
26876 say so explicitly. Suppose something strange is going on, such as, your
26877 copy of @value{GDBN} is out of synch, or you have encountered a bug in
26878 the C library on your system. (This has happened!) Your copy might
26879 crash and ours would not. If you told us to expect a crash, then when
26880 ours fails to crash, we would know that the bug was not happening for
26881 us. If you had not told us to expect a crash, then we would not be able
26882 to draw any conclusion from our observations.
26885 @cindex recording a session script
26886 To collect all this information, you can use a session recording program
26887 such as @command{script}, which is available on many Unix systems.
26888 Just run your @value{GDBN} session inside @command{script} and then
26889 include the @file{typescript} file with your bug report.
26891 Another way to record a @value{GDBN} session is to run @value{GDBN}
26892 inside Emacs and then save the entire buffer to a file.
26895 If you wish to suggest changes to the @value{GDBN} source, send us context
26896 diffs. If you even discuss something in the @value{GDBN} source, refer to
26897 it by context, not by line number.
26899 The line numbers in our development sources will not match those in your
26900 sources. Your line numbers would convey no useful information to us.
26904 Here are some things that are not necessary:
26908 A description of the envelope of the bug.
26910 Often people who encounter a bug spend a lot of time investigating
26911 which changes to the input file will make the bug go away and which
26912 changes will not affect it.
26914 This is often time consuming and not very useful, because the way we
26915 will find the bug is by running a single example under the debugger
26916 with breakpoints, not by pure deduction from a series of examples.
26917 We recommend that you save your time for something else.
26919 Of course, if you can find a simpler example to report @emph{instead}
26920 of the original one, that is a convenience for us. Errors in the
26921 output will be easier to spot, running under the debugger will take
26922 less time, and so on.
26924 However, simplification is not vital; if you do not want to do this,
26925 report the bug anyway and send us the entire test case you used.
26928 A patch for the bug.
26930 A patch for the bug does help us if it is a good one. But do not omit
26931 the necessary information, such as the test case, on the assumption that
26932 a patch is all we need. We might see problems with your patch and decide
26933 to fix the problem another way, or we might not understand it at all.
26935 Sometimes with a program as complicated as @value{GDBN} it is very hard to
26936 construct an example that will make the program follow a certain path
26937 through the code. If you do not send us the example, we will not be able
26938 to construct one, so we will not be able to verify that the bug is fixed.
26940 And if we cannot understand what bug you are trying to fix, or why your
26941 patch should be an improvement, we will not install it. A test case will
26942 help us to understand.
26945 A guess about what the bug is or what it depends on.
26947 Such guesses are usually wrong. Even we cannot guess right about such
26948 things without first using the debugger to find the facts.
26951 @c The readline documentation is distributed with the readline code
26952 @c and consists of the two following files:
26954 @c inc-hist.texinfo
26955 @c Use -I with makeinfo to point to the appropriate directory,
26956 @c environment var TEXINPUTS with TeX.
26957 @include rluser.texi
26958 @include inc-hist.texinfo
26961 @node Formatting Documentation
26962 @appendix Formatting Documentation
26964 @cindex @value{GDBN} reference card
26965 @cindex reference card
26966 The @value{GDBN} 4 release includes an already-formatted reference card, ready
26967 for printing with PostScript or Ghostscript, in the @file{gdb}
26968 subdirectory of the main source directory@footnote{In
26969 @file{gdb-@value{GDBVN}/gdb/refcard.ps} of the version @value{GDBVN}
26970 release.}. If you can use PostScript or Ghostscript with your printer,
26971 you can print the reference card immediately with @file{refcard.ps}.
26973 The release also includes the source for the reference card. You
26974 can format it, using @TeX{}, by typing:
26980 The @value{GDBN} reference card is designed to print in @dfn{landscape}
26981 mode on US ``letter'' size paper;
26982 that is, on a sheet 11 inches wide by 8.5 inches
26983 high. You will need to specify this form of printing as an option to
26984 your @sc{dvi} output program.
26986 @cindex documentation
26988 All the documentation for @value{GDBN} comes as part of the machine-readable
26989 distribution. The documentation is written in Texinfo format, which is
26990 a documentation system that uses a single source file to produce both
26991 on-line information and a printed manual. You can use one of the Info
26992 formatting commands to create the on-line version of the documentation
26993 and @TeX{} (or @code{texi2roff}) to typeset the printed version.
26995 @value{GDBN} includes an already formatted copy of the on-line Info
26996 version of this manual in the @file{gdb} subdirectory. The main Info
26997 file is @file{gdb-@value{GDBVN}/gdb/gdb.info}, and it refers to
26998 subordinate files matching @samp{gdb.info*} in the same directory. If
26999 necessary, you can print out these files, or read them with any editor;
27000 but they are easier to read using the @code{info} subsystem in @sc{gnu}
27001 Emacs or the standalone @code{info} program, available as part of the
27002 @sc{gnu} Texinfo distribution.
27004 If you want to format these Info files yourself, you need one of the
27005 Info formatting programs, such as @code{texinfo-format-buffer} or
27008 If you have @code{makeinfo} installed, and are in the top level
27009 @value{GDBN} source directory (@file{gdb-@value{GDBVN}}, in the case of
27010 version @value{GDBVN}), you can make the Info file by typing:
27017 If you want to typeset and print copies of this manual, you need @TeX{},
27018 a program to print its @sc{dvi} output files, and @file{texinfo.tex}, the
27019 Texinfo definitions file.
27021 @TeX{} is a typesetting program; it does not print files directly, but
27022 produces output files called @sc{dvi} files. To print a typeset
27023 document, you need a program to print @sc{dvi} files. If your system
27024 has @TeX{} installed, chances are it has such a program. The precise
27025 command to use depends on your system; @kbd{lpr -d} is common; another
27026 (for PostScript devices) is @kbd{dvips}. The @sc{dvi} print command may
27027 require a file name without any extension or a @samp{.dvi} extension.
27029 @TeX{} also requires a macro definitions file called
27030 @file{texinfo.tex}. This file tells @TeX{} how to typeset a document
27031 written in Texinfo format. On its own, @TeX{} cannot either read or
27032 typeset a Texinfo file. @file{texinfo.tex} is distributed with GDB
27033 and is located in the @file{gdb-@var{version-number}/texinfo}
27036 If you have @TeX{} and a @sc{dvi} printer program installed, you can
27037 typeset and print this manual. First switch to the @file{gdb}
27038 subdirectory of the main source directory (for example, to
27039 @file{gdb-@value{GDBVN}/gdb}) and type:
27045 Then give @file{gdb.dvi} to your @sc{dvi} printing program.
27047 @node Installing GDB
27048 @appendix Installing @value{GDBN}
27049 @cindex installation
27052 * Requirements:: Requirements for building @value{GDBN}
27053 * Running Configure:: Invoking the @value{GDBN} @file{configure} script
27054 * Separate Objdir:: Compiling @value{GDBN} in another directory
27055 * Config Names:: Specifying names for hosts and targets
27056 * Configure Options:: Summary of options for configure
27057 * System-wide configuration:: Having a system-wide init file
27061 @section Requirements for Building @value{GDBN}
27062 @cindex building @value{GDBN}, requirements for
27064 Building @value{GDBN} requires various tools and packages to be available.
27065 Other packages will be used only if they are found.
27067 @heading Tools/Packages Necessary for Building @value{GDBN}
27069 @item ISO C90 compiler
27070 @value{GDBN} is written in ISO C90. It should be buildable with any
27071 working C90 compiler, e.g.@: GCC.
27075 @heading Tools/Packages Optional for Building @value{GDBN}
27079 @value{GDBN} can use the Expat XML parsing library. This library may be
27080 included with your operating system distribution; if it is not, you
27081 can get the latest version from @url{http://expat.sourceforge.net}.
27082 The @file{configure} script will search for this library in several
27083 standard locations; if it is installed in an unusual path, you can
27084 use the @option{--with-libexpat-prefix} option to specify its location.
27090 Remote protocol memory maps (@pxref{Memory Map Format})
27092 Target descriptions (@pxref{Target Descriptions})
27094 Remote shared library lists (@pxref{Library List Format})
27096 MS-Windows shared libraries (@pxref{Shared Libraries})
27100 @cindex compressed debug sections
27101 @value{GDBN} will use the @samp{zlib} library, if available, to read
27102 compressed debug sections. Some linkers, such as GNU gold, are capable
27103 of producing binaries with compressed debug sections. If @value{GDBN}
27104 is compiled with @samp{zlib}, it will be able to read the debug
27105 information in such binaries.
27107 The @samp{zlib} library is likely included with your operating system
27108 distribution; if it is not, you can get the latest version from
27109 @url{http://zlib.net}.
27112 @value{GDBN}'s features related to character sets (@pxref{Character
27113 Sets}) require a functioning @code{iconv} implementation. If you are
27114 on a GNU system, then this is provided by the GNU C Library. Some
27115 other systems also provide a working @code{iconv}.
27117 On systems with @code{iconv}, you can install GNU Libiconv. If you
27118 have previously installed Libiconv, you can use the
27119 @option{--with-libiconv-prefix} option to configure.
27121 @value{GDBN}'s top-level @file{configure} and @file{Makefile} will
27122 arrange to build Libiconv if a directory named @file{libiconv} appears
27123 in the top-most source directory. If Libiconv is built this way, and
27124 if the operating system does not provide a suitable @code{iconv}
27125 implementation, then the just-built library will automatically be used
27126 by @value{GDBN}. One easy way to set this up is to download GNU
27127 Libiconv, unpack it, and then rename the directory holding the
27128 Libiconv source code to @samp{libiconv}.
27131 @node Running Configure
27132 @section Invoking the @value{GDBN} @file{configure} Script
27133 @cindex configuring @value{GDBN}
27134 @value{GDBN} comes with a @file{configure} script that automates the process
27135 of preparing @value{GDBN} for installation; you can then use @code{make} to
27136 build the @code{gdb} program.
27138 @c irrelevant in info file; it's as current as the code it lives with.
27139 @footnote{If you have a more recent version of @value{GDBN} than @value{GDBVN},
27140 look at the @file{README} file in the sources; we may have improved the
27141 installation procedures since publishing this manual.}
27144 The @value{GDBN} distribution includes all the source code you need for
27145 @value{GDBN} in a single directory, whose name is usually composed by
27146 appending the version number to @samp{gdb}.
27148 For example, the @value{GDBN} version @value{GDBVN} distribution is in the
27149 @file{gdb-@value{GDBVN}} directory. That directory contains:
27152 @item gdb-@value{GDBVN}/configure @r{(and supporting files)}
27153 script for configuring @value{GDBN} and all its supporting libraries
27155 @item gdb-@value{GDBVN}/gdb
27156 the source specific to @value{GDBN} itself
27158 @item gdb-@value{GDBVN}/bfd
27159 source for the Binary File Descriptor library
27161 @item gdb-@value{GDBVN}/include
27162 @sc{gnu} include files
27164 @item gdb-@value{GDBVN}/libiberty
27165 source for the @samp{-liberty} free software library
27167 @item gdb-@value{GDBVN}/opcodes
27168 source for the library of opcode tables and disassemblers
27170 @item gdb-@value{GDBVN}/readline
27171 source for the @sc{gnu} command-line interface
27173 @item gdb-@value{GDBVN}/glob
27174 source for the @sc{gnu} filename pattern-matching subroutine
27176 @item gdb-@value{GDBVN}/mmalloc
27177 source for the @sc{gnu} memory-mapped malloc package
27180 The simplest way to configure and build @value{GDBN} is to run @file{configure}
27181 from the @file{gdb-@var{version-number}} source directory, which in
27182 this example is the @file{gdb-@value{GDBVN}} directory.
27184 First switch to the @file{gdb-@var{version-number}} source directory
27185 if you are not already in it; then run @file{configure}. Pass the
27186 identifier for the platform on which @value{GDBN} will run as an
27192 cd gdb-@value{GDBVN}
27193 ./configure @var{host}
27198 where @var{host} is an identifier such as @samp{sun4} or
27199 @samp{decstation}, that identifies the platform where @value{GDBN} will run.
27200 (You can often leave off @var{host}; @file{configure} tries to guess the
27201 correct value by examining your system.)
27203 Running @samp{configure @var{host}} and then running @code{make} builds the
27204 @file{bfd}, @file{readline}, @file{mmalloc}, and @file{libiberty}
27205 libraries, then @code{gdb} itself. The configured source files, and the
27206 binaries, are left in the corresponding source directories.
27209 @file{configure} is a Bourne-shell (@code{/bin/sh}) script; if your
27210 system does not recognize this automatically when you run a different
27211 shell, you may need to run @code{sh} on it explicitly:
27214 sh configure @var{host}
27217 If you run @file{configure} from a directory that contains source
27218 directories for multiple libraries or programs, such as the
27219 @file{gdb-@value{GDBVN}} source directory for version @value{GDBVN},
27221 creates configuration files for every directory level underneath (unless
27222 you tell it not to, with the @samp{--norecursion} option).
27224 You should run the @file{configure} script from the top directory in the
27225 source tree, the @file{gdb-@var{version-number}} directory. If you run
27226 @file{configure} from one of the subdirectories, you will configure only
27227 that subdirectory. That is usually not what you want. In particular,
27228 if you run the first @file{configure} from the @file{gdb} subdirectory
27229 of the @file{gdb-@var{version-number}} directory, you will omit the
27230 configuration of @file{bfd}, @file{readline}, and other sibling
27231 directories of the @file{gdb} subdirectory. This leads to build errors
27232 about missing include files such as @file{bfd/bfd.h}.
27234 You can install @code{@value{GDBP}} anywhere; it has no hardwired paths.
27235 However, you should make sure that the shell on your path (named by
27236 the @samp{SHELL} environment variable) is publicly readable. Remember
27237 that @value{GDBN} uses the shell to start your program---some systems refuse to
27238 let @value{GDBN} debug child processes whose programs are not readable.
27240 @node Separate Objdir
27241 @section Compiling @value{GDBN} in Another Directory
27243 If you want to run @value{GDBN} versions for several host or target machines,
27244 you need a different @code{gdb} compiled for each combination of
27245 host and target. @file{configure} is designed to make this easy by
27246 allowing you to generate each configuration in a separate subdirectory,
27247 rather than in the source directory. If your @code{make} program
27248 handles the @samp{VPATH} feature (@sc{gnu} @code{make} does), running
27249 @code{make} in each of these directories builds the @code{gdb}
27250 program specified there.
27252 To build @code{gdb} in a separate directory, run @file{configure}
27253 with the @samp{--srcdir} option to specify where to find the source.
27254 (You also need to specify a path to find @file{configure}
27255 itself from your working directory. If the path to @file{configure}
27256 would be the same as the argument to @samp{--srcdir}, you can leave out
27257 the @samp{--srcdir} option; it is assumed.)
27259 For example, with version @value{GDBVN}, you can build @value{GDBN} in a
27260 separate directory for a Sun 4 like this:
27264 cd gdb-@value{GDBVN}
27267 ../gdb-@value{GDBVN}/configure sun4
27272 When @file{configure} builds a configuration using a remote source
27273 directory, it creates a tree for the binaries with the same structure
27274 (and using the same names) as the tree under the source directory. In
27275 the example, you'd find the Sun 4 library @file{libiberty.a} in the
27276 directory @file{gdb-sun4/libiberty}, and @value{GDBN} itself in
27277 @file{gdb-sun4/gdb}.
27279 Make sure that your path to the @file{configure} script has just one
27280 instance of @file{gdb} in it. If your path to @file{configure} looks
27281 like @file{../gdb-@value{GDBVN}/gdb/configure}, you are configuring only
27282 one subdirectory of @value{GDBN}, not the whole package. This leads to
27283 build errors about missing include files such as @file{bfd/bfd.h}.
27285 One popular reason to build several @value{GDBN} configurations in separate
27286 directories is to configure @value{GDBN} for cross-compiling (where
27287 @value{GDBN} runs on one machine---the @dfn{host}---while debugging
27288 programs that run on another machine---the @dfn{target}).
27289 You specify a cross-debugging target by
27290 giving the @samp{--target=@var{target}} option to @file{configure}.
27292 When you run @code{make} to build a program or library, you must run
27293 it in a configured directory---whatever directory you were in when you
27294 called @file{configure} (or one of its subdirectories).
27296 The @code{Makefile} that @file{configure} generates in each source
27297 directory also runs recursively. If you type @code{make} in a source
27298 directory such as @file{gdb-@value{GDBVN}} (or in a separate configured
27299 directory configured with @samp{--srcdir=@var{dirname}/gdb-@value{GDBVN}}), you
27300 will build all the required libraries, and then build GDB.
27302 When you have multiple hosts or targets configured in separate
27303 directories, you can run @code{make} on them in parallel (for example,
27304 if they are NFS-mounted on each of the hosts); they will not interfere
27308 @section Specifying Names for Hosts and Targets
27310 The specifications used for hosts and targets in the @file{configure}
27311 script are based on a three-part naming scheme, but some short predefined
27312 aliases are also supported. The full naming scheme encodes three pieces
27313 of information in the following pattern:
27316 @var{architecture}-@var{vendor}-@var{os}
27319 For example, you can use the alias @code{sun4} as a @var{host} argument,
27320 or as the value for @var{target} in a @code{--target=@var{target}}
27321 option. The equivalent full name is @samp{sparc-sun-sunos4}.
27323 The @file{configure} script accompanying @value{GDBN} does not provide
27324 any query facility to list all supported host and target names or
27325 aliases. @file{configure} calls the Bourne shell script
27326 @code{config.sub} to map abbreviations to full names; you can read the
27327 script, if you wish, or you can use it to test your guesses on
27328 abbreviations---for example:
27331 % sh config.sub i386-linux
27333 % sh config.sub alpha-linux
27334 alpha-unknown-linux-gnu
27335 % sh config.sub hp9k700
27337 % sh config.sub sun4
27338 sparc-sun-sunos4.1.1
27339 % sh config.sub sun3
27340 m68k-sun-sunos4.1.1
27341 % sh config.sub i986v
27342 Invalid configuration `i986v': machine `i986v' not recognized
27346 @code{config.sub} is also distributed in the @value{GDBN} source
27347 directory (@file{gdb-@value{GDBVN}}, for version @value{GDBVN}).
27349 @node Configure Options
27350 @section @file{configure} Options
27352 Here is a summary of the @file{configure} options and arguments that
27353 are most often useful for building @value{GDBN}. @file{configure} also has
27354 several other options not listed here. @inforef{What Configure
27355 Does,,configure.info}, for a full explanation of @file{configure}.
27358 configure @r{[}--help@r{]}
27359 @r{[}--prefix=@var{dir}@r{]}
27360 @r{[}--exec-prefix=@var{dir}@r{]}
27361 @r{[}--srcdir=@var{dirname}@r{]}
27362 @r{[}--norecursion@r{]} @r{[}--rm@r{]}
27363 @r{[}--target=@var{target}@r{]}
27368 You may introduce options with a single @samp{-} rather than
27369 @samp{--} if you prefer; but you may abbreviate option names if you use
27374 Display a quick summary of how to invoke @file{configure}.
27376 @item --prefix=@var{dir}
27377 Configure the source to install programs and files under directory
27380 @item --exec-prefix=@var{dir}
27381 Configure the source to install programs under directory
27384 @c avoid splitting the warning from the explanation:
27386 @item --srcdir=@var{dirname}
27387 @strong{Warning: using this option requires @sc{gnu} @code{make}, or another
27388 @code{make} that implements the @code{VPATH} feature.}@*
27389 Use this option to make configurations in directories separate from the
27390 @value{GDBN} source directories. Among other things, you can use this to
27391 build (or maintain) several configurations simultaneously, in separate
27392 directories. @file{configure} writes configuration-specific files in
27393 the current directory, but arranges for them to use the source in the
27394 directory @var{dirname}. @file{configure} creates directories under
27395 the working directory in parallel to the source directories below
27398 @item --norecursion
27399 Configure only the directory level where @file{configure} is executed; do not
27400 propagate configuration to subdirectories.
27402 @item --target=@var{target}
27403 Configure @value{GDBN} for cross-debugging programs running on the specified
27404 @var{target}. Without this option, @value{GDBN} is configured to debug
27405 programs that run on the same machine (@var{host}) as @value{GDBN} itself.
27407 There is no convenient way to generate a list of all available targets.
27409 @item @var{host} @dots{}
27410 Configure @value{GDBN} to run on the specified @var{host}.
27412 There is no convenient way to generate a list of all available hosts.
27415 There are many other options available as well, but they are generally
27416 needed for special purposes only.
27418 @node System-wide configuration
27419 @section System-wide configuration and settings
27420 @cindex system-wide init file
27422 @value{GDBN} can be configured to have a system-wide init file;
27423 this file will be read and executed at startup (@pxref{Startup, , What
27424 @value{GDBN} does during startup}).
27426 Here is the corresponding configure option:
27429 @item --with-system-gdbinit=@var{file}
27430 Specify that the default location of the system-wide init file is
27434 If @value{GDBN} has been configured with the option @option{--prefix=$prefix},
27435 it may be subject to relocation. Two possible cases:
27439 If the default location of this init file contains @file{$prefix},
27440 it will be subject to relocation. Suppose that the configure options
27441 are @option{--prefix=$prefix --with-system-gdbinit=$prefix/etc/gdbinit};
27442 if @value{GDBN} is moved from @file{$prefix} to @file{$install}, the system
27443 init file is looked for as @file{$install/etc/gdbinit} instead of
27444 @file{$prefix/etc/gdbinit}.
