1 \input texinfo @c -*-texinfo-*-
2 @c Copyright (C) 1988-1996, 1998-2012 Free Software Foundation, Inc.
5 @c makeinfo ignores cmds prev to setfilename, so its arg cannot make use
6 @c of @set vars. However, you can override filename with makeinfo -o.
11 @settitle Debugging with @value{GDBN}
12 @setchapternewpage odd
24 @c readline appendices use @vindex, @findex and @ftable,
25 @c annotate.texi and gdbmi use @findex.
29 @c !!set GDB manual's edition---not the same as GDB version!
30 @c This is updated by GNU Press.
33 @c !!set GDB edit command default editor
36 @c THIS MANUAL REQUIRES TEXINFO 4.0 OR LATER.
38 @c This is a dir.info fragment to support semi-automated addition of
39 @c manuals to an info tree.
40 @dircategory Software development
42 * Gdb: (gdb). The GNU debugger.
46 Copyright @copyright{} 1988, 1989, 1990, 1991, 1992, 1993, 1994, 1995, 1996,
47 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010
49 Free Software Foundation, Inc.
51 Permission is granted to copy, distribute and/or modify this document
52 under the terms of the GNU Free Documentation License, Version 1.3 or
53 any later version published by the Free Software Foundation; with the
54 Invariant Sections being ``Free Software'' and ``Free Software Needs
55 Free Documentation'', with the Front-Cover Texts being ``A GNU Manual,''
56 and with the Back-Cover Texts as in (a) below.
58 (a) The FSF's Back-Cover Text is: ``You are free to copy and modify
59 this GNU Manual. Buying copies from GNU Press supports the FSF in
60 developing GNU and promoting software freedom.''
64 This file documents the @sc{gnu} debugger @value{GDBN}.
66 This is the @value{EDITION} Edition, of @cite{Debugging with
67 @value{GDBN}: the @sc{gnu} Source-Level Debugger} for @value{GDBN}
68 @ifset VERSION_PACKAGE
69 @value{VERSION_PACKAGE}
71 Version @value{GDBVN}.
77 @title Debugging with @value{GDBN}
78 @subtitle The @sc{gnu} Source-Level Debugger
80 @subtitle @value{EDITION} Edition, for @value{GDBN} version @value{GDBVN}
81 @ifset VERSION_PACKAGE
83 @subtitle @value{VERSION_PACKAGE}
85 @author Richard Stallman, Roland Pesch, Stan Shebs, et al.
89 \hfill (Send bugs and comments on @value{GDBN} to @value{BUGURL}.)\par
90 \hfill {\it Debugging with @value{GDBN}}\par
91 \hfill \TeX{}info \texinfoversion\par
95 @vskip 0pt plus 1filll
96 Published by the Free Software Foundation @*
97 51 Franklin Street, Fifth Floor,
98 Boston, MA 02110-1301, USA@*
99 ISBN 978-0-9831592-3-0 @*
106 @node Top, Summary, (dir), (dir)
108 @top Debugging with @value{GDBN}
110 This file describes @value{GDBN}, the @sc{gnu} symbolic debugger.
112 This is the @value{EDITION} Edition, for @value{GDBN}
113 @ifset VERSION_PACKAGE
114 @value{VERSION_PACKAGE}
116 Version @value{GDBVN}.
118 Copyright (C) 1988-2012 Free Software Foundation, Inc.
120 This edition of the GDB manual is dedicated to the memory of Fred
121 Fish. Fred was a long-standing contributor to GDB and to Free
122 software in general. We will miss him.
125 * Summary:: Summary of @value{GDBN}
126 * Sample Session:: A sample @value{GDBN} session
128 * Invocation:: Getting in and out of @value{GDBN}
129 * Commands:: @value{GDBN} commands
130 * Running:: Running programs under @value{GDBN}
131 * Stopping:: Stopping and continuing
132 * Reverse Execution:: Running programs backward
133 * Process Record and Replay:: Recording inferior's execution and replaying it
134 * Stack:: Examining the stack
135 * Source:: Examining source files
136 * Data:: Examining data
137 * Optimized Code:: Debugging optimized code
138 * Macros:: Preprocessor Macros
139 * Tracepoints:: Debugging remote targets non-intrusively
140 * Overlays:: Debugging programs that use overlays
142 * Languages:: Using @value{GDBN} with different languages
144 * Symbols:: Examining the symbol table
145 * Altering:: Altering execution
146 * GDB Files:: @value{GDBN} files
147 * Targets:: Specifying a debugging target
148 * Remote Debugging:: Debugging remote programs
149 * Configurations:: Configuration-specific information
150 * Controlling GDB:: Controlling @value{GDBN}
151 * Extending GDB:: Extending @value{GDBN}
152 * Interpreters:: Command Interpreters
153 * TUI:: @value{GDBN} Text User Interface
154 * Emacs:: Using @value{GDBN} under @sc{gnu} Emacs
155 * GDB/MI:: @value{GDBN}'s Machine Interface.
156 * Annotations:: @value{GDBN}'s annotation interface.
157 * JIT Interface:: Using the JIT debugging interface.
158 * In-Process Agent:: In-Process Agent
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 -init-command @var{file}
995 @itemx -ix @var{file}
996 @cindex @code{--init-command}
998 Execute commands from file @var{file} before loading gdbinit files or the
1002 @item -init-eval-command @var{command}
1003 @itemx -iex @var{command}
1004 @cindex @code{--init-eval-command}
1006 Execute a single @value{GDBN} command before loading gdbinit files or the
1010 @item -directory @var{directory}
1011 @itemx -d @var{directory}
1012 @cindex @code{--directory}
1014 Add @var{directory} to the path to search for source and script files.
1018 @cindex @code{--readnow}
1020 Read each symbol file's entire symbol table immediately, rather than
1021 the default, which is to read it incrementally as it is needed.
1022 This makes startup slower, but makes future operations faster.
1027 @subsection Choosing Modes
1029 You can run @value{GDBN} in various alternative modes---for example, in
1030 batch mode or quiet mode.
1038 Do not execute commands found in any initialization files. Normally,
1039 @value{GDBN} executes the commands in these files after all the command
1040 options and arguments have been processed. @xref{Command Files,,Command
1046 @cindex @code{--quiet}
1047 @cindex @code{--silent}
1049 ``Quiet''. Do not print the introductory and copyright messages. These
1050 messages are also suppressed in batch mode.
1053 @cindex @code{--batch}
1054 Run in batch mode. Exit with status @code{0} after processing all the
1055 command files specified with @samp{-x} (and all commands from
1056 initialization files, if not inhibited with @samp{-n}). Exit with
1057 nonzero status if an error occurs in executing the @value{GDBN} commands
1058 in the command files. Batch mode also disables pagination, sets unlimited
1059 terminal width and height @pxref{Screen Size}, and acts as if @kbd{set confirm
1060 off} were in effect (@pxref{Messages/Warnings}).
1062 Batch mode may be useful for running @value{GDBN} as a filter, for
1063 example to download and run a program on another computer; in order to
1064 make this more useful, the message
1067 Program exited normally.
1071 (which is ordinarily issued whenever a program running under
1072 @value{GDBN} control terminates) is not issued when running in batch
1076 @cindex @code{--batch-silent}
1077 Run in batch mode exactly like @samp{-batch}, but totally silently. All
1078 @value{GDBN} output to @code{stdout} is prevented (@code{stderr} is
1079 unaffected). This is much quieter than @samp{-silent} and would be useless
1080 for an interactive session.
1082 This is particularly useful when using targets that give @samp{Loading section}
1083 messages, for example.
1085 Note that targets that give their output via @value{GDBN}, as opposed to
1086 writing directly to @code{stdout}, will also be made silent.
1088 @item -return-child-result
1089 @cindex @code{--return-child-result}
1090 The return code from @value{GDBN} will be the return code from the child
1091 process (the process being debugged), with the following exceptions:
1095 @value{GDBN} exits abnormally. E.g., due to an incorrect argument or an
1096 internal error. In this case the exit code is the same as it would have been
1097 without @samp{-return-child-result}.
1099 The user quits with an explicit value. E.g., @samp{quit 1}.
1101 The child process never runs, or is not allowed to terminate, in which case
1102 the exit code will be -1.
1105 This option is useful in conjunction with @samp{-batch} or @samp{-batch-silent},
1106 when @value{GDBN} is being used as a remote program loader or simulator
1111 @cindex @code{--nowindows}
1113 ``No windows''. If @value{GDBN} comes with a graphical user interface
1114 (GUI) built in, then this option tells @value{GDBN} to only use the command-line
1115 interface. If no GUI is available, this option has no effect.
1119 @cindex @code{--windows}
1121 If @value{GDBN} includes a GUI, then this option requires it to be
1124 @item -cd @var{directory}
1126 Run @value{GDBN} using @var{directory} as its working directory,
1127 instead of the current directory.
1129 @item -data-directory @var{directory}
1130 @cindex @code{--data-directory}
1131 Run @value{GDBN} using @var{directory} as its data directory.
1132 The data directory is where @value{GDBN} searches for its
1133 auxiliary files. @xref{Data Files}.
1137 @cindex @code{--fullname}
1139 @sc{gnu} Emacs sets this option when it runs @value{GDBN} as a
1140 subprocess. It tells @value{GDBN} to output the full file name and line
1141 number in a standard, recognizable fashion each time a stack frame is
1142 displayed (which includes each time your program stops). This
1143 recognizable format looks like two @samp{\032} characters, followed by
1144 the file name, line number and character position separated by colons,
1145 and a newline. The Emacs-to-@value{GDBN} interface program uses the two
1146 @samp{\032} characters as a signal to display the source code for the
1150 @cindex @code{--epoch}
1151 The Epoch Emacs-@value{GDBN} interface sets this option when it runs
1152 @value{GDBN} as a subprocess. It tells @value{GDBN} to modify its print
1153 routines so as to allow Epoch to display values of expressions in a
1156 @item -annotate @var{level}
1157 @cindex @code{--annotate}
1158 This option sets the @dfn{annotation level} inside @value{GDBN}. Its
1159 effect is identical to using @samp{set annotate @var{level}}
1160 (@pxref{Annotations}). The annotation @var{level} controls how much
1161 information @value{GDBN} prints together with its prompt, values of
1162 expressions, source lines, and other types of output. Level 0 is the
1163 normal, level 1 is for use when @value{GDBN} is run as a subprocess of
1164 @sc{gnu} Emacs, level 3 is the maximum annotation suitable for programs
1165 that control @value{GDBN}, and level 2 has been deprecated.
1167 The annotation mechanism has largely been superseded by @sc{gdb/mi}
1171 @cindex @code{--args}
1172 Change interpretation of command line so that arguments following the
1173 executable file are passed as command line arguments to the inferior.
1174 This option stops option processing.
1176 @item -baud @var{bps}
1178 @cindex @code{--baud}
1180 Set the line speed (baud rate or bits per second) of any serial
1181 interface used by @value{GDBN} for remote debugging.
1183 @item -l @var{timeout}
1185 Set the timeout (in seconds) of any communication used by @value{GDBN}
1186 for remote debugging.
1188 @item -tty @var{device}
1189 @itemx -t @var{device}
1190 @cindex @code{--tty}
1192 Run using @var{device} for your program's standard input and output.
1193 @c FIXME: kingdon thinks there is more to -tty. Investigate.
1195 @c resolve the situation of these eventually
1197 @cindex @code{--tui}
1198 Activate the @dfn{Text User Interface} when starting. The Text User
1199 Interface manages several text windows on the terminal, showing
1200 source, assembly, registers and @value{GDBN} command outputs
1201 (@pxref{TUI, ,@value{GDBN} Text User Interface}). Do not use this
1202 option if you run @value{GDBN} from Emacs (@pxref{Emacs, ,
1203 Using @value{GDBN} under @sc{gnu} Emacs}).
1206 @c @cindex @code{--xdb}
1207 @c Run in XDB compatibility mode, allowing the use of certain XDB commands.
1208 @c For information, see the file @file{xdb_trans.html}, which is usually
1209 @c installed in the directory @code{/opt/langtools/wdb/doc} on HP-UX
1212 @item -interpreter @var{interp}
1213 @cindex @code{--interpreter}
1214 Use the interpreter @var{interp} for interface with the controlling
1215 program or device. This option is meant to be set by programs which
1216 communicate with @value{GDBN} using it as a back end.
1217 @xref{Interpreters, , Command Interpreters}.
1219 @samp{--interpreter=mi} (or @samp{--interpreter=mi2}) causes
1220 @value{GDBN} to use the @dfn{@sc{gdb/mi} interface} (@pxref{GDB/MI, ,
1221 The @sc{gdb/mi} Interface}) included since @value{GDBN} version 6.0. The
1222 previous @sc{gdb/mi} interface, included in @value{GDBN} version 5.3 and
1223 selected with @samp{--interpreter=mi1}, is deprecated. Earlier
1224 @sc{gdb/mi} interfaces are no longer supported.
1227 @cindex @code{--write}
1228 Open the executable and core files for both reading and writing. This
1229 is equivalent to the @samp{set write on} command inside @value{GDBN}
1233 @cindex @code{--statistics}
1234 This option causes @value{GDBN} to print statistics about time and
1235 memory usage after it completes each command and returns to the prompt.
1238 @cindex @code{--version}
1239 This option causes @value{GDBN} to print its version number and
1240 no-warranty blurb, and exit.
1242 @item -use-deprecated-index-sections
1243 @cindex @code{--use-deprecated-index-sections}
1244 This option causes @value{GDBN} to read and use deprecated
1245 @samp{.gdb_index} sections from symbol files. This can speed up
1246 startup, but may result in some functionality being lost.
1247 @xref{Index Section Format}.
1252 @subsection What @value{GDBN} Does During Startup
1253 @cindex @value{GDBN} startup
1255 Here's the description of what @value{GDBN} does during session startup:
1259 Sets up the command interpreter as specified by the command line
1260 (@pxref{Mode Options, interpreter}).
1262 @anchor{Option -init-eval-command}
1264 Executes commands and command files specified by the @samp{-iex} and
1265 @samp{-ix} options in their specified order. Usually you should use the
1266 @samp{-ex} and @samp{-x} options instead, but this way you can apply
1267 settings before @value{GDBN} init files get executed and before inferior
1272 Reads the system-wide @dfn{init file} (if @option{--with-system-gdbinit} was
1273 used when building @value{GDBN}; @pxref{System-wide configuration,
1274 ,System-wide configuration and settings}) and executes all the commands in
1277 @anchor{Home Directory Init File}
1279 Reads the init file (if any) in your home directory@footnote{On
1280 DOS/Windows systems, the home directory is the one pointed to by the
1281 @code{HOME} environment variable.} and executes all the commands in
1285 Processes command line options and operands.
1287 @anchor{Init File in the Current Directory during Startup}
1289 Reads and executes the commands from init file (if any) in the current
1290 working directory as long as @samp{set auto-load local-gdbinit} is set to
1291 @samp{on} (@pxref{Init File in the Current Directory}).
1292 This is only done if the current directory is
1293 different from your home directory. Thus, you can have more than one
1294 init file, one generic in your home directory, and another, specific
1295 to the program you are debugging, in the directory where you invoke
1299 If the command line specified a program to debug, or a process to
1300 attach to, or a core file, @value{GDBN} loads any auto-loaded
1301 scripts provided for the program or for its loaded shared libraries.
1302 @xref{Auto-loading}.
1304 If you wish to disable the auto-loading during startup,
1305 you must do something like the following:
1308 $ gdb -iex "set auto-load python-scripts off" myprogram
1311 Option @samp{-ex} does not work because the auto-loading is then turned
1315 Executes commands and command files specified by the @samp{-ex} and
1316 @samp{-x} options in their specified order. @xref{Command Files}, for
1317 more details about @value{GDBN} command files.
1320 Reads the command history recorded in the @dfn{history file}.
1321 @xref{Command History}, for more details about the command history and the
1322 files where @value{GDBN} records it.
1325 Init files use the same syntax as @dfn{command files} (@pxref{Command
1326 Files}) and are processed by @value{GDBN} in the same way. The init
1327 file in your home directory can set options (such as @samp{set
1328 complaints}) that affect subsequent processing of command line options
1329 and operands. Init files are not executed if you use the @samp{-nx}
1330 option (@pxref{Mode Options, ,Choosing Modes}).
1332 To display the list of init files loaded by gdb at startup, you
1333 can use @kbd{gdb --help}.
1335 @cindex init file name
1336 @cindex @file{.gdbinit}
1337 @cindex @file{gdb.ini}
1338 The @value{GDBN} init files are normally called @file{.gdbinit}.
1339 The DJGPP port of @value{GDBN} uses the name @file{gdb.ini}, due to
1340 the limitations of file names imposed by DOS filesystems. The Windows
1341 ports of @value{GDBN} use the standard name, but if they find a
1342 @file{gdb.ini} file, they warn you about that and suggest to rename
1343 the file to the standard name.
1347 @section Quitting @value{GDBN}
1348 @cindex exiting @value{GDBN}
1349 @cindex leaving @value{GDBN}
1352 @kindex quit @r{[}@var{expression}@r{]}
1353 @kindex q @r{(@code{quit})}
1354 @item quit @r{[}@var{expression}@r{]}
1356 To exit @value{GDBN}, use the @code{quit} command (abbreviated
1357 @code{q}), or type an end-of-file character (usually @kbd{Ctrl-d}). If you
1358 do not supply @var{expression}, @value{GDBN} will terminate normally;
1359 otherwise it will terminate using the result of @var{expression} as the
1364 An interrupt (often @kbd{Ctrl-c}) does not exit from @value{GDBN}, but rather
1365 terminates the action of any @value{GDBN} command that is in progress and
1366 returns to @value{GDBN} command level. It is safe to type the interrupt
1367 character at any time because @value{GDBN} does not allow it to take effect
1368 until a time when it is safe.
1370 If you have been using @value{GDBN} to control an attached process or
1371 device, you can release it with the @code{detach} command
1372 (@pxref{Attach, ,Debugging an Already-running Process}).
1374 @node Shell Commands
1375 @section Shell Commands
1377 If you need to execute occasional shell commands during your
1378 debugging session, there is no need to leave or suspend @value{GDBN}; you can
1379 just use the @code{shell} command.
1384 @cindex shell escape
1385 @item shell @var{command-string}
1386 @itemx !@var{command-string}
1387 Invoke a standard shell to execute @var{command-string}.
1388 Note that no space is needed between @code{!} and @var{command-string}.
1389 If it exists, the environment variable @code{SHELL} determines which
1390 shell to run. Otherwise @value{GDBN} uses the default shell
1391 (@file{/bin/sh} on Unix systems, @file{COMMAND.COM} on MS-DOS, etc.).
1394 The utility @code{make} is often needed in development environments.
1395 You do not have to use the @code{shell} command for this purpose in
1400 @cindex calling make
1401 @item make @var{make-args}
1402 Execute the @code{make} program with the specified
1403 arguments. This is equivalent to @samp{shell make @var{make-args}}.
1406 @node Logging Output
1407 @section Logging Output
1408 @cindex logging @value{GDBN} output
1409 @cindex save @value{GDBN} output to a file
1411 You may want to save the output of @value{GDBN} commands to a file.
1412 There are several commands to control @value{GDBN}'s logging.
1416 @item set logging on
1418 @item set logging off
1420 @cindex logging file name
1421 @item set logging file @var{file}
1422 Change the name of the current logfile. The default logfile is @file{gdb.txt}.
1423 @item set logging overwrite [on|off]
1424 By default, @value{GDBN} will append to the logfile. Set @code{overwrite} if
1425 you want @code{set logging on} to overwrite the logfile instead.
1426 @item set logging redirect [on|off]
1427 By default, @value{GDBN} output will go to both the terminal and the logfile.
1428 Set @code{redirect} if you want output to go only to the log file.
1429 @kindex show logging
1431 Show the current values of the logging settings.
1435 @chapter @value{GDBN} Commands
1437 You can abbreviate a @value{GDBN} command to the first few letters of the command
1438 name, if that abbreviation is unambiguous; and you can repeat certain
1439 @value{GDBN} commands by typing just @key{RET}. You can also use the @key{TAB}
1440 key to get @value{GDBN} to fill out the rest of a word in a command (or to
1441 show you the alternatives available, if there is more than one possibility).
1444 * Command Syntax:: How to give commands to @value{GDBN}
1445 * Completion:: Command completion
1446 * Help:: How to ask @value{GDBN} for help
1449 @node Command Syntax
1450 @section Command Syntax
1452 A @value{GDBN} command is a single line of input. There is no limit on
1453 how long it can be. It starts with a command name, which is followed by
1454 arguments whose meaning depends on the command name. For example, the
1455 command @code{step} accepts an argument which is the number of times to
1456 step, as in @samp{step 5}. You can also use the @code{step} command
1457 with no arguments. Some commands do not allow any arguments.
1459 @cindex abbreviation
1460 @value{GDBN} command names may always be truncated if that abbreviation is
1461 unambiguous. Other possible command abbreviations are listed in the
1462 documentation for individual commands. In some cases, even ambiguous
1463 abbreviations are allowed; for example, @code{s} is specially defined as
1464 equivalent to @code{step} even though there are other commands whose
1465 names start with @code{s}. You can test abbreviations by using them as
1466 arguments to the @code{help} command.
1468 @cindex repeating commands
1469 @kindex RET @r{(repeat last command)}
1470 A blank line as input to @value{GDBN} (typing just @key{RET}) means to
1471 repeat the previous command. Certain commands (for example, @code{run})
1472 will not repeat this way; these are commands whose unintentional
1473 repetition might cause trouble and which you are unlikely to want to
1474 repeat. User-defined commands can disable this feature; see
1475 @ref{Define, dont-repeat}.
1477 The @code{list} and @code{x} commands, when you repeat them with
1478 @key{RET}, construct new arguments rather than repeating
1479 exactly as typed. This permits easy scanning of source or memory.
1481 @value{GDBN} can also use @key{RET} in another way: to partition lengthy
1482 output, in a way similar to the common utility @code{more}
1483 (@pxref{Screen Size,,Screen Size}). Since it is easy to press one
1484 @key{RET} too many in this situation, @value{GDBN} disables command
1485 repetition after any command that generates this sort of display.
1487 @kindex # @r{(a comment)}
1489 Any text from a @kbd{#} to the end of the line is a comment; it does
1490 nothing. This is useful mainly in command files (@pxref{Command
1491 Files,,Command Files}).
1493 @cindex repeating command sequences
1494 @kindex Ctrl-o @r{(operate-and-get-next)}
1495 The @kbd{Ctrl-o} binding is useful for repeating a complex sequence of
1496 commands. This command accepts the current line, like @key{RET}, and
1497 then fetches the next line relative to the current line from the history
1501 @section Command Completion
1504 @cindex word completion
1505 @value{GDBN} can fill in the rest of a word in a command for you, if there is
1506 only one possibility; it can also show you what the valid possibilities
1507 are for the next word in a command, at any time. This works for @value{GDBN}
1508 commands, @value{GDBN} subcommands, and the names of symbols in your program.
1510 Press the @key{TAB} key whenever you want @value{GDBN} to fill out the rest
1511 of a word. If there is only one possibility, @value{GDBN} fills in the
1512 word, and waits for you to finish the command (or press @key{RET} to
1513 enter it). For example, if you type
1515 @c FIXME "@key" does not distinguish its argument sufficiently to permit
1516 @c complete accuracy in these examples; space introduced for clarity.
1517 @c If texinfo enhancements make it unnecessary, it would be nice to
1518 @c replace " @key" by "@key" in the following...
1520 (@value{GDBP}) info bre @key{TAB}
1524 @value{GDBN} fills in the rest of the word @samp{breakpoints}, since that is
1525 the only @code{info} subcommand beginning with @samp{bre}:
1528 (@value{GDBP}) info breakpoints
1532 You can either press @key{RET} at this point, to run the @code{info
1533 breakpoints} command, or backspace and enter something else, if
1534 @samp{breakpoints} does not look like the command you expected. (If you
1535 were sure you wanted @code{info breakpoints} in the first place, you
1536 might as well just type @key{RET} immediately after @samp{info bre},
1537 to exploit command abbreviations rather than command completion).
1539 If there is more than one possibility for the next word when you press
1540 @key{TAB}, @value{GDBN} sounds a bell. You can either supply more
1541 characters and try again, or just press @key{TAB} a second time;
1542 @value{GDBN} displays all the possible completions for that word. For
1543 example, you might want to set a breakpoint on a subroutine whose name
1544 begins with @samp{make_}, but when you type @kbd{b make_@key{TAB}} @value{GDBN}
1545 just sounds the bell. Typing @key{TAB} again displays all the
1546 function names in your program that begin with those characters, for
1550 (@value{GDBP}) b make_ @key{TAB}
1551 @exdent @value{GDBN} sounds bell; press @key{TAB} again, to see:
1552 make_a_section_from_file make_environ
1553 make_abs_section make_function_type
1554 make_blockvector make_pointer_type
1555 make_cleanup make_reference_type
1556 make_command make_symbol_completion_list
1557 (@value{GDBP}) b make_
1561 After displaying the available possibilities, @value{GDBN} copies your
1562 partial input (@samp{b make_} in the example) so you can finish the
1565 If you just want to see the list of alternatives in the first place, you
1566 can press @kbd{M-?} rather than pressing @key{TAB} twice. @kbd{M-?}
1567 means @kbd{@key{META} ?}. You can type this either by holding down a
1568 key designated as the @key{META} shift on your keyboard (if there is
1569 one) while typing @kbd{?}, or as @key{ESC} followed by @kbd{?}.
1571 @cindex quotes in commands
1572 @cindex completion of quoted strings
1573 Sometimes the string you need, while logically a ``word'', may contain
1574 parentheses or other characters that @value{GDBN} normally excludes from
1575 its notion of a word. To permit word completion to work in this
1576 situation, you may enclose words in @code{'} (single quote marks) in
1577 @value{GDBN} commands.
1579 The most likely situation where you might need this is in typing the
1580 name of a C@t{++} function. This is because C@t{++} allows function
1581 overloading (multiple definitions of the same function, distinguished
1582 by argument type). For example, when you want to set a breakpoint you
1583 may need to distinguish whether you mean the version of @code{name}
1584 that takes an @code{int} parameter, @code{name(int)}, or the version
1585 that takes a @code{float} parameter, @code{name(float)}. To use the
1586 word-completion facilities in this situation, type a single quote
1587 @code{'} at the beginning of the function name. This alerts
1588 @value{GDBN} that it may need to consider more information than usual
1589 when you press @key{TAB} or @kbd{M-?} to request word completion:
1592 (@value{GDBP}) b 'bubble( @kbd{M-?}
1593 bubble(double,double) bubble(int,int)
1594 (@value{GDBP}) b 'bubble(
1597 In some cases, @value{GDBN} can tell that completing a name requires using
1598 quotes. When this happens, @value{GDBN} inserts the quote for you (while
1599 completing as much as it can) if you do not type the quote in the first
1603 (@value{GDBP}) b bub @key{TAB}
1604 @exdent @value{GDBN} alters your input line to the following, and rings a bell:
1605 (@value{GDBP}) b 'bubble(
1609 In general, @value{GDBN} can tell that a quote is needed (and inserts it) if
1610 you have not yet started typing the argument list when you ask for
1611 completion on an overloaded symbol.
1613 For more information about overloaded functions, see @ref{C Plus Plus
1614 Expressions, ,C@t{++} Expressions}. You can use the command @code{set
1615 overload-resolution off} to disable overload resolution;
1616 see @ref{Debugging C Plus Plus, ,@value{GDBN} Features for C@t{++}}.
1618 @cindex completion of structure field names
1619 @cindex structure field name completion
1620 @cindex completion of union field names
1621 @cindex union field name completion
1622 When completing in an expression which looks up a field in a
1623 structure, @value{GDBN} also tries@footnote{The completer can be
1624 confused by certain kinds of invalid expressions. Also, it only
1625 examines the static type of the expression, not the dynamic type.} to
1626 limit completions to the field names available in the type of the
1630 (@value{GDBP}) p gdb_stdout.@kbd{M-?}
1631 magic to_fputs to_rewind
1632 to_data to_isatty to_write
1633 to_delete to_put to_write_async_safe
1638 This is because the @code{gdb_stdout} is a variable of the type
1639 @code{struct ui_file} that is defined in @value{GDBN} sources as
1646 ui_file_flush_ftype *to_flush;
1647 ui_file_write_ftype *to_write;
1648 ui_file_write_async_safe_ftype *to_write_async_safe;
1649 ui_file_fputs_ftype *to_fputs;
1650 ui_file_read_ftype *to_read;
1651 ui_file_delete_ftype *to_delete;
1652 ui_file_isatty_ftype *to_isatty;
1653 ui_file_rewind_ftype *to_rewind;
1654 ui_file_put_ftype *to_put;
1661 @section Getting Help
1662 @cindex online documentation
1665 You can always ask @value{GDBN} itself for information on its commands,
1666 using the command @code{help}.
1669 @kindex h @r{(@code{help})}
1672 You can use @code{help} (abbreviated @code{h}) with no arguments to
1673 display a short list of named classes of commands:
1677 List of classes of commands:
1679 aliases -- Aliases of other commands
1680 breakpoints -- Making program stop at certain points
1681 data -- Examining data
1682 files -- Specifying and examining files
1683 internals -- Maintenance commands
1684 obscure -- Obscure features
1685 running -- Running the program
1686 stack -- Examining the stack
1687 status -- Status inquiries
1688 support -- Support facilities
1689 tracepoints -- Tracing of program execution without
1690 stopping the program
1691 user-defined -- User-defined commands
1693 Type "help" followed by a class name for a list of
1694 commands in that class.
1695 Type "help" followed by command name for full
1697 Command name abbreviations are allowed if unambiguous.
1700 @c the above line break eliminates huge line overfull...
1702 @item help @var{class}
1703 Using one of the general help classes as an argument, you can get a
1704 list of the individual commands in that class. For example, here is the
1705 help display for the class @code{status}:
1708 (@value{GDBP}) help status
1713 @c Line break in "show" line falsifies real output, but needed
1714 @c to fit in smallbook page size.
1715 info -- Generic command for showing things
1716 about the program being debugged
1717 show -- Generic command for showing things
1720 Type "help" followed by command name for full
1722 Command name abbreviations are allowed if unambiguous.
1726 @item help @var{command}
1727 With a command name as @code{help} argument, @value{GDBN} displays a
1728 short paragraph on how to use that command.
1731 @item apropos @var{args}
1732 The @code{apropos} command searches through all of the @value{GDBN}
1733 commands, and their documentation, for the regular expression specified in
1734 @var{args}. It prints out all matches found. For example:
1745 alias -- Define a new command that is an alias of an existing command
1746 aliases -- Aliases of other commands
1747 d -- Delete some breakpoints or auto-display expressions
1748 del -- Delete some breakpoints or auto-display expressions
1749 delete -- Delete some breakpoints or auto-display expressions
1754 @item complete @var{args}
1755 The @code{complete @var{args}} command lists all the possible completions
1756 for the beginning of a command. Use @var{args} to specify the beginning of the
1757 command you want completed. For example:
1763 @noindent results in:
1774 @noindent This is intended for use by @sc{gnu} Emacs.
1777 In addition to @code{help}, you can use the @value{GDBN} commands @code{info}
1778 and @code{show} to inquire about the state of your program, or the state
1779 of @value{GDBN} itself. Each command supports many topics of inquiry; this
1780 manual introduces each of them in the appropriate context. The listings
1781 under @code{info} and under @code{show} in the Index point to
1782 all the sub-commands. @xref{Index}.
1787 @kindex i @r{(@code{info})}
1789 This command (abbreviated @code{i}) is for describing the state of your
1790 program. For example, you can show the arguments passed to a function
1791 with @code{info args}, list the registers currently in use with @code{info
1792 registers}, or list the breakpoints you have set with @code{info breakpoints}.
1793 You can get a complete list of the @code{info} sub-commands with
1794 @w{@code{help info}}.
1798 You can assign the result of an expression to an environment variable with
1799 @code{set}. For example, you can set the @value{GDBN} prompt to a $-sign with
1800 @code{set prompt $}.
1804 In contrast to @code{info}, @code{show} is for describing the state of
1805 @value{GDBN} itself.
1806 You can change most of the things you can @code{show}, by using the
1807 related command @code{set}; for example, you can control what number
1808 system is used for displays with @code{set radix}, or simply inquire
1809 which is currently in use with @code{show radix}.
1812 To display all the settable parameters and their current
1813 values, you can use @code{show} with no arguments; you may also use
1814 @code{info set}. Both commands produce the same display.
1815 @c FIXME: "info set" violates the rule that "info" is for state of
1816 @c FIXME...program. Ck w/ GNU: "info set" to be called something else,
1817 @c FIXME...or change desc of rule---eg "state of prog and debugging session"?
1821 Here are three miscellaneous @code{show} subcommands, all of which are
1822 exceptional in lacking corresponding @code{set} commands:
1825 @kindex show version
1826 @cindex @value{GDBN} version number
1828 Show what version of @value{GDBN} is running. You should include this
1829 information in @value{GDBN} bug-reports. If multiple versions of
1830 @value{GDBN} are in use at your site, you may need to determine which
1831 version of @value{GDBN} you are running; as @value{GDBN} evolves, new
1832 commands are introduced, and old ones may wither away. Also, many
1833 system vendors ship variant versions of @value{GDBN}, and there are
1834 variant versions of @value{GDBN} in @sc{gnu}/Linux distributions as well.
1835 The version number is the same as the one announced when you start
1838 @kindex show copying
1839 @kindex info copying
1840 @cindex display @value{GDBN} copyright
1843 Display information about permission for copying @value{GDBN}.
1845 @kindex show warranty
1846 @kindex info warranty
1848 @itemx info warranty
1849 Display the @sc{gnu} ``NO WARRANTY'' statement, or a warranty,
1850 if your version of @value{GDBN} comes with one.
1855 @chapter Running Programs Under @value{GDBN}
1857 When you run a program under @value{GDBN}, you must first generate
1858 debugging information when you compile it.
1860 You may start @value{GDBN} with its arguments, if any, in an environment
1861 of your choice. If you are doing native debugging, you may redirect
1862 your program's input and output, debug an already running process, or
1863 kill a child process.
1866 * Compilation:: Compiling for debugging
1867 * Starting:: Starting your program
1868 * Arguments:: Your program's arguments
1869 * Environment:: Your program's environment
1871 * Working Directory:: Your program's working directory
1872 * Input/Output:: Your program's input and output
1873 * Attach:: Debugging an already-running process
1874 * Kill Process:: Killing the child process
1876 * Inferiors and Programs:: Debugging multiple inferiors and programs
1877 * Threads:: Debugging programs with multiple threads
1878 * Forks:: Debugging forks
1879 * Checkpoint/Restart:: Setting a @emph{bookmark} to return to later
1883 @section Compiling for Debugging
1885 In order to debug a program effectively, you need to generate
1886 debugging information when you compile it. This debugging information
1887 is stored in the object file; it describes the data type of each
1888 variable or function and the correspondence between source line numbers
1889 and addresses in the executable code.
1891 To request debugging information, specify the @samp{-g} option when you run
1894 Programs that are to be shipped to your customers are compiled with
1895 optimizations, using the @samp{-O} compiler option. However, some
1896 compilers are unable to handle the @samp{-g} and @samp{-O} options
1897 together. Using those compilers, you cannot generate optimized
1898 executables containing debugging information.
1900 @value{NGCC}, the @sc{gnu} C/C@t{++} compiler, supports @samp{-g} with or
1901 without @samp{-O}, making it possible to debug optimized code. We
1902 recommend that you @emph{always} use @samp{-g} whenever you compile a
1903 program. You may think your program is correct, but there is no sense
1904 in pushing your luck. For more information, see @ref{Optimized Code}.
1906 Older versions of the @sc{gnu} C compiler permitted a variant option
1907 @w{@samp{-gg}} for debugging information. @value{GDBN} no longer supports this
1908 format; if your @sc{gnu} C compiler has this option, do not use it.
1910 @value{GDBN} knows about preprocessor macros and can show you their
1911 expansion (@pxref{Macros}). Most compilers do not include information
1912 about preprocessor macros in the debugging information if you specify
1913 the @option{-g} flag alone. Version 3.1 and later of @value{NGCC},
1914 the @sc{gnu} C compiler, provides macro information if you are using
1915 the DWARF debugging format, and specify the option @option{-g3}.
1917 @xref{Debugging Options,,Options for Debugging Your Program or GCC,
1918 gcc.info, Using the @sc{gnu} Compiler Collection (GCC)}, for more
1919 information on @value{NGCC} options affecting debug information.
1921 You will have the best debugging experience if you use the latest
1922 version of the DWARF debugging format that your compiler supports.
1923 DWARF is currently the most expressive and best supported debugging
1924 format in @value{GDBN}.
1928 @section Starting your Program
1934 @kindex r @r{(@code{run})}
1937 Use the @code{run} command to start your program under @value{GDBN}.
1938 You must first specify the program name (except on VxWorks) with an
1939 argument to @value{GDBN} (@pxref{Invocation, ,Getting In and Out of
1940 @value{GDBN}}), or by using the @code{file} or @code{exec-file} command
1941 (@pxref{Files, ,Commands to Specify Files}).
1945 If you are running your program in an execution environment that
1946 supports processes, @code{run} creates an inferior process and makes
1947 that process run your program. In some environments without processes,
1948 @code{run} jumps to the start of your program. Other targets,
1949 like @samp{remote}, are always running. If you get an error
1950 message like this one:
1953 The "remote" target does not support "run".
1954 Try "help target" or "continue".
1958 then use @code{continue} to run your program. You may need @code{load}
1959 first (@pxref{load}).
1961 The execution of a program is affected by certain information it
1962 receives from its superior. @value{GDBN} provides ways to specify this
1963 information, which you must do @emph{before} starting your program. (You
1964 can change it after starting your program, but such changes only affect
1965 your program the next time you start it.) This information may be
1966 divided into four categories:
1969 @item The @emph{arguments.}
1970 Specify the arguments to give your program as the arguments of the
1971 @code{run} command. If a shell is available on your target, the shell
1972 is used to pass the arguments, so that you may use normal conventions
1973 (such as wildcard expansion or variable substitution) in describing
1975 In Unix systems, you can control which shell is used with the
1976 @code{SHELL} environment variable.
1977 @xref{Arguments, ,Your Program's Arguments}.
1979 @item The @emph{environment.}
1980 Your program normally inherits its environment from @value{GDBN}, but you can
1981 use the @value{GDBN} commands @code{set environment} and @code{unset
1982 environment} to change parts of the environment that affect
1983 your program. @xref{Environment, ,Your Program's Environment}.
1985 @item The @emph{working directory.}
1986 Your program inherits its working directory from @value{GDBN}. You can set
1987 the @value{GDBN} working directory with the @code{cd} command in @value{GDBN}.
1988 @xref{Working Directory, ,Your Program's Working Directory}.
1990 @item The @emph{standard input and output.}
1991 Your program normally uses the same device for standard input and
1992 standard output as @value{GDBN} is using. You can redirect input and output
1993 in the @code{run} command line, or you can use the @code{tty} command to
1994 set a different device for your program.
1995 @xref{Input/Output, ,Your Program's Input and Output}.
1998 @emph{Warning:} While input and output redirection work, you cannot use
1999 pipes to pass the output of the program you are debugging to another
2000 program; if you attempt this, @value{GDBN} is likely to wind up debugging the
2004 When you issue the @code{run} command, your program begins to execute
2005 immediately. @xref{Stopping, ,Stopping and Continuing}, for discussion
2006 of how to arrange for your program to stop. Once your program has
2007 stopped, you may call functions in your program, using the @code{print}
2008 or @code{call} commands. @xref{Data, ,Examining Data}.
2010 If the modification time of your symbol file has changed since the last
2011 time @value{GDBN} read its symbols, @value{GDBN} discards its symbol
2012 table, and reads it again. When it does this, @value{GDBN} tries to retain
2013 your current breakpoints.
2018 @cindex run to main procedure
2019 The name of the main procedure can vary from language to language.
2020 With C or C@t{++}, the main procedure name is always @code{main}, but
2021 other languages such as Ada do not require a specific name for their
2022 main procedure. The debugger provides a convenient way to start the
2023 execution of the program and to stop at the beginning of the main
2024 procedure, depending on the language used.
2026 The @samp{start} command does the equivalent of setting a temporary
2027 breakpoint at the beginning of the main procedure and then invoking
2028 the @samp{run} command.
2030 @cindex elaboration phase
2031 Some programs contain an @dfn{elaboration} phase where some startup code is
2032 executed before the main procedure is called. This depends on the
2033 languages used to write your program. In C@t{++}, for instance,
2034 constructors for static and global objects are executed before
2035 @code{main} is called. It is therefore possible that the debugger stops
2036 before reaching the main procedure. However, the temporary breakpoint
2037 will remain to halt execution.
2039 Specify the arguments to give to your program as arguments to the
2040 @samp{start} command. These arguments will be given verbatim to the
2041 underlying @samp{run} command. Note that the same arguments will be
2042 reused if no argument is provided during subsequent calls to
2043 @samp{start} or @samp{run}.
2045 It is sometimes necessary to debug the program during elaboration. In
2046 these cases, using the @code{start} command would stop the execution of
2047 your program too late, as the program would have already completed the
2048 elaboration phase. Under these circumstances, insert breakpoints in your
2049 elaboration code before running your program.
2051 @kindex set exec-wrapper
2052 @item set exec-wrapper @var{wrapper}
2053 @itemx show exec-wrapper
2054 @itemx unset exec-wrapper
2055 When @samp{exec-wrapper} is set, the specified wrapper is used to
2056 launch programs for debugging. @value{GDBN} starts your program
2057 with a shell command of the form @kbd{exec @var{wrapper}
2058 @var{program}}. Quoting is added to @var{program} and its
2059 arguments, but not to @var{wrapper}, so you should add quotes if
2060 appropriate for your shell. The wrapper runs until it executes
2061 your program, and then @value{GDBN} takes control.
2063 You can use any program that eventually calls @code{execve} with
2064 its arguments as a wrapper. Several standard Unix utilities do
2065 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
2066 with @code{exec "$@@"} will also work.
2068 For example, you can use @code{env} to pass an environment variable to
2069 the debugged program, without setting the variable in your shell's
2073 (@value{GDBP}) set exec-wrapper env 'LD_PRELOAD=libtest.so'
2077 This command is available when debugging locally on most targets, excluding
2078 @sc{djgpp}, Cygwin, MS Windows, and QNX Neutrino.
2080 @kindex set disable-randomization
2081 @item set disable-randomization
2082 @itemx set disable-randomization on
2083 This option (enabled by default in @value{GDBN}) will turn off the native
2084 randomization of the virtual address space of the started program. This option
2085 is useful for multiple debugging sessions to make the execution better
2086 reproducible and memory addresses reusable across debugging sessions.
2088 This feature is implemented only on certain targets, including @sc{gnu}/Linux.
2089 On @sc{gnu}/Linux you can get the same behavior using
2092 (@value{GDBP}) set exec-wrapper setarch `uname -m` -R
2095 @item set disable-randomization off
2096 Leave the behavior of the started executable unchanged. Some bugs rear their
2097 ugly heads only when the program is loaded at certain addresses. If your bug
2098 disappears when you run the program under @value{GDBN}, that might be because
2099 @value{GDBN} by default disables the address randomization on platforms, such
2100 as @sc{gnu}/Linux, which do that for stand-alone programs. Use @kbd{set
2101 disable-randomization off} to try to reproduce such elusive bugs.
2103 On targets where it is available, virtual address space randomization
2104 protects the programs against certain kinds of security attacks. In these
2105 cases the attacker needs to know the exact location of a concrete executable
2106 code. Randomizing its location makes it impossible to inject jumps misusing
2107 a code at its expected addresses.
2109 Prelinking shared libraries provides a startup performance advantage but it
2110 makes addresses in these libraries predictable for privileged processes by
2111 having just unprivileged access at the target system. Reading the shared
2112 library binary gives enough information for assembling the malicious code
2113 misusing it. Still even a prelinked shared library can get loaded at a new
2114 random address just requiring the regular relocation process during the
2115 startup. Shared libraries not already prelinked are always loaded at
2116 a randomly chosen address.
2118 Position independent executables (PIE) contain position independent code
2119 similar to the shared libraries and therefore such executables get loaded at
2120 a randomly chosen address upon startup. PIE executables always load even
2121 already prelinked shared libraries at a random address. You can build such
2122 executable using @command{gcc -fPIE -pie}.
2124 Heap (malloc storage), stack and custom mmap areas are always placed randomly
2125 (as long as the randomization is enabled).
2127 @item show disable-randomization
2128 Show the current setting of the explicit disable of the native randomization of
2129 the virtual address space of the started program.
2134 @section Your Program's Arguments
2136 @cindex arguments (to your program)
2137 The arguments to your program can be specified by the arguments of the
2139 They are passed to a shell, which expands wildcard characters and
2140 performs redirection of I/O, and thence to your program. Your
2141 @code{SHELL} environment variable (if it exists) specifies what shell
2142 @value{GDBN} uses. If you do not define @code{SHELL}, @value{GDBN} uses
2143 the default shell (@file{/bin/sh} on Unix).
2145 On non-Unix systems, the program is usually invoked directly by
2146 @value{GDBN}, which emulates I/O redirection via the appropriate system
2147 calls, and the wildcard characters are expanded by the startup code of
2148 the program, not by the shell.
2150 @code{run} with no arguments uses the same arguments used by the previous
2151 @code{run}, or those set by the @code{set args} command.
2156 Specify the arguments to be used the next time your program is run. If
2157 @code{set args} has no arguments, @code{run} executes your program
2158 with no arguments. Once you have run your program with arguments,
2159 using @code{set args} before the next @code{run} is the only way to run
2160 it again without arguments.
2164 Show the arguments to give your program when it is started.
2168 @section Your Program's Environment
2170 @cindex environment (of your program)
2171 The @dfn{environment} consists of a set of environment variables and
2172 their values. Environment variables conventionally record such things as
2173 your user name, your home directory, your terminal type, and your search
2174 path for programs to run. Usually you set up environment variables with
2175 the shell and they are inherited by all the other programs you run. When
2176 debugging, it can be useful to try running your program with a modified
2177 environment without having to start @value{GDBN} over again.
2181 @item path @var{directory}
2182 Add @var{directory} to the front of the @code{PATH} environment variable
2183 (the search path for executables) that will be passed to your program.
2184 The value of @code{PATH} used by @value{GDBN} does not change.
2185 You may specify several directory names, separated by whitespace or by a
2186 system-dependent separator character (@samp{:} on Unix, @samp{;} on
2187 MS-DOS and MS-Windows). If @var{directory} is already in the path, it
2188 is moved to the front, so it is searched sooner.
2190 You can use the string @samp{$cwd} to refer to whatever is the current
2191 working directory at the time @value{GDBN} searches the path. If you
2192 use @samp{.} instead, it refers to the directory where you executed the
2193 @code{path} command. @value{GDBN} replaces @samp{.} in the
2194 @var{directory} argument (with the current path) before adding
2195 @var{directory} to the search path.
2196 @c 'path' is explicitly nonrepeatable, but RMS points out it is silly to
2197 @c document that, since repeating it would be a no-op.
2201 Display the list of search paths for executables (the @code{PATH}
2202 environment variable).
2204 @kindex show environment
2205 @item show environment @r{[}@var{varname}@r{]}
2206 Print the value of environment variable @var{varname} to be given to
2207 your program when it starts. If you do not supply @var{varname},
2208 print the names and values of all environment variables to be given to
2209 your program. You can abbreviate @code{environment} as @code{env}.
2211 @kindex set environment
2212 @item set environment @var{varname} @r{[}=@var{value}@r{]}
2213 Set environment variable @var{varname} to @var{value}. The value
2214 changes for your program only, not for @value{GDBN} itself. @var{value} may
2215 be any string; the values of environment variables are just strings, and
2216 any interpretation is supplied by your program itself. The @var{value}
2217 parameter is optional; if it is eliminated, the variable is set to a
2219 @c "any string" here does not include leading, trailing
2220 @c blanks. Gnu asks: does anyone care?
2222 For example, this command:
2229 tells the debugged program, when subsequently run, that its user is named
2230 @samp{foo}. (The spaces around @samp{=} are used for clarity here; they
2231 are not actually required.)
2233 @kindex unset environment
2234 @item unset environment @var{varname}
2235 Remove variable @var{varname} from the environment to be passed to your
2236 program. This is different from @samp{set env @var{varname} =};
2237 @code{unset environment} removes the variable from the environment,
2238 rather than assigning it an empty value.
2241 @emph{Warning:} On Unix systems, @value{GDBN} runs your program using
2243 by your @code{SHELL} environment variable if it exists (or
2244 @code{/bin/sh} if not). If your @code{SHELL} variable names a shell
2245 that runs an initialization file---such as @file{.cshrc} for C-shell, or
2246 @file{.bashrc} for BASH---any variables you set in that file affect
2247 your program. You may wish to move setting of environment variables to
2248 files that are only run when you sign on, such as @file{.login} or
2251 @node Working Directory
2252 @section Your Program's Working Directory
2254 @cindex working directory (of your program)
2255 Each time you start your program with @code{run}, it inherits its
2256 working directory from the current working directory of @value{GDBN}.
2257 The @value{GDBN} working directory is initially whatever it inherited
2258 from its parent process (typically the shell), but you can specify a new
2259 working directory in @value{GDBN} with the @code{cd} command.
2261 The @value{GDBN} working directory also serves as a default for the commands
2262 that specify files for @value{GDBN} to operate on. @xref{Files, ,Commands to
2267 @cindex change working directory
2268 @item cd @var{directory}
2269 Set the @value{GDBN} working directory to @var{directory}.
2273 Print the @value{GDBN} working directory.
2276 It is generally impossible to find the current working directory of
2277 the process being debugged (since a program can change its directory
2278 during its run). If you work on a system where @value{GDBN} is
2279 configured with the @file{/proc} support, you can use the @code{info
2280 proc} command (@pxref{SVR4 Process Information}) to find out the
2281 current working directory of the debuggee.
2284 @section Your Program's Input and Output
2289 By default, the program you run under @value{GDBN} does input and output to
2290 the same terminal that @value{GDBN} uses. @value{GDBN} switches the terminal
2291 to its own terminal modes to interact with you, but it records the terminal
2292 modes your program was using and switches back to them when you continue
2293 running your program.
2296 @kindex info terminal
2298 Displays information recorded by @value{GDBN} about the terminal modes your
2302 You can redirect your program's input and/or output using shell
2303 redirection with the @code{run} command. For example,
2310 starts your program, diverting its output to the file @file{outfile}.
2313 @cindex controlling terminal
2314 Another way to specify where your program should do input and output is
2315 with the @code{tty} command. This command accepts a file name as
2316 argument, and causes this file to be the default for future @code{run}
2317 commands. It also resets the controlling terminal for the child
2318 process, for future @code{run} commands. For example,
2325 directs that processes started with subsequent @code{run} commands
2326 default to do input and output on the terminal @file{/dev/ttyb} and have
2327 that as their controlling terminal.
2329 An explicit redirection in @code{run} overrides the @code{tty} command's
2330 effect on the input/output device, but not its effect on the controlling
2333 When you use the @code{tty} command or redirect input in the @code{run}
2334 command, only the input @emph{for your program} is affected. The input
2335 for @value{GDBN} still comes from your terminal. @code{tty} is an alias
2336 for @code{set inferior-tty}.
2338 @cindex inferior tty
2339 @cindex set inferior controlling terminal
2340 You can use the @code{show inferior-tty} command to tell @value{GDBN} to
2341 display the name of the terminal that will be used for future runs of your
2345 @item set inferior-tty /dev/ttyb
2346 @kindex set inferior-tty
2347 Set the tty for the program being debugged to /dev/ttyb.
2349 @item show inferior-tty
2350 @kindex show inferior-tty
2351 Show the current tty for the program being debugged.
2355 @section Debugging an Already-running Process
2360 @item attach @var{process-id}
2361 This command attaches to a running process---one that was started
2362 outside @value{GDBN}. (@code{info files} shows your active
2363 targets.) The command takes as argument a process ID. The usual way to
2364 find out the @var{process-id} of a Unix process is with the @code{ps} utility,
2365 or with the @samp{jobs -l} shell command.
2367 @code{attach} does not repeat if you press @key{RET} a second time after
2368 executing the command.
2371 To use @code{attach}, your program must be running in an environment
2372 which supports processes; for example, @code{attach} does not work for
2373 programs on bare-board targets that lack an operating system. You must
2374 also have permission to send the process a signal.
2376 When you use @code{attach}, the debugger finds the program running in
2377 the process first by looking in the current working directory, then (if
2378 the program is not found) by using the source file search path
2379 (@pxref{Source Path, ,Specifying Source Directories}). You can also use
2380 the @code{file} command to load the program. @xref{Files, ,Commands to
2383 The first thing @value{GDBN} does after arranging to debug the specified
2384 process is to stop it. You can examine and modify an attached process
2385 with all the @value{GDBN} commands that are ordinarily available when
2386 you start processes with @code{run}. You can insert breakpoints; you
2387 can step and continue; you can modify storage. If you would rather the
2388 process continue running, you may use the @code{continue} command after
2389 attaching @value{GDBN} to the process.
2394 When you have finished debugging the attached process, you can use the
2395 @code{detach} command to release it from @value{GDBN} control. Detaching
2396 the process continues its execution. After the @code{detach} command,
2397 that process and @value{GDBN} become completely independent once more, and you
2398 are ready to @code{attach} another process or start one with @code{run}.
2399 @code{detach} does not repeat if you press @key{RET} again after
2400 executing the command.
2403 If you exit @value{GDBN} while you have an attached process, you detach
2404 that process. If you use the @code{run} command, you kill that process.
2405 By default, @value{GDBN} asks for confirmation if you try to do either of these
2406 things; you can control whether or not you need to confirm by using the
2407 @code{set confirm} command (@pxref{Messages/Warnings, ,Optional Warnings and
2411 @section Killing the Child Process
2416 Kill the child process in which your program is running under @value{GDBN}.
2419 This command is useful if you wish to debug a core dump instead of a
2420 running process. @value{GDBN} ignores any core dump file while your program
2423 On some operating systems, a program cannot be executed outside @value{GDBN}
2424 while you have breakpoints set on it inside @value{GDBN}. You can use the
2425 @code{kill} command in this situation to permit running your program
2426 outside the debugger.
2428 The @code{kill} command is also useful if you wish to recompile and
2429 relink your program, since on many systems it is impossible to modify an
2430 executable file while it is running in a process. In this case, when you
2431 next type @code{run}, @value{GDBN} notices that the file has changed, and
2432 reads the symbol table again (while trying to preserve your current
2433 breakpoint settings).
2435 @node Inferiors and Programs
2436 @section Debugging Multiple Inferiors and Programs
2438 @value{GDBN} lets you run and debug multiple programs in a single
2439 session. In addition, @value{GDBN} on some systems may let you run
2440 several programs simultaneously (otherwise you have to exit from one
2441 before starting another). In the most general case, you can have
2442 multiple threads of execution in each of multiple processes, launched
2443 from multiple executables.
2446 @value{GDBN} represents the state of each program execution with an
2447 object called an @dfn{inferior}. An inferior typically corresponds to
2448 a process, but is more general and applies also to targets that do not
2449 have processes. Inferiors may be created before a process runs, and
2450 may be retained after a process exits. Inferiors have unique
2451 identifiers that are different from process ids. Usually each
2452 inferior will also have its own distinct address space, although some
2453 embedded targets may have several inferiors running in different parts
2454 of a single address space. Each inferior may in turn have multiple
2455 threads running in it.
2457 To find out what inferiors exist at any moment, use @w{@code{info
2461 @kindex info inferiors
2462 @item info inferiors
2463 Print a list of all inferiors currently being managed by @value{GDBN}.
2465 @value{GDBN} displays for each inferior (in this order):
2469 the inferior number assigned by @value{GDBN}
2472 the target system's inferior identifier
2475 the name of the executable the inferior is running.
2480 An asterisk @samp{*} preceding the @value{GDBN} inferior number
2481 indicates the current inferior.
2485 @c end table here to get a little more width for example
2488 (@value{GDBP}) info inferiors
2489 Num Description Executable
2490 2 process 2307 hello
2491 * 1 process 3401 goodbye
2494 To switch focus between inferiors, use the @code{inferior} command:
2497 @kindex inferior @var{infno}
2498 @item inferior @var{infno}
2499 Make inferior number @var{infno} the current inferior. The argument
2500 @var{infno} is the inferior number assigned by @value{GDBN}, as shown
2501 in the first field of the @samp{info inferiors} display.
2505 You can get multiple executables into a debugging session via the
2506 @code{add-inferior} and @w{@code{clone-inferior}} commands. On some
2507 systems @value{GDBN} can add inferiors to the debug session
2508 automatically by following calls to @code{fork} and @code{exec}. To
2509 remove inferiors from the debugging session use the
2510 @w{@code{remove-inferiors}} command.
2513 @kindex add-inferior
2514 @item add-inferior [ -copies @var{n} ] [ -exec @var{executable} ]
2515 Adds @var{n} inferiors to be run using @var{executable} as the
2516 executable. @var{n} defaults to 1. If no executable is specified,
2517 the inferiors begins empty, with no program. You can still assign or
2518 change the program assigned to the inferior at any time by using the
2519 @code{file} command with the executable name as its argument.
2521 @kindex clone-inferior
2522 @item clone-inferior [ -copies @var{n} ] [ @var{infno} ]
2523 Adds @var{n} inferiors ready to execute the same program as inferior
2524 @var{infno}. @var{n} defaults to 1. @var{infno} defaults to the
2525 number of the current inferior. This is a convenient command when you
2526 want to run another instance of the inferior you are debugging.
2529 (@value{GDBP}) info inferiors
2530 Num Description Executable
2531 * 1 process 29964 helloworld
2532 (@value{GDBP}) clone-inferior
2535 (@value{GDBP}) info inferiors
2536 Num Description Executable
2538 * 1 process 29964 helloworld
2541 You can now simply switch focus to inferior 2 and run it.
2543 @kindex remove-inferiors
2544 @item remove-inferiors @var{infno}@dots{}
2545 Removes the inferior or inferiors @var{infno}@dots{}. It is not
2546 possible to remove an inferior that is running with this command. For
2547 those, use the @code{kill} or @code{detach} command first.
2551 To quit debugging one of the running inferiors that is not the current
2552 inferior, you can either detach from it by using the @w{@code{detach
2553 inferior}} command (allowing it to run independently), or kill it
2554 using the @w{@code{kill inferiors}} command:
2557 @kindex detach inferiors @var{infno}@dots{}
2558 @item detach inferior @var{infno}@dots{}
2559 Detach from the inferior or inferiors identified by @value{GDBN}
2560 inferior number(s) @var{infno}@dots{}. Note that the inferior's entry
2561 still stays on the list of inferiors shown by @code{info inferiors},
2562 but its Description will show @samp{<null>}.
2564 @kindex kill inferiors @var{infno}@dots{}
2565 @item kill inferiors @var{infno}@dots{}
2566 Kill the inferior or inferiors identified by @value{GDBN} inferior
2567 number(s) @var{infno}@dots{}. Note that the inferior's entry still
2568 stays on the list of inferiors shown by @code{info inferiors}, but its
2569 Description will show @samp{<null>}.
2572 After the successful completion of a command such as @code{detach},
2573 @code{detach inferiors}, @code{kill} or @code{kill inferiors}, or after
2574 a normal process exit, the inferior is still valid and listed with
2575 @code{info inferiors}, ready to be restarted.
2578 To be notified when inferiors are started or exit under @value{GDBN}'s
2579 control use @w{@code{set print inferior-events}}:
2582 @kindex set print inferior-events
2583 @cindex print messages on inferior start and exit
2584 @item set print inferior-events
2585 @itemx set print inferior-events on
2586 @itemx set print inferior-events off
2587 The @code{set print inferior-events} command allows you to enable or
2588 disable printing of messages when @value{GDBN} notices that new
2589 inferiors have started or that inferiors have exited or have been
2590 detached. By default, these messages will not be printed.
2592 @kindex show print inferior-events
2593 @item show print inferior-events
2594 Show whether messages will be printed when @value{GDBN} detects that
2595 inferiors have started, exited or have been detached.
2598 Many commands will work the same with multiple programs as with a
2599 single program: e.g., @code{print myglobal} will simply display the
2600 value of @code{myglobal} in the current inferior.
2603 Occasionaly, when debugging @value{GDBN} itself, it may be useful to
2604 get more info about the relationship of inferiors, programs, address
2605 spaces in a debug session. You can do that with the @w{@code{maint
2606 info program-spaces}} command.
2609 @kindex maint info program-spaces
2610 @item maint info program-spaces
2611 Print a list of all program spaces currently being managed by
2614 @value{GDBN} displays for each program space (in this order):
2618 the program space number assigned by @value{GDBN}
2621 the name of the executable loaded into the program space, with e.g.,
2622 the @code{file} command.
2627 An asterisk @samp{*} preceding the @value{GDBN} program space number
2628 indicates the current program space.
2630 In addition, below each program space line, @value{GDBN} prints extra
2631 information that isn't suitable to display in tabular form. For
2632 example, the list of inferiors bound to the program space.
2635 (@value{GDBP}) maint info program-spaces
2638 Bound inferiors: ID 1 (process 21561)
2642 Here we can see that no inferior is running the program @code{hello},
2643 while @code{process 21561} is running the program @code{goodbye}. On
2644 some targets, it is possible that multiple inferiors are bound to the
2645 same program space. The most common example is that of debugging both
2646 the parent and child processes of a @code{vfork} call. For example,
2649 (@value{GDBP}) maint info program-spaces
2652 Bound inferiors: ID 2 (process 18050), ID 1 (process 18045)
2655 Here, both inferior 2 and inferior 1 are running in the same program
2656 space as a result of inferior 1 having executed a @code{vfork} call.
2660 @section Debugging Programs with Multiple Threads
2662 @cindex threads of execution
2663 @cindex multiple threads
2664 @cindex switching threads
2665 In some operating systems, such as HP-UX and Solaris, a single program
2666 may have more than one @dfn{thread} of execution. The precise semantics
2667 of threads differ from one operating system to another, but in general
2668 the threads of a single program are akin to multiple processes---except
2669 that they share one address space (that is, they can all examine and
2670 modify the same variables). On the other hand, each thread has its own
2671 registers and execution stack, and perhaps private memory.
2673 @value{GDBN} provides these facilities for debugging multi-thread
2677 @item automatic notification of new threads
2678 @item @samp{thread @var{threadno}}, a command to switch among threads
2679 @item @samp{info threads}, a command to inquire about existing threads
2680 @item @samp{thread apply [@var{threadno}] [@var{all}] @var{args}},
2681 a command to apply a command to a list of threads
2682 @item thread-specific breakpoints
2683 @item @samp{set print thread-events}, which controls printing of
2684 messages on thread start and exit.
2685 @item @samp{set libthread-db-search-path @var{path}}, which lets
2686 the user specify which @code{libthread_db} to use if the default choice
2687 isn't compatible with the program.
2691 @emph{Warning:} These facilities are not yet available on every
2692 @value{GDBN} configuration where the operating system supports threads.
2693 If your @value{GDBN} does not support threads, these commands have no
2694 effect. For example, a system without thread support shows no output
2695 from @samp{info threads}, and always rejects the @code{thread} command,
2699 (@value{GDBP}) info threads
2700 (@value{GDBP}) thread 1
2701 Thread ID 1 not known. Use the "info threads" command to
2702 see the IDs of currently known threads.
2704 @c FIXME to implementors: how hard would it be to say "sorry, this GDB
2705 @c doesn't support threads"?
2708 @cindex focus of debugging
2709 @cindex current thread
2710 The @value{GDBN} thread debugging facility allows you to observe all
2711 threads while your program runs---but whenever @value{GDBN} takes
2712 control, one thread in particular is always the focus of debugging.
2713 This thread is called the @dfn{current thread}. Debugging commands show
2714 program information from the perspective of the current thread.
2716 @cindex @code{New} @var{systag} message
2717 @cindex thread identifier (system)
2718 @c FIXME-implementors!! It would be more helpful if the [New...] message
2719 @c included GDB's numeric thread handle, so you could just go to that
2720 @c thread without first checking `info threads'.
2721 Whenever @value{GDBN} detects a new thread in your program, it displays
2722 the target system's identification for the thread with a message in the
2723 form @samp{[New @var{systag}]}. @var{systag} is a thread identifier
2724 whose form varies depending on the particular system. For example, on
2725 @sc{gnu}/Linux, you might see
2728 [New Thread 0x41e02940 (LWP 25582)]
2732 when @value{GDBN} notices a new thread. In contrast, on an SGI system,
2733 the @var{systag} is simply something like @samp{process 368}, with no
2736 @c FIXME!! (1) Does the [New...] message appear even for the very first
2737 @c thread of a program, or does it only appear for the
2738 @c second---i.e.@: when it becomes obvious we have a multithread
2740 @c (2) *Is* there necessarily a first thread always? Or do some
2741 @c multithread systems permit starting a program with multiple
2742 @c threads ab initio?
2744 @cindex thread number
2745 @cindex thread identifier (GDB)
2746 For debugging purposes, @value{GDBN} associates its own thread
2747 number---always a single integer---with each thread in your program.
2750 @kindex info threads
2751 @item info threads @r{[}@var{id}@dots{}@r{]}
2752 Display a summary of all threads currently in your program. Optional
2753 argument @var{id}@dots{} is one or more thread ids separated by spaces, and
2754 means to print information only about the specified thread or threads.
2755 @value{GDBN} displays for each thread (in this order):
2759 the thread number assigned by @value{GDBN}
2762 the target system's thread identifier (@var{systag})
2765 the thread's name, if one is known. A thread can either be named by
2766 the user (see @code{thread name}, below), or, in some cases, by the
2770 the current stack frame summary for that thread
2774 An asterisk @samp{*} to the left of the @value{GDBN} thread number
2775 indicates the current thread.
2779 @c end table here to get a little more width for example
2782 (@value{GDBP}) info threads
2784 3 process 35 thread 27 0x34e5 in sigpause ()
2785 2 process 35 thread 23 0x34e5 in sigpause ()
2786 * 1 process 35 thread 13 main (argc=1, argv=0x7ffffff8)
2790 On Solaris, you can display more information about user threads with a
2791 Solaris-specific command:
2794 @item maint info sol-threads
2795 @kindex maint info sol-threads
2796 @cindex thread info (Solaris)
2797 Display info on Solaris user threads.
2801 @kindex thread @var{threadno}
2802 @item thread @var{threadno}
2803 Make thread number @var{threadno} the current thread. The command
2804 argument @var{threadno} is the internal @value{GDBN} thread number, as
2805 shown in the first field of the @samp{info threads} display.
2806 @value{GDBN} responds by displaying the system identifier of the thread
2807 you selected, and its current stack frame summary:
2810 (@value{GDBP}) thread 2
2811 [Switching to thread 2 (Thread 0xb7fdab70 (LWP 12747))]
2812 #0 some_function (ignore=0x0) at example.c:8
2813 8 printf ("hello\n");
2817 As with the @samp{[New @dots{}]} message, the form of the text after
2818 @samp{Switching to} depends on your system's conventions for identifying
2821 @vindex $_thread@r{, convenience variable}
2822 The debugger convenience variable @samp{$_thread} contains the number
2823 of the current thread. You may find this useful in writing breakpoint
2824 conditional expressions, command scripts, and so forth. See
2825 @xref{Convenience Vars,, Convenience Variables}, for general
2826 information on convenience variables.
2828 @kindex thread apply
2829 @cindex apply command to several threads
2830 @item thread apply [@var{threadno} | all] @var{command}
2831 The @code{thread apply} command allows you to apply the named
2832 @var{command} to one or more threads. Specify the numbers of the
2833 threads that you want affected with the command argument
2834 @var{threadno}. It can be a single thread number, one of the numbers
2835 shown in the first field of the @samp{info threads} display; or it
2836 could be a range of thread numbers, as in @code{2-4}. To apply a
2837 command to all threads, type @kbd{thread apply all @var{command}}.
2840 @cindex name a thread
2841 @item thread name [@var{name}]
2842 This command assigns a name to the current thread. If no argument is
2843 given, any existing user-specified name is removed. The thread name
2844 appears in the @samp{info threads} display.
2846 On some systems, such as @sc{gnu}/Linux, @value{GDBN} is able to
2847 determine the name of the thread as given by the OS. On these
2848 systems, a name specified with @samp{thread name} will override the
2849 system-give name, and removing the user-specified name will cause
2850 @value{GDBN} to once again display the system-specified name.
2853 @cindex search for a thread
2854 @item thread find [@var{regexp}]
2855 Search for and display thread ids whose name or @var{systag}
2856 matches the supplied regular expression.
2858 As well as being the complement to the @samp{thread name} command,
2859 this command also allows you to identify a thread by its target
2860 @var{systag}. For instance, on @sc{gnu}/Linux, the target @var{systag}
2864 (@value{GDBN}) thread find 26688
2865 Thread 4 has target id 'Thread 0x41e02940 (LWP 26688)'
2866 (@value{GDBN}) info thread 4
2868 4 Thread 0x41e02940 (LWP 26688) 0x00000031ca6cd372 in select ()
2871 @kindex set print thread-events
2872 @cindex print messages on thread start and exit
2873 @item set print thread-events
2874 @itemx set print thread-events on
2875 @itemx set print thread-events off
2876 The @code{set print thread-events} command allows you to enable or
2877 disable printing of messages when @value{GDBN} notices that new threads have
2878 started or that threads have exited. By default, these messages will
2879 be printed if detection of these events is supported by the target.
2880 Note that these messages cannot be disabled on all targets.
2882 @kindex show print thread-events
2883 @item show print thread-events
2884 Show whether messages will be printed when @value{GDBN} detects that threads
2885 have started and exited.
2888 @xref{Thread Stops,,Stopping and Starting Multi-thread Programs}, for
2889 more information about how @value{GDBN} behaves when you stop and start
2890 programs with multiple threads.
2892 @xref{Set Watchpoints,,Setting Watchpoints}, for information about
2893 watchpoints in programs with multiple threads.
2895 @anchor{set libthread-db-search-path}
2897 @kindex set libthread-db-search-path
2898 @cindex search path for @code{libthread_db}
2899 @item set libthread-db-search-path @r{[}@var{path}@r{]}
2900 If this variable is set, @var{path} is a colon-separated list of
2901 directories @value{GDBN} will use to search for @code{libthread_db}.
2902 If you omit @var{path}, @samp{libthread-db-search-path} will be reset to
2903 its default value (@code{$sdir:$pdir} on @sc{gnu}/Linux and Solaris systems).
2904 Internally, the default value comes from the @code{LIBTHREAD_DB_SEARCH_PATH}
2907 On @sc{gnu}/Linux and Solaris systems, @value{GDBN} uses a ``helper''
2908 @code{libthread_db} library to obtain information about threads in the
2909 inferior process. @value{GDBN} will use @samp{libthread-db-search-path}
2910 to find @code{libthread_db}. @value{GDBN} also consults first if inferior
2911 specific thread debugging library loading is enabled
2912 by @samp{set auto-load libthread-db} (@pxref{libthread_db.so.1 file}).
2914 A special entry @samp{$sdir} for @samp{libthread-db-search-path}
2915 refers to the default system directories that are
2916 normally searched for loading shared libraries. The @samp{$sdir} entry
2917 is the only kind not needing to be enabled by @samp{set auto-load libthread-db}
2918 (@pxref{libthread_db.so.1 file}).
2920 A special entry @samp{$pdir} for @samp{libthread-db-search-path}
2921 refers to the directory from which @code{libpthread}
2922 was loaded in the inferior process.
2924 For any @code{libthread_db} library @value{GDBN} finds in above directories,
2925 @value{GDBN} attempts to initialize it with the current inferior process.
2926 If this initialization fails (which could happen because of a version
2927 mismatch between @code{libthread_db} and @code{libpthread}), @value{GDBN}
2928 will unload @code{libthread_db}, and continue with the next directory.
2929 If none of @code{libthread_db} libraries initialize successfully,
2930 @value{GDBN} will issue a warning and thread debugging will be disabled.
2932 Setting @code{libthread-db-search-path} is currently implemented
2933 only on some platforms.
2935 @kindex show libthread-db-search-path
2936 @item show libthread-db-search-path
2937 Display current libthread_db search path.
2939 @kindex set debug libthread-db
2940 @kindex show debug libthread-db
2941 @cindex debugging @code{libthread_db}
2942 @item set debug libthread-db
2943 @itemx show debug libthread-db
2944 Turns on or off display of @code{libthread_db}-related events.
2945 Use @code{1} to enable, @code{0} to disable.
2949 @section Debugging Forks
2951 @cindex fork, debugging programs which call
2952 @cindex multiple processes
2953 @cindex processes, multiple
2954 On most systems, @value{GDBN} has no special support for debugging
2955 programs which create additional processes using the @code{fork}
2956 function. When a program forks, @value{GDBN} will continue to debug the
2957 parent process and the child process will run unimpeded. If you have
2958 set a breakpoint in any code which the child then executes, the child
2959 will get a @code{SIGTRAP} signal which (unless it catches the signal)
2960 will cause it to terminate.
2962 However, if you want to debug the child process there is a workaround
2963 which isn't too painful. Put a call to @code{sleep} in the code which
2964 the child process executes after the fork. It may be useful to sleep
2965 only if a certain environment variable is set, or a certain file exists,
2966 so that the delay need not occur when you don't want to run @value{GDBN}
2967 on the child. While the child is sleeping, use the @code{ps} program to
2968 get its process ID. Then tell @value{GDBN} (a new invocation of
2969 @value{GDBN} if you are also debugging the parent process) to attach to
2970 the child process (@pxref{Attach}). From that point on you can debug
2971 the child process just like any other process which you attached to.
2973 On some systems, @value{GDBN} provides support for debugging programs that
2974 create additional processes using the @code{fork} or @code{vfork} functions.
2975 Currently, the only platforms with this feature are HP-UX (11.x and later
2976 only?) and @sc{gnu}/Linux (kernel version 2.5.60 and later).
2978 By default, when a program forks, @value{GDBN} will continue to debug
2979 the parent process and the child process will run unimpeded.
2981 If you want to follow the child process instead of the parent process,
2982 use the command @w{@code{set follow-fork-mode}}.
2985 @kindex set follow-fork-mode
2986 @item set follow-fork-mode @var{mode}
2987 Set the debugger response to a program call of @code{fork} or
2988 @code{vfork}. A call to @code{fork} or @code{vfork} creates a new
2989 process. The @var{mode} argument can be:
2993 The original process is debugged after a fork. The child process runs
2994 unimpeded. This is the default.
2997 The new process is debugged after a fork. The parent process runs
3002 @kindex show follow-fork-mode
3003 @item show follow-fork-mode
3004 Display the current debugger response to a @code{fork} or @code{vfork} call.
3007 @cindex debugging multiple processes
3008 On Linux, if you want to debug both the parent and child processes, use the
3009 command @w{@code{set detach-on-fork}}.
3012 @kindex set detach-on-fork
3013 @item set detach-on-fork @var{mode}
3014 Tells gdb whether to detach one of the processes after a fork, or
3015 retain debugger control over them both.
3019 The child process (or parent process, depending on the value of
3020 @code{follow-fork-mode}) will be detached and allowed to run
3021 independently. This is the default.
3024 Both processes will be held under the control of @value{GDBN}.
3025 One process (child or parent, depending on the value of
3026 @code{follow-fork-mode}) is debugged as usual, while the other
3031 @kindex show detach-on-fork
3032 @item show detach-on-fork
3033 Show whether detach-on-fork mode is on/off.
3036 If you choose to set @samp{detach-on-fork} mode off, then @value{GDBN}
3037 will retain control of all forked processes (including nested forks).
3038 You can list the forked processes under the control of @value{GDBN} by
3039 using the @w{@code{info inferiors}} command, and switch from one fork
3040 to another by using the @code{inferior} command (@pxref{Inferiors and
3041 Programs, ,Debugging Multiple Inferiors and Programs}).
3043 To quit debugging one of the forked processes, you can either detach
3044 from it by using the @w{@code{detach inferiors}} command (allowing it
3045 to run independently), or kill it using the @w{@code{kill inferiors}}
3046 command. @xref{Inferiors and Programs, ,Debugging Multiple Inferiors
3049 If you ask to debug a child process and a @code{vfork} is followed by an
3050 @code{exec}, @value{GDBN} executes the new target up to the first
3051 breakpoint in the new target. If you have a breakpoint set on
3052 @code{main} in your original program, the breakpoint will also be set on
3053 the child process's @code{main}.
3055 On some systems, when a child process is spawned by @code{vfork}, you
3056 cannot debug the child or parent until an @code{exec} call completes.
3058 If you issue a @code{run} command to @value{GDBN} after an @code{exec}
3059 call executes, the new target restarts. To restart the parent
3060 process, use the @code{file} command with the parent executable name
3061 as its argument. By default, after an @code{exec} call executes,
3062 @value{GDBN} discards the symbols of the previous executable image.
3063 You can change this behaviour with the @w{@code{set follow-exec-mode}}
3067 @kindex set follow-exec-mode
3068 @item set follow-exec-mode @var{mode}
3070 Set debugger response to a program call of @code{exec}. An
3071 @code{exec} call replaces the program image of a process.
3073 @code{follow-exec-mode} can be:
3077 @value{GDBN} creates a new inferior and rebinds the process to this
3078 new inferior. The program the process was running before the
3079 @code{exec} call can be restarted afterwards by restarting the
3085 (@value{GDBP}) info inferiors
3087 Id Description Executable
3090 process 12020 is executing new program: prog2
3091 Program exited normally.
3092 (@value{GDBP}) info inferiors
3093 Id Description Executable
3099 @value{GDBN} keeps the process bound to the same inferior. The new
3100 executable image replaces the previous executable loaded in the
3101 inferior. Restarting the inferior after the @code{exec} call, with
3102 e.g., the @code{run} command, restarts the executable the process was
3103 running after the @code{exec} call. This is the default mode.
3108 (@value{GDBP}) info inferiors
3109 Id Description Executable
3112 process 12020 is executing new program: prog2
3113 Program exited normally.
3114 (@value{GDBP}) info inferiors
3115 Id Description Executable
3122 You can use the @code{catch} command to make @value{GDBN} stop whenever
3123 a @code{fork}, @code{vfork}, or @code{exec} call is made. @xref{Set
3124 Catchpoints, ,Setting Catchpoints}.
3126 @node Checkpoint/Restart
3127 @section Setting a @emph{Bookmark} to Return to Later
3132 @cindex snapshot of a process
3133 @cindex rewind program state
3135 On certain operating systems@footnote{Currently, only
3136 @sc{gnu}/Linux.}, @value{GDBN} is able to save a @dfn{snapshot} of a
3137 program's state, called a @dfn{checkpoint}, and come back to it
3140 Returning to a checkpoint effectively undoes everything that has
3141 happened in the program since the @code{checkpoint} was saved. This
3142 includes changes in memory, registers, and even (within some limits)
3143 system state. Effectively, it is like going back in time to the
3144 moment when the checkpoint was saved.
3146 Thus, if you're stepping thru a program and you think you're
3147 getting close to the point where things go wrong, you can save
3148 a checkpoint. Then, if you accidentally go too far and miss
3149 the critical statement, instead of having to restart your program
3150 from the beginning, you can just go back to the checkpoint and
3151 start again from there.
3153 This can be especially useful if it takes a lot of time or
3154 steps to reach the point where you think the bug occurs.
3156 To use the @code{checkpoint}/@code{restart} method of debugging:
3161 Save a snapshot of the debugged program's current execution state.
3162 The @code{checkpoint} command takes no arguments, but each checkpoint
3163 is assigned a small integer id, similar to a breakpoint id.
3165 @kindex info checkpoints
3166 @item info checkpoints
3167 List the checkpoints that have been saved in the current debugging
3168 session. For each checkpoint, the following information will be
3175 @item Source line, or label
3178 @kindex restart @var{checkpoint-id}
3179 @item restart @var{checkpoint-id}
3180 Restore the program state that was saved as checkpoint number
3181 @var{checkpoint-id}. All program variables, registers, stack frames
3182 etc.@: will be returned to the values that they had when the checkpoint
3183 was saved. In essence, gdb will ``wind back the clock'' to the point
3184 in time when the checkpoint was saved.
3186 Note that breakpoints, @value{GDBN} variables, command history etc.
3187 are not affected by restoring a checkpoint. In general, a checkpoint
3188 only restores things that reside in the program being debugged, not in
3191 @kindex delete checkpoint @var{checkpoint-id}
3192 @item delete checkpoint @var{checkpoint-id}
3193 Delete the previously-saved checkpoint identified by @var{checkpoint-id}.
3197 Returning to a previously saved checkpoint will restore the user state
3198 of the program being debugged, plus a significant subset of the system
3199 (OS) state, including file pointers. It won't ``un-write'' data from
3200 a file, but it will rewind the file pointer to the previous location,
3201 so that the previously written data can be overwritten. For files
3202 opened in read mode, the pointer will also be restored so that the
3203 previously read data can be read again.
3205 Of course, characters that have been sent to a printer (or other
3206 external device) cannot be ``snatched back'', and characters received
3207 from eg.@: a serial device can be removed from internal program buffers,
3208 but they cannot be ``pushed back'' into the serial pipeline, ready to
3209 be received again. Similarly, the actual contents of files that have
3210 been changed cannot be restored (at this time).
3212 However, within those constraints, you actually can ``rewind'' your
3213 program to a previously saved point in time, and begin debugging it
3214 again --- and you can change the course of events so as to debug a
3215 different execution path this time.
3217 @cindex checkpoints and process id
3218 Finally, there is one bit of internal program state that will be
3219 different when you return to a checkpoint --- the program's process
3220 id. Each checkpoint will have a unique process id (or @var{pid}),
3221 and each will be different from the program's original @var{pid}.
3222 If your program has saved a local copy of its process id, this could
3223 potentially pose a problem.
3225 @subsection A Non-obvious Benefit of Using Checkpoints
3227 On some systems such as @sc{gnu}/Linux, address space randomization
3228 is performed on new processes for security reasons. This makes it
3229 difficult or impossible to set a breakpoint, or watchpoint, on an
3230 absolute address if you have to restart the program, since the
3231 absolute location of a symbol will change from one execution to the
3234 A checkpoint, however, is an @emph{identical} copy of a process.
3235 Therefore if you create a checkpoint at (eg.@:) the start of main,
3236 and simply return to that checkpoint instead of restarting the
3237 process, you can avoid the effects of address randomization and
3238 your symbols will all stay in the same place.
3241 @chapter Stopping and Continuing
3243 The principal purposes of using a debugger are so that you can stop your
3244 program before it terminates; or so that, if your program runs into
3245 trouble, you can investigate and find out why.
3247 Inside @value{GDBN}, your program may stop for any of several reasons,
3248 such as a signal, a breakpoint, or reaching a new line after a
3249 @value{GDBN} command such as @code{step}. You may then examine and
3250 change variables, set new breakpoints or remove old ones, and then
3251 continue execution. Usually, the messages shown by @value{GDBN} provide
3252 ample explanation of the status of your program---but you can also
3253 explicitly request this information at any time.
3256 @kindex info program
3258 Display information about the status of your program: whether it is
3259 running or not, what process it is, and why it stopped.
3263 * Breakpoints:: Breakpoints, watchpoints, and catchpoints
3264 * Continuing and Stepping:: Resuming execution
3265 * Skipping Over Functions and Files::
3266 Skipping over functions and files
3268 * Thread Stops:: Stopping and starting multi-thread programs
3272 @section Breakpoints, Watchpoints, and Catchpoints
3275 A @dfn{breakpoint} makes your program stop whenever a certain point in
3276 the program is reached. For each breakpoint, you can add conditions to
3277 control in finer detail whether your program stops. You can set
3278 breakpoints with the @code{break} command and its variants (@pxref{Set
3279 Breaks, ,Setting Breakpoints}), to specify the place where your program
3280 should stop by line number, function name or exact address in the
3283 On some systems, you can set breakpoints in shared libraries before
3284 the executable is run. There is a minor limitation on HP-UX systems:
3285 you must wait until the executable is run in order to set breakpoints
3286 in shared library routines that are not called directly by the program
3287 (for example, routines that are arguments in a @code{pthread_create}
3291 @cindex data breakpoints
3292 @cindex memory tracing
3293 @cindex breakpoint on memory address
3294 @cindex breakpoint on variable modification
3295 A @dfn{watchpoint} is a special breakpoint that stops your program
3296 when the value of an expression changes. The expression may be a value
3297 of a variable, or it could involve values of one or more variables
3298 combined by operators, such as @samp{a + b}. This is sometimes called
3299 @dfn{data breakpoints}. You must use a different command to set
3300 watchpoints (@pxref{Set Watchpoints, ,Setting Watchpoints}), but aside
3301 from that, you can manage a watchpoint like any other breakpoint: you
3302 enable, disable, and delete both breakpoints and watchpoints using the
3305 You can arrange to have values from your program displayed automatically
3306 whenever @value{GDBN} stops at a breakpoint. @xref{Auto Display,,
3310 @cindex breakpoint on events
3311 A @dfn{catchpoint} is another special breakpoint that stops your program
3312 when a certain kind of event occurs, such as the throwing of a C@t{++}
3313 exception or the loading of a library. As with watchpoints, you use a
3314 different command to set a catchpoint (@pxref{Set Catchpoints, ,Setting
3315 Catchpoints}), but aside from that, you can manage a catchpoint like any
3316 other breakpoint. (To stop when your program receives a signal, use the
3317 @code{handle} command; see @ref{Signals, ,Signals}.)
3319 @cindex breakpoint numbers
3320 @cindex numbers for breakpoints
3321 @value{GDBN} assigns a number to each breakpoint, watchpoint, or
3322 catchpoint when you create it; these numbers are successive integers
3323 starting with one. In many of the commands for controlling various
3324 features of breakpoints you use the breakpoint number to say which
3325 breakpoint you want to change. Each breakpoint may be @dfn{enabled} or
3326 @dfn{disabled}; if disabled, it has no effect on your program until you
3329 @cindex breakpoint ranges
3330 @cindex ranges of breakpoints
3331 Some @value{GDBN} commands accept a range of breakpoints on which to
3332 operate. A breakpoint range is either a single breakpoint number, like
3333 @samp{5}, or two such numbers, in increasing order, separated by a
3334 hyphen, like @samp{5-7}. When a breakpoint range is given to a command,
3335 all breakpoints in that range are operated on.
3338 * Set Breaks:: Setting breakpoints
3339 * Set Watchpoints:: Setting watchpoints
3340 * Set Catchpoints:: Setting catchpoints
3341 * Delete Breaks:: Deleting breakpoints
3342 * Disabling:: Disabling breakpoints
3343 * Conditions:: Break conditions
3344 * Break Commands:: Breakpoint command lists
3345 * Dynamic Printf:: Dynamic printf
3346 * Save Breakpoints:: How to save breakpoints in a file
3347 * Static Probe Points:: Listing static probe points
3348 * Error in Breakpoints:: ``Cannot insert breakpoints''
3349 * Breakpoint-related Warnings:: ``Breakpoint address adjusted...''
3353 @subsection Setting Breakpoints
3355 @c FIXME LMB what does GDB do if no code on line of breakpt?
3356 @c consider in particular declaration with/without initialization.
3358 @c FIXME 2 is there stuff on this already? break at fun start, already init?
3361 @kindex b @r{(@code{break})}
3362 @vindex $bpnum@r{, convenience variable}
3363 @cindex latest breakpoint
3364 Breakpoints are set with the @code{break} command (abbreviated
3365 @code{b}). The debugger convenience variable @samp{$bpnum} records the
3366 number of the breakpoint you've set most recently; see @ref{Convenience
3367 Vars,, Convenience Variables}, for a discussion of what you can do with
3368 convenience variables.
3371 @item break @var{location}
3372 Set a breakpoint at the given @var{location}, which can specify a
3373 function name, a line number, or an address of an instruction.
3374 (@xref{Specify Location}, for a list of all the possible ways to
3375 specify a @var{location}.) The breakpoint will stop your program just
3376 before it executes any of the code in the specified @var{location}.
3378 When using source languages that permit overloading of symbols, such as
3379 C@t{++}, a function name may refer to more than one possible place to break.
3380 @xref{Ambiguous Expressions,,Ambiguous Expressions}, for a discussion of
3383 It is also possible to insert a breakpoint that will stop the program
3384 only if a specific thread (@pxref{Thread-Specific Breakpoints})
3385 or a specific task (@pxref{Ada Tasks}) hits that breakpoint.
3388 When called without any arguments, @code{break} sets a breakpoint at
3389 the next instruction to be executed in the selected stack frame
3390 (@pxref{Stack, ,Examining the Stack}). In any selected frame but the
3391 innermost, this makes your program stop as soon as control
3392 returns to that frame. This is similar to the effect of a
3393 @code{finish} command in the frame inside the selected frame---except
3394 that @code{finish} does not leave an active breakpoint. If you use
3395 @code{break} without an argument in the innermost frame, @value{GDBN} stops
3396 the next time it reaches the current location; this may be useful
3399 @value{GDBN} normally ignores breakpoints when it resumes execution, until at
3400 least one instruction has been executed. If it did not do this, you
3401 would be unable to proceed past a breakpoint without first disabling the
3402 breakpoint. This rule applies whether or not the breakpoint already
3403 existed when your program stopped.
3405 @item break @dots{} if @var{cond}
3406 Set a breakpoint with condition @var{cond}; evaluate the expression
3407 @var{cond} each time the breakpoint is reached, and stop only if the
3408 value is nonzero---that is, if @var{cond} evaluates as true.
3409 @samp{@dots{}} stands for one of the possible arguments described
3410 above (or no argument) specifying where to break. @xref{Conditions,
3411 ,Break Conditions}, for more information on breakpoint conditions.
3414 @item tbreak @var{args}
3415 Set a breakpoint enabled only for one stop. @var{args} are the
3416 same as for the @code{break} command, and the breakpoint is set in the same
3417 way, but the breakpoint is automatically deleted after the first time your
3418 program stops there. @xref{Disabling, ,Disabling Breakpoints}.
3421 @cindex hardware breakpoints
3422 @item hbreak @var{args}
3423 Set a hardware-assisted breakpoint. @var{args} are the same as for the
3424 @code{break} command and the breakpoint is set in the same way, but the
3425 breakpoint requires hardware support and some target hardware may not
3426 have this support. The main purpose of this is EPROM/ROM code
3427 debugging, so you can set a breakpoint at an instruction without
3428 changing the instruction. This can be used with the new trap-generation
3429 provided by SPARClite DSU and most x86-based targets. These targets
3430 will generate traps when a program accesses some data or instruction
3431 address that is assigned to the debug registers. However the hardware
3432 breakpoint registers can take a limited number of breakpoints. For
3433 example, on the DSU, only two data breakpoints can be set at a time, and
3434 @value{GDBN} will reject this command if more than two are used. Delete
3435 or disable unused hardware breakpoints before setting new ones
3436 (@pxref{Disabling, ,Disabling Breakpoints}).
3437 @xref{Conditions, ,Break Conditions}.
3438 For remote targets, you can restrict the number of hardware
3439 breakpoints @value{GDBN} will use, see @ref{set remote
3440 hardware-breakpoint-limit}.
3443 @item thbreak @var{args}
3444 Set a hardware-assisted breakpoint enabled only for one stop. @var{args}
3445 are the same as for the @code{hbreak} command and the breakpoint is set in
3446 the same way. However, like the @code{tbreak} command,
3447 the breakpoint is automatically deleted after the
3448 first time your program stops there. Also, like the @code{hbreak}
3449 command, the breakpoint requires hardware support and some target hardware
3450 may not have this support. @xref{Disabling, ,Disabling Breakpoints}.
3451 See also @ref{Conditions, ,Break Conditions}.
3454 @cindex regular expression
3455 @cindex breakpoints at functions matching a regexp
3456 @cindex set breakpoints in many functions
3457 @item rbreak @var{regex}
3458 Set breakpoints on all functions matching the regular expression
3459 @var{regex}. This command sets an unconditional breakpoint on all
3460 matches, printing a list of all breakpoints it set. Once these
3461 breakpoints are set, they are treated just like the breakpoints set with
3462 the @code{break} command. You can delete them, disable them, or make
3463 them conditional the same way as any other breakpoint.
3465 The syntax of the regular expression is the standard one used with tools
3466 like @file{grep}. Note that this is different from the syntax used by
3467 shells, so for instance @code{foo*} matches all functions that include
3468 an @code{fo} followed by zero or more @code{o}s. There is an implicit
3469 @code{.*} leading and trailing the regular expression you supply, so to
3470 match only functions that begin with @code{foo}, use @code{^foo}.
3472 @cindex non-member C@t{++} functions, set breakpoint in
3473 When debugging C@t{++} programs, @code{rbreak} is useful for setting
3474 breakpoints on overloaded functions that are not members of any special
3477 @cindex set breakpoints on all functions
3478 The @code{rbreak} command can be used to set breakpoints in
3479 @strong{all} the functions in a program, like this:
3482 (@value{GDBP}) rbreak .
3485 @item rbreak @var{file}:@var{regex}
3486 If @code{rbreak} is called with a filename qualification, it limits
3487 the search for functions matching the given regular expression to the
3488 specified @var{file}. This can be used, for example, to set breakpoints on
3489 every function in a given file:
3492 (@value{GDBP}) rbreak file.c:.
3495 The colon separating the filename qualifier from the regex may
3496 optionally be surrounded by spaces.
3498 @kindex info breakpoints
3499 @cindex @code{$_} and @code{info breakpoints}
3500 @item info breakpoints @r{[}@var{n}@dots{}@r{]}
3501 @itemx info break @r{[}@var{n}@dots{}@r{]}
3502 Print a table of all breakpoints, watchpoints, and catchpoints set and
3503 not deleted. Optional argument @var{n} means print information only
3504 about the specified breakpoint(s) (or watchpoint(s) or catchpoint(s)).
3505 For each breakpoint, following columns are printed:
3508 @item Breakpoint Numbers
3510 Breakpoint, watchpoint, or catchpoint.
3512 Whether the breakpoint is marked to be disabled or deleted when hit.
3513 @item Enabled or Disabled
3514 Enabled breakpoints are marked with @samp{y}. @samp{n} marks breakpoints
3515 that are not enabled.
3517 Where the breakpoint is in your program, as a memory address. For a
3518 pending breakpoint whose address is not yet known, this field will
3519 contain @samp{<PENDING>}. Such breakpoint won't fire until a shared
3520 library that has the symbol or line referred by breakpoint is loaded.
3521 See below for details. A breakpoint with several locations will
3522 have @samp{<MULTIPLE>} in this field---see below for details.
3524 Where the breakpoint is in the source for your program, as a file and
3525 line number. For a pending breakpoint, the original string passed to
3526 the breakpoint command will be listed as it cannot be resolved until
3527 the appropriate shared library is loaded in the future.
3531 If a breakpoint is conditional, there are two evaluation modes: ``host'' and
3532 ``target''. If mode is ``host'', breakpoint condition evaluation is done by
3533 @value{GDBN} on the host's side. If it is ``target'', then the condition
3534 is evaluated by the target. The @code{info break} command shows
3535 the condition on the line following the affected breakpoint, together with
3536 its condition evaluation mode in between parentheses.
3538 Breakpoint commands, if any, are listed after that. A pending breakpoint is
3539 allowed to have a condition specified for it. The condition is not parsed for
3540 validity until a shared library is loaded that allows the pending
3541 breakpoint to resolve to a valid location.
3544 @code{info break} with a breakpoint
3545 number @var{n} as argument lists only that breakpoint. The
3546 convenience variable @code{$_} and the default examining-address for
3547 the @code{x} command are set to the address of the last breakpoint
3548 listed (@pxref{Memory, ,Examining Memory}).
3551 @code{info break} displays a count of the number of times the breakpoint
3552 has been hit. This is especially useful in conjunction with the
3553 @code{ignore} command. You can ignore a large number of breakpoint
3554 hits, look at the breakpoint info to see how many times the breakpoint
3555 was hit, and then run again, ignoring one less than that number. This
3556 will get you quickly to the last hit of that breakpoint.
3559 For a breakpoints with an enable count (xref) greater than 1,
3560 @code{info break} also displays that count.
3564 @value{GDBN} allows you to set any number of breakpoints at the same place in
3565 your program. There is nothing silly or meaningless about this. When
3566 the breakpoints are conditional, this is even useful
3567 (@pxref{Conditions, ,Break Conditions}).
3569 @cindex multiple locations, breakpoints
3570 @cindex breakpoints, multiple locations
3571 It is possible that a breakpoint corresponds to several locations
3572 in your program. Examples of this situation are:
3576 Multiple functions in the program may have the same name.
3579 For a C@t{++} constructor, the @value{NGCC} compiler generates several
3580 instances of the function body, used in different cases.
3583 For a C@t{++} template function, a given line in the function can
3584 correspond to any number of instantiations.
3587 For an inlined function, a given source line can correspond to
3588 several places where that function is inlined.
3591 In all those cases, @value{GDBN} will insert a breakpoint at all
3592 the relevant locations.
3594 A breakpoint with multiple locations is displayed in the breakpoint
3595 table using several rows---one header row, followed by one row for
3596 each breakpoint location. The header row has @samp{<MULTIPLE>} in the
3597 address column. The rows for individual locations contain the actual
3598 addresses for locations, and show the functions to which those
3599 locations belong. The number column for a location is of the form
3600 @var{breakpoint-number}.@var{location-number}.
3605 Num Type Disp Enb Address What
3606 1 breakpoint keep y <MULTIPLE>
3608 breakpoint already hit 1 time
3609 1.1 y 0x080486a2 in void foo<int>() at t.cc:8
3610 1.2 y 0x080486ca in void foo<double>() at t.cc:8
3613 Each location can be individually enabled or disabled by passing
3614 @var{breakpoint-number}.@var{location-number} as argument to the
3615 @code{enable} and @code{disable} commands. Note that you cannot
3616 delete the individual locations from the list, you can only delete the
3617 entire list of locations that belong to their parent breakpoint (with
3618 the @kbd{delete @var{num}} command, where @var{num} is the number of
3619 the parent breakpoint, 1 in the above example). Disabling or enabling
3620 the parent breakpoint (@pxref{Disabling}) affects all of the locations
3621 that belong to that breakpoint.
3623 @cindex pending breakpoints
3624 It's quite common to have a breakpoint inside a shared library.
3625 Shared libraries can be loaded and unloaded explicitly,
3626 and possibly repeatedly, as the program is executed. To support
3627 this use case, @value{GDBN} updates breakpoint locations whenever
3628 any shared library is loaded or unloaded. Typically, you would
3629 set a breakpoint in a shared library at the beginning of your
3630 debugging session, when the library is not loaded, and when the
3631 symbols from the library are not available. When you try to set
3632 breakpoint, @value{GDBN} will ask you if you want to set
3633 a so called @dfn{pending breakpoint}---breakpoint whose address
3634 is not yet resolved.
3636 After the program is run, whenever a new shared library is loaded,
3637 @value{GDBN} reevaluates all the breakpoints. When a newly loaded
3638 shared library contains the symbol or line referred to by some
3639 pending breakpoint, that breakpoint is resolved and becomes an
3640 ordinary breakpoint. When a library is unloaded, all breakpoints
3641 that refer to its symbols or source lines become pending again.
3643 This logic works for breakpoints with multiple locations, too. For
3644 example, if you have a breakpoint in a C@t{++} template function, and
3645 a newly loaded shared library has an instantiation of that template,
3646 a new location is added to the list of locations for the breakpoint.
3648 Except for having unresolved address, pending breakpoints do not
3649 differ from regular breakpoints. You can set conditions or commands,
3650 enable and disable them and perform other breakpoint operations.
3652 @value{GDBN} provides some additional commands for controlling what
3653 happens when the @samp{break} command cannot resolve breakpoint
3654 address specification to an address:
3656 @kindex set breakpoint pending
3657 @kindex show breakpoint pending
3659 @item set breakpoint pending auto
3660 This is the default behavior. When @value{GDBN} cannot find the breakpoint
3661 location, it queries you whether a pending breakpoint should be created.
3663 @item set breakpoint pending on
3664 This indicates that an unrecognized breakpoint location should automatically
3665 result in a pending breakpoint being created.
3667 @item set breakpoint pending off
3668 This indicates that pending breakpoints are not to be created. Any
3669 unrecognized breakpoint location results in an error. This setting does
3670 not affect any pending breakpoints previously created.
3672 @item show breakpoint pending
3673 Show the current behavior setting for creating pending breakpoints.
3676 The settings above only affect the @code{break} command and its
3677 variants. Once breakpoint is set, it will be automatically updated
3678 as shared libraries are loaded and unloaded.
3680 @cindex automatic hardware breakpoints
3681 For some targets, @value{GDBN} can automatically decide if hardware or
3682 software breakpoints should be used, depending on whether the
3683 breakpoint address is read-only or read-write. This applies to
3684 breakpoints set with the @code{break} command as well as to internal
3685 breakpoints set by commands like @code{next} and @code{finish}. For
3686 breakpoints set with @code{hbreak}, @value{GDBN} will always use hardware
3689 You can control this automatic behaviour with the following commands::
3691 @kindex set breakpoint auto-hw
3692 @kindex show breakpoint auto-hw
3694 @item set breakpoint auto-hw on
3695 This is the default behavior. When @value{GDBN} sets a breakpoint, it
3696 will try to use the target memory map to decide if software or hardware
3697 breakpoint must be used.
3699 @item set breakpoint auto-hw off
3700 This indicates @value{GDBN} should not automatically select breakpoint
3701 type. If the target provides a memory map, @value{GDBN} will warn when
3702 trying to set software breakpoint at a read-only address.
3705 @value{GDBN} normally implements breakpoints by replacing the program code
3706 at the breakpoint address with a special instruction, which, when
3707 executed, given control to the debugger. By default, the program
3708 code is so modified only when the program is resumed. As soon as
3709 the program stops, @value{GDBN} restores the original instructions. This
3710 behaviour guards against leaving breakpoints inserted in the
3711 target should gdb abrubptly disconnect. However, with slow remote
3712 targets, inserting and removing breakpoint can reduce the performance.
3713 This behavior can be controlled with the following commands::
3715 @kindex set breakpoint always-inserted
3716 @kindex show breakpoint always-inserted
3718 @item set breakpoint always-inserted off
3719 All breakpoints, including newly added by the user, are inserted in
3720 the target only when the target is resumed. All breakpoints are
3721 removed from the target when it stops.
3723 @item set breakpoint always-inserted on
3724 Causes all breakpoints to be inserted in the target at all times. If
3725 the user adds a new breakpoint, or changes an existing breakpoint, the
3726 breakpoints in the target are updated immediately. A breakpoint is
3727 removed from the target only when breakpoint itself is removed.
3729 @cindex non-stop mode, and @code{breakpoint always-inserted}
3730 @item set breakpoint always-inserted auto
3731 This is the default mode. If @value{GDBN} is controlling the inferior
3732 in non-stop mode (@pxref{Non-Stop Mode}), gdb behaves as if
3733 @code{breakpoint always-inserted} mode is on. If @value{GDBN} is
3734 controlling the inferior in all-stop mode, @value{GDBN} behaves as if
3735 @code{breakpoint always-inserted} mode is off.
3738 @value{GDBN} handles conditional breakpoints by evaluating these conditions
3739 when a breakpoint breaks. If the condition is true, then the process being
3740 debugged stops, otherwise the process is resumed.
3742 If the target supports evaluating conditions on its end, @value{GDBN} may
3743 download the breakpoint, together with its conditions, to it.
3745 This feature can be controlled via the following commands:
3747 @kindex set breakpoint condition-evaluation
3748 @kindex show breakpoint condition-evaluation
3750 @item set breakpoint condition-evaluation host
3751 This option commands @value{GDBN} to evaluate the breakpoint
3752 conditions on the host's side. Unconditional breakpoints are sent to
3753 the target which in turn receives the triggers and reports them back to GDB
3754 for condition evaluation. This is the standard evaluation mode.
3756 @item set breakpoint condition-evaluation target
3757 This option commands @value{GDBN} to download breakpoint conditions
3758 to the target at the moment of their insertion. The target
3759 is responsible for evaluating the conditional expression and reporting
3760 breakpoint stop events back to @value{GDBN} whenever the condition
3761 is true. Due to limitations of target-side evaluation, some conditions
3762 cannot be evaluated there, e.g., conditions that depend on local data
3763 that is only known to the host. Examples include
3764 conditional expressions involving convenience variables, complex types
3765 that cannot be handled by the agent expression parser and expressions
3766 that are too long to be sent over to the target, specially when the
3767 target is a remote system. In these cases, the conditions will be
3768 evaluated by @value{GDBN}.
3770 @item set breakpoint condition-evaluation auto
3771 This is the default mode. If the target supports evaluating breakpoint
3772 conditions on its end, @value{GDBN} will download breakpoint conditions to
3773 the target (limitations mentioned previously apply). If the target does
3774 not support breakpoint condition evaluation, then @value{GDBN} will fallback
3775 to evaluating all these conditions on the host's side.
3779 @cindex negative breakpoint numbers
3780 @cindex internal @value{GDBN} breakpoints
3781 @value{GDBN} itself sometimes sets breakpoints in your program for
3782 special purposes, such as proper handling of @code{longjmp} (in C
3783 programs). These internal breakpoints are assigned negative numbers,
3784 starting with @code{-1}; @samp{info breakpoints} does not display them.
3785 You can see these breakpoints with the @value{GDBN} maintenance command
3786 @samp{maint info breakpoints} (@pxref{maint info breakpoints}).
3789 @node Set Watchpoints
3790 @subsection Setting Watchpoints
3792 @cindex setting watchpoints
3793 You can use a watchpoint to stop execution whenever the value of an
3794 expression changes, without having to predict a particular place where
3795 this may happen. (This is sometimes called a @dfn{data breakpoint}.)
3796 The expression may be as simple as the value of a single variable, or
3797 as complex as many variables combined by operators. Examples include:
3801 A reference to the value of a single variable.
3804 An address cast to an appropriate data type. For example,
3805 @samp{*(int *)0x12345678} will watch a 4-byte region at the specified
3806 address (assuming an @code{int} occupies 4 bytes).
3809 An arbitrarily complex expression, such as @samp{a*b + c/d}. The
3810 expression can use any operators valid in the program's native
3811 language (@pxref{Languages}).
3814 You can set a watchpoint on an expression even if the expression can
3815 not be evaluated yet. For instance, you can set a watchpoint on
3816 @samp{*global_ptr} before @samp{global_ptr} is initialized.
3817 @value{GDBN} will stop when your program sets @samp{global_ptr} and
3818 the expression produces a valid value. If the expression becomes
3819 valid in some other way than changing a variable (e.g.@: if the memory
3820 pointed to by @samp{*global_ptr} becomes readable as the result of a
3821 @code{malloc} call), @value{GDBN} may not stop until the next time
3822 the expression changes.
3824 @cindex software watchpoints
3825 @cindex hardware watchpoints
3826 Depending on your system, watchpoints may be implemented in software or
3827 hardware. @value{GDBN} does software watchpointing by single-stepping your
3828 program and testing the variable's value each time, which is hundreds of
3829 times slower than normal execution. (But this may still be worth it, to
3830 catch errors where you have no clue what part of your program is the
3833 On some systems, such as HP-UX, PowerPC, @sc{gnu}/Linux and most other
3834 x86-based targets, @value{GDBN} includes support for hardware
3835 watchpoints, which do not slow down the running of your program.
3839 @item watch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{threadnum}@r{]} @r{[}mask @var{maskvalue}@r{]}
3840 Set a watchpoint for an expression. @value{GDBN} will break when the
3841 expression @var{expr} is written into by the program and its value
3842 changes. The simplest (and the most popular) use of this command is
3843 to watch the value of a single variable:
3846 (@value{GDBP}) watch foo
3849 If the command includes a @code{@r{[}thread @var{threadnum}@r{]}}
3850 argument, @value{GDBN} breaks only when the thread identified by
3851 @var{threadnum} changes the value of @var{expr}. If any other threads
3852 change the value of @var{expr}, @value{GDBN} will not break. Note
3853 that watchpoints restricted to a single thread in this way only work
3854 with Hardware Watchpoints.
3856 Ordinarily a watchpoint respects the scope of variables in @var{expr}
3857 (see below). The @code{-location} argument tells @value{GDBN} to
3858 instead watch the memory referred to by @var{expr}. In this case,
3859 @value{GDBN} will evaluate @var{expr}, take the address of the result,
3860 and watch the memory at that address. The type of the result is used
3861 to determine the size of the watched memory. If the expression's
3862 result does not have an address, then @value{GDBN} will print an
3865 The @code{@r{[}mask @var{maskvalue}@r{]}} argument allows creation
3866 of masked watchpoints, if the current architecture supports this
3867 feature (e.g., PowerPC Embedded architecture, see @ref{PowerPC
3868 Embedded}.) A @dfn{masked watchpoint} specifies a mask in addition
3869 to an address to watch. The mask specifies that some bits of an address
3870 (the bits which are reset in the mask) should be ignored when matching
3871 the address accessed by the inferior against the watchpoint address.
3872 Thus, a masked watchpoint watches many addresses simultaneously---those
3873 addresses whose unmasked bits are identical to the unmasked bits in the
3874 watchpoint address. The @code{mask} argument implies @code{-location}.
3878 (@value{GDBP}) watch foo mask 0xffff00ff
3879 (@value{GDBP}) watch *0xdeadbeef mask 0xffffff00
3883 @item rwatch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{threadnum}@r{]} @r{[}mask @var{maskvalue}@r{]}
3884 Set a watchpoint that will break when the value of @var{expr} is read
3888 @item awatch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{threadnum}@r{]} @r{[}mask @var{maskvalue}@r{]}
3889 Set a watchpoint that will break when @var{expr} is either read from
3890 or written into by the program.
3892 @kindex info watchpoints @r{[}@var{n}@dots{}@r{]}
3893 @item info watchpoints @r{[}@var{n}@dots{}@r{]}
3894 This command prints a list of watchpoints, using the same format as
3895 @code{info break} (@pxref{Set Breaks}).
3898 If you watch for a change in a numerically entered address you need to
3899 dereference it, as the address itself is just a constant number which will
3900 never change. @value{GDBN} refuses to create a watchpoint that watches
3901 a never-changing value:
3904 (@value{GDBP}) watch 0x600850
3905 Cannot watch constant value 0x600850.
3906 (@value{GDBP}) watch *(int *) 0x600850
3907 Watchpoint 1: *(int *) 6293584
3910 @value{GDBN} sets a @dfn{hardware watchpoint} if possible. Hardware
3911 watchpoints execute very quickly, and the debugger reports a change in
3912 value at the exact instruction where the change occurs. If @value{GDBN}
3913 cannot set a hardware watchpoint, it sets a software watchpoint, which
3914 executes more slowly and reports the change in value at the next
3915 @emph{statement}, not the instruction, after the change occurs.
3917 @cindex use only software watchpoints
3918 You can force @value{GDBN} to use only software watchpoints with the
3919 @kbd{set can-use-hw-watchpoints 0} command. With this variable set to
3920 zero, @value{GDBN} will never try to use hardware watchpoints, even if
3921 the underlying system supports them. (Note that hardware-assisted
3922 watchpoints that were set @emph{before} setting
3923 @code{can-use-hw-watchpoints} to zero will still use the hardware
3924 mechanism of watching expression values.)
3927 @item set can-use-hw-watchpoints
3928 @kindex set can-use-hw-watchpoints
3929 Set whether or not to use hardware watchpoints.
3931 @item show can-use-hw-watchpoints
3932 @kindex show can-use-hw-watchpoints
3933 Show the current mode of using hardware watchpoints.
3936 For remote targets, you can restrict the number of hardware
3937 watchpoints @value{GDBN} will use, see @ref{set remote
3938 hardware-breakpoint-limit}.
3940 When you issue the @code{watch} command, @value{GDBN} reports
3943 Hardware watchpoint @var{num}: @var{expr}
3947 if it was able to set a hardware watchpoint.
3949 Currently, the @code{awatch} and @code{rwatch} commands can only set
3950 hardware watchpoints, because accesses to data that don't change the
3951 value of the watched expression cannot be detected without examining
3952 every instruction as it is being executed, and @value{GDBN} does not do
3953 that currently. If @value{GDBN} finds that it is unable to set a
3954 hardware breakpoint with the @code{awatch} or @code{rwatch} command, it
3955 will print a message like this:
3958 Expression cannot be implemented with read/access watchpoint.
3961 Sometimes, @value{GDBN} cannot set a hardware watchpoint because the
3962 data type of the watched expression is wider than what a hardware
3963 watchpoint on the target machine can handle. For example, some systems
3964 can only watch regions that are up to 4 bytes wide; on such systems you
3965 cannot set hardware watchpoints for an expression that yields a
3966 double-precision floating-point number (which is typically 8 bytes
3967 wide). As a work-around, it might be possible to break the large region
3968 into a series of smaller ones and watch them with separate watchpoints.
3970 If you set too many hardware watchpoints, @value{GDBN} might be unable
3971 to insert all of them when you resume the execution of your program.
3972 Since the precise number of active watchpoints is unknown until such
3973 time as the program is about to be resumed, @value{GDBN} might not be
3974 able to warn you about this when you set the watchpoints, and the
3975 warning will be printed only when the program is resumed:
3978 Hardware watchpoint @var{num}: Could not insert watchpoint
3982 If this happens, delete or disable some of the watchpoints.
3984 Watching complex expressions that reference many variables can also
3985 exhaust the resources available for hardware-assisted watchpoints.
3986 That's because @value{GDBN} needs to watch every variable in the
3987 expression with separately allocated resources.
3989 If you call a function interactively using @code{print} or @code{call},
3990 any watchpoints you have set will be inactive until @value{GDBN} reaches another
3991 kind of breakpoint or the call completes.
3993 @value{GDBN} automatically deletes watchpoints that watch local
3994 (automatic) variables, or expressions that involve such variables, when
3995 they go out of scope, that is, when the execution leaves the block in
3996 which these variables were defined. In particular, when the program
3997 being debugged terminates, @emph{all} local variables go out of scope,
3998 and so only watchpoints that watch global variables remain set. If you
3999 rerun the program, you will need to set all such watchpoints again. One
4000 way of doing that would be to set a code breakpoint at the entry to the
4001 @code{main} function and when it breaks, set all the watchpoints.
4003 @cindex watchpoints and threads
4004 @cindex threads and watchpoints
4005 In multi-threaded programs, watchpoints will detect changes to the
4006 watched expression from every thread.
4009 @emph{Warning:} In multi-threaded programs, software watchpoints
4010 have only limited usefulness. If @value{GDBN} creates a software
4011 watchpoint, it can only watch the value of an expression @emph{in a
4012 single thread}. If you are confident that the expression can only
4013 change due to the current thread's activity (and if you are also
4014 confident that no other thread can become current), then you can use
4015 software watchpoints as usual. However, @value{GDBN} may not notice
4016 when a non-current thread's activity changes the expression. (Hardware
4017 watchpoints, in contrast, watch an expression in all threads.)
4020 @xref{set remote hardware-watchpoint-limit}.
4022 @node Set Catchpoints
4023 @subsection Setting Catchpoints
4024 @cindex catchpoints, setting
4025 @cindex exception handlers
4026 @cindex event handling
4028 You can use @dfn{catchpoints} to cause the debugger to stop for certain
4029 kinds of program events, such as C@t{++} exceptions or the loading of a
4030 shared library. Use the @code{catch} command to set a catchpoint.
4034 @item catch @var{event}
4035 Stop when @var{event} occurs. @var{event} can be any of the following:
4038 @cindex stop on C@t{++} exceptions
4039 The throwing of a C@t{++} exception.
4042 The catching of a C@t{++} exception.
4045 @cindex Ada exception catching
4046 @cindex catch Ada exceptions
4047 An Ada exception being raised. If an exception name is specified
4048 at the end of the command (eg @code{catch exception Program_Error}),
4049 the debugger will stop only when this specific exception is raised.
4050 Otherwise, the debugger stops execution when any Ada exception is raised.
4052 When inserting an exception catchpoint on a user-defined exception whose
4053 name is identical to one of the exceptions defined by the language, the
4054 fully qualified name must be used as the exception name. Otherwise,
4055 @value{GDBN} will assume that it should stop on the pre-defined exception
4056 rather than the user-defined one. For instance, assuming an exception
4057 called @code{Constraint_Error} is defined in package @code{Pck}, then
4058 the command to use to catch such exceptions is @kbd{catch exception
4059 Pck.Constraint_Error}.
4061 @item exception unhandled
4062 An exception that was raised but is not handled by the program.
4065 A failed Ada assertion.
4068 @cindex break on fork/exec
4069 A call to @code{exec}. This is currently only available for HP-UX
4073 @itemx syscall @r{[}@var{name} @r{|} @var{number}@r{]} @dots{}
4074 @cindex break on a system call.
4075 A call to or return from a system call, a.k.a.@: @dfn{syscall}. A
4076 syscall is a mechanism for application programs to request a service
4077 from the operating system (OS) or one of the OS system services.
4078 @value{GDBN} can catch some or all of the syscalls issued by the
4079 debuggee, and show the related information for each syscall. If no
4080 argument is specified, calls to and returns from all system calls
4083 @var{name} can be any system call name that is valid for the
4084 underlying OS. Just what syscalls are valid depends on the OS. On
4085 GNU and Unix systems, you can find the full list of valid syscall
4086 names on @file{/usr/include/asm/unistd.h}.
4088 @c For MS-Windows, the syscall names and the corresponding numbers
4089 @c can be found, e.g., on this URL:
4090 @c http://www.metasploit.com/users/opcode/syscalls.html
4091 @c but we don't support Windows syscalls yet.
4093 Normally, @value{GDBN} knows in advance which syscalls are valid for
4094 each OS, so you can use the @value{GDBN} command-line completion
4095 facilities (@pxref{Completion,, command completion}) to list the
4098 You may also specify the system call numerically. A syscall's
4099 number is the value passed to the OS's syscall dispatcher to
4100 identify the requested service. When you specify the syscall by its
4101 name, @value{GDBN} uses its database of syscalls to convert the name
4102 into the corresponding numeric code, but using the number directly
4103 may be useful if @value{GDBN}'s database does not have the complete
4104 list of syscalls on your system (e.g., because @value{GDBN} lags
4105 behind the OS upgrades).
4107 The example below illustrates how this command works if you don't provide
4111 (@value{GDBP}) catch syscall
4112 Catchpoint 1 (syscall)
4114 Starting program: /tmp/catch-syscall
4116 Catchpoint 1 (call to syscall 'close'), \
4117 0xffffe424 in __kernel_vsyscall ()
4121 Catchpoint 1 (returned from syscall 'close'), \
4122 0xffffe424 in __kernel_vsyscall ()
4126 Here is an example of catching a system call by name:
4129 (@value{GDBP}) catch syscall chroot
4130 Catchpoint 1 (syscall 'chroot' [61])
4132 Starting program: /tmp/catch-syscall
4134 Catchpoint 1 (call to syscall 'chroot'), \
4135 0xffffe424 in __kernel_vsyscall ()
4139 Catchpoint 1 (returned from syscall 'chroot'), \
4140 0xffffe424 in __kernel_vsyscall ()
4144 An example of specifying a system call numerically. In the case
4145 below, the syscall number has a corresponding entry in the XML
4146 file, so @value{GDBN} finds its name and prints it:
4149 (@value{GDBP}) catch syscall 252
4150 Catchpoint 1 (syscall(s) 'exit_group')
4152 Starting program: /tmp/catch-syscall
4154 Catchpoint 1 (call to syscall 'exit_group'), \
4155 0xffffe424 in __kernel_vsyscall ()
4159 Program exited normally.
4163 However, there can be situations when there is no corresponding name
4164 in XML file for that syscall number. In this case, @value{GDBN} prints
4165 a warning message saying that it was not able to find the syscall name,
4166 but the catchpoint will be set anyway. See the example below:
4169 (@value{GDBP}) catch syscall 764
4170 warning: The number '764' does not represent a known syscall.
4171 Catchpoint 2 (syscall 764)
4175 If you configure @value{GDBN} using the @samp{--without-expat} option,
4176 it will not be able to display syscall names. Also, if your
4177 architecture does not have an XML file describing its system calls,
4178 you will not be able to see the syscall names. It is important to
4179 notice that these two features are used for accessing the syscall
4180 name database. In either case, you will see a warning like this:
4183 (@value{GDBP}) catch syscall
4184 warning: Could not open "syscalls/i386-linux.xml"
4185 warning: Could not load the syscall XML file 'syscalls/i386-linux.xml'.
4186 GDB will not be able to display syscall names.
4187 Catchpoint 1 (syscall)
4191 Of course, the file name will change depending on your architecture and system.
4193 Still using the example above, you can also try to catch a syscall by its
4194 number. In this case, you would see something like:
4197 (@value{GDBP}) catch syscall 252
4198 Catchpoint 1 (syscall(s) 252)
4201 Again, in this case @value{GDBN} would not be able to display syscall's names.
4204 A call to @code{fork}. This is currently only available for HP-UX
4208 A call to @code{vfork}. This is currently only available for HP-UX
4211 @item load @r{[}regexp@r{]}
4212 @itemx unload @r{[}regexp@r{]}
4213 The loading or unloading of a shared library. If @var{regexp} is
4214 given, then the catchpoint will stop only if the regular expression
4215 matches one of the affected libraries.
4219 @item tcatch @var{event}
4220 Set a catchpoint that is enabled only for one stop. The catchpoint is
4221 automatically deleted after the first time the event is caught.
4225 Use the @code{info break} command to list the current catchpoints.
4227 There are currently some limitations to C@t{++} exception handling
4228 (@code{catch throw} and @code{catch catch}) in @value{GDBN}:
4232 If you call a function interactively, @value{GDBN} normally returns
4233 control to you when the function has finished executing. If the call
4234 raises an exception, however, the call may bypass the mechanism that
4235 returns control to you and cause your program either to abort or to
4236 simply continue running until it hits a breakpoint, catches a signal
4237 that @value{GDBN} is listening for, or exits. This is the case even if
4238 you set a catchpoint for the exception; catchpoints on exceptions are
4239 disabled within interactive calls.
4242 You cannot raise an exception interactively.
4245 You cannot install an exception handler interactively.
4248 @cindex raise exceptions
4249 Sometimes @code{catch} is not the best way to debug exception handling:
4250 if you need to know exactly where an exception is raised, it is better to
4251 stop @emph{before} the exception handler is called, since that way you
4252 can see the stack before any unwinding takes place. If you set a
4253 breakpoint in an exception handler instead, it may not be easy to find
4254 out where the exception was raised.
4256 To stop just before an exception handler is called, you need some
4257 knowledge of the implementation. In the case of @sc{gnu} C@t{++}, exceptions are
4258 raised by calling a library function named @code{__raise_exception}
4259 which has the following ANSI C interface:
4262 /* @var{addr} is where the exception identifier is stored.
4263 @var{id} is the exception identifier. */
4264 void __raise_exception (void **addr, void *id);
4268 To make the debugger catch all exceptions before any stack
4269 unwinding takes place, set a breakpoint on @code{__raise_exception}
4270 (@pxref{Breakpoints, ,Breakpoints; Watchpoints; and Exceptions}).
4272 With a conditional breakpoint (@pxref{Conditions, ,Break Conditions})
4273 that depends on the value of @var{id}, you can stop your program when
4274 a specific exception is raised. You can use multiple conditional
4275 breakpoints to stop your program when any of a number of exceptions are
4280 @subsection Deleting Breakpoints
4282 @cindex clearing breakpoints, watchpoints, catchpoints
4283 @cindex deleting breakpoints, watchpoints, catchpoints
4284 It is often necessary to eliminate a breakpoint, watchpoint, or
4285 catchpoint once it has done its job and you no longer want your program
4286 to stop there. This is called @dfn{deleting} the breakpoint. A
4287 breakpoint that has been deleted no longer exists; it is forgotten.
4289 With the @code{clear} command you can delete breakpoints according to
4290 where they are in your program. With the @code{delete} command you can
4291 delete individual breakpoints, watchpoints, or catchpoints by specifying
4292 their breakpoint numbers.
4294 It is not necessary to delete a breakpoint to proceed past it. @value{GDBN}
4295 automatically ignores breakpoints on the first instruction to be executed
4296 when you continue execution without changing the execution address.
4301 Delete any breakpoints at the next instruction to be executed in the
4302 selected stack frame (@pxref{Selection, ,Selecting a Frame}). When
4303 the innermost frame is selected, this is a good way to delete a
4304 breakpoint where your program just stopped.
4306 @item clear @var{location}
4307 Delete any breakpoints set at the specified @var{location}.
4308 @xref{Specify Location}, for the various forms of @var{location}; the
4309 most useful ones are listed below:
4312 @item clear @var{function}
4313 @itemx clear @var{filename}:@var{function}
4314 Delete any breakpoints set at entry to the named @var{function}.
4316 @item clear @var{linenum}
4317 @itemx clear @var{filename}:@var{linenum}
4318 Delete any breakpoints set at or within the code of the specified
4319 @var{linenum} of the specified @var{filename}.
4322 @cindex delete breakpoints
4324 @kindex d @r{(@code{delete})}
4325 @item delete @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
4326 Delete the breakpoints, watchpoints, or catchpoints of the breakpoint
4327 ranges specified as arguments. If no argument is specified, delete all
4328 breakpoints (@value{GDBN} asks confirmation, unless you have @code{set
4329 confirm off}). You can abbreviate this command as @code{d}.
4333 @subsection Disabling Breakpoints
4335 @cindex enable/disable a breakpoint
4336 Rather than deleting a breakpoint, watchpoint, or catchpoint, you might
4337 prefer to @dfn{disable} it. This makes the breakpoint inoperative as if
4338 it had been deleted, but remembers the information on the breakpoint so
4339 that you can @dfn{enable} it again later.
4341 You disable and enable breakpoints, watchpoints, and catchpoints with
4342 the @code{enable} and @code{disable} commands, optionally specifying
4343 one or more breakpoint numbers as arguments. Use @code{info break} to
4344 print a list of all breakpoints, watchpoints, and catchpoints if you
4345 do not know which numbers to use.
4347 Disabling and enabling a breakpoint that has multiple locations
4348 affects all of its locations.
4350 A breakpoint, watchpoint, or catchpoint can have any of several
4351 different states of enablement:
4355 Enabled. The breakpoint stops your program. A breakpoint set
4356 with the @code{break} command starts out in this state.
4358 Disabled. The breakpoint has no effect on your program.
4360 Enabled once. The breakpoint stops your program, but then becomes
4363 Enabled for a count. The breakpoint stops your program for the next
4364 N times, then becomes disabled.
4366 Enabled for deletion. The breakpoint stops your program, but
4367 immediately after it does so it is deleted permanently. A breakpoint
4368 set with the @code{tbreak} command starts out in this state.
4371 You can use the following commands to enable or disable breakpoints,
4372 watchpoints, and catchpoints:
4376 @kindex dis @r{(@code{disable})}
4377 @item disable @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
4378 Disable the specified breakpoints---or all breakpoints, if none are
4379 listed. A disabled breakpoint has no effect but is not forgotten. All
4380 options such as ignore-counts, conditions and commands are remembered in
4381 case the breakpoint is enabled again later. You may abbreviate
4382 @code{disable} as @code{dis}.
4385 @item enable @r{[}breakpoints@r{]} @r{[}@var{range}@dots{}@r{]}
4386 Enable the specified breakpoints (or all defined breakpoints). They
4387 become effective once again in stopping your program.
4389 @item enable @r{[}breakpoints@r{]} once @var{range}@dots{}
4390 Enable the specified breakpoints temporarily. @value{GDBN} disables any
4391 of these breakpoints immediately after stopping your program.
4393 @item enable @r{[}breakpoints@r{]} count @var{count} @var{range}@dots{}
4394 Enable the specified breakpoints temporarily. @value{GDBN} records
4395 @var{count} with each of the specified breakpoints, and decrements a
4396 breakpoint's count when it is hit. When any count reaches 0,
4397 @value{GDBN} disables that breakpoint. If a breakpoint has an ignore
4398 count (@pxref{Conditions, ,Break Conditions}), that will be
4399 decremented to 0 before @var{count} is affected.
4401 @item enable @r{[}breakpoints@r{]} delete @var{range}@dots{}
4402 Enable the specified breakpoints to work once, then die. @value{GDBN}
4403 deletes any of these breakpoints as soon as your program stops there.
4404 Breakpoints set by the @code{tbreak} command start out in this state.
4407 @c FIXME: I think the following ``Except for [...] @code{tbreak}'' is
4408 @c confusing: tbreak is also initially enabled.
4409 Except for a breakpoint set with @code{tbreak} (@pxref{Set Breaks,
4410 ,Setting Breakpoints}), breakpoints that you set are initially enabled;
4411 subsequently, they become disabled or enabled only when you use one of
4412 the commands above. (The command @code{until} can set and delete a
4413 breakpoint of its own, but it does not change the state of your other
4414 breakpoints; see @ref{Continuing and Stepping, ,Continuing and
4418 @subsection Break Conditions
4419 @cindex conditional breakpoints
4420 @cindex breakpoint conditions
4422 @c FIXME what is scope of break condition expr? Context where wanted?
4423 @c in particular for a watchpoint?
4424 The simplest sort of breakpoint breaks every time your program reaches a
4425 specified place. You can also specify a @dfn{condition} for a
4426 breakpoint. A condition is just a Boolean expression in your
4427 programming language (@pxref{Expressions, ,Expressions}). A breakpoint with
4428 a condition evaluates the expression each time your program reaches it,
4429 and your program stops only if the condition is @emph{true}.
4431 This is the converse of using assertions for program validation; in that
4432 situation, you want to stop when the assertion is violated---that is,
4433 when the condition is false. In C, if you want to test an assertion expressed
4434 by the condition @var{assert}, you should set the condition
4435 @samp{! @var{assert}} on the appropriate breakpoint.
4437 Conditions are also accepted for watchpoints; you may not need them,
4438 since a watchpoint is inspecting the value of an expression anyhow---but
4439 it might be simpler, say, to just set a watchpoint on a variable name,
4440 and specify a condition that tests whether the new value is an interesting
4443 Break conditions can have side effects, and may even call functions in
4444 your program. This can be useful, for example, to activate functions
4445 that log program progress, or to use your own print functions to
4446 format special data structures. The effects are completely predictable
4447 unless there is another enabled breakpoint at the same address. (In
4448 that case, @value{GDBN} might see the other breakpoint first and stop your
4449 program without checking the condition of this one.) Note that
4450 breakpoint commands are usually more convenient and flexible than break
4452 purpose of performing side effects when a breakpoint is reached
4453 (@pxref{Break Commands, ,Breakpoint Command Lists}).
4455 Breakpoint conditions can also be evaluated on the target's side if
4456 the target supports it. Instead of evaluating the conditions locally,
4457 @value{GDBN} encodes the expression into an agent expression
4458 (@pxref{Agent Expressions}) suitable for execution on the target,
4459 independently of @value{GDBN}. Global variables become raw memory
4460 locations, locals become stack accesses, and so forth.
4462 In this case, @value{GDBN} will only be notified of a breakpoint trigger
4463 when its condition evaluates to true. This mechanism may provide faster
4464 response times depending on the performance characteristics of the target
4465 since it does not need to keep @value{GDBN} informed about
4466 every breakpoint trigger, even those with false conditions.
4468 Break conditions can be specified when a breakpoint is set, by using
4469 @samp{if} in the arguments to the @code{break} command. @xref{Set
4470 Breaks, ,Setting Breakpoints}. They can also be changed at any time
4471 with the @code{condition} command.
4473 You can also use the @code{if} keyword with the @code{watch} command.
4474 The @code{catch} command does not recognize the @code{if} keyword;
4475 @code{condition} is the only way to impose a further condition on a
4480 @item condition @var{bnum} @var{expression}
4481 Specify @var{expression} as the break condition for breakpoint,
4482 watchpoint, or catchpoint number @var{bnum}. After you set a condition,
4483 breakpoint @var{bnum} stops your program only if the value of
4484 @var{expression} is true (nonzero, in C). When you use
4485 @code{condition}, @value{GDBN} checks @var{expression} immediately for
4486 syntactic correctness, and to determine whether symbols in it have
4487 referents in the context of your breakpoint. If @var{expression} uses
4488 symbols not referenced in the context of the breakpoint, @value{GDBN}
4489 prints an error message:
4492 No symbol "foo" in current context.
4497 not actually evaluate @var{expression} at the time the @code{condition}
4498 command (or a command that sets a breakpoint with a condition, like
4499 @code{break if @dots{}}) is given, however. @xref{Expressions, ,Expressions}.
4501 @item condition @var{bnum}
4502 Remove the condition from breakpoint number @var{bnum}. It becomes
4503 an ordinary unconditional breakpoint.
4506 @cindex ignore count (of breakpoint)
4507 A special case of a breakpoint condition is to stop only when the
4508 breakpoint has been reached a certain number of times. This is so
4509 useful that there is a special way to do it, using the @dfn{ignore
4510 count} of the breakpoint. Every breakpoint has an ignore count, which
4511 is an integer. Most of the time, the ignore count is zero, and
4512 therefore has no effect. But if your program reaches a breakpoint whose
4513 ignore count is positive, then instead of stopping, it just decrements
4514 the ignore count by one and continues. As a result, if the ignore count
4515 value is @var{n}, the breakpoint does not stop the next @var{n} times
4516 your program reaches it.
4520 @item ignore @var{bnum} @var{count}
4521 Set the ignore count of breakpoint number @var{bnum} to @var{count}.
4522 The next @var{count} times the breakpoint is reached, your program's
4523 execution does not stop; other than to decrement the ignore count, @value{GDBN}
4526 To make the breakpoint stop the next time it is reached, specify
4529 When you use @code{continue} to resume execution of your program from a
4530 breakpoint, you can specify an ignore count directly as an argument to
4531 @code{continue}, rather than using @code{ignore}. @xref{Continuing and
4532 Stepping,,Continuing and Stepping}.
4534 If a breakpoint has a positive ignore count and a condition, the
4535 condition is not checked. Once the ignore count reaches zero,
4536 @value{GDBN} resumes checking the condition.
4538 You could achieve the effect of the ignore count with a condition such
4539 as @w{@samp{$foo-- <= 0}} using a debugger convenience variable that
4540 is decremented each time. @xref{Convenience Vars, ,Convenience
4544 Ignore counts apply to breakpoints, watchpoints, and catchpoints.
4547 @node Break Commands
4548 @subsection Breakpoint Command Lists
4550 @cindex breakpoint commands
4551 You can give any breakpoint (or watchpoint or catchpoint) a series of
4552 commands to execute when your program stops due to that breakpoint. For
4553 example, you might want to print the values of certain expressions, or
4554 enable other breakpoints.
4558 @kindex end@r{ (breakpoint commands)}
4559 @item commands @r{[}@var{range}@dots{}@r{]}
4560 @itemx @dots{} @var{command-list} @dots{}
4562 Specify a list of commands for the given breakpoints. The commands
4563 themselves appear on the following lines. Type a line containing just
4564 @code{end} to terminate the commands.
4566 To remove all commands from a breakpoint, type @code{commands} and
4567 follow it immediately with @code{end}; that is, give no commands.
4569 With no argument, @code{commands} refers to the last breakpoint,
4570 watchpoint, or catchpoint set (not to the breakpoint most recently
4571 encountered). If the most recent breakpoints were set with a single
4572 command, then the @code{commands} will apply to all the breakpoints
4573 set by that command. This applies to breakpoints set by
4574 @code{rbreak}, and also applies when a single @code{break} command
4575 creates multiple breakpoints (@pxref{Ambiguous Expressions,,Ambiguous
4579 Pressing @key{RET} as a means of repeating the last @value{GDBN} command is
4580 disabled within a @var{command-list}.
4582 You can use breakpoint commands to start your program up again. Simply
4583 use the @code{continue} command, or @code{step}, or any other command
4584 that resumes execution.
4586 Any other commands in the command list, after a command that resumes
4587 execution, are ignored. This is because any time you resume execution
4588 (even with a simple @code{next} or @code{step}), you may encounter
4589 another breakpoint---which could have its own command list, leading to
4590 ambiguities about which list to execute.
4593 If the first command you specify in a command list is @code{silent}, the
4594 usual message about stopping at a breakpoint is not printed. This may
4595 be desirable for breakpoints that are to print a specific message and
4596 then continue. If none of the remaining commands print anything, you
4597 see no sign that the breakpoint was reached. @code{silent} is
4598 meaningful only at the beginning of a breakpoint command list.
4600 The commands @code{echo}, @code{output}, and @code{printf} allow you to
4601 print precisely controlled output, and are often useful in silent
4602 breakpoints. @xref{Output, ,Commands for Controlled Output}.
4604 For example, here is how you could use breakpoint commands to print the
4605 value of @code{x} at entry to @code{foo} whenever @code{x} is positive.
4611 printf "x is %d\n",x
4616 One application for breakpoint commands is to compensate for one bug so
4617 you can test for another. Put a breakpoint just after the erroneous line
4618 of code, give it a condition to detect the case in which something
4619 erroneous has been done, and give it commands to assign correct values
4620 to any variables that need them. End with the @code{continue} command
4621 so that your program does not stop, and start with the @code{silent}
4622 command so that no output is produced. Here is an example:
4633 @node Dynamic Printf
4634 @subsection Dynamic Printf
4636 @cindex dynamic printf
4638 The dynamic printf command @code{dprintf} combines a breakpoint with
4639 formatted printing of your program's data to give you the effect of
4640 inserting @code{printf} calls into your program on-the-fly, without
4641 having to recompile it.
4643 In its most basic form, the output goes to the GDB console. However,
4644 you can set the variable @code{dprintf-style} for alternate handling.
4645 For instance, you can ask to format the output by calling your
4646 program's @code{printf} function. This has the advantage that the
4647 characters go to the program's output device, so they can recorded in
4648 redirects to files and so forth.
4652 @item dprintf @var{location},@var{template},@var{expression}[,@var{expression}@dots{}]
4653 Whenever execution reaches @var{location}, print the values of one or
4654 more @var{expressions} under the control of the string @var{template}.
4655 To print several values, separate them with commas.
4657 @item set dprintf-style @var{style}
4658 Set the dprintf output to be handled in one of several different
4659 styles enumerated below. A change of style affects all existing
4660 dynamic printfs immediately. (If you need individual control over the
4661 print commands, simply define normal breakpoints with
4662 explicitly-supplied command lists.)
4665 @kindex dprintf-style gdb
4666 Handle the output using the @value{GDBN} @code{printf} command.
4669 @kindex dprintf-style call
4670 Handle the output by calling a function in your program (normally
4673 @item set dprintf-function @var{function}
4674 Set the function to call if the dprintf style is @code{call}. By
4675 default its value is @code{printf}. You may set it to any expression.
4676 that @value{GDBN} can evaluate to a function, as per the @code{call}
4679 @item set dprintf-channel @var{channel}
4680 Set a ``channel'' for dprintf. If set to a non-empty value,
4681 @value{GDBN} will evaluate it as an expression and pass the result as
4682 a first argument to the @code{dprintf-function}, in the manner of
4683 @code{fprintf} and similar functions. Otherwise, the dprintf format
4684 string will be the first argument, in the manner of @code{printf}.
4686 As an example, if you wanted @code{dprintf} output to go to a logfile
4687 that is a standard I/O stream assigned to the variable @code{mylog},
4688 you could do the following:
4691 (gdb) set dprintf-style call
4692 (gdb) set dprintf-function fprintf
4693 (gdb) set dprintf-channel mylog
4694 (gdb) dprintf 25,"at line 25, glob=%d\n",glob
4695 Dprintf 1 at 0x123456: file main.c, line 25.
4697 1 dprintf keep y 0x00123456 in main at main.c:25
4698 call (void) fprintf (mylog,"at line 25, glob=%d\n",glob)
4703 Note that the @code{info break} displays the dynamic printf commands
4704 as normal breakpoint commands; you can thus easily see the effect of
4705 the variable settings.
4709 @value{GDBN} does not check the validity of function and channel,
4710 relying on you to supply values that are meaningful for the contexts
4711 in which they are being used. For instance, the function and channel
4712 may be the values of local variables, but if that is the case, then
4713 all enabled dynamic prints must be at locations within the scope of
4714 those locals. If evaluation fails, @value{GDBN} will report an error.
4716 @node Save Breakpoints
4717 @subsection How to save breakpoints to a file
4719 To save breakpoint definitions to a file use the @w{@code{save
4720 breakpoints}} command.
4723 @kindex save breakpoints
4724 @cindex save breakpoints to a file for future sessions
4725 @item save breakpoints [@var{filename}]
4726 This command saves all current breakpoint definitions together with
4727 their commands and ignore counts, into a file @file{@var{filename}}
4728 suitable for use in a later debugging session. This includes all
4729 types of breakpoints (breakpoints, watchpoints, catchpoints,
4730 tracepoints). To read the saved breakpoint definitions, use the
4731 @code{source} command (@pxref{Command Files}). Note that watchpoints
4732 with expressions involving local variables may fail to be recreated
4733 because it may not be possible to access the context where the
4734 watchpoint is valid anymore. Because the saved breakpoint definitions
4735 are simply a sequence of @value{GDBN} commands that recreate the
4736 breakpoints, you can edit the file in your favorite editing program,
4737 and remove the breakpoint definitions you're not interested in, or
4738 that can no longer be recreated.
4741 @node Static Probe Points
4742 @subsection Static Probe Points
4744 @cindex static probe point, SystemTap
4745 @value{GDBN} supports @dfn{SDT} probes in the code. @acronym{SDT} stands
4746 for Statically Defined Tracing, and the probes are designed to have a tiny
4747 runtime code and data footprint, and no dynamic relocations. They are
4748 usable from assembly, C and C@t{++} languages. See
4749 @uref{http://sourceware.org/systemtap/wiki/UserSpaceProbeImplementation}
4750 for a good reference on how the @acronym{SDT} probes are implemented.
4752 Currently, @code{SystemTap} (@uref{http://sourceware.org/systemtap/})
4753 @acronym{SDT} probes are supported on ELF-compatible systems. See
4754 @uref{http://sourceware.org/systemtap/wiki/AddingUserSpaceProbingToApps}
4755 for more information on how to add @code{SystemTap} @acronym{SDT} probes
4756 in your applications.
4758 @cindex semaphores on static probe points
4759 Some probes have an associated semaphore variable; for instance, this
4760 happens automatically if you defined your probe using a DTrace-style
4761 @file{.d} file. If your probe has a semaphore, @value{GDBN} will
4762 automatically enable it when you specify a breakpoint using the
4763 @samp{-probe-stap} notation. But, if you put a breakpoint at a probe's
4764 location by some other method (e.g., @code{break file:line}), then
4765 @value{GDBN} will not automatically set the semaphore.
4767 You can examine the available static static probes using @code{info
4768 probes}, with optional arguments:
4772 @item info probes stap @r{[}@var{provider} @r{[}@var{name} @r{[}@var{objfile}@r{]}@r{]}@r{]}
4773 If given, @var{provider} is a regular expression used to match against provider
4774 names when selecting which probes to list. If omitted, probes by all
4775 probes from all providers are listed.
4777 If given, @var{name} is a regular expression to match against probe names
4778 when selecting which probes to list. If omitted, probe names are not
4779 considered when deciding whether to display them.
4781 If given, @var{objfile} is a regular expression used to select which
4782 object files (executable or shared libraries) to examine. If not
4783 given, all object files are considered.
4785 @item info probes all
4786 List the available static probes, from all types.
4789 @vindex $_probe_arg@r{, convenience variable}
4790 A probe may specify up to twelve arguments. These are available at the
4791 point at which the probe is defined---that is, when the current PC is
4792 at the probe's location. The arguments are available using the
4793 convenience variables (@pxref{Convenience Vars})
4794 @code{$_probe_arg0}@dots{}@code{$_probe_arg11}. Each probe argument is
4795 an integer of the appropriate size; types are not preserved. The
4796 convenience variable @code{$_probe_argc} holds the number of arguments
4797 at the current probe point.
4799 These variables are always available, but attempts to access them at
4800 any location other than a probe point will cause @value{GDBN} to give
4804 @c @ifclear BARETARGET
4805 @node Error in Breakpoints
4806 @subsection ``Cannot insert breakpoints''
4808 If you request too many active hardware-assisted breakpoints and
4809 watchpoints, you will see this error message:
4811 @c FIXME: the precise wording of this message may change; the relevant
4812 @c source change is not committed yet (Sep 3, 1999).
4814 Stopped; cannot insert breakpoints.
4815 You may have requested too many hardware breakpoints and watchpoints.
4819 This message is printed when you attempt to resume the program, since
4820 only then @value{GDBN} knows exactly how many hardware breakpoints and
4821 watchpoints it needs to insert.
4823 When this message is printed, you need to disable or remove some of the
4824 hardware-assisted breakpoints and watchpoints, and then continue.
4826 @node Breakpoint-related Warnings
4827 @subsection ``Breakpoint address adjusted...''
4828 @cindex breakpoint address adjusted
4830 Some processor architectures place constraints on the addresses at
4831 which breakpoints may be placed. For architectures thus constrained,
4832 @value{GDBN} will attempt to adjust the breakpoint's address to comply
4833 with the constraints dictated by the architecture.
4835 One example of such an architecture is the Fujitsu FR-V. The FR-V is
4836 a VLIW architecture in which a number of RISC-like instructions may be
4837 bundled together for parallel execution. The FR-V architecture
4838 constrains the location of a breakpoint instruction within such a
4839 bundle to the instruction with the lowest address. @value{GDBN}
4840 honors this constraint by adjusting a breakpoint's address to the
4841 first in the bundle.
4843 It is not uncommon for optimized code to have bundles which contain
4844 instructions from different source statements, thus it may happen that
4845 a breakpoint's address will be adjusted from one source statement to
4846 another. Since this adjustment may significantly alter @value{GDBN}'s
4847 breakpoint related behavior from what the user expects, a warning is
4848 printed when the breakpoint is first set and also when the breakpoint
4851 A warning like the one below is printed when setting a breakpoint
4852 that's been subject to address adjustment:
4855 warning: Breakpoint address adjusted from 0x00010414 to 0x00010410.
4858 Such warnings are printed both for user settable and @value{GDBN}'s
4859 internal breakpoints. If you see one of these warnings, you should
4860 verify that a breakpoint set at the adjusted address will have the
4861 desired affect. If not, the breakpoint in question may be removed and
4862 other breakpoints may be set which will have the desired behavior.
4863 E.g., it may be sufficient to place the breakpoint at a later
4864 instruction. A conditional breakpoint may also be useful in some
4865 cases to prevent the breakpoint from triggering too often.
4867 @value{GDBN} will also issue a warning when stopping at one of these
4868 adjusted breakpoints:
4871 warning: Breakpoint 1 address previously adjusted from 0x00010414
4875 When this warning is encountered, it may be too late to take remedial
4876 action except in cases where the breakpoint is hit earlier or more
4877 frequently than expected.
4879 @node Continuing and Stepping
4880 @section Continuing and Stepping
4884 @cindex resuming execution
4885 @dfn{Continuing} means resuming program execution until your program
4886 completes normally. In contrast, @dfn{stepping} means executing just
4887 one more ``step'' of your program, where ``step'' may mean either one
4888 line of source code, or one machine instruction (depending on what
4889 particular command you use). Either when continuing or when stepping,
4890 your program may stop even sooner, due to a breakpoint or a signal. (If
4891 it stops due to a signal, you may want to use @code{handle}, or use
4892 @samp{signal 0} to resume execution. @xref{Signals, ,Signals}.)
4896 @kindex c @r{(@code{continue})}
4897 @kindex fg @r{(resume foreground execution)}
4898 @item continue @r{[}@var{ignore-count}@r{]}
4899 @itemx c @r{[}@var{ignore-count}@r{]}
4900 @itemx fg @r{[}@var{ignore-count}@r{]}
4901 Resume program execution, at the address where your program last stopped;
4902 any breakpoints set at that address are bypassed. The optional argument
4903 @var{ignore-count} allows you to specify a further number of times to
4904 ignore a breakpoint at this location; its effect is like that of
4905 @code{ignore} (@pxref{Conditions, ,Break Conditions}).
4907 The argument @var{ignore-count} is meaningful only when your program
4908 stopped due to a breakpoint. At other times, the argument to
4909 @code{continue} is ignored.
4911 The synonyms @code{c} and @code{fg} (for @dfn{foreground}, as the
4912 debugged program is deemed to be the foreground program) are provided
4913 purely for convenience, and have exactly the same behavior as
4917 To resume execution at a different place, you can use @code{return}
4918 (@pxref{Returning, ,Returning from a Function}) to go back to the
4919 calling function; or @code{jump} (@pxref{Jumping, ,Continuing at a
4920 Different Address}) to go to an arbitrary location in your program.
4922 A typical technique for using stepping is to set a breakpoint
4923 (@pxref{Breakpoints, ,Breakpoints; Watchpoints; and Catchpoints}) at the
4924 beginning of the function or the section of your program where a problem
4925 is believed to lie, run your program until it stops at that breakpoint,
4926 and then step through the suspect area, examining the variables that are
4927 interesting, until you see the problem happen.
4931 @kindex s @r{(@code{step})}
4933 Continue running your program until control reaches a different source
4934 line, then stop it and return control to @value{GDBN}. This command is
4935 abbreviated @code{s}.
4938 @c "without debugging information" is imprecise; actually "without line
4939 @c numbers in the debugging information". (gcc -g1 has debugging info but
4940 @c not line numbers). But it seems complex to try to make that
4941 @c distinction here.
4942 @emph{Warning:} If you use the @code{step} command while control is
4943 within a function that was compiled without debugging information,
4944 execution proceeds until control reaches a function that does have
4945 debugging information. Likewise, it will not step into a function which
4946 is compiled without debugging information. To step through functions
4947 without debugging information, use the @code{stepi} command, described
4951 The @code{step} command only stops at the first instruction of a source
4952 line. This prevents the multiple stops that could otherwise occur in
4953 @code{switch} statements, @code{for} loops, etc. @code{step} continues
4954 to stop if a function that has debugging information is called within
4955 the line. In other words, @code{step} @emph{steps inside} any functions
4956 called within the line.
4958 Also, the @code{step} command only enters a function if there is line
4959 number information for the function. Otherwise it acts like the
4960 @code{next} command. This avoids problems when using @code{cc -gl}
4961 on @acronym{MIPS} machines. Previously, @code{step} entered subroutines if there
4962 was any debugging information about the routine.
4964 @item step @var{count}
4965 Continue running as in @code{step}, but do so @var{count} times. If a
4966 breakpoint is reached, or a signal not related to stepping occurs before
4967 @var{count} steps, stepping stops right away.
4970 @kindex n @r{(@code{next})}
4971 @item next @r{[}@var{count}@r{]}
4972 Continue to the next source line in the current (innermost) stack frame.
4973 This is similar to @code{step}, but function calls that appear within
4974 the line of code are executed without stopping. Execution stops when
4975 control reaches a different line of code at the original stack level
4976 that was executing when you gave the @code{next} command. This command
4977 is abbreviated @code{n}.
4979 An argument @var{count} is a repeat count, as for @code{step}.
4982 @c FIX ME!! Do we delete this, or is there a way it fits in with
4983 @c the following paragraph? --- Vctoria
4985 @c @code{next} within a function that lacks debugging information acts like
4986 @c @code{step}, but any function calls appearing within the code of the
4987 @c function are executed without stopping.
4989 The @code{next} command only stops at the first instruction of a
4990 source line. This prevents multiple stops that could otherwise occur in
4991 @code{switch} statements, @code{for} loops, etc.
4993 @kindex set step-mode
4995 @cindex functions without line info, and stepping
4996 @cindex stepping into functions with no line info
4997 @itemx set step-mode on
4998 The @code{set step-mode on} command causes the @code{step} command to
4999 stop at the first instruction of a function which contains no debug line
5000 information rather than stepping over it.
5002 This is useful in cases where you may be interested in inspecting the
5003 machine instructions of a function which has no symbolic info and do not
5004 want @value{GDBN} to automatically skip over this function.
5006 @item set step-mode off
5007 Causes the @code{step} command to step over any functions which contains no
5008 debug information. This is the default.
5010 @item show step-mode
5011 Show whether @value{GDBN} will stop in or step over functions without
5012 source line debug information.
5015 @kindex fin @r{(@code{finish})}
5017 Continue running until just after function in the selected stack frame
5018 returns. Print the returned value (if any). This command can be
5019 abbreviated as @code{fin}.
5021 Contrast this with the @code{return} command (@pxref{Returning,
5022 ,Returning from a Function}).
5025 @kindex u @r{(@code{until})}
5026 @cindex run until specified location
5029 Continue running until a source line past the current line, in the
5030 current stack frame, is reached. This command is used to avoid single
5031 stepping through a loop more than once. It is like the @code{next}
5032 command, except that when @code{until} encounters a jump, it
5033 automatically continues execution until the program counter is greater
5034 than the address of the jump.
5036 This means that when you reach the end of a loop after single stepping
5037 though it, @code{until} makes your program continue execution until it
5038 exits the loop. In contrast, a @code{next} command at the end of a loop
5039 simply steps back to the beginning of the loop, which forces you to step
5040 through the next iteration.
5042 @code{until} always stops your program if it attempts to exit the current
5045 @code{until} may produce somewhat counterintuitive results if the order
5046 of machine code does not match the order of the source lines. For
5047 example, in the following excerpt from a debugging session, the @code{f}
5048 (@code{frame}) command shows that execution is stopped at line
5049 @code{206}; yet when we use @code{until}, we get to line @code{195}:
5053 #0 main (argc=4, argv=0xf7fffae8) at m4.c:206
5055 (@value{GDBP}) until
5056 195 for ( ; argc > 0; NEXTARG) @{
5059 This happened because, for execution efficiency, the compiler had
5060 generated code for the loop closure test at the end, rather than the
5061 start, of the loop---even though the test in a C @code{for}-loop is
5062 written before the body of the loop. The @code{until} command appeared
5063 to step back to the beginning of the loop when it advanced to this
5064 expression; however, it has not really gone to an earlier
5065 statement---not in terms of the actual machine code.
5067 @code{until} with no argument works by means of single
5068 instruction stepping, and hence is slower than @code{until} with an
5071 @item until @var{location}
5072 @itemx u @var{location}
5073 Continue running your program until either the specified location is
5074 reached, or the current stack frame returns. @var{location} is any of
5075 the forms described in @ref{Specify Location}.
5076 This form of the command uses temporary breakpoints, and
5077 hence is quicker than @code{until} without an argument. The specified
5078 location is actually reached only if it is in the current frame. This
5079 implies that @code{until} can be used to skip over recursive function
5080 invocations. For instance in the code below, if the current location is
5081 line @code{96}, issuing @code{until 99} will execute the program up to
5082 line @code{99} in the same invocation of factorial, i.e., after the inner
5083 invocations have returned.
5086 94 int factorial (int value)
5088 96 if (value > 1) @{
5089 97 value *= factorial (value - 1);
5096 @kindex advance @var{location}
5097 @itemx advance @var{location}
5098 Continue running the program up to the given @var{location}. An argument is
5099 required, which should be of one of the forms described in
5100 @ref{Specify Location}.
5101 Execution will also stop upon exit from the current stack
5102 frame. This command is similar to @code{until}, but @code{advance} will
5103 not skip over recursive function calls, and the target location doesn't
5104 have to be in the same frame as the current one.
5108 @kindex si @r{(@code{stepi})}
5110 @itemx stepi @var{arg}
5112 Execute one machine instruction, then stop and return to the debugger.
5114 It is often useful to do @samp{display/i $pc} when stepping by machine
5115 instructions. This makes @value{GDBN} automatically display the next
5116 instruction to be executed, each time your program stops. @xref{Auto
5117 Display,, Automatic Display}.
5119 An argument is a repeat count, as in @code{step}.
5123 @kindex ni @r{(@code{nexti})}
5125 @itemx nexti @var{arg}
5127 Execute one machine instruction, but if it is a function call,
5128 proceed until the function returns.
5130 An argument is a repeat count, as in @code{next}.
5133 @node Skipping Over Functions and Files
5134 @section Skipping Over Functions and Files
5135 @cindex skipping over functions and files
5137 The program you are debugging may contain some functions which are
5138 uninteresting to debug. The @code{skip} comand lets you tell @value{GDBN} to
5139 skip a function or all functions in a file when stepping.
5141 For example, consider the following C function:
5152 Suppose you wish to step into the functions @code{foo} and @code{bar}, but you
5153 are not interested in stepping through @code{boring}. If you run @code{step}
5154 at line 103, you'll enter @code{boring()}, but if you run @code{next}, you'll
5155 step over both @code{foo} and @code{boring}!
5157 One solution is to @code{step} into @code{boring} and use the @code{finish}
5158 command to immediately exit it. But this can become tedious if @code{boring}
5159 is called from many places.
5161 A more flexible solution is to execute @kbd{skip boring}. This instructs
5162 @value{GDBN} never to step into @code{boring}. Now when you execute
5163 @code{step} at line 103, you'll step over @code{boring} and directly into
5166 You can also instruct @value{GDBN} to skip all functions in a file, with, for
5167 example, @code{skip file boring.c}.
5170 @kindex skip function
5171 @item skip @r{[}@var{linespec}@r{]}
5172 @itemx skip function @r{[}@var{linespec}@r{]}
5173 After running this command, the function named by @var{linespec} or the
5174 function containing the line named by @var{linespec} will be skipped over when
5175 stepping. @xref{Specify Location}.
5177 If you do not specify @var{linespec}, the function you're currently debugging
5180 (If you have a function called @code{file} that you want to skip, use
5181 @kbd{skip function file}.)
5184 @item skip file @r{[}@var{filename}@r{]}
5185 After running this command, any function whose source lives in @var{filename}
5186 will be skipped over when stepping.
5188 If you do not specify @var{filename}, functions whose source lives in the file
5189 you're currently debugging will be skipped.
5192 Skips can be listed, deleted, disabled, and enabled, much like breakpoints.
5193 These are the commands for managing your list of skips:
5197 @item info skip @r{[}@var{range}@r{]}
5198 Print details about the specified skip(s). If @var{range} is not specified,
5199 print a table with details about all functions and files marked for skipping.
5200 @code{info skip} prints the following information about each skip:
5204 A number identifying this skip.
5206 The type of this skip, either @samp{function} or @samp{file}.
5207 @item Enabled or Disabled
5208 Enabled skips are marked with @samp{y}. Disabled skips are marked with @samp{n}.
5210 For function skips, this column indicates the address in memory of the function
5211 being skipped. If you've set a function skip on a function which has not yet
5212 been loaded, this field will contain @samp{<PENDING>}. Once a shared library
5213 which has the function is loaded, @code{info skip} will show the function's
5216 For file skips, this field contains the filename being skipped. For functions
5217 skips, this field contains the function name and its line number in the file
5218 where it is defined.
5222 @item skip delete @r{[}@var{range}@r{]}
5223 Delete the specified skip(s). If @var{range} is not specified, delete all
5227 @item skip enable @r{[}@var{range}@r{]}
5228 Enable the specified skip(s). If @var{range} is not specified, enable all
5231 @kindex skip disable
5232 @item skip disable @r{[}@var{range}@r{]}
5233 Disable the specified skip(s). If @var{range} is not specified, disable all
5242 A signal is an asynchronous event that can happen in a program. The
5243 operating system defines the possible kinds of signals, and gives each
5244 kind a name and a number. For example, in Unix @code{SIGINT} is the
5245 signal a program gets when you type an interrupt character (often @kbd{Ctrl-c});
5246 @code{SIGSEGV} is the signal a program gets from referencing a place in
5247 memory far away from all the areas in use; @code{SIGALRM} occurs when
5248 the alarm clock timer goes off (which happens only if your program has
5249 requested an alarm).
5251 @cindex fatal signals
5252 Some signals, including @code{SIGALRM}, are a normal part of the
5253 functioning of your program. Others, such as @code{SIGSEGV}, indicate
5254 errors; these signals are @dfn{fatal} (they kill your program immediately) if the
5255 program has not specified in advance some other way to handle the signal.
5256 @code{SIGINT} does not indicate an error in your program, but it is normally
5257 fatal so it can carry out the purpose of the interrupt: to kill the program.
5259 @value{GDBN} has the ability to detect any occurrence of a signal in your
5260 program. You can tell @value{GDBN} in advance what to do for each kind of
5263 @cindex handling signals
5264 Normally, @value{GDBN} is set up to let the non-erroneous signals like
5265 @code{SIGALRM} be silently passed to your program
5266 (so as not to interfere with their role in the program's functioning)
5267 but to stop your program immediately whenever an error signal happens.
5268 You can change these settings with the @code{handle} command.
5271 @kindex info signals
5275 Print a table of all the kinds of signals and how @value{GDBN} has been told to
5276 handle each one. You can use this to see the signal numbers of all
5277 the defined types of signals.
5279 @item info signals @var{sig}
5280 Similar, but print information only about the specified signal number.
5282 @code{info handle} is an alias for @code{info signals}.
5285 @item handle @var{signal} @r{[}@var{keywords}@dots{}@r{]}
5286 Change the way @value{GDBN} handles signal @var{signal}. @var{signal}
5287 can be the number of a signal or its name (with or without the
5288 @samp{SIG} at the beginning); a list of signal numbers of the form
5289 @samp{@var{low}-@var{high}}; or the word @samp{all}, meaning all the
5290 known signals. Optional arguments @var{keywords}, described below,
5291 say what change to make.
5295 The keywords allowed by the @code{handle} command can be abbreviated.
5296 Their full names are:
5300 @value{GDBN} should not stop your program when this signal happens. It may
5301 still print a message telling you that the signal has come in.
5304 @value{GDBN} should stop your program when this signal happens. This implies
5305 the @code{print} keyword as well.
5308 @value{GDBN} should print a message when this signal happens.
5311 @value{GDBN} should not mention the occurrence of the signal at all. This
5312 implies the @code{nostop} keyword as well.
5316 @value{GDBN} should allow your program to see this signal; your program
5317 can handle the signal, or else it may terminate if the signal is fatal
5318 and not handled. @code{pass} and @code{noignore} are synonyms.
5322 @value{GDBN} should not allow your program to see this signal.
5323 @code{nopass} and @code{ignore} are synonyms.
