1 \input texinfo @c -*-texinfo-*-
2 @c Copyright (C) 1988-2019 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.
13 @settitle Debugging with @value{GDBN}
14 @setchapternewpage odd
23 @c To avoid file-name clashes between index.html and Index.html, when
24 @c the manual is produced on a Posix host and then moved to a
25 @c case-insensitive filesystem (e.g., MS-Windows), we separate the
26 @c indices into two: Concept Index and all the rest.
30 @c readline appendices use @vindex, @findex and @ftable,
31 @c annotate.texi and gdbmi use @findex.
34 @c !!set GDB manual's edition---not the same as GDB version!
35 @c This is updated by GNU Press.
38 @c !!set GDB edit command default editor
41 @c THIS MANUAL REQUIRES TEXINFO 4.0 OR LATER.
43 @c This is a dir.info fragment to support semi-automated addition of
44 @c manuals to an info tree.
45 @dircategory Software development
47 * Gdb: (gdb). The GNU debugger.
48 * gdbserver: (gdb) Server. The GNU debugging server.
52 @c man begin COPYRIGHT
53 Copyright @copyright{} 1988-2019 Free Software Foundation, Inc.
55 Permission is granted to copy, distribute and/or modify this document
56 under the terms of the GNU Free Documentation License, Version 1.3 or
57 any later version published by the Free Software Foundation; with the
58 Invariant Sections being ``Free Software'' and ``Free Software Needs
59 Free Documentation'', with the Front-Cover Texts being ``A GNU Manual,''
60 and with the Back-Cover Texts as in (a) below.
62 (a) The FSF's Back-Cover Text is: ``You are free to copy and modify
63 this GNU Manual. Buying copies from GNU Press supports the FSF in
64 developing GNU and promoting software freedom.''
69 This file documents the @sc{gnu} debugger @value{GDBN}.
71 This is the @value{EDITION} Edition, of @cite{Debugging with
72 @value{GDBN}: the @sc{gnu} Source-Level Debugger} for @value{GDBN}
73 @ifset VERSION_PACKAGE
74 @value{VERSION_PACKAGE}
76 Version @value{GDBVN}.
82 @title Debugging with @value{GDBN}
83 @subtitle The @sc{gnu} Source-Level Debugger
85 @subtitle @value{EDITION} Edition, for @value{GDBN} version @value{GDBVN}
86 @ifset VERSION_PACKAGE
88 @subtitle @value{VERSION_PACKAGE}
90 @author Richard Stallman, Roland Pesch, Stan Shebs, et al.
94 \hfill (Send bugs and comments on @value{GDBN} to @value{BUGURL}.)\par
95 \hfill {\it Debugging with @value{GDBN}}\par
96 \hfill \TeX{}info \texinfoversion\par
100 @vskip 0pt plus 1filll
101 Published by the Free Software Foundation @*
102 51 Franklin Street, Fifth Floor,
103 Boston, MA 02110-1301, USA@*
104 ISBN 978-0-9831592-3-0 @*
111 @node Top, Summary, (dir), (dir)
113 @top Debugging with @value{GDBN}
115 This file describes @value{GDBN}, the @sc{gnu} symbolic debugger.
117 This is the @value{EDITION} Edition, for @value{GDBN}
118 @ifset VERSION_PACKAGE
119 @value{VERSION_PACKAGE}
121 Version @value{GDBVN}.
123 Copyright (C) 1988-2019 Free Software Foundation, Inc.
125 This edition of the GDB manual is dedicated to the memory of Fred
126 Fish. Fred was a long-standing contributor to GDB and to Free
127 software in general. We will miss him.
130 * Summary:: Summary of @value{GDBN}
131 * Sample Session:: A sample @value{GDBN} session
133 * Invocation:: Getting in and out of @value{GDBN}
134 * Commands:: @value{GDBN} commands
135 * Running:: Running programs under @value{GDBN}
136 * Stopping:: Stopping and continuing
137 * Reverse Execution:: Running programs backward
138 * Process Record and Replay:: Recording inferior's execution and replaying it
139 * Stack:: Examining the stack
140 * Source:: Examining source files
141 * Data:: Examining data
142 * Optimized Code:: Debugging optimized code
143 * Macros:: Preprocessor Macros
144 * Tracepoints:: Debugging remote targets non-intrusively
145 * Overlays:: Debugging programs that use overlays
147 * Languages:: Using @value{GDBN} with different languages
149 * Symbols:: Examining the symbol table
150 * Altering:: Altering execution
151 * GDB Files:: @value{GDBN} files
152 * Targets:: Specifying a debugging target
153 * Remote Debugging:: Debugging remote programs
154 * Configurations:: Configuration-specific information
155 * Controlling GDB:: Controlling @value{GDBN}
156 * Extending GDB:: Extending @value{GDBN}
157 * Interpreters:: Command Interpreters
158 * TUI:: @value{GDBN} Text User Interface
159 * Emacs:: Using @value{GDBN} under @sc{gnu} Emacs
160 * GDB/MI:: @value{GDBN}'s Machine Interface.
161 * Annotations:: @value{GDBN}'s annotation interface.
162 * JIT Interface:: Using the JIT debugging interface.
163 * In-Process Agent:: In-Process Agent
165 * GDB Bugs:: Reporting bugs in @value{GDBN}
167 @ifset SYSTEM_READLINE
168 * Command Line Editing: (rluserman). Command Line Editing
169 * Using History Interactively: (history). Using History Interactively
171 @ifclear SYSTEM_READLINE
172 * Command Line Editing:: Command Line Editing
173 * Using History Interactively:: Using History Interactively
175 * In Memoriam:: In Memoriam
176 * Formatting Documentation:: How to format and print @value{GDBN} documentation
177 * Installing GDB:: Installing GDB
178 * Maintenance Commands:: Maintenance Commands
179 * Remote Protocol:: GDB Remote Serial Protocol
180 * Agent Expressions:: The GDB Agent Expression Mechanism
181 * Target Descriptions:: How targets can describe themselves to
183 * Operating System Information:: Getting additional information from
185 * Trace File Format:: GDB trace file format
186 * Index Section Format:: .gdb_index section format
187 * Man Pages:: Manual pages
188 * Copying:: GNU General Public License says
189 how you can copy and share GDB
190 * GNU Free Documentation License:: The license for this documentation
191 * Concept Index:: Index of @value{GDBN} concepts
192 * Command and Variable Index:: Index of @value{GDBN} commands, variables,
193 functions, and Python data types
201 @unnumbered Summary of @value{GDBN}
203 The purpose of a debugger such as @value{GDBN} is to allow you to see what is
204 going on ``inside'' another program while it executes---or what another
205 program was doing at the moment it crashed.
207 @value{GDBN} can do four main kinds of things (plus other things in support of
208 these) to help you catch bugs in the act:
212 Start your program, specifying anything that might affect its behavior.
215 Make your program stop on specified conditions.
218 Examine what has happened, when your program has stopped.
221 Change things in your program, so you can experiment with correcting the
222 effects of one bug and go on to learn about another.
225 You can use @value{GDBN} to debug programs written in C and C@t{++}.
226 For more information, see @ref{Supported Languages,,Supported Languages}.
227 For more information, see @ref{C,,C and C++}.
229 Support for D is partial. For information on D, see
233 Support for Modula-2 is partial. For information on Modula-2, see
234 @ref{Modula-2,,Modula-2}.
236 Support for OpenCL C is partial. For information on OpenCL C, see
237 @ref{OpenCL C,,OpenCL C}.
240 Debugging Pascal programs which use sets, subranges, file variables, or
241 nested functions does not currently work. @value{GDBN} does not support
242 entering expressions, printing values, or similar features using Pascal
246 @value{GDBN} can be used to debug programs written in Fortran, although
247 it may be necessary to refer to some variables with a trailing
250 @value{GDBN} can be used to debug programs written in Objective-C,
251 using either the Apple/NeXT or the GNU Objective-C runtime.
254 * Free Software:: Freely redistributable software
255 * Free Documentation:: Free Software Needs Free Documentation
256 * Contributors:: Contributors to GDB
260 @unnumberedsec Free Software
262 @value{GDBN} is @dfn{free software}, protected by the @sc{gnu}
263 General Public License
264 (GPL). The GPL gives you the freedom to copy or adapt a licensed
265 program---but every person getting a copy also gets with it the
266 freedom to modify that copy (which means that they must get access to
267 the source code), and the freedom to distribute further copies.
268 Typical software companies use copyrights to limit your freedoms; the
269 Free Software Foundation uses the GPL to preserve these freedoms.
271 Fundamentally, the General Public License is a license which says that
272 you have these freedoms and that you cannot take these freedoms away
275 @node Free Documentation
276 @unnumberedsec Free Software Needs Free Documentation
278 The biggest deficiency in the free software community today is not in
279 the software---it is the lack of good free documentation that we can
280 include with the free software. Many of our most important
281 programs do not come with free reference manuals and free introductory
282 texts. Documentation is an essential part of any software package;
283 when an important free software package does not come with a free
284 manual and a free tutorial, that is a major gap. We have many such
287 Consider Perl, for instance. The tutorial manuals that people
288 normally use are non-free. How did this come about? Because the
289 authors of those manuals published them with restrictive terms---no
290 copying, no modification, source files not available---which exclude
291 them from the free software world.
293 That wasn't the first time this sort of thing happened, and it was far
294 from the last. Many times we have heard a GNU user eagerly describe a
295 manual that he is writing, his intended contribution to the community,
296 only to learn that he had ruined everything by signing a publication
297 contract to make it non-free.
299 Free documentation, like free software, is a matter of freedom, not
300 price. The problem with the non-free manual is not that publishers
301 charge a price for printed copies---that in itself is fine. (The Free
302 Software Foundation sells printed copies of manuals, too.) The
303 problem is the restrictions on the use of the manual. Free manuals
304 are available in source code form, and give you permission to copy and
305 modify. Non-free manuals do not allow this.
307 The criteria of freedom for a free manual are roughly the same as for
308 free software. Redistribution (including the normal kinds of
309 commercial redistribution) must be permitted, so that the manual can
310 accompany every copy of the program, both on-line and on paper.
312 Permission for modification of the technical content is crucial too.
313 When people modify the software, adding or changing features, if they
314 are conscientious they will change the manual too---so they can
315 provide accurate and clear documentation for the modified program. A
316 manual that leaves you no choice but to write a new manual to document
317 a changed version of the program is not really available to our
320 Some kinds of limits on the way modification is handled are
321 acceptable. For example, requirements to preserve the original
322 author's copyright notice, the distribution terms, or the list of
323 authors, are ok. It is also no problem to require modified versions
324 to include notice that they were modified. Even entire sections that
325 may not be deleted or changed are acceptable, as long as they deal
326 with nontechnical topics (like this one). These kinds of restrictions
327 are acceptable because they don't obstruct the community's normal use
330 However, it must be possible to modify all the @emph{technical}
331 content of the manual, and then distribute the result in all the usual
332 media, through all the usual channels. Otherwise, the restrictions
333 obstruct the use of the manual, it is not free, and we need another
334 manual to replace it.
336 Please spread the word about this issue. Our community continues to
337 lose manuals to proprietary publishing. If we spread the word that
338 free software needs free reference manuals and free tutorials, perhaps
339 the next person who wants to contribute by writing documentation will
340 realize, before it is too late, that only free manuals contribute to
341 the free software community.
343 If you are writing documentation, please insist on publishing it under
344 the GNU Free Documentation License or another free documentation
345 license. Remember that this decision requires your approval---you
346 don't have to let the publisher decide. Some commercial publishers
347 will use a free license if you insist, but they will not propose the
348 option; it is up to you to raise the issue and say firmly that this is
349 what you want. If the publisher you are dealing with refuses, please
350 try other publishers. If you're not sure whether a proposed license
351 is free, write to @email{licensing@@gnu.org}.
353 You can encourage commercial publishers to sell more free, copylefted
354 manuals and tutorials by buying them, and particularly by buying
355 copies from the publishers that paid for their writing or for major
356 improvements. Meanwhile, try to avoid buying non-free documentation
357 at all. Check the distribution terms of a manual before you buy it,
358 and insist that whoever seeks your business must respect your freedom.
359 Check the history of the book, and try to reward the publishers that
360 have paid or pay the authors to work on it.
362 The Free Software Foundation maintains a list of free documentation
363 published by other publishers, at
364 @url{http://www.fsf.org/doc/other-free-books.html}.
367 @unnumberedsec Contributors to @value{GDBN}
369 Richard Stallman was the original author of @value{GDBN}, and of many
370 other @sc{gnu} programs. Many others have contributed to its
371 development. This section attempts to credit major contributors. One
372 of the virtues of free software is that everyone is free to contribute
373 to it; with regret, we cannot actually acknowledge everyone here. The
374 file @file{ChangeLog} in the @value{GDBN} distribution approximates a
375 blow-by-blow account.
377 Changes much prior to version 2.0 are lost in the mists of time.
380 @emph{Plea:} Additions to this section are particularly welcome. If you
381 or your friends (or enemies, to be evenhanded) have been unfairly
382 omitted from this list, we would like to add your names!
385 So that they may not regard their many labors as thankless, we
386 particularly thank those who shepherded @value{GDBN} through major
388 Andrew Cagney (releases 6.3, 6.2, 6.1, 6.0, 5.3, 5.2, 5.1 and 5.0);
389 Jim Blandy (release 4.18);
390 Jason Molenda (release 4.17);
391 Stan Shebs (release 4.14);
392 Fred Fish (releases 4.16, 4.15, 4.13, 4.12, 4.11, 4.10, and 4.9);
393 Stu Grossman and John Gilmore (releases 4.8, 4.7, 4.6, 4.5, and 4.4);
394 John Gilmore (releases 4.3, 4.2, 4.1, 4.0, and 3.9);
395 Jim Kingdon (releases 3.5, 3.4, and 3.3);
396 and Randy Smith (releases 3.2, 3.1, and 3.0).
398 Richard Stallman, assisted at various times by Peter TerMaat, Chris
399 Hanson, and Richard Mlynarik, handled releases through 2.8.
401 Michael Tiemann is the author of most of the @sc{gnu} C@t{++} support
402 in @value{GDBN}, with significant additional contributions from Per
403 Bothner and Daniel Berlin. James Clark wrote the @sc{gnu} C@t{++}
404 demangler. Early work on C@t{++} was by Peter TerMaat (who also did
405 much general update work leading to release 3.0).
407 @value{GDBN} uses the BFD subroutine library to examine multiple
408 object-file formats; BFD was a joint project of David V.
409 Henkel-Wallace, Rich Pixley, Steve Chamberlain, and John Gilmore.
411 David Johnson wrote the original COFF support; Pace Willison did
412 the original support for encapsulated COFF.
414 Brent Benson of Harris Computer Systems contributed DWARF 2 support.
416 Adam de Boor and Bradley Davis contributed the ISI Optimum V support.
417 Per Bothner, Noboyuki Hikichi, and Alessandro Forin contributed MIPS
419 Jean-Daniel Fekete contributed Sun 386i support.
420 Chris Hanson improved the HP9000 support.
421 Noboyuki Hikichi and Tomoyuki Hasei contributed Sony/News OS 3 support.
422 David Johnson contributed Encore Umax support.
423 Jyrki Kuoppala contributed Altos 3068 support.
424 Jeff Law contributed HP PA and SOM support.
425 Keith Packard contributed NS32K support.
426 Doug Rabson contributed Acorn Risc Machine support.
427 Bob Rusk contributed Harris Nighthawk CX-UX support.
428 Chris Smith contributed Convex support (and Fortran debugging).
429 Jonathan Stone contributed Pyramid support.
430 Michael Tiemann contributed SPARC support.
431 Tim Tucker contributed support for the Gould NP1 and Gould Powernode.
432 Pace Willison contributed Intel 386 support.
433 Jay Vosburgh contributed Symmetry support.
434 Marko Mlinar contributed OpenRISC 1000 support.
436 Andreas Schwab contributed M68K @sc{gnu}/Linux support.
438 Rich Schaefer and Peter Schauer helped with support of SunOS shared
441 Jay Fenlason and Roland McGrath ensured that @value{GDBN} and GAS agree
442 about several machine instruction sets.
444 Patrick Duval, Ted Goldstein, Vikram Koka and Glenn Engel helped develop
445 remote debugging. Intel Corporation, Wind River Systems, AMD, and ARM
446 contributed remote debugging modules for the i960, VxWorks, A29K UDI,
447 and RDI targets, respectively.
449 Brian Fox is the author of the readline libraries providing
450 command-line editing and command history.
452 Andrew Beers of SUNY Buffalo wrote the language-switching code, the
453 Modula-2 support, and contributed the Languages chapter of this manual.
455 Fred Fish wrote most of the support for Unix System Vr4.
456 He also enhanced the command-completion support to cover C@t{++} overloaded
459 Hitachi America (now Renesas America), Ltd. sponsored the support for
460 H8/300, H8/500, and Super-H processors.
462 NEC sponsored the support for the v850, Vr4xxx, and Vr5xxx processors.
464 Mitsubishi (now Renesas) sponsored the support for D10V, D30V, and M32R/D
467 Toshiba sponsored the support for the TX39 Mips processor.
469 Matsushita sponsored the support for the MN10200 and MN10300 processors.
471 Fujitsu sponsored the support for SPARClite and FR30 processors.
473 Kung Hsu, Jeff Law, and Rick Sladkey added support for hardware
476 Michael Snyder added support for tracepoints.
478 Stu Grossman wrote gdbserver.
480 Jim Kingdon, Peter Schauer, Ian Taylor, and Stu Grossman made
481 nearly innumerable bug fixes and cleanups throughout @value{GDBN}.
483 The following people at the Hewlett-Packard Company contributed
484 support for the PA-RISC 2.0 architecture, HP-UX 10.20, 10.30, and 11.0
485 (narrow mode), HP's implementation of kernel threads, HP's aC@t{++}
486 compiler, and the Text User Interface (nee Terminal User Interface):
487 Ben Krepp, Richard Title, John Bishop, Susan Macchia, Kathy Mann,
488 Satish Pai, India Paul, Steve Rehrauer, and Elena Zannoni. Kim Haase
489 provided HP-specific information in this manual.
491 DJ Delorie ported @value{GDBN} to MS-DOS, for the DJGPP project.
492 Robert Hoehne made significant contributions to the DJGPP port.
494 Cygnus Solutions has sponsored @value{GDBN} maintenance and much of its
495 development since 1991. Cygnus engineers who have worked on @value{GDBN}
496 fulltime include Mark Alexander, Jim Blandy, Per Bothner, Kevin
497 Buettner, Edith Epstein, Chris Faylor, Fred Fish, Martin Hunt, Jim
498 Ingham, John Gilmore, Stu Grossman, Kung Hsu, Jim Kingdon, John Metzler,
499 Fernando Nasser, Geoffrey Noer, Dawn Perchik, Rich Pixley, Zdenek
500 Radouch, Keith Seitz, Stan Shebs, David Taylor, and Elena Zannoni. In
501 addition, Dave Brolley, Ian Carmichael, Steve Chamberlain, Nick Clifton,
502 JT Conklin, Stan Cox, DJ Delorie, Ulrich Drepper, Frank Eigler, Doug
503 Evans, Sean Fagan, David Henkel-Wallace, Richard Henderson, Jeff
504 Holcomb, Jeff Law, Jim Lemke, Tom Lord, Bob Manson, Michael Meissner,
505 Jason Merrill, Catherine Moore, Drew Moseley, Ken Raeburn, Gavin
506 Romig-Koch, Rob Savoye, Jamie Smith, Mike Stump, Ian Taylor, Angela
507 Thomas, Michael Tiemann, Tom Tromey, Ron Unrau, Jim Wilson, and David
508 Zuhn have made contributions both large and small.
510 Andrew Cagney, Fernando Nasser, and Elena Zannoni, while working for
511 Cygnus Solutions, implemented the original @sc{gdb/mi} interface.
513 Jim Blandy added support for preprocessor macros, while working for Red
516 Andrew Cagney designed @value{GDBN}'s architecture vector. Many
517 people including Andrew Cagney, Stephane Carrez, Randolph Chung, Nick
518 Duffek, Richard Henderson, Mark Kettenis, Grace Sainsbury, Kei
519 Sakamoto, Yoshinori Sato, Michael Snyder, Andreas Schwab, Jason
520 Thorpe, Corinna Vinschen, Ulrich Weigand, and Elena Zannoni, helped
521 with the migration of old architectures to this new framework.
523 Andrew Cagney completely re-designed and re-implemented @value{GDBN}'s
524 unwinder framework, this consisting of a fresh new design featuring
525 frame IDs, independent frame sniffers, and the sentinel frame. Mark
526 Kettenis implemented the @sc{dwarf 2} unwinder, Jeff Johnston the
527 libunwind unwinder, and Andrew Cagney the dummy, sentinel, tramp, and
528 trad unwinders. The architecture-specific changes, each involving a
529 complete rewrite of the architecture's frame code, were carried out by
530 Jim Blandy, Joel Brobecker, Kevin Buettner, Andrew Cagney, Stephane
531 Carrez, Randolph Chung, Orjan Friberg, Richard Henderson, Daniel
532 Jacobowitz, Jeff Johnston, Mark Kettenis, Theodore A. Roth, Kei
533 Sakamoto, Yoshinori Sato, Michael Snyder, Corinna Vinschen, and Ulrich
536 Christian Zankel, Ross Morley, Bob Wilson, and Maxim Grigoriev from
537 Tensilica, Inc.@: contributed support for Xtensa processors. Others
538 who have worked on the Xtensa port of @value{GDBN} in the past include
539 Steve Tjiang, John Newlin, and Scott Foehner.
541 Michael Eager and staff of Xilinx, Inc., contributed support for the
542 Xilinx MicroBlaze architecture.
544 Initial support for the FreeBSD/mips target and native configuration
545 was developed by SRI International and the University of Cambridge
546 Computer Laboratory under DARPA/AFRL contract FA8750-10-C-0237
547 ("CTSRD"), as part of the DARPA CRASH research programme.
549 Initial support for the FreeBSD/riscv target and native configuration
550 was developed by SRI International and the University of Cambridge
551 Computer Laboratory (Department of Computer Science and Technology)
552 under DARPA contract HR0011-18-C-0016 ("ECATS"), as part of the DARPA
553 SSITH research programme.
555 The original port to the OpenRISC 1000 is believed to be due to
556 Alessandro Forin and Per Bothner. More recent ports have been the work
557 of Jeremy Bennett, Franck Jullien, Stefan Wallentowitz and
561 @chapter A Sample @value{GDBN} Session
563 You can use this manual at your leisure to read all about @value{GDBN}.
564 However, a handful of commands are enough to get started using the
565 debugger. This chapter illustrates those commands.
568 In this sample session, we emphasize user input like this: @b{input},
569 to make it easier to pick out from the surrounding output.
572 @c FIXME: this example may not be appropriate for some configs, where
573 @c FIXME...primary interest is in remote use.
575 One of the preliminary versions of @sc{gnu} @code{m4} (a generic macro
576 processor) exhibits the following bug: sometimes, when we change its
577 quote strings from the default, the commands used to capture one macro
578 definition within another stop working. In the following short @code{m4}
579 session, we define a macro @code{foo} which expands to @code{0000}; we
580 then use the @code{m4} built-in @code{defn} to define @code{bar} as the
581 same thing. However, when we change the open quote string to
582 @code{<QUOTE>} and the close quote string to @code{<UNQUOTE>}, the same
583 procedure fails to define a new synonym @code{baz}:
592 @b{define(bar,defn(`foo'))}
596 @b{changequote(<QUOTE>,<UNQUOTE>)}
598 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
601 m4: End of input: 0: fatal error: EOF in string
605 Let us use @value{GDBN} to try to see what is going on.
608 $ @b{@value{GDBP} m4}
609 @c FIXME: this falsifies the exact text played out, to permit smallbook
610 @c FIXME... format to come out better.
611 @value{GDBN} is free software and you are welcome to distribute copies
612 of it under certain conditions; type "show copying" to see
614 There is absolutely no warranty for @value{GDBN}; type "show warranty"
617 @value{GDBN} @value{GDBVN}, Copyright 1999 Free Software Foundation, Inc...
622 @value{GDBN} reads only enough symbol data to know where to find the
623 rest when needed; as a result, the first prompt comes up very quickly.
624 We now tell @value{GDBN} to use a narrower display width than usual, so
625 that examples fit in this manual.
628 (@value{GDBP}) @b{set width 70}
632 We need to see how the @code{m4} built-in @code{changequote} works.
633 Having looked at the source, we know the relevant subroutine is
634 @code{m4_changequote}, so we set a breakpoint there with the @value{GDBN}
635 @code{break} command.
638 (@value{GDBP}) @b{break m4_changequote}
639 Breakpoint 1 at 0x62f4: file builtin.c, line 879.
643 Using the @code{run} command, we start @code{m4} running under @value{GDBN}
644 control; as long as control does not reach the @code{m4_changequote}
645 subroutine, the program runs as usual:
648 (@value{GDBP}) @b{run}
649 Starting program: /work/Editorial/gdb/gnu/m4/m4
657 To trigger the breakpoint, we call @code{changequote}. @value{GDBN}
658 suspends execution of @code{m4}, displaying information about the
659 context where it stops.
662 @b{changequote(<QUOTE>,<UNQUOTE>)}
664 Breakpoint 1, m4_changequote (argc=3, argv=0x33c70)
666 879 if (bad_argc(TOKEN_DATA_TEXT(argv[0]),argc,1,3))
670 Now we use the command @code{n} (@code{next}) to advance execution to
671 the next line of the current function.
675 882 set_quotes((argc >= 2) ? TOKEN_DATA_TEXT(argv[1])\
680 @code{set_quotes} looks like a promising subroutine. We can go into it
681 by using the command @code{s} (@code{step}) instead of @code{next}.
682 @code{step} goes to the next line to be executed in @emph{any}
683 subroutine, so it steps into @code{set_quotes}.
687 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
689 530 if (lquote != def_lquote)
693 The display that shows the subroutine where @code{m4} is now
694 suspended (and its arguments) is called a stack frame display. It
695 shows a summary of the stack. We can use the @code{backtrace}
696 command (which can also be spelled @code{bt}), to see where we are
697 in the stack as a whole: the @code{backtrace} command displays a
698 stack frame for each active subroutine.
701 (@value{GDBP}) @b{bt}
702 #0 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
704 #1 0x6344 in m4_changequote (argc=3, argv=0x33c70)
706 #2 0x8174 in expand_macro (sym=0x33320) at macro.c:242
707 #3 0x7a88 in expand_token (obs=0x0, t=209696, td=0xf7fffa30)
709 #4 0x79dc in expand_input () at macro.c:40
710 #5 0x2930 in main (argc=0, argv=0xf7fffb20) at m4.c:195
714 We step through a few more lines to see what happens. The first two
715 times, we can use @samp{s}; the next two times we use @code{n} to avoid
716 falling into the @code{xstrdup} subroutine.
720 0x3b5c 532 if (rquote != def_rquote)
722 0x3b80 535 lquote = (lq == nil || *lq == '\0') ? \
723 def_lquote : xstrdup(lq);
725 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
728 538 len_lquote = strlen(rquote);
732 The last line displayed looks a little odd; we can examine the variables
733 @code{lquote} and @code{rquote} to see if they are in fact the new left
734 and right quotes we specified. We use the command @code{p}
735 (@code{print}) to see their values.
738 (@value{GDBP}) @b{p lquote}
739 $1 = 0x35d40 "<QUOTE>"
740 (@value{GDBP}) @b{p rquote}
741 $2 = 0x35d50 "<UNQUOTE>"
745 @code{lquote} and @code{rquote} are indeed the new left and right quotes.
746 To look at some context, we can display ten lines of source
747 surrounding the current line with the @code{l} (@code{list}) command.
753 535 lquote = (lq == nil || *lq == '\0') ? def_lquote\
755 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
758 538 len_lquote = strlen(rquote);
759 539 len_rquote = strlen(lquote);
766 Let us step past the two lines that set @code{len_lquote} and
767 @code{len_rquote}, and then examine the values of those variables.
771 539 len_rquote = strlen(lquote);
774 (@value{GDBP}) @b{p len_lquote}
776 (@value{GDBP}) @b{p len_rquote}
781 That certainly looks wrong, assuming @code{len_lquote} and
782 @code{len_rquote} are meant to be the lengths of @code{lquote} and
783 @code{rquote} respectively. We can set them to better values using
784 the @code{p} command, since it can print the value of
785 any expression---and that expression can include subroutine calls and
789 (@value{GDBP}) @b{p len_lquote=strlen(lquote)}
791 (@value{GDBP}) @b{p len_rquote=strlen(rquote)}
796 Is that enough to fix the problem of using the new quotes with the
797 @code{m4} built-in @code{defn}? We can allow @code{m4} to continue
798 executing with the @code{c} (@code{continue}) command, and then try the
799 example that caused trouble initially:
805 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
812 Success! The new quotes now work just as well as the default ones. The
813 problem seems to have been just the two typos defining the wrong
814 lengths. We allow @code{m4} exit by giving it an EOF as input:
818 Program exited normally.
822 The message @samp{Program exited normally.} is from @value{GDBN}; it
823 indicates @code{m4} has finished executing. We can end our @value{GDBN}
824 session with the @value{GDBN} @code{quit} command.
827 (@value{GDBP}) @b{quit}
831 @chapter Getting In and Out of @value{GDBN}
833 This chapter discusses how to start @value{GDBN}, and how to get out of it.
837 type @samp{@value{GDBP}} to start @value{GDBN}.
839 type @kbd{quit} or @kbd{Ctrl-d} to exit.
843 * Invoking GDB:: How to start @value{GDBN}
844 * Quitting GDB:: How to quit @value{GDBN}
845 * Shell Commands:: How to use shell commands inside @value{GDBN}
846 * Logging Output:: How to log @value{GDBN}'s output to a file
850 @section Invoking @value{GDBN}
852 Invoke @value{GDBN} by running the program @code{@value{GDBP}}. Once started,
853 @value{GDBN} reads commands from the terminal until you tell it to exit.
855 You can also run @code{@value{GDBP}} with a variety of arguments and options,
856 to specify more of your debugging environment at the outset.
858 The command-line options described here are designed
859 to cover a variety of situations; in some environments, some of these
860 options may effectively be unavailable.
862 The most usual way to start @value{GDBN} is with one argument,
863 specifying an executable program:
866 @value{GDBP} @var{program}
870 You can also start with both an executable program and a core file
874 @value{GDBP} @var{program} @var{core}
877 You can, instead, specify a process ID as a second argument or use option
878 @code{-p}, if you want to debug a running process:
881 @value{GDBP} @var{program} 1234
886 would attach @value{GDBN} to process @code{1234}. With option @option{-p} you
887 can omit the @var{program} filename.
889 Taking advantage of the second command-line argument requires a fairly
890 complete operating system; when you use @value{GDBN} as a remote
891 debugger attached to a bare board, there may not be any notion of
892 ``process'', and there is often no way to get a core dump. @value{GDBN}
893 will warn you if it is unable to attach or to read core dumps.
895 You can optionally have @code{@value{GDBP}} pass any arguments after the
896 executable file to the inferior using @code{--args}. This option stops
899 @value{GDBP} --args gcc -O2 -c foo.c
901 This will cause @code{@value{GDBP}} to debug @code{gcc}, and to set
902 @code{gcc}'s command-line arguments (@pxref{Arguments}) to @samp{-O2 -c foo.c}.
904 You can run @code{@value{GDBP}} without printing the front material, which describes
905 @value{GDBN}'s non-warranty, by specifying @code{--silent}
906 (or @code{-q}/@code{--quiet}):
909 @value{GDBP} --silent
913 You can further control how @value{GDBN} starts up by using command-line
914 options. @value{GDBN} itself can remind you of the options available.
924 to display all available options and briefly describe their use
925 (@samp{@value{GDBP} -h} is a shorter equivalent).
927 All options and command line arguments you give are processed
928 in sequential order. The order makes a difference when the
929 @samp{-x} option is used.
933 * File Options:: Choosing files
934 * Mode Options:: Choosing modes
935 * Startup:: What @value{GDBN} does during startup
939 @subsection Choosing Files
941 When @value{GDBN} starts, it reads any arguments other than options as
942 specifying an executable file and core file (or process ID). This is
943 the same as if the arguments were specified by the @samp{-se} and
944 @samp{-c} (or @samp{-p}) options respectively. (@value{GDBN} reads the
945 first argument that does not have an associated option flag as
946 equivalent to the @samp{-se} option followed by that argument; and the
947 second argument that does not have an associated option flag, if any, as
948 equivalent to the @samp{-c}/@samp{-p} option followed by that argument.)
949 If the second argument begins with a decimal digit, @value{GDBN} will
950 first attempt to attach to it as a process, and if that fails, attempt
951 to open it as a corefile. If you have a corefile whose name begins with
952 a digit, you can prevent @value{GDBN} from treating it as a pid by
953 prefixing it with @file{./}, e.g.@: @file{./12345}.
955 If @value{GDBN} has not been configured to included core file support,
956 such as for most embedded targets, then it will complain about a second
957 argument and ignore it.
959 Many options have both long and short forms; both are shown in the
960 following list. @value{GDBN} also recognizes the long forms if you truncate
961 them, so long as enough of the option is present to be unambiguous.
962 (If you prefer, you can flag option arguments with @samp{--} rather
963 than @samp{-}, though we illustrate the more usual convention.)
965 @c NOTE: the @cindex entries here use double dashes ON PURPOSE. This
966 @c way, both those who look for -foo and --foo in the index, will find
970 @item -symbols @var{file}
972 @cindex @code{--symbols}
974 Read symbol table from file @var{file}.
976 @item -exec @var{file}
978 @cindex @code{--exec}
980 Use file @var{file} as the executable file to execute when appropriate,
981 and for examining pure data in conjunction with a core dump.
985 Read symbol table from file @var{file} and use it as the executable
988 @item -core @var{file}
990 @cindex @code{--core}
992 Use file @var{file} as a core dump to examine.
994 @item -pid @var{number}
995 @itemx -p @var{number}
998 Connect to process ID @var{number}, as with the @code{attach} command.
1000 @item -command @var{file}
1001 @itemx -x @var{file}
1002 @cindex @code{--command}
1004 Execute commands from file @var{file}. The contents of this file is
1005 evaluated exactly as the @code{source} command would.
1006 @xref{Command Files,, Command files}.
1008 @item -eval-command @var{command}
1009 @itemx -ex @var{command}
1010 @cindex @code{--eval-command}
1012 Execute a single @value{GDBN} command.
1014 This option may be used multiple times to call multiple commands. It may
1015 also be interleaved with @samp{-command} as required.
1018 @value{GDBP} -ex 'target sim' -ex 'load' \
1019 -x setbreakpoints -ex 'run' a.out
1022 @item -init-command @var{file}
1023 @itemx -ix @var{file}
1024 @cindex @code{--init-command}
1026 Execute commands from file @var{file} before loading the inferior (but
1027 after loading gdbinit files).
1030 @item -init-eval-command @var{command}
1031 @itemx -iex @var{command}
1032 @cindex @code{--init-eval-command}
1034 Execute a single @value{GDBN} command before loading the inferior (but
1035 after loading gdbinit files).
1038 @item -directory @var{directory}
1039 @itemx -d @var{directory}
1040 @cindex @code{--directory}
1042 Add @var{directory} to the path to search for source and script files.
1046 @cindex @code{--readnow}
1048 Read each symbol file's entire symbol table immediately, rather than
1049 the default, which is to read it incrementally as it is needed.
1050 This makes startup slower, but makes future operations faster.
1053 @anchor{--readnever}
1054 @cindex @code{--readnever}, command-line option
1055 Do not read each symbol file's symbolic debug information. This makes
1056 startup faster but at the expense of not being able to perform
1057 symbolic debugging. DWARF unwind information is also not read,
1058 meaning backtraces may become incomplete or inaccurate. One use of
1059 this is when a user simply wants to do the following sequence: attach,
1060 dump core, detach. Loading the debugging information in this case is
1061 an unnecessary cause of delay.
1065 @subsection Choosing Modes
1067 You can run @value{GDBN} in various alternative modes---for example, in
1068 batch mode or quiet mode.
1076 Do not execute commands found in any initialization file.
1077 There are three init files, loaded in the following order:
1080 @item @file{system.gdbinit}
1081 This is the system-wide init file.
1082 Its location is specified with the @code{--with-system-gdbinit}
1083 configure option (@pxref{System-wide configuration}).
1084 It is loaded first when @value{GDBN} starts, before command line options
1085 have been processed.
1086 @item @file{system.gdbinit.d}
1087 This is the system-wide init directory.
1088 Its location is specified with the @code{--with-system-gdbinit-dir}
1089 configure option (@pxref{System-wide configuration}).
1090 Files in this directory are loaded in alphabetical order immediately after
1091 system.gdbinit (if enabled) when @value{GDBN} starts, before command line
1092 options have been processed. Files need to have a recognized scripting
1093 language extension (@file{.py}/@file{.scm}) or be named with a @file{.gdb}
1094 extension to be interpreted as regular @value{GDBN} commands. @value{GDBN}
1095 will not recurse into any subdirectories of this directory.
1096 @item @file{~/.gdbinit}
1097 This is the init file in your home directory.
1098 It is loaded next, after @file{system.gdbinit}, and before
1099 command options have been processed.
1100 @item @file{./.gdbinit}
1101 This is the init file in the current directory.
1102 It is loaded last, after command line options other than @code{-x} and
1103 @code{-ex} have been processed. Command line options @code{-x} and
1104 @code{-ex} are processed last, after @file{./.gdbinit} has been loaded.
1107 For further documentation on startup processing, @xref{Startup}.
1108 For documentation on how to write command files,
1109 @xref{Command Files,,Command Files}.
1114 Do not execute commands found in @file{~/.gdbinit}, the init file
1115 in your home directory.
1121 @cindex @code{--quiet}
1122 @cindex @code{--silent}
1124 ``Quiet''. Do not print the introductory and copyright messages. These
1125 messages are also suppressed in batch mode.
1128 @cindex @code{--batch}
1129 Run in batch mode. Exit with status @code{0} after processing all the
1130 command files specified with @samp{-x} (and all commands from
1131 initialization files, if not inhibited with @samp{-n}). Exit with
1132 nonzero status if an error occurs in executing the @value{GDBN} commands
1133 in the command files. Batch mode also disables pagination, sets unlimited
1134 terminal width and height @pxref{Screen Size}, and acts as if @kbd{set confirm
1135 off} were in effect (@pxref{Messages/Warnings}).
1137 Batch mode may be useful for running @value{GDBN} as a filter, for
1138 example to download and run a program on another computer; in order to
1139 make this more useful, the message
1142 Program exited normally.
1146 (which is ordinarily issued whenever a program running under
1147 @value{GDBN} control terminates) is not issued when running in batch
1151 @cindex @code{--batch-silent}
1152 Run in batch mode exactly like @samp{-batch}, but totally silently. All
1153 @value{GDBN} output to @code{stdout} is prevented (@code{stderr} is
1154 unaffected). This is much quieter than @samp{-silent} and would be useless
1155 for an interactive session.
1157 This is particularly useful when using targets that give @samp{Loading section}
1158 messages, for example.
1160 Note that targets that give their output via @value{GDBN}, as opposed to
1161 writing directly to @code{stdout}, will also be made silent.
1163 @item -return-child-result
1164 @cindex @code{--return-child-result}
1165 The return code from @value{GDBN} will be the return code from the child
1166 process (the process being debugged), with the following exceptions:
1170 @value{GDBN} exits abnormally. E.g., due to an incorrect argument or an
1171 internal error. In this case the exit code is the same as it would have been
1172 without @samp{-return-child-result}.
1174 The user quits with an explicit value. E.g., @samp{quit 1}.
1176 The child process never runs, or is not allowed to terminate, in which case
1177 the exit code will be -1.
1180 This option is useful in conjunction with @samp{-batch} or @samp{-batch-silent},
1181 when @value{GDBN} is being used as a remote program loader or simulator
1186 @cindex @code{--nowindows}
1188 ``No windows''. If @value{GDBN} comes with a graphical user interface
1189 (GUI) built in, then this option tells @value{GDBN} to only use the command-line
1190 interface. If no GUI is available, this option has no effect.
1194 @cindex @code{--windows}
1196 If @value{GDBN} includes a GUI, then this option requires it to be
1199 @item -cd @var{directory}
1201 Run @value{GDBN} using @var{directory} as its working directory,
1202 instead of the current directory.
1204 @item -data-directory @var{directory}
1205 @itemx -D @var{directory}
1206 @cindex @code{--data-directory}
1208 Run @value{GDBN} using @var{directory} as its data directory.
1209 The data directory is where @value{GDBN} searches for its
1210 auxiliary files. @xref{Data Files}.
1214 @cindex @code{--fullname}
1216 @sc{gnu} Emacs sets this option when it runs @value{GDBN} as a
1217 subprocess. It tells @value{GDBN} to output the full file name and line
1218 number in a standard, recognizable fashion each time a stack frame is
1219 displayed (which includes each time your program stops). This
1220 recognizable format looks like two @samp{\032} characters, followed by
1221 the file name, line number and character position separated by colons,
1222 and a newline. The Emacs-to-@value{GDBN} interface program uses the two
1223 @samp{\032} characters as a signal to display the source code for the
1226 @item -annotate @var{level}
1227 @cindex @code{--annotate}
1228 This option sets the @dfn{annotation level} inside @value{GDBN}. Its
1229 effect is identical to using @samp{set annotate @var{level}}
1230 (@pxref{Annotations}). The annotation @var{level} controls how much
1231 information @value{GDBN} prints together with its prompt, values of
1232 expressions, source lines, and other types of output. Level 0 is the
1233 normal, level 1 is for use when @value{GDBN} is run as a subprocess of
1234 @sc{gnu} Emacs, level 3 is the maximum annotation suitable for programs
1235 that control @value{GDBN}, and level 2 has been deprecated.
1237 The annotation mechanism has largely been superseded by @sc{gdb/mi}
1241 @cindex @code{--args}
1242 Change interpretation of command line so that arguments following the
1243 executable file are passed as command line arguments to the inferior.
1244 This option stops option processing.
1246 @item -baud @var{bps}
1248 @cindex @code{--baud}
1250 Set the line speed (baud rate or bits per second) of any serial
1251 interface used by @value{GDBN} for remote debugging.
1253 @item -l @var{timeout}
1255 Set the timeout (in seconds) of any communication used by @value{GDBN}
1256 for remote debugging.
1258 @item -tty @var{device}
1259 @itemx -t @var{device}
1260 @cindex @code{--tty}
1262 Run using @var{device} for your program's standard input and output.
1263 @c FIXME: kingdon thinks there is more to -tty. Investigate.
1265 @c resolve the situation of these eventually
1267 @cindex @code{--tui}
1268 Activate the @dfn{Text User Interface} when starting. The Text User
1269 Interface manages several text windows on the terminal, showing
1270 source, assembly, registers and @value{GDBN} command outputs
1271 (@pxref{TUI, ,@value{GDBN} Text User Interface}). Do not use this
1272 option if you run @value{GDBN} from Emacs (@pxref{Emacs, ,
1273 Using @value{GDBN} under @sc{gnu} Emacs}).
1275 @item -interpreter @var{interp}
1276 @cindex @code{--interpreter}
1277 Use the interpreter @var{interp} for interface with the controlling
1278 program or device. This option is meant to be set by programs which
1279 communicate with @value{GDBN} using it as a back end.
1280 @xref{Interpreters, , Command Interpreters}.
1282 @samp{--interpreter=mi} (or @samp{--interpreter=mi3}) causes
1283 @value{GDBN} to use the @dfn{@sc{gdb/mi} interface} version 3 (@pxref{GDB/MI, ,
1284 The @sc{gdb/mi} Interface}) included since @value{GDBN} version 9.1. @sc{gdb/mi}
1285 version 2 (@code{mi2}), included in @value{GDBN} 6.0 and version 1 (@code{mi1}),
1286 included in @value{GDBN} 5.3, are also available. Earlier @sc{gdb/mi}
1287 interfaces are no longer supported.
1290 @cindex @code{--write}
1291 Open the executable and core files for both reading and writing. This
1292 is equivalent to the @samp{set write on} command inside @value{GDBN}
1296 @cindex @code{--statistics}
1297 This option causes @value{GDBN} to print statistics about time and
1298 memory usage after it completes each command and returns to the prompt.
1301 @cindex @code{--version}
1302 This option causes @value{GDBN} to print its version number and
1303 no-warranty blurb, and exit.
1305 @item -configuration
1306 @cindex @code{--configuration}
1307 This option causes @value{GDBN} to print details about its build-time
1308 configuration parameters, and then exit. These details can be
1309 important when reporting @value{GDBN} bugs (@pxref{GDB Bugs}).
1314 @subsection What @value{GDBN} Does During Startup
1315 @cindex @value{GDBN} startup
1317 Here's the description of what @value{GDBN} does during session startup:
1321 Sets up the command interpreter as specified by the command line
1322 (@pxref{Mode Options, interpreter}).
1326 Reads the system-wide @dfn{init file} (if @option{--with-system-gdbinit} was
1327 used when building @value{GDBN}; @pxref{System-wide configuration,
1328 ,System-wide configuration and settings}) and the files in the system-wide
1329 gdbinit directory (if @option{--with-system-gdbinit-dir} was used) and executes
1330 all the commands in those files. The files need to be named with a @file{.gdb}
1331 extension to be interpreted as @value{GDBN} commands, or they can be written
1332 in a supported scripting language with an appropriate file extension.
1334 @anchor{Home Directory Init File}
1336 Reads the init file (if any) in your home directory@footnote{On
1337 DOS/Windows systems, the home directory is the one pointed to by the
1338 @code{HOME} environment variable.} and executes all the commands in
1341 @anchor{Option -init-eval-command}
1343 Executes commands and command files specified by the @samp{-iex} and
1344 @samp{-ix} options in their specified order. Usually you should use the
1345 @samp{-ex} and @samp{-x} options instead, but this way you can apply
1346 settings before @value{GDBN} init files get executed and before inferior
1350 Processes command line options and operands.
1352 @anchor{Init File in the Current Directory during Startup}
1354 Reads and executes the commands from init file (if any) in the current
1355 working directory as long as @samp{set auto-load local-gdbinit} is set to
1356 @samp{on} (@pxref{Init File in the Current Directory}).
1357 This is only done if the current directory is
1358 different from your home directory. Thus, you can have more than one
1359 init file, one generic in your home directory, and another, specific
1360 to the program you are debugging, in the directory where you invoke
1364 If the command line specified a program to debug, or a process to
1365 attach to, or a core file, @value{GDBN} loads any auto-loaded
1366 scripts provided for the program or for its loaded shared libraries.
1367 @xref{Auto-loading}.
1369 If you wish to disable the auto-loading during startup,
1370 you must do something like the following:
1373 $ gdb -iex "set auto-load python-scripts off" myprogram
1376 Option @samp{-ex} does not work because the auto-loading is then turned
1380 Executes commands and command files specified by the @samp{-ex} and
1381 @samp{-x} options in their specified order. @xref{Command Files}, for
1382 more details about @value{GDBN} command files.
1385 Reads the command history recorded in the @dfn{history file}.
1386 @xref{Command History}, for more details about the command history and the
1387 files where @value{GDBN} records it.
1390 Init files use the same syntax as @dfn{command files} (@pxref{Command
1391 Files}) and are processed by @value{GDBN} in the same way. The init
1392 file in your home directory can set options (such as @samp{set
1393 complaints}) that affect subsequent processing of command line options
1394 and operands. Init files are not executed if you use the @samp{-nx}
1395 option (@pxref{Mode Options, ,Choosing Modes}).
1397 To display the list of init files loaded by gdb at startup, you
1398 can use @kbd{gdb --help}.
1400 @cindex init file name
1401 @cindex @file{.gdbinit}
1402 @cindex @file{gdb.ini}
1403 The @value{GDBN} init files are normally called @file{.gdbinit}.
1404 The DJGPP port of @value{GDBN} uses the name @file{gdb.ini}, due to
1405 the limitations of file names imposed by DOS filesystems. The Windows
1406 port of @value{GDBN} uses the standard name, but if it finds a
1407 @file{gdb.ini} file in your home directory, it warns you about that
1408 and suggests to rename the file to the standard name.
1412 @section Quitting @value{GDBN}
1413 @cindex exiting @value{GDBN}
1414 @cindex leaving @value{GDBN}
1417 @kindex quit @r{[}@var{expression}@r{]}
1418 @kindex q @r{(@code{quit})}
1419 @item quit @r{[}@var{expression}@r{]}
1421 To exit @value{GDBN}, use the @code{quit} command (abbreviated
1422 @code{q}), or type an end-of-file character (usually @kbd{Ctrl-d}). If you
1423 do not supply @var{expression}, @value{GDBN} will terminate normally;
1424 otherwise it will terminate using the result of @var{expression} as the
1429 An interrupt (often @kbd{Ctrl-c}) does not exit from @value{GDBN}, but rather
1430 terminates the action of any @value{GDBN} command that is in progress and
1431 returns to @value{GDBN} command level. It is safe to type the interrupt
1432 character at any time because @value{GDBN} does not allow it to take effect
1433 until a time when it is safe.
1435 If you have been using @value{GDBN} to control an attached process or
1436 device, you can release it with the @code{detach} command
1437 (@pxref{Attach, ,Debugging an Already-running Process}).
1439 @node Shell Commands
1440 @section Shell Commands
1442 If you need to execute occasional shell commands during your
1443 debugging session, there is no need to leave or suspend @value{GDBN}; you can
1444 just use the @code{shell} command.
1449 @cindex shell escape
1450 @item shell @var{command-string}
1451 @itemx !@var{command-string}
1452 Invoke a standard shell to execute @var{command-string}.
1453 Note that no space is needed between @code{!} and @var{command-string}.
1454 If it exists, the environment variable @code{SHELL} determines which
1455 shell to run. Otherwise @value{GDBN} uses the default shell
1456 (@file{/bin/sh} on Unix systems, @file{COMMAND.COM} on MS-DOS, etc.).
1459 The utility @code{make} is often needed in development environments.
1460 You do not have to use the @code{shell} command for this purpose in
1465 @cindex calling make
1466 @item make @var{make-args}
1467 Execute the @code{make} program with the specified
1468 arguments. This is equivalent to @samp{shell make @var{make-args}}.
1474 @cindex send the output of a gdb command to a shell command
1476 @item pipe [@var{command}] | @var{shell_command}
1477 @itemx | [@var{command}] | @var{shell_command}
1478 @itemx pipe -d @var{delim} @var{command} @var{delim} @var{shell_command}
1479 @itemx | -d @var{delim} @var{command} @var{delim} @var{shell_command}
1480 Executes @var{command} and sends its output to @var{shell_command}.
1481 Note that no space is needed around @code{|}.
1482 If no @var{command} is provided, the last command executed is repeated.
1484 In case the @var{command} contains a @code{|}, the option @code{-d @var{delim}}
1485 can be used to specify an alternate delimiter string @var{delim} that separates
1486 the @var{command} from the @var{shell_command}.
1519 (gdb) | -d ! echo this contains a | char\n ! sed -e 's/|/PIPE/'
1520 this contains a PIPE char
1521 (gdb) | -d xxx echo this contains a | char!\n xxx sed -e 's/|/PIPE/'
1522 this contains a PIPE char!
1528 The convenience variables @code{$_shell_exitcode} and @code{$_shell_exitsignal}
1529 can be used to examine the exit status of the last shell command launched
1530 by @code{shell}, @code{make}, @code{pipe} and @code{|}.
1531 @xref{Convenience Vars,, Convenience Variables}.
1533 @node Logging Output
1534 @section Logging Output
1535 @cindex logging @value{GDBN} output
1536 @cindex save @value{GDBN} output to a file
1538 You may want to save the output of @value{GDBN} commands to a file.
1539 There are several commands to control @value{GDBN}'s logging.
1543 @item set logging on
1545 @item set logging off
1547 @cindex logging file name
1548 @item set logging file @var{file}
1549 Change the name of the current logfile. The default logfile is @file{gdb.txt}.
1550 @item set logging overwrite [on|off]
1551 By default, @value{GDBN} will append to the logfile. Set @code{overwrite} if
1552 you want @code{set logging on} to overwrite the logfile instead.
1553 @item set logging redirect [on|off]
1554 By default, @value{GDBN} output will go to both the terminal and the logfile.
1555 Set @code{redirect} if you want output to go only to the log file.
1556 @item set logging debugredirect [on|off]
1557 By default, @value{GDBN} debug output will go to both the terminal and the logfile.
1558 Set @code{debugredirect} if you want debug output to go only to the log file.
1559 @kindex show logging
1561 Show the current values of the logging settings.
1564 You can also redirect the output of a @value{GDBN} command to a
1565 shell command. @xref{pipe}.
1567 @chapter @value{GDBN} Commands
1569 You can abbreviate a @value{GDBN} command to the first few letters of the command
1570 name, if that abbreviation is unambiguous; and you can repeat certain
1571 @value{GDBN} commands by typing just @key{RET}. You can also use the @key{TAB}
1572 key to get @value{GDBN} to fill out the rest of a word in a command (or to
1573 show you the alternatives available, if there is more than one possibility).
1576 * Command Syntax:: How to give commands to @value{GDBN}
1577 * Command Settings:: How to change default behavior of commands
1578 * Completion:: Command completion
1579 * Command Options:: Command options
1580 * Help:: How to ask @value{GDBN} for help
1583 @node Command Syntax
1584 @section Command Syntax
1586 A @value{GDBN} command is a single line of input. There is no limit on
1587 how long it can be. It starts with a command name, which is followed by
1588 arguments whose meaning depends on the command name. For example, the
1589 command @code{step} accepts an argument which is the number of times to
1590 step, as in @samp{step 5}. You can also use the @code{step} command
1591 with no arguments. Some commands do not allow any arguments.
1593 @cindex abbreviation
1594 @value{GDBN} command names may always be truncated if that abbreviation is
1595 unambiguous. Other possible command abbreviations are listed in the
1596 documentation for individual commands. In some cases, even ambiguous
1597 abbreviations are allowed; for example, @code{s} is specially defined as
1598 equivalent to @code{step} even though there are other commands whose
1599 names start with @code{s}. You can test abbreviations by using them as
1600 arguments to the @code{help} command.
1602 @cindex repeating commands
1603 @kindex RET @r{(repeat last command)}
1604 A blank line as input to @value{GDBN} (typing just @key{RET}) means to
1605 repeat the previous command. Certain commands (for example, @code{run})
1606 will not repeat this way; these are commands whose unintentional
1607 repetition might cause trouble and which you are unlikely to want to
1608 repeat. User-defined commands can disable this feature; see
1609 @ref{Define, dont-repeat}.
1611 The @code{list} and @code{x} commands, when you repeat them with
1612 @key{RET}, construct new arguments rather than repeating
1613 exactly as typed. This permits easy scanning of source or memory.
1615 @value{GDBN} can also use @key{RET} in another way: to partition lengthy
1616 output, in a way similar to the common utility @code{more}
1617 (@pxref{Screen Size,,Screen Size}). Since it is easy to press one
1618 @key{RET} too many in this situation, @value{GDBN} disables command
1619 repetition after any command that generates this sort of display.
1621 @kindex # @r{(a comment)}
1623 Any text from a @kbd{#} to the end of the line is a comment; it does
1624 nothing. This is useful mainly in command files (@pxref{Command
1625 Files,,Command Files}).
1627 @cindex repeating command sequences
1628 @kindex Ctrl-o @r{(operate-and-get-next)}
1629 The @kbd{Ctrl-o} binding is useful for repeating a complex sequence of
1630 commands. This command accepts the current line, like @key{RET}, and
1631 then fetches the next line relative to the current line from the history
1635 @node Command Settings
1636 @section Command Settings
1637 @cindex default behavior of commands, changing
1638 @cindex default settings, changing
1640 Many commands change their behavior according to command-specific
1641 variables or settings. These settings can be changed with the
1642 @code{set} subcommands. For example, the @code{print} command
1643 (@pxref{Data, ,Examining Data}) prints arrays differently depending on
1644 settings changeable with the commands @code{set print elements
1645 NUMBER-OF-ELEMENTS} and @code{set print array-indexes}, among others.
1647 You can change these settings to your preference in the gdbinit files
1648 loaded at @value{GDBN} startup. @xref{Startup}.
1650 The settings can also be changed interactively during the debugging
1651 session. For example, to change the limit of array elements to print,
1652 you can do the following:
1654 (@value{GDBN}) set print elements 10
1655 (@value{GDBN}) print some_array
1656 $1 = @{0, 10, 20, 30, 40, 50, 60, 70, 80, 90...@}
1659 The above @code{set print elements 10} command changes the number of
1660 elements to print from the default of 200 to 10. If you only intend
1661 this limit of 10 to be used for printing @code{some_array}, then you
1662 must restore the limit back to 200, with @code{set print elements
1665 Some commands allow overriding settings with command options. For
1666 example, the @code{print} command supports a number of options that
1667 allow overriding relevant global print settings as set by @code{set
1668 print} subcommands. @xref{print options}. The example above could be
1671 (@value{GDBN}) print -elements 10 -- some_array
1672 $1 = @{0, 10, 20, 30, 40, 50, 60, 70, 80, 90...@}
1675 Alternatively, you can use the @code{with} command to change a setting
1676 temporarily, for the duration of a command invocation.
1679 @kindex with command
1680 @kindex w @r{(@code{with})}
1682 @cindex temporarily change settings
1683 @item with @var{setting} [@var{value}] [-- @var{command}]
1684 @itemx w @var{setting} [@var{value}] [-- @var{command}]
1685 Temporarily set @var{setting} to @var{value} for the duration of
1688 @var{setting} is any setting you can change with the @code{set}
1689 subcommands. @var{value} is the value to assign to @code{setting}
1690 while running @code{command}.
1692 If no @var{command} is provided, the last command executed is
1695 If a @var{command} is provided, it must be preceded by a double dash
1696 (@code{--}) separator. This is required because some settings accept
1697 free-form arguments, such as expressions or filenames.
1699 For example, the command
1701 (@value{GDBN}) with print array on -- print some_array
1704 is equivalent to the following 3 commands:
1706 (@value{GDBN}) set print array on
1707 (@value{GDBN}) print some_array
1708 (@value{GDBN}) set print array off
1711 The @code{with} command is particularly useful when you want to
1712 override a setting while running user-defined commands, or commands
1713 defined in Python or Guile. @xref{Extending GDB,, Extending GDB}.
1716 (@value{GDBN}) with print pretty on -- my_complex_command
1719 To change several settings for the same command, you can nest
1720 @code{with} commands. For example, @code{with language ada -- with
1721 print elements 10} temporarily changes the language to Ada and sets a
1722 limit of 10 elements to print for arrays and strings.
1727 @section Command Completion
1730 @cindex word completion
1731 @value{GDBN} can fill in the rest of a word in a command for you, if there is
1732 only one possibility; it can also show you what the valid possibilities
1733 are for the next word in a command, at any time. This works for @value{GDBN}
1734 commands, @value{GDBN} subcommands, command options, and the names of symbols
1737 Press the @key{TAB} key whenever you want @value{GDBN} to fill out the rest
1738 of a word. If there is only one possibility, @value{GDBN} fills in the
1739 word, and waits for you to finish the command (or press @key{RET} to
1740 enter it). For example, if you type
1742 @c FIXME "@key" does not distinguish its argument sufficiently to permit
1743 @c complete accuracy in these examples; space introduced for clarity.
1744 @c If texinfo enhancements make it unnecessary, it would be nice to
1745 @c replace " @key" by "@key" in the following...
1747 (@value{GDBP}) info bre @key{TAB}
1751 @value{GDBN} fills in the rest of the word @samp{breakpoints}, since that is
1752 the only @code{info} subcommand beginning with @samp{bre}:
1755 (@value{GDBP}) info breakpoints
1759 You can either press @key{RET} at this point, to run the @code{info
1760 breakpoints} command, or backspace and enter something else, if
1761 @samp{breakpoints} does not look like the command you expected. (If you
1762 were sure you wanted @code{info breakpoints} in the first place, you
1763 might as well just type @key{RET} immediately after @samp{info bre},
1764 to exploit command abbreviations rather than command completion).
1766 If there is more than one possibility for the next word when you press
1767 @key{TAB}, @value{GDBN} sounds a bell. You can either supply more
1768 characters and try again, or just press @key{TAB} a second time;
1769 @value{GDBN} displays all the possible completions for that word. For
1770 example, you might want to set a breakpoint on a subroutine whose name
1771 begins with @samp{make_}, but when you type @kbd{b make_@key{TAB}} @value{GDBN}
1772 just sounds the bell. Typing @key{TAB} again displays all the
1773 function names in your program that begin with those characters, for
1777 (@value{GDBP}) b make_ @key{TAB}
1778 @exdent @value{GDBN} sounds bell; press @key{TAB} again, to see:
1779 make_a_section_from_file make_environ
1780 make_abs_section make_function_type
1781 make_blockvector make_pointer_type
1782 make_cleanup make_reference_type
1783 make_command make_symbol_completion_list
1784 (@value{GDBP}) b make_
1788 After displaying the available possibilities, @value{GDBN} copies your
1789 partial input (@samp{b make_} in the example) so you can finish the
1792 If you just want to see the list of alternatives in the first place, you
1793 can press @kbd{M-?} rather than pressing @key{TAB} twice. @kbd{M-?}
1794 means @kbd{@key{META} ?}. You can type this either by holding down a
1795 key designated as the @key{META} shift on your keyboard (if there is
1796 one) while typing @kbd{?}, or as @key{ESC} followed by @kbd{?}.
1798 If the number of possible completions is large, @value{GDBN} will
1799 print as much of the list as it has collected, as well as a message
1800 indicating that the list may be truncated.
1803 (@value{GDBP}) b m@key{TAB}@key{TAB}
1805 <... the rest of the possible completions ...>
1806 *** List may be truncated, max-completions reached. ***
1811 This behavior can be controlled with the following commands:
1814 @kindex set max-completions
1815 @item set max-completions @var{limit}
1816 @itemx set max-completions unlimited
1817 Set the maximum number of completion candidates. @value{GDBN} will
1818 stop looking for more completions once it collects this many candidates.
1819 This is useful when completing on things like function names as collecting
1820 all the possible candidates can be time consuming.
1821 The default value is 200. A value of zero disables tab-completion.
1822 Note that setting either no limit or a very large limit can make
1824 @kindex show max-completions
1825 @item show max-completions
1826 Show the maximum number of candidates that @value{GDBN} will collect and show
1830 @cindex quotes in commands
1831 @cindex completion of quoted strings
1832 Sometimes the string you need, while logically a ``word'', may contain
1833 parentheses or other characters that @value{GDBN} normally excludes from
1834 its notion of a word. To permit word completion to work in this
1835 situation, you may enclose words in @code{'} (single quote marks) in
1836 @value{GDBN} commands.
1838 A likely situation where you might need this is in typing an
1839 expression that involves a C@t{++} symbol name with template
1840 parameters. This is because when completing expressions, GDB treats
1841 the @samp{<} character as word delimiter, assuming that it's the
1842 less-than comparison operator (@pxref{C Operators, , C and C@t{++}
1845 For example, when you want to call a C@t{++} template function
1846 interactively using the @code{print} or @code{call} commands, you may
1847 need to distinguish whether you mean the version of @code{name} that
1848 was specialized for @code{int}, @code{name<int>()}, or the version
1849 that was specialized for @code{float}, @code{name<float>()}. To use
1850 the word-completion facilities in this situation, type a single quote
1851 @code{'} at the beginning of the function name. This alerts
1852 @value{GDBN} that it may need to consider more information than usual
1853 when you press @key{TAB} or @kbd{M-?} to request word completion:
1856 (@value{GDBP}) p 'func< @kbd{M-?}
1857 func<int>() func<float>()
1858 (@value{GDBP}) p 'func<
1861 When setting breakpoints however (@pxref{Specify Location}), you don't
1862 usually need to type a quote before the function name, because
1863 @value{GDBN} understands that you want to set a breakpoint on a
1867 (@value{GDBP}) b func< @kbd{M-?}
1868 func<int>() func<float>()
1869 (@value{GDBP}) b func<
1872 This is true even in the case of typing the name of C@t{++} overloaded
1873 functions (multiple definitions of the same function, distinguished by
1874 argument type). For example, when you want to set a breakpoint you
1875 don't need to distinguish whether you mean the version of @code{name}
1876 that takes an @code{int} parameter, @code{name(int)}, or the version
1877 that takes a @code{float} parameter, @code{name(float)}.
1880 (@value{GDBP}) b bubble( @kbd{M-?}
1881 bubble(int) bubble(double)
1882 (@value{GDBP}) b bubble(dou @kbd{M-?}
1886 See @ref{quoting names} for a description of other scenarios that
1889 For more information about overloaded functions, see @ref{C Plus Plus
1890 Expressions, ,C@t{++} Expressions}. You can use the command @code{set
1891 overload-resolution off} to disable overload resolution;
1892 see @ref{Debugging C Plus Plus, ,@value{GDBN} Features for C@t{++}}.
1894 @cindex completion of structure field names
1895 @cindex structure field name completion
1896 @cindex completion of union field names
1897 @cindex union field name completion
1898 When completing in an expression which looks up a field in a
1899 structure, @value{GDBN} also tries@footnote{The completer can be
1900 confused by certain kinds of invalid expressions. Also, it only
1901 examines the static type of the expression, not the dynamic type.} to
1902 limit completions to the field names available in the type of the
1906 (@value{GDBP}) p gdb_stdout.@kbd{M-?}
1907 magic to_fputs to_rewind
1908 to_data to_isatty to_write
1909 to_delete to_put to_write_async_safe
1914 This is because the @code{gdb_stdout} is a variable of the type
1915 @code{struct ui_file} that is defined in @value{GDBN} sources as
1922 ui_file_flush_ftype *to_flush;
1923 ui_file_write_ftype *to_write;
1924 ui_file_write_async_safe_ftype *to_write_async_safe;
1925 ui_file_fputs_ftype *to_fputs;
1926 ui_file_read_ftype *to_read;
1927 ui_file_delete_ftype *to_delete;
1928 ui_file_isatty_ftype *to_isatty;
1929 ui_file_rewind_ftype *to_rewind;
1930 ui_file_put_ftype *to_put;
1935 @node Command Options
1936 @section Command options
1938 @cindex command options
1939 Some commands accept options starting with a leading dash. For
1940 example, @code{print -pretty}. Similarly to command names, you can
1941 abbreviate a @value{GDBN} option to the first few letters of the
1942 option name, if that abbreviation is unambiguous, and you can also use
1943 the @key{TAB} key to get @value{GDBN} to fill out the rest of a word
1944 in an option (or to show you the alternatives available, if there is
1945 more than one possibility).
1947 @cindex command options, raw input
1948 Some commands take raw input as argument. For example, the print
1949 command processes arbitrary expressions in any of the languages
1950 supported by @value{GDBN}. With such commands, because raw input may
1951 start with a leading dash that would be confused with an option or any
1952 of its abbreviations, e.g.@: @code{print -r} (short for @code{print
1953 -raw} or printing negative @code{r}?), if you specify any command
1954 option, then you must use a double-dash (@code{--}) delimiter to
1955 indicate the end of options.
1957 @cindex command options, boolean
1959 Some options are described as accepting an argument which can be
1960 either @code{on} or @code{off}. These are known as @dfn{boolean
1961 options}. Similarly to boolean settings commands---@code{on} and
1962 @code{off} are the typical values, but any of @code{1}, @code{yes} and
1963 @code{enable} can also be used as ``true'' value, and any of @code{0},
1964 @code{no} and @code{disable} can also be used as ``false'' value. You
1965 can also omit a ``true'' value, as it is implied by default.
1967 For example, these are equivalent:
1970 (@value{GDBP}) print -object on -pretty off -element unlimited -- *myptr
1971 (@value{GDBP}) p -o -p 0 -e u -- *myptr
1974 You can discover the set of options some command accepts by completing
1975 on @code{-} after the command name. For example:
1978 (@value{GDBP}) print -@key{TAB}@key{TAB}
1979 -address -max-depth -repeats -vtbl
1980 -array -null-stop -static-members
1981 -array-indexes -object -symbol
1982 -elements -pretty -union
1985 Completion will in some cases guide you with a suggestion of what kind
1986 of argument an option expects. For example:
1989 (@value{GDBP}) print -elements @key{TAB}@key{TAB}
1993 Here, the option expects a number (e.g., @code{100}), not literal
1994 @code{NUMBER}. Such metasyntactical arguments are always presented in
1997 (For more on using the @code{print} command, see @ref{Data, ,Examining
2001 @section Getting Help
2002 @cindex online documentation
2005 You can always ask @value{GDBN} itself for information on its commands,
2006 using the command @code{help}.
2009 @kindex h @r{(@code{help})}
2012 You can use @code{help} (abbreviated @code{h}) with no arguments to
2013 display a short list of named classes of commands:
2017 List of classes of commands:
2019 aliases -- Aliases of other commands
2020 breakpoints -- Making program stop at certain points
2021 data -- Examining data
2022 files -- Specifying and examining files
2023 internals -- Maintenance commands
2024 obscure -- Obscure features
2025 running -- Running the program
2026 stack -- Examining the stack
2027 status -- Status inquiries
2028 support -- Support facilities
2029 tracepoints -- Tracing of program execution without
2030 stopping the program
2031 user-defined -- User-defined commands
2033 Type "help" followed by a class name for a list of
2034 commands in that class.
2035 Type "help" followed by command name for full
2037 Command name abbreviations are allowed if unambiguous.
2040 @c the above line break eliminates huge line overfull...
2042 @item help @var{class}
2043 Using one of the general help classes as an argument, you can get a
2044 list of the individual commands in that class. For example, here is the
2045 help display for the class @code{status}:
2048 (@value{GDBP}) help status
2053 @c Line break in "show" line falsifies real output, but needed
2054 @c to fit in smallbook page size.
2055 info -- Generic command for showing things
2056 about the program being debugged
2057 show -- Generic command for showing things
2060 Type "help" followed by command name for full
2062 Command name abbreviations are allowed if unambiguous.
2066 @item help @var{command}
2067 With a command name as @code{help} argument, @value{GDBN} displays a
2068 short paragraph on how to use that command.
2071 @item apropos [-v] @var{regexp}
2072 The @code{apropos} command searches through all of the @value{GDBN}
2073 commands, and their documentation, for the regular expression specified in
2074 @var{args}. It prints out all matches found. The optional flag @samp{-v},
2075 which stands for @samp{verbose}, indicates to output the full documentation
2076 of the matching commands and highlight the parts of the documentation
2077 matching @var{regexp}. For example:
2088 alias -- Define a new command that is an alias of an existing command
2089 aliases -- Aliases of other commands
2090 d -- Delete some breakpoints or auto-display expressions
2091 del -- Delete some breakpoints or auto-display expressions
2092 delete -- Delete some breakpoints or auto-display expressions
2100 apropos -v cut.*thread apply
2104 results in the below output, where @samp{cut for 'thread apply}
2105 is highlighted if styling is enabled.
2109 taas -- Apply a command to all threads (ignoring errors
2112 shortcut for 'thread apply all -s COMMAND'
2114 tfaas -- Apply a command to all frames of all threads
2115 (ignoring errors and empty output).
2116 Usage: tfaas COMMAND
2117 shortcut for 'thread apply all -s frame apply all -s COMMAND'
2122 @item complete @var{args}
2123 The @code{complete @var{args}} command lists all the possible completions
2124 for the beginning of a command. Use @var{args} to specify the beginning of the
2125 command you want completed. For example:
2131 @noindent results in:
2142 @noindent This is intended for use by @sc{gnu} Emacs.
2145 In addition to @code{help}, you can use the @value{GDBN} commands @code{info}
2146 and @code{show} to inquire about the state of your program, or the state
2147 of @value{GDBN} itself. Each command supports many topics of inquiry; this
2148 manual introduces each of them in the appropriate context. The listings
2149 under @code{info} and under @code{show} in the Command, Variable, and
2150 Function Index point to all the sub-commands. @xref{Command and Variable
2156 @kindex i @r{(@code{info})}
2158 This command (abbreviated @code{i}) is for describing the state of your
2159 program. For example, you can show the arguments passed to a function
2160 with @code{info args}, list the registers currently in use with @code{info
2161 registers}, or list the breakpoints you have set with @code{info breakpoints}.
2162 You can get a complete list of the @code{info} sub-commands with
2163 @w{@code{help info}}.
2167 You can assign the result of an expression to an environment variable with
2168 @code{set}. For example, you can set the @value{GDBN} prompt to a $-sign with
2169 @code{set prompt $}.
2173 In contrast to @code{info}, @code{show} is for describing the state of
2174 @value{GDBN} itself.
2175 You can change most of the things you can @code{show}, by using the
2176 related command @code{set}; for example, you can control what number
2177 system is used for displays with @code{set radix}, or simply inquire
2178 which is currently in use with @code{show radix}.
2181 To display all the settable parameters and their current
2182 values, you can use @code{show} with no arguments; you may also use
2183 @code{info set}. Both commands produce the same display.
2184 @c FIXME: "info set" violates the rule that "info" is for state of
2185 @c FIXME...program. Ck w/ GNU: "info set" to be called something else,
2186 @c FIXME...or change desc of rule---eg "state of prog and debugging session"?
2190 Here are several miscellaneous @code{show} subcommands, all of which are
2191 exceptional in lacking corresponding @code{set} commands:
2194 @kindex show version
2195 @cindex @value{GDBN} version number
2197 Show what version of @value{GDBN} is running. You should include this
2198 information in @value{GDBN} bug-reports. If multiple versions of
2199 @value{GDBN} are in use at your site, you may need to determine which
2200 version of @value{GDBN} you are running; as @value{GDBN} evolves, new
2201 commands are introduced, and old ones may wither away. Also, many
2202 system vendors ship variant versions of @value{GDBN}, and there are
2203 variant versions of @value{GDBN} in @sc{gnu}/Linux distributions as well.
2204 The version number is the same as the one announced when you start
2207 @kindex show copying
2208 @kindex info copying
2209 @cindex display @value{GDBN} copyright
2212 Display information about permission for copying @value{GDBN}.
2214 @kindex show warranty
2215 @kindex info warranty
2217 @itemx info warranty
2218 Display the @sc{gnu} ``NO WARRANTY'' statement, or a warranty,
2219 if your version of @value{GDBN} comes with one.
