1 \input texinfo @c -*-texinfo-*-
2 @c Copyright (C) 1988, 1989, 1990, 1991, 1992, 1993, 1994, 1995, 1996,
3 @c 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009,
4 @c 2010, 2011 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, 2010
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 1.3 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 @*
107 @node Top, Summary, (dir), (dir)
109 @top Debugging with @value{GDBN}
111 This file describes @value{GDBN}, the @sc{gnu} symbolic debugger.
113 This is the @value{EDITION} Edition, for @value{GDBN}
114 @ifset VERSION_PACKAGE
115 @value{VERSION_PACKAGE}
117 Version @value{GDBVN}.
119 Copyright (C) 1988-2010 Free Software Foundation, Inc.
121 This edition of the GDB manual is dedicated to the memory of Fred
122 Fish. Fred was a long-standing contributor to GDB and to Free
123 software in general. We will miss him.
126 * Summary:: Summary of @value{GDBN}
127 * Sample Session:: A sample @value{GDBN} session
129 * Invocation:: Getting in and out of @value{GDBN}
130 * Commands:: @value{GDBN} commands
131 * Running:: Running programs under @value{GDBN}
132 * Stopping:: Stopping and continuing
133 * Reverse Execution:: Running programs backward
134 * Process Record and Replay:: Recording inferior's execution and replaying it
135 * Stack:: Examining the stack
136 * Source:: Examining source files
137 * Data:: Examining data
138 * Optimized Code:: Debugging optimized code
139 * Macros:: Preprocessor Macros
140 * Tracepoints:: Debugging remote targets non-intrusively
141 * Overlays:: Debugging programs that use overlays
143 * Languages:: Using @value{GDBN} with different languages
145 * Symbols:: Examining the symbol table
146 * Altering:: Altering execution
147 * GDB Files:: @value{GDBN} files
148 * Targets:: Specifying a debugging target
149 * Remote Debugging:: Debugging remote programs
150 * Configurations:: Configuration-specific information
151 * Controlling GDB:: Controlling @value{GDBN}
152 * Extending GDB:: Extending @value{GDBN}
153 * Interpreters:: Command Interpreters
154 * TUI:: @value{GDBN} Text User Interface
155 * Emacs:: Using @value{GDBN} under @sc{gnu} Emacs
156 * GDB/MI:: @value{GDBN}'s Machine Interface.
157 * Annotations:: @value{GDBN}'s annotation interface.
158 * JIT Interface:: Using the JIT debugging interface.
160 * GDB Bugs:: Reporting bugs in @value{GDBN}
162 @ifset SYSTEM_READLINE
163 * Command Line Editing: (rluserman). Command Line Editing
164 * Using History Interactively: (history). Using History Interactively
166 @ifclear SYSTEM_READLINE
167 * Command Line Editing:: Command Line Editing
168 * Using History Interactively:: Using History Interactively
170 * In Memoriam:: In Memoriam
171 * Formatting Documentation:: How to format and print @value{GDBN} documentation
172 * Installing GDB:: Installing GDB
173 * Maintenance Commands:: Maintenance Commands
174 * Remote Protocol:: GDB Remote Serial Protocol
175 * Agent Expressions:: The GDB Agent Expression Mechanism
176 * Target Descriptions:: How targets can describe themselves to
178 * Operating System Information:: Getting additional information from
180 * Trace File Format:: GDB trace file format
181 * Index Section Format:: .gdb_index section format
182 * Copying:: GNU General Public License says
183 how you can copy and share GDB
184 * GNU Free Documentation License:: The license for this documentation
193 @unnumbered Summary of @value{GDBN}
195 The purpose of a debugger such as @value{GDBN} is to allow you to see what is
196 going on ``inside'' another program while it executes---or what another
197 program was doing at the moment it crashed.
199 @value{GDBN} can do four main kinds of things (plus other things in support of
200 these) to help you catch bugs in the act:
204 Start your program, specifying anything that might affect its behavior.
207 Make your program stop on specified conditions.
210 Examine what has happened, when your program has stopped.
213 Change things in your program, so you can experiment with correcting the
214 effects of one bug and go on to learn about another.
217 You can use @value{GDBN} to debug programs written in C and C@t{++}.
218 For more information, see @ref{Supported Languages,,Supported Languages}.
219 For more information, see @ref{C,,C and C++}.
221 Support for D is partial. For information on D, see
225 Support for Modula-2 is partial. For information on Modula-2, see
226 @ref{Modula-2,,Modula-2}.
228 Support for OpenCL C is partial. For information on OpenCL C, see
229 @ref{OpenCL C,,OpenCL C}.
232 Debugging Pascal programs which use sets, subranges, file variables, or
233 nested functions does not currently work. @value{GDBN} does not support
234 entering expressions, printing values, or similar features using Pascal
238 @value{GDBN} can be used to debug programs written in Fortran, although
239 it may be necessary to refer to some variables with a trailing
242 @value{GDBN} can be used to debug programs written in Objective-C,
243 using either the Apple/NeXT or the GNU Objective-C runtime.
246 * Free Software:: Freely redistributable software
247 * Contributors:: Contributors to GDB
251 @unnumberedsec Free Software
253 @value{GDBN} is @dfn{free software}, protected by the @sc{gnu}
254 General Public License
255 (GPL). The GPL gives you the freedom to copy or adapt a licensed
256 program---but every person getting a copy also gets with it the
257 freedom to modify that copy (which means that they must get access to
258 the source code), and the freedom to distribute further copies.
259 Typical software companies use copyrights to limit your freedoms; the
260 Free Software Foundation uses the GPL to preserve these freedoms.
262 Fundamentally, the General Public License is a license which says that
263 you have these freedoms and that you cannot take these freedoms away
266 @unnumberedsec Free Software Needs Free Documentation
268 The biggest deficiency in the free software community today is not in
269 the software---it is the lack of good free documentation that we can
270 include with the free software. Many of our most important
271 programs do not come with free reference manuals and free introductory
272 texts. Documentation is an essential part of any software package;
273 when an important free software package does not come with a free
274 manual and a free tutorial, that is a major gap. We have many such
277 Consider Perl, for instance. The tutorial manuals that people
278 normally use are non-free. How did this come about? Because the
279 authors of those manuals published them with restrictive terms---no
280 copying, no modification, source files not available---which exclude
281 them from the free software world.
283 That wasn't the first time this sort of thing happened, and it was far
284 from the last. Many times we have heard a GNU user eagerly describe a
285 manual that he is writing, his intended contribution to the community,
286 only to learn that he had ruined everything by signing a publication
287 contract to make it non-free.
289 Free documentation, like free software, is a matter of freedom, not
290 price. The problem with the non-free manual is not that publishers
291 charge a price for printed copies---that in itself is fine. (The Free
292 Software Foundation sells printed copies of manuals, too.) The
293 problem is the restrictions on the use of the manual. Free manuals
294 are available in source code form, and give you permission to copy and
295 modify. Non-free manuals do not allow this.
297 The criteria of freedom for a free manual are roughly the same as for
298 free software. Redistribution (including the normal kinds of
299 commercial redistribution) must be permitted, so that the manual can
300 accompany every copy of the program, both on-line and on paper.
302 Permission for modification of the technical content is crucial too.
303 When people modify the software, adding or changing features, if they
304 are conscientious they will change the manual too---so they can
305 provide accurate and clear documentation for the modified program. A
306 manual that leaves you no choice but to write a new manual to document
307 a changed version of the program is not really available to our
310 Some kinds of limits on the way modification is handled are
311 acceptable. For example, requirements to preserve the original
312 author's copyright notice, the distribution terms, or the list of
313 authors, are ok. It is also no problem to require modified versions
314 to include notice that they were modified. Even entire sections that
315 may not be deleted or changed are acceptable, as long as they deal
316 with nontechnical topics (like this one). These kinds of restrictions
317 are acceptable because they don't obstruct the community's normal use
320 However, it must be possible to modify all the @emph{technical}
321 content of the manual, and then distribute the result in all the usual
322 media, through all the usual channels. Otherwise, the restrictions
323 obstruct the use of the manual, it is not free, and we need another
324 manual to replace it.
326 Please spread the word about this issue. Our community continues to
327 lose manuals to proprietary publishing. If we spread the word that
328 free software needs free reference manuals and free tutorials, perhaps
329 the next person who wants to contribute by writing documentation will
330 realize, before it is too late, that only free manuals contribute to
331 the free software community.
333 If you are writing documentation, please insist on publishing it under
334 the GNU Free Documentation License or another free documentation
335 license. Remember that this decision requires your approval---you
336 don't have to let the publisher decide. Some commercial publishers
337 will use a free license if you insist, but they will not propose the
338 option; it is up to you to raise the issue and say firmly that this is
339 what you want. If the publisher you are dealing with refuses, please
340 try other publishers. If you're not sure whether a proposed license
341 is free, write to @email{licensing@@gnu.org}.
343 You can encourage commercial publishers to sell more free, copylefted
344 manuals and tutorials by buying them, and particularly by buying
345 copies from the publishers that paid for their writing or for major
346 improvements. Meanwhile, try to avoid buying non-free documentation
347 at all. Check the distribution terms of a manual before you buy it,
348 and insist that whoever seeks your business must respect your freedom.
349 Check the history of the book, and try to reward the publishers that
350 have paid or pay the authors to work on it.
352 The Free Software Foundation maintains a list of free documentation
353 published by other publishers, at
354 @url{http://www.fsf.org/doc/other-free-books.html}.
357 @unnumberedsec Contributors to @value{GDBN}
359 Richard Stallman was the original author of @value{GDBN}, and of many
360 other @sc{gnu} programs. Many others have contributed to its
361 development. This section attempts to credit major contributors. One
362 of the virtues of free software is that everyone is free to contribute
363 to it; with regret, we cannot actually acknowledge everyone here. The
364 file @file{ChangeLog} in the @value{GDBN} distribution approximates a
365 blow-by-blow account.
367 Changes much prior to version 2.0 are lost in the mists of time.
370 @emph{Plea:} Additions to this section are particularly welcome. If you
371 or your friends (or enemies, to be evenhanded) have been unfairly
372 omitted from this list, we would like to add your names!
375 So that they may not regard their many labors as thankless, we
376 particularly thank those who shepherded @value{GDBN} through major
378 Andrew Cagney (releases 6.3, 6.2, 6.1, 6.0, 5.3, 5.2, 5.1 and 5.0);
379 Jim Blandy (release 4.18);
380 Jason Molenda (release 4.17);
381 Stan Shebs (release 4.14);
382 Fred Fish (releases 4.16, 4.15, 4.13, 4.12, 4.11, 4.10, and 4.9);
383 Stu Grossman and John Gilmore (releases 4.8, 4.7, 4.6, 4.5, and 4.4);
384 John Gilmore (releases 4.3, 4.2, 4.1, 4.0, and 3.9);
385 Jim Kingdon (releases 3.5, 3.4, and 3.3);
386 and Randy Smith (releases 3.2, 3.1, and 3.0).
388 Richard Stallman, assisted at various times by Peter TerMaat, Chris
389 Hanson, and Richard Mlynarik, handled releases through 2.8.
391 Michael Tiemann is the author of most of the @sc{gnu} C@t{++} support
392 in @value{GDBN}, with significant additional contributions from Per
393 Bothner and Daniel Berlin. James Clark wrote the @sc{gnu} C@t{++}
394 demangler. Early work on C@t{++} was by Peter TerMaat (who also did
395 much general update work leading to release 3.0).
397 @value{GDBN} uses the BFD subroutine library to examine multiple
398 object-file formats; BFD was a joint project of David V.
399 Henkel-Wallace, Rich Pixley, Steve Chamberlain, and John Gilmore.
401 David Johnson wrote the original COFF support; Pace Willison did
402 the original support for encapsulated COFF.
404 Brent Benson of Harris Computer Systems contributed DWARF 2 support.
406 Adam de Boor and Bradley Davis contributed the ISI Optimum V support.
407 Per Bothner, Noboyuki Hikichi, and Alessandro Forin contributed MIPS
409 Jean-Daniel Fekete contributed Sun 386i support.
410 Chris Hanson improved the HP9000 support.
411 Noboyuki Hikichi and Tomoyuki Hasei contributed Sony/News OS 3 support.
412 David Johnson contributed Encore Umax support.
413 Jyrki Kuoppala contributed Altos 3068 support.
414 Jeff Law contributed HP PA and SOM support.
415 Keith Packard contributed NS32K support.
416 Doug Rabson contributed Acorn Risc Machine support.
417 Bob Rusk contributed Harris Nighthawk CX-UX support.
418 Chris Smith contributed Convex support (and Fortran debugging).
419 Jonathan Stone contributed Pyramid support.
420 Michael Tiemann contributed SPARC support.
421 Tim Tucker contributed support for the Gould NP1 and Gould Powernode.
422 Pace Willison contributed Intel 386 support.
423 Jay Vosburgh contributed Symmetry support.
424 Marko Mlinar contributed OpenRISC 1000 support.
426 Andreas Schwab contributed M68K @sc{gnu}/Linux support.
428 Rich Schaefer and Peter Schauer helped with support of SunOS shared
431 Jay Fenlason and Roland McGrath ensured that @value{GDBN} and GAS agree
432 about several machine instruction sets.
434 Patrick Duval, Ted Goldstein, Vikram Koka and Glenn Engel helped develop
435 remote debugging. Intel Corporation, Wind River Systems, AMD, and ARM
436 contributed remote debugging modules for the i960, VxWorks, A29K UDI,
437 and RDI targets, respectively.
439 Brian Fox is the author of the readline libraries providing
440 command-line editing and command history.
442 Andrew Beers of SUNY Buffalo wrote the language-switching code, the
443 Modula-2 support, and contributed the Languages chapter of this manual.
445 Fred Fish wrote most of the support for Unix System Vr4.
446 He also enhanced the command-completion support to cover C@t{++} overloaded
449 Hitachi America (now Renesas America), Ltd. sponsored the support for
450 H8/300, H8/500, and Super-H processors.
452 NEC sponsored the support for the v850, Vr4xxx, and Vr5xxx processors.
454 Mitsubishi (now Renesas) sponsored the support for D10V, D30V, and M32R/D
457 Toshiba sponsored the support for the TX39 Mips processor.
459 Matsushita sponsored the support for the MN10200 and MN10300 processors.
461 Fujitsu sponsored the support for SPARClite and FR30 processors.
463 Kung Hsu, Jeff Law, and Rick Sladkey added support for hardware
466 Michael Snyder added support for tracepoints.
468 Stu Grossman wrote gdbserver.
470 Jim Kingdon, Peter Schauer, Ian Taylor, and Stu Grossman made
471 nearly innumerable bug fixes and cleanups throughout @value{GDBN}.
473 The following people at the Hewlett-Packard Company contributed
474 support for the PA-RISC 2.0 architecture, HP-UX 10.20, 10.30, and 11.0
475 (narrow mode), HP's implementation of kernel threads, HP's aC@t{++}
476 compiler, and the Text User Interface (nee Terminal User Interface):
477 Ben Krepp, Richard Title, John Bishop, Susan Macchia, Kathy Mann,
478 Satish Pai, India Paul, Steve Rehrauer, and Elena Zannoni. Kim Haase
479 provided HP-specific information in this manual.
481 DJ Delorie ported @value{GDBN} to MS-DOS, for the DJGPP project.
482 Robert Hoehne made significant contributions to the DJGPP port.
484 Cygnus Solutions has sponsored @value{GDBN} maintenance and much of its
485 development since 1991. Cygnus engineers who have worked on @value{GDBN}
486 fulltime include Mark Alexander, Jim Blandy, Per Bothner, Kevin
487 Buettner, Edith Epstein, Chris Faylor, Fred Fish, Martin Hunt, Jim
488 Ingham, John Gilmore, Stu Grossman, Kung Hsu, Jim Kingdon, John Metzler,
489 Fernando Nasser, Geoffrey Noer, Dawn Perchik, Rich Pixley, Zdenek
490 Radouch, Keith Seitz, Stan Shebs, David Taylor, and Elena Zannoni. In
491 addition, Dave Brolley, Ian Carmichael, Steve Chamberlain, Nick Clifton,
492 JT Conklin, Stan Cox, DJ Delorie, Ulrich Drepper, Frank Eigler, Doug
493 Evans, Sean Fagan, David Henkel-Wallace, Richard Henderson, Jeff
494 Holcomb, Jeff Law, Jim Lemke, Tom Lord, Bob Manson, Michael Meissner,
495 Jason Merrill, Catherine Moore, Drew Moseley, Ken Raeburn, Gavin
496 Romig-Koch, Rob Savoye, Jamie Smith, Mike Stump, Ian Taylor, Angela
497 Thomas, Michael Tiemann, Tom Tromey, Ron Unrau, Jim Wilson, and David
498 Zuhn have made contributions both large and small.
500 Andrew Cagney, Fernando Nasser, and Elena Zannoni, while working for
501 Cygnus Solutions, implemented the original @sc{gdb/mi} interface.
503 Jim Blandy added support for preprocessor macros, while working for Red
506 Andrew Cagney designed @value{GDBN}'s architecture vector. Many
507 people including Andrew Cagney, Stephane Carrez, Randolph Chung, Nick
508 Duffek, Richard Henderson, Mark Kettenis, Grace Sainsbury, Kei
509 Sakamoto, Yoshinori Sato, Michael Snyder, Andreas Schwab, Jason
510 Thorpe, Corinna Vinschen, Ulrich Weigand, and Elena Zannoni, helped
511 with the migration of old architectures to this new framework.
513 Andrew Cagney completely re-designed and re-implemented @value{GDBN}'s
514 unwinder framework, this consisting of a fresh new design featuring
515 frame IDs, independent frame sniffers, and the sentinel frame. Mark
516 Kettenis implemented the @sc{dwarf 2} unwinder, Jeff Johnston the
517 libunwind unwinder, and Andrew Cagney the dummy, sentinel, tramp, and
518 trad unwinders. The architecture-specific changes, each involving a
519 complete rewrite of the architecture's frame code, were carried out by
520 Jim Blandy, Joel Brobecker, Kevin Buettner, Andrew Cagney, Stephane
521 Carrez, Randolph Chung, Orjan Friberg, Richard Henderson, Daniel
522 Jacobowitz, Jeff Johnston, Mark Kettenis, Theodore A. Roth, Kei
523 Sakamoto, Yoshinori Sato, Michael Snyder, Corinna Vinschen, and Ulrich
526 Christian Zankel, Ross Morley, Bob Wilson, and Maxim Grigoriev from
527 Tensilica, Inc.@: contributed support for Xtensa processors. Others
528 who have worked on the Xtensa port of @value{GDBN} in the past include
529 Steve Tjiang, John Newlin, and Scott Foehner.
531 Michael Eager and staff of Xilinx, Inc., contributed support for the
532 Xilinx MicroBlaze architecture.
535 @chapter A Sample @value{GDBN} Session
537 You can use this manual at your leisure to read all about @value{GDBN}.
538 However, a handful of commands are enough to get started using the
539 debugger. This chapter illustrates those commands.
542 In this sample session, we emphasize user input like this: @b{input},
543 to make it easier to pick out from the surrounding output.
546 @c FIXME: this example may not be appropriate for some configs, where
547 @c FIXME...primary interest is in remote use.
549 One of the preliminary versions of @sc{gnu} @code{m4} (a generic macro
550 processor) exhibits the following bug: sometimes, when we change its
551 quote strings from the default, the commands used to capture one macro
552 definition within another stop working. In the following short @code{m4}
553 session, we define a macro @code{foo} which expands to @code{0000}; we
554 then use the @code{m4} built-in @code{defn} to define @code{bar} as the
555 same thing. However, when we change the open quote string to
556 @code{<QUOTE>} and the close quote string to @code{<UNQUOTE>}, the same
557 procedure fails to define a new synonym @code{baz}:
566 @b{define(bar,defn(`foo'))}
570 @b{changequote(<QUOTE>,<UNQUOTE>)}
572 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
575 m4: End of input: 0: fatal error: EOF in string
579 Let us use @value{GDBN} to try to see what is going on.
582 $ @b{@value{GDBP} m4}
583 @c FIXME: this falsifies the exact text played out, to permit smallbook
584 @c FIXME... format to come out better.
585 @value{GDBN} is free software and you are welcome to distribute copies
586 of it under certain conditions; type "show copying" to see
588 There is absolutely no warranty for @value{GDBN}; type "show warranty"
591 @value{GDBN} @value{GDBVN}, Copyright 1999 Free Software Foundation, Inc...
596 @value{GDBN} reads only enough symbol data to know where to find the
597 rest when needed; as a result, the first prompt comes up very quickly.
598 We now tell @value{GDBN} to use a narrower display width than usual, so
599 that examples fit in this manual.
602 (@value{GDBP}) @b{set width 70}
606 We need to see how the @code{m4} built-in @code{changequote} works.
607 Having looked at the source, we know the relevant subroutine is
608 @code{m4_changequote}, so we set a breakpoint there with the @value{GDBN}
609 @code{break} command.
612 (@value{GDBP}) @b{break m4_changequote}
613 Breakpoint 1 at 0x62f4: file builtin.c, line 879.
617 Using the @code{run} command, we start @code{m4} running under @value{GDBN}
618 control; as long as control does not reach the @code{m4_changequote}
619 subroutine, the program runs as usual:
622 (@value{GDBP}) @b{run}
623 Starting program: /work/Editorial/gdb/gnu/m4/m4
631 To trigger the breakpoint, we call @code{changequote}. @value{GDBN}
632 suspends execution of @code{m4}, displaying information about the
633 context where it stops.
636 @b{changequote(<QUOTE>,<UNQUOTE>)}
638 Breakpoint 1, m4_changequote (argc=3, argv=0x33c70)
640 879 if (bad_argc(TOKEN_DATA_TEXT(argv[0]),argc,1,3))
644 Now we use the command @code{n} (@code{next}) to advance execution to
645 the next line of the current function.
649 882 set_quotes((argc >= 2) ? TOKEN_DATA_TEXT(argv[1])\
654 @code{set_quotes} looks like a promising subroutine. We can go into it
655 by using the command @code{s} (@code{step}) instead of @code{next}.
656 @code{step} goes to the next line to be executed in @emph{any}
657 subroutine, so it steps into @code{set_quotes}.
661 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
663 530 if (lquote != def_lquote)
667 The display that shows the subroutine where @code{m4} is now
668 suspended (and its arguments) is called a stack frame display. It
669 shows a summary of the stack. We can use the @code{backtrace}
670 command (which can also be spelled @code{bt}), to see where we are
671 in the stack as a whole: the @code{backtrace} command displays a
672 stack frame for each active subroutine.
675 (@value{GDBP}) @b{bt}
676 #0 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
678 #1 0x6344 in m4_changequote (argc=3, argv=0x33c70)
680 #2 0x8174 in expand_macro (sym=0x33320) at macro.c:242
681 #3 0x7a88 in expand_token (obs=0x0, t=209696, td=0xf7fffa30)
683 #4 0x79dc in expand_input () at macro.c:40
684 #5 0x2930 in main (argc=0, argv=0xf7fffb20) at m4.c:195
688 We step through a few more lines to see what happens. The first two
689 times, we can use @samp{s}; the next two times we use @code{n} to avoid
690 falling into the @code{xstrdup} subroutine.
694 0x3b5c 532 if (rquote != def_rquote)
696 0x3b80 535 lquote = (lq == nil || *lq == '\0') ? \
697 def_lquote : xstrdup(lq);
699 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
702 538 len_lquote = strlen(rquote);
706 The last line displayed looks a little odd; we can examine the variables
707 @code{lquote} and @code{rquote} to see if they are in fact the new left
708 and right quotes we specified. We use the command @code{p}
709 (@code{print}) to see their values.
712 (@value{GDBP}) @b{p lquote}
713 $1 = 0x35d40 "<QUOTE>"
714 (@value{GDBP}) @b{p rquote}
715 $2 = 0x35d50 "<UNQUOTE>"
719 @code{lquote} and @code{rquote} are indeed the new left and right quotes.
720 To look at some context, we can display ten lines of source
721 surrounding the current line with the @code{l} (@code{list}) command.
727 535 lquote = (lq == nil || *lq == '\0') ? def_lquote\
729 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
732 538 len_lquote = strlen(rquote);
733 539 len_rquote = strlen(lquote);
740 Let us step past the two lines that set @code{len_lquote} and
741 @code{len_rquote}, and then examine the values of those variables.
745 539 len_rquote = strlen(lquote);
748 (@value{GDBP}) @b{p len_lquote}
750 (@value{GDBP}) @b{p len_rquote}
755 That certainly looks wrong, assuming @code{len_lquote} and
756 @code{len_rquote} are meant to be the lengths of @code{lquote} and
757 @code{rquote} respectively. We can set them to better values using
758 the @code{p} command, since it can print the value of
759 any expression---and that expression can include subroutine calls and
763 (@value{GDBP}) @b{p len_lquote=strlen(lquote)}
765 (@value{GDBP}) @b{p len_rquote=strlen(rquote)}
770 Is that enough to fix the problem of using the new quotes with the
771 @code{m4} built-in @code{defn}? We can allow @code{m4} to continue
772 executing with the @code{c} (@code{continue}) command, and then try the
773 example that caused trouble initially:
779 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
786 Success! The new quotes now work just as well as the default ones. The
787 problem seems to have been just the two typos defining the wrong
788 lengths. We allow @code{m4} exit by giving it an EOF as input:
792 Program exited normally.
796 The message @samp{Program exited normally.} is from @value{GDBN}; it
797 indicates @code{m4} has finished executing. We can end our @value{GDBN}
798 session with the @value{GDBN} @code{quit} command.
801 (@value{GDBP}) @b{quit}
805 @chapter Getting In and Out of @value{GDBN}
807 This chapter discusses how to start @value{GDBN}, and how to get out of it.
811 type @samp{@value{GDBP}} to start @value{GDBN}.
813 type @kbd{quit} or @kbd{Ctrl-d} to exit.
817 * Invoking GDB:: How to start @value{GDBN}
818 * Quitting GDB:: How to quit @value{GDBN}
819 * Shell Commands:: How to use shell commands inside @value{GDBN}
820 * Logging Output:: How to log @value{GDBN}'s output to a file
824 @section Invoking @value{GDBN}
826 Invoke @value{GDBN} by running the program @code{@value{GDBP}}. Once started,
827 @value{GDBN} reads commands from the terminal until you tell it to exit.
829 You can also run @code{@value{GDBP}} with a variety of arguments and options,
830 to specify more of your debugging environment at the outset.
832 The command-line options described here are designed
833 to cover a variety of situations; in some environments, some of these
834 options may effectively be unavailable.
836 The most usual way to start @value{GDBN} is with one argument,
837 specifying an executable program:
840 @value{GDBP} @var{program}
844 You can also start with both an executable program and a core file
848 @value{GDBP} @var{program} @var{core}
851 You can, instead, specify a process ID as a second argument, if you want
852 to debug a running process:
855 @value{GDBP} @var{program} 1234
859 would attach @value{GDBN} to process @code{1234} (unless you also have a file
860 named @file{1234}; @value{GDBN} does check for a core file first).
862 Taking advantage of the second command-line argument requires a fairly
863 complete operating system; when you use @value{GDBN} as a remote
864 debugger attached to a bare board, there may not be any notion of
865 ``process'', and there is often no way to get a core dump. @value{GDBN}
866 will warn you if it is unable to attach or to read core dumps.
868 You can optionally have @code{@value{GDBP}} pass any arguments after the
869 executable file to the inferior using @code{--args}. This option stops
872 @value{GDBP} --args gcc -O2 -c foo.c
874 This will cause @code{@value{GDBP}} to debug @code{gcc}, and to set
875 @code{gcc}'s command-line arguments (@pxref{Arguments}) to @samp{-O2 -c foo.c}.
877 You can run @code{@value{GDBP}} without printing the front material, which describes
878 @value{GDBN}'s non-warranty, by specifying @code{-silent}:
885 You can further control how @value{GDBN} starts up by using command-line
886 options. @value{GDBN} itself can remind you of the options available.
896 to display all available options and briefly describe their use
897 (@samp{@value{GDBP} -h} is a shorter equivalent).
899 All options and command line arguments you give are processed
900 in sequential order. The order makes a difference when the
901 @samp{-x} option is used.
905 * File Options:: Choosing files
906 * Mode Options:: Choosing modes
907 * Startup:: What @value{GDBN} does during startup
911 @subsection Choosing Files
913 When @value{GDBN} starts, it reads any arguments other than options as
914 specifying an executable file and core file (or process ID). This is
915 the same as if the arguments were specified by the @samp{-se} and
916 @samp{-c} (or @samp{-p}) options respectively. (@value{GDBN} reads the
917 first argument that does not have an associated option flag as
918 equivalent to the @samp{-se} option followed by that argument; and the
919 second argument that does not have an associated option flag, if any, as
920 equivalent to the @samp{-c}/@samp{-p} option followed by that argument.)
921 If the second argument begins with a decimal digit, @value{GDBN} will
922 first attempt to attach to it as a process, and if that fails, attempt
923 to open it as a corefile. If you have a corefile whose name begins with
924 a digit, you can prevent @value{GDBN} from treating it as a pid by
925 prefixing it with @file{./}, e.g.@: @file{./12345}.
927 If @value{GDBN} has not been configured to included core file support,
928 such as for most embedded targets, then it will complain about a second
929 argument and ignore it.
931 Many options have both long and short forms; both are shown in the
932 following list. @value{GDBN} also recognizes the long forms if you truncate
933 them, so long as enough of the option is present to be unambiguous.
934 (If you prefer, you can flag option arguments with @samp{--} rather
935 than @samp{-}, though we illustrate the more usual convention.)
937 @c NOTE: the @cindex entries here use double dashes ON PURPOSE. This
938 @c way, both those who look for -foo and --foo in the index, will find
942 @item -symbols @var{file}
944 @cindex @code{--symbols}
946 Read symbol table from file @var{file}.
948 @item -exec @var{file}
950 @cindex @code{--exec}
952 Use file @var{file} as the executable file to execute when appropriate,
953 and for examining pure data in conjunction with a core dump.
957 Read symbol table from file @var{file} and use it as the executable
960 @item -core @var{file}
962 @cindex @code{--core}
964 Use file @var{file} as a core dump to examine.
966 @item -pid @var{number}
967 @itemx -p @var{number}
970 Connect to process ID @var{number}, as with the @code{attach} command.
972 @item -command @var{file}
974 @cindex @code{--command}
976 Execute commands from file @var{file}. The contents of this file is
977 evaluated exactly as the @code{source} command would.
978 @xref{Command Files,, Command files}.
980 @item -eval-command @var{command}
981 @itemx -ex @var{command}
982 @cindex @code{--eval-command}
984 Execute a single @value{GDBN} command.
986 This option may be used multiple times to call multiple commands. It may
987 also be interleaved with @samp{-command} as required.
990 @value{GDBP} -ex 'target sim' -ex 'load' \
991 -x setbreakpoints -ex 'run' a.out
994 @item -directory @var{directory}
995 @itemx -d @var{directory}
996 @cindex @code{--directory}
998 Add @var{directory} to the path to search for source and script files.
1002 @cindex @code{--readnow}
1004 Read each symbol file's entire symbol table immediately, rather than
1005 the default, which is to read it incrementally as it is needed.
1006 This makes startup slower, but makes future operations faster.
1011 @subsection Choosing Modes
1013 You can run @value{GDBN} in various alternative modes---for example, in
1014 batch mode or quiet mode.
1021 Do not execute commands found in any initialization files. Normally,
1022 @value{GDBN} executes the commands in these files after all the command
1023 options and arguments have been processed. @xref{Command Files,,Command
1029 @cindex @code{--quiet}
1030 @cindex @code{--silent}
1032 ``Quiet''. Do not print the introductory and copyright messages. These
1033 messages are also suppressed in batch mode.
1036 @cindex @code{--batch}
1037 Run in batch mode. Exit with status @code{0} after processing all the
1038 command files specified with @samp{-x} (and all commands from
1039 initialization files, if not inhibited with @samp{-n}). Exit with
1040 nonzero status if an error occurs in executing the @value{GDBN} commands
1041 in the command files. Batch mode also disables pagination, sets unlimited
1042 terminal width and height @pxref{Screen Size}, and acts as if @kbd{set confirm
1043 off} were in effect (@pxref{Messages/Warnings}).
1045 Batch mode may be useful for running @value{GDBN} as a filter, for
1046 example to download and run a program on another computer; in order to
1047 make this more useful, the message
1050 Program exited normally.
1054 (which is ordinarily issued whenever a program running under
1055 @value{GDBN} control terminates) is not issued when running in batch
1059 @cindex @code{--batch-silent}
1060 Run in batch mode exactly like @samp{-batch}, but totally silently. All
1061 @value{GDBN} output to @code{stdout} is prevented (@code{stderr} is
1062 unaffected). This is much quieter than @samp{-silent} and would be useless
1063 for an interactive session.
1065 This is particularly useful when using targets that give @samp{Loading section}
1066 messages, for example.
1068 Note that targets that give their output via @value{GDBN}, as opposed to
1069 writing directly to @code{stdout}, will also be made silent.
1071 @item -return-child-result
1072 @cindex @code{--return-child-result}
1073 The return code from @value{GDBN} will be the return code from the child
1074 process (the process being debugged), with the following exceptions:
1078 @value{GDBN} exits abnormally. E.g., due to an incorrect argument or an
1079 internal error. In this case the exit code is the same as it would have been
1080 without @samp{-return-child-result}.
1082 The user quits with an explicit value. E.g., @samp{quit 1}.
1084 The child process never runs, or is not allowed to terminate, in which case
1085 the exit code will be -1.
1088 This option is useful in conjunction with @samp{-batch} or @samp{-batch-silent},
1089 when @value{GDBN} is being used as a remote program loader or simulator
1094 @cindex @code{--nowindows}
1096 ``No windows''. If @value{GDBN} comes with a graphical user interface
1097 (GUI) built in, then this option tells @value{GDBN} to only use the command-line
1098 interface. If no GUI is available, this option has no effect.
1102 @cindex @code{--windows}
1104 If @value{GDBN} includes a GUI, then this option requires it to be
1107 @item -cd @var{directory}
1109 Run @value{GDBN} using @var{directory} as its working directory,
1110 instead of the current directory.
1112 @item -data-directory @var{directory}
1113 @cindex @code{--data-directory}
1114 Run @value{GDBN} using @var{directory} as its data directory.
1115 The data directory is where @value{GDBN} searches for its
1116 auxiliary files. @xref{Data Files}.
1120 @cindex @code{--fullname}
1122 @sc{gnu} Emacs sets this option when it runs @value{GDBN} as a
1123 subprocess. It tells @value{GDBN} to output the full file name and line
1124 number in a standard, recognizable fashion each time a stack frame is
1125 displayed (which includes each time your program stops). This
1126 recognizable format looks like two @samp{\032} characters, followed by
1127 the file name, line number and character position separated by colons,
1128 and a newline. The Emacs-to-@value{GDBN} interface program uses the two
1129 @samp{\032} characters as a signal to display the source code for the
1133 @cindex @code{--epoch}
1134 The Epoch Emacs-@value{GDBN} interface sets this option when it runs
1135 @value{GDBN} as a subprocess. It tells @value{GDBN} to modify its print
1136 routines so as to allow Epoch to display values of expressions in a
1139 @item -annotate @var{level}
1140 @cindex @code{--annotate}
1141 This option sets the @dfn{annotation level} inside @value{GDBN}. Its
1142 effect is identical to using @samp{set annotate @var{level}}
1143 (@pxref{Annotations}). The annotation @var{level} controls how much
1144 information @value{GDBN} prints together with its prompt, values of
1145 expressions, source lines, and other types of output. Level 0 is the
1146 normal, level 1 is for use when @value{GDBN} is run as a subprocess of
1147 @sc{gnu} Emacs, level 3 is the maximum annotation suitable for programs
1148 that control @value{GDBN}, and level 2 has been deprecated.
1150 The annotation mechanism has largely been superseded by @sc{gdb/mi}
1154 @cindex @code{--args}
1155 Change interpretation of command line so that arguments following the
1156 executable file are passed as command line arguments to the inferior.
1157 This option stops option processing.
1159 @item -baud @var{bps}
1161 @cindex @code{--baud}
1163 Set the line speed (baud rate or bits per second) of any serial
1164 interface used by @value{GDBN} for remote debugging.
1166 @item -l @var{timeout}
1168 Set the timeout (in seconds) of any communication used by @value{GDBN}
1169 for remote debugging.
1171 @item -tty @var{device}
1172 @itemx -t @var{device}
1173 @cindex @code{--tty}
1175 Run using @var{device} for your program's standard input and output.
1176 @c FIXME: kingdon thinks there is more to -tty. Investigate.
1178 @c resolve the situation of these eventually
1180 @cindex @code{--tui}
1181 Activate the @dfn{Text User Interface} when starting. The Text User
1182 Interface manages several text windows on the terminal, showing
1183 source, assembly, registers and @value{GDBN} command outputs
1184 (@pxref{TUI, ,@value{GDBN} Text User Interface}). Alternatively, the
1185 Text User Interface can be enabled by invoking the program
1186 @samp{@value{GDBTUI}}. Do not use this option if you run @value{GDBN} from
1187 Emacs (@pxref{Emacs, ,Using @value{GDBN} under @sc{gnu} Emacs}).
1190 @c @cindex @code{--xdb}
1191 @c Run in XDB compatibility mode, allowing the use of certain XDB commands.
1192 @c For information, see the file @file{xdb_trans.html}, which is usually
1193 @c installed in the directory @code{/opt/langtools/wdb/doc} on HP-UX
1196 @item -interpreter @var{interp}
1197 @cindex @code{--interpreter}
1198 Use the interpreter @var{interp} for interface with the controlling
1199 program or device. This option is meant to be set by programs which
1200 communicate with @value{GDBN} using it as a back end.
1201 @xref{Interpreters, , Command Interpreters}.
1203 @samp{--interpreter=mi} (or @samp{--interpreter=mi2}) causes
1204 @value{GDBN} to use the @dfn{@sc{gdb/mi} interface} (@pxref{GDB/MI, ,
1205 The @sc{gdb/mi} Interface}) included since @value{GDBN} version 6.0. The
1206 previous @sc{gdb/mi} interface, included in @value{GDBN} version 5.3 and
1207 selected with @samp{--interpreter=mi1}, is deprecated. Earlier
1208 @sc{gdb/mi} interfaces are no longer supported.
1211 @cindex @code{--write}
1212 Open the executable and core files for both reading and writing. This
1213 is equivalent to the @samp{set write on} command inside @value{GDBN}
1217 @cindex @code{--statistics}
1218 This option causes @value{GDBN} to print statistics about time and
1219 memory usage after it completes each command and returns to the prompt.
1222 @cindex @code{--version}
1223 This option causes @value{GDBN} to print its version number and
1224 no-warranty blurb, and exit.
1229 @subsection What @value{GDBN} Does During Startup
1230 @cindex @value{GDBN} startup
1232 Here's the description of what @value{GDBN} does during session startup:
1236 Sets up the command interpreter as specified by the command line
1237 (@pxref{Mode Options, interpreter}).
1241 Reads the system-wide @dfn{init file} (if @option{--with-system-gdbinit} was
1242 used when building @value{GDBN}; @pxref{System-wide configuration,
1243 ,System-wide configuration and settings}) and executes all the commands in
1247 Reads the init file (if any) in your home directory@footnote{On
1248 DOS/Windows systems, the home directory is the one pointed to by the
1249 @code{HOME} environment variable.} and executes all the commands in
1253 Processes command line options and operands.
1256 Reads and executes the commands from init file (if any) in the current
1257 working directory. This is only done if the current directory is
1258 different from your home directory. Thus, you can have more than one
1259 init file, one generic in your home directory, and another, specific
1260 to the program you are debugging, in the directory where you invoke
1264 If the command line specified a program to debug, or a process to
1265 attach to, or a core file, @value{GDBN} loads any auto-loaded
1266 scripts provided for the program or for its loaded shared libraries.
1267 @xref{Auto-loading}.
1269 If you wish to disable the auto-loading during startup,
1270 you must do something like the following:
1273 $ gdb -ex "set auto-load-scripts off" -ex "file myprogram"
1276 The following does not work because the auto-loading is turned off too late:
1279 $ gdb -ex "set auto-load-scripts off" myprogram
1283 Reads command files specified by the @samp{-x} option. @xref{Command
1284 Files}, for more details about @value{GDBN} command files.
1287 Reads the command history recorded in the @dfn{history file}.
1288 @xref{Command History}, for more details about the command history and the
1289 files where @value{GDBN} records it.
1292 Init files use the same syntax as @dfn{command files} (@pxref{Command
1293 Files}) and are processed by @value{GDBN} in the same way. The init
1294 file in your home directory can set options (such as @samp{set
1295 complaints}) that affect subsequent processing of command line options
1296 and operands. Init files are not executed if you use the @samp{-nx}
1297 option (@pxref{Mode Options, ,Choosing Modes}).
1299 To display the list of init files loaded by gdb at startup, you
1300 can use @kbd{gdb --help}.
1302 @cindex init file name
1303 @cindex @file{.gdbinit}
1304 @cindex @file{gdb.ini}
1305 The @value{GDBN} init files are normally called @file{.gdbinit}.
1306 The DJGPP port of @value{GDBN} uses the name @file{gdb.ini}, due to
1307 the limitations of file names imposed by DOS filesystems. The Windows
1308 ports of @value{GDBN} use the standard name, but if they find a
1309 @file{gdb.ini} file, they warn you about that and suggest to rename
1310 the file to the standard name.
1314 @section Quitting @value{GDBN}
1315 @cindex exiting @value{GDBN}
1316 @cindex leaving @value{GDBN}
1319 @kindex quit @r{[}@var{expression}@r{]}
1320 @kindex q @r{(@code{quit})}
1321 @item quit @r{[}@var{expression}@r{]}
1323 To exit @value{GDBN}, use the @code{quit} command (abbreviated
1324 @code{q}), or type an end-of-file character (usually @kbd{Ctrl-d}). If you
1325 do not supply @var{expression}, @value{GDBN} will terminate normally;
1326 otherwise it will terminate using the result of @var{expression} as the
1331 An interrupt (often @kbd{Ctrl-c}) does not exit from @value{GDBN}, but rather
1332 terminates the action of any @value{GDBN} command that is in progress and
1333 returns to @value{GDBN} command level. It is safe to type the interrupt
1334 character at any time because @value{GDBN} does not allow it to take effect
1335 until a time when it is safe.
1337 If you have been using @value{GDBN} to control an attached process or
1338 device, you can release it with the @code{detach} command
1339 (@pxref{Attach, ,Debugging an Already-running Process}).
1341 @node Shell Commands
1342 @section Shell Commands
1344 If you need to execute occasional shell commands during your
1345 debugging session, there is no need to leave or suspend @value{GDBN}; you can
1346 just use the @code{shell} command.
1350 @cindex shell escape
1351 @item shell @var{command string}
1352 Invoke a standard shell to execute @var{command string}.
1353 If it exists, the environment variable @code{SHELL} determines which
1354 shell to run. Otherwise @value{GDBN} uses the default shell
1355 (@file{/bin/sh} on Unix systems, @file{COMMAND.COM} on MS-DOS, etc.).
1358 The utility @code{make} is often needed in development environments.
1359 You do not have to use the @code{shell} command for this purpose in
1364 @cindex calling make
1365 @item make @var{make-args}
1366 Execute the @code{make} program with the specified
1367 arguments. This is equivalent to @samp{shell make @var{make-args}}.
1370 @node Logging Output
1371 @section Logging Output
1372 @cindex logging @value{GDBN} output
1373 @cindex save @value{GDBN} output to a file
1375 You may want to save the output of @value{GDBN} commands to a file.
1376 There are several commands to control @value{GDBN}'s logging.
1380 @item set logging on
1382 @item set logging off
1384 @cindex logging file name
1385 @item set logging file @var{file}
1386 Change the name of the current logfile. The default logfile is @file{gdb.txt}.
1387 @item set logging overwrite [on|off]
1388 By default, @value{GDBN} will append to the logfile. Set @code{overwrite} if
1389 you want @code{set logging on} to overwrite the logfile instead.
1390 @item set logging redirect [on|off]
1391 By default, @value{GDBN} output will go to both the terminal and the logfile.
1392 Set @code{redirect} if you want output to go only to the log file.
1393 @kindex show logging
1395 Show the current values of the logging settings.
1399 @chapter @value{GDBN} Commands
1401 You can abbreviate a @value{GDBN} command to the first few letters of the command
1402 name, if that abbreviation is unambiguous; and you can repeat certain
1403 @value{GDBN} commands by typing just @key{RET}. You can also use the @key{TAB}
1404 key to get @value{GDBN} to fill out the rest of a word in a command (or to
1405 show you the alternatives available, if there is more than one possibility).
1408 * Command Syntax:: How to give commands to @value{GDBN}
1409 * Completion:: Command completion
1410 * Help:: How to ask @value{GDBN} for help
1413 @node Command Syntax
1414 @section Command Syntax
1416 A @value{GDBN} command is a single line of input. There is no limit on
1417 how long it can be. It starts with a command name, which is followed by
1418 arguments whose meaning depends on the command name. For example, the
1419 command @code{step} accepts an argument which is the number of times to
1420 step, as in @samp{step 5}. You can also use the @code{step} command
1421 with no arguments. Some commands do not allow any arguments.
1423 @cindex abbreviation
1424 @value{GDBN} command names may always be truncated if that abbreviation is
1425 unambiguous. Other possible command abbreviations are listed in the
1426 documentation for individual commands. In some cases, even ambiguous
1427 abbreviations are allowed; for example, @code{s} is specially defined as
1428 equivalent to @code{step} even though there are other commands whose
1429 names start with @code{s}. You can test abbreviations by using them as
1430 arguments to the @code{help} command.
1432 @cindex repeating commands
1433 @kindex RET @r{(repeat last command)}
1434 A blank line as input to @value{GDBN} (typing just @key{RET}) means to
1435 repeat the previous command. Certain commands (for example, @code{run})
1436 will not repeat this way; these are commands whose unintentional
1437 repetition might cause trouble and which you are unlikely to want to
1438 repeat. User-defined commands can disable this feature; see
1439 @ref{Define, dont-repeat}.
1441 The @code{list} and @code{x} commands, when you repeat them with
1442 @key{RET}, construct new arguments rather than repeating
1443 exactly as typed. This permits easy scanning of source or memory.
1445 @value{GDBN} can also use @key{RET} in another way: to partition lengthy
1446 output, in a way similar to the common utility @code{more}
1447 (@pxref{Screen Size,,Screen Size}). Since it is easy to press one
1448 @key{RET} too many in this situation, @value{GDBN} disables command
1449 repetition after any command that generates this sort of display.
1451 @kindex # @r{(a comment)}
1453 Any text from a @kbd{#} to the end of the line is a comment; it does
1454 nothing. This is useful mainly in command files (@pxref{Command
1455 Files,,Command Files}).
1457 @cindex repeating command sequences
1458 @kindex Ctrl-o @r{(operate-and-get-next)}
1459 The @kbd{Ctrl-o} binding is useful for repeating a complex sequence of
1460 commands. This command accepts the current line, like @key{RET}, and
1461 then fetches the next line relative to the current line from the history
1465 @section Command Completion
1468 @cindex word completion
1469 @value{GDBN} can fill in the rest of a word in a command for you, if there is
1470 only one possibility; it can also show you what the valid possibilities
1471 are for the next word in a command, at any time. This works for @value{GDBN}
1472 commands, @value{GDBN} subcommands, and the names of symbols in your program.
1474 Press the @key{TAB} key whenever you want @value{GDBN} to fill out the rest
1475 of a word. If there is only one possibility, @value{GDBN} fills in the
1476 word, and waits for you to finish the command (or press @key{RET} to
1477 enter it). For example, if you type
1479 @c FIXME "@key" does not distinguish its argument sufficiently to permit
1480 @c complete accuracy in these examples; space introduced for clarity.
1481 @c If texinfo enhancements make it unnecessary, it would be nice to
1482 @c replace " @key" by "@key" in the following...
1484 (@value{GDBP}) info bre @key{TAB}
1488 @value{GDBN} fills in the rest of the word @samp{breakpoints}, since that is
1489 the only @code{info} subcommand beginning with @samp{bre}:
1492 (@value{GDBP}) info breakpoints
1496 You can either press @key{RET} at this point, to run the @code{info
1497 breakpoints} command, or backspace and enter something else, if
1498 @samp{breakpoints} does not look like the command you expected. (If you
1499 were sure you wanted @code{info breakpoints} in the first place, you
1500 might as well just type @key{RET} immediately after @samp{info bre},
1501 to exploit command abbreviations rather than command completion).
1503 If there is more than one possibility for the next word when you press
1504 @key{TAB}, @value{GDBN} sounds a bell. You can either supply more
1505 characters and try again, or just press @key{TAB} a second time;
1506 @value{GDBN} displays all the possible completions for that word. For
1507 example, you might want to set a breakpoint on a subroutine whose name
1508 begins with @samp{make_}, but when you type @kbd{b make_@key{TAB}} @value{GDBN}
1509 just sounds the bell. Typing @key{TAB} again displays all the
1510 function names in your program that begin with those characters, for
1514 (@value{GDBP}) b make_ @key{TAB}
1515 @exdent @value{GDBN} sounds bell; press @key{TAB} again, to see:
1516 make_a_section_from_file make_environ
1517 make_abs_section make_function_type
1518 make_blockvector make_pointer_type
1519 make_cleanup make_reference_type
1520 make_command make_symbol_completion_list
1521 (@value{GDBP}) b make_
1525 After displaying the available possibilities, @value{GDBN} copies your
1526 partial input (@samp{b make_} in the example) so you can finish the
1529 If you just want to see the list of alternatives in the first place, you
1530 can press @kbd{M-?} rather than pressing @key{TAB} twice. @kbd{M-?}
1531 means @kbd{@key{META} ?}. You can type this either by holding down a
1532 key designated as the @key{META} shift on your keyboard (if there is
1533 one) while typing @kbd{?}, or as @key{ESC} followed by @kbd{?}.
1535 @cindex quotes in commands
1536 @cindex completion of quoted strings
1537 Sometimes the string you need, while logically a ``word'', may contain
1538 parentheses or other characters that @value{GDBN} normally excludes from
1539 its notion of a word. To permit word completion to work in this
1540 situation, you may enclose words in @code{'} (single quote marks) in
1541 @value{GDBN} commands.
1543 The most likely situation where you might need this is in typing the
1544 name of a C@t{++} function. This is because C@t{++} allows function
1545 overloading (multiple definitions of the same function, distinguished
1546 by argument type). For example, when you want to set a breakpoint you
1547 may need to distinguish whether you mean the version of @code{name}
1548 that takes an @code{int} parameter, @code{name(int)}, or the version
1549 that takes a @code{float} parameter, @code{name(float)}. To use the
1550 word-completion facilities in this situation, type a single quote
1551 @code{'} at the beginning of the function name. This alerts
1552 @value{GDBN} that it may need to consider more information than usual
1553 when you press @key{TAB} or @kbd{M-?} to request word completion:
1556 (@value{GDBP}) b 'bubble( @kbd{M-?}
1557 bubble(double,double) bubble(int,int)
1558 (@value{GDBP}) b 'bubble(
1561 In some cases, @value{GDBN} can tell that completing a name requires using
1562 quotes. When this happens, @value{GDBN} inserts the quote for you (while
1563 completing as much as it can) if you do not type the quote in the first
1567 (@value{GDBP}) b bub @key{TAB}
1568 @exdent @value{GDBN} alters your input line to the following, and rings a bell:
1569 (@value{GDBP}) b 'bubble(
1573 In general, @value{GDBN} can tell that a quote is needed (and inserts it) if
1574 you have not yet started typing the argument list when you ask for
1575 completion on an overloaded symbol.
1577 For more information about overloaded functions, see @ref{C Plus Plus
1578 Expressions, ,C@t{++} Expressions}. You can use the command @code{set
1579 overload-resolution off} to disable overload resolution;
1580 see @ref{Debugging C Plus Plus, ,@value{GDBN} Features for C@t{++}}.
1582 @cindex completion of structure field names
1583 @cindex structure field name completion
1584 @cindex completion of union field names
1585 @cindex union field name completion
1586 When completing in an expression which looks up a field in a
1587 structure, @value{GDBN} also tries@footnote{The completer can be
1588 confused by certain kinds of invalid expressions. Also, it only
1589 examines the static type of the expression, not the dynamic type.} to
1590 limit completions to the field names available in the type of the
1594 (@value{GDBP}) p gdb_stdout.@kbd{M-?}
1595 magic to_delete to_fputs to_put to_rewind
1596 to_data to_flush to_isatty to_read to_write
1600 This is because the @code{gdb_stdout} is a variable of the type
1601 @code{struct ui_file} that is defined in @value{GDBN} sources as
1608 ui_file_flush_ftype *to_flush;
1609 ui_file_write_ftype *to_write;
1610 ui_file_fputs_ftype *to_fputs;
1611 ui_file_read_ftype *to_read;
1612 ui_file_delete_ftype *to_delete;
1613 ui_file_isatty_ftype *to_isatty;
1614 ui_file_rewind_ftype *to_rewind;
1615 ui_file_put_ftype *to_put;
1622 @section Getting Help
1623 @cindex online documentation
1626 You can always ask @value{GDBN} itself for information on its commands,
1627 using the command @code{help}.
1630 @kindex h @r{(@code{help})}
1633 You can use @code{help} (abbreviated @code{h}) with no arguments to
1634 display a short list of named classes of commands:
1638 List of classes of commands:
1640 aliases -- Aliases of other commands
1641 breakpoints -- Making program stop at certain points
1642 data -- Examining data
1643 files -- Specifying and examining files
1644 internals -- Maintenance commands
1645 obscure -- Obscure features
1646 running -- Running the program
1647 stack -- Examining the stack
1648 status -- Status inquiries
1649 support -- Support facilities
1650 tracepoints -- Tracing of program execution without
1651 stopping the program
1652 user-defined -- User-defined commands
1654 Type "help" followed by a class name for a list of
1655 commands in that class.
1656 Type "help" followed by command name for full
1658 Command name abbreviations are allowed if unambiguous.
1661 @c the above line break eliminates huge line overfull...
1663 @item help @var{class}
1664 Using one of the general help classes as an argument, you can get a
1665 list of the individual commands in that class. For example, here is the
1666 help display for the class @code{status}:
1669 (@value{GDBP}) help status
1674 @c Line break in "show" line falsifies real output, but needed
1675 @c to fit in smallbook page size.
1676 info -- Generic command for showing things
1677 about the program being debugged
1678 show -- Generic command for showing things
1681 Type "help" followed by command name for full
1683 Command name abbreviations are allowed if unambiguous.
1687 @item help @var{command}
1688 With a command name as @code{help} argument, @value{GDBN} displays a
1689 short paragraph on how to use that command.
1692 @item apropos @var{args}
1693 The @code{apropos} command searches through all of the @value{GDBN}
1694 commands, and their documentation, for the regular expression specified in
1695 @var{args}. It prints out all matches found. For example:
1706 set symbol-reloading -- Set dynamic symbol table reloading
1707 multiple times in one run
1708 show symbol-reloading -- Show dynamic symbol table reloading
1709 multiple times in one run
1714 @item complete @var{args}
1715 The @code{complete @var{args}} command lists all the possible completions
1716 for the beginning of a command. Use @var{args} to specify the beginning of the
1717 command you want completed. For example:
1723 @noindent results in:
1734 @noindent This is intended for use by @sc{gnu} Emacs.
1737 In addition to @code{help}, you can use the @value{GDBN} commands @code{info}
1738 and @code{show} to inquire about the state of your program, or the state
1739 of @value{GDBN} itself. Each command supports many topics of inquiry; this
1740 manual introduces each of them in the appropriate context. The listings
1741 under @code{info} and under @code{show} in the Index point to
1742 all the sub-commands. @xref{Index}.
1747 @kindex i @r{(@code{info})}
1749 This command (abbreviated @code{i}) is for describing the state of your
1750 program. For example, you can show the arguments passed to a function
1751 with @code{info args}, list the registers currently in use with @code{info
1752 registers}, or list the breakpoints you have set with @code{info breakpoints}.
1753 You can get a complete list of the @code{info} sub-commands with
1754 @w{@code{help info}}.
1758 You can assign the result of an expression to an environment variable with
1759 @code{set}. For example, you can set the @value{GDBN} prompt to a $-sign with
1760 @code{set prompt $}.
1764 In contrast to @code{info}, @code{show} is for describing the state of
1765 @value{GDBN} itself.
1766 You can change most of the things you can @code{show}, by using the
1767 related command @code{set}; for example, you can control what number
1768 system is used for displays with @code{set radix}, or simply inquire
1769 which is currently in use with @code{show radix}.
1772 To display all the settable parameters and their current
1773 values, you can use @code{show} with no arguments; you may also use
1774 @code{info set}. Both commands produce the same display.
1775 @c FIXME: "info set" violates the rule that "info" is for state of
1776 @c FIXME...program. Ck w/ GNU: "info set" to be called something else,
1777 @c FIXME...or change desc of rule---eg "state of prog and debugging session"?
1781 Here are three miscellaneous @code{show} subcommands, all of which are
1782 exceptional in lacking corresponding @code{set} commands:
1785 @kindex show version
1786 @cindex @value{GDBN} version number
1788 Show what version of @value{GDBN} is running. You should include this
1789 information in @value{GDBN} bug-reports. If multiple versions of
1790 @value{GDBN} are in use at your site, you may need to determine which
1791 version of @value{GDBN} you are running; as @value{GDBN} evolves, new
1792 commands are introduced, and old ones may wither away. Also, many
1793 system vendors ship variant versions of @value{GDBN}, and there are
1794 variant versions of @value{GDBN} in @sc{gnu}/Linux distributions as well.
1795 The version number is the same as the one announced when you start
1798 @kindex show copying
1799 @kindex info copying
1800 @cindex display @value{GDBN} copyright
1803 Display information about permission for copying @value{GDBN}.
1805 @kindex show warranty
1806 @kindex info warranty
1808 @itemx info warranty
1809 Display the @sc{gnu} ``NO WARRANTY'' statement, or a warranty,
1810 if your version of @value{GDBN} comes with one.
1815 @chapter Running Programs Under @value{GDBN}
1817 When you run a program under @value{GDBN}, you must first generate
1818 debugging information when you compile it.
1820 You may start @value{GDBN} with its arguments, if any, in an environment
1821 of your choice. If you are doing native debugging, you may redirect
1822 your program's input and output, debug an already running process, or
1823 kill a child process.
1826 * Compilation:: Compiling for debugging
1827 * Starting:: Starting your program
1828 * Arguments:: Your program's arguments
1829 * Environment:: Your program's environment
1831 * Working Directory:: Your program's working directory
1832 * Input/Output:: Your program's input and output
1833 * Attach:: Debugging an already-running process
1834 * Kill Process:: Killing the child process
1836 * Inferiors and Programs:: Debugging multiple inferiors and programs
1837 * Threads:: Debugging programs with multiple threads
1838 * Forks:: Debugging forks
1839 * Checkpoint/Restart:: Setting a @emph{bookmark} to return to later
1843 @section Compiling for Debugging
1845 In order to debug a program effectively, you need to generate
1846 debugging information when you compile it. This debugging information
1847 is stored in the object file; it describes the data type of each
1848 variable or function and the correspondence between source line numbers
1849 and addresses in the executable code.
1851 To request debugging information, specify the @samp{-g} option when you run
1854 Programs that are to be shipped to your customers are compiled with
1855 optimizations, using the @samp{-O} compiler option. However, some
1856 compilers are unable to handle the @samp{-g} and @samp{-O} options
1857 together. Using those compilers, you cannot generate optimized
1858 executables containing debugging information.
1860 @value{NGCC}, the @sc{gnu} C/C@t{++} compiler, supports @samp{-g} with or
1861 without @samp{-O}, making it possible to debug optimized code. We
1862 recommend that you @emph{always} use @samp{-g} whenever you compile a
1863 program. You may think your program is correct, but there is no sense
1864 in pushing your luck. For more information, see @ref{Optimized Code}.
1866 Older versions of the @sc{gnu} C compiler permitted a variant option
1867 @w{@samp{-gg}} for debugging information. @value{GDBN} no longer supports this
1868 format; if your @sc{gnu} C compiler has this option, do not use it.
1870 @value{GDBN} knows about preprocessor macros and can show you their
1871 expansion (@pxref{Macros}). Most compilers do not include information
1872 about preprocessor macros in the debugging information if you specify
1873 the @option{-g} flag alone, because this information is rather large.
1874 Version 3.1 and later of @value{NGCC}, the @sc{gnu} C compiler,
1875 provides macro information if you specify the options
1876 @option{-gdwarf-2} and @option{-g3}; the former option requests
1877 debugging information in the Dwarf 2 format, and the latter requests
1878 ``extra information''. In the future, we hope to find more compact
1879 ways to represent macro information, so that it can be included with
1884 @section Starting your Program
1890 @kindex r @r{(@code{run})}
1893 Use the @code{run} command to start your program under @value{GDBN}.
1894 You must first specify the program name (except on VxWorks) with an
1895 argument to @value{GDBN} (@pxref{Invocation, ,Getting In and Out of
1896 @value{GDBN}}), or by using the @code{file} or @code{exec-file} command
1897 (@pxref{Files, ,Commands to Specify Files}).
1901 If you are running your program in an execution environment that
1902 supports processes, @code{run} creates an inferior process and makes
1903 that process run your program. In some environments without processes,
1904 @code{run} jumps to the start of your program. Other targets,
1905 like @samp{remote}, are always running. If you get an error
1906 message like this one:
1909 The "remote" target does not support "run".
1910 Try "help target" or "continue".
1914 then use @code{continue} to run your program. You may need @code{load}
1915 first (@pxref{load}).
1917 The execution of a program is affected by certain information it
1918 receives from its superior. @value{GDBN} provides ways to specify this
1919 information, which you must do @emph{before} starting your program. (You
1920 can change it after starting your program, but such changes only affect
1921 your program the next time you start it.) This information may be
1922 divided into four categories:
1925 @item The @emph{arguments.}
1926 Specify the arguments to give your program as the arguments of the
1927 @code{run} command. If a shell is available on your target, the shell
1928 is used to pass the arguments, so that you may use normal conventions
1929 (such as wildcard expansion or variable substitution) in describing
1931 In Unix systems, you can control which shell is used with the
1932 @code{SHELL} environment variable.
1933 @xref{Arguments, ,Your Program's Arguments}.
1935 @item The @emph{environment.}
1936 Your program normally inherits its environment from @value{GDBN}, but you can
1937 use the @value{GDBN} commands @code{set environment} and @code{unset
1938 environment} to change parts of the environment that affect
1939 your program. @xref{Environment, ,Your Program's Environment}.
1941 @item The @emph{working directory.}
1942 Your program inherits its working directory from @value{GDBN}. You can set
1943 the @value{GDBN} working directory with the @code{cd} command in @value{GDBN}.
1944 @xref{Working Directory, ,Your Program's Working Directory}.
1946 @item The @emph{standard input and output.}
1947 Your program normally uses the same device for standard input and
1948 standard output as @value{GDBN} is using. You can redirect input and output
1949 in the @code{run} command line, or you can use the @code{tty} command to
1950 set a different device for your program.
1951 @xref{Input/Output, ,Your Program's Input and Output}.
1954 @emph{Warning:} While input and output redirection work, you cannot use
1955 pipes to pass the output of the program you are debugging to another
1956 program; if you attempt this, @value{GDBN} is likely to wind up debugging the
1960 When you issue the @code{run} command, your program begins to execute
1961 immediately. @xref{Stopping, ,Stopping and Continuing}, for discussion
1962 of how to arrange for your program to stop. Once your program has
1963 stopped, you may call functions in your program, using the @code{print}
1964 or @code{call} commands. @xref{Data, ,Examining Data}.
1966 If the modification time of your symbol file has changed since the last
1967 time @value{GDBN} read its symbols, @value{GDBN} discards its symbol
1968 table, and reads it again. When it does this, @value{GDBN} tries to retain
1969 your current breakpoints.
1974 @cindex run to main procedure
1975 The name of the main procedure can vary from language to language.
1976 With C or C@t{++}, the main procedure name is always @code{main}, but
1977 other languages such as Ada do not require a specific name for their
1978 main procedure. The debugger provides a convenient way to start the
1979 execution of the program and to stop at the beginning of the main
1980 procedure, depending on the language used.
1982 The @samp{start} command does the equivalent of setting a temporary
1983 breakpoint at the beginning of the main procedure and then invoking
1984 the @samp{run} command.
1986 @cindex elaboration phase
1987 Some programs contain an @dfn{elaboration} phase where some startup code is
1988 executed before the main procedure is called. This depends on the
1989 languages used to write your program. In C@t{++}, for instance,
1990 constructors for static and global objects are executed before
1991 @code{main} is called. It is therefore possible that the debugger stops
1992 before reaching the main procedure. However, the temporary breakpoint
1993 will remain to halt execution.
1995 Specify the arguments to give to your program as arguments to the
1996 @samp{start} command. These arguments will be given verbatim to the
1997 underlying @samp{run} command. Note that the same arguments will be
1998 reused if no argument is provided during subsequent calls to
1999 @samp{start} or @samp{run}.
2001 It is sometimes necessary to debug the program during elaboration. In
2002 these cases, using the @code{start} command would stop the execution of
2003 your program too late, as the program would have already completed the
2004 elaboration phase. Under these circumstances, insert breakpoints in your
2005 elaboration code before running your program.
2007 @kindex set exec-wrapper
2008 @item set exec-wrapper @var{wrapper}
2009 @itemx show exec-wrapper
2010 @itemx unset exec-wrapper
2011 When @samp{exec-wrapper} is set, the specified wrapper is used to
2012 launch programs for debugging. @value{GDBN} starts your program
2013 with a shell command of the form @kbd{exec @var{wrapper}
2014 @var{program}}. Quoting is added to @var{program} and its
2015 arguments, but not to @var{wrapper}, so you should add quotes if
2016 appropriate for your shell. The wrapper runs until it executes
2017 your program, and then @value{GDBN} takes control.
2019 You can use any program that eventually calls @code{execve} with
2020 its arguments as a wrapper. Several standard Unix utilities do
2021 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
2022 with @code{exec "$@@"} will also work.
2024 For example, you can use @code{env} to pass an environment variable to
2025 the debugged program, without setting the variable in your shell's
2029 (@value{GDBP}) set exec-wrapper env 'LD_PRELOAD=libtest.so'
2033 This command is available when debugging locally on most targets, excluding
2034 @sc{djgpp}, Cygwin, MS Windows, and QNX Neutrino.
2036 @kindex set disable-randomization
2037 @item set disable-randomization
2038 @itemx set disable-randomization on
2039 This option (enabled by default in @value{GDBN}) will turn off the native
2040 randomization of the virtual address space of the started program. This option
2041 is useful for multiple debugging sessions to make the execution better
2042 reproducible and memory addresses reusable across debugging sessions.
2044 This feature is implemented only on @sc{gnu}/Linux. You can get the same
2048 (@value{GDBP}) set exec-wrapper setarch `uname -m` -R
2051 @item set disable-randomization off
2052 Leave the behavior of the started executable unchanged. Some bugs rear their
2053 ugly heads only when the program is loaded at certain addresses. If your bug
2054 disappears when you run the program under @value{GDBN}, that might be because
2055 @value{GDBN} by default disables the address randomization on platforms, such
2056 as @sc{gnu}/Linux, which do that for stand-alone programs. Use @kbd{set
2057 disable-randomization off} to try to reproduce such elusive bugs.
2059 The virtual address space randomization is implemented only on @sc{gnu}/Linux.
2060 It protects the programs against some kinds of security attacks. In these
2061 cases the attacker needs to know the exact location of a concrete executable
2062 code. Randomizing its location makes it impossible to inject jumps misusing
2063 a code at its expected addresses.
2065 Prelinking shared libraries provides a startup performance advantage but it
2066 makes addresses in these libraries predictable for privileged processes by
2067 having just unprivileged access at the target system. Reading the shared
2068 library binary gives enough information for assembling the malicious code
2069 misusing it. Still even a prelinked shared library can get loaded at a new
2070 random address just requiring the regular relocation process during the
2071 startup. Shared libraries not already prelinked are always loaded at
2072 a randomly chosen address.
2074 Position independent executables (PIE) contain position independent code
2075 similar to the shared libraries and therefore such executables get loaded at
2076 a randomly chosen address upon startup. PIE executables always load even
2077 already prelinked shared libraries at a random address. You can build such
2078 executable using @command{gcc -fPIE -pie}.
2080 Heap (malloc storage), stack and custom mmap areas are always placed randomly
2081 (as long as the randomization is enabled).
2083 @item show disable-randomization
2084 Show the current setting of the explicit disable of the native randomization of
2085 the virtual address space of the started program.
2090 @section Your Program's Arguments
2092 @cindex arguments (to your program)
2093 The arguments to your program can be specified by the arguments of the
2095 They are passed to a shell, which expands wildcard characters and
2096 performs redirection of I/O, and thence to your program. Your
2097 @code{SHELL} environment variable (if it exists) specifies what shell
2098 @value{GDBN} uses. If you do not define @code{SHELL}, @value{GDBN} uses
2099 the default shell (@file{/bin/sh} on Unix).
2101 On non-Unix systems, the program is usually invoked directly by
2102 @value{GDBN}, which emulates I/O redirection via the appropriate system
2103 calls, and the wildcard characters are expanded by the startup code of
2104 the program, not by the shell.
2106 @code{run} with no arguments uses the same arguments used by the previous
2107 @code{run}, or those set by the @code{set args} command.
2112 Specify the arguments to be used the next time your program is run. If
2113 @code{set args} has no arguments, @code{run} executes your program
2114 with no arguments. Once you have run your program with arguments,
2115 using @code{set args} before the next @code{run} is the only way to run
2116 it again without arguments.
2120 Show the arguments to give your program when it is started.
2124 @section Your Program's Environment
2126 @cindex environment (of your program)
2127 The @dfn{environment} consists of a set of environment variables and
2128 their values. Environment variables conventionally record such things as
2129 your user name, your home directory, your terminal type, and your search
2130 path for programs to run. Usually you set up environment variables with
2131 the shell and they are inherited by all the other programs you run. When
2132 debugging, it can be useful to try running your program with a modified
2133 environment without having to start @value{GDBN} over again.
2137 @item path @var{directory}
2138 Add @var{directory} to the front of the @code{PATH} environment variable
2139 (the search path for executables) that will be passed to your program.
2140 The value of @code{PATH} used by @value{GDBN} does not change.
2141 You may specify several directory names, separated by whitespace or by a
2142 system-dependent separator character (@samp{:} on Unix, @samp{;} on
2143 MS-DOS and MS-Windows). If @var{directory} is already in the path, it
2144 is moved to the front, so it is searched sooner.
2146 You can use the string @samp{$cwd} to refer to whatever is the current
2147 working directory at the time @value{GDBN} searches the path. If you
2148 use @samp{.} instead, it refers to the directory where you executed the
2149 @code{path} command. @value{GDBN} replaces @samp{.} in the
2150 @var{directory} argument (with the current path) before adding
2151 @var{directory} to the search path.
2152 @c 'path' is explicitly nonrepeatable, but RMS points out it is silly to
2153 @c document that, since repeating it would be a no-op.
2157 Display the list of search paths for executables (the @code{PATH}
2158 environment variable).
2160 @kindex show environment
2161 @item show environment @r{[}@var{varname}@r{]}
2162 Print the value of environment variable @var{varname} to be given to
2163 your program when it starts. If you do not supply @var{varname},
2164 print the names and values of all environment variables to be given to
2165 your program. You can abbreviate @code{environment} as @code{env}.
2167 @kindex set environment
2168 @item set environment @var{varname} @r{[}=@var{value}@r{]}
2169 Set environment variable @var{varname} to @var{value}. The value
2170 changes for your program only, not for @value{GDBN} itself. @var{value} may
2171 be any string; the values of environment variables are just strings, and
2172 any interpretation is supplied by your program itself. The @var{value}
2173 parameter is optional; if it is eliminated, the variable is set to a
2175 @c "any string" here does not include leading, trailing
2176 @c blanks. Gnu asks: does anyone care?
2178 For example, this command:
2185 tells the debugged program, when subsequently run, that its user is named
2186 @samp{foo}. (The spaces around @samp{=} are used for clarity here; they
2187 are not actually required.)
2189 @kindex unset environment
2190 @item unset environment @var{varname}
2191 Remove variable @var{varname} from the environment to be passed to your
2192 program. This is different from @samp{set env @var{varname} =};
2193 @code{unset environment} removes the variable from the environment,
2194 rather than assigning it an empty value.
2197 @emph{Warning:} On Unix systems, @value{GDBN} runs your program using
2199 by your @code{SHELL} environment variable if it exists (or
2200 @code{/bin/sh} if not). If your @code{SHELL} variable names a shell
2201 that runs an initialization file---such as @file{.cshrc} for C-shell, or
2202 @file{.bashrc} for BASH---any variables you set in that file affect
2203 your program. You may wish to move setting of environment variables to
2204 files that are only run when you sign on, such as @file{.login} or
2207 @node Working Directory
2208 @section Your Program's Working Directory
2210 @cindex working directory (of your program)
2211 Each time you start your program with @code{run}, it inherits its
2212 working directory from the current working directory of @value{GDBN}.
2213 The @value{GDBN} working directory is initially whatever it inherited
2214 from its parent process (typically the shell), but you can specify a new
2215 working directory in @value{GDBN} with the @code{cd} command.
2217 The @value{GDBN} working directory also serves as a default for the commands
2218 that specify files for @value{GDBN} to operate on. @xref{Files, ,Commands to
2223 @cindex change working directory
2224 @item cd @var{directory}
2225 Set the @value{GDBN} working directory to @var{directory}.
2229 Print the @value{GDBN} working directory.
2232 It is generally impossible to find the current working directory of
2233 the process being debugged (since a program can change its directory
2234 during its run). If you work on a system where @value{GDBN} is
2235 configured with the @file{/proc} support, you can use the @code{info
2236 proc} command (@pxref{SVR4 Process Information}) to find out the
2237 current working directory of the debuggee.
2240 @section Your Program's Input and Output
2245 By default, the program you run under @value{GDBN} does input and output to
2246 the same terminal that @value{GDBN} uses. @value{GDBN} switches the terminal
2247 to its own terminal modes to interact with you, but it records the terminal
2248 modes your program was using and switches back to them when you continue
2249 running your program.
2252 @kindex info terminal
2254 Displays information recorded by @value{GDBN} about the terminal modes your
2258 You can redirect your program's input and/or output using shell
2259 redirection with the @code{run} command. For example,
2266 starts your program, diverting its output to the file @file{outfile}.
2269 @cindex controlling terminal
2270 Another way to specify where your program should do input and output is
2271 with the @code{tty} command. This command accepts a file name as
2272 argument, and causes this file to be the default for future @code{run}
2273 commands. It also resets the controlling terminal for the child
2274 process, for future @code{run} commands. For example,
2281 directs that processes started with subsequent @code{run} commands
2282 default to do input and output on the terminal @file{/dev/ttyb} and have
2283 that as their controlling terminal.
2285 An explicit redirection in @code{run} overrides the @code{tty} command's
2286 effect on the input/output device, but not its effect on the controlling
2289 When you use the @code{tty} command or redirect input in the @code{run}
2290 command, only the input @emph{for your program} is affected. The input
2291 for @value{GDBN} still comes from your terminal. @code{tty} is an alias
2292 for @code{set inferior-tty}.
2294 @cindex inferior tty
2295 @cindex set inferior controlling terminal
2296 You can use the @code{show inferior-tty} command to tell @value{GDBN} to
2297 display the name of the terminal that will be used for future runs of your
2301 @item set inferior-tty /dev/ttyb
2302 @kindex set inferior-tty
2303 Set the tty for the program being debugged to /dev/ttyb.
2305 @item show inferior-tty
2306 @kindex show inferior-tty
2307 Show the current tty for the program being debugged.
2311 @section Debugging an Already-running Process
2316 @item attach @var{process-id}
2317 This command attaches to a running process---one that was started
2318 outside @value{GDBN}. (@code{info files} shows your active
2319 targets.) The command takes as argument a process ID. The usual way to
2320 find out the @var{process-id} of a Unix process is with the @code{ps} utility,
2321 or with the @samp{jobs -l} shell command.
2323 @code{attach} does not repeat if you press @key{RET} a second time after
2324 executing the command.
2327 To use @code{attach}, your program must be running in an environment
2328 which supports processes; for example, @code{attach} does not work for
2329 programs on bare-board targets that lack an operating system. You must
2330 also have permission to send the process a signal.
2332 When you use @code{attach}, the debugger finds the program running in
2333 the process first by looking in the current working directory, then (if
2334 the program is not found) by using the source file search path
2335 (@pxref{Source Path, ,Specifying Source Directories}). You can also use
2336 the @code{file} command to load the program. @xref{Files, ,Commands to
2339 The first thing @value{GDBN} does after arranging to debug the specified
2340 process is to stop it. You can examine and modify an attached process
2341 with all the @value{GDBN} commands that are ordinarily available when
2342 you start processes with @code{run}. You can insert breakpoints; you
2343 can step and continue; you can modify storage. If you would rather the
2344 process continue running, you may use the @code{continue} command after
2345 attaching @value{GDBN} to the process.
2350 When you have finished debugging the attached process, you can use the
2351 @code{detach} command to release it from @value{GDBN} control. Detaching
2352 the process continues its execution. After the @code{detach} command,
2353 that process and @value{GDBN} become completely independent once more, and you
2354 are ready to @code{attach} another process or start one with @code{run}.
2355 @code{detach} does not repeat if you press @key{RET} again after
2356 executing the command.
2359 If you exit @value{GDBN} while you have an attached process, you detach
2360 that process. If you use the @code{run} command, you kill that process.
2361 By default, @value{GDBN} asks for confirmation if you try to do either of these
2362 things; you can control whether or not you need to confirm by using the
2363 @code{set confirm} command (@pxref{Messages/Warnings, ,Optional Warnings and
2367 @section Killing the Child Process
2372 Kill the child process in which your program is running under @value{GDBN}.
2375 This command is useful if you wish to debug a core dump instead of a
2376 running process. @value{GDBN} ignores any core dump file while your program
2379 On some operating systems, a program cannot be executed outside @value{GDBN}
2380 while you have breakpoints set on it inside @value{GDBN}. You can use the
2381 @code{kill} command in this situation to permit running your program
2382 outside the debugger.
2384 The @code{kill} command is also useful if you wish to recompile and
2385 relink your program, since on many systems it is impossible to modify an
2386 executable file while it is running in a process. In this case, when you
2387 next type @code{run}, @value{GDBN} notices that the file has changed, and
2388 reads the symbol table again (while trying to preserve your current
2389 breakpoint settings).
2391 @node Inferiors and Programs
2392 @section Debugging Multiple Inferiors and Programs
2394 @value{GDBN} lets you run and debug multiple programs in a single
2395 session. In addition, @value{GDBN} on some systems may let you run
2396 several programs simultaneously (otherwise you have to exit from one
2397 before starting another). In the most general case, you can have
2398 multiple threads of execution in each of multiple processes, launched
2399 from multiple executables.
2402 @value{GDBN} represents the state of each program execution with an
2403 object called an @dfn{inferior}. An inferior typically corresponds to
2404 a process, but is more general and applies also to targets that do not
2405 have processes. Inferiors may be created before a process runs, and
2406 may be retained after a process exits. Inferiors have unique
2407 identifiers that are different from process ids. Usually each
2408 inferior will also have its own distinct address space, although some
2409 embedded targets may have several inferiors running in different parts
2410 of a single address space. Each inferior may in turn have multiple
2411 threads running in it.
2413 To find out what inferiors exist at any moment, use @w{@code{info
2417 @kindex info inferiors
2418 @item info inferiors
2419 Print a list of all inferiors currently being managed by @value{GDBN}.
2421 @value{GDBN} displays for each inferior (in this order):
2425 the inferior number assigned by @value{GDBN}
2428 the target system's inferior identifier
2431 the name of the executable the inferior is running.
2436 An asterisk @samp{*} preceding the @value{GDBN} inferior number
2437 indicates the current inferior.
2441 @c end table here to get a little more width for example
2444 (@value{GDBP}) info inferiors
2445 Num Description Executable
2446 2 process 2307 hello
2447 * 1 process 3401 goodbye
2450 To switch focus between inferiors, use the @code{inferior} command:
2453 @kindex inferior @var{infno}
2454 @item inferior @var{infno}
2455 Make inferior number @var{infno} the current inferior. The argument
2456 @var{infno} is the inferior number assigned by @value{GDBN}, as shown
2457 in the first field of the @samp{info inferiors} display.
2461 You can get multiple executables into a debugging session via the
2462 @code{add-inferior} and @w{@code{clone-inferior}} commands. On some
2463 systems @value{GDBN} can add inferiors to the debug session
2464 automatically by following calls to @code{fork} and @code{exec}. To
2465 remove inferiors from the debugging session use the
2466 @w{@code{remove-inferiors}} command.
2469 @kindex add-inferior
2470 @item add-inferior [ -copies @var{n} ] [ -exec @var{executable} ]
2471 Adds @var{n} inferiors to be run using @var{executable} as the
2472 executable. @var{n} defaults to 1. If no executable is specified,
2473 the inferiors begins empty, with no program. You can still assign or
2474 change the program assigned to the inferior at any time by using the
2475 @code{file} command with the executable name as its argument.
2477 @kindex clone-inferior
2478 @item clone-inferior [ -copies @var{n} ] [ @var{infno} ]
2479 Adds @var{n} inferiors ready to execute the same program as inferior
2480 @var{infno}. @var{n} defaults to 1. @var{infno} defaults to the
2481 number of the current inferior. This is a convenient command when you
2482 want to run another instance of the inferior you are debugging.
2485 (@value{GDBP}) info inferiors
2486 Num Description Executable
2487 * 1 process 29964 helloworld
2488 (@value{GDBP}) clone-inferior
2491 (@value{GDBP}) info inferiors
2492 Num Description Executable
2494 * 1 process 29964 helloworld
2497 You can now simply switch focus to inferior 2 and run it.
2499 @kindex remove-inferiors
2500 @item remove-inferiors @var{infno}@dots{}
2501 Removes the inferior or inferiors @var{infno}@dots{}. It is not
2502 possible to remove an inferior that is running with this command. For
2503 those, use the @code{kill} or @code{detach} command first.
2507 To quit debugging one of the running inferiors that is not the current
2508 inferior, you can either detach from it by using the @w{@code{detach
2509 inferior}} command (allowing it to run independently), or kill it
2510 using the @w{@code{kill inferiors}} command:
2513 @kindex detach inferiors @var{infno}@dots{}
2514 @item detach inferior @var{infno}@dots{}
2515 Detach from the inferior or inferiors identified by @value{GDBN}
2516 inferior number(s) @var{infno}@dots{}. Note that the inferior's entry
2517 still stays on the list of inferiors shown by @code{info inferiors},
2518 but its Description will show @samp{<null>}.
2520 @kindex kill inferiors @var{infno}@dots{}
2521 @item kill inferiors @var{infno}@dots{}
2522 Kill the inferior or inferiors identified by @value{GDBN} inferior
2523 number(s) @var{infno}@dots{}. Note that the inferior's entry still
2524 stays on the list of inferiors shown by @code{info inferiors}, but its
2525 Description will show @samp{<null>}.
2528 After the successful completion of a command such as @code{detach},
2529 @code{detach inferiors}, @code{kill} or @code{kill inferiors}, or after
2530 a normal process exit, the inferior is still valid and listed with
2531 @code{info inferiors}, ready to be restarted.
2534 To be notified when inferiors are started or exit under @value{GDBN}'s
2535 control use @w{@code{set print inferior-events}}:
2538 @kindex set print inferior-events
2539 @cindex print messages on inferior start and exit
2540 @item set print inferior-events
2541 @itemx set print inferior-events on
2542 @itemx set print inferior-events off
2543 The @code{set print inferior-events} command allows you to enable or
2544 disable printing of messages when @value{GDBN} notices that new
2545 inferiors have started or that inferiors have exited or have been
2546 detached. By default, these messages will not be printed.
2548 @kindex show print inferior-events
2549 @item show print inferior-events
2550 Show whether messages will be printed when @value{GDBN} detects that
2551 inferiors have started, exited or have been detached.
2554 Many commands will work the same with multiple programs as with a
2555 single program: e.g., @code{print myglobal} will simply display the
2556 value of @code{myglobal} in the current inferior.
2559 Occasionaly, when debugging @value{GDBN} itself, it may be useful to
2560 get more info about the relationship of inferiors, programs, address
2561 spaces in a debug session. You can do that with the @w{@code{maint
2562 info program-spaces}} command.
2565 @kindex maint info program-spaces
2566 @item maint info program-spaces
2567 Print a list of all program spaces currently being managed by
2570 @value{GDBN} displays for each program space (in this order):
2574 the program space number assigned by @value{GDBN}
2577 the name of the executable loaded into the program space, with e.g.,
2578 the @code{file} command.
2583 An asterisk @samp{*} preceding the @value{GDBN} program space number
2584 indicates the current program space.
2586 In addition, below each program space line, @value{GDBN} prints extra
2587 information that isn't suitable to display in tabular form. For
2588 example, the list of inferiors bound to the program space.
2591 (@value{GDBP}) maint info program-spaces
2594 Bound inferiors: ID 1 (process 21561)
2598 Here we can see that no inferior is running the program @code{hello},
2599 while @code{process 21561} is running the program @code{goodbye}. On
2600 some targets, it is possible that multiple inferiors are bound to the
2601 same program space. The most common example is that of debugging both
2602 the parent and child processes of a @code{vfork} call. For example,
2605 (@value{GDBP}) maint info program-spaces
2608 Bound inferiors: ID 2 (process 18050), ID 1 (process 18045)
2611 Here, both inferior 2 and inferior 1 are running in the same program
2612 space as a result of inferior 1 having executed a @code{vfork} call.
2616 @section Debugging Programs with Multiple Threads
2618 @cindex threads of execution
2619 @cindex multiple threads
2620 @cindex switching threads
2621 In some operating systems, such as HP-UX and Solaris, a single program
2622 may have more than one @dfn{thread} of execution. The precise semantics
2623 of threads differ from one operating system to another, but in general
2624 the threads of a single program are akin to multiple processes---except
2625 that they share one address space (that is, they can all examine and
2626 modify the same variables). On the other hand, each thread has its own
2627 registers and execution stack, and perhaps private memory.
2629 @value{GDBN} provides these facilities for debugging multi-thread
2633 @item automatic notification of new threads
2634 @item @samp{thread @var{threadno}}, a command to switch among threads
2635 @item @samp{info threads}, a command to inquire about existing threads
2636 @item @samp{thread apply [@var{threadno}] [@var{all}] @var{args}},
2637 a command to apply a command to a list of threads
2638 @item thread-specific breakpoints
2639 @item @samp{set print thread-events}, which controls printing of
2640 messages on thread start and exit.
2641 @item @samp{set libthread-db-search-path @var{path}}, which lets
2642 the user specify which @code{libthread_db} to use if the default choice
2643 isn't compatible with the program.
2647 @emph{Warning:} These facilities are not yet available on every
2648 @value{GDBN} configuration where the operating system supports threads.
2649 If your @value{GDBN} does not support threads, these commands have no
2650 effect. For example, a system without thread support shows no output
2651 from @samp{info threads}, and always rejects the @code{thread} command,
2655 (@value{GDBP}) info threads
2656 (@value{GDBP}) thread 1
2657 Thread ID 1 not known. Use the "info threads" command to
2658 see the IDs of currently known threads.
2660 @c FIXME to implementors: how hard would it be to say "sorry, this GDB
2661 @c doesn't support threads"?
2664 @cindex focus of debugging
2665 @cindex current thread
2666 The @value{GDBN} thread debugging facility allows you to observe all
2667 threads while your program runs---but whenever @value{GDBN} takes
2668 control, one thread in particular is always the focus of debugging.
2669 This thread is called the @dfn{current thread}. Debugging commands show
2670 program information from the perspective of the current thread.
2672 @cindex @code{New} @var{systag} message
2673 @cindex thread identifier (system)
2674 @c FIXME-implementors!! It would be more helpful if the [New...] message
2675 @c included GDB's numeric thread handle, so you could just go to that
2676 @c thread without first checking `info threads'.
2677 Whenever @value{GDBN} detects a new thread in your program, it displays
2678 the target system's identification for the thread with a message in the
2679 form @samp{[New @var{systag}]}. @var{systag} is a thread identifier
2680 whose form varies depending on the particular system. For example, on
2681 @sc{gnu}/Linux, you might see
2684 [New Thread 0x41e02940 (LWP 25582)]
2688 when @value{GDBN} notices a new thread. In contrast, on an SGI system,
2689 the @var{systag} is simply something like @samp{process 368}, with no
2692 @c FIXME!! (1) Does the [New...] message appear even for the very first
2693 @c thread of a program, or does it only appear for the
2694 @c second---i.e.@: when it becomes obvious we have a multithread
2696 @c (2) *Is* there necessarily a first thread always? Or do some
2697 @c multithread systems permit starting a program with multiple
2698 @c threads ab initio?
2700 @cindex thread number
2701 @cindex thread identifier (GDB)
2702 For debugging purposes, @value{GDBN} associates its own thread
2703 number---always a single integer---with each thread in your program.
2706 @kindex info threads
2707 @item info threads @r{[}@var{id}@dots{}@r{]}
2708 Display a summary of all threads currently in your program. Optional
2709 argument @var{id}@dots{} is one or more thread ids separated by spaces, and
2710 means to print information only about the specified thread or threads.
2711 @value{GDBN} displays for each thread (in this order):
2715 the thread number assigned by @value{GDBN}
2718 the target system's thread identifier (@var{systag})
2721 the thread's name, if one is known. A thread can either be named by
2722 the user (see @code{thread name}, below), or, in some cases, by the
2726 the current stack frame summary for that thread
2730 An asterisk @samp{*} to the left of the @value{GDBN} thread number
2731 indicates the current thread.
2735 @c end table here to get a little more width for example
2738 (@value{GDBP}) info threads
2740 3 process 35 thread 27 0x34e5 in sigpause ()
2741 2 process 35 thread 23 0x34e5 in sigpause ()
2742 * 1 process 35 thread 13 main (argc=1, argv=0x7ffffff8)
2746 On Solaris, you can display more information about user threads with a
2747 Solaris-specific command:
2750 @item maint info sol-threads
2751 @kindex maint info sol-threads
2752 @cindex thread info (Solaris)
2753 Display info on Solaris user threads.
2757 @kindex thread @var{threadno}
2758 @item thread @var{threadno}
2759 Make thread number @var{threadno} the current thread. The command
2760 argument @var{threadno} is the internal @value{GDBN} thread number, as
2761 shown in the first field of the @samp{info threads} display.
2762 @value{GDBN} responds by displaying the system identifier of the thread
2763 you selected, and its current stack frame summary:
2766 (@value{GDBP}) thread 2
2767 [Switching to thread 2 (Thread 0xb7fdab70 (LWP 12747))]
2768 #0 some_function (ignore=0x0) at example.c:8
2769 8 printf ("hello\n");
2773 As with the @samp{[New @dots{}]} message, the form of the text after
2774 @samp{Switching to} depends on your system's conventions for identifying
2777 @vindex $_thread@r{, convenience variable}
2778 The debugger convenience variable @samp{$_thread} contains the number
2779 of the current thread. You may find this useful in writing breakpoint
2780 conditional expressions, command scripts, and so forth. See
2781 @xref{Convenience Vars,, Convenience Variables}, for general
2782 information on convenience variables.
2784 @kindex thread apply
2785 @cindex apply command to several threads
2786 @item thread apply [@var{threadno} | 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}}.
2796 @cindex name a thread
2797 @item thread name [@var{name}]
2798 This command assigns a name to the current thread. If no argument is
2799 given, any existing user-specified name is removed. The thread name
2800 appears in the @samp{info threads} display.
2802 On some systems, such as @sc{gnu}/Linux, @value{GDBN} is able to
2803 determine the name of the thread as given by the OS. On these
2804 systems, a name specified with @samp{thread name} will override the
2805 system-give name, and removing the user-specified name will cause
2806 @value{GDBN} to once again display the system-specified name.
2809 @cindex search for a thread
2810 @item thread find [@var{regexp}]
2811 Search for and display thread ids whose name or @var{systag}
2812 matches the supplied regular expression.
2814 As well as being the complement to the @samp{thread name} command,
2815 this command also allows you to identify a thread by its target
2816 @var{systag}. For instance, on @sc{gnu}/Linux, the target @var{systag}
2820 (@value{GDBN}) thread find 26688
2821 Thread 4 has target id 'Thread 0x41e02940 (LWP 26688)'
2822 (@value{GDBN}) info thread 4
2824 4 Thread 0x41e02940 (LWP 26688) 0x00000031ca6cd372 in select ()
2827 @kindex set print thread-events
2828 @cindex print messages on thread start and exit
2829 @item set print thread-events
2830 @itemx set print thread-events on
2831 @itemx set print thread-events off
2832 The @code{set print thread-events} command allows you to enable or
2833 disable printing of messages when @value{GDBN} notices that new threads have
2834 started or that threads have exited. By default, these messages will
2835 be printed if detection of these events is supported by the target.
2836 Note that these messages cannot be disabled on all targets.
2838 @kindex show print thread-events
2839 @item show print thread-events
2840 Show whether messages will be printed when @value{GDBN} detects that threads
2841 have started and exited.
2844 @xref{Thread Stops,,Stopping and Starting Multi-thread Programs}, for
2845 more information about how @value{GDBN} behaves when you stop and start
2846 programs with multiple threads.
2848 @xref{Set Watchpoints,,Setting Watchpoints}, for information about
2849 watchpoints in programs with multiple threads.
2852 @kindex set libthread-db-search-path
2853 @cindex search path for @code{libthread_db}
2854 @item set libthread-db-search-path @r{[}@var{path}@r{]}
2855 If this variable is set, @var{path} is a colon-separated list of
2856 directories @value{GDBN} will use to search for @code{libthread_db}.
2857 If you omit @var{path}, @samp{libthread-db-search-path} will be reset to
2860 On @sc{gnu}/Linux and Solaris systems, @value{GDBN} uses a ``helper''
2861 @code{libthread_db} library to obtain information about threads in the
2862 inferior process. @value{GDBN} will use @samp{libthread-db-search-path}
2863 to find @code{libthread_db}. If that fails, @value{GDBN} will continue
2864 with default system shared library directories, and finally the directory
2865 from which @code{libpthread} was loaded in the inferior process.
2867 For any @code{libthread_db} library @value{GDBN} finds in above directories,
2868 @value{GDBN} attempts to initialize it with the current inferior process.
2869 If this initialization fails (which could happen because of a version
2870 mismatch between @code{libthread_db} and @code{libpthread}), @value{GDBN}
2871 will unload @code{libthread_db}, and continue with the next directory.
2872 If none of @code{libthread_db} libraries initialize successfully,
2873 @value{GDBN} will issue a warning and thread debugging will be disabled.
2875 Setting @code{libthread-db-search-path} is currently implemented
2876 only on some platforms.
2878 @kindex show libthread-db-search-path
2879 @item show libthread-db-search-path
2880 Display current libthread_db search path.
2882 @kindex set debug libthread-db
2883 @kindex show debug libthread-db
2884 @cindex debugging @code{libthread_db}
2885 @item set debug libthread-db
2886 @itemx show debug libthread-db
2887 Turns on or off display of @code{libthread_db}-related events.
2888 Use @code{1} to enable, @code{0} to disable.
2892 @section Debugging Forks
2894 @cindex fork, debugging programs which call
2895 @cindex multiple processes
2896 @cindex processes, multiple
2897 On most systems, @value{GDBN} has no special support for debugging
2898 programs which create additional processes using the @code{fork}
2899 function. When a program forks, @value{GDBN} will continue to debug the
2900 parent process and the child process will run unimpeded. If you have
2901 set a breakpoint in any code which the child then executes, the child
2902 will get a @code{SIGTRAP} signal which (unless it catches the signal)
2903 will cause it to terminate.
2905 However, if you want to debug the child process there is a workaround
2906 which isn't too painful. Put a call to @code{sleep} in the code which
2907 the child process executes after the fork. It may be useful to sleep
2908 only if a certain environment variable is set, or a certain file exists,
2909 so that the delay need not occur when you don't want to run @value{GDBN}
2910 on the child. While the child is sleeping, use the @code{ps} program to
2911 get its process ID. Then tell @value{GDBN} (a new invocation of
2912 @value{GDBN} if you are also debugging the parent process) to attach to
2913 the child process (@pxref{Attach}). From that point on you can debug
2914 the child process just like any other process which you attached to.
2916 On some systems, @value{GDBN} provides support for debugging programs that
2917 create additional processes using the @code{fork} or @code{vfork} functions.
2918 Currently, the only platforms with this feature are HP-UX (11.x and later
2919 only?) and @sc{gnu}/Linux (kernel version 2.5.60 and later).
2921 By default, when a program forks, @value{GDBN} will continue to debug
2922 the parent process and the child process will run unimpeded.
2924 If you want to follow the child process instead of the parent process,
2925 use the command @w{@code{set follow-fork-mode}}.
2928 @kindex set follow-fork-mode
2929 @item set follow-fork-mode @var{mode}
2930 Set the debugger response to a program call of @code{fork} or
2931 @code{vfork}. A call to @code{fork} or @code{vfork} creates a new
2932 process. The @var{mode} argument can be:
2936 The original process is debugged after a fork. The child process runs
2937 unimpeded. This is the default.
2940 The new process is debugged after a fork. The parent process runs
2945 @kindex show follow-fork-mode
2946 @item show follow-fork-mode
2947 Display the current debugger response to a @code{fork} or @code{vfork} call.
2950 @cindex debugging multiple processes
2951 On Linux, if you want to debug both the parent and child processes, use the
2952 command @w{@code{set detach-on-fork}}.
2955 @kindex set detach-on-fork
2956 @item set detach-on-fork @var{mode}
2957 Tells gdb whether to detach one of the processes after a fork, or
2958 retain debugger control over them both.
2962 The child process (or parent process, depending on the value of
2963 @code{follow-fork-mode}) will be detached and allowed to run
2964 independently. This is the default.
2967 Both processes will be held under the control of @value{GDBN}.
2968 One process (child or parent, depending on the value of
2969 @code{follow-fork-mode}) is debugged as usual, while the other
2974 @kindex show detach-on-fork
2975 @item show detach-on-fork
2976 Show whether detach-on-fork mode is on/off.
2979 If you choose to set @samp{detach-on-fork} mode off, then @value{GDBN}
2980 will retain control of all forked processes (including nested forks).
2981 You can list the forked processes under the control of @value{GDBN} by
2982 using the @w{@code{info inferiors}} command, and switch from one fork
2983 to another by using the @code{inferior} command (@pxref{Inferiors and
2984 Programs, ,Debugging Multiple Inferiors and Programs}).
2986 To quit debugging one of the forked processes, you can either detach
2987 from it by using the @w{@code{detach inferiors}} command (allowing it
2988 to run independently), or kill it using the @w{@code{kill inferiors}}
2989 command. @xref{Inferiors and Programs, ,Debugging Multiple Inferiors
2992 If you ask to debug a child process and a @code{vfork} is followed by an
2993 @code{exec}, @value{GDBN} executes the new target up to the first
2994 breakpoint in the new target. If you have a breakpoint set on
2995 @code{main} in your original program, the breakpoint will also be set on
2996 the child process's @code{main}.
2998 On some systems, when a child process is spawned by @code{vfork}, you
2999 cannot debug the child or parent until an @code{exec} call completes.
3001 If you issue a @code{run} command to @value{GDBN} after an @code{exec}
3002 call executes, the new target restarts. To restart the parent
3003 process, use the @code{file} command with the parent executable name
3004 as its argument. By default, after an @code{exec} call executes,
3005 @value{GDBN} discards the symbols of the previous executable image.
3006 You can change this behaviour with the @w{@code{set follow-exec-mode}}
3010 @kindex set follow-exec-mode
3011 @item set follow-exec-mode @var{mode}
3013 Set debugger response to a program call of @code{exec}. An
3014 @code{exec} call replaces the program image of a process.
3016 @code{follow-exec-mode} can be:
3020 @value{GDBN} creates a new inferior and rebinds the process to this
3021 new inferior. The program the process was running before the
3022 @code{exec} call can be restarted afterwards by restarting the
3028 (@value{GDBP}) info inferiors
3030 Id Description Executable
3033 process 12020 is executing new program: prog2
3034 Program exited normally.
3035 (@value{GDBP}) info inferiors
3036 Id Description Executable
3042 @value{GDBN} keeps the process bound to the same inferior. The new
3043 executable image replaces the previous executable loaded in the
3044 inferior. Restarting the inferior after the @code{exec} call, with
3045 e.g., the @code{run} command, restarts the executable the process was
3046 running after the @code{exec} call. This is the default mode.
3051 (@value{GDBP}) info inferiors
3052 Id Description Executable
3055 process 12020 is executing new program: prog2
3056 Program exited normally.
3057 (@value{GDBP}) info inferiors
3058 Id Description Executable
3065 You can use the @code{catch} command to make @value{GDBN} stop whenever
3066 a @code{fork}, @code{vfork}, or @code{exec} call is made. @xref{Set
3067 Catchpoints, ,Setting Catchpoints}.
3069 @node Checkpoint/Restart
3070 @section Setting a @emph{Bookmark} to Return to Later
3075 @cindex snapshot of a process
3076 @cindex rewind program state
3078 On certain operating systems@footnote{Currently, only
3079 @sc{gnu}/Linux.}, @value{GDBN} is able to save a @dfn{snapshot} of a
3080 program's state, called a @dfn{checkpoint}, and come back to it
3083 Returning to a checkpoint effectively undoes everything that has
3084 happened in the program since the @code{checkpoint} was saved. This
3085 includes changes in memory, registers, and even (within some limits)
3086 system state. Effectively, it is like going back in time to the
3087 moment when the checkpoint was saved.
3089 Thus, if you're stepping thru a program and you think you're
3090 getting close to the point where things go wrong, you can save
3091 a checkpoint. Then, if you accidentally go too far and miss
3092 the critical statement, instead of having to restart your program
3093 from the beginning, you can just go back to the checkpoint and
3094 start again from there.
3096 This can be especially useful if it takes a lot of time or
3097 steps to reach the point where you think the bug occurs.
3099 To use the @code{checkpoint}/@code{restart} method of debugging:
3104 Save a snapshot of the debugged program's current execution state.
3105 The @code{checkpoint} command takes no arguments, but each checkpoint
3106 is assigned a small integer id, similar to a breakpoint id.
3108 @kindex info checkpoints
3109 @item info checkpoints
3110 List the checkpoints that have been saved in the current debugging
3111 session. For each checkpoint, the following information will be
3118 @item Source line, or label
3121 @kindex restart @var{checkpoint-id}
3122 @item restart @var{checkpoint-id}
3123 Restore the program state that was saved as checkpoint number
3124 @var{checkpoint-id}. All program variables, registers, stack frames
3125 etc.@: will be returned to the values that they had when the checkpoint
3126 was saved. In essence, gdb will ``wind back the clock'' to the point
3127 in time when the checkpoint was saved.
3129 Note that breakpoints, @value{GDBN} variables, command history etc.
3130 are not affected by restoring a checkpoint. In general, a checkpoint
3131 only restores things that reside in the program being debugged, not in
3134 @kindex delete checkpoint @var{checkpoint-id}
3135 @item delete checkpoint @var{checkpoint-id}
3136 Delete the previously-saved checkpoint identified by @var{checkpoint-id}.
3140 Returning to a previously saved checkpoint will restore the user state
3141 of the program being debugged, plus a significant subset of the system
3142 (OS) state, including file pointers. It won't ``un-write'' data from
3143 a file, but it will rewind the file pointer to the previous location,
3144 so that the previously written data can be overwritten. For files
3145 opened in read mode, the pointer will also be restored so that the
3146 previously read data can be read again.
3148 Of course, characters that have been sent to a printer (or other
3149 external device) cannot be ``snatched back'', and characters received
3150 from eg.@: a serial device can be removed from internal program buffers,
3151 but they cannot be ``pushed back'' into the serial pipeline, ready to
3152 be received again. Similarly, the actual contents of files that have
3153 been changed cannot be restored (at this time).
3155 However, within those constraints, you actually can ``rewind'' your
3156 program to a previously saved point in time, and begin debugging it
3157 again --- and you can change the course of events so as to debug a
3158 different execution path this time.
3160 @cindex checkpoints and process id
3161 Finally, there is one bit of internal program state that will be
3162 different when you return to a checkpoint --- the program's process
3163 id. Each checkpoint will have a unique process id (or @var{pid}),
3164 and each will be different from the program's original @var{pid}.
3165 If your program has saved a local copy of its process id, this could
3166 potentially pose a problem.
3168 @subsection A Non-obvious Benefit of Using Checkpoints
3170 On some systems such as @sc{gnu}/Linux, address space randomization
3171 is performed on new processes for security reasons. This makes it
3172 difficult or impossible to set a breakpoint, or watchpoint, on an
3173 absolute address if you have to restart the program, since the
3174 absolute location of a symbol will change from one execution to the
3177 A checkpoint, however, is an @emph{identical} copy of a process.
3178 Therefore if you create a checkpoint at (eg.@:) the start of main,
3179 and simply return to that checkpoint instead of restarting the
3180 process, you can avoid the effects of address randomization and
3181 your symbols will all stay in the same place.
3184 @chapter Stopping and Continuing
3186 The principal purposes of using a debugger are so that you can stop your
3187 program before it terminates; or so that, if your program runs into
3188 trouble, you can investigate and find out why.
3190 Inside @value{GDBN}, your program may stop for any of several reasons,
3191 such as a signal, a breakpoint, or reaching a new line after a
3192 @value{GDBN} command such as @code{step}. You may then examine and
3193 change variables, set new breakpoints or remove old ones, and then
3194 continue execution. Usually, the messages shown by @value{GDBN} provide
3195 ample explanation of the status of your program---but you can also
3196 explicitly request this information at any time.
3199 @kindex info program
3201 Display information about the status of your program: whether it is
3202 running or not, what process it is, and why it stopped.
3206 * Breakpoints:: Breakpoints, watchpoints, and catchpoints
3207 * Continuing and Stepping:: Resuming execution
3209 * Thread Stops:: Stopping and starting multi-thread programs
3213 @section Breakpoints, Watchpoints, and Catchpoints
3216 A @dfn{breakpoint} makes your program stop whenever a certain point in
3217 the program is reached. For each breakpoint, you can add conditions to
3218 control in finer detail whether your program stops. You can set
3219 breakpoints with the @code{break} command and its variants (@pxref{Set
3220 Breaks, ,Setting Breakpoints}), to specify the place where your program
3221 should stop by line number, function name or exact address in the
3224 On some systems, you can set breakpoints in shared libraries before
3225 the executable is run. There is a minor limitation on HP-UX systems:
3226 you must wait until the executable is run in order to set breakpoints
3227 in shared library routines that are not called directly by the program
3228 (for example, routines that are arguments in a @code{pthread_create}
3232 @cindex data breakpoints
3233 @cindex memory tracing
3234 @cindex breakpoint on memory address
3235 @cindex breakpoint on variable modification
3236 A @dfn{watchpoint} is a special breakpoint that stops your program
3237 when the value of an expression changes. The expression may be a value
3238 of a variable, or it could involve values of one or more variables
3239 combined by operators, such as @samp{a + b}. This is sometimes called
3240 @dfn{data breakpoints}. You must use a different command to set
3241 watchpoints (@pxref{Set Watchpoints, ,Setting Watchpoints}), but aside
3242 from that, you can manage a watchpoint like any other breakpoint: you
3243 enable, disable, and delete both breakpoints and watchpoints using the
3246 You can arrange to have values from your program displayed automatically
3247 whenever @value{GDBN} stops at a breakpoint. @xref{Auto Display,,
3251 @cindex breakpoint on events
3252 A @dfn{catchpoint} is another special breakpoint that stops your program
3253 when a certain kind of event occurs, such as the throwing of a C@t{++}
3254 exception or the loading of a library. As with watchpoints, you use a
3255 different command to set a catchpoint (@pxref{Set Catchpoints, ,Setting
3256 Catchpoints}), but aside from that, you can manage a catchpoint like any
3257 other breakpoint. (To stop when your program receives a signal, use the
3258 @code{handle} command; see @ref{Signals, ,Signals}.)
3260 @cindex breakpoint numbers
3261 @cindex numbers for breakpoints
3262 @value{GDBN} assigns a number to each breakpoint, watchpoint, or
3263 catchpoint when you create it; these numbers are successive integers
3264 starting with one. In many of the commands for controlling various
3265 features of breakpoints you use the breakpoint number to say which
3266 breakpoint you want to change. Each breakpoint may be @dfn{enabled} or
3267 @dfn{disabled}; if disabled, it has no effect on your program until you
3270 @cindex breakpoint ranges
3271 @cindex ranges of breakpoints
3272 Some @value{GDBN} commands accept a range of breakpoints on which to
3273 operate. A breakpoint range is either a single breakpoint number, like
3274 @samp{5}, or two such numbers, in increasing order, separated by a
3275 hyphen, like @samp{5-7}. When a breakpoint range is given to a command,
3276 all breakpoints in that range are operated on.
3279 * Set Breaks:: Setting breakpoints
3280 * Set Watchpoints:: Setting watchpoints
3281 * Set Catchpoints:: Setting catchpoints
3282 * Delete Breaks:: Deleting breakpoints
3283 * Disabling:: Disabling breakpoints
3284 * Conditions:: Break conditions
3285 * Break Commands:: Breakpoint command lists
3286 * Save Breakpoints:: How to save breakpoints in a file
3287 * Error in Breakpoints:: ``Cannot insert breakpoints''
3288 * Breakpoint-related Warnings:: ``Breakpoint address adjusted...''
3292 @subsection Setting Breakpoints
3294 @c FIXME LMB what does GDB do if no code on line of breakpt?
3295 @c consider in particular declaration with/without initialization.
3297 @c FIXME 2 is there stuff on this already? break at fun start, already init?
3300 @kindex b @r{(@code{break})}
3301 @vindex $bpnum@r{, convenience variable}
3302 @cindex latest breakpoint
3303 Breakpoints are set with the @code{break} command (abbreviated
3304 @code{b}). The debugger convenience variable @samp{$bpnum} records the
3305 number of the breakpoint you've set most recently; see @ref{Convenience
3306 Vars,, Convenience Variables}, for a discussion of what you can do with
3307 convenience variables.
3310 @item break @var{location}
3311 Set a breakpoint at the given @var{location}, which can specify a
3312 function name, a line number, or an address of an instruction.
3313 (@xref{Specify Location}, for a list of all the possible ways to
3314 specify a @var{location}.) The breakpoint will stop your program just
3315 before it executes any of the code in the specified @var{location}.
3317 When using source languages that permit overloading of symbols, such as
3318 C@t{++}, a function name may refer to more than one possible place to break.
3319 @xref{Ambiguous Expressions,,Ambiguous Expressions}, for a discussion of
3322 It is also possible to insert a breakpoint that will stop the program
3323 only if a specific thread (@pxref{Thread-Specific Breakpoints})
3324 or a specific task (@pxref{Ada Tasks}) hits that breakpoint.
3327 When called without any arguments, @code{break} sets a breakpoint at
3328 the next instruction to be executed in the selected stack frame
3329 (@pxref{Stack, ,Examining the Stack}). In any selected frame but the
3330 innermost, this makes your program stop as soon as control
3331 returns to that frame. This is similar to the effect of a
3332 @code{finish} command in the frame inside the selected frame---except
3333 that @code{finish} does not leave an active breakpoint. If you use
3334 @code{break} without an argument in the innermost frame, @value{GDBN} stops
3335 the next time it reaches the current location; this may be useful
3338 @value{GDBN} normally ignores breakpoints when it resumes execution, until at
3339 least one instruction has been executed. If it did not do this, you
3340 would be unable to proceed past a breakpoint without first disabling the
3341 breakpoint. This rule applies whether or not the breakpoint already
3342 existed when your program stopped.
3344 @item break @dots{} if @var{cond}
3345 Set a breakpoint with condition @var{cond}; evaluate the expression
3346 @var{cond} each time the breakpoint is reached, and stop only if the
3347 value is nonzero---that is, if @var{cond} evaluates as true.
3348 @samp{@dots{}} stands for one of the possible arguments described
3349 above (or no argument) specifying where to break. @xref{Conditions,
3350 ,Break Conditions}, for more information on breakpoint conditions.
3353 @item tbreak @var{args}
3354 Set a breakpoint enabled only for one stop. @var{args} are the
3355 same as for the @code{break} command, and the breakpoint is set in the same
3356 way, but the breakpoint is automatically deleted after the first time your
3357 program stops there. @xref{Disabling, ,Disabling Breakpoints}.
3360 @cindex hardware breakpoints
3361 @item hbreak @var{args}
3362 Set a hardware-assisted breakpoint. @var{args} are the same as for the
3363 @code{break} command and the breakpoint is set in the same way, but the
3364 breakpoint requires hardware support and some target hardware may not
3365 have this support. The main purpose of this is EPROM/ROM code
3366 debugging, so you can set a breakpoint at an instruction without
3367 changing the instruction. This can be used with the new trap-generation
3368 provided by SPARClite DSU and most x86-based targets. These targets
3369 will generate traps when a program accesses some data or instruction
3370 address that is assigned to the debug registers. However the hardware
3371 breakpoint registers can take a limited number of breakpoints. For
3372 example, on the DSU, only two data breakpoints can be set at a time, and
3373 @value{GDBN} will reject this command if more than two are used. Delete
3374 or disable unused hardware breakpoints before setting new ones
3375 (@pxref{Disabling, ,Disabling Breakpoints}).
3376 @xref{Conditions, ,Break Conditions}.
3377 For remote targets, you can restrict the number of hardware
3378 breakpoints @value{GDBN} will use, see @ref{set remote
3379 hardware-breakpoint-limit}.
3382 @item thbreak @var{args}
3383 Set a hardware-assisted breakpoint enabled only for one stop. @var{args}
3384 are the same as for the @code{hbreak} command and the breakpoint is set in
3385 the same way. However, like the @code{tbreak} command,
3386 the breakpoint is automatically deleted after the
3387 first time your program stops there. Also, like the @code{hbreak}
3388 command, the breakpoint requires hardware support and some target hardware
3389 may not have this support. @xref{Disabling, ,Disabling Breakpoints}.
3390 See also @ref{Conditions, ,Break Conditions}.
3393 @cindex regular expression
3394 @cindex breakpoints at functions matching a regexp
3395 @cindex set breakpoints in many functions
3396 @item rbreak @var{regex}
3397 Set breakpoints on all functions matching the regular expression
3398 @var{regex}. This command sets an unconditional breakpoint on all
3399 matches, printing a list of all breakpoints it set. Once these
3400 breakpoints are set, they are treated just like the breakpoints set with
3401 the @code{break} command. You can delete them, disable them, or make
3402 them conditional the same way as any other breakpoint.
3404 The syntax of the regular expression is the standard one used with tools
3405 like @file{grep}. Note that this is different from the syntax used by
3406 shells, so for instance @code{foo*} matches all functions that include
3407 an @code{fo} followed by zero or more @code{o}s. There is an implicit
3408 @code{.*} leading and trailing the regular expression you supply, so to
3409 match only functions that begin with @code{foo}, use @code{^foo}.
3411 @cindex non-member C@t{++} functions, set breakpoint in
3412 When debugging C@t{++} programs, @code{rbreak} is useful for setting
3413 breakpoints on overloaded functions that are not members of any special
3416 @cindex set breakpoints on all functions
3417 The @code{rbreak} command can be used to set breakpoints in
3418 @strong{all} the functions in a program, like this:
3421 (@value{GDBP}) rbreak .
3424 @item rbreak @var{file}:@var{regex}
3425 If @code{rbreak} is called with a filename qualification, it limits
3426 the search for functions matching the given regular expression to the
3427 specified @var{file}. This can be used, for example, to set breakpoints on
3428 every function in a given file:
3431 (@value{GDBP}) rbreak file.c:.
3434 The colon separating the filename qualifier from the regex may
3435 optionally be surrounded by spaces.
3437 @kindex info breakpoints
3438 @cindex @code{$_} and @code{info breakpoints}
3439 @item info breakpoints @r{[}@var{n}@dots{}@r{]}
3440 @itemx info break @r{[}@var{n}@dots{}@r{]}
3441 Print a table of all breakpoints, watchpoints, and catchpoints set and
3442 not deleted. Optional argument @var{n} means print information only
3443 about the specified breakpoint(s) (or watchpoint(s) or catchpoint(s)).
3444 For each breakpoint, following columns are printed:
3447 @item Breakpoint Numbers
3449 Breakpoint, watchpoint, or catchpoint.
3451 Whether the breakpoint is marked to be disabled or deleted when hit.
3452 @item Enabled or Disabled
3453 Enabled breakpoints are marked with @samp{y}. @samp{n} marks breakpoints
3454 that are not enabled.
3456 Where the breakpoint is in your program, as a memory address. For a
3457 pending breakpoint whose address is not yet known, this field will
3458 contain @samp{<PENDING>}. Such breakpoint won't fire until a shared
3459 library that has the symbol or line referred by breakpoint is loaded.
3460 See below for details. A breakpoint with several locations will
3461 have @samp{<MULTIPLE>} in this field---see below for details.
3463 Where the breakpoint is in the source for your program, as a file and
3464 line number. For a pending breakpoint, the original string passed to
3465 the breakpoint command will be listed as it cannot be resolved until
3466 the appropriate shared library is loaded in the future.
3470 If a breakpoint is conditional, @code{info break} shows the condition on
3471 the line following the affected breakpoint; breakpoint commands, if any,
3472 are listed after that. A pending breakpoint is allowed to have a condition
3473 specified for it. The condition is not parsed for validity until a shared
3474 library is loaded that allows the pending breakpoint to resolve to a
3478 @code{info break} with a breakpoint
3479 number @var{n} as argument lists only that breakpoint. The
3480 convenience variable @code{$_} and the default examining-address for
3481 the @code{x} command are set to the address of the last breakpoint
3482 listed (@pxref{Memory, ,Examining Memory}).
3485 @code{info break} displays a count of the number of times the breakpoint
3486 has been hit. This is especially useful in conjunction with the
3487 @code{ignore} command. You can ignore a large number of breakpoint
3488 hits, look at the breakpoint info to see how many times the breakpoint
3489 was hit, and then run again, ignoring one less than that number. This
3490 will get you quickly to the last hit of that breakpoint.
3493 @value{GDBN} allows you to set any number of breakpoints at the same place in
3494 your program. There is nothing silly or meaningless about this. When
3495 the breakpoints are conditional, this is even useful
3496 (@pxref{Conditions, ,Break Conditions}).
3498 @cindex multiple locations, breakpoints
3499 @cindex breakpoints, multiple locations
3500 It is possible that a breakpoint corresponds to several locations
3501 in your program. Examples of this situation are:
3505 For a C@t{++} constructor, the @value{NGCC} compiler generates several
3506 instances of the function body, used in different cases.
3509 For a C@t{++} template function, a given line in the function can
3510 correspond to any number of instantiations.
3513 For an inlined function, a given source line can correspond to
3514 several places where that function is inlined.
3517 In all those cases, @value{GDBN} will insert a breakpoint at all
3518 the relevant locations@footnote{
3519 As of this writing, multiple-location breakpoints work only if there's
3520 line number information for all the locations. This means that they
3521 will generally not work in system libraries, unless you have debug
3522 info with line numbers for them.}.
3524 A breakpoint with multiple locations is displayed in the breakpoint
3525 table using several rows---one header row, followed by one row for
3526 each breakpoint location. The header row has @samp{<MULTIPLE>} in the
3527 address column. The rows for individual locations contain the actual
3528 addresses for locations, and show the functions to which those
3529 locations belong. The number column for a location is of the form
3530 @var{breakpoint-number}.@var{location-number}.
3535 Num Type Disp Enb Address What
3536 1 breakpoint keep y <MULTIPLE>
3538 breakpoint already hit 1 time
3539 1.1 y 0x080486a2 in void foo<int>() at t.cc:8
3540 1.2 y 0x080486ca in void foo<double>() at t.cc:8
3543 Each location can be individually enabled or disabled by passing
3544 @var{breakpoint-number}.@var{location-number} as argument to the
3545 @code{enable} and @code{disable} commands. Note that you cannot
3546 delete the individual locations from the list, you can only delete the
3547 entire list of locations that belong to their parent breakpoint (with
3548 the @kbd{delete @var{num}} command, where @var{num} is the number of
3549 the parent breakpoint, 1 in the above example). Disabling or enabling
3550 the parent breakpoint (@pxref{Disabling}) affects all of the locations
3551 that belong to that breakpoint.
3553 @cindex pending breakpoints
3554 It's quite common to have a breakpoint inside a shared library.
3555 Shared libraries can be loaded and unloaded explicitly,
3556 and possibly repeatedly, as the program is executed. To support
3557 this use case, @value{GDBN} updates breakpoint locations whenever
3558 any shared library is loaded or unloaded. Typically, you would
3559 set a breakpoint in a shared library at the beginning of your
3560 debugging session, when the library is not loaded, and when the
3561 symbols from the library are not available. When you try to set
3562 breakpoint, @value{GDBN} will ask you if you want to set
3563 a so called @dfn{pending breakpoint}---breakpoint whose address
3564 is not yet resolved.
3566 After the program is run, whenever a new shared library is loaded,
3567 @value{GDBN} reevaluates all the breakpoints. When a newly loaded
3568 shared library contains the symbol or line referred to by some
3569 pending breakpoint, that breakpoint is resolved and becomes an
3570 ordinary breakpoint. When a library is unloaded, all breakpoints
3571 that refer to its symbols or source lines become pending again.
3573 This logic works for breakpoints with multiple locations, too. For
3574 example, if you have a breakpoint in a C@t{++} template function, and
3575 a newly loaded shared library has an instantiation of that template,
3576 a new location is added to the list of locations for the breakpoint.
3578 Except for having unresolved address, pending breakpoints do not
3579 differ from regular breakpoints. You can set conditions or commands,
3580 enable and disable them and perform other breakpoint operations.
3582 @value{GDBN} provides some additional commands for controlling what
3583 happens when the @samp{break} command cannot resolve breakpoint
3584 address specification to an address:
3586 @kindex set breakpoint pending
3587 @kindex show breakpoint pending
3589 @item set breakpoint pending auto
3590 This is the default behavior. When @value{GDBN} cannot find the breakpoint
3591 location, it queries you whether a pending breakpoint should be created.
3593 @item set breakpoint pending on
3594 This indicates that an unrecognized breakpoint location should automatically
3595 result in a pending breakpoint being created.
3597 @item set breakpoint pending off
3598 This indicates that pending breakpoints are not to be created. Any
3599 unrecognized breakpoint location results in an error. This setting does
3600 not affect any pending breakpoints previously created.
3602 @item show breakpoint pending
3603 Show the current behavior setting for creating pending breakpoints.
3606 The settings above only affect the @code{break} command and its
3607 variants. Once breakpoint is set, it will be automatically updated
3608 as shared libraries are loaded and unloaded.
3610 @cindex automatic hardware breakpoints
3611 For some targets, @value{GDBN} can automatically decide if hardware or
3612 software breakpoints should be used, depending on whether the
3613 breakpoint address is read-only or read-write. This applies to
3614 breakpoints set with the @code{break} command as well as to internal
3615 breakpoints set by commands like @code{next} and @code{finish}. For
3616 breakpoints set with @code{hbreak}, @value{GDBN} will always use hardware
3619 You can control this automatic behaviour with the following commands::
3621 @kindex set breakpoint auto-hw
3622 @kindex show breakpoint auto-hw
3624 @item set breakpoint auto-hw on
3625 This is the default behavior. When @value{GDBN} sets a breakpoint, it
3626 will try to use the target memory map to decide if software or hardware
3627 breakpoint must be used.
3629 @item set breakpoint auto-hw off
3630 This indicates @value{GDBN} should not automatically select breakpoint
3631 type. If the target provides a memory map, @value{GDBN} will warn when
3632 trying to set software breakpoint at a read-only address.
3635 @value{GDBN} normally implements breakpoints by replacing the program code
3636 at the breakpoint address with a special instruction, which, when
3637 executed, given control to the debugger. By default, the program
3638 code is so modified only when the program is resumed. As soon as
3639 the program stops, @value{GDBN} restores the original instructions. This
3640 behaviour guards against leaving breakpoints inserted in the
3641 target should gdb abrubptly disconnect. However, with slow remote
3642 targets, inserting and removing breakpoint can reduce the performance.
3643 This behavior can be controlled with the following commands::
3645 @kindex set breakpoint always-inserted
3646 @kindex show breakpoint always-inserted
3648 @item set breakpoint always-inserted off
3649 All breakpoints, including newly added by the user, are inserted in
3650 the target only when the target is resumed. All breakpoints are
3651 removed from the target when it stops.
3653 @item set breakpoint always-inserted on
3654 Causes all breakpoints to be inserted in the target at all times. If
3655 the user adds a new breakpoint, or changes an existing breakpoint, the
3656 breakpoints in the target are updated immediately. A breakpoint is
3657 removed from the target only when breakpoint itself is removed.
3659 @cindex non-stop mode, and @code{breakpoint always-inserted}
3660 @item set breakpoint always-inserted auto
3661 This is the default mode. If @value{GDBN} is controlling the inferior
3662 in non-stop mode (@pxref{Non-Stop Mode}), gdb behaves as if
3663 @code{breakpoint always-inserted} mode is on. If @value{GDBN} is
3664 controlling the inferior in all-stop mode, @value{GDBN} behaves as if
3665 @code{breakpoint always-inserted} mode is off.
3668 @cindex negative breakpoint numbers
3669 @cindex internal @value{GDBN} breakpoints
3670 @value{GDBN} itself sometimes sets breakpoints in your program for
3671 special purposes, such as proper handling of @code{longjmp} (in C
3672 programs). These internal breakpoints are assigned negative numbers,
3673 starting with @code{-1}; @samp{info breakpoints} does not display them.
3674 You can see these breakpoints with the @value{GDBN} maintenance command
3675 @samp{maint info breakpoints} (@pxref{maint info breakpoints}).
3678 @node Set Watchpoints
3679 @subsection Setting Watchpoints
3681 @cindex setting watchpoints
3682 You can use a watchpoint to stop execution whenever the value of an
3683 expression changes, without having to predict a particular place where
3684 this may happen. (This is sometimes called a @dfn{data breakpoint}.)
3685 The expression may be as simple as the value of a single variable, or
3686 as complex as many variables combined by operators. Examples include:
3690 A reference to the value of a single variable.
3693 An address cast to an appropriate data type. For example,
3694 @samp{*(int *)0x12345678} will watch a 4-byte region at the specified
3695 address (assuming an @code{int} occupies 4 bytes).
3698 An arbitrarily complex expression, such as @samp{a*b + c/d}. The
3699 expression can use any operators valid in the program's native
3700 language (@pxref{Languages}).
3703 You can set a watchpoint on an expression even if the expression can
3704 not be evaluated yet. For instance, you can set a watchpoint on
3705 @samp{*global_ptr} before @samp{global_ptr} is initialized.
3706 @value{GDBN} will stop when your program sets @samp{global_ptr} and
3707 the expression produces a valid value. If the expression becomes
3708 valid in some other way than changing a variable (e.g.@: if the memory
3709 pointed to by @samp{*global_ptr} becomes readable as the result of a
3710 @code{malloc} call), @value{GDBN} may not stop until the next time
3711 the expression changes.
3713 @cindex software watchpoints
3714 @cindex hardware watchpoints
3715 Depending on your system, watchpoints may be implemented in software or
3716 hardware. @value{GDBN} does software watchpointing by single-stepping your
3717 program and testing the variable's value each time, which is hundreds of
3718 times slower than normal execution. (But this may still be worth it, to
3719 catch errors where you have no clue what part of your program is the
3722 On some systems, such as HP-UX, PowerPC, @sc{gnu}/Linux and most other
3723 x86-based targets, @value{GDBN} includes support for hardware
3724 watchpoints, which do not slow down the running of your program.
3728 @item watch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{threadnum}@r{]} @r{[}mask @var{maskvalue}@r{]}
3729 Set a watchpoint for an expression. @value{GDBN} will break when the
3730 expression @var{expr} is written into by the program and its value
3731 changes. The simplest (and the most popular) use of this command is
3732 to watch the value of a single variable:
3735 (@value{GDBP}) watch foo
3738 If the command includes a @code{@r{[}thread @var{threadnum}@r{]}}
3739 argument, @value{GDBN} breaks only when the thread identified by
3740 @var{threadnum} changes the value of @var{expr}. If any other threads
3741 change the value of @var{expr}, @value{GDBN} will not break. Note
3742 that watchpoints restricted to a single thread in this way only work
3743 with Hardware Watchpoints.
3745 Ordinarily a watchpoint respects the scope of variables in @var{expr}
3746 (see below). The @code{-location} argument tells @value{GDBN} to
3747 instead watch the memory referred to by @var{expr}. In this case,
3748 @value{GDBN} will evaluate @var{expr}, take the address of the result,
3749 and watch the memory at that address. The type of the result is used
3750 to determine the size of the watched memory. If the expression's
3751 result does not have an address, then @value{GDBN} will print an
3754 The @code{@r{[}mask @var{maskvalue}@r{]}} argument allows creation
3755 of masked watchpoints, if the current architecture supports this
3756 feature (e.g., PowerPC Embedded architecture, see @ref{PowerPC
3757 Embedded}.) A @dfn{masked watchpoint} specifies a mask in addition
3758 to an address to watch. The mask specifies that some bits of an address
3759 (the bits which are reset in the mask) should be ignored when matching
3760 the address accessed by the inferior against the watchpoint address.
3761 Thus, a masked watchpoint watches many addresses simultaneously---those
3762 addresses whose unmasked bits are identical to the unmasked bits in the
3763 watchpoint address. The @code{mask} argument implies @code{-location}.
3767 (@value{GDBP}) watch foo mask 0xffff00ff
3768 (@value{GDBP}) watch *0xdeadbeef mask 0xffffff00
3772 @item rwatch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{threadnum}@r{]} @r{[}mask @var{maskvalue}@r{]}
3773 Set a watchpoint that will break when the value of @var{expr} is read
3777 @item awatch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{threadnum}@r{]} @r{[}mask @var{maskvalue}@r{]}
3778 Set a watchpoint that will break when @var{expr} is either read from
3779 or written into by the program.
3781 @kindex info watchpoints @r{[}@var{n}@dots{}@r{]}
3782 @item info watchpoints @r{[}@var{n}@dots{}@r{]}
3783 This command prints a list of watchpoints, using the same format as
3784 @code{info break} (@pxref{Set Breaks}).
3787 If you watch for a change in a numerically entered address you need to
3788 dereference it, as the address itself is just a constant number which will
3789 never change. @value{GDBN} refuses to create a watchpoint that watches
3790 a never-changing value:
3793 (@value{GDBP}) watch 0x600850
3794 Cannot watch constant value 0x600850.
3795 (@value{GDBP}) watch *(int *) 0x600850
3796 Watchpoint 1: *(int *) 6293584
3799 @value{GDBN} sets a @dfn{hardware watchpoint} if possible. Hardware
3800 watchpoints execute very quickly, and the debugger reports a change in
3801 value at the exact instruction where the change occurs. If @value{GDBN}
3802 cannot set a hardware watchpoint, it sets a software watchpoint, which
3803 executes more slowly and reports the change in value at the next
3804 @emph{statement}, not the instruction, after the change occurs.
3806 @cindex use only software watchpoints
3807 You can force @value{GDBN} to use only software watchpoints with the
3808 @kbd{set can-use-hw-watchpoints 0} command. With this variable set to
3809 zero, @value{GDBN} will never try to use hardware watchpoints, even if
3810 the underlying system supports them. (Note that hardware-assisted
3811 watchpoints that were set @emph{before} setting
3812 @code{can-use-hw-watchpoints} to zero will still use the hardware
3813 mechanism of watching expression values.)
3816 @item set can-use-hw-watchpoints
3817 @kindex set can-use-hw-watchpoints
3818 Set whether or not to use hardware watchpoints.
3820 @item show can-use-hw-watchpoints
3821 @kindex show can-use-hw-watchpoints
3822 Show the current mode of using hardware watchpoints.
3825 For remote targets, you can restrict the number of hardware
3826 watchpoints @value{GDBN} will use, see @ref{set remote
3827 hardware-breakpoint-limit}.
3829 When you issue the @code{watch} command, @value{GDBN} reports
3832 Hardware watchpoint @var{num}: @var{expr}
3836 if it was able to set a hardware watchpoint.
3838 Currently, the @code{awatch} and @code{rwatch} commands can only set
3839 hardware watchpoints, because accesses to data that don't change the
3840 value of the watched expression cannot be detected without examining
3841 every instruction as it is being executed, and @value{GDBN} does not do
3842 that currently. If @value{GDBN} finds that it is unable to set a
3843 hardware breakpoint with the @code{awatch} or @code{rwatch} command, it
3844 will print a message like this:
3847 Expression cannot be implemented with read/access watchpoint.
3850 Sometimes, @value{GDBN} cannot set a hardware watchpoint because the
3851 data type of the watched expression is wider than what a hardware
3852 watchpoint on the target machine can handle. For example, some systems
3853 can only watch regions that are up to 4 bytes wide; on such systems you
3854 cannot set hardware watchpoints for an expression that yields a
3855 double-precision floating-point number (which is typically 8 bytes
3856 wide). As a work-around, it might be possible to break the large region
3857 into a series of smaller ones and watch them with separate watchpoints.
3859 If you set too many hardware watchpoints, @value{GDBN} might be unable
3860 to insert all of them when you resume the execution of your program.
3861 Since the precise number of active watchpoints is unknown until such
3862 time as the program is about to be resumed, @value{GDBN} might not be
3863 able to warn you about this when you set the watchpoints, and the
3864 warning will be printed only when the program is resumed:
3867 Hardware watchpoint @var{num}: Could not insert watchpoint
3871 If this happens, delete or disable some of the watchpoints.
3873 Watching complex expressions that reference many variables can also
3874 exhaust the resources available for hardware-assisted watchpoints.
3875 That's because @value{GDBN} needs to watch every variable in the
3876 expression with separately allocated resources.
3878 If you call a function interactively using @code{print} or @code{call},
3879 any watchpoints you have set will be inactive until @value{GDBN} reaches another
3880 kind of breakpoint or the call completes.
3882 @value{GDBN} automatically deletes watchpoints that watch local
3883 (automatic) variables, or expressions that involve such variables, when
3884 they go out of scope, that is, when the execution leaves the block in
3885 which these variables were defined. In particular, when the program
3886 being debugged terminates, @emph{all} local variables go out of scope,
3887 and so only watchpoints that watch global variables remain set. If you
3888 rerun the program, you will need to set all such watchpoints again. One
3889 way of doing that would be to set a code breakpoint at the entry to the
3890 @code{main} function and when it breaks, set all the watchpoints.
3892 @cindex watchpoints and threads
3893 @cindex threads and watchpoints
3894 In multi-threaded programs, watchpoints will detect changes to the
3895 watched expression from every thread.
3898 @emph{Warning:} In multi-threaded programs, software watchpoints
3899 have only limited usefulness. If @value{GDBN} creates a software
3900 watchpoint, it can only watch the value of an expression @emph{in a
3901 single thread}. If you are confident that the expression can only
3902 change due to the current thread's activity (and if you are also
3903 confident that no other thread can become current), then you can use
3904 software watchpoints as usual. However, @value{GDBN} may not notice
3905 when a non-current thread's activity changes the expression. (Hardware
3906 watchpoints, in contrast, watch an expression in all threads.)
3909 @xref{set remote hardware-watchpoint-limit}.
3911 @node Set Catchpoints
3912 @subsection Setting Catchpoints
3913 @cindex catchpoints, setting
3914 @cindex exception handlers
3915 @cindex event handling
3917 You can use @dfn{catchpoints} to cause the debugger to stop for certain
3918 kinds of program events, such as C@t{++} exceptions or the loading of a
3919 shared library. Use the @code{catch} command to set a catchpoint.
3923 @item catch @var{event}
3924 Stop when @var{event} occurs. @var{event} can be any of the following:
3927 @cindex stop on C@t{++} exceptions
3928 The throwing of a C@t{++} exception.
3931 The catching of a C@t{++} exception.
3934 @cindex Ada exception catching
3935 @cindex catch Ada exceptions
3936 An Ada exception being raised. If an exception name is specified
3937 at the end of the command (eg @code{catch exception Program_Error}),
3938 the debugger will stop only when this specific exception is raised.
3939 Otherwise, the debugger stops execution when any Ada exception is raised.
3941 When inserting an exception catchpoint on a user-defined exception whose
3942 name is identical to one of the exceptions defined by the language, the
3943 fully qualified name must be used as the exception name. Otherwise,
3944 @value{GDBN} will assume that it should stop on the pre-defined exception
3945 rather than the user-defined one. For instance, assuming an exception
3946 called @code{Constraint_Error} is defined in package @code{Pck}, then
3947 the command to use to catch such exceptions is @kbd{catch exception
3948 Pck.Constraint_Error}.
3950 @item exception unhandled
3951 An exception that was raised but is not handled by the program.
3954 A failed Ada assertion.
3957 @cindex break on fork/exec
3958 A call to @code{exec}. This is currently only available for HP-UX
3962 @itemx syscall @r{[}@var{name} @r{|} @var{number}@r{]} @dots{}
3963 @cindex break on a system call.
3964 A call to or return from a system call, a.k.a.@: @dfn{syscall}. A
3965 syscall is a mechanism for application programs to request a service
3966 from the operating system (OS) or one of the OS system services.
3967 @value{GDBN} can catch some or all of the syscalls issued by the
3968 debuggee, and show the related information for each syscall. If no
3969 argument is specified, calls to and returns from all system calls
3972 @var{name} can be any system call name that is valid for the
3973 underlying OS. Just what syscalls are valid depends on the OS. On
3974 GNU and Unix systems, you can find the full list of valid syscall
3975 names on @file{/usr/include/asm/unistd.h}.
3977 @c For MS-Windows, the syscall names and the corresponding numbers
3978 @c can be found, e.g., on this URL:
3979 @c http://www.metasploit.com/users/opcode/syscalls.html
3980 @c but we don't support Windows syscalls yet.
3982 Normally, @value{GDBN} knows in advance which syscalls are valid for
3983 each OS, so you can use the @value{GDBN} command-line completion
3984 facilities (@pxref{Completion,, command completion}) to list the
3987 You may also specify the system call numerically. A syscall's
3988 number is the value passed to the OS's syscall dispatcher to
3989 identify the requested service. When you specify the syscall by its
3990 name, @value{GDBN} uses its database of syscalls to convert the name
3991 into the corresponding numeric code, but using the number directly
3992 may be useful if @value{GDBN}'s database does not have the complete
3993 list of syscalls on your system (e.g., because @value{GDBN} lags
3994 behind the OS upgrades).
3996 The example below illustrates how this command works if you don't provide
4000 (@value{GDBP}) catch syscall
4001 Catchpoint 1 (syscall)
4003 Starting program: /tmp/catch-syscall
4005 Catchpoint 1 (call to syscall 'close'), \
4006 0xffffe424 in __kernel_vsyscall ()
4010 Catchpoint 1 (returned from syscall 'close'), \
4011 0xffffe424 in __kernel_vsyscall ()
4015 Here is an example of catching a system call by name:
4018 (@value{GDBP}) catch syscall chroot
4019 Catchpoint 1 (syscall 'chroot' [61])
4021 Starting program: /tmp/catch-syscall
4023 Catchpoint 1 (call to syscall 'chroot'), \
4024 0xffffe424 in __kernel_vsyscall ()
4028 Catchpoint 1 (returned from syscall 'chroot'), \
4029 0xffffe424 in __kernel_vsyscall ()
4033 An example of specifying a system call numerically. In the case
4034 below, the syscall number has a corresponding entry in the XML
4035 file, so @value{GDBN} finds its name and prints it:
4038 (@value{GDBP}) catch syscall 252
4039 Catchpoint 1 (syscall(s) 'exit_group')
4041 Starting program: /tmp/catch-syscall
4043 Catchpoint 1 (call to syscall 'exit_group'), \
4044 0xffffe424 in __kernel_vsyscall ()
4048 Program exited normally.
4052 However, there can be situations when there is no corresponding name
4053 in XML file for that syscall number. In this case, @value{GDBN} prints
4054 a warning message saying that it was not able to find the syscall name,
4055 but the catchpoint will be set anyway. See the example below:
4058 (@value{GDBP}) catch syscall 764
4059 warning: The number '764' does not represent a known syscall.
4060 Catchpoint 2 (syscall 764)
4064 If you configure @value{GDBN} using the @samp{--without-expat} option,
4065 it will not be able to display syscall names. Also, if your
4066 architecture does not have an XML file describing its system calls,
4067 you will not be able to see the syscall names. It is important to
4068 notice that these two features are used for accessing the syscall
4069 name database. In either case, you will see a warning like this:
4072 (@value{GDBP}) catch syscall
4073 warning: Could not open "syscalls/i386-linux.xml"
4074 warning: Could not load the syscall XML file 'syscalls/i386-linux.xml'.
4075 GDB will not be able to display syscall names.
4076 Catchpoint 1 (syscall)
4080 Of course, the file name will change depending on your architecture and system.
4082 Still using the example above, you can also try to catch a syscall by its
4083 number. In this case, you would see something like:
4086 (@value{GDBP}) catch syscall 252
4087 Catchpoint 1 (syscall(s) 252)
4090 Again, in this case @value{GDBN} would not be able to display syscall's names.
4093 A call to @code{fork}. This is currently only available for HP-UX
4097 A call to @code{vfork}. This is currently only available for HP-UX
4102 @item tcatch @var{event}
4103 Set a catchpoint that is enabled only for one stop. The catchpoint is
4104 automatically deleted after the first time the event is caught.
4108 Use the @code{info break} command to list the current catchpoints.
4110 There are currently some limitations to C@t{++} exception handling
4111 (@code{catch throw} and @code{catch catch}) in @value{GDBN}:
4115 If you call a function interactively, @value{GDBN} normally returns
4116 control to you when the function has finished executing. If the call
4117 raises an exception, however, the call may bypass the mechanism that
4118 returns control to you and cause your program either to abort or to
4119 simply continue running until it hits a breakpoint, catches a signal
4120 that @value{GDBN} is listening for, or exits. This is the case even if
4121 you set a catchpoint for the exception; catchpoints on exceptions are
4122 disabled within interactive calls.
4125 You cannot raise an exception interactively.
4128 You cannot install an exception handler interactively.
4131 @cindex raise exceptions
4132 Sometimes @code{catch} is not the best way to debug exception handling:
4133 if you need to know exactly where an exception is raised, it is better to
4134 stop @emph{before} the exception handler is called, since that way you
4135 can see the stack before any unwinding takes place. If you set a
4136 breakpoint in an exception handler instead, it may not be easy to find
4137 out where the exception was raised.
4139 To stop just before an exception handler is called, you need some
4140 knowledge of the implementation. In the case of @sc{gnu} C@t{++}, exceptions are
4141 raised by calling a library function named @code{__raise_exception}
4142 which has the following ANSI C interface:
4145 /* @var{addr} is where the exception identifier is stored.
4146 @var{id} is the exception identifier. */
4147 void __raise_exception (void **addr, void *id);
4151 To make the debugger catch all exceptions before any stack
4152 unwinding takes place, set a breakpoint on @code{__raise_exception}
4153 (@pxref{Breakpoints, ,Breakpoints; Watchpoints; and Exceptions}).
4155 With a conditional breakpoint (@pxref{Conditions, ,Break Conditions})
4156 that depends on the value of @var{id}, you can stop your program when
4157 a specific exception is raised. You can use multiple conditional
4158 breakpoints to stop your program when any of a number of exceptions are
4163 @subsection Deleting Breakpoints
4165 @cindex clearing breakpoints, watchpoints, catchpoints
4166 @cindex deleting breakpoints, watchpoints, catchpoints
4167 It is often necessary to eliminate a breakpoint, watchpoint, or
4168 catchpoint once it has done its job and you no longer want your program
4169 to stop there. This is called @dfn{deleting} the breakpoint. A
4170 breakpoint that has been deleted no longer exists; it is forgotten.
4172 With the @code{clear} command you can delete breakpoints according to
4173 where they are in your program. With the @code{delete} command you can
4174 delete individual breakpoints, watchpoints, or catchpoints by specifying
4175 their breakpoint numbers.
4177 It is not necessary to delete a breakpoint to proceed past it. @value{GDBN}
4178 automatically ignores breakpoints on the first instruction to be executed
4179 when you continue execution without changing the execution address.
4184 Delete any breakpoints at the next instruction to be executed in the
4185 selected stack frame (@pxref{Selection, ,Selecting a Frame}). When
4186 the innermost frame is selected, this is a good way to delete a
4187 breakpoint where your program just stopped.
4189 @item clear @var{location}
4190 Delete any breakpoints set at the specified @var{location}.
4191 @xref{Specify Location}, for the various forms of @var{location}; the
4192 most useful ones are listed below:
4195 @item clear @var{function}
4196 @itemx clear @var{filename}:@var{function}
4197 Delete any breakpoints set at entry to the named @var{function}.
4199 @item clear @var{linenum}
4200 @itemx clear @var{filename}:@var{linenum}
4201 Delete any breakpoints set at or within the code of the specified
4202 @var{linenum} of the specified @var{filename}.
4205 @cindex delete breakpoints
4207 @kindex d @r{(@code{delete})}
4208 @item delete @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
4209 Delete the breakpoints, watchpoints, or catchpoints of the breakpoint
4210 ranges specified as arguments. If no argument is specified, delete all
4211 breakpoints (@value{GDBN} asks confirmation, unless you have @code{set
4212 confirm off}). You can abbreviate this command as @code{d}.
4216 @subsection Disabling Breakpoints
4218 @cindex enable/disable a breakpoint
4219 Rather than deleting a breakpoint, watchpoint, or catchpoint, you might
4220 prefer to @dfn{disable} it. This makes the breakpoint inoperative as if
4221 it had been deleted, but remembers the information on the breakpoint so
4222 that you can @dfn{enable} it again later.
4224 You disable and enable breakpoints, watchpoints, and catchpoints with
4225 the @code{enable} and @code{disable} commands, optionally specifying
4226 one or more breakpoint numbers as arguments. Use @code{info break} to
4227 print a list of all breakpoints, watchpoints, and catchpoints if you
4228 do not know which numbers to use.
4230 Disabling and enabling a breakpoint that has multiple locations
4231 affects all of its locations.
4233 A breakpoint, watchpoint, or catchpoint can have any of four different
4234 states of enablement:
4238 Enabled. The breakpoint stops your program. A breakpoint set
4239 with the @code{break} command starts out in this state.
4241 Disabled. The breakpoint has no effect on your program.
4243 Enabled once. The breakpoint stops your program, but then becomes
4246 Enabled for deletion. The breakpoint stops your program, but
4247 immediately after it does so it is deleted permanently. A breakpoint
4248 set with the @code{tbreak} command starts out in this state.
4251 You can use the following commands to enable or disable breakpoints,
4252 watchpoints, and catchpoints:
4256 @kindex dis @r{(@code{disable})}
4257 @item disable @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
4258 Disable the specified breakpoints---or all breakpoints, if none are
4259 listed. A disabled breakpoint has no effect but is not forgotten. All
4260 options such as ignore-counts, conditions and commands are remembered in
4261 case the breakpoint is enabled again later. You may abbreviate
4262 @code{disable} as @code{dis}.
4265 @item enable @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
4266 Enable the specified breakpoints (or all defined breakpoints). They
4267 become effective once again in stopping your program.
4269 @item enable @r{[}breakpoints@r{]} once @var{range}@dots{}
4270 Enable the specified breakpoints temporarily. @value{GDBN} disables any
4271 of these breakpoints immediately after stopping your program.
4273 @item enable @r{[}breakpoints@r{]} delete @var{range}@dots{}
4274 Enable the specified breakpoints to work once, then die. @value{GDBN}
4275 deletes any of these breakpoints as soon as your program stops there.
4276 Breakpoints set by the @code{tbreak} command start out in this state.
4279 @c FIXME: I think the following ``Except for [...] @code{tbreak}'' is
4280 @c confusing: tbreak is also initially enabled.
4281 Except for a breakpoint set with @code{tbreak} (@pxref{Set Breaks,
4282 ,Setting Breakpoints}), breakpoints that you set are initially enabled;
4283 subsequently, they become disabled or enabled only when you use one of
4284 the commands above. (The command @code{until} can set and delete a
4285 breakpoint of its own, but it does not change the state of your other
4286 breakpoints; see @ref{Continuing and Stepping, ,Continuing and
4290 @subsection Break Conditions
4291 @cindex conditional breakpoints
4292 @cindex breakpoint conditions
4294 @c FIXME what is scope of break condition expr? Context where wanted?
4295 @c in particular for a watchpoint?
4296 The simplest sort of breakpoint breaks every time your program reaches a
4297 specified place. You can also specify a @dfn{condition} for a
4298 breakpoint. A condition is just a Boolean expression in your
4299 programming language (@pxref{Expressions, ,Expressions}). A breakpoint with
4300 a condition evaluates the expression each time your program reaches it,
4301 and your program stops only if the condition is @emph{true}.
4303 This is the converse of using assertions for program validation; in that
4304 situation, you want to stop when the assertion is violated---that is,
4305 when the condition is false. In C, if you want to test an assertion expressed
4306 by the condition @var{assert}, you should set the condition
4307 @samp{! @var{assert}} on the appropriate breakpoint.
4309 Conditions are also accepted for watchpoints; you may not need them,
4310 since a watchpoint is inspecting the value of an expression anyhow---but
4311 it might be simpler, say, to just set a watchpoint on a variable name,
4312 and specify a condition that tests whether the new value is an interesting
4315 Break conditions can have side effects, and may even call functions in
4316 your program. This can be useful, for example, to activate functions
4317 that log program progress, or to use your own print functions to
4318 format special data structures. The effects are completely predictable
4319 unless there is another enabled breakpoint at the same address. (In
4320 that case, @value{GDBN} might see the other breakpoint first and stop your
4321 program without checking the condition of this one.) Note that
4322 breakpoint commands are usually more convenient and flexible than break
4324 purpose of performing side effects when a breakpoint is reached
4325 (@pxref{Break Commands, ,Breakpoint Command Lists}).
4327 Break conditions can be specified when a breakpoint is set, by using
4328 @samp{if} in the arguments to the @code{break} command. @xref{Set
4329 Breaks, ,Setting Breakpoints}. They can also be changed at any time
4330 with the @code{condition} command.