5327 When a signal stops your program, the signal is not visible to the
5329 continue. Your program sees the signal then, if @code{pass} is in
5330 effect for the signal in question @emph{at that time}. In other words,
5331 after @value{GDBN} reports a signal, you can use the @code{handle}
5332 command with @code{pass} or @code{nopass} to control whether your
5333 program sees that signal when you continue.
5335 The default is set to @code{nostop}, @code{noprint}, @code{pass} for
5336 non-erroneous signals such as @code{SIGALRM}, @code{SIGWINCH} and
5337 @code{SIGCHLD}, and to @code{stop}, @code{print}, @code{pass} for the
5340 You can also use the @code{signal} command to prevent your program from
5341 seeing a signal, or cause it to see a signal it normally would not see,
5342 or to give it any signal at any time. For example, if your program stopped
5343 due to some sort of memory reference error, you might store correct
5344 values into the erroneous variables and continue, hoping to see more
5345 execution; but your program would probably terminate immediately as
5346 a result of the fatal signal once it saw the signal. To prevent this,
5347 you can continue with @samp{signal 0}. @xref{Signaling, ,Giving your
5350 @cindex extra signal information
5351 @anchor{extra signal information}
5353 On some targets, @value{GDBN} can inspect extra signal information
5354 associated with the intercepted signal, before it is actually
5355 delivered to the program being debugged. This information is exported
5356 by the convenience variable @code{$_siginfo}, and consists of data
5357 that is passed by the kernel to the signal handler at the time of the
5358 receipt of a signal. The data type of the information itself is
5359 target dependent. You can see the data type using the @code{ptype
5360 $_siginfo} command. On Unix systems, it typically corresponds to the
5361 standard @code{siginfo_t} type, as defined in the @file{signal.h}
5364 Here's an example, on a @sc{gnu}/Linux system, printing the stray
5365 referenced address that raised a segmentation fault.
5369 (@value{GDBP}) continue
5370 Program received signal SIGSEGV, Segmentation fault.
5371 0x0000000000400766 in main ()
5373 (@value{GDBP}) ptype $_siginfo
5380 struct @{...@} _kill;
5381 struct @{...@} _timer;
5383 struct @{...@} _sigchld;
5384 struct @{...@} _sigfault;
5385 struct @{...@} _sigpoll;
5388 (@value{GDBP}) ptype $_siginfo._sifields._sigfault
5392 (@value{GDBP}) p $_siginfo._sifields._sigfault.si_addr
5393 $1 = (void *) 0x7ffff7ff7000
5397 Depending on target support, @code{$_siginfo} may also be writable.
5400 @section Stopping and Starting Multi-thread Programs
5402 @cindex stopped threads
5403 @cindex threads, stopped
5405 @cindex continuing threads
5406 @cindex threads, continuing
5408 @value{GDBN} supports debugging programs with multiple threads
5409 (@pxref{Threads,, Debugging Programs with Multiple Threads}). There
5410 are two modes of controlling execution of your program within the
5411 debugger. In the default mode, referred to as @dfn{all-stop mode},
5412 when any thread in your program stops (for example, at a breakpoint
5413 or while being stepped), all other threads in the program are also stopped by
5414 @value{GDBN}. On some targets, @value{GDBN} also supports
5415 @dfn{non-stop mode}, in which other threads can continue to run freely while
5416 you examine the stopped thread in the debugger.
5419 * All-Stop Mode:: All threads stop when GDB takes control
5420 * Non-Stop Mode:: Other threads continue to execute
5421 * Background Execution:: Running your program asynchronously
5422 * Thread-Specific Breakpoints:: Controlling breakpoints
5423 * Interrupted System Calls:: GDB may interfere with system calls
5424 * Observer Mode:: GDB does not alter program behavior
5428 @subsection All-Stop Mode
5430 @cindex all-stop mode
5432 In all-stop mode, whenever your program stops under @value{GDBN} for any reason,
5433 @emph{all} threads of execution stop, not just the current thread. This
5434 allows you to examine the overall state of the program, including
5435 switching between threads, without worrying that things may change
5438 Conversely, whenever you restart the program, @emph{all} threads start
5439 executing. @emph{This is true even when single-stepping} with commands
5440 like @code{step} or @code{next}.
5442 In particular, @value{GDBN} cannot single-step all threads in lockstep.
5443 Since thread scheduling is up to your debugging target's operating
5444 system (not controlled by @value{GDBN}), other threads may
5445 execute more than one statement while the current thread completes a
5446 single step. Moreover, in general other threads stop in the middle of a
5447 statement, rather than at a clean statement boundary, when the program
5450 You might even find your program stopped in another thread after
5451 continuing or even single-stepping. This happens whenever some other
5452 thread runs into a breakpoint, a signal, or an exception before the
5453 first thread completes whatever you requested.
5455 @cindex automatic thread selection
5456 @cindex switching threads automatically
5457 @cindex threads, automatic switching
5458 Whenever @value{GDBN} stops your program, due to a breakpoint or a
5459 signal, it automatically selects the thread where that breakpoint or
5460 signal happened. @value{GDBN} alerts you to the context switch with a
5461 message such as @samp{[Switching to Thread @var{n}]} to identify the
5464 On some OSes, you can modify @value{GDBN}'s default behavior by
5465 locking the OS scheduler to allow only a single thread to run.
5468 @item set scheduler-locking @var{mode}
5469 @cindex scheduler locking mode
5470 @cindex lock scheduler
5471 Set the scheduler locking mode. If it is @code{off}, then there is no
5472 locking and any thread may run at any time. If @code{on}, then only the
5473 current thread may run when the inferior is resumed. The @code{step}
5474 mode optimizes for single-stepping; it prevents other threads
5475 from preempting the current thread while you are stepping, so that
5476 the focus of debugging does not change unexpectedly.
5477 Other threads only rarely (or never) get a chance to run
5478 when you step. They are more likely to run when you @samp{next} over a
5479 function call, and they are completely free to run when you use commands
5480 like @samp{continue}, @samp{until}, or @samp{finish}. However, unless another
5481 thread hits a breakpoint during its timeslice, @value{GDBN} does not change
5482 the current thread away from the thread that you are debugging.
5484 @item show scheduler-locking
5485 Display the current scheduler locking mode.
5488 @cindex resume threads of multiple processes simultaneously
5489 By default, when you issue one of the execution commands such as
5490 @code{continue}, @code{next} or @code{step}, @value{GDBN} allows only
5491 threads of the current inferior to run. For example, if @value{GDBN}
5492 is attached to two inferiors, each with two threads, the
5493 @code{continue} command resumes only the two threads of the current
5494 inferior. This is useful, for example, when you debug a program that
5495 forks and you want to hold the parent stopped (so that, for instance,
5496 it doesn't run to exit), while you debug the child. In other
5497 situations, you may not be interested in inspecting the current state
5498 of any of the processes @value{GDBN} is attached to, and you may want
5499 to resume them all until some breakpoint is hit. In the latter case,
5500 you can instruct @value{GDBN} to allow all threads of all the
5501 inferiors to run with the @w{@code{set schedule-multiple}} command.
5504 @kindex set schedule-multiple
5505 @item set schedule-multiple
5506 Set the mode for allowing threads of multiple processes to be resumed
5507 when an execution command is issued. When @code{on}, all threads of
5508 all processes are allowed to run. When @code{off}, only the threads
5509 of the current process are resumed. The default is @code{off}. The
5510 @code{scheduler-locking} mode takes precedence when set to @code{on},
5511 or while you are stepping and set to @code{step}.
5513 @item show schedule-multiple
5514 Display the current mode for resuming the execution of threads of
5519 @subsection Non-Stop Mode
5521 @cindex non-stop mode
5523 @c This section is really only a place-holder, and needs to be expanded
5524 @c with more details.
5526 For some multi-threaded targets, @value{GDBN} supports an optional
5527 mode of operation in which you can examine stopped program threads in
5528 the debugger while other threads continue to execute freely. This
5529 minimizes intrusion when debugging live systems, such as programs
5530 where some threads have real-time constraints or must continue to
5531 respond to external events. This is referred to as @dfn{non-stop} mode.
5533 In non-stop mode, when a thread stops to report a debugging event,
5534 @emph{only} that thread is stopped; @value{GDBN} does not stop other
5535 threads as well, in contrast to the all-stop mode behavior. Additionally,
5536 execution commands such as @code{continue} and @code{step} apply by default
5537 only to the current thread in non-stop mode, rather than all threads as
5538 in all-stop mode. This allows you to control threads explicitly in
5539 ways that are not possible in all-stop mode --- for example, stepping
5540 one thread while allowing others to run freely, stepping
5541 one thread while holding all others stopped, or stepping several threads
5542 independently and simultaneously.
5544 To enter non-stop mode, use this sequence of commands before you run
5545 or attach to your program:
5548 # Enable the async interface.
5551 # If using the CLI, pagination breaks non-stop.
5554 # Finally, turn it on!
5558 You can use these commands to manipulate the non-stop mode setting:
5561 @kindex set non-stop
5562 @item set non-stop on
5563 Enable selection of non-stop mode.
5564 @item set non-stop off
5565 Disable selection of non-stop mode.
5566 @kindex show non-stop
5568 Show the current non-stop enablement setting.
5571 Note these commands only reflect whether non-stop mode is enabled,
5572 not whether the currently-executing program is being run in non-stop mode.
5573 In particular, the @code{set non-stop} preference is only consulted when
5574 @value{GDBN} starts or connects to the target program, and it is generally
5575 not possible to switch modes once debugging has started. Furthermore,
5576 since not all targets support non-stop mode, even when you have enabled
5577 non-stop mode, @value{GDBN} may still fall back to all-stop operation by
5580 In non-stop mode, all execution commands apply only to the current thread
5581 by default. That is, @code{continue} only continues one thread.
5582 To continue all threads, issue @code{continue -a} or @code{c -a}.
5584 You can use @value{GDBN}'s background execution commands
5585 (@pxref{Background Execution}) to run some threads in the background
5586 while you continue to examine or step others from @value{GDBN}.
5587 The MI execution commands (@pxref{GDB/MI Program Execution}) are
5588 always executed asynchronously in non-stop mode.
5590 Suspending execution is done with the @code{interrupt} command when
5591 running in the background, or @kbd{Ctrl-c} during foreground execution.
5592 In all-stop mode, this stops the whole process;
5593 but in non-stop mode the interrupt applies only to the current thread.
5594 To stop the whole program, use @code{interrupt -a}.
5596 Other execution commands do not currently support the @code{-a} option.
5598 In non-stop mode, when a thread stops, @value{GDBN} doesn't automatically make
5599 that thread current, as it does in all-stop mode. This is because the
5600 thread stop notifications are asynchronous with respect to @value{GDBN}'s
5601 command interpreter, and it would be confusing if @value{GDBN} unexpectedly
5602 changed to a different thread just as you entered a command to operate on the
5603 previously current thread.
5605 @node Background Execution
5606 @subsection Background Execution
5608 @cindex foreground execution
5609 @cindex background execution
5610 @cindex asynchronous execution
5611 @cindex execution, foreground, background and asynchronous
5613 @value{GDBN}'s execution commands have two variants: the normal
5614 foreground (synchronous) behavior, and a background
5615 (asynchronous) behavior. In foreground execution, @value{GDBN} waits for
5616 the program to report that some thread has stopped before prompting for
5617 another command. In background execution, @value{GDBN} immediately gives
5618 a command prompt so that you can issue other commands while your program runs.
5620 You need to explicitly enable asynchronous mode before you can use
5621 background execution commands. You can use these commands to
5622 manipulate the asynchronous mode setting:
5625 @kindex set target-async
5626 @item set target-async on
5627 Enable asynchronous mode.
5628 @item set target-async off
5629 Disable asynchronous mode.
5630 @kindex show target-async
5631 @item show target-async
5632 Show the current target-async setting.
5635 If the target doesn't support async mode, @value{GDBN} issues an error
5636 message if you attempt to use the background execution commands.
5638 To specify background execution, add a @code{&} to the command. For example,
5639 the background form of the @code{continue} command is @code{continue&}, or
5640 just @code{c&}. The execution commands that accept background execution
5646 @xref{Starting, , Starting your Program}.
5650 @xref{Attach, , Debugging an Already-running Process}.
5654 @xref{Continuing and Stepping, step}.
5658 @xref{Continuing and Stepping, stepi}.
5662 @xref{Continuing and Stepping, next}.
5666 @xref{Continuing and Stepping, nexti}.
5670 @xref{Continuing and Stepping, continue}.
5674 @xref{Continuing and Stepping, finish}.
5678 @xref{Continuing and Stepping, until}.
5682 Background execution is especially useful in conjunction with non-stop
5683 mode for debugging programs with multiple threads; see @ref{Non-Stop Mode}.
5684 However, you can also use these commands in the normal all-stop mode with
5685 the restriction that you cannot issue another execution command until the
5686 previous one finishes. Examples of commands that are valid in all-stop
5687 mode while the program is running include @code{help} and @code{info break}.
5689 You can interrupt your program while it is running in the background by
5690 using the @code{interrupt} command.
5697 Suspend execution of the running program. In all-stop mode,
5698 @code{interrupt} stops the whole process, but in non-stop mode, it stops
5699 only the current thread. To stop the whole program in non-stop mode,
5700 use @code{interrupt -a}.
5703 @node Thread-Specific Breakpoints
5704 @subsection Thread-Specific Breakpoints
5706 When your program has multiple threads (@pxref{Threads,, Debugging
5707 Programs with Multiple Threads}), you can choose whether to set
5708 breakpoints on all threads, or on a particular thread.
5711 @cindex breakpoints and threads
5712 @cindex thread breakpoints
5713 @kindex break @dots{} thread @var{threadno}
5714 @item break @var{linespec} thread @var{threadno}
5715 @itemx break @var{linespec} thread @var{threadno} if @dots{}
5716 @var{linespec} specifies source lines; there are several ways of
5717 writing them (@pxref{Specify Location}), but the effect is always to
5718 specify some source line.
5720 Use the qualifier @samp{thread @var{threadno}} with a breakpoint command
5721 to specify that you only want @value{GDBN} to stop the program when a
5722 particular thread reaches this breakpoint. @var{threadno} is one of the
5723 numeric thread identifiers assigned by @value{GDBN}, shown in the first
5724 column of the @samp{info threads} display.
5726 If you do not specify @samp{thread @var{threadno}} when you set a
5727 breakpoint, the breakpoint applies to @emph{all} threads of your
5730 You can use the @code{thread} qualifier on conditional breakpoints as
5731 well; in this case, place @samp{thread @var{threadno}} before or
5732 after the breakpoint condition, like this:
5735 (@value{GDBP}) break frik.c:13 thread 28 if bartab > lim
5740 @node Interrupted System Calls
5741 @subsection Interrupted System Calls
5743 @cindex thread breakpoints and system calls
5744 @cindex system calls and thread breakpoints
5745 @cindex premature return from system calls
5746 There is an unfortunate side effect when using @value{GDBN} to debug
5747 multi-threaded programs. If one thread stops for a
5748 breakpoint, or for some other reason, and another thread is blocked in a
5749 system call, then the system call may return prematurely. This is a
5750 consequence of the interaction between multiple threads and the signals
5751 that @value{GDBN} uses to implement breakpoints and other events that
5754 To handle this problem, your program should check the return value of
5755 each system call and react appropriately. This is good programming
5758 For example, do not write code like this:
5764 The call to @code{sleep} will return early if a different thread stops
5765 at a breakpoint or for some other reason.
5767 Instead, write this:
5772 unslept = sleep (unslept);
5775 A system call is allowed to return early, so the system is still
5776 conforming to its specification. But @value{GDBN} does cause your
5777 multi-threaded program to behave differently than it would without
5780 Also, @value{GDBN} uses internal breakpoints in the thread library to
5781 monitor certain events such as thread creation and thread destruction.
5782 When such an event happens, a system call in another thread may return
5783 prematurely, even though your program does not appear to stop.
5786 @subsection Observer Mode
5788 If you want to build on non-stop mode and observe program behavior
5789 without any chance of disruption by @value{GDBN}, you can set
5790 variables to disable all of the debugger's attempts to modify state,
5791 whether by writing memory, inserting breakpoints, etc. These operate
5792 at a low level, intercepting operations from all commands.
5794 When all of these are set to @code{off}, then @value{GDBN} is said to
5795 be @dfn{observer mode}. As a convenience, the variable
5796 @code{observer} can be set to disable these, plus enable non-stop
5799 Note that @value{GDBN} will not prevent you from making nonsensical
5800 combinations of these settings. For instance, if you have enabled
5801 @code{may-insert-breakpoints} but disabled @code{may-write-memory},
5802 then breakpoints that work by writing trap instructions into the code
5803 stream will still not be able to be placed.
5808 @item set observer on
5809 @itemx set observer off
5810 When set to @code{on}, this disables all the permission variables
5811 below (except for @code{insert-fast-tracepoints}), plus enables
5812 non-stop debugging. Setting this to @code{off} switches back to
5813 normal debugging, though remaining in non-stop mode.
5816 Show whether observer mode is on or off.
5818 @kindex may-write-registers
5819 @item set may-write-registers on
5820 @itemx set may-write-registers off
5821 This controls whether @value{GDBN} will attempt to alter the values of
5822 registers, such as with assignment expressions in @code{print}, or the
5823 @code{jump} command. It defaults to @code{on}.
5825 @item show may-write-registers
5826 Show the current permission to write registers.
5828 @kindex may-write-memory
5829 @item set may-write-memory on
5830 @itemx set may-write-memory off
5831 This controls whether @value{GDBN} will attempt to alter the contents
5832 of memory, such as with assignment expressions in @code{print}. It
5833 defaults to @code{on}.
5835 @item show may-write-memory
5836 Show the current permission to write memory.
5838 @kindex may-insert-breakpoints
5839 @item set may-insert-breakpoints on
5840 @itemx set may-insert-breakpoints off
5841 This controls whether @value{GDBN} will attempt to insert breakpoints.
5842 This affects all breakpoints, including internal breakpoints defined
5843 by @value{GDBN}. It defaults to @code{on}.
5845 @item show may-insert-breakpoints
5846 Show the current permission to insert breakpoints.
5848 @kindex may-insert-tracepoints
5849 @item set may-insert-tracepoints on
5850 @itemx set may-insert-tracepoints off
5851 This controls whether @value{GDBN} will attempt to insert (regular)
5852 tracepoints at the beginning of a tracing experiment. It affects only
5853 non-fast tracepoints, fast tracepoints being under the control of
5854 @code{may-insert-fast-tracepoints}. It defaults to @code{on}.
5856 @item show may-insert-tracepoints
5857 Show the current permission to insert tracepoints.
5859 @kindex may-insert-fast-tracepoints
5860 @item set may-insert-fast-tracepoints on
5861 @itemx set may-insert-fast-tracepoints off
5862 This controls whether @value{GDBN} will attempt to insert fast
5863 tracepoints at the beginning of a tracing experiment. It affects only
5864 fast tracepoints, regular (non-fast) tracepoints being under the
5865 control of @code{may-insert-tracepoints}. It defaults to @code{on}.
5867 @item show may-insert-fast-tracepoints
5868 Show the current permission to insert fast tracepoints.
5870 @kindex may-interrupt
5871 @item set may-interrupt on
5872 @itemx set may-interrupt off
5873 This controls whether @value{GDBN} will attempt to interrupt or stop
5874 program execution. When this variable is @code{off}, the
5875 @code{interrupt} command will have no effect, nor will
5876 @kbd{Ctrl-c}. It defaults to @code{on}.
5878 @item show may-interrupt
5879 Show the current permission to interrupt or stop the program.
5883 @node Reverse Execution
5884 @chapter Running programs backward
5885 @cindex reverse execution
5886 @cindex running programs backward
5888 When you are debugging a program, it is not unusual to realize that
5889 you have gone too far, and some event of interest has already happened.
5890 If the target environment supports it, @value{GDBN} can allow you to
5891 ``rewind'' the program by running it backward.
5893 A target environment that supports reverse execution should be able
5894 to ``undo'' the changes in machine state that have taken place as the
5895 program was executing normally. Variables, registers etc.@: should
5896 revert to their previous values. Obviously this requires a great
5897 deal of sophistication on the part of the target environment; not
5898 all target environments can support reverse execution.
5900 When a program is executed in reverse, the instructions that
5901 have most recently been executed are ``un-executed'', in reverse
5902 order. The program counter runs backward, following the previous
5903 thread of execution in reverse. As each instruction is ``un-executed'',
5904 the values of memory and/or registers that were changed by that
5905 instruction are reverted to their previous states. After executing
5906 a piece of source code in reverse, all side effects of that code
5907 should be ``undone'', and all variables should be returned to their
5908 prior values@footnote{
5909 Note that some side effects are easier to undo than others. For instance,
5910 memory and registers are relatively easy, but device I/O is hard. Some
5911 targets may be able undo things like device I/O, and some may not.
5913 The contract between @value{GDBN} and the reverse executing target
5914 requires only that the target do something reasonable when
5915 @value{GDBN} tells it to execute backwards, and then report the
5916 results back to @value{GDBN}. Whatever the target reports back to
5917 @value{GDBN}, @value{GDBN} will report back to the user. @value{GDBN}
5918 assumes that the memory and registers that the target reports are in a
5919 consistant state, but @value{GDBN} accepts whatever it is given.
5922 If you are debugging in a target environment that supports
5923 reverse execution, @value{GDBN} provides the following commands.
5926 @kindex reverse-continue
5927 @kindex rc @r{(@code{reverse-continue})}
5928 @item reverse-continue @r{[}@var{ignore-count}@r{]}
5929 @itemx rc @r{[}@var{ignore-count}@r{]}
5930 Beginning at the point where your program last stopped, start executing
5931 in reverse. Reverse execution will stop for breakpoints and synchronous
5932 exceptions (signals), just like normal execution. Behavior of
5933 asynchronous signals depends on the target environment.
5935 @kindex reverse-step
5936 @kindex rs @r{(@code{step})}
5937 @item reverse-step @r{[}@var{count}@r{]}
5938 Run the program backward until control reaches the start of a
5939 different source line; then stop it, and return control to @value{GDBN}.
5941 Like the @code{step} command, @code{reverse-step} will only stop
5942 at the beginning of a source line. It ``un-executes'' the previously
5943 executed source line. If the previous source line included calls to
5944 debuggable functions, @code{reverse-step} will step (backward) into
5945 the called function, stopping at the beginning of the @emph{last}
5946 statement in the called function (typically a return statement).
5948 Also, as with the @code{step} command, if non-debuggable functions are
5949 called, @code{reverse-step} will run thru them backward without stopping.
5951 @kindex reverse-stepi
5952 @kindex rsi @r{(@code{reverse-stepi})}
5953 @item reverse-stepi @r{[}@var{count}@r{]}
5954 Reverse-execute one machine instruction. Note that the instruction
5955 to be reverse-executed is @emph{not} the one pointed to by the program
5956 counter, but the instruction executed prior to that one. For instance,
5957 if the last instruction was a jump, @code{reverse-stepi} will take you
5958 back from the destination of the jump to the jump instruction itself.
5960 @kindex reverse-next
5961 @kindex rn @r{(@code{reverse-next})}
5962 @item reverse-next @r{[}@var{count}@r{]}
5963 Run backward to the beginning of the previous line executed in
5964 the current (innermost) stack frame. If the line contains function
5965 calls, they will be ``un-executed'' without stopping. Starting from
5966 the first line of a function, @code{reverse-next} will take you back
5967 to the caller of that function, @emph{before} the function was called,
5968 just as the normal @code{next} command would take you from the last
5969 line of a function back to its return to its caller
5970 @footnote{Unless the code is too heavily optimized.}.
5972 @kindex reverse-nexti
5973 @kindex rni @r{(@code{reverse-nexti})}
5974 @item reverse-nexti @r{[}@var{count}@r{]}
5975 Like @code{nexti}, @code{reverse-nexti} executes a single instruction
5976 in reverse, except that called functions are ``un-executed'' atomically.
5977 That is, if the previously executed instruction was a return from
5978 another function, @code{reverse-nexti} will continue to execute
5979 in reverse until the call to that function (from the current stack
5982 @kindex reverse-finish
5983 @item reverse-finish
5984 Just as the @code{finish} command takes you to the point where the
5985 current function returns, @code{reverse-finish} takes you to the point
5986 where it was called. Instead of ending up at the end of the current
5987 function invocation, you end up at the beginning.
5989 @kindex set exec-direction
5990 @item set exec-direction
5991 Set the direction of target execution.
5992 @itemx set exec-direction reverse
5993 @cindex execute forward or backward in time
5994 @value{GDBN} will perform all execution commands in reverse, until the
5995 exec-direction mode is changed to ``forward''. Affected commands include
5996 @code{step, stepi, next, nexti, continue, and finish}. The @code{return}
5997 command cannot be used in reverse mode.
5998 @item set exec-direction forward
5999 @value{GDBN} will perform all execution commands in the normal fashion.
6000 This is the default.
6004 @node Process Record and Replay
6005 @chapter Recording Inferior's Execution and Replaying It
6006 @cindex process record and replay
6007 @cindex recording inferior's execution and replaying it
6009 On some platforms, @value{GDBN} provides a special @dfn{process record
6010 and replay} target that can record a log of the process execution, and
6011 replay it later with both forward and reverse execution commands.
6014 When this target is in use, if the execution log includes the record
6015 for the next instruction, @value{GDBN} will debug in @dfn{replay
6016 mode}. In the replay mode, the inferior does not really execute code
6017 instructions. Instead, all the events that normally happen during
6018 code execution are taken from the execution log. While code is not
6019 really executed in replay mode, the values of registers (including the
6020 program counter register) and the memory of the inferior are still
6021 changed as they normally would. Their contents are taken from the
6025 If the record for the next instruction is not in the execution log,
6026 @value{GDBN} will debug in @dfn{record mode}. In this mode, the
6027 inferior executes normally, and @value{GDBN} records the execution log
6030 The process record and replay target supports reverse execution
6031 (@pxref{Reverse Execution}), even if the platform on which the
6032 inferior runs does not. However, the reverse execution is limited in
6033 this case by the range of the instructions recorded in the execution
6034 log. In other words, reverse execution on platforms that don't
6035 support it directly can only be done in the replay mode.
6037 When debugging in the reverse direction, @value{GDBN} will work in
6038 replay mode as long as the execution log includes the record for the
6039 previous instruction; otherwise, it will work in record mode, if the
6040 platform supports reverse execution, or stop if not.
6042 For architecture environments that support process record and replay,
6043 @value{GDBN} provides the following commands:
6046 @kindex target record
6050 This command starts the process record and replay target. The process
6051 record and replay target can only debug a process that is already
6052 running. Therefore, you need first to start the process with the
6053 @kbd{run} or @kbd{start} commands, and then start the recording with
6054 the @kbd{target record} command.
6056 Both @code{record} and @code{rec} are aliases of @code{target record}.
6058 @cindex displaced stepping, and process record and replay
6059 Displaced stepping (@pxref{Maintenance Commands,, displaced stepping})
6060 will be automatically disabled when process record and replay target
6061 is started. That's because the process record and replay target
6062 doesn't support displaced stepping.
6064 @cindex non-stop mode, and process record and replay
6065 @cindex asynchronous execution, and process record and replay
6066 If the inferior is in the non-stop mode (@pxref{Non-Stop Mode}) or in
6067 the asynchronous execution mode (@pxref{Background Execution}), the
6068 process record and replay target cannot be started because it doesn't
6069 support these two modes.
6074 Stop the process record and replay target. When process record and
6075 replay target stops, the entire execution log will be deleted and the
6076 inferior will either be terminated, or will remain in its final state.
6078 When you stop the process record and replay target in record mode (at
6079 the end of the execution log), the inferior will be stopped at the
6080 next instruction that would have been recorded. In other words, if
6081 you record for a while and then stop recording, the inferior process
6082 will be left in the same state as if the recording never happened.
6084 On the other hand, if the process record and replay target is stopped
6085 while in replay mode (that is, not at the end of the execution log,
6086 but at some earlier point), the inferior process will become ``live''
6087 at that earlier state, and it will then be possible to continue the
6088 usual ``live'' debugging of the process from that state.
6090 When the inferior process exits, or @value{GDBN} detaches from it,
6091 process record and replay target will automatically stop itself.
6094 @item record save @var{filename}
6095 Save the execution log to a file @file{@var{filename}}.
6096 Default filename is @file{gdb_record.@var{process_id}}, where
6097 @var{process_id} is the process ID of the inferior.
6099 @kindex record restore
6100 @item record restore @var{filename}
6101 Restore the execution log from a file @file{@var{filename}}.
6102 File must have been created with @code{record save}.
6104 @kindex set record insn-number-max
6105 @item set record insn-number-max @var{limit}
6106 Set the limit of instructions to be recorded. Default value is 200000.
6108 If @var{limit} is a positive number, then @value{GDBN} will start
6109 deleting instructions from the log once the number of the record
6110 instructions becomes greater than @var{limit}. For every new recorded
6111 instruction, @value{GDBN} will delete the earliest recorded
6112 instruction to keep the number of recorded instructions at the limit.
6113 (Since deleting recorded instructions loses information, @value{GDBN}
6114 lets you control what happens when the limit is reached, by means of
6115 the @code{stop-at-limit} option, described below.)
6117 If @var{limit} is zero, @value{GDBN} will never delete recorded
6118 instructions from the execution log. The number of recorded
6119 instructions is unlimited in this case.
6121 @kindex show record insn-number-max
6122 @item show record insn-number-max
6123 Show the limit of instructions to be recorded.
6125 @kindex set record stop-at-limit
6126 @item set record stop-at-limit
6127 Control the behavior when the number of recorded instructions reaches
6128 the limit. If ON (the default), @value{GDBN} will stop when the limit
6129 is reached for the first time and ask you whether you want to stop the
6130 inferior or continue running it and recording the execution log. If
6131 you decide to continue recording, each new recorded instruction will
6132 cause the oldest one to be deleted.
6134 If this option is OFF, @value{GDBN} will automatically delete the
6135 oldest record to make room for each new one, without asking.
6137 @kindex show record stop-at-limit
6138 @item show record stop-at-limit
6139 Show the current setting of @code{stop-at-limit}.
6141 @kindex set record memory-query
6142 @item set record memory-query
6143 Control the behavior when @value{GDBN} is unable to record memory
6144 changes caused by an instruction. If ON, @value{GDBN} will query
6145 whether to stop the inferior in that case.
6147 If this option is OFF (the default), @value{GDBN} will automatically
6148 ignore the effect of such instructions on memory. Later, when
6149 @value{GDBN} replays this execution log, it will mark the log of this
6150 instruction as not accessible, and it will not affect the replay
6153 @kindex show record memory-query
6154 @item show record memory-query
6155 Show the current setting of @code{memory-query}.
6159 Show various statistics about the state of process record and its
6160 in-memory execution log buffer, including:
6164 Whether in record mode or replay mode.
6166 Lowest recorded instruction number (counting from when the current execution log started recording instructions).
6168 Highest recorded instruction number.
6170 Current instruction about to be replayed (if in replay mode).
6172 Number of instructions contained in the execution log.
6174 Maximum number of instructions that may be contained in the execution log.
6177 @kindex record delete
6180 When record target runs in replay mode (``in the past''), delete the
6181 subsequent execution log and begin to record a new execution log starting
6182 from the current address. This means you will abandon the previously
6183 recorded ``future'' and begin recording a new ``future''.
6188 @chapter Examining the Stack
6190 When your program has stopped, the first thing you need to know is where it
6191 stopped and how it got there.
6194 Each time your program performs a function call, information about the call
6196 That information includes the location of the call in your program,
6197 the arguments of the call,
6198 and the local variables of the function being called.
6199 The information is saved in a block of data called a @dfn{stack frame}.
6200 The stack frames are allocated in a region of memory called the @dfn{call
6203 When your program stops, the @value{GDBN} commands for examining the
6204 stack allow you to see all of this information.
6206 @cindex selected frame
6207 One of the stack frames is @dfn{selected} by @value{GDBN} and many
6208 @value{GDBN} commands refer implicitly to the selected frame. In
6209 particular, whenever you ask @value{GDBN} for the value of a variable in
6210 your program, the value is found in the selected frame. There are
6211 special @value{GDBN} commands to select whichever frame you are
6212 interested in. @xref{Selection, ,Selecting a Frame}.
6214 When your program stops, @value{GDBN} automatically selects the
6215 currently executing frame and describes it briefly, similar to the
6216 @code{frame} command (@pxref{Frame Info, ,Information about a Frame}).
6219 * Frames:: Stack frames
6220 * Backtrace:: Backtraces
6221 * Selection:: Selecting a frame
6222 * Frame Info:: Information on a frame
6227 @section Stack Frames
6229 @cindex frame, definition
6231 The call stack is divided up into contiguous pieces called @dfn{stack
6232 frames}, or @dfn{frames} for short; each frame is the data associated
6233 with one call to one function. The frame contains the arguments given
6234 to the function, the function's local variables, and the address at
6235 which the function is executing.
6237 @cindex initial frame
6238 @cindex outermost frame
6239 @cindex innermost frame
6240 When your program is started, the stack has only one frame, that of the
6241 function @code{main}. This is called the @dfn{initial} frame or the
6242 @dfn{outermost} frame. Each time a function is called, a new frame is
6243 made. Each time a function returns, the frame for that function invocation
6244 is eliminated. If a function is recursive, there can be many frames for
6245 the same function. The frame for the function in which execution is
6246 actually occurring is called the @dfn{innermost} frame. This is the most
6247 recently created of all the stack frames that still exist.
6249 @cindex frame pointer
6250 Inside your program, stack frames are identified by their addresses. A
6251 stack frame consists of many bytes, each of which has its own address; each
6252 kind of computer has a convention for choosing one byte whose
6253 address serves as the address of the frame. Usually this address is kept
6254 in a register called the @dfn{frame pointer register}
6255 (@pxref{Registers, $fp}) while execution is going on in that frame.
6257 @cindex frame number
6258 @value{GDBN} assigns numbers to all existing stack frames, starting with
6259 zero for the innermost frame, one for the frame that called it,
6260 and so on upward. These numbers do not really exist in your program;
6261 they are assigned by @value{GDBN} to give you a way of designating stack
6262 frames in @value{GDBN} commands.
6264 @c The -fomit-frame-pointer below perennially causes hbox overflow
6265 @c underflow problems.
6266 @cindex frameless execution
6267 Some compilers provide a way to compile functions so that they operate
6268 without stack frames. (For example, the @value{NGCC} option
6270 @samp{-fomit-frame-pointer}
6272 generates functions without a frame.)
6273 This is occasionally done with heavily used library functions to save
6274 the frame setup time. @value{GDBN} has limited facilities for dealing
6275 with these function invocations. If the innermost function invocation
6276 has no stack frame, @value{GDBN} nevertheless regards it as though
6277 it had a separate frame, which is numbered zero as usual, allowing
6278 correct tracing of the function call chain. However, @value{GDBN} has
6279 no provision for frameless functions elsewhere in the stack.
6282 @kindex frame@r{, command}
6283 @cindex current stack frame
6284 @item frame @var{args}
6285 The @code{frame} command allows you to move from one stack frame to another,
6286 and to print the stack frame you select. @var{args} may be either the
6287 address of the frame or the stack frame number. Without an argument,
6288 @code{frame} prints the current stack frame.
6290 @kindex select-frame
6291 @cindex selecting frame silently
6293 The @code{select-frame} command allows you to move from one stack frame
6294 to another without printing the frame. This is the silent version of
6302 @cindex call stack traces
6303 A backtrace is a summary of how your program got where it is. It shows one
6304 line per frame, for many frames, starting with the currently executing
6305 frame (frame zero), followed by its caller (frame one), and on up the
6310 @kindex bt @r{(@code{backtrace})}
6313 Print a backtrace of the entire stack: one line per frame for all
6314 frames in the stack.
6316 You can stop the backtrace at any time by typing the system interrupt
6317 character, normally @kbd{Ctrl-c}.
6319 @item backtrace @var{n}
6321 Similar, but print only the innermost @var{n} frames.
6323 @item backtrace -@var{n}
6325 Similar, but print only the outermost @var{n} frames.
6327 @item backtrace full
6329 @itemx bt full @var{n}
6330 @itemx bt full -@var{n}
6331 Print the values of the local variables also. @var{n} specifies the
6332 number of frames to print, as described above.
6337 The names @code{where} and @code{info stack} (abbreviated @code{info s})
6338 are additional aliases for @code{backtrace}.
6340 @cindex multiple threads, backtrace
6341 In a multi-threaded program, @value{GDBN} by default shows the
6342 backtrace only for the current thread. To display the backtrace for
6343 several or all of the threads, use the command @code{thread apply}
6344 (@pxref{Threads, thread apply}). For example, if you type @kbd{thread
6345 apply all backtrace}, @value{GDBN} will display the backtrace for all
6346 the threads; this is handy when you debug a core dump of a
6347 multi-threaded program.
6349 Each line in the backtrace shows the frame number and the function name.
6350 The program counter value is also shown---unless you use @code{set
6351 print address off}. The backtrace also shows the source file name and
6352 line number, as well as the arguments to the function. The program
6353 counter value is omitted if it is at the beginning of the code for that
6356 Here is an example of a backtrace. It was made with the command
6357 @samp{bt 3}, so it shows the innermost three frames.
6361 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
6363 #1 0x6e38 in expand_macro (sym=0x2b600, data=...) at macro.c:242
6364 #2 0x6840 in expand_token (obs=0x0, t=177664, td=0xf7fffb08)
6366 (More stack frames follow...)
6371 The display for frame zero does not begin with a program counter
6372 value, indicating that your program has stopped at the beginning of the
6373 code for line @code{993} of @code{builtin.c}.
6376 The value of parameter @code{data} in frame 1 has been replaced by
6377 @code{@dots{}}. By default, @value{GDBN} prints the value of a parameter
6378 only if it is a scalar (integer, pointer, enumeration, etc). See command
6379 @kbd{set print frame-arguments} in @ref{Print Settings} for more details
6380 on how to configure the way function parameter values are printed.
6382 @cindex optimized out, in backtrace
6383 @cindex function call arguments, optimized out
6384 If your program was compiled with optimizations, some compilers will
6385 optimize away arguments passed to functions if those arguments are
6386 never used after the call. Such optimizations generate code that
6387 passes arguments through registers, but doesn't store those arguments
6388 in the stack frame. @value{GDBN} has no way of displaying such
6389 arguments in stack frames other than the innermost one. Here's what
6390 such a backtrace might look like:
6394 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
6396 #1 0x6e38 in expand_macro (sym=<optimized out>) at macro.c:242
6397 #2 0x6840 in expand_token (obs=0x0, t=<optimized out>, td=0xf7fffb08)
6399 (More stack frames follow...)
6404 The values of arguments that were not saved in their stack frames are
6405 shown as @samp{<optimized out>}.
6407 If you need to display the values of such optimized-out arguments,
6408 either deduce that from other variables whose values depend on the one
6409 you are interested in, or recompile without optimizations.
6411 @cindex backtrace beyond @code{main} function
6412 @cindex program entry point
6413 @cindex startup code, and backtrace
6414 Most programs have a standard user entry point---a place where system
6415 libraries and startup code transition into user code. For C this is
6416 @code{main}@footnote{
6417 Note that embedded programs (the so-called ``free-standing''
6418 environment) are not required to have a @code{main} function as the
6419 entry point. They could even have multiple entry points.}.
6420 When @value{GDBN} finds the entry function in a backtrace
6421 it will terminate the backtrace, to avoid tracing into highly
6422 system-specific (and generally uninteresting) code.
6424 If you need to examine the startup code, or limit the number of levels
6425 in a backtrace, you can change this behavior:
6428 @item set backtrace past-main
6429 @itemx set backtrace past-main on
6430 @kindex set backtrace
6431 Backtraces will continue past the user entry point.
6433 @item set backtrace past-main off
6434 Backtraces will stop when they encounter the user entry point. This is the
6437 @item show backtrace past-main
6438 @kindex show backtrace
6439 Display the current user entry point backtrace policy.
6441 @item set backtrace past-entry
6442 @itemx set backtrace past-entry on
6443 Backtraces will continue past the internal entry point of an application.
6444 This entry point is encoded by the linker when the application is built,
6445 and is likely before the user entry point @code{main} (or equivalent) is called.
6447 @item set backtrace past-entry off
6448 Backtraces will stop when they encounter the internal entry point of an
6449 application. This is the default.
6451 @item show backtrace past-entry
6452 Display the current internal entry point backtrace policy.
6454 @item set backtrace limit @var{n}
6455 @itemx set backtrace limit 0
6456 @cindex backtrace limit
6457 Limit the backtrace to @var{n} levels. A value of zero means
6460 @item show backtrace limit
6461 Display the current limit on backtrace levels.
6465 @section Selecting a Frame
6467 Most commands for examining the stack and other data in your program work on
6468 whichever stack frame is selected at the moment. Here are the commands for
6469 selecting a stack frame; all of them finish by printing a brief description
6470 of the stack frame just selected.
6473 @kindex frame@r{, selecting}
6474 @kindex f @r{(@code{frame})}
6477 Select frame number @var{n}. Recall that frame zero is the innermost
6478 (currently executing) frame, frame one is the frame that called the
6479 innermost one, and so on. The highest-numbered frame is the one for
6482 @item frame @var{addr}
6484 Select the frame at address @var{addr}. This is useful mainly if the
6485 chaining of stack frames has been damaged by a bug, making it
6486 impossible for @value{GDBN} to assign numbers properly to all frames. In
6487 addition, this can be useful when your program has multiple stacks and
6488 switches between them.
6490 On the SPARC architecture, @code{frame} needs two addresses to
6491 select an arbitrary frame: a frame pointer and a stack pointer.
6493 On the @acronym{MIPS} and Alpha architecture, it needs two addresses: a stack
6494 pointer and a program counter.
6496 On the 29k architecture, it needs three addresses: a register stack
6497 pointer, a program counter, and a memory stack pointer.
6501 Move @var{n} frames up the stack. For positive numbers @var{n}, this
6502 advances toward the outermost frame, to higher frame numbers, to frames
6503 that have existed longer. @var{n} defaults to one.
6506 @kindex do @r{(@code{down})}
6508 Move @var{n} frames down the stack. For positive numbers @var{n}, this
6509 advances toward the innermost frame, to lower frame numbers, to frames
6510 that were created more recently. @var{n} defaults to one. You may
6511 abbreviate @code{down} as @code{do}.
6514 All of these commands end by printing two lines of output describing the
6515 frame. The first line shows the frame number, the function name, the
6516 arguments, and the source file and line number of execution in that
6517 frame. The second line shows the text of that source line.
6525 #1 0x22f0 in main (argc=1, argv=0xf7fffbf4, env=0xf7fffbfc)
6527 10 read_input_file (argv[i]);
6531 After such a printout, the @code{list} command with no arguments
6532 prints ten lines centered on the point of execution in the frame.
6533 You can also edit the program at the point of execution with your favorite
6534 editing program by typing @code{edit}.
6535 @xref{List, ,Printing Source Lines},
6539 @kindex down-silently
6541 @item up-silently @var{n}
6542 @itemx down-silently @var{n}
6543 These two commands are variants of @code{up} and @code{down},
6544 respectively; they differ in that they do their work silently, without
6545 causing display of the new frame. They are intended primarily for use
6546 in @value{GDBN} command scripts, where the output might be unnecessary and
6551 @section Information About a Frame
6553 There are several other commands to print information about the selected
6559 When used without any argument, this command does not change which
6560 frame is selected, but prints a brief description of the currently
6561 selected stack frame. It can be abbreviated @code{f}. With an
6562 argument, this command is used to select a stack frame.
6563 @xref{Selection, ,Selecting a Frame}.
6566 @kindex info f @r{(@code{info frame})}
6569 This command prints a verbose description of the selected stack frame,
6574 the address of the frame
6576 the address of the next frame down (called by this frame)
6578 the address of the next frame up (caller of this frame)
6580 the language in which the source code corresponding to this frame is written
6582 the address of the frame's arguments
6584 the address of the frame's local variables
6586 the program counter saved in it (the address of execution in the caller frame)
6588 which registers were saved in the frame
6591 @noindent The verbose description is useful when
6592 something has gone wrong that has made the stack format fail to fit
6593 the usual conventions.
6595 @item info frame @var{addr}
6596 @itemx info f @var{addr}
6597 Print a verbose description of the frame at address @var{addr}, without
6598 selecting that frame. The selected frame remains unchanged by this
6599 command. This requires the same kind of address (more than one for some
6600 architectures) that you specify in the @code{frame} command.
6601 @xref{Selection, ,Selecting a Frame}.
6605 Print the arguments of the selected frame, each on a separate line.
6609 Print the local variables of the selected frame, each on a separate
6610 line. These are all variables (declared either static or automatic)
6611 accessible at the point of execution of the selected frame.
6617 @chapter Examining Source Files
6619 @value{GDBN} can print parts of your program's source, since the debugging
6620 information recorded in the program tells @value{GDBN} what source files were
6621 used to build it. When your program stops, @value{GDBN} spontaneously prints
6622 the line where it stopped. Likewise, when you select a stack frame
6623 (@pxref{Selection, ,Selecting a Frame}), @value{GDBN} prints the line where
6624 execution in that frame has stopped. You can print other portions of
6625 source files by explicit command.
6627 If you use @value{GDBN} through its @sc{gnu} Emacs interface, you may
6628 prefer to use Emacs facilities to view source; see @ref{Emacs, ,Using
6629 @value{GDBN} under @sc{gnu} Emacs}.
6632 * List:: Printing source lines
6633 * Specify Location:: How to specify code locations
6634 * Edit:: Editing source files
6635 * Search:: Searching source files
6636 * Source Path:: Specifying source directories
6637 * Machine Code:: Source and machine code
6641 @section Printing Source Lines
6644 @kindex l @r{(@code{list})}
6645 To print lines from a source file, use the @code{list} command
6646 (abbreviated @code{l}). By default, ten lines are printed.
6647 There are several ways to specify what part of the file you want to
6648 print; see @ref{Specify Location}, for the full list.
6650 Here are the forms of the @code{list} command most commonly used:
6653 @item list @var{linenum}
6654 Print lines centered around line number @var{linenum} in the
6655 current source file.
6657 @item list @var{function}
6658 Print lines centered around the beginning of function
6662 Print more lines. If the last lines printed were printed with a
6663 @code{list} command, this prints lines following the last lines
6664 printed; however, if the last line printed was a solitary line printed
6665 as part of displaying a stack frame (@pxref{Stack, ,Examining the
6666 Stack}), this prints lines centered around that line.
6669 Print lines just before the lines last printed.
6672 @cindex @code{list}, how many lines to display
6673 By default, @value{GDBN} prints ten source lines with any of these forms of
6674 the @code{list} command. You can change this using @code{set listsize}:
6677 @kindex set listsize
6678 @item set listsize @var{count}
6679 Make the @code{list} command display @var{count} source lines (unless
6680 the @code{list} argument explicitly specifies some other number).
6682 @kindex show listsize
6684 Display the number of lines that @code{list} prints.
6687 Repeating a @code{list} command with @key{RET} discards the argument,
6688 so it is equivalent to typing just @code{list}. This is more useful
6689 than listing the same lines again. An exception is made for an
6690 argument of @samp{-}; that argument is preserved in repetition so that
6691 each repetition moves up in the source file.
6693 In general, the @code{list} command expects you to supply zero, one or two
6694 @dfn{linespecs}. Linespecs specify source lines; there are several ways
6695 of writing them (@pxref{Specify Location}), but the effect is always
6696 to specify some source line.
6698 Here is a complete description of the possible arguments for @code{list}:
6701 @item list @var{linespec}
6702 Print lines centered around the line specified by @var{linespec}.
6704 @item list @var{first},@var{last}
6705 Print lines from @var{first} to @var{last}. Both arguments are
6706 linespecs. When a @code{list} command has two linespecs, and the
6707 source file of the second linespec is omitted, this refers to
6708 the same source file as the first linespec.
6710 @item list ,@var{last}
6711 Print lines ending with @var{last}.
6713 @item list @var{first},
6714 Print lines starting with @var{first}.
6717 Print lines just after the lines last printed.
6720 Print lines just before the lines last printed.
6723 As described in the preceding table.
6726 @node Specify Location
6727 @section Specifying a Location
6728 @cindex specifying location
6731 Several @value{GDBN} commands accept arguments that specify a location
6732 of your program's code. Since @value{GDBN} is a source-level
6733 debugger, a location usually specifies some line in the source code;
6734 for that reason, locations are also known as @dfn{linespecs}.
6736 Here are all the different ways of specifying a code location that
6737 @value{GDBN} understands:
6741 Specifies the line number @var{linenum} of the current source file.
6744 @itemx +@var{offset}
6745 Specifies the line @var{offset} lines before or after the @dfn{current
6746 line}. For the @code{list} command, the current line is the last one
6747 printed; for the breakpoint commands, this is the line at which
6748 execution stopped in the currently selected @dfn{stack frame}
6749 (@pxref{Frames, ,Frames}, for a description of stack frames.) When
6750 used as the second of the two linespecs in a @code{list} command,
6751 this specifies the line @var{offset} lines up or down from the first
6754 @item @var{filename}:@var{linenum}
6755 Specifies the line @var{linenum} in the source file @var{filename}.
6756 If @var{filename} is a relative file name, then it will match any
6757 source file name with the same trailing components. For example, if
6758 @var{filename} is @samp{gcc/expr.c}, then it will match source file
6759 name of @file{/build/trunk/gcc/expr.c}, but not
6760 @file{/build/trunk/libcpp/expr.c} or @file{/build/trunk/gcc/x-expr.c}.
6762 @item @var{function}
6763 Specifies the line that begins the body of the function @var{function}.
6764 For example, in C, this is the line with the open brace.
6766 @item @var{function}:@var{label}
6767 Specifies the line where @var{label} appears in @var{function}.
6769 @item @var{filename}:@var{function}
6770 Specifies the line that begins the body of the function @var{function}
6771 in the file @var{filename}. You only need the file name with a
6772 function name to avoid ambiguity when there are identically named
6773 functions in different source files.
6776 Specifies the line at which the label named @var{label} appears.
6777 @value{GDBN} searches for the label in the function corresponding to
6778 the currently selected stack frame. If there is no current selected
6779 stack frame (for instance, if the inferior is not running), then
6780 @value{GDBN} will not search for a label.
6782 @item *@var{address}
6783 Specifies the program address @var{address}. For line-oriented
6784 commands, such as @code{list} and @code{edit}, this specifies a source
6785 line that contains @var{address}. For @code{break} and other
6786 breakpoint oriented commands, this can be used to set breakpoints in
6787 parts of your program which do not have debugging information or
6790 Here @var{address} may be any expression valid in the current working
6791 language (@pxref{Languages, working language}) that specifies a code
6792 address. In addition, as a convenience, @value{GDBN} extends the
6793 semantics of expressions used in locations to cover the situations
6794 that frequently happen during debugging. Here are the various forms
6798 @item @var{expression}
6799 Any expression valid in the current working language.
6801 @item @var{funcaddr}
6802 An address of a function or procedure derived from its name. In C,
6803 C@t{++}, Java, Objective-C, Fortran, minimal, and assembly, this is
6804 simply the function's name @var{function} (and actually a special case
6805 of a valid expression). In Pascal and Modula-2, this is
6806 @code{&@var{function}}. In Ada, this is @code{@var{function}'Address}
6807 (although the Pascal form also works).
6809 This form specifies the address of the function's first instruction,
6810 before the stack frame and arguments have been set up.
6812 @item '@var{filename}'::@var{funcaddr}
6813 Like @var{funcaddr} above, but also specifies the name of the source
6814 file explicitly. This is useful if the name of the function does not
6815 specify the function unambiguously, e.g., if there are several
6816 functions with identical names in different source files.
6819 @cindex breakpoint at static probe point
6820 @item -pstap|-probe-stap @r{[}@var{objfile}:@r{[}@var{provider}:@r{]}@r{]}@var{name}
6821 The @sc{gnu}/Linux tool @code{SystemTap} provides a way for
6822 applications to embed static probes. @xref{Static Probe Points}, for more
6823 information on finding and using static probes. This form of linespec
6824 specifies the location of such a static probe.
6826 If @var{objfile} is given, only probes coming from that shared library
6827 or executable matching @var{objfile} as a regular expression are considered.
6828 If @var{provider} is given, then only probes from that provider are considered.
6829 If several probes match the spec, @value{GDBN} will insert a breakpoint at
6830 each one of those probes.
6836 @section Editing Source Files
6837 @cindex editing source files
6840 @kindex e @r{(@code{edit})}
6841 To edit the lines in a source file, use the @code{edit} command.
6842 The editing program of your choice
6843 is invoked with the current line set to
6844 the active line in the program.
6845 Alternatively, there are several ways to specify what part of the file you
6846 want to print if you want to see other parts of the program:
6849 @item edit @var{location}
6850 Edit the source file specified by @code{location}. Editing starts at
6851 that @var{location}, e.g., at the specified source line of the
6852 specified file. @xref{Specify Location}, for all the possible forms
6853 of the @var{location} argument; here are the forms of the @code{edit}
6854 command most commonly used:
6857 @item edit @var{number}
6858 Edit the current source file with @var{number} as the active line number.
6860 @item edit @var{function}
6861 Edit the file containing @var{function} at the beginning of its definition.
6866 @subsection Choosing your Editor
6867 You can customize @value{GDBN} to use any editor you want
6869 The only restriction is that your editor (say @code{ex}), recognizes the
6870 following command-line syntax:
6872 ex +@var{number} file
6874 The optional numeric value +@var{number} specifies the number of the line in
6875 the file where to start editing.}.
6876 By default, it is @file{@value{EDITOR}}, but you can change this
6877 by setting the environment variable @code{EDITOR} before using
6878 @value{GDBN}. For example, to configure @value{GDBN} to use the
6879 @code{vi} editor, you could use these commands with the @code{sh} shell:
6885 or in the @code{csh} shell,
6887 setenv EDITOR /usr/bin/vi
6892 @section Searching Source Files
6893 @cindex searching source files
6895 There are two commands for searching through the current source file for a
6900 @kindex forward-search
6901 @item forward-search @var{regexp}
6902 @itemx search @var{regexp}
6903 The command @samp{forward-search @var{regexp}} checks each line,
6904 starting with the one following the last line listed, for a match for
6905 @var{regexp}. It lists the line that is found. You can use the
6906 synonym @samp{search @var{regexp}} or abbreviate the command name as
6909 @kindex reverse-search
6910 @item reverse-search @var{regexp}
6911 The command @samp{reverse-search @var{regexp}} checks each line, starting
6912 with the one before the last line listed and going backward, for a match
6913 for @var{regexp}. It lists the line that is found. You can abbreviate
6914 this command as @code{rev}.
6918 @section Specifying Source Directories
6921 @cindex directories for source files
6922 Executable programs sometimes do not record the directories of the source
6923 files from which they were compiled, just the names. Even when they do,
6924 the directories could be moved between the compilation and your debugging
6925 session. @value{GDBN} has a list of directories to search for source files;
6926 this is called the @dfn{source path}. Each time @value{GDBN} wants a source file,
6927 it tries all the directories in the list, in the order they are present
6928 in the list, until it finds a file with the desired name.
6930 For example, suppose an executable references the file
6931 @file{/usr/src/foo-1.0/lib/foo.c}, and our source path is
6932 @file{/mnt/cross}. The file is first looked up literally; if this
6933 fails, @file{/mnt/cross/usr/src/foo-1.0/lib/foo.c} is tried; if this
6934 fails, @file{/mnt/cross/foo.c} is opened; if this fails, an error
6935 message is printed. @value{GDBN} does not look up the parts of the
6936 source file name, such as @file{/mnt/cross/src/foo-1.0/lib/foo.c}.
6937 Likewise, the subdirectories of the source path are not searched: if
6938 the source path is @file{/mnt/cross}, and the binary refers to
6939 @file{foo.c}, @value{GDBN} would not find it under
6940 @file{/mnt/cross/usr/src/foo-1.0/lib}.
6942 Plain file names, relative file names with leading directories, file
6943 names containing dots, etc.@: are all treated as described above; for
6944 instance, if the source path is @file{/mnt/cross}, and the source file
6945 is recorded as @file{../lib/foo.c}, @value{GDBN} would first try
6946 @file{../lib/foo.c}, then @file{/mnt/cross/../lib/foo.c}, and after
6947 that---@file{/mnt/cross/foo.c}.
6949 Note that the executable search path is @emph{not} used to locate the
6952 Whenever you reset or rearrange the source path, @value{GDBN} clears out
6953 any information it has cached about where source files are found and where
6954 each line is in the file.
6958 When you start @value{GDBN}, its source path includes only @samp{cdir}
6959 and @samp{cwd}, in that order.
6960 To add other directories, use the @code{directory} command.
6962 The search path is used to find both program source files and @value{GDBN}
6963 script files (read using the @samp{-command} option and @samp{source} command).
6965 In addition to the source path, @value{GDBN} provides a set of commands
6966 that manage a list of source path substitution rules. A @dfn{substitution
6967 rule} specifies how to rewrite source directories stored in the program's
6968 debug information in case the sources were moved to a different
6969 directory between compilation and debugging. A rule is made of
6970 two strings, the first specifying what needs to be rewritten in
6971 the path, and the second specifying how it should be rewritten.
6972 In @ref{set substitute-path}, we name these two parts @var{from} and
6973 @var{to} respectively. @value{GDBN} does a simple string replacement
6974 of @var{from} with @var{to} at the start of the directory part of the
6975 source file name, and uses that result instead of the original file
6976 name to look up the sources.
6978 Using the previous example, suppose the @file{foo-1.0} tree has been
6979 moved from @file{/usr/src} to @file{/mnt/cross}, then you can tell
6980 @value{GDBN} to replace @file{/usr/src} in all source path names with
6981 @file{/mnt/cross}. The first lookup will then be
6982 @file{/mnt/cross/foo-1.0/lib/foo.c} in place of the original location
6983 of @file{/usr/src/foo-1.0/lib/foo.c}. To define a source path
6984 substitution rule, use the @code{set substitute-path} command
6985 (@pxref{set substitute-path}).
6987 To avoid unexpected substitution results, a rule is applied only if the
6988 @var{from} part of the directory name ends at a directory separator.
6989 For instance, a rule substituting @file{/usr/source} into
6990 @file{/mnt/cross} will be applied to @file{/usr/source/foo-1.0} but
6991 not to @file{/usr/sourceware/foo-2.0}. And because the substitution
6992 is applied only at the beginning of the directory name, this rule will
6993 not be applied to @file{/root/usr/source/baz.c} either.
6995 In many cases, you can achieve the same result using the @code{directory}
6996 command. However, @code{set substitute-path} can be more efficient in
6997 the case where the sources are organized in a complex tree with multiple
6998 subdirectories. With the @code{directory} command, you need to add each
6999 subdirectory of your project. If you moved the entire tree while
7000 preserving its internal organization, then @code{set substitute-path}
7001 allows you to direct the debugger to all the sources with one single
7004 @code{set substitute-path} is also more than just a shortcut command.
7005 The source path is only used if the file at the original location no
7006 longer exists. On the other hand, @code{set substitute-path} modifies
7007 the debugger behavior to look at the rewritten location instead. So, if
7008 for any reason a source file that is not relevant to your executable is
7009 located at the original location, a substitution rule is the only
7010 method available to point @value{GDBN} at the new location.
7012 @cindex @samp{--with-relocated-sources}
7013 @cindex default source path substitution
7014 You can configure a default source path substitution rule by
7015 configuring @value{GDBN} with the
7016 @samp{--with-relocated-sources=@var{dir}} option. The @var{dir}
7017 should be the name of a directory under @value{GDBN}'s configured
7018 prefix (set with @samp{--prefix} or @samp{--exec-prefix}), and
7019 directory names in debug information under @var{dir} will be adjusted
7020 automatically if the installed @value{GDBN} is moved to a new
7021 location. This is useful if @value{GDBN}, libraries or executables
7022 with debug information and corresponding source code are being moved
7026 @item directory @var{dirname} @dots{}
7027 @item dir @var{dirname} @dots{}
7028 Add directory @var{dirname} to the front of the source path. Several
7029 directory names may be given to this command, separated by @samp{:}
7030 (@samp{;} on MS-DOS and MS-Windows, where @samp{:} usually appears as
7031 part of absolute file names) or
7032 whitespace. You may specify a directory that is already in the source
7033 path; this moves it forward, so @value{GDBN} searches it sooner.
7037 @vindex $cdir@r{, convenience variable}
7038 @vindex $cwd@r{, convenience variable}
7039 @cindex compilation directory
7040 @cindex current directory
7041 @cindex working directory
7042 @cindex directory, current
7043 @cindex directory, compilation
7044 You can use the string @samp{$cdir} to refer to the compilation
7045 directory (if one is recorded), and @samp{$cwd} to refer to the current
7046 working directory. @samp{$cwd} is not the same as @samp{.}---the former
7047 tracks the current working directory as it changes during your @value{GDBN}
7048 session, while the latter is immediately expanded to the current
7049 directory at the time you add an entry to the source path.
7052 Reset the source path to its default value (@samp{$cdir:$cwd} on Unix systems). This requires confirmation.
7054 @c RET-repeat for @code{directory} is explicitly disabled, but since
7055 @c repeating it would be a no-op we do not say that. (thanks to RMS)
7057 @item set directories @var{path-list}
7058 @kindex set directories
7059 Set the source path to @var{path-list}.
7060 @samp{$cdir:$cwd} are added if missing.
7062 @item show directories
7063 @kindex show directories
7064 Print the source path: show which directories it contains.
7066 @anchor{set substitute-path}
7067 @item set substitute-path @var{from} @var{to}
7068 @kindex set substitute-path
7069 Define a source path substitution rule, and add it at the end of the
7070 current list of existing substitution rules. If a rule with the same
7071 @var{from} was already defined, then the old rule is also deleted.
7073 For example, if the file @file{/foo/bar/baz.c} was moved to
7074 @file{/mnt/cross/baz.c}, then the command
7077 (@value{GDBP}) set substitute-path /usr/src /mnt/cross
7081 will tell @value{GDBN} to replace @samp{/usr/src} with
7082 @samp{/mnt/cross}, which will allow @value{GDBN} to find the file
7083 @file{baz.c} even though it was moved.
7085 In the case when more than one substitution rule have been defined,
7086 the rules are evaluated one by one in the order where they have been
7087 defined. The first one matching, if any, is selected to perform
7090 For instance, if we had entered the following commands:
7093 (@value{GDBP}) set substitute-path /usr/src/include /mnt/include
7094 (@value{GDBP}) set substitute-path /usr/src /mnt/src
7098 @value{GDBN} would then rewrite @file{/usr/src/include/defs.h} into
7099 @file{/mnt/include/defs.h} by using the first rule. However, it would
7100 use the second rule to rewrite @file{/usr/src/lib/foo.c} into
7101 @file{/mnt/src/lib/foo.c}.
7104 @item unset substitute-path [path]
7105 @kindex unset substitute-path
7106 If a path is specified, search the current list of substitution rules
7107 for a rule that would rewrite that path. Delete that rule if found.
7108 A warning is emitted by the debugger if no rule could be found.
7110 If no path is specified, then all substitution rules are deleted.
7112 @item show substitute-path [path]
7113 @kindex show substitute-path
7114 If a path is specified, then print the source path substitution rule
7115 which would rewrite that path, if any.
7117 If no path is specified, then print all existing source path substitution
7122 If your source path is cluttered with directories that are no longer of
7123 interest, @value{GDBN} may sometimes cause confusion by finding the wrong
7124 versions of source. You can correct the situation as follows:
7128 Use @code{directory} with no argument to reset the source path to its default value.
7131 Use @code{directory} with suitable arguments to reinstall the
7132 directories you want in the source path. You can add all the
7133 directories in one command.
7137 @section Source and Machine Code
7138 @cindex source line and its code address
7140 You can use the command @code{info line} to map source lines to program
7141 addresses (and vice versa), and the command @code{disassemble} to display
7142 a range of addresses as machine instructions. You can use the command
7143 @code{set disassemble-next-line} to set whether to disassemble next
7144 source line when execution stops. When run under @sc{gnu} Emacs
7145 mode, the @code{info line} command causes the arrow to point to the
7146 line specified. Also, @code{info line} prints addresses in symbolic form as
7151 @item info line @var{linespec}
7152 Print the starting and ending addresses of the compiled code for
7153 source line @var{linespec}. You can specify source lines in any of
7154 the ways documented in @ref{Specify Location}.
7157 For example, we can use @code{info line} to discover the location of
7158 the object code for the first line of function
7159 @code{m4_changequote}:
7161 @c FIXME: I think this example should also show the addresses in
7162 @c symbolic form, as they usually would be displayed.
7164 (@value{GDBP}) info line m4_changequote
7165 Line 895 of "builtin.c" starts at pc 0x634c and ends at 0x6350.
7169 @cindex code address and its source line
7170 We can also inquire (using @code{*@var{addr}} as the form for
7171 @var{linespec}) what source line covers a particular address:
7173 (@value{GDBP}) info line *0x63ff
7174 Line 926 of "builtin.c" starts at pc 0x63e4 and ends at 0x6404.
7177 @cindex @code{$_} and @code{info line}
7178 @cindex @code{x} command, default address
7179 @kindex x@r{(examine), and} info line
7180 After @code{info line}, the default address for the @code{x} command
7181 is changed to the starting address of the line, so that @samp{x/i} is
7182 sufficient to begin examining the machine code (@pxref{Memory,
7183 ,Examining Memory}). Also, this address is saved as the value of the
7184 convenience variable @code{$_} (@pxref{Convenience Vars, ,Convenience
7189 @cindex assembly instructions
7190 @cindex instructions, assembly
7191 @cindex machine instructions
7192 @cindex listing machine instructions
7194 @itemx disassemble /m
7195 @itemx disassemble /r
7196 This specialized command dumps a range of memory as machine
7197 instructions. It can also print mixed source+disassembly by specifying
7198 the @code{/m} modifier and print the raw instructions in hex as well as
7199 in symbolic form by specifying the @code{/r}.
7200 The default memory range is the function surrounding the
7201 program counter of the selected frame. A single argument to this
7202 command is a program counter value; @value{GDBN} dumps the function
7203 surrounding this value. When two arguments are given, they should
7204 be separated by a comma, possibly surrounded by whitespace. The
7205 arguments specify a range of addresses to dump, in one of two forms:
7208 @item @var{start},@var{end}
7209 the addresses from @var{start} (inclusive) to @var{end} (exclusive)
7210 @item @var{start},+@var{length}
7211 the addresses from @var{start} (inclusive) to
7212 @code{@var{start}+@var{length}} (exclusive).
7216 When 2 arguments are specified, the name of the function is also
7217 printed (since there could be several functions in the given range).
7219 The argument(s) can be any expression yielding a numeric value, such as
7220 @samp{0x32c4}, @samp{&main+10} or @samp{$pc - 8}.
7222 If the range of memory being disassembled contains current program counter,
7223 the instruction at that location is shown with a @code{=>} marker.
7226 The following example shows the disassembly of a range of addresses of
7227 HP PA-RISC 2.0 code:
7230 (@value{GDBP}) disas 0x32c4, 0x32e4
7231 Dump of assembler code from 0x32c4 to 0x32e4:
7232 0x32c4 <main+204>: addil 0,dp
7233 0x32c8 <main+208>: ldw 0x22c(sr0,r1),r26
7234 0x32cc <main+212>: ldil 0x3000,r31
7235 0x32d0 <main+216>: ble 0x3f8(sr4,r31)
7236 0x32d4 <main+220>: ldo 0(r31),rp
7237 0x32d8 <main+224>: addil -0x800,dp
7238 0x32dc <main+228>: ldo 0x588(r1),r26
7239 0x32e0 <main+232>: ldil 0x3000,r31
7240 End of assembler dump.
7243 Here is an example showing mixed source+assembly for Intel x86, when the
7244 program is stopped just after function prologue:
7247 (@value{GDBP}) disas /m main
7248 Dump of assembler code for function main:
7250 0x08048330 <+0>: push %ebp
7251 0x08048331 <+1>: mov %esp,%ebp
7252 0x08048333 <+3>: sub $0x8,%esp
7253 0x08048336 <+6>: and $0xfffffff0,%esp
7254 0x08048339 <+9>: sub $0x10,%esp
7256 6 printf ("Hello.\n");
7257 => 0x0804833c <+12>: movl $0x8048440,(%esp)
7258 0x08048343 <+19>: call 0x8048284 <puts@@plt>
7262 0x08048348 <+24>: mov $0x0,%eax
7263 0x0804834d <+29>: leave
7264 0x0804834e <+30>: ret
7266 End of assembler dump.
7269 Here is another example showing raw instructions in hex for AMD x86-64,
7272 (gdb) disas /r 0x400281,+10
7273 Dump of assembler code from 0x400281 to 0x40028b:
7274 0x0000000000400281: 38 36 cmp %dh,(%rsi)
7275 0x0000000000400283: 2d 36 34 2e 73 sub $0x732e3436,%eax
7276 0x0000000000400288: 6f outsl %ds:(%rsi),(%dx)
7277 0x0000000000400289: 2e 32 00 xor %cs:(%rax),%al
7278 End of assembler dump.
7281 Some architectures have more than one commonly-used set of instruction
7282 mnemonics or other syntax.
7284 For programs that were dynamically linked and use shared libraries,
7285 instructions that call functions or branch to locations in the shared
7286 libraries might show a seemingly bogus location---it's actually a
7287 location of the relocation table. On some architectures, @value{GDBN}
7288 might be able to resolve these to actual function names.
7291 @kindex set disassembly-flavor
7292 @cindex Intel disassembly flavor
7293 @cindex AT&T disassembly flavor
7294 @item set disassembly-flavor @var{instruction-set}
7295 Select the instruction set to use when disassembling the
7296 program via the @code{disassemble} or @code{x/i} commands.
7298 Currently this command is only defined for the Intel x86 family. You
7299 can set @var{instruction-set} to either @code{intel} or @code{att}.
7300 The default is @code{att}, the AT&T flavor used by default by Unix
7301 assemblers for x86-based targets.
7303 @kindex show disassembly-flavor
7304 @item show disassembly-flavor
7305 Show the current setting of the disassembly flavor.
7309 @kindex set disassemble-next-line
7310 @kindex show disassemble-next-line
7311 @item set disassemble-next-line
7312 @itemx show disassemble-next-line
7313 Control whether or not @value{GDBN} will disassemble the next source
7314 line or instruction when execution stops. If ON, @value{GDBN} will
7315 display disassembly of the next source line when execution of the
7316 program being debugged stops. This is @emph{in addition} to
7317 displaying the source line itself, which @value{GDBN} always does if
7318 possible. If the next source line cannot be displayed for some reason
7319 (e.g., if @value{GDBN} cannot find the source file, or there's no line
7320 info in the debug info), @value{GDBN} will display disassembly of the
7321 next @emph{instruction} instead of showing the next source line. If
7322 AUTO, @value{GDBN} will display disassembly of next instruction only
7323 if the source line cannot be displayed. This setting causes
7324 @value{GDBN} to display some feedback when you step through a function
7325 with no line info or whose source file is unavailable. The default is
7326 OFF, which means never display the disassembly of the next line or
7332 @chapter Examining Data
7334 @cindex printing data
7335 @cindex examining data
7338 @c "inspect" is not quite a synonym if you are using Epoch, which we do not
7339 @c document because it is nonstandard... Under Epoch it displays in a
7340 @c different window or something like that.
7341 The usual way to examine data in your program is with the @code{print}
7342 command (abbreviated @code{p}), or its synonym @code{inspect}. It
7343 evaluates and prints the value of an expression of the language your
7344 program is written in (@pxref{Languages, ,Using @value{GDBN} with
7345 Different Languages}). It may also print the expression using a
7346 Python-based pretty-printer (@pxref{Pretty Printing}).
7349 @item print @var{expr}
7350 @itemx print /@var{f} @var{expr}
7351 @var{expr} is an expression (in the source language). By default the
7352 value of @var{expr} is printed in a format appropriate to its data type;
7353 you can choose a different format by specifying @samp{/@var{f}}, where
7354 @var{f} is a letter specifying the format; see @ref{Output Formats,,Output
7358 @itemx print /@var{f}
7359 @cindex reprint the last value
7360 If you omit @var{expr}, @value{GDBN} displays the last value again (from the
7361 @dfn{value history}; @pxref{Value History, ,Value History}). This allows you to
7362 conveniently inspect the same value in an alternative format.
7365 A more low-level way of examining data is with the @code{x} command.
7366 It examines data in memory at a specified address and prints it in a
7367 specified format. @xref{Memory, ,Examining Memory}.
7369 If you are interested in information about types, or about how the
7370 fields of a struct or a class are declared, use the @code{ptype @var{exp}}
7371 command rather than @code{print}. @xref{Symbols, ,Examining the Symbol
7374 @cindex exploring hierarchical data structures
7376 Another way of examining values of expressions and type information is
7377 through the Python extension command @code{explore} (available only if
7378 the @value{GDBN} build is configured with @code{--with-python}). It
7379 offers an interactive way to start at the highest level (or, the most
7380 abstract level) of the data type of an expression (or, the data type
7381 itself) and explore all the way down to leaf scalar values/fields
7382 embedded in the higher level data types.
7385 @item explore @var{arg}
7386 @var{arg} is either an expression (in the source language), or a type
7387 visible in the current context of the program being debugged.
7390 The working of the @code{explore} command can be illustrated with an
7391 example. If a data type @code{struct ComplexStruct} is defined in your
7401 struct ComplexStruct
7403 struct SimpleStruct *ss_p;
7409 followed by variable declarations as
7412 struct SimpleStruct ss = @{ 10, 1.11 @};
7413 struct ComplexStruct cs = @{ &ss, @{ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9 @} @};
7417 then, the value of the variable @code{cs} can be explored using the
7418 @code{explore} command as follows.
7422 The value of `cs' is a struct/class of type `struct ComplexStruct' with
7423 the following fields:
7425 ss_p = <Enter 0 to explore this field of type `struct SimpleStruct *'>
7426 arr = <Enter 1 to explore this field of type `int [10]'>
7428 Enter the field number of choice:
7432 Since the fields of @code{cs} are not scalar values, you are being
7433 prompted to chose the field you want to explore. Let's say you choose
7434 the field @code{ss_p} by entering @code{0}. Then, since this field is a
7435 pointer, you will be asked if it is pointing to a single value. From
7436 the declaration of @code{cs} above, it is indeed pointing to a single
7437 value, hence you enter @code{y}. If you enter @code{n}, then you will
7438 be asked if it were pointing to an array of values, in which case this
7439 field will be explored as if it were an array.
7442 `cs.ss_p' is a pointer to a value of type `struct SimpleStruct'
7443 Continue exploring it as a pointer to a single value [y/n]: y
7444 The value of `*(cs.ss_p)' is a struct/class of type `struct
7445 SimpleStruct' with the following fields:
7447 i = 10 .. (Value of type `int')
7448 d = 1.1100000000000001 .. (Value of type `double')
7450 Press enter to return to parent value:
7454 If the field @code{arr} of @code{cs} was chosen for exploration by
7455 entering @code{1} earlier, then since it is as array, you will be
7456 prompted to enter the index of the element in the array that you want
7460 `cs.arr' is an array of `int'.
7461 Enter the index of the element you want to explore in `cs.arr': 5
7463 `(cs.arr)[5]' is a scalar value of type `int'.
7467 Press enter to return to parent value:
7470 In general, at any stage of exploration, you can go deeper towards the
7471 leaf values by responding to the prompts appropriately, or hit the
7472 return key to return to the enclosing data structure (the @i{higher}
7473 level data structure).
7475 Similar to exploring values, you can use the @code{explore} command to
7476 explore types. Instead of specifying a value (which is typically a
7477 variable name or an expression valid in the current context of the
7478 program being debugged), you specify a type name. If you consider the
7479 same example as above, your can explore the type
7480 @code{struct ComplexStruct} by passing the argument
7481 @code{struct ComplexStruct} to the @code{explore} command.
7484 (gdb) explore struct ComplexStruct
7488 By responding to the prompts appropriately in the subsequent interactive
7489 session, you can explore the type @code{struct ComplexStruct} in a
7490 manner similar to how the value @code{cs} was explored in the above
7493 The @code{explore} command also has two sub-commands,
7494 @code{explore value} and @code{explore type}. The former sub-command is
7495 a way to explicitly specify that value exploration of the argument is
7496 being invoked, while the latter is a way to explicitly specify that type
7497 exploration of the argument is being invoked.
7500 @item explore value @var{expr}
7501 @cindex explore value
7502 This sub-command of @code{explore} explores the value of the
7503 expression @var{expr} (if @var{expr} is an expression valid in the
7504 current context of the program being debugged). The behavior of this
7505 command is identical to that of the behavior of the @code{explore}
7506 command being passed the argument @var{expr}.
7508 @item explore type @var{arg}
7509 @cindex explore type
7510 This sub-command of @code{explore} explores the type of @var{arg} (if
7511 @var{arg} is a type visible in the current context of program being
7512 debugged), or the type of the value/expression @var{arg} (if @var{arg}
7513 is an expression valid in the current context of the program being
7514 debugged). If @var{arg} is a type, then the behavior of this command is
7515 identical to that of the @code{explore} command being passed the
7516 argument @var{arg}. If @var{arg} is an expression, then the behavior of
7517 this command will be identical to that of the @code{explore} command
7518 being passed the type of @var{arg} as the argument.
7522 * Expressions:: Expressions
7523 * Ambiguous Expressions:: Ambiguous Expressions
7524 * Variables:: Program variables
7525 * Arrays:: Artificial arrays
7526 * Output Formats:: Output formats
7527 * Memory:: Examining memory
7528 * Auto Display:: Automatic display
7529 * Print Settings:: Print settings
7530 * Pretty Printing:: Python pretty printing
7531 * Value History:: Value history
7532 * Convenience Vars:: Convenience variables
7533 * Registers:: Registers
7534 * Floating Point Hardware:: Floating point hardware
7535 * Vector Unit:: Vector Unit
7536 * OS Information:: Auxiliary data provided by operating system
7537 * Memory Region Attributes:: Memory region attributes
7538 * Dump/Restore Files:: Copy between memory and a file
7539 * Core File Generation:: Cause a program dump its core
7540 * Character Sets:: Debugging programs that use a different
7541 character set than GDB does
7542 * Caching Remote Data:: Data caching for remote targets
7543 * Searching Memory:: Searching memory for a sequence of bytes
7547 @section Expressions
7550 @code{print} and many other @value{GDBN} commands accept an expression and
7551 compute its value. Any kind of constant, variable or operator defined
7552 by the programming language you are using is valid in an expression in
7553 @value{GDBN}. This includes conditional expressions, function calls,
7554 casts, and string constants. It also includes preprocessor macros, if
7555 you compiled your program to include this information; see
7558 @cindex arrays in expressions
7559 @value{GDBN} supports array constants in expressions input by
7560 the user. The syntax is @{@var{element}, @var{element}@dots{}@}. For example,
7561 you can use the command @code{print @{1, 2, 3@}} to create an array
7562 of three integers. If you pass an array to a function or assign it
7563 to a program variable, @value{GDBN} copies the array to memory that
7564 is @code{malloc}ed in the target program.
7566 Because C is so widespread, most of the expressions shown in examples in
7567 this manual are in C. @xref{Languages, , Using @value{GDBN} with Different
7568 Languages}, for information on how to use expressions in other
7571 In this section, we discuss operators that you can use in @value{GDBN}
7572 expressions regardless of your programming language.
7574 @cindex casts, in expressions
7575 Casts are supported in all languages, not just in C, because it is so
7576 useful to cast a number into a pointer in order to examine a structure
7577 at that address in memory.
7578 @c FIXME: casts supported---Mod2 true?
7580 @value{GDBN} supports these operators, in addition to those common
7581 to programming languages:
7585 @samp{@@} is a binary operator for treating parts of memory as arrays.
7586 @xref{Arrays, ,Artificial Arrays}, for more information.
7589 @samp{::} allows you to specify a variable in terms of the file or
7590 function where it is defined. @xref{Variables, ,Program Variables}.
7592 @cindex @{@var{type}@}
7593 @cindex type casting memory
7594 @cindex memory, viewing as typed object
7595 @cindex casts, to view memory
7596 @item @{@var{type}@} @var{addr}
7597 Refers to an object of type @var{type} stored at address @var{addr} in
7598 memory. @var{addr} may be any expression whose value is an integer or
7599 pointer (but parentheses are required around binary operators, just as in
7600 a cast). This construct is allowed regardless of what kind of data is
7601 normally supposed to reside at @var{addr}.
7604 @node Ambiguous Expressions
7605 @section Ambiguous Expressions
7606 @cindex ambiguous expressions
7608 Expressions can sometimes contain some ambiguous elements. For instance,
7609 some programming languages (notably Ada, C@t{++} and Objective-C) permit
7610 a single function name to be defined several times, for application in
7611 different contexts. This is called @dfn{overloading}. Another example
7612 involving Ada is generics. A @dfn{generic package} is similar to C@t{++}
7613 templates and is typically instantiated several times, resulting in
7614 the same function name being defined in different contexts.
7616 In some cases and depending on the language, it is possible to adjust
7617 the expression to remove the ambiguity. For instance in C@t{++}, you
7618 can specify the signature of the function you want to break on, as in
7619 @kbd{break @var{function}(@var{types})}. In Ada, using the fully
7620 qualified name of your function often makes the expression unambiguous
7623 When an ambiguity that needs to be resolved is detected, the debugger
7624 has the capability to display a menu of numbered choices for each
7625 possibility, and then waits for the selection with the prompt @samp{>}.
7626 The first option is always @samp{[0] cancel}, and typing @kbd{0 @key{RET}}
7627 aborts the current command. If the command in which the expression was
7628 used allows more than one choice to be selected, the next option in the
7629 menu is @samp{[1] all}, and typing @kbd{1 @key{RET}} selects all possible
7632 For example, the following session excerpt shows an attempt to set a
7633 breakpoint at the overloaded symbol @code{String::after}.
7634 We choose three particular definitions of that function name:
7636 @c FIXME! This is likely to change to show arg type lists, at least
7639 (@value{GDBP}) b String::after
7642 [2] file:String.cc; line number:867
7643 [3] file:String.cc; line number:860
7644 [4] file:String.cc; line number:875
7645 [5] file:String.cc; line number:853
7646 [6] file:String.cc; line number:846
7647 [7] file:String.cc; line number:735
7649 Breakpoint 1 at 0xb26c: file String.cc, line 867.
7650 Breakpoint 2 at 0xb344: file String.cc, line 875.
7651 Breakpoint 3 at 0xafcc: file String.cc, line 846.
7652 Multiple breakpoints were set.
7653 Use the "delete" command to delete unwanted
7660 @kindex set multiple-symbols
7661 @item set multiple-symbols @var{mode}
7662 @cindex multiple-symbols menu
7664 This option allows you to adjust the debugger behavior when an expression
7667 By default, @var{mode} is set to @code{all}. If the command with which
7668 the expression is used allows more than one choice, then @value{GDBN}
7669 automatically selects all possible choices. For instance, inserting
7670 a breakpoint on a function using an ambiguous name results in a breakpoint
7671 inserted on each possible match. However, if a unique choice must be made,
7672 then @value{GDBN} uses the menu to help you disambiguate the expression.
7673 For instance, printing the address of an overloaded function will result
7674 in the use of the menu.
7676 When @var{mode} is set to @code{ask}, the debugger always uses the menu
7677 when an ambiguity is detected.
7679 Finally, when @var{mode} is set to @code{cancel}, the debugger reports
7680 an error due to the ambiguity and the command is aborted.
7682 @kindex show multiple-symbols
7683 @item show multiple-symbols
7684 Show the current value of the @code{multiple-symbols} setting.
7688 @section Program Variables
7690 The most common kind of expression to use is the name of a variable
7693 Variables in expressions are understood in the selected stack frame
7694 (@pxref{Selection, ,Selecting a Frame}); they must be either:
7698 global (or file-static)
7705 visible according to the scope rules of the
7706 programming language from the point of execution in that frame
7709 @noindent This means that in the function
7724 you can examine and use the variable @code{a} whenever your program is
7725 executing within the function @code{foo}, but you can only use or
7726 examine the variable @code{b} while your program is executing inside
7727 the block where @code{b} is declared.
7729 @cindex variable name conflict
7730 There is an exception: you can refer to a variable or function whose
7731 scope is a single source file even if the current execution point is not
7732 in this file. But it is possible to have more than one such variable or
7733 function with the same name (in different source files). If that
7734 happens, referring to that name has unpredictable effects. If you wish,
7735 you can specify a static variable in a particular function or file by
7736 using the colon-colon (@code{::}) notation:
7738 @cindex colon-colon, context for variables/functions
7740 @c info cannot cope with a :: index entry, but why deprive hard copy readers?
7741 @cindex @code{::}, context for variables/functions
7744 @var{file}::@var{variable}
7745 @var{function}::@var{variable}
7749 Here @var{file} or @var{function} is the name of the context for the
7750 static @var{variable}. In the case of file names, you can use quotes to
7751 make sure @value{GDBN} parses the file name as a single word---for example,
7752 to print a global value of @code{x} defined in @file{f2.c}:
7755 (@value{GDBP}) p 'f2.c'::x
7758 The @code{::} notation is normally used for referring to
7759 static variables, since you typically disambiguate uses of local variables
7760 in functions by selecting the appropriate frame and using the
7761 simple name of the variable. However, you may also use this notation
7762 to refer to local variables in frames enclosing the selected frame:
7771 process (a); /* Stop here */
7782 For example, if there is a breakpoint at the commented line,
7783 here is what you might see
7784 when the program stops after executing the call @code{bar(0)}:
7789 (@value{GDBP}) p bar::a
7792 #2 0x080483d0 in foo (a=5) at foobar.c:12
7795 (@value{GDBP}) p bar::a
7799 @cindex C@t{++} scope resolution
7800 These uses of @samp{::} are very rarely in conflict with the very similar
7801 use of the same notation in C@t{++}. @value{GDBN} also supports use of the C@t{++}
7802 scope resolution operator in @value{GDBN} expressions.
7803 @c FIXME: Um, so what happens in one of those rare cases where it's in
7806 @cindex wrong values
7807 @cindex variable values, wrong
7808 @cindex function entry/exit, wrong values of variables
7809 @cindex optimized code, wrong values of variables
7811 @emph{Warning:} Occasionally, a local variable may appear to have the
7812 wrong value at certain points in a function---just after entry to a new
7813 scope, and just before exit.
7815 You may see this problem when you are stepping by machine instructions.
7816 This is because, on most machines, it takes more than one instruction to
7817 set up a stack frame (including local variable definitions); if you are
7818 stepping by machine instructions, variables may appear to have the wrong
7819 values until the stack frame is completely built. On exit, it usually
7820 also takes more than one machine instruction to destroy a stack frame;
7821 after you begin stepping through that group of instructions, local
7822 variable definitions may be gone.
7824 This may also happen when the compiler does significant optimizations.
7825 To be sure of always seeing accurate values, turn off all optimization
7828 @cindex ``No symbol "foo" in current context''
7829 Another possible effect of compiler optimizations is to optimize
7830 unused variables out of existence, or assign variables to registers (as
7831 opposed to memory addresses). Depending on the support for such cases
7832 offered by the debug info format used by the compiler, @value{GDBN}
7833 might not be able to display values for such local variables. If that
7834 happens, @value{GDBN} will print a message like this:
7837 No symbol "foo" in current context.
7840 To solve such problems, either recompile without optimizations, or use a
7841 different debug info format, if the compiler supports several such
7842 formats. @xref{Compilation}, for more information on choosing compiler
7843 options. @xref{C, ,C and C@t{++}}, for more information about debug
7844 info formats that are best suited to C@t{++} programs.
7846 If you ask to print an object whose contents are unknown to
7847 @value{GDBN}, e.g., because its data type is not completely specified
7848 by the debug information, @value{GDBN} will say @samp{<incomplete
7849 type>}. @xref{Symbols, incomplete type}, for more about this.
7851 If you append @kbd{@@entry} string to a function parameter name you get its
7852 value at the time the function got called. If the value is not available an
7853 error message is printed. Entry values are available only with some compilers.
7854 Entry values are normally also printed at the function parameter list according
7855 to @ref{set print entry-values}.
7858 Breakpoint 1, d (i=30) at gdb.base/entry-value.c:29
7864 (gdb) print i@@entry
7868 Strings are identified as arrays of @code{char} values without specified
7869 signedness. Arrays of either @code{signed char} or @code{unsigned char} get
7870 printed as arrays of 1 byte sized integers. @code{-fsigned-char} or
7871 @code{-funsigned-char} @value{NGCC} options have no effect as @value{GDBN}
7872 defines literal string type @code{"char"} as @code{char} without a sign.
7877 signed char var1[] = "A";
7880 You get during debugging
7885 $2 = @{65 'A', 0 '\0'@}
7889 @section Artificial Arrays
7891 @cindex artificial array
7893 @kindex @@@r{, referencing memory as an array}
7894 It is often useful to print out several successive objects of the
7895 same type in memory; a section of an array, or an array of
7896 dynamically determined size for which only a pointer exists in the
7899 You can do this by referring to a contiguous span of memory as an
7900 @dfn{artificial array}, using the binary operator @samp{@@}. The left
7901 operand of @samp{@@} should be the first element of the desired array
7902 and be an individual object. The right operand should be the desired length
7903 of the array. The result is an array value whose elements are all of
7904 the type of the left argument. The first element is actually the left
7905 argument; the second element comes from bytes of memory immediately
7906 following those that hold the first element, and so on. Here is an
7907 example. If a program says
7910 int *array = (int *) malloc (len * sizeof (int));
7914 you can print the contents of @code{array} with
7920 The left operand of @samp{@@} must reside in memory. Array values made
7921 with @samp{@@} in this way behave just like other arrays in terms of
7922 subscripting, and are coerced to pointers when used in expressions.
7923 Artificial arrays most often appear in expressions via the value history
7924 (@pxref{Value History, ,Value History}), after printing one out.
7926 Another way to create an artificial array is to use a cast.
7927 This re-interprets a value as if it were an array.
7928 The value need not be in memory:
7930 (@value{GDBP}) p/x (short[2])0x12345678
7931 $1 = @{0x1234, 0x5678@}
7934 As a convenience, if you leave the array length out (as in
7935 @samp{(@var{type}[])@var{value}}) @value{GDBN} calculates the size to fill
7936 the value (as @samp{sizeof(@var{value})/sizeof(@var{type})}:
7938 (@value{GDBP}) p/x (short[])0x12345678
7939 $2 = @{0x1234, 0x5678@}
7942 Sometimes the artificial array mechanism is not quite enough; in
7943 moderately complex data structures, the elements of interest may not
7944 actually be adjacent---for example, if you are interested in the values
7945 of pointers in an array. One useful work-around in this situation is
7946 to use a convenience variable (@pxref{Convenience Vars, ,Convenience
7947 Variables}) as a counter in an expression that prints the first
7948 interesting value, and then repeat that expression via @key{RET}. For
7949 instance, suppose you have an array @code{dtab} of pointers to
7950 structures, and you are interested in the values of a field @code{fv}
7951 in each structure. Here is an example of what you might type:
7961 @node Output Formats
7962 @section Output Formats
7964 @cindex formatted output
7965 @cindex output formats
7966 By default, @value{GDBN} prints a value according to its data type. Sometimes
7967 this is not what you want. For example, you might want to print a number
7968 in hex, or a pointer in decimal. Or you might want to view data in memory
7969 at a certain address as a character string or as an instruction. To do
7970 these things, specify an @dfn{output format} when you print a value.
7972 The simplest use of output formats is to say how to print a value
7973 already computed. This is done by starting the arguments of the
7974 @code{print} command with a slash and a format letter. The format
7975 letters supported are:
7979 Regard the bits of the value as an integer, and print the integer in
7983 Print as integer in signed decimal.
7986 Print as integer in unsigned decimal.
7989 Print as integer in octal.
7992 Print as integer in binary. The letter @samp{t} stands for ``two''.
7993 @footnote{@samp{b} cannot be used because these format letters are also
7994 used with the @code{x} command, where @samp{b} stands for ``byte'';
7995 see @ref{Memory,,Examining Memory}.}
7998 @cindex unknown address, locating
7999 @cindex locate address
8000 Print as an address, both absolute in hexadecimal and as an offset from
8001 the nearest preceding symbol. You can use this format used to discover
8002 where (in what function) an unknown address is located:
8005 (@value{GDBP}) p/a 0x54320
8006 $3 = 0x54320 <_initialize_vx+396>
8010 The command @code{info symbol 0x54320} yields similar results.
8011 @xref{Symbols, info symbol}.
8014 Regard as an integer and print it as a character constant. This
8015 prints both the numerical value and its character representation. The
8016 character representation is replaced with the octal escape @samp{\nnn}
8017 for characters outside the 7-bit @sc{ascii} range.
8019 Without this format, @value{GDBN} displays @code{char},
8020 @w{@code{unsigned char}}, and @w{@code{signed char}} data as character
8021 constants. Single-byte members of vectors are displayed as integer
8025 Regard the bits of the value as a floating point number and print
8026 using typical floating point syntax.
8029 @cindex printing strings
8030 @cindex printing byte arrays
8031 Regard as a string, if possible. With this format, pointers to single-byte
8032 data are displayed as null-terminated strings and arrays of single-byte data
8033 are displayed as fixed-length strings. Other values are displayed in their
8036 Without this format, @value{GDBN} displays pointers to and arrays of
8037 @code{char}, @w{@code{unsigned char}}, and @w{@code{signed char}} as
8038 strings. Single-byte members of a vector are displayed as an integer
8042 @cindex raw printing
8043 Print using the @samp{raw} formatting. By default, @value{GDBN} will
8044 use a Python-based pretty-printer, if one is available (@pxref{Pretty
8045 Printing}). This typically results in a higher-level display of the
8046 value's contents. The @samp{r} format bypasses any Python
8047 pretty-printer which might exist.
8050 For example, to print the program counter in hex (@pxref{Registers}), type
8057 Note that no space is required before the slash; this is because command
8058 names in @value{GDBN} cannot contain a slash.
8060 To reprint the last value in the value history with a different format,
8061 you can use the @code{print} command with just a format and no
8062 expression. For example, @samp{p/x} reprints the last value in hex.
8065 @section Examining Memory
8067 You can use the command @code{x} (for ``examine'') to examine memory in
8068 any of several formats, independently of your program's data types.
8070 @cindex examining memory
8072 @kindex x @r{(examine memory)}
8073 @item x/@var{nfu} @var{addr}
8076 Use the @code{x} command to examine memory.
8079 @var{n}, @var{f}, and @var{u} are all optional parameters that specify how
8080 much memory to display and how to format it; @var{addr} is an
8081 expression giving the address where you want to start displaying memory.
8082 If you use defaults for @var{nfu}, you need not type the slash @samp{/}.
8083 Several commands set convenient defaults for @var{addr}.
8086 @item @var{n}, the repeat count
8087 The repeat count is a decimal integer; the default is 1. It specifies
8088 how much memory (counting by units @var{u}) to display.
8089 @c This really is **decimal**; unaffected by 'set radix' as of GDB
8092 @item @var{f}, the display format
8093 The display format is one of the formats used by @code{print}
8094 (@samp{x}, @samp{d}, @samp{u}, @samp{o}, @samp{t}, @samp{a}, @samp{c},
8095 @samp{f}, @samp{s}), and in addition @samp{i} (for machine instructions).
8096 The default is @samp{x} (hexadecimal) initially. The default changes
8097 each time you use either @code{x} or @code{print}.
8099 @item @var{u}, the unit size
8100 The unit size is any of
8106 Halfwords (two bytes).
8108 Words (four bytes). This is the initial default.
8110 Giant words (eight bytes).
8113 Each time you specify a unit size with @code{x}, that size becomes the
8114 default unit the next time you use @code{x}. For the @samp{i} format,
8115 the unit size is ignored and is normally not written. For the @samp{s} format,
8116 the unit size defaults to @samp{b}, unless it is explicitly given.
8117 Use @kbd{x /hs} to display 16-bit char strings and @kbd{x /ws} to display
8118 32-bit strings. The next use of @kbd{x /s} will again display 8-bit strings.
8119 Note that the results depend on the programming language of the
8120 current compilation unit. If the language is C, the @samp{s}
8121 modifier will use the UTF-16 encoding while @samp{w} will use
8122 UTF-32. The encoding is set by the programming language and cannot
8125 @item @var{addr}, starting display address
8126 @var{addr} is the address where you want @value{GDBN} to begin displaying
8127 memory. The expression need not have a pointer value (though it may);
8128 it is always interpreted as an integer address of a byte of memory.
8129 @xref{Expressions, ,Expressions}, for more information on expressions. The default for
8130 @var{addr} is usually just after the last address examined---but several
8131 other commands also set the default address: @code{info breakpoints} (to
8132 the address of the last breakpoint listed), @code{info line} (to the
8133 starting address of a line), and @code{print} (if you use it to display
8134 a value from memory).
8137 For example, @samp{x/3uh 0x54320} is a request to display three halfwords
8138 (@code{h}) of memory, formatted as unsigned decimal integers (@samp{u}),
8139 starting at address @code{0x54320}. @samp{x/4xw $sp} prints the four
8140 words (@samp{w}) of memory above the stack pointer (here, @samp{$sp};
8141 @pxref{Registers, ,Registers}) in hexadecimal (@samp{x}).
8143 Since the letters indicating unit sizes are all distinct from the
8144 letters specifying output formats, you do not have to remember whether
8145 unit size or format comes first; either order works. The output
8146 specifications @samp{4xw} and @samp{4wx} mean exactly the same thing.
8147 (However, the count @var{n} must come first; @samp{wx4} does not work.)
8149 Even though the unit size @var{u} is ignored for the formats @samp{s}
8150 and @samp{i}, you might still want to use a count @var{n}; for example,
8151 @samp{3i} specifies that you want to see three machine instructions,
8152 including any operands. For convenience, especially when used with
8153 the @code{display} command, the @samp{i} format also prints branch delay
8154 slot instructions, if any, beyond the count specified, which immediately
8155 follow the last instruction that is within the count. The command
8156 @code{disassemble} gives an alternative way of inspecting machine
8157 instructions; see @ref{Machine Code,,Source and Machine Code}.
8159 All the defaults for the arguments to @code{x} are designed to make it
8160 easy to continue scanning memory with minimal specifications each time
8161 you use @code{x}. For example, after you have inspected three machine
8162 instructions with @samp{x/3i @var{addr}}, you can inspect the next seven
8163 with just @samp{x/7}. If you use @key{RET} to repeat the @code{x} command,
8164 the repeat count @var{n} is used again; the other arguments default as
8165 for successive uses of @code{x}.
8167 When examining machine instructions, the instruction at current program
8168 counter is shown with a @code{=>} marker. For example:
8171 (@value{GDBP}) x/5i $pc-6
8172 0x804837f <main+11>: mov %esp,%ebp
8173 0x8048381 <main+13>: push %ecx
8174 0x8048382 <main+14>: sub $0x4,%esp
8175 => 0x8048385 <main+17>: movl $0x8048460,(%esp)
8176 0x804838c <main+24>: call 0x80482d4 <puts@@plt>
8179 @cindex @code{$_}, @code{$__}, and value history
8180 The addresses and contents printed by the @code{x} command are not saved
8181 in the value history because there is often too much of them and they
8182 would get in the way. Instead, @value{GDBN} makes these values available for
8183 subsequent use in expressions as values of the convenience variables
8184 @code{$_} and @code{$__}. After an @code{x} command, the last address
8185 examined is available for use in expressions in the convenience variable
8186 @code{$_}. The contents of that address, as examined, are available in
8187 the convenience variable @code{$__}.
8189 If the @code{x} command has a repeat count, the address and contents saved
8190 are from the last memory unit printed; this is not the same as the last
8191 address printed if several units were printed on the last line of output.
8193 @cindex remote memory comparison
8194 @cindex verify remote memory image
8195 When you are debugging a program running on a remote target machine
8196 (@pxref{Remote Debugging}), you may wish to verify the program's image in the
8197 remote machine's memory against the executable file you downloaded to
8198 the target. The @code{compare-sections} command is provided for such
8202 @kindex compare-sections
8203 @item compare-sections @r{[}@var{section-name}@r{]}
8204 Compare the data of a loadable section @var{section-name} in the
8205 executable file of the program being debugged with the same section in
8206 the remote machine's memory, and report any mismatches. With no
8207 arguments, compares all loadable sections. This command's
8208 availability depends on the target's support for the @code{"qCRC"}
8213 @section Automatic Display
8214 @cindex automatic display
8215 @cindex display of expressions
8217 If you find that you want to print the value of an expression frequently
8218 (to see how it changes), you might want to add it to the @dfn{automatic
8219 display list} so that @value{GDBN} prints its value each time your program stops.
8220 Each expression added to the list is given a number to identify it;
8221 to remove an expression from the list, you specify that number.
8222 The automatic display looks like this:
8226 3: bar[5] = (struct hack *) 0x3804
8230 This display shows item numbers, expressions and their current values. As with
8231 displays you request manually using @code{x} or @code{print}, you can
8232 specify the output format you prefer; in fact, @code{display} decides
8233 whether to use @code{print} or @code{x} depending your format
8234 specification---it uses @code{x} if you specify either the @samp{i}
8235 or @samp{s} format, or a unit size; otherwise it uses @code{print}.
8239 @item display @var{expr}
8240 Add the expression @var{expr} to the list of expressions to display
8241 each time your program stops. @xref{Expressions, ,Expressions}.
8243 @code{display} does not repeat if you press @key{RET} again after using it.
8245 @item display/@var{fmt} @var{expr}
8246 For @var{fmt} specifying only a display format and not a size or
8247 count, add the expression @var{expr} to the auto-display list but
8248 arrange to display it each time in the specified format @var{fmt}.
8249 @xref{Output Formats,,Output Formats}.
8251 @item display/@var{fmt} @var{addr}
8252 For @var{fmt} @samp{i} or @samp{s}, or including a unit-size or a
8253 number of units, add the expression @var{addr} as a memory address to
8254 be examined each time your program stops. Examining means in effect
8255 doing @samp{x/@var{fmt} @var{addr}}. @xref{Memory, ,Examining Memory}.
8258 For example, @samp{display/i $pc} can be helpful, to see the machine
8259 instruction about to be executed each time execution stops (@samp{$pc}
8260 is a common name for the program counter; @pxref{Registers, ,Registers}).
8263 @kindex delete display
8265 @item undisplay @var{dnums}@dots{}
8266 @itemx delete display @var{dnums}@dots{}
8267 Remove items from the list of expressions to display. Specify the
8268 numbers of the displays that you want affected with the command
8269 argument @var{dnums}. It can be a single display number, one of the
8270 numbers shown in the first field of the @samp{info display} display;
8271 or it could be a range of display numbers, as in @code{2-4}.
8273 @code{undisplay} does not repeat if you press @key{RET} after using it.
8274 (Otherwise you would just get the error @samp{No display number @dots{}}.)
8276 @kindex disable display
8277 @item disable display @var{dnums}@dots{}
8278 Disable the display of item numbers @var{dnums}. A disabled display
8279 item is not printed automatically, but is not forgotten. It may be
8280 enabled again later. Specify the numbers of the displays that you
8281 want affected with the command argument @var{dnums}. It can be a
8282 single display number, one of the numbers shown in the first field of
8283 the @samp{info display} display; or it could be a range of display
8284 numbers, as in @code{2-4}.
8286 @kindex enable display
8287 @item enable display @var{dnums}@dots{}
8288 Enable display of item numbers @var{dnums}. It becomes effective once
8289 again in auto display of its expression, until you specify otherwise.
8290 Specify the numbers of the displays that you want affected with the
8291 command argument @var{dnums}. It can be a single display number, one
8292 of the numbers shown in the first field of the @samp{info display}
8293 display; or it could be a range of display numbers, as in @code{2-4}.
8296 Display the current values of the expressions on the list, just as is
8297 done when your program stops.
8299 @kindex info display
8301 Print the list of expressions previously set up to display
8302 automatically, each one with its item number, but without showing the
8303 values. This includes disabled expressions, which are marked as such.
8304 It also includes expressions which would not be displayed right now
8305 because they refer to automatic variables not currently available.
8308 @cindex display disabled out of scope
8309 If a display expression refers to local variables, then it does not make
8310 sense outside the lexical context for which it was set up. Such an
8311 expression is disabled when execution enters a context where one of its
8312 variables is not defined. For example, if you give the command
8313 @code{display last_char} while inside a function with an argument
8314 @code{last_char}, @value{GDBN} displays this argument while your program
8315 continues to stop inside that function. When it stops elsewhere---where
8316 there is no variable @code{last_char}---the display is disabled
8317 automatically. The next time your program stops where @code{last_char}
8318 is meaningful, you can enable the display expression once again.
8320 @node Print Settings
8321 @section Print Settings
8323 @cindex format options
8324 @cindex print settings
8325 @value{GDBN} provides the following ways to control how arrays, structures,
8326 and symbols are printed.
8329 These settings are useful for debugging programs in any language:
8333 @item set print address
8334 @itemx set print address on
8335 @cindex print/don't print memory addresses
8336 @value{GDBN} prints memory addresses showing the location of stack
8337 traces, structure values, pointer values, breakpoints, and so forth,
8338 even when it also displays the contents of those addresses. The default
8339 is @code{on}. For example, this is what a stack frame display looks like with
8340 @code{set print address on}:
8345 #0 set_quotes (lq=0x34c78 "<<", rq=0x34c88 ">>")
8347 530 if (lquote != def_lquote)
8351 @item set print address off
8352 Do not print addresses when displaying their contents. For example,
8353 this is the same stack frame displayed with @code{set print address off}:
8357 (@value{GDBP}) set print addr off
8359 #0 set_quotes (lq="<<", rq=">>") at input.c:530
8360 530 if (lquote != def_lquote)
8364 You can use @samp{set print address off} to eliminate all machine
8365 dependent displays from the @value{GDBN} interface. For example, with
8366 @code{print address off}, you should get the same text for backtraces on
8367 all machines---whether or not they involve pointer arguments.
8370 @item show print address
8371 Show whether or not addresses are to be printed.
8374 When @value{GDBN} prints a symbolic address, it normally prints the
8375 closest earlier symbol plus an offset. If that symbol does not uniquely
8376 identify the address (for example, it is a name whose scope is a single
8377 source file), you may need to clarify. One way to do this is with
8378 @code{info line}, for example @samp{info line *0x4537}. Alternately,
8379 you can set @value{GDBN} to print the source file and line number when
8380 it prints a symbolic address:
8383 @item set print symbol-filename on
8384 @cindex source file and line of a symbol
8385 @cindex symbol, source file and line
8386 Tell @value{GDBN} to print the source file name and line number of a
8387 symbol in the symbolic form of an address.
8389 @item set print symbol-filename off
8390 Do not print source file name and line number of a symbol. This is the
8393 @item show print symbol-filename
8394 Show whether or not @value{GDBN} will print the source file name and
8395 line number of a symbol in the symbolic form of an address.
8398 Another situation where it is helpful to show symbol filenames and line
8399 numbers is when disassembling code; @value{GDBN} shows you the line
8400 number and source file that corresponds to each instruction.
8402 Also, you may wish to see the symbolic form only if the address being
8403 printed is reasonably close to the closest earlier symbol:
8406 @item set print max-symbolic-offset @var{max-offset}
8407 @cindex maximum value for offset of closest symbol
8408 Tell @value{GDBN} to only display the symbolic form of an address if the
8409 offset between the closest earlier symbol and the address is less than
8410 @var{max-offset}. The default is 0, which tells @value{GDBN}
8411 to always print the symbolic form of an address if any symbol precedes it.
8413 @item show print max-symbolic-offset
8414 Ask how large the maximum offset is that @value{GDBN} prints in a
8418 @cindex wild pointer, interpreting
8419 @cindex pointer, finding referent
8420 If you have a pointer and you are not sure where it points, try
8421 @samp{set print symbol-filename on}. Then you can determine the name
8422 and source file location of the variable where it points, using
8423 @samp{p/a @var{pointer}}. This interprets the address in symbolic form.
8424 For example, here @value{GDBN} shows that a variable @code{ptt} points
8425 at another variable @code{t}, defined in @file{hi2.c}:
8428 (@value{GDBP}) set print symbol-filename on
8429 (@value{GDBP}) p/a ptt
8430 $4 = 0xe008 <t in hi2.c>
8434 @emph{Warning:} For pointers that point to a local variable, @samp{p/a}
8435 does not show the symbol name and filename of the referent, even with
8436 the appropriate @code{set print} options turned on.
8439 You can also enable @samp{/a}-like formatting all the time using
8440 @samp{set print symbol on}:
8443 @item set print symbol on
8444 Tell @value{GDBN} to print the symbol corresponding to an address, if
8447 @item set print symbol off
8448 Tell @value{GDBN} not to print the symbol corresponding to an
8449 address. In this mode, @value{GDBN} will still print the symbol
8450 corresponding to pointers to functions. This is the default.
8452 @item show print symbol
8453 Show whether @value{GDBN} will display the symbol corresponding to an
8457 Other settings control how different kinds of objects are printed:
8460 @item set print array
8461 @itemx set print array on
8462 @cindex pretty print arrays
8463 Pretty print arrays. This format is more convenient to read,
8464 but uses more space. The default is off.
8466 @item set print array off
8467 Return to compressed format for arrays.
8469 @item show print array
8470 Show whether compressed or pretty format is selected for displaying
8473 @cindex print array indexes
8474 @item set print array-indexes
8475 @itemx set print array-indexes on
8476 Print the index of each element when displaying arrays. May be more
8477 convenient to locate a given element in the array or quickly find the
8478 index of a given element in that printed array. The default is off.
8480 @item set print array-indexes off
8481 Stop printing element indexes when displaying arrays.
8483 @item show print array-indexes
8484 Show whether the index of each element is printed when displaying
8487 @item set print elements @var{number-of-elements}
8488 @cindex number of array elements to print
8489 @cindex limit on number of printed array elements
8490 Set a limit on how many elements of an array @value{GDBN} will print.
8491 If @value{GDBN} is printing a large array, it stops printing after it has
8492 printed the number of elements set by the @code{set print elements} command.
8493 This limit also applies to the display of strings.
8494 When @value{GDBN} starts, this limit is set to 200.
8495 Setting @var{number-of-elements} to zero means that the printing is unlimited.
8497 @item show print elements
8498 Display the number of elements of a large array that @value{GDBN} will print.
8499 If the number is 0, then the printing is unlimited.
8501 @item set print frame-arguments @var{value}
8502 @kindex set print frame-arguments
8503 @cindex printing frame argument values
8504 @cindex print all frame argument values
8505 @cindex print frame argument values for scalars only
8506 @cindex do not print frame argument values
8507 This command allows to control how the values of arguments are printed
8508 when the debugger prints a frame (@pxref{Frames}). The possible
8513 The values of all arguments are printed.
8516 Print the value of an argument only if it is a scalar. The value of more
8517 complex arguments such as arrays, structures, unions, etc, is replaced
8518 by @code{@dots{}}. This is the default. Here is an example where
8519 only scalar arguments are shown:
8522 #1 0x08048361 in call_me (i=3, s=@dots{}, ss=0xbf8d508c, u=@dots{}, e=green)
8527 None of the argument values are printed. Instead, the value of each argument
8528 is replaced by @code{@dots{}}. In this case, the example above now becomes:
8531 #1 0x08048361 in call_me (i=@dots{}, s=@dots{}, ss=@dots{}, u=@dots{}, e=@dots{})
8536 By default, only scalar arguments are printed. This command can be used
8537 to configure the debugger to print the value of all arguments, regardless
8538 of their type. However, it is often advantageous to not print the value
8539 of more complex parameters. For instance, it reduces the amount of
8540 information printed in each frame, making the backtrace more readable.
8541 Also, it improves performance when displaying Ada frames, because
8542 the computation of large arguments can sometimes be CPU-intensive,
8543 especially in large applications. Setting @code{print frame-arguments}
8544 to @code{scalars} (the default) or @code{none} avoids this computation,
8545 thus speeding up the display of each Ada frame.
8547 @item show print frame-arguments
8548 Show how the value of arguments should be displayed when printing a frame.
8550 @anchor{set print entry-values}
8551 @item set print entry-values @var{value}
8552 @kindex set print entry-values
8553 Set printing of frame argument values at function entry. In some cases
8554 @value{GDBN} can determine the value of function argument which was passed by
8555 the function caller, even if the value was modified inside the called function
8556 and therefore is different. With optimized code, the current value could be
8557 unavailable, but the entry value may still be known.
8559 The default value is @code{default} (see below for its description). Older
8560 @value{GDBN} behaved as with the setting @code{no}. Compilers not supporting
8561 this feature will behave in the @code{default} setting the same way as with the
8564 This functionality is currently supported only by DWARF 2 debugging format and
8565 the compiler has to produce @samp{DW_TAG_GNU_call_site} tags. With
8566 @value{NGCC}, you need to specify @option{-O -g} during compilation, to get
8569 The @var{value} parameter can be one of the following:
8573 Print only actual parameter values, never print values from function entry
8577 #0 different (val=6)
8578 #0 lost (val=<optimized out>)
8580 #0 invalid (val=<optimized out>)
8584 Print only parameter values from function entry point. The actual parameter
8585 values are never printed.
8587 #0 equal (val@@entry=5)
8588 #0 different (val@@entry=5)
8589 #0 lost (val@@entry=5)
8590 #0 born (val@@entry=<optimized out>)
8591 #0 invalid (val@@entry=<optimized out>)
8595 Print only parameter values from function entry point. If value from function
8596 entry point is not known while the actual value is known, print the actual
8597 value for such parameter.
8599 #0 equal (val@@entry=5)
8600 #0 different (val@@entry=5)
8601 #0 lost (val@@entry=5)
8603 #0 invalid (val@@entry=<optimized out>)
8607 Print actual parameter values. If actual parameter value is not known while
8608 value from function entry point is known, print the entry point value for such
8612 #0 different (val=6)
8613 #0 lost (val@@entry=5)
8615 #0 invalid (val=<optimized out>)
8619 Always print both the actual parameter value and its value from function entry
8620 point, even if values of one or both are not available due to compiler
8623 #0 equal (val=5, val@@entry=5)
8624 #0 different (val=6, val@@entry=5)
8625 #0 lost (val=<optimized out>, val@@entry=5)
8626 #0 born (val=10, val@@entry=<optimized out>)
8627 #0 invalid (val=<optimized out>, val@@entry=<optimized out>)
8631 Print the actual parameter value if it is known and also its value from
8632 function entry point if it is known. If neither is known, print for the actual
8633 value @code{<optimized out>}. If not in MI mode (@pxref{GDB/MI}) and if both
8634 values are known and identical, print the shortened
8635 @code{param=param@@entry=VALUE} notation.
8637 #0 equal (val=val@@entry=5)
8638 #0 different (val=6, val@@entry=5)
8639 #0 lost (val@@entry=5)
8641 #0 invalid (val=<optimized out>)
8645 Always print the actual parameter value. Print also its value from function
8646 entry point, but only if it is known. If not in MI mode (@pxref{GDB/MI}) and
8647 if both values are known and identical, print the shortened
8648 @code{param=param@@entry=VALUE} notation.
8650 #0 equal (val=val@@entry=5)
8651 #0 different (val=6, val@@entry=5)
8652 #0 lost (val=<optimized out>, val@@entry=5)
8654 #0 invalid (val=<optimized out>)
8658 For analysis messages on possible failures of frame argument values at function
8659 entry resolution see @ref{set debug entry-values}.
8661 @item show print entry-values
8662 Show the method being used for printing of frame argument values at function
8665 @item set print repeats
8666 @cindex repeated array elements
8667 Set the threshold for suppressing display of repeated array
8668 elements. When the number of consecutive identical elements of an
8669 array exceeds the threshold, @value{GDBN} prints the string
8670 @code{"<repeats @var{n} times>"}, where @var{n} is the number of
8671 identical repetitions, instead of displaying the identical elements
8672 themselves. Setting the threshold to zero will cause all elements to
8673 be individually printed. The default threshold is 10.
8675 @item show print repeats
8676 Display the current threshold for printing repeated identical
8679 @item set print null-stop
8680 @cindex @sc{null} elements in arrays
8681 Cause @value{GDBN} to stop printing the characters of an array when the first
8682 @sc{null} is encountered. This is useful when large arrays actually
8683 contain only short strings.
8686 @item show print null-stop
8687 Show whether @value{GDBN} stops printing an array on the first
8688 @sc{null} character.
8690 @item set print pretty on
8691 @cindex print structures in indented form
8692 @cindex indentation in structure display
8693 Cause @value{GDBN} to print structures in an indented format with one member
8694 per line, like this:
8709 @item set print pretty off
8710 Cause @value{GDBN} to print structures in a compact format, like this:
8714 $1 = @{next = 0x0, flags = @{sweet = 1, sour = 1@}, \
8715 meat = 0x54 "Pork"@}
8720 This is the default format.
8722 @item show print pretty
8723 Show which format @value{GDBN} is using to print structures.
8725 @item set print sevenbit-strings on
8726 @cindex eight-bit characters in strings
8727 @cindex octal escapes in strings
8728 Print using only seven-bit characters; if this option is set,
8729 @value{GDBN} displays any eight-bit characters (in strings or
8730 character values) using the notation @code{\}@var{nnn}. This setting is
8731 best if you are working in English (@sc{ascii}) and you use the
8732 high-order bit of characters as a marker or ``meta'' bit.
8734 @item set print sevenbit-strings off
8735 Print full eight-bit characters. This allows the use of more
8736 international character sets, and is the default.
8738 @item show print sevenbit-strings
8739 Show whether or not @value{GDBN} is printing only seven-bit characters.
8741 @item set print union on
8742 @cindex unions in structures, printing
8743 Tell @value{GDBN} to print unions which are contained in structures
8744 and other unions. This is the default setting.
8746 @item set print union off
8747 Tell @value{GDBN} not to print unions which are contained in
8748 structures and other unions. @value{GDBN} will print @code{"@{...@}"}
8751 @item show print union
8752 Ask @value{GDBN} whether or not it will print unions which are contained in
8753 structures and other unions.
8755 For example, given the declarations
8758 typedef enum @{Tree, Bug@} Species;
8759 typedef enum @{Big_tree, Acorn, Seedling@} Tree_forms;
8760 typedef enum @{Caterpillar, Cocoon, Butterfly@}
8771 struct thing foo = @{Tree, @{Acorn@}@};
8775 with @code{set print union on} in effect @samp{p foo} would print
8778 $1 = @{it = Tree, form = @{tree = Acorn, bug = Cocoon@}@}
8782 and with @code{set print union off} in effect it would print
8785 $1 = @{it = Tree, form = @{...@}@}
8789 @code{set print union} affects programs written in C-like languages
8795 These settings are of interest when debugging C@t{++} programs:
8798 @cindex demangling C@t{++} names
8799 @item set print demangle
8800 @itemx set print demangle on
8801 Print C@t{++} names in their source form rather than in the encoded
8802 (``mangled'') form passed to the assembler and linker for type-safe
8803 linkage. The default is on.
8805 @item show print demangle
8806 Show whether C@t{++} names are printed in mangled or demangled form.
8808 @item set print asm-demangle
8809 @itemx set print asm-demangle on
8810 Print C@t{++} names in their source form rather than their mangled form, even
8811 in assembler code printouts such as instruction disassemblies.
8814 @item show print asm-demangle
8815 Show whether C@t{++} names in assembly listings are printed in mangled
8818 @cindex C@t{++} symbol decoding style
8819 @cindex symbol decoding style, C@t{++}
8820 @kindex set demangle-style
8821 @item set demangle-style @var{style}
8822 Choose among several encoding schemes used by different compilers to
8823 represent C@t{++} names. The choices for @var{style} are currently:
8827 Allow @value{GDBN} to choose a decoding style by inspecting your program.
8830 Decode based on the @sc{gnu} C@t{++} compiler (@code{g++}) encoding algorithm.
8831 This is the default.
8834 Decode based on the HP ANSI C@t{++} (@code{aCC}) encoding algorithm.
8837 Decode based on the Lucid C@t{++} compiler (@code{lcc}) encoding algorithm.
8840 Decode using the algorithm in the @cite{C@t{++} Annotated Reference Manual}.
8841 @strong{Warning:} this setting alone is not sufficient to allow
8842 debugging @code{cfront}-generated executables. @value{GDBN} would
8843 require further enhancement to permit that.
8846 If you omit @var{style}, you will see a list of possible formats.
8848 @item show demangle-style
8849 Display the encoding style currently in use for decoding C@t{++} symbols.
8851 @item set print object
8852 @itemx set print object on
8853 @cindex derived type of an object, printing
8854 @cindex display derived types
8855 When displaying a pointer to an object, identify the @emph{actual}
8856 (derived) type of the object rather than the @emph{declared} type, using
8857 the virtual function table. Note that the virtual function table is
8858 required---this feature can only work for objects that have run-time
8859 type identification; a single virtual method in the object's declared
8860 type is sufficient. Note that this setting is also taken into account when
8861 working with variable objects via MI (@pxref{GDB/MI}).
8863 @item set print object off
8864 Display only the declared type of objects, without reference to the
8865 virtual function table. This is the default setting.
8867 @item show print object
8868 Show whether actual, or declared, object types are displayed.
8870 @item set print static-members
8871 @itemx set print static-members on
8872 @cindex static members of C@t{++} objects
8873 Print static members when displaying a C@t{++} object. The default is on.
8875 @item set print static-members off
8876 Do not print static members when displaying a C@t{++} object.
8878 @item show print static-members
8879 Show whether C@t{++} static members are printed or not.
8881 @item set print pascal_static-members
8882 @itemx set print pascal_static-members on
8883 @cindex static members of Pascal objects
8884 @cindex Pascal objects, static members display
8885 Print static members when displaying a Pascal object. The default is on.
8887 @item set print pascal_static-members off
8888 Do not print static members when displaying a Pascal object.
8890 @item show print pascal_static-members
8891 Show whether Pascal static members are printed or not.
8893 @c These don't work with HP ANSI C++ yet.
8894 @item set print vtbl
8895 @itemx set print vtbl on
8896 @cindex pretty print C@t{++} virtual function tables
8897 @cindex virtual functions (C@t{++}) display
8898 @cindex VTBL display
8899 Pretty print C@t{++} virtual function tables. The default is off.
8900 (The @code{vtbl} commands do not work on programs compiled with the HP
8901 ANSI C@t{++} compiler (@code{aCC}).)
8903 @item set print vtbl off
8904 Do not pretty print C@t{++} virtual function tables.
8906 @item show print vtbl
8907 Show whether C@t{++} virtual function tables are pretty printed, or not.
8910 @node Pretty Printing
8911 @section Pretty Printing
8913 @value{GDBN} provides a mechanism to allow pretty-printing of values using
8914 Python code. It greatly simplifies the display of complex objects. This
8915 mechanism works for both MI and the CLI.
8918 * Pretty-Printer Introduction:: Introduction to pretty-printers
8919 * Pretty-Printer Example:: An example pretty-printer
8920 * Pretty-Printer Commands:: Pretty-printer commands
8923 @node Pretty-Printer Introduction
8924 @subsection Pretty-Printer Introduction
8926 When @value{GDBN} prints a value, it first sees if there is a pretty-printer
8927 registered for the value. If there is then @value{GDBN} invokes the
8928 pretty-printer to print the value. Otherwise the value is printed normally.
8930 Pretty-printers are normally named. This makes them easy to manage.
8931 The @samp{info pretty-printer} command will list all the installed
8932 pretty-printers with their names.
8933 If a pretty-printer can handle multiple data types, then its
8934 @dfn{subprinters} are the printers for the individual data types.
8935 Each such subprinter has its own name.
8936 The format of the name is @var{printer-name};@var{subprinter-name}.
8938 Pretty-printers are installed by @dfn{registering} them with @value{GDBN}.
8939 Typically they are automatically loaded and registered when the corresponding
8940 debug information is loaded, thus making them available without having to
8941 do anything special.
8943 There are three places where a pretty-printer can be registered.
8947 Pretty-printers registered globally are available when debugging
8951 Pretty-printers registered with a program space are available only
8952 when debugging that program.
8953 @xref{Progspaces In Python}, for more details on program spaces in Python.
8956 Pretty-printers registered with an objfile are loaded and unloaded
8957 with the corresponding objfile (e.g., shared library).
8958 @xref{Objfiles In Python}, for more details on objfiles in Python.
8961 @xref{Selecting Pretty-Printers}, for further information on how
8962 pretty-printers are selected,
8964 @xref{Writing a Pretty-Printer}, for implementing pretty printers
8967 @node Pretty-Printer Example
8968 @subsection Pretty-Printer Example
8970 Here is how a C@t{++} @code{std::string} looks without a pretty-printer:
8973 (@value{GDBP}) print s
8975 static npos = 4294967295,
8977 <std::allocator<char>> = @{
8978 <__gnu_cxx::new_allocator<char>> = @{
8979 <No data fields>@}, <No data fields>
8981 members of std::basic_string<char, std::char_traits<char>,
8982 std::allocator<char> >::_Alloc_hider:
8983 _M_p = 0x804a014 "abcd"
8988 With a pretty-printer for @code{std::string} only the contents are printed:
8991 (@value{GDBP}) print s
8995 @node Pretty-Printer Commands
8996 @subsection Pretty-Printer Commands
8997 @cindex pretty-printer commands
9000 @kindex info pretty-printer
9001 @item info pretty-printer [@var{object-regexp} [@var{name-regexp}]]
9002 Print the list of installed pretty-printers.
9003 This includes disabled pretty-printers, which are marked as such.
9005 @var{object-regexp} is a regular expression matching the objects
9006 whose pretty-printers to list.
9007 Objects can be @code{global}, the program space's file
9008 (@pxref{Progspaces In Python}),
9009 and the object files within that program space (@pxref{Objfiles In Python}).
9010 @xref{Selecting Pretty-Printers}, for details on how @value{GDBN}
9011 looks up a printer from these three objects.
9013 @var{name-regexp} is a regular expression matching the name of the printers
9016 @kindex disable pretty-printer
9017 @item disable pretty-printer [@var{object-regexp} [@var{name-regexp}]]
9018 Disable pretty-printers matching @var{object-regexp} and @var{name-regexp}.
9019 A disabled pretty-printer is not forgotten, it may be enabled again later.
9021 @kindex enable pretty-printer
9022 @item enable pretty-printer [@var{object-regexp} [@var{name-regexp}]]
9023 Enable pretty-printers matching @var{object-regexp} and @var{name-regexp}.
9028 Suppose we have three pretty-printers installed: one from library1.so
9029 named @code{foo} that prints objects of type @code{foo}, and
9030 another from library2.so named @code{bar} that prints two types of objects,
9031 @code{bar1} and @code{bar2}.
9034 (gdb) info pretty-printer
9041 (gdb) info pretty-printer library2
9046 (gdb) disable pretty-printer library1
9048 2 of 3 printers enabled
9049 (gdb) info pretty-printer
9056 (gdb) disable pretty-printer library2 bar:bar1
9058 1 of 3 printers enabled
9059 (gdb) info pretty-printer library2
9066 (gdb) disable pretty-printer library2 bar
9068 0 of 3 printers enabled
9069 (gdb) info pretty-printer library2
9078 Note that for @code{bar} the entire printer can be disabled,
9079 as can each individual subprinter.
9082 @section Value History
9084 @cindex value history
9085 @cindex history of values printed by @value{GDBN}
9086 Values printed by the @code{print} command are saved in the @value{GDBN}
9087 @dfn{value history}. This allows you to refer to them in other expressions.
9088 Values are kept until the symbol table is re-read or discarded
9089 (for example with the @code{file} or @code{symbol-file} commands).
9090 When the symbol table changes, the value history is discarded,
9091 since the values may contain pointers back to the types defined in the
9096 @cindex history number
9097 The values printed are given @dfn{history numbers} by which you can
9098 refer to them. These are successive integers starting with one.
9099 @code{print} shows you the history number assigned to a value by
9100 printing @samp{$@var{num} = } before the value; here @var{num} is the
9103 To refer to any previous value, use @samp{$} followed by the value's
9104 history number. The way @code{print} labels its output is designed to
9105 remind you of this. Just @code{$} refers to the most recent value in
9106 the history, and @code{$$} refers to the value before that.
9107 @code{$$@var{n}} refers to the @var{n}th value from the end; @code{$$2}
9108 is the value just prior to @code{$$}, @code{$$1} is equivalent to
9109 @code{$$}, and @code{$$0} is equivalent to @code{$}.
9111 For example, suppose you have just printed a pointer to a structure and
9112 want to see the contents of the structure. It suffices to type
9118 If you have a chain of structures where the component @code{next} points
9119 to the next one, you can print the contents of the next one with this:
9126 You can print successive links in the chain by repeating this
9127 command---which you can do by just typing @key{RET}.
9129 Note that the history records values, not expressions. If the value of
9130 @code{x} is 4 and you type these commands:
9138 then the value recorded in the value history by the @code{print} command
9139 remains 4 even though the value of @code{x} has changed.
9144 Print the last ten values in the value history, with their item numbers.
9145 This is like @samp{p@ $$9} repeated ten times, except that @code{show
9146 values} does not change the history.
9148 @item show values @var{n}
9149 Print ten history values centered on history item number @var{n}.
9152 Print ten history values just after the values last printed. If no more
9153 values are available, @code{show values +} produces no display.
9156 Pressing @key{RET} to repeat @code{show values @var{n}} has exactly the
9157 same effect as @samp{show values +}.
9159 @node Convenience Vars
9160 @section Convenience Variables
9162 @cindex convenience variables
9163 @cindex user-defined variables
9164 @value{GDBN} provides @dfn{convenience variables} that you can use within
9165 @value{GDBN} to hold on to a value and refer to it later. These variables
9166 exist entirely within @value{GDBN}; they are not part of your program, and
9167 setting a convenience variable has no direct effect on further execution
9168 of your program. That is why you can use them freely.
9170 Convenience variables are prefixed with @samp{$}. Any name preceded by
9171 @samp{$} can be used for a convenience variable, unless it is one of
9172 the predefined machine-specific register names (@pxref{Registers, ,Registers}).
9173 (Value history references, in contrast, are @emph{numbers} preceded
9174 by @samp{$}. @xref{Value History, ,Value History}.)
9176 You can save a value in a convenience variable with an assignment
9177 expression, just as you would set a variable in your program.
9181 set $foo = *object_ptr
9185 would save in @code{$foo} the value contained in the object pointed to by
9188 Using a convenience variable for the first time creates it, but its
9189 value is @code{void} until you assign a new value. You can alter the
9190 value with another assignment at any time.
9192 Convenience variables have no fixed types. You can assign a convenience
9193 variable any type of value, including structures and arrays, even if
9194 that variable already has a value of a different type. The convenience
9195 variable, when used as an expression, has the type of its current value.
9198 @kindex show convenience
9199 @cindex show all user variables
9200 @item show convenience
9201 Print a list of convenience variables used so far, and their values.
9202 Abbreviated @code{show conv}.
9204 @kindex init-if-undefined
9205 @cindex convenience variables, initializing
9206 @item init-if-undefined $@var{variable} = @var{expression}
9207 Set a convenience variable if it has not already been set. This is useful
9208 for user-defined commands that keep some state. It is similar, in concept,
9209 to using local static variables with initializers in C (except that
9210 convenience variables are global). It can also be used to allow users to
9211 override default values used in a command script.
9213 If the variable is already defined then the expression is not evaluated so
9214 any side-effects do not occur.
9217 One of the ways to use a convenience variable is as a counter to be
9218 incremented or a pointer to be advanced. For example, to print
9219 a field from successive elements of an array of structures:
9223 print bar[$i++]->contents
9227 Repeat that command by typing @key{RET}.
9229 Some convenience variables are created automatically by @value{GDBN} and given
9230 values likely to be useful.
9233 @vindex $_@r{, convenience variable}
9235 The variable @code{$_} is automatically set by the @code{x} command to
9236 the last address examined (@pxref{Memory, ,Examining Memory}). Other
9237 commands which provide a default address for @code{x} to examine also
9238 set @code{$_} to that address; these commands include @code{info line}
9239 and @code{info breakpoint}. The type of @code{$_} is @code{void *}
9240 except when set by the @code{x} command, in which case it is a pointer
9241 to the type of @code{$__}.
9243 @vindex $__@r{, convenience variable}
9245 The variable @code{$__} is automatically set by the @code{x} command
9246 to the value found in the last address examined. Its type is chosen
9247 to match the format in which the data was printed.
9250 @vindex $_exitcode@r{, convenience variable}
9251 The variable @code{$_exitcode} is automatically set to the exit code when
9252 the program being debugged terminates.
9255 @itemx $_probe_arg0@dots{}$_probe_arg11
9256 Arguments to a static probe. @xref{Static Probe Points}.
9259 @vindex $_sdata@r{, inspect, convenience variable}
9260 The variable @code{$_sdata} contains extra collected static tracepoint
9261 data. @xref{Tracepoint Actions,,Tracepoint Action Lists}. Note that
9262 @code{$_sdata} could be empty, if not inspecting a trace buffer, or
9263 if extra static tracepoint data has not been collected.
9266 @vindex $_siginfo@r{, convenience variable}
9267 The variable @code{$_siginfo} contains extra signal information
9268 (@pxref{extra signal information}). Note that @code{$_siginfo}
9269 could be empty, if the application has not yet received any signals.
9270 For example, it will be empty before you execute the @code{run} command.
9273 @vindex $_tlb@r{, convenience variable}
9274 The variable @code{$_tlb} is automatically set when debugging
9275 applications running on MS-Windows in native mode or connected to
9276 gdbserver that supports the @code{qGetTIBAddr} request.
9277 @xref{General Query Packets}.
9278 This variable contains the address of the thread information block.
9282 On HP-UX systems, if you refer to a function or variable name that
9283 begins with a dollar sign, @value{GDBN} searches for a user or system
9284 name first, before it searches for a convenience variable.
9286 @cindex convenience functions
9287 @value{GDBN} also supplies some @dfn{convenience functions}. These
9288 have a syntax similar to convenience variables. A convenience
9289 function can be used in an expression just like an ordinary function;
9290 however, a convenience function is implemented internally to
9295 @kindex help function
9296 @cindex show all convenience functions
9297 Print a list of all convenience functions.
9304 You can refer to machine register contents, in expressions, as variables
9305 with names starting with @samp{$}. The names of registers are different
9306 for each machine; use @code{info registers} to see the names used on
9310 @kindex info registers
9311 @item info registers
9312 Print the names and values of all registers except floating-point
9313 and vector registers (in the selected stack frame).
9315 @kindex info all-registers
9316 @cindex floating point registers
9317 @item info all-registers
9318 Print the names and values of all registers, including floating-point
9319 and vector registers (in the selected stack frame).
9321 @item info registers @var{regname} @dots{}
9322 Print the @dfn{relativized} value of each specified register @var{regname}.
9323 As discussed in detail below, register values are normally relative to
9324 the selected stack frame. @var{regname} may be any register name valid on
9325 the machine you are using, with or without the initial @samp{$}.
9328 @cindex stack pointer register
9329 @cindex program counter register
9330 @cindex process status register
9331 @cindex frame pointer register
9332 @cindex standard registers
9333 @value{GDBN} has four ``standard'' register names that are available (in
9334 expressions) on most machines---whenever they do not conflict with an
9335 architecture's canonical mnemonics for registers. The register names
9336 @code{$pc} and @code{$sp} are used for the program counter register and
9337 the stack pointer. @code{$fp} is used for a register that contains a
9338 pointer to the current stack frame, and @code{$ps} is used for a
9339 register that contains the processor status. For example,
9340 you could print the program counter in hex with
9347 or print the instruction to be executed next with
9354 or add four to the stack pointer@footnote{This is a way of removing
9355 one word from the stack, on machines where stacks grow downward in
9356 memory (most machines, nowadays). This assumes that the innermost
9357 stack frame is selected; setting @code{$sp} is not allowed when other
9358 stack frames are selected. To pop entire frames off the stack,
9359 regardless of machine architecture, use @code{return};
9360 see @ref{Returning, ,Returning from a Function}.} with
9366 Whenever possible, these four standard register names are available on
9367 your machine even though the machine has different canonical mnemonics,
9368 so long as there is no conflict. The @code{info registers} command
9369 shows the canonical names. For example, on the SPARC, @code{info
9370 registers} displays the processor status register as @code{$psr} but you
9371 can also refer to it as @code{$ps}; and on x86-based machines @code{$ps}
9372 is an alias for the @sc{eflags} register.
9374 @value{GDBN} always considers the contents of an ordinary register as an
9375 integer when the register is examined in this way. Some machines have
9376 special registers which can hold nothing but floating point; these
9377 registers are considered to have floating point values. There is no way
9378 to refer to the contents of an ordinary register as floating point value
9379 (although you can @emph{print} it as a floating point value with
9380 @samp{print/f $@var{regname}}).
9382 Some registers have distinct ``raw'' and ``virtual'' data formats. This
9383 means that the data format in which the register contents are saved by
9384 the operating system is not the same one that your program normally
9385 sees. For example, the registers of the 68881 floating point
9386 coprocessor are always saved in ``extended'' (raw) format, but all C
9387 programs expect to work with ``double'' (virtual) format. In such
9388 cases, @value{GDBN} normally works with the virtual format only (the format
9389 that makes sense for your program), but the @code{info registers} command
9390 prints the data in both formats.
9392 @cindex SSE registers (x86)
9393 @cindex MMX registers (x86)
9394 Some machines have special registers whose contents can be interpreted
9395 in several different ways. For example, modern x86-based machines
9396 have SSE and MMX registers that can hold several values packed
9397 together in several different formats. @value{GDBN} refers to such
9398 registers in @code{struct} notation:
9401 (@value{GDBP}) print $xmm1
9403 v4_float = @{0, 3.43859137e-038, 1.54142831e-044, 1.821688e-044@},
9404 v2_double = @{9.92129282474342e-303, 2.7585945287983262e-313@},
9405 v16_int8 = "\000\000\000\000\3706;\001\v\000\000\000\r\000\000",
9406 v8_int16 = @{0, 0, 14072, 315, 11, 0, 13, 0@},
9407 v4_int32 = @{0, 20657912, 11, 13@},
9408 v2_int64 = @{88725056443645952, 55834574859@},
9409 uint128 = 0x0000000d0000000b013b36f800000000
9414 To set values of such registers, you need to tell @value{GDBN} which
9415 view of the register you wish to change, as if you were assigning
9416 value to a @code{struct} member:
9419 (@value{GDBP}) set $xmm1.uint128 = 0x000000000000000000000000FFFFFFFF
9422 Normally, register values are relative to the selected stack frame
9423 (@pxref{Selection, ,Selecting a Frame}). This means that you get the
9424 value that the register would contain if all stack frames farther in
9425 were exited and their saved registers restored. In order to see the
9426 true contents of hardware registers, you must select the innermost
9427 frame (with @samp{frame 0}).
9429 However, @value{GDBN} must deduce where registers are saved, from the machine
9430 code generated by your compiler. If some registers are not saved, or if
9431 @value{GDBN} is unable to locate the saved registers, the selected stack
9432 frame makes no difference.
9434 @node Floating Point Hardware
9435 @section Floating Point Hardware
9436 @cindex floating point
9438 Depending on the configuration, @value{GDBN} may be able to give
9439 you more information about the status of the floating point hardware.
9444 Display hardware-dependent information about the floating
9445 point unit. The exact contents and layout vary depending on the
9446 floating point chip. Currently, @samp{info float} is supported on
9447 the ARM and x86 machines.
9451 @section Vector Unit
9454 Depending on the configuration, @value{GDBN} may be able to give you
9455 more information about the status of the vector unit.
9460 Display information about the vector unit. The exact contents and
9461 layout vary depending on the hardware.
9464 @node OS Information
9465 @section Operating System Auxiliary Information
9466 @cindex OS information
9468 @value{GDBN} provides interfaces to useful OS facilities that can help
9469 you debug your program.
9471 @cindex @code{ptrace} system call
9472 @cindex @code{struct user} contents
9473 When @value{GDBN} runs on a @dfn{Posix system} (such as GNU or Unix
9474 machines), it interfaces with the inferior via the @code{ptrace}
9475 system call. The operating system creates a special sata structure,
9476 called @code{struct user}, for this interface. You can use the
9477 command @code{info udot} to display the contents of this data
9483 Display the contents of the @code{struct user} maintained by the OS
9484 kernel for the program being debugged. @value{GDBN} displays the
9485 contents of @code{struct user} as a list of hex numbers, similar to
9486 the @code{examine} command.
9489 @cindex auxiliary vector
9490 @cindex vector, auxiliary
9491 Some operating systems supply an @dfn{auxiliary vector} to programs at
9492 startup. This is akin to the arguments and environment that you
9493 specify for a program, but contains a system-dependent variety of
9494 binary values that tell system libraries important details about the
9495 hardware, operating system, and process. Each value's purpose is
9496 identified by an integer tag; the meanings are well-known but system-specific.
9497 Depending on the configuration and operating system facilities,
9498 @value{GDBN} may be able to show you this information. For remote
9499 targets, this functionality may further depend on the remote stub's
9500 support of the @samp{qXfer:auxv:read} packet, see
9501 @ref{qXfer auxiliary vector read}.
9506 Display the auxiliary vector of the inferior, which can be either a
9507 live process or a core dump file. @value{GDBN} prints each tag value
9508 numerically, and also shows names and text descriptions for recognized
9509 tags. Some values in the vector are numbers, some bit masks, and some
9510 pointers to strings or other data. @value{GDBN} displays each value in the
9511 most appropriate form for a recognized tag, and in hexadecimal for
9512 an unrecognized tag.
9515 On some targets, @value{GDBN} can access operating system-specific
9516 information and show it to you. The types of information available
9517 will differ depending on the type of operating system running on the
9518 target. The mechanism used to fetch the data is described in
9519 @ref{Operating System Information}. For remote targets, this
9520 functionality depends on the remote stub's support of the
9521 @samp{qXfer:osdata:read} packet, see @ref{qXfer osdata read}.
9525 @item info os @var{infotype}
9527 Display OS information of the requested type.
9529 On @sc{gnu}/Linux, the following values of @var{infotype} are valid:
9531 @anchor{linux info os infotypes}
9533 @kindex info os processes
9535 Display the list of processes on the target. For each process,
9536 @value{GDBN} prints the process identifier, the name of the user, the
9537 command corresponding to the process, and the list of processor cores
9538 that the process is currently running on. (To understand what these
9539 properties mean, for this and the following info types, please consult
9540 the general @sc{gnu}/Linux documentation.)
9542 @kindex info os procgroups
9544 Display the list of process groups on the target. For each process,
9545 @value{GDBN} prints the identifier of the process group that it belongs
9546 to, the command corresponding to the process group leader, the process
9547 identifier, and the command line of the process. The list is sorted
9548 first by the process group identifier, then by the process identifier,
9549 so that processes belonging to the same process group are grouped together
9550 and the process group leader is listed first.
9552 @kindex info os threads
9554 Display the list of threads running on the target. For each thread,
9555 @value{GDBN} prints the identifier of the process that the thread
9556 belongs to, the command of the process, the thread identifier, and the
9557 processor core that it is currently running on. The main thread of a
9558 process is not listed.
9560 @kindex info os files
9562 Display the list of open file descriptors on the target. For each
9563 file descriptor, @value{GDBN} prints the identifier of the process
9564 owning the descriptor, the command of the owning process, the value
9565 of the descriptor, and the target of the descriptor.
9567 @kindex info os sockets
9569 Display the list of Internet-domain sockets on the target. For each
9570 socket, @value{GDBN} prints the address and port of the local and
9571 remote endpoints, the current state of the connection, the creator of
9572 the socket, the IP address family of the socket, and the type of the
9577 Display the list of all System V shared-memory regions on the target.
9578 For each shared-memory region, @value{GDBN} prints the region key,
9579 the shared-memory identifier, the access permissions, the size of the
9580 region, the process that created the region, the process that last
9581 attached to or detached from the region, the current number of live
9582 attaches to the region, and the times at which the region was last
9583 attached to, detach from, and changed.
9585 @kindex info os semaphores
9587 Display the list of all System V semaphore sets on the target. For each
9588 semaphore set, @value{GDBN} prints the semaphore set key, the semaphore
9589 set identifier, the access permissions, the number of semaphores in the
9590 set, the user and group of the owner and creator of the semaphore set,
9591 and the times at which the semaphore set was operated upon and changed.
9595 Display the list of all System V message queues on the target. For each
9596 message queue, @value{GDBN} prints the message queue key, the message
9597 queue identifier, the access permissions, the current number of bytes
9598 on the queue, the current number of messages on the queue, the processes
9599 that last sent and received a message on the queue, the user and group
9600 of the owner and creator of the message queue, the times at which a
9601 message was last sent and received on the queue, and the time at which
9602 the message queue was last changed.
9604 @kindex info os modules
9606 Display the list of all loaded kernel modules on the target. For each
9607 module, @value{GDBN} prints the module name, the size of the module in
9608 bytes, the number of times the module is used, the dependencies of the
9609 module, the status of the module, and the address of the loaded module
9614 If @var{infotype} is omitted, then list the possible values for
9615 @var{infotype} and the kind of OS information available for each
9616 @var{infotype}. If the target does not return a list of possible
9617 types, this command will report an error.
9620 @node Memory Region Attributes
9621 @section Memory Region Attributes
9622 @cindex memory region attributes
9624 @dfn{Memory region attributes} allow you to describe special handling
9625 required by regions of your target's memory. @value{GDBN} uses
9626 attributes to determine whether to allow certain types of memory
9627 accesses; whether to use specific width accesses; and whether to cache
9628 target memory. By default the description of memory regions is
9629 fetched from the target (if the current target supports this), but the
9630 user can override the fetched regions.
9632 Defined memory regions can be individually enabled and disabled. When a
9633 memory region is disabled, @value{GDBN} uses the default attributes when
9634 accessing memory in that region. Similarly, if no memory regions have
9635 been defined, @value{GDBN} uses the default attributes when accessing
9638 When a memory region is defined, it is given a number to identify it;
9639 to enable, disable, or remove a memory region, you specify that number.
9643 @item mem @var{lower} @var{upper} @var{attributes}@dots{}
9644 Define a memory region bounded by @var{lower} and @var{upper} with
9645 attributes @var{attributes}@dots{}, and add it to the list of regions
9646 monitored by @value{GDBN}. Note that @var{upper} == 0 is a special
9647 case: it is treated as the target's maximum memory address.
9648 (0xffff on 16 bit targets, 0xffffffff on 32 bit targets, etc.)
9651 Discard any user changes to the memory regions and use target-supplied
9652 regions, if available, or no regions if the target does not support.
9655 @item delete mem @var{nums}@dots{}
9656 Remove memory regions @var{nums}@dots{} from the list of regions
9657 monitored by @value{GDBN}.
9660 @item disable mem @var{nums}@dots{}
9661 Disable monitoring of memory regions @var{nums}@dots{}.
9662 A disabled memory region is not forgotten.
9663 It may be enabled again later.
9666 @item enable mem @var{nums}@dots{}
9667 Enable monitoring of memory regions @var{nums}@dots{}.
9671 Print a table of all defined memory regions, with the following columns
9675 @item Memory Region Number
9676 @item Enabled or Disabled.
9677 Enabled memory regions are marked with @samp{y}.
9678 Disabled memory regions are marked with @samp{n}.
9681 The address defining the inclusive lower bound of the memory region.
9684 The address defining the exclusive upper bound of the memory region.
9687 The list of attributes set for this memory region.
9692 @subsection Attributes
9694 @subsubsection Memory Access Mode
9695 The access mode attributes set whether @value{GDBN} may make read or
9696 write accesses to a memory region.
9698 While these attributes prevent @value{GDBN} from performing invalid
9699 memory accesses, they do nothing to prevent the target system, I/O DMA,
9700 etc.@: from accessing memory.
9704 Memory is read only.
9706 Memory is write only.
9708 Memory is read/write. This is the default.
9711 @subsubsection Memory Access Size
9712 The access size attribute tells @value{GDBN} to use specific sized
9713 accesses in the memory region. Often memory mapped device registers
9714 require specific sized accesses. If no access size attribute is
9715 specified, @value{GDBN} may use accesses of any size.
9719 Use 8 bit memory accesses.
9721 Use 16 bit memory accesses.
9723 Use 32 bit memory accesses.
9725 Use 64 bit memory accesses.
9728 @c @subsubsection Hardware/Software Breakpoints
9729 @c The hardware/software breakpoint attributes set whether @value{GDBN}
9730 @c will use hardware or software breakpoints for the internal breakpoints
9731 @c used by the step, next, finish, until, etc. commands.
9735 @c Always use hardware breakpoints
9736 @c @item swbreak (default)
9739 @subsubsection Data Cache
9740 The data cache attributes set whether @value{GDBN} will cache target
9741 memory. While this generally improves performance by reducing debug
9742 protocol overhead, it can lead to incorrect results because @value{GDBN}
9743 does not know about volatile variables or memory mapped device
9748 Enable @value{GDBN} to cache target memory.
9750 Disable @value{GDBN} from caching target memory. This is the default.
9753 @subsection Memory Access Checking
9754 @value{GDBN} can be instructed to refuse accesses to memory that is
9755 not explicitly described. This can be useful if accessing such
9756 regions has undesired effects for a specific target, or to provide
9757 better error checking. The following commands control this behaviour.
9760 @kindex set mem inaccessible-by-default
9761 @item set mem inaccessible-by-default [on|off]
9762 If @code{on} is specified, make @value{GDBN} treat memory not
9763 explicitly described by the memory ranges as non-existent and refuse accesses
9764 to such memory. The checks are only performed if there's at least one
9765 memory range defined. If @code{off} is specified, make @value{GDBN}
9766 treat the memory not explicitly described by the memory ranges as RAM.
9767 The default value is @code{on}.
9768 @kindex show mem inaccessible-by-default
9769 @item show mem inaccessible-by-default
9770 Show the current handling of accesses to unknown memory.
9774 @c @subsubsection Memory Write Verification
9775 @c The memory write verification attributes set whether @value{GDBN}
9776 @c will re-reads data after each write to verify the write was successful.
9780 @c @item noverify (default)
9783 @node Dump/Restore Files
9784 @section Copy Between Memory and a File
9785 @cindex dump/restore files
9786 @cindex append data to a file
9787 @cindex dump data to a file
9788 @cindex restore data from a file
9790 You can use the commands @code{dump}, @code{append}, and
9791 @code{restore} to copy data between target memory and a file. The
9792 @code{dump} and @code{append} commands write data to a file, and the
9793 @code{restore} command reads data from a file back into the inferior's
9794 memory. Files may be in binary, Motorola S-record, Intel hex, or
9795 Tektronix Hex format; however, @value{GDBN} can only append to binary
9801 @item dump @r{[}@var{format}@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
9802 @itemx dump @r{[}@var{format}@r{]} value @var{filename} @var{expr}
9803 Dump the contents of memory from @var{start_addr} to @var{end_addr},
9804 or the value of @var{expr}, to @var{filename} in the given format.
9806 The @var{format} parameter may be any one of:
9813 Motorola S-record format.
9815 Tektronix Hex format.
9818 @value{GDBN} uses the same definitions of these formats as the
9819 @sc{gnu} binary utilities, like @samp{objdump} and @samp{objcopy}. If
9820 @var{format} is omitted, @value{GDBN} dumps the data in raw binary
9824 @item append @r{[}binary@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
9825 @itemx append @r{[}binary@r{]} value @var{filename} @var{expr}
9826 Append the contents of memory from @var{start_addr} to @var{end_addr},
9827 or the value of @var{expr}, to the file @var{filename}, in raw binary form.
9828 (@value{GDBN} can only append data to files in raw binary form.)
9831 @item restore @var{filename} @r{[}binary@r{]} @var{bias} @var{start} @var{end}
9832 Restore the contents of file @var{filename} into memory. The
9833 @code{restore} command can automatically recognize any known @sc{bfd}
9834 file format, except for raw binary. To restore a raw binary file you
9835 must specify the optional keyword @code{binary} after the filename.
9837 If @var{bias} is non-zero, its value will be added to the addresses
9838 contained in the file. Binary files always start at address zero, so
9839 they will be restored at address @var{bias}. Other bfd files have
9840 a built-in location; they will be restored at offset @var{bias}
9843 If @var{start} and/or @var{end} are non-zero, then only data between
9844 file offset @var{start} and file offset @var{end} will be restored.
9845 These offsets are relative to the addresses in the file, before
9846 the @var{bias} argument is applied.
9850 @node Core File Generation
9851 @section How to Produce a Core File from Your Program
9852 @cindex dump core from inferior
9854 A @dfn{core file} or @dfn{core dump} is a file that records the memory
9855 image of a running process and its process status (register values
9856 etc.). Its primary use is post-mortem debugging of a program that
9857 crashed while it ran outside a debugger. A program that crashes
9858 automatically produces a core file, unless this feature is disabled by
9859 the user. @xref{Files}, for information on invoking @value{GDBN} in
9860 the post-mortem debugging mode.
9862 Occasionally, you may wish to produce a core file of the program you
9863 are debugging in order to preserve a snapshot of its state.
9864 @value{GDBN} has a special command for that.
9868 @kindex generate-core-file
9869 @item generate-core-file [@var{file}]
9870 @itemx gcore [@var{file}]
9871 Produce a core dump of the inferior process. The optional argument
9872 @var{file} specifies the file name where to put the core dump. If not
9873 specified, the file name defaults to @file{core.@var{pid}}, where
9874 @var{pid} is the inferior process ID.
9876 Note that this command is implemented only for some systems (as of
9877 this writing, @sc{gnu}/Linux, FreeBSD, Solaris, Unixware, and S390).
9880 @node Character Sets
9881 @section Character Sets
9882 @cindex character sets
9884 @cindex translating between character sets
9885 @cindex host character set
9886 @cindex target character set
9888 If the program you are debugging uses a different character set to
9889 represent characters and strings than the one @value{GDBN} uses itself,
9890 @value{GDBN} can automatically translate between the character sets for
9891 you. The character set @value{GDBN} uses we call the @dfn{host
9892 character set}; the one the inferior program uses we call the
9893 @dfn{target character set}.
9895 For example, if you are running @value{GDBN} on a @sc{gnu}/Linux system, which
9896 uses the ISO Latin 1 character set, but you are using @value{GDBN}'s
9897 remote protocol (@pxref{Remote Debugging}) to debug a program
9898 running on an IBM mainframe, which uses the @sc{ebcdic} character set,
9899 then the host character set is Latin-1, and the target character set is
9900 @sc{ebcdic}. If you give @value{GDBN} the command @code{set
9901 target-charset EBCDIC-US}, then @value{GDBN} translates between
9902 @sc{ebcdic} and Latin 1 as you print character or string values, or use
9903 character and string literals in expressions.
9905 @value{GDBN} has no way to automatically recognize which character set
9906 the inferior program uses; you must tell it, using the @code{set
9907 target-charset} command, described below.
9909 Here are the commands for controlling @value{GDBN}'s character set
9913 @item set target-charset @var{charset}
9914 @kindex set target-charset
9915 Set the current target character set to @var{charset}. To display the
9916 list of supported target character sets, type
9917 @kbd{@w{set target-charset @key{TAB}@key{TAB}}}.
9919 @item set host-charset @var{charset}
9920 @kindex set host-charset
9921 Set the current host character set to @var{charset}.
9923 By default, @value{GDBN} uses a host character set appropriate to the
9924 system it is running on; you can override that default using the
9925 @code{set host-charset} command. On some systems, @value{GDBN} cannot
9926 automatically determine the appropriate host character set. In this
9927 case, @value{GDBN} uses @samp{UTF-8}.
9929 @value{GDBN} can only use certain character sets as its host character
9930 set. If you type @kbd{@w{set host-charset @key{TAB}@key{TAB}}},
9931 @value{GDBN} will list the host character sets it supports.
9933 @item set charset @var{charset}
9935 Set the current host and target character sets to @var{charset}. As
9936 above, if you type @kbd{@w{set charset @key{TAB}@key{TAB}}},
9937 @value{GDBN} will list the names of the character sets that can be used
9938 for both host and target.
9941 @kindex show charset
9942 Show the names of the current host and target character sets.
9944 @item show host-charset
9945 @kindex show host-charset
9946 Show the name of the current host character set.
9948 @item show target-charset
9949 @kindex show target-charset
9950 Show the name of the current target character set.
9952 @item set target-wide-charset @var{charset}
9953 @kindex set target-wide-charset
9954 Set the current target's wide character set to @var{charset}. This is
9955 the character set used by the target's @code{wchar_t} type. To
9956 display the list of supported wide character sets, type
9957 @kbd{@w{set target-wide-charset @key{TAB}@key{TAB}}}.
9959 @item show target-wide-charset
9960 @kindex show target-wide-charset
9961 Show the name of the current target's wide character set.
9964 Here is an example of @value{GDBN}'s character set support in action.
9965 Assume that the following source code has been placed in the file
9966 @file{charset-test.c}:
9972 = @{72, 101, 108, 108, 111, 44, 32, 119,
9973 111, 114, 108, 100, 33, 10, 0@};
9974 char ibm1047_hello[]
9975 = @{200, 133, 147, 147, 150, 107, 64, 166,
9976 150, 153, 147, 132, 90, 37, 0@};
9980 printf ("Hello, world!\n");
9984 In this program, @code{ascii_hello} and @code{ibm1047_hello} are arrays
9985 containing the string @samp{Hello, world!} followed by a newline,
9986 encoded in the @sc{ascii} and @sc{ibm1047} character sets.
9988 We compile the program, and invoke the debugger on it:
9991 $ gcc -g charset-test.c -o charset-test
9992 $ gdb -nw charset-test
9993 GNU gdb 2001-12-19-cvs
9994 Copyright 2001 Free Software Foundation, Inc.
9999 We can use the @code{show charset} command to see what character sets
10000 @value{GDBN} is currently using to interpret and display characters and
10004 (@value{GDBP}) show charset
10005 The current host and target character set is `ISO-8859-1'.
10009 For the sake of printing this manual, let's use @sc{ascii} as our
10010 initial character set:
10012 (@value{GDBP}) set charset ASCII
10013 (@value{GDBP}) show charset
10014 The current host and target character set is `ASCII'.
10018 Let's assume that @sc{ascii} is indeed the correct character set for our
10019 host system --- in other words, let's assume that if @value{GDBN} prints
10020 characters using the @sc{ascii} character set, our terminal will display
10021 them properly. Since our current target character set is also
10022 @sc{ascii}, the contents of @code{ascii_hello} print legibly:
10025 (@value{GDBP}) print ascii_hello
10026 $1 = 0x401698 "Hello, world!\n"
10027 (@value{GDBP}) print ascii_hello[0]
10032 @value{GDBN} uses the target character set for character and string
10033 literals you use in expressions:
10036 (@value{GDBP}) print '+'
10041 The @sc{ascii} character set uses the number 43 to encode the @samp{+}
10044 @value{GDBN} relies on the user to tell it which character set the
10045 target program uses. If we print @code{ibm1047_hello} while our target
10046 character set is still @sc{ascii}, we get jibberish:
10049 (@value{GDBP}) print ibm1047_hello
10050 $4 = 0x4016a8 "\310\205\223\223\226k@@\246\226\231\223\204Z%"
10051 (@value{GDBP}) print ibm1047_hello[0]
10056 If we invoke the @code{set target-charset} followed by @key{TAB}@key{TAB},
10057 @value{GDBN} tells us the character sets it supports:
10060 (@value{GDBP}) set target-charset
10061 ASCII EBCDIC-US IBM1047 ISO-8859-1
10062 (@value{GDBP}) set target-charset
10065 We can select @sc{ibm1047} as our target character set, and examine the
10066 program's strings again. Now the @sc{ascii} string is wrong, but
10067 @value{GDBN} translates the contents of @code{ibm1047_hello} from the
10068 target character set, @sc{ibm1047}, to the host character set,
10069 @sc{ascii}, and they display correctly:
10072 (@value{GDBP}) set target-charset IBM1047
10073 (@value{GDBP}) show charset
10074 The current host character set is `ASCII'.
10075 The current target character set is `IBM1047'.
10076 (@value{GDBP}) print ascii_hello
10077 $6 = 0x401698 "\110\145%%?\054\040\167?\162%\144\041\012"
10078 (@value{GDBP}) print ascii_hello[0]
10080 (@value{GDBP}) print ibm1047_hello
10081 $8 = 0x4016a8 "Hello, world!\n"
10082 (@value{GDBP}) print ibm1047_hello[0]
10087 As above, @value{GDBN} uses the target character set for character and
10088 string literals you use in expressions:
10091 (@value{GDBP}) print '+'
10096 The @sc{ibm1047} character set uses the number 78 to encode the @samp{+}
10099 @node Caching Remote Data
10100 @section Caching Data of Remote Targets
10101 @cindex caching data of remote targets
10103 @value{GDBN} caches data exchanged between the debugger and a
10104 remote target (@pxref{Remote Debugging}). Such caching generally improves
10105 performance, because it reduces the overhead of the remote protocol by
10106 bundling memory reads and writes into large chunks. Unfortunately, simply
10107 caching everything would lead to incorrect results, since @value{GDBN}
10108 does not necessarily know anything about volatile values, memory-mapped I/O
10109 addresses, etc. Furthermore, in non-stop mode (@pxref{Non-Stop Mode})
10110 memory can be changed @emph{while} a gdb command is executing.
10111 Therefore, by default, @value{GDBN} only caches data
10112 known to be on the stack@footnote{In non-stop mode, it is moderately
10113 rare for a running thread to modify the stack of a stopped thread
10114 in a way that would interfere with a backtrace, and caching of
10115 stack reads provides a significant speed up of remote backtraces.}.
10116 Other regions of memory can be explicitly marked as
10117 cacheable; see @pxref{Memory Region Attributes}.
10120 @kindex set remotecache
10121 @item set remotecache on
10122 @itemx set remotecache off
10123 This option no longer does anything; it exists for compatibility
10126 @kindex show remotecache
10127 @item show remotecache
10128 Show the current state of the obsolete remotecache flag.
10130 @kindex set stack-cache
10131 @item set stack-cache on
10132 @itemx set stack-cache off
10133 Enable or disable caching of stack accesses. When @code{ON}, use
10134 caching. By default, this option is @code{ON}.