27447 By contrast, if the default location does not contain the prefix,
27448 it will not be relocated. E.g.@: if @value{GDBN} has been configured with
27449 @option{--prefix=/usr/local --with-system-gdbinit=/usr/share/gdb/gdbinit},
27450 then @value{GDBN} will always look for @file{/usr/share/gdb/gdbinit},
27451 wherever @value{GDBN} is installed.
27454 @node Maintenance Commands
27455 @appendix Maintenance Commands
27456 @cindex maintenance commands
27457 @cindex internal commands
27459 In addition to commands intended for @value{GDBN} users, @value{GDBN}
27460 includes a number of commands intended for @value{GDBN} developers,
27461 that are not documented elsewhere in this manual. These commands are
27462 provided here for reference. (For commands that turn on debugging
27463 messages, see @ref{Debugging Output}.)
27466 @kindex maint agent
27467 @kindex maint agent-eval
27468 @item maint agent @var{expression}
27469 @itemx maint agent-eval @var{expression}
27470 Translate the given @var{expression} into remote agent bytecodes.
27471 This command is useful for debugging the Agent Expression mechanism
27472 (@pxref{Agent Expressions}). The @samp{agent} version produces an
27473 expression useful for data collection, such as by tracepoints, while
27474 @samp{maint agent-eval} produces an expression that evaluates directly
27475 to a result. For instance, a collection expression for @code{globa +
27476 globb} will include bytecodes to record four bytes of memory at each
27477 of the addresses of @code{globa} and @code{globb}, while discarding
27478 the result of the addition, while an evaluation expression will do the
27479 addition and return the sum.
27481 @kindex maint info breakpoints
27482 @item @anchor{maint info breakpoints}maint info breakpoints
27483 Using the same format as @samp{info breakpoints}, display both the
27484 breakpoints you've set explicitly, and those @value{GDBN} is using for
27485 internal purposes. Internal breakpoints are shown with negative
27486 breakpoint numbers. The type column identifies what kind of breakpoint
27491 Normal, explicitly set breakpoint.
27494 Normal, explicitly set watchpoint.
27497 Internal breakpoint, used to handle correctly stepping through
27498 @code{longjmp} calls.
27500 @item longjmp resume
27501 Internal breakpoint at the target of a @code{longjmp}.
27504 Temporary internal breakpoint used by the @value{GDBN} @code{until} command.
27507 Temporary internal breakpoint used by the @value{GDBN} @code{finish} command.
27510 Shared library events.
27514 @kindex set displaced-stepping
27515 @kindex show displaced-stepping
27516 @cindex displaced stepping support
27517 @cindex out-of-line single-stepping
27518 @item set displaced-stepping
27519 @itemx show displaced-stepping
27520 Control whether or not @value{GDBN} will do @dfn{displaced stepping}
27521 if the target supports it. Displaced stepping is a way to single-step
27522 over breakpoints without removing them from the inferior, by executing
27523 an out-of-line copy of the instruction that was originally at the
27524 breakpoint location. It is also known as out-of-line single-stepping.
27527 @item set displaced-stepping on
27528 If the target architecture supports it, @value{GDBN} will use
27529 displaced stepping to step over breakpoints.
27531 @item set displaced-stepping off
27532 @value{GDBN} will not use displaced stepping to step over breakpoints,
27533 even if such is supported by the target architecture.
27535 @cindex non-stop mode, and @samp{set displaced-stepping}
27536 @item set displaced-stepping auto
27537 This is the default mode. @value{GDBN} will use displaced stepping
27538 only if non-stop mode is active (@pxref{Non-Stop Mode}) and the target
27539 architecture supports displaced stepping.
27542 @kindex maint check-symtabs
27543 @item maint check-symtabs
27544 Check the consistency of psymtabs and symtabs.
27546 @kindex maint cplus first_component
27547 @item maint cplus first_component @var{name}
27548 Print the first C@t{++} class/namespace component of @var{name}.
27550 @kindex maint cplus namespace
27551 @item maint cplus namespace
27552 Print the list of possible C@t{++} namespaces.
27554 @kindex maint demangle
27555 @item maint demangle @var{name}
27556 Demangle a C@t{++} or Objective-C mangled @var{name}.
27558 @kindex maint deprecate
27559 @kindex maint undeprecate
27560 @cindex deprecated commands
27561 @item maint deprecate @var{command} @r{[}@var{replacement}@r{]}
27562 @itemx maint undeprecate @var{command}
27563 Deprecate or undeprecate the named @var{command}. Deprecated commands
27564 cause @value{GDBN} to issue a warning when you use them. The optional
27565 argument @var{replacement} says which newer command should be used in
27566 favor of the deprecated one; if it is given, @value{GDBN} will mention
27567 the replacement as part of the warning.
27569 @kindex maint dump-me
27570 @item maint dump-me
27571 @cindex @code{SIGQUIT} signal, dump core of @value{GDBN}
27572 Cause a fatal signal in the debugger and force it to dump its core.
27573 This is supported only on systems which support aborting a program
27574 with the @code{SIGQUIT} signal.
27576 @kindex maint internal-error
27577 @kindex maint internal-warning
27578 @item maint internal-error @r{[}@var{message-text}@r{]}
27579 @itemx maint internal-warning @r{[}@var{message-text}@r{]}
27580 Cause @value{GDBN} to call the internal function @code{internal_error}
27581 or @code{internal_warning} and hence behave as though an internal error
27582 or internal warning has been detected. In addition to reporting the
27583 internal problem, these functions give the user the opportunity to
27584 either quit @value{GDBN} or create a core file of the current
27585 @value{GDBN} session.
27587 These commands take an optional parameter @var{message-text} that is
27588 used as the text of the error or warning message.
27590 Here's an example of using @code{internal-error}:
27593 (@value{GDBP}) @kbd{maint internal-error testing, 1, 2}
27594 @dots{}/maint.c:121: internal-error: testing, 1, 2
27595 A problem internal to GDB has been detected. Further
27596 debugging may prove unreliable.
27597 Quit this debugging session? (y or n) @kbd{n}
27598 Create a core file? (y or n) @kbd{n}
27602 @cindex @value{GDBN} internal error
27603 @cindex internal errors, control of @value{GDBN} behavior
27605 @kindex maint set internal-error
27606 @kindex maint show internal-error
27607 @kindex maint set internal-warning
27608 @kindex maint show internal-warning
27609 @item maint set internal-error @var{action} [ask|yes|no]
27610 @itemx maint show internal-error @var{action}
27611 @itemx maint set internal-warning @var{action} [ask|yes|no]
27612 @itemx maint show internal-warning @var{action}
27613 When @value{GDBN} reports an internal problem (error or warning) it
27614 gives the user the opportunity to both quit @value{GDBN} and create a
27615 core file of the current @value{GDBN} session. These commands let you
27616 override the default behaviour for each particular @var{action},
27617 described in the table below.
27621 You can specify that @value{GDBN} should always (yes) or never (no)
27622 quit. The default is to ask the user what to do.
27625 You can specify that @value{GDBN} should always (yes) or never (no)
27626 create a core file. The default is to ask the user what to do.
27629 @kindex maint packet
27630 @item maint packet @var{text}
27631 If @value{GDBN} is talking to an inferior via the serial protocol,
27632 then this command sends the string @var{text} to the inferior, and
27633 displays the response packet. @value{GDBN} supplies the initial
27634 @samp{$} character, the terminating @samp{#} character, and the
27637 @kindex maint print architecture
27638 @item maint print architecture @r{[}@var{file}@r{]}
27639 Print the entire architecture configuration. The optional argument
27640 @var{file} names the file where the output goes.
27642 @kindex maint print c-tdesc
27643 @item maint print c-tdesc
27644 Print the current target description (@pxref{Target Descriptions}) as
27645 a C source file. The created source file can be used in @value{GDBN}
27646 when an XML parser is not available to parse the description.
27648 @kindex maint print dummy-frames
27649 @item maint print dummy-frames
27650 Prints the contents of @value{GDBN}'s internal dummy-frame stack.
27653 (@value{GDBP}) @kbd{b add}
27655 (@value{GDBP}) @kbd{print add(2,3)}
27656 Breakpoint 2, add (a=2, b=3) at @dots{}
27658 The program being debugged stopped while in a function called from GDB.
27660 (@value{GDBP}) @kbd{maint print dummy-frames}
27661 0x1a57c80: pc=0x01014068 fp=0x0200bddc sp=0x0200bdd6
27662 top=0x0200bdd4 id=@{stack=0x200bddc,code=0x101405c@}
27663 call_lo=0x01014000 call_hi=0x01014001
27667 Takes an optional file parameter.
27669 @kindex maint print registers
27670 @kindex maint print raw-registers
27671 @kindex maint print cooked-registers
27672 @kindex maint print register-groups
27673 @item maint print registers @r{[}@var{file}@r{]}
27674 @itemx maint print raw-registers @r{[}@var{file}@r{]}
27675 @itemx maint print cooked-registers @r{[}@var{file}@r{]}
27676 @itemx maint print register-groups @r{[}@var{file}@r{]}
27677 Print @value{GDBN}'s internal register data structures.
27679 The command @code{maint print raw-registers} includes the contents of
27680 the raw register cache; the command @code{maint print cooked-registers}
27681 includes the (cooked) value of all registers; and the command
27682 @code{maint print register-groups} includes the groups that each
27683 register is a member of. @xref{Registers,, Registers, gdbint,
27684 @value{GDBN} Internals}.
27686 These commands take an optional parameter, a file name to which to
27687 write the information.
27689 @kindex maint print reggroups
27690 @item maint print reggroups @r{[}@var{file}@r{]}
27691 Print @value{GDBN}'s internal register group data structures. The
27692 optional argument @var{file} tells to what file to write the
27695 The register groups info looks like this:
27698 (@value{GDBP}) @kbd{maint print reggroups}
27711 This command forces @value{GDBN} to flush its internal register cache.
27713 @kindex maint print objfiles
27714 @cindex info for known object files
27715 @item maint print objfiles
27716 Print a dump of all known object files. For each object file, this
27717 command prints its name, address in memory, and all of its psymtabs
27720 @kindex maint print statistics
27721 @cindex bcache statistics
27722 @item maint print statistics
27723 This command prints, for each object file in the program, various data
27724 about that object file followed by the byte cache (@dfn{bcache})
27725 statistics for the object file. The objfile data includes the number
27726 of minimal, partial, full, and stabs symbols, the number of types
27727 defined by the objfile, the number of as yet unexpanded psym tables,
27728 the number of line tables and string tables, and the amount of memory
27729 used by the various tables. The bcache statistics include the counts,
27730 sizes, and counts of duplicates of all and unique objects, max,
27731 average, and median entry size, total memory used and its overhead and
27732 savings, and various measures of the hash table size and chain
27735 @kindex maint print target-stack
27736 @cindex target stack description
27737 @item maint print target-stack
27738 A @dfn{target} is an interface between the debugger and a particular
27739 kind of file or process. Targets can be stacked in @dfn{strata},
27740 so that more than one target can potentially respond to a request.
27741 In particular, memory accesses will walk down the stack of targets
27742 until they find a target that is interested in handling that particular
27745 This command prints a short description of each layer that was pushed on
27746 the @dfn{target stack}, starting from the top layer down to the bottom one.
27748 @kindex maint print type
27749 @cindex type chain of a data type
27750 @item maint print type @var{expr}
27751 Print the type chain for a type specified by @var{expr}. The argument
27752 can be either a type name or a symbol. If it is a symbol, the type of
27753 that symbol is described. The type chain produced by this command is
27754 a recursive definition of the data type as stored in @value{GDBN}'s
27755 data structures, including its flags and contained types.
27757 @kindex maint set dwarf2 max-cache-age
27758 @kindex maint show dwarf2 max-cache-age
27759 @item maint set dwarf2 max-cache-age
27760 @itemx maint show dwarf2 max-cache-age
27761 Control the DWARF 2 compilation unit cache.
27763 @cindex DWARF 2 compilation units cache
27764 In object files with inter-compilation-unit references, such as those
27765 produced by the GCC option @samp{-feliminate-dwarf2-dups}, the DWARF 2
27766 reader needs to frequently refer to previously read compilation units.
27767 This setting controls how long a compilation unit will remain in the
27768 cache if it is not referenced. A higher limit means that cached
27769 compilation units will be stored in memory longer, and more total
27770 memory will be used. Setting it to zero disables caching, which will
27771 slow down @value{GDBN} startup, but reduce memory consumption.
27773 @kindex maint set profile
27774 @kindex maint show profile
27775 @cindex profiling GDB
27776 @item maint set profile
27777 @itemx maint show profile
27778 Control profiling of @value{GDBN}.
27780 Profiling will be disabled until you use the @samp{maint set profile}
27781 command to enable it. When you enable profiling, the system will begin
27782 collecting timing and execution count data; when you disable profiling or
27783 exit @value{GDBN}, the results will be written to a log file. Remember that
27784 if you use profiling, @value{GDBN} will overwrite the profiling log file
27785 (often called @file{gmon.out}). If you have a record of important profiling
27786 data in a @file{gmon.out} file, be sure to move it to a safe location.
27788 Configuring with @samp{--enable-profiling} arranges for @value{GDBN} to be
27789 compiled with the @samp{-pg} compiler option.
27791 @kindex maint set show-debug-regs
27792 @kindex maint show show-debug-regs
27793 @cindex hardware debug registers
27794 @item maint set show-debug-regs
27795 @itemx maint show show-debug-regs
27796 Control whether to show variables that mirror the hardware debug
27797 registers. Use @code{ON} to enable, @code{OFF} to disable. If
27798 enabled, the debug registers values are shown when @value{GDBN} inserts or
27799 removes a hardware breakpoint or watchpoint, and when the inferior
27800 triggers a hardware-assisted breakpoint or watchpoint.
27802 @kindex maint space
27803 @cindex memory used by commands
27805 Control whether to display memory usage for each command. If set to a
27806 nonzero value, @value{GDBN} will display how much memory each command
27807 took, following the command's own output. This can also be requested
27808 by invoking @value{GDBN} with the @option{--statistics} command-line
27809 switch (@pxref{Mode Options}).
27812 @cindex time of command execution
27814 Control whether to display the execution time for each command. If
27815 set to a nonzero value, @value{GDBN} will display how much time it
27816 took to execute each command, following the command's own output.
27817 The time is not printed for the commands that run the target, since
27818 there's no mechanism currently to compute how much time was spend
27819 by @value{GDBN} and how much time was spend by the program been debugged.
27820 it's not possibly currently
27821 This can also be requested by invoking @value{GDBN} with the
27822 @option{--statistics} command-line switch (@pxref{Mode Options}).
27824 @kindex maint translate-address
27825 @item maint translate-address @r{[}@var{section}@r{]} @var{addr}
27826 Find the symbol stored at the location specified by the address
27827 @var{addr} and an optional section name @var{section}. If found,
27828 @value{GDBN} prints the name of the closest symbol and an offset from
27829 the symbol's location to the specified address. This is similar to
27830 the @code{info address} command (@pxref{Symbols}), except that this
27831 command also allows to find symbols in other sections.
27833 If section was not specified, the section in which the symbol was found
27834 is also printed. For dynamically linked executables, the name of
27835 executable or shared library containing the symbol is printed as well.
27839 The following command is useful for non-interactive invocations of
27840 @value{GDBN}, such as in the test suite.
27843 @item set watchdog @var{nsec}
27844 @kindex set watchdog
27845 @cindex watchdog timer
27846 @cindex timeout for commands
27847 Set the maximum number of seconds @value{GDBN} will wait for the
27848 target operation to finish. If this time expires, @value{GDBN}
27849 reports and error and the command is aborted.
27851 @item show watchdog
27852 Show the current setting of the target wait timeout.
27855 @node Remote Protocol
27856 @appendix @value{GDBN} Remote Serial Protocol
27861 * Stop Reply Packets::
27862 * General Query Packets::
27863 * Register Packet Format::
27864 * Tracepoint Packets::
27865 * Host I/O Packets::
27867 * Notification Packets::
27868 * Remote Non-Stop::
27869 * Packet Acknowledgment::
27871 * File-I/O Remote Protocol Extension::
27872 * Library List Format::
27873 * Memory Map Format::
27879 There may be occasions when you need to know something about the
27880 protocol---for example, if there is only one serial port to your target
27881 machine, you might want your program to do something special if it
27882 recognizes a packet meant for @value{GDBN}.
27884 In the examples below, @samp{->} and @samp{<-} are used to indicate
27885 transmitted and received data, respectively.
27887 @cindex protocol, @value{GDBN} remote serial
27888 @cindex serial protocol, @value{GDBN} remote
27889 @cindex remote serial protocol
27890 All @value{GDBN} commands and responses (other than acknowledgments
27891 and notifications, see @ref{Notification Packets}) are sent as a
27892 @var{packet}. A @var{packet} is introduced with the character
27893 @samp{$}, the actual @var{packet-data}, and the terminating character
27894 @samp{#} followed by a two-digit @var{checksum}:
27897 @code{$}@var{packet-data}@code{#}@var{checksum}
27901 @cindex checksum, for @value{GDBN} remote
27903 The two-digit @var{checksum} is computed as the modulo 256 sum of all
27904 characters between the leading @samp{$} and the trailing @samp{#} (an
27905 eight bit unsigned checksum).
27907 Implementors should note that prior to @value{GDBN} 5.0 the protocol
27908 specification also included an optional two-digit @var{sequence-id}:
27911 @code{$}@var{sequence-id}@code{:}@var{packet-data}@code{#}@var{checksum}
27914 @cindex sequence-id, for @value{GDBN} remote
27916 That @var{sequence-id} was appended to the acknowledgment. @value{GDBN}
27917 has never output @var{sequence-id}s. Stubs that handle packets added
27918 since @value{GDBN} 5.0 must not accept @var{sequence-id}.
27920 When either the host or the target machine receives a packet, the first
27921 response expected is an acknowledgment: either @samp{+} (to indicate
27922 the package was received correctly) or @samp{-} (to request
27926 -> @code{$}@var{packet-data}@code{#}@var{checksum}
27931 The @samp{+}/@samp{-} acknowledgments can be disabled
27932 once a connection is established.
27933 @xref{Packet Acknowledgment}, for details.
27935 The host (@value{GDBN}) sends @var{command}s, and the target (the
27936 debugging stub incorporated in your program) sends a @var{response}. In
27937 the case of step and continue @var{command}s, the response is only sent
27938 when the operation has completed, and the target has again stopped all
27939 threads in all attached processes. This is the default all-stop mode
27940 behavior, but the remote protocol also supports @value{GDBN}'s non-stop
27941 execution mode; see @ref{Remote Non-Stop}, for details.
27943 @var{packet-data} consists of a sequence of characters with the
27944 exception of @samp{#} and @samp{$} (see @samp{X} packet for additional
27947 @cindex remote protocol, field separator
27948 Fields within the packet should be separated using @samp{,} @samp{;} or
27949 @samp{:}. Except where otherwise noted all numbers are represented in
27950 @sc{hex} with leading zeros suppressed.
27952 Implementors should note that prior to @value{GDBN} 5.0, the character
27953 @samp{:} could not appear as the third character in a packet (as it
27954 would potentially conflict with the @var{sequence-id}).
27956 @cindex remote protocol, binary data
27957 @anchor{Binary Data}
27958 Binary data in most packets is encoded either as two hexadecimal
27959 digits per byte of binary data. This allowed the traditional remote
27960 protocol to work over connections which were only seven-bit clean.
27961 Some packets designed more recently assume an eight-bit clean
27962 connection, and use a more efficient encoding to send and receive
27965 The binary data representation uses @code{7d} (@sc{ascii} @samp{@}})
27966 as an escape character. Any escaped byte is transmitted as the escape
27967 character followed by the original character XORed with @code{0x20}.
27968 For example, the byte @code{0x7d} would be transmitted as the two
27969 bytes @code{0x7d 0x5d}. The bytes @code{0x23} (@sc{ascii} @samp{#}),
27970 @code{0x24} (@sc{ascii} @samp{$}), and @code{0x7d} (@sc{ascii}
27971 @samp{@}}) must always be escaped. Responses sent by the stub
27972 must also escape @code{0x2a} (@sc{ascii} @samp{*}), so that it
27973 is not interpreted as the start of a run-length encoded sequence
27976 Response @var{data} can be run-length encoded to save space.
27977 Run-length encoding replaces runs of identical characters with one
27978 instance of the repeated character, followed by a @samp{*} and a
27979 repeat count. The repeat count is itself sent encoded, to avoid
27980 binary characters in @var{data}: a value of @var{n} is sent as
27981 @code{@var{n}+29}. For a repeat count greater or equal to 3, this
27982 produces a printable @sc{ascii} character, e.g.@: a space (@sc{ascii}
27983 code 32) for a repeat count of 3. (This is because run-length
27984 encoding starts to win for counts 3 or more.) Thus, for example,
27985 @samp{0* } is a run-length encoding of ``0000'': the space character
27986 after @samp{*} means repeat the leading @code{0} @w{@code{32 - 29 =
27989 The printable characters @samp{#} and @samp{$} or with a numeric value
27990 greater than 126 must not be used. Runs of six repeats (@samp{#}) or
27991 seven repeats (@samp{$}) can be expanded using a repeat count of only
27992 five (@samp{"}). For example, @samp{00000000} can be encoded as
27995 The error response returned for some packets includes a two character
27996 error number. That number is not well defined.
27998 @cindex empty response, for unsupported packets
27999 For any @var{command} not supported by the stub, an empty response
28000 (@samp{$#00}) should be returned. That way it is possible to extend the
28001 protocol. A newer @value{GDBN} can tell if a packet is supported based
28004 A stub is required to support the @samp{g}, @samp{G}, @samp{m}, @samp{M},
28005 @samp{c}, and @samp{s} @var{command}s. All other @var{command}s are
28011 The following table provides a complete list of all currently defined
28012 @var{command}s and their corresponding response @var{data}.