2221 @kindex show configuration
2222 @item show configuration
2223 Display detailed information about the way @value{GDBN} was configured
2224 when it was built. This displays the optional arguments passed to the
2225 @file{configure} script and also configuration parameters detected
2226 automatically by @command{configure}. When reporting a @value{GDBN}
2227 bug (@pxref{GDB Bugs}), it is important to include this information in
2233 @chapter Running Programs Under @value{GDBN}
2235 When you run a program under @value{GDBN}, you must first generate
2236 debugging information when you compile it.
2238 You may start @value{GDBN} with its arguments, if any, in an environment
2239 of your choice. If you are doing native debugging, you may redirect
2240 your program's input and output, debug an already running process, or
2241 kill a child process.
2244 * Compilation:: Compiling for debugging
2245 * Starting:: Starting your program
2246 * Arguments:: Your program's arguments
2247 * Environment:: Your program's environment
2249 * Working Directory:: Your program's working directory
2250 * Input/Output:: Your program's input and output
2251 * Attach:: Debugging an already-running process
2252 * Kill Process:: Killing the child process
2254 * Inferiors and Programs:: Debugging multiple inferiors and programs
2255 * Threads:: Debugging programs with multiple threads
2256 * Forks:: Debugging forks
2257 * Checkpoint/Restart:: Setting a @emph{bookmark} to return to later
2261 @section Compiling for Debugging
2263 In order to debug a program effectively, you need to generate
2264 debugging information when you compile it. This debugging information
2265 is stored in the object file; it describes the data type of each
2266 variable or function and the correspondence between source line numbers
2267 and addresses in the executable code.
2269 To request debugging information, specify the @samp{-g} option when you run
2272 Programs that are to be shipped to your customers are compiled with
2273 optimizations, using the @samp{-O} compiler option. However, some
2274 compilers are unable to handle the @samp{-g} and @samp{-O} options
2275 together. Using those compilers, you cannot generate optimized
2276 executables containing debugging information.
2278 @value{NGCC}, the @sc{gnu} C/C@t{++} compiler, supports @samp{-g} with or
2279 without @samp{-O}, making it possible to debug optimized code. We
2280 recommend that you @emph{always} use @samp{-g} whenever you compile a
2281 program. You may think your program is correct, but there is no sense
2282 in pushing your luck. For more information, see @ref{Optimized Code}.
2284 Older versions of the @sc{gnu} C compiler permitted a variant option
2285 @w{@samp{-gg}} for debugging information. @value{GDBN} no longer supports this
2286 format; if your @sc{gnu} C compiler has this option, do not use it.
2288 @value{GDBN} knows about preprocessor macros and can show you their
2289 expansion (@pxref{Macros}). Most compilers do not include information
2290 about preprocessor macros in the debugging information if you specify
2291 the @option{-g} flag alone. Version 3.1 and later of @value{NGCC},
2292 the @sc{gnu} C compiler, provides macro information if you are using
2293 the DWARF debugging format, and specify the option @option{-g3}.
2295 @xref{Debugging Options,,Options for Debugging Your Program or GCC,
2296 gcc, Using the @sc{gnu} Compiler Collection (GCC)}, for more
2297 information on @value{NGCC} options affecting debug information.
2299 You will have the best debugging experience if you use the latest
2300 version of the DWARF debugging format that your compiler supports.
2301 DWARF is currently the most expressive and best supported debugging
2302 format in @value{GDBN}.
2306 @section Starting your Program
2312 @kindex r @r{(@code{run})}
2315 Use the @code{run} command to start your program under @value{GDBN}.
2316 You must first specify the program name with an argument to
2317 @value{GDBN} (@pxref{Invocation, ,Getting In and Out of
2318 @value{GDBN}}), or by using the @code{file} or @code{exec-file}
2319 command (@pxref{Files, ,Commands to Specify Files}).
2323 If you are running your program in an execution environment that
2324 supports processes, @code{run} creates an inferior process and makes
2325 that process run your program. In some environments without processes,
2326 @code{run} jumps to the start of your program. Other targets,
2327 like @samp{remote}, are always running. If you get an error
2328 message like this one:
2331 The "remote" target does not support "run".
2332 Try "help target" or "continue".
2336 then use @code{continue} to run your program. You may need @code{load}
2337 first (@pxref{load}).
2339 The execution of a program is affected by certain information it
2340 receives from its superior. @value{GDBN} provides ways to specify this
2341 information, which you must do @emph{before} starting your program. (You
2342 can change it after starting your program, but such changes only affect
2343 your program the next time you start it.) This information may be
2344 divided into four categories:
2347 @item The @emph{arguments.}
2348 Specify the arguments to give your program as the arguments of the
2349 @code{run} command. If a shell is available on your target, the shell
2350 is used to pass the arguments, so that you may use normal conventions
2351 (such as wildcard expansion or variable substitution) in describing
2353 In Unix systems, you can control which shell is used with the
2354 @code{SHELL} environment variable. If you do not define @code{SHELL},
2355 @value{GDBN} uses the default shell (@file{/bin/sh}). You can disable
2356 use of any shell with the @code{set startup-with-shell} command (see
2359 @item The @emph{environment.}
2360 Your program normally inherits its environment from @value{GDBN}, but you can
2361 use the @value{GDBN} commands @code{set environment} and @code{unset
2362 environment} to change parts of the environment that affect
2363 your program. @xref{Environment, ,Your Program's Environment}.
2365 @item The @emph{working directory.}
2366 You can set your program's working directory with the command
2367 @kbd{set cwd}. If you do not set any working directory with this
2368 command, your program will inherit @value{GDBN}'s working directory if
2369 native debugging, or the remote server's working directory if remote
2370 debugging. @xref{Working Directory, ,Your Program's Working
2373 @item The @emph{standard input and output.}
2374 Your program normally uses the same device for standard input and
2375 standard output as @value{GDBN} is using. You can redirect input and output
2376 in the @code{run} command line, or you can use the @code{tty} command to
2377 set a different device for your program.
2378 @xref{Input/Output, ,Your Program's Input and Output}.
2381 @emph{Warning:} While input and output redirection work, you cannot use
2382 pipes to pass the output of the program you are debugging to another
2383 program; if you attempt this, @value{GDBN} is likely to wind up debugging the
2387 When you issue the @code{run} command, your program begins to execute
2388 immediately. @xref{Stopping, ,Stopping and Continuing}, for discussion
2389 of how to arrange for your program to stop. Once your program has
2390 stopped, you may call functions in your program, using the @code{print}
2391 or @code{call} commands. @xref{Data, ,Examining Data}.
2393 If the modification time of your symbol file has changed since the last
2394 time @value{GDBN} read its symbols, @value{GDBN} discards its symbol
2395 table, and reads it again. When it does this, @value{GDBN} tries to retain
2396 your current breakpoints.
2401 @cindex run to main procedure
2402 The name of the main procedure can vary from language to language.
2403 With C or C@t{++}, the main procedure name is always @code{main}, but
2404 other languages such as Ada do not require a specific name for their
2405 main procedure. The debugger provides a convenient way to start the
2406 execution of the program and to stop at the beginning of the main
2407 procedure, depending on the language used.
2409 The @samp{start} command does the equivalent of setting a temporary
2410 breakpoint at the beginning of the main procedure and then invoking
2411 the @samp{run} command.
2413 @cindex elaboration phase
2414 Some programs contain an @dfn{elaboration} phase where some startup code is
2415 executed before the main procedure is called. This depends on the
2416 languages used to write your program. In C@t{++}, for instance,
2417 constructors for static and global objects are executed before
2418 @code{main} is called. It is therefore possible that the debugger stops
2419 before reaching the main procedure. However, the temporary breakpoint
2420 will remain to halt execution.
2422 Specify the arguments to give to your program as arguments to the
2423 @samp{start} command. These arguments will be given verbatim to the
2424 underlying @samp{run} command. Note that the same arguments will be
2425 reused if no argument is provided during subsequent calls to
2426 @samp{start} or @samp{run}.
2428 It is sometimes necessary to debug the program during elaboration. In
2429 these cases, using the @code{start} command would stop the execution
2430 of your program too late, as the program would have already completed
2431 the elaboration phase. Under these circumstances, either insert
2432 breakpoints in your elaboration code before running your program or
2433 use the @code{starti} command.
2437 @cindex run to first instruction
2438 The @samp{starti} command does the equivalent of setting a temporary
2439 breakpoint at the first instruction of a program's execution and then
2440 invoking the @samp{run} command. For programs containing an
2441 elaboration phase, the @code{starti} command will stop execution at
2442 the start of the elaboration phase.
2444 @anchor{set exec-wrapper}
2445 @kindex set exec-wrapper
2446 @item set exec-wrapper @var{wrapper}
2447 @itemx show exec-wrapper
2448 @itemx unset exec-wrapper
2449 When @samp{exec-wrapper} is set, the specified wrapper is used to
2450 launch programs for debugging. @value{GDBN} starts your program
2451 with a shell command of the form @kbd{exec @var{wrapper}
2452 @var{program}}. Quoting is added to @var{program} and its
2453 arguments, but not to @var{wrapper}, so you should add quotes if
2454 appropriate for your shell. The wrapper runs until it executes
2455 your program, and then @value{GDBN} takes control.
2457 You can use any program that eventually calls @code{execve} with
2458 its arguments as a wrapper. Several standard Unix utilities do
2459 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
2460 with @code{exec "$@@"} will also work.
2462 For example, you can use @code{env} to pass an environment variable to
2463 the debugged program, without setting the variable in your shell's
2467 (@value{GDBP}) set exec-wrapper env 'LD_PRELOAD=libtest.so'
2471 This command is available when debugging locally on most targets, excluding
2472 @sc{djgpp}, Cygwin, MS Windows, and QNX Neutrino.
2474 @kindex set startup-with-shell
2475 @anchor{set startup-with-shell}
2476 @item set startup-with-shell
2477 @itemx set startup-with-shell on
2478 @itemx set startup-with-shell off
2479 @itemx show startup-with-shell
2480 On Unix systems, by default, if a shell is available on your target,
2481 @value{GDBN}) uses it to start your program. Arguments of the
2482 @code{run} command are passed to the shell, which does variable
2483 substitution, expands wildcard characters and performs redirection of
2484 I/O. In some circumstances, it may be useful to disable such use of a
2485 shell, for example, when debugging the shell itself or diagnosing
2486 startup failures such as:
2490 Starting program: ./a.out
2491 During startup program terminated with signal SIGSEGV, Segmentation fault.
2495 which indicates the shell or the wrapper specified with
2496 @samp{exec-wrapper} crashed, not your program. Most often, this is
2497 caused by something odd in your shell's non-interactive mode
2498 initialization file---such as @file{.cshrc} for C-shell,
2499 $@file{.zshenv} for the Z shell, or the file specified in the
2500 @samp{BASH_ENV} environment variable for BASH.
2502 @anchor{set auto-connect-native-target}
2503 @kindex set auto-connect-native-target
2504 @item set auto-connect-native-target
2505 @itemx set auto-connect-native-target on
2506 @itemx set auto-connect-native-target off
2507 @itemx show auto-connect-native-target
2509 By default, if not connected to any target yet (e.g., with
2510 @code{target remote}), the @code{run} command starts your program as a
2511 native process under @value{GDBN}, on your local machine. If you're
2512 sure you don't want to debug programs on your local machine, you can
2513 tell @value{GDBN} to not connect to the native target automatically
2514 with the @code{set auto-connect-native-target off} command.
2516 If @code{on}, which is the default, and if @value{GDBN} is not
2517 connected to a target already, the @code{run} command automaticaly
2518 connects to the native target, if one is available.
2520 If @code{off}, and if @value{GDBN} is not connected to a target
2521 already, the @code{run} command fails with an error:
2525 Don't know how to run. Try "help target".
2528 If @value{GDBN} is already connected to a target, @value{GDBN} always
2529 uses it with the @code{run} command.
2531 In any case, you can explicitly connect to the native target with the
2532 @code{target native} command. For example,
2535 (@value{GDBP}) set auto-connect-native-target off
2537 Don't know how to run. Try "help target".
2538 (@value{GDBP}) target native
2540 Starting program: ./a.out
2541 [Inferior 1 (process 10421) exited normally]
2544 In case you connected explicitly to the @code{native} target,
2545 @value{GDBN} remains connected even if all inferiors exit, ready for
2546 the next @code{run} command. Use the @code{disconnect} command to
2549 Examples of other commands that likewise respect the
2550 @code{auto-connect-native-target} setting: @code{attach}, @code{info
2551 proc}, @code{info os}.
2553 @kindex set disable-randomization
2554 @item set disable-randomization
2555 @itemx set disable-randomization on
2556 This option (enabled by default in @value{GDBN}) will turn off the native
2557 randomization of the virtual address space of the started program. This option
2558 is useful for multiple debugging sessions to make the execution better
2559 reproducible and memory addresses reusable across debugging sessions.
2561 This feature is implemented only on certain targets, including @sc{gnu}/Linux.
2562 On @sc{gnu}/Linux you can get the same behavior using
2565 (@value{GDBP}) set exec-wrapper setarch `uname -m` -R
2568 @item set disable-randomization off
2569 Leave the behavior of the started executable unchanged. Some bugs rear their
2570 ugly heads only when the program is loaded at certain addresses. If your bug
2571 disappears when you run the program under @value{GDBN}, that might be because
2572 @value{GDBN} by default disables the address randomization on platforms, such
2573 as @sc{gnu}/Linux, which do that for stand-alone programs. Use @kbd{set
2574 disable-randomization off} to try to reproduce such elusive bugs.
2576 On targets where it is available, virtual address space randomization
2577 protects the programs against certain kinds of security attacks. In these
2578 cases the attacker needs to know the exact location of a concrete executable
2579 code. Randomizing its location makes it impossible to inject jumps misusing
2580 a code at its expected addresses.
2582 Prelinking shared libraries provides a startup performance advantage but it
2583 makes addresses in these libraries predictable for privileged processes by
2584 having just unprivileged access at the target system. Reading the shared
2585 library binary gives enough information for assembling the malicious code
2586 misusing it. Still even a prelinked shared library can get loaded at a new
2587 random address just requiring the regular relocation process during the
2588 startup. Shared libraries not already prelinked are always loaded at
2589 a randomly chosen address.
2591 Position independent executables (PIE) contain position independent code
2592 similar to the shared libraries and therefore such executables get loaded at
2593 a randomly chosen address upon startup. PIE executables always load even
2594 already prelinked shared libraries at a random address. You can build such
2595 executable using @command{gcc -fPIE -pie}.
2597 Heap (malloc storage), stack and custom mmap areas are always placed randomly
2598 (as long as the randomization is enabled).
2600 @item show disable-randomization
2601 Show the current setting of the explicit disable of the native randomization of
2602 the virtual address space of the started program.
2607 @section Your Program's Arguments
2609 @cindex arguments (to your program)
2610 The arguments to your program can be specified by the arguments of the
2612 They are passed to a shell, which expands wildcard characters and
2613 performs redirection of I/O, and thence to your program. Your
2614 @code{SHELL} environment variable (if it exists) specifies what shell
2615 @value{GDBN} uses. If you do not define @code{SHELL}, @value{GDBN} uses
2616 the default shell (@file{/bin/sh} on Unix).
2618 On non-Unix systems, the program is usually invoked directly by
2619 @value{GDBN}, which emulates I/O redirection via the appropriate system
2620 calls, and the wildcard characters are expanded by the startup code of
2621 the program, not by the shell.
2623 @code{run} with no arguments uses the same arguments used by the previous
2624 @code{run}, or those set by the @code{set args} command.
2629 Specify the arguments to be used the next time your program is run. If
2630 @code{set args} has no arguments, @code{run} executes your program
2631 with no arguments. Once you have run your program with arguments,
2632 using @code{set args} before the next @code{run} is the only way to run
2633 it again without arguments.
2637 Show the arguments to give your program when it is started.
2641 @section Your Program's Environment
2643 @cindex environment (of your program)
2644 The @dfn{environment} consists of a set of environment variables and
2645 their values. Environment variables conventionally record such things as
2646 your user name, your home directory, your terminal type, and your search
2647 path for programs to run. Usually you set up environment variables with
2648 the shell and they are inherited by all the other programs you run. When
2649 debugging, it can be useful to try running your program with a modified
2650 environment without having to start @value{GDBN} over again.
2654 @item path @var{directory}
2655 Add @var{directory} to the front of the @code{PATH} environment variable
2656 (the search path for executables) that will be passed to your program.
2657 The value of @code{PATH} used by @value{GDBN} does not change.
2658 You may specify several directory names, separated by whitespace or by a
2659 system-dependent separator character (@samp{:} on Unix, @samp{;} on
2660 MS-DOS and MS-Windows). If @var{directory} is already in the path, it
2661 is moved to the front, so it is searched sooner.
2663 You can use the string @samp{$cwd} to refer to whatever is the current
2664 working directory at the time @value{GDBN} searches the path. If you
2665 use @samp{.} instead, it refers to the directory where you executed the
2666 @code{path} command. @value{GDBN} replaces @samp{.} in the
2667 @var{directory} argument (with the current path) before adding
2668 @var{directory} to the search path.
2669 @c 'path' is explicitly nonrepeatable, but RMS points out it is silly to
2670 @c document that, since repeating it would be a no-op.
2674 Display the list of search paths for executables (the @code{PATH}
2675 environment variable).
2677 @kindex show environment
2678 @item show environment @r{[}@var{varname}@r{]}
2679 Print the value of environment variable @var{varname} to be given to
2680 your program when it starts. If you do not supply @var{varname},
2681 print the names and values of all environment variables to be given to
2682 your program. You can abbreviate @code{environment} as @code{env}.
2684 @kindex set environment
2685 @anchor{set environment}
2686 @item set environment @var{varname} @r{[}=@var{value}@r{]}
2687 Set environment variable @var{varname} to @var{value}. The value
2688 changes for your program (and the shell @value{GDBN} uses to launch
2689 it), not for @value{GDBN} itself. The @var{value} may be any string; the
2690 values of environment variables are just strings, and any
2691 interpretation is supplied by your program itself. The @var{value}
2692 parameter is optional; if it is eliminated, the variable is set to a
2694 @c "any string" here does not include leading, trailing
2695 @c blanks. Gnu asks: does anyone care?
2697 For example, this command:
2704 tells the debugged program, when subsequently run, that its user is named
2705 @samp{foo}. (The spaces around @samp{=} are used for clarity here; they
2706 are not actually required.)
2708 Note that on Unix systems, @value{GDBN} runs your program via a shell,
2709 which also inherits the environment set with @code{set environment}.
2710 If necessary, you can avoid that by using the @samp{env} program as a
2711 wrapper instead of using @code{set environment}. @xref{set
2712 exec-wrapper}, for an example doing just that.
2714 Environment variables that are set by the user are also transmitted to
2715 @command{gdbserver} to be used when starting the remote inferior.
2716 @pxref{QEnvironmentHexEncoded}.
2718 @kindex unset environment
2719 @anchor{unset environment}
2720 @item unset environment @var{varname}
2721 Remove variable @var{varname} from the environment to be passed to your
2722 program. This is different from @samp{set env @var{varname} =};
2723 @code{unset environment} removes the variable from the environment,
2724 rather than assigning it an empty value.
2726 Environment variables that are unset by the user are also unset on
2727 @command{gdbserver} when starting the remote inferior.
2728 @pxref{QEnvironmentUnset}.
2731 @emph{Warning:} On Unix systems, @value{GDBN} runs your program using
2732 the shell indicated by your @code{SHELL} environment variable if it
2733 exists (or @code{/bin/sh} if not). If your @code{SHELL} variable
2734 names a shell that runs an initialization file when started
2735 non-interactively---such as @file{.cshrc} for C-shell, $@file{.zshenv}
2736 for the Z shell, or the file specified in the @samp{BASH_ENV}
2737 environment variable for BASH---any variables you set in that file
2738 affect your program. You may wish to move setting of environment
2739 variables to files that are only run when you sign on, such as
2740 @file{.login} or @file{.profile}.
2742 @node Working Directory
2743 @section Your Program's Working Directory
2745 @cindex working directory (of your program)
2746 Each time you start your program with @code{run}, the inferior will be
2747 initialized with the current working directory specified by the
2748 @kbd{set cwd} command. If no directory has been specified by this
2749 command, then the inferior will inherit @value{GDBN}'s current working
2750 directory as its working directory if native debugging, or it will
2751 inherit the remote server's current working directory if remote
2756 @cindex change inferior's working directory
2757 @anchor{set cwd command}
2758 @item set cwd @r{[}@var{directory}@r{]}
2759 Set the inferior's working directory to @var{directory}, which will be
2760 @code{glob}-expanded in order to resolve tildes (@file{~}). If no
2761 argument has been specified, the command clears the setting and resets
2762 it to an empty state. This setting has no effect on @value{GDBN}'s
2763 working directory, and it only takes effect the next time you start
2764 the inferior. The @file{~} in @var{directory} is a short for the
2765 @dfn{home directory}, usually pointed to by the @env{HOME} environment
2766 variable. On MS-Windows, if @env{HOME} is not defined, @value{GDBN}
2767 uses the concatenation of @env{HOMEDRIVE} and @env{HOMEPATH} as
2770 You can also change @value{GDBN}'s current working directory by using
2771 the @code{cd} command.
2775 @cindex show inferior's working directory
2777 Show the inferior's working directory. If no directory has been
2778 specified by @kbd{set cwd}, then the default inferior's working
2779 directory is the same as @value{GDBN}'s working directory.
2782 @cindex change @value{GDBN}'s working directory
2784 @item cd @r{[}@var{directory}@r{]}
2785 Set the @value{GDBN} working directory to @var{directory}. If not
2786 given, @var{directory} uses @file{'~'}.
2788 The @value{GDBN} working directory serves as a default for the
2789 commands that specify files for @value{GDBN} to operate on.
2790 @xref{Files, ,Commands to Specify Files}.
2791 @xref{set cwd command}.
2795 Print the @value{GDBN} working directory.
2798 It is generally impossible to find the current working directory of
2799 the process being debugged (since a program can change its directory
2800 during its run). If you work on a system where @value{GDBN} supports
2801 the @code{info proc} command (@pxref{Process Information}), you can
2802 use the @code{info proc} command to find out the
2803 current working directory of the debuggee.
2806 @section Your Program's Input and Output
2811 By default, the program you run under @value{GDBN} does input and output to
2812 the same terminal that @value{GDBN} uses. @value{GDBN} switches the terminal
2813 to its own terminal modes to interact with you, but it records the terminal
2814 modes your program was using and switches back to them when you continue
2815 running your program.
2818 @kindex info terminal
2820 Displays information recorded by @value{GDBN} about the terminal modes your
2824 You can redirect your program's input and/or output using shell
2825 redirection with the @code{run} command. For example,
2832 starts your program, diverting its output to the file @file{outfile}.
2835 @cindex controlling terminal
2836 Another way to specify where your program should do input and output is
2837 with the @code{tty} command. This command accepts a file name as
2838 argument, and causes this file to be the default for future @code{run}
2839 commands. It also resets the controlling terminal for the child
2840 process, for future @code{run} commands. For example,
2847 directs that processes started with subsequent @code{run} commands
2848 default to do input and output on the terminal @file{/dev/ttyb} and have
2849 that as their controlling terminal.
2851 An explicit redirection in @code{run} overrides the @code{tty} command's
2852 effect on the input/output device, but not its effect on the controlling
2855 When you use the @code{tty} command or redirect input in the @code{run}
2856 command, only the input @emph{for your program} is affected. The input
2857 for @value{GDBN} still comes from your terminal. @code{tty} is an alias
2858 for @code{set inferior-tty}.
2860 @cindex inferior tty
2861 @cindex set inferior controlling terminal
2862 You can use the @code{show inferior-tty} command to tell @value{GDBN} to
2863 display the name of the terminal that will be used for future runs of your
2867 @item set inferior-tty [ @var{tty} ]
2868 @kindex set inferior-tty
2869 Set the tty for the program being debugged to @var{tty}. Omitting @var{tty}
2870 restores the default behavior, which is to use the same terminal as
2873 @item show inferior-tty
2874 @kindex show inferior-tty
2875 Show the current tty for the program being debugged.
2879 @section Debugging an Already-running Process
2884 @item attach @var{process-id}
2885 This command attaches to a running process---one that was started
2886 outside @value{GDBN}. (@code{info files} shows your active
2887 targets.) The command takes as argument a process ID. The usual way to
2888 find out the @var{process-id} of a Unix process is with the @code{ps} utility,
2889 or with the @samp{jobs -l} shell command.
2891 @code{attach} does not repeat if you press @key{RET} a second time after
2892 executing the command.
2895 To use @code{attach}, your program must be running in an environment
2896 which supports processes; for example, @code{attach} does not work for
2897 programs on bare-board targets that lack an operating system. You must
2898 also have permission to send the process a signal.
2900 When you use @code{attach}, the debugger finds the program running in
2901 the process first by looking in the current working directory, then (if
2902 the program is not found) by using the source file search path
2903 (@pxref{Source Path, ,Specifying Source Directories}). You can also use
2904 the @code{file} command to load the program. @xref{Files, ,Commands to
2907 The first thing @value{GDBN} does after arranging to debug the specified
2908 process is to stop it. You can examine and modify an attached process
2909 with all the @value{GDBN} commands that are ordinarily available when
2910 you start processes with @code{run}. You can insert breakpoints; you
2911 can step and continue; you can modify storage. If you would rather the
2912 process continue running, you may use the @code{continue} command after
2913 attaching @value{GDBN} to the process.
2918 When you have finished debugging the attached process, you can use the
2919 @code{detach} command to release it from @value{GDBN} control. Detaching
2920 the process continues its execution. After the @code{detach} command,
2921 that process and @value{GDBN} become completely independent once more, and you
2922 are ready to @code{attach} another process or start one with @code{run}.
2923 @code{detach} does not repeat if you press @key{RET} again after
2924 executing the command.
2927 If you exit @value{GDBN} while you have an attached process, you detach
2928 that process. If you use the @code{run} command, you kill that process.
2929 By default, @value{GDBN} asks for confirmation if you try to do either of these
2930 things; you can control whether or not you need to confirm by using the
2931 @code{set confirm} command (@pxref{Messages/Warnings, ,Optional Warnings and
2935 @section Killing the Child Process
2940 Kill the child process in which your program is running under @value{GDBN}.
2943 This command is useful if you wish to debug a core dump instead of a
2944 running process. @value{GDBN} ignores any core dump file while your program
2947 On some operating systems, a program cannot be executed outside @value{GDBN}
2948 while you have breakpoints set on it inside @value{GDBN}. You can use the
2949 @code{kill} command in this situation to permit running your program
2950 outside the debugger.
2952 The @code{kill} command is also useful if you wish to recompile and
2953 relink your program, since on many systems it is impossible to modify an
2954 executable file while it is running in a process. In this case, when you
2955 next type @code{run}, @value{GDBN} notices that the file has changed, and
2956 reads the symbol table again (while trying to preserve your current
2957 breakpoint settings).
2959 @node Inferiors and Programs
2960 @section Debugging Multiple Inferiors and Programs
2962 @value{GDBN} lets you run and debug multiple programs in a single
2963 session. In addition, @value{GDBN} on some systems may let you run
2964 several programs simultaneously (otherwise you have to exit from one
2965 before starting another). In the most general case, you can have
2966 multiple threads of execution in each of multiple processes, launched
2967 from multiple executables.
2970 @value{GDBN} represents the state of each program execution with an
2971 object called an @dfn{inferior}. An inferior typically corresponds to
2972 a process, but is more general and applies also to targets that do not
2973 have processes. Inferiors may be created before a process runs, and
2974 may be retained after a process exits. Inferiors have unique
2975 identifiers that are different from process ids. Usually each
2976 inferior will also have its own distinct address space, although some
2977 embedded targets may have several inferiors running in different parts
2978 of a single address space. Each inferior may in turn have multiple
2979 threads running in it.
2981 To find out what inferiors exist at any moment, use @w{@code{info
2985 @kindex info inferiors [ @var{id}@dots{} ]
2986 @item info inferiors
2987 Print a list of all inferiors currently being managed by @value{GDBN}.
2988 By default all inferiors are printed, but the argument @var{id}@dots{}
2989 -- a space separated list of inferior numbers -- can be used to limit
2990 the display to just the requested inferiors.
2992 @value{GDBN} displays for each inferior (in this order):
2996 the inferior number assigned by @value{GDBN}
2999 the target system's inferior identifier
3002 the name of the executable the inferior is running.
3007 An asterisk @samp{*} preceding the @value{GDBN} inferior number
3008 indicates the current inferior.
3012 @c end table here to get a little more width for example
3015 (@value{GDBP}) info inferiors
3016 Num Description Executable
3017 2 process 2307 hello
3018 * 1 process 3401 goodbye
3021 To switch focus between inferiors, use the @code{inferior} command:
3024 @kindex inferior @var{infno}
3025 @item inferior @var{infno}
3026 Make inferior number @var{infno} the current inferior. The argument
3027 @var{infno} is the inferior number assigned by @value{GDBN}, as shown
3028 in the first field of the @samp{info inferiors} display.
3031 @vindex $_inferior@r{, convenience variable}
3032 The debugger convenience variable @samp{$_inferior} contains the
3033 number of the current inferior. You may find this useful in writing
3034 breakpoint conditional expressions, command scripts, and so forth.
3035 @xref{Convenience Vars,, Convenience Variables}, for general
3036 information on convenience variables.
3038 You can get multiple executables into a debugging session via the
3039 @code{add-inferior} and @w{@code{clone-inferior}} commands. On some
3040 systems @value{GDBN} can add inferiors to the debug session
3041 automatically by following calls to @code{fork} and @code{exec}. To
3042 remove inferiors from the debugging session use the
3043 @w{@code{remove-inferiors}} command.
3046 @kindex add-inferior
3047 @item add-inferior [ -copies @var{n} ] [ -exec @var{executable} ]
3048 Adds @var{n} inferiors to be run using @var{executable} as the
3049 executable; @var{n} defaults to 1. If no executable is specified,
3050 the inferiors begins empty, with no program. You can still assign or
3051 change the program assigned to the inferior at any time by using the
3052 @code{file} command with the executable name as its argument.
3054 @kindex clone-inferior
3055 @item clone-inferior [ -copies @var{n} ] [ @var{infno} ]
3056 Adds @var{n} inferiors ready to execute the same program as inferior
3057 @var{infno}; @var{n} defaults to 1, and @var{infno} defaults to the
3058 number of the current inferior. This is a convenient command when you
3059 want to run another instance of the inferior you are debugging.
3062 (@value{GDBP}) info inferiors
3063 Num Description Executable
3064 * 1 process 29964 helloworld
3065 (@value{GDBP}) clone-inferior
3068 (@value{GDBP}) info inferiors
3069 Num Description Executable
3071 * 1 process 29964 helloworld
3074 You can now simply switch focus to inferior 2 and run it.
3076 @kindex remove-inferiors
3077 @item remove-inferiors @var{infno}@dots{}
3078 Removes the inferior or inferiors @var{infno}@dots{}. It is not
3079 possible to remove an inferior that is running with this command. For
3080 those, use the @code{kill} or @code{detach} command first.
3084 To quit debugging one of the running inferiors that is not the current
3085 inferior, you can either detach from it by using the @w{@code{detach
3086 inferior}} command (allowing it to run independently), or kill it
3087 using the @w{@code{kill inferiors}} command:
3090 @kindex detach inferiors @var{infno}@dots{}
3091 @item detach inferior @var{infno}@dots{}
3092 Detach from the inferior or inferiors identified by @value{GDBN}
3093 inferior number(s) @var{infno}@dots{}. Note that the inferior's entry
3094 still stays on the list of inferiors shown by @code{info inferiors},
3095 but its Description will show @samp{<null>}.
3097 @kindex kill inferiors @var{infno}@dots{}
3098 @item kill inferiors @var{infno}@dots{}
3099 Kill the inferior or inferiors identified by @value{GDBN} inferior
3100 number(s) @var{infno}@dots{}. Note that the inferior's entry still
3101 stays on the list of inferiors shown by @code{info inferiors}, but its
3102 Description will show @samp{<null>}.
3105 After the successful completion of a command such as @code{detach},
3106 @code{detach inferiors}, @code{kill} or @code{kill inferiors}, or after
3107 a normal process exit, the inferior is still valid and listed with
3108 @code{info inferiors}, ready to be restarted.
3111 To be notified when inferiors are started or exit under @value{GDBN}'s
3112 control use @w{@code{set print inferior-events}}:
3115 @kindex set print inferior-events
3116 @cindex print messages on inferior start and exit
3117 @item set print inferior-events
3118 @itemx set print inferior-events on
3119 @itemx set print inferior-events off
3120 The @code{set print inferior-events} command allows you to enable or
3121 disable printing of messages when @value{GDBN} notices that new
3122 inferiors have started or that inferiors have exited or have been
3123 detached. By default, these messages will not be printed.
3125 @kindex show print inferior-events
3126 @item show print inferior-events
3127 Show whether messages will be printed when @value{GDBN} detects that
3128 inferiors have started, exited or have been detached.
3131 Many commands will work the same with multiple programs as with a
3132 single program: e.g., @code{print myglobal} will simply display the
3133 value of @code{myglobal} in the current inferior.
3136 Occasionally, when debugging @value{GDBN} itself, it may be useful to
3137 get more info about the relationship of inferiors, programs, address
3138 spaces in a debug session. You can do that with the @w{@code{maint
3139 info program-spaces}} command.
3142 @kindex maint info program-spaces
3143 @item maint info program-spaces
3144 Print a list of all program spaces currently being managed by
3147 @value{GDBN} displays for each program space (in this order):
3151 the program space number assigned by @value{GDBN}
3154 the name of the executable loaded into the program space, with e.g.,
3155 the @code{file} command.
3160 An asterisk @samp{*} preceding the @value{GDBN} program space number
3161 indicates the current program space.
3163 In addition, below each program space line, @value{GDBN} prints extra
3164 information that isn't suitable to display in tabular form. For
3165 example, the list of inferiors bound to the program space.
3168 (@value{GDBP}) maint info program-spaces
3172 Bound inferiors: ID 1 (process 21561)
3175 Here we can see that no inferior is running the program @code{hello},
3176 while @code{process 21561} is running the program @code{goodbye}. On
3177 some targets, it is possible that multiple inferiors are bound to the
3178 same program space. The most common example is that of debugging both
3179 the parent and child processes of a @code{vfork} call. For example,
3182 (@value{GDBP}) maint info program-spaces
3185 Bound inferiors: ID 2 (process 18050), ID 1 (process 18045)
3188 Here, both inferior 2 and inferior 1 are running in the same program
3189 space as a result of inferior 1 having executed a @code{vfork} call.
3193 @section Debugging Programs with Multiple Threads
3195 @cindex threads of execution
3196 @cindex multiple threads
3197 @cindex switching threads
3198 In some operating systems, such as GNU/Linux and Solaris, a single program
3199 may have more than one @dfn{thread} of execution. The precise semantics
3200 of threads differ from one operating system to another, but in general
3201 the threads of a single program are akin to multiple processes---except
3202 that they share one address space (that is, they can all examine and
3203 modify the same variables). On the other hand, each thread has its own
3204 registers and execution stack, and perhaps private memory.
3206 @value{GDBN} provides these facilities for debugging multi-thread
3210 @item automatic notification of new threads
3211 @item @samp{thread @var{thread-id}}, a command to switch among threads
3212 @item @samp{info threads}, a command to inquire about existing threads
3213 @item @samp{thread apply [@var{thread-id-list} | all] @var{args}},
3214 a command to apply a command to a list of threads
3215 @item thread-specific breakpoints
3216 @item @samp{set print thread-events}, which controls printing of
3217 messages on thread start and exit.
3218 @item @samp{set libthread-db-search-path @var{path}}, which lets
3219 the user specify which @code{libthread_db} to use if the default choice
3220 isn't compatible with the program.
3223 @cindex focus of debugging
3224 @cindex current thread
3225 The @value{GDBN} thread debugging facility allows you to observe all
3226 threads while your program runs---but whenever @value{GDBN} takes
3227 control, one thread in particular is always the focus of debugging.
3228 This thread is called the @dfn{current thread}. Debugging commands show
3229 program information from the perspective of the current thread.
3231 @cindex @code{New} @var{systag} message
3232 @cindex thread identifier (system)
3233 @c FIXME-implementors!! It would be more helpful if the [New...] message
3234 @c included GDB's numeric thread handle, so you could just go to that
3235 @c thread without first checking `info threads'.
3236 Whenever @value{GDBN} detects a new thread in your program, it displays
3237 the target system's identification for the thread with a message in the
3238 form @samp{[New @var{systag}]}, where @var{systag} is a thread identifier
3239 whose form varies depending on the particular system. For example, on
3240 @sc{gnu}/Linux, you might see
3243 [New Thread 0x41e02940 (LWP 25582)]
3247 when @value{GDBN} notices a new thread. In contrast, on other systems,
3248 the @var{systag} is simply something like @samp{process 368}, with no
3251 @c FIXME!! (1) Does the [New...] message appear even for the very first
3252 @c thread of a program, or does it only appear for the
3253 @c second---i.e.@: when it becomes obvious we have a multithread
3255 @c (2) *Is* there necessarily a first thread always? Or do some
3256 @c multithread systems permit starting a program with multiple
3257 @c threads ab initio?
3259 @anchor{thread numbers}
3260 @cindex thread number, per inferior
3261 @cindex thread identifier (GDB)
3262 For debugging purposes, @value{GDBN} associates its own thread number
3263 ---always a single integer---with each thread of an inferior. This
3264 number is unique between all threads of an inferior, but not unique
3265 between threads of different inferiors.
3267 @cindex qualified thread ID
3268 You can refer to a given thread in an inferior using the qualified
3269 @var{inferior-num}.@var{thread-num} syntax, also known as
3270 @dfn{qualified thread ID}, with @var{inferior-num} being the inferior
3271 number and @var{thread-num} being the thread number of the given
3272 inferior. For example, thread @code{2.3} refers to thread number 3 of
3273 inferior 2. If you omit @var{inferior-num} (e.g., @code{thread 3}),
3274 then @value{GDBN} infers you're referring to a thread of the current
3277 Until you create a second inferior, @value{GDBN} does not show the
3278 @var{inferior-num} part of thread IDs, even though you can always use
3279 the full @var{inferior-num}.@var{thread-num} form to refer to threads
3280 of inferior 1, the initial inferior.
3282 @anchor{thread ID lists}
3283 @cindex thread ID lists
3284 Some commands accept a space-separated @dfn{thread ID list} as
3285 argument. A list element can be:
3289 A thread ID as shown in the first field of the @samp{info threads}
3290 display, with or without an inferior qualifier. E.g., @samp{2.1} or
3294 A range of thread numbers, again with or without an inferior
3295 qualifier, as in @var{inf}.@var{thr1}-@var{thr2} or
3296 @var{thr1}-@var{thr2}. E.g., @samp{1.2-4} or @samp{2-4}.
3299 All threads of an inferior, specified with a star wildcard, with or
3300 without an inferior qualifier, as in @var{inf}.@code{*} (e.g.,
3301 @samp{1.*}) or @code{*}. The former refers to all threads of the
3302 given inferior, and the latter form without an inferior qualifier
3303 refers to all threads of the current inferior.
3307 For example, if the current inferior is 1, and inferior 7 has one
3308 thread with ID 7.1, the thread list @samp{1 2-3 4.5 6.7-9 7.*}
3309 includes threads 1 to 3 of inferior 1, thread 5 of inferior 4, threads
3310 7 to 9 of inferior 6 and all threads of inferior 7. That is, in
3311 expanded qualified form, the same as @samp{1.1 1.2 1.3 4.5 6.7 6.8 6.9
3315 @anchor{global thread numbers}
3316 @cindex global thread number
3317 @cindex global thread identifier (GDB)
3318 In addition to a @emph{per-inferior} number, each thread is also
3319 assigned a unique @emph{global} number, also known as @dfn{global
3320 thread ID}, a single integer. Unlike the thread number component of
3321 the thread ID, no two threads have the same global ID, even when
3322 you're debugging multiple inferiors.
3324 From @value{GDBN}'s perspective, a process always has at least one
3325 thread. In other words, @value{GDBN} assigns a thread number to the
3326 program's ``main thread'' even if the program is not multi-threaded.
3328 @vindex $_thread@r{, convenience variable}
3329 @vindex $_gthread@r{, convenience variable}
3330 The debugger convenience variables @samp{$_thread} and
3331 @samp{$_gthread} contain, respectively, the per-inferior thread number
3332 and the global thread number of the current thread. You may find this
3333 useful in writing breakpoint conditional expressions, command scripts,
3334 and so forth. @xref{Convenience Vars,, Convenience Variables}, for
3335 general information on convenience variables.
3337 If @value{GDBN} detects the program is multi-threaded, it augments the
3338 usual message about stopping at a breakpoint with the ID and name of
3339 the thread that hit the breakpoint.
3342 Thread 2 "client" hit Breakpoint 1, send_message () at client.c:68
3345 Likewise when the program receives a signal:
3348 Thread 1 "main" received signal SIGINT, Interrupt.
3352 @kindex info threads
3353 @item info threads @r{[}@var{thread-id-list}@r{]}
3355 Display information about one or more threads. With no arguments
3356 displays information about all threads. You can specify the list of
3357 threads that you want to display using the thread ID list syntax
3358 (@pxref{thread ID lists}).
3360 @value{GDBN} displays for each thread (in this order):
3364 the per-inferior thread number assigned by @value{GDBN}
3367 the global thread number assigned by @value{GDBN}, if the @samp{-gid}
3368 option was specified
3371 the target system's thread identifier (@var{systag})
3374 the thread's name, if one is known. A thread can either be named by
3375 the user (see @code{thread name}, below), or, in some cases, by the
3379 the current stack frame summary for that thread
3383 An asterisk @samp{*} to the left of the @value{GDBN} thread number
3384 indicates the current thread.
3388 @c end table here to get a little more width for example
3391 (@value{GDBP}) info threads
3393 * 1 process 35 thread 13 main (argc=1, argv=0x7ffffff8)
3394 2 process 35 thread 23 0x34e5 in sigpause ()
3395 3 process 35 thread 27 0x34e5 in sigpause ()
3399 If you're debugging multiple inferiors, @value{GDBN} displays thread
3400 IDs using the qualified @var{inferior-num}.@var{thread-num} format.
3401 Otherwise, only @var{thread-num} is shown.
3403 If you specify the @samp{-gid} option, @value{GDBN} displays a column
3404 indicating each thread's global thread ID:
3407 (@value{GDBP}) info threads
3408 Id GId Target Id Frame
3409 1.1 1 process 35 thread 13 main (argc=1, argv=0x7ffffff8)
3410 1.2 3 process 35 thread 23 0x34e5 in sigpause ()
3411 1.3 4 process 35 thread 27 0x34e5 in sigpause ()
3412 * 2.1 2 process 65 thread 1 main (argc=1, argv=0x7ffffff8)
3415 On Solaris, you can display more information about user threads with a
3416 Solaris-specific command:
3419 @item maint info sol-threads
3420 @kindex maint info sol-threads
3421 @cindex thread info (Solaris)
3422 Display info on Solaris user threads.
3426 @kindex thread @var{thread-id}
3427 @item thread @var{thread-id}
3428 Make thread ID @var{thread-id} the current thread. The command
3429 argument @var{thread-id} is the @value{GDBN} thread ID, as shown in
3430 the first field of the @samp{info threads} display, with or without an
3431 inferior qualifier (e.g., @samp{2.1} or @samp{1}).
3433 @value{GDBN} responds by displaying the system identifier of the
3434 thread you selected, and its current stack frame summary:
3437 (@value{GDBP}) thread 2
3438 [Switching to thread 2 (Thread 0xb7fdab70 (LWP 12747))]
3439 #0 some_function (ignore=0x0) at example.c:8
3440 8 printf ("hello\n");
3444 As with the @samp{[New @dots{}]} message, the form of the text after
3445 @samp{Switching to} depends on your system's conventions for identifying
3448 @anchor{thread apply all}
3449 @kindex thread apply
3450 @cindex apply command to several threads
3451 @item thread apply [@var{thread-id-list} | all [-ascending]] [@var{flag}]@dots{} @var{command}
3452 The @code{thread apply} command allows you to apply the named
3453 @var{command} to one or more threads. Specify the threads that you
3454 want affected using the thread ID list syntax (@pxref{thread ID
3455 lists}), or specify @code{all} to apply to all threads. To apply a
3456 command to all threads in descending order, type @kbd{thread apply all
3457 @var{command}}. To apply a command to all threads in ascending order,
3458 type @kbd{thread apply all -ascending @var{command}}.
3460 The @var{flag} arguments control what output to produce and how to handle
3461 errors raised when applying @var{command} to a thread. @var{flag}
3462 must start with a @code{-} directly followed by one letter in
3463 @code{qcs}. If several flags are provided, they must be given
3464 individually, such as @code{-c -q}.
3466 By default, @value{GDBN} displays some thread information before the
3467 output produced by @var{command}, and an error raised during the
3468 execution of a @var{command} will abort @code{thread apply}. The
3469 following flags can be used to fine-tune this behavior:
3473 The flag @code{-c}, which stands for @samp{continue}, causes any
3474 errors in @var{command} to be displayed, and the execution of
3475 @code{thread apply} then continues.
3477 The flag @code{-s}, which stands for @samp{silent}, causes any errors
3478 or empty output produced by a @var{command} to be silently ignored.
3479 That is, the execution continues, but the thread information and errors
3482 The flag @code{-q} (@samp{quiet}) disables printing the thread
3486 Flags @code{-c} and @code{-s} cannot be used together.
3489 @cindex apply command to all threads (ignoring errors and empty output)
3490 @item taas [@var{option}]@dots{} @var{command}
3491 Shortcut for @code{thread apply all -s [@var{option}]@dots{} @var{command}}.
3492 Applies @var{command} on all threads, ignoring errors and empty output.
3494 The @code{taas} command accepts the same options as the @code{thread
3495 apply all} command. @xref{thread apply all}.
3498 @cindex apply a command to all frames of all threads (ignoring errors and empty output)
3499 @item tfaas [@var{option}]@dots{} @var{command}
3500 Shortcut for @code{thread apply all -s -- frame apply all -s [@var{option}]@dots{} @var{command}}.
3501 Applies @var{command} on all frames of all threads, ignoring errors
3502 and empty output. Note that the flag @code{-s} is specified twice:
3503 The first @code{-s} ensures that @code{thread apply} only shows the thread
3504 information of the threads for which @code{frame apply} produces
3505 some output. The second @code{-s} is needed to ensure that @code{frame
3506 apply} shows the frame information of a frame only if the
3507 @var{command} successfully produced some output.
3509 It can for example be used to print a local variable or a function
3510 argument without knowing the thread or frame where this variable or argument
3513 (@value{GDBP}) tfaas p some_local_var_i_do_not_remember_where_it_is
3516 The @code{tfaas} command accepts the same options as the @code{frame
3517 apply} command. @xref{frame apply}.
3520 @cindex name a thread
3521 @item thread name [@var{name}]
3522 This command assigns a name to the current thread. If no argument is
3523 given, any existing user-specified name is removed. The thread name
3524 appears in the @samp{info threads} display.
3526 On some systems, such as @sc{gnu}/Linux, @value{GDBN} is able to
3527 determine the name of the thread as given by the OS. On these
3528 systems, a name specified with @samp{thread name} will override the
3529 system-give name, and removing the user-specified name will cause
3530 @value{GDBN} to once again display the system-specified name.
3533 @cindex search for a thread
3534 @item thread find [@var{regexp}]
3535 Search for and display thread ids whose name or @var{systag}
3536 matches the supplied regular expression.
3538 As well as being the complement to the @samp{thread name} command,
3539 this command also allows you to identify a thread by its target
3540 @var{systag}. For instance, on @sc{gnu}/Linux, the target @var{systag}
3544 (@value{GDBN}) thread find 26688
3545 Thread 4 has target id 'Thread 0x41e02940 (LWP 26688)'
3546 (@value{GDBN}) info thread 4
3548 4 Thread 0x41e02940 (LWP 26688) 0x00000031ca6cd372 in select ()
3551 @kindex set print thread-events
3552 @cindex print messages on thread start and exit
3553 @item set print thread-events
3554 @itemx set print thread-events on
3555 @itemx set print thread-events off
3556 The @code{set print thread-events} command allows you to enable or
3557 disable printing of messages when @value{GDBN} notices that new threads have
3558 started or that threads have exited. By default, these messages will
3559 be printed if detection of these events is supported by the target.
3560 Note that these messages cannot be disabled on all targets.
3562 @kindex show print thread-events
3563 @item show print thread-events
3564 Show whether messages will be printed when @value{GDBN} detects that threads
3565 have started and exited.
3568 @xref{Thread Stops,,Stopping and Starting Multi-thread Programs}, for
3569 more information about how @value{GDBN} behaves when you stop and start
3570 programs with multiple threads.
3572 @xref{Set Watchpoints,,Setting Watchpoints}, for information about
3573 watchpoints in programs with multiple threads.
3575 @anchor{set libthread-db-search-path}
3577 @kindex set libthread-db-search-path
3578 @cindex search path for @code{libthread_db}
3579 @item set libthread-db-search-path @r{[}@var{path}@r{]}
3580 If this variable is set, @var{path} is a colon-separated list of
3581 directories @value{GDBN} will use to search for @code{libthread_db}.
3582 If you omit @var{path}, @samp{libthread-db-search-path} will be reset to
3583 its default value (@code{$sdir:$pdir} on @sc{gnu}/Linux and Solaris systems).
3584 Internally, the default value comes from the @code{LIBTHREAD_DB_SEARCH_PATH}
3587 On @sc{gnu}/Linux and Solaris systems, @value{GDBN} uses a ``helper''
3588 @code{libthread_db} library to obtain information about threads in the
3589 inferior process. @value{GDBN} will use @samp{libthread-db-search-path}
3590 to find @code{libthread_db}. @value{GDBN} also consults first if inferior
3591 specific thread debugging library loading is enabled
3592 by @samp{set auto-load libthread-db} (@pxref{libthread_db.so.1 file}).
3594 A special entry @samp{$sdir} for @samp{libthread-db-search-path}
3595 refers to the default system directories that are
3596 normally searched for loading shared libraries. The @samp{$sdir} entry
3597 is the only kind not needing to be enabled by @samp{set auto-load libthread-db}
3598 (@pxref{libthread_db.so.1 file}).
3600 A special entry @samp{$pdir} for @samp{libthread-db-search-path}
3601 refers to the directory from which @code{libpthread}
3602 was loaded in the inferior process.
3604 For any @code{libthread_db} library @value{GDBN} finds in above directories,
3605 @value{GDBN} attempts to initialize it with the current inferior process.
3606 If this initialization fails (which could happen because of a version
3607 mismatch between @code{libthread_db} and @code{libpthread}), @value{GDBN}
3608 will unload @code{libthread_db}, and continue with the next directory.
3609 If none of @code{libthread_db} libraries initialize successfully,
3610 @value{GDBN} will issue a warning and thread debugging will be disabled.
3612 Setting @code{libthread-db-search-path} is currently implemented
3613 only on some platforms.
3615 @kindex show libthread-db-search-path
3616 @item show libthread-db-search-path
3617 Display current libthread_db search path.
3619 @kindex set debug libthread-db
3620 @kindex show debug libthread-db
3621 @cindex debugging @code{libthread_db}
3622 @item set debug libthread-db
3623 @itemx show debug libthread-db
3624 Turns on or off display of @code{libthread_db}-related events.
3625 Use @code{1} to enable, @code{0} to disable.
3629 @section Debugging Forks
3631 @cindex fork, debugging programs which call
3632 @cindex multiple processes
3633 @cindex processes, multiple
3634 On most systems, @value{GDBN} has no special support for debugging
3635 programs which create additional processes using the @code{fork}
3636 function. When a program forks, @value{GDBN} will continue to debug the
3637 parent process and the child process will run unimpeded. If you have
3638 set a breakpoint in any code which the child then executes, the child
3639 will get a @code{SIGTRAP} signal which (unless it catches the signal)
3640 will cause it to terminate.
3642 However, if you want to debug the child process there is a workaround
3643 which isn't too painful. Put a call to @code{sleep} in the code which
3644 the child process executes after the fork. It may be useful to sleep
3645 only if a certain environment variable is set, or a certain file exists,
3646 so that the delay need not occur when you don't want to run @value{GDBN}
3647 on the child. While the child is sleeping, use the @code{ps} program to
3648 get its process ID. Then tell @value{GDBN} (a new invocation of
3649 @value{GDBN} if you are also debugging the parent process) to attach to
3650 the child process (@pxref{Attach}). From that point on you can debug
3651 the child process just like any other process which you attached to.
3653 On some systems, @value{GDBN} provides support for debugging programs
3654 that create additional processes using the @code{fork} or @code{vfork}
3655 functions. On @sc{gnu}/Linux platforms, this feature is supported
3656 with kernel version 2.5.46 and later.
3658 The fork debugging commands are supported in native mode and when
3659 connected to @code{gdbserver} in either @code{target remote} mode or
3660 @code{target extended-remote} mode.
3662 By default, when a program forks, @value{GDBN} will continue to debug
3663 the parent process and the child process will run unimpeded.
3665 If you want to follow the child process instead of the parent process,
3666 use the command @w{@code{set follow-fork-mode}}.
3669 @kindex set follow-fork-mode
3670 @item set follow-fork-mode @var{mode}
3671 Set the debugger response to a program call of @code{fork} or
3672 @code{vfork}. A call to @code{fork} or @code{vfork} creates a new
3673 process. The @var{mode} argument can be:
3677 The original process is debugged after a fork. The child process runs
3678 unimpeded. This is the default.
3681 The new process is debugged after a fork. The parent process runs
3686 @kindex show follow-fork-mode
3687 @item show follow-fork-mode
3688 Display the current debugger response to a @code{fork} or @code{vfork} call.
3691 @cindex debugging multiple processes
3692 On Linux, if you want to debug both the parent and child processes, use the
3693 command @w{@code{set detach-on-fork}}.
3696 @kindex set detach-on-fork
3697 @item set detach-on-fork @var{mode}
3698 Tells gdb whether to detach one of the processes after a fork, or
3699 retain debugger control over them both.
3703 The child process (or parent process, depending on the value of
3704 @code{follow-fork-mode}) will be detached and allowed to run
3705 independently. This is the default.
3708 Both processes will be held under the control of @value{GDBN}.
3709 One process (child or parent, depending on the value of
3710 @code{follow-fork-mode}) is debugged as usual, while the other
3715 @kindex show detach-on-fork
3716 @item show detach-on-fork
3717 Show whether detach-on-fork mode is on/off.
3720 If you choose to set @samp{detach-on-fork} mode off, then @value{GDBN}
3721 will retain control of all forked processes (including nested forks).
3722 You can list the forked processes under the control of @value{GDBN} by
3723 using the @w{@code{info inferiors}} command, and switch from one fork
3724 to another by using the @code{inferior} command (@pxref{Inferiors and
3725 Programs, ,Debugging Multiple Inferiors and Programs}).
3727 To quit debugging one of the forked processes, you can either detach
3728 from it by using the @w{@code{detach inferiors}} command (allowing it
3729 to run independently), or kill it using the @w{@code{kill inferiors}}
3730 command. @xref{Inferiors and Programs, ,Debugging Multiple Inferiors
3733 If you ask to debug a child process and a @code{vfork} is followed by an
3734 @code{exec}, @value{GDBN} executes the new target up to the first
3735 breakpoint in the new target. If you have a breakpoint set on
3736 @code{main} in your original program, the breakpoint will also be set on
3737 the child process's @code{main}.
3739 On some systems, when a child process is spawned by @code{vfork}, you
3740 cannot debug the child or parent until an @code{exec} call completes.
3742 If you issue a @code{run} command to @value{GDBN} after an @code{exec}
3743 call executes, the new target restarts. To restart the parent
3744 process, use the @code{file} command with the parent executable name
3745 as its argument. By default, after an @code{exec} call executes,
3746 @value{GDBN} discards the symbols of the previous executable image.
3747 You can change this behaviour with the @w{@code{set follow-exec-mode}}
3751 @kindex set follow-exec-mode
3752 @item set follow-exec-mode @var{mode}
3754 Set debugger response to a program call of @code{exec}. An
3755 @code{exec} call replaces the program image of a process.
3757 @code{follow-exec-mode} can be:
3761 @value{GDBN} creates a new inferior and rebinds the process to this
3762 new inferior. The program the process was running before the
3763 @code{exec} call can be restarted afterwards by restarting the
3769 (@value{GDBP}) info inferiors
3771 Id Description Executable
3774 process 12020 is executing new program: prog2
3775 Program exited normally.
3776 (@value{GDBP}) info inferiors
3777 Id Description Executable
3783 @value{GDBN} keeps the process bound to the same inferior. The new
3784 executable image replaces the previous executable loaded in the
3785 inferior. Restarting the inferior after the @code{exec} call, with
3786 e.g., the @code{run} command, restarts the executable the process was
3787 running after the @code{exec} call. This is the default mode.
3792 (@value{GDBP}) info inferiors
3793 Id Description Executable
3796 process 12020 is executing new program: prog2
3797 Program exited normally.
3798 (@value{GDBP}) info inferiors
3799 Id Description Executable
3806 @code{follow-exec-mode} is supported in native mode and
3807 @code{target extended-remote} mode.
3809 You can use the @code{catch} command to make @value{GDBN} stop whenever
3810 a @code{fork}, @code{vfork}, or @code{exec} call is made. @xref{Set
3811 Catchpoints, ,Setting Catchpoints}.
3813 @node Checkpoint/Restart
3814 @section Setting a @emph{Bookmark} to Return to Later
3819 @cindex snapshot of a process
3820 @cindex rewind program state
3822 On certain operating systems@footnote{Currently, only
3823 @sc{gnu}/Linux.}, @value{GDBN} is able to save a @dfn{snapshot} of a
3824 program's state, called a @dfn{checkpoint}, and come back to it
3827 Returning to a checkpoint effectively undoes everything that has
3828 happened in the program since the @code{checkpoint} was saved. This
3829 includes changes in memory, registers, and even (within some limits)
3830 system state. Effectively, it is like going back in time to the
3831 moment when the checkpoint was saved.
3833 Thus, if you're stepping thru a program and you think you're
3834 getting close to the point where things go wrong, you can save
3835 a checkpoint. Then, if you accidentally go too far and miss
3836 the critical statement, instead of having to restart your program
3837 from the beginning, you can just go back to the checkpoint and
3838 start again from there.
3840 This can be especially useful if it takes a lot of time or
3841 steps to reach the point where you think the bug occurs.
3843 To use the @code{checkpoint}/@code{restart} method of debugging:
3848 Save a snapshot of the debugged program's current execution state.
3849 The @code{checkpoint} command takes no arguments, but each checkpoint
3850 is assigned a small integer id, similar to a breakpoint id.
3852 @kindex info checkpoints
3853 @item info checkpoints
3854 List the checkpoints that have been saved in the current debugging
3855 session. For each checkpoint, the following information will be
3862 @item Source line, or label
3865 @kindex restart @var{checkpoint-id}
3866 @item restart @var{checkpoint-id}
3867 Restore the program state that was saved as checkpoint number
3868 @var{checkpoint-id}. All program variables, registers, stack frames
3869 etc.@: will be returned to the values that they had when the checkpoint
3870 was saved. In essence, gdb will ``wind back the clock'' to the point
3871 in time when the checkpoint was saved.
3873 Note that breakpoints, @value{GDBN} variables, command history etc.
3874 are not affected by restoring a checkpoint. In general, a checkpoint
3875 only restores things that reside in the program being debugged, not in
3878 @kindex delete checkpoint @var{checkpoint-id}
3879 @item delete checkpoint @var{checkpoint-id}
3880 Delete the previously-saved checkpoint identified by @var{checkpoint-id}.
3884 Returning to a previously saved checkpoint will restore the user state
3885 of the program being debugged, plus a significant subset of the system
3886 (OS) state, including file pointers. It won't ``un-write'' data from
3887 a file, but it will rewind the file pointer to the previous location,
3888 so that the previously written data can be overwritten. For files
3889 opened in read mode, the pointer will also be restored so that the
3890 previously read data can be read again.
3892 Of course, characters that have been sent to a printer (or other
3893 external device) cannot be ``snatched back'', and characters received
3894 from eg.@: a serial device can be removed from internal program buffers,
3895 but they cannot be ``pushed back'' into the serial pipeline, ready to
3896 be received again. Similarly, the actual contents of files that have
3897 been changed cannot be restored (at this time).
3899 However, within those constraints, you actually can ``rewind'' your
3900 program to a previously saved point in time, and begin debugging it
3901 again --- and you can change the course of events so as to debug a
3902 different execution path this time.
3904 @cindex checkpoints and process id
3905 Finally, there is one bit of internal program state that will be
3906 different when you return to a checkpoint --- the program's process
3907 id. Each checkpoint will have a unique process id (or @var{pid}),
3908 and each will be different from the program's original @var{pid}.
3909 If your program has saved a local copy of its process id, this could
3910 potentially pose a problem.
3912 @subsection A Non-obvious Benefit of Using Checkpoints
3914 On some systems such as @sc{gnu}/Linux, address space randomization
3915 is performed on new processes for security reasons. This makes it
3916 difficult or impossible to set a breakpoint, or watchpoint, on an
3917 absolute address if you have to restart the program, since the
3918 absolute location of a symbol will change from one execution to the
3921 A checkpoint, however, is an @emph{identical} copy of a process.
3922 Therefore if you create a checkpoint at (eg.@:) the start of main,
3923 and simply return to that checkpoint instead of restarting the
3924 process, you can avoid the effects of address randomization and
3925 your symbols will all stay in the same place.
3928 @chapter Stopping and Continuing
3930 The principal purposes of using a debugger are so that you can stop your
3931 program before it terminates; or so that, if your program runs into
3932 trouble, you can investigate and find out why.
3934 Inside @value{GDBN}, your program may stop for any of several reasons,
3935 such as a signal, a breakpoint, or reaching a new line after a
3936 @value{GDBN} command such as @code{step}. You may then examine and
3937 change variables, set new breakpoints or remove old ones, and then
3938 continue execution. Usually, the messages shown by @value{GDBN} provide
3939 ample explanation of the status of your program---but you can also
3940 explicitly request this information at any time.
3943 @kindex info program
3945 Display information about the status of your program: whether it is
3946 running or not, what process it is, and why it stopped.
3950 * Breakpoints:: Breakpoints, watchpoints, and catchpoints
3951 * Continuing and Stepping:: Resuming execution
3952 * Skipping Over Functions and Files::
3953 Skipping over functions and files
3955 * Thread Stops:: Stopping and starting multi-thread programs
3959 @section Breakpoints, Watchpoints, and Catchpoints
3962 A @dfn{breakpoint} makes your program stop whenever a certain point in
3963 the program is reached. For each breakpoint, you can add conditions to
3964 control in finer detail whether your program stops. You can set
3965 breakpoints with the @code{break} command and its variants (@pxref{Set
3966 Breaks, ,Setting Breakpoints}), to specify the place where your program
3967 should stop by line number, function name or exact address in the
3970 On some systems, you can set breakpoints in shared libraries before
3971 the executable is run.
3974 @cindex data breakpoints
3975 @cindex memory tracing
3976 @cindex breakpoint on memory address
3977 @cindex breakpoint on variable modification
3978 A @dfn{watchpoint} is a special breakpoint that stops your program
3979 when the value of an expression changes. The expression may be a value
3980 of a variable, or it could involve values of one or more variables
3981 combined by operators, such as @samp{a + b}. This is sometimes called
3982 @dfn{data breakpoints}. You must use a different command to set
3983 watchpoints (@pxref{Set Watchpoints, ,Setting Watchpoints}), but aside
3984 from that, you can manage a watchpoint like any other breakpoint: you
3985 enable, disable, and delete both breakpoints and watchpoints using the
3988 You can arrange to have values from your program displayed automatically
3989 whenever @value{GDBN} stops at a breakpoint. @xref{Auto Display,,
3993 @cindex breakpoint on events
3994 A @dfn{catchpoint} is another special breakpoint that stops your program
3995 when a certain kind of event occurs, such as the throwing of a C@t{++}
3996 exception or the loading of a library. As with watchpoints, you use a
3997 different command to set a catchpoint (@pxref{Set Catchpoints, ,Setting
3998 Catchpoints}), but aside from that, you can manage a catchpoint like any
3999 other breakpoint. (To stop when your program receives a signal, use the
4000 @code{handle} command; see @ref{Signals, ,Signals}.)
4002 @cindex breakpoint numbers
4003 @cindex numbers for breakpoints
4004 @value{GDBN} assigns a number to each breakpoint, watchpoint, or
4005 catchpoint when you create it; these numbers are successive integers
4006 starting with one. In many of the commands for controlling various
4007 features of breakpoints you use the breakpoint number to say which
4008 breakpoint you want to change. Each breakpoint may be @dfn{enabled} or
4009 @dfn{disabled}; if disabled, it has no effect on your program until you
4012 @cindex breakpoint ranges
4013 @cindex breakpoint lists
4014 @cindex ranges of breakpoints
4015 @cindex lists of breakpoints
4016 Some @value{GDBN} commands accept a space-separated list of breakpoints
4017 on which to operate. A list element can be either a single breakpoint number,
4018 like @samp{5}, or a range of such numbers, like @samp{5-7}.
4019 When a breakpoint list is given to a command, all breakpoints in that list
4023 * Set Breaks:: Setting breakpoints
4024 * Set Watchpoints:: Setting watchpoints
4025 * Set Catchpoints:: Setting catchpoints
4026 * Delete Breaks:: Deleting breakpoints
4027 * Disabling:: Disabling breakpoints
4028 * Conditions:: Break conditions
4029 * Break Commands:: Breakpoint command lists
4030 * Dynamic Printf:: Dynamic printf
4031 * Save Breakpoints:: How to save breakpoints in a file
4032 * Static Probe Points:: Listing static probe points
4033 * Error in Breakpoints:: ``Cannot insert breakpoints''
4034 * Breakpoint-related Warnings:: ``Breakpoint address adjusted...''
4038 @subsection Setting Breakpoints
4040 @c FIXME LMB what does GDB do if no code on line of breakpt?
4041 @c consider in particular declaration with/without initialization.
4043 @c FIXME 2 is there stuff on this already? break at fun start, already init?
4046 @kindex b @r{(@code{break})}
4047 @vindex $bpnum@r{, convenience variable}
4048 @cindex latest breakpoint
4049 Breakpoints are set with the @code{break} command (abbreviated
4050 @code{b}). The debugger convenience variable @samp{$bpnum} records the
4051 number of the breakpoint you've set most recently; see @ref{Convenience
4052 Vars,, Convenience Variables}, for a discussion of what you can do with
4053 convenience variables.
4056 @item break @var{location}
4057 Set a breakpoint at the given @var{location}, which can specify a
4058 function name, a line number, or an address of an instruction.
4059 (@xref{Specify Location}, for a list of all the possible ways to
4060 specify a @var{location}.) The breakpoint will stop your program just
4061 before it executes any of the code in the specified @var{location}.
4063 When using source languages that permit overloading of symbols, such as
4064 C@t{++}, a function name may refer to more than one possible place to break.
4065 @xref{Ambiguous Expressions,,Ambiguous Expressions}, for a discussion of
4068 It is also possible to insert a breakpoint that will stop the program
4069 only if a specific thread (@pxref{Thread-Specific Breakpoints})
4070 or a specific task (@pxref{Ada Tasks}) hits that breakpoint.
4073 When called without any arguments, @code{break} sets a breakpoint at
4074 the next instruction to be executed in the selected stack frame
4075 (@pxref{Stack, ,Examining the Stack}). In any selected frame but the
4076 innermost, this makes your program stop as soon as control
4077 returns to that frame. This is similar to the effect of a
4078 @code{finish} command in the frame inside the selected frame---except
4079 that @code{finish} does not leave an active breakpoint. If you use
4080 @code{break} without an argument in the innermost frame, @value{GDBN} stops
4081 the next time it reaches the current location; this may be useful
4084 @value{GDBN} normally ignores breakpoints when it resumes execution, until at
4085 least one instruction has been executed. If it did not do this, you
4086 would be unable to proceed past a breakpoint without first disabling the
4087 breakpoint. This rule applies whether or not the breakpoint already
4088 existed when your program stopped.
4090 @item break @dots{} if @var{cond}
4091 Set a breakpoint with condition @var{cond}; evaluate the expression
4092 @var{cond} each time the breakpoint is reached, and stop only if the
4093 value is nonzero---that is, if @var{cond} evaluates as true.
4094 @samp{@dots{}} stands for one of the possible arguments described
4095 above (or no argument) specifying where to break. @xref{Conditions,
4096 ,Break Conditions}, for more information on breakpoint conditions.
4099 @item tbreak @var{args}
4100 Set a breakpoint enabled only for one stop. The @var{args} are the
4101 same as for the @code{break} command, and the breakpoint is set in the same
4102 way, but the breakpoint is automatically deleted after the first time your
4103 program stops there. @xref{Disabling, ,Disabling Breakpoints}.
4106 @cindex hardware breakpoints
4107 @item hbreak @var{args}
4108 Set a hardware-assisted breakpoint. The @var{args} are the same as for the
4109 @code{break} command and the breakpoint is set in the same way, but the
4110 breakpoint requires hardware support and some target hardware may not
4111 have this support. The main purpose of this is EPROM/ROM code
4112 debugging, so you can set a breakpoint at an instruction without
4113 changing the instruction. This can be used with the new trap-generation
4114 provided by SPARClite DSU and most x86-based targets. These targets
4115 will generate traps when a program accesses some data or instruction
4116 address that is assigned to the debug registers. However the hardware
4117 breakpoint registers can take a limited number of breakpoints. For
4118 example, on the DSU, only two data breakpoints can be set at a time, and
4119 @value{GDBN} will reject this command if more than two are used. Delete
4120 or disable unused hardware breakpoints before setting new ones
4121 (@pxref{Disabling, ,Disabling Breakpoints}).
4122 @xref{Conditions, ,Break Conditions}.
4123 For remote targets, you can restrict the number of hardware
4124 breakpoints @value{GDBN} will use, see @ref{set remote
4125 hardware-breakpoint-limit}.
4128 @item thbreak @var{args}
4129 Set a hardware-assisted breakpoint enabled only for one stop. The @var{args}
4130 are the same as for the @code{hbreak} command and the breakpoint is set in
4131 the same way. However, like the @code{tbreak} command,
4132 the breakpoint is automatically deleted after the
4133 first time your program stops there. Also, like the @code{hbreak}
4134 command, the breakpoint requires hardware support and some target hardware
4135 may not have this support. @xref{Disabling, ,Disabling Breakpoints}.
4136 See also @ref{Conditions, ,Break Conditions}.
4139 @cindex regular expression
4140 @cindex breakpoints at functions matching a regexp
4141 @cindex set breakpoints in many functions
4142 @item rbreak @var{regex}
4143 Set breakpoints on all functions matching the regular expression
4144 @var{regex}. This command sets an unconditional breakpoint on all
4145 matches, printing a list of all breakpoints it set. Once these
4146 breakpoints are set, they are treated just like the breakpoints set with
4147 the @code{break} command. You can delete them, disable them, or make
4148 them conditional the same way as any other breakpoint.
4150 In programs using different languages, @value{GDBN} chooses the syntax
4151 to print the list of all breakpoints it sets according to the
4152 @samp{set language} value: using @samp{set language auto}
4153 (see @ref{Automatically, ,Set Language Automatically}) means to use the
4154 language of the breakpoint's function, other values mean to use
4155 the manually specified language (see @ref{Manually, ,Set Language Manually}).
4157 The syntax of the regular expression is the standard one used with tools
4158 like @file{grep}. Note that this is different from the syntax used by
4159 shells, so for instance @code{foo*} matches all functions that include
4160 an @code{fo} followed by zero or more @code{o}s. There is an implicit
4161 @code{.*} leading and trailing the regular expression you supply, so to
4162 match only functions that begin with @code{foo}, use @code{^foo}.
4164 @cindex non-member C@t{++} functions, set breakpoint in
4165 When debugging C@t{++} programs, @code{rbreak} is useful for setting
4166 breakpoints on overloaded functions that are not members of any special
4169 @cindex set breakpoints on all functions
4170 The @code{rbreak} command can be used to set breakpoints in
4171 @strong{all} the functions in a program, like this:
4174 (@value{GDBP}) rbreak .
4177 @item rbreak @var{file}:@var{regex}
4178 If @code{rbreak} is called with a filename qualification, it limits
4179 the search for functions matching the given regular expression to the
4180 specified @var{file}. This can be used, for example, to set breakpoints on
4181 every function in a given file:
4184 (@value{GDBP}) rbreak file.c:.
4187 The colon separating the filename qualifier from the regex may
4188 optionally be surrounded by spaces.
4190 @kindex info breakpoints
4191 @cindex @code{$_} and @code{info breakpoints}
4192 @item info breakpoints @r{[}@var{list}@dots{}@r{]}
4193 @itemx info break @r{[}@var{list}@dots{}@r{]}
4194 Print a table of all breakpoints, watchpoints, and catchpoints set and
4195 not deleted. Optional argument @var{n} means print information only
4196 about the specified breakpoint(s) (or watchpoint(s) or catchpoint(s)).
4197 For each breakpoint, following columns are printed:
4200 @item Breakpoint Numbers
4202 Breakpoint, watchpoint, or catchpoint.
4204 Whether the breakpoint is marked to be disabled or deleted when hit.
4205 @item Enabled or Disabled
4206 Enabled breakpoints are marked with @samp{y}. @samp{n} marks breakpoints
4207 that are not enabled.
4209 Where the breakpoint is in your program, as a memory address. For a
4210 pending breakpoint whose address is not yet known, this field will
4211 contain @samp{<PENDING>}. Such breakpoint won't fire until a shared
4212 library that has the symbol or line referred by breakpoint is loaded.
4213 See below for details. A breakpoint with several locations will
4214 have @samp{<MULTIPLE>} in this field---see below for details.