10136 @kindex show stack-cache
10137 @item show stack-cache
10138 Show the current state of data caching for memory accesses.
10140 @kindex info dcache
10141 @item info dcache @r{[}line@r{]}
10142 Print the information about the data cache performance. The
10143 information displayed includes the dcache width and depth, and for
10144 each cache line, its number, address, and how many times it was
10145 referenced. This command is useful for debugging the data cache
10148 If a line number is specified, the contents of that line will be
10151 @item set dcache size @var{size}
10152 @cindex dcache size
10153 @kindex set dcache size
10154 Set maximum number of entries in dcache (dcache depth above).
10156 @item set dcache line-size @var{line-size}
10157 @cindex dcache line-size
10158 @kindex set dcache line-size
10159 Set number of bytes each dcache entry caches (dcache width above).
10160 Must be a power of 2.
10162 @item show dcache size
10163 @kindex show dcache size
10164 Show maximum number of dcache entries. See also @ref{Caching Remote Data, info dcache}.
10166 @item show dcache line-size
10167 @kindex show dcache line-size
10168 Show default size of dcache lines. See also @ref{Caching Remote Data, info dcache}.
10172 @node Searching Memory
10173 @section Search Memory
10174 @cindex searching memory
10176 Memory can be searched for a particular sequence of bytes with the
10177 @code{find} command.
10181 @item find @r{[}/@var{sn}@r{]} @var{start_addr}, +@var{len}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
10182 @itemx find @r{[}/@var{sn}@r{]} @var{start_addr}, @var{end_addr}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
10183 Search memory for the sequence of bytes specified by @var{val1}, @var{val2},
10184 etc. The search begins at address @var{start_addr} and continues for either
10185 @var{len} bytes or through to @var{end_addr} inclusive.
10188 @var{s} and @var{n} are optional parameters.
10189 They may be specified in either order, apart or together.
10192 @item @var{s}, search query size
10193 The size of each search query value.
10199 halfwords (two bytes)
10203 giant words (eight bytes)
10206 All values are interpreted in the current language.
10207 This means, for example, that if the current source language is C/C@t{++}
10208 then searching for the string ``hello'' includes the trailing '\0'.
10210 If the value size is not specified, it is taken from the
10211 value's type in the current language.
10212 This is useful when one wants to specify the search
10213 pattern as a mixture of types.
10214 Note that this means, for example, that in the case of C-like languages
10215 a search for an untyped 0x42 will search for @samp{(int) 0x42}
10216 which is typically four bytes.
10218 @item @var{n}, maximum number of finds
10219 The maximum number of matches to print. The default is to print all finds.
10222 You can use strings as search values. Quote them with double-quotes
10224 The string value is copied into the search pattern byte by byte,
10225 regardless of the endianness of the target and the size specification.
10227 The address of each match found is printed as well as a count of the
10228 number of matches found.
10230 The address of the last value found is stored in convenience variable
10232 A count of the number of matches is stored in @samp{$numfound}.
10234 For example, if stopped at the @code{printf} in this function:
10240 static char hello[] = "hello-hello";
10241 static struct @{ char c; short s; int i; @}
10242 __attribute__ ((packed)) mixed
10243 = @{ 'c', 0x1234, 0x87654321 @};
10244 printf ("%s\n", hello);
10249 you get during debugging:
10252 (gdb) find &hello[0], +sizeof(hello), "hello"
10253 0x804956d <hello.1620+6>
10255 (gdb) find &hello[0], +sizeof(hello), 'h', 'e', 'l', 'l', 'o'
10256 0x8049567 <hello.1620>
10257 0x804956d <hello.1620+6>
10259 (gdb) find /b1 &hello[0], +sizeof(hello), 'h', 0x65, 'l'
10260 0x8049567 <hello.1620>
10262 (gdb) find &mixed, +sizeof(mixed), (char) 'c', (short) 0x1234, (int) 0x87654321
10263 0x8049560 <mixed.1625>
10265 (gdb) print $numfound
10268 $2 = (void *) 0x8049560
10271 @node Optimized Code
10272 @chapter Debugging Optimized Code
10273 @cindex optimized code, debugging
10274 @cindex debugging optimized code
10276 Almost all compilers support optimization. With optimization
10277 disabled, the compiler generates assembly code that corresponds
10278 directly to your source code, in a simplistic way. As the compiler
10279 applies more powerful optimizations, the generated assembly code
10280 diverges from your original source code. With help from debugging
10281 information generated by the compiler, @value{GDBN} can map from
10282 the running program back to constructs from your original source.
10284 @value{GDBN} is more accurate with optimization disabled. If you
10285 can recompile without optimization, it is easier to follow the
10286 progress of your program during debugging. But, there are many cases
10287 where you may need to debug an optimized version.
10289 When you debug a program compiled with @samp{-g -O}, remember that the
10290 optimizer has rearranged your code; the debugger shows you what is
10291 really there. Do not be too surprised when the execution path does not
10292 exactly match your source file! An extreme example: if you define a
10293 variable, but never use it, @value{GDBN} never sees that
10294 variable---because the compiler optimizes it out of existence.
10296 Some things do not work as well with @samp{-g -O} as with just
10297 @samp{-g}, particularly on machines with instruction scheduling. If in
10298 doubt, recompile with @samp{-g} alone, and if this fixes the problem,
10299 please report it to us as a bug (including a test case!).
10300 @xref{Variables}, for more information about debugging optimized code.
10303 * Inline Functions:: How @value{GDBN} presents inlining
10304 * Tail Call Frames:: @value{GDBN} analysis of jumps to functions
10307 @node Inline Functions
10308 @section Inline Functions
10309 @cindex inline functions, debugging
10311 @dfn{Inlining} is an optimization that inserts a copy of the function
10312 body directly at each call site, instead of jumping to a shared
10313 routine. @value{GDBN} displays inlined functions just like
10314 non-inlined functions. They appear in backtraces. You can view their
10315 arguments and local variables, step into them with @code{step}, skip
10316 them with @code{next}, and escape from them with @code{finish}.
10317 You can check whether a function was inlined by using the
10318 @code{info frame} command.
10320 For @value{GDBN} to support inlined functions, the compiler must
10321 record information about inlining in the debug information ---
10322 @value{NGCC} using the @sc{dwarf 2} format does this, and several
10323 other compilers do also. @value{GDBN} only supports inlined functions
10324 when using @sc{dwarf 2}. Versions of @value{NGCC} before 4.1
10325 do not emit two required attributes (@samp{DW_AT_call_file} and
10326 @samp{DW_AT_call_line}); @value{GDBN} does not display inlined
10327 function calls with earlier versions of @value{NGCC}. It instead
10328 displays the arguments and local variables of inlined functions as
10329 local variables in the caller.
10331 The body of an inlined function is directly included at its call site;
10332 unlike a non-inlined function, there are no instructions devoted to
10333 the call. @value{GDBN} still pretends that the call site and the
10334 start of the inlined function are different instructions. Stepping to
10335 the call site shows the call site, and then stepping again shows
10336 the first line of the inlined function, even though no additional
10337 instructions are executed.
10339 This makes source-level debugging much clearer; you can see both the
10340 context of the call and then the effect of the call. Only stepping by
10341 a single instruction using @code{stepi} or @code{nexti} does not do
10342 this; single instruction steps always show the inlined body.
10344 There are some ways that @value{GDBN} does not pretend that inlined
10345 function calls are the same as normal calls:
10349 Setting breakpoints at the call site of an inlined function may not
10350 work, because the call site does not contain any code. @value{GDBN}
10351 may incorrectly move the breakpoint to the next line of the enclosing
10352 function, after the call. This limitation will be removed in a future
10353 version of @value{GDBN}; until then, set a breakpoint on an earlier line
10354 or inside the inlined function instead.
10357 @value{GDBN} cannot locate the return value of inlined calls after
10358 using the @code{finish} command. This is a limitation of compiler-generated
10359 debugging information; after @code{finish}, you can step to the next line
10360 and print a variable where your program stored the return value.
10364 @node Tail Call Frames
10365 @section Tail Call Frames
10366 @cindex tail call frames, debugging
10368 Function @code{B} can call function @code{C} in its very last statement. In
10369 unoptimized compilation the call of @code{C} is immediately followed by return
10370 instruction at the end of @code{B} code. Optimizing compiler may replace the
10371 call and return in function @code{B} into one jump to function @code{C}
10372 instead. Such use of a jump instruction is called @dfn{tail call}.
10374 During execution of function @code{C}, there will be no indication in the
10375 function call stack frames that it was tail-called from @code{B}. If function
10376 @code{A} regularly calls function @code{B} which tail-calls function @code{C},
10377 then @value{GDBN} will see @code{A} as the caller of @code{C}. However, in
10378 some cases @value{GDBN} can determine that @code{C} was tail-called from
10379 @code{B}, and it will then create fictitious call frame for that, with the
10380 return address set up as if @code{B} called @code{C} normally.
10382 This functionality is currently supported only by DWARF 2 debugging format and
10383 the compiler has to produce @samp{DW_TAG_GNU_call_site} tags. With
10384 @value{NGCC}, you need to specify @option{-O -g} during compilation, to get
10387 @kbd{info frame} command (@pxref{Frame Info}) will indicate the tail call frame
10388 kind by text @code{tail call frame} such as in this sample @value{GDBN} output:
10392 0x40066b <b(int, double)+11>: jmp 0x400640 <c(int, double)>
10394 Stack level 1, frame at 0x7fffffffda30:
10395 rip = 0x40066d in b (amd64-entry-value.cc:59); saved rip 0x4004c5
10396 tail call frame, caller of frame at 0x7fffffffda30
10397 source language c++.
10398 Arglist at unknown address.
10399 Locals at unknown address, Previous frame's sp is 0x7fffffffda30
10402 The detection of all the possible code path executions can find them ambiguous.
10403 There is no execution history stored (possible @ref{Reverse Execution} is never
10404 used for this purpose) and the last known caller could have reached the known
10405 callee by multiple different jump sequences. In such case @value{GDBN} still
10406 tries to show at least all the unambiguous top tail callers and all the
10407 unambiguous bottom tail calees, if any.
10410 @anchor{set debug entry-values}
10411 @item set debug entry-values
10412 @kindex set debug entry-values
10413 When set to on, enables printing of analysis messages for both frame argument
10414 values at function entry and tail calls. It will show all the possible valid
10415 tail calls code paths it has considered. It will also print the intersection
10416 of them with the final unambiguous (possibly partial or even empty) code path
10419 @item show debug entry-values
10420 @kindex show debug entry-values
10421 Show the current state of analysis messages printing for both frame argument
10422 values at function entry and tail calls.
10425 The analysis messages for tail calls can for example show why the virtual tail
10426 call frame for function @code{c} has not been recognized (due to the indirect
10427 reference by variable @code{x}):
10430 static void __attribute__((noinline, noclone)) c (void);
10431 void (*x) (void) = c;
10432 static void __attribute__((noinline, noclone)) a (void) @{ x++; @}
10433 static void __attribute__((noinline, noclone)) c (void) @{ a (); @}
10434 int main (void) @{ x (); return 0; @}
10436 Breakpoint 1, DW_OP_GNU_entry_value resolving cannot find
10437 DW_TAG_GNU_call_site 0x40039a in main
10439 3 static void __attribute__((noinline, noclone)) a (void) @{ x++; @}
10442 #1 0x000000000040039a in main () at t.c:5
10445 Another possibility is an ambiguous virtual tail call frames resolution:
10449 static void __attribute__((noinline, noclone)) f (void) @{ i++; @}
10450 static void __attribute__((noinline, noclone)) e (void) @{ f (); @}
10451 static void __attribute__((noinline, noclone)) d (void) @{ f (); @}
10452 static void __attribute__((noinline, noclone)) c (void) @{ d (); @}
10453 static void __attribute__((noinline, noclone)) b (void)
10454 @{ if (i) c (); else e (); @}
10455 static void __attribute__((noinline, noclone)) a (void) @{ b (); @}
10456 int main (void) @{ a (); return 0; @}
10458 tailcall: initial: 0x4004d2(a) 0x4004ce(b) 0x4004b2(c) 0x4004a2(d)
10459 tailcall: compare: 0x4004d2(a) 0x4004cc(b) 0x400492(e)
10460 tailcall: reduced: 0x4004d2(a) |
10463 #1 0x00000000004004d2 in a () at t.c:8
10464 #2 0x0000000000400395 in main () at t.c:9
10467 @set CALLSEQ1A @code{main@value{ARROW}a@value{ARROW}b@value{ARROW}c@value{ARROW}d@value{ARROW}f}
10468 @set CALLSEQ2A @code{main@value{ARROW}a@value{ARROW}b@value{ARROW}e@value{ARROW}f}
10470 @c Convert CALLSEQ#A to CALLSEQ#B depending on HAVE_MAKEINFO_CLICK.
10471 @ifset HAVE_MAKEINFO_CLICK
10472 @set ARROW @click{}
10473 @set CALLSEQ1B @clicksequence{@value{CALLSEQ1A}}
10474 @set CALLSEQ2B @clicksequence{@value{CALLSEQ2A}}
10476 @ifclear HAVE_MAKEINFO_CLICK
10478 @set CALLSEQ1B @value{CALLSEQ1A}
10479 @set CALLSEQ2B @value{CALLSEQ2A}
10482 Frames #0 and #2 are real, #1 is a virtual tail call frame.
10483 The code can have possible execution paths @value{CALLSEQ1B} or
10484 @value{CALLSEQ2B}, @value{GDBN} cannot find which one from the inferior state.
10486 @code{initial:} state shows some random possible calling sequence @value{GDBN}
10487 has found. It then finds another possible calling sequcen - that one is
10488 prefixed by @code{compare:}. The non-ambiguous intersection of these two is
10489 printed as the @code{reduced:} calling sequence. That one could have many
10490 futher @code{compare:} and @code{reduced:} statements as long as there remain
10491 any non-ambiguous sequence entries.
10493 For the frame of function @code{b} in both cases there are different possible
10494 @code{$pc} values (@code{0x4004cc} or @code{0x4004ce}), therefore this frame is
10495 also ambigous. The only non-ambiguous frame is the one for function @code{a},
10496 therefore this one is displayed to the user while the ambiguous frames are
10499 There can be also reasons why printing of frame argument values at function
10504 static void __attribute__((noinline, noclone)) c (int i) @{ v++; @}
10505 static void __attribute__((noinline, noclone)) a (int i);
10506 static void __attribute__((noinline, noclone)) b (int i) @{ a (i); @}
10507 static void __attribute__((noinline, noclone)) a (int i)
10508 @{ if (i) b (i - 1); else c (0); @}
10509 int main (void) @{ a (5); return 0; @}
10512 #0 c (i=i@@entry=0) at t.c:2
10513 #1 0x0000000000400428 in a (DW_OP_GNU_entry_value resolving has found
10514 function "a" at 0x400420 can call itself via tail calls
10515 i=<optimized out>) at t.c:6
10516 #2 0x000000000040036e in main () at t.c:7
10519 @value{GDBN} cannot find out from the inferior state if and how many times did
10520 function @code{a} call itself (via function @code{b}) as these calls would be
10521 tail calls. Such tail calls would modify thue @code{i} variable, therefore
10522 @value{GDBN} cannot be sure the value it knows would be right - @value{GDBN}
10523 prints @code{<optimized out>} instead.
10526 @chapter C Preprocessor Macros
10528 Some languages, such as C and C@t{++}, provide a way to define and invoke
10529 ``preprocessor macros'' which expand into strings of tokens.
10530 @value{GDBN} can evaluate expressions containing macro invocations, show
10531 the result of macro expansion, and show a macro's definition, including
10532 where it was defined.
10534 You may need to compile your program specially to provide @value{GDBN}
10535 with information about preprocessor macros. Most compilers do not
10536 include macros in their debugging information, even when you compile
10537 with the @option{-g} flag. @xref{Compilation}.
10539 A program may define a macro at one point, remove that definition later,
10540 and then provide a different definition after that. Thus, at different
10541 points in the program, a macro may have different definitions, or have
10542 no definition at all. If there is a current stack frame, @value{GDBN}
10543 uses the macros in scope at that frame's source code line. Otherwise,
10544 @value{GDBN} uses the macros in scope at the current listing location;
10547 Whenever @value{GDBN} evaluates an expression, it always expands any
10548 macro invocations present in the expression. @value{GDBN} also provides
10549 the following commands for working with macros explicitly.
10553 @kindex macro expand
10554 @cindex macro expansion, showing the results of preprocessor
10555 @cindex preprocessor macro expansion, showing the results of
10556 @cindex expanding preprocessor macros
10557 @item macro expand @var{expression}
10558 @itemx macro exp @var{expression}
10559 Show the results of expanding all preprocessor macro invocations in
10560 @var{expression}. Since @value{GDBN} simply expands macros, but does
10561 not parse the result, @var{expression} need not be a valid expression;
10562 it can be any string of tokens.
10565 @item macro expand-once @var{expression}
10566 @itemx macro exp1 @var{expression}
10567 @cindex expand macro once
10568 @i{(This command is not yet implemented.)} Show the results of
10569 expanding those preprocessor macro invocations that appear explicitly in
10570 @var{expression}. Macro invocations appearing in that expansion are
10571 left unchanged. This command allows you to see the effect of a
10572 particular macro more clearly, without being confused by further
10573 expansions. Since @value{GDBN} simply expands macros, but does not
10574 parse the result, @var{expression} need not be a valid expression; it
10575 can be any string of tokens.
10578 @cindex macro definition, showing
10579 @cindex definition of a macro, showing
10580 @cindex macros, from debug info
10581 @item info macro [-a|-all] [--] @var{macro}
10582 Show the current definition or all definitions of the named @var{macro},
10583 and describe the source location or compiler command-line where that
10584 definition was established. The optional double dash is to signify the end of
10585 argument processing and the beginning of @var{macro} for non C-like macros where
10586 the macro may begin with a hyphen.
10588 @kindex info macros
10589 @item info macros @var{linespec}
10590 Show all macro definitions that are in effect at the location specified
10591 by @var{linespec}, and describe the source location or compiler
10592 command-line where those definitions were established.
10594 @kindex macro define
10595 @cindex user-defined macros
10596 @cindex defining macros interactively
10597 @cindex macros, user-defined
10598 @item macro define @var{macro} @var{replacement-list}
10599 @itemx macro define @var{macro}(@var{arglist}) @var{replacement-list}
10600 Introduce a definition for a preprocessor macro named @var{macro},
10601 invocations of which are replaced by the tokens given in
10602 @var{replacement-list}. The first form of this command defines an
10603 ``object-like'' macro, which takes no arguments; the second form
10604 defines a ``function-like'' macro, which takes the arguments given in
10607 A definition introduced by this command is in scope in every
10608 expression evaluated in @value{GDBN}, until it is removed with the
10609 @code{macro undef} command, described below. The definition overrides
10610 all definitions for @var{macro} present in the program being debugged,
10611 as well as any previous user-supplied definition.
10613 @kindex macro undef
10614 @item macro undef @var{macro}
10615 Remove any user-supplied definition for the macro named @var{macro}.
10616 This command only affects definitions provided with the @code{macro
10617 define} command, described above; it cannot remove definitions present
10618 in the program being debugged.
10622 List all the macros defined using the @code{macro define} command.
10625 @cindex macros, example of debugging with
10626 Here is a transcript showing the above commands in action. First, we
10627 show our source files:
10632 #include "sample.h"
10635 #define ADD(x) (M + x)
10640 printf ("Hello, world!\n");
10642 printf ("We're so creative.\n");
10644 printf ("Goodbye, world!\n");
10651 Now, we compile the program using the @sc{gnu} C compiler,
10652 @value{NGCC}. We pass the @option{-gdwarf-2}@footnote{This is the
10653 minimum. Recent versions of @value{NGCC} support @option{-gdwarf-3}
10654 and @option{-gdwarf-4}; we recommend always choosing the most recent
10655 version of DWARF.} @emph{and} @option{-g3} flags to ensure the compiler
10656 includes information about preprocessor macros in the debugging
10660 $ gcc -gdwarf-2 -g3 sample.c -o sample
10664 Now, we start @value{GDBN} on our sample program:
10668 GNU gdb 2002-05-06-cvs
10669 Copyright 2002 Free Software Foundation, Inc.
10670 GDB is free software, @dots{}
10674 We can expand macros and examine their definitions, even when the
10675 program is not running. @value{GDBN} uses the current listing position
10676 to decide which macro definitions are in scope:
10679 (@value{GDBP}) list main
10682 5 #define ADD(x) (M + x)
10687 10 printf ("Hello, world!\n");
10689 12 printf ("We're so creative.\n");
10690 (@value{GDBP}) info macro ADD
10691 Defined at /home/jimb/gdb/macros/play/sample.c:5
10692 #define ADD(x) (M + x)
10693 (@value{GDBP}) info macro Q
10694 Defined at /home/jimb/gdb/macros/play/sample.h:1
10695 included at /home/jimb/gdb/macros/play/sample.c:2
10697 (@value{GDBP}) macro expand ADD(1)
10698 expands to: (42 + 1)
10699 (@value{GDBP}) macro expand-once ADD(1)
10700 expands to: once (M + 1)
10704 In the example above, note that @code{macro expand-once} expands only
10705 the macro invocation explicit in the original text --- the invocation of
10706 @code{ADD} --- but does not expand the invocation of the macro @code{M},
10707 which was introduced by @code{ADD}.
10709 Once the program is running, @value{GDBN} uses the macro definitions in
10710 force at the source line of the current stack frame:
10713 (@value{GDBP}) break main
10714 Breakpoint 1 at 0x8048370: file sample.c, line 10.
10716 Starting program: /home/jimb/gdb/macros/play/sample
10718 Breakpoint 1, main () at sample.c:10
10719 10 printf ("Hello, world!\n");
10723 At line 10, the definition of the macro @code{N} at line 9 is in force:
10726 (@value{GDBP}) info macro N
10727 Defined at /home/jimb/gdb/macros/play/sample.c:9
10729 (@value{GDBP}) macro expand N Q M
10730 expands to: 28 < 42
10731 (@value{GDBP}) print N Q M
10736 As we step over directives that remove @code{N}'s definition, and then
10737 give it a new definition, @value{GDBN} finds the definition (or lack
10738 thereof) in force at each point:
10741 (@value{GDBP}) next
10743 12 printf ("We're so creative.\n");
10744 (@value{GDBP}) info macro N
10745 The symbol `N' has no definition as a C/C++ preprocessor macro
10746 at /home/jimb/gdb/macros/play/sample.c:12
10747 (@value{GDBP}) next
10749 14 printf ("Goodbye, world!\n");
10750 (@value{GDBP}) info macro N
10751 Defined at /home/jimb/gdb/macros/play/sample.c:13
10753 (@value{GDBP}) macro expand N Q M
10754 expands to: 1729 < 42
10755 (@value{GDBP}) print N Q M
10760 In addition to source files, macros can be defined on the compilation command
10761 line using the @option{-D@var{name}=@var{value}} syntax. For macros defined in
10762 such a way, @value{GDBN} displays the location of their definition as line zero
10763 of the source file submitted to the compiler.
10766 (@value{GDBP}) info macro __STDC__
10767 Defined at /home/jimb/gdb/macros/play/sample.c:0
10774 @chapter Tracepoints
10775 @c This chapter is based on the documentation written by Michael
10776 @c Snyder, David Taylor, Jim Blandy, and Elena Zannoni.
10778 @cindex tracepoints
10779 In some applications, it is not feasible for the debugger to interrupt
10780 the program's execution long enough for the developer to learn
10781 anything helpful about its behavior. If the program's correctness
10782 depends on its real-time behavior, delays introduced by a debugger
10783 might cause the program to change its behavior drastically, or perhaps
10784 fail, even when the code itself is correct. It is useful to be able
10785 to observe the program's behavior without interrupting it.
10787 Using @value{GDBN}'s @code{trace} and @code{collect} commands, you can
10788 specify locations in the program, called @dfn{tracepoints}, and
10789 arbitrary expressions to evaluate when those tracepoints are reached.
10790 Later, using the @code{tfind} command, you can examine the values
10791 those expressions had when the program hit the tracepoints. The
10792 expressions may also denote objects in memory---structures or arrays,
10793 for example---whose values @value{GDBN} should record; while visiting
10794 a particular tracepoint, you may inspect those objects as if they were
10795 in memory at that moment. However, because @value{GDBN} records these
10796 values without interacting with you, it can do so quickly and
10797 unobtrusively, hopefully not disturbing the program's behavior.
10799 The tracepoint facility is currently available only for remote
10800 targets. @xref{Targets}. In addition, your remote target must know
10801 how to collect trace data. This functionality is implemented in the
10802 remote stub; however, none of the stubs distributed with @value{GDBN}
10803 support tracepoints as of this writing. The format of the remote
10804 packets used to implement tracepoints are described in @ref{Tracepoint
10807 It is also possible to get trace data from a file, in a manner reminiscent
10808 of corefiles; you specify the filename, and use @code{tfind} to search
10809 through the file. @xref{Trace Files}, for more details.
10811 This chapter describes the tracepoint commands and features.
10814 * Set Tracepoints::
10815 * Analyze Collected Data::
10816 * Tracepoint Variables::
10820 @node Set Tracepoints
10821 @section Commands to Set Tracepoints
10823 Before running such a @dfn{trace experiment}, an arbitrary number of
10824 tracepoints can be set. A tracepoint is actually a special type of
10825 breakpoint (@pxref{Set Breaks}), so you can manipulate it using
10826 standard breakpoint commands. For instance, as with breakpoints,
10827 tracepoint numbers are successive integers starting from one, and many
10828 of the commands associated with tracepoints take the tracepoint number
10829 as their argument, to identify which tracepoint to work on.
10831 For each tracepoint, you can specify, in advance, some arbitrary set
10832 of data that you want the target to collect in the trace buffer when
10833 it hits that tracepoint. The collected data can include registers,
10834 local variables, or global data. Later, you can use @value{GDBN}
10835 commands to examine the values these data had at the time the
10836 tracepoint was hit.
10838 Tracepoints do not support every breakpoint feature. Ignore counts on
10839 tracepoints have no effect, and tracepoints cannot run @value{GDBN}
10840 commands when they are hit. Tracepoints may not be thread-specific
10843 @cindex fast tracepoints
10844 Some targets may support @dfn{fast tracepoints}, which are inserted in
10845 a different way (such as with a jump instead of a trap), that is
10846 faster but possibly restricted in where they may be installed.
10848 @cindex static tracepoints
10849 @cindex markers, static tracepoints
10850 @cindex probing markers, static tracepoints
10851 Regular and fast tracepoints are dynamic tracing facilities, meaning
10852 that they can be used to insert tracepoints at (almost) any location
10853 in the target. Some targets may also support controlling @dfn{static
10854 tracepoints} from @value{GDBN}. With static tracing, a set of
10855 instrumentation points, also known as @dfn{markers}, are embedded in
10856 the target program, and can be activated or deactivated by name or
10857 address. These are usually placed at locations which facilitate
10858 investigating what the target is actually doing. @value{GDBN}'s
10859 support for static tracing includes being able to list instrumentation
10860 points, and attach them with @value{GDBN} defined high level
10861 tracepoints that expose the whole range of convenience of
10862 @value{GDBN}'s tracepoints support. Namely, support for collecting
10863 registers values and values of global or local (to the instrumentation
10864 point) variables; tracepoint conditions and trace state variables.
10865 The act of installing a @value{GDBN} static tracepoint on an
10866 instrumentation point, or marker, is referred to as @dfn{probing} a
10867 static tracepoint marker.
10869 @code{gdbserver} supports tracepoints on some target systems.
10870 @xref{Server,,Tracepoints support in @code{gdbserver}}.
10872 This section describes commands to set tracepoints and associated
10873 conditions and actions.
10876 * Create and Delete Tracepoints::
10877 * Enable and Disable Tracepoints::
10878 * Tracepoint Passcounts::
10879 * Tracepoint Conditions::
10880 * Trace State Variables::
10881 * Tracepoint Actions::
10882 * Listing Tracepoints::
10883 * Listing Static Tracepoint Markers::
10884 * Starting and Stopping Trace Experiments::
10885 * Tracepoint Restrictions::
10888 @node Create and Delete Tracepoints
10889 @subsection Create and Delete Tracepoints
10892 @cindex set tracepoint
10894 @item trace @var{location}
10895 The @code{trace} command is very similar to the @code{break} command.
10896 Its argument @var{location} can be a source line, a function name, or
10897 an address in the target program. @xref{Specify Location}. The
10898 @code{trace} command defines a tracepoint, which is a point in the
10899 target program where the debugger will briefly stop, collect some
10900 data, and then allow the program to continue. Setting a tracepoint or
10901 changing its actions takes effect immediately if the remote stub
10902 supports the @samp{InstallInTrace} feature (@pxref{install tracepoint
10904 If remote stub doesn't support the @samp{InstallInTrace} feature, all
10905 these changes don't take effect until the next @code{tstart}
10906 command, and once a trace experiment is running, further changes will
10907 not have any effect until the next trace experiment starts. In addition,
10908 @value{GDBN} supports @dfn{pending tracepoints}---tracepoints whose
10909 address is not yet resolved. (This is similar to pending breakpoints.)
10910 Pending tracepoints are not downloaded to the target and not installed
10911 until they are resolved. The resolution of pending tracepoints requires
10912 @value{GDBN} support---when debugging with the remote target, and
10913 @value{GDBN} disconnects from the remote stub (@pxref{disconnected
10914 tracing}), pending tracepoints can not be resolved (and downloaded to
10915 the remote stub) while @value{GDBN} is disconnected.
10917 Here are some examples of using the @code{trace} command:
10920 (@value{GDBP}) @b{trace foo.c:121} // a source file and line number
10922 (@value{GDBP}) @b{trace +2} // 2 lines forward
10924 (@value{GDBP}) @b{trace my_function} // first source line of function
10926 (@value{GDBP}) @b{trace *my_function} // EXACT start address of function
10928 (@value{GDBP}) @b{trace *0x2117c4} // an address
10932 You can abbreviate @code{trace} as @code{tr}.
10934 @item trace @var{location} if @var{cond}
10935 Set a tracepoint with condition @var{cond}; evaluate the expression
10936 @var{cond} each time the tracepoint is reached, and collect data only
10937 if the value is nonzero---that is, if @var{cond} evaluates as true.
10938 @xref{Tracepoint Conditions, ,Tracepoint Conditions}, for more
10939 information on tracepoint conditions.
10941 @item ftrace @var{location} [ if @var{cond} ]
10942 @cindex set fast tracepoint
10943 @cindex fast tracepoints, setting
10945 The @code{ftrace} command sets a fast tracepoint. For targets that
10946 support them, fast tracepoints will use a more efficient but possibly
10947 less general technique to trigger data collection, such as a jump
10948 instruction instead of a trap, or some sort of hardware support. It
10949 may not be possible to create a fast tracepoint at the desired
10950 location, in which case the command will exit with an explanatory
10953 @value{GDBN} handles arguments to @code{ftrace} exactly as for
10956 On 32-bit x86-architecture systems, fast tracepoints normally need to
10957 be placed at an instruction that is 5 bytes or longer, but can be
10958 placed at 4-byte instructions if the low 64K of memory of the target
10959 program is available to install trampolines. Some Unix-type systems,
10960 such as @sc{gnu}/Linux, exclude low addresses from the program's
10961 address space; but for instance with the Linux kernel it is possible
10962 to let @value{GDBN} use this area by doing a @command{sysctl} command
10963 to set the @code{mmap_min_addr} kernel parameter, as in
10966 sudo sysctl -w vm.mmap_min_addr=32768
10970 which sets the low address to 32K, which leaves plenty of room for
10971 trampolines. The minimum address should be set to a page boundary.
10973 @item strace @var{location} [ if @var{cond} ]
10974 @cindex set static tracepoint
10975 @cindex static tracepoints, setting
10976 @cindex probe static tracepoint marker
10978 The @code{strace} command sets a static tracepoint. For targets that
10979 support it, setting a static tracepoint probes a static
10980 instrumentation point, or marker, found at @var{location}. It may not
10981 be possible to set a static tracepoint at the desired location, in
10982 which case the command will exit with an explanatory message.
10984 @value{GDBN} handles arguments to @code{strace} exactly as for
10985 @code{trace}, with the addition that the user can also specify
10986 @code{-m @var{marker}} as @var{location}. This probes the marker
10987 identified by the @var{marker} string identifier. This identifier
10988 depends on the static tracepoint backend library your program is
10989 using. You can find all the marker identifiers in the @samp{ID} field
10990 of the @code{info static-tracepoint-markers} command output.
10991 @xref{Listing Static Tracepoint Markers,,Listing Static Tracepoint
10992 Markers}. For example, in the following small program using the UST
10998 trace_mark(ust, bar33, "str %s", "FOOBAZ");
11003 the marker id is composed of joining the first two arguments to the
11004 @code{trace_mark} call with a slash, which translates to:
11007 (@value{GDBP}) info static-tracepoint-markers
11008 Cnt Enb ID Address What
11009 1 n ust/bar33 0x0000000000400ddc in main at stexample.c:22
11015 so you may probe the marker above with:
11018 (@value{GDBP}) strace -m ust/bar33
11021 Static tracepoints accept an extra collect action --- @code{collect
11022 $_sdata}. This collects arbitrary user data passed in the probe point
11023 call to the tracing library. In the UST example above, you'll see
11024 that the third argument to @code{trace_mark} is a printf-like format
11025 string. The user data is then the result of running that formating
11026 string against the following arguments. Note that @code{info
11027 static-tracepoint-markers} command output lists that format string in
11028 the @samp{Data:} field.
11030 You can inspect this data when analyzing the trace buffer, by printing
11031 the $_sdata variable like any other variable available to
11032 @value{GDBN}. @xref{Tracepoint Actions,,Tracepoint Action Lists}.
11035 @cindex last tracepoint number
11036 @cindex recent tracepoint number
11037 @cindex tracepoint number
11038 The convenience variable @code{$tpnum} records the tracepoint number
11039 of the most recently set tracepoint.
11041 @kindex delete tracepoint
11042 @cindex tracepoint deletion
11043 @item delete tracepoint @r{[}@var{num}@r{]}
11044 Permanently delete one or more tracepoints. With no argument, the
11045 default is to delete all tracepoints. Note that the regular
11046 @code{delete} command can remove tracepoints also.
11051 (@value{GDBP}) @b{delete trace 1 2 3} // remove three tracepoints
11053 (@value{GDBP}) @b{delete trace} // remove all tracepoints
11057 You can abbreviate this command as @code{del tr}.
11060 @node Enable and Disable Tracepoints
11061 @subsection Enable and Disable Tracepoints
11063 These commands are deprecated; they are equivalent to plain @code{disable} and @code{enable}.
11066 @kindex disable tracepoint
11067 @item disable tracepoint @r{[}@var{num}@r{]}
11068 Disable tracepoint @var{num}, or all tracepoints if no argument
11069 @var{num} is given. A disabled tracepoint will have no effect during
11070 a trace experiment, but it is not forgotten. You can re-enable
11071 a disabled tracepoint using the @code{enable tracepoint} command.
11072 If the command is issued during a trace experiment and the debug target
11073 has support for disabling tracepoints during a trace experiment, then the
11074 change will be effective immediately. Otherwise, it will be applied to the
11075 next trace experiment.
11077 @kindex enable tracepoint
11078 @item enable tracepoint @r{[}@var{num}@r{]}
11079 Enable tracepoint @var{num}, or all tracepoints. If this command is
11080 issued during a trace experiment and the debug target supports enabling
11081 tracepoints during a trace experiment, then the enabled tracepoints will
11082 become effective immediately. Otherwise, they will become effective the
11083 next time a trace experiment is run.
11086 @node Tracepoint Passcounts
11087 @subsection Tracepoint Passcounts
11091 @cindex tracepoint pass count
11092 @item passcount @r{[}@var{n} @r{[}@var{num}@r{]]}
11093 Set the @dfn{passcount} of a tracepoint. The passcount is a way to
11094 automatically stop a trace experiment. If a tracepoint's passcount is
11095 @var{n}, then the trace experiment will be automatically stopped on
11096 the @var{n}'th time that tracepoint is hit. If the tracepoint number
11097 @var{num} is not specified, the @code{passcount} command sets the
11098 passcount of the most recently defined tracepoint. If no passcount is
11099 given, the trace experiment will run until stopped explicitly by the
11105 (@value{GDBP}) @b{passcount 5 2} // Stop on the 5th execution of
11106 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// tracepoint 2}
11108 (@value{GDBP}) @b{passcount 12} // Stop on the 12th execution of the
11109 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// most recently defined tracepoint.}
11110 (@value{GDBP}) @b{trace foo}
11111 (@value{GDBP}) @b{pass 3}
11112 (@value{GDBP}) @b{trace bar}
11113 (@value{GDBP}) @b{pass 2}
11114 (@value{GDBP}) @b{trace baz}
11115 (@value{GDBP}) @b{pass 1} // Stop tracing when foo has been
11116 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// executed 3 times OR when bar has}
11117 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// been executed 2 times}
11118 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// OR when baz has been executed 1 time.}
11122 @node Tracepoint Conditions
11123 @subsection Tracepoint Conditions
11124 @cindex conditional tracepoints
11125 @cindex tracepoint conditions
11127 The simplest sort of tracepoint collects data every time your program
11128 reaches a specified place. You can also specify a @dfn{condition} for
11129 a tracepoint. A condition is just a Boolean expression in your
11130 programming language (@pxref{Expressions, ,Expressions}). A
11131 tracepoint with a condition evaluates the expression each time your
11132 program reaches it, and data collection happens only if the condition
11135 Tracepoint conditions can be specified when a tracepoint is set, by
11136 using @samp{if} in the arguments to the @code{trace} command.
11137 @xref{Create and Delete Tracepoints, ,Setting Tracepoints}. They can
11138 also be set or changed at any time with the @code{condition} command,
11139 just as with breakpoints.
11141 Unlike breakpoint conditions, @value{GDBN} does not actually evaluate
11142 the conditional expression itself. Instead, @value{GDBN} encodes the
11143 expression into an agent expression (@pxref{Agent Expressions})
11144 suitable for execution on the target, independently of @value{GDBN}.
11145 Global variables become raw memory locations, locals become stack
11146 accesses, and so forth.
11148 For instance, suppose you have a function that is usually called
11149 frequently, but should not be called after an error has occurred. You
11150 could use the following tracepoint command to collect data about calls
11151 of that function that happen while the error code is propagating
11152 through the program; an unconditional tracepoint could end up
11153 collecting thousands of useless trace frames that you would have to
11157 (@value{GDBP}) @kbd{trace normal_operation if errcode > 0}
11160 @node Trace State Variables
11161 @subsection Trace State Variables
11162 @cindex trace state variables
11164 A @dfn{trace state variable} is a special type of variable that is
11165 created and managed by target-side code. The syntax is the same as
11166 that for GDB's convenience variables (a string prefixed with ``$''),
11167 but they are stored on the target. They must be created explicitly,
11168 using a @code{tvariable} command. They are always 64-bit signed
11171 Trace state variables are remembered by @value{GDBN}, and downloaded
11172 to the target along with tracepoint information when the trace
11173 experiment starts. There are no intrinsic limits on the number of
11174 trace state variables, beyond memory limitations of the target.
11176 @cindex convenience variables, and trace state variables
11177 Although trace state variables are managed by the target, you can use
11178 them in print commands and expressions as if they were convenience
11179 variables; @value{GDBN} will get the current value from the target
11180 while the trace experiment is running. Trace state variables share
11181 the same namespace as other ``$'' variables, which means that you
11182 cannot have trace state variables with names like @code{$23} or
11183 @code{$pc}, nor can you have a trace state variable and a convenience
11184 variable with the same name.
11188 @item tvariable $@var{name} [ = @var{expression} ]
11190 The @code{tvariable} command creates a new trace state variable named
11191 @code{$@var{name}}, and optionally gives it an initial value of
11192 @var{expression}. @var{expression} is evaluated when this command is
11193 entered; the result will be converted to an integer if possible,
11194 otherwise @value{GDBN} will report an error. A subsequent
11195 @code{tvariable} command specifying the same name does not create a
11196 variable, but instead assigns the supplied initial value to the
11197 existing variable of that name, overwriting any previous initial
11198 value. The default initial value is 0.
11200 @item info tvariables
11201 @kindex info tvariables
11202 List all the trace state variables along with their initial values.
11203 Their current values may also be displayed, if the trace experiment is
11206 @item delete tvariable @r{[} $@var{name} @dots{} @r{]}
11207 @kindex delete tvariable
11208 Delete the given trace state variables, or all of them if no arguments
11213 @node Tracepoint Actions
11214 @subsection Tracepoint Action Lists
11218 @cindex tracepoint actions
11219 @item actions @r{[}@var{num}@r{]}
11220 This command will prompt for a list of actions to be taken when the
11221 tracepoint is hit. If the tracepoint number @var{num} is not
11222 specified, this command sets the actions for the one that was most
11223 recently defined (so that you can define a tracepoint and then say
11224 @code{actions} without bothering about its number). You specify the
11225 actions themselves on the following lines, one action at a time, and
11226 terminate the actions list with a line containing just @code{end}. So
11227 far, the only defined actions are @code{collect}, @code{teval}, and
11228 @code{while-stepping}.
11230 @code{actions} is actually equivalent to @code{commands} (@pxref{Break
11231 Commands, ,Breakpoint Command Lists}), except that only the defined
11232 actions are allowed; any other @value{GDBN} command is rejected.
11234 @cindex remove actions from a tracepoint
11235 To remove all actions from a tracepoint, type @samp{actions @var{num}}
11236 and follow it immediately with @samp{end}.
11239 (@value{GDBP}) @b{collect @var{data}} // collect some data
11241 (@value{GDBP}) @b{while-stepping 5} // single-step 5 times, collect data
11243 (@value{GDBP}) @b{end} // signals the end of actions.
11246 In the following example, the action list begins with @code{collect}
11247 commands indicating the things to be collected when the tracepoint is
11248 hit. Then, in order to single-step and collect additional data
11249 following the tracepoint, a @code{while-stepping} command is used,
11250 followed by the list of things to be collected after each step in a
11251 sequence of single steps. The @code{while-stepping} command is
11252 terminated by its own separate @code{end} command. Lastly, the action
11253 list is terminated by an @code{end} command.
11256 (@value{GDBP}) @b{trace foo}
11257 (@value{GDBP}) @b{actions}
11258 Enter actions for tracepoint 1, one per line:
11261 > while-stepping 12
11262 > collect $pc, arr[i]
11267 @kindex collect @r{(tracepoints)}
11268 @item collect@r{[}/@var{mods}@r{]} @var{expr1}, @var{expr2}, @dots{}
11269 Collect values of the given expressions when the tracepoint is hit.
11270 This command accepts a comma-separated list of any valid expressions.
11271 In addition to global, static, or local variables, the following
11272 special arguments are supported:
11276 Collect all registers.
11279 Collect all function arguments.
11282 Collect all local variables.
11285 Collect the return address. This is helpful if you want to see more
11289 Collects the number of arguments from the static probe at which the
11290 tracepoint is located.
11291 @xref{Static Probe Points}.
11293 @item $_probe_arg@var{n}
11294 @var{n} is an integer between 0 and 11. Collects the @var{n}th argument
11295 from the static probe at which the tracepoint is located.
11296 @xref{Static Probe Points}.
11299 @vindex $_sdata@r{, collect}
11300 Collect static tracepoint marker specific data. Only available for
11301 static tracepoints. @xref{Tracepoint Actions,,Tracepoint Action
11302 Lists}. On the UST static tracepoints library backend, an
11303 instrumentation point resembles a @code{printf} function call. The
11304 tracing library is able to collect user specified data formatted to a
11305 character string using the format provided by the programmer that
11306 instrumented the program. Other backends have similar mechanisms.
11307 Here's an example of a UST marker call:
11310 const char master_name[] = "$your_name";
11311 trace_mark(channel1, marker1, "hello %s", master_name)
11314 In this case, collecting @code{$_sdata} collects the string
11315 @samp{hello $yourname}. When analyzing the trace buffer, you can
11316 inspect @samp{$_sdata} like any other variable available to
11320 You can give several consecutive @code{collect} commands, each one
11321 with a single argument, or one @code{collect} command with several
11322 arguments separated by commas; the effect is the same.
11324 The optional @var{mods} changes the usual handling of the arguments.
11325 @code{s} requests that pointers to chars be handled as strings, in
11326 particular collecting the contents of the memory being pointed at, up
11327 to the first zero. The upper bound is by default the value of the
11328 @code{print elements} variable; if @code{s} is followed by a decimal
11329 number, that is the upper bound instead. So for instance
11330 @samp{collect/s25 mystr} collects as many as 25 characters at
11333 The command @code{info scope} (@pxref{Symbols, info scope}) is
11334 particularly useful for figuring out what data to collect.
11336 @kindex teval @r{(tracepoints)}
11337 @item teval @var{expr1}, @var{expr2}, @dots{}
11338 Evaluate the given expressions when the tracepoint is hit. This
11339 command accepts a comma-separated list of expressions. The results
11340 are discarded, so this is mainly useful for assigning values to trace
11341 state variables (@pxref{Trace State Variables}) without adding those
11342 values to the trace buffer, as would be the case if the @code{collect}
11345 @kindex while-stepping @r{(tracepoints)}
11346 @item while-stepping @var{n}
11347 Perform @var{n} single-step instruction traces after the tracepoint,
11348 collecting new data after each step. The @code{while-stepping}
11349 command is followed by the list of what to collect while stepping
11350 (followed by its own @code{end} command):
11353 > while-stepping 12
11354 > collect $regs, myglobal
11360 Note that @code{$pc} is not automatically collected by
11361 @code{while-stepping}; you need to explicitly collect that register if
11362 you need it. You may abbreviate @code{while-stepping} as @code{ws} or
11365 @item set default-collect @var{expr1}, @var{expr2}, @dots{}
11366 @kindex set default-collect
11367 @cindex default collection action
11368 This variable is a list of expressions to collect at each tracepoint
11369 hit. It is effectively an additional @code{collect} action prepended
11370 to every tracepoint action list. The expressions are parsed
11371 individually for each tracepoint, so for instance a variable named
11372 @code{xyz} may be interpreted as a global for one tracepoint, and a
11373 local for another, as appropriate to the tracepoint's location.
11375 @item show default-collect
11376 @kindex show default-collect
11377 Show the list of expressions that are collected by default at each
11382 @node Listing Tracepoints
11383 @subsection Listing Tracepoints
11386 @kindex info tracepoints @r{[}@var{n}@dots{}@r{]}
11387 @kindex info tp @r{[}@var{n}@dots{}@r{]}
11388 @cindex information about tracepoints
11389 @item info tracepoints @r{[}@var{num}@dots{}@r{]}
11390 Display information about the tracepoint @var{num}. If you don't
11391 specify a tracepoint number, displays information about all the
11392 tracepoints defined so far. The format is similar to that used for
11393 @code{info breakpoints}; in fact, @code{info tracepoints} is the same
11394 command, simply restricting itself to tracepoints.
11396 A tracepoint's listing may include additional information specific to
11401 its passcount as given by the @code{passcount @var{n}} command
11405 (@value{GDBP}) @b{info trace}
11406 Num Type Disp Enb Address What
11407 1 tracepoint keep y 0x0804ab57 in foo() at main.cxx:7
11409 collect globfoo, $regs
11418 This command can be abbreviated @code{info tp}.
11421 @node Listing Static Tracepoint Markers
11422 @subsection Listing Static Tracepoint Markers
11425 @kindex info static-tracepoint-markers
11426 @cindex information about static tracepoint markers
11427 @item info static-tracepoint-markers
11428 Display information about all static tracepoint markers defined in the
11431 For each marker, the following columns are printed:
11435 An incrementing counter, output to help readability. This is not a
11438 The marker ID, as reported by the target.
11439 @item Enabled or Disabled
11440 Probed markers are tagged with @samp{y}. @samp{n} identifies marks
11441 that are not enabled.
11443 Where the marker is in your program, as a memory address.
11445 Where the marker is in the source for your program, as a file and line
11446 number. If the debug information included in the program does not
11447 allow @value{GDBN} to locate the source of the marker, this column
11448 will be left blank.
11452 In addition, the following information may be printed for each marker:
11456 User data passed to the tracing library by the marker call. In the
11457 UST backend, this is the format string passed as argument to the
11459 @item Static tracepoints probing the marker
11460 The list of static tracepoints attached to the marker.
11464 (@value{GDBP}) info static-tracepoint-markers
11465 Cnt ID Enb Address What
11466 1 ust/bar2 y 0x0000000000400e1a in main at stexample.c:25
11467 Data: number1 %d number2 %d
11468 Probed by static tracepoints: #2
11469 2 ust/bar33 n 0x0000000000400c87 in main at stexample.c:24
11475 @node Starting and Stopping Trace Experiments
11476 @subsection Starting and Stopping Trace Experiments
11479 @kindex tstart [ @var{notes} ]
11480 @cindex start a new trace experiment
11481 @cindex collected data discarded
11483 This command starts the trace experiment, and begins collecting data.
11484 It has the side effect of discarding all the data collected in the
11485 trace buffer during the previous trace experiment. If any arguments
11486 are supplied, they are taken as a note and stored with the trace
11487 experiment's state. The notes may be arbitrary text, and are
11488 especially useful with disconnected tracing in a multi-user context;
11489 the notes can explain what the trace is doing, supply user contact
11490 information, and so forth.
11492 @kindex tstop [ @var{notes} ]
11493 @cindex stop a running trace experiment
11495 This command stops the trace experiment. If any arguments are
11496 supplied, they are recorded with the experiment as a note. This is
11497 useful if you are stopping a trace started by someone else, for
11498 instance if the trace is interfering with the system's behavior and
11499 needs to be stopped quickly.
11501 @strong{Note}: a trace experiment and data collection may stop
11502 automatically if any tracepoint's passcount is reached
11503 (@pxref{Tracepoint Passcounts}), or if the trace buffer becomes full.
11506 @cindex status of trace data collection
11507 @cindex trace experiment, status of
11509 This command displays the status of the current trace data
11513 Here is an example of the commands we described so far:
11516 (@value{GDBP}) @b{trace gdb_c_test}
11517 (@value{GDBP}) @b{actions}
11518 Enter actions for tracepoint #1, one per line.
11519 > collect $regs,$locals,$args
11520 > while-stepping 11
11524 (@value{GDBP}) @b{tstart}
11525 [time passes @dots{}]
11526 (@value{GDBP}) @b{tstop}
11529 @anchor{disconnected tracing}
11530 @cindex disconnected tracing
11531 You can choose to continue running the trace experiment even if
11532 @value{GDBN} disconnects from the target, voluntarily or
11533 involuntarily. For commands such as @code{detach}, the debugger will
11534 ask what you want to do with the trace. But for unexpected
11535 terminations (@value{GDBN} crash, network outage), it would be
11536 unfortunate to lose hard-won trace data, so the variable
11537 @code{disconnected-tracing} lets you decide whether the trace should
11538 continue running without @value{GDBN}.
11541 @item set disconnected-tracing on
11542 @itemx set disconnected-tracing off
11543 @kindex set disconnected-tracing
11544 Choose whether a tracing run should continue to run if @value{GDBN}
11545 has disconnected from the target. Note that @code{detach} or
11546 @code{quit} will ask you directly what to do about a running trace no
11547 matter what this variable's setting, so the variable is mainly useful
11548 for handling unexpected situations, such as loss of the network.
11550 @item show disconnected-tracing
11551 @kindex show disconnected-tracing
11552 Show the current choice for disconnected tracing.
11556 When you reconnect to the target, the trace experiment may or may not
11557 still be running; it might have filled the trace buffer in the
11558 meantime, or stopped for one of the other reasons. If it is running,
11559 it will continue after reconnection.
11561 Upon reconnection, the target will upload information about the
11562 tracepoints in effect. @value{GDBN} will then compare that
11563 information to the set of tracepoints currently defined, and attempt
11564 to match them up, allowing for the possibility that the numbers may
11565 have changed due to creation and deletion in the meantime. If one of
11566 the target's tracepoints does not match any in @value{GDBN}, the
11567 debugger will create a new tracepoint, so that you have a number with
11568 which to specify that tracepoint. This matching-up process is
11569 necessarily heuristic, and it may result in useless tracepoints being
11570 created; you may simply delete them if they are of no use.
11572 @cindex circular trace buffer
11573 If your target agent supports a @dfn{circular trace buffer}, then you
11574 can run a trace experiment indefinitely without filling the trace
11575 buffer; when space runs out, the agent deletes already-collected trace
11576 frames, oldest first, until there is enough room to continue
11577 collecting. This is especially useful if your tracepoints are being
11578 hit too often, and your trace gets terminated prematurely because the
11579 buffer is full. To ask for a circular trace buffer, simply set
11580 @samp{circular-trace-buffer} to on. You can set this at any time,
11581 including during tracing; if the agent can do it, it will change
11582 buffer handling on the fly, otherwise it will not take effect until
11586 @item set circular-trace-buffer on
11587 @itemx set circular-trace-buffer off
11588 @kindex set circular-trace-buffer
11589 Choose whether a tracing run should use a linear or circular buffer
11590 for trace data. A linear buffer will not lose any trace data, but may
11591 fill up prematurely, while a circular buffer will discard old trace
11592 data, but it will have always room for the latest tracepoint hits.
11594 @item show circular-trace-buffer
11595 @kindex show circular-trace-buffer
11596 Show the current choice for the trace buffer. Note that this may not
11597 match the agent's current buffer handling, nor is it guaranteed to
11598 match the setting that might have been in effect during a past run,
11599 for instance if you are looking at frames from a trace file.
11604 @item set trace-user @var{text}
11605 @kindex set trace-user
11607 @item show trace-user
11608 @kindex show trace-user
11610 @item set trace-notes @var{text}
11611 @kindex set trace-notes
11612 Set the trace run's notes.
11614 @item show trace-notes
11615 @kindex show trace-notes
11616 Show the trace run's notes.
11618 @item set trace-stop-notes @var{text}
11619 @kindex set trace-stop-notes
11620 Set the trace run's stop notes. The handling of the note is as for
11621 @code{tstop} arguments; the set command is convenient way to fix a
11622 stop note that is mistaken or incomplete.
11624 @item show trace-stop-notes
11625 @kindex show trace-stop-notes
11626 Show the trace run's stop notes.
11630 @node Tracepoint Restrictions
11631 @subsection Tracepoint Restrictions
11633 @cindex tracepoint restrictions
11634 There are a number of restrictions on the use of tracepoints. As
11635 described above, tracepoint data gathering occurs on the target
11636 without interaction from @value{GDBN}. Thus the full capabilities of
11637 the debugger are not available during data gathering, and then at data
11638 examination time, you will be limited by only having what was
11639 collected. The following items describe some common problems, but it
11640 is not exhaustive, and you may run into additional difficulties not
11646 Tracepoint expressions are intended to gather objects (lvalues). Thus
11647 the full flexibility of GDB's expression evaluator is not available.
11648 You cannot call functions, cast objects to aggregate types, access
11649 convenience variables or modify values (except by assignment to trace
11650 state variables). Some language features may implicitly call
11651 functions (for instance Objective-C fields with accessors), and therefore
11652 cannot be collected either.
11655 Collection of local variables, either individually or in bulk with
11656 @code{$locals} or @code{$args}, during @code{while-stepping} may
11657 behave erratically. The stepping action may enter a new scope (for
11658 instance by stepping into a function), or the location of the variable
11659 may change (for instance it is loaded into a register). The
11660 tracepoint data recorded uses the location information for the
11661 variables that is correct for the tracepoint location. When the
11662 tracepoint is created, it is not possible, in general, to determine
11663 where the steps of a @code{while-stepping} sequence will advance the
11664 program---particularly if a conditional branch is stepped.
11667 Collection of an incompletely-initialized or partially-destroyed object
11668 may result in something that @value{GDBN} cannot display, or displays
11669 in a misleading way.
11672 When @value{GDBN} displays a pointer to character it automatically
11673 dereferences the pointer to also display characters of the string
11674 being pointed to. However, collecting the pointer during tracing does
11675 not automatically collect the string. You need to explicitly
11676 dereference the pointer and provide size information if you want to
11677 collect not only the pointer, but the memory pointed to. For example,
11678 @code{*ptr@@50} can be used to collect the 50 element array pointed to
11682 It is not possible to collect a complete stack backtrace at a
11683 tracepoint. Instead, you may collect the registers and a few hundred
11684 bytes from the stack pointer with something like @code{*(unsigned char *)$esp@@300}
11685 (adjust to use the name of the actual stack pointer register on your
11686 target architecture, and the amount of stack you wish to capture).
11687 Then the @code{backtrace} command will show a partial backtrace when
11688 using a trace frame. The number of stack frames that can be examined
11689 depends on the sizes of the frames in the collected stack. Note that
11690 if you ask for a block so large that it goes past the bottom of the
11691 stack, the target agent may report an error trying to read from an
11695 If you do not collect registers at a tracepoint, @value{GDBN} can
11696 infer that the value of @code{$pc} must be the same as the address of
11697 the tracepoint and use that when you are looking at a trace frame
11698 for that tracepoint. However, this cannot work if the tracepoint has
11699 multiple locations (for instance if it was set in a function that was
11700 inlined), or if it has a @code{while-stepping} loop. In those cases
11701 @value{GDBN} will warn you that it can't infer @code{$pc}, and default
11706 @node Analyze Collected Data
11707 @section Using the Collected Data
11709 After the tracepoint experiment ends, you use @value{GDBN} commands
11710 for examining the trace data. The basic idea is that each tracepoint
11711 collects a trace @dfn{snapshot} every time it is hit and another
11712 snapshot every time it single-steps. All these snapshots are
11713 consecutively numbered from zero and go into a buffer, and you can
11714 examine them later. The way you examine them is to @dfn{focus} on a
11715 specific trace snapshot. When the remote stub is focused on a trace
11716 snapshot, it will respond to all @value{GDBN} requests for memory and
11717 registers by reading from the buffer which belongs to that snapshot,
11718 rather than from @emph{real} memory or registers of the program being
11719 debugged. This means that @strong{all} @value{GDBN} commands
11720 (@code{print}, @code{info registers}, @code{backtrace}, etc.) will
11721 behave as if we were currently debugging the program state as it was
11722 when the tracepoint occurred. Any requests for data that are not in
11723 the buffer will fail.
11726 * tfind:: How to select a trace snapshot
11727 * tdump:: How to display all data for a snapshot
11728 * save tracepoints:: How to save tracepoints for a future run
11732 @subsection @code{tfind @var{n}}
11735 @cindex select trace snapshot
11736 @cindex find trace snapshot
11737 The basic command for selecting a trace snapshot from the buffer is
11738 @code{tfind @var{n}}, which finds trace snapshot number @var{n},
11739 counting from zero. If no argument @var{n} is given, the next
11740 snapshot is selected.
11742 Here are the various forms of using the @code{tfind} command.
11746 Find the first snapshot in the buffer. This is a synonym for
11747 @code{tfind 0} (since 0 is the number of the first snapshot).
11750 Stop debugging trace snapshots, resume @emph{live} debugging.
11753 Same as @samp{tfind none}.
11756 No argument means find the next trace snapshot.
11759 Find the previous trace snapshot before the current one. This permits
11760 retracing earlier steps.
11762 @item tfind tracepoint @var{num}
11763 Find the next snapshot associated with tracepoint @var{num}. Search
11764 proceeds forward from the last examined trace snapshot. If no
11765 argument @var{num} is given, it means find the next snapshot collected
11766 for the same tracepoint as the current snapshot.
11768 @item tfind pc @var{addr}
11769 Find the next snapshot associated with the value @var{addr} of the
11770 program counter. Search proceeds forward from the last examined trace
11771 snapshot. If no argument @var{addr} is given, it means find the next
11772 snapshot with the same value of PC as the current snapshot.
11774 @item tfind outside @var{addr1}, @var{addr2}
11775 Find the next snapshot whose PC is outside the given range of
11776 addresses (exclusive).
11778 @item tfind range @var{addr1}, @var{addr2}
11779 Find the next snapshot whose PC is between @var{addr1} and
11780 @var{addr2} (inclusive).
11782 @item tfind line @r{[}@var{file}:@r{]}@var{n}
11783 Find the next snapshot associated with the source line @var{n}. If
11784 the optional argument @var{file} is given, refer to line @var{n} in
11785 that source file. Search proceeds forward from the last examined
11786 trace snapshot. If no argument @var{n} is given, it means find the
11787 next line other than the one currently being examined; thus saying
11788 @code{tfind line} repeatedly can appear to have the same effect as
11789 stepping from line to line in a @emph{live} debugging session.
11792 The default arguments for the @code{tfind} commands are specifically
11793 designed to make it easy to scan through the trace buffer. For
11794 instance, @code{tfind} with no argument selects the next trace
11795 snapshot, and @code{tfind -} with no argument selects the previous
11796 trace snapshot. So, by giving one @code{tfind} command, and then
11797 simply hitting @key{RET} repeatedly you can examine all the trace
11798 snapshots in order. Or, by saying @code{tfind -} and then hitting
11799 @key{RET} repeatedly you can examine the snapshots in reverse order.
11800 The @code{tfind line} command with no argument selects the snapshot
11801 for the next source line executed. The @code{tfind pc} command with
11802 no argument selects the next snapshot with the same program counter
11803 (PC) as the current frame. The @code{tfind tracepoint} command with
11804 no argument selects the next trace snapshot collected by the same
11805 tracepoint as the current one.
11807 In addition to letting you scan through the trace buffer manually,
11808 these commands make it easy to construct @value{GDBN} scripts that
11809 scan through the trace buffer and print out whatever collected data
11810 you are interested in. Thus, if we want to examine the PC, FP, and SP
11811 registers from each trace frame in the buffer, we can say this:
11814 (@value{GDBP}) @b{tfind start}
11815 (@value{GDBP}) @b{while ($trace_frame != -1)}
11816 > printf "Frame %d, PC = %08X, SP = %08X, FP = %08X\n", \
11817 $trace_frame, $pc, $sp, $fp
11821 Frame 0, PC = 0020DC64, SP = 0030BF3C, FP = 0030BF44
11822 Frame 1, PC = 0020DC6C, SP = 0030BF38, FP = 0030BF44
11823 Frame 2, PC = 0020DC70, SP = 0030BF34, FP = 0030BF44
11824 Frame 3, PC = 0020DC74, SP = 0030BF30, FP = 0030BF44
11825 Frame 4, PC = 0020DC78, SP = 0030BF2C, FP = 0030BF44
11826 Frame 5, PC = 0020DC7C, SP = 0030BF28, FP = 0030BF44
11827 Frame 6, PC = 0020DC80, SP = 0030BF24, FP = 0030BF44
11828 Frame 7, PC = 0020DC84, SP = 0030BF20, FP = 0030BF44
11829 Frame 8, PC = 0020DC88, SP = 0030BF1C, FP = 0030BF44
11830 Frame 9, PC = 0020DC8E, SP = 0030BF18, FP = 0030BF44
11831 Frame 10, PC = 00203F6C, SP = 0030BE3C, FP = 0030BF14
11834 Or, if we want to examine the variable @code{X} at each source line in
11838 (@value{GDBP}) @b{tfind start}
11839 (@value{GDBP}) @b{while ($trace_frame != -1)}
11840 > printf "Frame %d, X == %d\n", $trace_frame, X
11850 @subsection @code{tdump}
11852 @cindex dump all data collected at tracepoint
11853 @cindex tracepoint data, display
11855 This command takes no arguments. It prints all the data collected at
11856 the current trace snapshot.
11859 (@value{GDBP}) @b{trace 444}
11860 (@value{GDBP}) @b{actions}
11861 Enter actions for tracepoint #2, one per line:
11862 > collect $regs, $locals, $args, gdb_long_test
11865 (@value{GDBP}) @b{tstart}
11867 (@value{GDBP}) @b{tfind line 444}
11868 #0 gdb_test (p1=0x11, p2=0x22, p3=0x33, p4=0x44, p5=0x55, p6=0x66)
11870 444 printp( "%s: arguments = 0x%X 0x%X 0x%X 0x%X 0x%X 0x%X\n", )
11872 (@value{GDBP}) @b{tdump}
11873 Data collected at tracepoint 2, trace frame 1:
11874 d0 0xc4aa0085 -995491707
11878 d4 0x71aea3d 119204413
11881 d7 0x380035 3670069
11882 a0 0x19e24a 1696330
11883 a1 0x3000668 50333288
11885 a3 0x322000 3284992
11886 a4 0x3000698 50333336
11887 a5 0x1ad3cc 1758156
11888 fp 0x30bf3c 0x30bf3c
11889 sp 0x30bf34 0x30bf34
11891 pc 0x20b2c8 0x20b2c8
11895 p = 0x20e5b4 "gdb-test"
11902 gdb_long_test = 17 '\021'
11907 @code{tdump} works by scanning the tracepoint's current collection
11908 actions and printing the value of each expression listed. So
11909 @code{tdump} can fail, if after a run, you change the tracepoint's
11910 actions to mention variables that were not collected during the run.
11912 Also, for tracepoints with @code{while-stepping} loops, @code{tdump}
11913 uses the collected value of @code{$pc} to distinguish between trace
11914 frames that were collected at the tracepoint hit, and frames that were
11915 collected while stepping. This allows it to correctly choose whether
11916 to display the basic list of collections, or the collections from the
11917 body of the while-stepping loop. However, if @code{$pc} was not collected,
11918 then @code{tdump} will always attempt to dump using the basic collection
11919 list, and may fail if a while-stepping frame does not include all the
11920 same data that is collected at the tracepoint hit.
11921 @c This is getting pretty arcane, example would be good.
11923 @node save tracepoints
11924 @subsection @code{save tracepoints @var{filename}}
11925 @kindex save tracepoints
11926 @kindex save-tracepoints
11927 @cindex save tracepoints for future sessions
11929 This command saves all current tracepoint definitions together with
11930 their actions and passcounts, into a file @file{@var{filename}}
11931 suitable for use in a later debugging session. To read the saved
11932 tracepoint definitions, use the @code{source} command (@pxref{Command
11933 Files}). The @w{@code{save-tracepoints}} command is a deprecated
11934 alias for @w{@code{save tracepoints}}
11936 @node Tracepoint Variables
11937 @section Convenience Variables for Tracepoints
11938 @cindex tracepoint variables
11939 @cindex convenience variables for tracepoints
11942 @vindex $trace_frame
11943 @item (int) $trace_frame
11944 The current trace snapshot (a.k.a.@: @dfn{frame}) number, or -1 if no
11945 snapshot is selected.
11947 @vindex $tracepoint
11948 @item (int) $tracepoint
11949 The tracepoint for the current trace snapshot.
11951 @vindex $trace_line
11952 @item (int) $trace_line
11953 The line number for the current trace snapshot.
11955 @vindex $trace_file
11956 @item (char []) $trace_file
11957 The source file for the current trace snapshot.
11959 @vindex $trace_func
11960 @item (char []) $trace_func
11961 The name of the function containing @code{$tracepoint}.
11964 Note: @code{$trace_file} is not suitable for use in @code{printf},
11965 use @code{output} instead.
11967 Here's a simple example of using these convenience variables for
11968 stepping through all the trace snapshots and printing some of their
11969 data. Note that these are not the same as trace state variables,
11970 which are managed by the target.
11973 (@value{GDBP}) @b{tfind start}
11975 (@value{GDBP}) @b{while $trace_frame != -1}
11976 > output $trace_file
11977 > printf ", line %d (tracepoint #%d)\n", $trace_line, $tracepoint
11983 @section Using Trace Files
11984 @cindex trace files
11986 In some situations, the target running a trace experiment may no
11987 longer be available; perhaps it crashed, or the hardware was needed
11988 for a different activity. To handle these cases, you can arrange to
11989 dump the trace data into a file, and later use that file as a source
11990 of trace data, via the @code{target tfile} command.
11995 @item tsave [ -r ] @var{filename}
11996 Save the trace data to @var{filename}. By default, this command
11997 assumes that @var{filename} refers to the host filesystem, so if
11998 necessary @value{GDBN} will copy raw trace data up from the target and
11999 then save it. If the target supports it, you can also supply the
12000 optional argument @code{-r} (``remote'') to direct the target to save
12001 the data directly into @var{filename} in its own filesystem, which may be
12002 more efficient if the trace buffer is very large. (Note, however, that
12003 @code{target tfile} can only read from files accessible to the host.)
12005 @kindex target tfile
12007 @item target tfile @var{filename}
12008 Use the file named @var{filename} as a source of trace data. Commands
12009 that examine data work as they do with a live target, but it is not
12010 possible to run any new trace experiments. @code{tstatus} will report
12011 the state of the trace run at the moment the data was saved, as well
12012 as the current trace frame you are examining. @var{filename} must be
12013 on a filesystem accessible to the host.
12018 @chapter Debugging Programs That Use Overlays
12021 If your program is too large to fit completely in your target system's
12022 memory, you can sometimes use @dfn{overlays} to work around this
12023 problem. @value{GDBN} provides some support for debugging programs that
12027 * How Overlays Work:: A general explanation of overlays.
12028 * Overlay Commands:: Managing overlays in @value{GDBN}.
12029 * Automatic Overlay Debugging:: @value{GDBN} can find out which overlays are
12030 mapped by asking the inferior.
12031 * Overlay Sample Program:: A sample program using overlays.
12034 @node How Overlays Work
12035 @section How Overlays Work
12036 @cindex mapped overlays
12037 @cindex unmapped overlays
12038 @cindex load address, overlay's
12039 @cindex mapped address
12040 @cindex overlay area
12042 Suppose you have a computer whose instruction address space is only 64
12043 kilobytes long, but which has much more memory which can be accessed by
12044 other means: special instructions, segment registers, or memory
12045 management hardware, for example. Suppose further that you want to
12046 adapt a program which is larger than 64 kilobytes to run on this system.
12048 One solution is to identify modules of your program which are relatively
12049 independent, and need not call each other directly; call these modules
12050 @dfn{overlays}. Separate the overlays from the main program, and place
12051 their machine code in the larger memory. Place your main program in
12052 instruction memory, but leave at least enough space there to hold the
12053 largest overlay as well.
12055 Now, to call a function located in an overlay, you must first copy that
12056 overlay's machine code from the large memory into the space set aside
12057 for it in the instruction memory, and then jump to its entry point
12060 @c NB: In the below the mapped area's size is greater or equal to the
12061 @c size of all overlays. This is intentional to remind the developer
12062 @c that overlays don't necessarily need to be the same size.
12066 Data Instruction Larger
12067 Address Space Address Space Address Space
12068 +-----------+ +-----------+ +-----------+
12070 +-----------+ +-----------+ +-----------+<-- overlay 1
12071 | program | | main | .----| overlay 1 | load address
12072 | variables | | program | | +-----------+
12073 | and heap | | | | | |
12074 +-----------+ | | | +-----------+<-- overlay 2
12075 | | +-----------+ | | | load address
12076 +-----------+ | | | .-| overlay 2 |
12078 mapped --->+-----------+ | | +-----------+
12079 address | | | | | |
12080 | overlay | <-' | | |
12081 | area | <---' +-----------+<-- overlay 3
12082 | | <---. | | load address
12083 +-----------+ `--| overlay 3 |
12090 @anchor{A code overlay}A code overlay
12094 The diagram (@pxref{A code overlay}) shows a system with separate data
12095 and instruction address spaces. To map an overlay, the program copies
12096 its code from the larger address space to the instruction address space.
12097 Since the overlays shown here all use the same mapped address, only one
12098 may be mapped at a time. For a system with a single address space for
12099 data and instructions, the diagram would be similar, except that the
12100 program variables and heap would share an address space with the main
12101 program and the overlay area.
12103 An overlay loaded into instruction memory and ready for use is called a
12104 @dfn{mapped} overlay; its @dfn{mapped address} is its address in the
12105 instruction memory. An overlay not present (or only partially present)
12106 in instruction memory is called @dfn{unmapped}; its @dfn{load address}
12107 is its address in the larger memory. The mapped address is also called
12108 the @dfn{virtual memory address}, or @dfn{VMA}; the load address is also
12109 called the @dfn{load memory address}, or @dfn{LMA}.
12111 Unfortunately, overlays are not a completely transparent way to adapt a
12112 program to limited instruction memory. They introduce a new set of
12113 global constraints you must keep in mind as you design your program:
12118 Before calling or returning to a function in an overlay, your program
12119 must make sure that overlay is actually mapped. Otherwise, the call or
12120 return will transfer control to the right address, but in the wrong
12121 overlay, and your program will probably crash.
12124 If the process of mapping an overlay is expensive on your system, you
12125 will need to choose your overlays carefully to minimize their effect on
12126 your program's performance.
12129 The executable file you load onto your system must contain each
12130 overlay's instructions, appearing at the overlay's load address, not its
12131 mapped address. However, each overlay's instructions must be relocated
12132 and its symbols defined as if the overlay were at its mapped address.
12133 You can use GNU linker scripts to specify different load and relocation
12134 addresses for pieces of your program; see @ref{Overlay Description,,,
12135 ld.info, Using ld: the GNU linker}.
12138 The procedure for loading executable files onto your system must be able
12139 to load their contents into the larger address space as well as the
12140 instruction and data spaces.
12144 The overlay system described above is rather simple, and could be
12145 improved in many ways:
12150 If your system has suitable bank switch registers or memory management
12151 hardware, you could use those facilities to make an overlay's load area
12152 contents simply appear at their mapped address in instruction space.
12153 This would probably be faster than copying the overlay to its mapped
12154 area in the usual way.
12157 If your overlays are small enough, you could set aside more than one
12158 overlay area, and have more than one overlay mapped at a time.
12161 You can use overlays to manage data, as well as instructions. In
12162 general, data overlays are even less transparent to your design than
12163 code overlays: whereas code overlays only require care when you call or
12164 return to functions, data overlays require care every time you access
12165 the data. Also, if you change the contents of a data overlay, you
12166 must copy its contents back out to its load address before you can copy a
12167 different data overlay into the same mapped area.
12172 @node Overlay Commands
12173 @section Overlay Commands
12175 To use @value{GDBN}'s overlay support, each overlay in your program must
12176 correspond to a separate section of the executable file. The section's
12177 virtual memory address and load memory address must be the overlay's
12178 mapped and load addresses. Identifying overlays with sections allows
12179 @value{GDBN} to determine the appropriate address of a function or
12180 variable, depending on whether the overlay is mapped or not.
12182 @value{GDBN}'s overlay commands all start with the word @code{overlay};
12183 you can abbreviate this as @code{ov} or @code{ovly}. The commands are:
12188 Disable @value{GDBN}'s overlay support. When overlay support is
12189 disabled, @value{GDBN} assumes that all functions and variables are
12190 always present at their mapped addresses. By default, @value{GDBN}'s
12191 overlay support is disabled.
12193 @item overlay manual
12194 @cindex manual overlay debugging
12195 Enable @dfn{manual} overlay debugging. In this mode, @value{GDBN}
12196 relies on you to tell it which overlays are mapped, and which are not,
12197 using the @code{overlay map-overlay} and @code{overlay unmap-overlay}
12198 commands described below.
12200 @item overlay map-overlay @var{overlay}
12201 @itemx overlay map @var{overlay}
12202 @cindex map an overlay
12203 Tell @value{GDBN} that @var{overlay} is now mapped; @var{overlay} must
12204 be the name of the object file section containing the overlay. When an
12205 overlay is mapped, @value{GDBN} assumes it can find the overlay's
12206 functions and variables at their mapped addresses. @value{GDBN} assumes
12207 that any other overlays whose mapped ranges overlap that of
12208 @var{overlay} are now unmapped.
12210 @item overlay unmap-overlay @var{overlay}
12211 @itemx overlay unmap @var{overlay}
12212 @cindex unmap an overlay
12213 Tell @value{GDBN} that @var{overlay} is no longer mapped; @var{overlay}
12214 must be the name of the object file section containing the overlay.
12215 When an overlay is unmapped, @value{GDBN} assumes it can find the
12216 overlay's functions and variables at their load addresses.
12219 Enable @dfn{automatic} overlay debugging. In this mode, @value{GDBN}
12220 consults a data structure the overlay manager maintains in the inferior
12221 to see which overlays are mapped. For details, see @ref{Automatic
12222 Overlay Debugging}.
12224 @item overlay load-target
12225 @itemx overlay load
12226 @cindex reloading the overlay table
12227 Re-read the overlay table from the inferior. Normally, @value{GDBN}
12228 re-reads the table @value{GDBN} automatically each time the inferior
12229 stops, so this command should only be necessary if you have changed the
12230 overlay mapping yourself using @value{GDBN}. This command is only
12231 useful when using automatic overlay debugging.
12233 @item overlay list-overlays
12234 @itemx overlay list
12235 @cindex listing mapped overlays
12236 Display a list of the overlays currently mapped, along with their mapped
12237 addresses, load addresses, and sizes.
12241 Normally, when @value{GDBN} prints a code address, it includes the name
12242 of the function the address falls in:
12245 (@value{GDBP}) print main
12246 $3 = @{int ()@} 0x11a0 <main>
12249 When overlay debugging is enabled, @value{GDBN} recognizes code in
12250 unmapped overlays, and prints the names of unmapped functions with
12251 asterisks around them. For example, if @code{foo} is a function in an
12252 unmapped overlay, @value{GDBN} prints it this way:
12255 (@value{GDBP}) overlay list
12256 No sections are mapped.
12257 (@value{GDBP}) print foo
12258 $5 = @{int (int)@} 0x100000 <*foo*>
12261 When @code{foo}'s overlay is mapped, @value{GDBN} prints the function's
12265 (@value{GDBP}) overlay list
12266 Section .ov.foo.text, loaded at 0x100000 - 0x100034,
12267 mapped at 0x1016 - 0x104a
12268 (@value{GDBP}) print foo
12269 $6 = @{int (int)@} 0x1016 <foo>
12272 When overlay debugging is enabled, @value{GDBN} can find the correct
12273 address for functions and variables in an overlay, whether or not the
12274 overlay is mapped. This allows most @value{GDBN} commands, like
12275 @code{break} and @code{disassemble}, to work normally, even on unmapped
12276 code. However, @value{GDBN}'s breakpoint support has some limitations:
12280 @cindex breakpoints in overlays
12281 @cindex overlays, setting breakpoints in
12282 You can set breakpoints in functions in unmapped overlays, as long as
12283 @value{GDBN} can write to the overlay at its load address.
12285 @value{GDBN} can not set hardware or simulator-based breakpoints in
12286 unmapped overlays. However, if you set a breakpoint at the end of your
12287 overlay manager (and tell @value{GDBN} which overlays are now mapped, if
12288 you are using manual overlay management), @value{GDBN} will re-set its
12289 breakpoints properly.
12293 @node Automatic Overlay Debugging
12294 @section Automatic Overlay Debugging
12295 @cindex automatic overlay debugging
12297 @value{GDBN} can automatically track which overlays are mapped and which
12298 are not, given some simple co-operation from the overlay manager in the
12299 inferior. If you enable automatic overlay debugging with the
12300 @code{overlay auto} command (@pxref{Overlay Commands}), @value{GDBN}
12301 looks in the inferior's memory for certain variables describing the
12302 current state of the overlays.
12304 Here are the variables your overlay manager must define to support
12305 @value{GDBN}'s automatic overlay debugging:
12309 @item @code{_ovly_table}:
12310 This variable must be an array of the following structures:
12315 /* The overlay's mapped address. */
12318 /* The size of the overlay, in bytes. */
12319 unsigned long size;
12321 /* The overlay's load address. */
12324 /* Non-zero if the overlay is currently mapped;
12326 unsigned long mapped;
12330 @item @code{_novlys}:
12331 This variable must be a four-byte signed integer, holding the total
12332 number of elements in @code{_ovly_table}.
12336 To decide whether a particular overlay is mapped or not, @value{GDBN}
12337 looks for an entry in @w{@code{_ovly_table}} whose @code{vma} and
12338 @code{lma} members equal the VMA and LMA of the overlay's section in the
12339 executable file. When @value{GDBN} finds a matching entry, it consults
12340 the entry's @code{mapped} member to determine whether the overlay is
12343 In addition, your overlay manager may define a function called
12344 @code{_ovly_debug_event}. If this function is defined, @value{GDBN}
12345 will silently set a breakpoint there. If the overlay manager then
12346 calls this function whenever it has changed the overlay table, this
12347 will enable @value{GDBN} to accurately keep track of which overlays
12348 are in program memory, and update any breakpoints that may be set
12349 in overlays. This will allow breakpoints to work even if the
12350 overlays are kept in ROM or other non-writable memory while they
12351 are not being executed.
12353 @node Overlay Sample Program
12354 @section Overlay Sample Program
12355 @cindex overlay example program
12357 When linking a program which uses overlays, you must place the overlays
12358 at their load addresses, while relocating them to run at their mapped
12359 addresses. To do this, you must write a linker script (@pxref{Overlay
12360 Description,,, ld.info, Using ld: the GNU linker}). Unfortunately,
12361 since linker scripts are specific to a particular host system, target
12362 architecture, and target memory layout, this manual cannot provide
12363 portable sample code demonstrating @value{GDBN}'s overlay support.
12365 However, the @value{GDBN} source distribution does contain an overlaid
12366 program, with linker scripts for a few systems, as part of its test
12367 suite. The program consists of the following files from
12368 @file{gdb/testsuite/gdb.base}:
12372 The main program file.
12374 A simple overlay manager, used by @file{overlays.c}.
12379 Overlay modules, loaded and used by @file{overlays.c}.
12382 Linker scripts for linking the test program on the @code{d10v-elf}
12383 and @code{m32r-elf} targets.
12386 You can build the test program using the @code{d10v-elf} GCC
12387 cross-compiler like this:
12390 $ d10v-elf-gcc -g -c overlays.c
12391 $ d10v-elf-gcc -g -c ovlymgr.c
12392 $ d10v-elf-gcc -g -c foo.c
12393 $ d10v-elf-gcc -g -c bar.c
12394 $ d10v-elf-gcc -g -c baz.c
12395 $ d10v-elf-gcc -g -c grbx.c
12396 $ d10v-elf-gcc -g overlays.o ovlymgr.o foo.o bar.o \
12397 baz.o grbx.o -Wl,-Td10v.ld -o overlays
12400 The build process is identical for any other architecture, except that
12401 you must substitute the appropriate compiler and linker script for the
12402 target system for @code{d10v-elf-gcc} and @code{d10v.ld}.
12406 @chapter Using @value{GDBN} with Different Languages
12409 Although programming languages generally have common aspects, they are
12410 rarely expressed in the same manner. For instance, in ANSI C,
12411 dereferencing a pointer @code{p} is accomplished by @code{*p}, but in
12412 Modula-2, it is accomplished by @code{p^}. Values can also be
12413 represented (and displayed) differently. Hex numbers in C appear as
12414 @samp{0x1ae}, while in Modula-2 they appear as @samp{1AEH}.
12416 @cindex working language
12417 Language-specific information is built into @value{GDBN} for some languages,
12418 allowing you to express operations like the above in your program's
12419 native language, and allowing @value{GDBN} to output values in a manner
12420 consistent with the syntax of your program's native language. The
12421 language you use to build expressions is called the @dfn{working
12425 * Setting:: Switching between source languages
12426 * Show:: Displaying the language
12427 * Checks:: Type and range checks
12428 * Supported Languages:: Supported languages
12429 * Unsupported Languages:: Unsupported languages
12433 @section Switching Between Source Languages
12435 There are two ways to control the working language---either have @value{GDBN}
12436 set it automatically, or select it manually yourself. You can use the
12437 @code{set language} command for either purpose. On startup, @value{GDBN}
12438 defaults to setting the language automatically. The working language is
12439 used to determine how expressions you type are interpreted, how values
12442 In addition to the working language, every source file that
12443 @value{GDBN} knows about has its own working language. For some object
12444 file formats, the compiler might indicate which language a particular
12445 source file is in. However, most of the time @value{GDBN} infers the
12446 language from the name of the file. The language of a source file
12447 controls whether C@t{++} names are demangled---this way @code{backtrace} can
12448 show each frame appropriately for its own language. There is no way to
12449 set the language of a source file from within @value{GDBN}, but you can
12450 set the language associated with a filename extension. @xref{Show, ,
12451 Displaying the Language}.
12453 This is most commonly a problem when you use a program, such
12454 as @code{cfront} or @code{f2c}, that generates C but is written in
12455 another language. In that case, make the
12456 program use @code{#line} directives in its C output; that way
12457 @value{GDBN} will know the correct language of the source code of the original
12458 program, and will display that source code, not the generated C code.
12461 * Filenames:: Filename extensions and languages.
12462 * Manually:: Setting the working language manually
12463 * Automatically:: Having @value{GDBN} infer the source language
12467 @subsection List of Filename Extensions and Languages
12469 If a source file name ends in one of the following extensions, then
12470 @value{GDBN} infers that its language is the one indicated.
12488 C@t{++} source file
12494 Objective-C source file
12498 Fortran source file
12501 Modula-2 source file
12505 Assembler source file. This actually behaves almost like C, but
12506 @value{GDBN} does not skip over function prologues when stepping.
12509 In addition, you may set the language associated with a filename
12510 extension. @xref{Show, , Displaying the Language}.
12513 @subsection Setting the Working Language
12515 If you allow @value{GDBN} to set the language automatically,
12516 expressions are interpreted the same way in your debugging session and
12519 @kindex set language
12520 If you wish, you may set the language manually. To do this, issue the
12521 command @samp{set language @var{lang}}, where @var{lang} is the name of
12522 a language, such as
12523 @code{c} or @code{modula-2}.
12524 For a list of the supported languages, type @samp{set language}.
12526 Setting the language manually prevents @value{GDBN} from updating the working
12527 language automatically. This can lead to confusion if you try
12528 to debug a program when the working language is not the same as the
12529 source language, when an expression is acceptable to both
12530 languages---but means different things. For instance, if the current
12531 source file were written in C, and @value{GDBN} was parsing Modula-2, a
12539 might not have the effect you intended. In C, this means to add
12540 @code{b} and @code{c} and place the result in @code{a}. The result
12541 printed would be the value of @code{a}. In Modula-2, this means to compare
12542 @code{a} to the result of @code{b+c}, yielding a @code{BOOLEAN} value.
12544 @node Automatically
12545 @subsection Having @value{GDBN} Infer the Source Language
12547 To have @value{GDBN} set the working language automatically, use
12548 @samp{set language local} or @samp{set language auto}. @value{GDBN}
12549 then infers the working language. That is, when your program stops in a
12550 frame (usually by encountering a breakpoint), @value{GDBN} sets the
12551 working language to the language recorded for the function in that
12552 frame. If the language for a frame is unknown (that is, if the function
12553 or block corresponding to the frame was defined in a source file that
12554 does not have a recognized extension), the current working language is
12555 not changed, and @value{GDBN} issues a warning.
12557 This may not seem necessary for most programs, which are written
12558 entirely in one source language. However, program modules and libraries
12559 written in one source language can be used by a main program written in
12560 a different source language. Using @samp{set language auto} in this
12561 case frees you from having to set the working language manually.
12564 @section Displaying the Language
12566 The following commands help you find out which language is the
12567 working language, and also what language source files were written in.
12570 @item show language
12571 @kindex show language
12572 Display the current working language. This is the
12573 language you can use with commands such as @code{print} to
12574 build and compute expressions that may involve variables in your program.
12577 @kindex info frame@r{, show the source language}
12578 Display the source language for this frame. This language becomes the
12579 working language if you use an identifier from this frame.
12580 @xref{Frame Info, ,Information about a Frame}, to identify the other
12581 information listed here.
12584 @kindex info source@r{, show the source language}
12585 Display the source language of this source file.
12586 @xref{Symbols, ,Examining the Symbol Table}, to identify the other
12587 information listed here.
12590 In unusual circumstances, you may have source files with extensions
12591 not in the standard list. You can then set the extension associated
12592 with a language explicitly:
12595 @item set extension-language @var{ext} @var{language}
12596 @kindex set extension-language
12597 Tell @value{GDBN} that source files with extension @var{ext} are to be
12598 assumed as written in the source language @var{language}.
12600 @item info extensions
12601 @kindex info extensions
12602 List all the filename extensions and the associated languages.
12606 @section Type and Range Checking
12609 @emph{Warning:} In this release, the @value{GDBN} commands for type and range
12610 checking are included, but they do not yet have any effect. This
12611 section documents the intended facilities.
12613 @c FIXME remove warning when type/range code added
12615 Some languages are designed to guard you against making seemingly common
12616 errors through a series of compile- and run-time checks. These include
12617 checking the type of arguments to functions and operators, and making
12618 sure mathematical overflows are caught at run time. Checks such as
12619 these help to ensure a program's correctness once it has been compiled
12620 by eliminating type mismatches, and providing active checks for range
12621 errors when your program is running.
12623 @value{GDBN} can check for conditions like the above if you wish.
12624 Although @value{GDBN} does not check the statements in your program,
12625 it can check expressions entered directly into @value{GDBN} for
12626 evaluation via the @code{print} command, for example. As with the
12627 working language, @value{GDBN} can also decide whether or not to check
12628 automatically based on your program's source language.
12629 @xref{Supported Languages, ,Supported Languages}, for the default
12630 settings of supported languages.
12633 * Type Checking:: An overview of type checking
12634 * Range Checking:: An overview of range checking
12637 @cindex type checking
12638 @cindex checks, type
12639 @node Type Checking
12640 @subsection An Overview of Type Checking
12642 Some languages, such as Modula-2, are strongly typed, meaning that the
12643 arguments to operators and functions have to be of the correct type,
12644 otherwise an error occurs. These checks prevent type mismatch
12645 errors from ever causing any run-time problems. For example,
12653 The second example fails because the @code{CARDINAL} 1 is not
12654 type-compatible with the @code{REAL} 2.3.
12656 For the expressions you use in @value{GDBN} commands, you can tell the
12657 @value{GDBN} type checker to skip checking;
12658 to treat any mismatches as errors and abandon the expression;
12659 or to only issue warnings when type mismatches occur,
12660 but evaluate the expression anyway. When you choose the last of
12661 these, @value{GDBN} evaluates expressions like the second example above, but
12662 also issues a warning.
12664 Even if you turn type checking off, there may be other reasons
12665 related to type that prevent @value{GDBN} from evaluating an expression.
12666 For instance, @value{GDBN} does not know how to add an @code{int} and
12667 a @code{struct foo}. These particular type errors have nothing to do
12668 with the language in use, and usually arise from expressions, such as
12669 the one described above, which make little sense to evaluate anyway.
12671 Each language defines to what degree it is strict about type. For
12672 instance, both Modula-2 and C require the arguments to arithmetical
12673 operators to be numbers. In C, enumerated types and pointers can be
12674 represented as numbers, so that they are valid arguments to mathematical
12675 operators. @xref{Supported Languages, ,Supported Languages}, for further
12676 details on specific languages.
12678 @value{GDBN} provides some additional commands for controlling the type checker:
12680 @kindex set check type
12681 @kindex show check type
12683 @item set check type auto
12684 Set type checking on or off based on the current working language.
12685 @xref{Supported Languages, ,Supported Languages}, for the default settings for
12688 @item set check type on
12689 @itemx set check type off
12690 Set type checking on or off, overriding the default setting for the
12691 current working language. Issue a warning if the setting does not
12692 match the language default. If any type mismatches occur in
12693 evaluating an expression while type checking is on, @value{GDBN} prints a
12694 message and aborts evaluation of the expression.
12696 @item set check type warn
12697 Cause the type checker to issue warnings, but to always attempt to
12698 evaluate the expression. Evaluating the expression may still
12699 be impossible for other reasons. For example, @value{GDBN} cannot add
12700 numbers and structures.
12703 Show the current setting of the type checker, and whether or not @value{GDBN}
12704 is setting it automatically.
12707 @cindex range checking
12708 @cindex checks, range
12709 @node Range Checking
12710 @subsection An Overview of Range Checking
12712 In some languages (such as Modula-2), it is an error to exceed the
12713 bounds of a type; this is enforced with run-time checks. Such range
12714 checking is meant to ensure program correctness by making sure
12715 computations do not overflow, or indices on an array element access do
12716 not exceed the bounds of the array.
12718 For expressions you use in @value{GDBN} commands, you can tell
12719 @value{GDBN} to treat range errors in one of three ways: ignore them,
12720 always treat them as errors and abandon the expression, or issue
12721 warnings but evaluate the expression anyway.
12723 A range error can result from numerical overflow, from exceeding an
12724 array index bound, or when you type a constant that is not a member
12725 of any type. Some languages, however, do not treat overflows as an
12726 error. In many implementations of C, mathematical overflow causes the
12727 result to ``wrap around'' to lower values---for example, if @var{m} is
12728 the largest integer value, and @var{s} is the smallest, then
12731 @var{m} + 1 @result{} @var{s}
12734 This, too, is specific to individual languages, and in some cases
12735 specific to individual compilers or machines. @xref{Supported Languages, ,
12736 Supported Languages}, for further details on specific languages.
12738 @value{GDBN} provides some additional commands for controlling the range checker:
12740 @kindex set check range
12741 @kindex show check range
12743 @item set check range auto
12744 Set range checking on or off based on the current working language.
12745 @xref{Supported Languages, ,Supported Languages}, for the default settings for
12748 @item set check range on
12749 @itemx set check range off
12750 Set range checking on or off, overriding the default setting for the
12751 current working language. A warning is issued if the setting does not
12752 match the language default. If a range error occurs and range checking is on,
12753 then a message is printed and evaluation of the expression is aborted.
12755 @item set check range warn
12756 Output messages when the @value{GDBN} range checker detects a range error,
12757 but attempt to evaluate the expression anyway. Evaluating the
12758 expression may still be impossible for other reasons, such as accessing
12759 memory that the process does not own (a typical example from many Unix
12763 Show the current setting of the range checker, and whether or not it is
12764 being set automatically by @value{GDBN}.
12767 @node Supported Languages
12768 @section Supported Languages
12770 @value{GDBN} supports C, C@t{++}, D, Go, Objective-C, Fortran, Java,
12771 OpenCL C, Pascal, assembly, Modula-2, and Ada.
12772 @c This is false ...
12773 Some @value{GDBN} features may be used in expressions regardless of the
12774 language you use: the @value{GDBN} @code{@@} and @code{::} operators,
12775 and the @samp{@{type@}addr} construct (@pxref{Expressions,
12776 ,Expressions}) can be used with the constructs of any supported
12779 The following sections detail to what degree each source language is
12780 supported by @value{GDBN}. These sections are not meant to be language
12781 tutorials or references, but serve only as a reference guide to what the
12782 @value{GDBN} expression parser accepts, and what input and output
12783 formats should look like for different languages. There are many good
12784 books written on each of these languages; please look to these for a
12785 language reference or tutorial.
12788 * C:: C and C@t{++}
12791 * Objective-C:: Objective-C
12792 * OpenCL C:: OpenCL C
12793 * Fortran:: Fortran
12795 * Modula-2:: Modula-2
12800 @subsection C and C@t{++}
12802 @cindex C and C@t{++}
12803 @cindex expressions in C or C@t{++}
12805 Since C and C@t{++} are so closely related, many features of @value{GDBN} apply
12806 to both languages. Whenever this is the case, we discuss those languages
12810 @cindex @code{g++}, @sc{gnu} C@t{++} compiler
12811 @cindex @sc{gnu} C@t{++}
12812 The C@t{++} debugging facilities are jointly implemented by the C@t{++}
12813 compiler and @value{GDBN}. Therefore, to debug your C@t{++} code
12814 effectively, you must compile your C@t{++} programs with a supported
12815 C@t{++} compiler, such as @sc{gnu} @code{g++}, or the HP ANSI C@t{++}
12816 compiler (@code{aCC}).
12819 * C Operators:: C and C@t{++} operators
12820 * C Constants:: C and C@t{++} constants
12821 * C Plus Plus Expressions:: C@t{++} expressions
12822 * C Defaults:: Default settings for C and C@t{++}
12823 * C Checks:: C and C@t{++} type and range checks
12824 * Debugging C:: @value{GDBN} and C
12825 * Debugging C Plus Plus:: @value{GDBN} features for C@t{++}
12826 * Decimal Floating Point:: Numbers in Decimal Floating Point format
12830 @subsubsection C and C@t{++} Operators
12832 @cindex C and C@t{++} operators
12834 Operators must be defined on values of specific types. For instance,
12835 @code{+} is defined on numbers, but not on structures. Operators are
12836 often defined on groups of types.
12838 For the purposes of C and C@t{++}, the following definitions hold:
12843 @emph{Integral types} include @code{int} with any of its storage-class
12844 specifiers; @code{char}; @code{enum}; and, for C@t{++}, @code{bool}.
12847 @emph{Floating-point types} include @code{float}, @code{double}, and
12848 @code{long double} (if supported by the target platform).
12851 @emph{Pointer types} include all types defined as @code{(@var{type} *)}.
12854 @emph{Scalar types} include all of the above.
12859 The following operators are supported. They are listed here
12860 in order of increasing precedence:
12864 The comma or sequencing operator. Expressions in a comma-separated list
12865 are evaluated from left to right, with the result of the entire
12866 expression being the last expression evaluated.
12869 Assignment. The value of an assignment expression is the value
12870 assigned. Defined on scalar types.
12873 Used in an expression of the form @w{@code{@var{a} @var{op}= @var{b}}},
12874 and translated to @w{@code{@var{a} = @var{a op b}}}.
12875 @w{@code{@var{op}=}} and @code{=} have the same precedence.
12876 @var{op} is any one of the operators @code{|}, @code{^}, @code{&},
12877 @code{<<}, @code{>>}, @code{+}, @code{-}, @code{*}, @code{/}, @code{%}.
12880 The ternary operator. @code{@var{a} ? @var{b} : @var{c}} can be thought
12881 of as: if @var{a} then @var{b} else @var{c}. @var{a} should be of an
12885 Logical @sc{or}. Defined on integral types.
12888 Logical @sc{and}. Defined on integral types.
12891 Bitwise @sc{or}. Defined on integral types.
12894 Bitwise exclusive-@sc{or}. Defined on integral types.
12897 Bitwise @sc{and}. Defined on integral types.
12900 Equality and inequality. Defined on scalar types. The value of these
12901 expressions is 0 for false and non-zero for true.
12903 @item <@r{, }>@r{, }<=@r{, }>=
12904 Less than, greater than, less than or equal, greater than or equal.
12905 Defined on scalar types. The value of these expressions is 0 for false
12906 and non-zero for true.
12909 left shift, and right shift. Defined on integral types.
12912 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
12915 Addition and subtraction. Defined on integral types, floating-point types and
12918 @item *@r{, }/@r{, }%
12919 Multiplication, division, and modulus. Multiplication and division are
12920 defined on integral and floating-point types. Modulus is defined on
12924 Increment and decrement. When appearing before a variable, the
12925 operation is performed before the variable is used in an expression;
12926 when appearing after it, the variable's value is used before the
12927 operation takes place.
12930 Pointer dereferencing. Defined on pointer types. Same precedence as
12934 Address operator. Defined on variables. Same precedence as @code{++}.
12936 For debugging C@t{++}, @value{GDBN} implements a use of @samp{&} beyond what is
12937 allowed in the C@t{++} language itself: you can use @samp{&(&@var{ref})}
12938 to examine the address
12939 where a C@t{++} reference variable (declared with @samp{&@var{ref}}) is
12943 Negative. Defined on integral and floating-point types. Same
12944 precedence as @code{++}.
12947 Logical negation. Defined on integral types. Same precedence as
12951 Bitwise complement operator. Defined on integral types. Same precedence as
12956 Structure member, and pointer-to-structure member. For convenience,
12957 @value{GDBN} regards the two as equivalent, choosing whether to dereference a
12958 pointer based on the stored type information.
12959 Defined on @code{struct} and @code{union} data.
12962 Dereferences of pointers to members.
12965 Array indexing. @code{@var{a}[@var{i}]} is defined as
12966 @code{*(@var{a}+@var{i})}. Same precedence as @code{->}.
12969 Function parameter list. Same precedence as @code{->}.
12972 C@t{++} scope resolution operator. Defined on @code{struct}, @code{union},
12973 and @code{class} types.
12976 Doubled colons also represent the @value{GDBN} scope operator
12977 (@pxref{Expressions, ,Expressions}). Same precedence as @code{::},
12981 If an operator is redefined in the user code, @value{GDBN} usually
12982 attempts to invoke the redefined version instead of using the operator's
12983 predefined meaning.
12986 @subsubsection C and C@t{++} Constants
12988 @cindex C and C@t{++} constants
12990 @value{GDBN} allows you to express the constants of C and C@t{++} in the
12995 Integer constants are a sequence of digits. Octal constants are
12996 specified by a leading @samp{0} (i.e.@: zero), and hexadecimal constants
12997 by a leading @samp{0x} or @samp{0X}. Constants may also end with a letter
12998 @samp{l}, specifying that the constant should be treated as a
13002 Floating point constants are a sequence of digits, followed by a decimal
13003 point, followed by a sequence of digits, and optionally followed by an
13004 exponent. An exponent is of the form:
13005 @samp{@w{e@r{[[}+@r{]|}-@r{]}@var{nnn}}}, where @var{nnn} is another
13006 sequence of digits. The @samp{+} is optional for positive exponents.
13007 A floating-point constant may also end with a letter @samp{f} or
13008 @samp{F}, specifying that the constant should be treated as being of
13009 the @code{float} (as opposed to the default @code{double}) type; or with
13010 a letter @samp{l} or @samp{L}, which specifies a @code{long double}
13014 Enumerated constants consist of enumerated identifiers, or their
13015 integral equivalents.
13018 Character constants are a single character surrounded by single quotes
13019 (@code{'}), or a number---the ordinal value of the corresponding character
13020 (usually its @sc{ascii} value). Within quotes, the single character may
13021 be represented by a letter or by @dfn{escape sequences}, which are of
13022 the form @samp{\@var{nnn}}, where @var{nnn} is the octal representation
13023 of the character's ordinal value; or of the form @samp{\@var{x}}, where
13024 @samp{@var{x}} is a predefined special character---for example,
13025 @samp{\n} for newline.
13027 Wide character constants can be written by prefixing a character
13028 constant with @samp{L}, as in C. For example, @samp{L'x'} is the wide
13029 form of @samp{x}. The target wide character set is used when
13030 computing the value of this constant (@pxref{Character Sets}).
13033 String constants are a sequence of character constants surrounded by
13034 double quotes (@code{"}). Any valid character constant (as described
13035 above) may appear. Double quotes within the string must be preceded by
13036 a backslash, so for instance @samp{"a\"b'c"} is a string of five
13039 Wide string constants can be written by prefixing a string constant
13040 with @samp{L}, as in C. The target wide character set is used when
13041 computing the value of this constant (@pxref{Character Sets}).
13044 Pointer constants are an integral value. You can also write pointers
13045 to constants using the C operator @samp{&}.
13048 Array constants are comma-separated lists surrounded by braces @samp{@{}
13049 and @samp{@}}; for example, @samp{@{1,2,3@}} is a three-element array of
13050 integers, @samp{@{@{1,2@}, @{3,4@}, @{5,6@}@}} is a three-by-two array,
13051 and @samp{@{&"hi", &"there", &"fred"@}} is a three-element array of pointers.
13054 @node C Plus Plus Expressions
13055 @subsubsection C@t{++} Expressions
13057 @cindex expressions in C@t{++}
13058 @value{GDBN} expression handling can interpret most C@t{++} expressions.
13060 @cindex debugging C@t{++} programs
13061 @cindex C@t{++} compilers
13062 @cindex debug formats and C@t{++}
13063 @cindex @value{NGCC} and C@t{++}
13065 @emph{Warning:} @value{GDBN} can only debug C@t{++} code if you use
13066 the proper compiler and the proper debug format. Currently,
13067 @value{GDBN} works best when debugging C@t{++} code that is compiled
13068 with the most recent version of @value{NGCC} possible. The DWARF
13069 debugging format is preferred; @value{NGCC} defaults to this on most
13070 popular platforms. Other compilers and/or debug formats are likely to
13071 work badly or not at all when using @value{GDBN} to debug C@t{++}
13072 code. @xref{Compilation}.
13077 @cindex member functions
13079 Member function calls are allowed; you can use expressions like
13082 count = aml->GetOriginal(x, y)
13085 @vindex this@r{, inside C@t{++} member functions}
13086 @cindex namespace in C@t{++}
13088 While a member function is active (in the selected stack frame), your
13089 expressions have the same namespace available as the member function;
13090 that is, @value{GDBN} allows implicit references to the class instance
13091 pointer @code{this} following the same rules as C@t{++}. @code{using}
13092 declarations in the current scope are also respected by @value{GDBN}.
13094 @cindex call overloaded functions
13095 @cindex overloaded functions, calling
13096 @cindex type conversions in C@t{++}
13098 You can call overloaded functions; @value{GDBN} resolves the function
13099 call to the right definition, with some restrictions. @value{GDBN} does not
13100 perform overload resolution involving user-defined type conversions,
13101 calls to constructors, or instantiations of templates that do not exist
13102 in the program. It also cannot handle ellipsis argument lists or
13105 It does perform integral conversions and promotions, floating-point
13106 promotions, arithmetic conversions, pointer conversions, conversions of
13107 class objects to base classes, and standard conversions such as those of
13108 functions or arrays to pointers; it requires an exact match on the
13109 number of function arguments.
13111 Overload resolution is always performed, unless you have specified
13112 @code{set overload-resolution off}. @xref{Debugging C Plus Plus,
13113 ,@value{GDBN} Features for C@t{++}}.
13115 You must specify @code{set overload-resolution off} in order to use an
13116 explicit function signature to call an overloaded function, as in
13118 p 'foo(char,int)'('x', 13)
13121 The @value{GDBN} command-completion facility can simplify this;
13122 see @ref{Completion, ,Command Completion}.
13124 @cindex reference declarations
13126 @value{GDBN} understands variables declared as C@t{++} references; you can use
13127 them in expressions just as you do in C@t{++} source---they are automatically
13130 In the parameter list shown when @value{GDBN} displays a frame, the values of
13131 reference variables are not displayed (unlike other variables); this
13132 avoids clutter, since references are often used for large structures.
13133 The @emph{address} of a reference variable is always shown, unless
13134 you have specified @samp{set print address off}.
13137 @value{GDBN} supports the C@t{++} name resolution operator @code{::}---your
13138 expressions can use it just as expressions in your program do. Since
13139 one scope may be defined in another, you can use @code{::} repeatedly if
13140 necessary, for example in an expression like
13141 @samp{@var{scope1}::@var{scope2}::@var{name}}. @value{GDBN} also allows
13142 resolving name scope by reference to source files, in both C and C@t{++}
13143 debugging (@pxref{Variables, ,Program Variables}).
13146 @value{GDBN} performs argument-dependent lookup, following the C@t{++}
13151 @subsubsection C and C@t{++} Defaults
13153 @cindex C and C@t{++} defaults
13155 If you allow @value{GDBN} to set type and range checking automatically, they
13156 both default to @code{off} whenever the working language changes to
13157 C or C@t{++}. This happens regardless of whether you or @value{GDBN}
13158 selects the working language.
13160 If you allow @value{GDBN} to set the language automatically, it
13161 recognizes source files whose names end with @file{.c}, @file{.C}, or
13162 @file{.cc}, etc, and when @value{GDBN} enters code compiled from one of
13163 these files, it sets the working language to C or C@t{++}.
13164 @xref{Automatically, ,Having @value{GDBN} Infer the Source Language},
13165 for further details.
13167 @c Type checking is (a) primarily motivated by Modula-2, and (b)
13168 @c unimplemented. If (b) changes, it might make sense to let this node
13169 @c appear even if Mod-2 does not, but meanwhile ignore it. roland 16jul93.
13172 @subsubsection C and C@t{++} Type and Range Checks
13174 @cindex C and C@t{++} checks
13176 By default, when @value{GDBN} parses C or C@t{++} expressions, type checking
13177 is not used. However, if you turn type checking on, @value{GDBN}
13178 considers two variables type equivalent if:
13182 The two variables are structured and have the same structure, union, or
13186 The two variables have the same type name, or types that have been
13187 declared equivalent through @code{typedef}.
13190 @c leaving this out because neither J Gilmore nor R Pesch understand it.
13193 The two @code{struct}, @code{union}, or @code{enum} variables are
13194 declared in the same declaration. (Note: this may not be true for all C
13199 Range checking, if turned on, is done on mathematical operations. Array
13200 indices are not checked, since they are often used to index a pointer
13201 that is not itself an array.
13204 @subsubsection @value{GDBN} and C
13206 The @code{set print union} and @code{show print union} commands apply to
13207 the @code{union} type. When set to @samp{on}, any @code{union} that is
13208 inside a @code{struct} or @code{class} is also printed. Otherwise, it
13209 appears as @samp{@{...@}}.
13211 The @code{@@} operator aids in the debugging of dynamic arrays, formed
13212 with pointers and a memory allocation function. @xref{Expressions,
13215 @node Debugging C Plus Plus
13216 @subsubsection @value{GDBN} Features for C@t{++}
13218 @cindex commands for C@t{++}
13220 Some @value{GDBN} commands are particularly useful with C@t{++}, and some are
13221 designed specifically for use with C@t{++}. Here is a summary:
13224 @cindex break in overloaded functions
13225 @item @r{breakpoint menus}
13226 When you want a breakpoint in a function whose name is overloaded,
13227 @value{GDBN} has the capability to display a menu of possible breakpoint
13228 locations to help you specify which function definition you want.
13229 @xref{Ambiguous Expressions,,Ambiguous Expressions}.
13231 @cindex overloading in C@t{++}
13232 @item rbreak @var{regex}
13233 Setting breakpoints using regular expressions is helpful for setting
13234 breakpoints on overloaded functions that are not members of any special
13236 @xref{Set Breaks, ,Setting Breakpoints}.
13238 @cindex C@t{++} exception handling
13241 Debug C@t{++} exception handling using these commands. @xref{Set
13242 Catchpoints, , Setting Catchpoints}.
13244 @cindex inheritance
13245 @item ptype @var{typename}
13246 Print inheritance relationships as well as other information for type
13248 @xref{Symbols, ,Examining the Symbol Table}.
13250 @item info vtbl @var{expression}.
13251 The @code{info vtbl} command can be used to display the virtual
13252 method tables of the object computed by @var{expression}. This shows
13253 one entry per virtual table; there may be multiple virtual tables when
13254 multiple inheritance is in use.
13256 @cindex C@t{++} symbol display
13257 @item set print demangle
13258 @itemx show print demangle
13259 @itemx set print asm-demangle
13260 @itemx show print asm-demangle
13261 Control whether C@t{++} symbols display in their source form, both when
13262 displaying code as C@t{++} source and when displaying disassemblies.
13263 @xref{Print Settings, ,Print Settings}.
13265 @item set print object
13266 @itemx show print object
13267 Choose whether to print derived (actual) or declared types of objects.
13268 @xref{Print Settings, ,Print Settings}.
13270 @item set print vtbl
13271 @itemx show print vtbl
13272 Control the format for printing virtual function tables.
13273 @xref{Print Settings, ,Print Settings}.
13274 (The @code{vtbl} commands do not work on programs compiled with the HP
13275 ANSI C@t{++} compiler (@code{aCC}).)
13277 @kindex set overload-resolution
13278 @cindex overloaded functions, overload resolution
13279 @item set overload-resolution on
13280 Enable overload resolution for C@t{++} expression evaluation. The default
13281 is on. For overloaded functions, @value{GDBN} evaluates the arguments
13282 and searches for a function whose signature matches the argument types,
13283 using the standard C@t{++} conversion rules (see @ref{C Plus Plus
13284 Expressions, ,C@t{++} Expressions}, for details).
13285 If it cannot find a match, it emits a message.
13287 @item set overload-resolution off
13288 Disable overload resolution for C@t{++} expression evaluation. For
13289 overloaded functions that are not class member functions, @value{GDBN}
13290 chooses the first function of the specified name that it finds in the
13291 symbol table, whether or not its arguments are of the correct type. For
13292 overloaded functions that are class member functions, @value{GDBN}
13293 searches for a function whose signature @emph{exactly} matches the
13296 @kindex show overload-resolution
13297 @item show overload-resolution
13298 Show the current setting of overload resolution.
13300 @item @r{Overloaded symbol names}
13301 You can specify a particular definition of an overloaded symbol, using
13302 the same notation that is used to declare such symbols in C@t{++}: type
13303 @code{@var{symbol}(@var{types})} rather than just @var{symbol}. You can
13304 also use the @value{GDBN} command-line word completion facilities to list the
13305 available choices, or to finish the type list for you.
13306 @xref{Completion,, Command Completion}, for details on how to do this.
13309 @node Decimal Floating Point
13310 @subsubsection Decimal Floating Point format
13311 @cindex decimal floating point format
13313 @value{GDBN} can examine, set and perform computations with numbers in
13314 decimal floating point format, which in the C language correspond to the
13315 @code{_Decimal32}, @code{_Decimal64} and @code{_Decimal128} types as
13316 specified by the extension to support decimal floating-point arithmetic.
13318 There are two encodings in use, depending on the architecture: BID (Binary
13319 Integer Decimal) for x86 and x86-64, and DPD (Densely Packed Decimal) for
13320 PowerPC. @value{GDBN} will use the appropriate encoding for the configured
13323 Because of a limitation in @file{libdecnumber}, the library used by @value{GDBN}
13324 to manipulate decimal floating point numbers, it is not possible to convert
13325 (using a cast, for example) integers wider than 32-bit to decimal float.
13327 In addition, in order to imitate @value{GDBN}'s behaviour with binary floating
13328 point computations, error checking in decimal float operations ignores
13329 underflow, overflow and divide by zero exceptions.
13331 In the PowerPC architecture, @value{GDBN} provides a set of pseudo-registers
13332 to inspect @code{_Decimal128} values stored in floating point registers.
13333 See @ref{PowerPC,,PowerPC} for more details.
13339 @value{GDBN} can be used to debug programs written in D and compiled with
13340 GDC, LDC or DMD compilers. Currently @value{GDBN} supports only one D
13341 specific feature --- dynamic arrays.
13346 @cindex Go (programming language)
13347 @value{GDBN} can be used to debug programs written in Go and compiled with
13348 @file{gccgo} or @file{6g} compilers.
13350 Here is a summary of the Go-specific features and restrictions:
13353 @cindex current Go package
13354 @item The current Go package
13355 The name of the current package does not need to be specified when
13356 specifying global variables and functions.
13358 For example, given the program:
13362 var myglob = "Shall we?"
13368 When stopped inside @code{main} either of these work:
13372 (gdb) p main.myglob
13375 @cindex builtin Go types
13376 @item Builtin Go types
13377 The @code{string} type is recognized by @value{GDBN} and is printed
13380 @cindex builtin Go functions
13381 @item Builtin Go functions
13382 The @value{GDBN} expression parser recognizes the @code{unsafe.Sizeof}
13383 function and handles it internally.
13385 @cindex restrictions on Go expressions
13386 @item Restrictions on Go expressions
13387 All Go operators are supported except @code{&^}.
13388 The Go @code{_} ``blank identifier'' is not supported.
13389 Automatic dereferencing of pointers is not supported.
13393 @subsection Objective-C
13395 @cindex Objective-C
13396 This section provides information about some commands and command
13397 options that are useful for debugging Objective-C code. See also
13398 @ref{Symbols, info classes}, and @ref{Symbols, info selectors}, for a
13399 few more commands specific to Objective-C support.
13402 * Method Names in Commands::
13403 * The Print Command with Objective-C::
13406 @node Method Names in Commands
13407 @subsubsection Method Names in Commands
13409 The following commands have been extended to accept Objective-C method
13410 names as line specifications:
13412 @kindex clear@r{, and Objective-C}
13413 @kindex break@r{, and Objective-C}
13414 @kindex info line@r{, and Objective-C}
13415 @kindex jump@r{, and Objective-C}
13416 @kindex list@r{, and Objective-C}
13420 @item @code{info line}
13425 A fully qualified Objective-C method name is specified as
13428 -[@var{Class} @var{methodName}]
13431 where the minus sign is used to indicate an instance method and a
13432 plus sign (not shown) is used to indicate a class method. The class
13433 name @var{Class} and method name @var{methodName} are enclosed in
13434 brackets, similar to the way messages are specified in Objective-C
13435 source code. For example, to set a breakpoint at the @code{create}
13436 instance method of class @code{Fruit} in the program currently being
13440 break -[Fruit create]
13443 To list ten program lines around the @code{initialize} class method,
13447 list +[NSText initialize]
13450 In the current version of @value{GDBN}, the plus or minus sign is
13451 required. In future versions of @value{GDBN}, the plus or minus
13452 sign will be optional, but you can use it to narrow the search. It
13453 is also possible to specify just a method name:
13459 You must specify the complete method name, including any colons. If
13460 your program's source files contain more than one @code{create} method,
13461 you'll be presented with a numbered list of classes that implement that
13462 method. Indicate your choice by number, or type @samp{0} to exit if
13465 As another example, to clear a breakpoint established at the
13466 @code{makeKeyAndOrderFront:} method of the @code{NSWindow} class, enter:
13469 clear -[NSWindow makeKeyAndOrderFront:]
13472 @node The Print Command with Objective-C
13473 @subsubsection The Print Command With Objective-C
13474 @cindex Objective-C, print objects
13475 @kindex print-object
13476 @kindex po @r{(@code{print-object})}
13478 The print command has also been extended to accept methods. For example:
13481 print -[@var{object} hash]
13484 @cindex print an Objective-C object description
13485 @cindex @code{_NSPrintForDebugger}, and printing Objective-C objects
13487 will tell @value{GDBN} to send the @code{hash} message to @var{object}
13488 and print the result. Also, an additional command has been added,
13489 @code{print-object} or @code{po} for short, which is meant to print
13490 the description of an object. However, this command may only work
13491 with certain Objective-C libraries that have a particular hook
13492 function, @code{_NSPrintForDebugger}, defined.
13495 @subsection OpenCL C
13498 This section provides information about @value{GDBN}s OpenCL C support.
13501 * OpenCL C Datatypes::
13502 * OpenCL C Expressions::
13503 * OpenCL C Operators::
13506 @node OpenCL C Datatypes
13507 @subsubsection OpenCL C Datatypes
13509 @cindex OpenCL C Datatypes
13510 @value{GDBN} supports the builtin scalar and vector datatypes specified
13511 by OpenCL 1.1. In addition the half- and double-precision floating point
13512 data types of the @code{cl_khr_fp16} and @code{cl_khr_fp64} OpenCL
13513 extensions are also known to @value{GDBN}.
13515 @node OpenCL C Expressions
13516 @subsubsection OpenCL C Expressions
13518 @cindex OpenCL C Expressions
13519 @value{GDBN} supports accesses to vector components including the access as
13520 lvalue where possible. Since OpenCL C is based on C99 most C expressions
13521 supported by @value{GDBN} can be used as well.
13523 @node OpenCL C Operators
13524 @subsubsection OpenCL C Operators
13526 @cindex OpenCL C Operators
13527 @value{GDBN} supports the operators specified by OpenCL 1.1 for scalar and
13531 @subsection Fortran
13532 @cindex Fortran-specific support in @value{GDBN}
13534 @value{GDBN} can be used to debug programs written in Fortran, but it
13535 currently supports only the features of Fortran 77 language.
13537 @cindex trailing underscore, in Fortran symbols
13538 Some Fortran compilers (@sc{gnu} Fortran 77 and Fortran 95 compilers
13539 among them) append an underscore to the names of variables and
13540 functions. When you debug programs compiled by those compilers, you
13541 will need to refer to variables and functions with a trailing
13545 * Fortran Operators:: Fortran operators and expressions
13546 * Fortran Defaults:: Default settings for Fortran
13547 * Special Fortran Commands:: Special @value{GDBN} commands for Fortran
13550 @node Fortran Operators
13551 @subsubsection Fortran Operators and Expressions
13553 @cindex Fortran operators and expressions
13555 Operators must be defined on values of specific types. For instance,
13556 @code{+} is defined on numbers, but not on characters or other non-
13557 arithmetic types. Operators are often defined on groups of types.
13561 The exponentiation operator. It raises the first operand to the power
13565 The range operator. Normally used in the form of array(low:high) to
13566 represent a section of array.
13569 The access component operator. Normally used to access elements in derived
13570 types. Also suitable for unions. As unions aren't part of regular Fortran,
13571 this can only happen when accessing a register that uses a gdbarch-defined
13575 @node Fortran Defaults
13576 @subsubsection Fortran Defaults
13578 @cindex Fortran Defaults
13580 Fortran symbols are usually case-insensitive, so @value{GDBN} by
13581 default uses case-insensitive matches for Fortran symbols. You can
13582 change that with the @samp{set case-insensitive} command, see
13583 @ref{Symbols}, for the details.
13585 @node Special Fortran Commands
13586 @subsubsection Special Fortran Commands
13588 @cindex Special Fortran commands
13590 @value{GDBN} has some commands to support Fortran-specific features,
13591 such as displaying common blocks.
13594 @cindex @code{COMMON} blocks, Fortran
13595 @kindex info common
13596 @item info common @r{[}@var{common-name}@r{]}
13597 This command prints the values contained in the Fortran @code{COMMON}
13598 block whose name is @var{common-name}. With no argument, the names of
13599 all @code{COMMON} blocks visible at the current program location are
13606 @cindex Pascal support in @value{GDBN}, limitations
13607 Debugging Pascal programs which use sets, subranges, file variables, or
13608 nested functions does not currently work. @value{GDBN} does not support
13609 entering expressions, printing values, or similar features using Pascal
13612 The Pascal-specific command @code{set print pascal_static-members}
13613 controls whether static members of Pascal objects are displayed.
13614 @xref{Print Settings, pascal_static-members}.
13617 @subsection Modula-2
13619 @cindex Modula-2, @value{GDBN} support
13621 The extensions made to @value{GDBN} to support Modula-2 only support
13622 output from the @sc{gnu} Modula-2 compiler (which is currently being
13623 developed). Other Modula-2 compilers are not currently supported, and
13624 attempting to debug executables produced by them is most likely
13625 to give an error as @value{GDBN} reads in the executable's symbol
13628 @cindex expressions in Modula-2
13630 * M2 Operators:: Built-in operators
13631 * Built-In Func/Proc:: Built-in functions and procedures
13632 * M2 Constants:: Modula-2 constants
13633 * M2 Types:: Modula-2 types
13634 * M2 Defaults:: Default settings for Modula-2
13635 * Deviations:: Deviations from standard Modula-2
13636 * M2 Checks:: Modula-2 type and range checks
13637 * M2 Scope:: The scope operators @code{::} and @code{.}
13638 * GDB/M2:: @value{GDBN} and Modula-2
13642 @subsubsection Operators
13643 @cindex Modula-2 operators
13645 Operators must be defined on values of specific types. For instance,
13646 @code{+} is defined on numbers, but not on structures. Operators are
13647 often defined on groups of types. For the purposes of Modula-2, the
13648 following definitions hold:
13653 @emph{Integral types} consist of @code{INTEGER}, @code{CARDINAL}, and
13657 @emph{Character types} consist of @code{CHAR} and its subranges.
13660 @emph{Floating-point types} consist of @code{REAL}.
13663 @emph{Pointer types} consist of anything declared as @code{POINTER TO
13667 @emph{Scalar types} consist of all of the above.
13670 @emph{Set types} consist of @code{SET} and @code{BITSET} types.
13673 @emph{Boolean types} consist of @code{BOOLEAN}.
13677 The following operators are supported, and appear in order of
13678 increasing precedence:
13682 Function argument or array index separator.
13685 Assignment. The value of @var{var} @code{:=} @var{value} is
13689 Less than, greater than on integral, floating-point, or enumerated
13693 Less than or equal to, greater than or equal to
13694 on integral, floating-point and enumerated types, or set inclusion on
13695 set types. Same precedence as @code{<}.
13697 @item =@r{, }<>@r{, }#
13698 Equality and two ways of expressing inequality, valid on scalar types.
13699 Same precedence as @code{<}. In @value{GDBN} scripts, only @code{<>} is
13700 available for inequality, since @code{#} conflicts with the script
13704 Set membership. Defined on set types and the types of their members.
13705 Same precedence as @code{<}.
13708 Boolean disjunction. Defined on boolean types.
13711 Boolean conjunction. Defined on boolean types.
13714 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
13717 Addition and subtraction on integral and floating-point types, or union
13718 and difference on set types.
13721 Multiplication on integral and floating-point types, or set intersection
13725 Division on floating-point types, or symmetric set difference on set
13726 types. Same precedence as @code{*}.
13729 Integer division and remainder. Defined on integral types. Same
13730 precedence as @code{*}.
13733 Negative. Defined on @code{INTEGER} and @code{REAL} data.
13736 Pointer dereferencing. Defined on pointer types.
13739 Boolean negation. Defined on boolean types. Same precedence as
13743 @code{RECORD} field selector. Defined on @code{RECORD} data. Same
13744 precedence as @code{^}.
13747 Array indexing. Defined on @code{ARRAY} data. Same precedence as @code{^}.
13750 Procedure argument list. Defined on @code{PROCEDURE} objects. Same precedence
13754 @value{GDBN} and Modula-2 scope operators.
13758 @emph{Warning:} Set expressions and their operations are not yet supported, so @value{GDBN}
13759 treats the use of the operator @code{IN}, or the use of operators
13760 @code{+}, @code{-}, @code{*}, @code{/}, @code{=}, , @code{<>}, @code{#},
13761 @code{<=}, and @code{>=} on sets as an error.
13765 @node Built-In Func/Proc
13766 @subsubsection Built-in Functions and Procedures
13767 @cindex Modula-2 built-ins
13769 Modula-2 also makes available several built-in procedures and functions.
13770 In describing these, the following metavariables are used:
13775 represents an @code{ARRAY} variable.
13778 represents a @code{CHAR} constant or variable.
13781 represents a variable or constant of integral type.
13784 represents an identifier that belongs to a set. Generally used in the
13785 same function with the metavariable @var{s}. The type of @var{s} should
13786 be @code{SET OF @var{mtype}} (where @var{mtype} is the type of @var{m}).
13789 represents a variable or constant of integral or floating-point type.
13792 represents a variable or constant of floating-point type.
13798 represents a variable.
13801 represents a variable or constant of one of many types. See the
13802 explanation of the function for details.
13805 All Modula-2 built-in procedures also return a result, described below.
13809 Returns the absolute value of @var{n}.
13812 If @var{c} is a lower case letter, it returns its upper case
13813 equivalent, otherwise it returns its argument.
13816 Returns the character whose ordinal value is @var{i}.
13819 Decrements the value in the variable @var{v} by one. Returns the new value.
13821 @item DEC(@var{v},@var{i})
13822 Decrements the value in the variable @var{v} by @var{i}. Returns the
13825 @item EXCL(@var{m},@var{s})
13826 Removes the element @var{m} from the set @var{s}. Returns the new
13829 @item FLOAT(@var{i})
13830 Returns the floating point equivalent of the integer @var{i}.
13832 @item HIGH(@var{a})
13833 Returns the index of the last member of @var{a}.
13836 Increments the value in the variable @var{v} by one. Returns the new value.
13838 @item INC(@var{v},@var{i})
13839 Increments the value in the variable @var{v} by @var{i}. Returns the
13842 @item INCL(@var{m},@var{s})
13843 Adds the element @var{m} to the set @var{s} if it is not already
13844 there. Returns the new set.
13847 Returns the maximum value of the type @var{t}.
13850 Returns the minimum value of the type @var{t}.
13853 Returns boolean TRUE if @var{i} is an odd number.
13856 Returns the ordinal value of its argument. For example, the ordinal
13857 value of a character is its @sc{ascii} value (on machines supporting the
13858 @sc{ascii} character set). @var{x} must be of an ordered type, which include
13859 integral, character and enumerated types.
13861 @item SIZE(@var{x})
13862 Returns the size of its argument. @var{x} can be a variable or a type.
13864 @item TRUNC(@var{r})
13865 Returns the integral part of @var{r}.
13867 @item TSIZE(@var{x})
13868 Returns the size of its argument. @var{x} can be a variable or a type.
13870 @item VAL(@var{t},@var{i})
13871 Returns the member of the type @var{t} whose ordinal value is @var{i}.
13875 @emph{Warning:} Sets and their operations are not yet supported, so
13876 @value{GDBN} treats the use of procedures @code{INCL} and @code{EXCL} as
13880 @cindex Modula-2 constants
13882 @subsubsection Constants
13884 @value{GDBN} allows you to express the constants of Modula-2 in the following
13890 Integer constants are simply a sequence of digits. When used in an
13891 expression, a constant is interpreted to be type-compatible with the
13892 rest of the expression. Hexadecimal integers are specified by a
13893 trailing @samp{H}, and octal integers by a trailing @samp{B}.
13896 Floating point constants appear as a sequence of digits, followed by a
13897 decimal point and another sequence of digits. An optional exponent can
13898 then be specified, in the form @samp{E@r{[}+@r{|}-@r{]}@var{nnn}}, where
13899 @samp{@r{[}+@r{|}-@r{]}@var{nnn}} is the desired exponent. All of the
13900 digits of the floating point constant must be valid decimal (base 10)
13904 Character constants consist of a single character enclosed by a pair of
13905 like quotes, either single (@code{'}) or double (@code{"}). They may
13906 also be expressed by their ordinal value (their @sc{ascii} value, usually)
13907 followed by a @samp{C}.
13910 String constants consist of a sequence of characters enclosed by a
13911 pair of like quotes, either single (@code{'}) or double (@code{"}).
13912 Escape sequences in the style of C are also allowed. @xref{C
13913 Constants, ,C and C@t{++} Constants}, for a brief explanation of escape
13917 Enumerated constants consist of an enumerated identifier.
13920 Boolean constants consist of the identifiers @code{TRUE} and
13924 Pointer constants consist of integral values only.
13927 Set constants are not yet supported.
13931 @subsubsection Modula-2 Types
13932 @cindex Modula-2 types
13934 Currently @value{GDBN} can print the following data types in Modula-2
13935 syntax: array types, record types, set types, pointer types, procedure
13936 types, enumerated types, subrange types and base types. You can also
13937 print the contents of variables declared using these type.
13938 This section gives a number of simple source code examples together with
13939 sample @value{GDBN} sessions.
13941 The first example contains the following section of code:
13950 and you can request @value{GDBN} to interrogate the type and value of
13951 @code{r} and @code{s}.
13954 (@value{GDBP}) print s
13956 (@value{GDBP}) ptype s
13958 (@value{GDBP}) print r
13960 (@value{GDBP}) ptype r
13965 Likewise if your source code declares @code{s} as:
13969 s: SET ['A'..'Z'] ;
13973 then you may query the type of @code{s} by:
13976 (@value{GDBP}) ptype s
13977 type = SET ['A'..'Z']
13981 Note that at present you cannot interactively manipulate set
13982 expressions using the debugger.
13984 The following example shows how you might declare an array in Modula-2
13985 and how you can interact with @value{GDBN} to print its type and contents:
13989 s: ARRAY [-10..10] OF CHAR ;
13993 (@value{GDBP}) ptype s
13994 ARRAY [-10..10] OF CHAR
13997 Note that the array handling is not yet complete and although the type
13998 is printed correctly, expression handling still assumes that all
13999 arrays have a lower bound of zero and not @code{-10} as in the example
14002 Here are some more type related Modula-2 examples:
14006 colour = (blue, red, yellow, green) ;
14007 t = [blue..yellow] ;
14015 The @value{GDBN} interaction shows how you can query the data type
14016 and value of a variable.
14019 (@value{GDBP}) print s
14021 (@value{GDBP}) ptype t
14022 type = [blue..yellow]
14026 In this example a Modula-2 array is declared and its contents
14027 displayed. Observe that the contents are written in the same way as
14028 their @code{C} counterparts.
14032 s: ARRAY [1..5] OF CARDINAL ;
14038 (@value{GDBP}) print s
14039 $1 = @{1, 0, 0, 0, 0@}
14040 (@value{GDBP}) ptype s
14041 type = ARRAY [1..5] OF CARDINAL
14044 The Modula-2 language interface to @value{GDBN} also understands
14045 pointer types as shown in this example:
14049 s: POINTER TO ARRAY [1..5] OF CARDINAL ;
14056 and you can request that @value{GDBN} describes the type of @code{s}.
14059 (@value{GDBP}) ptype s
14060 type = POINTER TO ARRAY [1..5] OF CARDINAL
14063 @value{GDBN} handles compound types as we can see in this example.
14064 Here we combine array types, record types, pointer types and subrange
14075 myarray = ARRAY myrange OF CARDINAL ;
14076 myrange = [-2..2] ;
14078 s: POINTER TO ARRAY myrange OF foo ;
14082 and you can ask @value{GDBN} to describe the type of @code{s} as shown
14086 (@value{GDBP}) ptype s
14087 type = POINTER TO ARRAY [-2..2] OF foo = RECORD
14090 f3 : ARRAY [-2..2] OF CARDINAL;
14095 @subsubsection Modula-2 Defaults
14096 @cindex Modula-2 defaults
14098 If type and range checking are set automatically by @value{GDBN}, they
14099 both default to @code{on} whenever the working language changes to
14100 Modula-2. This happens regardless of whether you or @value{GDBN}
14101 selected the working language.
14103 If you allow @value{GDBN} to set the language automatically, then entering
14104 code compiled from a file whose name ends with @file{.mod} sets the
14105 working language to Modula-2. @xref{Automatically, ,Having @value{GDBN}
14106 Infer the Source Language}, for further details.
14109 @subsubsection Deviations from Standard Modula-2
14110 @cindex Modula-2, deviations from
14112 A few changes have been made to make Modula-2 programs easier to debug.
14113 This is done primarily via loosening its type strictness:
14117 Unlike in standard Modula-2, pointer constants can be formed by
14118 integers. This allows you to modify pointer variables during
14119 debugging. (In standard Modula-2, the actual address contained in a
14120 pointer variable is hidden from you; it can only be modified
14121 through direct assignment to another pointer variable or expression that
14122 returned a pointer.)
14125 C escape sequences can be used in strings and characters to represent
14126 non-printable characters. @value{GDBN} prints out strings with these
14127 escape sequences embedded. Single non-printable characters are
14128 printed using the @samp{CHR(@var{nnn})} format.
14131 The assignment operator (@code{:=}) returns the value of its right-hand
14135 All built-in procedures both modify @emph{and} return their argument.
14139 @subsubsection Modula-2 Type and Range Checks
14140 @cindex Modula-2 checks
14143 @emph{Warning:} in this release, @value{GDBN} does not yet perform type or
14146 @c FIXME remove warning when type/range checks added
14148 @value{GDBN} considers two Modula-2 variables type equivalent if:
14152 They are of types that have been declared equivalent via a @code{TYPE
14153 @var{t1} = @var{t2}} statement
14156 They have been declared on the same line. (Note: This is true of the
14157 @sc{gnu} Modula-2 compiler, but it may not be true of other compilers.)
14160 As long as type checking is enabled, any attempt to combine variables
14161 whose types are not equivalent is an error.
14163 Range checking is done on all mathematical operations, assignment, array
14164 index bounds, and all built-in functions and procedures.
14167 @subsubsection The Scope Operators @code{::} and @code{.}
14169 @cindex @code{.}, Modula-2 scope operator
14170 @cindex colon, doubled as scope operator
14172 @vindex colon-colon@r{, in Modula-2}
14173 @c Info cannot handle :: but TeX can.
14176 @vindex ::@r{, in Modula-2}
14179 There are a few subtle differences between the Modula-2 scope operator
14180 (@code{.}) and the @value{GDBN} scope operator (@code{::}). The two have
14185 @var{module} . @var{id}
14186 @var{scope} :: @var{id}
14190 where @var{scope} is the name of a module or a procedure,
14191 @var{module} the name of a module, and @var{id} is any declared
14192 identifier within your program, except another module.
14194 Using the @code{::} operator makes @value{GDBN} search the scope
14195 specified by @var{scope} for the identifier @var{id}. If it is not
14196 found in the specified scope, then @value{GDBN} searches all scopes
14197 enclosing the one specified by @var{scope}.
14199 Using the @code{.} operator makes @value{GDBN} search the current scope for
14200 the identifier specified by @var{id} that was imported from the
14201 definition module specified by @var{module}. With this operator, it is
14202 an error if the identifier @var{id} was not imported from definition
14203 module @var{module}, or if @var{id} is not an identifier in
14207 @subsubsection @value{GDBN} and Modula-2
14209 Some @value{GDBN} commands have little use when debugging Modula-2 programs.
14210 Five subcommands of @code{set print} and @code{show print} apply
14211 specifically to C and C@t{++}: @samp{vtbl}, @samp{demangle},
14212 @samp{asm-demangle}, @samp{object}, and @samp{union}. The first four
14213 apply to C@t{++}, and the last to the C @code{union} type, which has no direct
14214 analogue in Modula-2.
14216 The @code{@@} operator (@pxref{Expressions, ,Expressions}), while available
14217 with any language, is not useful with Modula-2. Its
14218 intent is to aid the debugging of @dfn{dynamic arrays}, which cannot be
14219 created in Modula-2 as they can in C or C@t{++}. However, because an
14220 address can be specified by an integral constant, the construct
14221 @samp{@{@var{type}@}@var{adrexp}} is still useful.
14223 @cindex @code{#} in Modula-2
14224 In @value{GDBN} scripts, the Modula-2 inequality operator @code{#} is
14225 interpreted as the beginning of a comment. Use @code{<>} instead.
14231 The extensions made to @value{GDBN} for Ada only support
14232 output from the @sc{gnu} Ada (GNAT) compiler.
14233 Other Ada compilers are not currently supported, and
14234 attempting to debug executables produced by them is most likely
14238 @cindex expressions in Ada
14240 * Ada Mode Intro:: General remarks on the Ada syntax
14241 and semantics supported by Ada mode
14243 * Omissions from Ada:: Restrictions on the Ada expression syntax.
14244 * Additions to Ada:: Extensions of the Ada expression syntax.
14245 * Stopping Before Main Program:: Debugging the program during elaboration.
14246 * Ada Tasks:: Listing and setting breakpoints in tasks.
14247 * Ada Tasks and Core Files:: Tasking Support when Debugging Core Files
14248 * Ravenscar Profile:: Tasking Support when using the Ravenscar
14250 * Ada Glitches:: Known peculiarities of Ada mode.
14253 @node Ada Mode Intro
14254 @subsubsection Introduction
14255 @cindex Ada mode, general
14257 The Ada mode of @value{GDBN} supports a fairly large subset of Ada expression
14258 syntax, with some extensions.
14259 The philosophy behind the design of this subset is
14263 That @value{GDBN} should provide basic literals and access to operations for
14264 arithmetic, dereferencing, field selection, indexing, and subprogram calls,
14265 leaving more sophisticated computations to subprograms written into the
14266 program (which therefore may be called from @value{GDBN}).
14269 That type safety and strict adherence to Ada language restrictions
14270 are not particularly important to the @value{GDBN} user.
14273 That brevity is important to the @value{GDBN} user.
14276 Thus, for brevity, the debugger acts as if all names declared in
14277 user-written packages are directly visible, even if they are not visible
14278 according to Ada rules, thus making it unnecessary to fully qualify most
14279 names with their packages, regardless of context. Where this causes
14280 ambiguity, @value{GDBN} asks the user's intent.
14282 The debugger will start in Ada mode if it detects an Ada main program.
14283 As for other languages, it will enter Ada mode when stopped in a program that
14284 was translated from an Ada source file.
14286 While in Ada mode, you may use `@t{--}' for comments. This is useful
14287 mostly for documenting command files. The standard @value{GDBN} comment
14288 (@samp{#}) still works at the beginning of a line in Ada mode, but not in the
14289 middle (to allow based literals).
14291 The debugger supports limited overloading. Given a subprogram call in which
14292 the function symbol has multiple definitions, it will use the number of
14293 actual parameters and some information about their types to attempt to narrow
14294 the set of definitions. It also makes very limited use of context, preferring
14295 procedures to functions in the context of the @code{call} command, and
14296 functions to procedures elsewhere.
14298 @node Omissions from Ada
14299 @subsubsection Omissions from Ada
14300 @cindex Ada, omissions from
14302 Here are the notable omissions from the subset:
14306 Only a subset of the attributes are supported:
14310 @t{'First}, @t{'Last}, and @t{'Length}
14311 on array objects (not on types and subtypes).
14314 @t{'Min} and @t{'Max}.
14317 @t{'Pos} and @t{'Val}.
14323 @t{'Range} on array objects (not subtypes), but only as the right
14324 operand of the membership (@code{in}) operator.
14327 @t{'Access}, @t{'Unchecked_Access}, and
14328 @t{'Unrestricted_Access} (a GNAT extension).
14336 @code{Characters.Latin_1} are not available and
14337 concatenation is not implemented. Thus, escape characters in strings are
14338 not currently available.
14341 Equality tests (@samp{=} and @samp{/=}) on arrays test for bitwise
14342 equality of representations. They will generally work correctly
14343 for strings and arrays whose elements have integer or enumeration types.
14344 They may not work correctly for arrays whose element
14345 types have user-defined equality, for arrays of real values
14346 (in particular, IEEE-conformant floating point, because of negative
14347 zeroes and NaNs), and for arrays whose elements contain unused bits with
14348 indeterminate values.
14351 The other component-by-component array operations (@code{and}, @code{or},
14352 @code{xor}, @code{not}, and relational tests other than equality)
14353 are not implemented.
14356 @cindex array aggregates (Ada)
14357 @cindex record aggregates (Ada)
14358 @cindex aggregates (Ada)
14359 There is limited support for array and record aggregates. They are
14360 permitted only on the right sides of assignments, as in these examples:
14363 (@value{GDBP}) set An_Array := (1, 2, 3, 4, 5, 6)
14364 (@value{GDBP}) set An_Array := (1, others => 0)
14365 (@value{GDBP}) set An_Array := (0|4 => 1, 1..3 => 2, 5 => 6)
14366 (@value{GDBP}) set A_2D_Array := ((1, 2, 3), (4, 5, 6), (7, 8, 9))
14367 (@value{GDBP}) set A_Record := (1, "Peter", True);
14368 (@value{GDBP}) set A_Record := (Name => "Peter", Id => 1, Alive => True)
14372 discriminant's value by assigning an aggregate has an
14373 undefined effect if that discriminant is used within the record.
14374 However, you can first modify discriminants by directly assigning to
14375 them (which normally would not be allowed in Ada), and then performing an
14376 aggregate assignment. For example, given a variable @code{A_Rec}
14377 declared to have a type such as:
14380 type Rec (Len : Small_Integer := 0) is record
14382 Vals : IntArray (1 .. Len);
14386 you can assign a value with a different size of @code{Vals} with two
14390 (@value{GDBP}) set A_Rec.Len := 4
14391 (@value{GDBP}) set A_Rec := (Id => 42, Vals => (1, 2, 3, 4))
14394 As this example also illustrates, @value{GDBN} is very loose about the usual
14395 rules concerning aggregates. You may leave out some of the
14396 components of an array or record aggregate (such as the @code{Len}
14397 component in the assignment to @code{A_Rec} above); they will retain their
14398 original values upon assignment. You may freely use dynamic values as
14399 indices in component associations. You may even use overlapping or
14400 redundant component associations, although which component values are
14401 assigned in such cases is not defined.
14404 Calls to dispatching subprograms are not implemented.
14407 The overloading algorithm is much more limited (i.e., less selective)
14408 than that of real Ada. It makes only limited use of the context in
14409 which a subexpression appears to resolve its meaning, and it is much
14410 looser in its rules for allowing type matches. As a result, some
14411 function calls will be ambiguous, and the user will be asked to choose
14412 the proper resolution.
14415 The @code{new} operator is not implemented.
14418 Entry calls are not implemented.
14421 Aside from printing, arithmetic operations on the native VAX floating-point
14422 formats are not supported.
14425 It is not possible to slice a packed array.
14428 The names @code{True} and @code{False}, when not part of a qualified name,
14429 are interpreted as if implicitly prefixed by @code{Standard}, regardless of
14431 Should your program
14432 redefine these names in a package or procedure (at best a dubious practice),
14433 you will have to use fully qualified names to access their new definitions.
14436 @node Additions to Ada
14437 @subsubsection Additions to Ada
14438 @cindex Ada, deviations from
14440 As it does for other languages, @value{GDBN} makes certain generic
14441 extensions to Ada (@pxref{Expressions}):
14445 If the expression @var{E} is a variable residing in memory (typically
14446 a local variable or array element) and @var{N} is a positive integer,
14447 then @code{@var{E}@@@var{N}} displays the values of @var{E} and the
14448 @var{N}-1 adjacent variables following it in memory as an array. In
14449 Ada, this operator is generally not necessary, since its prime use is
14450 in displaying parts of an array, and slicing will usually do this in
14451 Ada. However, there are occasional uses when debugging programs in
14452 which certain debugging information has been optimized away.
14455 @code{@var{B}::@var{var}} means ``the variable named @var{var} that
14456 appears in function or file @var{B}.'' When @var{B} is a file name,
14457 you must typically surround it in single quotes.
14460 The expression @code{@{@var{type}@} @var{addr}} means ``the variable of type
14461 @var{type} that appears at address @var{addr}.''
14464 A name starting with @samp{$} is a convenience variable
14465 (@pxref{Convenience Vars}) or a machine register (@pxref{Registers}).
14468 In addition, @value{GDBN} provides a few other shortcuts and outright
14469 additions specific to Ada:
14473 The assignment statement is allowed as an expression, returning
14474 its right-hand operand as its value. Thus, you may enter
14477 (@value{GDBP}) set x := y + 3
14478 (@value{GDBP}) print A(tmp := y + 1)
14482 The semicolon is allowed as an ``operator,'' returning as its value
14483 the value of its right-hand operand.
14484 This allows, for example,
14485 complex conditional breaks:
14488 (@value{GDBP}) break f
14489 (@value{GDBP}) condition 1 (report(i); k += 1; A(k) > 100)
14493 Rather than use catenation and symbolic character names to introduce special
14494 characters into strings, one may instead use a special bracket notation,
14495 which is also used to print strings. A sequence of characters of the form
14496 @samp{["@var{XX}"]} within a string or character literal denotes the
14497 (single) character whose numeric encoding is @var{XX} in hexadecimal. The
14498 sequence of characters @samp{["""]} also denotes a single quotation mark
14499 in strings. For example,
14501 "One line.["0a"]Next line.["0a"]"
14504 contains an ASCII newline character (@code{Ada.Characters.Latin_1.LF})
14508 The subtype used as a prefix for the attributes @t{'Pos}, @t{'Min}, and
14509 @t{'Max} is optional (and is ignored in any case). For example, it is valid
14513 (@value{GDBP}) print 'max(x, y)
14517 When printing arrays, @value{GDBN} uses positional notation when the
14518 array has a lower bound of 1, and uses a modified named notation otherwise.
14519 For example, a one-dimensional array of three integers with a lower bound
14520 of 3 might print as
14527 That is, in contrast to valid Ada, only the first component has a @code{=>}
14531 You may abbreviate attributes in expressions with any unique,
14532 multi-character subsequence of
14533 their names (an exact match gets preference).
14534 For example, you may use @t{a'len}, @t{a'gth}, or @t{a'lh}
14535 in place of @t{a'length}.
14538 @cindex quoting Ada internal identifiers
14539 Since Ada is case-insensitive, the debugger normally maps identifiers you type
14540 to lower case. The GNAT compiler uses upper-case characters for
14541 some of its internal identifiers, which are normally of no interest to users.
14542 For the rare occasions when you actually have to look at them,
14543 enclose them in angle brackets to avoid the lower-case mapping.
14546 (@value{GDBP}) print <JMPBUF_SAVE>[0]
14550 Printing an object of class-wide type or dereferencing an
14551 access-to-class-wide value will display all the components of the object's
14552 specific type (as indicated by its run-time tag). Likewise, component
14553 selection on such a value will operate on the specific type of the
14558 @node Stopping Before Main Program
14559 @subsubsection Stopping at the Very Beginning
14561 @cindex breakpointing Ada elaboration code
14562 It is sometimes necessary to debug the program during elaboration, and
14563 before reaching the main procedure.
14564 As defined in the Ada Reference
14565 Manual, the elaboration code is invoked from a procedure called
14566 @code{adainit}. To run your program up to the beginning of
14567 elaboration, simply use the following two commands:
14568 @code{tbreak adainit} and @code{run}.
14571 @subsubsection Extensions for Ada Tasks
14572 @cindex Ada, tasking
14574 Support for Ada tasks is analogous to that for threads (@pxref{Threads}).
14575 @value{GDBN} provides the following task-related commands:
14580 This command shows a list of current Ada tasks, as in the following example:
14587 (@value{GDBP}) info tasks
14588 ID TID P-ID Pri State Name
14589 1 8088000 0 15 Child Activation Wait main_task
14590 2 80a4000 1 15 Accept Statement b
14591 3 809a800 1 15 Child Activation Wait a
14592 * 4 80ae800 3 15 Runnable c
14597 In this listing, the asterisk before the last task indicates it to be the
14598 task currently being inspected.
14602 Represents @value{GDBN}'s internal task number.
14608 The parent's task ID (@value{GDBN}'s internal task number).
14611 The base priority of the task.
14614 Current state of the task.
14618 The task has been created but has not been activated. It cannot be
14622 The task is not blocked for any reason known to Ada. (It may be waiting
14623 for a mutex, though.) It is conceptually "executing" in normal mode.
14626 The task is terminated, in the sense of ARM 9.3 (5). Any dependents
14627 that were waiting on terminate alternatives have been awakened and have
14628 terminated themselves.
14630 @item Child Activation Wait
14631 The task is waiting for created tasks to complete activation.
14633 @item Accept Statement
14634 The task is waiting on an accept or selective wait statement.
14636 @item Waiting on entry call
14637 The task is waiting on an entry call.
14639 @item Async Select Wait
14640 The task is waiting to start the abortable part of an asynchronous
14644 The task is waiting on a select statement with only a delay
14647 @item Child Termination Wait
14648 The task is sleeping having completed a master within itself, and is
14649 waiting for the tasks dependent on that master to become terminated or
14650 waiting on a terminate Phase.
14652 @item Wait Child in Term Alt
14653 The task is sleeping waiting for tasks on terminate alternatives to
14654 finish terminating.
14656 @item Accepting RV with @var{taskno}
14657 The task is accepting a rendez-vous with the task @var{taskno}.
14661 Name of the task in the program.
14665 @kindex info task @var{taskno}
14666 @item info task @var{taskno}
14667 This command shows detailled informations on the specified task, as in
14668 the following example:
14673 (@value{GDBP}) info tasks
14674 ID TID P-ID Pri State Name
14675 1 8077880 0 15 Child Activation Wait main_task
14676 * 2 807c468 1 15 Runnable task_1
14677 (@value{GDBP}) info task 2
14678 Ada Task: 0x807c468
14681 Parent: 1 (main_task)
14687 @kindex task@r{ (Ada)}
14688 @cindex current Ada task ID
14689 This command prints the ID of the current task.
14695 (@value{GDBP}) info tasks
14696 ID TID P-ID Pri State Name
14697 1 8077870 0 15 Child Activation Wait main_task
14698 * 2 807c458 1 15 Runnable t
14699 (@value{GDBP}) task
14700 [Current task is 2]
14703 @item task @var{taskno}
14704 @cindex Ada task switching
14705 This command is like the @code{thread @var{threadno}}
14706 command (@pxref{Threads}). It switches the context of debugging
14707 from the current task to the given task.
14713 (@value{GDBP}) info tasks
14714 ID TID P-ID Pri State Name
14715 1 8077870 0 15 Child Activation Wait main_task
14716 * 2 807c458 1 15 Runnable t
14717 (@value{GDBP}) task 1
14718 [Switching to task 1]
14719 #0 0x8067726 in pthread_cond_wait ()
14721 #0 0x8067726 in pthread_cond_wait ()
14722 #1 0x8056714 in system.os_interface.pthread_cond_wait ()
14723 #2 0x805cb63 in system.task_primitives.operations.sleep ()
14724 #3 0x806153e in system.tasking.stages.activate_tasks ()
14725 #4 0x804aacc in un () at un.adb:5
14728 @item break @var{linespec} task @var{taskno}
14729 @itemx break @var{linespec} task @var{taskno} if @dots{}
14730 @cindex breakpoints and tasks, in Ada
14731 @cindex task breakpoints, in Ada
14732 @kindex break @dots{} task @var{taskno}@r{ (Ada)}
14733 These commands are like the @code{break @dots{} thread @dots{}}
14734 command (@pxref{Thread Stops}).
14735 @var{linespec} specifies source lines, as described
14736 in @ref{Specify Location}.
14738 Use the qualifier @samp{task @var{taskno}} with a breakpoint command
14739 to specify that you only want @value{GDBN} to stop the program when a
14740 particular Ada task reaches this breakpoint. @var{taskno} is one of the
14741 numeric task identifiers assigned by @value{GDBN}, shown in the first
14742 column of the @samp{info tasks} display.
14744 If you do not specify @samp{task @var{taskno}} when you set a
14745 breakpoint, the breakpoint applies to @emph{all} tasks of your
14748 You can use the @code{task} qualifier on conditional breakpoints as
14749 well; in this case, place @samp{task @var{taskno}} before the
14750 breakpoint condition (before the @code{if}).
14758 (@value{GDBP}) info tasks
14759 ID TID P-ID Pri State Name
14760 1 140022020 0 15 Child Activation Wait main_task
14761 2 140045060 1 15 Accept/Select Wait t2
14762 3 140044840 1 15 Runnable t1
14763 * 4 140056040 1 15 Runnable t3
14764 (@value{GDBP}) b 15 task 2
14765 Breakpoint 5 at 0x120044cb0: file test_task_debug.adb, line 15.
14766 (@value{GDBP}) cont
14771 Breakpoint 5, test_task_debug () at test_task_debug.adb:15
14773 (@value{GDBP}) info tasks
14774 ID TID P-ID Pri State Name
14775 1 140022020 0 15 Child Activation Wait main_task
14776 * 2 140045060 1 15 Runnable t2
14777 3 140044840 1 15 Runnable t1
14778 4 140056040 1 15 Delay Sleep t3
14782 @node Ada Tasks and Core Files
14783 @subsubsection Tasking Support when Debugging Core Files
14784 @cindex Ada tasking and core file debugging
14786 When inspecting a core file, as opposed to debugging a live program,
14787 tasking support may be limited or even unavailable, depending on
14788 the platform being used.
14789 For instance, on x86-linux, the list of tasks is available, but task
14790 switching is not supported. On Tru64, however, task switching will work
14793 On certain platforms, including Tru64, the debugger needs to perform some
14794 memory writes in order to provide Ada tasking support. When inspecting
14795 a core file, this means that the core file must be opened with read-write
14796 privileges, using the command @samp{"set write on"} (@pxref{Patching}).
14797 Under these circumstances, you should make a backup copy of the core
14798 file before inspecting it with @value{GDBN}.
14800 @node Ravenscar Profile
14801 @subsubsection Tasking Support when using the Ravenscar Profile
14802 @cindex Ravenscar Profile
14804 The @dfn{Ravenscar Profile} is a subset of the Ada tasking features,
14805 specifically designed for systems with safety-critical real-time
14809 @kindex set ravenscar task-switching on
14810 @cindex task switching with program using Ravenscar Profile
14811 @item set ravenscar task-switching on
14812 Allows task switching when debugging a program that uses the Ravenscar
14813 Profile. This is the default.
14815 @kindex set ravenscar task-switching off
14816 @item set ravenscar task-switching off
14817 Turn off task switching when debugging a program that uses the Ravenscar
14818 Profile. This is mostly intended to disable the code that adds support
14819 for the Ravenscar Profile, in case a bug in either @value{GDBN} or in
14820 the Ravenscar runtime is preventing @value{GDBN} from working properly.
14821 To be effective, this command should be run before the program is started.
14823 @kindex show ravenscar task-switching
14824 @item show ravenscar task-switching
14825 Show whether it is possible to switch from task to task in a program
14826 using the Ravenscar Profile.
14831 @subsubsection Known Peculiarities of Ada Mode
14832 @cindex Ada, problems
14834 Besides the omissions listed previously (@pxref{Omissions from Ada}),
14835 we know of several problems with and limitations of Ada mode in
14837 some of which will be fixed with planned future releases of the debugger
14838 and the GNU Ada compiler.
14842 Static constants that the compiler chooses not to materialize as objects in
14843 storage are invisible to the debugger.
14846 Named parameter associations in function argument lists are ignored (the
14847 argument lists are treated as positional).
14850 Many useful library packages are currently invisible to the debugger.
14853 Fixed-point arithmetic, conversions, input, and output is carried out using
14854 floating-point arithmetic, and may give results that only approximate those on
14858 The GNAT compiler never generates the prefix @code{Standard} for any of
14859 the standard symbols defined by the Ada language. @value{GDBN} knows about
14860 this: it will strip the prefix from names when you use it, and will never
14861 look for a name you have so qualified among local symbols, nor match against
14862 symbols in other packages or subprograms. If you have
14863 defined entities anywhere in your program other than parameters and
14864 local variables whose simple names match names in @code{Standard},
14865 GNAT's lack of qualification here can cause confusion. When this happens,
14866 you can usually resolve the confusion
14867 by qualifying the problematic names with package
14868 @code{Standard} explicitly.
14871 Older versions of the compiler sometimes generate erroneous debugging
14872 information, resulting in the debugger incorrectly printing the value
14873 of affected entities. In some cases, the debugger is able to work
14874 around an issue automatically. In other cases, the debugger is able
14875 to work around the issue, but the work-around has to be specifically
14878 @kindex set ada trust-PAD-over-XVS
14879 @kindex show ada trust-PAD-over-XVS
14882 @item set ada trust-PAD-over-XVS on
14883 Configure GDB to strictly follow the GNAT encoding when computing the
14884 value of Ada entities, particularly when @code{PAD} and @code{PAD___XVS}
14885 types are involved (see @code{ada/exp_dbug.ads} in the GCC sources for
14886 a complete description of the encoding used by the GNAT compiler).
14887 This is the default.
14889 @item set ada trust-PAD-over-XVS off
14890 This is related to the encoding using by the GNAT compiler. If @value{GDBN}
14891 sometimes prints the wrong value for certain entities, changing @code{ada
14892 trust-PAD-over-XVS} to @code{off} activates a work-around which may fix
14893 the issue. It is always safe to set @code{ada trust-PAD-over-XVS} to
14894 @code{off}, but this incurs a slight performance penalty, so it is
14895 recommended to leave this setting to @code{on} unless necessary.
14899 @node Unsupported Languages
14900 @section Unsupported Languages
14902 @cindex unsupported languages
14903 @cindex minimal language
14904 In addition to the other fully-supported programming languages,
14905 @value{GDBN} also provides a pseudo-language, called @code{minimal}.
14906 It does not represent a real programming language, but provides a set
14907 of capabilities close to what the C or assembly languages provide.
14908 This should allow most simple operations to be performed while debugging
14909 an application that uses a language currently not supported by @value{GDBN}.
14911 If the language is set to @code{auto}, @value{GDBN} will automatically
14912 select this language if the current frame corresponds to an unsupported
14916 @chapter Examining the Symbol Table
14918 The commands described in this chapter allow you to inquire about the
14919 symbols (names of variables, functions and types) defined in your
14920 program. This information is inherent in the text of your program and
14921 does not change as your program executes. @value{GDBN} finds it in your
14922 program's symbol table, in the file indicated when you started @value{GDBN}
14923 (@pxref{File Options, ,Choosing Files}), or by one of the
14924 file-management commands (@pxref{Files, ,Commands to Specify Files}).
14926 @cindex symbol names
14927 @cindex names of symbols
14928 @cindex quoting names
14929 Occasionally, you may need to refer to symbols that contain unusual
14930 characters, which @value{GDBN} ordinarily treats as word delimiters. The
14931 most frequent case is in referring to static variables in other
14932 source files (@pxref{Variables,,Program Variables}). File names
14933 are recorded in object files as debugging symbols, but @value{GDBN} would
14934 ordinarily parse a typical file name, like @file{foo.c}, as the three words
14935 @samp{foo} @samp{.} @samp{c}. To allow @value{GDBN} to recognize
14936 @samp{foo.c} as a single symbol, enclose it in single quotes; for example,
14943 looks up the value of @code{x} in the scope of the file @file{foo.c}.
14946 @cindex case-insensitive symbol names
14947 @cindex case sensitivity in symbol names
14948 @kindex set case-sensitive
14949 @item set case-sensitive on
14950 @itemx set case-sensitive off
14951 @itemx set case-sensitive auto
14952 Normally, when @value{GDBN} looks up symbols, it matches their names
14953 with case sensitivity determined by the current source language.
14954 Occasionally, you may wish to control that. The command @code{set
14955 case-sensitive} lets you do that by specifying @code{on} for
14956 case-sensitive matches or @code{off} for case-insensitive ones. If
14957 you specify @code{auto}, case sensitivity is reset to the default
14958 suitable for the source language. The default is case-sensitive
14959 matches for all languages except for Fortran, for which the default is
14960 case-insensitive matches.
14962 @kindex show case-sensitive
14963 @item show case-sensitive
14964 This command shows the current setting of case sensitivity for symbols
14967 @kindex info address
14968 @cindex address of a symbol
14969 @item info address @var{symbol}
14970 Describe where the data for @var{symbol} is stored. For a register
14971 variable, this says which register it is kept in. For a non-register
14972 local variable, this prints the stack-frame offset at which the variable
14975 Note the contrast with @samp{print &@var{symbol}}, which does not work
14976 at all for a register variable, and for a stack local variable prints
14977 the exact address of the current instantiation of the variable.
14979 @kindex info symbol
14980 @cindex symbol from address
14981 @cindex closest symbol and offset for an address
14982 @item info symbol @var{addr}
14983 Print the name of a symbol which is stored at the address @var{addr}.
14984 If no symbol is stored exactly at @var{addr}, @value{GDBN} prints the
14985 nearest symbol and an offset from it:
14988 (@value{GDBP}) info symbol 0x54320
14989 _initialize_vx + 396 in section .text
14993 This is the opposite of the @code{info address} command. You can use
14994 it to find out the name of a variable or a function given its address.