28013 @xref{File-I/O Remote Protocol Extension}, for details about the File
28014 I/O extension of the remote protocol.
28016 Each packet's description has a template showing the packet's overall
28017 syntax, followed by an explanation of the packet's meaning. We
28018 include spaces in some of the templates for clarity; these are not
28019 part of the packet's syntax. No @value{GDBN} packet uses spaces to
28020 separate its components. For example, a template like @samp{foo
28021 @var{bar} @var{baz}} describes a packet beginning with the three ASCII
28022 bytes @samp{foo}, followed by a @var{bar}, followed directly by a
28023 @var{baz}. @value{GDBN} does not transmit a space character between the
28024 @samp{foo} and the @var{bar}, or between the @var{bar} and the
28027 @cindex @var{thread-id}, in remote protocol
28028 @anchor{thread-id syntax}
28029 Several packets and replies include a @var{thread-id} field to identify
28030 a thread. Normally these are positive numbers with a target-specific
28031 interpretation, formatted as big-endian hex strings. A @var{thread-id}
28032 can also be a literal @samp{-1} to indicate all threads, or @samp{0} to
28035 In addition, the remote protocol supports a multiprocess feature in
28036 which the @var{thread-id} syntax is extended to optionally include both
28037 process and thread ID fields, as @samp{p@var{pid}.@var{tid}}.
28038 The @var{pid} (process) and @var{tid} (thread) components each have the
28039 format described above: a positive number with target-specific
28040 interpretation formatted as a big-endian hex string, literal @samp{-1}
28041 to indicate all processes or threads (respectively), or @samp{0} to
28042 indicate an arbitrary process or thread. Specifying just a process, as
28043 @samp{p@var{pid}}, is equivalent to @samp{p@var{pid}.-1}. It is an
28044 error to specify all processes but a specific thread, such as
28045 @samp{p-1.@var{tid}}. Note that the @samp{p} prefix is @emph{not} used
28046 for those packets and replies explicitly documented to include a process
28047 ID, rather than a @var{thread-id}.
28049 The multiprocess @var{thread-id} syntax extensions are only used if both
28050 @value{GDBN} and the stub report support for the @samp{multiprocess}
28051 feature using @samp{qSupported}. @xref{multiprocess extensions}, for
28054 Note that all packet forms beginning with an upper- or lower-case
28055 letter, other than those described here, are reserved for future use.
28057 Here are the packet descriptions.
28062 @cindex @samp{!} packet
28063 @anchor{extended mode}
28064 Enable extended mode. In extended mode, the remote server is made
28065 persistent. The @samp{R} packet is used to restart the program being
28071 The remote target both supports and has enabled extended mode.
28075 @cindex @samp{?} packet
28076 Indicate the reason the target halted. The reply is the same as for
28077 step and continue. This packet has a special interpretation when the
28078 target is in non-stop mode; see @ref{Remote Non-Stop}.
28081 @xref{Stop Reply Packets}, for the reply specifications.
28083 @item A @var{arglen},@var{argnum},@var{arg},@dots{}
28084 @cindex @samp{A} packet
28085 Initialized @code{argv[]} array passed into program. @var{arglen}
28086 specifies the number of bytes in the hex encoded byte stream
28087 @var{arg}. See @code{gdbserver} for more details.
28092 The arguments were set.
28098 @cindex @samp{b} packet
28099 (Don't use this packet; its behavior is not well-defined.)
28100 Change the serial line speed to @var{baud}.
28102 JTC: @emph{When does the transport layer state change? When it's
28103 received, or after the ACK is transmitted. In either case, there are
28104 problems if the command or the acknowledgment packet is dropped.}
28106 Stan: @emph{If people really wanted to add something like this, and get
28107 it working for the first time, they ought to modify ser-unix.c to send
28108 some kind of out-of-band message to a specially-setup stub and have the
28109 switch happen "in between" packets, so that from remote protocol's point
28110 of view, nothing actually happened.}
28112 @item B @var{addr},@var{mode}
28113 @cindex @samp{B} packet
28114 Set (@var{mode} is @samp{S}) or clear (@var{mode} is @samp{C}) a
28115 breakpoint at @var{addr}.
28117 Don't use this packet. Use the @samp{Z} and @samp{z} packets instead
28118 (@pxref{insert breakpoint or watchpoint packet}).
28120 @cindex @samp{bc} packet
28123 Backward continue. Execute the target system in reverse. No parameter.
28124 @xref{Reverse Execution}, for more information.
28127 @xref{Stop Reply Packets}, for the reply specifications.
28129 @cindex @samp{bs} packet
28132 Backward single step. Execute one instruction in reverse. No parameter.
28133 @xref{Reverse Execution}, for more information.
28136 @xref{Stop Reply Packets}, for the reply specifications.
28138 @item c @r{[}@var{addr}@r{]}
28139 @cindex @samp{c} packet
28140 Continue. @var{addr} is address to resume. If @var{addr} is omitted,
28141 resume at current address.
28144 @xref{Stop Reply Packets}, for the reply specifications.
28146 @item C @var{sig}@r{[};@var{addr}@r{]}
28147 @cindex @samp{C} packet
28148 Continue with signal @var{sig} (hex signal number). If
28149 @samp{;@var{addr}} is omitted, resume at same address.
28152 @xref{Stop Reply Packets}, for the reply specifications.
28155 @cindex @samp{d} packet
28158 Don't use this packet; instead, define a general set packet
28159 (@pxref{General Query Packets}).
28163 @cindex @samp{D} packet
28164 The first form of the packet is used to detach @value{GDBN} from the
28165 remote system. It is sent to the remote target
28166 before @value{GDBN} disconnects via the @code{detach} command.
28168 The second form, including a process ID, is used when multiprocess
28169 protocol extensions are enabled (@pxref{multiprocess extensions}), to
28170 detach only a specific process. The @var{pid} is specified as a
28171 big-endian hex string.
28181 @item F @var{RC},@var{EE},@var{CF};@var{XX}
28182 @cindex @samp{F} packet
28183 A reply from @value{GDBN} to an @samp{F} packet sent by the target.
28184 This is part of the File-I/O protocol extension. @xref{File-I/O
28185 Remote Protocol Extension}, for the specification.
28188 @anchor{read registers packet}
28189 @cindex @samp{g} packet
28190 Read general registers.
28194 @item @var{XX@dots{}}
28195 Each byte of register data is described by two hex digits. The bytes
28196 with the register are transmitted in target byte order. The size of
28197 each register and their position within the @samp{g} packet are
28198 determined by the @value{GDBN} internal gdbarch functions
28199 @code{DEPRECATED_REGISTER_RAW_SIZE} and @code{gdbarch_register_name}. The
28200 specification of several standard @samp{g} packets is specified below.
28205 @item G @var{XX@dots{}}
28206 @cindex @samp{G} packet
28207 Write general registers. @xref{read registers packet}, for a
28208 description of the @var{XX@dots{}} data.
28218 @item H @var{c} @var{thread-id}
28219 @cindex @samp{H} packet
28220 Set thread for subsequent operations (@samp{m}, @samp{M}, @samp{g},
28221 @samp{G}, et.al.). @var{c} depends on the operation to be performed: it
28222 should be @samp{c} for step and continue operations, @samp{g} for other
28223 operations. The thread designator @var{thread-id} has the format and
28224 interpretation described in @ref{thread-id syntax}.
28235 @c 'H': How restrictive (or permissive) is the thread model. If a
28236 @c thread is selected and stopped, are other threads allowed
28237 @c to continue to execute? As I mentioned above, I think the
28238 @c semantics of each command when a thread is selected must be
28239 @c described. For example:
28241 @c 'g': If the stub supports threads and a specific thread is
28242 @c selected, returns the register block from that thread;
28243 @c otherwise returns current registers.
28245 @c 'G' If the stub supports threads and a specific thread is
28246 @c selected, sets the registers of the register block of
28247 @c that thread; otherwise sets current registers.
28249 @item i @r{[}@var{addr}@r{[},@var{nnn}@r{]]}
28250 @anchor{cycle step packet}
28251 @cindex @samp{i} packet
28252 Step the remote target by a single clock cycle. If @samp{,@var{nnn}} is
28253 present, cycle step @var{nnn} cycles. If @var{addr} is present, cycle
28254 step starting at that address.
28257 @cindex @samp{I} packet
28258 Signal, then cycle step. @xref{step with signal packet}. @xref{cycle
28262 @cindex @samp{k} packet
28265 FIXME: @emph{There is no description of how to operate when a specific
28266 thread context has been selected (i.e.@: does 'k' kill only that
28269 @item m @var{addr},@var{length}
28270 @cindex @samp{m} packet
28271 Read @var{length} bytes of memory starting at address @var{addr}.
28272 Note that @var{addr} may not be aligned to any particular boundary.
28274 The stub need not use any particular size or alignment when gathering
28275 data from memory for the response; even if @var{addr} is word-aligned
28276 and @var{length} is a multiple of the word size, the stub is free to
28277 use byte accesses, or not. For this reason, this packet may not be
28278 suitable for accessing memory-mapped I/O devices.
28279 @cindex alignment of remote memory accesses
28280 @cindex size of remote memory accesses
28281 @cindex memory, alignment and size of remote accesses
28285 @item @var{XX@dots{}}
28286 Memory contents; each byte is transmitted as a two-digit hexadecimal
28287 number. The reply may contain fewer bytes than requested if the
28288 server was able to read only part of the region of memory.
28293 @item M @var{addr},@var{length}:@var{XX@dots{}}
28294 @cindex @samp{M} packet
28295 Write @var{length} bytes of memory starting at address @var{addr}.
28296 @var{XX@dots{}} is the data; each byte is transmitted as a two-digit
28297 hexadecimal number.
28304 for an error (this includes the case where only part of the data was
28309 @cindex @samp{p} packet
28310 Read the value of register @var{n}; @var{n} is in hex.
28311 @xref{read registers packet}, for a description of how the returned
28312 register value is encoded.
28316 @item @var{XX@dots{}}
28317 the register's value
28321 Indicating an unrecognized @var{query}.
28324 @item P @var{n@dots{}}=@var{r@dots{}}
28325 @anchor{write register packet}
28326 @cindex @samp{P} packet
28327 Write register @var{n@dots{}} with value @var{r@dots{}}. The register
28328 number @var{n} is in hexadecimal, and @var{r@dots{}} contains two hex
28329 digits for each byte in the register (target byte order).
28339 @item q @var{name} @var{params}@dots{}
28340 @itemx Q @var{name} @var{params}@dots{}
28341 @cindex @samp{q} packet
28342 @cindex @samp{Q} packet
28343 General query (@samp{q}) and set (@samp{Q}). These packets are
28344 described fully in @ref{General Query Packets}.
28347 @cindex @samp{r} packet
28348 Reset the entire system.
28350 Don't use this packet; use the @samp{R} packet instead.
28353 @cindex @samp{R} packet
28354 Restart the program being debugged. @var{XX}, while needed, is ignored.
28355 This packet is only available in extended mode (@pxref{extended mode}).
28357 The @samp{R} packet has no reply.
28359 @item s @r{[}@var{addr}@r{]}
28360 @cindex @samp{s} packet
28361 Single step. @var{addr} is the address at which to resume. If
28362 @var{addr} is omitted, resume at same address.
28365 @xref{Stop Reply Packets}, for the reply specifications.
28367 @item S @var{sig}@r{[};@var{addr}@r{]}
28368 @anchor{step with signal packet}
28369 @cindex @samp{S} packet
28370 Step with signal. This is analogous to the @samp{C} packet, but
28371 requests a single-step, rather than a normal resumption of execution.
28374 @xref{Stop Reply Packets}, for the reply specifications.
28376 @item t @var{addr}:@var{PP},@var{MM}
28377 @cindex @samp{t} packet
28378 Search backwards starting at address @var{addr} for a match with pattern
28379 @var{PP} and mask @var{MM}. @var{PP} and @var{MM} are 4 bytes.
28380 @var{addr} must be at least 3 digits.
28382 @item T @var{thread-id}
28383 @cindex @samp{T} packet
28384 Find out if the thread @var{thread-id} is alive. @xref{thread-id syntax}.
28389 thread is still alive
28395 Packets starting with @samp{v} are identified by a multi-letter name,
28396 up to the first @samp{;} or @samp{?} (or the end of the packet).
28398 @item vAttach;@var{pid}
28399 @cindex @samp{vAttach} packet
28400 Attach to a new process with the specified process ID @var{pid}.
28401 The process ID is a
28402 hexadecimal integer identifying the process. In all-stop mode, all
28403 threads in the attached process are stopped; in non-stop mode, it may be
28404 attached without being stopped if that is supported by the target.
28406 @c In non-stop mode, on a successful vAttach, the stub should set the
28407 @c current thread to a thread of the newly-attached process. After
28408 @c attaching, GDB queries for the attached process's thread ID with qC.
28409 @c Also note that, from a user perspective, whether or not the
28410 @c target is stopped on attach in non-stop mode depends on whether you
28411 @c use the foreground or background version of the attach command, not
28412 @c on what vAttach does; GDB does the right thing with respect to either
28413 @c stopping or restarting threads.
28415 This packet is only available in extended mode (@pxref{extended mode}).
28421 @item @r{Any stop packet}
28422 for success in all-stop mode (@pxref{Stop Reply Packets})
28424 for success in non-stop mode (@pxref{Remote Non-Stop})
28427 @item vCont@r{[};@var{action}@r{[}:@var{thread-id}@r{]]}@dots{}
28428 @cindex @samp{vCont} packet
28429 Resume the inferior, specifying different actions for each thread.
28430 If an action is specified with no @var{thread-id}, then it is applied to any
28431 threads that don't have a specific action specified; if no default action is
28432 specified then other threads should remain stopped in all-stop mode and
28433 in their current state in non-stop mode.
28434 Specifying multiple
28435 default actions is an error; specifying no actions is also an error.
28436 Thread IDs are specified using the syntax described in @ref{thread-id syntax}.
28438 Currently supported actions are:
28444 Continue with signal @var{sig}. The signal @var{sig} should be two hex digits.
28448 Step with signal @var{sig}. The signal @var{sig} should be two hex digits.
28453 The optional argument @var{addr} normally associated with the
28454 @samp{c}, @samp{C}, @samp{s}, and @samp{S} packets is
28455 not supported in @samp{vCont}.
28457 The @samp{t} action is only relevant in non-stop mode
28458 (@pxref{Remote Non-Stop}) and may be ignored by the stub otherwise.
28459 A stop reply should be generated for any affected thread not already stopped.
28460 When a thread is stopped by means of a @samp{t} action,
28461 the corresponding stop reply should indicate that the thread has stopped with
28462 signal @samp{0}, regardless of whether the target uses some other signal
28463 as an implementation detail.
28466 @xref{Stop Reply Packets}, for the reply specifications.
28469 @cindex @samp{vCont?} packet
28470 Request a list of actions supported by the @samp{vCont} packet.
28474 @item vCont@r{[};@var{action}@dots{}@r{]}
28475 The @samp{vCont} packet is supported. Each @var{action} is a supported
28476 command in the @samp{vCont} packet.
28478 The @samp{vCont} packet is not supported.
28481 @item vFile:@var{operation}:@var{parameter}@dots{}
28482 @cindex @samp{vFile} packet
28483 Perform a file operation on the target system. For details,
28484 see @ref{Host I/O Packets}.
28486 @item vFlashErase:@var{addr},@var{length}
28487 @cindex @samp{vFlashErase} packet
28488 Direct the stub to erase @var{length} bytes of flash starting at
28489 @var{addr}. The region may enclose any number of flash blocks, but
28490 its start and end must fall on block boundaries, as indicated by the
28491 flash block size appearing in the memory map (@pxref{Memory Map
28492 Format}). @value{GDBN} groups flash memory programming operations
28493 together, and sends a @samp{vFlashDone} request after each group; the
28494 stub is allowed to delay erase operation until the @samp{vFlashDone}
28495 packet is received.
28497 The stub must support @samp{vCont} if it reports support for
28498 multiprocess extensions (@pxref{multiprocess extensions}). Note that in
28499 this case @samp{vCont} actions can be specified to apply to all threads
28500 in a process by using the @samp{p@var{pid}.-1} form of the
28511 @item vFlashWrite:@var{addr}:@var{XX@dots{}}
28512 @cindex @samp{vFlashWrite} packet
28513 Direct the stub to write data to flash address @var{addr}. The data
28514 is passed in binary form using the same encoding as for the @samp{X}
28515 packet (@pxref{Binary Data}). The memory ranges specified by
28516 @samp{vFlashWrite} packets preceding a @samp{vFlashDone} packet must
28517 not overlap, and must appear in order of increasing addresses
28518 (although @samp{vFlashErase} packets for higher addresses may already
28519 have been received; the ordering is guaranteed only between
28520 @samp{vFlashWrite} packets). If a packet writes to an address that was
28521 neither erased by a preceding @samp{vFlashErase} packet nor by some other
28522 target-specific method, the results are unpredictable.
28530 for vFlashWrite addressing non-flash memory
28536 @cindex @samp{vFlashDone} packet
28537 Indicate to the stub that flash programming operation is finished.
28538 The stub is permitted to delay or batch the effects of a group of
28539 @samp{vFlashErase} and @samp{vFlashWrite} packets until a
28540 @samp{vFlashDone} packet is received. The contents of the affected
28541 regions of flash memory are unpredictable until the @samp{vFlashDone}
28542 request is completed.
28544 @item vKill;@var{pid}
28545 @cindex @samp{vKill} packet
28546 Kill the process with the specified process ID. @var{pid} is a
28547 hexadecimal integer identifying the process. This packet is used in
28548 preference to @samp{k} when multiprocess protocol extensions are
28549 supported; see @ref{multiprocess extensions}.
28559 @item vRun;@var{filename}@r{[};@var{argument}@r{]}@dots{}
28560 @cindex @samp{vRun} packet
28561 Run the program @var{filename}, passing it each @var{argument} on its
28562 command line. The file and arguments are hex-encoded strings. If
28563 @var{filename} is an empty string, the stub may use a default program
28564 (e.g.@: the last program run). The program is created in the stopped
28567 @c FIXME: What about non-stop mode?
28569 This packet is only available in extended mode (@pxref{extended mode}).
28575 @item @r{Any stop packet}
28576 for success (@pxref{Stop Reply Packets})
28580 @anchor{vStopped packet}
28581 @cindex @samp{vStopped} packet
28583 In non-stop mode (@pxref{Remote Non-Stop}), acknowledge a previous stop
28584 reply and prompt for the stub to report another one.
28588 @item @r{Any stop packet}
28589 if there is another unreported stop event (@pxref{Stop Reply Packets})
28591 if there are no unreported stop events
28594 @item X @var{addr},@var{length}:@var{XX@dots{}}
28596 @cindex @samp{X} packet
28597 Write data to memory, where the data is transmitted in binary.
28598 @var{addr} is address, @var{length} is number of bytes,
28599 @samp{@var{XX}@dots{}} is binary data (@pxref{Binary Data}).
28609 @item z @var{type},@var{addr},@var{length}
28610 @itemx Z @var{type},@var{addr},@var{length}
28611 @anchor{insert breakpoint or watchpoint packet}
28612 @cindex @samp{z} packet
28613 @cindex @samp{Z} packets
28614 Insert (@samp{Z}) or remove (@samp{z}) a @var{type} breakpoint or
28615 watchpoint starting at address @var{address} and covering the next
28616 @var{length} bytes.
28618 Each breakpoint and watchpoint packet @var{type} is documented
28621 @emph{Implementation notes: A remote target shall return an empty string
28622 for an unrecognized breakpoint or watchpoint packet @var{type}. A
28623 remote target shall support either both or neither of a given
28624 @samp{Z@var{type}@dots{}} and @samp{z@var{type}@dots{}} packet pair. To
28625 avoid potential problems with duplicate packets, the operations should
28626 be implemented in an idempotent way.}
28628 @item z0,@var{addr},@var{length}
28629 @itemx Z0,@var{addr},@var{length}
28630 @cindex @samp{z0} packet
28631 @cindex @samp{Z0} packet
28632 Insert (@samp{Z0}) or remove (@samp{z0}) a memory breakpoint at address
28633 @var{addr} of size @var{length}.
28635 A memory breakpoint is implemented by replacing the instruction at
28636 @var{addr} with a software breakpoint or trap instruction. The
28637 @var{length} is used by targets that indicates the size of the
28638 breakpoint (in bytes) that should be inserted (e.g., the @sc{arm} and
28639 @sc{mips} can insert either a 2 or 4 byte breakpoint).
28641 @emph{Implementation note: It is possible for a target to copy or move
28642 code that contains memory breakpoints (e.g., when implementing
28643 overlays). The behavior of this packet, in the presence of such a
28644 target, is not defined.}
28656 @item z1,@var{addr},@var{length}
28657 @itemx Z1,@var{addr},@var{length}
28658 @cindex @samp{z1} packet
28659 @cindex @samp{Z1} packet
28660 Insert (@samp{Z1}) or remove (@samp{z1}) a hardware breakpoint at
28661 address @var{addr} of size @var{length}.
28663 A hardware breakpoint is implemented using a mechanism that is not
28664 dependant on being able to modify the target's memory.