4216 Where the breakpoint is in the source for your program, as a file and
4217 line number. For a pending breakpoint, the original string passed to
4218 the breakpoint command will be listed as it cannot be resolved until
4219 the appropriate shared library is loaded in the future.
4223 If a breakpoint is conditional, there are two evaluation modes: ``host'' and
4224 ``target''. If mode is ``host'', breakpoint condition evaluation is done by
4225 @value{GDBN} on the host's side. If it is ``target'', then the condition
4226 is evaluated by the target. The @code{info break} command shows
4227 the condition on the line following the affected breakpoint, together with
4228 its condition evaluation mode in between parentheses.
4230 Breakpoint commands, if any, are listed after that. A pending breakpoint is
4231 allowed to have a condition specified for it. The condition is not parsed for
4232 validity until a shared library is loaded that allows the pending
4233 breakpoint to resolve to a valid location.
4236 @code{info break} with a breakpoint
4237 number @var{n} as argument lists only that breakpoint. The
4238 convenience variable @code{$_} and the default examining-address for
4239 the @code{x} command are set to the address of the last breakpoint
4240 listed (@pxref{Memory, ,Examining Memory}).
4243 @code{info break} displays a count of the number of times the breakpoint
4244 has been hit. This is especially useful in conjunction with the
4245 @code{ignore} command. You can ignore a large number of breakpoint
4246 hits, look at the breakpoint info to see how many times the breakpoint
4247 was hit, and then run again, ignoring one less than that number. This
4248 will get you quickly to the last hit of that breakpoint.
4251 For a breakpoints with an enable count (xref) greater than 1,
4252 @code{info break} also displays that count.
4256 @value{GDBN} allows you to set any number of breakpoints at the same place in
4257 your program. There is nothing silly or meaningless about this. When
4258 the breakpoints are conditional, this is even useful
4259 (@pxref{Conditions, ,Break Conditions}).
4261 @cindex multiple locations, breakpoints
4262 @cindex breakpoints, multiple locations
4263 It is possible that a breakpoint corresponds to several locations
4264 in your program. Examples of this situation are:
4268 Multiple functions in the program may have the same name.
4271 For a C@t{++} constructor, the @value{NGCC} compiler generates several
4272 instances of the function body, used in different cases.
4275 For a C@t{++} template function, a given line in the function can
4276 correspond to any number of instantiations.
4279 For an inlined function, a given source line can correspond to
4280 several places where that function is inlined.
4283 In all those cases, @value{GDBN} will insert a breakpoint at all
4284 the relevant locations.
4286 A breakpoint with multiple locations is displayed in the breakpoint
4287 table using several rows---one header row, followed by one row for
4288 each breakpoint location. The header row has @samp{<MULTIPLE>} in the
4289 address column. The rows for individual locations contain the actual
4290 addresses for locations, and show the functions to which those
4291 locations belong. The number column for a location is of the form
4292 @var{breakpoint-number}.@var{location-number}.
4297 Num Type Disp Enb Address What
4298 1 breakpoint keep y <MULTIPLE>
4300 breakpoint already hit 1 time
4301 1.1 y 0x080486a2 in void foo<int>() at t.cc:8
4302 1.2 y 0x080486ca in void foo<double>() at t.cc:8
4305 You cannot delete the individual locations from a breakpoint. However,
4306 each location can be individually enabled or disabled by passing
4307 @var{breakpoint-number}.@var{location-number} as argument to the
4308 @code{enable} and @code{disable} commands. It's also possible to
4309 @code{enable} and @code{disable} a range of @var{location-number}
4310 locations using a @var{breakpoint-number} and two @var{location-number}s,
4311 in increasing order, separated by a hyphen, like
4312 @kbd{@var{breakpoint-number}.@var{location-number1}-@var{location-number2}},
4313 in which case @value{GDBN} acts on all the locations in the range (inclusive).
4314 Disabling or enabling the parent breakpoint (@pxref{Disabling}) affects
4315 all of the locations that belong to that breakpoint.
4317 @cindex pending breakpoints
4318 It's quite common to have a breakpoint inside a shared library.
4319 Shared libraries can be loaded and unloaded explicitly,
4320 and possibly repeatedly, as the program is executed. To support
4321 this use case, @value{GDBN} updates breakpoint locations whenever
4322 any shared library is loaded or unloaded. Typically, you would
4323 set a breakpoint in a shared library at the beginning of your
4324 debugging session, when the library is not loaded, and when the
4325 symbols from the library are not available. When you try to set
4326 breakpoint, @value{GDBN} will ask you if you want to set
4327 a so called @dfn{pending breakpoint}---breakpoint whose address
4328 is not yet resolved.
4330 After the program is run, whenever a new shared library is loaded,
4331 @value{GDBN} reevaluates all the breakpoints. When a newly loaded
4332 shared library contains the symbol or line referred to by some
4333 pending breakpoint, that breakpoint is resolved and becomes an
4334 ordinary breakpoint. When a library is unloaded, all breakpoints
4335 that refer to its symbols or source lines become pending again.
4337 This logic works for breakpoints with multiple locations, too. For
4338 example, if you have a breakpoint in a C@t{++} template function, and
4339 a newly loaded shared library has an instantiation of that template,
4340 a new location is added to the list of locations for the breakpoint.
4342 Except for having unresolved address, pending breakpoints do not
4343 differ from regular breakpoints. You can set conditions or commands,
4344 enable and disable them and perform other breakpoint operations.
4346 @value{GDBN} provides some additional commands for controlling what
4347 happens when the @samp{break} command cannot resolve breakpoint
4348 address specification to an address:
4350 @kindex set breakpoint pending
4351 @kindex show breakpoint pending
4353 @item set breakpoint pending auto
4354 This is the default behavior. When @value{GDBN} cannot find the breakpoint
4355 location, it queries you whether a pending breakpoint should be created.
4357 @item set breakpoint pending on
4358 This indicates that an unrecognized breakpoint location should automatically
4359 result in a pending breakpoint being created.
4361 @item set breakpoint pending off
4362 This indicates that pending breakpoints are not to be created. Any
4363 unrecognized breakpoint location results in an error. This setting does
4364 not affect any pending breakpoints previously created.
4366 @item show breakpoint pending
4367 Show the current behavior setting for creating pending breakpoints.
4370 The settings above only affect the @code{break} command and its
4371 variants. Once breakpoint is set, it will be automatically updated
4372 as shared libraries are loaded and unloaded.
4374 @cindex automatic hardware breakpoints
4375 For some targets, @value{GDBN} can automatically decide if hardware or
4376 software breakpoints should be used, depending on whether the
4377 breakpoint address is read-only or read-write. This applies to
4378 breakpoints set with the @code{break} command as well as to internal
4379 breakpoints set by commands like @code{next} and @code{finish}. For
4380 breakpoints set with @code{hbreak}, @value{GDBN} will always use hardware
4383 You can control this automatic behaviour with the following commands:
4385 @kindex set breakpoint auto-hw
4386 @kindex show breakpoint auto-hw
4388 @item set breakpoint auto-hw on
4389 This is the default behavior. When @value{GDBN} sets a breakpoint, it
4390 will try to use the target memory map to decide if software or hardware
4391 breakpoint must be used.
4393 @item set breakpoint auto-hw off
4394 This indicates @value{GDBN} should not automatically select breakpoint
4395 type. If the target provides a memory map, @value{GDBN} will warn when
4396 trying to set software breakpoint at a read-only address.
4399 @value{GDBN} normally implements breakpoints by replacing the program code
4400 at the breakpoint address with a special instruction, which, when
4401 executed, given control to the debugger. By default, the program
4402 code is so modified only when the program is resumed. As soon as
4403 the program stops, @value{GDBN} restores the original instructions. This
4404 behaviour guards against leaving breakpoints inserted in the
4405 target should gdb abrubptly disconnect. However, with slow remote
4406 targets, inserting and removing breakpoint can reduce the performance.
4407 This behavior can be controlled with the following commands::
4409 @kindex set breakpoint always-inserted
4410 @kindex show breakpoint always-inserted
4412 @item set breakpoint always-inserted off
4413 All breakpoints, including newly added by the user, are inserted in
4414 the target only when the target is resumed. All breakpoints are
4415 removed from the target when it stops. This is the default mode.
4417 @item set breakpoint always-inserted on
4418 Causes all breakpoints to be inserted in the target at all times. If
4419 the user adds a new breakpoint, or changes an existing breakpoint, the
4420 breakpoints in the target are updated immediately. A breakpoint is
4421 removed from the target only when breakpoint itself is deleted.
4424 @value{GDBN} handles conditional breakpoints by evaluating these conditions
4425 when a breakpoint breaks. If the condition is true, then the process being
4426 debugged stops, otherwise the process is resumed.
4428 If the target supports evaluating conditions on its end, @value{GDBN} may
4429 download the breakpoint, together with its conditions, to it.
4431 This feature can be controlled via the following commands:
4433 @kindex set breakpoint condition-evaluation
4434 @kindex show breakpoint condition-evaluation
4436 @item set breakpoint condition-evaluation host
4437 This option commands @value{GDBN} to evaluate the breakpoint
4438 conditions on the host's side. Unconditional breakpoints are sent to
4439 the target which in turn receives the triggers and reports them back to GDB
4440 for condition evaluation. This is the standard evaluation mode.
4442 @item set breakpoint condition-evaluation target
4443 This option commands @value{GDBN} to download breakpoint conditions
4444 to the target at the moment of their insertion. The target
4445 is responsible for evaluating the conditional expression and reporting
4446 breakpoint stop events back to @value{GDBN} whenever the condition
4447 is true. Due to limitations of target-side evaluation, some conditions
4448 cannot be evaluated there, e.g., conditions that depend on local data
4449 that is only known to the host. Examples include
4450 conditional expressions involving convenience variables, complex types
4451 that cannot be handled by the agent expression parser and expressions
4452 that are too long to be sent over to the target, specially when the
4453 target is a remote system. In these cases, the conditions will be
4454 evaluated by @value{GDBN}.
4456 @item set breakpoint condition-evaluation auto
4457 This is the default mode. If the target supports evaluating breakpoint
4458 conditions on its end, @value{GDBN} will download breakpoint conditions to
4459 the target (limitations mentioned previously apply). If the target does
4460 not support breakpoint condition evaluation, then @value{GDBN} will fallback
4461 to evaluating all these conditions on the host's side.
4465 @cindex negative breakpoint numbers
4466 @cindex internal @value{GDBN} breakpoints
4467 @value{GDBN} itself sometimes sets breakpoints in your program for
4468 special purposes, such as proper handling of @code{longjmp} (in C
4469 programs). These internal breakpoints are assigned negative numbers,
4470 starting with @code{-1}; @samp{info breakpoints} does not display them.
4471 You can see these breakpoints with the @value{GDBN} maintenance command
4472 @samp{maint info breakpoints} (@pxref{maint info breakpoints}).
4475 @node Set Watchpoints
4476 @subsection Setting Watchpoints
4478 @cindex setting watchpoints
4479 You can use a watchpoint to stop execution whenever the value of an
4480 expression changes, without having to predict a particular place where
4481 this may happen. (This is sometimes called a @dfn{data breakpoint}.)
4482 The expression may be as simple as the value of a single variable, or
4483 as complex as many variables combined by operators. Examples include:
4487 A reference to the value of a single variable.
4490 An address cast to an appropriate data type. For example,
4491 @samp{*(int *)0x12345678} will watch a 4-byte region at the specified
4492 address (assuming an @code{int} occupies 4 bytes).
4495 An arbitrarily complex expression, such as @samp{a*b + c/d}. The
4496 expression can use any operators valid in the program's native
4497 language (@pxref{Languages}).
4500 You can set a watchpoint on an expression even if the expression can
4501 not be evaluated yet. For instance, you can set a watchpoint on
4502 @samp{*global_ptr} before @samp{global_ptr} is initialized.
4503 @value{GDBN} will stop when your program sets @samp{global_ptr} and
4504 the expression produces a valid value. If the expression becomes
4505 valid in some other way than changing a variable (e.g.@: if the memory
4506 pointed to by @samp{*global_ptr} becomes readable as the result of a
4507 @code{malloc} call), @value{GDBN} may not stop until the next time
4508 the expression changes.
4510 @cindex software watchpoints
4511 @cindex hardware watchpoints
4512 Depending on your system, watchpoints may be implemented in software or
4513 hardware. @value{GDBN} does software watchpointing by single-stepping your
4514 program and testing the variable's value each time, which is hundreds of
4515 times slower than normal execution. (But this may still be worth it, to
4516 catch errors where you have no clue what part of your program is the
4519 On some systems, such as most PowerPC or x86-based targets,
4520 @value{GDBN} includes support for hardware watchpoints, which do not
4521 slow down the running of your program.
4525 @item watch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{thread-id}@r{]} @r{[}mask @var{maskvalue}@r{]}
4526 Set a watchpoint for an expression. @value{GDBN} will break when the
4527 expression @var{expr} is written into by the program and its value
4528 changes. The simplest (and the most popular) use of this command is
4529 to watch the value of a single variable:
4532 (@value{GDBP}) watch foo
4535 If the command includes a @code{@r{[}thread @var{thread-id}@r{]}}
4536 argument, @value{GDBN} breaks only when the thread identified by
4537 @var{thread-id} changes the value of @var{expr}. If any other threads
4538 change the value of @var{expr}, @value{GDBN} will not break. Note
4539 that watchpoints restricted to a single thread in this way only work
4540 with Hardware Watchpoints.
4542 Ordinarily a watchpoint respects the scope of variables in @var{expr}
4543 (see below). The @code{-location} argument tells @value{GDBN} to
4544 instead watch the memory referred to by @var{expr}. In this case,
4545 @value{GDBN} will evaluate @var{expr}, take the address of the result,
4546 and watch the memory at that address. The type of the result is used
4547 to determine the size of the watched memory. If the expression's
4548 result does not have an address, then @value{GDBN} will print an
4551 The @code{@r{[}mask @var{maskvalue}@r{]}} argument allows creation
4552 of masked watchpoints, if the current architecture supports this
4553 feature (e.g., PowerPC Embedded architecture, see @ref{PowerPC
4554 Embedded}.) A @dfn{masked watchpoint} specifies a mask in addition
4555 to an address to watch. The mask specifies that some bits of an address
4556 (the bits which are reset in the mask) should be ignored when matching
4557 the address accessed by the inferior against the watchpoint address.
4558 Thus, a masked watchpoint watches many addresses simultaneously---those
4559 addresses whose unmasked bits are identical to the unmasked bits in the
4560 watchpoint address. The @code{mask} argument implies @code{-location}.
4564 (@value{GDBP}) watch foo mask 0xffff00ff
4565 (@value{GDBP}) watch *0xdeadbeef mask 0xffffff00
4569 @item rwatch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{thread-id}@r{]} @r{[}mask @var{maskvalue}@r{]}
4570 Set a watchpoint that will break when the value of @var{expr} is read
4574 @item awatch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{thread-id}@r{]} @r{[}mask @var{maskvalue}@r{]}
4575 Set a watchpoint that will break when @var{expr} is either read from
4576 or written into by the program.
4578 @kindex info watchpoints @r{[}@var{list}@dots{}@r{]}
4579 @item info watchpoints @r{[}@var{list}@dots{}@r{]}
4580 This command prints a list of watchpoints, using the same format as
4581 @code{info break} (@pxref{Set Breaks}).
4584 If you watch for a change in a numerically entered address you need to
4585 dereference it, as the address itself is just a constant number which will
4586 never change. @value{GDBN} refuses to create a watchpoint that watches
4587 a never-changing value:
4590 (@value{GDBP}) watch 0x600850
4591 Cannot watch constant value 0x600850.
4592 (@value{GDBP}) watch *(int *) 0x600850
4593 Watchpoint 1: *(int *) 6293584
4596 @value{GDBN} sets a @dfn{hardware watchpoint} if possible. Hardware
4597 watchpoints execute very quickly, and the debugger reports a change in
4598 value at the exact instruction where the change occurs. If @value{GDBN}
4599 cannot set a hardware watchpoint, it sets a software watchpoint, which
4600 executes more slowly and reports the change in value at the next
4601 @emph{statement}, not the instruction, after the change occurs.
4603 @cindex use only software watchpoints
4604 You can force @value{GDBN} to use only software watchpoints with the
4605 @kbd{set can-use-hw-watchpoints 0} command. With this variable set to
4606 zero, @value{GDBN} will never try to use hardware watchpoints, even if
4607 the underlying system supports them. (Note that hardware-assisted
4608 watchpoints that were set @emph{before} setting
4609 @code{can-use-hw-watchpoints} to zero will still use the hardware
4610 mechanism of watching expression values.)
4613 @item set can-use-hw-watchpoints
4614 @kindex set can-use-hw-watchpoints
4615 Set whether or not to use hardware watchpoints.
4617 @item show can-use-hw-watchpoints
4618 @kindex show can-use-hw-watchpoints
4619 Show the current mode of using hardware watchpoints.
4622 For remote targets, you can restrict the number of hardware
4623 watchpoints @value{GDBN} will use, see @ref{set remote
4624 hardware-breakpoint-limit}.
4626 When you issue the @code{watch} command, @value{GDBN} reports
4629 Hardware watchpoint @var{num}: @var{expr}
4633 if it was able to set a hardware watchpoint.
4635 Currently, the @code{awatch} and @code{rwatch} commands can only set
4636 hardware watchpoints, because accesses to data that don't change the
4637 value of the watched expression cannot be detected without examining
4638 every instruction as it is being executed, and @value{GDBN} does not do
4639 that currently. If @value{GDBN} finds that it is unable to set a
4640 hardware breakpoint with the @code{awatch} or @code{rwatch} command, it
4641 will print a message like this:
4644 Expression cannot be implemented with read/access watchpoint.
4647 Sometimes, @value{GDBN} cannot set a hardware watchpoint because the
4648 data type of the watched expression is wider than what a hardware
4649 watchpoint on the target machine can handle. For example, some systems
4650 can only watch regions that are up to 4 bytes wide; on such systems you
4651 cannot set hardware watchpoints for an expression that yields a
4652 double-precision floating-point number (which is typically 8 bytes
4653 wide). As a work-around, it might be possible to break the large region
4654 into a series of smaller ones and watch them with separate watchpoints.
4656 If you set too many hardware watchpoints, @value{GDBN} might be unable
4657 to insert all of them when you resume the execution of your program.
4658 Since the precise number of active watchpoints is unknown until such
4659 time as the program is about to be resumed, @value{GDBN} might not be
4660 able to warn you about this when you set the watchpoints, and the
4661 warning will be printed only when the program is resumed:
4664 Hardware watchpoint @var{num}: Could not insert watchpoint
4668 If this happens, delete or disable some of the watchpoints.
4670 Watching complex expressions that reference many variables can also
4671 exhaust the resources available for hardware-assisted watchpoints.
4672 That's because @value{GDBN} needs to watch every variable in the
4673 expression with separately allocated resources.
4675 If you call a function interactively using @code{print} or @code{call},
4676 any watchpoints you have set will be inactive until @value{GDBN} reaches another
4677 kind of breakpoint or the call completes.
4679 @value{GDBN} automatically deletes watchpoints that watch local
4680 (automatic) variables, or expressions that involve such variables, when
4681 they go out of scope, that is, when the execution leaves the block in
4682 which these variables were defined. In particular, when the program
4683 being debugged terminates, @emph{all} local variables go out of scope,
4684 and so only watchpoints that watch global variables remain set. If you
4685 rerun the program, you will need to set all such watchpoints again. One
4686 way of doing that would be to set a code breakpoint at the entry to the
4687 @code{main} function and when it breaks, set all the watchpoints.
4689 @cindex watchpoints and threads
4690 @cindex threads and watchpoints
4691 In multi-threaded programs, watchpoints will detect changes to the
4692 watched expression from every thread.
4695 @emph{Warning:} In multi-threaded programs, software watchpoints
4696 have only limited usefulness. If @value{GDBN} creates a software
4697 watchpoint, it can only watch the value of an expression @emph{in a
4698 single thread}. If you are confident that the expression can only
4699 change due to the current thread's activity (and if you are also
4700 confident that no other thread can become current), then you can use
4701 software watchpoints as usual. However, @value{GDBN} may not notice
4702 when a non-current thread's activity changes the expression. (Hardware
4703 watchpoints, in contrast, watch an expression in all threads.)
4706 @xref{set remote hardware-watchpoint-limit}.
4708 @node Set Catchpoints
4709 @subsection Setting Catchpoints
4710 @cindex catchpoints, setting
4711 @cindex exception handlers
4712 @cindex event handling
4714 You can use @dfn{catchpoints} to cause the debugger to stop for certain
4715 kinds of program events, such as C@t{++} exceptions or the loading of a
4716 shared library. Use the @code{catch} command to set a catchpoint.
4720 @item catch @var{event}
4721 Stop when @var{event} occurs. The @var{event} can be any of the following:
4724 @item throw @r{[}@var{regexp}@r{]}
4725 @itemx rethrow @r{[}@var{regexp}@r{]}
4726 @itemx catch @r{[}@var{regexp}@r{]}
4728 @kindex catch rethrow
4730 @cindex stop on C@t{++} exceptions
4731 The throwing, re-throwing, or catching of a C@t{++} exception.
4733 If @var{regexp} is given, then only exceptions whose type matches the
4734 regular expression will be caught.
4736 @vindex $_exception@r{, convenience variable}
4737 The convenience variable @code{$_exception} is available at an
4738 exception-related catchpoint, on some systems. This holds the
4739 exception being thrown.
4741 There are currently some limitations to C@t{++} exception handling in
4746 The support for these commands is system-dependent. Currently, only
4747 systems using the @samp{gnu-v3} C@t{++} ABI (@pxref{ABI}) are
4751 The regular expression feature and the @code{$_exception} convenience
4752 variable rely on the presence of some SDT probes in @code{libstdc++}.
4753 If these probes are not present, then these features cannot be used.
4754 These probes were first available in the GCC 4.8 release, but whether
4755 or not they are available in your GCC also depends on how it was
4759 The @code{$_exception} convenience variable is only valid at the
4760 instruction at which an exception-related catchpoint is set.
4763 When an exception-related catchpoint is hit, @value{GDBN} stops at a
4764 location in the system library which implements runtime exception
4765 support for C@t{++}, usually @code{libstdc++}. You can use @code{up}
4766 (@pxref{Selection}) to get to your code.
4769 If you call a function interactively, @value{GDBN} normally returns
4770 control to you when the function has finished executing. If the call
4771 raises an exception, however, the call may bypass the mechanism that
4772 returns control to you and cause your program either to abort or to
4773 simply continue running until it hits a breakpoint, catches a signal
4774 that @value{GDBN} is listening for, or exits. This is the case even if
4775 you set a catchpoint for the exception; catchpoints on exceptions are
4776 disabled within interactive calls. @xref{Calling}, for information on
4777 controlling this with @code{set unwind-on-terminating-exception}.
4780 You cannot raise an exception interactively.
4783 You cannot install an exception handler interactively.
4786 @item exception @r{[}@var{name}@r{]}
4787 @kindex catch exception
4788 @cindex Ada exception catching
4789 @cindex catch Ada exceptions
4790 An Ada exception being raised. If an exception name is specified
4791 at the end of the command (eg @code{catch exception Program_Error}),
4792 the debugger will stop only when this specific exception is raised.
4793 Otherwise, the debugger stops execution when any Ada exception is raised.
4795 When inserting an exception catchpoint on a user-defined exception whose
4796 name is identical to one of the exceptions defined by the language, the
4797 fully qualified name must be used as the exception name. Otherwise,
4798 @value{GDBN} will assume that it should stop on the pre-defined exception
4799 rather than the user-defined one. For instance, assuming an exception
4800 called @code{Constraint_Error} is defined in package @code{Pck}, then
4801 the command to use to catch such exceptions is @kbd{catch exception
4802 Pck.Constraint_Error}.
4804 @vindex $_ada_exception@r{, convenience variable}
4805 The convenience variable @code{$_ada_exception} holds the address of
4806 the exception being thrown. This can be useful when setting a
4807 condition for such a catchpoint.
4809 @item exception unhandled
4810 @kindex catch exception unhandled
4811 An exception that was raised but is not handled by the program. The
4812 convenience variable @code{$_ada_exception} is set as for @code{catch
4815 @item handlers @r{[}@var{name}@r{]}
4816 @kindex catch handlers
4817 @cindex Ada exception handlers catching
4818 @cindex catch Ada exceptions when handled
4819 An Ada exception being handled. If an exception name is
4820 specified at the end of the command
4821 (eg @kbd{catch handlers Program_Error}), the debugger will stop
4822 only when this specific exception is handled.
4823 Otherwise, the debugger stops execution when any Ada exception is handled.
4825 When inserting a handlers catchpoint on a user-defined
4826 exception whose name is identical to one of the exceptions
4827 defined by the language, the fully qualified name must be used
4828 as the exception name. Otherwise, @value{GDBN} will assume that it
4829 should stop on the pre-defined exception rather than the
4830 user-defined one. For instance, assuming an exception called
4831 @code{Constraint_Error} is defined in package @code{Pck}, then the
4832 command to use to catch such exceptions handling is
4833 @kbd{catch handlers Pck.Constraint_Error}.
4835 The convenience variable @code{$_ada_exception} is set as for
4836 @code{catch exception}.
4839 @kindex catch assert
4840 A failed Ada assertion. Note that the convenience variable
4841 @code{$_ada_exception} is @emph{not} set by this catchpoint.
4845 @cindex break on fork/exec
4846 A call to @code{exec}.
4848 @anchor{catch syscall}
4850 @itemx syscall @r{[}@var{name} @r{|} @var{number} @r{|} @r{group:}@var{groupname} @r{|} @r{g:}@var{groupname}@r{]} @dots{}
4851 @kindex catch syscall
4852 @cindex break on a system call.
4853 A call to or return from a system call, a.k.a.@: @dfn{syscall}. A
4854 syscall is a mechanism for application programs to request a service
4855 from the operating system (OS) or one of the OS system services.
4856 @value{GDBN} can catch some or all of the syscalls issued by the
4857 debuggee, and show the related information for each syscall. If no
4858 argument is specified, calls to and returns from all system calls
4861 @var{name} can be any system call name that is valid for the
4862 underlying OS. Just what syscalls are valid depends on the OS. On
4863 GNU and Unix systems, you can find the full list of valid syscall
4864 names on @file{/usr/include/asm/unistd.h}.
4866 @c For MS-Windows, the syscall names and the corresponding numbers
4867 @c can be found, e.g., on this URL:
4868 @c http://www.metasploit.com/users/opcode/syscalls.html
4869 @c but we don't support Windows syscalls yet.
4871 Normally, @value{GDBN} knows in advance which syscalls are valid for
4872 each OS, so you can use the @value{GDBN} command-line completion
4873 facilities (@pxref{Completion,, command completion}) to list the
4876 You may also specify the system call numerically. A syscall's
4877 number is the value passed to the OS's syscall dispatcher to
4878 identify the requested service. When you specify the syscall by its
4879 name, @value{GDBN} uses its database of syscalls to convert the name
4880 into the corresponding numeric code, but using the number directly
4881 may be useful if @value{GDBN}'s database does not have the complete
4882 list of syscalls on your system (e.g., because @value{GDBN} lags
4883 behind the OS upgrades).
4885 You may specify a group of related syscalls to be caught at once using
4886 the @code{group:} syntax (@code{g:} is a shorter equivalent). For
4887 instance, on some platforms @value{GDBN} allows you to catch all
4888 network related syscalls, by passing the argument @code{group:network}
4889 to @code{catch syscall}. Note that not all syscall groups are
4890 available in every system. You can use the command completion
4891 facilities (@pxref{Completion,, command completion}) to list the
4892 syscall groups available on your environment.
4894 The example below illustrates how this command works if you don't provide
4898 (@value{GDBP}) catch syscall
4899 Catchpoint 1 (syscall)
4901 Starting program: /tmp/catch-syscall
4903 Catchpoint 1 (call to syscall 'close'), \
4904 0xffffe424 in __kernel_vsyscall ()
4908 Catchpoint 1 (returned from syscall 'close'), \
4909 0xffffe424 in __kernel_vsyscall ()
4913 Here is an example of catching a system call by name:
4916 (@value{GDBP}) catch syscall chroot
4917 Catchpoint 1 (syscall 'chroot' [61])
4919 Starting program: /tmp/catch-syscall
4921 Catchpoint 1 (call to syscall 'chroot'), \
4922 0xffffe424 in __kernel_vsyscall ()
4926 Catchpoint 1 (returned from syscall 'chroot'), \
4927 0xffffe424 in __kernel_vsyscall ()
4931 An example of specifying a system call numerically. In the case
4932 below, the syscall number has a corresponding entry in the XML
4933 file, so @value{GDBN} finds its name and prints it:
4936 (@value{GDBP}) catch syscall 252
4937 Catchpoint 1 (syscall(s) 'exit_group')
4939 Starting program: /tmp/catch-syscall
4941 Catchpoint 1 (call to syscall 'exit_group'), \
4942 0xffffe424 in __kernel_vsyscall ()
4946 Program exited normally.
4950 Here is an example of catching a syscall group:
4953 (@value{GDBP}) catch syscall group:process
4954 Catchpoint 1 (syscalls 'exit' [1] 'fork' [2] 'waitpid' [7]
4955 'execve' [11] 'wait4' [114] 'clone' [120] 'vfork' [190]
4956 'exit_group' [252] 'waitid' [284] 'unshare' [310])
4958 Starting program: /tmp/catch-syscall
4960 Catchpoint 1 (call to syscall fork), 0x00007ffff7df4e27 in open64 ()
4961 from /lib64/ld-linux-x86-64.so.2
4967 However, there can be situations when there is no corresponding name
4968 in XML file for that syscall number. In this case, @value{GDBN} prints
4969 a warning message saying that it was not able to find the syscall name,
4970 but the catchpoint will be set anyway. See the example below:
4973 (@value{GDBP}) catch syscall 764
4974 warning: The number '764' does not represent a known syscall.
4975 Catchpoint 2 (syscall 764)
4979 If you configure @value{GDBN} using the @samp{--without-expat} option,
4980 it will not be able to display syscall names. Also, if your
4981 architecture does not have an XML file describing its system calls,
4982 you will not be able to see the syscall names. It is important to
4983 notice that these two features are used for accessing the syscall
4984 name database. In either case, you will see a warning like this:
4987 (@value{GDBP}) catch syscall
4988 warning: Could not open "syscalls/i386-linux.xml"
4989 warning: Could not load the syscall XML file 'syscalls/i386-linux.xml'.
4990 GDB will not be able to display syscall names.
4991 Catchpoint 1 (syscall)
4995 Of course, the file name will change depending on your architecture and system.
4997 Still using the example above, you can also try to catch a syscall by its
4998 number. In this case, you would see something like:
5001 (@value{GDBP}) catch syscall 252
5002 Catchpoint 1 (syscall(s) 252)
5005 Again, in this case @value{GDBN} would not be able to display syscall's names.
5009 A call to @code{fork}.
5013 A call to @code{vfork}.
5015 @item load @r{[}@var{regexp}@r{]}
5016 @itemx unload @r{[}@var{regexp}@r{]}
5018 @kindex catch unload
5019 The loading or unloading of a shared library. If @var{regexp} is
5020 given, then the catchpoint will stop only if the regular expression
5021 matches one of the affected libraries.
5023 @item signal @r{[}@var{signal}@dots{} @r{|} @samp{all}@r{]}
5024 @kindex catch signal
5025 The delivery of a signal.
5027 With no arguments, this catchpoint will catch any signal that is not
5028 used internally by @value{GDBN}, specifically, all signals except
5029 @samp{SIGTRAP} and @samp{SIGINT}.
5031 With the argument @samp{all}, all signals, including those used by
5032 @value{GDBN}, will be caught. This argument cannot be used with other
5035 Otherwise, the arguments are a list of signal names as given to
5036 @code{handle} (@pxref{Signals}). Only signals specified in this list
5039 One reason that @code{catch signal} can be more useful than
5040 @code{handle} is that you can attach commands and conditions to the
5043 When a signal is caught by a catchpoint, the signal's @code{stop} and
5044 @code{print} settings, as specified by @code{handle}, are ignored.
5045 However, whether the signal is still delivered to the inferior depends
5046 on the @code{pass} setting; this can be changed in the catchpoint's
5051 @item tcatch @var{event}
5053 Set a catchpoint that is enabled only for one stop. The catchpoint is
5054 automatically deleted after the first time the event is caught.
5058 Use the @code{info break} command to list the current catchpoints.
5062 @subsection Deleting Breakpoints
5064 @cindex clearing breakpoints, watchpoints, catchpoints
5065 @cindex deleting breakpoints, watchpoints, catchpoints
5066 It is often necessary to eliminate a breakpoint, watchpoint, or
5067 catchpoint once it has done its job and you no longer want your program
5068 to stop there. This is called @dfn{deleting} the breakpoint. A
5069 breakpoint that has been deleted no longer exists; it is forgotten.
5071 With the @code{clear} command you can delete breakpoints according to
5072 where they are in your program. With the @code{delete} command you can
5073 delete individual breakpoints, watchpoints, or catchpoints by specifying
5074 their breakpoint numbers.
5076 It is not necessary to delete a breakpoint to proceed past it. @value{GDBN}
5077 automatically ignores breakpoints on the first instruction to be executed
5078 when you continue execution without changing the execution address.
5083 Delete any breakpoints at the next instruction to be executed in the
5084 selected stack frame (@pxref{Selection, ,Selecting a Frame}). When
5085 the innermost frame is selected, this is a good way to delete a
5086 breakpoint where your program just stopped.
5088 @item clear @var{location}
5089 Delete any breakpoints set at the specified @var{location}.
5090 @xref{Specify Location}, for the various forms of @var{location}; the
5091 most useful ones are listed below:
5094 @item clear @var{function}
5095 @itemx clear @var{filename}:@var{function}
5096 Delete any breakpoints set at entry to the named @var{function}.
5098 @item clear @var{linenum}
5099 @itemx clear @var{filename}:@var{linenum}
5100 Delete any breakpoints set at or within the code of the specified
5101 @var{linenum} of the specified @var{filename}.
5104 @cindex delete breakpoints
5106 @kindex d @r{(@code{delete})}
5107 @item delete @r{[}breakpoints@r{]} @r{[}@var{list}@dots{}@r{]}
5108 Delete the breakpoints, watchpoints, or catchpoints of the breakpoint
5109 list specified as argument. If no argument is specified, delete all
5110 breakpoints (@value{GDBN} asks confirmation, unless you have @code{set
5111 confirm off}). You can abbreviate this command as @code{d}.
5115 @subsection Disabling Breakpoints
5117 @cindex enable/disable a breakpoint
5118 Rather than deleting a breakpoint, watchpoint, or catchpoint, you might
5119 prefer to @dfn{disable} it. This makes the breakpoint inoperative as if
5120 it had been deleted, but remembers the information on the breakpoint so
5121 that you can @dfn{enable} it again later.
5123 You disable and enable breakpoints, watchpoints, and catchpoints with
5124 the @code{enable} and @code{disable} commands, optionally specifying
5125 one or more breakpoint numbers as arguments. Use @code{info break} to
5126 print a list of all breakpoints, watchpoints, and catchpoints if you
5127 do not know which numbers to use.
5129 Disabling and enabling a breakpoint that has multiple locations
5130 affects all of its locations.
5132 A breakpoint, watchpoint, or catchpoint can have any of several
5133 different states of enablement:
5137 Enabled. The breakpoint stops your program. A breakpoint set
5138 with the @code{break} command starts out in this state.
5140 Disabled. The breakpoint has no effect on your program.
5142 Enabled once. The breakpoint stops your program, but then becomes
5145 Enabled for a count. The breakpoint stops your program for the next
5146 N times, then becomes disabled.
5148 Enabled for deletion. The breakpoint stops your program, but
5149 immediately after it does so it is deleted permanently. A breakpoint
5150 set with the @code{tbreak} command starts out in this state.
5153 You can use the following commands to enable or disable breakpoints,
5154 watchpoints, and catchpoints:
5158 @kindex dis @r{(@code{disable})}
5159 @item disable @r{[}breakpoints@r{]} @r{[}@var{list}@dots{}@r{]}
5160 Disable the specified breakpoints---or all breakpoints, if none are
5161 listed. A disabled breakpoint has no effect but is not forgotten. All
5162 options such as ignore-counts, conditions and commands are remembered in
5163 case the breakpoint is enabled again later. You may abbreviate
5164 @code{disable} as @code{dis}.
5167 @item enable @r{[}breakpoints@r{]} @r{[}@var{list}@dots{}@r{]}
5168 Enable the specified breakpoints (or all defined breakpoints). They
5169 become effective once again in stopping your program.
5171 @item enable @r{[}breakpoints@r{]} once @var{list}@dots{}
5172 Enable the specified breakpoints temporarily. @value{GDBN} disables any
5173 of these breakpoints immediately after stopping your program.
5175 @item enable @r{[}breakpoints@r{]} count @var{count} @var{list}@dots{}
5176 Enable the specified breakpoints temporarily. @value{GDBN} records
5177 @var{count} with each of the specified breakpoints, and decrements a
5178 breakpoint's count when it is hit. When any count reaches 0,
5179 @value{GDBN} disables that breakpoint. If a breakpoint has an ignore
5180 count (@pxref{Conditions, ,Break Conditions}), that will be
5181 decremented to 0 before @var{count} is affected.
5183 @item enable @r{[}breakpoints@r{]} delete @var{list}@dots{}
5184 Enable the specified breakpoints to work once, then die. @value{GDBN}
5185 deletes any of these breakpoints as soon as your program stops there.
5186 Breakpoints set by the @code{tbreak} command start out in this state.
5189 @c FIXME: I think the following ``Except for [...] @code{tbreak}'' is
5190 @c confusing: tbreak is also initially enabled.
5191 Except for a breakpoint set with @code{tbreak} (@pxref{Set Breaks,
5192 ,Setting Breakpoints}), breakpoints that you set are initially enabled;
5193 subsequently, they become disabled or enabled only when you use one of
5194 the commands above. (The command @code{until} can set and delete a
5195 breakpoint of its own, but it does not change the state of your other
5196 breakpoints; see @ref{Continuing and Stepping, ,Continuing and
5200 @subsection Break Conditions
5201 @cindex conditional breakpoints
5202 @cindex breakpoint conditions
5204 @c FIXME what is scope of break condition expr? Context where wanted?
5205 @c in particular for a watchpoint?
5206 The simplest sort of breakpoint breaks every time your program reaches a
5207 specified place. You can also specify a @dfn{condition} for a
5208 breakpoint. A condition is just a Boolean expression in your
5209 programming language (@pxref{Expressions, ,Expressions}). A breakpoint with
5210 a condition evaluates the expression each time your program reaches it,
5211 and your program stops only if the condition is @emph{true}.
5213 This is the converse of using assertions for program validation; in that
5214 situation, you want to stop when the assertion is violated---that is,
5215 when the condition is false. In C, if you want to test an assertion expressed
5216 by the condition @var{assert}, you should set the condition
5217 @samp{! @var{assert}} on the appropriate breakpoint.
5219 Conditions are also accepted for watchpoints; you may not need them,
5220 since a watchpoint is inspecting the value of an expression anyhow---but
5221 it might be simpler, say, to just set a watchpoint on a variable name,
5222 and specify a condition that tests whether the new value is an interesting
5225 Break conditions can have side effects, and may even call functions in
5226 your program. This can be useful, for example, to activate functions
5227 that log program progress, or to use your own print functions to
5228 format special data structures. The effects are completely predictable
5229 unless there is another enabled breakpoint at the same address. (In
5230 that case, @value{GDBN} might see the other breakpoint first and stop your
5231 program without checking the condition of this one.) Note that
5232 breakpoint commands are usually more convenient and flexible than break
5234 purpose of performing side effects when a breakpoint is reached
5235 (@pxref{Break Commands, ,Breakpoint Command Lists}).
5237 Breakpoint conditions can also be evaluated on the target's side if
5238 the target supports it. Instead of evaluating the conditions locally,
5239 @value{GDBN} encodes the expression into an agent expression
5240 (@pxref{Agent Expressions}) suitable for execution on the target,
5241 independently of @value{GDBN}. Global variables become raw memory
5242 locations, locals become stack accesses, and so forth.
5244 In this case, @value{GDBN} will only be notified of a breakpoint trigger
5245 when its condition evaluates to true. This mechanism may provide faster
5246 response times depending on the performance characteristics of the target
5247 since it does not need to keep @value{GDBN} informed about
5248 every breakpoint trigger, even those with false conditions.
5250 Break conditions can be specified when a breakpoint is set, by using
5251 @samp{if} in the arguments to the @code{break} command. @xref{Set
5252 Breaks, ,Setting Breakpoints}. They can also be changed at any time
5253 with the @code{condition} command.
5255 You can also use the @code{if} keyword with the @code{watch} command.
5256 The @code{catch} command does not recognize the @code{if} keyword;
5257 @code{condition} is the only way to impose a further condition on a
5262 @item condition @var{bnum} @var{expression}
5263 Specify @var{expression} as the break condition for breakpoint,
5264 watchpoint, or catchpoint number @var{bnum}. After you set a condition,
5265 breakpoint @var{bnum} stops your program only if the value of
5266 @var{expression} is true (nonzero, in C). When you use
5267 @code{condition}, @value{GDBN} checks @var{expression} immediately for
5268 syntactic correctness, and to determine whether symbols in it have
5269 referents in the context of your breakpoint. If @var{expression} uses
5270 symbols not referenced in the context of the breakpoint, @value{GDBN}
5271 prints an error message:
5274 No symbol "foo" in current context.
5279 not actually evaluate @var{expression} at the time the @code{condition}
5280 command (or a command that sets a breakpoint with a condition, like
5281 @code{break if @dots{}}) is given, however. @xref{Expressions, ,Expressions}.
5283 @item condition @var{bnum}
5284 Remove the condition from breakpoint number @var{bnum}. It becomes
5285 an ordinary unconditional breakpoint.
5288 @cindex ignore count (of breakpoint)
5289 A special case of a breakpoint condition is to stop only when the
5290 breakpoint has been reached a certain number of times. This is so
5291 useful that there is a special way to do it, using the @dfn{ignore
5292 count} of the breakpoint. Every breakpoint has an ignore count, which
5293 is an integer. Most of the time, the ignore count is zero, and
5294 therefore has no effect. But if your program reaches a breakpoint whose
5295 ignore count is positive, then instead of stopping, it just decrements
5296 the ignore count by one and continues. As a result, if the ignore count
5297 value is @var{n}, the breakpoint does not stop the next @var{n} times
5298 your program reaches it.
5302 @item ignore @var{bnum} @var{count}
5303 Set the ignore count of breakpoint number @var{bnum} to @var{count}.
5304 The next @var{count} times the breakpoint is reached, your program's
5305 execution does not stop; other than to decrement the ignore count, @value{GDBN}
5308 To make the breakpoint stop the next time it is reached, specify
5311 When you use @code{continue} to resume execution of your program from a
5312 breakpoint, you can specify an ignore count directly as an argument to
5313 @code{continue}, rather than using @code{ignore}. @xref{Continuing and
5314 Stepping,,Continuing and Stepping}.
5316 If a breakpoint has a positive ignore count and a condition, the
5317 condition is not checked. Once the ignore count reaches zero,
5318 @value{GDBN} resumes checking the condition.
5320 You could achieve the effect of the ignore count with a condition such
5321 as @w{@samp{$foo-- <= 0}} using a debugger convenience variable that
5322 is decremented each time. @xref{Convenience Vars, ,Convenience
5326 Ignore counts apply to breakpoints, watchpoints, and catchpoints.
5329 @node Break Commands
5330 @subsection Breakpoint Command Lists
5332 @cindex breakpoint commands
5333 You can give any breakpoint (or watchpoint or catchpoint) a series of
5334 commands to execute when your program stops due to that breakpoint. For
5335 example, you might want to print the values of certain expressions, or
5336 enable other breakpoints.
5340 @kindex end@r{ (breakpoint commands)}
5341 @item commands @r{[}@var{list}@dots{}@r{]}
5342 @itemx @dots{} @var{command-list} @dots{}
5344 Specify a list of commands for the given breakpoints. The commands
5345 themselves appear on the following lines. Type a line containing just
5346 @code{end} to terminate the commands.
5348 To remove all commands from a breakpoint, type @code{commands} and
5349 follow it immediately with @code{end}; that is, give no commands.
5351 With no argument, @code{commands} refers to the last breakpoint,
5352 watchpoint, or catchpoint set (not to the breakpoint most recently
5353 encountered). If the most recent breakpoints were set with a single
5354 command, then the @code{commands} will apply to all the breakpoints
5355 set by that command. This applies to breakpoints set by
5356 @code{rbreak}, and also applies when a single @code{break} command
5357 creates multiple breakpoints (@pxref{Ambiguous Expressions,,Ambiguous
5361 Pressing @key{RET} as a means of repeating the last @value{GDBN} command is
5362 disabled within a @var{command-list}.
5364 You can use breakpoint commands to start your program up again. Simply
5365 use the @code{continue} command, or @code{step}, or any other command
5366 that resumes execution.
5368 Any other commands in the command list, after a command that resumes
5369 execution, are ignored. This is because any time you resume execution
5370 (even with a simple @code{next} or @code{step}), you may encounter
5371 another breakpoint---which could have its own command list, leading to
5372 ambiguities about which list to execute.
5375 If the first command you specify in a command list is @code{silent}, the
5376 usual message about stopping at a breakpoint is not printed. This may
5377 be desirable for breakpoints that are to print a specific message and
5378 then continue. If none of the remaining commands print anything, you
5379 see no sign that the breakpoint was reached. @code{silent} is
5380 meaningful only at the beginning of a breakpoint command list.
5382 The commands @code{echo}, @code{output}, and @code{printf} allow you to
5383 print precisely controlled output, and are often useful in silent
5384 breakpoints. @xref{Output, ,Commands for Controlled Output}.
5386 For example, here is how you could use breakpoint commands to print the
5387 value of @code{x} at entry to @code{foo} whenever @code{x} is positive.
5393 printf "x is %d\n",x
5398 One application for breakpoint commands is to compensate for one bug so
5399 you can test for another. Put a breakpoint just after the erroneous line
5400 of code, give it a condition to detect the case in which something
5401 erroneous has been done, and give it commands to assign correct values
5402 to any variables that need them. End with the @code{continue} command
5403 so that your program does not stop, and start with the @code{silent}
5404 command so that no output is produced. Here is an example:
5415 @node Dynamic Printf
5416 @subsection Dynamic Printf
5418 @cindex dynamic printf
5420 The dynamic printf command @code{dprintf} combines a breakpoint with
5421 formatted printing of your program's data to give you the effect of
5422 inserting @code{printf} calls into your program on-the-fly, without
5423 having to recompile it.
5425 In its most basic form, the output goes to the GDB console. However,
5426 you can set the variable @code{dprintf-style} for alternate handling.
5427 For instance, you can ask to format the output by calling your
5428 program's @code{printf} function. This has the advantage that the
5429 characters go to the program's output device, so they can recorded in
5430 redirects to files and so forth.
5432 If you are doing remote debugging with a stub or agent, you can also
5433 ask to have the printf handled by the remote agent. In addition to
5434 ensuring that the output goes to the remote program's device along
5435 with any other output the program might produce, you can also ask that
5436 the dprintf remain active even after disconnecting from the remote
5437 target. Using the stub/agent is also more efficient, as it can do
5438 everything without needing to communicate with @value{GDBN}.
5442 @item dprintf @var{location},@var{template},@var{expression}[,@var{expression}@dots{}]
5443 Whenever execution reaches @var{location}, print the values of one or
5444 more @var{expressions} under the control of the string @var{template}.
5445 To print several values, separate them with commas.
5447 @item set dprintf-style @var{style}
5448 Set the dprintf output to be handled in one of several different
5449 styles enumerated below. A change of style affects all existing
5450 dynamic printfs immediately. (If you need individual control over the
5451 print commands, simply define normal breakpoints with
5452 explicitly-supplied command lists.)
5456 @kindex dprintf-style gdb
5457 Handle the output using the @value{GDBN} @code{printf} command.
5460 @kindex dprintf-style call
5461 Handle the output by calling a function in your program (normally
5465 @kindex dprintf-style agent
5466 Have the remote debugging agent (such as @code{gdbserver}) handle
5467 the output itself. This style is only available for agents that
5468 support running commands on the target.
5471 @item set dprintf-function @var{function}
5472 Set the function to call if the dprintf style is @code{call}. By
5473 default its value is @code{printf}. You may set it to any expression.
5474 that @value{GDBN} can evaluate to a function, as per the @code{call}
5477 @item set dprintf-channel @var{channel}
5478 Set a ``channel'' for dprintf. If set to a non-empty value,
5479 @value{GDBN} will evaluate it as an expression and pass the result as
5480 a first argument to the @code{dprintf-function}, in the manner of
5481 @code{fprintf} and similar functions. Otherwise, the dprintf format
5482 string will be the first argument, in the manner of @code{printf}.
5484 As an example, if you wanted @code{dprintf} output to go to a logfile
5485 that is a standard I/O stream assigned to the variable @code{mylog},
5486 you could do the following:
5489 (gdb) set dprintf-style call
5490 (gdb) set dprintf-function fprintf
5491 (gdb) set dprintf-channel mylog
5492 (gdb) dprintf 25,"at line 25, glob=%d\n",glob
5493 Dprintf 1 at 0x123456: file main.c, line 25.
5495 1 dprintf keep y 0x00123456 in main at main.c:25
5496 call (void) fprintf (mylog,"at line 25, glob=%d\n",glob)
5501 Note that the @code{info break} displays the dynamic printf commands
5502 as normal breakpoint commands; you can thus easily see the effect of
5503 the variable settings.
5505 @item set disconnected-dprintf on
5506 @itemx set disconnected-dprintf off
5507 @kindex set disconnected-dprintf
5508 Choose whether @code{dprintf} commands should continue to run if
5509 @value{GDBN} has disconnected from the target. This only applies
5510 if the @code{dprintf-style} is @code{agent}.
5512 @item show disconnected-dprintf off
5513 @kindex show disconnected-dprintf
5514 Show the current choice for disconnected @code{dprintf}.
5518 @value{GDBN} does not check the validity of function and channel,
5519 relying on you to supply values that are meaningful for the contexts
5520 in which they are being used. For instance, the function and channel
5521 may be the values of local variables, but if that is the case, then
5522 all enabled dynamic prints must be at locations within the scope of
5523 those locals. If evaluation fails, @value{GDBN} will report an error.
5525 @node Save Breakpoints
5526 @subsection How to save breakpoints to a file
5528 To save breakpoint definitions to a file use the @w{@code{save
5529 breakpoints}} command.
5532 @kindex save breakpoints
5533 @cindex save breakpoints to a file for future sessions
5534 @item save breakpoints [@var{filename}]
5535 This command saves all current breakpoint definitions together with
5536 their commands and ignore counts, into a file @file{@var{filename}}
5537 suitable for use in a later debugging session. This includes all
5538 types of breakpoints (breakpoints, watchpoints, catchpoints,
5539 tracepoints). To read the saved breakpoint definitions, use the
5540 @code{source} command (@pxref{Command Files}). Note that watchpoints
5541 with expressions involving local variables may fail to be recreated
5542 because it may not be possible to access the context where the
5543 watchpoint is valid anymore. Because the saved breakpoint definitions
5544 are simply a sequence of @value{GDBN} commands that recreate the
5545 breakpoints, you can edit the file in your favorite editing program,
5546 and remove the breakpoint definitions you're not interested in, or
5547 that can no longer be recreated.
5550 @node Static Probe Points
5551 @subsection Static Probe Points
5553 @cindex static probe point, SystemTap
5554 @cindex static probe point, DTrace
5555 @value{GDBN} supports @dfn{SDT} probes in the code. @acronym{SDT} stands
5556 for Statically Defined Tracing, and the probes are designed to have a tiny
5557 runtime code and data footprint, and no dynamic relocations.
5559 Currently, the following types of probes are supported on
5560 ELF-compatible systems:
5564 @item @code{SystemTap} (@uref{http://sourceware.org/systemtap/})
5565 @acronym{SDT} probes@footnote{See
5566 @uref{http://sourceware.org/systemtap/wiki/AddingUserSpaceProbingToApps}
5567 for more information on how to add @code{SystemTap} @acronym{SDT}
5568 probes in your applications.}. @code{SystemTap} probes are usable
5569 from assembly, C and C@t{++} languages@footnote{See
5570 @uref{http://sourceware.org/systemtap/wiki/UserSpaceProbeImplementation}
5571 for a good reference on how the @acronym{SDT} probes are implemented.}.
5573 @item @code{DTrace} (@uref{http://oss.oracle.com/projects/DTrace})
5574 @acronym{USDT} probes. @code{DTrace} probes are usable from C and
5578 @cindex semaphores on static probe points
5579 Some @code{SystemTap} probes have an associated semaphore variable;
5580 for instance, this happens automatically if you defined your probe
5581 using a DTrace-style @file{.d} file. If your probe has a semaphore,
5582 @value{GDBN} will automatically enable it when you specify a
5583 breakpoint using the @samp{-probe-stap} notation. But, if you put a
5584 breakpoint at a probe's location by some other method (e.g.,
5585 @code{break file:line}), then @value{GDBN} will not automatically set
5586 the semaphore. @code{DTrace} probes do not support semaphores.
5588 You can examine the available static static probes using @code{info
5589 probes}, with optional arguments:
5593 @item info probes @r{[}@var{type}@r{]} @r{[}@var{provider} @r{[}@var{name} @r{[}@var{objfile}@r{]}@r{]}@r{]}
5594 If given, @var{type} is either @code{stap} for listing
5595 @code{SystemTap} probes or @code{dtrace} for listing @code{DTrace}
5596 probes. If omitted all probes are listed regardless of their types.
5598 If given, @var{provider} is a regular expression used to match against provider
5599 names when selecting which probes to list. If omitted, probes by all
5600 probes from all providers are listed.
5602 If given, @var{name} is a regular expression to match against probe names
5603 when selecting which probes to list. If omitted, probe names are not
5604 considered when deciding whether to display them.
5606 If given, @var{objfile} is a regular expression used to select which
5607 object files (executable or shared libraries) to examine. If not
5608 given, all object files are considered.
5610 @item info probes all
5611 List the available static probes, from all types.
5614 @cindex enabling and disabling probes
5615 Some probe points can be enabled and/or disabled. The effect of
5616 enabling or disabling a probe depends on the type of probe being
5617 handled. Some @code{DTrace} probes can be enabled or
5618 disabled, but @code{SystemTap} probes cannot be disabled.
5620 You can enable (or disable) one or more probes using the following
5621 commands, with optional arguments:
5624 @kindex enable probes
5625 @item enable probes @r{[}@var{provider} @r{[}@var{name} @r{[}@var{objfile}@r{]}@r{]}@r{]}
5626 If given, @var{provider} is a regular expression used to match against
5627 provider names when selecting which probes to enable. If omitted,
5628 all probes from all providers are enabled.
5630 If given, @var{name} is a regular expression to match against probe
5631 names when selecting which probes to enable. If omitted, probe names
5632 are not considered when deciding whether to enable them.
5634 If given, @var{objfile} is a regular expression used to select which
5635 object files (executable or shared libraries) to examine. If not
5636 given, all object files are considered.
5638 @kindex disable probes
5639 @item disable probes @r{[}@var{provider} @r{[}@var{name} @r{[}@var{objfile}@r{]}@r{]}@r{]}
5640 See the @code{enable probes} command above for a description of the
5641 optional arguments accepted by this command.
5644 @vindex $_probe_arg@r{, convenience variable}
5645 A probe may specify up to twelve arguments. These are available at the
5646 point at which the probe is defined---that is, when the current PC is
5647 at the probe's location. The arguments are available using the
5648 convenience variables (@pxref{Convenience Vars})
5649 @code{$_probe_arg0}@dots{}@code{$_probe_arg11}. In @code{SystemTap}
5650 probes each probe argument is an integer of the appropriate size;
5651 types are not preserved. In @code{DTrace} probes types are preserved
5652 provided that they are recognized as such by @value{GDBN}; otherwise
5653 the value of the probe argument will be a long integer. The
5654 convenience variable @code{$_probe_argc} holds the number of arguments
5655 at the current probe point.
5657 These variables are always available, but attempts to access them at
5658 any location other than a probe point will cause @value{GDBN} to give
5662 @c @ifclear BARETARGET
5663 @node Error in Breakpoints
5664 @subsection ``Cannot insert breakpoints''
5666 If you request too many active hardware-assisted breakpoints and
5667 watchpoints, you will see this error message:
5669 @c FIXME: the precise wording of this message may change; the relevant
5670 @c source change is not committed yet (Sep 3, 1999).
5672 Stopped; cannot insert breakpoints.
5673 You may have requested too many hardware breakpoints and watchpoints.
5677 This message is printed when you attempt to resume the program, since
5678 only then @value{GDBN} knows exactly how many hardware breakpoints and
5679 watchpoints it needs to insert.
5681 When this message is printed, you need to disable or remove some of the
5682 hardware-assisted breakpoints and watchpoints, and then continue.
5684 @node Breakpoint-related Warnings
5685 @subsection ``Breakpoint address adjusted...''
5686 @cindex breakpoint address adjusted
5688 Some processor architectures place constraints on the addresses at
5689 which breakpoints may be placed. For architectures thus constrained,
5690 @value{GDBN} will attempt to adjust the breakpoint's address to comply
5691 with the constraints dictated by the architecture.
5693 One example of such an architecture is the Fujitsu FR-V. The FR-V is
5694 a VLIW architecture in which a number of RISC-like instructions may be
5695 bundled together for parallel execution. The FR-V architecture
5696 constrains the location of a breakpoint instruction within such a
5697 bundle to the instruction with the lowest address. @value{GDBN}
5698 honors this constraint by adjusting a breakpoint's address to the
5699 first in the bundle.
5701 It is not uncommon for optimized code to have bundles which contain
5702 instructions from different source statements, thus it may happen that
5703 a breakpoint's address will be adjusted from one source statement to
5704 another. Since this adjustment may significantly alter @value{GDBN}'s
5705 breakpoint related behavior from what the user expects, a warning is
5706 printed when the breakpoint is first set and also when the breakpoint
5709 A warning like the one below is printed when setting a breakpoint
5710 that's been subject to address adjustment:
5713 warning: Breakpoint address adjusted from 0x00010414 to 0x00010410.
5716 Such warnings are printed both for user settable and @value{GDBN}'s
5717 internal breakpoints. If you see one of these warnings, you should
5718 verify that a breakpoint set at the adjusted address will have the
5719 desired affect. If not, the breakpoint in question may be removed and
5720 other breakpoints may be set which will have the desired behavior.
5721 E.g., it may be sufficient to place the breakpoint at a later
5722 instruction. A conditional breakpoint may also be useful in some
5723 cases to prevent the breakpoint from triggering too often.
5725 @value{GDBN} will also issue a warning when stopping at one of these
5726 adjusted breakpoints:
5729 warning: Breakpoint 1 address previously adjusted from 0x00010414
5733 When this warning is encountered, it may be too late to take remedial
5734 action except in cases where the breakpoint is hit earlier or more
5735 frequently than expected.
5737 @node Continuing and Stepping
5738 @section Continuing and Stepping
5742 @cindex resuming execution
5743 @dfn{Continuing} means resuming program execution until your program
5744 completes normally. In contrast, @dfn{stepping} means executing just
5745 one more ``step'' of your program, where ``step'' may mean either one
5746 line of source code, or one machine instruction (depending on what
5747 particular command you use). Either when continuing or when stepping,
5748 your program may stop even sooner, due to a breakpoint or a signal. (If
5749 it stops due to a signal, you may want to use @code{handle}, or use
5750 @samp{signal 0} to resume execution (@pxref{Signals, ,Signals}),
5751 or you may step into the signal's handler (@pxref{stepping and signal
5756 @kindex c @r{(@code{continue})}
5757 @kindex fg @r{(resume foreground execution)}
5758 @item continue @r{[}@var{ignore-count}@r{]}
5759 @itemx c @r{[}@var{ignore-count}@r{]}
5760 @itemx fg @r{[}@var{ignore-count}@r{]}
5761 Resume program execution, at the address where your program last stopped;
5762 any breakpoints set at that address are bypassed. The optional argument
5763 @var{ignore-count} allows you to specify a further number of times to
5764 ignore a breakpoint at this location; its effect is like that of
5765 @code{ignore} (@pxref{Conditions, ,Break Conditions}).
5767 The argument @var{ignore-count} is meaningful only when your program
5768 stopped due to a breakpoint. At other times, the argument to
5769 @code{continue} is ignored.
5771 The synonyms @code{c} and @code{fg} (for @dfn{foreground}, as the
5772 debugged program is deemed to be the foreground program) are provided
5773 purely for convenience, and have exactly the same behavior as
5777 To resume execution at a different place, you can use @code{return}
5778 (@pxref{Returning, ,Returning from a Function}) to go back to the
5779 calling function; or @code{jump} (@pxref{Jumping, ,Continuing at a
5780 Different Address}) to go to an arbitrary location in your program.
5782 A typical technique for using stepping is to set a breakpoint
5783 (@pxref{Breakpoints, ,Breakpoints; Watchpoints; and Catchpoints}) at the
5784 beginning of the function or the section of your program where a problem
5785 is believed to lie, run your program until it stops at that breakpoint,
5786 and then step through the suspect area, examining the variables that are
5787 interesting, until you see the problem happen.
5791 @kindex s @r{(@code{step})}
5793 Continue running your program until control reaches a different source
5794 line, then stop it and return control to @value{GDBN}. This command is
5795 abbreviated @code{s}.
5798 @c "without debugging information" is imprecise; actually "without line
5799 @c numbers in the debugging information". (gcc -g1 has debugging info but
5800 @c not line numbers). But it seems complex to try to make that
5801 @c distinction here.
5802 @emph{Warning:} If you use the @code{step} command while control is
5803 within a function that was compiled without debugging information,
5804 execution proceeds until control reaches a function that does have
5805 debugging information. Likewise, it will not step into a function which
5806 is compiled without debugging information. To step through functions
5807 without debugging information, use the @code{stepi} command, described
5811 The @code{step} command only stops at the first instruction of a source
5812 line. This prevents the multiple stops that could otherwise occur in
5813 @code{switch} statements, @code{for} loops, etc. @code{step} continues
5814 to stop if a function that has debugging information is called within
5815 the line. In other words, @code{step} @emph{steps inside} any functions
5816 called within the line.
5818 Also, the @code{step} command only enters a function if there is line
5819 number information for the function. Otherwise it acts like the
5820 @code{next} command. This avoids problems when using @code{cc -gl}
5821 on @acronym{MIPS} machines. Previously, @code{step} entered subroutines if there
5822 was any debugging information about the routine.
5824 @item step @var{count}
5825 Continue running as in @code{step}, but do so @var{count} times. If a
5826 breakpoint is reached, or a signal not related to stepping occurs before
5827 @var{count} steps, stepping stops right away.
5830 @kindex n @r{(@code{next})}
5831 @item next @r{[}@var{count}@r{]}
5832 Continue to the next source line in the current (innermost) stack frame.
5833 This is similar to @code{step}, but function calls that appear within
5834 the line of code are executed without stopping. Execution stops when
5835 control reaches a different line of code at the original stack level
5836 that was executing when you gave the @code{next} command. This command
5837 is abbreviated @code{n}.
5839 An argument @var{count} is a repeat count, as for @code{step}.
5842 @c FIX ME!! Do we delete this, or is there a way it fits in with
5843 @c the following paragraph? --- Vctoria
5845 @c @code{next} within a function that lacks debugging information acts like
5846 @c @code{step}, but any function calls appearing within the code of the
5847 @c function are executed without stopping.
5849 The @code{next} command only stops at the first instruction of a
5850 source line. This prevents multiple stops that could otherwise occur in
5851 @code{switch} statements, @code{for} loops, etc.
5853 @kindex set step-mode
5855 @cindex functions without line info, and stepping
5856 @cindex stepping into functions with no line info
5857 @itemx set step-mode on
5858 The @code{set step-mode on} command causes the @code{step} command to
5859 stop at the first instruction of a function which contains no debug line
5860 information rather than stepping over it.
5862 This is useful in cases where you may be interested in inspecting the
5863 machine instructions of a function which has no symbolic info and do not
5864 want @value{GDBN} to automatically skip over this function.
5866 @item set step-mode off
5867 Causes the @code{step} command to step over any functions which contains no
5868 debug information. This is the default.
5870 @item show step-mode
5871 Show whether @value{GDBN} will stop in or step over functions without
5872 source line debug information.
5875 @kindex fin @r{(@code{finish})}
5877 Continue running until just after function in the selected stack frame
5878 returns. Print the returned value (if any). This command can be
5879 abbreviated as @code{fin}.
5881 Contrast this with the @code{return} command (@pxref{Returning,
5882 ,Returning from a Function}).
5884 @kindex set print finish
5885 @kindex show print finish
5886 @item set print finish @r{[}on|off@r{]}
5887 @itemx show print finish
5888 By default the @code{finish} command will show the value that is
5889 returned by the function. This can be disabled using @code{set print
5890 finish off}. When disabled, the value is still entered into the value
5891 history (@pxref{Value History}), but not displayed.
5894 @kindex u @r{(@code{until})}
5895 @cindex run until specified location
5898 Continue running until a source line past the current line, in the
5899 current stack frame, is reached. This command is used to avoid single
5900 stepping through a loop more than once. It is like the @code{next}
5901 command, except that when @code{until} encounters a jump, it
5902 automatically continues execution until the program counter is greater
5903 than the address of the jump.
5905 This means that when you reach the end of a loop after single stepping
5906 though it, @code{until} makes your program continue execution until it
5907 exits the loop. In contrast, a @code{next} command at the end of a loop
5908 simply steps back to the beginning of the loop, which forces you to step
5909 through the next iteration.
5911 @code{until} always stops your program if it attempts to exit the current
5914 @code{until} may produce somewhat counterintuitive results if the order
5915 of machine code does not match the order of the source lines. For
5916 example, in the following excerpt from a debugging session, the @code{f}
5917 (@code{frame}) command shows that execution is stopped at line
5918 @code{206}; yet when we use @code{until}, we get to line @code{195}:
5922 #0 main (argc=4, argv=0xf7fffae8) at m4.c:206
5924 (@value{GDBP}) until
5925 195 for ( ; argc > 0; NEXTARG) @{
5928 This happened because, for execution efficiency, the compiler had
5929 generated code for the loop closure test at the end, rather than the
5930 start, of the loop---even though the test in a C @code{for}-loop is
5931 written before the body of the loop. The @code{until} command appeared
5932 to step back to the beginning of the loop when it advanced to this
5933 expression; however, it has not really gone to an earlier
5934 statement---not in terms of the actual machine code.
5936 @code{until} with no argument works by means of single
5937 instruction stepping, and hence is slower than @code{until} with an
5940 @item until @var{location}
5941 @itemx u @var{location}
5942 Continue running your program until either the specified @var{location} is
5943 reached, or the current stack frame returns. The location is any of
5944 the forms described in @ref{Specify Location}.
5945 This form of the command uses temporary breakpoints, and
5946 hence is quicker than @code{until} without an argument. The specified
5947 location is actually reached only if it is in the current frame. This
5948 implies that @code{until} can be used to skip over recursive function
5949 invocations. For instance in the code below, if the current location is
5950 line @code{96}, issuing @code{until 99} will execute the program up to
5951 line @code{99} in the same invocation of factorial, i.e., after the inner
5952 invocations have returned.
5955 94 int factorial (int value)
5957 96 if (value > 1) @{
5958 97 value *= factorial (value - 1);
5965 @kindex advance @var{location}
5966 @item advance @var{location}
5967 Continue running the program up to the given @var{location}. An argument is
5968 required, which should be of one of the forms described in
5969 @ref{Specify Location}.
5970 Execution will also stop upon exit from the current stack
5971 frame. This command is similar to @code{until}, but @code{advance} will
5972 not skip over recursive function calls, and the target location doesn't
5973 have to be in the same frame as the current one.
5977 @kindex si @r{(@code{stepi})}
5979 @itemx stepi @var{arg}
5981 Execute one machine instruction, then stop and return to the debugger.
5983 It is often useful to do @samp{display/i $pc} when stepping by machine
5984 instructions. This makes @value{GDBN} automatically display the next
5985 instruction to be executed, each time your program stops. @xref{Auto
5986 Display,, Automatic Display}.
5988 An argument is a repeat count, as in @code{step}.
5992 @kindex ni @r{(@code{nexti})}
5994 @itemx nexti @var{arg}
5996 Execute one machine instruction, but if it is a function call,
5997 proceed until the function returns.
5999 An argument is a repeat count, as in @code{next}.
6003 @anchor{range stepping}
6004 @cindex range stepping
6005 @cindex target-assisted range stepping
6006 By default, and if available, @value{GDBN} makes use of
6007 target-assisted @dfn{range stepping}. In other words, whenever you
6008 use a stepping command (e.g., @code{step}, @code{next}), @value{GDBN}
6009 tells the target to step the corresponding range of instruction
6010 addresses instead of issuing multiple single-steps. This speeds up
6011 line stepping, particularly for remote targets. Ideally, there should
6012 be no reason you would want to turn range stepping off. However, it's
6013 possible that a bug in the debug info, a bug in the remote stub (for
6014 remote targets), or even a bug in @value{GDBN} could make line
6015 stepping behave incorrectly when target-assisted range stepping is
6016 enabled. You can use the following command to turn off range stepping
6020 @kindex set range-stepping
6021 @kindex show range-stepping
6022 @item set range-stepping
6023 @itemx show range-stepping
6024 Control whether range stepping is enabled.
6026 If @code{on}, and the target supports it, @value{GDBN} tells the
6027 target to step a range of addresses itself, instead of issuing
6028 multiple single-steps. If @code{off}, @value{GDBN} always issues
6029 single-steps, even if range stepping is supported by the target. The
6030 default is @code{on}.
6034 @node Skipping Over Functions and Files
6035 @section Skipping Over Functions and Files
6036 @cindex skipping over functions and files
6038 The program you are debugging may contain some functions which are
6039 uninteresting to debug. The @code{skip} command lets you tell @value{GDBN} to
6040 skip a function, all functions in a file or a particular function in
6041 a particular file when stepping.
6043 For example, consider the following C function:
6054 Suppose you wish to step into the functions @code{foo} and @code{bar}, but you
6055 are not interested in stepping through @code{boring}. If you run @code{step}
6056 at line 103, you'll enter @code{boring()}, but if you run @code{next}, you'll
6057 step over both @code{foo} and @code{boring}!
6059 One solution is to @code{step} into @code{boring} and use the @code{finish}
6060 command to immediately exit it. But this can become tedious if @code{boring}
6061 is called from many places.
6063 A more flexible solution is to execute @kbd{skip boring}. This instructs
6064 @value{GDBN} never to step into @code{boring}. Now when you execute
6065 @code{step} at line 103, you'll step over @code{boring} and directly into
6068 Functions may be skipped by providing either a function name, linespec
6069 (@pxref{Specify Location}), regular expression that matches the function's
6070 name, file name or a @code{glob}-style pattern that matches the file name.
6072 On Posix systems the form of the regular expression is
6073 ``Extended Regular Expressions''. See for example @samp{man 7 regex}
6074 on @sc{gnu}/Linux systems. On non-Posix systems the form of the regular
6075 expression is whatever is provided by the @code{regcomp} function of
6076 the underlying system.
6077 See for example @samp{man 7 glob} on @sc{gnu}/Linux systems for a
6078 description of @code{glob}-style patterns.
6082 @item skip @r{[}@var{options}@r{]}
6083 The basic form of the @code{skip} command takes zero or more options
6084 that specify what to skip.
6085 The @var{options} argument is any useful combination of the following:
6088 @item -file @var{file}
6089 @itemx -fi @var{file}
6090 Functions in @var{file} will be skipped over when stepping.
6092 @item -gfile @var{file-glob-pattern}
6093 @itemx -gfi @var{file-glob-pattern}
6094 @cindex skipping over files via glob-style patterns
6095 Functions in files matching @var{file-glob-pattern} will be skipped
6099 (gdb) skip -gfi utils/*.c
6102 @item -function @var{linespec}
6103 @itemx -fu @var{linespec}
6104 Functions named by @var{linespec} or the function containing the line
6105 named by @var{linespec} will be skipped over when stepping.
6106 @xref{Specify Location}.
6108 @item -rfunction @var{regexp}
6109 @itemx -rfu @var{regexp}
6110 @cindex skipping over functions via regular expressions
6111 Functions whose name matches @var{regexp} will be skipped over when stepping.
6113 This form is useful for complex function names.
6114 For example, there is generally no need to step into C@t{++} @code{std::string}
6115 constructors or destructors. Plus with C@t{++} templates it can be hard to
6116 write out the full name of the function, and often it doesn't matter what
6117 the template arguments are. Specifying the function to be skipped as a
6118 regular expression makes this easier.
6121 (gdb) skip -rfu ^std::(allocator|basic_string)<.*>::~?\1 *\(
6124 If you want to skip every templated C@t{++} constructor and destructor
6125 in the @code{std} namespace you can do:
6128 (gdb) skip -rfu ^std::([a-zA-z0-9_]+)<.*>::~?\1 *\(
6132 If no options are specified, the function you're currently debugging
6135 @kindex skip function
6136 @item skip function @r{[}@var{linespec}@r{]}
6137 After running this command, the function named by @var{linespec} or the
6138 function containing the line named by @var{linespec} will be skipped over when
6139 stepping. @xref{Specify Location}.
6141 If you do not specify @var{linespec}, the function you're currently debugging
6144 (If you have a function called @code{file} that you want to skip, use
6145 @kbd{skip function file}.)
6148 @item skip file @r{[}@var{filename}@r{]}
6149 After running this command, any function whose source lives in @var{filename}
6150 will be skipped over when stepping.
6153 (gdb) skip file boring.c
6154 File boring.c will be skipped when stepping.
6157 If you do not specify @var{filename}, functions whose source lives in the file
6158 you're currently debugging will be skipped.
6161 Skips can be listed, deleted, disabled, and enabled, much like breakpoints.
6162 These are the commands for managing your list of skips:
6166 @item info skip @r{[}@var{range}@r{]}
6167 Print details about the specified skip(s). If @var{range} is not specified,
6168 print a table with details about all functions and files marked for skipping.