14996 For dynamically linked executables, the name of executable or shared
14997 library containing the symbol is also printed:
15000 (@value{GDBP}) info symbol 0x400225
15001 _start + 5 in section .text of /tmp/a.out
15002 (@value{GDBP}) info symbol 0x2aaaac2811cf
15003 __read_nocancel + 6 in section .text of /usr/lib64/libc.so.6
15007 @item whatis [@var{arg}]
15008 Print the data type of @var{arg}, which can be either an expression
15009 or a name of a data type. With no argument, print the data type of
15010 @code{$}, the last value in the value history.
15012 If @var{arg} is an expression (@pxref{Expressions, ,Expressions}), it
15013 is not actually evaluated, and any side-effecting operations (such as
15014 assignments or function calls) inside it do not take place.
15016 If @var{arg} is a variable or an expression, @code{whatis} prints its
15017 literal type as it is used in the source code. If the type was
15018 defined using a @code{typedef}, @code{whatis} will @emph{not} print
15019 the data type underlying the @code{typedef}. If the type of the
15020 variable or the expression is a compound data type, such as
15021 @code{struct} or @code{class}, @code{whatis} never prints their
15022 fields or methods. It just prints the @code{struct}/@code{class}
15023 name (a.k.a.@: its @dfn{tag}). If you want to see the members of
15024 such a compound data type, use @code{ptype}.
15026 If @var{arg} is a type name that was defined using @code{typedef},
15027 @code{whatis} @dfn{unrolls} only one level of that @code{typedef}.
15028 Unrolling means that @code{whatis} will show the underlying type used
15029 in the @code{typedef} declaration of @var{arg}. However, if that
15030 underlying type is also a @code{typedef}, @code{whatis} will not
15033 For C code, the type names may also have the form @samp{class
15034 @var{class-name}}, @samp{struct @var{struct-tag}}, @samp{union
15035 @var{union-tag}} or @samp{enum @var{enum-tag}}.
15038 @item ptype [@var{arg}]
15039 @code{ptype} accepts the same arguments as @code{whatis}, but prints a
15040 detailed description of the type, instead of just the name of the type.
15041 @xref{Expressions, ,Expressions}.
15043 Contrary to @code{whatis}, @code{ptype} always unrolls any
15044 @code{typedef}s in its argument declaration, whether the argument is
15045 a variable, expression, or a data type. This means that @code{ptype}
15046 of a variable or an expression will not print literally its type as
15047 present in the source code---use @code{whatis} for that. @code{typedef}s at
15048 the pointer or reference targets are also unrolled. Only @code{typedef}s of
15049 fields, methods and inner @code{class typedef}s of @code{struct}s,
15050 @code{class}es and @code{union}s are not unrolled even with @code{ptype}.
15052 For example, for this variable declaration:
15055 typedef double real_t;
15056 struct complex @{ real_t real; double imag; @};
15057 typedef struct complex complex_t;
15059 real_t *real_pointer_var;
15063 the two commands give this output:
15067 (@value{GDBP}) whatis var
15069 (@value{GDBP}) ptype var
15070 type = struct complex @{
15074 (@value{GDBP}) whatis complex_t
15075 type = struct complex
15076 (@value{GDBP}) whatis struct complex
15077 type = struct complex
15078 (@value{GDBP}) ptype struct complex
15079 type = struct complex @{
15083 (@value{GDBP}) whatis real_pointer_var
15085 (@value{GDBP}) ptype real_pointer_var
15091 As with @code{whatis}, using @code{ptype} without an argument refers to
15092 the type of @code{$}, the last value in the value history.
15094 @cindex incomplete type
15095 Sometimes, programs use opaque data types or incomplete specifications
15096 of complex data structure. If the debug information included in the
15097 program does not allow @value{GDBN} to display a full declaration of
15098 the data type, it will say @samp{<incomplete type>}. For example,
15099 given these declarations:
15103 struct foo *fooptr;
15107 but no definition for @code{struct foo} itself, @value{GDBN} will say:
15110 (@value{GDBP}) ptype foo
15111 $1 = <incomplete type>
15115 ``Incomplete type'' is C terminology for data types that are not
15116 completely specified.
15119 @item info types @var{regexp}
15121 Print a brief description of all types whose names match the regular
15122 expression @var{regexp} (or all types in your program, if you supply
15123 no argument). Each complete typename is matched as though it were a
15124 complete line; thus, @samp{i type value} gives information on all
15125 types in your program whose names include the string @code{value}, but
15126 @samp{i type ^value$} gives information only on types whose complete
15127 name is @code{value}.
15129 This command differs from @code{ptype} in two ways: first, like
15130 @code{whatis}, it does not print a detailed description; second, it
15131 lists all source files where a type is defined.
15134 @cindex local variables
15135 @item info scope @var{location}
15136 List all the variables local to a particular scope. This command
15137 accepts a @var{location} argument---a function name, a source line, or
15138 an address preceded by a @samp{*}, and prints all the variables local
15139 to the scope defined by that location. (@xref{Specify Location}, for
15140 details about supported forms of @var{location}.) For example:
15143 (@value{GDBP}) @b{info scope command_line_handler}
15144 Scope for command_line_handler:
15145 Symbol rl is an argument at stack/frame offset 8, length 4.
15146 Symbol linebuffer is in static storage at address 0x150a18, length 4.
15147 Symbol linelength is in static storage at address 0x150a1c, length 4.
15148 Symbol p is a local variable in register $esi, length 4.
15149 Symbol p1 is a local variable in register $ebx, length 4.
15150 Symbol nline is a local variable in register $edx, length 4.
15151 Symbol repeat is a local variable at frame offset -8, length 4.
15155 This command is especially useful for determining what data to collect
15156 during a @dfn{trace experiment}, see @ref{Tracepoint Actions,
15159 @kindex info source
15161 Show information about the current source file---that is, the source file for
15162 the function containing the current point of execution:
15165 the name of the source file, and the directory containing it,
15167 the directory it was compiled in,
15169 its length, in lines,
15171 which programming language it is written in,
15173 whether the executable includes debugging information for that file, and
15174 if so, what format the information is in (e.g., STABS, Dwarf 2, etc.), and
15176 whether the debugging information includes information about
15177 preprocessor macros.
15181 @kindex info sources
15183 Print the names of all source files in your program for which there is
15184 debugging information, organized into two lists: files whose symbols
15185 have already been read, and files whose symbols will be read when needed.
15187 @kindex info functions
15188 @item info functions
15189 Print the names and data types of all defined functions.
15191 @item info functions @var{regexp}
15192 Print the names and data types of all defined functions
15193 whose names contain a match for regular expression @var{regexp}.
15194 Thus, @samp{info fun step} finds all functions whose names
15195 include @code{step}; @samp{info fun ^step} finds those whose names
15196 start with @code{step}. If a function name contains characters
15197 that conflict with the regular expression language (e.g.@:
15198 @samp{operator*()}), they may be quoted with a backslash.
15200 @kindex info variables
15201 @item info variables
15202 Print the names and data types of all variables that are defined
15203 outside of functions (i.e.@: excluding local variables).
15205 @item info variables @var{regexp}
15206 Print the names and data types of all variables (except for local
15207 variables) whose names contain a match for regular expression
15210 @kindex info classes
15211 @cindex Objective-C, classes and selectors
15213 @itemx info classes @var{regexp}
15214 Display all Objective-C classes in your program, or
15215 (with the @var{regexp} argument) all those matching a particular regular
15218 @kindex info selectors
15219 @item info selectors
15220 @itemx info selectors @var{regexp}
15221 Display all Objective-C selectors in your program, or
15222 (with the @var{regexp} argument) all those matching a particular regular
15226 This was never implemented.
15227 @kindex info methods
15229 @itemx info methods @var{regexp}
15230 The @code{info methods} command permits the user to examine all defined
15231 methods within C@t{++} program, or (with the @var{regexp} argument) a
15232 specific set of methods found in the various C@t{++} classes. Many
15233 C@t{++} classes provide a large number of methods. Thus, the output
15234 from the @code{ptype} command can be overwhelming and hard to use. The
15235 @code{info-methods} command filters the methods, printing only those
15236 which match the regular-expression @var{regexp}.
15239 @cindex opaque data types
15240 @kindex set opaque-type-resolution
15241 @item set opaque-type-resolution on
15242 Tell @value{GDBN} to resolve opaque types. An opaque type is a type
15243 declared as a pointer to a @code{struct}, @code{class}, or
15244 @code{union}---for example, @code{struct MyType *}---that is used in one
15245 source file although the full declaration of @code{struct MyType} is in
15246 another source file. The default is on.
15248 A change in the setting of this subcommand will not take effect until
15249 the next time symbols for a file are loaded.
15251 @item set opaque-type-resolution off
15252 Tell @value{GDBN} not to resolve opaque types. In this case, the type
15253 is printed as follows:
15255 @{<no data fields>@}
15258 @kindex show opaque-type-resolution
15259 @item show opaque-type-resolution
15260 Show whether opaque types are resolved or not.
15262 @kindex maint print symbols
15263 @cindex symbol dump
15264 @kindex maint print psymbols
15265 @cindex partial symbol dump
15266 @item maint print symbols @var{filename}
15267 @itemx maint print psymbols @var{filename}
15268 @itemx maint print msymbols @var{filename}
15269 Write a dump of debugging symbol data into the file @var{filename}.
15270 These commands are used to debug the @value{GDBN} symbol-reading code. Only
15271 symbols with debugging data are included. If you use @samp{maint print
15272 symbols}, @value{GDBN} includes all the symbols for which it has already
15273 collected full details: that is, @var{filename} reflects symbols for
15274 only those files whose symbols @value{GDBN} has read. You can use the
15275 command @code{info sources} to find out which files these are. If you
15276 use @samp{maint print psymbols} instead, the dump shows information about
15277 symbols that @value{GDBN} only knows partially---that is, symbols defined in
15278 files that @value{GDBN} has skimmed, but not yet read completely. Finally,
15279 @samp{maint print msymbols} dumps just the minimal symbol information
15280 required for each object file from which @value{GDBN} has read some symbols.
15281 @xref{Files, ,Commands to Specify Files}, for a discussion of how
15282 @value{GDBN} reads symbols (in the description of @code{symbol-file}).
15284 @kindex maint info symtabs
15285 @kindex maint info psymtabs
15286 @cindex listing @value{GDBN}'s internal symbol tables
15287 @cindex symbol tables, listing @value{GDBN}'s internal
15288 @cindex full symbol tables, listing @value{GDBN}'s internal
15289 @cindex partial symbol tables, listing @value{GDBN}'s internal
15290 @item maint info symtabs @r{[} @var{regexp} @r{]}
15291 @itemx maint info psymtabs @r{[} @var{regexp} @r{]}
15293 List the @code{struct symtab} or @code{struct partial_symtab}
15294 structures whose names match @var{regexp}. If @var{regexp} is not
15295 given, list them all. The output includes expressions which you can
15296 copy into a @value{GDBN} debugging this one to examine a particular
15297 structure in more detail. For example:
15300 (@value{GDBP}) maint info psymtabs dwarf2read
15301 @{ objfile /home/gnu/build/gdb/gdb
15302 ((struct objfile *) 0x82e69d0)
15303 @{ psymtab /home/gnu/src/gdb/dwarf2read.c
15304 ((struct partial_symtab *) 0x8474b10)
15307 text addresses 0x814d3c8 -- 0x8158074
15308 globals (* (struct partial_symbol **) 0x8507a08 @@ 9)
15309 statics (* (struct partial_symbol **) 0x40e95b78 @@ 2882)
15310 dependencies (none)
15313 (@value{GDBP}) maint info symtabs
15317 We see that there is one partial symbol table whose filename contains
15318 the string @samp{dwarf2read}, belonging to the @samp{gdb} executable;
15319 and we see that @value{GDBN} has not read in any symtabs yet at all.
15320 If we set a breakpoint on a function, that will cause @value{GDBN} to
15321 read the symtab for the compilation unit containing that function:
15324 (@value{GDBP}) break dwarf2_psymtab_to_symtab
15325 Breakpoint 1 at 0x814e5da: file /home/gnu/src/gdb/dwarf2read.c,
15327 (@value{GDBP}) maint info symtabs
15328 @{ objfile /home/gnu/build/gdb/gdb
15329 ((struct objfile *) 0x82e69d0)
15330 @{ symtab /home/gnu/src/gdb/dwarf2read.c
15331 ((struct symtab *) 0x86c1f38)
15334 blockvector ((struct blockvector *) 0x86c1bd0) (primary)
15335 linetable ((struct linetable *) 0x8370fa0)
15336 debugformat DWARF 2
15345 @chapter Altering Execution
15347 Once you think you have found an error in your program, you might want to
15348 find out for certain whether correcting the apparent error would lead to
15349 correct results in the rest of the run. You can find the answer by
15350 experiment, using the @value{GDBN} features for altering execution of the
15353 For example, you can store new values into variables or memory
15354 locations, give your program a signal, restart it at a different
15355 address, or even return prematurely from a function.
15358 * Assignment:: Assignment to variables
15359 * Jumping:: Continuing at a different address
15360 * Signaling:: Giving your program a signal
15361 * Returning:: Returning from a function
15362 * Calling:: Calling your program's functions
15363 * Patching:: Patching your program
15367 @section Assignment to Variables
15370 @cindex setting variables
15371 To alter the value of a variable, evaluate an assignment expression.
15372 @xref{Expressions, ,Expressions}. For example,
15379 stores the value 4 into the variable @code{x}, and then prints the
15380 value of the assignment expression (which is 4).
15381 @xref{Languages, ,Using @value{GDBN} with Different Languages}, for more
15382 information on operators in supported languages.
15384 @kindex set variable
15385 @cindex variables, setting
15386 If you are not interested in seeing the value of the assignment, use the
15387 @code{set} command instead of the @code{print} command. @code{set} is
15388 really the same as @code{print} except that the expression's value is
15389 not printed and is not put in the value history (@pxref{Value History,
15390 ,Value History}). The expression is evaluated only for its effects.
15392 If the beginning of the argument string of the @code{set} command
15393 appears identical to a @code{set} subcommand, use the @code{set
15394 variable} command instead of just @code{set}. This command is identical
15395 to @code{set} except for its lack of subcommands. For example, if your
15396 program has a variable @code{width}, you get an error if you try to set
15397 a new value with just @samp{set width=13}, because @value{GDBN} has the
15398 command @code{set width}:
15401 (@value{GDBP}) whatis width
15403 (@value{GDBP}) p width
15405 (@value{GDBP}) set width=47
15406 Invalid syntax in expression.
15410 The invalid expression, of course, is @samp{=47}. In
15411 order to actually set the program's variable @code{width}, use
15414 (@value{GDBP}) set var width=47
15417 Because the @code{set} command has many subcommands that can conflict
15418 with the names of program variables, it is a good idea to use the
15419 @code{set variable} command instead of just @code{set}. For example, if
15420 your program has a variable @code{g}, you run into problems if you try
15421 to set a new value with just @samp{set g=4}, because @value{GDBN} has
15422 the command @code{set gnutarget}, abbreviated @code{set g}:
15426 (@value{GDBP}) whatis g
15430 (@value{GDBP}) set g=4
15434 The program being debugged has been started already.
15435 Start it from the beginning? (y or n) y
15436 Starting program: /home/smith/cc_progs/a.out
15437 "/home/smith/cc_progs/a.out": can't open to read symbols:
15438 Invalid bfd target.
15439 (@value{GDBP}) show g
15440 The current BFD target is "=4".
15445 The program variable @code{g} did not change, and you silently set the
15446 @code{gnutarget} to an invalid value. In order to set the variable
15450 (@value{GDBP}) set var g=4
15453 @value{GDBN} allows more implicit conversions in assignments than C; you can
15454 freely store an integer value into a pointer variable or vice versa,
15455 and you can convert any structure to any other structure that is the
15456 same length or shorter.
15457 @comment FIXME: how do structs align/pad in these conversions?
15458 @comment /doc@cygnus.com 18dec1990
15460 To store values into arbitrary places in memory, use the @samp{@{@dots{}@}}
15461 construct to generate a value of specified type at a specified address
15462 (@pxref{Expressions, ,Expressions}). For example, @code{@{int@}0x83040} refers
15463 to memory location @code{0x83040} as an integer (which implies a certain size
15464 and representation in memory), and
15467 set @{int@}0x83040 = 4
15471 stores the value 4 into that memory location.
15474 @section Continuing at a Different Address
15476 Ordinarily, when you continue your program, you do so at the place where
15477 it stopped, with the @code{continue} command. You can instead continue at
15478 an address of your own choosing, with the following commands:
15482 @item jump @var{linespec}
15483 @itemx jump @var{location}
15484 Resume execution at line @var{linespec} or at address given by
15485 @var{location}. Execution stops again immediately if there is a
15486 breakpoint there. @xref{Specify Location}, for a description of the
15487 different forms of @var{linespec} and @var{location}. It is common
15488 practice to use the @code{tbreak} command in conjunction with
15489 @code{jump}. @xref{Set Breaks, ,Setting Breakpoints}.
15491 The @code{jump} command does not change the current stack frame, or
15492 the stack pointer, or the contents of any memory location or any
15493 register other than the program counter. If line @var{linespec} is in
15494 a different function from the one currently executing, the results may
15495 be bizarre if the two functions expect different patterns of arguments or
15496 of local variables. For this reason, the @code{jump} command requests
15497 confirmation if the specified line is not in the function currently
15498 executing. However, even bizarre results are predictable if you are
15499 well acquainted with the machine-language code of your program.
15502 @c Doesn't work on HP-UX; have to set $pcoqh and $pcoqt.
15503 On many systems, you can get much the same effect as the @code{jump}
15504 command by storing a new value into the register @code{$pc}. The
15505 difference is that this does not start your program running; it only
15506 changes the address of where it @emph{will} run when you continue. For
15514 makes the next @code{continue} command or stepping command execute at
15515 address @code{0x485}, rather than at the address where your program stopped.
15516 @xref{Continuing and Stepping, ,Continuing and Stepping}.
15518 The most common occasion to use the @code{jump} command is to back
15519 up---perhaps with more breakpoints set---over a portion of a program
15520 that has already executed, in order to examine its execution in more
15525 @section Giving your Program a Signal
15526 @cindex deliver a signal to a program
15530 @item signal @var{signal}
15531 Resume execution where your program stopped, but immediately give it the
15532 signal @var{signal}. @var{signal} can be the name or the number of a
15533 signal. For example, on many systems @code{signal 2} and @code{signal
15534 SIGINT} are both ways of sending an interrupt signal.
15536 Alternatively, if @var{signal} is zero, continue execution without
15537 giving a signal. This is useful when your program stopped on account of
15538 a signal and would ordinary see the signal when resumed with the
15539 @code{continue} command; @samp{signal 0} causes it to resume without a
15542 @code{signal} does not repeat when you press @key{RET} a second time
15543 after executing the command.
15547 Invoking the @code{signal} command is not the same as invoking the
15548 @code{kill} utility from the shell. Sending a signal with @code{kill}
15549 causes @value{GDBN} to decide what to do with the signal depending on
15550 the signal handling tables (@pxref{Signals}). The @code{signal} command
15551 passes the signal directly to your program.
15555 @section Returning from a Function
15558 @cindex returning from a function
15561 @itemx return @var{expression}
15562 You can cancel execution of a function call with the @code{return}
15563 command. If you give an
15564 @var{expression} argument, its value is used as the function's return
15568 When you use @code{return}, @value{GDBN} discards the selected stack frame
15569 (and all frames within it). You can think of this as making the
15570 discarded frame return prematurely. If you wish to specify a value to
15571 be returned, give that value as the argument to @code{return}.
15573 This pops the selected stack frame (@pxref{Selection, ,Selecting a
15574 Frame}), and any other frames inside of it, leaving its caller as the
15575 innermost remaining frame. That frame becomes selected. The
15576 specified value is stored in the registers used for returning values
15579 The @code{return} command does not resume execution; it leaves the
15580 program stopped in the state that would exist if the function had just
15581 returned. In contrast, the @code{finish} command (@pxref{Continuing
15582 and Stepping, ,Continuing and Stepping}) resumes execution until the
15583 selected stack frame returns naturally.
15585 @value{GDBN} needs to know how the @var{expression} argument should be set for
15586 the inferior. The concrete registers assignment depends on the OS ABI and the
15587 type being returned by the selected stack frame. For example it is common for
15588 OS ABI to return floating point values in FPU registers while integer values in
15589 CPU registers. Still some ABIs return even floating point values in CPU
15590 registers. Larger integer widths (such as @code{long long int}) also have
15591 specific placement rules. @value{GDBN} already knows the OS ABI from its
15592 current target so it needs to find out also the type being returned to make the
15593 assignment into the right register(s).
15595 Normally, the selected stack frame has debug info. @value{GDBN} will always
15596 use the debug info instead of the implicit type of @var{expression} when the
15597 debug info is available. For example, if you type @kbd{return -1}, and the
15598 function in the current stack frame is declared to return a @code{long long
15599 int}, @value{GDBN} transparently converts the implicit @code{int} value of -1
15600 into a @code{long long int}:
15603 Breakpoint 1, func () at gdb.base/return-nodebug.c:29
15605 (@value{GDBP}) return -1
15606 Make func return now? (y or n) y
15607 #0 0x004004f6 in main () at gdb.base/return-nodebug.c:43
15608 43 printf ("result=%lld\n", func ());
15612 However, if the selected stack frame does not have a debug info, e.g., if the
15613 function was compiled without debug info, @value{GDBN} has to find out the type
15614 to return from user. Specifying a different type by mistake may set the value
15615 in different inferior registers than the caller code expects. For example,
15616 typing @kbd{return -1} with its implicit type @code{int} would set only a part
15617 of a @code{long long int} result for a debug info less function (on 32-bit
15618 architectures). Therefore the user is required to specify the return type by
15619 an appropriate cast explicitly:
15622 Breakpoint 2, 0x0040050b in func ()
15623 (@value{GDBP}) return -1
15624 Return value type not available for selected stack frame.
15625 Please use an explicit cast of the value to return.
15626 (@value{GDBP}) return (long long int) -1
15627 Make selected stack frame return now? (y or n) y
15628 #0 0x00400526 in main ()
15633 @section Calling Program Functions
15636 @cindex calling functions
15637 @cindex inferior functions, calling
15638 @item print @var{expr}
15639 Evaluate the expression @var{expr} and display the resulting value.
15640 @var{expr} may include calls to functions in the program being
15644 @item call @var{expr}
15645 Evaluate the expression @var{expr} without displaying @code{void}
15648 You can use this variant of the @code{print} command if you want to
15649 execute a function from your program that does not return anything
15650 (a.k.a.@: @dfn{a void function}), but without cluttering the output
15651 with @code{void} returned values that @value{GDBN} will otherwise
15652 print. If the result is not void, it is printed and saved in the
15656 It is possible for the function you call via the @code{print} or
15657 @code{call} command to generate a signal (e.g., if there's a bug in
15658 the function, or if you passed it incorrect arguments). What happens
15659 in that case is controlled by the @code{set unwindonsignal} command.
15661 Similarly, with a C@t{++} program it is possible for the function you
15662 call via the @code{print} or @code{call} command to generate an
15663 exception that is not handled due to the constraints of the dummy
15664 frame. In this case, any exception that is raised in the frame, but has
15665 an out-of-frame exception handler will not be found. GDB builds a
15666 dummy-frame for the inferior function call, and the unwinder cannot
15667 seek for exception handlers outside of this dummy-frame. What happens
15668 in that case is controlled by the
15669 @code{set unwind-on-terminating-exception} command.
15672 @item set unwindonsignal
15673 @kindex set unwindonsignal
15674 @cindex unwind stack in called functions
15675 @cindex call dummy stack unwinding
15676 Set unwinding of the stack if a signal is received while in a function
15677 that @value{GDBN} called in the program being debugged. If set to on,
15678 @value{GDBN} unwinds the stack it created for the call and restores
15679 the context to what it was before the call. If set to off (the
15680 default), @value{GDBN} stops in the frame where the signal was
15683 @item show unwindonsignal
15684 @kindex show unwindonsignal
15685 Show the current setting of stack unwinding in the functions called by
15688 @item set unwind-on-terminating-exception
15689 @kindex set unwind-on-terminating-exception
15690 @cindex unwind stack in called functions with unhandled exceptions
15691 @cindex call dummy stack unwinding on unhandled exception.
15692 Set unwinding of the stack if a C@t{++} exception is raised, but left
15693 unhandled while in a function that @value{GDBN} called in the program being
15694 debugged. If set to on (the default), @value{GDBN} unwinds the stack
15695 it created for the call and restores the context to what it was before
15696 the call. If set to off, @value{GDBN} the exception is delivered to
15697 the default C@t{++} exception handler and the inferior terminated.
15699 @item show unwind-on-terminating-exception
15700 @kindex show unwind-on-terminating-exception
15701 Show the current setting of stack unwinding in the functions called by
15706 @cindex weak alias functions
15707 Sometimes, a function you wish to call is actually a @dfn{weak alias}
15708 for another function. In such case, @value{GDBN} might not pick up
15709 the type information, including the types of the function arguments,
15710 which causes @value{GDBN} to call the inferior function incorrectly.
15711 As a result, the called function will function erroneously and may
15712 even crash. A solution to that is to use the name of the aliased
15716 @section Patching Programs
15718 @cindex patching binaries
15719 @cindex writing into executables
15720 @cindex writing into corefiles
15722 By default, @value{GDBN} opens the file containing your program's
15723 executable code (or the corefile) read-only. This prevents accidental
15724 alterations to machine code; but it also prevents you from intentionally
15725 patching your program's binary.
15727 If you'd like to be able to patch the binary, you can specify that
15728 explicitly with the @code{set write} command. For example, you might
15729 want to turn on internal debugging flags, or even to make emergency
15735 @itemx set write off
15736 If you specify @samp{set write on}, @value{GDBN} opens executable and
15737 core files for both reading and writing; if you specify @kbd{set write
15738 off} (the default), @value{GDBN} opens them read-only.
15740 If you have already loaded a file, you must load it again (using the
15741 @code{exec-file} or @code{core-file} command) after changing @code{set
15742 write}, for your new setting to take effect.
15746 Display whether executable files and core files are opened for writing
15747 as well as reading.
15751 @chapter @value{GDBN} Files
15753 @value{GDBN} needs to know the file name of the program to be debugged,
15754 both in order to read its symbol table and in order to start your
15755 program. To debug a core dump of a previous run, you must also tell
15756 @value{GDBN} the name of the core dump file.
15759 * Files:: Commands to specify files
15760 * Separate Debug Files:: Debugging information in separate files
15761 * Index Files:: Index files speed up GDB
15762 * Symbol Errors:: Errors reading symbol files
15763 * Data Files:: GDB data files
15767 @section Commands to Specify Files
15769 @cindex symbol table
15770 @cindex core dump file
15772 You may want to specify executable and core dump file names. The usual
15773 way to do this is at start-up time, using the arguments to
15774 @value{GDBN}'s start-up commands (@pxref{Invocation, , Getting In and
15775 Out of @value{GDBN}}).
15777 Occasionally it is necessary to change to a different file during a
15778 @value{GDBN} session. Or you may run @value{GDBN} and forget to
15779 specify a file you want to use. Or you are debugging a remote target
15780 via @code{gdbserver} (@pxref{Server, file, Using the @code{gdbserver}
15781 Program}). In these situations the @value{GDBN} commands to specify
15782 new files are useful.
15785 @cindex executable file
15787 @item file @var{filename}
15788 Use @var{filename} as the program to be debugged. It is read for its
15789 symbols and for the contents of pure memory. It is also the program
15790 executed when you use the @code{run} command. If you do not specify a
15791 directory and the file is not found in the @value{GDBN} working directory,
15792 @value{GDBN} uses the environment variable @code{PATH} as a list of
15793 directories to search, just as the shell does when looking for a program
15794 to run. You can change the value of this variable, for both @value{GDBN}
15795 and your program, using the @code{path} command.
15797 @cindex unlinked object files
15798 @cindex patching object files
15799 You can load unlinked object @file{.o} files into @value{GDBN} using
15800 the @code{file} command. You will not be able to ``run'' an object
15801 file, but you can disassemble functions and inspect variables. Also,
15802 if the underlying BFD functionality supports it, you could use
15803 @kbd{gdb -write} to patch object files using this technique. Note
15804 that @value{GDBN} can neither interpret nor modify relocations in this
15805 case, so branches and some initialized variables will appear to go to
15806 the wrong place. But this feature is still handy from time to time.
15809 @code{file} with no argument makes @value{GDBN} discard any information it
15810 has on both executable file and the symbol table.
15813 @item exec-file @r{[} @var{filename} @r{]}
15814 Specify that the program to be run (but not the symbol table) is found
15815 in @var{filename}. @value{GDBN} searches the environment variable @code{PATH}
15816 if necessary to locate your program. Omitting @var{filename} means to
15817 discard information on the executable file.
15819 @kindex symbol-file
15820 @item symbol-file @r{[} @var{filename} @r{]}
15821 Read symbol table information from file @var{filename}. @code{PATH} is
15822 searched when necessary. Use the @code{file} command to get both symbol
15823 table and program to run from the same file.
15825 @code{symbol-file} with no argument clears out @value{GDBN} information on your
15826 program's symbol table.
15828 The @code{symbol-file} command causes @value{GDBN} to forget the contents of
15829 some breakpoints and auto-display expressions. This is because they may
15830 contain pointers to the internal data recording symbols and data types,
15831 which are part of the old symbol table data being discarded inside
15834 @code{symbol-file} does not repeat if you press @key{RET} again after
15837 When @value{GDBN} is configured for a particular environment, it
15838 understands debugging information in whatever format is the standard
15839 generated for that environment; you may use either a @sc{gnu} compiler, or
15840 other compilers that adhere to the local conventions.
15841 Best results are usually obtained from @sc{gnu} compilers; for example,
15842 using @code{@value{NGCC}} you can generate debugging information for
15845 For most kinds of object files, with the exception of old SVR3 systems
15846 using COFF, the @code{symbol-file} command does not normally read the
15847 symbol table in full right away. Instead, it scans the symbol table
15848 quickly to find which source files and which symbols are present. The
15849 details are read later, one source file at a time, as they are needed.
15851 The purpose of this two-stage reading strategy is to make @value{GDBN}
15852 start up faster. For the most part, it is invisible except for
15853 occasional pauses while the symbol table details for a particular source
15854 file are being read. (The @code{set verbose} command can turn these
15855 pauses into messages if desired. @xref{Messages/Warnings, ,Optional
15856 Warnings and Messages}.)
15858 We have not implemented the two-stage strategy for COFF yet. When the
15859 symbol table is stored in COFF format, @code{symbol-file} reads the
15860 symbol table data in full right away. Note that ``stabs-in-COFF''
15861 still does the two-stage strategy, since the debug info is actually
15865 @cindex reading symbols immediately
15866 @cindex symbols, reading immediately
15867 @item symbol-file @r{[} -readnow @r{]} @var{filename}
15868 @itemx file @r{[} -readnow @r{]} @var{filename}
15869 You can override the @value{GDBN} two-stage strategy for reading symbol
15870 tables by using the @samp{-readnow} option with any of the commands that
15871 load symbol table information, if you want to be sure @value{GDBN} has the
15872 entire symbol table available.
15874 @c FIXME: for now no mention of directories, since this seems to be in
15875 @c flux. 13mar1992 status is that in theory GDB would look either in
15876 @c current dir or in same dir as myprog; but issues like competing
15877 @c GDB's, or clutter in system dirs, mean that in practice right now
15878 @c only current dir is used. FFish says maybe a special GDB hierarchy
15879 @c (eg rooted in val of env var GDBSYMS) could exist for mappable symbol
15883 @item core-file @r{[}@var{filename}@r{]}
15885 Specify the whereabouts of a core dump file to be used as the ``contents
15886 of memory''. Traditionally, core files contain only some parts of the
15887 address space of the process that generated them; @value{GDBN} can access the
15888 executable file itself for other parts.
15890 @code{core-file} with no argument specifies that no core file is
15893 Note that the core file is ignored when your program is actually running
15894 under @value{GDBN}. So, if you have been running your program and you
15895 wish to debug a core file instead, you must kill the subprocess in which
15896 the program is running. To do this, use the @code{kill} command
15897 (@pxref{Kill Process, ,Killing the Child Process}).
15899 @kindex add-symbol-file
15900 @cindex dynamic linking
15901 @item add-symbol-file @var{filename} @var{address}
15902 @itemx add-symbol-file @var{filename} @var{address} @r{[} -readnow @r{]}
15903 @itemx add-symbol-file @var{filename} @var{address} -s @var{section} @var{address} @dots{}
15904 The @code{add-symbol-file} command reads additional symbol table
15905 information from the file @var{filename}. You would use this command
15906 when @var{filename} has been dynamically loaded (by some other means)
15907 into the program that is running. @var{address} should be the memory
15908 address at which the file has been loaded; @value{GDBN} cannot figure
15909 this out for itself. You can additionally specify an arbitrary number
15910 of @samp{-s @var{section} @var{address}} pairs, to give an explicit
15911 section name and base address for that section. You can specify any
15912 @var{address} as an expression.
15914 The symbol table of the file @var{filename} is added to the symbol table
15915 originally read with the @code{symbol-file} command. You can use the
15916 @code{add-symbol-file} command any number of times; the new symbol data
15917 thus read keeps adding to the old. To discard all old symbol data
15918 instead, use the @code{symbol-file} command without any arguments.
15920 @cindex relocatable object files, reading symbols from
15921 @cindex object files, relocatable, reading symbols from
15922 @cindex reading symbols from relocatable object files
15923 @cindex symbols, reading from relocatable object files
15924 @cindex @file{.o} files, reading symbols from
15925 Although @var{filename} is typically a shared library file, an
15926 executable file, or some other object file which has been fully
15927 relocated for loading into a process, you can also load symbolic
15928 information from relocatable @file{.o} files, as long as:
15932 the file's symbolic information refers only to linker symbols defined in
15933 that file, not to symbols defined by other object files,
15935 every section the file's symbolic information refers to has actually
15936 been loaded into the inferior, as it appears in the file, and
15938 you can determine the address at which every section was loaded, and
15939 provide these to the @code{add-symbol-file} command.
15943 Some embedded operating systems, like Sun Chorus and VxWorks, can load
15944 relocatable files into an already running program; such systems
15945 typically make the requirements above easy to meet. However, it's
15946 important to recognize that many native systems use complex link
15947 procedures (@code{.linkonce} section factoring and C@t{++} constructor table
15948 assembly, for example) that make the requirements difficult to meet. In
15949 general, one cannot assume that using @code{add-symbol-file} to read a
15950 relocatable object file's symbolic information will have the same effect
15951 as linking the relocatable object file into the program in the normal
15954 @code{add-symbol-file} does not repeat if you press @key{RET} after using it.
15956 @kindex add-symbol-file-from-memory
15957 @cindex @code{syscall DSO}
15958 @cindex load symbols from memory
15959 @item add-symbol-file-from-memory @var{address}
15960 Load symbols from the given @var{address} in a dynamically loaded
15961 object file whose image is mapped directly into the inferior's memory.
15962 For example, the Linux kernel maps a @code{syscall DSO} into each
15963 process's address space; this DSO provides kernel-specific code for
15964 some system calls. The argument can be any expression whose
15965 evaluation yields the address of the file's shared object file header.
15966 For this command to work, you must have used @code{symbol-file} or
15967 @code{exec-file} commands in advance.
15969 @kindex add-shared-symbol-files
15971 @item add-shared-symbol-files @var{library-file}
15972 @itemx assf @var{library-file}
15973 The @code{add-shared-symbol-files} command can currently be used only
15974 in the Cygwin build of @value{GDBN} on MS-Windows OS, where it is an
15975 alias for the @code{dll-symbols} command (@pxref{Cygwin Native}).
15976 @value{GDBN} automatically looks for shared libraries, however if
15977 @value{GDBN} does not find yours, you can invoke
15978 @code{add-shared-symbol-files}. It takes one argument: the shared
15979 library's file name. @code{assf} is a shorthand alias for
15980 @code{add-shared-symbol-files}.
15983 @item section @var{section} @var{addr}
15984 The @code{section} command changes the base address of the named
15985 @var{section} of the exec file to @var{addr}. This can be used if the
15986 exec file does not contain section addresses, (such as in the
15987 @code{a.out} format), or when the addresses specified in the file
15988 itself are wrong. Each section must be changed separately. The
15989 @code{info files} command, described below, lists all the sections and
15993 @kindex info target
15996 @code{info files} and @code{info target} are synonymous; both print the
15997 current target (@pxref{Targets, ,Specifying a Debugging Target}),
15998 including the names of the executable and core dump files currently in
15999 use by @value{GDBN}, and the files from which symbols were loaded. The
16000 command @code{help target} lists all possible targets rather than
16003 @kindex maint info sections
16004 @item maint info sections
16005 Another command that can give you extra information about program sections
16006 is @code{maint info sections}. In addition to the section information
16007 displayed by @code{info files}, this command displays the flags and file
16008 offset of each section in the executable and core dump files. In addition,
16009 @code{maint info sections} provides the following command options (which
16010 may be arbitrarily combined):
16014 Display sections for all loaded object files, including shared libraries.
16015 @item @var{sections}
16016 Display info only for named @var{sections}.
16017 @item @var{section-flags}
16018 Display info only for sections for which @var{section-flags} are true.
16019 The section flags that @value{GDBN} currently knows about are:
16022 Section will have space allocated in the process when loaded.
16023 Set for all sections except those containing debug information.
16025 Section will be loaded from the file into the child process memory.
16026 Set for pre-initialized code and data, clear for @code{.bss} sections.
16028 Section needs to be relocated before loading.
16030 Section cannot be modified by the child process.
16032 Section contains executable code only.
16034 Section contains data only (no executable code).
16036 Section will reside in ROM.
16038 Section contains data for constructor/destructor lists.
16040 Section is not empty.
16042 An instruction to the linker to not output the section.
16043 @item COFF_SHARED_LIBRARY
16044 A notification to the linker that the section contains
16045 COFF shared library information.
16047 Section contains common symbols.
16050 @kindex set trust-readonly-sections
16051 @cindex read-only sections
16052 @item set trust-readonly-sections on
16053 Tell @value{GDBN} that readonly sections in your object file
16054 really are read-only (i.e.@: that their contents will not change).
16055 In that case, @value{GDBN} can fetch values from these sections
16056 out of the object file, rather than from the target program.
16057 For some targets (notably embedded ones), this can be a significant
16058 enhancement to debugging performance.
16060 The default is off.
16062 @item set trust-readonly-sections off
16063 Tell @value{GDBN} not to trust readonly sections. This means that
16064 the contents of the section might change while the program is running,
16065 and must therefore be fetched from the target when needed.
16067 @item show trust-readonly-sections
16068 Show the current setting of trusting readonly sections.
16071 All file-specifying commands allow both absolute and relative file names
16072 as arguments. @value{GDBN} always converts the file name to an absolute file
16073 name and remembers it that way.
16075 @cindex shared libraries
16076 @anchor{Shared Libraries}
16077 @value{GDBN} supports @sc{gnu}/Linux, MS-Windows, HP-UX, SunOS, SVr4, Irix,
16078 and IBM RS/6000 AIX shared libraries.
16080 On MS-Windows @value{GDBN} must be linked with the Expat library to support
16081 shared libraries. @xref{Expat}.
16083 @value{GDBN} automatically loads symbol definitions from shared libraries
16084 when you use the @code{run} command, or when you examine a core file.
16085 (Before you issue the @code{run} command, @value{GDBN} does not understand
16086 references to a function in a shared library, however---unless you are
16087 debugging a core file).
16089 On HP-UX, if the program loads a library explicitly, @value{GDBN}
16090 automatically loads the symbols at the time of the @code{shl_load} call.
16092 @c FIXME: some @value{GDBN} release may permit some refs to undef
16093 @c FIXME...symbols---eg in a break cmd---assuming they are from a shared
16094 @c FIXME...lib; check this from time to time when updating manual
16096 There are times, however, when you may wish to not automatically load
16097 symbol definitions from shared libraries, such as when they are
16098 particularly large or there are many of them.
16100 To control the automatic loading of shared library symbols, use the
16104 @kindex set auto-solib-add
16105 @item set auto-solib-add @var{mode}
16106 If @var{mode} is @code{on}, symbols from all shared object libraries
16107 will be loaded automatically when the inferior begins execution, you
16108 attach to an independently started inferior, or when the dynamic linker
16109 informs @value{GDBN} that a new library has been loaded. If @var{mode}
16110 is @code{off}, symbols must be loaded manually, using the
16111 @code{sharedlibrary} command. The default value is @code{on}.
16113 @cindex memory used for symbol tables
16114 If your program uses lots of shared libraries with debug info that
16115 takes large amounts of memory, you can decrease the @value{GDBN}
16116 memory footprint by preventing it from automatically loading the
16117 symbols from shared libraries. To that end, type @kbd{set
16118 auto-solib-add off} before running the inferior, then load each
16119 library whose debug symbols you do need with @kbd{sharedlibrary
16120 @var{regexp}}, where @var{regexp} is a regular expression that matches
16121 the libraries whose symbols you want to be loaded.
16123 @kindex show auto-solib-add
16124 @item show auto-solib-add
16125 Display the current autoloading mode.
16128 @cindex load shared library
16129 To explicitly load shared library symbols, use the @code{sharedlibrary}
16133 @kindex info sharedlibrary
16135 @item info share @var{regex}
16136 @itemx info sharedlibrary @var{regex}
16137 Print the names of the shared libraries which are currently loaded
16138 that match @var{regex}. If @var{regex} is omitted then print
16139 all shared libraries that are loaded.
16141 @kindex sharedlibrary
16143 @item sharedlibrary @var{regex}
16144 @itemx share @var{regex}
16145 Load shared object library symbols for files matching a
16146 Unix regular expression.
16147 As with files loaded automatically, it only loads shared libraries
16148 required by your program for a core file or after typing @code{run}. If
16149 @var{regex} is omitted all shared libraries required by your program are
16152 @item nosharedlibrary
16153 @kindex nosharedlibrary
16154 @cindex unload symbols from shared libraries
16155 Unload all shared object library symbols. This discards all symbols
16156 that have been loaded from all shared libraries. Symbols from shared
16157 libraries that were loaded by explicit user requests are not
16161 Sometimes you may wish that @value{GDBN} stops and gives you control
16162 when any of shared library events happen. The best way to do this is
16163 to use @code{catch load} and @code{catch unload} (@pxref{Set
16166 @value{GDBN} also supports the the @code{set stop-on-solib-events}
16167 command for this. This command exists for historical reasons. It is
16168 less useful than setting a catchpoint, because it does not allow for
16169 conditions or commands as a catchpoint does.
16172 @item set stop-on-solib-events
16173 @kindex set stop-on-solib-events
16174 This command controls whether @value{GDBN} should give you control
16175 when the dynamic linker notifies it about some shared library event.
16176 The most common event of interest is loading or unloading of a new
16179 @item show stop-on-solib-events
16180 @kindex show stop-on-solib-events
16181 Show whether @value{GDBN} stops and gives you control when shared
16182 library events happen.
16185 Shared libraries are also supported in many cross or remote debugging
16186 configurations. @value{GDBN} needs to have access to the target's libraries;
16187 this can be accomplished either by providing copies of the libraries
16188 on the host system, or by asking @value{GDBN} to automatically retrieve the
16189 libraries from the target. If copies of the target libraries are
16190 provided, they need to be the same as the target libraries, although the
16191 copies on the target can be stripped as long as the copies on the host are
16194 @cindex where to look for shared libraries
16195 For remote debugging, you need to tell @value{GDBN} where the target
16196 libraries are, so that it can load the correct copies---otherwise, it
16197 may try to load the host's libraries. @value{GDBN} has two variables
16198 to specify the search directories for target libraries.
16201 @cindex prefix for shared library file names
16202 @cindex system root, alternate
16203 @kindex set solib-absolute-prefix
16204 @kindex set sysroot
16205 @item set sysroot @var{path}
16206 Use @var{path} as the system root for the program being debugged. Any
16207 absolute shared library paths will be prefixed with @var{path}; many
16208 runtime loaders store the absolute paths to the shared library in the
16209 target program's memory. If you use @code{set sysroot} to find shared
16210 libraries, they need to be laid out in the same way that they are on
16211 the target, with e.g.@: a @file{/lib} and @file{/usr/lib} hierarchy
16214 If @var{path} starts with the sequence @file{remote:}, @value{GDBN} will
16215 retrieve the target libraries from the remote system. This is only
16216 supported when using a remote target that supports the @code{remote get}
16217 command (@pxref{File Transfer,,Sending files to a remote system}).
16218 The part of @var{path} following the initial @file{remote:}
16219 (if present) is used as system root prefix on the remote file system.
16220 @footnote{If you want to specify a local system root using a directory
16221 that happens to be named @file{remote:}, you need to use some equivalent
16222 variant of the name like @file{./remote:}.}
16224 For targets with an MS-DOS based filesystem, such as MS-Windows and
16225 SymbianOS, @value{GDBN} tries prefixing a few variants of the target
16226 absolute file name with @var{path}. But first, on Unix hosts,
16227 @value{GDBN} converts all backslash directory separators into forward
16228 slashes, because the backslash is not a directory separator on Unix:
16231 c:\foo\bar.dll @result{} c:/foo/bar.dll
16234 Then, @value{GDBN} attempts prefixing the target file name with
16235 @var{path}, and looks for the resulting file name in the host file
16239 c:/foo/bar.dll @result{} /path/to/sysroot/c:/foo/bar.dll
16242 If that does not find the shared library, @value{GDBN} tries removing
16243 the @samp{:} character from the drive spec, both for convenience, and,
16244 for the case of the host file system not supporting file names with
16248 c:/foo/bar.dll @result{} /path/to/sysroot/c/foo/bar.dll
16251 This makes it possible to have a system root that mirrors a target
16252 with more than one drive. E.g., you may want to setup your local
16253 copies of the target system shared libraries like so (note @samp{c} vs
16257 @file{/path/to/sysroot/c/sys/bin/foo.dll}
16258 @file{/path/to/sysroot/c/sys/bin/bar.dll}
16259 @file{/path/to/sysroot/z/sys/bin/bar.dll}
16263 and point the system root at @file{/path/to/sysroot}, so that
16264 @value{GDBN} can find the correct copies of both
16265 @file{c:\sys\bin\foo.dll}, and @file{z:\sys\bin\bar.dll}.
16267 If that still does not find the shared library, @value{GDBN} tries
16268 removing the whole drive spec from the target file name:
16271 c:/foo/bar.dll @result{} /path/to/sysroot/foo/bar.dll
16274 This last lookup makes it possible to not care about the drive name,
16275 if you don't want or need to.
16277 The @code{set solib-absolute-prefix} command is an alias for @code{set
16280 @cindex default system root
16281 @cindex @samp{--with-sysroot}
16282 You can set the default system root by using the configure-time
16283 @samp{--with-sysroot} option. If the system root is inside
16284 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
16285 @samp{--exec-prefix}), then the default system root will be updated
16286 automatically if the installed @value{GDBN} is moved to a new
16289 @kindex show sysroot
16291 Display the current shared library prefix.
16293 @kindex set solib-search-path
16294 @item set solib-search-path @var{path}
16295 If this variable is set, @var{path} is a colon-separated list of
16296 directories to search for shared libraries. @samp{solib-search-path}
16297 is used after @samp{sysroot} fails to locate the library, or if the
16298 path to the library is relative instead of absolute. If you want to
16299 use @samp{solib-search-path} instead of @samp{sysroot}, be sure to set
16300 @samp{sysroot} to a nonexistent directory to prevent @value{GDBN} from
16301 finding your host's libraries. @samp{sysroot} is preferred; setting
16302 it to a nonexistent directory may interfere with automatic loading
16303 of shared library symbols.
16305 @kindex show solib-search-path
16306 @item show solib-search-path
16307 Display the current shared library search path.
16309 @cindex DOS file-name semantics of file names.
16310 @kindex set target-file-system-kind (unix|dos-based|auto)
16311 @kindex show target-file-system-kind
16312 @item set target-file-system-kind @var{kind}
16313 Set assumed file system kind for target reported file names.
16315 Shared library file names as reported by the target system may not
16316 make sense as is on the system @value{GDBN} is running on. For
16317 example, when remote debugging a target that has MS-DOS based file
16318 system semantics, from a Unix host, the target may be reporting to
16319 @value{GDBN} a list of loaded shared libraries with file names such as
16320 @file{c:\Windows\kernel32.dll}. On Unix hosts, there's no concept of
16321 drive letters, so the @samp{c:\} prefix is not normally understood as
16322 indicating an absolute file name, and neither is the backslash
16323 normally considered a directory separator character. In that case,
16324 the native file system would interpret this whole absolute file name
16325 as a relative file name with no directory components. This would make
16326 it impossible to point @value{GDBN} at a copy of the remote target's
16327 shared libraries on the host using @code{set sysroot}, and impractical
16328 with @code{set solib-search-path}. Setting
16329 @code{target-file-system-kind} to @code{dos-based} tells @value{GDBN}
16330 to interpret such file names similarly to how the target would, and to
16331 map them to file names valid on @value{GDBN}'s native file system
16332 semantics. The value of @var{kind} can be @code{"auto"}, in addition
16333 to one of the supported file system kinds. In that case, @value{GDBN}
16334 tries to determine the appropriate file system variant based on the
16335 current target's operating system (@pxref{ABI, ,Configuring the
16336 Current ABI}). The supported file system settings are:
16340 Instruct @value{GDBN} to assume the target file system is of Unix
16341 kind. Only file names starting the forward slash (@samp{/}) character
16342 are considered absolute, and the directory separator character is also
16346 Instruct @value{GDBN} to assume the target file system is DOS based.
16347 File names starting with either a forward slash, or a drive letter
16348 followed by a colon (e.g., @samp{c:}), are considered absolute, and
16349 both the slash (@samp{/}) and the backslash (@samp{\\}) characters are
16350 considered directory separators.
16353 Instruct @value{GDBN} to use the file system kind associated with the
16354 target operating system (@pxref{ABI, ,Configuring the Current ABI}).
16355 This is the default.
16359 @cindex file name canonicalization
16360 @cindex base name differences
16361 When processing file names provided by the user, @value{GDBN}
16362 frequently needs to compare them to the file names recorded in the
16363 program's debug info. Normally, @value{GDBN} compares just the
16364 @dfn{base names} of the files as strings, which is reasonably fast
16365 even for very large programs. (The base name of a file is the last
16366 portion of its name, after stripping all the leading directories.)
16367 This shortcut in comparison is based upon the assumption that files
16368 cannot have more than one base name. This is usually true, but
16369 references to files that use symlinks or similar filesystem
16370 facilities violate that assumption. If your program records files
16371 using such facilities, or if you provide file names to @value{GDBN}
16372 using symlinks etc., you can set @code{basenames-may-differ} to
16373 @code{true} to instruct @value{GDBN} to completely canonicalize each
16374 pair of file names it needs to compare. This will make file-name
16375 comparisons accurate, but at a price of a significant slowdown.
16378 @item set basenames-may-differ
16379 @kindex set basenames-may-differ
16380 Set whether a source file may have multiple base names.
16382 @item show basenames-may-differ
16383 @kindex show basenames-may-differ
16384 Show whether a source file may have multiple base names.
16387 @node Separate Debug Files
16388 @section Debugging Information in Separate Files
16389 @cindex separate debugging information files
16390 @cindex debugging information in separate files
16391 @cindex @file{.debug} subdirectories
16392 @cindex debugging information directory, global
16393 @cindex global debugging information directories
16394 @cindex build ID, and separate debugging files
16395 @cindex @file{.build-id} directory
16397 @value{GDBN} allows you to put a program's debugging information in a
16398 file separate from the executable itself, in a way that allows
16399 @value{GDBN} to find and load the debugging information automatically.
16400 Since debugging information can be very large---sometimes larger
16401 than the executable code itself---some systems distribute debugging
16402 information for their executables in separate files, which users can
16403 install only when they need to debug a problem.
16405 @value{GDBN} supports two ways of specifying the separate debug info
16410 The executable contains a @dfn{debug link} that specifies the name of
16411 the separate debug info file. The separate debug file's name is
16412 usually @file{@var{executable}.debug}, where @var{executable} is the
16413 name of the corresponding executable file without leading directories
16414 (e.g., @file{ls.debug} for @file{/usr/bin/ls}). In addition, the
16415 debug link specifies a 32-bit @dfn{Cyclic Redundancy Check} (CRC)
16416 checksum for the debug file, which @value{GDBN} uses to validate that
16417 the executable and the debug file came from the same build.
16420 The executable contains a @dfn{build ID}, a unique bit string that is
16421 also present in the corresponding debug info file. (This is supported
16422 only on some operating systems, notably those which use the ELF format
16423 for binary files and the @sc{gnu} Binutils.) For more details about
16424 this feature, see the description of the @option{--build-id}
16425 command-line option in @ref{Options, , Command Line Options, ld.info,
16426 The GNU Linker}. The debug info file's name is not specified
16427 explicitly by the build ID, but can be computed from the build ID, see
16431 Depending on the way the debug info file is specified, @value{GDBN}
16432 uses two different methods of looking for the debug file:
16436 For the ``debug link'' method, @value{GDBN} looks up the named file in
16437 the directory of the executable file, then in a subdirectory of that
16438 directory named @file{.debug}, and finally under each one of the global debug
16439 directories, in a subdirectory whose name is identical to the leading
16440 directories of the executable's absolute file name.
16443 For the ``build ID'' method, @value{GDBN} looks in the
16444 @file{.build-id} subdirectory of each one of the global debug directories for
16445 a file named @file{@var{nn}/@var{nnnnnnnn}.debug}, where @var{nn} are the
16446 first 2 hex characters of the build ID bit string, and @var{nnnnnnnn}
16447 are the rest of the bit string. (Real build ID strings are 32 or more
16448 hex characters, not 10.)
16451 So, for example, suppose you ask @value{GDBN} to debug
16452 @file{/usr/bin/ls}, which has a debug link that specifies the
16453 file @file{ls.debug}, and a build ID whose value in hex is
16454 @code{abcdef1234}. If the list of the global debug directories includes
16455 @file{/usr/lib/debug}, then @value{GDBN} will look for the following
16456 debug information files, in the indicated order:
16460 @file{/usr/lib/debug/.build-id/ab/cdef1234.debug}
16462 @file{/usr/bin/ls.debug}
16464 @file{/usr/bin/.debug/ls.debug}
16466 @file{/usr/lib/debug/usr/bin/ls.debug}.
16469 @anchor{debug-file-directory}
16470 Global debugging info directories default to what is set by @value{GDBN}
16471 configure option @option{--with-separate-debug-dir}. During @value{GDBN} run
16472 you can also set the global debugging info directories, and view the list
16473 @value{GDBN} is currently using.
16477 @kindex set debug-file-directory
16478 @item set debug-file-directory @var{directories}
16479 Set the directories which @value{GDBN} searches for separate debugging
16480 information files to @var{directory}. Multiple path components can be set
16481 concatenating them by a path separator.
16483 @kindex show debug-file-directory
16484 @item show debug-file-directory
16485 Show the directories @value{GDBN} searches for separate debugging
16490 @cindex @code{.gnu_debuglink} sections
16491 @cindex debug link sections
16492 A debug link is a special section of the executable file named
16493 @code{.gnu_debuglink}. The section must contain:
16497 A filename, with any leading directory components removed, followed by
16500 zero to three bytes of padding, as needed to reach the next four-byte
16501 boundary within the section, and
16503 a four-byte CRC checksum, stored in the same endianness used for the
16504 executable file itself. The checksum is computed on the debugging
16505 information file's full contents by the function given below, passing
16506 zero as the @var{crc} argument.
16509 Any executable file format can carry a debug link, as long as it can
16510 contain a section named @code{.gnu_debuglink} with the contents
16513 @cindex @code{.note.gnu.build-id} sections
16514 @cindex build ID sections
16515 The build ID is a special section in the executable file (and in other
16516 ELF binary files that @value{GDBN} may consider). This section is
16517 often named @code{.note.gnu.build-id}, but that name is not mandatory.
16518 It contains unique identification for the built files---the ID remains
16519 the same across multiple builds of the same build tree. The default
16520 algorithm SHA1 produces 160 bits (40 hexadecimal characters) of the
16521 content for the build ID string. The same section with an identical
16522 value is present in the original built binary with symbols, in its
16523 stripped variant, and in the separate debugging information file.
16525 The debugging information file itself should be an ordinary
16526 executable, containing a full set of linker symbols, sections, and
16527 debugging information. The sections of the debugging information file
16528 should have the same names, addresses, and sizes as the original file,
16529 but they need not contain any data---much like a @code{.bss} section
16530 in an ordinary executable.
16532 The @sc{gnu} binary utilities (Binutils) package includes the
16533 @samp{objcopy} utility that can produce
16534 the separated executable / debugging information file pairs using the
16535 following commands:
16538 @kbd{objcopy --only-keep-debug foo foo.debug}
16543 These commands remove the debugging
16544 information from the executable file @file{foo} and place it in the file
16545 @file{foo.debug}. You can use the first, second or both methods to link the
16550 The debug link method needs the following additional command to also leave
16551 behind a debug link in @file{foo}:
16554 @kbd{objcopy --add-gnu-debuglink=foo.debug foo}
16557 Ulrich Drepper's @file{elfutils} package, starting with version 0.53, contains
16558 a version of the @code{strip} command such that the command @kbd{strip foo -f
16559 foo.debug} has the same functionality as the two @code{objcopy} commands and
16560 the @code{ln -s} command above, together.
16563 Build ID gets embedded into the main executable using @code{ld --build-id} or
16564 the @value{NGCC} counterpart @code{gcc -Wl,--build-id}. Build ID support plus
16565 compatibility fixes for debug files separation are present in @sc{gnu} binary
16566 utilities (Binutils) package since version 2.18.
16571 @cindex CRC algorithm definition
16572 The CRC used in @code{.gnu_debuglink} is the CRC-32 defined in
16573 IEEE 802.3 using the polynomial:
16575 @c TexInfo requires naked braces for multi-digit exponents for Tex
16576 @c output, but this causes HTML output to barf. HTML has to be set using
16577 @c raw commands. So we end up having to specify this equation in 2
16582 <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>
16583 + <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
16589 @math{x^{32} + x^{26} + x^{23} + x^{22} + x^{16} + x^{12} + x^{11}}
16590 @math{+ x^{10} + x^8 + x^7 + x^5 + x^4 + x^2 + x + 1}
16594 The function is computed byte at a time, taking the least
16595 significant bit of each byte first. The initial pattern
16596 @code{0xffffffff} is used, to ensure leading zeros affect the CRC and
16597 the final result is inverted to ensure trailing zeros also affect the
16600 @emph{Note:} This is the same CRC polynomial as used in handling the
16601 @dfn{Remote Serial Protocol} @code{qCRC} packet (@pxref{Remote Protocol,
16602 , @value{GDBN} Remote Serial Protocol}). However in the
16603 case of the Remote Serial Protocol, the CRC is computed @emph{most}
16604 significant bit first, and the result is not inverted, so trailing
16605 zeros have no effect on the CRC value.
16607 To complete the description, we show below the code of the function
16608 which produces the CRC used in @code{.gnu_debuglink}. Inverting the
16609 initially supplied @code{crc} argument means that an initial call to
16610 this function passing in zero will start computing the CRC using
16613 @kindex gnu_debuglink_crc32
16616 gnu_debuglink_crc32 (unsigned long crc,
16617 unsigned char *buf, size_t len)
16619 static const unsigned long crc32_table[256] =
16621 0x00000000, 0x77073096, 0xee0e612c, 0x990951ba, 0x076dc419,
16622 0x706af48f, 0xe963a535, 0x9e6495a3, 0x0edb8832, 0x79dcb8a4,
16623 0xe0d5e91e, 0x97d2d988, 0x09b64c2b, 0x7eb17cbd, 0xe7b82d07,
16624 0x90bf1d91, 0x1db71064, 0x6ab020f2, 0xf3b97148, 0x84be41de,
16625 0x1adad47d, 0x6ddde4eb, 0xf4d4b551, 0x83d385c7, 0x136c9856,
16626 0x646ba8c0, 0xfd62f97a, 0x8a65c9ec, 0x14015c4f, 0x63066cd9,
16627 0xfa0f3d63, 0x8d080df5, 0x3b6e20c8, 0x4c69105e, 0xd56041e4,
16628 0xa2677172, 0x3c03e4d1, 0x4b04d447, 0xd20d85fd, 0xa50ab56b,
16629 0x35b5a8fa, 0x42b2986c, 0xdbbbc9d6, 0xacbcf940, 0x32d86ce3,
16630 0x45df5c75, 0xdcd60dcf, 0xabd13d59, 0x26d930ac, 0x51de003a,
16631 0xc8d75180, 0xbfd06116, 0x21b4f4b5, 0x56b3c423, 0xcfba9599,
16632 0xb8bda50f, 0x2802b89e, 0x5f058808, 0xc60cd9b2, 0xb10be924,
16633 0x2f6f7c87, 0x58684c11, 0xc1611dab, 0xb6662d3d, 0x76dc4190,
16634 0x01db7106, 0x98d220bc, 0xefd5102a, 0x71b18589, 0x06b6b51f,
16635 0x9fbfe4a5, 0xe8b8d433, 0x7807c9a2, 0x0f00f934, 0x9609a88e,
16636 0xe10e9818, 0x7f6a0dbb, 0x086d3d2d, 0x91646c97, 0xe6635c01,
16637 0x6b6b51f4, 0x1c6c6162, 0x856530d8, 0xf262004e, 0x6c0695ed,
16638 0x1b01a57b, 0x8208f4c1, 0xf50fc457, 0x65b0d9c6, 0x12b7e950,
16639 0x8bbeb8ea, 0xfcb9887c, 0x62dd1ddf, 0x15da2d49, 0x8cd37cf3,
16640 0xfbd44c65, 0x4db26158, 0x3ab551ce, 0xa3bc0074, 0xd4bb30e2,
16641 0x4adfa541, 0x3dd895d7, 0xa4d1c46d, 0xd3d6f4fb, 0x4369e96a,
16642 0x346ed9fc, 0xad678846, 0xda60b8d0, 0x44042d73, 0x33031de5,
16643 0xaa0a4c5f, 0xdd0d7cc9, 0x5005713c, 0x270241aa, 0xbe0b1010,
16644 0xc90c2086, 0x5768b525, 0x206f85b3, 0xb966d409, 0xce61e49f,
16645 0x5edef90e, 0x29d9c998, 0xb0d09822, 0xc7d7a8b4, 0x59b33d17,
16646 0x2eb40d81, 0xb7bd5c3b, 0xc0ba6cad, 0xedb88320, 0x9abfb3b6,
16647 0x03b6e20c, 0x74b1d29a, 0xead54739, 0x9dd277af, 0x04db2615,
16648 0x73dc1683, 0xe3630b12, 0x94643b84, 0x0d6d6a3e, 0x7a6a5aa8,
16649 0xe40ecf0b, 0x9309ff9d, 0x0a00ae27, 0x7d079eb1, 0xf00f9344,
16650 0x8708a3d2, 0x1e01f268, 0x6906c2fe, 0xf762575d, 0x806567cb,
16651 0x196c3671, 0x6e6b06e7, 0xfed41b76, 0x89d32be0, 0x10da7a5a,
16652 0x67dd4acc, 0xf9b9df6f, 0x8ebeeff9, 0x17b7be43, 0x60b08ed5,
16653 0xd6d6a3e8, 0xa1d1937e, 0x38d8c2c4, 0x4fdff252, 0xd1bb67f1,
16654 0xa6bc5767, 0x3fb506dd, 0x48b2364b, 0xd80d2bda, 0xaf0a1b4c,
16655 0x36034af6, 0x41047a60, 0xdf60efc3, 0xa867df55, 0x316e8eef,
16656 0x4669be79, 0xcb61b38c, 0xbc66831a, 0x256fd2a0, 0x5268e236,
16657 0xcc0c7795, 0xbb0b4703, 0x220216b9, 0x5505262f, 0xc5ba3bbe,
16658 0xb2bd0b28, 0x2bb45a92, 0x5cb36a04, 0xc2d7ffa7, 0xb5d0cf31,
16659 0x2cd99e8b, 0x5bdeae1d, 0x9b64c2b0, 0xec63f226, 0x756aa39c,
16660 0x026d930a, 0x9c0906a9, 0xeb0e363f, 0x72076785, 0x05005713,
16661 0x95bf4a82, 0xe2b87a14, 0x7bb12bae, 0x0cb61b38, 0x92d28e9b,
16662 0xe5d5be0d, 0x7cdcefb7, 0x0bdbdf21, 0x86d3d2d4, 0xf1d4e242,
16663 0x68ddb3f8, 0x1fda836e, 0x81be16cd, 0xf6b9265b, 0x6fb077e1,
16664 0x18b74777, 0x88085ae6, 0xff0f6a70, 0x66063bca, 0x11010b5c,
16665 0x8f659eff, 0xf862ae69, 0x616bffd3, 0x166ccf45, 0xa00ae278,
16666 0xd70dd2ee, 0x4e048354, 0x3903b3c2, 0xa7672661, 0xd06016f7,
16667 0x4969474d, 0x3e6e77db, 0xaed16a4a, 0xd9d65adc, 0x40df0b66,
16668 0x37d83bf0, 0xa9bcae53, 0xdebb9ec5, 0x47b2cf7f, 0x30b5ffe9,
16669 0xbdbdf21c, 0xcabac28a, 0x53b39330, 0x24b4a3a6, 0xbad03605,
16670 0xcdd70693, 0x54de5729, 0x23d967bf, 0xb3667a2e, 0xc4614ab8,
16671 0x5d681b02, 0x2a6f2b94, 0xb40bbe37, 0xc30c8ea1, 0x5a05df1b,
16674 unsigned char *end;
16676 crc = ~crc & 0xffffffff;
16677 for (end = buf + len; buf < end; ++buf)
16678 crc = crc32_table[(crc ^ *buf) & 0xff] ^ (crc >> 8);
16679 return ~crc & 0xffffffff;
16684 This computation does not apply to the ``build ID'' method.
16688 @section Index Files Speed Up @value{GDBN}
16689 @cindex index files
16690 @cindex @samp{.gdb_index} section
16692 When @value{GDBN} finds a symbol file, it scans the symbols in the
16693 file in order to construct an internal symbol table. This lets most
16694 @value{GDBN} operations work quickly---at the cost of a delay early
16695 on. For large programs, this delay can be quite lengthy, so
16696 @value{GDBN} provides a way to build an index, which speeds up
16699 The index is stored as a section in the symbol file. @value{GDBN} can
16700 write the index to a file, then you can put it into the symbol file
16701 using @command{objcopy}.
16703 To create an index file, use the @code{save gdb-index} command:
16706 @item save gdb-index @var{directory}
16707 @kindex save gdb-index
16708 Create an index file for each symbol file currently known by
16709 @value{GDBN}. Each file is named after its corresponding symbol file,
16710 with @samp{.gdb-index} appended, and is written into the given
16714 Once you have created an index file you can merge it into your symbol
16715 file, here named @file{symfile}, using @command{objcopy}:
16718 $ objcopy --add-section .gdb_index=symfile.gdb-index \
16719 --set-section-flags .gdb_index=readonly symfile symfile
16722 There are currently some limitation on indices. They only work when
16723 for DWARF debugging information, not stabs. And, they do not
16724 currently work for programs using Ada.
16726 @node Symbol Errors
16727 @section Errors Reading Symbol Files
16729 While reading a symbol file, @value{GDBN} occasionally encounters problems,
16730 such as symbol types it does not recognize, or known bugs in compiler
16731 output. By default, @value{GDBN} does not notify you of such problems, since
16732 they are relatively common and primarily of interest to people
16733 debugging compilers. If you are interested in seeing information
16734 about ill-constructed symbol tables, you can either ask @value{GDBN} to print
16735 only one message about each such type of problem, no matter how many
16736 times the problem occurs; or you can ask @value{GDBN} to print more messages,
16737 to see how many times the problems occur, with the @code{set
16738 complaints} command (@pxref{Messages/Warnings, ,Optional Warnings and
16741 The messages currently printed, and their meanings, include:
16744 @item inner block not inside outer block in @var{symbol}
16746 The symbol information shows where symbol scopes begin and end
16747 (such as at the start of a function or a block of statements). This
16748 error indicates that an inner scope block is not fully contained
16749 in its outer scope blocks.
16751 @value{GDBN} circumvents the problem by treating the inner block as if it had
16752 the same scope as the outer block. In the error message, @var{symbol}
16753 may be shown as ``@code{(don't know)}'' if the outer block is not a
16756 @item block at @var{address} out of order
16758 The symbol information for symbol scope blocks should occur in
16759 order of increasing addresses. This error indicates that it does not
16762 @value{GDBN} does not circumvent this problem, and has trouble
16763 locating symbols in the source file whose symbols it is reading. (You
16764 can often determine what source file is affected by specifying
16765 @code{set verbose on}. @xref{Messages/Warnings, ,Optional Warnings and
16768 @item bad block start address patched
16770 The symbol information for a symbol scope block has a start address
16771 smaller than the address of the preceding source line. This is known
16772 to occur in the SunOS 4.1.1 (and earlier) C compiler.
16774 @value{GDBN} circumvents the problem by treating the symbol scope block as
16775 starting on the previous source line.
16777 @item bad string table offset in symbol @var{n}
16780 Symbol number @var{n} contains a pointer into the string table which is
16781 larger than the size of the string table.
16783 @value{GDBN} circumvents the problem by considering the symbol to have the
16784 name @code{foo}, which may cause other problems if many symbols end up
16787 @item unknown symbol type @code{0x@var{nn}}
16789 The symbol information contains new data types that @value{GDBN} does
16790 not yet know how to read. @code{0x@var{nn}} is the symbol type of the
16791 uncomprehended information, in hexadecimal.
16793 @value{GDBN} circumvents the error by ignoring this symbol information.
16794 This usually allows you to debug your program, though certain symbols
16795 are not accessible. If you encounter such a problem and feel like
16796 debugging it, you can debug @code{@value{GDBP}} with itself, breakpoint
16797 on @code{complain}, then go up to the function @code{read_dbx_symtab}
16798 and examine @code{*bufp} to see the symbol.
16800 @item stub type has NULL name
16802 @value{GDBN} could not find the full definition for a struct or class.
16804 @item const/volatile indicator missing (ok if using g++ v1.x), got@dots{}
16805 The symbol information for a C@t{++} member function is missing some
16806 information that recent versions of the compiler should have output for
16809 @item info mismatch between compiler and debugger
16811 @value{GDBN} could not parse a type specification output by the compiler.
16816 @section GDB Data Files
16818 @cindex prefix for data files
16819 @value{GDBN} will sometimes read an auxiliary data file. These files
16820 are kept in a directory known as the @dfn{data directory}.
16822 You can set the data directory's name, and view the name @value{GDBN}
16823 is currently using.
16826 @kindex set data-directory
16827 @item set data-directory @var{directory}
16828 Set the directory which @value{GDBN} searches for auxiliary data files
16829 to @var{directory}.
16831 @kindex show data-directory
16832 @item show data-directory
16833 Show the directory @value{GDBN} searches for auxiliary data files.
16836 @cindex default data directory
16837 @cindex @samp{--with-gdb-datadir}
16838 You can set the default data directory by using the configure-time
16839 @samp{--with-gdb-datadir} option. If the data directory is inside
16840 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
16841 @samp{--exec-prefix}), then the default data directory will be updated
16842 automatically if the installed @value{GDBN} is moved to a new
16845 The data directory may also be specified with the
16846 @code{--data-directory} command line option.
16847 @xref{Mode Options}.
16850 @chapter Specifying a Debugging Target
16852 @cindex debugging target
16853 A @dfn{target} is the execution environment occupied by your program.
16855 Often, @value{GDBN} runs in the same host environment as your program;
16856 in that case, the debugging target is specified as a side effect when
16857 you use the @code{file} or @code{core} commands. When you need more
16858 flexibility---for example, running @value{GDBN} on a physically separate
16859 host, or controlling a standalone system over a serial port or a
16860 realtime system over a TCP/IP connection---you can use the @code{target}
16861 command to specify one of the target types configured for @value{GDBN}
16862 (@pxref{Target Commands, ,Commands for Managing Targets}).
16864 @cindex target architecture
16865 It is possible to build @value{GDBN} for several different @dfn{target
16866 architectures}. When @value{GDBN} is built like that, you can choose
16867 one of the available architectures with the @kbd{set architecture}
16871 @kindex set architecture
16872 @kindex show architecture
16873 @item set architecture @var{arch}
16874 This command sets the current target architecture to @var{arch}. The
16875 value of @var{arch} can be @code{"auto"}, in addition to one of the
16876 supported architectures.
16878 @item show architecture
16879 Show the current target architecture.
16881 @item set processor
16883 @kindex set processor
16884 @kindex show processor
16885 These are alias commands for, respectively, @code{set architecture}
16886 and @code{show architecture}.
16890 * Active Targets:: Active targets
16891 * Target Commands:: Commands for managing targets
16892 * Byte Order:: Choosing target byte order
16895 @node Active Targets
16896 @section Active Targets
16898 @cindex stacking targets
16899 @cindex active targets
16900 @cindex multiple targets
16902 There are multiple classes of targets such as: processes, executable files or
16903 recording sessions. Core files belong to the process class, making core file
16904 and process mutually exclusive. Otherwise, @value{GDBN} can work concurrently
16905 on multiple active targets, one in each class. This allows you to (for
16906 example) start a process and inspect its activity, while still having access to
16907 the executable file after the process finishes. Or if you start process
16908 recording (@pxref{Reverse Execution}) and @code{reverse-step} there, you are
16909 presented a virtual layer of the recording target, while the process target
16910 remains stopped at the chronologically last point of the process execution.
16912 Use the @code{core-file} and @code{exec-file} commands to select a new core
16913 file or executable target (@pxref{Files, ,Commands to Specify Files}). To
16914 specify as a target a process that is already running, use the @code{attach}
16915 command (@pxref{Attach, ,Debugging an Already-running Process}).
16917 @node Target Commands
16918 @section Commands for Managing Targets
16921 @item target @var{type} @var{parameters}
16922 Connects the @value{GDBN} host environment to a target machine or
16923 process. A target is typically a protocol for talking to debugging
16924 facilities. You use the argument @var{type} to specify the type or
16925 protocol of the target machine.
16927 Further @var{parameters} are interpreted by the target protocol, but
16928 typically include things like device names or host names to connect
16929 with, process numbers, and baud rates.
16931 The @code{target} command does not repeat if you press @key{RET} again
16932 after executing the command.
16934 @kindex help target
16936 Displays the names of all targets available. To display targets
16937 currently selected, use either @code{info target} or @code{info files}
16938 (@pxref{Files, ,Commands to Specify Files}).
16940 @item help target @var{name}
16941 Describe a particular target, including any parameters necessary to
16944 @kindex set gnutarget
16945 @item set gnutarget @var{args}
16946 @value{GDBN} uses its own library BFD to read your files. @value{GDBN}
16947 knows whether it is reading an @dfn{executable},
16948 a @dfn{core}, or a @dfn{.o} file; however, you can specify the file format
16949 with the @code{set gnutarget} command. Unlike most @code{target} commands,
16950 with @code{gnutarget} the @code{target} refers to a program, not a machine.
16953 @emph{Warning:} To specify a file format with @code{set gnutarget},
16954 you must know the actual BFD name.
16958 @xref{Files, , Commands to Specify Files}.
16960 @kindex show gnutarget
16961 @item show gnutarget
16962 Use the @code{show gnutarget} command to display what file format
16963 @code{gnutarget} is set to read. If you have not set @code{gnutarget},
16964 @value{GDBN} will determine the file format for each file automatically,
16965 and @code{show gnutarget} displays @samp{The current BDF target is "auto"}.
16968 @cindex common targets
16969 Here are some common targets (available, or not, depending on the GDB
16974 @item target exec @var{program}
16975 @cindex executable file target
16976 An executable file. @samp{target exec @var{program}} is the same as
16977 @samp{exec-file @var{program}}.
16979 @item target core @var{filename}
16980 @cindex core dump file target
16981 A core dump file. @samp{target core @var{filename}} is the same as
16982 @samp{core-file @var{filename}}.
16984 @item target remote @var{medium}
16985 @cindex remote target
16986 A remote system connected to @value{GDBN} via a serial line or network
16987 connection. This command tells @value{GDBN} to use its own remote
16988 protocol over @var{medium} for debugging. @xref{Remote Debugging}.
16990 For example, if you have a board connected to @file{/dev/ttya} on the
16991 machine running @value{GDBN}, you could say:
16994 target remote /dev/ttya
16997 @code{target remote} supports the @code{load} command. This is only
16998 useful if you have some other way of getting the stub to the target
16999 system, and you can put it somewhere in memory where it won't get
17000 clobbered by the download.
17002 @item target sim @r{[}@var{simargs}@r{]} @dots{}
17003 @cindex built-in simulator target
17004 Builtin CPU simulator. @value{GDBN} includes simulators for most architectures.
17012 works; however, you cannot assume that a specific memory map, device
17013 drivers, or even basic I/O is available, although some simulators do
17014 provide these. For info about any processor-specific simulator details,
17015 see the appropriate section in @ref{Embedded Processors, ,Embedded
17020 Some configurations may include these targets as well:
17024 @item target nrom @var{dev}
17025 @cindex NetROM ROM emulator target
17026 NetROM ROM emulator. This target only supports downloading.
17030 Different targets are available on different configurations of @value{GDBN};
17031 your configuration may have more or fewer targets.
17033 Many remote targets require you to download the executable's code once
17034 you've successfully established a connection. You may wish to control
17035 various aspects of this process.
17040 @kindex set hash@r{, for remote monitors}
17041 @cindex hash mark while downloading
17042 This command controls whether a hash mark @samp{#} is displayed while
17043 downloading a file to the remote monitor. If on, a hash mark is
17044 displayed after each S-record is successfully downloaded to the
17048 @kindex show hash@r{, for remote monitors}
17049 Show the current status of displaying the hash mark.
17051 @item set debug monitor
17052 @kindex set debug monitor
17053 @cindex display remote monitor communications
17054 Enable or disable display of communications messages between
17055 @value{GDBN} and the remote monitor.
17057 @item show debug monitor
17058 @kindex show debug monitor
17059 Show the current status of displaying communications between
17060 @value{GDBN} and the remote monitor.
17065 @kindex load @var{filename}
17066 @item load @var{filename}
17068 Depending on what remote debugging facilities are configured into
17069 @value{GDBN}, the @code{load} command may be available. Where it exists, it
17070 is meant to make @var{filename} (an executable) available for debugging
17071 on the remote system---by downloading, or dynamic linking, for example.
17072 @code{load} also records the @var{filename} symbol table in @value{GDBN}, like
17073 the @code{add-symbol-file} command.
17075 If your @value{GDBN} does not have a @code{load} command, attempting to
17076 execute it gets the error message ``@code{You can't do that when your
17077 target is @dots{}}''
17079 The file is loaded at whatever address is specified in the executable.
17080 For some object file formats, you can specify the load address when you
17081 link the program; for other formats, like a.out, the object file format
17082 specifies a fixed address.
17083 @c FIXME! This would be a good place for an xref to the GNU linker doc.
17085 Depending on the remote side capabilities, @value{GDBN} may be able to
17086 load programs into flash memory.
17088 @code{load} does not repeat if you press @key{RET} again after using it.
17092 @section Choosing Target Byte Order
17094 @cindex choosing target byte order
17095 @cindex target byte order
17097 Some types of processors, such as the @acronym{MIPS}, PowerPC, and Renesas SH,
17098 offer the ability to run either big-endian or little-endian byte
17099 orders. Usually the executable or symbol will include a bit to
17100 designate the endian-ness, and you will not need to worry about
17101 which to use. However, you may still find it useful to adjust
17102 @value{GDBN}'s idea of processor endian-ness manually.
17106 @item set endian big
17107 Instruct @value{GDBN} to assume the target is big-endian.
17109 @item set endian little
17110 Instruct @value{GDBN} to assume the target is little-endian.
17112 @item set endian auto
17113 Instruct @value{GDBN} to use the byte order associated with the
17117 Display @value{GDBN}'s current idea of the target byte order.
17121 Note that these commands merely adjust interpretation of symbolic
17122 data on the host, and that they have absolutely no effect on the
17126 @node Remote Debugging
17127 @chapter Debugging Remote Programs
17128 @cindex remote debugging
17130 If you are trying to debug a program running on a machine that cannot run
17131 @value{GDBN} in the usual way, it is often useful to use remote debugging.
17132 For example, you might use remote debugging on an operating system kernel,
17133 or on a small system which does not have a general purpose operating system
17134 powerful enough to run a full-featured debugger.
17136 Some configurations of @value{GDBN} have special serial or TCP/IP interfaces
17137 to make this work with particular debugging targets. In addition,
17138 @value{GDBN} comes with a generic serial protocol (specific to @value{GDBN},
17139 but not specific to any particular target system) which you can use if you
17140 write the remote stubs---the code that runs on the remote system to
17141 communicate with @value{GDBN}.
17143 Other remote targets may be available in your
17144 configuration of @value{GDBN}; use @code{help target} to list them.
17147 * Connecting:: Connecting to a remote target
17148 * File Transfer:: Sending files to a remote system
17149 * Server:: Using the gdbserver program
17150 * Remote Configuration:: Remote configuration
17151 * Remote Stub:: Implementing a remote stub
17155 @section Connecting to a Remote Target
17157 On the @value{GDBN} host machine, you will need an unstripped copy of
17158 your program, since @value{GDBN} needs symbol and debugging information.
17159 Start up @value{GDBN} as usual, using the name of the local copy of your
17160 program as the first argument.
17162 @cindex @code{target remote}
17163 @value{GDBN} can communicate with the target over a serial line, or
17164 over an @acronym{IP} network using @acronym{TCP} or @acronym{UDP}. In
17165 each case, @value{GDBN} uses the same protocol for debugging your
17166 program; only the medium carrying the debugging packets varies. The
17167 @code{target remote} command establishes a connection to the target.
17168 Its arguments indicate which medium to use:
17172 @item target remote @var{serial-device}
17173 @cindex serial line, @code{target remote}
17174 Use @var{serial-device} to communicate with the target. For example,
17175 to use a serial line connected to the device named @file{/dev/ttyb}:
17178 target remote /dev/ttyb
17181 If you're using a serial line, you may want to give @value{GDBN} the
17182 @w{@samp{--baud}} option, or use the @code{set remotebaud} command
17183 (@pxref{Remote Configuration, set remotebaud}) before the
17184 @code{target} command.
17186 @item target remote @code{@var{host}:@var{port}}
17187 @itemx target remote @code{tcp:@var{host}:@var{port}}
17188 @cindex @acronym{TCP} port, @code{target remote}
17189 Debug using a @acronym{TCP} connection to @var{port} on @var{host}.
17190 The @var{host} may be either a host name or a numeric @acronym{IP}
17191 address; @var{port} must be a decimal number. The @var{host} could be
17192 the target machine itself, if it is directly connected to the net, or
17193 it might be a terminal server which in turn has a serial line to the
17196 For example, to connect to port 2828 on a terminal server named
17200 target remote manyfarms:2828
17203 If your remote target is actually running on the same machine as your
17204 debugger session (e.g.@: a simulator for your target running on the
17205 same host), you can omit the hostname. For example, to connect to
17206 port 1234 on your local machine:
17209 target remote :1234
17213 Note that the colon is still required here.
17215 @item target remote @code{udp:@var{host}:@var{port}}
17216 @cindex @acronym{UDP} port, @code{target remote}
17217 Debug using @acronym{UDP} packets to @var{port} on @var{host}. For example, to
17218 connect to @acronym{UDP} port 2828 on a terminal server named @code{manyfarms}:
17221 target remote udp:manyfarms:2828
17224 When using a @acronym{UDP} connection for remote debugging, you should
17225 keep in mind that the `U' stands for ``Unreliable''. @acronym{UDP}
17226 can silently drop packets on busy or unreliable networks, which will
17227 cause havoc with your debugging session.
17229 @item target remote | @var{command}
17230 @cindex pipe, @code{target remote} to
17231 Run @var{command} in the background and communicate with it using a
17232 pipe. The @var{command} is a shell command, to be parsed and expanded
17233 by the system's command shell, @code{/bin/sh}; it should expect remote
17234 protocol packets on its standard input, and send replies on its
17235 standard output. You could use this to run a stand-alone simulator
17236 that speaks the remote debugging protocol, to make net connections
17237 using programs like @code{ssh}, or for other similar tricks.
17239 If @var{command} closes its standard output (perhaps by exiting),
17240 @value{GDBN} will try to send it a @code{SIGTERM} signal. (If the
17241 program has already exited, this will have no effect.)
17245 Once the connection has been established, you can use all the usual
17246 commands to examine and change data. The remote program is already
17247 running; you can use @kbd{step} and @kbd{continue}, and you do not
17248 need to use @kbd{run}.
17250 @cindex interrupting remote programs
17251 @cindex remote programs, interrupting
17252 Whenever @value{GDBN} is waiting for the remote program, if you type the
17253 interrupt character (often @kbd{Ctrl-c}), @value{GDBN} attempts to stop the
17254 program. This may or may not succeed, depending in part on the hardware
17255 and the serial drivers the remote system uses. If you type the
17256 interrupt character once again, @value{GDBN} displays this prompt:
17259 Interrupted while waiting for the program.
17260 Give up (and stop debugging it)? (y or n)
17263 If you type @kbd{y}, @value{GDBN} abandons the remote debugging session.
17264 (If you decide you want to try again later, you can use @samp{target
17265 remote} again to connect once more.) If you type @kbd{n}, @value{GDBN}
17266 goes back to waiting.
17269 @kindex detach (remote)
17271 When you have finished debugging the remote program, you can use the
17272 @code{detach} command to release it from @value{GDBN} control.
17273 Detaching from the target normally resumes its execution, but the results
17274 will depend on your particular remote stub. After the @code{detach}
17275 command, @value{GDBN} is free to connect to another target.
17279 The @code{disconnect} command behaves like @code{detach}, except that
17280 the target is generally not resumed. It will wait for @value{GDBN}
17281 (this instance or another one) to connect and continue debugging. After
17282 the @code{disconnect} command, @value{GDBN} is again free to connect to
17285 @cindex send command to remote monitor
17286 @cindex extend @value{GDBN} for remote targets
17287 @cindex add new commands for external monitor
17289 @item monitor @var{cmd}
17290 This command allows you to send arbitrary commands directly to the
17291 remote monitor. Since @value{GDBN} doesn't care about the commands it
17292 sends like this, this command is the way to extend @value{GDBN}---you
17293 can add new commands that only the external monitor will understand
17297 @node File Transfer
17298 @section Sending files to a remote system
17299 @cindex remote target, file transfer
17300 @cindex file transfer
17301 @cindex sending files to remote systems
17303 Some remote targets offer the ability to transfer files over the same
17304 connection used to communicate with @value{GDBN}. This is convenient
17305 for targets accessible through other means, e.g.@: @sc{gnu}/Linux systems
17306 running @code{gdbserver} over a network interface. For other targets,
17307 e.g.@: embedded devices with only a single serial port, this may be
17308 the only way to upload or download files.
17310 Not all remote targets support these commands.
17314 @item remote put @var{hostfile} @var{targetfile}
17315 Copy file @var{hostfile} from the host system (the machine running
17316 @value{GDBN}) to @var{targetfile} on the target system.
17319 @item remote get @var{targetfile} @var{hostfile}
17320 Copy file @var{targetfile} from the target system to @var{hostfile}
17321 on the host system.
17323 @kindex remote delete
17324 @item remote delete @var{targetfile}
17325 Delete @var{targetfile} from the target system.
17330 @section Using the @code{gdbserver} Program
17333 @cindex remote connection without stubs
17334 @code{gdbserver} is a control program for Unix-like systems, which
17335 allows you to connect your program with a remote @value{GDBN} via
17336 @code{target remote}---but without linking in the usual debugging stub.
17338 @code{gdbserver} is not a complete replacement for the debugging stubs,
17339 because it requires essentially the same operating-system facilities
17340 that @value{GDBN} itself does. In fact, a system that can run
17341 @code{gdbserver} to connect to a remote @value{GDBN} could also run
17342 @value{GDBN} locally! @code{gdbserver} is sometimes useful nevertheless,
17343 because it is a much smaller program than @value{GDBN} itself. It is
17344 also easier to port than all of @value{GDBN}, so you may be able to get
17345 started more quickly on a new system by using @code{gdbserver}.
17346 Finally, if you develop code for real-time systems, you may find that
17347 the tradeoffs involved in real-time operation make it more convenient to
17348 do as much development work as possible on another system, for example
17349 by cross-compiling. You can use @code{gdbserver} to make a similar
17350 choice for debugging.
17352 @value{GDBN} and @code{gdbserver} communicate via either a serial line
17353 or a TCP connection, using the standard @value{GDBN} remote serial
17357 @emph{Warning:} @code{gdbserver} does not have any built-in security.
17358 Do not run @code{gdbserver} connected to any public network; a
17359 @value{GDBN} connection to @code{gdbserver} provides access to the
17360 target system with the same privileges as the user running
17364 @subsection Running @code{gdbserver}
17365 @cindex arguments, to @code{gdbserver}
17366 @cindex @code{gdbserver}, command-line arguments
17368 Run @code{gdbserver} on the target system. You need a copy of the
17369 program you want to debug, including any libraries it requires.
17370 @code{gdbserver} does not need your program's symbol table, so you can
17371 strip the program if necessary to save space. @value{GDBN} on the host
17372 system does all the symbol handling.
17374 To use the server, you must tell it how to communicate with @value{GDBN};
17375 the name of your program; and the arguments for your program. The usual
17379 target> gdbserver @var{comm} @var{program} [ @var{args} @dots{} ]
17382 @var{comm} is either a device name (to use a serial line), or a TCP
17383 hostname and portnumber, or @code{-} or @code{stdio} to use
17384 stdin/stdout of @code{gdbserver}.
17385 For example, to debug Emacs with the argument
17386 @samp{foo.txt} and communicate with @value{GDBN} over the serial port
17390 target> gdbserver /dev/com1 emacs foo.txt
17393 @code{gdbserver} waits passively for the host @value{GDBN} to communicate
17396 To use a TCP connection instead of a serial line:
17399 target> gdbserver host:2345 emacs foo.txt
17402 The only difference from the previous example is the first argument,
17403 specifying that you are communicating with the host @value{GDBN} via
17404 TCP. The @samp{host:2345} argument means that @code{gdbserver} is to
17405 expect a TCP connection from machine @samp{host} to local TCP port 2345.
17406 (Currently, the @samp{host} part is ignored.) You can choose any number
17407 you want for the port number as long as it does not conflict with any
17408 TCP ports already in use on the target system (for example, @code{23} is
17409 reserved for @code{telnet}).@footnote{If you choose a port number that
17410 conflicts with another service, @code{gdbserver} prints an error message
17411 and exits.} You must use the same port number with the host @value{GDBN}
17412 @code{target remote} command.
17414 The @code{stdio} connection is useful when starting @code{gdbserver}
17418 (gdb) target remote | ssh -T hostname gdbserver - hello
17421 The @samp{-T} option to ssh is provided because we don't need a remote pty,
17422 and we don't want escape-character handling. Ssh does this by default when
17423 a command is provided, the flag is provided to make it explicit.
17424 You could elide it if you want to.
17426 Programs started with stdio-connected gdbserver have @file{/dev/null} for
17427 @code{stdin}, and @code{stdout},@code{stderr} are sent back to gdb for
17428 display through a pipe connected to gdbserver.
17429 Both @code{stdout} and @code{stderr} use the same pipe.
17431 @subsubsection Attaching to a Running Program
17432 @cindex attach to a program, @code{gdbserver}
17433 @cindex @option{--attach}, @code{gdbserver} option
17435 On some targets, @code{gdbserver} can also attach to running programs.
17436 This is accomplished via the @code{--attach} argument. The syntax is:
17439 target> gdbserver --attach @var{comm} @var{pid}
17442 @var{pid} is the process ID of a currently running process. It isn't necessary
17443 to point @code{gdbserver} at a binary for the running process.
17446 You can debug processes by name instead of process ID if your target has the
17447 @code{pidof} utility:
17450 target> gdbserver --attach @var{comm} `pidof @var{program}`
17453 In case more than one copy of @var{program} is running, or @var{program}
17454 has multiple threads, most versions of @code{pidof} support the
17455 @code{-s} option to only return the first process ID.
17457 @subsubsection Multi-Process Mode for @code{gdbserver}
17458 @cindex @code{gdbserver}, multiple processes
17459 @cindex multiple processes with @code{gdbserver}
17461 When you connect to @code{gdbserver} using @code{target remote},
17462 @code{gdbserver} debugs the specified program only once. When the
17463 program exits, or you detach from it, @value{GDBN} closes the connection
17464 and @code{gdbserver} exits.
17466 If you connect using @kbd{target extended-remote}, @code{gdbserver}
17467 enters multi-process mode. When the debugged program exits, or you
17468 detach from it, @value{GDBN} stays connected to @code{gdbserver} even
17469 though no program is running. The @code{run} and @code{attach}
17470 commands instruct @code{gdbserver} to run or attach to a new program.
17471 The @code{run} command uses @code{set remote exec-file} (@pxref{set
17472 remote exec-file}) to select the program to run. Command line
17473 arguments are supported, except for wildcard expansion and I/O
17474 redirection (@pxref{Arguments}).
17476 @cindex @option{--multi}, @code{gdbserver} option
17477 To start @code{gdbserver} without supplying an initial command to run
17478 or process ID to attach, use the @option{--multi} command line option.
17479 Then you can connect using @kbd{target extended-remote} and start
17480 the program you want to debug.
17482 In multi-process mode @code{gdbserver} does not automatically exit unless you
17483 use the option @option{--once}. You can terminate it by using
17484 @code{monitor exit} (@pxref{Monitor Commands for gdbserver}). Note that the
17485 conditions under which @code{gdbserver} terminates depend on how @value{GDBN}
17486 connects to it (@kbd{target remote} or @kbd{target extended-remote}). The
17487 @option{--multi} option to @code{gdbserver} has no influence on that.
17489 @subsubsection TCP port allocation lifecycle of @code{gdbserver}
17491 This section applies only when @code{gdbserver} is run to listen on a TCP port.
17493 @code{gdbserver} normally terminates after all of its debugged processes have
17494 terminated in @kbd{target remote} mode. On the other hand, for @kbd{target
17495 extended-remote}, @code{gdbserver} stays running even with no processes left.
17496 @value{GDBN} normally terminates the spawned debugged process on its exit,
17497 which normally also terminates @code{gdbserver} in the @kbd{target remote}
17498 mode. Therefore, when the connection drops unexpectedly, and @value{GDBN}
17499 cannot ask @code{gdbserver} to kill its debugged processes, @code{gdbserver}
17500 stays running even in the @kbd{target remote} mode.
17502 When @code{gdbserver} stays running, @value{GDBN} can connect to it again later.
17503 Such reconnecting is useful for features like @ref{disconnected tracing}. For
17504 completeness, at most one @value{GDBN} can be connected at a time.
17506 @cindex @option{--once}, @code{gdbserver} option
17507 By default, @code{gdbserver} keeps the listening TCP port open, so that
17508 additional connections are possible. However, if you start @code{gdbserver}
17509 with the @option{--once} option, it will stop listening for any further
17510 connection attempts after connecting to the first @value{GDBN} session. This
17511 means no further connections to @code{gdbserver} will be possible after the
17512 first one. It also means @code{gdbserver} will terminate after the first
17513 connection with remote @value{GDBN} has closed, even for unexpectedly closed
17514 connections and even in the @kbd{target extended-remote} mode. The
17515 @option{--once} option allows reusing the same port number for connecting to
17516 multiple instances of @code{gdbserver} running on the same host, since each
17517 instance closes its port after the first connection.
17519 @subsubsection Other Command-Line Arguments for @code{gdbserver}
17521 @cindex @option{--debug}, @code{gdbserver} option
17522 The @option{--debug} option tells @code{gdbserver} to display extra
17523 status information about the debugging process.
17524 @cindex @option{--remote-debug}, @code{gdbserver} option
17525 The @option{--remote-debug} option tells @code{gdbserver} to display
17526 remote protocol debug output. These options are intended for
17527 @code{gdbserver} development and for bug reports to the developers.
17529 @cindex @option{--wrapper}, @code{gdbserver} option
17530 The @option{--wrapper} option specifies a wrapper to launch programs
17531 for debugging. The option should be followed by the name of the
17532 wrapper, then any command-line arguments to pass to the wrapper, then
17533 @kbd{--} indicating the end of the wrapper arguments.
17535 @code{gdbserver} runs the specified wrapper program with a combined
17536 command line including the wrapper arguments, then the name of the
17537 program to debug, then any arguments to the program. The wrapper
17538 runs until it executes your program, and then @value{GDBN} gains control.
17540 You can use any program that eventually calls @code{execve} with
17541 its arguments as a wrapper. Several standard Unix utilities do
17542 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
17543 with @code{exec "$@@"} will also work.
17545 For example, you can use @code{env} to pass an environment variable to
17546 the debugged program, without setting the variable in @code{gdbserver}'s
17550 $ gdbserver --wrapper env LD_PRELOAD=libtest.so -- :2222 ./testprog
17553 @subsection Connecting to @code{gdbserver}
17555 Run @value{GDBN} on the host system.
17557 First make sure you have the necessary symbol files. Load symbols for
17558 your application using the @code{file} command before you connect. Use
17559 @code{set sysroot} to locate target libraries (unless your @value{GDBN}
17560 was compiled with the correct sysroot using @code{--with-sysroot}).
17562 The symbol file and target libraries must exactly match the executable
17563 and libraries on the target, with one exception: the files on the host
17564 system should not be stripped, even if the files on the target system
17565 are. Mismatched or missing files will lead to confusing results
17566 during debugging. On @sc{gnu}/Linux targets, mismatched or missing
17567 files may also prevent @code{gdbserver} from debugging multi-threaded
17570 Connect to your target (@pxref{Connecting,,Connecting to a Remote Target}).
17571 For TCP connections, you must start up @code{gdbserver} prior to using
17572 the @code{target remote} command. Otherwise you may get an error whose
17573 text depends on the host system, but which usually looks something like
17574 @samp{Connection refused}. Don't use the @code{load}
17575 command in @value{GDBN} when using @code{gdbserver}, since the program is
17576 already on the target.
17578 @subsection Monitor Commands for @code{gdbserver}
17579 @cindex monitor commands, for @code{gdbserver}
17580 @anchor{Monitor Commands for gdbserver}
17582 During a @value{GDBN} session using @code{gdbserver}, you can use the
17583 @code{monitor} command to send special requests to @code{gdbserver}.
17584 Here are the available commands.
17588 List the available monitor commands.
17590 @item monitor set debug 0
17591 @itemx monitor set debug 1
17592 Disable or enable general debugging messages.
17594 @item monitor set remote-debug 0
17595 @itemx monitor set remote-debug 1
17596 Disable or enable specific debugging messages associated with the remote
17597 protocol (@pxref{Remote Protocol}).
17599 @item monitor set libthread-db-search-path [PATH]
17600 @cindex gdbserver, search path for @code{libthread_db}
17601 When this command is issued, @var{path} is a colon-separated list of
17602 directories to search for @code{libthread_db} (@pxref{Threads,,set
17603 libthread-db-search-path}). If you omit @var{path},
17604 @samp{libthread-db-search-path} will be reset to its default value.
17606 The special entry @samp{$pdir} for @samp{libthread-db-search-path} is
17607 not supported in @code{gdbserver}.
17610 Tell gdbserver to exit immediately. This command should be followed by
17611 @code{disconnect} to close the debugging session. @code{gdbserver} will
17612 detach from any attached processes and kill any processes it created.
17613 Use @code{monitor exit} to terminate @code{gdbserver} at the end
17614 of a multi-process mode debug session.
17618 @subsection Tracepoints support in @code{gdbserver}
17619 @cindex tracepoints support in @code{gdbserver}
17621 On some targets, @code{gdbserver} supports tracepoints, fast
17622 tracepoints and static tracepoints.
17624 For fast or static tracepoints to work, a special library called the
17625 @dfn{in-process agent} (IPA), must be loaded in the inferior process.
17626 This library is built and distributed as an integral part of
17627 @code{gdbserver}. In addition, support for static tracepoints
17628 requires building the in-process agent library with static tracepoints
17629 support. At present, the UST (LTTng Userspace Tracer,
17630 @url{http://lttng.org/ust}) tracing engine is supported. This support
17631 is automatically available if UST development headers are found in the
17632 standard include path when @code{gdbserver} is built, or if
17633 @code{gdbserver} was explicitly configured using @option{--with-ust}
17634 to point at such headers. You can explicitly disable the support
17635 using @option{--with-ust=no}.
17637 There are several ways to load the in-process agent in your program:
17640 @item Specifying it as dependency at link time
17642 You can link your program dynamically with the in-process agent
17643 library. On most systems, this is accomplished by adding
17644 @code{-linproctrace} to the link command.
17646 @item Using the system's preloading mechanisms
17648 You can force loading the in-process agent at startup time by using
17649 your system's support for preloading shared libraries. Many Unixes
17650 support the concept of preloading user defined libraries. In most
17651 cases, you do that by specifying @code{LD_PRELOAD=libinproctrace.so}
17652 in the environment. See also the description of @code{gdbserver}'s
17653 @option{--wrapper} command line option.
17655 @item Using @value{GDBN} to force loading the agent at run time
17657 On some systems, you can force the inferior to load a shared library,
17658 by calling a dynamic loader function in the inferior that takes care
17659 of dynamically looking up and loading a shared library. On most Unix
17660 systems, the function is @code{dlopen}. You'll use the @code{call}
17661 command for that. For example:
17664 (@value{GDBP}) call dlopen ("libinproctrace.so", ...)
17667 Note that on most Unix systems, for the @code{dlopen} function to be
17668 available, the program needs to be linked with @code{-ldl}.
17671 On systems that have a userspace dynamic loader, like most Unix
17672 systems, when you connect to @code{gdbserver} using @code{target
17673 remote}, you'll find that the program is stopped at the dynamic
17674 loader's entry point, and no shared library has been loaded in the
17675 program's address space yet, including the in-process agent. In that
17676 case, before being able to use any of the fast or static tracepoints
17677 features, you need to let the loader run and load the shared
17678 libraries. The simplest way to do that is to run the program to the
17679 main procedure. E.g., if debugging a C or C@t{++} program, start
17680 @code{gdbserver} like so:
17683 $ gdbserver :9999 myprogram
17686 Start GDB and connect to @code{gdbserver} like so, and run to main:
17690 (@value{GDBP}) target remote myhost:9999
17691 0x00007f215893ba60 in ?? () from /lib64/ld-linux-x86-64.so.2
17692 (@value{GDBP}) b main
17693 (@value{GDBP}) continue
17696 The in-process tracing agent library should now be loaded into the
17697 process; you can confirm it with the @code{info sharedlibrary}
17698 command, which will list @file{libinproctrace.so} as loaded in the
17699 process. You are now ready to install fast tracepoints, list static
17700 tracepoint markers, probe static tracepoints markers, and start
17703 @node Remote Configuration
17704 @section Remote Configuration
17707 @kindex show remote
17708 This section documents the configuration options available when
17709 debugging remote programs. For the options related to the File I/O
17710 extensions of the remote protocol, see @ref{system,
17711 system-call-allowed}.
17714 @item set remoteaddresssize @var{bits}
17715 @cindex address size for remote targets
17716 @cindex bits in remote address
17717 Set the maximum size of address in a memory packet to the specified
17718 number of bits. @value{GDBN} will mask off the address bits above
17719 that number, when it passes addresses to the remote target. The
17720 default value is the number of bits in the target's address.
17722 @item show remoteaddresssize
17723 Show the current value of remote address size in bits.
17725 @item set remotebaud @var{n}
17726 @cindex baud rate for remote targets
17727 Set the baud rate for the remote serial I/O to @var{n} baud. The
17728 value is used to set the speed of the serial port used for debugging
17731 @item show remotebaud
17732 Show the current speed of the remote connection.
17734 @item set remotebreak
17735 @cindex interrupt remote programs
17736 @cindex BREAK signal instead of Ctrl-C
17737 @anchor{set remotebreak}
17738 If set to on, @value{GDBN} sends a @code{BREAK} signal to the remote
17739 when you type @kbd{Ctrl-c} to interrupt the program running
17740 on the remote. If set to off, @value{GDBN} sends the @samp{Ctrl-C}
17741 character instead. The default is off, since most remote systems
17742 expect to see @samp{Ctrl-C} as the interrupt signal.
17744 @item show remotebreak
17745 Show whether @value{GDBN} sends @code{BREAK} or @samp{Ctrl-C} to
17746 interrupt the remote program.
17748 @item set remoteflow on
17749 @itemx set remoteflow off
17750 @kindex set remoteflow
17751 Enable or disable hardware flow control (@code{RTS}/@code{CTS})
17752 on the serial port used to communicate to the remote target.
17754 @item show remoteflow
17755 @kindex show remoteflow
17756 Show the current setting of hardware flow control.
17758 @item set remotelogbase @var{base}
17759 Set the base (a.k.a.@: radix) of logging serial protocol
17760 communications to @var{base}. Supported values of @var{base} are:
17761 @code{ascii}, @code{octal}, and @code{hex}. The default is
17764 @item show remotelogbase
17765 Show the current setting of the radix for logging remote serial
17768 @item set remotelogfile @var{file}
17769 @cindex record serial communications on file
17770 Record remote serial communications on the named @var{file}. The
17771 default is not to record at all.
17773 @item show remotelogfile.
17774 Show the current setting of the file name on which to record the
17775 serial communications.
17777 @item set remotetimeout @var{num}
17778 @cindex timeout for serial communications
17779 @cindex remote timeout
17780 Set the timeout limit to wait for the remote target to respond to
17781 @var{num} seconds. The default is 2 seconds.
17783 @item show remotetimeout
17784 Show the current number of seconds to wait for the remote target
17787 @cindex limit hardware breakpoints and watchpoints
17788 @cindex remote target, limit break- and watchpoints
17789 @anchor{set remote hardware-watchpoint-limit}
17790 @anchor{set remote hardware-breakpoint-limit}
17791 @item set remote hardware-watchpoint-limit @var{limit}
17792 @itemx set remote hardware-breakpoint-limit @var{limit}
17793 Restrict @value{GDBN} to using @var{limit} remote hardware breakpoint or
17794 watchpoints. A limit of -1, the default, is treated as unlimited.
17796 @cindex limit hardware watchpoints length
17797 @cindex remote target, limit watchpoints length
17798 @anchor{set remote hardware-watchpoint-length-limit}
17799 @item set remote hardware-watchpoint-length-limit @var{limit}
17800 Restrict @value{GDBN} to using @var{limit} bytes for the maximum length of
17801 a remote hardware watchpoint. A limit of -1, the default, is treated
17804 @item show remote hardware-watchpoint-length-limit
17805 Show the current limit (in bytes) of the maximum length of
17806 a remote hardware watchpoint.
17808 @item set remote exec-file @var{filename}
17809 @itemx show remote exec-file
17810 @anchor{set remote exec-file}
17811 @cindex executable file, for remote target
17812 Select the file used for @code{run} with @code{target
17813 extended-remote}. This should be set to a filename valid on the
17814 target system. If it is not set, the target will use a default
17815 filename (e.g.@: the last program run).
17817 @item set remote interrupt-sequence
17818 @cindex interrupt remote programs
17819 @cindex select Ctrl-C, BREAK or BREAK-g
17820 Allow the user to select one of @samp{Ctrl-C}, a @code{BREAK} or
17821 @samp{BREAK-g} as the
17822 sequence to the remote target in order to interrupt the execution.
17823 @samp{Ctrl-C} is a default. Some system prefers @code{BREAK} which
17824 is high level of serial line for some certain time.
17825 Linux kernel prefers @samp{BREAK-g}, a.k.a Magic SysRq g.
17826 It is @code{BREAK} signal followed by character @code{g}.
17828 @item show interrupt-sequence
17829 Show which of @samp{Ctrl-C}, @code{BREAK} or @code{BREAK-g}
17830 is sent by @value{GDBN} to interrupt the remote program.
17831 @code{BREAK-g} is BREAK signal followed by @code{g} and
17832 also known as Magic SysRq g.
17834 @item set remote interrupt-on-connect
17835 @cindex send interrupt-sequence on start
17836 Specify whether interrupt-sequence is sent to remote target when
17837 @value{GDBN} connects to it. This is mostly needed when you debug
17838 Linux kernel. Linux kernel expects @code{BREAK} followed by @code{g}
17839 which is known as Magic SysRq g in order to connect @value{GDBN}.
17841 @item show interrupt-on-connect
17842 Show whether interrupt-sequence is sent
17843 to remote target when @value{GDBN} connects to it.
17847 @item set tcp auto-retry on
17848 @cindex auto-retry, for remote TCP target
17849 Enable auto-retry for remote TCP connections. This is useful if the remote
17850 debugging agent is launched in parallel with @value{GDBN}; there is a race
17851 condition because the agent may not become ready to accept the connection
17852 before @value{GDBN} attempts to connect. When auto-retry is
17853 enabled, if the initial attempt to connect fails, @value{GDBN} reattempts
17854 to establish the connection using the timeout specified by
17855 @code{set tcp connect-timeout}.
17857 @item set tcp auto-retry off
17858 Do not auto-retry failed TCP connections.
17860 @item show tcp auto-retry
17861 Show the current auto-retry setting.
17863 @item set tcp connect-timeout @var{seconds}
17864 @cindex connection timeout, for remote TCP target
17865 @cindex timeout, for remote target connection
17866 Set the timeout for establishing a TCP connection to the remote target to
17867 @var{seconds}. The timeout affects both polling to retry failed connections
17868 (enabled by @code{set tcp auto-retry on}) and waiting for connections
17869 that are merely slow to complete, and represents an approximate cumulative
17872 @item show tcp connect-timeout
17873 Show the current connection timeout setting.
17876 @cindex remote packets, enabling and disabling
17877 The @value{GDBN} remote protocol autodetects the packets supported by
17878 your debugging stub. If you need to override the autodetection, you
17879 can use these commands to enable or disable individual packets. Each
17880 packet can be set to @samp{on} (the remote target supports this
17881 packet), @samp{off} (the remote target does not support this packet),
17882 or @samp{auto} (detect remote target support for this packet). They
17883 all default to @samp{auto}. For more information about each packet,
17884 see @ref{Remote Protocol}.
17886 During normal use, you should not have to use any of these commands.
17887 If you do, that may be a bug in your remote debugging stub, or a bug
17888 in @value{GDBN}. You may want to report the problem to the
17889 @value{GDBN} developers.
17891 For each packet @var{name}, the command to enable or disable the
17892 packet is @code{set remote @var{name}-packet}. The available settings
17895 @multitable @columnfractions 0.28 0.32 0.25
17898 @tab Related Features
17900 @item @code{fetch-register}
17902 @tab @code{info registers}
17904 @item @code{set-register}
17908 @item @code{binary-download}
17910 @tab @code{load}, @code{set}
17912 @item @code{read-aux-vector}
17913 @tab @code{qXfer:auxv:read}
17914 @tab @code{info auxv}
17916 @item @code{symbol-lookup}
17917 @tab @code{qSymbol}
17918 @tab Detecting multiple threads
17920 @item @code{attach}
17921 @tab @code{vAttach}
17924 @item @code{verbose-resume}
17926 @tab Stepping or resuming multiple threads
17932 @item @code{software-breakpoint}
17936 @item @code{hardware-breakpoint}
17940 @item @code{write-watchpoint}
17944 @item @code{read-watchpoint}
17948 @item @code{access-watchpoint}
17952 @item @code{target-features}
17953 @tab @code{qXfer:features:read}
17954 @tab @code{set architecture}
17956 @item @code{library-info}
17957 @tab @code{qXfer:libraries:read}
17958 @tab @code{info sharedlibrary}
17960 @item @code{memory-map}
17961 @tab @code{qXfer:memory-map:read}
17962 @tab @code{info mem}
17964 @item @code{read-sdata-object}
17965 @tab @code{qXfer:sdata:read}
17966 @tab @code{print $_sdata}
17968 @item @code{read-spu-object}
17969 @tab @code{qXfer:spu:read}
17970 @tab @code{info spu}
17972 @item @code{write-spu-object}
17973 @tab @code{qXfer:spu:write}
17974 @tab @code{info spu}
17976 @item @code{read-siginfo-object}
17977 @tab @code{qXfer:siginfo:read}
17978 @tab @code{print $_siginfo}
17980 @item @code{write-siginfo-object}
17981 @tab @code{qXfer:siginfo:write}
17982 @tab @code{set $_siginfo}
17984 @item @code{threads}
17985 @tab @code{qXfer:threads:read}
17986 @tab @code{info threads}
17988 @item @code{get-thread-local-@*storage-address}
17989 @tab @code{qGetTLSAddr}
17990 @tab Displaying @code{__thread} variables
17992 @item @code{get-thread-information-block-address}
17993 @tab @code{qGetTIBAddr}
17994 @tab Display MS-Windows Thread Information Block.
17996 @item @code{search-memory}
17997 @tab @code{qSearch:memory}
18000 @item @code{supported-packets}
18001 @tab @code{qSupported}
18002 @tab Remote communications parameters
18004 @item @code{pass-signals}
18005 @tab @code{QPassSignals}
18006 @tab @code{handle @var{signal}}
18008 @item @code{program-signals}
18009 @tab @code{QProgramSignals}
18010 @tab @code{handle @var{signal}}
18012 @item @code{hostio-close-packet}
18013 @tab @code{vFile:close}
18014 @tab @code{remote get}, @code{remote put}
18016 @item @code{hostio-open-packet}
18017 @tab @code{vFile:open}
18018 @tab @code{remote get}, @code{remote put}
18020 @item @code{hostio-pread-packet}
18021 @tab @code{vFile:pread}
18022 @tab @code{remote get}, @code{remote put}
18024 @item @code{hostio-pwrite-packet}
18025 @tab @code{vFile:pwrite}
18026 @tab @code{remote get}, @code{remote put}
18028 @item @code{hostio-unlink-packet}
18029 @tab @code{vFile:unlink}
18030 @tab @code{remote delete}
18032 @item @code{hostio-readlink-packet}
18033 @tab @code{vFile:readlink}
18036 @item @code{noack-packet}
18037 @tab @code{QStartNoAckMode}
18038 @tab Packet acknowledgment
18040 @item @code{osdata}
18041 @tab @code{qXfer:osdata:read}
18042 @tab @code{info os}
18044 @item @code{query-attached}
18045 @tab @code{qAttached}
18046 @tab Querying remote process attach state.
18048 @item @code{traceframe-info}
18049 @tab @code{qXfer:traceframe-info:read}
18050 @tab Traceframe info
18052 @item @code{install-in-trace}
18053 @tab @code{InstallInTrace}
18054 @tab Install tracepoint in tracing
18056 @item @code{disable-randomization}
18057 @tab @code{QDisableRandomization}
18058 @tab @code{set disable-randomization}
18060 @item @code{conditional-breakpoints-packet}
18061 @tab @code{Z0 and Z1}
18062 @tab @code{Support for target-side breakpoint condition evaluation}
18066 @section Implementing a Remote Stub
18068 @cindex debugging stub, example
18069 @cindex remote stub, example
18070 @cindex stub example, remote debugging
18071 The stub files provided with @value{GDBN} implement the target side of the
18072 communication protocol, and the @value{GDBN} side is implemented in the
18073 @value{GDBN} source file @file{remote.c}. Normally, you can simply allow
18074 these subroutines to communicate, and ignore the details. (If you're
18075 implementing your own stub file, you can still ignore the details: start
18076 with one of the existing stub files. @file{sparc-stub.c} is the best
18077 organized, and therefore the easiest to read.)
18079 @cindex remote serial debugging, overview
18080 To debug a program running on another machine (the debugging
18081 @dfn{target} machine), you must first arrange for all the usual
18082 prerequisites for the program to run by itself. For example, for a C
18087 A startup routine to set up the C runtime environment; these usually
18088 have a name like @file{crt0}. The startup routine may be supplied by
18089 your hardware supplier, or you may have to write your own.
18092 A C subroutine library to support your program's
18093 subroutine calls, notably managing input and output.
18096 A way of getting your program to the other machine---for example, a
18097 download program. These are often supplied by the hardware
18098 manufacturer, but you may have to write your own from hardware
18102 The next step is to arrange for your program to use a serial port to
18103 communicate with the machine where @value{GDBN} is running (the @dfn{host}
18104 machine). In general terms, the scheme looks like this:
18108 @value{GDBN} already understands how to use this protocol; when everything
18109 else is set up, you can simply use the @samp{target remote} command
18110 (@pxref{Targets,,Specifying a Debugging Target}).
18112 @item On the target,
18113 you must link with your program a few special-purpose subroutines that
18114 implement the @value{GDBN} remote serial protocol. The file containing these
18115 subroutines is called a @dfn{debugging stub}.
18117 On certain remote targets, you can use an auxiliary program
18118 @code{gdbserver} instead of linking a stub into your program.
18119 @xref{Server,,Using the @code{gdbserver} Program}, for details.
18122 The debugging stub is specific to the architecture of the remote
18123 machine; for example, use @file{sparc-stub.c} to debug programs on
18126 @cindex remote serial stub list
18127 These working remote stubs are distributed with @value{GDBN}:
18132 @cindex @file{i386-stub.c}
18135 For Intel 386 and compatible architectures.
18138 @cindex @file{m68k-stub.c}
18139 @cindex Motorola 680x0
18141 For Motorola 680x0 architectures.
18144 @cindex @file{sh-stub.c}
18147 For Renesas SH architectures.
18150 @cindex @file{sparc-stub.c}
18152 For @sc{sparc} architectures.
18154 @item sparcl-stub.c
18155 @cindex @file{sparcl-stub.c}
18158 For Fujitsu @sc{sparclite} architectures.
18162 The @file{README} file in the @value{GDBN} distribution may list other
18163 recently added stubs.
18166 * Stub Contents:: What the stub can do for you
18167 * Bootstrapping:: What you must do for the stub
18168 * Debug Session:: Putting it all together
18171 @node Stub Contents
18172 @subsection What the Stub Can Do for You
18174 @cindex remote serial stub
18175 The debugging stub for your architecture supplies these three
18179 @item set_debug_traps
18180 @findex set_debug_traps
18181 @cindex remote serial stub, initialization
18182 This routine arranges for @code{handle_exception} to run when your
18183 program stops. You must call this subroutine explicitly in your
18184 program's startup code.
18186 @item handle_exception
18187 @findex handle_exception
18188 @cindex remote serial stub, main routine
18189 This is the central workhorse, but your program never calls it
18190 explicitly---the setup code arranges for @code{handle_exception} to
18191 run when a trap is triggered.
18193 @code{handle_exception} takes control when your program stops during
18194 execution (for example, on a breakpoint), and mediates communications
18195 with @value{GDBN} on the host machine. This is where the communications
18196 protocol is implemented; @code{handle_exception} acts as the @value{GDBN}
18197 representative on the target machine. It begins by sending summary
18198 information on the state of your program, then continues to execute,
18199 retrieving and transmitting any information @value{GDBN} needs, until you
18200 execute a @value{GDBN} command that makes your program resume; at that point,
18201 @code{handle_exception} returns control to your own code on the target
18205 @cindex @code{breakpoint} subroutine, remote
18206 Use this auxiliary subroutine to make your program contain a
18207 breakpoint. Depending on the particular situation, this may be the only
18208 way for @value{GDBN} to get control. For instance, if your target
18209 machine has some sort of interrupt button, you won't need to call this;
18210 pressing the interrupt button transfers control to
18211 @code{handle_exception}---in effect, to @value{GDBN}. On some machines,
18212 simply receiving characters on the serial port may also trigger a trap;
18213 again, in that situation, you don't need to call @code{breakpoint} from
18214 your own program---simply running @samp{target remote} from the host
18215 @value{GDBN} session gets control.
18217 Call @code{breakpoint} if none of these is true, or if you simply want
18218 to make certain your program stops at a predetermined point for the
18219 start of your debugging session.
18222 @node Bootstrapping
18223 @subsection What You Must Do for the Stub
18225 @cindex remote stub, support routines
18226 The debugging stubs that come with @value{GDBN} are set up for a particular
18227 chip architecture, but they have no information about the rest of your
18228 debugging target machine.
18230 First of all you need to tell the stub how to communicate with the
18234 @item int getDebugChar()
18235 @findex getDebugChar
18236 Write this subroutine to read a single character from the serial port.
18237 It may be identical to @code{getchar} for your target system; a
18238 different name is used to allow you to distinguish the two if you wish.
18240 @item void putDebugChar(int)
18241 @findex putDebugChar
18242 Write this subroutine to write a single character to the serial port.
18243 It may be identical to @code{putchar} for your target system; a
18244 different name is used to allow you to distinguish the two if you wish.
18247 @cindex control C, and remote debugging
18248 @cindex interrupting remote targets
18249 If you want @value{GDBN} to be able to stop your program while it is
18250 running, you need to use an interrupt-driven serial driver, and arrange
18251 for it to stop when it receives a @code{^C} (@samp{\003}, the control-C
18252 character). That is the character which @value{GDBN} uses to tell the
18253 remote system to stop.
18255 Getting the debugging target to return the proper status to @value{GDBN}
18256 probably requires changes to the standard stub; one quick and dirty way
18257 is to just execute a breakpoint instruction (the ``dirty'' part is that
18258 @value{GDBN} reports a @code{SIGTRAP} instead of a @code{SIGINT}).
18260 Other routines you need to supply are:
18263 @item void exceptionHandler (int @var{exception_number}, void *@var{exception_address})
18264 @findex exceptionHandler
18265 Write this function to install @var{exception_address} in the exception
18266 handling tables. You need to do this because the stub does not have any
18267 way of knowing what the exception handling tables on your target system
18268 are like (for example, the processor's table might be in @sc{rom},
18269 containing entries which point to a table in @sc{ram}).
18270 @var{exception_number} is the exception number which should be changed;
18271 its meaning is architecture-dependent (for example, different numbers
18272 might represent divide by zero, misaligned access, etc). When this
18273 exception occurs, control should be transferred directly to
18274 @var{exception_address}, and the processor state (stack, registers,
18275 and so on) should be just as it is when a processor exception occurs. So if
18276 you want to use a jump instruction to reach @var{exception_address}, it
18277 should be a simple jump, not a jump to subroutine.
18279 For the 386, @var{exception_address} should be installed as an interrupt
18280 gate so that interrupts are masked while the handler runs. The gate
18281 should be at privilege level 0 (the most privileged level). The
18282 @sc{sparc} and 68k stubs are able to mask interrupts themselves without
18283 help from @code{exceptionHandler}.
18285 @item void flush_i_cache()
18286 @findex flush_i_cache
18287 On @sc{sparc} and @sc{sparclite} only, write this subroutine to flush the
18288 instruction cache, if any, on your target machine. If there is no
18289 instruction cache, this subroutine may be a no-op.
18291 On target machines that have instruction caches, @value{GDBN} requires this
18292 function to make certain that the state of your program is stable.
18296 You must also make sure this library routine is available:
18299 @item void *memset(void *, int, int)
18301 This is the standard library function @code{memset} that sets an area of
18302 memory to a known value. If you have one of the free versions of
18303 @code{libc.a}, @code{memset} can be found there; otherwise, you must
18304 either obtain it from your hardware manufacturer, or write your own.
18307 If you do not use the GNU C compiler, you may need other standard
18308 library subroutines as well; this varies from one stub to another,
18309 but in general the stubs are likely to use any of the common library
18310 subroutines which @code{@value{NGCC}} generates as inline code.
18313 @node Debug Session
18314 @subsection Putting it All Together
18316 @cindex remote serial debugging summary
18317 In summary, when your program is ready to debug, you must follow these
18322 Make sure you have defined the supporting low-level routines
18323 (@pxref{Bootstrapping,,What You Must Do for the Stub}):
18325 @code{getDebugChar}, @code{putDebugChar},
18326 @code{flush_i_cache}, @code{memset}, @code{exceptionHandler}.
18330 Insert these lines in your program's startup code, before the main
18331 procedure is called:
18338 On some machines, when a breakpoint trap is raised, the hardware
18339 automatically makes the PC point to the instruction after the
18340 breakpoint. If your machine doesn't do that, you may need to adjust
18341 @code{handle_exception} to arrange for it to return to the instruction
18342 after the breakpoint on this first invocation, so that your program
18343 doesn't keep hitting the initial breakpoint instead of making
18347 For the 680x0 stub only, you need to provide a variable called
18348 @code{exceptionHook}. Normally you just use:
18351 void (*exceptionHook)() = 0;
18355 but if before calling @code{set_debug_traps}, you set it to point to a
18356 function in your program, that function is called when
18357 @code{@value{GDBN}} continues after stopping on a trap (for example, bus
18358 error). The function indicated by @code{exceptionHook} is called with
18359 one parameter: an @code{int} which is the exception number.
18362 Compile and link together: your program, the @value{GDBN} debugging stub for
18363 your target architecture, and the supporting subroutines.
18366 Make sure you have a serial connection between your target machine and
18367 the @value{GDBN} host, and identify the serial port on the host.
18370 @c The "remote" target now provides a `load' command, so we should
18371 @c document that. FIXME.
18372 Download your program to your target machine (or get it there by
18373 whatever means the manufacturer provides), and start it.
18376 Start @value{GDBN} on the host, and connect to the target
18377 (@pxref{Connecting,,Connecting to a Remote Target}).
18381 @node Configurations
18382 @chapter Configuration-Specific Information
18384 While nearly all @value{GDBN} commands are available for all native and
18385 cross versions of the debugger, there are some exceptions. This chapter
18386 describes things that are only available in certain configurations.
18388 There are three major categories of configurations: native
18389 configurations, where the host and target are the same, embedded
18390 operating system configurations, which are usually the same for several
18391 different processor architectures, and bare embedded processors, which
18392 are quite different from each other.
18397 * Embedded Processors::
18404 This section describes details specific to particular native
18409 * BSD libkvm Interface:: Debugging BSD kernel memory images
18410 * SVR4 Process Information:: SVR4 process information
18411 * DJGPP Native:: Features specific to the DJGPP port
18412 * Cygwin Native:: Features specific to the Cygwin port
18413 * Hurd Native:: Features specific to @sc{gnu} Hurd
18414 * Neutrino:: Features specific to QNX Neutrino
18415 * Darwin:: Features specific to Darwin
18421 On HP-UX systems, if you refer to a function or variable name that
18422 begins with a dollar sign, @value{GDBN} searches for a user or system
18423 name first, before it searches for a convenience variable.
18426 @node BSD libkvm Interface
18427 @subsection BSD libkvm Interface
18430 @cindex kernel memory image
18431 @cindex kernel crash dump
18433 BSD-derived systems (FreeBSD/NetBSD/OpenBSD) have a kernel memory
18434 interface that provides a uniform interface for accessing kernel virtual
18435 memory images, including live systems and crash dumps. @value{GDBN}
18436 uses this interface to allow you to debug live kernels and kernel crash
18437 dumps on many native BSD configurations. This is implemented as a
18438 special @code{kvm} debugging target. For debugging a live system, load
18439 the currently running kernel into @value{GDBN} and connect to the
18443 (@value{GDBP}) @b{target kvm}
18446 For debugging crash dumps, provide the file name of the crash dump as an
18450 (@value{GDBP}) @b{target kvm /var/crash/bsd.0}
18453 Once connected to the @code{kvm} target, the following commands are
18459 Set current context from the @dfn{Process Control Block} (PCB) address.
18462 Set current context from proc address. This command isn't available on
18463 modern FreeBSD systems.
18466 @node SVR4 Process Information
18467 @subsection SVR4 Process Information
18469 @cindex examine process image
18470 @cindex process info via @file{/proc}
18472 Many versions of SVR4 and compatible systems provide a facility called
18473 @samp{/proc} that can be used to examine the image of a running
18474 process using file-system subroutines. If @value{GDBN} is configured
18475 for an operating system with this facility, the command @code{info
18476 proc} is available to report information about the process running
18477 your program, or about any process running on your system. @code{info
18478 proc} works only on SVR4 systems that include the @code{procfs} code.
18479 This includes, as of this writing, @sc{gnu}/Linux, OSF/1 (Digital
18480 Unix), Solaris, Irix, and Unixware, but not HP-UX, for example.
18486 @itemx info proc @var{process-id}
18487 Summarize available information about any running process. If a
18488 process ID is specified by @var{process-id}, display information about
18489 that process; otherwise display information about the program being
18490 debugged. The summary includes the debugged process ID, the command
18491 line used to invoke it, its current working directory, and its
18492 executable file's absolute file name.
18494 On some systems, @var{process-id} can be of the form
18495 @samp{[@var{pid}]/@var{tid}} which specifies a certain thread ID
18496 within a process. If the optional @var{pid} part is missing, it means
18497 a thread from the process being debugged (the leading @samp{/} still
18498 needs to be present, or else @value{GDBN} will interpret the number as
18499 a process ID rather than a thread ID).
18501 @item info proc mappings
18502 @cindex memory address space mappings
18503 Report the memory address space ranges accessible in the program, with
18504 information on whether the process has read, write, or execute access
18505 rights to each range. On @sc{gnu}/Linux systems, each memory range
18506 includes the object file which is mapped to that range, instead of the
18507 memory access rights to that range.
18509 @item info proc stat
18510 @itemx info proc status
18511 @cindex process detailed status information
18512 These subcommands are specific to @sc{gnu}/Linux systems. They show
18513 the process-related information, including the user ID and group ID;
18514 how many threads are there in the process; its virtual memory usage;
18515 the signals that are pending, blocked, and ignored; its TTY; its
18516 consumption of system and user time; its stack size; its @samp{nice}
18517 value; etc. For more information, see the @samp{proc} man page
18518 (type @kbd{man 5 proc} from your shell prompt).
18520 @item info proc all
18521 Show all the information about the process described under all of the
18522 above @code{info proc} subcommands.
18525 @comment These sub-options of 'info proc' were not included when
18526 @comment procfs.c was re-written. Keep their descriptions around
18527 @comment against the day when someone finds the time to put them back in.
18528 @kindex info proc times
18529 @item info proc times
18530 Starting time, user CPU time, and system CPU time for your program and
18533 @kindex info proc id
18535 Report on the process IDs related to your program: its own process ID,
18536 the ID of its parent, the process group ID, and the session ID.
18539 @item set procfs-trace
18540 @kindex set procfs-trace
18541 @cindex @code{procfs} API calls
18542 This command enables and disables tracing of @code{procfs} API calls.
18544 @item show procfs-trace
18545 @kindex show procfs-trace
18546 Show the current state of @code{procfs} API call tracing.
18548 @item set procfs-file @var{file}
18549 @kindex set procfs-file
18550 Tell @value{GDBN} to write @code{procfs} API trace to the named
18551 @var{file}. @value{GDBN} appends the trace info to the previous
18552 contents of the file. The default is to display the trace on the
18555 @item show procfs-file
18556 @kindex show procfs-file
18557 Show the file to which @code{procfs} API trace is written.
18559 @item proc-trace-entry
18560 @itemx proc-trace-exit
18561 @itemx proc-untrace-entry
18562 @itemx proc-untrace-exit
18563 @kindex proc-trace-entry
18564 @kindex proc-trace-exit
18565 @kindex proc-untrace-entry
18566 @kindex proc-untrace-exit
18567 These commands enable and disable tracing of entries into and exits
18568 from the @code{syscall} interface.
18571 @kindex info pidlist
18572 @cindex process list, QNX Neutrino
18573 For QNX Neutrino only, this command displays the list of all the
18574 processes and all the threads within each process.
18577 @kindex info meminfo
18578 @cindex mapinfo list, QNX Neutrino
18579 For QNX Neutrino only, this command displays the list of all mapinfos.
18583 @subsection Features for Debugging @sc{djgpp} Programs
18584 @cindex @sc{djgpp} debugging
18585 @cindex native @sc{djgpp} debugging
18586 @cindex MS-DOS-specific commands
18589 @sc{djgpp} is a port of the @sc{gnu} development tools to MS-DOS and
18590 MS-Windows. @sc{djgpp} programs are 32-bit protected-mode programs
18591 that use the @dfn{DPMI} (DOS Protected-Mode Interface) API to run on
18592 top of real-mode DOS systems and their emulations.
18594 @value{GDBN} supports native debugging of @sc{djgpp} programs, and
18595 defines a few commands specific to the @sc{djgpp} port. This
18596 subsection describes those commands.
18601 This is a prefix of @sc{djgpp}-specific commands which print
18602 information about the target system and important OS structures.
18605 @cindex MS-DOS system info
18606 @cindex free memory information (MS-DOS)
18607 @item info dos sysinfo
18608 This command displays assorted information about the underlying
18609 platform: the CPU type and features, the OS version and flavor, the
18610 DPMI version, and the available conventional and DPMI memory.
18615 @cindex segment descriptor tables
18616 @cindex descriptor tables display
18618 @itemx info dos ldt
18619 @itemx info dos idt
18620 These 3 commands display entries from, respectively, Global, Local,
18621 and Interrupt Descriptor Tables (GDT, LDT, and IDT). The descriptor
18622 tables are data structures which store a descriptor for each segment
18623 that is currently in use. The segment's selector is an index into a
18624 descriptor table; the table entry for that index holds the
18625 descriptor's base address and limit, and its attributes and access
18628 A typical @sc{djgpp} program uses 3 segments: a code segment, a data
18629 segment (used for both data and the stack), and a DOS segment (which
18630 allows access to DOS/BIOS data structures and absolute addresses in
18631 conventional memory). However, the DPMI host will usually define
18632 additional segments in order to support the DPMI environment.
18634 @cindex garbled pointers
18635 These commands allow to display entries from the descriptor tables.
18636 Without an argument, all entries from the specified table are
18637 displayed. An argument, which should be an integer expression, means
18638 display a single entry whose index is given by the argument. For
18639 example, here's a convenient way to display information about the
18640 debugged program's data segment:
18643 @exdent @code{(@value{GDBP}) info dos ldt $ds}
18644 @exdent @code{0x13f: base=0x11970000 limit=0x0009ffff 32-Bit Data (Read/Write, Exp-up)}
18648 This comes in handy when you want to see whether a pointer is outside
18649 the data segment's limit (i.e.@: @dfn{garbled}).
18651 @cindex page tables display (MS-DOS)
18653 @itemx info dos pte
18654 These two commands display entries from, respectively, the Page
18655 Directory and the Page Tables. Page Directories and Page Tables are
18656 data structures which control how virtual memory addresses are mapped
18657 into physical addresses. A Page Table includes an entry for every
18658 page of memory that is mapped into the program's address space; there
18659 may be several Page Tables, each one holding up to 4096 entries. A
18660 Page Directory has up to 4096 entries, one each for every Page Table
18661 that is currently in use.
18663 Without an argument, @kbd{info dos pde} displays the entire Page
18664 Directory, and @kbd{info dos pte} displays all the entries in all of
18665 the Page Tables. An argument, an integer expression, given to the
18666 @kbd{info dos pde} command means display only that entry from the Page
18667 Directory table. An argument given to the @kbd{info dos pte} command
18668 means display entries from a single Page Table, the one pointed to by
18669 the specified entry in the Page Directory.
18671 @cindex direct memory access (DMA) on MS-DOS
18672 These commands are useful when your program uses @dfn{DMA} (Direct
18673 Memory Access), which needs physical addresses to program the DMA
18676 These commands are supported only with some DPMI servers.
18678 @cindex physical address from linear address
18679 @item info dos address-pte @var{addr}
18680 This command displays the Page Table entry for a specified linear
18681 address. The argument @var{addr} is a linear address which should
18682 already have the appropriate segment's base address added to it,
18683 because this command accepts addresses which may belong to @emph{any}
18684 segment. For example, here's how to display the Page Table entry for
18685 the page where a variable @code{i} is stored:
18688 @exdent @code{(@value{GDBP}) info dos address-pte __djgpp_base_address + (char *)&i}
18689 @exdent @code{Page Table entry for address 0x11a00d30:}
18690 @exdent @code{Base=0x02698000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0xd30}
18694 This says that @code{i} is stored at offset @code{0xd30} from the page
18695 whose physical base address is @code{0x02698000}, and shows all the
18696 attributes of that page.
18698 Note that you must cast the addresses of variables to a @code{char *},
18699 since otherwise the value of @code{__djgpp_base_address}, the base
18700 address of all variables and functions in a @sc{djgpp} program, will
18701 be added using the rules of C pointer arithmetics: if @code{i} is
18702 declared an @code{int}, @value{GDBN} will add 4 times the value of
18703 @code{__djgpp_base_address} to the address of @code{i}.
18705 Here's another example, it displays the Page Table entry for the
18709 @exdent @code{(@value{GDBP}) info dos address-pte *((unsigned *)&_go32_info_block + 3)}
18710 @exdent @code{Page Table entry for address 0x29110:}
18711 @exdent @code{Base=0x00029000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0x110}
18715 (The @code{+ 3} offset is because the transfer buffer's address is the
18716 3rd member of the @code{_go32_info_block} structure.) The output
18717 clearly shows that this DPMI server maps the addresses in conventional
18718 memory 1:1, i.e.@: the physical (@code{0x00029000} + @code{0x110}) and
18719 linear (@code{0x29110}) addresses are identical.
18721 This command is supported only with some DPMI servers.
18724 @cindex DOS serial data link, remote debugging
18725 In addition to native debugging, the DJGPP port supports remote
18726 debugging via a serial data link. The following commands are specific
18727 to remote serial debugging in the DJGPP port of @value{GDBN}.
18730 @kindex set com1base
18731 @kindex set com1irq
18732 @kindex set com2base
18733 @kindex set com2irq
18734 @kindex set com3base
18735 @kindex set com3irq
18736 @kindex set com4base
18737 @kindex set com4irq
18738 @item set com1base @var{addr}
18739 This command sets the base I/O port address of the @file{COM1} serial
18742 @item set com1irq @var{irq}
18743 This command sets the @dfn{Interrupt Request} (@code{IRQ}) line to use
18744 for the @file{COM1} serial port.
18746 There are similar commands @samp{set com2base}, @samp{set com3irq},
18747 etc.@: for setting the port address and the @code{IRQ} lines for the
18750 @kindex show com1base
18751 @kindex show com1irq
18752 @kindex show com2base
18753 @kindex show com2irq
18754 @kindex show com3base
18755 @kindex show com3irq
18756 @kindex show com4base
18757 @kindex show com4irq
18758 The related commands @samp{show com1base}, @samp{show com1irq} etc.@:
18759 display the current settings of the base address and the @code{IRQ}
18760 lines used by the COM ports.
18763 @kindex info serial
18764 @cindex DOS serial port status
18765 This command prints the status of the 4 DOS serial ports. For each
18766 port, it prints whether it's active or not, its I/O base address and
18767 IRQ number, whether it uses a 16550-style FIFO, its baudrate, and the
18768 counts of various errors encountered so far.
18772 @node Cygwin Native
18773 @subsection Features for Debugging MS Windows PE Executables
18774 @cindex MS Windows debugging
18775 @cindex native Cygwin debugging
18776 @cindex Cygwin-specific commands
18778 @value{GDBN} supports native debugging of MS Windows programs, including
18779 DLLs with and without symbolic debugging information.
18781 @cindex Ctrl-BREAK, MS-Windows
18782 @cindex interrupt debuggee on MS-Windows
18783 MS-Windows programs that call @code{SetConsoleMode} to switch off the
18784 special meaning of the @samp{Ctrl-C} keystroke cannot be interrupted
18785 by typing @kbd{C-c}. For this reason, @value{GDBN} on MS-Windows
18786 supports @kbd{C-@key{BREAK}} as an alternative interrupt key
18787 sequence, which can be used to interrupt the debuggee even if it
18790 There are various additional Cygwin-specific commands, described in
18791 this section. Working with DLLs that have no debugging symbols is
18792 described in @ref{Non-debug DLL Symbols}.
18797 This is a prefix of MS Windows-specific commands which print
18798 information about the target system and important OS structures.
18800 @item info w32 selector
18801 This command displays information returned by
18802 the Win32 API @code{GetThreadSelectorEntry} function.
18803 It takes an optional argument that is evaluated to
18804 a long value to give the information about this given selector.
18805 Without argument, this command displays information
18806 about the six segment registers.
18808 @item info w32 thread-information-block
18809 This command displays thread specific information stored in the
18810 Thread Information Block (readable on the X86 CPU family using @code{$fs}
18811 selector for 32-bit programs and @code{$gs} for 64-bit programs).
18815 This is a Cygwin-specific alias of @code{info shared}.
18817 @kindex dll-symbols
18819 This command loads symbols from a dll similarly to
18820 add-sym command but without the need to specify a base address.
18822 @kindex set cygwin-exceptions
18823 @cindex debugging the Cygwin DLL
18824 @cindex Cygwin DLL, debugging
18825 @item set cygwin-exceptions @var{mode}
18826 If @var{mode} is @code{on}, @value{GDBN} will break on exceptions that
18827 happen inside the Cygwin DLL. If @var{mode} is @code{off},
18828 @value{GDBN} will delay recognition of exceptions, and may ignore some
18829 exceptions which seem to be caused by internal Cygwin DLL
18830 ``bookkeeping''. This option is meant primarily for debugging the
18831 Cygwin DLL itself; the default value is @code{off} to avoid annoying
18832 @value{GDBN} users with false @code{SIGSEGV} signals.
18834 @kindex show cygwin-exceptions
18835 @item show cygwin-exceptions
18836 Displays whether @value{GDBN} will break on exceptions that happen
18837 inside the Cygwin DLL itself.
18839 @kindex set new-console
18840 @item set new-console @var{mode}
18841 If @var{mode} is @code{on} the debuggee will
18842 be started in a new console on next start.
18843 If @var{mode} is @code{off}, the debuggee will
18844 be started in the same console as the debugger.
18846 @kindex show new-console
18847 @item show new-console
18848 Displays whether a new console is used
18849 when the debuggee is started.
18851 @kindex set new-group
18852 @item set new-group @var{mode}
18853 This boolean value controls whether the debuggee should
18854 start a new group or stay in the same group as the debugger.
18855 This affects the way the Windows OS handles
18858 @kindex show new-group
18859 @item show new-group
18860 Displays current value of new-group boolean.
18862 @kindex set debugevents
18863 @item set debugevents
18864 This boolean value adds debug output concerning kernel events related
18865 to the debuggee seen by the debugger. This includes events that
18866 signal thread and process creation and exit, DLL loading and
18867 unloading, console interrupts, and debugging messages produced by the
18868 Windows @code{OutputDebugString} API call.
18870 @kindex set debugexec
18871 @item set debugexec
18872 This boolean value adds debug output concerning execute events
18873 (such as resume thread) seen by the debugger.
18875 @kindex set debugexceptions
18876 @item set debugexceptions
18877 This boolean value adds debug output concerning exceptions in the
18878 debuggee seen by the debugger.
18880 @kindex set debugmemory
18881 @item set debugmemory
18882 This boolean value adds debug output concerning debuggee memory reads
18883 and writes by the debugger.
18887 This boolean values specifies whether the debuggee is called
18888 via a shell or directly (default value is on).
18892 Displays if the debuggee will be started with a shell.
18897 * Non-debug DLL Symbols:: Support for DLLs without debugging symbols
18900 @node Non-debug DLL Symbols
18901 @subsubsection Support for DLLs without Debugging Symbols
18902 @cindex DLLs with no debugging symbols
18903 @cindex Minimal symbols and DLLs
18905 Very often on windows, some of the DLLs that your program relies on do
18906 not include symbolic debugging information (for example,
18907 @file{kernel32.dll}). When @value{GDBN} doesn't recognize any debugging
18908 symbols in a DLL, it relies on the minimal amount of symbolic
18909 information contained in the DLL's export table. This section
18910 describes working with such symbols, known internally to @value{GDBN} as
18911 ``minimal symbols''.
18913 Note that before the debugged program has started execution, no DLLs
18914 will have been loaded. The easiest way around this problem is simply to
18915 start the program --- either by setting a breakpoint or letting the
18916 program run once to completion. It is also possible to force
18917 @value{GDBN} to load a particular DLL before starting the executable ---
18918 see the shared library information in @ref{Files}, or the
18919 @code{dll-symbols} command in @ref{Cygwin Native}. Currently,
18920 explicitly loading symbols from a DLL with no debugging information will
18921 cause the symbol names to be duplicated in @value{GDBN}'s lookup table,
18922 which may adversely affect symbol lookup performance.
18924 @subsubsection DLL Name Prefixes
18926 In keeping with the naming conventions used by the Microsoft debugging
18927 tools, DLL export symbols are made available with a prefix based on the
18928 DLL name, for instance @code{KERNEL32!CreateFileA}. The plain name is
18929 also entered into the symbol table, so @code{CreateFileA} is often
18930 sufficient. In some cases there will be name clashes within a program
18931 (particularly if the executable itself includes full debugging symbols)
18932 necessitating the use of the fully qualified name when referring to the
18933 contents of the DLL. Use single-quotes around the name to avoid the
18934 exclamation mark (``!'') being interpreted as a language operator.
18936 Note that the internal name of the DLL may be all upper-case, even
18937 though the file name of the DLL is lower-case, or vice-versa. Since
18938 symbols within @value{GDBN} are @emph{case-sensitive} this may cause
18939 some confusion. If in doubt, try the @code{info functions} and
18940 @code{info variables} commands or even @code{maint print msymbols}
18941 (@pxref{Symbols}). Here's an example:
18944 (@value{GDBP}) info function CreateFileA
18945 All functions matching regular expression "CreateFileA":
18947 Non-debugging symbols:
18948 0x77e885f4 CreateFileA
18949 0x77e885f4 KERNEL32!CreateFileA
18953 (@value{GDBP}) info function !
18954 All functions matching regular expression "!":
18956 Non-debugging symbols:
18957 0x6100114c cygwin1!__assert
18958 0x61004034 cygwin1!_dll_crt0@@0
18959 0x61004240 cygwin1!dll_crt0(per_process *)
18963 @subsubsection Working with Minimal Symbols
18965 Symbols extracted from a DLL's export table do not contain very much
18966 type information. All that @value{GDBN} can do is guess whether a symbol
18967 refers to a function or variable depending on the linker section that
18968 contains the symbol. Also note that the actual contents of the memory
18969 contained in a DLL are not available unless the program is running. This
18970 means that you cannot examine the contents of a variable or disassemble
18971 a function within a DLL without a running program.
18973 Variables are generally treated as pointers and dereferenced
18974 automatically. For this reason, it is often necessary to prefix a
18975 variable name with the address-of operator (``&'') and provide explicit
18976 type information in the command. Here's an example of the type of
18980 (@value{GDBP}) print 'cygwin1!__argv'
18985 (@value{GDBP}) x 'cygwin1!__argv'
18986 0x10021610: "\230y\""
18989 And two possible solutions:
18992 (@value{GDBP}) print ((char **)'cygwin1!__argv')[0]
18993 $2 = 0x22fd98 "/cygdrive/c/mydirectory/myprogram"
18997 (@value{GDBP}) x/2x &'cygwin1!__argv'
18998 0x610c0aa8 <cygwin1!__argv>: 0x10021608 0x00000000
18999 (@value{GDBP}) x/x 0x10021608
19000 0x10021608: 0x0022fd98
19001 (@value{GDBP}) x/s 0x0022fd98
19002 0x22fd98: "/cygdrive/c/mydirectory/myprogram"
19005 Setting a break point within a DLL is possible even before the program
19006 starts execution. However, under these circumstances, @value{GDBN} can't
19007 examine the initial instructions of the function in order to skip the
19008 function's frame set-up code. You can work around this by using ``*&''
19009 to set the breakpoint at a raw memory address:
19012 (@value{GDBP}) break *&'python22!PyOS_Readline'
19013 Breakpoint 1 at 0x1e04eff0
19016 The author of these extensions is not entirely convinced that setting a
19017 break point within a shared DLL like @file{kernel32.dll} is completely
19021 @subsection Commands Specific to @sc{gnu} Hurd Systems
19022 @cindex @sc{gnu} Hurd debugging
19024 This subsection describes @value{GDBN} commands specific to the
19025 @sc{gnu} Hurd native debugging.
19030 @kindex set signals@r{, Hurd command}
19031 @kindex set sigs@r{, Hurd command}
19032 This command toggles the state of inferior signal interception by
19033 @value{GDBN}. Mach exceptions, such as breakpoint traps, are not
19034 affected by this command. @code{sigs} is a shorthand alias for
19039 @kindex show signals@r{, Hurd command}
19040 @kindex show sigs@r{, Hurd command}
19041 Show the current state of intercepting inferior's signals.
19043 @item set signal-thread
19044 @itemx set sigthread
19045 @kindex set signal-thread
19046 @kindex set sigthread
19047 This command tells @value{GDBN} which thread is the @code{libc} signal
19048 thread. That thread is run when a signal is delivered to a running
19049 process. @code{set sigthread} is the shorthand alias of @code{set
19052 @item show signal-thread
19053 @itemx show sigthread
19054 @kindex show signal-thread
19055 @kindex show sigthread
19056 These two commands show which thread will run when the inferior is
19057 delivered a signal.
19060 @kindex set stopped@r{, Hurd command}
19061 This commands tells @value{GDBN} that the inferior process is stopped,
19062 as with the @code{SIGSTOP} signal. The stopped process can be
19063 continued by delivering a signal to it.
19066 @kindex show stopped@r{, Hurd command}
19067 This command shows whether @value{GDBN} thinks the debuggee is
19070 @item set exceptions
19071 @kindex set exceptions@r{, Hurd command}
19072 Use this command to turn off trapping of exceptions in the inferior.
19073 When exception trapping is off, neither breakpoints nor
19074 single-stepping will work. To restore the default, set exception
19077 @item show exceptions
19078 @kindex show exceptions@r{, Hurd command}
19079 Show the current state of trapping exceptions in the inferior.
19081 @item set task pause
19082 @kindex set task@r{, Hurd commands}
19083 @cindex task attributes (@sc{gnu} Hurd)
19084 @cindex pause current task (@sc{gnu} Hurd)
19085 This command toggles task suspension when @value{GDBN} has control.
19086 Setting it to on takes effect immediately, and the task is suspended
19087 whenever @value{GDBN} gets control. Setting it to off will take
19088 effect the next time the inferior is continued. If this option is set
19089 to off, you can use @code{set thread default pause on} or @code{set
19090 thread pause on} (see below) to pause individual threads.
19092 @item show task pause
19093 @kindex show task@r{, Hurd commands}
19094 Show the current state of task suspension.
19096 @item set task detach-suspend-count
19097 @cindex task suspend count
19098 @cindex detach from task, @sc{gnu} Hurd
19099 This command sets the suspend count the task will be left with when
19100 @value{GDBN} detaches from it.
19102 @item show task detach-suspend-count
19103 Show the suspend count the task will be left with when detaching.
19105 @item set task exception-port
19106 @itemx set task excp
19107 @cindex task exception port, @sc{gnu} Hurd
19108 This command sets the task exception port to which @value{GDBN} will
19109 forward exceptions. The argument should be the value of the @dfn{send
19110 rights} of the task. @code{set task excp} is a shorthand alias.
19112 @item set noninvasive
19113 @cindex noninvasive task options
19114 This command switches @value{GDBN} to a mode that is the least
19115 invasive as far as interfering with the inferior is concerned. This
19116 is the same as using @code{set task pause}, @code{set exceptions}, and
19117 @code{set signals} to values opposite to the defaults.
19119 @item info send-rights
19120 @itemx info receive-rights
19121 @itemx info port-rights
19122 @itemx info port-sets
19123 @itemx info dead-names
19126 @cindex send rights, @sc{gnu} Hurd
19127 @cindex receive rights, @sc{gnu} Hurd
19128 @cindex port rights, @sc{gnu} Hurd
19129 @cindex port sets, @sc{gnu} Hurd
19130 @cindex dead names, @sc{gnu} Hurd
19131 These commands display information about, respectively, send rights,
19132 receive rights, port rights, port sets, and dead names of a task.
19133 There are also shorthand aliases: @code{info ports} for @code{info
19134 port-rights} and @code{info psets} for @code{info port-sets}.
19136 @item set thread pause
19137 @kindex set thread@r{, Hurd command}
19138 @cindex thread properties, @sc{gnu} Hurd
19139 @cindex pause current thread (@sc{gnu} Hurd)
19140 This command toggles current thread suspension when @value{GDBN} has
19141 control. Setting it to on takes effect immediately, and the current
19142 thread is suspended whenever @value{GDBN} gets control. Setting it to
19143 off will take effect the next time the inferior is continued.
19144 Normally, this command has no effect, since when @value{GDBN} has
19145 control, the whole task is suspended. However, if you used @code{set
19146 task pause off} (see above), this command comes in handy to suspend
19147 only the current thread.
19149 @item show thread pause
19150 @kindex show thread@r{, Hurd command}
19151 This command shows the state of current thread suspension.
19153 @item set thread run
19154 This command sets whether the current thread is allowed to run.
19156 @item show thread run
19157 Show whether the current thread is allowed to run.
19159 @item set thread detach-suspend-count
19160 @cindex thread suspend count, @sc{gnu} Hurd
19161 @cindex detach from thread, @sc{gnu} Hurd
19162 This command sets the suspend count @value{GDBN} will leave on a
19163 thread when detaching. This number is relative to the suspend count
19164 found by @value{GDBN} when it notices the thread; use @code{set thread
19165 takeover-suspend-count} to force it to an absolute value.
19167 @item show thread detach-suspend-count
19168 Show the suspend count @value{GDBN} will leave on the thread when
19171 @item set thread exception-port
19172 @itemx set thread excp
19173 Set the thread exception port to which to forward exceptions. This
19174 overrides the port set by @code{set task exception-port} (see above).
19175 @code{set thread excp} is the shorthand alias.
19177 @item set thread takeover-suspend-count
19178 Normally, @value{GDBN}'s thread suspend counts are relative to the
19179 value @value{GDBN} finds when it notices each thread. This command
19180 changes the suspend counts to be absolute instead.
19182 @item set thread default
19183 @itemx show thread default
19184 @cindex thread default settings, @sc{gnu} Hurd
19185 Each of the above @code{set thread} commands has a @code{set thread
19186 default} counterpart (e.g., @code{set thread default pause}, @code{set
19187 thread default exception-port}, etc.). The @code{thread default}
19188 variety of commands sets the default thread properties for all
19189 threads; you can then change the properties of individual threads with
19190 the non-default commands.
19195 @subsection QNX Neutrino
19196 @cindex QNX Neutrino
19198 @value{GDBN} provides the following commands specific to the QNX
19202 @item set debug nto-debug
19203 @kindex set debug nto-debug
19204 When set to on, enables debugging messages specific to the QNX
19207 @item show debug nto-debug
19208 @kindex show debug nto-debug
19209 Show the current state of QNX Neutrino messages.
19216 @value{GDBN} provides the following commands specific to the Darwin target:
19219 @item set debug darwin @var{num}
19220 @kindex set debug darwin
19221 When set to a non zero value, enables debugging messages specific to
19222 the Darwin support. Higher values produce more verbose output.
19224 @item show debug darwin
19225 @kindex show debug darwin
19226 Show the current state of Darwin messages.
19228 @item set debug mach-o @var{num}
19229 @kindex set debug mach-o
19230 When set to a non zero value, enables debugging messages while
19231 @value{GDBN} is reading Darwin object files. (@dfn{Mach-O} is the
19232 file format used on Darwin for object and executable files.) Higher
19233 values produce more verbose output. This is a command to diagnose
19234 problems internal to @value{GDBN} and should not be needed in normal
19237 @item show debug mach-o
19238 @kindex show debug mach-o
19239 Show the current state of Mach-O file messages.
19241 @item set mach-exceptions on
19242 @itemx set mach-exceptions off
19243 @kindex set mach-exceptions
19244 On Darwin, faults are first reported as a Mach exception and are then
19245 mapped to a Posix signal. Use this command to turn on trapping of
19246 Mach exceptions in the inferior. This might be sometimes useful to
19247 better understand the cause of a fault. The default is off.
19249 @item show mach-exceptions
19250 @kindex show mach-exceptions
19251 Show the current state of exceptions trapping.
19256 @section Embedded Operating Systems
19258 This section describes configurations involving the debugging of
19259 embedded operating systems that are available for several different
19263 * VxWorks:: Using @value{GDBN} with VxWorks
19266 @value{GDBN} includes the ability to debug programs running on
19267 various real-time operating systems.
19270 @subsection Using @value{GDBN} with VxWorks
19276 @kindex target vxworks
19277 @item target vxworks @var{machinename}
19278 A VxWorks system, attached via TCP/IP. The argument @var{machinename}
19279 is the target system's machine name or IP address.
19283 On VxWorks, @code{load} links @var{filename} dynamically on the
19284 current target system as well as adding its symbols in @value{GDBN}.
19286 @value{GDBN} enables developers to spawn and debug tasks running on networked
19287 VxWorks targets from a Unix host. Already-running tasks spawned from
19288 the VxWorks shell can also be debugged. @value{GDBN} uses code that runs on
19289 both the Unix host and on the VxWorks target. The program
19290 @code{@value{GDBP}} is installed and executed on the Unix host. (It may be
19291 installed with the name @code{vxgdb}, to distinguish it from a
19292 @value{GDBN} for debugging programs on the host itself.)
19295 @item VxWorks-timeout @var{args}
19296 @kindex vxworks-timeout
19297 All VxWorks-based targets now support the option @code{vxworks-timeout}.
19298 This option is set by the user, and @var{args} represents the number of
19299 seconds @value{GDBN} waits for responses to rpc's. You might use this if
19300 your VxWorks target is a slow software simulator or is on the far side
19301 of a thin network line.
19304 The following information on connecting to VxWorks was current when
19305 this manual was produced; newer releases of VxWorks may use revised
19308 @findex INCLUDE_RDB
19309 To use @value{GDBN} with VxWorks, you must rebuild your VxWorks kernel
19310 to include the remote debugging interface routines in the VxWorks
19311 library @file{rdb.a}. To do this, define @code{INCLUDE_RDB} in the
19312 VxWorks configuration file @file{configAll.h} and rebuild your VxWorks
19313 kernel. The resulting kernel contains @file{rdb.a}, and spawns the
19314 source debugging task @code{tRdbTask} when VxWorks is booted. For more
19315 information on configuring and remaking VxWorks, see the manufacturer's
19317 @c VxWorks, see the @cite{VxWorks Programmer's Guide}.
19319 Once you have included @file{rdb.a} in your VxWorks system image and set
19320 your Unix execution search path to find @value{GDBN}, you are ready to
19321 run @value{GDBN}. From your Unix host, run @code{@value{GDBP}} (or
19322 @code{vxgdb}, depending on your installation).
19324 @value{GDBN} comes up showing the prompt:
19331 * VxWorks Connection:: Connecting to VxWorks
19332 * VxWorks Download:: VxWorks download
19333 * VxWorks Attach:: Running tasks
19336 @node VxWorks Connection
19337 @subsubsection Connecting to VxWorks
19339 The @value{GDBN} command @code{target} lets you connect to a VxWorks target on the
19340 network. To connect to a target whose host name is ``@code{tt}'', type:
19343 (vxgdb) target vxworks tt
19347 @value{GDBN} displays messages like these:
19350 Attaching remote machine across net...
19355 @value{GDBN} then attempts to read the symbol tables of any object modules
19356 loaded into the VxWorks target since it was last booted. @value{GDBN} locates
19357 these files by searching the directories listed in the command search
19358 path (@pxref{Environment, ,Your Program's Environment}); if it fails
19359 to find an object file, it displays a message such as:
19362 prog.o: No such file or directory.
19365 When this happens, add the appropriate directory to the search path with
19366 the @value{GDBN} command @code{path}, and execute the @code{target}
19369 @node VxWorks Download
19370 @subsubsection VxWorks Download
19372 @cindex download to VxWorks
19373 If you have connected to the VxWorks target and you want to debug an
19374 object that has not yet been loaded, you can use the @value{GDBN}
19375 @code{load} command to download a file from Unix to VxWorks
19376 incrementally. The object file given as an argument to the @code{load}
19377 command is actually opened twice: first by the VxWorks target in order
19378 to download the code, then by @value{GDBN} in order to read the symbol
19379 table. This can lead to problems if the current working directories on
19380 the two systems differ. If both systems have NFS mounted the same
19381 filesystems, you can avoid these problems by using absolute paths.
19382 Otherwise, it is simplest to set the working directory on both systems
19383 to the directory in which the object file resides, and then to reference
19384 the file by its name, without any path. For instance, a program
19385 @file{prog.o} may reside in @file{@var{vxpath}/vw/demo/rdb} in VxWorks
19386 and in @file{@var{hostpath}/vw/demo/rdb} on the host. To load this
19387 program, type this on VxWorks:
19390 -> cd "@var{vxpath}/vw/demo/rdb"
19394 Then, in @value{GDBN}, type:
19397 (vxgdb) cd @var{hostpath}/vw/demo/rdb
19398 (vxgdb) load prog.o
19401 @value{GDBN} displays a response similar to this:
19404 Reading symbol data from wherever/vw/demo/rdb/prog.o... done.
19407 You can also use the @code{load} command to reload an object module
19408 after editing and recompiling the corresponding source file. Note that
19409 this makes @value{GDBN} delete all currently-defined breakpoints,
19410 auto-displays, and convenience variables, and to clear the value
19411 history. (This is necessary in order to preserve the integrity of
19412 debugger's data structures that reference the target system's symbol
19415 @node VxWorks Attach
19416 @subsubsection Running Tasks
19418 @cindex running VxWorks tasks
19419 You can also attach to an existing task using the @code{attach} command as
19423 (vxgdb) attach @var{task}
19427 where @var{task} is the VxWorks hexadecimal task ID. The task can be running
19428 or suspended when you attach to it. Running tasks are suspended at
19429 the time of attachment.
19431 @node Embedded Processors
19432 @section Embedded Processors
19434 This section goes into details specific to particular embedded
19437 @cindex send command to simulator
19438 Whenever a specific embedded processor has a simulator, @value{GDBN}
19439 allows to send an arbitrary command to the simulator.
19442 @item sim @var{command}
19443 @kindex sim@r{, a command}
19444 Send an arbitrary @var{command} string to the simulator. Consult the
19445 documentation for the specific simulator in use for information about
19446 acceptable commands.