28666 @emph{Implementation note: A hardware breakpoint is not affected by code
28679 @item z2,@var{addr},@var{length}
28680 @itemx Z2,@var{addr},@var{length}
28681 @cindex @samp{z2} packet
28682 @cindex @samp{Z2} packet
28683 Insert (@samp{Z2}) or remove (@samp{z2}) a write watchpoint.
28695 @item z3,@var{addr},@var{length}
28696 @itemx Z3,@var{addr},@var{length}
28697 @cindex @samp{z3} packet
28698 @cindex @samp{Z3} packet
28699 Insert (@samp{Z3}) or remove (@samp{z3}) a read watchpoint.
28711 @item z4,@var{addr},@var{length}
28712 @itemx Z4,@var{addr},@var{length}
28713 @cindex @samp{z4} packet
28714 @cindex @samp{Z4} packet
28715 Insert (@samp{Z4}) or remove (@samp{z4}) an access watchpoint.
28729 @node Stop Reply Packets
28730 @section Stop Reply Packets
28731 @cindex stop reply packets
28733 The @samp{C}, @samp{c}, @samp{S}, @samp{s}, @samp{vCont},
28734 @samp{vAttach}, @samp{vRun}, @samp{vStopped}, and @samp{?} packets can
28735 receive any of the below as a reply. Except for @samp{?}
28736 and @samp{vStopped}, that reply is only returned
28737 when the target halts. In the below the exact meaning of @dfn{signal
28738 number} is defined by the header @file{include/gdb/signals.h} in the
28739 @value{GDBN} source code.
28741 As in the description of request packets, we include spaces in the
28742 reply templates for clarity; these are not part of the reply packet's
28743 syntax. No @value{GDBN} stop reply packet uses spaces to separate its
28749 The program received signal number @var{AA} (a two-digit hexadecimal
28750 number). This is equivalent to a @samp{T} response with no
28751 @var{n}:@var{r} pairs.
28753 @item T @var{AA} @var{n1}:@var{r1};@var{n2}:@var{r2};@dots{}
28754 @cindex @samp{T} packet reply
28755 The program received signal number @var{AA} (a two-digit hexadecimal
28756 number). This is equivalent to an @samp{S} response, except that the
28757 @samp{@var{n}:@var{r}} pairs can carry values of important registers
28758 and other information directly in the stop reply packet, reducing
28759 round-trip latency. Single-step and breakpoint traps are reported
28760 this way. Each @samp{@var{n}:@var{r}} pair is interpreted as follows:
28764 If @var{n} is a hexadecimal number, it is a register number, and the
28765 corresponding @var{r} gives that register's value. @var{r} is a
28766 series of bytes in target byte order, with each byte given by a
28767 two-digit hex number.
28770 If @var{n} is @samp{thread}, then @var{r} is the @var{thread-id} of
28771 the stopped thread, as specified in @ref{thread-id syntax}.
28774 If @var{n} is a recognized @dfn{stop reason}, it describes a more
28775 specific event that stopped the target. The currently defined stop
28776 reasons are listed below. @var{aa} should be @samp{05}, the trap
28777 signal. At most one stop reason should be present.
28780 Otherwise, @value{GDBN} should ignore this @samp{@var{n}:@var{r}} pair
28781 and go on to the next; this allows us to extend the protocol in the
28785 The currently defined stop reasons are:
28791 The packet indicates a watchpoint hit, and @var{r} is the data address, in
28794 @cindex shared library events, remote reply
28796 The packet indicates that the loaded libraries have changed.
28797 @value{GDBN} should use @samp{qXfer:libraries:read} to fetch a new
28798 list of loaded libraries. @var{r} is ignored.
28800 @cindex replay log events, remote reply
28802 The packet indicates that the target cannot continue replaying
28803 logged execution events, because it has reached the end (or the
28804 beginning when executing backward) of the log. The value of @var{r}
28805 will be either @samp{begin} or @samp{end}. @xref{Reverse Execution},
28806 for more information.
28812 @itemx W @var{AA} ; process:@var{pid}
28813 The process exited, and @var{AA} is the exit status. This is only
28814 applicable to certain targets.
28816 The second form of the response, including the process ID of the exited
28817 process, can be used only when @value{GDBN} has reported support for
28818 multiprocess protocol extensions; see @ref{multiprocess extensions}.
28819 The @var{pid} is formatted as a big-endian hex string.
28822 @itemx X @var{AA} ; process:@var{pid}
28823 The process terminated with signal @var{AA}.
28825 The second form of the response, including the process ID of the
28826 terminated process, can be used only when @value{GDBN} has reported
28827 support for multiprocess protocol extensions; see @ref{multiprocess
28828 extensions}. The @var{pid} is formatted as a big-endian hex string.
28830 @item O @var{XX}@dots{}
28831 @samp{@var{XX}@dots{}} is hex encoding of @sc{ascii} data, to be
28832 written as the program's console output. This can happen at any time
28833 while the program is running and the debugger should continue to wait
28834 for @samp{W}, @samp{T}, etc. This reply is not permitted in non-stop mode.
28836 @item F @var{call-id},@var{parameter}@dots{}
28837 @var{call-id} is the identifier which says which host system call should
28838 be called. This is just the name of the function. Translation into the
28839 correct system call is only applicable as it's defined in @value{GDBN}.
28840 @xref{File-I/O Remote Protocol Extension}, for a list of implemented
28843 @samp{@var{parameter}@dots{}} is a list of parameters as defined for
28844 this very system call.
28846 The target replies with this packet when it expects @value{GDBN} to
28847 call a host system call on behalf of the target. @value{GDBN} replies
28848 with an appropriate @samp{F} packet and keeps up waiting for the next
28849 reply packet from the target. The latest @samp{C}, @samp{c}, @samp{S}
28850 or @samp{s} action is expected to be continued. @xref{File-I/O Remote
28851 Protocol Extension}, for more details.
28855 @node General Query Packets
28856 @section General Query Packets
28857 @cindex remote query requests
28859 Packets starting with @samp{q} are @dfn{general query packets};
28860 packets starting with @samp{Q} are @dfn{general set packets}. General
28861 query and set packets are a semi-unified form for retrieving and
28862 sending information to and from the stub.
28864 The initial letter of a query or set packet is followed by a name
28865 indicating what sort of thing the packet applies to. For example,
28866 @value{GDBN} may use a @samp{qSymbol} packet to exchange symbol
28867 definitions with the stub. These packet names follow some
28872 The name must not contain commas, colons or semicolons.
28874 Most @value{GDBN} query and set packets have a leading upper case
28877 The names of custom vendor packets should use a company prefix, in
28878 lower case, followed by a period. For example, packets designed at
28879 the Acme Corporation might begin with @samp{qacme.foo} (for querying
28880 foos) or @samp{Qacme.bar} (for setting bars).
28883 The name of a query or set packet should be separated from any
28884 parameters by a @samp{:}; the parameters themselves should be
28885 separated by @samp{,} or @samp{;}. Stubs must be careful to match the
28886 full packet name, and check for a separator or the end of the packet,
28887 in case two packet names share a common prefix. New packets should not begin
28888 with @samp{qC}, @samp{qP}, or @samp{qL}@footnote{The @samp{qP} and @samp{qL}
28889 packets predate these conventions, and have arguments without any terminator
28890 for the packet name; we suspect they are in widespread use in places that
28891 are difficult to upgrade. The @samp{qC} packet has no arguments, but some
28892 existing stubs (e.g.@: RedBoot) are known to not check for the end of the
28895 Like the descriptions of the other packets, each description here
28896 has a template showing the packet's overall syntax, followed by an
28897 explanation of the packet's meaning. We include spaces in some of the
28898 templates for clarity; these are not part of the packet's syntax. No
28899 @value{GDBN} packet uses spaces to separate its components.
28901 Here are the currently defined query and set packets:
28906 @cindex current thread, remote request
28907 @cindex @samp{qC} packet
28908 Return the current thread ID.
28912 @item QC @var{thread-id}
28913 Where @var{thread-id} is a thread ID as documented in
28914 @ref{thread-id syntax}.
28915 @item @r{(anything else)}
28916 Any other reply implies the old thread ID.
28919 @item qCRC:@var{addr},@var{length}
28920 @cindex CRC of memory block, remote request
28921 @cindex @samp{qCRC} packet
28922 Compute the CRC checksum of a block of memory using CRC-32 defined in
28923 IEEE 802.3. The CRC is computed byte at a time, taking the most
28924 significant bit of each byte first. The initial pattern code
28925 @code{0xffffffff} is used to ensure leading zeros affect the CRC.
28927 @emph{Note:} This is the same CRC used in validating separate debug
28928 files (@pxref{Separate Debug Files, , Debugging Information in Separate
28929 Files}). However the algorithm is slightly different. When validating
28930 separate debug files, the CRC is computed taking the @emph{least}
28931 significant bit of each byte first, and the final result is inverted to
28932 detect trailing zeros.
28937 An error (such as memory fault)
28938 @item C @var{crc32}
28939 The specified memory region's checksum is @var{crc32}.
28943 @itemx qsThreadInfo
28944 @cindex list active threads, remote request
28945 @cindex @samp{qfThreadInfo} packet
28946 @cindex @samp{qsThreadInfo} packet
28947 Obtain a list of all active thread IDs from the target (OS). Since there
28948 may be too many active threads to fit into one reply packet, this query
28949 works iteratively: it may require more than one query/reply sequence to
28950 obtain the entire list of threads. The first query of the sequence will
28951 be the @samp{qfThreadInfo} query; subsequent queries in the
28952 sequence will be the @samp{qsThreadInfo} query.
28954 NOTE: This packet replaces the @samp{qL} query (see below).
28958 @item m @var{thread-id}
28960 @item m @var{thread-id},@var{thread-id}@dots{}
28961 a comma-separated list of thread IDs
28963 (lower case letter @samp{L}) denotes end of list.
28966 In response to each query, the target will reply with a list of one or
28967 more thread IDs, separated by commas.
28968 @value{GDBN} will respond to each reply with a request for more thread
28969 ids (using the @samp{qs} form of the query), until the target responds
28970 with @samp{l} (lower-case el, for @dfn{last}).
28971 Refer to @ref{thread-id syntax}, for the format of the @var{thread-id}
28974 @item qGetTLSAddr:@var{thread-id},@var{offset},@var{lm}
28975 @cindex get thread-local storage address, remote request
28976 @cindex @samp{qGetTLSAddr} packet
28977 Fetch the address associated with thread local storage specified
28978 by @var{thread-id}, @var{offset}, and @var{lm}.
28980 @var{thread-id} is the thread ID associated with the
28981 thread for which to fetch the TLS address. @xref{thread-id syntax}.
28983 @var{offset} is the (big endian, hex encoded) offset associated with the
28984 thread local variable. (This offset is obtained from the debug
28985 information associated with the variable.)
28987 @var{lm} is the (big endian, hex encoded) OS/ABI-specific encoding of the
28988 the load module associated with the thread local storage. For example,
28989 a @sc{gnu}/Linux system will pass the link map address of the shared
28990 object associated with the thread local storage under consideration.
28991 Other operating environments may choose to represent the load module
28992 differently, so the precise meaning of this parameter will vary.
28996 @item @var{XX}@dots{}
28997 Hex encoded (big endian) bytes representing the address of the thread
28998 local storage requested.
29001 An error occurred. @var{nn} are hex digits.
29004 An empty reply indicates that @samp{qGetTLSAddr} is not supported by the stub.
29007 @item qL @var{startflag} @var{threadcount} @var{nextthread}
29008 Obtain thread information from RTOS. Where: @var{startflag} (one hex
29009 digit) is one to indicate the first query and zero to indicate a
29010 subsequent query; @var{threadcount} (two hex digits) is the maximum
29011 number of threads the response packet can contain; and @var{nextthread}
29012 (eight hex digits), for subsequent queries (@var{startflag} is zero), is
29013 returned in the response as @var{argthread}.
29015 Don't use this packet; use the @samp{qfThreadInfo} query instead (see above).
29019 @item qM @var{count} @var{done} @var{argthread} @var{thread}@dots{}
29020 Where: @var{count} (two hex digits) is the number of threads being
29021 returned; @var{done} (one hex digit) is zero to indicate more threads
29022 and one indicates no further threads; @var{argthreadid} (eight hex
29023 digits) is @var{nextthread} from the request packet; @var{thread}@dots{}
29024 is a sequence of thread IDs from the target. @var{threadid} (eight hex
29025 digits). See @code{remote.c:parse_threadlist_response()}.
29029 @cindex section offsets, remote request
29030 @cindex @samp{qOffsets} packet
29031 Get section offsets that the target used when relocating the downloaded
29036 @item Text=@var{xxx};Data=@var{yyy}@r{[};Bss=@var{zzz}@r{]}
29037 Relocate the @code{Text} section by @var{xxx} from its original address.
29038 Relocate the @code{Data} section by @var{yyy} from its original address.
29039 If the object file format provides segment information (e.g.@: @sc{elf}
29040 @samp{PT_LOAD} program headers), @value{GDBN} will relocate entire
29041 segments by the supplied offsets.
29043 @emph{Note: while a @code{Bss} offset may be included in the response,
29044 @value{GDBN} ignores this and instead applies the @code{Data} offset
29045 to the @code{Bss} section.}
29047 @item TextSeg=@var{xxx}@r{[};DataSeg=@var{yyy}@r{]}
29048 Relocate the first segment of the object file, which conventionally
29049 contains program code, to a starting address of @var{xxx}. If
29050 @samp{DataSeg} is specified, relocate the second segment, which
29051 conventionally contains modifiable data, to a starting address of
29052 @var{yyy}. @value{GDBN} will report an error if the object file
29053 does not contain segment information, or does not contain at least
29054 as many segments as mentioned in the reply. Extra segments are
29055 kept at fixed offsets relative to the last relocated segment.
29058 @item qP @var{mode} @var{thread-id}
29059 @cindex thread information, remote request
29060 @cindex @samp{qP} packet
29061 Returns information on @var{thread-id}. Where: @var{mode} is a hex
29062 encoded 32 bit mode; @var{thread-id} is a thread ID
29063 (@pxref{thread-id syntax}).
29065 Don't use this packet; use the @samp{qThreadExtraInfo} query instead
29068 Reply: see @code{remote.c:remote_unpack_thread_info_response()}.
29072 @cindex non-stop mode, remote request
29073 @cindex @samp{QNonStop} packet
29075 Enter non-stop (@samp{QNonStop:1}) or all-stop (@samp{QNonStop:0}) mode.
29076 @xref{Remote Non-Stop}, for more information.
29081 The request succeeded.
29084 An error occurred. @var{nn} are hex digits.
29087 An empty reply indicates that @samp{QNonStop} is not supported by
29091 This packet is not probed by default; the remote stub must request it,
29092 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
29093 Use of this packet is controlled by the @code{set non-stop} command;
29094 @pxref{Non-Stop Mode}.
29096 @item QPassSignals: @var{signal} @r{[};@var{signal}@r{]}@dots{}
29097 @cindex pass signals to inferior, remote request
29098 @cindex @samp{QPassSignals} packet
29099 @anchor{QPassSignals}
29100 Each listed @var{signal} should be passed directly to the inferior process.
29101 Signals are numbered identically to continue packets and stop replies
29102 (@pxref{Stop Reply Packets}). Each @var{signal} list item should be
29103 strictly greater than the previous item. These signals do not need to stop
29104 the inferior, or be reported to @value{GDBN}. All other signals should be
29105 reported to @value{GDBN}. Multiple @samp{QPassSignals} packets do not
29106 combine; any earlier @samp{QPassSignals} list is completely replaced by the
29107 new list. This packet improves performance when using @samp{handle
29108 @var{signal} nostop noprint pass}.
29113 The request succeeded.
29116 An error occurred. @var{nn} are hex digits.
29119 An empty reply indicates that @samp{QPassSignals} is not supported by
29123 Use of this packet is controlled by the @code{set remote pass-signals}
29124 command (@pxref{Remote Configuration, set remote pass-signals}).
29125 This packet is not probed by default; the remote stub must request it,
29126 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
29128 @item qRcmd,@var{command}
29129 @cindex execute remote command, remote request
29130 @cindex @samp{qRcmd} packet
29131 @var{command} (hex encoded) is passed to the local interpreter for
29132 execution. Invalid commands should be reported using the output
29133 string. Before the final result packet, the target may also respond
29134 with a number of intermediate @samp{O@var{output}} console output
29135 packets. @emph{Implementors should note that providing access to a
29136 stubs's interpreter may have security implications}.
29141 A command response with no output.
29143 A command response with the hex encoded output string @var{OUTPUT}.
29145 Indicate a badly formed request.
29147 An empty reply indicates that @samp{qRcmd} is not recognized.
29150 (Note that the @code{qRcmd} packet's name is separated from the
29151 command by a @samp{,}, not a @samp{:}, contrary to the naming
29152 conventions above. Please don't use this packet as a model for new
29155 @item qSearch:memory:@var{address};@var{length};@var{search-pattern}
29156 @cindex searching memory, in remote debugging
29157 @cindex @samp{qSearch:memory} packet
29158 @anchor{qSearch memory}
29159 Search @var{length} bytes at @var{address} for @var{search-pattern}.
29160 @var{address} and @var{length} are encoded in hex.
29161 @var{search-pattern} is a sequence of bytes, hex encoded.
29166 The pattern was not found.
29168 The pattern was found at @var{address}.
29170 A badly formed request or an error was encountered while searching memory.
29172 An empty reply indicates that @samp{qSearch:memory} is not recognized.
29175 @item QStartNoAckMode
29176 @cindex @samp{QStartNoAckMode} packet
29177 @anchor{QStartNoAckMode}
29178 Request that the remote stub disable the normal @samp{+}/@samp{-}
29179 protocol acknowledgments (@pxref{Packet Acknowledgment}).
29184 The stub has switched to no-acknowledgment mode.
29185 @value{GDBN} acknowledges this reponse,
29186 but neither the stub nor @value{GDBN} shall send or expect further
29187 @samp{+}/@samp{-} acknowledgments in the current connection.
29189 An empty reply indicates that the stub does not support no-acknowledgment mode.
29192 @item qSupported @r{[}:@var{gdbfeature} @r{[};@var{gdbfeature}@r{]}@dots{} @r{]}
29193 @cindex supported packets, remote query
29194 @cindex features of the remote protocol
29195 @cindex @samp{qSupported} packet
29196 @anchor{qSupported}
29197 Tell the remote stub about features supported by @value{GDBN}, and
29198 query the stub for features it supports. This packet allows
29199 @value{GDBN} and the remote stub to take advantage of each others'
29200 features. @samp{qSupported} also consolidates multiple feature probes
29201 at startup, to improve @value{GDBN} performance---a single larger
29202 packet performs better than multiple smaller probe packets on
29203 high-latency links. Some features may enable behavior which must not
29204 be on by default, e.g.@: because it would confuse older clients or
29205 stubs. Other features may describe packets which could be
29206 automatically probed for, but are not. These features must be
29207 reported before @value{GDBN} will use them. This ``default
29208 unsupported'' behavior is not appropriate for all packets, but it
29209 helps to keep the initial connection time under control with new
29210 versions of @value{GDBN} which support increasing numbers of packets.
29214 @item @var{stubfeature} @r{[};@var{stubfeature}@r{]}@dots{}
29215 The stub supports or does not support each returned @var{stubfeature},
29216 depending on the form of each @var{stubfeature} (see below for the
29219 An empty reply indicates that @samp{qSupported} is not recognized,
29220 or that no features needed to be reported to @value{GDBN}.
29223 The allowed forms for each feature (either a @var{gdbfeature} in the
29224 @samp{qSupported} packet, or a @var{stubfeature} in the response)
29228 @item @var{name}=@var{value}
29229 The remote protocol feature @var{name} is supported, and associated
29230 with the specified @var{value}. The format of @var{value} depends
29231 on the feature, but it must not include a semicolon.
29233 The remote protocol feature @var{name} is supported, and does not
29234 need an associated value.
29236 The remote protocol feature @var{name} is not supported.
29238 The remote protocol feature @var{name} may be supported, and
29239 @value{GDBN} should auto-detect support in some other way when it is
29240 needed. This form will not be used for @var{gdbfeature} notifications,
29241 but may be used for @var{stubfeature} responses.
29244 Whenever the stub receives a @samp{qSupported} request, the
29245 supplied set of @value{GDBN} features should override any previous
29246 request. This allows @value{GDBN} to put the stub in a known
29247 state, even if the stub had previously been communicating with
29248 a different version of @value{GDBN}.
29250 The following values of @var{gdbfeature} (for the packet sent by @value{GDBN})
29255 This feature indicates whether @value{GDBN} supports multiprocess
29256 extensions to the remote protocol. @value{GDBN} does not use such
29257 extensions unless the stub also reports that it supports them by
29258 including @samp{multiprocess+} in its @samp{qSupported} reply.
29259 @xref{multiprocess extensions}, for details.
29262 Stubs should ignore any unknown values for
29263 @var{gdbfeature}. Any @value{GDBN} which sends a @samp{qSupported}
29264 packet supports receiving packets of unlimited length (earlier
29265 versions of @value{GDBN} may reject overly long responses). Additional values
29266 for @var{gdbfeature} may be defined in the future to let the stub take
29267 advantage of new features in @value{GDBN}, e.g.@: incompatible
29268 improvements in the remote protocol---the @samp{multiprocess} feature is
29269 an example of such a feature. The stub's reply should be independent
29270 of the @var{gdbfeature} entries sent by @value{GDBN}; first @value{GDBN}
29271 describes all the features it supports, and then the stub replies with
29272 all the features it supports.
29274 Similarly, @value{GDBN} will silently ignore unrecognized stub feature
29275 responses, as long as each response uses one of the standard forms.