6169 @code{info skip} prints the following information about each skip:
6173 A number identifying this skip.
6174 @item Enabled or Disabled
6175 Enabled skips are marked with @samp{y}.
6176 Disabled skips are marked with @samp{n}.
6178 If the file name is a @samp{glob} pattern this is @samp{y}.
6179 Otherwise it is @samp{n}.
6181 The name or @samp{glob} pattern of the file to be skipped.
6182 If no file is specified this is @samp{<none>}.
6184 If the function name is a @samp{regular expression} this is @samp{y}.
6185 Otherwise it is @samp{n}.
6187 The name or regular expression of the function to skip.
6188 If no function is specified this is @samp{<none>}.
6192 @item skip delete @r{[}@var{range}@r{]}
6193 Delete the specified skip(s). If @var{range} is not specified, delete all
6197 @item skip enable @r{[}@var{range}@r{]}
6198 Enable the specified skip(s). If @var{range} is not specified, enable all
6201 @kindex skip disable
6202 @item skip disable @r{[}@var{range}@r{]}
6203 Disable the specified skip(s). If @var{range} is not specified, disable all
6206 @kindex set debug skip
6207 @item set debug skip @r{[}on|off@r{]}
6208 Set whether to print the debug output about skipping files and functions.
6210 @kindex show debug skip
6211 @item show debug skip
6212 Show whether the debug output about skipping files and functions is printed.
6220 A signal is an asynchronous event that can happen in a program. The
6221 operating system defines the possible kinds of signals, and gives each
6222 kind a name and a number. For example, in Unix @code{SIGINT} is the
6223 signal a program gets when you type an interrupt character (often @kbd{Ctrl-c});
6224 @code{SIGSEGV} is the signal a program gets from referencing a place in
6225 memory far away from all the areas in use; @code{SIGALRM} occurs when
6226 the alarm clock timer goes off (which happens only if your program has
6227 requested an alarm).
6229 @cindex fatal signals
6230 Some signals, including @code{SIGALRM}, are a normal part of the
6231 functioning of your program. Others, such as @code{SIGSEGV}, indicate
6232 errors; these signals are @dfn{fatal} (they kill your program immediately) if the
6233 program has not specified in advance some other way to handle the signal.
6234 @code{SIGINT} does not indicate an error in your program, but it is normally
6235 fatal so it can carry out the purpose of the interrupt: to kill the program.
6237 @value{GDBN} has the ability to detect any occurrence of a signal in your
6238 program. You can tell @value{GDBN} in advance what to do for each kind of
6241 @cindex handling signals
6242 Normally, @value{GDBN} is set up to let the non-erroneous signals like
6243 @code{SIGALRM} be silently passed to your program
6244 (so as not to interfere with their role in the program's functioning)
6245 but to stop your program immediately whenever an error signal happens.
6246 You can change these settings with the @code{handle} command.
6249 @kindex info signals
6253 Print a table of all the kinds of signals and how @value{GDBN} has been told to
6254 handle each one. You can use this to see the signal numbers of all
6255 the defined types of signals.
6257 @item info signals @var{sig}
6258 Similar, but print information only about the specified signal number.
6260 @code{info handle} is an alias for @code{info signals}.
6262 @item catch signal @r{[}@var{signal}@dots{} @r{|} @samp{all}@r{]}
6263 Set a catchpoint for the indicated signals. @xref{Set Catchpoints},
6264 for details about this command.
6267 @item handle @var{signal} @r{[}@var{keywords}@dots{}@r{]}
6268 Change the way @value{GDBN} handles signal @var{signal}. The @var{signal}
6269 can be the number of a signal or its name (with or without the
6270 @samp{SIG} at the beginning); a list of signal numbers of the form
6271 @samp{@var{low}-@var{high}}; or the word @samp{all}, meaning all the
6272 known signals. Optional arguments @var{keywords}, described below,
6273 say what change to make.
6277 The keywords allowed by the @code{handle} command can be abbreviated.
6278 Their full names are:
6282 @value{GDBN} should not stop your program when this signal happens. It may
6283 still print a message telling you that the signal has come in.
6286 @value{GDBN} should stop your program when this signal happens. This implies
6287 the @code{print} keyword as well.
6290 @value{GDBN} should print a message when this signal happens.
6293 @value{GDBN} should not mention the occurrence of the signal at all. This
6294 implies the @code{nostop} keyword as well.
6298 @value{GDBN} should allow your program to see this signal; your program
6299 can handle the signal, or else it may terminate if the signal is fatal
6300 and not handled. @code{pass} and @code{noignore} are synonyms.
6304 @value{GDBN} should not allow your program to see this signal.
6305 @code{nopass} and @code{ignore} are synonyms.
6309 When a signal stops your program, the signal is not visible to the
6311 continue. Your program sees the signal then, if @code{pass} is in
6312 effect for the signal in question @emph{at that time}. In other words,
6313 after @value{GDBN} reports a signal, you can use the @code{handle}
6314 command with @code{pass} or @code{nopass} to control whether your
6315 program sees that signal when you continue.
6317 The default is set to @code{nostop}, @code{noprint}, @code{pass} for
6318 non-erroneous signals such as @code{SIGALRM}, @code{SIGWINCH} and
6319 @code{SIGCHLD}, and to @code{stop}, @code{print}, @code{pass} for the
6322 You can also use the @code{signal} command to prevent your program from
6323 seeing a signal, or cause it to see a signal it normally would not see,
6324 or to give it any signal at any time. For example, if your program stopped
6325 due to some sort of memory reference error, you might store correct
6326 values into the erroneous variables and continue, hoping to see more
6327 execution; but your program would probably terminate immediately as
6328 a result of the fatal signal once it saw the signal. To prevent this,
6329 you can continue with @samp{signal 0}. @xref{Signaling, ,Giving your
6332 @cindex stepping and signal handlers
6333 @anchor{stepping and signal handlers}
6335 @value{GDBN} optimizes for stepping the mainline code. If a signal
6336 that has @code{handle nostop} and @code{handle pass} set arrives while
6337 a stepping command (e.g., @code{stepi}, @code{step}, @code{next}) is
6338 in progress, @value{GDBN} lets the signal handler run and then resumes
6339 stepping the mainline code once the signal handler returns. In other
6340 words, @value{GDBN} steps over the signal handler. This prevents
6341 signals that you've specified as not interesting (with @code{handle
6342 nostop}) from changing the focus of debugging unexpectedly. Note that
6343 the signal handler itself may still hit a breakpoint, stop for another
6344 signal that has @code{handle stop} in effect, or for any other event
6345 that normally results in stopping the stepping command sooner. Also
6346 note that @value{GDBN} still informs you that the program received a
6347 signal if @code{handle print} is set.
6349 @anchor{stepping into signal handlers}
6351 If you set @code{handle pass} for a signal, and your program sets up a
6352 handler for it, then issuing a stepping command, such as @code{step}
6353 or @code{stepi}, when your program is stopped due to the signal will
6354 step @emph{into} the signal handler (if the target supports that).
6356 Likewise, if you use the @code{queue-signal} command to queue a signal
6357 to be delivered to the current thread when execution of the thread
6358 resumes (@pxref{Signaling, ,Giving your Program a Signal}), then a
6359 stepping command will step into the signal handler.
6361 Here's an example, using @code{stepi} to step to the first instruction
6362 of @code{SIGUSR1}'s handler:
6365 (@value{GDBP}) handle SIGUSR1
6366 Signal Stop Print Pass to program Description
6367 SIGUSR1 Yes Yes Yes User defined signal 1
6371 Program received signal SIGUSR1, User defined signal 1.
6372 main () sigusr1.c:28
6375 sigusr1_handler () at sigusr1.c:9
6379 The same, but using @code{queue-signal} instead of waiting for the
6380 program to receive the signal first:
6385 (@value{GDBP}) queue-signal SIGUSR1
6387 sigusr1_handler () at sigusr1.c:9
6392 @cindex extra signal information
6393 @anchor{extra signal information}
6395 On some targets, @value{GDBN} can inspect extra signal information
6396 associated with the intercepted signal, before it is actually
6397 delivered to the program being debugged. This information is exported
6398 by the convenience variable @code{$_siginfo}, and consists of data
6399 that is passed by the kernel to the signal handler at the time of the
6400 receipt of a signal. The data type of the information itself is
6401 target dependent. You can see the data type using the @code{ptype
6402 $_siginfo} command. On Unix systems, it typically corresponds to the
6403 standard @code{siginfo_t} type, as defined in the @file{signal.h}
6406 Here's an example, on a @sc{gnu}/Linux system, printing the stray
6407 referenced address that raised a segmentation fault.
6411 (@value{GDBP}) continue
6412 Program received signal SIGSEGV, Segmentation fault.
6413 0x0000000000400766 in main ()
6415 (@value{GDBP}) ptype $_siginfo
6422 struct @{...@} _kill;
6423 struct @{...@} _timer;
6425 struct @{...@} _sigchld;
6426 struct @{...@} _sigfault;
6427 struct @{...@} _sigpoll;
6430 (@value{GDBP}) ptype $_siginfo._sifields._sigfault
6434 (@value{GDBP}) p $_siginfo._sifields._sigfault.si_addr
6435 $1 = (void *) 0x7ffff7ff7000
6439 Depending on target support, @code{$_siginfo} may also be writable.
6441 @cindex Intel MPX boundary violations
6442 @cindex boundary violations, Intel MPX
6443 On some targets, a @code{SIGSEGV} can be caused by a boundary
6444 violation, i.e., accessing an address outside of the allowed range.
6445 In those cases @value{GDBN} may displays additional information,
6446 depending on how @value{GDBN} has been told to handle the signal.
6447 With @code{handle stop SIGSEGV}, @value{GDBN} displays the violation
6448 kind: "Upper" or "Lower", the memory address accessed and the
6449 bounds, while with @code{handle nostop SIGSEGV} no additional
6450 information is displayed.
6452 The usual output of a segfault is:
6454 Program received signal SIGSEGV, Segmentation fault
6455 0x0000000000400d7c in upper () at i386-mpx-sigsegv.c:68
6456 68 value = *(p + len);
6459 While a bound violation is presented as:
6461 Program received signal SIGSEGV, Segmentation fault
6462 Upper bound violation while accessing address 0x7fffffffc3b3
6463 Bounds: [lower = 0x7fffffffc390, upper = 0x7fffffffc3a3]
6464 0x0000000000400d7c in upper () at i386-mpx-sigsegv.c:68
6465 68 value = *(p + len);
6469 @section Stopping and Starting Multi-thread Programs
6471 @cindex stopped threads
6472 @cindex threads, stopped
6474 @cindex continuing threads
6475 @cindex threads, continuing
6477 @value{GDBN} supports debugging programs with multiple threads
6478 (@pxref{Threads,, Debugging Programs with Multiple Threads}). There
6479 are two modes of controlling execution of your program within the
6480 debugger. In the default mode, referred to as @dfn{all-stop mode},
6481 when any thread in your program stops (for example, at a breakpoint
6482 or while being stepped), all other threads in the program are also stopped by
6483 @value{GDBN}. On some targets, @value{GDBN} also supports
6484 @dfn{non-stop mode}, in which other threads can continue to run freely while
6485 you examine the stopped thread in the debugger.
6488 * All-Stop Mode:: All threads stop when GDB takes control
6489 * Non-Stop Mode:: Other threads continue to execute
6490 * Background Execution:: Running your program asynchronously
6491 * Thread-Specific Breakpoints:: Controlling breakpoints
6492 * Interrupted System Calls:: GDB may interfere with system calls
6493 * Observer Mode:: GDB does not alter program behavior
6497 @subsection All-Stop Mode
6499 @cindex all-stop mode
6501 In all-stop mode, whenever your program stops under @value{GDBN} for any reason,
6502 @emph{all} threads of execution stop, not just the current thread. This
6503 allows you to examine the overall state of the program, including
6504 switching between threads, without worrying that things may change
6507 Conversely, whenever you restart the program, @emph{all} threads start
6508 executing. @emph{This is true even when single-stepping} with commands
6509 like @code{step} or @code{next}.
6511 In particular, @value{GDBN} cannot single-step all threads in lockstep.
6512 Since thread scheduling is up to your debugging target's operating
6513 system (not controlled by @value{GDBN}), other threads may
6514 execute more than one statement while the current thread completes a
6515 single step. Moreover, in general other threads stop in the middle of a
6516 statement, rather than at a clean statement boundary, when the program
6519 You might even find your program stopped in another thread after
6520 continuing or even single-stepping. This happens whenever some other
6521 thread runs into a breakpoint, a signal, or an exception before the
6522 first thread completes whatever you requested.
6524 @cindex automatic thread selection
6525 @cindex switching threads automatically
6526 @cindex threads, automatic switching
6527 Whenever @value{GDBN} stops your program, due to a breakpoint or a
6528 signal, it automatically selects the thread where that breakpoint or
6529 signal happened. @value{GDBN} alerts you to the context switch with a
6530 message such as @samp{[Switching to Thread @var{n}]} to identify the
6533 On some OSes, you can modify @value{GDBN}'s default behavior by
6534 locking the OS scheduler to allow only a single thread to run.
6537 @item set scheduler-locking @var{mode}
6538 @cindex scheduler locking mode
6539 @cindex lock scheduler
6540 Set the scheduler locking mode. It applies to normal execution,
6541 record mode, and replay mode. If it is @code{off}, then there is no
6542 locking and any thread may run at any time. If @code{on}, then only
6543 the current thread may run when the inferior is resumed. The
6544 @code{step} mode optimizes for single-stepping; it prevents other
6545 threads from preempting the current thread while you are stepping, so
6546 that the focus of debugging does not change unexpectedly. Other
6547 threads never get a chance to run when you step, and they are
6548 completely free to run when you use commands like @samp{continue},
6549 @samp{until}, or @samp{finish}. However, unless another thread hits a
6550 breakpoint during its timeslice, @value{GDBN} does not change the
6551 current thread away from the thread that you are debugging. The
6552 @code{replay} mode behaves like @code{off} in record mode and like
6553 @code{on} in replay mode.
6555 @item show scheduler-locking
6556 Display the current scheduler locking mode.
6559 @cindex resume threads of multiple processes simultaneously
6560 By default, when you issue one of the execution commands such as
6561 @code{continue}, @code{next} or @code{step}, @value{GDBN} allows only
6562 threads of the current inferior to run. For example, if @value{GDBN}
6563 is attached to two inferiors, each with two threads, the
6564 @code{continue} command resumes only the two threads of the current
6565 inferior. This is useful, for example, when you debug a program that
6566 forks and you want to hold the parent stopped (so that, for instance,
6567 it doesn't run to exit), while you debug the child. In other
6568 situations, you may not be interested in inspecting the current state
6569 of any of the processes @value{GDBN} is attached to, and you may want
6570 to resume them all until some breakpoint is hit. In the latter case,
6571 you can instruct @value{GDBN} to allow all threads of all the
6572 inferiors to run with the @w{@code{set schedule-multiple}} command.
6575 @kindex set schedule-multiple
6576 @item set schedule-multiple
6577 Set the mode for allowing threads of multiple processes to be resumed
6578 when an execution command is issued. When @code{on}, all threads of
6579 all processes are allowed to run. When @code{off}, only the threads
6580 of the current process are resumed. The default is @code{off}. The
6581 @code{scheduler-locking} mode takes precedence when set to @code{on},
6582 or while you are stepping and set to @code{step}.
6584 @item show schedule-multiple
6585 Display the current mode for resuming the execution of threads of
6590 @subsection Non-Stop Mode
6592 @cindex non-stop mode
6594 @c This section is really only a place-holder, and needs to be expanded
6595 @c with more details.
6597 For some multi-threaded targets, @value{GDBN} supports an optional
6598 mode of operation in which you can examine stopped program threads in
6599 the debugger while other threads continue to execute freely. This
6600 minimizes intrusion when debugging live systems, such as programs
6601 where some threads have real-time constraints or must continue to
6602 respond to external events. This is referred to as @dfn{non-stop} mode.
6604 In non-stop mode, when a thread stops to report a debugging event,
6605 @emph{only} that thread is stopped; @value{GDBN} does not stop other
6606 threads as well, in contrast to the all-stop mode behavior. Additionally,
6607 execution commands such as @code{continue} and @code{step} apply by default
6608 only to the current thread in non-stop mode, rather than all threads as
6609 in all-stop mode. This allows you to control threads explicitly in
6610 ways that are not possible in all-stop mode --- for example, stepping
6611 one thread while allowing others to run freely, stepping
6612 one thread while holding all others stopped, or stepping several threads
6613 independently and simultaneously.
6615 To enter non-stop mode, use this sequence of commands before you run
6616 or attach to your program:
6619 # If using the CLI, pagination breaks non-stop.
6622 # Finally, turn it on!
6626 You can use these commands to manipulate the non-stop mode setting:
6629 @kindex set non-stop
6630 @item set non-stop on
6631 Enable selection of non-stop mode.
6632 @item set non-stop off
6633 Disable selection of non-stop mode.
6634 @kindex show non-stop
6636 Show the current non-stop enablement setting.
6639 Note these commands only reflect whether non-stop mode is enabled,
6640 not whether the currently-executing program is being run in non-stop mode.
6641 In particular, the @code{set non-stop} preference is only consulted when
6642 @value{GDBN} starts or connects to the target program, and it is generally
6643 not possible to switch modes once debugging has started. Furthermore,
6644 since not all targets support non-stop mode, even when you have enabled
6645 non-stop mode, @value{GDBN} may still fall back to all-stop operation by
6648 In non-stop mode, all execution commands apply only to the current thread
6649 by default. That is, @code{continue} only continues one thread.
6650 To continue all threads, issue @code{continue -a} or @code{c -a}.
6652 You can use @value{GDBN}'s background execution commands
6653 (@pxref{Background Execution}) to run some threads in the background
6654 while you continue to examine or step others from @value{GDBN}.
6655 The MI execution commands (@pxref{GDB/MI Program Execution}) are
6656 always executed asynchronously in non-stop mode.
6658 Suspending execution is done with the @code{interrupt} command when
6659 running in the background, or @kbd{Ctrl-c} during foreground execution.
6660 In all-stop mode, this stops the whole process;
6661 but in non-stop mode the interrupt applies only to the current thread.
6662 To stop the whole program, use @code{interrupt -a}.
6664 Other execution commands do not currently support the @code{-a} option.
6666 In non-stop mode, when a thread stops, @value{GDBN} doesn't automatically make
6667 that thread current, as it does in all-stop mode. This is because the
6668 thread stop notifications are asynchronous with respect to @value{GDBN}'s
6669 command interpreter, and it would be confusing if @value{GDBN} unexpectedly
6670 changed to a different thread just as you entered a command to operate on the
6671 previously current thread.
6673 @node Background Execution
6674 @subsection Background Execution
6676 @cindex foreground execution
6677 @cindex background execution
6678 @cindex asynchronous execution
6679 @cindex execution, foreground, background and asynchronous
6681 @value{GDBN}'s execution commands have two variants: the normal
6682 foreground (synchronous) behavior, and a background
6683 (asynchronous) behavior. In foreground execution, @value{GDBN} waits for
6684 the program to report that some thread has stopped before prompting for
6685 another command. In background execution, @value{GDBN} immediately gives
6686 a command prompt so that you can issue other commands while your program runs.
6688 If the target doesn't support async mode, @value{GDBN} issues an error
6689 message if you attempt to use the background execution commands.
6691 @cindex @code{&}, background execution of commands
6692 To specify background execution, add a @code{&} to the command. For example,
6693 the background form of the @code{continue} command is @code{continue&}, or
6694 just @code{c&}. The execution commands that accept background execution
6700 @xref{Starting, , Starting your Program}.
6704 @xref{Attach, , Debugging an Already-running Process}.
6708 @xref{Continuing and Stepping, step}.
6712 @xref{Continuing and Stepping, stepi}.
6716 @xref{Continuing and Stepping, next}.
6720 @xref{Continuing and Stepping, nexti}.
6724 @xref{Continuing and Stepping, continue}.
6728 @xref{Continuing and Stepping, finish}.
6732 @xref{Continuing and Stepping, until}.
6736 Background execution is especially useful in conjunction with non-stop
6737 mode for debugging programs with multiple threads; see @ref{Non-Stop Mode}.
6738 However, you can also use these commands in the normal all-stop mode with
6739 the restriction that you cannot issue another execution command until the
6740 previous one finishes. Examples of commands that are valid in all-stop
6741 mode while the program is running include @code{help} and @code{info break}.
6743 You can interrupt your program while it is running in the background by
6744 using the @code{interrupt} command.
6751 Suspend execution of the running program. In all-stop mode,
6752 @code{interrupt} stops the whole process, but in non-stop mode, it stops
6753 only the current thread. To stop the whole program in non-stop mode,
6754 use @code{interrupt -a}.
6757 @node Thread-Specific Breakpoints
6758 @subsection Thread-Specific Breakpoints
6760 When your program has multiple threads (@pxref{Threads,, Debugging
6761 Programs with Multiple Threads}), you can choose whether to set
6762 breakpoints on all threads, or on a particular thread.
6765 @cindex breakpoints and threads
6766 @cindex thread breakpoints
6767 @kindex break @dots{} thread @var{thread-id}
6768 @item break @var{location} thread @var{thread-id}
6769 @itemx break @var{location} thread @var{thread-id} if @dots{}
6770 @var{location} specifies source lines; there are several ways of
6771 writing them (@pxref{Specify Location}), but the effect is always to
6772 specify some source line.
6774 Use the qualifier @samp{thread @var{thread-id}} with a breakpoint command
6775 to specify that you only want @value{GDBN} to stop the program when a
6776 particular thread reaches this breakpoint. The @var{thread-id} specifier
6777 is one of the thread identifiers assigned by @value{GDBN}, shown
6778 in the first column of the @samp{info threads} display.
6780 If you do not specify @samp{thread @var{thread-id}} when you set a
6781 breakpoint, the breakpoint applies to @emph{all} threads of your
6784 You can use the @code{thread} qualifier on conditional breakpoints as
6785 well; in this case, place @samp{thread @var{thread-id}} before or
6786 after the breakpoint condition, like this:
6789 (@value{GDBP}) break frik.c:13 thread 28 if bartab > lim
6794 Thread-specific breakpoints are automatically deleted when
6795 @value{GDBN} detects the corresponding thread is no longer in the
6796 thread list. For example:
6800 Thread-specific breakpoint 3 deleted - thread 28 no longer in the thread list.
6803 There are several ways for a thread to disappear, such as a regular
6804 thread exit, but also when you detach from the process with the
6805 @code{detach} command (@pxref{Attach, ,Debugging an Already-running
6806 Process}), or if @value{GDBN} loses the remote connection
6807 (@pxref{Remote Debugging}), etc. Note that with some targets,
6808 @value{GDBN} is only able to detect a thread has exited when the user
6809 explictly asks for the thread list with the @code{info threads}
6812 @node Interrupted System Calls
6813 @subsection Interrupted System Calls
6815 @cindex thread breakpoints and system calls
6816 @cindex system calls and thread breakpoints
6817 @cindex premature return from system calls
6818 There is an unfortunate side effect when using @value{GDBN} to debug
6819 multi-threaded programs. If one thread stops for a
6820 breakpoint, or for some other reason, and another thread is blocked in a
6821 system call, then the system call may return prematurely. This is a
6822 consequence of the interaction between multiple threads and the signals
6823 that @value{GDBN} uses to implement breakpoints and other events that
6826 To handle this problem, your program should check the return value of
6827 each system call and react appropriately. This is good programming
6830 For example, do not write code like this:
6836 The call to @code{sleep} will return early if a different thread stops
6837 at a breakpoint or for some other reason.
6839 Instead, write this:
6844 unslept = sleep (unslept);
6847 A system call is allowed to return early, so the system is still
6848 conforming to its specification. But @value{GDBN} does cause your
6849 multi-threaded program to behave differently than it would without
6852 Also, @value{GDBN} uses internal breakpoints in the thread library to
6853 monitor certain events such as thread creation and thread destruction.
6854 When such an event happens, a system call in another thread may return
6855 prematurely, even though your program does not appear to stop.
6858 @subsection Observer Mode
6860 If you want to build on non-stop mode and observe program behavior
6861 without any chance of disruption by @value{GDBN}, you can set
6862 variables to disable all of the debugger's attempts to modify state,
6863 whether by writing memory, inserting breakpoints, etc. These operate
6864 at a low level, intercepting operations from all commands.
6866 When all of these are set to @code{off}, then @value{GDBN} is said to
6867 be @dfn{observer mode}. As a convenience, the variable
6868 @code{observer} can be set to disable these, plus enable non-stop
6871 Note that @value{GDBN} will not prevent you from making nonsensical
6872 combinations of these settings. For instance, if you have enabled
6873 @code{may-insert-breakpoints} but disabled @code{may-write-memory},
6874 then breakpoints that work by writing trap instructions into the code
6875 stream will still not be able to be placed.
6880 @item set observer on
6881 @itemx set observer off
6882 When set to @code{on}, this disables all the permission variables
6883 below (except for @code{insert-fast-tracepoints}), plus enables
6884 non-stop debugging. Setting this to @code{off} switches back to
6885 normal debugging, though remaining in non-stop mode.
6888 Show whether observer mode is on or off.
6890 @kindex may-write-registers
6891 @item set may-write-registers on
6892 @itemx set may-write-registers off
6893 This controls whether @value{GDBN} will attempt to alter the values of
6894 registers, such as with assignment expressions in @code{print}, or the
6895 @code{jump} command. It defaults to @code{on}.
6897 @item show may-write-registers
6898 Show the current permission to write registers.
6900 @kindex may-write-memory
6901 @item set may-write-memory on
6902 @itemx set may-write-memory off
6903 This controls whether @value{GDBN} will attempt to alter the contents
6904 of memory, such as with assignment expressions in @code{print}. It
6905 defaults to @code{on}.
6907 @item show may-write-memory
6908 Show the current permission to write memory.
6910 @kindex may-insert-breakpoints
6911 @item set may-insert-breakpoints on
6912 @itemx set may-insert-breakpoints off
6913 This controls whether @value{GDBN} will attempt to insert breakpoints.
6914 This affects all breakpoints, including internal breakpoints defined
6915 by @value{GDBN}. It defaults to @code{on}.
6917 @item show may-insert-breakpoints
6918 Show the current permission to insert breakpoints.
6920 @kindex may-insert-tracepoints
6921 @item set may-insert-tracepoints on
6922 @itemx set may-insert-tracepoints off
6923 This controls whether @value{GDBN} will attempt to insert (regular)
6924 tracepoints at the beginning of a tracing experiment. It affects only
6925 non-fast tracepoints, fast tracepoints being under the control of
6926 @code{may-insert-fast-tracepoints}. It defaults to @code{on}.
6928 @item show may-insert-tracepoints
6929 Show the current permission to insert tracepoints.
6931 @kindex may-insert-fast-tracepoints
6932 @item set may-insert-fast-tracepoints on
6933 @itemx set may-insert-fast-tracepoints off
6934 This controls whether @value{GDBN} will attempt to insert fast
6935 tracepoints at the beginning of a tracing experiment. It affects only
6936 fast tracepoints, regular (non-fast) tracepoints being under the
6937 control of @code{may-insert-tracepoints}. It defaults to @code{on}.
6939 @item show may-insert-fast-tracepoints
6940 Show the current permission to insert fast tracepoints.
6942 @kindex may-interrupt
6943 @item set may-interrupt on
6944 @itemx set may-interrupt off
6945 This controls whether @value{GDBN} will attempt to interrupt or stop
6946 program execution. When this variable is @code{off}, the
6947 @code{interrupt} command will have no effect, nor will
6948 @kbd{Ctrl-c}. It defaults to @code{on}.
6950 @item show may-interrupt
6951 Show the current permission to interrupt or stop the program.
6955 @node Reverse Execution
6956 @chapter Running programs backward
6957 @cindex reverse execution
6958 @cindex running programs backward
6960 When you are debugging a program, it is not unusual to realize that
6961 you have gone too far, and some event of interest has already happened.
6962 If the target environment supports it, @value{GDBN} can allow you to
6963 ``rewind'' the program by running it backward.
6965 A target environment that supports reverse execution should be able
6966 to ``undo'' the changes in machine state that have taken place as the
6967 program was executing normally. Variables, registers etc.@: should
6968 revert to their previous values. Obviously this requires a great
6969 deal of sophistication on the part of the target environment; not
6970 all target environments can support reverse execution.
6972 When a program is executed in reverse, the instructions that
6973 have most recently been executed are ``un-executed'', in reverse
6974 order. The program counter runs backward, following the previous
6975 thread of execution in reverse. As each instruction is ``un-executed'',
6976 the values of memory and/or registers that were changed by that
6977 instruction are reverted to their previous states. After executing
6978 a piece of source code in reverse, all side effects of that code
6979 should be ``undone'', and all variables should be returned to their
6980 prior values@footnote{
6981 Note that some side effects are easier to undo than others. For instance,
6982 memory and registers are relatively easy, but device I/O is hard. Some
6983 targets may be able undo things like device I/O, and some may not.
6985 The contract between @value{GDBN} and the reverse executing target
6986 requires only that the target do something reasonable when
6987 @value{GDBN} tells it to execute backwards, and then report the
6988 results back to @value{GDBN}. Whatever the target reports back to
6989 @value{GDBN}, @value{GDBN} will report back to the user. @value{GDBN}
6990 assumes that the memory and registers that the target reports are in a
6991 consistent state, but @value{GDBN} accepts whatever it is given.
6994 On some platforms, @value{GDBN} has built-in support for reverse
6995 execution, activated with the @code{record} or @code{record btrace}
6996 commands. @xref{Process Record and Replay}. Some remote targets,
6997 typically full system emulators, support reverse execution directly
6998 without requiring any special command.
7000 If you are debugging in a target environment that supports
7001 reverse execution, @value{GDBN} provides the following commands.
7004 @kindex reverse-continue
7005 @kindex rc @r{(@code{reverse-continue})}
7006 @item reverse-continue @r{[}@var{ignore-count}@r{]}
7007 @itemx rc @r{[}@var{ignore-count}@r{]}
7008 Beginning at the point where your program last stopped, start executing
7009 in reverse. Reverse execution will stop for breakpoints and synchronous
7010 exceptions (signals), just like normal execution. Behavior of
7011 asynchronous signals depends on the target environment.
7013 @kindex reverse-step
7014 @kindex rs @r{(@code{step})}
7015 @item reverse-step @r{[}@var{count}@r{]}
7016 Run the program backward until control reaches the start of a
7017 different source line; then stop it, and return control to @value{GDBN}.
7019 Like the @code{step} command, @code{reverse-step} will only stop
7020 at the beginning of a source line. It ``un-executes'' the previously
7021 executed source line. If the previous source line included calls to
7022 debuggable functions, @code{reverse-step} will step (backward) into
7023 the called function, stopping at the beginning of the @emph{last}
7024 statement in the called function (typically a return statement).
7026 Also, as with the @code{step} command, if non-debuggable functions are
7027 called, @code{reverse-step} will run thru them backward without stopping.
7029 @kindex reverse-stepi
7030 @kindex rsi @r{(@code{reverse-stepi})}
7031 @item reverse-stepi @r{[}@var{count}@r{]}
7032 Reverse-execute one machine instruction. Note that the instruction
7033 to be reverse-executed is @emph{not} the one pointed to by the program
7034 counter, but the instruction executed prior to that one. For instance,
7035 if the last instruction was a jump, @code{reverse-stepi} will take you
7036 back from the destination of the jump to the jump instruction itself.
7038 @kindex reverse-next
7039 @kindex rn @r{(@code{reverse-next})}
7040 @item reverse-next @r{[}@var{count}@r{]}
7041 Run backward to the beginning of the previous line executed in
7042 the current (innermost) stack frame. If the line contains function
7043 calls, they will be ``un-executed'' without stopping. Starting from
7044 the first line of a function, @code{reverse-next} will take you back
7045 to the caller of that function, @emph{before} the function was called,
7046 just as the normal @code{next} command would take you from the last
7047 line of a function back to its return to its caller
7048 @footnote{Unless the code is too heavily optimized.}.
7050 @kindex reverse-nexti
7051 @kindex rni @r{(@code{reverse-nexti})}
7052 @item reverse-nexti @r{[}@var{count}@r{]}
7053 Like @code{nexti}, @code{reverse-nexti} executes a single instruction
7054 in reverse, except that called functions are ``un-executed'' atomically.
7055 That is, if the previously executed instruction was a return from
7056 another function, @code{reverse-nexti} will continue to execute
7057 in reverse until the call to that function (from the current stack
7060 @kindex reverse-finish
7061 @item reverse-finish
7062 Just as the @code{finish} command takes you to the point where the
7063 current function returns, @code{reverse-finish} takes you to the point
7064 where it was called. Instead of ending up at the end of the current
7065 function invocation, you end up at the beginning.
7067 @kindex set exec-direction
7068 @item set exec-direction
7069 Set the direction of target execution.
7070 @item set exec-direction reverse
7071 @cindex execute forward or backward in time
7072 @value{GDBN} will perform all execution commands in reverse, until the
7073 exec-direction mode is changed to ``forward''. Affected commands include
7074 @code{step, stepi, next, nexti, continue, and finish}. The @code{return}
7075 command cannot be used in reverse mode.
7076 @item set exec-direction forward
7077 @value{GDBN} will perform all execution commands in the normal fashion.
7078 This is the default.
7082 @node Process Record and Replay
7083 @chapter Recording Inferior's Execution and Replaying It
7084 @cindex process record and replay
7085 @cindex recording inferior's execution and replaying it
7087 On some platforms, @value{GDBN} provides a special @dfn{process record
7088 and replay} target that can record a log of the process execution, and
7089 replay it later with both forward and reverse execution commands.
7092 When this target is in use, if the execution log includes the record
7093 for the next instruction, @value{GDBN} will debug in @dfn{replay
7094 mode}. In the replay mode, the inferior does not really execute code
7095 instructions. Instead, all the events that normally happen during
7096 code execution are taken from the execution log. While code is not
7097 really executed in replay mode, the values of registers (including the
7098 program counter register) and the memory of the inferior are still
7099 changed as they normally would. Their contents are taken from the
7103 If the record for the next instruction is not in the execution log,
7104 @value{GDBN} will debug in @dfn{record mode}. In this mode, the
7105 inferior executes normally, and @value{GDBN} records the execution log
7108 The process record and replay target supports reverse execution
7109 (@pxref{Reverse Execution}), even if the platform on which the
7110 inferior runs does not. However, the reverse execution is limited in
7111 this case by the range of the instructions recorded in the execution
7112 log. In other words, reverse execution on platforms that don't
7113 support it directly can only be done in the replay mode.
7115 When debugging in the reverse direction, @value{GDBN} will work in
7116 replay mode as long as the execution log includes the record for the
7117 previous instruction; otherwise, it will work in record mode, if the
7118 platform supports reverse execution, or stop if not.
7120 Currently, process record and replay is supported on ARM, Aarch64,
7121 Moxie, PowerPC, PowerPC64, S/390, and x86 (i386/amd64) running
7122 GNU/Linux. Process record and replay can be used both when native
7123 debugging, and when remote debugging via @code{gdbserver}.
7125 For architecture environments that support process record and replay,
7126 @value{GDBN} provides the following commands:
7129 @kindex target record
7130 @kindex target record-full
7131 @kindex target record-btrace
7134 @kindex record btrace
7135 @kindex record btrace bts
7136 @kindex record btrace pt
7142 @kindex rec btrace bts
7143 @kindex rec btrace pt
7146 @item record @var{method}
7147 This command starts the process record and replay target. The
7148 recording method can be specified as parameter. Without a parameter
7149 the command uses the @code{full} recording method. The following
7150 recording methods are available:
7154 Full record/replay recording using @value{GDBN}'s software record and
7155 replay implementation. This method allows replaying and reverse
7158 @item btrace @var{format}
7159 Hardware-supported instruction recording, supported on Intel
7160 processors. This method does not record data. Further, the data is
7161 collected in a ring buffer so old data will be overwritten when the
7162 buffer is full. It allows limited reverse execution. Variables and
7163 registers are not available during reverse execution. In remote
7164 debugging, recording continues on disconnect. Recorded data can be
7165 inspected after reconnecting. The recording may be stopped using
7168 The recording format can be specified as parameter. Without a parameter
7169 the command chooses the recording format. The following recording
7170 formats are available:
7174 @cindex branch trace store
7175 Use the @dfn{Branch Trace Store} (@acronym{BTS}) recording format. In
7176 this format, the processor stores a from/to record for each executed
7177 branch in the btrace ring buffer.
7180 @cindex Intel Processor Trace
7181 Use the @dfn{Intel Processor Trace} recording format. In this
7182 format, the processor stores the execution trace in a compressed form
7183 that is afterwards decoded by @value{GDBN}.
7185 The trace can be recorded with very low overhead. The compressed
7186 trace format also allows small trace buffers to already contain a big
7187 number of instructions compared to @acronym{BTS}.
7189 Decoding the recorded execution trace, on the other hand, is more
7190 expensive than decoding @acronym{BTS} trace. This is mostly due to the
7191 increased number of instructions to process. You should increase the
7192 buffer-size with care.
7195 Not all recording formats may be available on all processors.
7198 The process record and replay target can only debug a process that is
7199 already running. Therefore, you need first to start the process with
7200 the @kbd{run} or @kbd{start} commands, and then start the recording
7201 with the @kbd{record @var{method}} command.
7203 @cindex displaced stepping, and process record and replay
7204 Displaced stepping (@pxref{Maintenance Commands,, displaced stepping})
7205 will be automatically disabled when process record and replay target
7206 is started. That's because the process record and replay target
7207 doesn't support displaced stepping.
7209 @cindex non-stop mode, and process record and replay
7210 @cindex asynchronous execution, and process record and replay
7211 If the inferior is in the non-stop mode (@pxref{Non-Stop Mode}) or in
7212 the asynchronous execution mode (@pxref{Background Execution}), not
7213 all recording methods are available. The @code{full} recording method
7214 does not support these two modes.
7219 Stop the process record and replay target. When process record and
7220 replay target stops, the entire execution log will be deleted and the
7221 inferior will either be terminated, or will remain in its final state.
7223 When you stop the process record and replay target in record mode (at
7224 the end of the execution log), the inferior will be stopped at the
7225 next instruction that would have been recorded. In other words, if
7226 you record for a while and then stop recording, the inferior process
7227 will be left in the same state as if the recording never happened.
7229 On the other hand, if the process record and replay target is stopped
7230 while in replay mode (that is, not at the end of the execution log,
7231 but at some earlier point), the inferior process will become ``live''
7232 at that earlier state, and it will then be possible to continue the
7233 usual ``live'' debugging of the process from that state.
7235 When the inferior process exits, or @value{GDBN} detaches from it,
7236 process record and replay target will automatically stop itself.
7240 Go to a specific location in the execution log. There are several
7241 ways to specify the location to go to:
7244 @item record goto begin
7245 @itemx record goto start
7246 Go to the beginning of the execution log.
7248 @item record goto end
7249 Go to the end of the execution log.
7251 @item record goto @var{n}
7252 Go to instruction number @var{n} in the execution log.
7256 @item record save @var{filename}
7257 Save the execution log to a file @file{@var{filename}}.
7258 Default filename is @file{gdb_record.@var{process_id}}, where
7259 @var{process_id} is the process ID of the inferior.
7261 This command may not be available for all recording methods.
7263 @kindex record restore
7264 @item record restore @var{filename}
7265 Restore the execution log from a file @file{@var{filename}}.
7266 File must have been created with @code{record save}.
7268 @kindex set record full
7269 @item set record full insn-number-max @var{limit}
7270 @itemx set record full insn-number-max unlimited
7271 Set the limit of instructions to be recorded for the @code{full}
7272 recording method. Default value is 200000.
7274 If @var{limit} is a positive number, then @value{GDBN} will start
7275 deleting instructions from the log once the number of the record
7276 instructions becomes greater than @var{limit}. For every new recorded
7277 instruction, @value{GDBN} will delete the earliest recorded
7278 instruction to keep the number of recorded instructions at the limit.
7279 (Since deleting recorded instructions loses information, @value{GDBN}
7280 lets you control what happens when the limit is reached, by means of
7281 the @code{stop-at-limit} option, described below.)
7283 If @var{limit} is @code{unlimited} or zero, @value{GDBN} will never
7284 delete recorded instructions from the execution log. The number of
7285 recorded instructions is limited only by the available memory.
7287 @kindex show record full
7288 @item show record full insn-number-max
7289 Show the limit of instructions to be recorded with the @code{full}
7292 @item set record full stop-at-limit
7293 Control the behavior of the @code{full} recording method when the
7294 number of recorded instructions reaches the limit. If ON (the
7295 default), @value{GDBN} will stop when the limit is reached for the
7296 first time and ask you whether you want to stop the inferior or
7297 continue running it and recording the execution log. If you decide
7298 to continue recording, each new recorded instruction will cause the
7299 oldest one to be deleted.
7301 If this option is OFF, @value{GDBN} will automatically delete the
7302 oldest record to make room for each new one, without asking.
7304 @item show record full stop-at-limit
7305 Show the current setting of @code{stop-at-limit}.
7307 @item set record full memory-query
7308 Control the behavior when @value{GDBN} is unable to record memory
7309 changes caused by an instruction for the @code{full} recording method.
7310 If ON, @value{GDBN} will query whether to stop the inferior in that
7313 If this option is OFF (the default), @value{GDBN} will automatically
7314 ignore the effect of such instructions on memory. Later, when
7315 @value{GDBN} replays this execution log, it will mark the log of this
7316 instruction as not accessible, and it will not affect the replay
7319 @item show record full memory-query
7320 Show the current setting of @code{memory-query}.
7322 @kindex set record btrace
7323 The @code{btrace} record target does not trace data. As a
7324 convenience, when replaying, @value{GDBN} reads read-only memory off
7325 the live program directly, assuming that the addresses of the
7326 read-only areas don't change. This for example makes it possible to
7327 disassemble code while replaying, but not to print variables.
7328 In some cases, being able to inspect variables might be useful.
7329 You can use the following command for that:
7331 @item set record btrace replay-memory-access
7332 Control the behavior of the @code{btrace} recording method when
7333 accessing memory during replay. If @code{read-only} (the default),
7334 @value{GDBN} will only allow accesses to read-only memory.
7335 If @code{read-write}, @value{GDBN} will allow accesses to read-only
7336 and to read-write memory. Beware that the accessed memory corresponds
7337 to the live target and not necessarily to the current replay
7340 @item set record btrace cpu @var{identifier}
7341 Set the processor to be used for enabling workarounds for processor
7342 errata when decoding the trace.
7344 Processor errata are defects in processor operation, caused by its
7345 design or manufacture. They can cause a trace not to match the
7346 specification. This, in turn, may cause trace decode to fail.
7347 @value{GDBN} can detect erroneous trace packets and correct them, thus
7348 avoiding the decoding failures. These corrections are known as
7349 @dfn{errata workarounds}, and are enabled based on the processor on
7350 which the trace was recorded.
7352 By default, @value{GDBN} attempts to detect the processor
7353 automatically, and apply the necessary workarounds for it. However,
7354 you may need to specify the processor if @value{GDBN} does not yet
7355 support it. This command allows you to do that, and also allows to
7356 disable the workarounds.
7358 The argument @var{identifier} identifies the @sc{cpu} and is of the
7359 form: @code{@var{vendor}:@var{processor identifier}}. In addition,
7360 there are two special identifiers, @code{none} and @code{auto}
7363 The following vendor identifiers and corresponding processor
7364 identifiers are currently supported:
7366 @multitable @columnfractions .1 .9
7369 @tab @var{family}/@var{model}[/@var{stepping}]
7373 On GNU/Linux systems, the processor @var{family}, @var{model}, and
7374 @var{stepping} can be obtained from @code{/proc/cpuinfo}.
7376 If @var{identifier} is @code{auto}, enable errata workarounds for the
7377 processor on which the trace was recorded. If @var{identifier} is
7378 @code{none}, errata workarounds are disabled.
7380 For example, when using an old @value{GDBN} on a new system, decode
7381 may fail because @value{GDBN} does not support the new processor. It
7382 often suffices to specify an older processor that @value{GDBN}
7387 Active record target: record-btrace
7388 Recording format: Intel Processor Trace.
7390 Failed to configure the Intel Processor Trace decoder: unknown cpu.
7391 (gdb) set record btrace cpu intel:6/158
7393 Active record target: record-btrace
7394 Recording format: Intel Processor Trace.
7396 Recorded 84872 instructions in 3189 functions (0 gaps) for thread 1 (...).
7399 @kindex show record btrace
7400 @item show record btrace replay-memory-access
7401 Show the current setting of @code{replay-memory-access}.
7403 @item show record btrace cpu
7404 Show the processor to be used for enabling trace decode errata
7407 @kindex set record btrace bts
7408 @item set record btrace bts buffer-size @var{size}
7409 @itemx set record btrace bts buffer-size unlimited
7410 Set the requested ring buffer size for branch tracing in @acronym{BTS}
7411 format. Default is 64KB.
7413 If @var{size} is a positive number, then @value{GDBN} will try to
7414 allocate a buffer of at least @var{size} bytes for each new thread
7415 that uses the btrace recording method and the @acronym{BTS} format.
7416 The actually obtained buffer size may differ from the requested
7417 @var{size}. Use the @code{info record} command to see the actual
7418 buffer size for each thread that uses the btrace recording method and
7419 the @acronym{BTS} format.
7421 If @var{limit} is @code{unlimited} or zero, @value{GDBN} will try to
7422 allocate a buffer of 4MB.
7424 Bigger buffers mean longer traces. On the other hand, @value{GDBN} will
7425 also need longer to process the branch trace data before it can be used.
7427 @item show record btrace bts buffer-size @var{size}
7428 Show the current setting of the requested ring buffer size for branch
7429 tracing in @acronym{BTS} format.
7431 @kindex set record btrace pt
7432 @item set record btrace pt buffer-size @var{size}
7433 @itemx set record btrace pt buffer-size unlimited
7434 Set the requested ring buffer size for branch tracing in Intel
7435 Processor Trace format. Default is 16KB.
7437 If @var{size} is a positive number, then @value{GDBN} will try to
7438 allocate a buffer of at least @var{size} bytes for each new thread
7439 that uses the btrace recording method and the Intel Processor Trace
7440 format. The actually obtained buffer size may differ from the
7441 requested @var{size}. Use the @code{info record} command to see the
7442 actual buffer size for each thread.
7444 If @var{limit} is @code{unlimited} or zero, @value{GDBN} will try to
7445 allocate a buffer of 4MB.
7447 Bigger buffers mean longer traces. On the other hand, @value{GDBN} will
7448 also need longer to process the branch trace data before it can be used.
7450 @item show record btrace pt buffer-size @var{size}
7451 Show the current setting of the requested ring buffer size for branch
7452 tracing in Intel Processor Trace format.
7456 Show various statistics about the recording depending on the recording
7461 For the @code{full} recording method, it shows the state of process
7462 record and its in-memory execution log buffer, including:
7466 Whether in record mode or replay mode.
7468 Lowest recorded instruction number (counting from when the current execution log started recording instructions).
7470 Highest recorded instruction number.
7472 Current instruction about to be replayed (if in replay mode).
7474 Number of instructions contained in the execution log.
7476 Maximum number of instructions that may be contained in the execution log.
7480 For the @code{btrace} recording method, it shows:
7486 Number of instructions that have been recorded.
7488 Number of blocks of sequential control-flow formed by the recorded
7491 Whether in record mode or replay mode.
7494 For the @code{bts} recording format, it also shows:
7497 Size of the perf ring buffer.
7500 For the @code{pt} recording format, it also shows:
7503 Size of the perf ring buffer.
7507 @kindex record delete
7510 When record target runs in replay mode (``in the past''), delete the
7511 subsequent execution log and begin to record a new execution log starting
7512 from the current address. This means you will abandon the previously
7513 recorded ``future'' and begin recording a new ``future''.
7515 @kindex record instruction-history
7516 @kindex rec instruction-history
7517 @item record instruction-history
7518 Disassembles instructions from the recorded execution log. By
7519 default, ten instructions are disassembled. This can be changed using
7520 the @code{set record instruction-history-size} command. Instructions
7521 are printed in execution order.
7523 It can also print mixed source+disassembly if you specify the the
7524 @code{/m} or @code{/s} modifier, and print the raw instructions in hex
7525 as well as in symbolic form by specifying the @code{/r} modifier.
7527 The current position marker is printed for the instruction at the
7528 current program counter value. This instruction can appear multiple
7529 times in the trace and the current position marker will be printed
7530 every time. To omit the current position marker, specify the
7533 To better align the printed instructions when the trace contains
7534 instructions from more than one function, the function name may be
7535 omitted by specifying the @code{/f} modifier.
7537 Speculatively executed instructions are prefixed with @samp{?}. This
7538 feature is not available for all recording formats.
7540 There are several ways to specify what part of the execution log to
7544 @item record instruction-history @var{insn}
7545 Disassembles ten instructions starting from instruction number
7548 @item record instruction-history @var{insn}, +/-@var{n}
7549 Disassembles @var{n} instructions around instruction number
7550 @var{insn}. If @var{n} is preceded with @code{+}, disassembles
7551 @var{n} instructions after instruction number @var{insn}. If
7552 @var{n} is preceded with @code{-}, disassembles @var{n}
7553 instructions before instruction number @var{insn}.
7555 @item record instruction-history
7556 Disassembles ten more instructions after the last disassembly.
7558 @item record instruction-history -
7559 Disassembles ten more instructions before the last disassembly.
7561 @item record instruction-history @var{begin}, @var{end}
7562 Disassembles instructions beginning with instruction number
7563 @var{begin} until instruction number @var{end}. The instruction
7564 number @var{end} is included.
7567 This command may not be available for all recording methods.
7570 @item set record instruction-history-size @var{size}
7571 @itemx set record instruction-history-size unlimited
7572 Define how many instructions to disassemble in the @code{record
7573 instruction-history} command. The default value is 10.
7574 A @var{size} of @code{unlimited} means unlimited instructions.
7577 @item show record instruction-history-size
7578 Show how many instructions to disassemble in the @code{record
7579 instruction-history} command.
7581 @kindex record function-call-history
7582 @kindex rec function-call-history
7583 @item record function-call-history
7584 Prints the execution history at function granularity. It prints one
7585 line for each sequence of instructions that belong to the same
7586 function giving the name of that function, the source lines
7587 for this instruction sequence (if the @code{/l} modifier is
7588 specified), and the instructions numbers that form the sequence (if
7589 the @code{/i} modifier is specified). The function names are indented
7590 to reflect the call stack depth if the @code{/c} modifier is
7591 specified. The @code{/l}, @code{/i}, and @code{/c} modifiers can be
7595 (@value{GDBP}) @b{list 1, 10}
7606 (@value{GDBP}) @b{record function-call-history /ilc}
7607 1 bar inst 1,4 at foo.c:6,8
7608 2 foo inst 5,10 at foo.c:2,3
7609 3 bar inst 11,13 at foo.c:9,10
7612 By default, ten lines are printed. This can be changed using the
7613 @code{set record function-call-history-size} command. Functions are
7614 printed in execution order. There are several ways to specify what
7618 @item record function-call-history @var{func}
7619 Prints ten functions starting from function number @var{func}.
7621 @item record function-call-history @var{func}, +/-@var{n}
7622 Prints @var{n} functions around function number @var{func}. If
7623 @var{n} is preceded with @code{+}, prints @var{n} functions after
7624 function number @var{func}. If @var{n} is preceded with @code{-},
7625 prints @var{n} functions before function number @var{func}.
7627 @item record function-call-history
7628 Prints ten more functions after the last ten-line print.
7630 @item record function-call-history -
7631 Prints ten more functions before the last ten-line print.
7633 @item record function-call-history @var{begin}, @var{end}
7634 Prints functions beginning with function number @var{begin} until
7635 function number @var{end}. The function number @var{end} is included.
7638 This command may not be available for all recording methods.
7640 @item set record function-call-history-size @var{size}
7641 @itemx set record function-call-history-size unlimited
7642 Define how many lines to print in the
7643 @code{record function-call-history} command. The default value is 10.
7644 A size of @code{unlimited} means unlimited lines.
7646 @item show record function-call-history-size
7647 Show how many lines to print in the
7648 @code{record function-call-history} command.
7653 @chapter Examining the Stack
7655 When your program has stopped, the first thing you need to know is where it
7656 stopped and how it got there.
7659 Each time your program performs a function call, information about the call
7661 That information includes the location of the call in your program,
7662 the arguments of the call,
7663 and the local variables of the function being called.
7664 The information is saved in a block of data called a @dfn{stack frame}.
7665 The stack frames are allocated in a region of memory called the @dfn{call
7668 When your program stops, the @value{GDBN} commands for examining the
7669 stack allow you to see all of this information.
7671 @cindex selected frame
7672 One of the stack frames is @dfn{selected} by @value{GDBN} and many
7673 @value{GDBN} commands refer implicitly to the selected frame. In
7674 particular, whenever you ask @value{GDBN} for the value of a variable in
7675 your program, the value is found in the selected frame. There are
7676 special @value{GDBN} commands to select whichever frame you are
7677 interested in. @xref{Selection, ,Selecting a Frame}.
7679 When your program stops, @value{GDBN} automatically selects the
7680 currently executing frame and describes it briefly, similar to the
7681 @code{frame} command (@pxref{Frame Info, ,Information about a Frame}).
7684 * Frames:: Stack frames
7685 * Backtrace:: Backtraces
7686 * Selection:: Selecting a frame
7687 * Frame Info:: Information on a frame
7688 * Frame Apply:: Applying a command to several frames
7689 * Frame Filter Management:: Managing frame filters
7694 @section Stack Frames
7696 @cindex frame, definition
7698 The call stack is divided up into contiguous pieces called @dfn{stack
7699 frames}, or @dfn{frames} for short; each frame is the data associated
7700 with one call to one function. The frame contains the arguments given
7701 to the function, the function's local variables, and the address at
7702 which the function is executing.
7704 @cindex initial frame
7705 @cindex outermost frame
7706 @cindex innermost frame
7707 When your program is started, the stack has only one frame, that of the
7708 function @code{main}. This is called the @dfn{initial} frame or the
7709 @dfn{outermost} frame. Each time a function is called, a new frame is
7710 made. Each time a function returns, the frame for that function invocation
7711 is eliminated. If a function is recursive, there can be many frames for
7712 the same function. The frame for the function in which execution is
7713 actually occurring is called the @dfn{innermost} frame. This is the most
7714 recently created of all the stack frames that still exist.
7716 @cindex frame pointer
7717 Inside your program, stack frames are identified by their addresses. A
7718 stack frame consists of many bytes, each of which has its own address; each
7719 kind of computer has a convention for choosing one byte whose
7720 address serves as the address of the frame. Usually this address is kept
7721 in a register called the @dfn{frame pointer register}
7722 (@pxref{Registers, $fp}) while execution is going on in that frame.
7725 @cindex frame number
7726 @value{GDBN} labels each existing stack frame with a @dfn{level}, a
7727 number that is zero for the innermost frame, one for the frame that
7728 called it, and so on upward. These level numbers give you a way of
7729 designating stack frames in @value{GDBN} commands. The terms
7730 @dfn{frame number} and @dfn{frame level} can be used interchangeably to
7731 describe this number.
7733 @c The -fomit-frame-pointer below perennially causes hbox overflow
7734 @c underflow problems.
7735 @cindex frameless execution
7736 Some compilers provide a way to compile functions so that they operate
7737 without stack frames. (For example, the @value{NGCC} option
7739 @samp{-fomit-frame-pointer}
7741 generates functions without a frame.)
7742 This is occasionally done with heavily used library functions to save
7743 the frame setup time. @value{GDBN} has limited facilities for dealing
7744 with these function invocations. If the innermost function invocation
7745 has no stack frame, @value{GDBN} nevertheless regards it as though
7746 it had a separate frame, which is numbered zero as usual, allowing
7747 correct tracing of the function call chain. However, @value{GDBN} has
7748 no provision for frameless functions elsewhere in the stack.
7754 @cindex call stack traces
7755 A backtrace is a summary of how your program got where it is. It shows one
7756 line per frame, for many frames, starting with the currently executing
7757 frame (frame zero), followed by its caller (frame one), and on up the
7760 @anchor{backtrace-command}
7762 @kindex bt @r{(@code{backtrace})}
7763 To print a backtrace of the entire stack, use the @code{backtrace}
7764 command, or its alias @code{bt}. This command will print one line per
7765 frame for frames in the stack. By default, all stack frames are
7766 printed. You can stop the backtrace at any time by typing the system
7767 interrupt character, normally @kbd{Ctrl-c}.
7770 @item backtrace [@var{option}]@dots{} [@var{qualifier}]@dots{} [@var{count}]
7771 @itemx bt [@var{option}]@dots{} [@var{qualifier}]@dots{} [@var{count}]
7772 Print the backtrace of the entire stack.
7774 The optional @var{count} can be one of the following:
7779 Print only the innermost @var{n} frames, where @var{n} is a positive
7784 Print only the outermost @var{n} frames, where @var{n} is a positive
7792 Print the values of the local variables also. This can be combined
7793 with the optional @var{count} to limit the number of frames shown.
7796 Do not run Python frame filters on this backtrace. @xref{Frame
7797 Filter API}, for more information. Additionally use @ref{disable
7798 frame-filter all} to turn off all frame filters. This is only
7799 relevant when @value{GDBN} has been configured with @code{Python}
7803 A Python frame filter might decide to ``elide'' some frames. Normally
7804 such elided frames are still printed, but they are indented relative
7805 to the filtered frames that cause them to be elided. The @code{-hide}
7806 option causes elided frames to not be printed at all.
7809 The @code{backtrace} command also supports a number of options that
7810 allow overriding relevant global print settings as set by @code{set
7811 backtrace} and @code{set print} subcommands:
7814 @item -past-main [@code{on}|@code{off}]
7815 Set whether backtraces should continue past @code{main}. Related setting:
7816 @ref{set backtrace past-main}.
7818 @item -past-entry [@code{on}|@code{off}]
7819 Set whether backtraces should continue past the entry point of a program.
7820 Related setting: @ref{set backtrace past-entry}.
7822 @item -entry-values @code{no}|@code{only}|@code{preferred}|@code{if-needed}|@code{both}|@code{compact}|@code{default}
7823 Set printing of function arguments at function entry.
7824 Related setting: @ref{set print entry-values}.
7826 @item -frame-arguments @code{all}|@code{scalars}|@code{none}
7827 Set printing of non-scalar frame arguments.
7828 Related setting: @ref{set print frame-arguments}.
7830 @item -raw-frame-arguments [@code{on}|@code{off}]
7831 Set whether to print frame arguments in raw form.
7832 Related setting: @ref{set print raw-frame-arguments}.
7834 @item -frame-info @code{auto}|@code{source-line}|@code{location}|@code{source-and-location}|@code{location-and-address}|@code{short-location}
7835 Set printing of frame information.
7836 Related setting: @ref{set print frame-info}.
7839 The optional @var{qualifier} is maintained for backward compatibility.
7840 It can be one of the following:
7844 Equivalent to the @code{-full} option.
7847 Equivalent to the @code{-no-filters} option.
7850 Equivalent to the @code{-hide} option.
7857 The names @code{where} and @code{info stack} (abbreviated @code{info s})
7858 are additional aliases for @code{backtrace}.
7860 @cindex multiple threads, backtrace
7861 In a multi-threaded program, @value{GDBN} by default shows the
7862 backtrace only for the current thread. To display the backtrace for
7863 several or all of the threads, use the command @code{thread apply}
7864 (@pxref{Threads, thread apply}). For example, if you type @kbd{thread
7865 apply all backtrace}, @value{GDBN} will display the backtrace for all
7866 the threads; this is handy when you debug a core dump of a
7867 multi-threaded program.
7869 Each line in the backtrace shows the frame number and the function name.
7870 The program counter value is also shown---unless you use @code{set
7871 print address off}. The backtrace also shows the source file name and
7872 line number, as well as the arguments to the function. The program
7873 counter value is omitted if it is at the beginning of the code for that
7876 Here is an example of a backtrace. It was made with the command
7877 @samp{bt 3}, so it shows the innermost three frames.
7881 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
7883 #1 0x6e38 in expand_macro (sym=0x2b600, data=...) at macro.c:242
7884 #2 0x6840 in expand_token (obs=0x0, t=177664, td=0xf7fffb08)
7886 (More stack frames follow...)
7891 The display for frame zero does not begin with a program counter
7892 value, indicating that your program has stopped at the beginning of the
7893 code for line @code{993} of @code{builtin.c}.
7896 The value of parameter @code{data} in frame 1 has been replaced by
7897 @code{@dots{}}. By default, @value{GDBN} prints the value of a parameter
7898 only if it is a scalar (integer, pointer, enumeration, etc). See command
7899 @kbd{set print frame-arguments} in @ref{Print Settings} for more details
7900 on how to configure the way function parameter values are printed.
7901 The command @kbd{set print frame-info} (@pxref{Print Settings}) controls
7902 what frame information is printed.
7904 @cindex optimized out, in backtrace
7905 @cindex function call arguments, optimized out
7906 If your program was compiled with optimizations, some compilers will
7907 optimize away arguments passed to functions if those arguments are
7908 never used after the call. Such optimizations generate code that
7909 passes arguments through registers, but doesn't store those arguments
7910 in the stack frame. @value{GDBN} has no way of displaying such
7911 arguments in stack frames other than the innermost one. Here's what
7912 such a backtrace might look like:
7916 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
7918 #1 0x6e38 in expand_macro (sym=<optimized out>) at macro.c:242
7919 #2 0x6840 in expand_token (obs=0x0, t=<optimized out>, td=0xf7fffb08)
7921 (More stack frames follow...)
7926 The values of arguments that were not saved in their stack frames are
7927 shown as @samp{<optimized out>}.
7929 If you need to display the values of such optimized-out arguments,
7930 either deduce that from other variables whose values depend on the one
7931 you are interested in, or recompile without optimizations.
7933 @cindex backtrace beyond @code{main} function
7934 @cindex program entry point
7935 @cindex startup code, and backtrace
7936 Most programs have a standard user entry point---a place where system
7937 libraries and startup code transition into user code. For C this is
7938 @code{main}@footnote{
7939 Note that embedded programs (the so-called ``free-standing''
7940 environment) are not required to have a @code{main} function as the
7941 entry point. They could even have multiple entry points.}.
7942 When @value{GDBN} finds the entry function in a backtrace
7943 it will terminate the backtrace, to avoid tracing into highly
7944 system-specific (and generally uninteresting) code.
7946 If you need to examine the startup code, or limit the number of levels
7947 in a backtrace, you can change this behavior:
7950 @item set backtrace past-main
7951 @itemx set backtrace past-main on
7952 @anchor{set backtrace past-main}
7953 @kindex set backtrace
7954 Backtraces will continue past the user entry point.
7956 @item set backtrace past-main off
7957 Backtraces will stop when they encounter the user entry point. This is the
7960 @item show backtrace past-main
7961 @kindex show backtrace
7962 Display the current user entry point backtrace policy.
7964 @item set backtrace past-entry
7965 @itemx set backtrace past-entry on
7966 @anchor{set backtrace past-entry}
7967 Backtraces will continue past the internal entry point of an application.
7968 This entry point is encoded by the linker when the application is built,
7969 and is likely before the user entry point @code{main} (or equivalent) is called.
7971 @item set backtrace past-entry off
7972 Backtraces will stop when they encounter the internal entry point of an
7973 application. This is the default.
7975 @item show backtrace past-entry
7976 Display the current internal entry point backtrace policy.
7978 @item set backtrace limit @var{n}
7979 @itemx set backtrace limit 0
7980 @itemx set backtrace limit unlimited
7981 @anchor{set backtrace limit}
7982 @cindex backtrace limit
7983 Limit the backtrace to @var{n} levels. A value of @code{unlimited}
7984 or zero means unlimited levels.
7986 @item show backtrace limit
7987 Display the current limit on backtrace levels.
7990 You can control how file names are displayed.
7993 @item set filename-display
7994 @itemx set filename-display relative
7995 @cindex filename-display
7996 Display file names relative to the compilation directory. This is the default.
7998 @item set filename-display basename
7999 Display only basename of a filename.
8001 @item set filename-display absolute
8002 Display an absolute filename.
8004 @item show filename-display
8005 Show the current way to display filenames.
8009 @section Selecting a Frame
8011 Most commands for examining the stack and other data in your program work on
8012 whichever stack frame is selected at the moment. Here are the commands for
8013 selecting a stack frame; all of them finish by printing a brief description
8014 of the stack frame just selected.
8017 @kindex frame@r{, selecting}
8018 @kindex f @r{(@code{frame})}
8019 @item frame @r{[} @var{frame-selection-spec} @r{]}
8020 @item f @r{[} @var{frame-selection-spec} @r{]}
8021 The @command{frame} command allows different stack frames to be
8022 selected. The @var{frame-selection-spec} can be any of the following:
8027 @item level @var{num}
8028 Select frame level @var{num}. Recall that frame zero is the innermost
8029 (currently executing) frame, frame one is the frame that called the
8030 innermost one, and so on. The highest level frame is usually the one
8033 As this is the most common method of navigating the frame stack, the
8034 string @command{level} can be omitted. For example, the following two
8035 commands are equivalent:
8038 (@value{GDBP}) frame 3
8039 (@value{GDBP}) frame level 3
8042 @kindex frame address
8043 @item address @var{stack-address}
8044 Select the frame with stack address @var{stack-address}. The
8045 @var{stack-address} for a frame can be seen in the output of
8046 @command{info frame}, for example:
8050 Stack level 1, frame at 0x7fffffffda30:
8051 rip = 0x40066d in b (amd64-entry-value.cc:59); saved rip 0x4004c5
8052 tail call frame, caller of frame at 0x7fffffffda30
8053 source language c++.
8054 Arglist at unknown address.
8055 Locals at unknown address, Previous frame's sp is 0x7fffffffda30
8058 The @var{stack-address} for this frame is @code{0x7fffffffda30} as
8059 indicated by the line:
8062 Stack level 1, frame at 0x7fffffffda30:
8065 @kindex frame function
8066 @item function @var{function-name}
8067 Select the stack frame for function @var{function-name}. If there are
8068 multiple stack frames for function @var{function-name} then the inner
8069 most stack frame is selected.
8072 @item view @var{stack-address} @r{[} @var{pc-addr} @r{]}
8073 View a frame that is not part of @value{GDBN}'s backtrace. The frame
8074 viewed has stack address @var{stack-addr}, and optionally, a program
8075 counter address of @var{pc-addr}.
8077 This is useful mainly if the chaining of stack frames has been
8078 damaged by a bug, making it impossible for @value{GDBN} to assign
8079 numbers properly to all frames. In addition, this can be useful
8080 when your program has multiple stacks and switches between them.
8082 When viewing a frame outside the current backtrace using
8083 @command{frame view} then you can always return to the original
8084 stack using one of the previous stack frame selection instructions,
8085 for example @command{frame level 0}.
8091 Move @var{n} frames up the stack; @var{n} defaults to 1. For positive
8092 numbers @var{n}, this advances toward the outermost frame, to higher
8093 frame numbers, to frames that have existed longer.
8096 @kindex do @r{(@code{down})}
8098 Move @var{n} frames down the stack; @var{n} defaults to 1. For
8099 positive numbers @var{n}, this advances toward the innermost frame, to
8100 lower frame numbers, to frames that were created more recently.
8101 You may abbreviate @code{down} as @code{do}.
8104 All of these commands end by printing two lines of output describing the
8105 frame. The first line shows the frame number, the function name, the
8106 arguments, and the source file and line number of execution in that
8107 frame. The second line shows the text of that source line.
8115 #1 0x22f0 in main (argc=1, argv=0xf7fffbf4, env=0xf7fffbfc)
8117 10 read_input_file (argv[i]);
8121 After such a printout, the @code{list} command with no arguments
8122 prints ten lines centered on the point of execution in the frame.
8123 You can also edit the program at the point of execution with your favorite
8124 editing program by typing @code{edit}.
8125 @xref{List, ,Printing Source Lines},
8129 @kindex select-frame
8130 @item select-frame @r{[} @var{frame-selection-spec} @r{]}
8131 The @code{select-frame} command is a variant of @code{frame} that does
8132 not display the new frame after selecting it. This command is
8133 intended primarily for use in @value{GDBN} command scripts, where the
8134 output might be unnecessary and distracting. The
8135 @var{frame-selection-spec} is as for the @command{frame} command
8136 described in @ref{Selection, ,Selecting a Frame}.
8138 @kindex down-silently
8140 @item up-silently @var{n}
8141 @itemx down-silently @var{n}
8142 These two commands are variants of @code{up} and @code{down},
8143 respectively; they differ in that they do their work silently, without
8144 causing display of the new frame. They are intended primarily for use
8145 in @value{GDBN} command scripts, where the output might be unnecessary and
8150 @section Information About a Frame
8152 There are several other commands to print information about the selected
8158 When used without any argument, this command does not change which
8159 frame is selected, but prints a brief description of the currently
8160 selected stack frame. It can be abbreviated @code{f}. With an
8161 argument, this command is used to select a stack frame.
8162 @xref{Selection, ,Selecting a Frame}.
8165 @kindex info f @r{(@code{info frame})}
8168 This command prints a verbose description of the selected stack frame,
8173 the address of the frame
8175 the address of the next frame down (called by this frame)
8177 the address of the next frame up (caller of this frame)
8179 the language in which the source code corresponding to this frame is written
8181 the address of the frame's arguments
8183 the address of the frame's local variables
8185 the program counter saved in it (the address of execution in the caller frame)
8187 which registers were saved in the frame
8190 @noindent The verbose description is useful when
8191 something has gone wrong that has made the stack format fail to fit
8192 the usual conventions.
8194 @item info frame @r{[} @var{frame-selection-spec} @r{]}
8195 @itemx info f @r{[} @var{frame-selection-spec} @r{]}
8196 Print a verbose description of the frame selected by
8197 @var{frame-selection-spec}. The @var{frame-selection-spec} is the
8198 same as for the @command{frame} command (@pxref{Selection, ,Selecting
8199 a Frame}). The selected frame remains unchanged by this command.
8202 @item info args [-q]
8203 Print the arguments of the selected frame, each on a separate line.
8205 The optional flag @samp{-q}, which stands for @samp{quiet}, disables
8206 printing header information and messages explaining why no argument
8209 @item info args [-q] [-t @var{type_regexp}] [@var{regexp}]
8210 Like @kbd{info args}, but only print the arguments selected
8211 with the provided regexp(s).
8213 If @var{regexp} is provided, print only the arguments whose names
8214 match the regular expression @var{regexp}.
8216 If @var{type_regexp} is provided, print only the arguments whose
8217 types, as printed by the @code{whatis} command, match
8218 the regular expression @var{type_regexp}.
8219 If @var{type_regexp} contains space(s), it should be enclosed in
8220 quote characters. If needed, use backslash to escape the meaning
8221 of special characters or quotes.
8223 If both @var{regexp} and @var{type_regexp} are provided, an argument
8224 is printed only if its name matches @var{regexp} and its type matches
8227 @item info locals [-q]
8229 Print the local variables of the selected frame, each on a separate
8230 line. These are all variables (declared either static or automatic)
8231 accessible at the point of execution of the selected frame.
8233 The optional flag @samp{-q}, which stands for @samp{quiet}, disables
8234 printing header information and messages explaining why no local variables
8237 @item info locals [-q] [-t @var{type_regexp}] [@var{regexp}]
8238 Like @kbd{info locals}, but only print the local variables selected
8239 with the provided regexp(s).
8241 If @var{regexp} is provided, print only the local variables whose names
8242 match the regular expression @var{regexp}.
8244 If @var{type_regexp} is provided, print only the local variables whose
8245 types, as printed by the @code{whatis} command, match
8246 the regular expression @var{type_regexp}.
8247 If @var{type_regexp} contains space(s), it should be enclosed in
8248 quote characters. If needed, use backslash to escape the meaning
8249 of special characters or quotes.
8251 If both @var{regexp} and @var{type_regexp} are provided, a local variable
8252 is printed only if its name matches @var{regexp} and its type matches
8255 The command @kbd{info locals -q -t @var{type_regexp}} can usefully be
8256 combined with the commands @kbd{frame apply} and @kbd{thread apply}.
8257 For example, your program might use Resource Acquisition Is
8258 Initialization types (RAII) such as @code{lock_something_t}: each
8259 local variable of type @code{lock_something_t} automatically places a
8260 lock that is destroyed when the variable goes out of scope. You can
8261 then list all acquired locks in your program by doing
8263 thread apply all -s frame apply all -s info locals -q -t lock_something_t
8266 or the equivalent shorter form
8268 tfaas i lo -q -t lock_something_t
8274 @section Applying a Command to Several Frames.
8275 @anchor{frame apply}
8277 @cindex apply command to several frames
8279 @item frame apply [all | @var{count} | @var{-count} | level @var{level}@dots{}] [@var{option}]@dots{} @var{command}
8280 The @code{frame apply} command allows you to apply the named
8281 @var{command} to one or more frames.
8285 Specify @code{all} to apply @var{command} to all frames.
8288 Use @var{count} to apply @var{command} to the innermost @var{count}
8289 frames, where @var{count} is a positive number.
8292 Use @var{-count} to apply @var{command} to the outermost @var{count}
8293 frames, where @var{count} is a positive number.
8296 Use @code{level} to apply @var{command} to the set of frames identified
8297 by the @var{level} list. @var{level} is a frame level or a range of frame
8298 levels as @var{level1}-@var{level2}. The frame level is the number shown
8299 in the first field of the @samp{backtrace} command output.
8300 E.g., @samp{2-4 6-8 3} indicates to apply @var{command} for the frames
8301 at levels 2, 3, 4, 6, 7, 8, and then again on frame at level 3.
8305 Note that the frames on which @code{frame apply} applies a command are
8306 also influenced by the @code{set backtrace} settings such as @code{set
8307 backtrace past-main} and @code{set backtrace limit N}.
8308 @xref{Backtrace,,Backtraces}.