19452 * M32R/D:: Renesas M32R/D
19453 * M68K:: Motorola M68K
19454 * MicroBlaze:: Xilinx MicroBlaze
19455 * MIPS Embedded:: MIPS Embedded
19456 * OpenRISC 1000:: OpenRisc 1000
19457 * PA:: HP PA Embedded
19458 * PowerPC Embedded:: PowerPC Embedded
19459 * Sparclet:: Tsqware Sparclet
19460 * Sparclite:: Fujitsu Sparclite
19461 * Z8000:: Zilog Z8000
19464 * Super-H:: Renesas Super-H
19473 @item target rdi @var{dev}
19474 ARM Angel monitor, via RDI library interface to ADP protocol. You may
19475 use this target to communicate with both boards running the Angel
19476 monitor, or with the EmbeddedICE JTAG debug device.
19479 @item target rdp @var{dev}
19484 @value{GDBN} provides the following ARM-specific commands:
19487 @item set arm disassembler
19489 This commands selects from a list of disassembly styles. The
19490 @code{"std"} style is the standard style.
19492 @item show arm disassembler
19494 Show the current disassembly style.
19496 @item set arm apcs32
19497 @cindex ARM 32-bit mode
19498 This command toggles ARM operation mode between 32-bit and 26-bit.
19500 @item show arm apcs32
19501 Display the current usage of the ARM 32-bit mode.
19503 @item set arm fpu @var{fputype}
19504 This command sets the ARM floating-point unit (FPU) type. The
19505 argument @var{fputype} can be one of these:
19509 Determine the FPU type by querying the OS ABI.
19511 Software FPU, with mixed-endian doubles on little-endian ARM
19514 GCC-compiled FPA co-processor.
19516 Software FPU with pure-endian doubles.
19522 Show the current type of the FPU.
19525 This command forces @value{GDBN} to use the specified ABI.
19528 Show the currently used ABI.
19530 @item set arm fallback-mode (arm|thumb|auto)
19531 @value{GDBN} uses the symbol table, when available, to determine
19532 whether instructions are ARM or Thumb. This command controls
19533 @value{GDBN}'s default behavior when the symbol table is not
19534 available. The default is @samp{auto}, which causes @value{GDBN} to
19535 use the current execution mode (from the @code{T} bit in the @code{CPSR}
19538 @item show arm fallback-mode
19539 Show the current fallback instruction mode.
19541 @item set arm force-mode (arm|thumb|auto)
19542 This command overrides use of the symbol table to determine whether
19543 instructions are ARM or Thumb. The default is @samp{auto}, which
19544 causes @value{GDBN} to use the symbol table and then the setting
19545 of @samp{set arm fallback-mode}.
19547 @item show arm force-mode
19548 Show the current forced instruction mode.
19550 @item set debug arm
19551 Toggle whether to display ARM-specific debugging messages from the ARM
19552 target support subsystem.
19554 @item show debug arm
19555 Show whether ARM-specific debugging messages are enabled.
19558 The following commands are available when an ARM target is debugged
19559 using the RDI interface:
19562 @item rdilogfile @r{[}@var{file}@r{]}
19564 @cindex ADP (Angel Debugger Protocol) logging
19565 Set the filename for the ADP (Angel Debugger Protocol) packet log.
19566 With an argument, sets the log file to the specified @var{file}. With
19567 no argument, show the current log file name. The default log file is
19570 @item rdilogenable @r{[}@var{arg}@r{]}
19571 @kindex rdilogenable
19572 Control logging of ADP packets. With an argument of 1 or @code{"yes"}
19573 enables logging, with an argument 0 or @code{"no"} disables it. With
19574 no arguments displays the current setting. When logging is enabled,
19575 ADP packets exchanged between @value{GDBN} and the RDI target device
19576 are logged to a file.
19578 @item set rdiromatzero
19579 @kindex set rdiromatzero
19580 @cindex ROM at zero address, RDI
19581 Tell @value{GDBN} whether the target has ROM at address 0. If on,
19582 vector catching is disabled, so that zero address can be used. If off
19583 (the default), vector catching is enabled. For this command to take
19584 effect, it needs to be invoked prior to the @code{target rdi} command.
19586 @item show rdiromatzero
19587 @kindex show rdiromatzero
19588 Show the current setting of ROM at zero address.
19590 @item set rdiheartbeat
19591 @kindex set rdiheartbeat
19592 @cindex RDI heartbeat
19593 Enable or disable RDI heartbeat packets. It is not recommended to
19594 turn on this option, since it confuses ARM and EPI JTAG interface, as
19595 well as the Angel monitor.
19597 @item show rdiheartbeat
19598 @kindex show rdiheartbeat
19599 Show the setting of RDI heartbeat packets.
19603 @item target sim @r{[}@var{simargs}@r{]} @dots{}
19604 The @value{GDBN} ARM simulator accepts the following optional arguments.
19607 @item --swi-support=@var{type}
19608 Tell the simulator which SWI interfaces to support.
19609 @var{type} may be a comma separated list of the following values.
19610 The default value is @code{all}.
19623 @subsection Renesas M32R/D and M32R/SDI
19626 @kindex target m32r
19627 @item target m32r @var{dev}
19628 Renesas M32R/D ROM monitor.
19630 @kindex target m32rsdi
19631 @item target m32rsdi @var{dev}
19632 Renesas M32R SDI server, connected via parallel port to the board.
19635 The following @value{GDBN} commands are specific to the M32R monitor:
19638 @item set download-path @var{path}
19639 @kindex set download-path
19640 @cindex find downloadable @sc{srec} files (M32R)
19641 Set the default path for finding downloadable @sc{srec} files.
19643 @item show download-path
19644 @kindex show download-path
19645 Show the default path for downloadable @sc{srec} files.
19647 @item set board-address @var{addr}
19648 @kindex set board-address
19649 @cindex M32-EVA target board address
19650 Set the IP address for the M32R-EVA target board.
19652 @item show board-address
19653 @kindex show board-address
19654 Show the current IP address of the target board.
19656 @item set server-address @var{addr}
19657 @kindex set server-address
19658 @cindex download server address (M32R)
19659 Set the IP address for the download server, which is the @value{GDBN}'s
19662 @item show server-address
19663 @kindex show server-address
19664 Display the IP address of the download server.
19666 @item upload @r{[}@var{file}@r{]}
19667 @kindex upload@r{, M32R}
19668 Upload the specified @sc{srec} @var{file} via the monitor's Ethernet
19669 upload capability. If no @var{file} argument is given, the current
19670 executable file is uploaded.
19672 @item tload @r{[}@var{file}@r{]}
19673 @kindex tload@r{, M32R}
19674 Test the @code{upload} command.
19677 The following commands are available for M32R/SDI:
19682 @cindex reset SDI connection, M32R
19683 This command resets the SDI connection.
19687 This command shows the SDI connection status.
19690 @kindex debug_chaos
19691 @cindex M32R/Chaos debugging
19692 Instructs the remote that M32R/Chaos debugging is to be used.
19694 @item use_debug_dma
19695 @kindex use_debug_dma
19696 Instructs the remote to use the DEBUG_DMA method of accessing memory.
19699 @kindex use_mon_code
19700 Instructs the remote to use the MON_CODE method of accessing memory.
19703 @kindex use_ib_break
19704 Instructs the remote to set breakpoints by IB break.
19706 @item use_dbt_break
19707 @kindex use_dbt_break
19708 Instructs the remote to set breakpoints by DBT.
19714 The Motorola m68k configuration includes ColdFire support, and a
19715 target command for the following ROM monitor.
19719 @kindex target dbug
19720 @item target dbug @var{dev}
19721 dBUG ROM monitor for Motorola ColdFire.
19726 @subsection MicroBlaze
19727 @cindex Xilinx MicroBlaze
19728 @cindex XMD, Xilinx Microprocessor Debugger
19730 The MicroBlaze is a soft-core processor supported on various Xilinx
19731 FPGAs, such as Spartan or Virtex series. Boards with these processors
19732 usually have JTAG ports which connect to a host system running the Xilinx
19733 Embedded Development Kit (EDK) or Software Development Kit (SDK).
19734 This host system is used to download the configuration bitstream to
19735 the target FPGA. The Xilinx Microprocessor Debugger (XMD) program
19736 communicates with the target board using the JTAG interface and
19737 presents a @code{gdbserver} interface to the board. By default
19738 @code{xmd} uses port @code{1234}. (While it is possible to change
19739 this default port, it requires the use of undocumented @code{xmd}
19740 commands. Contact Xilinx support if you need to do this.)
19742 Use these GDB commands to connect to the MicroBlaze target processor.
19745 @item target remote :1234
19746 Use this command to connect to the target if you are running @value{GDBN}
19747 on the same system as @code{xmd}.
19749 @item target remote @var{xmd-host}:1234
19750 Use this command to connect to the target if it is connected to @code{xmd}
19751 running on a different system named @var{xmd-host}.
19754 Use this command to download a program to the MicroBlaze target.
19756 @item set debug microblaze @var{n}
19757 Enable MicroBlaze-specific debugging messages if non-zero.
19759 @item show debug microblaze @var{n}
19760 Show MicroBlaze-specific debugging level.
19763 @node MIPS Embedded
19764 @subsection @acronym{MIPS} Embedded
19766 @cindex @acronym{MIPS} boards
19767 @value{GDBN} can use the @acronym{MIPS} remote debugging protocol to talk to a
19768 @acronym{MIPS} board attached to a serial line. This is available when
19769 you configure @value{GDBN} with @samp{--target=mips-elf}.
19772 Use these @value{GDBN} commands to specify the connection to your target board:
19775 @item target mips @var{port}
19776 @kindex target mips @var{port}
19777 To run a program on the board, start up @code{@value{GDBP}} with the
19778 name of your program as the argument. To connect to the board, use the
19779 command @samp{target mips @var{port}}, where @var{port} is the name of
19780 the serial port connected to the board. If the program has not already
19781 been downloaded to the board, you may use the @code{load} command to
19782 download it. You can then use all the usual @value{GDBN} commands.
19784 For example, this sequence connects to the target board through a serial
19785 port, and loads and runs a program called @var{prog} through the
19789 host$ @value{GDBP} @var{prog}
19790 @value{GDBN} is free software and @dots{}
19791 (@value{GDBP}) target mips /dev/ttyb
19792 (@value{GDBP}) load @var{prog}
19796 @item target mips @var{hostname}:@var{portnumber}
19797 On some @value{GDBN} host configurations, you can specify a TCP
19798 connection (for instance, to a serial line managed by a terminal
19799 concentrator) instead of a serial port, using the syntax
19800 @samp{@var{hostname}:@var{portnumber}}.
19802 @item target pmon @var{port}
19803 @kindex target pmon @var{port}
19806 @item target ddb @var{port}
19807 @kindex target ddb @var{port}
19808 NEC's DDB variant of PMON for Vr4300.
19810 @item target lsi @var{port}
19811 @kindex target lsi @var{port}
19812 LSI variant of PMON.
19814 @kindex target r3900
19815 @item target r3900 @var{dev}
19816 Densan DVE-R3900 ROM monitor for Toshiba R3900 Mips.
19818 @kindex target array
19819 @item target array @var{dev}
19820 Array Tech LSI33K RAID controller board.
19826 @value{GDBN} also supports these special commands for @acronym{MIPS} targets:
19829 @item set mipsfpu double
19830 @itemx set mipsfpu single
19831 @itemx set mipsfpu none
19832 @itemx set mipsfpu auto
19833 @itemx show mipsfpu
19834 @kindex set mipsfpu
19835 @kindex show mipsfpu
19836 @cindex @acronym{MIPS} remote floating point
19837 @cindex floating point, @acronym{MIPS} remote
19838 If your target board does not support the @acronym{MIPS} floating point
19839 coprocessor, you should use the command @samp{set mipsfpu none} (if you
19840 need this, you may wish to put the command in your @value{GDBN} init
19841 file). This tells @value{GDBN} how to find the return value of
19842 functions which return floating point values. It also allows
19843 @value{GDBN} to avoid saving the floating point registers when calling
19844 functions on the board. If you are using a floating point coprocessor
19845 with only single precision floating point support, as on the @sc{r4650}
19846 processor, use the command @samp{set mipsfpu single}. The default
19847 double precision floating point coprocessor may be selected using
19848 @samp{set mipsfpu double}.
19850 In previous versions the only choices were double precision or no
19851 floating point, so @samp{set mipsfpu on} will select double precision
19852 and @samp{set mipsfpu off} will select no floating point.
19854 As usual, you can inquire about the @code{mipsfpu} variable with
19855 @samp{show mipsfpu}.
19857 @item set timeout @var{seconds}
19858 @itemx set retransmit-timeout @var{seconds}
19859 @itemx show timeout
19860 @itemx show retransmit-timeout
19861 @cindex @code{timeout}, @acronym{MIPS} protocol
19862 @cindex @code{retransmit-timeout}, @acronym{MIPS} protocol
19863 @kindex set timeout
19864 @kindex show timeout
19865 @kindex set retransmit-timeout
19866 @kindex show retransmit-timeout
19867 You can control the timeout used while waiting for a packet, in the @acronym{MIPS}
19868 remote protocol, with the @code{set timeout @var{seconds}} command. The
19869 default is 5 seconds. Similarly, you can control the timeout used while
19870 waiting for an acknowledgment of a packet with the @code{set
19871 retransmit-timeout @var{seconds}} command. The default is 3 seconds.
19872 You can inspect both values with @code{show timeout} and @code{show
19873 retransmit-timeout}. (These commands are @emph{only} available when
19874 @value{GDBN} is configured for @samp{--target=mips-elf}.)
19876 The timeout set by @code{set timeout} does not apply when @value{GDBN}
19877 is waiting for your program to stop. In that case, @value{GDBN} waits
19878 forever because it has no way of knowing how long the program is going
19879 to run before stopping.
19881 @item set syn-garbage-limit @var{num}
19882 @kindex set syn-garbage-limit@r{, @acronym{MIPS} remote}
19883 @cindex synchronize with remote @acronym{MIPS} target
19884 Limit the maximum number of characters @value{GDBN} should ignore when
19885 it tries to synchronize with the remote target. The default is 10
19886 characters. Setting the limit to -1 means there's no limit.
19888 @item show syn-garbage-limit
19889 @kindex show syn-garbage-limit@r{, @acronym{MIPS} remote}
19890 Show the current limit on the number of characters to ignore when
19891 trying to synchronize with the remote system.
19893 @item set monitor-prompt @var{prompt}
19894 @kindex set monitor-prompt@r{, @acronym{MIPS} remote}
19895 @cindex remote monitor prompt
19896 Tell @value{GDBN} to expect the specified @var{prompt} string from the
19897 remote monitor. The default depends on the target:
19907 @item show monitor-prompt
19908 @kindex show monitor-prompt@r{, @acronym{MIPS} remote}
19909 Show the current strings @value{GDBN} expects as the prompt from the
19912 @item set monitor-warnings
19913 @kindex set monitor-warnings@r{, @acronym{MIPS} remote}
19914 Enable or disable monitor warnings about hardware breakpoints. This
19915 has effect only for the @code{lsi} target. When on, @value{GDBN} will
19916 display warning messages whose codes are returned by the @code{lsi}
19917 PMON monitor for breakpoint commands.
19919 @item show monitor-warnings
19920 @kindex show monitor-warnings@r{, @acronym{MIPS} remote}
19921 Show the current setting of printing monitor warnings.
19923 @item pmon @var{command}
19924 @kindex pmon@r{, @acronym{MIPS} remote}
19925 @cindex send PMON command
19926 This command allows sending an arbitrary @var{command} string to the
19927 monitor. The monitor must be in debug mode for this to work.
19930 @node OpenRISC 1000
19931 @subsection OpenRISC 1000
19932 @cindex OpenRISC 1000
19934 @cindex or1k boards
19935 See OR1k Architecture document (@uref{www.opencores.org}) for more information
19936 about platform and commands.
19940 @kindex target jtag
19941 @item target jtag jtag://@var{host}:@var{port}
19943 Connects to remote JTAG server.
19944 JTAG remote server can be either an or1ksim or JTAG server,
19945 connected via parallel port to the board.
19947 Example: @code{target jtag jtag://localhost:9999}
19950 @item or1ksim @var{command}
19951 If connected to @code{or1ksim} OpenRISC 1000 Architectural
19952 Simulator, proprietary commands can be executed.
19954 @kindex info or1k spr
19955 @item info or1k spr
19956 Displays spr groups.
19958 @item info or1k spr @var{group}
19959 @itemx info or1k spr @var{groupno}
19960 Displays register names in selected group.
19962 @item info or1k spr @var{group} @var{register}
19963 @itemx info or1k spr @var{register}
19964 @itemx info or1k spr @var{groupno} @var{registerno}
19965 @itemx info or1k spr @var{registerno}
19966 Shows information about specified spr register.
19969 @item spr @var{group} @var{register} @var{value}
19970 @itemx spr @var{register @var{value}}
19971 @itemx spr @var{groupno} @var{registerno @var{value}}
19972 @itemx spr @var{registerno @var{value}}
19973 Writes @var{value} to specified spr register.
19976 Some implementations of OpenRISC 1000 Architecture also have hardware trace.
19977 It is very similar to @value{GDBN} trace, except it does not interfere with normal
19978 program execution and is thus much faster. Hardware breakpoints/watchpoint
19979 triggers can be set using:
19982 Load effective address/data
19984 Store effective address/data
19986 Access effective address ($SEA or $LEA) or data ($SDATA/$LDATA)
19991 When triggered, it can capture low level data, like: @code{PC}, @code{LSEA},
19992 @code{LDATA}, @code{SDATA}, @code{READSPR}, @code{WRITESPR}, @code{INSTR}.
19994 @code{htrace} commands:
19995 @cindex OpenRISC 1000 htrace
19998 @item hwatch @var{conditional}
19999 Set hardware watchpoint on combination of Load/Store Effective Address(es)
20000 or Data. For example:
20002 @code{hwatch ($LEA == my_var) && ($LDATA < 50) || ($SEA == my_var) && ($SDATA >= 50)}
20004 @code{hwatch ($LEA == my_var) && ($LDATA < 50) || ($SEA == my_var) && ($SDATA >= 50)}
20008 Display information about current HW trace configuration.
20010 @item htrace trigger @var{conditional}
20011 Set starting criteria for HW trace.
20013 @item htrace qualifier @var{conditional}
20014 Set acquisition qualifier for HW trace.
20016 @item htrace stop @var{conditional}
20017 Set HW trace stopping criteria.
20019 @item htrace record [@var{data}]*
20020 Selects the data to be recorded, when qualifier is met and HW trace was
20023 @item htrace enable
20024 @itemx htrace disable
20025 Enables/disables the HW trace.
20027 @item htrace rewind [@var{filename}]
20028 Clears currently recorded trace data.
20030 If filename is specified, new trace file is made and any newly collected data
20031 will be written there.
20033 @item htrace print [@var{start} [@var{len}]]
20034 Prints trace buffer, using current record configuration.
20036 @item htrace mode continuous
20037 Set continuous trace mode.
20039 @item htrace mode suspend
20040 Set suspend trace mode.
20044 @node PowerPC Embedded
20045 @subsection PowerPC Embedded
20047 @cindex DVC register
20048 @value{GDBN} supports using the DVC (Data Value Compare) register to
20049 implement in hardware simple hardware watchpoint conditions of the form:
20052 (@value{GDBP}) watch @var{ADDRESS|VARIABLE} \
20053 if @var{ADDRESS|VARIABLE} == @var{CONSTANT EXPRESSION}
20056 The DVC register will be automatically used when @value{GDBN} detects
20057 such pattern in a condition expression, and the created watchpoint uses one
20058 debug register (either the @code{exact-watchpoints} option is on and the
20059 variable is scalar, or the variable has a length of one byte). This feature
20060 is available in native @value{GDBN} running on a Linux kernel version 2.6.34
20063 When running on PowerPC embedded processors, @value{GDBN} automatically uses
20064 ranged hardware watchpoints, unless the @code{exact-watchpoints} option is on,
20065 in which case watchpoints using only one debug register are created when
20066 watching variables of scalar types.
20068 You can create an artificial array to watch an arbitrary memory
20069 region using one of the following commands (@pxref{Expressions}):
20072 (@value{GDBP}) watch *((char *) @var{address})@@@var{length}
20073 (@value{GDBP}) watch @{char[@var{length}]@} @var{address}
20076 PowerPC embedded processors support masked watchpoints. See the discussion
20077 about the @code{mask} argument in @ref{Set Watchpoints}.
20079 @cindex ranged breakpoint
20080 PowerPC embedded processors support hardware accelerated
20081 @dfn{ranged breakpoints}. A ranged breakpoint stops execution of
20082 the inferior whenever it executes an instruction at any address within
20083 the range it specifies. To set a ranged breakpoint in @value{GDBN},
20084 use the @code{break-range} command.
20086 @value{GDBN} provides the following PowerPC-specific commands:
20089 @kindex break-range
20090 @item break-range @var{start-location}, @var{end-location}
20091 Set a breakpoint for an address range.
20092 @var{start-location} and @var{end-location} can specify a function name,
20093 a line number, an offset of lines from the current line or from the start
20094 location, or an address of an instruction (see @ref{Specify Location},
20095 for a list of all the possible ways to specify a @var{location}.)
20096 The breakpoint will stop execution of the inferior whenever it
20097 executes an instruction at any address within the specified range,
20098 (including @var{start-location} and @var{end-location}.)
20100 @kindex set powerpc
20101 @item set powerpc soft-float
20102 @itemx show powerpc soft-float
20103 Force @value{GDBN} to use (or not use) a software floating point calling
20104 convention. By default, @value{GDBN} selects the calling convention based
20105 on the selected architecture and the provided executable file.
20107 @item set powerpc vector-abi
20108 @itemx show powerpc vector-abi
20109 Force @value{GDBN} to use the specified calling convention for vector
20110 arguments and return values. The valid options are @samp{auto};
20111 @samp{generic}, to avoid vector registers even if they are present;
20112 @samp{altivec}, to use AltiVec registers; and @samp{spe} to use SPE
20113 registers. By default, @value{GDBN} selects the calling convention
20114 based on the selected architecture and the provided executable file.
20116 @item set powerpc exact-watchpoints
20117 @itemx show powerpc exact-watchpoints
20118 Allow @value{GDBN} to use only one debug register when watching a variable
20119 of scalar type, thus assuming that the variable is accessed through the
20120 address of its first byte.
20122 @kindex target dink32
20123 @item target dink32 @var{dev}
20124 DINK32 ROM monitor.
20126 @kindex target ppcbug
20127 @item target ppcbug @var{dev}
20128 @kindex target ppcbug1
20129 @item target ppcbug1 @var{dev}
20130 PPCBUG ROM monitor for PowerPC.
20133 @item target sds @var{dev}
20134 SDS monitor, running on a PowerPC board (such as Motorola's ADS).
20137 @cindex SDS protocol
20138 The following commands specific to the SDS protocol are supported
20142 @item set sdstimeout @var{nsec}
20143 @kindex set sdstimeout
20144 Set the timeout for SDS protocol reads to be @var{nsec} seconds. The
20145 default is 2 seconds.
20147 @item show sdstimeout
20148 @kindex show sdstimeout
20149 Show the current value of the SDS timeout.
20151 @item sds @var{command}
20152 @kindex sds@r{, a command}
20153 Send the specified @var{command} string to the SDS monitor.
20158 @subsection HP PA Embedded
20162 @kindex target op50n
20163 @item target op50n @var{dev}
20164 OP50N monitor, running on an OKI HPPA board.
20166 @kindex target w89k
20167 @item target w89k @var{dev}
20168 W89K monitor, running on a Winbond HPPA board.
20173 @subsection Tsqware Sparclet
20177 @value{GDBN} enables developers to debug tasks running on
20178 Sparclet targets from a Unix host.
20179 @value{GDBN} uses code that runs on
20180 both the Unix host and on the Sparclet target. The program
20181 @code{@value{GDBP}} is installed and executed on the Unix host.
20184 @item remotetimeout @var{args}
20185 @kindex remotetimeout
20186 @value{GDBN} supports the option @code{remotetimeout}.
20187 This option is set by the user, and @var{args} represents the number of
20188 seconds @value{GDBN} waits for responses.
20191 @cindex compiling, on Sparclet
20192 When compiling for debugging, include the options @samp{-g} to get debug
20193 information and @samp{-Ttext} to relocate the program to where you wish to
20194 load it on the target. You may also want to add the options @samp{-n} or
20195 @samp{-N} in order to reduce the size of the sections. Example:
20198 sparclet-aout-gcc prog.c -Ttext 0x12010000 -g -o prog -N
20201 You can use @code{objdump} to verify that the addresses are what you intended:
20204 sparclet-aout-objdump --headers --syms prog
20207 @cindex running, on Sparclet
20209 your Unix execution search path to find @value{GDBN}, you are ready to
20210 run @value{GDBN}. From your Unix host, run @code{@value{GDBP}}
20211 (or @code{sparclet-aout-gdb}, depending on your installation).
20213 @value{GDBN} comes up showing the prompt:
20220 * Sparclet File:: Setting the file to debug
20221 * Sparclet Connection:: Connecting to Sparclet
20222 * Sparclet Download:: Sparclet download
20223 * Sparclet Execution:: Running and debugging
20226 @node Sparclet File
20227 @subsubsection Setting File to Debug
20229 The @value{GDBN} command @code{file} lets you choose with program to debug.
20232 (gdbslet) file prog
20236 @value{GDBN} then attempts to read the symbol table of @file{prog}.
20237 @value{GDBN} locates
20238 the file by searching the directories listed in the command search
20240 If the file was compiled with debug information (option @samp{-g}), source
20241 files will be searched as well.
20242 @value{GDBN} locates
20243 the source files by searching the directories listed in the directory search
20244 path (@pxref{Environment, ,Your Program's Environment}).
20246 to find a file, it displays a message such as:
20249 prog: No such file or directory.
20252 When this happens, add the appropriate directories to the search paths with
20253 the @value{GDBN} commands @code{path} and @code{dir}, and execute the
20254 @code{target} command again.
20256 @node Sparclet Connection
20257 @subsubsection Connecting to Sparclet
20259 The @value{GDBN} command @code{target} lets you connect to a Sparclet target.
20260 To connect to a target on serial port ``@code{ttya}'', type:
20263 (gdbslet) target sparclet /dev/ttya
20264 Remote target sparclet connected to /dev/ttya
20265 main () at ../prog.c:3
20269 @value{GDBN} displays messages like these:
20275 @node Sparclet Download
20276 @subsubsection Sparclet Download
20278 @cindex download to Sparclet
20279 Once connected to the Sparclet target,
20280 you can use the @value{GDBN}
20281 @code{load} command to download the file from the host to the target.
20282 The file name and load offset should be given as arguments to the @code{load}
20284 Since the file format is aout, the program must be loaded to the starting
20285 address. You can use @code{objdump} to find out what this value is. The load
20286 offset is an offset which is added to the VMA (virtual memory address)
20287 of each of the file's sections.
20288 For instance, if the program
20289 @file{prog} was linked to text address 0x1201000, with data at 0x12010160
20290 and bss at 0x12010170, in @value{GDBN}, type:
20293 (gdbslet) load prog 0x12010000
20294 Loading section .text, size 0xdb0 vma 0x12010000
20297 If the code is loaded at a different address then what the program was linked
20298 to, you may need to use the @code{section} and @code{add-symbol-file} commands
20299 to tell @value{GDBN} where to map the symbol table.
20301 @node Sparclet Execution
20302 @subsubsection Running and Debugging
20304 @cindex running and debugging Sparclet programs
20305 You can now begin debugging the task using @value{GDBN}'s execution control
20306 commands, @code{b}, @code{step}, @code{run}, etc. See the @value{GDBN}
20307 manual for the list of commands.
20311 Breakpoint 1 at 0x12010000: file prog.c, line 3.
20313 Starting program: prog
20314 Breakpoint 1, main (argc=1, argv=0xeffff21c) at prog.c:3
20315 3 char *symarg = 0;
20317 4 char *execarg = "hello!";
20322 @subsection Fujitsu Sparclite
20326 @kindex target sparclite
20327 @item target sparclite @var{dev}
20328 Fujitsu sparclite boards, used only for the purpose of loading.
20329 You must use an additional command to debug the program.
20330 For example: target remote @var{dev} using @value{GDBN} standard
20336 @subsection Zilog Z8000
20339 @cindex simulator, Z8000
20340 @cindex Zilog Z8000 simulator
20342 When configured for debugging Zilog Z8000 targets, @value{GDBN} includes
20345 For the Z8000 family, @samp{target sim} simulates either the Z8002 (the
20346 unsegmented variant of the Z8000 architecture) or the Z8001 (the
20347 segmented variant). The simulator recognizes which architecture is
20348 appropriate by inspecting the object code.
20351 @item target sim @var{args}
20353 @kindex target sim@r{, with Z8000}
20354 Debug programs on a simulated CPU. If the simulator supports setup
20355 options, specify them via @var{args}.
20359 After specifying this target, you can debug programs for the simulated
20360 CPU in the same style as programs for your host computer; use the
20361 @code{file} command to load a new program image, the @code{run} command
20362 to run your program, and so on.
20364 As well as making available all the usual machine registers
20365 (@pxref{Registers, ,Registers}), the Z8000 simulator provides three
20366 additional items of information as specially named registers:
20371 Counts clock-ticks in the simulator.
20374 Counts instructions run in the simulator.
20377 Execution time in 60ths of a second.
20381 You can refer to these values in @value{GDBN} expressions with the usual
20382 conventions; for example, @w{@samp{b fputc if $cycles>5000}} sets a
20383 conditional breakpoint that suspends only after at least 5000
20384 simulated clock ticks.
20387 @subsection Atmel AVR
20390 When configured for debugging the Atmel AVR, @value{GDBN} supports the
20391 following AVR-specific commands:
20394 @item info io_registers
20395 @kindex info io_registers@r{, AVR}
20396 @cindex I/O registers (Atmel AVR)
20397 This command displays information about the AVR I/O registers. For
20398 each register, @value{GDBN} prints its number and value.
20405 When configured for debugging CRIS, @value{GDBN} provides the
20406 following CRIS-specific commands:
20409 @item set cris-version @var{ver}
20410 @cindex CRIS version
20411 Set the current CRIS version to @var{ver}, either @samp{10} or @samp{32}.
20412 The CRIS version affects register names and sizes. This command is useful in
20413 case autodetection of the CRIS version fails.
20415 @item show cris-version
20416 Show the current CRIS version.
20418 @item set cris-dwarf2-cfi
20419 @cindex DWARF-2 CFI and CRIS
20420 Set the usage of DWARF-2 CFI for CRIS debugging. The default is @samp{on}.
20421 Change to @samp{off} when using @code{gcc-cris} whose version is below
20424 @item show cris-dwarf2-cfi
20425 Show the current state of using DWARF-2 CFI.
20427 @item set cris-mode @var{mode}
20429 Set the current CRIS mode to @var{mode}. It should only be changed when
20430 debugging in guru mode, in which case it should be set to
20431 @samp{guru} (the default is @samp{normal}).
20433 @item show cris-mode
20434 Show the current CRIS mode.
20438 @subsection Renesas Super-H
20441 For the Renesas Super-H processor, @value{GDBN} provides these
20446 @kindex regs@r{, Super-H}
20447 Show the values of all Super-H registers.
20449 @item set sh calling-convention @var{convention}
20450 @kindex set sh calling-convention
20451 Set the calling-convention used when calling functions from @value{GDBN}.
20452 Allowed values are @samp{gcc}, which is the default setting, and @samp{renesas}.
20453 With the @samp{gcc} setting, functions are called using the @value{NGCC} calling
20454 convention. If the DWARF-2 information of the called function specifies
20455 that the function follows the Renesas calling convention, the function
20456 is called using the Renesas calling convention. If the calling convention
20457 is set to @samp{renesas}, the Renesas calling convention is always used,
20458 regardless of the DWARF-2 information. This can be used to override the
20459 default of @samp{gcc} if debug information is missing, or the compiler
20460 does not emit the DWARF-2 calling convention entry for a function.
20462 @item show sh calling-convention
20463 @kindex show sh calling-convention
20464 Show the current calling convention setting.
20469 @node Architectures
20470 @section Architectures
20472 This section describes characteristics of architectures that affect
20473 all uses of @value{GDBN} with the architecture, both native and cross.
20479 * HPPA:: HP PA architecture
20480 * SPU:: Cell Broadband Engine SPU architecture
20485 @subsection x86 Architecture-specific Issues
20488 @item set struct-convention @var{mode}
20489 @kindex set struct-convention
20490 @cindex struct return convention
20491 @cindex struct/union returned in registers
20492 Set the convention used by the inferior to return @code{struct}s and
20493 @code{union}s from functions to @var{mode}. Possible values of
20494 @var{mode} are @code{"pcc"}, @code{"reg"}, and @code{"default"} (the
20495 default). @code{"default"} or @code{"pcc"} means that @code{struct}s
20496 are returned on the stack, while @code{"reg"} means that a
20497 @code{struct} or a @code{union} whose size is 1, 2, 4, or 8 bytes will
20498 be returned in a register.
20500 @item show struct-convention
20501 @kindex show struct-convention
20502 Show the current setting of the convention to return @code{struct}s
20509 See the following section.
20512 @subsection @acronym{MIPS}
20514 @cindex stack on Alpha
20515 @cindex stack on @acronym{MIPS}
20516 @cindex Alpha stack
20517 @cindex @acronym{MIPS} stack
20518 Alpha- and @acronym{MIPS}-based computers use an unusual stack frame, which
20519 sometimes requires @value{GDBN} to search backward in the object code to
20520 find the beginning of a function.
20522 @cindex response time, @acronym{MIPS} debugging
20523 To improve response time (especially for embedded applications, where
20524 @value{GDBN} may be restricted to a slow serial line for this search)
20525 you may want to limit the size of this search, using one of these
20529 @cindex @code{heuristic-fence-post} (Alpha, @acronym{MIPS})
20530 @item set heuristic-fence-post @var{limit}
20531 Restrict @value{GDBN} to examining at most @var{limit} bytes in its
20532 search for the beginning of a function. A value of @var{0} (the
20533 default) means there is no limit. However, except for @var{0}, the
20534 larger the limit the more bytes @code{heuristic-fence-post} must search
20535 and therefore the longer it takes to run. You should only need to use
20536 this command when debugging a stripped executable.
20538 @item show heuristic-fence-post
20539 Display the current limit.
20543 These commands are available @emph{only} when @value{GDBN} is configured
20544 for debugging programs on Alpha or @acronym{MIPS} processors.
20546 Several @acronym{MIPS}-specific commands are available when debugging @acronym{MIPS}
20550 @item set mips abi @var{arg}
20551 @kindex set mips abi
20552 @cindex set ABI for @acronym{MIPS}
20553 Tell @value{GDBN} which @acronym{MIPS} ABI is used by the inferior. Possible
20554 values of @var{arg} are:
20558 The default ABI associated with the current binary (this is the
20568 @item show mips abi
20569 @kindex show mips abi
20570 Show the @acronym{MIPS} ABI used by @value{GDBN} to debug the inferior.
20572 @item set mips compression @var{arg}
20573 @kindex set mips compression
20574 @cindex code compression, @acronym{MIPS}
20575 Tell @value{GDBN} which @acronym{MIPS} compressed
20576 @acronym{ISA, Instruction Set Architecture} encoding is used by the
20577 inferior. @value{GDBN} uses this for code disassembly and other
20578 internal interpretation purposes. This setting is only referred to
20579 when no executable has been associated with the debugging session or
20580 the executable does not provide information about the encoding it uses.
20581 Otherwise this setting is automatically updated from information
20582 provided by the executable.
20584 Possible values of @var{arg} are @samp{mips16} and @samp{micromips}.
20585 The default compressed @acronym{ISA} encoding is @samp{mips16}, as
20586 executables containing @acronym{MIPS16} code frequently are not
20587 identified as such.
20589 This setting is ``sticky''; that is, it retains its value across
20590 debugging sessions until reset either explicitly with this command or
20591 implicitly from an executable.
20593 The compiler and/or assembler typically add symbol table annotations to
20594 identify functions compiled for the @acronym{MIPS16} or
20595 @acronym{microMIPS} @acronym{ISA}s. If these function-scope annotations
20596 are present, @value{GDBN} uses them in preference to the global
20597 compressed @acronym{ISA} encoding setting.
20599 @item show mips compression
20600 @kindex show mips compression
20601 Show the @acronym{MIPS} compressed @acronym{ISA} encoding used by
20602 @value{GDBN} to debug the inferior.
20605 @itemx show mipsfpu
20606 @xref{MIPS Embedded, set mipsfpu}.
20608 @item set mips mask-address @var{arg}
20609 @kindex set mips mask-address
20610 @cindex @acronym{MIPS} addresses, masking
20611 This command determines whether the most-significant 32 bits of 64-bit
20612 @acronym{MIPS} addresses are masked off. The argument @var{arg} can be
20613 @samp{on}, @samp{off}, or @samp{auto}. The latter is the default
20614 setting, which lets @value{GDBN} determine the correct value.
20616 @item show mips mask-address
20617 @kindex show mips mask-address
20618 Show whether the upper 32 bits of @acronym{MIPS} addresses are masked off or
20621 @item set remote-mips64-transfers-32bit-regs
20622 @kindex set remote-mips64-transfers-32bit-regs
20623 This command controls compatibility with 64-bit @acronym{MIPS} targets that
20624 transfer data in 32-bit quantities. If you have an old @acronym{MIPS} 64 target
20625 that transfers 32 bits for some registers, like @sc{sr} and @sc{fsr},
20626 and 64 bits for other registers, set this option to @samp{on}.
20628 @item show remote-mips64-transfers-32bit-regs
20629 @kindex show remote-mips64-transfers-32bit-regs
20630 Show the current setting of compatibility with older @acronym{MIPS} 64 targets.
20632 @item set debug mips
20633 @kindex set debug mips
20634 This command turns on and off debugging messages for the @acronym{MIPS}-specific
20635 target code in @value{GDBN}.
20637 @item show debug mips
20638 @kindex show debug mips
20639 Show the current setting of @acronym{MIPS} debugging messages.
20645 @cindex HPPA support
20647 When @value{GDBN} is debugging the HP PA architecture, it provides the
20648 following special commands:
20651 @item set debug hppa
20652 @kindex set debug hppa
20653 This command determines whether HPPA architecture-specific debugging
20654 messages are to be displayed.
20656 @item show debug hppa
20657 Show whether HPPA debugging messages are displayed.
20659 @item maint print unwind @var{address}
20660 @kindex maint print unwind@r{, HPPA}
20661 This command displays the contents of the unwind table entry at the
20662 given @var{address}.
20668 @subsection Cell Broadband Engine SPU architecture
20669 @cindex Cell Broadband Engine
20672 When @value{GDBN} is debugging the Cell Broadband Engine SPU architecture,
20673 it provides the following special commands:
20676 @item info spu event
20678 Display SPU event facility status. Shows current event mask
20679 and pending event status.
20681 @item info spu signal
20682 Display SPU signal notification facility status. Shows pending
20683 signal-control word and signal notification mode of both signal
20684 notification channels.
20686 @item info spu mailbox
20687 Display SPU mailbox facility status. Shows all pending entries,
20688 in order of processing, in each of the SPU Write Outbound,
20689 SPU Write Outbound Interrupt, and SPU Read Inbound mailboxes.
20692 Display MFC DMA status. Shows all pending commands in the MFC
20693 DMA queue. For each entry, opcode, tag, class IDs, effective
20694 and local store addresses and transfer size are shown.
20696 @item info spu proxydma
20697 Display MFC Proxy-DMA status. Shows all pending commands in the MFC
20698 Proxy-DMA queue. For each entry, opcode, tag, class IDs, effective
20699 and local store addresses and transfer size are shown.
20703 When @value{GDBN} is debugging a combined PowerPC/SPU application
20704 on the Cell Broadband Engine, it provides in addition the following
20708 @item set spu stop-on-load @var{arg}
20710 Set whether to stop for new SPE threads. When set to @code{on}, @value{GDBN}
20711 will give control to the user when a new SPE thread enters its @code{main}
20712 function. The default is @code{off}.
20714 @item show spu stop-on-load
20716 Show whether to stop for new SPE threads.
20718 @item set spu auto-flush-cache @var{arg}
20719 Set whether to automatically flush the software-managed cache. When set to
20720 @code{on}, @value{GDBN} will automatically cause the SPE software-managed
20721 cache to be flushed whenever SPE execution stops. This provides a consistent
20722 view of PowerPC memory that is accessed via the cache. If an application
20723 does not use the software-managed cache, this option has no effect.
20725 @item show spu auto-flush-cache
20726 Show whether to automatically flush the software-managed cache.
20731 @subsection PowerPC
20732 @cindex PowerPC architecture
20734 When @value{GDBN} is debugging the PowerPC architecture, it provides a set of
20735 pseudo-registers to enable inspection of 128-bit wide Decimal Floating Point
20736 numbers stored in the floating point registers. These values must be stored
20737 in two consecutive registers, always starting at an even register like
20738 @code{f0} or @code{f2}.
20740 The pseudo-registers go from @code{$dl0} through @code{$dl15}, and are formed
20741 by joining the even/odd register pairs @code{f0} and @code{f1} for @code{$dl0},
20742 @code{f2} and @code{f3} for @code{$dl1} and so on.
20744 For POWER7 processors, @value{GDBN} provides a set of pseudo-registers, the 64-bit
20745 wide Extended Floating Point Registers (@samp{f32} through @samp{f63}).
20748 @node Controlling GDB
20749 @chapter Controlling @value{GDBN}
20751 You can alter the way @value{GDBN} interacts with you by using the
20752 @code{set} command. For commands controlling how @value{GDBN} displays
20753 data, see @ref{Print Settings, ,Print Settings}. Other settings are
20758 * Editing:: Command editing
20759 * Command History:: Command history
20760 * Screen Size:: Screen size
20761 * Numbers:: Numbers
20762 * ABI:: Configuring the current ABI
20763 * Auto-loading:: Automatically loading associated files
20764 * Messages/Warnings:: Optional warnings and messages
20765 * Debugging Output:: Optional messages about internal happenings
20766 * Other Misc Settings:: Other Miscellaneous Settings
20774 @value{GDBN} indicates its readiness to read a command by printing a string
20775 called the @dfn{prompt}. This string is normally @samp{(@value{GDBP})}. You
20776 can change the prompt string with the @code{set prompt} command. For
20777 instance, when debugging @value{GDBN} with @value{GDBN}, it is useful to change
20778 the prompt in one of the @value{GDBN} sessions so that you can always tell
20779 which one you are talking to.
20781 @emph{Note:} @code{set prompt} does not add a space for you after the
20782 prompt you set. This allows you to set a prompt which ends in a space
20783 or a prompt that does not.
20787 @item set prompt @var{newprompt}
20788 Directs @value{GDBN} to use @var{newprompt} as its prompt string henceforth.
20790 @kindex show prompt
20792 Prints a line of the form: @samp{Gdb's prompt is: @var{your-prompt}}
20795 Versions of @value{GDBN} that ship with Python scripting enabled have
20796 prompt extensions. The commands for interacting with these extensions
20800 @kindex set extended-prompt
20801 @item set extended-prompt @var{prompt}
20802 Set an extended prompt that allows for substitutions.
20803 @xref{gdb.prompt}, for a list of escape sequences that can be used for
20804 substitution. Any escape sequences specified as part of the prompt
20805 string are replaced with the corresponding strings each time the prompt
20811 set extended-prompt Current working directory: \w (gdb)
20814 Note that when an extended-prompt is set, it takes control of the
20815 @var{prompt_hook} hook. @xref{prompt_hook}, for further information.
20817 @kindex show extended-prompt
20818 @item show extended-prompt
20819 Prints the extended prompt. Any escape sequences specified as part of
20820 the prompt string with @code{set extended-prompt}, are replaced with the
20821 corresponding strings each time the prompt is displayed.
20825 @section Command Editing
20827 @cindex command line editing
20829 @value{GDBN} reads its input commands via the @dfn{Readline} interface. This
20830 @sc{gnu} library provides consistent behavior for programs which provide a
20831 command line interface to the user. Advantages are @sc{gnu} Emacs-style
20832 or @dfn{vi}-style inline editing of commands, @code{csh}-like history
20833 substitution, and a storage and recall of command history across
20834 debugging sessions.
20836 You may control the behavior of command line editing in @value{GDBN} with the
20837 command @code{set}.
20840 @kindex set editing
20843 @itemx set editing on
20844 Enable command line editing (enabled by default).
20846 @item set editing off
20847 Disable command line editing.
20849 @kindex show editing
20851 Show whether command line editing is enabled.
20854 @ifset SYSTEM_READLINE
20855 @xref{Command Line Editing, , , rluserman, GNU Readline Library},
20857 @ifclear SYSTEM_READLINE
20858 @xref{Command Line Editing},
20860 for more details about the Readline
20861 interface. Users unfamiliar with @sc{gnu} Emacs or @code{vi} are
20862 encouraged to read that chapter.
20864 @node Command History
20865 @section Command History
20866 @cindex command history
20868 @value{GDBN} can keep track of the commands you type during your
20869 debugging sessions, so that you can be certain of precisely what
20870 happened. Use these commands to manage the @value{GDBN} command
20873 @value{GDBN} uses the @sc{gnu} History library, a part of the Readline
20874 package, to provide the history facility.
20875 @ifset SYSTEM_READLINE
20876 @xref{Using History Interactively, , , history, GNU History Library},
20878 @ifclear SYSTEM_READLINE
20879 @xref{Using History Interactively},
20881 for the detailed description of the History library.
20883 To issue a command to @value{GDBN} without affecting certain aspects of
20884 the state which is seen by users, prefix it with @samp{server }
20885 (@pxref{Server Prefix}). This
20886 means that this command will not affect the command history, nor will it
20887 affect @value{GDBN}'s notion of which command to repeat if @key{RET} is
20888 pressed on a line by itself.
20890 @cindex @code{server}, command prefix
20891 The server prefix does not affect the recording of values into the value
20892 history; to print a value without recording it into the value history,
20893 use the @code{output} command instead of the @code{print} command.
20895 Here is the description of @value{GDBN} commands related to command
20899 @cindex history substitution
20900 @cindex history file
20901 @kindex set history filename
20902 @cindex @env{GDBHISTFILE}, environment variable
20903 @item set history filename @var{fname}
20904 Set the name of the @value{GDBN} command history file to @var{fname}.
20905 This is the file where @value{GDBN} reads an initial command history
20906 list, and where it writes the command history from this session when it
20907 exits. You can access this list through history expansion or through
20908 the history command editing characters listed below. This file defaults
20909 to the value of the environment variable @code{GDBHISTFILE}, or to
20910 @file{./.gdb_history} (@file{./_gdb_history} on MS-DOS) if this variable
20913 @cindex save command history
20914 @kindex set history save
20915 @item set history save
20916 @itemx set history save on
20917 Record command history in a file, whose name may be specified with the
20918 @code{set history filename} command. By default, this option is disabled.
20920 @item set history save off
20921 Stop recording command history in a file.
20923 @cindex history size
20924 @kindex set history size
20925 @cindex @env{HISTSIZE}, environment variable
20926 @item set history size @var{size}
20927 Set the number of commands which @value{GDBN} keeps in its history list.
20928 This defaults to the value of the environment variable
20929 @code{HISTSIZE}, or to 256 if this variable is not set.
20932 History expansion assigns special meaning to the character @kbd{!}.
20933 @ifset SYSTEM_READLINE
20934 @xref{Event Designators, , , history, GNU History Library},
20936 @ifclear SYSTEM_READLINE
20937 @xref{Event Designators},
20941 @cindex history expansion, turn on/off
20942 Since @kbd{!} is also the logical not operator in C, history expansion
20943 is off by default. If you decide to enable history expansion with the
20944 @code{set history expansion on} command, you may sometimes need to
20945 follow @kbd{!} (when it is used as logical not, in an expression) with
20946 a space or a tab to prevent it from being expanded. The readline
20947 history facilities do not attempt substitution on the strings
20948 @kbd{!=} and @kbd{!(}, even when history expansion is enabled.
20950 The commands to control history expansion are:
20953 @item set history expansion on
20954 @itemx set history expansion
20955 @kindex set history expansion
20956 Enable history expansion. History expansion is off by default.
20958 @item set history expansion off
20959 Disable history expansion.
20962 @kindex show history
20964 @itemx show history filename
20965 @itemx show history save
20966 @itemx show history size
20967 @itemx show history expansion
20968 These commands display the state of the @value{GDBN} history parameters.
20969 @code{show history} by itself displays all four states.
20974 @kindex show commands
20975 @cindex show last commands
20976 @cindex display command history
20977 @item show commands
20978 Display the last ten commands in the command history.
20980 @item show commands @var{n}
20981 Print ten commands centered on command number @var{n}.
20983 @item show commands +
20984 Print ten commands just after the commands last printed.
20988 @section Screen Size
20989 @cindex size of screen
20990 @cindex pauses in output
20992 Certain commands to @value{GDBN} may produce large amounts of
20993 information output to the screen. To help you read all of it,
20994 @value{GDBN} pauses and asks you for input at the end of each page of
20995 output. Type @key{RET} when you want to continue the output, or @kbd{q}
20996 to discard the remaining output. Also, the screen width setting
20997 determines when to wrap lines of output. Depending on what is being
20998 printed, @value{GDBN} tries to break the line at a readable place,
20999 rather than simply letting it overflow onto the following line.
21001 Normally @value{GDBN} knows the size of the screen from the terminal
21002 driver software. For example, on Unix @value{GDBN} uses the termcap data base
21003 together with the value of the @code{TERM} environment variable and the
21004 @code{stty rows} and @code{stty cols} settings. If this is not correct,
21005 you can override it with the @code{set height} and @code{set
21012 @kindex show height
21013 @item set height @var{lpp}
21015 @itemx set width @var{cpl}
21017 These @code{set} commands specify a screen height of @var{lpp} lines and
21018 a screen width of @var{cpl} characters. The associated @code{show}
21019 commands display the current settings.
21021 If you specify a height of zero lines, @value{GDBN} does not pause during
21022 output no matter how long the output is. This is useful if output is to a
21023 file or to an editor buffer.
21025 Likewise, you can specify @samp{set width 0} to prevent @value{GDBN}
21026 from wrapping its output.
21028 @item set pagination on
21029 @itemx set pagination off
21030 @kindex set pagination
21031 Turn the output pagination on or off; the default is on. Turning
21032 pagination off is the alternative to @code{set height 0}. Note that
21033 running @value{GDBN} with the @option{--batch} option (@pxref{Mode
21034 Options, -batch}) also automatically disables pagination.
21036 @item show pagination
21037 @kindex show pagination
21038 Show the current pagination mode.
21043 @cindex number representation
21044 @cindex entering numbers
21046 You can always enter numbers in octal, decimal, or hexadecimal in
21047 @value{GDBN} by the usual conventions: octal numbers begin with
21048 @samp{0}, decimal numbers end with @samp{.}, and hexadecimal numbers
21049 begin with @samp{0x}. Numbers that neither begin with @samp{0} or
21050 @samp{0x}, nor end with a @samp{.} are, by default, entered in base
21051 10; likewise, the default display for numbers---when no particular
21052 format is specified---is base 10. You can change the default base for
21053 both input and output with the commands described below.
21056 @kindex set input-radix
21057 @item set input-radix @var{base}
21058 Set the default base for numeric input. Supported choices
21059 for @var{base} are decimal 8, 10, or 16. @var{base} must itself be
21060 specified either unambiguously or using the current input radix; for
21064 set input-radix 012
21065 set input-radix 10.
21066 set input-radix 0xa
21070 sets the input base to decimal. On the other hand, @samp{set input-radix 10}
21071 leaves the input radix unchanged, no matter what it was, since
21072 @samp{10}, being without any leading or trailing signs of its base, is
21073 interpreted in the current radix. Thus, if the current radix is 16,
21074 @samp{10} is interpreted in hex, i.e.@: as 16 decimal, which doesn't
21077 @kindex set output-radix
21078 @item set output-radix @var{base}
21079 Set the default base for numeric display. Supported choices
21080 for @var{base} are decimal 8, 10, or 16. @var{base} must itself be
21081 specified either unambiguously or using the current input radix.
21083 @kindex show input-radix
21084 @item show input-radix
21085 Display the current default base for numeric input.
21087 @kindex show output-radix
21088 @item show output-radix
21089 Display the current default base for numeric display.
21091 @item set radix @r{[}@var{base}@r{]}
21095 These commands set and show the default base for both input and output
21096 of numbers. @code{set radix} sets the radix of input and output to
21097 the same base; without an argument, it resets the radix back to its
21098 default value of 10.
21103 @section Configuring the Current ABI
21105 @value{GDBN} can determine the @dfn{ABI} (Application Binary Interface) of your
21106 application automatically. However, sometimes you need to override its
21107 conclusions. Use these commands to manage @value{GDBN}'s view of the
21114 One @value{GDBN} configuration can debug binaries for multiple operating
21115 system targets, either via remote debugging or native emulation.
21116 @value{GDBN} will autodetect the @dfn{OS ABI} (Operating System ABI) in use,
21117 but you can override its conclusion using the @code{set osabi} command.
21118 One example where this is useful is in debugging of binaries which use
21119 an alternate C library (e.g.@: @sc{uClibc} for @sc{gnu}/Linux) which does
21120 not have the same identifying marks that the standard C library for your
21125 Show the OS ABI currently in use.
21128 With no argument, show the list of registered available OS ABI's.
21130 @item set osabi @var{abi}
21131 Set the current OS ABI to @var{abi}.
21134 @cindex float promotion
21136 Generally, the way that an argument of type @code{float} is passed to a
21137 function depends on whether the function is prototyped. For a prototyped
21138 (i.e.@: ANSI/ISO style) function, @code{float} arguments are passed unchanged,
21139 according to the architecture's convention for @code{float}. For unprototyped
21140 (i.e.@: K&R style) functions, @code{float} arguments are first promoted to type
21141 @code{double} and then passed.
21143 Unfortunately, some forms of debug information do not reliably indicate whether
21144 a function is prototyped. If @value{GDBN} calls a function that is not marked
21145 as prototyped, it consults @kbd{set coerce-float-to-double}.
21148 @kindex set coerce-float-to-double
21149 @item set coerce-float-to-double
21150 @itemx set coerce-float-to-double on
21151 Arguments of type @code{float} will be promoted to @code{double} when passed
21152 to an unprototyped function. This is the default setting.
21154 @item set coerce-float-to-double off
21155 Arguments of type @code{float} will be passed directly to unprototyped
21158 @kindex show coerce-float-to-double
21159 @item show coerce-float-to-double
21160 Show the current setting of promoting @code{float} to @code{double}.
21164 @kindex show cp-abi
21165 @value{GDBN} needs to know the ABI used for your program's C@t{++}
21166 objects. The correct C@t{++} ABI depends on which C@t{++} compiler was
21167 used to build your application. @value{GDBN} only fully supports
21168 programs with a single C@t{++} ABI; if your program contains code using
21169 multiple C@t{++} ABI's or if @value{GDBN} can not identify your
21170 program's ABI correctly, you can tell @value{GDBN} which ABI to use.
21171 Currently supported ABI's include ``gnu-v2'', for @code{g++} versions
21172 before 3.0, ``gnu-v3'', for @code{g++} versions 3.0 and later, and
21173 ``hpaCC'' for the HP ANSI C@t{++} compiler. Other C@t{++} compilers may
21174 use the ``gnu-v2'' or ``gnu-v3'' ABI's as well. The default setting is
21179 Show the C@t{++} ABI currently in use.
21182 With no argument, show the list of supported C@t{++} ABI's.
21184 @item set cp-abi @var{abi}
21185 @itemx set cp-abi auto
21186 Set the current C@t{++} ABI to @var{abi}, or return to automatic detection.
21190 @section Automatically loading associated files
21191 @cindex auto-loading
21193 @value{GDBN} sometimes reads files with commands and settings automatically,
21194 without being explicitly told so by the user. We call this feature
21195 @dfn{auto-loading}. While auto-loading is useful for automatically adapting
21196 @value{GDBN} to the needs of your project, it can sometimes produce unexpected
21197 results or introduce security risks (e.g., if the file comes from untrusted
21200 Note that loading of these associated files (including the local @file{.gdbinit}
21201 file) requires accordingly configured @code{auto-load safe-path}
21202 (@pxref{Auto-loading safe path}).
21204 For these reasons, @value{GDBN} includes commands and options to let you
21205 control when to auto-load files and which files should be auto-loaded.
21208 @anchor{set auto-load off}
21209 @kindex set auto-load off
21210 @item set auto-load off
21211 Globally disable loading of all auto-loaded files.
21212 You may want to use this command with the @samp{-iex} option
21213 (@pxref{Option -init-eval-command}) such as:
21215 $ @kbd{gdb -iex "set auto-load off" untrusted-executable corefile}
21218 Be aware that system init file (@pxref{System-wide configuration})
21219 and init files from your home directory (@pxref{Home Directory Init File})
21220 still get read (as they come from generally trusted directories).
21221 To prevent @value{GDBN} from auto-loading even those init files, use the
21222 @option{-nx} option (@pxref{Mode Options}), in addition to
21223 @code{set auto-load no}.
21225 @anchor{show auto-load}
21226 @kindex show auto-load
21227 @item show auto-load
21228 Show whether auto-loading of each specific @samp{auto-load} file(s) is enabled
21232 (gdb) show auto-load
21233 gdb-scripts: Auto-loading of canned sequences of commands scripts is on.
21234 libthread-db: Auto-loading of inferior specific libthread_db is on.
21235 local-gdbinit: Auto-loading of .gdbinit script from current directory
21237 python-scripts: Auto-loading of Python scripts is on.
21238 safe-path: List of directories from which it is safe to auto-load files
21239 is $debugdir:$datadir/auto-load.
21240 scripts-directory: List of directories from which to load auto-loaded scripts
21241 is $debugdir:$datadir/auto-load.
21244 @anchor{info auto-load}
21245 @kindex info auto-load
21246 @item info auto-load
21247 Print whether each specific @samp{auto-load} file(s) have been auto-loaded or
21251 (gdb) info auto-load
21254 Yes /home/user/gdb/gdb-gdb.gdb
21255 libthread-db: No auto-loaded libthread-db.
21256 local-gdbinit: Local .gdbinit file "/home/user/gdb/.gdbinit" has been
21260 Yes /home/user/gdb/gdb-gdb.py
21264 These are various kinds of files @value{GDBN} can automatically load:
21268 @xref{objfile-gdb.py file}, controlled by @ref{set auto-load python-scripts}.
21270 @xref{objfile-gdb.gdb file}, controlled by @ref{set auto-load gdb-scripts}.
21272 @xref{dotdebug_gdb_scripts section},
21273 controlled by @ref{set auto-load python-scripts}.
21275 @xref{Init File in the Current Directory},
21276 controlled by @ref{set auto-load local-gdbinit}.
21278 @xref{libthread_db.so.1 file}, controlled by @ref{set auto-load libthread-db}.
21281 These are @value{GDBN} control commands for the auto-loading:
21283 @multitable @columnfractions .5 .5
21284 @item @xref{set auto-load off}.
21285 @tab Disable auto-loading globally.
21286 @item @xref{show auto-load}.
21287 @tab Show setting of all kinds of files.
21288 @item @xref{info auto-load}.
21289 @tab Show state of all kinds of files.
21290 @item @xref{set auto-load gdb-scripts}.
21291 @tab Control for @value{GDBN} command scripts.
21292 @item @xref{show auto-load gdb-scripts}.
21293 @tab Show setting of @value{GDBN} command scripts.
21294 @item @xref{info auto-load gdb-scripts}.
21295 @tab Show state of @value{GDBN} command scripts.
21296 @item @xref{set auto-load python-scripts}.
21297 @tab Control for @value{GDBN} Python scripts.
21298 @item @xref{show auto-load python-scripts}.
21299 @tab Show setting of @value{GDBN} Python scripts.
21300 @item @xref{info auto-load python-scripts}.
21301 @tab Show state of @value{GDBN} Python scripts.
21302 @item @xref{set auto-load scripts-directory}.
21303 @tab Control for @value{GDBN} auto-loaded scripts location.
21304 @item @xref{show auto-load scripts-directory}.
21305 @tab Show @value{GDBN} auto-loaded scripts location.
21306 @item @xref{set auto-load local-gdbinit}.
21307 @tab Control for init file in the current directory.
21308 @item @xref{show auto-load local-gdbinit}.
21309 @tab Show setting of init file in the current directory.
21310 @item @xref{info auto-load local-gdbinit}.
21311 @tab Show state of init file in the current directory.
21312 @item @xref{set auto-load libthread-db}.
21313 @tab Control for thread debugging library.
21314 @item @xref{show auto-load libthread-db}.
21315 @tab Show setting of thread debugging library.
21316 @item @xref{info auto-load libthread-db}.
21317 @tab Show state of thread debugging library.
21318 @item @xref{set auto-load safe-path}.
21319 @tab Control directories trusted for automatic loading.
21320 @item @xref{show auto-load safe-path}.
21321 @tab Show directories trusted for automatic loading.
21322 @item @xref{add-auto-load-safe-path}.
21323 @tab Add directory trusted for automatic loading.
21327 * Init File in the Current Directory:: @samp{set/show/info auto-load local-gdbinit}
21328 * libthread_db.so.1 file:: @samp{set/show/info auto-load libthread-db}
21329 * objfile-gdb.gdb file:: @samp{set/show/info auto-load gdb-script}
21330 * Auto-loading safe path:: @samp{set/show/info auto-load safe-path}
21331 * Auto-loading verbose mode:: @samp{set/show debug auto-load}
21332 @xref{Python Auto-loading}.
21335 @node Init File in the Current Directory
21336 @subsection Automatically loading init file in the current directory
21337 @cindex auto-loading init file in the current directory
21339 By default, @value{GDBN} reads and executes the canned sequences of commands
21340 from init file (if any) in the current working directory,
21341 see @ref{Init File in the Current Directory during Startup}.
21343 Note that loading of this local @file{.gdbinit} file also requires accordingly
21344 configured @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
21347 @anchor{set auto-load local-gdbinit}
21348 @kindex set auto-load local-gdbinit
21349 @item set auto-load local-gdbinit [on|off]
21350 Enable or disable the auto-loading of canned sequences of commands
21351 (@pxref{Sequences}) found in init file in the current directory.
21353 @anchor{show auto-load local-gdbinit}
21354 @kindex show auto-load local-gdbinit
21355 @item show auto-load local-gdbinit
21356 Show whether auto-loading of canned sequences of commands from init file in the
21357 current directory is enabled or disabled.
21359 @anchor{info auto-load local-gdbinit}
21360 @kindex info auto-load local-gdbinit
21361 @item info auto-load local-gdbinit
21362 Print whether canned sequences of commands from init file in the
21363 current directory have been auto-loaded.
21366 @node libthread_db.so.1 file
21367 @subsection Automatically loading thread debugging library
21368 @cindex auto-loading libthread_db.so.1
21370 This feature is currently present only on @sc{gnu}/Linux native hosts.
21372 @value{GDBN} reads in some cases thread debugging library from places specific
21373 to the inferior (@pxref{set libthread-db-search-path}).
21375 The special @samp{libthread-db-search-path} entry @samp{$sdir} is processed
21376 without checking this @samp{set auto-load libthread-db} switch as system
21377 libraries have to be trusted in general. In all other cases of
21378 @samp{libthread-db-search-path} entries @value{GDBN} checks first if @samp{set
21379 auto-load libthread-db} is enabled before trying to open such thread debugging
21382 Note that loading of this debugging library also requires accordingly configured
21383 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
21386 @anchor{set auto-load libthread-db}
21387 @kindex set auto-load libthread-db
21388 @item set auto-load libthread-db [on|off]
21389 Enable or disable the auto-loading of inferior specific thread debugging library.
21391 @anchor{show auto-load libthread-db}
21392 @kindex show auto-load libthread-db
21393 @item show auto-load libthread-db
21394 Show whether auto-loading of inferior specific thread debugging library is
21395 enabled or disabled.
21397 @anchor{info auto-load libthread-db}
21398 @kindex info auto-load libthread-db
21399 @item info auto-load libthread-db
21400 Print the list of all loaded inferior specific thread debugging libraries and
21401 for each such library print list of inferior @var{pid}s using it.
21404 @node objfile-gdb.gdb file
21405 @subsection The @file{@var{objfile}-gdb.gdb} file
21406 @cindex auto-loading @file{@var{objfile}-gdb.gdb}
21408 @value{GDBN} tries to load an @file{@var{objfile}-gdb.gdb} file containing
21409 canned sequences of commands (@pxref{Sequences}), as long as @samp{set
21410 auto-load gdb-scripts} is set to @samp{on}.
21412 Note that loading of this script file also requires accordingly configured
21413 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
21415 For more background refer to the similar Python scripts auto-loading
21416 description (@pxref{objfile-gdb.py file}).
21419 @anchor{set auto-load gdb-scripts}
21420 @kindex set auto-load gdb-scripts
21421 @item set auto-load gdb-scripts [on|off]
21422 Enable or disable the auto-loading of canned sequences of commands scripts.
21424 @anchor{show auto-load gdb-scripts}
21425 @kindex show auto-load gdb-scripts
21426 @item show auto-load gdb-scripts
21427 Show whether auto-loading of canned sequences of commands scripts is enabled or
21430 @anchor{info auto-load gdb-scripts}
21431 @kindex info auto-load gdb-scripts
21432 @cindex print list of auto-loaded canned sequences of commands scripts
21433 @item info auto-load gdb-scripts [@var{regexp}]
21434 Print the list of all canned sequences of commands scripts that @value{GDBN}
21438 If @var{regexp} is supplied only canned sequences of commands scripts with
21439 matching names are printed.
21441 @node Auto-loading safe path
21442 @subsection Security restriction for auto-loading
21443 @cindex auto-loading safe-path
21445 As the files of inferior can come from untrusted source (such as submitted by
21446 an application user) @value{GDBN} does not always load any files automatically.
21447 @value{GDBN} provides the @samp{set auto-load safe-path} setting to list
21448 directories trusted for loading files not explicitly requested by user.
21450 If the path is not set properly you will see a warning and the file will not
21455 Reading symbols from /home/user/gdb/gdb...done.
21456 warning: File "/home/user/gdb/gdb-gdb.gdb" auto-loading has been
21457 declined by your `auto-load safe-path' set
21458 to "$debugdir:$datadir/auto-load".
21459 warning: File "/home/user/gdb/gdb-gdb.py" auto-loading has been
21460 declined by your `auto-load safe-path' set
21461 to "$debugdir:$datadir/auto-load".
21464 The list of trusted directories is controlled by the following commands:
21467 @anchor{set auto-load safe-path}
21468 @kindex set auto-load safe-path
21469 @item set auto-load safe-path @r{[}@var{directories}@r{]}
21470 Set the list of directories (and their subdirectories) trusted for automatic
21471 loading and execution of scripts. You can also enter a specific trusted file.
21472 If you omit @var{directories}, @samp{auto-load safe-path} will be reset to
21473 its default value as specified during @value{GDBN} compilation.
21475 The list of directories uses path separator (@samp{:} on GNU and Unix
21476 systems, @samp{;} on MS-Windows and MS-DOS) to separate directories, similarly
21477 to the @env{PATH} environment variable.
21479 @anchor{show auto-load safe-path}
21480 @kindex show auto-load safe-path
21481 @item show auto-load safe-path
21482 Show the list of directories trusted for automatic loading and execution of
21485 @anchor{add-auto-load-safe-path}
21486 @kindex add-auto-load-safe-path
21487 @item add-auto-load-safe-path
21488 Add an entry (or list of entries) the list of directories trusted for automatic
21489 loading and execution of scripts. Multiple entries may be delimited by the
21490 host platform path separator in use.
21493 This variable defaults to what @code{--with-auto-load-dir} has been configured
21494 to (@pxref{with-auto-load-dir}). @file{$debugdir} and @file{$datadir}
21495 substitution applies the same as for @ref{set auto-load scripts-directory}.
21496 The default @code{set auto-load safe-path} value can be also overriden by
21497 @value{GDBN} configuration option @option{--with-auto-load-safe-path}.
21499 Setting this variable to @file{/} disables this security protection,
21500 corresponding @value{GDBN} configuration option is
21501 @option{--without-auto-load-safe-path}.
21502 This variable is supposed to be set to the system directories writable by the
21503 system superuser only. Users can add their source directories in init files in
21504 their home directories (@pxref{Home Directory Init File}). See also deprecated
21505 init file in the current directory
21506 (@pxref{Init File in the Current Directory during Startup}).
21508 To force @value{GDBN} to load the files it declined to load in the previous
21509 example, you could use one of the following ways:
21512 @item @file{~/.gdbinit}: @samp{add-auto-load-safe-path ~/src/gdb}
21513 Specify this trusted directory (or a file) as additional component of the list.
21514 You have to specify also any existing directories displayed by
21515 by @samp{show auto-load safe-path} (such as @samp{/usr:/bin} in this example).
21517 @item @kbd{gdb -iex "set auto-load safe-path /usr:/bin:~/src/gdb" @dots{}}
21518 Specify this directory as in the previous case but just for a single
21519 @value{GDBN} session.
21521 @item @kbd{gdb -iex "set auto-load safe-path /" @dots{}}
21522 Disable auto-loading safety for a single @value{GDBN} session.
21523 This assumes all the files you debug during this @value{GDBN} session will come
21524 from trusted sources.
21526 @item @kbd{./configure --without-auto-load-safe-path}
21527 During compilation of @value{GDBN} you may disable any auto-loading safety.
21528 This assumes all the files you will ever debug with this @value{GDBN} come from
21532 On the other hand you can also explicitly forbid automatic files loading which
21533 also suppresses any such warning messages:
21536 @item @kbd{gdb -iex "set auto-load no" @dots{}}
21537 You can use @value{GDBN} command-line option for a single @value{GDBN} session.
21539 @item @file{~/.gdbinit}: @samp{set auto-load no}
21540 Disable auto-loading globally for the user
21541 (@pxref{Home Directory Init File}). While it is improbable, you could also
21542 use system init file instead (@pxref{System-wide configuration}).
21545 This setting applies to the file names as entered by user. If no entry matches
21546 @value{GDBN} tries as a last resort to also resolve all the file names into
21547 their canonical form (typically resolving symbolic links) and compare the
21548 entries again. @value{GDBN} already canonicalizes most of the filenames on its
21549 own before starting the comparison so a canonical form of directories is
21550 recommended to be entered.
21552 @node Auto-loading verbose mode
21553 @subsection Displaying files tried for auto-load
21554 @cindex auto-loading verbose mode
21556 For better visibility of all the file locations where you can place scripts to
21557 be auto-loaded with inferior --- or to protect yourself against accidental
21558 execution of untrusted scripts --- @value{GDBN} provides a feature for printing
21559 all the files attempted to be loaded. Both existing and non-existing files may
21562 For example the list of directories from which it is safe to auto-load files
21563 (@pxref{Auto-loading safe path}) applies also to canonicalized filenames which
21564 may not be too obvious while setting it up.
21567 (gdb) set debug auto-load on
21568 (gdb) file ~/src/t/true
21569 auto-load: Loading canned sequences of commands script "/tmp/true-gdb.gdb"
21570 for objfile "/tmp/true".
21571 auto-load: Updating directories of "/usr:/opt".
21572 auto-load: Using directory "/usr".
21573 auto-load: Using directory "/opt".
21574 warning: File "/tmp/true-gdb.gdb" auto-loading has been declined
21575 by your `auto-load safe-path' set to "/usr:/opt".
21579 @anchor{set debug auto-load}
21580 @kindex set debug auto-load
21581 @item set debug auto-load [on|off]
21582 Set whether to print the filenames attempted to be auto-loaded.
21584 @anchor{show debug auto-load}
21585 @kindex show debug auto-load
21586 @item show debug auto-load
21587 Show whether printing of the filenames attempted to be auto-loaded is turned
21591 @node Messages/Warnings
21592 @section Optional Warnings and Messages
21594 @cindex verbose operation
21595 @cindex optional warnings
21596 By default, @value{GDBN} is silent about its inner workings. If you are
21597 running on a slow machine, you may want to use the @code{set verbose}
21598 command. This makes @value{GDBN} tell you when it does a lengthy
21599 internal operation, so you will not think it has crashed.
21601 Currently, the messages controlled by @code{set verbose} are those
21602 which announce that the symbol table for a source file is being read;
21603 see @code{symbol-file} in @ref{Files, ,Commands to Specify Files}.
21606 @kindex set verbose
21607 @item set verbose on
21608 Enables @value{GDBN} output of certain informational messages.
21610 @item set verbose off
21611 Disables @value{GDBN} output of certain informational messages.
21613 @kindex show verbose
21615 Displays whether @code{set verbose} is on or off.
21618 By default, if @value{GDBN} encounters bugs in the symbol table of an
21619 object file, it is silent; but if you are debugging a compiler, you may
21620 find this information useful (@pxref{Symbol Errors, ,Errors Reading
21625 @kindex set complaints
21626 @item set complaints @var{limit}
21627 Permits @value{GDBN} to output @var{limit} complaints about each type of
21628 unusual symbols before becoming silent about the problem. Set
21629 @var{limit} to zero to suppress all complaints; set it to a large number
21630 to prevent complaints from being suppressed.
21632 @kindex show complaints
21633 @item show complaints
21634 Displays how many symbol complaints @value{GDBN} is permitted to produce.
21638 @anchor{confirmation requests}
21639 By default, @value{GDBN} is cautious, and asks what sometimes seems to be a
21640 lot of stupid questions to confirm certain commands. For example, if
21641 you try to run a program which is already running:
21645 The program being debugged has been started already.
21646 Start it from the beginning? (y or n)
21649 If you are willing to unflinchingly face the consequences of your own
21650 commands, you can disable this ``feature'':
21654 @kindex set confirm
21656 @cindex confirmation
21657 @cindex stupid questions
21658 @item set confirm off
21659 Disables confirmation requests. Note that running @value{GDBN} with
21660 the @option{--batch} option (@pxref{Mode Options, -batch}) also
21661 automatically disables confirmation requests.
21663 @item set confirm on
21664 Enables confirmation requests (the default).
21666 @kindex show confirm
21668 Displays state of confirmation requests.
21672 @cindex command tracing
21673 If you need to debug user-defined commands or sourced files you may find it
21674 useful to enable @dfn{command tracing}. In this mode each command will be
21675 printed as it is executed, prefixed with one or more @samp{+} symbols, the
21676 quantity denoting the call depth of each command.
21679 @kindex set trace-commands
21680 @cindex command scripts, debugging
21681 @item set trace-commands on
21682 Enable command tracing.
21683 @item set trace-commands off
21684 Disable command tracing.
21685 @item show trace-commands
21686 Display the current state of command tracing.
21689 @node Debugging Output
21690 @section Optional Messages about Internal Happenings
21691 @cindex optional debugging messages
21693 @value{GDBN} has commands that enable optional debugging messages from
21694 various @value{GDBN} subsystems; normally these commands are of
21695 interest to @value{GDBN} maintainers, or when reporting a bug. This
21696 section documents those commands.
21699 @kindex set exec-done-display
21700 @item set exec-done-display
21701 Turns on or off the notification of asynchronous commands'
21702 completion. When on, @value{GDBN} will print a message when an
21703 asynchronous command finishes its execution. The default is off.
21704 @kindex show exec-done-display
21705 @item show exec-done-display
21706 Displays the current setting of asynchronous command completion
21709 @cindex gdbarch debugging info
21710 @cindex architecture debugging info
21711 @item set debug arch
21712 Turns on or off display of gdbarch debugging info. The default is off
21714 @item show debug arch
21715 Displays the current state of displaying gdbarch debugging info.
21716 @item set debug aix-thread
21717 @cindex AIX threads
21718 Display debugging messages about inner workings of the AIX thread
21720 @item show debug aix-thread
21721 Show the current state of AIX thread debugging info display.
21722 @item set debug check-physname
21724 Check the results of the ``physname'' computation. When reading DWARF
21725 debugging information for C@t{++}, @value{GDBN} attempts to compute
21726 each entity's name. @value{GDBN} can do this computation in two
21727 different ways, depending on exactly what information is present.
21728 When enabled, this setting causes @value{GDBN} to compute the names
21729 both ways and display any discrepancies.
21730 @item show debug check-physname
21731 Show the current state of ``physname'' checking.
21732 @item set debug dwarf2-die
21733 @cindex DWARF2 DIEs
21734 Dump DWARF2 DIEs after they are read in.
21735 The value is the number of nesting levels to print.
21736 A value of zero turns off the display.
21737 @item show debug dwarf2-die
21738 Show the current state of DWARF2 DIE debugging.
21739 @item set debug displaced
21740 @cindex displaced stepping debugging info
21741 Turns on or off display of @value{GDBN} debugging info for the
21742 displaced stepping support. The default is off.
21743 @item show debug displaced
21744 Displays the current state of displaying @value{GDBN} debugging info
21745 related to displaced stepping.
21746 @item set debug event
21747 @cindex event debugging info
21748 Turns on or off display of @value{GDBN} event debugging info. The
21750 @item show debug event
21751 Displays the current state of displaying @value{GDBN} event debugging
21753 @item set debug expression
21754 @cindex expression debugging info
21755 Turns on or off display of debugging info about @value{GDBN}
21756 expression parsing. The default is off.
21757 @item show debug expression
21758 Displays the current state of displaying debugging info about
21759 @value{GDBN} expression parsing.
21760 @item set debug frame
21761 @cindex frame debugging info
21762 Turns on or off display of @value{GDBN} frame debugging info. The
21764 @item show debug frame
21765 Displays the current state of displaying @value{GDBN} frame debugging
21767 @item set debug gnu-nat
21768 @cindex @sc{gnu}/Hurd debug messages
21769 Turns on or off debugging messages from the @sc{gnu}/Hurd debug support.
21770 @item show debug gnu-nat
21771 Show the current state of @sc{gnu}/Hurd debugging messages.
21772 @item set debug infrun
21773 @cindex inferior debugging info
21774 Turns on or off display of @value{GDBN} debugging info for running the inferior.
21775 The default is off. @file{infrun.c} contains GDB's runtime state machine used
21776 for implementing operations such as single-stepping the inferior.
21777 @item show debug infrun
21778 Displays the current state of @value{GDBN} inferior debugging.
21779 @item set debug jit
21780 @cindex just-in-time compilation, debugging messages
21781 Turns on or off debugging messages from JIT debug support.
21782 @item show debug jit
21783 Displays the current state of @value{GDBN} JIT debugging.
21784 @item set debug lin-lwp
21785 @cindex @sc{gnu}/Linux LWP debug messages
21786 @cindex Linux lightweight processes
21787 Turns on or off debugging messages from the Linux LWP debug support.
21788 @item show debug lin-lwp
21789 Show the current state of Linux LWP debugging messages.
21790 @item set debug observer
21791 @cindex observer debugging info
21792 Turns on or off display of @value{GDBN} observer debugging. This
21793 includes info such as the notification of observable events.
21794 @item show debug observer
21795 Displays the current state of observer debugging.
21796 @item set debug overload
21797 @cindex C@t{++} overload debugging info
21798 Turns on or off display of @value{GDBN} C@t{++} overload debugging
21799 info. This includes info such as ranking of functions, etc. The default
21801 @item show debug overload
21802 Displays the current state of displaying @value{GDBN} C@t{++} overload
21804 @cindex expression parser, debugging info
21805 @cindex debug expression parser
21806 @item set debug parser
21807 Turns on or off the display of expression parser debugging output.
21808 Internally, this sets the @code{yydebug} variable in the expression
21809 parser. @xref{Tracing, , Tracing Your Parser, bison, Bison}, for
21810 details. The default is off.
21811 @item show debug parser
21812 Show the current state of expression parser debugging.
21813 @cindex packets, reporting on stdout
21814 @cindex serial connections, debugging
21815 @cindex debug remote protocol
21816 @cindex remote protocol debugging
21817 @cindex display remote packets
21818 @item set debug remote
21819 Turns on or off display of reports on all packets sent back and forth across
21820 the serial line to the remote machine. The info is printed on the
21821 @value{GDBN} standard output stream. The default is off.
21822 @item show debug remote
21823 Displays the state of display of remote packets.
21824 @item set debug serial
21825 Turns on or off display of @value{GDBN} serial debugging info. The
21827 @item show debug serial
21828 Displays the current state of displaying @value{GDBN} serial debugging
21830 @item set debug solib-frv
21831 @cindex FR-V shared-library debugging
21832 Turns on or off debugging messages for FR-V shared-library code.
21833 @item show debug solib-frv
21834 Display the current state of FR-V shared-library code debugging
21836 @item set debug target
21837 @cindex target debugging info
21838 Turns on or off display of @value{GDBN} target debugging info. This info
21839 includes what is going on at the target level of GDB, as it happens. The
21840 default is 0. Set it to 1 to track events, and to 2 to also track the
21841 value of large memory transfers. Changes to this flag do not take effect
21842 until the next time you connect to a target or use the @code{run} command.
21843 @item show debug target
21844 Displays the current state of displaying @value{GDBN} target debugging
21846 @item set debug timestamp
21847 @cindex timestampping debugging info
21848 Turns on or off display of timestamps with @value{GDBN} debugging info.
21849 When enabled, seconds and microseconds are displayed before each debugging
21851 @item show debug timestamp
21852 Displays the current state of displaying timestamps with @value{GDBN}
21854 @item set debugvarobj
21855 @cindex variable object debugging info
21856 Turns on or off display of @value{GDBN} variable object debugging
21857 info. The default is off.
21858 @item show debugvarobj
21859 Displays the current state of displaying @value{GDBN} variable object
21861 @item set debug xml
21862 @cindex XML parser debugging
21863 Turns on or off debugging messages for built-in XML parsers.
21864 @item show debug xml
21865 Displays the current state of XML debugging messages.
21868 @node Other Misc Settings
21869 @section Other Miscellaneous Settings
21870 @cindex miscellaneous settings
21873 @kindex set interactive-mode
21874 @item set interactive-mode
21875 If @code{on}, forces @value{GDBN} to assume that GDB was started
21876 in a terminal. In practice, this means that @value{GDBN} should wait
21877 for the user to answer queries generated by commands entered at
21878 the command prompt. If @code{off}, forces @value{GDBN} to operate
21879 in the opposite mode, and it uses the default answers to all queries.
21880 If @code{auto} (the default), @value{GDBN} tries to determine whether
21881 its standard input is a terminal, and works in interactive-mode if it
21882 is, non-interactively otherwise.
21884 In the vast majority of cases, the debugger should be able to guess
21885 correctly which mode should be used. But this setting can be useful
21886 in certain specific cases, such as running a MinGW @value{GDBN}
21887 inside a cygwin window.
21889 @kindex show interactive-mode
21890 @item show interactive-mode
21891 Displays whether the debugger is operating in interactive mode or not.
21894 @node Extending GDB
21895 @chapter Extending @value{GDBN}
21896 @cindex extending GDB
21898 @value{GDBN} provides three mechanisms for extension. The first is based
21899 on composition of @value{GDBN} commands, the second is based on the
21900 Python scripting language, and the third is for defining new aliases of
21903 To facilitate the use of the first two extensions, @value{GDBN} is capable
21904 of evaluating the contents of a file. When doing so, @value{GDBN}
21905 can recognize which scripting language is being used by looking at
21906 the filename extension. Files with an unrecognized filename extension
21907 are always treated as a @value{GDBN} Command Files.
21908 @xref{Command Files,, Command files}.
21910 You can control how @value{GDBN} evaluates these files with the following
21914 @kindex set script-extension
21915 @kindex show script-extension
21916 @item set script-extension off
21917 All scripts are always evaluated as @value{GDBN} Command Files.
21919 @item set script-extension soft
21920 The debugger determines the scripting language based on filename
21921 extension. If this scripting language is supported, @value{GDBN}
21922 evaluates the script using that language. Otherwise, it evaluates
21923 the file as a @value{GDBN} Command File.
21925 @item set script-extension strict
21926 The debugger determines the scripting language based on filename
21927 extension, and evaluates the script using that language. If the
21928 language is not supported, then the evaluation fails.
21930 @item show script-extension
21931 Display the current value of the @code{script-extension} option.
21936 * Sequences:: Canned Sequences of Commands
21937 * Python:: Scripting @value{GDBN} using Python
21938 * Aliases:: Creating new spellings of existing commands
21942 @section Canned Sequences of Commands
21944 Aside from breakpoint commands (@pxref{Break Commands, ,Breakpoint
21945 Command Lists}), @value{GDBN} provides two ways to store sequences of
21946 commands for execution as a unit: user-defined commands and command
21950 * Define:: How to define your own commands
21951 * Hooks:: Hooks for user-defined commands
21952 * Command Files:: How to write scripts of commands to be stored in a file
21953 * Output:: Commands for controlled output
21957 @subsection User-defined Commands
21959 @cindex user-defined command
21960 @cindex arguments, to user-defined commands
21961 A @dfn{user-defined command} is a sequence of @value{GDBN} commands to
21962 which you assign a new name as a command. This is done with the
21963 @code{define} command. User commands may accept up to 10 arguments
21964 separated by whitespace. Arguments are accessed within the user command
21965 via @code{$arg0@dots{}$arg9}. A trivial example:
21969 print $arg0 + $arg1 + $arg2
21974 To execute the command use:
21981 This defines the command @code{adder}, which prints the sum of
21982 its three arguments. Note the arguments are text substitutions, so they may
21983 reference variables, use complex expressions, or even perform inferior
21986 @cindex argument count in user-defined commands
21987 @cindex how many arguments (user-defined commands)
21988 In addition, @code{$argc} may be used to find out how many arguments have
21989 been passed. This expands to a number in the range 0@dots{}10.
21994 print $arg0 + $arg1
21997 print $arg0 + $arg1 + $arg2
22005 @item define @var{commandname}
22006 Define a command named @var{commandname}. If there is already a command
22007 by that name, you are asked to confirm that you want to redefine it.
22008 @var{commandname} may be a bare command name consisting of letters,
22009 numbers, dashes, and underscores. It may also start with any predefined
22010 prefix command. For example, @samp{define target my-target} creates
22011 a user-defined @samp{target my-target} command.
22013 The definition of the command is made up of other @value{GDBN} command lines,
22014 which are given following the @code{define} command. The end of these
22015 commands is marked by a line containing @code{end}.
22018 @kindex end@r{ (user-defined commands)}
22019 @item document @var{commandname}
22020 Document the user-defined command @var{commandname}, so that it can be
22021 accessed by @code{help}. The command @var{commandname} must already be
22022 defined. This command reads lines of documentation just as @code{define}
22023 reads the lines of the command definition, ending with @code{end}.
22024 After the @code{document} command is finished, @code{help} on command
22025 @var{commandname} displays the documentation you have written.
22027 You may use the @code{document} command again to change the
22028 documentation of a command. Redefining the command with @code{define}
22029 does not change the documentation.
22031 @kindex dont-repeat
22032 @cindex don't repeat command
22034 Used inside a user-defined command, this tells @value{GDBN} that this
22035 command should not be repeated when the user hits @key{RET}
22036 (@pxref{Command Syntax, repeat last command}).
22038 @kindex help user-defined
22039 @item help user-defined
22040 List all user-defined commands and all python commands defined in class
22041 COMAND_USER. The first line of the documentation or docstring is
22046 @itemx show user @var{commandname}
22047 Display the @value{GDBN} commands used to define @var{commandname} (but
22048 not its documentation). If no @var{commandname} is given, display the
22049 definitions for all user-defined commands.
22050 This does not work for user-defined python commands.
22052 @cindex infinite recursion in user-defined commands
22053 @kindex show max-user-call-depth
22054 @kindex set max-user-call-depth
22055 @item show max-user-call-depth
22056 @itemx set max-user-call-depth
22057 The value of @code{max-user-call-depth} controls how many recursion
22058 levels are allowed in user-defined commands before @value{GDBN} suspects an
22059 infinite recursion and aborts the command.
22060 This does not apply to user-defined python commands.
22063 In addition to the above commands, user-defined commands frequently
22064 use control flow commands, described in @ref{Command Files}.
22066 When user-defined commands are executed, the
22067 commands of the definition are not printed. An error in any command
22068 stops execution of the user-defined command.
22070 If used interactively, commands that would ask for confirmation proceed
22071 without asking when used inside a user-defined command. Many @value{GDBN}
22072 commands that normally print messages to say what they are doing omit the
22073 messages when used in a user-defined command.
22076 @subsection User-defined Command Hooks
22077 @cindex command hooks
22078 @cindex hooks, for commands
22079 @cindex hooks, pre-command
22082 You may define @dfn{hooks}, which are a special kind of user-defined
22083 command. Whenever you run the command @samp{foo}, if the user-defined
22084 command @samp{hook-foo} exists, it is executed (with no arguments)
22085 before that command.
22087 @cindex hooks, post-command
22089 A hook may also be defined which is run after the command you executed.
22090 Whenever you run the command @samp{foo}, if the user-defined command
22091 @samp{hookpost-foo} exists, it is executed (with no arguments) after
22092 that command. Post-execution hooks may exist simultaneously with
22093 pre-execution hooks, for the same command.
22095 It is valid for a hook to call the command which it hooks. If this
22096 occurs, the hook is not re-executed, thereby avoiding infinite recursion.
22098 @c It would be nice if hookpost could be passed a parameter indicating
22099 @c if the command it hooks executed properly or not. FIXME!
22101 @kindex stop@r{, a pseudo-command}
22102 In addition, a pseudo-command, @samp{stop} exists. Defining
22103 (@samp{hook-stop}) makes the associated commands execute every time
22104 execution stops in your program: before breakpoint commands are run,
22105 displays are printed, or the stack frame is printed.
22107 For example, to ignore @code{SIGALRM} signals while
22108 single-stepping, but treat them normally during normal execution,
22113 handle SIGALRM nopass
22117 handle SIGALRM pass
22120 define hook-continue
22121 handle SIGALRM pass
22125 As a further example, to hook at the beginning and end of the @code{echo}
22126 command, and to add extra text to the beginning and end of the message,
22134 define hookpost-echo
22138 (@value{GDBP}) echo Hello World
22139 <<<---Hello World--->>>
22144 You can define a hook for any single-word command in @value{GDBN}, but
22145 not for command aliases; you should define a hook for the basic command
22146 name, e.g.@: @code{backtrace} rather than @code{bt}.
22147 @c FIXME! So how does Joe User discover whether a command is an alias
22149 You can hook a multi-word command by adding @code{hook-} or
22150 @code{hookpost-} to the last word of the command, e.g.@:
22151 @samp{define target hook-remote} to add a hook to @samp{target remote}.
22153 If an error occurs during the execution of your hook, execution of
22154 @value{GDBN} commands stops and @value{GDBN} issues a prompt
22155 (before the command that you actually typed had a chance to run).
22157 If you try to define a hook which does not match any known command, you
22158 get a warning from the @code{define} command.
22160 @node Command Files
22161 @subsection Command Files
22163 @cindex command files
22164 @cindex scripting commands
22165 A command file for @value{GDBN} is a text file made of lines that are
22166 @value{GDBN} commands. Comments (lines starting with @kbd{#}) may
22167 also be included. An empty line in a command file does nothing; it
22168 does not mean to repeat the last command, as it would from the
22171 You can request the execution of a command file with the @code{source}
22172 command. Note that the @code{source} command is also used to evaluate
22173 scripts that are not Command Files. The exact behavior can be configured
22174 using the @code{script-extension} setting.
22175 @xref{Extending GDB,, Extending GDB}.
22179 @cindex execute commands from a file
22180 @item source [-s] [-v] @var{filename}
22181 Execute the command file @var{filename}.
22184 The lines in a command file are generally executed sequentially,
22185 unless the order of execution is changed by one of the
22186 @emph{flow-control commands} described below. The commands are not
22187 printed as they are executed. An error in any command terminates
22188 execution of the command file and control is returned to the console.
22190 @value{GDBN} first searches for @var{filename} in the current directory.
22191 If the file is not found there, and @var{filename} does not specify a
22192 directory, then @value{GDBN} also looks for the file on the source search path
22193 (specified with the @samp{directory} command);
22194 except that @file{$cdir} is not searched because the compilation directory
22195 is not relevant to scripts.
22197 If @code{-s} is specified, then @value{GDBN} searches for @var{filename}
22198 on the search path even if @var{filename} specifies a directory.
22199 The search is done by appending @var{filename} to each element of the
22200 search path. So, for example, if @var{filename} is @file{mylib/myscript}
22201 and the search path contains @file{/home/user} then @value{GDBN} will
22202 look for the script @file{/home/user/mylib/myscript}.
22203 The search is also done if @var{filename} is an absolute path.
22204 For example, if @var{filename} is @file{/tmp/myscript} and
22205 the search path contains @file{/home/user} then @value{GDBN} will
22206 look for the script @file{/home/user/tmp/myscript}.
22207 For DOS-like systems, if @var{filename} contains a drive specification,
22208 it is stripped before concatenation. For example, if @var{filename} is
22209 @file{d:myscript} and the search path contains @file{c:/tmp} then @value{GDBN}
22210 will look for the script @file{c:/tmp/myscript}.
22212 If @code{-v}, for verbose mode, is given then @value{GDBN} displays
22213 each command as it is executed. The option must be given before
22214 @var{filename}, and is interpreted as part of the filename anywhere else.
22216 Commands that would ask for confirmation if used interactively proceed
22217 without asking when used in a command file. Many @value{GDBN} commands that
22218 normally print messages to say what they are doing omit the messages
22219 when called from command files.
22221 @value{GDBN} also accepts command input from standard input. In this
22222 mode, normal output goes to standard output and error output goes to
22223 standard error. Errors in a command file supplied on standard input do
22224 not terminate execution of the command file---execution continues with
22228 gdb < cmds > log 2>&1
22231 (The syntax above will vary depending on the shell used.) This example
22232 will execute commands from the file @file{cmds}. All output and errors
22233 would be directed to @file{log}.
22235 Since commands stored on command files tend to be more general than
22236 commands typed interactively, they frequently need to deal with
22237 complicated situations, such as different or unexpected values of
22238 variables and symbols, changes in how the program being debugged is
22239 built, etc. @value{GDBN} provides a set of flow-control commands to
22240 deal with these complexities. Using these commands, you can write
22241 complex scripts that loop over data structures, execute commands
22242 conditionally, etc.
22249 This command allows to include in your script conditionally executed
22250 commands. The @code{if} command takes a single argument, which is an
22251 expression to evaluate. It is followed by a series of commands that
22252 are executed only if the expression is true (its value is nonzero).
22253 There can then optionally be an @code{else} line, followed by a series
22254 of commands that are only executed if the expression was false. The
22255 end of the list is marked by a line containing @code{end}.
22259 This command allows to write loops. Its syntax is similar to
22260 @code{if}: the command takes a single argument, which is an expression
22261 to evaluate, and must be followed by the commands to execute, one per
22262 line, terminated by an @code{end}. These commands are called the
22263 @dfn{body} of the loop. The commands in the body of @code{while} are
22264 executed repeatedly as long as the expression evaluates to true.
22268 This command exits the @code{while} loop in whose body it is included.
22269 Execution of the script continues after that @code{while}s @code{end}
22272 @kindex loop_continue
22273 @item loop_continue
22274 This command skips the execution of the rest of the body of commands
22275 in the @code{while} loop in whose body it is included. Execution
22276 branches to the beginning of the @code{while} loop, where it evaluates
22277 the controlling expression.
22279 @kindex end@r{ (if/else/while commands)}
22281 Terminate the block of commands that are the body of @code{if},
22282 @code{else}, or @code{while} flow-control commands.
22287 @subsection Commands for Controlled Output
22289 During the execution of a command file or a user-defined command, normal
22290 @value{GDBN} output is suppressed; the only output that appears is what is
22291 explicitly printed by the commands in the definition. This section
22292 describes three commands useful for generating exactly the output you
22297 @item echo @var{text}
22298 @c I do not consider backslash-space a standard C escape sequence
22299 @c because it is not in ANSI.
22300 Print @var{text}. Nonprinting characters can be included in
22301 @var{text} using C escape sequences, such as @samp{\n} to print a
22302 newline. @strong{No newline is printed unless you specify one.}
22303 In addition to the standard C escape sequences, a backslash followed
22304 by a space stands for a space. This is useful for displaying a
22305 string with spaces at the beginning or the end, since leading and
22306 trailing spaces are otherwise trimmed from all arguments.
22307 To print @samp{@w{ }and foo =@w{ }}, use the command
22308 @samp{echo \@w{ }and foo = \@w{ }}.
22310 A backslash at the end of @var{text} can be used, as in C, to continue
22311 the command onto subsequent lines. For example,
22314 echo This is some text\n\
22315 which is continued\n\
22316 onto several lines.\n
22319 produces the same output as
22322 echo This is some text\n
22323 echo which is continued\n
22324 echo onto several lines.\n
22328 @item output @var{expression}
22329 Print the value of @var{expression} and nothing but that value: no
22330 newlines, no @samp{$@var{nn} = }. The value is not entered in the
22331 value history either. @xref{Expressions, ,Expressions}, for more information
22334 @item output/@var{fmt} @var{expression}
22335 Print the value of @var{expression} in format @var{fmt}. You can use
22336 the same formats as for @code{print}. @xref{Output Formats,,Output
22337 Formats}, for more information.
22340 @item printf @var{template}, @var{expressions}@dots{}
22341 Print the values of one or more @var{expressions} under the control of
22342 the string @var{template}. To print several values, make
22343 @var{expressions} be a comma-separated list of individual expressions,
22344 which may be either numbers or pointers. Their values are printed as
22345 specified by @var{template}, exactly as a C program would do by
22346 executing the code below:
22349 printf (@var{template}, @var{expressions}@dots{});
22352 As in @code{C} @code{printf}, ordinary characters in @var{template}
22353 are printed verbatim, while @dfn{conversion specification} introduced
22354 by the @samp{%} character cause subsequent @var{expressions} to be
22355 evaluated, their values converted and formatted according to type and
22356 style information encoded in the conversion specifications, and then
22359 For example, you can print two values in hex like this:
22362 printf "foo, bar-foo = 0x%x, 0x%x\n", foo, bar-foo
22365 @code{printf} supports all the standard @code{C} conversion
22366 specifications, including the flags and modifiers between the @samp{%}
22367 character and the conversion letter, with the following exceptions:
22371 The argument-ordering modifiers, such as @samp{2$}, are not supported.
22374 The modifier @samp{*} is not supported for specifying precision or
22378 The @samp{'} flag (for separation of digits into groups according to
22379 @code{LC_NUMERIC'}) is not supported.
22382 The type modifiers @samp{hh}, @samp{j}, @samp{t}, and @samp{z} are not
22386 The conversion letter @samp{n} (as in @samp{%n}) is not supported.
22389 The conversion letters @samp{a} and @samp{A} are not supported.
22393 Note that the @samp{ll} type modifier is supported only if the
22394 underlying @code{C} implementation used to build @value{GDBN} supports
22395 the @code{long long int} type, and the @samp{L} type modifier is
22396 supported only if @code{long double} type is available.
22398 As in @code{C}, @code{printf} supports simple backslash-escape
22399 sequences, such as @code{\n}, @samp{\t}, @samp{\\}, @samp{\"},
22400 @samp{\a}, and @samp{\f}, that consist of backslash followed by a
22401 single character. Octal and hexadecimal escape sequences are not
22404 Additionally, @code{printf} supports conversion specifications for DFP
22405 (@dfn{Decimal Floating Point}) types using the following length modifiers
22406 together with a floating point specifier.
22411 @samp{H} for printing @code{Decimal32} types.
22414 @samp{D} for printing @code{Decimal64} types.
22417 @samp{DD} for printing @code{Decimal128} types.
22420 If the underlying @code{C} implementation used to build @value{GDBN} has
22421 support for the three length modifiers for DFP types, other modifiers
22422 such as width and precision will also be available for @value{GDBN} to use.
22424 In case there is no such @code{C} support, no additional modifiers will be
22425 available and the value will be printed in the standard way.
22427 Here's an example of printing DFP types using the above conversion letters:
22429 printf "D32: %Hf - D64: %Df - D128: %DDf\n",1.2345df,1.2E10dd,1.2E1dl
22433 @item eval @var{template}, @var{expressions}@dots{}
22434 Convert the values of one or more @var{expressions} under the control of
22435 the string @var{template} to a command line, and call it.
22440 @section Scripting @value{GDBN} using Python
22441 @cindex python scripting
22442 @cindex scripting with python
22444 You can script @value{GDBN} using the @uref{http://www.python.org/,
22445 Python programming language}. This feature is available only if
22446 @value{GDBN} was configured using @option{--with-python}.
22448 @cindex python directory
22449 Python scripts used by @value{GDBN} should be installed in
22450 @file{@var{data-directory}/python}, where @var{data-directory} is
22451 the data directory as determined at @value{GDBN} startup (@pxref{Data Files}).
22452 This directory, known as the @dfn{python directory},
22453 is automatically added to the Python Search Path in order to allow
22454 the Python interpreter to locate all scripts installed at this location.
22456 Additionally, @value{GDBN} commands and convenience functions which
22457 are written in Python and are located in the
22458 @file{@var{data-directory}/python/gdb/command} or
22459 @file{@var{data-directory}/python/gdb/function} directories are
22460 automatically imported when @value{GDBN} starts.
22463 * Python Commands:: Accessing Python from @value{GDBN}.
22464 * Python API:: Accessing @value{GDBN} from Python.
22465 * Python Auto-loading:: Automatically loading Python code.
22466 * Python modules:: Python modules provided by @value{GDBN}.
22469 @node Python Commands
22470 @subsection Python Commands
22471 @cindex python commands
22472 @cindex commands to access python
22474 @value{GDBN} provides one command for accessing the Python interpreter,
22475 and one related setting:
22479 @item python @r{[}@var{code}@r{]}
22480 The @code{python} command can be used to evaluate Python code.
22482 If given an argument, the @code{python} command will evaluate the
22483 argument as a Python command. For example:
22486 (@value{GDBP}) python print 23
22490 If you do not provide an argument to @code{python}, it will act as a
22491 multi-line command, like @code{define}. In this case, the Python
22492 script is made up of subsequent command lines, given after the
22493 @code{python} command. This command list is terminated using a line
22494 containing @code{end}. For example:
22497 (@value{GDBP}) python
22499 End with a line saying just "end".
22505 @kindex set python print-stack
22506 @item set python print-stack
22507 By default, @value{GDBN} will print only the message component of a
22508 Python exception when an error occurs in a Python script. This can be
22509 controlled using @code{set python print-stack}: if @code{full}, then
22510 full Python stack printing is enabled; if @code{none}, then Python stack
22511 and message printing is disabled; if @code{message}, the default, only
22512 the message component of the error is printed.
22515 It is also possible to execute a Python script from the @value{GDBN}
22519 @item source @file{script-name}
22520 The script name must end with @samp{.py} and @value{GDBN} must be configured
22521 to recognize the script language based on filename extension using
22522 the @code{script-extension} setting. @xref{Extending GDB, ,Extending GDB}.
22524 @item python execfile ("script-name")
22525 This method is based on the @code{execfile} Python built-in function,
22526 and thus is always available.
22530 @subsection Python API
22532 @cindex programming in python
22534 @cindex python stdout
22535 @cindex python pagination
22536 At startup, @value{GDBN} overrides Python's @code{sys.stdout} and
22537 @code{sys.stderr} to print using @value{GDBN}'s output-paging streams.
22538 A Python program which outputs to one of these streams may have its
22539 output interrupted by the user (@pxref{Screen Size}). In this
22540 situation, a Python @code{KeyboardInterrupt} exception is thrown.
22543 * Basic Python:: Basic Python Functions.
22544 * Exception Handling:: How Python exceptions are translated.
22545 * Values From Inferior:: Python representation of values.
22546 * Types In Python:: Python representation of types.
22547 * Pretty Printing API:: Pretty-printing values.
22548 * Selecting Pretty-Printers:: How GDB chooses a pretty-printer.
22549 * Writing a Pretty-Printer:: Writing a Pretty-Printer.
22550 * Inferiors In Python:: Python representation of inferiors (processes)
22551 * Events In Python:: Listening for events from @value{GDBN}.
22552 * Threads In Python:: Accessing inferior threads from Python.
22553 * Commands In Python:: Implementing new commands in Python.
22554 * Parameters In Python:: Adding new @value{GDBN} parameters.
22555 * Functions In Python:: Writing new convenience functions.
22556 * Progspaces In Python:: Program spaces.
22557 * Objfiles In Python:: Object files.
22558 * Frames In Python:: Accessing inferior stack frames from Python.
22559 * Blocks In Python:: Accessing frame blocks from Python.
22560 * Symbols In Python:: Python representation of symbols.
22561 * Symbol Tables In Python:: Python representation of symbol tables.
22562 * Lazy Strings In Python:: Python representation of lazy strings.
22563 * Breakpoints In Python:: Manipulating breakpoints using Python.
22564 * Finish Breakpoints in Python:: Setting Breakpoints on function return
22569 @subsubsection Basic Python
22571 @cindex python functions
22572 @cindex python module
22574 @value{GDBN} introduces a new Python module, named @code{gdb}. All
22575 methods and classes added by @value{GDBN} are placed in this module.
22576 @value{GDBN} automatically @code{import}s the @code{gdb} module for
22577 use in all scripts evaluated by the @code{python} command.
22579 @findex gdb.PYTHONDIR
22580 @defvar gdb.PYTHONDIR
22581 A string containing the python directory (@pxref{Python}).
22584 @findex gdb.execute
22585 @defun gdb.execute (command @r{[}, from_tty @r{[}, to_string@r{]]})
22586 Evaluate @var{command}, a string, as a @value{GDBN} CLI command.
22587 If a GDB exception happens while @var{command} runs, it is
22588 translated as described in @ref{Exception Handling,,Exception Handling}.
22590 @var{from_tty} specifies whether @value{GDBN} ought to consider this
22591 command as having originated from the user invoking it interactively.
22592 It must be a boolean value. If omitted, it defaults to @code{False}.
22594 By default, any output produced by @var{command} is sent to
22595 @value{GDBN}'s standard output. If the @var{to_string} parameter is
22596 @code{True}, then output will be collected by @code{gdb.execute} and
22597 returned as a string. The default is @code{False}, in which case the
22598 return value is @code{None}. If @var{to_string} is @code{True}, the
22599 @value{GDBN} virtual terminal will be temporarily set to unlimited width
22600 and height, and its pagination will be disabled; @pxref{Screen Size}.
22603 @findex gdb.breakpoints
22604 @defun gdb.breakpoints ()
22605 Return a sequence holding all of @value{GDBN}'s breakpoints.
22606 @xref{Breakpoints In Python}, for more information.
22609 @findex gdb.parameter
22610 @defun gdb.parameter (parameter)
22611 Return the value of a @value{GDBN} parameter. @var{parameter} is a
22612 string naming the parameter to look up; @var{parameter} may contain
22613 spaces if the parameter has a multi-part name. For example,
22614 @samp{print object} is a valid parameter name.
22616 If the named parameter does not exist, this function throws a
22617 @code{gdb.error} (@pxref{Exception Handling}). Otherwise, the
22618 parameter's value is converted to a Python value of the appropriate
22619 type, and returned.
22622 @findex gdb.history
22623 @defun gdb.history (number)
22624 Return a value from @value{GDBN}'s value history (@pxref{Value
22625 History}). @var{number} indicates which history element to return.
22626 If @var{number} is negative, then @value{GDBN} will take its absolute value
22627 and count backward from the last element (i.e., the most recent element) to
22628 find the value to return. If @var{number} is zero, then @value{GDBN} will
22629 return the most recent element. If the element specified by @var{number}
22630 doesn't exist in the value history, a @code{gdb.error} exception will be
22633 If no exception is raised, the return value is always an instance of
22634 @code{gdb.Value} (@pxref{Values From Inferior}).
22637 @findex gdb.parse_and_eval
22638 @defun gdb.parse_and_eval (expression)
22639 Parse @var{expression} as an expression in the current language,
22640 evaluate it, and return the result as a @code{gdb.Value}.
22641 @var{expression} must be a string.
22643 This function can be useful when implementing a new command
22644 (@pxref{Commands In Python}), as it provides a way to parse the
22645 command's argument as an expression. It is also useful simply to
22646 compute values, for example, it is the only way to get the value of a
22647 convenience variable (@pxref{Convenience Vars}) as a @code{gdb.Value}.
22650 @findex gdb.find_pc_line
22651 @defun gdb.find_pc_line (pc)
22652 Return the @code{gdb.Symtab_and_line} object corresponding to the
22653 @var{pc} value. @xref{Symbol Tables In Python}. If an invalid
22654 value of @var{pc} is passed as an argument, then the @code{symtab} and
22655 @code{line} attributes of the returned @code{gdb.Symtab_and_line} object
22656 will be @code{None} and 0 respectively.
22659 @findex gdb.post_event
22660 @defun gdb.post_event (event)
22661 Put @var{event}, a callable object taking no arguments, into
22662 @value{GDBN}'s internal event queue. This callable will be invoked at
22663 some later point, during @value{GDBN}'s event processing. Events
22664 posted using @code{post_event} will be run in the order in which they
22665 were posted; however, there is no way to know when they will be
22666 processed relative to other events inside @value{GDBN}.
22668 @value{GDBN} is not thread-safe. If your Python program uses multiple
22669 threads, you must be careful to only call @value{GDBN}-specific
22670 functions in the main @value{GDBN} thread. @code{post_event} ensures
22674 (@value{GDBP}) python
22678 > def __init__(self, message):
22679 > self.message = message;
22680 > def __call__(self):
22681 > gdb.write(self.message)
22683 >class MyThread1 (threading.Thread):
22685 > gdb.post_event(Writer("Hello "))
22687 >class MyThread2 (threading.Thread):
22689 > gdb.post_event(Writer("World\n"))
22691 >MyThread1().start()
22692 >MyThread2().start()
22694 (@value{GDBP}) Hello World
22699 @defun gdb.write (string @r{[}, stream{]})
22700 Print a string to @value{GDBN}'s paginated output stream. The
22701 optional @var{stream} determines the stream to print to. The default
22702 stream is @value{GDBN}'s standard output stream. Possible stream
22709 @value{GDBN}'s standard output stream.
22714 @value{GDBN}'s standard error stream.
22719 @value{GDBN}'s log stream (@pxref{Logging Output}).
22722 Writing to @code{sys.stdout} or @code{sys.stderr} will automatically
22723 call this function and will automatically direct the output to the
22728 @defun gdb.flush ()
22729 Flush the buffer of a @value{GDBN} paginated stream so that the
22730 contents are displayed immediately. @value{GDBN} will flush the
22731 contents of a stream automatically when it encounters a newline in the
22732 buffer. The optional @var{stream} determines the stream to flush. The
22733 default stream is @value{GDBN}'s standard output stream. Possible
22740 @value{GDBN}'s standard output stream.
22745 @value{GDBN}'s standard error stream.
22750 @value{GDBN}'s log stream (@pxref{Logging Output}).
22754 Flushing @code{sys.stdout} or @code{sys.stderr} will automatically
22755 call this function for the relevant stream.
22758 @findex gdb.target_charset
22759 @defun gdb.target_charset ()
22760 Return the name of the current target character set (@pxref{Character
22761 Sets}). This differs from @code{gdb.parameter('target-charset')} in
22762 that @samp{auto} is never returned.
22765 @findex gdb.target_wide_charset
22766 @defun gdb.target_wide_charset ()
22767 Return the name of the current target wide character set
22768 (@pxref{Character Sets}). This differs from
22769 @code{gdb.parameter('target-wide-charset')} in that @samp{auto} is
22773 @findex gdb.solib_name
22774 @defun gdb.solib_name (address)
22775 Return the name of the shared library holding the given @var{address}
22776 as a string, or @code{None}.
22779 @findex gdb.decode_line
22780 @defun gdb.decode_line @r{[}expression@r{]}
22781 Return locations of the line specified by @var{expression}, or of the
22782 current line if no argument was given. This function returns a Python
22783 tuple containing two elements. The first element contains a string
22784 holding any unparsed section of @var{expression} (or @code{None} if
22785 the expression has been fully parsed). The second element contains
22786 either @code{None} or another tuple that contains all the locations
22787 that match the expression represented as @code{gdb.Symtab_and_line}
22788 objects (@pxref{Symbol Tables In Python}). If @var{expression} is
22789 provided, it is decoded the way that @value{GDBN}'s inbuilt
22790 @code{break} or @code{edit} commands do (@pxref{Specify Location}).
22793 @defun gdb.prompt_hook (current_prompt)
22794 @anchor{prompt_hook}
22796 If @var{prompt_hook} is callable, @value{GDBN} will call the method
22797 assigned to this operation before a prompt is displayed by
22800 The parameter @code{current_prompt} contains the current @value{GDBN}
22801 prompt. This method must return a Python string, or @code{None}. If
22802 a string is returned, the @value{GDBN} prompt will be set to that
22803 string. If @code{None} is returned, @value{GDBN} will continue to use
22804 the current prompt.
22806 Some prompts cannot be substituted in @value{GDBN}. Secondary prompts
22807 such as those used by readline for command input, and annotation
22808 related prompts are prohibited from being changed.
22811 @node Exception Handling
22812 @subsubsection Exception Handling
22813 @cindex python exceptions
22814 @cindex exceptions, python
22816 When executing the @code{python} command, Python exceptions
22817 uncaught within the Python code are translated to calls to
22818 @value{GDBN} error-reporting mechanism. If the command that called
22819 @code{python} does not handle the error, @value{GDBN} will
22820 terminate it and print an error message containing the Python
22821 exception name, the associated value, and the Python call stack
22822 backtrace at the point where the exception was raised. Example:
22825 (@value{GDBP}) python print foo
22826 Traceback (most recent call last):
22827 File "<string>", line 1, in <module>
22828 NameError: name 'foo' is not defined
22831 @value{GDBN} errors that happen in @value{GDBN} commands invoked by
22832 Python code are converted to Python exceptions. The type of the
22833 Python exception depends on the error.
22837 This is the base class for most exceptions generated by @value{GDBN}.
22838 It is derived from @code{RuntimeError}, for compatibility with earlier
22839 versions of @value{GDBN}.
22841 If an error occurring in @value{GDBN} does not fit into some more
22842 specific category, then the generated exception will have this type.
22844 @item gdb.MemoryError
22845 This is a subclass of @code{gdb.error} which is thrown when an
22846 operation tried to access invalid memory in the inferior.
22848 @item KeyboardInterrupt
22849 User interrupt (via @kbd{C-c} or by typing @kbd{q} at a pagination
22850 prompt) is translated to a Python @code{KeyboardInterrupt} exception.
22853 In all cases, your exception handler will see the @value{GDBN} error
22854 message as its value and the Python call stack backtrace at the Python
22855 statement closest to where the @value{GDBN} error occured as the
22858 @findex gdb.GdbError
22859 When implementing @value{GDBN} commands in Python via @code{gdb.Command},
22860 it is useful to be able to throw an exception that doesn't cause a
22861 traceback to be printed. For example, the user may have invoked the
22862 command incorrectly. Use the @code{gdb.GdbError} exception
22863 to handle this case. Example:
22867 >class HelloWorld (gdb.Command):
22868 > """Greet the whole world."""
22869 > def __init__ (self):
22870 > super (HelloWorld, self).__init__ ("hello-world", gdb.COMMAND_USER)
22871 > def invoke (self, args, from_tty):
22872 > argv = gdb.string_to_argv (args)
22873 > if len (argv) != 0:
22874 > raise gdb.GdbError ("hello-world takes no arguments")
22875 > print "Hello, World!"
22878 (gdb) hello-world 42
22879 hello-world takes no arguments
22882 @node Values From Inferior
22883 @subsubsection Values From Inferior
22884 @cindex values from inferior, with Python
22885 @cindex python, working with values from inferior
22887 @cindex @code{gdb.Value}
22888 @value{GDBN} provides values it obtains from the inferior program in
22889 an object of type @code{gdb.Value}. @value{GDBN} uses this object
22890 for its internal bookkeeping of the inferior's values, and for
22891 fetching values when necessary.
22893 Inferior values that are simple scalars can be used directly in
22894 Python expressions that are valid for the value's data type. Here's
22895 an example for an integer or floating-point value @code{some_val}:
22902 As result of this, @code{bar} will also be a @code{gdb.Value} object
22903 whose values are of the same type as those of @code{some_val}.
22905 Inferior values that are structures or instances of some class can
22906 be accessed using the Python @dfn{dictionary syntax}. For example, if
22907 @code{some_val} is a @code{gdb.Value} instance holding a structure, you
22908 can access its @code{foo} element with:
22911 bar = some_val['foo']
22914 Again, @code{bar} will also be a @code{gdb.Value} object.
22916 A @code{gdb.Value} that represents a function can be executed via
22917 inferior function call. Any arguments provided to the call must match
22918 the function's prototype, and must be provided in the order specified
22921 For example, @code{some_val} is a @code{gdb.Value} instance
22922 representing a function that takes two integers as arguments. To
22923 execute this function, call it like so:
22926 result = some_val (10,20)
22929 Any values returned from a function call will be stored as a
22932 The following attributes are provided:
22935 @defvar Value.address
22936 If this object is addressable, this read-only attribute holds a
22937 @code{gdb.Value} object representing the address. Otherwise,
22938 this attribute holds @code{None}.
22941 @cindex optimized out value in Python
22942 @defvar Value.is_optimized_out
22943 This read-only boolean attribute is true if the compiler optimized out
22944 this value, thus it is not available for fetching from the inferior.
22948 The type of this @code{gdb.Value}. The value of this attribute is a
22949 @code{gdb.Type} object (@pxref{Types In Python}).
22952 @defvar Value.dynamic_type
22953 The dynamic type of this @code{gdb.Value}. This uses C@t{++} run-time
22954 type information (@acronym{RTTI}) to determine the dynamic type of the
22955 value. If this value is of class type, it will return the class in
22956 which the value is embedded, if any. If this value is of pointer or
22957 reference to a class type, it will compute the dynamic type of the
22958 referenced object, and return a pointer or reference to that type,
22959 respectively. In all other cases, it will return the value's static
22962 Note that this feature will only work when debugging a C@t{++} program
22963 that includes @acronym{RTTI} for the object in question. Otherwise,
22964 it will just return the static type of the value as in @kbd{ptype foo}
22965 (@pxref{Symbols, ptype}).
22968 @defvar Value.is_lazy
22969 The value of this read-only boolean attribute is @code{True} if this
22970 @code{gdb.Value} has not yet been fetched from the inferior.
22971 @value{GDBN} does not fetch values until necessary, for efficiency.
22975 myval = gdb.parse_and_eval ('somevar')
22978 The value of @code{somevar} is not fetched at this time. It will be
22979 fetched when the value is needed, or when the @code{fetch_lazy}
22984 The following methods are provided:
22987 @defun Value.__init__ (@var{val})
22988 Many Python values can be converted directly to a @code{gdb.Value} via
22989 this object initializer. Specifically:
22992 @item Python boolean
22993 A Python boolean is converted to the boolean type from the current
22996 @item Python integer
22997 A Python integer is converted to the C @code{long} type for the
22998 current architecture.
23001 A Python long is converted to the C @code{long long} type for the
23002 current architecture.
23005 A Python float is converted to the C @code{double} type for the
23006 current architecture.
23008 @item Python string
23009 A Python string is converted to a target string, using the current
23012 @item @code{gdb.Value}
23013 If @code{val} is a @code{gdb.Value}, then a copy of the value is made.
23015 @item @code{gdb.LazyString}
23016 If @code{val} is a @code{gdb.LazyString} (@pxref{Lazy Strings In
23017 Python}), then the lazy string's @code{value} method is called, and
23018 its result is used.
23022 @defun Value.cast (type)
23023 Return a new instance of @code{gdb.Value} that is the result of
23024 casting this instance to the type described by @var{type}, which must
23025 be a @code{gdb.Type} object. If the cast cannot be performed for some
23026 reason, this method throws an exception.
23029 @defun Value.dereference ()
23030 For pointer data types, this method returns a new @code{gdb.Value} object
23031 whose contents is the object pointed to by the pointer. For example, if
23032 @code{foo} is a C pointer to an @code{int}, declared in your C program as
23039 then you can use the corresponding @code{gdb.Value} to access what
23040 @code{foo} points to like this:
23043 bar = foo.dereference ()
23046 The result @code{bar} will be a @code{gdb.Value} object holding the
23047 value pointed to by @code{foo}.
23049 A similar function @code{Value.referenced_value} exists which also
23050 returns @code{gdb.Value} objects corresonding to the values pointed to
23051 by pointer values (and additionally, values referenced by reference
23052 values). However, the behavior of @code{Value.dereference}
23053 differs from @code{Value.referenced_value} by the fact that the
23054 behavior of @code{Value.dereference} is identical to applying the C
23055 unary operator @code{*} on a given value. For example, consider a
23056 reference to a pointer @code{ptrref}, declared in your C@t{++} program
23060 typedef int *intptr;
23064 intptr &ptrref = ptr;
23067 Though @code{ptrref} is a reference value, one can apply the method
23068 @code{Value.dereference} to the @code{gdb.Value} object corresponding
23069 to it and obtain a @code{gdb.Value} which is identical to that
23070 corresponding to @code{val}. However, if you apply the method
23071 @code{Value.referenced_value}, the result would be a @code{gdb.Value}
23072 object identical to that corresponding to @code{ptr}.
23075 py_ptrref = gdb.parse_and_eval ("ptrref")
23076 py_val = py_ptrref.dereference ()
23077 py_ptr = py_ptrref.referenced_value ()
23080 The @code{gdb.Value} object @code{py_val} is identical to that
23081 corresponding to @code{val}, and @code{py_ptr} is identical to that
23082 corresponding to @code{ptr}. In general, @code{Value.dereference} can
23083 be applied whenever the C unary operator @code{*} can be applied
23084 to the corresponding C value. For those cases where applying both
23085 @code{Value.dereference} and @code{Value.referenced_value} is allowed,
23086 the results obtained need not be identical (as we have seen in the above
23087 example). The results are however identical when applied on
23088 @code{gdb.Value} objects corresponding to pointers (@code{gdb.Value}
23089 objects with type code @code{TYPE_CODE_PTR}) in a C/C@t{++} program.
23092 @defun Value.referenced_value ()
23093 For pointer or reference data types, this method returns a new
23094 @code{gdb.Value} object corresponding to the value referenced by the
23095 pointer/reference value. For pointer data types,
23096 @code{Value.dereference} and @code{Value.referenced_value} produce
23097 identical results. The difference between these methods is that
23098 @code{Value.dereference} cannot get the values referenced by reference
23099 values. For example, consider a reference to an @code{int}, declared
23100 in your C@t{++} program as
23108 then applying @code{Value.dereference} to the @code{gdb.Value} object
23109 corresponding to @code{ref} will result in an error, while applying
23110 @code{Value.referenced_value} will result in a @code{gdb.Value} object
23111 identical to that corresponding to @code{val}.
23114 py_ref = gdb.parse_and_eval ("ref")
23115 er_ref = py_ref.dereference () # Results in error
23116 py_val = py_ref.referenced_value () # Returns the referenced value
23119 The @code{gdb.Value} object @code{py_val} is identical to that
23120 corresponding to @code{val}.
23123 @defun Value.dynamic_cast (type)
23124 Like @code{Value.cast}, but works as if the C@t{++} @code{dynamic_cast}
23125 operator were used. Consult a C@t{++} reference for details.
23128 @defun Value.reinterpret_cast (type)
23129 Like @code{Value.cast}, but works as if the C@t{++} @code{reinterpret_cast}
23130 operator were used. Consult a C@t{++} reference for details.
23133 @defun Value.string (@r{[}encoding@r{[}, errors@r{[}, length@r{]]]})
23134 If this @code{gdb.Value} represents a string, then this method
23135 converts the contents to a Python string. Otherwise, this method will
23136 throw an exception.
23138 Strings are recognized in a language-specific way; whether a given
23139 @code{gdb.Value} represents a string is determined by the current
23142 For C-like languages, a value is a string if it is a pointer to or an
23143 array of characters or ints. The string is assumed to be terminated
23144 by a zero of the appropriate width. However if the optional length
23145 argument is given, the string will be converted to that given length,
23146 ignoring any embedded zeros that the string may contain.
23148 If the optional @var{encoding} argument is given, it must be a string
23149 naming the encoding of the string in the @code{gdb.Value}, such as
23150 @code{"ascii"}, @code{"iso-8859-6"} or @code{"utf-8"}. It accepts
23151 the same encodings as the corresponding argument to Python's
23152 @code{string.decode} method, and the Python codec machinery will be used
23153 to convert the string. If @var{encoding} is not given, or if
23154 @var{encoding} is the empty string, then either the @code{target-charset}
23155 (@pxref{Character Sets}) will be used, or a language-specific encoding
23156 will be used, if the current language is able to supply one.
23158 The optional @var{errors} argument is the same as the corresponding
23159 argument to Python's @code{string.decode} method.
23161 If the optional @var{length} argument is given, the string will be
23162 fetched and converted to the given length.
23165 @defun Value.lazy_string (@r{[}encoding @r{[}, length@r{]]})
23166 If this @code{gdb.Value} represents a string, then this method
23167 converts the contents to a @code{gdb.LazyString} (@pxref{Lazy Strings
23168 In Python}). Otherwise, this method will throw an exception.
23170 If the optional @var{encoding} argument is given, it must be a string
23171 naming the encoding of the @code{gdb.LazyString}. Some examples are:
23172 @samp{ascii}, @samp{iso-8859-6} or @samp{utf-8}. If the
23173 @var{encoding} argument is an encoding that @value{GDBN} does
23174 recognize, @value{GDBN} will raise an error.
23176 When a lazy string is printed, the @value{GDBN} encoding machinery is
23177 used to convert the string during printing. If the optional
23178 @var{encoding} argument is not provided, or is an empty string,
23179 @value{GDBN} will automatically select the encoding most suitable for
23180 the string type. For further information on encoding in @value{GDBN}
23181 please see @ref{Character Sets}.
23183 If the optional @var{length} argument is given, the string will be
23184 fetched and encoded to the length of characters specified. If
23185 the @var{length} argument is not provided, the string will be fetched
23186 and encoded until a null of appropriate width is found.
23189 @defun Value.fetch_lazy ()
23190 If the @code{gdb.Value} object is currently a lazy value
23191 (@code{gdb.Value.is_lazy} is @code{True}), then the value is
23192 fetched from the inferior. Any errors that occur in the process
23193 will produce a Python exception.
23195 If the @code{gdb.Value} object is not a lazy value, this method
23198 This method does not return a value.
23203 @node Types In Python
23204 @subsubsection Types In Python
23205 @cindex types in Python
23206 @cindex Python, working with types
23209 @value{GDBN} represents types from the inferior using the class
23212 The following type-related functions are available in the @code{gdb}
23215 @findex gdb.lookup_type
23216 @defun gdb.lookup_type (name @r{[}, block@r{]})
23217 This function looks up a type by name. @var{name} is the name of the
23218 type to look up. It must be a string.
23220 If @var{block} is given, then @var{name} is looked up in that scope.
23221 Otherwise, it is searched for globally.
23223 Ordinarily, this function will return an instance of @code{gdb.Type}.
23224 If the named type cannot be found, it will throw an exception.
23227 If the type is a structure or class type, or an enum type, the fields
23228 of that type can be accessed using the Python @dfn{dictionary syntax}.
23229 For example, if @code{some_type} is a @code{gdb.Type} instance holding
23230 a structure type, you can access its @code{foo} field with:
23233 bar = some_type['foo']
23236 @code{bar} will be a @code{gdb.Field} object; see below under the
23237 description of the @code{Type.fields} method for a description of the
23238 @code{gdb.Field} class.
23240 An instance of @code{Type} has the following attributes:
23244 The type code for this type. The type code will be one of the
23245 @code{TYPE_CODE_} constants defined below.
23248 @defvar Type.sizeof
23249 The size of this type, in target @code{char} units. Usually, a
23250 target's @code{char} type will be an 8-bit byte. However, on some
23251 unusual platforms, this type may have a different size.
23255 The tag name for this type. The tag name is the name after
23256 @code{struct}, @code{union}, or @code{enum} in C and C@t{++}; not all
23257 languages have this concept. If this type has no tag name, then
23258 @code{None} is returned.
23262 The following methods are provided:
23265 @defun Type.fields ()
23266 For structure and union types, this method returns the fields. Range
23267 types have two fields, the minimum and maximum values. Enum types
23268 have one field per enum constant. Function and method types have one
23269 field per parameter. The base types of C@t{++} classes are also
23270 represented as fields. If the type has no fields, or does not fit
23271 into one of these categories, an empty sequence will be returned.
23273 Each field is a @code{gdb.Field} object, with some pre-defined attributes:
23276 This attribute is not available for @code{static} fields (as in
23277 C@t{++} or Java). For non-@code{static} fields, the value is the bit
23278 position of the field. For @code{enum} fields, the value is the
23279 enumeration member's integer representation.
23282 The name of the field, or @code{None} for anonymous fields.
23285 This is @code{True} if the field is artificial, usually meaning that
23286 it was provided by the compiler and not the user. This attribute is
23287 always provided, and is @code{False} if the field is not artificial.
23289 @item is_base_class
23290 This is @code{True} if the field represents a base class of a C@t{++}
23291 structure. This attribute is always provided, and is @code{False}
23292 if the field is not a base class of the type that is the argument of
23293 @code{fields}, or if that type was not a C@t{++} class.
23296 If the field is packed, or is a bitfield, then this will have a
23297 non-zero value, which is the size of the field in bits. Otherwise,
23298 this will be zero; in this case the field's size is given by its type.
23301 The type of the field. This is usually an instance of @code{Type},
23302 but it can be @code{None} in some situations.
23306 @defun Type.array (@var{n1} @r{[}, @var{n2}@r{]})
23307 Return a new @code{gdb.Type} object which represents an array of this
23308 type. If one argument is given, it is the inclusive upper bound of
23309 the array; in this case the lower bound is zero. If two arguments are
23310 given, the first argument is the lower bound of the array, and the
23311 second argument is the upper bound of the array. An array's length
23312 must not be negative, but the bounds can be.
23315 @defun Type.const ()
23316 Return a new @code{gdb.Type} object which represents a
23317 @code{const}-qualified variant of this type.
23320 @defun Type.volatile ()
23321 Return a new @code{gdb.Type} object which represents a
23322 @code{volatile}-qualified variant of this type.
23325 @defun Type.unqualified ()
23326 Return a new @code{gdb.Type} object which represents an unqualified
23327 variant of this type. That is, the result is neither @code{const} nor
23331 @defun Type.range ()
23332 Return a Python @code{Tuple} object that contains two elements: the
23333 low bound of the argument type and the high bound of that type. If
23334 the type does not have a range, @value{GDBN} will raise a
23335 @code{gdb.error} exception (@pxref{Exception Handling}).
23338 @defun Type.reference ()
23339 Return a new @code{gdb.Type} object which represents a reference to this
23343 @defun Type.pointer ()
23344 Return a new @code{gdb.Type} object which represents a pointer to this
23348 @defun Type.strip_typedefs ()
23349 Return a new @code{gdb.Type} that represents the real type,
23350 after removing all layers of typedefs.
23353 @defun Type.target ()
23354 Return a new @code{gdb.Type} object which represents the target type
23357 For a pointer type, the target type is the type of the pointed-to
23358 object. For an array type (meaning C-like arrays), the target type is
23359 the type of the elements of the array. For a function or method type,
23360 the target type is the type of the return value. For a complex type,
23361 the target type is the type of the elements. For a typedef, the
23362 target type is the aliased type.
23364 If the type does not have a target, this method will throw an
23368 @defun Type.template_argument (n @r{[}, block@r{]})
23369 If this @code{gdb.Type} is an instantiation of a template, this will
23370 return a new @code{gdb.Type} which represents the type of the
23371 @var{n}th template argument.
23373 If this @code{gdb.Type} is not a template type, this will throw an
23374 exception. Ordinarily, only C@t{++} code will have template types.
23376 If @var{block} is given, then @var{name} is looked up in that scope.
23377 Otherwise, it is searched for globally.
23382 Each type has a code, which indicates what category this type falls
23383 into. The available type categories are represented by constants
23384 defined in the @code{gdb} module:
23387 @findex TYPE_CODE_PTR
23388 @findex gdb.TYPE_CODE_PTR
23389 @item gdb.TYPE_CODE_PTR
23390 The type is a pointer.
23392 @findex TYPE_CODE_ARRAY
23393 @findex gdb.TYPE_CODE_ARRAY
23394 @item gdb.TYPE_CODE_ARRAY
23395 The type is an array.
23397 @findex TYPE_CODE_STRUCT
23398 @findex gdb.TYPE_CODE_STRUCT
23399 @item gdb.TYPE_CODE_STRUCT
23400 The type is a structure.
23402 @findex TYPE_CODE_UNION
23403 @findex gdb.TYPE_CODE_UNION
23404 @item gdb.TYPE_CODE_UNION
23405 The type is a union.
23407 @findex TYPE_CODE_ENUM
23408 @findex gdb.TYPE_CODE_ENUM
23409 @item gdb.TYPE_CODE_ENUM
23410 The type is an enum.
23412 @findex TYPE_CODE_FLAGS
23413 @findex gdb.TYPE_CODE_FLAGS
23414 @item gdb.TYPE_CODE_FLAGS
23415 A bit flags type, used for things such as status registers.
23417 @findex TYPE_CODE_FUNC
23418 @findex gdb.TYPE_CODE_FUNC
23419 @item gdb.TYPE_CODE_FUNC
23420 The type is a function.
23422 @findex TYPE_CODE_INT
23423 @findex gdb.TYPE_CODE_INT
23424 @item gdb.TYPE_CODE_INT
23425 The type is an integer type.
23427 @findex TYPE_CODE_FLT
23428 @findex gdb.TYPE_CODE_FLT
23429 @item gdb.TYPE_CODE_FLT
23430 A floating point type.
23432 @findex TYPE_CODE_VOID
23433 @findex gdb.TYPE_CODE_VOID
23434 @item gdb.TYPE_CODE_VOID
23435 The special type @code{void}.
23437 @findex TYPE_CODE_SET
23438 @findex gdb.TYPE_CODE_SET
23439 @item gdb.TYPE_CODE_SET
23442 @findex TYPE_CODE_RANGE
23443 @findex gdb.TYPE_CODE_RANGE
23444 @item gdb.TYPE_CODE_RANGE
23445 A range type, that is, an integer type with bounds.
23447 @findex TYPE_CODE_STRING
23448 @findex gdb.TYPE_CODE_STRING
23449 @item gdb.TYPE_CODE_STRING
23450 A string type. Note that this is only used for certain languages with
23451 language-defined string types; C strings are not represented this way.
23453 @findex TYPE_CODE_BITSTRING
23454 @findex gdb.TYPE_CODE_BITSTRING
23455 @item gdb.TYPE_CODE_BITSTRING
23458 @findex TYPE_CODE_ERROR
23459 @findex gdb.TYPE_CODE_ERROR
23460 @item gdb.TYPE_CODE_ERROR
23461 An unknown or erroneous type.
23463 @findex TYPE_CODE_METHOD
23464 @findex gdb.TYPE_CODE_METHOD
23465 @item gdb.TYPE_CODE_METHOD
23466 A method type, as found in C@t{++} or Java.
23468 @findex TYPE_CODE_METHODPTR
23469 @findex gdb.TYPE_CODE_METHODPTR
23470 @item gdb.TYPE_CODE_METHODPTR
23471 A pointer-to-member-function.
23473 @findex TYPE_CODE_MEMBERPTR
23474 @findex gdb.TYPE_CODE_MEMBERPTR
23475 @item gdb.TYPE_CODE_MEMBERPTR
23476 A pointer-to-member.
23478 @findex TYPE_CODE_REF
23479 @findex gdb.TYPE_CODE_REF
23480 @item gdb.TYPE_CODE_REF
23483 @findex TYPE_CODE_CHAR
23484 @findex gdb.TYPE_CODE_CHAR
23485 @item gdb.TYPE_CODE_CHAR
23488 @findex TYPE_CODE_BOOL
23489 @findex gdb.TYPE_CODE_BOOL
23490 @item gdb.TYPE_CODE_BOOL
23493 @findex TYPE_CODE_COMPLEX
23494 @findex gdb.TYPE_CODE_COMPLEX
23495 @item gdb.TYPE_CODE_COMPLEX
23496 A complex float type.
23498 @findex TYPE_CODE_TYPEDEF
23499 @findex gdb.TYPE_CODE_TYPEDEF
23500 @item gdb.TYPE_CODE_TYPEDEF
23501 A typedef to some other type.
23503 @findex TYPE_CODE_NAMESPACE
23504 @findex gdb.TYPE_CODE_NAMESPACE
23505 @item gdb.TYPE_CODE_NAMESPACE
23506 A C@t{++} namespace.
23508 @findex TYPE_CODE_DECFLOAT
23509 @findex gdb.TYPE_CODE_DECFLOAT
23510 @item gdb.TYPE_CODE_DECFLOAT
23511 A decimal floating point type.
23513 @findex TYPE_CODE_INTERNAL_FUNCTION
23514 @findex gdb.TYPE_CODE_INTERNAL_FUNCTION
23515 @item gdb.TYPE_CODE_INTERNAL_FUNCTION
23516 A function internal to @value{GDBN}. This is the type used to represent
23517 convenience functions.
23520 Further support for types is provided in the @code{gdb.types}
23521 Python module (@pxref{gdb.types}).
23523 @node Pretty Printing API
23524 @subsubsection Pretty Printing API
23526 An example output is provided (@pxref{Pretty Printing}).
23528 A pretty-printer is just an object that holds a value and implements a
23529 specific interface, defined here.
23531 @defun pretty_printer.children (self)
23532 @value{GDBN} will call this method on a pretty-printer to compute the
23533 children of the pretty-printer's value.
23535 This method must return an object conforming to the Python iterator
23536 protocol. Each item returned by the iterator must be a tuple holding
23537 two elements. The first element is the ``name'' of the child; the
23538 second element is the child's value. The value can be any Python
23539 object which is convertible to a @value{GDBN} value.
23541 This method is optional. If it does not exist, @value{GDBN} will act
23542 as though the value has no children.
23545 @defun pretty_printer.display_hint (self)
23546 The CLI may call this method and use its result to change the
23547 formatting of a value. The result will also be supplied to an MI
23548 consumer as a @samp{displayhint} attribute of the variable being
23551 This method is optional. If it does exist, this method must return a
23554 Some display hints are predefined by @value{GDBN}:
23558 Indicate that the object being printed is ``array-like''. The CLI
23559 uses this to respect parameters such as @code{set print elements} and
23560 @code{set print array}.
23563 Indicate that the object being printed is ``map-like'', and that the
23564 children of this value can be assumed to alternate between keys and
23568 Indicate that the object being printed is ``string-like''. If the
23569 printer's @code{to_string} method returns a Python string of some
23570 kind, then @value{GDBN} will call its internal language-specific
23571 string-printing function to format the string. For the CLI this means
23572 adding quotation marks, possibly escaping some characters, respecting
23573 @code{set print elements}, and the like.
23577 @defun pretty_printer.to_string (self)
23578 @value{GDBN} will call this method to display the string
23579 representation of the value passed to the object's constructor.
23581 When printing from the CLI, if the @code{to_string} method exists,
23582 then @value{GDBN} will prepend its result to the values returned by
23583 @code{children}. Exactly how this formatting is done is dependent on
23584 the display hint, and may change as more hints are added. Also,
23585 depending on the print settings (@pxref{Print Settings}), the CLI may
23586 print just the result of @code{to_string} in a stack trace, omitting
23587 the result of @code{children}.
23589 If this method returns a string, it is printed verbatim.
23591 Otherwise, if this method returns an instance of @code{gdb.Value},
23592 then @value{GDBN} prints this value. This may result in a call to
23593 another pretty-printer.
23595 If instead the method returns a Python value which is convertible to a
23596 @code{gdb.Value}, then @value{GDBN} performs the conversion and prints
23597 the resulting value. Again, this may result in a call to another
23598 pretty-printer. Python scalars (integers, floats, and booleans) and
23599 strings are convertible to @code{gdb.Value}; other types are not.
23601 Finally, if this method returns @code{None} then no further operations
23602 are peformed in this method and nothing is printed.
23604 If the result is not one of these types, an exception is raised.
23607 @value{GDBN} provides a function which can be used to look up the
23608 default pretty-printer for a @code{gdb.Value}:
23610 @findex gdb.default_visualizer
23611 @defun gdb.default_visualizer (value)
23612 This function takes a @code{gdb.Value} object as an argument. If a
23613 pretty-printer for this value exists, then it is returned. If no such
23614 printer exists, then this returns @code{None}.
23617 @node Selecting Pretty-Printers
23618 @subsubsection Selecting Pretty-Printers
23620 The Python list @code{gdb.pretty_printers} contains an array of
23621 functions or callable objects that have been registered via addition
23622 as a pretty-printer. Printers in this list are called @code{global}
23623 printers, they're available when debugging all inferiors.
23624 Each @code{gdb.Progspace} contains a @code{pretty_printers} attribute.
23625 Each @code{gdb.Objfile} also contains a @code{pretty_printers}
23628 Each function on these lists is passed a single @code{gdb.Value}
23629 argument and should return a pretty-printer object conforming to the
23630 interface definition above (@pxref{Pretty Printing API}). If a function
23631 cannot create a pretty-printer for the value, it should return
23634 @value{GDBN} first checks the @code{pretty_printers} attribute of each
23635 @code{gdb.Objfile} in the current program space and iteratively calls
23636 each enabled lookup routine in the list for that @code{gdb.Objfile}
23637 until it receives a pretty-printer object.
23638 If no pretty-printer is found in the objfile lists, @value{GDBN} then
23639 searches the pretty-printer list of the current program space,
23640 calling each enabled function until an object is returned.
23641 After these lists have been exhausted, it tries the global
23642 @code{gdb.pretty_printers} list, again calling each enabled function until an
23643 object is returned.
23645 The order in which the objfiles are searched is not specified. For a
23646 given list, functions are always invoked from the head of the list,
23647 and iterated over sequentially until the end of the list, or a printer
23648 object is returned.
23650 For various reasons a pretty-printer may not work.
23651 For example, the underlying data structure may have changed and
23652 the pretty-printer is out of date.
23654 The consequences of a broken pretty-printer are severe enough that
23655 @value{GDBN} provides support for enabling and disabling individual
23656 printers. For example, if @code{print frame-arguments} is on,
23657 a backtrace can become highly illegible if any argument is printed
23658 with a broken printer.
23660 Pretty-printers are enabled and disabled by attaching an @code{enabled}
23661 attribute to the registered function or callable object. If this attribute
23662 is present and its value is @code{False}, the printer is disabled, otherwise
23663 the printer is enabled.
23665 @node Writing a Pretty-Printer
23666 @subsubsection Writing a Pretty-Printer
23667 @cindex writing a pretty-printer
23669 A pretty-printer consists of two parts: a lookup function to detect
23670 if the type is supported, and the printer itself.
23672 Here is an example showing how a @code{std::string} printer might be
23673 written. @xref{Pretty Printing API}, for details on the API this class
23677 class StdStringPrinter(object):
23678 "Print a std::string"
23680 def __init__(self, val):
23683 def to_string(self):
23684 return self.val['_M_dataplus']['_M_p']
23686 def display_hint(self):
23690 And here is an example showing how a lookup function for the printer
23691 example above might be written.
23694 def str_lookup_function(val):
23695 lookup_tag = val.type.tag
23696 if lookup_tag == None:
23698 regex = re.compile("^std::basic_string<char,.*>$")
23699 if regex.match(lookup_tag):
23700 return StdStringPrinter(val)
23704 The example lookup function extracts the value's type, and attempts to
23705 match it to a type that it can pretty-print. If it is a type the
23706 printer can pretty-print, it will return a printer object. If not, it
23707 returns @code{None}.
23709 We recommend that you put your core pretty-printers into a Python
23710 package. If your pretty-printers are for use with a library, we
23711 further recommend embedding a version number into the package name.
23712 This practice will enable @value{GDBN} to load multiple versions of
23713 your pretty-printers at the same time, because they will have
23716 You should write auto-loaded code (@pxref{Python Auto-loading}) such that it
23717 can be evaluated multiple times without changing its meaning. An
23718 ideal auto-load file will consist solely of @code{import}s of your
23719 printer modules, followed by a call to a register pretty-printers with
23720 the current objfile.
23722 Taken as a whole, this approach will scale nicely to multiple
23723 inferiors, each potentially using a different library version.
23724 Embedding a version number in the Python package name will ensure that
23725 @value{GDBN} is able to load both sets of printers simultaneously.
23726 Then, because the search for pretty-printers is done by objfile, and
23727 because your auto-loaded code took care to register your library's
23728 printers with a specific objfile, @value{GDBN} will find the correct
23729 printers for the specific version of the library used by each
23732 To continue the @code{std::string} example (@pxref{Pretty Printing API}),
23733 this code might appear in @code{gdb.libstdcxx.v6}:
23736 def register_printers(objfile):
23737 objfile.pretty_printers.append(str_lookup_function)
23741 And then the corresponding contents of the auto-load file would be:
23744 import gdb.libstdcxx.v6
23745 gdb.libstdcxx.v6.register_printers(gdb.current_objfile())
23748 The previous example illustrates a basic pretty-printer.
23749 There are a few things that can be improved on.
23750 The printer doesn't have a name, making it hard to identify in a
23751 list of installed printers. The lookup function has a name, but
23752 lookup functions can have arbitrary, even identical, names.
23754 Second, the printer only handles one type, whereas a library typically has
23755 several types. One could install a lookup function for each desired type
23756 in the library, but one could also have a single lookup function recognize
23757 several types. The latter is the conventional way this is handled.
23758 If a pretty-printer can handle multiple data types, then its
23759 @dfn{subprinters} are the printers for the individual data types.
23761 The @code{gdb.printing} module provides a formal way of solving these
23762 problems (@pxref{gdb.printing}).
23763 Here is another example that handles multiple types.
23765 These are the types we are going to pretty-print:
23768 struct foo @{ int a, b; @};
23769 struct bar @{ struct foo x, y; @};
23772 Here are the printers:
23776 """Print a foo object."""
23778 def __init__(self, val):
23781 def to_string(self):
23782 return ("a=<" + str(self.val["a"]) +
23783 "> b=<" + str(self.val["b"]) + ">")
23786 """Print a bar object."""
23788 def __init__(self, val):
23791 def to_string(self):
23792 return ("x=<" + str(self.val["x"]) +
23793 "> y=<" + str(self.val["y"]) + ">")
23796 This example doesn't need a lookup function, that is handled by the
23797 @code{gdb.printing} module. Instead a function is provided to build up
23798 the object that handles the lookup.
23801 import gdb.printing
23803 def build_pretty_printer():
23804 pp = gdb.printing.RegexpCollectionPrettyPrinter(
23806 pp.add_printer('foo', '^foo$', fooPrinter)
23807 pp.add_printer('bar', '^bar$', barPrinter)
23811 And here is the autoload support:
23814 import gdb.printing
23816 gdb.printing.register_pretty_printer(
23817 gdb.current_objfile(),
23818 my_library.build_pretty_printer())
23821 Finally, when this printer is loaded into @value{GDBN}, here is the
23822 corresponding output of @samp{info pretty-printer}:
23825 (gdb) info pretty-printer
23832 @node Inferiors In Python
23833 @subsubsection Inferiors In Python
23834 @cindex inferiors in Python
23836 @findex gdb.Inferior
23837 Programs which are being run under @value{GDBN} are called inferiors
23838 (@pxref{Inferiors and Programs}). Python scripts can access
23839 information about and manipulate inferiors controlled by @value{GDBN}
23840 via objects of the @code{gdb.Inferior} class.
23842 The following inferior-related functions are available in the @code{gdb}
23845 @defun gdb.inferiors ()
23846 Return a tuple containing all inferior objects.
23849 @defun gdb.selected_inferior ()
23850 Return an object representing the current inferior.
23853 A @code{gdb.Inferior} object has the following attributes:
23856 @defvar Inferior.num
23857 ID of inferior, as assigned by GDB.
23860 @defvar Inferior.pid
23861 Process ID of the inferior, as assigned by the underlying operating
23865 @defvar Inferior.was_attached
23866 Boolean signaling whether the inferior was created using `attach', or
23867 started by @value{GDBN} itself.
23871 A @code{gdb.Inferior} object has the following methods:
23874 @defun Inferior.is_valid ()
23875 Returns @code{True} if the @code{gdb.Inferior} object is valid,
23876 @code{False} if not. A @code{gdb.Inferior} object will become invalid
23877 if the inferior no longer exists within @value{GDBN}. All other
23878 @code{gdb.Inferior} methods will throw an exception if it is invalid
23879 at the time the method is called.
23882 @defun Inferior.threads ()
23883 This method returns a tuple holding all the threads which are valid
23884 when it is called. If there are no valid threads, the method will
23885 return an empty tuple.
23888 @findex gdb.read_memory
23889 @defun Inferior.read_memory (address, length)
23890 Read @var{length} bytes of memory from the inferior, starting at
23891 @var{address}. Returns a buffer object, which behaves much like an array
23892 or a string. It can be modified and given to the @code{gdb.write_memory}
23896 @findex gdb.write_memory
23897 @defun Inferior.write_memory (address, buffer @r{[}, length@r{]})
23898 Write the contents of @var{buffer} to the inferior, starting at
23899 @var{address}. The @var{buffer} parameter must be a Python object
23900 which supports the buffer protocol, i.e., a string, an array or the
23901 object returned from @code{gdb.read_memory}. If given, @var{length}
23902 determines the number of bytes from @var{buffer} to be written.
23905 @findex gdb.search_memory
23906 @defun Inferior.search_memory (address, length, pattern)
23907 Search a region of the inferior memory starting at @var{address} with
23908 the given @var{length} using the search pattern supplied in
23909 @var{pattern}. The @var{pattern} parameter must be a Python object
23910 which supports the buffer protocol, i.e., a string, an array or the
23911 object returned from @code{gdb.read_memory}. Returns a Python @code{Long}
23912 containing the address where the pattern was found, or @code{None} if
23913 the pattern could not be found.
23917 @node Events In Python
23918 @subsubsection Events In Python
23919 @cindex inferior events in Python
23921 @value{GDBN} provides a general event facility so that Python code can be
23922 notified of various state changes, particularly changes that occur in
23925 An @dfn{event} is just an object that describes some state change. The
23926 type of the object and its attributes will vary depending on the details
23927 of the change. All the existing events are described below.
23929 In order to be notified of an event, you must register an event handler
23930 with an @dfn{event registry}. An event registry is an object in the
23931 @code{gdb.events} module which dispatches particular events. A registry
23932 provides methods to register and unregister event handlers:
23935 @defun EventRegistry.connect (object)
23936 Add the given callable @var{object} to the registry. This object will be
23937 called when an event corresponding to this registry occurs.
23940 @defun EventRegistry.disconnect (object)
23941 Remove the given @var{object} from the registry. Once removed, the object
23942 will no longer receive notifications of events.
23946 Here is an example:
23949 def exit_handler (event):
23950 print "event type: exit"
23951 print "exit code: %d" % (event.exit_code)
23953 gdb.events.exited.connect (exit_handler)
23956 In the above example we connect our handler @code{exit_handler} to the
23957 registry @code{events.exited}. Once connected, @code{exit_handler} gets
23958 called when the inferior exits. The argument @dfn{event} in this example is
23959 of type @code{gdb.ExitedEvent}. As you can see in the example the
23960 @code{ExitedEvent} object has an attribute which indicates the exit code of
23963 The following is a listing of the event registries that are available and
23964 details of the events they emit:
23969 Emits @code{gdb.ThreadEvent}.
23971 Some events can be thread specific when @value{GDBN} is running in non-stop
23972 mode. When represented in Python, these events all extend
23973 @code{gdb.ThreadEvent}. Note, this event is not emitted directly; instead,
23974 events which are emitted by this or other modules might extend this event.
23975 Examples of these events are @code{gdb.BreakpointEvent} and
23976 @code{gdb.ContinueEvent}.
23979 @defvar ThreadEvent.inferior_thread
23980 In non-stop mode this attribute will be set to the specific thread which was
23981 involved in the emitted event. Otherwise, it will be set to @code{None}.
23985 Emits @code{gdb.ContinueEvent} which extends @code{gdb.ThreadEvent}.
23987 This event indicates that the inferior has been continued after a stop. For
23988 inherited attribute refer to @code{gdb.ThreadEvent} above.
23990 @item events.exited
23991 Emits @code{events.ExitedEvent} which indicates that the inferior has exited.
23992 @code{events.ExitedEvent} has two attributes:
23994 @defvar ExitedEvent.exit_code
23995 An integer representing the exit code, if available, which the inferior
23996 has returned. (The exit code could be unavailable if, for example,
23997 @value{GDBN} detaches from the inferior.) If the exit code is unavailable,
23998 the attribute does not exist.
24000 @defvar ExitedEvent inferior
24001 A reference to the inferior which triggered the @code{exited} event.
24006 Emits @code{gdb.StopEvent} which extends @code{gdb.ThreadEvent}.
24008 Indicates that the inferior has stopped. All events emitted by this registry
24009 extend StopEvent. As a child of @code{gdb.ThreadEvent}, @code{gdb.StopEvent}
24010 will indicate the stopped thread when @value{GDBN} is running in non-stop
24011 mode. Refer to @code{gdb.ThreadEvent} above for more details.
24013 Emits @code{gdb.SignalEvent} which extends @code{gdb.StopEvent}.
24015 This event indicates that the inferior or one of its threads has received as
24016 signal. @code{gdb.SignalEvent} has the following attributes:
24019 @defvar SignalEvent.stop_signal
24020 A string representing the signal received by the inferior. A list of possible
24021 signal values can be obtained by running the command @code{info signals} in
24022 the @value{GDBN} command prompt.
24026 Also emits @code{gdb.BreakpointEvent} which extends @code{gdb.StopEvent}.
24028 @code{gdb.BreakpointEvent} event indicates that one or more breakpoints have
24029 been hit, and has the following attributes:
24032 @defvar BreakpointEvent.breakpoints
24033 A sequence containing references to all the breakpoints (type
24034 @code{gdb.Breakpoint}) that were hit.
24035 @xref{Breakpoints In Python}, for details of the @code{gdb.Breakpoint} object.
24037 @defvar BreakpointEvent.breakpoint
24038 A reference to the first breakpoint that was hit.
24039 This function is maintained for backward compatibility and is now deprecated
24040 in favor of the @code{gdb.BreakpointEvent.breakpoints} attribute.
24044 @item events.new_objfile
24045 Emits @code{gdb.NewObjFileEvent} which indicates that a new object file has
24046 been loaded by @value{GDBN}. @code{gdb.NewObjFileEvent} has one attribute:
24049 @defvar NewObjFileEvent.new_objfile
24050 A reference to the object file (@code{gdb.Objfile}) which has been loaded.
24051 @xref{Objfiles In Python}, for details of the @code{gdb.Objfile} object.
24057 @node Threads In Python
24058 @subsubsection Threads In Python
24059 @cindex threads in python
24061 @findex gdb.InferiorThread
24062 Python scripts can access information about, and manipulate inferior threads
24063 controlled by @value{GDBN}, via objects of the @code{gdb.InferiorThread} class.
24065 The following thread-related functions are available in the @code{gdb}
24068 @findex gdb.selected_thread
24069 @defun gdb.selected_thread ()
24070 This function returns the thread object for the selected thread. If there
24071 is no selected thread, this will return @code{None}.
24074 A @code{gdb.InferiorThread} object has the following attributes:
24077 @defvar InferiorThread.name
24078 The name of the thread. If the user specified a name using
24079 @code{thread name}, then this returns that name. Otherwise, if an
24080 OS-supplied name is available, then it is returned. Otherwise, this
24081 returns @code{None}.
24083 This attribute can be assigned to. The new value must be a string
24084 object, which sets the new name, or @code{None}, which removes any
24085 user-specified thread name.
24088 @defvar InferiorThread.num
24089 ID of the thread, as assigned by GDB.
24092 @defvar InferiorThread.ptid
24093 ID of the thread, as assigned by the operating system. This attribute is a
24094 tuple containing three integers. The first is the Process ID (PID); the second
24095 is the Lightweight Process ID (LWPID), and the third is the Thread ID (TID).
24096 Either the LWPID or TID may be 0, which indicates that the operating system
24097 does not use that identifier.
24101 A @code{gdb.InferiorThread} object has the following methods:
24104 @defun InferiorThread.is_valid ()
24105 Returns @code{True} if the @code{gdb.InferiorThread} object is valid,
24106 @code{False} if not. A @code{gdb.InferiorThread} object will become
24107 invalid if the thread exits, or the inferior that the thread belongs
24108 is deleted. All other @code{gdb.InferiorThread} methods will throw an
24109 exception if it is invalid at the time the method is called.
24112 @defun InferiorThread.switch ()
24113 This changes @value{GDBN}'s currently selected thread to the one represented
24117 @defun InferiorThread.is_stopped ()
24118 Return a Boolean indicating whether the thread is stopped.
24121 @defun InferiorThread.is_running ()
24122 Return a Boolean indicating whether the thread is running.
24125 @defun InferiorThread.is_exited ()
24126 Return a Boolean indicating whether the thread is exited.
24130 @node Commands In Python
24131 @subsubsection Commands In Python
24133 @cindex commands in python
24134 @cindex python commands
24135 You can implement new @value{GDBN} CLI commands in Python. A CLI
24136 command is implemented using an instance of the @code{gdb.Command}
24137 class, most commonly using a subclass.
24139 @defun Command.__init__ (name, @var{command_class} @r{[}, @var{completer_class} @r{[}, @var{prefix}@r{]]})
24140 The object initializer for @code{Command} registers the new command
24141 with @value{GDBN}. This initializer is normally invoked from the
24142 subclass' own @code{__init__} method.
24144 @var{name} is the name of the command. If @var{name} consists of
24145 multiple words, then the initial words are looked for as prefix
24146 commands. In this case, if one of the prefix commands does not exist,
24147 an exception is raised.
24149 There is no support for multi-line commands.
24151 @var{command_class} should be one of the @samp{COMMAND_} constants
24152 defined below. This argument tells @value{GDBN} how to categorize the
24153 new command in the help system.
24155 @var{completer_class} is an optional argument. If given, it should be
24156 one of the @samp{COMPLETE_} constants defined below. This argument
24157 tells @value{GDBN} how to perform completion for this command. If not
24158 given, @value{GDBN} will attempt to complete using the object's
24159 @code{complete} method (see below); if no such method is found, an
24160 error will occur when completion is attempted.
24162 @var{prefix} is an optional argument. If @code{True}, then the new
24163 command is a prefix command; sub-commands of this command may be
24166 The help text for the new command is taken from the Python
24167 documentation string for the command's class, if there is one. If no
24168 documentation string is provided, the default value ``This command is
24169 not documented.'' is used.
24172 @cindex don't repeat Python command
24173 @defun Command.dont_repeat ()
24174 By default, a @value{GDBN} command is repeated when the user enters a
24175 blank line at the command prompt. A command can suppress this
24176 behavior by invoking the @code{dont_repeat} method. This is similar
24177 to the user command @code{dont-repeat}, see @ref{Define, dont-repeat}.
24180 @defun Command.invoke (argument, from_tty)
24181 This method is called by @value{GDBN} when this command is invoked.
24183 @var{argument} is a string. It is the argument to the command, after
24184 leading and trailing whitespace has been stripped.
24186 @var{from_tty} is a boolean argument. When true, this means that the
24187 command was entered by the user at the terminal; when false it means
24188 that the command came from elsewhere.
24190 If this method throws an exception, it is turned into a @value{GDBN}
24191 @code{error} call. Otherwise, the return value is ignored.
24193 @findex gdb.string_to_argv
24194 To break @var{argument} up into an argv-like string use
24195 @code{gdb.string_to_argv}. This function behaves identically to
24196 @value{GDBN}'s internal argument lexer @code{buildargv}.
24197 It is recommended to use this for consistency.
24198 Arguments are separated by spaces and may be quoted.
24202 print gdb.string_to_argv ("1 2\ \\\"3 '4 \"5' \"6 '7\"")
24203 ['1', '2 "3', '4 "5', "6 '7"]
24208 @cindex completion of Python commands
24209 @defun Command.complete (text, word)
24210 This method is called by @value{GDBN} when the user attempts
24211 completion on this command. All forms of completion are handled by
24212 this method, that is, the @key{TAB} and @key{M-?} key bindings
24213 (@pxref{Completion}), and the @code{complete} command (@pxref{Help,
24216 The arguments @var{text} and @var{word} are both strings. @var{text}
24217 holds the complete command line up to the cursor's location.
24218 @var{word} holds the last word of the command line; this is computed
24219 using a word-breaking heuristic.
24221 The @code{complete} method can return several values:
24224 If the return value is a sequence, the contents of the sequence are
24225 used as the completions. It is up to @code{complete} to ensure that the
24226 contents actually do complete the word. A zero-length sequence is
24227 allowed, it means that there were no completions available. Only
24228 string elements of the sequence are used; other elements in the
24229 sequence are ignored.
24232 If the return value is one of the @samp{COMPLETE_} constants defined
24233 below, then the corresponding @value{GDBN}-internal completion
24234 function is invoked, and its result is used.
24237 All other results are treated as though there were no available
24242 When a new command is registered, it must be declared as a member of
24243 some general class of commands. This is used to classify top-level
24244 commands in the on-line help system; note that prefix commands are not
24245 listed under their own category but rather that of their top-level
24246 command. The available classifications are represented by constants
24247 defined in the @code{gdb} module:
24250 @findex COMMAND_NONE
24251 @findex gdb.COMMAND_NONE
24252 @item gdb.COMMAND_NONE
24253 The command does not belong to any particular class. A command in
24254 this category will not be displayed in any of the help categories.
24256 @findex COMMAND_RUNNING
24257 @findex gdb.COMMAND_RUNNING
24258 @item gdb.COMMAND_RUNNING
24259 The command is related to running the inferior. For example,
24260 @code{start}, @code{step}, and @code{continue} are in this category.
24261 Type @kbd{help running} at the @value{GDBN} prompt to see a list of
24262 commands in this category.