29277 Some features are flags. A stub which supports a flag feature
29278 should respond with a @samp{+} form response. Other features
29279 require values, and the stub should respond with an @samp{=}
29282 Each feature has a default value, which @value{GDBN} will use if
29283 @samp{qSupported} is not available or if the feature is not mentioned
29284 in the @samp{qSupported} response. The default values are fixed; a
29285 stub is free to omit any feature responses that match the defaults.
29287 Not all features can be probed, but for those which can, the probing
29288 mechanism is useful: in some cases, a stub's internal
29289 architecture may not allow the protocol layer to know some information
29290 about the underlying target in advance. This is especially common in
29291 stubs which may be configured for multiple targets.
29293 These are the currently defined stub features and their properties:
29295 @multitable @columnfractions 0.35 0.2 0.12 0.2
29296 @c NOTE: The first row should be @headitem, but we do not yet require
29297 @c a new enough version of Texinfo (4.7) to use @headitem.
29299 @tab Value Required
29303 @item @samp{PacketSize}
29308 @item @samp{qXfer:auxv:read}
29313 @item @samp{qXfer:features:read}
29318 @item @samp{qXfer:libraries:read}
29323 @item @samp{qXfer:memory-map:read}
29328 @item @samp{qXfer:spu:read}
29333 @item @samp{qXfer:spu:write}
29338 @item @samp{qXfer:siginfo:read}
29343 @item @samp{qXfer:siginfo:write}
29348 @item @samp{QNonStop}
29353 @item @samp{QPassSignals}
29358 @item @samp{QStartNoAckMode}
29363 @item @samp{multiprocess}
29368 @item @samp{ConditionalTracepoints}
29373 @item @samp{ReverseContinue}
29378 @item @samp{ReverseStep}
29385 These are the currently defined stub features, in more detail:
29388 @cindex packet size, remote protocol
29389 @item PacketSize=@var{bytes}
29390 The remote stub can accept packets up to at least @var{bytes} in
29391 length. @value{GDBN} will send packets up to this size for bulk
29392 transfers, and will never send larger packets. This is a limit on the
29393 data characters in the packet, including the frame and checksum.
29394 There is no trailing NUL byte in a remote protocol packet; if the stub
29395 stores packets in a NUL-terminated format, it should allow an extra
29396 byte in its buffer for the NUL. If this stub feature is not supported,
29397 @value{GDBN} guesses based on the size of the @samp{g} packet response.
29399 @item qXfer:auxv:read
29400 The remote stub understands the @samp{qXfer:auxv:read} packet
29401 (@pxref{qXfer auxiliary vector read}).
29403 @item qXfer:features:read
29404 The remote stub understands the @samp{qXfer:features:read} packet
29405 (@pxref{qXfer target description read}).
29407 @item qXfer:libraries:read
29408 The remote stub understands the @samp{qXfer:libraries:read} packet
29409 (@pxref{qXfer library list read}).
29411 @item qXfer:memory-map:read
29412 The remote stub understands the @samp{qXfer:memory-map:read} packet
29413 (@pxref{qXfer memory map read}).
29415 @item qXfer:spu:read
29416 The remote stub understands the @samp{qXfer:spu:read} packet
29417 (@pxref{qXfer spu read}).
29419 @item qXfer:spu:write
29420 The remote stub understands the @samp{qXfer:spu:write} packet
29421 (@pxref{qXfer spu write}).
29423 @item qXfer:siginfo:read
29424 The remote stub understands the @samp{qXfer:siginfo:read} packet
29425 (@pxref{qXfer siginfo read}).
29427 @item qXfer:siginfo:write
29428 The remote stub understands the @samp{qXfer:siginfo:write} packet
29429 (@pxref{qXfer siginfo write}).
29432 The remote stub understands the @samp{QNonStop} packet
29433 (@pxref{QNonStop}).
29436 The remote stub understands the @samp{QPassSignals} packet
29437 (@pxref{QPassSignals}).
29439 @item QStartNoAckMode
29440 The remote stub understands the @samp{QStartNoAckMode} packet and
29441 prefers to operate in no-acknowledgment mode. @xref{Packet Acknowledgment}.
29444 @anchor{multiprocess extensions}
29445 @cindex multiprocess extensions, in remote protocol
29446 The remote stub understands the multiprocess extensions to the remote
29447 protocol syntax. The multiprocess extensions affect the syntax of
29448 thread IDs in both packets and replies (@pxref{thread-id syntax}), and
29449 add process IDs to the @samp{D} packet and @samp{W} and @samp{X}
29450 replies. Note that reporting this feature indicates support for the
29451 syntactic extensions only, not that the stub necessarily supports
29452 debugging of more than one process at a time. The stub must not use
29453 multiprocess extensions in packet replies unless @value{GDBN} has also
29454 indicated it supports them in its @samp{qSupported} request.
29456 @item qXfer:osdata:read
29457 The remote stub understands the @samp{qXfer:osdata:read} packet
29458 ((@pxref{qXfer osdata read}).
29460 @item ConditionalTracepoints
29461 The remote stub accepts and implements conditional expressions defined
29462 for tracepoints (@pxref{Tracepoint Conditions}).
29464 @item ReverseContinue
29465 The remote stub accepts and implements the reverse continue packet
29469 The remote stub accepts and implements the reverse step packet
29475 @cindex symbol lookup, remote request
29476 @cindex @samp{qSymbol} packet
29477 Notify the target that @value{GDBN} is prepared to serve symbol lookup
29478 requests. Accept requests from the target for the values of symbols.
29483 The target does not need to look up any (more) symbols.
29484 @item qSymbol:@var{sym_name}
29485 The target requests the value of symbol @var{sym_name} (hex encoded).
29486 @value{GDBN} may provide the value by using the
29487 @samp{qSymbol:@var{sym_value}:@var{sym_name}} message, described
29491 @item qSymbol:@var{sym_value}:@var{sym_name}
29492 Set the value of @var{sym_name} to @var{sym_value}.
29494 @var{sym_name} (hex encoded) is the name of a symbol whose value the
29495 target has previously requested.
29497 @var{sym_value} (hex) is the value for symbol @var{sym_name}. If
29498 @value{GDBN} cannot supply a value for @var{sym_name}, then this field
29504 The target does not need to look up any (more) symbols.
29505 @item qSymbol:@var{sym_name}
29506 The target requests the value of a new symbol @var{sym_name} (hex
29507 encoded). @value{GDBN} will continue to supply the values of symbols
29508 (if available), until the target ceases to request them.
29513 @xref{Tracepoint Packets}.
29515 @item qThreadExtraInfo,@var{thread-id}
29516 @cindex thread attributes info, remote request
29517 @cindex @samp{qThreadExtraInfo} packet
29518 Obtain a printable string description of a thread's attributes from
29519 the target OS. @var{thread-id} is a thread ID;
29520 see @ref{thread-id syntax}. This
29521 string may contain anything that the target OS thinks is interesting
29522 for @value{GDBN} to tell the user about the thread. The string is
29523 displayed in @value{GDBN}'s @code{info threads} display. Some
29524 examples of possible thread extra info strings are @samp{Runnable}, or
29525 @samp{Blocked on Mutex}.
29529 @item @var{XX}@dots{}
29530 Where @samp{@var{XX}@dots{}} is a hex encoding of @sc{ascii} data,
29531 comprising the printable string containing the extra information about
29532 the thread's attributes.
29535 (Note that the @code{qThreadExtraInfo} packet's name is separated from
29536 the command by a @samp{,}, not a @samp{:}, contrary to the naming
29537 conventions above. Please don't use this packet as a model for new
29545 @xref{Tracepoint Packets}.
29547 @item qXfer:@var{object}:read:@var{annex}:@var{offset},@var{length}
29548 @cindex read special object, remote request
29549 @cindex @samp{qXfer} packet
29550 @anchor{qXfer read}
29551 Read uninterpreted bytes from the target's special data area
29552 identified by the keyword @var{object}. Request @var{length} bytes
29553 starting at @var{offset} bytes into the data. The content and
29554 encoding of @var{annex} is specific to @var{object}; it can supply
29555 additional details about what data to access.
29557 Here are the specific requests of this form defined so far. All
29558 @samp{qXfer:@var{object}:read:@dots{}} requests use the same reply
29559 formats, listed below.
29562 @item qXfer:auxv:read::@var{offset},@var{length}
29563 @anchor{qXfer auxiliary vector read}
29564 Access the target's @dfn{auxiliary vector}. @xref{OS Information,
29565 auxiliary vector}. Note @var{annex} must be empty.
29567 This packet is not probed by default; the remote stub must request it,
29568 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
29570 @item qXfer:features:read:@var{annex}:@var{offset},@var{length}
29571 @anchor{qXfer target description read}
29572 Access the @dfn{target description}. @xref{Target Descriptions}. The
29573 annex specifies which XML document to access. The main description is
29574 always loaded from the @samp{target.xml} annex.
29576 This packet is not probed by default; the remote stub must request it,
29577 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
29579 @item qXfer:libraries:read:@var{annex}:@var{offset},@var{length}
29580 @anchor{qXfer library list read}
29581 Access the target's list of loaded libraries. @xref{Library List Format}.
29582 The annex part of the generic @samp{qXfer} packet must be empty
29583 (@pxref{qXfer read}).
29585 Targets which maintain a list of libraries in the program's memory do
29586 not need to implement this packet; it is designed for platforms where
29587 the operating system manages the list of loaded libraries.
29589 This packet is not probed by default; the remote stub must request it,
29590 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
29592 @item qXfer:memory-map:read::@var{offset},@var{length}
29593 @anchor{qXfer memory map read}
29594 Access the target's @dfn{memory-map}. @xref{Memory Map Format}. The
29595 annex part of the generic @samp{qXfer} packet must be empty
29596 (@pxref{qXfer read}).
29598 This packet is not probed by default; the remote stub must request it,
29599 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
29601 @item qXfer:siginfo:read::@var{offset},@var{length}
29602 @anchor{qXfer siginfo read}
29603 Read contents of the extra signal information on the target
29604 system. The annex part of the generic @samp{qXfer} packet must be
29605 empty (@pxref{qXfer read}).
29607 This packet is not probed by default; the remote stub must request it,
29608 by supplying an appropriate @samp{qSupported} response
29609 (@pxref{qSupported}).
29611 @item qXfer:spu:read:@var{annex}:@var{offset},@var{length}
29612 @anchor{qXfer spu read}
29613 Read contents of an @code{spufs} file on the target system. The
29614 annex specifies which file to read; it must be of the form
29615 @file{@var{id}/@var{name}}, where @var{id} specifies an SPU context ID
29616 in the target process, and @var{name} identifes the @code{spufs} file
29617 in that context to be accessed.
29619 This packet is not probed by default; the remote stub must request it,
29620 by supplying an appropriate @samp{qSupported} response
29621 (@pxref{qSupported}).
29623 @item qXfer:osdata:read::@var{offset},@var{length}
29624 @anchor{qXfer osdata read}
29625 Access the target's @dfn{operating system information}.
29626 @xref{Operating System Information}.
29633 Data @var{data} (@pxref{Binary Data}) has been read from the
29634 target. There may be more data at a higher address (although
29635 it is permitted to return @samp{m} even for the last valid
29636 block of data, as long as at least one byte of data was read).
29637 @var{data} may have fewer bytes than the @var{length} in the
29641 Data @var{data} (@pxref{Binary Data}) has been read from the target.
29642 There is no more data to be read. @var{data} may have fewer bytes
29643 than the @var{length} in the request.
29646 The @var{offset} in the request is at the end of the data.
29647 There is no more data to be read.
29650 The request was malformed, or @var{annex} was invalid.
29653 The offset was invalid, or there was an error encountered reading the data.
29654 @var{nn} is a hex-encoded @code{errno} value.
29657 An empty reply indicates the @var{object} string was not recognized by
29658 the stub, or that the object does not support reading.
29661 @item qXfer:@var{object}:write:@var{annex}:@var{offset}:@var{data}@dots{}
29662 @cindex write data into object, remote request
29663 @anchor{qXfer write}
29664 Write uninterpreted bytes into the target's special data area
29665 identified by the keyword @var{object}, starting at @var{offset} bytes
29666 into the data. @var{data}@dots{} is the binary-encoded data
29667 (@pxref{Binary Data}) to be written. The content and encoding of @var{annex}
29668 is specific to @var{object}; it can supply additional details about what data
29671 Here are the specific requests of this form defined so far. All
29672 @samp{qXfer:@var{object}:write:@dots{}} requests use the same reply
29673 formats, listed below.
29676 @item qXfer:siginfo:write::@var{offset}:@var{data}@dots{}
29677 @anchor{qXfer siginfo write}
29678 Write @var{data} to the extra signal information on the target system.
29679 The annex part of the generic @samp{qXfer} packet must be
29680 empty (@pxref{qXfer write}).
29682 This packet is not probed by default; the remote stub must request it,
29683 by supplying an appropriate @samp{qSupported} response
29684 (@pxref{qSupported}).
29686 @item qXfer:spu:write:@var{annex}:@var{offset}:@var{data}@dots{}
29687 @anchor{qXfer spu write}
29688 Write @var{data} to an @code{spufs} file on the target system. The
29689 annex specifies which file to write; it must be of the form
29690 @file{@var{id}/@var{name}}, where @var{id} specifies an SPU context ID
29691 in the target process, and @var{name} identifes the @code{spufs} file
29692 in that context to be accessed.
29694 This packet is not probed by default; the remote stub must request it,
29695 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
29701 @var{nn} (hex encoded) is the number of bytes written.
29702 This may be fewer bytes than supplied in the request.
29705 The request was malformed, or @var{annex} was invalid.
29708 The offset was invalid, or there was an error encountered writing the data.
29709 @var{nn} is a hex-encoded @code{errno} value.
29712 An empty reply indicates the @var{object} string was not
29713 recognized by the stub, or that the object does not support writing.
29716 @item qXfer:@var{object}:@var{operation}:@dots{}
29717 Requests of this form may be added in the future. When a stub does
29718 not recognize the @var{object} keyword, or its support for
29719 @var{object} does not recognize the @var{operation} keyword, the stub
29720 must respond with an empty packet.
29722 @item qAttached:@var{pid}
29723 @cindex query attached, remote request
29724 @cindex @samp{qAttached} packet
29725 Return an indication of whether the remote server attached to an
29726 existing process or created a new process. When the multiprocess
29727 protocol extensions are supported (@pxref{multiprocess extensions}),
29728 @var{pid} is an integer in hexadecimal format identifying the target
29729 process. Otherwise, @value{GDBN} will omit the @var{pid} field and
29730 the query packet will be simplified as @samp{qAttached}.
29732 This query is used, for example, to know whether the remote process
29733 should be detached or killed when a @value{GDBN} session is ended with
29734 the @code{quit} command.
29739 The remote server attached to an existing process.
29741 The remote server created a new process.
29743 A badly formed request or an error was encountered.
29748 @node Register Packet Format
29749 @section Register Packet Format
29751 The following @code{g}/@code{G} packets have previously been defined.
29752 In the below, some thirty-two bit registers are transferred as
29753 sixty-four bits. Those registers should be zero/sign extended (which?)
29754 to fill the space allocated. Register bytes are transferred in target
29755 byte order. The two nibbles within a register byte are transferred
29756 most-significant - least-significant.
29762 All registers are transferred as thirty-two bit quantities in the order:
29763 32 general-purpose; sr; lo; hi; bad; cause; pc; 32 floating-point
29764 registers; fsr; fir; fp.
29768 All registers are transferred as sixty-four bit quantities (including
29769 thirty-two bit registers such as @code{sr}). The ordering is the same
29774 @node Tracepoint Packets
29775 @section Tracepoint Packets
29776 @cindex tracepoint packets
29777 @cindex packets, tracepoint
29779 Here we describe the packets @value{GDBN} uses to implement
29780 tracepoints (@pxref{Tracepoints}).
29784 @item QTDP:@var{n}:@var{addr}:@var{ena}:@var{step}:@var{pass}[:X@var{len},@var{bytes}]@r{[}-@r{]}
29785 Create a new tracepoint, number @var{n}, at @var{addr}. If @var{ena}
29786 is @samp{E}, then the tracepoint is enabled; if it is @samp{D}, then
29787 the tracepoint is disabled. @var{step} is the tracepoint's step
29788 count, and @var{pass} is its pass count. If an @samp{X} is present,
29789 it introduces a tracepoint condition, which consists of a hexadecimal
29790 length, followed by a comma and hex-encoded bytes, in a manner similar
29791 to action encodings as described below. If the trailing @samp{-} is
29792 present, further @samp{QTDP} packets will follow to specify this
29793 tracepoint's actions.
29798 The packet was understood and carried out.
29800 The packet was not recognized.
29803 @item QTDP:-@var{n}:@var{addr}:@r{[}S@r{]}@var{action}@dots{}@r{[}-@r{]}
29804 Define actions to be taken when a tracepoint is hit. @var{n} and
29805 @var{addr} must be the same as in the initial @samp{QTDP} packet for
29806 this tracepoint. This packet may only be sent immediately after
29807 another @samp{QTDP} packet that ended with a @samp{-}. If the
29808 trailing @samp{-} is present, further @samp{QTDP} packets will follow,
29809 specifying more actions for this tracepoint.
29811 In the series of action packets for a given tracepoint, at most one
29812 can have an @samp{S} before its first @var{action}. If such a packet
29813 is sent, it and the following packets define ``while-stepping''
29814 actions. Any prior packets define ordinary actions --- that is, those
29815 taken when the tracepoint is first hit. If no action packet has an
29816 @samp{S}, then all the packets in the series specify ordinary
29817 tracepoint actions.
29819 The @samp{@var{action}@dots{}} portion of the packet is a series of
29820 actions, concatenated without separators. Each action has one of the
29826 Collect the registers whose bits are set in @var{mask}. @var{mask} is
29827 a hexadecimal number whose @var{i}'th bit is set if register number
29828 @var{i} should be collected. (The least significant bit is numbered
29829 zero.) Note that @var{mask} may be any number of digits long; it may
29830 not fit in a 32-bit word.
29832 @item M @var{basereg},@var{offset},@var{len}
29833 Collect @var{len} bytes of memory starting at the address in register
29834 number @var{basereg}, plus @var{offset}. If @var{basereg} is
29835 @samp{-1}, then the range has a fixed address: @var{offset} is the
29836 address of the lowest byte to collect. The @var{basereg},
29837 @var{offset}, and @var{len} parameters are all unsigned hexadecimal
29838 values (the @samp{-1} value for @var{basereg} is a special case).
29840 @item X @var{len},@var{expr}
29841 Evaluate @var{expr}, whose length is @var{len}, and collect memory as
29842 it directs. @var{expr} is an agent expression, as described in
29843 @ref{Agent Expressions}. Each byte of the expression is encoded as a
29844 two-digit hex number in the packet; @var{len} is the number of bytes
29845 in the expression (and thus one-half the number of hex digits in the
29850 Any number of actions may be packed together in a single @samp{QTDP}
29851 packet, as long as the packet does not exceed the maximum packet
29852 length (400 bytes, for many stubs). There may be only one @samp{R}
29853 action per tracepoint, and it must precede any @samp{M} or @samp{X}
29854 actions. Any registers referred to by @samp{M} and @samp{X} actions
29855 must be collected by a preceding @samp{R} action. (The
29856 ``while-stepping'' actions are treated as if they were attached to a
29857 separate tracepoint, as far as these restrictions are concerned.)
29862 The packet was understood and carried out.
29864 The packet was not recognized.
29867 @item QTFrame:@var{n}
29868 Select the @var{n}'th tracepoint frame from the buffer, and use the
29869 register and memory contents recorded there to answer subsequent
29870 request packets from @value{GDBN}.
29872 A successful reply from the stub indicates that the stub has found the
29873 requested frame. The response is a series of parts, concatenated
29874 without separators, describing the frame we selected. Each part has
29875 one of the following forms:
29879 The selected frame is number @var{n} in the trace frame buffer;
29880 @var{f} is a hexadecimal number. If @var{f} is @samp{-1}, then there
29881 was no frame matching the criteria in the request packet.
29884 The selected trace frame records a hit of tracepoint number @var{t};
29885 @var{t} is a hexadecimal number.
29889 @item QTFrame:pc:@var{addr}
29890 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
29891 currently selected frame whose PC is @var{addr};
29892 @var{addr} is a hexadecimal number.
29894 @item QTFrame:tdp:@var{t}
29895 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
29896 currently selected frame that is a hit of tracepoint @var{t}; @var{t}
29897 is a hexadecimal number.
29899 @item QTFrame:range:@var{start}:@var{end}
29900 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
29901 currently selected frame whose PC is between @var{start} (inclusive)
29902 and @var{end} (exclusive); @var{start} and @var{end} are hexadecimal
29905 @item QTFrame:outside:@var{start}:@var{end}
29906 Like @samp{QTFrame:range:@var{start}:@var{end}}, but select the first
29907 frame @emph{outside} the given range of addresses.
29910 Begin the tracepoint experiment. Begin collecting data from tracepoint
29911 hits in the trace frame buffer.
29914 End the tracepoint experiment. Stop collecting trace frames.
29917 Clear the table of tracepoints, and empty the trace frame buffer.
29919 @item QTro:@var{start1},@var{end1}:@var{start2},@var{end2}:@dots{}
29920 Establish the given ranges of memory as ``transparent''. The stub
29921 will answer requests for these ranges from memory's current contents,
29922 if they were not collected as part of the tracepoint hit.
29924 @value{GDBN} uses this to mark read-only regions of memory, like those
29925 containing program code. Since these areas never change, they should
29926 still have the same contents they did when the tracepoint was hit, so
29927 there's no reason for the stub to refuse to provide their contents.