8310 The @code{frame apply} command also supports a number of options that
8311 allow overriding relevant @code{set backtrace} settings:
8314 @item -past-main [@code{on}|@code{off}]
8315 Whether backtraces should continue past @code{main}.
8316 Related setting: @ref{set backtrace past-main}.
8318 @item -past-entry [@code{on}|@code{off}]
8319 Whether backtraces should continue past the entry point of a program.
8320 Related setting: @ref{set backtrace past-entry}.
8323 By default, @value{GDBN} displays some frame information before the
8324 output produced by @var{command}, and an error raised during the
8325 execution of a @var{command} will abort @code{frame apply}. The
8326 following options can be used to fine-tune these behaviors:
8330 The flag @code{-c}, which stands for @samp{continue}, causes any
8331 errors in @var{command} to be displayed, and the execution of
8332 @code{frame apply} then continues.
8334 The flag @code{-s}, which stands for @samp{silent}, causes any errors
8335 or empty output produced by a @var{command} to be silently ignored.
8336 That is, the execution continues, but the frame information and errors
8339 The flag @code{-q} (@samp{quiet}) disables printing the frame
8343 The following example shows how the flags @code{-c} and @code{-s} are
8344 working when applying the command @code{p j} to all frames, where
8345 variable @code{j} can only be successfully printed in the outermost
8346 @code{#1 main} frame.
8350 (gdb) frame apply all p j
8351 #0 some_function (i=5) at fun.c:4
8352 No symbol "j" in current context.
8353 (gdb) frame apply all -c p j
8354 #0 some_function (i=5) at fun.c:4
8355 No symbol "j" in current context.
8356 #1 0x565555fb in main (argc=1, argv=0xffffd2c4) at fun.c:11
8358 (gdb) frame apply all -s p j
8359 #1 0x565555fb in main (argc=1, argv=0xffffd2c4) at fun.c:11
8365 By default, @samp{frame apply}, prints the frame location
8366 information before the command output:
8370 (gdb) frame apply all p $sp
8371 #0 some_function (i=5) at fun.c:4
8372 $4 = (void *) 0xffffd1e0
8373 #1 0x565555fb in main (argc=1, argv=0xffffd2c4) at fun.c:11
8374 $5 = (void *) 0xffffd1f0
8379 If the flag @code{-q} is given, no frame information is printed:
8382 (gdb) frame apply all -q p $sp
8383 $12 = (void *) 0xffffd1e0
8384 $13 = (void *) 0xffffd1f0
8394 @cindex apply a command to all frames (ignoring errors and empty output)
8395 @item faas @var{command}
8396 Shortcut for @code{frame apply all -s @var{command}}.
8397 Applies @var{command} on all frames, ignoring errors and empty output.
8399 It can for example be used to print a local variable or a function
8400 argument without knowing the frame where this variable or argument
8403 (@value{GDBP}) faas p some_local_var_i_do_not_remember_where_it_is
8406 The @code{faas} command accepts the same options as the @code{frame
8407 apply} command. @xref{frame apply}.
8409 Note that the command @code{tfaas @var{command}} applies @var{command}
8410 on all frames of all threads. See @xref{Threads,,Threads}.
8414 @node Frame Filter Management
8415 @section Management of Frame Filters.
8416 @cindex managing frame filters
8418 Frame filters are Python based utilities to manage and decorate the
8419 output of frames. @xref{Frame Filter API}, for further information.
8421 Managing frame filters is performed by several commands available
8422 within @value{GDBN}, detailed here.
8425 @kindex info frame-filter
8426 @item info frame-filter
8427 Print a list of installed frame filters from all dictionaries, showing
8428 their name, priority and enabled status.
8430 @kindex disable frame-filter
8431 @anchor{disable frame-filter all}
8432 @item disable frame-filter @var{filter-dictionary} @var{filter-name}
8433 Disable a frame filter in the dictionary matching
8434 @var{filter-dictionary} and @var{filter-name}. The
8435 @var{filter-dictionary} may be @code{all}, @code{global},
8436 @code{progspace}, or the name of the object file where the frame filter
8437 dictionary resides. When @code{all} is specified, all frame filters
8438 across all dictionaries are disabled. The @var{filter-name} is the name
8439 of the frame filter and is used when @code{all} is not the option for
8440 @var{filter-dictionary}. A disabled frame-filter is not deleted, it
8441 may be enabled again later.
8443 @kindex enable frame-filter
8444 @item enable frame-filter @var{filter-dictionary} @var{filter-name}
8445 Enable a frame filter in the dictionary matching
8446 @var{filter-dictionary} and @var{filter-name}. The
8447 @var{filter-dictionary} may be @code{all}, @code{global},
8448 @code{progspace} or the name of the object file where the frame filter
8449 dictionary resides. When @code{all} is specified, all frame filters across
8450 all dictionaries are enabled. The @var{filter-name} is the name of the frame
8451 filter and is used when @code{all} is not the option for
8452 @var{filter-dictionary}.
8457 (gdb) info frame-filter
8459 global frame-filters:
8460 Priority Enabled Name
8461 1000 No PrimaryFunctionFilter
8464 progspace /build/test frame-filters:
8465 Priority Enabled Name
8466 100 Yes ProgspaceFilter
8468 objfile /build/test frame-filters:
8469 Priority Enabled Name
8470 999 Yes BuildProgramFilter
8472 (gdb) disable frame-filter /build/test BuildProgramFilter
8473 (gdb) info frame-filter
8475 global frame-filters:
8476 Priority Enabled Name
8477 1000 No PrimaryFunctionFilter
8480 progspace /build/test frame-filters:
8481 Priority Enabled Name
8482 100 Yes ProgspaceFilter
8484 objfile /build/test frame-filters:
8485 Priority Enabled Name
8486 999 No BuildProgramFilter
8488 (gdb) enable frame-filter global PrimaryFunctionFilter
8489 (gdb) info frame-filter
8491 global frame-filters:
8492 Priority Enabled Name
8493 1000 Yes PrimaryFunctionFilter
8496 progspace /build/test frame-filters:
8497 Priority Enabled Name
8498 100 Yes ProgspaceFilter
8500 objfile /build/test frame-filters:
8501 Priority Enabled Name
8502 999 No BuildProgramFilter
8505 @kindex set frame-filter priority
8506 @item set frame-filter priority @var{filter-dictionary} @var{filter-name} @var{priority}
8507 Set the @var{priority} of a frame filter in the dictionary matching
8508 @var{filter-dictionary}, and the frame filter name matching
8509 @var{filter-name}. The @var{filter-dictionary} may be @code{global},
8510 @code{progspace} or the name of the object file where the frame filter
8511 dictionary resides. The @var{priority} is an integer.
8513 @kindex show frame-filter priority
8514 @item show frame-filter priority @var{filter-dictionary} @var{filter-name}
8515 Show the @var{priority} of a frame filter in the dictionary matching
8516 @var{filter-dictionary}, and the frame filter name matching
8517 @var{filter-name}. The @var{filter-dictionary} may be @code{global},
8518 @code{progspace} or the name of the object file where the frame filter
8524 (gdb) info frame-filter
8526 global frame-filters:
8527 Priority Enabled Name
8528 1000 Yes PrimaryFunctionFilter
8531 progspace /build/test frame-filters:
8532 Priority Enabled Name
8533 100 Yes ProgspaceFilter
8535 objfile /build/test frame-filters:
8536 Priority Enabled Name
8537 999 No BuildProgramFilter
8539 (gdb) set frame-filter priority global Reverse 50
8540 (gdb) info frame-filter
8542 global frame-filters:
8543 Priority Enabled Name
8544 1000 Yes PrimaryFunctionFilter
8547 progspace /build/test frame-filters:
8548 Priority Enabled Name
8549 100 Yes ProgspaceFilter
8551 objfile /build/test frame-filters:
8552 Priority Enabled Name
8553 999 No BuildProgramFilter
8558 @chapter Examining Source Files
8560 @value{GDBN} can print parts of your program's source, since the debugging
8561 information recorded in the program tells @value{GDBN} what source files were
8562 used to build it. When your program stops, @value{GDBN} spontaneously prints
8563 the line where it stopped. Likewise, when you select a stack frame
8564 (@pxref{Selection, ,Selecting a Frame}), @value{GDBN} prints the line where
8565 execution in that frame has stopped. You can print other portions of
8566 source files by explicit command.
8568 If you use @value{GDBN} through its @sc{gnu} Emacs interface, you may
8569 prefer to use Emacs facilities to view source; see @ref{Emacs, ,Using
8570 @value{GDBN} under @sc{gnu} Emacs}.
8573 * List:: Printing source lines
8574 * Specify Location:: How to specify code locations
8575 * Edit:: Editing source files
8576 * Search:: Searching source files
8577 * Source Path:: Specifying source directories
8578 * Machine Code:: Source and machine code
8582 @section Printing Source Lines
8585 @kindex l @r{(@code{list})}
8586 To print lines from a source file, use the @code{list} command
8587 (abbreviated @code{l}). By default, ten lines are printed.
8588 There are several ways to specify what part of the file you want to
8589 print; see @ref{Specify Location}, for the full list.
8591 Here are the forms of the @code{list} command most commonly used:
8594 @item list @var{linenum}
8595 Print lines centered around line number @var{linenum} in the
8596 current source file.
8598 @item list @var{function}
8599 Print lines centered around the beginning of function
8603 Print more lines. If the last lines printed were printed with a
8604 @code{list} command, this prints lines following the last lines
8605 printed; however, if the last line printed was a solitary line printed
8606 as part of displaying a stack frame (@pxref{Stack, ,Examining the
8607 Stack}), this prints lines centered around that line.
8610 Print lines just before the lines last printed.
8613 @cindex @code{list}, how many lines to display
8614 By default, @value{GDBN} prints ten source lines with any of these forms of
8615 the @code{list} command. You can change this using @code{set listsize}:
8618 @kindex set listsize
8619 @item set listsize @var{count}
8620 @itemx set listsize unlimited
8621 Make the @code{list} command display @var{count} source lines (unless
8622 the @code{list} argument explicitly specifies some other number).
8623 Setting @var{count} to @code{unlimited} or 0 means there's no limit.
8625 @kindex show listsize
8627 Display the number of lines that @code{list} prints.
8630 Repeating a @code{list} command with @key{RET} discards the argument,
8631 so it is equivalent to typing just @code{list}. This is more useful
8632 than listing the same lines again. An exception is made for an
8633 argument of @samp{-}; that argument is preserved in repetition so that
8634 each repetition moves up in the source file.
8636 In general, the @code{list} command expects you to supply zero, one or two
8637 @dfn{locations}. Locations specify source lines; there are several ways
8638 of writing them (@pxref{Specify Location}), but the effect is always
8639 to specify some source line.
8641 Here is a complete description of the possible arguments for @code{list}:
8644 @item list @var{location}
8645 Print lines centered around the line specified by @var{location}.
8647 @item list @var{first},@var{last}
8648 Print lines from @var{first} to @var{last}. Both arguments are
8649 locations. When a @code{list} command has two locations, and the
8650 source file of the second location is omitted, this refers to
8651 the same source file as the first location.
8653 @item list ,@var{last}
8654 Print lines ending with @var{last}.
8656 @item list @var{first},
8657 Print lines starting with @var{first}.
8660 Print lines just after the lines last printed.
8663 Print lines just before the lines last printed.
8666 As described in the preceding table.
8669 @node Specify Location
8670 @section Specifying a Location
8671 @cindex specifying location
8673 @cindex source location
8676 * Linespec Locations:: Linespec locations
8677 * Explicit Locations:: Explicit locations
8678 * Address Locations:: Address locations
8681 Several @value{GDBN} commands accept arguments that specify a location
8682 of your program's code. Since @value{GDBN} is a source-level
8683 debugger, a location usually specifies some line in the source code.
8684 Locations may be specified using three different formats:
8685 linespec locations, explicit locations, or address locations.
8687 @node Linespec Locations
8688 @subsection Linespec Locations
8689 @cindex linespec locations
8691 A @dfn{linespec} is a colon-separated list of source location parameters such
8692 as file name, function name, etc. Here are all the different ways of
8693 specifying a linespec:
8697 Specifies the line number @var{linenum} of the current source file.
8700 @itemx +@var{offset}
8701 Specifies the line @var{offset} lines before or after the @dfn{current
8702 line}. For the @code{list} command, the current line is the last one
8703 printed; for the breakpoint commands, this is the line at which
8704 execution stopped in the currently selected @dfn{stack frame}
8705 (@pxref{Frames, ,Frames}, for a description of stack frames.) When
8706 used as the second of the two linespecs in a @code{list} command,
8707 this specifies the line @var{offset} lines up or down from the first
8710 @item @var{filename}:@var{linenum}
8711 Specifies the line @var{linenum} in the source file @var{filename}.
8712 If @var{filename} is a relative file name, then it will match any
8713 source file name with the same trailing components. For example, if
8714 @var{filename} is @samp{gcc/expr.c}, then it will match source file
8715 name of @file{/build/trunk/gcc/expr.c}, but not
8716 @file{/build/trunk/libcpp/expr.c} or @file{/build/trunk/gcc/x-expr.c}.
8718 @item @var{function}
8719 Specifies the line that begins the body of the function @var{function}.
8720 For example, in C, this is the line with the open brace.
8722 By default, in C@t{++} and Ada, @var{function} is interpreted as
8723 specifying all functions named @var{function} in all scopes. For
8724 C@t{++}, this means in all namespaces and classes. For Ada, this
8725 means in all packages.
8727 For example, assuming a program with C@t{++} symbols named
8728 @code{A::B::func} and @code{B::func}, both commands @w{@kbd{break
8729 func}} and @w{@kbd{break B::func}} set a breakpoint on both symbols.
8731 Commands that accept a linespec let you override this with the
8732 @code{-qualified} option. For example, @w{@kbd{break -qualified
8733 func}} sets a breakpoint on a free-function named @code{func} ignoring
8734 any C@t{++} class methods and namespace functions called @code{func}.
8736 @xref{Explicit Locations}.
8738 @item @var{function}:@var{label}
8739 Specifies the line where @var{label} appears in @var{function}.
8741 @item @var{filename}:@var{function}
8742 Specifies the line that begins the body of the function @var{function}
8743 in the file @var{filename}. You only need the file name with a
8744 function name to avoid ambiguity when there are identically named
8745 functions in different source files.
8748 Specifies the line at which the label named @var{label} appears
8749 in the function corresponding to the currently selected stack frame.
8750 If there is no current selected stack frame (for instance, if the inferior
8751 is not running), then @value{GDBN} will not search for a label.
8753 @cindex breakpoint at static probe point
8754 @item -pstap|-probe-stap @r{[}@var{objfile}:@r{[}@var{provider}:@r{]}@r{]}@var{name}
8755 The @sc{gnu}/Linux tool @code{SystemTap} provides a way for
8756 applications to embed static probes. @xref{Static Probe Points}, for more
8757 information on finding and using static probes. This form of linespec
8758 specifies the location of such a static probe.
8760 If @var{objfile} is given, only probes coming from that shared library
8761 or executable matching @var{objfile} as a regular expression are considered.
8762 If @var{provider} is given, then only probes from that provider are considered.
8763 If several probes match the spec, @value{GDBN} will insert a breakpoint at
8764 each one of those probes.
8767 @node Explicit Locations
8768 @subsection Explicit Locations
8769 @cindex explicit locations
8771 @dfn{Explicit locations} allow the user to directly specify the source
8772 location's parameters using option-value pairs.
8774 Explicit locations are useful when several functions, labels, or
8775 file names have the same name (base name for files) in the program's
8776 sources. In these cases, explicit locations point to the source
8777 line you meant more accurately and unambiguously. Also, using
8778 explicit locations might be faster in large programs.
8780 For example, the linespec @samp{foo:bar} may refer to a function @code{bar}
8781 defined in the file named @file{foo} or the label @code{bar} in a function
8782 named @code{foo}. @value{GDBN} must search either the file system or
8783 the symbol table to know.
8785 The list of valid explicit location options is summarized in the
8789 @item -source @var{filename}
8790 The value specifies the source file name. To differentiate between
8791 files with the same base name, prepend as many directories as is necessary
8792 to uniquely identify the desired file, e.g., @file{foo/bar/baz.c}. Otherwise
8793 @value{GDBN} will use the first file it finds with the given base
8794 name. This option requires the use of either @code{-function} or @code{-line}.
8796 @item -function @var{function}
8797 The value specifies the name of a function. Operations
8798 on function locations unmodified by other options (such as @code{-label}
8799 or @code{-line}) refer to the line that begins the body of the function.
8800 In C, for example, this is the line with the open brace.
8802 By default, in C@t{++} and Ada, @var{function} is interpreted as
8803 specifying all functions named @var{function} in all scopes. For
8804 C@t{++}, this means in all namespaces and classes. For Ada, this
8805 means in all packages.
8807 For example, assuming a program with C@t{++} symbols named
8808 @code{A::B::func} and @code{B::func}, both commands @w{@kbd{break
8809 -function func}} and @w{@kbd{break -function B::func}} set a
8810 breakpoint on both symbols.
8812 You can use the @kbd{-qualified} flag to override this (see below).
8816 This flag makes @value{GDBN} interpret a function name specified with
8817 @kbd{-function} as a complete fully-qualified name.
8819 For example, assuming a C@t{++} program with symbols named
8820 @code{A::B::func} and @code{B::func}, the @w{@kbd{break -qualified
8821 -function B::func}} command sets a breakpoint on @code{B::func}, only.
8823 (Note: the @kbd{-qualified} option can precede a linespec as well
8824 (@pxref{Linespec Locations}), so the particular example above could be
8825 simplified as @w{@kbd{break -qualified B::func}}.)
8827 @item -label @var{label}
8828 The value specifies the name of a label. When the function
8829 name is not specified, the label is searched in the function of the currently
8830 selected stack frame.
8832 @item -line @var{number}
8833 The value specifies a line offset for the location. The offset may either
8834 be absolute (@code{-line 3}) or relative (@code{-line +3}), depending on
8835 the command. When specified without any other options, the line offset is
8836 relative to the current line.
8839 Explicit location options may be abbreviated by omitting any non-unique
8840 trailing characters from the option name, e.g., @w{@kbd{break -s main.c -li 3}}.
8842 @node Address Locations
8843 @subsection Address Locations
8844 @cindex address locations
8846 @dfn{Address locations} indicate a specific program address. They have
8847 the generalized form *@var{address}.
8849 For line-oriented commands, such as @code{list} and @code{edit}, this
8850 specifies a source line that contains @var{address}. For @code{break} and
8851 other breakpoint-oriented commands, this can be used to set breakpoints in
8852 parts of your program which do not have debugging information or
8855 Here @var{address} may be any expression valid in the current working
8856 language (@pxref{Languages, working language}) that specifies a code
8857 address. In addition, as a convenience, @value{GDBN} extends the
8858 semantics of expressions used in locations to cover several situations
8859 that frequently occur during debugging. Here are the various forms
8863 @item @var{expression}
8864 Any expression valid in the current working language.
8866 @item @var{funcaddr}
8867 An address of a function or procedure derived from its name. In C,
8868 C@t{++}, Objective-C, Fortran, minimal, and assembly, this is
8869 simply the function's name @var{function} (and actually a special case
8870 of a valid expression). In Pascal and Modula-2, this is
8871 @code{&@var{function}}. In Ada, this is @code{@var{function}'Address}
8872 (although the Pascal form also works).
8874 This form specifies the address of the function's first instruction,
8875 before the stack frame and arguments have been set up.
8877 @item '@var{filename}':@var{funcaddr}
8878 Like @var{funcaddr} above, but also specifies the name of the source
8879 file explicitly. This is useful if the name of the function does not
8880 specify the function unambiguously, e.g., if there are several
8881 functions with identical names in different source files.
8885 @section Editing Source Files
8886 @cindex editing source files
8889 @kindex e @r{(@code{edit})}
8890 To edit the lines in a source file, use the @code{edit} command.
8891 The editing program of your choice
8892 is invoked with the current line set to
8893 the active line in the program.
8894 Alternatively, there are several ways to specify what part of the file you
8895 want to print if you want to see other parts of the program:
8898 @item edit @var{location}
8899 Edit the source file specified by @code{location}. Editing starts at
8900 that @var{location}, e.g., at the specified source line of the
8901 specified file. @xref{Specify Location}, for all the possible forms
8902 of the @var{location} argument; here are the forms of the @code{edit}
8903 command most commonly used:
8906 @item edit @var{number}
8907 Edit the current source file with @var{number} as the active line number.
8909 @item edit @var{function}
8910 Edit the file containing @var{function} at the beginning of its definition.
8915 @subsection Choosing your Editor
8916 You can customize @value{GDBN} to use any editor you want
8918 The only restriction is that your editor (say @code{ex}), recognizes the
8919 following command-line syntax:
8921 ex +@var{number} file
8923 The optional numeric value +@var{number} specifies the number of the line in
8924 the file where to start editing.}.
8925 By default, it is @file{@value{EDITOR}}, but you can change this
8926 by setting the environment variable @code{EDITOR} before using
8927 @value{GDBN}. For example, to configure @value{GDBN} to use the
8928 @code{vi} editor, you could use these commands with the @code{sh} shell:
8934 or in the @code{csh} shell,
8936 setenv EDITOR /usr/bin/vi
8941 @section Searching Source Files
8942 @cindex searching source files
8944 There are two commands for searching through the current source file for a
8949 @kindex forward-search
8950 @kindex fo @r{(@code{forward-search})}
8951 @item forward-search @var{regexp}
8952 @itemx search @var{regexp}
8953 The command @samp{forward-search @var{regexp}} checks each line,
8954 starting with the one following the last line listed, for a match for
8955 @var{regexp}. It lists the line that is found. You can use the
8956 synonym @samp{search @var{regexp}} or abbreviate the command name as
8959 @kindex reverse-search
8960 @item reverse-search @var{regexp}
8961 The command @samp{reverse-search @var{regexp}} checks each line, starting
8962 with the one before the last line listed and going backward, for a match
8963 for @var{regexp}. It lists the line that is found. You can abbreviate
8964 this command as @code{rev}.
8968 @section Specifying Source Directories
8971 @cindex directories for source files
8972 Executable programs sometimes do not record the directories of the source
8973 files from which they were compiled, just the names. Even when they do,
8974 the directories could be moved between the compilation and your debugging
8975 session. @value{GDBN} has a list of directories to search for source files;
8976 this is called the @dfn{source path}. Each time @value{GDBN} wants a source file,
8977 it tries all the directories in the list, in the order they are present
8978 in the list, until it finds a file with the desired name.
8980 For example, suppose an executable references the file
8981 @file{/usr/src/foo-1.0/lib/foo.c}, does not record a compilation
8982 directory, and the @dfn{source path} is @file{/mnt/cross}.
8983 @value{GDBN} would look for the source file in the following
8988 @item @file{/usr/src/foo-1.0/lib/foo.c}
8989 @item @file{/mnt/cross/usr/src/foo-1.0/lib/foo.c}
8990 @item @file{/mnt/cross/foo.c}
8994 If the source file is not present at any of the above locations then
8995 an error is printed. @value{GDBN} does not look up the parts of the
8996 source file name, such as @file{/mnt/cross/src/foo-1.0/lib/foo.c}.
8997 Likewise, the subdirectories of the source path are not searched: if
8998 the source path is @file{/mnt/cross}, and the binary refers to
8999 @file{foo.c}, @value{GDBN} would not find it under
9000 @file{/mnt/cross/usr/src/foo-1.0/lib}.
9002 Plain file names, relative file names with leading directories, file
9003 names containing dots, etc.@: are all treated as described above,
9004 except that non-absolute file names are not looked up literally. If
9005 the @dfn{source path} is @file{/mnt/cross}, the source file is
9006 recorded as @file{../lib/foo.c}, and no compilation directory is
9007 recorded, then @value{GDBN} will search in the following locations:
9011 @item @file{/mnt/cross/../lib/foo.c}
9012 @item @file{/mnt/cross/foo.c}
9018 @vindex $cdir@r{, convenience variable}
9019 @vindex $cwd@r{, convenience variable}
9020 @cindex compilation directory
9021 @cindex current directory
9022 @cindex working directory
9023 @cindex directory, current
9024 @cindex directory, compilation
9025 The @dfn{source path} will always include two special entries
9026 @samp{$cdir} and @samp{$cwd}, these refer to the compilation directory
9027 (if one is recorded) and the current working directory respectively.
9029 @samp{$cdir} causes @value{GDBN} to search within the compilation
9030 directory, if one is recorded in the debug information. If no
9031 compilation directory is recorded in the debug information then
9032 @samp{$cdir} is ignored.
9034 @samp{$cwd} is not the same as @samp{.}---the former tracks the
9035 current working directory as it changes during your @value{GDBN}
9036 session, while the latter is immediately expanded to the current
9037 directory at the time you add an entry to the source path.
9039 If a compilation directory is recorded in the debug information, and
9040 @value{GDBN} has not found the source file after the first search
9041 using @dfn{source path}, then @value{GDBN} will combine the
9042 compilation directory and the filename, and then search for the source
9043 file again using the @dfn{source path}.
9045 For example, if the executable records the source file as
9046 @file{/usr/src/foo-1.0/lib/foo.c}, the compilation directory is
9047 recorded as @file{/project/build}, and the @dfn{source path} is
9048 @file{/mnt/cross:$cdir:$cwd} while the current working directory of
9049 the @value{GDBN} session is @file{/home/user}, then @value{GDBN} will
9050 search for the source file in the following locations:
9054 @item @file{/usr/src/foo-1.0/lib/foo.c}
9055 @item @file{/mnt/cross/usr/src/foo-1.0/lib/foo.c}
9056 @item @file{/project/build/usr/src/foo-1.0/lib/foo.c}
9057 @item @file{/home/user/usr/src/foo-1.0/lib/foo.c}
9058 @item @file{/mnt/cross/project/build/usr/src/foo-1.0/lib/foo.c}
9059 @item @file{/project/build/project/build/usr/src/foo-1.0/lib/foo.c}
9060 @item @file{/home/user/project/build/usr/src/foo-1.0/lib/foo.c}
9061 @item @file{/mnt/cross/foo.c}
9062 @item @file{/project/build/foo.c}
9063 @item @file{/home/user/foo.c}
9067 If the file name in the previous example had been recorded in the
9068 executable as a relative path rather than an absolute path, then the
9069 first look up would not have occurred, but all of the remaining steps
9072 When searching for source files on MS-DOS and MS-Windows, where
9073 absolute paths start with a drive letter (e.g.
9074 @file{C:/project/foo.c}), @value{GDBN} will remove the drive letter
9075 from the file name before appending it to a search directory from
9076 @dfn{source path}; for instance if the executable references the
9077 source file @file{C:/project/foo.c} and @dfn{source path} is set to
9078 @file{D:/mnt/cross}, then @value{GDBN} will search in the following
9079 locations for the source file:
9083 @item @file{C:/project/foo.c}
9084 @item @file{D:/mnt/cross/project/foo.c}
9085 @item @file{D:/mnt/cross/foo.c}
9089 Note that the executable search path is @emph{not} used to locate the
9092 Whenever you reset or rearrange the source path, @value{GDBN} clears out
9093 any information it has cached about where source files are found and where
9094 each line is in the file.
9098 When you start @value{GDBN}, its source path includes only @samp{$cdir}
9099 and @samp{$cwd}, in that order.
9100 To add other directories, use the @code{directory} command.
9102 The search path is used to find both program source files and @value{GDBN}
9103 script files (read using the @samp{-command} option and @samp{source} command).
9105 In addition to the source path, @value{GDBN} provides a set of commands
9106 that manage a list of source path substitution rules. A @dfn{substitution
9107 rule} specifies how to rewrite source directories stored in the program's
9108 debug information in case the sources were moved to a different
9109 directory between compilation and debugging. A rule is made of
9110 two strings, the first specifying what needs to be rewritten in
9111 the path, and the second specifying how it should be rewritten.
9112 In @ref{set substitute-path}, we name these two parts @var{from} and
9113 @var{to} respectively. @value{GDBN} does a simple string replacement
9114 of @var{from} with @var{to} at the start of the directory part of the
9115 source file name, and uses that result instead of the original file
9116 name to look up the sources.
9118 Using the previous example, suppose the @file{foo-1.0} tree has been
9119 moved from @file{/usr/src} to @file{/mnt/cross}, then you can tell
9120 @value{GDBN} to replace @file{/usr/src} in all source path names with
9121 @file{/mnt/cross}. The first lookup will then be
9122 @file{/mnt/cross/foo-1.0/lib/foo.c} in place of the original location
9123 of @file{/usr/src/foo-1.0/lib/foo.c}. To define a source path
9124 substitution rule, use the @code{set substitute-path} command
9125 (@pxref{set substitute-path}).
9127 To avoid unexpected substitution results, a rule is applied only if the
9128 @var{from} part of the directory name ends at a directory separator.
9129 For instance, a rule substituting @file{/usr/source} into
9130 @file{/mnt/cross} will be applied to @file{/usr/source/foo-1.0} but
9131 not to @file{/usr/sourceware/foo-2.0}. And because the substitution
9132 is applied only at the beginning of the directory name, this rule will
9133 not be applied to @file{/root/usr/source/baz.c} either.
9135 In many cases, you can achieve the same result using the @code{directory}
9136 command. However, @code{set substitute-path} can be more efficient in
9137 the case where the sources are organized in a complex tree with multiple
9138 subdirectories. With the @code{directory} command, you need to add each
9139 subdirectory of your project. If you moved the entire tree while
9140 preserving its internal organization, then @code{set substitute-path}
9141 allows you to direct the debugger to all the sources with one single
9144 @code{set substitute-path} is also more than just a shortcut command.
9145 The source path is only used if the file at the original location no
9146 longer exists. On the other hand, @code{set substitute-path} modifies
9147 the debugger behavior to look at the rewritten location instead. So, if
9148 for any reason a source file that is not relevant to your executable is
9149 located at the original location, a substitution rule is the only
9150 method available to point @value{GDBN} at the new location.
9152 @cindex @samp{--with-relocated-sources}
9153 @cindex default source path substitution
9154 You can configure a default source path substitution rule by
9155 configuring @value{GDBN} with the
9156 @samp{--with-relocated-sources=@var{dir}} option. The @var{dir}
9157 should be the name of a directory under @value{GDBN}'s configured
9158 prefix (set with @samp{--prefix} or @samp{--exec-prefix}), and
9159 directory names in debug information under @var{dir} will be adjusted
9160 automatically if the installed @value{GDBN} is moved to a new
9161 location. This is useful if @value{GDBN}, libraries or executables
9162 with debug information and corresponding source code are being moved
9166 @item directory @var{dirname} @dots{}
9167 @item dir @var{dirname} @dots{}
9168 Add directory @var{dirname} to the front of the source path. Several
9169 directory names may be given to this command, separated by @samp{:}
9170 (@samp{;} on MS-DOS and MS-Windows, where @samp{:} usually appears as
9171 part of absolute file names) or
9172 whitespace. You may specify a directory that is already in the source
9173 path; this moves it forward, so @value{GDBN} searches it sooner.
9175 The special strings @samp{$cdir} (to refer to the compilation
9176 directory, if one is recorded), and @samp{$cwd} (to refer to the
9177 current working directory) can also be included in the list of
9178 directories @var{dirname}. Though these will already be in the source
9179 path they will be moved forward in the list so @value{GDBN} searches
9183 Reset the source path to its default value (@samp{$cdir:$cwd} on Unix systems). This requires confirmation.
9185 @c RET-repeat for @code{directory} is explicitly disabled, but since
9186 @c repeating it would be a no-op we do not say that. (thanks to RMS)
9188 @item set directories @var{path-list}
9189 @kindex set directories
9190 Set the source path to @var{path-list}.
9191 @samp{$cdir:$cwd} are added if missing.
9193 @item show directories
9194 @kindex show directories
9195 Print the source path: show which directories it contains.
9197 @anchor{set substitute-path}
9198 @item set substitute-path @var{from} @var{to}
9199 @kindex set substitute-path
9200 Define a source path substitution rule, and add it at the end of the
9201 current list of existing substitution rules. If a rule with the same
9202 @var{from} was already defined, then the old rule is also deleted.
9204 For example, if the file @file{/foo/bar/baz.c} was moved to
9205 @file{/mnt/cross/baz.c}, then the command
9208 (@value{GDBP}) set substitute-path /foo/bar /mnt/cross
9212 will tell @value{GDBN} to replace @samp{/foo/bar} with
9213 @samp{/mnt/cross}, which will allow @value{GDBN} to find the file
9214 @file{baz.c} even though it was moved.
9216 In the case when more than one substitution rule have been defined,
9217 the rules are evaluated one by one in the order where they have been
9218 defined. The first one matching, if any, is selected to perform
9221 For instance, if we had entered the following commands:
9224 (@value{GDBP}) set substitute-path /usr/src/include /mnt/include
9225 (@value{GDBP}) set substitute-path /usr/src /mnt/src
9229 @value{GDBN} would then rewrite @file{/usr/src/include/defs.h} into
9230 @file{/mnt/include/defs.h} by using the first rule. However, it would
9231 use the second rule to rewrite @file{/usr/src/lib/foo.c} into
9232 @file{/mnt/src/lib/foo.c}.
9235 @item unset substitute-path [path]
9236 @kindex unset substitute-path
9237 If a path is specified, search the current list of substitution rules
9238 for a rule that would rewrite that path. Delete that rule if found.
9239 A warning is emitted by the debugger if no rule could be found.
9241 If no path is specified, then all substitution rules are deleted.
9243 @item show substitute-path [path]
9244 @kindex show substitute-path
9245 If a path is specified, then print the source path substitution rule
9246 which would rewrite that path, if any.
9248 If no path is specified, then print all existing source path substitution
9253 If your source path is cluttered with directories that are no longer of
9254 interest, @value{GDBN} may sometimes cause confusion by finding the wrong
9255 versions of source. You can correct the situation as follows:
9259 Use @code{directory} with no argument to reset the source path to its default value.
9262 Use @code{directory} with suitable arguments to reinstall the
9263 directories you want in the source path. You can add all the
9264 directories in one command.
9268 @section Source and Machine Code
9269 @cindex source line and its code address
9271 You can use the command @code{info line} to map source lines to program
9272 addresses (and vice versa), and the command @code{disassemble} to display
9273 a range of addresses as machine instructions. You can use the command
9274 @code{set disassemble-next-line} to set whether to disassemble next
9275 source line when execution stops. When run under @sc{gnu} Emacs
9276 mode, the @code{info line} command causes the arrow to point to the
9277 line specified. Also, @code{info line} prints addresses in symbolic form as
9283 @itemx info line @var{location}
9284 Print the starting and ending addresses of the compiled code for
9285 source line @var{location}. You can specify source lines in any of
9286 the ways documented in @ref{Specify Location}. With no @var{location}
9287 information about the current source line is printed.
9290 For example, we can use @code{info line} to discover the location of
9291 the object code for the first line of function
9292 @code{m4_changequote}:
9295 (@value{GDBP}) info line m4_changequote
9296 Line 895 of "builtin.c" starts at pc 0x634c <m4_changequote> and \
9297 ends at 0x6350 <m4_changequote+4>.
9301 @cindex code address and its source line
9302 We can also inquire (using @code{*@var{addr}} as the form for
9303 @var{location}) what source line covers a particular address:
9305 (@value{GDBP}) info line *0x63ff
9306 Line 926 of "builtin.c" starts at pc 0x63e4 <m4_changequote+152> and \
9307 ends at 0x6404 <m4_changequote+184>.
9310 @cindex @code{$_} and @code{info line}
9311 @cindex @code{x} command, default address
9312 @kindex x@r{(examine), and} info line
9313 After @code{info line}, the default address for the @code{x} command
9314 is changed to the starting address of the line, so that @samp{x/i} is
9315 sufficient to begin examining the machine code (@pxref{Memory,
9316 ,Examining Memory}). Also, this address is saved as the value of the
9317 convenience variable @code{$_} (@pxref{Convenience Vars, ,Convenience
9320 @cindex info line, repeated calls
9321 After @code{info line}, using @code{info line} again without
9322 specifying a location will display information about the next source
9327 @cindex assembly instructions
9328 @cindex instructions, assembly
9329 @cindex machine instructions
9330 @cindex listing machine instructions
9332 @itemx disassemble /m
9333 @itemx disassemble /s
9334 @itemx disassemble /r
9335 This specialized command dumps a range of memory as machine
9336 instructions. It can also print mixed source+disassembly by specifying
9337 the @code{/m} or @code{/s} modifier and print the raw instructions in hex
9338 as well as in symbolic form by specifying the @code{/r} modifier.
9339 The default memory range is the function surrounding the
9340 program counter of the selected frame. A single argument to this
9341 command is a program counter value; @value{GDBN} dumps the function
9342 surrounding this value. When two arguments are given, they should
9343 be separated by a comma, possibly surrounded by whitespace. The
9344 arguments specify a range of addresses to dump, in one of two forms:
9347 @item @var{start},@var{end}
9348 the addresses from @var{start} (inclusive) to @var{end} (exclusive)
9349 @item @var{start},+@var{length}
9350 the addresses from @var{start} (inclusive) to
9351 @code{@var{start}+@var{length}} (exclusive).
9355 When 2 arguments are specified, the name of the function is also
9356 printed (since there could be several functions in the given range).
9358 The argument(s) can be any expression yielding a numeric value, such as
9359 @samp{0x32c4}, @samp{&main+10} or @samp{$pc - 8}.
9361 If the range of memory being disassembled contains current program counter,
9362 the instruction at that location is shown with a @code{=>} marker.
9365 The following example shows the disassembly of a range of addresses of
9366 HP PA-RISC 2.0 code:
9369 (@value{GDBP}) disas 0x32c4, 0x32e4
9370 Dump of assembler code from 0x32c4 to 0x32e4:
9371 0x32c4 <main+204>: addil 0,dp
9372 0x32c8 <main+208>: ldw 0x22c(sr0,r1),r26
9373 0x32cc <main+212>: ldil 0x3000,r31
9374 0x32d0 <main+216>: ble 0x3f8(sr4,r31)
9375 0x32d4 <main+220>: ldo 0(r31),rp
9376 0x32d8 <main+224>: addil -0x800,dp
9377 0x32dc <main+228>: ldo 0x588(r1),r26
9378 0x32e0 <main+232>: ldil 0x3000,r31
9379 End of assembler dump.
9382 Here is an example showing mixed source+assembly for Intel x86
9383 with @code{/m} or @code{/s}, when the program is stopped just after
9384 function prologue in a non-optimized function with no inline code.
9387 (@value{GDBP}) disas /m main
9388 Dump of assembler code for function main:
9390 0x08048330 <+0>: push %ebp
9391 0x08048331 <+1>: mov %esp,%ebp
9392 0x08048333 <+3>: sub $0x8,%esp
9393 0x08048336 <+6>: and $0xfffffff0,%esp
9394 0x08048339 <+9>: sub $0x10,%esp
9396 6 printf ("Hello.\n");
9397 => 0x0804833c <+12>: movl $0x8048440,(%esp)
9398 0x08048343 <+19>: call 0x8048284 <puts@@plt>
9402 0x08048348 <+24>: mov $0x0,%eax
9403 0x0804834d <+29>: leave
9404 0x0804834e <+30>: ret
9406 End of assembler dump.
9409 The @code{/m} option is deprecated as its output is not useful when
9410 there is either inlined code or re-ordered code.
9411 The @code{/s} option is the preferred choice.
9412 Here is an example for AMD x86-64 showing the difference between
9413 @code{/m} output and @code{/s} output.
9414 This example has one inline function defined in a header file,
9415 and the code is compiled with @samp{-O2} optimization.
9416 Note how the @code{/m} output is missing the disassembly of
9417 several instructions that are present in the @code{/s} output.
9447 (@value{GDBP}) disas /m main
9448 Dump of assembler code for function main:
9452 0x0000000000400400 <+0>: mov 0x200c2e(%rip),%eax # 0x601034 <y>
9453 0x0000000000400417 <+23>: mov %eax,0x200c13(%rip) # 0x601030 <x>
9457 0x000000000040041d <+29>: xor %eax,%eax
9458 0x000000000040041f <+31>: retq
9459 0x0000000000400420 <+32>: add %eax,%eax
9460 0x0000000000400422 <+34>: jmp 0x400417 <main+23>
9462 End of assembler dump.
9463 (@value{GDBP}) disas /s main
9464 Dump of assembler code for function main:
9468 0x0000000000400400 <+0>: mov 0x200c2e(%rip),%eax # 0x601034 <y>
9472 0x0000000000400406 <+6>: test %eax,%eax
9473 0x0000000000400408 <+8>: js 0x400420 <main+32>
9478 0x000000000040040a <+10>: lea 0xa(%rax),%edx
9479 0x000000000040040d <+13>: test %eax,%eax
9480 0x000000000040040f <+15>: mov $0x1,%eax
9481 0x0000000000400414 <+20>: cmovne %edx,%eax
9485 0x0000000000400417 <+23>: mov %eax,0x200c13(%rip) # 0x601030 <x>
9489 0x000000000040041d <+29>: xor %eax,%eax
9490 0x000000000040041f <+31>: retq
9494 0x0000000000400420 <+32>: add %eax,%eax
9495 0x0000000000400422 <+34>: jmp 0x400417 <main+23>
9496 End of assembler dump.
9499 Here is another example showing raw instructions in hex for AMD x86-64,
9502 (gdb) disas /r 0x400281,+10
9503 Dump of assembler code from 0x400281 to 0x40028b:
9504 0x0000000000400281: 38 36 cmp %dh,(%rsi)
9505 0x0000000000400283: 2d 36 34 2e 73 sub $0x732e3436,%eax
9506 0x0000000000400288: 6f outsl %ds:(%rsi),(%dx)
9507 0x0000000000400289: 2e 32 00 xor %cs:(%rax),%al
9508 End of assembler dump.
9511 Addresses cannot be specified as a location (@pxref{Specify Location}).
9512 So, for example, if you want to disassemble function @code{bar}
9513 in file @file{foo.c}, you must type @samp{disassemble 'foo.c'::bar}
9514 and not @samp{disassemble foo.c:bar}.
9516 Some architectures have more than one commonly-used set of instruction
9517 mnemonics or other syntax.
9519 For programs that were dynamically linked and use shared libraries,
9520 instructions that call functions or branch to locations in the shared
9521 libraries might show a seemingly bogus location---it's actually a
9522 location of the relocation table. On some architectures, @value{GDBN}
9523 might be able to resolve these to actual function names.
9526 @kindex set disassembler-options
9527 @cindex disassembler options
9528 @item set disassembler-options @var{option1}[,@var{option2}@dots{}]
9529 This command controls the passing of target specific information to
9530 the disassembler. For a list of valid options, please refer to the
9531 @code{-M}/@code{--disassembler-options} section of the @samp{objdump}
9532 manual and/or the output of @kbd{objdump --help}
9533 (@pxref{objdump,,objdump,binutils,The GNU Binary Utilities}).
9534 The default value is the empty string.
9536 If it is necessary to specify more than one disassembler option, then
9537 multiple options can be placed together into a comma separated list.
9538 Currently this command is only supported on targets ARM, MIPS, PowerPC
9541 @kindex show disassembler-options
9542 @item show disassembler-options
9543 Show the current setting of the disassembler options.
9547 @kindex set disassembly-flavor
9548 @cindex Intel disassembly flavor
9549 @cindex AT&T disassembly flavor
9550 @item set disassembly-flavor @var{instruction-set}
9551 Select the instruction set to use when disassembling the
9552 program via the @code{disassemble} or @code{x/i} commands.
9554 Currently this command is only defined for the Intel x86 family. You
9555 can set @var{instruction-set} to either @code{intel} or @code{att}.
9556 The default is @code{att}, the AT&T flavor used by default by Unix
9557 assemblers for x86-based targets.
9559 @kindex show disassembly-flavor
9560 @item show disassembly-flavor
9561 Show the current setting of the disassembly flavor.
9565 @kindex set disassemble-next-line
9566 @kindex show disassemble-next-line
9567 @item set disassemble-next-line
9568 @itemx show disassemble-next-line
9569 Control whether or not @value{GDBN} will disassemble the next source
9570 line or instruction when execution stops. If ON, @value{GDBN} will
9571 display disassembly of the next source line when execution of the
9572 program being debugged stops. This is @emph{in addition} to
9573 displaying the source line itself, which @value{GDBN} always does if
9574 possible. If the next source line cannot be displayed for some reason
9575 (e.g., if @value{GDBN} cannot find the source file, or there's no line
9576 info in the debug info), @value{GDBN} will display disassembly of the
9577 next @emph{instruction} instead of showing the next source line. If
9578 AUTO, @value{GDBN} will display disassembly of next instruction only
9579 if the source line cannot be displayed. This setting causes
9580 @value{GDBN} to display some feedback when you step through a function
9581 with no line info or whose source file is unavailable. The default is
9582 OFF, which means never display the disassembly of the next line or
9588 @chapter Examining Data
9590 @cindex printing data
9591 @cindex examining data
9594 The usual way to examine data in your program is with the @code{print}
9595 command (abbreviated @code{p}), or its synonym @code{inspect}. It
9596 evaluates and prints the value of an expression of the language your
9597 program is written in (@pxref{Languages, ,Using @value{GDBN} with
9598 Different Languages}). It may also print the expression using a
9599 Python-based pretty-printer (@pxref{Pretty Printing}).
9602 @item print [[@var{options}] --] @var{expr}
9603 @itemx print [[@var{options}] --] /@var{f} @var{expr}
9604 @var{expr} is an expression (in the source language). By default the
9605 value of @var{expr} is printed in a format appropriate to its data type;
9606 you can choose a different format by specifying @samp{/@var{f}}, where
9607 @var{f} is a letter specifying the format; see @ref{Output Formats,,Output
9610 @anchor{print options}
9611 The @code{print} command supports a number of options that allow
9612 overriding relevant global print settings as set by @code{set print}
9616 @item -address [@code{on}|@code{off}]
9617 Set printing of addresses.
9618 Related setting: @ref{set print address}.
9620 @item -array [@code{on}|@code{off}]
9621 Pretty formatting of arrays.
9622 Related setting: @ref{set print array}.
9624 @item -array-indexes [@code{on}|@code{off}]
9625 Set printing of array indexes.
9626 Related setting: @ref{set print array-indexes}.
9628 @item -elements @var{number-of-elements}|@code{unlimited}
9629 Set limit on string chars or array elements to print. The value
9630 @code{unlimited} causes there to be no limit. Related setting:
9631 @ref{set print elements}.
9633 @item -max-depth @var{depth}|@code{unlimited}
9634 Set the threshold after which nested structures are replaced with
9635 ellipsis. Related setting: @ref{set print max-depth}.
9637 @item -null-stop [@code{on}|@code{off}]
9638 Set printing of char arrays to stop at first null char. Related
9639 setting: @ref{set print null-stop}.
9641 @item -object [@code{on}|@code{off}]
9642 Set printing C@t{++} virtual function tables. Related setting:
9643 @ref{set print object}.
9645 @item -pretty [@code{on}|@code{off}]
9646 Set pretty formatting of structures. Related setting: @ref{set print
9649 @item -repeats @var{number-of-repeats}|@code{unlimited}
9650 Set threshold for repeated print elements. @code{unlimited} causes
9651 all elements to be individually printed. Related setting: @ref{set
9654 @item -static-members [@code{on}|@code{off}]
9655 Set printing C@t{++} static members. Related setting: @ref{set print
9658 @item -symbol [@code{on}|@code{off}]
9659 Set printing of symbol names when printing pointers. Related setting:
9660 @ref{set print symbol}.
9662 @item -union [@code{on}|@code{off}]
9663 Set printing of unions interior to structures. Related setting:
9664 @ref{set print union}.
9666 @item -vtbl [@code{on}|@code{off}]
9667 Set printing of C++ virtual function tables. Related setting:
9668 @ref{set print vtbl}.
9671 Because the @code{print} command accepts arbitrary expressions which
9672 may look like options (including abbreviations), if you specify any
9673 command option, then you must use a double dash (@code{--}) to mark
9674 the end of option processing.
9676 For example, this prints the value of the @code{-r} expression:
9679 (@value{GDBP}) print -r
9682 While this repeats the last value in the value history (see below)
9683 with the @code{-raw} option in effect:
9686 (@value{GDBP}) print -r --
9689 Here is an example including both on option and an expression:
9693 (@value{GDBP}) print -pretty -- *myptr
9705 @item print [@var{options}]
9706 @itemx print [@var{options}] /@var{f}
9707 @cindex reprint the last value
9708 If you omit @var{expr}, @value{GDBN} displays the last value again (from the
9709 @dfn{value history}; @pxref{Value History, ,Value History}). This allows you to
9710 conveniently inspect the same value in an alternative format.
9713 A more low-level way of examining data is with the @code{x} command.
9714 It examines data in memory at a specified address and prints it in a
9715 specified format. @xref{Memory, ,Examining Memory}.
9717 If you are interested in information about types, or about how the
9718 fields of a struct or a class are declared, use the @code{ptype @var{exp}}
9719 command rather than @code{print}. @xref{Symbols, ,Examining the Symbol
9722 @cindex exploring hierarchical data structures
9724 Another way of examining values of expressions and type information is
9725 through the Python extension command @code{explore} (available only if
9726 the @value{GDBN} build is configured with @code{--with-python}). It
9727 offers an interactive way to start at the highest level (or, the most
9728 abstract level) of the data type of an expression (or, the data type
9729 itself) and explore all the way down to leaf scalar values/fields
9730 embedded in the higher level data types.
9733 @item explore @var{arg}
9734 @var{arg} is either an expression (in the source language), or a type
9735 visible in the current context of the program being debugged.
9738 The working of the @code{explore} command can be illustrated with an
9739 example. If a data type @code{struct ComplexStruct} is defined in your
9749 struct ComplexStruct
9751 struct SimpleStruct *ss_p;
9757 followed by variable declarations as
9760 struct SimpleStruct ss = @{ 10, 1.11 @};
9761 struct ComplexStruct cs = @{ &ss, @{ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9 @} @};
9765 then, the value of the variable @code{cs} can be explored using the
9766 @code{explore} command as follows.
9770 The value of `cs' is a struct/class of type `struct ComplexStruct' with
9771 the following fields:
9773 ss_p = <Enter 0 to explore this field of type `struct SimpleStruct *'>
9774 arr = <Enter 1 to explore this field of type `int [10]'>
9776 Enter the field number of choice:
9780 Since the fields of @code{cs} are not scalar values, you are being
9781 prompted to chose the field you want to explore. Let's say you choose
9782 the field @code{ss_p} by entering @code{0}. Then, since this field is a
9783 pointer, you will be asked if it is pointing to a single value. From
9784 the declaration of @code{cs} above, it is indeed pointing to a single
9785 value, hence you enter @code{y}. If you enter @code{n}, then you will
9786 be asked if it were pointing to an array of values, in which case this
9787 field will be explored as if it were an array.
9790 `cs.ss_p' is a pointer to a value of type `struct SimpleStruct'
9791 Continue exploring it as a pointer to a single value [y/n]: y
9792 The value of `*(cs.ss_p)' is a struct/class of type `struct
9793 SimpleStruct' with the following fields:
9795 i = 10 .. (Value of type `int')
9796 d = 1.1100000000000001 .. (Value of type `double')
9798 Press enter to return to parent value:
9802 If the field @code{arr} of @code{cs} was chosen for exploration by
9803 entering @code{1} earlier, then since it is as array, you will be
9804 prompted to enter the index of the element in the array that you want
9808 `cs.arr' is an array of `int'.
9809 Enter the index of the element you want to explore in `cs.arr': 5
9811 `(cs.arr)[5]' is a scalar value of type `int'.
9815 Press enter to return to parent value:
9818 In general, at any stage of exploration, you can go deeper towards the
9819 leaf values by responding to the prompts appropriately, or hit the
9820 return key to return to the enclosing data structure (the @i{higher}
9821 level data structure).
9823 Similar to exploring values, you can use the @code{explore} command to
9824 explore types. Instead of specifying a value (which is typically a
9825 variable name or an expression valid in the current context of the
9826 program being debugged), you specify a type name. If you consider the
9827 same example as above, your can explore the type
9828 @code{struct ComplexStruct} by passing the argument
9829 @code{struct ComplexStruct} to the @code{explore} command.
9832 (gdb) explore struct ComplexStruct
9836 By responding to the prompts appropriately in the subsequent interactive
9837 session, you can explore the type @code{struct ComplexStruct} in a
9838 manner similar to how the value @code{cs} was explored in the above
9841 The @code{explore} command also has two sub-commands,
9842 @code{explore value} and @code{explore type}. The former sub-command is
9843 a way to explicitly specify that value exploration of the argument is
9844 being invoked, while the latter is a way to explicitly specify that type
9845 exploration of the argument is being invoked.
9848 @item explore value @var{expr}
9849 @cindex explore value
9850 This sub-command of @code{explore} explores the value of the
9851 expression @var{expr} (if @var{expr} is an expression valid in the
9852 current context of the program being debugged). The behavior of this
9853 command is identical to that of the behavior of the @code{explore}
9854 command being passed the argument @var{expr}.
9856 @item explore type @var{arg}
9857 @cindex explore type
9858 This sub-command of @code{explore} explores the type of @var{arg} (if
9859 @var{arg} is a type visible in the current context of program being
9860 debugged), or the type of the value/expression @var{arg} (if @var{arg}
9861 is an expression valid in the current context of the program being
9862 debugged). If @var{arg} is a type, then the behavior of this command is
9863 identical to that of the @code{explore} command being passed the
9864 argument @var{arg}. If @var{arg} is an expression, then the behavior of
9865 this command will be identical to that of the @code{explore} command
9866 being passed the type of @var{arg} as the argument.
9870 * Expressions:: Expressions
9871 * Ambiguous Expressions:: Ambiguous Expressions
9872 * Variables:: Program variables
9873 * Arrays:: Artificial arrays
9874 * Output Formats:: Output formats
9875 * Memory:: Examining memory
9876 * Auto Display:: Automatic display
9877 * Print Settings:: Print settings
9878 * Pretty Printing:: Python pretty printing
9879 * Value History:: Value history
9880 * Convenience Vars:: Convenience variables
9881 * Convenience Funs:: Convenience functions
9882 * Registers:: Registers
9883 * Floating Point Hardware:: Floating point hardware
9884 * Vector Unit:: Vector Unit
9885 * OS Information:: Auxiliary data provided by operating system
9886 * Memory Region Attributes:: Memory region attributes
9887 * Dump/Restore Files:: Copy between memory and a file
9888 * Core File Generation:: Cause a program dump its core
9889 * Character Sets:: Debugging programs that use a different
9890 character set than GDB does
9891 * Caching Target Data:: Data caching for targets
9892 * Searching Memory:: Searching memory for a sequence of bytes
9893 * Value Sizes:: Managing memory allocated for values
9897 @section Expressions
9900 @code{print} and many other @value{GDBN} commands accept an expression and
9901 compute its value. Any kind of constant, variable or operator defined
9902 by the programming language you are using is valid in an expression in
9903 @value{GDBN}. This includes conditional expressions, function calls,
9904 casts, and string constants. It also includes preprocessor macros, if
9905 you compiled your program to include this information; see
9908 @cindex arrays in expressions
9909 @value{GDBN} supports array constants in expressions input by
9910 the user. The syntax is @{@var{element}, @var{element}@dots{}@}. For example,
9911 you can use the command @code{print @{1, 2, 3@}} to create an array
9912 of three integers. If you pass an array to a function or assign it
9913 to a program variable, @value{GDBN} copies the array to memory that
9914 is @code{malloc}ed in the target program.
9916 Because C is so widespread, most of the expressions shown in examples in
9917 this manual are in C. @xref{Languages, , Using @value{GDBN} with Different
9918 Languages}, for information on how to use expressions in other
9921 In this section, we discuss operators that you can use in @value{GDBN}
9922 expressions regardless of your programming language.
9924 @cindex casts, in expressions
9925 Casts are supported in all languages, not just in C, because it is so
9926 useful to cast a number into a pointer in order to examine a structure
9927 at that address in memory.
9928 @c FIXME: casts supported---Mod2 true?
9930 @value{GDBN} supports these operators, in addition to those common
9931 to programming languages:
9935 @samp{@@} is a binary operator for treating parts of memory as arrays.
9936 @xref{Arrays, ,Artificial Arrays}, for more information.
9939 @samp{::} allows you to specify a variable in terms of the file or
9940 function where it is defined. @xref{Variables, ,Program Variables}.
9942 @cindex @{@var{type}@}
9943 @cindex type casting memory
9944 @cindex memory, viewing as typed object
9945 @cindex casts, to view memory
9946 @item @{@var{type}@} @var{addr}
9947 Refers to an object of type @var{type} stored at address @var{addr} in
9948 memory. The address @var{addr} may be any expression whose value is
9949 an integer or pointer (but parentheses are required around binary
9950 operators, just as in a cast). This construct is allowed regardless
9951 of what kind of data is normally supposed to reside at @var{addr}.
9954 @node Ambiguous Expressions
9955 @section Ambiguous Expressions
9956 @cindex ambiguous expressions
9958 Expressions can sometimes contain some ambiguous elements. For instance,
9959 some programming languages (notably Ada, C@t{++} and Objective-C) permit
9960 a single function name to be defined several times, for application in
9961 different contexts. This is called @dfn{overloading}. Another example
9962 involving Ada is generics. A @dfn{generic package} is similar to C@t{++}
9963 templates and is typically instantiated several times, resulting in
9964 the same function name being defined in different contexts.
9966 In some cases and depending on the language, it is possible to adjust
9967 the expression to remove the ambiguity. For instance in C@t{++}, you
9968 can specify the signature of the function you want to break on, as in
9969 @kbd{break @var{function}(@var{types})}. In Ada, using the fully
9970 qualified name of your function often makes the expression unambiguous
9973 When an ambiguity that needs to be resolved is detected, the debugger
9974 has the capability to display a menu of numbered choices for each
9975 possibility, and then waits for the selection with the prompt @samp{>}.
9976 The first option is always @samp{[0] cancel}, and typing @kbd{0 @key{RET}}
9977 aborts the current command. If the command in which the expression was
9978 used allows more than one choice to be selected, the next option in the
9979 menu is @samp{[1] all}, and typing @kbd{1 @key{RET}} selects all possible
9982 For example, the following session excerpt shows an attempt to set a
9983 breakpoint at the overloaded symbol @code{String::after}.
9984 We choose three particular definitions of that function name:
9986 @c FIXME! This is likely to change to show arg type lists, at least
9989 (@value{GDBP}) b String::after
9992 [2] file:String.cc; line number:867
9993 [3] file:String.cc; line number:860
9994 [4] file:String.cc; line number:875
9995 [5] file:String.cc; line number:853
9996 [6] file:String.cc; line number:846
9997 [7] file:String.cc; line number:735
9999 Breakpoint 1 at 0xb26c: file String.cc, line 867.
10000 Breakpoint 2 at 0xb344: file String.cc, line 875.
10001 Breakpoint 3 at 0xafcc: file String.cc, line 846.
10002 Multiple breakpoints were set.
10003 Use the "delete" command to delete unwanted
10010 @kindex set multiple-symbols
10011 @item set multiple-symbols @var{mode}
10012 @cindex multiple-symbols menu
10014 This option allows you to adjust the debugger behavior when an expression
10017 By default, @var{mode} is set to @code{all}. If the command with which
10018 the expression is used allows more than one choice, then @value{GDBN}
10019 automatically selects all possible choices. For instance, inserting
10020 a breakpoint on a function using an ambiguous name results in a breakpoint
10021 inserted on each possible match. However, if a unique choice must be made,
10022 then @value{GDBN} uses the menu to help you disambiguate the expression.
10023 For instance, printing the address of an overloaded function will result
10024 in the use of the menu.
10026 When @var{mode} is set to @code{ask}, the debugger always uses the menu
10027 when an ambiguity is detected.
10029 Finally, when @var{mode} is set to @code{cancel}, the debugger reports
10030 an error due to the ambiguity and the command is aborted.
10032 @kindex show multiple-symbols
10033 @item show multiple-symbols
10034 Show the current value of the @code{multiple-symbols} setting.
10038 @section Program Variables
10040 The most common kind of expression to use is the name of a variable
10043 Variables in expressions are understood in the selected stack frame
10044 (@pxref{Selection, ,Selecting a Frame}); they must be either:
10048 global (or file-static)
10055 visible according to the scope rules of the
10056 programming language from the point of execution in that frame
10059 @noindent This means that in the function
10074 you can examine and use the variable @code{a} whenever your program is
10075 executing within the function @code{foo}, but you can only use or
10076 examine the variable @code{b} while your program is executing inside
10077 the block where @code{b} is declared.
10079 @cindex variable name conflict
10080 There is an exception: you can refer to a variable or function whose
10081 scope is a single source file even if the current execution point is not
10082 in this file. But it is possible to have more than one such variable or
10083 function with the same name (in different source files). If that
10084 happens, referring to that name has unpredictable effects. If you wish,
10085 you can specify a static variable in a particular function or file by
10086 using the colon-colon (@code{::}) notation:
10088 @cindex colon-colon, context for variables/functions
10090 @c info cannot cope with a :: index entry, but why deprive hard copy readers?
10091 @cindex @code{::}, context for variables/functions
10094 @var{file}::@var{variable}
10095 @var{function}::@var{variable}
10099 Here @var{file} or @var{function} is the name of the context for the
10100 static @var{variable}. In the case of file names, you can use quotes to
10101 make sure @value{GDBN} parses the file name as a single word---for example,
10102 to print a global value of @code{x} defined in @file{f2.c}:
10105 (@value{GDBP}) p 'f2.c'::x
10108 The @code{::} notation is normally used for referring to
10109 static variables, since you typically disambiguate uses of local variables
10110 in functions by selecting the appropriate frame and using the
10111 simple name of the variable. However, you may also use this notation
10112 to refer to local variables in frames enclosing the selected frame:
10121 process (a); /* Stop here */
10132 For example, if there is a breakpoint at the commented line,
10133 here is what you might see
10134 when the program stops after executing the call @code{bar(0)}:
10139 (@value{GDBP}) p bar::a
10141 (@value{GDBP}) up 2
10142 #2 0x080483d0 in foo (a=5) at foobar.c:12
10145 (@value{GDBP}) p bar::a
10149 @cindex C@t{++} scope resolution
10150 These uses of @samp{::} are very rarely in conflict with the very
10151 similar use of the same notation in C@t{++}. When they are in
10152 conflict, the C@t{++} meaning takes precedence; however, this can be
10153 overridden by quoting the file or function name with single quotes.
10155 For example, suppose the program is stopped in a method of a class
10156 that has a field named @code{includefile}, and there is also an
10157 include file named @file{includefile} that defines a variable,
10158 @code{some_global}.
10161 (@value{GDBP}) p includefile
10163 (@value{GDBP}) p includefile::some_global
10164 A syntax error in expression, near `'.
10165 (@value{GDBP}) p 'includefile'::some_global
10169 @cindex wrong values
10170 @cindex variable values, wrong
10171 @cindex function entry/exit, wrong values of variables
10172 @cindex optimized code, wrong values of variables
10174 @emph{Warning:} Occasionally, a local variable may appear to have the
10175 wrong value at certain points in a function---just after entry to a new
10176 scope, and just before exit.
10178 You may see this problem when you are stepping by machine instructions.
10179 This is because, on most machines, it takes more than one instruction to
10180 set up a stack frame (including local variable definitions); if you are
10181 stepping by machine instructions, variables may appear to have the wrong
10182 values until the stack frame is completely built. On exit, it usually
10183 also takes more than one machine instruction to destroy a stack frame;
10184 after you begin stepping through that group of instructions, local
10185 variable definitions may be gone.
10187 This may also happen when the compiler does significant optimizations.
10188 To be sure of always seeing accurate values, turn off all optimization
10191 @cindex ``No symbol "foo" in current context''
10192 Another possible effect of compiler optimizations is to optimize
10193 unused variables out of existence, or assign variables to registers (as
10194 opposed to memory addresses). Depending on the support for such cases
10195 offered by the debug info format used by the compiler, @value{GDBN}
10196 might not be able to display values for such local variables. If that
10197 happens, @value{GDBN} will print a message like this:
10200 No symbol "foo" in current context.
10203 To solve such problems, either recompile without optimizations, or use a
10204 different debug info format, if the compiler supports several such
10205 formats. @xref{Compilation}, for more information on choosing compiler
10206 options. @xref{C, ,C and C@t{++}}, for more information about debug
10207 info formats that are best suited to C@t{++} programs.
10209 If you ask to print an object whose contents are unknown to
10210 @value{GDBN}, e.g., because its data type is not completely specified
10211 by the debug information, @value{GDBN} will say @samp{<incomplete
10212 type>}. @xref{Symbols, incomplete type}, for more about this.
10214 @cindex no debug info variables
10215 If you try to examine or use the value of a (global) variable for
10216 which @value{GDBN} has no type information, e.g., because the program
10217 includes no debug information, @value{GDBN} displays an error message.
10218 @xref{Symbols, unknown type}, for more about unknown types. If you
10219 cast the variable to its declared type, @value{GDBN} gets the
10220 variable's value using the cast-to type as the variable's type. For
10221 example, in a C program:
10224 (@value{GDBP}) p var
10225 'var' has unknown type; cast it to its declared type
10226 (@value{GDBP}) p (float) var
10230 If you append @kbd{@@entry} string to a function parameter name you get its
10231 value at the time the function got called. If the value is not available an
10232 error message is printed. Entry values are available only with some compilers.
10233 Entry values are normally also printed at the function parameter list according
10234 to @ref{set print entry-values}.
10237 Breakpoint 1, d (i=30) at gdb.base/entry-value.c:29
10243 (gdb) print i@@entry
10247 Strings are identified as arrays of @code{char} values without specified
10248 signedness. Arrays of either @code{signed char} or @code{unsigned char} get
10249 printed as arrays of 1 byte sized integers. @code{-fsigned-char} or
10250 @code{-funsigned-char} @value{NGCC} options have no effect as @value{GDBN}
10251 defines literal string type @code{"char"} as @code{char} without a sign.
10256 signed char var1[] = "A";
10259 You get during debugging
10264 $2 = @{65 'A', 0 '\0'@}
10268 @section Artificial Arrays
10270 @cindex artificial array
10272 @kindex @@@r{, referencing memory as an array}
10273 It is often useful to print out several successive objects of the
10274 same type in memory; a section of an array, or an array of
10275 dynamically determined size for which only a pointer exists in the
10278 You can do this by referring to a contiguous span of memory as an
10279 @dfn{artificial array}, using the binary operator @samp{@@}. The left
10280 operand of @samp{@@} should be the first element of the desired array
10281 and be an individual object. The right operand should be the desired length
10282 of the array. The result is an array value whose elements are all of
10283 the type of the left argument. The first element is actually the left
10284 argument; the second element comes from bytes of memory immediately
10285 following those that hold the first element, and so on. Here is an
10286 example. If a program says
10289 int *array = (int *) malloc (len * sizeof (int));
10293 you can print the contents of @code{array} with
10299 The left operand of @samp{@@} must reside in memory. Array values made
10300 with @samp{@@} in this way behave just like other arrays in terms of
10301 subscripting, and are coerced to pointers when used in expressions.
10302 Artificial arrays most often appear in expressions via the value history
10303 (@pxref{Value History, ,Value History}), after printing one out.
10305 Another way to create an artificial array is to use a cast.
10306 This re-interprets a value as if it were an array.
10307 The value need not be in memory:
10309 (@value{GDBP}) p/x (short[2])0x12345678
10310 $1 = @{0x1234, 0x5678@}
10313 As a convenience, if you leave the array length out (as in
10314 @samp{(@var{type}[])@var{value}}) @value{GDBN} calculates the size to fill
10315 the value (as @samp{sizeof(@var{value})/sizeof(@var{type})}:
10317 (@value{GDBP}) p/x (short[])0x12345678
10318 $2 = @{0x1234, 0x5678@}
10321 Sometimes the artificial array mechanism is not quite enough; in
10322 moderately complex data structures, the elements of interest may not
10323 actually be adjacent---for example, if you are interested in the values
10324 of pointers in an array. One useful work-around in this situation is
10325 to use a convenience variable (@pxref{Convenience Vars, ,Convenience
10326 Variables}) as a counter in an expression that prints the first
10327 interesting value, and then repeat that expression via @key{RET}. For
10328 instance, suppose you have an array @code{dtab} of pointers to
10329 structures, and you are interested in the values of a field @code{fv}
10330 in each structure. Here is an example of what you might type:
10340 @node Output Formats
10341 @section Output Formats
10343 @cindex formatted output
10344 @cindex output formats
10345 By default, @value{GDBN} prints a value according to its data type. Sometimes
10346 this is not what you want. For example, you might want to print a number
10347 in hex, or a pointer in decimal. Or you might want to view data in memory
10348 at a certain address as a character string or as an instruction. To do
10349 these things, specify an @dfn{output format} when you print a value.
10351 The simplest use of output formats is to say how to print a value
10352 already computed. This is done by starting the arguments of the
10353 @code{print} command with a slash and a format letter. The format
10354 letters supported are:
10358 Regard the bits of the value as an integer, and print the integer in
10362 Print as integer in signed decimal.
10365 Print as integer in unsigned decimal.
10368 Print as integer in octal.
10371 Print as integer in binary. The letter @samp{t} stands for ``two''.
10372 @footnote{@samp{b} cannot be used because these format letters are also
10373 used with the @code{x} command, where @samp{b} stands for ``byte'';
10374 see @ref{Memory,,Examining Memory}.}
10377 @cindex unknown address, locating
10378 @cindex locate address
10379 Print as an address, both absolute in hexadecimal and as an offset from
10380 the nearest preceding symbol. You can use this format used to discover
10381 where (in what function) an unknown address is located:
10384 (@value{GDBP}) p/a 0x54320
10385 $3 = 0x54320 <_initialize_vx+396>
10389 The command @code{info symbol 0x54320} yields similar results.
10390 @xref{Symbols, info symbol}.
10393 Regard as an integer and print it as a character constant. This
10394 prints both the numerical value and its character representation. The
10395 character representation is replaced with the octal escape @samp{\nnn}
10396 for characters outside the 7-bit @sc{ascii} range.
10398 Without this format, @value{GDBN} displays @code{char},
10399 @w{@code{unsigned char}}, and @w{@code{signed char}} data as character
10400 constants. Single-byte members of vectors are displayed as integer
10404 Regard the bits of the value as a floating point number and print
10405 using typical floating point syntax.
10408 @cindex printing strings
10409 @cindex printing byte arrays
10410 Regard as a string, if possible. With this format, pointers to single-byte
10411 data are displayed as null-terminated strings and arrays of single-byte data
10412 are displayed as fixed-length strings. Other values are displayed in their
10415 Without this format, @value{GDBN} displays pointers to and arrays of
10416 @code{char}, @w{@code{unsigned char}}, and @w{@code{signed char}} as
10417 strings. Single-byte members of a vector are displayed as an integer
10421 Like @samp{x} formatting, the value is treated as an integer and
10422 printed as hexadecimal, but leading zeros are printed to pad the value
10423 to the size of the integer type.
10426 @cindex raw printing
10427 Print using the @samp{raw} formatting. By default, @value{GDBN} will
10428 use a Python-based pretty-printer, if one is available (@pxref{Pretty
10429 Printing}). This typically results in a higher-level display of the
10430 value's contents. The @samp{r} format bypasses any Python
10431 pretty-printer which might exist.
10434 For example, to print the program counter in hex (@pxref{Registers}), type
10441 Note that no space is required before the slash; this is because command
10442 names in @value{GDBN} cannot contain a slash.
10444 To reprint the last value in the value history with a different format,
10445 you can use the @code{print} command with just a format and no
10446 expression. For example, @samp{p/x} reprints the last value in hex.
10449 @section Examining Memory
10451 You can use the command @code{x} (for ``examine'') to examine memory in
10452 any of several formats, independently of your program's data types.
10454 @cindex examining memory
10456 @kindex x @r{(examine memory)}
10457 @item x/@var{nfu} @var{addr}
10458 @itemx x @var{addr}
10460 Use the @code{x} command to examine memory.
10463 @var{n}, @var{f}, and @var{u} are all optional parameters that specify how
10464 much memory to display and how to format it; @var{addr} is an
10465 expression giving the address where you want to start displaying memory.
10466 If you use defaults for @var{nfu}, you need not type the slash @samp{/}.
10467 Several commands set convenient defaults for @var{addr}.
10470 @item @var{n}, the repeat count
10471 The repeat count is a decimal integer; the default is 1. It specifies
10472 how much memory (counting by units @var{u}) to display. If a negative
10473 number is specified, memory is examined backward from @var{addr}.
10474 @c This really is **decimal**; unaffected by 'set radix' as of GDB
10477 @item @var{f}, the display format
10478 The display format is one of the formats used by @code{print}
10479 (@samp{x}, @samp{d}, @samp{u}, @samp{o}, @samp{t}, @samp{a}, @samp{c},
10480 @samp{f}, @samp{s}), and in addition @samp{i} (for machine instructions).
10481 The default is @samp{x} (hexadecimal) initially. The default changes
10482 each time you use either @code{x} or @code{print}.
10484 @item @var{u}, the unit size
10485 The unit size is any of
10491 Halfwords (two bytes).
10493 Words (four bytes). This is the initial default.
10495 Giant words (eight bytes).
10498 Each time you specify a unit size with @code{x}, that size becomes the
10499 default unit the next time you use @code{x}. For the @samp{i} format,
10500 the unit size is ignored and is normally not written. For the @samp{s} format,
10501 the unit size defaults to @samp{b}, unless it is explicitly given.
10502 Use @kbd{x /hs} to display 16-bit char strings and @kbd{x /ws} to display
10503 32-bit strings. The next use of @kbd{x /s} will again display 8-bit strings.
10504 Note that the results depend on the programming language of the
10505 current compilation unit. If the language is C, the @samp{s}
10506 modifier will use the UTF-16 encoding while @samp{w} will use
10507 UTF-32. The encoding is set by the programming language and cannot
10510 @item @var{addr}, starting display address
10511 @var{addr} is the address where you want @value{GDBN} to begin displaying
10512 memory. The expression need not have a pointer value (though it may);
10513 it is always interpreted as an integer address of a byte of memory.
10514 @xref{Expressions, ,Expressions}, for more information on expressions. The default for
10515 @var{addr} is usually just after the last address examined---but several
10516 other commands also set the default address: @code{info breakpoints} (to
10517 the address of the last breakpoint listed), @code{info line} (to the
10518 starting address of a line), and @code{print} (if you use it to display
10519 a value from memory).