24264 @findex COMMAND_DATA
24265 @findex gdb.COMMAND_DATA
24266 @item gdb.COMMAND_DATA
24267 The command is related to data or variables. For example,
24268 @code{call}, @code{find}, and @code{print} are in this category. Type
24269 @kbd{help data} at the @value{GDBN} prompt to see a list of commands
24272 @findex COMMAND_STACK
24273 @findex gdb.COMMAND_STACK
24274 @item gdb.COMMAND_STACK
24275 The command has to do with manipulation of the stack. For example,
24276 @code{backtrace}, @code{frame}, and @code{return} are in this
24277 category. Type @kbd{help stack} at the @value{GDBN} prompt to see a
24278 list of commands in this category.
24280 @findex COMMAND_FILES
24281 @findex gdb.COMMAND_FILES
24282 @item gdb.COMMAND_FILES
24283 This class is used for file-related commands. For example,
24284 @code{file}, @code{list} and @code{section} are in this category.
24285 Type @kbd{help files} at the @value{GDBN} prompt to see a list of
24286 commands in this category.
24288 @findex COMMAND_SUPPORT
24289 @findex gdb.COMMAND_SUPPORT
24290 @item gdb.COMMAND_SUPPORT
24291 This should be used for ``support facilities'', generally meaning
24292 things that are useful to the user when interacting with @value{GDBN},
24293 but not related to the state of the inferior. For example,
24294 @code{help}, @code{make}, and @code{shell} are in this category. Type
24295 @kbd{help support} at the @value{GDBN} prompt to see a list of
24296 commands in this category.
24298 @findex COMMAND_STATUS
24299 @findex gdb.COMMAND_STATUS
24300 @item gdb.COMMAND_STATUS
24301 The command is an @samp{info}-related command, that is, related to the
24302 state of @value{GDBN} itself. For example, @code{info}, @code{macro},
24303 and @code{show} are in this category. Type @kbd{help status} at the
24304 @value{GDBN} prompt to see a list of commands in this category.
24306 @findex COMMAND_BREAKPOINTS
24307 @findex gdb.COMMAND_BREAKPOINTS
24308 @item gdb.COMMAND_BREAKPOINTS
24309 The command has to do with breakpoints. For example, @code{break},
24310 @code{clear}, and @code{delete} are in this category. Type @kbd{help
24311 breakpoints} at the @value{GDBN} prompt to see a list of commands in
24314 @findex COMMAND_TRACEPOINTS
24315 @findex gdb.COMMAND_TRACEPOINTS
24316 @item gdb.COMMAND_TRACEPOINTS
24317 The command has to do with tracepoints. For example, @code{trace},
24318 @code{actions}, and @code{tfind} are in this category. Type
24319 @kbd{help tracepoints} at the @value{GDBN} prompt to see a list of
24320 commands in this category.
24322 @findex COMMAND_USER
24323 @findex gdb.COMMAND_USER
24324 @item gdb.COMMAND_USER
24325 The command is a general purpose command for the user, and typically
24326 does not fit in one of the other categories.
24327 Type @kbd{help user-defined} at the @value{GDBN} prompt to see
24328 a list of commands in this category, as well as the list of gdb macros
24329 (@pxref{Sequences}).
24331 @findex COMMAND_OBSCURE
24332 @findex gdb.COMMAND_OBSCURE
24333 @item gdb.COMMAND_OBSCURE
24334 The command is only used in unusual circumstances, or is not of
24335 general interest to users. For example, @code{checkpoint},
24336 @code{fork}, and @code{stop} are in this category. Type @kbd{help
24337 obscure} at the @value{GDBN} prompt to see a list of commands in this
24340 @findex COMMAND_MAINTENANCE
24341 @findex gdb.COMMAND_MAINTENANCE
24342 @item gdb.COMMAND_MAINTENANCE
24343 The command is only useful to @value{GDBN} maintainers. The
24344 @code{maintenance} and @code{flushregs} commands are in this category.
24345 Type @kbd{help internals} at the @value{GDBN} prompt to see a list of
24346 commands in this category.
24349 A new command can use a predefined completion function, either by
24350 specifying it via an argument at initialization, or by returning it
24351 from the @code{complete} method. These predefined completion
24352 constants are all defined in the @code{gdb} module:
24355 @findex COMPLETE_NONE
24356 @findex gdb.COMPLETE_NONE
24357 @item gdb.COMPLETE_NONE
24358 This constant means that no completion should be done.
24360 @findex COMPLETE_FILENAME
24361 @findex gdb.COMPLETE_FILENAME
24362 @item gdb.COMPLETE_FILENAME
24363 This constant means that filename completion should be performed.
24365 @findex COMPLETE_LOCATION
24366 @findex gdb.COMPLETE_LOCATION
24367 @item gdb.COMPLETE_LOCATION
24368 This constant means that location completion should be done.
24369 @xref{Specify Location}.
24371 @findex COMPLETE_COMMAND
24372 @findex gdb.COMPLETE_COMMAND
24373 @item gdb.COMPLETE_COMMAND
24374 This constant means that completion should examine @value{GDBN}
24377 @findex COMPLETE_SYMBOL
24378 @findex gdb.COMPLETE_SYMBOL
24379 @item gdb.COMPLETE_SYMBOL
24380 This constant means that completion should be done using symbol names
24384 The following code snippet shows how a trivial CLI command can be
24385 implemented in Python:
24388 class HelloWorld (gdb.Command):
24389 """Greet the whole world."""
24391 def __init__ (self):
24392 super (HelloWorld, self).__init__ ("hello-world", gdb.COMMAND_USER)
24394 def invoke (self, arg, from_tty):
24395 print "Hello, World!"
24400 The last line instantiates the class, and is necessary to trigger the
24401 registration of the command with @value{GDBN}. Depending on how the
24402 Python code is read into @value{GDBN}, you may need to import the
24403 @code{gdb} module explicitly.
24405 @node Parameters In Python
24406 @subsubsection Parameters In Python
24408 @cindex parameters in python
24409 @cindex python parameters
24410 @tindex gdb.Parameter
24412 You can implement new @value{GDBN} parameters using Python. A new
24413 parameter is implemented as an instance of the @code{gdb.Parameter}
24416 Parameters are exposed to the user via the @code{set} and
24417 @code{show} commands. @xref{Help}.
24419 There are many parameters that already exist and can be set in
24420 @value{GDBN}. Two examples are: @code{set follow fork} and
24421 @code{set charset}. Setting these parameters influences certain
24422 behavior in @value{GDBN}. Similarly, you can define parameters that
24423 can be used to influence behavior in custom Python scripts and commands.
24425 @defun Parameter.__init__ (name, @var{command-class}, @var{parameter-class} @r{[}, @var{enum-sequence}@r{]})
24426 The object initializer for @code{Parameter} registers the new
24427 parameter with @value{GDBN}. This initializer is normally invoked
24428 from the subclass' own @code{__init__} method.
24430 @var{name} is the name of the new parameter. If @var{name} consists
24431 of multiple words, then the initial words are looked for as prefix
24432 parameters. An example of this can be illustrated with the
24433 @code{set print} set of parameters. If @var{name} is
24434 @code{print foo}, then @code{print} will be searched as the prefix
24435 parameter. In this case the parameter can subsequently be accessed in
24436 @value{GDBN} as @code{set print foo}.
24438 If @var{name} consists of multiple words, and no prefix parameter group
24439 can be found, an exception is raised.
24441 @var{command-class} should be one of the @samp{COMMAND_} constants
24442 (@pxref{Commands In Python}). This argument tells @value{GDBN} how to
24443 categorize the new parameter in the help system.
24445 @var{parameter-class} should be one of the @samp{PARAM_} constants
24446 defined below. This argument tells @value{GDBN} the type of the new
24447 parameter; this information is used for input validation and
24450 If @var{parameter-class} is @code{PARAM_ENUM}, then
24451 @var{enum-sequence} must be a sequence of strings. These strings
24452 represent the possible values for the parameter.
24454 If @var{parameter-class} is not @code{PARAM_ENUM}, then the presence
24455 of a fourth argument will cause an exception to be thrown.
24457 The help text for the new parameter is taken from the Python
24458 documentation string for the parameter's class, if there is one. If
24459 there is no documentation string, a default value is used.
24462 @defvar Parameter.set_doc
24463 If this attribute exists, and is a string, then its value is used as
24464 the help text for this parameter's @code{set} command. The value is
24465 examined when @code{Parameter.__init__} is invoked; subsequent changes
24469 @defvar Parameter.show_doc
24470 If this attribute exists, and is a string, then its value is used as
24471 the help text for this parameter's @code{show} command. The value is
24472 examined when @code{Parameter.__init__} is invoked; subsequent changes
24476 @defvar Parameter.value
24477 The @code{value} attribute holds the underlying value of the
24478 parameter. It can be read and assigned to just as any other
24479 attribute. @value{GDBN} does validation when assignments are made.
24482 There are two methods that should be implemented in any
24483 @code{Parameter} class. These are:
24485 @defun Parameter.get_set_string (self)
24486 @value{GDBN} will call this method when a @var{parameter}'s value has
24487 been changed via the @code{set} API (for example, @kbd{set foo off}).
24488 The @code{value} attribute has already been populated with the new
24489 value and may be used in output. This method must return a string.
24492 @defun Parameter.get_show_string (self, svalue)
24493 @value{GDBN} will call this method when a @var{parameter}'s
24494 @code{show} API has been invoked (for example, @kbd{show foo}). The
24495 argument @code{svalue} receives the string representation of the
24496 current value. This method must return a string.
24499 When a new parameter is defined, its type must be specified. The
24500 available types are represented by constants defined in the @code{gdb}
24504 @findex PARAM_BOOLEAN
24505 @findex gdb.PARAM_BOOLEAN
24506 @item gdb.PARAM_BOOLEAN
24507 The value is a plain boolean. The Python boolean values, @code{True}
24508 and @code{False} are the only valid values.
24510 @findex PARAM_AUTO_BOOLEAN
24511 @findex gdb.PARAM_AUTO_BOOLEAN
24512 @item gdb.PARAM_AUTO_BOOLEAN
24513 The value has three possible states: true, false, and @samp{auto}. In
24514 Python, true and false are represented using boolean constants, and
24515 @samp{auto} is represented using @code{None}.
24517 @findex PARAM_UINTEGER
24518 @findex gdb.PARAM_UINTEGER
24519 @item gdb.PARAM_UINTEGER
24520 The value is an unsigned integer. The value of 0 should be
24521 interpreted to mean ``unlimited''.
24523 @findex PARAM_INTEGER
24524 @findex gdb.PARAM_INTEGER
24525 @item gdb.PARAM_INTEGER
24526 The value is a signed integer. The value of 0 should be interpreted
24527 to mean ``unlimited''.
24529 @findex PARAM_STRING
24530 @findex gdb.PARAM_STRING
24531 @item gdb.PARAM_STRING
24532 The value is a string. When the user modifies the string, any escape
24533 sequences, such as @samp{\t}, @samp{\f}, and octal escapes, are
24534 translated into corresponding characters and encoded into the current
24537 @findex PARAM_STRING_NOESCAPE
24538 @findex gdb.PARAM_STRING_NOESCAPE
24539 @item gdb.PARAM_STRING_NOESCAPE
24540 The value is a string. When the user modifies the string, escapes are
24541 passed through untranslated.
24543 @findex PARAM_OPTIONAL_FILENAME
24544 @findex gdb.PARAM_OPTIONAL_FILENAME
24545 @item gdb.PARAM_OPTIONAL_FILENAME
24546 The value is a either a filename (a string), or @code{None}.
24548 @findex PARAM_FILENAME
24549 @findex gdb.PARAM_FILENAME
24550 @item gdb.PARAM_FILENAME
24551 The value is a filename. This is just like
24552 @code{PARAM_STRING_NOESCAPE}, but uses file names for completion.
24554 @findex PARAM_ZINTEGER
24555 @findex gdb.PARAM_ZINTEGER
24556 @item gdb.PARAM_ZINTEGER
24557 The value is an integer. This is like @code{PARAM_INTEGER}, except 0
24558 is interpreted as itself.
24561 @findex gdb.PARAM_ENUM
24562 @item gdb.PARAM_ENUM
24563 The value is a string, which must be one of a collection string
24564 constants provided when the parameter is created.
24567 @node Functions In Python
24568 @subsubsection Writing new convenience functions
24570 @cindex writing convenience functions
24571 @cindex convenience functions in python
24572 @cindex python convenience functions
24573 @tindex gdb.Function
24575 You can implement new convenience functions (@pxref{Convenience Vars})
24576 in Python. A convenience function is an instance of a subclass of the
24577 class @code{gdb.Function}.
24579 @defun Function.__init__ (name)
24580 The initializer for @code{Function} registers the new function with
24581 @value{GDBN}. The argument @var{name} is the name of the function,
24582 a string. The function will be visible to the user as a convenience
24583 variable of type @code{internal function}, whose name is the same as
24584 the given @var{name}.
24586 The documentation for the new function is taken from the documentation
24587 string for the new class.
24590 @defun Function.invoke (@var{*args})
24591 When a convenience function is evaluated, its arguments are converted
24592 to instances of @code{gdb.Value}, and then the function's
24593 @code{invoke} method is called. Note that @value{GDBN} does not
24594 predetermine the arity of convenience functions. Instead, all
24595 available arguments are passed to @code{invoke}, following the
24596 standard Python calling convention. In particular, a convenience
24597 function can have default values for parameters without ill effect.
24599 The return value of this method is used as its value in the enclosing
24600 expression. If an ordinary Python value is returned, it is converted
24601 to a @code{gdb.Value} following the usual rules.
24604 The following code snippet shows how a trivial convenience function can
24605 be implemented in Python:
24608 class Greet (gdb.Function):
24609 """Return string to greet someone.
24610 Takes a name as argument."""
24612 def __init__ (self):
24613 super (Greet, self).__init__ ("greet")
24615 def invoke (self, name):
24616 return "Hello, %s!" % name.string ()
24621 The last line instantiates the class, and is necessary to trigger the
24622 registration of the function with @value{GDBN}. Depending on how the
24623 Python code is read into @value{GDBN}, you may need to import the
24624 @code{gdb} module explicitly.
24626 @node Progspaces In Python
24627 @subsubsection Program Spaces In Python
24629 @cindex progspaces in python
24630 @tindex gdb.Progspace
24632 A program space, or @dfn{progspace}, represents a symbolic view
24633 of an address space.
24634 It consists of all of the objfiles of the program.
24635 @xref{Objfiles In Python}.
24636 @xref{Inferiors and Programs, program spaces}, for more details
24637 about program spaces.
24639 The following progspace-related functions are available in the
24642 @findex gdb.current_progspace
24643 @defun gdb.current_progspace ()
24644 This function returns the program space of the currently selected inferior.
24645 @xref{Inferiors and Programs}.
24648 @findex gdb.progspaces
24649 @defun gdb.progspaces ()
24650 Return a sequence of all the progspaces currently known to @value{GDBN}.
24653 Each progspace is represented by an instance of the @code{gdb.Progspace}
24656 @defvar Progspace.filename
24657 The file name of the progspace as a string.
24660 @defvar Progspace.pretty_printers
24661 The @code{pretty_printers} attribute is a list of functions. It is
24662 used to look up pretty-printers. A @code{Value} is passed to each
24663 function in order; if the function returns @code{None}, then the
24664 search continues. Otherwise, the return value should be an object
24665 which is used to format the value. @xref{Pretty Printing API}, for more
24669 @node Objfiles In Python
24670 @subsubsection Objfiles In Python
24672 @cindex objfiles in python
24673 @tindex gdb.Objfile
24675 @value{GDBN} loads symbols for an inferior from various
24676 symbol-containing files (@pxref{Files}). These include the primary
24677 executable file, any shared libraries used by the inferior, and any
24678 separate debug info files (@pxref{Separate Debug Files}).
24679 @value{GDBN} calls these symbol-containing files @dfn{objfiles}.
24681 The following objfile-related functions are available in the
24684 @findex gdb.current_objfile
24685 @defun gdb.current_objfile ()
24686 When auto-loading a Python script (@pxref{Python Auto-loading}), @value{GDBN}
24687 sets the ``current objfile'' to the corresponding objfile. This
24688 function returns the current objfile. If there is no current objfile,
24689 this function returns @code{None}.
24692 @findex gdb.objfiles
24693 @defun gdb.objfiles ()
24694 Return a sequence of all the objfiles current known to @value{GDBN}.
24695 @xref{Objfiles In Python}.
24698 Each objfile is represented by an instance of the @code{gdb.Objfile}
24701 @defvar Objfile.filename
24702 The file name of the objfile as a string.
24705 @defvar Objfile.pretty_printers
24706 The @code{pretty_printers} attribute is a list of functions. It is
24707 used to look up pretty-printers. A @code{Value} is passed to each
24708 function in order; if the function returns @code{None}, then the
24709 search continues. Otherwise, the return value should be an object
24710 which is used to format the value. @xref{Pretty Printing API}, for more
24714 A @code{gdb.Objfile} object has the following methods:
24716 @defun Objfile.is_valid ()
24717 Returns @code{True} if the @code{gdb.Objfile} object is valid,
24718 @code{False} if not. A @code{gdb.Objfile} object can become invalid
24719 if the object file it refers to is not loaded in @value{GDBN} any
24720 longer. All other @code{gdb.Objfile} methods will throw an exception
24721 if it is invalid at the time the method is called.
24724 @node Frames In Python
24725 @subsubsection Accessing inferior stack frames from Python.
24727 @cindex frames in python
24728 When the debugged program stops, @value{GDBN} is able to analyze its call
24729 stack (@pxref{Frames,,Stack frames}). The @code{gdb.Frame} class
24730 represents a frame in the stack. A @code{gdb.Frame} object is only valid
24731 while its corresponding frame exists in the inferior's stack. If you try
24732 to use an invalid frame object, @value{GDBN} will throw a @code{gdb.error}
24733 exception (@pxref{Exception Handling}).
24735 Two @code{gdb.Frame} objects can be compared for equality with the @code{==}
24739 (@value{GDBP}) python print gdb.newest_frame() == gdb.selected_frame ()
24743 The following frame-related functions are available in the @code{gdb} module:
24745 @findex gdb.selected_frame
24746 @defun gdb.selected_frame ()
24747 Return the selected frame object. (@pxref{Selection,,Selecting a Frame}).
24750 @findex gdb.newest_frame
24751 @defun gdb.newest_frame ()
24752 Return the newest frame object for the selected thread.
24755 @defun gdb.frame_stop_reason_string (reason)
24756 Return a string explaining the reason why @value{GDBN} stopped unwinding
24757 frames, as expressed by the given @var{reason} code (an integer, see the
24758 @code{unwind_stop_reason} method further down in this section).
24761 A @code{gdb.Frame} object has the following methods:
24764 @defun Frame.is_valid ()
24765 Returns true if the @code{gdb.Frame} object is valid, false if not.
24766 A frame object can become invalid if the frame it refers to doesn't
24767 exist anymore in the inferior. All @code{gdb.Frame} methods will throw
24768 an exception if it is invalid at the time the method is called.
24771 @defun Frame.name ()
24772 Returns the function name of the frame, or @code{None} if it can't be
24776 @defun Frame.type ()
24777 Returns the type of the frame. The value can be one of:
24779 @item gdb.NORMAL_FRAME
24780 An ordinary stack frame.
24782 @item gdb.DUMMY_FRAME
24783 A fake stack frame that was created by @value{GDBN} when performing an
24784 inferior function call.
24786 @item gdb.INLINE_FRAME
24787 A frame representing an inlined function. The function was inlined
24788 into a @code{gdb.NORMAL_FRAME} that is older than this one.
24790 @item gdb.TAILCALL_FRAME
24791 A frame representing a tail call. @xref{Tail Call Frames}.
24793 @item gdb.SIGTRAMP_FRAME
24794 A signal trampoline frame. This is the frame created by the OS when
24795 it calls into a signal handler.
24797 @item gdb.ARCH_FRAME
24798 A fake stack frame representing a cross-architecture call.
24800 @item gdb.SENTINEL_FRAME
24801 This is like @code{gdb.NORMAL_FRAME}, but it is only used for the
24806 @defun Frame.unwind_stop_reason ()
24807 Return an integer representing the reason why it's not possible to find
24808 more frames toward the outermost frame. Use
24809 @code{gdb.frame_stop_reason_string} to convert the value returned by this
24810 function to a string. The value can be one of:
24813 @item gdb.FRAME_UNWIND_NO_REASON
24814 No particular reason (older frames should be available).
24816 @item gdb.FRAME_UNWIND_NULL_ID
24817 The previous frame's analyzer returns an invalid result.
24819 @item gdb.FRAME_UNWIND_OUTERMOST
24820 This frame is the outermost.
24822 @item gdb.FRAME_UNWIND_UNAVAILABLE
24823 Cannot unwind further, because that would require knowing the
24824 values of registers or memory that have not been collected.
24826 @item gdb.FRAME_UNWIND_INNER_ID
24827 This frame ID looks like it ought to belong to a NEXT frame,
24828 but we got it for a PREV frame. Normally, this is a sign of
24829 unwinder failure. It could also indicate stack corruption.
24831 @item gdb.FRAME_UNWIND_SAME_ID
24832 This frame has the same ID as the previous one. That means
24833 that unwinding further would almost certainly give us another
24834 frame with exactly the same ID, so break the chain. Normally,
24835 this is a sign of unwinder failure. It could also indicate
24838 @item gdb.FRAME_UNWIND_NO_SAVED_PC
24839 The frame unwinder did not find any saved PC, but we needed
24840 one to unwind further.
24842 @item gdb.FRAME_UNWIND_FIRST_ERROR
24843 Any stop reason greater or equal to this value indicates some kind
24844 of error. This special value facilitates writing code that tests
24845 for errors in unwinding in a way that will work correctly even if
24846 the list of the other values is modified in future @value{GDBN}
24847 versions. Using it, you could write:
24849 reason = gdb.selected_frame().unwind_stop_reason ()
24850 reason_str = gdb.frame_stop_reason_string (reason)
24851 if reason >= gdb.FRAME_UNWIND_FIRST_ERROR:
24852 print "An error occured: %s" % reason_str
24859 Returns the frame's resume address.
24862 @defun Frame.block ()
24863 Return the frame's code block. @xref{Blocks In Python}.
24866 @defun Frame.function ()
24867 Return the symbol for the function corresponding to this frame.
24868 @xref{Symbols In Python}.
24871 @defun Frame.older ()
24872 Return the frame that called this frame.
24875 @defun Frame.newer ()
24876 Return the frame called by this frame.
24879 @defun Frame.find_sal ()
24880 Return the frame's symtab and line object.
24881 @xref{Symbol Tables In Python}.
24884 @defun Frame.read_var (variable @r{[}, block@r{]})
24885 Return the value of @var{variable} in this frame. If the optional
24886 argument @var{block} is provided, search for the variable from that
24887 block; otherwise start at the frame's current block (which is
24888 determined by the frame's current program counter). @var{variable}
24889 must be a string or a @code{gdb.Symbol} object. @var{block} must be a
24890 @code{gdb.Block} object.
24893 @defun Frame.select ()
24894 Set this frame to be the selected frame. @xref{Stack, ,Examining the
24899 @node Blocks In Python
24900 @subsubsection Accessing frame blocks from Python.
24902 @cindex blocks in python
24905 Within each frame, @value{GDBN} maintains information on each block
24906 stored in that frame. These blocks are organized hierarchically, and
24907 are represented individually in Python as a @code{gdb.Block}.
24908 Please see @ref{Frames In Python}, for a more in-depth discussion on
24909 frames. Furthermore, see @ref{Stack, ,Examining the Stack}, for more
24910 detailed technical information on @value{GDBN}'s book-keeping of the
24913 A @code{gdb.Block} is iterable. The iterator returns the symbols
24914 (@pxref{Symbols In Python}) local to the block. Python programs
24915 should not assume that a specific block object will always contain a
24916 given symbol, since changes in @value{GDBN} features and
24917 infrastructure may cause symbols move across blocks in a symbol
24920 The following block-related functions are available in the @code{gdb}
24923 @findex gdb.block_for_pc
24924 @defun gdb.block_for_pc (pc)
24925 Return the @code{gdb.Block} containing the given @var{pc} value. If the
24926 block cannot be found for the @var{pc} value specified, the function
24927 will return @code{None}.
24930 A @code{gdb.Block} object has the following methods:
24933 @defun Block.is_valid ()
24934 Returns @code{True} if the @code{gdb.Block} object is valid,
24935 @code{False} if not. A block object can become invalid if the block it
24936 refers to doesn't exist anymore in the inferior. All other
24937 @code{gdb.Block} methods will throw an exception if it is invalid at
24938 the time the method is called. The block's validity is also checked
24939 during iteration over symbols of the block.
24943 A @code{gdb.Block} object has the following attributes:
24946 @defvar Block.start
24947 The start address of the block. This attribute is not writable.
24951 The end address of the block. This attribute is not writable.
24954 @defvar Block.function
24955 The name of the block represented as a @code{gdb.Symbol}. If the
24956 block is not named, then this attribute holds @code{None}. This
24957 attribute is not writable.
24960 @defvar Block.superblock
24961 The block containing this block. If this parent block does not exist,
24962 this attribute holds @code{None}. This attribute is not writable.
24965 @defvar Block.global_block
24966 The global block associated with this block. This attribute is not
24970 @defvar Block.static_block
24971 The static block associated with this block. This attribute is not
24975 @defvar Block.is_global
24976 @code{True} if the @code{gdb.Block} object is a global block,
24977 @code{False} if not. This attribute is not
24981 @defvar Block.is_static
24982 @code{True} if the @code{gdb.Block} object is a static block,
24983 @code{False} if not. This attribute is not writable.
24987 @node Symbols In Python
24988 @subsubsection Python representation of Symbols.
24990 @cindex symbols in python
24993 @value{GDBN} represents every variable, function and type as an
24994 entry in a symbol table. @xref{Symbols, ,Examining the Symbol Table}.
24995 Similarly, Python represents these symbols in @value{GDBN} with the
24996 @code{gdb.Symbol} object.
24998 The following symbol-related functions are available in the @code{gdb}
25001 @findex gdb.lookup_symbol
25002 @defun gdb.lookup_symbol (name @r{[}, block @r{[}, domain@r{]]})
25003 This function searches for a symbol by name. The search scope can be
25004 restricted to the parameters defined in the optional domain and block
25007 @var{name} is the name of the symbol. It must be a string. The
25008 optional @var{block} argument restricts the search to symbols visible
25009 in that @var{block}. The @var{block} argument must be a
25010 @code{gdb.Block} object. If omitted, the block for the current frame
25011 is used. The optional @var{domain} argument restricts
25012 the search to the domain type. The @var{domain} argument must be a
25013 domain constant defined in the @code{gdb} module and described later
25016 The result is a tuple of two elements.
25017 The first element is a @code{gdb.Symbol} object or @code{None} if the symbol
25019 If the symbol is found, the second element is @code{True} if the symbol
25020 is a field of a method's object (e.g., @code{this} in C@t{++}),
25021 otherwise it is @code{False}.
25022 If the symbol is not found, the second element is @code{False}.
25025 @findex gdb.lookup_global_symbol
25026 @defun gdb.lookup_global_symbol (name @r{[}, domain@r{]})
25027 This function searches for a global symbol by name.
25028 The search scope can be restricted to by the domain argument.
25030 @var{name} is the name of the symbol. It must be a string.
25031 The optional @var{domain} argument restricts the search to the domain type.
25032 The @var{domain} argument must be a domain constant defined in the @code{gdb}
25033 module and described later in this chapter.
25035 The result is a @code{gdb.Symbol} object or @code{None} if the symbol
25039 A @code{gdb.Symbol} object has the following attributes:
25042 @defvar Symbol.type
25043 The type of the symbol or @code{None} if no type is recorded.
25044 This attribute is represented as a @code{gdb.Type} object.
25045 @xref{Types In Python}. This attribute is not writable.
25048 @defvar Symbol.symtab
25049 The symbol table in which the symbol appears. This attribute is
25050 represented as a @code{gdb.Symtab} object. @xref{Symbol Tables In
25051 Python}. This attribute is not writable.
25054 @defvar Symbol.line
25055 The line number in the source code at which the symbol was defined.
25056 This is an integer.
25059 @defvar Symbol.name
25060 The name of the symbol as a string. This attribute is not writable.
25063 @defvar Symbol.linkage_name
25064 The name of the symbol, as used by the linker (i.e., may be mangled).
25065 This attribute is not writable.
25068 @defvar Symbol.print_name
25069 The name of the symbol in a form suitable for output. This is either
25070 @code{name} or @code{linkage_name}, depending on whether the user
25071 asked @value{GDBN} to display demangled or mangled names.
25074 @defvar Symbol.addr_class
25075 The address class of the symbol. This classifies how to find the value
25076 of a symbol. Each address class is a constant defined in the
25077 @code{gdb} module and described later in this chapter.
25080 @defvar Symbol.needs_frame
25081 This is @code{True} if evaluating this symbol's value requires a frame
25082 (@pxref{Frames In Python}) and @code{False} otherwise. Typically,
25083 local variables will require a frame, but other symbols will not.
25086 @defvar Symbol.is_argument
25087 @code{True} if the symbol is an argument of a function.
25090 @defvar Symbol.is_constant
25091 @code{True} if the symbol is a constant.
25094 @defvar Symbol.is_function
25095 @code{True} if the symbol is a function or a method.
25098 @defvar Symbol.is_variable
25099 @code{True} if the symbol is a variable.
25103 A @code{gdb.Symbol} object has the following methods:
25106 @defun Symbol.is_valid ()
25107 Returns @code{True} if the @code{gdb.Symbol} object is valid,
25108 @code{False} if not. A @code{gdb.Symbol} object can become invalid if
25109 the symbol it refers to does not exist in @value{GDBN} any longer.
25110 All other @code{gdb.Symbol} methods will throw an exception if it is
25111 invalid at the time the method is called.
25114 @defun Symbol.value (@r{[}frame@r{]})
25115 Compute the value of the symbol, as a @code{gdb.Value}. For
25116 functions, this computes the address of the function, cast to the
25117 appropriate type. If the symbol requires a frame in order to compute
25118 its value, then @var{frame} must be given. If @var{frame} is not
25119 given, or if @var{frame} is invalid, then this method will throw an
25124 The available domain categories in @code{gdb.Symbol} are represented
25125 as constants in the @code{gdb} module:
25128 @findex SYMBOL_UNDEF_DOMAIN
25129 @findex gdb.SYMBOL_UNDEF_DOMAIN
25130 @item gdb.SYMBOL_UNDEF_DOMAIN
25131 This is used when a domain has not been discovered or none of the
25132 following domains apply. This usually indicates an error either
25133 in the symbol information or in @value{GDBN}'s handling of symbols.
25134 @findex SYMBOL_VAR_DOMAIN
25135 @findex gdb.SYMBOL_VAR_DOMAIN
25136 @item gdb.SYMBOL_VAR_DOMAIN
25137 This domain contains variables, function names, typedef names and enum
25139 @findex SYMBOL_STRUCT_DOMAIN
25140 @findex gdb.SYMBOL_STRUCT_DOMAIN
25141 @item gdb.SYMBOL_STRUCT_DOMAIN
25142 This domain holds struct, union and enum type names.
25143 @findex SYMBOL_LABEL_DOMAIN
25144 @findex gdb.SYMBOL_LABEL_DOMAIN
25145 @item gdb.SYMBOL_LABEL_DOMAIN
25146 This domain contains names of labels (for gotos).
25147 @findex SYMBOL_VARIABLES_DOMAIN
25148 @findex gdb.SYMBOL_VARIABLES_DOMAIN
25149 @item gdb.SYMBOL_VARIABLES_DOMAIN
25150 This domain holds a subset of the @code{SYMBOLS_VAR_DOMAIN}; it
25151 contains everything minus functions and types.
25152 @findex SYMBOL_FUNCTIONS_DOMAIN
25153 @findex gdb.SYMBOL_FUNCTIONS_DOMAIN
25154 @item gdb.SYMBOL_FUNCTION_DOMAIN
25155 This domain contains all functions.
25156 @findex SYMBOL_TYPES_DOMAIN
25157 @findex gdb.SYMBOL_TYPES_DOMAIN
25158 @item gdb.SYMBOL_TYPES_DOMAIN
25159 This domain contains all types.
25162 The available address class categories in @code{gdb.Symbol} are represented
25163 as constants in the @code{gdb} module:
25166 @findex SYMBOL_LOC_UNDEF
25167 @findex gdb.SYMBOL_LOC_UNDEF
25168 @item gdb.SYMBOL_LOC_UNDEF
25169 If this is returned by address class, it indicates an error either in
25170 the symbol information or in @value{GDBN}'s handling of symbols.
25171 @findex SYMBOL_LOC_CONST
25172 @findex gdb.SYMBOL_LOC_CONST
25173 @item gdb.SYMBOL_LOC_CONST
25174 Value is constant int.
25175 @findex SYMBOL_LOC_STATIC
25176 @findex gdb.SYMBOL_LOC_STATIC
25177 @item gdb.SYMBOL_LOC_STATIC
25178 Value is at a fixed address.
25179 @findex SYMBOL_LOC_REGISTER
25180 @findex gdb.SYMBOL_LOC_REGISTER
25181 @item gdb.SYMBOL_LOC_REGISTER
25182 Value is in a register.
25183 @findex SYMBOL_LOC_ARG
25184 @findex gdb.SYMBOL_LOC_ARG
25185 @item gdb.SYMBOL_LOC_ARG
25186 Value is an argument. This value is at the offset stored within the
25187 symbol inside the frame's argument list.
25188 @findex SYMBOL_LOC_REF_ARG
25189 @findex gdb.SYMBOL_LOC_REF_ARG
25190 @item gdb.SYMBOL_LOC_REF_ARG
25191 Value address is stored in the frame's argument list. Just like
25192 @code{LOC_ARG} except that the value's address is stored at the
25193 offset, not the value itself.
25194 @findex SYMBOL_LOC_REGPARM_ADDR
25195 @findex gdb.SYMBOL_LOC_REGPARM_ADDR
25196 @item gdb.SYMBOL_LOC_REGPARM_ADDR
25197 Value is a specified register. Just like @code{LOC_REGISTER} except
25198 the register holds the address of the argument instead of the argument
25200 @findex SYMBOL_LOC_LOCAL
25201 @findex gdb.SYMBOL_LOC_LOCAL
25202 @item gdb.SYMBOL_LOC_LOCAL
25203 Value is a local variable.
25204 @findex SYMBOL_LOC_TYPEDEF
25205 @findex gdb.SYMBOL_LOC_TYPEDEF
25206 @item gdb.SYMBOL_LOC_TYPEDEF
25207 Value not used. Symbols in the domain @code{SYMBOL_STRUCT_DOMAIN} all
25209 @findex SYMBOL_LOC_BLOCK
25210 @findex gdb.SYMBOL_LOC_BLOCK
25211 @item gdb.SYMBOL_LOC_BLOCK
25213 @findex SYMBOL_LOC_CONST_BYTES
25214 @findex gdb.SYMBOL_LOC_CONST_BYTES
25215 @item gdb.SYMBOL_LOC_CONST_BYTES
25216 Value is a byte-sequence.
25217 @findex SYMBOL_LOC_UNRESOLVED
25218 @findex gdb.SYMBOL_LOC_UNRESOLVED
25219 @item gdb.SYMBOL_LOC_UNRESOLVED
25220 Value is at a fixed address, but the address of the variable has to be
25221 determined from the minimal symbol table whenever the variable is
25223 @findex SYMBOL_LOC_OPTIMIZED_OUT
25224 @findex gdb.SYMBOL_LOC_OPTIMIZED_OUT
25225 @item gdb.SYMBOL_LOC_OPTIMIZED_OUT
25226 The value does not actually exist in the program.
25227 @findex SYMBOL_LOC_COMPUTED
25228 @findex gdb.SYMBOL_LOC_COMPUTED
25229 @item gdb.SYMBOL_LOC_COMPUTED
25230 The value's address is a computed location.
25233 @node Symbol Tables In Python
25234 @subsubsection Symbol table representation in Python.
25236 @cindex symbol tables in python
25238 @tindex gdb.Symtab_and_line
25240 Access to symbol table data maintained by @value{GDBN} on the inferior
25241 is exposed to Python via two objects: @code{gdb.Symtab_and_line} and
25242 @code{gdb.Symtab}. Symbol table and line data for a frame is returned
25243 from the @code{find_sal} method in @code{gdb.Frame} object.
25244 @xref{Frames In Python}.
25246 For more information on @value{GDBN}'s symbol table management, see
25247 @ref{Symbols, ,Examining the Symbol Table}, for more information.
25249 A @code{gdb.Symtab_and_line} object has the following attributes:
25252 @defvar Symtab_and_line.symtab
25253 The symbol table object (@code{gdb.Symtab}) for this frame.
25254 This attribute is not writable.
25257 @defvar Symtab_and_line.pc
25258 Indicates the current program counter address. This attribute is not
25262 @defvar Symtab_and_line.line
25263 Indicates the current line number for this object. This
25264 attribute is not writable.
25268 A @code{gdb.Symtab_and_line} object has the following methods:
25271 @defun Symtab_and_line.is_valid ()
25272 Returns @code{True} if the @code{gdb.Symtab_and_line} object is valid,
25273 @code{False} if not. A @code{gdb.Symtab_and_line} object can become
25274 invalid if the Symbol table and line object it refers to does not
25275 exist in @value{GDBN} any longer. All other
25276 @code{gdb.Symtab_and_line} methods will throw an exception if it is
25277 invalid at the time the method is called.
25281 A @code{gdb.Symtab} object has the following attributes:
25284 @defvar Symtab.filename
25285 The symbol table's source filename. This attribute is not writable.
25288 @defvar Symtab.objfile
25289 The symbol table's backing object file. @xref{Objfiles In Python}.
25290 This attribute is not writable.
25294 A @code{gdb.Symtab} object has the following methods:
25297 @defun Symtab.is_valid ()
25298 Returns @code{True} if the @code{gdb.Symtab} object is valid,
25299 @code{False} if not. A @code{gdb.Symtab} object can become invalid if
25300 the symbol table it refers to does not exist in @value{GDBN} any
25301 longer. All other @code{gdb.Symtab} methods will throw an exception
25302 if it is invalid at the time the method is called.
25305 @defun Symtab.fullname ()
25306 Return the symbol table's source absolute file name.
25309 @defun Symtab.global_block ()
25310 Return the global block of the underlying symbol table.
25311 @xref{Blocks In Python}.
25314 @defun Symtab.static_block ()
25315 Return the static block of the underlying symbol table.
25316 @xref{Blocks In Python}.
25320 @node Breakpoints In Python
25321 @subsubsection Manipulating breakpoints using Python
25323 @cindex breakpoints in python
25324 @tindex gdb.Breakpoint
25326 Python code can manipulate breakpoints via the @code{gdb.Breakpoint}
25329 @defun Breakpoint.__init__ (spec @r{[}, type @r{[}, wp_class @r{[},internal@r{]]]})
25330 Create a new breakpoint. @var{spec} is a string naming the
25331 location of the breakpoint, or an expression that defines a
25332 watchpoint. The contents can be any location recognized by the
25333 @code{break} command, or in the case of a watchpoint, by the @code{watch}
25334 command. The optional @var{type} denotes the breakpoint to create
25335 from the types defined later in this chapter. This argument can be
25336 either: @code{gdb.BP_BREAKPOINT} or @code{gdb.BP_WATCHPOINT}. @var{type}
25337 defaults to @code{gdb.BP_BREAKPOINT}. The optional @var{internal} argument
25338 allows the breakpoint to become invisible to the user. The breakpoint
25339 will neither be reported when created, nor will it be listed in the
25340 output from @code{info breakpoints} (but will be listed with the
25341 @code{maint info breakpoints} command). The optional @var{wp_class}
25342 argument defines the class of watchpoint to create, if @var{type} is
25343 @code{gdb.BP_WATCHPOINT}. If a watchpoint class is not provided, it is
25344 assumed to be a @code{gdb.WP_WRITE} class.
25347 @defun Breakpoint.stop (self)
25348 The @code{gdb.Breakpoint} class can be sub-classed and, in
25349 particular, you may choose to implement the @code{stop} method.
25350 If this method is defined as a sub-class of @code{gdb.Breakpoint},
25351 it will be called when the inferior reaches any location of a
25352 breakpoint which instantiates that sub-class. If the method returns
25353 @code{True}, the inferior will be stopped at the location of the
25354 breakpoint, otherwise the inferior will continue.
25356 If there are multiple breakpoints at the same location with a
25357 @code{stop} method, each one will be called regardless of the
25358 return status of the previous. This ensures that all @code{stop}
25359 methods have a chance to execute at that location. In this scenario
25360 if one of the methods returns @code{True} but the others return
25361 @code{False}, the inferior will still be stopped.
25363 You should not alter the execution state of the inferior (i.e.@:, step,
25364 next, etc.), alter the current frame context (i.e.@:, change the current
25365 active frame), or alter, add or delete any breakpoint. As a general
25366 rule, you should not alter any data within @value{GDBN} or the inferior
25369 Example @code{stop} implementation:
25372 class MyBreakpoint (gdb.Breakpoint):
25374 inf_val = gdb.parse_and_eval("foo")
25381 The available watchpoint types represented by constants are defined in the
25386 @findex gdb.WP_READ
25388 Read only watchpoint.
25391 @findex gdb.WP_WRITE
25393 Write only watchpoint.
25396 @findex gdb.WP_ACCESS
25397 @item gdb.WP_ACCESS
25398 Read/Write watchpoint.
25401 @defun Breakpoint.is_valid ()
25402 Return @code{True} if this @code{Breakpoint} object is valid,
25403 @code{False} otherwise. A @code{Breakpoint} object can become invalid
25404 if the user deletes the breakpoint. In this case, the object still
25405 exists, but the underlying breakpoint does not. In the cases of
25406 watchpoint scope, the watchpoint remains valid even if execution of the
25407 inferior leaves the scope of that watchpoint.
25410 @defun Breakpoint.delete
25411 Permanently deletes the @value{GDBN} breakpoint. This also
25412 invalidates the Python @code{Breakpoint} object. Any further access
25413 to this object's attributes or methods will raise an error.
25416 @defvar Breakpoint.enabled
25417 This attribute is @code{True} if the breakpoint is enabled, and
25418 @code{False} otherwise. This attribute is writable.
25421 @defvar Breakpoint.silent
25422 This attribute is @code{True} if the breakpoint is silent, and
25423 @code{False} otherwise. This attribute is writable.
25425 Note that a breakpoint can also be silent if it has commands and the
25426 first command is @code{silent}. This is not reported by the
25427 @code{silent} attribute.
25430 @defvar Breakpoint.thread
25431 If the breakpoint is thread-specific, this attribute holds the thread
25432 id. If the breakpoint is not thread-specific, this attribute is
25433 @code{None}. This attribute is writable.
25436 @defvar Breakpoint.task
25437 If the breakpoint is Ada task-specific, this attribute holds the Ada task
25438 id. If the breakpoint is not task-specific (or the underlying
25439 language is not Ada), this attribute is @code{None}. This attribute
25443 @defvar Breakpoint.ignore_count
25444 This attribute holds the ignore count for the breakpoint, an integer.
25445 This attribute is writable.
25448 @defvar Breakpoint.number
25449 This attribute holds the breakpoint's number --- the identifier used by
25450 the user to manipulate the breakpoint. This attribute is not writable.
25453 @defvar Breakpoint.type
25454 This attribute holds the breakpoint's type --- the identifier used to
25455 determine the actual breakpoint type or use-case. This attribute is not
25459 @defvar Breakpoint.visible
25460 This attribute tells whether the breakpoint is visible to the user
25461 when set, or when the @samp{info breakpoints} command is run. This
25462 attribute is not writable.
25465 The available types are represented by constants defined in the @code{gdb}
25469 @findex BP_BREAKPOINT
25470 @findex gdb.BP_BREAKPOINT
25471 @item gdb.BP_BREAKPOINT
25472 Normal code breakpoint.
25474 @findex BP_WATCHPOINT
25475 @findex gdb.BP_WATCHPOINT
25476 @item gdb.BP_WATCHPOINT
25477 Watchpoint breakpoint.
25479 @findex BP_HARDWARE_WATCHPOINT
25480 @findex gdb.BP_HARDWARE_WATCHPOINT
25481 @item gdb.BP_HARDWARE_WATCHPOINT
25482 Hardware assisted watchpoint.
25484 @findex BP_READ_WATCHPOINT
25485 @findex gdb.BP_READ_WATCHPOINT
25486 @item gdb.BP_READ_WATCHPOINT
25487 Hardware assisted read watchpoint.
25489 @findex BP_ACCESS_WATCHPOINT
25490 @findex gdb.BP_ACCESS_WATCHPOINT
25491 @item gdb.BP_ACCESS_WATCHPOINT
25492 Hardware assisted access watchpoint.
25495 @defvar Breakpoint.hit_count
25496 This attribute holds the hit count for the breakpoint, an integer.
25497 This attribute is writable, but currently it can only be set to zero.
25500 @defvar Breakpoint.location
25501 This attribute holds the location of the breakpoint, as specified by
25502 the user. It is a string. If the breakpoint does not have a location
25503 (that is, it is a watchpoint) the attribute's value is @code{None}. This
25504 attribute is not writable.
25507 @defvar Breakpoint.expression
25508 This attribute holds a breakpoint expression, as specified by
25509 the user. It is a string. If the breakpoint does not have an
25510 expression (the breakpoint is not a watchpoint) the attribute's value
25511 is @code{None}. This attribute is not writable.
25514 @defvar Breakpoint.condition
25515 This attribute holds the condition of the breakpoint, as specified by
25516 the user. It is a string. If there is no condition, this attribute's
25517 value is @code{None}. This attribute is writable.
25520 @defvar Breakpoint.commands
25521 This attribute holds the commands attached to the breakpoint. If
25522 there are commands, this attribute's value is a string holding all the
25523 commands, separated by newlines. If there are no commands, this
25524 attribute is @code{None}. This attribute is not writable.
25527 @node Finish Breakpoints in Python
25528 @subsubsection Finish Breakpoints
25530 @cindex python finish breakpoints
25531 @tindex gdb.FinishBreakpoint
25533 A finish breakpoint is a temporary breakpoint set at the return address of
25534 a frame, based on the @code{finish} command. @code{gdb.FinishBreakpoint}
25535 extends @code{gdb.Breakpoint}. The underlying breakpoint will be disabled
25536 and deleted when the execution will run out of the breakpoint scope (i.e.@:
25537 @code{Breakpoint.stop} or @code{FinishBreakpoint.out_of_scope} triggered).
25538 Finish breakpoints are thread specific and must be create with the right
25541 @defun FinishBreakpoint.__init__ (@r{[}frame@r{]} @r{[}, internal@r{]})
25542 Create a finish breakpoint at the return address of the @code{gdb.Frame}
25543 object @var{frame}. If @var{frame} is not provided, this defaults to the
25544 newest frame. The optional @var{internal} argument allows the breakpoint to
25545 become invisible to the user. @xref{Breakpoints In Python}, for further
25546 details about this argument.
25549 @defun FinishBreakpoint.out_of_scope (self)
25550 In some circumstances (e.g.@: @code{longjmp}, C@t{++} exceptions, @value{GDBN}
25551 @code{return} command, @dots{}), a function may not properly terminate, and
25552 thus never hit the finish breakpoint. When @value{GDBN} notices such a
25553 situation, the @code{out_of_scope} callback will be triggered.
25555 You may want to sub-class @code{gdb.FinishBreakpoint} and override this
25559 class MyFinishBreakpoint (gdb.FinishBreakpoint)
25561 print "normal finish"
25564 def out_of_scope ():
25565 print "abnormal finish"
25569 @defvar FinishBreakpoint.return_value
25570 When @value{GDBN} is stopped at a finish breakpoint and the frame
25571 used to build the @code{gdb.FinishBreakpoint} object had debug symbols, this
25572 attribute will contain a @code{gdb.Value} object corresponding to the return
25573 value of the function. The value will be @code{None} if the function return
25574 type is @code{void} or if the return value was not computable. This attribute
25578 @node Lazy Strings In Python
25579 @subsubsection Python representation of lazy strings.
25581 @cindex lazy strings in python
25582 @tindex gdb.LazyString
25584 A @dfn{lazy string} is a string whose contents is not retrieved or
25585 encoded until it is needed.
25587 A @code{gdb.LazyString} is represented in @value{GDBN} as an
25588 @code{address} that points to a region of memory, an @code{encoding}
25589 that will be used to encode that region of memory, and a @code{length}
25590 to delimit the region of memory that represents the string. The
25591 difference between a @code{gdb.LazyString} and a string wrapped within
25592 a @code{gdb.Value} is that a @code{gdb.LazyString} will be treated
25593 differently by @value{GDBN} when printing. A @code{gdb.LazyString} is
25594 retrieved and encoded during printing, while a @code{gdb.Value}
25595 wrapping a string is immediately retrieved and encoded on creation.
25597 A @code{gdb.LazyString} object has the following functions:
25599 @defun LazyString.value ()
25600 Convert the @code{gdb.LazyString} to a @code{gdb.Value}. This value
25601 will point to the string in memory, but will lose all the delayed
25602 retrieval, encoding and handling that @value{GDBN} applies to a
25603 @code{gdb.LazyString}.
25606 @defvar LazyString.address
25607 This attribute holds the address of the string. This attribute is not
25611 @defvar LazyString.length
25612 This attribute holds the length of the string in characters. If the
25613 length is -1, then the string will be fetched and encoded up to the
25614 first null of appropriate width. This attribute is not writable.
25617 @defvar LazyString.encoding
25618 This attribute holds the encoding that will be applied to the string
25619 when the string is printed by @value{GDBN}. If the encoding is not
25620 set, or contains an empty string, then @value{GDBN} will select the
25621 most appropriate encoding when the string is printed. This attribute
25625 @defvar LazyString.type
25626 This attribute holds the type that is represented by the lazy string's
25627 type. For a lazy string this will always be a pointer type. To
25628 resolve this to the lazy string's character type, use the type's
25629 @code{target} method. @xref{Types In Python}. This attribute is not
25633 @node Python Auto-loading
25634 @subsection Python Auto-loading
25635 @cindex Python auto-loading
25637 When a new object file is read (for example, due to the @code{file}
25638 command, or because the inferior has loaded a shared library),
25639 @value{GDBN} will look for Python support scripts in several ways:
25640 @file{@var{objfile}-gdb.py} (@pxref{objfile-gdb.py file})
25641 and @code{.debug_gdb_scripts} section
25642 (@pxref{dotdebug_gdb_scripts section}).
25644 The auto-loading feature is useful for supplying application-specific
25645 debugging commands and scripts.
25647 Auto-loading can be enabled or disabled,
25648 and the list of auto-loaded scripts can be printed.
25651 @anchor{set auto-load python-scripts}
25652 @kindex set auto-load python-scripts
25653 @item set auto-load python-scripts [on|off]
25654 Enable or disable the auto-loading of Python scripts.
25656 @anchor{show auto-load python-scripts}
25657 @kindex show auto-load python-scripts
25658 @item show auto-load python-scripts
25659 Show whether auto-loading of Python scripts is enabled or disabled.
25661 @anchor{info auto-load python-scripts}
25662 @kindex info auto-load python-scripts
25663 @cindex print list of auto-loaded Python scripts
25664 @item info auto-load python-scripts [@var{regexp}]
25665 Print the list of all Python scripts that @value{GDBN} auto-loaded.
25667 Also printed is the list of Python scripts that were mentioned in
25668 the @code{.debug_gdb_scripts} section and were not found
25669 (@pxref{dotdebug_gdb_scripts section}).
25670 This is useful because their names are not printed when @value{GDBN}
25671 tries to load them and fails. There may be many of them, and printing
25672 an error message for each one is problematic.
25674 If @var{regexp} is supplied only Python scripts with matching names are printed.
25679 (gdb) info auto-load python-scripts
25681 Yes py-section-script.py
25682 full name: /tmp/py-section-script.py
25683 No my-foo-pretty-printers.py
25687 When reading an auto-loaded file, @value{GDBN} sets the
25688 @dfn{current objfile}. This is available via the @code{gdb.current_objfile}
25689 function (@pxref{Objfiles In Python}). This can be useful for
25690 registering objfile-specific pretty-printers.
25693 * objfile-gdb.py file:: The @file{@var{objfile}-gdb.py} file
25694 * dotdebug_gdb_scripts section:: The @code{.debug_gdb_scripts} section
25695 * Which flavor to choose?::
25698 @node objfile-gdb.py file
25699 @subsubsection The @file{@var{objfile}-gdb.py} file
25700 @cindex @file{@var{objfile}-gdb.py}
25702 When a new object file is read, @value{GDBN} looks for
25703 a file named @file{@var{objfile}-gdb.py} (we call it @var{script-name} below),
25704 where @var{objfile} is the object file's real name, formed by ensuring
25705 that the file name is absolute, following all symlinks, and resolving
25706 @code{.} and @code{..} components. If this file exists and is
25707 readable, @value{GDBN} will evaluate it as a Python script.
25709 If this file does not exist, then @value{GDBN} will look for
25710 @var{script-name} file in all of the directories as specified below.
25712 Note that loading of this script file also requires accordingly configured
25713 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
25716 @anchor{set auto-load scripts-directory}
25717 @kindex set auto-load scripts-directory
25718 @item set auto-load scripts-directory @r{[}@var{directories}@r{]}
25719 Control @value{GDBN} auto-loaded scripts location. Multiple directory entries
25720 may be delimited by the host platform path separator in use
25721 (@samp{:} on Unix, @samp{;} on MS-Windows and MS-DOS).
25723 Each entry here needs to be covered also by the security setting
25724 @code{set auto-load safe-path} (@pxref{set auto-load safe-path}).
25726 @anchor{with-auto-load-dir}
25727 This variable defaults to @file{$debugdir:$datadir/auto-load}. The default
25728 @code{set auto-load safe-path} value can be also overriden by @value{GDBN}
25729 configuration option @option{--with-auto-load-dir}.
25731 Any reference to @file{$debugdir} will get replaced by
25732 @var{debug-file-directory} value (@pxref{Separate Debug Files}) and any
25733 reference to @file{$datadir} will get replaced by @var{data-directory} which is
25734 determined at @value{GDBN} startup (@pxref{Data Files}). @file{$debugdir} and
25735 @file{$datadir} must be placed as a directory component --- either alone or
25736 delimited by @file{/} or @file{\} directory separators, depending on the host
25739 The list of directories uses path separator (@samp{:} on GNU and Unix
25740 systems, @samp{;} on MS-Windows and MS-DOS) to separate directories, similarly
25741 to the @env{PATH} environment variable.
25743 @anchor{show auto-load scripts-directory}
25744 @kindex show auto-load scripts-directory
25745 @item show auto-load scripts-directory
25746 Show @value{GDBN} auto-loaded scripts location.
25749 @value{GDBN} does not track which files it has already auto-loaded this way.
25750 @value{GDBN} will load the associated script every time the corresponding
25751 @var{objfile} is opened.
25752 So your @file{-gdb.py} file should be careful to avoid errors if it
25753 is evaluated more than once.
25755 @node dotdebug_gdb_scripts section
25756 @subsubsection The @code{.debug_gdb_scripts} section
25757 @cindex @code{.debug_gdb_scripts} section
25759 For systems using file formats like ELF and COFF,
25760 when @value{GDBN} loads a new object file
25761 it will look for a special section named @samp{.debug_gdb_scripts}.
25762 If this section exists, its contents is a list of names of scripts to load.
25764 @value{GDBN} will look for each specified script file first in the
25765 current directory and then along the source search path
25766 (@pxref{Source Path, ,Specifying Source Directories}),
25767 except that @file{$cdir} is not searched, since the compilation
25768 directory is not relevant to scripts.
25770 Entries can be placed in section @code{.debug_gdb_scripts} with,
25771 for example, this GCC macro:
25774 /* Note: The "MS" section flags are to remove duplicates. */
25775 #define DEFINE_GDB_SCRIPT(script_name) \
25777 .pushsection \".debug_gdb_scripts\", \"MS\",@@progbits,1\n\
25779 .asciz \"" script_name "\"\n\
25785 Then one can reference the macro in a header or source file like this:
25788 DEFINE_GDB_SCRIPT ("my-app-scripts.py")
25791 The script name may include directories if desired.
25793 Note that loading of this script file also requires accordingly configured
25794 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
25796 If the macro is put in a header, any application or library
25797 using this header will get a reference to the specified script.
25799 @node Which flavor to choose?
25800 @subsubsection Which flavor to choose?
25802 Given the multiple ways of auto-loading Python scripts, it might not always
25803 be clear which one to choose. This section provides some guidance.
25805 Benefits of the @file{-gdb.py} way:
25809 Can be used with file formats that don't support multiple sections.
25812 Ease of finding scripts for public libraries.
25814 Scripts specified in the @code{.debug_gdb_scripts} section are searched for
25815 in the source search path.
25816 For publicly installed libraries, e.g., @file{libstdc++}, there typically
25817 isn't a source directory in which to find the script.
25820 Doesn't require source code additions.
25823 Benefits of the @code{.debug_gdb_scripts} way:
25827 Works with static linking.
25829 Scripts for libraries done the @file{-gdb.py} way require an objfile to
25830 trigger their loading. When an application is statically linked the only
25831 objfile available is the executable, and it is cumbersome to attach all the
25832 scripts from all the input libraries to the executable's @file{-gdb.py} script.
25835 Works with classes that are entirely inlined.
25837 Some classes can be entirely inlined, and thus there may not be an associated
25838 shared library to attach a @file{-gdb.py} script to.
25841 Scripts needn't be copied out of the source tree.
25843 In some circumstances, apps can be built out of large collections of internal
25844 libraries, and the build infrastructure necessary to install the
25845 @file{-gdb.py} scripts in a place where @value{GDBN} can find them is
25846 cumbersome. It may be easier to specify the scripts in the
25847 @code{.debug_gdb_scripts} section as relative paths, and add a path to the
25848 top of the source tree to the source search path.
25851 @node Python modules
25852 @subsection Python modules
25853 @cindex python modules
25855 @value{GDBN} comes with several modules to assist writing Python code.
25858 * gdb.printing:: Building and registering pretty-printers.
25859 * gdb.types:: Utilities for working with types.
25860 * gdb.prompt:: Utilities for prompt value substitution.
25864 @subsubsection gdb.printing
25865 @cindex gdb.printing
25867 This module provides a collection of utilities for working with
25871 @item PrettyPrinter (@var{name}, @var{subprinters}=None)
25872 This class specifies the API that makes @samp{info pretty-printer},
25873 @samp{enable pretty-printer} and @samp{disable pretty-printer} work.
25874 Pretty-printers should generally inherit from this class.
25876 @item SubPrettyPrinter (@var{name})
25877 For printers that handle multiple types, this class specifies the
25878 corresponding API for the subprinters.
25880 @item RegexpCollectionPrettyPrinter (@var{name})
25881 Utility class for handling multiple printers, all recognized via
25882 regular expressions.
25883 @xref{Writing a Pretty-Printer}, for an example.
25885 @item FlagEnumerationPrinter (@var{name})
25886 A pretty-printer which handles printing of @code{enum} values. Unlike
25887 @value{GDBN}'s built-in @code{enum} printing, this printer attempts to
25888 work properly when there is some overlap between the enumeration
25889 constants. @var{name} is the name of the printer and also the name of
25890 the @code{enum} type to look up.
25892 @item register_pretty_printer (@var{obj}, @var{printer}, @var{replace}=False)
25893 Register @var{printer} with the pretty-printer list of @var{obj}.
25894 If @var{replace} is @code{True} then any existing copy of the printer
25895 is replaced. Otherwise a @code{RuntimeError} exception is raised
25896 if a printer with the same name already exists.
25900 @subsubsection gdb.types
25903 This module provides a collection of utilities for working with
25904 @code{gdb.Types} objects.
25907 @item get_basic_type (@var{type})
25908 Return @var{type} with const and volatile qualifiers stripped,
25909 and with typedefs and C@t{++} references converted to the underlying type.
25914 typedef const int const_int;
25916 const_int& foo_ref (foo);
25917 int main () @{ return 0; @}
25924 (gdb) python import gdb.types
25925 (gdb) python foo_ref = gdb.parse_and_eval("foo_ref")
25926 (gdb) python print gdb.types.get_basic_type(foo_ref.type)
25930 @item has_field (@var{type}, @var{field})
25931 Return @code{True} if @var{type}, assumed to be a type with fields
25932 (e.g., a structure or union), has field @var{field}.
25934 @item make_enum_dict (@var{enum_type})
25935 Return a Python @code{dictionary} type produced from @var{enum_type}.
25937 @item deep_items (@var{type})
25938 Returns a Python iterator similar to the standard
25939 @code{gdb.Type.iteritems} method, except that the iterator returned
25940 by @code{deep_items} will recursively traverse anonymous struct or
25941 union fields. For example:
25955 Then in @value{GDBN}:
25957 (@value{GDBP}) python import gdb.types
25958 (@value{GDBP}) python struct_a = gdb.lookup_type("struct A")
25959 (@value{GDBP}) python print struct_a.keys ()
25961 (@value{GDBP}) python print [k for k,v in gdb.types.deep_items(struct_a)]
25962 @{['a', 'b0', 'b1']@}
25968 @subsubsection gdb.prompt
25971 This module provides a method for prompt value-substitution.
25974 @item substitute_prompt (@var{string})
25975 Return @var{string} with escape sequences substituted by values. Some
25976 escape sequences take arguments. You can specify arguments inside
25977 ``@{@}'' immediately following the escape sequence.
25979 The escape sequences you can pass to this function are:
25983 Substitute a backslash.
25985 Substitute an ESC character.
25987 Substitute the selected frame; an argument names a frame parameter.
25989 Substitute a newline.
25991 Substitute a parameter's value; the argument names the parameter.
25993 Substitute a carriage return.
25995 Substitute the selected thread; an argument names a thread parameter.
25997 Substitute the version of GDB.
25999 Substitute the current working directory.
26001 Begin a sequence of non-printing characters. These sequences are
26002 typically used with the ESC character, and are not counted in the string
26003 length. Example: ``\[\e[0;34m\](gdb)\[\e[0m\]'' will return a
26004 blue-colored ``(gdb)'' prompt where the length is five.
26006 End a sequence of non-printing characters.
26012 substitute_prompt (``frame: \f,
26013 print arguments: \p@{print frame-arguments@}'')
26016 @exdent will return the string:
26019 "frame: main, print arguments: scalars"
26024 @section Creating new spellings of existing commands
26025 @cindex aliases for commands
26027 It is often useful to define alternate spellings of existing commands.
26028 For example, if a new @value{GDBN} command defined in Python has
26029 a long name to type, it is handy to have an abbreviated version of it
26030 that involves less typing.
26032 @value{GDBN} itself uses aliases. For example @samp{s} is an alias
26033 of the @samp{step} command even though it is otherwise an ambiguous
26034 abbreviation of other commands like @samp{set} and @samp{show}.
26036 Aliases are also used to provide shortened or more common versions
26037 of multi-word commands. For example, @value{GDBN} provides the
26038 @samp{tty} alias of the @samp{set inferior-tty} command.
26040 You can define a new alias with the @samp{alias} command.
26045 @item alias [-a] [--] @var{ALIAS} = @var{COMMAND}
26049 @var{ALIAS} specifies the name of the new alias.
26050 Each word of @var{ALIAS} must consist of letters, numbers, dashes and
26053 @var{COMMAND} specifies the name of an existing command
26054 that is being aliased.
26056 The @samp{-a} option specifies that the new alias is an abbreviation
26057 of the command. Abbreviations are not shown in command
26058 lists displayed by the @samp{help} command.
26060 The @samp{--} option specifies the end of options,
26061 and is useful when @var{ALIAS} begins with a dash.
26063 Here is a simple example showing how to make an abbreviation
26064 of a command so that there is less to type.
26065 Suppose you were tired of typing @samp{disas}, the current
26066 shortest unambiguous abbreviation of the @samp{disassemble} command
26067 and you wanted an even shorter version named @samp{di}.
26068 The following will accomplish this.
26071 (gdb) alias -a di = disas
26074 Note that aliases are different from user-defined commands.
26075 With a user-defined command, you also need to write documentation
26076 for it with the @samp{document} command.
26077 An alias automatically picks up the documentation of the existing command.
26079 Here is an example where we make @samp{elms} an abbreviation of
26080 @samp{elements} in the @samp{set print elements} command.
26081 This is to show that you can make an abbreviation of any part
26085 (gdb) alias -a set print elms = set print elements
26086 (gdb) alias -a show print elms = show print elements
26087 (gdb) set p elms 20
26089 Limit on string chars or array elements to print is 200.
26092 Note that if you are defining an alias of a @samp{set} command,
26093 and you want to have an alias for the corresponding @samp{show}
26094 command, then you need to define the latter separately.
26096 Unambiguously abbreviated commands are allowed in @var{COMMAND} and
26097 @var{ALIAS}, just as they are normally.
26100 (gdb) alias -a set pr elms = set p ele
26103 Finally, here is an example showing the creation of a one word
26104 alias for a more complex command.
26105 This creates alias @samp{spe} of the command @samp{set print elements}.
26108 (gdb) alias spe = set print elements
26113 @chapter Command Interpreters
26114 @cindex command interpreters
26116 @value{GDBN} supports multiple command interpreters, and some command
26117 infrastructure to allow users or user interface writers to switch
26118 between interpreters or run commands in other interpreters.
26120 @value{GDBN} currently supports two command interpreters, the console
26121 interpreter (sometimes called the command-line interpreter or @sc{cli})
26122 and the machine interface interpreter (or @sc{gdb/mi}). This manual
26123 describes both of these interfaces in great detail.
26125 By default, @value{GDBN} will start with the console interpreter.
26126 However, the user may choose to start @value{GDBN} with another
26127 interpreter by specifying the @option{-i} or @option{--interpreter}
26128 startup options. Defined interpreters include:
26132 @cindex console interpreter
26133 The traditional console or command-line interpreter. This is the most often
26134 used interpreter with @value{GDBN}. With no interpreter specified at runtime,
26135 @value{GDBN} will use this interpreter.
26138 @cindex mi interpreter
26139 The newest @sc{gdb/mi} interface (currently @code{mi2}). Used primarily
26140 by programs wishing to use @value{GDBN} as a backend for a debugger GUI
26141 or an IDE. For more information, see @ref{GDB/MI, ,The @sc{gdb/mi}
26145 @cindex mi2 interpreter
26146 The current @sc{gdb/mi} interface.
26149 @cindex mi1 interpreter
26150 The @sc{gdb/mi} interface included in @value{GDBN} 5.1, 5.2, and 5.3.
26154 @cindex invoke another interpreter
26155 The interpreter being used by @value{GDBN} may not be dynamically
26156 switched at runtime. Although possible, this could lead to a very
26157 precarious situation. Consider an IDE using @sc{gdb/mi}. If a user
26158 enters the command "interpreter-set console" in a console view,
26159 @value{GDBN} would switch to using the console interpreter, rendering
26160 the IDE inoperable!
26162 @kindex interpreter-exec
26163 Although you may only choose a single interpreter at startup, you may execute
26164 commands in any interpreter from the current interpreter using the appropriate
26165 command. If you are running the console interpreter, simply use the
26166 @code{interpreter-exec} command:
26169 interpreter-exec mi "-data-list-register-names"
26172 @sc{gdb/mi} has a similar command, although it is only available in versions of
26173 @value{GDBN} which support @sc{gdb/mi} version 2 (or greater).
26176 @chapter @value{GDBN} Text User Interface
26178 @cindex Text User Interface
26181 * TUI Overview:: TUI overview
26182 * TUI Keys:: TUI key bindings
26183 * TUI Single Key Mode:: TUI single key mode
26184 * TUI Commands:: TUI-specific commands
26185 * TUI Configuration:: TUI configuration variables
26188 The @value{GDBN} Text User Interface (TUI) is a terminal
26189 interface which uses the @code{curses} library to show the source
26190 file, the assembly output, the program registers and @value{GDBN}
26191 commands in separate text windows. The TUI mode is supported only
26192 on platforms where a suitable version of the @code{curses} library
26195 The TUI mode is enabled by default when you invoke @value{GDBN} as
26196 @samp{@value{GDBP} -tui}.
26197 You can also switch in and out of TUI mode while @value{GDBN} runs by
26198 using various TUI commands and key bindings, such as @kbd{C-x C-a}.
26199 @xref{TUI Keys, ,TUI Key Bindings}.
26202 @section TUI Overview
26204 In TUI mode, @value{GDBN} can display several text windows:
26208 This window is the @value{GDBN} command window with the @value{GDBN}
26209 prompt and the @value{GDBN} output. The @value{GDBN} input is still
26210 managed using readline.
26213 The source window shows the source file of the program. The current
26214 line and active breakpoints are displayed in this window.
26217 The assembly window shows the disassembly output of the program.
26220 This window shows the processor registers. Registers are highlighted
26221 when their values change.
26224 The source and assembly windows show the current program position
26225 by highlighting the current line and marking it with a @samp{>} marker.
26226 Breakpoints are indicated with two markers. The first marker
26227 indicates the breakpoint type:
26231 Breakpoint which was hit at least once.
26234 Breakpoint which was never hit.
26237 Hardware breakpoint which was hit at least once.
26240 Hardware breakpoint which was never hit.
26243 The second marker indicates whether the breakpoint is enabled or not:
26247 Breakpoint is enabled.
26250 Breakpoint is disabled.
26253 The source, assembly and register windows are updated when the current
26254 thread changes, when the frame changes, or when the program counter
26257 These windows are not all visible at the same time. The command
26258 window is always visible. The others can be arranged in several
26269 source and assembly,
26272 source and registers, or
26275 assembly and registers.
26278 A status line above the command window shows the following information:
26282 Indicates the current @value{GDBN} target.
26283 (@pxref{Targets, ,Specifying a Debugging Target}).
26286 Gives the current process or thread number.
26287 When no process is being debugged, this field is set to @code{No process}.
26290 Gives the current function name for the selected frame.
26291 The name is demangled if demangling is turned on (@pxref{Print Settings}).
26292 When there is no symbol corresponding to the current program counter,
26293 the string @code{??} is displayed.
26296 Indicates the current line number for the selected frame.
26297 When the current line number is not known, the string @code{??} is displayed.
26300 Indicates the current program counter address.
26304 @section TUI Key Bindings
26305 @cindex TUI key bindings
26307 The TUI installs several key bindings in the readline keymaps
26308 @ifset SYSTEM_READLINE
26309 (@pxref{Command Line Editing, , , rluserman, GNU Readline Library}).
26311 @ifclear SYSTEM_READLINE
26312 (@pxref{Command Line Editing}).
26314 The following key bindings are installed for both TUI mode and the
26315 @value{GDBN} standard mode.
26324 Enter or leave the TUI mode. When leaving the TUI mode,
26325 the curses window management stops and @value{GDBN} operates using
26326 its standard mode, writing on the terminal directly. When reentering
26327 the TUI mode, control is given back to the curses windows.
26328 The screen is then refreshed.
26332 Use a TUI layout with only one window. The layout will
26333 either be @samp{source} or @samp{assembly}. When the TUI mode
26334 is not active, it will switch to the TUI mode.
26336 Think of this key binding as the Emacs @kbd{C-x 1} binding.
26340 Use a TUI layout with at least two windows. When the current
26341 layout already has two windows, the next layout with two windows is used.
26342 When a new layout is chosen, one window will always be common to the
26343 previous layout and the new one.
26345 Think of it as the Emacs @kbd{C-x 2} binding.
26349 Change the active window. The TUI associates several key bindings
26350 (like scrolling and arrow keys) with the active window. This command
26351 gives the focus to the next TUI window.
26353 Think of it as the Emacs @kbd{C-x o} binding.
26357 Switch in and out of the TUI SingleKey mode that binds single
26358 keys to @value{GDBN} commands (@pxref{TUI Single Key Mode}).
26361 The following key bindings only work in the TUI mode:
26366 Scroll the active window one page up.
26370 Scroll the active window one page down.
26374 Scroll the active window one line up.
26378 Scroll the active window one line down.
26382 Scroll the active window one column left.
26386 Scroll the active window one column right.
26390 Refresh the screen.
26393 Because the arrow keys scroll the active window in the TUI mode, they
26394 are not available for their normal use by readline unless the command
26395 window has the focus. When another window is active, you must use
26396 other readline key bindings such as @kbd{C-p}, @kbd{C-n}, @kbd{C-b}
26397 and @kbd{C-f} to control the command window.
26399 @node TUI Single Key Mode
26400 @section TUI Single Key Mode
26401 @cindex TUI single key mode
26403 The TUI also provides a @dfn{SingleKey} mode, which binds several
26404 frequently used @value{GDBN} commands to single keys. Type @kbd{C-x s} to
26405 switch into this mode, where the following key bindings are used:
26408 @kindex c @r{(SingleKey TUI key)}
26412 @kindex d @r{(SingleKey TUI key)}
26416 @kindex f @r{(SingleKey TUI key)}
26420 @kindex n @r{(SingleKey TUI key)}
26424 @kindex q @r{(SingleKey TUI key)}
26426 exit the SingleKey mode.
26428 @kindex r @r{(SingleKey TUI key)}
26432 @kindex s @r{(SingleKey TUI key)}
26436 @kindex u @r{(SingleKey TUI key)}
26440 @kindex v @r{(SingleKey TUI key)}
26444 @kindex w @r{(SingleKey TUI key)}
26449 Other keys temporarily switch to the @value{GDBN} command prompt.
26450 The key that was pressed is inserted in the editing buffer so that
26451 it is possible to type most @value{GDBN} commands without interaction
26452 with the TUI SingleKey mode. Once the command is entered the TUI
26453 SingleKey mode is restored. The only way to permanently leave
26454 this mode is by typing @kbd{q} or @kbd{C-x s}.
26458 @section TUI-specific Commands
26459 @cindex TUI commands
26461 The TUI has specific commands to control the text windows.
26462 These commands are always available, even when @value{GDBN} is not in
26463 the TUI mode. When @value{GDBN} is in the standard mode, most
26464 of these commands will automatically switch to the TUI mode.
26466 Note that if @value{GDBN}'s @code{stdout} is not connected to a
26467 terminal, or @value{GDBN} has been started with the machine interface
26468 interpreter (@pxref{GDB/MI, ,The @sc{gdb/mi} Interface}), most of
26469 these commands will fail with an error, because it would not be
26470 possible or desirable to enable curses window management.
26475 List and give the size of all displayed windows.
26479 Display the next layout.
26482 Display the previous layout.
26485 Display the source window only.
26488 Display the assembly window only.
26491 Display the source and assembly window.
26494 Display the register window together with the source or assembly window.
26498 Make the next window active for scrolling.
26501 Make the previous window active for scrolling.
26504 Make the source window active for scrolling.
26507 Make the assembly window active for scrolling.
26510 Make the register window active for scrolling.
26513 Make the command window active for scrolling.
26517 Refresh the screen. This is similar to typing @kbd{C-L}.
26519 @item tui reg float
26521 Show the floating point registers in the register window.
26523 @item tui reg general
26524 Show the general registers in the register window.
26527 Show the next register group. The list of register groups as well as
26528 their order is target specific. The predefined register groups are the
26529 following: @code{general}, @code{float}, @code{system}, @code{vector},
26530 @code{all}, @code{save}, @code{restore}.
26532 @item tui reg system
26533 Show the system registers in the register window.
26537 Update the source window and the current execution point.
26539 @item winheight @var{name} +@var{count}
26540 @itemx winheight @var{name} -@var{count}
26542 Change the height of the window @var{name} by @var{count}
26543 lines. Positive counts increase the height, while negative counts
26546 @item tabset @var{nchars}
26548 Set the width of tab stops to be @var{nchars} characters.
26551 @node TUI Configuration
26552 @section TUI Configuration Variables
26553 @cindex TUI configuration variables
26555 Several configuration variables control the appearance of TUI windows.
26558 @item set tui border-kind @var{kind}
26559 @kindex set tui border-kind
26560 Select the border appearance for the source, assembly and register windows.
26561 The possible values are the following:
26564 Use a space character to draw the border.
26567 Use @sc{ascii} characters @samp{+}, @samp{-} and @samp{|} to draw the border.
26570 Use the Alternate Character Set to draw the border. The border is
26571 drawn using character line graphics if the terminal supports them.
26574 @item set tui border-mode @var{mode}
26575 @kindex set tui border-mode
26576 @itemx set tui active-border-mode @var{mode}
26577 @kindex set tui active-border-mode
26578 Select the display attributes for the borders of the inactive windows
26579 or the active window. The @var{mode} can be one of the following:
26582 Use normal attributes to display the border.
26588 Use reverse video mode.
26591 Use half bright mode.
26593 @item half-standout
26594 Use half bright and standout mode.
26597 Use extra bright or bold mode.
26599 @item bold-standout
26600 Use extra bright or bold and standout mode.
26605 @chapter Using @value{GDBN} under @sc{gnu} Emacs
26608 @cindex @sc{gnu} Emacs
26609 A special interface allows you to use @sc{gnu} Emacs to view (and
26610 edit) the source files for the program you are debugging with
26613 To use this interface, use the command @kbd{M-x gdb} in Emacs. Give the
26614 executable file you want to debug as an argument. This command starts
26615 @value{GDBN} as a subprocess of Emacs, with input and output through a newly
26616 created Emacs buffer.
26617 @c (Do not use the @code{-tui} option to run @value{GDBN} from Emacs.)
26619 Running @value{GDBN} under Emacs can be just like running @value{GDBN} normally except for two
26624 All ``terminal'' input and output goes through an Emacs buffer, called
26627 This applies both to @value{GDBN} commands and their output, and to the input
26628 and output done by the program you are debugging.
26630 This is useful because it means that you can copy the text of previous
26631 commands and input them again; you can even use parts of the output
26634 All the facilities of Emacs' Shell mode are available for interacting
26635 with your program. In particular, you can send signals the usual
26636 way---for example, @kbd{C-c C-c} for an interrupt, @kbd{C-c C-z} for a
26640 @value{GDBN} displays source code through Emacs.
26642 Each time @value{GDBN} displays a stack frame, Emacs automatically finds the
26643 source file for that frame and puts an arrow (@samp{=>}) at the
26644 left margin of the current line. Emacs uses a separate buffer for
26645 source display, and splits the screen to show both your @value{GDBN} session
26648 Explicit @value{GDBN} @code{list} or search commands still produce output as
26649 usual, but you probably have no reason to use them from Emacs.
26652 We call this @dfn{text command mode}. Emacs 22.1, and later, also uses
26653 a graphical mode, enabled by default, which provides further buffers
26654 that can control the execution and describe the state of your program.
26655 @xref{GDB Graphical Interface,,, Emacs, The @sc{gnu} Emacs Manual}.
26657 If you specify an absolute file name when prompted for the @kbd{M-x
26658 gdb} argument, then Emacs sets your current working directory to where
26659 your program resides. If you only specify the file name, then Emacs
26660 sets your current working directory to the directory associated
26661 with the previous buffer. In this case, @value{GDBN} may find your
26662 program by searching your environment's @code{PATH} variable, but on
26663 some operating systems it might not find the source. So, although the
26664 @value{GDBN} input and output session proceeds normally, the auxiliary
26665 buffer does not display the current source and line of execution.
26667 The initial working directory of @value{GDBN} is printed on the top
26668 line of the GUD buffer and this serves as a default for the commands
26669 that specify files for @value{GDBN} to operate on. @xref{Files,
26670 ,Commands to Specify Files}.
26672 By default, @kbd{M-x gdb} calls the program called @file{gdb}. If you
26673 need to call @value{GDBN} by a different name (for example, if you
26674 keep several configurations around, with different names) you can
26675 customize the Emacs variable @code{gud-gdb-command-name} to run the
26678 In the GUD buffer, you can use these special Emacs commands in
26679 addition to the standard Shell mode commands:
26683 Describe the features of Emacs' GUD Mode.
26686 Execute to another source line, like the @value{GDBN} @code{step} command; also
26687 update the display window to show the current file and location.
26690 Execute to next source line in this function, skipping all function
26691 calls, like the @value{GDBN} @code{next} command. Then update the display window
26692 to show the current file and location.
26695 Execute one instruction, like the @value{GDBN} @code{stepi} command; update
26696 display window accordingly.
26699 Execute until exit from the selected stack frame, like the @value{GDBN}
26700 @code{finish} command.
26703 Continue execution of your program, like the @value{GDBN} @code{continue}
26707 Go up the number of frames indicated by the numeric argument
26708 (@pxref{Arguments, , Numeric Arguments, Emacs, The @sc{gnu} Emacs Manual}),
26709 like the @value{GDBN} @code{up} command.
26712 Go down the number of frames indicated by the numeric argument, like the
26713 @value{GDBN} @code{down} command.
26716 In any source file, the Emacs command @kbd{C-x @key{SPC}} (@code{gud-break})
26717 tells @value{GDBN} to set a breakpoint on the source line point is on.
26719 In text command mode, if you type @kbd{M-x speedbar}, Emacs displays a
26720 separate frame which shows a backtrace when the GUD buffer is current.
26721 Move point to any frame in the stack and type @key{RET} to make it
26722 become the current frame and display the associated source in the
26723 source buffer. Alternatively, click @kbd{Mouse-2} to make the
26724 selected frame become the current one. In graphical mode, the
26725 speedbar displays watch expressions.
26727 If you accidentally delete the source-display buffer, an easy way to get
26728 it back is to type the command @code{f} in the @value{GDBN} buffer, to
26729 request a frame display; when you run under Emacs, this recreates
26730 the source buffer if necessary to show you the context of the current
26733 The source files displayed in Emacs are in ordinary Emacs buffers
26734 which are visiting the source files in the usual way. You can edit
26735 the files with these buffers if you wish; but keep in mind that @value{GDBN}
26736 communicates with Emacs in terms of line numbers. If you add or
26737 delete lines from the text, the line numbers that @value{GDBN} knows cease
26738 to correspond properly with the code.
26740 A more detailed description of Emacs' interaction with @value{GDBN} is
26741 given in the Emacs manual (@pxref{Debuggers,,, Emacs, The @sc{gnu}
26744 @c The following dropped because Epoch is nonstandard. Reactivate
26745 @c if/when v19 does something similar. ---doc@cygnus.com 19dec1990
26747 @kindex Emacs Epoch environment
26751 Version 18 of @sc{gnu} Emacs has a built-in window system
26752 called the @code{epoch}
26753 environment. Users of this environment can use a new command,
26754 @code{inspect} which performs identically to @code{print} except that
26755 each value is printed in its own window.
26760 @chapter The @sc{gdb/mi} Interface
26762 @unnumberedsec Function and Purpose
26764 @cindex @sc{gdb/mi}, its purpose
26765 @sc{gdb/mi} is a line based machine oriented text interface to
26766 @value{GDBN} and is activated by specifying using the
26767 @option{--interpreter} command line option (@pxref{Mode Options}). It
26768 is specifically intended to support the development of systems which
26769 use the debugger as just one small component of a larger system.
26771 This chapter is a specification of the @sc{gdb/mi} interface. It is written
26772 in the form of a reference manual.
26774 Note that @sc{gdb/mi} is still under construction, so some of the
26775 features described below are incomplete and subject to change
26776 (@pxref{GDB/MI Development and Front Ends, , @sc{gdb/mi} Development and Front Ends}).
26778 @unnumberedsec Notation and Terminology
26780 @cindex notational conventions, for @sc{gdb/mi}
26781 This chapter uses the following notation:
26785 @code{|} separates two alternatives.
26788 @code{[ @var{something} ]} indicates that @var{something} is optional:
26789 it may or may not be given.
26792 @code{( @var{group} )*} means that @var{group} inside the parentheses
26793 may repeat zero or more times.
26796 @code{( @var{group} )+} means that @var{group} inside the parentheses
26797 may repeat one or more times.
26800 @code{"@var{string}"} means a literal @var{string}.
26804 @heading Dependencies
26808 * GDB/MI General Design::
26809 * GDB/MI Command Syntax::
26810 * GDB/MI Compatibility with CLI::
26811 * GDB/MI Development and Front Ends::
26812 * GDB/MI Output Records::
26813 * GDB/MI Simple Examples::
26814 * GDB/MI Command Description Format::
26815 * GDB/MI Breakpoint Commands::
26816 * GDB/MI Program Context::
26817 * GDB/MI Thread Commands::
26818 * GDB/MI Ada Tasking Commands::
26819 * GDB/MI Program Execution::
26820 * GDB/MI Stack Manipulation::
26821 * GDB/MI Variable Objects::
26822 * GDB/MI Data Manipulation::
26823 * GDB/MI Tracepoint Commands::
26824 * GDB/MI Symbol Query::
26825 * GDB/MI File Commands::
26827 * GDB/MI Kod Commands::
26828 * GDB/MI Memory Overlay Commands::
26829 * GDB/MI Signal Handling Commands::
26831 * GDB/MI Target Manipulation::
26832 * GDB/MI File Transfer Commands::
26833 * GDB/MI Miscellaneous Commands::
26836 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
26837 @node GDB/MI General Design
26838 @section @sc{gdb/mi} General Design
26839 @cindex GDB/MI General Design
26841 Interaction of a @sc{GDB/MI} frontend with @value{GDBN} involves three
26842 parts---commands sent to @value{GDBN}, responses to those commands
26843 and notifications. Each command results in exactly one response,
26844 indicating either successful completion of the command, or an error.
26845 For the commands that do not resume the target, the response contains the
26846 requested information. For the commands that resume the target, the
26847 response only indicates whether the target was successfully resumed.
26848 Notifications is the mechanism for reporting changes in the state of the
26849 target, or in @value{GDBN} state, that cannot conveniently be associated with
26850 a command and reported as part of that command response.
26852 The important examples of notifications are:
26856 Exec notifications. These are used to report changes in
26857 target state---when a target is resumed, or stopped. It would not
26858 be feasible to include this information in response of resuming
26859 commands, because one resume commands can result in multiple events in
26860 different threads. Also, quite some time may pass before any event
26861 happens in the target, while a frontend needs to know whether the resuming
26862 command itself was successfully executed.
26865 Console output, and status notifications. Console output
26866 notifications are used to report output of CLI commands, as well as
26867 diagnostics for other commands. Status notifications are used to
26868 report the progress of a long-running operation. Naturally, including
26869 this information in command response would mean no output is produced
26870 until the command is finished, which is undesirable.
26873 General notifications. Commands may have various side effects on
26874 the @value{GDBN} or target state beyond their official purpose. For example,
26875 a command may change the selected thread. Although such changes can
26876 be included in command response, using notification allows for more
26877 orthogonal frontend design.
26881 There's no guarantee that whenever an MI command reports an error,
26882 @value{GDBN} or the target are in any specific state, and especially,
26883 the state is not reverted to the state before the MI command was
26884 processed. Therefore, whenever an MI command results in an error,
26885 we recommend that the frontend refreshes all the information shown in
26886 the user interface.
26890 * Context management::
26891 * Asynchronous and non-stop modes::
26895 @node Context management
26896 @subsection Context management
26898 In most cases when @value{GDBN} accesses the target, this access is
26899 done in context of a specific thread and frame (@pxref{Frames}).
26900 Often, even when accessing global data, the target requires that a thread
26901 be specified. The CLI interface maintains the selected thread and frame,
26902 and supplies them to target on each command. This is convenient,
26903 because a command line user would not want to specify that information
26904 explicitly on each command, and because user interacts with
26905 @value{GDBN} via a single terminal, so no confusion is possible as
26906 to what thread and frame are the current ones.
26908 In the case of MI, the concept of selected thread and frame is less
26909 useful. First, a frontend can easily remember this information
26910 itself. Second, a graphical frontend can have more than one window,
26911 each one used for debugging a different thread, and the frontend might
26912 want to access additional threads for internal purposes. This
26913 increases the risk that by relying on implicitly selected thread, the
26914 frontend may be operating on a wrong one. Therefore, each MI command
26915 should explicitly specify which thread and frame to operate on. To
26916 make it possible, each MI command accepts the @samp{--thread} and
26917 @samp{--frame} options, the value to each is @value{GDBN} identifier
26918 for thread and frame to operate on.
26920 Usually, each top-level window in a frontend allows the user to select
26921 a thread and a frame, and remembers the user selection for further
26922 operations. However, in some cases @value{GDBN} may suggest that the
26923 current thread be changed. For example, when stopping on a breakpoint
26924 it is reasonable to switch to the thread where breakpoint is hit. For
26925 another example, if the user issues the CLI @samp{thread} command via
26926 the frontend, it is desirable to change the frontend's selected thread to the
26927 one specified by user. @value{GDBN} communicates the suggestion to
26928 change current thread using the @samp{=thread-selected} notification.
26929 No such notification is available for the selected frame at the moment.
26931 Note that historically, MI shares the selected thread with CLI, so
26932 frontends used the @code{-thread-select} to execute commands in the
26933 right context. However, getting this to work right is cumbersome. The
26934 simplest way is for frontend to emit @code{-thread-select} command
26935 before every command. This doubles the number of commands that need
26936 to be sent. The alternative approach is to suppress @code{-thread-select}
26937 if the selected thread in @value{GDBN} is supposed to be identical to the
26938 thread the frontend wants to operate on. However, getting this
26939 optimization right can be tricky. In particular, if the frontend
26940 sends several commands to @value{GDBN}, and one of the commands changes the
26941 selected thread, then the behaviour of subsequent commands will
26942 change. So, a frontend should either wait for response from such
26943 problematic commands, or explicitly add @code{-thread-select} for
26944 all subsequent commands. No frontend is known to do this exactly
26945 right, so it is suggested to just always pass the @samp{--thread} and
26946 @samp{--frame} options.
26948 @node Asynchronous and non-stop modes
26949 @subsection Asynchronous command execution and non-stop mode
26951 On some targets, @value{GDBN} is capable of processing MI commands
26952 even while the target is running. This is called @dfn{asynchronous
26953 command execution} (@pxref{Background Execution}). The frontend may
26954 specify a preferrence for asynchronous execution using the
26955 @code{-gdb-set target-async 1} command, which should be emitted before
26956 either running the executable or attaching to the target. After the
26957 frontend has started the executable or attached to the target, it can
26958 find if asynchronous execution is enabled using the
26959 @code{-list-target-features} command.
26961 Even if @value{GDBN} can accept a command while target is running,
26962 many commands that access the target do not work when the target is
26963 running. Therefore, asynchronous command execution is most useful
26964 when combined with non-stop mode (@pxref{Non-Stop Mode}). Then,
26965 it is possible to examine the state of one thread, while other threads
26968 When a given thread is running, MI commands that try to access the
26969 target in the context of that thread may not work, or may work only on
26970 some targets. In particular, commands that try to operate on thread's
26971 stack will not work, on any target. Commands that read memory, or
26972 modify breakpoints, may work or not work, depending on the target. Note
26973 that even commands that operate on global state, such as @code{print},
26974 @code{set}, and breakpoint commands, still access the target in the
26975 context of a specific thread, so frontend should try to find a
26976 stopped thread and perform the operation on that thread (using the
26977 @samp{--thread} option).
26979 Which commands will work in the context of a running thread is
26980 highly target dependent. However, the two commands
26981 @code{-exec-interrupt}, to stop a thread, and @code{-thread-info},
26982 to find the state of a thread, will always work.
26984 @node Thread groups
26985 @subsection Thread groups
26986 @value{GDBN} may be used to debug several processes at the same time.
26987 On some platfroms, @value{GDBN} may support debugging of several
26988 hardware systems, each one having several cores with several different
26989 processes running on each core. This section describes the MI
26990 mechanism to support such debugging scenarios.
26992 The key observation is that regardless of the structure of the
26993 target, MI can have a global list of threads, because most commands that
26994 accept the @samp{--thread} option do not need to know what process that
26995 thread belongs to. Therefore, it is not necessary to introduce
26996 neither additional @samp{--process} option, nor an notion of the
26997 current process in the MI interface. The only strictly new feature
26998 that is required is the ability to find how the threads are grouped
27001 To allow the user to discover such grouping, and to support arbitrary
27002 hierarchy of machines/cores/processes, MI introduces the concept of a
27003 @dfn{thread group}. Thread group is a collection of threads and other
27004 thread groups. A thread group always has a string identifier, a type,
27005 and may have additional attributes specific to the type. A new
27006 command, @code{-list-thread-groups}, returns the list of top-level
27007 thread groups, which correspond to processes that @value{GDBN} is
27008 debugging at the moment. By passing an identifier of a thread group
27009 to the @code{-list-thread-groups} command, it is possible to obtain
27010 the members of specific thread group.
27012 To allow the user to easily discover processes, and other objects, he
27013 wishes to debug, a concept of @dfn{available thread group} is
27014 introduced. Available thread group is an thread group that
27015 @value{GDBN} is not debugging, but that can be attached to, using the
27016 @code{-target-attach} command. The list of available top-level thread
27017 groups can be obtained using @samp{-list-thread-groups --available}.
27018 In general, the content of a thread group may be only retrieved only
27019 after attaching to that thread group.
27021 Thread groups are related to inferiors (@pxref{Inferiors and
27022 Programs}). Each inferior corresponds to a thread group of a special
27023 type @samp{process}, and some additional operations are permitted on
27024 such thread groups.
27026 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
27027 @node GDB/MI Command Syntax
27028 @section @sc{gdb/mi} Command Syntax
27031 * GDB/MI Input Syntax::
27032 * GDB/MI Output Syntax::
27035 @node GDB/MI Input Syntax
27036 @subsection @sc{gdb/mi} Input Syntax
27038 @cindex input syntax for @sc{gdb/mi}
27039 @cindex @sc{gdb/mi}, input syntax
27041 @item @var{command} @expansion{}
27042 @code{@var{cli-command} | @var{mi-command}}
27044 @item @var{cli-command} @expansion{}
27045 @code{[ @var{token} ] @var{cli-command} @var{nl}}, where
27046 @var{cli-command} is any existing @value{GDBN} CLI command.
27048 @item @var{mi-command} @expansion{}
27049 @code{[ @var{token} ] "-" @var{operation} ( " " @var{option} )*
27050 @code{[} " --" @code{]} ( " " @var{parameter} )* @var{nl}}
27052 @item @var{token} @expansion{}
27053 "any sequence of digits"
27055 @item @var{option} @expansion{}
27056 @code{"-" @var{parameter} [ " " @var{parameter} ]}
27058 @item @var{parameter} @expansion{}
27059 @code{@var{non-blank-sequence} | @var{c-string}}
27061 @item @var{operation} @expansion{}
27062 @emph{any of the operations described in this chapter}
27064 @item @var{non-blank-sequence} @expansion{}
27065 @emph{anything, provided it doesn't contain special characters such as
27066 "-", @var{nl}, """ and of course " "}
27068 @item @var{c-string} @expansion{}
27069 @code{""" @var{seven-bit-iso-c-string-content} """}
27071 @item @var{nl} @expansion{}
27080 The CLI commands are still handled by the @sc{mi} interpreter; their
27081 output is described below.
27084 The @code{@var{token}}, when present, is passed back when the command
27088 Some @sc{mi} commands accept optional arguments as part of the parameter
27089 list. Each option is identified by a leading @samp{-} (dash) and may be
27090 followed by an optional argument parameter. Options occur first in the
27091 parameter list and can be delimited from normal parameters using
27092 @samp{--} (this is useful when some parameters begin with a dash).
27099 We want easy access to the existing CLI syntax (for debugging).
27102 We want it to be easy to spot a @sc{mi} operation.
27105 @node GDB/MI Output Syntax
27106 @subsection @sc{gdb/mi} Output Syntax
27108 @cindex output syntax of @sc{gdb/mi}
27109 @cindex @sc{gdb/mi}, output syntax
27110 The output from @sc{gdb/mi} consists of zero or more out-of-band records
27111 followed, optionally, by a single result record. This result record
27112 is for the most recent command. The sequence of output records is
27113 terminated by @samp{(gdb)}.
27115 If an input command was prefixed with a @code{@var{token}} then the
27116 corresponding output for that command will also be prefixed by that same
27120 @item @var{output} @expansion{}
27121 @code{( @var{out-of-band-record} )* [ @var{result-record} ] "(gdb)" @var{nl}}
27123 @item @var{result-record} @expansion{}
27124 @code{ [ @var{token} ] "^" @var{result-class} ( "," @var{result} )* @var{nl}}
27126 @item @var{out-of-band-record} @expansion{}
27127 @code{@var{async-record} | @var{stream-record}}
27129 @item @var{async-record} @expansion{}
27130 @code{@var{exec-async-output} | @var{status-async-output} | @var{notify-async-output}}
27132 @item @var{exec-async-output} @expansion{}
27133 @code{[ @var{token} ] "*" @var{async-output}}
27135 @item @var{status-async-output} @expansion{}
27136 @code{[ @var{token} ] "+" @var{async-output}}
27138 @item @var{notify-async-output} @expansion{}
27139 @code{[ @var{token} ] "=" @var{async-output}}
27141 @item @var{async-output} @expansion{}
27142 @code{@var{async-class} ( "," @var{result} )* @var{nl}}
27144 @item @var{result-class} @expansion{}
27145 @code{"done" | "running" | "connected" | "error" | "exit"}
27147 @item @var{async-class} @expansion{}
27148 @code{"stopped" | @var{others}} (where @var{others} will be added
27149 depending on the needs---this is still in development).
27151 @item @var{result} @expansion{}
27152 @code{ @var{variable} "=" @var{value}}
27154 @item @var{variable} @expansion{}
27155 @code{ @var{string} }
27157 @item @var{value} @expansion{}
27158 @code{ @var{const} | @var{tuple} | @var{list} }
27160 @item @var{const} @expansion{}
27161 @code{@var{c-string}}
27163 @item @var{tuple} @expansion{}
27164 @code{ "@{@}" | "@{" @var{result} ( "," @var{result} )* "@}" }
27166 @item @var{list} @expansion{}
27167 @code{ "[]" | "[" @var{value} ( "," @var{value} )* "]" | "["
27168 @var{result} ( "," @var{result} )* "]" }
27170 @item @var{stream-record} @expansion{}
27171 @code{@var{console-stream-output} | @var{target-stream-output} | @var{log-stream-output}}
27173 @item @var{console-stream-output} @expansion{}
27174 @code{"~" @var{c-string}}
27176 @item @var{target-stream-output} @expansion{}
27177 @code{"@@" @var{c-string}}
27179 @item @var{log-stream-output} @expansion{}
27180 @code{"&" @var{c-string}}
27182 @item @var{nl} @expansion{}
27185 @item @var{token} @expansion{}
27186 @emph{any sequence of digits}.
27194 All output sequences end in a single line containing a period.
27197 The @code{@var{token}} is from the corresponding request. Note that
27198 for all async output, while the token is allowed by the grammar and
27199 may be output by future versions of @value{GDBN} for select async
27200 output messages, it is generally omitted. Frontends should treat
27201 all async output as reporting general changes in the state of the
27202 target and there should be no need to associate async output to any
27206 @cindex status output in @sc{gdb/mi}
27207 @var{status-async-output} contains on-going status information about the
27208 progress of a slow operation. It can be discarded. All status output is
27209 prefixed by @samp{+}.
27212 @cindex async output in @sc{gdb/mi}
27213 @var{exec-async-output} contains asynchronous state change on the target
27214 (stopped, started, disappeared). All async output is prefixed by
27218 @cindex notify output in @sc{gdb/mi}
27219 @var{notify-async-output} contains supplementary information that the
27220 client should handle (e.g., a new breakpoint information). All notify
27221 output is prefixed by @samp{=}.
27224 @cindex console output in @sc{gdb/mi}
27225 @var{console-stream-output} is output that should be displayed as is in the
27226 console. It is the textual response to a CLI command. All the console
27227 output is prefixed by @samp{~}.
27230 @cindex target output in @sc{gdb/mi}
27231 @var{target-stream-output} is the output produced by the target program.
27232 All the target output is prefixed by @samp{@@}.
27235 @cindex log output in @sc{gdb/mi}
27236 @var{log-stream-output} is output text coming from @value{GDBN}'s internals, for
27237 instance messages that should be displayed as part of an error log. All
27238 the log output is prefixed by @samp{&}.
27241 @cindex list output in @sc{gdb/mi}
27242 New @sc{gdb/mi} commands should only output @var{lists} containing
27248 @xref{GDB/MI Stream Records, , @sc{gdb/mi} Stream Records}, for more
27249 details about the various output records.
27251 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
27252 @node GDB/MI Compatibility with CLI
27253 @section @sc{gdb/mi} Compatibility with CLI
27255 @cindex compatibility, @sc{gdb/mi} and CLI
27256 @cindex @sc{gdb/mi}, compatibility with CLI
27258 For the developers convenience CLI commands can be entered directly,
27259 but there may be some unexpected behaviour. For example, commands
27260 that query the user will behave as if the user replied yes, breakpoint
27261 command lists are not executed and some CLI commands, such as
27262 @code{if}, @code{when} and @code{define}, prompt for further input with
27263 @samp{>}, which is not valid MI output.
27265 This feature may be removed at some stage in the future and it is
27266 recommended that front ends use the @code{-interpreter-exec} command
27267 (@pxref{-interpreter-exec}).
27269 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
27270 @node GDB/MI Development and Front Ends
27271 @section @sc{gdb/mi} Development and Front Ends
27272 @cindex @sc{gdb/mi} development
27274 The application which takes the MI output and presents the state of the
27275 program being debugged to the user is called a @dfn{front end}.
27277 Although @sc{gdb/mi} is still incomplete, it is currently being used
27278 by a variety of front ends to @value{GDBN}. This makes it difficult
27279 to introduce new functionality without breaking existing usage. This
27280 section tries to minimize the problems by describing how the protocol
27283 Some changes in MI need not break a carefully designed front end, and
27284 for these the MI version will remain unchanged. The following is a
27285 list of changes that may occur within one level, so front ends should
27286 parse MI output in a way that can handle them:
27290 New MI commands may be added.
27293 New fields may be added to the output of any MI command.
27296 The range of values for fields with specified values, e.g.,
27297 @code{in_scope} (@pxref{-var-update}) may be extended.
27299 @c The format of field's content e.g type prefix, may change so parse it
27300 @c at your own risk. Yes, in general?
27302 @c The order of fields may change? Shouldn't really matter but it might
27303 @c resolve inconsistencies.
27306 If the changes are likely to break front ends, the MI version level
27307 will be increased by one. This will allow the front end to parse the
27308 output according to the MI version. Apart from mi0, new versions of
27309 @value{GDBN} will not support old versions of MI and it will be the
27310 responsibility of the front end to work with the new one.
27312 @c Starting with mi3, add a new command -mi-version that prints the MI
27315 The best way to avoid unexpected changes in MI that might break your front
27316 end is to make your project known to @value{GDBN} developers and
27317 follow development on @email{gdb@@sourceware.org} and
27318 @email{gdb-patches@@sourceware.org}.
27319 @cindex mailing lists
27321 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
27322 @node GDB/MI Output Records
27323 @section @sc{gdb/mi} Output Records
27326 * GDB/MI Result Records::
27327 * GDB/MI Stream Records::
27328 * GDB/MI Async Records::
27329 * GDB/MI Frame Information::
27330 * GDB/MI Thread Information::
27331 * GDB/MI Ada Exception Information::
27334 @node GDB/MI Result Records
27335 @subsection @sc{gdb/mi} Result Records
27337 @cindex result records in @sc{gdb/mi}
27338 @cindex @sc{gdb/mi}, result records
27339 In addition to a number of out-of-band notifications, the response to a
27340 @sc{gdb/mi} command includes one of the following result indications:
27344 @item "^done" [ "," @var{results} ]
27345 The synchronous operation was successful, @code{@var{results}} are the return
27350 This result record is equivalent to @samp{^done}. Historically, it
27351 was output instead of @samp{^done} if the command has resumed the
27352 target. This behaviour is maintained for backward compatibility, but
27353 all frontends should treat @samp{^done} and @samp{^running}
27354 identically and rely on the @samp{*running} output record to determine
27355 which threads are resumed.
27359 @value{GDBN} has connected to a remote target.
27361 @item "^error" "," @var{c-string}
27363 The operation failed. The @code{@var{c-string}} contains the corresponding
27368 @value{GDBN} has terminated.
27372 @node GDB/MI Stream Records
27373 @subsection @sc{gdb/mi} Stream Records
27375 @cindex @sc{gdb/mi}, stream records
27376 @cindex stream records in @sc{gdb/mi}
27377 @value{GDBN} internally maintains a number of output streams: the console, the
27378 target, and the log. The output intended for each of these streams is
27379 funneled through the @sc{gdb/mi} interface using @dfn{stream records}.
27381 Each stream record begins with a unique @dfn{prefix character} which
27382 identifies its stream (@pxref{GDB/MI Output Syntax, , @sc{gdb/mi} Output
27383 Syntax}). In addition to the prefix, each stream record contains a
27384 @code{@var{string-output}}. This is either raw text (with an implicit new
27385 line) or a quoted C string (which does not contain an implicit newline).
27388 @item "~" @var{string-output}
27389 The console output stream contains text that should be displayed in the
27390 CLI console window. It contains the textual responses to CLI commands.
27392 @item "@@" @var{string-output}
27393 The target output stream contains any textual output from the running
27394 target. This is only present when GDB's event loop is truly
27395 asynchronous, which is currently only the case for remote targets.
27397 @item "&" @var{string-output}
27398 The log stream contains debugging messages being produced by @value{GDBN}'s
27402 @node GDB/MI Async Records
27403 @subsection @sc{gdb/mi} Async Records
27405 @cindex async records in @sc{gdb/mi}
27406 @cindex @sc{gdb/mi}, async records
27407 @dfn{Async} records are used to notify the @sc{gdb/mi} client of
27408 additional changes that have occurred. Those changes can either be a
27409 consequence of @sc{gdb/mi} commands (e.g., a breakpoint modified) or a result of
27410 target activity (e.g., target stopped).
27412 The following is the list of possible async records:
27416 @item *running,thread-id="@var{thread}"
27417 The target is now running. The @var{thread} field tells which
27418 specific thread is now running, and can be @samp{all} if all threads
27419 are running. The frontend should assume that no interaction with a
27420 running thread is possible after this notification is produced.
27421 The frontend should not assume that this notification is output
27422 only once for any command. @value{GDBN} may emit this notification
27423 several times, either for different threads, because it cannot resume
27424 all threads together, or even for a single thread, if the thread must
27425 be stepped though some code before letting it run freely.
27427 @item *stopped,reason="@var{reason}",thread-id="@var{id}",stopped-threads="@var{stopped}",core="@var{core}"
27428 The target has stopped. The @var{reason} field can have one of the
27432 @item breakpoint-hit
27433 A breakpoint was reached.
27434 @item watchpoint-trigger
27435 A watchpoint was triggered.
27436 @item read-watchpoint-trigger
27437 A read watchpoint was triggered.
27438 @item access-watchpoint-trigger
27439 An access watchpoint was triggered.
27440 @item function-finished
27441 An -exec-finish or similar CLI command was accomplished.
27442 @item location-reached
27443 An -exec-until or similar CLI command was accomplished.
27444 @item watchpoint-scope
27445 A watchpoint has gone out of scope.
27446 @item end-stepping-range
27447 An -exec-next, -exec-next-instruction, -exec-step, -exec-step-instruction or
27448 similar CLI command was accomplished.
27449 @item exited-signalled
27450 The inferior exited because of a signal.
27452 The inferior exited.
27453 @item exited-normally
27454 The inferior exited normally.
27455 @item signal-received
27456 A signal was received by the inferior.
27458 The inferior has stopped due to a library being loaded or unloaded.
27459 This can happen when @code{stop-on-solib-events} (@pxref{Files}) is
27460 set or when a @code{catch load} or @code{catch unload} catchpoint is
27461 in use (@pxref{Set Catchpoints}).
27463 The inferior has forked. This is reported when @code{catch fork}
27464 (@pxref{Set Catchpoints}) has been used.
27466 The inferior has vforked. This is reported in when @code{catch vfork}
27467 (@pxref{Set Catchpoints}) has been used.
27468 @item syscall-entry
27469 The inferior entered a system call. This is reported when @code{catch
27470 syscall} (@pxref{Set Catchpoints}) has been used.
27471 @item syscall-entry
27472 The inferior returned from a system call. This is reported when
27473 @code{catch syscall} (@pxref{Set Catchpoints}) has been used.
27475 The inferior called @code{exec}. This is reported when @code{catch exec}
27476 (@pxref{Set Catchpoints}) has been used.
27479 The @var{id} field identifies the thread that directly caused the stop
27480 -- for example by hitting a breakpoint. Depending on whether all-stop
27481 mode is in effect (@pxref{All-Stop Mode}), @value{GDBN} may either
27482 stop all threads, or only the thread that directly triggered the stop.
27483 If all threads are stopped, the @var{stopped} field will have the
27484 value of @code{"all"}. Otherwise, the value of the @var{stopped}
27485 field will be a list of thread identifiers. Presently, this list will
27486 always include a single thread, but frontend should be prepared to see
27487 several threads in the list. The @var{core} field reports the
27488 processor core on which the stop event has happened. This field may be absent
27489 if such information is not available.
27491 @item =thread-group-added,id="@var{id}"
27492 @itemx =thread-group-removed,id="@var{id}"
27493 A thread group was either added or removed. The @var{id} field
27494 contains the @value{GDBN} identifier of the thread group. When a thread
27495 group is added, it generally might not be associated with a running
27496 process. When a thread group is removed, its id becomes invalid and
27497 cannot be used in any way.
27499 @item =thread-group-started,id="@var{id}",pid="@var{pid}"
27500 A thread group became associated with a running program,
27501 either because the program was just started or the thread group
27502 was attached to a program. The @var{id} field contains the
27503 @value{GDBN} identifier of the thread group. The @var{pid} field
27504 contains process identifier, specific to the operating system.
27506 @item =thread-group-exited,id="@var{id}"[,exit-code="@var{code}"]
27507 A thread group is no longer associated with a running program,
27508 either because the program has exited, or because it was detached
27509 from. The @var{id} field contains the @value{GDBN} identifier of the
27510 thread group. @var{code} is the exit code of the inferior; it exists
27511 only when the inferior exited with some code.
27513 @item =thread-created,id="@var{id}",group-id="@var{gid}"
27514 @itemx =thread-exited,id="@var{id}",group-id="@var{gid}"
27515 A thread either was created, or has exited. The @var{id} field
27516 contains the @value{GDBN} identifier of the thread. The @var{gid}
27517 field identifies the thread group this thread belongs to.
27519 @item =thread-selected,id="@var{id}"
27520 Informs that the selected thread was changed as result of the last
27521 command. This notification is not emitted as result of @code{-thread-select}
27522 command but is emitted whenever an MI command that is not documented
27523 to change the selected thread actually changes it. In particular,
27524 invoking, directly or indirectly (via user-defined command), the CLI
27525 @code{thread} command, will generate this notification.
27527 We suggest that in response to this notification, front ends
27528 highlight the selected thread and cause subsequent commands to apply to
27531 @item =library-loaded,...
27532 Reports that a new library file was loaded by the program. This
27533 notification has 4 fields---@var{id}, @var{target-name},
27534 @var{host-name}, and @var{symbols-loaded}. The @var{id} field is an
27535 opaque identifier of the library. For remote debugging case,
27536 @var{target-name} and @var{host-name} fields give the name of the
27537 library file on the target, and on the host respectively. For native
27538 debugging, both those fields have the same value. The
27539 @var{symbols-loaded} field is emitted only for backward compatibility
27540 and should not be relied on to convey any useful information. The
27541 @var{thread-group} field, if present, specifies the id of the thread
27542 group in whose context the library was loaded. If the field is
27543 absent, it means the library was loaded in the context of all present
27546 @item =library-unloaded,...
27547 Reports that a library was unloaded by the program. This notification
27548 has 3 fields---@var{id}, @var{target-name} and @var{host-name} with
27549 the same meaning as for the @code{=library-loaded} notification.
27550 The @var{thread-group} field, if present, specifies the id of the
27551 thread group in whose context the library was unloaded. If the field is
27552 absent, it means the library was unloaded in the context of all present
27555 @item =breakpoint-created,bkpt=@{...@}
27556 @itemx =breakpoint-modified,bkpt=@{...@}
27557 @itemx =breakpoint-deleted,bkpt=@{...@}
27558 Reports that a breakpoint was created, modified, or deleted,
27559 respectively. Only user-visible breakpoints are reported to the MI
27562 The @var{bkpt} argument is of the same form as returned by the various
27563 breakpoint commands; @xref{GDB/MI Breakpoint Commands}.
27565 Note that if a breakpoint is emitted in the result record of a
27566 command, then it will not also be emitted in an async record.
27570 @node GDB/MI Frame Information
27571 @subsection @sc{gdb/mi} Frame Information
27573 Response from many MI commands includes an information about stack
27574 frame. This information is a tuple that may have the following
27579 The level of the stack frame. The innermost frame has the level of
27580 zero. This field is always present.
27583 The name of the function corresponding to the frame. This field may
27584 be absent if @value{GDBN} is unable to determine the function name.
27587 The code address for the frame. This field is always present.
27590 The name of the source files that correspond to the frame's code
27591 address. This field may be absent.
27594 The source line corresponding to the frames' code address. This field
27598 The name of the binary file (either executable or shared library) the
27599 corresponds to the frame's code address. This field may be absent.
27603 @node GDB/MI Thread Information
27604 @subsection @sc{gdb/mi} Thread Information
27606 Whenever @value{GDBN} has to report an information about a thread, it
27607 uses a tuple with the following fields:
27611 The numeric id assigned to the thread by @value{GDBN}. This field is
27615 Target-specific string identifying the thread. This field is always present.
27618 Additional information about the thread provided by the target.
27619 It is supposed to be human-readable and not interpreted by the
27620 frontend. This field is optional.
27623 Either @samp{stopped} or @samp{running}, depending on whether the
27624 thread is presently running. This field is always present.
27627 The value of this field is an integer number of the processor core the
27628 thread was last seen on. This field is optional.
27631 @node GDB/MI Ada Exception Information
27632 @subsection @sc{gdb/mi} Ada Exception Information
27634 Whenever a @code{*stopped} record is emitted because the program
27635 stopped after hitting an exception catchpoint (@pxref{Set Catchpoints}),
27636 @value{GDBN} provides the name of the exception that was raised via
27637 the @code{exception-name} field.
27639 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
27640 @node GDB/MI Simple Examples
27641 @section Simple Examples of @sc{gdb/mi} Interaction
27642 @cindex @sc{gdb/mi}, simple examples
27644 This subsection presents several simple examples of interaction using
27645 the @sc{gdb/mi} interface. In these examples, @samp{->} means that the
27646 following line is passed to @sc{gdb/mi} as input, while @samp{<-} means
27647 the output received from @sc{gdb/mi}.
27649 Note the line breaks shown in the examples are here only for
27650 readability, they don't appear in the real output.
27652 @subheading Setting a Breakpoint
27654 Setting a breakpoint generates synchronous output which contains detailed
27655 information of the breakpoint.
27658 -> -break-insert main
27659 <- ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
27660 enabled="y",addr="0x08048564",func="main",file="myprog.c",
27661 fullname="/home/nickrob/myprog.c",line="68",times="0"@}
27665 @subheading Program Execution
27667 Program execution generates asynchronous records and MI gives the
27668 reason that execution stopped.
27674 <- *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
27675 frame=@{addr="0x08048564",func="main",
27676 args=[@{name="argc",value="1"@},@{name="argv",value="0xbfc4d4d4"@}],
27677 file="myprog.c",fullname="/home/nickrob/myprog.c",line="68"@}
27682 <- *stopped,reason="exited-normally"
27686 @subheading Quitting @value{GDBN}
27688 Quitting @value{GDBN} just prints the result class @samp{^exit}.
27696 Please note that @samp{^exit} is printed immediately, but it might
27697 take some time for @value{GDBN} to actually exit. During that time, @value{GDBN}
27698 performs necessary cleanups, including killing programs being debugged
27699 or disconnecting from debug hardware, so the frontend should wait till
27700 @value{GDBN} exits and should only forcibly kill @value{GDBN} if it
27701 fails to exit in reasonable time.
27703 @subheading A Bad Command
27705 Here's what happens if you pass a non-existent command:
27709 <- ^error,msg="Undefined MI command: rubbish"
27714 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
27715 @node GDB/MI Command Description Format
27716 @section @sc{gdb/mi} Command Description Format
27718 The remaining sections describe blocks of commands. Each block of
27719 commands is laid out in a fashion similar to this section.
27721 @subheading Motivation
27723 The motivation for this collection of commands.
27725 @subheading Introduction
27727 A brief introduction to this collection of commands as a whole.
27729 @subheading Commands
27731 For each command in the block, the following is described:
27733 @subsubheading Synopsis
27736 -command @var{args}@dots{}
27739 @subsubheading Result
27741 @subsubheading @value{GDBN} Command
27743 The corresponding @value{GDBN} CLI command(s), if any.
27745 @subsubheading Example
27747 Example(s) formatted for readability. Some of the described commands have
27748 not been implemented yet and these are labeled N.A.@: (not available).
27751 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
27752 @node GDB/MI Breakpoint Commands
27753 @section @sc{gdb/mi} Breakpoint Commands
27755 @cindex breakpoint commands for @sc{gdb/mi}
27756 @cindex @sc{gdb/mi}, breakpoint commands
27757 This section documents @sc{gdb/mi} commands for manipulating
27760 @subheading The @code{-break-after} Command
27761 @findex -break-after
27763 @subsubheading Synopsis
27766 -break-after @var{number} @var{count}
27769 The breakpoint number @var{number} is not in effect until it has been
27770 hit @var{count} times. To see how this is reflected in the output of
27771 the @samp{-break-list} command, see the description of the
27772 @samp{-break-list} command below.
27774 @subsubheading @value{GDBN} Command
27776 The corresponding @value{GDBN} command is @samp{ignore}.
27778 @subsubheading Example
27783 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
27784 enabled="y",addr="0x000100d0",func="main",file="hello.c",
27785 fullname="/home/foo/hello.c",line="5",times="0"@}
27792 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
27793 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
27794 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
27795 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
27796 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
27797 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
27798 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
27799 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
27800 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
27801 line="5",times="0",ignore="3"@}]@}
27806 @subheading The @code{-break-catch} Command
27807 @findex -break-catch
27810 @subheading The @code{-break-commands} Command
27811 @findex -break-commands
27813 @subsubheading Synopsis
27816 -break-commands @var{number} [ @var{command1} ... @var{commandN} ]
27819 Specifies the CLI commands that should be executed when breakpoint
27820 @var{number} is hit. The parameters @var{command1} to @var{commandN}
27821 are the commands. If no command is specified, any previously-set
27822 commands are cleared. @xref{Break Commands}. Typical use of this
27823 functionality is tracing a program, that is, printing of values of
27824 some variables whenever breakpoint is hit and then continuing.
27826 @subsubheading @value{GDBN} Command
27828 The corresponding @value{GDBN} command is @samp{commands}.
27830 @subsubheading Example
27835 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
27836 enabled="y",addr="0x000100d0",func="main",file="hello.c",
27837 fullname="/home/foo/hello.c",line="5",times="0"@}
27839 -break-commands 1 "print v" "continue"
27844 @subheading The @code{-break-condition} Command
27845 @findex -break-condition
27847 @subsubheading Synopsis
27850 -break-condition @var{number} @var{expr}
27853 Breakpoint @var{number} will stop the program only if the condition in
27854 @var{expr} is true. The condition becomes part of the
27855 @samp{-break-list} output (see the description of the @samp{-break-list}
27858 @subsubheading @value{GDBN} Command
27860 The corresponding @value{GDBN} command is @samp{condition}.
27862 @subsubheading Example
27866 -break-condition 1 1
27870 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
27871 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
27872 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
27873 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
27874 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
27875 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
27876 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
27877 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
27878 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
27879 line="5",cond="1",times="0",ignore="3"@}]@}
27883 @subheading The @code{-break-delete} Command
27884 @findex -break-delete
27886 @subsubheading Synopsis
27889 -break-delete ( @var{breakpoint} )+
27892 Delete the breakpoint(s) whose number(s) are specified in the argument
27893 list. This is obviously reflected in the breakpoint list.
27895 @subsubheading @value{GDBN} Command
27897 The corresponding @value{GDBN} command is @samp{delete}.
27899 @subsubheading Example
27907 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
27908 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
27909 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
27910 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
27911 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
27912 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
27913 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
27918 @subheading The @code{-break-disable} Command
27919 @findex -break-disable
27921 @subsubheading Synopsis
27924 -break-disable ( @var{breakpoint} )+
27927 Disable the named @var{breakpoint}(s). The field @samp{enabled} in the
27928 break list is now set to @samp{n} for the named @var{breakpoint}(s).
27930 @subsubheading @value{GDBN} Command
27932 The corresponding @value{GDBN} command is @samp{disable}.
27934 @subsubheading Example
27942 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
27943 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
27944 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
27945 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
27946 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
27947 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
27948 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
27949 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="n",
27950 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
27951 line="5",times="0"@}]@}
27955 @subheading The @code{-break-enable} Command
27956 @findex -break-enable
27958 @subsubheading Synopsis
27961 -break-enable ( @var{breakpoint} )+
27964 Enable (previously disabled) @var{breakpoint}(s).
27966 @subsubheading @value{GDBN} Command
27968 The corresponding @value{GDBN} command is @samp{enable}.
27970 @subsubheading Example
27978 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
27979 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
27980 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
27981 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
27982 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
27983 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
27984 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
27985 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
27986 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
27987 line="5",times="0"@}]@}
27991 @subheading The @code{-break-info} Command
27992 @findex -break-info
27994 @subsubheading Synopsis
27997 -break-info @var{breakpoint}
28001 Get information about a single breakpoint.
28003 @subsubheading @value{GDBN} Command
28005 The corresponding @value{GDBN} command is @samp{info break @var{breakpoint}}.
28007 @subsubheading Example
28010 @subheading The @code{-break-insert} Command
28011 @findex -break-insert
28013 @subsubheading Synopsis
28016 -break-insert [ -t ] [ -h ] [ -f ] [ -d ] [ -a ]
28017 [ -c @var{condition} ] [ -i @var{ignore-count} ]
28018 [ -p @var{thread} ] [ @var{location} ]
28022 If specified, @var{location}, can be one of:
28029 @item filename:linenum
28030 @item filename:function
28034 The possible optional parameters of this command are:
28038 Insert a temporary breakpoint.
28040 Insert a hardware breakpoint.
28041 @item -c @var{condition}
28042 Make the breakpoint conditional on @var{condition}.
28043 @item -i @var{ignore-count}
28044 Initialize the @var{ignore-count}.
28046 If @var{location} cannot be parsed (for example if it
28047 refers to unknown files or functions), create a pending
28048 breakpoint. Without this flag, @value{GDBN} will report
28049 an error, and won't create a breakpoint, if @var{location}
28052 Create a disabled breakpoint.
28054 Create a tracepoint. @xref{Tracepoints}. When this parameter
28055 is used together with @samp{-h}, a fast tracepoint is created.
28058 @subsubheading Result
28060 The result is in the form:
28063 ^done,bkpt=@{number="@var{number}",type="@var{type}",disp="del"|"keep",
28064 enabled="y"|"n",addr="@var{hex}",func="@var{funcname}",file="@var{filename}",
28065 fullname="@var{full_filename}",line="@var{lineno}",[thread="@var{threadno},]
28066 times="@var{times}"@}
28070 where @var{number} is the @value{GDBN} number for this breakpoint,
28071 @var{funcname} is the name of the function where the breakpoint was
28072 inserted, @var{filename} is the name of the source file which contains
28073 this function, @var{lineno} is the source line number within that file
28074 and @var{times} the number of times that the breakpoint has been hit
28075 (always 0 for -break-insert but may be greater for -break-info or -break-list
28076 which use the same output).
28078 Note: this format is open to change.
28079 @c An out-of-band breakpoint instead of part of the result?
28081 @subsubheading @value{GDBN} Command
28083 The corresponding @value{GDBN} commands are @samp{break}, @samp{tbreak},
28084 @samp{hbreak}, @samp{thbreak}, and @samp{rbreak}.
28086 @subsubheading Example
28091 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",
28092 fullname="/home/foo/recursive2.c,line="4",times="0"@}
28094 -break-insert -t foo
28095 ^done,bkpt=@{number="2",addr="0x00010774",file="recursive2.c",
28096 fullname="/home/foo/recursive2.c,line="11",times="0"@}
28099 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
28100 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
28101 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
28102 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
28103 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
28104 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
28105 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
28106 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
28107 addr="0x0001072c", func="main",file="recursive2.c",
28108 fullname="/home/foo/recursive2.c,"line="4",times="0"@},
28109 bkpt=@{number="2",type="breakpoint",disp="del",enabled="y",
28110 addr="0x00010774",func="foo",file="recursive2.c",
28111 fullname="/home/foo/recursive2.c",line="11",times="0"@}]@}
28113 -break-insert -r foo.*
28114 ~int foo(int, int);
28115 ^done,bkpt=@{number="3",addr="0x00010774",file="recursive2.c,
28116 "fullname="/home/foo/recursive2.c",line="11",times="0"@}
28120 @subheading The @code{-break-list} Command
28121 @findex -break-list
28123 @subsubheading Synopsis
28129 Displays the list of inserted breakpoints, showing the following fields:
28133 number of the breakpoint
28135 type of the breakpoint: @samp{breakpoint} or @samp{watchpoint}
28137 should the breakpoint be deleted or disabled when it is hit: @samp{keep}
28140 is the breakpoint enabled or no: @samp{y} or @samp{n}
28142 memory location at which the breakpoint is set
28144 logical location of the breakpoint, expressed by function name, file
28147 number of times the breakpoint has been hit
28150 If there are no breakpoints or watchpoints, the @code{BreakpointTable}
28151 @code{body} field is an empty list.
28153 @subsubheading @value{GDBN} Command
28155 The corresponding @value{GDBN} command is @samp{info break}.
28157 @subsubheading Example
28162 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
28163 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
28164 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
28165 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
28166 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
28167 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
28168 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
28169 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
28170 addr="0x000100d0",func="main",file="hello.c",line="5",times="0"@},
28171 bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
28172 addr="0x00010114",func="foo",file="hello.c",fullname="/home/foo/hello.c",
28173 line="13",times="0"@}]@}
28177 Here's an example of the result when there are no breakpoints:
28182 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
28183 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
28184 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
28185 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
28186 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
28187 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
28188 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
28193 @subheading The @code{-break-passcount} Command
28194 @findex -break-passcount
28196 @subsubheading Synopsis
28199 -break-passcount @var{tracepoint-number} @var{passcount}
28202 Set the passcount for tracepoint @var{tracepoint-number} to
28203 @var{passcount}. If the breakpoint referred to by @var{tracepoint-number}
28204 is not a tracepoint, error is emitted. This corresponds to CLI
28205 command @samp{passcount}.
28207 @subheading The @code{-break-watch} Command
28208 @findex -break-watch
28210 @subsubheading Synopsis
28213 -break-watch [ -a | -r ]
28216 Create a watchpoint. With the @samp{-a} option it will create an
28217 @dfn{access} watchpoint, i.e., a watchpoint that triggers either on a
28218 read from or on a write to the memory location. With the @samp{-r}
28219 option, the watchpoint created is a @dfn{read} watchpoint, i.e., it will
28220 trigger only when the memory location is accessed for reading. Without
28221 either of the options, the watchpoint created is a regular watchpoint,
28222 i.e., it will trigger when the memory location is accessed for writing.
28223 @xref{Set Watchpoints, , Setting Watchpoints}.
28225 Note that @samp{-break-list} will report a single list of watchpoints and
28226 breakpoints inserted.
28228 @subsubheading @value{GDBN} Command
28230 The corresponding @value{GDBN} commands are @samp{watch}, @samp{awatch}, and
28233 @subsubheading Example
28235 Setting a watchpoint on a variable in the @code{main} function:
28240 ^done,wpt=@{number="2",exp="x"@}
28245 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="x"@},
28246 value=@{old="-268439212",new="55"@},
28247 frame=@{func="main",args=[],file="recursive2.c",
28248 fullname="/home/foo/bar/recursive2.c",line="5"@}
28252 Setting a watchpoint on a variable local to a function. @value{GDBN} will stop
28253 the program execution twice: first for the variable changing value, then
28254 for the watchpoint going out of scope.
28259 ^done,wpt=@{number="5",exp="C"@}
28264 *stopped,reason="watchpoint-trigger",
28265 wpt=@{number="5",exp="C"@},value=@{old="-276895068",new="3"@},
28266 frame=@{func="callee4",args=[],
28267 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
28268 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13"@}
28273 *stopped,reason="watchpoint-scope",wpnum="5",
28274 frame=@{func="callee3",args=[@{name="strarg",
28275 value="0x11940 \"A string argument.\""@}],
28276 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
28277 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
28281 Listing breakpoints and watchpoints, at different points in the program
28282 execution. Note that once the watchpoint goes out of scope, it is
28288 ^done,wpt=@{number="2",exp="C"@}
28291 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
28292 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
28293 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
28294 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
28295 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
28296 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
28297 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
28298 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
28299 addr="0x00010734",func="callee4",
28300 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
28301 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c"line="8",times="1"@},
28302 bkpt=@{number="2",type="watchpoint",disp="keep",
28303 enabled="y",addr="",what="C",times="0"@}]@}
28308 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="C"@},
28309 value=@{old="-276895068",new="3"@},
28310 frame=@{func="callee4",args=[],
28311 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
28312 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13"@}
28315 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
28316 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
28317 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
28318 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
28319 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
28320 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
28321 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
28322 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
28323 addr="0x00010734",func="callee4",
28324 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
28325 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",times="1"@},
28326 bkpt=@{number="2",type="watchpoint",disp="keep",
28327 enabled="y",addr="",what="C",times="-5"@}]@}
28331 ^done,reason="watchpoint-scope",wpnum="2",
28332 frame=@{func="callee3",args=[@{name="strarg",
28333 value="0x11940 \"A string argument.\""@}],
28334 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
28335 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
28338 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
28339 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
28340 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
28341 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
28342 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
28343 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
28344 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
28345 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
28346 addr="0x00010734",func="callee4",
28347 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
28348 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",
28353 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28354 @node GDB/MI Program Context
28355 @section @sc{gdb/mi} Program Context
28357 @subheading The @code{-exec-arguments} Command
28358 @findex -exec-arguments
28361 @subsubheading Synopsis
28364 -exec-arguments @var{args}
28367 Set the inferior program arguments, to be used in the next
28370 @subsubheading @value{GDBN} Command
28372 The corresponding @value{GDBN} command is @samp{set args}.
28374 @subsubheading Example
28378 -exec-arguments -v word
28385 @subheading The @code{-exec-show-arguments} Command
28386 @findex -exec-show-arguments
28388 @subsubheading Synopsis
28391 -exec-show-arguments
28394 Print the arguments of the program.
28396 @subsubheading @value{GDBN} Command
28398 The corresponding @value{GDBN} command is @samp{show args}.
28400 @subsubheading Example
28405 @subheading The @code{-environment-cd} Command
28406 @findex -environment-cd
28408 @subsubheading Synopsis
28411 -environment-cd @var{pathdir}
28414 Set @value{GDBN}'s working directory.
28416 @subsubheading @value{GDBN} Command
28418 The corresponding @value{GDBN} command is @samp{cd}.
28420 @subsubheading Example
28424 -environment-cd /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
28430 @subheading The @code{-environment-directory} Command
28431 @findex -environment-directory
28433 @subsubheading Synopsis
28436 -environment-directory [ -r ] [ @var{pathdir} ]+
28439 Add directories @var{pathdir} to beginning of search path for source files.
28440 If the @samp{-r} option is used, the search path is reset to the default
28441 search path. If directories @var{pathdir} are supplied in addition to the
28442 @samp{-r} option, the search path is first reset and then addition
28444 Multiple directories may be specified, separated by blanks. Specifying
28445 multiple directories in a single command
28446 results in the directories added to the beginning of the
28447 search path in the same order they were presented in the command.
28448 If blanks are needed as
28449 part of a directory name, double-quotes should be used around
28450 the name. In the command output, the path will show up separated
28451 by the system directory-separator character. The directory-separator
28452 character must not be used
28453 in any directory name.
28454 If no directories are specified, the current search path is displayed.
28456 @subsubheading @value{GDBN} Command
28458 The corresponding @value{GDBN} command is @samp{dir}.
28460 @subsubheading Example
28464 -environment-directory /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
28465 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
28467 -environment-directory ""
28468 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
28470 -environment-directory -r /home/jjohnstn/src/gdb /usr/src
28471 ^done,source-path="/home/jjohnstn/src/gdb:/usr/src:$cdir:$cwd"
28473 -environment-directory -r
28474 ^done,source-path="$cdir:$cwd"
28479 @subheading The @code{-environment-path} Command
28480 @findex -environment-path
28482 @subsubheading Synopsis
28485 -environment-path [ -r ] [ @var{pathdir} ]+
28488 Add directories @var{pathdir} to beginning of search path for object files.
28489 If the @samp{-r} option is used, the search path is reset to the original
28490 search path that existed at gdb start-up. If directories @var{pathdir} are
28491 supplied in addition to the
28492 @samp{-r} option, the search path is first reset and then addition
28494 Multiple directories may be specified, separated by blanks. Specifying
28495 multiple directories in a single command
28496 results in the directories added to the beginning of the
28497 search path in the same order they were presented in the command.
28498 If blanks are needed as
28499 part of a directory name, double-quotes should be used around
28500 the name. In the command output, the path will show up separated
28501 by the system directory-separator character. The directory-separator
28502 character must not be used
28503 in any directory name.
28504 If no directories are specified, the current path is displayed.
28507 @subsubheading @value{GDBN} Command
28509 The corresponding @value{GDBN} command is @samp{path}.
28511 @subsubheading Example
28516 ^done,path="/usr/bin"
28518 -environment-path /kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb /bin
28519 ^done,path="/kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb:/bin:/usr/bin"
28521 -environment-path -r /usr/local/bin
28522 ^done,path="/usr/local/bin:/usr/bin"
28527 @subheading The @code{-environment-pwd} Command
28528 @findex -environment-pwd
28530 @subsubheading Synopsis
28536 Show the current working directory.
28538 @subsubheading @value{GDBN} Command
28540 The corresponding @value{GDBN} command is @samp{pwd}.
28542 @subsubheading Example
28547 ^done,cwd="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb"
28551 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28552 @node GDB/MI Thread Commands
28553 @section @sc{gdb/mi} Thread Commands
28556 @subheading The @code{-thread-info} Command
28557 @findex -thread-info
28559 @subsubheading Synopsis
28562 -thread-info [ @var{thread-id} ]
28565 Reports information about either a specific thread, if
28566 the @var{thread-id} parameter is present, or about all
28567 threads. When printing information about all threads,
28568 also reports the current thread.
28570 @subsubheading @value{GDBN} Command
28572 The @samp{info thread} command prints the same information
28575 @subsubheading Result
28577 The result is a list of threads. The following attributes are
28578 defined for a given thread:
28582 This field exists only for the current thread. It has the value @samp{*}.
28585 The identifier that @value{GDBN} uses to refer to the thread.
28588 The identifier that the target uses to refer to the thread.
28591 Extra information about the thread, in a target-specific format. This
28595 The name of the thread. If the user specified a name using the
28596 @code{thread name} command, then this name is given. Otherwise, if
28597 @value{GDBN} can extract the thread name from the target, then that
28598 name is given. If @value{GDBN} cannot find the thread name, then this
28602 The stack frame currently executing in the thread.
28605 The thread's state. The @samp{state} field may have the following
28610 The thread is stopped. Frame information is available for stopped
28614 The thread is running. There's no frame information for running
28620 If @value{GDBN} can find the CPU core on which this thread is running,
28621 then this field is the core identifier. This field is optional.
28625 @subsubheading Example
28630 @{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
28631 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",
28632 args=[]@},state="running"@},
28633 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
28634 frame=@{level="0",addr="0x0804891f",func="foo",
28635 args=[@{name="i",value="10"@}],
28636 file="/tmp/a.c",fullname="/tmp/a.c",line="158"@},
28637 state="running"@}],
28638 current-thread-id="1"
28642 @subheading The @code{-thread-list-ids} Command
28643 @findex -thread-list-ids
28645 @subsubheading Synopsis
28651 Produces a list of the currently known @value{GDBN} thread ids. At the
28652 end of the list it also prints the total number of such threads.
28654 This command is retained for historical reasons, the
28655 @code{-thread-info} command should be used instead.
28657 @subsubheading @value{GDBN} Command
28659 Part of @samp{info threads} supplies the same information.
28661 @subsubheading Example
28666 ^done,thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
28667 current-thread-id="1",number-of-threads="3"
28672 @subheading The @code{-thread-select} Command
28673 @findex -thread-select
28675 @subsubheading Synopsis
28678 -thread-select @var{threadnum}
28681 Make @var{threadnum} the current thread. It prints the number of the new
28682 current thread, and the topmost frame for that thread.
28684 This command is deprecated in favor of explicitly using the
28685 @samp{--thread} option to each command.
28687 @subsubheading @value{GDBN} Command
28689 The corresponding @value{GDBN} command is @samp{thread}.
28691 @subsubheading Example
28698 *stopped,reason="end-stepping-range",thread-id="2",line="187",
28699 file="../../../devo/gdb/testsuite/gdb.threads/linux-dp.c"
28703 thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
28704 number-of-threads="3"
28707 ^done,new-thread-id="3",
28708 frame=@{level="0",func="vprintf",
28709 args=[@{name="format",value="0x8048e9c \"%*s%c %d %c\\n\""@},
28710 @{name="arg",value="0x2"@}],file="vprintf.c",line="31"@}
28714 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28715 @node GDB/MI Ada Tasking Commands
28716 @section @sc{gdb/mi} Ada Tasking Commands
28718 @subheading The @code{-ada-task-info} Command
28719 @findex -ada-task-info
28721 @subsubheading Synopsis
28724 -ada-task-info [ @var{task-id} ]
28727 Reports information about either a specific Ada task, if the
28728 @var{task-id} parameter is present, or about all Ada tasks.
28730 @subsubheading @value{GDBN} Command
28732 The @samp{info tasks} command prints the same information
28733 about all Ada tasks (@pxref{Ada Tasks}).
28735 @subsubheading Result
28737 The result is a table of Ada tasks. The following columns are
28738 defined for each Ada task:
28742 This field exists only for the current thread. It has the value @samp{*}.
28745 The identifier that @value{GDBN} uses to refer to the Ada task.
28748 The identifier that the target uses to refer to the Ada task.
28751 The identifier of the thread corresponding to the Ada task.
28753 This field should always exist, as Ada tasks are always implemented
28754 on top of a thread. But if @value{GDBN} cannot find this corresponding
28755 thread for any reason, the field is omitted.
28758 This field exists only when the task was created by another task.
28759 In this case, it provides the ID of the parent task.
28762 The base priority of the task.
28765 The current state of the task. For a detailed description of the
28766 possible states, see @ref{Ada Tasks}.
28769 The name of the task.
28773 @subsubheading Example
28777 ^done,tasks=@{nr_rows="3",nr_cols="8",
28778 hdr=[@{width="1",alignment="-1",col_name="current",colhdr=""@},
28779 @{width="3",alignment="1",col_name="id",colhdr="ID"@},
28780 @{width="9",alignment="1",col_name="task-id",colhdr="TID"@},
28781 @{width="4",alignment="1",col_name="thread-id",colhdr=""@},
28782 @{width="4",alignment="1",col_name="parent-id",colhdr="P-ID"@},
28783 @{width="3",alignment="1",col_name="priority",colhdr="Pri"@},
28784 @{width="22",alignment="-1",col_name="state",colhdr="State"@},
28785 @{width="1",alignment="2",col_name="name",colhdr="Name"@}],
28786 body=[@{current="*",id="1",task-id=" 644010",thread-id="1",priority="48",
28787 state="Child Termination Wait",name="main_task"@}]@}
28791 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28792 @node GDB/MI Program Execution
28793 @section @sc{gdb/mi} Program Execution
28795 These are the asynchronous commands which generate the out-of-band
28796 record @samp{*stopped}. Currently @value{GDBN} only really executes
28797 asynchronously with remote targets and this interaction is mimicked in
28800 @subheading The @code{-exec-continue} Command
28801 @findex -exec-continue
28803 @subsubheading Synopsis
28806 -exec-continue [--reverse] [--all|--thread-group N]
28809 Resumes the execution of the inferior program, which will continue
28810 to execute until it reaches a debugger stop event. If the
28811 @samp{--reverse} option is specified, execution resumes in reverse until
28812 it reaches a stop event. Stop events may include
28815 breakpoints or watchpoints
28817 signals or exceptions
28819 the end of the process (or its beginning under @samp{--reverse})
28821 the end or beginning of a replay log if one is being used.
28823 In all-stop mode (@pxref{All-Stop
28824 Mode}), may resume only one thread, or all threads, depending on the
28825 value of the @samp{scheduler-locking} variable. If @samp{--all} is
28826 specified, all threads (in all inferiors) will be resumed. The @samp{--all} option is
28827 ignored in all-stop mode. If the @samp{--thread-group} options is
28828 specified, then all threads in that thread group are resumed.
28830 @subsubheading @value{GDBN} Command
28832 The corresponding @value{GDBN} corresponding is @samp{continue}.
28834 @subsubheading Example
28841 *stopped,reason="breakpoint-hit",disp="keep",bkptno="2",frame=@{
28842 func="foo",args=[],file="hello.c",fullname="/home/foo/bar/hello.c",
28848 @subheading The @code{-exec-finish} Command
28849 @findex -exec-finish
28851 @subsubheading Synopsis
28854 -exec-finish [--reverse]
28857 Resumes the execution of the inferior program until the current
28858 function is exited. Displays the results returned by the function.
28859 If the @samp{--reverse} option is specified, resumes the reverse
28860 execution of the inferior program until the point where current
28861 function was called.
28863 @subsubheading @value{GDBN} Command
28865 The corresponding @value{GDBN} command is @samp{finish}.
28867 @subsubheading Example
28869 Function returning @code{void}.
28876 *stopped,reason="function-finished",frame=@{func="main",args=[],
28877 file="hello.c",fullname="/home/foo/bar/hello.c",line="7"@}
28881 Function returning other than @code{void}. The name of the internal
28882 @value{GDBN} variable storing the result is printed, together with the
28889 *stopped,reason="function-finished",frame=@{addr="0x000107b0",func="foo",
28890 args=[@{name="a",value="1"],@{name="b",value="9"@}@},
28891 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
28892 gdb-result-var="$1",return-value="0"
28897 @subheading The @code{-exec-interrupt} Command
28898 @findex -exec-interrupt
28900 @subsubheading Synopsis
28903 -exec-interrupt [--all|--thread-group N]
28906 Interrupts the background execution of the target. Note how the token
28907 associated with the stop message is the one for the execution command
28908 that has been interrupted. The token for the interrupt itself only
28909 appears in the @samp{^done} output. If the user is trying to
28910 interrupt a non-running program, an error message will be printed.
28912 Note that when asynchronous execution is enabled, this command is
28913 asynchronous just like other execution commands. That is, first the
28914 @samp{^done} response will be printed, and the target stop will be
28915 reported after that using the @samp{*stopped} notification.
28917 In non-stop mode, only the context thread is interrupted by default.
28918 All threads (in all inferiors) will be interrupted if the
28919 @samp{--all} option is specified. If the @samp{--thread-group}
28920 option is specified, all threads in that group will be interrupted.
28922 @subsubheading @value{GDBN} Command
28924 The corresponding @value{GDBN} command is @samp{interrupt}.
28926 @subsubheading Example
28937 111*stopped,signal-name="SIGINT",signal-meaning="Interrupt",
28938 frame=@{addr="0x00010140",func="foo",args=[],file="try.c",
28939 fullname="/home/foo/bar/try.c",line="13"@}
28944 ^error,msg="mi_cmd_exec_interrupt: Inferior not executing."
28948 @subheading The @code{-exec-jump} Command
28951 @subsubheading Synopsis
28954 -exec-jump @var{location}
28957 Resumes execution of the inferior program at the location specified by
28958 parameter. @xref{Specify Location}, for a description of the
28959 different forms of @var{location}.
28961 @subsubheading @value{GDBN} Command
28963 The corresponding @value{GDBN} command is @samp{jump}.
28965 @subsubheading Example
28968 -exec-jump foo.c:10
28969 *running,thread-id="all"
28974 @subheading The @code{-exec-next} Command
28977 @subsubheading Synopsis
28980 -exec-next [--reverse]
28983 Resumes execution of the inferior program, stopping when the beginning
28984 of the next source line is reached.
28986 If the @samp{--reverse} option is specified, resumes reverse execution
28987 of the inferior program, stopping at the beginning of the previous
28988 source line. If you issue this command on the first line of a
28989 function, it will take you back to the caller of that function, to the
28990 source line where the function was called.
28993 @subsubheading @value{GDBN} Command
28995 The corresponding @value{GDBN} command is @samp{next}.
28997 @subsubheading Example
29003 *stopped,reason="end-stepping-range",line="8",file="hello.c"
29008 @subheading The @code{-exec-next-instruction} Command
29009 @findex -exec-next-instruction
29011 @subsubheading Synopsis
29014 -exec-next-instruction [--reverse]
29017 Executes one machine instruction. If the instruction is a function
29018 call, continues until the function returns. If the program stops at an
29019 instruction in the middle of a source line, the address will be
29022 If the @samp{--reverse} option is specified, resumes reverse execution
29023 of the inferior program, stopping at the previous instruction. If the
29024 previously executed instruction was a return from another function,
29025 it will continue to execute in reverse until the call to that function
29026 (from the current stack frame) is reached.
29028 @subsubheading @value{GDBN} Command
29030 The corresponding @value{GDBN} command is @samp{nexti}.
29032 @subsubheading Example
29036 -exec-next-instruction
29040 *stopped,reason="end-stepping-range",
29041 addr="0x000100d4",line="5",file="hello.c"
29046 @subheading The @code{-exec-return} Command
29047 @findex -exec-return
29049 @subsubheading Synopsis
29055 Makes current function return immediately. Doesn't execute the inferior.
29056 Displays the new current frame.
29058 @subsubheading @value{GDBN} Command
29060 The corresponding @value{GDBN} command is @samp{return}.
29062 @subsubheading Example
29066 200-break-insert callee4
29067 200^done,bkpt=@{number="1",addr="0x00010734",
29068 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",line="8"@}
29073 000*stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
29074 frame=@{func="callee4",args=[],
29075 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29076 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8"@}
29082 111^done,frame=@{level="0",func="callee3",
29083 args=[@{name="strarg",
29084 value="0x11940 \"A string argument.\""@}],
29085 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29086 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
29091 @subheading The @code{-exec-run} Command
29094 @subsubheading Synopsis
29097 -exec-run [--all | --thread-group N]
29100 Starts execution of the inferior from the beginning. The inferior
29101 executes until either a breakpoint is encountered or the program
29102 exits. In the latter case the output will include an exit code, if
29103 the program has exited exceptionally.
29105 When no option is specified, the current inferior is started. If the
29106 @samp{--thread-group} option is specified, it should refer to a thread
29107 group of type @samp{process}, and that thread group will be started.
29108 If the @samp{--all} option is specified, then all inferiors will be started.
29110 @subsubheading @value{GDBN} Command
29112 The corresponding @value{GDBN} command is @samp{run}.
29114 @subsubheading Examples
29119 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",line="4"@}
29124 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
29125 frame=@{func="main",args=[],file="recursive2.c",
29126 fullname="/home/foo/bar/recursive2.c",line="4"@}
29131 Program exited normally:
29139 *stopped,reason="exited-normally"
29144 Program exited exceptionally:
29152 *stopped,reason="exited",exit-code="01"
29156 Another way the program can terminate is if it receives a signal such as
29157 @code{SIGINT}. In this case, @sc{gdb/mi} displays this:
29161 *stopped,reason="exited-signalled",signal-name="SIGINT",
29162 signal-meaning="Interrupt"
29166 @c @subheading -exec-signal
29169 @subheading The @code{-exec-step} Command
29172 @subsubheading Synopsis
29175 -exec-step [--reverse]
29178 Resumes execution of the inferior program, stopping when the beginning
29179 of the next source line is reached, if the next source line is not a
29180 function call. If it is, stop at the first instruction of the called
29181 function. If the @samp{--reverse} option is specified, resumes reverse
29182 execution of the inferior program, stopping at the beginning of the
29183 previously executed source line.
29185 @subsubheading @value{GDBN} Command
29187 The corresponding @value{GDBN} command is @samp{step}.
29189 @subsubheading Example
29191 Stepping into a function:
29197 *stopped,reason="end-stepping-range",
29198 frame=@{func="foo",args=[@{name="a",value="10"@},
29199 @{name="b",value="0"@}],file="recursive2.c",
29200 fullname="/home/foo/bar/recursive2.c",line="11"@}
29210 *stopped,reason="end-stepping-range",line="14",file="recursive2.c"
29215 @subheading The @code{-exec-step-instruction} Command
29216 @findex -exec-step-instruction
29218 @subsubheading Synopsis
29221 -exec-step-instruction [--reverse]
29224 Resumes the inferior which executes one machine instruction. If the
29225 @samp{--reverse} option is specified, resumes reverse execution of the
29226 inferior program, stopping at the previously executed instruction.
29227 The output, once @value{GDBN} has stopped, will vary depending on
29228 whether we have stopped in the middle of a source line or not. In the
29229 former case, the address at which the program stopped will be printed
29232 @subsubheading @value{GDBN} Command
29234 The corresponding @value{GDBN} command is @samp{stepi}.
29236 @subsubheading Example
29240 -exec-step-instruction
29244 *stopped,reason="end-stepping-range",
29245 frame=@{func="foo",args=[],file="try.c",
29246 fullname="/home/foo/bar/try.c",line="10"@}
29248 -exec-step-instruction
29252 *stopped,reason="end-stepping-range",
29253 frame=@{addr="0x000100f4",func="foo",args=[],file="try.c",
29254 fullname="/home/foo/bar/try.c",line="10"@}
29259 @subheading The @code{-exec-until} Command
29260 @findex -exec-until
29262 @subsubheading Synopsis
29265 -exec-until [ @var{location} ]
29268 Executes the inferior until the @var{location} specified in the
29269 argument is reached. If there is no argument, the inferior executes
29270 until a source line greater than the current one is reached. The
29271 reason for stopping in this case will be @samp{location-reached}.
29273 @subsubheading @value{GDBN} Command
29275 The corresponding @value{GDBN} command is @samp{until}.
29277 @subsubheading Example
29281 -exec-until recursive2.c:6
29285 *stopped,reason="location-reached",frame=@{func="main",args=[],
29286 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="6"@}
29291 @subheading -file-clear
29292 Is this going away????
29295 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
29296 @node GDB/MI Stack Manipulation
29297 @section @sc{gdb/mi} Stack Manipulation Commands
29300 @subheading The @code{-stack-info-frame} Command
29301 @findex -stack-info-frame
29303 @subsubheading Synopsis
29309 Get info on the selected frame.
29311 @subsubheading @value{GDBN} Command
29313 The corresponding @value{GDBN} command is @samp{info frame} or @samp{frame}
29314 (without arguments).
29316 @subsubheading Example
29321 ^done,frame=@{level="1",addr="0x0001076c",func="callee3",
29322 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29323 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17"@}
29327 @subheading The @code{-stack-info-depth} Command
29328 @findex -stack-info-depth
29330 @subsubheading Synopsis
29333 -stack-info-depth [ @var{max-depth} ]
29336 Return the depth of the stack. If the integer argument @var{max-depth}
29337 is specified, do not count beyond @var{max-depth} frames.
29339 @subsubheading @value{GDBN} Command
29341 There's no equivalent @value{GDBN} command.
29343 @subsubheading Example
29345 For a stack with frame levels 0 through 11:
29352 -stack-info-depth 4
29355 -stack-info-depth 12
29358 -stack-info-depth 11
29361 -stack-info-depth 13
29366 @subheading The @code{-stack-list-arguments} Command
29367 @findex -stack-list-arguments
29369 @subsubheading Synopsis
29372 -stack-list-arguments @var{print-values}
29373 [ @var{low-frame} @var{high-frame} ]
29376 Display a list of the arguments for the frames between @var{low-frame}
29377 and @var{high-frame} (inclusive). If @var{low-frame} and
29378 @var{high-frame} are not provided, list the arguments for the whole
29379 call stack. If the two arguments are equal, show the single frame
29380 at the corresponding level. It is an error if @var{low-frame} is
29381 larger than the actual number of frames. On the other hand,
29382 @var{high-frame} may be larger than the actual number of frames, in
29383 which case only existing frames will be returned.
29385 If @var{print-values} is 0 or @code{--no-values}, print only the names of
29386 the variables; if it is 1 or @code{--all-values}, print also their
29387 values; and if it is 2 or @code{--simple-values}, print the name,
29388 type and value for simple data types, and the name and type for arrays,
29389 structures and unions.
29391 Use of this command to obtain arguments in a single frame is
29392 deprecated in favor of the @samp{-stack-list-variables} command.
29394 @subsubheading @value{GDBN} Command
29396 @value{GDBN} does not have an equivalent command. @code{gdbtk} has a
29397 @samp{gdb_get_args} command which partially overlaps with the
29398 functionality of @samp{-stack-list-arguments}.
29400 @subsubheading Example
29407 frame=@{level="0",addr="0x00010734",func="callee4",
29408 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29409 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8"@},
29410 frame=@{level="1",addr="0x0001076c",func="callee3",
29411 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29412 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17"@},
29413 frame=@{level="2",addr="0x0001078c",func="callee2",
29414 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29415 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="22"@},
29416 frame=@{level="3",addr="0x000107b4",func="callee1",
29417 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29418 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="27"@},
29419 frame=@{level="4",addr="0x000107e0",func="main",
29420 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29421 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="32"@}]
29423 -stack-list-arguments 0
29426 frame=@{level="0",args=[]@},
29427 frame=@{level="1",args=[name="strarg"]@},
29428 frame=@{level="2",args=[name="intarg",name="strarg"]@},
29429 frame=@{level="3",args=[name="intarg",name="strarg",name="fltarg"]@},
29430 frame=@{level="4",args=[]@}]
29432 -stack-list-arguments 1
29435 frame=@{level="0",args=[]@},
29437 args=[@{name="strarg",value="0x11940 \"A string argument.\""@}]@},
29438 frame=@{level="2",args=[
29439 @{name="intarg",value="2"@},
29440 @{name="strarg",value="0x11940 \"A string argument.\""@}]@},
29441 @{frame=@{level="3",args=[
29442 @{name="intarg",value="2"@},
29443 @{name="strarg",value="0x11940 \"A string argument.\""@},
29444 @{name="fltarg",value="3.5"@}]@},
29445 frame=@{level="4",args=[]@}]
29447 -stack-list-arguments 0 2 2
29448 ^done,stack-args=[frame=@{level="2",args=[name="intarg",name="strarg"]@}]
29450 -stack-list-arguments 1 2 2
29451 ^done,stack-args=[frame=@{level="2",
29452 args=[@{name="intarg",value="2"@},
29453 @{name="strarg",value="0x11940 \"A string argument.\""@}]@}]
29457 @c @subheading -stack-list-exception-handlers
29460 @subheading The @code{-stack-list-frames} Command
29461 @findex -stack-list-frames
29463 @subsubheading Synopsis
29466 -stack-list-frames [ @var{low-frame} @var{high-frame} ]
29469 List the frames currently on the stack. For each frame it displays the
29474 The frame number, 0 being the topmost frame, i.e., the innermost function.
29476 The @code{$pc} value for that frame.
29480 File name of the source file where the function lives.
29481 @item @var{fullname}
29482 The full file name of the source file where the function lives.
29484 Line number corresponding to the @code{$pc}.
29486 The shared library where this function is defined. This is only given
29487 if the frame's function is not known.
29490 If invoked without arguments, this command prints a backtrace for the
29491 whole stack. If given two integer arguments, it shows the frames whose
29492 levels are between the two arguments (inclusive). If the two arguments
29493 are equal, it shows the single frame at the corresponding level. It is
29494 an error if @var{low-frame} is larger than the actual number of
29495 frames. On the other hand, @var{high-frame} may be larger than the
29496 actual number of frames, in which case only existing frames will be returned.
29498 @subsubheading @value{GDBN} Command
29500 The corresponding @value{GDBN} commands are @samp{backtrace} and @samp{where}.
29502 @subsubheading Example
29504 Full stack backtrace:
29510 [frame=@{level="0",addr="0x0001076c",func="foo",
29511 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="11"@},
29512 frame=@{level="1",addr="0x000107a4",func="foo",
29513 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29514 frame=@{level="2",addr="0x000107a4",func="foo",
29515 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29516 frame=@{level="3",addr="0x000107a4",func="foo",
29517 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29518 frame=@{level="4",addr="0x000107a4",func="foo",
29519 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29520 frame=@{level="5",addr="0x000107a4",func="foo",
29521 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29522 frame=@{level="6",addr="0x000107a4",func="foo",
29523 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29524 frame=@{level="7",addr="0x000107a4",func="foo",
29525 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29526 frame=@{level="8",addr="0x000107a4",func="foo",
29527 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29528 frame=@{level="9",addr="0x000107a4",func="foo",
29529 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29530 frame=@{level="10",addr="0x000107a4",func="foo",
29531 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29532 frame=@{level="11",addr="0x00010738",func="main",
29533 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="4"@}]
29537 Show frames between @var{low_frame} and @var{high_frame}:
29541 -stack-list-frames 3 5
29543 [frame=@{level="3",addr="0x000107a4",func="foo",
29544 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29545 frame=@{level="4",addr="0x000107a4",func="foo",
29546 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29547 frame=@{level="5",addr="0x000107a4",func="foo",
29548 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@}]
29552 Show a single frame:
29556 -stack-list-frames 3 3
29558 [frame=@{level="3",addr="0x000107a4",func="foo",
29559 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@}]
29564 @subheading The @code{-stack-list-locals} Command
29565 @findex -stack-list-locals
29567 @subsubheading Synopsis
29570 -stack-list-locals @var{print-values}
29573 Display the local variable names for the selected frame. If
29574 @var{print-values} is 0 or @code{--no-values}, print only the names of
29575 the variables; if it is 1 or @code{--all-values}, print also their
29576 values; and if it is 2 or @code{--simple-values}, print the name,
29577 type and value for simple data types, and the name and type for arrays,
29578 structures and unions. In this last case, a frontend can immediately
29579 display the value of simple data types and create variable objects for
29580 other data types when the user wishes to explore their values in
29583 This command is deprecated in favor of the
29584 @samp{-stack-list-variables} command.
29586 @subsubheading @value{GDBN} Command
29588 @samp{info locals} in @value{GDBN}, @samp{gdb_get_locals} in @code{gdbtk}.
29590 @subsubheading Example
29594 -stack-list-locals 0
29595 ^done,locals=[name="A",name="B",name="C"]
29597 -stack-list-locals --all-values
29598 ^done,locals=[@{name="A",value="1"@},@{name="B",value="2"@},
29599 @{name="C",value="@{1, 2, 3@}"@}]
29600 -stack-list-locals --simple-values
29601 ^done,locals=[@{name="A",type="int",value="1"@},
29602 @{name="B",type="int",value="2"@},@{name="C",type="int [3]"@}]
29606 @subheading The @code{-stack-list-variables} Command
29607 @findex -stack-list-variables
29609 @subsubheading Synopsis
29612 -stack-list-variables @var{print-values}
29615 Display the names of local variables and function arguments for the selected frame. If
29616 @var{print-values} is 0 or @code{--no-values}, print only the names of
29617 the variables; if it is 1 or @code{--all-values}, print also their
29618 values; and if it is 2 or @code{--simple-values}, print the name,
29619 type and value for simple data types, and the name and type for arrays,
29620 structures and unions.
29622 @subsubheading Example
29626 -stack-list-variables --thread 1 --frame 0 --all-values
29627 ^done,variables=[@{name="x",value="11"@},@{name="s",value="@{a = 1, b = 2@}"@}]
29632 @subheading The @code{-stack-select-frame} Command
29633 @findex -stack-select-frame
29635 @subsubheading Synopsis
29638 -stack-select-frame @var{framenum}
29641 Change the selected frame. Select a different frame @var{framenum} on
29644 This command in deprecated in favor of passing the @samp{--frame}
29645 option to every command.
29647 @subsubheading @value{GDBN} Command
29649 The corresponding @value{GDBN} commands are @samp{frame}, @samp{up},
29650 @samp{down}, @samp{select-frame}, @samp{up-silent}, and @samp{down-silent}.
29652 @subsubheading Example
29656 -stack-select-frame 2
29661 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
29662 @node GDB/MI Variable Objects
29663 @section @sc{gdb/mi} Variable Objects
29667 @subheading Motivation for Variable Objects in @sc{gdb/mi}
29669 For the implementation of a variable debugger window (locals, watched
29670 expressions, etc.), we are proposing the adaptation of the existing code
29671 used by @code{Insight}.
29673 The two main reasons for that are:
29677 It has been proven in practice (it is already on its second generation).
29680 It will shorten development time (needless to say how important it is
29684 The original interface was designed to be used by Tcl code, so it was
29685 slightly changed so it could be used through @sc{gdb/mi}. This section
29686 describes the @sc{gdb/mi} operations that will be available and gives some
29687 hints about their use.
29689 @emph{Note}: In addition to the set of operations described here, we
29690 expect the @sc{gui} implementation of a variable window to require, at
29691 least, the following operations:
29694 @item @code{-gdb-show} @code{output-radix}
29695 @item @code{-stack-list-arguments}
29696 @item @code{-stack-list-locals}
29697 @item @code{-stack-select-frame}
29702 @subheading Introduction to Variable Objects
29704 @cindex variable objects in @sc{gdb/mi}
29706 Variable objects are "object-oriented" MI interface for examining and
29707 changing values of expressions. Unlike some other MI interfaces that
29708 work with expressions, variable objects are specifically designed for
29709 simple and efficient presentation in the frontend. A variable object
29710 is identified by string name. When a variable object is created, the
29711 frontend specifies the expression for that variable object. The
29712 expression can be a simple variable, or it can be an arbitrary complex
29713 expression, and can even involve CPU registers. After creating a
29714 variable object, the frontend can invoke other variable object
29715 operations---for example to obtain or change the value of a variable
29716 object, or to change display format.
29718 Variable objects have hierarchical tree structure. Any variable object
29719 that corresponds to a composite type, such as structure in C, has
29720 a number of child variable objects, for example corresponding to each
29721 element of a structure. A child variable object can itself have
29722 children, recursively. Recursion ends when we reach
29723 leaf variable objects, which always have built-in types. Child variable
29724 objects are created only by explicit request, so if a frontend
29725 is not interested in the children of a particular variable object, no
29726 child will be created.