4332 You can also use the @code{if} keyword with the @code{watch} command.
4333 The @code{catch} command does not recognize the @code{if} keyword;
4334 @code{condition} is the only way to impose a further condition on a
4339 @item condition @var{bnum} @var{expression}
4340 Specify @var{expression} as the break condition for breakpoint,
4341 watchpoint, or catchpoint number @var{bnum}. After you set a condition,
4342 breakpoint @var{bnum} stops your program only if the value of
4343 @var{expression} is true (nonzero, in C). When you use
4344 @code{condition}, @value{GDBN} checks @var{expression} immediately for
4345 syntactic correctness, and to determine whether symbols in it have
4346 referents in the context of your breakpoint. If @var{expression} uses
4347 symbols not referenced in the context of the breakpoint, @value{GDBN}
4348 prints an error message:
4351 No symbol "foo" in current context.
4356 not actually evaluate @var{expression} at the time the @code{condition}
4357 command (or a command that sets a breakpoint with a condition, like
4358 @code{break if @dots{}}) is given, however. @xref{Expressions, ,Expressions}.
4360 @item condition @var{bnum}
4361 Remove the condition from breakpoint number @var{bnum}. It becomes
4362 an ordinary unconditional breakpoint.
4365 @cindex ignore count (of breakpoint)
4366 A special case of a breakpoint condition is to stop only when the
4367 breakpoint has been reached a certain number of times. This is so
4368 useful that there is a special way to do it, using the @dfn{ignore
4369 count} of the breakpoint. Every breakpoint has an ignore count, which
4370 is an integer. Most of the time, the ignore count is zero, and
4371 therefore has no effect. But if your program reaches a breakpoint whose
4372 ignore count is positive, then instead of stopping, it just decrements
4373 the ignore count by one and continues. As a result, if the ignore count
4374 value is @var{n}, the breakpoint does not stop the next @var{n} times
4375 your program reaches it.
4379 @item ignore @var{bnum} @var{count}
4380 Set the ignore count of breakpoint number @var{bnum} to @var{count}.
4381 The next @var{count} times the breakpoint is reached, your program's
4382 execution does not stop; other than to decrement the ignore count, @value{GDBN}
4385 To make the breakpoint stop the next time it is reached, specify
4388 When you use @code{continue} to resume execution of your program from a
4389 breakpoint, you can specify an ignore count directly as an argument to
4390 @code{continue}, rather than using @code{ignore}. @xref{Continuing and
4391 Stepping,,Continuing and Stepping}.
4393 If a breakpoint has a positive ignore count and a condition, the
4394 condition is not checked. Once the ignore count reaches zero,
4395 @value{GDBN} resumes checking the condition.
4397 You could achieve the effect of the ignore count with a condition such
4398 as @w{@samp{$foo-- <= 0}} using a debugger convenience variable that
4399 is decremented each time. @xref{Convenience Vars, ,Convenience
4403 Ignore counts apply to breakpoints, watchpoints, and catchpoints.
4406 @node Break Commands
4407 @subsection Breakpoint Command Lists
4409 @cindex breakpoint commands
4410 You can give any breakpoint (or watchpoint or catchpoint) a series of
4411 commands to execute when your program stops due to that breakpoint. For
4412 example, you might want to print the values of certain expressions, or
4413 enable other breakpoints.
4417 @kindex end@r{ (breakpoint commands)}
4418 @item commands @r{[}@var{range}@dots{}@r{]}
4419 @itemx @dots{} @var{command-list} @dots{}
4421 Specify a list of commands for the given breakpoints. The commands
4422 themselves appear on the following lines. Type a line containing just
4423 @code{end} to terminate the commands.
4425 To remove all commands from a breakpoint, type @code{commands} and
4426 follow it immediately with @code{end}; that is, give no commands.
4428 With no argument, @code{commands} refers to the last breakpoint,
4429 watchpoint, or catchpoint set (not to the breakpoint most recently
4430 encountered). If the most recent breakpoints were set with a single
4431 command, then the @code{commands} will apply to all the breakpoints
4432 set by that command. This applies to breakpoints set by
4433 @code{rbreak}, and also applies when a single @code{break} command
4434 creates multiple breakpoints (@pxref{Ambiguous Expressions,,Ambiguous
4438 Pressing @key{RET} as a means of repeating the last @value{GDBN} command is
4439 disabled within a @var{command-list}.
4441 You can use breakpoint commands to start your program up again. Simply
4442 use the @code{continue} command, or @code{step}, or any other command
4443 that resumes execution.
4445 Any other commands in the command list, after a command that resumes
4446 execution, are ignored. This is because any time you resume execution
4447 (even with a simple @code{next} or @code{step}), you may encounter
4448 another breakpoint---which could have its own command list, leading to
4449 ambiguities about which list to execute.
4452 If the first command you specify in a command list is @code{silent}, the
4453 usual message about stopping at a breakpoint is not printed. This may
4454 be desirable for breakpoints that are to print a specific message and
4455 then continue. If none of the remaining commands print anything, you
4456 see no sign that the breakpoint was reached. @code{silent} is
4457 meaningful only at the beginning of a breakpoint command list.
4459 The commands @code{echo}, @code{output}, and @code{printf} allow you to
4460 print precisely controlled output, and are often useful in silent
4461 breakpoints. @xref{Output, ,Commands for Controlled Output}.
4463 For example, here is how you could use breakpoint commands to print the
4464 value of @code{x} at entry to @code{foo} whenever @code{x} is positive.
4470 printf "x is %d\n",x
4475 One application for breakpoint commands is to compensate for one bug so
4476 you can test for another. Put a breakpoint just after the erroneous line
4477 of code, give it a condition to detect the case in which something
4478 erroneous has been done, and give it commands to assign correct values
4479 to any variables that need them. End with the @code{continue} command
4480 so that your program does not stop, and start with the @code{silent}
4481 command so that no output is produced. Here is an example:
4492 @node Save Breakpoints
4493 @subsection How to save breakpoints to a file
4495 To save breakpoint definitions to a file use the @w{@code{save
4496 breakpoints}} command.
4499 @kindex save breakpoints
4500 @cindex save breakpoints to a file for future sessions
4501 @item save breakpoints [@var{filename}]
4502 This command saves all current breakpoint definitions together with
4503 their commands and ignore counts, into a file @file{@var{filename}}
4504 suitable for use in a later debugging session. This includes all
4505 types of breakpoints (breakpoints, watchpoints, catchpoints,
4506 tracepoints). To read the saved breakpoint definitions, use the
4507 @code{source} command (@pxref{Command Files}). Note that watchpoints
4508 with expressions involving local variables may fail to be recreated
4509 because it may not be possible to access the context where the
4510 watchpoint is valid anymore. Because the saved breakpoint definitions
4511 are simply a sequence of @value{GDBN} commands that recreate the
4512 breakpoints, you can edit the file in your favorite editing program,
4513 and remove the breakpoint definitions you're not interested in, or
4514 that can no longer be recreated.
4517 @c @ifclear BARETARGET
4518 @node Error in Breakpoints
4519 @subsection ``Cannot insert breakpoints''
4521 If you request too many active hardware-assisted breakpoints and
4522 watchpoints, you will see this error message:
4524 @c FIXME: the precise wording of this message may change; the relevant
4525 @c source change is not committed yet (Sep 3, 1999).
4527 Stopped; cannot insert breakpoints.
4528 You may have requested too many hardware breakpoints and watchpoints.
4532 This message is printed when you attempt to resume the program, since
4533 only then @value{GDBN} knows exactly how many hardware breakpoints and
4534 watchpoints it needs to insert.
4536 When this message is printed, you need to disable or remove some of the
4537 hardware-assisted breakpoints and watchpoints, and then continue.
4539 @node Breakpoint-related Warnings
4540 @subsection ``Breakpoint address adjusted...''
4541 @cindex breakpoint address adjusted
4543 Some processor architectures place constraints on the addresses at
4544 which breakpoints may be placed. For architectures thus constrained,
4545 @value{GDBN} will attempt to adjust the breakpoint's address to comply
4546 with the constraints dictated by the architecture.
4548 One example of such an architecture is the Fujitsu FR-V. The FR-V is
4549 a VLIW architecture in which a number of RISC-like instructions may be
4550 bundled together for parallel execution. The FR-V architecture
4551 constrains the location of a breakpoint instruction within such a
4552 bundle to the instruction with the lowest address. @value{GDBN}
4553 honors this constraint by adjusting a breakpoint's address to the
4554 first in the bundle.
4556 It is not uncommon for optimized code to have bundles which contain
4557 instructions from different source statements, thus it may happen that
4558 a breakpoint's address will be adjusted from one source statement to
4559 another. Since this adjustment may significantly alter @value{GDBN}'s
4560 breakpoint related behavior from what the user expects, a warning is
4561 printed when the breakpoint is first set and also when the breakpoint
4564 A warning like the one below is printed when setting a breakpoint
4565 that's been subject to address adjustment:
4568 warning: Breakpoint address adjusted from 0x00010414 to 0x00010410.
4571 Such warnings are printed both for user settable and @value{GDBN}'s
4572 internal breakpoints. If you see one of these warnings, you should
4573 verify that a breakpoint set at the adjusted address will have the
4574 desired affect. If not, the breakpoint in question may be removed and
4575 other breakpoints may be set which will have the desired behavior.
4576 E.g., it may be sufficient to place the breakpoint at a later
4577 instruction. A conditional breakpoint may also be useful in some
4578 cases to prevent the breakpoint from triggering too often.
4580 @value{GDBN} will also issue a warning when stopping at one of these
4581 adjusted breakpoints:
4584 warning: Breakpoint 1 address previously adjusted from 0x00010414
4588 When this warning is encountered, it may be too late to take remedial
4589 action except in cases where the breakpoint is hit earlier or more
4590 frequently than expected.
4592 @node Continuing and Stepping
4593 @section Continuing and Stepping
4597 @cindex resuming execution
4598 @dfn{Continuing} means resuming program execution until your program
4599 completes normally. In contrast, @dfn{stepping} means executing just
4600 one more ``step'' of your program, where ``step'' may mean either one
4601 line of source code, or one machine instruction (depending on what
4602 particular command you use). Either when continuing or when stepping,
4603 your program may stop even sooner, due to a breakpoint or a signal. (If
4604 it stops due to a signal, you may want to use @code{handle}, or use
4605 @samp{signal 0} to resume execution. @xref{Signals, ,Signals}.)
4609 @kindex c @r{(@code{continue})}
4610 @kindex fg @r{(resume foreground execution)}
4611 @item continue @r{[}@var{ignore-count}@r{]}
4612 @itemx c @r{[}@var{ignore-count}@r{]}
4613 @itemx fg @r{[}@var{ignore-count}@r{]}
4614 Resume program execution, at the address where your program last stopped;
4615 any breakpoints set at that address are bypassed. The optional argument
4616 @var{ignore-count} allows you to specify a further number of times to
4617 ignore a breakpoint at this location; its effect is like that of
4618 @code{ignore} (@pxref{Conditions, ,Break Conditions}).
4620 The argument @var{ignore-count} is meaningful only when your program
4621 stopped due to a breakpoint. At other times, the argument to
4622 @code{continue} is ignored.
4624 The synonyms @code{c} and @code{fg} (for @dfn{foreground}, as the
4625 debugged program is deemed to be the foreground program) are provided
4626 purely for convenience, and have exactly the same behavior as
4630 To resume execution at a different place, you can use @code{return}
4631 (@pxref{Returning, ,Returning from a Function}) to go back to the
4632 calling function; or @code{jump} (@pxref{Jumping, ,Continuing at a
4633 Different Address}) to go to an arbitrary location in your program.
4635 A typical technique for using stepping is to set a breakpoint
4636 (@pxref{Breakpoints, ,Breakpoints; Watchpoints; and Catchpoints}) at the
4637 beginning of the function or the section of your program where a problem
4638 is believed to lie, run your program until it stops at that breakpoint,
4639 and then step through the suspect area, examining the variables that are
4640 interesting, until you see the problem happen.
4644 @kindex s @r{(@code{step})}
4646 Continue running your program until control reaches a different source
4647 line, then stop it and return control to @value{GDBN}. This command is
4648 abbreviated @code{s}.
4651 @c "without debugging information" is imprecise; actually "without line
4652 @c numbers in the debugging information". (gcc -g1 has debugging info but
4653 @c not line numbers). But it seems complex to try to make that
4654 @c distinction here.
4655 @emph{Warning:} If you use the @code{step} command while control is
4656 within a function that was compiled without debugging information,
4657 execution proceeds until control reaches a function that does have
4658 debugging information. Likewise, it will not step into a function which
4659 is compiled without debugging information. To step through functions
4660 without debugging information, use the @code{stepi} command, described
4664 The @code{step} command only stops at the first instruction of a source
4665 line. This prevents the multiple stops that could otherwise occur in
4666 @code{switch} statements, @code{for} loops, etc. @code{step} continues
4667 to stop if a function that has debugging information is called within
4668 the line. In other words, @code{step} @emph{steps inside} any functions
4669 called within the line.
4671 Also, the @code{step} command only enters a function if there is line
4672 number information for the function. Otherwise it acts like the
4673 @code{next} command. This avoids problems when using @code{cc -gl}
4674 on MIPS machines. Previously, @code{step} entered subroutines if there
4675 was any debugging information about the routine.
4677 @item step @var{count}
4678 Continue running as in @code{step}, but do so @var{count} times. If a
4679 breakpoint is reached, or a signal not related to stepping occurs before
4680 @var{count} steps, stepping stops right away.
4683 @kindex n @r{(@code{next})}
4684 @item next @r{[}@var{count}@r{]}
4685 Continue to the next source line in the current (innermost) stack frame.
4686 This is similar to @code{step}, but function calls that appear within
4687 the line of code are executed without stopping. Execution stops when
4688 control reaches a different line of code at the original stack level
4689 that was executing when you gave the @code{next} command. This command
4690 is abbreviated @code{n}.
4692 An argument @var{count} is a repeat count, as for @code{step}.
4695 @c FIX ME!! Do we delete this, or is there a way it fits in with
4696 @c the following paragraph? --- Vctoria
4698 @c @code{next} within a function that lacks debugging information acts like
4699 @c @code{step}, but any function calls appearing within the code of the
4700 @c function are executed without stopping.
4702 The @code{next} command only stops at the first instruction of a
4703 source line. This prevents multiple stops that could otherwise occur in
4704 @code{switch} statements, @code{for} loops, etc.
4706 @kindex set step-mode
4708 @cindex functions without line info, and stepping
4709 @cindex stepping into functions with no line info
4710 @itemx set step-mode on
4711 The @code{set step-mode on} command causes the @code{step} command to
4712 stop at the first instruction of a function which contains no debug line
4713 information rather than stepping over it.
4715 This is useful in cases where you may be interested in inspecting the
4716 machine instructions of a function which has no symbolic info and do not
4717 want @value{GDBN} to automatically skip over this function.
4719 @item set step-mode off
4720 Causes the @code{step} command to step over any functions which contains no
4721 debug information. This is the default.
4723 @item show step-mode
4724 Show whether @value{GDBN} will stop in or step over functions without
4725 source line debug information.
4728 @kindex fin @r{(@code{finish})}
4730 Continue running until just after function in the selected stack frame
4731 returns. Print the returned value (if any). This command can be
4732 abbreviated as @code{fin}.
4734 Contrast this with the @code{return} command (@pxref{Returning,
4735 ,Returning from a Function}).
4738 @kindex u @r{(@code{until})}
4739 @cindex run until specified location
4742 Continue running until a source line past the current line, in the
4743 current stack frame, is reached. This command is used to avoid single
4744 stepping through a loop more than once. It is like the @code{next}
4745 command, except that when @code{until} encounters a jump, it
4746 automatically continues execution until the program counter is greater
4747 than the address of the jump.
4749 This means that when you reach the end of a loop after single stepping
4750 though it, @code{until} makes your program continue execution until it
4751 exits the loop. In contrast, a @code{next} command at the end of a loop
4752 simply steps back to the beginning of the loop, which forces you to step
4753 through the next iteration.
4755 @code{until} always stops your program if it attempts to exit the current
4758 @code{until} may produce somewhat counterintuitive results if the order
4759 of machine code does not match the order of the source lines. For
4760 example, in the following excerpt from a debugging session, the @code{f}
4761 (@code{frame}) command shows that execution is stopped at line
4762 @code{206}; yet when we use @code{until}, we get to line @code{195}:
4766 #0 main (argc=4, argv=0xf7fffae8) at m4.c:206
4768 (@value{GDBP}) until
4769 195 for ( ; argc > 0; NEXTARG) @{
4772 This happened because, for execution efficiency, the compiler had
4773 generated code for the loop closure test at the end, rather than the
4774 start, of the loop---even though the test in a C @code{for}-loop is
4775 written before the body of the loop. The @code{until} command appeared
4776 to step back to the beginning of the loop when it advanced to this
4777 expression; however, it has not really gone to an earlier
4778 statement---not in terms of the actual machine code.
4780 @code{until} with no argument works by means of single
4781 instruction stepping, and hence is slower than @code{until} with an
4784 @item until @var{location}
4785 @itemx u @var{location}
4786 Continue running your program until either the specified location is
4787 reached, or the current stack frame returns. @var{location} is any of
4788 the forms described in @ref{Specify Location}.
4789 This form of the command uses temporary breakpoints, and
4790 hence is quicker than @code{until} without an argument. The specified
4791 location is actually reached only if it is in the current frame. This
4792 implies that @code{until} can be used to skip over recursive function
4793 invocations. For instance in the code below, if the current location is
4794 line @code{96}, issuing @code{until 99} will execute the program up to
4795 line @code{99} in the same invocation of factorial, i.e., after the inner
4796 invocations have returned.
4799 94 int factorial (int value)
4801 96 if (value > 1) @{
4802 97 value *= factorial (value - 1);
4809 @kindex advance @var{location}
4810 @itemx advance @var{location}
4811 Continue running the program up to the given @var{location}. An argument is
4812 required, which should be of one of the forms described in
4813 @ref{Specify Location}.
4814 Execution will also stop upon exit from the current stack
4815 frame. This command is similar to @code{until}, but @code{advance} will
4816 not skip over recursive function calls, and the target location doesn't
4817 have to be in the same frame as the current one.
4821 @kindex si @r{(@code{stepi})}
4823 @itemx stepi @var{arg}
4825 Execute one machine instruction, then stop and return to the debugger.
4827 It is often useful to do @samp{display/i $pc} when stepping by machine
4828 instructions. This makes @value{GDBN} automatically display the next
4829 instruction to be executed, each time your program stops. @xref{Auto
4830 Display,, Automatic Display}.
4832 An argument is a repeat count, as in @code{step}.
4836 @kindex ni @r{(@code{nexti})}
4838 @itemx nexti @var{arg}
4840 Execute one machine instruction, but if it is a function call,
4841 proceed until the function returns.
4843 An argument is a repeat count, as in @code{next}.
4850 A signal is an asynchronous event that can happen in a program. The
4851 operating system defines the possible kinds of signals, and gives each
4852 kind a name and a number. For example, in Unix @code{SIGINT} is the
4853 signal a program gets when you type an interrupt character (often @kbd{Ctrl-c});
4854 @code{SIGSEGV} is the signal a program gets from referencing a place in
4855 memory far away from all the areas in use; @code{SIGALRM} occurs when
4856 the alarm clock timer goes off (which happens only if your program has
4857 requested an alarm).
4859 @cindex fatal signals
4860 Some signals, including @code{SIGALRM}, are a normal part of the
4861 functioning of your program. Others, such as @code{SIGSEGV}, indicate
4862 errors; these signals are @dfn{fatal} (they kill your program immediately) if the
4863 program has not specified in advance some other way to handle the signal.
4864 @code{SIGINT} does not indicate an error in your program, but it is normally
4865 fatal so it can carry out the purpose of the interrupt: to kill the program.
4867 @value{GDBN} has the ability to detect any occurrence of a signal in your
4868 program. You can tell @value{GDBN} in advance what to do for each kind of
4871 @cindex handling signals
4872 Normally, @value{GDBN} is set up to let the non-erroneous signals like
4873 @code{SIGALRM} be silently passed to your program
4874 (so as not to interfere with their role in the program's functioning)
4875 but to stop your program immediately whenever an error signal happens.
4876 You can change these settings with the @code{handle} command.
4879 @kindex info signals
4883 Print a table of all the kinds of signals and how @value{GDBN} has been told to
4884 handle each one. You can use this to see the signal numbers of all
4885 the defined types of signals.
4887 @item info signals @var{sig}
4888 Similar, but print information only about the specified signal number.
4890 @code{info handle} is an alias for @code{info signals}.
4893 @item handle @var{signal} @r{[}@var{keywords}@dots{}@r{]}
4894 Change the way @value{GDBN} handles signal @var{signal}. @var{signal}
4895 can be the number of a signal or its name (with or without the
4896 @samp{SIG} at the beginning); a list of signal numbers of the form
4897 @samp{@var{low}-@var{high}}; or the word @samp{all}, meaning all the
4898 known signals. Optional arguments @var{keywords}, described below,
4899 say what change to make.
4903 The keywords allowed by the @code{handle} command can be abbreviated.
4904 Their full names are:
4908 @value{GDBN} should not stop your program when this signal happens. It may
4909 still print a message telling you that the signal has come in.
4912 @value{GDBN} should stop your program when this signal happens. This implies
4913 the @code{print} keyword as well.
4916 @value{GDBN} should print a message when this signal happens.
4919 @value{GDBN} should not mention the occurrence of the signal at all. This
4920 implies the @code{nostop} keyword as well.
4924 @value{GDBN} should allow your program to see this signal; your program
4925 can handle the signal, or else it may terminate if the signal is fatal
4926 and not handled. @code{pass} and @code{noignore} are synonyms.
4930 @value{GDBN} should not allow your program to see this signal.
4931 @code{nopass} and @code{ignore} are synonyms.
4935 When a signal stops your program, the signal is not visible to the
4937 continue. Your program sees the signal then, if @code{pass} is in
4938 effect for the signal in question @emph{at that time}. In other words,
4939 after @value{GDBN} reports a signal, you can use the @code{handle}
4940 command with @code{pass} or @code{nopass} to control whether your
4941 program sees that signal when you continue.
4943 The default is set to @code{nostop}, @code{noprint}, @code{pass} for
4944 non-erroneous signals such as @code{SIGALRM}, @code{SIGWINCH} and
4945 @code{SIGCHLD}, and to @code{stop}, @code{print}, @code{pass} for the
4948 You can also use the @code{signal} command to prevent your program from
4949 seeing a signal, or cause it to see a signal it normally would not see,
4950 or to give it any signal at any time. For example, if your program stopped
4951 due to some sort of memory reference error, you might store correct
4952 values into the erroneous variables and continue, hoping to see more
4953 execution; but your program would probably terminate immediately as
4954 a result of the fatal signal once it saw the signal. To prevent this,
4955 you can continue with @samp{signal 0}. @xref{Signaling, ,Giving your
4958 @cindex extra signal information
4959 @anchor{extra signal information}
4961 On some targets, @value{GDBN} can inspect extra signal information
4962 associated with the intercepted signal, before it is actually
4963 delivered to the program being debugged. This information is exported
4964 by the convenience variable @code{$_siginfo}, and consists of data
4965 that is passed by the kernel to the signal handler at the time of the
4966 receipt of a signal. The data type of the information itself is
4967 target dependent. You can see the data type using the @code{ptype
4968 $_siginfo} command. On Unix systems, it typically corresponds to the
4969 standard @code{siginfo_t} type, as defined in the @file{signal.h}
4972 Here's an example, on a @sc{gnu}/Linux system, printing the stray
4973 referenced address that raised a segmentation fault.
4977 (@value{GDBP}) continue
4978 Program received signal SIGSEGV, Segmentation fault.
4979 0x0000000000400766 in main ()
4981 (@value{GDBP}) ptype $_siginfo
4988 struct @{...@} _kill;
4989 struct @{...@} _timer;
4991 struct @{...@} _sigchld;
4992 struct @{...@} _sigfault;
4993 struct @{...@} _sigpoll;
4996 (@value{GDBP}) ptype $_siginfo._sifields._sigfault
5000 (@value{GDBP}) p $_siginfo._sifields._sigfault.si_addr
5001 $1 = (void *) 0x7ffff7ff7000
5005 Depending on target support, @code{$_siginfo} may also be writable.
5008 @section Stopping and Starting Multi-thread Programs
5010 @cindex stopped threads
5011 @cindex threads, stopped
5013 @cindex continuing threads
5014 @cindex threads, continuing
5016 @value{GDBN} supports debugging programs with multiple threads
5017 (@pxref{Threads,, Debugging Programs with Multiple Threads}). There
5018 are two modes of controlling execution of your program within the
5019 debugger. In the default mode, referred to as @dfn{all-stop mode},
5020 when any thread in your program stops (for example, at a breakpoint
5021 or while being stepped), all other threads in the program are also stopped by
5022 @value{GDBN}. On some targets, @value{GDBN} also supports
5023 @dfn{non-stop mode}, in which other threads can continue to run freely while
5024 you examine the stopped thread in the debugger.
5027 * All-Stop Mode:: All threads stop when GDB takes control
5028 * Non-Stop Mode:: Other threads continue to execute
5029 * Background Execution:: Running your program asynchronously
5030 * Thread-Specific Breakpoints:: Controlling breakpoints
5031 * Interrupted System Calls:: GDB may interfere with system calls
5032 * Observer Mode:: GDB does not alter program behavior
5036 @subsection All-Stop Mode
5038 @cindex all-stop mode
5040 In all-stop mode, whenever your program stops under @value{GDBN} for any reason,
5041 @emph{all} threads of execution stop, not just the current thread. This
5042 allows you to examine the overall state of the program, including
5043 switching between threads, without worrying that things may change
5046 Conversely, whenever you restart the program, @emph{all} threads start
5047 executing. @emph{This is true even when single-stepping} with commands
5048 like @code{step} or @code{next}.
5050 In particular, @value{GDBN} cannot single-step all threads in lockstep.
5051 Since thread scheduling is up to your debugging target's operating
5052 system (not controlled by @value{GDBN}), other threads may
5053 execute more than one statement while the current thread completes a
5054 single step. Moreover, in general other threads stop in the middle of a
5055 statement, rather than at a clean statement boundary, when the program
5058 You might even find your program stopped in another thread after
5059 continuing or even single-stepping. This happens whenever some other
5060 thread runs into a breakpoint, a signal, or an exception before the
5061 first thread completes whatever you requested.
5063 @cindex automatic thread selection
5064 @cindex switching threads automatically
5065 @cindex threads, automatic switching
5066 Whenever @value{GDBN} stops your program, due to a breakpoint or a
5067 signal, it automatically selects the thread where that breakpoint or
5068 signal happened. @value{GDBN} alerts you to the context switch with a
5069 message such as @samp{[Switching to Thread @var{n}]} to identify the
5072 On some OSes, you can modify @value{GDBN}'s default behavior by
5073 locking the OS scheduler to allow only a single thread to run.
5076 @item set scheduler-locking @var{mode}
5077 @cindex scheduler locking mode
5078 @cindex lock scheduler
5079 Set the scheduler locking mode. If it is @code{off}, then there is no
5080 locking and any thread may run at any time. If @code{on}, then only the
5081 current thread may run when the inferior is resumed. The @code{step}
5082 mode optimizes for single-stepping; it prevents other threads
5083 from preempting the current thread while you are stepping, so that
5084 the focus of debugging does not change unexpectedly.
5085 Other threads only rarely (or never) get a chance to run
5086 when you step. They are more likely to run when you @samp{next} over a
5087 function call, and they are completely free to run when you use commands
5088 like @samp{continue}, @samp{until}, or @samp{finish}. However, unless another
5089 thread hits a breakpoint during its timeslice, @value{GDBN} does not change
5090 the current thread away from the thread that you are debugging.
5092 @item show scheduler-locking
5093 Display the current scheduler locking mode.
5096 @cindex resume threads of multiple processes simultaneously
5097 By default, when you issue one of the execution commands such as
5098 @code{continue}, @code{next} or @code{step}, @value{GDBN} allows only
5099 threads of the current inferior to run. For example, if @value{GDBN}
5100 is attached to two inferiors, each with two threads, the
5101 @code{continue} command resumes only the two threads of the current
5102 inferior. This is useful, for example, when you debug a program that
5103 forks and you want to hold the parent stopped (so that, for instance,
5104 it doesn't run to exit), while you debug the child. In other
5105 situations, you may not be interested in inspecting the current state
5106 of any of the processes @value{GDBN} is attached to, and you may want
5107 to resume them all until some breakpoint is hit. In the latter case,
5108 you can instruct @value{GDBN} to allow all threads of all the
5109 inferiors to run with the @w{@code{set schedule-multiple}} command.
5112 @kindex set schedule-multiple
5113 @item set schedule-multiple
5114 Set the mode for allowing threads of multiple processes to be resumed
5115 when an execution command is issued. When @code{on}, all threads of
5116 all processes are allowed to run. When @code{off}, only the threads
5117 of the current process are resumed. The default is @code{off}. The
5118 @code{scheduler-locking} mode takes precedence when set to @code{on},
5119 or while you are stepping and set to @code{step}.
5121 @item show schedule-multiple
5122 Display the current mode for resuming the execution of threads of
5127 @subsection Non-Stop Mode
5129 @cindex non-stop mode
5131 @c This section is really only a place-holder, and needs to be expanded
5132 @c with more details.
5134 For some multi-threaded targets, @value{GDBN} supports an optional
5135 mode of operation in which you can examine stopped program threads in
5136 the debugger while other threads continue to execute freely. This
5137 minimizes intrusion when debugging live systems, such as programs
5138 where some threads have real-time constraints or must continue to
5139 respond to external events. This is referred to as @dfn{non-stop} mode.
5141 In non-stop mode, when a thread stops to report a debugging event,
5142 @emph{only} that thread is stopped; @value{GDBN} does not stop other
5143 threads as well, in contrast to the all-stop mode behavior. Additionally,
5144 execution commands such as @code{continue} and @code{step} apply by default
5145 only to the current thread in non-stop mode, rather than all threads as
5146 in all-stop mode. This allows you to control threads explicitly in
5147 ways that are not possible in all-stop mode --- for example, stepping
5148 one thread while allowing others to run freely, stepping
5149 one thread while holding all others stopped, or stepping several threads
5150 independently and simultaneously.
5152 To enter non-stop mode, use this sequence of commands before you run
5153 or attach to your program:
5156 # Enable the async interface.
5159 # If using the CLI, pagination breaks non-stop.
5162 # Finally, turn it on!
5166 You can use these commands to manipulate the non-stop mode setting:
5169 @kindex set non-stop
5170 @item set non-stop on
5171 Enable selection of non-stop mode.
5172 @item set non-stop off
5173 Disable selection of non-stop mode.
5174 @kindex show non-stop
5176 Show the current non-stop enablement setting.
5179 Note these commands only reflect whether non-stop mode is enabled,
5180 not whether the currently-executing program is being run in non-stop mode.
5181 In particular, the @code{set non-stop} preference is only consulted when
5182 @value{GDBN} starts or connects to the target program, and it is generally
5183 not possible to switch modes once debugging has started. Furthermore,
5184 since not all targets support non-stop mode, even when you have enabled
5185 non-stop mode, @value{GDBN} may still fall back to all-stop operation by
5188 In non-stop mode, all execution commands apply only to the current thread
5189 by default. That is, @code{continue} only continues one thread.
5190 To continue all threads, issue @code{continue -a} or @code{c -a}.
5192 You can use @value{GDBN}'s background execution commands
5193 (@pxref{Background Execution}) to run some threads in the background
5194 while you continue to examine or step others from @value{GDBN}.
5195 The MI execution commands (@pxref{GDB/MI Program Execution}) are
5196 always executed asynchronously in non-stop mode.
5198 Suspending execution is done with the @code{interrupt} command when
5199 running in the background, or @kbd{Ctrl-c} during foreground execution.
5200 In all-stop mode, this stops the whole process;
5201 but in non-stop mode the interrupt applies only to the current thread.
5202 To stop the whole program, use @code{interrupt -a}.
5204 Other execution commands do not currently support the @code{-a} option.
5206 In non-stop mode, when a thread stops, @value{GDBN} doesn't automatically make
5207 that thread current, as it does in all-stop mode. This is because the
5208 thread stop notifications are asynchronous with respect to @value{GDBN}'s
5209 command interpreter, and it would be confusing if @value{GDBN} unexpectedly
5210 changed to a different thread just as you entered a command to operate on the
5211 previously current thread.
5213 @node Background Execution
5214 @subsection Background Execution
5216 @cindex foreground execution
5217 @cindex background execution
5218 @cindex asynchronous execution
5219 @cindex execution, foreground, background and asynchronous
5221 @value{GDBN}'s execution commands have two variants: the normal
5222 foreground (synchronous) behavior, and a background
5223 (asynchronous) behavior. In foreground execution, @value{GDBN} waits for
5224 the program to report that some thread has stopped before prompting for
5225 another command. In background execution, @value{GDBN} immediately gives
5226 a command prompt so that you can issue other commands while your program runs.
5228 You need to explicitly enable asynchronous mode before you can use
5229 background execution commands. You can use these commands to
5230 manipulate the asynchronous mode setting:
5233 @kindex set target-async
5234 @item set target-async on
5235 Enable asynchronous mode.
5236 @item set target-async off
5237 Disable asynchronous mode.
5238 @kindex show target-async
5239 @item show target-async
5240 Show the current target-async setting.
5243 If the target doesn't support async mode, @value{GDBN} issues an error
5244 message if you attempt to use the background execution commands.
5246 To specify background execution, add a @code{&} to the command. For example,
5247 the background form of the @code{continue} command is @code{continue&}, or
5248 just @code{c&}. The execution commands that accept background execution
5254 @xref{Starting, , Starting your Program}.
5258 @xref{Attach, , Debugging an Already-running Process}.
5262 @xref{Continuing and Stepping, step}.
5266 @xref{Continuing and Stepping, stepi}.
5270 @xref{Continuing and Stepping, next}.
5274 @xref{Continuing and Stepping, nexti}.
5278 @xref{Continuing and Stepping, continue}.
5282 @xref{Continuing and Stepping, finish}.
5286 @xref{Continuing and Stepping, until}.
5290 Background execution is especially useful in conjunction with non-stop
5291 mode for debugging programs with multiple threads; see @ref{Non-Stop Mode}.
5292 However, you can also use these commands in the normal all-stop mode with
5293 the restriction that you cannot issue another execution command until the
5294 previous one finishes. Examples of commands that are valid in all-stop
5295 mode while the program is running include @code{help} and @code{info break}.
5297 You can interrupt your program while it is running in the background by
5298 using the @code{interrupt} command.
5305 Suspend execution of the running program. In all-stop mode,
5306 @code{interrupt} stops the whole process, but in non-stop mode, it stops
5307 only the current thread. To stop the whole program in non-stop mode,
5308 use @code{interrupt -a}.
5311 @node Thread-Specific Breakpoints
5312 @subsection Thread-Specific Breakpoints
5314 When your program has multiple threads (@pxref{Threads,, Debugging
5315 Programs with Multiple Threads}), you can choose whether to set
5316 breakpoints on all threads, or on a particular thread.
5319 @cindex breakpoints and threads
5320 @cindex thread breakpoints
5321 @kindex break @dots{} thread @var{threadno}
5322 @item break @var{linespec} thread @var{threadno}
5323 @itemx break @var{linespec} thread @var{threadno} if @dots{}
5324 @var{linespec} specifies source lines; there are several ways of
5325 writing them (@pxref{Specify Location}), but the effect is always to
5326 specify some source line.
5328 Use the qualifier @samp{thread @var{threadno}} with a breakpoint command
5329 to specify that you only want @value{GDBN} to stop the program when a
5330 particular thread reaches this breakpoint. @var{threadno} is one of the
5331 numeric thread identifiers assigned by @value{GDBN}, shown in the first
5332 column of the @samp{info threads} display.
5334 If you do not specify @samp{thread @var{threadno}} when you set a
5335 breakpoint, the breakpoint applies to @emph{all} threads of your
5338 You can use the @code{thread} qualifier on conditional breakpoints as
5339 well; in this case, place @samp{thread @var{threadno}} before or
5340 after the breakpoint condition, like this:
5343 (@value{GDBP}) break frik.c:13 thread 28 if bartab > lim
5348 @node Interrupted System Calls
5349 @subsection Interrupted System Calls
5351 @cindex thread breakpoints and system calls
5352 @cindex system calls and thread breakpoints
5353 @cindex premature return from system calls
5354 There is an unfortunate side effect when using @value{GDBN} to debug
5355 multi-threaded programs. If one thread stops for a
5356 breakpoint, or for some other reason, and another thread is blocked in a
5357 system call, then the system call may return prematurely. This is a
5358 consequence of the interaction between multiple threads and the signals
5359 that @value{GDBN} uses to implement breakpoints and other events that
5362 To handle this problem, your program should check the return value of
5363 each system call and react appropriately. This is good programming
5366 For example, do not write code like this:
5372 The call to @code{sleep} will return early if a different thread stops
5373 at a breakpoint or for some other reason.
5375 Instead, write this:
5380 unslept = sleep (unslept);
5383 A system call is allowed to return early, so the system is still
5384 conforming to its specification. But @value{GDBN} does cause your
5385 multi-threaded program to behave differently than it would without
5388 Also, @value{GDBN} uses internal breakpoints in the thread library to
5389 monitor certain events such as thread creation and thread destruction.
5390 When such an event happens, a system call in another thread may return
5391 prematurely, even though your program does not appear to stop.
5394 @subsection Observer Mode
5396 If you want to build on non-stop mode and observe program behavior
5397 without any chance of disruption by @value{GDBN}, you can set
5398 variables to disable all of the debugger's attempts to modify state,
5399 whether by writing memory, inserting breakpoints, etc. These operate
5400 at a low level, intercepting operations from all commands.
5402 When all of these are set to @code{off}, then @value{GDBN} is said to
5403 be @dfn{observer mode}. As a convenience, the variable
5404 @code{observer} can be set to disable these, plus enable non-stop
5407 Note that @value{GDBN} will not prevent you from making nonsensical
5408 combinations of these settings. For instance, if you have enabled
5409 @code{may-insert-breakpoints} but disabled @code{may-write-memory},
5410 then breakpoints that work by writing trap instructions into the code
5411 stream will still not be able to be placed.
5416 @item set observer on
5417 @itemx set observer off
5418 When set to @code{on}, this disables all the permission variables
5419 below (except for @code{insert-fast-tracepoints}), plus enables
5420 non-stop debugging. Setting this to @code{off} switches back to
5421 normal debugging, though remaining in non-stop mode.
5424 Show whether observer mode is on or off.
5426 @kindex may-write-registers
5427 @item set may-write-registers on
5428 @itemx set may-write-registers off
5429 This controls whether @value{GDBN} will attempt to alter the values of
5430 registers, such as with assignment expressions in @code{print}, or the
5431 @code{jump} command. It defaults to @code{on}.
5433 @item show may-write-registers
5434 Show the current permission to write registers.
5436 @kindex may-write-memory
5437 @item set may-write-memory on
5438 @itemx set may-write-memory off
5439 This controls whether @value{GDBN} will attempt to alter the contents
5440 of memory, such as with assignment expressions in @code{print}. It
5441 defaults to @code{on}.
5443 @item show may-write-memory
5444 Show the current permission to write memory.
5446 @kindex may-insert-breakpoints
5447 @item set may-insert-breakpoints on
5448 @itemx set may-insert-breakpoints off
5449 This controls whether @value{GDBN} will attempt to insert breakpoints.
5450 This affects all breakpoints, including internal breakpoints defined
5451 by @value{GDBN}. It defaults to @code{on}.
5453 @item show may-insert-breakpoints
5454 Show the current permission to insert breakpoints.
5456 @kindex may-insert-tracepoints
5457 @item set may-insert-tracepoints on
5458 @itemx set may-insert-tracepoints off
5459 This controls whether @value{GDBN} will attempt to insert (regular)
5460 tracepoints at the beginning of a tracing experiment. It affects only
5461 non-fast tracepoints, fast tracepoints being under the control of
5462 @code{may-insert-fast-tracepoints}. It defaults to @code{on}.
5464 @item show may-insert-tracepoints
5465 Show the current permission to insert tracepoints.
5467 @kindex may-insert-fast-tracepoints
5468 @item set may-insert-fast-tracepoints on
5469 @itemx set may-insert-fast-tracepoints off
5470 This controls whether @value{GDBN} will attempt to insert fast
5471 tracepoints at the beginning of a tracing experiment. It affects only
5472 fast tracepoints, regular (non-fast) tracepoints being under the
5473 control of @code{may-insert-tracepoints}. It defaults to @code{on}.
5475 @item show may-insert-fast-tracepoints
5476 Show the current permission to insert fast tracepoints.
5478 @kindex may-interrupt
5479 @item set may-interrupt on
5480 @itemx set may-interrupt off
5481 This controls whether @value{GDBN} will attempt to interrupt or stop
5482 program execution. When this variable is @code{off}, the
5483 @code{interrupt} command will have no effect, nor will
5484 @kbd{Ctrl-c}. It defaults to @code{on}.
5486 @item show may-interrupt
5487 Show the current permission to interrupt or stop the program.
5491 @node Reverse Execution
5492 @chapter Running programs backward
5493 @cindex reverse execution
5494 @cindex running programs backward
5496 When you are debugging a program, it is not unusual to realize that
5497 you have gone too far, and some event of interest has already happened.
5498 If the target environment supports it, @value{GDBN} can allow you to
5499 ``rewind'' the program by running it backward.
5501 A target environment that supports reverse execution should be able
5502 to ``undo'' the changes in machine state that have taken place as the
5503 program was executing normally. Variables, registers etc.@: should
5504 revert to their previous values. Obviously this requires a great
5505 deal of sophistication on the part of the target environment; not
5506 all target environments can support reverse execution.
5508 When a program is executed in reverse, the instructions that
5509 have most recently been executed are ``un-executed'', in reverse
5510 order. The program counter runs backward, following the previous
5511 thread of execution in reverse. As each instruction is ``un-executed'',
5512 the values of memory and/or registers that were changed by that
5513 instruction are reverted to their previous states. After executing
5514 a piece of source code in reverse, all side effects of that code
5515 should be ``undone'', and all variables should be returned to their
5516 prior values@footnote{
5517 Note that some side effects are easier to undo than others. For instance,
5518 memory and registers are relatively easy, but device I/O is hard. Some
5519 targets may be able undo things like device I/O, and some may not.
5521 The contract between @value{GDBN} and the reverse executing target
5522 requires only that the target do something reasonable when
5523 @value{GDBN} tells it to execute backwards, and then report the
5524 results back to @value{GDBN}. Whatever the target reports back to
5525 @value{GDBN}, @value{GDBN} will report back to the user. @value{GDBN}
5526 assumes that the memory and registers that the target reports are in a
5527 consistant state, but @value{GDBN} accepts whatever it is given.
5530 If you are debugging in a target environment that supports
5531 reverse execution, @value{GDBN} provides the following commands.
5534 @kindex reverse-continue
5535 @kindex rc @r{(@code{reverse-continue})}
5536 @item reverse-continue @r{[}@var{ignore-count}@r{]}
5537 @itemx rc @r{[}@var{ignore-count}@r{]}
5538 Beginning at the point where your program last stopped, start executing
5539 in reverse. Reverse execution will stop for breakpoints and synchronous
5540 exceptions (signals), just like normal execution. Behavior of
5541 asynchronous signals depends on the target environment.
5543 @kindex reverse-step
5544 @kindex rs @r{(@code{step})}
5545 @item reverse-step @r{[}@var{count}@r{]}
5546 Run the program backward until control reaches the start of a
5547 different source line; then stop it, and return control to @value{GDBN}.
5549 Like the @code{step} command, @code{reverse-step} will only stop
5550 at the beginning of a source line. It ``un-executes'' the previously
5551 executed source line. If the previous source line included calls to
5552 debuggable functions, @code{reverse-step} will step (backward) into
5553 the called function, stopping at the beginning of the @emph{last}
5554 statement in the called function (typically a return statement).
5556 Also, as with the @code{step} command, if non-debuggable functions are
5557 called, @code{reverse-step} will run thru them backward without stopping.
5559 @kindex reverse-stepi
5560 @kindex rsi @r{(@code{reverse-stepi})}
5561 @item reverse-stepi @r{[}@var{count}@r{]}
5562 Reverse-execute one machine instruction. Note that the instruction
5563 to be reverse-executed is @emph{not} the one pointed to by the program
5564 counter, but the instruction executed prior to that one. For instance,
5565 if the last instruction was a jump, @code{reverse-stepi} will take you
5566 back from the destination of the jump to the jump instruction itself.
5568 @kindex reverse-next
5569 @kindex rn @r{(@code{reverse-next})}
5570 @item reverse-next @r{[}@var{count}@r{]}
5571 Run backward to the beginning of the previous line executed in
5572 the current (innermost) stack frame. If the line contains function
5573 calls, they will be ``un-executed'' without stopping. Starting from
5574 the first line of a function, @code{reverse-next} will take you back
5575 to the caller of that function, @emph{before} the function was called,
5576 just as the normal @code{next} command would take you from the last
5577 line of a function back to its return to its caller
5578 @footnote{Unless the code is too heavily optimized.}.
5580 @kindex reverse-nexti
5581 @kindex rni @r{(@code{reverse-nexti})}
5582 @item reverse-nexti @r{[}@var{count}@r{]}
5583 Like @code{nexti}, @code{reverse-nexti} executes a single instruction
5584 in reverse, except that called functions are ``un-executed'' atomically.
5585 That is, if the previously executed instruction was a return from
5586 another function, @code{reverse-nexti} will continue to execute
5587 in reverse until the call to that function (from the current stack
5590 @kindex reverse-finish
5591 @item reverse-finish
5592 Just as the @code{finish} command takes you to the point where the
5593 current function returns, @code{reverse-finish} takes you to the point
5594 where it was called. Instead of ending up at the end of the current
5595 function invocation, you end up at the beginning.
5597 @kindex set exec-direction
5598 @item set exec-direction
5599 Set the direction of target execution.
5600 @itemx set exec-direction reverse
5601 @cindex execute forward or backward in time
5602 @value{GDBN} will perform all execution commands in reverse, until the
5603 exec-direction mode is changed to ``forward''. Affected commands include
5604 @code{step, stepi, next, nexti, continue, and finish}. The @code{return}
5605 command cannot be used in reverse mode.
5606 @item set exec-direction forward
5607 @value{GDBN} will perform all execution commands in the normal fashion.
5608 This is the default.
5612 @node Process Record and Replay
5613 @chapter Recording Inferior's Execution and Replaying It
5614 @cindex process record and replay
5615 @cindex recording inferior's execution and replaying it
5617 On some platforms, @value{GDBN} provides a special @dfn{process record
5618 and replay} target that can record a log of the process execution, and
5619 replay it later with both forward and reverse execution commands.
5622 When this target is in use, if the execution log includes the record
5623 for the next instruction, @value{GDBN} will debug in @dfn{replay
5624 mode}. In the replay mode, the inferior does not really execute code
5625 instructions. Instead, all the events that normally happen during
5626 code execution are taken from the execution log. While code is not
5627 really executed in replay mode, the values of registers (including the
5628 program counter register) and the memory of the inferior are still
5629 changed as they normally would. Their contents are taken from the
5633 If the record for the next instruction is not in the execution log,
5634 @value{GDBN} will debug in @dfn{record mode}. In this mode, the
5635 inferior executes normally, and @value{GDBN} records the execution log
5638 The process record and replay target supports reverse execution
5639 (@pxref{Reverse Execution}), even if the platform on which the
5640 inferior runs does not. However, the reverse execution is limited in
5641 this case by the range of the instructions recorded in the execution
5642 log. In other words, reverse execution on platforms that don't
5643 support it directly can only be done in the replay mode.
5645 When debugging in the reverse direction, @value{GDBN} will work in
5646 replay mode as long as the execution log includes the record for the
5647 previous instruction; otherwise, it will work in record mode, if the
5648 platform supports reverse execution, or stop if not.
5650 For architecture environments that support process record and replay,
5651 @value{GDBN} provides the following commands:
5654 @kindex target record
5658 This command starts the process record and replay target. The process
5659 record and replay target can only debug a process that is already
5660 running. Therefore, you need first to start the process with the
5661 @kbd{run} or @kbd{start} commands, and then start the recording with
5662 the @kbd{target record} command.
5664 Both @code{record} and @code{rec} are aliases of @code{target record}.
5666 @cindex displaced stepping, and process record and replay
5667 Displaced stepping (@pxref{Maintenance Commands,, displaced stepping})
5668 will be automatically disabled when process record and replay target
5669 is started. That's because the process record and replay target
5670 doesn't support displaced stepping.
5672 @cindex non-stop mode, and process record and replay
5673 @cindex asynchronous execution, and process record and replay
5674 If the inferior is in the non-stop mode (@pxref{Non-Stop Mode}) or in
5675 the asynchronous execution mode (@pxref{Background Execution}), the
5676 process record and replay target cannot be started because it doesn't
5677 support these two modes.
5682 Stop the process record and replay target. When process record and
5683 replay target stops, the entire execution log will be deleted and the
5684 inferior will either be terminated, or will remain in its final state.
5686 When you stop the process record and replay target in record mode (at
5687 the end of the execution log), the inferior will be stopped at the
5688 next instruction that would have been recorded. In other words, if
5689 you record for a while and then stop recording, the inferior process
5690 will be left in the same state as if the recording never happened.
5692 On the other hand, if the process record and replay target is stopped
5693 while in replay mode (that is, not at the end of the execution log,
5694 but at some earlier point), the inferior process will become ``live''
5695 at that earlier state, and it will then be possible to continue the
5696 usual ``live'' debugging of the process from that state.
5698 When the inferior process exits, or @value{GDBN} detaches from it,
5699 process record and replay target will automatically stop itself.
5702 @item record save @var{filename}
5703 Save the execution log to a file @file{@var{filename}}.
5704 Default filename is @file{gdb_record.@var{process_id}}, where
5705 @var{process_id} is the process ID of the inferior.
5707 @kindex record restore
5708 @item record restore @var{filename}
5709 Restore the execution log from a file @file{@var{filename}}.
5710 File must have been created with @code{record save}.
5712 @kindex set record insn-number-max
5713 @item set record insn-number-max @var{limit}
5714 Set the limit of instructions to be recorded. Default value is 200000.
5716 If @var{limit} is a positive number, then @value{GDBN} will start
5717 deleting instructions from the log once the number of the record
5718 instructions becomes greater than @var{limit}. For every new recorded
5719 instruction, @value{GDBN} will delete the earliest recorded
5720 instruction to keep the number of recorded instructions at the limit.
5721 (Since deleting recorded instructions loses information, @value{GDBN}
5722 lets you control what happens when the limit is reached, by means of
5723 the @code{stop-at-limit} option, described below.)
5725 If @var{limit} is zero, @value{GDBN} will never delete recorded
5726 instructions from the execution log. The number of recorded
5727 instructions is unlimited in this case.
5729 @kindex show record insn-number-max
5730 @item show record insn-number-max
5731 Show the limit of instructions to be recorded.
5733 @kindex set record stop-at-limit
5734 @item set record stop-at-limit
5735 Control the behavior when the number of recorded instructions reaches
5736 the limit. If ON (the default), @value{GDBN} will stop when the limit
5737 is reached for the first time and ask you whether you want to stop the
5738 inferior or continue running it and recording the execution log. If
5739 you decide to continue recording, each new recorded instruction will
5740 cause the oldest one to be deleted.
5742 If this option is OFF, @value{GDBN} will automatically delete the
5743 oldest record to make room for each new one, without asking.
5745 @kindex show record stop-at-limit
5746 @item show record stop-at-limit
5747 Show the current setting of @code{stop-at-limit}.
5749 @kindex set record memory-query
5750 @item set record memory-query
5751 Control the behavior when @value{GDBN} is unable to record memory
5752 changes caused by an instruction. If ON, @value{GDBN} will query
5753 whether to stop the inferior in that case.
5755 If this option is OFF (the default), @value{GDBN} will automatically
5756 ignore the effect of such instructions on memory. Later, when
5757 @value{GDBN} replays this execution log, it will mark the log of this
5758 instruction as not accessible, and it will not affect the replay
5761 @kindex show record memory-query
5762 @item show record memory-query
5763 Show the current setting of @code{memory-query}.
5767 Show various statistics about the state of process record and its
5768 in-memory execution log buffer, including:
5772 Whether in record mode or replay mode.
5774 Lowest recorded instruction number (counting from when the current execution log started recording instructions).
5776 Highest recorded instruction number.
5778 Current instruction about to be replayed (if in replay mode).
5780 Number of instructions contained in the execution log.
5782 Maximum number of instructions that may be contained in the execution log.
5785 @kindex record delete
5788 When record target runs in replay mode (``in the past''), delete the
5789 subsequent execution log and begin to record a new execution log starting
5790 from the current address. This means you will abandon the previously
5791 recorded ``future'' and begin recording a new ``future''.
5796 @chapter Examining the Stack
5798 When your program has stopped, the first thing you need to know is where it
5799 stopped and how it got there.
5802 Each time your program performs a function call, information about the call
5804 That information includes the location of the call in your program,
5805 the arguments of the call,
5806 and the local variables of the function being called.
5807 The information is saved in a block of data called a @dfn{stack frame}.
5808 The stack frames are allocated in a region of memory called the @dfn{call
5811 When your program stops, the @value{GDBN} commands for examining the
5812 stack allow you to see all of this information.
5814 @cindex selected frame
5815 One of the stack frames is @dfn{selected} by @value{GDBN} and many
5816 @value{GDBN} commands refer implicitly to the selected frame. In
5817 particular, whenever you ask @value{GDBN} for the value of a variable in
5818 your program, the value is found in the selected frame. There are
5819 special @value{GDBN} commands to select whichever frame you are
5820 interested in. @xref{Selection, ,Selecting a Frame}.
5822 When your program stops, @value{GDBN} automatically selects the
5823 currently executing frame and describes it briefly, similar to the
5824 @code{frame} command (@pxref{Frame Info, ,Information about a Frame}).
5827 * Frames:: Stack frames
5828 * Backtrace:: Backtraces
5829 * Selection:: Selecting a frame
5830 * Frame Info:: Information on a frame
5835 @section Stack Frames
5837 @cindex frame, definition
5839 The call stack is divided up into contiguous pieces called @dfn{stack
5840 frames}, or @dfn{frames} for short; each frame is the data associated
5841 with one call to one function. The frame contains the arguments given
5842 to the function, the function's local variables, and the address at
5843 which the function is executing.
5845 @cindex initial frame
5846 @cindex outermost frame
5847 @cindex innermost frame
5848 When your program is started, the stack has only one frame, that of the
5849 function @code{main}. This is called the @dfn{initial} frame or the
5850 @dfn{outermost} frame. Each time a function is called, a new frame is
5851 made. Each time a function returns, the frame for that function invocation
5852 is eliminated. If a function is recursive, there can be many frames for
5853 the same function. The frame for the function in which execution is
5854 actually occurring is called the @dfn{innermost} frame. This is the most
5855 recently created of all the stack frames that still exist.
5857 @cindex frame pointer
5858 Inside your program, stack frames are identified by their addresses. A
5859 stack frame consists of many bytes, each of which has its own address; each
5860 kind of computer has a convention for choosing one byte whose
5861 address serves as the address of the frame. Usually this address is kept
5862 in a register called the @dfn{frame pointer register}
5863 (@pxref{Registers, $fp}) while execution is going on in that frame.
5865 @cindex frame number
5866 @value{GDBN} assigns numbers to all existing stack frames, starting with
5867 zero for the innermost frame, one for the frame that called it,
5868 and so on upward. These numbers do not really exist in your program;
5869 they are assigned by @value{GDBN} to give you a way of designating stack
5870 frames in @value{GDBN} commands.
5872 @c The -fomit-frame-pointer below perennially causes hbox overflow
5873 @c underflow problems.
5874 @cindex frameless execution
5875 Some compilers provide a way to compile functions so that they operate
5876 without stack frames. (For example, the @value{NGCC} option
5878 @samp{-fomit-frame-pointer}
5880 generates functions without a frame.)
5881 This is occasionally done with heavily used library functions to save
5882 the frame setup time. @value{GDBN} has limited facilities for dealing
5883 with these function invocations. If the innermost function invocation
5884 has no stack frame, @value{GDBN} nevertheless regards it as though
5885 it had a separate frame, which is numbered zero as usual, allowing
5886 correct tracing of the function call chain. However, @value{GDBN} has
5887 no provision for frameless functions elsewhere in the stack.
5890 @kindex frame@r{, command}
5891 @cindex current stack frame
5892 @item frame @var{args}
5893 The @code{frame} command allows you to move from one stack frame to another,
5894 and to print the stack frame you select. @var{args} may be either the
5895 address of the frame or the stack frame number. Without an argument,
5896 @code{frame} prints the current stack frame.
5898 @kindex select-frame
5899 @cindex selecting frame silently
5901 The @code{select-frame} command allows you to move from one stack frame
5902 to another without printing the frame. This is the silent version of
5910 @cindex call stack traces
5911 A backtrace is a summary of how your program got where it is. It shows one
5912 line per frame, for many frames, starting with the currently executing
5913 frame (frame zero), followed by its caller (frame one), and on up the
5918 @kindex bt @r{(@code{backtrace})}
5921 Print a backtrace of the entire stack: one line per frame for all
5922 frames in the stack.
5924 You can stop the backtrace at any time by typing the system interrupt
5925 character, normally @kbd{Ctrl-c}.
5927 @item backtrace @var{n}
5929 Similar, but print only the innermost @var{n} frames.
5931 @item backtrace -@var{n}
5933 Similar, but print only the outermost @var{n} frames.
5935 @item backtrace full
5937 @itemx bt full @var{n}
5938 @itemx bt full -@var{n}
5939 Print the values of the local variables also. @var{n} specifies the
5940 number of frames to print, as described above.
5945 The names @code{where} and @code{info stack} (abbreviated @code{info s})
5946 are additional aliases for @code{backtrace}.
5948 @cindex multiple threads, backtrace
5949 In a multi-threaded program, @value{GDBN} by default shows the
5950 backtrace only for the current thread. To display the backtrace for
5951 several or all of the threads, use the command @code{thread apply}
5952 (@pxref{Threads, thread apply}). For example, if you type @kbd{thread
5953 apply all backtrace}, @value{GDBN} will display the backtrace for all
5954 the threads; this is handy when you debug a core dump of a
5955 multi-threaded program.
5957 Each line in the backtrace shows the frame number and the function name.
5958 The program counter value is also shown---unless you use @code{set
5959 print address off}. The backtrace also shows the source file name and
5960 line number, as well as the arguments to the function. The program
5961 counter value is omitted if it is at the beginning of the code for that
5964 Here is an example of a backtrace. It was made with the command
5965 @samp{bt 3}, so it shows the innermost three frames.
5969 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
5971 #1 0x6e38 in expand_macro (sym=0x2b600, data=...) at macro.c:242
5972 #2 0x6840 in expand_token (obs=0x0, t=177664, td=0xf7fffb08)
5974 (More stack frames follow...)
5979 The display for frame zero does not begin with a program counter
5980 value, indicating that your program has stopped at the beginning of the
5981 code for line @code{993} of @code{builtin.c}.
5984 The value of parameter @code{data} in frame 1 has been replaced by
5985 @code{@dots{}}. By default, @value{GDBN} prints the value of a parameter
5986 only if it is a scalar (integer, pointer, enumeration, etc). See command
5987 @kbd{set print frame-arguments} in @ref{Print Settings} for more details
5988 on how to configure the way function parameter values are printed.
5990 @cindex optimized out, in backtrace
5991 @cindex function call arguments, optimized out
5992 If your program was compiled with optimizations, some compilers will
5993 optimize away arguments passed to functions if those arguments are
5994 never used after the call. Such optimizations generate code that
5995 passes arguments through registers, but doesn't store those arguments
5996 in the stack frame. @value{GDBN} has no way of displaying such
5997 arguments in stack frames other than the innermost one. Here's what
5998 such a backtrace might look like:
6002 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
6004 #1 0x6e38 in expand_macro (sym=<optimized out>) at macro.c:242
6005 #2 0x6840 in expand_token (obs=0x0, t=<optimized out>, td=0xf7fffb08)
6007 (More stack frames follow...)
6012 The values of arguments that were not saved in their stack frames are
6013 shown as @samp{<optimized out>}.
6015 If you need to display the values of such optimized-out arguments,
6016 either deduce that from other variables whose values depend on the one
6017 you are interested in, or recompile without optimizations.
6019 @cindex backtrace beyond @code{main} function
6020 @cindex program entry point
6021 @cindex startup code, and backtrace
6022 Most programs have a standard user entry point---a place where system
6023 libraries and startup code transition into user code. For C this is
6024 @code{main}@footnote{
6025 Note that embedded programs (the so-called ``free-standing''
6026 environment) are not required to have a @code{main} function as the
6027 entry point. They could even have multiple entry points.}.
6028 When @value{GDBN} finds the entry function in a backtrace
6029 it will terminate the backtrace, to avoid tracing into highly
6030 system-specific (and generally uninteresting) code.
6032 If you need to examine the startup code, or limit the number of levels
6033 in a backtrace, you can change this behavior:
6036 @item set backtrace past-main
6037 @itemx set backtrace past-main on
6038 @kindex set backtrace
6039 Backtraces will continue past the user entry point.
6041 @item set backtrace past-main off
6042 Backtraces will stop when they encounter the user entry point. This is the
6045 @item show backtrace past-main
6046 @kindex show backtrace
6047 Display the current user entry point backtrace policy.
6049 @item set backtrace past-entry
6050 @itemx set backtrace past-entry on
6051 Backtraces will continue past the internal entry point of an application.
6052 This entry point is encoded by the linker when the application is built,
6053 and is likely before the user entry point @code{main} (or equivalent) is called.
6055 @item set backtrace past-entry off
6056 Backtraces will stop when they encounter the internal entry point of an
6057 application. This is the default.
6059 @item show backtrace past-entry
6060 Display the current internal entry point backtrace policy.
6062 @item set backtrace limit @var{n}
6063 @itemx set backtrace limit 0
6064 @cindex backtrace limit
6065 Limit the backtrace to @var{n} levels. A value of zero means
6068 @item show backtrace limit
6069 Display the current limit on backtrace levels.
6073 @section Selecting a Frame
6075 Most commands for examining the stack and other data in your program work on
6076 whichever stack frame is selected at the moment. Here are the commands for
6077 selecting a stack frame; all of them finish by printing a brief description
6078 of the stack frame just selected.
6081 @kindex frame@r{, selecting}
6082 @kindex f @r{(@code{frame})}
6085 Select frame number @var{n}. Recall that frame zero is the innermost
6086 (currently executing) frame, frame one is the frame that called the
6087 innermost one, and so on. The highest-numbered frame is the one for
6090 @item frame @var{addr}
6092 Select the frame at address @var{addr}. This is useful mainly if the
6093 chaining of stack frames has been damaged by a bug, making it
6094 impossible for @value{GDBN} to assign numbers properly to all frames. In
6095 addition, this can be useful when your program has multiple stacks and
6096 switches between them.
6098 On the SPARC architecture, @code{frame} needs two addresses to
6099 select an arbitrary frame: a frame pointer and a stack pointer.
6101 On the MIPS and Alpha architecture, it needs two addresses: a stack
6102 pointer and a program counter.
6104 On the 29k architecture, it needs three addresses: a register stack
6105 pointer, a program counter, and a memory stack pointer.
6109 Move @var{n} frames up the stack. For positive numbers @var{n}, this
6110 advances toward the outermost frame, to higher frame numbers, to frames
6111 that have existed longer. @var{n} defaults to one.
6114 @kindex do @r{(@code{down})}
6116 Move @var{n} frames down the stack. For positive numbers @var{n}, this
6117 advances toward the innermost frame, to lower frame numbers, to frames
6118 that were created more recently. @var{n} defaults to one. You may
6119 abbreviate @code{down} as @code{do}.
6122 All of these commands end by printing two lines of output describing the
6123 frame. The first line shows the frame number, the function name, the
6124 arguments, and the source file and line number of execution in that
6125 frame. The second line shows the text of that source line.
6133 #1 0x22f0 in main (argc=1, argv=0xf7fffbf4, env=0xf7fffbfc)
6135 10 read_input_file (argv[i]);
6139 After such a printout, the @code{list} command with no arguments
6140 prints ten lines centered on the point of execution in the frame.
6141 You can also edit the program at the point of execution with your favorite
6142 editing program by typing @code{edit}.
6143 @xref{List, ,Printing Source Lines},
6147 @kindex down-silently
6149 @item up-silently @var{n}
6150 @itemx down-silently @var{n}
6151 These two commands are variants of @code{up} and @code{down},
6152 respectively; they differ in that they do their work silently, without
6153 causing display of the new frame. They are intended primarily for use
6154 in @value{GDBN} command scripts, where the output might be unnecessary and
6159 @section Information About a Frame
6161 There are several other commands to print information about the selected
6167 When used without any argument, this command does not change which
6168 frame is selected, but prints a brief description of the currently
6169 selected stack frame. It can be abbreviated @code{f}. With an
6170 argument, this command is used to select a stack frame.
6171 @xref{Selection, ,Selecting a Frame}.
6174 @kindex info f @r{(@code{info frame})}
6177 This command prints a verbose description of the selected stack frame,
6182 the address of the frame
6184 the address of the next frame down (called by this frame)
6186 the address of the next frame up (caller of this frame)
6188 the language in which the source code corresponding to this frame is written
6190 the address of the frame's arguments
6192 the address of the frame's local variables
6194 the program counter saved in it (the address of execution in the caller frame)
6196 which registers were saved in the frame
6199 @noindent The verbose description is useful when
6200 something has gone wrong that has made the stack format fail to fit
6201 the usual conventions.
6203 @item info frame @var{addr}
6204 @itemx info f @var{addr}
6205 Print a verbose description of the frame at address @var{addr}, without
6206 selecting that frame. The selected frame remains unchanged by this
6207 command. This requires the same kind of address (more than one for some
6208 architectures) that you specify in the @code{frame} command.
6209 @xref{Selection, ,Selecting a Frame}.
6213 Print the arguments of the selected frame, each on a separate line.
6217 Print the local variables of the selected frame, each on a separate
6218 line. These are all variables (declared either static or automatic)
6219 accessible at the point of execution of the selected frame.
6222 @cindex catch exceptions, list active handlers
6223 @cindex exception handlers, how to list
6225 Print a list of all the exception handlers that are active in the
6226 current stack frame at the current point of execution. To see other
6227 exception handlers, visit the associated frame (using the @code{up},
6228 @code{down}, or @code{frame} commands); then type @code{info catch}.
6229 @xref{Set Catchpoints, , Setting Catchpoints}.
6235 @chapter Examining Source Files
6237 @value{GDBN} can print parts of your program's source, since the debugging
6238 information recorded in the program tells @value{GDBN} what source files were
6239 used to build it. When your program stops, @value{GDBN} spontaneously prints
6240 the line where it stopped. Likewise, when you select a stack frame
6241 (@pxref{Selection, ,Selecting a Frame}), @value{GDBN} prints the line where
6242 execution in that frame has stopped. You can print other portions of
6243 source files by explicit command.
6245 If you use @value{GDBN} through its @sc{gnu} Emacs interface, you may
6246 prefer to use Emacs facilities to view source; see @ref{Emacs, ,Using
6247 @value{GDBN} under @sc{gnu} Emacs}.
6250 * List:: Printing source lines
6251 * Specify Location:: How to specify code locations
6252 * Edit:: Editing source files
6253 * Search:: Searching source files
6254 * Source Path:: Specifying source directories
6255 * Machine Code:: Source and machine code
6259 @section Printing Source Lines
6262 @kindex l @r{(@code{list})}
6263 To print lines from a source file, use the @code{list} command
6264 (abbreviated @code{l}). By default, ten lines are printed.
6265 There are several ways to specify what part of the file you want to
6266 print; see @ref{Specify Location}, for the full list.
6268 Here are the forms of the @code{list} command most commonly used:
6271 @item list @var{linenum}
6272 Print lines centered around line number @var{linenum} in the
6273 current source file.
6275 @item list @var{function}
6276 Print lines centered around the beginning of function
6280 Print more lines. If the last lines printed were printed with a
6281 @code{list} command, this prints lines following the last lines
6282 printed; however, if the last line printed was a solitary line printed
6283 as part of displaying a stack frame (@pxref{Stack, ,Examining the
6284 Stack}), this prints lines centered around that line.
6287 Print lines just before the lines last printed.
6290 @cindex @code{list}, how many lines to display
6291 By default, @value{GDBN} prints ten source lines with any of these forms of
6292 the @code{list} command. You can change this using @code{set listsize}:
6295 @kindex set listsize
6296 @item set listsize @var{count}
6297 Make the @code{list} command display @var{count} source lines (unless
6298 the @code{list} argument explicitly specifies some other number).
6300 @kindex show listsize
6302 Display the number of lines that @code{list} prints.
6305 Repeating a @code{list} command with @key{RET} discards the argument,
6306 so it is equivalent to typing just @code{list}. This is more useful
6307 than listing the same lines again. An exception is made for an
6308 argument of @samp{-}; that argument is preserved in repetition so that
6309 each repetition moves up in the source file.
6311 In general, the @code{list} command expects you to supply zero, one or two
6312 @dfn{linespecs}. Linespecs specify source lines; there are several ways
6313 of writing them (@pxref{Specify Location}), but the effect is always
6314 to specify some source line.
6316 Here is a complete description of the possible arguments for @code{list}:
6319 @item list @var{linespec}
6320 Print lines centered around the line specified by @var{linespec}.
6322 @item list @var{first},@var{last}
6323 Print lines from @var{first} to @var{last}. Both arguments are
6324 linespecs. When a @code{list} command has two linespecs, and the
6325 source file of the second linespec is omitted, this refers to
6326 the same source file as the first linespec.
6328 @item list ,@var{last}
6329 Print lines ending with @var{last}.
6331 @item list @var{first},
6332 Print lines starting with @var{first}.
6335 Print lines just after the lines last printed.
6338 Print lines just before the lines last printed.
6341 As described in the preceding table.
6344 @node Specify Location
6345 @section Specifying a Location
6346 @cindex specifying location
6349 Several @value{GDBN} commands accept arguments that specify a location
6350 of your program's code. Since @value{GDBN} is a source-level
6351 debugger, a location usually specifies some line in the source code;
6352 for that reason, locations are also known as @dfn{linespecs}.
6354 Here are all the different ways of specifying a code location that
6355 @value{GDBN} understands:
6359 Specifies the line number @var{linenum} of the current source file.
6362 @itemx +@var{offset}
6363 Specifies the line @var{offset} lines before or after the @dfn{current
6364 line}. For the @code{list} command, the current line is the last one
6365 printed; for the breakpoint commands, this is the line at which
6366 execution stopped in the currently selected @dfn{stack frame}
6367 (@pxref{Frames, ,Frames}, for a description of stack frames.) When
6368 used as the second of the two linespecs in a @code{list} command,
6369 this specifies the line @var{offset} lines up or down from the first
6372 @item @var{filename}:@var{linenum}
6373 Specifies the line @var{linenum} in the source file @var{filename}.
6375 @item @var{function}
6376 Specifies the line that begins the body of the function @var{function}.
6377 For example, in C, this is the line with the open brace.
6379 @item @var{function}:@var{label}
6380 Specifies the line where @var{label} appears in @var{function}.
6382 @item @var{filename}:@var{function}
6383 Specifies the line that begins the body of the function @var{function}
6384 in the file @var{filename}. You only need the file name with a
6385 function name to avoid ambiguity when there are identically named
6386 functions in different source files.
6389 Specifies the line at which the label named @var{label} appears.
6390 @value{GDBN} searches for the label in the function corresponding to
6391 the currently selected stack frame. If there is no current selected
6392 stack frame (for instance, if the inferior is not running), then
6393 @value{GDBN} will not search for a label.
6395 @item *@var{address}
6396 Specifies the program address @var{address}. For line-oriented
6397 commands, such as @code{list} and @code{edit}, this specifies a source
6398 line that contains @var{address}. For @code{break} and other
6399 breakpoint oriented commands, this can be used to set breakpoints in
6400 parts of your program which do not have debugging information or
6403 Here @var{address} may be any expression valid in the current working
6404 language (@pxref{Languages, working language}) that specifies a code
6405 address. In addition, as a convenience, @value{GDBN} extends the
6406 semantics of expressions used in locations to cover the situations
6407 that frequently happen during debugging. Here are the various forms
6411 @item @var{expression}
6412 Any expression valid in the current working language.
6414 @item @var{funcaddr}
6415 An address of a function or procedure derived from its name. In C,
6416 C@t{++}, Java, Objective-C, Fortran, minimal, and assembly, this is
6417 simply the function's name @var{function} (and actually a special case
6418 of a valid expression). In Pascal and Modula-2, this is
6419 @code{&@var{function}}. In Ada, this is @code{@var{function}'Address}
6420 (although the Pascal form also works).
6422 This form specifies the address of the function's first instruction,
6423 before the stack frame and arguments have been set up.
6425 @item '@var{filename}'::@var{funcaddr}
6426 Like @var{funcaddr} above, but also specifies the name of the source
6427 file explicitly. This is useful if the name of the function does not
6428 specify the function unambiguously, e.g., if there are several
6429 functions with identical names in different source files.
6436 @section Editing Source Files
6437 @cindex editing source files
6440 @kindex e @r{(@code{edit})}
6441 To edit the lines in a source file, use the @code{edit} command.
6442 The editing program of your choice
6443 is invoked with the current line set to
6444 the active line in the program.
6445 Alternatively, there are several ways to specify what part of the file you
6446 want to print if you want to see other parts of the program:
6449 @item edit @var{location}
6450 Edit the source file specified by @code{location}. Editing starts at
6451 that @var{location}, e.g., at the specified source line of the
6452 specified file. @xref{Specify Location}, for all the possible forms
6453 of the @var{location} argument; here are the forms of the @code{edit}
6454 command most commonly used:
6457 @item edit @var{number}
6458 Edit the current source file with @var{number} as the active line number.
6460 @item edit @var{function}
6461 Edit the file containing @var{function} at the beginning of its definition.
6466 @subsection Choosing your Editor
6467 You can customize @value{GDBN} to use any editor you want
6469 The only restriction is that your editor (say @code{ex}), recognizes the
6470 following command-line syntax:
6472 ex +@var{number} file
6474 The optional numeric value +@var{number} specifies the number of the line in
6475 the file where to start editing.}.
6476 By default, it is @file{@value{EDITOR}}, but you can change this
6477 by setting the environment variable @code{EDITOR} before using
6478 @value{GDBN}. For example, to configure @value{GDBN} to use the
6479 @code{vi} editor, you could use these commands with the @code{sh} shell:
6485 or in the @code{csh} shell,
6487 setenv EDITOR /usr/bin/vi
6492 @section Searching Source Files
6493 @cindex searching source files
6495 There are two commands for searching through the current source file for a
6500 @kindex forward-search
6501 @item forward-search @var{regexp}
6502 @itemx search @var{regexp}
6503 The command @samp{forward-search @var{regexp}} checks each line,
6504 starting with the one following the last line listed, for a match for
6505 @var{regexp}. It lists the line that is found. You can use the
6506 synonym @samp{search @var{regexp}} or abbreviate the command name as
6509 @kindex reverse-search
6510 @item reverse-search @var{regexp}
6511 The command @samp{reverse-search @var{regexp}} checks each line, starting
6512 with the one before the last line listed and going backward, for a match
6513 for @var{regexp}. It lists the line that is found. You can abbreviate
6514 this command as @code{rev}.
6518 @section Specifying Source Directories
6521 @cindex directories for source files
6522 Executable programs sometimes do not record the directories of the source
6523 files from which they were compiled, just the names. Even when they do,
6524 the directories could be moved between the compilation and your debugging
6525 session. @value{GDBN} has a list of directories to search for source files;
6526 this is called the @dfn{source path}. Each time @value{GDBN} wants a source file,
6527 it tries all the directories in the list, in the order they are present
6528 in the list, until it finds a file with the desired name.
6530 For example, suppose an executable references the file
6531 @file{/usr/src/foo-1.0/lib/foo.c}, and our source path is
6532 @file{/mnt/cross}. The file is first looked up literally; if this
6533 fails, @file{/mnt/cross/usr/src/foo-1.0/lib/foo.c} is tried; if this
6534 fails, @file{/mnt/cross/foo.c} is opened; if this fails, an error
6535 message is printed. @value{GDBN} does not look up the parts of the
6536 source file name, such as @file{/mnt/cross/src/foo-1.0/lib/foo.c}.
6537 Likewise, the subdirectories of the source path are not searched: if
6538 the source path is @file{/mnt/cross}, and the binary refers to
6539 @file{foo.c}, @value{GDBN} would not find it under
6540 @file{/mnt/cross/usr/src/foo-1.0/lib}.
6542 Plain file names, relative file names with leading directories, file
6543 names containing dots, etc.@: are all treated as described above; for
6544 instance, if the source path is @file{/mnt/cross}, and the source file
6545 is recorded as @file{../lib/foo.c}, @value{GDBN} would first try
6546 @file{../lib/foo.c}, then @file{/mnt/cross/../lib/foo.c}, and after
6547 that---@file{/mnt/cross/foo.c}.
6549 Note that the executable search path is @emph{not} used to locate the
6552 Whenever you reset or rearrange the source path, @value{GDBN} clears out
6553 any information it has cached about where source files are found and where
6554 each line is in the file.
6558 When you start @value{GDBN}, its source path includes only @samp{cdir}
6559 and @samp{cwd}, in that order.
6560 To add other directories, use the @code{directory} command.
6562 The search path is used to find both program source files and @value{GDBN}
6563 script files (read using the @samp{-command} option and @samp{source} command).
6565 In addition to the source path, @value{GDBN} provides a set of commands
6566 that manage a list of source path substitution rules. A @dfn{substitution
6567 rule} specifies how to rewrite source directories stored in the program's
6568 debug information in case the sources were moved to a different
6569 directory between compilation and debugging. A rule is made of
6570 two strings, the first specifying what needs to be rewritten in
6571 the path, and the second specifying how it should be rewritten.
6572 In @ref{set substitute-path}, we name these two parts @var{from} and
6573 @var{to} respectively. @value{GDBN} does a simple string replacement
6574 of @var{from} with @var{to} at the start of the directory part of the
6575 source file name, and uses that result instead of the original file
6576 name to look up the sources.
6578 Using the previous example, suppose the @file{foo-1.0} tree has been
6579 moved from @file{/usr/src} to @file{/mnt/cross}, then you can tell
6580 @value{GDBN} to replace @file{/usr/src} in all source path names with
6581 @file{/mnt/cross}. The first lookup will then be
6582 @file{/mnt/cross/foo-1.0/lib/foo.c} in place of the original location
6583 of @file{/usr/src/foo-1.0/lib/foo.c}. To define a source path
6584 substitution rule, use the @code{set substitute-path} command
6585 (@pxref{set substitute-path}).
6587 To avoid unexpected substitution results, a rule is applied only if the
6588 @var{from} part of the directory name ends at a directory separator.
6589 For instance, a rule substituting @file{/usr/source} into
6590 @file{/mnt/cross} will be applied to @file{/usr/source/foo-1.0} but
6591 not to @file{/usr/sourceware/foo-2.0}. And because the substitution
6592 is applied only at the beginning of the directory name, this rule will
6593 not be applied to @file{/root/usr/source/baz.c} either.
6595 In many cases, you can achieve the same result using the @code{directory}
6596 command. However, @code{set substitute-path} can be more efficient in
6597 the case where the sources are organized in a complex tree with multiple
6598 subdirectories. With the @code{directory} command, you need to add each
6599 subdirectory of your project. If you moved the entire tree while
6600 preserving its internal organization, then @code{set substitute-path}
6601 allows you to direct the debugger to all the sources with one single
6604 @code{set substitute-path} is also more than just a shortcut command.
6605 The source path is only used if the file at the original location no
6606 longer exists. On the other hand, @code{set substitute-path} modifies
6607 the debugger behavior to look at the rewritten location instead. So, if
6608 for any reason a source file that is not relevant to your executable is
6609 located at the original location, a substitution rule is the only
6610 method available to point @value{GDBN} at the new location.
6612 @cindex @samp{--with-relocated-sources}
6613 @cindex default source path substitution
6614 You can configure a default source path substitution rule by
6615 configuring @value{GDBN} with the
6616 @samp{--with-relocated-sources=@var{dir}} option. The @var{dir}
6617 should be the name of a directory under @value{GDBN}'s configured
6618 prefix (set with @samp{--prefix} or @samp{--exec-prefix}), and
6619 directory names in debug information under @var{dir} will be adjusted
6620 automatically if the installed @value{GDBN} is moved to a new
6621 location. This is useful if @value{GDBN}, libraries or executables
6622 with debug information and corresponding source code are being moved
6626 @item directory @var{dirname} @dots{}
6627 @item dir @var{dirname} @dots{}
6628 Add directory @var{dirname} to the front of the source path. Several
6629 directory names may be given to this command, separated by @samp{:}
6630 (@samp{;} on MS-DOS and MS-Windows, where @samp{:} usually appears as
6631 part of absolute file names) or
6632 whitespace. You may specify a directory that is already in the source
6633 path; this moves it forward, so @value{GDBN} searches it sooner.
6637 @vindex $cdir@r{, convenience variable}
6638 @vindex $cwd@r{, convenience variable}
6639 @cindex compilation directory
6640 @cindex current directory
6641 @cindex working directory
6642 @cindex directory, current
6643 @cindex directory, compilation
6644 You can use the string @samp{$cdir} to refer to the compilation
6645 directory (if one is recorded), and @samp{$cwd} to refer to the current
6646 working directory. @samp{$cwd} is not the same as @samp{.}---the former
6647 tracks the current working directory as it changes during your @value{GDBN}
6648 session, while the latter is immediately expanded to the current
6649 directory at the time you add an entry to the source path.
6652 Reset the source path to its default value (@samp{$cdir:$cwd} on Unix systems). This requires confirmation.
6654 @c RET-repeat for @code{directory} is explicitly disabled, but since
6655 @c repeating it would be a no-op we do not say that. (thanks to RMS)
6657 @item set directories @var{path-list}
6658 @kindex set directories
6659 Set the source path to @var{path-list}.
6660 @samp{$cdir:$cwd} are added if missing.
6662 @item show directories
6663 @kindex show directories
6664 Print the source path: show which directories it contains.
6666 @anchor{set substitute-path}
6667 @item set substitute-path @var{from} @var{to}
6668 @kindex set substitute-path
6669 Define a source path substitution rule, and add it at the end of the
6670 current list of existing substitution rules. If a rule with the same
6671 @var{from} was already defined, then the old rule is also deleted.
6673 For example, if the file @file{/foo/bar/baz.c} was moved to
6674 @file{/mnt/cross/baz.c}, then the command
6677 (@value{GDBP}) set substitute-path /usr/src /mnt/cross
6681 will tell @value{GDBN} to replace @samp{/usr/src} with
6682 @samp{/mnt/cross}, which will allow @value{GDBN} to find the file
6683 @file{baz.c} even though it was moved.
6685 In the case when more than one substitution rule have been defined,
6686 the rules are evaluated one by one in the order where they have been
6687 defined. The first one matching, if any, is selected to perform
6690 For instance, if we had entered the following commands:
6693 (@value{GDBP}) set substitute-path /usr/src/include /mnt/include
6694 (@value{GDBP}) set substitute-path /usr/src /mnt/src
6698 @value{GDBN} would then rewrite @file{/usr/src/include/defs.h} into
6699 @file{/mnt/include/defs.h} by using the first rule. However, it would
6700 use the second rule to rewrite @file{/usr/src/lib/foo.c} into
6701 @file{/mnt/src/lib/foo.c}.
6704 @item unset substitute-path [path]
6705 @kindex unset substitute-path
6706 If a path is specified, search the current list of substitution rules
6707 for a rule that would rewrite that path. Delete that rule if found.
6708 A warning is emitted by the debugger if no rule could be found.
6710 If no path is specified, then all substitution rules are deleted.
6712 @item show substitute-path [path]
6713 @kindex show substitute-path
6714 If a path is specified, then print the source path substitution rule
6715 which would rewrite that path, if any.
6717 If no path is specified, then print all existing source path substitution
6722 If your source path is cluttered with directories that are no longer of
6723 interest, @value{GDBN} may sometimes cause confusion by finding the wrong
6724 versions of source. You can correct the situation as follows:
6728 Use @code{directory} with no argument to reset the source path to its default value.
6731 Use @code{directory} with suitable arguments to reinstall the
6732 directories you want in the source path. You can add all the
6733 directories in one command.
6737 @section Source and Machine Code
6738 @cindex source line and its code address
6740 You can use the command @code{info line} to map source lines to program
6741 addresses (and vice versa), and the command @code{disassemble} to display
6742 a range of addresses as machine instructions. You can use the command
6743 @code{set disassemble-next-line} to set whether to disassemble next
6744 source line when execution stops. When run under @sc{gnu} Emacs
6745 mode, the @code{info line} command causes the arrow to point to the
6746 line specified. Also, @code{info line} prints addresses in symbolic form as
6751 @item info line @var{linespec}
6752 Print the starting and ending addresses of the compiled code for
6753 source line @var{linespec}. You can specify source lines in any of
6754 the ways documented in @ref{Specify Location}.
6757 For example, we can use @code{info line} to discover the location of
6758 the object code for the first line of function
6759 @code{m4_changequote}:
6761 @c FIXME: I think this example should also show the addresses in
6762 @c symbolic form, as they usually would be displayed.
6764 (@value{GDBP}) info line m4_changequote
6765 Line 895 of "builtin.c" starts at pc 0x634c and ends at 0x6350.
6769 @cindex code address and its source line
6770 We can also inquire (using @code{*@var{addr}} as the form for
6771 @var{linespec}) what source line covers a particular address:
6773 (@value{GDBP}) info line *0x63ff
6774 Line 926 of "builtin.c" starts at pc 0x63e4 and ends at 0x6404.
6777 @cindex @code{$_} and @code{info line}
6778 @cindex @code{x} command, default address
6779 @kindex x@r{(examine), and} info line
6780 After @code{info line}, the default address for the @code{x} command
6781 is changed to the starting address of the line, so that @samp{x/i} is
6782 sufficient to begin examining the machine code (@pxref{Memory,
6783 ,Examining Memory}). Also, this address is saved as the value of the
6784 convenience variable @code{$_} (@pxref{Convenience Vars, ,Convenience
6789 @cindex assembly instructions
6790 @cindex instructions, assembly
6791 @cindex machine instructions
6792 @cindex listing machine instructions
6794 @itemx disassemble /m
6795 @itemx disassemble /r
6796 This specialized command dumps a range of memory as machine
6797 instructions. It can also print mixed source+disassembly by specifying
6798 the @code{/m} modifier and print the raw instructions in hex as well as
6799 in symbolic form by specifying the @code{/r}.
6800 The default memory range is the function surrounding the
6801 program counter of the selected frame. A single argument to this
6802 command is a program counter value; @value{GDBN} dumps the function
6803 surrounding this value. When two arguments are given, they should
6804 be separated by a comma, possibly surrounded by whitespace. The
6805 arguments specify a range of addresses to dump, in one of two forms:
6808 @item @var{start},@var{end}
6809 the addresses from @var{start} (inclusive) to @var{end} (exclusive)
6810 @item @var{start},+@var{length}
6811 the addresses from @var{start} (inclusive) to
6812 @code{@var{start}+@var{length}} (exclusive).
6816 When 2 arguments are specified, the name of the function is also
6817 printed (since there could be several functions in the given range).
6819 The argument(s) can be any expression yielding a numeric value, such as
6820 @samp{0x32c4}, @samp{&main+10} or @samp{$pc - 8}.
6822 If the range of memory being disassembled contains current program counter,
6823 the instruction at that location is shown with a @code{=>} marker.
6826 The following example shows the disassembly of a range of addresses of
6827 HP PA-RISC 2.0 code:
6830 (@value{GDBP}) disas 0x32c4, 0x32e4
6831 Dump of assembler code from 0x32c4 to 0x32e4:
6832 0x32c4 <main+204>: addil 0,dp
6833 0x32c8 <main+208>: ldw 0x22c(sr0,r1),r26
6834 0x32cc <main+212>: ldil 0x3000,r31
6835 0x32d0 <main+216>: ble 0x3f8(sr4,r31)
6836 0x32d4 <main+220>: ldo 0(r31),rp
6837 0x32d8 <main+224>: addil -0x800,dp
6838 0x32dc <main+228>: ldo 0x588(r1),r26
6839 0x32e0 <main+232>: ldil 0x3000,r31
6840 End of assembler dump.
6843 Here is an example showing mixed source+assembly for Intel x86, when the
6844 program is stopped just after function prologue:
6847 (@value{GDBP}) disas /m main
6848 Dump of assembler code for function main:
6850 0x08048330 <+0>: push %ebp
6851 0x08048331 <+1>: mov %esp,%ebp
6852 0x08048333 <+3>: sub $0x8,%esp
6853 0x08048336 <+6>: and $0xfffffff0,%esp
6854 0x08048339 <+9>: sub $0x10,%esp
6856 6 printf ("Hello.\n");
6857 => 0x0804833c <+12>: movl $0x8048440,(%esp)
6858 0x08048343 <+19>: call 0x8048284 <puts@@plt>
6862 0x08048348 <+24>: mov $0x0,%eax
6863 0x0804834d <+29>: leave
6864 0x0804834e <+30>: ret
6866 End of assembler dump.
6869 Here is another example showing raw instructions in hex for AMD x86-64,
6872 (gdb) disas /r 0x400281,+10
6873 Dump of assembler code from 0x400281 to 0x40028b:
6874 0x0000000000400281: 38 36 cmp %dh,(%rsi)
6875 0x0000000000400283: 2d 36 34 2e 73 sub $0x732e3436,%eax
6876 0x0000000000400288: 6f outsl %ds:(%rsi),(%dx)
6877 0x0000000000400289: 2e 32 00 xor %cs:(%rax),%al
6878 End of assembler dump.
6881 Some architectures have more than one commonly-used set of instruction
6882 mnemonics or other syntax.
6884 For programs that were dynamically linked and use shared libraries,
6885 instructions that call functions or branch to locations in the shared
6886 libraries might show a seemingly bogus location---it's actually a
6887 location of the relocation table. On some architectures, @value{GDBN}
6888 might be able to resolve these to actual function names.
6891 @kindex set disassembly-flavor
6892 @cindex Intel disassembly flavor
6893 @cindex AT&T disassembly flavor
6894 @item set disassembly-flavor @var{instruction-set}
6895 Select the instruction set to use when disassembling the
6896 program via the @code{disassemble} or @code{x/i} commands.
6898 Currently this command is only defined for the Intel x86 family. You
6899 can set @var{instruction-set} to either @code{intel} or @code{att}.
6900 The default is @code{att}, the AT&T flavor used by default by Unix
6901 assemblers for x86-based targets.
6903 @kindex show disassembly-flavor
6904 @item show disassembly-flavor
6905 Show the current setting of the disassembly flavor.
6909 @kindex set disassemble-next-line
6910 @kindex show disassemble-next-line
6911 @item set disassemble-next-line
6912 @itemx show disassemble-next-line
6913 Control whether or not @value{GDBN} will disassemble the next source
6914 line or instruction when execution stops. If ON, @value{GDBN} will
6915 display disassembly of the next source line when execution of the
6916 program being debugged stops. This is @emph{in addition} to
6917 displaying the source line itself, which @value{GDBN} always does if
6918 possible. If the next source line cannot be displayed for some reason
6919 (e.g., if @value{GDBN} cannot find the source file, or there's no line
6920 info in the debug info), @value{GDBN} will display disassembly of the
6921 next @emph{instruction} instead of showing the next source line. If
6922 AUTO, @value{GDBN} will display disassembly of next instruction only
6923 if the source line cannot be displayed. This setting causes
6924 @value{GDBN} to display some feedback when you step through a function
6925 with no line info or whose source file is unavailable. The default is
6926 OFF, which means never display the disassembly of the next line or
6932 @chapter Examining Data
6934 @cindex printing data
6935 @cindex examining data
6938 @c "inspect" is not quite a synonym if you are using Epoch, which we do not
6939 @c document because it is nonstandard... Under Epoch it displays in a
6940 @c different window or something like that.
6941 The usual way to examine data in your program is with the @code{print}
6942 command (abbreviated @code{p}), or its synonym @code{inspect}. It
6943 evaluates and prints the value of an expression of the language your
6944 program is written in (@pxref{Languages, ,Using @value{GDBN} with
6945 Different Languages}). It may also print the expression using a
6946 Python-based pretty-printer (@pxref{Pretty Printing}).
6949 @item print @var{expr}
6950 @itemx print /@var{f} @var{expr}
6951 @var{expr} is an expression (in the source language). By default the
6952 value of @var{expr} is printed in a format appropriate to its data type;
6953 you can choose a different format by specifying @samp{/@var{f}}, where
6954 @var{f} is a letter specifying the format; see @ref{Output Formats,,Output
6958 @itemx print /@var{f}
6959 @cindex reprint the last value
6960 If you omit @var{expr}, @value{GDBN} displays the last value again (from the
6961 @dfn{value history}; @pxref{Value History, ,Value History}). This allows you to
6962 conveniently inspect the same value in an alternative format.
6965 A more low-level way of examining data is with the @code{x} command.
6966 It examines data in memory at a specified address and prints it in a
6967 specified format. @xref{Memory, ,Examining Memory}.
6969 If you are interested in information about types, or about how the
6970 fields of a struct or a class are declared, use the @code{ptype @var{exp}}
6971 command rather than @code{print}. @xref{Symbols, ,Examining the Symbol
6975 * Expressions:: Expressions
6976 * Ambiguous Expressions:: Ambiguous Expressions
6977 * Variables:: Program variables
6978 * Arrays:: Artificial arrays
6979 * Output Formats:: Output formats
6980 * Memory:: Examining memory
6981 * Auto Display:: Automatic display
6982 * Print Settings:: Print settings
6983 * Pretty Printing:: Python pretty printing
6984 * Value History:: Value history
6985 * Convenience Vars:: Convenience variables
6986 * Registers:: Registers
6987 * Floating Point Hardware:: Floating point hardware
6988 * Vector Unit:: Vector Unit
6989 * OS Information:: Auxiliary data provided by operating system
6990 * Memory Region Attributes:: Memory region attributes
6991 * Dump/Restore Files:: Copy between memory and a file
6992 * Core File Generation:: Cause a program dump its core
6993 * Character Sets:: Debugging programs that use a different
6994 character set than GDB does
6995 * Caching Remote Data:: Data caching for remote targets
6996 * Searching Memory:: Searching memory for a sequence of bytes
7000 @section Expressions
7003 @code{print} and many other @value{GDBN} commands accept an expression and
7004 compute its value. Any kind of constant, variable or operator defined
7005 by the programming language you are using is valid in an expression in
7006 @value{GDBN}. This includes conditional expressions, function calls,
7007 casts, and string constants. It also includes preprocessor macros, if
7008 you compiled your program to include this information; see
7011 @cindex arrays in expressions
7012 @value{GDBN} supports array constants in expressions input by
7013 the user. The syntax is @{@var{element}, @var{element}@dots{}@}. For example,
7014 you can use the command @code{print @{1, 2, 3@}} to create an array
7015 of three integers. If you pass an array to a function or assign it
7016 to a program variable, @value{GDBN} copies the array to memory that
7017 is @code{malloc}ed in the target program.
7019 Because C is so widespread, most of the expressions shown in examples in
7020 this manual are in C. @xref{Languages, , Using @value{GDBN} with Different
7021 Languages}, for information on how to use expressions in other
7024 In this section, we discuss operators that you can use in @value{GDBN}
7025 expressions regardless of your programming language.
7027 @cindex casts, in expressions
7028 Casts are supported in all languages, not just in C, because it is so
7029 useful to cast a number into a pointer in order to examine a structure
7030 at that address in memory.
7031 @c FIXME: casts supported---Mod2 true?
7033 @value{GDBN} supports these operators, in addition to those common
7034 to programming languages:
7038 @samp{@@} is a binary operator for treating parts of memory as arrays.
7039 @xref{Arrays, ,Artificial Arrays}, for more information.
7042 @samp{::} allows you to specify a variable in terms of the file or
7043 function where it is defined. @xref{Variables, ,Program Variables}.
7045 @cindex @{@var{type}@}
7046 @cindex type casting memory
7047 @cindex memory, viewing as typed object
7048 @cindex casts, to view memory
7049 @item @{@var{type}@} @var{addr}
7050 Refers to an object of type @var{type} stored at address @var{addr} in
7051 memory. @var{addr} may be any expression whose value is an integer or
7052 pointer (but parentheses are required around binary operators, just as in
7053 a cast). This construct is allowed regardless of what kind of data is
7054 normally supposed to reside at @var{addr}.
7057 @node Ambiguous Expressions
7058 @section Ambiguous Expressions
7059 @cindex ambiguous expressions
7061 Expressions can sometimes contain some ambiguous elements. For instance,
7062 some programming languages (notably Ada, C@t{++} and Objective-C) permit
7063 a single function name to be defined several times, for application in
7064 different contexts. This is called @dfn{overloading}. Another example
7065 involving Ada is generics. A @dfn{generic package} is similar to C@t{++}
7066 templates and is typically instantiated several times, resulting in
7067 the same function name being defined in different contexts.
7069 In some cases and depending on the language, it is possible to adjust
7070 the expression to remove the ambiguity. For instance in C@t{++}, you
7071 can specify the signature of the function you want to break on, as in
7072 @kbd{break @var{function}(@var{types})}. In Ada, using the fully
7073 qualified name of your function often makes the expression unambiguous
7076 When an ambiguity that needs to be resolved is detected, the debugger
7077 has the capability to display a menu of numbered choices for each
7078 possibility, and then waits for the selection with the prompt @samp{>}.
7079 The first option is always @samp{[0] cancel}, and typing @kbd{0 @key{RET}}
7080 aborts the current command. If the command in which the expression was
7081 used allows more than one choice to be selected, the next option in the
7082 menu is @samp{[1] all}, and typing @kbd{1 @key{RET}} selects all possible
7085 For example, the following session excerpt shows an attempt to set a
7086 breakpoint at the overloaded symbol @code{String::after}.
7087 We choose three particular definitions of that function name:
7089 @c FIXME! This is likely to change to show arg type lists, at least
7092 (@value{GDBP}) b String::after
7095 [2] file:String.cc; line number:867
7096 [3] file:String.cc; line number:860
7097 [4] file:String.cc; line number:875
7098 [5] file:String.cc; line number:853
7099 [6] file:String.cc; line number:846
7100 [7] file:String.cc; line number:735
7102 Breakpoint 1 at 0xb26c: file String.cc, line 867.
7103 Breakpoint 2 at 0xb344: file String.cc, line 875.
7104 Breakpoint 3 at 0xafcc: file String.cc, line 846.
7105 Multiple breakpoints were set.
7106 Use the "delete" command to delete unwanted
7113 @kindex set multiple-symbols
7114 @item set multiple-symbols @var{mode}
7115 @cindex multiple-symbols menu
7117 This option allows you to adjust the debugger behavior when an expression
7120 By default, @var{mode} is set to @code{all}. If the command with which
7121 the expression is used allows more than one choice, then @value{GDBN}
7122 automatically selects all possible choices. For instance, inserting
7123 a breakpoint on a function using an ambiguous name results in a breakpoint
7124 inserted on each possible match. However, if a unique choice must be made,
7125 then @value{GDBN} uses the menu to help you disambiguate the expression.
7126 For instance, printing the address of an overloaded function will result
7127 in the use of the menu.
7129 When @var{mode} is set to @code{ask}, the debugger always uses the menu
7130 when an ambiguity is detected.
7132 Finally, when @var{mode} is set to @code{cancel}, the debugger reports
7133 an error due to the ambiguity and the command is aborted.
7135 @kindex show multiple-symbols
7136 @item show multiple-symbols
7137 Show the current value of the @code{multiple-symbols} setting.
7141 @section Program Variables
7143 The most common kind of expression to use is the name of a variable
7146 Variables in expressions are understood in the selected stack frame
7147 (@pxref{Selection, ,Selecting a Frame}); they must be either:
7151 global (or file-static)
7158 visible according to the scope rules of the
7159 programming language from the point of execution in that frame
7162 @noindent This means that in the function
7177 you can examine and use the variable @code{a} whenever your program is
7178 executing within the function @code{foo}, but you can only use or
7179 examine the variable @code{b} while your program is executing inside
7180 the block where @code{b} is declared.
7182 @cindex variable name conflict
7183 There is an exception: you can refer to a variable or function whose
7184 scope is a single source file even if the current execution point is not
7185 in this file. But it is possible to have more than one such variable or
7186 function with the same name (in different source files). If that
7187 happens, referring to that name has unpredictable effects. If you wish,
7188 you can specify a static variable in a particular function or file,
7189 using the colon-colon (@code{::}) notation:
7191 @cindex colon-colon, context for variables/functions
7193 @c info cannot cope with a :: index entry, but why deprive hard copy readers?
7194 @cindex @code{::}, context for variables/functions
7197 @var{file}::@var{variable}
7198 @var{function}::@var{variable}
7202 Here @var{file} or @var{function} is the name of the context for the
7203 static @var{variable}. In the case of file names, you can use quotes to
7204 make sure @value{GDBN} parses the file name as a single word---for example,
7205 to print a global value of @code{x} defined in @file{f2.c}:
7208 (@value{GDBP}) p 'f2.c'::x
7211 @cindex C@t{++} scope resolution
7212 This use of @samp{::} is very rarely in conflict with the very similar
7213 use of the same notation in C@t{++}. @value{GDBN} also supports use of the C@t{++}
7214 scope resolution operator in @value{GDBN} expressions.
7215 @c FIXME: Um, so what happens in one of those rare cases where it's in
7218 @cindex wrong values
7219 @cindex variable values, wrong
7220 @cindex function entry/exit, wrong values of variables
7221 @cindex optimized code, wrong values of variables
7223 @emph{Warning:} Occasionally, a local variable may appear to have the
7224 wrong value at certain points in a function---just after entry to a new
7225 scope, and just before exit.
7227 You may see this problem when you are stepping by machine instructions.
7228 This is because, on most machines, it takes more than one instruction to
7229 set up a stack frame (including local variable definitions); if you are
7230 stepping by machine instructions, variables may appear to have the wrong
7231 values until the stack frame is completely built. On exit, it usually
7232 also takes more than one machine instruction to destroy a stack frame;
7233 after you begin stepping through that group of instructions, local
7234 variable definitions may be gone.
7236 This may also happen when the compiler does significant optimizations.
7237 To be sure of always seeing accurate values, turn off all optimization
7240 @cindex ``No symbol "foo" in current context''
7241 Another possible effect of compiler optimizations is to optimize
7242 unused variables out of existence, or assign variables to registers (as
7243 opposed to memory addresses). Depending on the support for such cases
7244 offered by the debug info format used by the compiler, @value{GDBN}
7245 might not be able to display values for such local variables. If that
7246 happens, @value{GDBN} will print a message like this:
7249 No symbol "foo" in current context.
7252 To solve such problems, either recompile without optimizations, or use a
7253 different debug info format, if the compiler supports several such
7254 formats. For example, @value{NGCC}, the @sc{gnu} C/C@t{++} compiler,
7255 usually supports the @option{-gstabs+} option. @option{-gstabs+}
7256 produces debug info in a format that is superior to formats such as
7257 COFF. You may be able to use DWARF 2 (@option{-gdwarf-2}), which is also
7258 an effective form for debug info. @xref{Debugging Options,,Options
7259 for Debugging Your Program or GCC, gcc.info, Using the @sc{gnu}
7260 Compiler Collection (GCC)}.
7261 @xref{C, ,C and C@t{++}}, for more information about debug info formats
7262 that are best suited to C@t{++} programs.
7264 If you ask to print an object whose contents are unknown to
7265 @value{GDBN}, e.g., because its data type is not completely specified
7266 by the debug information, @value{GDBN} will say @samp{<incomplete
7267 type>}. @xref{Symbols, incomplete type}, for more about this.
7269 Strings are identified as arrays of @code{char} values without specified
7270 signedness. Arrays of either @code{signed char} or @code{unsigned char} get
7271 printed as arrays of 1 byte sized integers. @code{-fsigned-char} or
7272 @code{-funsigned-char} @value{NGCC} options have no effect as @value{GDBN}
7273 defines literal string type @code{"char"} as @code{char} without a sign.
7278 signed char var1[] = "A";
7281 You get during debugging
7286 $2 = @{65 'A', 0 '\0'@}
7290 @section Artificial Arrays
7292 @cindex artificial array
7294 @kindex @@@r{, referencing memory as an array}
7295 It is often useful to print out several successive objects of the
7296 same type in memory; a section of an array, or an array of
7297 dynamically determined size for which only a pointer exists in the
7300 You can do this by referring to a contiguous span of memory as an
7301 @dfn{artificial array}, using the binary operator @samp{@@}. The left
7302 operand of @samp{@@} should be the first element of the desired array
7303 and be an individual object. The right operand should be the desired length
7304 of the array. The result is an array value whose elements are all of
7305 the type of the left argument. The first element is actually the left
7306 argument; the second element comes from bytes of memory immediately
7307 following those that hold the first element, and so on. Here is an
7308 example. If a program says
7311 int *array = (int *) malloc (len * sizeof (int));
7315 you can print the contents of @code{array} with
7321 The left operand of @samp{@@} must reside in memory. Array values made
7322 with @samp{@@} in this way behave just like other arrays in terms of
7323 subscripting, and are coerced to pointers when used in expressions.
7324 Artificial arrays most often appear in expressions via the value history
7325 (@pxref{Value History, ,Value History}), after printing one out.
7327 Another way to create an artificial array is to use a cast.
7328 This re-interprets a value as if it were an array.
7329 The value need not be in memory:
7331 (@value{GDBP}) p/x (short[2])0x12345678
7332 $1 = @{0x1234, 0x5678@}
7335 As a convenience, if you leave the array length out (as in
7336 @samp{(@var{type}[])@var{value}}) @value{GDBN} calculates the size to fill
7337 the value (as @samp{sizeof(@var{value})/sizeof(@var{type})}:
7339 (@value{GDBP}) p/x (short[])0x12345678
7340 $2 = @{0x1234, 0x5678@}
7343 Sometimes the artificial array mechanism is not quite enough; in
7344 moderately complex data structures, the elements of interest may not
7345 actually be adjacent---for example, if you are interested in the values
7346 of pointers in an array. One useful work-around in this situation is
7347 to use a convenience variable (@pxref{Convenience Vars, ,Convenience
7348 Variables}) as a counter in an expression that prints the first
7349 interesting value, and then repeat that expression via @key{RET}. For
7350 instance, suppose you have an array @code{dtab} of pointers to
7351 structures, and you are interested in the values of a field @code{fv}
7352 in each structure. Here is an example of what you might type:
7362 @node Output Formats
7363 @section Output Formats
7365 @cindex formatted output
7366 @cindex output formats
7367 By default, @value{GDBN} prints a value according to its data type. Sometimes
7368 this is not what you want. For example, you might want to print a number
7369 in hex, or a pointer in decimal. Or you might want to view data in memory
7370 at a certain address as a character string or as an instruction. To do
7371 these things, specify an @dfn{output format} when you print a value.
7373 The simplest use of output formats is to say how to print a value
7374 already computed. This is done by starting the arguments of the
7375 @code{print} command with a slash and a format letter. The format
7376 letters supported are:
7380 Regard the bits of the value as an integer, and print the integer in
7384 Print as integer in signed decimal.
7387 Print as integer in unsigned decimal.
7390 Print as integer in octal.
7393 Print as integer in binary. The letter @samp{t} stands for ``two''.
7394 @footnote{@samp{b} cannot be used because these format letters are also
7395 used with the @code{x} command, where @samp{b} stands for ``byte'';
7396 see @ref{Memory,,Examining Memory}.}
7399 @cindex unknown address, locating
7400 @cindex locate address
7401 Print as an address, both absolute in hexadecimal and as an offset from
7402 the nearest preceding symbol. You can use this format used to discover
7403 where (in what function) an unknown address is located:
7406 (@value{GDBP}) p/a 0x54320
7407 $3 = 0x54320 <_initialize_vx+396>
7411 The command @code{info symbol 0x54320} yields similar results.
7412 @xref{Symbols, info symbol}.
7415 Regard as an integer and print it as a character constant. This
7416 prints both the numerical value and its character representation. The
7417 character representation is replaced with the octal escape @samp{\nnn}
7418 for characters outside the 7-bit @sc{ascii} range.
7420 Without this format, @value{GDBN} displays @code{char},
7421 @w{@code{unsigned char}}, and @w{@code{signed char}} data as character
7422 constants. Single-byte members of vectors are displayed as integer
7426 Regard the bits of the value as a floating point number and print
7427 using typical floating point syntax.
7430 @cindex printing strings
7431 @cindex printing byte arrays
7432 Regard as a string, if possible. With this format, pointers to single-byte
7433 data are displayed as null-terminated strings and arrays of single-byte data
7434 are displayed as fixed-length strings. Other values are displayed in their
7437 Without this format, @value{GDBN} displays pointers to and arrays of
7438 @code{char}, @w{@code{unsigned char}}, and @w{@code{signed char}} as
7439 strings. Single-byte members of a vector are displayed as an integer
7443 @cindex raw printing
7444 Print using the @samp{raw} formatting. By default, @value{GDBN} will
7445 use a Python-based pretty-printer, if one is available (@pxref{Pretty
7446 Printing}). This typically results in a higher-level display of the
7447 value's contents. The @samp{r} format bypasses any Python
7448 pretty-printer which might exist.
7451 For example, to print the program counter in hex (@pxref{Registers}), type
7458 Note that no space is required before the slash; this is because command
7459 names in @value{GDBN} cannot contain a slash.
7461 To reprint the last value in the value history with a different format,
7462 you can use the @code{print} command with just a format and no
7463 expression. For example, @samp{p/x} reprints the last value in hex.
7466 @section Examining Memory
7468 You can use the command @code{x} (for ``examine'') to examine memory in
7469 any of several formats, independently of your program's data types.
7471 @cindex examining memory
7473 @kindex x @r{(examine memory)}
7474 @item x/@var{nfu} @var{addr}
7477 Use the @code{x} command to examine memory.
7480 @var{n}, @var{f}, and @var{u} are all optional parameters that specify how
7481 much memory to display and how to format it; @var{addr} is an
7482 expression giving the address where you want to start displaying memory.
7483 If you use defaults for @var{nfu}, you need not type the slash @samp{/}.
7484 Several commands set convenient defaults for @var{addr}.
7487 @item @var{n}, the repeat count
7488 The repeat count is a decimal integer; the default is 1. It specifies
7489 how much memory (counting by units @var{u}) to display.
7490 @c This really is **decimal**; unaffected by 'set radix' as of GDB
7493 @item @var{f}, the display format
7494 The display format is one of the formats used by @code{print}
7495 (@samp{x}, @samp{d}, @samp{u}, @samp{o}, @samp{t}, @samp{a}, @samp{c},
7496 @samp{f}, @samp{s}), and in addition @samp{i} (for machine instructions).
7497 The default is @samp{x} (hexadecimal) initially. The default changes
7498 each time you use either @code{x} or @code{print}.
7500 @item @var{u}, the unit size
7501 The unit size is any of
7507 Halfwords (two bytes).
7509 Words (four bytes). This is the initial default.
7511 Giant words (eight bytes).
7514 Each time you specify a unit size with @code{x}, that size becomes the
7515 default unit the next time you use @code{x}. For the @samp{i} format,
7516 the unit size is ignored and is normally not written. For the @samp{s} format,
7517 the unit size defaults to @samp{b}, unless it is explicitly given.
7518 Use @kbd{x /hs} to display 16-bit char strings and @kbd{x /ws} to display
7519 32-bit strings. The next use of @kbd{x /s} will again display 8-bit strings.
7520 Note that the results depend on the programming language of the
7521 current compilation unit. If the language is C, the @samp{s}
7522 modifier will use the UTF-16 encoding while @samp{w} will use
7523 UTF-32. The encoding is set by the programming language and cannot
7526 @item @var{addr}, starting display address
7527 @var{addr} is the address where you want @value{GDBN} to begin displaying
7528 memory. The expression need not have a pointer value (though it may);
7529 it is always interpreted as an integer address of a byte of memory.
7530 @xref{Expressions, ,Expressions}, for more information on expressions. The default for
7531 @var{addr} is usually just after the last address examined---but several
7532 other commands also set the default address: @code{info breakpoints} (to
7533 the address of the last breakpoint listed), @code{info line} (to the
7534 starting address of a line), and @code{print} (if you use it to display
7535 a value from memory).
7538 For example, @samp{x/3uh 0x54320} is a request to display three halfwords
7539 (@code{h}) of memory, formatted as unsigned decimal integers (@samp{u}),
7540 starting at address @code{0x54320}. @samp{x/4xw $sp} prints the four
7541 words (@samp{w}) of memory above the stack pointer (here, @samp{$sp};
7542 @pxref{Registers, ,Registers}) in hexadecimal (@samp{x}).
7544 Since the letters indicating unit sizes are all distinct from the
7545 letters specifying output formats, you do not have to remember whether
7546 unit size or format comes first; either order works. The output
7547 specifications @samp{4xw} and @samp{4wx} mean exactly the same thing.
7548 (However, the count @var{n} must come first; @samp{wx4} does not work.)
7550 Even though the unit size @var{u} is ignored for the formats @samp{s}
7551 and @samp{i}, you might still want to use a count @var{n}; for example,
7552 @samp{3i} specifies that you want to see three machine instructions,
7553 including any operands. For convenience, especially when used with
7554 the @code{display} command, the @samp{i} format also prints branch delay
7555 slot instructions, if any, beyond the count specified, which immediately
7556 follow the last instruction that is within the count. The command
7557 @code{disassemble} gives an alternative way of inspecting machine
7558 instructions; see @ref{Machine Code,,Source and Machine Code}.
7560 All the defaults for the arguments to @code{x} are designed to make it
7561 easy to continue scanning memory with minimal specifications each time
7562 you use @code{x}. For example, after you have inspected three machine
7563 instructions with @samp{x/3i @var{addr}}, you can inspect the next seven
7564 with just @samp{x/7}. If you use @key{RET} to repeat the @code{x} command,
7565 the repeat count @var{n} is used again; the other arguments default as
7566 for successive uses of @code{x}.
7568 When examining machine instructions, the instruction at current program
7569 counter is shown with a @code{=>} marker. For example:
7572 (@value{GDBP}) x/5i $pc-6
7573 0x804837f <main+11>: mov %esp,%ebp
7574 0x8048381 <main+13>: push %ecx
7575 0x8048382 <main+14>: sub $0x4,%esp
7576 => 0x8048385 <main+17>: movl $0x8048460,(%esp)
7577 0x804838c <main+24>: call 0x80482d4 <puts@@plt>
7580 @cindex @code{$_}, @code{$__}, and value history
7581 The addresses and contents printed by the @code{x} command are not saved
7582 in the value history because there is often too much of them and they
7583 would get in the way. Instead, @value{GDBN} makes these values available for
7584 subsequent use in expressions as values of the convenience variables
7585 @code{$_} and @code{$__}. After an @code{x} command, the last address
7586 examined is available for use in expressions in the convenience variable
7587 @code{$_}. The contents of that address, as examined, are available in
7588 the convenience variable @code{$__}.
7590 If the @code{x} command has a repeat count, the address and contents saved
7591 are from the last memory unit printed; this is not the same as the last
7592 address printed if several units were printed on the last line of output.
7594 @cindex remote memory comparison
7595 @cindex verify remote memory image
7596 When you are debugging a program running on a remote target machine
7597 (@pxref{Remote Debugging}), you may wish to verify the program's image in the
7598 remote machine's memory against the executable file you downloaded to
7599 the target. The @code{compare-sections} command is provided for such
7603 @kindex compare-sections
7604 @item compare-sections @r{[}@var{section-name}@r{]}
7605 Compare the data of a loadable section @var{section-name} in the
7606 executable file of the program being debugged with the same section in
7607 the remote machine's memory, and report any mismatches. With no
7608 arguments, compares all loadable sections. This command's
7609 availability depends on the target's support for the @code{"qCRC"}
7614 @section Automatic Display
7615 @cindex automatic display
7616 @cindex display of expressions
7618 If you find that you want to print the value of an expression frequently
7619 (to see how it changes), you might want to add it to the @dfn{automatic
7620 display list} so that @value{GDBN} prints its value each time your program stops.
7621 Each expression added to the list is given a number to identify it;
7622 to remove an expression from the list, you specify that number.
7623 The automatic display looks like this:
7627 3: bar[5] = (struct hack *) 0x3804
7631 This display shows item numbers, expressions and their current values. As with
7632 displays you request manually using @code{x} or @code{print}, you can
7633 specify the output format you prefer; in fact, @code{display} decides
7634 whether to use @code{print} or @code{x} depending your format
7635 specification---it uses @code{x} if you specify either the @samp{i}
7636 or @samp{s} format, or a unit size; otherwise it uses @code{print}.
7640 @item display @var{expr}
7641 Add the expression @var{expr} to the list of expressions to display
7642 each time your program stops. @xref{Expressions, ,Expressions}.
7644 @code{display} does not repeat if you press @key{RET} again after using it.
7646 @item display/@var{fmt} @var{expr}
7647 For @var{fmt} specifying only a display format and not a size or
7648 count, add the expression @var{expr} to the auto-display list but
7649 arrange to display it each time in the specified format @var{fmt}.
7650 @xref{Output Formats,,Output Formats}.
7652 @item display/@var{fmt} @var{addr}
7653 For @var{fmt} @samp{i} or @samp{s}, or including a unit-size or a
7654 number of units, add the expression @var{addr} as a memory address to
7655 be examined each time your program stops. Examining means in effect
7656 doing @samp{x/@var{fmt} @var{addr}}. @xref{Memory, ,Examining Memory}.
7659 For example, @samp{display/i $pc} can be helpful, to see the machine
7660 instruction about to be executed each time execution stops (@samp{$pc}
7661 is a common name for the program counter; @pxref{Registers, ,Registers}).
7664 @kindex delete display
7666 @item undisplay @var{dnums}@dots{}
7667 @itemx delete display @var{dnums}@dots{}
7668 Remove items from the list of expressions to display. Specify the
7669 numbers of the displays that you want affected with the command
7670 argument @var{dnums}. It can be a single display number, one of the
7671 numbers shown in the first field of the @samp{info display} display;
7672 or it could be a range of display numbers, as in @code{2-4}.
7674 @code{undisplay} does not repeat if you press @key{RET} after using it.
7675 (Otherwise you would just get the error @samp{No display number @dots{}}.)
7677 @kindex disable display
7678 @item disable display @var{dnums}@dots{}
7679 Disable the display of item numbers @var{dnums}. A disabled display
7680 item is not printed automatically, but is not forgotten. It may be
7681 enabled again later. Specify the numbers of the displays that you
7682 want affected with the command argument @var{dnums}. It can be a
7683 single display number, one of the numbers shown in the first field of
7684 the @samp{info display} display; or it could be a range of display
7685 numbers, as in @code{2-4}.
7687 @kindex enable display
7688 @item enable display @var{dnums}@dots{}
7689 Enable display of item numbers @var{dnums}. It becomes effective once
7690 again in auto display of its expression, until you specify otherwise.
7691 Specify the numbers of the displays that you want affected with the
7692 command argument @var{dnums}. It can be a single display number, one
7693 of the numbers shown in the first field of the @samp{info display}
7694 display; or it could be a range of display numbers, as in @code{2-4}.
7697 Display the current values of the expressions on the list, just as is
7698 done when your program stops.
7700 @kindex info display
7702 Print the list of expressions previously set up to display
7703 automatically, each one with its item number, but without showing the
7704 values. This includes disabled expressions, which are marked as such.
7705 It also includes expressions which would not be displayed right now
7706 because they refer to automatic variables not currently available.
7709 @cindex display disabled out of scope
7710 If a display expression refers to local variables, then it does not make
7711 sense outside the lexical context for which it was set up. Such an
7712 expression is disabled when execution enters a context where one of its
7713 variables is not defined. For example, if you give the command
7714 @code{display last_char} while inside a function with an argument
7715 @code{last_char}, @value{GDBN} displays this argument while your program
7716 continues to stop inside that function. When it stops elsewhere---where
7717 there is no variable @code{last_char}---the display is disabled
7718 automatically. The next time your program stops where @code{last_char}
7719 is meaningful, you can enable the display expression once again.
7721 @node Print Settings
7722 @section Print Settings
7724 @cindex format options
7725 @cindex print settings
7726 @value{GDBN} provides the following ways to control how arrays, structures,
7727 and symbols are printed.
7730 These settings are useful for debugging programs in any language:
7734 @item set print address
7735 @itemx set print address on
7736 @cindex print/don't print memory addresses
7737 @value{GDBN} prints memory addresses showing the location of stack
7738 traces, structure values, pointer values, breakpoints, and so forth,
7739 even when it also displays the contents of those addresses. The default
7740 is @code{on}. For example, this is what a stack frame display looks like with
7741 @code{set print address on}:
7746 #0 set_quotes (lq=0x34c78 "<<", rq=0x34c88 ">>")
7748 530 if (lquote != def_lquote)
7752 @item set print address off
7753 Do not print addresses when displaying their contents. For example,
7754 this is the same stack frame displayed with @code{set print address off}:
7758 (@value{GDBP}) set print addr off
7760 #0 set_quotes (lq="<<", rq=">>") at input.c:530
7761 530 if (lquote != def_lquote)
7765 You can use @samp{set print address off} to eliminate all machine
7766 dependent displays from the @value{GDBN} interface. For example, with
7767 @code{print address off}, you should get the same text for backtraces on
7768 all machines---whether or not they involve pointer arguments.
7771 @item show print address
7772 Show whether or not addresses are to be printed.
7775 When @value{GDBN} prints a symbolic address, it normally prints the
7776 closest earlier symbol plus an offset. If that symbol does not uniquely
7777 identify the address (for example, it is a name whose scope is a single
7778 source file), you may need to clarify. One way to do this is with
7779 @code{info line}, for example @samp{info line *0x4537}. Alternately,
7780 you can set @value{GDBN} to print the source file and line number when
7781 it prints a symbolic address:
7784 @item set print symbol-filename on
7785 @cindex source file and line of a symbol
7786 @cindex symbol, source file and line
7787 Tell @value{GDBN} to print the source file name and line number of a
7788 symbol in the symbolic form of an address.
7790 @item set print symbol-filename off
7791 Do not print source file name and line number of a symbol. This is the
7794 @item show print symbol-filename
7795 Show whether or not @value{GDBN} will print the source file name and
7796 line number of a symbol in the symbolic form of an address.
7799 Another situation where it is helpful to show symbol filenames and line
7800 numbers is when disassembling code; @value{GDBN} shows you the line
7801 number and source file that corresponds to each instruction.
7803 Also, you may wish to see the symbolic form only if the address being
7804 printed is reasonably close to the closest earlier symbol:
7807 @item set print max-symbolic-offset @var{max-offset}
7808 @cindex maximum value for offset of closest symbol
7809 Tell @value{GDBN} to only display the symbolic form of an address if the
7810 offset between the closest earlier symbol and the address is less than
7811 @var{max-offset}. The default is 0, which tells @value{GDBN}
7812 to always print the symbolic form of an address if any symbol precedes it.
7814 @item show print max-symbolic-offset
7815 Ask how large the maximum offset is that @value{GDBN} prints in a
7819 @cindex wild pointer, interpreting
7820 @cindex pointer, finding referent
7821 If you have a pointer and you are not sure where it points, try
7822 @samp{set print symbol-filename on}. Then you can determine the name
7823 and source file location of the variable where it points, using
7824 @samp{p/a @var{pointer}}. This interprets the address in symbolic form.
7825 For example, here @value{GDBN} shows that a variable @code{ptt} points
7826 at another variable @code{t}, defined in @file{hi2.c}:
7829 (@value{GDBP}) set print symbol-filename on
7830 (@value{GDBP}) p/a ptt
7831 $4 = 0xe008 <t in hi2.c>
7835 @emph{Warning:} For pointers that point to a local variable, @samp{p/a}
7836 does not show the symbol name and filename of the referent, even with
7837 the appropriate @code{set print} options turned on.
7840 Other settings control how different kinds of objects are printed:
7843 @item set print array
7844 @itemx set print array on
7845 @cindex pretty print arrays
7846 Pretty print arrays. This format is more convenient to read,
7847 but uses more space. The default is off.
7849 @item set print array off
7850 Return to compressed format for arrays.
7852 @item show print array
7853 Show whether compressed or pretty format is selected for displaying
7856 @cindex print array indexes
7857 @item set print array-indexes
7858 @itemx set print array-indexes on
7859 Print the index of each element when displaying arrays. May be more
7860 convenient to locate a given element in the array or quickly find the
7861 index of a given element in that printed array. The default is off.
7863 @item set print array-indexes off
7864 Stop printing element indexes when displaying arrays.
7866 @item show print array-indexes
7867 Show whether the index of each element is printed when displaying
7870 @item set print elements @var{number-of-elements}
7871 @cindex number of array elements to print
7872 @cindex limit on number of printed array elements
7873 Set a limit on how many elements of an array @value{GDBN} will print.
7874 If @value{GDBN} is printing a large array, it stops printing after it has
7875 printed the number of elements set by the @code{set print elements} command.
7876 This limit also applies to the display of strings.
7877 When @value{GDBN} starts, this limit is set to 200.
7878 Setting @var{number-of-elements} to zero means that the printing is unlimited.
7880 @item show print elements
7881 Display the number of elements of a large array that @value{GDBN} will print.
7882 If the number is 0, then the printing is unlimited.
7884 @item set print frame-arguments @var{value}
7885 @kindex set print frame-arguments
7886 @cindex printing frame argument values
7887 @cindex print all frame argument values
7888 @cindex print frame argument values for scalars only
7889 @cindex do not print frame argument values
7890 This command allows to control how the values of arguments are printed
7891 when the debugger prints a frame (@pxref{Frames}). The possible
7896 The values of all arguments are printed.
7899 Print the value of an argument only if it is a scalar. The value of more
7900 complex arguments such as arrays, structures, unions, etc, is replaced
7901 by @code{@dots{}}. This is the default. Here is an example where
7902 only scalar arguments are shown:
7905 #1 0x08048361 in call_me (i=3, s=@dots{}, ss=0xbf8d508c, u=@dots{}, e=green)
7910 None of the argument values are printed. Instead, the value of each argument
7911 is replaced by @code{@dots{}}. In this case, the example above now becomes:
7914 #1 0x08048361 in call_me (i=@dots{}, s=@dots{}, ss=@dots{}, u=@dots{}, e=@dots{})
7919 By default, only scalar arguments are printed. This command can be used
7920 to configure the debugger to print the value of all arguments, regardless
7921 of their type. However, it is often advantageous to not print the value
7922 of more complex parameters. For instance, it reduces the amount of
7923 information printed in each frame, making the backtrace more readable.
7924 Also, it improves performance when displaying Ada frames, because
7925 the computation of large arguments can sometimes be CPU-intensive,
7926 especially in large applications. Setting @code{print frame-arguments}
7927 to @code{scalars} (the default) or @code{none} avoids this computation,
7928 thus speeding up the display of each Ada frame.
7930 @item show print frame-arguments
7931 Show how the value of arguments should be displayed when printing a frame.
7933 @item set print repeats
7934 @cindex repeated array elements
7935 Set the threshold for suppressing display of repeated array
7936 elements. When the number of consecutive identical elements of an
7937 array exceeds the threshold, @value{GDBN} prints the string
7938 @code{"<repeats @var{n} times>"}, where @var{n} is the number of
7939 identical repetitions, instead of displaying the identical elements
7940 themselves. Setting the threshold to zero will cause all elements to
7941 be individually printed. The default threshold is 10.
7943 @item show print repeats
7944 Display the current threshold for printing repeated identical
7947 @item set print null-stop
7948 @cindex @sc{null} elements in arrays
7949 Cause @value{GDBN} to stop printing the characters of an array when the first
7950 @sc{null} is encountered. This is useful when large arrays actually
7951 contain only short strings.
7954 @item show print null-stop
7955 Show whether @value{GDBN} stops printing an array on the first
7956 @sc{null} character.
7958 @item set print pretty on
7959 @cindex print structures in indented form
7960 @cindex indentation in structure display
7961 Cause @value{GDBN} to print structures in an indented format with one member
7962 per line, like this:
7977 @item set print pretty off
7978 Cause @value{GDBN} to print structures in a compact format, like this:
7982 $1 = @{next = 0x0, flags = @{sweet = 1, sour = 1@}, \
7983 meat = 0x54 "Pork"@}
7988 This is the default format.
7990 @item show print pretty
7991 Show which format @value{GDBN} is using to print structures.
7993 @item set print sevenbit-strings on
7994 @cindex eight-bit characters in strings
7995 @cindex octal escapes in strings
7996 Print using only seven-bit characters; if this option is set,
7997 @value{GDBN} displays any eight-bit characters (in strings or
7998 character values) using the notation @code{\}@var{nnn}. This setting is
7999 best if you are working in English (@sc{ascii}) and you use the
8000 high-order bit of characters as a marker or ``meta'' bit.
8002 @item set print sevenbit-strings off
8003 Print full eight-bit characters. This allows the use of more
8004 international character sets, and is the default.
8006 @item show print sevenbit-strings
8007 Show whether or not @value{GDBN} is printing only seven-bit characters.
8009 @item set print union on
8010 @cindex unions in structures, printing
8011 Tell @value{GDBN} to print unions which are contained in structures
8012 and other unions. This is the default setting.
8014 @item set print union off
8015 Tell @value{GDBN} not to print unions which are contained in
8016 structures and other unions. @value{GDBN} will print @code{"@{...@}"}
8019 @item show print union
8020 Ask @value{GDBN} whether or not it will print unions which are contained in
8021 structures and other unions.
8023 For example, given the declarations
8026 typedef enum @{Tree, Bug@} Species;
8027 typedef enum @{Big_tree, Acorn, Seedling@} Tree_forms;
8028 typedef enum @{Caterpillar, Cocoon, Butterfly@}
8039 struct thing foo = @{Tree, @{Acorn@}@};
8043 with @code{set print union on} in effect @samp{p foo} would print
8046 $1 = @{it = Tree, form = @{tree = Acorn, bug = Cocoon@}@}
8050 and with @code{set print union off} in effect it would print
8053 $1 = @{it = Tree, form = @{...@}@}
8057 @code{set print union} affects programs written in C-like languages
8063 These settings are of interest when debugging C@t{++} programs:
8066 @cindex demangling C@t{++} names
8067 @item set print demangle
8068 @itemx set print demangle on
8069 Print C@t{++} names in their source form rather than in the encoded
8070 (``mangled'') form passed to the assembler and linker for type-safe
8071 linkage. The default is on.
8073 @item show print demangle
8074 Show whether C@t{++} names are printed in mangled or demangled form.
8076 @item set print asm-demangle
8077 @itemx set print asm-demangle on
8078 Print C@t{++} names in their source form rather than their mangled form, even
8079 in assembler code printouts such as instruction disassemblies.
8082 @item show print asm-demangle
8083 Show whether C@t{++} names in assembly listings are printed in mangled
8086 @cindex C@t{++} symbol decoding style
8087 @cindex symbol decoding style, C@t{++}
8088 @kindex set demangle-style
8089 @item set demangle-style @var{style}
8090 Choose among several encoding schemes used by different compilers to
8091 represent C@t{++} names. The choices for @var{style} are currently:
8095 Allow @value{GDBN} to choose a decoding style by inspecting your program.
8098 Decode based on the @sc{gnu} C@t{++} compiler (@code{g++}) encoding algorithm.
8099 This is the default.
8102 Decode based on the HP ANSI C@t{++} (@code{aCC}) encoding algorithm.
8105 Decode based on the Lucid C@t{++} compiler (@code{lcc}) encoding algorithm.
8108 Decode using the algorithm in the @cite{C@t{++} Annotated Reference Manual}.
8109 @strong{Warning:} this setting alone is not sufficient to allow
8110 debugging @code{cfront}-generated executables. @value{GDBN} would
8111 require further enhancement to permit that.
8114 If you omit @var{style}, you will see a list of possible formats.
8116 @item show demangle-style
8117 Display the encoding style currently in use for decoding C@t{++} symbols.
8119 @item set print object
8120 @itemx set print object on
8121 @cindex derived type of an object, printing
8122 @cindex display derived types
8123 When displaying a pointer to an object, identify the @emph{actual}
8124 (derived) type of the object rather than the @emph{declared} type, using
8125 the virtual function table.
8127 @item set print object off
8128 Display only the declared type of objects, without reference to the
8129 virtual function table. This is the default setting.
8131 @item show print object
8132 Show whether actual, or declared, object types are displayed.
8134 @item set print static-members
8135 @itemx set print static-members on
8136 @cindex static members of C@t{++} objects
8137 Print static members when displaying a C@t{++} object. The default is on.
8139 @item set print static-members off
8140 Do not print static members when displaying a C@t{++} object.
8142 @item show print static-members
8143 Show whether C@t{++} static members are printed or not.
8145 @item set print pascal_static-members
8146 @itemx set print pascal_static-members on
8147 @cindex static members of Pascal objects
8148 @cindex Pascal objects, static members display
8149 Print static members when displaying a Pascal object. The default is on.
8151 @item set print pascal_static-members off
8152 Do not print static members when displaying a Pascal object.
8154 @item show print pascal_static-members
8155 Show whether Pascal static members are printed or not.
8157 @c These don't work with HP ANSI C++ yet.
8158 @item set print vtbl
8159 @itemx set print vtbl on
8160 @cindex pretty print C@t{++} virtual function tables
8161 @cindex virtual functions (C@t{++}) display
8162 @cindex VTBL display
8163 Pretty print C@t{++} virtual function tables. The default is off.
8164 (The @code{vtbl} commands do not work on programs compiled with the HP
8165 ANSI C@t{++} compiler (@code{aCC}).)
8167 @item set print vtbl off
8168 Do not pretty print C@t{++} virtual function tables.
8170 @item show print vtbl
8171 Show whether C@t{++} virtual function tables are pretty printed, or not.
8174 @node Pretty Printing
8175 @section Pretty Printing
8177 @value{GDBN} provides a mechanism to allow pretty-printing of values using
8178 Python code. It greatly simplifies the display of complex objects. This
8179 mechanism works for both MI and the CLI.
8182 * Pretty-Printer Introduction:: Introduction to pretty-printers
8183 * Pretty-Printer Example:: An example pretty-printer
8184 * Pretty-Printer Commands:: Pretty-printer commands
8187 @node Pretty-Printer Introduction
8188 @subsection Pretty-Printer Introduction
8190 When @value{GDBN} prints a value, it first sees if there is a pretty-printer
8191 registered for the value. If there is then @value{GDBN} invokes the
8192 pretty-printer to print the value. Otherwise the value is printed normally.
8194 Pretty-printers are normally named. This makes them easy to manage.
8195 The @samp{info pretty-printer} command will list all the installed
8196 pretty-printers with their names.
8197 If a pretty-printer can handle multiple data types, then its
8198 @dfn{subprinters} are the printers for the individual data types.
8199 Each such subprinter has its own name.
8200 The format of the name is @var{printer-name};@var{subprinter-name}.
8202 Pretty-printers are installed by @dfn{registering} them with @value{GDBN}.
8203 Typically they are automatically loaded and registered when the corresponding
8204 debug information is loaded, thus making them available without having to
8205 do anything special.
8207 There are three places where a pretty-printer can be registered.
8211 Pretty-printers registered globally are available when debugging
8215 Pretty-printers registered with a program space are available only
8216 when debugging that program.
8217 @xref{Progspaces In Python}, for more details on program spaces in Python.
8220 Pretty-printers registered with an objfile are loaded and unloaded
8221 with the corresponding objfile (e.g., shared library).
8222 @xref{Objfiles In Python}, for more details on objfiles in Python.
8225 @xref{Selecting Pretty-Printers}, for further information on how
8226 pretty-printers are selected,
8228 @xref{Writing a Pretty-Printer}, for implementing pretty printers
8231 @node Pretty-Printer Example
8232 @subsection Pretty-Printer Example
8234 Here is how a C@t{++} @code{std::string} looks without a pretty-printer:
8237 (@value{GDBP}) print s
8239 static npos = 4294967295,
8241 <std::allocator<char>> = @{
8242 <__gnu_cxx::new_allocator<char>> = @{
8243 <No data fields>@}, <No data fields>
8245 members of std::basic_string<char, std::char_traits<char>,
8246 std::allocator<char> >::_Alloc_hider:
8247 _M_p = 0x804a014 "abcd"
8252 With a pretty-printer for @code{std::string} only the contents are printed:
8255 (@value{GDBP}) print s
8259 @node Pretty-Printer Commands
8260 @subsection Pretty-Printer Commands
8261 @cindex pretty-printer commands
8264 @kindex info pretty-printer
8265 @item info pretty-printer [@var{object-regexp} [@var{name-regexp}]]
8266 Print the list of installed pretty-printers.
8267 This includes disabled pretty-printers, which are marked as such.
8269 @var{object-regexp} is a regular expression matching the objects
8270 whose pretty-printers to list.
8271 Objects can be @code{global}, the program space's file
8272 (@pxref{Progspaces In Python}),
8273 and the object files within that program space (@pxref{Objfiles In Python}).
8274 @xref{Selecting Pretty-Printers}, for details on how @value{GDBN}
8275 looks up a printer from these three objects.
8277 @var{name-regexp} is a regular expression matching the name of the printers
8280 @kindex disable pretty-printer
8281 @item disable pretty-printer [@var{object-regexp} [@var{name-regexp}]]
8282 Disable pretty-printers matching @var{object-regexp} and @var{name-regexp}.
8283 A disabled pretty-printer is not forgotten, it may be enabled again later.
8285 @kindex enable pretty-printer
8286 @item enable pretty-printer [@var{object-regexp} [@var{name-regexp}]]
8287 Enable pretty-printers matching @var{object-regexp} and @var{name-regexp}.
8292 Suppose we have three pretty-printers installed: one from library1.so
8293 named @code{foo} that prints objects of type @code{foo}, and
8294 another from library2.so named @code{bar} that prints two types of objects,
8295 @code{bar1} and @code{bar2}.
8298 (gdb) info pretty-printer
8305 (gdb) info pretty-printer library2
8310 (gdb) disable pretty-printer library1
8312 2 of 3 printers enabled
8313 (gdb) info pretty-printer
8320 (gdb) disable pretty-printer library2 bar:bar1
8322 1 of 3 printers enabled
8323 (gdb) info pretty-printer library2
8330 (gdb) disable pretty-printer library2 bar
8332 0 of 3 printers enabled
8333 (gdb) info pretty-printer library2
8342 Note that for @code{bar} the entire printer can be disabled,
8343 as can each individual subprinter.
8346 @section Value History
8348 @cindex value history
8349 @cindex history of values printed by @value{GDBN}
8350 Values printed by the @code{print} command are saved in the @value{GDBN}
8351 @dfn{value history}. This allows you to refer to them in other expressions.
8352 Values are kept until the symbol table is re-read or discarded
8353 (for example with the @code{file} or @code{symbol-file} commands).
8354 When the symbol table changes, the value history is discarded,
8355 since the values may contain pointers back to the types defined in the
8360 @cindex history number
8361 The values printed are given @dfn{history numbers} by which you can
8362 refer to them. These are successive integers starting with one.
8363 @code{print} shows you the history number assigned to a value by
8364 printing @samp{$@var{num} = } before the value; here @var{num} is the
8367 To refer to any previous value, use @samp{$} followed by the value's
8368 history number. The way @code{print} labels its output is designed to
8369 remind you of this. Just @code{$} refers to the most recent value in
8370 the history, and @code{$$} refers to the value before that.
8371 @code{$$@var{n}} refers to the @var{n}th value from the end; @code{$$2}
8372 is the value just prior to @code{$$}, @code{$$1} is equivalent to
8373 @code{$$}, and @code{$$0} is equivalent to @code{$}.
8375 For example, suppose you have just printed a pointer to a structure and
8376 want to see the contents of the structure. It suffices to type
8382 If you have a chain of structures where the component @code{next} points
8383 to the next one, you can print the contents of the next one with this:
8390 You can print successive links in the chain by repeating this
8391 command---which you can do by just typing @key{RET}.
8393 Note that the history records values, not expressions. If the value of
8394 @code{x} is 4 and you type these commands:
8402 then the value recorded in the value history by the @code{print} command
8403 remains 4 even though the value of @code{x} has changed.
8408 Print the last ten values in the value history, with their item numbers.
8409 This is like @samp{p@ $$9} repeated ten times, except that @code{show
8410 values} does not change the history.
8412 @item show values @var{n}
8413 Print ten history values centered on history item number @var{n}.
8416 Print ten history values just after the values last printed. If no more
8417 values are available, @code{show values +} produces no display.
8420 Pressing @key{RET} to repeat @code{show values @var{n}} has exactly the
8421 same effect as @samp{show values +}.
8423 @node Convenience Vars
8424 @section Convenience Variables
8426 @cindex convenience variables
8427 @cindex user-defined variables
8428 @value{GDBN} provides @dfn{convenience variables} that you can use within
8429 @value{GDBN} to hold on to a value and refer to it later. These variables
8430 exist entirely within @value{GDBN}; they are not part of your program, and
8431 setting a convenience variable has no direct effect on further execution
8432 of your program. That is why you can use them freely.
8434 Convenience variables are prefixed with @samp{$}. Any name preceded by
8435 @samp{$} can be used for a convenience variable, unless it is one of
8436 the predefined machine-specific register names (@pxref{Registers, ,Registers}).
8437 (Value history references, in contrast, are @emph{numbers} preceded
8438 by @samp{$}. @xref{Value History, ,Value History}.)
8440 You can save a value in a convenience variable with an assignment
8441 expression, just as you would set a variable in your program.
8445 set $foo = *object_ptr
8449 would save in @code{$foo} the value contained in the object pointed to by
8452 Using a convenience variable for the first time creates it, but its
8453 value is @code{void} until you assign a new value. You can alter the
8454 value with another assignment at any time.
8456 Convenience variables have no fixed types. You can assign a convenience
8457 variable any type of value, including structures and arrays, even if
8458 that variable already has a value of a different type. The convenience
8459 variable, when used as an expression, has the type of its current value.
8462 @kindex show convenience
8463 @cindex show all user variables
8464 @item show convenience
8465 Print a list of convenience variables used so far, and their values.
8466 Abbreviated @code{show conv}.
8468 @kindex init-if-undefined
8469 @cindex convenience variables, initializing
8470 @item init-if-undefined $@var{variable} = @var{expression}
8471 Set a convenience variable if it has not already been set. This is useful
8472 for user-defined commands that keep some state. It is similar, in concept,
8473 to using local static variables with initializers in C (except that
8474 convenience variables are global). It can also be used to allow users to
8475 override default values used in a command script.
8477 If the variable is already defined then the expression is not evaluated so
8478 any side-effects do not occur.
8481 One of the ways to use a convenience variable is as a counter to be
8482 incremented or a pointer to be advanced. For example, to print
8483 a field from successive elements of an array of structures:
8487 print bar[$i++]->contents
8491 Repeat that command by typing @key{RET}.
8493 Some convenience variables are created automatically by @value{GDBN} and given
8494 values likely to be useful.
8497 @vindex $_@r{, convenience variable}
8499 The variable @code{$_} is automatically set by the @code{x} command to
8500 the last address examined (@pxref{Memory, ,Examining Memory}). Other
8501 commands which provide a default address for @code{x} to examine also
8502 set @code{$_} to that address; these commands include @code{info line}
8503 and @code{info breakpoint}. The type of @code{$_} is @code{void *}
8504 except when set by the @code{x} command, in which case it is a pointer
8505 to the type of @code{$__}.
8507 @vindex $__@r{, convenience variable}
8509 The variable @code{$__} is automatically set by the @code{x} command
8510 to the value found in the last address examined. Its type is chosen
8511 to match the format in which the data was printed.
8514 @vindex $_exitcode@r{, convenience variable}
8515 The variable @code{$_exitcode} is automatically set to the exit code when
8516 the program being debugged terminates.
8519 @vindex $_sdata@r{, inspect, convenience variable}
8520 The variable @code{$_sdata} contains extra collected static tracepoint
8521 data. @xref{Tracepoint Actions,,Tracepoint Action Lists}. Note that
8522 @code{$_sdata} could be empty, if not inspecting a trace buffer, or
8523 if extra static tracepoint data has not been collected.
8526 @vindex $_siginfo@r{, convenience variable}
8527 The variable @code{$_siginfo} contains extra signal information
8528 (@pxref{extra signal information}). Note that @code{$_siginfo}
8529 could be empty, if the application has not yet received any signals.
8530 For example, it will be empty before you execute the @code{run} command.
8533 @vindex $_tlb@r{, convenience variable}
8534 The variable @code{$_tlb} is automatically set when debugging
8535 applications running on MS-Windows in native mode or connected to
8536 gdbserver that supports the @code{qGetTIBAddr} request.
8537 @xref{General Query Packets}.
8538 This variable contains the address of the thread information block.
8542 On HP-UX systems, if you refer to a function or variable name that
8543 begins with a dollar sign, @value{GDBN} searches for a user or system
8544 name first, before it searches for a convenience variable.
8546 @cindex convenience functions
8547 @value{GDBN} also supplies some @dfn{convenience functions}. These
8548 have a syntax similar to convenience variables. A convenience
8549 function can be used in an expression just like an ordinary function;
8550 however, a convenience function is implemented internally to
8555 @kindex help function
8556 @cindex show all convenience functions
8557 Print a list of all convenience functions.
8564 You can refer to machine register contents, in expressions, as variables
8565 with names starting with @samp{$}. The names of registers are different
8566 for each machine; use @code{info registers} to see the names used on
8570 @kindex info registers
8571 @item info registers
8572 Print the names and values of all registers except floating-point
8573 and vector registers (in the selected stack frame).
8575 @kindex info all-registers
8576 @cindex floating point registers
8577 @item info all-registers
8578 Print the names and values of all registers, including floating-point
8579 and vector registers (in the selected stack frame).
8581 @item info registers @var{regname} @dots{}
8582 Print the @dfn{relativized} value of each specified register @var{regname}.
8583 As discussed in detail below, register values are normally relative to
8584 the selected stack frame. @var{regname} may be any register name valid on
8585 the machine you are using, with or without the initial @samp{$}.
8588 @cindex stack pointer register
8589 @cindex program counter register
8590 @cindex process status register
8591 @cindex frame pointer register
8592 @cindex standard registers
8593 @value{GDBN} has four ``standard'' register names that are available (in
8594 expressions) on most machines---whenever they do not conflict with an
8595 architecture's canonical mnemonics for registers. The register names
8596 @code{$pc} and @code{$sp} are used for the program counter register and
8597 the stack pointer. @code{$fp} is used for a register that contains a
8598 pointer to the current stack frame, and @code{$ps} is used for a
8599 register that contains the processor status. For example,
8600 you could print the program counter in hex with
8607 or print the instruction to be executed next with
8614 or add four to the stack pointer@footnote{This is a way of removing
8615 one word from the stack, on machines where stacks grow downward in
8616 memory (most machines, nowadays). This assumes that the innermost
8617 stack frame is selected; setting @code{$sp} is not allowed when other
8618 stack frames are selected. To pop entire frames off the stack,
8619 regardless of machine architecture, use @code{return};
8620 see @ref{Returning, ,Returning from a Function}.} with
8626 Whenever possible, these four standard register names are available on
8627 your machine even though the machine has different canonical mnemonics,
8628 so long as there is no conflict. The @code{info registers} command
8629 shows the canonical names. For example, on the SPARC, @code{info
8630 registers} displays the processor status register as @code{$psr} but you
8631 can also refer to it as @code{$ps}; and on x86-based machines @code{$ps}
8632 is an alias for the @sc{eflags} register.
8634 @value{GDBN} always considers the contents of an ordinary register as an
8635 integer when the register is examined in this way. Some machines have
8636 special registers which can hold nothing but floating point; these
8637 registers are considered to have floating point values. There is no way
8638 to refer to the contents of an ordinary register as floating point value
8639 (although you can @emph{print} it as a floating point value with
8640 @samp{print/f $@var{regname}}).
8642 Some registers have distinct ``raw'' and ``virtual'' data formats. This
8643 means that the data format in which the register contents are saved by
8644 the operating system is not the same one that your program normally
8645 sees. For example, the registers of the 68881 floating point
8646 coprocessor are always saved in ``extended'' (raw) format, but all C
8647 programs expect to work with ``double'' (virtual) format. In such
8648 cases, @value{GDBN} normally works with the virtual format only (the format
8649 that makes sense for your program), but the @code{info registers} command
8650 prints the data in both formats.
8652 @cindex SSE registers (x86)
8653 @cindex MMX registers (x86)
8654 Some machines have special registers whose contents can be interpreted
8655 in several different ways. For example, modern x86-based machines
8656 have SSE and MMX registers that can hold several values packed
8657 together in several different formats. @value{GDBN} refers to such
8658 registers in @code{struct} notation:
8661 (@value{GDBP}) print $xmm1
8663 v4_float = @{0, 3.43859137e-038, 1.54142831e-044, 1.821688e-044@},
8664 v2_double = @{9.92129282474342e-303, 2.7585945287983262e-313@},
8665 v16_int8 = "\000\000\000\000\3706;\001\v\000\000\000\r\000\000",
8666 v8_int16 = @{0, 0, 14072, 315, 11, 0, 13, 0@},
8667 v4_int32 = @{0, 20657912, 11, 13@},
8668 v2_int64 = @{88725056443645952, 55834574859@},
8669 uint128 = 0x0000000d0000000b013b36f800000000
8674 To set values of such registers, you need to tell @value{GDBN} which
8675 view of the register you wish to change, as if you were assigning
8676 value to a @code{struct} member:
8679 (@value{GDBP}) set $xmm1.uint128 = 0x000000000000000000000000FFFFFFFF
8682 Normally, register values are relative to the selected stack frame
8683 (@pxref{Selection, ,Selecting a Frame}). This means that you get the
8684 value that the register would contain if all stack frames farther in
8685 were exited and their saved registers restored. In order to see the
8686 true contents of hardware registers, you must select the innermost
8687 frame (with @samp{frame 0}).
8689 However, @value{GDBN} must deduce where registers are saved, from the machine
8690 code generated by your compiler. If some registers are not saved, or if
8691 @value{GDBN} is unable to locate the saved registers, the selected stack
8692 frame makes no difference.
8694 @node Floating Point Hardware
8695 @section Floating Point Hardware
8696 @cindex floating point
8698 Depending on the configuration, @value{GDBN} may be able to give
8699 you more information about the status of the floating point hardware.
8704 Display hardware-dependent information about the floating
8705 point unit. The exact contents and layout vary depending on the
8706 floating point chip. Currently, @samp{info float} is supported on
8707 the ARM and x86 machines.
8711 @section Vector Unit
8714 Depending on the configuration, @value{GDBN} may be able to give you
8715 more information about the status of the vector unit.
8720 Display information about the vector unit. The exact contents and
8721 layout vary depending on the hardware.