10522 For example, @samp{x/3uh 0x54320} is a request to display three halfwords
10523 (@code{h}) of memory, formatted as unsigned decimal integers (@samp{u}),
10524 starting at address @code{0x54320}. @samp{x/4xw $sp} prints the four
10525 words (@samp{w}) of memory above the stack pointer (here, @samp{$sp};
10526 @pxref{Registers, ,Registers}) in hexadecimal (@samp{x}).
10528 You can also specify a negative repeat count to examine memory backward
10529 from the given address. For example, @samp{x/-3uh 0x54320} prints three
10530 halfwords (@code{h}) at @code{0x54314}, @code{0x54328}, and @code{0x5431c}.
10532 Since the letters indicating unit sizes are all distinct from the
10533 letters specifying output formats, you do not have to remember whether
10534 unit size or format comes first; either order works. The output
10535 specifications @samp{4xw} and @samp{4wx} mean exactly the same thing.
10536 (However, the count @var{n} must come first; @samp{wx4} does not work.)
10538 Even though the unit size @var{u} is ignored for the formats @samp{s}
10539 and @samp{i}, you might still want to use a count @var{n}; for example,
10540 @samp{3i} specifies that you want to see three machine instructions,
10541 including any operands. For convenience, especially when used with
10542 the @code{display} command, the @samp{i} format also prints branch delay
10543 slot instructions, if any, beyond the count specified, which immediately
10544 follow the last instruction that is within the count. The command
10545 @code{disassemble} gives an alternative way of inspecting machine
10546 instructions; see @ref{Machine Code,,Source and Machine Code}.
10548 If a negative repeat count is specified for the formats @samp{s} or @samp{i},
10549 the command displays null-terminated strings or instructions before the given
10550 address as many as the absolute value of the given number. For the @samp{i}
10551 format, we use line number information in the debug info to accurately locate
10552 instruction boundaries while disassembling backward. If line info is not
10553 available, the command stops examining memory with an error message.
10555 All the defaults for the arguments to @code{x} are designed to make it
10556 easy to continue scanning memory with minimal specifications each time
10557 you use @code{x}. For example, after you have inspected three machine
10558 instructions with @samp{x/3i @var{addr}}, you can inspect the next seven
10559 with just @samp{x/7}. If you use @key{RET} to repeat the @code{x} command,
10560 the repeat count @var{n} is used again; the other arguments default as
10561 for successive uses of @code{x}.
10563 When examining machine instructions, the instruction at current program
10564 counter is shown with a @code{=>} marker. For example:
10567 (@value{GDBP}) x/5i $pc-6
10568 0x804837f <main+11>: mov %esp,%ebp
10569 0x8048381 <main+13>: push %ecx
10570 0x8048382 <main+14>: sub $0x4,%esp
10571 => 0x8048385 <main+17>: movl $0x8048460,(%esp)
10572 0x804838c <main+24>: call 0x80482d4 <puts@@plt>
10575 @cindex @code{$_}, @code{$__}, and value history
10576 The addresses and contents printed by the @code{x} command are not saved
10577 in the value history because there is often too much of them and they
10578 would get in the way. Instead, @value{GDBN} makes these values available for
10579 subsequent use in expressions as values of the convenience variables
10580 @code{$_} and @code{$__}. After an @code{x} command, the last address
10581 examined is available for use in expressions in the convenience variable
10582 @code{$_}. The contents of that address, as examined, are available in
10583 the convenience variable @code{$__}.
10585 If the @code{x} command has a repeat count, the address and contents saved
10586 are from the last memory unit printed; this is not the same as the last
10587 address printed if several units were printed on the last line of output.
10589 @anchor{addressable memory unit}
10590 @cindex addressable memory unit
10591 Most targets have an addressable memory unit size of 8 bits. This means
10592 that to each memory address are associated 8 bits of data. Some
10593 targets, however, have other addressable memory unit sizes.
10594 Within @value{GDBN} and this document, the term
10595 @dfn{addressable memory unit} (or @dfn{memory unit} for short) is used
10596 when explicitly referring to a chunk of data of that size. The word
10597 @dfn{byte} is used to refer to a chunk of data of 8 bits, regardless of
10598 the addressable memory unit size of the target. For most systems,
10599 addressable memory unit is a synonym of byte.
10601 @cindex remote memory comparison
10602 @cindex target memory comparison
10603 @cindex verify remote memory image
10604 @cindex verify target memory image
10605 When you are debugging a program running on a remote target machine
10606 (@pxref{Remote Debugging}), you may wish to verify the program's image
10607 in the remote machine's memory against the executable file you
10608 downloaded to the target. Or, on any target, you may want to check
10609 whether the program has corrupted its own read-only sections. The
10610 @code{compare-sections} command is provided for such situations.
10613 @kindex compare-sections
10614 @item compare-sections @r{[}@var{section-name}@r{|}@code{-r}@r{]}
10615 Compare the data of a loadable section @var{section-name} in the
10616 executable file of the program being debugged with the same section in
10617 the target machine's memory, and report any mismatches. With no
10618 arguments, compares all loadable sections. With an argument of
10619 @code{-r}, compares all loadable read-only sections.
10621 Note: for remote targets, this command can be accelerated if the
10622 target supports computing the CRC checksum of a block of memory
10623 (@pxref{qCRC packet}).
10627 @section Automatic Display
10628 @cindex automatic display
10629 @cindex display of expressions
10631 If you find that you want to print the value of an expression frequently
10632 (to see how it changes), you might want to add it to the @dfn{automatic
10633 display list} so that @value{GDBN} prints its value each time your program stops.
10634 Each expression added to the list is given a number to identify it;
10635 to remove an expression from the list, you specify that number.
10636 The automatic display looks like this:
10640 3: bar[5] = (struct hack *) 0x3804
10644 This display shows item numbers, expressions and their current values. As with
10645 displays you request manually using @code{x} or @code{print}, you can
10646 specify the output format you prefer; in fact, @code{display} decides
10647 whether to use @code{print} or @code{x} depending your format
10648 specification---it uses @code{x} if you specify either the @samp{i}
10649 or @samp{s} format, or a unit size; otherwise it uses @code{print}.
10653 @item display @var{expr}
10654 Add the expression @var{expr} to the list of expressions to display
10655 each time your program stops. @xref{Expressions, ,Expressions}.
10657 @code{display} does not repeat if you press @key{RET} again after using it.
10659 @item display/@var{fmt} @var{expr}
10660 For @var{fmt} specifying only a display format and not a size or
10661 count, add the expression @var{expr} to the auto-display list but
10662 arrange to display it each time in the specified format @var{fmt}.
10663 @xref{Output Formats,,Output Formats}.
10665 @item display/@var{fmt} @var{addr}
10666 For @var{fmt} @samp{i} or @samp{s}, or including a unit-size or a
10667 number of units, add the expression @var{addr} as a memory address to
10668 be examined each time your program stops. Examining means in effect
10669 doing @samp{x/@var{fmt} @var{addr}}. @xref{Memory, ,Examining Memory}.
10672 For example, @samp{display/i $pc} can be helpful, to see the machine
10673 instruction about to be executed each time execution stops (@samp{$pc}
10674 is a common name for the program counter; @pxref{Registers, ,Registers}).
10677 @kindex delete display
10679 @item undisplay @var{dnums}@dots{}
10680 @itemx delete display @var{dnums}@dots{}
10681 Remove items from the list of expressions to display. Specify the
10682 numbers of the displays that you want affected with the command
10683 argument @var{dnums}. It can be a single display number, one of the
10684 numbers shown in the first field of the @samp{info display} display;
10685 or it could be a range of display numbers, as in @code{2-4}.
10687 @code{undisplay} does not repeat if you press @key{RET} after using it.
10688 (Otherwise you would just get the error @samp{No display number @dots{}}.)
10690 @kindex disable display
10691 @item disable display @var{dnums}@dots{}
10692 Disable the display of item numbers @var{dnums}. A disabled display
10693 item is not printed automatically, but is not forgotten. It may be
10694 enabled again later. Specify the numbers of the displays that you
10695 want affected with the command argument @var{dnums}. It can be a
10696 single display number, one of the numbers shown in the first field of
10697 the @samp{info display} display; or it could be a range of display
10698 numbers, as in @code{2-4}.
10700 @kindex enable display
10701 @item enable display @var{dnums}@dots{}
10702 Enable display of item numbers @var{dnums}. It becomes effective once
10703 again in auto display of its expression, until you specify otherwise.
10704 Specify the numbers of the displays that you want affected with the
10705 command argument @var{dnums}. It can be a single display number, one
10706 of the numbers shown in the first field of the @samp{info display}
10707 display; or it could be a range of display numbers, as in @code{2-4}.
10710 Display the current values of the expressions on the list, just as is
10711 done when your program stops.
10713 @kindex info display
10715 Print the list of expressions previously set up to display
10716 automatically, each one with its item number, but without showing the
10717 values. This includes disabled expressions, which are marked as such.
10718 It also includes expressions which would not be displayed right now
10719 because they refer to automatic variables not currently available.
10722 @cindex display disabled out of scope
10723 If a display expression refers to local variables, then it does not make
10724 sense outside the lexical context for which it was set up. Such an
10725 expression is disabled when execution enters a context where one of its
10726 variables is not defined. For example, if you give the command
10727 @code{display last_char} while inside a function with an argument
10728 @code{last_char}, @value{GDBN} displays this argument while your program
10729 continues to stop inside that function. When it stops elsewhere---where
10730 there is no variable @code{last_char}---the display is disabled
10731 automatically. The next time your program stops where @code{last_char}
10732 is meaningful, you can enable the display expression once again.
10734 @node Print Settings
10735 @section Print Settings
10737 @cindex format options
10738 @cindex print settings
10739 @value{GDBN} provides the following ways to control how arrays, structures,
10740 and symbols are printed.
10743 These settings are useful for debugging programs in any language:
10747 @anchor{set print address}
10748 @item set print address
10749 @itemx set print address on
10750 @cindex print/don't print memory addresses
10751 @value{GDBN} prints memory addresses showing the location of stack
10752 traces, structure values, pointer values, breakpoints, and so forth,
10753 even when it also displays the contents of those addresses. The default
10754 is @code{on}. For example, this is what a stack frame display looks like with
10755 @code{set print address on}:
10760 #0 set_quotes (lq=0x34c78 "<<", rq=0x34c88 ">>")
10762 530 if (lquote != def_lquote)
10766 @item set print address off
10767 Do not print addresses when displaying their contents. For example,
10768 this is the same stack frame displayed with @code{set print address off}:
10772 (@value{GDBP}) set print addr off
10774 #0 set_quotes (lq="<<", rq=">>") at input.c:530
10775 530 if (lquote != def_lquote)
10779 You can use @samp{set print address off} to eliminate all machine
10780 dependent displays from the @value{GDBN} interface. For example, with
10781 @code{print address off}, you should get the same text for backtraces on
10782 all machines---whether or not they involve pointer arguments.
10785 @item show print address
10786 Show whether or not addresses are to be printed.
10789 When @value{GDBN} prints a symbolic address, it normally prints the
10790 closest earlier symbol plus an offset. If that symbol does not uniquely
10791 identify the address (for example, it is a name whose scope is a single
10792 source file), you may need to clarify. One way to do this is with
10793 @code{info line}, for example @samp{info line *0x4537}. Alternately,
10794 you can set @value{GDBN} to print the source file and line number when
10795 it prints a symbolic address:
10798 @item set print symbol-filename on
10799 @cindex source file and line of a symbol
10800 @cindex symbol, source file and line
10801 Tell @value{GDBN} to print the source file name and line number of a
10802 symbol in the symbolic form of an address.
10804 @item set print symbol-filename off
10805 Do not print source file name and line number of a symbol. This is the
10808 @item show print symbol-filename
10809 Show whether or not @value{GDBN} will print the source file name and
10810 line number of a symbol in the symbolic form of an address.
10813 Another situation where it is helpful to show symbol filenames and line
10814 numbers is when disassembling code; @value{GDBN} shows you the line
10815 number and source file that corresponds to each instruction.
10817 Also, you may wish to see the symbolic form only if the address being
10818 printed is reasonably close to the closest earlier symbol:
10821 @item set print max-symbolic-offset @var{max-offset}
10822 @itemx set print max-symbolic-offset unlimited
10823 @cindex maximum value for offset of closest symbol
10824 Tell @value{GDBN} to only display the symbolic form of an address if the
10825 offset between the closest earlier symbol and the address is less than
10826 @var{max-offset}. The default is @code{unlimited}, which tells @value{GDBN}
10827 to always print the symbolic form of an address if any symbol precedes
10828 it. Zero is equivalent to @code{unlimited}.
10830 @item show print max-symbolic-offset
10831 Ask how large the maximum offset is that @value{GDBN} prints in a
10835 @cindex wild pointer, interpreting
10836 @cindex pointer, finding referent
10837 If you have a pointer and you are not sure where it points, try
10838 @samp{set print symbol-filename on}. Then you can determine the name
10839 and source file location of the variable where it points, using
10840 @samp{p/a @var{pointer}}. This interprets the address in symbolic form.
10841 For example, here @value{GDBN} shows that a variable @code{ptt} points
10842 at another variable @code{t}, defined in @file{hi2.c}:
10845 (@value{GDBP}) set print symbol-filename on
10846 (@value{GDBP}) p/a ptt
10847 $4 = 0xe008 <t in hi2.c>
10851 @emph{Warning:} For pointers that point to a local variable, @samp{p/a}
10852 does not show the symbol name and filename of the referent, even with
10853 the appropriate @code{set print} options turned on.
10856 You can also enable @samp{/a}-like formatting all the time using
10857 @samp{set print symbol on}:
10859 @anchor{set print symbol}
10861 @item set print symbol on
10862 Tell @value{GDBN} to print the symbol corresponding to an address, if
10865 @item set print symbol off
10866 Tell @value{GDBN} not to print the symbol corresponding to an
10867 address. In this mode, @value{GDBN} will still print the symbol
10868 corresponding to pointers to functions. This is the default.
10870 @item show print symbol
10871 Show whether @value{GDBN} will display the symbol corresponding to an
10875 Other settings control how different kinds of objects are printed:
10878 @anchor{set print array}
10879 @item set print array
10880 @itemx set print array on
10881 @cindex pretty print arrays
10882 Pretty print arrays. This format is more convenient to read,
10883 but uses more space. The default is off.
10885 @item set print array off
10886 Return to compressed format for arrays.
10888 @item show print array
10889 Show whether compressed or pretty format is selected for displaying
10892 @cindex print array indexes
10893 @anchor{set print array-indexes}
10894 @item set print array-indexes
10895 @itemx set print array-indexes on
10896 Print the index of each element when displaying arrays. May be more
10897 convenient to locate a given element in the array or quickly find the
10898 index of a given element in that printed array. The default is off.
10900 @item set print array-indexes off
10901 Stop printing element indexes when displaying arrays.
10903 @item show print array-indexes
10904 Show whether the index of each element is printed when displaying
10907 @anchor{set print elements}
10908 @item set print elements @var{number-of-elements}
10909 @itemx set print elements unlimited
10910 @cindex number of array elements to print
10911 @cindex limit on number of printed array elements
10912 Set a limit on how many elements of an array @value{GDBN} will print.
10913 If @value{GDBN} is printing a large array, it stops printing after it has
10914 printed the number of elements set by the @code{set print elements} command.
10915 This limit also applies to the display of strings.
10916 When @value{GDBN} starts, this limit is set to 200.
10917 Setting @var{number-of-elements} to @code{unlimited} or zero means
10918 that the number of elements to print is unlimited.
10920 @item show print elements
10921 Display the number of elements of a large array that @value{GDBN} will print.
10922 If the number is 0, then the printing is unlimited.
10924 @anchor{set print frame-arguments}
10925 @item set print frame-arguments @var{value}
10926 @kindex set print frame-arguments
10927 @cindex printing frame argument values
10928 @cindex print all frame argument values
10929 @cindex print frame argument values for scalars only
10930 @cindex do not print frame arguments
10931 This command allows to control how the values of arguments are printed
10932 when the debugger prints a frame (@pxref{Frames}). The possible
10937 The values of all arguments are printed.
10940 Print the value of an argument only if it is a scalar. The value of more
10941 complex arguments such as arrays, structures, unions, etc, is replaced
10942 by @code{@dots{}}. This is the default. Here is an example where
10943 only scalar arguments are shown:
10946 #1 0x08048361 in call_me (i=3, s=@dots{}, ss=0xbf8d508c, u=@dots{}, e=green)
10951 None of the argument values are printed. Instead, the value of each argument
10952 is replaced by @code{@dots{}}. In this case, the example above now becomes:
10955 #1 0x08048361 in call_me (i=@dots{}, s=@dots{}, ss=@dots{}, u=@dots{}, e=@dots{})
10960 Only the presence of arguments is indicated by @code{@dots{}}.
10961 The @code{@dots{}} are not printed for function without any arguments.
10962 None of the argument names and values are printed.
10963 In this case, the example above now becomes:
10966 #1 0x08048361 in call_me (@dots{}) at frame-args.c:23
10971 By default, only scalar arguments are printed. This command can be used
10972 to configure the debugger to print the value of all arguments, regardless
10973 of their type. However, it is often advantageous to not print the value
10974 of more complex parameters. For instance, it reduces the amount of
10975 information printed in each frame, making the backtrace more readable.
10976 Also, it improves performance when displaying Ada frames, because
10977 the computation of large arguments can sometimes be CPU-intensive,
10978 especially in large applications. Setting @code{print frame-arguments}
10979 to @code{scalars} (the default), @code{none} or @code{presence} avoids
10980 this computation, thus speeding up the display of each Ada frame.
10982 @item show print frame-arguments
10983 Show how the value of arguments should be displayed when printing a frame.
10985 @anchor{set print raw-frame-arguments}
10986 @item set print raw-frame-arguments on
10987 Print frame arguments in raw, non pretty-printed, form.
10989 @item set print raw-frame-arguments off
10990 Print frame arguments in pretty-printed form, if there is a pretty-printer
10991 for the value (@pxref{Pretty Printing}),
10992 otherwise print the value in raw form.
10993 This is the default.
10995 @item show print raw-frame-arguments
10996 Show whether to print frame arguments in raw form.
10998 @anchor{set print entry-values}
10999 @item set print entry-values @var{value}
11000 @kindex set print entry-values
11001 Set printing of frame argument values at function entry. In some cases
11002 @value{GDBN} can determine the value of function argument which was passed by
11003 the function caller, even if the value was modified inside the called function
11004 and therefore is different. With optimized code, the current value could be
11005 unavailable, but the entry value may still be known.
11007 The default value is @code{default} (see below for its description). Older
11008 @value{GDBN} behaved as with the setting @code{no}. Compilers not supporting
11009 this feature will behave in the @code{default} setting the same way as with the
11012 This functionality is currently supported only by DWARF 2 debugging format and
11013 the compiler has to produce @samp{DW_TAG_call_site} tags. With
11014 @value{NGCC}, you need to specify @option{-O -g} during compilation, to get
11017 The @var{value} parameter can be one of the following:
11021 Print only actual parameter values, never print values from function entry
11025 #0 different (val=6)
11026 #0 lost (val=<optimized out>)
11028 #0 invalid (val=<optimized out>)
11032 Print only parameter values from function entry point. The actual parameter
11033 values are never printed.
11035 #0 equal (val@@entry=5)
11036 #0 different (val@@entry=5)
11037 #0 lost (val@@entry=5)
11038 #0 born (val@@entry=<optimized out>)
11039 #0 invalid (val@@entry=<optimized out>)
11043 Print only parameter values from function entry point. If value from function
11044 entry point is not known while the actual value is known, print the actual
11045 value for such parameter.
11047 #0 equal (val@@entry=5)
11048 #0 different (val@@entry=5)
11049 #0 lost (val@@entry=5)
11051 #0 invalid (val@@entry=<optimized out>)
11055 Print actual parameter values. If actual parameter value is not known while
11056 value from function entry point is known, print the entry point value for such
11060 #0 different (val=6)
11061 #0 lost (val@@entry=5)
11063 #0 invalid (val=<optimized out>)
11067 Always print both the actual parameter value and its value from function entry
11068 point, even if values of one or both are not available due to compiler
11071 #0 equal (val=5, val@@entry=5)
11072 #0 different (val=6, val@@entry=5)
11073 #0 lost (val=<optimized out>, val@@entry=5)
11074 #0 born (val=10, val@@entry=<optimized out>)
11075 #0 invalid (val=<optimized out>, val@@entry=<optimized out>)
11079 Print the actual parameter value if it is known and also its value from
11080 function entry point if it is known. If neither is known, print for the actual
11081 value @code{<optimized out>}. If not in MI mode (@pxref{GDB/MI}) and if both
11082 values are known and identical, print the shortened
11083 @code{param=param@@entry=VALUE} notation.
11085 #0 equal (val=val@@entry=5)
11086 #0 different (val=6, val@@entry=5)
11087 #0 lost (val@@entry=5)
11089 #0 invalid (val=<optimized out>)
11093 Always print the actual parameter value. Print also its value from function
11094 entry point, but only if it is known. If not in MI mode (@pxref{GDB/MI}) and
11095 if both values are known and identical, print the shortened
11096 @code{param=param@@entry=VALUE} notation.
11098 #0 equal (val=val@@entry=5)
11099 #0 different (val=6, val@@entry=5)
11100 #0 lost (val=<optimized out>, val@@entry=5)
11102 #0 invalid (val=<optimized out>)
11106 For analysis messages on possible failures of frame argument values at function
11107 entry resolution see @ref{set debug entry-values}.
11109 @item show print entry-values
11110 Show the method being used for printing of frame argument values at function
11113 @anchor{set print frame-info}
11114 @item set print frame-info @var{value}
11115 @kindex set print frame-info
11116 @cindex printing frame information
11117 @cindex frame information, printing
11118 This command allows to control the information printed when
11119 the debugger prints a frame. See @ref{Frames}, @ref{Backtrace},
11120 for a general explanation about frames and frame information.
11121 Note that some other settings (such as @code{set print frame-arguments}
11122 and @code{set print address}) are also influencing if and how some frame
11123 information is displayed. In particular, the frame program counter is never
11124 printed if @code{set print address} is off.
11126 The possible values for @code{set print frame-info} are:
11128 @item short-location
11129 Print the frame level, the program counter (if not at the
11130 beginning of the location source line), the function, the function
11133 Same as @code{short-location} but also print the source file and source line
11135 @item location-and-address
11136 Same as @code{location} but print the program counter even if located at the
11137 beginning of the location source line.
11139 Print the program counter (if not at the beginning of the location
11140 source line), the line number and the source line.
11141 @item source-and-location
11142 Print what @code{location} and @code{source-line} are printing.
11144 The information printed for a frame is decided automatically
11145 by the @value{GDBN} command that prints a frame.
11146 For example, @code{frame} prints the information printed by
11147 @code{source-and-location} while @code{stepi} will switch between
11148 @code{source-line} and @code{source-and-location} depending on the program
11150 The default value is @code{auto}.
11153 @anchor{set print repeats}
11154 @item set print repeats @var{number-of-repeats}
11155 @itemx set print repeats unlimited
11156 @cindex repeated array elements
11157 Set the threshold for suppressing display of repeated array
11158 elements. When the number of consecutive identical elements of an
11159 array exceeds the threshold, @value{GDBN} prints the string
11160 @code{"<repeats @var{n} times>"}, where @var{n} is the number of
11161 identical repetitions, instead of displaying the identical elements
11162 themselves. Setting the threshold to @code{unlimited} or zero will
11163 cause all elements to be individually printed. The default threshold
11166 @item show print repeats
11167 Display the current threshold for printing repeated identical
11170 @anchor{set print max-depth}
11171 @item set print max-depth @var{depth}
11172 @item set print max-depth unlimited
11173 @cindex printing nested structures
11174 Set the threshold after which nested structures are replaced with
11175 ellipsis, this can make visualising deeply nested structures easier.
11177 For example, given this C code
11180 typedef struct s1 @{ int a; @} s1;
11181 typedef struct s2 @{ s1 b; @} s2;
11182 typedef struct s3 @{ s2 c; @} s3;
11183 typedef struct s4 @{ s3 d; @} s4;
11185 s4 var = @{ @{ @{ @{ 3 @} @} @} @};
11188 The following table shows how different values of @var{depth} will
11189 effect how @code{var} is printed by @value{GDBN}:
11191 @multitable @columnfractions .3 .7
11192 @headitem @var{depth} setting @tab Result of @samp{p var}
11194 @tab @code{$1 = @{d = @{c = @{b = @{a = 3@}@}@}@}}
11196 @tab @code{$1 = @{...@}}
11198 @tab @code{$1 = @{d = @{...@}@}}
11200 @tab @code{$1 = @{d = @{c = @{...@}@}@}}
11202 @tab @code{$1 = @{d = @{c = @{b = @{...@}@}@}@}}
11204 @tab @code{$1 = @{d = @{c = @{b = @{a = 3@}@}@}@}}
11207 To see the contents of structures that have been hidden the user can
11208 either increase the print max-depth, or they can print the elements of
11209 the structure that are visible, for example
11212 (gdb) set print max-depth 2
11214 $1 = @{d = @{c = @{...@}@}@}
11216 $2 = @{c = @{b = @{...@}@}@}
11218 $3 = @{b = @{a = 3@}@}
11221 The pattern used to replace nested structures varies based on
11222 language, for most languages @code{@{...@}} is used, but Fortran uses
11225 @item show print max-depth
11226 Display the current threshold after which nested structures are
11227 replaces with ellipsis.
11229 @anchor{set print null-stop}
11230 @item set print null-stop
11231 @cindex @sc{null} elements in arrays
11232 Cause @value{GDBN} to stop printing the characters of an array when the first
11233 @sc{null} is encountered. This is useful when large arrays actually
11234 contain only short strings.
11235 The default is off.
11237 @item show print null-stop
11238 Show whether @value{GDBN} stops printing an array on the first
11239 @sc{null} character.
11241 @anchor{set print pretty}
11242 @item set print pretty on
11243 @cindex print structures in indented form
11244 @cindex indentation in structure display
11245 Cause @value{GDBN} to print structures in an indented format with one member
11246 per line, like this:
11261 @item set print pretty off
11262 Cause @value{GDBN} to print structures in a compact format, like this:
11266 $1 = @{next = 0x0, flags = @{sweet = 1, sour = 1@}, \
11267 meat = 0x54 "Pork"@}
11272 This is the default format.
11274 @item show print pretty
11275 Show which format @value{GDBN} is using to print structures.
11277 @item set print sevenbit-strings on
11278 @cindex eight-bit characters in strings
11279 @cindex octal escapes in strings
11280 Print using only seven-bit characters; if this option is set,
11281 @value{GDBN} displays any eight-bit characters (in strings or
11282 character values) using the notation @code{\}@var{nnn}. This setting is
11283 best if you are working in English (@sc{ascii}) and you use the
11284 high-order bit of characters as a marker or ``meta'' bit.
11286 @item set print sevenbit-strings off
11287 Print full eight-bit characters. This allows the use of more
11288 international character sets, and is the default.
11290 @item show print sevenbit-strings
11291 Show whether or not @value{GDBN} is printing only seven-bit characters.
11293 @anchor{set print union}
11294 @item set print union on
11295 @cindex unions in structures, printing
11296 Tell @value{GDBN} to print unions which are contained in structures
11297 and other unions. This is the default setting.
11299 @item set print union off
11300 Tell @value{GDBN} not to print unions which are contained in
11301 structures and other unions. @value{GDBN} will print @code{"@{...@}"}
11304 @item show print union
11305 Ask @value{GDBN} whether or not it will print unions which are contained in
11306 structures and other unions.
11308 For example, given the declarations
11311 typedef enum @{Tree, Bug@} Species;
11312 typedef enum @{Big_tree, Acorn, Seedling@} Tree_forms;
11313 typedef enum @{Caterpillar, Cocoon, Butterfly@}
11324 struct thing foo = @{Tree, @{Acorn@}@};
11328 with @code{set print union on} in effect @samp{p foo} would print
11331 $1 = @{it = Tree, form = @{tree = Acorn, bug = Cocoon@}@}
11335 and with @code{set print union off} in effect it would print
11338 $1 = @{it = Tree, form = @{...@}@}
11342 @code{set print union} affects programs written in C-like languages
11348 These settings are of interest when debugging C@t{++} programs:
11351 @cindex demangling C@t{++} names
11352 @item set print demangle
11353 @itemx set print demangle on
11354 Print C@t{++} names in their source form rather than in the encoded
11355 (``mangled'') form passed to the assembler and linker for type-safe
11356 linkage. The default is on.
11358 @item show print demangle
11359 Show whether C@t{++} names are printed in mangled or demangled form.
11361 @item set print asm-demangle
11362 @itemx set print asm-demangle on
11363 Print C@t{++} names in their source form rather than their mangled form, even
11364 in assembler code printouts such as instruction disassemblies.
11365 The default is off.
11367 @item show print asm-demangle
11368 Show whether C@t{++} names in assembly listings are printed in mangled
11371 @cindex C@t{++} symbol decoding style
11372 @cindex symbol decoding style, C@t{++}
11373 @kindex set demangle-style
11374 @item set demangle-style @var{style}
11375 Choose among several encoding schemes used by different compilers to represent
11376 C@t{++} names. If you omit @var{style}, you will see a list of possible
11377 formats. The default value is @var{auto}, which lets @value{GDBN} choose a
11378 decoding style by inspecting your program.
11380 @item show demangle-style
11381 Display the encoding style currently in use for decoding C@t{++} symbols.
11383 @anchor{set print object}
11384 @item set print object
11385 @itemx set print object on
11386 @cindex derived type of an object, printing
11387 @cindex display derived types
11388 When displaying a pointer to an object, identify the @emph{actual}
11389 (derived) type of the object rather than the @emph{declared} type, using
11390 the virtual function table. Note that the virtual function table is
11391 required---this feature can only work for objects that have run-time
11392 type identification; a single virtual method in the object's declared
11393 type is sufficient. Note that this setting is also taken into account when
11394 working with variable objects via MI (@pxref{GDB/MI}).
11396 @item set print object off
11397 Display only the declared type of objects, without reference to the
11398 virtual function table. This is the default setting.
11400 @item show print object
11401 Show whether actual, or declared, object types are displayed.
11403 @anchor{set print static-members}
11404 @item set print static-members
11405 @itemx set print static-members on
11406 @cindex static members of C@t{++} objects
11407 Print static members when displaying a C@t{++} object. The default is on.
11409 @item set print static-members off
11410 Do not print static members when displaying a C@t{++} object.
11412 @item show print static-members
11413 Show whether C@t{++} static members are printed or not.
11415 @item set print pascal_static-members
11416 @itemx set print pascal_static-members on
11417 @cindex static members of Pascal objects
11418 @cindex Pascal objects, static members display
11419 Print static members when displaying a Pascal object. The default is on.
11421 @item set print pascal_static-members off
11422 Do not print static members when displaying a Pascal object.
11424 @item show print pascal_static-members
11425 Show whether Pascal static members are printed or not.
11427 @c These don't work with HP ANSI C++ yet.
11428 @anchor{set print vtbl}
11429 @item set print vtbl
11430 @itemx set print vtbl on
11431 @cindex pretty print C@t{++} virtual function tables
11432 @cindex virtual functions (C@t{++}) display
11433 @cindex VTBL display
11434 Pretty print C@t{++} virtual function tables. The default is off.
11435 (The @code{vtbl} commands do not work on programs compiled with the HP
11436 ANSI C@t{++} compiler (@code{aCC}).)
11438 @item set print vtbl off
11439 Do not pretty print C@t{++} virtual function tables.
11441 @item show print vtbl
11442 Show whether C@t{++} virtual function tables are pretty printed, or not.
11445 @node Pretty Printing
11446 @section Pretty Printing
11448 @value{GDBN} provides a mechanism to allow pretty-printing of values using
11449 Python code. It greatly simplifies the display of complex objects. This
11450 mechanism works for both MI and the CLI.
11453 * Pretty-Printer Introduction:: Introduction to pretty-printers
11454 * Pretty-Printer Example:: An example pretty-printer
11455 * Pretty-Printer Commands:: Pretty-printer commands
11458 @node Pretty-Printer Introduction
11459 @subsection Pretty-Printer Introduction
11461 When @value{GDBN} prints a value, it first sees if there is a pretty-printer
11462 registered for the value. If there is then @value{GDBN} invokes the
11463 pretty-printer to print the value. Otherwise the value is printed normally.
11465 Pretty-printers are normally named. This makes them easy to manage.
11466 The @samp{info pretty-printer} command will list all the installed
11467 pretty-printers with their names.
11468 If a pretty-printer can handle multiple data types, then its
11469 @dfn{subprinters} are the printers for the individual data types.
11470 Each such subprinter has its own name.
11471 The format of the name is @var{printer-name};@var{subprinter-name}.
11473 Pretty-printers are installed by @dfn{registering} them with @value{GDBN}.
11474 Typically they are automatically loaded and registered when the corresponding
11475 debug information is loaded, thus making them available without having to
11476 do anything special.
11478 There are three places where a pretty-printer can be registered.
11482 Pretty-printers registered globally are available when debugging
11486 Pretty-printers registered with a program space are available only
11487 when debugging that program.
11488 @xref{Progspaces In Python}, for more details on program spaces in Python.
11491 Pretty-printers registered with an objfile are loaded and unloaded
11492 with the corresponding objfile (e.g., shared library).
11493 @xref{Objfiles In Python}, for more details on objfiles in Python.
11496 @xref{Selecting Pretty-Printers}, for further information on how
11497 pretty-printers are selected,
11499 @xref{Writing a Pretty-Printer}, for implementing pretty printers
11502 @node Pretty-Printer Example
11503 @subsection Pretty-Printer Example
11505 Here is how a C@t{++} @code{std::string} looks without a pretty-printer:
11508 (@value{GDBP}) print s
11510 static npos = 4294967295,
11512 <std::allocator<char>> = @{
11513 <__gnu_cxx::new_allocator<char>> = @{
11514 <No data fields>@}, <No data fields>
11516 members of std::basic_string<char, std::char_traits<char>,
11517 std::allocator<char> >::_Alloc_hider:
11518 _M_p = 0x804a014 "abcd"
11523 With a pretty-printer for @code{std::string} only the contents are printed:
11526 (@value{GDBP}) print s
11530 @node Pretty-Printer Commands
11531 @subsection Pretty-Printer Commands
11532 @cindex pretty-printer commands
11535 @kindex info pretty-printer
11536 @item info pretty-printer [@var{object-regexp} [@var{name-regexp}]]
11537 Print the list of installed pretty-printers.
11538 This includes disabled pretty-printers, which are marked as such.
11540 @var{object-regexp} is a regular expression matching the objects
11541 whose pretty-printers to list.
11542 Objects can be @code{global}, the program space's file
11543 (@pxref{Progspaces In Python}),
11544 and the object files within that program space (@pxref{Objfiles In Python}).
11545 @xref{Selecting Pretty-Printers}, for details on how @value{GDBN}
11546 looks up a printer from these three objects.
11548 @var{name-regexp} is a regular expression matching the name of the printers
11551 @kindex disable pretty-printer
11552 @item disable pretty-printer [@var{object-regexp} [@var{name-regexp}]]
11553 Disable pretty-printers matching @var{object-regexp} and @var{name-regexp}.
11554 A disabled pretty-printer is not forgotten, it may be enabled again later.
11556 @kindex enable pretty-printer
11557 @item enable pretty-printer [@var{object-regexp} [@var{name-regexp}]]
11558 Enable pretty-printers matching @var{object-regexp} and @var{name-regexp}.
11563 Suppose we have three pretty-printers installed: one from library1.so
11564 named @code{foo} that prints objects of type @code{foo}, and
11565 another from library2.so named @code{bar} that prints two types of objects,
11566 @code{bar1} and @code{bar2}.
11569 (gdb) info pretty-printer
11576 (gdb) info pretty-printer library2
11581 (gdb) disable pretty-printer library1
11583 2 of 3 printers enabled
11584 (gdb) info pretty-printer
11591 (gdb) disable pretty-printer library2 bar;bar1
11593 1 of 3 printers enabled
11594 (gdb) info pretty-printer library2
11601 (gdb) disable pretty-printer library2 bar
11603 0 of 3 printers enabled
11604 (gdb) info pretty-printer library2
11613 Note that for @code{bar} the entire printer can be disabled,
11614 as can each individual subprinter.
11616 @node Value History
11617 @section Value History
11619 @cindex value history
11620 @cindex history of values printed by @value{GDBN}
11621 Values printed by the @code{print} command are saved in the @value{GDBN}
11622 @dfn{value history}. This allows you to refer to them in other expressions.
11623 Values are kept until the symbol table is re-read or discarded
11624 (for example with the @code{file} or @code{symbol-file} commands).
11625 When the symbol table changes, the value history is discarded,
11626 since the values may contain pointers back to the types defined in the
11631 @cindex history number
11632 The values printed are given @dfn{history numbers} by which you can
11633 refer to them. These are successive integers starting with one.
11634 @code{print} shows you the history number assigned to a value by
11635 printing @samp{$@var{num} = } before the value; here @var{num} is the
11638 To refer to any previous value, use @samp{$} followed by the value's
11639 history number. The way @code{print} labels its output is designed to
11640 remind you of this. Just @code{$} refers to the most recent value in
11641 the history, and @code{$$} refers to the value before that.
11642 @code{$$@var{n}} refers to the @var{n}th value from the end; @code{$$2}
11643 is the value just prior to @code{$$}, @code{$$1} is equivalent to
11644 @code{$$}, and @code{$$0} is equivalent to @code{$}.
11646 For example, suppose you have just printed a pointer to a structure and
11647 want to see the contents of the structure. It suffices to type
11653 If you have a chain of structures where the component @code{next} points
11654 to the next one, you can print the contents of the next one with this:
11661 You can print successive links in the chain by repeating this
11662 command---which you can do by just typing @key{RET}.
11664 Note that the history records values, not expressions. If the value of
11665 @code{x} is 4 and you type these commands:
11673 then the value recorded in the value history by the @code{print} command
11674 remains 4 even though the value of @code{x} has changed.
11677 @kindex show values
11679 Print the last ten values in the value history, with their item numbers.
11680 This is like @samp{p@ $$9} repeated ten times, except that @code{show
11681 values} does not change the history.
11683 @item show values @var{n}
11684 Print ten history values centered on history item number @var{n}.
11686 @item show values +
11687 Print ten history values just after the values last printed. If no more
11688 values are available, @code{show values +} produces no display.
11691 Pressing @key{RET} to repeat @code{show values @var{n}} has exactly the
11692 same effect as @samp{show values +}.
11694 @node Convenience Vars
11695 @section Convenience Variables
11697 @cindex convenience variables
11698 @cindex user-defined variables
11699 @value{GDBN} provides @dfn{convenience variables} that you can use within
11700 @value{GDBN} to hold on to a value and refer to it later. These variables
11701 exist entirely within @value{GDBN}; they are not part of your program, and
11702 setting a convenience variable has no direct effect on further execution
11703 of your program. That is why you can use them freely.
11705 Convenience variables are prefixed with @samp{$}. Any name preceded by
11706 @samp{$} can be used for a convenience variable, unless it is one of
11707 the predefined machine-specific register names (@pxref{Registers, ,Registers}).
11708 (Value history references, in contrast, are @emph{numbers} preceded
11709 by @samp{$}. @xref{Value History, ,Value History}.)
11711 You can save a value in a convenience variable with an assignment
11712 expression, just as you would set a variable in your program.
11716 set $foo = *object_ptr
11720 would save in @code{$foo} the value contained in the object pointed to by
11723 Using a convenience variable for the first time creates it, but its
11724 value is @code{void} until you assign a new value. You can alter the
11725 value with another assignment at any time.
11727 Convenience variables have no fixed types. You can assign a convenience
11728 variable any type of value, including structures and arrays, even if
11729 that variable already has a value of a different type. The convenience
11730 variable, when used as an expression, has the type of its current value.
11733 @kindex show convenience
11734 @cindex show all user variables and functions
11735 @item show convenience
11736 Print a list of convenience variables used so far, and their values,
11737 as well as a list of the convenience functions.
11738 Abbreviated @code{show conv}.
11740 @kindex init-if-undefined
11741 @cindex convenience variables, initializing
11742 @item init-if-undefined $@var{variable} = @var{expression}
11743 Set a convenience variable if it has not already been set. This is useful
11744 for user-defined commands that keep some state. It is similar, in concept,
11745 to using local static variables with initializers in C (except that
11746 convenience variables are global). It can also be used to allow users to
11747 override default values used in a command script.
11749 If the variable is already defined then the expression is not evaluated so
11750 any side-effects do not occur.
11753 One of the ways to use a convenience variable is as a counter to be
11754 incremented or a pointer to be advanced. For example, to print
11755 a field from successive elements of an array of structures:
11759 print bar[$i++]->contents
11763 Repeat that command by typing @key{RET}.
11765 Some convenience variables are created automatically by @value{GDBN} and given
11766 values likely to be useful.
11769 @vindex $_@r{, convenience variable}
11771 The variable @code{$_} is automatically set by the @code{x} command to
11772 the last address examined (@pxref{Memory, ,Examining Memory}). Other
11773 commands which provide a default address for @code{x} to examine also
11774 set @code{$_} to that address; these commands include @code{info line}
11775 and @code{info breakpoint}. The type of @code{$_} is @code{void *}
11776 except when set by the @code{x} command, in which case it is a pointer
11777 to the type of @code{$__}.
11779 @vindex $__@r{, convenience variable}
11781 The variable @code{$__} is automatically set by the @code{x} command
11782 to the value found in the last address examined. Its type is chosen
11783 to match the format in which the data was printed.
11786 @vindex $_exitcode@r{, convenience variable}
11787 When the program being debugged terminates normally, @value{GDBN}
11788 automatically sets this variable to the exit code of the program, and
11789 resets @code{$_exitsignal} to @code{void}.
11792 @vindex $_exitsignal@r{, convenience variable}
11793 When the program being debugged dies due to an uncaught signal,
11794 @value{GDBN} automatically sets this variable to that signal's number,
11795 and resets @code{$_exitcode} to @code{void}.
11797 To distinguish between whether the program being debugged has exited
11798 (i.e., @code{$_exitcode} is not @code{void}) or signalled (i.e.,
11799 @code{$_exitsignal} is not @code{void}), the convenience function
11800 @code{$_isvoid} can be used (@pxref{Convenience Funs,, Convenience
11801 Functions}). For example, considering the following source code:
11804 #include <signal.h>
11807 main (int argc, char *argv[])
11814 A valid way of telling whether the program being debugged has exited
11815 or signalled would be:
11818 (@value{GDBP}) define has_exited_or_signalled
11819 Type commands for definition of ``has_exited_or_signalled''.
11820 End with a line saying just ``end''.
11821 >if $_isvoid ($_exitsignal)
11822 >echo The program has exited\n
11824 >echo The program has signalled\n
11830 Program terminated with signal SIGALRM, Alarm clock.
11831 The program no longer exists.
11832 (@value{GDBP}) has_exited_or_signalled
11833 The program has signalled
11836 As can be seen, @value{GDBN} correctly informs that the program being
11837 debugged has signalled, since it calls @code{raise} and raises a
11838 @code{SIGALRM} signal. If the program being debugged had not called
11839 @code{raise}, then @value{GDBN} would report a normal exit:
11842 (@value{GDBP}) has_exited_or_signalled
11843 The program has exited
11847 The variable @code{$_exception} is set to the exception object being
11848 thrown at an exception-related catchpoint. @xref{Set Catchpoints}.
11850 @item $_ada_exception
11851 The variable @code{$_ada_exception} is set to the address of the
11852 exception being caught or thrown at an Ada exception-related
11853 catchpoint. @xref{Set Catchpoints}.
11856 @itemx $_probe_arg0@dots{}$_probe_arg11
11857 Arguments to a static probe. @xref{Static Probe Points}.
11860 @vindex $_sdata@r{, inspect, convenience variable}
11861 The variable @code{$_sdata} contains extra collected static tracepoint
11862 data. @xref{Tracepoint Actions,,Tracepoint Action Lists}. Note that
11863 @code{$_sdata} could be empty, if not inspecting a trace buffer, or
11864 if extra static tracepoint data has not been collected.
11867 @vindex $_siginfo@r{, convenience variable}
11868 The variable @code{$_siginfo} contains extra signal information
11869 (@pxref{extra signal information}). Note that @code{$_siginfo}
11870 could be empty, if the application has not yet received any signals.
11871 For example, it will be empty before you execute the @code{run} command.
11874 @vindex $_tlb@r{, convenience variable}
11875 The variable @code{$_tlb} is automatically set when debugging
11876 applications running on MS-Windows in native mode or connected to
11877 gdbserver that supports the @code{qGetTIBAddr} request.
11878 @xref{General Query Packets}.
11879 This variable contains the address of the thread information block.
11882 The number of the current inferior. @xref{Inferiors and
11883 Programs, ,Debugging Multiple Inferiors and Programs}.
11886 The thread number of the current thread. @xref{thread numbers}.
11889 The global number of the current thread. @xref{global thread numbers}.
11893 @vindex $_gdb_major@r{, convenience variable}
11894 @vindex $_gdb_minor@r{, convenience variable}
11895 The major and minor version numbers of the running @value{GDBN}.
11896 Development snapshots and pretest versions have their minor version
11897 incremented by one; thus, @value{GDBN} pretest 9.11.90 will produce
11898 the value 12 for @code{$_gdb_minor}. These variables allow you to
11899 write scripts that work with different versions of @value{GDBN}
11900 without errors caused by features unavailable in some of those
11903 @item $_shell_exitcode
11904 @itemx $_shell_exitsignal
11905 @vindex $_shell_exitcode@r{, convenience variable}
11906 @vindex $_shell_exitsignal@r{, convenience variable}
11907 @cindex shell command, exit code
11908 @cindex shell command, exit signal
11909 @cindex exit status of shell commands
11910 @value{GDBN} commands such as @code{shell} and @code{|} are launching
11911 shell commands. When a launched command terminates, @value{GDBN}
11912 automatically maintains the variables @code{$_shell_exitcode}
11913 and @code{$_shell_exitsignal} according to the exit status of the last
11914 launched command. These variables are set and used similarly to
11915 the variables @code{$_exitcode} and @code{$_exitsignal}.
11919 @node Convenience Funs
11920 @section Convenience Functions
11922 @cindex convenience functions
11923 @value{GDBN} also supplies some @dfn{convenience functions}. These
11924 have a syntax similar to convenience variables. A convenience
11925 function can be used in an expression just like an ordinary function;
11926 however, a convenience function is implemented internally to
11929 These functions do not require @value{GDBN} to be configured with
11930 @code{Python} support, which means that they are always available.
11934 @item $_isvoid (@var{expr})
11935 @findex $_isvoid@r{, convenience function}
11936 Return one if the expression @var{expr} is @code{void}. Otherwise it
11939 A @code{void} expression is an expression where the type of the result
11940 is @code{void}. For example, you can examine a convenience variable
11941 (see @ref{Convenience Vars,, Convenience Variables}) to check whether
11945 (@value{GDBP}) print $_exitcode
11947 (@value{GDBP}) print $_isvoid ($_exitcode)
11950 Starting program: ./a.out
11951 [Inferior 1 (process 29572) exited normally]
11952 (@value{GDBP}) print $_exitcode
11954 (@value{GDBP}) print $_isvoid ($_exitcode)
11958 In the example above, we used @code{$_isvoid} to check whether
11959 @code{$_exitcode} is @code{void} before and after the execution of the
11960 program being debugged. Before the execution there is no exit code to
11961 be examined, therefore @code{$_exitcode} is @code{void}. After the
11962 execution the program being debugged returned zero, therefore
11963 @code{$_exitcode} is zero, which means that it is not @code{void}
11966 The @code{void} expression can also be a call of a function from the
11967 program being debugged. For example, given the following function:
11976 The result of calling it inside @value{GDBN} is @code{void}:
11979 (@value{GDBP}) print foo ()
11981 (@value{GDBP}) print $_isvoid (foo ())
11983 (@value{GDBP}) set $v = foo ()
11984 (@value{GDBP}) print $v
11986 (@value{GDBP}) print $_isvoid ($v)
11990 @item $_gdb_setting_str (@var{setting})
11991 @findex $_gdb_setting_str@r{, convenience function}
11992 Return the value of the @value{GDBN} @var{setting} as a string.
11993 @var{setting} is any setting that can be used in a @code{set} or
11994 @code{show} command (@pxref{Controlling GDB}).
11997 (@value{GDBP}) show print frame-arguments
11998 Printing of non-scalar frame arguments is "scalars".
11999 (@value{GDBP}) p $_gdb_setting_str("print frame-arguments")
12001 (@value{GDBP}) p $_gdb_setting_str("height")
12006 @item $_gdb_setting (@var{setting})
12007 @findex $_gdb_setting@r{, convenience function}
12008 Return the value of the @value{GDBN} @var{setting}.
12009 The type of the returned value depends on the setting.
12011 The value type for boolean and auto boolean settings is @code{int}.
12012 The boolean values @code{off} and @code{on} are converted to
12013 the integer values @code{0} and @code{1}. The value @code{auto} is
12014 converted to the value @code{-1}.
12016 The value type for integer settings is either @code{unsigned int}
12017 or @code{int}, depending on the setting.
12019 Some integer settings accept an @code{unlimited} value.
12020 Depending on the setting, the @code{set} command also accepts
12021 the value @code{0} or the value @code{@minus{}1} as a synonym for
12023 For example, @code{set height unlimited} is equivalent to
12024 @code{set height 0}.
12026 Some other settings that accept the @code{unlimited} value
12027 use the value @code{0} to literally mean zero.
12028 For example, @code{set history size 0} indicates to not
12029 record any @value{GDBN} commands in the command history.
12030 For such settings, @code{@minus{}1} is the synonym
12031 for @code{unlimited}.
12033 See the documentation of the corresponding @code{set} command for
12034 the numerical value equivalent to @code{unlimited}.
12036 The @code{$_gdb_setting} function converts the unlimited value
12037 to a @code{0} or a @code{@minus{}1} value according to what the
12038 @code{set} command uses.
12042 (@value{GDBP}) p $_gdb_setting_str("height")
12044 (@value{GDBP}) p $_gdb_setting("height")
12046 (@value{GDBP}) set height unlimited
12047 (@value{GDBP}) p $_gdb_setting_str("height")
12049 (@value{GDBP}) p $_gdb_setting("height")
12053 (@value{GDBP}) p $_gdb_setting_str("history size")
12055 (@value{GDBP}) p $_gdb_setting("history size")
12057 (@value{GDBP}) p $_gdb_setting_str("disassemble-next-line")
12059 (@value{GDBP}) p $_gdb_setting("disassemble-next-line")
12065 Other setting types (enum, filename, optional filename, string, string noescape)
12066 are returned as string values.
12069 @item $_gdb_maint_setting_str (@var{setting})
12070 @findex $_gdb_maint_setting_str@r{, convenience function}
12071 Like the @code{$_gdb_setting_str} function, but works with
12072 @code{maintenance set} variables.
12074 @item $_gdb_maint_setting (@var{setting})
12075 @findex $_gdb_maint_setting@r{, convenience function}
12076 Like the @code{$_gdb_setting} function, but works with
12077 @code{maintenance set} variables.
12081 The following functions require @value{GDBN} to be configured with
12082 @code{Python} support.
12086 @item $_memeq(@var{buf1}, @var{buf2}, @var{length})
12087 @findex $_memeq@r{, convenience function}
12088 Returns one if the @var{length} bytes at the addresses given by
12089 @var{buf1} and @var{buf2} are equal.
12090 Otherwise it returns zero.
12092 @item $_regex(@var{str}, @var{regex})
12093 @findex $_regex@r{, convenience function}
12094 Returns one if the string @var{str} matches the regular expression
12095 @var{regex}. Otherwise it returns zero.
12096 The syntax of the regular expression is that specified by @code{Python}'s
12097 regular expression support.
12099 @item $_streq(@var{str1}, @var{str2})
12100 @findex $_streq@r{, convenience function}
12101 Returns one if the strings @var{str1} and @var{str2} are equal.
12102 Otherwise it returns zero.
12104 @item $_strlen(@var{str})
12105 @findex $_strlen@r{, convenience function}
12106 Returns the length of string @var{str}.
12108 @item $_caller_is(@var{name}@r{[}, @var{number_of_frames}@r{]})
12109 @findex $_caller_is@r{, convenience function}
12110 Returns one if the calling function's name is equal to @var{name}.
12111 Otherwise it returns zero.
12113 If the optional argument @var{number_of_frames} is provided,
12114 it is the number of frames up in the stack to look.
12122 at testsuite/gdb.python/py-caller-is.c:21
12123 #1 0x00000000004005a0 in middle_func ()
12124 at testsuite/gdb.python/py-caller-is.c:27
12125 #2 0x00000000004005ab in top_func ()
12126 at testsuite/gdb.python/py-caller-is.c:33
12127 #3 0x00000000004005b6 in main ()
12128 at testsuite/gdb.python/py-caller-is.c:39
12129 (gdb) print $_caller_is ("middle_func")
12131 (gdb) print $_caller_is ("top_func", 2)
12135 @item $_caller_matches(@var{regexp}@r{[}, @var{number_of_frames}@r{]})
12136 @findex $_caller_matches@r{, convenience function}
12137 Returns one if the calling function's name matches the regular expression
12138 @var{regexp}. Otherwise it returns zero.
12140 If the optional argument @var{number_of_frames} is provided,
12141 it is the number of frames up in the stack to look.
12144 @item $_any_caller_is(@var{name}@r{[}, @var{number_of_frames}@r{]})
12145 @findex $_any_caller_is@r{, convenience function}
12146 Returns one if any calling function's name is equal to @var{name}.
12147 Otherwise it returns zero.
12149 If the optional argument @var{number_of_frames} is provided,
12150 it is the number of frames up in the stack to look.
12153 This function differs from @code{$_caller_is} in that this function
12154 checks all stack frames from the immediate caller to the frame specified
12155 by @var{number_of_frames}, whereas @code{$_caller_is} only checks the
12156 frame specified by @var{number_of_frames}.
12158 @item $_any_caller_matches(@var{regexp}@r{[}, @var{number_of_frames}@r{]})
12159 @findex $_any_caller_matches@r{, convenience function}
12160 Returns one if any calling function's name matches the regular expression
12161 @var{regexp}. Otherwise it returns zero.
12163 If the optional argument @var{number_of_frames} is provided,
12164 it is the number of frames up in the stack to look.
12167 This function differs from @code{$_caller_matches} in that this function
12168 checks all stack frames from the immediate caller to the frame specified
12169 by @var{number_of_frames}, whereas @code{$_caller_matches} only checks the
12170 frame specified by @var{number_of_frames}.
12172 @item $_as_string(@var{value})
12173 @findex $_as_string@r{, convenience function}
12174 Return the string representation of @var{value}.
12176 This function is useful to obtain the textual label (enumerator) of an
12177 enumeration value. For example, assuming the variable @var{node} is of
12178 an enumerated type:
12181 (gdb) printf "Visiting node of type %s\n", $_as_string(node)
12182 Visiting node of type NODE_INTEGER
12185 @item $_cimag(@var{value})
12186 @itemx $_creal(@var{value})
12187 @findex $_cimag@r{, convenience function}
12188 @findex $_creal@r{, convenience function}
12189 Return the imaginary (@code{$_cimag}) or real (@code{$_creal}) part of
12190 the complex number @var{value}.
12192 The type of the imaginary or real part depends on the type of the
12193 complex number, e.g., using @code{$_cimag} on a @code{float complex}
12194 will return an imaginary part of type @code{float}.
12198 @value{GDBN} provides the ability to list and get help on
12199 convenience functions.
12202 @item help function
12203 @kindex help function
12204 @cindex show all convenience functions
12205 Print a list of all convenience functions.
12212 You can refer to machine register contents, in expressions, as variables
12213 with names starting with @samp{$}. The names of registers are different
12214 for each machine; use @code{info registers} to see the names used on
12218 @kindex info registers
12219 @item info registers
12220 Print the names and values of all registers except floating-point
12221 and vector registers (in the selected stack frame).
12223 @kindex info all-registers
12224 @cindex floating point registers
12225 @item info all-registers
12226 Print the names and values of all registers, including floating-point
12227 and vector registers (in the selected stack frame).
12229 @item info registers @var{reggroup} @dots{}
12230 Print the name and value of the registers in each of the specified
12231 @var{reggroup}s. The @var{reggroup} can be any of those returned by
12232 @code{maint print reggroups} (@pxref{Maintenance Commands}).
12234 @item info registers @var{regname} @dots{}
12235 Print the @dfn{relativized} value of each specified register @var{regname}.
12236 As discussed in detail below, register values are normally relative to
12237 the selected stack frame. The @var{regname} may be any register name valid on
12238 the machine you are using, with or without the initial @samp{$}.
12241 @anchor{standard registers}
12242 @cindex stack pointer register
12243 @cindex program counter register
12244 @cindex process status register
12245 @cindex frame pointer register
12246 @cindex standard registers
12247 @value{GDBN} has four ``standard'' register names that are available (in
12248 expressions) on most machines---whenever they do not conflict with an
12249 architecture's canonical mnemonics for registers. The register names
12250 @code{$pc} and @code{$sp} are used for the program counter register and
12251 the stack pointer. @code{$fp} is used for a register that contains a
12252 pointer to the current stack frame, and @code{$ps} is used for a
12253 register that contains the processor status. For example,
12254 you could print the program counter in hex with
12261 or print the instruction to be executed next with
12268 or add four to the stack pointer@footnote{This is a way of removing
12269 one word from the stack, on machines where stacks grow downward in
12270 memory (most machines, nowadays). This assumes that the innermost
12271 stack frame is selected; setting @code{$sp} is not allowed when other
12272 stack frames are selected. To pop entire frames off the stack,
12273 regardless of machine architecture, use @code{return};
12274 see @ref{Returning, ,Returning from a Function}.} with
12280 Whenever possible, these four standard register names are available on
12281 your machine even though the machine has different canonical mnemonics,
12282 so long as there is no conflict. The @code{info registers} command
12283 shows the canonical names. For example, on the SPARC, @code{info
12284 registers} displays the processor status register as @code{$psr} but you
12285 can also refer to it as @code{$ps}; and on x86-based machines @code{$ps}
12286 is an alias for the @sc{eflags} register.
12288 @value{GDBN} always considers the contents of an ordinary register as an
12289 integer when the register is examined in this way. Some machines have
12290 special registers which can hold nothing but floating point; these
12291 registers are considered to have floating point values. There is no way
12292 to refer to the contents of an ordinary register as floating point value
12293 (although you can @emph{print} it as a floating point value with
12294 @samp{print/f $@var{regname}}).
12296 Some registers have distinct ``raw'' and ``virtual'' data formats. This
12297 means that the data format in which the register contents are saved by
12298 the operating system is not the same one that your program normally
12299 sees. For example, the registers of the 68881 floating point
12300 coprocessor are always saved in ``extended'' (raw) format, but all C
12301 programs expect to work with ``double'' (virtual) format. In such
12302 cases, @value{GDBN} normally works with the virtual format only (the format
12303 that makes sense for your program), but the @code{info registers} command
12304 prints the data in both formats.
12306 @cindex SSE registers (x86)
12307 @cindex MMX registers (x86)
12308 Some machines have special registers whose contents can be interpreted
12309 in several different ways. For example, modern x86-based machines
12310 have SSE and MMX registers that can hold several values packed
12311 together in several different formats. @value{GDBN} refers to such
12312 registers in @code{struct} notation:
12315 (@value{GDBP}) print $xmm1
12317 v4_float = @{0, 3.43859137e-038, 1.54142831e-044, 1.821688e-044@},
12318 v2_double = @{9.92129282474342e-303, 2.7585945287983262e-313@},
12319 v16_int8 = "\000\000\000\000\3706;\001\v\000\000\000\r\000\000",
12320 v8_int16 = @{0, 0, 14072, 315, 11, 0, 13, 0@},
12321 v4_int32 = @{0, 20657912, 11, 13@},
12322 v2_int64 = @{88725056443645952, 55834574859@},
12323 uint128 = 0x0000000d0000000b013b36f800000000
12328 To set values of such registers, you need to tell @value{GDBN} which
12329 view of the register you wish to change, as if you were assigning
12330 value to a @code{struct} member:
12333 (@value{GDBP}) set $xmm1.uint128 = 0x000000000000000000000000FFFFFFFF
12336 Normally, register values are relative to the selected stack frame
12337 (@pxref{Selection, ,Selecting a Frame}). This means that you get the
12338 value that the register would contain if all stack frames farther in
12339 were exited and their saved registers restored. In order to see the
12340 true contents of hardware registers, you must select the innermost
12341 frame (with @samp{frame 0}).
12343 @cindex caller-saved registers
12344 @cindex call-clobbered registers
12345 @cindex volatile registers
12346 @cindex <not saved> values
12347 Usually ABIs reserve some registers as not needed to be saved by the
12348 callee (a.k.a.: ``caller-saved'', ``call-clobbered'' or ``volatile''
12349 registers). It may therefore not be possible for @value{GDBN} to know
12350 the value a register had before the call (in other words, in the outer
12351 frame), if the register value has since been changed by the callee.
12352 @value{GDBN} tries to deduce where the inner frame saved
12353 (``callee-saved'') registers, from the debug info, unwind info, or the
12354 machine code generated by your compiler. If some register is not
12355 saved, and @value{GDBN} knows the register is ``caller-saved'' (via
12356 its own knowledge of the ABI, or because the debug/unwind info
12357 explicitly says the register's value is undefined), @value{GDBN}
12358 displays @w{@samp{<not saved>}} as the register's value. With targets
12359 that @value{GDBN} has no knowledge of the register saving convention,
12360 if a register was not saved by the callee, then its value and location
12361 in the outer frame are assumed to be the same of the inner frame.
12362 This is usually harmless, because if the register is call-clobbered,
12363 the caller either does not care what is in the register after the
12364 call, or has code to restore the value that it does care about. Note,
12365 however, that if you change such a register in the outer frame, you
12366 may also be affecting the inner frame. Also, the more ``outer'' the
12367 frame is you're looking at, the more likely a call-clobbered
12368 register's value is to be wrong, in the sense that it doesn't actually
12369 represent the value the register had just before the call.
12371 @node Floating Point Hardware
12372 @section Floating Point Hardware
12373 @cindex floating point
12375 Depending on the configuration, @value{GDBN} may be able to give
12376 you more information about the status of the floating point hardware.
12381 Display hardware-dependent information about the floating
12382 point unit. The exact contents and layout vary depending on the
12383 floating point chip. Currently, @samp{info float} is supported on
12384 the ARM and x86 machines.
12388 @section Vector Unit
12389 @cindex vector unit
12391 Depending on the configuration, @value{GDBN} may be able to give you
12392 more information about the status of the vector unit.
12395 @kindex info vector
12397 Display information about the vector unit. The exact contents and
12398 layout vary depending on the hardware.
12401 @node OS Information
12402 @section Operating System Auxiliary Information
12403 @cindex OS information
12405 @value{GDBN} provides interfaces to useful OS facilities that can help
12406 you debug your program.
12408 @cindex auxiliary vector
12409 @cindex vector, auxiliary
12410 Some operating systems supply an @dfn{auxiliary vector} to programs at
12411 startup. This is akin to the arguments and environment that you
12412 specify for a program, but contains a system-dependent variety of
12413 binary values that tell system libraries important details about the
12414 hardware, operating system, and process. Each value's purpose is
12415 identified by an integer tag; the meanings are well-known but system-specific.
12416 Depending on the configuration and operating system facilities,
12417 @value{GDBN} may be able to show you this information. For remote
12418 targets, this functionality may further depend on the remote stub's
12419 support of the @samp{qXfer:auxv:read} packet, see
12420 @ref{qXfer auxiliary vector read}.
12425 Display the auxiliary vector of the inferior, which can be either a
12426 live process or a core dump file. @value{GDBN} prints each tag value
12427 numerically, and also shows names and text descriptions for recognized
12428 tags. Some values in the vector are numbers, some bit masks, and some
12429 pointers to strings or other data. @value{GDBN} displays each value in the
12430 most appropriate form for a recognized tag, and in hexadecimal for
12431 an unrecognized tag.
12434 On some targets, @value{GDBN} can access operating system-specific
12435 information and show it to you. The types of information available
12436 will differ depending on the type of operating system running on the
12437 target. The mechanism used to fetch the data is described in
12438 @ref{Operating System Information}. For remote targets, this
12439 functionality depends on the remote stub's support of the
12440 @samp{qXfer:osdata:read} packet, see @ref{qXfer osdata read}.
12444 @item info os @var{infotype}
12446 Display OS information of the requested type.
12448 On @sc{gnu}/Linux, the following values of @var{infotype} are valid:
12450 @anchor{linux info os infotypes}
12452 @kindex info os cpus
12454 Display the list of all CPUs/cores. For each CPU/core, @value{GDBN} prints
12455 the available fields from /proc/cpuinfo. For each supported architecture
12456 different fields are available. Two common entries are processor which gives
12457 CPU number and bogomips; a system constant that is calculated during
12458 kernel initialization.
12460 @kindex info os files
12462 Display the list of open file descriptors on the target. For each
12463 file descriptor, @value{GDBN} prints the identifier of the process
12464 owning the descriptor, the command of the owning process, the value
12465 of the descriptor, and the target of the descriptor.
12467 @kindex info os modules
12469 Display the list of all loaded kernel modules on the target. For each
12470 module, @value{GDBN} prints the module name, the size of the module in
12471 bytes, the number of times the module is used, the dependencies of the
12472 module, the status of the module, and the address of the loaded module
12475 @kindex info os msg
12477 Display the list of all System V message queues on the target. For each
12478 message queue, @value{GDBN} prints the message queue key, the message
12479 queue identifier, the access permissions, the current number of bytes
12480 on the queue, the current number of messages on the queue, the processes
12481 that last sent and received a message on the queue, the user and group
12482 of the owner and creator of the message queue, the times at which a
12483 message was last sent and received on the queue, and the time at which
12484 the message queue was last changed.
12486 @kindex info os processes
12488 Display the list of processes on the target. For each process,
12489 @value{GDBN} prints the process identifier, the name of the user, the
12490 command corresponding to the process, and the list of processor cores
12491 that the process is currently running on. (To understand what these
12492 properties mean, for this and the following info types, please consult
12493 the general @sc{gnu}/Linux documentation.)
12495 @kindex info os procgroups
12497 Display the list of process groups on the target. For each process,
12498 @value{GDBN} prints the identifier of the process group that it belongs
12499 to, the command corresponding to the process group leader, the process
12500 identifier, and the command line of the process. The list is sorted
12501 first by the process group identifier, then by the process identifier,
12502 so that processes belonging to the same process group are grouped together
12503 and the process group leader is listed first.
12505 @kindex info os semaphores
12507 Display the list of all System V semaphore sets on the target. For each
12508 semaphore set, @value{GDBN} prints the semaphore set key, the semaphore
12509 set identifier, the access permissions, the number of semaphores in the
12510 set, the user and group of the owner and creator of the semaphore set,
12511 and the times at which the semaphore set was operated upon and changed.
12513 @kindex info os shm
12515 Display the list of all System V shared-memory regions on the target.
12516 For each shared-memory region, @value{GDBN} prints the region key,
12517 the shared-memory identifier, the access permissions, the size of the
12518 region, the process that created the region, the process that last
12519 attached to or detached from the region, the current number of live
12520 attaches to the region, and the times at which the region was last
12521 attached to, detach from, and changed.
12523 @kindex info os sockets
12525 Display the list of Internet-domain sockets on the target. For each
12526 socket, @value{GDBN} prints the address and port of the local and
12527 remote endpoints, the current state of the connection, the creator of
12528 the socket, the IP address family of the socket, and the type of the
12531 @kindex info os threads
12533 Display the list of threads running on the target. For each thread,
12534 @value{GDBN} prints the identifier of the process that the thread
12535 belongs to, the command of the process, the thread identifier, and the
12536 processor core that it is currently running on. The main thread of a
12537 process is not listed.
12541 If @var{infotype} is omitted, then list the possible values for
12542 @var{infotype} and the kind of OS information available for each
12543 @var{infotype}. If the target does not return a list of possible
12544 types, this command will report an error.
12547 @node Memory Region Attributes
12548 @section Memory Region Attributes
12549 @cindex memory region attributes
12551 @dfn{Memory region attributes} allow you to describe special handling
12552 required by regions of your target's memory. @value{GDBN} uses
12553 attributes to determine whether to allow certain types of memory
12554 accesses; whether to use specific width accesses; and whether to cache
12555 target memory. By default the description of memory regions is
12556 fetched from the target (if the current target supports this), but the
12557 user can override the fetched regions.
12559 Defined memory regions can be individually enabled and disabled. When a
12560 memory region is disabled, @value{GDBN} uses the default attributes when
12561 accessing memory in that region. Similarly, if no memory regions have
12562 been defined, @value{GDBN} uses the default attributes when accessing
12565 When a memory region is defined, it is given a number to identify it;
12566 to enable, disable, or remove a memory region, you specify that number.
12570 @item mem @var{lower} @var{upper} @var{attributes}@dots{}
12571 Define a memory region bounded by @var{lower} and @var{upper} with
12572 attributes @var{attributes}@dots{}, and add it to the list of regions
12573 monitored by @value{GDBN}. Note that @var{upper} == 0 is a special
12574 case: it is treated as the target's maximum memory address.
12575 (0xffff on 16 bit targets, 0xffffffff on 32 bit targets, etc.)
12578 Discard any user changes to the memory regions and use target-supplied
12579 regions, if available, or no regions if the target does not support.
12582 @item delete mem @var{nums}@dots{}
12583 Remove memory regions @var{nums}@dots{} from the list of regions
12584 monitored by @value{GDBN}.
12586 @kindex disable mem
12587 @item disable mem @var{nums}@dots{}
12588 Disable monitoring of memory regions @var{nums}@dots{}.
12589 A disabled memory region is not forgotten.
12590 It may be enabled again later.
12593 @item enable mem @var{nums}@dots{}
12594 Enable monitoring of memory regions @var{nums}@dots{}.
12598 Print a table of all defined memory regions, with the following columns
12602 @item Memory Region Number
12603 @item Enabled or Disabled.
12604 Enabled memory regions are marked with @samp{y}.
12605 Disabled memory regions are marked with @samp{n}.
12608 The address defining the inclusive lower bound of the memory region.
12611 The address defining the exclusive upper bound of the memory region.
12614 The list of attributes set for this memory region.
12619 @subsection Attributes
12621 @subsubsection Memory Access Mode
12622 The access mode attributes set whether @value{GDBN} may make read or
12623 write accesses to a memory region.
12625 While these attributes prevent @value{GDBN} from performing invalid
12626 memory accesses, they do nothing to prevent the target system, I/O DMA,
12627 etc.@: from accessing memory.
12631 Memory is read only.
12633 Memory is write only.
12635 Memory is read/write. This is the default.
12638 @subsubsection Memory Access Size
12639 The access size attribute tells @value{GDBN} to use specific sized
12640 accesses in the memory region. Often memory mapped device registers
12641 require specific sized accesses. If no access size attribute is
12642 specified, @value{GDBN} may use accesses of any size.
12646 Use 8 bit memory accesses.
12648 Use 16 bit memory accesses.
12650 Use 32 bit memory accesses.
12652 Use 64 bit memory accesses.
12655 @c @subsubsection Hardware/Software Breakpoints
12656 @c The hardware/software breakpoint attributes set whether @value{GDBN}
12657 @c will use hardware or software breakpoints for the internal breakpoints
12658 @c used by the step, next, finish, until, etc. commands.
12662 @c Always use hardware breakpoints
12663 @c @item swbreak (default)
12666 @subsubsection Data Cache
12667 The data cache attributes set whether @value{GDBN} will cache target
12668 memory. While this generally improves performance by reducing debug
12669 protocol overhead, it can lead to incorrect results because @value{GDBN}
12670 does not know about volatile variables or memory mapped device
12675 Enable @value{GDBN} to cache target memory.
12677 Disable @value{GDBN} from caching target memory. This is the default.
12680 @subsection Memory Access Checking
12681 @value{GDBN} can be instructed to refuse accesses to memory that is
12682 not explicitly described. This can be useful if accessing such
12683 regions has undesired effects for a specific target, or to provide
12684 better error checking. The following commands control this behaviour.
12687 @kindex set mem inaccessible-by-default
12688 @item set mem inaccessible-by-default [on|off]
12689 If @code{on} is specified, make @value{GDBN} treat memory not
12690 explicitly described by the memory ranges as non-existent and refuse accesses
12691 to such memory. The checks are only performed if there's at least one
12692 memory range defined. If @code{off} is specified, make @value{GDBN}
12693 treat the memory not explicitly described by the memory ranges as RAM.
12694 The default value is @code{on}.
12695 @kindex show mem inaccessible-by-default
12696 @item show mem inaccessible-by-default
12697 Show the current handling of accesses to unknown memory.
12701 @c @subsubsection Memory Write Verification
12702 @c The memory write verification attributes set whether @value{GDBN}
12703 @c will re-reads data after each write to verify the write was successful.
12707 @c @item noverify (default)
12710 @node Dump/Restore Files
12711 @section Copy Between Memory and a File
12712 @cindex dump/restore files
12713 @cindex append data to a file
12714 @cindex dump data to a file
12715 @cindex restore data from a file
12717 You can use the commands @code{dump}, @code{append}, and
12718 @code{restore} to copy data between target memory and a file. The
12719 @code{dump} and @code{append} commands write data to a file, and the
12720 @code{restore} command reads data from a file back into the inferior's
12721 memory. Files may be in binary, Motorola S-record, Intel hex,
12722 Tektronix Hex, or Verilog Hex format; however, @value{GDBN} can only
12723 append to binary files, and cannot read from Verilog Hex files.
12728 @item dump @r{[}@var{format}@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
12729 @itemx dump @r{[}@var{format}@r{]} value @var{filename} @var{expr}
12730 Dump the contents of memory from @var{start_addr} to @var{end_addr},
12731 or the value of @var{expr}, to @var{filename} in the given format.