29728 For a leaf variable object it is possible to obtain its value as a
29729 string, or set the value from a string. String value can be also
29730 obtained for a non-leaf variable object, but it's generally a string
29731 that only indicates the type of the object, and does not list its
29732 contents. Assignment to a non-leaf variable object is not allowed.
29734 A frontend does not need to read the values of all variable objects each time
29735 the program stops. Instead, MI provides an update command that lists all
29736 variable objects whose values has changed since the last update
29737 operation. This considerably reduces the amount of data that must
29738 be transferred to the frontend. As noted above, children variable
29739 objects are created on demand, and only leaf variable objects have a
29740 real value. As result, gdb will read target memory only for leaf
29741 variables that frontend has created.
29743 The automatic update is not always desirable. For example, a frontend
29744 might want to keep a value of some expression for future reference,
29745 and never update it. For another example, fetching memory is
29746 relatively slow for embedded targets, so a frontend might want
29747 to disable automatic update for the variables that are either not
29748 visible on the screen, or ``closed''. This is possible using so
29749 called ``frozen variable objects''. Such variable objects are never
29750 implicitly updated.
29752 Variable objects can be either @dfn{fixed} or @dfn{floating}. For the
29753 fixed variable object, the expression is parsed when the variable
29754 object is created, including associating identifiers to specific
29755 variables. The meaning of expression never changes. For a floating
29756 variable object the values of variables whose names appear in the
29757 expressions are re-evaluated every time in the context of the current
29758 frame. Consider this example:
29763 struct work_state state;
29770 If a fixed variable object for the @code{state} variable is created in
29771 this function, and we enter the recursive call, the variable
29772 object will report the value of @code{state} in the top-level
29773 @code{do_work} invocation. On the other hand, a floating variable
29774 object will report the value of @code{state} in the current frame.
29776 If an expression specified when creating a fixed variable object
29777 refers to a local variable, the variable object becomes bound to the
29778 thread and frame in which the variable object is created. When such
29779 variable object is updated, @value{GDBN} makes sure that the
29780 thread/frame combination the variable object is bound to still exists,
29781 and re-evaluates the variable object in context of that thread/frame.
29783 The following is the complete set of @sc{gdb/mi} operations defined to
29784 access this functionality:
29786 @multitable @columnfractions .4 .6
29787 @item @strong{Operation}
29788 @tab @strong{Description}
29790 @item @code{-enable-pretty-printing}
29791 @tab enable Python-based pretty-printing
29792 @item @code{-var-create}
29793 @tab create a variable object
29794 @item @code{-var-delete}
29795 @tab delete the variable object and/or its children
29796 @item @code{-var-set-format}
29797 @tab set the display format of this variable
29798 @item @code{-var-show-format}
29799 @tab show the display format of this variable
29800 @item @code{-var-info-num-children}
29801 @tab tells how many children this object has
29802 @item @code{-var-list-children}
29803 @tab return a list of the object's children
29804 @item @code{-var-info-type}
29805 @tab show the type of this variable object
29806 @item @code{-var-info-expression}
29807 @tab print parent-relative expression that this variable object represents
29808 @item @code{-var-info-path-expression}
29809 @tab print full expression that this variable object represents
29810 @item @code{-var-show-attributes}
29811 @tab is this variable editable? does it exist here?
29812 @item @code{-var-evaluate-expression}
29813 @tab get the value of this variable
29814 @item @code{-var-assign}
29815 @tab set the value of this variable
29816 @item @code{-var-update}
29817 @tab update the variable and its children
29818 @item @code{-var-set-frozen}
29819 @tab set frozeness attribute
29820 @item @code{-var-set-update-range}
29821 @tab set range of children to display on update
29824 In the next subsection we describe each operation in detail and suggest
29825 how it can be used.
29827 @subheading Description And Use of Operations on Variable Objects
29829 @subheading The @code{-enable-pretty-printing} Command
29830 @findex -enable-pretty-printing
29833 -enable-pretty-printing
29836 @value{GDBN} allows Python-based visualizers to affect the output of the
29837 MI variable object commands. However, because there was no way to
29838 implement this in a fully backward-compatible way, a front end must
29839 request that this functionality be enabled.
29841 Once enabled, this feature cannot be disabled.
29843 Note that if Python support has not been compiled into @value{GDBN},
29844 this command will still succeed (and do nothing).
29846 This feature is currently (as of @value{GDBN} 7.0) experimental, and
29847 may work differently in future versions of @value{GDBN}.
29849 @subheading The @code{-var-create} Command
29850 @findex -var-create
29852 @subsubheading Synopsis
29855 -var-create @{@var{name} | "-"@}
29856 @{@var{frame-addr} | "*" | "@@"@} @var{expression}
29859 This operation creates a variable object, which allows the monitoring of
29860 a variable, the result of an expression, a memory cell or a CPU
29863 The @var{name} parameter is the string by which the object can be
29864 referenced. It must be unique. If @samp{-} is specified, the varobj
29865 system will generate a string ``varNNNNNN'' automatically. It will be
29866 unique provided that one does not specify @var{name} of that format.
29867 The command fails if a duplicate name is found.
29869 The frame under which the expression should be evaluated can be
29870 specified by @var{frame-addr}. A @samp{*} indicates that the current
29871 frame should be used. A @samp{@@} indicates that a floating variable
29872 object must be created.
29874 @var{expression} is any expression valid on the current language set (must not
29875 begin with a @samp{*}), or one of the following:
29879 @samp{*@var{addr}}, where @var{addr} is the address of a memory cell
29882 @samp{*@var{addr}-@var{addr}} --- a memory address range (TBD)
29885 @samp{$@var{regname}} --- a CPU register name
29888 @cindex dynamic varobj
29889 A varobj's contents may be provided by a Python-based pretty-printer. In this
29890 case the varobj is known as a @dfn{dynamic varobj}. Dynamic varobjs
29891 have slightly different semantics in some cases. If the
29892 @code{-enable-pretty-printing} command is not sent, then @value{GDBN}
29893 will never create a dynamic varobj. This ensures backward
29894 compatibility for existing clients.
29896 @subsubheading Result
29898 This operation returns attributes of the newly-created varobj. These
29903 The name of the varobj.
29906 The number of children of the varobj. This number is not necessarily
29907 reliable for a dynamic varobj. Instead, you must examine the
29908 @samp{has_more} attribute.
29911 The varobj's scalar value. For a varobj whose type is some sort of
29912 aggregate (e.g., a @code{struct}), or for a dynamic varobj, this value
29913 will not be interesting.
29916 The varobj's type. This is a string representation of the type, as
29917 would be printed by the @value{GDBN} CLI. If @samp{print object}
29918 (@pxref{Print Settings, set print object}) is set to @code{on}, the
29919 @emph{actual} (derived) type of the object is shown rather than the
29920 @emph{declared} one.
29923 If a variable object is bound to a specific thread, then this is the
29924 thread's identifier.
29927 For a dynamic varobj, this indicates whether there appear to be any
29928 children available. For a non-dynamic varobj, this will be 0.
29931 This attribute will be present and have the value @samp{1} if the
29932 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
29933 then this attribute will not be present.
29936 A dynamic varobj can supply a display hint to the front end. The
29937 value comes directly from the Python pretty-printer object's
29938 @code{display_hint} method. @xref{Pretty Printing API}.
29941 Typical output will look like this:
29944 name="@var{name}",numchild="@var{N}",type="@var{type}",thread-id="@var{M}",
29945 has_more="@var{has_more}"
29949 @subheading The @code{-var-delete} Command
29950 @findex -var-delete
29952 @subsubheading Synopsis
29955 -var-delete [ -c ] @var{name}
29958 Deletes a previously created variable object and all of its children.
29959 With the @samp{-c} option, just deletes the children.
29961 Returns an error if the object @var{name} is not found.
29964 @subheading The @code{-var-set-format} Command
29965 @findex -var-set-format
29967 @subsubheading Synopsis
29970 -var-set-format @var{name} @var{format-spec}
29973 Sets the output format for the value of the object @var{name} to be
29976 @anchor{-var-set-format}
29977 The syntax for the @var{format-spec} is as follows:
29980 @var{format-spec} @expansion{}
29981 @{binary | decimal | hexadecimal | octal | natural@}
29984 The natural format is the default format choosen automatically
29985 based on the variable type (like decimal for an @code{int}, hex
29986 for pointers, etc.).
29988 For a variable with children, the format is set only on the
29989 variable itself, and the children are not affected.
29991 @subheading The @code{-var-show-format} Command
29992 @findex -var-show-format
29994 @subsubheading Synopsis
29997 -var-show-format @var{name}
30000 Returns the format used to display the value of the object @var{name}.
30003 @var{format} @expansion{}
30008 @subheading The @code{-var-info-num-children} Command
30009 @findex -var-info-num-children
30011 @subsubheading Synopsis
30014 -var-info-num-children @var{name}
30017 Returns the number of children of a variable object @var{name}:
30023 Note that this number is not completely reliable for a dynamic varobj.
30024 It will return the current number of children, but more children may
30028 @subheading The @code{-var-list-children} Command
30029 @findex -var-list-children
30031 @subsubheading Synopsis
30034 -var-list-children [@var{print-values}] @var{name} [@var{from} @var{to}]
30036 @anchor{-var-list-children}
30038 Return a list of the children of the specified variable object and
30039 create variable objects for them, if they do not already exist. With
30040 a single argument or if @var{print-values} has a value of 0 or
30041 @code{--no-values}, print only the names of the variables; if
30042 @var{print-values} is 1 or @code{--all-values}, also print their
30043 values; and if it is 2 or @code{--simple-values} print the name and
30044 value for simple data types and just the name for arrays, structures
30047 @var{from} and @var{to}, if specified, indicate the range of children
30048 to report. If @var{from} or @var{to} is less than zero, the range is
30049 reset and all children will be reported. Otherwise, children starting
30050 at @var{from} (zero-based) and up to and excluding @var{to} will be
30053 If a child range is requested, it will only affect the current call to
30054 @code{-var-list-children}, but not future calls to @code{-var-update}.
30055 For this, you must instead use @code{-var-set-update-range}. The
30056 intent of this approach is to enable a front end to implement any
30057 update approach it likes; for example, scrolling a view may cause the
30058 front end to request more children with @code{-var-list-children}, and
30059 then the front end could call @code{-var-set-update-range} with a
30060 different range to ensure that future updates are restricted to just
30063 For each child the following results are returned:
30068 Name of the variable object created for this child.
30071 The expression to be shown to the user by the front end to designate this child.
30072 For example this may be the name of a structure member.
30074 For a dynamic varobj, this value cannot be used to form an
30075 expression. There is no way to do this at all with a dynamic varobj.
30077 For C/C@t{++} structures there are several pseudo children returned to
30078 designate access qualifiers. For these pseudo children @var{exp} is
30079 @samp{public}, @samp{private}, or @samp{protected}. In this case the
30080 type and value are not present.
30082 A dynamic varobj will not report the access qualifying
30083 pseudo-children, regardless of the language. This information is not
30084 available at all with a dynamic varobj.
30087 Number of children this child has. For a dynamic varobj, this will be
30091 The type of the child. If @samp{print object}
30092 (@pxref{Print Settings, set print object}) is set to @code{on}, the
30093 @emph{actual} (derived) type of the object is shown rather than the
30094 @emph{declared} one.
30097 If values were requested, this is the value.
30100 If this variable object is associated with a thread, this is the thread id.
30101 Otherwise this result is not present.
30104 If the variable object is frozen, this variable will be present with a value of 1.
30107 The result may have its own attributes:
30111 A dynamic varobj can supply a display hint to the front end. The
30112 value comes directly from the Python pretty-printer object's
30113 @code{display_hint} method. @xref{Pretty Printing API}.
30116 This is an integer attribute which is nonzero if there are children
30117 remaining after the end of the selected range.
30120 @subsubheading Example
30124 -var-list-children n
30125 ^done,numchild=@var{n},children=[child=@{name=@var{name},exp=@var{exp},
30126 numchild=@var{n},type=@var{type}@},@r{(repeats N times)}]
30128 -var-list-children --all-values n
30129 ^done,numchild=@var{n},children=[child=@{name=@var{name},exp=@var{exp},
30130 numchild=@var{n},value=@var{value},type=@var{type}@},@r{(repeats N times)}]
30134 @subheading The @code{-var-info-type} Command
30135 @findex -var-info-type
30137 @subsubheading Synopsis
30140 -var-info-type @var{name}
30143 Returns the type of the specified variable @var{name}. The type is
30144 returned as a string in the same format as it is output by the
30148 type=@var{typename}
30152 @subheading The @code{-var-info-expression} Command
30153 @findex -var-info-expression
30155 @subsubheading Synopsis
30158 -var-info-expression @var{name}
30161 Returns a string that is suitable for presenting this
30162 variable object in user interface. The string is generally
30163 not valid expression in the current language, and cannot be evaluated.
30165 For example, if @code{a} is an array, and variable object
30166 @code{A} was created for @code{a}, then we'll get this output:
30169 (gdb) -var-info-expression A.1
30170 ^done,lang="C",exp="1"
30174 Here, the values of @code{lang} can be @code{@{"C" | "C++" | "Java"@}}.
30176 Note that the output of the @code{-var-list-children} command also
30177 includes those expressions, so the @code{-var-info-expression} command
30180 @subheading The @code{-var-info-path-expression} Command
30181 @findex -var-info-path-expression
30183 @subsubheading Synopsis
30186 -var-info-path-expression @var{name}
30189 Returns an expression that can be evaluated in the current
30190 context and will yield the same value that a variable object has.
30191 Compare this with the @code{-var-info-expression} command, which
30192 result can be used only for UI presentation. Typical use of
30193 the @code{-var-info-path-expression} command is creating a
30194 watchpoint from a variable object.
30196 This command is currently not valid for children of a dynamic varobj,
30197 and will give an error when invoked on one.
30199 For example, suppose @code{C} is a C@t{++} class, derived from class
30200 @code{Base}, and that the @code{Base} class has a member called
30201 @code{m_size}. Assume a variable @code{c} is has the type of
30202 @code{C} and a variable object @code{C} was created for variable
30203 @code{c}. Then, we'll get this output:
30205 (gdb) -var-info-path-expression C.Base.public.m_size
30206 ^done,path_expr=((Base)c).m_size)
30209 @subheading The @code{-var-show-attributes} Command
30210 @findex -var-show-attributes
30212 @subsubheading Synopsis
30215 -var-show-attributes @var{name}
30218 List attributes of the specified variable object @var{name}:
30221 status=@var{attr} [ ( ,@var{attr} )* ]
30225 where @var{attr} is @code{@{ @{ editable | noneditable @} | TBD @}}.
30227 @subheading The @code{-var-evaluate-expression} Command
30228 @findex -var-evaluate-expression
30230 @subsubheading Synopsis
30233 -var-evaluate-expression [-f @var{format-spec}] @var{name}
30236 Evaluates the expression that is represented by the specified variable
30237 object and returns its value as a string. The format of the string
30238 can be specified with the @samp{-f} option. The possible values of
30239 this option are the same as for @code{-var-set-format}
30240 (@pxref{-var-set-format}). If the @samp{-f} option is not specified,
30241 the current display format will be used. The current display format
30242 can be changed using the @code{-var-set-format} command.
30248 Note that one must invoke @code{-var-list-children} for a variable
30249 before the value of a child variable can be evaluated.
30251 @subheading The @code{-var-assign} Command
30252 @findex -var-assign
30254 @subsubheading Synopsis
30257 -var-assign @var{name} @var{expression}
30260 Assigns the value of @var{expression} to the variable object specified
30261 by @var{name}. The object must be @samp{editable}. If the variable's
30262 value is altered by the assign, the variable will show up in any
30263 subsequent @code{-var-update} list.
30265 @subsubheading Example
30273 ^done,changelist=[@{name="var1",in_scope="true",type_changed="false"@}]
30277 @subheading The @code{-var-update} Command
30278 @findex -var-update
30280 @subsubheading Synopsis
30283 -var-update [@var{print-values}] @{@var{name} | "*"@}
30286 Reevaluate the expressions corresponding to the variable object
30287 @var{name} and all its direct and indirect children, and return the
30288 list of variable objects whose values have changed; @var{name} must
30289 be a root variable object. Here, ``changed'' means that the result of
30290 @code{-var-evaluate-expression} before and after the
30291 @code{-var-update} is different. If @samp{*} is used as the variable
30292 object names, all existing variable objects are updated, except
30293 for frozen ones (@pxref{-var-set-frozen}). The option
30294 @var{print-values} determines whether both names and values, or just
30295 names are printed. The possible values of this option are the same
30296 as for @code{-var-list-children} (@pxref{-var-list-children}). It is
30297 recommended to use the @samp{--all-values} option, to reduce the
30298 number of MI commands needed on each program stop.
30300 With the @samp{*} parameter, if a variable object is bound to a
30301 currently running thread, it will not be updated, without any
30304 If @code{-var-set-update-range} was previously used on a varobj, then
30305 only the selected range of children will be reported.
30307 @code{-var-update} reports all the changed varobjs in a tuple named
30310 Each item in the change list is itself a tuple holding:
30314 The name of the varobj.
30317 If values were requested for this update, then this field will be
30318 present and will hold the value of the varobj.
30321 @anchor{-var-update}
30322 This field is a string which may take one of three values:
30326 The variable object's current value is valid.
30329 The variable object does not currently hold a valid value but it may
30330 hold one in the future if its associated expression comes back into
30334 The variable object no longer holds a valid value.
30335 This can occur when the executable file being debugged has changed,
30336 either through recompilation or by using the @value{GDBN} @code{file}
30337 command. The front end should normally choose to delete these variable
30341 In the future new values may be added to this list so the front should
30342 be prepared for this possibility. @xref{GDB/MI Development and Front Ends, ,@sc{GDB/MI} Development and Front Ends}.
30345 This is only present if the varobj is still valid. If the type
30346 changed, then this will be the string @samp{true}; otherwise it will
30349 When a varobj's type changes, its children are also likely to have
30350 become incorrect. Therefore, the varobj's children are automatically
30351 deleted when this attribute is @samp{true}. Also, the varobj's update
30352 range, when set using the @code{-var-set-update-range} command, is
30356 If the varobj's type changed, then this field will be present and will
30359 @item new_num_children
30360 For a dynamic varobj, if the number of children changed, or if the
30361 type changed, this will be the new number of children.
30363 The @samp{numchild} field in other varobj responses is generally not
30364 valid for a dynamic varobj -- it will show the number of children that
30365 @value{GDBN} knows about, but because dynamic varobjs lazily
30366 instantiate their children, this will not reflect the number of
30367 children which may be available.
30369 The @samp{new_num_children} attribute only reports changes to the
30370 number of children known by @value{GDBN}. This is the only way to
30371 detect whether an update has removed children (which necessarily can
30372 only happen at the end of the update range).
30375 The display hint, if any.
30378 This is an integer value, which will be 1 if there are more children
30379 available outside the varobj's update range.
30382 This attribute will be present and have the value @samp{1} if the
30383 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
30384 then this attribute will not be present.
30387 If new children were added to a dynamic varobj within the selected
30388 update range (as set by @code{-var-set-update-range}), then they will
30389 be listed in this attribute.
30392 @subsubheading Example
30399 -var-update --all-values var1
30400 ^done,changelist=[@{name="var1",value="3",in_scope="true",
30401 type_changed="false"@}]
30405 @subheading The @code{-var-set-frozen} Command
30406 @findex -var-set-frozen
30407 @anchor{-var-set-frozen}
30409 @subsubheading Synopsis
30412 -var-set-frozen @var{name} @var{flag}
30415 Set the frozenness flag on the variable object @var{name}. The
30416 @var{flag} parameter should be either @samp{1} to make the variable
30417 frozen or @samp{0} to make it unfrozen. If a variable object is
30418 frozen, then neither itself, nor any of its children, are
30419 implicitly updated by @code{-var-update} of
30420 a parent variable or by @code{-var-update *}. Only
30421 @code{-var-update} of the variable itself will update its value and
30422 values of its children. After a variable object is unfrozen, it is
30423 implicitly updated by all subsequent @code{-var-update} operations.
30424 Unfreezing a variable does not update it, only subsequent
30425 @code{-var-update} does.
30427 @subsubheading Example
30431 -var-set-frozen V 1
30436 @subheading The @code{-var-set-update-range} command
30437 @findex -var-set-update-range
30438 @anchor{-var-set-update-range}
30440 @subsubheading Synopsis
30443 -var-set-update-range @var{name} @var{from} @var{to}
30446 Set the range of children to be returned by future invocations of
30447 @code{-var-update}.
30449 @var{from} and @var{to} indicate the range of children to report. If
30450 @var{from} or @var{to} is less than zero, the range is reset and all
30451 children will be reported. Otherwise, children starting at @var{from}
30452 (zero-based) and up to and excluding @var{to} will be reported.
30454 @subsubheading Example
30458 -var-set-update-range V 1 2
30462 @subheading The @code{-var-set-visualizer} command
30463 @findex -var-set-visualizer
30464 @anchor{-var-set-visualizer}
30466 @subsubheading Synopsis
30469 -var-set-visualizer @var{name} @var{visualizer}
30472 Set a visualizer for the variable object @var{name}.
30474 @var{visualizer} is the visualizer to use. The special value
30475 @samp{None} means to disable any visualizer in use.
30477 If not @samp{None}, @var{visualizer} must be a Python expression.
30478 This expression must evaluate to a callable object which accepts a
30479 single argument. @value{GDBN} will call this object with the value of
30480 the varobj @var{name} as an argument (this is done so that the same
30481 Python pretty-printing code can be used for both the CLI and MI).
30482 When called, this object must return an object which conforms to the
30483 pretty-printing interface (@pxref{Pretty Printing API}).
30485 The pre-defined function @code{gdb.default_visualizer} may be used to
30486 select a visualizer by following the built-in process
30487 (@pxref{Selecting Pretty-Printers}). This is done automatically when
30488 a varobj is created, and so ordinarily is not needed.
30490 This feature is only available if Python support is enabled. The MI
30491 command @code{-list-features} (@pxref{GDB/MI Miscellaneous Commands})
30492 can be used to check this.
30494 @subsubheading Example
30496 Resetting the visualizer:
30500 -var-set-visualizer V None
30504 Reselecting the default (type-based) visualizer:
30508 -var-set-visualizer V gdb.default_visualizer
30512 Suppose @code{SomeClass} is a visualizer class. A lambda expression
30513 can be used to instantiate this class for a varobj:
30517 -var-set-visualizer V "lambda val: SomeClass()"
30521 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30522 @node GDB/MI Data Manipulation
30523 @section @sc{gdb/mi} Data Manipulation
30525 @cindex data manipulation, in @sc{gdb/mi}
30526 @cindex @sc{gdb/mi}, data manipulation
30527 This section describes the @sc{gdb/mi} commands that manipulate data:
30528 examine memory and registers, evaluate expressions, etc.
30530 @c REMOVED FROM THE INTERFACE.
30531 @c @subheading -data-assign
30532 @c Change the value of a program variable. Plenty of side effects.
30533 @c @subsubheading GDB Command
30535 @c @subsubheading Example
30538 @subheading The @code{-data-disassemble} Command
30539 @findex -data-disassemble
30541 @subsubheading Synopsis
30545 [ -s @var{start-addr} -e @var{end-addr} ]
30546 | [ -f @var{filename} -l @var{linenum} [ -n @var{lines} ] ]
30554 @item @var{start-addr}
30555 is the beginning address (or @code{$pc})
30556 @item @var{end-addr}
30558 @item @var{filename}
30559 is the name of the file to disassemble
30560 @item @var{linenum}
30561 is the line number to disassemble around
30563 is the number of disassembly lines to be produced. If it is -1,
30564 the whole function will be disassembled, in case no @var{end-addr} is
30565 specified. If @var{end-addr} is specified as a non-zero value, and
30566 @var{lines} is lower than the number of disassembly lines between
30567 @var{start-addr} and @var{end-addr}, only @var{lines} lines are
30568 displayed; if @var{lines} is higher than the number of lines between
30569 @var{start-addr} and @var{end-addr}, only the lines up to @var{end-addr}
30572 is either 0 (meaning only disassembly), 1 (meaning mixed source and
30573 disassembly), 2 (meaning disassembly with raw opcodes), or 3 (meaning
30574 mixed source and disassembly with raw opcodes).
30577 @subsubheading Result
30579 The output for each instruction is composed of four fields:
30588 Note that whatever included in the instruction field, is not manipulated
30589 directly by @sc{gdb/mi}, i.e., it is not possible to adjust its format.
30591 @subsubheading @value{GDBN} Command
30593 There's no direct mapping from this command to the CLI.
30595 @subsubheading Example
30597 Disassemble from the current value of @code{$pc} to @code{$pc + 20}:
30601 -data-disassemble -s $pc -e "$pc + 20" -- 0
30604 @{address="0x000107c0",func-name="main",offset="4",
30605 inst="mov 2, %o0"@},
30606 @{address="0x000107c4",func-name="main",offset="8",
30607 inst="sethi %hi(0x11800), %o2"@},
30608 @{address="0x000107c8",func-name="main",offset="12",
30609 inst="or %o2, 0x140, %o1\t! 0x11940 <_lib_version+8>"@},
30610 @{address="0x000107cc",func-name="main",offset="16",
30611 inst="sethi %hi(0x11800), %o2"@},
30612 @{address="0x000107d0",func-name="main",offset="20",
30613 inst="or %o2, 0x168, %o4\t! 0x11968 <_lib_version+48>"@}]
30617 Disassemble the whole @code{main} function. Line 32 is part of
30621 -data-disassemble -f basics.c -l 32 -- 0
30623 @{address="0x000107bc",func-name="main",offset="0",
30624 inst="save %sp, -112, %sp"@},
30625 @{address="0x000107c0",func-name="main",offset="4",
30626 inst="mov 2, %o0"@},
30627 @{address="0x000107c4",func-name="main",offset="8",
30628 inst="sethi %hi(0x11800), %o2"@},
30630 @{address="0x0001081c",func-name="main",offset="96",inst="ret "@},
30631 @{address="0x00010820",func-name="main",offset="100",inst="restore "@}]
30635 Disassemble 3 instructions from the start of @code{main}:
30639 -data-disassemble -f basics.c -l 32 -n 3 -- 0
30641 @{address="0x000107bc",func-name="main",offset="0",
30642 inst="save %sp, -112, %sp"@},
30643 @{address="0x000107c0",func-name="main",offset="4",
30644 inst="mov 2, %o0"@},
30645 @{address="0x000107c4",func-name="main",offset="8",
30646 inst="sethi %hi(0x11800), %o2"@}]
30650 Disassemble 3 instructions from the start of @code{main} in mixed mode:
30654 -data-disassemble -f basics.c -l 32 -n 3 -- 1
30656 src_and_asm_line=@{line="31",
30657 file="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb/ \
30658 testsuite/gdb.mi/basics.c",line_asm_insn=[
30659 @{address="0x000107bc",func-name="main",offset="0",
30660 inst="save %sp, -112, %sp"@}]@},
30661 src_and_asm_line=@{line="32",
30662 file="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb/ \
30663 testsuite/gdb.mi/basics.c",line_asm_insn=[
30664 @{address="0x000107c0",func-name="main",offset="4",
30665 inst="mov 2, %o0"@},
30666 @{address="0x000107c4",func-name="main",offset="8",
30667 inst="sethi %hi(0x11800), %o2"@}]@}]
30672 @subheading The @code{-data-evaluate-expression} Command
30673 @findex -data-evaluate-expression
30675 @subsubheading Synopsis
30678 -data-evaluate-expression @var{expr}
30681 Evaluate @var{expr} as an expression. The expression could contain an
30682 inferior function call. The function call will execute synchronously.
30683 If the expression contains spaces, it must be enclosed in double quotes.
30685 @subsubheading @value{GDBN} Command
30687 The corresponding @value{GDBN} commands are @samp{print}, @samp{output}, and
30688 @samp{call}. In @code{gdbtk} only, there's a corresponding
30689 @samp{gdb_eval} command.
30691 @subsubheading Example
30693 In the following example, the numbers that precede the commands are the
30694 @dfn{tokens} described in @ref{GDB/MI Command Syntax, ,@sc{gdb/mi}
30695 Command Syntax}. Notice how @sc{gdb/mi} returns the same tokens in its
30699 211-data-evaluate-expression A
30702 311-data-evaluate-expression &A
30703 311^done,value="0xefffeb7c"
30705 411-data-evaluate-expression A+3
30708 511-data-evaluate-expression "A + 3"
30714 @subheading The @code{-data-list-changed-registers} Command
30715 @findex -data-list-changed-registers
30717 @subsubheading Synopsis
30720 -data-list-changed-registers
30723 Display a list of the registers that have changed.
30725 @subsubheading @value{GDBN} Command
30727 @value{GDBN} doesn't have a direct analog for this command; @code{gdbtk}
30728 has the corresponding command @samp{gdb_changed_register_list}.
30730 @subsubheading Example
30732 On a PPC MBX board:
30740 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",frame=@{
30741 func="main",args=[],file="try.c",fullname="/home/foo/bar/try.c",
30744 -data-list-changed-registers
30745 ^done,changed-registers=["0","1","2","4","5","6","7","8","9",
30746 "10","11","13","14","15","16","17","18","19","20","21","22","23",
30747 "24","25","26","27","28","30","31","64","65","66","67","69"]
30752 @subheading The @code{-data-list-register-names} Command
30753 @findex -data-list-register-names
30755 @subsubheading Synopsis
30758 -data-list-register-names [ ( @var{regno} )+ ]
30761 Show a list of register names for the current target. If no arguments
30762 are given, it shows a list of the names of all the registers. If
30763 integer numbers are given as arguments, it will print a list of the
30764 names of the registers corresponding to the arguments. To ensure
30765 consistency between a register name and its number, the output list may
30766 include empty register names.
30768 @subsubheading @value{GDBN} Command
30770 @value{GDBN} does not have a command which corresponds to
30771 @samp{-data-list-register-names}. In @code{gdbtk} there is a
30772 corresponding command @samp{gdb_regnames}.
30774 @subsubheading Example
30776 For the PPC MBX board:
30779 -data-list-register-names
30780 ^done,register-names=["r0","r1","r2","r3","r4","r5","r6","r7",
30781 "r8","r9","r10","r11","r12","r13","r14","r15","r16","r17","r18",
30782 "r19","r20","r21","r22","r23","r24","r25","r26","r27","r28","r29",
30783 "r30","r31","f0","f1","f2","f3","f4","f5","f6","f7","f8","f9",
30784 "f10","f11","f12","f13","f14","f15","f16","f17","f18","f19","f20",
30785 "f21","f22","f23","f24","f25","f26","f27","f28","f29","f30","f31",
30786 "", "pc","ps","cr","lr","ctr","xer"]
30788 -data-list-register-names 1 2 3
30789 ^done,register-names=["r1","r2","r3"]
30793 @subheading The @code{-data-list-register-values} Command
30794 @findex -data-list-register-values
30796 @subsubheading Synopsis
30799 -data-list-register-values @var{fmt} [ ( @var{regno} )*]
30802 Display the registers' contents. @var{fmt} is the format according to
30803 which the registers' contents are to be returned, followed by an optional
30804 list of numbers specifying the registers to display. A missing list of
30805 numbers indicates that the contents of all the registers must be returned.
30807 Allowed formats for @var{fmt} are:
30824 @subsubheading @value{GDBN} Command
30826 The corresponding @value{GDBN} commands are @samp{info reg}, @samp{info
30827 all-reg}, and (in @code{gdbtk}) @samp{gdb_fetch_registers}.
30829 @subsubheading Example
30831 For a PPC MBX board (note: line breaks are for readability only, they
30832 don't appear in the actual output):
30836 -data-list-register-values r 64 65
30837 ^done,register-values=[@{number="64",value="0xfe00a300"@},
30838 @{number="65",value="0x00029002"@}]
30840 -data-list-register-values x
30841 ^done,register-values=[@{number="0",value="0xfe0043c8"@},
30842 @{number="1",value="0x3fff88"@},@{number="2",value="0xfffffffe"@},
30843 @{number="3",value="0x0"@},@{number="4",value="0xa"@},
30844 @{number="5",value="0x3fff68"@},@{number="6",value="0x3fff58"@},
30845 @{number="7",value="0xfe011e98"@},@{number="8",value="0x2"@},
30846 @{number="9",value="0xfa202820"@},@{number="10",value="0xfa202808"@},
30847 @{number="11",value="0x1"@},@{number="12",value="0x0"@},
30848 @{number="13",value="0x4544"@},@{number="14",value="0xffdfffff"@},
30849 @{number="15",value="0xffffffff"@},@{number="16",value="0xfffffeff"@},
30850 @{number="17",value="0xefffffed"@},@{number="18",value="0xfffffffe"@},
30851 @{number="19",value="0xffffffff"@},@{number="20",value="0xffffffff"@},
30852 @{number="21",value="0xffffffff"@},@{number="22",value="0xfffffff7"@},
30853 @{number="23",value="0xffffffff"@},@{number="24",value="0xffffffff"@},
30854 @{number="25",value="0xffffffff"@},@{number="26",value="0xfffffffb"@},
30855 @{number="27",value="0xffffffff"@},@{number="28",value="0xf7bfffff"@},
30856 @{number="29",value="0x0"@},@{number="30",value="0xfe010000"@},
30857 @{number="31",value="0x0"@},@{number="32",value="0x0"@},
30858 @{number="33",value="0x0"@},@{number="34",value="0x0"@},
30859 @{number="35",value="0x0"@},@{number="36",value="0x0"@},
30860 @{number="37",value="0x0"@},@{number="38",value="0x0"@},
30861 @{number="39",value="0x0"@},@{number="40",value="0x0"@},
30862 @{number="41",value="0x0"@},@{number="42",value="0x0"@},
30863 @{number="43",value="0x0"@},@{number="44",value="0x0"@},
30864 @{number="45",value="0x0"@},@{number="46",value="0x0"@},
30865 @{number="47",value="0x0"@},@{number="48",value="0x0"@},
30866 @{number="49",value="0x0"@},@{number="50",value="0x0"@},
30867 @{number="51",value="0x0"@},@{number="52",value="0x0"@},
30868 @{number="53",value="0x0"@},@{number="54",value="0x0"@},
30869 @{number="55",value="0x0"@},@{number="56",value="0x0"@},
30870 @{number="57",value="0x0"@},@{number="58",value="0x0"@},
30871 @{number="59",value="0x0"@},@{number="60",value="0x0"@},
30872 @{number="61",value="0x0"@},@{number="62",value="0x0"@},
30873 @{number="63",value="0x0"@},@{number="64",value="0xfe00a300"@},
30874 @{number="65",value="0x29002"@},@{number="66",value="0x202f04b5"@},
30875 @{number="67",value="0xfe0043b0"@},@{number="68",value="0xfe00b3e4"@},
30876 @{number="69",value="0x20002b03"@}]
30881 @subheading The @code{-data-read-memory} Command
30882 @findex -data-read-memory
30884 This command is deprecated, use @code{-data-read-memory-bytes} instead.
30886 @subsubheading Synopsis
30889 -data-read-memory [ -o @var{byte-offset} ]
30890 @var{address} @var{word-format} @var{word-size}
30891 @var{nr-rows} @var{nr-cols} [ @var{aschar} ]
30898 @item @var{address}
30899 An expression specifying the address of the first memory word to be
30900 read. Complex expressions containing embedded white space should be
30901 quoted using the C convention.
30903 @item @var{word-format}
30904 The format to be used to print the memory words. The notation is the
30905 same as for @value{GDBN}'s @code{print} command (@pxref{Output Formats,
30908 @item @var{word-size}
30909 The size of each memory word in bytes.
30911 @item @var{nr-rows}
30912 The number of rows in the output table.
30914 @item @var{nr-cols}
30915 The number of columns in the output table.
30918 If present, indicates that each row should include an @sc{ascii} dump. The
30919 value of @var{aschar} is used as a padding character when a byte is not a
30920 member of the printable @sc{ascii} character set (printable @sc{ascii}
30921 characters are those whose code is between 32 and 126, inclusively).
30923 @item @var{byte-offset}
30924 An offset to add to the @var{address} before fetching memory.
30927 This command displays memory contents as a table of @var{nr-rows} by
30928 @var{nr-cols} words, each word being @var{word-size} bytes. In total,
30929 @code{@var{nr-rows} * @var{nr-cols} * @var{word-size}} bytes are read
30930 (returned as @samp{total-bytes}). Should less than the requested number
30931 of bytes be returned by the target, the missing words are identified
30932 using @samp{N/A}. The number of bytes read from the target is returned
30933 in @samp{nr-bytes} and the starting address used to read memory in
30936 The address of the next/previous row or page is available in
30937 @samp{next-row} and @samp{prev-row}, @samp{next-page} and
30940 @subsubheading @value{GDBN} Command
30942 The corresponding @value{GDBN} command is @samp{x}. @code{gdbtk} has
30943 @samp{gdb_get_mem} memory read command.
30945 @subsubheading Example
30947 Read six bytes of memory starting at @code{bytes+6} but then offset by
30948 @code{-6} bytes. Format as three rows of two columns. One byte per
30949 word. Display each word in hex.
30953 9-data-read-memory -o -6 -- bytes+6 x 1 3 2
30954 9^done,addr="0x00001390",nr-bytes="6",total-bytes="6",
30955 next-row="0x00001396",prev-row="0x0000138e",next-page="0x00001396",
30956 prev-page="0x0000138a",memory=[
30957 @{addr="0x00001390",data=["0x00","0x01"]@},
30958 @{addr="0x00001392",data=["0x02","0x03"]@},
30959 @{addr="0x00001394",data=["0x04","0x05"]@}]
30963 Read two bytes of memory starting at address @code{shorts + 64} and
30964 display as a single word formatted in decimal.
30968 5-data-read-memory shorts+64 d 2 1 1
30969 5^done,addr="0x00001510",nr-bytes="2",total-bytes="2",
30970 next-row="0x00001512",prev-row="0x0000150e",
30971 next-page="0x00001512",prev-page="0x0000150e",memory=[
30972 @{addr="0x00001510",data=["128"]@}]
30976 Read thirty two bytes of memory starting at @code{bytes+16} and format
30977 as eight rows of four columns. Include a string encoding with @samp{x}
30978 used as the non-printable character.
30982 4-data-read-memory bytes+16 x 1 8 4 x
30983 4^done,addr="0x000013a0",nr-bytes="32",total-bytes="32",
30984 next-row="0x000013c0",prev-row="0x0000139c",
30985 next-page="0x000013c0",prev-page="0x00001380",memory=[
30986 @{addr="0x000013a0",data=["0x10","0x11","0x12","0x13"],ascii="xxxx"@},
30987 @{addr="0x000013a4",data=["0x14","0x15","0x16","0x17"],ascii="xxxx"@},
30988 @{addr="0x000013a8",data=["0x18","0x19","0x1a","0x1b"],ascii="xxxx"@},
30989 @{addr="0x000013ac",data=["0x1c","0x1d","0x1e","0x1f"],ascii="xxxx"@},
30990 @{addr="0x000013b0",data=["0x20","0x21","0x22","0x23"],ascii=" !\"#"@},
30991 @{addr="0x000013b4",data=["0x24","0x25","0x26","0x27"],ascii="$%&'"@},
30992 @{addr="0x000013b8",data=["0x28","0x29","0x2a","0x2b"],ascii="()*+"@},
30993 @{addr="0x000013bc",data=["0x2c","0x2d","0x2e","0x2f"],ascii=",-./"@}]
30997 @subheading The @code{-data-read-memory-bytes} Command
30998 @findex -data-read-memory-bytes
31000 @subsubheading Synopsis
31003 -data-read-memory-bytes [ -o @var{byte-offset} ]
31004 @var{address} @var{count}
31011 @item @var{address}
31012 An expression specifying the address of the first memory word to be
31013 read. Complex expressions containing embedded white space should be
31014 quoted using the C convention.
31017 The number of bytes to read. This should be an integer literal.
31019 @item @var{byte-offset}
31020 The offsets in bytes relative to @var{address} at which to start
31021 reading. This should be an integer literal. This option is provided
31022 so that a frontend is not required to first evaluate address and then
31023 perform address arithmetics itself.
31027 This command attempts to read all accessible memory regions in the
31028 specified range. First, all regions marked as unreadable in the memory
31029 map (if one is defined) will be skipped. @xref{Memory Region
31030 Attributes}. Second, @value{GDBN} will attempt to read the remaining
31031 regions. For each one, if reading full region results in an errors,
31032 @value{GDBN} will try to read a subset of the region.
31034 In general, every single byte in the region may be readable or not,
31035 and the only way to read every readable byte is to try a read at
31036 every address, which is not practical. Therefore, @value{GDBN} will
31037 attempt to read all accessible bytes at either beginning or the end
31038 of the region, using a binary division scheme. This heuristic works
31039 well for reading accross a memory map boundary. Note that if a region
31040 has a readable range that is neither at the beginning or the end,
31041 @value{GDBN} will not read it.
31043 The result record (@pxref{GDB/MI Result Records}) that is output of
31044 the command includes a field named @samp{memory} whose content is a
31045 list of tuples. Each tuple represent a successfully read memory block
31046 and has the following fields:
31050 The start address of the memory block, as hexadecimal literal.
31053 The end address of the memory block, as hexadecimal literal.
31056 The offset of the memory block, as hexadecimal literal, relative to
31057 the start address passed to @code{-data-read-memory-bytes}.
31060 The contents of the memory block, in hex.
31066 @subsubheading @value{GDBN} Command
31068 The corresponding @value{GDBN} command is @samp{x}.
31070 @subsubheading Example
31074 -data-read-memory-bytes &a 10
31075 ^done,memory=[@{begin="0xbffff154",offset="0x00000000",
31077 contents="01000000020000000300"@}]
31082 @subheading The @code{-data-write-memory-bytes} Command
31083 @findex -data-write-memory-bytes
31085 @subsubheading Synopsis
31088 -data-write-memory-bytes @var{address} @var{contents}
31095 @item @var{address}
31096 An expression specifying the address of the first memory word to be
31097 read. Complex expressions containing embedded white space should be
31098 quoted using the C convention.
31100 @item @var{contents}
31101 The hex-encoded bytes to write.
31105 @subsubheading @value{GDBN} Command
31107 There's no corresponding @value{GDBN} command.
31109 @subsubheading Example
31113 -data-write-memory-bytes &a "aabbccdd"
31119 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31120 @node GDB/MI Tracepoint Commands
31121 @section @sc{gdb/mi} Tracepoint Commands
31123 The commands defined in this section implement MI support for
31124 tracepoints. For detailed introduction, see @ref{Tracepoints}.
31126 @subheading The @code{-trace-find} Command
31127 @findex -trace-find
31129 @subsubheading Synopsis
31132 -trace-find @var{mode} [@var{parameters}@dots{}]
31135 Find a trace frame using criteria defined by @var{mode} and
31136 @var{parameters}. The following table lists permissible
31137 modes and their parameters. For details of operation, see @ref{tfind}.
31142 No parameters are required. Stops examining trace frames.
31145 An integer is required as parameter. Selects tracepoint frame with
31148 @item tracepoint-number
31149 An integer is required as parameter. Finds next
31150 trace frame that corresponds to tracepoint with the specified number.
31153 An address is required as parameter. Finds
31154 next trace frame that corresponds to any tracepoint at the specified
31157 @item pc-inside-range
31158 Two addresses are required as parameters. Finds next trace
31159 frame that corresponds to a tracepoint at an address inside the
31160 specified range. Both bounds are considered to be inside the range.
31162 @item pc-outside-range
31163 Two addresses are required as parameters. Finds
31164 next trace frame that corresponds to a tracepoint at an address outside
31165 the specified range. Both bounds are considered to be inside the range.
31168 Line specification is required as parameter. @xref{Specify Location}.
31169 Finds next trace frame that corresponds to a tracepoint at
31170 the specified location.
31174 If @samp{none} was passed as @var{mode}, the response does not
31175 have fields. Otherwise, the response may have the following fields:
31179 This field has either @samp{0} or @samp{1} as the value, depending
31180 on whether a matching tracepoint was found.
31183 The index of the found traceframe. This field is present iff
31184 the @samp{found} field has value of @samp{1}.
31187 The index of the found tracepoint. This field is present iff
31188 the @samp{found} field has value of @samp{1}.
31191 The information about the frame corresponding to the found trace
31192 frame. This field is present only if a trace frame was found.
31193 @xref{GDB/MI Frame Information}, for description of this field.
31197 @subsubheading @value{GDBN} Command
31199 The corresponding @value{GDBN} command is @samp{tfind}.
31201 @subheading -trace-define-variable
31202 @findex -trace-define-variable
31204 @subsubheading Synopsis
31207 -trace-define-variable @var{name} [ @var{value} ]
31210 Create trace variable @var{name} if it does not exist. If
31211 @var{value} is specified, sets the initial value of the specified
31212 trace variable to that value. Note that the @var{name} should start
31213 with the @samp{$} character.
31215 @subsubheading @value{GDBN} Command
31217 The corresponding @value{GDBN} command is @samp{tvariable}.
31219 @subheading -trace-list-variables
31220 @findex -trace-list-variables
31222 @subsubheading Synopsis
31225 -trace-list-variables
31228 Return a table of all defined trace variables. Each element of the
31229 table has the following fields:
31233 The name of the trace variable. This field is always present.
31236 The initial value. This is a 64-bit signed integer. This
31237 field is always present.
31240 The value the trace variable has at the moment. This is a 64-bit
31241 signed integer. This field is absent iff current value is
31242 not defined, for example if the trace was never run, or is
31247 @subsubheading @value{GDBN} Command
31249 The corresponding @value{GDBN} command is @samp{tvariables}.
31251 @subsubheading Example
31255 -trace-list-variables
31256 ^done,trace-variables=@{nr_rows="1",nr_cols="3",
31257 hdr=[@{width="15",alignment="-1",col_name="name",colhdr="Name"@},
31258 @{width="11",alignment="-1",col_name="initial",colhdr="Initial"@},
31259 @{width="11",alignment="-1",col_name="current",colhdr="Current"@}],
31260 body=[variable=@{name="$trace_timestamp",initial="0"@}
31261 variable=@{name="$foo",initial="10",current="15"@}]@}
31265 @subheading -trace-save
31266 @findex -trace-save
31268 @subsubheading Synopsis
31271 -trace-save [-r ] @var{filename}
31274 Saves the collected trace data to @var{filename}. Without the
31275 @samp{-r} option, the data is downloaded from the target and saved
31276 in a local file. With the @samp{-r} option the target is asked
31277 to perform the save.
31279 @subsubheading @value{GDBN} Command
31281 The corresponding @value{GDBN} command is @samp{tsave}.
31284 @subheading -trace-start
31285 @findex -trace-start
31287 @subsubheading Synopsis
31293 Starts a tracing experiments. The result of this command does not
31296 @subsubheading @value{GDBN} Command
31298 The corresponding @value{GDBN} command is @samp{tstart}.
31300 @subheading -trace-status
31301 @findex -trace-status
31303 @subsubheading Synopsis
31309 Obtains the status of a tracing experiment. The result may include
31310 the following fields:
31315 May have a value of either @samp{0}, when no tracing operations are
31316 supported, @samp{1}, when all tracing operations are supported, or
31317 @samp{file} when examining trace file. In the latter case, examining
31318 of trace frame is possible but new tracing experiement cannot be
31319 started. This field is always present.
31322 May have a value of either @samp{0} or @samp{1} depending on whether
31323 tracing experiement is in progress on target. This field is present
31324 if @samp{supported} field is not @samp{0}.
31327 Report the reason why the tracing was stopped last time. This field
31328 may be absent iff tracing was never stopped on target yet. The
31329 value of @samp{request} means the tracing was stopped as result of
31330 the @code{-trace-stop} command. The value of @samp{overflow} means
31331 the tracing buffer is full. The value of @samp{disconnection} means
31332 tracing was automatically stopped when @value{GDBN} has disconnected.
31333 The value of @samp{passcount} means tracing was stopped when a
31334 tracepoint was passed a maximal number of times for that tracepoint.
31335 This field is present if @samp{supported} field is not @samp{0}.
31337 @item stopping-tracepoint
31338 The number of tracepoint whose passcount as exceeded. This field is
31339 present iff the @samp{stop-reason} field has the value of
31343 @itemx frames-created
31344 The @samp{frames} field is a count of the total number of trace frames
31345 in the trace buffer, while @samp{frames-created} is the total created
31346 during the run, including ones that were discarded, such as when a
31347 circular trace buffer filled up. Both fields are optional.
31351 These fields tell the current size of the tracing buffer and the
31352 remaining space. These fields are optional.
31355 The value of the circular trace buffer flag. @code{1} means that the
31356 trace buffer is circular and old trace frames will be discarded if
31357 necessary to make room, @code{0} means that the trace buffer is linear
31361 The value of the disconnected tracing flag. @code{1} means that
31362 tracing will continue after @value{GDBN} disconnects, @code{0} means
31363 that the trace run will stop.
31367 @subsubheading @value{GDBN} Command
31369 The corresponding @value{GDBN} command is @samp{tstatus}.
31371 @subheading -trace-stop
31372 @findex -trace-stop
31374 @subsubheading Synopsis
31380 Stops a tracing experiment. The result of this command has the same
31381 fields as @code{-trace-status}, except that the @samp{supported} and
31382 @samp{running} fields are not output.
31384 @subsubheading @value{GDBN} Command
31386 The corresponding @value{GDBN} command is @samp{tstop}.
31389 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31390 @node GDB/MI Symbol Query
31391 @section @sc{gdb/mi} Symbol Query Commands
31395 @subheading The @code{-symbol-info-address} Command
31396 @findex -symbol-info-address
31398 @subsubheading Synopsis
31401 -symbol-info-address @var{symbol}
31404 Describe where @var{symbol} is stored.
31406 @subsubheading @value{GDBN} Command
31408 The corresponding @value{GDBN} command is @samp{info address}.
31410 @subsubheading Example
31414 @subheading The @code{-symbol-info-file} Command
31415 @findex -symbol-info-file
31417 @subsubheading Synopsis
31423 Show the file for the symbol.
31425 @subsubheading @value{GDBN} Command
31427 There's no equivalent @value{GDBN} command. @code{gdbtk} has
31428 @samp{gdb_find_file}.
31430 @subsubheading Example
31434 @subheading The @code{-symbol-info-function} Command
31435 @findex -symbol-info-function
31437 @subsubheading Synopsis
31440 -symbol-info-function
31443 Show which function the symbol lives in.
31445 @subsubheading @value{GDBN} Command
31447 @samp{gdb_get_function} in @code{gdbtk}.
31449 @subsubheading Example
31453 @subheading The @code{-symbol-info-line} Command
31454 @findex -symbol-info-line
31456 @subsubheading Synopsis
31462 Show the core addresses of the code for a source line.
31464 @subsubheading @value{GDBN} Command
31466 The corresponding @value{GDBN} command is @samp{info line}.
31467 @code{gdbtk} has the @samp{gdb_get_line} and @samp{gdb_get_file} commands.
31469 @subsubheading Example
31473 @subheading The @code{-symbol-info-symbol} Command
31474 @findex -symbol-info-symbol
31476 @subsubheading Synopsis
31479 -symbol-info-symbol @var{addr}
31482 Describe what symbol is at location @var{addr}.
31484 @subsubheading @value{GDBN} Command
31486 The corresponding @value{GDBN} command is @samp{info symbol}.
31488 @subsubheading Example
31492 @subheading The @code{-symbol-list-functions} Command
31493 @findex -symbol-list-functions
31495 @subsubheading Synopsis
31498 -symbol-list-functions
31501 List the functions in the executable.
31503 @subsubheading @value{GDBN} Command
31505 @samp{info functions} in @value{GDBN}, @samp{gdb_listfunc} and
31506 @samp{gdb_search} in @code{gdbtk}.
31508 @subsubheading Example
31513 @subheading The @code{-symbol-list-lines} Command
31514 @findex -symbol-list-lines
31516 @subsubheading Synopsis
31519 -symbol-list-lines @var{filename}
31522 Print the list of lines that contain code and their associated program
31523 addresses for the given source filename. The entries are sorted in
31524 ascending PC order.
31526 @subsubheading @value{GDBN} Command
31528 There is no corresponding @value{GDBN} command.
31530 @subsubheading Example
31533 -symbol-list-lines basics.c
31534 ^done,lines=[@{pc="0x08048554",line="7"@},@{pc="0x0804855a",line="8"@}]
31540 @subheading The @code{-symbol-list-types} Command
31541 @findex -symbol-list-types
31543 @subsubheading Synopsis
31549 List all the type names.
31551 @subsubheading @value{GDBN} Command
31553 The corresponding commands are @samp{info types} in @value{GDBN},
31554 @samp{gdb_search} in @code{gdbtk}.
31556 @subsubheading Example
31560 @subheading The @code{-symbol-list-variables} Command
31561 @findex -symbol-list-variables
31563 @subsubheading Synopsis
31566 -symbol-list-variables
31569 List all the global and static variable names.
31571 @subsubheading @value{GDBN} Command
31573 @samp{info variables} in @value{GDBN}, @samp{gdb_search} in @code{gdbtk}.
31575 @subsubheading Example
31579 @subheading The @code{-symbol-locate} Command
31580 @findex -symbol-locate
31582 @subsubheading Synopsis
31588 @subsubheading @value{GDBN} Command
31590 @samp{gdb_loc} in @code{gdbtk}.
31592 @subsubheading Example
31596 @subheading The @code{-symbol-type} Command
31597 @findex -symbol-type
31599 @subsubheading Synopsis
31602 -symbol-type @var{variable}
31605 Show type of @var{variable}.
31607 @subsubheading @value{GDBN} Command
31609 The corresponding @value{GDBN} command is @samp{ptype}, @code{gdbtk} has
31610 @samp{gdb_obj_variable}.
31612 @subsubheading Example
31617 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31618 @node GDB/MI File Commands
31619 @section @sc{gdb/mi} File Commands
31621 This section describes the GDB/MI commands to specify executable file names
31622 and to read in and obtain symbol table information.
31624 @subheading The @code{-file-exec-and-symbols} Command
31625 @findex -file-exec-and-symbols
31627 @subsubheading Synopsis
31630 -file-exec-and-symbols @var{file}
31633 Specify the executable file to be debugged. This file is the one from
31634 which the symbol table is also read. If no file is specified, the
31635 command clears the executable and symbol information. If breakpoints
31636 are set when using this command with no arguments, @value{GDBN} will produce
31637 error messages. Otherwise, no output is produced, except a completion
31640 @subsubheading @value{GDBN} Command
31642 The corresponding @value{GDBN} command is @samp{file}.
31644 @subsubheading Example
31648 -file-exec-and-symbols /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
31654 @subheading The @code{-file-exec-file} Command
31655 @findex -file-exec-file
31657 @subsubheading Synopsis
31660 -file-exec-file @var{file}
31663 Specify the executable file to be debugged. Unlike
31664 @samp{-file-exec-and-symbols}, the symbol table is @emph{not} read
31665 from this file. If used without argument, @value{GDBN} clears the information
31666 about the executable file. No output is produced, except a completion
31669 @subsubheading @value{GDBN} Command
31671 The corresponding @value{GDBN} command is @samp{exec-file}.
31673 @subsubheading Example
31677 -file-exec-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
31684 @subheading The @code{-file-list-exec-sections} Command
31685 @findex -file-list-exec-sections
31687 @subsubheading Synopsis
31690 -file-list-exec-sections
31693 List the sections of the current executable file.
31695 @subsubheading @value{GDBN} Command
31697 The @value{GDBN} command @samp{info file} shows, among the rest, the same
31698 information as this command. @code{gdbtk} has a corresponding command
31699 @samp{gdb_load_info}.
31701 @subsubheading Example
31706 @subheading The @code{-file-list-exec-source-file} Command
31707 @findex -file-list-exec-source-file
31709 @subsubheading Synopsis
31712 -file-list-exec-source-file
31715 List the line number, the current source file, and the absolute path
31716 to the current source file for the current executable. The macro
31717 information field has a value of @samp{1} or @samp{0} depending on
31718 whether or not the file includes preprocessor macro information.
31720 @subsubheading @value{GDBN} Command
31722 The @value{GDBN} equivalent is @samp{info source}
31724 @subsubheading Example
31728 123-file-list-exec-source-file
31729 123^done,line="1",file="foo.c",fullname="/home/bar/foo.c,macro-info="1"
31734 @subheading The @code{-file-list-exec-source-files} Command
31735 @findex -file-list-exec-source-files
31737 @subsubheading Synopsis
31740 -file-list-exec-source-files
31743 List the source files for the current executable.
31745 It will always output the filename, but only when @value{GDBN} can find
31746 the absolute file name of a source file, will it output the fullname.
31748 @subsubheading @value{GDBN} Command
31750 The @value{GDBN} equivalent is @samp{info sources}.
31751 @code{gdbtk} has an analogous command @samp{gdb_listfiles}.
31753 @subsubheading Example
31756 -file-list-exec-source-files
31758 @{file=foo.c,fullname=/home/foo.c@},
31759 @{file=/home/bar.c,fullname=/home/bar.c@},
31760 @{file=gdb_could_not_find_fullpath.c@}]
31765 @subheading The @code{-file-list-shared-libraries} Command
31766 @findex -file-list-shared-libraries
31768 @subsubheading Synopsis
31771 -file-list-shared-libraries
31774 List the shared libraries in the program.
31776 @subsubheading @value{GDBN} Command
31778 The corresponding @value{GDBN} command is @samp{info shared}.
31780 @subsubheading Example
31784 @subheading The @code{-file-list-symbol-files} Command
31785 @findex -file-list-symbol-files
31787 @subsubheading Synopsis
31790 -file-list-symbol-files
31795 @subsubheading @value{GDBN} Command
31797 The corresponding @value{GDBN} command is @samp{info file} (part of it).
31799 @subsubheading Example
31804 @subheading The @code{-file-symbol-file} Command
31805 @findex -file-symbol-file
31807 @subsubheading Synopsis
31810 -file-symbol-file @var{file}
31813 Read symbol table info from the specified @var{file} argument. When
31814 used without arguments, clears @value{GDBN}'s symbol table info. No output is
31815 produced, except for a completion notification.
31817 @subsubheading @value{GDBN} Command
31819 The corresponding @value{GDBN} command is @samp{symbol-file}.
31821 @subsubheading Example
31825 -file-symbol-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
31831 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31832 @node GDB/MI Memory Overlay Commands
31833 @section @sc{gdb/mi} Memory Overlay Commands
31835 The memory overlay commands are not implemented.
31837 @c @subheading -overlay-auto
31839 @c @subheading -overlay-list-mapping-state
31841 @c @subheading -overlay-list-overlays
31843 @c @subheading -overlay-map
31845 @c @subheading -overlay-off
31847 @c @subheading -overlay-on
31849 @c @subheading -overlay-unmap
31851 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31852 @node GDB/MI Signal Handling Commands
31853 @section @sc{gdb/mi} Signal Handling Commands
31855 Signal handling commands are not implemented.
31857 @c @subheading -signal-handle
31859 @c @subheading -signal-list-handle-actions
31861 @c @subheading -signal-list-signal-types
31865 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31866 @node GDB/MI Target Manipulation
31867 @section @sc{gdb/mi} Target Manipulation Commands
31870 @subheading The @code{-target-attach} Command
31871 @findex -target-attach
31873 @subsubheading Synopsis
31876 -target-attach @var{pid} | @var{gid} | @var{file}
31879 Attach to a process @var{pid} or a file @var{file} outside of
31880 @value{GDBN}, or a thread group @var{gid}. If attaching to a thread
31881 group, the id previously returned by
31882 @samp{-list-thread-groups --available} must be used.
31884 @subsubheading @value{GDBN} Command
31886 The corresponding @value{GDBN} command is @samp{attach}.
31888 @subsubheading Example
31892 =thread-created,id="1"
31893 *stopped,thread-id="1",frame=@{addr="0xb7f7e410",func="bar",args=[]@}
31899 @subheading The @code{-target-compare-sections} Command
31900 @findex -target-compare-sections
31902 @subsubheading Synopsis
31905 -target-compare-sections [ @var{section} ]
31908 Compare data of section @var{section} on target to the exec file.
31909 Without the argument, all sections are compared.
31911 @subsubheading @value{GDBN} Command
31913 The @value{GDBN} equivalent is @samp{compare-sections}.
31915 @subsubheading Example
31920 @subheading The @code{-target-detach} Command
31921 @findex -target-detach
31923 @subsubheading Synopsis
31926 -target-detach [ @var{pid} | @var{gid} ]
31929 Detach from the remote target which normally resumes its execution.
31930 If either @var{pid} or @var{gid} is specified, detaches from either
31931 the specified process, or specified thread group. There's no output.
31933 @subsubheading @value{GDBN} Command
31935 The corresponding @value{GDBN} command is @samp{detach}.
31937 @subsubheading Example
31947 @subheading The @code{-target-disconnect} Command
31948 @findex -target-disconnect
31950 @subsubheading Synopsis
31956 Disconnect from the remote target. There's no output and the target is
31957 generally not resumed.
31959 @subsubheading @value{GDBN} Command
31961 The corresponding @value{GDBN} command is @samp{disconnect}.
31963 @subsubheading Example
31973 @subheading The @code{-target-download} Command
31974 @findex -target-download
31976 @subsubheading Synopsis
31982 Loads the executable onto the remote target.
31983 It prints out an update message every half second, which includes the fields:
31987 The name of the section.
31989 The size of what has been sent so far for that section.
31991 The size of the section.
31993 The total size of what was sent so far (the current and the previous sections).
31995 The size of the overall executable to download.
31999 Each message is sent as status record (@pxref{GDB/MI Output Syntax, ,
32000 @sc{gdb/mi} Output Syntax}).
32002 In addition, it prints the name and size of the sections, as they are
32003 downloaded. These messages include the following fields:
32007 The name of the section.
32009 The size of the section.
32011 The size of the overall executable to download.
32015 At the end, a summary is printed.
32017 @subsubheading @value{GDBN} Command
32019 The corresponding @value{GDBN} command is @samp{load}.
32021 @subsubheading Example
32023 Note: each status message appears on a single line. Here the messages
32024 have been broken down so that they can fit onto a page.
32029 +download,@{section=".text",section-size="6668",total-size="9880"@}
32030 +download,@{section=".text",section-sent="512",section-size="6668",
32031 total-sent="512",total-size="9880"@}
32032 +download,@{section=".text",section-sent="1024",section-size="6668",
32033 total-sent="1024",total-size="9880"@}
32034 +download,@{section=".text",section-sent="1536",section-size="6668",
32035 total-sent="1536",total-size="9880"@}
32036 +download,@{section=".text",section-sent="2048",section-size="6668",
32037 total-sent="2048",total-size="9880"@}
32038 +download,@{section=".text",section-sent="2560",section-size="6668",
32039 total-sent="2560",total-size="9880"@}
32040 +download,@{section=".text",section-sent="3072",section-size="6668",
32041 total-sent="3072",total-size="9880"@}
32042 +download,@{section=".text",section-sent="3584",section-size="6668",
32043 total-sent="3584",total-size="9880"@}
32044 +download,@{section=".text",section-sent="4096",section-size="6668",
32045 total-sent="4096",total-size="9880"@}
32046 +download,@{section=".text",section-sent="4608",section-size="6668",
32047 total-sent="4608",total-size="9880"@}
32048 +download,@{section=".text",section-sent="5120",section-size="6668",
32049 total-sent="5120",total-size="9880"@}
32050 +download,@{section=".text",section-sent="5632",section-size="6668",
32051 total-sent="5632",total-size="9880"@}
32052 +download,@{section=".text",section-sent="6144",section-size="6668",
32053 total-sent="6144",total-size="9880"@}
32054 +download,@{section=".text",section-sent="6656",section-size="6668",
32055 total-sent="6656",total-size="9880"@}
32056 +download,@{section=".init",section-size="28",total-size="9880"@}
32057 +download,@{section=".fini",section-size="28",total-size="9880"@}
32058 +download,@{section=".data",section-size="3156",total-size="9880"@}
32059 +download,@{section=".data",section-sent="512",section-size="3156",
32060 total-sent="7236",total-size="9880"@}
32061 +download,@{section=".data",section-sent="1024",section-size="3156",
32062 total-sent="7748",total-size="9880"@}
32063 +download,@{section=".data",section-sent="1536",section-size="3156",
32064 total-sent="8260",total-size="9880"@}
32065 +download,@{section=".data",section-sent="2048",section-size="3156",
32066 total-sent="8772",total-size="9880"@}
32067 +download,@{section=".data",section-sent="2560",section-size="3156",
32068 total-sent="9284",total-size="9880"@}
32069 +download,@{section=".data",section-sent="3072",section-size="3156",
32070 total-sent="9796",total-size="9880"@}
32071 ^done,address="0x10004",load-size="9880",transfer-rate="6586",
32078 @subheading The @code{-target-exec-status} Command
32079 @findex -target-exec-status
32081 @subsubheading Synopsis
32084 -target-exec-status
32087 Provide information on the state of the target (whether it is running or
32088 not, for instance).
32090 @subsubheading @value{GDBN} Command
32092 There's no equivalent @value{GDBN} command.
32094 @subsubheading Example
32098 @subheading The @code{-target-list-available-targets} Command
32099 @findex -target-list-available-targets
32101 @subsubheading Synopsis
32104 -target-list-available-targets
32107 List the possible targets to connect to.
32109 @subsubheading @value{GDBN} Command
32111 The corresponding @value{GDBN} command is @samp{help target}.
32113 @subsubheading Example
32117 @subheading The @code{-target-list-current-targets} Command
32118 @findex -target-list-current-targets
32120 @subsubheading Synopsis
32123 -target-list-current-targets
32126 Describe the current target.
32128 @subsubheading @value{GDBN} Command
32130 The corresponding information is printed by @samp{info file} (among
32133 @subsubheading Example
32137 @subheading The @code{-target-list-parameters} Command
32138 @findex -target-list-parameters
32140 @subsubheading Synopsis
32143 -target-list-parameters
32149 @subsubheading @value{GDBN} Command
32153 @subsubheading Example
32157 @subheading The @code{-target-select} Command
32158 @findex -target-select
32160 @subsubheading Synopsis
32163 -target-select @var{type} @var{parameters @dots{}}
32166 Connect @value{GDBN} to the remote target. This command takes two args:
32170 The type of target, for instance @samp{remote}, etc.
32171 @item @var{parameters}
32172 Device names, host names and the like. @xref{Target Commands, ,
32173 Commands for Managing Targets}, for more details.
32176 The output is a connection notification, followed by the address at
32177 which the target program is, in the following form:
32180 ^connected,addr="@var{address}",func="@var{function name}",
32181 args=[@var{arg list}]
32184 @subsubheading @value{GDBN} Command
32186 The corresponding @value{GDBN} command is @samp{target}.
32188 @subsubheading Example
32192 -target-select remote /dev/ttya
32193 ^connected,addr="0xfe00a300",func="??",args=[]
32197 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
32198 @node GDB/MI File Transfer Commands
32199 @section @sc{gdb/mi} File Transfer Commands
32202 @subheading The @code{-target-file-put} Command
32203 @findex -target-file-put
32205 @subsubheading Synopsis
32208 -target-file-put @var{hostfile} @var{targetfile}
32211 Copy file @var{hostfile} from the host system (the machine running
32212 @value{GDBN}) to @var{targetfile} on the target system.
32214 @subsubheading @value{GDBN} Command
32216 The corresponding @value{GDBN} command is @samp{remote put}.
32218 @subsubheading Example
32222 -target-file-put localfile remotefile
32228 @subheading The @code{-target-file-get} Command
32229 @findex -target-file-get
32231 @subsubheading Synopsis
32234 -target-file-get @var{targetfile} @var{hostfile}
32237 Copy file @var{targetfile} from the target system to @var{hostfile}
32238 on the host system.
32240 @subsubheading @value{GDBN} Command
32242 The corresponding @value{GDBN} command is @samp{remote get}.
32244 @subsubheading Example
32248 -target-file-get remotefile localfile
32254 @subheading The @code{-target-file-delete} Command
32255 @findex -target-file-delete
32257 @subsubheading Synopsis
32260 -target-file-delete @var{targetfile}
32263 Delete @var{targetfile} from the target system.
32265 @subsubheading @value{GDBN} Command
32267 The corresponding @value{GDBN} command is @samp{remote delete}.
32269 @subsubheading Example
32273 -target-file-delete remotefile
32279 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
32280 @node GDB/MI Miscellaneous Commands
32281 @section Miscellaneous @sc{gdb/mi} Commands
32283 @c @subheading -gdb-complete
32285 @subheading The @code{-gdb-exit} Command
32288 @subsubheading Synopsis
32294 Exit @value{GDBN} immediately.
32296 @subsubheading @value{GDBN} Command
32298 Approximately corresponds to @samp{quit}.
32300 @subsubheading Example
32310 @subheading The @code{-exec-abort} Command
32311 @findex -exec-abort
32313 @subsubheading Synopsis
32319 Kill the inferior running program.
32321 @subsubheading @value{GDBN} Command
32323 The corresponding @value{GDBN} command is @samp{kill}.
32325 @subsubheading Example
32330 @subheading The @code{-gdb-set} Command
32333 @subsubheading Synopsis
32339 Set an internal @value{GDBN} variable.
32340 @c IS THIS A DOLLAR VARIABLE? OR SOMETHING LIKE ANNOTATE ?????
32342 @subsubheading @value{GDBN} Command
32344 The corresponding @value{GDBN} command is @samp{set}.
32346 @subsubheading Example
32356 @subheading The @code{-gdb-show} Command
32359 @subsubheading Synopsis
32365 Show the current value of a @value{GDBN} variable.
32367 @subsubheading @value{GDBN} Command
32369 The corresponding @value{GDBN} command is @samp{show}.
32371 @subsubheading Example
32380 @c @subheading -gdb-source
32383 @subheading The @code{-gdb-version} Command
32384 @findex -gdb-version
32386 @subsubheading Synopsis
32392 Show version information for @value{GDBN}. Used mostly in testing.
32394 @subsubheading @value{GDBN} Command
32396 The @value{GDBN} equivalent is @samp{show version}. @value{GDBN} by
32397 default shows this information when you start an interactive session.
32399 @subsubheading Example
32401 @c This example modifies the actual output from GDB to avoid overfull
32407 ~Copyright 2000 Free Software Foundation, Inc.
32408 ~GDB is free software, covered by the GNU General Public License, and
32409 ~you are welcome to change it and/or distribute copies of it under
32410 ~ certain conditions.
32411 ~Type "show copying" to see the conditions.
32412 ~There is absolutely no warranty for GDB. Type "show warranty" for
32414 ~This GDB was configured as
32415 "--host=sparc-sun-solaris2.5.1 --target=ppc-eabi".
32420 @subheading The @code{-list-features} Command
32421 @findex -list-features
32423 Returns a list of particular features of the MI protocol that
32424 this version of gdb implements. A feature can be a command,
32425 or a new field in an output of some command, or even an
32426 important bugfix. While a frontend can sometimes detect presence
32427 of a feature at runtime, it is easier to perform detection at debugger
32430 The command returns a list of strings, with each string naming an
32431 available feature. Each returned string is just a name, it does not
32432 have any internal structure. The list of possible feature names
32438 (gdb) -list-features
32439 ^done,result=["feature1","feature2"]
32442 The current list of features is:
32445 @item frozen-varobjs
32446 Indicates support for the @code{-var-set-frozen} command, as well
32447 as possible presense of the @code{frozen} field in the output
32448 of @code{-varobj-create}.
32449 @item pending-breakpoints
32450 Indicates support for the @option{-f} option to the @code{-break-insert}
32453 Indicates Python scripting support, Python-based
32454 pretty-printing commands, and possible presence of the
32455 @samp{display_hint} field in the output of @code{-var-list-children}
32457 Indicates support for the @code{-thread-info} command.
32458 @item data-read-memory-bytes
32459 Indicates support for the @code{-data-read-memory-bytes} and the
32460 @code{-data-write-memory-bytes} commands.
32461 @item breakpoint-notifications
32462 Indicates that changes to breakpoints and breakpoints created via the
32463 CLI will be announced via async records.
32464 @item ada-task-info
32465 Indicates support for the @code{-ada-task-info} command.
32468 @subheading The @code{-list-target-features} Command
32469 @findex -list-target-features
32471 Returns a list of particular features that are supported by the
32472 target. Those features affect the permitted MI commands, but
32473 unlike the features reported by the @code{-list-features} command, the
32474 features depend on which target GDB is using at the moment. Whenever
32475 a target can change, due to commands such as @code{-target-select},
32476 @code{-target-attach} or @code{-exec-run}, the list of target features
32477 may change, and the frontend should obtain it again.
32481 (gdb) -list-features
32482 ^done,result=["async"]
32485 The current list of features is:
32489 Indicates that the target is capable of asynchronous command
32490 execution, which means that @value{GDBN} will accept further commands
32491 while the target is running.
32494 Indicates that the target is capable of reverse execution.
32495 @xref{Reverse Execution}, for more information.
32499 @subheading The @code{-list-thread-groups} Command
32500 @findex -list-thread-groups
32502 @subheading Synopsis
32505 -list-thread-groups [ --available ] [ --recurse 1 ] [ @var{group} ... ]
32508 Lists thread groups (@pxref{Thread groups}). When a single thread
32509 group is passed as the argument, lists the children of that group.
32510 When several thread group are passed, lists information about those
32511 thread groups. Without any parameters, lists information about all
32512 top-level thread groups.
32514 Normally, thread groups that are being debugged are reported.
32515 With the @samp{--available} option, @value{GDBN} reports thread groups
32516 available on the target.
32518 The output of this command may have either a @samp{threads} result or
32519 a @samp{groups} result. The @samp{thread} result has a list of tuples
32520 as value, with each tuple describing a thread (@pxref{GDB/MI Thread
32521 Information}). The @samp{groups} result has a list of tuples as value,
32522 each tuple describing a thread group. If top-level groups are
32523 requested (that is, no parameter is passed), or when several groups
32524 are passed, the output always has a @samp{groups} result. The format
32525 of the @samp{group} result is described below.
32527 To reduce the number of roundtrips it's possible to list thread groups
32528 together with their children, by passing the @samp{--recurse} option
32529 and the recursion depth. Presently, only recursion depth of 1 is
32530 permitted. If this option is present, then every reported thread group
32531 will also include its children, either as @samp{group} or
32532 @samp{threads} field.
32534 In general, any combination of option and parameters is permitted, with
32535 the following caveats:
32539 When a single thread group is passed, the output will typically
32540 be the @samp{threads} result. Because threads may not contain
32541 anything, the @samp{recurse} option will be ignored.
32544 When the @samp{--available} option is passed, limited information may
32545 be available. In particular, the list of threads of a process might
32546 be inaccessible. Further, specifying specific thread groups might
32547 not give any performance advantage over listing all thread groups.
32548 The frontend should assume that @samp{-list-thread-groups --available}
32549 is always an expensive operation and cache the results.
32553 The @samp{groups} result is a list of tuples, where each tuple may
32554 have the following fields:
32558 Identifier of the thread group. This field is always present.
32559 The identifier is an opaque string; frontends should not try to
32560 convert it to an integer, even though it might look like one.
32563 The type of the thread group. At present, only @samp{process} is a
32567 The target-specific process identifier. This field is only present
32568 for thread groups of type @samp{process} and only if the process exists.
32571 The number of children this thread group has. This field may be
32572 absent for an available thread group.
32575 This field has a list of tuples as value, each tuple describing a
32576 thread. It may be present if the @samp{--recurse} option is
32577 specified, and it's actually possible to obtain the threads.
32580 This field is a list of integers, each identifying a core that one
32581 thread of the group is running on. This field may be absent if
32582 such information is not available.
32585 The name of the executable file that corresponds to this thread group.
32586 The field is only present for thread groups of type @samp{process},
32587 and only if there is a corresponding executable file.
32591 @subheading Example
32595 -list-thread-groups
32596 ^done,groups=[@{id="17",type="process",pid="yyy",num_children="2"@}]
32597 -list-thread-groups 17
32598 ^done,threads=[@{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
32599 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",args=[]@},state="running"@},
32600 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
32601 frame=@{level="0",addr="0x0804891f",func="foo",args=[@{name="i",value="10"@}],
32602 file="/tmp/a.c",fullname="/tmp/a.c",line="158"@},state="running"@}]]
32603 -list-thread-groups --available
32604 ^done,groups=[@{id="17",type="process",pid="yyy",num_children="2",cores=[1,2]@}]
32605 -list-thread-groups --available --recurse 1
32606 ^done,groups=[@{id="17", types="process",pid="yyy",num_children="2",cores=[1,2],
32607 threads=[@{id="1",target-id="Thread 0xb7e14b90",cores=[1]@},
32608 @{id="2",target-id="Thread 0xb7e14b90",cores=[2]@}]@},..]
32609 -list-thread-groups --available --recurse 1 17 18
32610 ^done,groups=[@{id="17", types="process",pid="yyy",num_children="2",cores=[1,2],
32611 threads=[@{id="1",target-id="Thread 0xb7e14b90",cores=[1]@},
32612 @{id="2",target-id="Thread 0xb7e14b90",cores=[2]@}]@},...]
32615 @subheading The @code{-info-os} Command
32618 @subsubheading Synopsis
32621 -info-os [ @var{type} ]
32624 If no argument is supplied, the command returns a table of available
32625 operating-system-specific information types. If one of these types is
32626 supplied as an argument @var{type}, then the command returns a table
32627 of data of that type.
32629 The types of information available depend on the target operating
32632 @subsubheading @value{GDBN} Command
32634 The corresponding @value{GDBN} command is @samp{info os}.
32636 @subsubheading Example
32638 When run on a @sc{gnu}/Linux system, the output will look something
32644 ^done,OSDataTable=@{nr_rows="9",nr_cols="2",
32645 hdr=[@{width="10",alignment="-1",col_name="col0",colhdr="Type"@},
32646 @{width="10",alignment="-1",col_name="col1",colhdr="Description"@}],
32647 body=[item=@{col0="processes",col1="Listing of all processes"@},
32648 item=@{col0="procgroups",col1="Listing of all process groups"@},
32649 item=@{col0="threads",col1="Listing of all threads"@},
32650 item=@{col0="files",col1="Listing of all file descriptors"@},
32651 item=@{col0="sockets",col1="Listing of all internet-domain sockets"@},
32652 item=@{col0="shm",col1="Listing of all shared-memory regions"@},
32653 item=@{col0="semaphores",col1="Listing of all semaphores"@},
32654 item=@{col0="msg",col1="Listing of all message queues"@},
32655 item=@{col0="modules",col1="Listing of all loaded kernel modules"@}]@}
32658 ^done,OSDataTable=@{nr_rows="190",nr_cols="4",
32659 hdr=[@{width="10",alignment="-1",col_name="col0",colhdr="pid"@},
32660 @{width="10",alignment="-1",col_name="col1",colhdr="user"@},
32661 @{width="10",alignment="-1",col_name="col2",colhdr="command"@},
32662 @{width="10",alignment="-1",col_name="col3",colhdr="cores"@}],
32663 body=[item=@{col0="1",col1="root",col2="/sbin/init",col3="0"@},
32664 item=@{col0="2",col1="root",col2="[kthreadd]",col3="1"@},
32665 item=@{col0="3",col1="root",col2="[ksoftirqd/0]",col3="0"@},
32667 item=@{col0="26446",col1="stan",col2="bash",col3="0"@},
32668 item=@{col0="28152",col1="stan",col2="bash",col3="1"@}]@}
32672 @subheading The @code{-add-inferior} Command
32673 @findex -add-inferior
32675 @subheading Synopsis
32681 Creates a new inferior (@pxref{Inferiors and Programs}). The created
32682 inferior is not associated with any executable. Such association may
32683 be established with the @samp{-file-exec-and-symbols} command
32684 (@pxref{GDB/MI File Commands}). The command response has a single
32685 field, @samp{thread-group}, whose value is the identifier of the
32686 thread group corresponding to the new inferior.
32688 @subheading Example
32693 ^done,thread-group="i3"
32696 @subheading The @code{-interpreter-exec} Command
32697 @findex -interpreter-exec
32699 @subheading Synopsis
32702 -interpreter-exec @var{interpreter} @var{command}
32704 @anchor{-interpreter-exec}
32706 Execute the specified @var{command} in the given @var{interpreter}.
32708 @subheading @value{GDBN} Command
32710 The corresponding @value{GDBN} command is @samp{interpreter-exec}.
32712 @subheading Example
32716 -interpreter-exec console "break main"
32717 &"During symbol reading, couldn't parse type; debugger out of date?.\n"
32718 &"During symbol reading, bad structure-type format.\n"
32719 ~"Breakpoint 1 at 0x8074fc6: file ../../src/gdb/main.c, line 743.\n"
32724 @subheading The @code{-inferior-tty-set} Command
32725 @findex -inferior-tty-set
32727 @subheading Synopsis
32730 -inferior-tty-set /dev/pts/1
32733 Set terminal for future runs of the program being debugged.
32735 @subheading @value{GDBN} Command
32737 The corresponding @value{GDBN} command is @samp{set inferior-tty} /dev/pts/1.
32739 @subheading Example
32743 -inferior-tty-set /dev/pts/1
32748 @subheading The @code{-inferior-tty-show} Command
32749 @findex -inferior-tty-show
32751 @subheading Synopsis
32757 Show terminal for future runs of program being debugged.
32759 @subheading @value{GDBN} Command
32761 The corresponding @value{GDBN} command is @samp{show inferior-tty}.
32763 @subheading Example
32767 -inferior-tty-set /dev/pts/1
32771 ^done,inferior_tty_terminal="/dev/pts/1"
32775 @subheading The @code{-enable-timings} Command
32776 @findex -enable-timings
32778 @subheading Synopsis
32781 -enable-timings [yes | no]
32784 Toggle the printing of the wallclock, user and system times for an MI
32785 command as a field in its output. This command is to help frontend
32786 developers optimize the performance of their code. No argument is
32787 equivalent to @samp{yes}.
32789 @subheading @value{GDBN} Command
32793 @subheading Example
32801 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
32802 addr="0x080484ed",func="main",file="myprog.c",
32803 fullname="/home/nickrob/myprog.c",line="73",times="0"@},
32804 time=@{wallclock="0.05185",user="0.00800",system="0.00000"@}
32812 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
32813 frame=@{addr="0x080484ed",func="main",args=[@{name="argc",value="1"@},
32814 @{name="argv",value="0xbfb60364"@}],file="myprog.c",
32815 fullname="/home/nickrob/myprog.c",line="73"@}
32820 @chapter @value{GDBN} Annotations
32822 This chapter describes annotations in @value{GDBN}. Annotations were
32823 designed to interface @value{GDBN} to graphical user interfaces or other
32824 similar programs which want to interact with @value{GDBN} at a
32825 relatively high level.
32827 The annotation mechanism has largely been superseded by @sc{gdb/mi}
32831 This is Edition @value{EDITION}, @value{DATE}.
32835 * Annotations Overview:: What annotations are; the general syntax.
32836 * Server Prefix:: Issuing a command without affecting user state.
32837 * Prompting:: Annotations marking @value{GDBN}'s need for input.
32838 * Errors:: Annotations for error messages.
32839 * Invalidation:: Some annotations describe things now invalid.
32840 * Annotations for Running::
32841 Whether the program is running, how it stopped, etc.
32842 * Source Annotations:: Annotations describing source code.
32845 @node Annotations Overview
32846 @section What is an Annotation?
32847 @cindex annotations
32849 Annotations start with a newline character, two @samp{control-z}
32850 characters, and the name of the annotation. If there is no additional
32851 information associated with this annotation, the name of the annotation
32852 is followed immediately by a newline. If there is additional
32853 information, the name of the annotation is followed by a space, the
32854 additional information, and a newline. The additional information
32855 cannot contain newline characters.
32857 Any output not beginning with a newline and two @samp{control-z}
32858 characters denotes literal output from @value{GDBN}. Currently there is
32859 no need for @value{GDBN} to output a newline followed by two
32860 @samp{control-z} characters, but if there was such a need, the
32861 annotations could be extended with an @samp{escape} annotation which
32862 means those three characters as output.
32864 The annotation @var{level}, which is specified using the
32865 @option{--annotate} command line option (@pxref{Mode Options}), controls
32866 how much information @value{GDBN} prints together with its prompt,
32867 values of expressions, source lines, and other types of output. Level 0
32868 is for no annotations, level 1 is for use when @value{GDBN} is run as a
32869 subprocess of @sc{gnu} Emacs, level 3 is the maximum annotation suitable
32870 for programs that control @value{GDBN}, and level 2 annotations have
32871 been made obsolete (@pxref{Limitations, , Limitations of the Annotation
32872 Interface, annotate, GDB's Obsolete Annotations}).
32875 @kindex set annotate
32876 @item set annotate @var{level}
32877 The @value{GDBN} command @code{set annotate} sets the level of
32878 annotations to the specified @var{level}.
32880 @item show annotate
32881 @kindex show annotate
32882 Show the current annotation level.
32885 This chapter describes level 3 annotations.
32887 A simple example of starting up @value{GDBN} with annotations is:
32890 $ @kbd{gdb --annotate=3}
32892 Copyright 2003 Free Software Foundation, Inc.
32893 GDB is free software, covered by the GNU General Public License,
32894 and you are welcome to change it and/or distribute copies of it
32895 under certain conditions.
32896 Type "show copying" to see the conditions.
32897 There is absolutely no warranty for GDB. Type "show warranty"
32899 This GDB was configured as "i386-pc-linux-gnu"
32910 Here @samp{quit} is input to @value{GDBN}; the rest is output from
32911 @value{GDBN}. The three lines beginning @samp{^Z^Z} (where @samp{^Z}
32912 denotes a @samp{control-z} character) are annotations; the rest is
32913 output from @value{GDBN}.
32915 @node Server Prefix
32916 @section The Server Prefix
32917 @cindex server prefix
32919 If you prefix a command with @samp{server } then it will not affect
32920 the command history, nor will it affect @value{GDBN}'s notion of which
32921 command to repeat if @key{RET} is pressed on a line by itself. This
32922 means that commands can be run behind a user's back by a front-end in
32923 a transparent manner.
32925 The @code{server } prefix does not affect the recording of values into
32926 the value history; to print a value without recording it into the
32927 value history, use the @code{output} command instead of the
32928 @code{print} command.
32930 Using this prefix also disables confirmation requests
32931 (@pxref{confirmation requests}).
32934 @section Annotation for @value{GDBN} Input
32936 @cindex annotations for prompts
32937 When @value{GDBN} prompts for input, it annotates this fact so it is possible
32938 to know when to send output, when the output from a given command is
32941 Different kinds of input each have a different @dfn{input type}. Each
32942 input type has three annotations: a @code{pre-} annotation, which
32943 denotes the beginning of any prompt which is being output, a plain
32944 annotation, which denotes the end of the prompt, and then a @code{post-}
32945 annotation which denotes the end of any echo which may (or may not) be
32946 associated with the input. For example, the @code{prompt} input type
32947 features the following annotations:
32955 The input types are
32958 @findex pre-prompt annotation
32959 @findex prompt annotation
32960 @findex post-prompt annotation
32962 When @value{GDBN} is prompting for a command (the main @value{GDBN} prompt).
32964 @findex pre-commands annotation
32965 @findex commands annotation
32966 @findex post-commands annotation
32968 When @value{GDBN} prompts for a set of commands, like in the @code{commands}
32969 command. The annotations are repeated for each command which is input.
32971 @findex pre-overload-choice annotation
32972 @findex overload-choice annotation
32973 @findex post-overload-choice annotation
32974 @item overload-choice
32975 When @value{GDBN} wants the user to select between various overloaded functions.
32977 @findex pre-query annotation
32978 @findex query annotation
32979 @findex post-query annotation
32981 When @value{GDBN} wants the user to confirm a potentially dangerous operation.
32983 @findex pre-prompt-for-continue annotation
32984 @findex prompt-for-continue annotation
32985 @findex post-prompt-for-continue annotation
32986 @item prompt-for-continue
32987 When @value{GDBN} is asking the user to press return to continue. Note: Don't
32988 expect this to work well; instead use @code{set height 0} to disable
32989 prompting. This is because the counting of lines is buggy in the
32990 presence of annotations.
32995 @cindex annotations for errors, warnings and interrupts
32997 @findex quit annotation
33002 This annotation occurs right before @value{GDBN} responds to an interrupt.
33004 @findex error annotation
33009 This annotation occurs right before @value{GDBN} responds to an error.
33011 Quit and error annotations indicate that any annotations which @value{GDBN} was
33012 in the middle of may end abruptly. For example, if a
33013 @code{value-history-begin} annotation is followed by a @code{error}, one
33014 cannot expect to receive the matching @code{value-history-end}. One
33015 cannot expect not to receive it either, however; an error annotation
33016 does not necessarily mean that @value{GDBN} is immediately returning all the way
33019 @findex error-begin annotation
33020 A quit or error annotation may be preceded by
33026 Any output between that and the quit or error annotation is the error
33029 Warning messages are not yet annotated.
33030 @c If we want to change that, need to fix warning(), type_error(),
33031 @c range_error(), and possibly other places.
33034 @section Invalidation Notices
33036 @cindex annotations for invalidation messages
33037 The following annotations say that certain pieces of state may have
33041 @findex frames-invalid annotation
33042 @item ^Z^Zframes-invalid
33044 The frames (for example, output from the @code{backtrace} command) may
33047 @findex breakpoints-invalid annotation
33048 @item ^Z^Zbreakpoints-invalid
33050 The breakpoints may have changed. For example, the user just added or
33051 deleted a breakpoint.
33054 @node Annotations for Running
33055 @section Running the Program
33056 @cindex annotations for running programs
33058 @findex starting annotation
33059 @findex stopping annotation
33060 When the program starts executing due to a @value{GDBN} command such as
33061 @code{step} or @code{continue},
33067 is output. When the program stops,
33073 is output. Before the @code{stopped} annotation, a variety of
33074 annotations describe how the program stopped.
33077 @findex exited annotation
33078 @item ^Z^Zexited @var{exit-status}
33079 The program exited, and @var{exit-status} is the exit status (zero for
33080 successful exit, otherwise nonzero).
33082 @findex signalled annotation
33083 @findex signal-name annotation
33084 @findex signal-name-end annotation
33085 @findex signal-string annotation
33086 @findex signal-string-end annotation
33087 @item ^Z^Zsignalled
33088 The program exited with a signal. After the @code{^Z^Zsignalled}, the
33089 annotation continues:
33095 ^Z^Zsignal-name-end
33099 ^Z^Zsignal-string-end
33104 where @var{name} is the name of the signal, such as @code{SIGILL} or
33105 @code{SIGSEGV}, and @var{string} is the explanation of the signal, such
33106 as @code{Illegal Instruction} or @code{Segmentation fault}.
33107 @var{intro-text}, @var{middle-text}, and @var{end-text} are for the
33108 user's benefit and have no particular format.
33110 @findex signal annotation
33112 The syntax of this annotation is just like @code{signalled}, but @value{GDBN} is
33113 just saying that the program received the signal, not that it was
33114 terminated with it.
33116 @findex breakpoint annotation
33117 @item ^Z^Zbreakpoint @var{number}
33118 The program hit breakpoint number @var{number}.
33120 @findex watchpoint annotation
33121 @item ^Z^Zwatchpoint @var{number}
33122 The program hit watchpoint number @var{number}.
33125 @node Source Annotations
33126 @section Displaying Source
33127 @cindex annotations for source display
33129 @findex source annotation
33130 The following annotation is used instead of displaying source code:
33133 ^Z^Zsource @var{filename}:@var{line}:@var{character}:@var{middle}:@var{addr}
33136 where @var{filename} is an absolute file name indicating which source
33137 file, @var{line} is the line number within that file (where 1 is the
33138 first line in the file), @var{character} is the character position
33139 within the file (where 0 is the first character in the file) (for most
33140 debug formats this will necessarily point to the beginning of a line),
33141 @var{middle} is @samp{middle} if @var{addr} is in the middle of the
33142 line, or @samp{beg} if @var{addr} is at the beginning of the line, and
33143 @var{addr} is the address in the target program associated with the
33144 source which is being displayed. @var{addr} is in the form @samp{0x}
33145 followed by one or more lowercase hex digits (note that this does not
33146 depend on the language).
33148 @node JIT Interface
33149 @chapter JIT Compilation Interface
33150 @cindex just-in-time compilation
33151 @cindex JIT compilation interface
33153 This chapter documents @value{GDBN}'s @dfn{just-in-time} (JIT) compilation
33154 interface. A JIT compiler is a program or library that generates native
33155 executable code at runtime and executes it, usually in order to achieve good
33156 performance while maintaining platform independence.
33158 Programs that use JIT compilation are normally difficult to debug because
33159 portions of their code are generated at runtime, instead of being loaded from
33160 object files, which is where @value{GDBN} normally finds the program's symbols
33161 and debug information. In order to debug programs that use JIT compilation,
33162 @value{GDBN} has an interface that allows the program to register in-memory
33163 symbol files with @value{GDBN} at runtime.
33165 If you are using @value{GDBN} to debug a program that uses this interface, then
33166 it should work transparently so long as you have not stripped the binary. If
33167 you are developing a JIT compiler, then the interface is documented in the rest
33168 of this chapter. At this time, the only known client of this interface is the
33171 Broadly speaking, the JIT interface mirrors the dynamic loader interface. The
33172 JIT compiler communicates with @value{GDBN} by writing data into a global
33173 variable and calling a fuction at a well-known symbol. When @value{GDBN}
33174 attaches, it reads a linked list of symbol files from the global variable to
33175 find existing code, and puts a breakpoint in the function so that it can find
33176 out about additional code.
33179 * Declarations:: Relevant C struct declarations
33180 * Registering Code:: Steps to register code
33181 * Unregistering Code:: Steps to unregister code
33182 * Custom Debug Info:: Emit debug information in a custom format
33186 @section JIT Declarations
33188 These are the relevant struct declarations that a C program should include to
33189 implement the interface:
33199 struct jit_code_entry
33201 struct jit_code_entry *next_entry;
33202 struct jit_code_entry *prev_entry;
33203 const char *symfile_addr;
33204 uint64_t symfile_size;
33207 struct jit_descriptor
33210 /* This type should be jit_actions_t, but we use uint32_t
33211 to be explicit about the bitwidth. */
33212 uint32_t action_flag;
33213 struct jit_code_entry *relevant_entry;
33214 struct jit_code_entry *first_entry;
33217 /* GDB puts a breakpoint in this function. */
33218 void __attribute__((noinline)) __jit_debug_register_code() @{ @};
33220 /* Make sure to specify the version statically, because the
33221 debugger may check the version before we can set it. */
33222 struct jit_descriptor __jit_debug_descriptor = @{ 1, 0, 0, 0 @};
33225 If the JIT is multi-threaded, then it is important that the JIT synchronize any
33226 modifications to this global data properly, which can easily be done by putting
33227 a global mutex around modifications to these structures.
33229 @node Registering Code
33230 @section Registering Code
33232 To register code with @value{GDBN}, the JIT should follow this protocol:
33236 Generate an object file in memory with symbols and other desired debug
33237 information. The file must include the virtual addresses of the sections.
33240 Create a code entry for the file, which gives the start and size of the symbol
33244 Add it to the linked list in the JIT descriptor.
33247 Point the relevant_entry field of the descriptor at the entry.
33250 Set @code{action_flag} to @code{JIT_REGISTER} and call
33251 @code{__jit_debug_register_code}.
33254 When @value{GDBN} is attached and the breakpoint fires, @value{GDBN} uses the
33255 @code{relevant_entry} pointer so it doesn't have to walk the list looking for
33256 new code. However, the linked list must still be maintained in order to allow
33257 @value{GDBN} to attach to a running process and still find the symbol files.
33259 @node Unregistering Code
33260 @section Unregistering Code
33262 If code is freed, then the JIT should use the following protocol:
33266 Remove the code entry corresponding to the code from the linked list.
33269 Point the @code{relevant_entry} field of the descriptor at the code entry.
33272 Set @code{action_flag} to @code{JIT_UNREGISTER} and call
33273 @code{__jit_debug_register_code}.
33276 If the JIT frees or recompiles code without unregistering it, then @value{GDBN}
33277 and the JIT will leak the memory used for the associated symbol files.
33279 @node Custom Debug Info
33280 @section Custom Debug Info
33281 @cindex custom JIT debug info
33282 @cindex JIT debug info reader
33284 Generating debug information in platform-native file formats (like ELF
33285 or COFF) may be an overkill for JIT compilers; especially if all the
33286 debug info is used for is displaying a meaningful backtrace. The
33287 issue can be resolved by having the JIT writers decide on a debug info
33288 format and also provide a reader that parses the debug info generated
33289 by the JIT compiler. This section gives a brief overview on writing
33290 such a parser. More specific details can be found in the source file
33291 @file{gdb/jit-reader.in}, which is also installed as a header at
33292 @file{@var{includedir}/gdb/jit-reader.h} for easy inclusion.
33294 The reader is implemented as a shared object (so this functionality is
33295 not available on platforms which don't allow loading shared objects at
33296 runtime). Two @value{GDBN} commands, @code{jit-reader-load} and
33297 @code{jit-reader-unload} are provided, to be used to load and unload
33298 the readers from a preconfigured directory. Once loaded, the shared
33299 object is used the parse the debug information emitted by the JIT
33303 * Using JIT Debug Info Readers:: How to use supplied readers correctly
33304 * Writing JIT Debug Info Readers:: Creating a debug-info reader
33307 @node Using JIT Debug Info Readers
33308 @subsection Using JIT Debug Info Readers
33309 @kindex jit-reader-load
33310 @kindex jit-reader-unload
33312 Readers can be loaded and unloaded using the @code{jit-reader-load}
33313 and @code{jit-reader-unload} commands.
33316 @item jit-reader-load @var{reader-name}
33317 Load the JIT reader named @var{reader-name}. On a UNIX system, this
33318 will usually load @file{@var{libdir}/gdb/@var{reader-name}}, where
33319 @var{libdir} is the system library directory, usually
33320 @file{/usr/local/lib}. Only one reader can be active at a time;
33321 trying to load a second reader when one is already loaded will result
33322 in @value{GDBN} reporting an error. A new JIT reader can be loaded by
33323 first unloading the current one using @code{jit-reader-load} and then
33324 invoking @code{jit-reader-load}.
33326 @item jit-reader-unload
33327 Unload the currently loaded JIT reader.
33331 @node Writing JIT Debug Info Readers
33332 @subsection Writing JIT Debug Info Readers
33333 @cindex writing JIT debug info readers
33335 As mentioned, a reader is essentially a shared object conforming to a
33336 certain ABI. This ABI is described in @file{jit-reader.h}.
33338 @file{jit-reader.h} defines the structures, macros and functions
33339 required to write a reader. It is installed (along with
33340 @value{GDBN}), in @file{@var{includedir}/gdb} where @var{includedir} is
33341 the system include directory.
33343 Readers need to be released under a GPL compatible license. A reader
33344 can be declared as released under such a license by placing the macro
33345 @code{GDB_DECLARE_GPL_COMPATIBLE_READER} in a source file.
33347 The entry point for readers is the symbol @code{gdb_init_reader},
33348 which is expected to be a function with the prototype
33350 @findex gdb_init_reader
33352 extern struct gdb_reader_funcs *gdb_init_reader (void);
33355 @cindex @code{struct gdb_reader_funcs}
33357 @code{struct gdb_reader_funcs} contains a set of pointers to callback
33358 functions. These functions are executed to read the debug info
33359 generated by the JIT compiler (@code{read}), to unwind stack frames
33360 (@code{unwind}) and to create canonical frame IDs
33361 (@code{get_Frame_id}). It also has a callback that is called when the
33362 reader is being unloaded (@code{destroy}). The struct looks like this
33365 struct gdb_reader_funcs
33367 /* Must be set to GDB_READER_INTERFACE_VERSION. */
33368 int reader_version;
33370 /* For use by the reader. */
33373 gdb_read_debug_info *read;
33374 gdb_unwind_frame *unwind;
33375 gdb_get_frame_id *get_frame_id;
33376 gdb_destroy_reader *destroy;
33380 @cindex @code{struct gdb_symbol_callbacks}
33381 @cindex @code{struct gdb_unwind_callbacks}
33383 The callbacks are provided with another set of callbacks by
33384 @value{GDBN} to do their job. For @code{read}, these callbacks are
33385 passed in a @code{struct gdb_symbol_callbacks} and for @code{unwind}
33386 and @code{get_frame_id}, in a @code{struct gdb_unwind_callbacks}.
33387 @code{struct gdb_symbol_callbacks} has callbacks to create new object
33388 files and new symbol tables inside those object files. @code{struct
33389 gdb_unwind_callbacks} has callbacks to read registers off the current
33390 frame and to write out the values of the registers in the previous
33391 frame. Both have a callback (@code{target_read}) to read bytes off the
33392 target's address space.
33394 @node In-Process Agent
33395 @chapter In-Process Agent
33396 @cindex debugging agent
33397 The traditional debugging model is conceptually low-speed, but works fine,
33398 because most bugs can be reproduced in debugging-mode execution. However,
33399 as multi-core or many-core processors are becoming mainstream, and
33400 multi-threaded programs become more and more popular, there should be more
33401 and more bugs that only manifest themselves at normal-mode execution, for
33402 example, thread races, because debugger's interference with the program's
33403 timing may conceal the bugs. On the other hand, in some applications,
33404 it is not feasible for the debugger to interrupt the program's execution
33405 long enough for the developer to learn anything helpful about its behavior.
33406 If the program's correctness depends on its real-time behavior, delays
33407 introduced by a debugger might cause the program to fail, even when the
33408 code itself is correct. It is useful to be able to observe the program's
33409 behavior without interrupting it.
33411 Therefore, traditional debugging model is too intrusive to reproduce
33412 some bugs. In order to reduce the interference with the program, we can
33413 reduce the number of operations performed by debugger. The
33414 @dfn{In-Process Agent}, a shared library, is running within the same
33415 process with inferior, and is able to perform some debugging operations
33416 itself. As a result, debugger is only involved when necessary, and
33417 performance of debugging can be improved accordingly. Note that
33418 interference with program can be reduced but can't be removed completely,
33419 because the in-process agent will still stop or slow down the program.
33421 The in-process agent can interpret and execute Agent Expressions
33422 (@pxref{Agent Expressions}) during performing debugging operations. The
33423 agent expressions can be used for different purposes, such as collecting
33424 data in tracepoints, and condition evaluation in breakpoints.
33426 @anchor{Control Agent}
33427 You can control whether the in-process agent is used as an aid for
33428 debugging with the following commands:
33431 @kindex set agent on
33433 Causes the in-process agent to perform some operations on behalf of the
33434 debugger. Just which operations requested by the user will be done
33435 by the in-process agent depends on the its capabilities. For example,
33436 if you request to evaluate breakpoint conditions in the in-process agent,
33437 and the in-process agent has such capability as well, then breakpoint
33438 conditions will be evaluated in the in-process agent.
33440 @kindex set agent off
33441 @item set agent off
33442 Disables execution of debugging operations by the in-process agent. All
33443 of the operations will be performed by @value{GDBN}.
33447 Display the current setting of execution of debugging operations by
33448 the in-process agent.
33452 @chapter Reporting Bugs in @value{GDBN}
33453 @cindex bugs in @value{GDBN}
33454 @cindex reporting bugs in @value{GDBN}
33456 Your bug reports play an essential role in making @value{GDBN} reliable.
33458 Reporting a bug may help you by bringing a solution to your problem, or it
33459 may not. But in any case the principal function of a bug report is to help
33460 the entire community by making the next version of @value{GDBN} work better. Bug
33461 reports are your contribution to the maintenance of @value{GDBN}.
33463 In order for a bug report to serve its purpose, you must include the
33464 information that enables us to fix the bug.
33467 * Bug Criteria:: Have you found a bug?
33468 * Bug Reporting:: How to report bugs
33472 @section Have You Found a Bug?
33473 @cindex bug criteria
33475 If you are not sure whether you have found a bug, here are some guidelines:
33478 @cindex fatal signal
33479 @cindex debugger crash
33480 @cindex crash of debugger
33482 If the debugger gets a fatal signal, for any input whatever, that is a
33483 @value{GDBN} bug. Reliable debuggers never crash.
33485 @cindex error on valid input
33487 If @value{GDBN} produces an error message for valid input, that is a
33488 bug. (Note that if you're cross debugging, the problem may also be
33489 somewhere in the connection to the target.)
33491 @cindex invalid input
33493 If @value{GDBN} does not produce an error message for invalid input,
33494 that is a bug. However, you should note that your idea of
33495 ``invalid input'' might be our idea of ``an extension'' or ``support
33496 for traditional practice''.
33499 If you are an experienced user of debugging tools, your suggestions
33500 for improvement of @value{GDBN} are welcome in any case.
33503 @node Bug Reporting
33504 @section How to Report Bugs
33505 @cindex bug reports
33506 @cindex @value{GDBN} bugs, reporting
33508 A number of companies and individuals offer support for @sc{gnu} products.
33509 If you obtained @value{GDBN} from a support organization, we recommend you
33510 contact that organization first.
33512 You can find contact information for many support companies and
33513 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
33515 @c should add a web page ref...
33518 @ifset BUGURL_DEFAULT
33519 In any event, we also recommend that you submit bug reports for
33520 @value{GDBN}. The preferred method is to submit them directly using
33521 @uref{http://www.gnu.org/software/gdb/bugs/, @value{GDBN}'s Bugs web
33522 page}. Alternatively, the @email{bug-gdb@@gnu.org, e-mail gateway} can
33525 @strong{Do not send bug reports to @samp{info-gdb}, or to
33526 @samp{help-gdb}, or to any newsgroups.} Most users of @value{GDBN} do
33527 not want to receive bug reports. Those that do have arranged to receive
33530 The mailing list @samp{bug-gdb} has a newsgroup @samp{gnu.gdb.bug} which
33531 serves as a repeater. The mailing list and the newsgroup carry exactly
33532 the same messages. Often people think of posting bug reports to the
33533 newsgroup instead of mailing them. This appears to work, but it has one
33534 problem which can be crucial: a newsgroup posting often lacks a mail
33535 path back to the sender. Thus, if we need to ask for more information,
33536 we may be unable to reach you. For this reason, it is better to send
33537 bug reports to the mailing list.
33539 @ifclear BUGURL_DEFAULT
33540 In any event, we also recommend that you submit bug reports for
33541 @value{GDBN} to @value{BUGURL}.
33545 The fundamental principle of reporting bugs usefully is this:
33546 @strong{report all the facts}. If you are not sure whether to state a
33547 fact or leave it out, state it!
33549 Often people omit facts because they think they know what causes the
33550 problem and assume that some details do not matter. Thus, you might
33551 assume that the name of the variable you use in an example does not matter.
33552 Well, probably it does not, but one cannot be sure. Perhaps the bug is a
33553 stray memory reference which happens to fetch from the location where that
33554 name is stored in memory; perhaps, if the name were different, the contents
33555 of that location would fool the debugger into doing the right thing despite
33556 the bug. Play it safe and give a specific, complete example. That is the
33557 easiest thing for you to do, and the most helpful.
33559 Keep in mind that the purpose of a bug report is to enable us to fix the
33560 bug. It may be that the bug has been reported previously, but neither
33561 you nor we can know that unless your bug report is complete and
33564 Sometimes people give a few sketchy facts and ask, ``Does this ring a
33565 bell?'' Those bug reports are useless, and we urge everyone to
33566 @emph{refuse to respond to them} except to chide the sender to report
33569 To enable us to fix the bug, you should include all these things:
33573 The version of @value{GDBN}. @value{GDBN} announces it if you start
33574 with no arguments; you can also print it at any time using @code{show
33577 Without this, we will not know whether there is any point in looking for
33578 the bug in the current version of @value{GDBN}.
33581 The type of machine you are using, and the operating system name and
33585 What compiler (and its version) was used to compile @value{GDBN}---e.g.@:
33586 ``@value{GCC}--2.8.1''.
33589 What compiler (and its version) was used to compile the program you are
33590 debugging---e.g.@: ``@value{GCC}--2.8.1'', or ``HP92453-01 A.10.32.03 HP
33591 C Compiler''. For @value{NGCC}, you can say @kbd{@value{GCC} --version}
33592 to get this information; for other compilers, see the documentation for
33596 The command arguments you gave the compiler to compile your example and
33597 observe the bug. For example, did you use @samp{-O}? To guarantee
33598 you will not omit something important, list them all. A copy of the
33599 Makefile (or the output from make) is sufficient.
33601 If we were to try to guess the arguments, we would probably guess wrong
33602 and then we might not encounter the bug.
33605 A complete input script, and all necessary source files, that will
33609 A description of what behavior you observe that you believe is
33610 incorrect. For example, ``It gets a fatal signal.''
33612 Of course, if the bug is that @value{GDBN} gets a fatal signal, then we
33613 will certainly notice it. But if the bug is incorrect output, we might
33614 not notice unless it is glaringly wrong. You might as well not give us
33615 a chance to make a mistake.
33617 Even if the problem you experience is a fatal signal, you should still
33618 say so explicitly. Suppose something strange is going on, such as, your
33619 copy of @value{GDBN} is out of synch, or you have encountered a bug in
33620 the C library on your system. (This has happened!) Your copy might
33621 crash and ours would not. If you told us to expect a crash, then when
33622 ours fails to crash, we would know that the bug was not happening for
33623 us. If you had not told us to expect a crash, then we would not be able
33624 to draw any conclusion from our observations.
33627 @cindex recording a session script
33628 To collect all this information, you can use a session recording program
33629 such as @command{script}, which is available on many Unix systems.
33630 Just run your @value{GDBN} session inside @command{script} and then
33631 include the @file{typescript} file with your bug report.
33633 Another way to record a @value{GDBN} session is to run @value{GDBN}
33634 inside Emacs and then save the entire buffer to a file.
33637 If you wish to suggest changes to the @value{GDBN} source, send us context
33638 diffs. If you even discuss something in the @value{GDBN} source, refer to
33639 it by context, not by line number.
33641 The line numbers in our development sources will not match those in your
33642 sources. Your line numbers would convey no useful information to us.
33646 Here are some things that are not necessary:
33650 A description of the envelope of the bug.
33652 Often people who encounter a bug spend a lot of time investigating
33653 which changes to the input file will make the bug go away and which
33654 changes will not affect it.
33656 This is often time consuming and not very useful, because the way we
33657 will find the bug is by running a single example under the debugger
33658 with breakpoints, not by pure deduction from a series of examples.
33659 We recommend that you save your time for something else.
33661 Of course, if you can find a simpler example to report @emph{instead}
33662 of the original one, that is a convenience for us. Errors in the
33663 output will be easier to spot, running under the debugger will take
33664 less time, and so on.
33666 However, simplification is not vital; if you do not want to do this,
33667 report the bug anyway and send us the entire test case you used.
33670 A patch for the bug.
33672 A patch for the bug does help us if it is a good one. But do not omit
33673 the necessary information, such as the test case, on the assumption that
33674 a patch is all we need. We might see problems with your patch and decide
33675 to fix the problem another way, or we might not understand it at all.
33677 Sometimes with a program as complicated as @value{GDBN} it is very hard to
33678 construct an example that will make the program follow a certain path
33679 through the code. If you do not send us the example, we will not be able
33680 to construct one, so we will not be able to verify that the bug is fixed.
33682 And if we cannot understand what bug you are trying to fix, or why your
33683 patch should be an improvement, we will not install it. A test case will
33684 help us to understand.
33687 A guess about what the bug is or what it depends on.
33689 Such guesses are usually wrong. Even we cannot guess right about such
33690 things without first using the debugger to find the facts.
33693 @c The readline documentation is distributed with the readline code
33694 @c and consists of the two following files:
33697 @c Use -I with makeinfo to point to the appropriate directory,
33698 @c environment var TEXINPUTS with TeX.
33699 @ifclear SYSTEM_READLINE
33700 @include rluser.texi
33701 @include hsuser.texi
33705 @appendix In Memoriam
33707 The @value{GDBN} project mourns the loss of the following long-time
33712 Fred was a long-standing contributor to @value{GDBN} (1991-2006), and
33713 to Free Software in general. Outside of @value{GDBN}, he was known in
33714 the Amiga world for his series of Fish Disks, and the GeekGadget project.
33716 @item Michael Snyder
33717 Michael was one of the Global Maintainers of the @value{GDBN} project,
33718 with contributions recorded as early as 1996, until 2011. In addition
33719 to his day to day participation, he was a large driving force behind
33720 adding Reverse Debugging to @value{GDBN}.
33723 Beyond their technical contributions to the project, they were also
33724 enjoyable members of the Free Software Community. We will miss them.
33726 @node Formatting Documentation
33727 @appendix Formatting Documentation
33729 @cindex @value{GDBN} reference card
33730 @cindex reference card
33731 The @value{GDBN} 4 release includes an already-formatted reference card, ready
33732 for printing with PostScript or Ghostscript, in the @file{gdb}
33733 subdirectory of the main source directory@footnote{In
33734 @file{gdb-@value{GDBVN}/gdb/refcard.ps} of the version @value{GDBVN}
33735 release.}. If you can use PostScript or Ghostscript with your printer,
33736 you can print the reference card immediately with @file{refcard.ps}.
33738 The release also includes the source for the reference card. You
33739 can format it, using @TeX{}, by typing:
33745 The @value{GDBN} reference card is designed to print in @dfn{landscape}
33746 mode on US ``letter'' size paper;
33747 that is, on a sheet 11 inches wide by 8.5 inches
33748 high. You will need to specify this form of printing as an option to
33749 your @sc{dvi} output program.
33751 @cindex documentation
33753 All the documentation for @value{GDBN} comes as part of the machine-readable
33754 distribution. The documentation is written in Texinfo format, which is
33755 a documentation system that uses a single source file to produce both
33756 on-line information and a printed manual. You can use one of the Info
33757 formatting commands to create the on-line version of the documentation
33758 and @TeX{} (or @code{texi2roff}) to typeset the printed version.
33760 @value{GDBN} includes an already formatted copy of the on-line Info
33761 version of this manual in the @file{gdb} subdirectory. The main Info
33762 file is @file{gdb-@value{GDBVN}/gdb/gdb.info}, and it refers to
33763 subordinate files matching @samp{gdb.info*} in the same directory. If
33764 necessary, you can print out these files, or read them with any editor;
33765 but they are easier to read using the @code{info} subsystem in @sc{gnu}
33766 Emacs or the standalone @code{info} program, available as part of the
33767 @sc{gnu} Texinfo distribution.
33769 If you want to format these Info files yourself, you need one of the
33770 Info formatting programs, such as @code{texinfo-format-buffer} or
33773 If you have @code{makeinfo} installed, and are in the top level
33774 @value{GDBN} source directory (@file{gdb-@value{GDBVN}}, in the case of
33775 version @value{GDBVN}), you can make the Info file by typing:
33782 If you want to typeset and print copies of this manual, you need @TeX{},
33783 a program to print its @sc{dvi} output files, and @file{texinfo.tex}, the
33784 Texinfo definitions file.
33786 @TeX{} is a typesetting program; it does not print files directly, but
33787 produces output files called @sc{dvi} files. To print a typeset
33788 document, you need a program to print @sc{dvi} files. If your system
33789 has @TeX{} installed, chances are it has such a program. The precise
33790 command to use depends on your system; @kbd{lpr -d} is common; another
33791 (for PostScript devices) is @kbd{dvips}. The @sc{dvi} print command may
33792 require a file name without any extension or a @samp{.dvi} extension.
33794 @TeX{} also requires a macro definitions file called
33795 @file{texinfo.tex}. This file tells @TeX{} how to typeset a document
33796 written in Texinfo format. On its own, @TeX{} cannot either read or
33797 typeset a Texinfo file. @file{texinfo.tex} is distributed with GDB
33798 and is located in the @file{gdb-@var{version-number}/texinfo}
33801 If you have @TeX{} and a @sc{dvi} printer program installed, you can
33802 typeset and print this manual. First switch to the @file{gdb}
33803 subdirectory of the main source directory (for example, to
33804 @file{gdb-@value{GDBVN}/gdb}) and type:
33810 Then give @file{gdb.dvi} to your @sc{dvi} printing program.
33812 @node Installing GDB
33813 @appendix Installing @value{GDBN}
33814 @cindex installation
33817 * Requirements:: Requirements for building @value{GDBN}
33818 * Running Configure:: Invoking the @value{GDBN} @file{configure} script
33819 * Separate Objdir:: Compiling @value{GDBN} in another directory
33820 * Config Names:: Specifying names for hosts and targets
33821 * Configure Options:: Summary of options for configure
33822 * System-wide configuration:: Having a system-wide init file
33826 @section Requirements for Building @value{GDBN}
33827 @cindex building @value{GDBN}, requirements for
33829 Building @value{GDBN} requires various tools and packages to be available.
33830 Other packages will be used only if they are found.
33832 @heading Tools/Packages Necessary for Building @value{GDBN}
33834 @item ISO C90 compiler
33835 @value{GDBN} is written in ISO C90. It should be buildable with any
33836 working C90 compiler, e.g.@: GCC.
33840 @heading Tools/Packages Optional for Building @value{GDBN}
33844 @value{GDBN} can use the Expat XML parsing library. This library may be
33845 included with your operating system distribution; if it is not, you
33846 can get the latest version from @url{http://expat.sourceforge.net}.
33847 The @file{configure} script will search for this library in several
33848 standard locations; if it is installed in an unusual path, you can
33849 use the @option{--with-libexpat-prefix} option to specify its location.
33855 Remote protocol memory maps (@pxref{Memory Map Format})
33857 Target descriptions (@pxref{Target Descriptions})
33859 Remote shared library lists (@xref{Library List Format},
33860 or alternatively @pxref{Library List Format for SVR4 Targets})
33862 MS-Windows shared libraries (@pxref{Shared Libraries})
33864 Traceframe info (@pxref{Traceframe Info Format})
33868 @cindex compressed debug sections
33869 @value{GDBN} will use the @samp{zlib} library, if available, to read
33870 compressed debug sections. Some linkers, such as GNU gold, are capable
33871 of producing binaries with compressed debug sections. If @value{GDBN}
33872 is compiled with @samp{zlib}, it will be able to read the debug
33873 information in such binaries.
33875 The @samp{zlib} library is likely included with your operating system
33876 distribution; if it is not, you can get the latest version from
33877 @url{http://zlib.net}.
33880 @value{GDBN}'s features related to character sets (@pxref{Character
33881 Sets}) require a functioning @code{iconv} implementation. If you are
33882 on a GNU system, then this is provided by the GNU C Library. Some
33883 other systems also provide a working @code{iconv}.
33885 If @value{GDBN} is using the @code{iconv} program which is installed
33886 in a non-standard place, you will need to tell @value{GDBN} where to find it.
33887 This is done with @option{--with-iconv-bin} which specifies the
33888 directory that contains the @code{iconv} program.
33890 On systems without @code{iconv}, you can install GNU Libiconv. If you
33891 have previously installed Libiconv, you can use the
33892 @option{--with-libiconv-prefix} option to configure.
33894 @value{GDBN}'s top-level @file{configure} and @file{Makefile} will
33895 arrange to build Libiconv if a directory named @file{libiconv} appears
33896 in the top-most source directory. If Libiconv is built this way, and
33897 if the operating system does not provide a suitable @code{iconv}
33898 implementation, then the just-built library will automatically be used
33899 by @value{GDBN}. One easy way to set this up is to download GNU
33900 Libiconv, unpack it, and then rename the directory holding the
33901 Libiconv source code to @samp{libiconv}.
33904 @node Running Configure
33905 @section Invoking the @value{GDBN} @file{configure} Script
33906 @cindex configuring @value{GDBN}
33907 @value{GDBN} comes with a @file{configure} script that automates the process
33908 of preparing @value{GDBN} for installation; you can then use @code{make} to
33909 build the @code{gdb} program.
33911 @c irrelevant in info file; it's as current as the code it lives with.
33912 @footnote{If you have a more recent version of @value{GDBN} than @value{GDBVN},
33913 look at the @file{README} file in the sources; we may have improved the
33914 installation procedures since publishing this manual.}
33917 The @value{GDBN} distribution includes all the source code you need for
33918 @value{GDBN} in a single directory, whose name is usually composed by
33919 appending the version number to @samp{gdb}.
33921 For example, the @value{GDBN} version @value{GDBVN} distribution is in the
33922 @file{gdb-@value{GDBVN}} directory. That directory contains:
33925 @item gdb-@value{GDBVN}/configure @r{(and supporting files)}
33926 script for configuring @value{GDBN} and all its supporting libraries
33928 @item gdb-@value{GDBVN}/gdb
33929 the source specific to @value{GDBN} itself
33931 @item gdb-@value{GDBVN}/bfd
33932 source for the Binary File Descriptor library
33934 @item gdb-@value{GDBVN}/include
33935 @sc{gnu} include files
33937 @item gdb-@value{GDBVN}/libiberty
33938 source for the @samp{-liberty} free software library
33940 @item gdb-@value{GDBVN}/opcodes
33941 source for the library of opcode tables and disassemblers
33943 @item gdb-@value{GDBVN}/readline
33944 source for the @sc{gnu} command-line interface
33946 @item gdb-@value{GDBVN}/glob
33947 source for the @sc{gnu} filename pattern-matching subroutine
33949 @item gdb-@value{GDBVN}/mmalloc
33950 source for the @sc{gnu} memory-mapped malloc package
33953 The simplest way to configure and build @value{GDBN} is to run @file{configure}
33954 from the @file{gdb-@var{version-number}} source directory, which in
33955 this example is the @file{gdb-@value{GDBVN}} directory.
33957 First switch to the @file{gdb-@var{version-number}} source directory
33958 if you are not already in it; then run @file{configure}. Pass the
33959 identifier for the platform on which @value{GDBN} will run as an
33965 cd gdb-@value{GDBVN}
33966 ./configure @var{host}
33971 where @var{host} is an identifier such as @samp{sun4} or
33972 @samp{decstation}, that identifies the platform where @value{GDBN} will run.
33973 (You can often leave off @var{host}; @file{configure} tries to guess the
33974 correct value by examining your system.)
33976 Running @samp{configure @var{host}} and then running @code{make} builds the
33977 @file{bfd}, @file{readline}, @file{mmalloc}, and @file{libiberty}
33978 libraries, then @code{gdb} itself. The configured source files, and the
33979 binaries, are left in the corresponding source directories.
33982 @file{configure} is a Bourne-shell (@code{/bin/sh}) script; if your
33983 system does not recognize this automatically when you run a different
33984 shell, you may need to run @code{sh} on it explicitly:
33987 sh configure @var{host}
33990 If you run @file{configure} from a directory that contains source
33991 directories for multiple libraries or programs, such as the
33992 @file{gdb-@value{GDBVN}} source directory for version @value{GDBVN},
33994 creates configuration files for every directory level underneath (unless
33995 you tell it not to, with the @samp{--norecursion} option).
33997 You should run the @file{configure} script from the top directory in the
33998 source tree, the @file{gdb-@var{version-number}} directory. If you run
33999 @file{configure} from one of the subdirectories, you will configure only
34000 that subdirectory. That is usually not what you want. In particular,
34001 if you run the first @file{configure} from the @file{gdb} subdirectory
34002 of the @file{gdb-@var{version-number}} directory, you will omit the
34003 configuration of @file{bfd}, @file{readline}, and other sibling
34004 directories of the @file{gdb} subdirectory. This leads to build errors
34005 about missing include files such as @file{bfd/bfd.h}.
34007 You can install @code{@value{GDBP}} anywhere; it has no hardwired paths.
34008 However, you should make sure that the shell on your path (named by
34009 the @samp{SHELL} environment variable) is publicly readable. Remember
34010 that @value{GDBN} uses the shell to start your program---some systems refuse to
34011 let @value{GDBN} debug child processes whose programs are not readable.
34013 @node Separate Objdir
34014 @section Compiling @value{GDBN} in Another Directory
34016 If you want to run @value{GDBN} versions for several host or target machines,
34017 you need a different @code{gdb} compiled for each combination of
34018 host and target. @file{configure} is designed to make this easy by
34019 allowing you to generate each configuration in a separate subdirectory,
34020 rather than in the source directory. If your @code{make} program
34021 handles the @samp{VPATH} feature (@sc{gnu} @code{make} does), running
34022 @code{make} in each of these directories builds the @code{gdb}
34023 program specified there.
34025 To build @code{gdb} in a separate directory, run @file{configure}
34026 with the @samp{--srcdir} option to specify where to find the source.
34027 (You also need to specify a path to find @file{configure}
34028 itself from your working directory. If the path to @file{configure}
34029 would be the same as the argument to @samp{--srcdir}, you can leave out
34030 the @samp{--srcdir} option; it is assumed.)
34032 For example, with version @value{GDBVN}, you can build @value{GDBN} in a
34033 separate directory for a Sun 4 like this:
34037 cd gdb-@value{GDBVN}
34040 ../gdb-@value{GDBVN}/configure sun4
34045 When @file{configure} builds a configuration using a remote source
34046 directory, it creates a tree for the binaries with the same structure
34047 (and using the same names) as the tree under the source directory. In
34048 the example, you'd find the Sun 4 library @file{libiberty.a} in the
34049 directory @file{gdb-sun4/libiberty}, and @value{GDBN} itself in
34050 @file{gdb-sun4/gdb}.
34052 Make sure that your path to the @file{configure} script has just one
34053 instance of @file{gdb} in it. If your path to @file{configure} looks
34054 like @file{../gdb-@value{GDBVN}/gdb/configure}, you are configuring only
34055 one subdirectory of @value{GDBN}, not the whole package. This leads to
34056 build errors about missing include files such as @file{bfd/bfd.h}.
34058 One popular reason to build several @value{GDBN} configurations in separate
34059 directories is to configure @value{GDBN} for cross-compiling (where
34060 @value{GDBN} runs on one machine---the @dfn{host}---while debugging
34061 programs that run on another machine---the @dfn{target}).
34062 You specify a cross-debugging target by
34063 giving the @samp{--target=@var{target}} option to @file{configure}.
34065 When you run @code{make} to build a program or library, you must run
34066 it in a configured directory---whatever directory you were in when you
34067 called @file{configure} (or one of its subdirectories).
34069 The @code{Makefile} that @file{configure} generates in each source
34070 directory also runs recursively. If you type @code{make} in a source
34071 directory such as @file{gdb-@value{GDBVN}} (or in a separate configured
34072 directory configured with @samp{--srcdir=@var{dirname}/gdb-@value{GDBVN}}), you
34073 will build all the required libraries, and then build GDB.
34075 When you have multiple hosts or targets configured in separate
34076 directories, you can run @code{make} on them in parallel (for example,
34077 if they are NFS-mounted on each of the hosts); they will not interfere
34081 @section Specifying Names for Hosts and Targets
34083 The specifications used for hosts and targets in the @file{configure}
34084 script are based on a three-part naming scheme, but some short predefined
34085 aliases are also supported. The full naming scheme encodes three pieces
34086 of information in the following pattern:
34089 @var{architecture}-@var{vendor}-@var{os}
34092 For example, you can use the alias @code{sun4} as a @var{host} argument,
34093 or as the value for @var{target} in a @code{--target=@var{target}}
34094 option. The equivalent full name is @samp{sparc-sun-sunos4}.
34096 The @file{configure} script accompanying @value{GDBN} does not provide
34097 any query facility to list all supported host and target names or
34098 aliases. @file{configure} calls the Bourne shell script
34099 @code{config.sub} to map abbreviations to full names; you can read the
34100 script, if you wish, or you can use it to test your guesses on
34101 abbreviations---for example:
34104 % sh config.sub i386-linux
34106 % sh config.sub alpha-linux
34107 alpha-unknown-linux-gnu
34108 % sh config.sub hp9k700
34110 % sh config.sub sun4
34111 sparc-sun-sunos4.1.1
34112 % sh config.sub sun3
34113 m68k-sun-sunos4.1.1
34114 % sh config.sub i986v
34115 Invalid configuration `i986v': machine `i986v' not recognized
34119 @code{config.sub} is also distributed in the @value{GDBN} source
34120 directory (@file{gdb-@value{GDBVN}}, for version @value{GDBVN}).
34122 @node Configure Options
34123 @section @file{configure} Options
34125 Here is a summary of the @file{configure} options and arguments that
34126 are most often useful for building @value{GDBN}. @file{configure} also has
34127 several other options not listed here. @inforef{What Configure
34128 Does,,configure.info}, for a full explanation of @file{configure}.
34131 configure @r{[}--help@r{]}
34132 @r{[}--prefix=@var{dir}@r{]}
34133 @r{[}--exec-prefix=@var{dir}@r{]}
34134 @r{[}--srcdir=@var{dirname}@r{]}
34135 @r{[}--norecursion@r{]} @r{[}--rm@r{]}
34136 @r{[}--target=@var{target}@r{]}
34141 You may introduce options with a single @samp{-} rather than
34142 @samp{--} if you prefer; but you may abbreviate option names if you use
34147 Display a quick summary of how to invoke @file{configure}.
34149 @item --prefix=@var{dir}
34150 Configure the source to install programs and files under directory
34153 @item --exec-prefix=@var{dir}
34154 Configure the source to install programs under directory
34157 @c avoid splitting the warning from the explanation:
34159 @item --srcdir=@var{dirname}
34160 @strong{Warning: using this option requires @sc{gnu} @code{make}, or another
34161 @code{make} that implements the @code{VPATH} feature.}@*
34162 Use this option to make configurations in directories separate from the
34163 @value{GDBN} source directories. Among other things, you can use this to
34164 build (or maintain) several configurations simultaneously, in separate
34165 directories. @file{configure} writes configuration-specific files in
34166 the current directory, but arranges for them to use the source in the
34167 directory @var{dirname}. @file{configure} creates directories under
34168 the working directory in parallel to the source directories below
34171 @item --norecursion
34172 Configure only the directory level where @file{configure} is executed; do not
34173 propagate configuration to subdirectories.
34175 @item --target=@var{target}
34176 Configure @value{GDBN} for cross-debugging programs running on the specified
34177 @var{target}. Without this option, @value{GDBN} is configured to debug
34178 programs that run on the same machine (@var{host}) as @value{GDBN} itself.
34180 There is no convenient way to generate a list of all available targets.
34182 @item @var{host} @dots{}
34183 Configure @value{GDBN} to run on the specified @var{host}.
34185 There is no convenient way to generate a list of all available hosts.
34188 There are many other options available as well, but they are generally
34189 needed for special purposes only.
34191 @node System-wide configuration
34192 @section System-wide configuration and settings
34193 @cindex system-wide init file
34195 @value{GDBN} can be configured to have a system-wide init file;
34196 this file will be read and executed at startup (@pxref{Startup, , What
34197 @value{GDBN} does during startup}).
34199 Here is the corresponding configure option:
34202 @item --with-system-gdbinit=@var{file}
34203 Specify that the default location of the system-wide init file is
34207 If @value{GDBN} has been configured with the option @option{--prefix=$prefix},
34208 it may be subject to relocation. Two possible cases:
34212 If the default location of this init file contains @file{$prefix},
34213 it will be subject to relocation. Suppose that the configure options
34214 are @option{--prefix=$prefix --with-system-gdbinit=$prefix/etc/gdbinit};
34215 if @value{GDBN} is moved from @file{$prefix} to @file{$install}, the system
34216 init file is looked for as @file{$install/etc/gdbinit} instead of
34217 @file{$prefix/etc/gdbinit}.
34220 By contrast, if the default location does not contain the prefix,
34221 it will not be relocated. E.g.@: if @value{GDBN} has been configured with
34222 @option{--prefix=/usr/local --with-system-gdbinit=/usr/share/gdb/gdbinit},
34223 then @value{GDBN} will always look for @file{/usr/share/gdb/gdbinit},
34224 wherever @value{GDBN} is installed.
34227 @node Maintenance Commands
34228 @appendix Maintenance Commands
34229 @cindex maintenance commands
34230 @cindex internal commands
34232 In addition to commands intended for @value{GDBN} users, @value{GDBN}
34233 includes a number of commands intended for @value{GDBN} developers,
34234 that are not documented elsewhere in this manual. These commands are
34235 provided here for reference. (For commands that turn on debugging
34236 messages, see @ref{Debugging Output}.)
34239 @kindex maint agent
34240 @kindex maint agent-eval
34241 @item maint agent @var{expression}
34242 @itemx maint agent-eval @var{expression}
34243 Translate the given @var{expression} into remote agent bytecodes.
34244 This command is useful for debugging the Agent Expression mechanism
34245 (@pxref{Agent Expressions}). The @samp{agent} version produces an
34246 expression useful for data collection, such as by tracepoints, while
34247 @samp{maint agent-eval} produces an expression that evaluates directly
34248 to a result. For instance, a collection expression for @code{globa +
34249 globb} will include bytecodes to record four bytes of memory at each
34250 of the addresses of @code{globa} and @code{globb}, while discarding
34251 the result of the addition, while an evaluation expression will do the
34252 addition and return the sum.
34254 @kindex maint info breakpoints
34255 @item @anchor{maint info breakpoints}maint info breakpoints
34256 Using the same format as @samp{info breakpoints}, display both the
34257 breakpoints you've set explicitly, and those @value{GDBN} is using for
34258 internal purposes. Internal breakpoints are shown with negative
34259 breakpoint numbers. The type column identifies what kind of breakpoint
34264 Normal, explicitly set breakpoint.
34267 Normal, explicitly set watchpoint.
34270 Internal breakpoint, used to handle correctly stepping through
34271 @code{longjmp} calls.
34273 @item longjmp resume
34274 Internal breakpoint at the target of a @code{longjmp}.
34277 Temporary internal breakpoint used by the @value{GDBN} @code{until} command.
34280 Temporary internal breakpoint used by the @value{GDBN} @code{finish} command.
34283 Shared library events.
34287 @kindex set displaced-stepping
34288 @kindex show displaced-stepping
34289 @cindex displaced stepping support
34290 @cindex out-of-line single-stepping
34291 @item set displaced-stepping
34292 @itemx show displaced-stepping
34293 Control whether or not @value{GDBN} will do @dfn{displaced stepping}
34294 if the target supports it. Displaced stepping is a way to single-step
34295 over breakpoints without removing them from the inferior, by executing
34296 an out-of-line copy of the instruction that was originally at the
34297 breakpoint location. It is also known as out-of-line single-stepping.
34300 @item set displaced-stepping on
34301 If the target architecture supports it, @value{GDBN} will use
34302 displaced stepping to step over breakpoints.
34304 @item set displaced-stepping off
34305 @value{GDBN} will not use displaced stepping to step over breakpoints,
34306 even if such is supported by the target architecture.
34308 @cindex non-stop mode, and @samp{set displaced-stepping}
34309 @item set displaced-stepping auto
34310 This is the default mode. @value{GDBN} will use displaced stepping
34311 only if non-stop mode is active (@pxref{Non-Stop Mode}) and the target
34312 architecture supports displaced stepping.
34315 @kindex maint check-symtabs
34316 @item maint check-symtabs
34317 Check the consistency of psymtabs and symtabs.
34319 @kindex maint cplus first_component
34320 @item maint cplus first_component @var{name}
34321 Print the first C@t{++} class/namespace component of @var{name}.
34323 @kindex maint cplus namespace
34324 @item maint cplus namespace
34325 Print the list of possible C@t{++} namespaces.
34327 @kindex maint demangle
34328 @item maint demangle @var{name}
34329 Demangle a C@t{++} or Objective-C mangled @var{name}.
34331 @kindex maint deprecate
34332 @kindex maint undeprecate
34333 @cindex deprecated commands
34334 @item maint deprecate @var{command} @r{[}@var{replacement}@r{]}
34335 @itemx maint undeprecate @var{command}
34336 Deprecate or undeprecate the named @var{command}. Deprecated commands
34337 cause @value{GDBN} to issue a warning when you use them. The optional
34338 argument @var{replacement} says which newer command should be used in
34339 favor of the deprecated one; if it is given, @value{GDBN} will mention
34340 the replacement as part of the warning.
34342 @kindex maint dump-me
34343 @item maint dump-me
34344 @cindex @code{SIGQUIT} signal, dump core of @value{GDBN}
34345 Cause a fatal signal in the debugger and force it to dump its core.
34346 This is supported only on systems which support aborting a program
34347 with the @code{SIGQUIT} signal.
34349 @kindex maint internal-error
34350 @kindex maint internal-warning
34351 @item maint internal-error @r{[}@var{message-text}@r{]}
34352 @itemx maint internal-warning @r{[}@var{message-text}@r{]}
34353 Cause @value{GDBN} to call the internal function @code{internal_error}
34354 or @code{internal_warning} and hence behave as though an internal error
34355 or internal warning has been detected. In addition to reporting the
34356 internal problem, these functions give the user the opportunity to
34357 either quit @value{GDBN} or create a core file of the current
34358 @value{GDBN} session.
34360 These commands take an optional parameter @var{message-text} that is
34361 used as the text of the error or warning message.
34363 Here's an example of using @code{internal-error}:
34366 (@value{GDBP}) @kbd{maint internal-error testing, 1, 2}
34367 @dots{}/maint.c:121: internal-error: testing, 1, 2
34368 A problem internal to GDB has been detected. Further
34369 debugging may prove unreliable.
34370 Quit this debugging session? (y or n) @kbd{n}
34371 Create a core file? (y or n) @kbd{n}
34375 @cindex @value{GDBN} internal error
34376 @cindex internal errors, control of @value{GDBN} behavior
34378 @kindex maint set internal-error
34379 @kindex maint show internal-error
34380 @kindex maint set internal-warning
34381 @kindex maint show internal-warning
34382 @item maint set internal-error @var{action} [ask|yes|no]
34383 @itemx maint show internal-error @var{action}
34384 @itemx maint set internal-warning @var{action} [ask|yes|no]
34385 @itemx maint show internal-warning @var{action}
34386 When @value{GDBN} reports an internal problem (error or warning) it
34387 gives the user the opportunity to both quit @value{GDBN} and create a
34388 core file of the current @value{GDBN} session. These commands let you
34389 override the default behaviour for each particular @var{action},
34390 described in the table below.
34394 You can specify that @value{GDBN} should always (yes) or never (no)
34395 quit. The default is to ask the user what to do.
34398 You can specify that @value{GDBN} should always (yes) or never (no)
34399 create a core file. The default is to ask the user what to do.
34402 @kindex maint packet
34403 @item maint packet @var{text}
34404 If @value{GDBN} is talking to an inferior via the serial protocol,
34405 then this command sends the string @var{text} to the inferior, and
34406 displays the response packet. @value{GDBN} supplies the initial
34407 @samp{$} character, the terminating @samp{#} character, and the
34410 @kindex maint print architecture
34411 @item maint print architecture @r{[}@var{file}@r{]}
34412 Print the entire architecture configuration. The optional argument
34413 @var{file} names the file where the output goes.
34415 @kindex maint print c-tdesc
34416 @item maint print c-tdesc
34417 Print the current target description (@pxref{Target Descriptions}) as
34418 a C source file. The created source file can be used in @value{GDBN}
34419 when an XML parser is not available to parse the description.
34421 @kindex maint print dummy-frames
34422 @item maint print dummy-frames
34423 Prints the contents of @value{GDBN}'s internal dummy-frame stack.
34426 (@value{GDBP}) @kbd{b add}
34428 (@value{GDBP}) @kbd{print add(2,3)}
34429 Breakpoint 2, add (a=2, b=3) at @dots{}
34431 The program being debugged stopped while in a function called from GDB.
34433 (@value{GDBP}) @kbd{maint print dummy-frames}
34434 0x1a57c80: pc=0x01014068 fp=0x0200bddc sp=0x0200bdd6
34435 top=0x0200bdd4 id=@{stack=0x200bddc,code=0x101405c@}
34436 call_lo=0x01014000 call_hi=0x01014001
34440 Takes an optional file parameter.
34442 @kindex maint print registers
34443 @kindex maint print raw-registers
34444 @kindex maint print cooked-registers
34445 @kindex maint print register-groups
34446 @kindex maint print remote-registers
34447 @item maint print registers @r{[}@var{file}@r{]}
34448 @itemx maint print raw-registers @r{[}@var{file}@r{]}
34449 @itemx maint print cooked-registers @r{[}@var{file}@r{]}
34450 @itemx maint print register-groups @r{[}@var{file}@r{]}
34451 @itemx maint print remote-registers @r{[}@var{file}@r{]}
34452 Print @value{GDBN}'s internal register data structures.
34454 The command @code{maint print raw-registers} includes the contents of
34455 the raw register cache; the command @code{maint print
34456 cooked-registers} includes the (cooked) value of all registers,
34457 including registers which aren't available on the target nor visible
34458 to user; the command @code{maint print register-groups} includes the
34459 groups that each register is a member of; and the command @code{maint
34460 print remote-registers} includes the remote target's register numbers
34461 and offsets in the `G' packets. @xref{Registers,, Registers, gdbint,
34462 @value{GDBN} Internals}.
34464 These commands take an optional parameter, a file name to which to
34465 write the information.
34467 @kindex maint print reggroups
34468 @item maint print reggroups @r{[}@var{file}@r{]}
34469 Print @value{GDBN}'s internal register group data structures. The
34470 optional argument @var{file} tells to what file to write the
34473 The register groups info looks like this:
34476 (@value{GDBP}) @kbd{maint print reggroups}
34489 This command forces @value{GDBN} to flush its internal register cache.
34491 @kindex maint print objfiles
34492 @cindex info for known object files
34493 @item maint print objfiles
34494 Print a dump of all known object files. For each object file, this
34495 command prints its name, address in memory, and all of its psymtabs
34498 @kindex maint print section-scripts
34499 @cindex info for known .debug_gdb_scripts-loaded scripts
34500 @item maint print section-scripts [@var{regexp}]
34501 Print a dump of scripts specified in the @code{.debug_gdb_section} section.
34502 If @var{regexp} is specified, only print scripts loaded by object files
34503 matching @var{regexp}.
34504 For each script, this command prints its name as specified in the objfile,
34505 and the full path if known.
34506 @xref{dotdebug_gdb_scripts section}.
34508 @kindex maint print statistics
34509 @cindex bcache statistics
34510 @item maint print statistics
34511 This command prints, for each object file in the program, various data
34512 about that object file followed by the byte cache (@dfn{bcache})
34513 statistics for the object file. The objfile data includes the number
34514 of minimal, partial, full, and stabs symbols, the number of types
34515 defined by the objfile, the number of as yet unexpanded psym tables,
34516 the number of line tables and string tables, and the amount of memory
34517 used by the various tables. The bcache statistics include the counts,
34518 sizes, and counts of duplicates of all and unique objects, max,
34519 average, and median entry size, total memory used and its overhead and
34520 savings, and various measures of the hash table size and chain
34523 @kindex maint print target-stack
34524 @cindex target stack description
34525 @item maint print target-stack
34526 A @dfn{target} is an interface between the debugger and a particular
34527 kind of file or process. Targets can be stacked in @dfn{strata},
34528 so that more than one target can potentially respond to a request.
34529 In particular, memory accesses will walk down the stack of targets
34530 until they find a target that is interested in handling that particular
34533 This command prints a short description of each layer that was pushed on
34534 the @dfn{target stack}, starting from the top layer down to the bottom one.
34536 @kindex maint print type
34537 @cindex type chain of a data type
34538 @item maint print type @var{expr}
34539 Print the type chain for a type specified by @var{expr}. The argument
34540 can be either a type name or a symbol. If it is a symbol, the type of
34541 that symbol is described. The type chain produced by this command is
34542 a recursive definition of the data type as stored in @value{GDBN}'s
34543 data structures, including its flags and contained types.
34545 @kindex maint set dwarf2 always-disassemble
34546 @kindex maint show dwarf2 always-disassemble
34547 @item maint set dwarf2 always-disassemble
34548 @item maint show dwarf2 always-disassemble
34549 Control the behavior of @code{info address} when using DWARF debugging
34552 The default is @code{off}, which means that @value{GDBN} should try to
34553 describe a variable's location in an easily readable format. When
34554 @code{on}, @value{GDBN} will instead display the DWARF location
34555 expression in an assembly-like format. Note that some locations are
34556 too complex for @value{GDBN} to describe simply; in this case you will
34557 always see the disassembly form.
34559 Here is an example of the resulting disassembly:
34562 (gdb) info addr argc
34563 Symbol "argc" is a complex DWARF expression:
34567 For more information on these expressions, see
34568 @uref{http://www.dwarfstd.org/, the DWARF standard}.
34570 @kindex maint set dwarf2 max-cache-age
34571 @kindex maint show dwarf2 max-cache-age
34572 @item maint set dwarf2 max-cache-age
34573 @itemx maint show dwarf2 max-cache-age
34574 Control the DWARF 2 compilation unit cache.
34576 @cindex DWARF 2 compilation units cache
34577 In object files with inter-compilation-unit references, such as those
34578 produced by the GCC option @samp{-feliminate-dwarf2-dups}, the DWARF 2
34579 reader needs to frequently refer to previously read compilation units.
34580 This setting controls how long a compilation unit will remain in the
34581 cache if it is not referenced. A higher limit means that cached
34582 compilation units will be stored in memory longer, and more total
34583 memory will be used. Setting it to zero disables caching, which will
34584 slow down @value{GDBN} startup, but reduce memory consumption.
34586 @kindex maint set profile
34587 @kindex maint show profile
34588 @cindex profiling GDB
34589 @item maint set profile
34590 @itemx maint show profile
34591 Control profiling of @value{GDBN}.
34593 Profiling will be disabled until you use the @samp{maint set profile}
34594 command to enable it. When you enable profiling, the system will begin
34595 collecting timing and execution count data; when you disable profiling or
34596 exit @value{GDBN}, the results will be written to a log file. Remember that
34597 if you use profiling, @value{GDBN} will overwrite the profiling log file
34598 (often called @file{gmon.out}). If you have a record of important profiling
34599 data in a @file{gmon.out} file, be sure to move it to a safe location.
34601 Configuring with @samp{--enable-profiling} arranges for @value{GDBN} to be
34602 compiled with the @samp{-pg} compiler option.
34604 @kindex maint set show-debug-regs
34605 @kindex maint show show-debug-regs
34606 @cindex hardware debug registers
34607 @item maint set show-debug-regs
34608 @itemx maint show show-debug-regs
34609 Control whether to show variables that mirror the hardware debug
34610 registers. Use @code{ON} to enable, @code{OFF} to disable. If
34611 enabled, the debug registers values are shown when @value{GDBN} inserts or
34612 removes a hardware breakpoint or watchpoint, and when the inferior
34613 triggers a hardware-assisted breakpoint or watchpoint.
34615 @kindex maint set show-all-tib
34616 @kindex maint show show-all-tib
34617 @item maint set show-all-tib
34618 @itemx maint show show-all-tib
34619 Control whether to show all non zero areas within a 1k block starting
34620 at thread local base, when using the @samp{info w32 thread-information-block}
34623 @kindex maint space
34624 @cindex memory used by commands
34626 Control whether to display memory usage for each command. If set to a
34627 nonzero value, @value{GDBN} will display how much memory each command
34628 took, following the command's own output. This can also be requested
34629 by invoking @value{GDBN} with the @option{--statistics} command-line
34630 switch (@pxref{Mode Options}).
34633 @cindex time of command execution
34635 Control whether to display the execution time of @value{GDBN} for each command.
34636 If set to a nonzero value, @value{GDBN} will display how much time it
34637 took to execute each command, following the command's own output.
34638 Both CPU time and wallclock time are printed.
34639 Printing both is useful when trying to determine whether the cost is
34640 CPU or, e.g., disk/network, latency.
34641 Note that the CPU time printed is for @value{GDBN} only, it does not include
34642 the execution time of the inferior because there's no mechanism currently
34643 to compute how much time was spent by @value{GDBN} and how much time was
34644 spent by the program been debugged.
34645 This can also be requested by invoking @value{GDBN} with the
34646 @option{--statistics} command-line switch (@pxref{Mode Options}).
34648 @kindex maint translate-address
34649 @item maint translate-address @r{[}@var{section}@r{]} @var{addr}
34650 Find the symbol stored at the location specified by the address
34651 @var{addr} and an optional section name @var{section}. If found,
34652 @value{GDBN} prints the name of the closest symbol and an offset from
34653 the symbol's location to the specified address. This is similar to
34654 the @code{info address} command (@pxref{Symbols}), except that this
34655 command also allows to find symbols in other sections.
34657 If section was not specified, the section in which the symbol was found
34658 is also printed. For dynamically linked executables, the name of
34659 executable or shared library containing the symbol is printed as well.
34663 The following command is useful for non-interactive invocations of
34664 @value{GDBN}, such as in the test suite.
34667 @item set watchdog @var{nsec}
34668 @kindex set watchdog
34669 @cindex watchdog timer
34670 @cindex timeout for commands
34671 Set the maximum number of seconds @value{GDBN} will wait for the
34672 target operation to finish. If this time expires, @value{GDBN}
34673 reports and error and the command is aborted.
34675 @item show watchdog
34676 Show the current setting of the target wait timeout.
34679 @node Remote Protocol
34680 @appendix @value{GDBN} Remote Serial Protocol
34685 * Stop Reply Packets::
34686 * General Query Packets::
34687 * Architecture-Specific Protocol Details::
34688 * Tracepoint Packets::
34689 * Host I/O Packets::
34691 * Notification Packets::
34692 * Remote Non-Stop::
34693 * Packet Acknowledgment::
34695 * File-I/O Remote Protocol Extension::
34696 * Library List Format::
34697 * Library List Format for SVR4 Targets::
34698 * Memory Map Format::
34699 * Thread List Format::
34700 * Traceframe Info Format::
34706 There may be occasions when you need to know something about the
34707 protocol---for example, if there is only one serial port to your target
34708 machine, you might want your program to do something special if it
34709 recognizes a packet meant for @value{GDBN}.
34711 In the examples below, @samp{->} and @samp{<-} are used to indicate
34712 transmitted and received data, respectively.
34714 @cindex protocol, @value{GDBN} remote serial
34715 @cindex serial protocol, @value{GDBN} remote
34716 @cindex remote serial protocol
34717 All @value{GDBN} commands and responses (other than acknowledgments
34718 and notifications, see @ref{Notification Packets}) are sent as a
34719 @var{packet}. A @var{packet} is introduced with the character
34720 @samp{$}, the actual @var{packet-data}, and the terminating character
34721 @samp{#} followed by a two-digit @var{checksum}:
34724 @code{$}@var{packet-data}@code{#}@var{checksum}
34728 @cindex checksum, for @value{GDBN} remote
34730 The two-digit @var{checksum} is computed as the modulo 256 sum of all
34731 characters between the leading @samp{$} and the trailing @samp{#} (an
34732 eight bit unsigned checksum).
34734 Implementors should note that prior to @value{GDBN} 5.0 the protocol
34735 specification also included an optional two-digit @var{sequence-id}:
34738 @code{$}@var{sequence-id}@code{:}@var{packet-data}@code{#}@var{checksum}
34741 @cindex sequence-id, for @value{GDBN} remote
34743 That @var{sequence-id} was appended to the acknowledgment. @value{GDBN}
34744 has never output @var{sequence-id}s. Stubs that handle packets added
34745 since @value{GDBN} 5.0 must not accept @var{sequence-id}.
34747 When either the host or the target machine receives a packet, the first
34748 response expected is an acknowledgment: either @samp{+} (to indicate
34749 the package was received correctly) or @samp{-} (to request
34753 -> @code{$}@var{packet-data}@code{#}@var{checksum}
34758 The @samp{+}/@samp{-} acknowledgments can be disabled
34759 once a connection is established.
34760 @xref{Packet Acknowledgment}, for details.
34762 The host (@value{GDBN}) sends @var{command}s, and the target (the
34763 debugging stub incorporated in your program) sends a @var{response}. In
34764 the case of step and continue @var{command}s, the response is only sent
34765 when the operation has completed, and the target has again stopped all
34766 threads in all attached processes. This is the default all-stop mode
34767 behavior, but the remote protocol also supports @value{GDBN}'s non-stop
34768 execution mode; see @ref{Remote Non-Stop}, for details.
34770 @var{packet-data} consists of a sequence of characters with the
34771 exception of @samp{#} and @samp{$} (see @samp{X} packet for additional
34774 @cindex remote protocol, field separator
34775 Fields within the packet should be separated using @samp{,} @samp{;} or
34776 @samp{:}. Except where otherwise noted all numbers are represented in
34777 @sc{hex} with leading zeros suppressed.
34779 Implementors should note that prior to @value{GDBN} 5.0, the character
34780 @samp{:} could not appear as the third character in a packet (as it
34781 would potentially conflict with the @var{sequence-id}).
34783 @cindex remote protocol, binary data
34784 @anchor{Binary Data}
34785 Binary data in most packets is encoded either as two hexadecimal
34786 digits per byte of binary data. This allowed the traditional remote
34787 protocol to work over connections which were only seven-bit clean.
34788 Some packets designed more recently assume an eight-bit clean
34789 connection, and use a more efficient encoding to send and receive
34792 The binary data representation uses @code{7d} (@sc{ascii} @samp{@}})
34793 as an escape character. Any escaped byte is transmitted as the escape
34794 character followed by the original character XORed with @code{0x20}.
34795 For example, the byte @code{0x7d} would be transmitted as the two
34796 bytes @code{0x7d 0x5d}. The bytes @code{0x23} (@sc{ascii} @samp{#}),
34797 @code{0x24} (@sc{ascii} @samp{$}), and @code{0x7d} (@sc{ascii}
34798 @samp{@}}) must always be escaped. Responses sent by the stub
34799 must also escape @code{0x2a} (@sc{ascii} @samp{*}), so that it
34800 is not interpreted as the start of a run-length encoded sequence
34803 Response @var{data} can be run-length encoded to save space.
34804 Run-length encoding replaces runs of identical characters with one
34805 instance of the repeated character, followed by a @samp{*} and a
34806 repeat count. The repeat count is itself sent encoded, to avoid
34807 binary characters in @var{data}: a value of @var{n} is sent as
34808 @code{@var{n}+29}. For a repeat count greater or equal to 3, this
34809 produces a printable @sc{ascii} character, e.g.@: a space (@sc{ascii}
34810 code 32) for a repeat count of 3. (This is because run-length
34811 encoding starts to win for counts 3 or more.) Thus, for example,
34812 @samp{0* } is a run-length encoding of ``0000'': the space character
34813 after @samp{*} means repeat the leading @code{0} @w{@code{32 - 29 =
34816 The printable characters @samp{#} and @samp{$} or with a numeric value
34817 greater than 126 must not be used. Runs of six repeats (@samp{#}) or
34818 seven repeats (@samp{$}) can be expanded using a repeat count of only
34819 five (@samp{"}). For example, @samp{00000000} can be encoded as
34822 The error response returned for some packets includes a two character
34823 error number. That number is not well defined.
34825 @cindex empty response, for unsupported packets
34826 For any @var{command} not supported by the stub, an empty response
34827 (@samp{$#00}) should be returned. That way it is possible to extend the
34828 protocol. A newer @value{GDBN} can tell if a packet is supported based
34831 At a minimum, a stub is required to support the @samp{g} and @samp{G}
34832 commands for register access, and the @samp{m} and @samp{M} commands
34833 for memory access. Stubs that only control single-threaded targets
34834 can implement run control with the @samp{c} (continue), and @samp{s}
34835 (step) commands. Stubs that support multi-threading targets should
34836 support the @samp{vCont} command. All other commands are optional.
34841 The following table provides a complete list of all currently defined
34842 @var{command}s and their corresponding response @var{data}.
34843 @xref{File-I/O Remote Protocol Extension}, for details about the File
34844 I/O extension of the remote protocol.
34846 Each packet's description has a template showing the packet's overall
34847 syntax, followed by an explanation of the packet's meaning. We
34848 include spaces in some of the templates for clarity; these are not
34849 part of the packet's syntax. No @value{GDBN} packet uses spaces to
34850 separate its components. For example, a template like @samp{foo
34851 @var{bar} @var{baz}} describes a packet beginning with the three ASCII
34852 bytes @samp{foo}, followed by a @var{bar}, followed directly by a
34853 @var{baz}. @value{GDBN} does not transmit a space character between the
34854 @samp{foo} and the @var{bar}, or between the @var{bar} and the
34857 @cindex @var{thread-id}, in remote protocol
34858 @anchor{thread-id syntax}
34859 Several packets and replies include a @var{thread-id} field to identify
34860 a thread. Normally these are positive numbers with a target-specific
34861 interpretation, formatted as big-endian hex strings. A @var{thread-id}
34862 can also be a literal @samp{-1} to indicate all threads, or @samp{0} to
34865 In addition, the remote protocol supports a multiprocess feature in
34866 which the @var{thread-id} syntax is extended to optionally include both
34867 process and thread ID fields, as @samp{p@var{pid}.@var{tid}}.
34868 The @var{pid} (process) and @var{tid} (thread) components each have the
34869 format described above: a positive number with target-specific
34870 interpretation formatted as a big-endian hex string, literal @samp{-1}
34871 to indicate all processes or threads (respectively), or @samp{0} to
34872 indicate an arbitrary process or thread. Specifying just a process, as
34873 @samp{p@var{pid}}, is equivalent to @samp{p@var{pid}.-1}. It is an
34874 error to specify all processes but a specific thread, such as
34875 @samp{p-1.@var{tid}}. Note that the @samp{p} prefix is @emph{not} used
34876 for those packets and replies explicitly documented to include a process
34877 ID, rather than a @var{thread-id}.
34879 The multiprocess @var{thread-id} syntax extensions are only used if both
34880 @value{GDBN} and the stub report support for the @samp{multiprocess}
34881 feature using @samp{qSupported}. @xref{multiprocess extensions}, for
34884 Note that all packet forms beginning with an upper- or lower-case
34885 letter, other than those described here, are reserved for future use.
34887 Here are the packet descriptions.
34892 @cindex @samp{!} packet
34893 @anchor{extended mode}
34894 Enable extended mode. In extended mode, the remote server is made
34895 persistent. The @samp{R} packet is used to restart the program being
34901 The remote target both supports and has enabled extended mode.
34905 @cindex @samp{?} packet
34906 Indicate the reason the target halted. The reply is the same as for
34907 step and continue. This packet has a special interpretation when the
34908 target is in non-stop mode; see @ref{Remote Non-Stop}.
34911 @xref{Stop Reply Packets}, for the reply specifications.
34913 @item A @var{arglen},@var{argnum},@var{arg},@dots{}
34914 @cindex @samp{A} packet
34915 Initialized @code{argv[]} array passed into program. @var{arglen}
34916 specifies the number of bytes in the hex encoded byte stream
34917 @var{arg}. See @code{gdbserver} for more details.
34922 The arguments were set.
34928 @cindex @samp{b} packet
34929 (Don't use this packet; its behavior is not well-defined.)
34930 Change the serial line speed to @var{baud}.
34932 JTC: @emph{When does the transport layer state change? When it's
34933 received, or after the ACK is transmitted. In either case, there are
34934 problems if the command or the acknowledgment packet is dropped.}
34936 Stan: @emph{If people really wanted to add something like this, and get
34937 it working for the first time, they ought to modify ser-unix.c to send
34938 some kind of out-of-band message to a specially-setup stub and have the
34939 switch happen "in between" packets, so that from remote protocol's point
34940 of view, nothing actually happened.}
34942 @item B @var{addr},@var{mode}
34943 @cindex @samp{B} packet
34944 Set (@var{mode} is @samp{S}) or clear (@var{mode} is @samp{C}) a
34945 breakpoint at @var{addr}.
34947 Don't use this packet. Use the @samp{Z} and @samp{z} packets instead
34948 (@pxref{insert breakpoint or watchpoint packet}).
34950 @cindex @samp{bc} packet
34953 Backward continue. Execute the target system in reverse. No parameter.
34954 @xref{Reverse Execution}, for more information.
34957 @xref{Stop Reply Packets}, for the reply specifications.
34959 @cindex @samp{bs} packet
34962 Backward single step. Execute one instruction in reverse. No parameter.
34963 @xref{Reverse Execution}, for more information.
34966 @xref{Stop Reply Packets}, for the reply specifications.
34968 @item c @r{[}@var{addr}@r{]}
34969 @cindex @samp{c} packet
34970 Continue. @var{addr} is address to resume. If @var{addr} is omitted,
34971 resume at current address.
34973 This packet is deprecated for multi-threading support. @xref{vCont
34977 @xref{Stop Reply Packets}, for the reply specifications.
34979 @item C @var{sig}@r{[};@var{addr}@r{]}
34980 @cindex @samp{C} packet
34981 Continue with signal @var{sig} (hex signal number). If
34982 @samp{;@var{addr}} is omitted, resume at same address.
34984 This packet is deprecated for multi-threading support. @xref{vCont
34988 @xref{Stop Reply Packets}, for the reply specifications.
34991 @cindex @samp{d} packet
34994 Don't use this packet; instead, define a general set packet
34995 (@pxref{General Query Packets}).
34999 @cindex @samp{D} packet
35000 The first form of the packet is used to detach @value{GDBN} from the
35001 remote system. It is sent to the remote target
35002 before @value{GDBN} disconnects via the @code{detach} command.
35004 The second form, including a process ID, is used when multiprocess
35005 protocol extensions are enabled (@pxref{multiprocess extensions}), to
35006 detach only a specific process. The @var{pid} is specified as a
35007 big-endian hex string.
35017 @item F @var{RC},@var{EE},@var{CF};@var{XX}
35018 @cindex @samp{F} packet
35019 A reply from @value{GDBN} to an @samp{F} packet sent by the target.
35020 This is part of the File-I/O protocol extension. @xref{File-I/O
35021 Remote Protocol Extension}, for the specification.
35024 @anchor{read registers packet}
35025 @cindex @samp{g} packet
35026 Read general registers.
35030 @item @var{XX@dots{}}
35031 Each byte of register data is described by two hex digits. The bytes
35032 with the register are transmitted in target byte order. The size of
35033 each register and their position within the @samp{g} packet are
35034 determined by the @value{GDBN} internal gdbarch functions
35035 @code{DEPRECATED_REGISTER_RAW_SIZE} and @code{gdbarch_register_name}. The
35036 specification of several standard @samp{g} packets is specified below.