8724 @node OS Information
8725 @section Operating System Auxiliary Information
8726 @cindex OS information
8728 @value{GDBN} provides interfaces to useful OS facilities that can help
8729 you debug your program.
8731 @cindex @code{ptrace} system call
8732 @cindex @code{struct user} contents
8733 When @value{GDBN} runs on a @dfn{Posix system} (such as GNU or Unix
8734 machines), it interfaces with the inferior via the @code{ptrace}
8735 system call. The operating system creates a special sata structure,
8736 called @code{struct user}, for this interface. You can use the
8737 command @code{info udot} to display the contents of this data
8743 Display the contents of the @code{struct user} maintained by the OS
8744 kernel for the program being debugged. @value{GDBN} displays the
8745 contents of @code{struct user} as a list of hex numbers, similar to
8746 the @code{examine} command.
8749 @cindex auxiliary vector
8750 @cindex vector, auxiliary
8751 Some operating systems supply an @dfn{auxiliary vector} to programs at
8752 startup. This is akin to the arguments and environment that you
8753 specify for a program, but contains a system-dependent variety of
8754 binary values that tell system libraries important details about the
8755 hardware, operating system, and process. Each value's purpose is
8756 identified by an integer tag; the meanings are well-known but system-specific.
8757 Depending on the configuration and operating system facilities,
8758 @value{GDBN} may be able to show you this information. For remote
8759 targets, this functionality may further depend on the remote stub's
8760 support of the @samp{qXfer:auxv:read} packet, see
8761 @ref{qXfer auxiliary vector read}.
8766 Display the auxiliary vector of the inferior, which can be either a
8767 live process or a core dump file. @value{GDBN} prints each tag value
8768 numerically, and also shows names and text descriptions for recognized
8769 tags. Some values in the vector are numbers, some bit masks, and some
8770 pointers to strings or other data. @value{GDBN} displays each value in the
8771 most appropriate form for a recognized tag, and in hexadecimal for
8772 an unrecognized tag.
8775 On some targets, @value{GDBN} can access operating-system-specific information
8776 and display it to user, without interpretation. For remote targets,
8777 this functionality depends on the remote stub's support of the
8778 @samp{qXfer:osdata:read} packet, see @ref{qXfer osdata read}.
8783 List the types of OS information available for the target. If the
8784 target does not return a list of possible types, this command will
8787 @kindex info os processes
8788 @item info os processes
8789 Display the list of processes on the target. For each process,
8790 @value{GDBN} prints the process identifier, the name of the user, and
8791 the command corresponding to the process.
8794 @node Memory Region Attributes
8795 @section Memory Region Attributes
8796 @cindex memory region attributes
8798 @dfn{Memory region attributes} allow you to describe special handling
8799 required by regions of your target's memory. @value{GDBN} uses
8800 attributes to determine whether to allow certain types of memory
8801 accesses; whether to use specific width accesses; and whether to cache
8802 target memory. By default the description of memory regions is
8803 fetched from the target (if the current target supports this), but the
8804 user can override the fetched regions.
8806 Defined memory regions can be individually enabled and disabled. When a
8807 memory region is disabled, @value{GDBN} uses the default attributes when
8808 accessing memory in that region. Similarly, if no memory regions have
8809 been defined, @value{GDBN} uses the default attributes when accessing
8812 When a memory region is defined, it is given a number to identify it;
8813 to enable, disable, or remove a memory region, you specify that number.
8817 @item mem @var{lower} @var{upper} @var{attributes}@dots{}
8818 Define a memory region bounded by @var{lower} and @var{upper} with
8819 attributes @var{attributes}@dots{}, and add it to the list of regions
8820 monitored by @value{GDBN}. Note that @var{upper} == 0 is a special
8821 case: it is treated as the target's maximum memory address.
8822 (0xffff on 16 bit targets, 0xffffffff on 32 bit targets, etc.)
8825 Discard any user changes to the memory regions and use target-supplied
8826 regions, if available, or no regions if the target does not support.
8829 @item delete mem @var{nums}@dots{}
8830 Remove memory regions @var{nums}@dots{} from the list of regions
8831 monitored by @value{GDBN}.
8834 @item disable mem @var{nums}@dots{}
8835 Disable monitoring of memory regions @var{nums}@dots{}.
8836 A disabled memory region is not forgotten.
8837 It may be enabled again later.
8840 @item enable mem @var{nums}@dots{}
8841 Enable monitoring of memory regions @var{nums}@dots{}.
8845 Print a table of all defined memory regions, with the following columns
8849 @item Memory Region Number
8850 @item Enabled or Disabled.
8851 Enabled memory regions are marked with @samp{y}.
8852 Disabled memory regions are marked with @samp{n}.
8855 The address defining the inclusive lower bound of the memory region.
8858 The address defining the exclusive upper bound of the memory region.
8861 The list of attributes set for this memory region.
8866 @subsection Attributes
8868 @subsubsection Memory Access Mode
8869 The access mode attributes set whether @value{GDBN} may make read or
8870 write accesses to a memory region.
8872 While these attributes prevent @value{GDBN} from performing invalid
8873 memory accesses, they do nothing to prevent the target system, I/O DMA,
8874 etc.@: from accessing memory.
8878 Memory is read only.
8880 Memory is write only.
8882 Memory is read/write. This is the default.
8885 @subsubsection Memory Access Size
8886 The access size attribute tells @value{GDBN} to use specific sized
8887 accesses in the memory region. Often memory mapped device registers
8888 require specific sized accesses. If no access size attribute is
8889 specified, @value{GDBN} may use accesses of any size.
8893 Use 8 bit memory accesses.
8895 Use 16 bit memory accesses.
8897 Use 32 bit memory accesses.
8899 Use 64 bit memory accesses.
8902 @c @subsubsection Hardware/Software Breakpoints
8903 @c The hardware/software breakpoint attributes set whether @value{GDBN}
8904 @c will use hardware or software breakpoints for the internal breakpoints
8905 @c used by the step, next, finish, until, etc. commands.
8909 @c Always use hardware breakpoints
8910 @c @item swbreak (default)
8913 @subsubsection Data Cache
8914 The data cache attributes set whether @value{GDBN} will cache target
8915 memory. While this generally improves performance by reducing debug
8916 protocol overhead, it can lead to incorrect results because @value{GDBN}
8917 does not know about volatile variables or memory mapped device
8922 Enable @value{GDBN} to cache target memory.
8924 Disable @value{GDBN} from caching target memory. This is the default.
8927 @subsection Memory Access Checking
8928 @value{GDBN} can be instructed to refuse accesses to memory that is
8929 not explicitly described. This can be useful if accessing such
8930 regions has undesired effects for a specific target, or to provide
8931 better error checking. The following commands control this behaviour.
8934 @kindex set mem inaccessible-by-default
8935 @item set mem inaccessible-by-default [on|off]
8936 If @code{on} is specified, make @value{GDBN} treat memory not
8937 explicitly described by the memory ranges as non-existent and refuse accesses
8938 to such memory. The checks are only performed if there's at least one
8939 memory range defined. If @code{off} is specified, make @value{GDBN}
8940 treat the memory not explicitly described by the memory ranges as RAM.
8941 The default value is @code{on}.
8942 @kindex show mem inaccessible-by-default
8943 @item show mem inaccessible-by-default
8944 Show the current handling of accesses to unknown memory.
8948 @c @subsubsection Memory Write Verification
8949 @c The memory write verification attributes set whether @value{GDBN}
8950 @c will re-reads data after each write to verify the write was successful.
8954 @c @item noverify (default)
8957 @node Dump/Restore Files
8958 @section Copy Between Memory and a File
8959 @cindex dump/restore files
8960 @cindex append data to a file
8961 @cindex dump data to a file
8962 @cindex restore data from a file
8964 You can use the commands @code{dump}, @code{append}, and
8965 @code{restore} to copy data between target memory and a file. The
8966 @code{dump} and @code{append} commands write data to a file, and the
8967 @code{restore} command reads data from a file back into the inferior's
8968 memory. Files may be in binary, Motorola S-record, Intel hex, or
8969 Tektronix Hex format; however, @value{GDBN} can only append to binary
8975 @item dump @r{[}@var{format}@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
8976 @itemx dump @r{[}@var{format}@r{]} value @var{filename} @var{expr}
8977 Dump the contents of memory from @var{start_addr} to @var{end_addr},
8978 or the value of @var{expr}, to @var{filename} in the given format.
8980 The @var{format} parameter may be any one of:
8987 Motorola S-record format.
8989 Tektronix Hex format.
8992 @value{GDBN} uses the same definitions of these formats as the
8993 @sc{gnu} binary utilities, like @samp{objdump} and @samp{objcopy}. If
8994 @var{format} is omitted, @value{GDBN} dumps the data in raw binary
8998 @item append @r{[}binary@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
8999 @itemx append @r{[}binary@r{]} value @var{filename} @var{expr}
9000 Append the contents of memory from @var{start_addr} to @var{end_addr},
9001 or the value of @var{expr}, to the file @var{filename}, in raw binary form.
9002 (@value{GDBN} can only append data to files in raw binary form.)
9005 @item restore @var{filename} @r{[}binary@r{]} @var{bias} @var{start} @var{end}
9006 Restore the contents of file @var{filename} into memory. The
9007 @code{restore} command can automatically recognize any known @sc{bfd}
9008 file format, except for raw binary. To restore a raw binary file you
9009 must specify the optional keyword @code{binary} after the filename.
9011 If @var{bias} is non-zero, its value will be added to the addresses
9012 contained in the file. Binary files always start at address zero, so
9013 they will be restored at address @var{bias}. Other bfd files have
9014 a built-in location; they will be restored at offset @var{bias}
9017 If @var{start} and/or @var{end} are non-zero, then only data between
9018 file offset @var{start} and file offset @var{end} will be restored.
9019 These offsets are relative to the addresses in the file, before
9020 the @var{bias} argument is applied.
9024 @node Core File Generation
9025 @section How to Produce a Core File from Your Program
9026 @cindex dump core from inferior
9028 A @dfn{core file} or @dfn{core dump} is a file that records the memory
9029 image of a running process and its process status (register values
9030 etc.). Its primary use is post-mortem debugging of a program that
9031 crashed while it ran outside a debugger. A program that crashes
9032 automatically produces a core file, unless this feature is disabled by
9033 the user. @xref{Files}, for information on invoking @value{GDBN} in
9034 the post-mortem debugging mode.
9036 Occasionally, you may wish to produce a core file of the program you
9037 are debugging in order to preserve a snapshot of its state.
9038 @value{GDBN} has a special command for that.
9042 @kindex generate-core-file
9043 @item generate-core-file [@var{file}]
9044 @itemx gcore [@var{file}]
9045 Produce a core dump of the inferior process. The optional argument
9046 @var{file} specifies the file name where to put the core dump. If not
9047 specified, the file name defaults to @file{core.@var{pid}}, where
9048 @var{pid} is the inferior process ID.
9050 Note that this command is implemented only for some systems (as of
9051 this writing, @sc{gnu}/Linux, FreeBSD, Solaris, Unixware, and S390).
9054 @node Character Sets
9055 @section Character Sets
9056 @cindex character sets
9058 @cindex translating between character sets
9059 @cindex host character set
9060 @cindex target character set
9062 If the program you are debugging uses a different character set to
9063 represent characters and strings than the one @value{GDBN} uses itself,
9064 @value{GDBN} can automatically translate between the character sets for
9065 you. The character set @value{GDBN} uses we call the @dfn{host
9066 character set}; the one the inferior program uses we call the
9067 @dfn{target character set}.
9069 For example, if you are running @value{GDBN} on a @sc{gnu}/Linux system, which
9070 uses the ISO Latin 1 character set, but you are using @value{GDBN}'s
9071 remote protocol (@pxref{Remote Debugging}) to debug a program
9072 running on an IBM mainframe, which uses the @sc{ebcdic} character set,
9073 then the host character set is Latin-1, and the target character set is
9074 @sc{ebcdic}. If you give @value{GDBN} the command @code{set
9075 target-charset EBCDIC-US}, then @value{GDBN} translates between
9076 @sc{ebcdic} and Latin 1 as you print character or string values, or use
9077 character and string literals in expressions.
9079 @value{GDBN} has no way to automatically recognize which character set
9080 the inferior program uses; you must tell it, using the @code{set
9081 target-charset} command, described below.
9083 Here are the commands for controlling @value{GDBN}'s character set
9087 @item set target-charset @var{charset}
9088 @kindex set target-charset
9089 Set the current target character set to @var{charset}. To display the
9090 list of supported target character sets, type
9091 @kbd{@w{set target-charset @key{TAB}@key{TAB}}}.
9093 @item set host-charset @var{charset}
9094 @kindex set host-charset
9095 Set the current host character set to @var{charset}.
9097 By default, @value{GDBN} uses a host character set appropriate to the
9098 system it is running on; you can override that default using the
9099 @code{set host-charset} command. On some systems, @value{GDBN} cannot
9100 automatically determine the appropriate host character set. In this
9101 case, @value{GDBN} uses @samp{UTF-8}.
9103 @value{GDBN} can only use certain character sets as its host character
9104 set. If you type @kbd{@w{set host-charset @key{TAB}@key{TAB}}},
9105 @value{GDBN} will list the host character sets it supports.
9107 @item set charset @var{charset}
9109 Set the current host and target character sets to @var{charset}. As
9110 above, if you type @kbd{@w{set charset @key{TAB}@key{TAB}}},
9111 @value{GDBN} will list the names of the character sets that can be used
9112 for both host and target.
9115 @kindex show charset
9116 Show the names of the current host and target character sets.
9118 @item show host-charset
9119 @kindex show host-charset
9120 Show the name of the current host character set.
9122 @item show target-charset
9123 @kindex show target-charset
9124 Show the name of the current target character set.
9126 @item set target-wide-charset @var{charset}
9127 @kindex set target-wide-charset
9128 Set the current target's wide character set to @var{charset}. This is
9129 the character set used by the target's @code{wchar_t} type. To
9130 display the list of supported wide character sets, type
9131 @kbd{@w{set target-wide-charset @key{TAB}@key{TAB}}}.
9133 @item show target-wide-charset
9134 @kindex show target-wide-charset
9135 Show the name of the current target's wide character set.
9138 Here is an example of @value{GDBN}'s character set support in action.
9139 Assume that the following source code has been placed in the file
9140 @file{charset-test.c}:
9146 = @{72, 101, 108, 108, 111, 44, 32, 119,
9147 111, 114, 108, 100, 33, 10, 0@};
9148 char ibm1047_hello[]
9149 = @{200, 133, 147, 147, 150, 107, 64, 166,
9150 150, 153, 147, 132, 90, 37, 0@};
9154 printf ("Hello, world!\n");
9158 In this program, @code{ascii_hello} and @code{ibm1047_hello} are arrays
9159 containing the string @samp{Hello, world!} followed by a newline,
9160 encoded in the @sc{ascii} and @sc{ibm1047} character sets.
9162 We compile the program, and invoke the debugger on it:
9165 $ gcc -g charset-test.c -o charset-test
9166 $ gdb -nw charset-test
9167 GNU gdb 2001-12-19-cvs
9168 Copyright 2001 Free Software Foundation, Inc.
9173 We can use the @code{show charset} command to see what character sets
9174 @value{GDBN} is currently using to interpret and display characters and
9178 (@value{GDBP}) show charset
9179 The current host and target character set is `ISO-8859-1'.
9183 For the sake of printing this manual, let's use @sc{ascii} as our
9184 initial character set:
9186 (@value{GDBP}) set charset ASCII
9187 (@value{GDBP}) show charset
9188 The current host and target character set is `ASCII'.
9192 Let's assume that @sc{ascii} is indeed the correct character set for our
9193 host system --- in other words, let's assume that if @value{GDBN} prints
9194 characters using the @sc{ascii} character set, our terminal will display
9195 them properly. Since our current target character set is also
9196 @sc{ascii}, the contents of @code{ascii_hello} print legibly:
9199 (@value{GDBP}) print ascii_hello
9200 $1 = 0x401698 "Hello, world!\n"
9201 (@value{GDBP}) print ascii_hello[0]
9206 @value{GDBN} uses the target character set for character and string
9207 literals you use in expressions:
9210 (@value{GDBP}) print '+'
9215 The @sc{ascii} character set uses the number 43 to encode the @samp{+}
9218 @value{GDBN} relies on the user to tell it which character set the
9219 target program uses. If we print @code{ibm1047_hello} while our target
9220 character set is still @sc{ascii}, we get jibberish:
9223 (@value{GDBP}) print ibm1047_hello
9224 $4 = 0x4016a8 "\310\205\223\223\226k@@\246\226\231\223\204Z%"
9225 (@value{GDBP}) print ibm1047_hello[0]
9230 If we invoke the @code{set target-charset} followed by @key{TAB}@key{TAB},
9231 @value{GDBN} tells us the character sets it supports:
9234 (@value{GDBP}) set target-charset
9235 ASCII EBCDIC-US IBM1047 ISO-8859-1
9236 (@value{GDBP}) set target-charset
9239 We can select @sc{ibm1047} as our target character set, and examine the
9240 program's strings again. Now the @sc{ascii} string is wrong, but
9241 @value{GDBN} translates the contents of @code{ibm1047_hello} from the
9242 target character set, @sc{ibm1047}, to the host character set,
9243 @sc{ascii}, and they display correctly:
9246 (@value{GDBP}) set target-charset IBM1047
9247 (@value{GDBP}) show charset
9248 The current host character set is `ASCII'.
9249 The current target character set is `IBM1047'.
9250 (@value{GDBP}) print ascii_hello
9251 $6 = 0x401698 "\110\145%%?\054\040\167?\162%\144\041\012"
9252 (@value{GDBP}) print ascii_hello[0]
9254 (@value{GDBP}) print ibm1047_hello
9255 $8 = 0x4016a8 "Hello, world!\n"
9256 (@value{GDBP}) print ibm1047_hello[0]
9261 As above, @value{GDBN} uses the target character set for character and
9262 string literals you use in expressions:
9265 (@value{GDBP}) print '+'
9270 The @sc{ibm1047} character set uses the number 78 to encode the @samp{+}
9273 @node Caching Remote Data
9274 @section Caching Data of Remote Targets
9275 @cindex caching data of remote targets
9277 @value{GDBN} caches data exchanged between the debugger and a
9278 remote target (@pxref{Remote Debugging}). Such caching generally improves
9279 performance, because it reduces the overhead of the remote protocol by
9280 bundling memory reads and writes into large chunks. Unfortunately, simply
9281 caching everything would lead to incorrect results, since @value{GDBN}
9282 does not necessarily know anything about volatile values, memory-mapped I/O
9283 addresses, etc. Furthermore, in non-stop mode (@pxref{Non-Stop Mode})
9284 memory can be changed @emph{while} a gdb command is executing.
9285 Therefore, by default, @value{GDBN} only caches data
9286 known to be on the stack@footnote{In non-stop mode, it is moderately
9287 rare for a running thread to modify the stack of a stopped thread
9288 in a way that would interfere with a backtrace, and caching of
9289 stack reads provides a significant speed up of remote backtraces.}.
9290 Other regions of memory can be explicitly marked as
9291 cacheable; see @pxref{Memory Region Attributes}.
9294 @kindex set remotecache
9295 @item set remotecache on
9296 @itemx set remotecache off
9297 This option no longer does anything; it exists for compatibility
9300 @kindex show remotecache
9301 @item show remotecache
9302 Show the current state of the obsolete remotecache flag.
9304 @kindex set stack-cache
9305 @item set stack-cache on
9306 @itemx set stack-cache off
9307 Enable or disable caching of stack accesses. When @code{ON}, use
9308 caching. By default, this option is @code{ON}.
9310 @kindex show stack-cache
9311 @item show stack-cache
9312 Show the current state of data caching for memory accesses.
9315 @item info dcache @r{[}line@r{]}
9316 Print the information about the data cache performance. The
9317 information displayed includes the dcache width and depth, and for
9318 each cache line, its number, address, and how many times it was
9319 referenced. This command is useful for debugging the data cache
9322 If a line number is specified, the contents of that line will be
9326 @node Searching Memory
9327 @section Search Memory
9328 @cindex searching memory
9330 Memory can be searched for a particular sequence of bytes with the
9331 @code{find} command.
9335 @item find @r{[}/@var{sn}@r{]} @var{start_addr}, +@var{len}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
9336 @itemx find @r{[}/@var{sn}@r{]} @var{start_addr}, @var{end_addr}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
9337 Search memory for the sequence of bytes specified by @var{val1}, @var{val2},
9338 etc. The search begins at address @var{start_addr} and continues for either
9339 @var{len} bytes or through to @var{end_addr} inclusive.
9342 @var{s} and @var{n} are optional parameters.
9343 They may be specified in either order, apart or together.
9346 @item @var{s}, search query size
9347 The size of each search query value.
9353 halfwords (two bytes)
9357 giant words (eight bytes)
9360 All values are interpreted in the current language.
9361 This means, for example, that if the current source language is C/C@t{++}
9362 then searching for the string ``hello'' includes the trailing '\0'.
9364 If the value size is not specified, it is taken from the
9365 value's type in the current language.
9366 This is useful when one wants to specify the search
9367 pattern as a mixture of types.
9368 Note that this means, for example, that in the case of C-like languages
9369 a search for an untyped 0x42 will search for @samp{(int) 0x42}
9370 which is typically four bytes.
9372 @item @var{n}, maximum number of finds
9373 The maximum number of matches to print. The default is to print all finds.
9376 You can use strings as search values. Quote them with double-quotes
9378 The string value is copied into the search pattern byte by byte,
9379 regardless of the endianness of the target and the size specification.
9381 The address of each match found is printed as well as a count of the
9382 number of matches found.
9384 The address of the last value found is stored in convenience variable
9386 A count of the number of matches is stored in @samp{$numfound}.
9388 For example, if stopped at the @code{printf} in this function:
9394 static char hello[] = "hello-hello";
9395 static struct @{ char c; short s; int i; @}
9396 __attribute__ ((packed)) mixed
9397 = @{ 'c', 0x1234, 0x87654321 @};
9398 printf ("%s\n", hello);
9403 you get during debugging:
9406 (gdb) find &hello[0], +sizeof(hello), "hello"
9407 0x804956d <hello.1620+6>
9409 (gdb) find &hello[0], +sizeof(hello), 'h', 'e', 'l', 'l', 'o'
9410 0x8049567 <hello.1620>
9411 0x804956d <hello.1620+6>
9413 (gdb) find /b1 &hello[0], +sizeof(hello), 'h', 0x65, 'l'
9414 0x8049567 <hello.1620>
9416 (gdb) find &mixed, +sizeof(mixed), (char) 'c', (short) 0x1234, (int) 0x87654321
9417 0x8049560 <mixed.1625>
9419 (gdb) print $numfound
9422 $2 = (void *) 0x8049560
9425 @node Optimized Code
9426 @chapter Debugging Optimized Code
9427 @cindex optimized code, debugging
9428 @cindex debugging optimized code
9430 Almost all compilers support optimization. With optimization
9431 disabled, the compiler generates assembly code that corresponds
9432 directly to your source code, in a simplistic way. As the compiler
9433 applies more powerful optimizations, the generated assembly code
9434 diverges from your original source code. With help from debugging
9435 information generated by the compiler, @value{GDBN} can map from
9436 the running program back to constructs from your original source.
9438 @value{GDBN} is more accurate with optimization disabled. If you
9439 can recompile without optimization, it is easier to follow the
9440 progress of your program during debugging. But, there are many cases
9441 where you may need to debug an optimized version.
9443 When you debug a program compiled with @samp{-g -O}, remember that the
9444 optimizer has rearranged your code; the debugger shows you what is
9445 really there. Do not be too surprised when the execution path does not
9446 exactly match your source file! An extreme example: if you define a
9447 variable, but never use it, @value{GDBN} never sees that
9448 variable---because the compiler optimizes it out of existence.
9450 Some things do not work as well with @samp{-g -O} as with just
9451 @samp{-g}, particularly on machines with instruction scheduling. If in
9452 doubt, recompile with @samp{-g} alone, and if this fixes the problem,
9453 please report it to us as a bug (including a test case!).
9454 @xref{Variables}, for more information about debugging optimized code.
9457 * Inline Functions:: How @value{GDBN} presents inlining
9460 @node Inline Functions
9461 @section Inline Functions
9462 @cindex inline functions, debugging
9464 @dfn{Inlining} is an optimization that inserts a copy of the function
9465 body directly at each call site, instead of jumping to a shared
9466 routine. @value{GDBN} displays inlined functions just like
9467 non-inlined functions. They appear in backtraces. You can view their
9468 arguments and local variables, step into them with @code{step}, skip
9469 them with @code{next}, and escape from them with @code{finish}.
9470 You can check whether a function was inlined by using the
9471 @code{info frame} command.
9473 For @value{GDBN} to support inlined functions, the compiler must
9474 record information about inlining in the debug information ---
9475 @value{NGCC} using the @sc{dwarf 2} format does this, and several
9476 other compilers do also. @value{GDBN} only supports inlined functions
9477 when using @sc{dwarf 2}. Versions of @value{NGCC} before 4.1
9478 do not emit two required attributes (@samp{DW_AT_call_file} and
9479 @samp{DW_AT_call_line}); @value{GDBN} does not display inlined
9480 function calls with earlier versions of @value{NGCC}. It instead
9481 displays the arguments and local variables of inlined functions as
9482 local variables in the caller.
9484 The body of an inlined function is directly included at its call site;
9485 unlike a non-inlined function, there are no instructions devoted to
9486 the call. @value{GDBN} still pretends that the call site and the
9487 start of the inlined function are different instructions. Stepping to
9488 the call site shows the call site, and then stepping again shows
9489 the first line of the inlined function, even though no additional
9490 instructions are executed.
9492 This makes source-level debugging much clearer; you can see both the
9493 context of the call and then the effect of the call. Only stepping by
9494 a single instruction using @code{stepi} or @code{nexti} does not do
9495 this; single instruction steps always show the inlined body.
9497 There are some ways that @value{GDBN} does not pretend that inlined
9498 function calls are the same as normal calls:
9502 You cannot set breakpoints on inlined functions. @value{GDBN}
9503 either reports that there is no symbol with that name, or else sets the
9504 breakpoint only on non-inlined copies of the function. This limitation
9505 will be removed in a future version of @value{GDBN}; until then,
9506 set a breakpoint by line number on the first line of the inlined
9510 Setting breakpoints at the call site of an inlined function may not
9511 work, because the call site does not contain any code. @value{GDBN}
9512 may incorrectly move the breakpoint to the next line of the enclosing
9513 function, after the call. This limitation will be removed in a future
9514 version of @value{GDBN}; until then, set a breakpoint on an earlier line
9515 or inside the inlined function instead.
9518 @value{GDBN} cannot locate the return value of inlined calls after
9519 using the @code{finish} command. This is a limitation of compiler-generated
9520 debugging information; after @code{finish}, you can step to the next line
9521 and print a variable where your program stored the return value.
9527 @chapter C Preprocessor Macros
9529 Some languages, such as C and C@t{++}, provide a way to define and invoke
9530 ``preprocessor macros'' which expand into strings of tokens.
9531 @value{GDBN} can evaluate expressions containing macro invocations, show
9532 the result of macro expansion, and show a macro's definition, including
9533 where it was defined.
9535 You may need to compile your program specially to provide @value{GDBN}
9536 with information about preprocessor macros. Most compilers do not
9537 include macros in their debugging information, even when you compile
9538 with the @option{-g} flag. @xref{Compilation}.
9540 A program may define a macro at one point, remove that definition later,
9541 and then provide a different definition after that. Thus, at different
9542 points in the program, a macro may have different definitions, or have
9543 no definition at all. If there is a current stack frame, @value{GDBN}
9544 uses the macros in scope at that frame's source code line. Otherwise,
9545 @value{GDBN} uses the macros in scope at the current listing location;
9548 Whenever @value{GDBN} evaluates an expression, it always expands any
9549 macro invocations present in the expression. @value{GDBN} also provides
9550 the following commands for working with macros explicitly.
9554 @kindex macro expand
9555 @cindex macro expansion, showing the results of preprocessor
9556 @cindex preprocessor macro expansion, showing the results of
9557 @cindex expanding preprocessor macros
9558 @item macro expand @var{expression}
9559 @itemx macro exp @var{expression}
9560 Show the results of expanding all preprocessor macro invocations in
9561 @var{expression}. Since @value{GDBN} simply expands macros, but does
9562 not parse the result, @var{expression} need not be a valid expression;
9563 it can be any string of tokens.
9566 @item macro expand-once @var{expression}
9567 @itemx macro exp1 @var{expression}
9568 @cindex expand macro once
9569 @i{(This command is not yet implemented.)} Show the results of
9570 expanding those preprocessor macro invocations that appear explicitly in
9571 @var{expression}. Macro invocations appearing in that expansion are
9572 left unchanged. This command allows you to see the effect of a
9573 particular macro more clearly, without being confused by further
9574 expansions. Since @value{GDBN} simply expands macros, but does not
9575 parse the result, @var{expression} need not be a valid expression; it
9576 can be any string of tokens.
9579 @cindex macro definition, showing
9580 @cindex definition, showing a macro's
9581 @item info macro @var{macro}
9582 Show the definition of the macro named @var{macro}, and describe the
9583 source location or compiler command-line where that definition was established.
9585 @kindex macro define
9586 @cindex user-defined macros
9587 @cindex defining macros interactively
9588 @cindex macros, user-defined
9589 @item macro define @var{macro} @var{replacement-list}
9590 @itemx macro define @var{macro}(@var{arglist}) @var{replacement-list}
9591 Introduce a definition for a preprocessor macro named @var{macro},
9592 invocations of which are replaced by the tokens given in
9593 @var{replacement-list}. The first form of this command defines an
9594 ``object-like'' macro, which takes no arguments; the second form
9595 defines a ``function-like'' macro, which takes the arguments given in
9598 A definition introduced by this command is in scope in every
9599 expression evaluated in @value{GDBN}, until it is removed with the
9600 @code{macro undef} command, described below. The definition overrides
9601 all definitions for @var{macro} present in the program being debugged,
9602 as well as any previous user-supplied definition.
9605 @item macro undef @var{macro}
9606 Remove any user-supplied definition for the macro named @var{macro}.
9607 This command only affects definitions provided with the @code{macro
9608 define} command, described above; it cannot remove definitions present
9609 in the program being debugged.
9613 List all the macros defined using the @code{macro define} command.
9616 @cindex macros, example of debugging with
9617 Here is a transcript showing the above commands in action. First, we
9618 show our source files:
9626 #define ADD(x) (M + x)
9631 printf ("Hello, world!\n");
9633 printf ("We're so creative.\n");
9635 printf ("Goodbye, world!\n");
9642 Now, we compile the program using the @sc{gnu} C compiler, @value{NGCC}.
9643 We pass the @option{-gdwarf-2} and @option{-g3} flags to ensure the
9644 compiler includes information about preprocessor macros in the debugging
9648 $ gcc -gdwarf-2 -g3 sample.c -o sample
9652 Now, we start @value{GDBN} on our sample program:
9656 GNU gdb 2002-05-06-cvs
9657 Copyright 2002 Free Software Foundation, Inc.
9658 GDB is free software, @dots{}
9662 We can expand macros and examine their definitions, even when the
9663 program is not running. @value{GDBN} uses the current listing position
9664 to decide which macro definitions are in scope:
9667 (@value{GDBP}) list main
9670 5 #define ADD(x) (M + x)
9675 10 printf ("Hello, world!\n");
9677 12 printf ("We're so creative.\n");
9678 (@value{GDBP}) info macro ADD
9679 Defined at /home/jimb/gdb/macros/play/sample.c:5
9680 #define ADD(x) (M + x)
9681 (@value{GDBP}) info macro Q
9682 Defined at /home/jimb/gdb/macros/play/sample.h:1
9683 included at /home/jimb/gdb/macros/play/sample.c:2
9685 (@value{GDBP}) macro expand ADD(1)
9686 expands to: (42 + 1)
9687 (@value{GDBP}) macro expand-once ADD(1)
9688 expands to: once (M + 1)
9692 In the example above, note that @code{macro expand-once} expands only
9693 the macro invocation explicit in the original text --- the invocation of
9694 @code{ADD} --- but does not expand the invocation of the macro @code{M},
9695 which was introduced by @code{ADD}.
9697 Once the program is running, @value{GDBN} uses the macro definitions in
9698 force at the source line of the current stack frame:
9701 (@value{GDBP}) break main
9702 Breakpoint 1 at 0x8048370: file sample.c, line 10.
9704 Starting program: /home/jimb/gdb/macros/play/sample
9706 Breakpoint 1, main () at sample.c:10
9707 10 printf ("Hello, world!\n");
9711 At line 10, the definition of the macro @code{N} at line 9 is in force:
9714 (@value{GDBP}) info macro N
9715 Defined at /home/jimb/gdb/macros/play/sample.c:9
9717 (@value{GDBP}) macro expand N Q M
9719 (@value{GDBP}) print N Q M
9724 As we step over directives that remove @code{N}'s definition, and then
9725 give it a new definition, @value{GDBN} finds the definition (or lack
9726 thereof) in force at each point:
9731 12 printf ("We're so creative.\n");
9732 (@value{GDBP}) info macro N
9733 The symbol `N' has no definition as a C/C++ preprocessor macro
9734 at /home/jimb/gdb/macros/play/sample.c:12
9737 14 printf ("Goodbye, world!\n");
9738 (@value{GDBP}) info macro N
9739 Defined at /home/jimb/gdb/macros/play/sample.c:13
9741 (@value{GDBP}) macro expand N Q M
9742 expands to: 1729 < 42
9743 (@value{GDBP}) print N Q M
9748 In addition to source files, macros can be defined on the compilation command
9749 line using the @option{-D@var{name}=@var{value}} syntax. For macros defined in
9750 such a way, @value{GDBN} displays the location of their definition as line zero
9751 of the source file submitted to the compiler.
9754 (@value{GDBP}) info macro __STDC__
9755 Defined at /home/jimb/gdb/macros/play/sample.c:0
9762 @chapter Tracepoints
9763 @c This chapter is based on the documentation written by Michael
9764 @c Snyder, David Taylor, Jim Blandy, and Elena Zannoni.
9767 In some applications, it is not feasible for the debugger to interrupt
9768 the program's execution long enough for the developer to learn
9769 anything helpful about its behavior. If the program's correctness
9770 depends on its real-time behavior, delays introduced by a debugger
9771 might cause the program to change its behavior drastically, or perhaps
9772 fail, even when the code itself is correct. It is useful to be able
9773 to observe the program's behavior without interrupting it.
9775 Using @value{GDBN}'s @code{trace} and @code{collect} commands, you can
9776 specify locations in the program, called @dfn{tracepoints}, and
9777 arbitrary expressions to evaluate when those tracepoints are reached.
9778 Later, using the @code{tfind} command, you can examine the values
9779 those expressions had when the program hit the tracepoints. The
9780 expressions may also denote objects in memory---structures or arrays,
9781 for example---whose values @value{GDBN} should record; while visiting
9782 a particular tracepoint, you may inspect those objects as if they were
9783 in memory at that moment. However, because @value{GDBN} records these
9784 values without interacting with you, it can do so quickly and
9785 unobtrusively, hopefully not disturbing the program's behavior.
9787 The tracepoint facility is currently available only for remote
9788 targets. @xref{Targets}. In addition, your remote target must know
9789 how to collect trace data. This functionality is implemented in the
9790 remote stub; however, none of the stubs distributed with @value{GDBN}
9791 support tracepoints as of this writing. The format of the remote
9792 packets used to implement tracepoints are described in @ref{Tracepoint
9795 It is also possible to get trace data from a file, in a manner reminiscent
9796 of corefiles; you specify the filename, and use @code{tfind} to search
9797 through the file. @xref{Trace Files}, for more details.
9799 This chapter describes the tracepoint commands and features.
9803 * Analyze Collected Data::
9804 * Tracepoint Variables::
9808 @node Set Tracepoints
9809 @section Commands to Set Tracepoints
9811 Before running such a @dfn{trace experiment}, an arbitrary number of
9812 tracepoints can be set. A tracepoint is actually a special type of
9813 breakpoint (@pxref{Set Breaks}), so you can manipulate it using
9814 standard breakpoint commands. For instance, as with breakpoints,
9815 tracepoint numbers are successive integers starting from one, and many
9816 of the commands associated with tracepoints take the tracepoint number
9817 as their argument, to identify which tracepoint to work on.
9819 For each tracepoint, you can specify, in advance, some arbitrary set
9820 of data that you want the target to collect in the trace buffer when
9821 it hits that tracepoint. The collected data can include registers,
9822 local variables, or global data. Later, you can use @value{GDBN}
9823 commands to examine the values these data had at the time the
9826 Tracepoints do not support every breakpoint feature. Ignore counts on
9827 tracepoints have no effect, and tracepoints cannot run @value{GDBN}
9828 commands when they are hit. Tracepoints may not be thread-specific
9831 @cindex fast tracepoints
9832 Some targets may support @dfn{fast tracepoints}, which are inserted in
9833 a different way (such as with a jump instead of a trap), that is
9834 faster but possibly restricted in where they may be installed.
9836 @cindex static tracepoints
9837 @cindex markers, static tracepoints
9838 @cindex probing markers, static tracepoints
9839 Regular and fast tracepoints are dynamic tracing facilities, meaning
9840 that they can be used to insert tracepoints at (almost) any location
9841 in the target. Some targets may also support controlling @dfn{static
9842 tracepoints} from @value{GDBN}. With static tracing, a set of
9843 instrumentation points, also known as @dfn{markers}, are embedded in
9844 the target program, and can be activated or deactivated by name or
9845 address. These are usually placed at locations which facilitate
9846 investigating what the target is actually doing. @value{GDBN}'s
9847 support for static tracing includes being able to list instrumentation
9848 points, and attach them with @value{GDBN} defined high level
9849 tracepoints that expose the whole range of convenience of
9850 @value{GDBN}'s tracepoints support. Namely, support for collecting
9851 registers values and values of global or local (to the instrumentation
9852 point) variables; tracepoint conditions and trace state variables.
9853 The act of installing a @value{GDBN} static tracepoint on an
9854 instrumentation point, or marker, is referred to as @dfn{probing} a
9855 static tracepoint marker.
9857 @code{gdbserver} supports tracepoints on some target systems.
9858 @xref{Server,,Tracepoints support in @code{gdbserver}}.
9860 This section describes commands to set tracepoints and associated
9861 conditions and actions.
9864 * Create and Delete Tracepoints::
9865 * Enable and Disable Tracepoints::
9866 * Tracepoint Passcounts::
9867 * Tracepoint Conditions::
9868 * Trace State Variables::
9869 * Tracepoint Actions::
9870 * Listing Tracepoints::
9871 * Listing Static Tracepoint Markers::
9872 * Starting and Stopping Trace Experiments::
9873 * Tracepoint Restrictions::
9876 @node Create and Delete Tracepoints
9877 @subsection Create and Delete Tracepoints
9880 @cindex set tracepoint
9882 @item trace @var{location}
9883 The @code{trace} command is very similar to the @code{break} command.
9884 Its argument @var{location} can be a source line, a function name, or
9885 an address in the target program. @xref{Specify Location}. The
9886 @code{trace} command defines a tracepoint, which is a point in the
9887 target program where the debugger will briefly stop, collect some
9888 data, and then allow the program to continue. Setting a tracepoint or
9889 changing its actions doesn't take effect until the next @code{tstart}
9890 command, and once a trace experiment is running, further changes will
9891 not have any effect until the next trace experiment starts.
9893 Here are some examples of using the @code{trace} command:
9896 (@value{GDBP}) @b{trace foo.c:121} // a source file and line number
9898 (@value{GDBP}) @b{trace +2} // 2 lines forward
9900 (@value{GDBP}) @b{trace my_function} // first source line of function
9902 (@value{GDBP}) @b{trace *my_function} // EXACT start address of function
9904 (@value{GDBP}) @b{trace *0x2117c4} // an address
9908 You can abbreviate @code{trace} as @code{tr}.
9910 @item trace @var{location} if @var{cond}
9911 Set a tracepoint with condition @var{cond}; evaluate the expression
9912 @var{cond} each time the tracepoint is reached, and collect data only
9913 if the value is nonzero---that is, if @var{cond} evaluates as true.
9914 @xref{Tracepoint Conditions, ,Tracepoint Conditions}, for more
9915 information on tracepoint conditions.
9917 @item ftrace @var{location} [ if @var{cond} ]
9918 @cindex set fast tracepoint
9919 @cindex fast tracepoints, setting
9921 The @code{ftrace} command sets a fast tracepoint. For targets that
9922 support them, fast tracepoints will use a more efficient but possibly
9923 less general technique to trigger data collection, such as a jump
9924 instruction instead of a trap, or some sort of hardware support. It
9925 may not be possible to create a fast tracepoint at the desired
9926 location, in which case the command will exit with an explanatory
9929 @value{GDBN} handles arguments to @code{ftrace} exactly as for
9932 @item strace @var{location} [ if @var{cond} ]
9933 @cindex set static tracepoint
9934 @cindex static tracepoints, setting
9935 @cindex probe static tracepoint marker
9937 The @code{strace} command sets a static tracepoint. For targets that
9938 support it, setting a static tracepoint probes a static
9939 instrumentation point, or marker, found at @var{location}. It may not
9940 be possible to set a static tracepoint at the desired location, in
9941 which case the command will exit with an explanatory message.
9943 @value{GDBN} handles arguments to @code{strace} exactly as for
9944 @code{trace}, with the addition that the user can also specify
9945 @code{-m @var{marker}} as @var{location}. This probes the marker
9946 identified by the @var{marker} string identifier. This identifier
9947 depends on the static tracepoint backend library your program is
9948 using. You can find all the marker identifiers in the @samp{ID} field
9949 of the @code{info static-tracepoint-markers} command output.
9950 @xref{Listing Static Tracepoint Markers,,Listing Static Tracepoint
9951 Markers}. For example, in the following small program using the UST
9957 trace_mark(ust, bar33, "str %s", "FOOBAZ");
9962 the marker id is composed of joining the first two arguments to the
9963 @code{trace_mark} call with a slash, which translates to:
9966 (@value{GDBP}) info static-tracepoint-markers
9967 Cnt Enb ID Address What
9968 1 n ust/bar33 0x0000000000400ddc in main at stexample.c:22
9974 so you may probe the marker above with:
9977 (@value{GDBP}) strace -m ust/bar33
9980 Static tracepoints accept an extra collect action --- @code{collect
9981 $_sdata}. This collects arbitrary user data passed in the probe point
9982 call to the tracing library. In the UST example above, you'll see
9983 that the third argument to @code{trace_mark} is a printf-like format
9984 string. The user data is then the result of running that formating
9985 string against the following arguments. Note that @code{info
9986 static-tracepoint-markers} command output lists that format string in
9987 the @samp{Data:} field.
9989 You can inspect this data when analyzing the trace buffer, by printing
9990 the $_sdata variable like any other variable available to
9991 @value{GDBN}. @xref{Tracepoint Actions,,Tracepoint Action Lists}.
9994 @cindex last tracepoint number
9995 @cindex recent tracepoint number
9996 @cindex tracepoint number
9997 The convenience variable @code{$tpnum} records the tracepoint number
9998 of the most recently set tracepoint.
10000 @kindex delete tracepoint
10001 @cindex tracepoint deletion
10002 @item delete tracepoint @r{[}@var{num}@r{]}
10003 Permanently delete one or more tracepoints. With no argument, the
10004 default is to delete all tracepoints. Note that the regular
10005 @code{delete} command can remove tracepoints also.
10010 (@value{GDBP}) @b{delete trace 1 2 3} // remove three tracepoints
10012 (@value{GDBP}) @b{delete trace} // remove all tracepoints
10016 You can abbreviate this command as @code{del tr}.
10019 @node Enable and Disable Tracepoints
10020 @subsection Enable and Disable Tracepoints
10022 These commands are deprecated; they are equivalent to plain @code{disable} and @code{enable}.
10025 @kindex disable tracepoint
10026 @item disable tracepoint @r{[}@var{num}@r{]}
10027 Disable tracepoint @var{num}, or all tracepoints if no argument
10028 @var{num} is given. A disabled tracepoint will have no effect during
10029 a trace experiment, but it is not forgotten. You can re-enable
10030 a disabled tracepoint using the @code{enable tracepoint} command.
10031 If the command is issued during a trace experiment and the debug target
10032 has support for disabling tracepoints during a trace experiment, then the
10033 change will be effective immediately. Otherwise, it will be applied to the
10034 next trace experiment.
10036 @kindex enable tracepoint
10037 @item enable tracepoint @r{[}@var{num}@r{]}
10038 Enable tracepoint @var{num}, or all tracepoints. If this command is
10039 issued during a trace experiment and the debug target supports enabling
10040 tracepoints during a trace experiment, then the enabled tracepoints will
10041 become effective immediately. Otherwise, they will become effective the
10042 next time a trace experiment is run.
10045 @node Tracepoint Passcounts
10046 @subsection Tracepoint Passcounts
10050 @cindex tracepoint pass count
10051 @item passcount @r{[}@var{n} @r{[}@var{num}@r{]]}
10052 Set the @dfn{passcount} of a tracepoint. The passcount is a way to
10053 automatically stop a trace experiment. If a tracepoint's passcount is
10054 @var{n}, then the trace experiment will be automatically stopped on
10055 the @var{n}'th time that tracepoint is hit. If the tracepoint number
10056 @var{num} is not specified, the @code{passcount} command sets the
10057 passcount of the most recently defined tracepoint. If no passcount is
10058 given, the trace experiment will run until stopped explicitly by the
10064 (@value{GDBP}) @b{passcount 5 2} // Stop on the 5th execution of
10065 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// tracepoint 2}
10067 (@value{GDBP}) @b{passcount 12} // Stop on the 12th execution of the
10068 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// most recently defined tracepoint.}
10069 (@value{GDBP}) @b{trace foo}
10070 (@value{GDBP}) @b{pass 3}
10071 (@value{GDBP}) @b{trace bar}
10072 (@value{GDBP}) @b{pass 2}
10073 (@value{GDBP}) @b{trace baz}
10074 (@value{GDBP}) @b{pass 1} // Stop tracing when foo has been
10075 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// executed 3 times OR when bar has}
10076 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// been executed 2 times}
10077 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// OR when baz has been executed 1 time.}
10081 @node Tracepoint Conditions
10082 @subsection Tracepoint Conditions
10083 @cindex conditional tracepoints
10084 @cindex tracepoint conditions
10086 The simplest sort of tracepoint collects data every time your program
10087 reaches a specified place. You can also specify a @dfn{condition} for
10088 a tracepoint. A condition is just a Boolean expression in your
10089 programming language (@pxref{Expressions, ,Expressions}). A
10090 tracepoint with a condition evaluates the expression each time your
10091 program reaches it, and data collection happens only if the condition
10094 Tracepoint conditions can be specified when a tracepoint is set, by
10095 using @samp{if} in the arguments to the @code{trace} command.
10096 @xref{Create and Delete Tracepoints, ,Setting Tracepoints}. They can
10097 also be set or changed at any time with the @code{condition} command,
10098 just as with breakpoints.
10100 Unlike breakpoint conditions, @value{GDBN} does not actually evaluate
10101 the conditional expression itself. Instead, @value{GDBN} encodes the
10102 expression into an agent expression (@pxref{Agent Expressions})
10103 suitable for execution on the target, independently of @value{GDBN}.
10104 Global variables become raw memory locations, locals become stack
10105 accesses, and so forth.
10107 For instance, suppose you have a function that is usually called
10108 frequently, but should not be called after an error has occurred. You
10109 could use the following tracepoint command to collect data about calls
10110 of that function that happen while the error code is propagating
10111 through the program; an unconditional tracepoint could end up
10112 collecting thousands of useless trace frames that you would have to
10116 (@value{GDBP}) @kbd{trace normal_operation if errcode > 0}
10119 @node Trace State Variables
10120 @subsection Trace State Variables
10121 @cindex trace state variables
10123 A @dfn{trace state variable} is a special type of variable that is
10124 created and managed by target-side code. The syntax is the same as
10125 that for GDB's convenience variables (a string prefixed with ``$''),
10126 but they are stored on the target. They must be created explicitly,
10127 using a @code{tvariable} command. They are always 64-bit signed
10130 Trace state variables are remembered by @value{GDBN}, and downloaded
10131 to the target along with tracepoint information when the trace
10132 experiment starts. There are no intrinsic limits on the number of
10133 trace state variables, beyond memory limitations of the target.
10135 @cindex convenience variables, and trace state variables
10136 Although trace state variables are managed by the target, you can use
10137 them in print commands and expressions as if they were convenience
10138 variables; @value{GDBN} will get the current value from the target
10139 while the trace experiment is running. Trace state variables share
10140 the same namespace as other ``$'' variables, which means that you
10141 cannot have trace state variables with names like @code{$23} or
10142 @code{$pc}, nor can you have a trace state variable and a convenience
10143 variable with the same name.
10147 @item tvariable $@var{name} [ = @var{expression} ]
10149 The @code{tvariable} command creates a new trace state variable named
10150 @code{$@var{name}}, and optionally gives it an initial value of
10151 @var{expression}. @var{expression} is evaluated when this command is
10152 entered; the result will be converted to an integer if possible,
10153 otherwise @value{GDBN} will report an error. A subsequent
10154 @code{tvariable} command specifying the same name does not create a
10155 variable, but instead assigns the supplied initial value to the
10156 existing variable of that name, overwriting any previous initial
10157 value. The default initial value is 0.
10159 @item info tvariables
10160 @kindex info tvariables
10161 List all the trace state variables along with their initial values.
10162 Their current values may also be displayed, if the trace experiment is
10165 @item delete tvariable @r{[} $@var{name} @dots{} @r{]}
10166 @kindex delete tvariable
10167 Delete the given trace state variables, or all of them if no arguments
10172 @node Tracepoint Actions
10173 @subsection Tracepoint Action Lists
10177 @cindex tracepoint actions
10178 @item actions @r{[}@var{num}@r{]}
10179 This command will prompt for a list of actions to be taken when the
10180 tracepoint is hit. If the tracepoint number @var{num} is not
10181 specified, this command sets the actions for the one that was most
10182 recently defined (so that you can define a tracepoint and then say
10183 @code{actions} without bothering about its number). You specify the
10184 actions themselves on the following lines, one action at a time, and
10185 terminate the actions list with a line containing just @code{end}. So
10186 far, the only defined actions are @code{collect}, @code{teval}, and
10187 @code{while-stepping}.
10189 @code{actions} is actually equivalent to @code{commands} (@pxref{Break
10190 Commands, ,Breakpoint Command Lists}), except that only the defined
10191 actions are allowed; any other @value{GDBN} command is rejected.
10193 @cindex remove actions from a tracepoint
10194 To remove all actions from a tracepoint, type @samp{actions @var{num}}
10195 and follow it immediately with @samp{end}.
10198 (@value{GDBP}) @b{collect @var{data}} // collect some data
10200 (@value{GDBP}) @b{while-stepping 5} // single-step 5 times, collect data
10202 (@value{GDBP}) @b{end} // signals the end of actions.
10205 In the following example, the action list begins with @code{collect}
10206 commands indicating the things to be collected when the tracepoint is
10207 hit. Then, in order to single-step and collect additional data
10208 following the tracepoint, a @code{while-stepping} command is used,
10209 followed by the list of things to be collected after each step in a
10210 sequence of single steps. The @code{while-stepping} command is
10211 terminated by its own separate @code{end} command. Lastly, the action
10212 list is terminated by an @code{end} command.
10215 (@value{GDBP}) @b{trace foo}
10216 (@value{GDBP}) @b{actions}
10217 Enter actions for tracepoint 1, one per line:
10220 > while-stepping 12
10221 > collect $pc, arr[i]
10226 @kindex collect @r{(tracepoints)}
10227 @item collect @var{expr1}, @var{expr2}, @dots{}
10228 Collect values of the given expressions when the tracepoint is hit.
10229 This command accepts a comma-separated list of any valid expressions.
10230 In addition to global, static, or local variables, the following
10231 special arguments are supported:
10235 Collect all registers.
10238 Collect all function arguments.
10241 Collect all local variables.
10244 @vindex $_sdata@r{, collect}
10245 Collect static tracepoint marker specific data. Only available for
10246 static tracepoints. @xref{Tracepoint Actions,,Tracepoint Action
10247 Lists}. On the UST static tracepoints library backend, an
10248 instrumentation point resembles a @code{printf} function call. The
10249 tracing library is able to collect user specified data formatted to a
10250 character string using the format provided by the programmer that
10251 instrumented the program. Other backends have similar mechanisms.
10252 Here's an example of a UST marker call:
10255 const char master_name[] = "$your_name";
10256 trace_mark(channel1, marker1, "hello %s", master_name)
10259 In this case, collecting @code{$_sdata} collects the string
10260 @samp{hello $yourname}. When analyzing the trace buffer, you can
10261 inspect @samp{$_sdata} like any other variable available to
10265 You can give several consecutive @code{collect} commands, each one
10266 with a single argument, or one @code{collect} command with several
10267 arguments separated by commas; the effect is the same.
10269 The command @code{info scope} (@pxref{Symbols, info scope}) is
10270 particularly useful for figuring out what data to collect.
10272 @kindex teval @r{(tracepoints)}
10273 @item teval @var{expr1}, @var{expr2}, @dots{}
10274 Evaluate the given expressions when the tracepoint is hit. This
10275 command accepts a comma-separated list of expressions. The results
10276 are discarded, so this is mainly useful for assigning values to trace
10277 state variables (@pxref{Trace State Variables}) without adding those
10278 values to the trace buffer, as would be the case if the @code{collect}
10281 @kindex while-stepping @r{(tracepoints)}
10282 @item while-stepping @var{n}
10283 Perform @var{n} single-step instruction traces after the tracepoint,
10284 collecting new data after each step. The @code{while-stepping}
10285 command is followed by the list of what to collect while stepping
10286 (followed by its own @code{end} command):
10289 > while-stepping 12
10290 > collect $regs, myglobal
10296 Note that @code{$pc} is not automatically collected by
10297 @code{while-stepping}; you need to explicitly collect that register if
10298 you need it. You may abbreviate @code{while-stepping} as @code{ws} or
10301 @item set default-collect @var{expr1}, @var{expr2}, @dots{}
10302 @kindex set default-collect
10303 @cindex default collection action
10304 This variable is a list of expressions to collect at each tracepoint
10305 hit. It is effectively an additional @code{collect} action prepended
10306 to every tracepoint action list. The expressions are parsed
10307 individually for each tracepoint, so for instance a variable named
10308 @code{xyz} may be interpreted as a global for one tracepoint, and a
10309 local for another, as appropriate to the tracepoint's location.
10311 @item show default-collect
10312 @kindex show default-collect
10313 Show the list of expressions that are collected by default at each
10318 @node Listing Tracepoints
10319 @subsection Listing Tracepoints
10322 @kindex info tracepoints @r{[}@var{n}@dots{}@r{]}
10323 @kindex info tp @r{[}@var{n}@dots{}@r{]}
10324 @cindex information about tracepoints
10325 @item info tracepoints @r{[}@var{num}@dots{}@r{]}
10326 Display information about the tracepoint @var{num}. If you don't
10327 specify a tracepoint number, displays information about all the
10328 tracepoints defined so far. The format is similar to that used for
10329 @code{info breakpoints}; in fact, @code{info tracepoints} is the same
10330 command, simply restricting itself to tracepoints.
10332 A tracepoint's listing may include additional information specific to
10337 its passcount as given by the @code{passcount @var{n}} command
10341 (@value{GDBP}) @b{info trace}
10342 Num Type Disp Enb Address What
10343 1 tracepoint keep y 0x0804ab57 in foo() at main.cxx:7
10345 collect globfoo, $regs
10354 This command can be abbreviated @code{info tp}.
10357 @node Listing Static Tracepoint Markers
10358 @subsection Listing Static Tracepoint Markers
10361 @kindex info static-tracepoint-markers
10362 @cindex information about static tracepoint markers
10363 @item info static-tracepoint-markers
10364 Display information about all static tracepoint markers defined in the
10367 For each marker, the following columns are printed:
10371 An incrementing counter, output to help readability. This is not a
10374 The marker ID, as reported by the target.
10375 @item Enabled or Disabled
10376 Probed markers are tagged with @samp{y}. @samp{n} identifies marks
10377 that are not enabled.
10379 Where the marker is in your program, as a memory address.
10381 Where the marker is in the source for your program, as a file and line
10382 number. If the debug information included in the program does not
10383 allow @value{GDBN} to locate the source of the marker, this column
10384 will be left blank.
10388 In addition, the following information may be printed for each marker:
10392 User data passed to the tracing library by the marker call. In the
10393 UST backend, this is the format string passed as argument to the
10395 @item Static tracepoints probing the marker
10396 The list of static tracepoints attached to the marker.
10400 (@value{GDBP}) info static-tracepoint-markers
10401 Cnt ID Enb Address What
10402 1 ust/bar2 y 0x0000000000400e1a in main at stexample.c:25
10403 Data: number1 %d number2 %d
10404 Probed by static tracepoints: #2
10405 2 ust/bar33 n 0x0000000000400c87 in main at stexample.c:24
10411 @node Starting and Stopping Trace Experiments
10412 @subsection Starting and Stopping Trace Experiments
10416 @cindex start a new trace experiment
10417 @cindex collected data discarded
10419 This command takes no arguments. It starts the trace experiment, and
10420 begins collecting data. This has the side effect of discarding all
10421 the data collected in the trace buffer during the previous trace
10425 @cindex stop a running trace experiment
10427 This command takes no arguments. It ends the trace experiment, and
10428 stops collecting data.
10430 @strong{Note}: a trace experiment and data collection may stop
10431 automatically if any tracepoint's passcount is reached
10432 (@pxref{Tracepoint Passcounts}), or if the trace buffer becomes full.
10435 @cindex status of trace data collection
10436 @cindex trace experiment, status of
10438 This command displays the status of the current trace data
10442 Here is an example of the commands we described so far:
10445 (@value{GDBP}) @b{trace gdb_c_test}
10446 (@value{GDBP}) @b{actions}
10447 Enter actions for tracepoint #1, one per line.
10448 > collect $regs,$locals,$args
10449 > while-stepping 11
10453 (@value{GDBP}) @b{tstart}
10454 [time passes @dots{}]
10455 (@value{GDBP}) @b{tstop}
10458 @anchor{disconnected tracing}
10459 @cindex disconnected tracing
10460 You can choose to continue running the trace experiment even if
10461 @value{GDBN} disconnects from the target, voluntarily or
10462 involuntarily. For commands such as @code{detach}, the debugger will
10463 ask what you want to do with the trace. But for unexpected
10464 terminations (@value{GDBN} crash, network outage), it would be
10465 unfortunate to lose hard-won trace data, so the variable
10466 @code{disconnected-tracing} lets you decide whether the trace should
10467 continue running without @value{GDBN}.
10470 @item set disconnected-tracing on
10471 @itemx set disconnected-tracing off
10472 @kindex set disconnected-tracing
10473 Choose whether a tracing run should continue to run if @value{GDBN}
10474 has disconnected from the target. Note that @code{detach} or
10475 @code{quit} will ask you directly what to do about a running trace no
10476 matter what this variable's setting, so the variable is mainly useful
10477 for handling unexpected situations, such as loss of the network.
10479 @item show disconnected-tracing
10480 @kindex show disconnected-tracing
10481 Show the current choice for disconnected tracing.
10485 When you reconnect to the target, the trace experiment may or may not
10486 still be running; it might have filled the trace buffer in the
10487 meantime, or stopped for one of the other reasons. If it is running,
10488 it will continue after reconnection.
10490 Upon reconnection, the target will upload information about the
10491 tracepoints in effect. @value{GDBN} will then compare that
10492 information to the set of tracepoints currently defined, and attempt
10493 to match them up, allowing for the possibility that the numbers may
10494 have changed due to creation and deletion in the meantime. If one of
10495 the target's tracepoints does not match any in @value{GDBN}, the
10496 debugger will create a new tracepoint, so that you have a number with
10497 which to specify that tracepoint. This matching-up process is
10498 necessarily heuristic, and it may result in useless tracepoints being
10499 created; you may simply delete them if they are of no use.
10501 @cindex circular trace buffer
10502 If your target agent supports a @dfn{circular trace buffer}, then you
10503 can run a trace experiment indefinitely without filling the trace
10504 buffer; when space runs out, the agent deletes already-collected trace
10505 frames, oldest first, until there is enough room to continue
10506 collecting. This is especially useful if your tracepoints are being
10507 hit too often, and your trace gets terminated prematurely because the
10508 buffer is full. To ask for a circular trace buffer, simply set
10509 @samp{circular-trace-buffer} to on. You can set this at any time,
10510 including during tracing; if the agent can do it, it will change
10511 buffer handling on the fly, otherwise it will not take effect until
10515 @item set circular-trace-buffer on
10516 @itemx set circular-trace-buffer off
10517 @kindex set circular-trace-buffer
10518 Choose whether a tracing run should use a linear or circular buffer
10519 for trace data. A linear buffer will not lose any trace data, but may
10520 fill up prematurely, while a circular buffer will discard old trace
10521 data, but it will have always room for the latest tracepoint hits.
10523 @item show circular-trace-buffer
10524 @kindex show circular-trace-buffer
10525 Show the current choice for the trace buffer. Note that this may not
10526 match the agent's current buffer handling, nor is it guaranteed to
10527 match the setting that might have been in effect during a past run,
10528 for instance if you are looking at frames from a trace file.
10532 @node Tracepoint Restrictions
10533 @subsection Tracepoint Restrictions
10535 @cindex tracepoint restrictions
10536 There are a number of restrictions on the use of tracepoints. As
10537 described above, tracepoint data gathering occurs on the target
10538 without interaction from @value{GDBN}. Thus the full capabilities of
10539 the debugger are not available during data gathering, and then at data
10540 examination time, you will be limited by only having what was
10541 collected. The following items describe some common problems, but it
10542 is not exhaustive, and you may run into additional difficulties not
10548 Tracepoint expressions are intended to gather objects (lvalues). Thus
10549 the full flexibility of GDB's expression evaluator is not available.
10550 You cannot call functions, cast objects to aggregate types, access
10551 convenience variables or modify values (except by assignment to trace
10552 state variables). Some language features may implicitly call
10553 functions (for instance Objective-C fields with accessors), and therefore
10554 cannot be collected either.
10557 Collection of local variables, either individually or in bulk with
10558 @code{$locals} or @code{$args}, during @code{while-stepping} may
10559 behave erratically. The stepping action may enter a new scope (for
10560 instance by stepping into a function), or the location of the variable
10561 may change (for instance it is loaded into a register). The
10562 tracepoint data recorded uses the location information for the
10563 variables that is correct for the tracepoint location. When the
10564 tracepoint is created, it is not possible, in general, to determine
10565 where the steps of a @code{while-stepping} sequence will advance the
10566 program---particularly if a conditional branch is stepped.
10569 Collection of an incompletely-initialized or partially-destroyed object
10570 may result in something that @value{GDBN} cannot display, or displays
10571 in a misleading way.
10574 When @value{GDBN} displays a pointer to character it automatically
10575 dereferences the pointer to also display characters of the string
10576 being pointed to. However, collecting the pointer during tracing does
10577 not automatically collect the string. You need to explicitly
10578 dereference the pointer and provide size information if you want to
10579 collect not only the pointer, but the memory pointed to. For example,
10580 @code{*ptr@@50} can be used to collect the 50 element array pointed to
10584 It is not possible to collect a complete stack backtrace at a
10585 tracepoint. Instead, you may collect the registers and a few hundred
10586 bytes from the stack pointer with something like @code{*$esp@@300}
10587 (adjust to use the name of the actual stack pointer register on your
10588 target architecture, and the amount of stack you wish to capture).
10589 Then the @code{backtrace} command will show a partial backtrace when
10590 using a trace frame. The number of stack frames that can be examined
10591 depends on the sizes of the frames in the collected stack. Note that
10592 if you ask for a block so large that it goes past the bottom of the
10593 stack, the target agent may report an error trying to read from an
10597 If you do not collect registers at a tracepoint, @value{GDBN} can
10598 infer that the value of @code{$pc} must be the same as the address of
10599 the tracepoint and use that when you are looking at a trace frame
10600 for that tracepoint. However, this cannot work if the tracepoint has
10601 multiple locations (for instance if it was set in a function that was
10602 inlined), or if it has a @code{while-stepping} loop. In those cases
10603 @value{GDBN} will warn you that it can't infer @code{$pc}, and default
10608 @node Analyze Collected Data
10609 @section Using the Collected Data
10611 After the tracepoint experiment ends, you use @value{GDBN} commands
10612 for examining the trace data. The basic idea is that each tracepoint
10613 collects a trace @dfn{snapshot} every time it is hit and another
10614 snapshot every time it single-steps. All these snapshots are
10615 consecutively numbered from zero and go into a buffer, and you can
10616 examine them later. The way you examine them is to @dfn{focus} on a
10617 specific trace snapshot. When the remote stub is focused on a trace
10618 snapshot, it will respond to all @value{GDBN} requests for memory and
10619 registers by reading from the buffer which belongs to that snapshot,
10620 rather than from @emph{real} memory or registers of the program being
10621 debugged. This means that @strong{all} @value{GDBN} commands
10622 (@code{print}, @code{info registers}, @code{backtrace}, etc.) will
10623 behave as if we were currently debugging the program state as it was
10624 when the tracepoint occurred. Any requests for data that are not in
10625 the buffer will fail.
10628 * tfind:: How to select a trace snapshot
10629 * tdump:: How to display all data for a snapshot
10630 * save tracepoints:: How to save tracepoints for a future run
10634 @subsection @code{tfind @var{n}}
10637 @cindex select trace snapshot
10638 @cindex find trace snapshot
10639 The basic command for selecting a trace snapshot from the buffer is
10640 @code{tfind @var{n}}, which finds trace snapshot number @var{n},
10641 counting from zero. If no argument @var{n} is given, the next
10642 snapshot is selected.
10644 Here are the various forms of using the @code{tfind} command.
10648 Find the first snapshot in the buffer. This is a synonym for
10649 @code{tfind 0} (since 0 is the number of the first snapshot).
10652 Stop debugging trace snapshots, resume @emph{live} debugging.
10655 Same as @samp{tfind none}.
10658 No argument means find the next trace snapshot.
10661 Find the previous trace snapshot before the current one. This permits
10662 retracing earlier steps.
10664 @item tfind tracepoint @var{num}
10665 Find the next snapshot associated with tracepoint @var{num}. Search
10666 proceeds forward from the last examined trace snapshot. If no
10667 argument @var{num} is given, it means find the next snapshot collected
10668 for the same tracepoint as the current snapshot.
10670 @item tfind pc @var{addr}
10671 Find the next snapshot associated with the value @var{addr} of the
10672 program counter. Search proceeds forward from the last examined trace
10673 snapshot. If no argument @var{addr} is given, it means find the next
10674 snapshot with the same value of PC as the current snapshot.
10676 @item tfind outside @var{addr1}, @var{addr2}
10677 Find the next snapshot whose PC is outside the given range of
10678 addresses (exclusive).
10680 @item tfind range @var{addr1}, @var{addr2}
10681 Find the next snapshot whose PC is between @var{addr1} and
10682 @var{addr2} (inclusive).
10684 @item tfind line @r{[}@var{file}:@r{]}@var{n}
10685 Find the next snapshot associated with the source line @var{n}. If
10686 the optional argument @var{file} is given, refer to line @var{n} in
10687 that source file. Search proceeds forward from the last examined
10688 trace snapshot. If no argument @var{n} is given, it means find the
10689 next line other than the one currently being examined; thus saying
10690 @code{tfind line} repeatedly can appear to have the same effect as
10691 stepping from line to line in a @emph{live} debugging session.
10694 The default arguments for the @code{tfind} commands are specifically
10695 designed to make it easy to scan through the trace buffer. For
10696 instance, @code{tfind} with no argument selects the next trace
10697 snapshot, and @code{tfind -} with no argument selects the previous
10698 trace snapshot. So, by giving one @code{tfind} command, and then
10699 simply hitting @key{RET} repeatedly you can examine all the trace
10700 snapshots in order. Or, by saying @code{tfind -} and then hitting
10701 @key{RET} repeatedly you can examine the snapshots in reverse order.
10702 The @code{tfind line} command with no argument selects the snapshot
10703 for the next source line executed. The @code{tfind pc} command with
10704 no argument selects the next snapshot with the same program counter
10705 (PC) as the current frame. The @code{tfind tracepoint} command with
10706 no argument selects the next trace snapshot collected by the same
10707 tracepoint as the current one.
10709 In addition to letting you scan through the trace buffer manually,
10710 these commands make it easy to construct @value{GDBN} scripts that
10711 scan through the trace buffer and print out whatever collected data
10712 you are interested in. Thus, if we want to examine the PC, FP, and SP
10713 registers from each trace frame in the buffer, we can say this:
10716 (@value{GDBP}) @b{tfind start}
10717 (@value{GDBP}) @b{while ($trace_frame != -1)}
10718 > printf "Frame %d, PC = %08X, SP = %08X, FP = %08X\n", \
10719 $trace_frame, $pc, $sp, $fp
10723 Frame 0, PC = 0020DC64, SP = 0030BF3C, FP = 0030BF44
10724 Frame 1, PC = 0020DC6C, SP = 0030BF38, FP = 0030BF44
10725 Frame 2, PC = 0020DC70, SP = 0030BF34, FP = 0030BF44
10726 Frame 3, PC = 0020DC74, SP = 0030BF30, FP = 0030BF44
10727 Frame 4, PC = 0020DC78, SP = 0030BF2C, FP = 0030BF44
10728 Frame 5, PC = 0020DC7C, SP = 0030BF28, FP = 0030BF44
10729 Frame 6, PC = 0020DC80, SP = 0030BF24, FP = 0030BF44
10730 Frame 7, PC = 0020DC84, SP = 0030BF20, FP = 0030BF44
10731 Frame 8, PC = 0020DC88, SP = 0030BF1C, FP = 0030BF44
10732 Frame 9, PC = 0020DC8E, SP = 0030BF18, FP = 0030BF44
10733 Frame 10, PC = 00203F6C, SP = 0030BE3C, FP = 0030BF14
10736 Or, if we want to examine the variable @code{X} at each source line in
10740 (@value{GDBP}) @b{tfind start}
10741 (@value{GDBP}) @b{while ($trace_frame != -1)}
10742 > printf "Frame %d, X == %d\n", $trace_frame, X
10752 @subsection @code{tdump}
10754 @cindex dump all data collected at tracepoint
10755 @cindex tracepoint data, display
10757 This command takes no arguments. It prints all the data collected at
10758 the current trace snapshot.
10761 (@value{GDBP}) @b{trace 444}
10762 (@value{GDBP}) @b{actions}
10763 Enter actions for tracepoint #2, one per line:
10764 > collect $regs, $locals, $args, gdb_long_test
10767 (@value{GDBP}) @b{tstart}
10769 (@value{GDBP}) @b{tfind line 444}
10770 #0 gdb_test (p1=0x11, p2=0x22, p3=0x33, p4=0x44, p5=0x55, p6=0x66)
10772 444 printp( "%s: arguments = 0x%X 0x%X 0x%X 0x%X 0x%X 0x%X\n", )
10774 (@value{GDBP}) @b{tdump}
10775 Data collected at tracepoint 2, trace frame 1:
10776 d0 0xc4aa0085 -995491707
10780 d4 0x71aea3d 119204413
10783 d7 0x380035 3670069
10784 a0 0x19e24a 1696330
10785 a1 0x3000668 50333288
10787 a3 0x322000 3284992
10788 a4 0x3000698 50333336
10789 a5 0x1ad3cc 1758156
10790 fp 0x30bf3c 0x30bf3c
10791 sp 0x30bf34 0x30bf34
10793 pc 0x20b2c8 0x20b2c8
10797 p = 0x20e5b4 "gdb-test"
10804 gdb_long_test = 17 '\021'
10809 @code{tdump} works by scanning the tracepoint's current collection
10810 actions and printing the value of each expression listed. So
10811 @code{tdump} can fail, if after a run, you change the tracepoint's
10812 actions to mention variables that were not collected during the run.
10814 Also, for tracepoints with @code{while-stepping} loops, @code{tdump}
10815 uses the collected value of @code{$pc} to distinguish between trace
10816 frames that were collected at the tracepoint hit, and frames that were
10817 collected while stepping. This allows it to correctly choose whether
10818 to display the basic list of collections, or the collections from the
10819 body of the while-stepping loop. However, if @code{$pc} was not collected,
10820 then @code{tdump} will always attempt to dump using the basic collection
10821 list, and may fail if a while-stepping frame does not include all the
10822 same data that is collected at the tracepoint hit.
10823 @c This is getting pretty arcane, example would be good.
10825 @node save tracepoints
10826 @subsection @code{save tracepoints @var{filename}}
10827 @kindex save tracepoints
10828 @kindex save-tracepoints
10829 @cindex save tracepoints for future sessions
10831 This command saves all current tracepoint definitions together with
10832 their actions and passcounts, into a file @file{@var{filename}}
10833 suitable for use in a later debugging session. To read the saved
10834 tracepoint definitions, use the @code{source} command (@pxref{Command
10835 Files}). The @w{@code{save-tracepoints}} command is a deprecated
10836 alias for @w{@code{save tracepoints}}
10838 @node Tracepoint Variables
10839 @section Convenience Variables for Tracepoints
10840 @cindex tracepoint variables
10841 @cindex convenience variables for tracepoints
10844 @vindex $trace_frame
10845 @item (int) $trace_frame
10846 The current trace snapshot (a.k.a.@: @dfn{frame}) number, or -1 if no
10847 snapshot is selected.
10849 @vindex $tracepoint
10850 @item (int) $tracepoint
10851 The tracepoint for the current trace snapshot.
10853 @vindex $trace_line
10854 @item (int) $trace_line
10855 The line number for the current trace snapshot.
10857 @vindex $trace_file
10858 @item (char []) $trace_file
10859 The source file for the current trace snapshot.
10861 @vindex $trace_func
10862 @item (char []) $trace_func
10863 The name of the function containing @code{$tracepoint}.
10866 Note: @code{$trace_file} is not suitable for use in @code{printf},
10867 use @code{output} instead.
10869 Here's a simple example of using these convenience variables for
10870 stepping through all the trace snapshots and printing some of their
10871 data. Note that these are not the same as trace state variables,
10872 which are managed by the target.
10875 (@value{GDBP}) @b{tfind start}
10877 (@value{GDBP}) @b{while $trace_frame != -1}
10878 > output $trace_file
10879 > printf ", line %d (tracepoint #%d)\n", $trace_line, $tracepoint
10885 @section Using Trace Files
10886 @cindex trace files
10888 In some situations, the target running a trace experiment may no
10889 longer be available; perhaps it crashed, or the hardware was needed
10890 for a different activity. To handle these cases, you can arrange to
10891 dump the trace data into a file, and later use that file as a source
10892 of trace data, via the @code{target tfile} command.
10897 @item tsave [ -r ] @var{filename}
10898 Save the trace data to @var{filename}. By default, this command
10899 assumes that @var{filename} refers to the host filesystem, so if
10900 necessary @value{GDBN} will copy raw trace data up from the target and
10901 then save it. If the target supports it, you can also supply the
10902 optional argument @code{-r} (``remote'') to direct the target to save
10903 the data directly into @var{filename} in its own filesystem, which may be
10904 more efficient if the trace buffer is very large. (Note, however, that
10905 @code{target tfile} can only read from files accessible to the host.)
10907 @kindex target tfile
10909 @item target tfile @var{filename}
10910 Use the file named @var{filename} as a source of trace data. Commands
10911 that examine data work as they do with a live target, but it is not
10912 possible to run any new trace experiments. @code{tstatus} will report
10913 the state of the trace run at the moment the data was saved, as well
10914 as the current trace frame you are examining. @var{filename} must be
10915 on a filesystem accessible to the host.
10920 @chapter Debugging Programs That Use Overlays
10923 If your program is too large to fit completely in your target system's
10924 memory, you can sometimes use @dfn{overlays} to work around this
10925 problem. @value{GDBN} provides some support for debugging programs that
10929 * How Overlays Work:: A general explanation of overlays.
10930 * Overlay Commands:: Managing overlays in @value{GDBN}.
10931 * Automatic Overlay Debugging:: @value{GDBN} can find out which overlays are
10932 mapped by asking the inferior.
10933 * Overlay Sample Program:: A sample program using overlays.
10936 @node How Overlays Work
10937 @section How Overlays Work
10938 @cindex mapped overlays
10939 @cindex unmapped overlays
10940 @cindex load address, overlay's
10941 @cindex mapped address
10942 @cindex overlay area
10944 Suppose you have a computer whose instruction address space is only 64
10945 kilobytes long, but which has much more memory which can be accessed by
10946 other means: special instructions, segment registers, or memory
10947 management hardware, for example. Suppose further that you want to
10948 adapt a program which is larger than 64 kilobytes to run on this system.
10950 One solution is to identify modules of your program which are relatively
10951 independent, and need not call each other directly; call these modules
10952 @dfn{overlays}. Separate the overlays from the main program, and place
10953 their machine code in the larger memory. Place your main program in
10954 instruction memory, but leave at least enough space there to hold the
10955 largest overlay as well.
10957 Now, to call a function located in an overlay, you must first copy that
10958 overlay's machine code from the large memory into the space set aside
10959 for it in the instruction memory, and then jump to its entry point
10962 @c NB: In the below the mapped area's size is greater or equal to the
10963 @c size of all overlays. This is intentional to remind the developer
10964 @c that overlays don't necessarily need to be the same size.
10968 Data Instruction Larger
10969 Address Space Address Space Address Space
10970 +-----------+ +-----------+ +-----------+
10972 +-----------+ +-----------+ +-----------+<-- overlay 1
10973 | program | | main | .----| overlay 1 | load address
10974 | variables | | program | | +-----------+
10975 | and heap | | | | | |
10976 +-----------+ | | | +-----------+<-- overlay 2
10977 | | +-----------+ | | | load address
10978 +-----------+ | | | .-| overlay 2 |
10980 mapped --->+-----------+ | | +-----------+
10981 address | | | | | |
10982 | overlay | <-' | | |
10983 | area | <---' +-----------+<-- overlay 3
10984 | | <---. | | load address
10985 +-----------+ `--| overlay 3 |
10992 @anchor{A code overlay}A code overlay
10996 The diagram (@pxref{A code overlay}) shows a system with separate data
10997 and instruction address spaces. To map an overlay, the program copies
10998 its code from the larger address space to the instruction address space.
10999 Since the overlays shown here all use the same mapped address, only one
11000 may be mapped at a time. For a system with a single address space for
11001 data and instructions, the diagram would be similar, except that the
11002 program variables and heap would share an address space with the main
11003 program and the overlay area.
11005 An overlay loaded into instruction memory and ready for use is called a
11006 @dfn{mapped} overlay; its @dfn{mapped address} is its address in the
11007 instruction memory. An overlay not present (or only partially present)
11008 in instruction memory is called @dfn{unmapped}; its @dfn{load address}
11009 is its address in the larger memory. The mapped address is also called
11010 the @dfn{virtual memory address}, or @dfn{VMA}; the load address is also
11011 called the @dfn{load memory address}, or @dfn{LMA}.
11013 Unfortunately, overlays are not a completely transparent way to adapt a
11014 program to limited instruction memory. They introduce a new set of
11015 global constraints you must keep in mind as you design your program:
11020 Before calling or returning to a function in an overlay, your program
11021 must make sure that overlay is actually mapped. Otherwise, the call or
11022 return will transfer control to the right address, but in the wrong
11023 overlay, and your program will probably crash.
11026 If the process of mapping an overlay is expensive on your system, you
11027 will need to choose your overlays carefully to minimize their effect on
11028 your program's performance.
11031 The executable file you load onto your system must contain each
11032 overlay's instructions, appearing at the overlay's load address, not its
11033 mapped address. However, each overlay's instructions must be relocated
11034 and its symbols defined as if the overlay were at its mapped address.
11035 You can use GNU linker scripts to specify different load and relocation
11036 addresses for pieces of your program; see @ref{Overlay Description,,,
11037 ld.info, Using ld: the GNU linker}.
11040 The procedure for loading executable files onto your system must be able
11041 to load their contents into the larger address space as well as the
11042 instruction and data spaces.
11046 The overlay system described above is rather simple, and could be
11047 improved in many ways:
11052 If your system has suitable bank switch registers or memory management
11053 hardware, you could use those facilities to make an overlay's load area
11054 contents simply appear at their mapped address in instruction space.
11055 This would probably be faster than copying the overlay to its mapped
11056 area in the usual way.
11059 If your overlays are small enough, you could set aside more than one
11060 overlay area, and have more than one overlay mapped at a time.
11063 You can use overlays to manage data, as well as instructions. In
11064 general, data overlays are even less transparent to your design than
11065 code overlays: whereas code overlays only require care when you call or
11066 return to functions, data overlays require care every time you access
11067 the data. Also, if you change the contents of a data overlay, you
11068 must copy its contents back out to its load address before you can copy a
11069 different data overlay into the same mapped area.
11074 @node Overlay Commands
11075 @section Overlay Commands
11077 To use @value{GDBN}'s overlay support, each overlay in your program must
11078 correspond to a separate section of the executable file. The section's
11079 virtual memory address and load memory address must be the overlay's
11080 mapped and load addresses. Identifying overlays with sections allows
11081 @value{GDBN} to determine the appropriate address of a function or
11082 variable, depending on whether the overlay is mapped or not.
11084 @value{GDBN}'s overlay commands all start with the word @code{overlay};
11085 you can abbreviate this as @code{ov} or @code{ovly}. The commands are:
11090 Disable @value{GDBN}'s overlay support. When overlay support is
11091 disabled, @value{GDBN} assumes that all functions and variables are
11092 always present at their mapped addresses. By default, @value{GDBN}'s
11093 overlay support is disabled.
11095 @item overlay manual
11096 @cindex manual overlay debugging
11097 Enable @dfn{manual} overlay debugging. In this mode, @value{GDBN}
11098 relies on you to tell it which overlays are mapped, and which are not,
11099 using the @code{overlay map-overlay} and @code{overlay unmap-overlay}
11100 commands described below.
11102 @item overlay map-overlay @var{overlay}
11103 @itemx overlay map @var{overlay}
11104 @cindex map an overlay
11105 Tell @value{GDBN} that @var{overlay} is now mapped; @var{overlay} must
11106 be the name of the object file section containing the overlay. When an
11107 overlay is mapped, @value{GDBN} assumes it can find the overlay's
11108 functions and variables at their mapped addresses. @value{GDBN} assumes
11109 that any other overlays whose mapped ranges overlap that of
11110 @var{overlay} are now unmapped.
11112 @item overlay unmap-overlay @var{overlay}
11113 @itemx overlay unmap @var{overlay}
11114 @cindex unmap an overlay
11115 Tell @value{GDBN} that @var{overlay} is no longer mapped; @var{overlay}
11116 must be the name of the object file section containing the overlay.
11117 When an overlay is unmapped, @value{GDBN} assumes it can find the
11118 overlay's functions and variables at their load addresses.
11121 Enable @dfn{automatic} overlay debugging. In this mode, @value{GDBN}
11122 consults a data structure the overlay manager maintains in the inferior
11123 to see which overlays are mapped. For details, see @ref{Automatic
11124 Overlay Debugging}.
11126 @item overlay load-target
11127 @itemx overlay load
11128 @cindex reloading the overlay table
11129 Re-read the overlay table from the inferior. Normally, @value{GDBN}
11130 re-reads the table @value{GDBN} automatically each time the inferior
11131 stops, so this command should only be necessary if you have changed the
11132 overlay mapping yourself using @value{GDBN}. This command is only
11133 useful when using automatic overlay debugging.
11135 @item overlay list-overlays
11136 @itemx overlay list
11137 @cindex listing mapped overlays
11138 Display a list of the overlays currently mapped, along with their mapped
11139 addresses, load addresses, and sizes.
11143 Normally, when @value{GDBN} prints a code address, it includes the name
11144 of the function the address falls in:
11147 (@value{GDBP}) print main
11148 $3 = @{int ()@} 0x11a0 <main>
11151 When overlay debugging is enabled, @value{GDBN} recognizes code in
11152 unmapped overlays, and prints the names of unmapped functions with
11153 asterisks around them. For example, if @code{foo} is a function in an
11154 unmapped overlay, @value{GDBN} prints it this way:
11157 (@value{GDBP}) overlay list
11158 No sections are mapped.
11159 (@value{GDBP}) print foo
11160 $5 = @{int (int)@} 0x100000 <*foo*>
11163 When @code{foo}'s overlay is mapped, @value{GDBN} prints the function's
11167 (@value{GDBP}) overlay list
11168 Section .ov.foo.text, loaded at 0x100000 - 0x100034,
11169 mapped at 0x1016 - 0x104a
11170 (@value{GDBP}) print foo
11171 $6 = @{int (int)@} 0x1016 <foo>
11174 When overlay debugging is enabled, @value{GDBN} can find the correct
11175 address for functions and variables in an overlay, whether or not the
11176 overlay is mapped. This allows most @value{GDBN} commands, like
11177 @code{break} and @code{disassemble}, to work normally, even on unmapped
11178 code. However, @value{GDBN}'s breakpoint support has some limitations:
11182 @cindex breakpoints in overlays
11183 @cindex overlays, setting breakpoints in
11184 You can set breakpoints in functions in unmapped overlays, as long as
11185 @value{GDBN} can write to the overlay at its load address.
11187 @value{GDBN} can not set hardware or simulator-based breakpoints in
11188 unmapped overlays. However, if you set a breakpoint at the end of your
11189 overlay manager (and tell @value{GDBN} which overlays are now mapped, if
11190 you are using manual overlay management), @value{GDBN} will re-set its
11191 breakpoints properly.
11195 @node Automatic Overlay Debugging
11196 @section Automatic Overlay Debugging
11197 @cindex automatic overlay debugging
11199 @value{GDBN} can automatically track which overlays are mapped and which
11200 are not, given some simple co-operation from the overlay manager in the
11201 inferior. If you enable automatic overlay debugging with the
11202 @code{overlay auto} command (@pxref{Overlay Commands}), @value{GDBN}
11203 looks in the inferior's memory for certain variables describing the
11204 current state of the overlays.
11206 Here are the variables your overlay manager must define to support
11207 @value{GDBN}'s automatic overlay debugging:
11211 @item @code{_ovly_table}:
11212 This variable must be an array of the following structures:
11217 /* The overlay's mapped address. */
11220 /* The size of the overlay, in bytes. */
11221 unsigned long size;
11223 /* The overlay's load address. */
11226 /* Non-zero if the overlay is currently mapped;
11228 unsigned long mapped;
11232 @item @code{_novlys}:
11233 This variable must be a four-byte signed integer, holding the total
11234 number of elements in @code{_ovly_table}.
11238 To decide whether a particular overlay is mapped or not, @value{GDBN}
11239 looks for an entry in @w{@code{_ovly_table}} whose @code{vma} and
11240 @code{lma} members equal the VMA and LMA of the overlay's section in the
11241 executable file. When @value{GDBN} finds a matching entry, it consults
11242 the entry's @code{mapped} member to determine whether the overlay is
11245 In addition, your overlay manager may define a function called
11246 @code{_ovly_debug_event}. If this function is defined, @value{GDBN}
11247 will silently set a breakpoint there. If the overlay manager then
11248 calls this function whenever it has changed the overlay table, this
11249 will enable @value{GDBN} to accurately keep track of which overlays
11250 are in program memory, and update any breakpoints that may be set
11251 in overlays. This will allow breakpoints to work even if the
11252 overlays are kept in ROM or other non-writable memory while they
11253 are not being executed.
11255 @node Overlay Sample Program
11256 @section Overlay Sample Program
11257 @cindex overlay example program
11259 When linking a program which uses overlays, you must place the overlays
11260 at their load addresses, while relocating them to run at their mapped
11261 addresses. To do this, you must write a linker script (@pxref{Overlay
11262 Description,,, ld.info, Using ld: the GNU linker}). Unfortunately,
11263 since linker scripts are specific to a particular host system, target
11264 architecture, and target memory layout, this manual cannot provide
11265 portable sample code demonstrating @value{GDBN}'s overlay support.
11267 However, the @value{GDBN} source distribution does contain an overlaid
11268 program, with linker scripts for a few systems, as part of its test
11269 suite. The program consists of the following files from
11270 @file{gdb/testsuite/gdb.base}:
11274 The main program file.
11276 A simple overlay manager, used by @file{overlays.c}.
11281 Overlay modules, loaded and used by @file{overlays.c}.
11284 Linker scripts for linking the test program on the @code{d10v-elf}
11285 and @code{m32r-elf} targets.
11288 You can build the test program using the @code{d10v-elf} GCC
11289 cross-compiler like this:
11292 $ d10v-elf-gcc -g -c overlays.c
11293 $ d10v-elf-gcc -g -c ovlymgr.c
11294 $ d10v-elf-gcc -g -c foo.c
11295 $ d10v-elf-gcc -g -c bar.c
11296 $ d10v-elf-gcc -g -c baz.c
11297 $ d10v-elf-gcc -g -c grbx.c
11298 $ d10v-elf-gcc -g overlays.o ovlymgr.o foo.o bar.o \
11299 baz.o grbx.o -Wl,-Td10v.ld -o overlays
11302 The build process is identical for any other architecture, except that
11303 you must substitute the appropriate compiler and linker script for the
11304 target system for @code{d10v-elf-gcc} and @code{d10v.ld}.
11308 @chapter Using @value{GDBN} with Different Languages
11311 Although programming languages generally have common aspects, they are
11312 rarely expressed in the same manner. For instance, in ANSI C,
11313 dereferencing a pointer @code{p} is accomplished by @code{*p}, but in
11314 Modula-2, it is accomplished by @code{p^}. Values can also be
11315 represented (and displayed) differently. Hex numbers in C appear as
11316 @samp{0x1ae}, while in Modula-2 they appear as @samp{1AEH}.
11318 @cindex working language
11319 Language-specific information is built into @value{GDBN} for some languages,
11320 allowing you to express operations like the above in your program's
11321 native language, and allowing @value{GDBN} to output values in a manner
11322 consistent with the syntax of your program's native language. The
11323 language you use to build expressions is called the @dfn{working
11327 * Setting:: Switching between source languages
11328 * Show:: Displaying the language
11329 * Checks:: Type and range checks
11330 * Supported Languages:: Supported languages
11331 * Unsupported Languages:: Unsupported languages
11335 @section Switching Between Source Languages
11337 There are two ways to control the working language---either have @value{GDBN}
11338 set it automatically, or select it manually yourself. You can use the
11339 @code{set language} command for either purpose. On startup, @value{GDBN}
11340 defaults to setting the language automatically. The working language is
11341 used to determine how expressions you type are interpreted, how values
11344 In addition to the working language, every source file that
11345 @value{GDBN} knows about has its own working language. For some object
11346 file formats, the compiler might indicate which language a particular
11347 source file is in. However, most of the time @value{GDBN} infers the
11348 language from the name of the file. The language of a source file
11349 controls whether C@t{++} names are demangled---this way @code{backtrace} can
11350 show each frame appropriately for its own language. There is no way to
11351 set the language of a source file from within @value{GDBN}, but you can
11352 set the language associated with a filename extension. @xref{Show, ,
11353 Displaying the Language}.
11355 This is most commonly a problem when you use a program, such
11356 as @code{cfront} or @code{f2c}, that generates C but is written in
11357 another language. In that case, make the
11358 program use @code{#line} directives in its C output; that way
11359 @value{GDBN} will know the correct language of the source code of the original
11360 program, and will display that source code, not the generated C code.
11363 * Filenames:: Filename extensions and languages.
11364 * Manually:: Setting the working language manually
11365 * Automatically:: Having @value{GDBN} infer the source language
11369 @subsection List of Filename Extensions and Languages
11371 If a source file name ends in one of the following extensions, then
11372 @value{GDBN} infers that its language is the one indicated.
11390 C@t{++} source file
11396 Objective-C source file
11400 Fortran source file
11403 Modula-2 source file
11407 Assembler source file. This actually behaves almost like C, but
11408 @value{GDBN} does not skip over function prologues when stepping.
11411 In addition, you may set the language associated with a filename
11412 extension. @xref{Show, , Displaying the Language}.
11415 @subsection Setting the Working Language
11417 If you allow @value{GDBN} to set the language automatically,
11418 expressions are interpreted the same way in your debugging session and
11421 @kindex set language
11422 If you wish, you may set the language manually. To do this, issue the
11423 command @samp{set language @var{lang}}, where @var{lang} is the name of
11424 a language, such as
11425 @code{c} or @code{modula-2}.
11426 For a list of the supported languages, type @samp{set language}.
11428 Setting the language manually prevents @value{GDBN} from updating the working
11429 language automatically. This can lead to confusion if you try
11430 to debug a program when the working language is not the same as the
11431 source language, when an expression is acceptable to both
11432 languages---but means different things. For instance, if the current
11433 source file were written in C, and @value{GDBN} was parsing Modula-2, a
11441 might not have the effect you intended. In C, this means to add
11442 @code{b} and @code{c} and place the result in @code{a}. The result
11443 printed would be the value of @code{a}. In Modula-2, this means to compare
11444 @code{a} to the result of @code{b+c}, yielding a @code{BOOLEAN} value.
11446 @node Automatically
11447 @subsection Having @value{GDBN} Infer the Source Language
11449 To have @value{GDBN} set the working language automatically, use
11450 @samp{set language local} or @samp{set language auto}. @value{GDBN}
11451 then infers the working language. That is, when your program stops in a
11452 frame (usually by encountering a breakpoint), @value{GDBN} sets the
11453 working language to the language recorded for the function in that
11454 frame. If the language for a frame is unknown (that is, if the function
11455 or block corresponding to the frame was defined in a source file that
11456 does not have a recognized extension), the current working language is
11457 not changed, and @value{GDBN} issues a warning.
11459 This may not seem necessary for most programs, which are written
11460 entirely in one source language. However, program modules and libraries
11461 written in one source language can be used by a main program written in
11462 a different source language. Using @samp{set language auto} in this
11463 case frees you from having to set the working language manually.
11466 @section Displaying the Language
11468 The following commands help you find out which language is the
11469 working language, and also what language source files were written in.
11472 @item show language
11473 @kindex show language
11474 Display the current working language. This is the
11475 language you can use with commands such as @code{print} to
11476 build and compute expressions that may involve variables in your program.
11479 @kindex info frame@r{, show the source language}
11480 Display the source language for this frame. This language becomes the
11481 working language if you use an identifier from this frame.
11482 @xref{Frame Info, ,Information about a Frame}, to identify the other
11483 information listed here.
11486 @kindex info source@r{, show the source language}
11487 Display the source language of this source file.
11488 @xref{Symbols, ,Examining the Symbol Table}, to identify the other
11489 information listed here.
11492 In unusual circumstances, you may have source files with extensions
11493 not in the standard list. You can then set the extension associated
11494 with a language explicitly:
11497 @item set extension-language @var{ext} @var{language}
11498 @kindex set extension-language
11499 Tell @value{GDBN} that source files with extension @var{ext} are to be
11500 assumed as written in the source language @var{language}.
11502 @item info extensions
11503 @kindex info extensions
11504 List all the filename extensions and the associated languages.
11508 @section Type and Range Checking
11511 @emph{Warning:} In this release, the @value{GDBN} commands for type and range
11512 checking are included, but they do not yet have any effect. This
11513 section documents the intended facilities.
11515 @c FIXME remove warning when type/range code added
11517 Some languages are designed to guard you against making seemingly common
11518 errors through a series of compile- and run-time checks. These include
11519 checking the type of arguments to functions and operators, and making
11520 sure mathematical overflows are caught at run time. Checks such as
11521 these help to ensure a program's correctness once it has been compiled
11522 by eliminating type mismatches, and providing active checks for range
11523 errors when your program is running.
11525 @value{GDBN} can check for conditions like the above if you wish.
11526 Although @value{GDBN} does not check the statements in your program,
11527 it can check expressions entered directly into @value{GDBN} for
11528 evaluation via the @code{print} command, for example. As with the
11529 working language, @value{GDBN} can also decide whether or not to check
11530 automatically based on your program's source language.
11531 @xref{Supported Languages, ,Supported Languages}, for the default
11532 settings of supported languages.
11535 * Type Checking:: An overview of type checking
11536 * Range Checking:: An overview of range checking
11539 @cindex type checking
11540 @cindex checks, type
11541 @node Type Checking
11542 @subsection An Overview of Type Checking
11544 Some languages, such as Modula-2, are strongly typed, meaning that the
11545 arguments to operators and functions have to be of the correct type,
11546 otherwise an error occurs. These checks prevent type mismatch
11547 errors from ever causing any run-time problems. For example,
11555 The second example fails because the @code{CARDINAL} 1 is not
11556 type-compatible with the @code{REAL} 2.3.
11558 For the expressions you use in @value{GDBN} commands, you can tell the
11559 @value{GDBN} type checker to skip checking;
11560 to treat any mismatches as errors and abandon the expression;
11561 or to only issue warnings when type mismatches occur,
11562 but evaluate the expression anyway. When you choose the last of
11563 these, @value{GDBN} evaluates expressions like the second example above, but
11564 also issues a warning.
11566 Even if you turn type checking off, there may be other reasons
11567 related to type that prevent @value{GDBN} from evaluating an expression.
11568 For instance, @value{GDBN} does not know how to add an @code{int} and
11569 a @code{struct foo}. These particular type errors have nothing to do
11570 with the language in use, and usually arise from expressions, such as
11571 the one described above, which make little sense to evaluate anyway.
11573 Each language defines to what degree it is strict about type. For
11574 instance, both Modula-2 and C require the arguments to arithmetical
11575 operators to be numbers. In C, enumerated types and pointers can be
11576 represented as numbers, so that they are valid arguments to mathematical
11577 operators. @xref{Supported Languages, ,Supported Languages}, for further
11578 details on specific languages.
11580 @value{GDBN} provides some additional commands for controlling the type checker:
11582 @kindex set check type
11583 @kindex show check type
11585 @item set check type auto
11586 Set type checking on or off based on the current working language.
11587 @xref{Supported Languages, ,Supported Languages}, for the default settings for
11590 @item set check type on
11591 @itemx set check type off
11592 Set type checking on or off, overriding the default setting for the
11593 current working language. Issue a warning if the setting does not
11594 match the language default. If any type mismatches occur in
11595 evaluating an expression while type checking is on, @value{GDBN} prints a
11596 message and aborts evaluation of the expression.
11598 @item set check type warn
11599 Cause the type checker to issue warnings, but to always attempt to
11600 evaluate the expression. Evaluating the expression may still
11601 be impossible for other reasons. For example, @value{GDBN} cannot add
11602 numbers and structures.
11605 Show the current setting of the type checker, and whether or not @value{GDBN}
11606 is setting it automatically.
11609 @cindex range checking
11610 @cindex checks, range
11611 @node Range Checking
11612 @subsection An Overview of Range Checking
11614 In some languages (such as Modula-2), it is an error to exceed the
11615 bounds of a type; this is enforced with run-time checks. Such range
11616 checking is meant to ensure program correctness by making sure
11617 computations do not overflow, or indices on an array element access do
11618 not exceed the bounds of the array.
11620 For expressions you use in @value{GDBN} commands, you can tell
11621 @value{GDBN} to treat range errors in one of three ways: ignore them,
11622 always treat them as errors and abandon the expression, or issue
11623 warnings but evaluate the expression anyway.
11625 A range error can result from numerical overflow, from exceeding an
11626 array index bound, or when you type a constant that is not a member
11627 of any type. Some languages, however, do not treat overflows as an
11628 error. In many implementations of C, mathematical overflow causes the
11629 result to ``wrap around'' to lower values---for example, if @var{m} is
11630 the largest integer value, and @var{s} is the smallest, then
11633 @var{m} + 1 @result{} @var{s}
11636 This, too, is specific to individual languages, and in some cases
11637 specific to individual compilers or machines. @xref{Supported Languages, ,
11638 Supported Languages}, for further details on specific languages.
11640 @value{GDBN} provides some additional commands for controlling the range checker:
11642 @kindex set check range
11643 @kindex show check range
11645 @item set check range auto
11646 Set range checking on or off based on the current working language.
11647 @xref{Supported Languages, ,Supported Languages}, for the default settings for
11650 @item set check range on
11651 @itemx set check range off
11652 Set range checking on or off, overriding the default setting for the
11653 current working language. A warning is issued if the setting does not
11654 match the language default. If a range error occurs and range checking is on,
11655 then a message is printed and evaluation of the expression is aborted.
11657 @item set check range warn
11658 Output messages when the @value{GDBN} range checker detects a range error,
11659 but attempt to evaluate the expression anyway. Evaluating the
11660 expression may still be impossible for other reasons, such as accessing
11661 memory that the process does not own (a typical example from many Unix
11665 Show the current setting of the range checker, and whether or not it is
11666 being set automatically by @value{GDBN}.
11669 @node Supported Languages
11670 @section Supported Languages
11672 @value{GDBN} supports C, C@t{++}, D, Objective-C, Fortran, Java, OpenCL C, Pascal,
11673 assembly, Modula-2, and Ada.
11674 @c This is false ...
11675 Some @value{GDBN} features may be used in expressions regardless of the
11676 language you use: the @value{GDBN} @code{@@} and @code{::} operators,
11677 and the @samp{@{type@}addr} construct (@pxref{Expressions,
11678 ,Expressions}) can be used with the constructs of any supported
11681 The following sections detail to what degree each source language is
11682 supported by @value{GDBN}. These sections are not meant to be language
11683 tutorials or references, but serve only as a reference guide to what the
11684 @value{GDBN} expression parser accepts, and what input and output
11685 formats should look like for different languages. There are many good
11686 books written on each of these languages; please look to these for a
11687 language reference or tutorial.
11690 * C:: C and C@t{++}
11692 * Objective-C:: Objective-C
11693 * OpenCL C:: OpenCL C
11694 * Fortran:: Fortran
11696 * Modula-2:: Modula-2
11701 @subsection C and C@t{++}
11703 @cindex C and C@t{++}
11704 @cindex expressions in C or C@t{++}
11706 Since C and C@t{++} are so closely related, many features of @value{GDBN} apply
11707 to both languages. Whenever this is the case, we discuss those languages
11711 @cindex @code{g++}, @sc{gnu} C@t{++} compiler
11712 @cindex @sc{gnu} C@t{++}
11713 The C@t{++} debugging facilities are jointly implemented by the C@t{++}
11714 compiler and @value{GDBN}. Therefore, to debug your C@t{++} code
11715 effectively, you must compile your C@t{++} programs with a supported
11716 C@t{++} compiler, such as @sc{gnu} @code{g++}, or the HP ANSI C@t{++}
11717 compiler (@code{aCC}).
11719 For best results when using @sc{gnu} C@t{++}, use the DWARF 2 debugging
11720 format; if it doesn't work on your system, try the stabs+ debugging
11721 format. You can select those formats explicitly with the @code{g++}
11722 command-line options @option{-gdwarf-2} and @option{-gstabs+}.
11723 @xref{Debugging Options,,Options for Debugging Your Program or GCC,
11724 gcc.info, Using the @sc{gnu} Compiler Collection (GCC)}.
11727 * C Operators:: C and C@t{++} operators
11728 * C Constants:: C and C@t{++} constants
11729 * C Plus Plus Expressions:: C@t{++} expressions
11730 * C Defaults:: Default settings for C and C@t{++}
11731 * C Checks:: C and C@t{++} type and range checks
11732 * Debugging C:: @value{GDBN} and C
11733 * Debugging C Plus Plus:: @value{GDBN} features for C@t{++}
11734 * Decimal Floating Point:: Numbers in Decimal Floating Point format
11738 @subsubsection C and C@t{++} Operators
11740 @cindex C and C@t{++} operators
11742 Operators must be defined on values of specific types. For instance,
11743 @code{+} is defined on numbers, but not on structures. Operators are
11744 often defined on groups of types.
11746 For the purposes of C and C@t{++}, the following definitions hold:
11751 @emph{Integral types} include @code{int} with any of its storage-class
11752 specifiers; @code{char}; @code{enum}; and, for C@t{++}, @code{bool}.
11755 @emph{Floating-point types} include @code{float}, @code{double}, and
11756 @code{long double} (if supported by the target platform).
11759 @emph{Pointer types} include all types defined as @code{(@var{type} *)}.
11762 @emph{Scalar types} include all of the above.
11767 The following operators are supported. They are listed here
11768 in order of increasing precedence:
11772 The comma or sequencing operator. Expressions in a comma-separated list
11773 are evaluated from left to right, with the result of the entire
11774 expression being the last expression evaluated.
11777 Assignment. The value of an assignment expression is the value
11778 assigned. Defined on scalar types.
11781 Used in an expression of the form @w{@code{@var{a} @var{op}= @var{b}}},
11782 and translated to @w{@code{@var{a} = @var{a op b}}}.
11783 @w{@code{@var{op}=}} and @code{=} have the same precedence.
11784 @var{op} is any one of the operators @code{|}, @code{^}, @code{&},
11785 @code{<<}, @code{>>}, @code{+}, @code{-}, @code{*}, @code{/}, @code{%}.
11788 The ternary operator. @code{@var{a} ? @var{b} : @var{c}} can be thought
11789 of as: if @var{a} then @var{b} else @var{c}. @var{a} should be of an
11793 Logical @sc{or}. Defined on integral types.
11796 Logical @sc{and}. Defined on integral types.
11799 Bitwise @sc{or}. Defined on integral types.
11802 Bitwise exclusive-@sc{or}. Defined on integral types.
11805 Bitwise @sc{and}. Defined on integral types.
11808 Equality and inequality. Defined on scalar types. The value of these
11809 expressions is 0 for false and non-zero for true.
11811 @item <@r{, }>@r{, }<=@r{, }>=
11812 Less than, greater than, less than or equal, greater than or equal.
11813 Defined on scalar types. The value of these expressions is 0 for false
11814 and non-zero for true.
11817 left shift, and right shift. Defined on integral types.
11820 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
11823 Addition and subtraction. Defined on integral types, floating-point types and
11826 @item *@r{, }/@r{, }%
11827 Multiplication, division, and modulus. Multiplication and division are
11828 defined on integral and floating-point types. Modulus is defined on
11832 Increment and decrement. When appearing before a variable, the
11833 operation is performed before the variable is used in an expression;
11834 when appearing after it, the variable's value is used before the
11835 operation takes place.
11838 Pointer dereferencing. Defined on pointer types. Same precedence as
11842 Address operator. Defined on variables. Same precedence as @code{++}.
11844 For debugging C@t{++}, @value{GDBN} implements a use of @samp{&} beyond what is
11845 allowed in the C@t{++} language itself: you can use @samp{&(&@var{ref})}
11846 to examine the address
11847 where a C@t{++} reference variable (declared with @samp{&@var{ref}}) is
11851 Negative. Defined on integral and floating-point types. Same
11852 precedence as @code{++}.
11855 Logical negation. Defined on integral types. Same precedence as
11859 Bitwise complement operator. Defined on integral types. Same precedence as
11864 Structure member, and pointer-to-structure member. For convenience,
11865 @value{GDBN} regards the two as equivalent, choosing whether to dereference a
11866 pointer based on the stored type information.
11867 Defined on @code{struct} and @code{union} data.
11870 Dereferences of pointers to members.
11873 Array indexing. @code{@var{a}[@var{i}]} is defined as
11874 @code{*(@var{a}+@var{i})}. Same precedence as @code{->}.
11877 Function parameter list. Same precedence as @code{->}.
11880 C@t{++} scope resolution operator. Defined on @code{struct}, @code{union},
11881 and @code{class} types.
11884 Doubled colons also represent the @value{GDBN} scope operator
11885 (@pxref{Expressions, ,Expressions}). Same precedence as @code{::},
11889 If an operator is redefined in the user code, @value{GDBN} usually
11890 attempts to invoke the redefined version instead of using the operator's
11891 predefined meaning.
11894 @subsubsection C and C@t{++} Constants
11896 @cindex C and C@t{++} constants
11898 @value{GDBN} allows you to express the constants of C and C@t{++} in the
11903 Integer constants are a sequence of digits. Octal constants are
11904 specified by a leading @samp{0} (i.e.@: zero), and hexadecimal constants
11905 by a leading @samp{0x} or @samp{0X}. Constants may also end with a letter
11906 @samp{l}, specifying that the constant should be treated as a
11910 Floating point constants are a sequence of digits, followed by a decimal
11911 point, followed by a sequence of digits, and optionally followed by an
11912 exponent. An exponent is of the form:
11913 @samp{@w{e@r{[[}+@r{]|}-@r{]}@var{nnn}}}, where @var{nnn} is another
11914 sequence of digits. The @samp{+} is optional for positive exponents.
11915 A floating-point constant may also end with a letter @samp{f} or
11916 @samp{F}, specifying that the constant should be treated as being of
11917 the @code{float} (as opposed to the default @code{double}) type; or with
11918 a letter @samp{l} or @samp{L}, which specifies a @code{long double}
11922 Enumerated constants consist of enumerated identifiers, or their
11923 integral equivalents.
11926 Character constants are a single character surrounded by single quotes
11927 (@code{'}), or a number---the ordinal value of the corresponding character
11928 (usually its @sc{ascii} value). Within quotes, the single character may
11929 be represented by a letter or by @dfn{escape sequences}, which are of
11930 the form @samp{\@var{nnn}}, where @var{nnn} is the octal representation
11931 of the character's ordinal value; or of the form @samp{\@var{x}}, where
11932 @samp{@var{x}} is a predefined special character---for example,
11933 @samp{\n} for newline.
11936 String constants are a sequence of character constants surrounded by
11937 double quotes (@code{"}). Any valid character constant (as described
11938 above) may appear. Double quotes within the string must be preceded by
11939 a backslash, so for instance @samp{"a\"b'c"} is a string of five
11943 Pointer constants are an integral value. You can also write pointers
11944 to constants using the C operator @samp{&}.
11947 Array constants are comma-separated lists surrounded by braces @samp{@{}
11948 and @samp{@}}; for example, @samp{@{1,2,3@}} is a three-element array of
11949 integers, @samp{@{@{1,2@}, @{3,4@}, @{5,6@}@}} is a three-by-two array,
11950 and @samp{@{&"hi", &"there", &"fred"@}} is a three-element array of pointers.
11953 @node C Plus Plus Expressions
11954 @subsubsection C@t{++} Expressions
11956 @cindex expressions in C@t{++}
11957 @value{GDBN} expression handling can interpret most C@t{++} expressions.
11959 @cindex debugging C@t{++} programs
11960 @cindex C@t{++} compilers
11961 @cindex debug formats and C@t{++}
11962 @cindex @value{NGCC} and C@t{++}
11964 @emph{Warning:} @value{GDBN} can only debug C@t{++} code if you use the
11965 proper compiler and the proper debug format. Currently, @value{GDBN}
11966 works best when debugging C@t{++} code that is compiled with
11967 @value{NGCC} 2.95.3 or with @value{NGCC} 3.1 or newer, using the options
11968 @option{-gdwarf-2} or @option{-gstabs+}. DWARF 2 is preferred over
11969 stabs+. Most configurations of @value{NGCC} emit either DWARF 2 or
11970 stabs+ as their default debug format, so you usually don't need to
11971 specify a debug format explicitly. Other compilers and/or debug formats
11972 are likely to work badly or not at all when using @value{GDBN} to debug
11978 @cindex member functions
11980 Member function calls are allowed; you can use expressions like
11983 count = aml->GetOriginal(x, y)
11986 @vindex this@r{, inside C@t{++} member functions}
11987 @cindex namespace in C@t{++}
11989 While a member function is active (in the selected stack frame), your
11990 expressions have the same namespace available as the member function;
11991 that is, @value{GDBN} allows implicit references to the class instance
11992 pointer @code{this} following the same rules as C@t{++}.
11994 @cindex call overloaded functions
11995 @cindex overloaded functions, calling
11996 @cindex type conversions in C@t{++}
11998 You can call overloaded functions; @value{GDBN} resolves the function
11999 call to the right definition, with some restrictions. @value{GDBN} does not
12000 perform overload resolution involving user-defined type conversions,
12001 calls to constructors, or instantiations of templates that do not exist
12002 in the program. It also cannot handle ellipsis argument lists or
12005 It does perform integral conversions and promotions, floating-point
12006 promotions, arithmetic conversions, pointer conversions, conversions of
12007 class objects to base classes, and standard conversions such as those of
12008 functions or arrays to pointers; it requires an exact match on the
12009 number of function arguments.
12011 Overload resolution is always performed, unless you have specified
12012 @code{set overload-resolution off}. @xref{Debugging C Plus Plus,
12013 ,@value{GDBN} Features for C@t{++}}.
12015 You must specify @code{set overload-resolution off} in order to use an
12016 explicit function signature to call an overloaded function, as in
12018 p 'foo(char,int)'('x', 13)
12021 The @value{GDBN} command-completion facility can simplify this;
12022 see @ref{Completion, ,Command Completion}.
12024 @cindex reference declarations
12026 @value{GDBN} understands variables declared as C@t{++} references; you can use
12027 them in expressions just as you do in C@t{++} source---they are automatically
12030 In the parameter list shown when @value{GDBN} displays a frame, the values of
12031 reference variables are not displayed (unlike other variables); this
12032 avoids clutter, since references are often used for large structures.
12033 The @emph{address} of a reference variable is always shown, unless
12034 you have specified @samp{set print address off}.
12037 @value{GDBN} supports the C@t{++} name resolution operator @code{::}---your
12038 expressions can use it just as expressions in your program do. Since
12039 one scope may be defined in another, you can use @code{::} repeatedly if
12040 necessary, for example in an expression like
12041 @samp{@var{scope1}::@var{scope2}::@var{name}}. @value{GDBN} also allows
12042 resolving name scope by reference to source files, in both C and C@t{++}
12043 debugging (@pxref{Variables, ,Program Variables}).
12046 In addition, when used with HP's C@t{++} compiler, @value{GDBN} supports
12047 calling virtual functions correctly, printing out virtual bases of
12048 objects, calling functions in a base subobject, casting objects, and
12049 invoking user-defined operators.
12052 @subsubsection C and C@t{++} Defaults
12054 @cindex C and C@t{++} defaults
12056 If you allow @value{GDBN} to set type and range checking automatically, they
12057 both default to @code{off} whenever the working language changes to
12058 C or C@t{++}. This happens regardless of whether you or @value{GDBN}
12059 selects the working language.
12061 If you allow @value{GDBN} to set the language automatically, it
12062 recognizes source files whose names end with @file{.c}, @file{.C}, or
12063 @file{.cc}, etc, and when @value{GDBN} enters code compiled from one of
12064 these files, it sets the working language to C or C@t{++}.
12065 @xref{Automatically, ,Having @value{GDBN} Infer the Source Language},
12066 for further details.
12068 @c Type checking is (a) primarily motivated by Modula-2, and (b)
12069 @c unimplemented. If (b) changes, it might make sense to let this node
12070 @c appear even if Mod-2 does not, but meanwhile ignore it. roland 16jul93.
12073 @subsubsection C and C@t{++} Type and Range Checks
12075 @cindex C and C@t{++} checks
12077 By default, when @value{GDBN} parses C or C@t{++} expressions, type checking
12078 is not used. However, if you turn type checking on, @value{GDBN}
12079 considers two variables type equivalent if:
12083 The two variables are structured and have the same structure, union, or
12087 The two variables have the same type name, or types that have been
12088 declared equivalent through @code{typedef}.
12091 @c leaving this out because neither J Gilmore nor R Pesch understand it.
12094 The two @code{struct}, @code{union}, or @code{enum} variables are
12095 declared in the same declaration. (Note: this may not be true for all C
12100 Range checking, if turned on, is done on mathematical operations. Array
12101 indices are not checked, since they are often used to index a pointer
12102 that is not itself an array.
12105 @subsubsection @value{GDBN} and C
12107 The @code{set print union} and @code{show print union} commands apply to
12108 the @code{union} type. When set to @samp{on}, any @code{union} that is
12109 inside a @code{struct} or @code{class} is also printed. Otherwise, it
12110 appears as @samp{@{...@}}.
12112 The @code{@@} operator aids in the debugging of dynamic arrays, formed
12113 with pointers and a memory allocation function. @xref{Expressions,
12116 @node Debugging C Plus Plus
12117 @subsubsection @value{GDBN} Features for C@t{++}
12119 @cindex commands for C@t{++}
12121 Some @value{GDBN} commands are particularly useful with C@t{++}, and some are
12122 designed specifically for use with C@t{++}. Here is a summary:
12125 @cindex break in overloaded functions
12126 @item @r{breakpoint menus}
12127 When you want a breakpoint in a function whose name is overloaded,
12128 @value{GDBN} has the capability to display a menu of possible breakpoint
12129 locations to help you specify which function definition you want.
12130 @xref{Ambiguous Expressions,,Ambiguous Expressions}.
12132 @cindex overloading in C@t{++}
12133 @item rbreak @var{regex}
12134 Setting breakpoints using regular expressions is helpful for setting
12135 breakpoints on overloaded functions that are not members of any special
12137 @xref{Set Breaks, ,Setting Breakpoints}.
12139 @cindex C@t{++} exception handling
12142 Debug C@t{++} exception handling using these commands. @xref{Set
12143 Catchpoints, , Setting Catchpoints}.
12145 @cindex inheritance
12146 @item ptype @var{typename}
12147 Print inheritance relationships as well as other information for type
12149 @xref{Symbols, ,Examining the Symbol Table}.
12151 @cindex C@t{++} symbol display
12152 @item set print demangle
12153 @itemx show print demangle
12154 @itemx set print asm-demangle
12155 @itemx show print asm-demangle
12156 Control whether C@t{++} symbols display in their source form, both when
12157 displaying code as C@t{++} source and when displaying disassemblies.
12158 @xref{Print Settings, ,Print Settings}.
12160 @item set print object
12161 @itemx show print object
12162 Choose whether to print derived (actual) or declared types of objects.
12163 @xref{Print Settings, ,Print Settings}.
12165 @item set print vtbl
12166 @itemx show print vtbl
12167 Control the format for printing virtual function tables.
12168 @xref{Print Settings, ,Print Settings}.
12169 (The @code{vtbl} commands do not work on programs compiled with the HP
12170 ANSI C@t{++} compiler (@code{aCC}).)
12172 @kindex set overload-resolution
12173 @cindex overloaded functions, overload resolution
12174 @item set overload-resolution on
12175 Enable overload resolution for C@t{++} expression evaluation. The default
12176 is on. For overloaded functions, @value{GDBN} evaluates the arguments
12177 and searches for a function whose signature matches the argument types,
12178 using the standard C@t{++} conversion rules (see @ref{C Plus Plus
12179 Expressions, ,C@t{++} Expressions}, for details).
12180 If it cannot find a match, it emits a message.
12182 @item set overload-resolution off
12183 Disable overload resolution for C@t{++} expression evaluation. For
12184 overloaded functions that are not class member functions, @value{GDBN}
12185 chooses the first function of the specified name that it finds in the
12186 symbol table, whether or not its arguments are of the correct type. For
12187 overloaded functions that are class member functions, @value{GDBN}
12188 searches for a function whose signature @emph{exactly} matches the
12191 @kindex show overload-resolution
12192 @item show overload-resolution
12193 Show the current setting of overload resolution.
12195 @item @r{Overloaded symbol names}
12196 You can specify a particular definition of an overloaded symbol, using
12197 the same notation that is used to declare such symbols in C@t{++}: type
12198 @code{@var{symbol}(@var{types})} rather than just @var{symbol}. You can
12199 also use the @value{GDBN} command-line word completion facilities to list the
12200 available choices, or to finish the type list for you.
12201 @xref{Completion,, Command Completion}, for details on how to do this.
12204 @node Decimal Floating Point
12205 @subsubsection Decimal Floating Point format
12206 @cindex decimal floating point format
12208 @value{GDBN} can examine, set and perform computations with numbers in
12209 decimal floating point format, which in the C language correspond to the
12210 @code{_Decimal32}, @code{_Decimal64} and @code{_Decimal128} types as
12211 specified by the extension to support decimal floating-point arithmetic.
12213 There are two encodings in use, depending on the architecture: BID (Binary
12214 Integer Decimal) for x86 and x86-64, and DPD (Densely Packed Decimal) for
12215 PowerPC. @value{GDBN} will use the appropriate encoding for the configured
12218 Because of a limitation in @file{libdecnumber}, the library used by @value{GDBN}
12219 to manipulate decimal floating point numbers, it is not possible to convert
12220 (using a cast, for example) integers wider than 32-bit to decimal float.
12222 In addition, in order to imitate @value{GDBN}'s behaviour with binary floating
12223 point computations, error checking in decimal float operations ignores
12224 underflow, overflow and divide by zero exceptions.
12226 In the PowerPC architecture, @value{GDBN} provides a set of pseudo-registers
12227 to inspect @code{_Decimal128} values stored in floating point registers.
12228 See @ref{PowerPC,,PowerPC} for more details.
12234 @value{GDBN} can be used to debug programs written in D and compiled with
12235 GDC, LDC or DMD compilers. Currently @value{GDBN} supports only one D
12236 specific feature --- dynamic arrays.
12239 @subsection Objective-C
12241 @cindex Objective-C
12242 This section provides information about some commands and command
12243 options that are useful for debugging Objective-C code. See also
12244 @ref{Symbols, info classes}, and @ref{Symbols, info selectors}, for a
12245 few more commands specific to Objective-C support.
12248 * Method Names in Commands::
12249 * The Print Command with Objective-C::
12252 @node Method Names in Commands
12253 @subsubsection Method Names in Commands
12255 The following commands have been extended to accept Objective-C method
12256 names as line specifications:
12258 @kindex clear@r{, and Objective-C}
12259 @kindex break@r{, and Objective-C}
12260 @kindex info line@r{, and Objective-C}
12261 @kindex jump@r{, and Objective-C}
12262 @kindex list@r{, and Objective-C}
12266 @item @code{info line}
12271 A fully qualified Objective-C method name is specified as
12274 -[@var{Class} @var{methodName}]
12277 where the minus sign is used to indicate an instance method and a
12278 plus sign (not shown) is used to indicate a class method. The class
12279 name @var{Class} and method name @var{methodName} are enclosed in
12280 brackets, similar to the way messages are specified in Objective-C
12281 source code. For example, to set a breakpoint at the @code{create}
12282 instance method of class @code{Fruit} in the program currently being
12286 break -[Fruit create]
12289 To list ten program lines around the @code{initialize} class method,
12293 list +[NSText initialize]
12296 In the current version of @value{GDBN}, the plus or minus sign is
12297 required. In future versions of @value{GDBN}, the plus or minus
12298 sign will be optional, but you can use it to narrow the search. It
12299 is also possible to specify just a method name:
12305 You must specify the complete method name, including any colons. If
12306 your program's source files contain more than one @code{create} method,
12307 you'll be presented with a numbered list of classes that implement that
12308 method. Indicate your choice by number, or type @samp{0} to exit if
12311 As another example, to clear a breakpoint established at the
12312 @code{makeKeyAndOrderFront:} method of the @code{NSWindow} class, enter:
12315 clear -[NSWindow makeKeyAndOrderFront:]
12318 @node The Print Command with Objective-C
12319 @subsubsection The Print Command With Objective-C
12320 @cindex Objective-C, print objects
12321 @kindex print-object
12322 @kindex po @r{(@code{print-object})}
12324 The print command has also been extended to accept methods. For example:
12327 print -[@var{object} hash]
12330 @cindex print an Objective-C object description
12331 @cindex @code{_NSPrintForDebugger}, and printing Objective-C objects
12333 will tell @value{GDBN} to send the @code{hash} message to @var{object}
12334 and print the result. Also, an additional command has been added,
12335 @code{print-object} or @code{po} for short, which is meant to print
12336 the description of an object. However, this command may only work
12337 with certain Objective-C libraries that have a particular hook
12338 function, @code{_NSPrintForDebugger}, defined.
12341 @subsection OpenCL C
12344 This section provides information about @value{GDBN}s OpenCL C support.
12347 * OpenCL C Datatypes::
12348 * OpenCL C Expressions::
12349 * OpenCL C Operators::
12352 @node OpenCL C Datatypes
12353 @subsubsection OpenCL C Datatypes
12355 @cindex OpenCL C Datatypes
12356 @value{GDBN} supports the builtin scalar and vector datatypes specified
12357 by OpenCL 1.1. In addition the half- and double-precision floating point
12358 data types of the @code{cl_khr_fp16} and @code{cl_khr_fp64} OpenCL
12359 extensions are also known to @value{GDBN}.
12361 @node OpenCL C Expressions
12362 @subsubsection OpenCL C Expressions
12364 @cindex OpenCL C Expressions
12365 @value{GDBN} supports accesses to vector components including the access as
12366 lvalue where possible. Since OpenCL C is based on C99 most C expressions
12367 supported by @value{GDBN} can be used as well.
12369 @node OpenCL C Operators
12370 @subsubsection OpenCL C Operators
12372 @cindex OpenCL C Operators
12373 @value{GDBN} supports the operators specified by OpenCL 1.1 for scalar and
12377 @subsection Fortran
12378 @cindex Fortran-specific support in @value{GDBN}
12380 @value{GDBN} can be used to debug programs written in Fortran, but it
12381 currently supports only the features of Fortran 77 language.
12383 @cindex trailing underscore, in Fortran symbols
12384 Some Fortran compilers (@sc{gnu} Fortran 77 and Fortran 95 compilers
12385 among them) append an underscore to the names of variables and
12386 functions. When you debug programs compiled by those compilers, you
12387 will need to refer to variables and functions with a trailing
12391 * Fortran Operators:: Fortran operators and expressions
12392 * Fortran Defaults:: Default settings for Fortran
12393 * Special Fortran Commands:: Special @value{GDBN} commands for Fortran
12396 @node Fortran Operators
12397 @subsubsection Fortran Operators and Expressions
12399 @cindex Fortran operators and expressions
12401 Operators must be defined on values of specific types. For instance,
12402 @code{+} is defined on numbers, but not on characters or other non-
12403 arithmetic types. Operators are often defined on groups of types.
12407 The exponentiation operator. It raises the first operand to the power
12411 The range operator. Normally used in the form of array(low:high) to
12412 represent a section of array.
12415 The access component operator. Normally used to access elements in derived
12416 types. Also suitable for unions. As unions aren't part of regular Fortran,
12417 this can only happen when accessing a register that uses a gdbarch-defined
12421 @node Fortran Defaults
12422 @subsubsection Fortran Defaults
12424 @cindex Fortran Defaults
12426 Fortran symbols are usually case-insensitive, so @value{GDBN} by
12427 default uses case-insensitive matches for Fortran symbols. You can
12428 change that with the @samp{set case-insensitive} command, see
12429 @ref{Symbols}, for the details.
12431 @node Special Fortran Commands
12432 @subsubsection Special Fortran Commands
12434 @cindex Special Fortran commands
12436 @value{GDBN} has some commands to support Fortran-specific features,
12437 such as displaying common blocks.
12440 @cindex @code{COMMON} blocks, Fortran
12441 @kindex info common
12442 @item info common @r{[}@var{common-name}@r{]}
12443 This command prints the values contained in the Fortran @code{COMMON}
12444 block whose name is @var{common-name}. With no argument, the names of
12445 all @code{COMMON} blocks visible at the current program location are
12452 @cindex Pascal support in @value{GDBN}, limitations
12453 Debugging Pascal programs which use sets, subranges, file variables, or
12454 nested functions does not currently work. @value{GDBN} does not support
12455 entering expressions, printing values, or similar features using Pascal
12458 The Pascal-specific command @code{set print pascal_static-members}
12459 controls whether static members of Pascal objects are displayed.
12460 @xref{Print Settings, pascal_static-members}.
12463 @subsection Modula-2
12465 @cindex Modula-2, @value{GDBN} support
12467 The extensions made to @value{GDBN} to support Modula-2 only support
12468 output from the @sc{gnu} Modula-2 compiler (which is currently being
12469 developed). Other Modula-2 compilers are not currently supported, and
12470 attempting to debug executables produced by them is most likely
12471 to give an error as @value{GDBN} reads in the executable's symbol
12474 @cindex expressions in Modula-2
12476 * M2 Operators:: Built-in operators
12477 * Built-In Func/Proc:: Built-in functions and procedures
12478 * M2 Constants:: Modula-2 constants
12479 * M2 Types:: Modula-2 types
12480 * M2 Defaults:: Default settings for Modula-2
12481 * Deviations:: Deviations from standard Modula-2
12482 * M2 Checks:: Modula-2 type and range checks
12483 * M2 Scope:: The scope operators @code{::} and @code{.}
12484 * GDB/M2:: @value{GDBN} and Modula-2
12488 @subsubsection Operators
12489 @cindex Modula-2 operators
12491 Operators must be defined on values of specific types. For instance,
12492 @code{+} is defined on numbers, but not on structures. Operators are
12493 often defined on groups of types. For the purposes of Modula-2, the
12494 following definitions hold:
12499 @emph{Integral types} consist of @code{INTEGER}, @code{CARDINAL}, and
12503 @emph{Character types} consist of @code{CHAR} and its subranges.
12506 @emph{Floating-point types} consist of @code{REAL}.
12509 @emph{Pointer types} consist of anything declared as @code{POINTER TO
12513 @emph{Scalar types} consist of all of the above.
12516 @emph{Set types} consist of @code{SET} and @code{BITSET} types.
12519 @emph{Boolean types} consist of @code{BOOLEAN}.
12523 The following operators are supported, and appear in order of
12524 increasing precedence:
12528 Function argument or array index separator.
12531 Assignment. The value of @var{var} @code{:=} @var{value} is
12535 Less than, greater than on integral, floating-point, or enumerated
12539 Less than or equal to, greater than or equal to
12540 on integral, floating-point and enumerated types, or set inclusion on
12541 set types. Same precedence as @code{<}.
12543 @item =@r{, }<>@r{, }#
12544 Equality and two ways of expressing inequality, valid on scalar types.
12545 Same precedence as @code{<}. In @value{GDBN} scripts, only @code{<>} is
12546 available for inequality, since @code{#} conflicts with the script
12550 Set membership. Defined on set types and the types of their members.
12551 Same precedence as @code{<}.
12554 Boolean disjunction. Defined on boolean types.
12557 Boolean conjunction. Defined on boolean types.
12560 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
12563 Addition and subtraction on integral and floating-point types, or union
12564 and difference on set types.
12567 Multiplication on integral and floating-point types, or set intersection
12571 Division on floating-point types, or symmetric set difference on set
12572 types. Same precedence as @code{*}.
12575 Integer division and remainder. Defined on integral types. Same
12576 precedence as @code{*}.
12579 Negative. Defined on @code{INTEGER} and @code{REAL} data.
12582 Pointer dereferencing. Defined on pointer types.
12585 Boolean negation. Defined on boolean types. Same precedence as
12589 @code{RECORD} field selector. Defined on @code{RECORD} data. Same
12590 precedence as @code{^}.
12593 Array indexing. Defined on @code{ARRAY} data. Same precedence as @code{^}.
12596 Procedure argument list. Defined on @code{PROCEDURE} objects. Same precedence
12600 @value{GDBN} and Modula-2 scope operators.
12604 @emph{Warning:} Set expressions and their operations are not yet supported, so @value{GDBN}
12605 treats the use of the operator @code{IN}, or the use of operators
12606 @code{+}, @code{-}, @code{*}, @code{/}, @code{=}, , @code{<>}, @code{#},
12607 @code{<=}, and @code{>=} on sets as an error.
12611 @node Built-In Func/Proc
12612 @subsubsection Built-in Functions and Procedures
12613 @cindex Modula-2 built-ins
12615 Modula-2 also makes available several built-in procedures and functions.
12616 In describing these, the following metavariables are used:
12621 represents an @code{ARRAY} variable.
12624 represents a @code{CHAR} constant or variable.
12627 represents a variable or constant of integral type.
12630 represents an identifier that belongs to a set. Generally used in the
12631 same function with the metavariable @var{s}. The type of @var{s} should
12632 be @code{SET OF @var{mtype}} (where @var{mtype} is the type of @var{m}).
12635 represents a variable or constant of integral or floating-point type.
12638 represents a variable or constant of floating-point type.
12644 represents a variable.
12647 represents a variable or constant of one of many types. See the
12648 explanation of the function for details.
12651 All Modula-2 built-in procedures also return a result, described below.
12655 Returns the absolute value of @var{n}.
12658 If @var{c} is a lower case letter, it returns its upper case
12659 equivalent, otherwise it returns its argument.
12662 Returns the character whose ordinal value is @var{i}.
12665 Decrements the value in the variable @var{v} by one. Returns the new value.
12667 @item DEC(@var{v},@var{i})
12668 Decrements the value in the variable @var{v} by @var{i}. Returns the
12671 @item EXCL(@var{m},@var{s})
12672 Removes the element @var{m} from the set @var{s}. Returns the new
12675 @item FLOAT(@var{i})
12676 Returns the floating point equivalent of the integer @var{i}.
12678 @item HIGH(@var{a})
12679 Returns the index of the last member of @var{a}.
12682 Increments the value in the variable @var{v} by one. Returns the new value.
12684 @item INC(@var{v},@var{i})
12685 Increments the value in the variable @var{v} by @var{i}. Returns the
12688 @item INCL(@var{m},@var{s})
12689 Adds the element @var{m} to the set @var{s} if it is not already
12690 there. Returns the new set.
12693 Returns the maximum value of the type @var{t}.
12696 Returns the minimum value of the type @var{t}.
12699 Returns boolean TRUE if @var{i} is an odd number.
12702 Returns the ordinal value of its argument. For example, the ordinal
12703 value of a character is its @sc{ascii} value (on machines supporting the
12704 @sc{ascii} character set). @var{x} must be of an ordered type, which include
12705 integral, character and enumerated types.
12707 @item SIZE(@var{x})
12708 Returns the size of its argument. @var{x} can be a variable or a type.
12710 @item TRUNC(@var{r})
12711 Returns the integral part of @var{r}.
12713 @item TSIZE(@var{x})
12714 Returns the size of its argument. @var{x} can be a variable or a type.
12716 @item VAL(@var{t},@var{i})
12717 Returns the member of the type @var{t} whose ordinal value is @var{i}.
12721 @emph{Warning:} Sets and their operations are not yet supported, so
12722 @value{GDBN} treats the use of procedures @code{INCL} and @code{EXCL} as
12726 @cindex Modula-2 constants
12728 @subsubsection Constants
12730 @value{GDBN} allows you to express the constants of Modula-2 in the following
12736 Integer constants are simply a sequence of digits. When used in an
12737 expression, a constant is interpreted to be type-compatible with the
12738 rest of the expression. Hexadecimal integers are specified by a
12739 trailing @samp{H}, and octal integers by a trailing @samp{B}.
12742 Floating point constants appear as a sequence of digits, followed by a
12743 decimal point and another sequence of digits. An optional exponent can
12744 then be specified, in the form @samp{E@r{[}+@r{|}-@r{]}@var{nnn}}, where
12745 @samp{@r{[}+@r{|}-@r{]}@var{nnn}} is the desired exponent. All of the
12746 digits of the floating point constant must be valid decimal (base 10)
12750 Character constants consist of a single character enclosed by a pair of
12751 like quotes, either single (@code{'}) or double (@code{"}). They may
12752 also be expressed by their ordinal value (their @sc{ascii} value, usually)
12753 followed by a @samp{C}.
12756 String constants consist of a sequence of characters enclosed by a
12757 pair of like quotes, either single (@code{'}) or double (@code{"}).
12758 Escape sequences in the style of C are also allowed. @xref{C
12759 Constants, ,C and C@t{++} Constants}, for a brief explanation of escape
12763 Enumerated constants consist of an enumerated identifier.
12766 Boolean constants consist of the identifiers @code{TRUE} and
12770 Pointer constants consist of integral values only.
12773 Set constants are not yet supported.
12777 @subsubsection Modula-2 Types
12778 @cindex Modula-2 types
12780 Currently @value{GDBN} can print the following data types in Modula-2
12781 syntax: array types, record types, set types, pointer types, procedure
12782 types, enumerated types, subrange types and base types. You can also
12783 print the contents of variables declared using these type.
12784 This section gives a number of simple source code examples together with
12785 sample @value{GDBN} sessions.
12787 The first example contains the following section of code:
12796 and you can request @value{GDBN} to interrogate the type and value of
12797 @code{r} and @code{s}.
12800 (@value{GDBP}) print s
12802 (@value{GDBP}) ptype s
12804 (@value{GDBP}) print r
12806 (@value{GDBP}) ptype r
12811 Likewise if your source code declares @code{s} as:
12815 s: SET ['A'..'Z'] ;
12819 then you may query the type of @code{s} by:
12822 (@value{GDBP}) ptype s
12823 type = SET ['A'..'Z']
12827 Note that at present you cannot interactively manipulate set
12828 expressions using the debugger.
12830 The following example shows how you might declare an array in Modula-2
12831 and how you can interact with @value{GDBN} to print its type and contents:
12835 s: ARRAY [-10..10] OF CHAR ;
12839 (@value{GDBP}) ptype s
12840 ARRAY [-10..10] OF CHAR
12843 Note that the array handling is not yet complete and although the type
12844 is printed correctly, expression handling still assumes that all
12845 arrays have a lower bound of zero and not @code{-10} as in the example
12848 Here are some more type related Modula-2 examples:
12852 colour = (blue, red, yellow, green) ;
12853 t = [blue..yellow] ;
12861 The @value{GDBN} interaction shows how you can query the data type
12862 and value of a variable.
12865 (@value{GDBP}) print s
12867 (@value{GDBP}) ptype t
12868 type = [blue..yellow]
12872 In this example a Modula-2 array is declared and its contents
12873 displayed. Observe that the contents are written in the same way as
12874 their @code{C} counterparts.
12878 s: ARRAY [1..5] OF CARDINAL ;
12884 (@value{GDBP}) print s
12885 $1 = @{1, 0, 0, 0, 0@}
12886 (@value{GDBP}) ptype s
12887 type = ARRAY [1..5] OF CARDINAL
12890 The Modula-2 language interface to @value{GDBN} also understands
12891 pointer types as shown in this example:
12895 s: POINTER TO ARRAY [1..5] OF CARDINAL ;
12902 and you can request that @value{GDBN} describes the type of @code{s}.
12905 (@value{GDBP}) ptype s
12906 type = POINTER TO ARRAY [1..5] OF CARDINAL
12909 @value{GDBN} handles compound types as we can see in this example.
12910 Here we combine array types, record types, pointer types and subrange
12921 myarray = ARRAY myrange OF CARDINAL ;
12922 myrange = [-2..2] ;
12924 s: POINTER TO ARRAY myrange OF foo ;
12928 and you can ask @value{GDBN} to describe the type of @code{s} as shown
12932 (@value{GDBP}) ptype s
12933 type = POINTER TO ARRAY [-2..2] OF foo = RECORD
12936 f3 : ARRAY [-2..2] OF CARDINAL;
12941 @subsubsection Modula-2 Defaults
12942 @cindex Modula-2 defaults
12944 If type and range checking are set automatically by @value{GDBN}, they
12945 both default to @code{on} whenever the working language changes to
12946 Modula-2. This happens regardless of whether you or @value{GDBN}
12947 selected the working language.
12949 If you allow @value{GDBN} to set the language automatically, then entering
12950 code compiled from a file whose name ends with @file{.mod} sets the
12951 working language to Modula-2. @xref{Automatically, ,Having @value{GDBN}
12952 Infer the Source Language}, for further details.
12955 @subsubsection Deviations from Standard Modula-2
12956 @cindex Modula-2, deviations from
12958 A few changes have been made to make Modula-2 programs easier to debug.
12959 This is done primarily via loosening its type strictness:
12963 Unlike in standard Modula-2, pointer constants can be formed by
12964 integers. This allows you to modify pointer variables during
12965 debugging. (In standard Modula-2, the actual address contained in a
12966 pointer variable is hidden from you; it can only be modified
12967 through direct assignment to another pointer variable or expression that
12968 returned a pointer.)
12971 C escape sequences can be used in strings and characters to represent
12972 non-printable characters. @value{GDBN} prints out strings with these
12973 escape sequences embedded. Single non-printable characters are
12974 printed using the @samp{CHR(@var{nnn})} format.
12977 The assignment operator (@code{:=}) returns the value of its right-hand
12981 All built-in procedures both modify @emph{and} return their argument.
12985 @subsubsection Modula-2 Type and Range Checks
12986 @cindex Modula-2 checks
12989 @emph{Warning:} in this release, @value{GDBN} does not yet perform type or
12992 @c FIXME remove warning when type/range checks added
12994 @value{GDBN} considers two Modula-2 variables type equivalent if:
12998 They are of types that have been declared equivalent via a @code{TYPE
12999 @var{t1} = @var{t2}} statement
13002 They have been declared on the same line. (Note: This is true of the
13003 @sc{gnu} Modula-2 compiler, but it may not be true of other compilers.)
13006 As long as type checking is enabled, any attempt to combine variables
13007 whose types are not equivalent is an error.
13009 Range checking is done on all mathematical operations, assignment, array
13010 index bounds, and all built-in functions and procedures.
13013 @subsubsection The Scope Operators @code{::} and @code{.}
13015 @cindex @code{.}, Modula-2 scope operator
13016 @cindex colon, doubled as scope operator
13018 @vindex colon-colon@r{, in Modula-2}
13019 @c Info cannot handle :: but TeX can.
13022 @vindex ::@r{, in Modula-2}
13025 There are a few subtle differences between the Modula-2 scope operator
13026 (@code{.}) and the @value{GDBN} scope operator (@code{::}). The two have
13031 @var{module} . @var{id}
13032 @var{scope} :: @var{id}
13036 where @var{scope} is the name of a module or a procedure,
13037 @var{module} the name of a module, and @var{id} is any declared
13038 identifier within your program, except another module.
13040 Using the @code{::} operator makes @value{GDBN} search the scope
13041 specified by @var{scope} for the identifier @var{id}. If it is not
13042 found in the specified scope, then @value{GDBN} searches all scopes
13043 enclosing the one specified by @var{scope}.
13045 Using the @code{.} operator makes @value{GDBN} search the current scope for
13046 the identifier specified by @var{id} that was imported from the
13047 definition module specified by @var{module}. With this operator, it is
13048 an error if the identifier @var{id} was not imported from definition
13049 module @var{module}, or if @var{id} is not an identifier in
13053 @subsubsection @value{GDBN} and Modula-2
13055 Some @value{GDBN} commands have little use when debugging Modula-2 programs.
13056 Five subcommands of @code{set print} and @code{show print} apply
13057 specifically to C and C@t{++}: @samp{vtbl}, @samp{demangle},
13058 @samp{asm-demangle}, @samp{object}, and @samp{union}. The first four
13059 apply to C@t{++}, and the last to the C @code{union} type, which has no direct
13060 analogue in Modula-2.
13062 The @code{@@} operator (@pxref{Expressions, ,Expressions}), while available
13063 with any language, is not useful with Modula-2. Its
13064 intent is to aid the debugging of @dfn{dynamic arrays}, which cannot be
13065 created in Modula-2 as they can in C or C@t{++}. However, because an
13066 address can be specified by an integral constant, the construct
13067 @samp{@{@var{type}@}@var{adrexp}} is still useful.
13069 @cindex @code{#} in Modula-2
13070 In @value{GDBN} scripts, the Modula-2 inequality operator @code{#} is
13071 interpreted as the beginning of a comment. Use @code{<>} instead.
13077 The extensions made to @value{GDBN} for Ada only support
13078 output from the @sc{gnu} Ada (GNAT) compiler.
13079 Other Ada compilers are not currently supported, and
13080 attempting to debug executables produced by them is most likely
13084 @cindex expressions in Ada
13086 * Ada Mode Intro:: General remarks on the Ada syntax
13087 and semantics supported by Ada mode
13089 * Omissions from Ada:: Restrictions on the Ada expression syntax.
13090 * Additions to Ada:: Extensions of the Ada expression syntax.
13091 * Stopping Before Main Program:: Debugging the program during elaboration.
13092 * Ada Tasks:: Listing and setting breakpoints in tasks.
13093 * Ada Tasks and Core Files:: Tasking Support when Debugging Core Files
13094 * Ravenscar Profile:: Tasking Support when using the Ravenscar
13096 * Ada Glitches:: Known peculiarities of Ada mode.
13099 @node Ada Mode Intro
13100 @subsubsection Introduction
13101 @cindex Ada mode, general
13103 The Ada mode of @value{GDBN} supports a fairly large subset of Ada expression
13104 syntax, with some extensions.
13105 The philosophy behind the design of this subset is
13109 That @value{GDBN} should provide basic literals and access to operations for
13110 arithmetic, dereferencing, field selection, indexing, and subprogram calls,
13111 leaving more sophisticated computations to subprograms written into the
13112 program (which therefore may be called from @value{GDBN}).
13115 That type safety and strict adherence to Ada language restrictions
13116 are not particularly important to the @value{GDBN} user.
13119 That brevity is important to the @value{GDBN} user.
13122 Thus, for brevity, the debugger acts as if all names declared in
13123 user-written packages are directly visible, even if they are not visible
13124 according to Ada rules, thus making it unnecessary to fully qualify most
13125 names with their packages, regardless of context. Where this causes
13126 ambiguity, @value{GDBN} asks the user's intent.
13128 The debugger will start in Ada mode if it detects an Ada main program.
13129 As for other languages, it will enter Ada mode when stopped in a program that
13130 was translated from an Ada source file.
13132 While in Ada mode, you may use `@t{--}' for comments. This is useful
13133 mostly for documenting command files. The standard @value{GDBN} comment
13134 (@samp{#}) still works at the beginning of a line in Ada mode, but not in the
13135 middle (to allow based literals).
13137 The debugger supports limited overloading. Given a subprogram call in which
13138 the function symbol has multiple definitions, it will use the number of
13139 actual parameters and some information about their types to attempt to narrow
13140 the set of definitions. It also makes very limited use of context, preferring
13141 procedures to functions in the context of the @code{call} command, and
13142 functions to procedures elsewhere.
13144 @node Omissions from Ada
13145 @subsubsection Omissions from Ada
13146 @cindex Ada, omissions from
13148 Here are the notable omissions from the subset:
13152 Only a subset of the attributes are supported:
13156 @t{'First}, @t{'Last}, and @t{'Length}
13157 on array objects (not on types and subtypes).
13160 @t{'Min} and @t{'Max}.
13163 @t{'Pos} and @t{'Val}.
13169 @t{'Range} on array objects (not subtypes), but only as the right
13170 operand of the membership (@code{in}) operator.
13173 @t{'Access}, @t{'Unchecked_Access}, and
13174 @t{'Unrestricted_Access} (a GNAT extension).
13182 @code{Characters.Latin_1} are not available and
13183 concatenation is not implemented. Thus, escape characters in strings are
13184 not currently available.
13187 Equality tests (@samp{=} and @samp{/=}) on arrays test for bitwise
13188 equality of representations. They will generally work correctly
13189 for strings and arrays whose elements have integer or enumeration types.
13190 They may not work correctly for arrays whose element
13191 types have user-defined equality, for arrays of real values
13192 (in particular, IEEE-conformant floating point, because of negative
13193 zeroes and NaNs), and for arrays whose elements contain unused bits with
13194 indeterminate values.
13197 The other component-by-component array operations (@code{and}, @code{or},
13198 @code{xor}, @code{not}, and relational tests other than equality)
13199 are not implemented.
13202 @cindex array aggregates (Ada)
13203 @cindex record aggregates (Ada)
13204 @cindex aggregates (Ada)
13205 There is limited support for array and record aggregates. They are
13206 permitted only on the right sides of assignments, as in these examples:
13209 (@value{GDBP}) set An_Array := (1, 2, 3, 4, 5, 6)
13210 (@value{GDBP}) set An_Array := (1, others => 0)
13211 (@value{GDBP}) set An_Array := (0|4 => 1, 1..3 => 2, 5 => 6)
13212 (@value{GDBP}) set A_2D_Array := ((1, 2, 3), (4, 5, 6), (7, 8, 9))
13213 (@value{GDBP}) set A_Record := (1, "Peter", True);
13214 (@value{GDBP}) set A_Record := (Name => "Peter", Id => 1, Alive => True)
13218 discriminant's value by assigning an aggregate has an
13219 undefined effect if that discriminant is used within the record.
13220 However, you can first modify discriminants by directly assigning to
13221 them (which normally would not be allowed in Ada), and then performing an
13222 aggregate assignment. For example, given a variable @code{A_Rec}
13223 declared to have a type such as:
13226 type Rec (Len : Small_Integer := 0) is record
13228 Vals : IntArray (1 .. Len);
13232 you can assign a value with a different size of @code{Vals} with two
13236 (@value{GDBP}) set A_Rec.Len := 4
13237 (@value{GDBP}) set A_Rec := (Id => 42, Vals => (1, 2, 3, 4))
13240 As this example also illustrates, @value{GDBN} is very loose about the usual
13241 rules concerning aggregates. You may leave out some of the
13242 components of an array or record aggregate (such as the @code{Len}
13243 component in the assignment to @code{A_Rec} above); they will retain their
13244 original values upon assignment. You may freely use dynamic values as
13245 indices in component associations. You may even use overlapping or
13246 redundant component associations, although which component values are
13247 assigned in such cases is not defined.
13250 Calls to dispatching subprograms are not implemented.
13253 The overloading algorithm is much more limited (i.e., less selective)
13254 than that of real Ada. It makes only limited use of the context in
13255 which a subexpression appears to resolve its meaning, and it is much
13256 looser in its rules for allowing type matches. As a result, some
13257 function calls will be ambiguous, and the user will be asked to choose
13258 the proper resolution.
13261 The @code{new} operator is not implemented.
13264 Entry calls are not implemented.
13267 Aside from printing, arithmetic operations on the native VAX floating-point
13268 formats are not supported.
13271 It is not possible to slice a packed array.
13274 The names @code{True} and @code{False}, when not part of a qualified name,
13275 are interpreted as if implicitly prefixed by @code{Standard}, regardless of
13277 Should your program
13278 redefine these names in a package or procedure (at best a dubious practice),
13279 you will have to use fully qualified names to access their new definitions.
13282 @node Additions to Ada
13283 @subsubsection Additions to Ada
13284 @cindex Ada, deviations from
13286 As it does for other languages, @value{GDBN} makes certain generic
13287 extensions to Ada (@pxref{Expressions}):
13291 If the expression @var{E} is a variable residing in memory (typically
13292 a local variable or array element) and @var{N} is a positive integer,
13293 then @code{@var{E}@@@var{N}} displays the values of @var{E} and the
13294 @var{N}-1 adjacent variables following it in memory as an array. In
13295 Ada, this operator is generally not necessary, since its prime use is
13296 in displaying parts of an array, and slicing will usually do this in
13297 Ada. However, there are occasional uses when debugging programs in
13298 which certain debugging information has been optimized away.
13301 @code{@var{B}::@var{var}} means ``the variable named @var{var} that
13302 appears in function or file @var{B}.'' When @var{B} is a file name,
13303 you must typically surround it in single quotes.