29930 Ask the stub if there is a trace experiment running right now.
29935 There is no trace experiment running.
29937 There is a trace experiment running.
29943 @node Host I/O Packets
29944 @section Host I/O Packets
29945 @cindex Host I/O, remote protocol
29946 @cindex file transfer, remote protocol
29948 The @dfn{Host I/O} packets allow @value{GDBN} to perform I/O
29949 operations on the far side of a remote link. For example, Host I/O is
29950 used to upload and download files to a remote target with its own
29951 filesystem. Host I/O uses the same constant values and data structure
29952 layout as the target-initiated File-I/O protocol. However, the
29953 Host I/O packets are structured differently. The target-initiated
29954 protocol relies on target memory to store parameters and buffers.
29955 Host I/O requests are initiated by @value{GDBN}, and the
29956 target's memory is not involved. @xref{File-I/O Remote Protocol
29957 Extension}, for more details on the target-initiated protocol.
29959 The Host I/O request packets all encode a single operation along with
29960 its arguments. They have this format:
29964 @item vFile:@var{operation}: @var{parameter}@dots{}
29965 @var{operation} is the name of the particular request; the target
29966 should compare the entire packet name up to the second colon when checking
29967 for a supported operation. The format of @var{parameter} depends on
29968 the operation. Numbers are always passed in hexadecimal. Negative
29969 numbers have an explicit minus sign (i.e.@: two's complement is not
29970 used). Strings (e.g.@: filenames) are encoded as a series of
29971 hexadecimal bytes. The last argument to a system call may be a
29972 buffer of escaped binary data (@pxref{Binary Data}).
29976 The valid responses to Host I/O packets are:
29980 @item F @var{result} [, @var{errno}] [; @var{attachment}]
29981 @var{result} is the integer value returned by this operation, usually
29982 non-negative for success and -1 for errors. If an error has occured,
29983 @var{errno} will be included in the result. @var{errno} will have a
29984 value defined by the File-I/O protocol (@pxref{Errno Values}). For
29985 operations which return data, @var{attachment} supplies the data as a
29986 binary buffer. Binary buffers in response packets are escaped in the
29987 normal way (@pxref{Binary Data}). See the individual packet
29988 documentation for the interpretation of @var{result} and
29992 An empty response indicates that this operation is not recognized.
29996 These are the supported Host I/O operations:
29999 @item vFile:open: @var{pathname}, @var{flags}, @var{mode}
30000 Open a file at @var{pathname} and return a file descriptor for it, or
30001 return -1 if an error occurs. @var{pathname} is a string,
30002 @var{flags} is an integer indicating a mask of open flags
30003 (@pxref{Open Flags}), and @var{mode} is an integer indicating a mask
30004 of mode bits to use if the file is created (@pxref{mode_t Values}).
30005 @xref{open}, for details of the open flags and mode values.
30007 @item vFile:close: @var{fd}
30008 Close the open file corresponding to @var{fd} and return 0, or
30009 -1 if an error occurs.
30011 @item vFile:pread: @var{fd}, @var{count}, @var{offset}
30012 Read data from the open file corresponding to @var{fd}. Up to
30013 @var{count} bytes will be read from the file, starting at @var{offset}
30014 relative to the start of the file. The target may read fewer bytes;
30015 common reasons include packet size limits and an end-of-file
30016 condition. The number of bytes read is returned. Zero should only be
30017 returned for a successful read at the end of the file, or if
30018 @var{count} was zero.
30020 The data read should be returned as a binary attachment on success.
30021 If zero bytes were read, the response should include an empty binary
30022 attachment (i.e.@: a trailing semicolon). The return value is the
30023 number of target bytes read; the binary attachment may be longer if
30024 some characters were escaped.
30026 @item vFile:pwrite: @var{fd}, @var{offset}, @var{data}
30027 Write @var{data} (a binary buffer) to the open file corresponding
30028 to @var{fd}. Start the write at @var{offset} from the start of the
30029 file. Unlike many @code{write} system calls, there is no
30030 separate @var{count} argument; the length of @var{data} in the
30031 packet is used. @samp{vFile:write} returns the number of bytes written,
30032 which may be shorter than the length of @var{data}, or -1 if an
30035 @item vFile:unlink: @var{pathname}
30036 Delete the file at @var{pathname} on the target. Return 0,
30037 or -1 if an error occurs. @var{pathname} is a string.
30042 @section Interrupts
30043 @cindex interrupts (remote protocol)
30045 When a program on the remote target is running, @value{GDBN} may
30046 attempt to interrupt it by sending a @samp{Ctrl-C} or a @code{BREAK},
30047 control of which is specified via @value{GDBN}'s @samp{remotebreak}
30048 setting (@pxref{set remotebreak}).
30050 The precise meaning of @code{BREAK} is defined by the transport
30051 mechanism and may, in fact, be undefined. @value{GDBN} does not
30052 currently define a @code{BREAK} mechanism for any of the network
30053 interfaces except for TCP, in which case @value{GDBN} sends the
30054 @code{telnet} BREAK sequence.
30056 @samp{Ctrl-C}, on the other hand, is defined and implemented for all
30057 transport mechanisms. It is represented by sending the single byte
30058 @code{0x03} without any of the usual packet overhead described in
30059 the Overview section (@pxref{Overview}). When a @code{0x03} byte is
30060 transmitted as part of a packet, it is considered to be packet data
30061 and does @emph{not} represent an interrupt. E.g., an @samp{X} packet
30062 (@pxref{X packet}), used for binary downloads, may include an unescaped
30063 @code{0x03} as part of its packet.
30065 Stubs are not required to recognize these interrupt mechanisms and the
30066 precise meaning associated with receipt of the interrupt is
30067 implementation defined. If the target supports debugging of multiple
30068 threads and/or processes, it should attempt to interrupt all
30069 currently-executing threads and processes.
30070 If the stub is successful at interrupting the
30071 running program, it should send one of the stop
30072 reply packets (@pxref{Stop Reply Packets}) to @value{GDBN} as a result
30073 of successfully stopping the program in all-stop mode, and a stop reply
30074 for each stopped thread in non-stop mode.
30075 Interrupts received while the
30076 program is stopped are discarded.
30078 @node Notification Packets
30079 @section Notification Packets
30080 @cindex notification packets
30081 @cindex packets, notification
30083 The @value{GDBN} remote serial protocol includes @dfn{notifications},
30084 packets that require no acknowledgment. Both the GDB and the stub
30085 may send notifications (although the only notifications defined at
30086 present are sent by the stub). Notifications carry information
30087 without incurring the round-trip latency of an acknowledgment, and so
30088 are useful for low-impact communications where occasional packet loss
30091 A notification packet has the form @samp{% @var{data} #
30092 @var{checksum}}, where @var{data} is the content of the notification,
30093 and @var{checksum} is a checksum of @var{data}, computed and formatted
30094 as for ordinary @value{GDBN} packets. A notification's @var{data}
30095 never contains @samp{$}, @samp{%} or @samp{#} characters. Upon
30096 receiving a notification, the recipient sends no @samp{+} or @samp{-}
30097 to acknowledge the notification's receipt or to report its corruption.
30099 Every notification's @var{data} begins with a name, which contains no
30100 colon characters, followed by a colon character.
30102 Recipients should silently ignore corrupted notifications and
30103 notifications they do not understand. Recipients should restart
30104 timeout periods on receipt of a well-formed notification, whether or
30105 not they understand it.
30107 Senders should only send the notifications described here when this
30108 protocol description specifies that they are permitted. In the
30109 future, we may extend the protocol to permit existing notifications in
30110 new contexts; this rule helps older senders avoid confusing newer
30113 (Older versions of @value{GDBN} ignore bytes received until they see
30114 the @samp{$} byte that begins an ordinary packet, so new stubs may
30115 transmit notifications without fear of confusing older clients. There
30116 are no notifications defined for @value{GDBN} to send at the moment, but we
30117 assume that most older stubs would ignore them, as well.)
30119 The following notification packets from the stub to @value{GDBN} are
30123 @item Stop: @var{reply}
30124 Report an asynchronous stop event in non-stop mode.
30125 The @var{reply} has the form of a stop reply, as
30126 described in @ref{Stop Reply Packets}. Refer to @ref{Remote Non-Stop},
30127 for information on how these notifications are acknowledged by
30131 @node Remote Non-Stop
30132 @section Remote Protocol Support for Non-Stop Mode
30134 @value{GDBN}'s remote protocol supports non-stop debugging of
30135 multi-threaded programs, as described in @ref{Non-Stop Mode}. If the stub
30136 supports non-stop mode, it should report that to @value{GDBN} by including
30137 @samp{QNonStop+} in its @samp{qSupported} response (@pxref{qSupported}).
30139 @value{GDBN} typically sends a @samp{QNonStop} packet only when
30140 establishing a new connection with the stub. Entering non-stop mode
30141 does not alter the state of any currently-running threads, but targets
30142 must stop all threads in any already-attached processes when entering
30143 all-stop mode. @value{GDBN} uses the @samp{?} packet as necessary to
30144 probe the target state after a mode change.
30146 In non-stop mode, when an attached process encounters an event that
30147 would otherwise be reported with a stop reply, it uses the
30148 asynchronous notification mechanism (@pxref{Notification Packets}) to
30149 inform @value{GDBN}. In contrast to all-stop mode, where all threads
30150 in all processes are stopped when a stop reply is sent, in non-stop
30151 mode only the thread reporting the stop event is stopped. That is,
30152 when reporting a @samp{S} or @samp{T} response to indicate completion
30153 of a step operation, hitting a breakpoint, or a fault, only the
30154 affected thread is stopped; any other still-running threads continue
30155 to run. When reporting a @samp{W} or @samp{X} response, all running
30156 threads belonging to other attached processes continue to run.
30158 Only one stop reply notification at a time may be pending; if
30159 additional stop events occur before @value{GDBN} has acknowledged the
30160 previous notification, they must be queued by the stub for later
30161 synchronous transmission in response to @samp{vStopped} packets from
30162 @value{GDBN}. Because the notification mechanism is unreliable,
30163 the stub is permitted to resend a stop reply notification
30164 if it believes @value{GDBN} may not have received it. @value{GDBN}
30165 ignores additional stop reply notifications received before it has
30166 finished processing a previous notification and the stub has completed
30167 sending any queued stop events.
30169 Otherwise, @value{GDBN} must be prepared to receive a stop reply
30170 notification at any time. Specifically, they may appear when
30171 @value{GDBN} is not otherwise reading input from the stub, or when
30172 @value{GDBN} is expecting to read a normal synchronous response or a
30173 @samp{+}/@samp{-} acknowledgment to a packet it has sent.
30174 Notification packets are distinct from any other communication from
30175 the stub so there is no ambiguity.
30177 After receiving a stop reply notification, @value{GDBN} shall
30178 acknowledge it by sending a @samp{vStopped} packet (@pxref{vStopped packet})
30179 as a regular, synchronous request to the stub. Such acknowledgment
30180 is not required to happen immediately, as @value{GDBN} is permitted to
30181 send other, unrelated packets to the stub first, which the stub should
30184 Upon receiving a @samp{vStopped} packet, if the stub has other queued
30185 stop events to report to @value{GDBN}, it shall respond by sending a
30186 normal stop reply response. @value{GDBN} shall then send another
30187 @samp{vStopped} packet to solicit further responses; again, it is
30188 permitted to send other, unrelated packets as well which the stub
30189 should process normally.
30191 If the stub receives a @samp{vStopped} packet and there are no
30192 additional stop events to report, the stub shall return an @samp{OK}
30193 response. At this point, if further stop events occur, the stub shall
30194 send a new stop reply notification, @value{GDBN} shall accept the
30195 notification, and the process shall be repeated.
30197 In non-stop mode, the target shall respond to the @samp{?} packet as
30198 follows. First, any incomplete stop reply notification/@samp{vStopped}
30199 sequence in progress is abandoned. The target must begin a new
30200 sequence reporting stop events for all stopped threads, whether or not
30201 it has previously reported those events to @value{GDBN}. The first
30202 stop reply is sent as a synchronous reply to the @samp{?} packet, and
30203 subsequent stop replies are sent as responses to @samp{vStopped} packets
30204 using the mechanism described above. The target must not send
30205 asynchronous stop reply notifications until the sequence is complete.
30206 If all threads are running when the target receives the @samp{?} packet,
30207 or if the target is not attached to any process, it shall respond
30210 @node Packet Acknowledgment
30211 @section Packet Acknowledgment
30213 @cindex acknowledgment, for @value{GDBN} remote
30214 @cindex packet acknowledgment, for @value{GDBN} remote
30215 By default, when either the host or the target machine receives a packet,
30216 the first response expected is an acknowledgment: either @samp{+} (to indicate
30217 the package was received correctly) or @samp{-} (to request retransmission).
30218 This mechanism allows the @value{GDBN} remote protocol to operate over
30219 unreliable transport mechanisms, such as a serial line.
30221 In cases where the transport mechanism is itself reliable (such as a pipe or
30222 TCP connection), the @samp{+}/@samp{-} acknowledgments are redundant.
30223 It may be desirable to disable them in that case to reduce communication
30224 overhead, or for other reasons. This can be accomplished by means of the
30225 @samp{QStartNoAckMode} packet; @pxref{QStartNoAckMode}.
30227 When in no-acknowledgment mode, neither the stub nor @value{GDBN} shall send or
30228 expect @samp{+}/@samp{-} protocol acknowledgments. The packet
30229 and response format still includes the normal checksum, as described in
30230 @ref{Overview}, but the checksum may be ignored by the receiver.
30232 If the stub supports @samp{QStartNoAckMode} and prefers to operate in
30233 no-acknowledgment mode, it should report that to @value{GDBN}
30234 by including @samp{QStartNoAckMode+} in its response to @samp{qSupported};
30235 @pxref{qSupported}.
30236 If @value{GDBN} also supports @samp{QStartNoAckMode} and it has not been
30237 disabled via the @code{set remote noack-packet off} command
30238 (@pxref{Remote Configuration}),
30239 @value{GDBN} may then send a @samp{QStartNoAckMode} packet to the stub.
30240 Only then may the stub actually turn off packet acknowledgments.
30241 @value{GDBN} sends a final @samp{+} acknowledgment of the stub's @samp{OK}
30242 response, which can be safely ignored by the stub.
30244 Note that @code{set remote noack-packet} command only affects negotiation
30245 between @value{GDBN} and the stub when subsequent connections are made;
30246 it does not affect the protocol acknowledgment state for any current
30248 Since @samp{+}/@samp{-} acknowledgments are enabled by default when a
30249 new connection is established,
30250 there is also no protocol request to re-enable the acknowledgments
30251 for the current connection, once disabled.
30256 Example sequence of a target being re-started. Notice how the restart
30257 does not get any direct output:
30262 @emph{target restarts}
30265 <- @code{T001:1234123412341234}
30269 Example sequence of a target being stepped by a single instruction:
30272 -> @code{G1445@dots{}}
30277 <- @code{T001:1234123412341234}
30281 <- @code{1455@dots{}}
30285 @node File-I/O Remote Protocol Extension
30286 @section File-I/O Remote Protocol Extension
30287 @cindex File-I/O remote protocol extension
30290 * File-I/O Overview::
30291 * Protocol Basics::
30292 * The F Request Packet::
30293 * The F Reply Packet::
30294 * The Ctrl-C Message::
30296 * List of Supported Calls::
30297 * Protocol-specific Representation of Datatypes::
30299 * File-I/O Examples::
30302 @node File-I/O Overview
30303 @subsection File-I/O Overview
30304 @cindex file-i/o overview
30306 The @dfn{File I/O remote protocol extension} (short: File-I/O) allows the
30307 target to use the host's file system and console I/O to perform various
30308 system calls. System calls on the target system are translated into a
30309 remote protocol packet to the host system, which then performs the needed
30310 actions and returns a response packet to the target system.
30311 This simulates file system operations even on targets that lack file systems.
30313 The protocol is defined to be independent of both the host and target systems.
30314 It uses its own internal representation of datatypes and values. Both
30315 @value{GDBN} and the target's @value{GDBN} stub are responsible for
30316 translating the system-dependent value representations into the internal
30317 protocol representations when data is transmitted.
30319 The communication is synchronous. A system call is possible only when
30320 @value{GDBN} is waiting for a response from the @samp{C}, @samp{c}, @samp{S}
30321 or @samp{s} packets. While @value{GDBN} handles the request for a system call,
30322 the target is stopped to allow deterministic access to the target's
30323 memory. Therefore File-I/O is not interruptible by target signals. On
30324 the other hand, it is possible to interrupt File-I/O by a user interrupt
30325 (@samp{Ctrl-C}) within @value{GDBN}.
30327 The target's request to perform a host system call does not finish
30328 the latest @samp{C}, @samp{c}, @samp{S} or @samp{s} action. That means,
30329 after finishing the system call, the target returns to continuing the
30330 previous activity (continue, step). No additional continue or step
30331 request from @value{GDBN} is required.
30334 (@value{GDBP}) continue
30335 <- target requests 'system call X'
30336 target is stopped, @value{GDBN} executes system call
30337 -> @value{GDBN} returns result
30338 ... target continues, @value{GDBN} returns to wait for the target
30339 <- target hits breakpoint and sends a Txx packet
30342 The protocol only supports I/O on the console and to regular files on
30343 the host file system. Character or block special devices, pipes,
30344 named pipes, sockets or any other communication method on the host
30345 system are not supported by this protocol.
30347 File I/O is not supported in non-stop mode.
30349 @node Protocol Basics
30350 @subsection Protocol Basics
30351 @cindex protocol basics, file-i/o
30353 The File-I/O protocol uses the @code{F} packet as the request as well
30354 as reply packet. Since a File-I/O system call can only occur when
30355 @value{GDBN} is waiting for a response from the continuing or stepping target,
30356 the File-I/O request is a reply that @value{GDBN} has to expect as a result
30357 of a previous @samp{C}, @samp{c}, @samp{S} or @samp{s} packet.
30358 This @code{F} packet contains all information needed to allow @value{GDBN}
30359 to call the appropriate host system call:
30363 A unique identifier for the requested system call.
30366 All parameters to the system call. Pointers are given as addresses
30367 in the target memory address space. Pointers to strings are given as
30368 pointer/length pair. Numerical values are given as they are.
30369 Numerical control flags are given in a protocol-specific representation.
30373 At this point, @value{GDBN} has to perform the following actions.
30377 If the parameters include pointer values to data needed as input to a
30378 system call, @value{GDBN} requests this data from the target with a
30379 standard @code{m} packet request. This additional communication has to be
30380 expected by the target implementation and is handled as any other @code{m}
30384 @value{GDBN} translates all value from protocol representation to host
30385 representation as needed. Datatypes are coerced into the host types.
30388 @value{GDBN} calls the system call.
30391 It then coerces datatypes back to protocol representation.
30394 If the system call is expected to return data in buffer space specified
30395 by pointer parameters to the call, the data is transmitted to the
30396 target using a @code{M} or @code{X} packet. This packet has to be expected
30397 by the target implementation and is handled as any other @code{M} or @code{X}
30402 Eventually @value{GDBN} replies with another @code{F} packet which contains all
30403 necessary information for the target to continue. This at least contains
30410 @code{errno}, if has been changed by the system call.
30417 After having done the needed type and value coercion, the target continues
30418 the latest continue or step action.
30420 @node The F Request Packet
30421 @subsection The @code{F} Request Packet
30422 @cindex file-i/o request packet
30423 @cindex @code{F} request packet
30425 The @code{F} request packet has the following format:
30428 @item F@var{call-id},@var{parameter@dots{}}
30430 @var{call-id} is the identifier to indicate the host system call to be called.
30431 This is just the name of the function.
30433 @var{parameter@dots{}} are the parameters to the system call.
30434 Parameters are hexadecimal integer values, either the actual values in case
30435 of scalar datatypes, pointers to target buffer space in case of compound
30436 datatypes and unspecified memory areas, or pointer/length pairs in case
30437 of string parameters. These are appended to the @var{call-id} as a
30438 comma-delimited list. All values are transmitted in ASCII
30439 string representation, pointer/length pairs separated by a slash.
30445 @node The F Reply Packet
30446 @subsection The @code{F} Reply Packet
30447 @cindex file-i/o reply packet
30448 @cindex @code{F} reply packet
30450 The @code{F} reply packet has the following format:
30454 @item F@var{retcode},@var{errno},@var{Ctrl-C flag};@var{call-specific attachment}
30456 @var{retcode} is the return code of the system call as hexadecimal value.
30458 @var{errno} is the @code{errno} set by the call, in protocol-specific
30460 This parameter can be omitted if the call was successful.
30462 @var{Ctrl-C flag} is only sent if the user requested a break. In this
30463 case, @var{errno} must be sent as well, even if the call was successful.
30464 The @var{Ctrl-C flag} itself consists of the character @samp{C}:
30471 or, if the call was interrupted before the host call has been performed:
30478 assuming 4 is the protocol-specific representation of @code{EINTR}.
30483 @node The Ctrl-C Message
30484 @subsection The @samp{Ctrl-C} Message
30485 @cindex ctrl-c message, in file-i/o protocol
30487 If the @samp{Ctrl-C} flag is set in the @value{GDBN}
30488 reply packet (@pxref{The F Reply Packet}),
30489 the target should behave as if it had
30490 gotten a break message. The meaning for the target is ``system call
30491 interrupted by @code{SIGINT}''. Consequentially, the target should actually stop
30492 (as with a break message) and return to @value{GDBN} with a @code{T02}
30495 It's important for the target to know in which
30496 state the system call was interrupted. There are two possible cases:
30500 The system call hasn't been performed on the host yet.
30503 The system call on the host has been finished.
30507 These two states can be distinguished by the target by the value of the
30508 returned @code{errno}. If it's the protocol representation of @code{EINTR}, the system
30509 call hasn't been performed. This is equivalent to the @code{EINTR} handling
30510 on POSIX systems. In any other case, the target may presume that the
30511 system call has been finished --- successfully or not --- and should behave
30512 as if the break message arrived right after the system call.