12733 The @var{format} parameter may be any one of:
12740 Motorola S-record format.
12742 Tektronix Hex format.
12744 Verilog Hex format.
12747 @value{GDBN} uses the same definitions of these formats as the
12748 @sc{gnu} binary utilities, like @samp{objdump} and @samp{objcopy}. If
12749 @var{format} is omitted, @value{GDBN} dumps the data in raw binary
12753 @item append @r{[}binary@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
12754 @itemx append @r{[}binary@r{]} value @var{filename} @var{expr}
12755 Append the contents of memory from @var{start_addr} to @var{end_addr},
12756 or the value of @var{expr}, to the file @var{filename}, in raw binary form.
12757 (@value{GDBN} can only append data to files in raw binary form.)
12760 @item restore @var{filename} @r{[}binary@r{]} @var{bias} @var{start} @var{end}
12761 Restore the contents of file @var{filename} into memory. The
12762 @code{restore} command can automatically recognize any known @sc{bfd}
12763 file format, except for raw binary. To restore a raw binary file you
12764 must specify the optional keyword @code{binary} after the filename.
12766 If @var{bias} is non-zero, its value will be added to the addresses
12767 contained in the file. Binary files always start at address zero, so
12768 they will be restored at address @var{bias}. Other bfd files have
12769 a built-in location; they will be restored at offset @var{bias}
12770 from that location.
12772 If @var{start} and/or @var{end} are non-zero, then only data between
12773 file offset @var{start} and file offset @var{end} will be restored.
12774 These offsets are relative to the addresses in the file, before
12775 the @var{bias} argument is applied.
12779 @node Core File Generation
12780 @section How to Produce a Core File from Your Program
12781 @cindex dump core from inferior
12783 A @dfn{core file} or @dfn{core dump} is a file that records the memory
12784 image of a running process and its process status (register values
12785 etc.). Its primary use is post-mortem debugging of a program that
12786 crashed while it ran outside a debugger. A program that crashes
12787 automatically produces a core file, unless this feature is disabled by
12788 the user. @xref{Files}, for information on invoking @value{GDBN} in
12789 the post-mortem debugging mode.
12791 Occasionally, you may wish to produce a core file of the program you
12792 are debugging in order to preserve a snapshot of its state.
12793 @value{GDBN} has a special command for that.
12797 @kindex generate-core-file
12798 @item generate-core-file [@var{file}]
12799 @itemx gcore [@var{file}]
12800 Produce a core dump of the inferior process. The optional argument
12801 @var{file} specifies the file name where to put the core dump. If not
12802 specified, the file name defaults to @file{core.@var{pid}}, where
12803 @var{pid} is the inferior process ID.
12805 Note that this command is implemented only for some systems (as of
12806 this writing, @sc{gnu}/Linux, FreeBSD, Solaris, and S390).
12808 On @sc{gnu}/Linux, this command can take into account the value of the
12809 file @file{/proc/@var{pid}/coredump_filter} when generating the core
12810 dump (@pxref{set use-coredump-filter}), and by default honors the
12811 @code{VM_DONTDUMP} flag for mappings where it is present in the file
12812 @file{/proc/@var{pid}/smaps} (@pxref{set dump-excluded-mappings}).
12814 @kindex set use-coredump-filter
12815 @anchor{set use-coredump-filter}
12816 @item set use-coredump-filter on
12817 @itemx set use-coredump-filter off
12818 Enable or disable the use of the file
12819 @file{/proc/@var{pid}/coredump_filter} when generating core dump
12820 files. This file is used by the Linux kernel to decide what types of
12821 memory mappings will be dumped or ignored when generating a core dump
12822 file. @var{pid} is the process ID of a currently running process.
12824 To make use of this feature, you have to write in the
12825 @file{/proc/@var{pid}/coredump_filter} file a value, in hexadecimal,
12826 which is a bit mask representing the memory mapping types. If a bit
12827 is set in the bit mask, then the memory mappings of the corresponding
12828 types will be dumped; otherwise, they will be ignored. This
12829 configuration is inherited by child processes. For more information
12830 about the bits that can be set in the
12831 @file{/proc/@var{pid}/coredump_filter} file, please refer to the
12832 manpage of @code{core(5)}.
12834 By default, this option is @code{on}. If this option is turned
12835 @code{off}, @value{GDBN} does not read the @file{coredump_filter} file
12836 and instead uses the same default value as the Linux kernel in order
12837 to decide which pages will be dumped in the core dump file. This
12838 value is currently @code{0x33}, which means that bits @code{0}
12839 (anonymous private mappings), @code{1} (anonymous shared mappings),
12840 @code{4} (ELF headers) and @code{5} (private huge pages) are active.
12841 This will cause these memory mappings to be dumped automatically.
12843 @kindex set dump-excluded-mappings
12844 @anchor{set dump-excluded-mappings}
12845 @item set dump-excluded-mappings on
12846 @itemx set dump-excluded-mappings off
12847 If @code{on} is specified, @value{GDBN} will dump memory mappings
12848 marked with the @code{VM_DONTDUMP} flag. This flag is represented in
12849 the file @file{/proc/@var{pid}/smaps} with the acronym @code{dd}.
12851 The default value is @code{off}.
12854 @node Character Sets
12855 @section Character Sets
12856 @cindex character sets
12858 @cindex translating between character sets
12859 @cindex host character set
12860 @cindex target character set
12862 If the program you are debugging uses a different character set to
12863 represent characters and strings than the one @value{GDBN} uses itself,
12864 @value{GDBN} can automatically translate between the character sets for
12865 you. The character set @value{GDBN} uses we call the @dfn{host
12866 character set}; the one the inferior program uses we call the
12867 @dfn{target character set}.
12869 For example, if you are running @value{GDBN} on a @sc{gnu}/Linux system, which
12870 uses the ISO Latin 1 character set, but you are using @value{GDBN}'s
12871 remote protocol (@pxref{Remote Debugging}) to debug a program
12872 running on an IBM mainframe, which uses the @sc{ebcdic} character set,
12873 then the host character set is Latin-1, and the target character set is
12874 @sc{ebcdic}. If you give @value{GDBN} the command @code{set
12875 target-charset EBCDIC-US}, then @value{GDBN} translates between
12876 @sc{ebcdic} and Latin 1 as you print character or string values, or use
12877 character and string literals in expressions.
12879 @value{GDBN} has no way to automatically recognize which character set
12880 the inferior program uses; you must tell it, using the @code{set
12881 target-charset} command, described below.
12883 Here are the commands for controlling @value{GDBN}'s character set
12887 @item set target-charset @var{charset}
12888 @kindex set target-charset
12889 Set the current target character set to @var{charset}. To display the
12890 list of supported target character sets, type
12891 @kbd{@w{set target-charset @key{TAB}@key{TAB}}}.
12893 @item set host-charset @var{charset}
12894 @kindex set host-charset
12895 Set the current host character set to @var{charset}.
12897 By default, @value{GDBN} uses a host character set appropriate to the
12898 system it is running on; you can override that default using the
12899 @code{set host-charset} command. On some systems, @value{GDBN} cannot
12900 automatically determine the appropriate host character set. In this
12901 case, @value{GDBN} uses @samp{UTF-8}.
12903 @value{GDBN} can only use certain character sets as its host character
12904 set. If you type @kbd{@w{set host-charset @key{TAB}@key{TAB}}},
12905 @value{GDBN} will list the host character sets it supports.
12907 @item set charset @var{charset}
12908 @kindex set charset
12909 Set the current host and target character sets to @var{charset}. As
12910 above, if you type @kbd{@w{set charset @key{TAB}@key{TAB}}},
12911 @value{GDBN} will list the names of the character sets that can be used
12912 for both host and target.
12915 @kindex show charset
12916 Show the names of the current host and target character sets.
12918 @item show host-charset
12919 @kindex show host-charset
12920 Show the name of the current host character set.
12922 @item show target-charset
12923 @kindex show target-charset
12924 Show the name of the current target character set.
12926 @item set target-wide-charset @var{charset}
12927 @kindex set target-wide-charset
12928 Set the current target's wide character set to @var{charset}. This is
12929 the character set used by the target's @code{wchar_t} type. To
12930 display the list of supported wide character sets, type
12931 @kbd{@w{set target-wide-charset @key{TAB}@key{TAB}}}.
12933 @item show target-wide-charset
12934 @kindex show target-wide-charset
12935 Show the name of the current target's wide character set.
12938 Here is an example of @value{GDBN}'s character set support in action.
12939 Assume that the following source code has been placed in the file
12940 @file{charset-test.c}:
12946 = @{72, 101, 108, 108, 111, 44, 32, 119,
12947 111, 114, 108, 100, 33, 10, 0@};
12948 char ibm1047_hello[]
12949 = @{200, 133, 147, 147, 150, 107, 64, 166,
12950 150, 153, 147, 132, 90, 37, 0@};
12954 printf ("Hello, world!\n");
12958 In this program, @code{ascii_hello} and @code{ibm1047_hello} are arrays
12959 containing the string @samp{Hello, world!} followed by a newline,
12960 encoded in the @sc{ascii} and @sc{ibm1047} character sets.
12962 We compile the program, and invoke the debugger on it:
12965 $ gcc -g charset-test.c -o charset-test
12966 $ gdb -nw charset-test
12967 GNU gdb 2001-12-19-cvs
12968 Copyright 2001 Free Software Foundation, Inc.
12973 We can use the @code{show charset} command to see what character sets
12974 @value{GDBN} is currently using to interpret and display characters and
12978 (@value{GDBP}) show charset
12979 The current host and target character set is `ISO-8859-1'.
12983 For the sake of printing this manual, let's use @sc{ascii} as our
12984 initial character set:
12986 (@value{GDBP}) set charset ASCII
12987 (@value{GDBP}) show charset
12988 The current host and target character set is `ASCII'.
12992 Let's assume that @sc{ascii} is indeed the correct character set for our
12993 host system --- in other words, let's assume that if @value{GDBN} prints
12994 characters using the @sc{ascii} character set, our terminal will display
12995 them properly. Since our current target character set is also
12996 @sc{ascii}, the contents of @code{ascii_hello} print legibly:
12999 (@value{GDBP}) print ascii_hello
13000 $1 = 0x401698 "Hello, world!\n"
13001 (@value{GDBP}) print ascii_hello[0]
13006 @value{GDBN} uses the target character set for character and string
13007 literals you use in expressions:
13010 (@value{GDBP}) print '+'
13015 The @sc{ascii} character set uses the number 43 to encode the @samp{+}
13018 @value{GDBN} relies on the user to tell it which character set the
13019 target program uses. If we print @code{ibm1047_hello} while our target
13020 character set is still @sc{ascii}, we get jibberish:
13023 (@value{GDBP}) print ibm1047_hello
13024 $4 = 0x4016a8 "\310\205\223\223\226k@@\246\226\231\223\204Z%"
13025 (@value{GDBP}) print ibm1047_hello[0]
13030 If we invoke the @code{set target-charset} followed by @key{TAB}@key{TAB},
13031 @value{GDBN} tells us the character sets it supports:
13034 (@value{GDBP}) set target-charset
13035 ASCII EBCDIC-US IBM1047 ISO-8859-1
13036 (@value{GDBP}) set target-charset
13039 We can select @sc{ibm1047} as our target character set, and examine the
13040 program's strings again. Now the @sc{ascii} string is wrong, but
13041 @value{GDBN} translates the contents of @code{ibm1047_hello} from the
13042 target character set, @sc{ibm1047}, to the host character set,
13043 @sc{ascii}, and they display correctly:
13046 (@value{GDBP}) set target-charset IBM1047
13047 (@value{GDBP}) show charset
13048 The current host character set is `ASCII'.
13049 The current target character set is `IBM1047'.
13050 (@value{GDBP}) print ascii_hello
13051 $6 = 0x401698 "\110\145%%?\054\040\167?\162%\144\041\012"
13052 (@value{GDBP}) print ascii_hello[0]
13054 (@value{GDBP}) print ibm1047_hello
13055 $8 = 0x4016a8 "Hello, world!\n"
13056 (@value{GDBP}) print ibm1047_hello[0]
13061 As above, @value{GDBN} uses the target character set for character and
13062 string literals you use in expressions:
13065 (@value{GDBP}) print '+'
13070 The @sc{ibm1047} character set uses the number 78 to encode the @samp{+}
13073 @node Caching Target Data
13074 @section Caching Data of Targets
13075 @cindex caching data of targets
13077 @value{GDBN} caches data exchanged between the debugger and a target.
13078 Each cache is associated with the address space of the inferior.
13079 @xref{Inferiors and Programs}, about inferior and address space.
13080 Such caching generally improves performance in remote debugging
13081 (@pxref{Remote Debugging}), because it reduces the overhead of the
13082 remote protocol by bundling memory reads and writes into large chunks.
13083 Unfortunately, simply caching everything would lead to incorrect results,
13084 since @value{GDBN} does not necessarily know anything about volatile
13085 values, memory-mapped I/O addresses, etc. Furthermore, in non-stop mode
13086 (@pxref{Non-Stop Mode}) memory can be changed @emph{while} a gdb command
13088 Therefore, by default, @value{GDBN} only caches data
13089 known to be on the stack@footnote{In non-stop mode, it is moderately
13090 rare for a running thread to modify the stack of a stopped thread
13091 in a way that would interfere with a backtrace, and caching of
13092 stack reads provides a significant speed up of remote backtraces.} or
13093 in the code segment.
13094 Other regions of memory can be explicitly marked as
13095 cacheable; @pxref{Memory Region Attributes}.
13098 @kindex set remotecache
13099 @item set remotecache on
13100 @itemx set remotecache off
13101 This option no longer does anything; it exists for compatibility
13104 @kindex show remotecache
13105 @item show remotecache
13106 Show the current state of the obsolete remotecache flag.
13108 @kindex set stack-cache
13109 @item set stack-cache on
13110 @itemx set stack-cache off
13111 Enable or disable caching of stack accesses. When @code{on}, use
13112 caching. By default, this option is @code{on}.
13114 @kindex show stack-cache
13115 @item show stack-cache
13116 Show the current state of data caching for memory accesses.
13118 @kindex set code-cache
13119 @item set code-cache on
13120 @itemx set code-cache off
13121 Enable or disable caching of code segment accesses. When @code{on},
13122 use caching. By default, this option is @code{on}. This improves
13123 performance of disassembly in remote debugging.
13125 @kindex show code-cache
13126 @item show code-cache
13127 Show the current state of target memory cache for code segment
13130 @kindex info dcache
13131 @item info dcache @r{[}line@r{]}
13132 Print the information about the performance of data cache of the
13133 current inferior's address space. The information displayed
13134 includes the dcache width and depth, and for each cache line, its
13135 number, address, and how many times it was referenced. This
13136 command is useful for debugging the data cache operation.
13138 If a line number is specified, the contents of that line will be
13141 @item set dcache size @var{size}
13142 @cindex dcache size
13143 @kindex set dcache size
13144 Set maximum number of entries in dcache (dcache depth above).
13146 @item set dcache line-size @var{line-size}
13147 @cindex dcache line-size
13148 @kindex set dcache line-size
13149 Set number of bytes each dcache entry caches (dcache width above).
13150 Must be a power of 2.
13152 @item show dcache size
13153 @kindex show dcache size
13154 Show maximum number of dcache entries. @xref{Caching Target Data, info dcache}.
13156 @item show dcache line-size
13157 @kindex show dcache line-size
13158 Show default size of dcache lines.
13162 @node Searching Memory
13163 @section Search Memory
13164 @cindex searching memory
13166 Memory can be searched for a particular sequence of bytes with the
13167 @code{find} command.
13171 @item find @r{[}/@var{sn}@r{]} @var{start_addr}, +@var{len}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
13172 @itemx find @r{[}/@var{sn}@r{]} @var{start_addr}, @var{end_addr}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
13173 Search memory for the sequence of bytes specified by @var{val1}, @var{val2},
13174 etc. The search begins at address @var{start_addr} and continues for either
13175 @var{len} bytes or through to @var{end_addr} inclusive.
13178 @var{s} and @var{n} are optional parameters.
13179 They may be specified in either order, apart or together.
13182 @item @var{s}, search query size
13183 The size of each search query value.
13189 halfwords (two bytes)
13193 giant words (eight bytes)
13196 All values are interpreted in the current language.
13197 This means, for example, that if the current source language is C/C@t{++}
13198 then searching for the string ``hello'' includes the trailing '\0'.
13199 The null terminator can be removed from searching by using casts,
13200 e.g.: @samp{@{char[5]@}"hello"}.
13202 If the value size is not specified, it is taken from the
13203 value's type in the current language.
13204 This is useful when one wants to specify the search
13205 pattern as a mixture of types.
13206 Note that this means, for example, that in the case of C-like languages
13207 a search for an untyped 0x42 will search for @samp{(int) 0x42}
13208 which is typically four bytes.
13210 @item @var{n}, maximum number of finds
13211 The maximum number of matches to print. The default is to print all finds.
13214 You can use strings as search values. Quote them with double-quotes
13216 The string value is copied into the search pattern byte by byte,
13217 regardless of the endianness of the target and the size specification.
13219 The address of each match found is printed as well as a count of the
13220 number of matches found.
13222 The address of the last value found is stored in convenience variable
13224 A count of the number of matches is stored in @samp{$numfound}.
13226 For example, if stopped at the @code{printf} in this function:
13232 static char hello[] = "hello-hello";
13233 static struct @{ char c; short s; int i; @}
13234 __attribute__ ((packed)) mixed
13235 = @{ 'c', 0x1234, 0x87654321 @};
13236 printf ("%s\n", hello);
13241 you get during debugging:
13244 (gdb) find &hello[0], +sizeof(hello), "hello"
13245 0x804956d <hello.1620+6>
13247 (gdb) find &hello[0], +sizeof(hello), 'h', 'e', 'l', 'l', 'o'
13248 0x8049567 <hello.1620>
13249 0x804956d <hello.1620+6>
13251 (gdb) find &hello[0], +sizeof(hello), @{char[5]@}"hello"
13252 0x8049567 <hello.1620>
13253 0x804956d <hello.1620+6>
13255 (gdb) find /b1 &hello[0], +sizeof(hello), 'h', 0x65, 'l'
13256 0x8049567 <hello.1620>
13258 (gdb) find &mixed, +sizeof(mixed), (char) 'c', (short) 0x1234, (int) 0x87654321
13259 0x8049560 <mixed.1625>
13261 (gdb) print $numfound
13264 $2 = (void *) 0x8049560
13268 @section Value Sizes
13270 Whenever @value{GDBN} prints a value memory will be allocated within
13271 @value{GDBN} to hold the contents of the value. It is possible in
13272 some languages with dynamic typing systems, that an invalid program
13273 may indicate a value that is incorrectly large, this in turn may cause
13274 @value{GDBN} to try and allocate an overly large amount of memory.
13277 @kindex set max-value-size
13278 @item set max-value-size @var{bytes}
13279 @itemx set max-value-size unlimited
13280 Set the maximum size of memory that @value{GDBN} will allocate for the
13281 contents of a value to @var{bytes}, trying to display a value that
13282 requires more memory than that will result in an error.
13284 Setting this variable does not effect values that have already been
13285 allocated within @value{GDBN}, only future allocations.
13287 There's a minimum size that @code{max-value-size} can be set to in
13288 order that @value{GDBN} can still operate correctly, this minimum is
13289 currently 16 bytes.
13291 The limit applies to the results of some subexpressions as well as to
13292 complete expressions. For example, an expression denoting a simple
13293 integer component, such as @code{x.y.z}, may fail if the size of
13294 @var{x.y} is dynamic and exceeds @var{bytes}. On the other hand,
13295 @value{GDBN} is sometimes clever; the expression @code{A[i]}, where
13296 @var{A} is an array variable with non-constant size, will generally
13297 succeed regardless of the bounds on @var{A}, as long as the component
13298 size is less than @var{bytes}.
13300 The default value of @code{max-value-size} is currently 64k.
13302 @kindex show max-value-size
13303 @item show max-value-size
13304 Show the maximum size of memory, in bytes, that @value{GDBN} will
13305 allocate for the contents of a value.
13308 @node Optimized Code
13309 @chapter Debugging Optimized Code
13310 @cindex optimized code, debugging
13311 @cindex debugging optimized code
13313 Almost all compilers support optimization. With optimization
13314 disabled, the compiler generates assembly code that corresponds
13315 directly to your source code, in a simplistic way. As the compiler
13316 applies more powerful optimizations, the generated assembly code
13317 diverges from your original source code. With help from debugging
13318 information generated by the compiler, @value{GDBN} can map from
13319 the running program back to constructs from your original source.
13321 @value{GDBN} is more accurate with optimization disabled. If you
13322 can recompile without optimization, it is easier to follow the
13323 progress of your program during debugging. But, there are many cases
13324 where you may need to debug an optimized version.
13326 When you debug a program compiled with @samp{-g -O}, remember that the
13327 optimizer has rearranged your code; the debugger shows you what is
13328 really there. Do not be too surprised when the execution path does not
13329 exactly match your source file! An extreme example: if you define a
13330 variable, but never use it, @value{GDBN} never sees that
13331 variable---because the compiler optimizes it out of existence.
13333 Some things do not work as well with @samp{-g -O} as with just
13334 @samp{-g}, particularly on machines with instruction scheduling. If in
13335 doubt, recompile with @samp{-g} alone, and if this fixes the problem,
13336 please report it to us as a bug (including a test case!).
13337 @xref{Variables}, for more information about debugging optimized code.
13340 * Inline Functions:: How @value{GDBN} presents inlining
13341 * Tail Call Frames:: @value{GDBN} analysis of jumps to functions
13344 @node Inline Functions
13345 @section Inline Functions
13346 @cindex inline functions, debugging
13348 @dfn{Inlining} is an optimization that inserts a copy of the function
13349 body directly at each call site, instead of jumping to a shared
13350 routine. @value{GDBN} displays inlined functions just like
13351 non-inlined functions. They appear in backtraces. You can view their
13352 arguments and local variables, step into them with @code{step}, skip
13353 them with @code{next}, and escape from them with @code{finish}.
13354 You can check whether a function was inlined by using the
13355 @code{info frame} command.
13357 For @value{GDBN} to support inlined functions, the compiler must
13358 record information about inlining in the debug information ---
13359 @value{NGCC} using the @sc{dwarf 2} format does this, and several
13360 other compilers do also. @value{GDBN} only supports inlined functions
13361 when using @sc{dwarf 2}. Versions of @value{NGCC} before 4.1
13362 do not emit two required attributes (@samp{DW_AT_call_file} and
13363 @samp{DW_AT_call_line}); @value{GDBN} does not display inlined
13364 function calls with earlier versions of @value{NGCC}. It instead
13365 displays the arguments and local variables of inlined functions as
13366 local variables in the caller.
13368 The body of an inlined function is directly included at its call site;
13369 unlike a non-inlined function, there are no instructions devoted to
13370 the call. @value{GDBN} still pretends that the call site and the
13371 start of the inlined function are different instructions. Stepping to
13372 the call site shows the call site, and then stepping again shows
13373 the first line of the inlined function, even though no additional
13374 instructions are executed.
13376 This makes source-level debugging much clearer; you can see both the
13377 context of the call and then the effect of the call. Only stepping by
13378 a single instruction using @code{stepi} or @code{nexti} does not do
13379 this; single instruction steps always show the inlined body.
13381 There are some ways that @value{GDBN} does not pretend that inlined
13382 function calls are the same as normal calls:
13386 Setting breakpoints at the call site of an inlined function may not
13387 work, because the call site does not contain any code. @value{GDBN}
13388 may incorrectly move the breakpoint to the next line of the enclosing
13389 function, after the call. This limitation will be removed in a future
13390 version of @value{GDBN}; until then, set a breakpoint on an earlier line
13391 or inside the inlined function instead.
13394 @value{GDBN} cannot locate the return value of inlined calls after
13395 using the @code{finish} command. This is a limitation of compiler-generated
13396 debugging information; after @code{finish}, you can step to the next line
13397 and print a variable where your program stored the return value.
13401 @node Tail Call Frames
13402 @section Tail Call Frames
13403 @cindex tail call frames, debugging
13405 Function @code{B} can call function @code{C} in its very last statement. In
13406 unoptimized compilation the call of @code{C} is immediately followed by return
13407 instruction at the end of @code{B} code. Optimizing compiler may replace the
13408 call and return in function @code{B} into one jump to function @code{C}
13409 instead. Such use of a jump instruction is called @dfn{tail call}.
13411 During execution of function @code{C}, there will be no indication in the
13412 function call stack frames that it was tail-called from @code{B}. If function
13413 @code{A} regularly calls function @code{B} which tail-calls function @code{C},
13414 then @value{GDBN} will see @code{A} as the caller of @code{C}. However, in
13415 some cases @value{GDBN} can determine that @code{C} was tail-called from
13416 @code{B}, and it will then create fictitious call frame for that, with the
13417 return address set up as if @code{B} called @code{C} normally.
13419 This functionality is currently supported only by DWARF 2 debugging format and
13420 the compiler has to produce @samp{DW_TAG_call_site} tags. With
13421 @value{NGCC}, you need to specify @option{-O -g} during compilation, to get
13424 @kbd{info frame} command (@pxref{Frame Info}) will indicate the tail call frame
13425 kind by text @code{tail call frame} such as in this sample @value{GDBN} output:
13429 0x40066b <b(int, double)+11>: jmp 0x400640 <c(int, double)>
13431 Stack level 1, frame at 0x7fffffffda30:
13432 rip = 0x40066d in b (amd64-entry-value.cc:59); saved rip 0x4004c5
13433 tail call frame, caller of frame at 0x7fffffffda30
13434 source language c++.
13435 Arglist at unknown address.
13436 Locals at unknown address, Previous frame's sp is 0x7fffffffda30
13439 The detection of all the possible code path executions can find them ambiguous.
13440 There is no execution history stored (possible @ref{Reverse Execution} is never
13441 used for this purpose) and the last known caller could have reached the known
13442 callee by multiple different jump sequences. In such case @value{GDBN} still
13443 tries to show at least all the unambiguous top tail callers and all the
13444 unambiguous bottom tail calees, if any.
13447 @anchor{set debug entry-values}
13448 @item set debug entry-values
13449 @kindex set debug entry-values
13450 When set to on, enables printing of analysis messages for both frame argument
13451 values at function entry and tail calls. It will show all the possible valid
13452 tail calls code paths it has considered. It will also print the intersection
13453 of them with the final unambiguous (possibly partial or even empty) code path
13456 @item show debug entry-values
13457 @kindex show debug entry-values
13458 Show the current state of analysis messages printing for both frame argument
13459 values at function entry and tail calls.
13462 The analysis messages for tail calls can for example show why the virtual tail
13463 call frame for function @code{c} has not been recognized (due to the indirect
13464 reference by variable @code{x}):
13467 static void __attribute__((noinline, noclone)) c (void);
13468 void (*x) (void) = c;
13469 static void __attribute__((noinline, noclone)) a (void) @{ x++; @}
13470 static void __attribute__((noinline, noclone)) c (void) @{ a (); @}
13471 int main (void) @{ x (); return 0; @}
13473 Breakpoint 1, DW_OP_entry_value resolving cannot find
13474 DW_TAG_call_site 0x40039a in main
13476 3 static void __attribute__((noinline, noclone)) a (void) @{ x++; @}
13479 #1 0x000000000040039a in main () at t.c:5
13482 Another possibility is an ambiguous virtual tail call frames resolution:
13486 static void __attribute__((noinline, noclone)) f (void) @{ i++; @}
13487 static void __attribute__((noinline, noclone)) e (void) @{ f (); @}
13488 static void __attribute__((noinline, noclone)) d (void) @{ f (); @}
13489 static void __attribute__((noinline, noclone)) c (void) @{ d (); @}
13490 static void __attribute__((noinline, noclone)) b (void)
13491 @{ if (i) c (); else e (); @}
13492 static void __attribute__((noinline, noclone)) a (void) @{ b (); @}
13493 int main (void) @{ a (); return 0; @}
13495 tailcall: initial: 0x4004d2(a) 0x4004ce(b) 0x4004b2(c) 0x4004a2(d)
13496 tailcall: compare: 0x4004d2(a) 0x4004cc(b) 0x400492(e)
13497 tailcall: reduced: 0x4004d2(a) |
13500 #1 0x00000000004004d2 in a () at t.c:8
13501 #2 0x0000000000400395 in main () at t.c:9
13504 @set CALLSEQ1A @code{main@value{ARROW}a@value{ARROW}b@value{ARROW}c@value{ARROW}d@value{ARROW}f}
13505 @set CALLSEQ2A @code{main@value{ARROW}a@value{ARROW}b@value{ARROW}e@value{ARROW}f}
13507 @c Convert CALLSEQ#A to CALLSEQ#B depending on HAVE_MAKEINFO_CLICK.
13508 @ifset HAVE_MAKEINFO_CLICK
13509 @set ARROW @click{}
13510 @set CALLSEQ1B @clicksequence{@value{CALLSEQ1A}}
13511 @set CALLSEQ2B @clicksequence{@value{CALLSEQ2A}}
13513 @ifclear HAVE_MAKEINFO_CLICK
13515 @set CALLSEQ1B @value{CALLSEQ1A}
13516 @set CALLSEQ2B @value{CALLSEQ2A}
13519 Frames #0 and #2 are real, #1 is a virtual tail call frame.
13520 The code can have possible execution paths @value{CALLSEQ1B} or
13521 @value{CALLSEQ2B}, @value{GDBN} cannot find which one from the inferior state.
13523 @code{initial:} state shows some random possible calling sequence @value{GDBN}
13524 has found. It then finds another possible calling sequence - that one is
13525 prefixed by @code{compare:}. The non-ambiguous intersection of these two is
13526 printed as the @code{reduced:} calling sequence. That one could have many
13527 further @code{compare:} and @code{reduced:} statements as long as there remain
13528 any non-ambiguous sequence entries.
13530 For the frame of function @code{b} in both cases there are different possible
13531 @code{$pc} values (@code{0x4004cc} or @code{0x4004ce}), therefore this frame is
13532 also ambiguous. The only non-ambiguous frame is the one for function @code{a},
13533 therefore this one is displayed to the user while the ambiguous frames are
13536 There can be also reasons why printing of frame argument values at function
13541 static void __attribute__((noinline, noclone)) c (int i) @{ v++; @}
13542 static void __attribute__((noinline, noclone)) a (int i);
13543 static void __attribute__((noinline, noclone)) b (int i) @{ a (i); @}
13544 static void __attribute__((noinline, noclone)) a (int i)
13545 @{ if (i) b (i - 1); else c (0); @}
13546 int main (void) @{ a (5); return 0; @}
13549 #0 c (i=i@@entry=0) at t.c:2
13550 #1 0x0000000000400428 in a (DW_OP_entry_value resolving has found
13551 function "a" at 0x400420 can call itself via tail calls
13552 i=<optimized out>) at t.c:6
13553 #2 0x000000000040036e in main () at t.c:7
13556 @value{GDBN} cannot find out from the inferior state if and how many times did
13557 function @code{a} call itself (via function @code{b}) as these calls would be
13558 tail calls. Such tail calls would modify the @code{i} variable, therefore
13559 @value{GDBN} cannot be sure the value it knows would be right - @value{GDBN}
13560 prints @code{<optimized out>} instead.
13563 @chapter C Preprocessor Macros
13565 Some languages, such as C and C@t{++}, provide a way to define and invoke
13566 ``preprocessor macros'' which expand into strings of tokens.
13567 @value{GDBN} can evaluate expressions containing macro invocations, show
13568 the result of macro expansion, and show a macro's definition, including
13569 where it was defined.
13571 You may need to compile your program specially to provide @value{GDBN}
13572 with information about preprocessor macros. Most compilers do not
13573 include macros in their debugging information, even when you compile
13574 with the @option{-g} flag. @xref{Compilation}.
13576 A program may define a macro at one point, remove that definition later,
13577 and then provide a different definition after that. Thus, at different
13578 points in the program, a macro may have different definitions, or have
13579 no definition at all. If there is a current stack frame, @value{GDBN}
13580 uses the macros in scope at that frame's source code line. Otherwise,
13581 @value{GDBN} uses the macros in scope at the current listing location;
13584 Whenever @value{GDBN} evaluates an expression, it always expands any
13585 macro invocations present in the expression. @value{GDBN} also provides
13586 the following commands for working with macros explicitly.
13590 @kindex macro expand
13591 @cindex macro expansion, showing the results of preprocessor
13592 @cindex preprocessor macro expansion, showing the results of
13593 @cindex expanding preprocessor macros
13594 @item macro expand @var{expression}
13595 @itemx macro exp @var{expression}
13596 Show the results of expanding all preprocessor macro invocations in
13597 @var{expression}. Since @value{GDBN} simply expands macros, but does
13598 not parse the result, @var{expression} need not be a valid expression;
13599 it can be any string of tokens.
13602 @item macro expand-once @var{expression}
13603 @itemx macro exp1 @var{expression}
13604 @cindex expand macro once
13605 @i{(This command is not yet implemented.)} Show the results of
13606 expanding those preprocessor macro invocations that appear explicitly in
13607 @var{expression}. Macro invocations appearing in that expansion are
13608 left unchanged. This command allows you to see the effect of a
13609 particular macro more clearly, without being confused by further
13610 expansions. Since @value{GDBN} simply expands macros, but does not
13611 parse the result, @var{expression} need not be a valid expression; it
13612 can be any string of tokens.
13615 @cindex macro definition, showing
13616 @cindex definition of a macro, showing
13617 @cindex macros, from debug info
13618 @item info macro [-a|-all] [--] @var{macro}
13619 Show the current definition or all definitions of the named @var{macro},
13620 and describe the source location or compiler command-line where that
13621 definition was established. The optional double dash is to signify the end of
13622 argument processing and the beginning of @var{macro} for non C-like macros where
13623 the macro may begin with a hyphen.
13625 @kindex info macros
13626 @item info macros @var{location}
13627 Show all macro definitions that are in effect at the location specified
13628 by @var{location}, and describe the source location or compiler
13629 command-line where those definitions were established.
13631 @kindex macro define
13632 @cindex user-defined macros
13633 @cindex defining macros interactively
13634 @cindex macros, user-defined
13635 @item macro define @var{macro} @var{replacement-list}
13636 @itemx macro define @var{macro}(@var{arglist}) @var{replacement-list}
13637 Introduce a definition for a preprocessor macro named @var{macro},
13638 invocations of which are replaced by the tokens given in
13639 @var{replacement-list}. The first form of this command defines an
13640 ``object-like'' macro, which takes no arguments; the second form
13641 defines a ``function-like'' macro, which takes the arguments given in
13644 A definition introduced by this command is in scope in every
13645 expression evaluated in @value{GDBN}, until it is removed with the
13646 @code{macro undef} command, described below. The definition overrides
13647 all definitions for @var{macro} present in the program being debugged,
13648 as well as any previous user-supplied definition.
13650 @kindex macro undef
13651 @item macro undef @var{macro}
13652 Remove any user-supplied definition for the macro named @var{macro}.
13653 This command only affects definitions provided with the @code{macro
13654 define} command, described above; it cannot remove definitions present
13655 in the program being debugged.
13659 List all the macros defined using the @code{macro define} command.
13662 @cindex macros, example of debugging with
13663 Here is a transcript showing the above commands in action. First, we
13664 show our source files:
13669 #include "sample.h"
13672 #define ADD(x) (M + x)
13677 printf ("Hello, world!\n");
13679 printf ("We're so creative.\n");
13681 printf ("Goodbye, world!\n");
13688 Now, we compile the program using the @sc{gnu} C compiler,
13689 @value{NGCC}. We pass the @option{-gdwarf-2}@footnote{This is the
13690 minimum. Recent versions of @value{NGCC} support @option{-gdwarf-3}
13691 and @option{-gdwarf-4}; we recommend always choosing the most recent
13692 version of DWARF.} @emph{and} @option{-g3} flags to ensure the compiler
13693 includes information about preprocessor macros in the debugging
13697 $ gcc -gdwarf-2 -g3 sample.c -o sample
13701 Now, we start @value{GDBN} on our sample program:
13705 GNU gdb 2002-05-06-cvs
13706 Copyright 2002 Free Software Foundation, Inc.
13707 GDB is free software, @dots{}
13711 We can expand macros and examine their definitions, even when the
13712 program is not running. @value{GDBN} uses the current listing position
13713 to decide which macro definitions are in scope:
13716 (@value{GDBP}) list main
13719 5 #define ADD(x) (M + x)
13724 10 printf ("Hello, world!\n");
13726 12 printf ("We're so creative.\n");
13727 (@value{GDBP}) info macro ADD
13728 Defined at /home/jimb/gdb/macros/play/sample.c:5
13729 #define ADD(x) (M + x)
13730 (@value{GDBP}) info macro Q
13731 Defined at /home/jimb/gdb/macros/play/sample.h:1
13732 included at /home/jimb/gdb/macros/play/sample.c:2
13734 (@value{GDBP}) macro expand ADD(1)
13735 expands to: (42 + 1)
13736 (@value{GDBP}) macro expand-once ADD(1)
13737 expands to: once (M + 1)
13741 In the example above, note that @code{macro expand-once} expands only
13742 the macro invocation explicit in the original text --- the invocation of
13743 @code{ADD} --- but does not expand the invocation of the macro @code{M},
13744 which was introduced by @code{ADD}.
13746 Once the program is running, @value{GDBN} uses the macro definitions in
13747 force at the source line of the current stack frame:
13750 (@value{GDBP}) break main
13751 Breakpoint 1 at 0x8048370: file sample.c, line 10.
13753 Starting program: /home/jimb/gdb/macros/play/sample
13755 Breakpoint 1, main () at sample.c:10
13756 10 printf ("Hello, world!\n");
13760 At line 10, the definition of the macro @code{N} at line 9 is in force:
13763 (@value{GDBP}) info macro N
13764 Defined at /home/jimb/gdb/macros/play/sample.c:9
13766 (@value{GDBP}) macro expand N Q M
13767 expands to: 28 < 42
13768 (@value{GDBP}) print N Q M
13773 As we step over directives that remove @code{N}'s definition, and then
13774 give it a new definition, @value{GDBN} finds the definition (or lack
13775 thereof) in force at each point:
13778 (@value{GDBP}) next
13780 12 printf ("We're so creative.\n");
13781 (@value{GDBP}) info macro N
13782 The symbol `N' has no definition as a C/C++ preprocessor macro
13783 at /home/jimb/gdb/macros/play/sample.c:12
13784 (@value{GDBP}) next
13786 14 printf ("Goodbye, world!\n");
13787 (@value{GDBP}) info macro N
13788 Defined at /home/jimb/gdb/macros/play/sample.c:13
13790 (@value{GDBP}) macro expand N Q M
13791 expands to: 1729 < 42
13792 (@value{GDBP}) print N Q M
13797 In addition to source files, macros can be defined on the compilation command
13798 line using the @option{-D@var{name}=@var{value}} syntax. For macros defined in
13799 such a way, @value{GDBN} displays the location of their definition as line zero
13800 of the source file submitted to the compiler.
13803 (@value{GDBP}) info macro __STDC__
13804 Defined at /home/jimb/gdb/macros/play/sample.c:0
13811 @chapter Tracepoints
13812 @c This chapter is based on the documentation written by Michael
13813 @c Snyder, David Taylor, Jim Blandy, and Elena Zannoni.
13815 @cindex tracepoints
13816 In some applications, it is not feasible for the debugger to interrupt
13817 the program's execution long enough for the developer to learn
13818 anything helpful about its behavior. If the program's correctness
13819 depends on its real-time behavior, delays introduced by a debugger
13820 might cause the program to change its behavior drastically, or perhaps
13821 fail, even when the code itself is correct. It is useful to be able
13822 to observe the program's behavior without interrupting it.
13824 Using @value{GDBN}'s @code{trace} and @code{collect} commands, you can
13825 specify locations in the program, called @dfn{tracepoints}, and
13826 arbitrary expressions to evaluate when those tracepoints are reached.
13827 Later, using the @code{tfind} command, you can examine the values
13828 those expressions had when the program hit the tracepoints. The
13829 expressions may also denote objects in memory---structures or arrays,
13830 for example---whose values @value{GDBN} should record; while visiting
13831 a particular tracepoint, you may inspect those objects as if they were
13832 in memory at that moment. However, because @value{GDBN} records these
13833 values without interacting with you, it can do so quickly and
13834 unobtrusively, hopefully not disturbing the program's behavior.
13836 The tracepoint facility is currently available only for remote
13837 targets. @xref{Targets}. In addition, your remote target must know
13838 how to collect trace data. This functionality is implemented in the
13839 remote stub; however, none of the stubs distributed with @value{GDBN}
13840 support tracepoints as of this writing. The format of the remote
13841 packets used to implement tracepoints are described in @ref{Tracepoint
13844 It is also possible to get trace data from a file, in a manner reminiscent
13845 of corefiles; you specify the filename, and use @code{tfind} to search
13846 through the file. @xref{Trace Files}, for more details.
13848 This chapter describes the tracepoint commands and features.
13851 * Set Tracepoints::
13852 * Analyze Collected Data::
13853 * Tracepoint Variables::
13857 @node Set Tracepoints
13858 @section Commands to Set Tracepoints
13860 Before running such a @dfn{trace experiment}, an arbitrary number of
13861 tracepoints can be set. A tracepoint is actually a special type of
13862 breakpoint (@pxref{Set Breaks}), so you can manipulate it using
13863 standard breakpoint commands. For instance, as with breakpoints,
13864 tracepoint numbers are successive integers starting from one, and many
13865 of the commands associated with tracepoints take the tracepoint number
13866 as their argument, to identify which tracepoint to work on.
13868 For each tracepoint, you can specify, in advance, some arbitrary set
13869 of data that you want the target to collect in the trace buffer when
13870 it hits that tracepoint. The collected data can include registers,
13871 local variables, or global data. Later, you can use @value{GDBN}
13872 commands to examine the values these data had at the time the
13873 tracepoint was hit.
13875 Tracepoints do not support every breakpoint feature. Ignore counts on
13876 tracepoints have no effect, and tracepoints cannot run @value{GDBN}
13877 commands when they are hit. Tracepoints may not be thread-specific
13880 @cindex fast tracepoints
13881 Some targets may support @dfn{fast tracepoints}, which are inserted in
13882 a different way (such as with a jump instead of a trap), that is
13883 faster but possibly restricted in where they may be installed.
13885 @cindex static tracepoints
13886 @cindex markers, static tracepoints
13887 @cindex probing markers, static tracepoints
13888 Regular and fast tracepoints are dynamic tracing facilities, meaning
13889 that they can be used to insert tracepoints at (almost) any location
13890 in the target. Some targets may also support controlling @dfn{static
13891 tracepoints} from @value{GDBN}. With static tracing, a set of
13892 instrumentation points, also known as @dfn{markers}, are embedded in
13893 the target program, and can be activated or deactivated by name or
13894 address. These are usually placed at locations which facilitate
13895 investigating what the target is actually doing. @value{GDBN}'s
13896 support for static tracing includes being able to list instrumentation
13897 points, and attach them with @value{GDBN} defined high level
13898 tracepoints that expose the whole range of convenience of
13899 @value{GDBN}'s tracepoints support. Namely, support for collecting
13900 registers values and values of global or local (to the instrumentation
13901 point) variables; tracepoint conditions and trace state variables.
13902 The act of installing a @value{GDBN} static tracepoint on an
13903 instrumentation point, or marker, is referred to as @dfn{probing} a
13904 static tracepoint marker.
13906 @code{gdbserver} supports tracepoints on some target systems.
13907 @xref{Server,,Tracepoints support in @code{gdbserver}}.
13909 This section describes commands to set tracepoints and associated
13910 conditions and actions.
13913 * Create and Delete Tracepoints::
13914 * Enable and Disable Tracepoints::
13915 * Tracepoint Passcounts::
13916 * Tracepoint Conditions::
13917 * Trace State Variables::
13918 * Tracepoint Actions::
13919 * Listing Tracepoints::
13920 * Listing Static Tracepoint Markers::
13921 * Starting and Stopping Trace Experiments::
13922 * Tracepoint Restrictions::
13925 @node Create and Delete Tracepoints
13926 @subsection Create and Delete Tracepoints
13929 @cindex set tracepoint
13931 @item trace @var{location}
13932 The @code{trace} command is very similar to the @code{break} command.
13933 Its argument @var{location} can be any valid location.
13934 @xref{Specify Location}. The @code{trace} command defines a tracepoint,
13935 which is a point in the target program where the debugger will briefly stop,
13936 collect some data, and then allow the program to continue. Setting a tracepoint
13937 or changing its actions takes effect immediately if the remote stub
13938 supports the @samp{InstallInTrace} feature (@pxref{install tracepoint
13940 If remote stub doesn't support the @samp{InstallInTrace} feature, all
13941 these changes don't take effect until the next @code{tstart}
13942 command, and once a trace experiment is running, further changes will
13943 not have any effect until the next trace experiment starts. In addition,
13944 @value{GDBN} supports @dfn{pending tracepoints}---tracepoints whose
13945 address is not yet resolved. (This is similar to pending breakpoints.)
13946 Pending tracepoints are not downloaded to the target and not installed
13947 until they are resolved. The resolution of pending tracepoints requires
13948 @value{GDBN} support---when debugging with the remote target, and
13949 @value{GDBN} disconnects from the remote stub (@pxref{disconnected
13950 tracing}), pending tracepoints can not be resolved (and downloaded to
13951 the remote stub) while @value{GDBN} is disconnected.
13953 Here are some examples of using the @code{trace} command:
13956 (@value{GDBP}) @b{trace foo.c:121} // a source file and line number
13958 (@value{GDBP}) @b{trace +2} // 2 lines forward
13960 (@value{GDBP}) @b{trace my_function} // first source line of function
13962 (@value{GDBP}) @b{trace *my_function} // EXACT start address of function
13964 (@value{GDBP}) @b{trace *0x2117c4} // an address
13968 You can abbreviate @code{trace} as @code{tr}.
13970 @item trace @var{location} if @var{cond}
13971 Set a tracepoint with condition @var{cond}; evaluate the expression
13972 @var{cond} each time the tracepoint is reached, and collect data only
13973 if the value is nonzero---that is, if @var{cond} evaluates as true.
13974 @xref{Tracepoint Conditions, ,Tracepoint Conditions}, for more
13975 information on tracepoint conditions.
13977 @item ftrace @var{location} [ if @var{cond} ]
13978 @cindex set fast tracepoint
13979 @cindex fast tracepoints, setting
13981 The @code{ftrace} command sets a fast tracepoint. For targets that
13982 support them, fast tracepoints will use a more efficient but possibly
13983 less general technique to trigger data collection, such as a jump
13984 instruction instead of a trap, or some sort of hardware support. It
13985 may not be possible to create a fast tracepoint at the desired
13986 location, in which case the command will exit with an explanatory
13989 @value{GDBN} handles arguments to @code{ftrace} exactly as for
13992 On 32-bit x86-architecture systems, fast tracepoints normally need to
13993 be placed at an instruction that is 5 bytes or longer, but can be
13994 placed at 4-byte instructions if the low 64K of memory of the target
13995 program is available to install trampolines. Some Unix-type systems,
13996 such as @sc{gnu}/Linux, exclude low addresses from the program's
13997 address space; but for instance with the Linux kernel it is possible
13998 to let @value{GDBN} use this area by doing a @command{sysctl} command
13999 to set the @code{mmap_min_addr} kernel parameter, as in
14002 sudo sysctl -w vm.mmap_min_addr=32768
14006 which sets the low address to 32K, which leaves plenty of room for
14007 trampolines. The minimum address should be set to a page boundary.
14009 @item strace @var{location} [ if @var{cond} ]
14010 @cindex set static tracepoint
14011 @cindex static tracepoints, setting
14012 @cindex probe static tracepoint marker
14014 The @code{strace} command sets a static tracepoint. For targets that
14015 support it, setting a static tracepoint probes a static
14016 instrumentation point, or marker, found at @var{location}. It may not
14017 be possible to set a static tracepoint at the desired location, in
14018 which case the command will exit with an explanatory message.
14020 @value{GDBN} handles arguments to @code{strace} exactly as for
14021 @code{trace}, with the addition that the user can also specify
14022 @code{-m @var{marker}} as @var{location}. This probes the marker
14023 identified by the @var{marker} string identifier. This identifier
14024 depends on the static tracepoint backend library your program is
14025 using. You can find all the marker identifiers in the @samp{ID} field
14026 of the @code{info static-tracepoint-markers} command output.
14027 @xref{Listing Static Tracepoint Markers,,Listing Static Tracepoint
14028 Markers}. For example, in the following small program using the UST
14034 trace_mark(ust, bar33, "str %s", "FOOBAZ");
14039 the marker id is composed of joining the first two arguments to the
14040 @code{trace_mark} call with a slash, which translates to:
14043 (@value{GDBP}) info static-tracepoint-markers
14044 Cnt Enb ID Address What
14045 1 n ust/bar33 0x0000000000400ddc in main at stexample.c:22
14051 so you may probe the marker above with:
14054 (@value{GDBP}) strace -m ust/bar33
14057 Static tracepoints accept an extra collect action --- @code{collect
14058 $_sdata}. This collects arbitrary user data passed in the probe point
14059 call to the tracing library. In the UST example above, you'll see
14060 that the third argument to @code{trace_mark} is a printf-like format
14061 string. The user data is then the result of running that formatting
14062 string against the following arguments. Note that @code{info
14063 static-tracepoint-markers} command output lists that format string in
14064 the @samp{Data:} field.
14066 You can inspect this data when analyzing the trace buffer, by printing
14067 the $_sdata variable like any other variable available to
14068 @value{GDBN}. @xref{Tracepoint Actions,,Tracepoint Action Lists}.
14071 @cindex last tracepoint number
14072 @cindex recent tracepoint number
14073 @cindex tracepoint number
14074 The convenience variable @code{$tpnum} records the tracepoint number
14075 of the most recently set tracepoint.
14077 @kindex delete tracepoint
14078 @cindex tracepoint deletion
14079 @item delete tracepoint @r{[}@var{num}@r{]}
14080 Permanently delete one or more tracepoints. With no argument, the
14081 default is to delete all tracepoints. Note that the regular
14082 @code{delete} command can remove tracepoints also.
14087 (@value{GDBP}) @b{delete trace 1 2 3} // remove three tracepoints
14089 (@value{GDBP}) @b{delete trace} // remove all tracepoints
14093 You can abbreviate this command as @code{del tr}.
14096 @node Enable and Disable Tracepoints
14097 @subsection Enable and Disable Tracepoints
14099 These commands are deprecated; they are equivalent to plain @code{disable} and @code{enable}.
14102 @kindex disable tracepoint
14103 @item disable tracepoint @r{[}@var{num}@r{]}
14104 Disable tracepoint @var{num}, or all tracepoints if no argument
14105 @var{num} is given. A disabled tracepoint will have no effect during
14106 a trace experiment, but it is not forgotten. You can re-enable
14107 a disabled tracepoint using the @code{enable tracepoint} command.
14108 If the command is issued during a trace experiment and the debug target
14109 has support for disabling tracepoints during a trace experiment, then the
14110 change will be effective immediately. Otherwise, it will be applied to the
14111 next trace experiment.
14113 @kindex enable tracepoint
14114 @item enable tracepoint @r{[}@var{num}@r{]}
14115 Enable tracepoint @var{num}, or all tracepoints. If this command is
14116 issued during a trace experiment and the debug target supports enabling
14117 tracepoints during a trace experiment, then the enabled tracepoints will
14118 become effective immediately. Otherwise, they will become effective the
14119 next time a trace experiment is run.
14122 @node Tracepoint Passcounts
14123 @subsection Tracepoint Passcounts
14127 @cindex tracepoint pass count
14128 @item passcount @r{[}@var{n} @r{[}@var{num}@r{]]}
14129 Set the @dfn{passcount} of a tracepoint. The passcount is a way to
14130 automatically stop a trace experiment. If a tracepoint's passcount is
14131 @var{n}, then the trace experiment will be automatically stopped on
14132 the @var{n}'th time that tracepoint is hit. If the tracepoint number
14133 @var{num} is not specified, the @code{passcount} command sets the
14134 passcount of the most recently defined tracepoint. If no passcount is
14135 given, the trace experiment will run until stopped explicitly by the
14141 (@value{GDBP}) @b{passcount 5 2} // Stop on the 5th execution of
14142 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// tracepoint 2}
14144 (@value{GDBP}) @b{passcount 12} // Stop on the 12th execution of the
14145 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// most recently defined tracepoint.}
14146 (@value{GDBP}) @b{trace foo}
14147 (@value{GDBP}) @b{pass 3}
14148 (@value{GDBP}) @b{trace bar}
14149 (@value{GDBP}) @b{pass 2}
14150 (@value{GDBP}) @b{trace baz}
14151 (@value{GDBP}) @b{pass 1} // Stop tracing when foo has been
14152 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// executed 3 times OR when bar has}
14153 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// been executed 2 times}
14154 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// OR when baz has been executed 1 time.}
14158 @node Tracepoint Conditions
14159 @subsection Tracepoint Conditions
14160 @cindex conditional tracepoints
14161 @cindex tracepoint conditions
14163 The simplest sort of tracepoint collects data every time your program
14164 reaches a specified place. You can also specify a @dfn{condition} for
14165 a tracepoint. A condition is just a Boolean expression in your
14166 programming language (@pxref{Expressions, ,Expressions}). A
14167 tracepoint with a condition evaluates the expression each time your
14168 program reaches it, and data collection happens only if the condition
14171 Tracepoint conditions can be specified when a tracepoint is set, by
14172 using @samp{if} in the arguments to the @code{trace} command.
14173 @xref{Create and Delete Tracepoints, ,Setting Tracepoints}. They can
14174 also be set or changed at any time with the @code{condition} command,
14175 just as with breakpoints.
14177 Unlike breakpoint conditions, @value{GDBN} does not actually evaluate
14178 the conditional expression itself. Instead, @value{GDBN} encodes the
14179 expression into an agent expression (@pxref{Agent Expressions})
14180 suitable for execution on the target, independently of @value{GDBN}.
14181 Global variables become raw memory locations, locals become stack
14182 accesses, and so forth.
14184 For instance, suppose you have a function that is usually called
14185 frequently, but should not be called after an error has occurred. You
14186 could use the following tracepoint command to collect data about calls
14187 of that function that happen while the error code is propagating
14188 through the program; an unconditional tracepoint could end up
14189 collecting thousands of useless trace frames that you would have to
14193 (@value{GDBP}) @kbd{trace normal_operation if errcode > 0}
14196 @node Trace State Variables
14197 @subsection Trace State Variables
14198 @cindex trace state variables
14200 A @dfn{trace state variable} is a special type of variable that is
14201 created and managed by target-side code. The syntax is the same as
14202 that for GDB's convenience variables (a string prefixed with ``$''),
14203 but they are stored on the target. They must be created explicitly,
14204 using a @code{tvariable} command. They are always 64-bit signed
14207 Trace state variables are remembered by @value{GDBN}, and downloaded
14208 to the target along with tracepoint information when the trace
14209 experiment starts. There are no intrinsic limits on the number of
14210 trace state variables, beyond memory limitations of the target.
14212 @cindex convenience variables, and trace state variables
14213 Although trace state variables are managed by the target, you can use
14214 them in print commands and expressions as if they were convenience
14215 variables; @value{GDBN} will get the current value from the target
14216 while the trace experiment is running. Trace state variables share
14217 the same namespace as other ``$'' variables, which means that you
14218 cannot have trace state variables with names like @code{$23} or
14219 @code{$pc}, nor can you have a trace state variable and a convenience
14220 variable with the same name.
14224 @item tvariable $@var{name} [ = @var{expression} ]
14226 The @code{tvariable} command creates a new trace state variable named
14227 @code{$@var{name}}, and optionally gives it an initial value of
14228 @var{expression}. The @var{expression} is evaluated when this command is
14229 entered; the result will be converted to an integer if possible,
14230 otherwise @value{GDBN} will report an error. A subsequent
14231 @code{tvariable} command specifying the same name does not create a
14232 variable, but instead assigns the supplied initial value to the
14233 existing variable of that name, overwriting any previous initial
14234 value. The default initial value is 0.
14236 @item info tvariables
14237 @kindex info tvariables
14238 List all the trace state variables along with their initial values.
14239 Their current values may also be displayed, if the trace experiment is
14242 @item delete tvariable @r{[} $@var{name} @dots{} @r{]}
14243 @kindex delete tvariable
14244 Delete the given trace state variables, or all of them if no arguments
14249 @node Tracepoint Actions
14250 @subsection Tracepoint Action Lists
14254 @cindex tracepoint actions
14255 @item actions @r{[}@var{num}@r{]}
14256 This command will prompt for a list of actions to be taken when the
14257 tracepoint is hit. If the tracepoint number @var{num} is not
14258 specified, this command sets the actions for the one that was most
14259 recently defined (so that you can define a tracepoint and then say
14260 @code{actions} without bothering about its number). You specify the
14261 actions themselves on the following lines, one action at a time, and
14262 terminate the actions list with a line containing just @code{end}. So
14263 far, the only defined actions are @code{collect}, @code{teval}, and
14264 @code{while-stepping}.
14266 @code{actions} is actually equivalent to @code{commands} (@pxref{Break
14267 Commands, ,Breakpoint Command Lists}), except that only the defined
14268 actions are allowed; any other @value{GDBN} command is rejected.
14270 @cindex remove actions from a tracepoint
14271 To remove all actions from a tracepoint, type @samp{actions @var{num}}
14272 and follow it immediately with @samp{end}.
14275 (@value{GDBP}) @b{collect @var{data}} // collect some data
14277 (@value{GDBP}) @b{while-stepping 5} // single-step 5 times, collect data
14279 (@value{GDBP}) @b{end} // signals the end of actions.
14282 In the following example, the action list begins with @code{collect}
14283 commands indicating the things to be collected when the tracepoint is
14284 hit. Then, in order to single-step and collect additional data
14285 following the tracepoint, a @code{while-stepping} command is used,
14286 followed by the list of things to be collected after each step in a
14287 sequence of single steps. The @code{while-stepping} command is
14288 terminated by its own separate @code{end} command. Lastly, the action
14289 list is terminated by an @code{end} command.
14292 (@value{GDBP}) @b{trace foo}
14293 (@value{GDBP}) @b{actions}
14294 Enter actions for tracepoint 1, one per line:
14297 > while-stepping 12
14298 > collect $pc, arr[i]
14303 @kindex collect @r{(tracepoints)}
14304 @item collect@r{[}/@var{mods}@r{]} @var{expr1}, @var{expr2}, @dots{}
14305 Collect values of the given expressions when the tracepoint is hit.
14306 This command accepts a comma-separated list of any valid expressions.
14307 In addition to global, static, or local variables, the following
14308 special arguments are supported:
14312 Collect all registers.
14315 Collect all function arguments.
14318 Collect all local variables.
14321 Collect the return address. This is helpful if you want to see more
14324 @emph{Note:} The return address location can not always be reliably
14325 determined up front, and the wrong address / registers may end up
14326 collected instead. On some architectures the reliability is higher
14327 for tracepoints at function entry, while on others it's the opposite.
14328 When this happens, backtracing will stop because the return address is
14329 found unavailable (unless another collect rule happened to match it).
14332 Collects the number of arguments from the static probe at which the
14333 tracepoint is located.
14334 @xref{Static Probe Points}.
14336 @item $_probe_arg@var{n}
14337 @var{n} is an integer between 0 and 11. Collects the @var{n}th argument
14338 from the static probe at which the tracepoint is located.
14339 @xref{Static Probe Points}.
14342 @vindex $_sdata@r{, collect}
14343 Collect static tracepoint marker specific data. Only available for
14344 static tracepoints. @xref{Tracepoint Actions,,Tracepoint Action
14345 Lists}. On the UST static tracepoints library backend, an
14346 instrumentation point resembles a @code{printf} function call. The
14347 tracing library is able to collect user specified data formatted to a
14348 character string using the format provided by the programmer that
14349 instrumented the program. Other backends have similar mechanisms.
14350 Here's an example of a UST marker call:
14353 const char master_name[] = "$your_name";
14354 trace_mark(channel1, marker1, "hello %s", master_name)
14357 In this case, collecting @code{$_sdata} collects the string
14358 @samp{hello $yourname}. When analyzing the trace buffer, you can
14359 inspect @samp{$_sdata} like any other variable available to
14363 You can give several consecutive @code{collect} commands, each one
14364 with a single argument, or one @code{collect} command with several
14365 arguments separated by commas; the effect is the same.
14367 The optional @var{mods} changes the usual handling of the arguments.
14368 @code{s} requests that pointers to chars be handled as strings, in
14369 particular collecting the contents of the memory being pointed at, up
14370 to the first zero. The upper bound is by default the value of the
14371 @code{print elements} variable; if @code{s} is followed by a decimal
14372 number, that is the upper bound instead. So for instance
14373 @samp{collect/s25 mystr} collects as many as 25 characters at
14376 The command @code{info scope} (@pxref{Symbols, info scope}) is
14377 particularly useful for figuring out what data to collect.
14379 @kindex teval @r{(tracepoints)}
14380 @item teval @var{expr1}, @var{expr2}, @dots{}
14381 Evaluate the given expressions when the tracepoint is hit. This
14382 command accepts a comma-separated list of expressions. The results
14383 are discarded, so this is mainly useful for assigning values to trace
14384 state variables (@pxref{Trace State Variables}) without adding those
14385 values to the trace buffer, as would be the case if the @code{collect}
14388 @kindex while-stepping @r{(tracepoints)}
14389 @item while-stepping @var{n}
14390 Perform @var{n} single-step instruction traces after the tracepoint,
14391 collecting new data after each step. The @code{while-stepping}
14392 command is followed by the list of what to collect while stepping
14393 (followed by its own @code{end} command):
14396 > while-stepping 12
14397 > collect $regs, myglobal
14403 Note that @code{$pc} is not automatically collected by
14404 @code{while-stepping}; you need to explicitly collect that register if
14405 you need it. You may abbreviate @code{while-stepping} as @code{ws} or
14408 @item set default-collect @var{expr1}, @var{expr2}, @dots{}
14409 @kindex set default-collect
14410 @cindex default collection action
14411 This variable is a list of expressions to collect at each tracepoint
14412 hit. It is effectively an additional @code{collect} action prepended
14413 to every tracepoint action list. The expressions are parsed
14414 individually for each tracepoint, so for instance a variable named
14415 @code{xyz} may be interpreted as a global for one tracepoint, and a
14416 local for another, as appropriate to the tracepoint's location.
14418 @item show default-collect
14419 @kindex show default-collect
14420 Show the list of expressions that are collected by default at each
14425 @node Listing Tracepoints
14426 @subsection Listing Tracepoints
14429 @kindex info tracepoints @r{[}@var{n}@dots{}@r{]}
14430 @kindex info tp @r{[}@var{n}@dots{}@r{]}
14431 @cindex information about tracepoints
14432 @item info tracepoints @r{[}@var{num}@dots{}@r{]}
14433 Display information about the tracepoint @var{num}. If you don't
14434 specify a tracepoint number, displays information about all the
14435 tracepoints defined so far. The format is similar to that used for
14436 @code{info breakpoints}; in fact, @code{info tracepoints} is the same
14437 command, simply restricting itself to tracepoints.
14439 A tracepoint's listing may include additional information specific to
14444 its passcount as given by the @code{passcount @var{n}} command
14447 the state about installed on target of each location
14451 (@value{GDBP}) @b{info trace}
14452 Num Type Disp Enb Address What
14453 1 tracepoint keep y 0x0804ab57 in foo() at main.cxx:7
14455 collect globfoo, $regs
14460 2 tracepoint keep y <MULTIPLE>
14462 2.1 y 0x0804859c in func4 at change-loc.h:35
14463 installed on target
14464 2.2 y 0xb7ffc480 in func4 at change-loc.h:35
14465 installed on target
14466 2.3 y <PENDING> set_tracepoint
14467 3 tracepoint keep y 0x080485b1 in foo at change-loc.c:29
14468 not installed on target
14473 This command can be abbreviated @code{info tp}.
14476 @node Listing Static Tracepoint Markers
14477 @subsection Listing Static Tracepoint Markers
14480 @kindex info static-tracepoint-markers
14481 @cindex information about static tracepoint markers
14482 @item info static-tracepoint-markers
14483 Display information about all static tracepoint markers defined in the
14486 For each marker, the following columns are printed:
14490 An incrementing counter, output to help readability. This is not a
14493 The marker ID, as reported by the target.
14494 @item Enabled or Disabled
14495 Probed markers are tagged with @samp{y}. @samp{n} identifies marks
14496 that are not enabled.
14498 Where the marker is in your program, as a memory address.
14500 Where the marker is in the source for your program, as a file and line
14501 number. If the debug information included in the program does not
14502 allow @value{GDBN} to locate the source of the marker, this column
14503 will be left blank.
14507 In addition, the following information may be printed for each marker:
14511 User data passed to the tracing library by the marker call. In the
14512 UST backend, this is the format string passed as argument to the
14514 @item Static tracepoints probing the marker
14515 The list of static tracepoints attached to the marker.
14519 (@value{GDBP}) info static-tracepoint-markers
14520 Cnt ID Enb Address What
14521 1 ust/bar2 y 0x0000000000400e1a in main at stexample.c:25
14522 Data: number1 %d number2 %d
14523 Probed by static tracepoints: #2
14524 2 ust/bar33 n 0x0000000000400c87 in main at stexample.c:24
14530 @node Starting and Stopping Trace Experiments
14531 @subsection Starting and Stopping Trace Experiments
14534 @kindex tstart [ @var{notes} ]
14535 @cindex start a new trace experiment
14536 @cindex collected data discarded
14538 This command starts the trace experiment, and begins collecting data.
14539 It has the side effect of discarding all the data collected in the
14540 trace buffer during the previous trace experiment. If any arguments
14541 are supplied, they are taken as a note and stored with the trace
14542 experiment's state. The notes may be arbitrary text, and are
14543 especially useful with disconnected tracing in a multi-user context;
14544 the notes can explain what the trace is doing, supply user contact
14545 information, and so forth.
14547 @kindex tstop [ @var{notes} ]
14548 @cindex stop a running trace experiment
14550 This command stops the trace experiment. If any arguments are
14551 supplied, they are recorded with the experiment as a note. This is
14552 useful if you are stopping a trace started by someone else, for
14553 instance if the trace is interfering with the system's behavior and
14554 needs to be stopped quickly.
14556 @strong{Note}: a trace experiment and data collection may stop
14557 automatically if any tracepoint's passcount is reached
14558 (@pxref{Tracepoint Passcounts}), or if the trace buffer becomes full.
14561 @cindex status of trace data collection
14562 @cindex trace experiment, status of
14564 This command displays the status of the current trace data
14568 Here is an example of the commands we described so far:
14571 (@value{GDBP}) @b{trace gdb_c_test}
14572 (@value{GDBP}) @b{actions}
14573 Enter actions for tracepoint #1, one per line.
14574 > collect $regs,$locals,$args
14575 > while-stepping 11
14579 (@value{GDBP}) @b{tstart}
14580 [time passes @dots{}]
14581 (@value{GDBP}) @b{tstop}
14584 @anchor{disconnected tracing}
14585 @cindex disconnected tracing
14586 You can choose to continue running the trace experiment even if
14587 @value{GDBN} disconnects from the target, voluntarily or
14588 involuntarily. For commands such as @code{detach}, the debugger will
14589 ask what you want to do with the trace. But for unexpected
14590 terminations (@value{GDBN} crash, network outage), it would be
14591 unfortunate to lose hard-won trace data, so the variable
14592 @code{disconnected-tracing} lets you decide whether the trace should
14593 continue running without @value{GDBN}.
14596 @item set disconnected-tracing on
14597 @itemx set disconnected-tracing off
14598 @kindex set disconnected-tracing
14599 Choose whether a tracing run should continue to run if @value{GDBN}
14600 has disconnected from the target. Note that @code{detach} or
14601 @code{quit} will ask you directly what to do about a running trace no
14602 matter what this variable's setting, so the variable is mainly useful
14603 for handling unexpected situations, such as loss of the network.
14605 @item show disconnected-tracing
14606 @kindex show disconnected-tracing
14607 Show the current choice for disconnected tracing.
14611 When you reconnect to the target, the trace experiment may or may not
14612 still be running; it might have filled the trace buffer in the
14613 meantime, or stopped for one of the other reasons. If it is running,
14614 it will continue after reconnection.
14616 Upon reconnection, the target will upload information about the
14617 tracepoints in effect. @value{GDBN} will then compare that
14618 information to the set of tracepoints currently defined, and attempt
14619 to match them up, allowing for the possibility that the numbers may
14620 have changed due to creation and deletion in the meantime. If one of
14621 the target's tracepoints does not match any in @value{GDBN}, the
14622 debugger will create a new tracepoint, so that you have a number with
14623 which to specify that tracepoint. This matching-up process is
14624 necessarily heuristic, and it may result in useless tracepoints being
14625 created; you may simply delete them if they are of no use.
14627 @cindex circular trace buffer
14628 If your target agent supports a @dfn{circular trace buffer}, then you
14629 can run a trace experiment indefinitely without filling the trace
14630 buffer; when space runs out, the agent deletes already-collected trace
14631 frames, oldest first, until there is enough room to continue
14632 collecting. This is especially useful if your tracepoints are being
14633 hit too often, and your trace gets terminated prematurely because the
14634 buffer is full. To ask for a circular trace buffer, simply set
14635 @samp{circular-trace-buffer} to on. You can set this at any time,
14636 including during tracing; if the agent can do it, it will change
14637 buffer handling on the fly, otherwise it will not take effect until
14641 @item set circular-trace-buffer on
14642 @itemx set circular-trace-buffer off
14643 @kindex set circular-trace-buffer
14644 Choose whether a tracing run should use a linear or circular buffer
14645 for trace data. A linear buffer will not lose any trace data, but may
14646 fill up prematurely, while a circular buffer will discard old trace
14647 data, but it will have always room for the latest tracepoint hits.
14649 @item show circular-trace-buffer
14650 @kindex show circular-trace-buffer
14651 Show the current choice for the trace buffer. Note that this may not
14652 match the agent's current buffer handling, nor is it guaranteed to
14653 match the setting that might have been in effect during a past run,
14654 for instance if you are looking at frames from a trace file.
14659 @item set trace-buffer-size @var{n}
14660 @itemx set trace-buffer-size unlimited
14661 @kindex set trace-buffer-size
14662 Request that the target use a trace buffer of @var{n} bytes. Not all
14663 targets will honor the request; they may have a compiled-in size for
14664 the trace buffer, or some other limitation. Set to a value of
14665 @code{unlimited} or @code{-1} to let the target use whatever size it
14666 likes. This is also the default.
14668 @item show trace-buffer-size
14669 @kindex show trace-buffer-size
14670 Show the current requested size for the trace buffer. Note that this
14671 will only match the actual size if the target supports size-setting,
14672 and was able to handle the requested size. For instance, if the
14673 target can only change buffer size between runs, this variable will
14674 not reflect the change until the next run starts. Use @code{tstatus}
14675 to get a report of the actual buffer size.
14679 @item set trace-user @var{text}
14680 @kindex set trace-user
14682 @item show trace-user
14683 @kindex show trace-user
14685 @item set trace-notes @var{text}
14686 @kindex set trace-notes
14687 Set the trace run's notes.
14689 @item show trace-notes
14690 @kindex show trace-notes
14691 Show the trace run's notes.
14693 @item set trace-stop-notes @var{text}
14694 @kindex set trace-stop-notes
14695 Set the trace run's stop notes. The handling of the note is as for
14696 @code{tstop} arguments; the set command is convenient way to fix a
14697 stop note that is mistaken or incomplete.
14699 @item show trace-stop-notes
14700 @kindex show trace-stop-notes
14701 Show the trace run's stop notes.
14705 @node Tracepoint Restrictions
14706 @subsection Tracepoint Restrictions
14708 @cindex tracepoint restrictions
14709 There are a number of restrictions on the use of tracepoints. As
14710 described above, tracepoint data gathering occurs on the target
14711 without interaction from @value{GDBN}. Thus the full capabilities of
14712 the debugger are not available during data gathering, and then at data
14713 examination time, you will be limited by only having what was
14714 collected. The following items describe some common problems, but it
14715 is not exhaustive, and you may run into additional difficulties not
14721 Tracepoint expressions are intended to gather objects (lvalues). Thus
14722 the full flexibility of GDB's expression evaluator is not available.
14723 You cannot call functions, cast objects to aggregate types, access
14724 convenience variables or modify values (except by assignment to trace
14725 state variables). Some language features may implicitly call
14726 functions (for instance Objective-C fields with accessors), and therefore
14727 cannot be collected either.
14730 Collection of local variables, either individually or in bulk with
14731 @code{$locals} or @code{$args}, during @code{while-stepping} may
14732 behave erratically. The stepping action may enter a new scope (for
14733 instance by stepping into a function), or the location of the variable
14734 may change (for instance it is loaded into a register). The
14735 tracepoint data recorded uses the location information for the
14736 variables that is correct for the tracepoint location. When the
14737 tracepoint is created, it is not possible, in general, to determine
14738 where the steps of a @code{while-stepping} sequence will advance the
14739 program---particularly if a conditional branch is stepped.
14742 Collection of an incompletely-initialized or partially-destroyed object
14743 may result in something that @value{GDBN} cannot display, or displays
14744 in a misleading way.
14747 When @value{GDBN} displays a pointer to character it automatically
14748 dereferences the pointer to also display characters of the string
14749 being pointed to. However, collecting the pointer during tracing does
14750 not automatically collect the string. You need to explicitly
14751 dereference the pointer and provide size information if you want to
14752 collect not only the pointer, but the memory pointed to. For example,
14753 @code{*ptr@@50} can be used to collect the 50 element array pointed to
14757 It is not possible to collect a complete stack backtrace at a
14758 tracepoint. Instead, you may collect the registers and a few hundred
14759 bytes from the stack pointer with something like @code{*(unsigned char *)$esp@@300}
14760 (adjust to use the name of the actual stack pointer register on your
14761 target architecture, and the amount of stack you wish to capture).