13306 The expression @code{@{@var{type}@} @var{addr}} means ``the variable of type
13307 @var{type} that appears at address @var{addr}.''
13310 A name starting with @samp{$} is a convenience variable
13311 (@pxref{Convenience Vars}) or a machine register (@pxref{Registers}).
13314 In addition, @value{GDBN} provides a few other shortcuts and outright
13315 additions specific to Ada:
13319 The assignment statement is allowed as an expression, returning
13320 its right-hand operand as its value. Thus, you may enter
13323 (@value{GDBP}) set x := y + 3
13324 (@value{GDBP}) print A(tmp := y + 1)
13328 The semicolon is allowed as an ``operator,'' returning as its value
13329 the value of its right-hand operand.
13330 This allows, for example,
13331 complex conditional breaks:
13334 (@value{GDBP}) break f
13335 (@value{GDBP}) condition 1 (report(i); k += 1; A(k) > 100)
13339 Rather than use catenation and symbolic character names to introduce special
13340 characters into strings, one may instead use a special bracket notation,
13341 which is also used to print strings. A sequence of characters of the form
13342 @samp{["@var{XX}"]} within a string or character literal denotes the
13343 (single) character whose numeric encoding is @var{XX} in hexadecimal. The
13344 sequence of characters @samp{["""]} also denotes a single quotation mark
13345 in strings. For example,
13347 "One line.["0a"]Next line.["0a"]"
13350 contains an ASCII newline character (@code{Ada.Characters.Latin_1.LF})
13354 The subtype used as a prefix for the attributes @t{'Pos}, @t{'Min}, and
13355 @t{'Max} is optional (and is ignored in any case). For example, it is valid
13359 (@value{GDBP}) print 'max(x, y)
13363 When printing arrays, @value{GDBN} uses positional notation when the
13364 array has a lower bound of 1, and uses a modified named notation otherwise.
13365 For example, a one-dimensional array of three integers with a lower bound
13366 of 3 might print as
13373 That is, in contrast to valid Ada, only the first component has a @code{=>}
13377 You may abbreviate attributes in expressions with any unique,
13378 multi-character subsequence of
13379 their names (an exact match gets preference).
13380 For example, you may use @t{a'len}, @t{a'gth}, or @t{a'lh}
13381 in place of @t{a'length}.
13384 @cindex quoting Ada internal identifiers
13385 Since Ada is case-insensitive, the debugger normally maps identifiers you type
13386 to lower case. The GNAT compiler uses upper-case characters for
13387 some of its internal identifiers, which are normally of no interest to users.
13388 For the rare occasions when you actually have to look at them,
13389 enclose them in angle brackets to avoid the lower-case mapping.
13392 (@value{GDBP}) print <JMPBUF_SAVE>[0]
13396 Printing an object of class-wide type or dereferencing an
13397 access-to-class-wide value will display all the components of the object's
13398 specific type (as indicated by its run-time tag). Likewise, component
13399 selection on such a value will operate on the specific type of the
13404 @node Stopping Before Main Program
13405 @subsubsection Stopping at the Very Beginning
13407 @cindex breakpointing Ada elaboration code
13408 It is sometimes necessary to debug the program during elaboration, and
13409 before reaching the main procedure.
13410 As defined in the Ada Reference
13411 Manual, the elaboration code is invoked from a procedure called
13412 @code{adainit}. To run your program up to the beginning of
13413 elaboration, simply use the following two commands:
13414 @code{tbreak adainit} and @code{run}.
13417 @subsubsection Extensions for Ada Tasks
13418 @cindex Ada, tasking
13420 Support for Ada tasks is analogous to that for threads (@pxref{Threads}).
13421 @value{GDBN} provides the following task-related commands:
13426 This command shows a list of current Ada tasks, as in the following example:
13433 (@value{GDBP}) info tasks
13434 ID TID P-ID Pri State Name
13435 1 8088000 0 15 Child Activation Wait main_task
13436 2 80a4000 1 15 Accept Statement b
13437 3 809a800 1 15 Child Activation Wait a
13438 * 4 80ae800 3 15 Runnable c
13443 In this listing, the asterisk before the last task indicates it to be the
13444 task currently being inspected.
13448 Represents @value{GDBN}'s internal task number.
13454 The parent's task ID (@value{GDBN}'s internal task number).
13457 The base priority of the task.
13460 Current state of the task.
13464 The task has been created but has not been activated. It cannot be
13468 The task is not blocked for any reason known to Ada. (It may be waiting
13469 for a mutex, though.) It is conceptually "executing" in normal mode.
13472 The task is terminated, in the sense of ARM 9.3 (5). Any dependents
13473 that were waiting on terminate alternatives have been awakened and have
13474 terminated themselves.
13476 @item Child Activation Wait
13477 The task is waiting for created tasks to complete activation.
13479 @item Accept Statement
13480 The task is waiting on an accept or selective wait statement.
13482 @item Waiting on entry call
13483 The task is waiting on an entry call.
13485 @item Async Select Wait
13486 The task is waiting to start the abortable part of an asynchronous
13490 The task is waiting on a select statement with only a delay
13493 @item Child Termination Wait
13494 The task is sleeping having completed a master within itself, and is
13495 waiting for the tasks dependent on that master to become terminated or
13496 waiting on a terminate Phase.
13498 @item Wait Child in Term Alt
13499 The task is sleeping waiting for tasks on terminate alternatives to
13500 finish terminating.
13502 @item Accepting RV with @var{taskno}
13503 The task is accepting a rendez-vous with the task @var{taskno}.
13507 Name of the task in the program.
13511 @kindex info task @var{taskno}
13512 @item info task @var{taskno}
13513 This command shows detailled informations on the specified task, as in
13514 the following example:
13519 (@value{GDBP}) info tasks
13520 ID TID P-ID Pri State Name
13521 1 8077880 0 15 Child Activation Wait main_task
13522 * 2 807c468 1 15 Runnable task_1
13523 (@value{GDBP}) info task 2
13524 Ada Task: 0x807c468
13527 Parent: 1 (main_task)
13533 @kindex task@r{ (Ada)}
13534 @cindex current Ada task ID
13535 This command prints the ID of the current task.
13541 (@value{GDBP}) info tasks
13542 ID TID P-ID Pri State Name
13543 1 8077870 0 15 Child Activation Wait main_task
13544 * 2 807c458 1 15 Runnable t
13545 (@value{GDBP}) task
13546 [Current task is 2]
13549 @item task @var{taskno}
13550 @cindex Ada task switching
13551 This command is like the @code{thread @var{threadno}}
13552 command (@pxref{Threads}). It switches the context of debugging
13553 from the current task to the given task.
13559 (@value{GDBP}) info tasks
13560 ID TID P-ID Pri State Name
13561 1 8077870 0 15 Child Activation Wait main_task
13562 * 2 807c458 1 15 Runnable t
13563 (@value{GDBP}) task 1
13564 [Switching to task 1]
13565 #0 0x8067726 in pthread_cond_wait ()
13567 #0 0x8067726 in pthread_cond_wait ()
13568 #1 0x8056714 in system.os_interface.pthread_cond_wait ()
13569 #2 0x805cb63 in system.task_primitives.operations.sleep ()
13570 #3 0x806153e in system.tasking.stages.activate_tasks ()
13571 #4 0x804aacc in un () at un.adb:5
13574 @item break @var{linespec} task @var{taskno}
13575 @itemx break @var{linespec} task @var{taskno} if @dots{}
13576 @cindex breakpoints and tasks, in Ada
13577 @cindex task breakpoints, in Ada
13578 @kindex break @dots{} task @var{taskno}@r{ (Ada)}
13579 These commands are like the @code{break @dots{} thread @dots{}}
13580 command (@pxref{Thread Stops}).
13581 @var{linespec} specifies source lines, as described
13582 in @ref{Specify Location}.
13584 Use the qualifier @samp{task @var{taskno}} with a breakpoint command
13585 to specify that you only want @value{GDBN} to stop the program when a
13586 particular Ada task reaches this breakpoint. @var{taskno} is one of the
13587 numeric task identifiers assigned by @value{GDBN}, shown in the first
13588 column of the @samp{info tasks} display.
13590 If you do not specify @samp{task @var{taskno}} when you set a
13591 breakpoint, the breakpoint applies to @emph{all} tasks of your
13594 You can use the @code{task} qualifier on conditional breakpoints as
13595 well; in this case, place @samp{task @var{taskno}} before the
13596 breakpoint condition (before the @code{if}).
13604 (@value{GDBP}) info tasks
13605 ID TID P-ID Pri State Name
13606 1 140022020 0 15 Child Activation Wait main_task
13607 2 140045060 1 15 Accept/Select Wait t2
13608 3 140044840 1 15 Runnable t1
13609 * 4 140056040 1 15 Runnable t3
13610 (@value{GDBP}) b 15 task 2
13611 Breakpoint 5 at 0x120044cb0: file test_task_debug.adb, line 15.
13612 (@value{GDBP}) cont
13617 Breakpoint 5, test_task_debug () at test_task_debug.adb:15
13619 (@value{GDBP}) info tasks
13620 ID TID P-ID Pri State Name
13621 1 140022020 0 15 Child Activation Wait main_task
13622 * 2 140045060 1 15 Runnable t2
13623 3 140044840 1 15 Runnable t1
13624 4 140056040 1 15 Delay Sleep t3
13628 @node Ada Tasks and Core Files
13629 @subsubsection Tasking Support when Debugging Core Files
13630 @cindex Ada tasking and core file debugging
13632 When inspecting a core file, as opposed to debugging a live program,
13633 tasking support may be limited or even unavailable, depending on
13634 the platform being used.
13635 For instance, on x86-linux, the list of tasks is available, but task
13636 switching is not supported. On Tru64, however, task switching will work
13639 On certain platforms, including Tru64, the debugger needs to perform some
13640 memory writes in order to provide Ada tasking support. When inspecting
13641 a core file, this means that the core file must be opened with read-write
13642 privileges, using the command @samp{"set write on"} (@pxref{Patching}).
13643 Under these circumstances, you should make a backup copy of the core
13644 file before inspecting it with @value{GDBN}.
13646 @node Ravenscar Profile
13647 @subsubsection Tasking Support when using the Ravenscar Profile
13648 @cindex Ravenscar Profile
13650 The @dfn{Ravenscar Profile} is a subset of the Ada tasking features,
13651 specifically designed for systems with safety-critical real-time
13655 @kindex set ravenscar task-switching on
13656 @cindex task switching with program using Ravenscar Profile
13657 @item set ravenscar task-switching on
13658 Allows task switching when debugging a program that uses the Ravenscar
13659 Profile. This is the default.
13661 @kindex set ravenscar task-switching off
13662 @item set ravenscar task-switching off
13663 Turn off task switching when debugging a program that uses the Ravenscar
13664 Profile. This is mostly intended to disable the code that adds support
13665 for the Ravenscar Profile, in case a bug in either @value{GDBN} or in
13666 the Ravenscar runtime is preventing @value{GDBN} from working properly.
13667 To be effective, this command should be run before the program is started.
13669 @kindex show ravenscar task-switching
13670 @item show ravenscar task-switching
13671 Show whether it is possible to switch from task to task in a program
13672 using the Ravenscar Profile.
13677 @subsubsection Known Peculiarities of Ada Mode
13678 @cindex Ada, problems
13680 Besides the omissions listed previously (@pxref{Omissions from Ada}),
13681 we know of several problems with and limitations of Ada mode in
13683 some of which will be fixed with planned future releases of the debugger
13684 and the GNU Ada compiler.
13688 Static constants that the compiler chooses not to materialize as objects in
13689 storage are invisible to the debugger.
13692 Named parameter associations in function argument lists are ignored (the
13693 argument lists are treated as positional).
13696 Many useful library packages are currently invisible to the debugger.
13699 Fixed-point arithmetic, conversions, input, and output is carried out using
13700 floating-point arithmetic, and may give results that only approximate those on
13704 The GNAT compiler never generates the prefix @code{Standard} for any of
13705 the standard symbols defined by the Ada language. @value{GDBN} knows about
13706 this: it will strip the prefix from names when you use it, and will never
13707 look for a name you have so qualified among local symbols, nor match against
13708 symbols in other packages or subprograms. If you have
13709 defined entities anywhere in your program other than parameters and
13710 local variables whose simple names match names in @code{Standard},
13711 GNAT's lack of qualification here can cause confusion. When this happens,
13712 you can usually resolve the confusion
13713 by qualifying the problematic names with package
13714 @code{Standard} explicitly.
13717 Older versions of the compiler sometimes generate erroneous debugging
13718 information, resulting in the debugger incorrectly printing the value
13719 of affected entities. In some cases, the debugger is able to work
13720 around an issue automatically. In other cases, the debugger is able
13721 to work around the issue, but the work-around has to be specifically
13724 @kindex set ada trust-PAD-over-XVS
13725 @kindex show ada trust-PAD-over-XVS
13728 @item set ada trust-PAD-over-XVS on
13729 Configure GDB to strictly follow the GNAT encoding when computing the
13730 value of Ada entities, particularly when @code{PAD} and @code{PAD___XVS}
13731 types are involved (see @code{ada/exp_dbug.ads} in the GCC sources for
13732 a complete description of the encoding used by the GNAT compiler).
13733 This is the default.
13735 @item set ada trust-PAD-over-XVS off
13736 This is related to the encoding using by the GNAT compiler. If @value{GDBN}
13737 sometimes prints the wrong value for certain entities, changing @code{ada
13738 trust-PAD-over-XVS} to @code{off} activates a work-around which may fix
13739 the issue. It is always safe to set @code{ada trust-PAD-over-XVS} to
13740 @code{off}, but this incurs a slight performance penalty, so it is
13741 recommended to leave this setting to @code{on} unless necessary.
13745 @node Unsupported Languages
13746 @section Unsupported Languages
13748 @cindex unsupported languages
13749 @cindex minimal language
13750 In addition to the other fully-supported programming languages,
13751 @value{GDBN} also provides a pseudo-language, called @code{minimal}.
13752 It does not represent a real programming language, but provides a set
13753 of capabilities close to what the C or assembly languages provide.
13754 This should allow most simple operations to be performed while debugging
13755 an application that uses a language currently not supported by @value{GDBN}.
13757 If the language is set to @code{auto}, @value{GDBN} will automatically
13758 select this language if the current frame corresponds to an unsupported
13762 @chapter Examining the Symbol Table
13764 The commands described in this chapter allow you to inquire about the
13765 symbols (names of variables, functions and types) defined in your
13766 program. This information is inherent in the text of your program and
13767 does not change as your program executes. @value{GDBN} finds it in your
13768 program's symbol table, in the file indicated when you started @value{GDBN}
13769 (@pxref{File Options, ,Choosing Files}), or by one of the
13770 file-management commands (@pxref{Files, ,Commands to Specify Files}).
13772 @cindex symbol names
13773 @cindex names of symbols
13774 @cindex quoting names
13775 Occasionally, you may need to refer to symbols that contain unusual
13776 characters, which @value{GDBN} ordinarily treats as word delimiters. The
13777 most frequent case is in referring to static variables in other
13778 source files (@pxref{Variables,,Program Variables}). File names
13779 are recorded in object files as debugging symbols, but @value{GDBN} would
13780 ordinarily parse a typical file name, like @file{foo.c}, as the three words
13781 @samp{foo} @samp{.} @samp{c}. To allow @value{GDBN} to recognize
13782 @samp{foo.c} as a single symbol, enclose it in single quotes; for example,
13789 looks up the value of @code{x} in the scope of the file @file{foo.c}.
13792 @cindex case-insensitive symbol names
13793 @cindex case sensitivity in symbol names
13794 @kindex set case-sensitive
13795 @item set case-sensitive on
13796 @itemx set case-sensitive off
13797 @itemx set case-sensitive auto
13798 Normally, when @value{GDBN} looks up symbols, it matches their names
13799 with case sensitivity determined by the current source language.
13800 Occasionally, you may wish to control that. The command @code{set
13801 case-sensitive} lets you do that by specifying @code{on} for
13802 case-sensitive matches or @code{off} for case-insensitive ones. If
13803 you specify @code{auto}, case sensitivity is reset to the default
13804 suitable for the source language. The default is case-sensitive
13805 matches for all languages except for Fortran, for which the default is
13806 case-insensitive matches.
13808 @kindex show case-sensitive
13809 @item show case-sensitive
13810 This command shows the current setting of case sensitivity for symbols
13813 @kindex info address
13814 @cindex address of a symbol
13815 @item info address @var{symbol}
13816 Describe where the data for @var{symbol} is stored. For a register
13817 variable, this says which register it is kept in. For a non-register
13818 local variable, this prints the stack-frame offset at which the variable
13821 Note the contrast with @samp{print &@var{symbol}}, which does not work
13822 at all for a register variable, and for a stack local variable prints
13823 the exact address of the current instantiation of the variable.
13825 @kindex info symbol
13826 @cindex symbol from address
13827 @cindex closest symbol and offset for an address
13828 @item info symbol @var{addr}
13829 Print the name of a symbol which is stored at the address @var{addr}.
13830 If no symbol is stored exactly at @var{addr}, @value{GDBN} prints the
13831 nearest symbol and an offset from it:
13834 (@value{GDBP}) info symbol 0x54320
13835 _initialize_vx + 396 in section .text
13839 This is the opposite of the @code{info address} command. You can use
13840 it to find out the name of a variable or a function given its address.
13842 For dynamically linked executables, the name of executable or shared
13843 library containing the symbol is also printed:
13846 (@value{GDBP}) info symbol 0x400225
13847 _start + 5 in section .text of /tmp/a.out
13848 (@value{GDBP}) info symbol 0x2aaaac2811cf
13849 __read_nocancel + 6 in section .text of /usr/lib64/libc.so.6
13853 @item whatis [@var{arg}]
13854 Print the data type of @var{arg}, which can be either an expression or
13855 a data type. With no argument, print the data type of @code{$}, the
13856 last value in the value history. If @var{arg} is an expression, it is
13857 not actually evaluated, and any side-effecting operations (such as
13858 assignments or function calls) inside it do not take place. If
13859 @var{arg} is a type name, it may be the name of a type or typedef, or
13860 for C code it may have the form @samp{class @var{class-name}},
13861 @samp{struct @var{struct-tag}}, @samp{union @var{union-tag}} or
13862 @samp{enum @var{enum-tag}}.
13863 @xref{Expressions, ,Expressions}.
13866 @item ptype [@var{arg}]
13867 @code{ptype} accepts the same arguments as @code{whatis}, but prints a
13868 detailed description of the type, instead of just the name of the type.
13869 @xref{Expressions, ,Expressions}.
13871 For example, for this variable declaration:
13874 struct complex @{double real; double imag;@} v;
13878 the two commands give this output:
13882 (@value{GDBP}) whatis v
13883 type = struct complex
13884 (@value{GDBP}) ptype v
13885 type = struct complex @{
13893 As with @code{whatis}, using @code{ptype} without an argument refers to
13894 the type of @code{$}, the last value in the value history.
13896 @cindex incomplete type
13897 Sometimes, programs use opaque data types or incomplete specifications
13898 of complex data structure. If the debug information included in the
13899 program does not allow @value{GDBN} to display a full declaration of
13900 the data type, it will say @samp{<incomplete type>}. For example,
13901 given these declarations:
13905 struct foo *fooptr;
13909 but no definition for @code{struct foo} itself, @value{GDBN} will say:
13912 (@value{GDBP}) ptype foo
13913 $1 = <incomplete type>
13917 ``Incomplete type'' is C terminology for data types that are not
13918 completely specified.
13921 @item info types @var{regexp}
13923 Print a brief description of all types whose names match the regular
13924 expression @var{regexp} (or all types in your program, if you supply
13925 no argument). Each complete typename is matched as though it were a
13926 complete line; thus, @samp{i type value} gives information on all
13927 types in your program whose names include the string @code{value}, but
13928 @samp{i type ^value$} gives information only on types whose complete
13929 name is @code{value}.
13931 This command differs from @code{ptype} in two ways: first, like
13932 @code{whatis}, it does not print a detailed description; second, it
13933 lists all source files where a type is defined.
13936 @cindex local variables
13937 @item info scope @var{location}
13938 List all the variables local to a particular scope. This command
13939 accepts a @var{location} argument---a function name, a source line, or
13940 an address preceded by a @samp{*}, and prints all the variables local
13941 to the scope defined by that location. (@xref{Specify Location}, for
13942 details about supported forms of @var{location}.) For example:
13945 (@value{GDBP}) @b{info scope command_line_handler}
13946 Scope for command_line_handler:
13947 Symbol rl is an argument at stack/frame offset 8, length 4.
13948 Symbol linebuffer is in static storage at address 0x150a18, length 4.
13949 Symbol linelength is in static storage at address 0x150a1c, length 4.
13950 Symbol p is a local variable in register $esi, length 4.
13951 Symbol p1 is a local variable in register $ebx, length 4.
13952 Symbol nline is a local variable in register $edx, length 4.
13953 Symbol repeat is a local variable at frame offset -8, length 4.
13957 This command is especially useful for determining what data to collect
13958 during a @dfn{trace experiment}, see @ref{Tracepoint Actions,
13961 @kindex info source
13963 Show information about the current source file---that is, the source file for
13964 the function containing the current point of execution:
13967 the name of the source file, and the directory containing it,
13969 the directory it was compiled in,
13971 its length, in lines,
13973 which programming language it is written in,
13975 whether the executable includes debugging information for that file, and
13976 if so, what format the information is in (e.g., STABS, Dwarf 2, etc.), and
13978 whether the debugging information includes information about
13979 preprocessor macros.
13983 @kindex info sources
13985 Print the names of all source files in your program for which there is
13986 debugging information, organized into two lists: files whose symbols
13987 have already been read, and files whose symbols will be read when needed.
13989 @kindex info functions
13990 @item info functions
13991 Print the names and data types of all defined functions.
13993 @item info functions @var{regexp}
13994 Print the names and data types of all defined functions
13995 whose names contain a match for regular expression @var{regexp}.
13996 Thus, @samp{info fun step} finds all functions whose names
13997 include @code{step}; @samp{info fun ^step} finds those whose names
13998 start with @code{step}. If a function name contains characters
13999 that conflict with the regular expression language (e.g.@:
14000 @samp{operator*()}), they may be quoted with a backslash.
14002 @kindex info variables
14003 @item info variables
14004 Print the names and data types of all variables that are defined
14005 outside of functions (i.e.@: excluding local variables).
14007 @item info variables @var{regexp}
14008 Print the names and data types of all variables (except for local
14009 variables) whose names contain a match for regular expression
14012 @kindex info classes
14013 @cindex Objective-C, classes and selectors
14015 @itemx info classes @var{regexp}
14016 Display all Objective-C classes in your program, or
14017 (with the @var{regexp} argument) all those matching a particular regular
14020 @kindex info selectors
14021 @item info selectors
14022 @itemx info selectors @var{regexp}
14023 Display all Objective-C selectors in your program, or
14024 (with the @var{regexp} argument) all those matching a particular regular
14028 This was never implemented.
14029 @kindex info methods
14031 @itemx info methods @var{regexp}
14032 The @code{info methods} command permits the user to examine all defined
14033 methods within C@t{++} program, or (with the @var{regexp} argument) a
14034 specific set of methods found in the various C@t{++} classes. Many
14035 C@t{++} classes provide a large number of methods. Thus, the output
14036 from the @code{ptype} command can be overwhelming and hard to use. The
14037 @code{info-methods} command filters the methods, printing only those
14038 which match the regular-expression @var{regexp}.
14041 @cindex reloading symbols
14042 Some systems allow individual object files that make up your program to
14043 be replaced without stopping and restarting your program. For example,
14044 in VxWorks you can simply recompile a defective object file and keep on
14045 running. If you are running on one of these systems, you can allow
14046 @value{GDBN} to reload the symbols for automatically relinked modules:
14049 @kindex set symbol-reloading
14050 @item set symbol-reloading on
14051 Replace symbol definitions for the corresponding source file when an
14052 object file with a particular name is seen again.
14054 @item set symbol-reloading off
14055 Do not replace symbol definitions when encountering object files of the
14056 same name more than once. This is the default state; if you are not
14057 running on a system that permits automatic relinking of modules, you
14058 should leave @code{symbol-reloading} off, since otherwise @value{GDBN}
14059 may discard symbols when linking large programs, that may contain
14060 several modules (from different directories or libraries) with the same
14063 @kindex show symbol-reloading
14064 @item show symbol-reloading
14065 Show the current @code{on} or @code{off} setting.
14068 @cindex opaque data types
14069 @kindex set opaque-type-resolution
14070 @item set opaque-type-resolution on
14071 Tell @value{GDBN} to resolve opaque types. An opaque type is a type
14072 declared as a pointer to a @code{struct}, @code{class}, or
14073 @code{union}---for example, @code{struct MyType *}---that is used in one
14074 source file although the full declaration of @code{struct MyType} is in
14075 another source file. The default is on.
14077 A change in the setting of this subcommand will not take effect until
14078 the next time symbols for a file are loaded.
14080 @item set opaque-type-resolution off
14081 Tell @value{GDBN} not to resolve opaque types. In this case, the type
14082 is printed as follows:
14084 @{<no data fields>@}
14087 @kindex show opaque-type-resolution
14088 @item show opaque-type-resolution
14089 Show whether opaque types are resolved or not.
14091 @kindex maint print symbols
14092 @cindex symbol dump
14093 @kindex maint print psymbols
14094 @cindex partial symbol dump
14095 @item maint print symbols @var{filename}
14096 @itemx maint print psymbols @var{filename}
14097 @itemx maint print msymbols @var{filename}
14098 Write a dump of debugging symbol data into the file @var{filename}.
14099 These commands are used to debug the @value{GDBN} symbol-reading code. Only
14100 symbols with debugging data are included. If you use @samp{maint print
14101 symbols}, @value{GDBN} includes all the symbols for which it has already
14102 collected full details: that is, @var{filename} reflects symbols for
14103 only those files whose symbols @value{GDBN} has read. You can use the
14104 command @code{info sources} to find out which files these are. If you
14105 use @samp{maint print psymbols} instead, the dump shows information about
14106 symbols that @value{GDBN} only knows partially---that is, symbols defined in
14107 files that @value{GDBN} has skimmed, but not yet read completely. Finally,
14108 @samp{maint print msymbols} dumps just the minimal symbol information
14109 required for each object file from which @value{GDBN} has read some symbols.
14110 @xref{Files, ,Commands to Specify Files}, for a discussion of how
14111 @value{GDBN} reads symbols (in the description of @code{symbol-file}).
14113 @kindex maint info symtabs
14114 @kindex maint info psymtabs
14115 @cindex listing @value{GDBN}'s internal symbol tables
14116 @cindex symbol tables, listing @value{GDBN}'s internal
14117 @cindex full symbol tables, listing @value{GDBN}'s internal
14118 @cindex partial symbol tables, listing @value{GDBN}'s internal
14119 @item maint info symtabs @r{[} @var{regexp} @r{]}
14120 @itemx maint info psymtabs @r{[} @var{regexp} @r{]}
14122 List the @code{struct symtab} or @code{struct partial_symtab}
14123 structures whose names match @var{regexp}. If @var{regexp} is not
14124 given, list them all. The output includes expressions which you can
14125 copy into a @value{GDBN} debugging this one to examine a particular
14126 structure in more detail. For example:
14129 (@value{GDBP}) maint info psymtabs dwarf2read
14130 @{ objfile /home/gnu/build/gdb/gdb
14131 ((struct objfile *) 0x82e69d0)
14132 @{ psymtab /home/gnu/src/gdb/dwarf2read.c
14133 ((struct partial_symtab *) 0x8474b10)
14136 text addresses 0x814d3c8 -- 0x8158074
14137 globals (* (struct partial_symbol **) 0x8507a08 @@ 9)
14138 statics (* (struct partial_symbol **) 0x40e95b78 @@ 2882)
14139 dependencies (none)
14142 (@value{GDBP}) maint info symtabs
14146 We see that there is one partial symbol table whose filename contains
14147 the string @samp{dwarf2read}, belonging to the @samp{gdb} executable;
14148 and we see that @value{GDBN} has not read in any symtabs yet at all.
14149 If we set a breakpoint on a function, that will cause @value{GDBN} to
14150 read the symtab for the compilation unit containing that function:
14153 (@value{GDBP}) break dwarf2_psymtab_to_symtab
14154 Breakpoint 1 at 0x814e5da: file /home/gnu/src/gdb/dwarf2read.c,
14156 (@value{GDBP}) maint info symtabs
14157 @{ objfile /home/gnu/build/gdb/gdb
14158 ((struct objfile *) 0x82e69d0)
14159 @{ symtab /home/gnu/src/gdb/dwarf2read.c
14160 ((struct symtab *) 0x86c1f38)
14163 blockvector ((struct blockvector *) 0x86c1bd0) (primary)
14164 linetable ((struct linetable *) 0x8370fa0)
14165 debugformat DWARF 2
14174 @chapter Altering Execution
14176 Once you think you have found an error in your program, you might want to
14177 find out for certain whether correcting the apparent error would lead to
14178 correct results in the rest of the run. You can find the answer by
14179 experiment, using the @value{GDBN} features for altering execution of the
14182 For example, you can store new values into variables or memory
14183 locations, give your program a signal, restart it at a different
14184 address, or even return prematurely from a function.
14187 * Assignment:: Assignment to variables
14188 * Jumping:: Continuing at a different address
14189 * Signaling:: Giving your program a signal
14190 * Returning:: Returning from a function
14191 * Calling:: Calling your program's functions
14192 * Patching:: Patching your program
14196 @section Assignment to Variables
14199 @cindex setting variables
14200 To alter the value of a variable, evaluate an assignment expression.
14201 @xref{Expressions, ,Expressions}. For example,
14208 stores the value 4 into the variable @code{x}, and then prints the
14209 value of the assignment expression (which is 4).
14210 @xref{Languages, ,Using @value{GDBN} with Different Languages}, for more
14211 information on operators in supported languages.
14213 @kindex set variable
14214 @cindex variables, setting
14215 If you are not interested in seeing the value of the assignment, use the
14216 @code{set} command instead of the @code{print} command. @code{set} is
14217 really the same as @code{print} except that the expression's value is
14218 not printed and is not put in the value history (@pxref{Value History,
14219 ,Value History}). The expression is evaluated only for its effects.
14221 If the beginning of the argument string of the @code{set} command
14222 appears identical to a @code{set} subcommand, use the @code{set
14223 variable} command instead of just @code{set}. This command is identical
14224 to @code{set} except for its lack of subcommands. For example, if your
14225 program has a variable @code{width}, you get an error if you try to set
14226 a new value with just @samp{set width=13}, because @value{GDBN} has the
14227 command @code{set width}:
14230 (@value{GDBP}) whatis width
14232 (@value{GDBP}) p width
14234 (@value{GDBP}) set width=47
14235 Invalid syntax in expression.
14239 The invalid expression, of course, is @samp{=47}. In
14240 order to actually set the program's variable @code{width}, use
14243 (@value{GDBP}) set var width=47
14246 Because the @code{set} command has many subcommands that can conflict
14247 with the names of program variables, it is a good idea to use the
14248 @code{set variable} command instead of just @code{set}. For example, if
14249 your program has a variable @code{g}, you run into problems if you try
14250 to set a new value with just @samp{set g=4}, because @value{GDBN} has
14251 the command @code{set gnutarget}, abbreviated @code{set g}:
14255 (@value{GDBP}) whatis g
14259 (@value{GDBP}) set g=4
14263 The program being debugged has been started already.
14264 Start it from the beginning? (y or n) y
14265 Starting program: /home/smith/cc_progs/a.out
14266 "/home/smith/cc_progs/a.out": can't open to read symbols:
14267 Invalid bfd target.
14268 (@value{GDBP}) show g
14269 The current BFD target is "=4".
14274 The program variable @code{g} did not change, and you silently set the
14275 @code{gnutarget} to an invalid value. In order to set the variable
14279 (@value{GDBP}) set var g=4
14282 @value{GDBN} allows more implicit conversions in assignments than C; you can
14283 freely store an integer value into a pointer variable or vice versa,
14284 and you can convert any structure to any other structure that is the
14285 same length or shorter.
14286 @comment FIXME: how do structs align/pad in these conversions?
14287 @comment /doc@cygnus.com 18dec1990
14289 To store values into arbitrary places in memory, use the @samp{@{@dots{}@}}
14290 construct to generate a value of specified type at a specified address
14291 (@pxref{Expressions, ,Expressions}). For example, @code{@{int@}0x83040} refers
14292 to memory location @code{0x83040} as an integer (which implies a certain size
14293 and representation in memory), and
14296 set @{int@}0x83040 = 4
14300 stores the value 4 into that memory location.
14303 @section Continuing at a Different Address
14305 Ordinarily, when you continue your program, you do so at the place where
14306 it stopped, with the @code{continue} command. You can instead continue at
14307 an address of your own choosing, with the following commands:
14311 @item jump @var{linespec}
14312 @itemx jump @var{location}
14313 Resume execution at line @var{linespec} or at address given by
14314 @var{location}. Execution stops again immediately if there is a
14315 breakpoint there. @xref{Specify Location}, for a description of the
14316 different forms of @var{linespec} and @var{location}. It is common
14317 practice to use the @code{tbreak} command in conjunction with
14318 @code{jump}. @xref{Set Breaks, ,Setting Breakpoints}.
14320 The @code{jump} command does not change the current stack frame, or
14321 the stack pointer, or the contents of any memory location or any
14322 register other than the program counter. If line @var{linespec} is in
14323 a different function from the one currently executing, the results may
14324 be bizarre if the two functions expect different patterns of arguments or
14325 of local variables. For this reason, the @code{jump} command requests
14326 confirmation if the specified line is not in the function currently
14327 executing. However, even bizarre results are predictable if you are
14328 well acquainted with the machine-language code of your program.
14331 @c Doesn't work on HP-UX; have to set $pcoqh and $pcoqt.
14332 On many systems, you can get much the same effect as the @code{jump}
14333 command by storing a new value into the register @code{$pc}. The
14334 difference is that this does not start your program running; it only
14335 changes the address of where it @emph{will} run when you continue. For
14343 makes the next @code{continue} command or stepping command execute at
14344 address @code{0x485}, rather than at the address where your program stopped.
14345 @xref{Continuing and Stepping, ,Continuing and Stepping}.
14347 The most common occasion to use the @code{jump} command is to back
14348 up---perhaps with more breakpoints set---over a portion of a program
14349 that has already executed, in order to examine its execution in more
14354 @section Giving your Program a Signal
14355 @cindex deliver a signal to a program
14359 @item signal @var{signal}
14360 Resume execution where your program stopped, but immediately give it the
14361 signal @var{signal}. @var{signal} can be the name or the number of a
14362 signal. For example, on many systems @code{signal 2} and @code{signal
14363 SIGINT} are both ways of sending an interrupt signal.
14365 Alternatively, if @var{signal} is zero, continue execution without
14366 giving a signal. This is useful when your program stopped on account of
14367 a signal and would ordinary see the signal when resumed with the
14368 @code{continue} command; @samp{signal 0} causes it to resume without a
14371 @code{signal} does not repeat when you press @key{RET} a second time
14372 after executing the command.
14376 Invoking the @code{signal} command is not the same as invoking the
14377 @code{kill} utility from the shell. Sending a signal with @code{kill}
14378 causes @value{GDBN} to decide what to do with the signal depending on
14379 the signal handling tables (@pxref{Signals}). The @code{signal} command
14380 passes the signal directly to your program.
14384 @section Returning from a Function
14387 @cindex returning from a function
14390 @itemx return @var{expression}
14391 You can cancel execution of a function call with the @code{return}
14392 command. If you give an
14393 @var{expression} argument, its value is used as the function's return
14397 When you use @code{return}, @value{GDBN} discards the selected stack frame
14398 (and all frames within it). You can think of this as making the
14399 discarded frame return prematurely. If you wish to specify a value to
14400 be returned, give that value as the argument to @code{return}.
14402 This pops the selected stack frame (@pxref{Selection, ,Selecting a
14403 Frame}), and any other frames inside of it, leaving its caller as the
14404 innermost remaining frame. That frame becomes selected. The
14405 specified value is stored in the registers used for returning values
14408 The @code{return} command does not resume execution; it leaves the
14409 program stopped in the state that would exist if the function had just
14410 returned. In contrast, the @code{finish} command (@pxref{Continuing
14411 and Stepping, ,Continuing and Stepping}) resumes execution until the
14412 selected stack frame returns naturally.
14414 @value{GDBN} needs to know how the @var{expression} argument should be set for
14415 the inferior. The concrete registers assignment depends on the OS ABI and the
14416 type being returned by the selected stack frame. For example it is common for
14417 OS ABI to return floating point values in FPU registers while integer values in
14418 CPU registers. Still some ABIs return even floating point values in CPU
14419 registers. Larger integer widths (such as @code{long long int}) also have
14420 specific placement rules. @value{GDBN} already knows the OS ABI from its
14421 current target so it needs to find out also the type being returned to make the
14422 assignment into the right register(s).
14424 Normally, the selected stack frame has debug info. @value{GDBN} will always
14425 use the debug info instead of the implicit type of @var{expression} when the
14426 debug info is available. For example, if you type @kbd{return -1}, and the
14427 function in the current stack frame is declared to return a @code{long long
14428 int}, @value{GDBN} transparently converts the implicit @code{int} value of -1
14429 into a @code{long long int}:
14432 Breakpoint 1, func () at gdb.base/return-nodebug.c:29
14434 (@value{GDBP}) return -1
14435 Make func return now? (y or n) y
14436 #0 0x004004f6 in main () at gdb.base/return-nodebug.c:43
14437 43 printf ("result=%lld\n", func ());
14441 However, if the selected stack frame does not have a debug info, e.g., if the
14442 function was compiled without debug info, @value{GDBN} has to find out the type
14443 to return from user. Specifying a different type by mistake may set the value
14444 in different inferior registers than the caller code expects. For example,
14445 typing @kbd{return -1} with its implicit type @code{int} would set only a part
14446 of a @code{long long int} result for a debug info less function (on 32-bit
14447 architectures). Therefore the user is required to specify the return type by
14448 an appropriate cast explicitly:
14451 Breakpoint 2, 0x0040050b in func ()
14452 (@value{GDBP}) return -1
14453 Return value type not available for selected stack frame.
14454 Please use an explicit cast of the value to return.
14455 (@value{GDBP}) return (long long int) -1
14456 Make selected stack frame return now? (y or n) y
14457 #0 0x00400526 in main ()
14462 @section Calling Program Functions
14465 @cindex calling functions
14466 @cindex inferior functions, calling
14467 @item print @var{expr}
14468 Evaluate the expression @var{expr} and display the resulting value.
14469 @var{expr} may include calls to functions in the program being
14473 @item call @var{expr}
14474 Evaluate the expression @var{expr} without displaying @code{void}
14477 You can use this variant of the @code{print} command if you want to
14478 execute a function from your program that does not return anything
14479 (a.k.a.@: @dfn{a void function}), but without cluttering the output
14480 with @code{void} returned values that @value{GDBN} will otherwise
14481 print. If the result is not void, it is printed and saved in the
14485 It is possible for the function you call via the @code{print} or
14486 @code{call} command to generate a signal (e.g., if there's a bug in
14487 the function, or if you passed it incorrect arguments). What happens
14488 in that case is controlled by the @code{set unwindonsignal} command.
14490 Similarly, with a C@t{++} program it is possible for the function you
14491 call via the @code{print} or @code{call} command to generate an
14492 exception that is not handled due to the constraints of the dummy
14493 frame. In this case, any exception that is raised in the frame, but has
14494 an out-of-frame exception handler will not be found. GDB builds a
14495 dummy-frame for the inferior function call, and the unwinder cannot
14496 seek for exception handlers outside of this dummy-frame. What happens
14497 in that case is controlled by the
14498 @code{set unwind-on-terminating-exception} command.
14501 @item set unwindonsignal
14502 @kindex set unwindonsignal
14503 @cindex unwind stack in called functions
14504 @cindex call dummy stack unwinding
14505 Set unwinding of the stack if a signal is received while in a function
14506 that @value{GDBN} called in the program being debugged. If set to on,
14507 @value{GDBN} unwinds the stack it created for the call and restores
14508 the context to what it was before the call. If set to off (the
14509 default), @value{GDBN} stops in the frame where the signal was
14512 @item show unwindonsignal
14513 @kindex show unwindonsignal
14514 Show the current setting of stack unwinding in the functions called by
14517 @item set unwind-on-terminating-exception
14518 @kindex set unwind-on-terminating-exception
14519 @cindex unwind stack in called functions with unhandled exceptions
14520 @cindex call dummy stack unwinding on unhandled exception.
14521 Set unwinding of the stack if a C@t{++} exception is raised, but left
14522 unhandled while in a function that @value{GDBN} called in the program being
14523 debugged. If set to on (the default), @value{GDBN} unwinds the stack
14524 it created for the call and restores the context to what it was before
14525 the call. If set to off, @value{GDBN} the exception is delivered to
14526 the default C@t{++} exception handler and the inferior terminated.
14528 @item show unwind-on-terminating-exception
14529 @kindex show unwind-on-terminating-exception
14530 Show the current setting of stack unwinding in the functions called by
14535 @cindex weak alias functions
14536 Sometimes, a function you wish to call is actually a @dfn{weak alias}
14537 for another function. In such case, @value{GDBN} might not pick up
14538 the type information, including the types of the function arguments,
14539 which causes @value{GDBN} to call the inferior function incorrectly.
14540 As a result, the called function will function erroneously and may
14541 even crash. A solution to that is to use the name of the aliased
14545 @section Patching Programs
14547 @cindex patching binaries
14548 @cindex writing into executables
14549 @cindex writing into corefiles
14551 By default, @value{GDBN} opens the file containing your program's
14552 executable code (or the corefile) read-only. This prevents accidental
14553 alterations to machine code; but it also prevents you from intentionally
14554 patching your program's binary.
14556 If you'd like to be able to patch the binary, you can specify that
14557 explicitly with the @code{set write} command. For example, you might
14558 want to turn on internal debugging flags, or even to make emergency
14564 @itemx set write off
14565 If you specify @samp{set write on}, @value{GDBN} opens executable and
14566 core files for both reading and writing; if you specify @kbd{set write
14567 off} (the default), @value{GDBN} opens them read-only.
14569 If you have already loaded a file, you must load it again (using the
14570 @code{exec-file} or @code{core-file} command) after changing @code{set
14571 write}, for your new setting to take effect.
14575 Display whether executable files and core files are opened for writing
14576 as well as reading.
14580 @chapter @value{GDBN} Files
14582 @value{GDBN} needs to know the file name of the program to be debugged,
14583 both in order to read its symbol table and in order to start your
14584 program. To debug a core dump of a previous run, you must also tell
14585 @value{GDBN} the name of the core dump file.
14588 * Files:: Commands to specify files
14589 * Separate Debug Files:: Debugging information in separate files
14590 * Index Files:: Index files speed up GDB
14591 * Symbol Errors:: Errors reading symbol files
14592 * Data Files:: GDB data files
14596 @section Commands to Specify Files
14598 @cindex symbol table
14599 @cindex core dump file
14601 You may want to specify executable and core dump file names. The usual
14602 way to do this is at start-up time, using the arguments to
14603 @value{GDBN}'s start-up commands (@pxref{Invocation, , Getting In and
14604 Out of @value{GDBN}}).
14606 Occasionally it is necessary to change to a different file during a
14607 @value{GDBN} session. Or you may run @value{GDBN} and forget to
14608 specify a file you want to use. Or you are debugging a remote target
14609 via @code{gdbserver} (@pxref{Server, file, Using the @code{gdbserver}
14610 Program}). In these situations the @value{GDBN} commands to specify
14611 new files are useful.
14614 @cindex executable file
14616 @item file @var{filename}
14617 Use @var{filename} as the program to be debugged. It is read for its
14618 symbols and for the contents of pure memory. It is also the program
14619 executed when you use the @code{run} command. If you do not specify a
14620 directory and the file is not found in the @value{GDBN} working directory,
14621 @value{GDBN} uses the environment variable @code{PATH} as a list of
14622 directories to search, just as the shell does when looking for a program
14623 to run. You can change the value of this variable, for both @value{GDBN}
14624 and your program, using the @code{path} command.
14626 @cindex unlinked object files
14627 @cindex patching object files
14628 You can load unlinked object @file{.o} files into @value{GDBN} using
14629 the @code{file} command. You will not be able to ``run'' an object
14630 file, but you can disassemble functions and inspect variables. Also,
14631 if the underlying BFD functionality supports it, you could use
14632 @kbd{gdb -write} to patch object files using this technique. Note
14633 that @value{GDBN} can neither interpret nor modify relocations in this
14634 case, so branches and some initialized variables will appear to go to
14635 the wrong place. But this feature is still handy from time to time.
14638 @code{file} with no argument makes @value{GDBN} discard any information it
14639 has on both executable file and the symbol table.
14642 @item exec-file @r{[} @var{filename} @r{]}
14643 Specify that the program to be run (but not the symbol table) is found
14644 in @var{filename}. @value{GDBN} searches the environment variable @code{PATH}
14645 if necessary to locate your program. Omitting @var{filename} means to
14646 discard information on the executable file.
14648 @kindex symbol-file
14649 @item symbol-file @r{[} @var{filename} @r{]}
14650 Read symbol table information from file @var{filename}. @code{PATH} is
14651 searched when necessary. Use the @code{file} command to get both symbol
14652 table and program to run from the same file.
14654 @code{symbol-file} with no argument clears out @value{GDBN} information on your
14655 program's symbol table.
14657 The @code{symbol-file} command causes @value{GDBN} to forget the contents of
14658 some breakpoints and auto-display expressions. This is because they may
14659 contain pointers to the internal data recording symbols and data types,
14660 which are part of the old symbol table data being discarded inside
14663 @code{symbol-file} does not repeat if you press @key{RET} again after
14666 When @value{GDBN} is configured for a particular environment, it
14667 understands debugging information in whatever format is the standard
14668 generated for that environment; you may use either a @sc{gnu} compiler, or
14669 other compilers that adhere to the local conventions.
14670 Best results are usually obtained from @sc{gnu} compilers; for example,
14671 using @code{@value{NGCC}} you can generate debugging information for
14674 For most kinds of object files, with the exception of old SVR3 systems
14675 using COFF, the @code{symbol-file} command does not normally read the
14676 symbol table in full right away. Instead, it scans the symbol table
14677 quickly to find which source files and which symbols are present. The
14678 details are read later, one source file at a time, as they are needed.
14680 The purpose of this two-stage reading strategy is to make @value{GDBN}
14681 start up faster. For the most part, it is invisible except for
14682 occasional pauses while the symbol table details for a particular source
14683 file are being read. (The @code{set verbose} command can turn these
14684 pauses into messages if desired. @xref{Messages/Warnings, ,Optional
14685 Warnings and Messages}.)
14687 We have not implemented the two-stage strategy for COFF yet. When the
14688 symbol table is stored in COFF format, @code{symbol-file} reads the
14689 symbol table data in full right away. Note that ``stabs-in-COFF''
14690 still does the two-stage strategy, since the debug info is actually
14694 @cindex reading symbols immediately
14695 @cindex symbols, reading immediately
14696 @item symbol-file @r{[} -readnow @r{]} @var{filename}
14697 @itemx file @r{[} -readnow @r{]} @var{filename}
14698 You can override the @value{GDBN} two-stage strategy for reading symbol
14699 tables by using the @samp{-readnow} option with any of the commands that
14700 load symbol table information, if you want to be sure @value{GDBN} has the
14701 entire symbol table available.
14703 @c FIXME: for now no mention of directories, since this seems to be in
14704 @c flux. 13mar1992 status is that in theory GDB would look either in
14705 @c current dir or in same dir as myprog; but issues like competing
14706 @c GDB's, or clutter in system dirs, mean that in practice right now
14707 @c only current dir is used. FFish says maybe a special GDB hierarchy
14708 @c (eg rooted in val of env var GDBSYMS) could exist for mappable symbol
14712 @item core-file @r{[}@var{filename}@r{]}
14714 Specify the whereabouts of a core dump file to be used as the ``contents
14715 of memory''. Traditionally, core files contain only some parts of the
14716 address space of the process that generated them; @value{GDBN} can access the
14717 executable file itself for other parts.
14719 @code{core-file} with no argument specifies that no core file is
14722 Note that the core file is ignored when your program is actually running
14723 under @value{GDBN}. So, if you have been running your program and you
14724 wish to debug a core file instead, you must kill the subprocess in which
14725 the program is running. To do this, use the @code{kill} command
14726 (@pxref{Kill Process, ,Killing the Child Process}).
14728 @kindex add-symbol-file
14729 @cindex dynamic linking
14730 @item add-symbol-file @var{filename} @var{address}
14731 @itemx add-symbol-file @var{filename} @var{address} @r{[} -readnow @r{]}
14732 @itemx add-symbol-file @var{filename} @r{-s}@var{section} @var{address} @dots{}
14733 The @code{add-symbol-file} command reads additional symbol table
14734 information from the file @var{filename}. You would use this command
14735 when @var{filename} has been dynamically loaded (by some other means)
14736 into the program that is running. @var{address} should be the memory
14737 address at which the file has been loaded; @value{GDBN} cannot figure
14738 this out for itself. You can additionally specify an arbitrary number
14739 of @samp{@r{-s}@var{section} @var{address}} pairs, to give an explicit
14740 section name and base address for that section. You can specify any
14741 @var{address} as an expression.
14743 The symbol table of the file @var{filename} is added to the symbol table
14744 originally read with the @code{symbol-file} command. You can use the
14745 @code{add-symbol-file} command any number of times; the new symbol data
14746 thus read keeps adding to the old. To discard all old symbol data
14747 instead, use the @code{symbol-file} command without any arguments.
14749 @cindex relocatable object files, reading symbols from
14750 @cindex object files, relocatable, reading symbols from
14751 @cindex reading symbols from relocatable object files
14752 @cindex symbols, reading from relocatable object files
14753 @cindex @file{.o} files, reading symbols from
14754 Although @var{filename} is typically a shared library file, an
14755 executable file, or some other object file which has been fully
14756 relocated for loading into a process, you can also load symbolic
14757 information from relocatable @file{.o} files, as long as:
14761 the file's symbolic information refers only to linker symbols defined in
14762 that file, not to symbols defined by other object files,
14764 every section the file's symbolic information refers to has actually
14765 been loaded into the inferior, as it appears in the file, and
14767 you can determine the address at which every section was loaded, and
14768 provide these to the @code{add-symbol-file} command.
14772 Some embedded operating systems, like Sun Chorus and VxWorks, can load
14773 relocatable files into an already running program; such systems
14774 typically make the requirements above easy to meet. However, it's
14775 important to recognize that many native systems use complex link
14776 procedures (@code{.linkonce} section factoring and C@t{++} constructor table
14777 assembly, for example) that make the requirements difficult to meet. In
14778 general, one cannot assume that using @code{add-symbol-file} to read a
14779 relocatable object file's symbolic information will have the same effect
14780 as linking the relocatable object file into the program in the normal
14783 @code{add-symbol-file} does not repeat if you press @key{RET} after using it.
14785 @kindex add-symbol-file-from-memory
14786 @cindex @code{syscall DSO}
14787 @cindex load symbols from memory
14788 @item add-symbol-file-from-memory @var{address}
14789 Load symbols from the given @var{address} in a dynamically loaded
14790 object file whose image is mapped directly into the inferior's memory.
14791 For example, the Linux kernel maps a @code{syscall DSO} into each
14792 process's address space; this DSO provides kernel-specific code for
14793 some system calls. The argument can be any expression whose
14794 evaluation yields the address of the file's shared object file header.
14795 For this command to work, you must have used @code{symbol-file} or
14796 @code{exec-file} commands in advance.
14798 @kindex add-shared-symbol-files
14800 @item add-shared-symbol-files @var{library-file}
14801 @itemx assf @var{library-file}
14802 The @code{add-shared-symbol-files} command can currently be used only
14803 in the Cygwin build of @value{GDBN} on MS-Windows OS, where it is an
14804 alias for the @code{dll-symbols} command (@pxref{Cygwin Native}).
14805 @value{GDBN} automatically looks for shared libraries, however if
14806 @value{GDBN} does not find yours, you can invoke
14807 @code{add-shared-symbol-files}. It takes one argument: the shared
14808 library's file name. @code{assf} is a shorthand alias for
14809 @code{add-shared-symbol-files}.
14812 @item section @var{section} @var{addr}
14813 The @code{section} command changes the base address of the named
14814 @var{section} of the exec file to @var{addr}. This can be used if the
14815 exec file does not contain section addresses, (such as in the
14816 @code{a.out} format), or when the addresses specified in the file
14817 itself are wrong. Each section must be changed separately. The
14818 @code{info files} command, described below, lists all the sections and
14822 @kindex info target
14825 @code{info files} and @code{info target} are synonymous; both print the
14826 current target (@pxref{Targets, ,Specifying a Debugging Target}),
14827 including the names of the executable and core dump files currently in
14828 use by @value{GDBN}, and the files from which symbols were loaded. The
14829 command @code{help target} lists all possible targets rather than
14832 @kindex maint info sections
14833 @item maint info sections
14834 Another command that can give you extra information about program sections
14835 is @code{maint info sections}. In addition to the section information
14836 displayed by @code{info files}, this command displays the flags and file
14837 offset of each section in the executable and core dump files. In addition,
14838 @code{maint info sections} provides the following command options (which
14839 may be arbitrarily combined):
14843 Display sections for all loaded object files, including shared libraries.
14844 @item @var{sections}
14845 Display info only for named @var{sections}.
14846 @item @var{section-flags}
14847 Display info only for sections for which @var{section-flags} are true.
14848 The section flags that @value{GDBN} currently knows about are:
14851 Section will have space allocated in the process when loaded.
14852 Set for all sections except those containing debug information.
14854 Section will be loaded from the file into the child process memory.
14855 Set for pre-initialized code and data, clear for @code{.bss} sections.
14857 Section needs to be relocated before loading.
14859 Section cannot be modified by the child process.
14861 Section contains executable code only.
14863 Section contains data only (no executable code).
14865 Section will reside in ROM.
14867 Section contains data for constructor/destructor lists.
14869 Section is not empty.
14871 An instruction to the linker to not output the section.
14872 @item COFF_SHARED_LIBRARY
14873 A notification to the linker that the section contains
14874 COFF shared library information.
14876 Section contains common symbols.
14879 @kindex set trust-readonly-sections
14880 @cindex read-only sections
14881 @item set trust-readonly-sections on
14882 Tell @value{GDBN} that readonly sections in your object file
14883 really are read-only (i.e.@: that their contents will not change).
14884 In that case, @value{GDBN} can fetch values from these sections
14885 out of the object file, rather than from the target program.
14886 For some targets (notably embedded ones), this can be a significant
14887 enhancement to debugging performance.
14889 The default is off.
14891 @item set trust-readonly-sections off
14892 Tell @value{GDBN} not to trust readonly sections. This means that
14893 the contents of the section might change while the program is running,
14894 and must therefore be fetched from the target when needed.
14896 @item show trust-readonly-sections
14897 Show the current setting of trusting readonly sections.
14900 All file-specifying commands allow both absolute and relative file names
14901 as arguments. @value{GDBN} always converts the file name to an absolute file
14902 name and remembers it that way.
14904 @cindex shared libraries
14905 @anchor{Shared Libraries}
14906 @value{GDBN} supports @sc{gnu}/Linux, MS-Windows, HP-UX, SunOS, SVr4, Irix,
14907 and IBM RS/6000 AIX shared libraries.
14909 On MS-Windows @value{GDBN} must be linked with the Expat library to support
14910 shared libraries. @xref{Expat}.
14912 @value{GDBN} automatically loads symbol definitions from shared libraries
14913 when you use the @code{run} command, or when you examine a core file.
14914 (Before you issue the @code{run} command, @value{GDBN} does not understand
14915 references to a function in a shared library, however---unless you are
14916 debugging a core file).
14918 On HP-UX, if the program loads a library explicitly, @value{GDBN}
14919 automatically loads the symbols at the time of the @code{shl_load} call.
14921 @c FIXME: some @value{GDBN} release may permit some refs to undef
14922 @c FIXME...symbols---eg in a break cmd---assuming they are from a shared
14923 @c FIXME...lib; check this from time to time when updating manual
14925 There are times, however, when you may wish to not automatically load
14926 symbol definitions from shared libraries, such as when they are
14927 particularly large or there are many of them.
14929 To control the automatic loading of shared library symbols, use the
14933 @kindex set auto-solib-add
14934 @item set auto-solib-add @var{mode}
14935 If @var{mode} is @code{on}, symbols from all shared object libraries
14936 will be loaded automatically when the inferior begins execution, you
14937 attach to an independently started inferior, or when the dynamic linker
14938 informs @value{GDBN} that a new library has been loaded. If @var{mode}
14939 is @code{off}, symbols must be loaded manually, using the
14940 @code{sharedlibrary} command. The default value is @code{on}.
14942 @cindex memory used for symbol tables
14943 If your program uses lots of shared libraries with debug info that
14944 takes large amounts of memory, you can decrease the @value{GDBN}
14945 memory footprint by preventing it from automatically loading the
14946 symbols from shared libraries. To that end, type @kbd{set
14947 auto-solib-add off} before running the inferior, then load each
14948 library whose debug symbols you do need with @kbd{sharedlibrary
14949 @var{regexp}}, where @var{regexp} is a regular expression that matches
14950 the libraries whose symbols you want to be loaded.
14952 @kindex show auto-solib-add
14953 @item show auto-solib-add
14954 Display the current autoloading mode.
14957 @cindex load shared library
14958 To explicitly load shared library symbols, use the @code{sharedlibrary}
14962 @kindex info sharedlibrary
14964 @item info share @var{regex}
14965 @itemx info sharedlibrary @var{regex}
14966 Print the names of the shared libraries which are currently loaded
14967 that match @var{regex}. If @var{regex} is omitted then print
14968 all shared libraries that are loaded.
14970 @kindex sharedlibrary
14972 @item sharedlibrary @var{regex}
14973 @itemx share @var{regex}
14974 Load shared object library symbols for files matching a
14975 Unix regular expression.
14976 As with files loaded automatically, it only loads shared libraries
14977 required by your program for a core file or after typing @code{run}. If
14978 @var{regex} is omitted all shared libraries required by your program are
14981 @item nosharedlibrary
14982 @kindex nosharedlibrary
14983 @cindex unload symbols from shared libraries
14984 Unload all shared object library symbols. This discards all symbols
14985 that have been loaded from all shared libraries. Symbols from shared
14986 libraries that were loaded by explicit user requests are not
14990 Sometimes you may wish that @value{GDBN} stops and gives you control
14991 when any of shared library events happen. Use the @code{set
14992 stop-on-solib-events} command for this:
14995 @item set stop-on-solib-events
14996 @kindex set stop-on-solib-events
14997 This command controls whether @value{GDBN} should give you control
14998 when the dynamic linker notifies it about some shared library event.
14999 The most common event of interest is loading or unloading of a new
15002 @item show stop-on-solib-events
15003 @kindex show stop-on-solib-events
15004 Show whether @value{GDBN} stops and gives you control when shared
15005 library events happen.
15008 Shared libraries are also supported in many cross or remote debugging
15009 configurations. @value{GDBN} needs to have access to the target's libraries;
15010 this can be accomplished either by providing copies of the libraries
15011 on the host system, or by asking @value{GDBN} to automatically retrieve the
15012 libraries from the target. If copies of the target libraries are
15013 provided, they need to be the same as the target libraries, although the
15014 copies on the target can be stripped as long as the copies on the host are
15017 @cindex where to look for shared libraries
15018 For remote debugging, you need to tell @value{GDBN} where the target
15019 libraries are, so that it can load the correct copies---otherwise, it
15020 may try to load the host's libraries. @value{GDBN} has two variables
15021 to specify the search directories for target libraries.
15024 @cindex prefix for shared library file names
15025 @cindex system root, alternate
15026 @kindex set solib-absolute-prefix
15027 @kindex set sysroot
15028 @item set sysroot @var{path}
15029 Use @var{path} as the system root for the program being debugged. Any
15030 absolute shared library paths will be prefixed with @var{path}; many
15031 runtime loaders store the absolute paths to the shared library in the
15032 target program's memory. If you use @code{set sysroot} to find shared
15033 libraries, they need to be laid out in the same way that they are on
15034 the target, with e.g.@: a @file{/lib} and @file{/usr/lib} hierarchy
15037 If @var{path} starts with the sequence @file{remote:}, @value{GDBN} will
15038 retrieve the target libraries from the remote system. This is only
15039 supported when using a remote target that supports the @code{remote get}
15040 command (@pxref{File Transfer,,Sending files to a remote system}).
15041 The part of @var{path} following the initial @file{remote:}
15042 (if present) is used as system root prefix on the remote file system.
15043 @footnote{If you want to specify a local system root using a directory
15044 that happens to be named @file{remote:}, you need to use some equivalent
15045 variant of the name like @file{./remote:}.}
15047 For targets with an MS-DOS based filesystem, such as MS-Windows and
15048 SymbianOS, @value{GDBN} tries prefixing a few variants of the target
15049 absolute file name with @var{path}. But first, on Unix hosts,
15050 @value{GDBN} converts all backslash directory separators into forward
15051 slashes, because the backslash is not a directory separator on Unix:
15054 c:\foo\bar.dll @result{} c:/foo/bar.dll
15057 Then, @value{GDBN} attempts prefixing the target file name with
15058 @var{path}, and looks for the resulting file name in the host file
15062 c:/foo/bar.dll @result{} /path/to/sysroot/c:/foo/bar.dll
15065 If that does not find the shared library, @value{GDBN} tries removing
15066 the @samp{:} character from the drive spec, both for convenience, and,
15067 for the case of the host file system not supporting file names with
15071 c:/foo/bar.dll @result{} /path/to/sysroot/c/foo/bar.dll
15074 This makes it possible to have a system root that mirrors a target
15075 with more than one drive. E.g., you may want to setup your local
15076 copies of the target system shared libraries like so (note @samp{c} vs
15080 @file{/path/to/sysroot/c/sys/bin/foo.dll}
15081 @file{/path/to/sysroot/c/sys/bin/bar.dll}
15082 @file{/path/to/sysroot/z/sys/bin/bar.dll}
15086 and point the system root at @file{/path/to/sysroot}, so that
15087 @value{GDBN} can find the correct copies of both
15088 @file{c:\sys\bin\foo.dll}, and @file{z:\sys\bin\bar.dll}.
15090 If that still does not find the shared library, @value{GDBN} tries
15091 removing the whole drive spec from the target file name:
15094 c:/foo/bar.dll @result{} /path/to/sysroot/foo/bar.dll
15097 This last lookup makes it possible to not care about the drive name,
15098 if you don't want or need to.
15100 The @code{set solib-absolute-prefix} command is an alias for @code{set
15103 @cindex default system root
15104 @cindex @samp{--with-sysroot}
15105 You can set the default system root by using the configure-time
15106 @samp{--with-sysroot} option. If the system root is inside
15107 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
15108 @samp{--exec-prefix}), then the default system root will be updated
15109 automatically if the installed @value{GDBN} is moved to a new
15112 @kindex show sysroot
15114 Display the current shared library prefix.
15116 @kindex set solib-search-path
15117 @item set solib-search-path @var{path}
15118 If this variable is set, @var{path} is a colon-separated list of
15119 directories to search for shared libraries. @samp{solib-search-path}
15120 is used after @samp{sysroot} fails to locate the library, or if the
15121 path to the library is relative instead of absolute. If you want to
15122 use @samp{solib-search-path} instead of @samp{sysroot}, be sure to set
15123 @samp{sysroot} to a nonexistent directory to prevent @value{GDBN} from
15124 finding your host's libraries. @samp{sysroot} is preferred; setting
15125 it to a nonexistent directory may interfere with automatic loading
15126 of shared library symbols.
15128 @kindex show solib-search-path
15129 @item show solib-search-path
15130 Display the current shared library search path.
15132 @cindex DOS file-name semantics of file names.
15133 @kindex set target-file-system-kind (unix|dos-based|auto)
15134 @kindex show target-file-system-kind
15135 @item set target-file-system-kind @var{kind}
15136 Set assumed file system kind for target reported file names.
15138 Shared library file names as reported by the target system may not
15139 make sense as is on the system @value{GDBN} is running on. For
15140 example, when remote debugging a target that has MS-DOS based file
15141 system semantics, from a Unix host, the target may be reporting to
15142 @value{GDBN} a list of loaded shared libraries with file names such as
15143 @file{c:\Windows\kernel32.dll}. On Unix hosts, there's no concept of
15144 drive letters, so the @samp{c:\} prefix is not normally understood as
15145 indicating an absolute file name, and neither is the backslash
15146 normally considered a directory separator character. In that case,
15147 the native file system would interpret this whole absolute file name
15148 as a relative file name with no directory components. This would make
15149 it impossible to point @value{GDBN} at a copy of the remote target's
15150 shared libraries on the host using @code{set sysroot}, and impractical
15151 with @code{set solib-search-path}. Setting
15152 @code{target-file-system-kind} to @code{dos-based} tells @value{GDBN}
15153 to interpret such file names similarly to how the target would, and to
15154 map them to file names valid on @value{GDBN}'s native file system
15155 semantics. The value of @var{kind} can be @code{"auto"}, in addition
15156 to one of the supported file system kinds. In that case, @value{GDBN}
15157 tries to determine the appropriate file system variant based on the
15158 current target's operating system (@pxref{ABI, ,Configuring the
15159 Current ABI}). The supported file system settings are:
15163 Instruct @value{GDBN} to assume the target file system is of Unix
15164 kind. Only file names starting the forward slash (@samp{/}) character
15165 are considered absolute, and the directory separator character is also
15169 Instruct @value{GDBN} to assume the target file system is DOS based.
15170 File names starting with either a forward slash, or a drive letter
15171 followed by a colon (e.g., @samp{c:}), are considered absolute, and
15172 both the slash (@samp{/}) and the backslash (@samp{\\}) characters are
15173 considered directory separators.
15176 Instruct @value{GDBN} to use the file system kind associated with the
15177 target operating system (@pxref{ABI, ,Configuring the Current ABI}).
15178 This is the default.
15183 @node Separate Debug Files
15184 @section Debugging Information in Separate Files
15185 @cindex separate debugging information files
15186 @cindex debugging information in separate files
15187 @cindex @file{.debug} subdirectories
15188 @cindex debugging information directory, global
15189 @cindex global debugging information directory
15190 @cindex build ID, and separate debugging files
15191 @cindex @file{.build-id} directory
15193 @value{GDBN} allows you to put a program's debugging information in a
15194 file separate from the executable itself, in a way that allows
15195 @value{GDBN} to find and load the debugging information automatically.
15196 Since debugging information can be very large---sometimes larger
15197 than the executable code itself---some systems distribute debugging
15198 information for their executables in separate files, which users can
15199 install only when they need to debug a problem.
15201 @value{GDBN} supports two ways of specifying the separate debug info
15206 The executable contains a @dfn{debug link} that specifies the name of
15207 the separate debug info file. The separate debug file's name is
15208 usually @file{@var{executable}.debug}, where @var{executable} is the
15209 name of the corresponding executable file without leading directories
15210 (e.g., @file{ls.debug} for @file{/usr/bin/ls}). In addition, the
15211 debug link specifies a 32-bit @dfn{Cyclic Redundancy Check} (CRC)
15212 checksum for the debug file, which @value{GDBN} uses to validate that
15213 the executable and the debug file came from the same build.
15216 The executable contains a @dfn{build ID}, a unique bit string that is
15217 also present in the corresponding debug info file. (This is supported
15218 only on some operating systems, notably those which use the ELF format
15219 for binary files and the @sc{gnu} Binutils.) For more details about
15220 this feature, see the description of the @option{--build-id}
15221 command-line option in @ref{Options, , Command Line Options, ld.info,
15222 The GNU Linker}. The debug info file's name is not specified
15223 explicitly by the build ID, but can be computed from the build ID, see
15227 Depending on the way the debug info file is specified, @value{GDBN}
15228 uses two different methods of looking for the debug file:
15232 For the ``debug link'' method, @value{GDBN} looks up the named file in
15233 the directory of the executable file, then in a subdirectory of that
15234 directory named @file{.debug}, and finally under the global debug
15235 directory, in a subdirectory whose name is identical to the leading
15236 directories of the executable's absolute file name.
15239 For the ``build ID'' method, @value{GDBN} looks in the
15240 @file{.build-id} subdirectory of the global debug directory for a file
15241 named @file{@var{nn}/@var{nnnnnnnn}.debug}, where @var{nn} are the
15242 first 2 hex characters of the build ID bit string, and @var{nnnnnnnn}
15243 are the rest of the bit string. (Real build ID strings are 32 or more
15244 hex characters, not 10.)
15247 So, for example, suppose you ask @value{GDBN} to debug
15248 @file{/usr/bin/ls}, which has a debug link that specifies the
15249 file @file{ls.debug}, and a build ID whose value in hex is
15250 @code{abcdef1234}. If the global debug directory is
15251 @file{/usr/lib/debug}, then @value{GDBN} will look for the following
15252 debug information files, in the indicated order:
15256 @file{/usr/lib/debug/.build-id/ab/cdef1234.debug}
15258 @file{/usr/bin/ls.debug}
15260 @file{/usr/bin/.debug/ls.debug}
15262 @file{/usr/lib/debug/usr/bin/ls.debug}.
15265 You can set the global debugging info directory's name, and view the
15266 name @value{GDBN} is currently using.
15270 @kindex set debug-file-directory
15271 @item set debug-file-directory @var{directories}
15272 Set the directories which @value{GDBN} searches for separate debugging
15273 information files to @var{directory}. Multiple directory components can be set
15274 concatenating them by a directory separator.
15276 @kindex show debug-file-directory
15277 @item show debug-file-directory
15278 Show the directories @value{GDBN} searches for separate debugging
15283 @cindex @code{.gnu_debuglink} sections
15284 @cindex debug link sections
15285 A debug link is a special section of the executable file named
15286 @code{.gnu_debuglink}. The section must contain:
15290 A filename, with any leading directory components removed, followed by
15293 zero to three bytes of padding, as needed to reach the next four-byte
15294 boundary within the section, and
15296 a four-byte CRC checksum, stored in the same endianness used for the
15297 executable file itself. The checksum is computed on the debugging
15298 information file's full contents by the function given below, passing
15299 zero as the @var{crc} argument.
15302 Any executable file format can carry a debug link, as long as it can
15303 contain a section named @code{.gnu_debuglink} with the contents
15306 @cindex @code{.note.gnu.build-id} sections
15307 @cindex build ID sections
15308 The build ID is a special section in the executable file (and in other
15309 ELF binary files that @value{GDBN} may consider). This section is
15310 often named @code{.note.gnu.build-id}, but that name is not mandatory.
15311 It contains unique identification for the built files---the ID remains
15312 the same across multiple builds of the same build tree. The default
15313 algorithm SHA1 produces 160 bits (40 hexadecimal characters) of the
15314 content for the build ID string. The same section with an identical
15315 value is present in the original built binary with symbols, in its
15316 stripped variant, and in the separate debugging information file.
15318 The debugging information file itself should be an ordinary
15319 executable, containing a full set of linker symbols, sections, and
15320 debugging information. The sections of the debugging information file
15321 should have the same names, addresses, and sizes as the original file,
15322 but they need not contain any data---much like a @code{.bss} section
15323 in an ordinary executable.
15325 The @sc{gnu} binary utilities (Binutils) package includes the
15326 @samp{objcopy} utility that can produce
15327 the separated executable / debugging information file pairs using the
15328 following commands:
15331 @kbd{objcopy --only-keep-debug foo foo.debug}
15336 These commands remove the debugging
15337 information from the executable file @file{foo} and place it in the file
15338 @file{foo.debug}. You can use the first, second or both methods to link the
15343 The debug link method needs the following additional command to also leave
15344 behind a debug link in @file{foo}:
15347 @kbd{objcopy --add-gnu-debuglink=foo.debug foo}
15350 Ulrich Drepper's @file{elfutils} package, starting with version 0.53, contains
15351 a version of the @code{strip} command such that the command @kbd{strip foo -f
15352 foo.debug} has the same functionality as the two @code{objcopy} commands and
15353 the @code{ln -s} command above, together.
15356 Build ID gets embedded into the main executable using @code{ld --build-id} or
15357 the @value{NGCC} counterpart @code{gcc -Wl,--build-id}. Build ID support plus
15358 compatibility fixes for debug files separation are present in @sc{gnu} binary
15359 utilities (Binutils) package since version 2.18.
15364 @cindex CRC algorithm definition
15365 The CRC used in @code{.gnu_debuglink} is the CRC-32 defined in
15366 IEEE 802.3 using the polynomial:
15368 @c TexInfo requires naked braces for multi-digit exponents for Tex
15369 @c output, but this causes HTML output to barf. HTML has to be set using
15370 @c raw commands. So we end up having to specify this equation in 2
15375 <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>
15376 + <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
15382 @math{x^{32} + x^{26} + x^{23} + x^{22} + x^{16} + x^{12} + x^{11}}
15383 @math{+ x^{10} + x^8 + x^7 + x^5 + x^4 + x^2 + x + 1}
15387 The function is computed byte at a time, taking the least
15388 significant bit of each byte first. The initial pattern
15389 @code{0xffffffff} is used, to ensure leading zeros affect the CRC and
15390 the final result is inverted to ensure trailing zeros also affect the
15393 @emph{Note:} This is the same CRC polynomial as used in handling the
15394 @dfn{Remote Serial Protocol} @code{qCRC} packet (@pxref{Remote Protocol,
15395 , @value{GDBN} Remote Serial Protocol}). However in the
15396 case of the Remote Serial Protocol, the CRC is computed @emph{most}
15397 significant bit first, and the result is not inverted, so trailing
15398 zeros have no effect on the CRC value.
15400 To complete the description, we show below the code of the function
15401 which produces the CRC used in @code{.gnu_debuglink}. Inverting the
15402 initially supplied @code{crc} argument means that an initial call to
15403 this function passing in zero will start computing the CRC using
15406 @kindex gnu_debuglink_crc32
15409 gnu_debuglink_crc32 (unsigned long crc,
15410 unsigned char *buf, size_t len)
15412 static const unsigned long crc32_table[256] =
15414 0x00000000, 0x77073096, 0xee0e612c, 0x990951ba, 0x076dc419,
15415 0x706af48f, 0xe963a535, 0x9e6495a3, 0x0edb8832, 0x79dcb8a4,
15416 0xe0d5e91e, 0x97d2d988, 0x09b64c2b, 0x7eb17cbd, 0xe7b82d07,
15417 0x90bf1d91, 0x1db71064, 0x6ab020f2, 0xf3b97148, 0x84be41de,
15418 0x1adad47d, 0x6ddde4eb, 0xf4d4b551, 0x83d385c7, 0x136c9856,
15419 0x646ba8c0, 0xfd62f97a, 0x8a65c9ec, 0x14015c4f, 0x63066cd9,
15420 0xfa0f3d63, 0x8d080df5, 0x3b6e20c8, 0x4c69105e, 0xd56041e4,
15421 0xa2677172, 0x3c03e4d1, 0x4b04d447, 0xd20d85fd, 0xa50ab56b,
15422 0x35b5a8fa, 0x42b2986c, 0xdbbbc9d6, 0xacbcf940, 0x32d86ce3,
15423 0x45df5c75, 0xdcd60dcf, 0xabd13d59, 0x26d930ac, 0x51de003a,
15424 0xc8d75180, 0xbfd06116, 0x21b4f4b5, 0x56b3c423, 0xcfba9599,
15425 0xb8bda50f, 0x2802b89e, 0x5f058808, 0xc60cd9b2, 0xb10be924,
15426 0x2f6f7c87, 0x58684c11, 0xc1611dab, 0xb6662d3d, 0x76dc4190,
15427 0x01db7106, 0x98d220bc, 0xefd5102a, 0x71b18589, 0x06b6b51f,
15428 0x9fbfe4a5, 0xe8b8d433, 0x7807c9a2, 0x0f00f934, 0x9609a88e,
15429 0xe10e9818, 0x7f6a0dbb, 0x086d3d2d, 0x91646c97, 0xe6635c01,
15430 0x6b6b51f4, 0x1c6c6162, 0x856530d8, 0xf262004e, 0x6c0695ed,
15431 0x1b01a57b, 0x8208f4c1, 0xf50fc457, 0x65b0d9c6, 0x12b7e950,
15432 0x8bbeb8ea, 0xfcb9887c, 0x62dd1ddf, 0x15da2d49, 0x8cd37cf3,
15433 0xfbd44c65, 0x4db26158, 0x3ab551ce, 0xa3bc0074, 0xd4bb30e2,
15434 0x4adfa541, 0x3dd895d7, 0xa4d1c46d, 0xd3d6f4fb, 0x4369e96a,
15435 0x346ed9fc, 0xad678846, 0xda60b8d0, 0x44042d73, 0x33031de5,
15436 0xaa0a4c5f, 0xdd0d7cc9, 0x5005713c, 0x270241aa, 0xbe0b1010,
15437 0xc90c2086, 0x5768b525, 0x206f85b3, 0xb966d409, 0xce61e49f,
15438 0x5edef90e, 0x29d9c998, 0xb0d09822, 0xc7d7a8b4, 0x59b33d17,
15439 0x2eb40d81, 0xb7bd5c3b, 0xc0ba6cad, 0xedb88320, 0x9abfb3b6,
15440 0x03b6e20c, 0x74b1d29a, 0xead54739, 0x9dd277af, 0x04db2615,
15441 0x73dc1683, 0xe3630b12, 0x94643b84, 0x0d6d6a3e, 0x7a6a5aa8,
15442 0xe40ecf0b, 0x9309ff9d, 0x0a00ae27, 0x7d079eb1, 0xf00f9344,
15443 0x8708a3d2, 0x1e01f268, 0x6906c2fe, 0xf762575d, 0x806567cb,
15444 0x196c3671, 0x6e6b06e7, 0xfed41b76, 0x89d32be0, 0x10da7a5a,
15445 0x67dd4acc, 0xf9b9df6f, 0x8ebeeff9, 0x17b7be43, 0x60b08ed5,
15446 0xd6d6a3e8, 0xa1d1937e, 0x38d8c2c4, 0x4fdff252, 0xd1bb67f1,
15447 0xa6bc5767, 0x3fb506dd, 0x48b2364b, 0xd80d2bda, 0xaf0a1b4c,
15448 0x36034af6, 0x41047a60, 0xdf60efc3, 0xa867df55, 0x316e8eef,
15449 0x4669be79, 0xcb61b38c, 0xbc66831a, 0x256fd2a0, 0x5268e236,
15450 0xcc0c7795, 0xbb0b4703, 0x220216b9, 0x5505262f, 0xc5ba3bbe,
15451 0xb2bd0b28, 0x2bb45a92, 0x5cb36a04, 0xc2d7ffa7, 0xb5d0cf31,
15452 0x2cd99e8b, 0x5bdeae1d, 0x9b64c2b0, 0xec63f226, 0x756aa39c,
15453 0x026d930a, 0x9c0906a9, 0xeb0e363f, 0x72076785, 0x05005713,
15454 0x95bf4a82, 0xe2b87a14, 0x7bb12bae, 0x0cb61b38, 0x92d28e9b,
15455 0xe5d5be0d, 0x7cdcefb7, 0x0bdbdf21, 0x86d3d2d4, 0xf1d4e242,
15456 0x68ddb3f8, 0x1fda836e, 0x81be16cd, 0xf6b9265b, 0x6fb077e1,
15457 0x18b74777, 0x88085ae6, 0xff0f6a70, 0x66063bca, 0x11010b5c,
15458 0x8f659eff, 0xf862ae69, 0x616bffd3, 0x166ccf45, 0xa00ae278,
15459 0xd70dd2ee, 0x4e048354, 0x3903b3c2, 0xa7672661, 0xd06016f7,
15460 0x4969474d, 0x3e6e77db, 0xaed16a4a, 0xd9d65adc, 0x40df0b66,
15461 0x37d83bf0, 0xa9bcae53, 0xdebb9ec5, 0x47b2cf7f, 0x30b5ffe9,
15462 0xbdbdf21c, 0xcabac28a, 0x53b39330, 0x24b4a3a6, 0xbad03605,
15463 0xcdd70693, 0x54de5729, 0x23d967bf, 0xb3667a2e, 0xc4614ab8,
15464 0x5d681b02, 0x2a6f2b94, 0xb40bbe37, 0xc30c8ea1, 0x5a05df1b,
15467 unsigned char *end;
15469 crc = ~crc & 0xffffffff;
15470 for (end = buf + len; buf < end; ++buf)
15471 crc = crc32_table[(crc ^ *buf) & 0xff] ^ (crc >> 8);
15472 return ~crc & 0xffffffff;
15477 This computation does not apply to the ``build ID'' method.
15481 @section Index Files Speed Up @value{GDBN}
15482 @cindex index files
15483 @cindex @samp{.gdb_index} section
15485 When @value{GDBN} finds a symbol file, it scans the symbols in the
15486 file in order to construct an internal symbol table. This lets most
15487 @value{GDBN} operations work quickly---at the cost of a delay early
15488 on. For large programs, this delay can be quite lengthy, so
15489 @value{GDBN} provides a way to build an index, which speeds up
15492 The index is stored as a section in the symbol file. @value{GDBN} can
15493 write the index to a file, then you can put it into the symbol file
15494 using @command{objcopy}.
15496 To create an index file, use the @code{save gdb-index} command:
15499 @item save gdb-index @var{directory}
15500 @kindex save gdb-index
15501 Create an index file for each symbol file currently known by
15502 @value{GDBN}. Each file is named after its corresponding symbol file,
15503 with @samp{.gdb-index} appended, and is written into the given
15507 Once you have created an index file you can merge it into your symbol
15508 file, here named @file{symfile}, using @command{objcopy}:
15511 $ objcopy --add-section .gdb_index=symfile.gdb-index \
15512 --set-section-flags .gdb_index=readonly symfile symfile
15515 There are currently some limitation on indices. They only work when
15516 for DWARF debugging information, not stabs. And, they do not
15517 currently work for programs using Ada.
15519 @node Symbol Errors
15520 @section Errors Reading Symbol Files
15522 While reading a symbol file, @value{GDBN} occasionally encounters problems,
15523 such as symbol types it does not recognize, or known bugs in compiler
15524 output. By default, @value{GDBN} does not notify you of such problems, since
15525 they are relatively common and primarily of interest to people
15526 debugging compilers. If you are interested in seeing information
15527 about ill-constructed symbol tables, you can either ask @value{GDBN} to print
15528 only one message about each such type of problem, no matter how many
15529 times the problem occurs; or you can ask @value{GDBN} to print more messages,
15530 to see how many times the problems occur, with the @code{set
15531 complaints} command (@pxref{Messages/Warnings, ,Optional Warnings and
15534 The messages currently printed, and their meanings, include:
15537 @item inner block not inside outer block in @var{symbol}
15539 The symbol information shows where symbol scopes begin and end
15540 (such as at the start of a function or a block of statements). This
15541 error indicates that an inner scope block is not fully contained
15542 in its outer scope blocks.
15544 @value{GDBN} circumvents the problem by treating the inner block as if it had
15545 the same scope as the outer block. In the error message, @var{symbol}
15546 may be shown as ``@code{(don't know)}'' if the outer block is not a
15549 @item block at @var{address} out of order
15551 The symbol information for symbol scope blocks should occur in
15552 order of increasing addresses. This error indicates that it does not
15555 @value{GDBN} does not circumvent this problem, and has trouble
15556 locating symbols in the source file whose symbols it is reading. (You
15557 can often determine what source file is affected by specifying
15558 @code{set verbose on}. @xref{Messages/Warnings, ,Optional Warnings and
15561 @item bad block start address patched
15563 The symbol information for a symbol scope block has a start address
15564 smaller than the address of the preceding source line. This is known
15565 to occur in the SunOS 4.1.1 (and earlier) C compiler.
15567 @value{GDBN} circumvents the problem by treating the symbol scope block as
15568 starting on the previous source line.
15570 @item bad string table offset in symbol @var{n}
15573 Symbol number @var{n} contains a pointer into the string table which is
15574 larger than the size of the string table.
15576 @value{GDBN} circumvents the problem by considering the symbol to have the
15577 name @code{foo}, which may cause other problems if many symbols end up
15580 @item unknown symbol type @code{0x@var{nn}}
15582 The symbol information contains new data types that @value{GDBN} does
15583 not yet know how to read. @code{0x@var{nn}} is the symbol type of the
15584 uncomprehended information, in hexadecimal.
15586 @value{GDBN} circumvents the error by ignoring this symbol information.
15587 This usually allows you to debug your program, though certain symbols
15588 are not accessible. If you encounter such a problem and feel like
15589 debugging it, you can debug @code{@value{GDBP}} with itself, breakpoint
15590 on @code{complain}, then go up to the function @code{read_dbx_symtab}
15591 and examine @code{*bufp} to see the symbol.
15593 @item stub type has NULL name
15595 @value{GDBN} could not find the full definition for a struct or class.
15597 @item const/volatile indicator missing (ok if using g++ v1.x), got@dots{}
15598 The symbol information for a C@t{++} member function is missing some
15599 information that recent versions of the compiler should have output for
15602 @item info mismatch between compiler and debugger
15604 @value{GDBN} could not parse a type specification output by the compiler.
15609 @section GDB Data Files
15611 @cindex prefix for data files
15612 @value{GDBN} will sometimes read an auxiliary data file. These files
15613 are kept in a directory known as the @dfn{data directory}.
15615 You can set the data directory's name, and view the name @value{GDBN}
15616 is currently using.
15619 @kindex set data-directory
15620 @item set data-directory @var{directory}
15621 Set the directory which @value{GDBN} searches for auxiliary data files
15622 to @var{directory}.
15624 @kindex show data-directory
15625 @item show data-directory
15626 Show the directory @value{GDBN} searches for auxiliary data files.
15629 @cindex default data directory
15630 @cindex @samp{--with-gdb-datadir}
15631 You can set the default data directory by using the configure-time
15632 @samp{--with-gdb-datadir} option. If the data directory is inside
15633 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
15634 @samp{--exec-prefix}), then the default data directory will be updated
15635 automatically if the installed @value{GDBN} is moved to a new
15638 The data directory may also be specified with the
15639 @code{--data-directory} command line option.
15640 @xref{Mode Options}.
15643 @chapter Specifying a Debugging Target
15645 @cindex debugging target
15646 A @dfn{target} is the execution environment occupied by your program.
15648 Often, @value{GDBN} runs in the same host environment as your program;
15649 in that case, the debugging target is specified as a side effect when
15650 you use the @code{file} or @code{core} commands. When you need more
15651 flexibility---for example, running @value{GDBN} on a physically separate
15652 host, or controlling a standalone system over a serial port or a
15653 realtime system over a TCP/IP connection---you can use the @code{target}
15654 command to specify one of the target types configured for @value{GDBN}
15655 (@pxref{Target Commands, ,Commands for Managing Targets}).
15657 @cindex target architecture
15658 It is possible to build @value{GDBN} for several different @dfn{target
15659 architectures}. When @value{GDBN} is built like that, you can choose
15660 one of the available architectures with the @kbd{set architecture}
15664 @kindex set architecture
15665 @kindex show architecture
15666 @item set architecture @var{arch}
15667 This command sets the current target architecture to @var{arch}. The
15668 value of @var{arch} can be @code{"auto"}, in addition to one of the
15669 supported architectures.
15671 @item show architecture
15672 Show the current target architecture.
15674 @item set processor
15676 @kindex set processor
15677 @kindex show processor
15678 These are alias commands for, respectively, @code{set architecture}
15679 and @code{show architecture}.
15683 * Active Targets:: Active targets
15684 * Target Commands:: Commands for managing targets
15685 * Byte Order:: Choosing target byte order
15688 @node Active Targets
15689 @section Active Targets
15691 @cindex stacking targets
15692 @cindex active targets
15693 @cindex multiple targets
15695 There are multiple classes of targets such as: processes, executable files or
15696 recording sessions. Core files belong to the process class, making core file
15697 and process mutually exclusive. Otherwise, @value{GDBN} can work concurrently
15698 on multiple active targets, one in each class. This allows you to (for
15699 example) start a process and inspect its activity, while still having access to
15700 the executable file after the process finishes. Or if you start process
15701 recording (@pxref{Reverse Execution}) and @code{reverse-step} there, you are
15702 presented a virtual layer of the recording target, while the process target
15703 remains stopped at the chronologically last point of the process execution.
15705 Use the @code{core-file} and @code{exec-file} commands to select a new core
15706 file or executable target (@pxref{Files, ,Commands to Specify Files}). To
15707 specify as a target a process that is already running, use the @code{attach}
15708 command (@pxref{Attach, ,Debugging an Already-running Process}).
15710 @node Target Commands
15711 @section Commands for Managing Targets
15714 @item target @var{type} @var{parameters}
15715 Connects the @value{GDBN} host environment to a target machine or
15716 process. A target is typically a protocol for talking to debugging
15717 facilities. You use the argument @var{type} to specify the type or
15718 protocol of the target machine.
15720 Further @var{parameters} are interpreted by the target protocol, but
15721 typically include things like device names or host names to connect
15722 with, process numbers, and baud rates.
15724 The @code{target} command does not repeat if you press @key{RET} again
15725 after executing the command.
15727 @kindex help target
15729 Displays the names of all targets available. To display targets
15730 currently selected, use either @code{info target} or @code{info files}
15731 (@pxref{Files, ,Commands to Specify Files}).
15733 @item help target @var{name}
15734 Describe a particular target, including any parameters necessary to
15737 @kindex set gnutarget
15738 @item set gnutarget @var{args}
15739 @value{GDBN} uses its own library BFD to read your files. @value{GDBN}
15740 knows whether it is reading an @dfn{executable},
15741 a @dfn{core}, or a @dfn{.o} file; however, you can specify the file format
15742 with the @code{set gnutarget} command. Unlike most @code{target} commands,
15743 with @code{gnutarget} the @code{target} refers to a program, not a machine.
15746 @emph{Warning:} To specify a file format with @code{set gnutarget},
15747 you must know the actual BFD name.
15751 @xref{Files, , Commands to Specify Files}.
15753 @kindex show gnutarget
15754 @item show gnutarget
15755 Use the @code{show gnutarget} command to display what file format
15756 @code{gnutarget} is set to read. If you have not set @code{gnutarget},
15757 @value{GDBN} will determine the file format for each file automatically,
15758 and @code{show gnutarget} displays @samp{The current BDF target is "auto"}.
15761 @cindex common targets
15762 Here are some common targets (available, or not, depending on the GDB
15767 @item target exec @var{program}
15768 @cindex executable file target
15769 An executable file. @samp{target exec @var{program}} is the same as
15770 @samp{exec-file @var{program}}.
15772 @item target core @var{filename}
15773 @cindex core dump file target
15774 A core dump file. @samp{target core @var{filename}} is the same as
15775 @samp{core-file @var{filename}}.
15777 @item target remote @var{medium}
15778 @cindex remote target
15779 A remote system connected to @value{GDBN} via a serial line or network
15780 connection. This command tells @value{GDBN} to use its own remote
15781 protocol over @var{medium} for debugging. @xref{Remote Debugging}.
15783 For example, if you have a board connected to @file{/dev/ttya} on the
15784 machine running @value{GDBN}, you could say:
15787 target remote /dev/ttya
15790 @code{target remote} supports the @code{load} command. This is only
15791 useful if you have some other way of getting the stub to the target
15792 system, and you can put it somewhere in memory where it won't get
15793 clobbered by the download.
15795 @item target sim @r{[}@var{simargs}@r{]} @dots{}
15796 @cindex built-in simulator target
15797 Builtin CPU simulator. @value{GDBN} includes simulators for most architectures.
15805 works; however, you cannot assume that a specific memory map, device
15806 drivers, or even basic I/O is available, although some simulators do
15807 provide these. For info about any processor-specific simulator details,
15808 see the appropriate section in @ref{Embedded Processors, ,Embedded
15813 Some configurations may include these targets as well:
15817 @item target nrom @var{dev}
15818 @cindex NetROM ROM emulator target
15819 NetROM ROM emulator. This target only supports downloading.
15823 Different targets are available on different configurations of @value{GDBN};
15824 your configuration may have more or fewer targets.
15826 Many remote targets require you to download the executable's code once
15827 you've successfully established a connection. You may wish to control
15828 various aspects of this process.
15833 @kindex set hash@r{, for remote monitors}
15834 @cindex hash mark while downloading
15835 This command controls whether a hash mark @samp{#} is displayed while
15836 downloading a file to the remote monitor. If on, a hash mark is
15837 displayed after each S-record is successfully downloaded to the
15841 @kindex show hash@r{, for remote monitors}
15842 Show the current status of displaying the hash mark.
15844 @item set debug monitor
15845 @kindex set debug monitor
15846 @cindex display remote monitor communications
15847 Enable or disable display of communications messages between
15848 @value{GDBN} and the remote monitor.
15850 @item show debug monitor
15851 @kindex show debug monitor
15852 Show the current status of displaying communications between
15853 @value{GDBN} and the remote monitor.
15858 @kindex load @var{filename}
15859 @item load @var{filename}
15861 Depending on what remote debugging facilities are configured into
15862 @value{GDBN}, the @code{load} command may be available. Where it exists, it
15863 is meant to make @var{filename} (an executable) available for debugging
15864 on the remote system---by downloading, or dynamic linking, for example.
15865 @code{load} also records the @var{filename} symbol table in @value{GDBN}, like
15866 the @code{add-symbol-file} command.
15868 If your @value{GDBN} does not have a @code{load} command, attempting to
15869 execute it gets the error message ``@code{You can't do that when your
15870 target is @dots{}}''
15872 The file is loaded at whatever address is specified in the executable.
15873 For some object file formats, you can specify the load address when you
15874 link the program; for other formats, like a.out, the object file format
15875 specifies a fixed address.
15876 @c FIXME! This would be a good place for an xref to the GNU linker doc.
15878 Depending on the remote side capabilities, @value{GDBN} may be able to
15879 load programs into flash memory.
15881 @code{load} does not repeat if you press @key{RET} again after using it.
15885 @section Choosing Target Byte Order
15887 @cindex choosing target byte order
15888 @cindex target byte order
15890 Some types of processors, such as the MIPS, PowerPC, and Renesas SH,
15891 offer the ability to run either big-endian or little-endian byte
15892 orders. Usually the executable or symbol will include a bit to
15893 designate the endian-ness, and you will not need to worry about
15894 which to use. However, you may still find it useful to adjust
15895 @value{GDBN}'s idea of processor endian-ness manually.
15899 @item set endian big
15900 Instruct @value{GDBN} to assume the target is big-endian.
15902 @item set endian little
15903 Instruct @value{GDBN} to assume the target is little-endian.
15905 @item set endian auto
15906 Instruct @value{GDBN} to use the byte order associated with the
15910 Display @value{GDBN}'s current idea of the target byte order.
15914 Note that these commands merely adjust interpretation of symbolic
15915 data on the host, and that they have absolutely no effect on the
15919 @node Remote Debugging
15920 @chapter Debugging Remote Programs
15921 @cindex remote debugging
15923 If you are trying to debug a program running on a machine that cannot run
15924 @value{GDBN} in the usual way, it is often useful to use remote debugging.
15925 For example, you might use remote debugging on an operating system kernel,
15926 or on a small system which does not have a general purpose operating system
15927 powerful enough to run a full-featured debugger.
15929 Some configurations of @value{GDBN} have special serial or TCP/IP interfaces
15930 to make this work with particular debugging targets. In addition,
15931 @value{GDBN} comes with a generic serial protocol (specific to @value{GDBN},
15932 but not specific to any particular target system) which you can use if you
15933 write the remote stubs---the code that runs on the remote system to
15934 communicate with @value{GDBN}.
15936 Other remote targets may be available in your
15937 configuration of @value{GDBN}; use @code{help target} to list them.
15940 * Connecting:: Connecting to a remote target
15941 * File Transfer:: Sending files to a remote system
15942 * Server:: Using the gdbserver program
15943 * Remote Configuration:: Remote configuration
15944 * Remote Stub:: Implementing a remote stub
15948 @section Connecting to a Remote Target
15950 On the @value{GDBN} host machine, you will need an unstripped copy of
15951 your program, since @value{GDBN} needs symbol and debugging information.
15952 Start up @value{GDBN} as usual, using the name of the local copy of your
15953 program as the first argument.
15955 @cindex @code{target remote}
15956 @value{GDBN} can communicate with the target over a serial line, or
15957 over an @acronym{IP} network using @acronym{TCP} or @acronym{UDP}. In
15958 each case, @value{GDBN} uses the same protocol for debugging your
15959 program; only the medium carrying the debugging packets varies. The
15960 @code{target remote} command establishes a connection to the target.
15961 Its arguments indicate which medium to use:
15965 @item target remote @var{serial-device}
15966 @cindex serial line, @code{target remote}
15967 Use @var{serial-device} to communicate with the target. For example,
15968 to use a serial line connected to the device named @file{/dev/ttyb}:
15971 target remote /dev/ttyb
15974 If you're using a serial line, you may want to give @value{GDBN} the
15975 @w{@samp{--baud}} option, or use the @code{set remotebaud} command
15976 (@pxref{Remote Configuration, set remotebaud}) before the
15977 @code{target} command.
15979 @item target remote @code{@var{host}:@var{port}}
15980 @itemx target remote @code{tcp:@var{host}:@var{port}}
15981 @cindex @acronym{TCP} port, @code{target remote}
15982 Debug using a @acronym{TCP} connection to @var{port} on @var{host}.
15983 The @var{host} may be either a host name or a numeric @acronym{IP}
15984 address; @var{port} must be a decimal number. The @var{host} could be
15985 the target machine itself, if it is directly connected to the net, or
15986 it might be a terminal server which in turn has a serial line to the
15989 For example, to connect to port 2828 on a terminal server named
15993 target remote manyfarms:2828
15996 If your remote target is actually running on the same machine as your
15997 debugger session (e.g.@: a simulator for your target running on the
15998 same host), you can omit the hostname. For example, to connect to
15999 port 1234 on your local machine:
16002 target remote :1234
16006 Note that the colon is still required here.
16008 @item target remote @code{udp:@var{host}:@var{port}}
16009 @cindex @acronym{UDP} port, @code{target remote}
16010 Debug using @acronym{UDP} packets to @var{port} on @var{host}. For example, to
16011 connect to @acronym{UDP} port 2828 on a terminal server named @code{manyfarms}:
16014 target remote udp:manyfarms:2828
16017 When using a @acronym{UDP} connection for remote debugging, you should
16018 keep in mind that the `U' stands for ``Unreliable''. @acronym{UDP}
16019 can silently drop packets on busy or unreliable networks, which will
16020 cause havoc with your debugging session.
16022 @item target remote | @var{command}
16023 @cindex pipe, @code{target remote} to
16024 Run @var{command} in the background and communicate with it using a
16025 pipe. The @var{command} is a shell command, to be parsed and expanded
16026 by the system's command shell, @code{/bin/sh}; it should expect remote
16027 protocol packets on its standard input, and send replies on its
16028 standard output. You could use this to run a stand-alone simulator
16029 that speaks the remote debugging protocol, to make net connections
16030 using programs like @code{ssh}, or for other similar tricks.
16032 If @var{command} closes its standard output (perhaps by exiting),
16033 @value{GDBN} will try to send it a @code{SIGTERM} signal. (If the
16034 program has already exited, this will have no effect.)
16038 Once the connection has been established, you can use all the usual
16039 commands to examine and change data. The remote program is already
16040 running; you can use @kbd{step} and @kbd{continue}, and you do not
16041 need to use @kbd{run}.
16043 @cindex interrupting remote programs
16044 @cindex remote programs, interrupting
16045 Whenever @value{GDBN} is waiting for the remote program, if you type the
16046 interrupt character (often @kbd{Ctrl-c}), @value{GDBN} attempts to stop the
16047 program. This may or may not succeed, depending in part on the hardware
16048 and the serial drivers the remote system uses. If you type the
16049 interrupt character once again, @value{GDBN} displays this prompt:
16052 Interrupted while waiting for the program.
16053 Give up (and stop debugging it)? (y or n)
16056 If you type @kbd{y}, @value{GDBN} abandons the remote debugging session.
16057 (If you decide you want to try again later, you can use @samp{target
16058 remote} again to connect once more.) If you type @kbd{n}, @value{GDBN}
16059 goes back to waiting.
16062 @kindex detach (remote)
16064 When you have finished debugging the remote program, you can use the
16065 @code{detach} command to release it from @value{GDBN} control.
16066 Detaching from the target normally resumes its execution, but the results
16067 will depend on your particular remote stub. After the @code{detach}
16068 command, @value{GDBN} is free to connect to another target.
16072 The @code{disconnect} command behaves like @code{detach}, except that
16073 the target is generally not resumed. It will wait for @value{GDBN}
16074 (this instance or another one) to connect and continue debugging. After
16075 the @code{disconnect} command, @value{GDBN} is again free to connect to
16078 @cindex send command to remote monitor
16079 @cindex extend @value{GDBN} for remote targets
16080 @cindex add new commands for external monitor
16082 @item monitor @var{cmd}
16083 This command allows you to send arbitrary commands directly to the
16084 remote monitor. Since @value{GDBN} doesn't care about the commands it
16085 sends like this, this command is the way to extend @value{GDBN}---you
16086 can add new commands that only the external monitor will understand
16090 @node File Transfer
16091 @section Sending files to a remote system
16092 @cindex remote target, file transfer
16093 @cindex file transfer
16094 @cindex sending files to remote systems
16096 Some remote targets offer the ability to transfer files over the same
16097 connection used to communicate with @value{GDBN}. This is convenient
16098 for targets accessible through other means, e.g.@: @sc{gnu}/Linux systems
16099 running @code{gdbserver} over a network interface. For other targets,
16100 e.g.@: embedded devices with only a single serial port, this may be
16101 the only way to upload or download files.
16103 Not all remote targets support these commands.
16107 @item remote put @var{hostfile} @var{targetfile}
16108 Copy file @var{hostfile} from the host system (the machine running
16109 @value{GDBN}) to @var{targetfile} on the target system.
16112 @item remote get @var{targetfile} @var{hostfile}
16113 Copy file @var{targetfile} from the target system to @var{hostfile}
16114 on the host system.
16116 @kindex remote delete
16117 @item remote delete @var{targetfile}
16118 Delete @var{targetfile} from the target system.
16123 @section Using the @code{gdbserver} Program
16126 @cindex remote connection without stubs
16127 @code{gdbserver} is a control program for Unix-like systems, which
16128 allows you to connect your program with a remote @value{GDBN} via
16129 @code{target remote}---but without linking in the usual debugging stub.
16131 @code{gdbserver} is not a complete replacement for the debugging stubs,
16132 because it requires essentially the same operating-system facilities
16133 that @value{GDBN} itself does. In fact, a system that can run
16134 @code{gdbserver} to connect to a remote @value{GDBN} could also run
16135 @value{GDBN} locally! @code{gdbserver} is sometimes useful nevertheless,
16136 because it is a much smaller program than @value{GDBN} itself. It is
16137 also easier to port than all of @value{GDBN}, so you may be able to get
16138 started more quickly on a new system by using @code{gdbserver}.
16139 Finally, if you develop code for real-time systems, you may find that
16140 the tradeoffs involved in real-time operation make it more convenient to
16141 do as much development work as possible on another system, for example
16142 by cross-compiling. You can use @code{gdbserver} to make a similar
16143 choice for debugging.
16145 @value{GDBN} and @code{gdbserver} communicate via either a serial line
16146 or a TCP connection, using the standard @value{GDBN} remote serial
16150 @emph{Warning:} @code{gdbserver} does not have any built-in security.
16151 Do not run @code{gdbserver} connected to any public network; a
16152 @value{GDBN} connection to @code{gdbserver} provides access to the
16153 target system with the same privileges as the user running
16157 @subsection Running @code{gdbserver}
16158 @cindex arguments, to @code{gdbserver}
16159 @cindex @code{gdbserver}, command-line arguments
16161 Run @code{gdbserver} on the target system. You need a copy of the
16162 program you want to debug, including any libraries it requires.
16163 @code{gdbserver} does not need your program's symbol table, so you can
16164 strip the program if necessary to save space. @value{GDBN} on the host
16165 system does all the symbol handling.
16167 To use the server, you must tell it how to communicate with @value{GDBN};
16168 the name of your program; and the arguments for your program. The usual
16172 target> gdbserver @var{comm} @var{program} [ @var{args} @dots{} ]
16175 @var{comm} is either a device name (to use a serial line) or a TCP
16176 hostname and portnumber. For example, to debug Emacs with the argument
16177 @samp{foo.txt} and communicate with @value{GDBN} over the serial port
16181 target> gdbserver /dev/com1 emacs foo.txt
16184 @code{gdbserver} waits passively for the host @value{GDBN} to communicate
16187 To use a TCP connection instead of a serial line:
16190 target> gdbserver host:2345 emacs foo.txt
16193 The only difference from the previous example is the first argument,
16194 specifying that you are communicating with the host @value{GDBN} via
16195 TCP. The @samp{host:2345} argument means that @code{gdbserver} is to
16196 expect a TCP connection from machine @samp{host} to local TCP port 2345.
16197 (Currently, the @samp{host} part is ignored.) You can choose any number
16198 you want for the port number as long as it does not conflict with any
16199 TCP ports already in use on the target system (for example, @code{23} is
16200 reserved for @code{telnet}).@footnote{If you choose a port number that
16201 conflicts with another service, @code{gdbserver} prints an error message
16202 and exits.} You must use the same port number with the host @value{GDBN}
16203 @code{target remote} command.
16205 @subsubsection Attaching to a Running Program
16206 @cindex attach to a program, @code{gdbserver}
16207 @cindex @option{--attach}, @code{gdbserver} option
16209 On some targets, @code{gdbserver} can also attach to running programs.
16210 This is accomplished via the @code{--attach} argument. The syntax is:
16213 target> gdbserver --attach @var{comm} @var{pid}
16216 @var{pid} is the process ID of a currently running process. It isn't necessary
16217 to point @code{gdbserver} at a binary for the running process.
16220 You can debug processes by name instead of process ID if your target has the
16221 @code{pidof} utility:
16224 target> gdbserver --attach @var{comm} `pidof @var{program}`
16227 In case more than one copy of @var{program} is running, or @var{program}
16228 has multiple threads, most versions of @code{pidof} support the
16229 @code{-s} option to only return the first process ID.
16231 @subsubsection Multi-Process Mode for @code{gdbserver}
16232 @cindex @code{gdbserver}, multiple processes
16233 @cindex multiple processes with @code{gdbserver}
16235 When you connect to @code{gdbserver} using @code{target remote},
16236 @code{gdbserver} debugs the specified program only once. When the
16237 program exits, or you detach from it, @value{GDBN} closes the connection
16238 and @code{gdbserver} exits.
16240 If you connect using @kbd{target extended-remote}, @code{gdbserver}
16241 enters multi-process mode. When the debugged program exits, or you
16242 detach from it, @value{GDBN} stays connected to @code{gdbserver} even
16243 though no program is running. The @code{run} and @code{attach}
16244 commands instruct @code{gdbserver} to run or attach to a new program.
16245 The @code{run} command uses @code{set remote exec-file} (@pxref{set
16246 remote exec-file}) to select the program to run. Command line
16247 arguments are supported, except for wildcard expansion and I/O
16248 redirection (@pxref{Arguments}).
16250 @cindex @option{--multi}, @code{gdbserver} option
16251 To start @code{gdbserver} without supplying an initial command to run
16252 or process ID to attach, use the @option{--multi} command line option.
16253 Then you can connect using @kbd{target extended-remote} and start
16254 the program you want to debug.
16256 In multi-process mode @code{gdbserver} does not automatically exit unless you
16257 use the option @option{--once}. You can terminate it by using
16258 @code{monitor exit} (@pxref{Monitor Commands for gdbserver}). Note that the
16259 conditions under which @code{gdbserver} terminates depend on how @value{GDBN}
16260 connects to it (@kbd{target remote} or @kbd{target extended-remote}). The
16261 @option{--multi} option to @code{gdbserver} has no influence on that.
16263 @subsubsection TCP port allocation lifecycle of @code{gdbserver}
16265 This section applies only when @code{gdbserver} is run to listen on a TCP port.
16267 @code{gdbserver} normally terminates after all of its debugged processes have
16268 terminated in @kbd{target remote} mode. On the other hand, for @kbd{target
16269 extended-remote}, @code{gdbserver} stays running even with no processes left.
16270 @value{GDBN} normally terminates the spawned debugged process on its exit,
16271 which normally also terminates @code{gdbserver} in the @kbd{target remote}
16272 mode. Therefore, when the connection drops unexpectedly, and @value{GDBN}
16273 cannot ask @code{gdbserver} to kill its debugged processes, @code{gdbserver}
16274 stays running even in the @kbd{target remote} mode.
16276 When @code{gdbserver} stays running, @value{GDBN} can connect to it again later.
16277 Such reconnecting is useful for features like @ref{disconnected tracing}. For
16278 completeness, at most one @value{GDBN} can be connected at a time.
16280 @cindex @option{--once}, @code{gdbserver} option
16281 By default, @code{gdbserver} keeps the listening TCP port open, so that
16282 additional connections are possible. However, if you start @code{gdbserver}
16283 with the @option{--once} option, it will stop listening for any further
16284 connection attempts after connecting to the first @value{GDBN} session. This
16285 means no further connections to @code{gdbserver} will be possible after the
16286 first one. It also means @code{gdbserver} will terminate after the first
16287 connection with remote @value{GDBN} has closed, even for unexpectedly closed
16288 connections and even in the @kbd{target extended-remote} mode. The
16289 @option{--once} option allows reusing the same port number for connecting to
16290 multiple instances of @code{gdbserver} running on the same host, since each
16291 instance closes its port after the first connection.
16293 @subsubsection Other Command-Line Arguments for @code{gdbserver}
16295 @cindex @option{--debug}, @code{gdbserver} option
16296 The @option{--debug} option tells @code{gdbserver} to display extra
16297 status information about the debugging process.
16298 @cindex @option{--remote-debug}, @code{gdbserver} option
16299 The @option{--remote-debug} option tells @code{gdbserver} to display
16300 remote protocol debug output. These options are intended for
16301 @code{gdbserver} development and for bug reports to the developers.
16303 @cindex @option{--wrapper}, @code{gdbserver} option
16304 The @option{--wrapper} option specifies a wrapper to launch programs
16305 for debugging. The option should be followed by the name of the
16306 wrapper, then any command-line arguments to pass to the wrapper, then
16307 @kbd{--} indicating the end of the wrapper arguments.
16309 @code{gdbserver} runs the specified wrapper program with a combined
16310 command line including the wrapper arguments, then the name of the
16311 program to debug, then any arguments to the program. The wrapper
16312 runs until it executes your program, and then @value{GDBN} gains control.
16314 You can use any program that eventually calls @code{execve} with
16315 its arguments as a wrapper. Several standard Unix utilities do
16316 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
16317 with @code{exec "$@@"} will also work.
16319 For example, you can use @code{env} to pass an environment variable to
16320 the debugged program, without setting the variable in @code{gdbserver}'s
16324 $ gdbserver --wrapper env LD_PRELOAD=libtest.so -- :2222 ./testprog
16327 @subsection Connecting to @code{gdbserver}
16329 Run @value{GDBN} on the host system.
16331 First make sure you have the necessary symbol files. Load symbols for
16332 your application using the @code{file} command before you connect. Use
16333 @code{set sysroot} to locate target libraries (unless your @value{GDBN}
16334 was compiled with the correct sysroot using @code{--with-sysroot}).
16336 The symbol file and target libraries must exactly match the executable
16337 and libraries on the target, with one exception: the files on the host
16338 system should not be stripped, even if the files on the target system
16339 are. Mismatched or missing files will lead to confusing results
16340 during debugging. On @sc{gnu}/Linux targets, mismatched or missing
16341 files may also prevent @code{gdbserver} from debugging multi-threaded
16344 Connect to your target (@pxref{Connecting,,Connecting to a Remote Target}).
16345 For TCP connections, you must start up @code{gdbserver} prior to using
16346 the @code{target remote} command. Otherwise you may get an error whose
16347 text depends on the host system, but which usually looks something like
16348 @samp{Connection refused}. Don't use the @code{load}
16349 command in @value{GDBN} when using @code{gdbserver}, since the program is
16350 already on the target.
16352 @subsection Monitor Commands for @code{gdbserver}
16353 @cindex monitor commands, for @code{gdbserver}
16354 @anchor{Monitor Commands for gdbserver}
16356 During a @value{GDBN} session using @code{gdbserver}, you can use the
16357 @code{monitor} command to send special requests to @code{gdbserver}.
16358 Here are the available commands.
16362 List the available monitor commands.
16364 @item monitor set debug 0
16365 @itemx monitor set debug 1
16366 Disable or enable general debugging messages.
16368 @item monitor set remote-debug 0
16369 @itemx monitor set remote-debug 1
16370 Disable or enable specific debugging messages associated with the remote
16371 protocol (@pxref{Remote Protocol}).
16373 @item monitor set libthread-db-search-path [PATH]
16374 @cindex gdbserver, search path for @code{libthread_db}
16375 When this command is issued, @var{path} is a colon-separated list of
16376 directories to search for @code{libthread_db} (@pxref{Threads,,set
16377 libthread-db-search-path}). If you omit @var{path},
16378 @samp{libthread-db-search-path} will be reset to its default value.
16381 Tell gdbserver to exit immediately. This command should be followed by
16382 @code{disconnect} to close the debugging session. @code{gdbserver} will
16383 detach from any attached processes and kill any processes it created.
16384 Use @code{monitor exit} to terminate @code{gdbserver} at the end
16385 of a multi-process mode debug session.
16389 @subsection Tracepoints support in @code{gdbserver}
16390 @cindex tracepoints support in @code{gdbserver}
16392 On some targets, @code{gdbserver} supports tracepoints, fast
16393 tracepoints and static tracepoints.
16395 For fast or static tracepoints to work, a special library called the
16396 @dfn{in-process agent} (IPA), must be loaded in the inferior process.
16397 This library is built and distributed as an integral part of
16398 @code{gdbserver}. In addition, support for static tracepoints
16399 requires building the in-process agent library with static tracepoints
16400 support. At present, the UST (LTTng Userspace Tracer,
16401 @url{http://lttng.org/ust}) tracing engine is supported. This support
16402 is automatically available if UST development headers are found in the
16403 standard include path when @code{gdbserver} is built, or if
16404 @code{gdbserver} was explicitly configured using @option{--with-ust}
16405 to point at such headers. You can explicitly disable the support
16406 using @option{--with-ust=no}.
16408 There are several ways to load the in-process agent in your program:
16411 @item Specifying it as dependency at link time
16413 You can link your program dynamically with the in-process agent
16414 library. On most systems, this is accomplished by adding
16415 @code{-linproctrace} to the link command.
16417 @item Using the system's preloading mechanisms
16419 You can force loading the in-process agent at startup time by using
16420 your system's support for preloading shared libraries. Many Unixes
16421 support the concept of preloading user defined libraries. In most
16422 cases, you do that by specifying @code{LD_PRELOAD=libinproctrace.so}
16423 in the environment. See also the description of @code{gdbserver}'s
16424 @option{--wrapper} command line option.
16426 @item Using @value{GDBN} to force loading the agent at run time
16428 On some systems, you can force the inferior to load a shared library,
16429 by calling a dynamic loader function in the inferior that takes care
16430 of dynamically looking up and loading a shared library. On most Unix
16431 systems, the function is @code{dlopen}. You'll use the @code{call}
16432 command for that. For example:
16435 (@value{GDBP}) call dlopen ("libinproctrace.so", ...)
16438 Note that on most Unix systems, for the @code{dlopen} function to be
16439 available, the program needs to be linked with @code{-ldl}.
16442 On systems that have a userspace dynamic loader, like most Unix
16443 systems, when you connect to @code{gdbserver} using @code{target
16444 remote}, you'll find that the program is stopped at the dynamic
16445 loader's entry point, and no shared library has been loaded in the
16446 program's address space yet, including the in-process agent. In that
16447 case, before being able to use any of the fast or static tracepoints
16448 features, you need to let the loader run and load the shared
16449 libraries. The simplest way to do that is to run the program to the
16450 main procedure. E.g., if debugging a C or C@t{++} program, start
16451 @code{gdbserver} like so:
16454 $ gdbserver :9999 myprogram
16457 Start GDB and connect to @code{gdbserver} like so, and run to main:
16461 (@value{GDBP}) target remote myhost:9999
16462 0x00007f215893ba60 in ?? () from /lib64/ld-linux-x86-64.so.2
16463 (@value{GDBP}) b main
16464 (@value{GDBP}) continue
16467 The in-process tracing agent library should now be loaded into the
16468 process; you can confirm it with the @code{info sharedlibrary}
16469 command, which will list @file{libinproctrace.so} as loaded in the
16470 process. You are now ready to install fast tracepoints, list static
16471 tracepoint markers, probe static tracepoints markers, and start
16474 @node Remote Configuration
16475 @section Remote Configuration
16478 @kindex show remote
16479 This section documents the configuration options available when
16480 debugging remote programs. For the options related to the File I/O
16481 extensions of the remote protocol, see @ref{system,
16482 system-call-allowed}.
16485 @item set remoteaddresssize @var{bits}
16486 @cindex address size for remote targets
16487 @cindex bits in remote address
16488 Set the maximum size of address in a memory packet to the specified
16489 number of bits. @value{GDBN} will mask off the address bits above
16490 that number, when it passes addresses to the remote target. The
16491 default value is the number of bits in the target's address.
16493 @item show remoteaddresssize
16494 Show the current value of remote address size in bits.
16496 @item set remotebaud @var{n}
16497 @cindex baud rate for remote targets
16498 Set the baud rate for the remote serial I/O to @var{n} baud. The
16499 value is used to set the speed of the serial port used for debugging
16502 @item show remotebaud
16503 Show the current speed of the remote connection.
16505 @item set remotebreak
16506 @cindex interrupt remote programs
16507 @cindex BREAK signal instead of Ctrl-C
16508 @anchor{set remotebreak}
16509 If set to on, @value{GDBN} sends a @code{BREAK} signal to the remote
16510 when you type @kbd{Ctrl-c} to interrupt the program running
16511 on the remote. If set to off, @value{GDBN} sends the @samp{Ctrl-C}
16512 character instead. The default is off, since most remote systems
16513 expect to see @samp{Ctrl-C} as the interrupt signal.
16515 @item show remotebreak
16516 Show whether @value{GDBN} sends @code{BREAK} or @samp{Ctrl-C} to
16517 interrupt the remote program.
16519 @item set remoteflow on
16520 @itemx set remoteflow off
16521 @kindex set remoteflow
16522 Enable or disable hardware flow control (@code{RTS}/@code{CTS})
16523 on the serial port used to communicate to the remote target.
16525 @item show remoteflow
16526 @kindex show remoteflow
16527 Show the current setting of hardware flow control.
16529 @item set remotelogbase @var{base}
16530 Set the base (a.k.a.@: radix) of logging serial protocol
16531 communications to @var{base}. Supported values of @var{base} are:
16532 @code{ascii}, @code{octal}, and @code{hex}. The default is
16535 @item show remotelogbase
16536 Show the current setting of the radix for logging remote serial
16539 @item set remotelogfile @var{file}
16540 @cindex record serial communications on file
16541 Record remote serial communications on the named @var{file}. The
16542 default is not to record at all.
16544 @item show remotelogfile.
16545 Show the current setting of the file name on which to record the
16546 serial communications.
16548 @item set remotetimeout @var{num}
16549 @cindex timeout for serial communications
16550 @cindex remote timeout
16551 Set the timeout limit to wait for the remote target to respond to
16552 @var{num} seconds. The default is 2 seconds.
16554 @item show remotetimeout
16555 Show the current number of seconds to wait for the remote target
16558 @cindex limit hardware breakpoints and watchpoints
16559 @cindex remote target, limit break- and watchpoints
16560 @anchor{set remote hardware-watchpoint-limit}
16561 @anchor{set remote hardware-breakpoint-limit}
16562 @item set remote hardware-watchpoint-limit @var{limit}
16563 @itemx set remote hardware-breakpoint-limit @var{limit}
16564 Restrict @value{GDBN} to using @var{limit} remote hardware breakpoint or
16565 watchpoints. A limit of -1, the default, is treated as unlimited.
16567 @item set remote exec-file @var{filename}
16568 @itemx show remote exec-file
16569 @anchor{set remote exec-file}
16570 @cindex executable file, for remote target
16571 Select the file used for @code{run} with @code{target
16572 extended-remote}. This should be set to a filename valid on the
16573 target system. If it is not set, the target will use a default
16574 filename (e.g.@: the last program run).
16576 @item set remote interrupt-sequence
16577 @cindex interrupt remote programs
16578 @cindex select Ctrl-C, BREAK or BREAK-g
16579 Allow the user to select one of @samp{Ctrl-C}, a @code{BREAK} or
16580 @samp{BREAK-g} as the
16581 sequence to the remote target in order to interrupt the execution.
16582 @samp{Ctrl-C} is a default. Some system prefers @code{BREAK} which
16583 is high level of serial line for some certain time.
16584 Linux kernel prefers @samp{BREAK-g}, a.k.a Magic SysRq g.
16585 It is @code{BREAK} signal followed by character @code{g}.
16587 @item show interrupt-sequence
16588 Show which of @samp{Ctrl-C}, @code{BREAK} or @code{BREAK-g}
16589 is sent by @value{GDBN} to interrupt the remote program.
16590 @code{BREAK-g} is BREAK signal followed by @code{g} and
16591 also known as Magic SysRq g.
16593 @item set remote interrupt-on-connect
16594 @cindex send interrupt-sequence on start
16595 Specify whether interrupt-sequence is sent to remote target when
16596 @value{GDBN} connects to it. This is mostly needed when you debug
16597 Linux kernel. Linux kernel expects @code{BREAK} followed by @code{g}
16598 which is known as Magic SysRq g in order to connect @value{GDBN}.
16600 @item show interrupt-on-connect
16601 Show whether interrupt-sequence is sent
16602 to remote target when @value{GDBN} connects to it.
16606 @item set tcp auto-retry on
16607 @cindex auto-retry, for remote TCP target
16608 Enable auto-retry for remote TCP connections. This is useful if the remote
16609 debugging agent is launched in parallel with @value{GDBN}; there is a race
16610 condition because the agent may not become ready to accept the connection
16611 before @value{GDBN} attempts to connect. When auto-retry is
16612 enabled, if the initial attempt to connect fails, @value{GDBN} reattempts
16613 to establish the connection using the timeout specified by
16614 @code{set tcp connect-timeout}.
16616 @item set tcp auto-retry off
16617 Do not auto-retry failed TCP connections.
16619 @item show tcp auto-retry
16620 Show the current auto-retry setting.
16622 @item set tcp connect-timeout @var{seconds}
16623 @cindex connection timeout, for remote TCP target
16624 @cindex timeout, for remote target connection
16625 Set the timeout for establishing a TCP connection to the remote target to
16626 @var{seconds}. The timeout affects both polling to retry failed connections
16627 (enabled by @code{set tcp auto-retry on}) and waiting for connections
16628 that are merely slow to complete, and represents an approximate cumulative
16631 @item show tcp connect-timeout
16632 Show the current connection timeout setting.
16635 @cindex remote packets, enabling and disabling
16636 The @value{GDBN} remote protocol autodetects the packets supported by
16637 your debugging stub. If you need to override the autodetection, you
16638 can use these commands to enable or disable individual packets. Each
16639 packet can be set to @samp{on} (the remote target supports this
16640 packet), @samp{off} (the remote target does not support this packet),
16641 or @samp{auto} (detect remote target support for this packet). They
16642 all default to @samp{auto}. For more information about each packet,
16643 see @ref{Remote Protocol}.
16645 During normal use, you should not have to use any of these commands.
16646 If you do, that may be a bug in your remote debugging stub, or a bug
16647 in @value{GDBN}. You may want to report the problem to the
16648 @value{GDBN} developers.
16650 For each packet @var{name}, the command to enable or disable the
16651 packet is @code{set remote @var{name}-packet}. The available settings
16654 @multitable @columnfractions 0.28 0.32 0.25
16657 @tab Related Features
16659 @item @code{fetch-register}
16661 @tab @code{info registers}
16663 @item @code{set-register}
16667 @item @code{binary-download}
16669 @tab @code{load}, @code{set}
16671 @item @code{read-aux-vector}
16672 @tab @code{qXfer:auxv:read}
16673 @tab @code{info auxv}
16675 @item @code{symbol-lookup}
16676 @tab @code{qSymbol}
16677 @tab Detecting multiple threads
16679 @item @code{attach}
16680 @tab @code{vAttach}
16683 @item @code{verbose-resume}
16685 @tab Stepping or resuming multiple threads
16691 @item @code{software-breakpoint}
16695 @item @code{hardware-breakpoint}
16699 @item @code{write-watchpoint}
16703 @item @code{read-watchpoint}
16707 @item @code{access-watchpoint}
16711 @item @code{target-features}
16712 @tab @code{qXfer:features:read}
16713 @tab @code{set architecture}
16715 @item @code{library-info}
16716 @tab @code{qXfer:libraries:read}
16717 @tab @code{info sharedlibrary}
16719 @item @code{memory-map}
16720 @tab @code{qXfer:memory-map:read}
16721 @tab @code{info mem}
16723 @item @code{read-sdata-object}
16724 @tab @code{qXfer:sdata:read}
16725 @tab @code{print $_sdata}
16727 @item @code{read-spu-object}
16728 @tab @code{qXfer:spu:read}
16729 @tab @code{info spu}
16731 @item @code{write-spu-object}
16732 @tab @code{qXfer:spu:write}
16733 @tab @code{info spu}
16735 @item @code{read-siginfo-object}
16736 @tab @code{qXfer:siginfo:read}
16737 @tab @code{print $_siginfo}
16739 @item @code{write-siginfo-object}
16740 @tab @code{qXfer:siginfo:write}
16741 @tab @code{set $_siginfo}
16743 @item @code{threads}
16744 @tab @code{qXfer:threads:read}
16745 @tab @code{info threads}
16747 @item @code{get-thread-local-@*storage-address}
16748 @tab @code{qGetTLSAddr}
16749 @tab Displaying @code{__thread} variables
16751 @item @code{get-thread-information-block-address}
16752 @tab @code{qGetTIBAddr}
16753 @tab Display MS-Windows Thread Information Block.
16755 @item @code{search-memory}
16756 @tab @code{qSearch:memory}
16759 @item @code{supported-packets}
16760 @tab @code{qSupported}
16761 @tab Remote communications parameters
16763 @item @code{pass-signals}
16764 @tab @code{QPassSignals}
16765 @tab @code{handle @var{signal}}
16767 @item @code{hostio-close-packet}
16768 @tab @code{vFile:close}
16769 @tab @code{remote get}, @code{remote put}
16771 @item @code{hostio-open-packet}
16772 @tab @code{vFile:open}
16773 @tab @code{remote get}, @code{remote put}
16775 @item @code{hostio-pread-packet}
16776 @tab @code{vFile:pread}
16777 @tab @code{remote get}, @code{remote put}
16779 @item @code{hostio-pwrite-packet}
16780 @tab @code{vFile:pwrite}
16781 @tab @code{remote get}, @code{remote put}
16783 @item @code{hostio-unlink-packet}
16784 @tab @code{vFile:unlink}
16785 @tab @code{remote delete}
16787 @item @code{noack-packet}
16788 @tab @code{QStartNoAckMode}
16789 @tab Packet acknowledgment
16791 @item @code{osdata}
16792 @tab @code{qXfer:osdata:read}
16793 @tab @code{info os}
16795 @item @code{query-attached}
16796 @tab @code{qAttached}
16797 @tab Querying remote process attach state.
16799 @item @code{traceframe-info}
16800 @tab @code{qXfer:traceframe-info:read}
16801 @tab Traceframe info
16805 @section Implementing a Remote Stub
16807 @cindex debugging stub, example
16808 @cindex remote stub, example
16809 @cindex stub example, remote debugging
16810 The stub files provided with @value{GDBN} implement the target side of the
16811 communication protocol, and the @value{GDBN} side is implemented in the
16812 @value{GDBN} source file @file{remote.c}. Normally, you can simply allow
16813 these subroutines to communicate, and ignore the details. (If you're
16814 implementing your own stub file, you can still ignore the details: start
16815 with one of the existing stub files. @file{sparc-stub.c} is the best
16816 organized, and therefore the easiest to read.)
16818 @cindex remote serial debugging, overview
16819 To debug a program running on another machine (the debugging
16820 @dfn{target} machine), you must first arrange for all the usual
16821 prerequisites for the program to run by itself. For example, for a C
16826 A startup routine to set up the C runtime environment; these usually
16827 have a name like @file{crt0}. The startup routine may be supplied by
16828 your hardware supplier, or you may have to write your own.
16831 A C subroutine library to support your program's
16832 subroutine calls, notably managing input and output.
16835 A way of getting your program to the other machine---for example, a
16836 download program. These are often supplied by the hardware
16837 manufacturer, but you may have to write your own from hardware
16841 The next step is to arrange for your program to use a serial port to
16842 communicate with the machine where @value{GDBN} is running (the @dfn{host}
16843 machine). In general terms, the scheme looks like this:
16847 @value{GDBN} already understands how to use this protocol; when everything
16848 else is set up, you can simply use the @samp{target remote} command
16849 (@pxref{Targets,,Specifying a Debugging Target}).
16851 @item On the target,
16852 you must link with your program a few special-purpose subroutines that
16853 implement the @value{GDBN} remote serial protocol. The file containing these
16854 subroutines is called a @dfn{debugging stub}.
16856 On certain remote targets, you can use an auxiliary program
16857 @code{gdbserver} instead of linking a stub into your program.
16858 @xref{Server,,Using the @code{gdbserver} Program}, for details.
16861 The debugging stub is specific to the architecture of the remote
16862 machine; for example, use @file{sparc-stub.c} to debug programs on
16865 @cindex remote serial stub list
16866 These working remote stubs are distributed with @value{GDBN}:
16871 @cindex @file{i386-stub.c}
16874 For Intel 386 and compatible architectures.
16877 @cindex @file{m68k-stub.c}
16878 @cindex Motorola 680x0
16880 For Motorola 680x0 architectures.
16883 @cindex @file{sh-stub.c}
16886 For Renesas SH architectures.
16889 @cindex @file{sparc-stub.c}
16891 For @sc{sparc} architectures.
16893 @item sparcl-stub.c
16894 @cindex @file{sparcl-stub.c}
16897 For Fujitsu @sc{sparclite} architectures.
16901 The @file{README} file in the @value{GDBN} distribution may list other
16902 recently added stubs.
16905 * Stub Contents:: What the stub can do for you
16906 * Bootstrapping:: What you must do for the stub
16907 * Debug Session:: Putting it all together
16910 @node Stub Contents
16911 @subsection What the Stub Can Do for You
16913 @cindex remote serial stub
16914 The debugging stub for your architecture supplies these three
16918 @item set_debug_traps
16919 @findex set_debug_traps
16920 @cindex remote serial stub, initialization
16921 This routine arranges for @code{handle_exception} to run when your
16922 program stops. You must call this subroutine explicitly near the
16923 beginning of your program.
16925 @item handle_exception
16926 @findex handle_exception
16927 @cindex remote serial stub, main routine
16928 This is the central workhorse, but your program never calls it
16929 explicitly---the setup code arranges for @code{handle_exception} to
16930 run when a trap is triggered.
16932 @code{handle_exception} takes control when your program stops during
16933 execution (for example, on a breakpoint), and mediates communications
16934 with @value{GDBN} on the host machine. This is where the communications
16935 protocol is implemented; @code{handle_exception} acts as the @value{GDBN}
16936 representative on the target machine. It begins by sending summary
16937 information on the state of your program, then continues to execute,
16938 retrieving and transmitting any information @value{GDBN} needs, until you
16939 execute a @value{GDBN} command that makes your program resume; at that point,
16940 @code{handle_exception} returns control to your own code on the target
16944 @cindex @code{breakpoint} subroutine, remote
16945 Use this auxiliary subroutine to make your program contain a
16946 breakpoint. Depending on the particular situation, this may be the only
16947 way for @value{GDBN} to get control. For instance, if your target
16948 machine has some sort of interrupt button, you won't need to call this;
16949 pressing the interrupt button transfers control to
16950 @code{handle_exception}---in effect, to @value{GDBN}. On some machines,
16951 simply receiving characters on the serial port may also trigger a trap;
16952 again, in that situation, you don't need to call @code{breakpoint} from
16953 your own program---simply running @samp{target remote} from the host
16954 @value{GDBN} session gets control.
16956 Call @code{breakpoint} if none of these is true, or if you simply want
16957 to make certain your program stops at a predetermined point for the
16958 start of your debugging session.
16961 @node Bootstrapping
16962 @subsection What You Must Do for the Stub
16964 @cindex remote stub, support routines
16965 The debugging stubs that come with @value{GDBN} are set up for a particular
16966 chip architecture, but they have no information about the rest of your
16967 debugging target machine.
16969 First of all you need to tell the stub how to communicate with the
16973 @item int getDebugChar()
16974 @findex getDebugChar
16975 Write this subroutine to read a single character from the serial port.
16976 It may be identical to @code{getchar} for your target system; a
16977 different name is used to allow you to distinguish the two if you wish.
16979 @item void putDebugChar(int)
16980 @findex putDebugChar
16981 Write this subroutine to write a single character to the serial port.
16982 It may be identical to @code{putchar} for your target system; a
16983 different name is used to allow you to distinguish the two if you wish.
16986 @cindex control C, and remote debugging
16987 @cindex interrupting remote targets
16988 If you want @value{GDBN} to be able to stop your program while it is
16989 running, you need to use an interrupt-driven serial driver, and arrange
16990 for it to stop when it receives a @code{^C} (@samp{\003}, the control-C
16991 character). That is the character which @value{GDBN} uses to tell the
16992 remote system to stop.
16994 Getting the debugging target to return the proper status to @value{GDBN}
16995 probably requires changes to the standard stub; one quick and dirty way
16996 is to just execute a breakpoint instruction (the ``dirty'' part is that
16997 @value{GDBN} reports a @code{SIGTRAP} instead of a @code{SIGINT}).
16999 Other routines you need to supply are:
17002 @item void exceptionHandler (int @var{exception_number}, void *@var{exception_address})
17003 @findex exceptionHandler
17004 Write this function to install @var{exception_address} in the exception
17005 handling tables. You need to do this because the stub does not have any
17006 way of knowing what the exception handling tables on your target system
17007 are like (for example, the processor's table might be in @sc{rom},
17008 containing entries which point to a table in @sc{ram}).
17009 @var{exception_number} is the exception number which should be changed;
17010 its meaning is architecture-dependent (for example, different numbers
17011 might represent divide by zero, misaligned access, etc). When this
17012 exception occurs, control should be transferred directly to
17013 @var{exception_address}, and the processor state (stack, registers,
17014 and so on) should be just as it is when a processor exception occurs. So if
17015 you want to use a jump instruction to reach @var{exception_address}, it
17016 should be a simple jump, not a jump to subroutine.
17018 For the 386, @var{exception_address} should be installed as an interrupt
17019 gate so that interrupts are masked while the handler runs. The gate
17020 should be at privilege level 0 (the most privileged level). The
17021 @sc{sparc} and 68k stubs are able to mask interrupts themselves without
17022 help from @code{exceptionHandler}.
17024 @item void flush_i_cache()
17025 @findex flush_i_cache
17026 On @sc{sparc} and @sc{sparclite} only, write this subroutine to flush the
17027 instruction cache, if any, on your target machine. If there is no
17028 instruction cache, this subroutine may be a no-op.
17030 On target machines that have instruction caches, @value{GDBN} requires this
17031 function to make certain that the state of your program is stable.
17035 You must also make sure this library routine is available:
17038 @item void *memset(void *, int, int)
17040 This is the standard library function @code{memset} that sets an area of
17041 memory to a known value. If you have one of the free versions of
17042 @code{libc.a}, @code{memset} can be found there; otherwise, you must
17043 either obtain it from your hardware manufacturer, or write your own.
17046 If you do not use the GNU C compiler, you may need other standard
17047 library subroutines as well; this varies from one stub to another,
17048 but in general the stubs are likely to use any of the common library
17049 subroutines which @code{@value{NGCC}} generates as inline code.
17052 @node Debug Session
17053 @subsection Putting it All Together
17055 @cindex remote serial debugging summary
17056 In summary, when your program is ready to debug, you must follow these
17061 Make sure you have defined the supporting low-level routines
17062 (@pxref{Bootstrapping,,What You Must Do for the Stub}):
17064 @code{getDebugChar}, @code{putDebugChar},
17065 @code{flush_i_cache}, @code{memset}, @code{exceptionHandler}.
17069 Insert these lines near the top of your program:
17077 For the 680x0 stub only, you need to provide a variable called
17078 @code{exceptionHook}. Normally you just use:
17081 void (*exceptionHook)() = 0;
17085 but if before calling @code{set_debug_traps}, you set it to point to a
17086 function in your program, that function is called when
17087 @code{@value{GDBN}} continues after stopping on a trap (for example, bus
17088 error). The function indicated by @code{exceptionHook} is called with
17089 one parameter: an @code{int} which is the exception number.
17092 Compile and link together: your program, the @value{GDBN} debugging stub for
17093 your target architecture, and the supporting subroutines.
17096 Make sure you have a serial connection between your target machine and
17097 the @value{GDBN} host, and identify the serial port on the host.
17100 @c The "remote" target now provides a `load' command, so we should
17101 @c document that. FIXME.
17102 Download your program to your target machine (or get it there by
17103 whatever means the manufacturer provides), and start it.
17106 Start @value{GDBN} on the host, and connect to the target
17107 (@pxref{Connecting,,Connecting to a Remote Target}).
17111 @node Configurations
17112 @chapter Configuration-Specific Information
17114 While nearly all @value{GDBN} commands are available for all native and
17115 cross versions of the debugger, there are some exceptions. This chapter
17116 describes things that are only available in certain configurations.
17118 There are three major categories of configurations: native
17119 configurations, where the host and target are the same, embedded
17120 operating system configurations, which are usually the same for several
17121 different processor architectures, and bare embedded processors, which
17122 are quite different from each other.
17127 * Embedded Processors::
17134 This section describes details specific to particular native
17139 * BSD libkvm Interface:: Debugging BSD kernel memory images
17140 * SVR4 Process Information:: SVR4 process information
17141 * DJGPP Native:: Features specific to the DJGPP port
17142 * Cygwin Native:: Features specific to the Cygwin port
17143 * Hurd Native:: Features specific to @sc{gnu} Hurd
17144 * Neutrino:: Features specific to QNX Neutrino
17145 * Darwin:: Features specific to Darwin
17151 On HP-UX systems, if you refer to a function or variable name that
17152 begins with a dollar sign, @value{GDBN} searches for a user or system
17153 name first, before it searches for a convenience variable.
17156 @node BSD libkvm Interface
17157 @subsection BSD libkvm Interface
17160 @cindex kernel memory image
17161 @cindex kernel crash dump
17163 BSD-derived systems (FreeBSD/NetBSD/OpenBSD) have a kernel memory
17164 interface that provides a uniform interface for accessing kernel virtual
17165 memory images, including live systems and crash dumps. @value{GDBN}
17166 uses this interface to allow you to debug live kernels and kernel crash
17167 dumps on many native BSD configurations. This is implemented as a
17168 special @code{kvm} debugging target. For debugging a live system, load
17169 the currently running kernel into @value{GDBN} and connect to the
17173 (@value{GDBP}) @b{target kvm}
17176 For debugging crash dumps, provide the file name of the crash dump as an
17180 (@value{GDBP}) @b{target kvm /var/crash/bsd.0}
17183 Once connected to the @code{kvm} target, the following commands are
17189 Set current context from the @dfn{Process Control Block} (PCB) address.
17192 Set current context from proc address. This command isn't available on
17193 modern FreeBSD systems.
17196 @node SVR4 Process Information
17197 @subsection SVR4 Process Information
17199 @cindex examine process image
17200 @cindex process info via @file{/proc}
17202 Many versions of SVR4 and compatible systems provide a facility called
17203 @samp{/proc} that can be used to examine the image of a running
17204 process using file-system subroutines. If @value{GDBN} is configured
17205 for an operating system with this facility, the command @code{info
17206 proc} is available to report information about the process running
17207 your program, or about any process running on your system. @code{info
17208 proc} works only on SVR4 systems that include the @code{procfs} code.
17209 This includes, as of this writing, @sc{gnu}/Linux, OSF/1 (Digital
17210 Unix), Solaris, Irix, and Unixware, but not HP-UX, for example.
17216 @itemx info proc @var{process-id}
17217 Summarize available information about any running process. If a
17218 process ID is specified by @var{process-id}, display information about
17219 that process; otherwise display information about the program being
17220 debugged. The summary includes the debugged process ID, the command
17221 line used to invoke it, its current working directory, and its
17222 executable file's absolute file name.
17224 On some systems, @var{process-id} can be of the form
17225 @samp{[@var{pid}]/@var{tid}} which specifies a certain thread ID
17226 within a process. If the optional @var{pid} part is missing, it means
17227 a thread from the process being debugged (the leading @samp{/} still
17228 needs to be present, or else @value{GDBN} will interpret the number as
17229 a process ID rather than a thread ID).
17231 @item info proc mappings
17232 @cindex memory address space mappings
17233 Report the memory address space ranges accessible in the program, with
17234 information on whether the process has read, write, or execute access
17235 rights to each range. On @sc{gnu}/Linux systems, each memory range
17236 includes the object file which is mapped to that range, instead of the
17237 memory access rights to that range.
17239 @item info proc stat
17240 @itemx info proc status
17241 @cindex process detailed status information
17242 These subcommands are specific to @sc{gnu}/Linux systems. They show
17243 the process-related information, including the user ID and group ID;
17244 how many threads are there in the process; its virtual memory usage;
17245 the signals that are pending, blocked, and ignored; its TTY; its
17246 consumption of system and user time; its stack size; its @samp{nice}
17247 value; etc. For more information, see the @samp{proc} man page
17248 (type @kbd{man 5 proc} from your shell prompt).
17250 @item info proc all
17251 Show all the information about the process described under all of the
17252 above @code{info proc} subcommands.
17255 @comment These sub-options of 'info proc' were not included when
17256 @comment procfs.c was re-written. Keep their descriptions around
17257 @comment against the day when someone finds the time to put them back in.
17258 @kindex info proc times
17259 @item info proc times
17260 Starting time, user CPU time, and system CPU time for your program and
17263 @kindex info proc id
17265 Report on the process IDs related to your program: its own process ID,
17266 the ID of its parent, the process group ID, and the session ID.
17269 @item set procfs-trace
17270 @kindex set procfs-trace
17271 @cindex @code{procfs} API calls
17272 This command enables and disables tracing of @code{procfs} API calls.
17274 @item show procfs-trace
17275 @kindex show procfs-trace
17276 Show the current state of @code{procfs} API call tracing.
17278 @item set procfs-file @var{file}
17279 @kindex set procfs-file
17280 Tell @value{GDBN} to write @code{procfs} API trace to the named
17281 @var{file}. @value{GDBN} appends the trace info to the previous
17282 contents of the file. The default is to display the trace on the
17285 @item show procfs-file
17286 @kindex show procfs-file
17287 Show the file to which @code{procfs} API trace is written.
17289 @item proc-trace-entry
17290 @itemx proc-trace-exit
17291 @itemx proc-untrace-entry
17292 @itemx proc-untrace-exit
17293 @kindex proc-trace-entry
17294 @kindex proc-trace-exit
17295 @kindex proc-untrace-entry
17296 @kindex proc-untrace-exit
17297 These commands enable and disable tracing of entries into and exits
17298 from the @code{syscall} interface.
17301 @kindex info pidlist
17302 @cindex process list, QNX Neutrino
17303 For QNX Neutrino only, this command displays the list of all the
17304 processes and all the threads within each process.
17307 @kindex info meminfo
17308 @cindex mapinfo list, QNX Neutrino
17309 For QNX Neutrino only, this command displays the list of all mapinfos.
17313 @subsection Features for Debugging @sc{djgpp} Programs
17314 @cindex @sc{djgpp} debugging
17315 @cindex native @sc{djgpp} debugging
17316 @cindex MS-DOS-specific commands
17319 @sc{djgpp} is a port of the @sc{gnu} development tools to MS-DOS and
17320 MS-Windows. @sc{djgpp} programs are 32-bit protected-mode programs
17321 that use the @dfn{DPMI} (DOS Protected-Mode Interface) API to run on
17322 top of real-mode DOS systems and their emulations.
17324 @value{GDBN} supports native debugging of @sc{djgpp} programs, and
17325 defines a few commands specific to the @sc{djgpp} port. This
17326 subsection describes those commands.
17331 This is a prefix of @sc{djgpp}-specific commands which print
17332 information about the target system and important OS structures.
17335 @cindex MS-DOS system info
17336 @cindex free memory information (MS-DOS)
17337 @item info dos sysinfo
17338 This command displays assorted information about the underlying
17339 platform: the CPU type and features, the OS version and flavor, the
17340 DPMI version, and the available conventional and DPMI memory.
17345 @cindex segment descriptor tables
17346 @cindex descriptor tables display
17348 @itemx info dos ldt
17349 @itemx info dos idt
17350 These 3 commands display entries from, respectively, Global, Local,
17351 and Interrupt Descriptor Tables (GDT, LDT, and IDT). The descriptor
17352 tables are data structures which store a descriptor for each segment
17353 that is currently in use. The segment's selector is an index into a
17354 descriptor table; the table entry for that index holds the
17355 descriptor's base address and limit, and its attributes and access
17358 A typical @sc{djgpp} program uses 3 segments: a code segment, a data
17359 segment (used for both data and the stack), and a DOS segment (which
17360 allows access to DOS/BIOS data structures and absolute addresses in
17361 conventional memory). However, the DPMI host will usually define
17362 additional segments in order to support the DPMI environment.
17364 @cindex garbled pointers
17365 These commands allow to display entries from the descriptor tables.
17366 Without an argument, all entries from the specified table are
17367 displayed. An argument, which should be an integer expression, means
17368 display a single entry whose index is given by the argument. For
17369 example, here's a convenient way to display information about the
17370 debugged program's data segment:
17373 @exdent @code{(@value{GDBP}) info dos ldt $ds}
17374 @exdent @code{0x13f: base=0x11970000 limit=0x0009ffff 32-Bit Data (Read/Write, Exp-up)}
17378 This comes in handy when you want to see whether a pointer is outside
17379 the data segment's limit (i.e.@: @dfn{garbled}).
17381 @cindex page tables display (MS-DOS)
17383 @itemx info dos pte
17384 These two commands display entries from, respectively, the Page
17385 Directory and the Page Tables. Page Directories and Page Tables are
17386 data structures which control how virtual memory addresses are mapped
17387 into physical addresses. A Page Table includes an entry for every
17388 page of memory that is mapped into the program's address space; there
17389 may be several Page Tables, each one holding up to 4096 entries. A
17390 Page Directory has up to 4096 entries, one each for every Page Table
17391 that is currently in use.
17393 Without an argument, @kbd{info dos pde} displays the entire Page
17394 Directory, and @kbd{info dos pte} displays all the entries in all of
17395 the Page Tables. An argument, an integer expression, given to the
17396 @kbd{info dos pde} command means display only that entry from the Page
17397 Directory table. An argument given to the @kbd{info dos pte} command
17398 means display entries from a single Page Table, the one pointed to by
17399 the specified entry in the Page Directory.
17401 @cindex direct memory access (DMA) on MS-DOS
17402 These commands are useful when your program uses @dfn{DMA} (Direct
17403 Memory Access), which needs physical addresses to program the DMA
17406 These commands are supported only with some DPMI servers.
17408 @cindex physical address from linear address
17409 @item info dos address-pte @var{addr}
17410 This command displays the Page Table entry for a specified linear
17411 address. The argument @var{addr} is a linear address which should
17412 already have the appropriate segment's base address added to it,
17413 because this command accepts addresses which may belong to @emph{any}
17414 segment. For example, here's how to display the Page Table entry for
17415 the page where a variable @code{i} is stored:
17418 @exdent @code{(@value{GDBP}) info dos address-pte __djgpp_base_address + (char *)&i}
17419 @exdent @code{Page Table entry for address 0x11a00d30:}
17420 @exdent @code{Base=0x02698000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0xd30}
17424 This says that @code{i} is stored at offset @code{0xd30} from the page
17425 whose physical base address is @code{0x02698000}, and shows all the
17426 attributes of that page.
17428 Note that you must cast the addresses of variables to a @code{char *},
17429 since otherwise the value of @code{__djgpp_base_address}, the base
17430 address of all variables and functions in a @sc{djgpp} program, will
17431 be added using the rules of C pointer arithmetics: if @code{i} is
17432 declared an @code{int}, @value{GDBN} will add 4 times the value of
17433 @code{__djgpp_base_address} to the address of @code{i}.
17435 Here's another example, it displays the Page Table entry for the
17439 @exdent @code{(@value{GDBP}) info dos address-pte *((unsigned *)&_go32_info_block + 3)}
17440 @exdent @code{Page Table entry for address 0x29110:}
17441 @exdent @code{Base=0x00029000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0x110}
17445 (The @code{+ 3} offset is because the transfer buffer's address is the
17446 3rd member of the @code{_go32_info_block} structure.) The output
17447 clearly shows that this DPMI server maps the addresses in conventional
17448 memory 1:1, i.e.@: the physical (@code{0x00029000} + @code{0x110}) and
17449 linear (@code{0x29110}) addresses are identical.
17451 This command is supported only with some DPMI servers.
17454 @cindex DOS serial data link, remote debugging
17455 In addition to native debugging, the DJGPP port supports remote
17456 debugging via a serial data link. The following commands are specific
17457 to remote serial debugging in the DJGPP port of @value{GDBN}.
17460 @kindex set com1base
17461 @kindex set com1irq
17462 @kindex set com2base
17463 @kindex set com2irq
17464 @kindex set com3base
17465 @kindex set com3irq
17466 @kindex set com4base
17467 @kindex set com4irq
17468 @item set com1base @var{addr}
17469 This command sets the base I/O port address of the @file{COM1} serial
17472 @item set com1irq @var{irq}
17473 This command sets the @dfn{Interrupt Request} (@code{IRQ}) line to use
17474 for the @file{COM1} serial port.
17476 There are similar commands @samp{set com2base}, @samp{set com3irq},
17477 etc.@: for setting the port address and the @code{IRQ} lines for the
17480 @kindex show com1base
17481 @kindex show com1irq
17482 @kindex show com2base
17483 @kindex show com2irq
17484 @kindex show com3base
17485 @kindex show com3irq
17486 @kindex show com4base
17487 @kindex show com4irq
17488 The related commands @samp{show com1base}, @samp{show com1irq} etc.@:
17489 display the current settings of the base address and the @code{IRQ}
17490 lines used by the COM ports.