35038 When reading registers from a trace frame (@pxref{Analyze Collected
35039 Data,,Using the Collected Data}), the stub may also return a string of
35040 literal @samp{x}'s in place of the register data digits, to indicate
35041 that the corresponding register has not been collected, thus its value
35042 is unavailable. For example, for an architecture with 4 registers of
35043 4 bytes each, the following reply indicates to @value{GDBN} that
35044 registers 0 and 2 have not been collected, while registers 1 and 3
35045 have been collected, and both have zero value:
35049 <- @code{xxxxxxxx00000000xxxxxxxx00000000}
35056 @item G @var{XX@dots{}}
35057 @cindex @samp{G} packet
35058 Write general registers. @xref{read registers packet}, for a
35059 description of the @var{XX@dots{}} data.
35069 @item H @var{op} @var{thread-id}
35070 @cindex @samp{H} packet
35071 Set thread for subsequent operations (@samp{m}, @samp{M}, @samp{g},
35072 @samp{G}, et.al.). @var{op} depends on the operation to be performed:
35073 it should be @samp{c} for step and continue operations (note that this
35074 is deprecated, supporting the @samp{vCont} command is a better
35075 option), @samp{g} for other operations. The thread designator
35076 @var{thread-id} has the format and interpretation described in
35077 @ref{thread-id syntax}.
35088 @c 'H': How restrictive (or permissive) is the thread model. If a
35089 @c thread is selected and stopped, are other threads allowed
35090 @c to continue to execute? As I mentioned above, I think the
35091 @c semantics of each command when a thread is selected must be
35092 @c described. For example:
35094 @c 'g': If the stub supports threads and a specific thread is
35095 @c selected, returns the register block from that thread;
35096 @c otherwise returns current registers.
35098 @c 'G' If the stub supports threads and a specific thread is
35099 @c selected, sets the registers of the register block of
35100 @c that thread; otherwise sets current registers.
35102 @item i @r{[}@var{addr}@r{[},@var{nnn}@r{]]}
35103 @anchor{cycle step packet}
35104 @cindex @samp{i} packet
35105 Step the remote target by a single clock cycle. If @samp{,@var{nnn}} is
35106 present, cycle step @var{nnn} cycles. If @var{addr} is present, cycle
35107 step starting at that address.
35110 @cindex @samp{I} packet
35111 Signal, then cycle step. @xref{step with signal packet}. @xref{cycle
35115 @cindex @samp{k} packet
35118 FIXME: @emph{There is no description of how to operate when a specific
35119 thread context has been selected (i.e.@: does 'k' kill only that
35122 @item m @var{addr},@var{length}
35123 @cindex @samp{m} packet
35124 Read @var{length} bytes of memory starting at address @var{addr}.
35125 Note that @var{addr} may not be aligned to any particular boundary.
35127 The stub need not use any particular size or alignment when gathering
35128 data from memory for the response; even if @var{addr} is word-aligned
35129 and @var{length} is a multiple of the word size, the stub is free to
35130 use byte accesses, or not. For this reason, this packet may not be
35131 suitable for accessing memory-mapped I/O devices.
35132 @cindex alignment of remote memory accesses
35133 @cindex size of remote memory accesses
35134 @cindex memory, alignment and size of remote accesses
35138 @item @var{XX@dots{}}
35139 Memory contents; each byte is transmitted as a two-digit hexadecimal
35140 number. The reply may contain fewer bytes than requested if the
35141 server was able to read only part of the region of memory.
35146 @item M @var{addr},@var{length}:@var{XX@dots{}}
35147 @cindex @samp{M} packet
35148 Write @var{length} bytes of memory starting at address @var{addr}.
35149 @var{XX@dots{}} is the data; each byte is transmitted as a two-digit
35150 hexadecimal number.
35157 for an error (this includes the case where only part of the data was
35162 @cindex @samp{p} packet
35163 Read the value of register @var{n}; @var{n} is in hex.
35164 @xref{read registers packet}, for a description of how the returned
35165 register value is encoded.
35169 @item @var{XX@dots{}}
35170 the register's value
35174 Indicating an unrecognized @var{query}.
35177 @item P @var{n@dots{}}=@var{r@dots{}}
35178 @anchor{write register packet}
35179 @cindex @samp{P} packet
35180 Write register @var{n@dots{}} with value @var{r@dots{}}. The register
35181 number @var{n} is in hexadecimal, and @var{r@dots{}} contains two hex
35182 digits for each byte in the register (target byte order).
35192 @item q @var{name} @var{params}@dots{}
35193 @itemx Q @var{name} @var{params}@dots{}
35194 @cindex @samp{q} packet
35195 @cindex @samp{Q} packet
35196 General query (@samp{q}) and set (@samp{Q}). These packets are
35197 described fully in @ref{General Query Packets}.
35200 @cindex @samp{r} packet
35201 Reset the entire system.
35203 Don't use this packet; use the @samp{R} packet instead.
35206 @cindex @samp{R} packet
35207 Restart the program being debugged. @var{XX}, while needed, is ignored.
35208 This packet is only available in extended mode (@pxref{extended mode}).
35210 The @samp{R} packet has no reply.
35212 @item s @r{[}@var{addr}@r{]}
35213 @cindex @samp{s} packet
35214 Single step. @var{addr} is the address at which to resume. If
35215 @var{addr} is omitted, resume at same address.
35217 This packet is deprecated for multi-threading support. @xref{vCont
35221 @xref{Stop Reply Packets}, for the reply specifications.
35223 @item S @var{sig}@r{[};@var{addr}@r{]}
35224 @anchor{step with signal packet}
35225 @cindex @samp{S} packet
35226 Step with signal. This is analogous to the @samp{C} packet, but
35227 requests a single-step, rather than a normal resumption of execution.
35229 This packet is deprecated for multi-threading support. @xref{vCont
35233 @xref{Stop Reply Packets}, for the reply specifications.
35235 @item t @var{addr}:@var{PP},@var{MM}
35236 @cindex @samp{t} packet
35237 Search backwards starting at address @var{addr} for a match with pattern
35238 @var{PP} and mask @var{MM}. @var{PP} and @var{MM} are 4 bytes.
35239 @var{addr} must be at least 3 digits.
35241 @item T @var{thread-id}
35242 @cindex @samp{T} packet
35243 Find out if the thread @var{thread-id} is alive. @xref{thread-id syntax}.
35248 thread is still alive
35254 Packets starting with @samp{v} are identified by a multi-letter name,
35255 up to the first @samp{;} or @samp{?} (or the end of the packet).
35257 @item vAttach;@var{pid}
35258 @cindex @samp{vAttach} packet
35259 Attach to a new process with the specified process ID @var{pid}.
35260 The process ID is a
35261 hexadecimal integer identifying the process. In all-stop mode, all
35262 threads in the attached process are stopped; in non-stop mode, it may be
35263 attached without being stopped if that is supported by the target.
35265 @c In non-stop mode, on a successful vAttach, the stub should set the
35266 @c current thread to a thread of the newly-attached process. After
35267 @c attaching, GDB queries for the attached process's thread ID with qC.
35268 @c Also note that, from a user perspective, whether or not the
35269 @c target is stopped on attach in non-stop mode depends on whether you
35270 @c use the foreground or background version of the attach command, not
35271 @c on what vAttach does; GDB does the right thing with respect to either
35272 @c stopping or restarting threads.
35274 This packet is only available in extended mode (@pxref{extended mode}).
35280 @item @r{Any stop packet}
35281 for success in all-stop mode (@pxref{Stop Reply Packets})
35283 for success in non-stop mode (@pxref{Remote Non-Stop})
35286 @item vCont@r{[};@var{action}@r{[}:@var{thread-id}@r{]]}@dots{}
35287 @cindex @samp{vCont} packet
35288 @anchor{vCont packet}
35289 Resume the inferior, specifying different actions for each thread.
35290 If an action is specified with no @var{thread-id}, then it is applied to any
35291 threads that don't have a specific action specified; if no default action is
35292 specified then other threads should remain stopped in all-stop mode and
35293 in their current state in non-stop mode.
35294 Specifying multiple
35295 default actions is an error; specifying no actions is also an error.
35296 Thread IDs are specified using the syntax described in @ref{thread-id syntax}.
35298 Currently supported actions are:
35304 Continue with signal @var{sig}. The signal @var{sig} should be two hex digits.
35308 Step with signal @var{sig}. The signal @var{sig} should be two hex digits.
35313 The optional argument @var{addr} normally associated with the
35314 @samp{c}, @samp{C}, @samp{s}, and @samp{S} packets is
35315 not supported in @samp{vCont}.
35317 The @samp{t} action is only relevant in non-stop mode
35318 (@pxref{Remote Non-Stop}) and may be ignored by the stub otherwise.
35319 A stop reply should be generated for any affected thread not already stopped.
35320 When a thread is stopped by means of a @samp{t} action,
35321 the corresponding stop reply should indicate that the thread has stopped with
35322 signal @samp{0}, regardless of whether the target uses some other signal
35323 as an implementation detail.
35325 The stub must support @samp{vCont} if it reports support for
35326 multiprocess extensions (@pxref{multiprocess extensions}). Note that in
35327 this case @samp{vCont} actions can be specified to apply to all threads
35328 in a process by using the @samp{p@var{pid}.-1} form of the
35332 @xref{Stop Reply Packets}, for the reply specifications.
35335 @cindex @samp{vCont?} packet
35336 Request a list of actions supported by the @samp{vCont} packet.
35340 @item vCont@r{[};@var{action}@dots{}@r{]}
35341 The @samp{vCont} packet is supported. Each @var{action} is a supported
35342 command in the @samp{vCont} packet.
35344 The @samp{vCont} packet is not supported.
35347 @item vFile:@var{operation}:@var{parameter}@dots{}
35348 @cindex @samp{vFile} packet
35349 Perform a file operation on the target system. For details,
35350 see @ref{Host I/O Packets}.
35352 @item vFlashErase:@var{addr},@var{length}
35353 @cindex @samp{vFlashErase} packet
35354 Direct the stub to erase @var{length} bytes of flash starting at
35355 @var{addr}. The region may enclose any number of flash blocks, but
35356 its start and end must fall on block boundaries, as indicated by the
35357 flash block size appearing in the memory map (@pxref{Memory Map
35358 Format}). @value{GDBN} groups flash memory programming operations
35359 together, and sends a @samp{vFlashDone} request after each group; the
35360 stub is allowed to delay erase operation until the @samp{vFlashDone}
35361 packet is received.
35371 @item vFlashWrite:@var{addr}:@var{XX@dots{}}
35372 @cindex @samp{vFlashWrite} packet
35373 Direct the stub to write data to flash address @var{addr}. The data
35374 is passed in binary form using the same encoding as for the @samp{X}
35375 packet (@pxref{Binary Data}). The memory ranges specified by
35376 @samp{vFlashWrite} packets preceding a @samp{vFlashDone} packet must
35377 not overlap, and must appear in order of increasing addresses
35378 (although @samp{vFlashErase} packets for higher addresses may already
35379 have been received; the ordering is guaranteed only between
35380 @samp{vFlashWrite} packets). If a packet writes to an address that was
35381 neither erased by a preceding @samp{vFlashErase} packet nor by some other
35382 target-specific method, the results are unpredictable.
35390 for vFlashWrite addressing non-flash memory
35396 @cindex @samp{vFlashDone} packet
35397 Indicate to the stub that flash programming operation is finished.
35398 The stub is permitted to delay or batch the effects of a group of
35399 @samp{vFlashErase} and @samp{vFlashWrite} packets until a
35400 @samp{vFlashDone} packet is received. The contents of the affected
35401 regions of flash memory are unpredictable until the @samp{vFlashDone}
35402 request is completed.
35404 @item vKill;@var{pid}
35405 @cindex @samp{vKill} packet
35406 Kill the process with the specified process ID. @var{pid} is a
35407 hexadecimal integer identifying the process. This packet is used in
35408 preference to @samp{k} when multiprocess protocol extensions are
35409 supported; see @ref{multiprocess extensions}.
35419 @item vRun;@var{filename}@r{[};@var{argument}@r{]}@dots{}
35420 @cindex @samp{vRun} packet
35421 Run the program @var{filename}, passing it each @var{argument} on its
35422 command line. The file and arguments are hex-encoded strings. If
35423 @var{filename} is an empty string, the stub may use a default program
35424 (e.g.@: the last program run). The program is created in the stopped
35427 @c FIXME: What about non-stop mode?
35429 This packet is only available in extended mode (@pxref{extended mode}).
35435 @item @r{Any stop packet}
35436 for success (@pxref{Stop Reply Packets})
35440 @anchor{vStopped packet}
35441 @cindex @samp{vStopped} packet
35443 In non-stop mode (@pxref{Remote Non-Stop}), acknowledge a previous stop
35444 reply and prompt for the stub to report another one.
35448 @item @r{Any stop packet}
35449 if there is another unreported stop event (@pxref{Stop Reply Packets})
35451 if there are no unreported stop events
35454 @item X @var{addr},@var{length}:@var{XX@dots{}}
35456 @cindex @samp{X} packet
35457 Write data to memory, where the data is transmitted in binary.
35458 @var{addr} is address, @var{length} is number of bytes,
35459 @samp{@var{XX}@dots{}} is binary data (@pxref{Binary Data}).
35469 @item z @var{type},@var{addr},@var{kind}
35470 @itemx Z @var{type},@var{addr},@var{kind}
35471 @anchor{insert breakpoint or watchpoint packet}
35472 @cindex @samp{z} packet
35473 @cindex @samp{Z} packets
35474 Insert (@samp{Z}) or remove (@samp{z}) a @var{type} breakpoint or
35475 watchpoint starting at address @var{address} of kind @var{kind}.
35477 Each breakpoint and watchpoint packet @var{type} is documented
35480 @emph{Implementation notes: A remote target shall return an empty string
35481 for an unrecognized breakpoint or watchpoint packet @var{type}. A
35482 remote target shall support either both or neither of a given
35483 @samp{Z@var{type}@dots{}} and @samp{z@var{type}@dots{}} packet pair. To
35484 avoid potential problems with duplicate packets, the operations should
35485 be implemented in an idempotent way.}
35487 @item z0,@var{addr},@var{kind}
35488 @itemx Z0,@var{addr},@var{kind}@r{[};@var{cond_list}@dots{}@r{]}
35489 @cindex @samp{z0} packet
35490 @cindex @samp{Z0} packet
35491 Insert (@samp{Z0}) or remove (@samp{z0}) a memory breakpoint at address
35492 @var{addr} of type @var{kind}.
35494 A memory breakpoint is implemented by replacing the instruction at
35495 @var{addr} with a software breakpoint or trap instruction. The
35496 @var{kind} is target-specific and typically indicates the size of
35497 the breakpoint in bytes that should be inserted. E.g., the @sc{arm}
35498 and @sc{mips} can insert either a 2 or 4 byte breakpoint. Some
35499 architectures have additional meanings for @var{kind};
35500 @var{cond_list} is an optional list of conditional expressions in bytecode
35501 form that should be evaluated on the target's side. These are the
35502 conditions that should be taken into consideration when deciding if
35503 the breakpoint trigger should be reported back to @var{GDBN}.
35505 The @var{cond_list} parameter is comprised of a series of expressions,
35506 concatenated without separators. Each expression has the following form:
35510 @item X @var{len},@var{expr}
35511 @var{len} is the length of the bytecode expression and @var{expr} is the
35512 actual conditional expression in bytecode form.
35516 see @ref{Architecture-Specific Protocol Details}.
35518 @emph{Implementation note: It is possible for a target to copy or move
35519 code that contains memory breakpoints (e.g., when implementing
35520 overlays). The behavior of this packet, in the presence of such a
35521 target, is not defined.}
35533 @item z1,@var{addr},@var{kind}
35534 @itemx Z1,@var{addr},@var{kind}@r{[};@var{cond_list}@dots{}@r{]}
35535 @cindex @samp{z1} packet
35536 @cindex @samp{Z1} packet
35537 Insert (@samp{Z1}) or remove (@samp{z1}) a hardware breakpoint at
35538 address @var{addr}.
35540 A hardware breakpoint is implemented using a mechanism that is not
35541 dependant on being able to modify the target's memory. @var{kind}
35542 and @var{cond_list} have the same meaning as in @samp{Z0} packets.
35544 @emph{Implementation note: A hardware breakpoint is not affected by code
35557 @item z2,@var{addr},@var{kind}
35558 @itemx Z2,@var{addr},@var{kind}
35559 @cindex @samp{z2} packet
35560 @cindex @samp{Z2} packet
35561 Insert (@samp{Z2}) or remove (@samp{z2}) a write watchpoint at @var{addr}.
35562 @var{kind} is interpreted as the number of bytes to watch.
35574 @item z3,@var{addr},@var{kind}
35575 @itemx Z3,@var{addr},@var{kind}
35576 @cindex @samp{z3} packet
35577 @cindex @samp{Z3} packet
35578 Insert (@samp{Z3}) or remove (@samp{z3}) a read watchpoint at @var{addr}.
35579 @var{kind} is interpreted as the number of bytes to watch.
35591 @item z4,@var{addr},@var{kind}
35592 @itemx Z4,@var{addr},@var{kind}
35593 @cindex @samp{z4} packet
35594 @cindex @samp{Z4} packet
35595 Insert (@samp{Z4}) or remove (@samp{z4}) an access watchpoint at @var{addr}.
35596 @var{kind} is interpreted as the number of bytes to watch.
35610 @node Stop Reply Packets
35611 @section Stop Reply Packets
35612 @cindex stop reply packets
35614 The @samp{C}, @samp{c}, @samp{S}, @samp{s}, @samp{vCont},
35615 @samp{vAttach}, @samp{vRun}, @samp{vStopped}, and @samp{?} packets can
35616 receive any of the below as a reply. Except for @samp{?}
35617 and @samp{vStopped}, that reply is only returned
35618 when the target halts. In the below the exact meaning of @dfn{signal
35619 number} is defined by the header @file{include/gdb/signals.h} in the
35620 @value{GDBN} source code.
35622 As in the description of request packets, we include spaces in the
35623 reply templates for clarity; these are not part of the reply packet's
35624 syntax. No @value{GDBN} stop reply packet uses spaces to separate its
35630 The program received signal number @var{AA} (a two-digit hexadecimal
35631 number). This is equivalent to a @samp{T} response with no
35632 @var{n}:@var{r} pairs.
35634 @item T @var{AA} @var{n1}:@var{r1};@var{n2}:@var{r2};@dots{}
35635 @cindex @samp{T} packet reply
35636 The program received signal number @var{AA} (a two-digit hexadecimal
35637 number). This is equivalent to an @samp{S} response, except that the
35638 @samp{@var{n}:@var{r}} pairs can carry values of important registers
35639 and other information directly in the stop reply packet, reducing
35640 round-trip latency. Single-step and breakpoint traps are reported
35641 this way. Each @samp{@var{n}:@var{r}} pair is interpreted as follows:
35645 If @var{n} is a hexadecimal number, it is a register number, and the
35646 corresponding @var{r} gives that register's value. @var{r} is a
35647 series of bytes in target byte order, with each byte given by a
35648 two-digit hex number.
35651 If @var{n} is @samp{thread}, then @var{r} is the @var{thread-id} of
35652 the stopped thread, as specified in @ref{thread-id syntax}.
35655 If @var{n} is @samp{core}, then @var{r} is the hexadecimal number of
35656 the core on which the stop event was detected.
35659 If @var{n} is a recognized @dfn{stop reason}, it describes a more
35660 specific event that stopped the target. The currently defined stop
35661 reasons are listed below. @var{aa} should be @samp{05}, the trap
35662 signal. At most one stop reason should be present.
35665 Otherwise, @value{GDBN} should ignore this @samp{@var{n}:@var{r}} pair
35666 and go on to the next; this allows us to extend the protocol in the
35670 The currently defined stop reasons are:
35676 The packet indicates a watchpoint hit, and @var{r} is the data address, in
35679 @cindex shared library events, remote reply
35681 The packet indicates that the loaded libraries have changed.
35682 @value{GDBN} should use @samp{qXfer:libraries:read} to fetch a new
35683 list of loaded libraries. @var{r} is ignored.
35685 @cindex replay log events, remote reply
35687 The packet indicates that the target cannot continue replaying
35688 logged execution events, because it has reached the end (or the
35689 beginning when executing backward) of the log. The value of @var{r}
35690 will be either @samp{begin} or @samp{end}. @xref{Reverse Execution},
35691 for more information.
35695 @itemx W @var{AA} ; process:@var{pid}
35696 The process exited, and @var{AA} is the exit status. This is only
35697 applicable to certain targets.
35699 The second form of the response, including the process ID of the exited
35700 process, can be used only when @value{GDBN} has reported support for
35701 multiprocess protocol extensions; see @ref{multiprocess extensions}.
35702 The @var{pid} is formatted as a big-endian hex string.
35705 @itemx X @var{AA} ; process:@var{pid}
35706 The process terminated with signal @var{AA}.
35708 The second form of the response, including the process ID of the
35709 terminated process, can be used only when @value{GDBN} has reported
35710 support for multiprocess protocol extensions; see @ref{multiprocess
35711 extensions}. The @var{pid} is formatted as a big-endian hex string.
35713 @item O @var{XX}@dots{}
35714 @samp{@var{XX}@dots{}} is hex encoding of @sc{ascii} data, to be
35715 written as the program's console output. This can happen at any time
35716 while the program is running and the debugger should continue to wait
35717 for @samp{W}, @samp{T}, etc. This reply is not permitted in non-stop mode.
35719 @item F @var{call-id},@var{parameter}@dots{}
35720 @var{call-id} is the identifier which says which host system call should
35721 be called. This is just the name of the function. Translation into the
35722 correct system call is only applicable as it's defined in @value{GDBN}.
35723 @xref{File-I/O Remote Protocol Extension}, for a list of implemented
35726 @samp{@var{parameter}@dots{}} is a list of parameters as defined for
35727 this very system call.
35729 The target replies with this packet when it expects @value{GDBN} to
35730 call a host system call on behalf of the target. @value{GDBN} replies
35731 with an appropriate @samp{F} packet and keeps up waiting for the next
35732 reply packet from the target. The latest @samp{C}, @samp{c}, @samp{S}
35733 or @samp{s} action is expected to be continued. @xref{File-I/O Remote
35734 Protocol Extension}, for more details.
35738 @node General Query Packets
35739 @section General Query Packets
35740 @cindex remote query requests
35742 Packets starting with @samp{q} are @dfn{general query packets};
35743 packets starting with @samp{Q} are @dfn{general set packets}. General
35744 query and set packets are a semi-unified form for retrieving and
35745 sending information to and from the stub.
35747 The initial letter of a query or set packet is followed by a name
35748 indicating what sort of thing the packet applies to. For example,
35749 @value{GDBN} may use a @samp{qSymbol} packet to exchange symbol
35750 definitions with the stub. These packet names follow some
35755 The name must not contain commas, colons or semicolons.
35757 Most @value{GDBN} query and set packets have a leading upper case
35760 The names of custom vendor packets should use a company prefix, in
35761 lower case, followed by a period. For example, packets designed at
35762 the Acme Corporation might begin with @samp{qacme.foo} (for querying
35763 foos) or @samp{Qacme.bar} (for setting bars).
35766 The name of a query or set packet should be separated from any
35767 parameters by a @samp{:}; the parameters themselves should be
35768 separated by @samp{,} or @samp{;}. Stubs must be careful to match the
35769 full packet name, and check for a separator or the end of the packet,
35770 in case two packet names share a common prefix. New packets should not begin
35771 with @samp{qC}, @samp{qP}, or @samp{qL}@footnote{The @samp{qP} and @samp{qL}
35772 packets predate these conventions, and have arguments without any terminator
35773 for the packet name; we suspect they are in widespread use in places that
35774 are difficult to upgrade. The @samp{qC} packet has no arguments, but some
35775 existing stubs (e.g.@: RedBoot) are known to not check for the end of the
35778 Like the descriptions of the other packets, each description here
35779 has a template showing the packet's overall syntax, followed by an
35780 explanation of the packet's meaning. We include spaces in some of the
35781 templates for clarity; these are not part of the packet's syntax. No
35782 @value{GDBN} packet uses spaces to separate its components.
35784 Here are the currently defined query and set packets:
35790 Turn on or off the agent as a helper to perform some debugging operations
35791 delegated from @value{GDBN} (@pxref{Control Agent}).
35793 @item QAllow:@var{op}:@var{val}@dots{}
35794 @cindex @samp{QAllow} packet
35795 Specify which operations @value{GDBN} expects to request of the
35796 target, as a semicolon-separated list of operation name and value
35797 pairs. Possible values for @var{op} include @samp{WriteReg},
35798 @samp{WriteMem}, @samp{InsertBreak}, @samp{InsertTrace},
35799 @samp{InsertFastTrace}, and @samp{Stop}. @var{val} is either 0,
35800 indicating that @value{GDBN} will not request the operation, or 1,
35801 indicating that it may. (The target can then use this to set up its
35802 own internals optimally, for instance if the debugger never expects to
35803 insert breakpoints, it may not need to install its own trap handler.)
35806 @cindex current thread, remote request
35807 @cindex @samp{qC} packet
35808 Return the current thread ID.
35812 @item QC @var{thread-id}
35813 Where @var{thread-id} is a thread ID as documented in
35814 @ref{thread-id syntax}.
35815 @item @r{(anything else)}
35816 Any other reply implies the old thread ID.
35819 @item qCRC:@var{addr},@var{length}
35820 @cindex CRC of memory block, remote request
35821 @cindex @samp{qCRC} packet
35822 Compute the CRC checksum of a block of memory using CRC-32 defined in
35823 IEEE 802.3. The CRC is computed byte at a time, taking the most
35824 significant bit of each byte first. The initial pattern code
35825 @code{0xffffffff} is used to ensure leading zeros affect the CRC.
35827 @emph{Note:} This is the same CRC used in validating separate debug
35828 files (@pxref{Separate Debug Files, , Debugging Information in Separate
35829 Files}). However the algorithm is slightly different. When validating
35830 separate debug files, the CRC is computed taking the @emph{least}
35831 significant bit of each byte first, and the final result is inverted to
35832 detect trailing zeros.
35837 An error (such as memory fault)
35838 @item C @var{crc32}
35839 The specified memory region's checksum is @var{crc32}.
35842 @item QDisableRandomization:@var{value}
35843 @cindex disable address space randomization, remote request
35844 @cindex @samp{QDisableRandomization} packet
35845 Some target operating systems will randomize the virtual address space
35846 of the inferior process as a security feature, but provide a feature
35847 to disable such randomization, e.g.@: to allow for a more deterministic
35848 debugging experience. On such systems, this packet with a @var{value}
35849 of 1 directs the target to disable address space randomization for
35850 processes subsequently started via @samp{vRun} packets, while a packet
35851 with a @var{value} of 0 tells the target to enable address space
35854 This packet is only available in extended mode (@pxref{extended mode}).
35859 The request succeeded.
35862 An error occurred. @var{nn} are hex digits.
35865 An empty reply indicates that @samp{QDisableRandomization} is not supported
35869 This packet is not probed by default; the remote stub must request it,
35870 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
35871 This should only be done on targets that actually support disabling
35872 address space randomization.
35875 @itemx qsThreadInfo
35876 @cindex list active threads, remote request
35877 @cindex @samp{qfThreadInfo} packet
35878 @cindex @samp{qsThreadInfo} packet
35879 Obtain a list of all active thread IDs from the target (OS). Since there
35880 may be too many active threads to fit into one reply packet, this query
35881 works iteratively: it may require more than one query/reply sequence to
35882 obtain the entire list of threads. The first query of the sequence will
35883 be the @samp{qfThreadInfo} query; subsequent queries in the
35884 sequence will be the @samp{qsThreadInfo} query.
35886 NOTE: This packet replaces the @samp{qL} query (see below).
35890 @item m @var{thread-id}
35892 @item m @var{thread-id},@var{thread-id}@dots{}
35893 a comma-separated list of thread IDs
35895 (lower case letter @samp{L}) denotes end of list.
35898 In response to each query, the target will reply with a list of one or
35899 more thread IDs, separated by commas.
35900 @value{GDBN} will respond to each reply with a request for more thread
35901 ids (using the @samp{qs} form of the query), until the target responds
35902 with @samp{l} (lower-case ell, for @dfn{last}).
35903 Refer to @ref{thread-id syntax}, for the format of the @var{thread-id}
35906 @item qGetTLSAddr:@var{thread-id},@var{offset},@var{lm}
35907 @cindex get thread-local storage address, remote request
35908 @cindex @samp{qGetTLSAddr} packet
35909 Fetch the address associated with thread local storage specified
35910 by @var{thread-id}, @var{offset}, and @var{lm}.
35912 @var{thread-id} is the thread ID associated with the
35913 thread for which to fetch the TLS address. @xref{thread-id syntax}.
35915 @var{offset} is the (big endian, hex encoded) offset associated with the
35916 thread local variable. (This offset is obtained from the debug
35917 information associated with the variable.)
35919 @var{lm} is the (big endian, hex encoded) OS/ABI-specific encoding of the
35920 load module associated with the thread local storage. For example,
35921 a @sc{gnu}/Linux system will pass the link map address of the shared
35922 object associated with the thread local storage under consideration.
35923 Other operating environments may choose to represent the load module
35924 differently, so the precise meaning of this parameter will vary.
35928 @item @var{XX}@dots{}
35929 Hex encoded (big endian) bytes representing the address of the thread
35930 local storage requested.
35933 An error occurred. @var{nn} are hex digits.
35936 An empty reply indicates that @samp{qGetTLSAddr} is not supported by the stub.
35939 @item qGetTIBAddr:@var{thread-id}
35940 @cindex get thread information block address
35941 @cindex @samp{qGetTIBAddr} packet
35942 Fetch address of the Windows OS specific Thread Information Block.
35944 @var{thread-id} is the thread ID associated with the thread.
35948 @item @var{XX}@dots{}
35949 Hex encoded (big endian) bytes representing the linear address of the
35950 thread information block.
35953 An error occured. This means that either the thread was not found, or the
35954 address could not be retrieved.
35957 An empty reply indicates that @samp{qGetTIBAddr} is not supported by the stub.
35960 @item qL @var{startflag} @var{threadcount} @var{nextthread}
35961 Obtain thread information from RTOS. Where: @var{startflag} (one hex
35962 digit) is one to indicate the first query and zero to indicate a
35963 subsequent query; @var{threadcount} (two hex digits) is the maximum
35964 number of threads the response packet can contain; and @var{nextthread}
35965 (eight hex digits), for subsequent queries (@var{startflag} is zero), is
35966 returned in the response as @var{argthread}.
35968 Don't use this packet; use the @samp{qfThreadInfo} query instead (see above).
35972 @item qM @var{count} @var{done} @var{argthread} @var{thread}@dots{}
35973 Where: @var{count} (two hex digits) is the number of threads being
35974 returned; @var{done} (one hex digit) is zero to indicate more threads
35975 and one indicates no further threads; @var{argthreadid} (eight hex
35976 digits) is @var{nextthread} from the request packet; @var{thread}@dots{}
35977 is a sequence of thread IDs from the target. @var{threadid} (eight hex
35978 digits). See @code{remote.c:parse_threadlist_response()}.
35982 @cindex section offsets, remote request
35983 @cindex @samp{qOffsets} packet
35984 Get section offsets that the target used when relocating the downloaded
35989 @item Text=@var{xxx};Data=@var{yyy}@r{[};Bss=@var{zzz}@r{]}
35990 Relocate the @code{Text} section by @var{xxx} from its original address.
35991 Relocate the @code{Data} section by @var{yyy} from its original address.
35992 If the object file format provides segment information (e.g.@: @sc{elf}
35993 @samp{PT_LOAD} program headers), @value{GDBN} will relocate entire
35994 segments by the supplied offsets.
35996 @emph{Note: while a @code{Bss} offset may be included in the response,
35997 @value{GDBN} ignores this and instead applies the @code{Data} offset
35998 to the @code{Bss} section.}
36000 @item TextSeg=@var{xxx}@r{[};DataSeg=@var{yyy}@r{]}
36001 Relocate the first segment of the object file, which conventionally
36002 contains program code, to a starting address of @var{xxx}. If
36003 @samp{DataSeg} is specified, relocate the second segment, which
36004 conventionally contains modifiable data, to a starting address of
36005 @var{yyy}. @value{GDBN} will report an error if the object file
36006 does not contain segment information, or does not contain at least
36007 as many segments as mentioned in the reply. Extra segments are
36008 kept at fixed offsets relative to the last relocated segment.
36011 @item qP @var{mode} @var{thread-id}
36012 @cindex thread information, remote request
36013 @cindex @samp{qP} packet
36014 Returns information on @var{thread-id}. Where: @var{mode} is a hex
36015 encoded 32 bit mode; @var{thread-id} is a thread ID
36016 (@pxref{thread-id syntax}).
36018 Don't use this packet; use the @samp{qThreadExtraInfo} query instead
36021 Reply: see @code{remote.c:remote_unpack_thread_info_response()}.
36025 @cindex non-stop mode, remote request
36026 @cindex @samp{QNonStop} packet
36028 Enter non-stop (@samp{QNonStop:1}) or all-stop (@samp{QNonStop:0}) mode.
36029 @xref{Remote Non-Stop}, for more information.
36034 The request succeeded.
36037 An error occurred. @var{nn} are hex digits.
36040 An empty reply indicates that @samp{QNonStop} is not supported by
36044 This packet is not probed by default; the remote stub must request it,
36045 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
36046 Use of this packet is controlled by the @code{set non-stop} command;
36047 @pxref{Non-Stop Mode}.
36049 @item QPassSignals: @var{signal} @r{[};@var{signal}@r{]}@dots{}
36050 @cindex pass signals to inferior, remote request
36051 @cindex @samp{QPassSignals} packet
36052 @anchor{QPassSignals}
36053 Each listed @var{signal} should be passed directly to the inferior process.
36054 Signals are numbered identically to continue packets and stop replies
36055 (@pxref{Stop Reply Packets}). Each @var{signal} list item should be
36056 strictly greater than the previous item. These signals do not need to stop
36057 the inferior, or be reported to @value{GDBN}. All other signals should be
36058 reported to @value{GDBN}. Multiple @samp{QPassSignals} packets do not
36059 combine; any earlier @samp{QPassSignals} list is completely replaced by the
36060 new list. This packet improves performance when using @samp{handle
36061 @var{signal} nostop noprint pass}.
36066 The request succeeded.
36069 An error occurred. @var{nn} are hex digits.
36072 An empty reply indicates that @samp{QPassSignals} is not supported by
36076 Use of this packet is controlled by the @code{set remote pass-signals}
36077 command (@pxref{Remote Configuration, set remote pass-signals}).
36078 This packet is not probed by default; the remote stub must request it,
36079 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
36081 @item QProgramSignals: @var{signal} @r{[};@var{signal}@r{]}@dots{}
36082 @cindex signals the inferior may see, remote request
36083 @cindex @samp{QProgramSignals} packet
36084 @anchor{QProgramSignals}
36085 Each listed @var{signal} may be delivered to the inferior process.
36086 Others should be silently discarded.
36088 In some cases, the remote stub may need to decide whether to deliver a
36089 signal to the program or not without @value{GDBN} involvement. One
36090 example of that is while detaching --- the program's threads may have
36091 stopped for signals that haven't yet had a chance of being reported to
36092 @value{GDBN}, and so the remote stub can use the signal list specified
36093 by this packet to know whether to deliver or ignore those pending
36096 This does not influence whether to deliver a signal as requested by a
36097 resumption packet (@pxref{vCont packet}).
36099 Signals are numbered identically to continue packets and stop replies
36100 (@pxref{Stop Reply Packets}). Each @var{signal} list item should be
36101 strictly greater than the previous item. Multiple
36102 @samp{QProgramSignals} packets do not combine; any earlier
36103 @samp{QProgramSignals} list is completely replaced by the new list.
36108 The request succeeded.
36111 An error occurred. @var{nn} are hex digits.
36114 An empty reply indicates that @samp{QProgramSignals} is not supported
36118 Use of this packet is controlled by the @code{set remote program-signals}
36119 command (@pxref{Remote Configuration, set remote program-signals}).
36120 This packet is not probed by default; the remote stub must request it,
36121 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
36123 @item qRcmd,@var{command}
36124 @cindex execute remote command, remote request
36125 @cindex @samp{qRcmd} packet
36126 @var{command} (hex encoded) is passed to the local interpreter for
36127 execution. Invalid commands should be reported using the output
36128 string. Before the final result packet, the target may also respond
36129 with a number of intermediate @samp{O@var{output}} console output
36130 packets. @emph{Implementors should note that providing access to a
36131 stubs's interpreter may have security implications}.
36136 A command response with no output.
36138 A command response with the hex encoded output string @var{OUTPUT}.
36140 Indicate a badly formed request.
36142 An empty reply indicates that @samp{qRcmd} is not recognized.
36145 (Note that the @code{qRcmd} packet's name is separated from the
36146 command by a @samp{,}, not a @samp{:}, contrary to the naming
36147 conventions above. Please don't use this packet as a model for new
36150 @item qSearch:memory:@var{address};@var{length};@var{search-pattern}
36151 @cindex searching memory, in remote debugging
36152 @cindex @samp{qSearch:memory} packet
36153 @anchor{qSearch memory}
36154 Search @var{length} bytes at @var{address} for @var{search-pattern}.
36155 @var{address} and @var{length} are encoded in hex.
36156 @var{search-pattern} is a sequence of bytes, hex encoded.
36161 The pattern was not found.
36163 The pattern was found at @var{address}.
36165 A badly formed request or an error was encountered while searching memory.
36167 An empty reply indicates that @samp{qSearch:memory} is not recognized.
36170 @item QStartNoAckMode
36171 @cindex @samp{QStartNoAckMode} packet
36172 @anchor{QStartNoAckMode}
36173 Request that the remote stub disable the normal @samp{+}/@samp{-}
36174 protocol acknowledgments (@pxref{Packet Acknowledgment}).
36179 The stub has switched to no-acknowledgment mode.
36180 @value{GDBN} acknowledges this reponse,
36181 but neither the stub nor @value{GDBN} shall send or expect further
36182 @samp{+}/@samp{-} acknowledgments in the current connection.
36184 An empty reply indicates that the stub does not support no-acknowledgment mode.
36187 @item qSupported @r{[}:@var{gdbfeature} @r{[};@var{gdbfeature}@r{]}@dots{} @r{]}
36188 @cindex supported packets, remote query
36189 @cindex features of the remote protocol
36190 @cindex @samp{qSupported} packet
36191 @anchor{qSupported}
36192 Tell the remote stub about features supported by @value{GDBN}, and
36193 query the stub for features it supports. This packet allows
36194 @value{GDBN} and the remote stub to take advantage of each others'
36195 features. @samp{qSupported} also consolidates multiple feature probes
36196 at startup, to improve @value{GDBN} performance---a single larger
36197 packet performs better than multiple smaller probe packets on
36198 high-latency links. Some features may enable behavior which must not
36199 be on by default, e.g.@: because it would confuse older clients or
36200 stubs. Other features may describe packets which could be
36201 automatically probed for, but are not. These features must be
36202 reported before @value{GDBN} will use them. This ``default
36203 unsupported'' behavior is not appropriate for all packets, but it
36204 helps to keep the initial connection time under control with new
36205 versions of @value{GDBN} which support increasing numbers of packets.
36209 @item @var{stubfeature} @r{[};@var{stubfeature}@r{]}@dots{}
36210 The stub supports or does not support each returned @var{stubfeature},
36211 depending on the form of each @var{stubfeature} (see below for the
36214 An empty reply indicates that @samp{qSupported} is not recognized,
36215 or that no features needed to be reported to @value{GDBN}.
36218 The allowed forms for each feature (either a @var{gdbfeature} in the
36219 @samp{qSupported} packet, or a @var{stubfeature} in the response)
36223 @item @var{name}=@var{value}
36224 The remote protocol feature @var{name} is supported, and associated
36225 with the specified @var{value}. The format of @var{value} depends
36226 on the feature, but it must not include a semicolon.
36228 The remote protocol feature @var{name} is supported, and does not
36229 need an associated value.
36231 The remote protocol feature @var{name} is not supported.
36233 The remote protocol feature @var{name} may be supported, and
36234 @value{GDBN} should auto-detect support in some other way when it is
36235 needed. This form will not be used for @var{gdbfeature} notifications,
36236 but may be used for @var{stubfeature} responses.
36239 Whenever the stub receives a @samp{qSupported} request, the
36240 supplied set of @value{GDBN} features should override any previous
36241 request. This allows @value{GDBN} to put the stub in a known
36242 state, even if the stub had previously been communicating with
36243 a different version of @value{GDBN}.
36245 The following values of @var{gdbfeature} (for the packet sent by @value{GDBN})
36250 This feature indicates whether @value{GDBN} supports multiprocess
36251 extensions to the remote protocol. @value{GDBN} does not use such
36252 extensions unless the stub also reports that it supports them by
36253 including @samp{multiprocess+} in its @samp{qSupported} reply.
36254 @xref{multiprocess extensions}, for details.
36257 This feature indicates that @value{GDBN} supports the XML target
36258 description. If the stub sees @samp{xmlRegisters=} with target
36259 specific strings separated by a comma, it will report register
36263 This feature indicates whether @value{GDBN} supports the
36264 @samp{qRelocInsn} packet (@pxref{Tracepoint Packets,,Relocate
36265 instruction reply packet}).
36268 Stubs should ignore any unknown values for
36269 @var{gdbfeature}. Any @value{GDBN} which sends a @samp{qSupported}
36270 packet supports receiving packets of unlimited length (earlier
36271 versions of @value{GDBN} may reject overly long responses). Additional values
36272 for @var{gdbfeature} may be defined in the future to let the stub take
36273 advantage of new features in @value{GDBN}, e.g.@: incompatible
36274 improvements in the remote protocol---the @samp{multiprocess} feature is
36275 an example of such a feature. The stub's reply should be independent
36276 of the @var{gdbfeature} entries sent by @value{GDBN}; first @value{GDBN}
36277 describes all the features it supports, and then the stub replies with
36278 all the features it supports.
36280 Similarly, @value{GDBN} will silently ignore unrecognized stub feature
36281 responses, as long as each response uses one of the standard forms.
36283 Some features are flags. A stub which supports a flag feature
36284 should respond with a @samp{+} form response. Other features
36285 require values, and the stub should respond with an @samp{=}
36288 Each feature has a default value, which @value{GDBN} will use if
36289 @samp{qSupported} is not available or if the feature is not mentioned
36290 in the @samp{qSupported} response. The default values are fixed; a
36291 stub is free to omit any feature responses that match the defaults.
36293 Not all features can be probed, but for those which can, the probing
36294 mechanism is useful: in some cases, a stub's internal
36295 architecture may not allow the protocol layer to know some information
36296 about the underlying target in advance. This is especially common in
36297 stubs which may be configured for multiple targets.
36299 These are the currently defined stub features and their properties:
36301 @multitable @columnfractions 0.35 0.2 0.12 0.2
36302 @c NOTE: The first row should be @headitem, but we do not yet require
36303 @c a new enough version of Texinfo (4.7) to use @headitem.
36305 @tab Value Required
36309 @item @samp{PacketSize}
36314 @item @samp{qXfer:auxv:read}
36319 @item @samp{qXfer:features:read}
36324 @item @samp{qXfer:libraries:read}
36329 @item @samp{qXfer:memory-map:read}
36334 @item @samp{qXfer:sdata:read}
36339 @item @samp{qXfer:spu:read}
36344 @item @samp{qXfer:spu:write}
36349 @item @samp{qXfer:siginfo:read}
36354 @item @samp{qXfer:siginfo:write}
36359 @item @samp{qXfer:threads:read}
36364 @item @samp{qXfer:traceframe-info:read}
36369 @item @samp{qXfer:uib:read}
36374 @item @samp{qXfer:fdpic:read}
36379 @item @samp{QNonStop}
36384 @item @samp{QPassSignals}
36389 @item @samp{QStartNoAckMode}
36394 @item @samp{multiprocess}
36399 @item @samp{ConditionalBreakpoints}
36404 @item @samp{ConditionalTracepoints}
36409 @item @samp{ReverseContinue}
36414 @item @samp{ReverseStep}
36419 @item @samp{TracepointSource}
36424 @item @samp{QAgent}
36429 @item @samp{QAllow}
36434 @item @samp{QDisableRandomization}
36439 @item @samp{EnableDisableTracepoints}
36444 @item @samp{tracenz}
36451 These are the currently defined stub features, in more detail:
36454 @cindex packet size, remote protocol
36455 @item PacketSize=@var{bytes}
36456 The remote stub can accept packets up to at least @var{bytes} in
36457 length. @value{GDBN} will send packets up to this size for bulk
36458 transfers, and will never send larger packets. This is a limit on the
36459 data characters in the packet, including the frame and checksum.
36460 There is no trailing NUL byte in a remote protocol packet; if the stub
36461 stores packets in a NUL-terminated format, it should allow an extra
36462 byte in its buffer for the NUL. If this stub feature is not supported,
36463 @value{GDBN} guesses based on the size of the @samp{g} packet response.
36465 @item qXfer:auxv:read
36466 The remote stub understands the @samp{qXfer:auxv:read} packet
36467 (@pxref{qXfer auxiliary vector read}).
36469 @item qXfer:features:read
36470 The remote stub understands the @samp{qXfer:features:read} packet
36471 (@pxref{qXfer target description read}).
36473 @item qXfer:libraries:read
36474 The remote stub understands the @samp{qXfer:libraries:read} packet
36475 (@pxref{qXfer library list read}).
36477 @item qXfer:libraries-svr4:read
36478 The remote stub understands the @samp{qXfer:libraries-svr4:read} packet
36479 (@pxref{qXfer svr4 library list read}).
36481 @item qXfer:memory-map:read
36482 The remote stub understands the @samp{qXfer:memory-map:read} packet
36483 (@pxref{qXfer memory map read}).
36485 @item qXfer:sdata:read
36486 The remote stub understands the @samp{qXfer:sdata:read} packet
36487 (@pxref{qXfer sdata read}).
36489 @item qXfer:spu:read
36490 The remote stub understands the @samp{qXfer:spu:read} packet
36491 (@pxref{qXfer spu read}).
36493 @item qXfer:spu:write
36494 The remote stub understands the @samp{qXfer:spu:write} packet
36495 (@pxref{qXfer spu write}).
36497 @item qXfer:siginfo:read
36498 The remote stub understands the @samp{qXfer:siginfo:read} packet
36499 (@pxref{qXfer siginfo read}).
36501 @item qXfer:siginfo:write
36502 The remote stub understands the @samp{qXfer:siginfo:write} packet
36503 (@pxref{qXfer siginfo write}).
36505 @item qXfer:threads:read
36506 The remote stub understands the @samp{qXfer:threads:read} packet
36507 (@pxref{qXfer threads read}).
36509 @item qXfer:traceframe-info:read
36510 The remote stub understands the @samp{qXfer:traceframe-info:read}
36511 packet (@pxref{qXfer traceframe info read}).
36513 @item qXfer:uib:read
36514 The remote stub understands the @samp{qXfer:uib:read}
36515 packet (@pxref{qXfer unwind info block}).
36517 @item qXfer:fdpic:read
36518 The remote stub understands the @samp{qXfer:fdpic:read}
36519 packet (@pxref{qXfer fdpic loadmap read}).
36522 The remote stub understands the @samp{QNonStop} packet
36523 (@pxref{QNonStop}).
36526 The remote stub understands the @samp{QPassSignals} packet
36527 (@pxref{QPassSignals}).
36529 @item QStartNoAckMode
36530 The remote stub understands the @samp{QStartNoAckMode} packet and
36531 prefers to operate in no-acknowledgment mode. @xref{Packet Acknowledgment}.
36534 @anchor{multiprocess extensions}
36535 @cindex multiprocess extensions, in remote protocol
36536 The remote stub understands the multiprocess extensions to the remote
36537 protocol syntax. The multiprocess extensions affect the syntax of
36538 thread IDs in both packets and replies (@pxref{thread-id syntax}), and
36539 add process IDs to the @samp{D} packet and @samp{W} and @samp{X}
36540 replies. Note that reporting this feature indicates support for the
36541 syntactic extensions only, not that the stub necessarily supports
36542 debugging of more than one process at a time. The stub must not use
36543 multiprocess extensions in packet replies unless @value{GDBN} has also
36544 indicated it supports them in its @samp{qSupported} request.
36546 @item qXfer:osdata:read
36547 The remote stub understands the @samp{qXfer:osdata:read} packet
36548 ((@pxref{qXfer osdata read}).
36550 @item ConditionalBreakpoints
36551 The target accepts and implements evaluation of conditional expressions
36552 defined for breakpoints. The target will only report breakpoint triggers
36553 when such conditions are true (@pxref{Conditions, ,Break Conditions}).
36555 @item ConditionalTracepoints
36556 The remote stub accepts and implements conditional expressions defined
36557 for tracepoints (@pxref{Tracepoint Conditions}).
36559 @item ReverseContinue
36560 The remote stub accepts and implements the reverse continue packet
36564 The remote stub accepts and implements the reverse step packet
36567 @item TracepointSource
36568 The remote stub understands the @samp{QTDPsrc} packet that supplies
36569 the source form of tracepoint definitions.
36572 The remote stub understands the @samp{QAgent} packet.
36575 The remote stub understands the @samp{QAllow} packet.
36577 @item QDisableRandomization
36578 The remote stub understands the @samp{QDisableRandomization} packet.
36580 @item StaticTracepoint
36581 @cindex static tracepoints, in remote protocol
36582 The remote stub supports static tracepoints.
36584 @item InstallInTrace
36585 @anchor{install tracepoint in tracing}
36586 The remote stub supports installing tracepoint in tracing.
36588 @item EnableDisableTracepoints
36589 The remote stub supports the @samp{QTEnable} (@pxref{QTEnable}) and
36590 @samp{QTDisable} (@pxref{QTDisable}) packets that allow tracepoints
36591 to be enabled and disabled while a trace experiment is running.
36594 @cindex string tracing, in remote protocol
36595 The remote stub supports the @samp{tracenz} bytecode for collecting strings.
36596 See @ref{Bytecode Descriptions} for details about the bytecode.
36601 @cindex symbol lookup, remote request
36602 @cindex @samp{qSymbol} packet
36603 Notify the target that @value{GDBN} is prepared to serve symbol lookup
36604 requests. Accept requests from the target for the values of symbols.
36609 The target does not need to look up any (more) symbols.
36610 @item qSymbol:@var{sym_name}
36611 The target requests the value of symbol @var{sym_name} (hex encoded).
36612 @value{GDBN} may provide the value by using the
36613 @samp{qSymbol:@var{sym_value}:@var{sym_name}} message, described
36617 @item qSymbol:@var{sym_value}:@var{sym_name}
36618 Set the value of @var{sym_name} to @var{sym_value}.
36620 @var{sym_name} (hex encoded) is the name of a symbol whose value the
36621 target has previously requested.
36623 @var{sym_value} (hex) is the value for symbol @var{sym_name}. If
36624 @value{GDBN} cannot supply a value for @var{sym_name}, then this field
36630 The target does not need to look up any (more) symbols.
36631 @item qSymbol:@var{sym_name}
36632 The target requests the value of a new symbol @var{sym_name} (hex
36633 encoded). @value{GDBN} will continue to supply the values of symbols
36634 (if available), until the target ceases to request them.
36639 @item QTDisconnected
36646 @itemx qTMinFTPILen
36648 @xref{Tracepoint Packets}.
36650 @item qThreadExtraInfo,@var{thread-id}
36651 @cindex thread attributes info, remote request
36652 @cindex @samp{qThreadExtraInfo} packet
36653 Obtain a printable string description of a thread's attributes from
36654 the target OS. @var{thread-id} is a thread ID;
36655 see @ref{thread-id syntax}. This
36656 string may contain anything that the target OS thinks is interesting
36657 for @value{GDBN} to tell the user about the thread. The string is
36658 displayed in @value{GDBN}'s @code{info threads} display. Some
36659 examples of possible thread extra info strings are @samp{Runnable}, or
36660 @samp{Blocked on Mutex}.
36664 @item @var{XX}@dots{}
36665 Where @samp{@var{XX}@dots{}} is a hex encoding of @sc{ascii} data,
36666 comprising the printable string containing the extra information about
36667 the thread's attributes.
36670 (Note that the @code{qThreadExtraInfo} packet's name is separated from
36671 the command by a @samp{,}, not a @samp{:}, contrary to the naming
36672 conventions above. Please don't use this packet as a model for new
36691 @xref{Tracepoint Packets}.
36693 @item qXfer:@var{object}:read:@var{annex}:@var{offset},@var{length}
36694 @cindex read special object, remote request
36695 @cindex @samp{qXfer} packet
36696 @anchor{qXfer read}
36697 Read uninterpreted bytes from the target's special data area
36698 identified by the keyword @var{object}. Request @var{length} bytes
36699 starting at @var{offset} bytes into the data. The content and
36700 encoding of @var{annex} is specific to @var{object}; it can supply
36701 additional details about what data to access.
36703 Here are the specific requests of this form defined so far. All
36704 @samp{qXfer:@var{object}:read:@dots{}} requests use the same reply
36705 formats, listed below.
36708 @item qXfer:auxv:read::@var{offset},@var{length}
36709 @anchor{qXfer auxiliary vector read}
36710 Access the target's @dfn{auxiliary vector}. @xref{OS Information,
36711 auxiliary vector}. Note @var{annex} must be empty.
36713 This packet is not probed by default; the remote stub must request it,
36714 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
36716 @item qXfer:features:read:@var{annex}:@var{offset},@var{length}
36717 @anchor{qXfer target description read}
36718 Access the @dfn{target description}. @xref{Target Descriptions}. The
36719 annex specifies which XML document to access. The main description is
36720 always loaded from the @samp{target.xml} annex.
36722 This packet is not probed by default; the remote stub must request it,
36723 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
36725 @item qXfer:libraries:read:@var{annex}:@var{offset},@var{length}
36726 @anchor{qXfer library list read}
36727 Access the target's list of loaded libraries. @xref{Library List Format}.
36728 The annex part of the generic @samp{qXfer} packet must be empty
36729 (@pxref{qXfer read}).
36731 Targets which maintain a list of libraries in the program's memory do
36732 not need to implement this packet; it is designed for platforms where
36733 the operating system manages the list of loaded libraries.
36735 This packet is not probed by default; the remote stub must request it,
36736 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
36738 @item qXfer:libraries-svr4:read:@var{annex}:@var{offset},@var{length}
36739 @anchor{qXfer svr4 library list read}
36740 Access the target's list of loaded libraries when the target is an SVR4
36741 platform. @xref{Library List Format for SVR4 Targets}. The annex part
36742 of the generic @samp{qXfer} packet must be empty (@pxref{qXfer read}).
36744 This packet is optional for better performance on SVR4 targets.
36745 @value{GDBN} uses memory read packets to read the SVR4 library list otherwise.
36747 This packet is not probed by default; the remote stub must request it,
36748 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
36750 @item qXfer:memory-map:read::@var{offset},@var{length}
36751 @anchor{qXfer memory map read}
36752 Access the target's @dfn{memory-map}. @xref{Memory Map Format}. The
36753 annex part of the generic @samp{qXfer} packet must be empty
36754 (@pxref{qXfer read}).
36756 This packet is not probed by default; the remote stub must request it,
36757 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
36759 @item qXfer:sdata:read::@var{offset},@var{length}
36760 @anchor{qXfer sdata read}
36762 Read contents of the extra collected static tracepoint marker
36763 information. The annex part of the generic @samp{qXfer} packet must
36764 be empty (@pxref{qXfer read}). @xref{Tracepoint Actions,,Tracepoint
36767 This packet is not probed by default; the remote stub must request it,
36768 by supplying an appropriate @samp{qSupported} response
36769 (@pxref{qSupported}).
36771 @item qXfer:siginfo:read::@var{offset},@var{length}
36772 @anchor{qXfer siginfo read}
36773 Read contents of the extra signal information on the target
36774 system. The annex part of the generic @samp{qXfer} packet must be
36775 empty (@pxref{qXfer read}).
36777 This packet is not probed by default; the remote stub must request it,
36778 by supplying an appropriate @samp{qSupported} response
36779 (@pxref{qSupported}).
36781 @item qXfer:spu:read:@var{annex}:@var{offset},@var{length}
36782 @anchor{qXfer spu read}
36783 Read contents of an @code{spufs} file on the target system. The
36784 annex specifies which file to read; it must be of the form
36785 @file{@var{id}/@var{name}}, where @var{id} specifies an SPU context ID
36786 in the target process, and @var{name} identifes the @code{spufs} file
36787 in that context to be accessed.
36789 This packet is not probed by default; the remote stub must request it,
36790 by supplying an appropriate @samp{qSupported} response
36791 (@pxref{qSupported}).
36793 @item qXfer:threads:read::@var{offset},@var{length}
36794 @anchor{qXfer threads read}
36795 Access the list of threads on target. @xref{Thread List Format}. The
36796 annex part of the generic @samp{qXfer} packet must be empty
36797 (@pxref{qXfer read}).
36799 This packet is not probed by default; the remote stub must request it,
36800 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
36802 @item qXfer:traceframe-info:read::@var{offset},@var{length}
36803 @anchor{qXfer traceframe info read}
36805 Return a description of the current traceframe's contents.
36806 @xref{Traceframe Info Format}. The annex part of the generic
36807 @samp{qXfer} packet must be empty (@pxref{qXfer read}).
36809 This packet is not probed by default; the remote stub must request it,
36810 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
36812 @item qXfer:uib:read:@var{pc}:@var{offset},@var{length}
36813 @anchor{qXfer unwind info block}
36815 Return the unwind information block for @var{pc}. This packet is used
36816 on OpenVMS/ia64 to ask the kernel unwind information.
36818 This packet is not probed by default.
36820 @item qXfer:fdpic:read:@var{annex}:@var{offset},@var{length}
36821 @anchor{qXfer fdpic loadmap read}
36822 Read contents of @code{loadmap}s on the target system. The
36823 annex, either @samp{exec} or @samp{interp}, specifies which @code{loadmap},
36824 executable @code{loadmap} or interpreter @code{loadmap} to read.
36826 This packet is not probed by default; the remote stub must request it,
36827 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
36829 @item qXfer:osdata:read::@var{offset},@var{length}
36830 @anchor{qXfer osdata read}
36831 Access the target's @dfn{operating system information}.
36832 @xref{Operating System Information}.
36839 Data @var{data} (@pxref{Binary Data}) has been read from the
36840 target. There may be more data at a higher address (although
36841 it is permitted to return @samp{m} even for the last valid
36842 block of data, as long as at least one byte of data was read).
36843 @var{data} may have fewer bytes than the @var{length} in the
36847 Data @var{data} (@pxref{Binary Data}) has been read from the target.
36848 There is no more data to be read. @var{data} may have fewer bytes
36849 than the @var{length} in the request.
36852 The @var{offset} in the request is at the end of the data.
36853 There is no more data to be read.
36856 The request was malformed, or @var{annex} was invalid.
36859 The offset was invalid, or there was an error encountered reading the data.
36860 @var{nn} is a hex-encoded @code{errno} value.
36863 An empty reply indicates the @var{object} string was not recognized by
36864 the stub, or that the object does not support reading.
36867 @item qXfer:@var{object}:write:@var{annex}:@var{offset}:@var{data}@dots{}
36868 @cindex write data into object, remote request
36869 @anchor{qXfer write}
36870 Write uninterpreted bytes into the target's special data area
36871 identified by the keyword @var{object}, starting at @var{offset} bytes
36872 into the data. @var{data}@dots{} is the binary-encoded data
36873 (@pxref{Binary Data}) to be written. The content and encoding of @var{annex}
36874 is specific to @var{object}; it can supply additional details about what data
36877 Here are the specific requests of this form defined so far. All
36878 @samp{qXfer:@var{object}:write:@dots{}} requests use the same reply
36879 formats, listed below.
36882 @item qXfer:siginfo:write::@var{offset}:@var{data}@dots{}
36883 @anchor{qXfer siginfo write}
36884 Write @var{data} to the extra signal information on the target system.
36885 The annex part of the generic @samp{qXfer} packet must be
36886 empty (@pxref{qXfer write}).
36888 This packet is not probed by default; the remote stub must request it,
36889 by supplying an appropriate @samp{qSupported} response
36890 (@pxref{qSupported}).
36892 @item qXfer:spu:write:@var{annex}:@var{offset}:@var{data}@dots{}
36893 @anchor{qXfer spu write}
36894 Write @var{data} to an @code{spufs} file on the target system. The
36895 annex specifies which file to write; it must be of the form
36896 @file{@var{id}/@var{name}}, where @var{id} specifies an SPU context ID
36897 in the target process, and @var{name} identifes the @code{spufs} file
36898 in that context to be accessed.
36900 This packet is not probed by default; the remote stub must request it,
36901 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
36907 @var{nn} (hex encoded) is the number of bytes written.
36908 This may be fewer bytes than supplied in the request.
36911 The request was malformed, or @var{annex} was invalid.
36914 The offset was invalid, or there was an error encountered writing the data.
36915 @var{nn} is a hex-encoded @code{errno} value.
36918 An empty reply indicates the @var{object} string was not
36919 recognized by the stub, or that the object does not support writing.
36922 @item qXfer:@var{object}:@var{operation}:@dots{}
36923 Requests of this form may be added in the future. When a stub does
36924 not recognize the @var{object} keyword, or its support for
36925 @var{object} does not recognize the @var{operation} keyword, the stub
36926 must respond with an empty packet.
36928 @item qAttached:@var{pid}
36929 @cindex query attached, remote request
36930 @cindex @samp{qAttached} packet
36931 Return an indication of whether the remote server attached to an
36932 existing process or created a new process. When the multiprocess
36933 protocol extensions are supported (@pxref{multiprocess extensions}),
36934 @var{pid} is an integer in hexadecimal format identifying the target
36935 process. Otherwise, @value{GDBN} will omit the @var{pid} field and
36936 the query packet will be simplified as @samp{qAttached}.
36938 This query is used, for example, to know whether the remote process
36939 should be detached or killed when a @value{GDBN} session is ended with
36940 the @code{quit} command.
36945 The remote server attached to an existing process.
36947 The remote server created a new process.
36949 A badly formed request or an error was encountered.
36954 @node Architecture-Specific Protocol Details
36955 @section Architecture-Specific Protocol Details
36957 This section describes how the remote protocol is applied to specific
36958 target architectures. Also see @ref{Standard Target Features}, for
36959 details of XML target descriptions for each architecture.
36962 * ARM-Specific Protocol Details::
36963 * MIPS-Specific Protocol Details::
36966 @node ARM-Specific Protocol Details
36967 @subsection @acronym{ARM}-specific Protocol Details
36970 * ARM Breakpoint Kinds::
36973 @node ARM Breakpoint Kinds
36974 @subsubsection @acronym{ARM} Breakpoint Kinds
36975 @cindex breakpoint kinds, @acronym{ARM}
36977 These breakpoint kinds are defined for the @samp{Z0} and @samp{Z1} packets.
36982 16-bit Thumb mode breakpoint.
36985 32-bit Thumb mode (Thumb-2) breakpoint.
36988 32-bit @acronym{ARM} mode breakpoint.
36992 @node MIPS-Specific Protocol Details
36993 @subsection @acronym{MIPS}-specific Protocol Details
36996 * MIPS Register packet Format::
36997 * MIPS Breakpoint Kinds::
37000 @node MIPS Register packet Format
37001 @subsubsection @acronym{MIPS} Register Packet Format
37002 @cindex register packet format, @acronym{MIPS}
37004 The following @code{g}/@code{G} packets have previously been defined.
37005 In the below, some thirty-two bit registers are transferred as
37006 sixty-four bits. Those registers should be zero/sign extended (which?)
37007 to fill the space allocated. Register bytes are transferred in target
37008 byte order. The two nibbles within a register byte are transferred
37009 most-significant -- least-significant.
37014 All registers are transferred as thirty-two bit quantities in the order:
37015 32 general-purpose; sr; lo; hi; bad; cause; pc; 32 floating-point
37016 registers; fsr; fir; fp.
37019 All registers are transferred as sixty-four bit quantities (including
37020 thirty-two bit registers such as @code{sr}). The ordering is the same
37025 @node MIPS Breakpoint Kinds
37026 @subsubsection @acronym{MIPS} Breakpoint Kinds
37027 @cindex breakpoint kinds, @acronym{MIPS}
37029 These breakpoint kinds are defined for the @samp{Z0} and @samp{Z1} packets.
37034 16-bit @acronym{MIPS16} mode breakpoint.
37037 16-bit @acronym{microMIPS} mode breakpoint.
37040 32-bit standard @acronym{MIPS} mode breakpoint.
37043 32-bit @acronym{microMIPS} mode breakpoint.
37047 @node Tracepoint Packets
37048 @section Tracepoint Packets
37049 @cindex tracepoint packets
37050 @cindex packets, tracepoint
37052 Here we describe the packets @value{GDBN} uses to implement
37053 tracepoints (@pxref{Tracepoints}).
37057 @item QTDP:@var{n}:@var{addr}:@var{ena}:@var{step}:@var{pass}[:F@var{flen}][:X@var{len},@var{bytes}]@r{[}-@r{]}
37058 Create a new tracepoint, number @var{n}, at @var{addr}. If @var{ena}
37059 is @samp{E}, then the tracepoint is enabled; if it is @samp{D}, then
37060 the tracepoint is disabled. @var{step} is the tracepoint's step
37061 count, and @var{pass} is its pass count. If an @samp{F} is present,
37062 then the tracepoint is to be a fast tracepoint, and the @var{flen} is
37063 the number of bytes that the target should copy elsewhere to make room
37064 for the tracepoint. If an @samp{X} is present, it introduces a
37065 tracepoint condition, which consists of a hexadecimal length, followed
37066 by a comma and hex-encoded bytes, in a manner similar to action
37067 encodings as described below. If the trailing @samp{-} is present,
37068 further @samp{QTDP} packets will follow to specify this tracepoint's
37074 The packet was understood and carried out.
37076 @xref{Tracepoint Packets,,Relocate instruction reply packet}.
37078 The packet was not recognized.
37081 @item QTDP:-@var{n}:@var{addr}:@r{[}S@r{]}@var{action}@dots{}@r{[}-@r{]}
37082 Define actions to be taken when a tracepoint is hit. @var{n} and
37083 @var{addr} must be the same as in the initial @samp{QTDP} packet for
37084 this tracepoint. This packet may only be sent immediately after
37085 another @samp{QTDP} packet that ended with a @samp{-}. If the
37086 trailing @samp{-} is present, further @samp{QTDP} packets will follow,
37087 specifying more actions for this tracepoint.
37089 In the series of action packets for a given tracepoint, at most one
37090 can have an @samp{S} before its first @var{action}. If such a packet
37091 is sent, it and the following packets define ``while-stepping''
37092 actions. Any prior packets define ordinary actions --- that is, those
37093 taken when the tracepoint is first hit. If no action packet has an
37094 @samp{S}, then all the packets in the series specify ordinary
37095 tracepoint actions.
37097 The @samp{@var{action}@dots{}} portion of the packet is a series of
37098 actions, concatenated without separators. Each action has one of the
37104 Collect the registers whose bits are set in @var{mask}. @var{mask} is
37105 a hexadecimal number whose @var{i}'th bit is set if register number
37106 @var{i} should be collected. (The least significant bit is numbered
37107 zero.) Note that @var{mask} may be any number of digits long; it may
37108 not fit in a 32-bit word.
37110 @item M @var{basereg},@var{offset},@var{len}
37111 Collect @var{len} bytes of memory starting at the address in register
37112 number @var{basereg}, plus @var{offset}. If @var{basereg} is
37113 @samp{-1}, then the range has a fixed address: @var{offset} is the
37114 address of the lowest byte to collect. The @var{basereg},
37115 @var{offset}, and @var{len} parameters are all unsigned hexadecimal
37116 values (the @samp{-1} value for @var{basereg} is a special case).
37118 @item X @var{len},@var{expr}
37119 Evaluate @var{expr}, whose length is @var{len}, and collect memory as
37120 it directs. @var{expr} is an agent expression, as described in
37121 @ref{Agent Expressions}. Each byte of the expression is encoded as a
37122 two-digit hex number in the packet; @var{len} is the number of bytes
37123 in the expression (and thus one-half the number of hex digits in the
37128 Any number of actions may be packed together in a single @samp{QTDP}
37129 packet, as long as the packet does not exceed the maximum packet
37130 length (400 bytes, for many stubs). There may be only one @samp{R}
37131 action per tracepoint, and it must precede any @samp{M} or @samp{X}
37132 actions. Any registers referred to by @samp{M} and @samp{X} actions
37133 must be collected by a preceding @samp{R} action. (The
37134 ``while-stepping'' actions are treated as if they were attached to a
37135 separate tracepoint, as far as these restrictions are concerned.)
37140 The packet was understood and carried out.
37142 @xref{Tracepoint Packets,,Relocate instruction reply packet}.
37144 The packet was not recognized.
37147 @item QTDPsrc:@var{n}:@var{addr}:@var{type}:@var{start}:@var{slen}:@var{bytes}
37148 @cindex @samp{QTDPsrc} packet
37149 Specify a source string of tracepoint @var{n} at address @var{addr}.
37150 This is useful to get accurate reproduction of the tracepoints
37151 originally downloaded at the beginning of the trace run. @var{type}
37152 is the name of the tracepoint part, such as @samp{cond} for the
37153 tracepoint's conditional expression (see below for a list of types), while
37154 @var{bytes} is the string, encoded in hexadecimal.
37156 @var{start} is the offset of the @var{bytes} within the overall source
37157 string, while @var{slen} is the total length of the source string.
37158 This is intended for handling source strings that are longer than will
37159 fit in a single packet.
37160 @c Add detailed example when this info is moved into a dedicated
37161 @c tracepoint descriptions section.
37163 The available string types are @samp{at} for the location,
37164 @samp{cond} for the conditional, and @samp{cmd} for an action command.
37165 @value{GDBN} sends a separate packet for each command in the action
37166 list, in the same order in which the commands are stored in the list.
37168 The target does not need to do anything with source strings except
37169 report them back as part of the replies to the @samp{qTfP}/@samp{qTsP}
37172 Although this packet is optional, and @value{GDBN} will only send it
37173 if the target replies with @samp{TracepointSource} @xref{General
37174 Query Packets}, it makes both disconnected tracing and trace files
37175 much easier to use. Otherwise the user must be careful that the
37176 tracepoints in effect while looking at trace frames are identical to
37177 the ones in effect during the trace run; even a small discrepancy
37178 could cause @samp{tdump} not to work, or a particular trace frame not
37181 @item QTDV:@var{n}:@var{value}
37182 @cindex define trace state variable, remote request
37183 @cindex @samp{QTDV} packet
37184 Create a new trace state variable, number @var{n}, with an initial
37185 value of @var{value}, which is a 64-bit signed integer. Both @var{n}
37186 and @var{value} are encoded as hexadecimal values. @value{GDBN} has
37187 the option of not using this packet for initial values of zero; the
37188 target should simply create the trace state variables as they are
37189 mentioned in expressions.
37191 @item QTFrame:@var{n}
37192 Select the @var{n}'th tracepoint frame from the buffer, and use the
37193 register and memory contents recorded there to answer subsequent
37194 request packets from @value{GDBN}.
37196 A successful reply from the stub indicates that the stub has found the
37197 requested frame. The response is a series of parts, concatenated
37198 without separators, describing the frame we selected. Each part has
37199 one of the following forms:
37203 The selected frame is number @var{n} in the trace frame buffer;
37204 @var{f} is a hexadecimal number. If @var{f} is @samp{-1}, then there
37205 was no frame matching the criteria in the request packet.
37208 The selected trace frame records a hit of tracepoint number @var{t};
37209 @var{t} is a hexadecimal number.
37213 @item QTFrame:pc:@var{addr}
37214 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
37215 currently selected frame whose PC is @var{addr};
37216 @var{addr} is a hexadecimal number.
37218 @item QTFrame:tdp:@var{t}
37219 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
37220 currently selected frame that is a hit of tracepoint @var{t}; @var{t}
37221 is a hexadecimal number.
37223 @item QTFrame:range:@var{start}:@var{end}
37224 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
37225 currently selected frame whose PC is between @var{start} (inclusive)
37226 and @var{end} (inclusive); @var{start} and @var{end} are hexadecimal
37229 @item QTFrame:outside:@var{start}:@var{end}
37230 Like @samp{QTFrame:range:@var{start}:@var{end}}, but select the first
37231 frame @emph{outside} the given range of addresses (exclusive).
37234 This packet requests the minimum length of instruction at which a fast
37235 tracepoint (@pxref{Set Tracepoints}) may be placed. For instance, on
37236 the 32-bit x86 architecture, it is possible to use a 4-byte jump, but
37237 it depends on the target system being able to create trampolines in
37238 the first 64K of memory, which might or might not be possible for that
37239 system. So the reply to this packet will be 4 if it is able to
37246 The minimum instruction length is currently unknown.
37248 The minimum instruction length is @var{length}, where @var{length} is greater
37249 or equal to 1. @var{length} is a hexadecimal number. A reply of 1 means
37250 that a fast tracepoint may be placed on any instruction regardless of size.
37252 An error has occurred.
37254 An empty reply indicates that the request is not supported by the stub.
37258 Begin the tracepoint experiment. Begin collecting data from
37259 tracepoint hits in the trace frame buffer. This packet supports the
37260 @samp{qRelocInsn} reply (@pxref{Tracepoint Packets,,Relocate
37261 instruction reply packet}).
37264 End the tracepoint experiment. Stop collecting trace frames.
37266 @item QTEnable:@var{n}:@var{addr}
37268 Enable tracepoint @var{n} at address @var{addr} in a started tracepoint
37269 experiment. If the tracepoint was previously disabled, then collection
37270 of data from it will resume.
37272 @item QTDisable:@var{n}:@var{addr}
37274 Disable tracepoint @var{n} at address @var{addr} in a started tracepoint
37275 experiment. No more data will be collected from the tracepoint unless
37276 @samp{QTEnable:@var{n}:@var{addr}} is subsequently issued.
37279 Clear the table of tracepoints, and empty the trace frame buffer.
37281 @item QTro:@var{start1},@var{end1}:@var{start2},@var{end2}:@dots{}
37282 Establish the given ranges of memory as ``transparent''. The stub
37283 will answer requests for these ranges from memory's current contents,
37284 if they were not collected as part of the tracepoint hit.
37286 @value{GDBN} uses this to mark read-only regions of memory, like those
37287 containing program code. Since these areas never change, they should
37288 still have the same contents they did when the tracepoint was hit, so
37289 there's no reason for the stub to refuse to provide their contents.
37291 @item QTDisconnected:@var{value}
37292 Set the choice to what to do with the tracing run when @value{GDBN}
37293 disconnects from the target. A @var{value} of 1 directs the target to
37294 continue the tracing run, while 0 tells the target to stop tracing if
37295 @value{GDBN} is no longer in the picture.
37298 Ask the stub if there is a trace experiment running right now.
37300 The reply has the form:
37304 @item T@var{running}@r{[};@var{field}@r{]}@dots{}
37305 @var{running} is a single digit @code{1} if the trace is presently
37306 running, or @code{0} if not. It is followed by semicolon-separated
37307 optional fields that an agent may use to report additional status.
37311 If the trace is not running, the agent may report any of several
37312 explanations as one of the optional fields:
37317 No trace has been run yet.
37319 @item tstop[:@var{text}]:0
37320 The trace was stopped by a user-originated stop command. The optional
37321 @var{text} field is a user-supplied string supplied as part of the
37322 stop command (for instance, an explanation of why the trace was
37323 stopped manually). It is hex-encoded.
37326 The trace stopped because the trace buffer filled up.
37328 @item tdisconnected:0
37329 The trace stopped because @value{GDBN} disconnected from the target.
37331 @item tpasscount:@var{tpnum}
37332 The trace stopped because tracepoint @var{tpnum} exceeded its pass count.
37334 @item terror:@var{text}:@var{tpnum}
37335 The trace stopped because tracepoint @var{tpnum} had an error. The
37336 string @var{text} is available to describe the nature of the error
37337 (for instance, a divide by zero in the condition expression).
37338 @var{text} is hex encoded.
37341 The trace stopped for some other reason.
37345 Additional optional fields supply statistical and other information.
37346 Although not required, they are extremely useful for users monitoring
37347 the progress of a trace run. If a trace has stopped, and these
37348 numbers are reported, they must reflect the state of the just-stopped
37353 @item tframes:@var{n}
37354 The number of trace frames in the buffer.
37356 @item tcreated:@var{n}
37357 The total number of trace frames created during the run. This may
37358 be larger than the trace frame count, if the buffer is circular.
37360 @item tsize:@var{n}
37361 The total size of the trace buffer, in bytes.
37363 @item tfree:@var{n}
37364 The number of bytes still unused in the buffer.
37366 @item circular:@var{n}
37367 The value of the circular trace buffer flag. @code{1} means that the
37368 trace buffer is circular and old trace frames will be discarded if
37369 necessary to make room, @code{0} means that the trace buffer is linear
37372 @item disconn:@var{n}
37373 The value of the disconnected tracing flag. @code{1} means that
37374 tracing will continue after @value{GDBN} disconnects, @code{0} means
37375 that the trace run will stop.
37379 @item qTP:@var{tp}:@var{addr}
37380 @cindex tracepoint status, remote request
37381 @cindex @samp{qTP} packet
37382 Ask the stub for the current state of tracepoint number @var{tp} at
37383 address @var{addr}.
37387 @item V@var{hits}:@var{usage}
37388 The tracepoint has been hit @var{hits} times so far during the trace
37389 run, and accounts for @var{usage} in the trace buffer. Note that
37390 @code{while-stepping} steps are not counted as separate hits, but the
37391 steps' space consumption is added into the usage number.
37395 @item qTV:@var{var}
37396 @cindex trace state variable value, remote request
37397 @cindex @samp{qTV} packet
37398 Ask the stub for the value of the trace state variable number @var{var}.
37403 The value of the variable is @var{value}. This will be the current
37404 value of the variable if the user is examining a running target, or a
37405 saved value if the variable was collected in the trace frame that the
37406 user is looking at. Note that multiple requests may result in
37407 different reply values, such as when requesting values while the
37408 program is running.
37411 The value of the variable is unknown. This would occur, for example,
37412 if the user is examining a trace frame in which the requested variable
37418 These packets request data about tracepoints that are being used by
37419 the target. @value{GDBN} sends @code{qTfP} to get the first piece
37420 of data, and multiple @code{qTsP} to get additional pieces. Replies
37421 to these packets generally take the form of the @code{QTDP} packets
37422 that define tracepoints. (FIXME add detailed syntax)
37426 These packets request data about trace state variables that are on the
37427 target. @value{GDBN} sends @code{qTfV} to get the first vari of data,
37428 and multiple @code{qTsV} to get additional variables. Replies to
37429 these packets follow the syntax of the @code{QTDV} packets that define
37430 trace state variables.
37434 These packets request data about static tracepoint markers that exist
37435 in the target program. @value{GDBN} sends @code{qTfSTM} to get the
37436 first piece of data, and multiple @code{qTsSTM} to get additional
37437 pieces. Replies to these packets take the following form:
37441 @item m @var{address}:@var{id}:@var{extra}
37443 @item m @var{address}:@var{id}:@var{extra},@var{address}:@var{id}:@var{extra}@dots{}
37444 a comma-separated list of markers
37446 (lower case letter @samp{L}) denotes end of list.
37448 An error occurred. @var{nn} are hex digits.
37450 An empty reply indicates that the request is not supported by the
37454 @var{address} is encoded in hex.
37455 @var{id} and @var{extra} are strings encoded in hex.
37457 In response to each query, the target will reply with a list of one or
37458 more markers, separated by commas. @value{GDBN} will respond to each
37459 reply with a request for more markers (using the @samp{qs} form of the
37460 query), until the target responds with @samp{l} (lower-case ell, for
37463 @item qTSTMat:@var{address}
37464 This packets requests data about static tracepoint markers in the
37465 target program at @var{address}. Replies to this packet follow the
37466 syntax of the @samp{qTfSTM} and @code{qTsSTM} packets that list static
37467 tracepoint markers.
37469 @item QTSave:@var{filename}
37470 This packet directs the target to save trace data to the file name
37471 @var{filename} in the target's filesystem. @var{filename} is encoded
37472 as a hex string; the interpretation of the file name (relative vs
37473 absolute, wild cards, etc) is up to the target.
37475 @item qTBuffer:@var{offset},@var{len}
37476 Return up to @var{len} bytes of the current contents of trace buffer,
37477 starting at @var{offset}. The trace buffer is treated as if it were
37478 a contiguous collection of traceframes, as per the trace file format.
37479 The reply consists as many hex-encoded bytes as the target can deliver
37480 in a packet; it is not an error to return fewer than were asked for.
37481 A reply consisting of just @code{l} indicates that no bytes are
37484 @item QTBuffer:circular:@var{value}
37485 This packet directs the target to use a circular trace buffer if
37486 @var{value} is 1, or a linear buffer if the value is 0.
37488 @item QTNotes:@r{[}@var{type}:@var{text}@r{]}@r{[};@var{type}:@var{text}@r{]}@dots{}
37489 This packet adds optional textual notes to the trace run. Allowable
37490 types include @code{user}, @code{notes}, and @code{tstop}, the
37491 @var{text} fields are arbitrary strings, hex-encoded.
37495 @subsection Relocate instruction reply packet
37496 When installing fast tracepoints in memory, the target may need to
37497 relocate the instruction currently at the tracepoint address to a
37498 different address in memory. For most instructions, a simple copy is
37499 enough, but, for example, call instructions that implicitly push the
37500 return address on the stack, and relative branches or other
37501 PC-relative instructions require offset adjustment, so that the effect
37502 of executing the instruction at a different address is the same as if
37503 it had executed in the original location.
37505 In response to several of the tracepoint packets, the target may also
37506 respond with a number of intermediate @samp{qRelocInsn} request
37507 packets before the final result packet, to have @value{GDBN} handle
37508 this relocation operation. If a packet supports this mechanism, its
37509 documentation will explicitly say so. See for example the above
37510 descriptions for the @samp{QTStart} and @samp{QTDP} packets. The
37511 format of the request is:
37514 @item qRelocInsn:@var{from};@var{to}
37516 This requests @value{GDBN} to copy instruction at address @var{from}
37517 to address @var{to}, possibly adjusted so that executing the
37518 instruction at @var{to} has the same effect as executing it at
37519 @var{from}. @value{GDBN} writes the adjusted instruction to target
37520 memory starting at @var{to}.
37525 @item qRelocInsn:@var{adjusted_size}
37526 Informs the stub the relocation is complete. @var{adjusted_size} is
37527 the length in bytes of resulting relocated instruction sequence.
37529 A badly formed request was detected, or an error was encountered while
37530 relocating the instruction.
37533 @node Host I/O Packets
37534 @section Host I/O Packets
37535 @cindex Host I/O, remote protocol
37536 @cindex file transfer, remote protocol
37538 The @dfn{Host I/O} packets allow @value{GDBN} to perform I/O
37539 operations on the far side of a remote link. For example, Host I/O is
37540 used to upload and download files to a remote target with its own
37541 filesystem. Host I/O uses the same constant values and data structure
37542 layout as the target-initiated File-I/O protocol. However, the
37543 Host I/O packets are structured differently. The target-initiated
37544 protocol relies on target memory to store parameters and buffers.
37545 Host I/O requests are initiated by @value{GDBN}, and the
37546 target's memory is not involved. @xref{File-I/O Remote Protocol
37547 Extension}, for more details on the target-initiated protocol.
37549 The Host I/O request packets all encode a single operation along with
37550 its arguments. They have this format:
37554 @item vFile:@var{operation}: @var{parameter}@dots{}
37555 @var{operation} is the name of the particular request; the target
37556 should compare the entire packet name up to the second colon when checking
37557 for a supported operation. The format of @var{parameter} depends on
37558 the operation. Numbers are always passed in hexadecimal. Negative
37559 numbers have an explicit minus sign (i.e.@: two's complement is not
37560 used). Strings (e.g.@: filenames) are encoded as a series of
37561 hexadecimal bytes. The last argument to a system call may be a
37562 buffer of escaped binary data (@pxref{Binary Data}).
37566 The valid responses to Host I/O packets are:
37570 @item F @var{result} [, @var{errno}] [; @var{attachment}]
37571 @var{result} is the integer value returned by this operation, usually
37572 non-negative for success and -1 for errors. If an error has occured,
37573 @var{errno} will be included in the result. @var{errno} will have a
37574 value defined by the File-I/O protocol (@pxref{Errno Values}). For
37575 operations which return data, @var{attachment} supplies the data as a
37576 binary buffer. Binary buffers in response packets are escaped in the
37577 normal way (@pxref{Binary Data}). See the individual packet
37578 documentation for the interpretation of @var{result} and
37582 An empty response indicates that this operation is not recognized.
37586 These are the supported Host I/O operations:
37589 @item vFile:open: @var{pathname}, @var{flags}, @var{mode}
37590 Open a file at @var{pathname} and return a file descriptor for it, or
37591 return -1 if an error occurs. @var{pathname} is a string,
37592 @var{flags} is an integer indicating a mask of open flags
37593 (@pxref{Open Flags}), and @var{mode} is an integer indicating a mask
37594 of mode bits to use if the file is created (@pxref{mode_t Values}).
37595 @xref{open}, for details of the open flags and mode values.
37597 @item vFile:close: @var{fd}
37598 Close the open file corresponding to @var{fd} and return 0, or
37599 -1 if an error occurs.
37601 @item vFile:pread: @var{fd}, @var{count}, @var{offset}
37602 Read data from the open file corresponding to @var{fd}. Up to
37603 @var{count} bytes will be read from the file, starting at @var{offset}
37604 relative to the start of the file. The target may read fewer bytes;
37605 common reasons include packet size limits and an end-of-file
37606 condition. The number of bytes read is returned. Zero should only be
37607 returned for a successful read at the end of the file, or if
37608 @var{count} was zero.
37610 The data read should be returned as a binary attachment on success.
37611 If zero bytes were read, the response should include an empty binary
37612 attachment (i.e.@: a trailing semicolon). The return value is the
37613 number of target bytes read; the binary attachment may be longer if
37614 some characters were escaped.
37616 @item vFile:pwrite: @var{fd}, @var{offset}, @var{data}
37617 Write @var{data} (a binary buffer) to the open file corresponding
37618 to @var{fd}. Start the write at @var{offset} from the start of the
37619 file. Unlike many @code{write} system calls, there is no
37620 separate @var{count} argument; the length of @var{data} in the
37621 packet is used. @samp{vFile:write} returns the number of bytes written,
37622 which may be shorter than the length of @var{data}, or -1 if an
37625 @item vFile:unlink: @var{pathname}
37626 Delete the file at @var{pathname} on the target. Return 0,
37627 or -1 if an error occurs. @var{pathname} is a string.
37629 @item vFile:readlink: @var{filename}
37630 Read value of symbolic link @var{filename} on the target. Return
37631 the number of bytes read, or -1 if an error occurs.
37633 The data read should be returned as a binary attachment on success.
37634 If zero bytes were read, the response should include an empty binary
37635 attachment (i.e.@: a trailing semicolon). The return value is the
37636 number of target bytes read; the binary attachment may be longer if
37637 some characters were escaped.
37642 @section Interrupts
37643 @cindex interrupts (remote protocol)
37645 When a program on the remote target is running, @value{GDBN} may
37646 attempt to interrupt it by sending a @samp{Ctrl-C}, @code{BREAK} or
37647 a @code{BREAK} followed by @code{g},
37648 control of which is specified via @value{GDBN}'s @samp{interrupt-sequence}.
37650 The precise meaning of @code{BREAK} is defined by the transport
37651 mechanism and may, in fact, be undefined. @value{GDBN} does not
37652 currently define a @code{BREAK} mechanism for any of the network
37653 interfaces except for TCP, in which case @value{GDBN} sends the
37654 @code{telnet} BREAK sequence.
37656 @samp{Ctrl-C}, on the other hand, is defined and implemented for all
37657 transport mechanisms. It is represented by sending the single byte
37658 @code{0x03} without any of the usual packet overhead described in
37659 the Overview section (@pxref{Overview}). When a @code{0x03} byte is
37660 transmitted as part of a packet, it is considered to be packet data
37661 and does @emph{not} represent an interrupt. E.g., an @samp{X} packet
37662 (@pxref{X packet}), used for binary downloads, may include an unescaped
37663 @code{0x03} as part of its packet.
37665 @code{BREAK} followed by @code{g} is also known as Magic SysRq g.
37666 When Linux kernel receives this sequence from serial port,
37667 it stops execution and connects to gdb.
37669 Stubs are not required to recognize these interrupt mechanisms and the
37670 precise meaning associated with receipt of the interrupt is
37671 implementation defined. If the target supports debugging of multiple
37672 threads and/or processes, it should attempt to interrupt all
37673 currently-executing threads and processes.
37674 If the stub is successful at interrupting the
37675 running program, it should send one of the stop
37676 reply packets (@pxref{Stop Reply Packets}) to @value{GDBN} as a result
37677 of successfully stopping the program in all-stop mode, and a stop reply
37678 for each stopped thread in non-stop mode.
37679 Interrupts received while the
37680 program is stopped are discarded.
37682 @node Notification Packets
37683 @section Notification Packets
37684 @cindex notification packets
37685 @cindex packets, notification
37687 The @value{GDBN} remote serial protocol includes @dfn{notifications},
37688 packets that require no acknowledgment. Both the GDB and the stub
37689 may send notifications (although the only notifications defined at
37690 present are sent by the stub). Notifications carry information
37691 without incurring the round-trip latency of an acknowledgment, and so
37692 are useful for low-impact communications where occasional packet loss
37695 A notification packet has the form @samp{% @var{data} #
37696 @var{checksum}}, where @var{data} is the content of the notification,
37697 and @var{checksum} is a checksum of @var{data}, computed and formatted
37698 as for ordinary @value{GDBN} packets. A notification's @var{data}
37699 never contains @samp{$}, @samp{%} or @samp{#} characters. Upon
37700 receiving a notification, the recipient sends no @samp{+} or @samp{-}
37701 to acknowledge the notification's receipt or to report its corruption.
37703 Every notification's @var{data} begins with a name, which contains no
37704 colon characters, followed by a colon character.
37706 Recipients should silently ignore corrupted notifications and
37707 notifications they do not understand. Recipients should restart
37708 timeout periods on receipt of a well-formed notification, whether or
37709 not they understand it.
37711 Senders should only send the notifications described here when this
37712 protocol description specifies that they are permitted. In the
37713 future, we may extend the protocol to permit existing notifications in
37714 new contexts; this rule helps older senders avoid confusing newer
37717 (Older versions of @value{GDBN} ignore bytes received until they see
37718 the @samp{$} byte that begins an ordinary packet, so new stubs may
37719 transmit notifications without fear of confusing older clients. There
37720 are no notifications defined for @value{GDBN} to send at the moment, but we
37721 assume that most older stubs would ignore them, as well.)
37723 The following notification packets from the stub to @value{GDBN} are
37727 @item Stop: @var{reply}
37728 Report an asynchronous stop event in non-stop mode.
37729 The @var{reply} has the form of a stop reply, as
37730 described in @ref{Stop Reply Packets}. Refer to @ref{Remote Non-Stop},
37731 for information on how these notifications are acknowledged by
37735 @node Remote Non-Stop
37736 @section Remote Protocol Support for Non-Stop Mode
37738 @value{GDBN}'s remote protocol supports non-stop debugging of
37739 multi-threaded programs, as described in @ref{Non-Stop Mode}. If the stub
37740 supports non-stop mode, it should report that to @value{GDBN} by including
37741 @samp{QNonStop+} in its @samp{qSupported} response (@pxref{qSupported}).
37743 @value{GDBN} typically sends a @samp{QNonStop} packet only when
37744 establishing a new connection with the stub. Entering non-stop mode
37745 does not alter the state of any currently-running threads, but targets
37746 must stop all threads in any already-attached processes when entering
37747 all-stop mode. @value{GDBN} uses the @samp{?} packet as necessary to
37748 probe the target state after a mode change.
37750 In non-stop mode, when an attached process encounters an event that
37751 would otherwise be reported with a stop reply, it uses the
37752 asynchronous notification mechanism (@pxref{Notification Packets}) to
37753 inform @value{GDBN}. In contrast to all-stop mode, where all threads
37754 in all processes are stopped when a stop reply is sent, in non-stop
37755 mode only the thread reporting the stop event is stopped. That is,
37756 when reporting a @samp{S} or @samp{T} response to indicate completion
37757 of a step operation, hitting a breakpoint, or a fault, only the
37758 affected thread is stopped; any other still-running threads continue
37759 to run. When reporting a @samp{W} or @samp{X} response, all running
37760 threads belonging to other attached processes continue to run.
37762 Only one stop reply notification at a time may be pending; if
37763 additional stop events occur before @value{GDBN} has acknowledged the
37764 previous notification, they must be queued by the stub for later
37765 synchronous transmission in response to @samp{vStopped} packets from
37766 @value{GDBN}. Because the notification mechanism is unreliable,
37767 the stub is permitted to resend a stop reply notification
37768 if it believes @value{GDBN} may not have received it. @value{GDBN}
37769 ignores additional stop reply notifications received before it has
37770 finished processing a previous notification and the stub has completed
37771 sending any queued stop events.
37773 Otherwise, @value{GDBN} must be prepared to receive a stop reply
37774 notification at any time. Specifically, they may appear when
37775 @value{GDBN} is not otherwise reading input from the stub, or when
37776 @value{GDBN} is expecting to read a normal synchronous response or a
37777 @samp{+}/@samp{-} acknowledgment to a packet it has sent.
37778 Notification packets are distinct from any other communication from
37779 the stub so there is no ambiguity.
37781 After receiving a stop reply notification, @value{GDBN} shall
37782 acknowledge it by sending a @samp{vStopped} packet (@pxref{vStopped packet})
37783 as a regular, synchronous request to the stub. Such acknowledgment
37784 is not required to happen immediately, as @value{GDBN} is permitted to
37785 send other, unrelated packets to the stub first, which the stub should
37788 Upon receiving a @samp{vStopped} packet, if the stub has other queued
37789 stop events to report to @value{GDBN}, it shall respond by sending a
37790 normal stop reply response. @value{GDBN} shall then send another
37791 @samp{vStopped} packet to solicit further responses; again, it is
37792 permitted to send other, unrelated packets as well which the stub
37793 should process normally.
37795 If the stub receives a @samp{vStopped} packet and there are no
37796 additional stop events to report, the stub shall return an @samp{OK}
37797 response. At this point, if further stop events occur, the stub shall
37798 send a new stop reply notification, @value{GDBN} shall accept the
37799 notification, and the process shall be repeated.
37801 In non-stop mode, the target shall respond to the @samp{?} packet as
37802 follows. First, any incomplete stop reply notification/@samp{vStopped}
37803 sequence in progress is abandoned. The target must begin a new
37804 sequence reporting stop events for all stopped threads, whether or not
37805 it has previously reported those events to @value{GDBN}. The first
37806 stop reply is sent as a synchronous reply to the @samp{?} packet, and
37807 subsequent stop replies are sent as responses to @samp{vStopped} packets
37808 using the mechanism described above. The target must not send
37809 asynchronous stop reply notifications until the sequence is complete.
37810 If all threads are running when the target receives the @samp{?} packet,
37811 or if the target is not attached to any process, it shall respond
37814 @node Packet Acknowledgment
37815 @section Packet Acknowledgment
37817 @cindex acknowledgment, for @value{GDBN} remote
37818 @cindex packet acknowledgment, for @value{GDBN} remote
37819 By default, when either the host or the target machine receives a packet,
37820 the first response expected is an acknowledgment: either @samp{+} (to indicate
37821 the package was received correctly) or @samp{-} (to request retransmission).
37822 This mechanism allows the @value{GDBN} remote protocol to operate over
37823 unreliable transport mechanisms, such as a serial line.
37825 In cases where the transport mechanism is itself reliable (such as a pipe or
37826 TCP connection), the @samp{+}/@samp{-} acknowledgments are redundant.
37827 It may be desirable to disable them in that case to reduce communication
37828 overhead, or for other reasons. This can be accomplished by means of the
37829 @samp{QStartNoAckMode} packet; @pxref{QStartNoAckMode}.
37831 When in no-acknowledgment mode, neither the stub nor @value{GDBN} shall send or
37832 expect @samp{+}/@samp{-} protocol acknowledgments. The packet
37833 and response format still includes the normal checksum, as described in
37834 @ref{Overview}, but the checksum may be ignored by the receiver.
37836 If the stub supports @samp{QStartNoAckMode} and prefers to operate in
37837 no-acknowledgment mode, it should report that to @value{GDBN}
37838 by including @samp{QStartNoAckMode+} in its response to @samp{qSupported};
37839 @pxref{qSupported}.
37840 If @value{GDBN} also supports @samp{QStartNoAckMode} and it has not been
37841 disabled via the @code{set remote noack-packet off} command
37842 (@pxref{Remote Configuration}),
37843 @value{GDBN} may then send a @samp{QStartNoAckMode} packet to the stub.
37844 Only then may the stub actually turn off packet acknowledgments.
37845 @value{GDBN} sends a final @samp{+} acknowledgment of the stub's @samp{OK}
37846 response, which can be safely ignored by the stub.
37848 Note that @code{set remote noack-packet} command only affects negotiation
37849 between @value{GDBN} and the stub when subsequent connections are made;
37850 it does not affect the protocol acknowledgment state for any current
37852 Since @samp{+}/@samp{-} acknowledgments are enabled by default when a
37853 new connection is established,
37854 there is also no protocol request to re-enable the acknowledgments
37855 for the current connection, once disabled.
37860 Example sequence of a target being re-started. Notice how the restart
37861 does not get any direct output:
37866 @emph{target restarts}
37869 <- @code{T001:1234123412341234}
37873 Example sequence of a target being stepped by a single instruction:
37876 -> @code{G1445@dots{}}
37881 <- @code{T001:1234123412341234}
37885 <- @code{1455@dots{}}
37889 @node File-I/O Remote Protocol Extension
37890 @section File-I/O Remote Protocol Extension
37891 @cindex File-I/O remote protocol extension
37894 * File-I/O Overview::
37895 * Protocol Basics::
37896 * The F Request Packet::
37897 * The F Reply Packet::
37898 * The Ctrl-C Message::
37900 * List of Supported Calls::
37901 * Protocol-specific Representation of Datatypes::
37903 * File-I/O Examples::
37906 @node File-I/O Overview
37907 @subsection File-I/O Overview
37908 @cindex file-i/o overview
37910 The @dfn{File I/O remote protocol extension} (short: File-I/O) allows the
37911 target to use the host's file system and console I/O to perform various
37912 system calls. System calls on the target system are translated into a
37913 remote protocol packet to the host system, which then performs the needed
37914 actions and returns a response packet to the target system.
37915 This simulates file system operations even on targets that lack file systems.
37917 The protocol is defined to be independent of both the host and target systems.
37918 It uses its own internal representation of datatypes and values. Both
37919 @value{GDBN} and the target's @value{GDBN} stub are responsible for
37920 translating the system-dependent value representations into the internal
37921 protocol representations when data is transmitted.
37923 The communication is synchronous. A system call is possible only when
37924 @value{GDBN} is waiting for a response from the @samp{C}, @samp{c}, @samp{S}
37925 or @samp{s} packets. While @value{GDBN} handles the request for a system call,
37926 the target is stopped to allow deterministic access to the target's
37927 memory. Therefore File-I/O is not interruptible by target signals. On
37928 the other hand, it is possible to interrupt File-I/O by a user interrupt
37929 (@samp{Ctrl-C}) within @value{GDBN}.
37931 The target's request to perform a host system call does not finish
37932 the latest @samp{C}, @samp{c}, @samp{S} or @samp{s} action. That means,
37933 after finishing the system call, the target returns to continuing the
37934 previous activity (continue, step). No additional continue or step
37935 request from @value{GDBN} is required.
37938 (@value{GDBP}) continue
37939 <- target requests 'system call X'
37940 target is stopped, @value{GDBN} executes system call
37941 -> @value{GDBN} returns result
37942 ... target continues, @value{GDBN} returns to wait for the target
37943 <- target hits breakpoint and sends a Txx packet
37946 The protocol only supports I/O on the console and to regular files on
37947 the host file system. Character or block special devices, pipes,
37948 named pipes, sockets or any other communication method on the host
37949 system are not supported by this protocol.
37951 File I/O is not supported in non-stop mode.
37953 @node Protocol Basics
37954 @subsection Protocol Basics
37955 @cindex protocol basics, file-i/o
37957 The File-I/O protocol uses the @code{F} packet as the request as well
37958 as reply packet. Since a File-I/O system call can only occur when
37959 @value{GDBN} is waiting for a response from the continuing or stepping target,
37960 the File-I/O request is a reply that @value{GDBN} has to expect as a result
37961 of a previous @samp{C}, @samp{c}, @samp{S} or @samp{s} packet.
37962 This @code{F} packet contains all information needed to allow @value{GDBN}
37963 to call the appropriate host system call:
37967 A unique identifier for the requested system call.
37970 All parameters to the system call. Pointers are given as addresses
37971 in the target memory address space. Pointers to strings are given as
37972 pointer/length pair. Numerical values are given as they are.
37973 Numerical control flags are given in a protocol-specific representation.
37977 At this point, @value{GDBN} has to perform the following actions.
37981 If the parameters include pointer values to data needed as input to a
37982 system call, @value{GDBN} requests this data from the target with a
37983 standard @code{m} packet request. This additional communication has to be
37984 expected by the target implementation and is handled as any other @code{m}
37988 @value{GDBN} translates all value from protocol representation to host
37989 representation as needed. Datatypes are coerced into the host types.
37992 @value{GDBN} calls the system call.
37995 It then coerces datatypes back to protocol representation.
37998 If the system call is expected to return data in buffer space specified
37999 by pointer parameters to the call, the data is transmitted to the
38000 target using a @code{M} or @code{X} packet. This packet has to be expected
38001 by the target implementation and is handled as any other @code{M} or @code{X}
38006 Eventually @value{GDBN} replies with another @code{F} packet which contains all
38007 necessary information for the target to continue. This at least contains
38014 @code{errno}, if has been changed by the system call.
38021 After having done the needed type and value coercion, the target continues
38022 the latest continue or step action.
38024 @node The F Request Packet
38025 @subsection The @code{F} Request Packet
38026 @cindex file-i/o request packet
38027 @cindex @code{F} request packet
38029 The @code{F} request packet has the following format:
38032 @item F@var{call-id},@var{parameter@dots{}}
38034 @var{call-id} is the identifier to indicate the host system call to be called.
38035 This is just the name of the function.
38037 @var{parameter@dots{}} are the parameters to the system call.
38038 Parameters are hexadecimal integer values, either the actual values in case
38039 of scalar datatypes, pointers to target buffer space in case of compound
38040 datatypes and unspecified memory areas, or pointer/length pairs in case
38041 of string parameters. These are appended to the @var{call-id} as a
38042 comma-delimited list. All values are transmitted in ASCII
38043 string representation, pointer/length pairs separated by a slash.
38049 @node The F Reply Packet
38050 @subsection The @code{F} Reply Packet
38051 @cindex file-i/o reply packet
38052 @cindex @code{F} reply packet
38054 The @code{F} reply packet has the following format:
38058 @item F@var{retcode},@var{errno},@var{Ctrl-C flag};@var{call-specific attachment}
38060 @var{retcode} is the return code of the system call as hexadecimal value.
38062 @var{errno} is the @code{errno} set by the call, in protocol-specific
38064 This parameter can be omitted if the call was successful.
38066 @var{Ctrl-C flag} is only sent if the user requested a break. In this
38067 case, @var{errno} must be sent as well, even if the call was successful.
38068 The @var{Ctrl-C flag} itself consists of the character @samp{C}:
38075 or, if the call was interrupted before the host call has been performed:
38082 assuming 4 is the protocol-specific representation of @code{EINTR}.
38087 @node The Ctrl-C Message
38088 @subsection The @samp{Ctrl-C} Message
38089 @cindex ctrl-c message, in file-i/o protocol
38091 If the @samp{Ctrl-C} flag is set in the @value{GDBN}
38092 reply packet (@pxref{The F Reply Packet}),
38093 the target should behave as if it had
38094 gotten a break message. The meaning for the target is ``system call
38095 interrupted by @code{SIGINT}''. Consequentially, the target should actually stop
38096 (as with a break message) and return to @value{GDBN} with a @code{T02}
38099 It's important for the target to know in which
38100 state the system call was interrupted. There are two possible cases:
38104 The system call hasn't been performed on the host yet.
38107 The system call on the host has been finished.
38111 These two states can be distinguished by the target by the value of the
38112 returned @code{errno}. If it's the protocol representation of @code{EINTR}, the system
38113 call hasn't been performed. This is equivalent to the @code{EINTR} handling
38114 on POSIX systems. In any other case, the target may presume that the
38115 system call has been finished --- successfully or not --- and should behave
38116 as if the break message arrived right after the system call.
38118 @value{GDBN} must behave reliably. If the system call has not been called
38119 yet, @value{GDBN} may send the @code{F} reply immediately, setting @code{EINTR} as
38120 @code{errno} in the packet. If the system call on the host has been finished
38121 before the user requests a break, the full action must be finished by
38122 @value{GDBN}. This requires sending @code{M} or @code{X} packets as necessary.
38123 The @code{F} packet may only be sent when either nothing has happened
38124 or the full action has been completed.
38127 @subsection Console I/O
38128 @cindex console i/o as part of file-i/o
38130 By default and if not explicitly closed by the target system, the file
38131 descriptors 0, 1 and 2 are connected to the @value{GDBN} console. Output
38132 on the @value{GDBN} console is handled as any other file output operation
38133 (@code{write(1, @dots{})} or @code{write(2, @dots{})}). Console input is handled
38134 by @value{GDBN} so that after the target read request from file descriptor
38135 0 all following typing is buffered until either one of the following
38140 The user types @kbd{Ctrl-c}. The behaviour is as explained above, and the
38142 system call is treated as finished.
38145 The user presses @key{RET}. This is treated as end of input with a trailing
38149 The user types @kbd{Ctrl-d}. This is treated as end of input. No trailing
38150 character (neither newline nor @samp{Ctrl-D}) is appended to the input.
38154 If the user has typed more characters than fit in the buffer given to
38155 the @code{read} call, the trailing characters are buffered in @value{GDBN} until
38156 either another @code{read(0, @dots{})} is requested by the target, or debugging
38157 is stopped at the user's request.
38160 @node List of Supported Calls
38161 @subsection List of Supported Calls
38162 @cindex list of supported file-i/o calls
38179 @unnumberedsubsubsec open
38180 @cindex open, file-i/o system call
38185 int open(const char *pathname, int flags);
38186 int open(const char *pathname, int flags, mode_t mode);
38190 @samp{Fopen,@var{pathptr}/@var{len},@var{flags},@var{mode}}
38193 @var{flags} is the bitwise @code{OR} of the following values:
38197 If the file does not exist it will be created. The host
38198 rules apply as far as file ownership and time stamps
38202 When used with @code{O_CREAT}, if the file already exists it is
38203 an error and open() fails.
38206 If the file already exists and the open mode allows
38207 writing (@code{O_RDWR} or @code{O_WRONLY} is given) it will be
38208 truncated to zero length.
38211 The file is opened in append mode.
38214 The file is opened for reading only.
38217 The file is opened for writing only.
38220 The file is opened for reading and writing.
38224 Other bits are silently ignored.
38228 @var{mode} is the bitwise @code{OR} of the following values:
38232 User has read permission.
38235 User has write permission.
38238 Group has read permission.
38241 Group has write permission.
38244 Others have read permission.
38247 Others have write permission.
38251 Other bits are silently ignored.
38254 @item Return value:
38255 @code{open} returns the new file descriptor or -1 if an error
38262 @var{pathname} already exists and @code{O_CREAT} and @code{O_EXCL} were used.
38265 @var{pathname} refers to a directory.
38268 The requested access is not allowed.
38271 @var{pathname} was too long.
38274 A directory component in @var{pathname} does not exist.
38277 @var{pathname} refers to a device, pipe, named pipe or socket.
38280 @var{pathname} refers to a file on a read-only filesystem and
38281 write access was requested.
38284 @var{pathname} is an invalid pointer value.
38287 No space on device to create the file.
38290 The process already has the maximum number of files open.
38293 The limit on the total number of files open on the system
38297 The call was interrupted by the user.
38303 @unnumberedsubsubsec close
38304 @cindex close, file-i/o system call
38313 @samp{Fclose,@var{fd}}
38315 @item Return value:
38316 @code{close} returns zero on success, or -1 if an error occurred.
38322 @var{fd} isn't a valid open file descriptor.
38325 The call was interrupted by the user.
38331 @unnumberedsubsubsec read
38332 @cindex read, file-i/o system call
38337 int read(int fd, void *buf, unsigned int count);
38341 @samp{Fread,@var{fd},@var{bufptr},@var{count}}
38343 @item Return value:
38344 On success, the number of bytes read is returned.
38345 Zero indicates end of file. If count is zero, read
38346 returns zero as well. On error, -1 is returned.
38352 @var{fd} is not a valid file descriptor or is not open for
38356 @var{bufptr} is an invalid pointer value.
38359 The call was interrupted by the user.
38365 @unnumberedsubsubsec write
38366 @cindex write, file-i/o system call
38371 int write(int fd, const void *buf, unsigned int count);
38375 @samp{Fwrite,@var{fd},@var{bufptr},@var{count}}
38377 @item Return value:
38378 On success, the number of bytes written are returned.
38379 Zero indicates nothing was written. On error, -1
38386 @var{fd} is not a valid file descriptor or is not open for
38390 @var{bufptr} is an invalid pointer value.
38393 An attempt was made to write a file that exceeds the
38394 host-specific maximum file size allowed.
38397 No space on device to write the data.
38400 The call was interrupted by the user.
38406 @unnumberedsubsubsec lseek
38407 @cindex lseek, file-i/o system call
38412 long lseek (int fd, long offset, int flag);
38416 @samp{Flseek,@var{fd},@var{offset},@var{flag}}
38418 @var{flag} is one of:
38422 The offset is set to @var{offset} bytes.
38425 The offset is set to its current location plus @var{offset}
38429 The offset is set to the size of the file plus @var{offset}
38433 @item Return value:
38434 On success, the resulting unsigned offset in bytes from
38435 the beginning of the file is returned. Otherwise, a
38436 value of -1 is returned.
38442 @var{fd} is not a valid open file descriptor.
38445 @var{fd} is associated with the @value{GDBN} console.
38448 @var{flag} is not a proper value.
38451 The call was interrupted by the user.
38457 @unnumberedsubsubsec rename
38458 @cindex rename, file-i/o system call
38463 int rename(const char *oldpath, const char *newpath);
38467 @samp{Frename,@var{oldpathptr}/@var{len},@var{newpathptr}/@var{len}}
38469 @item Return value:
38470 On success, zero is returned. On error, -1 is returned.
38476 @var{newpath} is an existing directory, but @var{oldpath} is not a
38480 @var{newpath} is a non-empty directory.
38483 @var{oldpath} or @var{newpath} is a directory that is in use by some
38487 An attempt was made to make a directory a subdirectory
38491 A component used as a directory in @var{oldpath} or new
38492 path is not a directory. Or @var{oldpath} is a directory
38493 and @var{newpath} exists but is not a directory.
38496 @var{oldpathptr} or @var{newpathptr} are invalid pointer values.
38499 No access to the file or the path of the file.
38503 @var{oldpath} or @var{newpath} was too long.
38506 A directory component in @var{oldpath} or @var{newpath} does not exist.
38509 The file is on a read-only filesystem.
38512 The device containing the file has no room for the new
38516 The call was interrupted by the user.
38522 @unnumberedsubsubsec unlink
38523 @cindex unlink, file-i/o system call
38528 int unlink(const char *pathname);
38532 @samp{Funlink,@var{pathnameptr}/@var{len}}
38534 @item Return value:
38535 On success, zero is returned. On error, -1 is returned.
38541 No access to the file or the path of the file.
38544 The system does not allow unlinking of directories.
38547 The file @var{pathname} cannot be unlinked because it's
38548 being used by another process.
38551 @var{pathnameptr} is an invalid pointer value.
38554 @var{pathname} was too long.
38557 A directory component in @var{pathname} does not exist.
38560 A component of the path is not a directory.
38563 The file is on a read-only filesystem.
38566 The call was interrupted by the user.
38572 @unnumberedsubsubsec stat/fstat
38573 @cindex fstat, file-i/o system call
38574 @cindex stat, file-i/o system call
38579 int stat(const char *pathname, struct stat *buf);
38580 int fstat(int fd, struct stat *buf);
38584 @samp{Fstat,@var{pathnameptr}/@var{len},@var{bufptr}}@*
38585 @samp{Ffstat,@var{fd},@var{bufptr}}
38587 @item Return value:
38588 On success, zero is returned. On error, -1 is returned.
38594 @var{fd} is not a valid open file.
38597 A directory component in @var{pathname} does not exist or the
38598 path is an empty string.
38601 A component of the path is not a directory.
38604 @var{pathnameptr} is an invalid pointer value.
38607 No access to the file or the path of the file.
38610 @var{pathname} was too long.
38613 The call was interrupted by the user.
38619 @unnumberedsubsubsec gettimeofday
38620 @cindex gettimeofday, file-i/o system call
38625 int gettimeofday(struct timeval *tv, void *tz);
38629 @samp{Fgettimeofday,@var{tvptr},@var{tzptr}}
38631 @item Return value:
38632 On success, 0 is returned, -1 otherwise.
38638 @var{tz} is a non-NULL pointer.
38641 @var{tvptr} and/or @var{tzptr} is an invalid pointer value.
38647 @unnumberedsubsubsec isatty
38648 @cindex isatty, file-i/o system call
38653 int isatty(int fd);
38657 @samp{Fisatty,@var{fd}}
38659 @item Return value:
38660 Returns 1 if @var{fd} refers to the @value{GDBN} console, 0 otherwise.
38666 The call was interrupted by the user.
38671 Note that the @code{isatty} call is treated as a special case: it returns
38672 1 to the target if the file descriptor is attached
38673 to the @value{GDBN} console, 0 otherwise. Implementing through system calls
38674 would require implementing @code{ioctl} and would be more complex than
38679 @unnumberedsubsubsec system
38680 @cindex system, file-i/o system call
38685 int system(const char *command);
38689 @samp{Fsystem,@var{commandptr}/@var{len}}
38691 @item Return value:
38692 If @var{len} is zero, the return value indicates whether a shell is
38693 available. A zero return value indicates a shell is not available.
38694 For non-zero @var{len}, the value returned is -1 on error and the
38695 return status of the command otherwise. Only the exit status of the
38696 command is returned, which is extracted from the host's @code{system}
38697 return value by calling @code{WEXITSTATUS(retval)}. In case
38698 @file{/bin/sh} could not be executed, 127 is returned.
38704 The call was interrupted by the user.
38709 @value{GDBN} takes over the full task of calling the necessary host calls
38710 to perform the @code{system} call. The return value of @code{system} on
38711 the host is simplified before it's returned
38712 to the target. Any termination signal information from the child process
38713 is discarded, and the return value consists
38714 entirely of the exit status of the called command.
38716 Due to security concerns, the @code{system} call is by default refused
38717 by @value{GDBN}. The user has to allow this call explicitly with the
38718 @code{set remote system-call-allowed 1} command.
38721 @item set remote system-call-allowed
38722 @kindex set remote system-call-allowed
38723 Control whether to allow the @code{system} calls in the File I/O
38724 protocol for the remote target. The default is zero (disabled).
38726 @item show remote system-call-allowed
38727 @kindex show remote system-call-allowed
38728 Show whether the @code{system} calls are allowed in the File I/O
38732 @node Protocol-specific Representation of Datatypes
38733 @subsection Protocol-specific Representation of Datatypes
38734 @cindex protocol-specific representation of datatypes, in file-i/o protocol
38737 * Integral Datatypes::
38739 * Memory Transfer::
38744 @node Integral Datatypes
38745 @unnumberedsubsubsec Integral Datatypes
38746 @cindex integral datatypes, in file-i/o protocol
38748 The integral datatypes used in the system calls are @code{int},
38749 @code{unsigned int}, @code{long}, @code{unsigned long},
38750 @code{mode_t}, and @code{time_t}.
38752 @code{int}, @code{unsigned int}, @code{mode_t} and @code{time_t} are
38753 implemented as 32 bit values in this protocol.
38755 @code{long} and @code{unsigned long} are implemented as 64 bit types.
38757 @xref{Limits}, for corresponding MIN and MAX values (similar to those
38758 in @file{limits.h}) to allow range checking on host and target.
38760 @code{time_t} datatypes are defined as seconds since the Epoch.
38762 All integral datatypes transferred as part of a memory read or write of a
38763 structured datatype e.g.@: a @code{struct stat} have to be given in big endian
38766 @node Pointer Values
38767 @unnumberedsubsubsec Pointer Values
38768 @cindex pointer values, in file-i/o protocol
38770 Pointers to target data are transmitted as they are. An exception
38771 is made for pointers to buffers for which the length isn't
38772 transmitted as part of the function call, namely strings. Strings
38773 are transmitted as a pointer/length pair, both as hex values, e.g.@:
38780 which is a pointer to data of length 18 bytes at position 0x1aaf.
38781 The length is defined as the full string length in bytes, including
38782 the trailing null byte. For example, the string @code{"hello world"}
38783 at address 0x123456 is transmitted as
38789 @node Memory Transfer
38790 @unnumberedsubsubsec Memory Transfer
38791 @cindex memory transfer, in file-i/o protocol
38793 Structured data which is transferred using a memory read or write (for
38794 example, a @code{struct stat}) is expected to be in a protocol-specific format
38795 with all scalar multibyte datatypes being big endian. Translation to
38796 this representation needs to be done both by the target before the @code{F}
38797 packet is sent, and by @value{GDBN} before
38798 it transfers memory to the target. Transferred pointers to structured
38799 data should point to the already-coerced data at any time.
38803 @unnumberedsubsubsec struct stat
38804 @cindex struct stat, in file-i/o protocol
38806 The buffer of type @code{struct stat} used by the target and @value{GDBN}
38807 is defined as follows:
38811 unsigned int st_dev; /* device */
38812 unsigned int st_ino; /* inode */
38813 mode_t st_mode; /* protection */
38814 unsigned int st_nlink; /* number of hard links */
38815 unsigned int st_uid; /* user ID of owner */
38816 unsigned int st_gid; /* group ID of owner */
38817 unsigned int st_rdev; /* device type (if inode device) */
38818 unsigned long st_size; /* total size, in bytes */
38819 unsigned long st_blksize; /* blocksize for filesystem I/O */
38820 unsigned long st_blocks; /* number of blocks allocated */
38821 time_t st_atime; /* time of last access */
38822 time_t st_mtime; /* time of last modification */
38823 time_t st_ctime; /* time of last change */
38827 The integral datatypes conform to the definitions given in the
38828 appropriate section (see @ref{Integral Datatypes}, for details) so this
38829 structure is of size 64 bytes.
38831 The values of several fields have a restricted meaning and/or
38837 A value of 0 represents a file, 1 the console.
38840 No valid meaning for the target. Transmitted unchanged.
38843 Valid mode bits are described in @ref{Constants}. Any other
38844 bits have currently no meaning for the target.
38849 No valid meaning for the target. Transmitted unchanged.
38854 These values have a host and file system dependent
38855 accuracy. Especially on Windows hosts, the file system may not
38856 support exact timing values.
38859 The target gets a @code{struct stat} of the above representation and is
38860 responsible for coercing it to the target representation before
38863 Note that due to size differences between the host, target, and protocol
38864 representations of @code{struct stat} members, these members could eventually
38865 get truncated on the target.
38867 @node struct timeval
38868 @unnumberedsubsubsec struct timeval
38869 @cindex struct timeval, in file-i/o protocol
38871 The buffer of type @code{struct timeval} used by the File-I/O protocol
38872 is defined as follows:
38876 time_t tv_sec; /* second */
38877 long tv_usec; /* microsecond */
38881 The integral datatypes conform to the definitions given in the
38882 appropriate section (see @ref{Integral Datatypes}, for details) so this
38883 structure is of size 8 bytes.
38886 @subsection Constants
38887 @cindex constants, in file-i/o protocol
38889 The following values are used for the constants inside of the
38890 protocol. @value{GDBN} and target are responsible for translating these
38891 values before and after the call as needed.
38902 @unnumberedsubsubsec Open Flags
38903 @cindex open flags, in file-i/o protocol
38905 All values are given in hexadecimal representation.
38917 @node mode_t Values
38918 @unnumberedsubsubsec mode_t Values
38919 @cindex mode_t values, in file-i/o protocol
38921 All values are given in octal representation.
38938 @unnumberedsubsubsec Errno Values
38939 @cindex errno values, in file-i/o protocol
38941 All values are given in decimal representation.
38966 @code{EUNKNOWN} is used as a fallback error value if a host system returns
38967 any error value not in the list of supported error numbers.
38970 @unnumberedsubsubsec Lseek Flags
38971 @cindex lseek flags, in file-i/o protocol
38980 @unnumberedsubsubsec Limits
38981 @cindex limits, in file-i/o protocol
38983 All values are given in decimal representation.
38986 INT_MIN -2147483648
38988 UINT_MAX 4294967295
38989 LONG_MIN -9223372036854775808
38990 LONG_MAX 9223372036854775807
38991 ULONG_MAX 18446744073709551615
38994 @node File-I/O Examples
38995 @subsection File-I/O Examples
38996 @cindex file-i/o examples
38998 Example sequence of a write call, file descriptor 3, buffer is at target
38999 address 0x1234, 6 bytes should be written:
39002 <- @code{Fwrite,3,1234,6}
39003 @emph{request memory read from target}
39006 @emph{return "6 bytes written"}
39010 Example sequence of a read call, file descriptor 3, buffer is at target
39011 address 0x1234, 6 bytes should be read:
39014 <- @code{Fread,3,1234,6}
39015 @emph{request memory write to target}
39016 -> @code{X1234,6:XXXXXX}
39017 @emph{return "6 bytes read"}
39021 Example sequence of a read call, call fails on the host due to invalid
39022 file descriptor (@code{EBADF}):
39025 <- @code{Fread,3,1234,6}
39029 Example sequence of a read call, user presses @kbd{Ctrl-c} before syscall on
39033 <- @code{Fread,3,1234,6}
39038 Example sequence of a read call, user presses @kbd{Ctrl-c} after syscall on
39042 <- @code{Fread,3,1234,6}
39043 -> @code{X1234,6:XXXXXX}
39047 @node Library List Format
39048 @section Library List Format
39049 @cindex library list format, remote protocol
39051 On some platforms, a dynamic loader (e.g.@: @file{ld.so}) runs in the
39052 same process as your application to manage libraries. In this case,
39053 @value{GDBN} can use the loader's symbol table and normal memory
39054 operations to maintain a list of shared libraries. On other
39055 platforms, the operating system manages loaded libraries.
39056 @value{GDBN} can not retrieve the list of currently loaded libraries
39057 through memory operations, so it uses the @samp{qXfer:libraries:read}
39058 packet (@pxref{qXfer library list read}) instead. The remote stub
39059 queries the target's operating system and reports which libraries
39062 The @samp{qXfer:libraries:read} packet returns an XML document which
39063 lists loaded libraries and their offsets. Each library has an
39064 associated name and one or more segment or section base addresses,
39065 which report where the library was loaded in memory.
39067 For the common case of libraries that are fully linked binaries, the
39068 library should have a list of segments. If the target supports
39069 dynamic linking of a relocatable object file, its library XML element
39070 should instead include a list of allocated sections. The segment or
39071 section bases are start addresses, not relocation offsets; they do not
39072 depend on the library's link-time base addresses.
39074 @value{GDBN} must be linked with the Expat library to support XML
39075 library lists. @xref{Expat}.
39077 A simple memory map, with one loaded library relocated by a single
39078 offset, looks like this:
39082 <library name="/lib/libc.so.6">
39083 <segment address="0x10000000"/>
39088 Another simple memory map, with one loaded library with three
39089 allocated sections (.text, .data, .bss), looks like this:
39093 <library name="sharedlib.o">
39094 <section address="0x10000000"/>
39095 <section address="0x20000000"/>
39096 <section address="0x30000000"/>
39101 The format of a library list is described by this DTD:
39104 <!-- library-list: Root element with versioning -->
39105 <!ELEMENT library-list (library)*>
39106 <!ATTLIST library-list version CDATA #FIXED "1.0">
39107 <!ELEMENT library (segment*, section*)>
39108 <!ATTLIST library name CDATA #REQUIRED>
39109 <!ELEMENT segment EMPTY>
39110 <!ATTLIST segment address CDATA #REQUIRED>
39111 <!ELEMENT section EMPTY>
39112 <!ATTLIST section address CDATA #REQUIRED>
39115 In addition, segments and section descriptors cannot be mixed within a
39116 single library element, and you must supply at least one segment or
39117 section for each library.
39119 @node Library List Format for SVR4 Targets
39120 @section Library List Format for SVR4 Targets
39121 @cindex library list format, remote protocol
39123 On SVR4 platforms @value{GDBN} can use the symbol table of a dynamic loader
39124 (e.g.@: @file{ld.so}) and normal memory operations to maintain a list of
39125 shared libraries. Still a special library list provided by this packet is
39126 more efficient for the @value{GDBN} remote protocol.
39128 The @samp{qXfer:libraries-svr4:read} packet returns an XML document which lists
39129 loaded libraries and their SVR4 linker parameters. For each library on SVR4
39130 target, the following parameters are reported:
39134 @code{name}, the absolute file name from the @code{l_name} field of
39135 @code{struct link_map}.
39137 @code{lm} with address of @code{struct link_map} used for TLS
39138 (Thread Local Storage) access.
39140 @code{l_addr}, the displacement as read from the field @code{l_addr} of
39141 @code{struct link_map}. For prelinked libraries this is not an absolute
39142 memory address. It is a displacement of absolute memory address against
39143 address the file was prelinked to during the library load.
39145 @code{l_ld}, which is memory address of the @code{PT_DYNAMIC} segment
39148 Additionally the single @code{main-lm} attribute specifies address of
39149 @code{struct link_map} used for the main executable. This parameter is used
39150 for TLS access and its presence is optional.
39152 @value{GDBN} must be linked with the Expat library to support XML
39153 SVR4 library lists. @xref{Expat}.
39155 A simple memory map, with two loaded libraries (which do not use prelink),
39159 <library-list-svr4 version="1.0" main-lm="0xe4f8f8">
39160 <library name="/lib/ld-linux.so.2" lm="0xe4f51c" l_addr="0xe2d000"
39162 <library name="/lib/libc.so.6" lm="0xe4fbe8" l_addr="0x154000"
39164 </library-list-svr>
39167 The format of an SVR4 library list is described by this DTD:
39170 <!-- library-list-svr4: Root element with versioning -->
39171 <!ELEMENT library-list-svr4 (library)*>
39172 <!ATTLIST library-list-svr4 version CDATA #FIXED "1.0">
39173 <!ATTLIST library-list-svr4 main-lm CDATA #IMPLIED>
39174 <!ELEMENT library EMPTY>
39175 <!ATTLIST library name CDATA #REQUIRED>
39176 <!ATTLIST library lm CDATA #REQUIRED>
39177 <!ATTLIST library l_addr CDATA #REQUIRED>
39178 <!ATTLIST library l_ld CDATA #REQUIRED>
39181 @node Memory Map Format
39182 @section Memory Map Format
39183 @cindex memory map format
39185 To be able to write into flash memory, @value{GDBN} needs to obtain a
39186 memory map from the target. This section describes the format of the
39189 The memory map is obtained using the @samp{qXfer:memory-map:read}
39190 (@pxref{qXfer memory map read}) packet and is an XML document that
39191 lists memory regions.
39193 @value{GDBN} must be linked with the Expat library to support XML
39194 memory maps. @xref{Expat}.
39196 The top-level structure of the document is shown below:
39199 <?xml version="1.0"?>
39200 <!DOCTYPE memory-map
39201 PUBLIC "+//IDN gnu.org//DTD GDB Memory Map V1.0//EN"
39202 "http://sourceware.org/gdb/gdb-memory-map.dtd">
39208 Each region can be either:
39213 A region of RAM starting at @var{addr} and extending for @var{length}
39217 <memory type="ram" start="@var{addr}" length="@var{length}"/>
39222 A region of read-only memory:
39225 <memory type="rom" start="@var{addr}" length="@var{length}"/>
39230 A region of flash memory, with erasure blocks @var{blocksize}
39234 <memory type="flash" start="@var{addr}" length="@var{length}">
39235 <property name="blocksize">@var{blocksize}</property>
39241 Regions must not overlap. @value{GDBN} assumes that areas of memory not covered
39242 by the memory map are RAM, and uses the ordinary @samp{M} and @samp{X}
39243 packets to write to addresses in such ranges.
39245 The formal DTD for memory map format is given below:
39248 <!-- ................................................... -->
39249 <!-- Memory Map XML DTD ................................ -->
39250 <!-- File: memory-map.dtd .............................. -->
39251 <!-- .................................... .............. -->
39252 <!-- memory-map.dtd -->
39253 <!-- memory-map: Root element with versioning -->
39254 <!ELEMENT memory-map (memory | property)>
39255 <!ATTLIST memory-map version CDATA #FIXED "1.0.0">
39256 <!ELEMENT memory (property)>
39257 <!-- memory: Specifies a memory region,
39258 and its type, or device. -->
39259 <!ATTLIST memory type CDATA #REQUIRED
39260 start CDATA #REQUIRED
39261 length CDATA #REQUIRED
39262 device CDATA #IMPLIED>
39263 <!-- property: Generic attribute tag -->
39264 <!ELEMENT property (#PCDATA | property)*>
39265 <!ATTLIST property name CDATA #REQUIRED>
39268 @node Thread List Format
39269 @section Thread List Format
39270 @cindex thread list format
39272 To efficiently update the list of threads and their attributes,
39273 @value{GDBN} issues the @samp{qXfer:threads:read} packet
39274 (@pxref{qXfer threads read}) and obtains the XML document with
39275 the following structure:
39278 <?xml version="1.0"?>
39280 <thread id="id" core="0">
39281 ... description ...
39286 Each @samp{thread} element must have the @samp{id} attribute that
39287 identifies the thread (@pxref{thread-id syntax}). The
39288 @samp{core} attribute, if present, specifies which processor core
39289 the thread was last executing on. The content of the of @samp{thread}
39290 element is interpreted as human-readable auxilliary information.
39292 @node Traceframe Info Format
39293 @section Traceframe Info Format
39294 @cindex traceframe info format
39296 To be able to know which objects in the inferior can be examined when
39297 inspecting a tracepoint hit, @value{GDBN} needs to obtain the list of
39298 memory ranges, registers and trace state variables that have been
39299 collected in a traceframe.
39301 This list is obtained using the @samp{qXfer:traceframe-info:read}
39302 (@pxref{qXfer traceframe info read}) packet and is an XML document.
39304 @value{GDBN} must be linked with the Expat library to support XML
39305 traceframe info discovery. @xref{Expat}.
39307 The top-level structure of the document is shown below:
39310 <?xml version="1.0"?>
39311 <!DOCTYPE traceframe-info
39312 PUBLIC "+//IDN gnu.org//DTD GDB Memory Map V1.0//EN"
39313 "http://sourceware.org/gdb/gdb-traceframe-info.dtd">
39319 Each traceframe block can be either:
39324 A region of collected memory starting at @var{addr} and extending for
39325 @var{length} bytes from there:
39328 <memory start="@var{addr}" length="@var{length}"/>
39333 The formal DTD for the traceframe info format is given below:
39336 <!ELEMENT traceframe-info (memory)* >
39337 <!ATTLIST traceframe-info version CDATA #FIXED "1.0">
39339 <!ELEMENT memory EMPTY>
39340 <!ATTLIST memory start CDATA #REQUIRED
39341 length CDATA #REQUIRED>
39344 @include agentexpr.texi
39346 @node Target Descriptions
39347 @appendix Target Descriptions
39348 @cindex target descriptions
39350 One of the challenges of using @value{GDBN} to debug embedded systems
39351 is that there are so many minor variants of each processor
39352 architecture in use. It is common practice for vendors to start with
39353 a standard processor core --- ARM, PowerPC, or @acronym{MIPS}, for example ---
39354 and then make changes to adapt it to a particular market niche. Some
39355 architectures have hundreds of variants, available from dozens of
39356 vendors. This leads to a number of problems:
39360 With so many different customized processors, it is difficult for
39361 the @value{GDBN} maintainers to keep up with the changes.
39363 Since individual variants may have short lifetimes or limited
39364 audiences, it may not be worthwhile to carry information about every
39365 variant in the @value{GDBN} source tree.
39367 When @value{GDBN} does support the architecture of the embedded system
39368 at hand, the task of finding the correct architecture name to give the
39369 @command{set architecture} command can be error-prone.
39372 To address these problems, the @value{GDBN} remote protocol allows a
39373 target system to not only identify itself to @value{GDBN}, but to
39374 actually describe its own features. This lets @value{GDBN} support
39375 processor variants it has never seen before --- to the extent that the
39376 descriptions are accurate, and that @value{GDBN} understands them.
39378 @value{GDBN} must be linked with the Expat library to support XML
39379 target descriptions. @xref{Expat}.
39382 * Retrieving Descriptions:: How descriptions are fetched from a target.
39383 * Target Description Format:: The contents of a target description.
39384 * Predefined Target Types:: Standard types available for target
39386 * Standard Target Features:: Features @value{GDBN} knows about.
39389 @node Retrieving Descriptions
39390 @section Retrieving Descriptions
39392 Target descriptions can be read from the target automatically, or
39393 specified by the user manually. The default behavior is to read the
39394 description from the target. @value{GDBN} retrieves it via the remote
39395 protocol using @samp{qXfer} requests (@pxref{General Query Packets,
39396 qXfer}). The @var{annex} in the @samp{qXfer} packet will be
39397 @samp{target.xml}. The contents of the @samp{target.xml} annex are an
39398 XML document, of the form described in @ref{Target Description
39401 Alternatively, you can specify a file to read for the target description.
39402 If a file is set, the target will not be queried. The commands to
39403 specify a file are:
39406 @cindex set tdesc filename
39407 @item set tdesc filename @var{path}
39408 Read the target description from @var{path}.
39410 @cindex unset tdesc filename
39411 @item unset tdesc filename
39412 Do not read the XML target description from a file. @value{GDBN}
39413 will use the description supplied by the current target.
39415 @cindex show tdesc filename
39416 @item show tdesc filename
39417 Show the filename to read for a target description, if any.
39421 @node Target Description Format
39422 @section Target Description Format
39423 @cindex target descriptions, XML format
39425 A target description annex is an @uref{http://www.w3.org/XML/, XML}
39426 document which complies with the Document Type Definition provided in
39427 the @value{GDBN} sources in @file{gdb/features/gdb-target.dtd}. This
39428 means you can use generally available tools like @command{xmllint} to
39429 check that your feature descriptions are well-formed and valid.
39430 However, to help people unfamiliar with XML write descriptions for
39431 their targets, we also describe the grammar here.
39433 Target descriptions can identify the architecture of the remote target
39434 and (for some architectures) provide information about custom register
39435 sets. They can also identify the OS ABI of the remote target.
39436 @value{GDBN} can use this information to autoconfigure for your
39437 target, or to warn you if you connect to an unsupported target.
39439 Here is a simple target description:
39442 <target version="1.0">
39443 <architecture>i386:x86-64</architecture>
39448 This minimal description only says that the target uses
39449 the x86-64 architecture.
39451 A target description has the following overall form, with [ ] marking
39452 optional elements and @dots{} marking repeatable elements. The elements
39453 are explained further below.
39456 <?xml version="1.0"?>
39457 <!DOCTYPE target SYSTEM "gdb-target.dtd">
39458 <target version="1.0">
39459 @r{[}@var{architecture}@r{]}
39460 @r{[}@var{osabi}@r{]}
39461 @r{[}@var{compatible}@r{]}
39462 @r{[}@var{feature}@dots{}@r{]}
39467 The description is generally insensitive to whitespace and line
39468 breaks, under the usual common-sense rules. The XML version
39469 declaration and document type declaration can generally be omitted
39470 (@value{GDBN} does not require them), but specifying them may be
39471 useful for XML validation tools. The @samp{version} attribute for
39472 @samp{<target>} may also be omitted, but we recommend
39473 including it; if future versions of @value{GDBN} use an incompatible
39474 revision of @file{gdb-target.dtd}, they will detect and report
39475 the version mismatch.
39477 @subsection Inclusion
39478 @cindex target descriptions, inclusion
39481 @cindex <xi:include>
39484 It can sometimes be valuable to split a target description up into
39485 several different annexes, either for organizational purposes, or to
39486 share files between different possible target descriptions. You can
39487 divide a description into multiple files by replacing any element of
39488 the target description with an inclusion directive of the form:
39491 <xi:include href="@var{document}"/>
39495 When @value{GDBN} encounters an element of this form, it will retrieve
39496 the named XML @var{document}, and replace the inclusion directive with
39497 the contents of that document. If the current description was read
39498 using @samp{qXfer}, then so will be the included document;
39499 @var{document} will be interpreted as the name of an annex. If the
39500 current description was read from a file, @value{GDBN} will look for
39501 @var{document} as a file in the same directory where it found the
39502 original description.
39504 @subsection Architecture
39505 @cindex <architecture>
39507 An @samp{<architecture>} element has this form:
39510 <architecture>@var{arch}</architecture>
39513 @var{arch} is one of the architectures from the set accepted by
39514 @code{set architecture} (@pxref{Targets, ,Specifying a Debugging Target}).
39517 @cindex @code{<osabi>}
39519 This optional field was introduced in @value{GDBN} version 7.0.
39520 Previous versions of @value{GDBN} ignore it.
39522 An @samp{<osabi>} element has this form:
39525 <osabi>@var{abi-name}</osabi>
39528 @var{abi-name} is an OS ABI name from the same selection accepted by
39529 @w{@code{set osabi}} (@pxref{ABI, ,Configuring the Current ABI}).
39531 @subsection Compatible Architecture
39532 @cindex @code{<compatible>}
39534 This optional field was introduced in @value{GDBN} version 7.0.
39535 Previous versions of @value{GDBN} ignore it.
39537 A @samp{<compatible>} element has this form:
39540 <compatible>@var{arch}</compatible>
39543 @var{arch} is one of the architectures from the set accepted by
39544 @code{set architecture} (@pxref{Targets, ,Specifying a Debugging Target}).
39546 A @samp{<compatible>} element is used to specify that the target
39547 is able to run binaries in some other than the main target architecture
39548 given by the @samp{<architecture>} element. For example, on the
39549 Cell Broadband Engine, the main architecture is @code{powerpc:common}
39550 or @code{powerpc:common64}, but the system is able to run binaries
39551 in the @code{spu} architecture as well. The way to describe this
39552 capability with @samp{<compatible>} is as follows:
39555 <architecture>powerpc:common</architecture>
39556 <compatible>spu</compatible>
39559 @subsection Features
39562 Each @samp{<feature>} describes some logical portion of the target
39563 system. Features are currently used to describe available CPU
39564 registers and the types of their contents. A @samp{<feature>} element
39568 <feature name="@var{name}">
39569 @r{[}@var{type}@dots{}@r{]}
39575 Each feature's name should be unique within the description. The name
39576 of a feature does not matter unless @value{GDBN} has some special
39577 knowledge of the contents of that feature; if it does, the feature
39578 should have its standard name. @xref{Standard Target Features}.
39582 Any register's value is a collection of bits which @value{GDBN} must
39583 interpret. The default interpretation is a two's complement integer,
39584 but other types can be requested by name in the register description.
39585 Some predefined types are provided by @value{GDBN} (@pxref{Predefined
39586 Target Types}), and the description can define additional composite types.
39588 Each type element must have an @samp{id} attribute, which gives
39589 a unique (within the containing @samp{<feature>}) name to the type.
39590 Types must be defined before they are used.
39593 Some targets offer vector registers, which can be treated as arrays
39594 of scalar elements. These types are written as @samp{<vector>} elements,
39595 specifying the array element type, @var{type}, and the number of elements,
39599 <vector id="@var{id}" type="@var{type}" count="@var{count}"/>
39603 If a register's value is usefully viewed in multiple ways, define it
39604 with a union type containing the useful representations. The
39605 @samp{<union>} element contains one or more @samp{<field>} elements,
39606 each of which has a @var{name} and a @var{type}:
39609 <union id="@var{id}">
39610 <field name="@var{name}" type="@var{type}"/>
39616 If a register's value is composed from several separate values, define
39617 it with a structure type. There are two forms of the @samp{<struct>}
39618 element; a @samp{<struct>} element must either contain only bitfields
39619 or contain no bitfields. If the structure contains only bitfields,
39620 its total size in bytes must be specified, each bitfield must have an
39621 explicit start and end, and bitfields are automatically assigned an
39622 integer type. The field's @var{start} should be less than or
39623 equal to its @var{end}, and zero represents the least significant bit.
39626 <struct id="@var{id}" size="@var{size}">
39627 <field name="@var{name}" start="@var{start}" end="@var{end}"/>
39632 If the structure contains no bitfields, then each field has an
39633 explicit type, and no implicit padding is added.
39636 <struct id="@var{id}">
39637 <field name="@var{name}" type="@var{type}"/>
39643 If a register's value is a series of single-bit flags, define it with
39644 a flags type. The @samp{<flags>} element has an explicit @var{size}
39645 and contains one or more @samp{<field>} elements. Each field has a
39646 @var{name}, a @var{start}, and an @var{end}. Only single-bit flags
39650 <flags id="@var{id}" size="@var{size}">
39651 <field name="@var{name}" start="@var{start}" end="@var{end}"/>
39656 @subsection Registers
39659 Each register is represented as an element with this form:
39662 <reg name="@var{name}"
39663 bitsize="@var{size}"
39664 @r{[}regnum="@var{num}"@r{]}
39665 @r{[}save-restore="@var{save-restore}"@r{]}
39666 @r{[}type="@var{type}"@r{]}
39667 @r{[}group="@var{group}"@r{]}/>
39671 The components are as follows:
39676 The register's name; it must be unique within the target description.
39679 The register's size, in bits.
39682 The register's number. If omitted, a register's number is one greater
39683 than that of the previous register (either in the current feature or in
39684 a preceding feature); the first register in the target description
39685 defaults to zero. This register number is used to read or write
39686 the register; e.g.@: it is used in the remote @code{p} and @code{P}
39687 packets, and registers appear in the @code{g} and @code{G} packets
39688 in order of increasing register number.
39691 Whether the register should be preserved across inferior function
39692 calls; this must be either @code{yes} or @code{no}. The default is
39693 @code{yes}, which is appropriate for most registers except for
39694 some system control registers; this is not related to the target's
39698 The type of the register. @var{type} may be a predefined type, a type
39699 defined in the current feature, or one of the special types @code{int}
39700 and @code{float}. @code{int} is an integer type of the correct size
39701 for @var{bitsize}, and @code{float} is a floating point type (in the
39702 architecture's normal floating point format) of the correct size for
39703 @var{bitsize}. The default is @code{int}.
39706 The register group to which this register belongs. @var{group} must
39707 be either @code{general}, @code{float}, or @code{vector}. If no
39708 @var{group} is specified, @value{GDBN} will not display the register
39709 in @code{info registers}.
39713 @node Predefined Target Types
39714 @section Predefined Target Types
39715 @cindex target descriptions, predefined types
39717 Type definitions in the self-description can build up composite types
39718 from basic building blocks, but can not define fundamental types. Instead,
39719 standard identifiers are provided by @value{GDBN} for the fundamental
39720 types. The currently supported types are:
39729 Signed integer types holding the specified number of bits.
39736 Unsigned integer types holding the specified number of bits.
39740 Pointers to unspecified code and data. The program counter and
39741 any dedicated return address register may be marked as code
39742 pointers; printing a code pointer converts it into a symbolic
39743 address. The stack pointer and any dedicated address registers
39744 may be marked as data pointers.
39747 Single precision IEEE floating point.
39750 Double precision IEEE floating point.
39753 The 12-byte extended precision format used by ARM FPA registers.
39756 The 10-byte extended precision format used by x87 registers.
39759 32bit @sc{eflags} register used by x86.
39762 32bit @sc{mxcsr} register used by x86.
39766 @node Standard Target Features
39767 @section Standard Target Features
39768 @cindex target descriptions, standard features
39770 A target description must contain either no registers or all the
39771 target's registers. If the description contains no registers, then
39772 @value{GDBN} will assume a default register layout, selected based on
39773 the architecture. If the description contains any registers, the
39774 default layout will not be used; the standard registers must be
39775 described in the target description, in such a way that @value{GDBN}
39776 can recognize them.
39778 This is accomplished by giving specific names to feature elements
39779 which contain standard registers. @value{GDBN} will look for features
39780 with those names and verify that they contain the expected registers;
39781 if any known feature is missing required registers, or if any required
39782 feature is missing, @value{GDBN} will reject the target
39783 description. You can add additional registers to any of the
39784 standard features --- @value{GDBN} will display them just as if
39785 they were added to an unrecognized feature.
39787 This section lists the known features and their expected contents.
39788 Sample XML documents for these features are included in the
39789 @value{GDBN} source tree, in the directory @file{gdb/features}.
39791 Names recognized by @value{GDBN} should include the name of the
39792 company or organization which selected the name, and the overall
39793 architecture to which the feature applies; so e.g.@: the feature
39794 containing ARM core registers is named @samp{org.gnu.gdb.arm.core}.
39796 The names of registers are not case sensitive for the purpose
39797 of recognizing standard features, but @value{GDBN} will only display
39798 registers using the capitalization used in the description.
39805 * PowerPC Features::
39811 @subsection ARM Features
39812 @cindex target descriptions, ARM features
39814 The @samp{org.gnu.gdb.arm.core} feature is required for non-M-profile
39816 It should contain registers @samp{r0} through @samp{r13}, @samp{sp},
39817 @samp{lr}, @samp{pc}, and @samp{cpsr}.
39819 For M-profile targets (e.g. Cortex-M3), the @samp{org.gnu.gdb.arm.core}
39820 feature is replaced by @samp{org.gnu.gdb.arm.m-profile}. It should contain
39821 registers @samp{r0} through @samp{r13}, @samp{sp}, @samp{lr}, @samp{pc},
39824 The @samp{org.gnu.gdb.arm.fpa} feature is optional. If present, it
39825 should contain registers @samp{f0} through @samp{f7} and @samp{fps}.
39827 The @samp{org.gnu.gdb.xscale.iwmmxt} feature is optional. If present,
39828 it should contain at least registers @samp{wR0} through @samp{wR15} and
39829 @samp{wCGR0} through @samp{wCGR3}. The @samp{wCID}, @samp{wCon},
39830 @samp{wCSSF}, and @samp{wCASF} registers are optional.
39832 The @samp{org.gnu.gdb.arm.vfp} feature is optional. If present, it
39833 should contain at least registers @samp{d0} through @samp{d15}. If
39834 they are present, @samp{d16} through @samp{d31} should also be included.
39835 @value{GDBN} will synthesize the single-precision registers from
39836 halves of the double-precision registers.
39838 The @samp{org.gnu.gdb.arm.neon} feature is optional. It does not
39839 need to contain registers; it instructs @value{GDBN} to display the
39840 VFP double-precision registers as vectors and to synthesize the
39841 quad-precision registers from pairs of double-precision registers.
39842 If this feature is present, @samp{org.gnu.gdb.arm.vfp} must also
39843 be present and include 32 double-precision registers.
39845 @node i386 Features
39846 @subsection i386 Features
39847 @cindex target descriptions, i386 features
39849 The @samp{org.gnu.gdb.i386.core} feature is required for i386/amd64
39850 targets. It should describe the following registers:
39854 @samp{eax} through @samp{edi} plus @samp{eip} for i386
39856 @samp{rax} through @samp{r15} plus @samp{rip} for amd64
39858 @samp{eflags}, @samp{cs}, @samp{ss}, @samp{ds}, @samp{es},
39859 @samp{fs}, @samp{gs}
39861 @samp{st0} through @samp{st7}
39863 @samp{fctrl}, @samp{fstat}, @samp{ftag}, @samp{fiseg}, @samp{fioff},
39864 @samp{foseg}, @samp{fooff} and @samp{fop}
39867 The register sets may be different, depending on the target.
39869 The @samp{org.gnu.gdb.i386.sse} feature is optional. It should
39870 describe registers:
39874 @samp{xmm0} through @samp{xmm7} for i386
39876 @samp{xmm0} through @samp{xmm15} for amd64
39881 The @samp{org.gnu.gdb.i386.avx} feature is optional and requires the
39882 @samp{org.gnu.gdb.i386.sse} feature. It should
39883 describe the upper 128 bits of @sc{ymm} registers:
39887 @samp{ymm0h} through @samp{ymm7h} for i386
39889 @samp{ymm0h} through @samp{ymm15h} for amd64
39892 The @samp{org.gnu.gdb.i386.linux} feature is optional. It should
39893 describe a single register, @samp{orig_eax}.
39895 @node MIPS Features
39896 @subsection @acronym{MIPS} Features
39897 @cindex target descriptions, @acronym{MIPS} features
39899 The @samp{org.gnu.gdb.mips.cpu} feature is required for @acronym{MIPS} targets.
39900 It should contain registers @samp{r0} through @samp{r31}, @samp{lo},
39901 @samp{hi}, and @samp{pc}. They may be 32-bit or 64-bit depending
39904 The @samp{org.gnu.gdb.mips.cp0} feature is also required. It should
39905 contain at least the @samp{status}, @samp{badvaddr}, and @samp{cause}
39906 registers. They may be 32-bit or 64-bit depending on the target.
39908 The @samp{org.gnu.gdb.mips.fpu} feature is currently required, though
39909 it may be optional in a future version of @value{GDBN}. It should
39910 contain registers @samp{f0} through @samp{f31}, @samp{fcsr}, and
39911 @samp{fir}. They may be 32-bit or 64-bit depending on the target.
39913 The @samp{org.gnu.gdb.mips.dsp} feature is optional. It should
39914 contain registers @samp{hi1} through @samp{hi3}, @samp{lo1} through
39915 @samp{lo3}, and @samp{dspctl}. The @samp{dspctl} register should
39916 be 32-bit and the rest may be 32-bit or 64-bit depending on the target.
39918 The @samp{org.gnu.gdb.mips.linux} feature is optional. It should
39919 contain a single register, @samp{restart}, which is used by the
39920 Linux kernel to control restartable syscalls.
39922 @node M68K Features
39923 @subsection M68K Features
39924 @cindex target descriptions, M68K features
39927 @item @samp{org.gnu.gdb.m68k.core}
39928 @itemx @samp{org.gnu.gdb.coldfire.core}
39929 @itemx @samp{org.gnu.gdb.fido.core}
39930 One of those features must be always present.
39931 The feature that is present determines which flavor of m68k is
39932 used. The feature that is present should contain registers
39933 @samp{d0} through @samp{d7}, @samp{a0} through @samp{a5}, @samp{fp},
39934 @samp{sp}, @samp{ps} and @samp{pc}.
39936 @item @samp{org.gnu.gdb.coldfire.fp}
39937 This feature is optional. If present, it should contain registers
39938 @samp{fp0} through @samp{fp7}, @samp{fpcontrol}, @samp{fpstatus} and
39942 @node PowerPC Features
39943 @subsection PowerPC Features
39944 @cindex target descriptions, PowerPC features
39946 The @samp{org.gnu.gdb.power.core} feature is required for PowerPC
39947 targets. It should contain registers @samp{r0} through @samp{r31},
39948 @samp{pc}, @samp{msr}, @samp{cr}, @samp{lr}, @samp{ctr}, and
39949 @samp{xer}. They may be 32-bit or 64-bit depending on the target.
39951 The @samp{org.gnu.gdb.power.fpu} feature is optional. It should
39952 contain registers @samp{f0} through @samp{f31} and @samp{fpscr}.
39954 The @samp{org.gnu.gdb.power.altivec} feature is optional. It should
39955 contain registers @samp{vr0} through @samp{vr31}, @samp{vscr},
39958 The @samp{org.gnu.gdb.power.vsx} feature is optional. It should
39959 contain registers @samp{vs0h} through @samp{vs31h}. @value{GDBN}
39960 will combine these registers with the floating point registers
39961 (@samp{f0} through @samp{f31}) and the altivec registers (@samp{vr0}
39962 through @samp{vr31}) to present the 128-bit wide registers @samp{vs0}
39963 through @samp{vs63}, the set of vector registers for POWER7.
39965 The @samp{org.gnu.gdb.power.spe} feature is optional. It should
39966 contain registers @samp{ev0h} through @samp{ev31h}, @samp{acc}, and
39967 @samp{spefscr}. SPE targets should provide 32-bit registers in
39968 @samp{org.gnu.gdb.power.core} and provide the upper halves in
39969 @samp{ev0h} through @samp{ev31h}. @value{GDBN} will combine
39970 these to present registers @samp{ev0} through @samp{ev31} to the
39973 @node TIC6x Features
39974 @subsection TMS320C6x Features
39975 @cindex target descriptions, TIC6x features
39976 @cindex target descriptions, TMS320C6x features
39977 The @samp{org.gnu.gdb.tic6x.core} feature is required for TMS320C6x
39978 targets. It should contain registers @samp{A0} through @samp{A15},
39979 registers @samp{B0} through @samp{B15}, @samp{CSR} and @samp{PC}.
39981 The @samp{org.gnu.gdb.tic6x.gp} feature is optional. It should
39982 contain registers @samp{A16} through @samp{A31} and @samp{B16}
39983 through @samp{B31}.
39985 The @samp{org.gnu.gdb.tic6x.c6xp} feature is optional. It should
39986 contain registers @samp{TSR}, @samp{ILC} and @samp{RILC}.
39988 @node Operating System Information
39989 @appendix Operating System Information
39990 @cindex operating system information
39996 Users of @value{GDBN} often wish to obtain information about the state of
39997 the operating system running on the target---for example the list of
39998 processes, or the list of open files. This section describes the
39999 mechanism that makes it possible. This mechanism is similar to the
40000 target features mechanism (@pxref{Target Descriptions}), but focuses
40001 on a different aspect of target.
40003 Operating system information is retrived from the target via the
40004 remote protocol, using @samp{qXfer} requests (@pxref{qXfer osdata
40005 read}). The object name in the request should be @samp{osdata}, and
40006 the @var{annex} identifies the data to be fetched.
40009 @appendixsection Process list
40010 @cindex operating system information, process list
40012 When requesting the process list, the @var{annex} field in the
40013 @samp{qXfer} request should be @samp{processes}. The returned data is
40014 an XML document. The formal syntax of this document is defined in
40015 @file{gdb/features/osdata.dtd}.
40017 An example document is:
40020 <?xml version="1.0"?>
40021 <!DOCTYPE target SYSTEM "osdata.dtd">
40022 <osdata type="processes">
40024 <column name="pid">1</column>
40025 <column name="user">root</column>
40026 <column name="command">/sbin/init</column>
40027 <column name="cores">1,2,3</column>
40032 Each item should include a column whose name is @samp{pid}. The value
40033 of that column should identify the process on the target. The
40034 @samp{user} and @samp{command} columns are optional, and will be
40035 displayed by @value{GDBN}. The @samp{cores} column, if present,
40036 should contain a comma-separated list of cores that this process
40037 is running on. Target may provide additional columns,
40038 which @value{GDBN} currently ignores.
40040 @node Trace File Format
40041 @appendix Trace File Format
40042 @cindex trace file format
40044 The trace file comes in three parts: a header, a textual description
40045 section, and a trace frame section with binary data.
40047 The header has the form @code{\x7fTRACE0\n}. The first byte is
40048 @code{0x7f} so as to indicate that the file contains binary data,
40049 while the @code{0} is a version number that may have different values
40052 The description section consists of multiple lines of @sc{ascii} text
40053 separated by newline characters (@code{0xa}). The lines may include a
40054 variety of optional descriptive or context-setting information, such
40055 as tracepoint definitions or register set size. @value{GDBN} will
40056 ignore any line that it does not recognize. An empty line marks the end
40059 @c FIXME add some specific types of data
40061 The trace frame section consists of a number of consecutive frames.
40062 Each frame begins with a two-byte tracepoint number, followed by a
40063 four-byte size giving the amount of data in the frame. The data in
40064 the frame consists of a number of blocks, each introduced by a
40065 character indicating its type (at least register, memory, and trace
40066 state variable). The data in this section is raw binary, not a
40067 hexadecimal or other encoding; its endianness matches the target's
40070 @c FIXME bi-arch may require endianness/arch info in description section
40073 @item R @var{bytes}
40074 Register block. The number and ordering of bytes matches that of a
40075 @code{g} packet in the remote protocol. Note that these are the
40076 actual bytes, in target order and @value{GDBN} register order, not a
40077 hexadecimal encoding.
40079 @item M @var{address} @var{length} @var{bytes}...
40080 Memory block. This is a contiguous block of memory, at the 8-byte
40081 address @var{address}, with a 2-byte length @var{length}, followed by
40082 @var{length} bytes.
40084 @item V @var{number} @var{value}
40085 Trace state variable block. This records the 8-byte signed value
40086 @var{value} of trace state variable numbered @var{number}.
40090 Future enhancements of the trace file format may include additional types
40093 @node Index Section Format
40094 @appendix @code{.gdb_index} section format
40095 @cindex .gdb_index section format
40096 @cindex index section format
40098 This section documents the index section that is created by @code{save
40099 gdb-index} (@pxref{Index Files}). The index section is
40100 DWARF-specific; some knowledge of DWARF is assumed in this
40103 The mapped index file format is designed to be directly
40104 @code{mmap}able on any architecture. In most cases, a datum is
40105 represented using a little-endian 32-bit integer value, called an
40106 @code{offset_type}. Big endian machines must byte-swap the values
40107 before using them. Exceptions to this rule are noted. The data is
40108 laid out such that alignment is always respected.
40110 A mapped index consists of several areas, laid out in order.
40114 The file header. This is a sequence of values, of @code{offset_type}
40115 unless otherwise noted:
40119 The version number, currently 6. Versions 1, 2 and 3 are obsolete.
40120 Version 4 uses a different hashing function from versions 5 and 6.
40121 Version 6 includes symbols for inlined functions, whereas versions
40122 4 and 5 do not. @value{GDBN} will only read version 4 and 5 indices
40123 if the @code{--use-deprecated-index-sections} option is used.
40126 The offset, from the start of the file, of the CU list.
40129 The offset, from the start of the file, of the types CU list. Note
40130 that this area can be empty, in which case this offset will be equal
40131 to the next offset.
40134 The offset, from the start of the file, of the address area.
40137 The offset, from the start of the file, of the symbol table.
40140 The offset, from the start of the file, of the constant pool.
40144 The CU list. This is a sequence of pairs of 64-bit little-endian
40145 values, sorted by the CU offset. The first element in each pair is
40146 the offset of a CU in the @code{.debug_info} section. The second
40147 element in each pair is the length of that CU. References to a CU
40148 elsewhere in the map are done using a CU index, which is just the
40149 0-based index into this table. Note that if there are type CUs, then
40150 conceptually CUs and type CUs form a single list for the purposes of
40154 The types CU list. This is a sequence of triplets of 64-bit
40155 little-endian values. In a triplet, the first value is the CU offset,
40156 the second value is the type offset in the CU, and the third value is
40157 the type signature. The types CU list is not sorted.
40160 The address area. The address area consists of a sequence of address
40161 entries. Each address entry has three elements:
40165 The low address. This is a 64-bit little-endian value.
40168 The high address. This is a 64-bit little-endian value. Like
40169 @code{DW_AT_high_pc}, the value is one byte beyond the end.
40172 The CU index. This is an @code{offset_type} value.
40176 The symbol table. This is an open-addressed hash table. The size of
40177 the hash table is always a power of 2.
40179 Each slot in the hash table consists of a pair of @code{offset_type}
40180 values. The first value is the offset of the symbol's name in the
40181 constant pool. The second value is the offset of the CU vector in the
40184 If both values are 0, then this slot in the hash table is empty. This
40185 is ok because while 0 is a valid constant pool index, it cannot be a
40186 valid index for both a string and a CU vector.
40188 The hash value for a table entry is computed by applying an
40189 iterative hash function to the symbol's name. Starting with an
40190 initial value of @code{r = 0}, each (unsigned) character @samp{c} in
40191 the string is incorporated into the hash using the formula depending on the
40196 The formula is @code{r = r * 67 + c - 113}.
40198 @item Versions 5 and 6
40199 The formula is @code{r = r * 67 + tolower (c) - 113}.
40202 The terminating @samp{\0} is not incorporated into the hash.
40204 The step size used in the hash table is computed via
40205 @code{((hash * 17) & (size - 1)) | 1}, where @samp{hash} is the hash
40206 value, and @samp{size} is the size of the hash table. The step size
40207 is used to find the next candidate slot when handling a hash
40210 The names of C@t{++} symbols in the hash table are canonicalized. We
40211 don't currently have a simple description of the canonicalization
40212 algorithm; if you intend to create new index sections, you must read
40216 The constant pool. This is simply a bunch of bytes. It is organized
40217 so that alignment is correct: CU vectors are stored first, followed by
40220 A CU vector in the constant pool is a sequence of @code{offset_type}
40221 values. The first value is the number of CU indices in the vector.
40222 Each subsequent value is the index of a CU in the CU list. This
40223 element in the hash table is used to indicate which CUs define the
40226 A string in the constant pool is zero-terminated.
40231 @node GNU Free Documentation License
40232 @appendix GNU Free Documentation License
40241 % I think something like @colophon should be in texinfo. In the
40243 \long\def\colophon{\hbox to0pt{}\vfill
40244 \centerline{The body of this manual is set in}
40245 \centerline{\fontname\tenrm,}
40246 \centerline{with headings in {\bf\fontname\tenbf}}
40247 \centerline{and examples in {\tt\fontname\tentt}.}
40248 \centerline{{\it\fontname\tenit\/},}
40249 \centerline{{\bf\fontname\tenbf}, and}
40250 \centerline{{\sl\fontname\tensl\/}}
40251 \centerline{are used for emphasis.}\vfill}
40253 % Blame: doc@cygnus.com, 1991.