30514 @value{GDBN} must behave reliably. If the system call has not been called
30515 yet, @value{GDBN} may send the @code{F} reply immediately, setting @code{EINTR} as
30516 @code{errno} in the packet. If the system call on the host has been finished
30517 before the user requests a break, the full action must be finished by
30518 @value{GDBN}. This requires sending @code{M} or @code{X} packets as necessary.
30519 The @code{F} packet may only be sent when either nothing has happened
30520 or the full action has been completed.
30523 @subsection Console I/O
30524 @cindex console i/o as part of file-i/o
30526 By default and if not explicitly closed by the target system, the file
30527 descriptors 0, 1 and 2 are connected to the @value{GDBN} console. Output
30528 on the @value{GDBN} console is handled as any other file output operation
30529 (@code{write(1, @dots{})} or @code{write(2, @dots{})}). Console input is handled
30530 by @value{GDBN} so that after the target read request from file descriptor
30531 0 all following typing is buffered until either one of the following
30536 The user types @kbd{Ctrl-c}. The behaviour is as explained above, and the
30538 system call is treated as finished.
30541 The user presses @key{RET}. This is treated as end of input with a trailing
30545 The user types @kbd{Ctrl-d}. This is treated as end of input. No trailing
30546 character (neither newline nor @samp{Ctrl-D}) is appended to the input.
30550 If the user has typed more characters than fit in the buffer given to
30551 the @code{read} call, the trailing characters are buffered in @value{GDBN} until
30552 either another @code{read(0, @dots{})} is requested by the target, or debugging
30553 is stopped at the user's request.
30556 @node List of Supported Calls
30557 @subsection List of Supported Calls
30558 @cindex list of supported file-i/o calls
30575 @unnumberedsubsubsec open
30576 @cindex open, file-i/o system call
30581 int open(const char *pathname, int flags);
30582 int open(const char *pathname, int flags, mode_t mode);
30586 @samp{Fopen,@var{pathptr}/@var{len},@var{flags},@var{mode}}
30589 @var{flags} is the bitwise @code{OR} of the following values:
30593 If the file does not exist it will be created. The host
30594 rules apply as far as file ownership and time stamps
30598 When used with @code{O_CREAT}, if the file already exists it is
30599 an error and open() fails.
30602 If the file already exists and the open mode allows
30603 writing (@code{O_RDWR} or @code{O_WRONLY} is given) it will be
30604 truncated to zero length.
30607 The file is opened in append mode.
30610 The file is opened for reading only.
30613 The file is opened for writing only.
30616 The file is opened for reading and writing.
30620 Other bits are silently ignored.
30624 @var{mode} is the bitwise @code{OR} of the following values:
30628 User has read permission.
30631 User has write permission.
30634 Group has read permission.
30637 Group has write permission.
30640 Others have read permission.
30643 Others have write permission.
30647 Other bits are silently ignored.
30650 @item Return value:
30651 @code{open} returns the new file descriptor or -1 if an error
30658 @var{pathname} already exists and @code{O_CREAT} and @code{O_EXCL} were used.
30661 @var{pathname} refers to a directory.
30664 The requested access is not allowed.
30667 @var{pathname} was too long.
30670 A directory component in @var{pathname} does not exist.
30673 @var{pathname} refers to a device, pipe, named pipe or socket.
30676 @var{pathname} refers to a file on a read-only filesystem and
30677 write access was requested.
30680 @var{pathname} is an invalid pointer value.
30683 No space on device to create the file.
30686 The process already has the maximum number of files open.
30689 The limit on the total number of files open on the system
30693 The call was interrupted by the user.
30699 @unnumberedsubsubsec close
30700 @cindex close, file-i/o system call
30709 @samp{Fclose,@var{fd}}
30711 @item Return value:
30712 @code{close} returns zero on success, or -1 if an error occurred.
30718 @var{fd} isn't a valid open file descriptor.
30721 The call was interrupted by the user.
30727 @unnumberedsubsubsec read
30728 @cindex read, file-i/o system call
30733 int read(int fd, void *buf, unsigned int count);
30737 @samp{Fread,@var{fd},@var{bufptr},@var{count}}
30739 @item Return value:
30740 On success, the number of bytes read is returned.
30741 Zero indicates end of file. If count is zero, read
30742 returns zero as well. On error, -1 is returned.
30748 @var{fd} is not a valid file descriptor or is not open for
30752 @var{bufptr} is an invalid pointer value.
30755 The call was interrupted by the user.
30761 @unnumberedsubsubsec write
30762 @cindex write, file-i/o system call
30767 int write(int fd, const void *buf, unsigned int count);
30771 @samp{Fwrite,@var{fd},@var{bufptr},@var{count}}
30773 @item Return value:
30774 On success, the number of bytes written are returned.
30775 Zero indicates nothing was written. On error, -1
30782 @var{fd} is not a valid file descriptor or is not open for
30786 @var{bufptr} is an invalid pointer value.
30789 An attempt was made to write a file that exceeds the
30790 host-specific maximum file size allowed.
30793 No space on device to write the data.
30796 The call was interrupted by the user.
30802 @unnumberedsubsubsec lseek
30803 @cindex lseek, file-i/o system call
30808 long lseek (int fd, long offset, int flag);
30812 @samp{Flseek,@var{fd},@var{offset},@var{flag}}
30814 @var{flag} is one of:
30818 The offset is set to @var{offset} bytes.
30821 The offset is set to its current location plus @var{offset}
30825 The offset is set to the size of the file plus @var{offset}
30829 @item Return value:
30830 On success, the resulting unsigned offset in bytes from
30831 the beginning of the file is returned. Otherwise, a
30832 value of -1 is returned.
30838 @var{fd} is not a valid open file descriptor.
30841 @var{fd} is associated with the @value{GDBN} console.
30844 @var{flag} is not a proper value.
30847 The call was interrupted by the user.
30853 @unnumberedsubsubsec rename
30854 @cindex rename, file-i/o system call
30859 int rename(const char *oldpath, const char *newpath);
30863 @samp{Frename,@var{oldpathptr}/@var{len},@var{newpathptr}/@var{len}}
30865 @item Return value:
30866 On success, zero is returned. On error, -1 is returned.
30872 @var{newpath} is an existing directory, but @var{oldpath} is not a
30876 @var{newpath} is a non-empty directory.
30879 @var{oldpath} or @var{newpath} is a directory that is in use by some
30883 An attempt was made to make a directory a subdirectory
30887 A component used as a directory in @var{oldpath} or new
30888 path is not a directory. Or @var{oldpath} is a directory
30889 and @var{newpath} exists but is not a directory.
30892 @var{oldpathptr} or @var{newpathptr} are invalid pointer values.
30895 No access to the file or the path of the file.
30899 @var{oldpath} or @var{newpath} was too long.
30902 A directory component in @var{oldpath} or @var{newpath} does not exist.
30905 The file is on a read-only filesystem.
30908 The device containing the file has no room for the new
30912 The call was interrupted by the user.
30918 @unnumberedsubsubsec unlink
30919 @cindex unlink, file-i/o system call
30924 int unlink(const char *pathname);
30928 @samp{Funlink,@var{pathnameptr}/@var{len}}
30930 @item Return value:
30931 On success, zero is returned. On error, -1 is returned.
30937 No access to the file or the path of the file.
30940 The system does not allow unlinking of directories.
30943 The file @var{pathname} cannot be unlinked because it's
30944 being used by another process.
30947 @var{pathnameptr} is an invalid pointer value.
30950 @var{pathname} was too long.
30953 A directory component in @var{pathname} does not exist.
30956 A component of the path is not a directory.
30959 The file is on a read-only filesystem.
30962 The call was interrupted by the user.
30968 @unnumberedsubsubsec stat/fstat
30969 @cindex fstat, file-i/o system call
30970 @cindex stat, file-i/o system call
30975 int stat(const char *pathname, struct stat *buf);
30976 int fstat(int fd, struct stat *buf);
30980 @samp{Fstat,@var{pathnameptr}/@var{len},@var{bufptr}}@*
30981 @samp{Ffstat,@var{fd},@var{bufptr}}
30983 @item Return value:
30984 On success, zero is returned. On error, -1 is returned.
30990 @var{fd} is not a valid open file.
30993 A directory component in @var{pathname} does not exist or the
30994 path is an empty string.
30997 A component of the path is not a directory.
31000 @var{pathnameptr} is an invalid pointer value.
31003 No access to the file or the path of the file.
31006 @var{pathname} was too long.
31009 The call was interrupted by the user.
31015 @unnumberedsubsubsec gettimeofday
31016 @cindex gettimeofday, file-i/o system call
31021 int gettimeofday(struct timeval *tv, void *tz);
31025 @samp{Fgettimeofday,@var{tvptr},@var{tzptr}}
31027 @item Return value:
31028 On success, 0 is returned, -1 otherwise.
31034 @var{tz} is a non-NULL pointer.
31037 @var{tvptr} and/or @var{tzptr} is an invalid pointer value.
31043 @unnumberedsubsubsec isatty
31044 @cindex isatty, file-i/o system call
31049 int isatty(int fd);
31053 @samp{Fisatty,@var{fd}}
31055 @item Return value:
31056 Returns 1 if @var{fd} refers to the @value{GDBN} console, 0 otherwise.
31062 The call was interrupted by the user.
31067 Note that the @code{isatty} call is treated as a special case: it returns
31068 1 to the target if the file descriptor is attached
31069 to the @value{GDBN} console, 0 otherwise. Implementing through system calls
31070 would require implementing @code{ioctl} and would be more complex than
31075 @unnumberedsubsubsec system
31076 @cindex system, file-i/o system call
31081 int system(const char *command);
31085 @samp{Fsystem,@var{commandptr}/@var{len}}
31087 @item Return value:
31088 If @var{len} is zero, the return value indicates whether a shell is
31089 available. A zero return value indicates a shell is not available.
31090 For non-zero @var{len}, the value returned is -1 on error and the
31091 return status of the command otherwise. Only the exit status of the
31092 command is returned, which is extracted from the host's @code{system}
31093 return value by calling @code{WEXITSTATUS(retval)}. In case
31094 @file{/bin/sh} could not be executed, 127 is returned.
31100 The call was interrupted by the user.
31105 @value{GDBN} takes over the full task of calling the necessary host calls
31106 to perform the @code{system} call. The return value of @code{system} on
31107 the host is simplified before it's returned
31108 to the target. Any termination signal information from the child process
31109 is discarded, and the return value consists
31110 entirely of the exit status of the called command.
31112 Due to security concerns, the @code{system} call is by default refused
31113 by @value{GDBN}. The user has to allow this call explicitly with the
31114 @code{set remote system-call-allowed 1} command.
31117 @item set remote system-call-allowed
31118 @kindex set remote system-call-allowed
31119 Control whether to allow the @code{system} calls in the File I/O
31120 protocol for the remote target. The default is zero (disabled).
31122 @item show remote system-call-allowed
31123 @kindex show remote system-call-allowed
31124 Show whether the @code{system} calls are allowed in the File I/O
31128 @node Protocol-specific Representation of Datatypes
31129 @subsection Protocol-specific Representation of Datatypes
31130 @cindex protocol-specific representation of datatypes, in file-i/o protocol
31133 * Integral Datatypes::
31135 * Memory Transfer::
31140 @node Integral Datatypes
31141 @unnumberedsubsubsec Integral Datatypes
31142 @cindex integral datatypes, in file-i/o protocol
31144 The integral datatypes used in the system calls are @code{int},
31145 @code{unsigned int}, @code{long}, @code{unsigned long},
31146 @code{mode_t}, and @code{time_t}.
31148 @code{int}, @code{unsigned int}, @code{mode_t} and @code{time_t} are
31149 implemented as 32 bit values in this protocol.
31151 @code{long} and @code{unsigned long} are implemented as 64 bit types.
31153 @xref{Limits}, for corresponding MIN and MAX values (similar to those
31154 in @file{limits.h}) to allow range checking on host and target.
31156 @code{time_t} datatypes are defined as seconds since the Epoch.
31158 All integral datatypes transferred as part of a memory read or write of a
31159 structured datatype e.g.@: a @code{struct stat} have to be given in big endian
31162 @node Pointer Values
31163 @unnumberedsubsubsec Pointer Values
31164 @cindex pointer values, in file-i/o protocol
31166 Pointers to target data are transmitted as they are. An exception
31167 is made for pointers to buffers for which the length isn't
31168 transmitted as part of the function call, namely strings. Strings
31169 are transmitted as a pointer/length pair, both as hex values, e.g.@:
31176 which is a pointer to data of length 18 bytes at position 0x1aaf.
31177 The length is defined as the full string length in bytes, including
31178 the trailing null byte. For example, the string @code{"hello world"}
31179 at address 0x123456 is transmitted as
31185 @node Memory Transfer
31186 @unnumberedsubsubsec Memory Transfer
31187 @cindex memory transfer, in file-i/o protocol
31189 Structured data which is transferred using a memory read or write (for
31190 example, a @code{struct stat}) is expected to be in a protocol-specific format
31191 with all scalar multibyte datatypes being big endian. Translation to
31192 this representation needs to be done both by the target before the @code{F}
31193 packet is sent, and by @value{GDBN} before
31194 it transfers memory to the target. Transferred pointers to structured
31195 data should point to the already-coerced data at any time.
31199 @unnumberedsubsubsec struct stat
31200 @cindex struct stat, in file-i/o protocol
31202 The buffer of type @code{struct stat} used by the target and @value{GDBN}
31203 is defined as follows:
31207 unsigned int st_dev; /* device */
31208 unsigned int st_ino; /* inode */
31209 mode_t st_mode; /* protection */
31210 unsigned int st_nlink; /* number of hard links */
31211 unsigned int st_uid; /* user ID of owner */
31212 unsigned int st_gid; /* group ID of owner */
31213 unsigned int st_rdev; /* device type (if inode device) */
31214 unsigned long st_size; /* total size, in bytes */
31215 unsigned long st_blksize; /* blocksize for filesystem I/O */
31216 unsigned long st_blocks; /* number of blocks allocated */
31217 time_t st_atime; /* time of last access */
31218 time_t st_mtime; /* time of last modification */
31219 time_t st_ctime; /* time of last change */
31223 The integral datatypes conform to the definitions given in the
31224 appropriate section (see @ref{Integral Datatypes}, for details) so this
31225 structure is of size 64 bytes.
31227 The values of several fields have a restricted meaning and/or
31233 A value of 0 represents a file, 1 the console.
31236 No valid meaning for the target. Transmitted unchanged.
31239 Valid mode bits are described in @ref{Constants}. Any other
31240 bits have currently no meaning for the target.
31245 No valid meaning for the target. Transmitted unchanged.
31250 These values have a host and file system dependent
31251 accuracy. Especially on Windows hosts, the file system may not
31252 support exact timing values.
31255 The target gets a @code{struct stat} of the above representation and is
31256 responsible for coercing it to the target representation before
31259 Note that due to size differences between the host, target, and protocol
31260 representations of @code{struct stat} members, these members could eventually
31261 get truncated on the target.
31263 @node struct timeval
31264 @unnumberedsubsubsec struct timeval
31265 @cindex struct timeval, in file-i/o protocol
31267 The buffer of type @code{struct timeval} used by the File-I/O protocol
31268 is defined as follows:
31272 time_t tv_sec; /* second */
31273 long tv_usec; /* microsecond */
31277 The integral datatypes conform to the definitions given in the
31278 appropriate section (see @ref{Integral Datatypes}, for details) so this
31279 structure is of size 8 bytes.
31282 @subsection Constants
31283 @cindex constants, in file-i/o protocol
31285 The following values are used for the constants inside of the
31286 protocol. @value{GDBN} and target are responsible for translating these
31287 values before and after the call as needed.
31298 @unnumberedsubsubsec Open Flags
31299 @cindex open flags, in file-i/o protocol
31301 All values are given in hexadecimal representation.
31313 @node mode_t Values
31314 @unnumberedsubsubsec mode_t Values
31315 @cindex mode_t values, in file-i/o protocol
31317 All values are given in octal representation.
31334 @unnumberedsubsubsec Errno Values
31335 @cindex errno values, in file-i/o protocol
31337 All values are given in decimal representation.
31362 @code{EUNKNOWN} is used as a fallback error value if a host system returns
31363 any error value not in the list of supported error numbers.
31366 @unnumberedsubsubsec Lseek Flags
31367 @cindex lseek flags, in file-i/o protocol
31376 @unnumberedsubsubsec Limits
31377 @cindex limits, in file-i/o protocol
31379 All values are given in decimal representation.
31382 INT_MIN -2147483648
31384 UINT_MAX 4294967295
31385 LONG_MIN -9223372036854775808
31386 LONG_MAX 9223372036854775807
31387 ULONG_MAX 18446744073709551615
31390 @node File-I/O Examples
31391 @subsection File-I/O Examples
31392 @cindex file-i/o examples
31394 Example sequence of a write call, file descriptor 3, buffer is at target
31395 address 0x1234, 6 bytes should be written:
31398 <- @code{Fwrite,3,1234,6}
31399 @emph{request memory read from target}
31402 @emph{return "6 bytes written"}
31406 Example sequence of a read call, file descriptor 3, buffer is at target
31407 address 0x1234, 6 bytes should be read:
31410 <- @code{Fread,3,1234,6}
31411 @emph{request memory write to target}
31412 -> @code{X1234,6:XXXXXX}
31413 @emph{return "6 bytes read"}
31417 Example sequence of a read call, call fails on the host due to invalid
31418 file descriptor (@code{EBADF}):
31421 <- @code{Fread,3,1234,6}
31425 Example sequence of a read call, user presses @kbd{Ctrl-c} before syscall on
31429 <- @code{Fread,3,1234,6}
31434 Example sequence of a read call, user presses @kbd{Ctrl-c} after syscall on
31438 <- @code{Fread,3,1234,6}
31439 -> @code{X1234,6:XXXXXX}
31443 @node Library List Format
31444 @section Library List Format
31445 @cindex library list format, remote protocol
31447 On some platforms, a dynamic loader (e.g.@: @file{ld.so}) runs in the
31448 same process as your application to manage libraries. In this case,
31449 @value{GDBN} can use the loader's symbol table and normal memory
31450 operations to maintain a list of shared libraries. On other
31451 platforms, the operating system manages loaded libraries.
31452 @value{GDBN} can not retrieve the list of currently loaded libraries
31453 through memory operations, so it uses the @samp{qXfer:libraries:read}
31454 packet (@pxref{qXfer library list read}) instead. The remote stub
31455 queries the target's operating system and reports which libraries
31458 The @samp{qXfer:libraries:read} packet returns an XML document which
31459 lists loaded libraries and their offsets. Each library has an
31460 associated name and one or more segment or section base addresses,
31461 which report where the library was loaded in memory.
31463 For the common case of libraries that are fully linked binaries, the
31464 library should have a list of segments. If the target supports
31465 dynamic linking of a relocatable object file, its library XML element
31466 should instead include a list of allocated sections. The segment or
31467 section bases are start addresses, not relocation offsets; they do not
31468 depend on the library's link-time base addresses.
31470 @value{GDBN} must be linked with the Expat library to support XML
31471 library lists. @xref{Expat}.
31473 A simple memory map, with one loaded library relocated by a single
31474 offset, looks like this:
31478 <library name="/lib/libc.so.6">
31479 <segment address="0x10000000"/>
31484 Another simple memory map, with one loaded library with three
31485 allocated sections (.text, .data, .bss), looks like this:
31489 <library name="sharedlib.o">
31490 <section address="0x10000000"/>
31491 <section address="0x20000000"/>
31492 <section address="0x30000000"/>
31497 The format of a library list is described by this DTD:
31500 <!-- library-list: Root element with versioning -->
31501 <!ELEMENT library-list (library)*>
31502 <!ATTLIST library-list version CDATA #FIXED "1.0">
31503 <!ELEMENT library (segment*, section*)>
31504 <!ATTLIST library name CDATA #REQUIRED>
31505 <!ELEMENT segment EMPTY>
31506 <!ATTLIST segment address CDATA #REQUIRED>
31507 <!ELEMENT section EMPTY>
31508 <!ATTLIST section address CDATA #REQUIRED>
31511 In addition, segments and section descriptors cannot be mixed within a
31512 single library element, and you must supply at least one segment or
31513 section for each library.
31515 @node Memory Map Format
31516 @section Memory Map Format
31517 @cindex memory map format
31519 To be able to write into flash memory, @value{GDBN} needs to obtain a
31520 memory map from the target. This section describes the format of the
31523 The memory map is obtained using the @samp{qXfer:memory-map:read}
31524 (@pxref{qXfer memory map read}) packet and is an XML document that
31525 lists memory regions.
31527 @value{GDBN} must be linked with the Expat library to support XML
31528 memory maps. @xref{Expat}.