17493 @kindex info serial
17494 @cindex DOS serial port status
17495 This command prints the status of the 4 DOS serial ports. For each
17496 port, it prints whether it's active or not, its I/O base address and
17497 IRQ number, whether it uses a 16550-style FIFO, its baudrate, and the
17498 counts of various errors encountered so far.
17502 @node Cygwin Native
17503 @subsection Features for Debugging MS Windows PE Executables
17504 @cindex MS Windows debugging
17505 @cindex native Cygwin debugging
17506 @cindex Cygwin-specific commands
17508 @value{GDBN} supports native debugging of MS Windows programs, including
17509 DLLs with and without symbolic debugging information.
17511 @cindex Ctrl-BREAK, MS-Windows
17512 @cindex interrupt debuggee on MS-Windows
17513 MS-Windows programs that call @code{SetConsoleMode} to switch off the
17514 special meaning of the @samp{Ctrl-C} keystroke cannot be interrupted
17515 by typing @kbd{C-c}. For this reason, @value{GDBN} on MS-Windows
17516 supports @kbd{C-@key{BREAK}} as an alternative interrupt key
17517 sequence, which can be used to interrupt the debuggee even if it
17520 There are various additional Cygwin-specific commands, described in
17521 this section. Working with DLLs that have no debugging symbols is
17522 described in @ref{Non-debug DLL Symbols}.
17527 This is a prefix of MS Windows-specific commands which print
17528 information about the target system and important OS structures.
17530 @item info w32 selector
17531 This command displays information returned by
17532 the Win32 API @code{GetThreadSelectorEntry} function.
17533 It takes an optional argument that is evaluated to
17534 a long value to give the information about this given selector.
17535 Without argument, this command displays information
17536 about the six segment registers.
17538 @item info w32 thread-information-block
17539 This command displays thread specific information stored in the
17540 Thread Information Block (readable on the X86 CPU family using @code{$fs}
17541 selector for 32-bit programs and @code{$gs} for 64-bit programs).
17545 This is a Cygwin-specific alias of @code{info shared}.
17547 @kindex dll-symbols
17549 This command loads symbols from a dll similarly to
17550 add-sym command but without the need to specify a base address.
17552 @kindex set cygwin-exceptions
17553 @cindex debugging the Cygwin DLL
17554 @cindex Cygwin DLL, debugging
17555 @item set cygwin-exceptions @var{mode}
17556 If @var{mode} is @code{on}, @value{GDBN} will break on exceptions that
17557 happen inside the Cygwin DLL. If @var{mode} is @code{off},
17558 @value{GDBN} will delay recognition of exceptions, and may ignore some
17559 exceptions which seem to be caused by internal Cygwin DLL
17560 ``bookkeeping''. This option is meant primarily for debugging the
17561 Cygwin DLL itself; the default value is @code{off} to avoid annoying
17562 @value{GDBN} users with false @code{SIGSEGV} signals.
17564 @kindex show cygwin-exceptions
17565 @item show cygwin-exceptions
17566 Displays whether @value{GDBN} will break on exceptions that happen
17567 inside the Cygwin DLL itself.
17569 @kindex set new-console
17570 @item set new-console @var{mode}
17571 If @var{mode} is @code{on} the debuggee will
17572 be started in a new console on next start.
17573 If @var{mode} is @code{off}, the debuggee will
17574 be started in the same console as the debugger.
17576 @kindex show new-console
17577 @item show new-console
17578 Displays whether a new console is used
17579 when the debuggee is started.
17581 @kindex set new-group
17582 @item set new-group @var{mode}
17583 This boolean value controls whether the debuggee should
17584 start a new group or stay in the same group as the debugger.
17585 This affects the way the Windows OS handles
17588 @kindex show new-group
17589 @item show new-group
17590 Displays current value of new-group boolean.
17592 @kindex set debugevents
17593 @item set debugevents
17594 This boolean value adds debug output concerning kernel events related
17595 to the debuggee seen by the debugger. This includes events that
17596 signal thread and process creation and exit, DLL loading and
17597 unloading, console interrupts, and debugging messages produced by the
17598 Windows @code{OutputDebugString} API call.
17600 @kindex set debugexec
17601 @item set debugexec
17602 This boolean value adds debug output concerning execute events
17603 (such as resume thread) seen by the debugger.
17605 @kindex set debugexceptions
17606 @item set debugexceptions
17607 This boolean value adds debug output concerning exceptions in the
17608 debuggee seen by the debugger.
17610 @kindex set debugmemory
17611 @item set debugmemory
17612 This boolean value adds debug output concerning debuggee memory reads
17613 and writes by the debugger.
17617 This boolean values specifies whether the debuggee is called
17618 via a shell or directly (default value is on).
17622 Displays if the debuggee will be started with a shell.
17627 * Non-debug DLL Symbols:: Support for DLLs without debugging symbols
17630 @node Non-debug DLL Symbols
17631 @subsubsection Support for DLLs without Debugging Symbols
17632 @cindex DLLs with no debugging symbols
17633 @cindex Minimal symbols and DLLs
17635 Very often on windows, some of the DLLs that your program relies on do
17636 not include symbolic debugging information (for example,
17637 @file{kernel32.dll}). When @value{GDBN} doesn't recognize any debugging
17638 symbols in a DLL, it relies on the minimal amount of symbolic
17639 information contained in the DLL's export table. This section
17640 describes working with such symbols, known internally to @value{GDBN} as
17641 ``minimal symbols''.
17643 Note that before the debugged program has started execution, no DLLs
17644 will have been loaded. The easiest way around this problem is simply to
17645 start the program --- either by setting a breakpoint or letting the
17646 program run once to completion. It is also possible to force
17647 @value{GDBN} to load a particular DLL before starting the executable ---
17648 see the shared library information in @ref{Files}, or the
17649 @code{dll-symbols} command in @ref{Cygwin Native}. Currently,
17650 explicitly loading symbols from a DLL with no debugging information will
17651 cause the symbol names to be duplicated in @value{GDBN}'s lookup table,
17652 which may adversely affect symbol lookup performance.
17654 @subsubsection DLL Name Prefixes
17656 In keeping with the naming conventions used by the Microsoft debugging
17657 tools, DLL export symbols are made available with a prefix based on the
17658 DLL name, for instance @code{KERNEL32!CreateFileA}. The plain name is
17659 also entered into the symbol table, so @code{CreateFileA} is often
17660 sufficient. In some cases there will be name clashes within a program
17661 (particularly if the executable itself includes full debugging symbols)
17662 necessitating the use of the fully qualified name when referring to the
17663 contents of the DLL. Use single-quotes around the name to avoid the
17664 exclamation mark (``!'') being interpreted as a language operator.
17666 Note that the internal name of the DLL may be all upper-case, even
17667 though the file name of the DLL is lower-case, or vice-versa. Since
17668 symbols within @value{GDBN} are @emph{case-sensitive} this may cause
17669 some confusion. If in doubt, try the @code{info functions} and
17670 @code{info variables} commands or even @code{maint print msymbols}
17671 (@pxref{Symbols}). Here's an example:
17674 (@value{GDBP}) info function CreateFileA
17675 All functions matching regular expression "CreateFileA":
17677 Non-debugging symbols:
17678 0x77e885f4 CreateFileA
17679 0x77e885f4 KERNEL32!CreateFileA
17683 (@value{GDBP}) info function !
17684 All functions matching regular expression "!":
17686 Non-debugging symbols:
17687 0x6100114c cygwin1!__assert
17688 0x61004034 cygwin1!_dll_crt0@@0
17689 0x61004240 cygwin1!dll_crt0(per_process *)
17693 @subsubsection Working with Minimal Symbols
17695 Symbols extracted from a DLL's export table do not contain very much
17696 type information. All that @value{GDBN} can do is guess whether a symbol
17697 refers to a function or variable depending on the linker section that
17698 contains the symbol. Also note that the actual contents of the memory
17699 contained in a DLL are not available unless the program is running. This
17700 means that you cannot examine the contents of a variable or disassemble
17701 a function within a DLL without a running program.
17703 Variables are generally treated as pointers and dereferenced
17704 automatically. For this reason, it is often necessary to prefix a
17705 variable name with the address-of operator (``&'') and provide explicit
17706 type information in the command. Here's an example of the type of
17710 (@value{GDBP}) print 'cygwin1!__argv'
17715 (@value{GDBP}) x 'cygwin1!__argv'
17716 0x10021610: "\230y\""
17719 And two possible solutions:
17722 (@value{GDBP}) print ((char **)'cygwin1!__argv')[0]
17723 $2 = 0x22fd98 "/cygdrive/c/mydirectory/myprogram"
17727 (@value{GDBP}) x/2x &'cygwin1!__argv'
17728 0x610c0aa8 <cygwin1!__argv>: 0x10021608 0x00000000
17729 (@value{GDBP}) x/x 0x10021608
17730 0x10021608: 0x0022fd98
17731 (@value{GDBP}) x/s 0x0022fd98
17732 0x22fd98: "/cygdrive/c/mydirectory/myprogram"
17735 Setting a break point within a DLL is possible even before the program
17736 starts execution. However, under these circumstances, @value{GDBN} can't
17737 examine the initial instructions of the function in order to skip the
17738 function's frame set-up code. You can work around this by using ``*&''
17739 to set the breakpoint at a raw memory address:
17742 (@value{GDBP}) break *&'python22!PyOS_Readline'
17743 Breakpoint 1 at 0x1e04eff0
17746 The author of these extensions is not entirely convinced that setting a
17747 break point within a shared DLL like @file{kernel32.dll} is completely
17751 @subsection Commands Specific to @sc{gnu} Hurd Systems
17752 @cindex @sc{gnu} Hurd debugging
17754 This subsection describes @value{GDBN} commands specific to the
17755 @sc{gnu} Hurd native debugging.
17760 @kindex set signals@r{, Hurd command}
17761 @kindex set sigs@r{, Hurd command}
17762 This command toggles the state of inferior signal interception by
17763 @value{GDBN}. Mach exceptions, such as breakpoint traps, are not
17764 affected by this command. @code{sigs} is a shorthand alias for
17769 @kindex show signals@r{, Hurd command}
17770 @kindex show sigs@r{, Hurd command}
17771 Show the current state of intercepting inferior's signals.
17773 @item set signal-thread
17774 @itemx set sigthread
17775 @kindex set signal-thread
17776 @kindex set sigthread
17777 This command tells @value{GDBN} which thread is the @code{libc} signal
17778 thread. That thread is run when a signal is delivered to a running
17779 process. @code{set sigthread} is the shorthand alias of @code{set
17782 @item show signal-thread
17783 @itemx show sigthread
17784 @kindex show signal-thread
17785 @kindex show sigthread
17786 These two commands show which thread will run when the inferior is
17787 delivered a signal.
17790 @kindex set stopped@r{, Hurd command}
17791 This commands tells @value{GDBN} that the inferior process is stopped,
17792 as with the @code{SIGSTOP} signal. The stopped process can be
17793 continued by delivering a signal to it.
17796 @kindex show stopped@r{, Hurd command}
17797 This command shows whether @value{GDBN} thinks the debuggee is
17800 @item set exceptions
17801 @kindex set exceptions@r{, Hurd command}
17802 Use this command to turn off trapping of exceptions in the inferior.
17803 When exception trapping is off, neither breakpoints nor
17804 single-stepping will work. To restore the default, set exception
17807 @item show exceptions
17808 @kindex show exceptions@r{, Hurd command}
17809 Show the current state of trapping exceptions in the inferior.
17811 @item set task pause
17812 @kindex set task@r{, Hurd commands}
17813 @cindex task attributes (@sc{gnu} Hurd)
17814 @cindex pause current task (@sc{gnu} Hurd)
17815 This command toggles task suspension when @value{GDBN} has control.
17816 Setting it to on takes effect immediately, and the task is suspended
17817 whenever @value{GDBN} gets control. Setting it to off will take
17818 effect the next time the inferior is continued. If this option is set
17819 to off, you can use @code{set thread default pause on} or @code{set
17820 thread pause on} (see below) to pause individual threads.
17822 @item show task pause
17823 @kindex show task@r{, Hurd commands}
17824 Show the current state of task suspension.
17826 @item set task detach-suspend-count
17827 @cindex task suspend count
17828 @cindex detach from task, @sc{gnu} Hurd
17829 This command sets the suspend count the task will be left with when
17830 @value{GDBN} detaches from it.
17832 @item show task detach-suspend-count
17833 Show the suspend count the task will be left with when detaching.
17835 @item set task exception-port
17836 @itemx set task excp
17837 @cindex task exception port, @sc{gnu} Hurd
17838 This command sets the task exception port to which @value{GDBN} will
17839 forward exceptions. The argument should be the value of the @dfn{send
17840 rights} of the task. @code{set task excp} is a shorthand alias.
17842 @item set noninvasive
17843 @cindex noninvasive task options
17844 This command switches @value{GDBN} to a mode that is the least
17845 invasive as far as interfering with the inferior is concerned. This
17846 is the same as using @code{set task pause}, @code{set exceptions}, and
17847 @code{set signals} to values opposite to the defaults.
17849 @item info send-rights
17850 @itemx info receive-rights
17851 @itemx info port-rights
17852 @itemx info port-sets
17853 @itemx info dead-names
17856 @cindex send rights, @sc{gnu} Hurd
17857 @cindex receive rights, @sc{gnu} Hurd
17858 @cindex port rights, @sc{gnu} Hurd
17859 @cindex port sets, @sc{gnu} Hurd
17860 @cindex dead names, @sc{gnu} Hurd
17861 These commands display information about, respectively, send rights,
17862 receive rights, port rights, port sets, and dead names of a task.
17863 There are also shorthand aliases: @code{info ports} for @code{info
17864 port-rights} and @code{info psets} for @code{info port-sets}.
17866 @item set thread pause
17867 @kindex set thread@r{, Hurd command}
17868 @cindex thread properties, @sc{gnu} Hurd
17869 @cindex pause current thread (@sc{gnu} Hurd)
17870 This command toggles current thread suspension when @value{GDBN} has
17871 control. Setting it to on takes effect immediately, and the current
17872 thread is suspended whenever @value{GDBN} gets control. Setting it to
17873 off will take effect the next time the inferior is continued.
17874 Normally, this command has no effect, since when @value{GDBN} has
17875 control, the whole task is suspended. However, if you used @code{set
17876 task pause off} (see above), this command comes in handy to suspend
17877 only the current thread.
17879 @item show thread pause
17880 @kindex show thread@r{, Hurd command}
17881 This command shows the state of current thread suspension.
17883 @item set thread run
17884 This command sets whether the current thread is allowed to run.
17886 @item show thread run
17887 Show whether the current thread is allowed to run.
17889 @item set thread detach-suspend-count
17890 @cindex thread suspend count, @sc{gnu} Hurd
17891 @cindex detach from thread, @sc{gnu} Hurd
17892 This command sets the suspend count @value{GDBN} will leave on a
17893 thread when detaching. This number is relative to the suspend count
17894 found by @value{GDBN} when it notices the thread; use @code{set thread
17895 takeover-suspend-count} to force it to an absolute value.
17897 @item show thread detach-suspend-count
17898 Show the suspend count @value{GDBN} will leave on the thread when
17901 @item set thread exception-port
17902 @itemx set thread excp
17903 Set the thread exception port to which to forward exceptions. This
17904 overrides the port set by @code{set task exception-port} (see above).
17905 @code{set thread excp} is the shorthand alias.
17907 @item set thread takeover-suspend-count
17908 Normally, @value{GDBN}'s thread suspend counts are relative to the
17909 value @value{GDBN} finds when it notices each thread. This command
17910 changes the suspend counts to be absolute instead.
17912 @item set thread default
17913 @itemx show thread default
17914 @cindex thread default settings, @sc{gnu} Hurd
17915 Each of the above @code{set thread} commands has a @code{set thread
17916 default} counterpart (e.g., @code{set thread default pause}, @code{set
17917 thread default exception-port}, etc.). The @code{thread default}
17918 variety of commands sets the default thread properties for all
17919 threads; you can then change the properties of individual threads with
17920 the non-default commands.
17925 @subsection QNX Neutrino
17926 @cindex QNX Neutrino
17928 @value{GDBN} provides the following commands specific to the QNX
17932 @item set debug nto-debug
17933 @kindex set debug nto-debug
17934 When set to on, enables debugging messages specific to the QNX
17937 @item show debug nto-debug
17938 @kindex show debug nto-debug
17939 Show the current state of QNX Neutrino messages.
17946 @value{GDBN} provides the following commands specific to the Darwin target:
17949 @item set debug darwin @var{num}
17950 @kindex set debug darwin
17951 When set to a non zero value, enables debugging messages specific to
17952 the Darwin support. Higher values produce more verbose output.
17954 @item show debug darwin
17955 @kindex show debug darwin
17956 Show the current state of Darwin messages.
17958 @item set debug mach-o @var{num}
17959 @kindex set debug mach-o
17960 When set to a non zero value, enables debugging messages while
17961 @value{GDBN} is reading Darwin object files. (@dfn{Mach-O} is the
17962 file format used on Darwin for object and executable files.) Higher
17963 values produce more verbose output. This is a command to diagnose
17964 problems internal to @value{GDBN} and should not be needed in normal
17967 @item show debug mach-o
17968 @kindex show debug mach-o
17969 Show the current state of Mach-O file messages.
17971 @item set mach-exceptions on
17972 @itemx set mach-exceptions off
17973 @kindex set mach-exceptions
17974 On Darwin, faults are first reported as a Mach exception and are then
17975 mapped to a Posix signal. Use this command to turn on trapping of
17976 Mach exceptions in the inferior. This might be sometimes useful to
17977 better understand the cause of a fault. The default is off.
17979 @item show mach-exceptions
17980 @kindex show mach-exceptions
17981 Show the current state of exceptions trapping.
17986 @section Embedded Operating Systems
17988 This section describes configurations involving the debugging of
17989 embedded operating systems that are available for several different
17993 * VxWorks:: Using @value{GDBN} with VxWorks
17996 @value{GDBN} includes the ability to debug programs running on
17997 various real-time operating systems.
18000 @subsection Using @value{GDBN} with VxWorks
18006 @kindex target vxworks
18007 @item target vxworks @var{machinename}
18008 A VxWorks system, attached via TCP/IP. The argument @var{machinename}
18009 is the target system's machine name or IP address.
18013 On VxWorks, @code{load} links @var{filename} dynamically on the
18014 current target system as well as adding its symbols in @value{GDBN}.
18016 @value{GDBN} enables developers to spawn and debug tasks running on networked
18017 VxWorks targets from a Unix host. Already-running tasks spawned from
18018 the VxWorks shell can also be debugged. @value{GDBN} uses code that runs on
18019 both the Unix host and on the VxWorks target. The program
18020 @code{@value{GDBP}} is installed and executed on the Unix host. (It may be
18021 installed with the name @code{vxgdb}, to distinguish it from a
18022 @value{GDBN} for debugging programs on the host itself.)
18025 @item VxWorks-timeout @var{args}
18026 @kindex vxworks-timeout
18027 All VxWorks-based targets now support the option @code{vxworks-timeout}.
18028 This option is set by the user, and @var{args} represents the number of
18029 seconds @value{GDBN} waits for responses to rpc's. You might use this if
18030 your VxWorks target is a slow software simulator or is on the far side
18031 of a thin network line.
18034 The following information on connecting to VxWorks was current when
18035 this manual was produced; newer releases of VxWorks may use revised
18038 @findex INCLUDE_RDB
18039 To use @value{GDBN} with VxWorks, you must rebuild your VxWorks kernel
18040 to include the remote debugging interface routines in the VxWorks
18041 library @file{rdb.a}. To do this, define @code{INCLUDE_RDB} in the
18042 VxWorks configuration file @file{configAll.h} and rebuild your VxWorks
18043 kernel. The resulting kernel contains @file{rdb.a}, and spawns the
18044 source debugging task @code{tRdbTask} when VxWorks is booted. For more
18045 information on configuring and remaking VxWorks, see the manufacturer's
18047 @c VxWorks, see the @cite{VxWorks Programmer's Guide}.
18049 Once you have included @file{rdb.a} in your VxWorks system image and set
18050 your Unix execution search path to find @value{GDBN}, you are ready to
18051 run @value{GDBN}. From your Unix host, run @code{@value{GDBP}} (or
18052 @code{vxgdb}, depending on your installation).
18054 @value{GDBN} comes up showing the prompt:
18061 * VxWorks Connection:: Connecting to VxWorks
18062 * VxWorks Download:: VxWorks download
18063 * VxWorks Attach:: Running tasks
18066 @node VxWorks Connection
18067 @subsubsection Connecting to VxWorks
18069 The @value{GDBN} command @code{target} lets you connect to a VxWorks target on the
18070 network. To connect to a target whose host name is ``@code{tt}'', type:
18073 (vxgdb) target vxworks tt
18077 @value{GDBN} displays messages like these:
18080 Attaching remote machine across net...
18085 @value{GDBN} then attempts to read the symbol tables of any object modules
18086 loaded into the VxWorks target since it was last booted. @value{GDBN} locates
18087 these files by searching the directories listed in the command search
18088 path (@pxref{Environment, ,Your Program's Environment}); if it fails
18089 to find an object file, it displays a message such as:
18092 prog.o: No such file or directory.
18095 When this happens, add the appropriate directory to the search path with
18096 the @value{GDBN} command @code{path}, and execute the @code{target}
18099 @node VxWorks Download
18100 @subsubsection VxWorks Download
18102 @cindex download to VxWorks
18103 If you have connected to the VxWorks target and you want to debug an
18104 object that has not yet been loaded, you can use the @value{GDBN}
18105 @code{load} command to download a file from Unix to VxWorks
18106 incrementally. The object file given as an argument to the @code{load}
18107 command is actually opened twice: first by the VxWorks target in order
18108 to download the code, then by @value{GDBN} in order to read the symbol
18109 table. This can lead to problems if the current working directories on
18110 the two systems differ. If both systems have NFS mounted the same
18111 filesystems, you can avoid these problems by using absolute paths.
18112 Otherwise, it is simplest to set the working directory on both systems
18113 to the directory in which the object file resides, and then to reference
18114 the file by its name, without any path. For instance, a program
18115 @file{prog.o} may reside in @file{@var{vxpath}/vw/demo/rdb} in VxWorks
18116 and in @file{@var{hostpath}/vw/demo/rdb} on the host. To load this
18117 program, type this on VxWorks:
18120 -> cd "@var{vxpath}/vw/demo/rdb"
18124 Then, in @value{GDBN}, type:
18127 (vxgdb) cd @var{hostpath}/vw/demo/rdb
18128 (vxgdb) load prog.o
18131 @value{GDBN} displays a response similar to this:
18134 Reading symbol data from wherever/vw/demo/rdb/prog.o... done.
18137 You can also use the @code{load} command to reload an object module
18138 after editing and recompiling the corresponding source file. Note that
18139 this makes @value{GDBN} delete all currently-defined breakpoints,
18140 auto-displays, and convenience variables, and to clear the value
18141 history. (This is necessary in order to preserve the integrity of
18142 debugger's data structures that reference the target system's symbol
18145 @node VxWorks Attach
18146 @subsubsection Running Tasks
18148 @cindex running VxWorks tasks
18149 You can also attach to an existing task using the @code{attach} command as
18153 (vxgdb) attach @var{task}
18157 where @var{task} is the VxWorks hexadecimal task ID. The task can be running
18158 or suspended when you attach to it. Running tasks are suspended at
18159 the time of attachment.
18161 @node Embedded Processors
18162 @section Embedded Processors
18164 This section goes into details specific to particular embedded
18167 @cindex send command to simulator
18168 Whenever a specific embedded processor has a simulator, @value{GDBN}
18169 allows to send an arbitrary command to the simulator.
18172 @item sim @var{command}
18173 @kindex sim@r{, a command}
18174 Send an arbitrary @var{command} string to the simulator. Consult the
18175 documentation for the specific simulator in use for information about
18176 acceptable commands.
18182 * M32R/D:: Renesas M32R/D
18183 * M68K:: Motorola M68K
18184 * MicroBlaze:: Xilinx MicroBlaze
18185 * MIPS Embedded:: MIPS Embedded
18186 * OpenRISC 1000:: OpenRisc 1000
18187 * PA:: HP PA Embedded
18188 * PowerPC Embedded:: PowerPC Embedded
18189 * Sparclet:: Tsqware Sparclet
18190 * Sparclite:: Fujitsu Sparclite
18191 * Z8000:: Zilog Z8000
18194 * Super-H:: Renesas Super-H
18203 @item target rdi @var{dev}
18204 ARM Angel monitor, via RDI library interface to ADP protocol. You may
18205 use this target to communicate with both boards running the Angel
18206 monitor, or with the EmbeddedICE JTAG debug device.
18209 @item target rdp @var{dev}
18214 @value{GDBN} provides the following ARM-specific commands:
18217 @item set arm disassembler
18219 This commands selects from a list of disassembly styles. The
18220 @code{"std"} style is the standard style.
18222 @item show arm disassembler
18224 Show the current disassembly style.
18226 @item set arm apcs32
18227 @cindex ARM 32-bit mode
18228 This command toggles ARM operation mode between 32-bit and 26-bit.
18230 @item show arm apcs32
18231 Display the current usage of the ARM 32-bit mode.
18233 @item set arm fpu @var{fputype}
18234 This command sets the ARM floating-point unit (FPU) type. The
18235 argument @var{fputype} can be one of these:
18239 Determine the FPU type by querying the OS ABI.
18241 Software FPU, with mixed-endian doubles on little-endian ARM
18244 GCC-compiled FPA co-processor.
18246 Software FPU with pure-endian doubles.
18252 Show the current type of the FPU.
18255 This command forces @value{GDBN} to use the specified ABI.
18258 Show the currently used ABI.
18260 @item set arm fallback-mode (arm|thumb|auto)
18261 @value{GDBN} uses the symbol table, when available, to determine
18262 whether instructions are ARM or Thumb. This command controls
18263 @value{GDBN}'s default behavior when the symbol table is not
18264 available. The default is @samp{auto}, which causes @value{GDBN} to
18265 use the current execution mode (from the @code{T} bit in the @code{CPSR}
18268 @item show arm fallback-mode
18269 Show the current fallback instruction mode.
18271 @item set arm force-mode (arm|thumb|auto)
18272 This command overrides use of the symbol table to determine whether
18273 instructions are ARM or Thumb. The default is @samp{auto}, which
18274 causes @value{GDBN} to use the symbol table and then the setting
18275 of @samp{set arm fallback-mode}.
18277 @item show arm force-mode
18278 Show the current forced instruction mode.
18280 @item set debug arm
18281 Toggle whether to display ARM-specific debugging messages from the ARM
18282 target support subsystem.
18284 @item show debug arm
18285 Show whether ARM-specific debugging messages are enabled.
18288 The following commands are available when an ARM target is debugged
18289 using the RDI interface:
18292 @item rdilogfile @r{[}@var{file}@r{]}
18294 @cindex ADP (Angel Debugger Protocol) logging
18295 Set the filename for the ADP (Angel Debugger Protocol) packet log.
18296 With an argument, sets the log file to the specified @var{file}. With
18297 no argument, show the current log file name. The default log file is
18300 @item rdilogenable @r{[}@var{arg}@r{]}
18301 @kindex rdilogenable
18302 Control logging of ADP packets. With an argument of 1 or @code{"yes"}
18303 enables logging, with an argument 0 or @code{"no"} disables it. With
18304 no arguments displays the current setting. When logging is enabled,
18305 ADP packets exchanged between @value{GDBN} and the RDI target device
18306 are logged to a file.
18308 @item set rdiromatzero
18309 @kindex set rdiromatzero
18310 @cindex ROM at zero address, RDI
18311 Tell @value{GDBN} whether the target has ROM at address 0. If on,
18312 vector catching is disabled, so that zero address can be used. If off
18313 (the default), vector catching is enabled. For this command to take
18314 effect, it needs to be invoked prior to the @code{target rdi} command.
18316 @item show rdiromatzero
18317 @kindex show rdiromatzero
18318 Show the current setting of ROM at zero address.
18320 @item set rdiheartbeat
18321 @kindex set rdiheartbeat
18322 @cindex RDI heartbeat
18323 Enable or disable RDI heartbeat packets. It is not recommended to
18324 turn on this option, since it confuses ARM and EPI JTAG interface, as
18325 well as the Angel monitor.
18327 @item show rdiheartbeat
18328 @kindex show rdiheartbeat
18329 Show the setting of RDI heartbeat packets.
18333 @item target sim @r{[}@var{simargs}@r{]} @dots{}
18334 The @value{GDBN} ARM simulator accepts the following optional arguments.
18337 @item --swi-support=@var{type}
18338 Tell the simulator which SWI interfaces to support.
18339 @var{type} may be a comma separated list of the following values.
18340 The default value is @code{all}.
18353 @subsection Renesas M32R/D and M32R/SDI
18356 @kindex target m32r
18357 @item target m32r @var{dev}
18358 Renesas M32R/D ROM monitor.
18360 @kindex target m32rsdi
18361 @item target m32rsdi @var{dev}
18362 Renesas M32R SDI server, connected via parallel port to the board.
18365 The following @value{GDBN} commands are specific to the M32R monitor:
18368 @item set download-path @var{path}
18369 @kindex set download-path
18370 @cindex find downloadable @sc{srec} files (M32R)
18371 Set the default path for finding downloadable @sc{srec} files.
18373 @item show download-path
18374 @kindex show download-path
18375 Show the default path for downloadable @sc{srec} files.
18377 @item set board-address @var{addr}
18378 @kindex set board-address
18379 @cindex M32-EVA target board address
18380 Set the IP address for the M32R-EVA target board.
18382 @item show board-address
18383 @kindex show board-address
18384 Show the current IP address of the target board.
18386 @item set server-address @var{addr}
18387 @kindex set server-address
18388 @cindex download server address (M32R)
18389 Set the IP address for the download server, which is the @value{GDBN}'s
18392 @item show server-address
18393 @kindex show server-address
18394 Display the IP address of the download server.
18396 @item upload @r{[}@var{file}@r{]}
18397 @kindex upload@r{, M32R}
18398 Upload the specified @sc{srec} @var{file} via the monitor's Ethernet
18399 upload capability. If no @var{file} argument is given, the current
18400 executable file is uploaded.
18402 @item tload @r{[}@var{file}@r{]}
18403 @kindex tload@r{, M32R}
18404 Test the @code{upload} command.
18407 The following commands are available for M32R/SDI:
18412 @cindex reset SDI connection, M32R
18413 This command resets the SDI connection.
18417 This command shows the SDI connection status.
18420 @kindex debug_chaos
18421 @cindex M32R/Chaos debugging
18422 Instructs the remote that M32R/Chaos debugging is to be used.
18424 @item use_debug_dma
18425 @kindex use_debug_dma
18426 Instructs the remote to use the DEBUG_DMA method of accessing memory.
18429 @kindex use_mon_code
18430 Instructs the remote to use the MON_CODE method of accessing memory.
18433 @kindex use_ib_break
18434 Instructs the remote to set breakpoints by IB break.
18436 @item use_dbt_break
18437 @kindex use_dbt_break
18438 Instructs the remote to set breakpoints by DBT.
18444 The Motorola m68k configuration includes ColdFire support, and a
18445 target command for the following ROM monitor.
18449 @kindex target dbug
18450 @item target dbug @var{dev}
18451 dBUG ROM monitor for Motorola ColdFire.
18456 @subsection MicroBlaze
18457 @cindex Xilinx MicroBlaze
18458 @cindex XMD, Xilinx Microprocessor Debugger
18460 The MicroBlaze is a soft-core processor supported on various Xilinx
18461 FPGAs, such as Spartan or Virtex series. Boards with these processors
18462 usually have JTAG ports which connect to a host system running the Xilinx
18463 Embedded Development Kit (EDK) or Software Development Kit (SDK).
18464 This host system is used to download the configuration bitstream to
18465 the target FPGA. The Xilinx Microprocessor Debugger (XMD) program
18466 communicates with the target board using the JTAG interface and
18467 presents a @code{gdbserver} interface to the board. By default
18468 @code{xmd} uses port @code{1234}. (While it is possible to change
18469 this default port, it requires the use of undocumented @code{xmd}
18470 commands. Contact Xilinx support if you need to do this.)
18472 Use these GDB commands to connect to the MicroBlaze target processor.
18475 @item target remote :1234
18476 Use this command to connect to the target if you are running @value{GDBN}
18477 on the same system as @code{xmd}.
18479 @item target remote @var{xmd-host}:1234
18480 Use this command to connect to the target if it is connected to @code{xmd}
18481 running on a different system named @var{xmd-host}.
18484 Use this command to download a program to the MicroBlaze target.
18486 @item set debug microblaze @var{n}
18487 Enable MicroBlaze-specific debugging messages if non-zero.
18489 @item show debug microblaze @var{n}
18490 Show MicroBlaze-specific debugging level.
18493 @node MIPS Embedded
18494 @subsection MIPS Embedded
18496 @cindex MIPS boards
18497 @value{GDBN} can use the MIPS remote debugging protocol to talk to a
18498 MIPS board attached to a serial line. This is available when
18499 you configure @value{GDBN} with @samp{--target=mips-idt-ecoff}.
18502 Use these @value{GDBN} commands to specify the connection to your target board:
18505 @item target mips @var{port}
18506 @kindex target mips @var{port}
18507 To run a program on the board, start up @code{@value{GDBP}} with the
18508 name of your program as the argument. To connect to the board, use the
18509 command @samp{target mips @var{port}}, where @var{port} is the name of
18510 the serial port connected to the board. If the program has not already
18511 been downloaded to the board, you may use the @code{load} command to
18512 download it. You can then use all the usual @value{GDBN} commands.
18514 For example, this sequence connects to the target board through a serial
18515 port, and loads and runs a program called @var{prog} through the
18519 host$ @value{GDBP} @var{prog}
18520 @value{GDBN} is free software and @dots{}
18521 (@value{GDBP}) target mips /dev/ttyb
18522 (@value{GDBP}) load @var{prog}
18526 @item target mips @var{hostname}:@var{portnumber}
18527 On some @value{GDBN} host configurations, you can specify a TCP
18528 connection (for instance, to a serial line managed by a terminal
18529 concentrator) instead of a serial port, using the syntax
18530 @samp{@var{hostname}:@var{portnumber}}.
18532 @item target pmon @var{port}
18533 @kindex target pmon @var{port}
18536 @item target ddb @var{port}
18537 @kindex target ddb @var{port}
18538 NEC's DDB variant of PMON for Vr4300.
18540 @item target lsi @var{port}
18541 @kindex target lsi @var{port}
18542 LSI variant of PMON.
18544 @kindex target r3900
18545 @item target r3900 @var{dev}
18546 Densan DVE-R3900 ROM monitor for Toshiba R3900 Mips.
18548 @kindex target array
18549 @item target array @var{dev}
18550 Array Tech LSI33K RAID controller board.
18556 @value{GDBN} also supports these special commands for MIPS targets:
18559 @item set mipsfpu double
18560 @itemx set mipsfpu single
18561 @itemx set mipsfpu none
18562 @itemx set mipsfpu auto
18563 @itemx show mipsfpu
18564 @kindex set mipsfpu
18565 @kindex show mipsfpu
18566 @cindex MIPS remote floating point
18567 @cindex floating point, MIPS remote
18568 If your target board does not support the MIPS floating point
18569 coprocessor, you should use the command @samp{set mipsfpu none} (if you
18570 need this, you may wish to put the command in your @value{GDBN} init
18571 file). This tells @value{GDBN} how to find the return value of
18572 functions which return floating point values. It also allows
18573 @value{GDBN} to avoid saving the floating point registers when calling
18574 functions on the board. If you are using a floating point coprocessor
18575 with only single precision floating point support, as on the @sc{r4650}
18576 processor, use the command @samp{set mipsfpu single}. The default
18577 double precision floating point coprocessor may be selected using
18578 @samp{set mipsfpu double}.
18580 In previous versions the only choices were double precision or no
18581 floating point, so @samp{set mipsfpu on} will select double precision
18582 and @samp{set mipsfpu off} will select no floating point.
18584 As usual, you can inquire about the @code{mipsfpu} variable with
18585 @samp{show mipsfpu}.
18587 @item set timeout @var{seconds}
18588 @itemx set retransmit-timeout @var{seconds}
18589 @itemx show timeout
18590 @itemx show retransmit-timeout
18591 @cindex @code{timeout}, MIPS protocol
18592 @cindex @code{retransmit-timeout}, MIPS protocol
18593 @kindex set timeout
18594 @kindex show timeout
18595 @kindex set retransmit-timeout
18596 @kindex show retransmit-timeout
18597 You can control the timeout used while waiting for a packet, in the MIPS
18598 remote protocol, with the @code{set timeout @var{seconds}} command. The
18599 default is 5 seconds. Similarly, you can control the timeout used while
18600 waiting for an acknowledgment of a packet with the @code{set
18601 retransmit-timeout @var{seconds}} command. The default is 3 seconds.
18602 You can inspect both values with @code{show timeout} and @code{show
18603 retransmit-timeout}. (These commands are @emph{only} available when
18604 @value{GDBN} is configured for @samp{--target=mips-idt-ecoff}.)
18606 The timeout set by @code{set timeout} does not apply when @value{GDBN}
18607 is waiting for your program to stop. In that case, @value{GDBN} waits
18608 forever because it has no way of knowing how long the program is going
18609 to run before stopping.
18611 @item set syn-garbage-limit @var{num}
18612 @kindex set syn-garbage-limit@r{, MIPS remote}
18613 @cindex synchronize with remote MIPS target
18614 Limit the maximum number of characters @value{GDBN} should ignore when
18615 it tries to synchronize with the remote target. The default is 10
18616 characters. Setting the limit to -1 means there's no limit.
18618 @item show syn-garbage-limit
18619 @kindex show syn-garbage-limit@r{, MIPS remote}
18620 Show the current limit on the number of characters to ignore when
18621 trying to synchronize with the remote system.
18623 @item set monitor-prompt @var{prompt}
18624 @kindex set monitor-prompt@r{, MIPS remote}
18625 @cindex remote monitor prompt
18626 Tell @value{GDBN} to expect the specified @var{prompt} string from the
18627 remote monitor. The default depends on the target:
18637 @item show monitor-prompt
18638 @kindex show monitor-prompt@r{, MIPS remote}
18639 Show the current strings @value{GDBN} expects as the prompt from the
18642 @item set monitor-warnings
18643 @kindex set monitor-warnings@r{, MIPS remote}
18644 Enable or disable monitor warnings about hardware breakpoints. This
18645 has effect only for the @code{lsi} target. When on, @value{GDBN} will
18646 display warning messages whose codes are returned by the @code{lsi}
18647 PMON monitor for breakpoint commands.
18649 @item show monitor-warnings
18650 @kindex show monitor-warnings@r{, MIPS remote}
18651 Show the current setting of printing monitor warnings.
18653 @item pmon @var{command}
18654 @kindex pmon@r{, MIPS remote}
18655 @cindex send PMON command
18656 This command allows sending an arbitrary @var{command} string to the
18657 monitor. The monitor must be in debug mode for this to work.
18660 @node OpenRISC 1000
18661 @subsection OpenRISC 1000
18662 @cindex OpenRISC 1000
18664 @cindex or1k boards
18665 See OR1k Architecture document (@uref{www.opencores.org}) for more information
18666 about platform and commands.
18670 @kindex target jtag
18671 @item target jtag jtag://@var{host}:@var{port}
18673 Connects to remote JTAG server.
18674 JTAG remote server can be either an or1ksim or JTAG server,
18675 connected via parallel port to the board.
18677 Example: @code{target jtag jtag://localhost:9999}
18680 @item or1ksim @var{command}
18681 If connected to @code{or1ksim} OpenRISC 1000 Architectural
18682 Simulator, proprietary commands can be executed.
18684 @kindex info or1k spr
18685 @item info or1k spr
18686 Displays spr groups.
18688 @item info or1k spr @var{group}
18689 @itemx info or1k spr @var{groupno}
18690 Displays register names in selected group.
18692 @item info or1k spr @var{group} @var{register}
18693 @itemx info or1k spr @var{register}
18694 @itemx info or1k spr @var{groupno} @var{registerno}
18695 @itemx info or1k spr @var{registerno}
18696 Shows information about specified spr register.
18699 @item spr @var{group} @var{register} @var{value}
18700 @itemx spr @var{register @var{value}}
18701 @itemx spr @var{groupno} @var{registerno @var{value}}
18702 @itemx spr @var{registerno @var{value}}
18703 Writes @var{value} to specified spr register.
18706 Some implementations of OpenRISC 1000 Architecture also have hardware trace.
18707 It is very similar to @value{GDBN} trace, except it does not interfere with normal
18708 program execution and is thus much faster. Hardware breakpoints/watchpoint
18709 triggers can be set using:
18712 Load effective address/data
18714 Store effective address/data
18716 Access effective address ($SEA or $LEA) or data ($SDATA/$LDATA)
18721 When triggered, it can capture low level data, like: @code{PC}, @code{LSEA},
18722 @code{LDATA}, @code{SDATA}, @code{READSPR}, @code{WRITESPR}, @code{INSTR}.
18724 @code{htrace} commands:
18725 @cindex OpenRISC 1000 htrace
18728 @item hwatch @var{conditional}
18729 Set hardware watchpoint on combination of Load/Store Effective Address(es)
18730 or Data. For example:
18732 @code{hwatch ($LEA == my_var) && ($LDATA < 50) || ($SEA == my_var) && ($SDATA >= 50)}
18734 @code{hwatch ($LEA == my_var) && ($LDATA < 50) || ($SEA == my_var) && ($SDATA >= 50)}
18738 Display information about current HW trace configuration.
18740 @item htrace trigger @var{conditional}
18741 Set starting criteria for HW trace.
18743 @item htrace qualifier @var{conditional}
18744 Set acquisition qualifier for HW trace.
18746 @item htrace stop @var{conditional}
18747 Set HW trace stopping criteria.
18749 @item htrace record [@var{data}]*
18750 Selects the data to be recorded, when qualifier is met and HW trace was
18753 @item htrace enable
18754 @itemx htrace disable
18755 Enables/disables the HW trace.
18757 @item htrace rewind [@var{filename}]
18758 Clears currently recorded trace data.
18760 If filename is specified, new trace file is made and any newly collected data
18761 will be written there.
18763 @item htrace print [@var{start} [@var{len}]]
18764 Prints trace buffer, using current record configuration.
18766 @item htrace mode continuous
18767 Set continuous trace mode.
18769 @item htrace mode suspend
18770 Set suspend trace mode.
18774 @node PowerPC Embedded
18775 @subsection PowerPC Embedded
18777 @cindex DVC register
18778 @value{GDBN} supports using the DVC (Data Value Compare) register to
18779 implement in hardware simple hardware watchpoint conditions of the form:
18782 (@value{GDBP}) watch @var{ADDRESS|VARIABLE} \
18783 if @var{ADDRESS|VARIABLE} == @var{CONSTANT EXPRESSION}
18786 The DVC register will be automatically used when @value{GDBN} detects
18787 such pattern in a condition expression, and the created watchpoint uses one
18788 debug register (either the @code{exact-watchpoints} option is on and the
18789 variable is scalar, or the variable has a length of one byte). This feature
18790 is available in native @value{GDBN} running on a Linux kernel version 2.6.34
18793 When running on PowerPC embedded processors, @value{GDBN} automatically uses
18794 ranged hardware watchpoints, unless the @code{exact-watchpoints} option is on,
18795 in which case watchpoints using only one debug register are created when
18796 watching variables of scalar types.
18798 You can create an artificial array to watch an arbitrary memory
18799 region using one of the following commands (@pxref{Expressions}):
18802 (@value{GDBP}) watch *((char *) @var{address})@@@var{length}
18803 (@value{GDBP}) watch @{char[@var{length}]@} @var{address}
18806 PowerPC embedded processors support masked watchpoints. See the discussion
18807 about the @code{mask} argument in @ref{Set Watchpoints}.
18809 @cindex ranged breakpoint
18810 PowerPC embedded processors support hardware accelerated
18811 @dfn{ranged breakpoints}. A ranged breakpoint stops execution of
18812 the inferior whenever it executes an instruction at any address within
18813 the range it specifies. To set a ranged breakpoint in @value{GDBN},
18814 use the @code{break-range} command.
18816 @value{GDBN} provides the following PowerPC-specific commands:
18819 @kindex break-range
18820 @item break-range @var{start-location}, @var{end-location}
18821 Set a breakpoint for an address range.
18822 @var{start-location} and @var{end-location} can specify a function name,
18823 a line number, an offset of lines from the current line or from the start
18824 location, or an address of an instruction (see @ref{Specify Location},
18825 for a list of all the possible ways to specify a @var{location}.)
18826 The breakpoint will stop execution of the inferior whenever it
18827 executes an instruction at any address within the specified range,
18828 (including @var{start-location} and @var{end-location}.)
18830 @kindex set powerpc
18831 @item set powerpc soft-float
18832 @itemx show powerpc soft-float
18833 Force @value{GDBN} to use (or not use) a software floating point calling
18834 convention. By default, @value{GDBN} selects the calling convention based
18835 on the selected architecture and the provided executable file.
18837 @item set powerpc vector-abi
18838 @itemx show powerpc vector-abi
18839 Force @value{GDBN} to use the specified calling convention for vector
18840 arguments and return values. The valid options are @samp{auto};
18841 @samp{generic}, to avoid vector registers even if they are present;
18842 @samp{altivec}, to use AltiVec registers; and @samp{spe} to use SPE
18843 registers. By default, @value{GDBN} selects the calling convention
18844 based on the selected architecture and the provided executable file.
18846 @item set powerpc exact-watchpoints
18847 @itemx show powerpc exact-watchpoints
18848 Allow @value{GDBN} to use only one debug register when watching a variable
18849 of scalar type, thus assuming that the variable is accessed through the
18850 address of its first byte.
18852 @kindex target dink32
18853 @item target dink32 @var{dev}
18854 DINK32 ROM monitor.
18856 @kindex target ppcbug
18857 @item target ppcbug @var{dev}
18858 @kindex target ppcbug1
18859 @item target ppcbug1 @var{dev}
18860 PPCBUG ROM monitor for PowerPC.
18863 @item target sds @var{dev}
18864 SDS monitor, running on a PowerPC board (such as Motorola's ADS).
18867 @cindex SDS protocol
18868 The following commands specific to the SDS protocol are supported
18872 @item set sdstimeout @var{nsec}
18873 @kindex set sdstimeout
18874 Set the timeout for SDS protocol reads to be @var{nsec} seconds. The
18875 default is 2 seconds.
18877 @item show sdstimeout
18878 @kindex show sdstimeout
18879 Show the current value of the SDS timeout.
18881 @item sds @var{command}
18882 @kindex sds@r{, a command}
18883 Send the specified @var{command} string to the SDS monitor.
18888 @subsection HP PA Embedded
18892 @kindex target op50n
18893 @item target op50n @var{dev}
18894 OP50N monitor, running on an OKI HPPA board.
18896 @kindex target w89k
18897 @item target w89k @var{dev}
18898 W89K monitor, running on a Winbond HPPA board.
18903 @subsection Tsqware Sparclet
18907 @value{GDBN} enables developers to debug tasks running on
18908 Sparclet targets from a Unix host.
18909 @value{GDBN} uses code that runs on
18910 both the Unix host and on the Sparclet target. The program
18911 @code{@value{GDBP}} is installed and executed on the Unix host.
18914 @item remotetimeout @var{args}
18915 @kindex remotetimeout
18916 @value{GDBN} supports the option @code{remotetimeout}.
18917 This option is set by the user, and @var{args} represents the number of
18918 seconds @value{GDBN} waits for responses.
18921 @cindex compiling, on Sparclet
18922 When compiling for debugging, include the options @samp{-g} to get debug
18923 information and @samp{-Ttext} to relocate the program to where you wish to
18924 load it on the target. You may also want to add the options @samp{-n} or
18925 @samp{-N} in order to reduce the size of the sections. Example:
18928 sparclet-aout-gcc prog.c -Ttext 0x12010000 -g -o prog -N
18931 You can use @code{objdump} to verify that the addresses are what you intended:
18934 sparclet-aout-objdump --headers --syms prog
18937 @cindex running, on Sparclet
18939 your Unix execution search path to find @value{GDBN}, you are ready to
18940 run @value{GDBN}. From your Unix host, run @code{@value{GDBP}}
18941 (or @code{sparclet-aout-gdb}, depending on your installation).
18943 @value{GDBN} comes up showing the prompt:
18950 * Sparclet File:: Setting the file to debug
18951 * Sparclet Connection:: Connecting to Sparclet
18952 * Sparclet Download:: Sparclet download
18953 * Sparclet Execution:: Running and debugging
18956 @node Sparclet File
18957 @subsubsection Setting File to Debug
18959 The @value{GDBN} command @code{file} lets you choose with program to debug.
18962 (gdbslet) file prog
18966 @value{GDBN} then attempts to read the symbol table of @file{prog}.
18967 @value{GDBN} locates
18968 the file by searching the directories listed in the command search
18970 If the file was compiled with debug information (option @samp{-g}), source
18971 files will be searched as well.
18972 @value{GDBN} locates
18973 the source files by searching the directories listed in the directory search
18974 path (@pxref{Environment, ,Your Program's Environment}).
18976 to find a file, it displays a message such as:
18979 prog: No such file or directory.
18982 When this happens, add the appropriate directories to the search paths with
18983 the @value{GDBN} commands @code{path} and @code{dir}, and execute the
18984 @code{target} command again.
18986 @node Sparclet Connection
18987 @subsubsection Connecting to Sparclet
18989 The @value{GDBN} command @code{target} lets you connect to a Sparclet target.
18990 To connect to a target on serial port ``@code{ttya}'', type:
18993 (gdbslet) target sparclet /dev/ttya
18994 Remote target sparclet connected to /dev/ttya
18995 main () at ../prog.c:3
18999 @value{GDBN} displays messages like these:
19005 @node Sparclet Download
19006 @subsubsection Sparclet Download
19008 @cindex download to Sparclet
19009 Once connected to the Sparclet target,
19010 you can use the @value{GDBN}
19011 @code{load} command to download the file from the host to the target.
19012 The file name and load offset should be given as arguments to the @code{load}
19014 Since the file format is aout, the program must be loaded to the starting
19015 address. You can use @code{objdump} to find out what this value is. The load
19016 offset is an offset which is added to the VMA (virtual memory address)
19017 of each of the file's sections.
19018 For instance, if the program
19019 @file{prog} was linked to text address 0x1201000, with data at 0x12010160
19020 and bss at 0x12010170, in @value{GDBN}, type:
19023 (gdbslet) load prog 0x12010000
19024 Loading section .text, size 0xdb0 vma 0x12010000
19027 If the code is loaded at a different address then what the program was linked
19028 to, you may need to use the @code{section} and @code{add-symbol-file} commands
19029 to tell @value{GDBN} where to map the symbol table.
19031 @node Sparclet Execution
19032 @subsubsection Running and Debugging
19034 @cindex running and debugging Sparclet programs
19035 You can now begin debugging the task using @value{GDBN}'s execution control
19036 commands, @code{b}, @code{step}, @code{run}, etc. See the @value{GDBN}
19037 manual for the list of commands.
19041 Breakpoint 1 at 0x12010000: file prog.c, line 3.
19043 Starting program: prog
19044 Breakpoint 1, main (argc=1, argv=0xeffff21c) at prog.c:3
19045 3 char *symarg = 0;
19047 4 char *execarg = "hello!";
19052 @subsection Fujitsu Sparclite
19056 @kindex target sparclite
19057 @item target sparclite @var{dev}
19058 Fujitsu sparclite boards, used only for the purpose of loading.
19059 You must use an additional command to debug the program.
19060 For example: target remote @var{dev} using @value{GDBN} standard
19066 @subsection Zilog Z8000
19069 @cindex simulator, Z8000
19070 @cindex Zilog Z8000 simulator
19072 When configured for debugging Zilog Z8000 targets, @value{GDBN} includes
19075 For the Z8000 family, @samp{target sim} simulates either the Z8002 (the
19076 unsegmented variant of the Z8000 architecture) or the Z8001 (the
19077 segmented variant). The simulator recognizes which architecture is
19078 appropriate by inspecting the object code.
19081 @item target sim @var{args}
19083 @kindex target sim@r{, with Z8000}
19084 Debug programs on a simulated CPU. If the simulator supports setup
19085 options, specify them via @var{args}.
19089 After specifying this target, you can debug programs for the simulated
19090 CPU in the same style as programs for your host computer; use the
19091 @code{file} command to load a new program image, the @code{run} command
19092 to run your program, and so on.
19094 As well as making available all the usual machine registers
19095 (@pxref{Registers, ,Registers}), the Z8000 simulator provides three
19096 additional items of information as specially named registers:
19101 Counts clock-ticks in the simulator.
19104 Counts instructions run in the simulator.
19107 Execution time in 60ths of a second.
19111 You can refer to these values in @value{GDBN} expressions with the usual
19112 conventions; for example, @w{@samp{b fputc if $cycles>5000}} sets a
19113 conditional breakpoint that suspends only after at least 5000
19114 simulated clock ticks.
19117 @subsection Atmel AVR
19120 When configured for debugging the Atmel AVR, @value{GDBN} supports the
19121 following AVR-specific commands:
19124 @item info io_registers
19125 @kindex info io_registers@r{, AVR}
19126 @cindex I/O registers (Atmel AVR)
19127 This command displays information about the AVR I/O registers. For
19128 each register, @value{GDBN} prints its number and value.
19135 When configured for debugging CRIS, @value{GDBN} provides the
19136 following CRIS-specific commands:
19139 @item set cris-version @var{ver}
19140 @cindex CRIS version
19141 Set the current CRIS version to @var{ver}, either @samp{10} or @samp{32}.
19142 The CRIS version affects register names and sizes. This command is useful in
19143 case autodetection of the CRIS version fails.
19145 @item show cris-version
19146 Show the current CRIS version.
19148 @item set cris-dwarf2-cfi
19149 @cindex DWARF-2 CFI and CRIS
19150 Set the usage of DWARF-2 CFI for CRIS debugging. The default is @samp{on}.
19151 Change to @samp{off} when using @code{gcc-cris} whose version is below
19154 @item show cris-dwarf2-cfi
19155 Show the current state of using DWARF-2 CFI.
19157 @item set cris-mode @var{mode}
19159 Set the current CRIS mode to @var{mode}. It should only be changed when
19160 debugging in guru mode, in which case it should be set to
19161 @samp{guru} (the default is @samp{normal}).
19163 @item show cris-mode
19164 Show the current CRIS mode.
19168 @subsection Renesas Super-H
19171 For the Renesas Super-H processor, @value{GDBN} provides these
19176 @kindex regs@r{, Super-H}
19177 Show the values of all Super-H registers.
19179 @item set sh calling-convention @var{convention}
19180 @kindex set sh calling-convention
19181 Set the calling-convention used when calling functions from @value{GDBN}.
19182 Allowed values are @samp{gcc}, which is the default setting, and @samp{renesas}.
19183 With the @samp{gcc} setting, functions are called using the @value{NGCC} calling
19184 convention. If the DWARF-2 information of the called function specifies
19185 that the function follows the Renesas calling convention, the function
19186 is called using the Renesas calling convention. If the calling convention
19187 is set to @samp{renesas}, the Renesas calling convention is always used,
19188 regardless of the DWARF-2 information. This can be used to override the
19189 default of @samp{gcc} if debug information is missing, or the compiler
19190 does not emit the DWARF-2 calling convention entry for a function.
19192 @item show sh calling-convention
19193 @kindex show sh calling-convention
19194 Show the current calling convention setting.
19199 @node Architectures
19200 @section Architectures
19202 This section describes characteristics of architectures that affect
19203 all uses of @value{GDBN} with the architecture, both native and cross.
19210 * HPPA:: HP PA architecture
19211 * SPU:: Cell Broadband Engine SPU architecture
19216 @subsection x86 Architecture-specific Issues
19219 @item set struct-convention @var{mode}
19220 @kindex set struct-convention
19221 @cindex struct return convention
19222 @cindex struct/union returned in registers
19223 Set the convention used by the inferior to return @code{struct}s and
19224 @code{union}s from functions to @var{mode}. Possible values of
19225 @var{mode} are @code{"pcc"}, @code{"reg"}, and @code{"default"} (the
19226 default). @code{"default"} or @code{"pcc"} means that @code{struct}s
19227 are returned on the stack, while @code{"reg"} means that a
19228 @code{struct} or a @code{union} whose size is 1, 2, 4, or 8 bytes will
19229 be returned in a register.
19231 @item show struct-convention
19232 @kindex show struct-convention
19233 Show the current setting of the convention to return @code{struct}s
19242 @kindex set rstack_high_address
19243 @cindex AMD 29K register stack
19244 @cindex register stack, AMD29K
19245 @item set rstack_high_address @var{address}
19246 On AMD 29000 family processors, registers are saved in a separate
19247 @dfn{register stack}. There is no way for @value{GDBN} to determine the
19248 extent of this stack. Normally, @value{GDBN} just assumes that the
19249 stack is ``large enough''. This may result in @value{GDBN} referencing
19250 memory locations that do not exist. If necessary, you can get around
19251 this problem by specifying the ending address of the register stack with
19252 the @code{set rstack_high_address} command. The argument should be an
19253 address, which you probably want to precede with @samp{0x} to specify in
19256 @kindex show rstack_high_address
19257 @item show rstack_high_address
19258 Display the current limit of the register stack, on AMD 29000 family
19266 See the following section.
19271 @cindex stack on Alpha
19272 @cindex stack on MIPS
19273 @cindex Alpha stack
19275 Alpha- and MIPS-based computers use an unusual stack frame, which
19276 sometimes requires @value{GDBN} to search backward in the object code to
19277 find the beginning of a function.
19279 @cindex response time, MIPS debugging
19280 To improve response time (especially for embedded applications, where
19281 @value{GDBN} may be restricted to a slow serial line for this search)
19282 you may want to limit the size of this search, using one of these
19286 @cindex @code{heuristic-fence-post} (Alpha, MIPS)
19287 @item set heuristic-fence-post @var{limit}
19288 Restrict @value{GDBN} to examining at most @var{limit} bytes in its
19289 search for the beginning of a function. A value of @var{0} (the
19290 default) means there is no limit. However, except for @var{0}, the
19291 larger the limit the more bytes @code{heuristic-fence-post} must search
19292 and therefore the longer it takes to run. You should only need to use
19293 this command when debugging a stripped executable.
19295 @item show heuristic-fence-post
19296 Display the current limit.
19300 These commands are available @emph{only} when @value{GDBN} is configured
19301 for debugging programs on Alpha or MIPS processors.
19303 Several MIPS-specific commands are available when debugging MIPS
19307 @item set mips abi @var{arg}
19308 @kindex set mips abi
19309 @cindex set ABI for MIPS
19310 Tell @value{GDBN} which MIPS ABI is used by the inferior. Possible
19311 values of @var{arg} are:
19315 The default ABI associated with the current binary (this is the
19326 @item show mips abi
19327 @kindex show mips abi
19328 Show the MIPS ABI used by @value{GDBN} to debug the inferior.
19331 @itemx show mipsfpu
19332 @xref{MIPS Embedded, set mipsfpu}.
19334 @item set mips mask-address @var{arg}
19335 @kindex set mips mask-address
19336 @cindex MIPS addresses, masking
19337 This command determines whether the most-significant 32 bits of 64-bit
19338 MIPS addresses are masked off. The argument @var{arg} can be
19339 @samp{on}, @samp{off}, or @samp{auto}. The latter is the default
19340 setting, which lets @value{GDBN} determine the correct value.
19342 @item show mips mask-address
19343 @kindex show mips mask-address
19344 Show whether the upper 32 bits of MIPS addresses are masked off or
19347 @item set remote-mips64-transfers-32bit-regs
19348 @kindex set remote-mips64-transfers-32bit-regs
19349 This command controls compatibility with 64-bit MIPS targets that
19350 transfer data in 32-bit quantities. If you have an old MIPS 64 target
19351 that transfers 32 bits for some registers, like @sc{sr} and @sc{fsr},
19352 and 64 bits for other registers, set this option to @samp{on}.
19354 @item show remote-mips64-transfers-32bit-regs
19355 @kindex show remote-mips64-transfers-32bit-regs
19356 Show the current setting of compatibility with older MIPS 64 targets.
19358 @item set debug mips
19359 @kindex set debug mips
19360 This command turns on and off debugging messages for the MIPS-specific
19361 target code in @value{GDBN}.
19363 @item show debug mips
19364 @kindex show debug mips
19365 Show the current setting of MIPS debugging messages.
19371 @cindex HPPA support
19373 When @value{GDBN} is debugging the HP PA architecture, it provides the
19374 following special commands:
19377 @item set debug hppa
19378 @kindex set debug hppa
19379 This command determines whether HPPA architecture-specific debugging
19380 messages are to be displayed.
19382 @item show debug hppa
19383 Show whether HPPA debugging messages are displayed.
19385 @item maint print unwind @var{address}
19386 @kindex maint print unwind@r{, HPPA}
19387 This command displays the contents of the unwind table entry at the
19388 given @var{address}.
19394 @subsection Cell Broadband Engine SPU architecture
19395 @cindex Cell Broadband Engine
19398 When @value{GDBN} is debugging the Cell Broadband Engine SPU architecture,
19399 it provides the following special commands:
19402 @item info spu event
19404 Display SPU event facility status. Shows current event mask
19405 and pending event status.
19407 @item info spu signal
19408 Display SPU signal notification facility status. Shows pending
19409 signal-control word and signal notification mode of both signal
19410 notification channels.
19412 @item info spu mailbox
19413 Display SPU mailbox facility status. Shows all pending entries,
19414 in order of processing, in each of the SPU Write Outbound,
19415 SPU Write Outbound Interrupt, and SPU Read Inbound mailboxes.
19418 Display MFC DMA status. Shows all pending commands in the MFC
19419 DMA queue. For each entry, opcode, tag, class IDs, effective
19420 and local store addresses and transfer size are shown.
19422 @item info spu proxydma
19423 Display MFC Proxy-DMA status. Shows all pending commands in the MFC
19424 Proxy-DMA queue. For each entry, opcode, tag, class IDs, effective
19425 and local store addresses and transfer size are shown.
19429 When @value{GDBN} is debugging a combined PowerPC/SPU application
19430 on the Cell Broadband Engine, it provides in addition the following
19434 @item set spu stop-on-load @var{arg}
19436 Set whether to stop for new SPE threads. When set to @code{on}, @value{GDBN}
19437 will give control to the user when a new SPE thread enters its @code{main}
19438 function. The default is @code{off}.
19440 @item show spu stop-on-load
19442 Show whether to stop for new SPE threads.
19444 @item set spu auto-flush-cache @var{arg}
19445 Set whether to automatically flush the software-managed cache. When set to
19446 @code{on}, @value{GDBN} will automatically cause the SPE software-managed
19447 cache to be flushed whenever SPE execution stops. This provides a consistent
19448 view of PowerPC memory that is accessed via the cache. If an application
19449 does not use the software-managed cache, this option has no effect.
19451 @item show spu auto-flush-cache
19452 Show whether to automatically flush the software-managed cache.
19457 @subsection PowerPC
19458 @cindex PowerPC architecture
19460 When @value{GDBN} is debugging the PowerPC architecture, it provides a set of
19461 pseudo-registers to enable inspection of 128-bit wide Decimal Floating Point
19462 numbers stored in the floating point registers. These values must be stored
19463 in two consecutive registers, always starting at an even register like
19464 @code{f0} or @code{f2}.
19466 The pseudo-registers go from @code{$dl0} through @code{$dl15}, and are formed
19467 by joining the even/odd register pairs @code{f0} and @code{f1} for @code{$dl0},
19468 @code{f2} and @code{f3} for @code{$dl1} and so on.
19470 For POWER7 processors, @value{GDBN} provides a set of pseudo-registers, the 64-bit
19471 wide Extended Floating Point Registers (@samp{f32} through @samp{f63}).
19474 @node Controlling GDB
19475 @chapter Controlling @value{GDBN}
19477 You can alter the way @value{GDBN} interacts with you by using the
19478 @code{set} command. For commands controlling how @value{GDBN} displays
19479 data, see @ref{Print Settings, ,Print Settings}. Other settings are
19484 * Editing:: Command editing
19485 * Command History:: Command history
19486 * Screen Size:: Screen size
19487 * Numbers:: Numbers
19488 * ABI:: Configuring the current ABI
19489 * Messages/Warnings:: Optional warnings and messages
19490 * Debugging Output:: Optional messages about internal happenings
19491 * Other Misc Settings:: Other Miscellaneous Settings
19499 @value{GDBN} indicates its readiness to read a command by printing a string
19500 called the @dfn{prompt}. This string is normally @samp{(@value{GDBP})}. You
19501 can change the prompt string with the @code{set prompt} command. For
19502 instance, when debugging @value{GDBN} with @value{GDBN}, it is useful to change
19503 the prompt in one of the @value{GDBN} sessions so that you can always tell
19504 which one you are talking to.
19506 @emph{Note:} @code{set prompt} does not add a space for you after the
19507 prompt you set. This allows you to set a prompt which ends in a space
19508 or a prompt that does not.
19512 @item set prompt @var{newprompt}
19513 Directs @value{GDBN} to use @var{newprompt} as its prompt string henceforth.
19515 @kindex show prompt
19517 Prints a line of the form: @samp{Gdb's prompt is: @var{your-prompt}}
19521 @section Command Editing
19523 @cindex command line editing
19525 @value{GDBN} reads its input commands via the @dfn{Readline} interface. This
19526 @sc{gnu} library provides consistent behavior for programs which provide a
19527 command line interface to the user. Advantages are @sc{gnu} Emacs-style
19528 or @dfn{vi}-style inline editing of commands, @code{csh}-like history
19529 substitution, and a storage and recall of command history across
19530 debugging sessions.
19532 You may control the behavior of command line editing in @value{GDBN} with the
19533 command @code{set}.
19536 @kindex set editing
19539 @itemx set editing on
19540 Enable command line editing (enabled by default).
19542 @item set editing off
19543 Disable command line editing.
19545 @kindex show editing
19547 Show whether command line editing is enabled.
19550 @ifset SYSTEM_READLINE
19551 @xref{Command Line Editing, , , rluserman, GNU Readline Library},
19553 @ifclear SYSTEM_READLINE
19554 @xref{Command Line Editing},
19556 for more details about the Readline
19557 interface. Users unfamiliar with @sc{gnu} Emacs or @code{vi} are
19558 encouraged to read that chapter.
19560 @node Command History
19561 @section Command History
19562 @cindex command history
19564 @value{GDBN} can keep track of the commands you type during your
19565 debugging sessions, so that you can be certain of precisely what
19566 happened. Use these commands to manage the @value{GDBN} command
19569 @value{GDBN} uses the @sc{gnu} History library, a part of the Readline
19570 package, to provide the history facility.
19571 @ifset SYSTEM_READLINE
19572 @xref{Using History Interactively, , , history, GNU History Library},
19574 @ifclear SYSTEM_READLINE
19575 @xref{Using History Interactively},
19577 for the detailed description of the History library.
19579 To issue a command to @value{GDBN} without affecting certain aspects of
19580 the state which is seen by users, prefix it with @samp{server }
19581 (@pxref{Server Prefix}). This
19582 means that this command will not affect the command history, nor will it
19583 affect @value{GDBN}'s notion of which command to repeat if @key{RET} is
19584 pressed on a line by itself.
19586 @cindex @code{server}, command prefix
19587 The server prefix does not affect the recording of values into the value
19588 history; to print a value without recording it into the value history,
19589 use the @code{output} command instead of the @code{print} command.
19591 Here is the description of @value{GDBN} commands related to command
19595 @cindex history substitution
19596 @cindex history file
19597 @kindex set history filename
19598 @cindex @env{GDBHISTFILE}, environment variable
19599 @item set history filename @var{fname}
19600 Set the name of the @value{GDBN} command history file to @var{fname}.
19601 This is the file where @value{GDBN} reads an initial command history
19602 list, and where it writes the command history from this session when it
19603 exits. You can access this list through history expansion or through
19604 the history command editing characters listed below. This file defaults
19605 to the value of the environment variable @code{GDBHISTFILE}, or to
19606 @file{./.gdb_history} (@file{./_gdb_history} on MS-DOS) if this variable
19609 @cindex save command history
19610 @kindex set history save
19611 @item set history save
19612 @itemx set history save on
19613 Record command history in a file, whose name may be specified with the
19614 @code{set history filename} command. By default, this option is disabled.
19616 @item set history save off
19617 Stop recording command history in a file.
19619 @cindex history size
19620 @kindex set history size
19621 @cindex @env{HISTSIZE}, environment variable
19622 @item set history size @var{size}
19623 Set the number of commands which @value{GDBN} keeps in its history list.
19624 This defaults to the value of the environment variable
19625 @code{HISTSIZE}, or to 256 if this variable is not set.
19628 History expansion assigns special meaning to the character @kbd{!}.
19629 @ifset SYSTEM_READLINE
19630 @xref{Event Designators, , , history, GNU History Library},
19632 @ifclear SYSTEM_READLINE
19633 @xref{Event Designators},
19637 @cindex history expansion, turn on/off
19638 Since @kbd{!} is also the logical not operator in C, history expansion
19639 is off by default. If you decide to enable history expansion with the
19640 @code{set history expansion on} command, you may sometimes need to
19641 follow @kbd{!} (when it is used as logical not, in an expression) with
19642 a space or a tab to prevent it from being expanded. The readline
19643 history facilities do not attempt substitution on the strings
19644 @kbd{!=} and @kbd{!(}, even when history expansion is enabled.
19646 The commands to control history expansion are:
19649 @item set history expansion on
19650 @itemx set history expansion
19651 @kindex set history expansion
19652 Enable history expansion. History expansion is off by default.
19654 @item set history expansion off
19655 Disable history expansion.
19658 @kindex show history
19660 @itemx show history filename
19661 @itemx show history save
19662 @itemx show history size
19663 @itemx show history expansion
19664 These commands display the state of the @value{GDBN} history parameters.
19665 @code{show history} by itself displays all four states.
19670 @kindex show commands
19671 @cindex show last commands
19672 @cindex display command history
19673 @item show commands
19674 Display the last ten commands in the command history.
19676 @item show commands @var{n}
19677 Print ten commands centered on command number @var{n}.
19679 @item show commands +
19680 Print ten commands just after the commands last printed.
19684 @section Screen Size
19685 @cindex size of screen
19686 @cindex pauses in output
19688 Certain commands to @value{GDBN} may produce large amounts of
19689 information output to the screen. To help you read all of it,
19690 @value{GDBN} pauses and asks you for input at the end of each page of
19691 output. Type @key{RET} when you want to continue the output, or @kbd{q}
19692 to discard the remaining output. Also, the screen width setting
19693 determines when to wrap lines of output. Depending on what is being
19694 printed, @value{GDBN} tries to break the line at a readable place,
19695 rather than simply letting it overflow onto the following line.
19697 Normally @value{GDBN} knows the size of the screen from the terminal
19698 driver software. For example, on Unix @value{GDBN} uses the termcap data base
19699 together with the value of the @code{TERM} environment variable and the
19700 @code{stty rows} and @code{stty cols} settings. If this is not correct,
19701 you can override it with the @code{set height} and @code{set
19708 @kindex show height
19709 @item set height @var{lpp}
19711 @itemx set width @var{cpl}
19713 These @code{set} commands specify a screen height of @var{lpp} lines and
19714 a screen width of @var{cpl} characters. The associated @code{show}
19715 commands display the current settings.
19717 If you specify a height of zero lines, @value{GDBN} does not pause during
19718 output no matter how long the output is. This is useful if output is to a
19719 file or to an editor buffer.
19721 Likewise, you can specify @samp{set width 0} to prevent @value{GDBN}
19722 from wrapping its output.
19724 @item set pagination on
19725 @itemx set pagination off
19726 @kindex set pagination
19727 Turn the output pagination on or off; the default is on. Turning
19728 pagination off is the alternative to @code{set height 0}. Note that
19729 running @value{GDBN} with the @option{--batch} option (@pxref{Mode
19730 Options, -batch}) also automatically disables pagination.
19732 @item show pagination
19733 @kindex show pagination
19734 Show the current pagination mode.
19739 @cindex number representation
19740 @cindex entering numbers
19742 You can always enter numbers in octal, decimal, or hexadecimal in
19743 @value{GDBN} by the usual conventions: octal numbers begin with
19744 @samp{0}, decimal numbers end with @samp{.}, and hexadecimal numbers
19745 begin with @samp{0x}. Numbers that neither begin with @samp{0} or
19746 @samp{0x}, nor end with a @samp{.} are, by default, entered in base
19747 10; likewise, the default display for numbers---when no particular
19748 format is specified---is base 10. You can change the default base for
19749 both input and output with the commands described below.
19752 @kindex set input-radix
19753 @item set input-radix @var{base}
19754 Set the default base for numeric input. Supported choices
19755 for @var{base} are decimal 8, 10, or 16. @var{base} must itself be
19756 specified either unambiguously or using the current input radix; for
19760 set input-radix 012
19761 set input-radix 10.
19762 set input-radix 0xa
19766 sets the input base to decimal. On the other hand, @samp{set input-radix 10}
19767 leaves the input radix unchanged, no matter what it was, since
19768 @samp{10}, being without any leading or trailing signs of its base, is
19769 interpreted in the current radix. Thus, if the current radix is 16,
19770 @samp{10} is interpreted in hex, i.e.@: as 16 decimal, which doesn't
19773 @kindex set output-radix
19774 @item set output-radix @var{base}
19775 Set the default base for numeric display. Supported choices
19776 for @var{base} are decimal 8, 10, or 16. @var{base} must itself be
19777 specified either unambiguously or using the current input radix.
19779 @kindex show input-radix
19780 @item show input-radix
19781 Display the current default base for numeric input.
19783 @kindex show output-radix
19784 @item show output-radix
19785 Display the current default base for numeric display.
19787 @item set radix @r{[}@var{base}@r{]}
19791 These commands set and show the default base for both input and output
19792 of numbers. @code{set radix} sets the radix of input and output to
19793 the same base; without an argument, it resets the radix back to its
19794 default value of 10.
19799 @section Configuring the Current ABI
19801 @value{GDBN} can determine the @dfn{ABI} (Application Binary Interface) of your
19802 application automatically. However, sometimes you need to override its
19803 conclusions. Use these commands to manage @value{GDBN}'s view of the
19810 One @value{GDBN} configuration can debug binaries for multiple operating
19811 system targets, either via remote debugging or native emulation.
19812 @value{GDBN} will autodetect the @dfn{OS ABI} (Operating System ABI) in use,
19813 but you can override its conclusion using the @code{set osabi} command.
19814 One example where this is useful is in debugging of binaries which use
19815 an alternate C library (e.g.@: @sc{uClibc} for @sc{gnu}/Linux) which does
19816 not have the same identifying marks that the standard C library for your
19821 Show the OS ABI currently in use.
19824 With no argument, show the list of registered available OS ABI's.
19826 @item set osabi @var{abi}
19827 Set the current OS ABI to @var{abi}.
19830 @cindex float promotion
19832 Generally, the way that an argument of type @code{float} is passed to a
19833 function depends on whether the function is prototyped. For a prototyped
19834 (i.e.@: ANSI/ISO style) function, @code{float} arguments are passed unchanged,
19835 according to the architecture's convention for @code{float}. For unprototyped
19836 (i.e.@: K&R style) functions, @code{float} arguments are first promoted to type
19837 @code{double} and then passed.
19839 Unfortunately, some forms of debug information do not reliably indicate whether
19840 a function is prototyped. If @value{GDBN} calls a function that is not marked
19841 as prototyped, it consults @kbd{set coerce-float-to-double}.
19844 @kindex set coerce-float-to-double
19845 @item set coerce-float-to-double
19846 @itemx set coerce-float-to-double on
19847 Arguments of type @code{float} will be promoted to @code{double} when passed
19848 to an unprototyped function. This is the default setting.
19850 @item set coerce-float-to-double off
19851 Arguments of type @code{float} will be passed directly to unprototyped
19854 @kindex show coerce-float-to-double
19855 @item show coerce-float-to-double
19856 Show the current setting of promoting @code{float} to @code{double}.
19860 @kindex show cp-abi
19861 @value{GDBN} needs to know the ABI used for your program's C@t{++}
19862 objects. The correct C@t{++} ABI depends on which C@t{++} compiler was
19863 used to build your application. @value{GDBN} only fully supports
19864 programs with a single C@t{++} ABI; if your program contains code using
19865 multiple C@t{++} ABI's or if @value{GDBN} can not identify your
19866 program's ABI correctly, you can tell @value{GDBN} which ABI to use.
19867 Currently supported ABI's include ``gnu-v2'', for @code{g++} versions
19868 before 3.0, ``gnu-v3'', for @code{g++} versions 3.0 and later, and
19869 ``hpaCC'' for the HP ANSI C@t{++} compiler. Other C@t{++} compilers may
19870 use the ``gnu-v2'' or ``gnu-v3'' ABI's as well. The default setting is
19875 Show the C@t{++} ABI currently in use.
19878 With no argument, show the list of supported C@t{++} ABI's.
19880 @item set cp-abi @var{abi}
19881 @itemx set cp-abi auto
19882 Set the current C@t{++} ABI to @var{abi}, or return to automatic detection.
19885 @node Messages/Warnings
19886 @section Optional Warnings and Messages
19888 @cindex verbose operation
19889 @cindex optional warnings
19890 By default, @value{GDBN} is silent about its inner workings. If you are
19891 running on a slow machine, you may want to use the @code{set verbose}
19892 command. This makes @value{GDBN} tell you when it does a lengthy
19893 internal operation, so you will not think it has crashed.
19895 Currently, the messages controlled by @code{set verbose} are those
19896 which announce that the symbol table for a source file is being read;
19897 see @code{symbol-file} in @ref{Files, ,Commands to Specify Files}.
19900 @kindex set verbose
19901 @item set verbose on
19902 Enables @value{GDBN} output of certain informational messages.
19904 @item set verbose off
19905 Disables @value{GDBN} output of certain informational messages.
19907 @kindex show verbose
19909 Displays whether @code{set verbose} is on or off.
19912 By default, if @value{GDBN} encounters bugs in the symbol table of an
19913 object file, it is silent; but if you are debugging a compiler, you may
19914 find this information useful (@pxref{Symbol Errors, ,Errors Reading
19919 @kindex set complaints
19920 @item set complaints @var{limit}
19921 Permits @value{GDBN} to output @var{limit} complaints about each type of
19922 unusual symbols before becoming silent about the problem. Set
19923 @var{limit} to zero to suppress all complaints; set it to a large number
19924 to prevent complaints from being suppressed.
19926 @kindex show complaints
19927 @item show complaints
19928 Displays how many symbol complaints @value{GDBN} is permitted to produce.
19932 @anchor{confirmation requests}
19933 By default, @value{GDBN} is cautious, and asks what sometimes seems to be a
19934 lot of stupid questions to confirm certain commands. For example, if
19935 you try to run a program which is already running:
19939 The program being debugged has been started already.
19940 Start it from the beginning? (y or n)
19943 If you are willing to unflinchingly face the consequences of your own
19944 commands, you can disable this ``feature'':
19948 @kindex set confirm
19950 @cindex confirmation
19951 @cindex stupid questions
19952 @item set confirm off
19953 Disables confirmation requests. Note that running @value{GDBN} with
19954 the @option{--batch} option (@pxref{Mode Options, -batch}) also
19955 automatically disables confirmation requests.
19957 @item set confirm on
19958 Enables confirmation requests (the default).
19960 @kindex show confirm
19962 Displays state of confirmation requests.
19966 @cindex command tracing
19967 If you need to debug user-defined commands or sourced files you may find it
19968 useful to enable @dfn{command tracing}. In this mode each command will be
19969 printed as it is executed, prefixed with one or more @samp{+} symbols, the
19970 quantity denoting the call depth of each command.
19973 @kindex set trace-commands
19974 @cindex command scripts, debugging
19975 @item set trace-commands on
19976 Enable command tracing.
19977 @item set trace-commands off
19978 Disable command tracing.
19979 @item show trace-commands
19980 Display the current state of command tracing.
19983 @node Debugging Output
19984 @section Optional Messages about Internal Happenings
19985 @cindex optional debugging messages
19987 @value{GDBN} has commands that enable optional debugging messages from
19988 various @value{GDBN} subsystems; normally these commands are of
19989 interest to @value{GDBN} maintainers, or when reporting a bug. This
19990 section documents those commands.
19993 @kindex set exec-done-display
19994 @item set exec-done-display
19995 Turns on or off the notification of asynchronous commands'
19996 completion. When on, @value{GDBN} will print a message when an
19997 asynchronous command finishes its execution. The default is off.
19998 @kindex show exec-done-display
19999 @item show exec-done-display
20000 Displays the current setting of asynchronous command completion
20003 @cindex gdbarch debugging info
20004 @cindex architecture debugging info
20005 @item set debug arch
20006 Turns on or off display of gdbarch debugging info. The default is off
20008 @item show debug arch
20009 Displays the current state of displaying gdbarch debugging info.
20010 @item set debug aix-thread
20011 @cindex AIX threads
20012 Display debugging messages about inner workings of the AIX thread
20014 @item show debug aix-thread
20015 Show the current state of AIX thread debugging info display.
20016 @item set debug dwarf2-die
20017 @cindex DWARF2 DIEs
20018 Dump DWARF2 DIEs after they are read in.
20019 The value is the number of nesting levels to print.
20020 A value of zero turns off the display.
20021 @item show debug dwarf2-die
20022 Show the current state of DWARF2 DIE debugging.
20023 @item set debug displaced
20024 @cindex displaced stepping debugging info
20025 Turns on or off display of @value{GDBN} debugging info for the
20026 displaced stepping support. The default is off.
20027 @item show debug displaced
20028 Displays the current state of displaying @value{GDBN} debugging info
20029 related to displaced stepping.
20030 @item set debug event
20031 @cindex event debugging info
20032 Turns on or off display of @value{GDBN} event debugging info. The
20034 @item show debug event
20035 Displays the current state of displaying @value{GDBN} event debugging
20037 @item set debug expression
20038 @cindex expression debugging info
20039 Turns on or off display of debugging info about @value{GDBN}
20040 expression parsing. The default is off.
20041 @item show debug expression
20042 Displays the current state of displaying debugging info about
20043 @value{GDBN} expression parsing.
20044 @item set debug frame
20045 @cindex frame debugging info
20046 Turns on or off display of @value{GDBN} frame debugging info. The
20048 @item show debug frame
20049 Displays the current state of displaying @value{GDBN} frame debugging
20051 @item set debug gnu-nat
20052 @cindex @sc{gnu}/Hurd debug messages
20053 Turns on or off debugging messages from the @sc{gnu}/Hurd debug support.
20054 @item show debug gnu-nat
20055 Show the current state of @sc{gnu}/Hurd debugging messages.
20056 @item set debug infrun
20057 @cindex inferior debugging info
20058 Turns on or off display of @value{GDBN} debugging info for running the inferior.
20059 The default is off. @file{infrun.c} contains GDB's runtime state machine used
20060 for implementing operations such as single-stepping the inferior.
20061 @item show debug infrun
20062 Displays the current state of @value{GDBN} inferior debugging.
20063 @item set debug jit
20064 @cindex just-in-time compilation, debugging messages
20065 Turns on or off debugging messages from JIT debug support.
20066 @item show debug jit
20067 Displays the current state of @value{GDBN} JIT debugging.
20068 @item set debug lin-lwp
20069 @cindex @sc{gnu}/Linux LWP debug messages
20070 @cindex Linux lightweight processes
20071 Turns on or off debugging messages from the Linux LWP debug support.
20072 @item show debug lin-lwp
20073 Show the current state of Linux LWP debugging messages.
20074 @item set debug lin-lwp-async
20075 @cindex @sc{gnu}/Linux LWP async debug messages
20076 @cindex Linux lightweight processes
20077 Turns on or off debugging messages from the Linux LWP async debug support.
20078 @item show debug lin-lwp-async
20079 Show the current state of Linux LWP async debugging messages.
20080 @item set debug observer
20081 @cindex observer debugging info
20082 Turns on or off display of @value{GDBN} observer debugging. This
20083 includes info such as the notification of observable events.
20084 @item show debug observer
20085 Displays the current state of observer debugging.
20086 @item set debug overload
20087 @cindex C@t{++} overload debugging info
20088 Turns on or off display of @value{GDBN} C@t{++} overload debugging
20089 info. This includes info such as ranking of functions, etc. The default
20091 @item show debug overload
20092 Displays the current state of displaying @value{GDBN} C@t{++} overload
20094 @cindex expression parser, debugging info
20095 @cindex debug expression parser
20096 @item set debug parser
20097 Turns on or off the display of expression parser debugging output.
20098 Internally, this sets the @code{yydebug} variable in the expression
20099 parser. @xref{Tracing, , Tracing Your Parser, bison, Bison}, for
20100 details. The default is off.
20101 @item show debug parser
20102 Show the current state of expression parser debugging.
20103 @cindex packets, reporting on stdout
20104 @cindex serial connections, debugging
20105 @cindex debug remote protocol
20106 @cindex remote protocol debugging
20107 @cindex display remote packets
20108 @item set debug remote
20109 Turns on or off display of reports on all packets sent back and forth across
20110 the serial line to the remote machine. The info is printed on the
20111 @value{GDBN} standard output stream. The default is off.
20112 @item show debug remote
20113 Displays the state of display of remote packets.
20114 @item set debug serial
20115 Turns on or off display of @value{GDBN} serial debugging info. The
20117 @item show debug serial
20118 Displays the current state of displaying @value{GDBN} serial debugging
20120 @item set debug solib-frv
20121 @cindex FR-V shared-library debugging
20122 Turns on or off debugging messages for FR-V shared-library code.
20123 @item show debug solib-frv
20124 Display the current state of FR-V shared-library code debugging
20126 @item set debug target
20127 @cindex target debugging info
20128 Turns on or off display of @value{GDBN} target debugging info. This info
20129 includes what is going on at the target level of GDB, as it happens. The
20130 default is 0. Set it to 1 to track events, and to 2 to also track the
20131 value of large memory transfers. Changes to this flag do not take effect
20132 until the next time you connect to a target or use the @code{run} command.
20133 @item show debug target
20134 Displays the current state of displaying @value{GDBN} target debugging
20136 @item set debug timestamp
20137 @cindex timestampping debugging info
20138 Turns on or off display of timestamps with @value{GDBN} debugging info.
20139 When enabled, seconds and microseconds are displayed before each debugging
20141 @item show debug timestamp
20142 Displays the current state of displaying timestamps with @value{GDBN}
20144 @item set debugvarobj
20145 @cindex variable object debugging info
20146 Turns on or off display of @value{GDBN} variable object debugging
20147 info. The default is off.
20148 @item show debugvarobj
20149 Displays the current state of displaying @value{GDBN} variable object
20151 @item set debug xml
20152 @cindex XML parser debugging
20153 Turns on or off debugging messages for built-in XML parsers.
20154 @item show debug xml
20155 Displays the current state of XML debugging messages.
20158 @node Other Misc Settings
20159 @section Other Miscellaneous Settings
20160 @cindex miscellaneous settings
20163 @kindex set interactive-mode
20164 @item set interactive-mode
20165 If @code{on}, forces @value{GDBN} to assume that GDB was started
20166 in a terminal. In practice, this means that @value{GDBN} should wait
20167 for the user to answer queries generated by commands entered at
20168 the command prompt. If @code{off}, forces @value{GDBN} to operate
20169 in the opposite mode, and it uses the default answers to all queries.
20170 If @code{auto} (the default), @value{GDBN} tries to determine whether
20171 its standard input is a terminal, and works in interactive-mode if it
20172 is, non-interactively otherwise.
20174 In the vast majority of cases, the debugger should be able to guess
20175 correctly which mode should be used. But this setting can be useful
20176 in certain specific cases, such as running a MinGW @value{GDBN}
20177 inside a cygwin window.
20179 @kindex show interactive-mode
20180 @item show interactive-mode
20181 Displays whether the debugger is operating in interactive mode or not.
20184 @node Extending GDB
20185 @chapter Extending @value{GDBN}
20186 @cindex extending GDB
20188 @value{GDBN} provides two mechanisms for extension. The first is based
20189 on composition of @value{GDBN} commands, and the second is based on the
20190 Python scripting language.
20192 To facilitate the use of these extensions, @value{GDBN} is capable
20193 of evaluating the contents of a file. When doing so, @value{GDBN}
20194 can recognize which scripting language is being used by looking at
20195 the filename extension. Files with an unrecognized filename extension
20196 are always treated as a @value{GDBN} Command Files.
20197 @xref{Command Files,, Command files}.
20199 You can control how @value{GDBN} evaluates these files with the following
20203 @kindex set script-extension
20204 @kindex show script-extension
20205 @item set script-extension off
20206 All scripts are always evaluated as @value{GDBN} Command Files.
20208 @item set script-extension soft
20209 The debugger determines the scripting language based on filename
20210 extension. If this scripting language is supported, @value{GDBN}
20211 evaluates the script using that language. Otherwise, it evaluates
20212 the file as a @value{GDBN} Command File.
20214 @item set script-extension strict
20215 The debugger determines the scripting language based on filename
20216 extension, and evaluates the script using that language. If the
20217 language is not supported, then the evaluation fails.
20219 @item show script-extension
20220 Display the current value of the @code{script-extension} option.
20225 * Sequences:: Canned Sequences of Commands
20226 * Python:: Scripting @value{GDBN} using Python
20230 @section Canned Sequences of Commands
20232 Aside from breakpoint commands (@pxref{Break Commands, ,Breakpoint
20233 Command Lists}), @value{GDBN} provides two ways to store sequences of
20234 commands for execution as a unit: user-defined commands and command
20238 * Define:: How to define your own commands
20239 * Hooks:: Hooks for user-defined commands
20240 * Command Files:: How to write scripts of commands to be stored in a file
20241 * Output:: Commands for controlled output
20245 @subsection User-defined Commands
20247 @cindex user-defined command
20248 @cindex arguments, to user-defined commands
20249 A @dfn{user-defined command} is a sequence of @value{GDBN} commands to
20250 which you assign a new name as a command. This is done with the
20251 @code{define} command. User commands may accept up to 10 arguments
20252 separated by whitespace. Arguments are accessed within the user command
20253 via @code{$arg0@dots{}$arg9}. A trivial example:
20257 print $arg0 + $arg1 + $arg2
20262 To execute the command use:
20269 This defines the command @code{adder}, which prints the sum of
20270 its three arguments. Note the arguments are text substitutions, so they may
20271 reference variables, use complex expressions, or even perform inferior
20274 @cindex argument count in user-defined commands
20275 @cindex how many arguments (user-defined commands)
20276 In addition, @code{$argc} may be used to find out how many arguments have
20277 been passed. This expands to a number in the range 0@dots{}10.
20282 print $arg0 + $arg1
20285 print $arg0 + $arg1 + $arg2
20293 @item define @var{commandname}
20294 Define a command named @var{commandname}. If there is already a command
20295 by that name, you are asked to confirm that you want to redefine it.
20296 @var{commandname} may be a bare command name consisting of letters,
20297 numbers, dashes, and underscores. It may also start with any predefined
20298 prefix command. For example, @samp{define target my-target} creates
20299 a user-defined @samp{target my-target} command.
20301 The definition of the command is made up of other @value{GDBN} command lines,
20302 which are given following the @code{define} command. The end of these
20303 commands is marked by a line containing @code{end}.
20306 @kindex end@r{ (user-defined commands)}
20307 @item document @var{commandname}
20308 Document the user-defined command @var{commandname}, so that it can be
20309 accessed by @code{help}. The command @var{commandname} must already be
20310 defined. This command reads lines of documentation just as @code{define}
20311 reads the lines of the command definition, ending with @code{end}.
20312 After the @code{document} command is finished, @code{help} on command
20313 @var{commandname} displays the documentation you have written.
20315 You may use the @code{document} command again to change the
20316 documentation of a command. Redefining the command with @code{define}
20317 does not change the documentation.
20319 @kindex dont-repeat
20320 @cindex don't repeat command
20322 Used inside a user-defined command, this tells @value{GDBN} that this
20323 command should not be repeated when the user hits @key{RET}
20324 (@pxref{Command Syntax, repeat last command}).
20326 @kindex help user-defined
20327 @item help user-defined
20328 List all user-defined commands, with the first line of the documentation
20333 @itemx show user @var{commandname}
20334 Display the @value{GDBN} commands used to define @var{commandname} (but
20335 not its documentation). If no @var{commandname} is given, display the
20336 definitions for all user-defined commands.
20338 @cindex infinite recursion in user-defined commands
20339 @kindex show max-user-call-depth
20340 @kindex set max-user-call-depth
20341 @item show max-user-call-depth
20342 @itemx set max-user-call-depth
20343 The value of @code{max-user-call-depth} controls how many recursion
20344 levels are allowed in user-defined commands before @value{GDBN} suspects an
20345 infinite recursion and aborts the command.
20348 In addition to the above commands, user-defined commands frequently
20349 use control flow commands, described in @ref{Command Files}.
20351 When user-defined commands are executed, the
20352 commands of the definition are not printed. An error in any command
20353 stops execution of the user-defined command.
20355 If used interactively, commands that would ask for confirmation proceed
20356 without asking when used inside a user-defined command. Many @value{GDBN}
20357 commands that normally print messages to say what they are doing omit the
20358 messages when used in a user-defined command.
20361 @subsection User-defined Command Hooks
20362 @cindex command hooks
20363 @cindex hooks, for commands
20364 @cindex hooks, pre-command
20367 You may define @dfn{hooks}, which are a special kind of user-defined
20368 command. Whenever you run the command @samp{foo}, if the user-defined
20369 command @samp{hook-foo} exists, it is executed (with no arguments)
20370 before that command.
20372 @cindex hooks, post-command
20374 A hook may also be defined which is run after the command you executed.
20375 Whenever you run the command @samp{foo}, if the user-defined command
20376 @samp{hookpost-foo} exists, it is executed (with no arguments) after
20377 that command. Post-execution hooks may exist simultaneously with
20378 pre-execution hooks, for the same command.
20380 It is valid for a hook to call the command which it hooks. If this
20381 occurs, the hook is not re-executed, thereby avoiding infinite recursion.
20383 @c It would be nice if hookpost could be passed a parameter indicating
20384 @c if the command it hooks executed properly or not. FIXME!
20386 @kindex stop@r{, a pseudo-command}
20387 In addition, a pseudo-command, @samp{stop} exists. Defining
20388 (@samp{hook-stop}) makes the associated commands execute every time
20389 execution stops in your program: before breakpoint commands are run,
20390 displays are printed, or the stack frame is printed.
20392 For example, to ignore @code{SIGALRM} signals while
20393 single-stepping, but treat them normally during normal execution,
20398 handle SIGALRM nopass
20402 handle SIGALRM pass
20405 define hook-continue
20406 handle SIGALRM pass
20410 As a further example, to hook at the beginning and end of the @code{echo}
20411 command, and to add extra text to the beginning and end of the message,
20419 define hookpost-echo
20423 (@value{GDBP}) echo Hello World
20424 <<<---Hello World--->>>
20429 You can define a hook for any single-word command in @value{GDBN}, but
20430 not for command aliases; you should define a hook for the basic command
20431 name, e.g.@: @code{backtrace} rather than @code{bt}.
20432 @c FIXME! So how does Joe User discover whether a command is an alias
20434 You can hook a multi-word command by adding @code{hook-} or
20435 @code{hookpost-} to the last word of the command, e.g.@:
20436 @samp{define target hook-remote} to add a hook to @samp{target remote}.
20438 If an error occurs during the execution of your hook, execution of
20439 @value{GDBN} commands stops and @value{GDBN} issues a prompt
20440 (before the command that you actually typed had a chance to run).
20442 If you try to define a hook which does not match any known command, you
20443 get a warning from the @code{define} command.
20445 @node Command Files
20446 @subsection Command Files
20448 @cindex command files
20449 @cindex scripting commands
20450 A command file for @value{GDBN} is a text file made of lines that are
20451 @value{GDBN} commands. Comments (lines starting with @kbd{#}) may
20452 also be included. An empty line in a command file does nothing; it
20453 does not mean to repeat the last command, as it would from the
20456 You can request the execution of a command file with the @code{source}
20457 command. Note that the @code{source} command is also used to evaluate
20458 scripts that are not Command Files. The exact behavior can be configured
20459 using the @code{script-extension} setting.
20460 @xref{Extending GDB,, Extending GDB}.
20464 @cindex execute commands from a file
20465 @item source [-s] [-v] @var{filename}
20466 Execute the command file @var{filename}.
20469 The lines in a command file are generally executed sequentially,
20470 unless the order of execution is changed by one of the
20471 @emph{flow-control commands} described below. The commands are not
20472 printed as they are executed. An error in any command terminates
20473 execution of the command file and control is returned to the console.
20475 @value{GDBN} first searches for @var{filename} in the current directory.
20476 If the file is not found there, and @var{filename} does not specify a
20477 directory, then @value{GDBN} also looks for the file on the source search path
20478 (specified with the @samp{directory} command);
20479 except that @file{$cdir} is not searched because the compilation directory
20480 is not relevant to scripts.
20482 If @code{-s} is specified, then @value{GDBN} searches for @var{filename}
20483 on the search path even if @var{filename} specifies a directory.
20484 The search is done by appending @var{filename} to each element of the
20485 search path. So, for example, if @var{filename} is @file{mylib/myscript}
20486 and the search path contains @file{/home/user} then @value{GDBN} will
20487 look for the script @file{/home/user/mylib/myscript}.
20488 The search is also done if @var{filename} is an absolute path.
20489 For example, if @var{filename} is @file{/tmp/myscript} and
20490 the search path contains @file{/home/user} then @value{GDBN} will
20491 look for the script @file{/home/user/tmp/myscript}.
20492 For DOS-like systems, if @var{filename} contains a drive specification,
20493 it is stripped before concatenation. For example, if @var{filename} is
20494 @file{d:myscript} and the search path contains @file{c:/tmp} then @value{GDBN}
20495 will look for the script @file{c:/tmp/myscript}.
20497 If @code{-v}, for verbose mode, is given then @value{GDBN} displays
20498 each command as it is executed. The option must be given before
20499 @var{filename}, and is interpreted as part of the filename anywhere else.
20501 Commands that would ask for confirmation if used interactively proceed
20502 without asking when used in a command file. Many @value{GDBN} commands that
20503 normally print messages to say what they are doing omit the messages
20504 when called from command files.
20506 @value{GDBN} also accepts command input from standard input. In this
20507 mode, normal output goes to standard output and error output goes to
20508 standard error. Errors in a command file supplied on standard input do
20509 not terminate execution of the command file---execution continues with
20513 gdb < cmds > log 2>&1
20516 (The syntax above will vary depending on the shell used.) This example
20517 will execute commands from the file @file{cmds}. All output and errors
20518 would be directed to @file{log}.
20520 Since commands stored on command files tend to be more general than
20521 commands typed interactively, they frequently need to deal with
20522 complicated situations, such as different or unexpected values of
20523 variables and symbols, changes in how the program being debugged is
20524 built, etc. @value{GDBN} provides a set of flow-control commands to
20525 deal with these complexities. Using these commands, you can write
20526 complex scripts that loop over data structures, execute commands
20527 conditionally, etc.
20534 This command allows to include in your script conditionally executed
20535 commands. The @code{if} command takes a single argument, which is an
20536 expression to evaluate. It is followed by a series of commands that
20537 are executed only if the expression is true (its value is nonzero).
20538 There can then optionally be an @code{else} line, followed by a series
20539 of commands that are only executed if the expression was false. The
20540 end of the list is marked by a line containing @code{end}.
20544 This command allows to write loops. Its syntax is similar to
20545 @code{if}: the command takes a single argument, which is an expression
20546 to evaluate, and must be followed by the commands to execute, one per
20547 line, terminated by an @code{end}. These commands are called the
20548 @dfn{body} of the loop. The commands in the body of @code{while} are
20549 executed repeatedly as long as the expression evaluates to true.
20553 This command exits the @code{while} loop in whose body it is included.
20554 Execution of the script continues after that @code{while}s @code{end}
20557 @kindex loop_continue
20558 @item loop_continue
20559 This command skips the execution of the rest of the body of commands
20560 in the @code{while} loop in whose body it is included. Execution
20561 branches to the beginning of the @code{while} loop, where it evaluates
20562 the controlling expression.
20564 @kindex end@r{ (if/else/while commands)}
20566 Terminate the block of commands that are the body of @code{if},
20567 @code{else}, or @code{while} flow-control commands.
20572 @subsection Commands for Controlled Output
20574 During the execution of a command file or a user-defined command, normal
20575 @value{GDBN} output is suppressed; the only output that appears is what is
20576 explicitly printed by the commands in the definition. This section
20577 describes three commands useful for generating exactly the output you
20582 @item echo @var{text}
20583 @c I do not consider backslash-space a standard C escape sequence
20584 @c because it is not in ANSI.
20585 Print @var{text}. Nonprinting characters can be included in
20586 @var{text} using C escape sequences, such as @samp{\n} to print a
20587 newline. @strong{No newline is printed unless you specify one.}
20588 In addition to the standard C escape sequences, a backslash followed
20589 by a space stands for a space. This is useful for displaying a
20590 string with spaces at the beginning or the end, since leading and
20591 trailing spaces are otherwise trimmed from all arguments.
20592 To print @samp{@w{ }and foo =@w{ }}, use the command
20593 @samp{echo \@w{ }and foo = \@w{ }}.
20595 A backslash at the end of @var{text} can be used, as in C, to continue
20596 the command onto subsequent lines. For example,
20599 echo This is some text\n\
20600 which is continued\n\
20601 onto several lines.\n
20604 produces the same output as
20607 echo This is some text\n
20608 echo which is continued\n
20609 echo onto several lines.\n
20613 @item output @var{expression}
20614 Print the value of @var{expression} and nothing but that value: no
20615 newlines, no @samp{$@var{nn} = }. The value is not entered in the
20616 value history either. @xref{Expressions, ,Expressions}, for more information
20619 @item output/@var{fmt} @var{expression}
20620 Print the value of @var{expression} in format @var{fmt}. You can use
20621 the same formats as for @code{print}. @xref{Output Formats,,Output
20622 Formats}, for more information.
20625 @item printf @var{template}, @var{expressions}@dots{}
20626 Print the values of one or more @var{expressions} under the control of
20627 the string @var{template}. To print several values, make
20628 @var{expressions} be a comma-separated list of individual expressions,
20629 which may be either numbers or pointers. Their values are printed as
20630 specified by @var{template}, exactly as a C program would do by
20631 executing the code below:
20634 printf (@var{template}, @var{expressions}@dots{});
20637 As in @code{C} @code{printf}, ordinary characters in @var{template}
20638 are printed verbatim, while @dfn{conversion specification} introduced
20639 by the @samp{%} character cause subsequent @var{expressions} to be
20640 evaluated, their values converted and formatted according to type and
20641 style information encoded in the conversion specifications, and then
20644 For example, you can print two values in hex like this:
20647 printf "foo, bar-foo = 0x%x, 0x%x\n", foo, bar-foo
20650 @code{printf} supports all the standard @code{C} conversion
20651 specifications, including the flags and modifiers between the @samp{%}
20652 character and the conversion letter, with the following exceptions:
20656 The argument-ordering modifiers, such as @samp{2$}, are not supported.
20659 The modifier @samp{*} is not supported for specifying precision or
20663 The @samp{'} flag (for separation of digits into groups according to
20664 @code{LC_NUMERIC'}) is not supported.
20667 The type modifiers @samp{hh}, @samp{j}, @samp{t}, and @samp{z} are not
20671 The conversion letter @samp{n} (as in @samp{%n}) is not supported.
20674 The conversion letters @samp{a} and @samp{A} are not supported.
20678 Note that the @samp{ll} type modifier is supported only if the
20679 underlying @code{C} implementation used to build @value{GDBN} supports
20680 the @code{long long int} type, and the @samp{L} type modifier is
20681 supported only if @code{long double} type is available.
20683 As in @code{C}, @code{printf} supports simple backslash-escape
20684 sequences, such as @code{\n}, @samp{\t}, @samp{\\}, @samp{\"},
20685 @samp{\a}, and @samp{\f}, that consist of backslash followed by a
20686 single character. Octal and hexadecimal escape sequences are not
20689 Additionally, @code{printf} supports conversion specifications for DFP
20690 (@dfn{Decimal Floating Point}) types using the following length modifiers
20691 together with a floating point specifier.
20696 @samp{H} for printing @code{Decimal32} types.
20699 @samp{D} for printing @code{Decimal64} types.
20702 @samp{DD} for printing @code{Decimal128} types.
20705 If the underlying @code{C} implementation used to build @value{GDBN} has
20706 support for the three length modifiers for DFP types, other modifiers
20707 such as width and precision will also be available for @value{GDBN} to use.
20709 In case there is no such @code{C} support, no additional modifiers will be
20710 available and the value will be printed in the standard way.
20712 Here's an example of printing DFP types using the above conversion letters:
20714 printf "D32: %Hf - D64: %Df - D128: %DDf\n",1.2345df,1.2E10dd,1.2E1dl
20718 @item eval @var{template}, @var{expressions}@dots{}
20719 Convert the values of one or more @var{expressions} under the control of
20720 the string @var{template} to a command line, and call it.
20725 @section Scripting @value{GDBN} using Python
20726 @cindex python scripting
20727 @cindex scripting with python
20729 You can script @value{GDBN} using the @uref{http://www.python.org/,
20730 Python programming language}. This feature is available only if
20731 @value{GDBN} was configured using @option{--with-python}.
20733 @cindex python directory
20734 Python scripts used by @value{GDBN} should be installed in
20735 @file{@var{data-directory}/python}, where @var{data-directory} is
20736 the data directory as determined at @value{GDBN} startup (@pxref{Data Files}).
20737 This directory, known as the @dfn{python directory},
20738 is automatically added to the Python Search Path in order to allow
20739 the Python interpreter to locate all scripts installed at this location.
20742 * Python Commands:: Accessing Python from @value{GDBN}.
20743 * Python API:: Accessing @value{GDBN} from Python.
20744 * Auto-loading:: Automatically loading Python code.
20745 * Python modules:: Python modules provided by @value{GDBN}.
20748 @node Python Commands
20749 @subsection Python Commands
20750 @cindex python commands
20751 @cindex commands to access python
20753 @value{GDBN} provides one command for accessing the Python interpreter,
20754 and one related setting:
20758 @item python @r{[}@var{code}@r{]}
20759 The @code{python} command can be used to evaluate Python code.
20761 If given an argument, the @code{python} command will evaluate the
20762 argument as a Python command. For example:
20765 (@value{GDBP}) python print 23
20769 If you do not provide an argument to @code{python}, it will act as a
20770 multi-line command, like @code{define}. In this case, the Python
20771 script is made up of subsequent command lines, given after the
20772 @code{python} command. This command list is terminated using a line
20773 containing @code{end}. For example:
20776 (@value{GDBP}) python
20778 End with a line saying just "end".
20784 @kindex maint set python print-stack
20785 @item maint set python print-stack
20786 By default, @value{GDBN} will print a stack trace when an error occurs
20787 in a Python script. This can be controlled using @code{maint set
20788 python print-stack}: if @code{on}, the default, then Python stack
20789 printing is enabled; if @code{off}, then Python stack printing is
20793 It is also possible to execute a Python script from the @value{GDBN}
20797 @item source @file{script-name}
20798 The script name must end with @samp{.py} and @value{GDBN} must be configured
20799 to recognize the script language based on filename extension using
20800 the @code{script-extension} setting. @xref{Extending GDB, ,Extending GDB}.
20802 @item python execfile ("script-name")
20803 This method is based on the @code{execfile} Python built-in function,
20804 and thus is always available.
20808 @subsection Python API
20810 @cindex programming in python
20812 @cindex python stdout
20813 @cindex python pagination
20814 At startup, @value{GDBN} overrides Python's @code{sys.stdout} and
20815 @code{sys.stderr} to print using @value{GDBN}'s output-paging streams.
20816 A Python program which outputs to one of these streams may have its
20817 output interrupted by the user (@pxref{Screen Size}). In this
20818 situation, a Python @code{KeyboardInterrupt} exception is thrown.
20821 * Basic Python:: Basic Python Functions.
20822 * Exception Handling:: How Python exceptions are translated.
20823 * Values From Inferior:: Python representation of values.
20824 * Types In Python:: Python representation of types.
20825 * Pretty Printing API:: Pretty-printing values.
20826 * Selecting Pretty-Printers:: How GDB chooses a pretty-printer.
20827 * Writing a Pretty-Printer:: Writing a Pretty-Printer.
20828 * Inferiors In Python:: Python representation of inferiors (processes)
20829 * Events In Python:: Listening for events from @value{GDBN}.
20830 * Threads In Python:: Accessing inferior threads from Python.
20831 * Commands In Python:: Implementing new commands in Python.
20832 * Parameters In Python:: Adding new @value{GDBN} parameters.
20833 * Functions In Python:: Writing new convenience functions.
20834 * Progspaces In Python:: Program spaces.
20835 * Objfiles In Python:: Object files.
20836 * Frames In Python:: Accessing inferior stack frames from Python.
20837 * Blocks In Python:: Accessing frame blocks from Python.
20838 * Symbols In Python:: Python representation of symbols.
20839 * Symbol Tables In Python:: Python representation of symbol tables.
20840 * Lazy Strings In Python:: Python representation of lazy strings.
20841 * Breakpoints In Python:: Manipulating breakpoints using Python.
20845 @subsubsection Basic Python
20847 @cindex python functions
20848 @cindex python module
20850 @value{GDBN} introduces a new Python module, named @code{gdb}. All
20851 methods and classes added by @value{GDBN} are placed in this module.
20852 @value{GDBN} automatically @code{import}s the @code{gdb} module for
20853 use in all scripts evaluated by the @code{python} command.
20855 @findex gdb.PYTHONDIR
20857 A string containing the python directory (@pxref{Python}).
20860 @findex gdb.execute
20861 @defun execute command [from_tty] [to_string]
20862 Evaluate @var{command}, a string, as a @value{GDBN} CLI command.
20863 If a GDB exception happens while @var{command} runs, it is
20864 translated as described in @ref{Exception Handling,,Exception Handling}.
20866 @var{from_tty} specifies whether @value{GDBN} ought to consider this
20867 command as having originated from the user invoking it interactively.
20868 It must be a boolean value. If omitted, it defaults to @code{False}.
20870 By default, any output produced by @var{command} is sent to
20871 @value{GDBN}'s standard output. If the @var{to_string} parameter is
20872 @code{True}, then output will be collected by @code{gdb.execute} and
20873 returned as a string. The default is @code{False}, in which case the
20874 return value is @code{None}. If @var{to_string} is @code{True}, the
20875 @value{GDBN} virtual terminal will be temporarily set to unlimited width
20876 and height, and its pagination will be disabled; @pxref{Screen Size}.
20879 @findex gdb.breakpoints
20881 Return a sequence holding all of @value{GDBN}'s breakpoints.
20882 @xref{Breakpoints In Python}, for more information.
20885 @findex gdb.parameter
20886 @defun parameter parameter
20887 Return the value of a @value{GDBN} parameter. @var{parameter} is a
20888 string naming the parameter to look up; @var{parameter} may contain
20889 spaces if the parameter has a multi-part name. For example,
20890 @samp{print object} is a valid parameter name.
20892 If the named parameter does not exist, this function throws a
20893 @code{gdb.error} (@pxref{Exception Handling}). Otherwise, the
20894 parameter's value is converted to a Python value of the appropriate
20895 type, and returned.
20898 @findex gdb.history
20899 @defun history number
20900 Return a value from @value{GDBN}'s value history (@pxref{Value
20901 History}). @var{number} indicates which history element to return.
20902 If @var{number} is negative, then @value{GDBN} will take its absolute value
20903 and count backward from the last element (i.e., the most recent element) to
20904 find the value to return. If @var{number} is zero, then @value{GDBN} will
20905 return the most recent element. If the element specified by @var{number}
20906 doesn't exist in the value history, a @code{gdb.error} exception will be
20909 If no exception is raised, the return value is always an instance of
20910 @code{gdb.Value} (@pxref{Values From Inferior}).
20913 @findex gdb.parse_and_eval
20914 @defun parse_and_eval expression
20915 Parse @var{expression} as an expression in the current language,
20916 evaluate it, and return the result as a @code{gdb.Value}.
20917 @var{expression} must be a string.
20919 This function can be useful when implementing a new command
20920 (@pxref{Commands In Python}), as it provides a way to parse the
20921 command's argument as an expression. It is also useful simply to
20922 compute values, for example, it is the only way to get the value of a
20923 convenience variable (@pxref{Convenience Vars}) as a @code{gdb.Value}.
20926 @findex gdb.post_event
20927 @defun post_event event
20928 Put @var{event}, a callable object taking no arguments, into
20929 @value{GDBN}'s internal event queue. This callable will be invoked at
20930 some later point, during @value{GDBN}'s event processing. Events
20931 posted using @code{post_event} will be run in the order in which they
20932 were posted; however, there is no way to know when they will be
20933 processed relative to other events inside @value{GDBN}.
20935 @value{GDBN} is not thread-safe. If your Python program uses multiple
20936 threads, you must be careful to only call @value{GDBN}-specific
20937 functions in the main @value{GDBN} thread. @code{post_event} ensures
20941 (@value{GDBP}) python
20945 > def __init__(self, message):
20946 > self.message = message;
20947 > def __call__(self):
20948 > gdb.write(self.message)
20950 >class MyThread1 (threading.Thread):
20952 > gdb.post_event(Writer("Hello "))
20954 >class MyThread2 (threading.Thread):
20956 > gdb.post_event(Writer("World\n"))
20958 >MyThread1().start()
20959 >MyThread2().start()
20961 (@value{GDBP}) Hello World
20966 @defun write string @r{[}stream{]}
20967 Print a string to @value{GDBN}'s paginated output stream. The
20968 optional @var{stream} determines the stream to print to. The default
20969 stream is @value{GDBN}'s standard output stream. Possible stream
20976 @value{GDBN}'s standard output stream.
20981 @value{GDBN}'s standard error stream.
20986 @value{GDBN}'s log stream (@pxref{Logging Output}).
20989 Writing to @code{sys.stdout} or @code{sys.stderr} will automatically
20990 call this function and will automatically direct the output to the
20996 Flush the buffer of a @value{GDBN} paginated stream so that the
20997 contents are displayed immediately. @value{GDBN} will flush the
20998 contents of a stream automatically when it encounters a newline in the
20999 buffer. The optional @var{stream} determines the stream to flush. The
21000 default stream is @value{GDBN}'s standard output stream. Possible
21007 @value{GDBN}'s standard output stream.
21012 @value{GDBN}'s standard error stream.
21017 @value{GDBN}'s log stream (@pxref{Logging Output}).
21021 Flushing @code{sys.stdout} or @code{sys.stderr} will automatically
21022 call this function for the relevant stream.
21025 @findex gdb.target_charset
21026 @defun target_charset
21027 Return the name of the current target character set (@pxref{Character
21028 Sets}). This differs from @code{gdb.parameter('target-charset')} in
21029 that @samp{auto} is never returned.
21032 @findex gdb.target_wide_charset
21033 @defun target_wide_charset
21034 Return the name of the current target wide character set
21035 (@pxref{Character Sets}). This differs from
21036 @code{gdb.parameter('target-wide-charset')} in that @samp{auto} is
21040 @findex gdb.solib_name
21041 @defun solib_name address
21042 Return the name of the shared library holding the given @var{address}
21043 as a string, or @code{None}.
21046 @findex gdb.decode_line
21047 @defun decode_line @r{[}expression@r{]}
21048 Return locations of the line specified by @var{expression}, or of the
21049 current line if no argument was given. This function returns a Python
21050 tuple containing two elements. The first element contains a string
21051 holding any unparsed section of @var{expression} (or @code{None} if
21052 the expression has been fully parsed). The second element contains
21053 either @code{None} or another tuple that contains all the locations
21054 that match the expression represented as @code{gdb.Symtab_and_line}
21055 objects (@pxref{Symbol Tables In Python}). If @var{expression} is
21056 provided, it is decoded the way that @value{GDBN}'s inbuilt
21057 @code{break} or @code{edit} commands do (@pxref{Specify Location}).
21060 @node Exception Handling
21061 @subsubsection Exception Handling
21062 @cindex python exceptions
21063 @cindex exceptions, python
21065 When executing the @code{python} command, Python exceptions
21066 uncaught within the Python code are translated to calls to
21067 @value{GDBN} error-reporting mechanism. If the command that called
21068 @code{python} does not handle the error, @value{GDBN} will
21069 terminate it and print an error message containing the Python
21070 exception name, the associated value, and the Python call stack
21071 backtrace at the point where the exception was raised. Example:
21074 (@value{GDBP}) python print foo
21075 Traceback (most recent call last):
21076 File "<string>", line 1, in <module>
21077 NameError: name 'foo' is not defined
21080 @value{GDBN} errors that happen in @value{GDBN} commands invoked by
21081 Python code are converted to Python exceptions. The type of the
21082 Python exception depends on the error.
21086 This is the base class for most exceptions generated by @value{GDBN}.
21087 It is derived from @code{RuntimeError}, for compatibility with earlier
21088 versions of @value{GDBN}.
21090 If an error occurring in @value{GDBN} does not fit into some more
21091 specific category, then the generated exception will have this type.
21093 @item gdb.MemoryError
21094 This is a subclass of @code{gdb.error} which is thrown when an
21095 operation tried to access invalid memory in the inferior.
21097 @item KeyboardInterrupt
21098 User interrupt (via @kbd{C-c} or by typing @kbd{q} at a pagination
21099 prompt) is translated to a Python @code{KeyboardInterrupt} exception.
21102 In all cases, your exception handler will see the @value{GDBN} error
21103 message as its value and the Python call stack backtrace at the Python
21104 statement closest to where the @value{GDBN} error occured as the
21107 @findex gdb.GdbError
21108 When implementing @value{GDBN} commands in Python via @code{gdb.Command},
21109 it is useful to be able to throw an exception that doesn't cause a
21110 traceback to be printed. For example, the user may have invoked the
21111 command incorrectly. Use the @code{gdb.GdbError} exception
21112 to handle this case. Example:
21116 >class HelloWorld (gdb.Command):
21117 > """Greet the whole world."""
21118 > def __init__ (self):
21119 > super (HelloWorld, self).__init__ ("hello-world", gdb.COMMAND_OBSCURE)
21120 > def invoke (self, args, from_tty):
21121 > argv = gdb.string_to_argv (args)
21122 > if len (argv) != 0:
21123 > raise gdb.GdbError ("hello-world takes no arguments")
21124 > print "Hello, World!"
21127 (gdb) hello-world 42
21128 hello-world takes no arguments
21131 @node Values From Inferior
21132 @subsubsection Values From Inferior
21133 @cindex values from inferior, with Python
21134 @cindex python, working with values from inferior
21136 @cindex @code{gdb.Value}
21137 @value{GDBN} provides values it obtains from the inferior program in
21138 an object of type @code{gdb.Value}. @value{GDBN} uses this object
21139 for its internal bookkeeping of the inferior's values, and for
21140 fetching values when necessary.
21142 Inferior values that are simple scalars can be used directly in
21143 Python expressions that are valid for the value's data type. Here's
21144 an example for an integer or floating-point value @code{some_val}:
21151 As result of this, @code{bar} will also be a @code{gdb.Value} object
21152 whose values are of the same type as those of @code{some_val}.
21154 Inferior values that are structures or instances of some class can
21155 be accessed using the Python @dfn{dictionary syntax}. For example, if
21156 @code{some_val} is a @code{gdb.Value} instance holding a structure, you
21157 can access its @code{foo} element with:
21160 bar = some_val['foo']
21163 Again, @code{bar} will also be a @code{gdb.Value} object.
21165 A @code{gdb.Value} that represents a function can be executed via
21166 inferior function call. Any arguments provided to the call must match
21167 the function's prototype, and must be provided in the order specified
21170 For example, @code{some_val} is a @code{gdb.Value} instance
21171 representing a function that takes two integers as arguments. To
21172 execute this function, call it like so:
21175 result = some_val (10,20)
21178 Any values returned from a function call will be stored as a
21181 The following attributes are provided:
21184 @defivar Value address
21185 If this object is addressable, this read-only attribute holds a
21186 @code{gdb.Value} object representing the address. Otherwise,
21187 this attribute holds @code{None}.
21190 @cindex optimized out value in Python
21191 @defivar Value is_optimized_out
21192 This read-only boolean attribute is true if the compiler optimized out
21193 this value, thus it is not available for fetching from the inferior.
21196 @defivar Value type
21197 The type of this @code{gdb.Value}. The value of this attribute is a
21198 @code{gdb.Type} object (@pxref{Types In Python}).
21201 @defivar Value dynamic_type
21202 The dynamic type of this @code{gdb.Value}. This uses C@t{++} run-time
21203 type information (@acronym{RTTI}) to determine the dynamic type of the
21204 value. If this value is of class type, it will return the class in
21205 which the value is embedded, if any. If this value is of pointer or
21206 reference to a class type, it will compute the dynamic type of the
21207 referenced object, and return a pointer or reference to that type,
21208 respectively. In all other cases, it will return the value's static
21211 Note that this feature will only work when debugging a C@t{++} program
21212 that includes @acronym{RTTI} for the object in question. Otherwise,
21213 it will just return the static type of the value as in @kbd{ptype foo}
21214 (@pxref{Symbols, ptype}).
21218 The following methods are provided:
21221 @defmethod Value __init__ @var{val}
21222 Many Python values can be converted directly to a @code{gdb.Value} via
21223 this object initializer. Specifically:
21226 @item Python boolean
21227 A Python boolean is converted to the boolean type from the current
21230 @item Python integer
21231 A Python integer is converted to the C @code{long} type for the
21232 current architecture.
21235 A Python long is converted to the C @code{long long} type for the
21236 current architecture.
21239 A Python float is converted to the C @code{double} type for the
21240 current architecture.
21242 @item Python string
21243 A Python string is converted to a target string, using the current
21246 @item @code{gdb.Value}
21247 If @code{val} is a @code{gdb.Value}, then a copy of the value is made.
21249 @item @code{gdb.LazyString}
21250 If @code{val} is a @code{gdb.LazyString} (@pxref{Lazy Strings In
21251 Python}), then the lazy string's @code{value} method is called, and
21252 its result is used.
21256 @defmethod Value cast type
21257 Return a new instance of @code{gdb.Value} that is the result of
21258 casting this instance to the type described by @var{type}, which must
21259 be a @code{gdb.Type} object. If the cast cannot be performed for some
21260 reason, this method throws an exception.
21263 @defmethod Value dereference
21264 For pointer data types, this method returns a new @code{gdb.Value} object
21265 whose contents is the object pointed to by the pointer. For example, if
21266 @code{foo} is a C pointer to an @code{int}, declared in your C program as
21273 then you can use the corresponding @code{gdb.Value} to access what
21274 @code{foo} points to like this:
21277 bar = foo.dereference ()
21280 The result @code{bar} will be a @code{gdb.Value} object holding the
21281 value pointed to by @code{foo}.
21284 @defmethod Value dynamic_cast type
21285 Like @code{Value.cast}, but works as if the C@t{++} @code{dynamic_cast}
21286 operator were used. Consult a C@t{++} reference for details.
21289 @defmethod Value reinterpret_cast type
21290 Like @code{Value.cast}, but works as if the C@t{++} @code{reinterpret_cast}
21291 operator were used. Consult a C@t{++} reference for details.
21294 @defmethod Value string @r{[}encoding@r{]} @r{[}errors@r{]} @r{[}length@r{]}
21295 If this @code{gdb.Value} represents a string, then this method
21296 converts the contents to a Python string. Otherwise, this method will
21297 throw an exception.
21299 Strings are recognized in a language-specific way; whether a given
21300 @code{gdb.Value} represents a string is determined by the current
21303 For C-like languages, a value is a string if it is a pointer to or an
21304 array of characters or ints. The string is assumed to be terminated
21305 by a zero of the appropriate width. However if the optional length
21306 argument is given, the string will be converted to that given length,
21307 ignoring any embedded zeros that the string may contain.
21309 If the optional @var{encoding} argument is given, it must be a string
21310 naming the encoding of the string in the @code{gdb.Value}, such as
21311 @code{"ascii"}, @code{"iso-8859-6"} or @code{"utf-8"}. It accepts
21312 the same encodings as the corresponding argument to Python's
21313 @code{string.decode} method, and the Python codec machinery will be used
21314 to convert the string. If @var{encoding} is not given, or if
21315 @var{encoding} is the empty string, then either the @code{target-charset}
21316 (@pxref{Character Sets}) will be used, or a language-specific encoding
21317 will be used, if the current language is able to supply one.
21319 The optional @var{errors} argument is the same as the corresponding
21320 argument to Python's @code{string.decode} method.
21322 If the optional @var{length} argument is given, the string will be
21323 fetched and converted to the given length.
21326 @defmethod Value lazy_string @r{[}encoding@r{]} @r{[}length@r{]}
21327 If this @code{gdb.Value} represents a string, then this method
21328 converts the contents to a @code{gdb.LazyString} (@pxref{Lazy Strings
21329 In Python}). Otherwise, this method will throw an exception.
21331 If the optional @var{encoding} argument is given, it must be a string
21332 naming the encoding of the @code{gdb.LazyString}. Some examples are:
21333 @samp{ascii}, @samp{iso-8859-6} or @samp{utf-8}. If the
21334 @var{encoding} argument is an encoding that @value{GDBN} does
21335 recognize, @value{GDBN} will raise an error.
21337 When a lazy string is printed, the @value{GDBN} encoding machinery is
21338 used to convert the string during printing. If the optional
21339 @var{encoding} argument is not provided, or is an empty string,
21340 @value{GDBN} will automatically select the encoding most suitable for
21341 the string type. For further information on encoding in @value{GDBN}
21342 please see @ref{Character Sets}.
21344 If the optional @var{length} argument is given, the string will be
21345 fetched and encoded to the length of characters specified. If
21346 the @var{length} argument is not provided, the string will be fetched
21347 and encoded until a null of appropriate width is found.
21351 @node Types In Python
21352 @subsubsection Types In Python
21353 @cindex types in Python
21354 @cindex Python, working with types
21357 @value{GDBN} represents types from the inferior using the class
21360 The following type-related functions are available in the @code{gdb}
21363 @findex gdb.lookup_type
21364 @defun lookup_type name [block]
21365 This function looks up a type by name. @var{name} is the name of the
21366 type to look up. It must be a string.
21368 If @var{block} is given, then @var{name} is looked up in that scope.
21369 Otherwise, it is searched for globally.
21371 Ordinarily, this function will return an instance of @code{gdb.Type}.
21372 If the named type cannot be found, it will throw an exception.
21375 An instance of @code{Type} has the following attributes:
21379 The type code for this type. The type code will be one of the
21380 @code{TYPE_CODE_} constants defined below.
21383 @defivar Type sizeof
21384 The size of this type, in target @code{char} units. Usually, a
21385 target's @code{char} type will be an 8-bit byte. However, on some
21386 unusual platforms, this type may have a different size.
21390 The tag name for this type. The tag name is the name after
21391 @code{struct}, @code{union}, or @code{enum} in C and C@t{++}; not all
21392 languages have this concept. If this type has no tag name, then
21393 @code{None} is returned.
21397 The following methods are provided:
21400 @defmethod Type fields
21401 For structure and union types, this method returns the fields. Range
21402 types have two fields, the minimum and maximum values. Enum types
21403 have one field per enum constant. Function and method types have one
21404 field per parameter. The base types of C@t{++} classes are also
21405 represented as fields. If the type has no fields, or does not fit
21406 into one of these categories, an empty sequence will be returned.
21408 Each field is an object, with some pre-defined attributes:
21411 This attribute is not available for @code{static} fields (as in
21412 C@t{++} or Java). For non-@code{static} fields, the value is the bit
21413 position of the field.
21416 The name of the field, or @code{None} for anonymous fields.
21419 This is @code{True} if the field is artificial, usually meaning that
21420 it was provided by the compiler and not the user. This attribute is
21421 always provided, and is @code{False} if the field is not artificial.
21423 @item is_base_class
21424 This is @code{True} if the field represents a base class of a C@t{++}
21425 structure. This attribute is always provided, and is @code{False}
21426 if the field is not a base class of the type that is the argument of
21427 @code{fields}, or if that type was not a C@t{++} class.
21430 If the field is packed, or is a bitfield, then this will have a
21431 non-zero value, which is the size of the field in bits. Otherwise,
21432 this will be zero; in this case the field's size is given by its type.
21435 The type of the field. This is usually an instance of @code{Type},
21436 but it can be @code{None} in some situations.
21440 @defmethod Type array @var{n1} @r{[}@var{n2}@r{]}
21441 Return a new @code{gdb.Type} object which represents an array of this
21442 type. If one argument is given, it is the inclusive upper bound of
21443 the array; in this case the lower bound is zero. If two arguments are
21444 given, the first argument is the lower bound of the array, and the
21445 second argument is the upper bound of the array. An array's length
21446 must not be negative, but the bounds can be.
21449 @defmethod Type const
21450 Return a new @code{gdb.Type} object which represents a
21451 @code{const}-qualified variant of this type.
21454 @defmethod Type volatile
21455 Return a new @code{gdb.Type} object which represents a
21456 @code{volatile}-qualified variant of this type.
21459 @defmethod Type unqualified
21460 Return a new @code{gdb.Type} object which represents an unqualified
21461 variant of this type. That is, the result is neither @code{const} nor
21465 @defmethod Type range
21466 Return a Python @code{Tuple} object that contains two elements: the
21467 low bound of the argument type and the high bound of that type. If
21468 the type does not have a range, @value{GDBN} will raise a
21469 @code{gdb.error} exception (@pxref{Exception Handling}).
21472 @defmethod Type reference
21473 Return a new @code{gdb.Type} object which represents a reference to this
21477 @defmethod Type pointer
21478 Return a new @code{gdb.Type} object which represents a pointer to this
21482 @defmethod Type strip_typedefs
21483 Return a new @code{gdb.Type} that represents the real type,
21484 after removing all layers of typedefs.
21487 @defmethod Type target
21488 Return a new @code{gdb.Type} object which represents the target type
21491 For a pointer type, the target type is the type of the pointed-to
21492 object. For an array type (meaning C-like arrays), the target type is
21493 the type of the elements of the array. For a function or method type,
21494 the target type is the type of the return value. For a complex type,
21495 the target type is the type of the elements. For a typedef, the
21496 target type is the aliased type.
21498 If the type does not have a target, this method will throw an
21502 @defmethod Type template_argument n [block]
21503 If this @code{gdb.Type} is an instantiation of a template, this will
21504 return a new @code{gdb.Type} which represents the type of the
21505 @var{n}th template argument.
21507 If this @code{gdb.Type} is not a template type, this will throw an
21508 exception. Ordinarily, only C@t{++} code will have template types.
21510 If @var{block} is given, then @var{name} is looked up in that scope.
21511 Otherwise, it is searched for globally.
21516 Each type has a code, which indicates what category this type falls
21517 into. The available type categories are represented by constants
21518 defined in the @code{gdb} module:
21521 @findex TYPE_CODE_PTR
21522 @findex gdb.TYPE_CODE_PTR
21523 @item TYPE_CODE_PTR
21524 The type is a pointer.
21526 @findex TYPE_CODE_ARRAY
21527 @findex gdb.TYPE_CODE_ARRAY
21528 @item TYPE_CODE_ARRAY
21529 The type is an array.
21531 @findex TYPE_CODE_STRUCT
21532 @findex gdb.TYPE_CODE_STRUCT
21533 @item TYPE_CODE_STRUCT
21534 The type is a structure.
21536 @findex TYPE_CODE_UNION
21537 @findex gdb.TYPE_CODE_UNION
21538 @item TYPE_CODE_UNION
21539 The type is a union.
21541 @findex TYPE_CODE_ENUM
21542 @findex gdb.TYPE_CODE_ENUM
21543 @item TYPE_CODE_ENUM
21544 The type is an enum.
21546 @findex TYPE_CODE_FLAGS
21547 @findex gdb.TYPE_CODE_FLAGS
21548 @item TYPE_CODE_FLAGS
21549 A bit flags type, used for things such as status registers.
21551 @findex TYPE_CODE_FUNC
21552 @findex gdb.TYPE_CODE_FUNC
21553 @item TYPE_CODE_FUNC
21554 The type is a function.
21556 @findex TYPE_CODE_INT
21557 @findex gdb.TYPE_CODE_INT
21558 @item TYPE_CODE_INT
21559 The type is an integer type.
21561 @findex TYPE_CODE_FLT
21562 @findex gdb.TYPE_CODE_FLT
21563 @item TYPE_CODE_FLT
21564 A floating point type.
21566 @findex TYPE_CODE_VOID
21567 @findex gdb.TYPE_CODE_VOID
21568 @item TYPE_CODE_VOID
21569 The special type @code{void}.
21571 @findex TYPE_CODE_SET
21572 @findex gdb.TYPE_CODE_SET
21573 @item TYPE_CODE_SET
21576 @findex TYPE_CODE_RANGE
21577 @findex gdb.TYPE_CODE_RANGE
21578 @item TYPE_CODE_RANGE
21579 A range type, that is, an integer type with bounds.
21581 @findex TYPE_CODE_STRING
21582 @findex gdb.TYPE_CODE_STRING
21583 @item TYPE_CODE_STRING
21584 A string type. Note that this is only used for certain languages with
21585 language-defined string types; C strings are not represented this way.
21587 @findex TYPE_CODE_BITSTRING
21588 @findex gdb.TYPE_CODE_BITSTRING
21589 @item TYPE_CODE_BITSTRING
21592 @findex TYPE_CODE_ERROR
21593 @findex gdb.TYPE_CODE_ERROR
21594 @item TYPE_CODE_ERROR
21595 An unknown or erroneous type.
21597 @findex TYPE_CODE_METHOD
21598 @findex gdb.TYPE_CODE_METHOD
21599 @item TYPE_CODE_METHOD
21600 A method type, as found in C@t{++} or Java.
21602 @findex TYPE_CODE_METHODPTR
21603 @findex gdb.TYPE_CODE_METHODPTR
21604 @item TYPE_CODE_METHODPTR
21605 A pointer-to-member-function.
21607 @findex TYPE_CODE_MEMBERPTR
21608 @findex gdb.TYPE_CODE_MEMBERPTR
21609 @item TYPE_CODE_MEMBERPTR
21610 A pointer-to-member.
21612 @findex TYPE_CODE_REF
21613 @findex gdb.TYPE_CODE_REF
21614 @item TYPE_CODE_REF
21617 @findex TYPE_CODE_CHAR
21618 @findex gdb.TYPE_CODE_CHAR
21619 @item TYPE_CODE_CHAR
21622 @findex TYPE_CODE_BOOL
21623 @findex gdb.TYPE_CODE_BOOL
21624 @item TYPE_CODE_BOOL
21627 @findex TYPE_CODE_COMPLEX
21628 @findex gdb.TYPE_CODE_COMPLEX
21629 @item TYPE_CODE_COMPLEX
21630 A complex float type.
21632 @findex TYPE_CODE_TYPEDEF
21633 @findex gdb.TYPE_CODE_TYPEDEF
21634 @item TYPE_CODE_TYPEDEF
21635 A typedef to some other type.
21637 @findex TYPE_CODE_NAMESPACE
21638 @findex gdb.TYPE_CODE_NAMESPACE
21639 @item TYPE_CODE_NAMESPACE
21640 A C@t{++} namespace.
21642 @findex TYPE_CODE_DECFLOAT
21643 @findex gdb.TYPE_CODE_DECFLOAT
21644 @item TYPE_CODE_DECFLOAT
21645 A decimal floating point type.
21647 @findex TYPE_CODE_INTERNAL_FUNCTION
21648 @findex gdb.TYPE_CODE_INTERNAL_FUNCTION
21649 @item TYPE_CODE_INTERNAL_FUNCTION
21650 A function internal to @value{GDBN}. This is the type used to represent
21651 convenience functions.
21654 Further support for types is provided in the @code{gdb.types}
21655 Python module (@pxref{gdb.types}).
21657 @node Pretty Printing API
21658 @subsubsection Pretty Printing API
21660 An example output is provided (@pxref{Pretty Printing}).
21662 A pretty-printer is just an object that holds a value and implements a
21663 specific interface, defined here.
21665 @defop Operation {pretty printer} children (self)
21666 @value{GDBN} will call this method on a pretty-printer to compute the
21667 children of the pretty-printer's value.
21669 This method must return an object conforming to the Python iterator
21670 protocol. Each item returned by the iterator must be a tuple holding
21671 two elements. The first element is the ``name'' of the child; the
21672 second element is the child's value. The value can be any Python
21673 object which is convertible to a @value{GDBN} value.
21675 This method is optional. If it does not exist, @value{GDBN} will act
21676 as though the value has no children.
21679 @defop Operation {pretty printer} display_hint (self)
21680 The CLI may call this method and use its result to change the
21681 formatting of a value. The result will also be supplied to an MI
21682 consumer as a @samp{displayhint} attribute of the variable being
21685 This method is optional. If it does exist, this method must return a
21688 Some display hints are predefined by @value{GDBN}:
21692 Indicate that the object being printed is ``array-like''. The CLI
21693 uses this to respect parameters such as @code{set print elements} and
21694 @code{set print array}.
21697 Indicate that the object being printed is ``map-like'', and that the
21698 children of this value can be assumed to alternate between keys and
21702 Indicate that the object being printed is ``string-like''. If the
21703 printer's @code{to_string} method returns a Python string of some
21704 kind, then @value{GDBN} will call its internal language-specific
21705 string-printing function to format the string. For the CLI this means
21706 adding quotation marks, possibly escaping some characters, respecting
21707 @code{set print elements}, and the like.
21711 @defop Operation {pretty printer} to_string (self)
21712 @value{GDBN} will call this method to display the string
21713 representation of the value passed to the object's constructor.
21715 When printing from the CLI, if the @code{to_string} method exists,
21716 then @value{GDBN} will prepend its result to the values returned by
21717 @code{children}. Exactly how this formatting is done is dependent on
21718 the display hint, and may change as more hints are added. Also,
21719 depending on the print settings (@pxref{Print Settings}), the CLI may
21720 print just the result of @code{to_string} in a stack trace, omitting
21721 the result of @code{children}.
21723 If this method returns a string, it is printed verbatim.
21725 Otherwise, if this method returns an instance of @code{gdb.Value},
21726 then @value{GDBN} prints this value. This may result in a call to
21727 another pretty-printer.
21729 If instead the method returns a Python value which is convertible to a
21730 @code{gdb.Value}, then @value{GDBN} performs the conversion and prints
21731 the resulting value. Again, this may result in a call to another
21732 pretty-printer. Python scalars (integers, floats, and booleans) and
21733 strings are convertible to @code{gdb.Value}; other types are not.
21735 Finally, if this method returns @code{None} then no further operations
21736 are peformed in this method and nothing is printed.
21738 If the result is not one of these types, an exception is raised.
21741 @value{GDBN} provides a function which can be used to look up the
21742 default pretty-printer for a @code{gdb.Value}:
21744 @findex gdb.default_visualizer
21745 @defun default_visualizer value
21746 This function takes a @code{gdb.Value} object as an argument. If a
21747 pretty-printer for this value exists, then it is returned. If no such
21748 printer exists, then this returns @code{None}.
21751 @node Selecting Pretty-Printers
21752 @subsubsection Selecting Pretty-Printers
21754 The Python list @code{gdb.pretty_printers} contains an array of
21755 functions or callable objects that have been registered via addition
21756 as a pretty-printer. Printers in this list are called @code{global}
21757 printers, they're available when debugging all inferiors.
21758 Each @code{gdb.Progspace} contains a @code{pretty_printers} attribute.
21759 Each @code{gdb.Objfile} also contains a @code{pretty_printers}
21762 Each function on these lists is passed a single @code{gdb.Value}
21763 argument and should return a pretty-printer object conforming to the
21764 interface definition above (@pxref{Pretty Printing API}). If a function
21765 cannot create a pretty-printer for the value, it should return
21768 @value{GDBN} first checks the @code{pretty_printers} attribute of each
21769 @code{gdb.Objfile} in the current program space and iteratively calls
21770 each enabled lookup routine in the list for that @code{gdb.Objfile}
21771 until it receives a pretty-printer object.
21772 If no pretty-printer is found in the objfile lists, @value{GDBN} then
21773 searches the pretty-printer list of the current program space,
21774 calling each enabled function until an object is returned.
21775 After these lists have been exhausted, it tries the global
21776 @code{gdb.pretty_printers} list, again calling each enabled function until an
21777 object is returned.
21779 The order in which the objfiles are searched is not specified. For a
21780 given list, functions are always invoked from the head of the list,
21781 and iterated over sequentially until the end of the list, or a printer
21782 object is returned.
21784 For various reasons a pretty-printer may not work.
21785 For example, the underlying data structure may have changed and
21786 the pretty-printer is out of date.
21788 The consequences of a broken pretty-printer are severe enough that
21789 @value{GDBN} provides support for enabling and disabling individual
21790 printers. For example, if @code{print frame-arguments} is on,
21791 a backtrace can become highly illegible if any argument is printed
21792 with a broken printer.
21794 Pretty-printers are enabled and disabled by attaching an @code{enabled}
21795 attribute to the registered function or callable object. If this attribute
21796 is present and its value is @code{False}, the printer is disabled, otherwise
21797 the printer is enabled.
21799 @node Writing a Pretty-Printer
21800 @subsubsection Writing a Pretty-Printer
21801 @cindex writing a pretty-printer
21803 A pretty-printer consists of two parts: a lookup function to detect
21804 if the type is supported, and the printer itself.
21806 Here is an example showing how a @code{std::string} printer might be
21807 written. @xref{Pretty Printing API}, for details on the API this class
21811 class StdStringPrinter(object):
21812 "Print a std::string"
21814 def __init__(self, val):
21817 def to_string(self):
21818 return self.val['_M_dataplus']['_M_p']
21820 def display_hint(self):
21824 And here is an example showing how a lookup function for the printer
21825 example above might be written.
21828 def str_lookup_function(val):
21829 lookup_tag = val.type.tag
21830 if lookup_tag == None:
21832 regex = re.compile("^std::basic_string<char,.*>$")
21833 if regex.match(lookup_tag):
21834 return StdStringPrinter(val)
21838 The example lookup function extracts the value's type, and attempts to
21839 match it to a type that it can pretty-print. If it is a type the
21840 printer can pretty-print, it will return a printer object. If not, it
21841 returns @code{None}.
21843 We recommend that you put your core pretty-printers into a Python
21844 package. If your pretty-printers are for use with a library, we
21845 further recommend embedding a version number into the package name.
21846 This practice will enable @value{GDBN} to load multiple versions of
21847 your pretty-printers at the same time, because they will have
21850 You should write auto-loaded code (@pxref{Auto-loading}) such that it
21851 can be evaluated multiple times without changing its meaning. An
21852 ideal auto-load file will consist solely of @code{import}s of your
21853 printer modules, followed by a call to a register pretty-printers with
21854 the current objfile.
21856 Taken as a whole, this approach will scale nicely to multiple
21857 inferiors, each potentially using a different library version.
21858 Embedding a version number in the Python package name will ensure that
21859 @value{GDBN} is able to load both sets of printers simultaneously.
21860 Then, because the search for pretty-printers is done by objfile, and
21861 because your auto-loaded code took care to register your library's
21862 printers with a specific objfile, @value{GDBN} will find the correct
21863 printers for the specific version of the library used by each
21866 To continue the @code{std::string} example (@pxref{Pretty Printing API}),
21867 this code might appear in @code{gdb.libstdcxx.v6}:
21870 def register_printers(objfile):
21871 objfile.pretty_printers.add(str_lookup_function)
21875 And then the corresponding contents of the auto-load file would be:
21878 import gdb.libstdcxx.v6
21879 gdb.libstdcxx.v6.register_printers(gdb.current_objfile())
21882 The previous example illustrates a basic pretty-printer.
21883 There are a few things that can be improved on.
21884 The printer doesn't have a name, making it hard to identify in a
21885 list of installed printers. The lookup function has a name, but
21886 lookup functions can have arbitrary, even identical, names.
21888 Second, the printer only handles one type, whereas a library typically has
21889 several types. One could install a lookup function for each desired type
21890 in the library, but one could also have a single lookup function recognize
21891 several types. The latter is the conventional way this is handled.
21892 If a pretty-printer can handle multiple data types, then its
21893 @dfn{subprinters} are the printers for the individual data types.
21895 The @code{gdb.printing} module provides a formal way of solving these
21896 problems (@pxref{gdb.printing}).
21897 Here is another example that handles multiple types.
21899 These are the types we are going to pretty-print:
21902 struct foo @{ int a, b; @};
21903 struct bar @{ struct foo x, y; @};
21906 Here are the printers:
21910 """Print a foo object."""
21912 def __init__(self, val):
21915 def to_string(self):
21916 return ("a=<" + str(self.val["a"]) +
21917 "> b=<" + str(self.val["b"]) + ">")
21920 """Print a bar object."""
21922 def __init__(self, val):
21925 def to_string(self):
21926 return ("x=<" + str(self.val["x"]) +
21927 "> y=<" + str(self.val["y"]) + ">")
21930 This example doesn't need a lookup function, that is handled by the
21931 @code{gdb.printing} module. Instead a function is provided to build up
21932 the object that handles the lookup.
21935 import gdb.printing
21937 def build_pretty_printer():
21938 pp = gdb.printing.RegexpCollectionPrettyPrinter(
21940 pp.add_printer('foo', '^foo$', fooPrinter)
21941 pp.add_printer('bar', '^bar$', barPrinter)
21945 And here is the autoload support:
21948 import gdb.printing
21950 gdb.printing.register_pretty_printer(
21951 gdb.current_objfile(),
21952 my_library.build_pretty_printer())
21955 Finally, when this printer is loaded into @value{GDBN}, here is the
21956 corresponding output of @samp{info pretty-printer}:
21959 (gdb) info pretty-printer
21966 @node Inferiors In Python
21967 @subsubsection Inferiors In Python
21968 @cindex inferiors in Python
21970 @findex gdb.Inferior
21971 Programs which are being run under @value{GDBN} are called inferiors
21972 (@pxref{Inferiors and Programs}). Python scripts can access
21973 information about and manipulate inferiors controlled by @value{GDBN}
21974 via objects of the @code{gdb.Inferior} class.
21976 The following inferior-related functions are available in the @code{gdb}
21980 Return a tuple containing all inferior objects.
21983 A @code{gdb.Inferior} object has the following attributes:
21986 @defivar Inferior num
21987 ID of inferior, as assigned by GDB.
21990 @defivar Inferior pid
21991 Process ID of the inferior, as assigned by the underlying operating
21995 @defivar Inferior was_attached
21996 Boolean signaling whether the inferior was created using `attach', or
21997 started by @value{GDBN} itself.
22001 A @code{gdb.Inferior} object has the following methods:
22004 @defmethod Inferior is_valid
22005 Returns @code{True} if the @code{gdb.Inferior} object is valid,
22006 @code{False} if not. A @code{gdb.Inferior} object will become invalid
22007 if the inferior no longer exists within @value{GDBN}. All other
22008 @code{gdb.Inferior} methods will throw an exception if it is invalid
22009 at the time the method is called.
22012 @defmethod Inferior threads
22013 This method returns a tuple holding all the threads which are valid
22014 when it is called. If there are no valid threads, the method will
22015 return an empty tuple.
22018 @findex gdb.read_memory
22019 @defmethod Inferior read_memory address length
22020 Read @var{length} bytes of memory from the inferior, starting at
22021 @var{address}. Returns a buffer object, which behaves much like an array
22022 or a string. It can be modified and given to the @code{gdb.write_memory}
22026 @findex gdb.write_memory
22027 @defmethod Inferior write_memory address buffer @r{[}length@r{]}
22028 Write the contents of @var{buffer} to the inferior, starting at
22029 @var{address}. The @var{buffer} parameter must be a Python object
22030 which supports the buffer protocol, i.e., a string, an array or the
22031 object returned from @code{gdb.read_memory}. If given, @var{length}
22032 determines the number of bytes from @var{buffer} to be written.
22035 @findex gdb.search_memory
22036 @defmethod Inferior search_memory address length pattern
22037 Search a region of the inferior memory starting at @var{address} with
22038 the given @var{length} using the search pattern supplied in
22039 @var{pattern}. The @var{pattern} parameter must be a Python object
22040 which supports the buffer protocol, i.e., a string, an array or the
22041 object returned from @code{gdb.read_memory}. Returns a Python @code{Long}
22042 containing the address where the pattern was found, or @code{None} if
22043 the pattern could not be found.
22047 @node Events In Python
22048 @subsubsection Events In Python
22049 @cindex inferior events in Python
22051 @value{GDBN} provides a general event facility so that Python code can be
22052 notified of various state changes, particularly changes that occur in
22055 An @dfn{event} is just an object that describes some state change. The
22056 type of the object and its attributes will vary depending on the details
22057 of the change. All the existing events are described below.
22059 In order to be notified of an event, you must register an event handler
22060 with an @dfn{event registry}. An event registry is an object in the
22061 @code{gdb.events} module which dispatches particular events. A registry
22062 provides methods to register and unregister event handlers:
22065 @defmethod EventRegistry connect object
22066 Add the given callable @var{object} to the registry. This object will be
22067 called when an event corresponding to this registry occurs.
22070 @defmethod EventRegistry disconnect object
22071 Remove the given @var{object} from the registry. Once removed, the object
22072 will no longer receive notifications of events.
22076 Here is an example:
22079 def exit_handler (event):
22080 print "event type: exit"
22081 print "exit code: %d" % (event.exit_code)
22083 gdb.events.exited.connect (exit_handler)
22086 In the above example we connect our handler @code{exit_handler} to the
22087 registry @code{events.exited}. Once connected, @code{exit_handler} gets
22088 called when the inferior exits. The argument @dfn{event} in this example is
22089 of type @code{gdb.ExitedEvent}. As you can see in the example the
22090 @code{ExitedEvent} object has an attribute which indicates the exit code of
22093 The following is a listing of the event registries that are available and
22094 details of the events they emit:
22099 Emits @code{gdb.ThreadEvent}.
22101 Some events can be thread specific when @value{GDBN} is running in non-stop
22102 mode. When represented in Python, these events all extend
22103 @code{gdb.ThreadEvent}. Note, this event is not emitted directly; instead,
22104 events which are emitted by this or other modules might extend this event.
22105 Examples of these events are @code{gdb.BreakpointEvent} and
22106 @code{gdb.ContinueEvent}.
22109 @defivar ThreadEvent inferior_thread
22110 In non-stop mode this attribute will be set to the specific thread which was
22111 involved in the emitted event. Otherwise, it will be set to @code{None}.
22115 Emits @code{gdb.ContinueEvent} which extends @code{gdb.ThreadEvent}.
22117 This event indicates that the inferior has been continued after a stop. For
22118 inherited attribute refer to @code{gdb.ThreadEvent} above.
22120 @item events.exited
22121 Emits @code{events.ExitedEvent} which indicates that the inferior has exited.
22122 @code{events.ExitedEvent} has one attribute:
22124 @defivar ExitedEvent exit_code
22125 An integer representing the exit code which the inferior has returned.
22130 Emits @code{gdb.StopEvent} which extends @code{gdb.ThreadEvent}.
22132 Indicates that the inferior has stopped. All events emitted by this registry
22133 extend StopEvent. As a child of @code{gdb.ThreadEvent}, @code{gdb.StopEvent}
22134 will indicate the stopped thread when @value{GDBN} is running in non-stop
22135 mode. Refer to @code{gdb.ThreadEvent} above for more details.
22137 Emits @code{gdb.SignalEvent} which extends @code{gdb.StopEvent}.
22139 This event indicates that the inferior or one of its threads has received as
22140 signal. @code{gdb.SignalEvent} has the following attributes:
22143 @defivar SignalEvent stop_signal
22144 A string representing the signal received by the inferior. A list of possible
22145 signal values can be obtained by running the command @code{info signals} in
22146 the @value{GDBN} command prompt.
22150 Also emits @code{gdb.BreakpointEvent} which extends @code{gdb.StopEvent}.
22152 @code{gdb.BreakpointEvent} event indicates that a breakpoint has been hit, and
22153 has the following attributes:
22156 @defivar BreakpointEvent breakpoint
22157 A reference to the breakpoint that was hit of type @code{gdb.Breakpoint}.
22158 @xref{Breakpoints In Python}, for details of the @code{gdb.Breakpoint} object.
22164 @node Threads In Python
22165 @subsubsection Threads In Python
22166 @cindex threads in python
22168 @findex gdb.InferiorThread
22169 Python scripts can access information about, and manipulate inferior threads
22170 controlled by @value{GDBN}, via objects of the @code{gdb.InferiorThread} class.
22172 The following thread-related functions are available in the @code{gdb}
22175 @findex gdb.selected_thread
22176 @defun selected_thread
22177 This function returns the thread object for the selected thread. If there
22178 is no selected thread, this will return @code{None}.
22181 A @code{gdb.InferiorThread} object has the following attributes:
22184 @defivar InferiorThread name
22185 The name of the thread. If the user specified a name using
22186 @code{thread name}, then this returns that name. Otherwise, if an
22187 OS-supplied name is available, then it is returned. Otherwise, this
22188 returns @code{None}.
22190 This attribute can be assigned to. The new value must be a string
22191 object, which sets the new name, or @code{None}, which removes any
22192 user-specified thread name.
22195 @defivar InferiorThread num
22196 ID of the thread, as assigned by GDB.
22199 @defivar InferiorThread ptid
22200 ID of the thread, as assigned by the operating system. This attribute is a
22201 tuple containing three integers. The first is the Process ID (PID); the second
22202 is the Lightweight Process ID (LWPID), and the third is the Thread ID (TID).
22203 Either the LWPID or TID may be 0, which indicates that the operating system
22204 does not use that identifier.
22208 A @code{gdb.InferiorThread} object has the following methods:
22211 @defmethod InferiorThread is_valid
22212 Returns @code{True} if the @code{gdb.InferiorThread} object is valid,
22213 @code{False} if not. A @code{gdb.InferiorThread} object will become
22214 invalid if the thread exits, or the inferior that the thread belongs
22215 is deleted. All other @code{gdb.InferiorThread} methods will throw an
22216 exception if it is invalid at the time the method is called.
22219 @defmethod InferiorThread switch
22220 This changes @value{GDBN}'s currently selected thread to the one represented
22224 @defmethod InferiorThread is_stopped
22225 Return a Boolean indicating whether the thread is stopped.
22228 @defmethod InferiorThread is_running
22229 Return a Boolean indicating whether the thread is running.
22232 @defmethod InferiorThread is_exited
22233 Return a Boolean indicating whether the thread is exited.
22237 @node Commands In Python
22238 @subsubsection Commands In Python
22240 @cindex commands in python
22241 @cindex python commands
22242 You can implement new @value{GDBN} CLI commands in Python. A CLI
22243 command is implemented using an instance of the @code{gdb.Command}
22244 class, most commonly using a subclass.
22246 @defmethod Command __init__ name @var{command_class} @r{[}@var{completer_class}@r{]} @r{[}@var{prefix}@r{]}
22247 The object initializer for @code{Command} registers the new command
22248 with @value{GDBN}. This initializer is normally invoked from the
22249 subclass' own @code{__init__} method.
22251 @var{name} is the name of the command. If @var{name} consists of
22252 multiple words, then the initial words are looked for as prefix
22253 commands. In this case, if one of the prefix commands does not exist,
22254 an exception is raised.
22256 There is no support for multi-line commands.
22258 @var{command_class} should be one of the @samp{COMMAND_} constants
22259 defined below. This argument tells @value{GDBN} how to categorize the
22260 new command in the help system.
22262 @var{completer_class} is an optional argument. If given, it should be
22263 one of the @samp{COMPLETE_} constants defined below. This argument
22264 tells @value{GDBN} how to perform completion for this command. If not
22265 given, @value{GDBN} will attempt to complete using the object's
22266 @code{complete} method (see below); if no such method is found, an
22267 error will occur when completion is attempted.
22269 @var{prefix} is an optional argument. If @code{True}, then the new
22270 command is a prefix command; sub-commands of this command may be
22273 The help text for the new command is taken from the Python
22274 documentation string for the command's class, if there is one. If no
22275 documentation string is provided, the default value ``This command is
22276 not documented.'' is used.
22279 @cindex don't repeat Python command
22280 @defmethod Command dont_repeat
22281 By default, a @value{GDBN} command is repeated when the user enters a
22282 blank line at the command prompt. A command can suppress this
22283 behavior by invoking the @code{dont_repeat} method. This is similar
22284 to the user command @code{dont-repeat}, see @ref{Define, dont-repeat}.
22287 @defmethod Command invoke argument from_tty
22288 This method is called by @value{GDBN} when this command is invoked.
22290 @var{argument} is a string. It is the argument to the command, after
22291 leading and trailing whitespace has been stripped.
22293 @var{from_tty} is a boolean argument. When true, this means that the
22294 command was entered by the user at the terminal; when false it means
22295 that the command came from elsewhere.
22297 If this method throws an exception, it is turned into a @value{GDBN}
22298 @code{error} call. Otherwise, the return value is ignored.
22300 @findex gdb.string_to_argv
22301 To break @var{argument} up into an argv-like string use
22302 @code{gdb.string_to_argv}. This function behaves identically to
22303 @value{GDBN}'s internal argument lexer @code{buildargv}.
22304 It is recommended to use this for consistency.
22305 Arguments are separated by spaces and may be quoted.
22309 print gdb.string_to_argv ("1 2\ \\\"3 '4 \"5' \"6 '7\"")
22310 ['1', '2 "3', '4 "5', "6 '7"]
22315 @cindex completion of Python commands
22316 @defmethod Command complete text word
22317 This method is called by @value{GDBN} when the user attempts
22318 completion on this command. All forms of completion are handled by
22319 this method, that is, the @key{TAB} and @key{M-?} key bindings
22320 (@pxref{Completion}), and the @code{complete} command (@pxref{Help,
22323 The arguments @var{text} and @var{word} are both strings. @var{text}
22324 holds the complete command line up to the cursor's location.
22325 @var{word} holds the last word of the command line; this is computed
22326 using a word-breaking heuristic.
22328 The @code{complete} method can return several values:
22331 If the return value is a sequence, the contents of the sequence are
22332 used as the completions. It is up to @code{complete} to ensure that the
22333 contents actually do complete the word. A zero-length sequence is
22334 allowed, it means that there were no completions available. Only
22335 string elements of the sequence are used; other elements in the
22336 sequence are ignored.
22339 If the return value is one of the @samp{COMPLETE_} constants defined
22340 below, then the corresponding @value{GDBN}-internal completion
22341 function is invoked, and its result is used.
22344 All other results are treated as though there were no available
22349 When a new command is registered, it must be declared as a member of
22350 some general class of commands. This is used to classify top-level
22351 commands in the on-line help system; note that prefix commands are not
22352 listed under their own category but rather that of their top-level
22353 command. The available classifications are represented by constants
22354 defined in the @code{gdb} module:
22357 @findex COMMAND_NONE
22358 @findex gdb.COMMAND_NONE
22360 The command does not belong to any particular class. A command in
22361 this category will not be displayed in any of the help categories.
22363 @findex COMMAND_RUNNING
22364 @findex gdb.COMMAND_RUNNING
22365 @item COMMAND_RUNNING
22366 The command is related to running the inferior. For example,
22367 @code{start}, @code{step}, and @code{continue} are in this category.
22368 Type @kbd{help running} at the @value{GDBN} prompt to see a list of
22369 commands in this category.
22371 @findex COMMAND_DATA
22372 @findex gdb.COMMAND_DATA
22374 The command is related to data or variables. For example,
22375 @code{call}, @code{find}, and @code{print} are in this category. Type
22376 @kbd{help data} at the @value{GDBN} prompt to see a list of commands
22379 @findex COMMAND_STACK
22380 @findex gdb.COMMAND_STACK
22381 @item COMMAND_STACK
22382 The command has to do with manipulation of the stack. For example,
22383 @code{backtrace}, @code{frame}, and @code{return} are in this
22384 category. Type @kbd{help stack} at the @value{GDBN} prompt to see a
22385 list of commands in this category.
22387 @findex COMMAND_FILES
22388 @findex gdb.COMMAND_FILES
22389 @item COMMAND_FILES
22390 This class is used for file-related commands. For example,
22391 @code{file}, @code{list} and @code{section} are in this category.
22392 Type @kbd{help files} at the @value{GDBN} prompt to see a list of
22393 commands in this category.
22395 @findex COMMAND_SUPPORT
22396 @findex gdb.COMMAND_SUPPORT
22397 @item COMMAND_SUPPORT
22398 This should be used for ``support facilities'', generally meaning
22399 things that are useful to the user when interacting with @value{GDBN},
22400 but not related to the state of the inferior. For example,
22401 @code{help}, @code{make}, and @code{shell} are in this category. Type
22402 @kbd{help support} at the @value{GDBN} prompt to see a list of
22403 commands in this category.
22405 @findex COMMAND_STATUS
22406 @findex gdb.COMMAND_STATUS
22407 @item COMMAND_STATUS
22408 The command is an @samp{info}-related command, that is, related to the
22409 state of @value{GDBN} itself. For example, @code{info}, @code{macro},
22410 and @code{show} are in this category. Type @kbd{help status} at the
22411 @value{GDBN} prompt to see a list of commands in this category.
22413 @findex COMMAND_BREAKPOINTS
22414 @findex gdb.COMMAND_BREAKPOINTS
22415 @item COMMAND_BREAKPOINTS
22416 The command has to do with breakpoints. For example, @code{break},
22417 @code{clear}, and @code{delete} are in this category. Type @kbd{help
22418 breakpoints} at the @value{GDBN} prompt to see a list of commands in
22421 @findex COMMAND_TRACEPOINTS
22422 @findex gdb.COMMAND_TRACEPOINTS
22423 @item COMMAND_TRACEPOINTS
22424 The command has to do with tracepoints. For example, @code{trace},
22425 @code{actions}, and @code{tfind} are in this category. Type
22426 @kbd{help tracepoints} at the @value{GDBN} prompt to see a list of
22427 commands in this category.
22429 @findex COMMAND_OBSCURE
22430 @findex gdb.COMMAND_OBSCURE
22431 @item COMMAND_OBSCURE
22432 The command is only used in unusual circumstances, or is not of
22433 general interest to users. For example, @code{checkpoint},
22434 @code{fork}, and @code{stop} are in this category. Type @kbd{help
22435 obscure} at the @value{GDBN} prompt to see a list of commands in this
22438 @findex COMMAND_MAINTENANCE
22439 @findex gdb.COMMAND_MAINTENANCE
22440 @item COMMAND_MAINTENANCE
22441 The command is only useful to @value{GDBN} maintainers. The
22442 @code{maintenance} and @code{flushregs} commands are in this category.
22443 Type @kbd{help internals} at the @value{GDBN} prompt to see a list of
22444 commands in this category.
22447 A new command can use a predefined completion function, either by
22448 specifying it via an argument at initialization, or by returning it
22449 from the @code{complete} method. These predefined completion
22450 constants are all defined in the @code{gdb} module:
22453 @findex COMPLETE_NONE
22454 @findex gdb.COMPLETE_NONE
22455 @item COMPLETE_NONE
22456 This constant means that no completion should be done.
22458 @findex COMPLETE_FILENAME
22459 @findex gdb.COMPLETE_FILENAME
22460 @item COMPLETE_FILENAME
22461 This constant means that filename completion should be performed.
22463 @findex COMPLETE_LOCATION
22464 @findex gdb.COMPLETE_LOCATION
22465 @item COMPLETE_LOCATION
22466 This constant means that location completion should be done.
22467 @xref{Specify Location}.
22469 @findex COMPLETE_COMMAND
22470 @findex gdb.COMPLETE_COMMAND
22471 @item COMPLETE_COMMAND
22472 This constant means that completion should examine @value{GDBN}
22475 @findex COMPLETE_SYMBOL
22476 @findex gdb.COMPLETE_SYMBOL
22477 @item COMPLETE_SYMBOL
22478 This constant means that completion should be done using symbol names
22482 The following code snippet shows how a trivial CLI command can be
22483 implemented in Python:
22486 class HelloWorld (gdb.Command):
22487 """Greet the whole world."""
22489 def __init__ (self):
22490 super (HelloWorld, self).__init__ ("hello-world", gdb.COMMAND_OBSCURE)
22492 def invoke (self, arg, from_tty):
22493 print "Hello, World!"
22498 The last line instantiates the class, and is necessary to trigger the
22499 registration of the command with @value{GDBN}. Depending on how the
22500 Python code is read into @value{GDBN}, you may need to import the
22501 @code{gdb} module explicitly.
22503 @node Parameters In Python
22504 @subsubsection Parameters In Python
22506 @cindex parameters in python
22507 @cindex python parameters
22508 @tindex gdb.Parameter
22510 You can implement new @value{GDBN} parameters using Python. A new
22511 parameter is implemented as an instance of the @code{gdb.Parameter}
22514 Parameters are exposed to the user via the @code{set} and
22515 @code{show} commands. @xref{Help}.
22517 There are many parameters that already exist and can be set in
22518 @value{GDBN}. Two examples are: @code{set follow fork} and
22519 @code{set charset}. Setting these parameters influences certain
22520 behavior in @value{GDBN}. Similarly, you can define parameters that
22521 can be used to influence behavior in custom Python scripts and commands.
22523 @defmethod Parameter __init__ name @var{command-class} @var{parameter-class} @r{[}@var{enum-sequence}@r{]}
22524 The object initializer for @code{Parameter} registers the new
22525 parameter with @value{GDBN}. This initializer is normally invoked
22526 from the subclass' own @code{__init__} method.
22528 @var{name} is the name of the new parameter. If @var{name} consists
22529 of multiple words, then the initial words are looked for as prefix
22530 parameters. An example of this can be illustrated with the
22531 @code{set print} set of parameters. If @var{name} is
22532 @code{print foo}, then @code{print} will be searched as the prefix
22533 parameter. In this case the parameter can subsequently be accessed in
22534 @value{GDBN} as @code{set print foo}.
22536 If @var{name} consists of multiple words, and no prefix parameter group
22537 can be found, an exception is raised.
22539 @var{command-class} should be one of the @samp{COMMAND_} constants
22540 (@pxref{Commands In Python}). This argument tells @value{GDBN} how to
22541 categorize the new parameter in the help system.
22543 @var{parameter-class} should be one of the @samp{PARAM_} constants
22544 defined below. This argument tells @value{GDBN} the type of the new
22545 parameter; this information is used for input validation and
22548 If @var{parameter-class} is @code{PARAM_ENUM}, then
22549 @var{enum-sequence} must be a sequence of strings. These strings
22550 represent the possible values for the parameter.
22552 If @var{parameter-class} is not @code{PARAM_ENUM}, then the presence
22553 of a fourth argument will cause an exception to be thrown.
22555 The help text for the new parameter is taken from the Python
22556 documentation string for the parameter's class, if there is one. If
22557 there is no documentation string, a default value is used.
22560 @defivar Parameter set_doc
22561 If this attribute exists, and is a string, then its value is used as
22562 the help text for this parameter's @code{set} command. The value is
22563 examined when @code{Parameter.__init__} is invoked; subsequent changes
22567 @defivar Parameter show_doc
22568 If this attribute exists, and is a string, then its value is used as
22569 the help text for this parameter's @code{show} command. The value is
22570 examined when @code{Parameter.__init__} is invoked; subsequent changes
22574 @defivar Parameter value
22575 The @code{value} attribute holds the underlying value of the
22576 parameter. It can be read and assigned to just as any other
22577 attribute. @value{GDBN} does validation when assignments are made.
22580 There are two methods that should be implemented in any
22581 @code{Parameter} class. These are:
22583 @defop Operation {parameter} get_set_string self
22584 @value{GDBN} will call this method when a @var{parameter}'s value has
22585 been changed via the @code{set} API (for example, @kbd{set foo off}).
22586 The @code{value} attribute has already been populated with the new
22587 value and may be used in output. This method must return a string.
22590 @defop Operation {parameter} get_show_string self svalue
22591 @value{GDBN} will call this method when a @var{parameter}'s
22592 @code{show} API has been invoked (for example, @kbd{show foo}). The
22593 argument @code{svalue} receives the string representation of the
22594 current value. This method must return a string.
22597 When a new parameter is defined, its type must be specified. The
22598 available types are represented by constants defined in the @code{gdb}
22602 @findex PARAM_BOOLEAN
22603 @findex gdb.PARAM_BOOLEAN
22604 @item PARAM_BOOLEAN
22605 The value is a plain boolean. The Python boolean values, @code{True}
22606 and @code{False} are the only valid values.
22608 @findex PARAM_AUTO_BOOLEAN
22609 @findex gdb.PARAM_AUTO_BOOLEAN
22610 @item PARAM_AUTO_BOOLEAN
22611 The value has three possible states: true, false, and @samp{auto}. In
22612 Python, true and false are represented using boolean constants, and
22613 @samp{auto} is represented using @code{None}.
22615 @findex PARAM_UINTEGER
22616 @findex gdb.PARAM_UINTEGER
22617 @item PARAM_UINTEGER
22618 The value is an unsigned integer. The value of 0 should be
22619 interpreted to mean ``unlimited''.
22621 @findex PARAM_INTEGER
22622 @findex gdb.PARAM_INTEGER
22623 @item PARAM_INTEGER
22624 The value is a signed integer. The value of 0 should be interpreted
22625 to mean ``unlimited''.
22627 @findex PARAM_STRING
22628 @findex gdb.PARAM_STRING
22630 The value is a string. When the user modifies the string, any escape
22631 sequences, such as @samp{\t}, @samp{\f}, and octal escapes, are
22632 translated into corresponding characters and encoded into the current
22635 @findex PARAM_STRING_NOESCAPE
22636 @findex gdb.PARAM_STRING_NOESCAPE
22637 @item PARAM_STRING_NOESCAPE
22638 The value is a string. When the user modifies the string, escapes are
22639 passed through untranslated.
22641 @findex PARAM_OPTIONAL_FILENAME
22642 @findex gdb.PARAM_OPTIONAL_FILENAME
22643 @item PARAM_OPTIONAL_FILENAME
22644 The value is a either a filename (a string), or @code{None}.
22646 @findex PARAM_FILENAME
22647 @findex gdb.PARAM_FILENAME
22648 @item PARAM_FILENAME
22649 The value is a filename. This is just like
22650 @code{PARAM_STRING_NOESCAPE}, but uses file names for completion.
22652 @findex PARAM_ZINTEGER
22653 @findex gdb.PARAM_ZINTEGER
22654 @item PARAM_ZINTEGER
22655 The value is an integer. This is like @code{PARAM_INTEGER}, except 0
22656 is interpreted as itself.
22659 @findex gdb.PARAM_ENUM
22661 The value is a string, which must be one of a collection string
22662 constants provided when the parameter is created.
22665 @node Functions In Python
22666 @subsubsection Writing new convenience functions
22668 @cindex writing convenience functions
22669 @cindex convenience functions in python
22670 @cindex python convenience functions
22671 @tindex gdb.Function
22673 You can implement new convenience functions (@pxref{Convenience Vars})
22674 in Python. A convenience function is an instance of a subclass of the
22675 class @code{gdb.Function}.
22677 @defmethod Function __init__ name
22678 The initializer for @code{Function} registers the new function with
22679 @value{GDBN}. The argument @var{name} is the name of the function,
22680 a string. The function will be visible to the user as a convenience
22681 variable of type @code{internal function}, whose name is the same as
22682 the given @var{name}.
22684 The documentation for the new function is taken from the documentation
22685 string for the new class.
22688 @defmethod Function invoke @var{*args}
22689 When a convenience function is evaluated, its arguments are converted
22690 to instances of @code{gdb.Value}, and then the function's
22691 @code{invoke} method is called. Note that @value{GDBN} does not
22692 predetermine the arity of convenience functions. Instead, all
22693 available arguments are passed to @code{invoke}, following the
22694 standard Python calling convention. In particular, a convenience
22695 function can have default values for parameters without ill effect.
22697 The return value of this method is used as its value in the enclosing
22698 expression. If an ordinary Python value is returned, it is converted
22699 to a @code{gdb.Value} following the usual rules.
22702 The following code snippet shows how a trivial convenience function can
22703 be implemented in Python:
22706 class Greet (gdb.Function):
22707 """Return string to greet someone.
22708 Takes a name as argument."""
22710 def __init__ (self):
22711 super (Greet, self).__init__ ("greet")
22713 def invoke (self, name):
22714 return "Hello, %s!" % name.string ()
22719 The last line instantiates the class, and is necessary to trigger the
22720 registration of the function with @value{GDBN}. Depending on how the
22721 Python code is read into @value{GDBN}, you may need to import the
22722 @code{gdb} module explicitly.
22724 @node Progspaces In Python
22725 @subsubsection Program Spaces In Python
22727 @cindex progspaces in python
22728 @tindex gdb.Progspace
22730 A program space, or @dfn{progspace}, represents a symbolic view
22731 of an address space.
22732 It consists of all of the objfiles of the program.
22733 @xref{Objfiles In Python}.
22734 @xref{Inferiors and Programs, program spaces}, for more details
22735 about program spaces.
22737 The following progspace-related functions are available in the
22740 @findex gdb.current_progspace
22741 @defun current_progspace
22742 This function returns the program space of the currently selected inferior.
22743 @xref{Inferiors and Programs}.
22746 @findex gdb.progspaces
22748 Return a sequence of all the progspaces currently known to @value{GDBN}.
22751 Each progspace is represented by an instance of the @code{gdb.Progspace}
22754 @defivar Progspace filename
22755 The file name of the progspace as a string.
22758 @defivar Progspace pretty_printers
22759 The @code{pretty_printers} attribute is a list of functions. It is
22760 used to look up pretty-printers. A @code{Value} is passed to each
22761 function in order; if the function returns @code{None}, then the
22762 search continues. Otherwise, the return value should be an object
22763 which is used to format the value. @xref{Pretty Printing API}, for more
22767 @node Objfiles In Python
22768 @subsubsection Objfiles In Python
22770 @cindex objfiles in python
22771 @tindex gdb.Objfile
22773 @value{GDBN} loads symbols for an inferior from various
22774 symbol-containing files (@pxref{Files}). These include the primary
22775 executable file, any shared libraries used by the inferior, and any
22776 separate debug info files (@pxref{Separate Debug Files}).
22777 @value{GDBN} calls these symbol-containing files @dfn{objfiles}.
22779 The following objfile-related functions are available in the
22782 @findex gdb.current_objfile
22783 @defun current_objfile
22784 When auto-loading a Python script (@pxref{Auto-loading}), @value{GDBN}
22785 sets the ``current objfile'' to the corresponding objfile. This
22786 function returns the current objfile. If there is no current objfile,
22787 this function returns @code{None}.
22790 @findex gdb.objfiles
22792 Return a sequence of all the objfiles current known to @value{GDBN}.
22793 @xref{Objfiles In Python}.
22796 Each objfile is represented by an instance of the @code{gdb.Objfile}
22799 @defivar Objfile filename
22800 The file name of the objfile as a string.
22803 @defivar Objfile pretty_printers
22804 The @code{pretty_printers} attribute is a list of functions. It is
22805 used to look up pretty-printers. A @code{Value} is passed to each
22806 function in order; if the function returns @code{None}, then the
22807 search continues. Otherwise, the return value should be an object
22808 which is used to format the value. @xref{Pretty Printing API}, for more
22812 A @code{gdb.Objfile} object has the following methods:
22814 @defmethod Objfile is_valid
22815 Returns @code{True} if the @code{gdb.Objfile} object is valid,
22816 @code{False} if not. A @code{gdb.Objfile} object can become invalid
22817 if the object file it refers to is not loaded in @value{GDBN} any
22818 longer. All other @code{gdb.Objfile} methods will throw an exception
22819 if it is invalid at the time the method is called.
22822 @node Frames In Python
22823 @subsubsection Accessing inferior stack frames from Python.
22825 @cindex frames in python
22826 When the debugged program stops, @value{GDBN} is able to analyze its call
22827 stack (@pxref{Frames,,Stack frames}). The @code{gdb.Frame} class
22828 represents a frame in the stack. A @code{gdb.Frame} object is only valid
22829 while its corresponding frame exists in the inferior's stack. If you try
22830 to use an invalid frame object, @value{GDBN} will throw a @code{gdb.error}
22831 exception (@pxref{Exception Handling}).
22833 Two @code{gdb.Frame} objects can be compared for equality with the @code{==}
22837 (@value{GDBP}) python print gdb.newest_frame() == gdb.selected_frame ()
22841 The following frame-related functions are available in the @code{gdb} module:
22843 @findex gdb.selected_frame
22844 @defun selected_frame
22845 Return the selected frame object. (@pxref{Selection,,Selecting a Frame}).
22848 @findex gdb.newest_frame
22849 @defun newest_frame
22850 Return the newest frame object for the selected thread.
22853 @defun frame_stop_reason_string reason
22854 Return a string explaining the reason why @value{GDBN} stopped unwinding
22855 frames, as expressed by the given @var{reason} code (an integer, see the
22856 @code{unwind_stop_reason} method further down in this section).
22859 A @code{gdb.Frame} object has the following methods:
22862 @defmethod Frame is_valid
22863 Returns true if the @code{gdb.Frame} object is valid, false if not.
22864 A frame object can become invalid if the frame it refers to doesn't
22865 exist anymore in the inferior. All @code{gdb.Frame} methods will throw
22866 an exception if it is invalid at the time the method is called.
22869 @defmethod Frame name
22870 Returns the function name of the frame, or @code{None} if it can't be
22874 @defmethod Frame type
22875 Returns the type of the frame. The value can be one of:
22877 @item gdb.NORMAL_FRAME
22878 An ordinary stack frame.
22880 @item gdb.DUMMY_FRAME
22881 A fake stack frame that was created by @value{GDBN} when performing an
22882 inferior function call.
22884 @item gdb.INLINE_FRAME
22885 A frame representing an inlined function. The function was inlined
22886 into a @code{gdb.NORMAL_FRAME} that is older than this one.
22888 @item gdb.SIGTRAMP_FRAME
22889 A signal trampoline frame. This is the frame created by the OS when
22890 it calls into a signal handler.
22892 @item gdb.ARCH_FRAME
22893 A fake stack frame representing a cross-architecture call.
22895 @item gdb.SENTINEL_FRAME
22896 This is like @code{gdb.NORMAL_FRAME}, but it is only used for the
22901 @defmethod Frame unwind_stop_reason
22902 Return an integer representing the reason why it's not possible to find
22903 more frames toward the outermost frame. Use
22904 @code{gdb.frame_stop_reason_string} to convert the value returned by this
22905 function to a string.
22908 @defmethod Frame pc
22909 Returns the frame's resume address.
22912 @defmethod Frame block
22913 Return the frame's code block. @xref{Blocks In Python}.
22916 @defmethod Frame function
22917 Return the symbol for the function corresponding to this frame.
22918 @xref{Symbols In Python}.
22921 @defmethod Frame older
22922 Return the frame that called this frame.
22925 @defmethod Frame newer
22926 Return the frame called by this frame.
22929 @defmethod Frame find_sal
22930 Return the frame's symtab and line object.
22931 @xref{Symbol Tables In Python}.
22934 @defmethod Frame read_var variable @r{[}block@r{]}
22935 Return the value of @var{variable} in this frame. If the optional
22936 argument @var{block} is provided, search for the variable from that
22937 block; otherwise start at the frame's current block (which is
22938 determined by the frame's current program counter). @var{variable}
22939 must be a string or a @code{gdb.Symbol} object. @var{block} must be a
22940 @code{gdb.Block} object.
22943 @defmethod Frame select
22944 Set this frame to be the selected frame. @xref{Stack, ,Examining the
22949 @node Blocks In Python
22950 @subsubsection Accessing frame blocks from Python.
22952 @cindex blocks in python
22955 Within each frame, @value{GDBN} maintains information on each block
22956 stored in that frame. These blocks are organized hierarchically, and
22957 are represented individually in Python as a @code{gdb.Block}.
22958 Please see @ref{Frames In Python}, for a more in-depth discussion on
22959 frames. Furthermore, see @ref{Stack, ,Examining the Stack}, for more
22960 detailed technical information on @value{GDBN}'s book-keeping of the
22963 The following block-related functions are available in the @code{gdb}
22966 @findex gdb.block_for_pc
22967 @defun block_for_pc pc
22968 Return the @code{gdb.Block} containing the given @var{pc} value. If the
22969 block cannot be found for the @var{pc} value specified, the function
22970 will return @code{None}.
22973 A @code{gdb.Block} object has the following methods:
22976 @defmethod Block is_valid
22977 Returns @code{True} if the @code{gdb.Block} object is valid,
22978 @code{False} if not. A block object can become invalid if the block it
22979 refers to doesn't exist anymore in the inferior. All other
22980 @code{gdb.Block} methods will throw an exception if it is invalid at
22981 the time the method is called. This method is also made available to
22982 the Python iterator object that @code{gdb.Block} provides in an iteration
22983 context and via the Python @code{iter} built-in function.
22987 A @code{gdb.Block} object has the following attributes:
22990 @defivar Block start
22991 The start address of the block. This attribute is not writable.
22995 The end address of the block. This attribute is not writable.
22998 @defivar Block function
22999 The name of the block represented as a @code{gdb.Symbol}. If the
23000 block is not named, then this attribute holds @code{None}. This
23001 attribute is not writable.
23004 @defivar Block superblock
23005 The block containing this block. If this parent block does not exist,
23006 this attribute holds @code{None}. This attribute is not writable.
23010 @node Symbols In Python
23011 @subsubsection Python representation of Symbols.
23013 @cindex symbols in python
23016 @value{GDBN} represents every variable, function and type as an
23017 entry in a symbol table. @xref{Symbols, ,Examining the Symbol Table}.
23018 Similarly, Python represents these symbols in @value{GDBN} with the
23019 @code{gdb.Symbol} object.
23021 The following symbol-related functions are available in the @code{gdb}
23024 @findex gdb.lookup_symbol
23025 @defun lookup_symbol name @r{[}block@r{]} @r{[}domain@r{]}
23026 This function searches for a symbol by name. The search scope can be
23027 restricted to the parameters defined in the optional domain and block
23030 @var{name} is the name of the symbol. It must be a string. The
23031 optional @var{block} argument restricts the search to symbols visible
23032 in that @var{block}. The @var{block} argument must be a
23033 @code{gdb.Block} object. If omitted, the block for the current frame
23034 is used. The optional @var{domain} argument restricts
23035 the search to the domain type. The @var{domain} argument must be a
23036 domain constant defined in the @code{gdb} module and described later
23039 The result is a tuple of two elements.
23040 The first element is a @code{gdb.Symbol} object or @code{None} if the symbol
23042 If the symbol is found, the second element is @code{True} if the symbol
23043 is a field of a method's object (e.g., @code{this} in C@t{++}),
23044 otherwise it is @code{False}.
23045 If the symbol is not found, the second element is @code{False}.
23048 @findex gdb.lookup_global_symbol
23049 @defun lookup_global_symbol name @r{[}domain@r{]}
23050 This function searches for a global symbol by name.
23051 The search scope can be restricted to by the domain argument.
23053 @var{name} is the name of the symbol. It must be a string.
23054 The optional @var{domain} argument restricts the search to the domain type.
23055 The @var{domain} argument must be a domain constant defined in the @code{gdb}
23056 module and described later in this chapter.
23058 The result is a @code{gdb.Symbol} object or @code{None} if the symbol
23062 A @code{gdb.Symbol} object has the following attributes:
23065 @defivar Symbol symtab
23066 The symbol table in which the symbol appears. This attribute is
23067 represented as a @code{gdb.Symtab} object. @xref{Symbol Tables In
23068 Python}. This attribute is not writable.
23071 @defivar Symbol name
23072 The name of the symbol as a string. This attribute is not writable.
23075 @defivar Symbol linkage_name
23076 The name of the symbol, as used by the linker (i.e., may be mangled).
23077 This attribute is not writable.
23080 @defivar Symbol print_name
23081 The name of the symbol in a form suitable for output. This is either
23082 @code{name} or @code{linkage_name}, depending on whether the user
23083 asked @value{GDBN} to display demangled or mangled names.
23086 @defivar Symbol addr_class
23087 The address class of the symbol. This classifies how to find the value
23088 of a symbol. Each address class is a constant defined in the
23089 @code{gdb} module and described later in this chapter.
23092 @defivar Symbol is_argument
23093 @code{True} if the symbol is an argument of a function.
23096 @defivar Symbol is_constant
23097 @code{True} if the symbol is a constant.
23100 @defivar Symbol is_function
23101 @code{True} if the symbol is a function or a method.
23104 @defivar Symbol is_variable
23105 @code{True} if the symbol is a variable.
23109 A @code{gdb.Symbol} object has the following methods:
23112 @defmethod Symbol is_valid
23113 Returns @code{True} if the @code{gdb.Symbol} object is valid,
23114 @code{False} if not. A @code{gdb.Symbol} object can become invalid if
23115 the symbol it refers to does not exist in @value{GDBN} any longer.
23116 All other @code{gdb.Symbol} methods will throw an exception if it is
23117 invalid at the time the method is called.
23121 The available domain categories in @code{gdb.Symbol} are represented
23122 as constants in the @code{gdb} module:
23125 @findex SYMBOL_UNDEF_DOMAIN
23126 @findex gdb.SYMBOL_UNDEF_DOMAIN
23127 @item SYMBOL_UNDEF_DOMAIN
23128 This is used when a domain has not been discovered or none of the
23129 following domains apply. This usually indicates an error either
23130 in the symbol information or in @value{GDBN}'s handling of symbols.
23131 @findex SYMBOL_VAR_DOMAIN
23132 @findex gdb.SYMBOL_VAR_DOMAIN
23133 @item SYMBOL_VAR_DOMAIN
23134 This domain contains variables, function names, typedef names and enum
23136 @findex SYMBOL_STRUCT_DOMAIN
23137 @findex gdb.SYMBOL_STRUCT_DOMAIN
23138 @item SYMBOL_STRUCT_DOMAIN
23139 This domain holds struct, union and enum type names.
23140 @findex SYMBOL_LABEL_DOMAIN
23141 @findex gdb.SYMBOL_LABEL_DOMAIN
23142 @item SYMBOL_LABEL_DOMAIN
23143 This domain contains names of labels (for gotos).
23144 @findex SYMBOL_VARIABLES_DOMAIN
23145 @findex gdb.SYMBOL_VARIABLES_DOMAIN
23146 @item SYMBOL_VARIABLES_DOMAIN
23147 This domain holds a subset of the @code{SYMBOLS_VAR_DOMAIN}; it
23148 contains everything minus functions and types.
23149 @findex SYMBOL_FUNCTIONS_DOMAIN
23150 @findex gdb.SYMBOL_FUNCTIONS_DOMAIN
23151 @item SYMBOL_FUNCTION_DOMAIN
23152 This domain contains all functions.
23153 @findex SYMBOL_TYPES_DOMAIN
23154 @findex gdb.SYMBOL_TYPES_DOMAIN
23155 @item SYMBOL_TYPES_DOMAIN
23156 This domain contains all types.
23159 The available address class categories in @code{gdb.Symbol} are represented
23160 as constants in the @code{gdb} module:
23163 @findex SYMBOL_LOC_UNDEF
23164 @findex gdb.SYMBOL_LOC_UNDEF
23165 @item SYMBOL_LOC_UNDEF
23166 If this is returned by address class, it indicates an error either in
23167 the symbol information or in @value{GDBN}'s handling of symbols.
23168 @findex SYMBOL_LOC_CONST
23169 @findex gdb.SYMBOL_LOC_CONST
23170 @item SYMBOL_LOC_CONST
23171 Value is constant int.
23172 @findex SYMBOL_LOC_STATIC
23173 @findex gdb.SYMBOL_LOC_STATIC
23174 @item SYMBOL_LOC_STATIC
23175 Value is at a fixed address.
23176 @findex SYMBOL_LOC_REGISTER
23177 @findex gdb.SYMBOL_LOC_REGISTER
23178 @item SYMBOL_LOC_REGISTER
23179 Value is in a register.
23180 @findex SYMBOL_LOC_ARG
23181 @findex gdb.SYMBOL_LOC_ARG
23182 @item SYMBOL_LOC_ARG
23183 Value is an argument. This value is at the offset stored within the
23184 symbol inside the frame's argument list.
23185 @findex SYMBOL_LOC_REF_ARG
23186 @findex gdb.SYMBOL_LOC_REF_ARG
23187 @item SYMBOL_LOC_REF_ARG
23188 Value address is stored in the frame's argument list. Just like
23189 @code{LOC_ARG} except that the value's address is stored at the
23190 offset, not the value itself.
23191 @findex SYMBOL_LOC_REGPARM_ADDR
23192 @findex gdb.SYMBOL_LOC_REGPARM_ADDR
23193 @item SYMBOL_LOC_REGPARM_ADDR
23194 Value is a specified register. Just like @code{LOC_REGISTER} except
23195 the register holds the address of the argument instead of the argument
23197 @findex SYMBOL_LOC_LOCAL
23198 @findex gdb.SYMBOL_LOC_LOCAL
23199 @item SYMBOL_LOC_LOCAL
23200 Value is a local variable.
23201 @findex SYMBOL_LOC_TYPEDEF
23202 @findex gdb.SYMBOL_LOC_TYPEDEF
23203 @item SYMBOL_LOC_TYPEDEF
23204 Value not used. Symbols in the domain @code{SYMBOL_STRUCT_DOMAIN} all
23206 @findex SYMBOL_LOC_BLOCK
23207 @findex gdb.SYMBOL_LOC_BLOCK
23208 @item SYMBOL_LOC_BLOCK
23210 @findex SYMBOL_LOC_CONST_BYTES
23211 @findex gdb.SYMBOL_LOC_CONST_BYTES
23212 @item SYMBOL_LOC_CONST_BYTES
23213 Value is a byte-sequence.
23214 @findex SYMBOL_LOC_UNRESOLVED
23215 @findex gdb.SYMBOL_LOC_UNRESOLVED
23216 @item SYMBOL_LOC_UNRESOLVED
23217 Value is at a fixed address, but the address of the variable has to be
23218 determined from the minimal symbol table whenever the variable is
23220 @findex SYMBOL_LOC_OPTIMIZED_OUT
23221 @findex gdb.SYMBOL_LOC_OPTIMIZED_OUT
23222 @item SYMBOL_LOC_OPTIMIZED_OUT
23223 The value does not actually exist in the program.
23224 @findex SYMBOL_LOC_COMPUTED
23225 @findex gdb.SYMBOL_LOC_COMPUTED
23226 @item SYMBOL_LOC_COMPUTED
23227 The value's address is a computed location.
23230 @node Symbol Tables In Python
23231 @subsubsection Symbol table representation in Python.
23233 @cindex symbol tables in python
23235 @tindex gdb.Symtab_and_line
23237 Access to symbol table data maintained by @value{GDBN} on the inferior
23238 is exposed to Python via two objects: @code{gdb.Symtab_and_line} and
23239 @code{gdb.Symtab}. Symbol table and line data for a frame is returned
23240 from the @code{find_sal} method in @code{gdb.Frame} object.
23241 @xref{Frames In Python}.
23243 For more information on @value{GDBN}'s symbol table management, see
23244 @ref{Symbols, ,Examining the Symbol Table}, for more information.
23246 A @code{gdb.Symtab_and_line} object has the following attributes:
23249 @defivar Symtab_and_line symtab
23250 The symbol table object (@code{gdb.Symtab}) for this frame.
23251 This attribute is not writable.
23254 @defivar Symtab_and_line pc
23255 Indicates the current program counter address. This attribute is not
23259 @defivar Symtab_and_line line
23260 Indicates the current line number for this object. This
23261 attribute is not writable.
23265 A @code{gdb.Symtab_and_line} object has the following methods:
23268 @defmethod Symtab_and_line is_valid
23269 Returns @code{True} if the @code{gdb.Symtab_and_line} object is valid,
23270 @code{False} if not. A @code{gdb.Symtab_and_line} object can become
23271 invalid if the Symbol table and line object it refers to does not
23272 exist in @value{GDBN} any longer. All other
23273 @code{gdb.Symtab_and_line} methods will throw an exception if it is
23274 invalid at the time the method is called.
23278 A @code{gdb.Symtab} object has the following attributes:
23281 @defivar Symtab filename
23282 The symbol table's source filename. This attribute is not writable.
23285 @defivar Symtab objfile
23286 The symbol table's backing object file. @xref{Objfiles In Python}.
23287 This attribute is not writable.
23291 A @code{gdb.Symtab} object has the following methods:
23294 @defmethod Symtab is_valid
23295 Returns @code{True} if the @code{gdb.Symtab} object is valid,
23296 @code{False} if not. A @code{gdb.Symtab} object can become invalid if
23297 the symbol table it refers to does not exist in @value{GDBN} any
23298 longer. All other @code{gdb.Symtab} methods will throw an exception
23299 if it is invalid at the time the method is called.
23302 @defmethod Symtab fullname
23303 Return the symbol table's source absolute file name.
23307 @node Breakpoints In Python
23308 @subsubsection Manipulating breakpoints using Python
23310 @cindex breakpoints in python
23311 @tindex gdb.Breakpoint
23313 Python code can manipulate breakpoints via the @code{gdb.Breakpoint}
23316 @defmethod Breakpoint __init__ spec @r{[}type@r{]} @r{[}wp_class@r{]} @r{[}internal@r{]}
23317 Create a new breakpoint. @var{spec} is a string naming the
23318 location of the breakpoint, or an expression that defines a
23319 watchpoint. The contents can be any location recognized by the
23320 @code{break} command, or in the case of a watchpoint, by the @code{watch}
23321 command. The optional @var{type} denotes the breakpoint to create
23322 from the types defined later in this chapter. This argument can be
23323 either: @code{BP_BREAKPOINT} or @code{BP_WATCHPOINT}. @var{type}
23324 defaults to @code{BP_BREAKPOINT}. The optional @var{internal} argument
23325 allows the breakpoint to become invisible to the user. The breakpoint
23326 will neither be reported when created, nor will it be listed in the
23327 output from @code{info breakpoints} (but will be listed with the
23328 @code{maint info breakpoints} command). The optional @var{wp_class}
23329 argument defines the class of watchpoint to create, if @var{type} is
23330 @code{BP_WATCHPOINT}. If a watchpoint class is not provided, it is
23331 assumed to be a @var{WP_WRITE} class.
23334 @defop Operation {gdb.Breakpoint} stop (self)
23335 The @code{gdb.Breakpoint} class can be sub-classed and, in
23336 particular, you may choose to implement the @code{stop} method.
23337 If this method is defined as a sub-class of @code{gdb.Breakpoint},
23338 it will be called when the inferior reaches any location of a
23339 breakpoint which instantiates that sub-class. If the method returns
23340 @code{True}, the inferior will be stopped at the location of the
23341 breakpoint, otherwise the inferior will continue.
23343 If there are multiple breakpoints at the same location with a
23344 @code{stop} method, each one will be called regardless of the
23345 return status of the previous. This ensures that all @code{stop}
23346 methods have a chance to execute at that location. In this scenario
23347 if one of the methods returns @code{True} but the others return
23348 @code{False}, the inferior will still be stopped.
23350 Example @code{stop} implementation:
23353 class MyBreakpoint (gdb.Breakpoint):
23355 inf_val = gdb.parse_and_eval("foo")
23362 The available watchpoint types represented by constants are defined in the
23367 @findex gdb.WP_READ
23369 Read only watchpoint.
23372 @findex gdb.WP_WRITE
23374 Write only watchpoint.
23377 @findex gdb.WP_ACCESS
23379 Read/Write watchpoint.
23382 @defmethod Breakpoint is_valid
23383 Return @code{True} if this @code{Breakpoint} object is valid,
23384 @code{False} otherwise. A @code{Breakpoint} object can become invalid
23385 if the user deletes the breakpoint. In this case, the object still
23386 exists, but the underlying breakpoint does not. In the cases of
23387 watchpoint scope, the watchpoint remains valid even if execution of the
23388 inferior leaves the scope of that watchpoint.
23391 @defmethod Breakpoint delete
23392 Permanently deletes the @value{GDBN} breakpoint. This also
23393 invalidates the Python @code{Breakpoint} object. Any further access
23394 to this object's attributes or methods will raise an error.
23397 @defivar Breakpoint enabled
23398 This attribute is @code{True} if the breakpoint is enabled, and
23399 @code{False} otherwise. This attribute is writable.
23402 @defivar Breakpoint silent
23403 This attribute is @code{True} if the breakpoint is silent, and
23404 @code{False} otherwise. This attribute is writable.
23406 Note that a breakpoint can also be silent if it has commands and the
23407 first command is @code{silent}. This is not reported by the
23408 @code{silent} attribute.
23411 @defivar Breakpoint thread
23412 If the breakpoint is thread-specific, this attribute holds the thread
23413 id. If the breakpoint is not thread-specific, this attribute is
23414 @code{None}. This attribute is writable.
23417 @defivar Breakpoint task
23418 If the breakpoint is Ada task-specific, this attribute holds the Ada task
23419 id. If the breakpoint is not task-specific (or the underlying
23420 language is not Ada), this attribute is @code{None}. This attribute
23424 @defivar Breakpoint ignore_count
23425 This attribute holds the ignore count for the breakpoint, an integer.
23426 This attribute is writable.
23429 @defivar Breakpoint number
23430 This attribute holds the breakpoint's number --- the identifier used by
23431 the user to manipulate the breakpoint. This attribute is not writable.
23434 @defivar Breakpoint type
23435 This attribute holds the breakpoint's type --- the identifier used to
23436 determine the actual breakpoint type or use-case. This attribute is not
23440 @defivar Breakpoint visible
23441 This attribute tells whether the breakpoint is visible to the user
23442 when set, or when the @samp{info breakpoints} command is run. This
23443 attribute is not writable.
23446 The available types are represented by constants defined in the @code{gdb}
23450 @findex BP_BREAKPOINT
23451 @findex gdb.BP_BREAKPOINT
23452 @item BP_BREAKPOINT
23453 Normal code breakpoint.
23455 @findex BP_WATCHPOINT
23456 @findex gdb.BP_WATCHPOINT
23457 @item BP_WATCHPOINT
23458 Watchpoint breakpoint.
23460 @findex BP_HARDWARE_WATCHPOINT
23461 @findex gdb.BP_HARDWARE_WATCHPOINT
23462 @item BP_HARDWARE_WATCHPOINT
23463 Hardware assisted watchpoint.
23465 @findex BP_READ_WATCHPOINT
23466 @findex gdb.BP_READ_WATCHPOINT
23467 @item BP_READ_WATCHPOINT
23468 Hardware assisted read watchpoint.
23470 @findex BP_ACCESS_WATCHPOINT
23471 @findex gdb.BP_ACCESS_WATCHPOINT
23472 @item BP_ACCESS_WATCHPOINT
23473 Hardware assisted access watchpoint.
23476 @defivar Breakpoint hit_count
23477 This attribute holds the hit count for the breakpoint, an integer.
23478 This attribute is writable, but currently it can only be set to zero.
23481 @defivar Breakpoint location
23482 This attribute holds the location of the breakpoint, as specified by
23483 the user. It is a string. If the breakpoint does not have a location
23484 (that is, it is a watchpoint) the attribute's value is @code{None}. This
23485 attribute is not writable.
23488 @defivar Breakpoint expression
23489 This attribute holds a breakpoint expression, as specified by
23490 the user. It is a string. If the breakpoint does not have an
23491 expression (the breakpoint is not a watchpoint) the attribute's value
23492 is @code{None}. This attribute is not writable.
23495 @defivar Breakpoint condition
23496 This attribute holds the condition of the breakpoint, as specified by
23497 the user. It is a string. If there is no condition, this attribute's
23498 value is @code{None}. This attribute is writable.
23501 @defivar Breakpoint commands
23502 This attribute holds the commands attached to the breakpoint. If
23503 there are commands, this attribute's value is a string holding all the
23504 commands, separated by newlines. If there are no commands, this
23505 attribute is @code{None}. This attribute is not writable.
23508 @node Lazy Strings In Python
23509 @subsubsection Python representation of lazy strings.
23511 @cindex lazy strings in python
23512 @tindex gdb.LazyString
23514 A @dfn{lazy string} is a string whose contents is not retrieved or
23515 encoded until it is needed.
23517 A @code{gdb.LazyString} is represented in @value{GDBN} as an
23518 @code{address} that points to a region of memory, an @code{encoding}
23519 that will be used to encode that region of memory, and a @code{length}
23520 to delimit the region of memory that represents the string. The
23521 difference between a @code{gdb.LazyString} and a string wrapped within
23522 a @code{gdb.Value} is that a @code{gdb.LazyString} will be treated
23523 differently by @value{GDBN} when printing. A @code{gdb.LazyString} is
23524 retrieved and encoded during printing, while a @code{gdb.Value}
23525 wrapping a string is immediately retrieved and encoded on creation.
23527 A @code{gdb.LazyString} object has the following functions:
23529 @defmethod LazyString value
23530 Convert the @code{gdb.LazyString} to a @code{gdb.Value}. This value
23531 will point to the string in memory, but will lose all the delayed
23532 retrieval, encoding and handling that @value{GDBN} applies to a
23533 @code{gdb.LazyString}.
23536 @defivar LazyString address
23537 This attribute holds the address of the string. This attribute is not
23541 @defivar LazyString length
23542 This attribute holds the length of the string in characters. If the
23543 length is -1, then the string will be fetched and encoded up to the
23544 first null of appropriate width. This attribute is not writable.
23547 @defivar LazyString encoding
23548 This attribute holds the encoding that will be applied to the string
23549 when the string is printed by @value{GDBN}. If the encoding is not
23550 set, or contains an empty string, then @value{GDBN} will select the
23551 most appropriate encoding when the string is printed. This attribute
23555 @defivar LazyString type
23556 This attribute holds the type that is represented by the lazy string's
23557 type. For a lazy string this will always be a pointer type. To
23558 resolve this to the lazy string's character type, use the type's
23559 @code{target} method. @xref{Types In Python}. This attribute is not
23564 @subsection Auto-loading
23565 @cindex auto-loading, Python
23567 When a new object file is read (for example, due to the @code{file}
23568 command, or because the inferior has loaded a shared library),
23569 @value{GDBN} will look for Python support scripts in several ways:
23570 @file{@var{objfile}-gdb.py} and @code{.debug_gdb_scripts} section.
23573 * objfile-gdb.py file:: The @file{@var{objfile}-gdb.py} file
23574 * .debug_gdb_scripts section:: The @code{.debug_gdb_scripts} section
23575 * Which flavor to choose?::
23578 The auto-loading feature is useful for supplying application-specific
23579 debugging commands and scripts.
23581 Auto-loading can be enabled or disabled.
23584 @kindex set auto-load-scripts
23585 @item set auto-load-scripts [yes|no]
23586 Enable or disable the auto-loading of Python scripts.
23588 @kindex show auto-load-scripts
23589 @item show auto-load-scripts
23590 Show whether auto-loading of Python scripts is enabled or disabled.
23593 When reading an auto-loaded file, @value{GDBN} sets the
23594 @dfn{current objfile}. This is available via the @code{gdb.current_objfile}
23595 function (@pxref{Objfiles In Python}). This can be useful for
23596 registering objfile-specific pretty-printers.
23598 @node objfile-gdb.py file
23599 @subsubsection The @file{@var{objfile}-gdb.py} file
23600 @cindex @file{@var{objfile}-gdb.py}
23602 When a new object file is read, @value{GDBN} looks for
23603 a file named @file{@var{objfile}-gdb.py},
23604 where @var{objfile} is the object file's real name, formed by ensuring
23605 that the file name is absolute, following all symlinks, and resolving
23606 @code{.} and @code{..} components. If this file exists and is
23607 readable, @value{GDBN} will evaluate it as a Python script.
23609 If this file does not exist, and if the parameter
23610 @code{debug-file-directory} is set (@pxref{Separate Debug Files}),
23611 then @value{GDBN} will look for @var{real-name} in all of the
23612 directories mentioned in the value of @code{debug-file-directory}.
23614 Finally, if this file does not exist, then @value{GDBN} will look for
23615 a file named @file{@var{data-directory}/python/auto-load/@var{real-name}}, where
23616 @var{data-directory} is @value{GDBN}'s data directory (available via
23617 @code{show data-directory}, @pxref{Data Files}), and @var{real-name}
23618 is the object file's real name, as described above.
23620 @value{GDBN} does not track which files it has already auto-loaded this way.
23621 @value{GDBN} will load the associated script every time the corresponding
23622 @var{objfile} is opened.
23623 So your @file{-gdb.py} file should be careful to avoid errors if it
23624 is evaluated more than once.
23626 @node .debug_gdb_scripts section
23627 @subsubsection The @code{.debug_gdb_scripts} section
23628 @cindex @code{.debug_gdb_scripts} section
23630 For systems using file formats like ELF and COFF,
23631 when @value{GDBN} loads a new object file
23632 it will look for a special section named @samp{.debug_gdb_scripts}.
23633 If this section exists, its contents is a list of names of scripts to load.
23635 @value{GDBN} will look for each specified script file first in the
23636 current directory and then along the source search path
23637 (@pxref{Source Path, ,Specifying Source Directories}),
23638 except that @file{$cdir} is not searched, since the compilation
23639 directory is not relevant to scripts.
23641 Entries can be placed in section @code{.debug_gdb_scripts} with,
23642 for example, this GCC macro:
23645 /* Note: The "MS" section flags are to remove duplicates. */
23646 #define DEFINE_GDB_SCRIPT(script_name) \
23648 .pushsection \".debug_gdb_scripts\", \"MS\",@@progbits,1\n\
23650 .asciz \"" script_name "\"\n\
23656 Then one can reference the macro in a header or source file like this:
23659 DEFINE_GDB_SCRIPT ("my-app-scripts.py")
23662 The script name may include directories if desired.
23664 If the macro is put in a header, any application or library
23665 using this header will get a reference to the specified script.
23667 @node Which flavor to choose?
23668 @subsubsection Which flavor to choose?
23670 Given the multiple ways of auto-loading Python scripts, it might not always
23671 be clear which one to choose. This section provides some guidance.
23673 Benefits of the @file{-gdb.py} way:
23677 Can be used with file formats that don't support multiple sections.
23680 Ease of finding scripts for public libraries.
23682 Scripts specified in the @code{.debug_gdb_scripts} section are searched for
23683 in the source search path.
23684 For publicly installed libraries, e.g., @file{libstdc++}, there typically
23685 isn't a source directory in which to find the script.
23688 Doesn't require source code additions.
23691 Benefits of the @code{.debug_gdb_scripts} way:
23695 Works with static linking.
23697 Scripts for libraries done the @file{-gdb.py} way require an objfile to
23698 trigger their loading. When an application is statically linked the only
23699 objfile available is the executable, and it is cumbersome to attach all the
23700 scripts from all the input libraries to the executable's @file{-gdb.py} script.
23703 Works with classes that are entirely inlined.
23705 Some classes can be entirely inlined, and thus there may not be an associated
23706 shared library to attach a @file{-gdb.py} script to.
23709 Scripts needn't be copied out of the source tree.
23711 In some circumstances, apps can be built out of large collections of internal
23712 libraries, and the build infrastructure necessary to install the
23713 @file{-gdb.py} scripts in a place where @value{GDBN} can find them is
23714 cumbersome. It may be easier to specify the scripts in the
23715 @code{.debug_gdb_scripts} section as relative paths, and add a path to the
23716 top of the source tree to the source search path.
23719 @node Python modules
23720 @subsection Python modules
23721 @cindex python modules
23723 @value{GDBN} comes with a module to assist writing Python code.
23726 * gdb.printing:: Building and registering pretty-printers.
23727 * gdb.types:: Utilities for working with types.
23731 @subsubsection gdb.printing
23732 @cindex gdb.printing
23734 This module provides a collection of utilities for working with
23738 @item PrettyPrinter (@var{name}, @var{subprinters}=None)
23739 This class specifies the API that makes @samp{info pretty-printer},
23740 @samp{enable pretty-printer} and @samp{disable pretty-printer} work.
23741 Pretty-printers should generally inherit from this class.
23743 @item SubPrettyPrinter (@var{name})
23744 For printers that handle multiple types, this class specifies the
23745 corresponding API for the subprinters.
23747 @item RegexpCollectionPrettyPrinter (@var{name})
23748 Utility class for handling multiple printers, all recognized via
23749 regular expressions.
23750 @xref{Writing a Pretty-Printer}, for an example.
23752 @item register_pretty_printer (@var{obj}, @var{printer})
23753 Register @var{printer} with the pretty-printer list of @var{obj}.
23757 @subsubsection gdb.types
23760 This module provides a collection of utilities for working with
23761 @code{gdb.Types} objects.
23764 @item get_basic_type (@var{type})
23765 Return @var{type} with const and volatile qualifiers stripped,
23766 and with typedefs and C@t{++} references converted to the underlying type.
23771 typedef const int const_int;
23773 const_int& foo_ref (foo);
23774 int main () @{ return 0; @}
23781 (gdb) python import gdb.types
23782 (gdb) python foo_ref = gdb.parse_and_eval("foo_ref")
23783 (gdb) python print gdb.types.get_basic_type(foo_ref.type)
23787 @item has_field (@var{type}, @var{field})
23788 Return @code{True} if @var{type}, assumed to be a type with fields
23789 (e.g., a structure or union), has field @var{field}.
23791 @item make_enum_dict (@var{enum_type})
23792 Return a Python @code{dictionary} type produced from @var{enum_type}.
23796 @chapter Command Interpreters
23797 @cindex command interpreters
23799 @value{GDBN} supports multiple command interpreters, and some command
23800 infrastructure to allow users or user interface writers to switch
23801 between interpreters or run commands in other interpreters.
23803 @value{GDBN} currently supports two command interpreters, the console
23804 interpreter (sometimes called the command-line interpreter or @sc{cli})
23805 and the machine interface interpreter (or @sc{gdb/mi}). This manual
23806 describes both of these interfaces in great detail.
23808 By default, @value{GDBN} will start with the console interpreter.
23809 However, the user may choose to start @value{GDBN} with another
23810 interpreter by specifying the @option{-i} or @option{--interpreter}
23811 startup options. Defined interpreters include:
23815 @cindex console interpreter
23816 The traditional console or command-line interpreter. This is the most often
23817 used interpreter with @value{GDBN}. With no interpreter specified at runtime,
23818 @value{GDBN} will use this interpreter.
23821 @cindex mi interpreter
23822 The newest @sc{gdb/mi} interface (currently @code{mi2}). Used primarily
23823 by programs wishing to use @value{GDBN} as a backend for a debugger GUI
23824 or an IDE. For more information, see @ref{GDB/MI, ,The @sc{gdb/mi}
23828 @cindex mi2 interpreter
23829 The current @sc{gdb/mi} interface.
23832 @cindex mi1 interpreter
23833 The @sc{gdb/mi} interface included in @value{GDBN} 5.1, 5.2, and 5.3.
23837 @cindex invoke another interpreter
23838 The interpreter being used by @value{GDBN} may not be dynamically
23839 switched at runtime. Although possible, this could lead to a very
23840 precarious situation. Consider an IDE using @sc{gdb/mi}. If a user
23841 enters the command "interpreter-set console" in a console view,
23842 @value{GDBN} would switch to using the console interpreter, rendering
23843 the IDE inoperable!
23845 @kindex interpreter-exec
23846 Although you may only choose a single interpreter at startup, you may execute
23847 commands in any interpreter from the current interpreter using the appropriate
23848 command. If you are running the console interpreter, simply use the
23849 @code{interpreter-exec} command:
23852 interpreter-exec mi "-data-list-register-names"
23855 @sc{gdb/mi} has a similar command, although it is only available in versions of
23856 @value{GDBN} which support @sc{gdb/mi} version 2 (or greater).
23859 @chapter @value{GDBN} Text User Interface
23861 @cindex Text User Interface
23864 * TUI Overview:: TUI overview
23865 * TUI Keys:: TUI key bindings
23866 * TUI Single Key Mode:: TUI single key mode
23867 * TUI Commands:: TUI-specific commands
23868 * TUI Configuration:: TUI configuration variables
23871 The @value{GDBN} Text User Interface (TUI) is a terminal
23872 interface which uses the @code{curses} library to show the source
23873 file, the assembly output, the program registers and @value{GDBN}
23874 commands in separate text windows. The TUI mode is supported only
23875 on platforms where a suitable version of the @code{curses} library
23878 @pindex @value{GDBTUI}
23879 The TUI mode is enabled by default when you invoke @value{GDBN} as
23880 either @samp{@value{GDBTUI}} or @samp{@value{GDBP} -tui}.
23881 You can also switch in and out of TUI mode while @value{GDBN} runs by
23882 using various TUI commands and key bindings, such as @kbd{C-x C-a}.
23883 @xref{TUI Keys, ,TUI Key Bindings}.
23886 @section TUI Overview
23888 In TUI mode, @value{GDBN} can display several text windows:
23892 This window is the @value{GDBN} command window with the @value{GDBN}
23893 prompt and the @value{GDBN} output. The @value{GDBN} input is still
23894 managed using readline.
23897 The source window shows the source file of the program. The current
23898 line and active breakpoints are displayed in this window.
23901 The assembly window shows the disassembly output of the program.
23904 This window shows the processor registers. Registers are highlighted
23905 when their values change.
23908 The source and assembly windows show the current program position
23909 by highlighting the current line and marking it with a @samp{>} marker.
23910 Breakpoints are indicated with two markers. The first marker
23911 indicates the breakpoint type:
23915 Breakpoint which was hit at least once.
23918 Breakpoint which was never hit.
23921 Hardware breakpoint which was hit at least once.
23924 Hardware breakpoint which was never hit.
23927 The second marker indicates whether the breakpoint is enabled or not:
23931 Breakpoint is enabled.
23934 Breakpoint is disabled.
23937 The source, assembly and register windows are updated when the current
23938 thread changes, when the frame changes, or when the program counter
23941 These windows are not all visible at the same time. The command
23942 window is always visible. The others can be arranged in several
23953 source and assembly,
23956 source and registers, or
23959 assembly and registers.
23962 A status line above the command window shows the following information:
23966 Indicates the current @value{GDBN} target.
23967 (@pxref{Targets, ,Specifying a Debugging Target}).
23970 Gives the current process or thread number.
23971 When no process is being debugged, this field is set to @code{No process}.
23974 Gives the current function name for the selected frame.
23975 The name is demangled if demangling is turned on (@pxref{Print Settings}).
23976 When there is no symbol corresponding to the current program counter,
23977 the string @code{??} is displayed.
23980 Indicates the current line number for the selected frame.
23981 When the current line number is not known, the string @code{??} is displayed.
23984 Indicates the current program counter address.
23988 @section TUI Key Bindings
23989 @cindex TUI key bindings
23991 The TUI installs several key bindings in the readline keymaps
23992 @ifset SYSTEM_READLINE
23993 (@pxref{Command Line Editing, , , rluserman, GNU Readline Library}).
23995 @ifclear SYSTEM_READLINE
23996 (@pxref{Command Line Editing}).
23998 The following key bindings are installed for both TUI mode and the
23999 @value{GDBN} standard mode.
24008 Enter or leave the TUI mode. When leaving the TUI mode,
24009 the curses window management stops and @value{GDBN} operates using
24010 its standard mode, writing on the terminal directly. When reentering
24011 the TUI mode, control is given back to the curses windows.
24012 The screen is then refreshed.
24016 Use a TUI layout with only one window. The layout will
24017 either be @samp{source} or @samp{assembly}. When the TUI mode
24018 is not active, it will switch to the TUI mode.
24020 Think of this key binding as the Emacs @kbd{C-x 1} binding.
24024 Use a TUI layout with at least two windows. When the current
24025 layout already has two windows, the next layout with two windows is used.
24026 When a new layout is chosen, one window will always be common to the
24027 previous layout and the new one.
24029 Think of it as the Emacs @kbd{C-x 2} binding.
24033 Change the active window. The TUI associates several key bindings
24034 (like scrolling and arrow keys) with the active window. This command
24035 gives the focus to the next TUI window.
24037 Think of it as the Emacs @kbd{C-x o} binding.
24041 Switch in and out of the TUI SingleKey mode that binds single
24042 keys to @value{GDBN} commands (@pxref{TUI Single Key Mode}).
24045 The following key bindings only work in the TUI mode:
24050 Scroll the active window one page up.
24054 Scroll the active window one page down.
24058 Scroll the active window one line up.
24062 Scroll the active window one line down.
24066 Scroll the active window one column left.
24070 Scroll the active window one column right.
24074 Refresh the screen.
24077 Because the arrow keys scroll the active window in the TUI mode, they
24078 are not available for their normal use by readline unless the command
24079 window has the focus. When another window is active, you must use
24080 other readline key bindings such as @kbd{C-p}, @kbd{C-n}, @kbd{C-b}
24081 and @kbd{C-f} to control the command window.
24083 @node TUI Single Key Mode
24084 @section TUI Single Key Mode
24085 @cindex TUI single key mode
24087 The TUI also provides a @dfn{SingleKey} mode, which binds several
24088 frequently used @value{GDBN} commands to single keys. Type @kbd{C-x s} to
24089 switch into this mode, where the following key bindings are used:
24092 @kindex c @r{(SingleKey TUI key)}
24096 @kindex d @r{(SingleKey TUI key)}
24100 @kindex f @r{(SingleKey TUI key)}
24104 @kindex n @r{(SingleKey TUI key)}
24108 @kindex q @r{(SingleKey TUI key)}
24110 exit the SingleKey mode.
24112 @kindex r @r{(SingleKey TUI key)}
24116 @kindex s @r{(SingleKey TUI key)}
24120 @kindex u @r{(SingleKey TUI key)}
24124 @kindex v @r{(SingleKey TUI key)}
24128 @kindex w @r{(SingleKey TUI key)}
24133 Other keys temporarily switch to the @value{GDBN} command prompt.
24134 The key that was pressed is inserted in the editing buffer so that
24135 it is possible to type most @value{GDBN} commands without interaction
24136 with the TUI SingleKey mode. Once the command is entered the TUI
24137 SingleKey mode is restored. The only way to permanently leave
24138 this mode is by typing @kbd{q} or @kbd{C-x s}.
24142 @section TUI-specific Commands
24143 @cindex TUI commands
24145 The TUI has specific commands to control the text windows.
24146 These commands are always available, even when @value{GDBN} is not in
24147 the TUI mode. When @value{GDBN} is in the standard mode, most
24148 of these commands will automatically switch to the TUI mode.
24150 Note that if @value{GDBN}'s @code{stdout} is not connected to a
24151 terminal, or @value{GDBN} has been started with the machine interface
24152 interpreter (@pxref{GDB/MI, ,The @sc{gdb/mi} Interface}), most of
24153 these commands will fail with an error, because it would not be
24154 possible or desirable to enable curses window management.
24159 List and give the size of all displayed windows.
24163 Display the next layout.
24166 Display the previous layout.
24169 Display the source window only.
24172 Display the assembly window only.
24175 Display the source and assembly window.
24178 Display the register window together with the source or assembly window.
24182 Make the next window active for scrolling.
24185 Make the previous window active for scrolling.
24188 Make the source window active for scrolling.
24191 Make the assembly window active for scrolling.
24194 Make the register window active for scrolling.
24197 Make the command window active for scrolling.
24201 Refresh the screen. This is similar to typing @kbd{C-L}.
24203 @item tui reg float
24205 Show the floating point registers in the register window.
24207 @item tui reg general
24208 Show the general registers in the register window.
24211 Show the next register group. The list of register groups as well as
24212 their order is target specific. The predefined register groups are the
24213 following: @code{general}, @code{float}, @code{system}, @code{vector},
24214 @code{all}, @code{save}, @code{restore}.
24216 @item tui reg system
24217 Show the system registers in the register window.
24221 Update the source window and the current execution point.
24223 @item winheight @var{name} +@var{count}
24224 @itemx winheight @var{name} -@var{count}
24226 Change the height of the window @var{name} by @var{count}
24227 lines. Positive counts increase the height, while negative counts
24230 @item tabset @var{nchars}
24232 Set the width of tab stops to be @var{nchars} characters.
24235 @node TUI Configuration
24236 @section TUI Configuration Variables
24237 @cindex TUI configuration variables
24239 Several configuration variables control the appearance of TUI windows.
24242 @item set tui border-kind @var{kind}
24243 @kindex set tui border-kind
24244 Select the border appearance for the source, assembly and register windows.
24245 The possible values are the following:
24248 Use a space character to draw the border.
24251 Use @sc{ascii} characters @samp{+}, @samp{-} and @samp{|} to draw the border.
24254 Use the Alternate Character Set to draw the border. The border is
24255 drawn using character line graphics if the terminal supports them.
24258 @item set tui border-mode @var{mode}
24259 @kindex set tui border-mode
24260 @itemx set tui active-border-mode @var{mode}
24261 @kindex set tui active-border-mode
24262 Select the display attributes for the borders of the inactive windows
24263 or the active window. The @var{mode} can be one of the following:
24266 Use normal attributes to display the border.
24272 Use reverse video mode.
24275 Use half bright mode.
24277 @item half-standout
24278 Use half bright and standout mode.
24281 Use extra bright or bold mode.
24283 @item bold-standout
24284 Use extra bright or bold and standout mode.
24289 @chapter Using @value{GDBN} under @sc{gnu} Emacs
24292 @cindex @sc{gnu} Emacs
24293 A special interface allows you to use @sc{gnu} Emacs to view (and
24294 edit) the source files for the program you are debugging with
24297 To use this interface, use the command @kbd{M-x gdb} in Emacs. Give the
24298 executable file you want to debug as an argument. This command starts
24299 @value{GDBN} as a subprocess of Emacs, with input and output through a newly
24300 created Emacs buffer.
24301 @c (Do not use the @code{-tui} option to run @value{GDBN} from Emacs.)
24303 Running @value{GDBN} under Emacs can be just like running @value{GDBN} normally except for two
24308 All ``terminal'' input and output goes through an Emacs buffer, called
24311 This applies both to @value{GDBN} commands and their output, and to the input
24312 and output done by the program you are debugging.
24314 This is useful because it means that you can copy the text of previous
24315 commands and input them again; you can even use parts of the output
24318 All the facilities of Emacs' Shell mode are available for interacting
24319 with your program. In particular, you can send signals the usual
24320 way---for example, @kbd{C-c C-c} for an interrupt, @kbd{C-c C-z} for a
24324 @value{GDBN} displays source code through Emacs.
24326 Each time @value{GDBN} displays a stack frame, Emacs automatically finds the
24327 source file for that frame and puts an arrow (@samp{=>}) at the
24328 left margin of the current line. Emacs uses a separate buffer for
24329 source display, and splits the screen to show both your @value{GDBN} session
24332 Explicit @value{GDBN} @code{list} or search commands still produce output as
24333 usual, but you probably have no reason to use them from Emacs.
24336 We call this @dfn{text command mode}. Emacs 22.1, and later, also uses
24337 a graphical mode, enabled by default, which provides further buffers
24338 that can control the execution and describe the state of your program.
24339 @xref{GDB Graphical Interface,,, Emacs, The @sc{gnu} Emacs Manual}.
24341 If you specify an absolute file name when prompted for the @kbd{M-x
24342 gdb} argument, then Emacs sets your current working directory to where
24343 your program resides. If you only specify the file name, then Emacs
24344 sets your current working directory to the directory associated
24345 with the previous buffer. In this case, @value{GDBN} may find your
24346 program by searching your environment's @code{PATH} variable, but on
24347 some operating systems it might not find the source. So, although the
24348 @value{GDBN} input and output session proceeds normally, the auxiliary
24349 buffer does not display the current source and line of execution.
24351 The initial working directory of @value{GDBN} is printed on the top
24352 line of the GUD buffer and this serves as a default for the commands
24353 that specify files for @value{GDBN} to operate on. @xref{Files,
24354 ,Commands to Specify Files}.
24356 By default, @kbd{M-x gdb} calls the program called @file{gdb}. If you
24357 need to call @value{GDBN} by a different name (for example, if you
24358 keep several configurations around, with different names) you can
24359 customize the Emacs variable @code{gud-gdb-command-name} to run the
24362 In the GUD buffer, you can use these special Emacs commands in
24363 addition to the standard Shell mode commands:
24367 Describe the features of Emacs' GUD Mode.
24370 Execute to another source line, like the @value{GDBN} @code{step} command; also
24371 update the display window to show the current file and location.
24374 Execute to next source line in this function, skipping all function
24375 calls, like the @value{GDBN} @code{next} command. Then update the display window
24376 to show the current file and location.
24379 Execute one instruction, like the @value{GDBN} @code{stepi} command; update
24380 display window accordingly.
24383 Execute until exit from the selected stack frame, like the @value{GDBN}
24384 @code{finish} command.
24387 Continue execution of your program, like the @value{GDBN} @code{continue}
24391 Go up the number of frames indicated by the numeric argument
24392 (@pxref{Arguments, , Numeric Arguments, Emacs, The @sc{gnu} Emacs Manual}),
24393 like the @value{GDBN} @code{up} command.
24396 Go down the number of frames indicated by the numeric argument, like the
24397 @value{GDBN} @code{down} command.
24400 In any source file, the Emacs command @kbd{C-x @key{SPC}} (@code{gud-break})
24401 tells @value{GDBN} to set a breakpoint on the source line point is on.
24403 In text command mode, if you type @kbd{M-x speedbar}, Emacs displays a
24404 separate frame which shows a backtrace when the GUD buffer is current.
24405 Move point to any frame in the stack and type @key{RET} to make it
24406 become the current frame and display the associated source in the
24407 source buffer. Alternatively, click @kbd{Mouse-2} to make the
24408 selected frame become the current one. In graphical mode, the
24409 speedbar displays watch expressions.
24411 If you accidentally delete the source-display buffer, an easy way to get
24412 it back is to type the command @code{f} in the @value{GDBN} buffer, to
24413 request a frame display; when you run under Emacs, this recreates
24414 the source buffer if necessary to show you the context of the current
24417 The source files displayed in Emacs are in ordinary Emacs buffers
24418 which are visiting the source files in the usual way. You can edit
24419 the files with these buffers if you wish; but keep in mind that @value{GDBN}
24420 communicates with Emacs in terms of line numbers. If you add or
24421 delete lines from the text, the line numbers that @value{GDBN} knows cease
24422 to correspond properly with the code.
24424 A more detailed description of Emacs' interaction with @value{GDBN} is
24425 given in the Emacs manual (@pxref{Debuggers,,, Emacs, The @sc{gnu}
24428 @c The following dropped because Epoch is nonstandard. Reactivate
24429 @c if/when v19 does something similar. ---doc@cygnus.com 19dec1990
24431 @kindex Emacs Epoch environment
24435 Version 18 of @sc{gnu} Emacs has a built-in window system
24436 called the @code{epoch}
24437 environment. Users of this environment can use a new command,
24438 @code{inspect} which performs identically to @code{print} except that
24439 each value is printed in its own window.
24444 @chapter The @sc{gdb/mi} Interface
24446 @unnumberedsec Function and Purpose
24448 @cindex @sc{gdb/mi}, its purpose
24449 @sc{gdb/mi} is a line based machine oriented text interface to
24450 @value{GDBN} and is activated by specifying using the
24451 @option{--interpreter} command line option (@pxref{Mode Options}). It
24452 is specifically intended to support the development of systems which
24453 use the debugger as just one small component of a larger system.
24455 This chapter is a specification of the @sc{gdb/mi} interface. It is written
24456 in the form of a reference manual.
24458 Note that @sc{gdb/mi} is still under construction, so some of the
24459 features described below are incomplete and subject to change
24460 (@pxref{GDB/MI Development and Front Ends, , @sc{gdb/mi} Development and Front Ends}).
24462 @unnumberedsec Notation and Terminology
24464 @cindex notational conventions, for @sc{gdb/mi}
24465 This chapter uses the following notation:
24469 @code{|} separates two alternatives.
24472 @code{[ @var{something} ]} indicates that @var{something} is optional:
24473 it may or may not be given.
24476 @code{( @var{group} )*} means that @var{group} inside the parentheses
24477 may repeat zero or more times.
24480 @code{( @var{group} )+} means that @var{group} inside the parentheses
24481 may repeat one or more times.
24484 @code{"@var{string}"} means a literal @var{string}.
24488 @heading Dependencies
24492 * GDB/MI General Design::
24493 * GDB/MI Command Syntax::
24494 * GDB/MI Compatibility with CLI::
24495 * GDB/MI Development and Front Ends::
24496 * GDB/MI Output Records::
24497 * GDB/MI Simple Examples::
24498 * GDB/MI Command Description Format::
24499 * GDB/MI Breakpoint Commands::
24500 * GDB/MI Program Context::
24501 * GDB/MI Thread Commands::
24502 * GDB/MI Program Execution::
24503 * GDB/MI Stack Manipulation::
24504 * GDB/MI Variable Objects::
24505 * GDB/MI Data Manipulation::
24506 * GDB/MI Tracepoint Commands::
24507 * GDB/MI Symbol Query::
24508 * GDB/MI File Commands::
24510 * GDB/MI Kod Commands::
24511 * GDB/MI Memory Overlay Commands::
24512 * GDB/MI Signal Handling Commands::
24514 * GDB/MI Target Manipulation::
24515 * GDB/MI File Transfer Commands::
24516 * GDB/MI Miscellaneous Commands::
24519 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
24520 @node GDB/MI General Design
24521 @section @sc{gdb/mi} General Design
24522 @cindex GDB/MI General Design
24524 Interaction of a @sc{GDB/MI} frontend with @value{GDBN} involves three
24525 parts---commands sent to @value{GDBN}, responses to those commands
24526 and notifications. Each command results in exactly one response,
24527 indicating either successful completion of the command, or an error.
24528 For the commands that do not resume the target, the response contains the
24529 requested information. For the commands that resume the target, the
24530 response only indicates whether the target was successfully resumed.
24531 Notifications is the mechanism for reporting changes in the state of the
24532 target, or in @value{GDBN} state, that cannot conveniently be associated with
24533 a command and reported as part of that command response.
24535 The important examples of notifications are:
24539 Exec notifications. These are used to report changes in
24540 target state---when a target is resumed, or stopped. It would not
24541 be feasible to include this information in response of resuming
24542 commands, because one resume commands can result in multiple events in
24543 different threads. Also, quite some time may pass before any event
24544 happens in the target, while a frontend needs to know whether the resuming
24545 command itself was successfully executed.
24548 Console output, and status notifications. Console output
24549 notifications are used to report output of CLI commands, as well as
24550 diagnostics for other commands. Status notifications are used to
24551 report the progress of a long-running operation. Naturally, including
24552 this information in command response would mean no output is produced
24553 until the command is finished, which is undesirable.
24556 General notifications. Commands may have various side effects on
24557 the @value{GDBN} or target state beyond their official purpose. For example,
24558 a command may change the selected thread. Although such changes can
24559 be included in command response, using notification allows for more
24560 orthogonal frontend design.
24564 There's no guarantee that whenever an MI command reports an error,
24565 @value{GDBN} or the target are in any specific state, and especially,
24566 the state is not reverted to the state before the MI command was
24567 processed. Therefore, whenever an MI command results in an error,
24568 we recommend that the frontend refreshes all the information shown in
24569 the user interface.
24573 * Context management::
24574 * Asynchronous and non-stop modes::
24578 @node Context management
24579 @subsection Context management
24581 In most cases when @value{GDBN} accesses the target, this access is
24582 done in context of a specific thread and frame (@pxref{Frames}).
24583 Often, even when accessing global data, the target requires that a thread
24584 be specified. The CLI interface maintains the selected thread and frame,
24585 and supplies them to target on each command. This is convenient,
24586 because a command line user would not want to specify that information
24587 explicitly on each command, and because user interacts with
24588 @value{GDBN} via a single terminal, so no confusion is possible as
24589 to what thread and frame are the current ones.
24591 In the case of MI, the concept of selected thread and frame is less
24592 useful. First, a frontend can easily remember this information
24593 itself. Second, a graphical frontend can have more than one window,
24594 each one used for debugging a different thread, and the frontend might
24595 want to access additional threads for internal purposes. This
24596 increases the risk that by relying on implicitly selected thread, the
24597 frontend may be operating on a wrong one. Therefore, each MI command
24598 should explicitly specify which thread and frame to operate on. To
24599 make it possible, each MI command accepts the @samp{--thread} and
24600 @samp{--frame} options, the value to each is @value{GDBN} identifier
24601 for thread and frame to operate on.
24603 Usually, each top-level window in a frontend allows the user to select
24604 a thread and a frame, and remembers the user selection for further
24605 operations. However, in some cases @value{GDBN} may suggest that the
24606 current thread be changed. For example, when stopping on a breakpoint
24607 it is reasonable to switch to the thread where breakpoint is hit. For
24608 another example, if the user issues the CLI @samp{thread} command via
24609 the frontend, it is desirable to change the frontend's selected thread to the
24610 one specified by user. @value{GDBN} communicates the suggestion to
24611 change current thread using the @samp{=thread-selected} notification.
24612 No such notification is available for the selected frame at the moment.
24614 Note that historically, MI shares the selected thread with CLI, so
24615 frontends used the @code{-thread-select} to execute commands in the
24616 right context. However, getting this to work right is cumbersome. The
24617 simplest way is for frontend to emit @code{-thread-select} command
24618 before every command. This doubles the number of commands that need
24619 to be sent. The alternative approach is to suppress @code{-thread-select}
24620 if the selected thread in @value{GDBN} is supposed to be identical to the
24621 thread the frontend wants to operate on. However, getting this
24622 optimization right can be tricky. In particular, if the frontend
24623 sends several commands to @value{GDBN}, and one of the commands changes the
24624 selected thread, then the behaviour of subsequent commands will
24625 change. So, a frontend should either wait for response from such
24626 problematic commands, or explicitly add @code{-thread-select} for
24627 all subsequent commands. No frontend is known to do this exactly
24628 right, so it is suggested to just always pass the @samp{--thread} and
24629 @samp{--frame} options.
24631 @node Asynchronous and non-stop modes
24632 @subsection Asynchronous command execution and non-stop mode
24634 On some targets, @value{GDBN} is capable of processing MI commands
24635 even while the target is running. This is called @dfn{asynchronous
24636 command execution} (@pxref{Background Execution}). The frontend may
24637 specify a preferrence for asynchronous execution using the
24638 @code{-gdb-set target-async 1} command, which should be emitted before
24639 either running the executable or attaching to the target. After the
24640 frontend has started the executable or attached to the target, it can
24641 find if asynchronous execution is enabled using the
24642 @code{-list-target-features} command.
24644 Even if @value{GDBN} can accept a command while target is running,
24645 many commands that access the target do not work when the target is
24646 running. Therefore, asynchronous command execution is most useful
24647 when combined with non-stop mode (@pxref{Non-Stop Mode}). Then,
24648 it is possible to examine the state of one thread, while other threads
24651 When a given thread is running, MI commands that try to access the
24652 target in the context of that thread may not work, or may work only on
24653 some targets. In particular, commands that try to operate on thread's
24654 stack will not work, on any target. Commands that read memory, or
24655 modify breakpoints, may work or not work, depending on the target. Note
24656 that even commands that operate on global state, such as @code{print},
24657 @code{set}, and breakpoint commands, still access the target in the
24658 context of a specific thread, so frontend should try to find a
24659 stopped thread and perform the operation on that thread (using the
24660 @samp{--thread} option).
24662 Which commands will work in the context of a running thread is
24663 highly target dependent. However, the two commands
24664 @code{-exec-interrupt}, to stop a thread, and @code{-thread-info},
24665 to find the state of a thread, will always work.
24667 @node Thread groups
24668 @subsection Thread groups
24669 @value{GDBN} may be used to debug several processes at the same time.
24670 On some platfroms, @value{GDBN} may support debugging of several
24671 hardware systems, each one having several cores with several different
24672 processes running on each core. This section describes the MI
24673 mechanism to support such debugging scenarios.
24675 The key observation is that regardless of the structure of the
24676 target, MI can have a global list of threads, because most commands that
24677 accept the @samp{--thread} option do not need to know what process that
24678 thread belongs to. Therefore, it is not necessary to introduce
24679 neither additional @samp{--process} option, nor an notion of the
24680 current process in the MI interface. The only strictly new feature
24681 that is required is the ability to find how the threads are grouped
24684 To allow the user to discover such grouping, and to support arbitrary
24685 hierarchy of machines/cores/processes, MI introduces the concept of a
24686 @dfn{thread group}. Thread group is a collection of threads and other
24687 thread groups. A thread group always has a string identifier, a type,
24688 and may have additional attributes specific to the type. A new
24689 command, @code{-list-thread-groups}, returns the list of top-level
24690 thread groups, which correspond to processes that @value{GDBN} is
24691 debugging at the moment. By passing an identifier of a thread group
24692 to the @code{-list-thread-groups} command, it is possible to obtain
24693 the members of specific thread group.
24695 To allow the user to easily discover processes, and other objects, he
24696 wishes to debug, a concept of @dfn{available thread group} is
24697 introduced. Available thread group is an thread group that
24698 @value{GDBN} is not debugging, but that can be attached to, using the
24699 @code{-target-attach} command. The list of available top-level thread
24700 groups can be obtained using @samp{-list-thread-groups --available}.
24701 In general, the content of a thread group may be only retrieved only
24702 after attaching to that thread group.
24704 Thread groups are related to inferiors (@pxref{Inferiors and
24705 Programs}). Each inferior corresponds to a thread group of a special
24706 type @samp{process}, and some additional operations are permitted on
24707 such thread groups.
24709 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
24710 @node GDB/MI Command Syntax
24711 @section @sc{gdb/mi} Command Syntax
24714 * GDB/MI Input Syntax::
24715 * GDB/MI Output Syntax::
24718 @node GDB/MI Input Syntax
24719 @subsection @sc{gdb/mi} Input Syntax
24721 @cindex input syntax for @sc{gdb/mi}
24722 @cindex @sc{gdb/mi}, input syntax
24724 @item @var{command} @expansion{}
24725 @code{@var{cli-command} | @var{mi-command}}
24727 @item @var{cli-command} @expansion{}
24728 @code{[ @var{token} ] @var{cli-command} @var{nl}}, where
24729 @var{cli-command} is any existing @value{GDBN} CLI command.
24731 @item @var{mi-command} @expansion{}
24732 @code{[ @var{token} ] "-" @var{operation} ( " " @var{option} )*
24733 @code{[} " --" @code{]} ( " " @var{parameter} )* @var{nl}}
24735 @item @var{token} @expansion{}
24736 "any sequence of digits"
24738 @item @var{option} @expansion{}
24739 @code{"-" @var{parameter} [ " " @var{parameter} ]}
24741 @item @var{parameter} @expansion{}
24742 @code{@var{non-blank-sequence} | @var{c-string}}
24744 @item @var{operation} @expansion{}
24745 @emph{any of the operations described in this chapter}
24747 @item @var{non-blank-sequence} @expansion{}
24748 @emph{anything, provided it doesn't contain special characters such as
24749 "-", @var{nl}, """ and of course " "}
24751 @item @var{c-string} @expansion{}
24752 @code{""" @var{seven-bit-iso-c-string-content} """}
24754 @item @var{nl} @expansion{}
24763 The CLI commands are still handled by the @sc{mi} interpreter; their
24764 output is described below.
24767 The @code{@var{token}}, when present, is passed back when the command
24771 Some @sc{mi} commands accept optional arguments as part of the parameter
24772 list. Each option is identified by a leading @samp{-} (dash) and may be
24773 followed by an optional argument parameter. Options occur first in the
24774 parameter list and can be delimited from normal parameters using
24775 @samp{--} (this is useful when some parameters begin with a dash).
24782 We want easy access to the existing CLI syntax (for debugging).
24785 We want it to be easy to spot a @sc{mi} operation.
24788 @node GDB/MI Output Syntax
24789 @subsection @sc{gdb/mi} Output Syntax
24791 @cindex output syntax of @sc{gdb/mi}
24792 @cindex @sc{gdb/mi}, output syntax
24793 The output from @sc{gdb/mi} consists of zero or more out-of-band records
24794 followed, optionally, by a single result record. This result record
24795 is for the most recent command. The sequence of output records is
24796 terminated by @samp{(gdb)}.
24798 If an input command was prefixed with a @code{@var{token}} then the
24799 corresponding output for that command will also be prefixed by that same
24803 @item @var{output} @expansion{}
24804 @code{( @var{out-of-band-record} )* [ @var{result-record} ] "(gdb)" @var{nl}}
24806 @item @var{result-record} @expansion{}
24807 @code{ [ @var{token} ] "^" @var{result-class} ( "," @var{result} )* @var{nl}}
24809 @item @var{out-of-band-record} @expansion{}
24810 @code{@var{async-record} | @var{stream-record}}
24812 @item @var{async-record} @expansion{}
24813 @code{@var{exec-async-output} | @var{status-async-output} | @var{notify-async-output}}
24815 @item @var{exec-async-output} @expansion{}
24816 @code{[ @var{token} ] "*" @var{async-output}}
24818 @item @var{status-async-output} @expansion{}
24819 @code{[ @var{token} ] "+" @var{async-output}}
24821 @item @var{notify-async-output} @expansion{}
24822 @code{[ @var{token} ] "=" @var{async-output}}
24824 @item @var{async-output} @expansion{}
24825 @code{@var{async-class} ( "," @var{result} )* @var{nl}}
24827 @item @var{result-class} @expansion{}
24828 @code{"done" | "running" | "connected" | "error" | "exit"}
24830 @item @var{async-class} @expansion{}
24831 @code{"stopped" | @var{others}} (where @var{others} will be added
24832 depending on the needs---this is still in development).
24834 @item @var{result} @expansion{}
24835 @code{ @var{variable} "=" @var{value}}
24837 @item @var{variable} @expansion{}
24838 @code{ @var{string} }
24840 @item @var{value} @expansion{}
24841 @code{ @var{const} | @var{tuple} | @var{list} }
24843 @item @var{const} @expansion{}
24844 @code{@var{c-string}}
24846 @item @var{tuple} @expansion{}
24847 @code{ "@{@}" | "@{" @var{result} ( "," @var{result} )* "@}" }
24849 @item @var{list} @expansion{}
24850 @code{ "[]" | "[" @var{value} ( "," @var{value} )* "]" | "["
24851 @var{result} ( "," @var{result} )* "]" }
24853 @item @var{stream-record} @expansion{}
24854 @code{@var{console-stream-output} | @var{target-stream-output} | @var{log-stream-output}}
24856 @item @var{console-stream-output} @expansion{}
24857 @code{"~" @var{c-string}}
24859 @item @var{target-stream-output} @expansion{}
24860 @code{"@@" @var{c-string}}
24862 @item @var{log-stream-output} @expansion{}
24863 @code{"&" @var{c-string}}
24865 @item @var{nl} @expansion{}
24868 @item @var{token} @expansion{}
24869 @emph{any sequence of digits}.
24877 All output sequences end in a single line containing a period.
24880 The @code{@var{token}} is from the corresponding request. Note that
24881 for all async output, while the token is allowed by the grammar and
24882 may be output by future versions of @value{GDBN} for select async
24883 output messages, it is generally omitted. Frontends should treat
24884 all async output as reporting general changes in the state of the
24885 target and there should be no need to associate async output to any
24889 @cindex status output in @sc{gdb/mi}
24890 @var{status-async-output} contains on-going status information about the
24891 progress of a slow operation. It can be discarded. All status output is
24892 prefixed by @samp{+}.
24895 @cindex async output in @sc{gdb/mi}
24896 @var{exec-async-output} contains asynchronous state change on the target
24897 (stopped, started, disappeared). All async output is prefixed by
24901 @cindex notify output in @sc{gdb/mi}
24902 @var{notify-async-output} contains supplementary information that the
24903 client should handle (e.g., a new breakpoint information). All notify
24904 output is prefixed by @samp{=}.
24907 @cindex console output in @sc{gdb/mi}
24908 @var{console-stream-output} is output that should be displayed as is in the
24909 console. It is the textual response to a CLI command. All the console
24910 output is prefixed by @samp{~}.
24913 @cindex target output in @sc{gdb/mi}
24914 @var{target-stream-output} is the output produced by the target program.
24915 All the target output is prefixed by @samp{@@}.
24918 @cindex log output in @sc{gdb/mi}
24919 @var{log-stream-output} is output text coming from @value{GDBN}'s internals, for
24920 instance messages that should be displayed as part of an error log. All
24921 the log output is prefixed by @samp{&}.
24924 @cindex list output in @sc{gdb/mi}
24925 New @sc{gdb/mi} commands should only output @var{lists} containing
24931 @xref{GDB/MI Stream Records, , @sc{gdb/mi} Stream Records}, for more
24932 details about the various output records.
24934 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
24935 @node GDB/MI Compatibility with CLI
24936 @section @sc{gdb/mi} Compatibility with CLI
24938 @cindex compatibility, @sc{gdb/mi} and CLI
24939 @cindex @sc{gdb/mi}, compatibility with CLI
24941 For the developers convenience CLI commands can be entered directly,
24942 but there may be some unexpected behaviour. For example, commands
24943 that query the user will behave as if the user replied yes, breakpoint
24944 command lists are not executed and some CLI commands, such as
24945 @code{if}, @code{when} and @code{define}, prompt for further input with
24946 @samp{>}, which is not valid MI output.
24948 This feature may be removed at some stage in the future and it is
24949 recommended that front ends use the @code{-interpreter-exec} command
24950 (@pxref{-interpreter-exec}).
24952 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
24953 @node GDB/MI Development and Front Ends
24954 @section @sc{gdb/mi} Development and Front Ends
24955 @cindex @sc{gdb/mi} development
24957 The application which takes the MI output and presents the state of the
24958 program being debugged to the user is called a @dfn{front end}.
24960 Although @sc{gdb/mi} is still incomplete, it is currently being used
24961 by a variety of front ends to @value{GDBN}. This makes it difficult
24962 to introduce new functionality without breaking existing usage. This
24963 section tries to minimize the problems by describing how the protocol
24966 Some changes in MI need not break a carefully designed front end, and
24967 for these the MI version will remain unchanged. The following is a
24968 list of changes that may occur within one level, so front ends should
24969 parse MI output in a way that can handle them:
24973 New MI commands may be added.
24976 New fields may be added to the output of any MI command.
24979 The range of values for fields with specified values, e.g.,
24980 @code{in_scope} (@pxref{-var-update}) may be extended.
24982 @c The format of field's content e.g type prefix, may change so parse it
24983 @c at your own risk. Yes, in general?
24985 @c The order of fields may change? Shouldn't really matter but it might
24986 @c resolve inconsistencies.
24989 If the changes are likely to break front ends, the MI version level
24990 will be increased by one. This will allow the front end to parse the
24991 output according to the MI version. Apart from mi0, new versions of
24992 @value{GDBN} will not support old versions of MI and it will be the
24993 responsibility of the front end to work with the new one.
24995 @c Starting with mi3, add a new command -mi-version that prints the MI
24998 The best way to avoid unexpected changes in MI that might break your front
24999 end is to make your project known to @value{GDBN} developers and
25000 follow development on @email{gdb@@sourceware.org} and
25001 @email{gdb-patches@@sourceware.org}.
25002 @cindex mailing lists
25004 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
25005 @node GDB/MI Output Records
25006 @section @sc{gdb/mi} Output Records
25009 * GDB/MI Result Records::
25010 * GDB/MI Stream Records::
25011 * GDB/MI Async Records::
25012 * GDB/MI Frame Information::
25013 * GDB/MI Thread Information::
25014 * GDB/MI Ada Exception Information::
25017 @node GDB/MI Result Records
25018 @subsection @sc{gdb/mi} Result Records
25020 @cindex result records in @sc{gdb/mi}
25021 @cindex @sc{gdb/mi}, result records
25022 In addition to a number of out-of-band notifications, the response to a
25023 @sc{gdb/mi} command includes one of the following result indications:
25027 @item "^done" [ "," @var{results} ]
25028 The synchronous operation was successful, @code{@var{results}} are the return
25033 This result record is equivalent to @samp{^done}. Historically, it
25034 was output instead of @samp{^done} if the command has resumed the
25035 target. This behaviour is maintained for backward compatibility, but
25036 all frontends should treat @samp{^done} and @samp{^running}
25037 identically and rely on the @samp{*running} output record to determine
25038 which threads are resumed.
25042 @value{GDBN} has connected to a remote target.
25044 @item "^error" "," @var{c-string}
25046 The operation failed. The @code{@var{c-string}} contains the corresponding
25051 @value{GDBN} has terminated.
25055 @node GDB/MI Stream Records
25056 @subsection @sc{gdb/mi} Stream Records
25058 @cindex @sc{gdb/mi}, stream records
25059 @cindex stream records in @sc{gdb/mi}
25060 @value{GDBN} internally maintains a number of output streams: the console, the
25061 target, and the log. The output intended for each of these streams is
25062 funneled through the @sc{gdb/mi} interface using @dfn{stream records}.
25064 Each stream record begins with a unique @dfn{prefix character} which
25065 identifies its stream (@pxref{GDB/MI Output Syntax, , @sc{gdb/mi} Output
25066 Syntax}). In addition to the prefix, each stream record contains a
25067 @code{@var{string-output}}. This is either raw text (with an implicit new
25068 line) or a quoted C string (which does not contain an implicit newline).
25071 @item "~" @var{string-output}
25072 The console output stream contains text that should be displayed in the
25073 CLI console window. It contains the textual responses to CLI commands.
25075 @item "@@" @var{string-output}
25076 The target output stream contains any textual output from the running
25077 target. This is only present when GDB's event loop is truly
25078 asynchronous, which is currently only the case for remote targets.
25080 @item "&" @var{string-output}
25081 The log stream contains debugging messages being produced by @value{GDBN}'s
25085 @node GDB/MI Async Records
25086 @subsection @sc{gdb/mi} Async Records
25088 @cindex async records in @sc{gdb/mi}
25089 @cindex @sc{gdb/mi}, async records
25090 @dfn{Async} records are used to notify the @sc{gdb/mi} client of
25091 additional changes that have occurred. Those changes can either be a
25092 consequence of @sc{gdb/mi} commands (e.g., a breakpoint modified) or a result of
25093 target activity (e.g., target stopped).
25095 The following is the list of possible async records:
25099 @item *running,thread-id="@var{thread}"
25100 The target is now running. The @var{thread} field tells which
25101 specific thread is now running, and can be @samp{all} if all threads
25102 are running. The frontend should assume that no interaction with a
25103 running thread is possible after this notification is produced.
25104 The frontend should not assume that this notification is output
25105 only once for any command. @value{GDBN} may emit this notification
25106 several times, either for different threads, because it cannot resume
25107 all threads together, or even for a single thread, if the thread must
25108 be stepped though some code before letting it run freely.
25110 @item *stopped,reason="@var{reason}",thread-id="@var{id}",stopped-threads="@var{stopped}",core="@var{core}"
25111 The target has stopped. The @var{reason} field can have one of the
25115 @item breakpoint-hit
25116 A breakpoint was reached.
25117 @item watchpoint-trigger
25118 A watchpoint was triggered.
25119 @item read-watchpoint-trigger
25120 A read watchpoint was triggered.
25121 @item access-watchpoint-trigger
25122 An access watchpoint was triggered.
25123 @item function-finished
25124 An -exec-finish or similar CLI command was accomplished.
25125 @item location-reached
25126 An -exec-until or similar CLI command was accomplished.
25127 @item watchpoint-scope
25128 A watchpoint has gone out of scope.
25129 @item end-stepping-range
25130 An -exec-next, -exec-next-instruction, -exec-step, -exec-step-instruction or
25131 similar CLI command was accomplished.
25132 @item exited-signalled
25133 The inferior exited because of a signal.
25135 The inferior exited.
25136 @item exited-normally
25137 The inferior exited normally.
25138 @item signal-received
25139 A signal was received by the inferior.
25142 The @var{id} field identifies the thread that directly caused the stop
25143 -- for example by hitting a breakpoint. Depending on whether all-stop
25144 mode is in effect (@pxref{All-Stop Mode}), @value{GDBN} may either
25145 stop all threads, or only the thread that directly triggered the stop.
25146 If all threads are stopped, the @var{stopped} field will have the
25147 value of @code{"all"}. Otherwise, the value of the @var{stopped}
25148 field will be a list of thread identifiers. Presently, this list will
25149 always include a single thread, but frontend should be prepared to see
25150 several threads in the list. The @var{core} field reports the
25151 processor core on which the stop event has happened. This field may be absent
25152 if such information is not available.
25154 @item =thread-group-added,id="@var{id}"
25155 @itemx =thread-group-removed,id="@var{id}"
25156 A thread group was either added or removed. The @var{id} field
25157 contains the @value{GDBN} identifier of the thread group. When a thread
25158 group is added, it generally might not be associated with a running
25159 process. When a thread group is removed, its id becomes invalid and
25160 cannot be used in any way.
25162 @item =thread-group-started,id="@var{id}",pid="@var{pid}"
25163 A thread group became associated with a running program,
25164 either because the program was just started or the thread group
25165 was attached to a program. The @var{id} field contains the
25166 @value{GDBN} identifier of the thread group. The @var{pid} field
25167 contains process identifier, specific to the operating system.
25169 @itemx =thread-group-exited,id="@var{id}"
25170 A thread group is no longer associated with a running program,
25171 either because the program has exited, or because it was detached
25172 from. The @var{id} field contains the @value{GDBN} identifier of the
25175 @item =thread-created,id="@var{id}",group-id="@var{gid}"
25176 @itemx =thread-exited,id="@var{id}",group-id="@var{gid}"
25177 A thread either was created, or has exited. The @var{id} field
25178 contains the @value{GDBN} identifier of the thread. The @var{gid}
25179 field identifies the thread group this thread belongs to.
25181 @item =thread-selected,id="@var{id}"
25182 Informs that the selected thread was changed as result of the last
25183 command. This notification is not emitted as result of @code{-thread-select}
25184 command but is emitted whenever an MI command that is not documented
25185 to change the selected thread actually changes it. In particular,
25186 invoking, directly or indirectly (via user-defined command), the CLI
25187 @code{thread} command, will generate this notification.
25189 We suggest that in response to this notification, front ends
25190 highlight the selected thread and cause subsequent commands to apply to
25193 @item =library-loaded,...
25194 Reports that a new library file was loaded by the program. This
25195 notification has 4 fields---@var{id}, @var{target-name},
25196 @var{host-name}, and @var{symbols-loaded}. The @var{id} field is an
25197 opaque identifier of the library. For remote debugging case,
25198 @var{target-name} and @var{host-name} fields give the name of the
25199 library file on the target, and on the host respectively. For native
25200 debugging, both those fields have the same value. The
25201 @var{symbols-loaded} field is emitted only for backward compatibility
25202 and should not be relied on to convey any useful information. The
25203 @var{thread-group} field, if present, specifies the id of the thread
25204 group in whose context the library was loaded. If the field is
25205 absent, it means the library was loaded in the context of all present
25208 @item =library-unloaded,...
25209 Reports that a library was unloaded by the program. This notification
25210 has 3 fields---@var{id}, @var{target-name} and @var{host-name} with
25211 the same meaning as for the @code{=library-loaded} notification.
25212 The @var{thread-group} field, if present, specifies the id of the
25213 thread group in whose context the library was unloaded. If the field is
25214 absent, it means the library was unloaded in the context of all present
25217 @item =breakpoint-created,bkpt=@{...@}
25218 @itemx =breakpoint-modified,bkpt=@{...@}
25219 @itemx =breakpoint-deleted,bkpt=@{...@}
25220 Reports that a breakpoint was created, modified, or deleted,
25221 respectively. Only user-visible breakpoints are reported to the MI
25224 The @var{bkpt} argument is of the same form as returned by the various
25225 breakpoint commands; @xref{GDB/MI Breakpoint Commands}.
25227 Note that if a breakpoint is emitted in the result record of a
25228 command, then it will not also be emitted in an async record.
25232 @node GDB/MI Frame Information
25233 @subsection @sc{gdb/mi} Frame Information
25235 Response from many MI commands includes an information about stack
25236 frame. This information is a tuple that may have the following
25241 The level of the stack frame. The innermost frame has the level of
25242 zero. This field is always present.
25245 The name of the function corresponding to the frame. This field may
25246 be absent if @value{GDBN} is unable to determine the function name.
25249 The code address for the frame. This field is always present.
25252 The name of the source files that correspond to the frame's code
25253 address. This field may be absent.
25256 The source line corresponding to the frames' code address. This field
25260 The name of the binary file (either executable or shared library) the
25261 corresponds to the frame's code address. This field may be absent.
25265 @node GDB/MI Thread Information
25266 @subsection @sc{gdb/mi} Thread Information
25268 Whenever @value{GDBN} has to report an information about a thread, it
25269 uses a tuple with the following fields:
25273 The numeric id assigned to the thread by @value{GDBN}. This field is
25277 Target-specific string identifying the thread. This field is always present.
25280 Additional information about the thread provided by the target.
25281 It is supposed to be human-readable and not interpreted by the
25282 frontend. This field is optional.
25285 Either @samp{stopped} or @samp{running}, depending on whether the
25286 thread is presently running. This field is always present.
25289 The value of this field is an integer number of the processor core the
25290 thread was last seen on. This field is optional.
25293 @node GDB/MI Ada Exception Information
25294 @subsection @sc{gdb/mi} Ada Exception Information
25296 Whenever a @code{*stopped} record is emitted because the program
25297 stopped after hitting an exception catchpoint (@pxref{Set Catchpoints}),
25298 @value{GDBN} provides the name of the exception that was raised via
25299 the @code{exception-name} field.
25301 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
25302 @node GDB/MI Simple Examples
25303 @section Simple Examples of @sc{gdb/mi} Interaction
25304 @cindex @sc{gdb/mi}, simple examples
25306 This subsection presents several simple examples of interaction using
25307 the @sc{gdb/mi} interface. In these examples, @samp{->} means that the
25308 following line is passed to @sc{gdb/mi} as input, while @samp{<-} means
25309 the output received from @sc{gdb/mi}.
25311 Note the line breaks shown in the examples are here only for
25312 readability, they don't appear in the real output.
25314 @subheading Setting a Breakpoint
25316 Setting a breakpoint generates synchronous output which contains detailed
25317 information of the breakpoint.
25320 -> -break-insert main
25321 <- ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
25322 enabled="y",addr="0x08048564",func="main",file="myprog.c",
25323 fullname="/home/nickrob/myprog.c",line="68",times="0"@}
25327 @subheading Program Execution
25329 Program execution generates asynchronous records and MI gives the
25330 reason that execution stopped.
25336 <- *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
25337 frame=@{addr="0x08048564",func="main",
25338 args=[@{name="argc",value="1"@},@{name="argv",value="0xbfc4d4d4"@}],
25339 file="myprog.c",fullname="/home/nickrob/myprog.c",line="68"@}
25344 <- *stopped,reason="exited-normally"
25348 @subheading Quitting @value{GDBN}
25350 Quitting @value{GDBN} just prints the result class @samp{^exit}.
25358 Please note that @samp{^exit} is printed immediately, but it might
25359 take some time for @value{GDBN} to actually exit. During that time, @value{GDBN}
25360 performs necessary cleanups, including killing programs being debugged
25361 or disconnecting from debug hardware, so the frontend should wait till
25362 @value{GDBN} exits and should only forcibly kill @value{GDBN} if it
25363 fails to exit in reasonable time.
25365 @subheading A Bad Command
25367 Here's what happens if you pass a non-existent command:
25371 <- ^error,msg="Undefined MI command: rubbish"
25376 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
25377 @node GDB/MI Command Description Format
25378 @section @sc{gdb/mi} Command Description Format
25380 The remaining sections describe blocks of commands. Each block of
25381 commands is laid out in a fashion similar to this section.
25383 @subheading Motivation
25385 The motivation for this collection of commands.
25387 @subheading Introduction
25389 A brief introduction to this collection of commands as a whole.
25391 @subheading Commands
25393 For each command in the block, the following is described:
25395 @subsubheading Synopsis
25398 -command @var{args}@dots{}
25401 @subsubheading Result
25403 @subsubheading @value{GDBN} Command
25405 The corresponding @value{GDBN} CLI command(s), if any.
25407 @subsubheading Example
25409 Example(s) formatted for readability. Some of the described commands have
25410 not been implemented yet and these are labeled N.A.@: (not available).
25413 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
25414 @node GDB/MI Breakpoint Commands
25415 @section @sc{gdb/mi} Breakpoint Commands
25417 @cindex breakpoint commands for @sc{gdb/mi}
25418 @cindex @sc{gdb/mi}, breakpoint commands
25419 This section documents @sc{gdb/mi} commands for manipulating
25422 @subheading The @code{-break-after} Command
25423 @findex -break-after
25425 @subsubheading Synopsis
25428 -break-after @var{number} @var{count}
25431 The breakpoint number @var{number} is not in effect until it has been
25432 hit @var{count} times. To see how this is reflected in the output of
25433 the @samp{-break-list} command, see the description of the
25434 @samp{-break-list} command below.
25436 @subsubheading @value{GDBN} Command
25438 The corresponding @value{GDBN} command is @samp{ignore}.
25440 @subsubheading Example
25445 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
25446 enabled="y",addr="0x000100d0",func="main",file="hello.c",
25447 fullname="/home/foo/hello.c",line="5",times="0"@}
25454 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
25455 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
25456 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
25457 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
25458 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
25459 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
25460 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
25461 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
25462 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
25463 line="5",times="0",ignore="3"@}]@}
25468 @subheading The @code{-break-catch} Command
25469 @findex -break-catch
25472 @subheading The @code{-break-commands} Command
25473 @findex -break-commands
25475 @subsubheading Synopsis
25478 -break-commands @var{number} [ @var{command1} ... @var{commandN} ]
25481 Specifies the CLI commands that should be executed when breakpoint
25482 @var{number} is hit. The parameters @var{command1} to @var{commandN}
25483 are the commands. If no command is specified, any previously-set
25484 commands are cleared. @xref{Break Commands}. Typical use of this
25485 functionality is tracing a program, that is, printing of values of
25486 some variables whenever breakpoint is hit and then continuing.
25488 @subsubheading @value{GDBN} Command
25490 The corresponding @value{GDBN} command is @samp{commands}.
25492 @subsubheading Example
25497 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
25498 enabled="y",addr="0x000100d0",func="main",file="hello.c",
25499 fullname="/home/foo/hello.c",line="5",times="0"@}
25501 -break-commands 1 "print v" "continue"
25506 @subheading The @code{-break-condition} Command
25507 @findex -break-condition
25509 @subsubheading Synopsis
25512 -break-condition @var{number} @var{expr}
25515 Breakpoint @var{number} will stop the program only if the condition in
25516 @var{expr} is true. The condition becomes part of the
25517 @samp{-break-list} output (see the description of the @samp{-break-list}
25520 @subsubheading @value{GDBN} Command
25522 The corresponding @value{GDBN} command is @samp{condition}.
25524 @subsubheading Example
25528 -break-condition 1 1
25532 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
25533 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
25534 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
25535 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
25536 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
25537 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
25538 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
25539 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
25540 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
25541 line="5",cond="1",times="0",ignore="3"@}]@}
25545 @subheading The @code{-break-delete} Command
25546 @findex -break-delete
25548 @subsubheading Synopsis
25551 -break-delete ( @var{breakpoint} )+
25554 Delete the breakpoint(s) whose number(s) are specified in the argument
25555 list. This is obviously reflected in the breakpoint list.
25557 @subsubheading @value{GDBN} Command
25559 The corresponding @value{GDBN} command is @samp{delete}.
25561 @subsubheading Example
25569 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
25570 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
25571 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
25572 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
25573 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
25574 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
25575 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
25580 @subheading The @code{-break-disable} Command
25581 @findex -break-disable
25583 @subsubheading Synopsis
25586 -break-disable ( @var{breakpoint} )+
25589 Disable the named @var{breakpoint}(s). The field @samp{enabled} in the
25590 break list is now set to @samp{n} for the named @var{breakpoint}(s).
25592 @subsubheading @value{GDBN} Command
25594 The corresponding @value{GDBN} command is @samp{disable}.
25596 @subsubheading Example
25604 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
25605 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
25606 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
25607 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
25608 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
25609 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
25610 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
25611 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="n",
25612 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
25613 line="5",times="0"@}]@}
25617 @subheading The @code{-break-enable} Command
25618 @findex -break-enable
25620 @subsubheading Synopsis
25623 -break-enable ( @var{breakpoint} )+
25626 Enable (previously disabled) @var{breakpoint}(s).
25628 @subsubheading @value{GDBN} Command
25630 The corresponding @value{GDBN} command is @samp{enable}.
25632 @subsubheading Example
25640 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
25641 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
25642 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
25643 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
25644 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
25645 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
25646 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
25647 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
25648 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
25649 line="5",times="0"@}]@}
25653 @subheading The @code{-break-info} Command
25654 @findex -break-info
25656 @subsubheading Synopsis
25659 -break-info @var{breakpoint}
25663 Get information about a single breakpoint.
25665 @subsubheading @value{GDBN} Command
25667 The corresponding @value{GDBN} command is @samp{info break @var{breakpoint}}.
25669 @subsubheading Example
25672 @subheading The @code{-break-insert} Command
25673 @findex -break-insert
25675 @subsubheading Synopsis
25678 -break-insert [ -t ] [ -h ] [ -f ] [ -d ] [ -a ]
25679 [ -c @var{condition} ] [ -i @var{ignore-count} ]
25680 [ -p @var{thread} ] [ @var{location} ]
25684 If specified, @var{location}, can be one of:
25691 @item filename:linenum
25692 @item filename:function
25696 The possible optional parameters of this command are:
25700 Insert a temporary breakpoint.
25702 Insert a hardware breakpoint.
25703 @item -c @var{condition}
25704 Make the breakpoint conditional on @var{condition}.
25705 @item -i @var{ignore-count}
25706 Initialize the @var{ignore-count}.
25708 If @var{location} cannot be parsed (for example if it
25709 refers to unknown files or functions), create a pending
25710 breakpoint. Without this flag, @value{GDBN} will report
25711 an error, and won't create a breakpoint, if @var{location}
25714 Create a disabled breakpoint.
25716 Create a tracepoint. @xref{Tracepoints}. When this parameter
25717 is used together with @samp{-h}, a fast tracepoint is created.
25720 @subsubheading Result
25722 The result is in the form:
25725 ^done,bkpt=@{number="@var{number}",type="@var{type}",disp="del"|"keep",
25726 enabled="y"|"n",addr="@var{hex}",func="@var{funcname}",file="@var{filename}",
25727 fullname="@var{full_filename}",line="@var{lineno}",[thread="@var{threadno},]
25728 times="@var{times}"@}
25732 where @var{number} is the @value{GDBN} number for this breakpoint,
25733 @var{funcname} is the name of the function where the breakpoint was
25734 inserted, @var{filename} is the name of the source file which contains
25735 this function, @var{lineno} is the source line number within that file
25736 and @var{times} the number of times that the breakpoint has been hit
25737 (always 0 for -break-insert but may be greater for -break-info or -break-list
25738 which use the same output).
25740 Note: this format is open to change.
25741 @c An out-of-band breakpoint instead of part of the result?
25743 @subsubheading @value{GDBN} Command
25745 The corresponding @value{GDBN} commands are @samp{break}, @samp{tbreak},
25746 @samp{hbreak}, @samp{thbreak}, and @samp{rbreak}.
25748 @subsubheading Example
25753 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",
25754 fullname="/home/foo/recursive2.c,line="4",times="0"@}
25756 -break-insert -t foo
25757 ^done,bkpt=@{number="2",addr="0x00010774",file="recursive2.c",
25758 fullname="/home/foo/recursive2.c,line="11",times="0"@}
25761 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
25762 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
25763 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
25764 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
25765 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
25766 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
25767 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
25768 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
25769 addr="0x0001072c", func="main",file="recursive2.c",
25770 fullname="/home/foo/recursive2.c,"line="4",times="0"@},
25771 bkpt=@{number="2",type="breakpoint",disp="del",enabled="y",
25772 addr="0x00010774",func="foo",file="recursive2.c",
25773 fullname="/home/foo/recursive2.c",line="11",times="0"@}]@}
25775 -break-insert -r foo.*
25776 ~int foo(int, int);
25777 ^done,bkpt=@{number="3",addr="0x00010774",file="recursive2.c,
25778 "fullname="/home/foo/recursive2.c",line="11",times="0"@}
25782 @subheading The @code{-break-list} Command
25783 @findex -break-list
25785 @subsubheading Synopsis
25791 Displays the list of inserted breakpoints, showing the following fields:
25795 number of the breakpoint
25797 type of the breakpoint: @samp{breakpoint} or @samp{watchpoint}
25799 should the breakpoint be deleted or disabled when it is hit: @samp{keep}
25802 is the breakpoint enabled or no: @samp{y} or @samp{n}
25804 memory location at which the breakpoint is set
25806 logical location of the breakpoint, expressed by function name, file
25809 number of times the breakpoint has been hit
25812 If there are no breakpoints or watchpoints, the @code{BreakpointTable}
25813 @code{body} field is an empty list.
25815 @subsubheading @value{GDBN} Command
25817 The corresponding @value{GDBN} command is @samp{info break}.
25819 @subsubheading Example
25824 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
25825 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
25826 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
25827 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
25828 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
25829 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
25830 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
25831 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
25832 addr="0x000100d0",func="main",file="hello.c",line="5",times="0"@},
25833 bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
25834 addr="0x00010114",func="foo",file="hello.c",fullname="/home/foo/hello.c",
25835 line="13",times="0"@}]@}
25839 Here's an example of the result when there are no breakpoints:
25844 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
25845 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
25846 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
25847 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
25848 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
25849 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
25850 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
25855 @subheading The @code{-break-passcount} Command
25856 @findex -break-passcount
25858 @subsubheading Synopsis
25861 -break-passcount @var{tracepoint-number} @var{passcount}
25864 Set the passcount for tracepoint @var{tracepoint-number} to
25865 @var{passcount}. If the breakpoint referred to by @var{tracepoint-number}
25866 is not a tracepoint, error is emitted. This corresponds to CLI
25867 command @samp{passcount}.
25869 @subheading The @code{-break-watch} Command
25870 @findex -break-watch
25872 @subsubheading Synopsis
25875 -break-watch [ -a | -r ]
25878 Create a watchpoint. With the @samp{-a} option it will create an
25879 @dfn{access} watchpoint, i.e., a watchpoint that triggers either on a
25880 read from or on a write to the memory location. With the @samp{-r}
25881 option, the watchpoint created is a @dfn{read} watchpoint, i.e., it will
25882 trigger only when the memory location is accessed for reading. Without
25883 either of the options, the watchpoint created is a regular watchpoint,
25884 i.e., it will trigger when the memory location is accessed for writing.
25885 @xref{Set Watchpoints, , Setting Watchpoints}.
25887 Note that @samp{-break-list} will report a single list of watchpoints and
25888 breakpoints inserted.
25890 @subsubheading @value{GDBN} Command
25892 The corresponding @value{GDBN} commands are @samp{watch}, @samp{awatch}, and
25895 @subsubheading Example
25897 Setting a watchpoint on a variable in the @code{main} function:
25902 ^done,wpt=@{number="2",exp="x"@}
25907 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="x"@},
25908 value=@{old="-268439212",new="55"@},
25909 frame=@{func="main",args=[],file="recursive2.c",
25910 fullname="/home/foo/bar/recursive2.c",line="5"@}
25914 Setting a watchpoint on a variable local to a function. @value{GDBN} will stop
25915 the program execution twice: first for the variable changing value, then
25916 for the watchpoint going out of scope.
25921 ^done,wpt=@{number="5",exp="C"@}
25926 *stopped,reason="watchpoint-trigger",
25927 wpt=@{number="5",exp="C"@},value=@{old="-276895068",new="3"@},
25928 frame=@{func="callee4",args=[],
25929 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
25930 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13"@}
25935 *stopped,reason="watchpoint-scope",wpnum="5",
25936 frame=@{func="callee3",args=[@{name="strarg",
25937 value="0x11940 \"A string argument.\""@}],
25938 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
25939 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
25943 Listing breakpoints and watchpoints, at different points in the program
25944 execution. Note that once the watchpoint goes out of scope, it is
25950 ^done,wpt=@{number="2",exp="C"@}
25953 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
25954 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
25955 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
25956 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
25957 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
25958 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
25959 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
25960 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
25961 addr="0x00010734",func="callee4",
25962 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
25963 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c"line="8",times="1"@},
25964 bkpt=@{number="2",type="watchpoint",disp="keep",
25965 enabled="y",addr="",what="C",times="0"@}]@}
25970 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="C"@},
25971 value=@{old="-276895068",new="3"@},
25972 frame=@{func="callee4",args=[],
25973 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
25974 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13"@}
25977 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
25978 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
25979 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
25980 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
25981 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
25982 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
25983 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
25984 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
25985 addr="0x00010734",func="callee4",
25986 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
25987 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",times="1"@},
25988 bkpt=@{number="2",type="watchpoint",disp="keep",
25989 enabled="y",addr="",what="C",times="-5"@}]@}
25993 ^done,reason="watchpoint-scope",wpnum="2",
25994 frame=@{func="callee3",args=[@{name="strarg",
25995 value="0x11940 \"A string argument.\""@}],
25996 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
25997 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
26000 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
26001 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
26002 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
26003 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
26004 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
26005 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
26006 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
26007 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
26008 addr="0x00010734",func="callee4",
26009 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
26010 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",
26015 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
26016 @node GDB/MI Program Context
26017 @section @sc{gdb/mi} Program Context
26019 @subheading The @code{-exec-arguments} Command
26020 @findex -exec-arguments
26023 @subsubheading Synopsis
26026 -exec-arguments @var{args}
26029 Set the inferior program arguments, to be used in the next
26032 @subsubheading @value{GDBN} Command
26034 The corresponding @value{GDBN} command is @samp{set args}.
26036 @subsubheading Example
26040 -exec-arguments -v word
26047 @subheading The @code{-exec-show-arguments} Command
26048 @findex -exec-show-arguments
26050 @subsubheading Synopsis
26053 -exec-show-arguments
26056 Print the arguments of the program.
26058 @subsubheading @value{GDBN} Command
26060 The corresponding @value{GDBN} command is @samp{show args}.
26062 @subsubheading Example
26067 @subheading The @code{-environment-cd} Command
26068 @findex -environment-cd
26070 @subsubheading Synopsis
26073 -environment-cd @var{pathdir}
26076 Set @value{GDBN}'s working directory.
26078 @subsubheading @value{GDBN} Command
26080 The corresponding @value{GDBN} command is @samp{cd}.
26082 @subsubheading Example
26086 -environment-cd /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
26092 @subheading The @code{-environment-directory} Command
26093 @findex -environment-directory
26095 @subsubheading Synopsis
26098 -environment-directory [ -r ] [ @var{pathdir} ]+
26101 Add directories @var{pathdir} to beginning of search path for source files.
26102 If the @samp{-r} option is used, the search path is reset to the default
26103 search path. If directories @var{pathdir} are supplied in addition to the
26104 @samp{-r} option, the search path is first reset and then addition
26106 Multiple directories may be specified, separated by blanks. Specifying
26107 multiple directories in a single command
26108 results in the directories added to the beginning of the
26109 search path in the same order they were presented in the command.
26110 If blanks are needed as
26111 part of a directory name, double-quotes should be used around
26112 the name. In the command output, the path will show up separated
26113 by the system directory-separator character. The directory-separator
26114 character must not be used
26115 in any directory name.
26116 If no directories are specified, the current search path is displayed.
26118 @subsubheading @value{GDBN} Command
26120 The corresponding @value{GDBN} command is @samp{dir}.
26122 @subsubheading Example
26126 -environment-directory /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
26127 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
26129 -environment-directory ""
26130 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
26132 -environment-directory -r /home/jjohnstn/src/gdb /usr/src
26133 ^done,source-path="/home/jjohnstn/src/gdb:/usr/src:$cdir:$cwd"
26135 -environment-directory -r
26136 ^done,source-path="$cdir:$cwd"
26141 @subheading The @code{-environment-path} Command
26142 @findex -environment-path
26144 @subsubheading Synopsis
26147 -environment-path [ -r ] [ @var{pathdir} ]+
26150 Add directories @var{pathdir} to beginning of search path for object files.
26151 If the @samp{-r} option is used, the search path is reset to the original
26152 search path that existed at gdb start-up. If directories @var{pathdir} are
26153 supplied in addition to the
26154 @samp{-r} option, the search path is first reset and then addition
26156 Multiple directories may be specified, separated by blanks. Specifying
26157 multiple directories in a single command
26158 results in the directories added to the beginning of the
26159 search path in the same order they were presented in the command.
26160 If blanks are needed as
26161 part of a directory name, double-quotes should be used around
26162 the name. In the command output, the path will show up separated
26163 by the system directory-separator character. The directory-separator
26164 character must not be used
26165 in any directory name.
26166 If no directories are specified, the current path is displayed.
26169 @subsubheading @value{GDBN} Command
26171 The corresponding @value{GDBN} command is @samp{path}.
26173 @subsubheading Example
26178 ^done,path="/usr/bin"
26180 -environment-path /kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb /bin
26181 ^done,path="/kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb:/bin:/usr/bin"
26183 -environment-path -r /usr/local/bin
26184 ^done,path="/usr/local/bin:/usr/bin"
26189 @subheading The @code{-environment-pwd} Command
26190 @findex -environment-pwd
26192 @subsubheading Synopsis
26198 Show the current working directory.
26200 @subsubheading @value{GDBN} Command
26202 The corresponding @value{GDBN} command is @samp{pwd}.
26204 @subsubheading Example
26209 ^done,cwd="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb"
26213 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
26214 @node GDB/MI Thread Commands
26215 @section @sc{gdb/mi} Thread Commands
26218 @subheading The @code{-thread-info} Command
26219 @findex -thread-info
26221 @subsubheading Synopsis
26224 -thread-info [ @var{thread-id} ]
26227 Reports information about either a specific thread, if
26228 the @var{thread-id} parameter is present, or about all
26229 threads. When printing information about all threads,
26230 also reports the current thread.
26232 @subsubheading @value{GDBN} Command
26234 The @samp{info thread} command prints the same information
26237 @subsubheading Result
26239 The result is a list of threads. The following attributes are
26240 defined for a given thread:
26244 This field exists only for the current thread. It has the value @samp{*}.
26247 The identifier that @value{GDBN} uses to refer to the thread.
26250 The identifier that the target uses to refer to the thread.
26253 Extra information about the thread, in a target-specific format. This
26257 The name of the thread. If the user specified a name using the
26258 @code{thread name} command, then this name is given. Otherwise, if
26259 @value{GDBN} can extract the thread name from the target, then that
26260 name is given. If @value{GDBN} cannot find the thread name, then this
26264 The stack frame currently executing in the thread.
26267 The thread's state. The @samp{state} field may have the following
26272 The thread is stopped. Frame information is available for stopped
26276 The thread is running. There's no frame information for running
26282 If @value{GDBN} can find the CPU core on which this thread is running,
26283 then this field is the core identifier. This field is optional.
26287 @subsubheading Example
26292 @{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
26293 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",
26294 args=[]@},state="running"@},
26295 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
26296 frame=@{level="0",addr="0x0804891f",func="foo",
26297 args=[@{name="i",value="10"@}],
26298 file="/tmp/a.c",fullname="/tmp/a.c",line="158"@},
26299 state="running"@}],
26300 current-thread-id="1"
26304 @subheading The @code{-thread-list-ids} Command
26305 @findex -thread-list-ids
26307 @subsubheading Synopsis
26313 Produces a list of the currently known @value{GDBN} thread ids. At the
26314 end of the list it also prints the total number of such threads.
26316 This command is retained for historical reasons, the
26317 @code{-thread-info} command should be used instead.
26319 @subsubheading @value{GDBN} Command
26321 Part of @samp{info threads} supplies the same information.
26323 @subsubheading Example
26328 ^done,thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
26329 current-thread-id="1",number-of-threads="3"
26334 @subheading The @code{-thread-select} Command
26335 @findex -thread-select
26337 @subsubheading Synopsis
26340 -thread-select @var{threadnum}
26343 Make @var{threadnum} the current thread. It prints the number of the new
26344 current thread, and the topmost frame for that thread.
26346 This command is deprecated in favor of explicitly using the
26347 @samp{--thread} option to each command.
26349 @subsubheading @value{GDBN} Command
26351 The corresponding @value{GDBN} command is @samp{thread}.
26353 @subsubheading Example
26360 *stopped,reason="end-stepping-range",thread-id="2",line="187",
26361 file="../../../devo/gdb/testsuite/gdb.threads/linux-dp.c"
26365 thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
26366 number-of-threads="3"
26369 ^done,new-thread-id="3",
26370 frame=@{level="0",func="vprintf",
26371 args=[@{name="format",value="0x8048e9c \"%*s%c %d %c\\n\""@},
26372 @{name="arg",value="0x2"@}],file="vprintf.c",line="31"@}
26376 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
26377 @node GDB/MI Program Execution
26378 @section @sc{gdb/mi} Program Execution
26380 These are the asynchronous commands which generate the out-of-band
26381 record @samp{*stopped}. Currently @value{GDBN} only really executes
26382 asynchronously with remote targets and this interaction is mimicked in
26385 @subheading The @code{-exec-continue} Command
26386 @findex -exec-continue
26388 @subsubheading Synopsis
26391 -exec-continue [--reverse] [--all|--thread-group N]
26394 Resumes the execution of the inferior program, which will continue
26395 to execute until it reaches a debugger stop event. If the
26396 @samp{--reverse} option is specified, execution resumes in reverse until
26397 it reaches a stop event. Stop events may include
26400 breakpoints or watchpoints
26402 signals or exceptions
26404 the end of the process (or its beginning under @samp{--reverse})
26406 the end or beginning of a replay log if one is being used.
26408 In all-stop mode (@pxref{All-Stop
26409 Mode}), may resume only one thread, or all threads, depending on the
26410 value of the @samp{scheduler-locking} variable. If @samp{--all} is
26411 specified, all threads (in all inferiors) will be resumed. The @samp{--all} option is
26412 ignored in all-stop mode. If the @samp{--thread-group} options is
26413 specified, then all threads in that thread group are resumed.
26415 @subsubheading @value{GDBN} Command
26417 The corresponding @value{GDBN} corresponding is @samp{continue}.
26419 @subsubheading Example
26426 *stopped,reason="breakpoint-hit",disp="keep",bkptno="2",frame=@{
26427 func="foo",args=[],file="hello.c",fullname="/home/foo/bar/hello.c",
26433 @subheading The @code{-exec-finish} Command
26434 @findex -exec-finish
26436 @subsubheading Synopsis
26439 -exec-finish [--reverse]
26442 Resumes the execution of the inferior program until the current
26443 function is exited. Displays the results returned by the function.
26444 If the @samp{--reverse} option is specified, resumes the reverse
26445 execution of the inferior program until the point where current
26446 function was called.
26448 @subsubheading @value{GDBN} Command
26450 The corresponding @value{GDBN} command is @samp{finish}.
26452 @subsubheading Example
26454 Function returning @code{void}.
26461 *stopped,reason="function-finished",frame=@{func="main",args=[],
26462 file="hello.c",fullname="/home/foo/bar/hello.c",line="7"@}
26466 Function returning other than @code{void}. The name of the internal
26467 @value{GDBN} variable storing the result is printed, together with the
26474 *stopped,reason="function-finished",frame=@{addr="0x000107b0",func="foo",
26475 args=[@{name="a",value="1"],@{name="b",value="9"@}@},
26476 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
26477 gdb-result-var="$1",return-value="0"
26482 @subheading The @code{-exec-interrupt} Command
26483 @findex -exec-interrupt
26485 @subsubheading Synopsis
26488 -exec-interrupt [--all|--thread-group N]
26491 Interrupts the background execution of the target. Note how the token
26492 associated with the stop message is the one for the execution command
26493 that has been interrupted. The token for the interrupt itself only
26494 appears in the @samp{^done} output. If the user is trying to
26495 interrupt a non-running program, an error message will be printed.
26497 Note that when asynchronous execution is enabled, this command is
26498 asynchronous just like other execution commands. That is, first the
26499 @samp{^done} response will be printed, and the target stop will be
26500 reported after that using the @samp{*stopped} notification.
26502 In non-stop mode, only the context thread is interrupted by default.
26503 All threads (in all inferiors) will be interrupted if the
26504 @samp{--all} option is specified. If the @samp{--thread-group}
26505 option is specified, all threads in that group will be interrupted.
26507 @subsubheading @value{GDBN} Command
26509 The corresponding @value{GDBN} command is @samp{interrupt}.
26511 @subsubheading Example
26522 111*stopped,signal-name="SIGINT",signal-meaning="Interrupt",
26523 frame=@{addr="0x00010140",func="foo",args=[],file="try.c",
26524 fullname="/home/foo/bar/try.c",line="13"@}
26529 ^error,msg="mi_cmd_exec_interrupt: Inferior not executing."
26533 @subheading The @code{-exec-jump} Command
26536 @subsubheading Synopsis
26539 -exec-jump @var{location}
26542 Resumes execution of the inferior program at the location specified by
26543 parameter. @xref{Specify Location}, for a description of the
26544 different forms of @var{location}.
26546 @subsubheading @value{GDBN} Command
26548 The corresponding @value{GDBN} command is @samp{jump}.
26550 @subsubheading Example
26553 -exec-jump foo.c:10
26554 *running,thread-id="all"
26559 @subheading The @code{-exec-next} Command
26562 @subsubheading Synopsis
26565 -exec-next [--reverse]
26568 Resumes execution of the inferior program, stopping when the beginning
26569 of the next source line is reached.
26571 If the @samp{--reverse} option is specified, resumes reverse execution
26572 of the inferior program, stopping at the beginning of the previous
26573 source line. If you issue this command on the first line of a
26574 function, it will take you back to the caller of that function, to the
26575 source line where the function was called.
26578 @subsubheading @value{GDBN} Command
26580 The corresponding @value{GDBN} command is @samp{next}.
26582 @subsubheading Example
26588 *stopped,reason="end-stepping-range",line="8",file="hello.c"
26593 @subheading The @code{-exec-next-instruction} Command
26594 @findex -exec-next-instruction
26596 @subsubheading Synopsis
26599 -exec-next-instruction [--reverse]
26602 Executes one machine instruction. If the instruction is a function
26603 call, continues until the function returns. If the program stops at an
26604 instruction in the middle of a source line, the address will be
26607 If the @samp{--reverse} option is specified, resumes reverse execution
26608 of the inferior program, stopping at the previous instruction. If the
26609 previously executed instruction was a return from another function,
26610 it will continue to execute in reverse until the call to that function
26611 (from the current stack frame) is reached.
26613 @subsubheading @value{GDBN} Command
26615 The corresponding @value{GDBN} command is @samp{nexti}.
26617 @subsubheading Example
26621 -exec-next-instruction
26625 *stopped,reason="end-stepping-range",
26626 addr="0x000100d4",line="5",file="hello.c"
26631 @subheading The @code{-exec-return} Command
26632 @findex -exec-return
26634 @subsubheading Synopsis
26640 Makes current function return immediately. Doesn't execute the inferior.
26641 Displays the new current frame.
26643 @subsubheading @value{GDBN} Command
26645 The corresponding @value{GDBN} command is @samp{return}.
26647 @subsubheading Example
26651 200-break-insert callee4
26652 200^done,bkpt=@{number="1",addr="0x00010734",
26653 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",line="8"@}
26658 000*stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
26659 frame=@{func="callee4",args=[],
26660 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
26661 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8"@}
26667 111^done,frame=@{level="0",func="callee3",
26668 args=[@{name="strarg",
26669 value="0x11940 \"A string argument.\""@}],
26670 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
26671 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
26676 @subheading The @code{-exec-run} Command
26679 @subsubheading Synopsis
26682 -exec-run [--all | --thread-group N]
26685 Starts execution of the inferior from the beginning. The inferior
26686 executes until either a breakpoint is encountered or the program
26687 exits. In the latter case the output will include an exit code, if
26688 the program has exited exceptionally.
26690 When no option is specified, the current inferior is started. If the
26691 @samp{--thread-group} option is specified, it should refer to a thread
26692 group of type @samp{process}, and that thread group will be started.
26693 If the @samp{--all} option is specified, then all inferiors will be started.
26695 @subsubheading @value{GDBN} Command
26697 The corresponding @value{GDBN} command is @samp{run}.
26699 @subsubheading Examples
26704 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",line="4"@}
26709 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
26710 frame=@{func="main",args=[],file="recursive2.c",
26711 fullname="/home/foo/bar/recursive2.c",line="4"@}
26716 Program exited normally:
26724 *stopped,reason="exited-normally"
26729 Program exited exceptionally:
26737 *stopped,reason="exited",exit-code="01"
26741 Another way the program can terminate is if it receives a signal such as
26742 @code{SIGINT}. In this case, @sc{gdb/mi} displays this:
26746 *stopped,reason="exited-signalled",signal-name="SIGINT",
26747 signal-meaning="Interrupt"
26751 @c @subheading -exec-signal
26754 @subheading The @code{-exec-step} Command
26757 @subsubheading Synopsis
26760 -exec-step [--reverse]
26763 Resumes execution of the inferior program, stopping when the beginning
26764 of the next source line is reached, if the next source line is not a
26765 function call. If it is, stop at the first instruction of the called
26766 function. If the @samp{--reverse} option is specified, resumes reverse
26767 execution of the inferior program, stopping at the beginning of the
26768 previously executed source line.
26770 @subsubheading @value{GDBN} Command
26772 The corresponding @value{GDBN} command is @samp{step}.
26774 @subsubheading Example
26776 Stepping into a function:
26782 *stopped,reason="end-stepping-range",
26783 frame=@{func="foo",args=[@{name="a",value="10"@},
26784 @{name="b",value="0"@}],file="recursive2.c",
26785 fullname="/home/foo/bar/recursive2.c",line="11"@}
26795 *stopped,reason="end-stepping-range",line="14",file="recursive2.c"
26800 @subheading The @code{-exec-step-instruction} Command
26801 @findex -exec-step-instruction
26803 @subsubheading Synopsis
26806 -exec-step-instruction [--reverse]
26809 Resumes the inferior which executes one machine instruction. If the
26810 @samp{--reverse} option is specified, resumes reverse execution of the
26811 inferior program, stopping at the previously executed instruction.
26812 The output, once @value{GDBN} has stopped, will vary depending on
26813 whether we have stopped in the middle of a source line or not. In the
26814 former case, the address at which the program stopped will be printed
26817 @subsubheading @value{GDBN} Command
26819 The corresponding @value{GDBN} command is @samp{stepi}.
26821 @subsubheading Example
26825 -exec-step-instruction
26829 *stopped,reason="end-stepping-range",
26830 frame=@{func="foo",args=[],file="try.c",
26831 fullname="/home/foo/bar/try.c",line="10"@}
26833 -exec-step-instruction
26837 *stopped,reason="end-stepping-range",
26838 frame=@{addr="0x000100f4",func="foo",args=[],file="try.c",
26839 fullname="/home/foo/bar/try.c",line="10"@}
26844 @subheading The @code{-exec-until} Command
26845 @findex -exec-until
26847 @subsubheading Synopsis
26850 -exec-until [ @var{location} ]
26853 Executes the inferior until the @var{location} specified in the
26854 argument is reached. If there is no argument, the inferior executes
26855 until a source line greater than the current one is reached. The
26856 reason for stopping in this case will be @samp{location-reached}.
26858 @subsubheading @value{GDBN} Command
26860 The corresponding @value{GDBN} command is @samp{until}.
26862 @subsubheading Example
26866 -exec-until recursive2.c:6
26870 *stopped,reason="location-reached",frame=@{func="main",args=[],
26871 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="6"@}
26876 @subheading -file-clear
26877 Is this going away????
26880 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
26881 @node GDB/MI Stack Manipulation
26882 @section @sc{gdb/mi} Stack Manipulation Commands
26885 @subheading The @code{-stack-info-frame} Command
26886 @findex -stack-info-frame
26888 @subsubheading Synopsis
26894 Get info on the selected frame.
26896 @subsubheading @value{GDBN} Command
26898 The corresponding @value{GDBN} command is @samp{info frame} or @samp{frame}
26899 (without arguments).
26901 @subsubheading Example
26906 ^done,frame=@{level="1",addr="0x0001076c",func="callee3",
26907 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
26908 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17"@}
26912 @subheading The @code{-stack-info-depth} Command
26913 @findex -stack-info-depth
26915 @subsubheading Synopsis
26918 -stack-info-depth [ @var{max-depth} ]
26921 Return the depth of the stack. If the integer argument @var{max-depth}
26922 is specified, do not count beyond @var{max-depth} frames.
26924 @subsubheading @value{GDBN} Command
26926 There's no equivalent @value{GDBN} command.
26928 @subsubheading Example
26930 For a stack with frame levels 0 through 11:
26937 -stack-info-depth 4
26940 -stack-info-depth 12
26943 -stack-info-depth 11
26946 -stack-info-depth 13
26951 @subheading The @code{-stack-list-arguments} Command
26952 @findex -stack-list-arguments
26954 @subsubheading Synopsis
26957 -stack-list-arguments @var{print-values}
26958 [ @var{low-frame} @var{high-frame} ]
26961 Display a list of the arguments for the frames between @var{low-frame}
26962 and @var{high-frame} (inclusive). If @var{low-frame} and
26963 @var{high-frame} are not provided, list the arguments for the whole
26964 call stack. If the two arguments are equal, show the single frame
26965 at the corresponding level. It is an error if @var{low-frame} is
26966 larger than the actual number of frames. On the other hand,
26967 @var{high-frame} may be larger than the actual number of frames, in
26968 which case only existing frames will be returned.
26970 If @var{print-values} is 0 or @code{--no-values}, print only the names of
26971 the variables; if it is 1 or @code{--all-values}, print also their
26972 values; and if it is 2 or @code{--simple-values}, print the name,
26973 type and value for simple data types, and the name and type for arrays,
26974 structures and unions.
26976 Use of this command to obtain arguments in a single frame is
26977 deprecated in favor of the @samp{-stack-list-variables} command.
26979 @subsubheading @value{GDBN} Command
26981 @value{GDBN} does not have an equivalent command. @code{gdbtk} has a
26982 @samp{gdb_get_args} command which partially overlaps with the
26983 functionality of @samp{-stack-list-arguments}.
26985 @subsubheading Example
26992 frame=@{level="0",addr="0x00010734",func="callee4",
26993 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
26994 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8"@},
26995 frame=@{level="1",addr="0x0001076c",func="callee3",
26996 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
26997 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17"@},
26998 frame=@{level="2",addr="0x0001078c",func="callee2",
26999 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
27000 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="22"@},
27001 frame=@{level="3",addr="0x000107b4",func="callee1",
27002 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
27003 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="27"@},
27004 frame=@{level="4",addr="0x000107e0",func="main",
27005 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
27006 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="32"@}]
27008 -stack-list-arguments 0
27011 frame=@{level="0",args=[]@},
27012 frame=@{level="1",args=[name="strarg"]@},
27013 frame=@{level="2",args=[name="intarg",name="strarg"]@},
27014 frame=@{level="3",args=[name="intarg",name="strarg",name="fltarg"]@},
27015 frame=@{level="4",args=[]@}]
27017 -stack-list-arguments 1
27020 frame=@{level="0",args=[]@},
27022 args=[@{name="strarg",value="0x11940 \"A string argument.\""@}]@},
27023 frame=@{level="2",args=[
27024 @{name="intarg",value="2"@},
27025 @{name="strarg",value="0x11940 \"A string argument.\""@}]@},
27026 @{frame=@{level="3",args=[
27027 @{name="intarg",value="2"@},
27028 @{name="strarg",value="0x11940 \"A string argument.\""@},
27029 @{name="fltarg",value="3.5"@}]@},
27030 frame=@{level="4",args=[]@}]
27032 -stack-list-arguments 0 2 2
27033 ^done,stack-args=[frame=@{level="2",args=[name="intarg",name="strarg"]@}]
27035 -stack-list-arguments 1 2 2
27036 ^done,stack-args=[frame=@{level="2",
27037 args=[@{name="intarg",value="2"@},
27038 @{name="strarg",value="0x11940 \"A string argument.\""@}]@}]
27042 @c @subheading -stack-list-exception-handlers
27045 @subheading The @code{-stack-list-frames} Command
27046 @findex -stack-list-frames
27048 @subsubheading Synopsis
27051 -stack-list-frames [ @var{low-frame} @var{high-frame} ]
27054 List the frames currently on the stack. For each frame it displays the
27059 The frame number, 0 being the topmost frame, i.e., the innermost function.
27061 The @code{$pc} value for that frame.
27065 File name of the source file where the function lives.
27066 @item @var{fullname}
27067 The full file name of the source file where the function lives.
27069 Line number corresponding to the @code{$pc}.
27071 The shared library where this function is defined. This is only given
27072 if the frame's function is not known.
27075 If invoked without arguments, this command prints a backtrace for the
27076 whole stack. If given two integer arguments, it shows the frames whose
27077 levels are between the two arguments (inclusive). If the two arguments
27078 are equal, it shows the single frame at the corresponding level. It is
27079 an error if @var{low-frame} is larger than the actual number of
27080 frames. On the other hand, @var{high-frame} may be larger than the
27081 actual number of frames, in which case only existing frames will be returned.
27083 @subsubheading @value{GDBN} Command
27085 The corresponding @value{GDBN} commands are @samp{backtrace} and @samp{where}.
27087 @subsubheading Example
27089 Full stack backtrace:
27095 [frame=@{level="0",addr="0x0001076c",func="foo",
27096 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="11"@},
27097 frame=@{level="1",addr="0x000107a4",func="foo",
27098 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
27099 frame=@{level="2",addr="0x000107a4",func="foo",
27100 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
27101 frame=@{level="3",addr="0x000107a4",func="foo",
27102 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
27103 frame=@{level="4",addr="0x000107a4",func="foo",
27104 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
27105 frame=@{level="5",addr="0x000107a4",func="foo",
27106 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
27107 frame=@{level="6",addr="0x000107a4",func="foo",
27108 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
27109 frame=@{level="7",addr="0x000107a4",func="foo",
27110 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
27111 frame=@{level="8",addr="0x000107a4",func="foo",
27112 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
27113 frame=@{level="9",addr="0x000107a4",func="foo",
27114 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
27115 frame=@{level="10",addr="0x000107a4",func="foo",
27116 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
27117 frame=@{level="11",addr="0x00010738",func="main",
27118 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="4"@}]
27122 Show frames between @var{low_frame} and @var{high_frame}:
27126 -stack-list-frames 3 5
27128 [frame=@{level="3",addr="0x000107a4",func="foo",
27129 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
27130 frame=@{level="4",addr="0x000107a4",func="foo",
27131 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
27132 frame=@{level="5",addr="0x000107a4",func="foo",
27133 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@}]
27137 Show a single frame:
27141 -stack-list-frames 3 3
27143 [frame=@{level="3",addr="0x000107a4",func="foo",
27144 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@}]
27149 @subheading The @code{-stack-list-locals} Command
27150 @findex -stack-list-locals
27152 @subsubheading Synopsis
27155 -stack-list-locals @var{print-values}
27158 Display the local variable names for the selected frame. If
27159 @var{print-values} is 0 or @code{--no-values}, print only the names of
27160 the variables; if it is 1 or @code{--all-values}, print also their
27161 values; and if it is 2 or @code{--simple-values}, print the name,
27162 type and value for simple data types, and the name and type for arrays,
27163 structures and unions. In this last case, a frontend can immediately
27164 display the value of simple data types and create variable objects for
27165 other data types when the user wishes to explore their values in
27168 This command is deprecated in favor of the
27169 @samp{-stack-list-variables} command.
27171 @subsubheading @value{GDBN} Command
27173 @samp{info locals} in @value{GDBN}, @samp{gdb_get_locals} in @code{gdbtk}.
27175 @subsubheading Example
27179 -stack-list-locals 0
27180 ^done,locals=[name="A",name="B",name="C"]
27182 -stack-list-locals --all-values
27183 ^done,locals=[@{name="A",value="1"@},@{name="B",value="2"@},
27184 @{name="C",value="@{1, 2, 3@}"@}]
27185 -stack-list-locals --simple-values
27186 ^done,locals=[@{name="A",type="int",value="1"@},
27187 @{name="B",type="int",value="2"@},@{name="C",type="int [3]"@}]
27191 @subheading The @code{-stack-list-variables} Command
27192 @findex -stack-list-variables
27194 @subsubheading Synopsis
27197 -stack-list-variables @var{print-values}
27200 Display the names of local variables and function arguments for the selected frame. If
27201 @var{print-values} is 0 or @code{--no-values}, print only the names of
27202 the variables; if it is 1 or @code{--all-values}, print also their
27203 values; and if it is 2 or @code{--simple-values}, print the name,
27204 type and value for simple data types, and the name and type for arrays,
27205 structures and unions.
27207 @subsubheading Example
27211 -stack-list-variables --thread 1 --frame 0 --all-values
27212 ^done,variables=[@{name="x",value="11"@},@{name="s",value="@{a = 1, b = 2@}"@}]
27217 @subheading The @code{-stack-select-frame} Command
27218 @findex -stack-select-frame
27220 @subsubheading Synopsis
27223 -stack-select-frame @var{framenum}
27226 Change the selected frame. Select a different frame @var{framenum} on
27229 This command in deprecated in favor of passing the @samp{--frame}
27230 option to every command.
27232 @subsubheading @value{GDBN} Command
27234 The corresponding @value{GDBN} commands are @samp{frame}, @samp{up},
27235 @samp{down}, @samp{select-frame}, @samp{up-silent}, and @samp{down-silent}.
27237 @subsubheading Example
27241 -stack-select-frame 2
27246 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
27247 @node GDB/MI Variable Objects
27248 @section @sc{gdb/mi} Variable Objects
27252 @subheading Motivation for Variable Objects in @sc{gdb/mi}
27254 For the implementation of a variable debugger window (locals, watched
27255 expressions, etc.), we are proposing the adaptation of the existing code
27256 used by @code{Insight}.