31530 The top-level structure of the document is shown below:
31533 <?xml version="1.0"?>
31534 <!DOCTYPE memory-map
31535 PUBLIC "+//IDN gnu.org//DTD GDB Memory Map V1.0//EN"
31536 "http://sourceware.org/gdb/gdb-memory-map.dtd">
31542 Each region can be either:
31547 A region of RAM starting at @var{addr} and extending for @var{length}
31551 <memory type="ram" start="@var{addr}" length="@var{length}"/>
31556 A region of read-only memory:
31559 <memory type="rom" start="@var{addr}" length="@var{length}"/>
31564 A region of flash memory, with erasure blocks @var{blocksize}
31568 <memory type="flash" start="@var{addr}" length="@var{length}">
31569 <property name="blocksize">@var{blocksize}</property>
31575 Regions must not overlap. @value{GDBN} assumes that areas of memory not covered
31576 by the memory map are RAM, and uses the ordinary @samp{M} and @samp{X}
31577 packets to write to addresses in such ranges.
31579 The formal DTD for memory map format is given below:
31582 <!-- ................................................... -->
31583 <!-- Memory Map XML DTD ................................ -->
31584 <!-- File: memory-map.dtd .............................. -->
31585 <!-- .................................... .............. -->
31586 <!-- memory-map.dtd -->
31587 <!-- memory-map: Root element with versioning -->
31588 <!ELEMENT memory-map (memory | property)>
31589 <!ATTLIST memory-map version CDATA #FIXED "1.0.0">
31590 <!ELEMENT memory (property)>
31591 <!-- memory: Specifies a memory region,
31592 and its type, or device. -->
31593 <!ATTLIST memory type CDATA #REQUIRED
31594 start CDATA #REQUIRED
31595 length CDATA #REQUIRED
31596 device CDATA #IMPLIED>
31597 <!-- property: Generic attribute tag -->
31598 <!ELEMENT property (#PCDATA | property)*>
31599 <!ATTLIST property name CDATA #REQUIRED>
31602 @include agentexpr.texi
31604 @node Target Descriptions
31605 @appendix Target Descriptions
31606 @cindex target descriptions
31608 @strong{Warning:} target descriptions are still under active development,
31609 and the contents and format may change between @value{GDBN} releases.
31610 The format is expected to stabilize in the future.
31612 One of the challenges of using @value{GDBN} to debug embedded systems
31613 is that there are so many minor variants of each processor
31614 architecture in use. It is common practice for vendors to start with
31615 a standard processor core --- ARM, PowerPC, or MIPS, for example ---
31616 and then make changes to adapt it to a particular market niche. Some
31617 architectures have hundreds of variants, available from dozens of
31618 vendors. This leads to a number of problems:
31622 With so many different customized processors, it is difficult for
31623 the @value{GDBN} maintainers to keep up with the changes.
31625 Since individual variants may have short lifetimes or limited
31626 audiences, it may not be worthwhile to carry information about every
31627 variant in the @value{GDBN} source tree.
31629 When @value{GDBN} does support the architecture of the embedded system
31630 at hand, the task of finding the correct architecture name to give the
31631 @command{set architecture} command can be error-prone.
31634 To address these problems, the @value{GDBN} remote protocol allows a
31635 target system to not only identify itself to @value{GDBN}, but to
31636 actually describe its own features. This lets @value{GDBN} support
31637 processor variants it has never seen before --- to the extent that the
31638 descriptions are accurate, and that @value{GDBN} understands them.
31640 @value{GDBN} must be linked with the Expat library to support XML
31641 target descriptions. @xref{Expat}.
31644 * Retrieving Descriptions:: How descriptions are fetched from a target.
31645 * Target Description Format:: The contents of a target description.
31646 * Predefined Target Types:: Standard types available for target
31648 * Standard Target Features:: Features @value{GDBN} knows about.
31651 @node Retrieving Descriptions
31652 @section Retrieving Descriptions
31654 Target descriptions can be read from the target automatically, or
31655 specified by the user manually. The default behavior is to read the
31656 description from the target. @value{GDBN} retrieves it via the remote
31657 protocol using @samp{qXfer} requests (@pxref{General Query Packets,
31658 qXfer}). The @var{annex} in the @samp{qXfer} packet will be
31659 @samp{target.xml}. The contents of the @samp{target.xml} annex are an
31660 XML document, of the form described in @ref{Target Description
31663 Alternatively, you can specify a file to read for the target description.
31664 If a file is set, the target will not be queried. The commands to
31665 specify a file are:
31668 @cindex set tdesc filename
31669 @item set tdesc filename @var{path}
31670 Read the target description from @var{path}.
31672 @cindex unset tdesc filename
31673 @item unset tdesc filename
31674 Do not read the XML target description from a file. @value{GDBN}
31675 will use the description supplied by the current target.
31677 @cindex show tdesc filename
31678 @item show tdesc filename
31679 Show the filename to read for a target description, if any.
31683 @node Target Description Format
31684 @section Target Description Format
31685 @cindex target descriptions, XML format
31687 A target description annex is an @uref{http://www.w3.org/XML/, XML}
31688 document which complies with the Document Type Definition provided in
31689 the @value{GDBN} sources in @file{gdb/features/gdb-target.dtd}. This
31690 means you can use generally available tools like @command{xmllint} to
31691 check that your feature descriptions are well-formed and valid.
31692 However, to help people unfamiliar with XML write descriptions for
31693 their targets, we also describe the grammar here.
31695 Target descriptions can identify the architecture of the remote target
31696 and (for some architectures) provide information about custom register
31697 sets. They can also identify the OS ABI of the remote target.
31698 @value{GDBN} can use this information to autoconfigure for your
31699 target, or to warn you if you connect to an unsupported target.
31701 Here is a simple target description:
31704 <target version="1.0">
31705 <architecture>i386:x86-64</architecture>
31710 This minimal description only says that the target uses
31711 the x86-64 architecture.
31713 A target description has the following overall form, with [ ] marking
31714 optional elements and @dots{} marking repeatable elements. The elements
31715 are explained further below.
31718 <?xml version="1.0"?>
31719 <!DOCTYPE target SYSTEM "gdb-target.dtd">
31720 <target version="1.0">
31721 @r{[}@var{architecture}@r{]}
31722 @r{[}@var{osabi}@r{]}
31723 @r{[}@var{compatible}@r{]}
31724 @r{[}@var{feature}@dots{}@r{]}
31729 The description is generally insensitive to whitespace and line
31730 breaks, under the usual common-sense rules. The XML version
31731 declaration and document type declaration can generally be omitted
31732 (@value{GDBN} does not require them), but specifying them may be
31733 useful for XML validation tools. The @samp{version} attribute for
31734 @samp{<target>} may also be omitted, but we recommend
31735 including it; if future versions of @value{GDBN} use an incompatible
31736 revision of @file{gdb-target.dtd}, they will detect and report
31737 the version mismatch.
31739 @subsection Inclusion
31740 @cindex target descriptions, inclusion
31743 @cindex <xi:include>
31746 It can sometimes be valuable to split a target description up into
31747 several different annexes, either for organizational purposes, or to
31748 share files between different possible target descriptions. You can
31749 divide a description into multiple files by replacing any element of
31750 the target description with an inclusion directive of the form:
31753 <xi:include href="@var{document}"/>
31757 When @value{GDBN} encounters an element of this form, it will retrieve
31758 the named XML @var{document}, and replace the inclusion directive with
31759 the contents of that document. If the current description was read
31760 using @samp{qXfer}, then so will be the included document;
31761 @var{document} will be interpreted as the name of an annex. If the
31762 current description was read from a file, @value{GDBN} will look for
31763 @var{document} as a file in the same directory where it found the
31764 original description.
31766 @subsection Architecture
31767 @cindex <architecture>
31769 An @samp{<architecture>} element has this form:
31772 <architecture>@var{arch}</architecture>
31775 @var{arch} is one of the architectures from the set accepted by
31776 @code{set architecture} (@pxref{Targets, ,Specifying a Debugging Target}).
31779 @cindex @code{<osabi>}
31781 This optional field was introduced in @value{GDBN} version 7.0.
31782 Previous versions of @value{GDBN} ignore it.
31784 An @samp{<osabi>} element has this form:
31787 <osabi>@var{abi-name}</osabi>
31790 @var{abi-name} is an OS ABI name from the same selection accepted by
31791 @w{@code{set osabi}} (@pxref{ABI, ,Configuring the Current ABI}).
31793 @subsection Compatible Architecture
31794 @cindex @code{<compatible>}
31796 This optional field was introduced in @value{GDBN} version 7.0.
31797 Previous versions of @value{GDBN} ignore it.
31799 A @samp{<compatible>} element has this form:
31802 <compatible>@var{arch}</compatible>
31805 @var{arch} is one of the architectures from the set accepted by
31806 @code{set architecture} (@pxref{Targets, ,Specifying a Debugging Target}).
31808 A @samp{<compatible>} element is used to specify that the target
31809 is able to run binaries in some other than the main target architecture
31810 given by the @samp{<architecture>} element. For example, on the
31811 Cell Broadband Engine, the main architecture is @code{powerpc:common}
31812 or @code{powerpc:common64}, but the system is able to run binaries
31813 in the @code{spu} architecture as well. The way to describe this
31814 capability with @samp{<compatible>} is as follows:
31817 <architecture>powerpc:common</architecture>
31818 <compatible>spu</compatible>
31821 @subsection Features
31824 Each @samp{<feature>} describes some logical portion of the target
31825 system. Features are currently used to describe available CPU
31826 registers and the types of their contents. A @samp{<feature>} element
31830 <feature name="@var{name}">
31831 @r{[}@var{type}@dots{}@r{]}
31837 Each feature's name should be unique within the description. The name
31838 of a feature does not matter unless @value{GDBN} has some special
31839 knowledge of the contents of that feature; if it does, the feature
31840 should have its standard name. @xref{Standard Target Features}.
31844 Any register's value is a collection of bits which @value{GDBN} must
31845 interpret. The default interpretation is a two's complement integer,
31846 but other types can be requested by name in the register description.
31847 Some predefined types are provided by @value{GDBN} (@pxref{Predefined
31848 Target Types}), and the description can define additional composite types.
31850 Each type element must have an @samp{id} attribute, which gives
31851 a unique (within the containing @samp{<feature>}) name to the type.
31852 Types must be defined before they are used.
31855 Some targets offer vector registers, which can be treated as arrays
31856 of scalar elements. These types are written as @samp{<vector>} elements,
31857 specifying the array element type, @var{type}, and the number of elements,
31861 <vector id="@var{id}" type="@var{type}" count="@var{count}"/>
31865 If a register's value is usefully viewed in multiple ways, define it
31866 with a union type containing the useful representations. The
31867 @samp{<union>} element contains one or more @samp{<field>} elements,
31868 each of which has a @var{name} and a @var{type}:
31871 <union id="@var{id}">
31872 <field name="@var{name}" type="@var{type}"/>
31877 @subsection Registers
31880 Each register is represented as an element with this form:
31883 <reg name="@var{name}"
31884 bitsize="@var{size}"
31885 @r{[}regnum="@var{num}"@r{]}
31886 @r{[}save-restore="@var{save-restore}"@r{]}
31887 @r{[}type="@var{type}"@r{]}
31888 @r{[}group="@var{group}"@r{]}/>
31892 The components are as follows:
31897 The register's name; it must be unique within the target description.
31900 The register's size, in bits.
31903 The register's number. If omitted, a register's number is one greater
31904 than that of the previous register (either in the current feature or in
31905 a preceeding feature); the first register in the target description
31906 defaults to zero. This register number is used to read or write
31907 the register; e.g.@: it is used in the remote @code{p} and @code{P}
31908 packets, and registers appear in the @code{g} and @code{G} packets
31909 in order of increasing register number.
31912 Whether the register should be preserved across inferior function
31913 calls; this must be either @code{yes} or @code{no}. The default is
31914 @code{yes}, which is appropriate for most registers except for
31915 some system control registers; this is not related to the target's
31919 The type of the register. @var{type} may be a predefined type, a type
31920 defined in the current feature, or one of the special types @code{int}
31921 and @code{float}. @code{int} is an integer type of the correct size
31922 for @var{bitsize}, and @code{float} is a floating point type (in the
31923 architecture's normal floating point format) of the correct size for
31924 @var{bitsize}. The default is @code{int}.
31927 The register group to which this register belongs. @var{group} must
31928 be either @code{general}, @code{float}, or @code{vector}. If no
31929 @var{group} is specified, @value{GDBN} will not display the register
31930 in @code{info registers}.
31934 @node Predefined Target Types
31935 @section Predefined Target Types
31936 @cindex target descriptions, predefined types
31938 Type definitions in the self-description can build up composite types
31939 from basic building blocks, but can not define fundamental types. Instead,
31940 standard identifiers are provided by @value{GDBN} for the fundamental
31941 types. The currently supported types are:
31950 Signed integer types holding the specified number of bits.
31957 Unsigned integer types holding the specified number of bits.
31961 Pointers to unspecified code and data. The program counter and
31962 any dedicated return address register may be marked as code
31963 pointers; printing a code pointer converts it into a symbolic
31964 address. The stack pointer and any dedicated address registers
31965 may be marked as data pointers.
31968 Single precision IEEE floating point.
31971 Double precision IEEE floating point.
31974 The 12-byte extended precision format used by ARM FPA registers.
31978 @node Standard Target Features
31979 @section Standard Target Features
31980 @cindex target descriptions, standard features
31982 A target description must contain either no registers or all the
31983 target's registers. If the description contains no registers, then
31984 @value{GDBN} will assume a default register layout, selected based on
31985 the architecture. If the description contains any registers, the
31986 default layout will not be used; the standard registers must be
31987 described in the target description, in such a way that @value{GDBN}
31988 can recognize them.
31990 This is accomplished by giving specific names to feature elements
31991 which contain standard registers. @value{GDBN} will look for features
31992 with those names and verify that they contain the expected registers;
31993 if any known feature is missing required registers, or if any required
31994 feature is missing, @value{GDBN} will reject the target
31995 description. You can add additional registers to any of the
31996 standard features --- @value{GDBN} will display them just as if
31997 they were added to an unrecognized feature.
31999 This section lists the known features and their expected contents.
32000 Sample XML documents for these features are included in the
32001 @value{GDBN} source tree, in the directory @file{gdb/features}.
32003 Names recognized by @value{GDBN} should include the name of the
32004 company or organization which selected the name, and the overall
32005 architecture to which the feature applies; so e.g.@: the feature
32006 containing ARM core registers is named @samp{org.gnu.gdb.arm.core}.
32008 The names of registers are not case sensitive for the purpose
32009 of recognizing standard features, but @value{GDBN} will only display
32010 registers using the capitalization used in the description.
32016 * PowerPC Features::
32021 @subsection ARM Features
32022 @cindex target descriptions, ARM features
32024 The @samp{org.gnu.gdb.arm.core} feature is required for ARM targets.
32025 It should contain registers @samp{r0} through @samp{r13}, @samp{sp},
32026 @samp{lr}, @samp{pc}, and @samp{cpsr}.
32028 The @samp{org.gnu.gdb.arm.fpa} feature is optional. If present, it
32029 should contain registers @samp{f0} through @samp{f7} and @samp{fps}.
32031 The @samp{org.gnu.gdb.xscale.iwmmxt} feature is optional. If present,
32032 it should contain at least registers @samp{wR0} through @samp{wR15} and
32033 @samp{wCGR0} through @samp{wCGR3}. The @samp{wCID}, @samp{wCon},
32034 @samp{wCSSF}, and @samp{wCASF} registers are optional.
32036 The @samp{org.gnu.gdb.arm.vfp} feature is optional. If present, it
32037 should contain at least registers @samp{d0} through @samp{d15}. If
32038 they are present, @samp{d16} through @samp{d31} should also be included.
32039 @value{GDBN} will synthesize the single-precision registers from
32040 halves of the double-precision registers.
32042 The @samp{org.gnu.gdb.arm.neon} feature is optional. It does not
32043 need to contain registers; it instructs @value{GDBN} to display the
32044 VFP double-precision registers as vectors and to synthesize the
32045 quad-precision registers from pairs of double-precision registers.
32046 If this feature is present, @samp{org.gnu.gdb.arm.vfp} must also
32047 be present and include 32 double-precision registers.
32049 @node MIPS Features
32050 @subsection MIPS Features
32051 @cindex target descriptions, MIPS features
32053 The @samp{org.gnu.gdb.mips.cpu} feature is required for MIPS targets.
32054 It should contain registers @samp{r0} through @samp{r31}, @samp{lo},
32055 @samp{hi}, and @samp{pc}. They may be 32-bit or 64-bit depending
32058 The @samp{org.gnu.gdb.mips.cp0} feature is also required. It should
32059 contain at least the @samp{status}, @samp{badvaddr}, and @samp{cause}
32060 registers. They may be 32-bit or 64-bit depending on the target.
32062 The @samp{org.gnu.gdb.mips.fpu} feature is currently required, though
32063 it may be optional in a future version of @value{GDBN}. It should
32064 contain registers @samp{f0} through @samp{f31}, @samp{fcsr}, and
32065 @samp{fir}. They may be 32-bit or 64-bit depending on the target.
32067 The @samp{org.gnu.gdb.mips.linux} feature is optional. It should
32068 contain a single register, @samp{restart}, which is used by the
32069 Linux kernel to control restartable syscalls.
32071 @node M68K Features
32072 @subsection M68K Features
32073 @cindex target descriptions, M68K features
32076 @item @samp{org.gnu.gdb.m68k.core}
32077 @itemx @samp{org.gnu.gdb.coldfire.core}
32078 @itemx @samp{org.gnu.gdb.fido.core}
32079 One of those features must be always present.
32080 The feature that is present determines which flavor of m68k is
32081 used. The feature that is present should contain registers
32082 @samp{d0} through @samp{d7}, @samp{a0} through @samp{a5}, @samp{fp},
32083 @samp{sp}, @samp{ps} and @samp{pc}.
32085 @item @samp{org.gnu.gdb.coldfire.fp}
32086 This feature is optional. If present, it should contain registers
32087 @samp{fp0} through @samp{fp7}, @samp{fpcontrol}, @samp{fpstatus} and
32091 @node PowerPC Features
32092 @subsection PowerPC Features
32093 @cindex target descriptions, PowerPC features
32095 The @samp{org.gnu.gdb.power.core} feature is required for PowerPC
32096 targets. It should contain registers @samp{r0} through @samp{r31},
32097 @samp{pc}, @samp{msr}, @samp{cr}, @samp{lr}, @samp{ctr}, and
32098 @samp{xer}. They may be 32-bit or 64-bit depending on the target.
32100 The @samp{org.gnu.gdb.power.fpu} feature is optional. It should
32101 contain registers @samp{f0} through @samp{f31} and @samp{fpscr}.
32103 The @samp{org.gnu.gdb.power.altivec} feature is optional. It should
32104 contain registers @samp{vr0} through @samp{vr31}, @samp{vscr},
32107 The @samp{org.gnu.gdb.power.vsx} feature is optional. It should
32108 contain registers @samp{vs0h} through @samp{vs31h}. @value{GDBN}
32109 will combine these registers with the floating point registers
32110 (@samp{f0} through @samp{f31}) and the altivec registers (@samp{vr0}
32111 through @samp{vr31}) to present the 128-bit wide registers @samp{vs0}
32112 through @samp{vs63}, the set of vector registers for POWER7.
32114 The @samp{org.gnu.gdb.power.spe} feature is optional. It should
32115 contain registers @samp{ev0h} through @samp{ev31h}, @samp{acc}, and
32116 @samp{spefscr}. SPE targets should provide 32-bit registers in
32117 @samp{org.gnu.gdb.power.core} and provide the upper halves in
32118 @samp{ev0h} through @samp{ev31h}. @value{GDBN} will combine
32119 these to present registers @samp{ev0} through @samp{ev31} to the
32122 @node Operating System Information
32123 @appendix Operating System Information
32124 @cindex operating system information
32130 Users of @value{GDBN} often wish to obtain information about the state of
32131 the operating system running on the target---for example the list of
32132 processes, or the list of open files. This section describes the
32133 mechanism that makes it possible. This mechanism is similar to the
32134 target features mechanism (@pxref{Target Descriptions}), but focuses
32135 on a different aspect of target.
32137 Operating system information is retrived from the target via the
32138 remote protocol, using @samp{qXfer} requests (@pxref{qXfer osdata
32139 read}). The object name in the request should be @samp{osdata}, and
32140 the @var{annex} identifies the data to be fetched.
32143 @appendixsection Process list
32144 @cindex operating system information, process list
32146 When requesting the process list, the @var{annex} field in the
32147 @samp{qXfer} request should be @samp{processes}. The returned data is
32148 an XML document. The formal syntax of this document is defined in
32149 @file{gdb/features/osdata.dtd}.
32151 An example document is:
32154 <?xml version="1.0"?>
32155 <!DOCTYPE target SYSTEM "osdata.dtd">
32156 <osdata type="processes">
32158 <column name="pid">1</column>
32159 <column name="user">root</column>
32160 <column name="command">/sbin/init</column>
32165 Each item should include a column whose name is @samp{pid}. The value
32166 of that column should identify the process on the target. The
32167 @samp{user} and @samp{command} columns are optional, and will be
32168 displayed by @value{GDBN}. Target may provide additional columns,
32169 which @value{GDBN} currently ignores.
32183 % I think something like @colophon should be in texinfo. In the
32185 \long\def\colophon{\hbox to0pt{}\vfill
32186 \centerline{The body of this manual is set in}
32187 \centerline{\fontname\tenrm,}
32188 \centerline{with headings in {\bf\fontname\tenbf}}
32189 \centerline{and examples in {\tt\fontname\tentt}.}
32190 \centerline{{\it\fontname\tenit\/},}
32191 \centerline{{\bf\fontname\tenbf}, and}
32192 \centerline{{\sl\fontname\tensl\/}}
32193 \centerline{are used for emphasis.}\vfill}
32195 % Blame: doc@cygnus.com, 1991.