27258 The two main reasons for that are:
27262 It has been proven in practice (it is already on its second generation).
27265 It will shorten development time (needless to say how important it is
27269 The original interface was designed to be used by Tcl code, so it was
27270 slightly changed so it could be used through @sc{gdb/mi}. This section
27271 describes the @sc{gdb/mi} operations that will be available and gives some
27272 hints about their use.
27274 @emph{Note}: In addition to the set of operations described here, we
27275 expect the @sc{gui} implementation of a variable window to require, at
27276 least, the following operations:
27279 @item @code{-gdb-show} @code{output-radix}
27280 @item @code{-stack-list-arguments}
27281 @item @code{-stack-list-locals}
27282 @item @code{-stack-select-frame}
27287 @subheading Introduction to Variable Objects
27289 @cindex variable objects in @sc{gdb/mi}
27291 Variable objects are "object-oriented" MI interface for examining and
27292 changing values of expressions. Unlike some other MI interfaces that
27293 work with expressions, variable objects are specifically designed for
27294 simple and efficient presentation in the frontend. A variable object
27295 is identified by string name. When a variable object is created, the
27296 frontend specifies the expression for that variable object. The
27297 expression can be a simple variable, or it can be an arbitrary complex
27298 expression, and can even involve CPU registers. After creating a
27299 variable object, the frontend can invoke other variable object
27300 operations---for example to obtain or change the value of a variable
27301 object, or to change display format.
27303 Variable objects have hierarchical tree structure. Any variable object
27304 that corresponds to a composite type, such as structure in C, has
27305 a number of child variable objects, for example corresponding to each
27306 element of a structure. A child variable object can itself have
27307 children, recursively. Recursion ends when we reach
27308 leaf variable objects, which always have built-in types. Child variable
27309 objects are created only by explicit request, so if a frontend
27310 is not interested in the children of a particular variable object, no
27311 child will be created.
27313 For a leaf variable object it is possible to obtain its value as a
27314 string, or set the value from a string. String value can be also
27315 obtained for a non-leaf variable object, but it's generally a string
27316 that only indicates the type of the object, and does not list its
27317 contents. Assignment to a non-leaf variable object is not allowed.
27319 A frontend does not need to read the values of all variable objects each time
27320 the program stops. Instead, MI provides an update command that lists all
27321 variable objects whose values has changed since the last update
27322 operation. This considerably reduces the amount of data that must
27323 be transferred to the frontend. As noted above, children variable
27324 objects are created on demand, and only leaf variable objects have a
27325 real value. As result, gdb will read target memory only for leaf
27326 variables that frontend has created.
27328 The automatic update is not always desirable. For example, a frontend
27329 might want to keep a value of some expression for future reference,
27330 and never update it. For another example, fetching memory is
27331 relatively slow for embedded targets, so a frontend might want
27332 to disable automatic update for the variables that are either not
27333 visible on the screen, or ``closed''. This is possible using so
27334 called ``frozen variable objects''. Such variable objects are never
27335 implicitly updated.
27337 Variable objects can be either @dfn{fixed} or @dfn{floating}. For the
27338 fixed variable object, the expression is parsed when the variable
27339 object is created, including associating identifiers to specific
27340 variables. The meaning of expression never changes. For a floating
27341 variable object the values of variables whose names appear in the
27342 expressions are re-evaluated every time in the context of the current
27343 frame. Consider this example:
27348 struct work_state state;
27355 If a fixed variable object for the @code{state} variable is created in
27356 this function, and we enter the recursive call, the variable
27357 object will report the value of @code{state} in the top-level
27358 @code{do_work} invocation. On the other hand, a floating variable
27359 object will report the value of @code{state} in the current frame.
27361 If an expression specified when creating a fixed variable object
27362 refers to a local variable, the variable object becomes bound to the
27363 thread and frame in which the variable object is created. When such
27364 variable object is updated, @value{GDBN} makes sure that the
27365 thread/frame combination the variable object is bound to still exists,
27366 and re-evaluates the variable object in context of that thread/frame.
27368 The following is the complete set of @sc{gdb/mi} operations defined to
27369 access this functionality:
27371 @multitable @columnfractions .4 .6
27372 @item @strong{Operation}
27373 @tab @strong{Description}
27375 @item @code{-enable-pretty-printing}
27376 @tab enable Python-based pretty-printing
27377 @item @code{-var-create}
27378 @tab create a variable object
27379 @item @code{-var-delete}
27380 @tab delete the variable object and/or its children
27381 @item @code{-var-set-format}
27382 @tab set the display format of this variable
27383 @item @code{-var-show-format}
27384 @tab show the display format of this variable
27385 @item @code{-var-info-num-children}
27386 @tab tells how many children this object has
27387 @item @code{-var-list-children}
27388 @tab return a list of the object's children
27389 @item @code{-var-info-type}
27390 @tab show the type of this variable object
27391 @item @code{-var-info-expression}
27392 @tab print parent-relative expression that this variable object represents
27393 @item @code{-var-info-path-expression}
27394 @tab print full expression that this variable object represents
27395 @item @code{-var-show-attributes}
27396 @tab is this variable editable? does it exist here?
27397 @item @code{-var-evaluate-expression}
27398 @tab get the value of this variable
27399 @item @code{-var-assign}
27400 @tab set the value of this variable
27401 @item @code{-var-update}
27402 @tab update the variable and its children
27403 @item @code{-var-set-frozen}
27404 @tab set frozeness attribute
27405 @item @code{-var-set-update-range}
27406 @tab set range of children to display on update
27409 In the next subsection we describe each operation in detail and suggest
27410 how it can be used.
27412 @subheading Description And Use of Operations on Variable Objects
27414 @subheading The @code{-enable-pretty-printing} Command
27415 @findex -enable-pretty-printing
27418 -enable-pretty-printing
27421 @value{GDBN} allows Python-based visualizers to affect the output of the
27422 MI variable object commands. However, because there was no way to
27423 implement this in a fully backward-compatible way, a front end must
27424 request that this functionality be enabled.
27426 Once enabled, this feature cannot be disabled.
27428 Note that if Python support has not been compiled into @value{GDBN},
27429 this command will still succeed (and do nothing).
27431 This feature is currently (as of @value{GDBN} 7.0) experimental, and
27432 may work differently in future versions of @value{GDBN}.
27434 @subheading The @code{-var-create} Command
27435 @findex -var-create
27437 @subsubheading Synopsis
27440 -var-create @{@var{name} | "-"@}
27441 @{@var{frame-addr} | "*" | "@@"@} @var{expression}
27444 This operation creates a variable object, which allows the monitoring of
27445 a variable, the result of an expression, a memory cell or a CPU
27448 The @var{name} parameter is the string by which the object can be
27449 referenced. It must be unique. If @samp{-} is specified, the varobj
27450 system will generate a string ``varNNNNNN'' automatically. It will be
27451 unique provided that one does not specify @var{name} of that format.
27452 The command fails if a duplicate name is found.
27454 The frame under which the expression should be evaluated can be
27455 specified by @var{frame-addr}. A @samp{*} indicates that the current
27456 frame should be used. A @samp{@@} indicates that a floating variable
27457 object must be created.
27459 @var{expression} is any expression valid on the current language set (must not
27460 begin with a @samp{*}), or one of the following:
27464 @samp{*@var{addr}}, where @var{addr} is the address of a memory cell
27467 @samp{*@var{addr}-@var{addr}} --- a memory address range (TBD)
27470 @samp{$@var{regname}} --- a CPU register name
27473 @cindex dynamic varobj
27474 A varobj's contents may be provided by a Python-based pretty-printer. In this
27475 case the varobj is known as a @dfn{dynamic varobj}. Dynamic varobjs
27476 have slightly different semantics in some cases. If the
27477 @code{-enable-pretty-printing} command is not sent, then @value{GDBN}
27478 will never create a dynamic varobj. This ensures backward
27479 compatibility for existing clients.
27481 @subsubheading Result
27483 This operation returns attributes of the newly-created varobj. These
27488 The name of the varobj.
27491 The number of children of the varobj. This number is not necessarily
27492 reliable for a dynamic varobj. Instead, you must examine the
27493 @samp{has_more} attribute.
27496 The varobj's scalar value. For a varobj whose type is some sort of
27497 aggregate (e.g., a @code{struct}), or for a dynamic varobj, this value
27498 will not be interesting.
27501 The varobj's type. This is a string representation of the type, as
27502 would be printed by the @value{GDBN} CLI.
27505 If a variable object is bound to a specific thread, then this is the
27506 thread's identifier.
27509 For a dynamic varobj, this indicates whether there appear to be any
27510 children available. For a non-dynamic varobj, this will be 0.
27513 This attribute will be present and have the value @samp{1} if the
27514 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
27515 then this attribute will not be present.
27518 A dynamic varobj can supply a display hint to the front end. The
27519 value comes directly from the Python pretty-printer object's
27520 @code{display_hint} method. @xref{Pretty Printing API}.
27523 Typical output will look like this:
27526 name="@var{name}",numchild="@var{N}",type="@var{type}",thread-id="@var{M}",
27527 has_more="@var{has_more}"
27531 @subheading The @code{-var-delete} Command
27532 @findex -var-delete
27534 @subsubheading Synopsis
27537 -var-delete [ -c ] @var{name}
27540 Deletes a previously created variable object and all of its children.
27541 With the @samp{-c} option, just deletes the children.
27543 Returns an error if the object @var{name} is not found.
27546 @subheading The @code{-var-set-format} Command
27547 @findex -var-set-format
27549 @subsubheading Synopsis
27552 -var-set-format @var{name} @var{format-spec}
27555 Sets the output format for the value of the object @var{name} to be
27558 @anchor{-var-set-format}
27559 The syntax for the @var{format-spec} is as follows:
27562 @var{format-spec} @expansion{}
27563 @{binary | decimal | hexadecimal | octal | natural@}
27566 The natural format is the default format choosen automatically
27567 based on the variable type (like decimal for an @code{int}, hex
27568 for pointers, etc.).
27570 For a variable with children, the format is set only on the
27571 variable itself, and the children are not affected.
27573 @subheading The @code{-var-show-format} Command
27574 @findex -var-show-format
27576 @subsubheading Synopsis
27579 -var-show-format @var{name}
27582 Returns the format used to display the value of the object @var{name}.
27585 @var{format} @expansion{}
27590 @subheading The @code{-var-info-num-children} Command
27591 @findex -var-info-num-children
27593 @subsubheading Synopsis
27596 -var-info-num-children @var{name}
27599 Returns the number of children of a variable object @var{name}:
27605 Note that this number is not completely reliable for a dynamic varobj.
27606 It will return the current number of children, but more children may
27610 @subheading The @code{-var-list-children} Command
27611 @findex -var-list-children
27613 @subsubheading Synopsis
27616 -var-list-children [@var{print-values}] @var{name} [@var{from} @var{to}]
27618 @anchor{-var-list-children}
27620 Return a list of the children of the specified variable object and
27621 create variable objects for them, if they do not already exist. With
27622 a single argument or if @var{print-values} has a value of 0 or
27623 @code{--no-values}, print only the names of the variables; if
27624 @var{print-values} is 1 or @code{--all-values}, also print their
27625 values; and if it is 2 or @code{--simple-values} print the name and
27626 value for simple data types and just the name for arrays, structures
27629 @var{from} and @var{to}, if specified, indicate the range of children
27630 to report. If @var{from} or @var{to} is less than zero, the range is
27631 reset and all children will be reported. Otherwise, children starting
27632 at @var{from} (zero-based) and up to and excluding @var{to} will be
27635 If a child range is requested, it will only affect the current call to
27636 @code{-var-list-children}, but not future calls to @code{-var-update}.
27637 For this, you must instead use @code{-var-set-update-range}. The
27638 intent of this approach is to enable a front end to implement any
27639 update approach it likes; for example, scrolling a view may cause the
27640 front end to request more children with @code{-var-list-children}, and
27641 then the front end could call @code{-var-set-update-range} with a
27642 different range to ensure that future updates are restricted to just
27645 For each child the following results are returned:
27650 Name of the variable object created for this child.
27653 The expression to be shown to the user by the front end to designate this child.
27654 For example this may be the name of a structure member.
27656 For a dynamic varobj, this value cannot be used to form an
27657 expression. There is no way to do this at all with a dynamic varobj.
27659 For C/C@t{++} structures there are several pseudo children returned to
27660 designate access qualifiers. For these pseudo children @var{exp} is
27661 @samp{public}, @samp{private}, or @samp{protected}. In this case the
27662 type and value are not present.
27664 A dynamic varobj will not report the access qualifying
27665 pseudo-children, regardless of the language. This information is not
27666 available at all with a dynamic varobj.
27669 Number of children this child has. For a dynamic varobj, this will be
27673 The type of the child.
27676 If values were requested, this is the value.
27679 If this variable object is associated with a thread, this is the thread id.
27680 Otherwise this result is not present.
27683 If the variable object is frozen, this variable will be present with a value of 1.
27686 The result may have its own attributes:
27690 A dynamic varobj can supply a display hint to the front end. The
27691 value comes directly from the Python pretty-printer object's
27692 @code{display_hint} method. @xref{Pretty Printing API}.
27695 This is an integer attribute which is nonzero if there are children
27696 remaining after the end of the selected range.
27699 @subsubheading Example
27703 -var-list-children n
27704 ^done,numchild=@var{n},children=[child=@{name=@var{name},exp=@var{exp},
27705 numchild=@var{n},type=@var{type}@},@r{(repeats N times)}]
27707 -var-list-children --all-values n
27708 ^done,numchild=@var{n},children=[child=@{name=@var{name},exp=@var{exp},
27709 numchild=@var{n},value=@var{value},type=@var{type}@},@r{(repeats N times)}]
27713 @subheading The @code{-var-info-type} Command
27714 @findex -var-info-type
27716 @subsubheading Synopsis
27719 -var-info-type @var{name}
27722 Returns the type of the specified variable @var{name}. The type is
27723 returned as a string in the same format as it is output by the
27727 type=@var{typename}
27731 @subheading The @code{-var-info-expression} Command
27732 @findex -var-info-expression
27734 @subsubheading Synopsis
27737 -var-info-expression @var{name}
27740 Returns a string that is suitable for presenting this
27741 variable object in user interface. The string is generally
27742 not valid expression in the current language, and cannot be evaluated.
27744 For example, if @code{a} is an array, and variable object
27745 @code{A} was created for @code{a}, then we'll get this output:
27748 (gdb) -var-info-expression A.1
27749 ^done,lang="C",exp="1"
27753 Here, the values of @code{lang} can be @code{@{"C" | "C++" | "Java"@}}.
27755 Note that the output of the @code{-var-list-children} command also
27756 includes those expressions, so the @code{-var-info-expression} command
27759 @subheading The @code{-var-info-path-expression} Command
27760 @findex -var-info-path-expression
27762 @subsubheading Synopsis
27765 -var-info-path-expression @var{name}
27768 Returns an expression that can be evaluated in the current
27769 context and will yield the same value that a variable object has.
27770 Compare this with the @code{-var-info-expression} command, which
27771 result can be used only for UI presentation. Typical use of
27772 the @code{-var-info-path-expression} command is creating a
27773 watchpoint from a variable object.
27775 This command is currently not valid for children of a dynamic varobj,
27776 and will give an error when invoked on one.
27778 For example, suppose @code{C} is a C@t{++} class, derived from class
27779 @code{Base}, and that the @code{Base} class has a member called
27780 @code{m_size}. Assume a variable @code{c} is has the type of
27781 @code{C} and a variable object @code{C} was created for variable
27782 @code{c}. Then, we'll get this output:
27784 (gdb) -var-info-path-expression C.Base.public.m_size
27785 ^done,path_expr=((Base)c).m_size)
27788 @subheading The @code{-var-show-attributes} Command
27789 @findex -var-show-attributes
27791 @subsubheading Synopsis
27794 -var-show-attributes @var{name}
27797 List attributes of the specified variable object @var{name}:
27800 status=@var{attr} [ ( ,@var{attr} )* ]
27804 where @var{attr} is @code{@{ @{ editable | noneditable @} | TBD @}}.
27806 @subheading The @code{-var-evaluate-expression} Command
27807 @findex -var-evaluate-expression
27809 @subsubheading Synopsis
27812 -var-evaluate-expression [-f @var{format-spec}] @var{name}
27815 Evaluates the expression that is represented by the specified variable
27816 object and returns its value as a string. The format of the string
27817 can be specified with the @samp{-f} option. The possible values of
27818 this option are the same as for @code{-var-set-format}
27819 (@pxref{-var-set-format}). If the @samp{-f} option is not specified,
27820 the current display format will be used. The current display format
27821 can be changed using the @code{-var-set-format} command.
27827 Note that one must invoke @code{-var-list-children} for a variable
27828 before the value of a child variable can be evaluated.
27830 @subheading The @code{-var-assign} Command
27831 @findex -var-assign
27833 @subsubheading Synopsis
27836 -var-assign @var{name} @var{expression}
27839 Assigns the value of @var{expression} to the variable object specified
27840 by @var{name}. The object must be @samp{editable}. If the variable's
27841 value is altered by the assign, the variable will show up in any
27842 subsequent @code{-var-update} list.
27844 @subsubheading Example
27852 ^done,changelist=[@{name="var1",in_scope="true",type_changed="false"@}]
27856 @subheading The @code{-var-update} Command
27857 @findex -var-update
27859 @subsubheading Synopsis
27862 -var-update [@var{print-values}] @{@var{name} | "*"@}
27865 Reevaluate the expressions corresponding to the variable object
27866 @var{name} and all its direct and indirect children, and return the
27867 list of variable objects whose values have changed; @var{name} must
27868 be a root variable object. Here, ``changed'' means that the result of
27869 @code{-var-evaluate-expression} before and after the
27870 @code{-var-update} is different. If @samp{*} is used as the variable
27871 object names, all existing variable objects are updated, except
27872 for frozen ones (@pxref{-var-set-frozen}). The option
27873 @var{print-values} determines whether both names and values, or just
27874 names are printed. The possible values of this option are the same
27875 as for @code{-var-list-children} (@pxref{-var-list-children}). It is
27876 recommended to use the @samp{--all-values} option, to reduce the
27877 number of MI commands needed on each program stop.
27879 With the @samp{*} parameter, if a variable object is bound to a
27880 currently running thread, it will not be updated, without any
27883 If @code{-var-set-update-range} was previously used on a varobj, then
27884 only the selected range of children will be reported.
27886 @code{-var-update} reports all the changed varobjs in a tuple named
27889 Each item in the change list is itself a tuple holding:
27893 The name of the varobj.
27896 If values were requested for this update, then this field will be
27897 present and will hold the value of the varobj.
27900 @anchor{-var-update}
27901 This field is a string which may take one of three values:
27905 The variable object's current value is valid.
27908 The variable object does not currently hold a valid value but it may
27909 hold one in the future if its associated expression comes back into
27913 The variable object no longer holds a valid value.
27914 This can occur when the executable file being debugged has changed,
27915 either through recompilation or by using the @value{GDBN} @code{file}
27916 command. The front end should normally choose to delete these variable
27920 In the future new values may be added to this list so the front should
27921 be prepared for this possibility. @xref{GDB/MI Development and Front Ends, ,@sc{GDB/MI} Development and Front Ends}.
27924 This is only present if the varobj is still valid. If the type
27925 changed, then this will be the string @samp{true}; otherwise it will
27929 If the varobj's type changed, then this field will be present and will
27932 @item new_num_children
27933 For a dynamic varobj, if the number of children changed, or if the
27934 type changed, this will be the new number of children.
27936 The @samp{numchild} field in other varobj responses is generally not
27937 valid for a dynamic varobj -- it will show the number of children that
27938 @value{GDBN} knows about, but because dynamic varobjs lazily
27939 instantiate their children, this will not reflect the number of
27940 children which may be available.
27942 The @samp{new_num_children} attribute only reports changes to the
27943 number of children known by @value{GDBN}. This is the only way to
27944 detect whether an update has removed children (which necessarily can
27945 only happen at the end of the update range).
27948 The display hint, if any.
27951 This is an integer value, which will be 1 if there are more children
27952 available outside the varobj's update range.
27955 This attribute will be present and have the value @samp{1} if the
27956 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
27957 then this attribute will not be present.
27960 If new children were added to a dynamic varobj within the selected
27961 update range (as set by @code{-var-set-update-range}), then they will
27962 be listed in this attribute.
27965 @subsubheading Example
27972 -var-update --all-values var1
27973 ^done,changelist=[@{name="var1",value="3",in_scope="true",
27974 type_changed="false"@}]
27978 @subheading The @code{-var-set-frozen} Command
27979 @findex -var-set-frozen
27980 @anchor{-var-set-frozen}
27982 @subsubheading Synopsis
27985 -var-set-frozen @var{name} @var{flag}
27988 Set the frozenness flag on the variable object @var{name}. The
27989 @var{flag} parameter should be either @samp{1} to make the variable
27990 frozen or @samp{0} to make it unfrozen. If a variable object is
27991 frozen, then neither itself, nor any of its children, are
27992 implicitly updated by @code{-var-update} of
27993 a parent variable or by @code{-var-update *}. Only
27994 @code{-var-update} of the variable itself will update its value and
27995 values of its children. After a variable object is unfrozen, it is
27996 implicitly updated by all subsequent @code{-var-update} operations.
27997 Unfreezing a variable does not update it, only subsequent
27998 @code{-var-update} does.
28000 @subsubheading Example
28004 -var-set-frozen V 1
28009 @subheading The @code{-var-set-update-range} command
28010 @findex -var-set-update-range
28011 @anchor{-var-set-update-range}
28013 @subsubheading Synopsis
28016 -var-set-update-range @var{name} @var{from} @var{to}
28019 Set the range of children to be returned by future invocations of
28020 @code{-var-update}.
28022 @var{from} and @var{to} indicate the range of children to report. If
28023 @var{from} or @var{to} is less than zero, the range is reset and all
28024 children will be reported. Otherwise, children starting at @var{from}
28025 (zero-based) and up to and excluding @var{to} will be reported.
28027 @subsubheading Example
28031 -var-set-update-range V 1 2
28035 @subheading The @code{-var-set-visualizer} command
28036 @findex -var-set-visualizer
28037 @anchor{-var-set-visualizer}
28039 @subsubheading Synopsis
28042 -var-set-visualizer @var{name} @var{visualizer}
28045 Set a visualizer for the variable object @var{name}.
28047 @var{visualizer} is the visualizer to use. The special value
28048 @samp{None} means to disable any visualizer in use.
28050 If not @samp{None}, @var{visualizer} must be a Python expression.
28051 This expression must evaluate to a callable object which accepts a
28052 single argument. @value{GDBN} will call this object with the value of
28053 the varobj @var{name} as an argument (this is done so that the same
28054 Python pretty-printing code can be used for both the CLI and MI).
28055 When called, this object must return an object which conforms to the
28056 pretty-printing interface (@pxref{Pretty Printing API}).
28058 The pre-defined function @code{gdb.default_visualizer} may be used to
28059 select a visualizer by following the built-in process
28060 (@pxref{Selecting Pretty-Printers}). This is done automatically when
28061 a varobj is created, and so ordinarily is not needed.
28063 This feature is only available if Python support is enabled. The MI
28064 command @code{-list-features} (@pxref{GDB/MI Miscellaneous Commands})
28065 can be used to check this.
28067 @subsubheading Example
28069 Resetting the visualizer:
28073 -var-set-visualizer V None
28077 Reselecting the default (type-based) visualizer:
28081 -var-set-visualizer V gdb.default_visualizer
28085 Suppose @code{SomeClass} is a visualizer class. A lambda expression
28086 can be used to instantiate this class for a varobj:
28090 -var-set-visualizer V "lambda val: SomeClass()"
28094 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28095 @node GDB/MI Data Manipulation
28096 @section @sc{gdb/mi} Data Manipulation
28098 @cindex data manipulation, in @sc{gdb/mi}
28099 @cindex @sc{gdb/mi}, data manipulation
28100 This section describes the @sc{gdb/mi} commands that manipulate data:
28101 examine memory and registers, evaluate expressions, etc.
28103 @c REMOVED FROM THE INTERFACE.
28104 @c @subheading -data-assign
28105 @c Change the value of a program variable. Plenty of side effects.
28106 @c @subsubheading GDB Command
28108 @c @subsubheading Example
28111 @subheading The @code{-data-disassemble} Command
28112 @findex -data-disassemble
28114 @subsubheading Synopsis
28118 [ -s @var{start-addr} -e @var{end-addr} ]
28119 | [ -f @var{filename} -l @var{linenum} [ -n @var{lines} ] ]
28127 @item @var{start-addr}
28128 is the beginning address (or @code{$pc})
28129 @item @var{end-addr}
28131 @item @var{filename}
28132 is the name of the file to disassemble
28133 @item @var{linenum}
28134 is the line number to disassemble around
28136 is the number of disassembly lines to be produced. If it is -1,
28137 the whole function will be disassembled, in case no @var{end-addr} is
28138 specified. If @var{end-addr} is specified as a non-zero value, and
28139 @var{lines} is lower than the number of disassembly lines between
28140 @var{start-addr} and @var{end-addr}, only @var{lines} lines are
28141 displayed; if @var{lines} is higher than the number of lines between
28142 @var{start-addr} and @var{end-addr}, only the lines up to @var{end-addr}
28145 is either 0 (meaning only disassembly), 1 (meaning mixed source and
28146 disassembly), 2 (meaning disassembly with raw opcodes), or 3 (meaning
28147 mixed source and disassembly with raw opcodes).
28150 @subsubheading Result
28152 The output for each instruction is composed of four fields:
28161 Note that whatever included in the instruction field, is not manipulated
28162 directly by @sc{gdb/mi}, i.e., it is not possible to adjust its format.
28164 @subsubheading @value{GDBN} Command
28166 There's no direct mapping from this command to the CLI.
28168 @subsubheading Example
28170 Disassemble from the current value of @code{$pc} to @code{$pc + 20}:
28174 -data-disassemble -s $pc -e "$pc + 20" -- 0
28177 @{address="0x000107c0",func-name="main",offset="4",
28178 inst="mov 2, %o0"@},
28179 @{address="0x000107c4",func-name="main",offset="8",
28180 inst="sethi %hi(0x11800), %o2"@},
28181 @{address="0x000107c8",func-name="main",offset="12",
28182 inst="or %o2, 0x140, %o1\t! 0x11940 <_lib_version+8>"@},
28183 @{address="0x000107cc",func-name="main",offset="16",
28184 inst="sethi %hi(0x11800), %o2"@},
28185 @{address="0x000107d0",func-name="main",offset="20",
28186 inst="or %o2, 0x168, %o4\t! 0x11968 <_lib_version+48>"@}]
28190 Disassemble the whole @code{main} function. Line 32 is part of
28194 -data-disassemble -f basics.c -l 32 -- 0
28196 @{address="0x000107bc",func-name="main",offset="0",
28197 inst="save %sp, -112, %sp"@},
28198 @{address="0x000107c0",func-name="main",offset="4",
28199 inst="mov 2, %o0"@},
28200 @{address="0x000107c4",func-name="main",offset="8",
28201 inst="sethi %hi(0x11800), %o2"@},
28203 @{address="0x0001081c",func-name="main",offset="96",inst="ret "@},
28204 @{address="0x00010820",func-name="main",offset="100",inst="restore "@}]
28208 Disassemble 3 instructions from the start of @code{main}:
28212 -data-disassemble -f basics.c -l 32 -n 3 -- 0
28214 @{address="0x000107bc",func-name="main",offset="0",
28215 inst="save %sp, -112, %sp"@},
28216 @{address="0x000107c0",func-name="main",offset="4",
28217 inst="mov 2, %o0"@},
28218 @{address="0x000107c4",func-name="main",offset="8",
28219 inst="sethi %hi(0x11800), %o2"@}]
28223 Disassemble 3 instructions from the start of @code{main} in mixed mode:
28227 -data-disassemble -f basics.c -l 32 -n 3 -- 1
28229 src_and_asm_line=@{line="31",
28230 file="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb/ \
28231 testsuite/gdb.mi/basics.c",line_asm_insn=[
28232 @{address="0x000107bc",func-name="main",offset="0",
28233 inst="save %sp, -112, %sp"@}]@},
28234 src_and_asm_line=@{line="32",
28235 file="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb/ \
28236 testsuite/gdb.mi/basics.c",line_asm_insn=[
28237 @{address="0x000107c0",func-name="main",offset="4",
28238 inst="mov 2, %o0"@},
28239 @{address="0x000107c4",func-name="main",offset="8",
28240 inst="sethi %hi(0x11800), %o2"@}]@}]
28245 @subheading The @code{-data-evaluate-expression} Command
28246 @findex -data-evaluate-expression
28248 @subsubheading Synopsis
28251 -data-evaluate-expression @var{expr}
28254 Evaluate @var{expr} as an expression. The expression could contain an
28255 inferior function call. The function call will execute synchronously.
28256 If the expression contains spaces, it must be enclosed in double quotes.
28258 @subsubheading @value{GDBN} Command
28260 The corresponding @value{GDBN} commands are @samp{print}, @samp{output}, and
28261 @samp{call}. In @code{gdbtk} only, there's a corresponding
28262 @samp{gdb_eval} command.
28264 @subsubheading Example
28266 In the following example, the numbers that precede the commands are the
28267 @dfn{tokens} described in @ref{GDB/MI Command Syntax, ,@sc{gdb/mi}
28268 Command Syntax}. Notice how @sc{gdb/mi} returns the same tokens in its
28272 211-data-evaluate-expression A
28275 311-data-evaluate-expression &A
28276 311^done,value="0xefffeb7c"
28278 411-data-evaluate-expression A+3
28281 511-data-evaluate-expression "A + 3"
28287 @subheading The @code{-data-list-changed-registers} Command
28288 @findex -data-list-changed-registers
28290 @subsubheading Synopsis
28293 -data-list-changed-registers
28296 Display a list of the registers that have changed.
28298 @subsubheading @value{GDBN} Command
28300 @value{GDBN} doesn't have a direct analog for this command; @code{gdbtk}
28301 has the corresponding command @samp{gdb_changed_register_list}.
28303 @subsubheading Example
28305 On a PPC MBX board:
28313 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",frame=@{
28314 func="main",args=[],file="try.c",fullname="/home/foo/bar/try.c",
28317 -data-list-changed-registers
28318 ^done,changed-registers=["0","1","2","4","5","6","7","8","9",
28319 "10","11","13","14","15","16","17","18","19","20","21","22","23",
28320 "24","25","26","27","28","30","31","64","65","66","67","69"]
28325 @subheading The @code{-data-list-register-names} Command
28326 @findex -data-list-register-names
28328 @subsubheading Synopsis
28331 -data-list-register-names [ ( @var{regno} )+ ]
28334 Show a list of register names for the current target. If no arguments
28335 are given, it shows a list of the names of all the registers. If
28336 integer numbers are given as arguments, it will print a list of the
28337 names of the registers corresponding to the arguments. To ensure
28338 consistency between a register name and its number, the output list may
28339 include empty register names.
28341 @subsubheading @value{GDBN} Command
28343 @value{GDBN} does not have a command which corresponds to
28344 @samp{-data-list-register-names}. In @code{gdbtk} there is a
28345 corresponding command @samp{gdb_regnames}.
28347 @subsubheading Example
28349 For the PPC MBX board:
28352 -data-list-register-names
28353 ^done,register-names=["r0","r1","r2","r3","r4","r5","r6","r7",
28354 "r8","r9","r10","r11","r12","r13","r14","r15","r16","r17","r18",
28355 "r19","r20","r21","r22","r23","r24","r25","r26","r27","r28","r29",
28356 "r30","r31","f0","f1","f2","f3","f4","f5","f6","f7","f8","f9",
28357 "f10","f11","f12","f13","f14","f15","f16","f17","f18","f19","f20",
28358 "f21","f22","f23","f24","f25","f26","f27","f28","f29","f30","f31",
28359 "", "pc","ps","cr","lr","ctr","xer"]
28361 -data-list-register-names 1 2 3
28362 ^done,register-names=["r1","r2","r3"]
28366 @subheading The @code{-data-list-register-values} Command
28367 @findex -data-list-register-values
28369 @subsubheading Synopsis
28372 -data-list-register-values @var{fmt} [ ( @var{regno} )*]
28375 Display the registers' contents. @var{fmt} is the format according to
28376 which the registers' contents are to be returned, followed by an optional
28377 list of numbers specifying the registers to display. A missing list of
28378 numbers indicates that the contents of all the registers must be returned.
28380 Allowed formats for @var{fmt} are:
28397 @subsubheading @value{GDBN} Command
28399 The corresponding @value{GDBN} commands are @samp{info reg}, @samp{info
28400 all-reg}, and (in @code{gdbtk}) @samp{gdb_fetch_registers}.
28402 @subsubheading Example
28404 For a PPC MBX board (note: line breaks are for readability only, they
28405 don't appear in the actual output):
28409 -data-list-register-values r 64 65
28410 ^done,register-values=[@{number="64",value="0xfe00a300"@},
28411 @{number="65",value="0x00029002"@}]
28413 -data-list-register-values x
28414 ^done,register-values=[@{number="0",value="0xfe0043c8"@},
28415 @{number="1",value="0x3fff88"@},@{number="2",value="0xfffffffe"@},
28416 @{number="3",value="0x0"@},@{number="4",value="0xa"@},
28417 @{number="5",value="0x3fff68"@},@{number="6",value="0x3fff58"@},
28418 @{number="7",value="0xfe011e98"@},@{number="8",value="0x2"@},
28419 @{number="9",value="0xfa202820"@},@{number="10",value="0xfa202808"@},
28420 @{number="11",value="0x1"@},@{number="12",value="0x0"@},
28421 @{number="13",value="0x4544"@},@{number="14",value="0xffdfffff"@},
28422 @{number="15",value="0xffffffff"@},@{number="16",value="0xfffffeff"@},
28423 @{number="17",value="0xefffffed"@},@{number="18",value="0xfffffffe"@},
28424 @{number="19",value="0xffffffff"@},@{number="20",value="0xffffffff"@},
28425 @{number="21",value="0xffffffff"@},@{number="22",value="0xfffffff7"@},
28426 @{number="23",value="0xffffffff"@},@{number="24",value="0xffffffff"@},
28427 @{number="25",value="0xffffffff"@},@{number="26",value="0xfffffffb"@},
28428 @{number="27",value="0xffffffff"@},@{number="28",value="0xf7bfffff"@},
28429 @{number="29",value="0x0"@},@{number="30",value="0xfe010000"@},
28430 @{number="31",value="0x0"@},@{number="32",value="0x0"@},
28431 @{number="33",value="0x0"@},@{number="34",value="0x0"@},
28432 @{number="35",value="0x0"@},@{number="36",value="0x0"@},
28433 @{number="37",value="0x0"@},@{number="38",value="0x0"@},
28434 @{number="39",value="0x0"@},@{number="40",value="0x0"@},
28435 @{number="41",value="0x0"@},@{number="42",value="0x0"@},
28436 @{number="43",value="0x0"@},@{number="44",value="0x0"@},
28437 @{number="45",value="0x0"@},@{number="46",value="0x0"@},
28438 @{number="47",value="0x0"@},@{number="48",value="0x0"@},
28439 @{number="49",value="0x0"@},@{number="50",value="0x0"@},
28440 @{number="51",value="0x0"@},@{number="52",value="0x0"@},
28441 @{number="53",value="0x0"@},@{number="54",value="0x0"@},
28442 @{number="55",value="0x0"@},@{number="56",value="0x0"@},
28443 @{number="57",value="0x0"@},@{number="58",value="0x0"@},
28444 @{number="59",value="0x0"@},@{number="60",value="0x0"@},
28445 @{number="61",value="0x0"@},@{number="62",value="0x0"@},
28446 @{number="63",value="0x0"@},@{number="64",value="0xfe00a300"@},
28447 @{number="65",value="0x29002"@},@{number="66",value="0x202f04b5"@},
28448 @{number="67",value="0xfe0043b0"@},@{number="68",value="0xfe00b3e4"@},
28449 @{number="69",value="0x20002b03"@}]
28454 @subheading The @code{-data-read-memory} Command
28455 @findex -data-read-memory
28457 This command is deprecated, use @code{-data-read-memory-bytes} instead.
28459 @subsubheading Synopsis
28462 -data-read-memory [ -o @var{byte-offset} ]
28463 @var{address} @var{word-format} @var{word-size}
28464 @var{nr-rows} @var{nr-cols} [ @var{aschar} ]
28471 @item @var{address}
28472 An expression specifying the address of the first memory word to be
28473 read. Complex expressions containing embedded white space should be
28474 quoted using the C convention.
28476 @item @var{word-format}
28477 The format to be used to print the memory words. The notation is the
28478 same as for @value{GDBN}'s @code{print} command (@pxref{Output Formats,
28481 @item @var{word-size}
28482 The size of each memory word in bytes.
28484 @item @var{nr-rows}
28485 The number of rows in the output table.
28487 @item @var{nr-cols}
28488 The number of columns in the output table.
28491 If present, indicates that each row should include an @sc{ascii} dump. The
28492 value of @var{aschar} is used as a padding character when a byte is not a
28493 member of the printable @sc{ascii} character set (printable @sc{ascii}
28494 characters are those whose code is between 32 and 126, inclusively).
28496 @item @var{byte-offset}
28497 An offset to add to the @var{address} before fetching memory.
28500 This command displays memory contents as a table of @var{nr-rows} by
28501 @var{nr-cols} words, each word being @var{word-size} bytes. In total,
28502 @code{@var{nr-rows} * @var{nr-cols} * @var{word-size}} bytes are read
28503 (returned as @samp{total-bytes}). Should less than the requested number
28504 of bytes be returned by the target, the missing words are identified
28505 using @samp{N/A}. The number of bytes read from the target is returned
28506 in @samp{nr-bytes} and the starting address used to read memory in
28509 The address of the next/previous row or page is available in
28510 @samp{next-row} and @samp{prev-row}, @samp{next-page} and
28513 @subsubheading @value{GDBN} Command
28515 The corresponding @value{GDBN} command is @samp{x}. @code{gdbtk} has
28516 @samp{gdb_get_mem} memory read command.
28518 @subsubheading Example
28520 Read six bytes of memory starting at @code{bytes+6} but then offset by
28521 @code{-6} bytes. Format as three rows of two columns. One byte per
28522 word. Display each word in hex.
28526 9-data-read-memory -o -6 -- bytes+6 x 1 3 2
28527 9^done,addr="0x00001390",nr-bytes="6",total-bytes="6",
28528 next-row="0x00001396",prev-row="0x0000138e",next-page="0x00001396",
28529 prev-page="0x0000138a",memory=[
28530 @{addr="0x00001390",data=["0x00","0x01"]@},
28531 @{addr="0x00001392",data=["0x02","0x03"]@},
28532 @{addr="0x00001394",data=["0x04","0x05"]@}]
28536 Read two bytes of memory starting at address @code{shorts + 64} and
28537 display as a single word formatted in decimal.
28541 5-data-read-memory shorts+64 d 2 1 1
28542 5^done,addr="0x00001510",nr-bytes="2",total-bytes="2",
28543 next-row="0x00001512",prev-row="0x0000150e",
28544 next-page="0x00001512",prev-page="0x0000150e",memory=[
28545 @{addr="0x00001510",data=["128"]@}]
28549 Read thirty two bytes of memory starting at @code{bytes+16} and format
28550 as eight rows of four columns. Include a string encoding with @samp{x}
28551 used as the non-printable character.
28555 4-data-read-memory bytes+16 x 1 8 4 x
28556 4^done,addr="0x000013a0",nr-bytes="32",total-bytes="32",
28557 next-row="0x000013c0",prev-row="0x0000139c",
28558 next-page="0x000013c0",prev-page="0x00001380",memory=[
28559 @{addr="0x000013a0",data=["0x10","0x11","0x12","0x13"],ascii="xxxx"@},
28560 @{addr="0x000013a4",data=["0x14","0x15","0x16","0x17"],ascii="xxxx"@},
28561 @{addr="0x000013a8",data=["0x18","0x19","0x1a","0x1b"],ascii="xxxx"@},
28562 @{addr="0x000013ac",data=["0x1c","0x1d","0x1e","0x1f"],ascii="xxxx"@},
28563 @{addr="0x000013b0",data=["0x20","0x21","0x22","0x23"],ascii=" !\"#"@},
28564 @{addr="0x000013b4",data=["0x24","0x25","0x26","0x27"],ascii="$%&'"@},
28565 @{addr="0x000013b8",data=["0x28","0x29","0x2a","0x2b"],ascii="()*+"@},
28566 @{addr="0x000013bc",data=["0x2c","0x2d","0x2e","0x2f"],ascii=",-./"@}]
28570 @subheading The @code{-data-read-memory-bytes} Command
28571 @findex -data-read-memory-bytes
28573 @subsubheading Synopsis
28576 -data-read-memory-bytes [ -o @var{byte-offset} ]
28577 @var{address} @var{count}
28584 @item @var{address}
28585 An expression specifying the address of the first memory word to be
28586 read. Complex expressions containing embedded white space should be
28587 quoted using the C convention.
28590 The number of bytes to read. This should be an integer literal.
28592 @item @var{byte-offset}
28593 The offsets in bytes relative to @var{address} at which to start
28594 reading. This should be an integer literal. This option is provided
28595 so that a frontend is not required to first evaluate address and then
28596 perform address arithmetics itself.
28600 This command attempts to read all accessible memory regions in the
28601 specified range. First, all regions marked as unreadable in the memory
28602 map (if one is defined) will be skipped. @xref{Memory Region
28603 Attributes}. Second, @value{GDBN} will attempt to read the remaining
28604 regions. For each one, if reading full region results in an errors,
28605 @value{GDBN} will try to read a subset of the region.
28607 In general, every single byte in the region may be readable or not,
28608 and the only way to read every readable byte is to try a read at
28609 every address, which is not practical. Therefore, @value{GDBN} will
28610 attempt to read all accessible bytes at either beginning or the end
28611 of the region, using a binary division scheme. This heuristic works
28612 well for reading accross a memory map boundary. Note that if a region
28613 has a readable range that is neither at the beginning or the end,
28614 @value{GDBN} will not read it.
28616 The result record (@pxref{GDB/MI Result Records}) that is output of
28617 the command includes a field named @samp{memory} whose content is a
28618 list of tuples. Each tuple represent a successfully read memory block
28619 and has the following fields:
28623 The start address of the memory block, as hexadecimal literal.
28626 The end address of the memory block, as hexadecimal literal.
28629 The offset of the memory block, as hexadecimal literal, relative to
28630 the start address passed to @code{-data-read-memory-bytes}.
28633 The contents of the memory block, in hex.
28639 @subsubheading @value{GDBN} Command
28641 The corresponding @value{GDBN} command is @samp{x}.
28643 @subsubheading Example
28647 -data-read-memory-bytes &a 10
28648 ^done,memory=[@{begin="0xbffff154",offset="0x00000000",
28650 contents="01000000020000000300"@}]
28655 @subheading The @code{-data-write-memory-bytes} Command
28656 @findex -data-write-memory-bytes
28658 @subsubheading Synopsis
28661 -data-write-memory-bytes @var{address} @var{contents}
28668 @item @var{address}
28669 An expression specifying the address of the first memory word to be
28670 read. Complex expressions containing embedded white space should be
28671 quoted using the C convention.
28673 @item @var{contents}
28674 The hex-encoded bytes to write.
28678 @subsubheading @value{GDBN} Command
28680 There's no corresponding @value{GDBN} command.
28682 @subsubheading Example
28686 -data-write-memory-bytes &a "aabbccdd"
28692 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28693 @node GDB/MI Tracepoint Commands
28694 @section @sc{gdb/mi} Tracepoint Commands
28696 The commands defined in this section implement MI support for
28697 tracepoints. For detailed introduction, see @ref{Tracepoints}.
28699 @subheading The @code{-trace-find} Command
28700 @findex -trace-find
28702 @subsubheading Synopsis
28705 -trace-find @var{mode} [@var{parameters}@dots{}]
28708 Find a trace frame using criteria defined by @var{mode} and
28709 @var{parameters}. The following table lists permissible
28710 modes and their parameters. For details of operation, see @ref{tfind}.
28715 No parameters are required. Stops examining trace frames.
28718 An integer is required as parameter. Selects tracepoint frame with
28721 @item tracepoint-number
28722 An integer is required as parameter. Finds next
28723 trace frame that corresponds to tracepoint with the specified number.
28726 An address is required as parameter. Finds
28727 next trace frame that corresponds to any tracepoint at the specified
28730 @item pc-inside-range
28731 Two addresses are required as parameters. Finds next trace
28732 frame that corresponds to a tracepoint at an address inside the
28733 specified range. Both bounds are considered to be inside the range.
28735 @item pc-outside-range
28736 Two addresses are required as parameters. Finds
28737 next trace frame that corresponds to a tracepoint at an address outside
28738 the specified range. Both bounds are considered to be inside the range.
28741 Line specification is required as parameter. @xref{Specify Location}.
28742 Finds next trace frame that corresponds to a tracepoint at
28743 the specified location.
28747 If @samp{none} was passed as @var{mode}, the response does not
28748 have fields. Otherwise, the response may have the following fields:
28752 This field has either @samp{0} or @samp{1} as the value, depending
28753 on whether a matching tracepoint was found.
28756 The index of the found traceframe. This field is present iff
28757 the @samp{found} field has value of @samp{1}.
28760 The index of the found tracepoint. This field is present iff
28761 the @samp{found} field has value of @samp{1}.
28764 The information about the frame corresponding to the found trace
28765 frame. This field is present only if a trace frame was found.
28766 @xref{GDB/MI Frame Information}, for description of this field.
28770 @subsubheading @value{GDBN} Command
28772 The corresponding @value{GDBN} command is @samp{tfind}.
28774 @subheading -trace-define-variable
28775 @findex -trace-define-variable
28777 @subsubheading Synopsis
28780 -trace-define-variable @var{name} [ @var{value} ]
28783 Create trace variable @var{name} if it does not exist. If
28784 @var{value} is specified, sets the initial value of the specified
28785 trace variable to that value. Note that the @var{name} should start
28786 with the @samp{$} character.
28788 @subsubheading @value{GDBN} Command
28790 The corresponding @value{GDBN} command is @samp{tvariable}.
28792 @subheading -trace-list-variables
28793 @findex -trace-list-variables
28795 @subsubheading Synopsis
28798 -trace-list-variables
28801 Return a table of all defined trace variables. Each element of the
28802 table has the following fields:
28806 The name of the trace variable. This field is always present.
28809 The initial value. This is a 64-bit signed integer. This
28810 field is always present.
28813 The value the trace variable has at the moment. This is a 64-bit
28814 signed integer. This field is absent iff current value is
28815 not defined, for example if the trace was never run, or is
28820 @subsubheading @value{GDBN} Command
28822 The corresponding @value{GDBN} command is @samp{tvariables}.
28824 @subsubheading Example
28828 -trace-list-variables
28829 ^done,trace-variables=@{nr_rows="1",nr_cols="3",
28830 hdr=[@{width="15",alignment="-1",col_name="name",colhdr="Name"@},
28831 @{width="11",alignment="-1",col_name="initial",colhdr="Initial"@},
28832 @{width="11",alignment="-1",col_name="current",colhdr="Current"@}],
28833 body=[variable=@{name="$trace_timestamp",initial="0"@}
28834 variable=@{name="$foo",initial="10",current="15"@}]@}
28838 @subheading -trace-save
28839 @findex -trace-save
28841 @subsubheading Synopsis
28844 -trace-save [-r ] @var{filename}
28847 Saves the collected trace data to @var{filename}. Without the
28848 @samp{-r} option, the data is downloaded from the target and saved
28849 in a local file. With the @samp{-r} option the target is asked
28850 to perform the save.
28852 @subsubheading @value{GDBN} Command
28854 The corresponding @value{GDBN} command is @samp{tsave}.
28857 @subheading -trace-start
28858 @findex -trace-start
28860 @subsubheading Synopsis
28866 Starts a tracing experiments. The result of this command does not
28869 @subsubheading @value{GDBN} Command
28871 The corresponding @value{GDBN} command is @samp{tstart}.
28873 @subheading -trace-status
28874 @findex -trace-status
28876 @subsubheading Synopsis
28882 Obtains the status of a tracing experiment. The result may include
28883 the following fields:
28888 May have a value of either @samp{0}, when no tracing operations are
28889 supported, @samp{1}, when all tracing operations are supported, or
28890 @samp{file} when examining trace file. In the latter case, examining
28891 of trace frame is possible but new tracing experiement cannot be
28892 started. This field is always present.
28895 May have a value of either @samp{0} or @samp{1} depending on whether
28896 tracing experiement is in progress on target. This field is present
28897 if @samp{supported} field is not @samp{0}.
28900 Report the reason why the tracing was stopped last time. This field
28901 may be absent iff tracing was never stopped on target yet. The
28902 value of @samp{request} means the tracing was stopped as result of
28903 the @code{-trace-stop} command. The value of @samp{overflow} means
28904 the tracing buffer is full. The value of @samp{disconnection} means
28905 tracing was automatically stopped when @value{GDBN} has disconnected.
28906 The value of @samp{passcount} means tracing was stopped when a
28907 tracepoint was passed a maximal number of times for that tracepoint.
28908 This field is present if @samp{supported} field is not @samp{0}.
28910 @item stopping-tracepoint
28911 The number of tracepoint whose passcount as exceeded. This field is
28912 present iff the @samp{stop-reason} field has the value of
28916 @itemx frames-created
28917 The @samp{frames} field is a count of the total number of trace frames
28918 in the trace buffer, while @samp{frames-created} is the total created
28919 during the run, including ones that were discarded, such as when a
28920 circular trace buffer filled up. Both fields are optional.
28924 These fields tell the current size of the tracing buffer and the
28925 remaining space. These fields are optional.
28928 The value of the circular trace buffer flag. @code{1} means that the
28929 trace buffer is circular and old trace frames will be discarded if
28930 necessary to make room, @code{0} means that the trace buffer is linear
28934 The value of the disconnected tracing flag. @code{1} means that
28935 tracing will continue after @value{GDBN} disconnects, @code{0} means
28936 that the trace run will stop.
28940 @subsubheading @value{GDBN} Command
28942 The corresponding @value{GDBN} command is @samp{tstatus}.
28944 @subheading -trace-stop
28945 @findex -trace-stop
28947 @subsubheading Synopsis
28953 Stops a tracing experiment. The result of this command has the same
28954 fields as @code{-trace-status}, except that the @samp{supported} and
28955 @samp{running} fields are not output.
28957 @subsubheading @value{GDBN} Command
28959 The corresponding @value{GDBN} command is @samp{tstop}.
28962 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28963 @node GDB/MI Symbol Query
28964 @section @sc{gdb/mi} Symbol Query Commands
28968 @subheading The @code{-symbol-info-address} Command
28969 @findex -symbol-info-address
28971 @subsubheading Synopsis
28974 -symbol-info-address @var{symbol}
28977 Describe where @var{symbol} is stored.
28979 @subsubheading @value{GDBN} Command
28981 The corresponding @value{GDBN} command is @samp{info address}.
28983 @subsubheading Example
28987 @subheading The @code{-symbol-info-file} Command
28988 @findex -symbol-info-file
28990 @subsubheading Synopsis
28996 Show the file for the symbol.
28998 @subsubheading @value{GDBN} Command
29000 There's no equivalent @value{GDBN} command. @code{gdbtk} has
29001 @samp{gdb_find_file}.
29003 @subsubheading Example
29007 @subheading The @code{-symbol-info-function} Command
29008 @findex -symbol-info-function
29010 @subsubheading Synopsis
29013 -symbol-info-function
29016 Show which function the symbol lives in.
29018 @subsubheading @value{GDBN} Command
29020 @samp{gdb_get_function} in @code{gdbtk}.
29022 @subsubheading Example
29026 @subheading The @code{-symbol-info-line} Command
29027 @findex -symbol-info-line
29029 @subsubheading Synopsis
29035 Show the core addresses of the code for a source line.
29037 @subsubheading @value{GDBN} Command
29039 The corresponding @value{GDBN} command is @samp{info line}.
29040 @code{gdbtk} has the @samp{gdb_get_line} and @samp{gdb_get_file} commands.
29042 @subsubheading Example
29046 @subheading The @code{-symbol-info-symbol} Command
29047 @findex -symbol-info-symbol
29049 @subsubheading Synopsis
29052 -symbol-info-symbol @var{addr}
29055 Describe what symbol is at location @var{addr}.
29057 @subsubheading @value{GDBN} Command
29059 The corresponding @value{GDBN} command is @samp{info symbol}.
29061 @subsubheading Example
29065 @subheading The @code{-symbol-list-functions} Command
29066 @findex -symbol-list-functions
29068 @subsubheading Synopsis
29071 -symbol-list-functions
29074 List the functions in the executable.
29076 @subsubheading @value{GDBN} Command
29078 @samp{info functions} in @value{GDBN}, @samp{gdb_listfunc} and
29079 @samp{gdb_search} in @code{gdbtk}.
29081 @subsubheading Example
29086 @subheading The @code{-symbol-list-lines} Command
29087 @findex -symbol-list-lines
29089 @subsubheading Synopsis
29092 -symbol-list-lines @var{filename}
29095 Print the list of lines that contain code and their associated program
29096 addresses for the given source filename. The entries are sorted in
29097 ascending PC order.
29099 @subsubheading @value{GDBN} Command
29101 There is no corresponding @value{GDBN} command.
29103 @subsubheading Example
29106 -symbol-list-lines basics.c
29107 ^done,lines=[@{pc="0x08048554",line="7"@},@{pc="0x0804855a",line="8"@}]
29113 @subheading The @code{-symbol-list-types} Command
29114 @findex -symbol-list-types
29116 @subsubheading Synopsis
29122 List all the type names.
29124 @subsubheading @value{GDBN} Command
29126 The corresponding commands are @samp{info types} in @value{GDBN},
29127 @samp{gdb_search} in @code{gdbtk}.
29129 @subsubheading Example
29133 @subheading The @code{-symbol-list-variables} Command
29134 @findex -symbol-list-variables
29136 @subsubheading Synopsis
29139 -symbol-list-variables
29142 List all the global and static variable names.
29144 @subsubheading @value{GDBN} Command
29146 @samp{info variables} in @value{GDBN}, @samp{gdb_search} in @code{gdbtk}.
29148 @subsubheading Example
29152 @subheading The @code{-symbol-locate} Command
29153 @findex -symbol-locate
29155 @subsubheading Synopsis
29161 @subsubheading @value{GDBN} Command
29163 @samp{gdb_loc} in @code{gdbtk}.
29165 @subsubheading Example
29169 @subheading The @code{-symbol-type} Command
29170 @findex -symbol-type
29172 @subsubheading Synopsis
29175 -symbol-type @var{variable}
29178 Show type of @var{variable}.
29180 @subsubheading @value{GDBN} Command
29182 The corresponding @value{GDBN} command is @samp{ptype}, @code{gdbtk} has
29183 @samp{gdb_obj_variable}.
29185 @subsubheading Example
29190 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
29191 @node GDB/MI File Commands
29192 @section @sc{gdb/mi} File Commands
29194 This section describes the GDB/MI commands to specify executable file names
29195 and to read in and obtain symbol table information.
29197 @subheading The @code{-file-exec-and-symbols} Command
29198 @findex -file-exec-and-symbols
29200 @subsubheading Synopsis
29203 -file-exec-and-symbols @var{file}
29206 Specify the executable file to be debugged. This file is the one from
29207 which the symbol table is also read. If no file is specified, the
29208 command clears the executable and symbol information. If breakpoints
29209 are set when using this command with no arguments, @value{GDBN} will produce
29210 error messages. Otherwise, no output is produced, except a completion
29213 @subsubheading @value{GDBN} Command
29215 The corresponding @value{GDBN} command is @samp{file}.
29217 @subsubheading Example
29221 -file-exec-and-symbols /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
29227 @subheading The @code{-file-exec-file} Command
29228 @findex -file-exec-file
29230 @subsubheading Synopsis
29233 -file-exec-file @var{file}
29236 Specify the executable file to be debugged. Unlike
29237 @samp{-file-exec-and-symbols}, the symbol table is @emph{not} read
29238 from this file. If used without argument, @value{GDBN} clears the information
29239 about the executable file. No output is produced, except a completion
29242 @subsubheading @value{GDBN} Command
29244 The corresponding @value{GDBN} command is @samp{exec-file}.
29246 @subsubheading Example
29250 -file-exec-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
29257 @subheading The @code{-file-list-exec-sections} Command
29258 @findex -file-list-exec-sections
29260 @subsubheading Synopsis
29263 -file-list-exec-sections
29266 List the sections of the current executable file.
29268 @subsubheading @value{GDBN} Command
29270 The @value{GDBN} command @samp{info file} shows, among the rest, the same
29271 information as this command. @code{gdbtk} has a corresponding command
29272 @samp{gdb_load_info}.
29274 @subsubheading Example
29279 @subheading The @code{-file-list-exec-source-file} Command
29280 @findex -file-list-exec-source-file
29282 @subsubheading Synopsis
29285 -file-list-exec-source-file
29288 List the line number, the current source file, and the absolute path
29289 to the current source file for the current executable. The macro
29290 information field has a value of @samp{1} or @samp{0} depending on
29291 whether or not the file includes preprocessor macro information.
29293 @subsubheading @value{GDBN} Command
29295 The @value{GDBN} equivalent is @samp{info source}
29297 @subsubheading Example
29301 123-file-list-exec-source-file
29302 123^done,line="1",file="foo.c",fullname="/home/bar/foo.c,macro-info="1"
29307 @subheading The @code{-file-list-exec-source-files} Command
29308 @findex -file-list-exec-source-files
29310 @subsubheading Synopsis
29313 -file-list-exec-source-files
29316 List the source files for the current executable.
29318 It will always output the filename, but only when @value{GDBN} can find
29319 the absolute file name of a source file, will it output the fullname.
29321 @subsubheading @value{GDBN} Command
29323 The @value{GDBN} equivalent is @samp{info sources}.
29324 @code{gdbtk} has an analogous command @samp{gdb_listfiles}.
29326 @subsubheading Example
29329 -file-list-exec-source-files
29331 @{file=foo.c,fullname=/home/foo.c@},
29332 @{file=/home/bar.c,fullname=/home/bar.c@},
29333 @{file=gdb_could_not_find_fullpath.c@}]
29338 @subheading The @code{-file-list-shared-libraries} Command
29339 @findex -file-list-shared-libraries
29341 @subsubheading Synopsis
29344 -file-list-shared-libraries
29347 List the shared libraries in the program.
29349 @subsubheading @value{GDBN} Command
29351 The corresponding @value{GDBN} command is @samp{info shared}.
29353 @subsubheading Example
29357 @subheading The @code{-file-list-symbol-files} Command
29358 @findex -file-list-symbol-files
29360 @subsubheading Synopsis
29363 -file-list-symbol-files
29368 @subsubheading @value{GDBN} Command
29370 The corresponding @value{GDBN} command is @samp{info file} (part of it).
29372 @subsubheading Example
29377 @subheading The @code{-file-symbol-file} Command
29378 @findex -file-symbol-file
29380 @subsubheading Synopsis
29383 -file-symbol-file @var{file}
29386 Read symbol table info from the specified @var{file} argument. When
29387 used without arguments, clears @value{GDBN}'s symbol table info. No output is
29388 produced, except for a completion notification.
29390 @subsubheading @value{GDBN} Command
29392 The corresponding @value{GDBN} command is @samp{symbol-file}.
29394 @subsubheading Example
29398 -file-symbol-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
29404 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
29405 @node GDB/MI Memory Overlay Commands
29406 @section @sc{gdb/mi} Memory Overlay Commands
29408 The memory overlay commands are not implemented.
29410 @c @subheading -overlay-auto
29412 @c @subheading -overlay-list-mapping-state
29414 @c @subheading -overlay-list-overlays
29416 @c @subheading -overlay-map
29418 @c @subheading -overlay-off
29420 @c @subheading -overlay-on
29422 @c @subheading -overlay-unmap
29424 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
29425 @node GDB/MI Signal Handling Commands
29426 @section @sc{gdb/mi} Signal Handling Commands
29428 Signal handling commands are not implemented.
29430 @c @subheading -signal-handle
29432 @c @subheading -signal-list-handle-actions
29434 @c @subheading -signal-list-signal-types
29438 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
29439 @node GDB/MI Target Manipulation
29440 @section @sc{gdb/mi} Target Manipulation Commands
29443 @subheading The @code{-target-attach} Command
29444 @findex -target-attach
29446 @subsubheading Synopsis
29449 -target-attach @var{pid} | @var{gid} | @var{file}
29452 Attach to a process @var{pid} or a file @var{file} outside of
29453 @value{GDBN}, or a thread group @var{gid}. If attaching to a thread
29454 group, the id previously returned by
29455 @samp{-list-thread-groups --available} must be used.
29457 @subsubheading @value{GDBN} Command
29459 The corresponding @value{GDBN} command is @samp{attach}.
29461 @subsubheading Example
29465 =thread-created,id="1"
29466 *stopped,thread-id="1",frame=@{addr="0xb7f7e410",func="bar",args=[]@}
29472 @subheading The @code{-target-compare-sections} Command
29473 @findex -target-compare-sections
29475 @subsubheading Synopsis
29478 -target-compare-sections [ @var{section} ]
29481 Compare data of section @var{section} on target to the exec file.
29482 Without the argument, all sections are compared.
29484 @subsubheading @value{GDBN} Command
29486 The @value{GDBN} equivalent is @samp{compare-sections}.
29488 @subsubheading Example
29493 @subheading The @code{-target-detach} Command
29494 @findex -target-detach
29496 @subsubheading Synopsis
29499 -target-detach [ @var{pid} | @var{gid} ]
29502 Detach from the remote target which normally resumes its execution.
29503 If either @var{pid} or @var{gid} is specified, detaches from either
29504 the specified process, or specified thread group. There's no output.
29506 @subsubheading @value{GDBN} Command
29508 The corresponding @value{GDBN} command is @samp{detach}.
29510 @subsubheading Example
29520 @subheading The @code{-target-disconnect} Command
29521 @findex -target-disconnect
29523 @subsubheading Synopsis
29529 Disconnect from the remote target. There's no output and the target is
29530 generally not resumed.
29532 @subsubheading @value{GDBN} Command
29534 The corresponding @value{GDBN} command is @samp{disconnect}.
29536 @subsubheading Example
29546 @subheading The @code{-target-download} Command
29547 @findex -target-download
29549 @subsubheading Synopsis
29555 Loads the executable onto the remote target.
29556 It prints out an update message every half second, which includes the fields:
29560 The name of the section.
29562 The size of what has been sent so far for that section.
29564 The size of the section.
29566 The total size of what was sent so far (the current and the previous sections).
29568 The size of the overall executable to download.
29572 Each message is sent as status record (@pxref{GDB/MI Output Syntax, ,
29573 @sc{gdb/mi} Output Syntax}).
29575 In addition, it prints the name and size of the sections, as they are
29576 downloaded. These messages include the following fields:
29580 The name of the section.
29582 The size of the section.
29584 The size of the overall executable to download.
29588 At the end, a summary is printed.
29590 @subsubheading @value{GDBN} Command
29592 The corresponding @value{GDBN} command is @samp{load}.
29594 @subsubheading Example
29596 Note: each status message appears on a single line. Here the messages
29597 have been broken down so that they can fit onto a page.
29602 +download,@{section=".text",section-size="6668",total-size="9880"@}
29603 +download,@{section=".text",section-sent="512",section-size="6668",
29604 total-sent="512",total-size="9880"@}
29605 +download,@{section=".text",section-sent="1024",section-size="6668",
29606 total-sent="1024",total-size="9880"@}
29607 +download,@{section=".text",section-sent="1536",section-size="6668",
29608 total-sent="1536",total-size="9880"@}
29609 +download,@{section=".text",section-sent="2048",section-size="6668",
29610 total-sent="2048",total-size="9880"@}
29611 +download,@{section=".text",section-sent="2560",section-size="6668",
29612 total-sent="2560",total-size="9880"@}
29613 +download,@{section=".text",section-sent="3072",section-size="6668",
29614 total-sent="3072",total-size="9880"@}
29615 +download,@{section=".text",section-sent="3584",section-size="6668",
29616 total-sent="3584",total-size="9880"@}
29617 +download,@{section=".text",section-sent="4096",section-size="6668",
29618 total-sent="4096",total-size="9880"@}
29619 +download,@{section=".text",section-sent="4608",section-size="6668",
29620 total-sent="4608",total-size="9880"@}
29621 +download,@{section=".text",section-sent="5120",section-size="6668",
29622 total-sent="5120",total-size="9880"@}
29623 +download,@{section=".text",section-sent="5632",section-size="6668",
29624 total-sent="5632",total-size="9880"@}
29625 +download,@{section=".text",section-sent="6144",section-size="6668",
29626 total-sent="6144",total-size="9880"@}
29627 +download,@{section=".text",section-sent="6656",section-size="6668",
29628 total-sent="6656",total-size="9880"@}
29629 +download,@{section=".init",section-size="28",total-size="9880"@}
29630 +download,@{section=".fini",section-size="28",total-size="9880"@}
29631 +download,@{section=".data",section-size="3156",total-size="9880"@}
29632 +download,@{section=".data",section-sent="512",section-size="3156",
29633 total-sent="7236",total-size="9880"@}
29634 +download,@{section=".data",section-sent="1024",section-size="3156",
29635 total-sent="7748",total-size="9880"@}
29636 +download,@{section=".data",section-sent="1536",section-size="3156",
29637 total-sent="8260",total-size="9880"@}
29638 +download,@{section=".data",section-sent="2048",section-size="3156",
29639 total-sent="8772",total-size="9880"@}
29640 +download,@{section=".data",section-sent="2560",section-size="3156",
29641 total-sent="9284",total-size="9880"@}
29642 +download,@{section=".data",section-sent="3072",section-size="3156",
29643 total-sent="9796",total-size="9880"@}
29644 ^done,address="0x10004",load-size="9880",transfer-rate="6586",
29651 @subheading The @code{-target-exec-status} Command
29652 @findex -target-exec-status
29654 @subsubheading Synopsis
29657 -target-exec-status
29660 Provide information on the state of the target (whether it is running or
29661 not, for instance).
29663 @subsubheading @value{GDBN} Command
29665 There's no equivalent @value{GDBN} command.
29667 @subsubheading Example
29671 @subheading The @code{-target-list-available-targets} Command
29672 @findex -target-list-available-targets
29674 @subsubheading Synopsis
29677 -target-list-available-targets
29680 List the possible targets to connect to.
29682 @subsubheading @value{GDBN} Command
29684 The corresponding @value{GDBN} command is @samp{help target}.
29686 @subsubheading Example
29690 @subheading The @code{-target-list-current-targets} Command
29691 @findex -target-list-current-targets
29693 @subsubheading Synopsis
29696 -target-list-current-targets
29699 Describe the current target.
29701 @subsubheading @value{GDBN} Command
29703 The corresponding information is printed by @samp{info file} (among
29706 @subsubheading Example
29710 @subheading The @code{-target-list-parameters} Command
29711 @findex -target-list-parameters
29713 @subsubheading Synopsis
29716 -target-list-parameters
29722 @subsubheading @value{GDBN} Command
29726 @subsubheading Example
29730 @subheading The @code{-target-select} Command
29731 @findex -target-select
29733 @subsubheading Synopsis
29736 -target-select @var{type} @var{parameters @dots{}}
29739 Connect @value{GDBN} to the remote target. This command takes two args:
29743 The type of target, for instance @samp{remote}, etc.
29744 @item @var{parameters}
29745 Device names, host names and the like. @xref{Target Commands, ,
29746 Commands for Managing Targets}, for more details.
29749 The output is a connection notification, followed by the address at
29750 which the target program is, in the following form:
29753 ^connected,addr="@var{address}",func="@var{function name}",
29754 args=[@var{arg list}]
29757 @subsubheading @value{GDBN} Command
29759 The corresponding @value{GDBN} command is @samp{target}.
29761 @subsubheading Example
29765 -target-select remote /dev/ttya
29766 ^connected,addr="0xfe00a300",func="??",args=[]
29770 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
29771 @node GDB/MI File Transfer Commands
29772 @section @sc{gdb/mi} File Transfer Commands
29775 @subheading The @code{-target-file-put} Command
29776 @findex -target-file-put
29778 @subsubheading Synopsis
29781 -target-file-put @var{hostfile} @var{targetfile}
29784 Copy file @var{hostfile} from the host system (the machine running
29785 @value{GDBN}) to @var{targetfile} on the target system.
29787 @subsubheading @value{GDBN} Command
29789 The corresponding @value{GDBN} command is @samp{remote put}.
29791 @subsubheading Example
29795 -target-file-put localfile remotefile
29801 @subheading The @code{-target-file-get} Command
29802 @findex -target-file-get
29804 @subsubheading Synopsis
29807 -target-file-get @var{targetfile} @var{hostfile}
29810 Copy file @var{targetfile} from the target system to @var{hostfile}
29811 on the host system.
29813 @subsubheading @value{GDBN} Command
29815 The corresponding @value{GDBN} command is @samp{remote get}.
29817 @subsubheading Example
29821 -target-file-get remotefile localfile
29827 @subheading The @code{-target-file-delete} Command
29828 @findex -target-file-delete
29830 @subsubheading Synopsis
29833 -target-file-delete @var{targetfile}
29836 Delete @var{targetfile} from the target system.
29838 @subsubheading @value{GDBN} Command
29840 The corresponding @value{GDBN} command is @samp{remote delete}.
29842 @subsubheading Example
29846 -target-file-delete remotefile
29852 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
29853 @node GDB/MI Miscellaneous Commands
29854 @section Miscellaneous @sc{gdb/mi} Commands
29856 @c @subheading -gdb-complete
29858 @subheading The @code{-gdb-exit} Command
29861 @subsubheading Synopsis
29867 Exit @value{GDBN} immediately.
29869 @subsubheading @value{GDBN} Command
29871 Approximately corresponds to @samp{quit}.
29873 @subsubheading Example
29883 @subheading The @code{-exec-abort} Command
29884 @findex -exec-abort
29886 @subsubheading Synopsis
29892 Kill the inferior running program.
29894 @subsubheading @value{GDBN} Command
29896 The corresponding @value{GDBN} command is @samp{kill}.
29898 @subsubheading Example
29903 @subheading The @code{-gdb-set} Command
29906 @subsubheading Synopsis
29912 Set an internal @value{GDBN} variable.
29913 @c IS THIS A DOLLAR VARIABLE? OR SOMETHING LIKE ANNOTATE ?????
29915 @subsubheading @value{GDBN} Command
29917 The corresponding @value{GDBN} command is @samp{set}.
29919 @subsubheading Example
29929 @subheading The @code{-gdb-show} Command
29932 @subsubheading Synopsis
29938 Show the current value of a @value{GDBN} variable.
29940 @subsubheading @value{GDBN} Command
29942 The corresponding @value{GDBN} command is @samp{show}.
29944 @subsubheading Example
29953 @c @subheading -gdb-source
29956 @subheading The @code{-gdb-version} Command
29957 @findex -gdb-version
29959 @subsubheading Synopsis
29965 Show version information for @value{GDBN}. Used mostly in testing.
29967 @subsubheading @value{GDBN} Command
29969 The @value{GDBN} equivalent is @samp{show version}. @value{GDBN} by
29970 default shows this information when you start an interactive session.
29972 @subsubheading Example
29974 @c This example modifies the actual output from GDB to avoid overfull
29980 ~Copyright 2000 Free Software Foundation, Inc.
29981 ~GDB is free software, covered by the GNU General Public License, and
29982 ~you are welcome to change it and/or distribute copies of it under
29983 ~ certain conditions.
29984 ~Type "show copying" to see the conditions.
29985 ~There is absolutely no warranty for GDB. Type "show warranty" for
29987 ~This GDB was configured as
29988 "--host=sparc-sun-solaris2.5.1 --target=ppc-eabi".
29993 @subheading The @code{-list-features} Command
29994 @findex -list-features
29996 Returns a list of particular features of the MI protocol that
29997 this version of gdb implements. A feature can be a command,
29998 or a new field in an output of some command, or even an
29999 important bugfix. While a frontend can sometimes detect presence
30000 of a feature at runtime, it is easier to perform detection at debugger
30003 The command returns a list of strings, with each string naming an
30004 available feature. Each returned string is just a name, it does not
30005 have any internal structure. The list of possible feature names
30011 (gdb) -list-features
30012 ^done,result=["feature1","feature2"]
30015 The current list of features is:
30018 @item frozen-varobjs
30019 Indicates presence of the @code{-var-set-frozen} command, as well
30020 as possible presense of the @code{frozen} field in the output
30021 of @code{-varobj-create}.
30022 @item pending-breakpoints
30023 Indicates presence of the @option{-f} option to the @code{-break-insert} command.
30025 Indicates presence of Python scripting support, Python-based
30026 pretty-printing commands, and possible presence of the
30027 @samp{display_hint} field in the output of @code{-var-list-children}
30029 Indicates presence of the @code{-thread-info} command.
30030 @item data-read-memory-bytes
30031 Indicates presense of the @code{-data-read-memory-bytes} and the
30032 @code{-data-write-memory-bytes} commands.
30036 @subheading The @code{-list-target-features} Command
30037 @findex -list-target-features
30039 Returns a list of particular features that are supported by the
30040 target. Those features affect the permitted MI commands, but
30041 unlike the features reported by the @code{-list-features} command, the
30042 features depend on which target GDB is using at the moment. Whenever
30043 a target can change, due to commands such as @code{-target-select},
30044 @code{-target-attach} or @code{-exec-run}, the list of target features
30045 may change, and the frontend should obtain it again.
30049 (gdb) -list-features
30050 ^done,result=["async"]
30053 The current list of features is:
30057 Indicates that the target is capable of asynchronous command
30058 execution, which means that @value{GDBN} will accept further commands
30059 while the target is running.
30062 Indicates that the target is capable of reverse execution.
30063 @xref{Reverse Execution}, for more information.
30067 @subheading The @code{-list-thread-groups} Command
30068 @findex -list-thread-groups
30070 @subheading Synopsis
30073 -list-thread-groups [ --available ] [ --recurse 1 ] [ @var{group} ... ]
30076 Lists thread groups (@pxref{Thread groups}). When a single thread
30077 group is passed as the argument, lists the children of that group.
30078 When several thread group are passed, lists information about those
30079 thread groups. Without any parameters, lists information about all
30080 top-level thread groups.
30082 Normally, thread groups that are being debugged are reported.
30083 With the @samp{--available} option, @value{GDBN} reports thread groups
30084 available on the target.
30086 The output of this command may have either a @samp{threads} result or
30087 a @samp{groups} result. The @samp{thread} result has a list of tuples
30088 as value, with each tuple describing a thread (@pxref{GDB/MI Thread
30089 Information}). The @samp{groups} result has a list of tuples as value,
30090 each tuple describing a thread group. If top-level groups are
30091 requested (that is, no parameter is passed), or when several groups
30092 are passed, the output always has a @samp{groups} result. The format
30093 of the @samp{group} result is described below.
30095 To reduce the number of roundtrips it's possible to list thread groups
30096 together with their children, by passing the @samp{--recurse} option
30097 and the recursion depth. Presently, only recursion depth of 1 is
30098 permitted. If this option is present, then every reported thread group
30099 will also include its children, either as @samp{group} or
30100 @samp{threads} field.
30102 In general, any combination of option and parameters is permitted, with
30103 the following caveats:
30107 When a single thread group is passed, the output will typically
30108 be the @samp{threads} result. Because threads may not contain
30109 anything, the @samp{recurse} option will be ignored.
30112 When the @samp{--available} option is passed, limited information may
30113 be available. In particular, the list of threads of a process might
30114 be inaccessible. Further, specifying specific thread groups might
30115 not give any performance advantage over listing all thread groups.
30116 The frontend should assume that @samp{-list-thread-groups --available}
30117 is always an expensive operation and cache the results.
30121 The @samp{groups} result is a list of tuples, where each tuple may
30122 have the following fields:
30126 Identifier of the thread group. This field is always present.
30127 The identifier is an opaque string; frontends should not try to
30128 convert it to an integer, even though it might look like one.
30131 The type of the thread group. At present, only @samp{process} is a
30135 The target-specific process identifier. This field is only present
30136 for thread groups of type @samp{process} and only if the process exists.
30139 The number of children this thread group has. This field may be
30140 absent for an available thread group.
30143 This field has a list of tuples as value, each tuple describing a
30144 thread. It may be present if the @samp{--recurse} option is
30145 specified, and it's actually possible to obtain the threads.
30148 This field is a list of integers, each identifying a core that one
30149 thread of the group is running on. This field may be absent if
30150 such information is not available.
30153 The name of the executable file that corresponds to this thread group.
30154 The field is only present for thread groups of type @samp{process},
30155 and only if there is a corresponding executable file.
30159 @subheading Example
30163 -list-thread-groups
30164 ^done,groups=[@{id="17",type="process",pid="yyy",num_children="2"@}]
30165 -list-thread-groups 17
30166 ^done,threads=[@{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
30167 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",args=[]@},state="running"@},
30168 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
30169 frame=@{level="0",addr="0x0804891f",func="foo",args=[@{name="i",value="10"@}],
30170 file="/tmp/a.c",fullname="/tmp/a.c",line="158"@},state="running"@}]]
30171 -list-thread-groups --available
30172 ^done,groups=[@{id="17",type="process",pid="yyy",num_children="2",cores=[1,2]@}]
30173 -list-thread-groups --available --recurse 1
30174 ^done,groups=[@{id="17", types="process",pid="yyy",num_children="2",cores=[1,2],
30175 threads=[@{id="1",target-id="Thread 0xb7e14b90",cores=[1]@},
30176 @{id="2",target-id="Thread 0xb7e14b90",cores=[2]@}]@},..]
30177 -list-thread-groups --available --recurse 1 17 18
30178 ^done,groups=[@{id="17", types="process",pid="yyy",num_children="2",cores=[1,2],
30179 threads=[@{id="1",target-id="Thread 0xb7e14b90",cores=[1]@},
30180 @{id="2",target-id="Thread 0xb7e14b90",cores=[2]@}]@},...]
30184 @subheading The @code{-add-inferior} Command
30185 @findex -add-inferior
30187 @subheading Synopsis
30193 Creates a new inferior (@pxref{Inferiors and Programs}). The created
30194 inferior is not associated with any executable. Such association may
30195 be established with the @samp{-file-exec-and-symbols} command
30196 (@pxref{GDB/MI File Commands}). The command response has a single
30197 field, @samp{thread-group}, whose value is the identifier of the
30198 thread group corresponding to the new inferior.
30200 @subheading Example
30205 ^done,thread-group="i3"
30208 @subheading The @code{-interpreter-exec} Command
30209 @findex -interpreter-exec
30211 @subheading Synopsis
30214 -interpreter-exec @var{interpreter} @var{command}
30216 @anchor{-interpreter-exec}
30218 Execute the specified @var{command} in the given @var{interpreter}.
30220 @subheading @value{GDBN} Command
30222 The corresponding @value{GDBN} command is @samp{interpreter-exec}.
30224 @subheading Example
30228 -interpreter-exec console "break main"
30229 &"During symbol reading, couldn't parse type; debugger out of date?.\n"
30230 &"During symbol reading, bad structure-type format.\n"
30231 ~"Breakpoint 1 at 0x8074fc6: file ../../src/gdb/main.c, line 743.\n"
30236 @subheading The @code{-inferior-tty-set} Command
30237 @findex -inferior-tty-set
30239 @subheading Synopsis
30242 -inferior-tty-set /dev/pts/1
30245 Set terminal for future runs of the program being debugged.
30247 @subheading @value{GDBN} Command
30249 The corresponding @value{GDBN} command is @samp{set inferior-tty} /dev/pts/1.
30251 @subheading Example
30255 -inferior-tty-set /dev/pts/1
30260 @subheading The @code{-inferior-tty-show} Command
30261 @findex -inferior-tty-show
30263 @subheading Synopsis
30269 Show terminal for future runs of program being debugged.
30271 @subheading @value{GDBN} Command
30273 The corresponding @value{GDBN} command is @samp{show inferior-tty}.
30275 @subheading Example
30279 -inferior-tty-set /dev/pts/1
30283 ^done,inferior_tty_terminal="/dev/pts/1"
30287 @subheading The @code{-enable-timings} Command
30288 @findex -enable-timings
30290 @subheading Synopsis
30293 -enable-timings [yes | no]
30296 Toggle the printing of the wallclock, user and system times for an MI
30297 command as a field in its output. This command is to help frontend
30298 developers optimize the performance of their code. No argument is
30299 equivalent to @samp{yes}.
30301 @subheading @value{GDBN} Command
30305 @subheading Example
30313 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
30314 addr="0x080484ed",func="main",file="myprog.c",
30315 fullname="/home/nickrob/myprog.c",line="73",times="0"@},
30316 time=@{wallclock="0.05185",user="0.00800",system="0.00000"@}
30324 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
30325 frame=@{addr="0x080484ed",func="main",args=[@{name="argc",value="1"@},
30326 @{name="argv",value="0xbfb60364"@}],file="myprog.c",
30327 fullname="/home/nickrob/myprog.c",line="73"@}
30332 @chapter @value{GDBN} Annotations
30334 This chapter describes annotations in @value{GDBN}. Annotations were
30335 designed to interface @value{GDBN} to graphical user interfaces or other
30336 similar programs which want to interact with @value{GDBN} at a
30337 relatively high level.
30339 The annotation mechanism has largely been superseded by @sc{gdb/mi}
30343 This is Edition @value{EDITION}, @value{DATE}.
30347 * Annotations Overview:: What annotations are; the general syntax.
30348 * Server Prefix:: Issuing a command without affecting user state.
30349 * Prompting:: Annotations marking @value{GDBN}'s need for input.
30350 * Errors:: Annotations for error messages.
30351 * Invalidation:: Some annotations describe things now invalid.
30352 * Annotations for Running::
30353 Whether the program is running, how it stopped, etc.
30354 * Source Annotations:: Annotations describing source code.
30357 @node Annotations Overview
30358 @section What is an Annotation?
30359 @cindex annotations
30361 Annotations start with a newline character, two @samp{control-z}
30362 characters, and the name of the annotation. If there is no additional
30363 information associated with this annotation, the name of the annotation
30364 is followed immediately by a newline. If there is additional
30365 information, the name of the annotation is followed by a space, the
30366 additional information, and a newline. The additional information
30367 cannot contain newline characters.
30369 Any output not beginning with a newline and two @samp{control-z}
30370 characters denotes literal output from @value{GDBN}. Currently there is
30371 no need for @value{GDBN} to output a newline followed by two
30372 @samp{control-z} characters, but if there was such a need, the
30373 annotations could be extended with an @samp{escape} annotation which
30374 means those three characters as output.
30376 The annotation @var{level}, which is specified using the
30377 @option{--annotate} command line option (@pxref{Mode Options}), controls
30378 how much information @value{GDBN} prints together with its prompt,
30379 values of expressions, source lines, and other types of output. Level 0
30380 is for no annotations, level 1 is for use when @value{GDBN} is run as a
30381 subprocess of @sc{gnu} Emacs, level 3 is the maximum annotation suitable
30382 for programs that control @value{GDBN}, and level 2 annotations have
30383 been made obsolete (@pxref{Limitations, , Limitations of the Annotation
30384 Interface, annotate, GDB's Obsolete Annotations}).
30387 @kindex set annotate
30388 @item set annotate @var{level}
30389 The @value{GDBN} command @code{set annotate} sets the level of
30390 annotations to the specified @var{level}.
30392 @item show annotate
30393 @kindex show annotate
30394 Show the current annotation level.
30397 This chapter describes level 3 annotations.
30399 A simple example of starting up @value{GDBN} with annotations is:
30402 $ @kbd{gdb --annotate=3}
30404 Copyright 2003 Free Software Foundation, Inc.
30405 GDB is free software, covered by the GNU General Public License,
30406 and you are welcome to change it and/or distribute copies of it
30407 under certain conditions.
30408 Type "show copying" to see the conditions.
30409 There is absolutely no warranty for GDB. Type "show warranty"
30411 This GDB was configured as "i386-pc-linux-gnu"
30422 Here @samp{quit} is input to @value{GDBN}; the rest is output from
30423 @value{GDBN}. The three lines beginning @samp{^Z^Z} (where @samp{^Z}
30424 denotes a @samp{control-z} character) are annotations; the rest is
30425 output from @value{GDBN}.
30427 @node Server Prefix
30428 @section The Server Prefix
30429 @cindex server prefix
30431 If you prefix a command with @samp{server } then it will not affect
30432 the command history, nor will it affect @value{GDBN}'s notion of which
30433 command to repeat if @key{RET} is pressed on a line by itself. This
30434 means that commands can be run behind a user's back by a front-end in
30435 a transparent manner.
30437 The @code{server } prefix does not affect the recording of values into
30438 the value history; to print a value without recording it into the
30439 value history, use the @code{output} command instead of the
30440 @code{print} command.
30442 Using this prefix also disables confirmation requests
30443 (@pxref{confirmation requests}).
30446 @section Annotation for @value{GDBN} Input
30448 @cindex annotations for prompts
30449 When @value{GDBN} prompts for input, it annotates this fact so it is possible
30450 to know when to send output, when the output from a given command is
30453 Different kinds of input each have a different @dfn{input type}. Each
30454 input type has three annotations: a @code{pre-} annotation, which
30455 denotes the beginning of any prompt which is being output, a plain
30456 annotation, which denotes the end of the prompt, and then a @code{post-}
30457 annotation which denotes the end of any echo which may (or may not) be
30458 associated with the input. For example, the @code{prompt} input type
30459 features the following annotations:
30467 The input types are
30470 @findex pre-prompt annotation
30471 @findex prompt annotation
30472 @findex post-prompt annotation
30474 When @value{GDBN} is prompting for a command (the main @value{GDBN} prompt).
30476 @findex pre-commands annotation
30477 @findex commands annotation
30478 @findex post-commands annotation
30480 When @value{GDBN} prompts for a set of commands, like in the @code{commands}
30481 command. The annotations are repeated for each command which is input.
30483 @findex pre-overload-choice annotation
30484 @findex overload-choice annotation
30485 @findex post-overload-choice annotation
30486 @item overload-choice
30487 When @value{GDBN} wants the user to select between various overloaded functions.
30489 @findex pre-query annotation
30490 @findex query annotation
30491 @findex post-query annotation
30493 When @value{GDBN} wants the user to confirm a potentially dangerous operation.
30495 @findex pre-prompt-for-continue annotation
30496 @findex prompt-for-continue annotation
30497 @findex post-prompt-for-continue annotation
30498 @item prompt-for-continue
30499 When @value{GDBN} is asking the user to press return to continue. Note: Don't
30500 expect this to work well; instead use @code{set height 0} to disable
30501 prompting. This is because the counting of lines is buggy in the
30502 presence of annotations.
30507 @cindex annotations for errors, warnings and interrupts
30509 @findex quit annotation
30514 This annotation occurs right before @value{GDBN} responds to an interrupt.
30516 @findex error annotation
30521 This annotation occurs right before @value{GDBN} responds to an error.
30523 Quit and error annotations indicate that any annotations which @value{GDBN} was
30524 in the middle of may end abruptly. For example, if a
30525 @code{value-history-begin} annotation is followed by a @code{error}, one
30526 cannot expect to receive the matching @code{value-history-end}. One
30527 cannot expect not to receive it either, however; an error annotation
30528 does not necessarily mean that @value{GDBN} is immediately returning all the way
30531 @findex error-begin annotation
30532 A quit or error annotation may be preceded by
30538 Any output between that and the quit or error annotation is the error
30541 Warning messages are not yet annotated.
30542 @c If we want to change that, need to fix warning(), type_error(),
30543 @c range_error(), and possibly other places.
30546 @section Invalidation Notices
30548 @cindex annotations for invalidation messages
30549 The following annotations say that certain pieces of state may have
30553 @findex frames-invalid annotation
30554 @item ^Z^Zframes-invalid
30556 The frames (for example, output from the @code{backtrace} command) may
30559 @findex breakpoints-invalid annotation
30560 @item ^Z^Zbreakpoints-invalid
30562 The breakpoints may have changed. For example, the user just added or
30563 deleted a breakpoint.
30566 @node Annotations for Running
30567 @section Running the Program
30568 @cindex annotations for running programs
30570 @findex starting annotation
30571 @findex stopping annotation
30572 When the program starts executing due to a @value{GDBN} command such as
30573 @code{step} or @code{continue},
30579 is output. When the program stops,
30585 is output. Before the @code{stopped} annotation, a variety of
30586 annotations describe how the program stopped.
30589 @findex exited annotation
30590 @item ^Z^Zexited @var{exit-status}
30591 The program exited, and @var{exit-status} is the exit status (zero for
30592 successful exit, otherwise nonzero).
30594 @findex signalled annotation
30595 @findex signal-name annotation
30596 @findex signal-name-end annotation
30597 @findex signal-string annotation
30598 @findex signal-string-end annotation
30599 @item ^Z^Zsignalled
30600 The program exited with a signal. After the @code{^Z^Zsignalled}, the
30601 annotation continues:
30607 ^Z^Zsignal-name-end
30611 ^Z^Zsignal-string-end
30616 where @var{name} is the name of the signal, such as @code{SIGILL} or
30617 @code{SIGSEGV}, and @var{string} is the explanation of the signal, such
30618 as @code{Illegal Instruction} or @code{Segmentation fault}.
30619 @var{intro-text}, @var{middle-text}, and @var{end-text} are for the
30620 user's benefit and have no particular format.
30622 @findex signal annotation
30624 The syntax of this annotation is just like @code{signalled}, but @value{GDBN} is
30625 just saying that the program received the signal, not that it was
30626 terminated with it.
30628 @findex breakpoint annotation
30629 @item ^Z^Zbreakpoint @var{number}
30630 The program hit breakpoint number @var{number}.
30632 @findex watchpoint annotation
30633 @item ^Z^Zwatchpoint @var{number}
30634 The program hit watchpoint number @var{number}.
30637 @node Source Annotations
30638 @section Displaying Source
30639 @cindex annotations for source display
30641 @findex source annotation
30642 The following annotation is used instead of displaying source code:
30645 ^Z^Zsource @var{filename}:@var{line}:@var{character}:@var{middle}:@var{addr}
30648 where @var{filename} is an absolute file name indicating which source
30649 file, @var{line} is the line number within that file (where 1 is the
30650 first line in the file), @var{character} is the character position
30651 within the file (where 0 is the first character in the file) (for most
30652 debug formats this will necessarily point to the beginning of a line),
30653 @var{middle} is @samp{middle} if @var{addr} is in the middle of the
30654 line, or @samp{beg} if @var{addr} is at the beginning of the line, and
30655 @var{addr} is the address in the target program associated with the
30656 source which is being displayed. @var{addr} is in the form @samp{0x}
30657 followed by one or more lowercase hex digits (note that this does not
30658 depend on the language).
30660 @node JIT Interface
30661 @chapter JIT Compilation Interface
30662 @cindex just-in-time compilation
30663 @cindex JIT compilation interface
30665 This chapter documents @value{GDBN}'s @dfn{just-in-time} (JIT) compilation
30666 interface. A JIT compiler is a program or library that generates native
30667 executable code at runtime and executes it, usually in order to achieve good
30668 performance while maintaining platform independence.
30670 Programs that use JIT compilation are normally difficult to debug because
30671 portions of their code are generated at runtime, instead of being loaded from
30672 object files, which is where @value{GDBN} normally finds the program's symbols
30673 and debug information. In order to debug programs that use JIT compilation,
30674 @value{GDBN} has an interface that allows the program to register in-memory
30675 symbol files with @value{GDBN} at runtime.
30677 If you are using @value{GDBN} to debug a program that uses this interface, then
30678 it should work transparently so long as you have not stripped the binary. If
30679 you are developing a JIT compiler, then the interface is documented in the rest
30680 of this chapter. At this time, the only known client of this interface is the
30683 Broadly speaking, the JIT interface mirrors the dynamic loader interface. The
30684 JIT compiler communicates with @value{GDBN} by writing data into a global
30685 variable and calling a fuction at a well-known symbol. When @value{GDBN}
30686 attaches, it reads a linked list of symbol files from the global variable to
30687 find existing code, and puts a breakpoint in the function so that it can find
30688 out about additional code.
30691 * Declarations:: Relevant C struct declarations
30692 * Registering Code:: Steps to register code
30693 * Unregistering Code:: Steps to unregister code
30697 @section JIT Declarations
30699 These are the relevant struct declarations that a C program should include to
30700 implement the interface:
30710 struct jit_code_entry
30712 struct jit_code_entry *next_entry;
30713 struct jit_code_entry *prev_entry;
30714 const char *symfile_addr;
30715 uint64_t symfile_size;
30718 struct jit_descriptor
30721 /* This type should be jit_actions_t, but we use uint32_t
30722 to be explicit about the bitwidth. */
30723 uint32_t action_flag;
30724 struct jit_code_entry *relevant_entry;
30725 struct jit_code_entry *first_entry;
30728 /* GDB puts a breakpoint in this function. */
30729 void __attribute__((noinline)) __jit_debug_register_code() @{ @};
30731 /* Make sure to specify the version statically, because the
30732 debugger may check the version before we can set it. */
30733 struct jit_descriptor __jit_debug_descriptor = @{ 1, 0, 0, 0 @};
30736 If the JIT is multi-threaded, then it is important that the JIT synchronize any
30737 modifications to this global data properly, which can easily be done by putting
30738 a global mutex around modifications to these structures.
30740 @node Registering Code
30741 @section Registering Code
30743 To register code with @value{GDBN}, the JIT should follow this protocol:
30747 Generate an object file in memory with symbols and other desired debug
30748 information. The file must include the virtual addresses of the sections.
30751 Create a code entry for the file, which gives the start and size of the symbol
30755 Add it to the linked list in the JIT descriptor.
30758 Point the relevant_entry field of the descriptor at the entry.
30761 Set @code{action_flag} to @code{JIT_REGISTER} and call
30762 @code{__jit_debug_register_code}.
30765 When @value{GDBN} is attached and the breakpoint fires, @value{GDBN} uses the
30766 @code{relevant_entry} pointer so it doesn't have to walk the list looking for
30767 new code. However, the linked list must still be maintained in order to allow
30768 @value{GDBN} to attach to a running process and still find the symbol files.
30770 @node Unregistering Code
30771 @section Unregistering Code
30773 If code is freed, then the JIT should use the following protocol:
30777 Remove the code entry corresponding to the code from the linked list.
30780 Point the @code{relevant_entry} field of the descriptor at the code entry.
30783 Set @code{action_flag} to @code{JIT_UNREGISTER} and call
30784 @code{__jit_debug_register_code}.
30787 If the JIT frees or recompiles code without unregistering it, then @value{GDBN}
30788 and the JIT will leak the memory used for the associated symbol files.
30791 @chapter Reporting Bugs in @value{GDBN}
30792 @cindex bugs in @value{GDBN}
30793 @cindex reporting bugs in @value{GDBN}
30795 Your bug reports play an essential role in making @value{GDBN} reliable.
30797 Reporting a bug may help you by bringing a solution to your problem, or it
30798 may not. But in any case the principal function of a bug report is to help
30799 the entire community by making the next version of @value{GDBN} work better. Bug
30800 reports are your contribution to the maintenance of @value{GDBN}.
30802 In order for a bug report to serve its purpose, you must include the
30803 information that enables us to fix the bug.
30806 * Bug Criteria:: Have you found a bug?
30807 * Bug Reporting:: How to report bugs
30811 @section Have You Found a Bug?
30812 @cindex bug criteria
30814 If you are not sure whether you have found a bug, here are some guidelines:
30817 @cindex fatal signal
30818 @cindex debugger crash
30819 @cindex crash of debugger
30821 If the debugger gets a fatal signal, for any input whatever, that is a
30822 @value{GDBN} bug. Reliable debuggers never crash.
30824 @cindex error on valid input
30826 If @value{GDBN} produces an error message for valid input, that is a
30827 bug. (Note that if you're cross debugging, the problem may also be
30828 somewhere in the connection to the target.)
30830 @cindex invalid input
30832 If @value{GDBN} does not produce an error message for invalid input,
30833 that is a bug. However, you should note that your idea of
30834 ``invalid input'' might be our idea of ``an extension'' or ``support
30835 for traditional practice''.
30838 If you are an experienced user of debugging tools, your suggestions
30839 for improvement of @value{GDBN} are welcome in any case.
30842 @node Bug Reporting
30843 @section How to Report Bugs
30844 @cindex bug reports
30845 @cindex @value{GDBN} bugs, reporting
30847 A number of companies and individuals offer support for @sc{gnu} products.
30848 If you obtained @value{GDBN} from a support organization, we recommend you
30849 contact that organization first.
30851 You can find contact information for many support companies and
30852 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
30854 @c should add a web page ref...
30857 @ifset BUGURL_DEFAULT
30858 In any event, we also recommend that you submit bug reports for
30859 @value{GDBN}. The preferred method is to submit them directly using
30860 @uref{http://www.gnu.org/software/gdb/bugs/, @value{GDBN}'s Bugs web
30861 page}. Alternatively, the @email{bug-gdb@@gnu.org, e-mail gateway} can
30864 @strong{Do not send bug reports to @samp{info-gdb}, or to
30865 @samp{help-gdb}, or to any newsgroups.} Most users of @value{GDBN} do
30866 not want to receive bug reports. Those that do have arranged to receive
30869 The mailing list @samp{bug-gdb} has a newsgroup @samp{gnu.gdb.bug} which
30870 serves as a repeater. The mailing list and the newsgroup carry exactly
30871 the same messages. Often people think of posting bug reports to the
30872 newsgroup instead of mailing them. This appears to work, but it has one
30873 problem which can be crucial: a newsgroup posting often lacks a mail
30874 path back to the sender. Thus, if we need to ask for more information,
30875 we may be unable to reach you. For this reason, it is better to send
30876 bug reports to the mailing list.
30878 @ifclear BUGURL_DEFAULT
30879 In any event, we also recommend that you submit bug reports for
30880 @value{GDBN} to @value{BUGURL}.
30884 The fundamental principle of reporting bugs usefully is this:
30885 @strong{report all the facts}. If you are not sure whether to state a
30886 fact or leave it out, state it!
30888 Often people omit facts because they think they know what causes the
30889 problem and assume that some details do not matter. Thus, you might
30890 assume that the name of the variable you use in an example does not matter.
30891 Well, probably it does not, but one cannot be sure. Perhaps the bug is a
30892 stray memory reference which happens to fetch from the location where that
30893 name is stored in memory; perhaps, if the name were different, the contents
30894 of that location would fool the debugger into doing the right thing despite
30895 the bug. Play it safe and give a specific, complete example. That is the
30896 easiest thing for you to do, and the most helpful.
30898 Keep in mind that the purpose of a bug report is to enable us to fix the
30899 bug. It may be that the bug has been reported previously, but neither
30900 you nor we can know that unless your bug report is complete and
30903 Sometimes people give a few sketchy facts and ask, ``Does this ring a
30904 bell?'' Those bug reports are useless, and we urge everyone to
30905 @emph{refuse to respond to them} except to chide the sender to report
30908 To enable us to fix the bug, you should include all these things:
30912 The version of @value{GDBN}. @value{GDBN} announces it if you start
30913 with no arguments; you can also print it at any time using @code{show
30916 Without this, we will not know whether there is any point in looking for
30917 the bug in the current version of @value{GDBN}.
30920 The type of machine you are using, and the operating system name and
30924 What compiler (and its version) was used to compile @value{GDBN}---e.g.@:
30925 ``@value{GCC}--2.8.1''.
30928 What compiler (and its version) was used to compile the program you are
30929 debugging---e.g.@: ``@value{GCC}--2.8.1'', or ``HP92453-01 A.10.32.03 HP
30930 C Compiler''. For @value{NGCC}, you can say @kbd{@value{GCC} --version}
30931 to get this information; for other compilers, see the documentation for
30935 The command arguments you gave the compiler to compile your example and
30936 observe the bug. For example, did you use @samp{-O}? To guarantee
30937 you will not omit something important, list them all. A copy of the
30938 Makefile (or the output from make) is sufficient.
30940 If we were to try to guess the arguments, we would probably guess wrong
30941 and then we might not encounter the bug.
30944 A complete input script, and all necessary source files, that will
30948 A description of what behavior you observe that you believe is
30949 incorrect. For example, ``It gets a fatal signal.''
30951 Of course, if the bug is that @value{GDBN} gets a fatal signal, then we
30952 will certainly notice it. But if the bug is incorrect output, we might
30953 not notice unless it is glaringly wrong. You might as well not give us
30954 a chance to make a mistake.
30956 Even if the problem you experience is a fatal signal, you should still
30957 say so explicitly. Suppose something strange is going on, such as, your
30958 copy of @value{GDBN} is out of synch, or you have encountered a bug in
30959 the C library on your system. (This has happened!) Your copy might
30960 crash and ours would not. If you told us to expect a crash, then when
30961 ours fails to crash, we would know that the bug was not happening for
30962 us. If you had not told us to expect a crash, then we would not be able
30963 to draw any conclusion from our observations.
30966 @cindex recording a session script
30967 To collect all this information, you can use a session recording program
30968 such as @command{script}, which is available on many Unix systems.
30969 Just run your @value{GDBN} session inside @command{script} and then
30970 include the @file{typescript} file with your bug report.
30972 Another way to record a @value{GDBN} session is to run @value{GDBN}
30973 inside Emacs and then save the entire buffer to a file.
30976 If you wish to suggest changes to the @value{GDBN} source, send us context
30977 diffs. If you even discuss something in the @value{GDBN} source, refer to
30978 it by context, not by line number.
30980 The line numbers in our development sources will not match those in your
30981 sources. Your line numbers would convey no useful information to us.
30985 Here are some things that are not necessary:
30989 A description of the envelope of the bug.
30991 Often people who encounter a bug spend a lot of time investigating
30992 which changes to the input file will make the bug go away and which
30993 changes will not affect it.
30995 This is often time consuming and not very useful, because the way we
30996 will find the bug is by running a single example under the debugger
30997 with breakpoints, not by pure deduction from a series of examples.
30998 We recommend that you save your time for something else.
31000 Of course, if you can find a simpler example to report @emph{instead}
31001 of the original one, that is a convenience for us. Errors in the
31002 output will be easier to spot, running under the debugger will take
31003 less time, and so on.
31005 However, simplification is not vital; if you do not want to do this,
31006 report the bug anyway and send us the entire test case you used.
31009 A patch for the bug.
31011 A patch for the bug does help us if it is a good one. But do not omit
31012 the necessary information, such as the test case, on the assumption that
31013 a patch is all we need. We might see problems with your patch and decide
31014 to fix the problem another way, or we might not understand it at all.
31016 Sometimes with a program as complicated as @value{GDBN} it is very hard to
31017 construct an example that will make the program follow a certain path
31018 through the code. If you do not send us the example, we will not be able
31019 to construct one, so we will not be able to verify that the bug is fixed.
31021 And if we cannot understand what bug you are trying to fix, or why your
31022 patch should be an improvement, we will not install it. A test case will
31023 help us to understand.
31026 A guess about what the bug is or what it depends on.
31028 Such guesses are usually wrong. Even we cannot guess right about such
31029 things without first using the debugger to find the facts.
31032 @c The readline documentation is distributed with the readline code
31033 @c and consists of the two following files:
31036 @c Use -I with makeinfo to point to the appropriate directory,
31037 @c environment var TEXINPUTS with TeX.
31038 @ifclear SYSTEM_READLINE
31039 @include rluser.texi
31040 @include hsuser.texi
31044 @appendix In Memoriam
31046 The @value{GDBN} project mourns the loss of the following long-time
31051 Fred was a long-standing contributor to @value{GDBN} (1991-2006), and
31052 to Free Software in general. Outside of @value{GDBN}, he was known in
31053 the Amiga world for his series of Fish Disks, and the GeekGadget project.
31055 @item Michael Snyder
31056 Michael was one of the Global Maintainers of the @value{GDBN} project,
31057 with contributions recorded as early as 1996, until 2011. In addition
31058 to his day to day participation, he was a large driving force behind
31059 adding Reverse Debugging to @value{GDBN}.
31062 Beyond their technical contributions to the project, they were also
31063 enjoyable members of the Free Software Community. We will miss them.
31065 @node Formatting Documentation
31066 @appendix Formatting Documentation
31068 @cindex @value{GDBN} reference card
31069 @cindex reference card
31070 The @value{GDBN} 4 release includes an already-formatted reference card, ready
31071 for printing with PostScript or Ghostscript, in the @file{gdb}
31072 subdirectory of the main source directory@footnote{In
31073 @file{gdb-@value{GDBVN}/gdb/refcard.ps} of the version @value{GDBVN}
31074 release.}. If you can use PostScript or Ghostscript with your printer,
31075 you can print the reference card immediately with @file{refcard.ps}.
31077 The release also includes the source for the reference card. You
31078 can format it, using @TeX{}, by typing:
31084 The @value{GDBN} reference card is designed to print in @dfn{landscape}
31085 mode on US ``letter'' size paper;
31086 that is, on a sheet 11 inches wide by 8.5 inches
31087 high. You will need to specify this form of printing as an option to
31088 your @sc{dvi} output program.
31090 @cindex documentation
31092 All the documentation for @value{GDBN} comes as part of the machine-readable
31093 distribution. The documentation is written in Texinfo format, which is
31094 a documentation system that uses a single source file to produce both
31095 on-line information and a printed manual. You can use one of the Info
31096 formatting commands to create the on-line version of the documentation
31097 and @TeX{} (or @code{texi2roff}) to typeset the printed version.
31099 @value{GDBN} includes an already formatted copy of the on-line Info
31100 version of this manual in the @file{gdb} subdirectory. The main Info
31101 file is @file{gdb-@value{GDBVN}/gdb/gdb.info}, and it refers to
31102 subordinate files matching @samp{gdb.info*} in the same directory. If
31103 necessary, you can print out these files, or read them with any editor;
31104 but they are easier to read using the @code{info} subsystem in @sc{gnu}
31105 Emacs or the standalone @code{info} program, available as part of the
31106 @sc{gnu} Texinfo distribution.
31108 If you want to format these Info files yourself, you need one of the
31109 Info formatting programs, such as @code{texinfo-format-buffer} or
31112 If you have @code{makeinfo} installed, and are in the top level
31113 @value{GDBN} source directory (@file{gdb-@value{GDBVN}}, in the case of
31114 version @value{GDBVN}), you can make the Info file by typing:
31121 If you want to typeset and print copies of this manual, you need @TeX{},
31122 a program to print its @sc{dvi} output files, and @file{texinfo.tex}, the
31123 Texinfo definitions file.
31125 @TeX{} is a typesetting program; it does not print files directly, but
31126 produces output files called @sc{dvi} files. To print a typeset
31127 document, you need a program to print @sc{dvi} files. If your system
31128 has @TeX{} installed, chances are it has such a program. The precise
31129 command to use depends on your system; @kbd{lpr -d} is common; another
31130 (for PostScript devices) is @kbd{dvips}. The @sc{dvi} print command may
31131 require a file name without any extension or a @samp{.dvi} extension.
31133 @TeX{} also requires a macro definitions file called
31134 @file{texinfo.tex}. This file tells @TeX{} how to typeset a document
31135 written in Texinfo format. On its own, @TeX{} cannot either read or
31136 typeset a Texinfo file. @file{texinfo.tex} is distributed with GDB
31137 and is located in the @file{gdb-@var{version-number}/texinfo}
31140 If you have @TeX{} and a @sc{dvi} printer program installed, you can
31141 typeset and print this manual. First switch to the @file{gdb}
31142 subdirectory of the main source directory (for example, to
31143 @file{gdb-@value{GDBVN}/gdb}) and type:
31149 Then give @file{gdb.dvi} to your @sc{dvi} printing program.
31151 @node Installing GDB
31152 @appendix Installing @value{GDBN}
31153 @cindex installation
31156 * Requirements:: Requirements for building @value{GDBN}
31157 * Running Configure:: Invoking the @value{GDBN} @file{configure} script
31158 * Separate Objdir:: Compiling @value{GDBN} in another directory
31159 * Config Names:: Specifying names for hosts and targets
31160 * Configure Options:: Summary of options for configure
31161 * System-wide configuration:: Having a system-wide init file
31165 @section Requirements for Building @value{GDBN}
31166 @cindex building @value{GDBN}, requirements for
31168 Building @value{GDBN} requires various tools and packages to be available.
31169 Other packages will be used only if they are found.
31171 @heading Tools/Packages Necessary for Building @value{GDBN}
31173 @item ISO C90 compiler
31174 @value{GDBN} is written in ISO C90. It should be buildable with any
31175 working C90 compiler, e.g.@: GCC.
31179 @heading Tools/Packages Optional for Building @value{GDBN}
31183 @value{GDBN} can use the Expat XML parsing library. This library may be
31184 included with your operating system distribution; if it is not, you
31185 can get the latest version from @url{http://expat.sourceforge.net}.
31186 The @file{configure} script will search for this library in several
31187 standard locations; if it is installed in an unusual path, you can
31188 use the @option{--with-libexpat-prefix} option to specify its location.
31194 Remote protocol memory maps (@pxref{Memory Map Format})
31196 Target descriptions (@pxref{Target Descriptions})
31198 Remote shared library lists (@pxref{Library List Format})
31200 MS-Windows shared libraries (@pxref{Shared Libraries})
31202 Traceframe info (@pxref{Traceframe Info Format})
31206 @cindex compressed debug sections
31207 @value{GDBN} will use the @samp{zlib} library, if available, to read
31208 compressed debug sections. Some linkers, such as GNU gold, are capable
31209 of producing binaries with compressed debug sections. If @value{GDBN}
31210 is compiled with @samp{zlib}, it will be able to read the debug
31211 information in such binaries.
31213 The @samp{zlib} library is likely included with your operating system
31214 distribution; if it is not, you can get the latest version from
31215 @url{http://zlib.net}.
31218 @value{GDBN}'s features related to character sets (@pxref{Character
31219 Sets}) require a functioning @code{iconv} implementation. If you are
31220 on a GNU system, then this is provided by the GNU C Library. Some
31221 other systems also provide a working @code{iconv}.
31223 If @value{GDBN} is using the @code{iconv} program which is installed
31224 in a non-standard place, you will need to tell @value{GDBN} where to find it.
31225 This is done with @option{--with-iconv-bin} which specifies the
31226 directory that contains the @code{iconv} program.
31228 On systems without @code{iconv}, you can install GNU Libiconv. If you
31229 have previously installed Libiconv, you can use the
31230 @option{--with-libiconv-prefix} option to configure.
31232 @value{GDBN}'s top-level @file{configure} and @file{Makefile} will
31233 arrange to build Libiconv if a directory named @file{libiconv} appears
31234 in the top-most source directory. If Libiconv is built this way, and
31235 if the operating system does not provide a suitable @code{iconv}
31236 implementation, then the just-built library will automatically be used
31237 by @value{GDBN}. One easy way to set this up is to download GNU
31238 Libiconv, unpack it, and then rename the directory holding the
31239 Libiconv source code to @samp{libiconv}.
31242 @node Running Configure
31243 @section Invoking the @value{GDBN} @file{configure} Script
31244 @cindex configuring @value{GDBN}
31245 @value{GDBN} comes with a @file{configure} script that automates the process
31246 of preparing @value{GDBN} for installation; you can then use @code{make} to
31247 build the @code{gdb} program.
31249 @c irrelevant in info file; it's as current as the code it lives with.
31250 @footnote{If you have a more recent version of @value{GDBN} than @value{GDBVN},
31251 look at the @file{README} file in the sources; we may have improved the
31252 installation procedures since publishing this manual.}
31255 The @value{GDBN} distribution includes all the source code you need for
31256 @value{GDBN} in a single directory, whose name is usually composed by
31257 appending the version number to @samp{gdb}.
31259 For example, the @value{GDBN} version @value{GDBVN} distribution is in the
31260 @file{gdb-@value{GDBVN}} directory. That directory contains:
31263 @item gdb-@value{GDBVN}/configure @r{(and supporting files)}
31264 script for configuring @value{GDBN} and all its supporting libraries
31266 @item gdb-@value{GDBVN}/gdb
31267 the source specific to @value{GDBN} itself
31269 @item gdb-@value{GDBVN}/bfd
31270 source for the Binary File Descriptor library
31272 @item gdb-@value{GDBVN}/include
31273 @sc{gnu} include files
31275 @item gdb-@value{GDBVN}/libiberty
31276 source for the @samp{-liberty} free software library
31278 @item gdb-@value{GDBVN}/opcodes
31279 source for the library of opcode tables and disassemblers
31281 @item gdb-@value{GDBVN}/readline
31282 source for the @sc{gnu} command-line interface
31284 @item gdb-@value{GDBVN}/glob
31285 source for the @sc{gnu} filename pattern-matching subroutine
31287 @item gdb-@value{GDBVN}/mmalloc
31288 source for the @sc{gnu} memory-mapped malloc package
31291 The simplest way to configure and build @value{GDBN} is to run @file{configure}
31292 from the @file{gdb-@var{version-number}} source directory, which in
31293 this example is the @file{gdb-@value{GDBVN}} directory.
31295 First switch to the @file{gdb-@var{version-number}} source directory
31296 if you are not already in it; then run @file{configure}. Pass the
31297 identifier for the platform on which @value{GDBN} will run as an
31303 cd gdb-@value{GDBVN}
31304 ./configure @var{host}
31309 where @var{host} is an identifier such as @samp{sun4} or
31310 @samp{decstation}, that identifies the platform where @value{GDBN} will run.
31311 (You can often leave off @var{host}; @file{configure} tries to guess the
31312 correct value by examining your system.)
31314 Running @samp{configure @var{host}} and then running @code{make} builds the
31315 @file{bfd}, @file{readline}, @file{mmalloc}, and @file{libiberty}
31316 libraries, then @code{gdb} itself. The configured source files, and the
31317 binaries, are left in the corresponding source directories.
31320 @file{configure} is a Bourne-shell (@code{/bin/sh}) script; if your
31321 system does not recognize this automatically when you run a different
31322 shell, you may need to run @code{sh} on it explicitly:
31325 sh configure @var{host}
31328 If you run @file{configure} from a directory that contains source
31329 directories for multiple libraries or programs, such as the
31330 @file{gdb-@value{GDBVN}} source directory for version @value{GDBVN},
31332 creates configuration files for every directory level underneath (unless
31333 you tell it not to, with the @samp{--norecursion} option).
31335 You should run the @file{configure} script from the top directory in the
31336 source tree, the @file{gdb-@var{version-number}} directory. If you run
31337 @file{configure} from one of the subdirectories, you will configure only
31338 that subdirectory. That is usually not what you want. In particular,
31339 if you run the first @file{configure} from the @file{gdb} subdirectory
31340 of the @file{gdb-@var{version-number}} directory, you will omit the
31341 configuration of @file{bfd}, @file{readline}, and other sibling
31342 directories of the @file{gdb} subdirectory. This leads to build errors
31343 about missing include files such as @file{bfd/bfd.h}.
31345 You can install @code{@value{GDBP}} anywhere; it has no hardwired paths.
31346 However, you should make sure that the shell on your path (named by
31347 the @samp{SHELL} environment variable) is publicly readable. Remember
31348 that @value{GDBN} uses the shell to start your program---some systems refuse to
31349 let @value{GDBN} debug child processes whose programs are not readable.
31351 @node Separate Objdir
31352 @section Compiling @value{GDBN} in Another Directory
31354 If you want to run @value{GDBN} versions for several host or target machines,
31355 you need a different @code{gdb} compiled for each combination of
31356 host and target. @file{configure} is designed to make this easy by
31357 allowing you to generate each configuration in a separate subdirectory,
31358 rather than in the source directory. If your @code{make} program
31359 handles the @samp{VPATH} feature (@sc{gnu} @code{make} does), running
31360 @code{make} in each of these directories builds the @code{gdb}
31361 program specified there.
31363 To build @code{gdb} in a separate directory, run @file{configure}
31364 with the @samp{--srcdir} option to specify where to find the source.
31365 (You also need to specify a path to find @file{configure}
31366 itself from your working directory. If the path to @file{configure}
31367 would be the same as the argument to @samp{--srcdir}, you can leave out
31368 the @samp{--srcdir} option; it is assumed.)
31370 For example, with version @value{GDBVN}, you can build @value{GDBN} in a
31371 separate directory for a Sun 4 like this:
31375 cd gdb-@value{GDBVN}
31378 ../gdb-@value{GDBVN}/configure sun4
31383 When @file{configure} builds a configuration using a remote source
31384 directory, it creates a tree for the binaries with the same structure
31385 (and using the same names) as the tree under the source directory. In
31386 the example, you'd find the Sun 4 library @file{libiberty.a} in the
31387 directory @file{gdb-sun4/libiberty}, and @value{GDBN} itself in
31388 @file{gdb-sun4/gdb}.
31390 Make sure that your path to the @file{configure} script has just one
31391 instance of @file{gdb} in it. If your path to @file{configure} looks
31392 like @file{../gdb-@value{GDBVN}/gdb/configure}, you are configuring only
31393 one subdirectory of @value{GDBN}, not the whole package. This leads to
31394 build errors about missing include files such as @file{bfd/bfd.h}.
31396 One popular reason to build several @value{GDBN} configurations in separate
31397 directories is to configure @value{GDBN} for cross-compiling (where
31398 @value{GDBN} runs on one machine---the @dfn{host}---while debugging
31399 programs that run on another machine---the @dfn{target}).
31400 You specify a cross-debugging target by
31401 giving the @samp{--target=@var{target}} option to @file{configure}.
31403 When you run @code{make} to build a program or library, you must run
31404 it in a configured directory---whatever directory you were in when you
31405 called @file{configure} (or one of its subdirectories).
31407 The @code{Makefile} that @file{configure} generates in each source
31408 directory also runs recursively. If you type @code{make} in a source
31409 directory such as @file{gdb-@value{GDBVN}} (or in a separate configured
31410 directory configured with @samp{--srcdir=@var{dirname}/gdb-@value{GDBVN}}), you
31411 will build all the required libraries, and then build GDB.
31413 When you have multiple hosts or targets configured in separate
31414 directories, you can run @code{make} on them in parallel (for example,
31415 if they are NFS-mounted on each of the hosts); they will not interfere
31419 @section Specifying Names for Hosts and Targets
31421 The specifications used for hosts and targets in the @file{configure}
31422 script are based on a three-part naming scheme, but some short predefined
31423 aliases are also supported. The full naming scheme encodes three pieces
31424 of information in the following pattern:
31427 @var{architecture}-@var{vendor}-@var{os}
31430 For example, you can use the alias @code{sun4} as a @var{host} argument,
31431 or as the value for @var{target} in a @code{--target=@var{target}}
31432 option. The equivalent full name is @samp{sparc-sun-sunos4}.
31434 The @file{configure} script accompanying @value{GDBN} does not provide
31435 any query facility to list all supported host and target names or
31436 aliases. @file{configure} calls the Bourne shell script
31437 @code{config.sub} to map abbreviations to full names; you can read the
31438 script, if you wish, or you can use it to test your guesses on
31439 abbreviations---for example:
31442 % sh config.sub i386-linux
31444 % sh config.sub alpha-linux
31445 alpha-unknown-linux-gnu
31446 % sh config.sub hp9k700
31448 % sh config.sub sun4
31449 sparc-sun-sunos4.1.1
31450 % sh config.sub sun3
31451 m68k-sun-sunos4.1.1
31452 % sh config.sub i986v
31453 Invalid configuration `i986v': machine `i986v' not recognized
31457 @code{config.sub} is also distributed in the @value{GDBN} source
31458 directory (@file{gdb-@value{GDBVN}}, for version @value{GDBVN}).
31460 @node Configure Options
31461 @section @file{configure} Options
31463 Here is a summary of the @file{configure} options and arguments that
31464 are most often useful for building @value{GDBN}. @file{configure} also has
31465 several other options not listed here. @inforef{What Configure
31466 Does,,configure.info}, for a full explanation of @file{configure}.
31469 configure @r{[}--help@r{]}
31470 @r{[}--prefix=@var{dir}@r{]}
31471 @r{[}--exec-prefix=@var{dir}@r{]}
31472 @r{[}--srcdir=@var{dirname}@r{]}
31473 @r{[}--norecursion@r{]} @r{[}--rm@r{]}
31474 @r{[}--target=@var{target}@r{]}
31479 You may introduce options with a single @samp{-} rather than
31480 @samp{--} if you prefer; but you may abbreviate option names if you use
31485 Display a quick summary of how to invoke @file{configure}.
31487 @item --prefix=@var{dir}
31488 Configure the source to install programs and files under directory
31491 @item --exec-prefix=@var{dir}
31492 Configure the source to install programs under directory
31495 @c avoid splitting the warning from the explanation:
31497 @item --srcdir=@var{dirname}
31498 @strong{Warning: using this option requires @sc{gnu} @code{make}, or another
31499 @code{make} that implements the @code{VPATH} feature.}@*
31500 Use this option to make configurations in directories separate from the
31501 @value{GDBN} source directories. Among other things, you can use this to
31502 build (or maintain) several configurations simultaneously, in separate
31503 directories. @file{configure} writes configuration-specific files in
31504 the current directory, but arranges for them to use the source in the
31505 directory @var{dirname}. @file{configure} creates directories under
31506 the working directory in parallel to the source directories below
31509 @item --norecursion
31510 Configure only the directory level where @file{configure} is executed; do not
31511 propagate configuration to subdirectories.
31513 @item --target=@var{target}
31514 Configure @value{GDBN} for cross-debugging programs running on the specified
31515 @var{target}. Without this option, @value{GDBN} is configured to debug
31516 programs that run on the same machine (@var{host}) as @value{GDBN} itself.
31518 There is no convenient way to generate a list of all available targets.
31520 @item @var{host} @dots{}
31521 Configure @value{GDBN} to run on the specified @var{host}.
31523 There is no convenient way to generate a list of all available hosts.
31526 There are many other options available as well, but they are generally
31527 needed for special purposes only.
31529 @node System-wide configuration
31530 @section System-wide configuration and settings
31531 @cindex system-wide init file
31533 @value{GDBN} can be configured to have a system-wide init file;
31534 this file will be read and executed at startup (@pxref{Startup, , What
31535 @value{GDBN} does during startup}).
31537 Here is the corresponding configure option:
31540 @item --with-system-gdbinit=@var{file}
31541 Specify that the default location of the system-wide init file is
31545 If @value{GDBN} has been configured with the option @option{--prefix=$prefix},
31546 it may be subject to relocation. Two possible cases:
31550 If the default location of this init file contains @file{$prefix},
31551 it will be subject to relocation. Suppose that the configure options
31552 are @option{--prefix=$prefix --with-system-gdbinit=$prefix/etc/gdbinit};
31553 if @value{GDBN} is moved from @file{$prefix} to @file{$install}, the system
31554 init file is looked for as @file{$install/etc/gdbinit} instead of
31555 @file{$prefix/etc/gdbinit}.
31558 By contrast, if the default location does not contain the prefix,
31559 it will not be relocated. E.g.@: if @value{GDBN} has been configured with
31560 @option{--prefix=/usr/local --with-system-gdbinit=/usr/share/gdb/gdbinit},
31561 then @value{GDBN} will always look for @file{/usr/share/gdb/gdbinit},
31562 wherever @value{GDBN} is installed.
31565 @node Maintenance Commands
31566 @appendix Maintenance Commands
31567 @cindex maintenance commands
31568 @cindex internal commands
31570 In addition to commands intended for @value{GDBN} users, @value{GDBN}
31571 includes a number of commands intended for @value{GDBN} developers,
31572 that are not documented elsewhere in this manual. These commands are
31573 provided here for reference. (For commands that turn on debugging
31574 messages, see @ref{Debugging Output}.)
31577 @kindex maint agent
31578 @kindex maint agent-eval
31579 @item maint agent @var{expression}
31580 @itemx maint agent-eval @var{expression}
31581 Translate the given @var{expression} into remote agent bytecodes.
31582 This command is useful for debugging the Agent Expression mechanism
31583 (@pxref{Agent Expressions}). The @samp{agent} version produces an
31584 expression useful for data collection, such as by tracepoints, while
31585 @samp{maint agent-eval} produces an expression that evaluates directly
31586 to a result. For instance, a collection expression for @code{globa +
31587 globb} will include bytecodes to record four bytes of memory at each
31588 of the addresses of @code{globa} and @code{globb}, while discarding
31589 the result of the addition, while an evaluation expression will do the
31590 addition and return the sum.
31592 @kindex maint info breakpoints
31593 @item @anchor{maint info breakpoints}maint info breakpoints
31594 Using the same format as @samp{info breakpoints}, display both the
31595 breakpoints you've set explicitly, and those @value{GDBN} is using for
31596 internal purposes. Internal breakpoints are shown with negative
31597 breakpoint numbers. The type column identifies what kind of breakpoint
31602 Normal, explicitly set breakpoint.
31605 Normal, explicitly set watchpoint.
31608 Internal breakpoint, used to handle correctly stepping through
31609 @code{longjmp} calls.
31611 @item longjmp resume
31612 Internal breakpoint at the target of a @code{longjmp}.
31615 Temporary internal breakpoint used by the @value{GDBN} @code{until} command.
31618 Temporary internal breakpoint used by the @value{GDBN} @code{finish} command.
31621 Shared library events.
31625 @kindex set displaced-stepping
31626 @kindex show displaced-stepping
31627 @cindex displaced stepping support
31628 @cindex out-of-line single-stepping
31629 @item set displaced-stepping
31630 @itemx show displaced-stepping
31631 Control whether or not @value{GDBN} will do @dfn{displaced stepping}
31632 if the target supports it. Displaced stepping is a way to single-step
31633 over breakpoints without removing them from the inferior, by executing
31634 an out-of-line copy of the instruction that was originally at the
31635 breakpoint location. It is also known as out-of-line single-stepping.
31638 @item set displaced-stepping on
31639 If the target architecture supports it, @value{GDBN} will use
31640 displaced stepping to step over breakpoints.
31642 @item set displaced-stepping off
31643 @value{GDBN} will not use displaced stepping to step over breakpoints,
31644 even if such is supported by the target architecture.
31646 @cindex non-stop mode, and @samp{set displaced-stepping}
31647 @item set displaced-stepping auto
31648 This is the default mode. @value{GDBN} will use displaced stepping
31649 only if non-stop mode is active (@pxref{Non-Stop Mode}) and the target
31650 architecture supports displaced stepping.
31653 @kindex maint check-symtabs
31654 @item maint check-symtabs
31655 Check the consistency of psymtabs and symtabs.
31657 @kindex maint cplus first_component
31658 @item maint cplus first_component @var{name}
31659 Print the first C@t{++} class/namespace component of @var{name}.
31661 @kindex maint cplus namespace
31662 @item maint cplus namespace
31663 Print the list of possible C@t{++} namespaces.
31665 @kindex maint demangle
31666 @item maint demangle @var{name}
31667 Demangle a C@t{++} or Objective-C mangled @var{name}.
31669 @kindex maint deprecate
31670 @kindex maint undeprecate
31671 @cindex deprecated commands
31672 @item maint deprecate @var{command} @r{[}@var{replacement}@r{]}
31673 @itemx maint undeprecate @var{command}
31674 Deprecate or undeprecate the named @var{command}. Deprecated commands
31675 cause @value{GDBN} to issue a warning when you use them. The optional
31676 argument @var{replacement} says which newer command should be used in
31677 favor of the deprecated one; if it is given, @value{GDBN} will mention
31678 the replacement as part of the warning.
31680 @kindex maint dump-me
31681 @item maint dump-me
31682 @cindex @code{SIGQUIT} signal, dump core of @value{GDBN}
31683 Cause a fatal signal in the debugger and force it to dump its core.
31684 This is supported only on systems which support aborting a program
31685 with the @code{SIGQUIT} signal.
31687 @kindex maint internal-error
31688 @kindex maint internal-warning
31689 @item maint internal-error @r{[}@var{message-text}@r{]}
31690 @itemx maint internal-warning @r{[}@var{message-text}@r{]}
31691 Cause @value{GDBN} to call the internal function @code{internal_error}
31692 or @code{internal_warning} and hence behave as though an internal error
31693 or internal warning has been detected. In addition to reporting the
31694 internal problem, these functions give the user the opportunity to
31695 either quit @value{GDBN} or create a core file of the current
31696 @value{GDBN} session.
31698 These commands take an optional parameter @var{message-text} that is
31699 used as the text of the error or warning message.
31701 Here's an example of using @code{internal-error}:
31704 (@value{GDBP}) @kbd{maint internal-error testing, 1, 2}
31705 @dots{}/maint.c:121: internal-error: testing, 1, 2
31706 A problem internal to GDB has been detected. Further
31707 debugging may prove unreliable.
31708 Quit this debugging session? (y or n) @kbd{n}
31709 Create a core file? (y or n) @kbd{n}
31713 @cindex @value{GDBN} internal error
31714 @cindex internal errors, control of @value{GDBN} behavior
31716 @kindex maint set internal-error
31717 @kindex maint show internal-error
31718 @kindex maint set internal-warning
31719 @kindex maint show internal-warning
31720 @item maint set internal-error @var{action} [ask|yes|no]
31721 @itemx maint show internal-error @var{action}
31722 @itemx maint set internal-warning @var{action} [ask|yes|no]
31723 @itemx maint show internal-warning @var{action}
31724 When @value{GDBN} reports an internal problem (error or warning) it
31725 gives the user the opportunity to both quit @value{GDBN} and create a
31726 core file of the current @value{GDBN} session. These commands let you
31727 override the default behaviour for each particular @var{action},
31728 described in the table below.
31732 You can specify that @value{GDBN} should always (yes) or never (no)
31733 quit. The default is to ask the user what to do.
31736 You can specify that @value{GDBN} should always (yes) or never (no)
31737 create a core file. The default is to ask the user what to do.
31740 @kindex maint packet
31741 @item maint packet @var{text}
31742 If @value{GDBN} is talking to an inferior via the serial protocol,
31743 then this command sends the string @var{text} to the inferior, and
31744 displays the response packet. @value{GDBN} supplies the initial
31745 @samp{$} character, the terminating @samp{#} character, and the
31748 @kindex maint print architecture
31749 @item maint print architecture @r{[}@var{file}@r{]}
31750 Print the entire architecture configuration. The optional argument
31751 @var{file} names the file where the output goes.
31753 @kindex maint print c-tdesc
31754 @item maint print c-tdesc
31755 Print the current target description (@pxref{Target Descriptions}) as
31756 a C source file. The created source file can be used in @value{GDBN}
31757 when an XML parser is not available to parse the description.
31759 @kindex maint print dummy-frames
31760 @item maint print dummy-frames
31761 Prints the contents of @value{GDBN}'s internal dummy-frame stack.
31764 (@value{GDBP}) @kbd{b add}
31766 (@value{GDBP}) @kbd{print add(2,3)}
31767 Breakpoint 2, add (a=2, b=3) at @dots{}
31769 The program being debugged stopped while in a function called from GDB.
31771 (@value{GDBP}) @kbd{maint print dummy-frames}
31772 0x1a57c80: pc=0x01014068 fp=0x0200bddc sp=0x0200bdd6
31773 top=0x0200bdd4 id=@{stack=0x200bddc,code=0x101405c@}
31774 call_lo=0x01014000 call_hi=0x01014001
31778 Takes an optional file parameter.
31780 @kindex maint print registers
31781 @kindex maint print raw-registers
31782 @kindex maint print cooked-registers
31783 @kindex maint print register-groups
31784 @kindex maint print remote-registers
31785 @item maint print registers @r{[}@var{file}@r{]}
31786 @itemx maint print raw-registers @r{[}@var{file}@r{]}
31787 @itemx maint print cooked-registers @r{[}@var{file}@r{]}
31788 @itemx maint print register-groups @r{[}@var{file}@r{]}
31789 @itemx maint print remote-registers @r{[}@var{file}@r{]}
31790 Print @value{GDBN}'s internal register data structures.
31792 The command @code{maint print raw-registers} includes the contents of
31793 the raw register cache; the command @code{maint print
31794 cooked-registers} includes the (cooked) value of all registers,
31795 including registers which aren't available on the target nor visible
31796 to user; the command @code{maint print register-groups} includes the
31797 groups that each register is a member of; and the command @code{maint
31798 print remote-registers} includes the remote target's register numbers
31799 and offsets in the `G' packets. @xref{Registers,, Registers, gdbint,
31800 @value{GDBN} Internals}.
31802 These commands take an optional parameter, a file name to which to
31803 write the information.
31805 @kindex maint print reggroups
31806 @item maint print reggroups @r{[}@var{file}@r{]}
31807 Print @value{GDBN}'s internal register group data structures. The
31808 optional argument @var{file} tells to what file to write the
31811 The register groups info looks like this:
31814 (@value{GDBP}) @kbd{maint print reggroups}
31827 This command forces @value{GDBN} to flush its internal register cache.
31829 @kindex maint print objfiles
31830 @cindex info for known object files
31831 @item maint print objfiles
31832 Print a dump of all known object files. For each object file, this
31833 command prints its name, address in memory, and all of its psymtabs
31836 @kindex maint print section-scripts
31837 @cindex info for known .debug_gdb_scripts-loaded scripts
31838 @item maint print section-scripts [@var{regexp}]
31839 Print a dump of scripts specified in the @code{.debug_gdb_section} section.
31840 If @var{regexp} is specified, only print scripts loaded by object files
31841 matching @var{regexp}.
31842 For each script, this command prints its name as specified in the objfile,
31843 and the full path if known.
31844 @xref{.debug_gdb_scripts section}.
31846 @kindex maint print statistics
31847 @cindex bcache statistics
31848 @item maint print statistics
31849 This command prints, for each object file in the program, various data
31850 about that object file followed by the byte cache (@dfn{bcache})
31851 statistics for the object file. The objfile data includes the number
31852 of minimal, partial, full, and stabs symbols, the number of types
31853 defined by the objfile, the number of as yet unexpanded psym tables,
31854 the number of line tables and string tables, and the amount of memory
31855 used by the various tables. The bcache statistics include the counts,
31856 sizes, and counts of duplicates of all and unique objects, max,
31857 average, and median entry size, total memory used and its overhead and
31858 savings, and various measures of the hash table size and chain
31861 @kindex maint print target-stack
31862 @cindex target stack description
31863 @item maint print target-stack
31864 A @dfn{target} is an interface between the debugger and a particular
31865 kind of file or process. Targets can be stacked in @dfn{strata},
31866 so that more than one target can potentially respond to a request.
31867 In particular, memory accesses will walk down the stack of targets
31868 until they find a target that is interested in handling that particular
31871 This command prints a short description of each layer that was pushed on
31872 the @dfn{target stack}, starting from the top layer down to the bottom one.
31874 @kindex maint print type
31875 @cindex type chain of a data type
31876 @item maint print type @var{expr}
31877 Print the type chain for a type specified by @var{expr}. The argument
31878 can be either a type name or a symbol. If it is a symbol, the type of
31879 that symbol is described. The type chain produced by this command is
31880 a recursive definition of the data type as stored in @value{GDBN}'s
31881 data structures, including its flags and contained types.
31883 @kindex maint set dwarf2 always-disassemble
31884 @kindex maint show dwarf2 always-disassemble
31885 @item maint set dwarf2 always-disassemble
31886 @item maint show dwarf2 always-disassemble
31887 Control the behavior of @code{info address} when using DWARF debugging
31890 The default is @code{off}, which means that @value{GDBN} should try to
31891 describe a variable's location in an easily readable format. When
31892 @code{on}, @value{GDBN} will instead display the DWARF location
31893 expression in an assembly-like format. Note that some locations are
31894 too complex for @value{GDBN} to describe simply; in this case you will
31895 always see the disassembly form.
31897 Here is an example of the resulting disassembly:
31900 (gdb) info addr argc
31901 Symbol "argc" is a complex DWARF expression:
31905 For more information on these expressions, see
31906 @uref{http://www.dwarfstd.org/, the DWARF standard}.
31908 @kindex maint set dwarf2 max-cache-age
31909 @kindex maint show dwarf2 max-cache-age
31910 @item maint set dwarf2 max-cache-age
31911 @itemx maint show dwarf2 max-cache-age
31912 Control the DWARF 2 compilation unit cache.
31914 @cindex DWARF 2 compilation units cache
31915 In object files with inter-compilation-unit references, such as those
31916 produced by the GCC option @samp{-feliminate-dwarf2-dups}, the DWARF 2
31917 reader needs to frequently refer to previously read compilation units.
31918 This setting controls how long a compilation unit will remain in the
31919 cache if it is not referenced. A higher limit means that cached
31920 compilation units will be stored in memory longer, and more total
31921 memory will be used. Setting it to zero disables caching, which will
31922 slow down @value{GDBN} startup, but reduce memory consumption.
31924 @kindex maint set profile
31925 @kindex maint show profile
31926 @cindex profiling GDB
31927 @item maint set profile
31928 @itemx maint show profile
31929 Control profiling of @value{GDBN}.
31931 Profiling will be disabled until you use the @samp{maint set profile}
31932 command to enable it. When you enable profiling, the system will begin
31933 collecting timing and execution count data; when you disable profiling or
31934 exit @value{GDBN}, the results will be written to a log file. Remember that
31935 if you use profiling, @value{GDBN} will overwrite the profiling log file
31936 (often called @file{gmon.out}). If you have a record of important profiling
31937 data in a @file{gmon.out} file, be sure to move it to a safe location.
31939 Configuring with @samp{--enable-profiling} arranges for @value{GDBN} to be
31940 compiled with the @samp{-pg} compiler option.
31942 @kindex maint set show-debug-regs
31943 @kindex maint show show-debug-regs
31944 @cindex hardware debug registers
31945 @item maint set show-debug-regs
31946 @itemx maint show show-debug-regs
31947 Control whether to show variables that mirror the hardware debug
31948 registers. Use @code{ON} to enable, @code{OFF} to disable. If
31949 enabled, the debug registers values are shown when @value{GDBN} inserts or
31950 removes a hardware breakpoint or watchpoint, and when the inferior
31951 triggers a hardware-assisted breakpoint or watchpoint.
31953 @kindex maint set show-all-tib
31954 @kindex maint show show-all-tib
31955 @item maint set show-all-tib
31956 @itemx maint show show-all-tib
31957 Control whether to show all non zero areas within a 1k block starting
31958 at thread local base, when using the @samp{info w32 thread-information-block}
31961 @kindex maint space
31962 @cindex memory used by commands
31964 Control whether to display memory usage for each command. If set to a
31965 nonzero value, @value{GDBN} will display how much memory each command
31966 took, following the command's own output. This can also be requested
31967 by invoking @value{GDBN} with the @option{--statistics} command-line
31968 switch (@pxref{Mode Options}).
31971 @cindex time of command execution
31973 Control whether to display the execution time for each command. If
31974 set to a nonzero value, @value{GDBN} will display how much time it
31975 took to execute each command, following the command's own output.
31976 The time is not printed for the commands that run the target, since
31977 there's no mechanism currently to compute how much time was spend
31978 by @value{GDBN} and how much time was spend by the program been debugged.
31979 it's not possibly currently
31980 This can also be requested by invoking @value{GDBN} with the
31981 @option{--statistics} command-line switch (@pxref{Mode Options}).
31983 @kindex maint translate-address
31984 @item maint translate-address @r{[}@var{section}@r{]} @var{addr}
31985 Find the symbol stored at the location specified by the address
31986 @var{addr} and an optional section name @var{section}. If found,
31987 @value{GDBN} prints the name of the closest symbol and an offset from
31988 the symbol's location to the specified address. This is similar to
31989 the @code{info address} command (@pxref{Symbols}), except that this
31990 command also allows to find symbols in other sections.
31992 If section was not specified, the section in which the symbol was found
31993 is also printed. For dynamically linked executables, the name of
31994 executable or shared library containing the symbol is printed as well.
31998 The following command is useful for non-interactive invocations of
31999 @value{GDBN}, such as in the test suite.
32002 @item set watchdog @var{nsec}
32003 @kindex set watchdog
32004 @cindex watchdog timer
32005 @cindex timeout for commands
32006 Set the maximum number of seconds @value{GDBN} will wait for the
32007 target operation to finish. If this time expires, @value{GDBN}
32008 reports and error and the command is aborted.
32010 @item show watchdog
32011 Show the current setting of the target wait timeout.
32014 @node Remote Protocol
32015 @appendix @value{GDBN} Remote Serial Protocol
32020 * Stop Reply Packets::
32021 * General Query Packets::
32022 * Architecture-Specific Protocol Details::
32023 * Tracepoint Packets::
32024 * Host I/O Packets::
32026 * Notification Packets::
32027 * Remote Non-Stop::
32028 * Packet Acknowledgment::
32030 * File-I/O Remote Protocol Extension::
32031 * Library List Format::
32032 * Memory Map Format::
32033 * Thread List Format::
32034 * Traceframe Info Format::
32040 There may be occasions when you need to know something about the
32041 protocol---for example, if there is only one serial port to your target
32042 machine, you might want your program to do something special if it
32043 recognizes a packet meant for @value{GDBN}.
32045 In the examples below, @samp{->} and @samp{<-} are used to indicate
32046 transmitted and received data, respectively.
32048 @cindex protocol, @value{GDBN} remote serial
32049 @cindex serial protocol, @value{GDBN} remote
32050 @cindex remote serial protocol
32051 All @value{GDBN} commands and responses (other than acknowledgments
32052 and notifications, see @ref{Notification Packets}) are sent as a
32053 @var{packet}. A @var{packet} is introduced with the character
32054 @samp{$}, the actual @var{packet-data}, and the terminating character
32055 @samp{#} followed by a two-digit @var{checksum}:
32058 @code{$}@var{packet-data}@code{#}@var{checksum}
32062 @cindex checksum, for @value{GDBN} remote
32064 The two-digit @var{checksum} is computed as the modulo 256 sum of all
32065 characters between the leading @samp{$} and the trailing @samp{#} (an
32066 eight bit unsigned checksum).
32068 Implementors should note that prior to @value{GDBN} 5.0 the protocol
32069 specification also included an optional two-digit @var{sequence-id}:
32072 @code{$}@var{sequence-id}@code{:}@var{packet-data}@code{#}@var{checksum}
32075 @cindex sequence-id, for @value{GDBN} remote
32077 That @var{sequence-id} was appended to the acknowledgment. @value{GDBN}
32078 has never output @var{sequence-id}s. Stubs that handle packets added
32079 since @value{GDBN} 5.0 must not accept @var{sequence-id}.
32081 When either the host or the target machine receives a packet, the first
32082 response expected is an acknowledgment: either @samp{+} (to indicate
32083 the package was received correctly) or @samp{-} (to request
32087 -> @code{$}@var{packet-data}@code{#}@var{checksum}
32092 The @samp{+}/@samp{-} acknowledgments can be disabled
32093 once a connection is established.
32094 @xref{Packet Acknowledgment}, for details.
32096 The host (@value{GDBN}) sends @var{command}s, and the target (the
32097 debugging stub incorporated in your program) sends a @var{response}. In
32098 the case of step and continue @var{command}s, the response is only sent
32099 when the operation has completed, and the target has again stopped all
32100 threads in all attached processes. This is the default all-stop mode
32101 behavior, but the remote protocol also supports @value{GDBN}'s non-stop
32102 execution mode; see @ref{Remote Non-Stop}, for details.
32104 @var{packet-data} consists of a sequence of characters with the
32105 exception of @samp{#} and @samp{$} (see @samp{X} packet for additional
32108 @cindex remote protocol, field separator
32109 Fields within the packet should be separated using @samp{,} @samp{;} or
32110 @samp{:}. Except where otherwise noted all numbers are represented in
32111 @sc{hex} with leading zeros suppressed.
32113 Implementors should note that prior to @value{GDBN} 5.0, the character
32114 @samp{:} could not appear as the third character in a packet (as it
32115 would potentially conflict with the @var{sequence-id}).
32117 @cindex remote protocol, binary data
32118 @anchor{Binary Data}
32119 Binary data in most packets is encoded either as two hexadecimal
32120 digits per byte of binary data. This allowed the traditional remote
32121 protocol to work over connections which were only seven-bit clean.
32122 Some packets designed more recently assume an eight-bit clean
32123 connection, and use a more efficient encoding to send and receive
32126 The binary data representation uses @code{7d} (@sc{ascii} @samp{@}})
32127 as an escape character. Any escaped byte is transmitted as the escape
32128 character followed by the original character XORed with @code{0x20}.
32129 For example, the byte @code{0x7d} would be transmitted as the two
32130 bytes @code{0x7d 0x5d}. The bytes @code{0x23} (@sc{ascii} @samp{#}),
32131 @code{0x24} (@sc{ascii} @samp{$}), and @code{0x7d} (@sc{ascii}
32132 @samp{@}}) must always be escaped. Responses sent by the stub
32133 must also escape @code{0x2a} (@sc{ascii} @samp{*}), so that it
32134 is not interpreted as the start of a run-length encoded sequence
32137 Response @var{data} can be run-length encoded to save space.
32138 Run-length encoding replaces runs of identical characters with one
32139 instance of the repeated character, followed by a @samp{*} and a
32140 repeat count. The repeat count is itself sent encoded, to avoid
32141 binary characters in @var{data}: a value of @var{n} is sent as
32142 @code{@var{n}+29}. For a repeat count greater or equal to 3, this
32143 produces a printable @sc{ascii} character, e.g.@: a space (@sc{ascii}
32144 code 32) for a repeat count of 3. (This is because run-length
32145 encoding starts to win for counts 3 or more.) Thus, for example,
32146 @samp{0* } is a run-length encoding of ``0000'': the space character
32147 after @samp{*} means repeat the leading @code{0} @w{@code{32 - 29 =
32150 The printable characters @samp{#} and @samp{$} or with a numeric value
32151 greater than 126 must not be used. Runs of six repeats (@samp{#}) or
32152 seven repeats (@samp{$}) can be expanded using a repeat count of only
32153 five (@samp{"}). For example, @samp{00000000} can be encoded as
32156 The error response returned for some packets includes a two character
32157 error number. That number is not well defined.
32159 @cindex empty response, for unsupported packets
32160 For any @var{command} not supported by the stub, an empty response
32161 (@samp{$#00}) should be returned. That way it is possible to extend the
32162 protocol. A newer @value{GDBN} can tell if a packet is supported based
32165 A stub is required to support the @samp{g}, @samp{G}, @samp{m}, @samp{M},
32166 @samp{c}, and @samp{s} @var{command}s. All other @var{command}s are
32172 The following table provides a complete list of all currently defined
32173 @var{command}s and their corresponding response @var{data}.
32174 @xref{File-I/O Remote Protocol Extension}, for details about the File
32175 I/O extension of the remote protocol.
32177 Each packet's description has a template showing the packet's overall
32178 syntax, followed by an explanation of the packet's meaning. We
32179 include spaces in some of the templates for clarity; these are not
32180 part of the packet's syntax. No @value{GDBN} packet uses spaces to
32181 separate its components. For example, a template like @samp{foo
32182 @var{bar} @var{baz}} describes a packet beginning with the three ASCII
32183 bytes @samp{foo}, followed by a @var{bar}, followed directly by a
32184 @var{baz}. @value{GDBN} does not transmit a space character between the
32185 @samp{foo} and the @var{bar}, or between the @var{bar} and the
32188 @cindex @var{thread-id}, in remote protocol
32189 @anchor{thread-id syntax}
32190 Several packets and replies include a @var{thread-id} field to identify
32191 a thread. Normally these are positive numbers with a target-specific
32192 interpretation, formatted as big-endian hex strings. A @var{thread-id}
32193 can also be a literal @samp{-1} to indicate all threads, or @samp{0} to
32196 In addition, the remote protocol supports a multiprocess feature in
32197 which the @var{thread-id} syntax is extended to optionally include both
32198 process and thread ID fields, as @samp{p@var{pid}.@var{tid}}.
32199 The @var{pid} (process) and @var{tid} (thread) components each have the
32200 format described above: a positive number with target-specific
32201 interpretation formatted as a big-endian hex string, literal @samp{-1}
32202 to indicate all processes or threads (respectively), or @samp{0} to
32203 indicate an arbitrary process or thread. Specifying just a process, as
32204 @samp{p@var{pid}}, is equivalent to @samp{p@var{pid}.-1}. It is an
32205 error to specify all processes but a specific thread, such as
32206 @samp{p-1.@var{tid}}. Note that the @samp{p} prefix is @emph{not} used
32207 for those packets and replies explicitly documented to include a process
32208 ID, rather than a @var{thread-id}.
32210 The multiprocess @var{thread-id} syntax extensions are only used if both
32211 @value{GDBN} and the stub report support for the @samp{multiprocess}
32212 feature using @samp{qSupported}. @xref{multiprocess extensions}, for
32215 Note that all packet forms beginning with an upper- or lower-case
32216 letter, other than those described here, are reserved for future use.
32218 Here are the packet descriptions.
32223 @cindex @samp{!} packet
32224 @anchor{extended mode}
32225 Enable extended mode. In extended mode, the remote server is made
32226 persistent. The @samp{R} packet is used to restart the program being
32232 The remote target both supports and has enabled extended mode.
32236 @cindex @samp{?} packet
32237 Indicate the reason the target halted. The reply is the same as for
32238 step and continue. This packet has a special interpretation when the
32239 target is in non-stop mode; see @ref{Remote Non-Stop}.
32242 @xref{Stop Reply Packets}, for the reply specifications.
32244 @item A @var{arglen},@var{argnum},@var{arg},@dots{}
32245 @cindex @samp{A} packet
32246 Initialized @code{argv[]} array passed into program. @var{arglen}
32247 specifies the number of bytes in the hex encoded byte stream
32248 @var{arg}. See @code{gdbserver} for more details.
32253 The arguments were set.
32259 @cindex @samp{b} packet
32260 (Don't use this packet; its behavior is not well-defined.)
32261 Change the serial line speed to @var{baud}.
32263 JTC: @emph{When does the transport layer state change? When it's
32264 received, or after the ACK is transmitted. In either case, there are
32265 problems if the command or the acknowledgment packet is dropped.}
32267 Stan: @emph{If people really wanted to add something like this, and get
32268 it working for the first time, they ought to modify ser-unix.c to send
32269 some kind of out-of-band message to a specially-setup stub and have the
32270 switch happen "in between" packets, so that from remote protocol's point
32271 of view, nothing actually happened.}
32273 @item B @var{addr},@var{mode}
32274 @cindex @samp{B} packet
32275 Set (@var{mode} is @samp{S}) or clear (@var{mode} is @samp{C}) a
32276 breakpoint at @var{addr}.
32278 Don't use this packet. Use the @samp{Z} and @samp{z} packets instead
32279 (@pxref{insert breakpoint or watchpoint packet}).
32281 @cindex @samp{bc} packet
32284 Backward continue. Execute the target system in reverse. No parameter.
32285 @xref{Reverse Execution}, for more information.
32288 @xref{Stop Reply Packets}, for the reply specifications.
32290 @cindex @samp{bs} packet
32293 Backward single step. Execute one instruction in reverse. No parameter.
32294 @xref{Reverse Execution}, for more information.
32297 @xref{Stop Reply Packets}, for the reply specifications.
32299 @item c @r{[}@var{addr}@r{]}
32300 @cindex @samp{c} packet
32301 Continue. @var{addr} is address to resume. If @var{addr} is omitted,
32302 resume at current address.
32305 @xref{Stop Reply Packets}, for the reply specifications.
32307 @item C @var{sig}@r{[};@var{addr}@r{]}
32308 @cindex @samp{C} packet
32309 Continue with signal @var{sig} (hex signal number). If
32310 @samp{;@var{addr}} is omitted, resume at same address.
32313 @xref{Stop Reply Packets}, for the reply specifications.
32316 @cindex @samp{d} packet
32319 Don't use this packet; instead, define a general set packet
32320 (@pxref{General Query Packets}).
32324 @cindex @samp{D} packet
32325 The first form of the packet is used to detach @value{GDBN} from the
32326 remote system. It is sent to the remote target
32327 before @value{GDBN} disconnects via the @code{detach} command.
32329 The second form, including a process ID, is used when multiprocess
32330 protocol extensions are enabled (@pxref{multiprocess extensions}), to
32331 detach only a specific process. The @var{pid} is specified as a
32332 big-endian hex string.
32342 @item F @var{RC},@var{EE},@var{CF};@var{XX}
32343 @cindex @samp{F} packet
32344 A reply from @value{GDBN} to an @samp{F} packet sent by the target.
32345 This is part of the File-I/O protocol extension. @xref{File-I/O
32346 Remote Protocol Extension}, for the specification.
32349 @anchor{read registers packet}
32350 @cindex @samp{g} packet
32351 Read general registers.
32355 @item @var{XX@dots{}}
32356 Each byte of register data is described by two hex digits. The bytes
32357 with the register are transmitted in target byte order. The size of
32358 each register and their position within the @samp{g} packet are
32359 determined by the @value{GDBN} internal gdbarch functions
32360 @code{DEPRECATED_REGISTER_RAW_SIZE} and @code{gdbarch_register_name}. The
32361 specification of several standard @samp{g} packets is specified below.
32363 When reading registers from a trace frame (@pxref{Analyze Collected
32364 Data,,Using the Collected Data}), the stub may also return a string of
32365 literal @samp{x}'s in place of the register data digits, to indicate
32366 that the corresponding register has not been collected, thus its value
32367 is unavailable. For example, for an architecture with 4 registers of
32368 4 bytes each, the following reply indicates to @value{GDBN} that
32369 registers 0 and 2 have not been collected, while registers 1 and 3
32370 have been collected, and both have zero value:
32374 <- @code{xxxxxxxx00000000xxxxxxxx00000000}
32381 @item G @var{XX@dots{}}
32382 @cindex @samp{G} packet
32383 Write general registers. @xref{read registers packet}, for a
32384 description of the @var{XX@dots{}} data.
32394 @item H @var{c} @var{thread-id}
32395 @cindex @samp{H} packet
32396 Set thread for subsequent operations (@samp{m}, @samp{M}, @samp{g},
32397 @samp{G}, et.al.). @var{c} depends on the operation to be performed: it
32398 should be @samp{c} for step and continue operations, @samp{g} for other
32399 operations. The thread designator @var{thread-id} has the format and
32400 interpretation described in @ref{thread-id syntax}.
32411 @c 'H': How restrictive (or permissive) is the thread model. If a
32412 @c thread is selected and stopped, are other threads allowed
32413 @c to continue to execute? As I mentioned above, I think the
32414 @c semantics of each command when a thread is selected must be
32415 @c described. For example:
32417 @c 'g': If the stub supports threads and a specific thread is
32418 @c selected, returns the register block from that thread;
32419 @c otherwise returns current registers.
32421 @c 'G' If the stub supports threads and a specific thread is
32422 @c selected, sets the registers of the register block of
32423 @c that thread; otherwise sets current registers.
32425 @item i @r{[}@var{addr}@r{[},@var{nnn}@r{]]}
32426 @anchor{cycle step packet}
32427 @cindex @samp{i} packet
32428 Step the remote target by a single clock cycle. If @samp{,@var{nnn}} is
32429 present, cycle step @var{nnn} cycles. If @var{addr} is present, cycle
32430 step starting at that address.
32433 @cindex @samp{I} packet
32434 Signal, then cycle step. @xref{step with signal packet}. @xref{cycle
32438 @cindex @samp{k} packet
32441 FIXME: @emph{There is no description of how to operate when a specific
32442 thread context has been selected (i.e.@: does 'k' kill only that
32445 @item m @var{addr},@var{length}
32446 @cindex @samp{m} packet
32447 Read @var{length} bytes of memory starting at address @var{addr}.
32448 Note that @var{addr} may not be aligned to any particular boundary.
32450 The stub need not use any particular size or alignment when gathering
32451 data from memory for the response; even if @var{addr} is word-aligned
32452 and @var{length} is a multiple of the word size, the stub is free to
32453 use byte accesses, or not. For this reason, this packet may not be
32454 suitable for accessing memory-mapped I/O devices.
32455 @cindex alignment of remote memory accesses
32456 @cindex size of remote memory accesses
32457 @cindex memory, alignment and size of remote accesses
32461 @item @var{XX@dots{}}
32462 Memory contents; each byte is transmitted as a two-digit hexadecimal
32463 number. The reply may contain fewer bytes than requested if the
32464 server was able to read only part of the region of memory.
32469 @item M @var{addr},@var{length}:@var{XX@dots{}}
32470 @cindex @samp{M} packet
32471 Write @var{length} bytes of memory starting at address @var{addr}.
32472 @var{XX@dots{}} is the data; each byte is transmitted as a two-digit
32473 hexadecimal number.
32480 for an error (this includes the case where only part of the data was
32485 @cindex @samp{p} packet
32486 Read the value of register @var{n}; @var{n} is in hex.
32487 @xref{read registers packet}, for a description of how the returned
32488 register value is encoded.
32492 @item @var{XX@dots{}}
32493 the register's value
32497 Indicating an unrecognized @var{query}.
32500 @item P @var{n@dots{}}=@var{r@dots{}}
32501 @anchor{write register packet}
32502 @cindex @samp{P} packet
32503 Write register @var{n@dots{}} with value @var{r@dots{}}. The register
32504 number @var{n} is in hexadecimal, and @var{r@dots{}} contains two hex
32505 digits for each byte in the register (target byte order).
32515 @item q @var{name} @var{params}@dots{}
32516 @itemx Q @var{name} @var{params}@dots{}
32517 @cindex @samp{q} packet
32518 @cindex @samp{Q} packet
32519 General query (@samp{q}) and set (@samp{Q}). These packets are
32520 described fully in @ref{General Query Packets}.
32523 @cindex @samp{r} packet
32524 Reset the entire system.
32526 Don't use this packet; use the @samp{R} packet instead.
32529 @cindex @samp{R} packet
32530 Restart the program being debugged. @var{XX}, while needed, is ignored.
32531 This packet is only available in extended mode (@pxref{extended mode}).
32533 The @samp{R} packet has no reply.
32535 @item s @r{[}@var{addr}@r{]}
32536 @cindex @samp{s} packet
32537 Single step. @var{addr} is the address at which to resume. If
32538 @var{addr} is omitted, resume at same address.
32541 @xref{Stop Reply Packets}, for the reply specifications.
32543 @item S @var{sig}@r{[};@var{addr}@r{]}
32544 @anchor{step with signal packet}
32545 @cindex @samp{S} packet
32546 Step with signal. This is analogous to the @samp{C} packet, but
32547 requests a single-step, rather than a normal resumption of execution.
32550 @xref{Stop Reply Packets}, for the reply specifications.
32552 @item t @var{addr}:@var{PP},@var{MM}
32553 @cindex @samp{t} packet
32554 Search backwards starting at address @var{addr} for a match with pattern
32555 @var{PP} and mask @var{MM}. @var{PP} and @var{MM} are 4 bytes.
32556 @var{addr} must be at least 3 digits.
32558 @item T @var{thread-id}
32559 @cindex @samp{T} packet
32560 Find out if the thread @var{thread-id} is alive. @xref{thread-id syntax}.
32565 thread is still alive
32571 Packets starting with @samp{v} are identified by a multi-letter name,
32572 up to the first @samp{;} or @samp{?} (or the end of the packet).
32574 @item vAttach;@var{pid}
32575 @cindex @samp{vAttach} packet
32576 Attach to a new process with the specified process ID @var{pid}.
32577 The process ID is a
32578 hexadecimal integer identifying the process. In all-stop mode, all
32579 threads in the attached process are stopped; in non-stop mode, it may be
32580 attached without being stopped if that is supported by the target.
32582 @c In non-stop mode, on a successful vAttach, the stub should set the
32583 @c current thread to a thread of the newly-attached process. After
32584 @c attaching, GDB queries for the attached process's thread ID with qC.
32585 @c Also note that, from a user perspective, whether or not the
32586 @c target is stopped on attach in non-stop mode depends on whether you
32587 @c use the foreground or background version of the attach command, not
32588 @c on what vAttach does; GDB does the right thing with respect to either
32589 @c stopping or restarting threads.
32591 This packet is only available in extended mode (@pxref{extended mode}).
32597 @item @r{Any stop packet}
32598 for success in all-stop mode (@pxref{Stop Reply Packets})
32600 for success in non-stop mode (@pxref{Remote Non-Stop})
32603 @item vCont@r{[};@var{action}@r{[}:@var{thread-id}@r{]]}@dots{}
32604 @cindex @samp{vCont} packet
32605 Resume the inferior, specifying different actions for each thread.
32606 If an action is specified with no @var{thread-id}, then it is applied to any
32607 threads that don't have a specific action specified; if no default action is
32608 specified then other threads should remain stopped in all-stop mode and
32609 in their current state in non-stop mode.
32610 Specifying multiple
32611 default actions is an error; specifying no actions is also an error.
32612 Thread IDs are specified using the syntax described in @ref{thread-id syntax}.
32614 Currently supported actions are:
32620 Continue with signal @var{sig}. The signal @var{sig} should be two hex digits.
32624 Step with signal @var{sig}. The signal @var{sig} should be two hex digits.
32629 The optional argument @var{addr} normally associated with the
32630 @samp{c}, @samp{C}, @samp{s}, and @samp{S} packets is
32631 not supported in @samp{vCont}.
32633 The @samp{t} action is only relevant in non-stop mode
32634 (@pxref{Remote Non-Stop}) and may be ignored by the stub otherwise.
32635 A stop reply should be generated for any affected thread not already stopped.
32636 When a thread is stopped by means of a @samp{t} action,
32637 the corresponding stop reply should indicate that the thread has stopped with
32638 signal @samp{0}, regardless of whether the target uses some other signal
32639 as an implementation detail.
32642 @xref{Stop Reply Packets}, for the reply specifications.
32645 @cindex @samp{vCont?} packet
32646 Request a list of actions supported by the @samp{vCont} packet.
32650 @item vCont@r{[};@var{action}@dots{}@r{]}
32651 The @samp{vCont} packet is supported. Each @var{action} is a supported
32652 command in the @samp{vCont} packet.
32654 The @samp{vCont} packet is not supported.
32657 @item vFile:@var{operation}:@var{parameter}@dots{}
32658 @cindex @samp{vFile} packet
32659 Perform a file operation on the target system. For details,
32660 see @ref{Host I/O Packets}.
32662 @item vFlashErase:@var{addr},@var{length}
32663 @cindex @samp{vFlashErase} packet
32664 Direct the stub to erase @var{length} bytes of flash starting at
32665 @var{addr}. The region may enclose any number of flash blocks, but
32666 its start and end must fall on block boundaries, as indicated by the
32667 flash block size appearing in the memory map (@pxref{Memory Map
32668 Format}). @value{GDBN} groups flash memory programming operations
32669 together, and sends a @samp{vFlashDone} request after each group; the
32670 stub is allowed to delay erase operation until the @samp{vFlashDone}
32671 packet is received.
32673 The stub must support @samp{vCont} if it reports support for
32674 multiprocess extensions (@pxref{multiprocess extensions}). Note that in
32675 this case @samp{vCont} actions can be specified to apply to all threads
32676 in a process by using the @samp{p@var{pid}.-1} form of the
32687 @item vFlashWrite:@var{addr}:@var{XX@dots{}}
32688 @cindex @samp{vFlashWrite} packet
32689 Direct the stub to write data to flash address @var{addr}. The data
32690 is passed in binary form using the same encoding as for the @samp{X}
32691 packet (@pxref{Binary Data}). The memory ranges specified by
32692 @samp{vFlashWrite} packets preceding a @samp{vFlashDone} packet must
32693 not overlap, and must appear in order of increasing addresses
32694 (although @samp{vFlashErase} packets for higher addresses may already
32695 have been received; the ordering is guaranteed only between
32696 @samp{vFlashWrite} packets). If a packet writes to an address that was
32697 neither erased by a preceding @samp{vFlashErase} packet nor by some other
32698 target-specific method, the results are unpredictable.
32706 for vFlashWrite addressing non-flash memory
32712 @cindex @samp{vFlashDone} packet
32713 Indicate to the stub that flash programming operation is finished.
32714 The stub is permitted to delay or batch the effects of a group of
32715 @samp{vFlashErase} and @samp{vFlashWrite} packets until a
32716 @samp{vFlashDone} packet is received. The contents of the affected
32717 regions of flash memory are unpredictable until the @samp{vFlashDone}
32718 request is completed.
32720 @item vKill;@var{pid}
32721 @cindex @samp{vKill} packet
32722 Kill the process with the specified process ID. @var{pid} is a
32723 hexadecimal integer identifying the process. This packet is used in
32724 preference to @samp{k} when multiprocess protocol extensions are
32725 supported; see @ref{multiprocess extensions}.
32735 @item vRun;@var{filename}@r{[};@var{argument}@r{]}@dots{}
32736 @cindex @samp{vRun} packet
32737 Run the program @var{filename}, passing it each @var{argument} on its
32738 command line. The file and arguments are hex-encoded strings. If
32739 @var{filename} is an empty string, the stub may use a default program
32740 (e.g.@: the last program run). The program is created in the stopped
32743 @c FIXME: What about non-stop mode?
32745 This packet is only available in extended mode (@pxref{extended mode}).
32751 @item @r{Any stop packet}
32752 for success (@pxref{Stop Reply Packets})
32756 @anchor{vStopped packet}
32757 @cindex @samp{vStopped} packet
32759 In non-stop mode (@pxref{Remote Non-Stop}), acknowledge a previous stop
32760 reply and prompt for the stub to report another one.
32764 @item @r{Any stop packet}
32765 if there is another unreported stop event (@pxref{Stop Reply Packets})
32767 if there are no unreported stop events
32770 @item X @var{addr},@var{length}:@var{XX@dots{}}
32772 @cindex @samp{X} packet
32773 Write data to memory, where the data is transmitted in binary.
32774 @var{addr} is address, @var{length} is number of bytes,
32775 @samp{@var{XX}@dots{}} is binary data (@pxref{Binary Data}).
32785 @item z @var{type},@var{addr},@var{kind}
32786 @itemx Z @var{type},@var{addr},@var{kind}
32787 @anchor{insert breakpoint or watchpoint packet}
32788 @cindex @samp{z} packet
32789 @cindex @samp{Z} packets
32790 Insert (@samp{Z}) or remove (@samp{z}) a @var{type} breakpoint or
32791 watchpoint starting at address @var{address} of kind @var{kind}.
32793 Each breakpoint and watchpoint packet @var{type} is documented
32796 @emph{Implementation notes: A remote target shall return an empty string
32797 for an unrecognized breakpoint or watchpoint packet @var{type}. A
32798 remote target shall support either both or neither of a given
32799 @samp{Z@var{type}@dots{}} and @samp{z@var{type}@dots{}} packet pair. To
32800 avoid potential problems with duplicate packets, the operations should
32801 be implemented in an idempotent way.}
32803 @item z0,@var{addr},@var{kind}
32804 @itemx Z0,@var{addr},@var{kind}
32805 @cindex @samp{z0} packet
32806 @cindex @samp{Z0} packet
32807 Insert (@samp{Z0}) or remove (@samp{z0}) a memory breakpoint at address
32808 @var{addr} of type @var{kind}.
32810 A memory breakpoint is implemented by replacing the instruction at
32811 @var{addr} with a software breakpoint or trap instruction. The
32812 @var{kind} is target-specific and typically indicates the size of
32813 the breakpoint in bytes that should be inserted. E.g., the @sc{arm}
32814 and @sc{mips} can insert either a 2 or 4 byte breakpoint. Some
32815 architectures have additional meanings for @var{kind};
32816 see @ref{Architecture-Specific Protocol Details}.
32818 @emph{Implementation note: It is possible for a target to copy or move
32819 code that contains memory breakpoints (e.g., when implementing
32820 overlays). The behavior of this packet, in the presence of such a
32821 target, is not defined.}
32833 @item z1,@var{addr},@var{kind}
32834 @itemx Z1,@var{addr},@var{kind}
32835 @cindex @samp{z1} packet
32836 @cindex @samp{Z1} packet
32837 Insert (@samp{Z1}) or remove (@samp{z1}) a hardware breakpoint at
32838 address @var{addr}.
32840 A hardware breakpoint is implemented using a mechanism that is not
32841 dependant on being able to modify the target's memory. @var{kind}
32842 has the same meaning as in @samp{Z0} packets.
32844 @emph{Implementation note: A hardware breakpoint is not affected by code
32857 @item z2,@var{addr},@var{kind}
32858 @itemx Z2,@var{addr},@var{kind}
32859 @cindex @samp{z2} packet
32860 @cindex @samp{Z2} packet
32861 Insert (@samp{Z2}) or remove (@samp{z2}) a write watchpoint at @var{addr}.
32862 @var{kind} is interpreted as the number of bytes to watch.
32874 @item z3,@var{addr},@var{kind}
32875 @itemx Z3,@var{addr},@var{kind}
32876 @cindex @samp{z3} packet
32877 @cindex @samp{Z3} packet
32878 Insert (@samp{Z3}) or remove (@samp{z3}) a read watchpoint at @var{addr}.
32879 @var{kind} is interpreted as the number of bytes to watch.
32891 @item z4,@var{addr},@var{kind}
32892 @itemx Z4,@var{addr},@var{kind}
32893 @cindex @samp{z4} packet
32894 @cindex @samp{Z4} packet
32895 Insert (@samp{Z4}) or remove (@samp{z4}) an access watchpoint at @var{addr}.
32896 @var{kind} is interpreted as the number of bytes to watch.
32910 @node Stop Reply Packets
32911 @section Stop Reply Packets
32912 @cindex stop reply packets
32914 The @samp{C}, @samp{c}, @samp{S}, @samp{s}, @samp{vCont},
32915 @samp{vAttach}, @samp{vRun}, @samp{vStopped}, and @samp{?} packets can
32916 receive any of the below as a reply. Except for @samp{?}
32917 and @samp{vStopped}, that reply is only returned
32918 when the target halts. In the below the exact meaning of @dfn{signal
32919 number} is defined by the header @file{include/gdb/signals.h} in the
32920 @value{GDBN} source code.
32922 As in the description of request packets, we include spaces in the
32923 reply templates for clarity; these are not part of the reply packet's
32924 syntax. No @value{GDBN} stop reply packet uses spaces to separate its
32930 The program received signal number @var{AA} (a two-digit hexadecimal
32931 number). This is equivalent to a @samp{T} response with no
32932 @var{n}:@var{r} pairs.
32934 @item T @var{AA} @var{n1}:@var{r1};@var{n2}:@var{r2};@dots{}
32935 @cindex @samp{T} packet reply
32936 The program received signal number @var{AA} (a two-digit hexadecimal
32937 number). This is equivalent to an @samp{S} response, except that the
32938 @samp{@var{n}:@var{r}} pairs can carry values of important registers
32939 and other information directly in the stop reply packet, reducing
32940 round-trip latency. Single-step and breakpoint traps are reported
32941 this way. Each @samp{@var{n}:@var{r}} pair is interpreted as follows:
32945 If @var{n} is a hexadecimal number, it is a register number, and the
32946 corresponding @var{r} gives that register's value. @var{r} is a
32947 series of bytes in target byte order, with each byte given by a
32948 two-digit hex number.
32951 If @var{n} is @samp{thread}, then @var{r} is the @var{thread-id} of
32952 the stopped thread, as specified in @ref{thread-id syntax}.
32955 If @var{n} is @samp{core}, then @var{r} is the hexadecimal number of
32956 the core on which the stop event was detected.
32959 If @var{n} is a recognized @dfn{stop reason}, it describes a more
32960 specific event that stopped the target. The currently defined stop
32961 reasons are listed below. @var{aa} should be @samp{05}, the trap
32962 signal. At most one stop reason should be present.
32965 Otherwise, @value{GDBN} should ignore this @samp{@var{n}:@var{r}} pair
32966 and go on to the next; this allows us to extend the protocol in the
32970 The currently defined stop reasons are:
32976 The packet indicates a watchpoint hit, and @var{r} is the data address, in
32979 @cindex shared library events, remote reply
32981 The packet indicates that the loaded libraries have changed.
32982 @value{GDBN} should use @samp{qXfer:libraries:read} to fetch a new
32983 list of loaded libraries. @var{r} is ignored.
32985 @cindex replay log events, remote reply
32987 The packet indicates that the target cannot continue replaying
32988 logged execution events, because it has reached the end (or the
32989 beginning when executing backward) of the log. The value of @var{r}
32990 will be either @samp{begin} or @samp{end}. @xref{Reverse Execution},
32991 for more information.
32995 @itemx W @var{AA} ; process:@var{pid}
32996 The process exited, and @var{AA} is the exit status. This is only
32997 applicable to certain targets.
32999 The second form of the response, including the process ID of the exited
33000 process, can be used only when @value{GDBN} has reported support for
33001 multiprocess protocol extensions; see @ref{multiprocess extensions}.
33002 The @var{pid} is formatted as a big-endian hex string.
33005 @itemx X @var{AA} ; process:@var{pid}
33006 The process terminated with signal @var{AA}.
33008 The second form of the response, including the process ID of the
33009 terminated process, can be used only when @value{GDBN} has reported
33010 support for multiprocess protocol extensions; see @ref{multiprocess
33011 extensions}. The @var{pid} is formatted as a big-endian hex string.
33013 @item O @var{XX}@dots{}
33014 @samp{@var{XX}@dots{}} is hex encoding of @sc{ascii} data, to be
33015 written as the program's console output. This can happen at any time
33016 while the program is running and the debugger should continue to wait
33017 for @samp{W}, @samp{T}, etc. This reply is not permitted in non-stop mode.
33019 @item F @var{call-id},@var{parameter}@dots{}
33020 @var{call-id} is the identifier which says which host system call should
33021 be called. This is just the name of the function. Translation into the
33022 correct system call is only applicable as it's defined in @value{GDBN}.
33023 @xref{File-I/O Remote Protocol Extension}, for a list of implemented
33026 @samp{@var{parameter}@dots{}} is a list of parameters as defined for
33027 this very system call.
33029 The target replies with this packet when it expects @value{GDBN} to
33030 call a host system call on behalf of the target. @value{GDBN} replies
33031 with an appropriate @samp{F} packet and keeps up waiting for the next
33032 reply packet from the target. The latest @samp{C}, @samp{c}, @samp{S}
33033 or @samp{s} action is expected to be continued. @xref{File-I/O Remote
33034 Protocol Extension}, for more details.
33038 @node General Query Packets
33039 @section General Query Packets
33040 @cindex remote query requests
33042 Packets starting with @samp{q} are @dfn{general query packets};
33043 packets starting with @samp{Q} are @dfn{general set packets}. General
33044 query and set packets are a semi-unified form for retrieving and
33045 sending information to and from the stub.
33047 The initial letter of a query or set packet is followed by a name
33048 indicating what sort of thing the packet applies to. For example,
33049 @value{GDBN} may use a @samp{qSymbol} packet to exchange symbol
33050 definitions with the stub. These packet names follow some
33055 The name must not contain commas, colons or semicolons.
33057 Most @value{GDBN} query and set packets have a leading upper case
33060 The names of custom vendor packets should use a company prefix, in
33061 lower case, followed by a period. For example, packets designed at
33062 the Acme Corporation might begin with @samp{qacme.foo} (for querying
33063 foos) or @samp{Qacme.bar} (for setting bars).
33066 The name of a query or set packet should be separated from any
33067 parameters by a @samp{:}; the parameters themselves should be
33068 separated by @samp{,} or @samp{;}. Stubs must be careful to match the
33069 full packet name, and check for a separator or the end of the packet,
33070 in case two packet names share a common prefix. New packets should not begin
33071 with @samp{qC}, @samp{qP}, or @samp{qL}@footnote{The @samp{qP} and @samp{qL}
33072 packets predate these conventions, and have arguments without any terminator
33073 for the packet name; we suspect they are in widespread use in places that
33074 are difficult to upgrade. The @samp{qC} packet has no arguments, but some
33075 existing stubs (e.g.@: RedBoot) are known to not check for the end of the
33078 Like the descriptions of the other packets, each description here
33079 has a template showing the packet's overall syntax, followed by an
33080 explanation of the packet's meaning. We include spaces in some of the
33081 templates for clarity; these are not part of the packet's syntax. No
33082 @value{GDBN} packet uses spaces to separate its components.
33084 Here are the currently defined query and set packets:
33088 @item QAllow:@var{op}:@var{val}@dots{}
33089 @cindex @samp{QAllow} packet
33090 Specify which operations @value{GDBN} expects to request of the
33091 target, as a semicolon-separated list of operation name and value
33092 pairs. Possible values for @var{op} include @samp{WriteReg},
33093 @samp{WriteMem}, @samp{InsertBreak}, @samp{InsertTrace},
33094 @samp{InsertFastTrace}, and @samp{Stop}. @var{val} is either 0,
33095 indicating that @value{GDBN} will not request the operation, or 1,
33096 indicating that it may. (The target can then use this to set up its
33097 own internals optimally, for instance if the debugger never expects to
33098 insert breakpoints, it may not need to install its own trap handler.)
33101 @cindex current thread, remote request
33102 @cindex @samp{qC} packet
33103 Return the current thread ID.
33107 @item QC @var{thread-id}
33108 Where @var{thread-id} is a thread ID as documented in
33109 @ref{thread-id syntax}.
33110 @item @r{(anything else)}
33111 Any other reply implies the old thread ID.
33114 @item qCRC:@var{addr},@var{length}
33115 @cindex CRC of memory block, remote request
33116 @cindex @samp{qCRC} packet
33117 Compute the CRC checksum of a block of memory using CRC-32 defined in
33118 IEEE 802.3. The CRC is computed byte at a time, taking the most
33119 significant bit of each byte first. The initial pattern code
33120 @code{0xffffffff} is used to ensure leading zeros affect the CRC.
33122 @emph{Note:} This is the same CRC used in validating separate debug
33123 files (@pxref{Separate Debug Files, , Debugging Information in Separate
33124 Files}). However the algorithm is slightly different. When validating
33125 separate debug files, the CRC is computed taking the @emph{least}
33126 significant bit of each byte first, and the final result is inverted to
33127 detect trailing zeros.
33132 An error (such as memory fault)
33133 @item C @var{crc32}
33134 The specified memory region's checksum is @var{crc32}.
33138 @itemx qsThreadInfo
33139 @cindex list active threads, remote request
33140 @cindex @samp{qfThreadInfo} packet
33141 @cindex @samp{qsThreadInfo} packet
33142 Obtain a list of all active thread IDs from the target (OS). Since there
33143 may be too many active threads to fit into one reply packet, this query
33144 works iteratively: it may require more than one query/reply sequence to
33145 obtain the entire list of threads. The first query of the sequence will
33146 be the @samp{qfThreadInfo} query; subsequent queries in the
33147 sequence will be the @samp{qsThreadInfo} query.
33149 NOTE: This packet replaces the @samp{qL} query (see below).
33153 @item m @var{thread-id}
33155 @item m @var{thread-id},@var{thread-id}@dots{}
33156 a comma-separated list of thread IDs
33158 (lower case letter @samp{L}) denotes end of list.
33161 In response to each query, the target will reply with a list of one or
33162 more thread IDs, separated by commas.
33163 @value{GDBN} will respond to each reply with a request for more thread
33164 ids (using the @samp{qs} form of the query), until the target responds
33165 with @samp{l} (lower-case ell, for @dfn{last}).
33166 Refer to @ref{thread-id syntax}, for the format of the @var{thread-id}
33169 @item qGetTLSAddr:@var{thread-id},@var{offset},@var{lm}
33170 @cindex get thread-local storage address, remote request
33171 @cindex @samp{qGetTLSAddr} packet
33172 Fetch the address associated with thread local storage specified
33173 by @var{thread-id}, @var{offset}, and @var{lm}.
33175 @var{thread-id} is the thread ID associated with the
33176 thread for which to fetch the TLS address. @xref{thread-id syntax}.
33178 @var{offset} is the (big endian, hex encoded) offset associated with the
33179 thread local variable. (This offset is obtained from the debug
33180 information associated with the variable.)
33182 @var{lm} is the (big endian, hex encoded) OS/ABI-specific encoding of the
33183 load module associated with the thread local storage. For example,
33184 a @sc{gnu}/Linux system will pass the link map address of the shared
33185 object associated with the thread local storage under consideration.
33186 Other operating environments may choose to represent the load module
33187 differently, so the precise meaning of this parameter will vary.
33191 @item @var{XX}@dots{}
33192 Hex encoded (big endian) bytes representing the address of the thread
33193 local storage requested.
33196 An error occurred. @var{nn} are hex digits.
33199 An empty reply indicates that @samp{qGetTLSAddr} is not supported by the stub.
33202 @item qGetTIBAddr:@var{thread-id}
33203 @cindex get thread information block address
33204 @cindex @samp{qGetTIBAddr} packet
33205 Fetch address of the Windows OS specific Thread Information Block.
33207 @var{thread-id} is the thread ID associated with the thread.
33211 @item @var{XX}@dots{}
33212 Hex encoded (big endian) bytes representing the linear address of the
33213 thread information block.
33216 An error occured. This means that either the thread was not found, or the
33217 address could not be retrieved.
33220 An empty reply indicates that @samp{qGetTIBAddr} is not supported by the stub.
33223 @item qL @var{startflag} @var{threadcount} @var{nextthread}
33224 Obtain thread information from RTOS. Where: @var{startflag} (one hex
33225 digit) is one to indicate the first query and zero to indicate a
33226 subsequent query; @var{threadcount} (two hex digits) is the maximum
33227 number of threads the response packet can contain; and @var{nextthread}
33228 (eight hex digits), for subsequent queries (@var{startflag} is zero), is
33229 returned in the response as @var{argthread}.
33231 Don't use this packet; use the @samp{qfThreadInfo} query instead (see above).
33235 @item qM @var{count} @var{done} @var{argthread} @var{thread}@dots{}
33236 Where: @var{count} (two hex digits) is the number of threads being
33237 returned; @var{done} (one hex digit) is zero to indicate more threads
33238 and one indicates no further threads; @var{argthreadid} (eight hex
33239 digits) is @var{nextthread} from the request packet; @var{thread}@dots{}
33240 is a sequence of thread IDs from the target. @var{threadid} (eight hex
33241 digits). See @code{remote.c:parse_threadlist_response()}.
33245 @cindex section offsets, remote request
33246 @cindex @samp{qOffsets} packet
33247 Get section offsets that the target used when relocating the downloaded
33252 @item Text=@var{xxx};Data=@var{yyy}@r{[};Bss=@var{zzz}@r{]}
33253 Relocate the @code{Text} section by @var{xxx} from its original address.
33254 Relocate the @code{Data} section by @var{yyy} from its original address.
33255 If the object file format provides segment information (e.g.@: @sc{elf}
33256 @samp{PT_LOAD} program headers), @value{GDBN} will relocate entire
33257 segments by the supplied offsets.
33259 @emph{Note: while a @code{Bss} offset may be included in the response,
33260 @value{GDBN} ignores this and instead applies the @code{Data} offset
33261 to the @code{Bss} section.}
33263 @item TextSeg=@var{xxx}@r{[};DataSeg=@var{yyy}@r{]}
33264 Relocate the first segment of the object file, which conventionally
33265 contains program code, to a starting address of @var{xxx}. If
33266 @samp{DataSeg} is specified, relocate the second segment, which
33267 conventionally contains modifiable data, to a starting address of
33268 @var{yyy}. @value{GDBN} will report an error if the object file
33269 does not contain segment information, or does not contain at least
33270 as many segments as mentioned in the reply. Extra segments are
33271 kept at fixed offsets relative to the last relocated segment.
33274 @item qP @var{mode} @var{thread-id}
33275 @cindex thread information, remote request
33276 @cindex @samp{qP} packet
33277 Returns information on @var{thread-id}. Where: @var{mode} is a hex
33278 encoded 32 bit mode; @var{thread-id} is a thread ID
33279 (@pxref{thread-id syntax}).
33281 Don't use this packet; use the @samp{qThreadExtraInfo} query instead
33284 Reply: see @code{remote.c:remote_unpack_thread_info_response()}.
33288 @cindex non-stop mode, remote request
33289 @cindex @samp{QNonStop} packet
33291 Enter non-stop (@samp{QNonStop:1}) or all-stop (@samp{QNonStop:0}) mode.
33292 @xref{Remote Non-Stop}, for more information.
33297 The request succeeded.
33300 An error occurred. @var{nn} are hex digits.
33303 An empty reply indicates that @samp{QNonStop} is not supported by
33307 This packet is not probed by default; the remote stub must request it,
33308 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
33309 Use of this packet is controlled by the @code{set non-stop} command;
33310 @pxref{Non-Stop Mode}.
33312 @item QPassSignals: @var{signal} @r{[};@var{signal}@r{]}@dots{}
33313 @cindex pass signals to inferior, remote request
33314 @cindex @samp{QPassSignals} packet
33315 @anchor{QPassSignals}
33316 Each listed @var{signal} should be passed directly to the inferior process.
33317 Signals are numbered identically to continue packets and stop replies
33318 (@pxref{Stop Reply Packets}). Each @var{signal} list item should be
33319 strictly greater than the previous item. These signals do not need to stop
33320 the inferior, or be reported to @value{GDBN}. All other signals should be
33321 reported to @value{GDBN}. Multiple @samp{QPassSignals} packets do not
33322 combine; any earlier @samp{QPassSignals} list is completely replaced by the
33323 new list. This packet improves performance when using @samp{handle
33324 @var{signal} nostop noprint pass}.
33329 The request succeeded.
33332 An error occurred. @var{nn} are hex digits.
33335 An empty reply indicates that @samp{QPassSignals} is not supported by
33339 Use of this packet is controlled by the @code{set remote pass-signals}
33340 command (@pxref{Remote Configuration, set remote pass-signals}).
33341 This packet is not probed by default; the remote stub must request it,
33342 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
33344 @item qRcmd,@var{command}
33345 @cindex execute remote command, remote request
33346 @cindex @samp{qRcmd} packet
33347 @var{command} (hex encoded) is passed to the local interpreter for
33348 execution. Invalid commands should be reported using the output
33349 string. Before the final result packet, the target may also respond
33350 with a number of intermediate @samp{O@var{output}} console output
33351 packets. @emph{Implementors should note that providing access to a
33352 stubs's interpreter may have security implications}.
33357 A command response with no output.
33359 A command response with the hex encoded output string @var{OUTPUT}.
33361 Indicate a badly formed request.
33363 An empty reply indicates that @samp{qRcmd} is not recognized.
33366 (Note that the @code{qRcmd} packet's name is separated from the
33367 command by a @samp{,}, not a @samp{:}, contrary to the naming
33368 conventions above. Please don't use this packet as a model for new
33371 @item qSearch:memory:@var{address};@var{length};@var{search-pattern}
33372 @cindex searching memory, in remote debugging
33373 @cindex @samp{qSearch:memory} packet
33374 @anchor{qSearch memory}
33375 Search @var{length} bytes at @var{address} for @var{search-pattern}.
33376 @var{address} and @var{length} are encoded in hex.
33377 @var{search-pattern} is a sequence of bytes, hex encoded.
33382 The pattern was not found.
33384 The pattern was found at @var{address}.
33386 A badly formed request or an error was encountered while searching memory.
33388 An empty reply indicates that @samp{qSearch:memory} is not recognized.
33391 @item QStartNoAckMode
33392 @cindex @samp{QStartNoAckMode} packet
33393 @anchor{QStartNoAckMode}
33394 Request that the remote stub disable the normal @samp{+}/@samp{-}
33395 protocol acknowledgments (@pxref{Packet Acknowledgment}).
33400 The stub has switched to no-acknowledgment mode.
33401 @value{GDBN} acknowledges this reponse,
33402 but neither the stub nor @value{GDBN} shall send or expect further
33403 @samp{+}/@samp{-} acknowledgments in the current connection.
33405 An empty reply indicates that the stub does not support no-acknowledgment mode.
33408 @item qSupported @r{[}:@var{gdbfeature} @r{[};@var{gdbfeature}@r{]}@dots{} @r{]}
33409 @cindex supported packets, remote query
33410 @cindex features of the remote protocol
33411 @cindex @samp{qSupported} packet
33412 @anchor{qSupported}
33413 Tell the remote stub about features supported by @value{GDBN}, and
33414 query the stub for features it supports. This packet allows
33415 @value{GDBN} and the remote stub to take advantage of each others'
33416 features. @samp{qSupported} also consolidates multiple feature probes
33417 at startup, to improve @value{GDBN} performance---a single larger
33418 packet performs better than multiple smaller probe packets on
33419 high-latency links. Some features may enable behavior which must not
33420 be on by default, e.g.@: because it would confuse older clients or
33421 stubs. Other features may describe packets which could be
33422 automatically probed for, but are not. These features must be
33423 reported before @value{GDBN} will use them. This ``default
33424 unsupported'' behavior is not appropriate for all packets, but it
33425 helps to keep the initial connection time under control with new
33426 versions of @value{GDBN} which support increasing numbers of packets.
33430 @item @var{stubfeature} @r{[};@var{stubfeature}@r{]}@dots{}
33431 The stub supports or does not support each returned @var{stubfeature},
33432 depending on the form of each @var{stubfeature} (see below for the
33435 An empty reply indicates that @samp{qSupported} is not recognized,
33436 or that no features needed to be reported to @value{GDBN}.
33439 The allowed forms for each feature (either a @var{gdbfeature} in the
33440 @samp{qSupported} packet, or a @var{stubfeature} in the response)
33444 @item @var{name}=@var{value}
33445 The remote protocol feature @var{name} is supported, and associated
33446 with the specified @var{value}. The format of @var{value} depends
33447 on the feature, but it must not include a semicolon.
33449 The remote protocol feature @var{name} is supported, and does not
33450 need an associated value.
33452 The remote protocol feature @var{name} is not supported.
33454 The remote protocol feature @var{name} may be supported, and
33455 @value{GDBN} should auto-detect support in some other way when it is
33456 needed. This form will not be used for @var{gdbfeature} notifications,
33457 but may be used for @var{stubfeature} responses.
33460 Whenever the stub receives a @samp{qSupported} request, the
33461 supplied set of @value{GDBN} features should override any previous
33462 request. This allows @value{GDBN} to put the stub in a known
33463 state, even if the stub had previously been communicating with
33464 a different version of @value{GDBN}.
33466 The following values of @var{gdbfeature} (for the packet sent by @value{GDBN})
33471 This feature indicates whether @value{GDBN} supports multiprocess
33472 extensions to the remote protocol. @value{GDBN} does not use such
33473 extensions unless the stub also reports that it supports them by
33474 including @samp{multiprocess+} in its @samp{qSupported} reply.
33475 @xref{multiprocess extensions}, for details.
33478 This feature indicates that @value{GDBN} supports the XML target
33479 description. If the stub sees @samp{xmlRegisters=} with target
33480 specific strings separated by a comma, it will report register
33484 This feature indicates whether @value{GDBN} supports the
33485 @samp{qRelocInsn} packet (@pxref{Tracepoint Packets,,Relocate
33486 instruction reply packet}).
33489 Stubs should ignore any unknown values for
33490 @var{gdbfeature}. Any @value{GDBN} which sends a @samp{qSupported}
33491 packet supports receiving packets of unlimited length (earlier
33492 versions of @value{GDBN} may reject overly long responses). Additional values
33493 for @var{gdbfeature} may be defined in the future to let the stub take
33494 advantage of new features in @value{GDBN}, e.g.@: incompatible
33495 improvements in the remote protocol---the @samp{multiprocess} feature is
33496 an example of such a feature. The stub's reply should be independent
33497 of the @var{gdbfeature} entries sent by @value{GDBN}; first @value{GDBN}
33498 describes all the features it supports, and then the stub replies with
33499 all the features it supports.
33501 Similarly, @value{GDBN} will silently ignore unrecognized stub feature
33502 responses, as long as each response uses one of the standard forms.
33504 Some features are flags. A stub which supports a flag feature
33505 should respond with a @samp{+} form response. Other features
33506 require values, and the stub should respond with an @samp{=}
33509 Each feature has a default value, which @value{GDBN} will use if
33510 @samp{qSupported} is not available or if the feature is not mentioned
33511 in the @samp{qSupported} response. The default values are fixed; a
33512 stub is free to omit any feature responses that match the defaults.
33514 Not all features can be probed, but for those which can, the probing
33515 mechanism is useful: in some cases, a stub's internal
33516 architecture may not allow the protocol layer to know some information
33517 about the underlying target in advance. This is especially common in
33518 stubs which may be configured for multiple targets.
33520 These are the currently defined stub features and their properties:
33522 @multitable @columnfractions 0.35 0.2 0.12 0.2
33523 @c NOTE: The first row should be @headitem, but we do not yet require
33524 @c a new enough version of Texinfo (4.7) to use @headitem.
33526 @tab Value Required
33530 @item @samp{PacketSize}
33535 @item @samp{qXfer:auxv:read}
33540 @item @samp{qXfer:features:read}
33545 @item @samp{qXfer:libraries:read}
33550 @item @samp{qXfer:memory-map:read}
33555 @item @samp{qXfer:sdata:read}
33560 @item @samp{qXfer:spu:read}
33565 @item @samp{qXfer:spu:write}
33570 @item @samp{qXfer:siginfo:read}
33575 @item @samp{qXfer:siginfo:write}
33580 @item @samp{qXfer:threads:read}
33585 @item @samp{qXfer:traceframe-info:read}
33591 @item @samp{QNonStop}
33596 @item @samp{QPassSignals}
33601 @item @samp{QStartNoAckMode}
33606 @item @samp{multiprocess}
33611 @item @samp{ConditionalTracepoints}
33616 @item @samp{ReverseContinue}
33621 @item @samp{ReverseStep}
33626 @item @samp{TracepointSource}
33631 @item @samp{QAllow}
33636 @item @samp{EnableDisableTracepoints}
33643 These are the currently defined stub features, in more detail:
33646 @cindex packet size, remote protocol
33647 @item PacketSize=@var{bytes}
33648 The remote stub can accept packets up to at least @var{bytes} in
33649 length. @value{GDBN} will send packets up to this size for bulk
33650 transfers, and will never send larger packets. This is a limit on the
33651 data characters in the packet, including the frame and checksum.
33652 There is no trailing NUL byte in a remote protocol packet; if the stub
33653 stores packets in a NUL-terminated format, it should allow an extra
33654 byte in its buffer for the NUL. If this stub feature is not supported,
33655 @value{GDBN} guesses based on the size of the @samp{g} packet response.
33657 @item qXfer:auxv:read
33658 The remote stub understands the @samp{qXfer:auxv:read} packet
33659 (@pxref{qXfer auxiliary vector read}).
33661 @item qXfer:features:read
33662 The remote stub understands the @samp{qXfer:features:read} packet
33663 (@pxref{qXfer target description read}).
33665 @item qXfer:libraries:read
33666 The remote stub understands the @samp{qXfer:libraries:read} packet
33667 (@pxref{qXfer library list read}).
33669 @item qXfer:memory-map:read
33670 The remote stub understands the @samp{qXfer:memory-map:read} packet
33671 (@pxref{qXfer memory map read}).
33673 @item qXfer:sdata:read
33674 The remote stub understands the @samp{qXfer:sdata:read} packet
33675 (@pxref{qXfer sdata read}).
33677 @item qXfer:spu:read
33678 The remote stub understands the @samp{qXfer:spu:read} packet
33679 (@pxref{qXfer spu read}).
33681 @item qXfer:spu:write
33682 The remote stub understands the @samp{qXfer:spu:write} packet
33683 (@pxref{qXfer spu write}).
33685 @item qXfer:siginfo:read
33686 The remote stub understands the @samp{qXfer:siginfo:read} packet
33687 (@pxref{qXfer siginfo read}).
33689 @item qXfer:siginfo:write
33690 The remote stub understands the @samp{qXfer:siginfo:write} packet
33691 (@pxref{qXfer siginfo write}).
33693 @item qXfer:threads:read
33694 The remote stub understands the @samp{qXfer:threads:read} packet
33695 (@pxref{qXfer threads read}).
33697 @item qXfer:traceframe-info:read
33698 The remote stub understands the @samp{qXfer:traceframe-info:read}
33699 packet (@pxref{qXfer traceframe info read}).
33702 The remote stub understands the @samp{QNonStop} packet
33703 (@pxref{QNonStop}).
33706 The remote stub understands the @samp{QPassSignals} packet
33707 (@pxref{QPassSignals}).
33709 @item QStartNoAckMode
33710 The remote stub understands the @samp{QStartNoAckMode} packet and
33711 prefers to operate in no-acknowledgment mode. @xref{Packet Acknowledgment}.
33714 @anchor{multiprocess extensions}
33715 @cindex multiprocess extensions, in remote protocol
33716 The remote stub understands the multiprocess extensions to the remote
33717 protocol syntax. The multiprocess extensions affect the syntax of
33718 thread IDs in both packets and replies (@pxref{thread-id syntax}), and
33719 add process IDs to the @samp{D} packet and @samp{W} and @samp{X}
33720 replies. Note that reporting this feature indicates support for the
33721 syntactic extensions only, not that the stub necessarily supports
33722 debugging of more than one process at a time. The stub must not use
33723 multiprocess extensions in packet replies unless @value{GDBN} has also
33724 indicated it supports them in its @samp{qSupported} request.
33726 @item qXfer:osdata:read
33727 The remote stub understands the @samp{qXfer:osdata:read} packet
33728 ((@pxref{qXfer osdata read}).
33730 @item ConditionalTracepoints
33731 The remote stub accepts and implements conditional expressions defined
33732 for tracepoints (@pxref{Tracepoint Conditions}).
33734 @item ReverseContinue
33735 The remote stub accepts and implements the reverse continue packet
33739 The remote stub accepts and implements the reverse step packet
33742 @item TracepointSource
33743 The remote stub understands the @samp{QTDPsrc} packet that supplies
33744 the source form of tracepoint definitions.
33747 The remote stub understands the @samp{QAllow} packet.
33749 @item StaticTracepoint
33750 @cindex static tracepoints, in remote protocol
33751 The remote stub supports static tracepoints.
33753 @item EnableDisableTracepoints
33754 The remote stub supports the @samp{QTEnable} (@pxref{QTEnable}) and
33755 @samp{QTDisable} (@pxref{QTDisable}) packets that allow tracepoints
33756 to be enabled and disabled while a trace experiment is running.
33761 @cindex symbol lookup, remote request
33762 @cindex @samp{qSymbol} packet
33763 Notify the target that @value{GDBN} is prepared to serve symbol lookup
33764 requests. Accept requests from the target for the values of symbols.
33769 The target does not need to look up any (more) symbols.
33770 @item qSymbol:@var{sym_name}
33771 The target requests the value of symbol @var{sym_name} (hex encoded).
33772 @value{GDBN} may provide the value by using the
33773 @samp{qSymbol:@var{sym_value}:@var{sym_name}} message, described
33777 @item qSymbol:@var{sym_value}:@var{sym_name}
33778 Set the value of @var{sym_name} to @var{sym_value}.
33780 @var{sym_name} (hex encoded) is the name of a symbol whose value the
33781 target has previously requested.
33783 @var{sym_value} (hex) is the value for symbol @var{sym_name}. If
33784 @value{GDBN} cannot supply a value for @var{sym_name}, then this field
33790 The target does not need to look up any (more) symbols.
33791 @item qSymbol:@var{sym_name}
33792 The target requests the value of a new symbol @var{sym_name} (hex
33793 encoded). @value{GDBN} will continue to supply the values of symbols
33794 (if available), until the target ceases to request them.
33799 @item QTDisconnected
33806 @xref{Tracepoint Packets}.
33808 @item qThreadExtraInfo,@var{thread-id}
33809 @cindex thread attributes info, remote request
33810 @cindex @samp{qThreadExtraInfo} packet
33811 Obtain a printable string description of a thread's attributes from
33812 the target OS. @var{thread-id} is a thread ID;
33813 see @ref{thread-id syntax}. This
33814 string may contain anything that the target OS thinks is interesting
33815 for @value{GDBN} to tell the user about the thread. The string is
33816 displayed in @value{GDBN}'s @code{info threads} display. Some
33817 examples of possible thread extra info strings are @samp{Runnable}, or
33818 @samp{Blocked on Mutex}.
33822 @item @var{XX}@dots{}
33823 Where @samp{@var{XX}@dots{}} is a hex encoding of @sc{ascii} data,
33824 comprising the printable string containing the extra information about
33825 the thread's attributes.
33828 (Note that the @code{qThreadExtraInfo} packet's name is separated from
33829 the command by a @samp{,}, not a @samp{:}, contrary to the naming
33830 conventions above. Please don't use this packet as a model for new
33847 @xref{Tracepoint Packets}.
33849 @item qXfer:@var{object}:read:@var{annex}:@var{offset},@var{length}
33850 @cindex read special object, remote request
33851 @cindex @samp{qXfer} packet
33852 @anchor{qXfer read}
33853 Read uninterpreted bytes from the target's special data area
33854 identified by the keyword @var{object}. Request @var{length} bytes
33855 starting at @var{offset} bytes into the data. The content and
33856 encoding of @var{annex} is specific to @var{object}; it can supply
33857 additional details about what data to access.
33859 Here are the specific requests of this form defined so far. All
33860 @samp{qXfer:@var{object}:read:@dots{}} requests use the same reply
33861 formats, listed below.
33864 @item qXfer:auxv:read::@var{offset},@var{length}
33865 @anchor{qXfer auxiliary vector read}
33866 Access the target's @dfn{auxiliary vector}. @xref{OS Information,
33867 auxiliary vector}. Note @var{annex} must be empty.
33869 This packet is not probed by default; the remote stub must request it,
33870 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
33872 @item qXfer:features:read:@var{annex}:@var{offset},@var{length}
33873 @anchor{qXfer target description read}
33874 Access the @dfn{target description}. @xref{Target Descriptions}. The
33875 annex specifies which XML document to access. The main description is
33876 always loaded from the @samp{target.xml} annex.
33878 This packet is not probed by default; the remote stub must request it,
33879 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
33881 @item qXfer:libraries:read:@var{annex}:@var{offset},@var{length}
33882 @anchor{qXfer library list read}
33883 Access the target's list of loaded libraries. @xref{Library List Format}.
33884 The annex part of the generic @samp{qXfer} packet must be empty
33885 (@pxref{qXfer read}).
33887 Targets which maintain a list of libraries in the program's memory do
33888 not need to implement this packet; it is designed for platforms where
33889 the operating system manages the list of loaded libraries.
33891 This packet is not probed by default; the remote stub must request it,
33892 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
33894 @item qXfer:memory-map:read::@var{offset},@var{length}
33895 @anchor{qXfer memory map read}
33896 Access the target's @dfn{memory-map}. @xref{Memory Map Format}. The
33897 annex part of the generic @samp{qXfer} packet must be empty
33898 (@pxref{qXfer read}).
33900 This packet is not probed by default; the remote stub must request it,
33901 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
33903 @item qXfer:sdata:read::@var{offset},@var{length}
33904 @anchor{qXfer sdata read}
33906 Read contents of the extra collected static tracepoint marker
33907 information. The annex part of the generic @samp{qXfer} packet must
33908 be empty (@pxref{qXfer read}). @xref{Tracepoint Actions,,Tracepoint
33911 This packet is not probed by default; the remote stub must request it,
33912 by supplying an appropriate @samp{qSupported} response
33913 (@pxref{qSupported}).
33915 @item qXfer:siginfo:read::@var{offset},@var{length}
33916 @anchor{qXfer siginfo read}
33917 Read contents of the extra signal information on the target
33918 system. The annex part of the generic @samp{qXfer} packet must be
33919 empty (@pxref{qXfer read}).
33921 This packet is not probed by default; the remote stub must request it,
33922 by supplying an appropriate @samp{qSupported} response
33923 (@pxref{qSupported}).
33925 @item qXfer:spu:read:@var{annex}:@var{offset},@var{length}
33926 @anchor{qXfer spu read}
33927 Read contents of an @code{spufs} file on the target system. The
33928 annex specifies which file to read; it must be of the form
33929 @file{@var{id}/@var{name}}, where @var{id} specifies an SPU context ID
33930 in the target process, and @var{name} identifes the @code{spufs} file
33931 in that context to be accessed.
33933 This packet is not probed by default; the remote stub must request it,
33934 by supplying an appropriate @samp{qSupported} response
33935 (@pxref{qSupported}).
33937 @item qXfer:threads:read::@var{offset},@var{length}
33938 @anchor{qXfer threads read}
33939 Access the list of threads on target. @xref{Thread List Format}. The
33940 annex part of the generic @samp{qXfer} packet must be empty
33941 (@pxref{qXfer read}).
33943 This packet is not probed by default; the remote stub must request it,
33944 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
33946 @item qXfer:traceframe-info:read::@var{offset},@var{length}
33947 @anchor{qXfer traceframe info read}
33949 Return a description of the current traceframe's contents.
33950 @xref{Traceframe Info Format}. The annex part of the generic
33951 @samp{qXfer} packet must be empty (@pxref{qXfer read}).
33953 This packet is not probed by default; the remote stub must request it,
33954 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
33956 @item qXfer:osdata:read::@var{offset},@var{length}
33957 @anchor{qXfer osdata read}
33958 Access the target's @dfn{operating system information}.
33959 @xref{Operating System Information}.
33966 Data @var{data} (@pxref{Binary Data}) has been read from the
33967 target. There may be more data at a higher address (although
33968 it is permitted to return @samp{m} even for the last valid
33969 block of data, as long as at least one byte of data was read).
33970 @var{data} may have fewer bytes than the @var{length} in the
33974 Data @var{data} (@pxref{Binary Data}) has been read from the target.
33975 There is no more data to be read. @var{data} may have fewer bytes
33976 than the @var{length} in the request.
33979 The @var{offset} in the request is at the end of the data.
33980 There is no more data to be read.
33983 The request was malformed, or @var{annex} was invalid.
33986 The offset was invalid, or there was an error encountered reading the data.
33987 @var{nn} is a hex-encoded @code{errno} value.
33990 An empty reply indicates the @var{object} string was not recognized by
33991 the stub, or that the object does not support reading.
33994 @item qXfer:@var{object}:write:@var{annex}:@var{offset}:@var{data}@dots{}
33995 @cindex write data into object, remote request
33996 @anchor{qXfer write}
33997 Write uninterpreted bytes into the target's special data area
33998 identified by the keyword @var{object}, starting at @var{offset} bytes
33999 into the data. @var{data}@dots{} is the binary-encoded data
34000 (@pxref{Binary Data}) to be written. The content and encoding of @var{annex}
34001 is specific to @var{object}; it can supply additional details about what data
34004 Here are the specific requests of this form defined so far. All
34005 @samp{qXfer:@var{object}:write:@dots{}} requests use the same reply
34006 formats, listed below.
34009 @item qXfer:siginfo:write::@var{offset}:@var{data}@dots{}
34010 @anchor{qXfer siginfo write}
34011 Write @var{data} to the extra signal information on the target system.
34012 The annex part of the generic @samp{qXfer} packet must be
34013 empty (@pxref{qXfer write}).
34015 This packet is not probed by default; the remote stub must request it,
34016 by supplying an appropriate @samp{qSupported} response
34017 (@pxref{qSupported}).
34019 @item qXfer:spu:write:@var{annex}:@var{offset}:@var{data}@dots{}
34020 @anchor{qXfer spu write}
34021 Write @var{data} to an @code{spufs} file on the target system. The
34022 annex specifies which file to write; it must be of the form
34023 @file{@var{id}/@var{name}}, where @var{id} specifies an SPU context ID
34024 in the target process, and @var{name} identifes the @code{spufs} file
34025 in that context to be accessed.
34027 This packet is not probed by default; the remote stub must request it,
34028 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
34034 @var{nn} (hex encoded) is the number of bytes written.
34035 This may be fewer bytes than supplied in the request.
34038 The request was malformed, or @var{annex} was invalid.
34041 The offset was invalid, or there was an error encountered writing the data.
34042 @var{nn} is a hex-encoded @code{errno} value.
34045 An empty reply indicates the @var{object} string was not
34046 recognized by the stub, or that the object does not support writing.
34049 @item qXfer:@var{object}:@var{operation}:@dots{}
34050 Requests of this form may be added in the future. When a stub does
34051 not recognize the @var{object} keyword, or its support for
34052 @var{object} does not recognize the @var{operation} keyword, the stub
34053 must respond with an empty packet.
34055 @item qAttached:@var{pid}
34056 @cindex query attached, remote request
34057 @cindex @samp{qAttached} packet
34058 Return an indication of whether the remote server attached to an
34059 existing process or created a new process. When the multiprocess
34060 protocol extensions are supported (@pxref{multiprocess extensions}),
34061 @var{pid} is an integer in hexadecimal format identifying the target
34062 process. Otherwise, @value{GDBN} will omit the @var{pid} field and
34063 the query packet will be simplified as @samp{qAttached}.
34065 This query is used, for example, to know whether the remote process
34066 should be detached or killed when a @value{GDBN} session is ended with
34067 the @code{quit} command.
34072 The remote server attached to an existing process.
34074 The remote server created a new process.
34076 A badly formed request or an error was encountered.
34081 @node Architecture-Specific Protocol Details
34082 @section Architecture-Specific Protocol Details
34084 This section describes how the remote protocol is applied to specific
34085 target architectures. Also see @ref{Standard Target Features}, for
34086 details of XML target descriptions for each architecture.
34090 @subsubsection Breakpoint Kinds
34092 These breakpoint kinds are defined for the @samp{Z0} and @samp{Z1} packets.
34097 16-bit Thumb mode breakpoint.
34100 32-bit Thumb mode (Thumb-2) breakpoint.
34103 32-bit ARM mode breakpoint.
34109 @subsubsection Register Packet Format
34111 The following @code{g}/@code{G} packets have previously been defined.
34112 In the below, some thirty-two bit registers are transferred as
34113 sixty-four bits. Those registers should be zero/sign extended (which?)
34114 to fill the space allocated. Register bytes are transferred in target
34115 byte order. The two nibbles within a register byte are transferred
34116 most-significant - least-significant.
34122 All registers are transferred as thirty-two bit quantities in the order:
34123 32 general-purpose; sr; lo; hi; bad; cause; pc; 32 floating-point
34124 registers; fsr; fir; fp.
34128 All registers are transferred as sixty-four bit quantities (including
34129 thirty-two bit registers such as @code{sr}). The ordering is the same
34134 @node Tracepoint Packets
34135 @section Tracepoint Packets
34136 @cindex tracepoint packets
34137 @cindex packets, tracepoint
34139 Here we describe the packets @value{GDBN} uses to implement
34140 tracepoints (@pxref{Tracepoints}).
34144 @item QTDP:@var{n}:@var{addr}:@var{ena}:@var{step}:@var{pass}[:F@var{flen}][:X@var{len},@var{bytes}]@r{[}-@r{]}
34145 Create a new tracepoint, number @var{n}, at @var{addr}. If @var{ena}
34146 is @samp{E}, then the tracepoint is enabled; if it is @samp{D}, then
34147 the tracepoint is disabled. @var{step} is the tracepoint's step
34148 count, and @var{pass} is its pass count. If an @samp{F} is present,
34149 then the tracepoint is to be a fast tracepoint, and the @var{flen} is
34150 the number of bytes that the target should copy elsewhere to make room
34151 for the tracepoint. If an @samp{X} is present, it introduces a
34152 tracepoint condition, which consists of a hexadecimal length, followed
34153 by a comma and hex-encoded bytes, in a manner similar to action
34154 encodings as described below. If the trailing @samp{-} is present,
34155 further @samp{QTDP} packets will follow to specify this tracepoint's
34161 The packet was understood and carried out.
34163 @xref{Tracepoint Packets,,Relocate instruction reply packet}.
34165 The packet was not recognized.
34168 @item QTDP:-@var{n}:@var{addr}:@r{[}S@r{]}@var{action}@dots{}@r{[}-@r{]}
34169 Define actions to be taken when a tracepoint is hit. @var{n} and
34170 @var{addr} must be the same as in the initial @samp{QTDP} packet for
34171 this tracepoint. This packet may only be sent immediately after
34172 another @samp{QTDP} packet that ended with a @samp{-}. If the
34173 trailing @samp{-} is present, further @samp{QTDP} packets will follow,
34174 specifying more actions for this tracepoint.
34176 In the series of action packets for a given tracepoint, at most one
34177 can have an @samp{S} before its first @var{action}. If such a packet
34178 is sent, it and the following packets define ``while-stepping''
34179 actions. Any prior packets define ordinary actions --- that is, those
34180 taken when the tracepoint is first hit. If no action packet has an
34181 @samp{S}, then all the packets in the series specify ordinary
34182 tracepoint actions.
34184 The @samp{@var{action}@dots{}} portion of the packet is a series of
34185 actions, concatenated without separators. Each action has one of the
34191 Collect the registers whose bits are set in @var{mask}. @var{mask} is
34192 a hexadecimal number whose @var{i}'th bit is set if register number
34193 @var{i} should be collected. (The least significant bit is numbered
34194 zero.) Note that @var{mask} may be any number of digits long; it may
34195 not fit in a 32-bit word.
34197 @item M @var{basereg},@var{offset},@var{len}
34198 Collect @var{len} bytes of memory starting at the address in register
34199 number @var{basereg}, plus @var{offset}. If @var{basereg} is
34200 @samp{-1}, then the range has a fixed address: @var{offset} is the
34201 address of the lowest byte to collect. The @var{basereg},
34202 @var{offset}, and @var{len} parameters are all unsigned hexadecimal
34203 values (the @samp{-1} value for @var{basereg} is a special case).
34205 @item X @var{len},@var{expr}
34206 Evaluate @var{expr}, whose length is @var{len}, and collect memory as
34207 it directs. @var{expr} is an agent expression, as described in
34208 @ref{Agent Expressions}. Each byte of the expression is encoded as a
34209 two-digit hex number in the packet; @var{len} is the number of bytes
34210 in the expression (and thus one-half the number of hex digits in the
34215 Any number of actions may be packed together in a single @samp{QTDP}
34216 packet, as long as the packet does not exceed the maximum packet
34217 length (400 bytes, for many stubs). There may be only one @samp{R}
34218 action per tracepoint, and it must precede any @samp{M} or @samp{X}
34219 actions. Any registers referred to by @samp{M} and @samp{X} actions
34220 must be collected by a preceding @samp{R} action. (The
34221 ``while-stepping'' actions are treated as if they were attached to a
34222 separate tracepoint, as far as these restrictions are concerned.)
34227 The packet was understood and carried out.
34229 @xref{Tracepoint Packets,,Relocate instruction reply packet}.
34231 The packet was not recognized.
34234 @item QTDPsrc:@var{n}:@var{addr}:@var{type}:@var{start}:@var{slen}:@var{bytes}
34235 @cindex @samp{QTDPsrc} packet
34236 Specify a source string of tracepoint @var{n} at address @var{addr}.
34237 This is useful to get accurate reproduction of the tracepoints
34238 originally downloaded at the beginning of the trace run. @var{type}
34239 is the name of the tracepoint part, such as @samp{cond} for the
34240 tracepoint's conditional expression (see below for a list of types), while
34241 @var{bytes} is the string, encoded in hexadecimal.
34243 @var{start} is the offset of the @var{bytes} within the overall source
34244 string, while @var{slen} is the total length of the source string.
34245 This is intended for handling source strings that are longer than will
34246 fit in a single packet.
34247 @c Add detailed example when this info is moved into a dedicated
34248 @c tracepoint descriptions section.
34250 The available string types are @samp{at} for the location,
34251 @samp{cond} for the conditional, and @samp{cmd} for an action command.
34252 @value{GDBN} sends a separate packet for each command in the action
34253 list, in the same order in which the commands are stored in the list.
34255 The target does not need to do anything with source strings except
34256 report them back as part of the replies to the @samp{qTfP}/@samp{qTsP}
34259 Although this packet is optional, and @value{GDBN} will only send it
34260 if the target replies with @samp{TracepointSource} @xref{General
34261 Query Packets}, it makes both disconnected tracing and trace files
34262 much easier to use. Otherwise the user must be careful that the
34263 tracepoints in effect while looking at trace frames are identical to
34264 the ones in effect during the trace run; even a small discrepancy
34265 could cause @samp{tdump} not to work, or a particular trace frame not
34268 @item QTDV:@var{n}:@var{value}
34269 @cindex define trace state variable, remote request
34270 @cindex @samp{QTDV} packet
34271 Create a new trace state variable, number @var{n}, with an initial
34272 value of @var{value}, which is a 64-bit signed integer. Both @var{n}
34273 and @var{value} are encoded as hexadecimal values. @value{GDBN} has
34274 the option of not using this packet for initial values of zero; the
34275 target should simply create the trace state variables as they are
34276 mentioned in expressions.
34278 @item QTFrame:@var{n}
34279 Select the @var{n}'th tracepoint frame from the buffer, and use the
34280 register and memory contents recorded there to answer subsequent
34281 request packets from @value{GDBN}.
34283 A successful reply from the stub indicates that the stub has found the
34284 requested frame. The response is a series of parts, concatenated
34285 without separators, describing the frame we selected. Each part has
34286 one of the following forms:
34290 The selected frame is number @var{n} in the trace frame buffer;
34291 @var{f} is a hexadecimal number. If @var{f} is @samp{-1}, then there
34292 was no frame matching the criteria in the request packet.
34295 The selected trace frame records a hit of tracepoint number @var{t};
34296 @var{t} is a hexadecimal number.
34300 @item QTFrame:pc:@var{addr}
34301 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
34302 currently selected frame whose PC is @var{addr};
34303 @var{addr} is a hexadecimal number.
34305 @item QTFrame:tdp:@var{t}
34306 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
34307 currently selected frame that is a hit of tracepoint @var{t}; @var{t}
34308 is a hexadecimal number.
34310 @item QTFrame:range:@var{start}:@var{end}
34311 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
34312 currently selected frame whose PC is between @var{start} (inclusive)
34313 and @var{end} (inclusive); @var{start} and @var{end} are hexadecimal
34316 @item QTFrame:outside:@var{start}:@var{end}
34317 Like @samp{QTFrame:range:@var{start}:@var{end}}, but select the first
34318 frame @emph{outside} the given range of addresses (exclusive).
34321 Begin the tracepoint experiment. Begin collecting data from
34322 tracepoint hits in the trace frame buffer. This packet supports the
34323 @samp{qRelocInsn} reply (@pxref{Tracepoint Packets,,Relocate
34324 instruction reply packet}).
34327 End the tracepoint experiment. Stop collecting trace frames.
34329 @item QTEnable:@var{n}:@var{addr}
34331 Enable tracepoint @var{n} at address @var{addr} in a started tracepoint
34332 experiment. If the tracepoint was previously disabled, then collection
34333 of data from it will resume.
34335 @item QTDisable:@var{n}:@var{addr}
34337 Disable tracepoint @var{n} at address @var{addr} in a started tracepoint
34338 experiment. No more data will be collected from the tracepoint unless
34339 @samp{QTEnable:@var{n}:@var{addr}} is subsequently issued.
34342 Clear the table of tracepoints, and empty the trace frame buffer.
34344 @item QTro:@var{start1},@var{end1}:@var{start2},@var{end2}:@dots{}
34345 Establish the given ranges of memory as ``transparent''. The stub
34346 will answer requests for these ranges from memory's current contents,
34347 if they were not collected as part of the tracepoint hit.
34349 @value{GDBN} uses this to mark read-only regions of memory, like those
34350 containing program code. Since these areas never change, they should
34351 still have the same contents they did when the tracepoint was hit, so
34352 there's no reason for the stub to refuse to provide their contents.
34354 @item QTDisconnected:@var{value}
34355 Set the choice to what to do with the tracing run when @value{GDBN}
34356 disconnects from the target. A @var{value} of 1 directs the target to
34357 continue the tracing run, while 0 tells the target to stop tracing if
34358 @value{GDBN} is no longer in the picture.
34361 Ask the stub if there is a trace experiment running right now.
34363 The reply has the form:
34367 @item T@var{running}@r{[};@var{field}@r{]}@dots{}
34368 @var{running} is a single digit @code{1} if the trace is presently
34369 running, or @code{0} if not. It is followed by semicolon-separated
34370 optional fields that an agent may use to report additional status.
34374 If the trace is not running, the agent may report any of several
34375 explanations as one of the optional fields:
34380 No trace has been run yet.
34383 The trace was stopped by a user-originated stop command.
34386 The trace stopped because the trace buffer filled up.
34388 @item tdisconnected:0
34389 The trace stopped because @value{GDBN} disconnected from the target.
34391 @item tpasscount:@var{tpnum}
34392 The trace stopped because tracepoint @var{tpnum} exceeded its pass count.
34394 @item terror:@var{text}:@var{tpnum}
34395 The trace stopped because tracepoint @var{tpnum} had an error. The
34396 string @var{text} is available to describe the nature of the error
34397 (for instance, a divide by zero in the condition expression).
34398 @var{text} is hex encoded.
34401 The trace stopped for some other reason.
34405 Additional optional fields supply statistical and other information.
34406 Although not required, they are extremely useful for users monitoring
34407 the progress of a trace run. If a trace has stopped, and these
34408 numbers are reported, they must reflect the state of the just-stopped
34413 @item tframes:@var{n}
34414 The number of trace frames in the buffer.
34416 @item tcreated:@var{n}
34417 The total number of trace frames created during the run. This may
34418 be larger than the trace frame count, if the buffer is circular.
34420 @item tsize:@var{n}
34421 The total size of the trace buffer, in bytes.
34423 @item tfree:@var{n}
34424 The number of bytes still unused in the buffer.
34426 @item circular:@var{n}
34427 The value of the circular trace buffer flag. @code{1} means that the
34428 trace buffer is circular and old trace frames will be discarded if
34429 necessary to make room, @code{0} means that the trace buffer is linear
34432 @item disconn:@var{n}
34433 The value of the disconnected tracing flag. @code{1} means that
34434 tracing will continue after @value{GDBN} disconnects, @code{0} means
34435 that the trace run will stop.
34439 @item qTV:@var{var}
34440 @cindex trace state variable value, remote request
34441 @cindex @samp{qTV} packet
34442 Ask the stub for the value of the trace state variable number @var{var}.
34447 The value of the variable is @var{value}. This will be the current
34448 value of the variable if the user is examining a running target, or a
34449 saved value if the variable was collected in the trace frame that the
34450 user is looking at. Note that multiple requests may result in
34451 different reply values, such as when requesting values while the
34452 program is running.
34455 The value of the variable is unknown. This would occur, for example,
34456 if the user is examining a trace frame in which the requested variable
34462 These packets request data about tracepoints that are being used by
34463 the target. @value{GDBN} sends @code{qTfP} to get the first piece
34464 of data, and multiple @code{qTsP} to get additional pieces. Replies
34465 to these packets generally take the form of the @code{QTDP} packets
34466 that define tracepoints. (FIXME add detailed syntax)
34470 These packets request data about trace state variables that are on the
34471 target. @value{GDBN} sends @code{qTfV} to get the first vari of data,
34472 and multiple @code{qTsV} to get additional variables. Replies to
34473 these packets follow the syntax of the @code{QTDV} packets that define
34474 trace state variables.
34478 These packets request data about static tracepoint markers that exist
34479 in the target program. @value{GDBN} sends @code{qTfSTM} to get the
34480 first piece of data, and multiple @code{qTsSTM} to get additional
34481 pieces. Replies to these packets take the following form:
34485 @item m @var{address}:@var{id}:@var{extra}
34487 @item m @var{address}:@var{id}:@var{extra},@var{address}:@var{id}:@var{extra}@dots{}
34488 a comma-separated list of markers
34490 (lower case letter @samp{L}) denotes end of list.
34492 An error occurred. @var{nn} are hex digits.
34494 An empty reply indicates that the request is not supported by the
34498 @var{address} is encoded in hex.
34499 @var{id} and @var{extra} are strings encoded in hex.
34501 In response to each query, the target will reply with a list of one or
34502 more markers, separated by commas. @value{GDBN} will respond to each
34503 reply with a request for more markers (using the @samp{qs} form of the
34504 query), until the target responds with @samp{l} (lower-case ell, for
34507 @item qTSTMat:@var{address}
34508 This packets requests data about static tracepoint markers in the
34509 target program at @var{address}. Replies to this packet follow the
34510 syntax of the @samp{qTfSTM} and @code{qTsSTM} packets that list static
34511 tracepoint markers.
34513 @item QTSave:@var{filename}
34514 This packet directs the target to save trace data to the file name
34515 @var{filename} in the target's filesystem. @var{filename} is encoded
34516 as a hex string; the interpretation of the file name (relative vs
34517 absolute, wild cards, etc) is up to the target.
34519 @item qTBuffer:@var{offset},@var{len}
34520 Return up to @var{len} bytes of the current contents of trace buffer,
34521 starting at @var{offset}. The trace buffer is treated as if it were
34522 a contiguous collection of traceframes, as per the trace file format.
34523 The reply consists as many hex-encoded bytes as the target can deliver
34524 in a packet; it is not an error to return fewer than were asked for.
34525 A reply consisting of just @code{l} indicates that no bytes are
34528 @item QTBuffer:circular:@var{value}
34529 This packet directs the target to use a circular trace buffer if
34530 @var{value} is 1, or a linear buffer if the value is 0.
34534 @subsection Relocate instruction reply packet
34535 When installing fast tracepoints in memory, the target may need to
34536 relocate the instruction currently at the tracepoint address to a
34537 different address in memory. For most instructions, a simple copy is
34538 enough, but, for example, call instructions that implicitly push the
34539 return address on the stack, and relative branches or other
34540 PC-relative instructions require offset adjustment, so that the effect
34541 of executing the instruction at a different address is the same as if
34542 it had executed in the original location.
34544 In response to several of the tracepoint packets, the target may also
34545 respond with a number of intermediate @samp{qRelocInsn} request
34546 packets before the final result packet, to have @value{GDBN} handle
34547 this relocation operation. If a packet supports this mechanism, its
34548 documentation will explicitly say so. See for example the above
34549 descriptions for the @samp{QTStart} and @samp{QTDP} packets. The
34550 format of the request is:
34553 @item qRelocInsn:@var{from};@var{to}
34555 This requests @value{GDBN} to copy instruction at address @var{from}
34556 to address @var{to}, possibly adjusted so that executing the
34557 instruction at @var{to} has the same effect as executing it at
34558 @var{from}. @value{GDBN} writes the adjusted instruction to target
34559 memory starting at @var{to}.
34564 @item qRelocInsn:@var{adjusted_size}
34565 Informs the stub the relocation is complete. @var{adjusted_size} is
34566 the length in bytes of resulting relocated instruction sequence.
34568 A badly formed request was detected, or an error was encountered while
34569 relocating the instruction.
34572 @node Host I/O Packets
34573 @section Host I/O Packets
34574 @cindex Host I/O, remote protocol
34575 @cindex file transfer, remote protocol
34577 The @dfn{Host I/O} packets allow @value{GDBN} to perform I/O
34578 operations on the far side of a remote link. For example, Host I/O is
34579 used to upload and download files to a remote target with its own
34580 filesystem. Host I/O uses the same constant values and data structure
34581 layout as the target-initiated File-I/O protocol. However, the
34582 Host I/O packets are structured differently. The target-initiated
34583 protocol relies on target memory to store parameters and buffers.
34584 Host I/O requests are initiated by @value{GDBN}, and the
34585 target's memory is not involved. @xref{File-I/O Remote Protocol
34586 Extension}, for more details on the target-initiated protocol.
34588 The Host I/O request packets all encode a single operation along with
34589 its arguments. They have this format:
34593 @item vFile:@var{operation}: @var{parameter}@dots{}
34594 @var{operation} is the name of the particular request; the target
34595 should compare the entire packet name up to the second colon when checking
34596 for a supported operation. The format of @var{parameter} depends on
34597 the operation. Numbers are always passed in hexadecimal. Negative
34598 numbers have an explicit minus sign (i.e.@: two's complement is not
34599 used). Strings (e.g.@: filenames) are encoded as a series of
34600 hexadecimal bytes. The last argument to a system call may be a
34601 buffer of escaped binary data (@pxref{Binary Data}).
34605 The valid responses to Host I/O packets are:
34609 @item F @var{result} [, @var{errno}] [; @var{attachment}]
34610 @var{result} is the integer value returned by this operation, usually
34611 non-negative for success and -1 for errors. If an error has occured,
34612 @var{errno} will be included in the result. @var{errno} will have a
34613 value defined by the File-I/O protocol (@pxref{Errno Values}). For
34614 operations which return data, @var{attachment} supplies the data as a
34615 binary buffer. Binary buffers in response packets are escaped in the
34616 normal way (@pxref{Binary Data}). See the individual packet
34617 documentation for the interpretation of @var{result} and
34621 An empty response indicates that this operation is not recognized.
34625 These are the supported Host I/O operations:
34628 @item vFile:open: @var{pathname}, @var{flags}, @var{mode}
34629 Open a file at @var{pathname} and return a file descriptor for it, or
34630 return -1 if an error occurs. @var{pathname} is a string,
34631 @var{flags} is an integer indicating a mask of open flags
34632 (@pxref{Open Flags}), and @var{mode} is an integer indicating a mask
34633 of mode bits to use if the file is created (@pxref{mode_t Values}).
34634 @xref{open}, for details of the open flags and mode values.
34636 @item vFile:close: @var{fd}
34637 Close the open file corresponding to @var{fd} and return 0, or
34638 -1 if an error occurs.
34640 @item vFile:pread: @var{fd}, @var{count}, @var{offset}
34641 Read data from the open file corresponding to @var{fd}. Up to
34642 @var{count} bytes will be read from the file, starting at @var{offset}
34643 relative to the start of the file. The target may read fewer bytes;
34644 common reasons include packet size limits and an end-of-file
34645 condition. The number of bytes read is returned. Zero should only be
34646 returned for a successful read at the end of the file, or if
34647 @var{count} was zero.
34649 The data read should be returned as a binary attachment on success.
34650 If zero bytes were read, the response should include an empty binary
34651 attachment (i.e.@: a trailing semicolon). The return value is the
34652 number of target bytes read; the binary attachment may be longer if
34653 some characters were escaped.
34655 @item vFile:pwrite: @var{fd}, @var{offset}, @var{data}
34656 Write @var{data} (a binary buffer) to the open file corresponding
34657 to @var{fd}. Start the write at @var{offset} from the start of the
34658 file. Unlike many @code{write} system calls, there is no
34659 separate @var{count} argument; the length of @var{data} in the
34660 packet is used. @samp{vFile:write} returns the number of bytes written,
34661 which may be shorter than the length of @var{data}, or -1 if an
34664 @item vFile:unlink: @var{pathname}
34665 Delete the file at @var{pathname} on the target. Return 0,
34666 or -1 if an error occurs. @var{pathname} is a string.
34671 @section Interrupts
34672 @cindex interrupts (remote protocol)
34674 When a program on the remote target is running, @value{GDBN} may
34675 attempt to interrupt it by sending a @samp{Ctrl-C}, @code{BREAK} or
34676 a @code{BREAK} followed by @code{g},
34677 control of which is specified via @value{GDBN}'s @samp{interrupt-sequence}.
34679 The precise meaning of @code{BREAK} is defined by the transport
34680 mechanism and may, in fact, be undefined. @value{GDBN} does not
34681 currently define a @code{BREAK} mechanism for any of the network
34682 interfaces except for TCP, in which case @value{GDBN} sends the
34683 @code{telnet} BREAK sequence.
34685 @samp{Ctrl-C}, on the other hand, is defined and implemented for all
34686 transport mechanisms. It is represented by sending the single byte
34687 @code{0x03} without any of the usual packet overhead described in
34688 the Overview section (@pxref{Overview}). When a @code{0x03} byte is
34689 transmitted as part of a packet, it is considered to be packet data
34690 and does @emph{not} represent an interrupt. E.g., an @samp{X} packet
34691 (@pxref{X packet}), used for binary downloads, may include an unescaped
34692 @code{0x03} as part of its packet.
34694 @code{BREAK} followed by @code{g} is also known as Magic SysRq g.
34695 When Linux kernel receives this sequence from serial port,
34696 it stops execution and connects to gdb.
34698 Stubs are not required to recognize these interrupt mechanisms and the
34699 precise meaning associated with receipt of the interrupt is
34700 implementation defined. If the target supports debugging of multiple
34701 threads and/or processes, it should attempt to interrupt all
34702 currently-executing threads and processes.
34703 If the stub is successful at interrupting the
34704 running program, it should send one of the stop
34705 reply packets (@pxref{Stop Reply Packets}) to @value{GDBN} as a result
34706 of successfully stopping the program in all-stop mode, and a stop reply
34707 for each stopped thread in non-stop mode.
34708 Interrupts received while the
34709 program is stopped are discarded.
34711 @node Notification Packets
34712 @section Notification Packets
34713 @cindex notification packets
34714 @cindex packets, notification
34716 The @value{GDBN} remote serial protocol includes @dfn{notifications},
34717 packets that require no acknowledgment. Both the GDB and the stub
34718 may send notifications (although the only notifications defined at
34719 present are sent by the stub). Notifications carry information
34720 without incurring the round-trip latency of an acknowledgment, and so
34721 are useful for low-impact communications where occasional packet loss
34724 A notification packet has the form @samp{% @var{data} #
34725 @var{checksum}}, where @var{data} is the content of the notification,
34726 and @var{checksum} is a checksum of @var{data}, computed and formatted
34727 as for ordinary @value{GDBN} packets. A notification's @var{data}
34728 never contains @samp{$}, @samp{%} or @samp{#} characters. Upon
34729 receiving a notification, the recipient sends no @samp{+} or @samp{-}
34730 to acknowledge the notification's receipt or to report its corruption.
34732 Every notification's @var{data} begins with a name, which contains no
34733 colon characters, followed by a colon character.
34735 Recipients should silently ignore corrupted notifications and
34736 notifications they do not understand. Recipients should restart
34737 timeout periods on receipt of a well-formed notification, whether or
34738 not they understand it.
34740 Senders should only send the notifications described here when this
34741 protocol description specifies that they are permitted. In the
34742 future, we may extend the protocol to permit existing notifications in
34743 new contexts; this rule helps older senders avoid confusing newer
34746 (Older versions of @value{GDBN} ignore bytes received until they see
34747 the @samp{$} byte that begins an ordinary packet, so new stubs may
34748 transmit notifications without fear of confusing older clients. There
34749 are no notifications defined for @value{GDBN} to send at the moment, but we
34750 assume that most older stubs would ignore them, as well.)
34752 The following notification packets from the stub to @value{GDBN} are
34756 @item Stop: @var{reply}
34757 Report an asynchronous stop event in non-stop mode.
34758 The @var{reply} has the form of a stop reply, as
34759 described in @ref{Stop Reply Packets}. Refer to @ref{Remote Non-Stop},
34760 for information on how these notifications are acknowledged by
34764 @node Remote Non-Stop
34765 @section Remote Protocol Support for Non-Stop Mode
34767 @value{GDBN}'s remote protocol supports non-stop debugging of
34768 multi-threaded programs, as described in @ref{Non-Stop Mode}. If the stub
34769 supports non-stop mode, it should report that to @value{GDBN} by including
34770 @samp{QNonStop+} in its @samp{qSupported} response (@pxref{qSupported}).
34772 @value{GDBN} typically sends a @samp{QNonStop} packet only when
34773 establishing a new connection with the stub. Entering non-stop mode
34774 does not alter the state of any currently-running threads, but targets
34775 must stop all threads in any already-attached processes when entering
34776 all-stop mode. @value{GDBN} uses the @samp{?} packet as necessary to
34777 probe the target state after a mode change.
34779 In non-stop mode, when an attached process encounters an event that
34780 would otherwise be reported with a stop reply, it uses the
34781 asynchronous notification mechanism (@pxref{Notification Packets}) to
34782 inform @value{GDBN}. In contrast to all-stop mode, where all threads
34783 in all processes are stopped when a stop reply is sent, in non-stop
34784 mode only the thread reporting the stop event is stopped. That is,
34785 when reporting a @samp{S} or @samp{T} response to indicate completion
34786 of a step operation, hitting a breakpoint, or a fault, only the
34787 affected thread is stopped; any other still-running threads continue
34788 to run. When reporting a @samp{W} or @samp{X} response, all running
34789 threads belonging to other attached processes continue to run.
34791 Only one stop reply notification at a time may be pending; if
34792 additional stop events occur before @value{GDBN} has acknowledged the
34793 previous notification, they must be queued by the stub for later
34794 synchronous transmission in response to @samp{vStopped} packets from
34795 @value{GDBN}. Because the notification mechanism is unreliable,
34796 the stub is permitted to resend a stop reply notification
34797 if it believes @value{GDBN} may not have received it. @value{GDBN}
34798 ignores additional stop reply notifications received before it has
34799 finished processing a previous notification and the stub has completed
34800 sending any queued stop events.
34802 Otherwise, @value{GDBN} must be prepared to receive a stop reply
34803 notification at any time. Specifically, they may appear when
34804 @value{GDBN} is not otherwise reading input from the stub, or when
34805 @value{GDBN} is expecting to read a normal synchronous response or a
34806 @samp{+}/@samp{-} acknowledgment to a packet it has sent.
34807 Notification packets are distinct from any other communication from
34808 the stub so there is no ambiguity.
34810 After receiving a stop reply notification, @value{GDBN} shall
34811 acknowledge it by sending a @samp{vStopped} packet (@pxref{vStopped packet})
34812 as a regular, synchronous request to the stub. Such acknowledgment
34813 is not required to happen immediately, as @value{GDBN} is permitted to
34814 send other, unrelated packets to the stub first, which the stub should
34817 Upon receiving a @samp{vStopped} packet, if the stub has other queued
34818 stop events to report to @value{GDBN}, it shall respond by sending a
34819 normal stop reply response. @value{GDBN} shall then send another
34820 @samp{vStopped} packet to solicit further responses; again, it is
34821 permitted to send other, unrelated packets as well which the stub
34822 should process normally.
34824 If the stub receives a @samp{vStopped} packet and there are no
34825 additional stop events to report, the stub shall return an @samp{OK}
34826 response. At this point, if further stop events occur, the stub shall
34827 send a new stop reply notification, @value{GDBN} shall accept the
34828 notification, and the process shall be repeated.
34830 In non-stop mode, the target shall respond to the @samp{?} packet as
34831 follows. First, any incomplete stop reply notification/@samp{vStopped}
34832 sequence in progress is abandoned. The target must begin a new
34833 sequence reporting stop events for all stopped threads, whether or not
34834 it has previously reported those events to @value{GDBN}. The first
34835 stop reply is sent as a synchronous reply to the @samp{?} packet, and
34836 subsequent stop replies are sent as responses to @samp{vStopped} packets
34837 using the mechanism described above. The target must not send
34838 asynchronous stop reply notifications until the sequence is complete.
34839 If all threads are running when the target receives the @samp{?} packet,
34840 or if the target is not attached to any process, it shall respond
34843 @node Packet Acknowledgment
34844 @section Packet Acknowledgment
34846 @cindex acknowledgment, for @value{GDBN} remote
34847 @cindex packet acknowledgment, for @value{GDBN} remote
34848 By default, when either the host or the target machine receives a packet,
34849 the first response expected is an acknowledgment: either @samp{+} (to indicate
34850 the package was received correctly) or @samp{-} (to request retransmission).
34851 This mechanism allows the @value{GDBN} remote protocol to operate over
34852 unreliable transport mechanisms, such as a serial line.
34854 In cases where the transport mechanism is itself reliable (such as a pipe or
34855 TCP connection), the @samp{+}/@samp{-} acknowledgments are redundant.
34856 It may be desirable to disable them in that case to reduce communication
34857 overhead, or for other reasons. This can be accomplished by means of the
34858 @samp{QStartNoAckMode} packet; @pxref{QStartNoAckMode}.
34860 When in no-acknowledgment mode, neither the stub nor @value{GDBN} shall send or
34861 expect @samp{+}/@samp{-} protocol acknowledgments. The packet
34862 and response format still includes the normal checksum, as described in
34863 @ref{Overview}, but the checksum may be ignored by the receiver.
34865 If the stub supports @samp{QStartNoAckMode} and prefers to operate in
34866 no-acknowledgment mode, it should report that to @value{GDBN}
34867 by including @samp{QStartNoAckMode+} in its response to @samp{qSupported};
34868 @pxref{qSupported}.
34869 If @value{GDBN} also supports @samp{QStartNoAckMode} and it has not been
34870 disabled via the @code{set remote noack-packet off} command
34871 (@pxref{Remote Configuration}),
34872 @value{GDBN} may then send a @samp{QStartNoAckMode} packet to the stub.
34873 Only then may the stub actually turn off packet acknowledgments.
34874 @value{GDBN} sends a final @samp{+} acknowledgment of the stub's @samp{OK}
34875 response, which can be safely ignored by the stub.
34877 Note that @code{set remote noack-packet} command only affects negotiation
34878 between @value{GDBN} and the stub when subsequent connections are made;
34879 it does not affect the protocol acknowledgment state for any current
34881 Since @samp{+}/@samp{-} acknowledgments are enabled by default when a
34882 new connection is established,
34883 there is also no protocol request to re-enable the acknowledgments
34884 for the current connection, once disabled.
34889 Example sequence of a target being re-started. Notice how the restart
34890 does not get any direct output:
34895 @emph{target restarts}
34898 <- @code{T001:1234123412341234}
34902 Example sequence of a target being stepped by a single instruction:
34905 -> @code{G1445@dots{}}
34910 <- @code{T001:1234123412341234}
34914 <- @code{1455@dots{}}
34918 @node File-I/O Remote Protocol Extension
34919 @section File-I/O Remote Protocol Extension
34920 @cindex File-I/O remote protocol extension
34923 * File-I/O Overview::
34924 * Protocol Basics::
34925 * The F Request Packet::
34926 * The F Reply Packet::
34927 * The Ctrl-C Message::
34929 * List of Supported Calls::
34930 * Protocol-specific Representation of Datatypes::
34932 * File-I/O Examples::
34935 @node File-I/O Overview
34936 @subsection File-I/O Overview
34937 @cindex file-i/o overview
34939 The @dfn{File I/O remote protocol extension} (short: File-I/O) allows the
34940 target to use the host's file system and console I/O to perform various
34941 system calls. System calls on the target system are translated into a
34942 remote protocol packet to the host system, which then performs the needed
34943 actions and returns a response packet to the target system.
34944 This simulates file system operations even on targets that lack file systems.
34946 The protocol is defined to be independent of both the host and target systems.
34947 It uses its own internal representation of datatypes and values. Both
34948 @value{GDBN} and the target's @value{GDBN} stub are responsible for
34949 translating the system-dependent value representations into the internal
34950 protocol representations when data is transmitted.
34952 The communication is synchronous. A system call is possible only when
34953 @value{GDBN} is waiting for a response from the @samp{C}, @samp{c}, @samp{S}
34954 or @samp{s} packets. While @value{GDBN} handles the request for a system call,
34955 the target is stopped to allow deterministic access to the target's
34956 memory. Therefore File-I/O is not interruptible by target signals. On
34957 the other hand, it is possible to interrupt File-I/O by a user interrupt
34958 (@samp{Ctrl-C}) within @value{GDBN}.
34960 The target's request to perform a host system call does not finish
34961 the latest @samp{C}, @samp{c}, @samp{S} or @samp{s} action. That means,
34962 after finishing the system call, the target returns to continuing the
34963 previous activity (continue, step). No additional continue or step
34964 request from @value{GDBN} is required.
34967 (@value{GDBP}) continue
34968 <- target requests 'system call X'
34969 target is stopped, @value{GDBN} executes system call
34970 -> @value{GDBN} returns result
34971 ... target continues, @value{GDBN} returns to wait for the target
34972 <- target hits breakpoint and sends a Txx packet
34975 The protocol only supports I/O on the console and to regular files on
34976 the host file system. Character or block special devices, pipes,
34977 named pipes, sockets or any other communication method on the host
34978 system are not supported by this protocol.
34980 File I/O is not supported in non-stop mode.
34982 @node Protocol Basics
34983 @subsection Protocol Basics
34984 @cindex protocol basics, file-i/o
34986 The File-I/O protocol uses the @code{F} packet as the request as well
34987 as reply packet. Since a File-I/O system call can only occur when
34988 @value{GDBN} is waiting for a response from the continuing or stepping target,
34989 the File-I/O request is a reply that @value{GDBN} has to expect as a result
34990 of a previous @samp{C}, @samp{c}, @samp{S} or @samp{s} packet.
34991 This @code{F} packet contains all information needed to allow @value{GDBN}
34992 to call the appropriate host system call:
34996 A unique identifier for the requested system call.
34999 All parameters to the system call. Pointers are given as addresses
35000 in the target memory address space. Pointers to strings are given as
35001 pointer/length pair. Numerical values are given as they are.
35002 Numerical control flags are given in a protocol-specific representation.
35006 At this point, @value{GDBN} has to perform the following actions.
35010 If the parameters include pointer values to data needed as input to a
35011 system call, @value{GDBN} requests this data from the target with a
35012 standard @code{m} packet request. This additional communication has to be
35013 expected by the target implementation and is handled as any other @code{m}
35017 @value{GDBN} translates all value from protocol representation to host
35018 representation as needed. Datatypes are coerced into the host types.
35021 @value{GDBN} calls the system call.
35024 It then coerces datatypes back to protocol representation.
35027 If the system call is expected to return data in buffer space specified
35028 by pointer parameters to the call, the data is transmitted to the
35029 target using a @code{M} or @code{X} packet. This packet has to be expected
35030 by the target implementation and is handled as any other @code{M} or @code{X}
35035 Eventually @value{GDBN} replies with another @code{F} packet which contains all
35036 necessary information for the target to continue. This at least contains
35043 @code{errno}, if has been changed by the system call.
35050 After having done the needed type and value coercion, the target continues
35051 the latest continue or step action.
35053 @node The F Request Packet
35054 @subsection The @code{F} Request Packet
35055 @cindex file-i/o request packet
35056 @cindex @code{F} request packet
35058 The @code{F} request packet has the following format:
35061 @item F@var{call-id},@var{parameter@dots{}}
35063 @var{call-id} is the identifier to indicate the host system call to be called.
35064 This is just the name of the function.
35066 @var{parameter@dots{}} are the parameters to the system call.
35067 Parameters are hexadecimal integer values, either the actual values in case
35068 of scalar datatypes, pointers to target buffer space in case of compound
35069 datatypes and unspecified memory areas, or pointer/length pairs in case
35070 of string parameters. These are appended to the @var{call-id} as a
35071 comma-delimited list. All values are transmitted in ASCII
35072 string representation, pointer/length pairs separated by a slash.
35078 @node The F Reply Packet
35079 @subsection The @code{F} Reply Packet
35080 @cindex file-i/o reply packet
35081 @cindex @code{F} reply packet
35083 The @code{F} reply packet has the following format:
35087 @item F@var{retcode},@var{errno},@var{Ctrl-C flag};@var{call-specific attachment}
35089 @var{retcode} is the return code of the system call as hexadecimal value.
35091 @var{errno} is the @code{errno} set by the call, in protocol-specific
35093 This parameter can be omitted if the call was successful.
35095 @var{Ctrl-C flag} is only sent if the user requested a break. In this
35096 case, @var{errno} must be sent as well, even if the call was successful.
35097 The @var{Ctrl-C flag} itself consists of the character @samp{C}:
35104 or, if the call was interrupted before the host call has been performed:
35111 assuming 4 is the protocol-specific representation of @code{EINTR}.
35116 @node The Ctrl-C Message
35117 @subsection The @samp{Ctrl-C} Message
35118 @cindex ctrl-c message, in file-i/o protocol
35120 If the @samp{Ctrl-C} flag is set in the @value{GDBN}
35121 reply packet (@pxref{The F Reply Packet}),
35122 the target should behave as if it had
35123 gotten a break message. The meaning for the target is ``system call
35124 interrupted by @code{SIGINT}''. Consequentially, the target should actually stop
35125 (as with a break message) and return to @value{GDBN} with a @code{T02}
35128 It's important for the target to know in which
35129 state the system call was interrupted. There are two possible cases:
35133 The system call hasn't been performed on the host yet.
35136 The system call on the host has been finished.
35140 These two states can be distinguished by the target by the value of the
35141 returned @code{errno}. If it's the protocol representation of @code{EINTR}, the system
35142 call hasn't been performed. This is equivalent to the @code{EINTR} handling
35143 on POSIX systems. In any other case, the target may presume that the
35144 system call has been finished --- successfully or not --- and should behave
35145 as if the break message arrived right after the system call.
35147 @value{GDBN} must behave reliably. If the system call has not been called
35148 yet, @value{GDBN} may send the @code{F} reply immediately, setting @code{EINTR} as
35149 @code{errno} in the packet. If the system call on the host has been finished
35150 before the user requests a break, the full action must be finished by
35151 @value{GDBN}. This requires sending @code{M} or @code{X} packets as necessary.
35152 The @code{F} packet may only be sent when either nothing has happened
35153 or the full action has been completed.
35156 @subsection Console I/O
35157 @cindex console i/o as part of file-i/o
35159 By default and if not explicitly closed by the target system, the file
35160 descriptors 0, 1 and 2 are connected to the @value{GDBN} console. Output
35161 on the @value{GDBN} console is handled as any other file output operation
35162 (@code{write(1, @dots{})} or @code{write(2, @dots{})}). Console input is handled
35163 by @value{GDBN} so that after the target read request from file descriptor
35164 0 all following typing is buffered until either one of the following
35169 The user types @kbd{Ctrl-c}. The behaviour is as explained above, and the
35171 system call is treated as finished.
35174 The user presses @key{RET}. This is treated as end of input with a trailing
35178 The user types @kbd{Ctrl-d}. This is treated as end of input. No trailing
35179 character (neither newline nor @samp{Ctrl-D}) is appended to the input.
35183 If the user has typed more characters than fit in the buffer given to
35184 the @code{read} call, the trailing characters are buffered in @value{GDBN} until
35185 either another @code{read(0, @dots{})} is requested by the target, or debugging
35186 is stopped at the user's request.
35189 @node List of Supported Calls
35190 @subsection List of Supported Calls
35191 @cindex list of supported file-i/o calls
35208 @unnumberedsubsubsec open
35209 @cindex open, file-i/o system call
35214 int open(const char *pathname, int flags);
35215 int open(const char *pathname, int flags, mode_t mode);
35219 @samp{Fopen,@var{pathptr}/@var{len},@var{flags},@var{mode}}
35222 @var{flags} is the bitwise @code{OR} of the following values:
35226 If the file does not exist it will be created. The host
35227 rules apply as far as file ownership and time stamps
35231 When used with @code{O_CREAT}, if the file already exists it is
35232 an error and open() fails.
35235 If the file already exists and the open mode allows
35236 writing (@code{O_RDWR} or @code{O_WRONLY} is given) it will be
35237 truncated to zero length.
35240 The file is opened in append mode.
35243 The file is opened for reading only.
35246 The file is opened for writing only.
35249 The file is opened for reading and writing.
35253 Other bits are silently ignored.
35257 @var{mode} is the bitwise @code{OR} of the following values:
35261 User has read permission.
35264 User has write permission.
35267 Group has read permission.
35270 Group has write permission.
35273 Others have read permission.
35276 Others have write permission.
35280 Other bits are silently ignored.
35283 @item Return value:
35284 @code{open} returns the new file descriptor or -1 if an error
35291 @var{pathname} already exists and @code{O_CREAT} and @code{O_EXCL} were used.
35294 @var{pathname} refers to a directory.
35297 The requested access is not allowed.
35300 @var{pathname} was too long.
35303 A directory component in @var{pathname} does not exist.
35306 @var{pathname} refers to a device, pipe, named pipe or socket.
35309 @var{pathname} refers to a file on a read-only filesystem and
35310 write access was requested.
35313 @var{pathname} is an invalid pointer value.
35316 No space on device to create the file.
35319 The process already has the maximum number of files open.
35322 The limit on the total number of files open on the system
35326 The call was interrupted by the user.
35332 @unnumberedsubsubsec close
35333 @cindex close, file-i/o system call
35342 @samp{Fclose,@var{fd}}
35344 @item Return value:
35345 @code{close} returns zero on success, or -1 if an error occurred.
35351 @var{fd} isn't a valid open file descriptor.
35354 The call was interrupted by the user.
35360 @unnumberedsubsubsec read
35361 @cindex read, file-i/o system call
35366 int read(int fd, void *buf, unsigned int count);
35370 @samp{Fread,@var{fd},@var{bufptr},@var{count}}
35372 @item Return value:
35373 On success, the number of bytes read is returned.
35374 Zero indicates end of file. If count is zero, read
35375 returns zero as well. On error, -1 is returned.
35381 @var{fd} is not a valid file descriptor or is not open for
35385 @var{bufptr} is an invalid pointer value.
35388 The call was interrupted by the user.
35394 @unnumberedsubsubsec write
35395 @cindex write, file-i/o system call
35400 int write(int fd, const void *buf, unsigned int count);
35404 @samp{Fwrite,@var{fd},@var{bufptr},@var{count}}
35406 @item Return value:
35407 On success, the number of bytes written are returned.
35408 Zero indicates nothing was written. On error, -1
35415 @var{fd} is not a valid file descriptor or is not open for
35419 @var{bufptr} is an invalid pointer value.
35422 An attempt was made to write a file that exceeds the
35423 host-specific maximum file size allowed.
35426 No space on device to write the data.
35429 The call was interrupted by the user.
35435 @unnumberedsubsubsec lseek
35436 @cindex lseek, file-i/o system call
35441 long lseek (int fd, long offset, int flag);
35445 @samp{Flseek,@var{fd},@var{offset},@var{flag}}
35447 @var{flag} is one of:
35451 The offset is set to @var{offset} bytes.
35454 The offset is set to its current location plus @var{offset}
35458 The offset is set to the size of the file plus @var{offset}
35462 @item Return value:
35463 On success, the resulting unsigned offset in bytes from
35464 the beginning of the file is returned. Otherwise, a
35465 value of -1 is returned.
35471 @var{fd} is not a valid open file descriptor.
35474 @var{fd} is associated with the @value{GDBN} console.
35477 @var{flag} is not a proper value.
35480 The call was interrupted by the user.
35486 @unnumberedsubsubsec rename
35487 @cindex rename, file-i/o system call
35492 int rename(const char *oldpath, const char *newpath);
35496 @samp{Frename,@var{oldpathptr}/@var{len},@var{newpathptr}/@var{len}}
35498 @item Return value:
35499 On success, zero is returned. On error, -1 is returned.
35505 @var{newpath} is an existing directory, but @var{oldpath} is not a
35509 @var{newpath} is a non-empty directory.
35512 @var{oldpath} or @var{newpath} is a directory that is in use by some
35516 An attempt was made to make a directory a subdirectory
35520 A component used as a directory in @var{oldpath} or new
35521 path is not a directory. Or @var{oldpath} is a directory
35522 and @var{newpath} exists but is not a directory.
35525 @var{oldpathptr} or @var{newpathptr} are invalid pointer values.
35528 No access to the file or the path of the file.
35532 @var{oldpath} or @var{newpath} was too long.
35535 A directory component in @var{oldpath} or @var{newpath} does not exist.
35538 The file is on a read-only filesystem.
35541 The device containing the file has no room for the new
35545 The call was interrupted by the user.
35551 @unnumberedsubsubsec unlink
35552 @cindex unlink, file-i/o system call
35557 int unlink(const char *pathname);
35561 @samp{Funlink,@var{pathnameptr}/@var{len}}
35563 @item Return value:
35564 On success, zero is returned. On error, -1 is returned.
35570 No access to the file or the path of the file.
35573 The system does not allow unlinking of directories.
35576 The file @var{pathname} cannot be unlinked because it's
35577 being used by another process.
35580 @var{pathnameptr} is an invalid pointer value.
35583 @var{pathname} was too long.
35586 A directory component in @var{pathname} does not exist.
35589 A component of the path is not a directory.
35592 The file is on a read-only filesystem.
35595 The call was interrupted by the user.
35601 @unnumberedsubsubsec stat/fstat
35602 @cindex fstat, file-i/o system call
35603 @cindex stat, file-i/o system call
35608 int stat(const char *pathname, struct stat *buf);
35609 int fstat(int fd, struct stat *buf);
35613 @samp{Fstat,@var{pathnameptr}/@var{len},@var{bufptr}}@*
35614 @samp{Ffstat,@var{fd},@var{bufptr}}
35616 @item Return value:
35617 On success, zero is returned. On error, -1 is returned.
35623 @var{fd} is not a valid open file.
35626 A directory component in @var{pathname} does not exist or the
35627 path is an empty string.
35630 A component of the path is not a directory.
35633 @var{pathnameptr} is an invalid pointer value.
35636 No access to the file or the path of the file.
35639 @var{pathname} was too long.
35642 The call was interrupted by the user.
35648 @unnumberedsubsubsec gettimeofday
35649 @cindex gettimeofday, file-i/o system call
35654 int gettimeofday(struct timeval *tv, void *tz);
35658 @samp{Fgettimeofday,@var{tvptr},@var{tzptr}}
35660 @item Return value:
35661 On success, 0 is returned, -1 otherwise.
35667 @var{tz} is a non-NULL pointer.
35670 @var{tvptr} and/or @var{tzptr} is an invalid pointer value.
35676 @unnumberedsubsubsec isatty
35677 @cindex isatty, file-i/o system call
35682 int isatty(int fd);
35686 @samp{Fisatty,@var{fd}}
35688 @item Return value:
35689 Returns 1 if @var{fd} refers to the @value{GDBN} console, 0 otherwise.
35695 The call was interrupted by the user.
35700 Note that the @code{isatty} call is treated as a special case: it returns
35701 1 to the target if the file descriptor is attached
35702 to the @value{GDBN} console, 0 otherwise. Implementing through system calls
35703 would require implementing @code{ioctl} and would be more complex than
35708 @unnumberedsubsubsec system
35709 @cindex system, file-i/o system call
35714 int system(const char *command);
35718 @samp{Fsystem,@var{commandptr}/@var{len}}
35720 @item Return value:
35721 If @var{len} is zero, the return value indicates whether a shell is
35722 available. A zero return value indicates a shell is not available.
35723 For non-zero @var{len}, the value returned is -1 on error and the
35724 return status of the command otherwise. Only the exit status of the
35725 command is returned, which is extracted from the host's @code{system}
35726 return value by calling @code{WEXITSTATUS(retval)}. In case
35727 @file{/bin/sh} could not be executed, 127 is returned.
35733 The call was interrupted by the user.
35738 @value{GDBN} takes over the full task of calling the necessary host calls
35739 to perform the @code{system} call. The return value of @code{system} on
35740 the host is simplified before it's returned
35741 to the target. Any termination signal information from the child process
35742 is discarded, and the return value consists
35743 entirely of the exit status of the called command.
35745 Due to security concerns, the @code{system} call is by default refused
35746 by @value{GDBN}. The user has to allow this call explicitly with the
35747 @code{set remote system-call-allowed 1} command.
35750 @item set remote system-call-allowed
35751 @kindex set remote system-call-allowed
35752 Control whether to allow the @code{system} calls in the File I/O
35753 protocol for the remote target. The default is zero (disabled).
35755 @item show remote system-call-allowed
35756 @kindex show remote system-call-allowed
35757 Show whether the @code{system} calls are allowed in the File I/O
35761 @node Protocol-specific Representation of Datatypes
35762 @subsection Protocol-specific Representation of Datatypes
35763 @cindex protocol-specific representation of datatypes, in file-i/o protocol
35766 * Integral Datatypes::
35768 * Memory Transfer::
35773 @node Integral Datatypes
35774 @unnumberedsubsubsec Integral Datatypes
35775 @cindex integral datatypes, in file-i/o protocol
35777 The integral datatypes used in the system calls are @code{int},
35778 @code{unsigned int}, @code{long}, @code{unsigned long},
35779 @code{mode_t}, and @code{time_t}.
35781 @code{int}, @code{unsigned int}, @code{mode_t} and @code{time_t} are
35782 implemented as 32 bit values in this protocol.
35784 @code{long} and @code{unsigned long} are implemented as 64 bit types.
35786 @xref{Limits}, for corresponding MIN and MAX values (similar to those
35787 in @file{limits.h}) to allow range checking on host and target.
35789 @code{time_t} datatypes are defined as seconds since the Epoch.
35791 All integral datatypes transferred as part of a memory read or write of a
35792 structured datatype e.g.@: a @code{struct stat} have to be given in big endian
35795 @node Pointer Values
35796 @unnumberedsubsubsec Pointer Values
35797 @cindex pointer values, in file-i/o protocol
35799 Pointers to target data are transmitted as they are. An exception
35800 is made for pointers to buffers for which the length isn't
35801 transmitted as part of the function call, namely strings. Strings
35802 are transmitted as a pointer/length pair, both as hex values, e.g.@:
35809 which is a pointer to data of length 18 bytes at position 0x1aaf.
35810 The length is defined as the full string length in bytes, including
35811 the trailing null byte. For example, the string @code{"hello world"}
35812 at address 0x123456 is transmitted as
35818 @node Memory Transfer
35819 @unnumberedsubsubsec Memory Transfer
35820 @cindex memory transfer, in file-i/o protocol
35822 Structured data which is transferred using a memory read or write (for
35823 example, a @code{struct stat}) is expected to be in a protocol-specific format
35824 with all scalar multibyte datatypes being big endian. Translation to
35825 this representation needs to be done both by the target before the @code{F}
35826 packet is sent, and by @value{GDBN} before
35827 it transfers memory to the target. Transferred pointers to structured
35828 data should point to the already-coerced data at any time.
35832 @unnumberedsubsubsec struct stat
35833 @cindex struct stat, in file-i/o protocol
35835 The buffer of type @code{struct stat} used by the target and @value{GDBN}
35836 is defined as follows:
35840 unsigned int st_dev; /* device */
35841 unsigned int st_ino; /* inode */
35842 mode_t st_mode; /* protection */
35843 unsigned int st_nlink; /* number of hard links */
35844 unsigned int st_uid; /* user ID of owner */
35845 unsigned int st_gid; /* group ID of owner */
35846 unsigned int st_rdev; /* device type (if inode device) */
35847 unsigned long st_size; /* total size, in bytes */
35848 unsigned long st_blksize; /* blocksize for filesystem I/O */
35849 unsigned long st_blocks; /* number of blocks allocated */
35850 time_t st_atime; /* time of last access */
35851 time_t st_mtime; /* time of last modification */
35852 time_t st_ctime; /* time of last change */
35856 The integral datatypes conform to the definitions given in the
35857 appropriate section (see @ref{Integral Datatypes}, for details) so this
35858 structure is of size 64 bytes.
35860 The values of several fields have a restricted meaning and/or
35866 A value of 0 represents a file, 1 the console.
35869 No valid meaning for the target. Transmitted unchanged.
35872 Valid mode bits are described in @ref{Constants}. Any other
35873 bits have currently no meaning for the target.
35878 No valid meaning for the target. Transmitted unchanged.
35883 These values have a host and file system dependent
35884 accuracy. Especially on Windows hosts, the file system may not
35885 support exact timing values.
35888 The target gets a @code{struct stat} of the above representation and is
35889 responsible for coercing it to the target representation before
35892 Note that due to size differences between the host, target, and protocol
35893 representations of @code{struct stat} members, these members could eventually
35894 get truncated on the target.
35896 @node struct timeval
35897 @unnumberedsubsubsec struct timeval
35898 @cindex struct timeval, in file-i/o protocol
35900 The buffer of type @code{struct timeval} used by the File-I/O protocol
35901 is defined as follows:
35905 time_t tv_sec; /* second */
35906 long tv_usec; /* microsecond */
35910 The integral datatypes conform to the definitions given in the
35911 appropriate section (see @ref{Integral Datatypes}, for details) so this
35912 structure is of size 8 bytes.
35915 @subsection Constants
35916 @cindex constants, in file-i/o protocol
35918 The following values are used for the constants inside of the
35919 protocol. @value{GDBN} and target are responsible for translating these
35920 values before and after the call as needed.
35931 @unnumberedsubsubsec Open Flags
35932 @cindex open flags, in file-i/o protocol
35934 All values are given in hexadecimal representation.
35946 @node mode_t Values
35947 @unnumberedsubsubsec mode_t Values
35948 @cindex mode_t values, in file-i/o protocol
35950 All values are given in octal representation.
35967 @unnumberedsubsubsec Errno Values
35968 @cindex errno values, in file-i/o protocol
35970 All values are given in decimal representation.
35995 @code{EUNKNOWN} is used as a fallback error value if a host system returns
35996 any error value not in the list of supported error numbers.
35999 @unnumberedsubsubsec Lseek Flags
36000 @cindex lseek flags, in file-i/o protocol
36009 @unnumberedsubsubsec Limits
36010 @cindex limits, in file-i/o protocol
36012 All values are given in decimal representation.
36015 INT_MIN -2147483648
36017 UINT_MAX 4294967295
36018 LONG_MIN -9223372036854775808
36019 LONG_MAX 9223372036854775807
36020 ULONG_MAX 18446744073709551615
36023 @node File-I/O Examples
36024 @subsection File-I/O Examples
36025 @cindex file-i/o examples
36027 Example sequence of a write call, file descriptor 3, buffer is at target
36028 address 0x1234, 6 bytes should be written:
36031 <- @code{Fwrite,3,1234,6}
36032 @emph{request memory read from target}
36035 @emph{return "6 bytes written"}
36039 Example sequence of a read call, file descriptor 3, buffer is at target
36040 address 0x1234, 6 bytes should be read:
36043 <- @code{Fread,3,1234,6}
36044 @emph{request memory write to target}
36045 -> @code{X1234,6:XXXXXX}
36046 @emph{return "6 bytes read"}
36050 Example sequence of a read call, call fails on the host due to invalid
36051 file descriptor (@code{EBADF}):
36054 <- @code{Fread,3,1234,6}
36058 Example sequence of a read call, user presses @kbd{Ctrl-c} before syscall on
36062 <- @code{Fread,3,1234,6}
36067 Example sequence of a read call, user presses @kbd{Ctrl-c} after syscall on
36071 <- @code{Fread,3,1234,6}
36072 -> @code{X1234,6:XXXXXX}
36076 @node Library List Format
36077 @section Library List Format
36078 @cindex library list format, remote protocol
36080 On some platforms, a dynamic loader (e.g.@: @file{ld.so}) runs in the
36081 same process as your application to manage libraries. In this case,
36082 @value{GDBN} can use the loader's symbol table and normal memory
36083 operations to maintain a list of shared libraries. On other
36084 platforms, the operating system manages loaded libraries.
36085 @value{GDBN} can not retrieve the list of currently loaded libraries
36086 through memory operations, so it uses the @samp{qXfer:libraries:read}
36087 packet (@pxref{qXfer library list read}) instead. The remote stub
36088 queries the target's operating system and reports which libraries
36091 The @samp{qXfer:libraries:read} packet returns an XML document which
36092 lists loaded libraries and their offsets. Each library has an
36093 associated name and one or more segment or section base addresses,
36094 which report where the library was loaded in memory.
36096 For the common case of libraries that are fully linked binaries, the
36097 library should have a list of segments. If the target supports
36098 dynamic linking of a relocatable object file, its library XML element
36099 should instead include a list of allocated sections. The segment or
36100 section bases are start addresses, not relocation offsets; they do not
36101 depend on the library's link-time base addresses.
36103 @value{GDBN} must be linked with the Expat library to support XML
36104 library lists. @xref{Expat}.
36106 A simple memory map, with one loaded library relocated by a single
36107 offset, looks like this:
36111 <library name="/lib/libc.so.6">
36112 <segment address="0x10000000"/>
36117 Another simple memory map, with one loaded library with three
36118 allocated sections (.text, .data, .bss), looks like this:
36122 <library name="sharedlib.o">
36123 <section address="0x10000000"/>
36124 <section address="0x20000000"/>
36125 <section address="0x30000000"/>
36130 The format of a library list is described by this DTD:
36133 <!-- library-list: Root element with versioning -->
36134 <!ELEMENT library-list (library)*>
36135 <!ATTLIST library-list version CDATA #FIXED "1.0">
36136 <!ELEMENT library (segment*, section*)>
36137 <!ATTLIST library name CDATA #REQUIRED>
36138 <!ELEMENT segment EMPTY>
36139 <!ATTLIST segment address CDATA #REQUIRED>
36140 <!ELEMENT section EMPTY>
36141 <!ATTLIST section address CDATA #REQUIRED>
36144 In addition, segments and section descriptors cannot be mixed within a
36145 single library element, and you must supply at least one segment or
36146 section for each library.
36148 @node Memory Map Format
36149 @section Memory Map Format
36150 @cindex memory map format
36152 To be able to write into flash memory, @value{GDBN} needs to obtain a
36153 memory map from the target. This section describes the format of the
36156 The memory map is obtained using the @samp{qXfer:memory-map:read}
36157 (@pxref{qXfer memory map read}) packet and is an XML document that
36158 lists memory regions.
36160 @value{GDBN} must be linked with the Expat library to support XML
36161 memory maps. @xref{Expat}.
36163 The top-level structure of the document is shown below:
36166 <?xml version="1.0"?>
36167 <!DOCTYPE memory-map
36168 PUBLIC "+//IDN gnu.org//DTD GDB Memory Map V1.0//EN"
36169 "http://sourceware.org/gdb/gdb-memory-map.dtd">
36175 Each region can be either:
36180 A region of RAM starting at @var{addr} and extending for @var{length}
36184 <memory type="ram" start="@var{addr}" length="@var{length}"/>
36189 A region of read-only memory:
36192 <memory type="rom" start="@var{addr}" length="@var{length}"/>
36197 A region of flash memory, with erasure blocks @var{blocksize}
36201 <memory type="flash" start="@var{addr}" length="@var{length}">
36202 <property name="blocksize">@var{blocksize}</property>
36208 Regions must not overlap. @value{GDBN} assumes that areas of memory not covered
36209 by the memory map are RAM, and uses the ordinary @samp{M} and @samp{X}
36210 packets to write to addresses in such ranges.
36212 The formal DTD for memory map format is given below:
36215 <!-- ................................................... -->
36216 <!-- Memory Map XML DTD ................................ -->
36217 <!-- File: memory-map.dtd .............................. -->
36218 <!-- .................................... .............. -->
36219 <!-- memory-map.dtd -->
36220 <!-- memory-map: Root element with versioning -->
36221 <!ELEMENT memory-map (memory | property)>
36222 <!ATTLIST memory-map version CDATA #FIXED "1.0.0">
36223 <!ELEMENT memory (property)>
36224 <!-- memory: Specifies a memory region,
36225 and its type, or device. -->
36226 <!ATTLIST memory type CDATA #REQUIRED
36227 start CDATA #REQUIRED
36228 length CDATA #REQUIRED
36229 device CDATA #IMPLIED>
36230 <!-- property: Generic attribute tag -->
36231 <!ELEMENT property (#PCDATA | property)*>
36232 <!ATTLIST property name CDATA #REQUIRED>
36235 @node Thread List Format
36236 @section Thread List Format
36237 @cindex thread list format
36239 To efficiently update the list of threads and their attributes,
36240 @value{GDBN} issues the @samp{qXfer:threads:read} packet
36241 (@pxref{qXfer threads read}) and obtains the XML document with
36242 the following structure:
36245 <?xml version="1.0"?>
36247 <thread id="id" core="0">
36248 ... description ...
36253 Each @samp{thread} element must have the @samp{id} attribute that
36254 identifies the thread (@pxref{thread-id syntax}). The
36255 @samp{core} attribute, if present, specifies which processor core
36256 the thread was last executing on. The content of the of @samp{thread}
36257 element is interpreted as human-readable auxilliary information.
36259 @node Traceframe Info Format
36260 @section Traceframe Info Format
36261 @cindex traceframe info format
36263 To be able to know which objects in the inferior can be examined when
36264 inspecting a tracepoint hit, @value{GDBN} needs to obtain the list of
36265 memory ranges, registers and trace state variables that have been
36266 collected in a traceframe.
36268 This list is obtained using the @samp{qXfer:traceframe-info:read}
36269 (@pxref{qXfer traceframe info read}) packet and is an XML document.
36271 @value{GDBN} must be linked with the Expat library to support XML
36272 traceframe info discovery. @xref{Expat}.
36274 The top-level structure of the document is shown below:
36277 <?xml version="1.0"?>
36278 <!DOCTYPE traceframe-info
36279 PUBLIC "+//IDN gnu.org//DTD GDB Memory Map V1.0//EN"
36280 "http://sourceware.org/gdb/gdb-traceframe-info.dtd">
36286 Each traceframe block can be either:
36291 A region of collected memory starting at @var{addr} and extending for
36292 @var{length} bytes from there:
36295 <memory start="@var{addr}" length="@var{length}"/>
36300 The formal DTD for the traceframe info format is given below:
36303 <!ELEMENT traceframe-info (memory)* >
36304 <!ATTLIST traceframe-info version CDATA #FIXED "1.0">
36306 <!ELEMENT memory EMPTY>
36307 <!ATTLIST memory start CDATA #REQUIRED
36308 length CDATA #REQUIRED>
36311 @include agentexpr.texi
36313 @node Target Descriptions
36314 @appendix Target Descriptions
36315 @cindex target descriptions
36317 @strong{Warning:} target descriptions are still under active development,
36318 and the contents and format may change between @value{GDBN} releases.
36319 The format is expected to stabilize in the future.
36321 One of the challenges of using @value{GDBN} to debug embedded systems
36322 is that there are so many minor variants of each processor
36323 architecture in use. It is common practice for vendors to start with
36324 a standard processor core --- ARM, PowerPC, or MIPS, for example ---
36325 and then make changes to adapt it to a particular market niche. Some
36326 architectures have hundreds of variants, available from dozens of
36327 vendors. This leads to a number of problems:
36331 With so many different customized processors, it is difficult for
36332 the @value{GDBN} maintainers to keep up with the changes.
36334 Since individual variants may have short lifetimes or limited
36335 audiences, it may not be worthwhile to carry information about every
36336 variant in the @value{GDBN} source tree.
36338 When @value{GDBN} does support the architecture of the embedded system
36339 at hand, the task of finding the correct architecture name to give the
36340 @command{set architecture} command can be error-prone.
36343 To address these problems, the @value{GDBN} remote protocol allows a
36344 target system to not only identify itself to @value{GDBN}, but to
36345 actually describe its own features. This lets @value{GDBN} support
36346 processor variants it has never seen before --- to the extent that the
36347 descriptions are accurate, and that @value{GDBN} understands them.
36349 @value{GDBN} must be linked with the Expat library to support XML
36350 target descriptions. @xref{Expat}.
36353 * Retrieving Descriptions:: How descriptions are fetched from a target.
36354 * Target Description Format:: The contents of a target description.
36355 * Predefined Target Types:: Standard types available for target
36357 * Standard Target Features:: Features @value{GDBN} knows about.
36360 @node Retrieving Descriptions
36361 @section Retrieving Descriptions
36363 Target descriptions can be read from the target automatically, or
36364 specified by the user manually. The default behavior is to read the
36365 description from the target. @value{GDBN} retrieves it via the remote
36366 protocol using @samp{qXfer} requests (@pxref{General Query Packets,
36367 qXfer}). The @var{annex} in the @samp{qXfer} packet will be
36368 @samp{target.xml}. The contents of the @samp{target.xml} annex are an
36369 XML document, of the form described in @ref{Target Description
36372 Alternatively, you can specify a file to read for the target description.
36373 If a file is set, the target will not be queried. The commands to
36374 specify a file are:
36377 @cindex set tdesc filename
36378 @item set tdesc filename @var{path}
36379 Read the target description from @var{path}.
36381 @cindex unset tdesc filename
36382 @item unset tdesc filename
36383 Do not read the XML target description from a file. @value{GDBN}
36384 will use the description supplied by the current target.
36386 @cindex show tdesc filename
36387 @item show tdesc filename
36388 Show the filename to read for a target description, if any.
36392 @node Target Description Format
36393 @section Target Description Format
36394 @cindex target descriptions, XML format
36396 A target description annex is an @uref{http://www.w3.org/XML/, XML}
36397 document which complies with the Document Type Definition provided in
36398 the @value{GDBN} sources in @file{gdb/features/gdb-target.dtd}. This
36399 means you can use generally available tools like @command{xmllint} to
36400 check that your feature descriptions are well-formed and valid.
36401 However, to help people unfamiliar with XML write descriptions for
36402 their targets, we also describe the grammar here.
36404 Target descriptions can identify the architecture of the remote target
36405 and (for some architectures) provide information about custom register
36406 sets. They can also identify the OS ABI of the remote target.
36407 @value{GDBN} can use this information to autoconfigure for your
36408 target, or to warn you if you connect to an unsupported target.
36410 Here is a simple target description:
36413 <target version="1.0">
36414 <architecture>i386:x86-64</architecture>
36419 This minimal description only says that the target uses
36420 the x86-64 architecture.
36422 A target description has the following overall form, with [ ] marking
36423 optional elements and @dots{} marking repeatable elements. The elements
36424 are explained further below.
36427 <?xml version="1.0"?>
36428 <!DOCTYPE target SYSTEM "gdb-target.dtd">
36429 <target version="1.0">
36430 @r{[}@var{architecture}@r{]}
36431 @r{[}@var{osabi}@r{]}
36432 @r{[}@var{compatible}@r{]}
36433 @r{[}@var{feature}@dots{}@r{]}
36438 The description is generally insensitive to whitespace and line
36439 breaks, under the usual common-sense rules. The XML version
36440 declaration and document type declaration can generally be omitted
36441 (@value{GDBN} does not require them), but specifying them may be
36442 useful for XML validation tools. The @samp{version} attribute for
36443 @samp{<target>} may also be omitted, but we recommend
36444 including it; if future versions of @value{GDBN} use an incompatible
36445 revision of @file{gdb-target.dtd}, they will detect and report
36446 the version mismatch.
36448 @subsection Inclusion
36449 @cindex target descriptions, inclusion
36452 @cindex <xi:include>
36455 It can sometimes be valuable to split a target description up into
36456 several different annexes, either for organizational purposes, or to
36457 share files between different possible target descriptions. You can
36458 divide a description into multiple files by replacing any element of
36459 the target description with an inclusion directive of the form:
36462 <xi:include href="@var{document}"/>
36466 When @value{GDBN} encounters an element of this form, it will retrieve
36467 the named XML @var{document}, and replace the inclusion directive with
36468 the contents of that document. If the current description was read
36469 using @samp{qXfer}, then so will be the included document;
36470 @var{document} will be interpreted as the name of an annex. If the
36471 current description was read from a file, @value{GDBN} will look for
36472 @var{document} as a file in the same directory where it found the
36473 original description.
36475 @subsection Architecture
36476 @cindex <architecture>
36478 An @samp{<architecture>} element has this form:
36481 <architecture>@var{arch}</architecture>
36484 @var{arch} is one of the architectures from the set accepted by
36485 @code{set architecture} (@pxref{Targets, ,Specifying a Debugging Target}).
36488 @cindex @code{<osabi>}
36490 This optional field was introduced in @value{GDBN} version 7.0.
36491 Previous versions of @value{GDBN} ignore it.
36493 An @samp{<osabi>} element has this form:
36496 <osabi>@var{abi-name}</osabi>
36499 @var{abi-name} is an OS ABI name from the same selection accepted by
36500 @w{@code{set osabi}} (@pxref{ABI, ,Configuring the Current ABI}).
36502 @subsection Compatible Architecture
36503 @cindex @code{<compatible>}
36505 This optional field was introduced in @value{GDBN} version 7.0.
36506 Previous versions of @value{GDBN} ignore it.
36508 A @samp{<compatible>} element has this form:
36511 <compatible>@var{arch}</compatible>
36514 @var{arch} is one of the architectures from the set accepted by
36515 @code{set architecture} (@pxref{Targets, ,Specifying a Debugging Target}).
36517 A @samp{<compatible>} element is used to specify that the target
36518 is able to run binaries in some other than the main target architecture
36519 given by the @samp{<architecture>} element. For example, on the
36520 Cell Broadband Engine, the main architecture is @code{powerpc:common}
36521 or @code{powerpc:common64}, but the system is able to run binaries
36522 in the @code{spu} architecture as well. The way to describe this
36523 capability with @samp{<compatible>} is as follows:
36526 <architecture>powerpc:common</architecture>
36527 <compatible>spu</compatible>
36530 @subsection Features
36533 Each @samp{<feature>} describes some logical portion of the target
36534 system. Features are currently used to describe available CPU
36535 registers and the types of their contents. A @samp{<feature>} element
36539 <feature name="@var{name}">
36540 @r{[}@var{type}@dots{}@r{]}
36546 Each feature's name should be unique within the description. The name
36547 of a feature does not matter unless @value{GDBN} has some special
36548 knowledge of the contents of that feature; if it does, the feature
36549 should have its standard name. @xref{Standard Target Features}.
36553 Any register's value is a collection of bits which @value{GDBN} must
36554 interpret. The default interpretation is a two's complement integer,
36555 but other types can be requested by name in the register description.
36556 Some predefined types are provided by @value{GDBN} (@pxref{Predefined
36557 Target Types}), and the description can define additional composite types.
36559 Each type element must have an @samp{id} attribute, which gives
36560 a unique (within the containing @samp{<feature>}) name to the type.
36561 Types must be defined before they are used.
36564 Some targets offer vector registers, which can be treated as arrays
36565 of scalar elements. These types are written as @samp{<vector>} elements,
36566 specifying the array element type, @var{type}, and the number of elements,
36570 <vector id="@var{id}" type="@var{type}" count="@var{count}"/>
36574 If a register's value is usefully viewed in multiple ways, define it
36575 with a union type containing the useful representations. The
36576 @samp{<union>} element contains one or more @samp{<field>} elements,
36577 each of which has a @var{name} and a @var{type}:
36580 <union id="@var{id}">
36581 <field name="@var{name}" type="@var{type}"/>
36587 If a register's value is composed from several separate values, define
36588 it with a structure type. There are two forms of the @samp{<struct>}
36589 element; a @samp{<struct>} element must either contain only bitfields
36590 or contain no bitfields. If the structure contains only bitfields,
36591 its total size in bytes must be specified, each bitfield must have an
36592 explicit start and end, and bitfields are automatically assigned an
36593 integer type. The field's @var{start} should be less than or
36594 equal to its @var{end}, and zero represents the least significant bit.
36597 <struct id="@var{id}" size="@var{size}">
36598 <field name="@var{name}" start="@var{start}" end="@var{end}"/>
36603 If the structure contains no bitfields, then each field has an
36604 explicit type, and no implicit padding is added.
36607 <struct id="@var{id}">
36608 <field name="@var{name}" type="@var{type}"/>
36614 If a register's value is a series of single-bit flags, define it with
36615 a flags type. The @samp{<flags>} element has an explicit @var{size}
36616 and contains one or more @samp{<field>} elements. Each field has a
36617 @var{name}, a @var{start}, and an @var{end}. Only single-bit flags
36621 <flags id="@var{id}" size="@var{size}">
36622 <field name="@var{name}" start="@var{start}" end="@var{end}"/>
36627 @subsection Registers
36630 Each register is represented as an element with this form:
36633 <reg name="@var{name}"
36634 bitsize="@var{size}"
36635 @r{[}regnum="@var{num}"@r{]}
36636 @r{[}save-restore="@var{save-restore}"@r{]}
36637 @r{[}type="@var{type}"@r{]}
36638 @r{[}group="@var{group}"@r{]}/>
36642 The components are as follows:
36647 The register's name; it must be unique within the target description.
36650 The register's size, in bits.
36653 The register's number. If omitted, a register's number is one greater
36654 than that of the previous register (either in the current feature or in
36655 a preceeding feature); the first register in the target description
36656 defaults to zero. This register number is used to read or write
36657 the register; e.g.@: it is used in the remote @code{p} and @code{P}
36658 packets, and registers appear in the @code{g} and @code{G} packets
36659 in order of increasing register number.
36662 Whether the register should be preserved across inferior function
36663 calls; this must be either @code{yes} or @code{no}. The default is
36664 @code{yes}, which is appropriate for most registers except for
36665 some system control registers; this is not related to the target's
36669 The type of the register. @var{type} may be a predefined type, a type
36670 defined in the current feature, or one of the special types @code{int}
36671 and @code{float}. @code{int} is an integer type of the correct size
36672 for @var{bitsize}, and @code{float} is a floating point type (in the
36673 architecture's normal floating point format) of the correct size for
36674 @var{bitsize}. The default is @code{int}.
36677 The register group to which this register belongs. @var{group} must
36678 be either @code{general}, @code{float}, or @code{vector}. If no
36679 @var{group} is specified, @value{GDBN} will not display the register
36680 in @code{info registers}.
36684 @node Predefined Target Types
36685 @section Predefined Target Types
36686 @cindex target descriptions, predefined types
36688 Type definitions in the self-description can build up composite types
36689 from basic building blocks, but can not define fundamental types. Instead,
36690 standard identifiers are provided by @value{GDBN} for the fundamental
36691 types. The currently supported types are:
36700 Signed integer types holding the specified number of bits.
36707 Unsigned integer types holding the specified number of bits.
36711 Pointers to unspecified code and data. The program counter and
36712 any dedicated return address register may be marked as code
36713 pointers; printing a code pointer converts it into a symbolic
36714 address. The stack pointer and any dedicated address registers
36715 may be marked as data pointers.
36718 Single precision IEEE floating point.
36721 Double precision IEEE floating point.
36724 The 12-byte extended precision format used by ARM FPA registers.
36727 The 10-byte extended precision format used by x87 registers.
36730 32bit @sc{eflags} register used by x86.
36733 32bit @sc{mxcsr} register used by x86.
36737 @node Standard Target Features
36738 @section Standard Target Features
36739 @cindex target descriptions, standard features
36741 A target description must contain either no registers or all the
36742 target's registers. If the description contains no registers, then
36743 @value{GDBN} will assume a default register layout, selected based on
36744 the architecture. If the description contains any registers, the
36745 default layout will not be used; the standard registers must be
36746 described in the target description, in such a way that @value{GDBN}
36747 can recognize them.
36749 This is accomplished by giving specific names to feature elements
36750 which contain standard registers. @value{GDBN} will look for features
36751 with those names and verify that they contain the expected registers;
36752 if any known feature is missing required registers, or if any required
36753 feature is missing, @value{GDBN} will reject the target
36754 description. You can add additional registers to any of the
36755 standard features --- @value{GDBN} will display them just as if
36756 they were added to an unrecognized feature.
36758 This section lists the known features and their expected contents.
36759 Sample XML documents for these features are included in the
36760 @value{GDBN} source tree, in the directory @file{gdb/features}.
36762 Names recognized by @value{GDBN} should include the name of the
36763 company or organization which selected the name, and the overall
36764 architecture to which the feature applies; so e.g.@: the feature
36765 containing ARM core registers is named @samp{org.gnu.gdb.arm.core}.
36767 The names of registers are not case sensitive for the purpose
36768 of recognizing standard features, but @value{GDBN} will only display
36769 registers using the capitalization used in the description.
36776 * PowerPC Features::
36781 @subsection ARM Features
36782 @cindex target descriptions, ARM features
36784 The @samp{org.gnu.gdb.arm.core} feature is required for non-M-profile
36786 It should contain registers @samp{r0} through @samp{r13}, @samp{sp},
36787 @samp{lr}, @samp{pc}, and @samp{cpsr}.
36789 For M-profile targets (e.g. Cortex-M3), the @samp{org.gnu.gdb.arm.core}
36790 feature is replaced by @samp{org.gnu.gdb.arm.m-profile}. It should contain
36791 registers @samp{r0} through @samp{r13}, @samp{sp}, @samp{lr}, @samp{pc},
36794 The @samp{org.gnu.gdb.arm.fpa} feature is optional. If present, it
36795 should contain registers @samp{f0} through @samp{f7} and @samp{fps}.
36797 The @samp{org.gnu.gdb.xscale.iwmmxt} feature is optional. If present,
36798 it should contain at least registers @samp{wR0} through @samp{wR15} and
36799 @samp{wCGR0} through @samp{wCGR3}. The @samp{wCID}, @samp{wCon},
36800 @samp{wCSSF}, and @samp{wCASF} registers are optional.
36802 The @samp{org.gnu.gdb.arm.vfp} feature is optional. If present, it
36803 should contain at least registers @samp{d0} through @samp{d15}. If
36804 they are present, @samp{d16} through @samp{d31} should also be included.
36805 @value{GDBN} will synthesize the single-precision registers from
36806 halves of the double-precision registers.
36808 The @samp{org.gnu.gdb.arm.neon} feature is optional. It does not
36809 need to contain registers; it instructs @value{GDBN} to display the
36810 VFP double-precision registers as vectors and to synthesize the
36811 quad-precision registers from pairs of double-precision registers.
36812 If this feature is present, @samp{org.gnu.gdb.arm.vfp} must also
36813 be present and include 32 double-precision registers.
36815 @node i386 Features
36816 @subsection i386 Features
36817 @cindex target descriptions, i386 features
36819 The @samp{org.gnu.gdb.i386.core} feature is required for i386/amd64
36820 targets. It should describe the following registers:
36824 @samp{eax} through @samp{edi} plus @samp{eip} for i386
36826 @samp{rax} through @samp{r15} plus @samp{rip} for amd64
36828 @samp{eflags}, @samp{cs}, @samp{ss}, @samp{ds}, @samp{es},
36829 @samp{fs}, @samp{gs}
36831 @samp{st0} through @samp{st7}
36833 @samp{fctrl}, @samp{fstat}, @samp{ftag}, @samp{fiseg}, @samp{fioff},
36834 @samp{foseg}, @samp{fooff} and @samp{fop}
36837 The register sets may be different, depending on the target.
36839 The @samp{org.gnu.gdb.i386.sse} feature is optional. It should
36840 describe registers:
36844 @samp{xmm0} through @samp{xmm7} for i386
36846 @samp{xmm0} through @samp{xmm15} for amd64
36851 The @samp{org.gnu.gdb.i386.avx} feature is optional and requires the
36852 @samp{org.gnu.gdb.i386.sse} feature. It should
36853 describe the upper 128 bits of @sc{ymm} registers:
36857 @samp{ymm0h} through @samp{ymm7h} for i386
36859 @samp{ymm0h} through @samp{ymm15h} for amd64
36862 The @samp{org.gnu.gdb.i386.linux} feature is optional. It should
36863 describe a single register, @samp{orig_eax}.
36865 @node MIPS Features
36866 @subsection MIPS Features
36867 @cindex target descriptions, MIPS features
36869 The @samp{org.gnu.gdb.mips.cpu} feature is required for MIPS targets.
36870 It should contain registers @samp{r0} through @samp{r31}, @samp{lo},
36871 @samp{hi}, and @samp{pc}. They may be 32-bit or 64-bit depending
36874 The @samp{org.gnu.gdb.mips.cp0} feature is also required. It should
36875 contain at least the @samp{status}, @samp{badvaddr}, and @samp{cause}
36876 registers. They may be 32-bit or 64-bit depending on the target.
36878 The @samp{org.gnu.gdb.mips.fpu} feature is currently required, though
36879 it may be optional in a future version of @value{GDBN}. It should
36880 contain registers @samp{f0} through @samp{f31}, @samp{fcsr}, and
36881 @samp{fir}. They may be 32-bit or 64-bit depending on the target.
36883 The @samp{org.gnu.gdb.mips.linux} feature is optional. It should
36884 contain a single register, @samp{restart}, which is used by the
36885 Linux kernel to control restartable syscalls.
36887 @node M68K Features
36888 @subsection M68K Features
36889 @cindex target descriptions, M68K features
36892 @item @samp{org.gnu.gdb.m68k.core}
36893 @itemx @samp{org.gnu.gdb.coldfire.core}
36894 @itemx @samp{org.gnu.gdb.fido.core}
36895 One of those features must be always present.
36896 The feature that is present determines which flavor of m68k is
36897 used. The feature that is present should contain registers
36898 @samp{d0} through @samp{d7}, @samp{a0} through @samp{a5}, @samp{fp},
36899 @samp{sp}, @samp{ps} and @samp{pc}.
36901 @item @samp{org.gnu.gdb.coldfire.fp}
36902 This feature is optional. If present, it should contain registers
36903 @samp{fp0} through @samp{fp7}, @samp{fpcontrol}, @samp{fpstatus} and
36907 @node PowerPC Features
36908 @subsection PowerPC Features
36909 @cindex target descriptions, PowerPC features
36911 The @samp{org.gnu.gdb.power.core} feature is required for PowerPC
36912 targets. It should contain registers @samp{r0} through @samp{r31},
36913 @samp{pc}, @samp{msr}, @samp{cr}, @samp{lr}, @samp{ctr}, and
36914 @samp{xer}. They may be 32-bit or 64-bit depending on the target.
36916 The @samp{org.gnu.gdb.power.fpu} feature is optional. It should
36917 contain registers @samp{f0} through @samp{f31} and @samp{fpscr}.
36919 The @samp{org.gnu.gdb.power.altivec} feature is optional. It should
36920 contain registers @samp{vr0} through @samp{vr31}, @samp{vscr},
36923 The @samp{org.gnu.gdb.power.vsx} feature is optional. It should
36924 contain registers @samp{vs0h} through @samp{vs31h}. @value{GDBN}
36925 will combine these registers with the floating point registers
36926 (@samp{f0} through @samp{f31}) and the altivec registers (@samp{vr0}
36927 through @samp{vr31}) to present the 128-bit wide registers @samp{vs0}
36928 through @samp{vs63}, the set of vector registers for POWER7.
36930 The @samp{org.gnu.gdb.power.spe} feature is optional. It should
36931 contain registers @samp{ev0h} through @samp{ev31h}, @samp{acc}, and
36932 @samp{spefscr}. SPE targets should provide 32-bit registers in
36933 @samp{org.gnu.gdb.power.core} and provide the upper halves in
36934 @samp{ev0h} through @samp{ev31h}. @value{GDBN} will combine
36935 these to present registers @samp{ev0} through @samp{ev31} to the
36938 @node Operating System Information
36939 @appendix Operating System Information
36940 @cindex operating system information
36946 Users of @value{GDBN} often wish to obtain information about the state of
36947 the operating system running on the target---for example the list of
36948 processes, or the list of open files. This section describes the
36949 mechanism that makes it possible. This mechanism is similar to the
36950 target features mechanism (@pxref{Target Descriptions}), but focuses
36951 on a different aspect of target.
36953 Operating system information is retrived from the target via the
36954 remote protocol, using @samp{qXfer} requests (@pxref{qXfer osdata
36955 read}). The object name in the request should be @samp{osdata}, and
36956 the @var{annex} identifies the data to be fetched.
36959 @appendixsection Process list
36960 @cindex operating system information, process list
36962 When requesting the process list, the @var{annex} field in the
36963 @samp{qXfer} request should be @samp{processes}. The returned data is
36964 an XML document. The formal syntax of this document is defined in
36965 @file{gdb/features/osdata.dtd}.
36967 An example document is:
36970 <?xml version="1.0"?>
36971 <!DOCTYPE target SYSTEM "osdata.dtd">
36972 <osdata type="processes">
36974 <column name="pid">1</column>
36975 <column name="user">root</column>
36976 <column name="command">/sbin/init</column>
36977 <column name="cores">1,2,3</column>
36982 Each item should include a column whose name is @samp{pid}. The value
36983 of that column should identify the process on the target. The
36984 @samp{user} and @samp{command} columns are optional, and will be
36985 displayed by @value{GDBN}. The @samp{cores} column, if present,
36986 should contain a comma-separated list of cores that this process
36987 is running on. Target may provide additional columns,
36988 which @value{GDBN} currently ignores.
36990 @node Trace File Format
36991 @appendix Trace File Format
36992 @cindex trace file format
36994 The trace file comes in three parts: a header, a textual description
36995 section, and a trace frame section with binary data.
36997 The header has the form @code{\x7fTRACE0\n}. The first byte is
36998 @code{0x7f} so as to indicate that the file contains binary data,
36999 while the @code{0} is a version number that may have different values
37002 The description section consists of multiple lines of @sc{ascii} text
37003 separated by newline characters (@code{0xa}). The lines may include a
37004 variety of optional descriptive or context-setting information, such
37005 as tracepoint definitions or register set size. @value{GDBN} will
37006 ignore any line that it does not recognize. An empty line marks the end
37009 @c FIXME add some specific types of data
37011 The trace frame section consists of a number of consecutive frames.
37012 Each frame begins with a two-byte tracepoint number, followed by a
37013 four-byte size giving the amount of data in the frame. The data in
37014 the frame consists of a number of blocks, each introduced by a
37015 character indicating its type (at least register, memory, and trace
37016 state variable). The data in this section is raw binary, not a
37017 hexadecimal or other encoding; its endianness matches the target's
37020 @c FIXME bi-arch may require endianness/arch info in description section
37023 @item R @var{bytes}
37024 Register block. The number and ordering of bytes matches that of a
37025 @code{g} packet in the remote protocol. Note that these are the
37026 actual bytes, in target order and @value{GDBN} register order, not a
37027 hexadecimal encoding.
37029 @item M @var{address} @var{length} @var{bytes}...
37030 Memory block. This is a contiguous block of memory, at the 8-byte
37031 address @var{address}, with a 2-byte length @var{length}, followed by
37032 @var{length} bytes.
37034 @item V @var{number} @var{value}
37035 Trace state variable block. This records the 8-byte signed value
37036 @var{value} of trace state variable numbered @var{number}.
37040 Future enhancements of the trace file format may include additional types
37043 @node Index Section Format
37044 @appendix @code{.gdb_index} section format
37045 @cindex .gdb_index section format
37046 @cindex index section format
37048 This section documents the index section that is created by @code{save
37049 gdb-index} (@pxref{Index Files}). The index section is
37050 DWARF-specific; some knowledge of DWARF is assumed in this
37053 The mapped index file format is designed to be directly
37054 @code{mmap}able on any architecture. In most cases, a datum is
37055 represented using a little-endian 32-bit integer value, called an
37056 @code{offset_type}. Big endian machines must byte-swap the values
37057 before using them. Exceptions to this rule are noted. The data is
37058 laid out such that alignment is always respected.
37060 A mapped index consists of several areas, laid out in order.
37064 The file header. This is a sequence of values, of @code{offset_type}
37065 unless otherwise noted:
37069 The version number, currently 5. Versions 1, 2 and 3 are obsolete.
37070 Version 4 differs by its hashing function.
37073 The offset, from the start of the file, of the CU list.
37076 The offset, from the start of the file, of the types CU list. Note
37077 that this area can be empty, in which case this offset will be equal
37078 to the next offset.
37081 The offset, from the start of the file, of the address area.
37084 The offset, from the start of the file, of the symbol table.
37087 The offset, from the start of the file, of the constant pool.
37091 The CU list. This is a sequence of pairs of 64-bit little-endian
37092 values, sorted by the CU offset. The first element in each pair is
37093 the offset of a CU in the @code{.debug_info} section. The second
37094 element in each pair is the length of that CU. References to a CU
37095 elsewhere in the map are done using a CU index, which is just the
37096 0-based index into this table. Note that if there are type CUs, then
37097 conceptually CUs and type CUs form a single list for the purposes of
37101 The types CU list. This is a sequence of triplets of 64-bit
37102 little-endian values. In a triplet, the first value is the CU offset,
37103 the second value is the type offset in the CU, and the third value is
37104 the type signature. The types CU list is not sorted.
37107 The address area. The address area consists of a sequence of address
37108 entries. Each address entry has three elements:
37112 The low address. This is a 64-bit little-endian value.
37115 The high address. This is a 64-bit little-endian value. Like
37116 @code{DW_AT_high_pc}, the value is one byte beyond the end.
37119 The CU index. This is an @code{offset_type} value.
37123 The symbol table. This is an open-addressed hash table. The size of
37124 the hash table is always a power of 2.
37126 Each slot in the hash table consists of a pair of @code{offset_type}
37127 values. The first value is the offset of the symbol's name in the
37128 constant pool. The second value is the offset of the CU vector in the
37131 If both values are 0, then this slot in the hash table is empty. This
37132 is ok because while 0 is a valid constant pool index, it cannot be a
37133 valid index for both a string and a CU vector.
37135 The hash value for a table entry is computed by applying an
37136 iterative hash function to the symbol's name. Starting with an
37137 initial value of @code{r = 0}, each (unsigned) character @samp{c} in
37138 the string is incorporated into the hash using the formula depending on the
37143 The formula is @code{r = r * 67 + c - 113}.
37146 The formula is @code{r = r * 67 + tolower (c) - 113}.
37149 The terminating @samp{\0} is not incorporated into the hash.
37151 The step size used in the hash table is computed via
37152 @code{((hash * 17) & (size - 1)) | 1}, where @samp{hash} is the hash
37153 value, and @samp{size} is the size of the hash table. The step size
37154 is used to find the next candidate slot when handling a hash
37157 The names of C@t{++} symbols in the hash table are canonicalized. We
37158 don't currently have a simple description of the canonicalization
37159 algorithm; if you intend to create new index sections, you must read
37163 The constant pool. This is simply a bunch of bytes. It is organized
37164 so that alignment is correct: CU vectors are stored first, followed by
37167 A CU vector in the constant pool is a sequence of @code{offset_type}
37168 values. The first value is the number of CU indices in the vector.
37169 Each subsequent value is the index of a CU in the CU list. This
37170 element in the hash table is used to indicate which CUs define the
37173 A string in the constant pool is zero-terminated.
37178 @node GNU Free Documentation License
37179 @appendix GNU Free Documentation License
37188 % I think something like @colophon should be in texinfo. In the
37190 \long\def\colophon{\hbox to0pt{}\vfill
37191 \centerline{The body of this manual is set in}
37192 \centerline{\fontname\tenrm,}
37193 \centerline{with headings in {\bf\fontname\tenbf}}
37194 \centerline{and examples in {\tt\fontname\tentt}.}
37195 \centerline{{\it\fontname\tenit\/},}
37196 \centerline{{\bf\fontname\tenbf}, and}
37197 \centerline{{\sl\fontname\tensl\/}}
37198 \centerline{are used for emphasis.}\vfill}
37200 % Blame: doc@cygnus.com, 1991.