14762 Then the @code{backtrace} command will show a partial backtrace when
14763 using a trace frame. The number of stack frames that can be examined
14764 depends on the sizes of the frames in the collected stack. Note that
14765 if you ask for a block so large that it goes past the bottom of the
14766 stack, the target agent may report an error trying to read from an
14770 If you do not collect registers at a tracepoint, @value{GDBN} can
14771 infer that the value of @code{$pc} must be the same as the address of
14772 the tracepoint and use that when you are looking at a trace frame
14773 for that tracepoint. However, this cannot work if the tracepoint has
14774 multiple locations (for instance if it was set in a function that was
14775 inlined), or if it has a @code{while-stepping} loop. In those cases
14776 @value{GDBN} will warn you that it can't infer @code{$pc}, and default
14781 @node Analyze Collected Data
14782 @section Using the Collected Data
14784 After the tracepoint experiment ends, you use @value{GDBN} commands
14785 for examining the trace data. The basic idea is that each tracepoint
14786 collects a trace @dfn{snapshot} every time it is hit and another
14787 snapshot every time it single-steps. All these snapshots are
14788 consecutively numbered from zero and go into a buffer, and you can
14789 examine them later. The way you examine them is to @dfn{focus} on a
14790 specific trace snapshot. When the remote stub is focused on a trace
14791 snapshot, it will respond to all @value{GDBN} requests for memory and
14792 registers by reading from the buffer which belongs to that snapshot,
14793 rather than from @emph{real} memory or registers of the program being
14794 debugged. This means that @strong{all} @value{GDBN} commands
14795 (@code{print}, @code{info registers}, @code{backtrace}, etc.) will
14796 behave as if we were currently debugging the program state as it was
14797 when the tracepoint occurred. Any requests for data that are not in
14798 the buffer will fail.
14801 * tfind:: How to select a trace snapshot
14802 * tdump:: How to display all data for a snapshot
14803 * save tracepoints:: How to save tracepoints for a future run
14807 @subsection @code{tfind @var{n}}
14810 @cindex select trace snapshot
14811 @cindex find trace snapshot
14812 The basic command for selecting a trace snapshot from the buffer is
14813 @code{tfind @var{n}}, which finds trace snapshot number @var{n},
14814 counting from zero. If no argument @var{n} is given, the next
14815 snapshot is selected.
14817 Here are the various forms of using the @code{tfind} command.
14821 Find the first snapshot in the buffer. This is a synonym for
14822 @code{tfind 0} (since 0 is the number of the first snapshot).
14825 Stop debugging trace snapshots, resume @emph{live} debugging.
14828 Same as @samp{tfind none}.
14831 No argument means find the next trace snapshot or find the first
14832 one if no trace snapshot is selected.
14835 Find the previous trace snapshot before the current one. This permits
14836 retracing earlier steps.
14838 @item tfind tracepoint @var{num}
14839 Find the next snapshot associated with tracepoint @var{num}. Search
14840 proceeds forward from the last examined trace snapshot. If no
14841 argument @var{num} is given, it means find the next snapshot collected
14842 for the same tracepoint as the current snapshot.
14844 @item tfind pc @var{addr}
14845 Find the next snapshot associated with the value @var{addr} of the
14846 program counter. Search proceeds forward from the last examined trace
14847 snapshot. If no argument @var{addr} is given, it means find the next
14848 snapshot with the same value of PC as the current snapshot.
14850 @item tfind outside @var{addr1}, @var{addr2}
14851 Find the next snapshot whose PC is outside the given range of
14852 addresses (exclusive).
14854 @item tfind range @var{addr1}, @var{addr2}
14855 Find the next snapshot whose PC is between @var{addr1} and
14856 @var{addr2} (inclusive).
14858 @item tfind line @r{[}@var{file}:@r{]}@var{n}
14859 Find the next snapshot associated with the source line @var{n}. If
14860 the optional argument @var{file} is given, refer to line @var{n} in
14861 that source file. Search proceeds forward from the last examined
14862 trace snapshot. If no argument @var{n} is given, it means find the
14863 next line other than the one currently being examined; thus saying
14864 @code{tfind line} repeatedly can appear to have the same effect as
14865 stepping from line to line in a @emph{live} debugging session.
14868 The default arguments for the @code{tfind} commands are specifically
14869 designed to make it easy to scan through the trace buffer. For
14870 instance, @code{tfind} with no argument selects the next trace
14871 snapshot, and @code{tfind -} with no argument selects the previous
14872 trace snapshot. So, by giving one @code{tfind} command, and then
14873 simply hitting @key{RET} repeatedly you can examine all the trace
14874 snapshots in order. Or, by saying @code{tfind -} and then hitting
14875 @key{RET} repeatedly you can examine the snapshots in reverse order.
14876 The @code{tfind line} command with no argument selects the snapshot
14877 for the next source line executed. The @code{tfind pc} command with
14878 no argument selects the next snapshot with the same program counter
14879 (PC) as the current frame. The @code{tfind tracepoint} command with
14880 no argument selects the next trace snapshot collected by the same
14881 tracepoint as the current one.
14883 In addition to letting you scan through the trace buffer manually,
14884 these commands make it easy to construct @value{GDBN} scripts that
14885 scan through the trace buffer and print out whatever collected data
14886 you are interested in. Thus, if we want to examine the PC, FP, and SP
14887 registers from each trace frame in the buffer, we can say this:
14890 (@value{GDBP}) @b{tfind start}
14891 (@value{GDBP}) @b{while ($trace_frame != -1)}
14892 > printf "Frame %d, PC = %08X, SP = %08X, FP = %08X\n", \
14893 $trace_frame, $pc, $sp, $fp
14897 Frame 0, PC = 0020DC64, SP = 0030BF3C, FP = 0030BF44
14898 Frame 1, PC = 0020DC6C, SP = 0030BF38, FP = 0030BF44
14899 Frame 2, PC = 0020DC70, SP = 0030BF34, FP = 0030BF44
14900 Frame 3, PC = 0020DC74, SP = 0030BF30, FP = 0030BF44
14901 Frame 4, PC = 0020DC78, SP = 0030BF2C, FP = 0030BF44
14902 Frame 5, PC = 0020DC7C, SP = 0030BF28, FP = 0030BF44
14903 Frame 6, PC = 0020DC80, SP = 0030BF24, FP = 0030BF44
14904 Frame 7, PC = 0020DC84, SP = 0030BF20, FP = 0030BF44
14905 Frame 8, PC = 0020DC88, SP = 0030BF1C, FP = 0030BF44
14906 Frame 9, PC = 0020DC8E, SP = 0030BF18, FP = 0030BF44
14907 Frame 10, PC = 00203F6C, SP = 0030BE3C, FP = 0030BF14
14910 Or, if we want to examine the variable @code{X} at each source line in
14914 (@value{GDBP}) @b{tfind start}
14915 (@value{GDBP}) @b{while ($trace_frame != -1)}
14916 > printf "Frame %d, X == %d\n", $trace_frame, X
14926 @subsection @code{tdump}
14928 @cindex dump all data collected at tracepoint
14929 @cindex tracepoint data, display
14931 This command takes no arguments. It prints all the data collected at
14932 the current trace snapshot.
14935 (@value{GDBP}) @b{trace 444}
14936 (@value{GDBP}) @b{actions}
14937 Enter actions for tracepoint #2, one per line:
14938 > collect $regs, $locals, $args, gdb_long_test
14941 (@value{GDBP}) @b{tstart}
14943 (@value{GDBP}) @b{tfind line 444}
14944 #0 gdb_test (p1=0x11, p2=0x22, p3=0x33, p4=0x44, p5=0x55, p6=0x66)
14946 444 printp( "%s: arguments = 0x%X 0x%X 0x%X 0x%X 0x%X 0x%X\n", )
14948 (@value{GDBP}) @b{tdump}
14949 Data collected at tracepoint 2, trace frame 1:
14950 d0 0xc4aa0085 -995491707
14954 d4 0x71aea3d 119204413
14957 d7 0x380035 3670069
14958 a0 0x19e24a 1696330
14959 a1 0x3000668 50333288
14961 a3 0x322000 3284992
14962 a4 0x3000698 50333336
14963 a5 0x1ad3cc 1758156
14964 fp 0x30bf3c 0x30bf3c
14965 sp 0x30bf34 0x30bf34
14967 pc 0x20b2c8 0x20b2c8
14971 p = 0x20e5b4 "gdb-test"
14978 gdb_long_test = 17 '\021'
14983 @code{tdump} works by scanning the tracepoint's current collection
14984 actions and printing the value of each expression listed. So
14985 @code{tdump} can fail, if after a run, you change the tracepoint's
14986 actions to mention variables that were not collected during the run.
14988 Also, for tracepoints with @code{while-stepping} loops, @code{tdump}
14989 uses the collected value of @code{$pc} to distinguish between trace
14990 frames that were collected at the tracepoint hit, and frames that were
14991 collected while stepping. This allows it to correctly choose whether
14992 to display the basic list of collections, or the collections from the
14993 body of the while-stepping loop. However, if @code{$pc} was not collected,
14994 then @code{tdump} will always attempt to dump using the basic collection
14995 list, and may fail if a while-stepping frame does not include all the
14996 same data that is collected at the tracepoint hit.
14997 @c This is getting pretty arcane, example would be good.
14999 @node save tracepoints
15000 @subsection @code{save tracepoints @var{filename}}
15001 @kindex save tracepoints
15002 @kindex save-tracepoints
15003 @cindex save tracepoints for future sessions
15005 This command saves all current tracepoint definitions together with
15006 their actions and passcounts, into a file @file{@var{filename}}
15007 suitable for use in a later debugging session. To read the saved
15008 tracepoint definitions, use the @code{source} command (@pxref{Command
15009 Files}). The @w{@code{save-tracepoints}} command is a deprecated
15010 alias for @w{@code{save tracepoints}}
15012 @node Tracepoint Variables
15013 @section Convenience Variables for Tracepoints
15014 @cindex tracepoint variables
15015 @cindex convenience variables for tracepoints
15018 @vindex $trace_frame
15019 @item (int) $trace_frame
15020 The current trace snapshot (a.k.a.@: @dfn{frame}) number, or -1 if no
15021 snapshot is selected.
15023 @vindex $tracepoint
15024 @item (int) $tracepoint
15025 The tracepoint for the current trace snapshot.
15027 @vindex $trace_line
15028 @item (int) $trace_line
15029 The line number for the current trace snapshot.
15031 @vindex $trace_file
15032 @item (char []) $trace_file
15033 The source file for the current trace snapshot.
15035 @vindex $trace_func
15036 @item (char []) $trace_func
15037 The name of the function containing @code{$tracepoint}.
15040 Note: @code{$trace_file} is not suitable for use in @code{printf},
15041 use @code{output} instead.
15043 Here's a simple example of using these convenience variables for
15044 stepping through all the trace snapshots and printing some of their
15045 data. Note that these are not the same as trace state variables,
15046 which are managed by the target.
15049 (@value{GDBP}) @b{tfind start}
15051 (@value{GDBP}) @b{while $trace_frame != -1}
15052 > output $trace_file
15053 > printf ", line %d (tracepoint #%d)\n", $trace_line, $tracepoint
15059 @section Using Trace Files
15060 @cindex trace files
15062 In some situations, the target running a trace experiment may no
15063 longer be available; perhaps it crashed, or the hardware was needed
15064 for a different activity. To handle these cases, you can arrange to
15065 dump the trace data into a file, and later use that file as a source
15066 of trace data, via the @code{target tfile} command.
15071 @item tsave [ -r ] @var{filename}
15072 @itemx tsave [-ctf] @var{dirname}
15073 Save the trace data to @var{filename}. By default, this command
15074 assumes that @var{filename} refers to the host filesystem, so if
15075 necessary @value{GDBN} will copy raw trace data up from the target and
15076 then save it. If the target supports it, you can also supply the
15077 optional argument @code{-r} (``remote'') to direct the target to save
15078 the data directly into @var{filename} in its own filesystem, which may be
15079 more efficient if the trace buffer is very large. (Note, however, that
15080 @code{target tfile} can only read from files accessible to the host.)
15081 By default, this command will save trace frame in tfile format.
15082 You can supply the optional argument @code{-ctf} to save data in CTF
15083 format. The @dfn{Common Trace Format} (CTF) is proposed as a trace format
15084 that can be shared by multiple debugging and tracing tools. Please go to
15085 @indicateurl{http://www.efficios.com/ctf} to get more information.
15087 @kindex target tfile
15091 @item target tfile @var{filename}
15092 @itemx target ctf @var{dirname}
15093 Use the file named @var{filename} or directory named @var{dirname} as
15094 a source of trace data. Commands that examine data work as they do with
15095 a live target, but it is not possible to run any new trace experiments.
15096 @code{tstatus} will report the state of the trace run at the moment
15097 the data was saved, as well as the current trace frame you are examining.
15098 Both @var{filename} and @var{dirname} must be on a filesystem accessible to
15102 (@value{GDBP}) target ctf ctf.ctf
15103 (@value{GDBP}) tfind
15104 Found trace frame 0, tracepoint 2
15105 39 ++a; /* set tracepoint 1 here */
15106 (@value{GDBP}) tdump
15107 Data collected at tracepoint 2, trace frame 0:
15111 c = @{"123", "456", "789", "123", "456", "789"@}
15112 d = @{@{@{a = 1, b = 2@}, @{a = 3, b = 4@}@}, @{@{a = 5, b = 6@}, @{a = 7, b = 8@}@}@}
15120 @chapter Debugging Programs That Use Overlays
15123 If your program is too large to fit completely in your target system's
15124 memory, you can sometimes use @dfn{overlays} to work around this
15125 problem. @value{GDBN} provides some support for debugging programs that
15129 * How Overlays Work:: A general explanation of overlays.
15130 * Overlay Commands:: Managing overlays in @value{GDBN}.
15131 * Automatic Overlay Debugging:: @value{GDBN} can find out which overlays are
15132 mapped by asking the inferior.
15133 * Overlay Sample Program:: A sample program using overlays.
15136 @node How Overlays Work
15137 @section How Overlays Work
15138 @cindex mapped overlays
15139 @cindex unmapped overlays
15140 @cindex load address, overlay's
15141 @cindex mapped address
15142 @cindex overlay area
15144 Suppose you have a computer whose instruction address space is only 64
15145 kilobytes long, but which has much more memory which can be accessed by
15146 other means: special instructions, segment registers, or memory
15147 management hardware, for example. Suppose further that you want to
15148 adapt a program which is larger than 64 kilobytes to run on this system.
15150 One solution is to identify modules of your program which are relatively
15151 independent, and need not call each other directly; call these modules
15152 @dfn{overlays}. Separate the overlays from the main program, and place
15153 their machine code in the larger memory. Place your main program in
15154 instruction memory, but leave at least enough space there to hold the
15155 largest overlay as well.
15157 Now, to call a function located in an overlay, you must first copy that
15158 overlay's machine code from the large memory into the space set aside
15159 for it in the instruction memory, and then jump to its entry point
15162 @c NB: In the below the mapped area's size is greater or equal to the
15163 @c size of all overlays. This is intentional to remind the developer
15164 @c that overlays don't necessarily need to be the same size.
15168 Data Instruction Larger
15169 Address Space Address Space Address Space
15170 +-----------+ +-----------+ +-----------+
15172 +-----------+ +-----------+ +-----------+<-- overlay 1
15173 | program | | main | .----| overlay 1 | load address
15174 | variables | | program | | +-----------+
15175 | and heap | | | | | |
15176 +-----------+ | | | +-----------+<-- overlay 2
15177 | | +-----------+ | | | load address
15178 +-----------+ | | | .-| overlay 2 |
15180 mapped --->+-----------+ | | +-----------+
15181 address | | | | | |
15182 | overlay | <-' | | |
15183 | area | <---' +-----------+<-- overlay 3
15184 | | <---. | | load address
15185 +-----------+ `--| overlay 3 |
15192 @anchor{A code overlay}A code overlay
15196 The diagram (@pxref{A code overlay}) shows a system with separate data
15197 and instruction address spaces. To map an overlay, the program copies
15198 its code from the larger address space to the instruction address space.
15199 Since the overlays shown here all use the same mapped address, only one
15200 may be mapped at a time. For a system with a single address space for
15201 data and instructions, the diagram would be similar, except that the
15202 program variables and heap would share an address space with the main
15203 program and the overlay area.
15205 An overlay loaded into instruction memory and ready for use is called a
15206 @dfn{mapped} overlay; its @dfn{mapped address} is its address in the
15207 instruction memory. An overlay not present (or only partially present)
15208 in instruction memory is called @dfn{unmapped}; its @dfn{load address}
15209 is its address in the larger memory. The mapped address is also called
15210 the @dfn{virtual memory address}, or @dfn{VMA}; the load address is also
15211 called the @dfn{load memory address}, or @dfn{LMA}.
15213 Unfortunately, overlays are not a completely transparent way to adapt a
15214 program to limited instruction memory. They introduce a new set of
15215 global constraints you must keep in mind as you design your program:
15220 Before calling or returning to a function in an overlay, your program
15221 must make sure that overlay is actually mapped. Otherwise, the call or
15222 return will transfer control to the right address, but in the wrong
15223 overlay, and your program will probably crash.
15226 If the process of mapping an overlay is expensive on your system, you
15227 will need to choose your overlays carefully to minimize their effect on
15228 your program's performance.
15231 The executable file you load onto your system must contain each
15232 overlay's instructions, appearing at the overlay's load address, not its
15233 mapped address. However, each overlay's instructions must be relocated
15234 and its symbols defined as if the overlay were at its mapped address.
15235 You can use GNU linker scripts to specify different load and relocation
15236 addresses for pieces of your program; see @ref{Overlay Description,,,
15237 ld.info, Using ld: the GNU linker}.
15240 The procedure for loading executable files onto your system must be able
15241 to load their contents into the larger address space as well as the
15242 instruction and data spaces.
15246 The overlay system described above is rather simple, and could be
15247 improved in many ways:
15252 If your system has suitable bank switch registers or memory management
15253 hardware, you could use those facilities to make an overlay's load area
15254 contents simply appear at their mapped address in instruction space.
15255 This would probably be faster than copying the overlay to its mapped
15256 area in the usual way.
15259 If your overlays are small enough, you could set aside more than one
15260 overlay area, and have more than one overlay mapped at a time.
15263 You can use overlays to manage data, as well as instructions. In
15264 general, data overlays are even less transparent to your design than
15265 code overlays: whereas code overlays only require care when you call or
15266 return to functions, data overlays require care every time you access
15267 the data. Also, if you change the contents of a data overlay, you
15268 must copy its contents back out to its load address before you can copy a
15269 different data overlay into the same mapped area.
15274 @node Overlay Commands
15275 @section Overlay Commands
15277 To use @value{GDBN}'s overlay support, each overlay in your program must
15278 correspond to a separate section of the executable file. The section's
15279 virtual memory address and load memory address must be the overlay's
15280 mapped and load addresses. Identifying overlays with sections allows
15281 @value{GDBN} to determine the appropriate address of a function or
15282 variable, depending on whether the overlay is mapped or not.
15284 @value{GDBN}'s overlay commands all start with the word @code{overlay};
15285 you can abbreviate this as @code{ov} or @code{ovly}. The commands are:
15290 Disable @value{GDBN}'s overlay support. When overlay support is
15291 disabled, @value{GDBN} assumes that all functions and variables are
15292 always present at their mapped addresses. By default, @value{GDBN}'s
15293 overlay support is disabled.
15295 @item overlay manual
15296 @cindex manual overlay debugging
15297 Enable @dfn{manual} overlay debugging. In this mode, @value{GDBN}
15298 relies on you to tell it which overlays are mapped, and which are not,
15299 using the @code{overlay map-overlay} and @code{overlay unmap-overlay}
15300 commands described below.
15302 @item overlay map-overlay @var{overlay}
15303 @itemx overlay map @var{overlay}
15304 @cindex map an overlay
15305 Tell @value{GDBN} that @var{overlay} is now mapped; @var{overlay} must
15306 be the name of the object file section containing the overlay. When an
15307 overlay is mapped, @value{GDBN} assumes it can find the overlay's
15308 functions and variables at their mapped addresses. @value{GDBN} assumes
15309 that any other overlays whose mapped ranges overlap that of
15310 @var{overlay} are now unmapped.
15312 @item overlay unmap-overlay @var{overlay}
15313 @itemx overlay unmap @var{overlay}
15314 @cindex unmap an overlay
15315 Tell @value{GDBN} that @var{overlay} is no longer mapped; @var{overlay}
15316 must be the name of the object file section containing the overlay.
15317 When an overlay is unmapped, @value{GDBN} assumes it can find the
15318 overlay's functions and variables at their load addresses.
15321 Enable @dfn{automatic} overlay debugging. In this mode, @value{GDBN}
15322 consults a data structure the overlay manager maintains in the inferior
15323 to see which overlays are mapped. For details, see @ref{Automatic
15324 Overlay Debugging}.
15326 @item overlay load-target
15327 @itemx overlay load
15328 @cindex reloading the overlay table
15329 Re-read the overlay table from the inferior. Normally, @value{GDBN}
15330 re-reads the table @value{GDBN} automatically each time the inferior
15331 stops, so this command should only be necessary if you have changed the
15332 overlay mapping yourself using @value{GDBN}. This command is only
15333 useful when using automatic overlay debugging.
15335 @item overlay list-overlays
15336 @itemx overlay list
15337 @cindex listing mapped overlays
15338 Display a list of the overlays currently mapped, along with their mapped
15339 addresses, load addresses, and sizes.
15343 Normally, when @value{GDBN} prints a code address, it includes the name
15344 of the function the address falls in:
15347 (@value{GDBP}) print main
15348 $3 = @{int ()@} 0x11a0 <main>
15351 When overlay debugging is enabled, @value{GDBN} recognizes code in
15352 unmapped overlays, and prints the names of unmapped functions with
15353 asterisks around them. For example, if @code{foo} is a function in an
15354 unmapped overlay, @value{GDBN} prints it this way:
15357 (@value{GDBP}) overlay list
15358 No sections are mapped.
15359 (@value{GDBP}) print foo
15360 $5 = @{int (int)@} 0x100000 <*foo*>
15363 When @code{foo}'s overlay is mapped, @value{GDBN} prints the function's
15367 (@value{GDBP}) overlay list
15368 Section .ov.foo.text, loaded at 0x100000 - 0x100034,
15369 mapped at 0x1016 - 0x104a
15370 (@value{GDBP}) print foo
15371 $6 = @{int (int)@} 0x1016 <foo>
15374 When overlay debugging is enabled, @value{GDBN} can find the correct
15375 address for functions and variables in an overlay, whether or not the
15376 overlay is mapped. This allows most @value{GDBN} commands, like
15377 @code{break} and @code{disassemble}, to work normally, even on unmapped
15378 code. However, @value{GDBN}'s breakpoint support has some limitations:
15382 @cindex breakpoints in overlays
15383 @cindex overlays, setting breakpoints in
15384 You can set breakpoints in functions in unmapped overlays, as long as
15385 @value{GDBN} can write to the overlay at its load address.
15387 @value{GDBN} can not set hardware or simulator-based breakpoints in
15388 unmapped overlays. However, if you set a breakpoint at the end of your
15389 overlay manager (and tell @value{GDBN} which overlays are now mapped, if
15390 you are using manual overlay management), @value{GDBN} will re-set its
15391 breakpoints properly.
15395 @node Automatic Overlay Debugging
15396 @section Automatic Overlay Debugging
15397 @cindex automatic overlay debugging
15399 @value{GDBN} can automatically track which overlays are mapped and which
15400 are not, given some simple co-operation from the overlay manager in the
15401 inferior. If you enable automatic overlay debugging with the
15402 @code{overlay auto} command (@pxref{Overlay Commands}), @value{GDBN}
15403 looks in the inferior's memory for certain variables describing the
15404 current state of the overlays.
15406 Here are the variables your overlay manager must define to support
15407 @value{GDBN}'s automatic overlay debugging:
15411 @item @code{_ovly_table}:
15412 This variable must be an array of the following structures:
15417 /* The overlay's mapped address. */
15420 /* The size of the overlay, in bytes. */
15421 unsigned long size;
15423 /* The overlay's load address. */
15426 /* Non-zero if the overlay is currently mapped;
15428 unsigned long mapped;
15432 @item @code{_novlys}:
15433 This variable must be a four-byte signed integer, holding the total
15434 number of elements in @code{_ovly_table}.
15438 To decide whether a particular overlay is mapped or not, @value{GDBN}
15439 looks for an entry in @w{@code{_ovly_table}} whose @code{vma} and
15440 @code{lma} members equal the VMA and LMA of the overlay's section in the
15441 executable file. When @value{GDBN} finds a matching entry, it consults
15442 the entry's @code{mapped} member to determine whether the overlay is
15445 In addition, your overlay manager may define a function called
15446 @code{_ovly_debug_event}. If this function is defined, @value{GDBN}
15447 will silently set a breakpoint there. If the overlay manager then
15448 calls this function whenever it has changed the overlay table, this
15449 will enable @value{GDBN} to accurately keep track of which overlays
15450 are in program memory, and update any breakpoints that may be set
15451 in overlays. This will allow breakpoints to work even if the
15452 overlays are kept in ROM or other non-writable memory while they
15453 are not being executed.
15455 @node Overlay Sample Program
15456 @section Overlay Sample Program
15457 @cindex overlay example program
15459 When linking a program which uses overlays, you must place the overlays
15460 at their load addresses, while relocating them to run at their mapped
15461 addresses. To do this, you must write a linker script (@pxref{Overlay
15462 Description,,, ld.info, Using ld: the GNU linker}). Unfortunately,
15463 since linker scripts are specific to a particular host system, target
15464 architecture, and target memory layout, this manual cannot provide
15465 portable sample code demonstrating @value{GDBN}'s overlay support.
15467 However, the @value{GDBN} source distribution does contain an overlaid
15468 program, with linker scripts for a few systems, as part of its test
15469 suite. The program consists of the following files from
15470 @file{gdb/testsuite/gdb.base}:
15474 The main program file.
15476 A simple overlay manager, used by @file{overlays.c}.
15481 Overlay modules, loaded and used by @file{overlays.c}.
15484 Linker scripts for linking the test program on the @code{d10v-elf}
15485 and @code{m32r-elf} targets.
15488 You can build the test program using the @code{d10v-elf} GCC
15489 cross-compiler like this:
15492 $ d10v-elf-gcc -g -c overlays.c
15493 $ d10v-elf-gcc -g -c ovlymgr.c
15494 $ d10v-elf-gcc -g -c foo.c
15495 $ d10v-elf-gcc -g -c bar.c
15496 $ d10v-elf-gcc -g -c baz.c
15497 $ d10v-elf-gcc -g -c grbx.c
15498 $ d10v-elf-gcc -g overlays.o ovlymgr.o foo.o bar.o \
15499 baz.o grbx.o -Wl,-Td10v.ld -o overlays
15502 The build process is identical for any other architecture, except that
15503 you must substitute the appropriate compiler and linker script for the
15504 target system for @code{d10v-elf-gcc} and @code{d10v.ld}.
15508 @chapter Using @value{GDBN} with Different Languages
15511 Although programming languages generally have common aspects, they are
15512 rarely expressed in the same manner. For instance, in ANSI C,
15513 dereferencing a pointer @code{p} is accomplished by @code{*p}, but in
15514 Modula-2, it is accomplished by @code{p^}. Values can also be
15515 represented (and displayed) differently. Hex numbers in C appear as
15516 @samp{0x1ae}, while in Modula-2 they appear as @samp{1AEH}.
15518 @cindex working language
15519 Language-specific information is built into @value{GDBN} for some languages,
15520 allowing you to express operations like the above in your program's
15521 native language, and allowing @value{GDBN} to output values in a manner
15522 consistent with the syntax of your program's native language. The
15523 language you use to build expressions is called the @dfn{working
15527 * Setting:: Switching between source languages
15528 * Show:: Displaying the language
15529 * Checks:: Type and range checks
15530 * Supported Languages:: Supported languages
15531 * Unsupported Languages:: Unsupported languages
15535 @section Switching Between Source Languages
15537 There are two ways to control the working language---either have @value{GDBN}
15538 set it automatically, or select it manually yourself. You can use the
15539 @code{set language} command for either purpose. On startup, @value{GDBN}
15540 defaults to setting the language automatically. The working language is
15541 used to determine how expressions you type are interpreted, how values
15544 In addition to the working language, every source file that
15545 @value{GDBN} knows about has its own working language. For some object
15546 file formats, the compiler might indicate which language a particular
15547 source file is in. However, most of the time @value{GDBN} infers the
15548 language from the name of the file. The language of a source file
15549 controls whether C@t{++} names are demangled---this way @code{backtrace} can
15550 show each frame appropriately for its own language. There is no way to
15551 set the language of a source file from within @value{GDBN}, but you can
15552 set the language associated with a filename extension. @xref{Show, ,
15553 Displaying the Language}.
15555 This is most commonly a problem when you use a program, such
15556 as @code{cfront} or @code{f2c}, that generates C but is written in
15557 another language. In that case, make the
15558 program use @code{#line} directives in its C output; that way
15559 @value{GDBN} will know the correct language of the source code of the original
15560 program, and will display that source code, not the generated C code.
15563 * Filenames:: Filename extensions and languages.
15564 * Manually:: Setting the working language manually
15565 * Automatically:: Having @value{GDBN} infer the source language
15569 @subsection List of Filename Extensions and Languages
15571 If a source file name ends in one of the following extensions, then
15572 @value{GDBN} infers that its language is the one indicated.
15590 C@t{++} source file
15596 Objective-C source file
15600 Fortran source file
15603 Modula-2 source file
15607 Assembler source file. This actually behaves almost like C, but
15608 @value{GDBN} does not skip over function prologues when stepping.
15611 In addition, you may set the language associated with a filename
15612 extension. @xref{Show, , Displaying the Language}.
15615 @subsection Setting the Working Language
15617 If you allow @value{GDBN} to set the language automatically,
15618 expressions are interpreted the same way in your debugging session and
15621 @kindex set language
15622 If you wish, you may set the language manually. To do this, issue the
15623 command @samp{set language @var{lang}}, where @var{lang} is the name of
15624 a language, such as
15625 @code{c} or @code{modula-2}.
15626 For a list of the supported languages, type @samp{set language}.
15628 Setting the language manually prevents @value{GDBN} from updating the working
15629 language automatically. This can lead to confusion if you try
15630 to debug a program when the working language is not the same as the
15631 source language, when an expression is acceptable to both
15632 languages---but means different things. For instance, if the current
15633 source file were written in C, and @value{GDBN} was parsing Modula-2, a
15641 might not have the effect you intended. In C, this means to add
15642 @code{b} and @code{c} and place the result in @code{a}. The result
15643 printed would be the value of @code{a}. In Modula-2, this means to compare
15644 @code{a} to the result of @code{b+c}, yielding a @code{BOOLEAN} value.
15646 @node Automatically
15647 @subsection Having @value{GDBN} Infer the Source Language
15649 To have @value{GDBN} set the working language automatically, use
15650 @samp{set language local} or @samp{set language auto}. @value{GDBN}
15651 then infers the working language. That is, when your program stops in a
15652 frame (usually by encountering a breakpoint), @value{GDBN} sets the
15653 working language to the language recorded for the function in that
15654 frame. If the language for a frame is unknown (that is, if the function
15655 or block corresponding to the frame was defined in a source file that
15656 does not have a recognized extension), the current working language is
15657 not changed, and @value{GDBN} issues a warning.
15659 This may not seem necessary for most programs, which are written
15660 entirely in one source language. However, program modules and libraries
15661 written in one source language can be used by a main program written in
15662 a different source language. Using @samp{set language auto} in this
15663 case frees you from having to set the working language manually.
15666 @section Displaying the Language
15668 The following commands help you find out which language is the
15669 working language, and also what language source files were written in.
15672 @item show language
15673 @anchor{show language}
15674 @kindex show language
15675 Display the current working language. This is the
15676 language you can use with commands such as @code{print} to
15677 build and compute expressions that may involve variables in your program.
15680 @kindex info frame@r{, show the source language}
15681 Display the source language for this frame. This language becomes the
15682 working language if you use an identifier from this frame.
15683 @xref{Frame Info, ,Information about a Frame}, to identify the other
15684 information listed here.
15687 @kindex info source@r{, show the source language}
15688 Display the source language of this source file.
15689 @xref{Symbols, ,Examining the Symbol Table}, to identify the other
15690 information listed here.
15693 In unusual circumstances, you may have source files with extensions
15694 not in the standard list. You can then set the extension associated
15695 with a language explicitly:
15698 @item set extension-language @var{ext} @var{language}
15699 @kindex set extension-language
15700 Tell @value{GDBN} that source files with extension @var{ext} are to be
15701 assumed as written in the source language @var{language}.
15703 @item info extensions
15704 @kindex info extensions
15705 List all the filename extensions and the associated languages.
15709 @section Type and Range Checking
15711 Some languages are designed to guard you against making seemingly common
15712 errors through a series of compile- and run-time checks. These include
15713 checking the type of arguments to functions and operators and making
15714 sure mathematical overflows are caught at run time. Checks such as
15715 these help to ensure a program's correctness once it has been compiled
15716 by eliminating type mismatches and providing active checks for range
15717 errors when your program is running.
15719 By default @value{GDBN} checks for these errors according to the
15720 rules of the current source language. Although @value{GDBN} does not check
15721 the statements in your program, it can check expressions entered directly
15722 into @value{GDBN} for evaluation via the @code{print} command, for example.
15725 * Type Checking:: An overview of type checking
15726 * Range Checking:: An overview of range checking
15729 @cindex type checking
15730 @cindex checks, type
15731 @node Type Checking
15732 @subsection An Overview of Type Checking
15734 Some languages, such as C and C@t{++}, are strongly typed, meaning that the
15735 arguments to operators and functions have to be of the correct type,
15736 otherwise an error occurs. These checks prevent type mismatch
15737 errors from ever causing any run-time problems. For example,
15740 int klass::my_method(char *b) @{ return b ? 1 : 2; @}
15742 (@value{GDBP}) print obj.my_method (0)
15745 (@value{GDBP}) print obj.my_method (0x1234)
15746 Cannot resolve method klass::my_method to any overloaded instance
15749 The second example fails because in C@t{++} the integer constant
15750 @samp{0x1234} is not type-compatible with the pointer parameter type.
15752 For the expressions you use in @value{GDBN} commands, you can tell
15753 @value{GDBN} to not enforce strict type checking or
15754 to treat any mismatches as errors and abandon the expression;
15755 When type checking is disabled, @value{GDBN} successfully evaluates
15756 expressions like the second example above.
15758 Even if type checking is off, there may be other reasons
15759 related to type that prevent @value{GDBN} from evaluating an expression.
15760 For instance, @value{GDBN} does not know how to add an @code{int} and
15761 a @code{struct foo}. These particular type errors have nothing to do
15762 with the language in use and usually arise from expressions which make
15763 little sense to evaluate anyway.
15765 @value{GDBN} provides some additional commands for controlling type checking:
15767 @kindex set check type
15768 @kindex show check type
15770 @item set check type on
15771 @itemx set check type off
15772 Set strict type checking on or off. If any type mismatches occur in
15773 evaluating an expression while type checking is on, @value{GDBN} prints a
15774 message and aborts evaluation of the expression.
15776 @item show check type
15777 Show the current setting of type checking and whether @value{GDBN}
15778 is enforcing strict type checking rules.
15781 @cindex range checking
15782 @cindex checks, range
15783 @node Range Checking
15784 @subsection An Overview of Range Checking
15786 In some languages (such as Modula-2), it is an error to exceed the
15787 bounds of a type; this is enforced with run-time checks. Such range
15788 checking is meant to ensure program correctness by making sure
15789 computations do not overflow, or indices on an array element access do
15790 not exceed the bounds of the array.
15792 For expressions you use in @value{GDBN} commands, you can tell
15793 @value{GDBN} to treat range errors in one of three ways: ignore them,
15794 always treat them as errors and abandon the expression, or issue
15795 warnings but evaluate the expression anyway.
15797 A range error can result from numerical overflow, from exceeding an
15798 array index bound, or when you type a constant that is not a member
15799 of any type. Some languages, however, do not treat overflows as an
15800 error. In many implementations of C, mathematical overflow causes the
15801 result to ``wrap around'' to lower values---for example, if @var{m} is
15802 the largest integer value, and @var{s} is the smallest, then
15805 @var{m} + 1 @result{} @var{s}
15808 This, too, is specific to individual languages, and in some cases
15809 specific to individual compilers or machines. @xref{Supported Languages, ,
15810 Supported Languages}, for further details on specific languages.
15812 @value{GDBN} provides some additional commands for controlling the range checker:
15814 @kindex set check range
15815 @kindex show check range
15817 @item set check range auto
15818 Set range checking on or off based on the current working language.
15819 @xref{Supported Languages, ,Supported Languages}, for the default settings for
15822 @item set check range on
15823 @itemx set check range off
15824 Set range checking on or off, overriding the default setting for the
15825 current working language. A warning is issued if the setting does not
15826 match the language default. If a range error occurs and range checking is on,
15827 then a message is printed and evaluation of the expression is aborted.
15829 @item set check range warn
15830 Output messages when the @value{GDBN} range checker detects a range error,
15831 but attempt to evaluate the expression anyway. Evaluating the
15832 expression may still be impossible for other reasons, such as accessing
15833 memory that the process does not own (a typical example from many Unix
15837 Show the current setting of the range checker, and whether or not it is
15838 being set automatically by @value{GDBN}.
15841 @node Supported Languages
15842 @section Supported Languages
15844 @value{GDBN} supports C, C@t{++}, D, Go, Objective-C, Fortran,
15845 OpenCL C, Pascal, Rust, assembly, Modula-2, and Ada.
15846 @c This is false ...
15847 Some @value{GDBN} features may be used in expressions regardless of the
15848 language you use: the @value{GDBN} @code{@@} and @code{::} operators,
15849 and the @samp{@{type@}addr} construct (@pxref{Expressions,
15850 ,Expressions}) can be used with the constructs of any supported
15853 The following sections detail to what degree each source language is
15854 supported by @value{GDBN}. These sections are not meant to be language
15855 tutorials or references, but serve only as a reference guide to what the
15856 @value{GDBN} expression parser accepts, and what input and output
15857 formats should look like for different languages. There are many good
15858 books written on each of these languages; please look to these for a
15859 language reference or tutorial.
15862 * C:: C and C@t{++}
15865 * Objective-C:: Objective-C
15866 * OpenCL C:: OpenCL C
15867 * Fortran:: Fortran
15870 * Modula-2:: Modula-2
15875 @subsection C and C@t{++}
15877 @cindex C and C@t{++}
15878 @cindex expressions in C or C@t{++}
15880 Since C and C@t{++} are so closely related, many features of @value{GDBN} apply
15881 to both languages. Whenever this is the case, we discuss those languages
15885 @cindex @code{g++}, @sc{gnu} C@t{++} compiler
15886 @cindex @sc{gnu} C@t{++}
15887 The C@t{++} debugging facilities are jointly implemented by the C@t{++}
15888 compiler and @value{GDBN}. Therefore, to debug your C@t{++} code
15889 effectively, you must compile your C@t{++} programs with a supported
15890 C@t{++} compiler, such as @sc{gnu} @code{g++}, or the HP ANSI C@t{++}
15891 compiler (@code{aCC}).
15894 * C Operators:: C and C@t{++} operators
15895 * C Constants:: C and C@t{++} constants
15896 * C Plus Plus Expressions:: C@t{++} expressions
15897 * C Defaults:: Default settings for C and C@t{++}
15898 * C Checks:: C and C@t{++} type and range checks
15899 * Debugging C:: @value{GDBN} and C
15900 * Debugging C Plus Plus:: @value{GDBN} features for C@t{++}
15901 * Decimal Floating Point:: Numbers in Decimal Floating Point format
15905 @subsubsection C and C@t{++} Operators
15907 @cindex C and C@t{++} operators
15909 Operators must be defined on values of specific types. For instance,
15910 @code{+} is defined on numbers, but not on structures. Operators are
15911 often defined on groups of types.
15913 For the purposes of C and C@t{++}, the following definitions hold:
15918 @emph{Integral types} include @code{int} with any of its storage-class
15919 specifiers; @code{char}; @code{enum}; and, for C@t{++}, @code{bool}.
15922 @emph{Floating-point types} include @code{float}, @code{double}, and
15923 @code{long double} (if supported by the target platform).
15926 @emph{Pointer types} include all types defined as @code{(@var{type} *)}.
15929 @emph{Scalar types} include all of the above.
15934 The following operators are supported. They are listed here
15935 in order of increasing precedence:
15939 The comma or sequencing operator. Expressions in a comma-separated list
15940 are evaluated from left to right, with the result of the entire
15941 expression being the last expression evaluated.
15944 Assignment. The value of an assignment expression is the value
15945 assigned. Defined on scalar types.
15948 Used in an expression of the form @w{@code{@var{a} @var{op}= @var{b}}},
15949 and translated to @w{@code{@var{a} = @var{a op b}}}.
15950 @w{@code{@var{op}=}} and @code{=} have the same precedence. The operator
15951 @var{op} is any one of the operators @code{|}, @code{^}, @code{&},
15952 @code{<<}, @code{>>}, @code{+}, @code{-}, @code{*}, @code{/}, @code{%}.
15955 The ternary operator. @code{@var{a} ? @var{b} : @var{c}} can be thought
15956 of as: if @var{a} then @var{b} else @var{c}. The argument @var{a}
15957 should be of an integral type.
15960 Logical @sc{or}. Defined on integral types.
15963 Logical @sc{and}. Defined on integral types.
15966 Bitwise @sc{or}. Defined on integral types.
15969 Bitwise exclusive-@sc{or}. Defined on integral types.
15972 Bitwise @sc{and}. Defined on integral types.
15975 Equality and inequality. Defined on scalar types. The value of these
15976 expressions is 0 for false and non-zero for true.
15978 @item <@r{, }>@r{, }<=@r{, }>=
15979 Less than, greater than, less than or equal, greater than or equal.
15980 Defined on scalar types. The value of these expressions is 0 for false
15981 and non-zero for true.
15984 left shift, and right shift. Defined on integral types.
15987 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
15990 Addition and subtraction. Defined on integral types, floating-point types and
15993 @item *@r{, }/@r{, }%
15994 Multiplication, division, and modulus. Multiplication and division are
15995 defined on integral and floating-point types. Modulus is defined on
15999 Increment and decrement. When appearing before a variable, the
16000 operation is performed before the variable is used in an expression;
16001 when appearing after it, the variable's value is used before the
16002 operation takes place.
16005 Pointer dereferencing. Defined on pointer types. Same precedence as
16009 Address operator. Defined on variables. Same precedence as @code{++}.
16011 For debugging C@t{++}, @value{GDBN} implements a use of @samp{&} beyond what is
16012 allowed in the C@t{++} language itself: you can use @samp{&(&@var{ref})}
16013 to examine the address
16014 where a C@t{++} reference variable (declared with @samp{&@var{ref}}) is
16018 Negative. Defined on integral and floating-point types. Same
16019 precedence as @code{++}.
16022 Logical negation. Defined on integral types. Same precedence as
16026 Bitwise complement operator. Defined on integral types. Same precedence as
16031 Structure member, and pointer-to-structure member. For convenience,
16032 @value{GDBN} regards the two as equivalent, choosing whether to dereference a
16033 pointer based on the stored type information.
16034 Defined on @code{struct} and @code{union} data.
16037 Dereferences of pointers to members.
16040 Array indexing. @code{@var{a}[@var{i}]} is defined as
16041 @code{*(@var{a}+@var{i})}. Same precedence as @code{->}.
16044 Function parameter list. Same precedence as @code{->}.
16047 C@t{++} scope resolution operator. Defined on @code{struct}, @code{union},
16048 and @code{class} types.
16051 Doubled colons also represent the @value{GDBN} scope operator
16052 (@pxref{Expressions, ,Expressions}). Same precedence as @code{::},
16056 If an operator is redefined in the user code, @value{GDBN} usually
16057 attempts to invoke the redefined version instead of using the operator's
16058 predefined meaning.
16061 @subsubsection C and C@t{++} Constants
16063 @cindex C and C@t{++} constants
16065 @value{GDBN} allows you to express the constants of C and C@t{++} in the
16070 Integer constants are a sequence of digits. Octal constants are
16071 specified by a leading @samp{0} (i.e.@: zero), and hexadecimal constants
16072 by a leading @samp{0x} or @samp{0X}. Constants may also end with a letter
16073 @samp{l}, specifying that the constant should be treated as a
16077 Floating point constants are a sequence of digits, followed by a decimal
16078 point, followed by a sequence of digits, and optionally followed by an
16079 exponent. An exponent is of the form:
16080 @samp{@w{e@r{[[}+@r{]|}-@r{]}@var{nnn}}}, where @var{nnn} is another
16081 sequence of digits. The @samp{+} is optional for positive exponents.
16082 A floating-point constant may also end with a letter @samp{f} or
16083 @samp{F}, specifying that the constant should be treated as being of
16084 the @code{float} (as opposed to the default @code{double}) type; or with
16085 a letter @samp{l} or @samp{L}, which specifies a @code{long double}
16089 Enumerated constants consist of enumerated identifiers, or their
16090 integral equivalents.
16093 Character constants are a single character surrounded by single quotes
16094 (@code{'}), or a number---the ordinal value of the corresponding character
16095 (usually its @sc{ascii} value). Within quotes, the single character may
16096 be represented by a letter or by @dfn{escape sequences}, which are of
16097 the form @samp{\@var{nnn}}, where @var{nnn} is the octal representation
16098 of the character's ordinal value; or of the form @samp{\@var{x}}, where
16099 @samp{@var{x}} is a predefined special character---for example,
16100 @samp{\n} for newline.
16102 Wide character constants can be written by prefixing a character
16103 constant with @samp{L}, as in C. For example, @samp{L'x'} is the wide
16104 form of @samp{x}. The target wide character set is used when
16105 computing the value of this constant (@pxref{Character Sets}).
16108 String constants are a sequence of character constants surrounded by
16109 double quotes (@code{"}). Any valid character constant (as described
16110 above) may appear. Double quotes within the string must be preceded by
16111 a backslash, so for instance @samp{"a\"b'c"} is a string of five
16114 Wide string constants can be written by prefixing a string constant
16115 with @samp{L}, as in C. The target wide character set is used when
16116 computing the value of this constant (@pxref{Character Sets}).
16119 Pointer constants are an integral value. You can also write pointers
16120 to constants using the C operator @samp{&}.
16123 Array constants are comma-separated lists surrounded by braces @samp{@{}
16124 and @samp{@}}; for example, @samp{@{1,2,3@}} is a three-element array of
16125 integers, @samp{@{@{1,2@}, @{3,4@}, @{5,6@}@}} is a three-by-two array,
16126 and @samp{@{&"hi", &"there", &"fred"@}} is a three-element array of pointers.
16129 @node C Plus Plus Expressions
16130 @subsubsection C@t{++} Expressions
16132 @cindex expressions in C@t{++}
16133 @value{GDBN} expression handling can interpret most C@t{++} expressions.
16135 @cindex debugging C@t{++} programs
16136 @cindex C@t{++} compilers
16137 @cindex debug formats and C@t{++}
16138 @cindex @value{NGCC} and C@t{++}
16140 @emph{Warning:} @value{GDBN} can only debug C@t{++} code if you use
16141 the proper compiler and the proper debug format. Currently,
16142 @value{GDBN} works best when debugging C@t{++} code that is compiled
16143 with the most recent version of @value{NGCC} possible. The DWARF
16144 debugging format is preferred; @value{NGCC} defaults to this on most
16145 popular platforms. Other compilers and/or debug formats are likely to
16146 work badly or not at all when using @value{GDBN} to debug C@t{++}
16147 code. @xref{Compilation}.
16152 @cindex member functions
16154 Member function calls are allowed; you can use expressions like
16157 count = aml->GetOriginal(x, y)
16160 @vindex this@r{, inside C@t{++} member functions}
16161 @cindex namespace in C@t{++}
16163 While a member function is active (in the selected stack frame), your
16164 expressions have the same namespace available as the member function;
16165 that is, @value{GDBN} allows implicit references to the class instance
16166 pointer @code{this} following the same rules as C@t{++}. @code{using}
16167 declarations in the current scope are also respected by @value{GDBN}.
16169 @cindex call overloaded functions
16170 @cindex overloaded functions, calling
16171 @cindex type conversions in C@t{++}
16173 You can call overloaded functions; @value{GDBN} resolves the function
16174 call to the right definition, with some restrictions. @value{GDBN} does not
16175 perform overload resolution involving user-defined type conversions,
16176 calls to constructors, or instantiations of templates that do not exist
16177 in the program. It also cannot handle ellipsis argument lists or
16180 It does perform integral conversions and promotions, floating-point
16181 promotions, arithmetic conversions, pointer conversions, conversions of
16182 class objects to base classes, and standard conversions such as those of
16183 functions or arrays to pointers; it requires an exact match on the
16184 number of function arguments.
16186 Overload resolution is always performed, unless you have specified
16187 @code{set overload-resolution off}. @xref{Debugging C Plus Plus,
16188 ,@value{GDBN} Features for C@t{++}}.
16190 You must specify @code{set overload-resolution off} in order to use an
16191 explicit function signature to call an overloaded function, as in
16193 p 'foo(char,int)'('x', 13)
16196 The @value{GDBN} command-completion facility can simplify this;
16197 see @ref{Completion, ,Command Completion}.
16199 @cindex reference declarations
16201 @value{GDBN} understands variables declared as C@t{++} lvalue or rvalue
16202 references; you can use them in expressions just as you do in C@t{++}
16203 source---they are automatically dereferenced.
16205 In the parameter list shown when @value{GDBN} displays a frame, the values of
16206 reference variables are not displayed (unlike other variables); this
16207 avoids clutter, since references are often used for large structures.
16208 The @emph{address} of a reference variable is always shown, unless
16209 you have specified @samp{set print address off}.
16212 @value{GDBN} supports the C@t{++} name resolution operator @code{::}---your
16213 expressions can use it just as expressions in your program do. Since
16214 one scope may be defined in another, you can use @code{::} repeatedly if
16215 necessary, for example in an expression like
16216 @samp{@var{scope1}::@var{scope2}::@var{name}}. @value{GDBN} also allows
16217 resolving name scope by reference to source files, in both C and C@t{++}
16218 debugging (@pxref{Variables, ,Program Variables}).
16221 @value{GDBN} performs argument-dependent lookup, following the C@t{++}
16226 @subsubsection C and C@t{++} Defaults
16228 @cindex C and C@t{++} defaults
16230 If you allow @value{GDBN} to set range checking automatically, it
16231 defaults to @code{off} whenever the working language changes to
16232 C or C@t{++}. This happens regardless of whether you or @value{GDBN}
16233 selects the working language.
16235 If you allow @value{GDBN} to set the language automatically, it
16236 recognizes source files whose names end with @file{.c}, @file{.C}, or
16237 @file{.cc}, etc, and when @value{GDBN} enters code compiled from one of
16238 these files, it sets the working language to C or C@t{++}.
16239 @xref{Automatically, ,Having @value{GDBN} Infer the Source Language},
16240 for further details.
16243 @subsubsection C and C@t{++} Type and Range Checks
16245 @cindex C and C@t{++} checks
16247 By default, when @value{GDBN} parses C or C@t{++} expressions, strict type
16248 checking is used. However, if you turn type checking off, @value{GDBN}
16249 will allow certain non-standard conversions, such as promoting integer
16250 constants to pointers.
16252 Range checking, if turned on, is done on mathematical operations. Array
16253 indices are not checked, since they are often used to index a pointer
16254 that is not itself an array.
16257 @subsubsection @value{GDBN} and C
16259 The @code{set print union} and @code{show print union} commands apply to
16260 the @code{union} type. When set to @samp{on}, any @code{union} that is
16261 inside a @code{struct} or @code{class} is also printed. Otherwise, it
16262 appears as @samp{@{...@}}.
16264 The @code{@@} operator aids in the debugging of dynamic arrays, formed
16265 with pointers and a memory allocation function. @xref{Expressions,
16268 @node Debugging C Plus Plus
16269 @subsubsection @value{GDBN} Features for C@t{++}
16271 @cindex commands for C@t{++}
16273 Some @value{GDBN} commands are particularly useful with C@t{++}, and some are
16274 designed specifically for use with C@t{++}. Here is a summary:
16277 @cindex break in overloaded functions
16278 @item @r{breakpoint menus}
16279 When you want a breakpoint in a function whose name is overloaded,
16280 @value{GDBN} has the capability to display a menu of possible breakpoint
16281 locations to help you specify which function definition you want.
16282 @xref{Ambiguous Expressions,,Ambiguous Expressions}.
16284 @cindex overloading in C@t{++}
16285 @item rbreak @var{regex}
16286 Setting breakpoints using regular expressions is helpful for setting
16287 breakpoints on overloaded functions that are not members of any special
16289 @xref{Set Breaks, ,Setting Breakpoints}.
16291 @cindex C@t{++} exception handling
16293 @itemx catch rethrow
16295 Debug C@t{++} exception handling using these commands. @xref{Set
16296 Catchpoints, , Setting Catchpoints}.
16298 @cindex inheritance
16299 @item ptype @var{typename}
16300 Print inheritance relationships as well as other information for type
16302 @xref{Symbols, ,Examining the Symbol Table}.
16304 @item info vtbl @var{expression}.
16305 The @code{info vtbl} command can be used to display the virtual
16306 method tables of the object computed by @var{expression}. This shows
16307 one entry per virtual table; there may be multiple virtual tables when
16308 multiple inheritance is in use.
16310 @cindex C@t{++} demangling
16311 @item demangle @var{name}
16312 Demangle @var{name}.
16313 @xref{Symbols}, for a more complete description of the @code{demangle} command.
16315 @cindex C@t{++} symbol display
16316 @item set print demangle
16317 @itemx show print demangle
16318 @itemx set print asm-demangle
16319 @itemx show print asm-demangle
16320 Control whether C@t{++} symbols display in their source form, both when
16321 displaying code as C@t{++} source and when displaying disassemblies.
16322 @xref{Print Settings, ,Print Settings}.
16324 @item set print object
16325 @itemx show print object
16326 Choose whether to print derived (actual) or declared types of objects.
16327 @xref{Print Settings, ,Print Settings}.
16329 @item set print vtbl
16330 @itemx show print vtbl
16331 Control the format for printing virtual function tables.
16332 @xref{Print Settings, ,Print Settings}.
16333 (The @code{vtbl} commands do not work on programs compiled with the HP
16334 ANSI C@t{++} compiler (@code{aCC}).)
16336 @kindex set overload-resolution
16337 @cindex overloaded functions, overload resolution
16338 @item set overload-resolution on
16339 Enable overload resolution for C@t{++} expression evaluation. The default
16340 is on. For overloaded functions, @value{GDBN} evaluates the arguments
16341 and searches for a function whose signature matches the argument types,
16342 using the standard C@t{++} conversion rules (see @ref{C Plus Plus
16343 Expressions, ,C@t{++} Expressions}, for details).
16344 If it cannot find a match, it emits a message.
16346 @item set overload-resolution off
16347 Disable overload resolution for C@t{++} expression evaluation. For
16348 overloaded functions that are not class member functions, @value{GDBN}
16349 chooses the first function of the specified name that it finds in the
16350 symbol table, whether or not its arguments are of the correct type. For
16351 overloaded functions that are class member functions, @value{GDBN}
16352 searches for a function whose signature @emph{exactly} matches the
16355 @kindex show overload-resolution
16356 @item show overload-resolution
16357 Show the current setting of overload resolution.
16359 @item @r{Overloaded symbol names}
16360 You can specify a particular definition of an overloaded symbol, using
16361 the same notation that is used to declare such symbols in C@t{++}: type
16362 @code{@var{symbol}(@var{types})} rather than just @var{symbol}. You can
16363 also use the @value{GDBN} command-line word completion facilities to list the
16364 available choices, or to finish the type list for you.
16365 @xref{Completion,, Command Completion}, for details on how to do this.
16367 @item @r{Breakpoints in functions with ABI tags}
16369 The GNU C@t{++} compiler introduced the notion of ABI ``tags'', which
16370 correspond to changes in the ABI of a type, function, or variable that
16371 would not otherwise be reflected in a mangled name. See
16372 @url{https://developers.redhat.com/blog/2015/02/05/gcc5-and-the-c11-abi/}
16375 The ABI tags are visible in C@t{++} demangled names. For example, a
16376 function that returns a std::string:
16379 std::string function(int);
16383 when compiled for the C++11 ABI is marked with the @code{cxx11} ABI
16384 tag, and @value{GDBN} displays the symbol like this:
16387 function[abi:cxx11](int)
16390 You can set a breakpoint on such functions simply as if they had no
16394 (gdb) b function(int)
16395 Breakpoint 2 at 0x40060d: file main.cc, line 10.
16396 (gdb) info breakpoints
16397 Num Type Disp Enb Address What
16398 1 breakpoint keep y 0x0040060d in function[abi:cxx11](int)
16402 On the rare occasion you need to disambiguate between different ABI
16403 tags, you can do so by simply including the ABI tag in the function
16407 (@value{GDBP}) b ambiguous[abi:other_tag](int)
16411 @node Decimal Floating Point
16412 @subsubsection Decimal Floating Point format
16413 @cindex decimal floating point format
16415 @value{GDBN} can examine, set and perform computations with numbers in
16416 decimal floating point format, which in the C language correspond to the
16417 @code{_Decimal32}, @code{_Decimal64} and @code{_Decimal128} types as
16418 specified by the extension to support decimal floating-point arithmetic.
16420 There are two encodings in use, depending on the architecture: BID (Binary
16421 Integer Decimal) for x86 and x86-64, and DPD (Densely Packed Decimal) for
16422 PowerPC and S/390. @value{GDBN} will use the appropriate encoding for the
16425 Because of a limitation in @file{libdecnumber}, the library used by @value{GDBN}
16426 to manipulate decimal floating point numbers, it is not possible to convert
16427 (using a cast, for example) integers wider than 32-bit to decimal float.
16429 In addition, in order to imitate @value{GDBN}'s behaviour with binary floating
16430 point computations, error checking in decimal float operations ignores
16431 underflow, overflow and divide by zero exceptions.
16433 In the PowerPC architecture, @value{GDBN} provides a set of pseudo-registers
16434 to inspect @code{_Decimal128} values stored in floating point registers.
16435 See @ref{PowerPC,,PowerPC} for more details.
16441 @value{GDBN} can be used to debug programs written in D and compiled with
16442 GDC, LDC or DMD compilers. Currently @value{GDBN} supports only one D
16443 specific feature --- dynamic arrays.
16448 @cindex Go (programming language)
16449 @value{GDBN} can be used to debug programs written in Go and compiled with
16450 @file{gccgo} or @file{6g} compilers.
16452 Here is a summary of the Go-specific features and restrictions:
16455 @cindex current Go package
16456 @item The current Go package
16457 The name of the current package does not need to be specified when
16458 specifying global variables and functions.
16460 For example, given the program:
16464 var myglob = "Shall we?"
16470 When stopped inside @code{main} either of these work:
16474 (gdb) p main.myglob
16477 @cindex builtin Go types
16478 @item Builtin Go types
16479 The @code{string} type is recognized by @value{GDBN} and is printed
16482 @cindex builtin Go functions
16483 @item Builtin Go functions
16484 The @value{GDBN} expression parser recognizes the @code{unsafe.Sizeof}
16485 function and handles it internally.
16487 @cindex restrictions on Go expressions
16488 @item Restrictions on Go expressions
16489 All Go operators are supported except @code{&^}.
16490 The Go @code{_} ``blank identifier'' is not supported.
16491 Automatic dereferencing of pointers is not supported.
16495 @subsection Objective-C
16497 @cindex Objective-C
16498 This section provides information about some commands and command
16499 options that are useful for debugging Objective-C code. See also
16500 @ref{Symbols, info classes}, and @ref{Symbols, info selectors}, for a
16501 few more commands specific to Objective-C support.
16504 * Method Names in Commands::
16505 * The Print Command with Objective-C::
16508 @node Method Names in Commands
16509 @subsubsection Method Names in Commands
16511 The following commands have been extended to accept Objective-C method
16512 names as line specifications:
16514 @kindex clear@r{, and Objective-C}
16515 @kindex break@r{, and Objective-C}
16516 @kindex info line@r{, and Objective-C}
16517 @kindex jump@r{, and Objective-C}
16518 @kindex list@r{, and Objective-C}
16522 @item @code{info line}
16527 A fully qualified Objective-C method name is specified as
16530 -[@var{Class} @var{methodName}]
16533 where the minus sign is used to indicate an instance method and a
16534 plus sign (not shown) is used to indicate a class method. The class
16535 name @var{Class} and method name @var{methodName} are enclosed in
16536 brackets, similar to the way messages are specified in Objective-C
16537 source code. For example, to set a breakpoint at the @code{create}
16538 instance method of class @code{Fruit} in the program currently being
16542 break -[Fruit create]
16545 To list ten program lines around the @code{initialize} class method,
16549 list +[NSText initialize]
16552 In the current version of @value{GDBN}, the plus or minus sign is
16553 required. In future versions of @value{GDBN}, the plus or minus
16554 sign will be optional, but you can use it to narrow the search. It
16555 is also possible to specify just a method name:
16561 You must specify the complete method name, including any colons. If
16562 your program's source files contain more than one @code{create} method,
16563 you'll be presented with a numbered list of classes that implement that
16564 method. Indicate your choice by number, or type @samp{0} to exit if
16567 As another example, to clear a breakpoint established at the
16568 @code{makeKeyAndOrderFront:} method of the @code{NSWindow} class, enter:
16571 clear -[NSWindow makeKeyAndOrderFront:]
16574 @node The Print Command with Objective-C
16575 @subsubsection The Print Command With Objective-C
16576 @cindex Objective-C, print objects
16577 @kindex print-object
16578 @kindex po @r{(@code{print-object})}
16580 The print command has also been extended to accept methods. For example:
16583 print -[@var{object} hash]
16586 @cindex print an Objective-C object description
16587 @cindex @code{_NSPrintForDebugger}, and printing Objective-C objects
16589 will tell @value{GDBN} to send the @code{hash} message to @var{object}
16590 and print the result. Also, an additional command has been added,
16591 @code{print-object} or @code{po} for short, which is meant to print
16592 the description of an object. However, this command may only work
16593 with certain Objective-C libraries that have a particular hook
16594 function, @code{_NSPrintForDebugger}, defined.
16597 @subsection OpenCL C
16600 This section provides information about @value{GDBN}s OpenCL C support.
16603 * OpenCL C Datatypes::
16604 * OpenCL C Expressions::
16605 * OpenCL C Operators::
16608 @node OpenCL C Datatypes
16609 @subsubsection OpenCL C Datatypes
16611 @cindex OpenCL C Datatypes
16612 @value{GDBN} supports the builtin scalar and vector datatypes specified
16613 by OpenCL 1.1. In addition the half- and double-precision floating point
16614 data types of the @code{cl_khr_fp16} and @code{cl_khr_fp64} OpenCL
16615 extensions are also known to @value{GDBN}.
16617 @node OpenCL C Expressions
16618 @subsubsection OpenCL C Expressions
16620 @cindex OpenCL C Expressions
16621 @value{GDBN} supports accesses to vector components including the access as
16622 lvalue where possible. Since OpenCL C is based on C99 most C expressions
16623 supported by @value{GDBN} can be used as well.
16625 @node OpenCL C Operators
16626 @subsubsection OpenCL C Operators
16628 @cindex OpenCL C Operators
16629 @value{GDBN} supports the operators specified by OpenCL 1.1 for scalar and
16633 @subsection Fortran
16634 @cindex Fortran-specific support in @value{GDBN}
16636 @value{GDBN} can be used to debug programs written in Fortran, but it
16637 currently supports only the features of Fortran 77 language.
16639 @cindex trailing underscore, in Fortran symbols
16640 Some Fortran compilers (@sc{gnu} Fortran 77 and Fortran 95 compilers
16641 among them) append an underscore to the names of variables and
16642 functions. When you debug programs compiled by those compilers, you
16643 will need to refer to variables and functions with a trailing
16647 * Fortran Operators:: Fortran operators and expressions
16648 * Fortran Defaults:: Default settings for Fortran
16649 * Special Fortran Commands:: Special @value{GDBN} commands for Fortran
16652 @node Fortran Operators
16653 @subsubsection Fortran Operators and Expressions
16655 @cindex Fortran operators and expressions
16657 Operators must be defined on values of specific types. For instance,
16658 @code{+} is defined on numbers, but not on characters or other non-
16659 arithmetic types. Operators are often defined on groups of types.
16663 The exponentiation operator. It raises the first operand to the power
16667 The range operator. Normally used in the form of array(low:high) to
16668 represent a section of array.
16671 The access component operator. Normally used to access elements in derived
16672 types. Also suitable for unions. As unions aren't part of regular Fortran,
16673 this can only happen when accessing a register that uses a gdbarch-defined
16676 The scope operator. Normally used to access variables in modules or
16677 to set breakpoints on subroutines nested in modules or in other
16678 subroutines (internal subroutines).
16681 @node Fortran Defaults
16682 @subsubsection Fortran Defaults
16684 @cindex Fortran Defaults
16686 Fortran symbols are usually case-insensitive, so @value{GDBN} by
16687 default uses case-insensitive matches for Fortran symbols. You can
16688 change that with the @samp{set case-insensitive} command, see
16689 @ref{Symbols}, for the details.
16691 @node Special Fortran Commands
16692 @subsubsection Special Fortran Commands
16694 @cindex Special Fortran commands
16696 @value{GDBN} has some commands to support Fortran-specific features,
16697 such as displaying common blocks.
16700 @cindex @code{COMMON} blocks, Fortran
16701 @kindex info common
16702 @item info common @r{[}@var{common-name}@r{]}
16703 This command prints the values contained in the Fortran @code{COMMON}
16704 block whose name is @var{common-name}. With no argument, the names of
16705 all @code{COMMON} blocks visible at the current program location are
16712 @cindex Pascal support in @value{GDBN}, limitations
16713 Debugging Pascal programs which use sets, subranges, file variables, or
16714 nested functions does not currently work. @value{GDBN} does not support
16715 entering expressions, printing values, or similar features using Pascal
16718 The Pascal-specific command @code{set print pascal_static-members}
16719 controls whether static members of Pascal objects are displayed.
16720 @xref{Print Settings, pascal_static-members}.
16725 @value{GDBN} supports the @url{https://www.rust-lang.org/, Rust
16726 Programming Language}. Type- and value-printing, and expression
16727 parsing, are reasonably complete. However, there are a few
16728 peculiarities and holes to be aware of.
16732 Linespecs (@pxref{Specify Location}) are never relative to the current
16733 crate. Instead, they act as if there were a global namespace of
16734 crates, somewhat similar to the way @code{extern crate} behaves.
16736 That is, if @value{GDBN} is stopped at a breakpoint in a function in
16737 crate @samp{A}, module @samp{B}, then @code{break B::f} will attempt
16738 to set a breakpoint in a function named @samp{f} in a crate named
16741 As a consequence of this approach, linespecs also cannot refer to
16742 items using @samp{self::} or @samp{super::}.
16745 Because @value{GDBN} implements Rust name-lookup semantics in
16746 expressions, it will sometimes prepend the current crate to a name.
16747 For example, if @value{GDBN} is stopped at a breakpoint in the crate
16748 @samp{K}, then @code{print ::x::y} will try to find the symbol
16751 However, since it is useful to be able to refer to other crates when
16752 debugging, @value{GDBN} provides the @code{extern} extension to
16753 circumvent this. To use the extension, just put @code{extern} before
16754 a path expression to refer to the otherwise unavailable ``global''
16757 In the above example, if you wanted to refer to the symbol @samp{y} in
16758 the crate @samp{x}, you would use @code{print extern x::y}.
16761 The Rust expression evaluator does not support ``statement-like''
16762 expressions such as @code{if} or @code{match}, or lambda expressions.
16765 Tuple expressions are not implemented.
16768 The Rust expression evaluator does not currently implement the
16769 @code{Drop} trait. Objects that may be created by the evaluator will
16770 never be destroyed.
16773 @value{GDBN} does not implement type inference for generics. In order
16774 to call generic functions or otherwise refer to generic items, you
16775 will have to specify the type parameters manually.
16778 @value{GDBN} currently uses the C@t{++} demangler for Rust. In most
16779 cases this does not cause any problems. However, in an expression
16780 context, completing a generic function name will give syntactically
16781 invalid results. This happens because Rust requires the @samp{::}
16782 operator between the function name and its generic arguments. For
16783 example, @value{GDBN} might provide a completion like
16784 @code{crate::f<u32>}, where the parser would require
16785 @code{crate::f::<u32>}.
16788 As of this writing, the Rust compiler (version 1.8) has a few holes in
16789 the debugging information it generates. These holes prevent certain
16790 features from being implemented by @value{GDBN}:
16794 Method calls cannot be made via traits.
16797 Operator overloading is not implemented.
16800 When debugging in a monomorphized function, you cannot use the generic
16804 The type @code{Self} is not available.
16807 @code{use} statements are not available, so some names may not be
16808 available in the crate.
16813 @subsection Modula-2
16815 @cindex Modula-2, @value{GDBN} support
16817 The extensions made to @value{GDBN} to support Modula-2 only support
16818 output from the @sc{gnu} Modula-2 compiler (which is currently being
16819 developed). Other Modula-2 compilers are not currently supported, and
16820 attempting to debug executables produced by them is most likely
16821 to give an error as @value{GDBN} reads in the executable's symbol
16824 @cindex expressions in Modula-2
16826 * M2 Operators:: Built-in operators
16827 * Built-In Func/Proc:: Built-in functions and procedures
16828 * M2 Constants:: Modula-2 constants
16829 * M2 Types:: Modula-2 types
16830 * M2 Defaults:: Default settings for Modula-2
16831 * Deviations:: Deviations from standard Modula-2
16832 * M2 Checks:: Modula-2 type and range checks
16833 * M2 Scope:: The scope operators @code{::} and @code{.}
16834 * GDB/M2:: @value{GDBN} and Modula-2
16838 @subsubsection Operators
16839 @cindex Modula-2 operators
16841 Operators must be defined on values of specific types. For instance,
16842 @code{+} is defined on numbers, but not on structures. Operators are
16843 often defined on groups of types. For the purposes of Modula-2, the
16844 following definitions hold:
16849 @emph{Integral types} consist of @code{INTEGER}, @code{CARDINAL}, and
16853 @emph{Character types} consist of @code{CHAR} and its subranges.
16856 @emph{Floating-point types} consist of @code{REAL}.
16859 @emph{Pointer types} consist of anything declared as @code{POINTER TO
16863 @emph{Scalar types} consist of all of the above.
16866 @emph{Set types} consist of @code{SET} and @code{BITSET} types.
16869 @emph{Boolean types} consist of @code{BOOLEAN}.
16873 The following operators are supported, and appear in order of
16874 increasing precedence:
16878 Function argument or array index separator.
16881 Assignment. The value of @var{var} @code{:=} @var{value} is
16885 Less than, greater than on integral, floating-point, or enumerated
16889 Less than or equal to, greater than or equal to
16890 on integral, floating-point and enumerated types, or set inclusion on
16891 set types. Same precedence as @code{<}.
16893 @item =@r{, }<>@r{, }#
16894 Equality and two ways of expressing inequality, valid on scalar types.
16895 Same precedence as @code{<}. In @value{GDBN} scripts, only @code{<>} is
16896 available for inequality, since @code{#} conflicts with the script
16900 Set membership. Defined on set types and the types of their members.
16901 Same precedence as @code{<}.
16904 Boolean disjunction. Defined on boolean types.
16907 Boolean conjunction. Defined on boolean types.
16910 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
16913 Addition and subtraction on integral and floating-point types, or union
16914 and difference on set types.
16917 Multiplication on integral and floating-point types, or set intersection
16921 Division on floating-point types, or symmetric set difference on set
16922 types. Same precedence as @code{*}.
16925 Integer division and remainder. Defined on integral types. Same
16926 precedence as @code{*}.
16929 Negative. Defined on @code{INTEGER} and @code{REAL} data.
16932 Pointer dereferencing. Defined on pointer types.
16935 Boolean negation. Defined on boolean types. Same precedence as
16939 @code{RECORD} field selector. Defined on @code{RECORD} data. Same
16940 precedence as @code{^}.
16943 Array indexing. Defined on @code{ARRAY} data. Same precedence as @code{^}.
16946 Procedure argument list. Defined on @code{PROCEDURE} objects. Same precedence
16950 @value{GDBN} and Modula-2 scope operators.
16954 @emph{Warning:} Set expressions and their operations are not yet supported, so @value{GDBN}
16955 treats the use of the operator @code{IN}, or the use of operators
16956 @code{+}, @code{-}, @code{*}, @code{/}, @code{=}, , @code{<>}, @code{#},
16957 @code{<=}, and @code{>=} on sets as an error.
16961 @node Built-In Func/Proc
16962 @subsubsection Built-in Functions and Procedures
16963 @cindex Modula-2 built-ins
16965 Modula-2 also makes available several built-in procedures and functions.
16966 In describing these, the following metavariables are used:
16971 represents an @code{ARRAY} variable.
16974 represents a @code{CHAR} constant or variable.
16977 represents a variable or constant of integral type.
16980 represents an identifier that belongs to a set. Generally used in the
16981 same function with the metavariable @var{s}. The type of @var{s} should
16982 be @code{SET OF @var{mtype}} (where @var{mtype} is the type of @var{m}).
16985 represents a variable or constant of integral or floating-point type.
16988 represents a variable or constant of floating-point type.
16994 represents a variable.
16997 represents a variable or constant of one of many types. See the
16998 explanation of the function for details.
17001 All Modula-2 built-in procedures also return a result, described below.
17005 Returns the absolute value of @var{n}.
17008 If @var{c} is a lower case letter, it returns its upper case
17009 equivalent, otherwise it returns its argument.
17012 Returns the character whose ordinal value is @var{i}.
17015 Decrements the value in the variable @var{v} by one. Returns the new value.
17017 @item DEC(@var{v},@var{i})
17018 Decrements the value in the variable @var{v} by @var{i}. Returns the
17021 @item EXCL(@var{m},@var{s})
17022 Removes the element @var{m} from the set @var{s}. Returns the new
17025 @item FLOAT(@var{i})
17026 Returns the floating point equivalent of the integer @var{i}.
17028 @item HIGH(@var{a})
17029 Returns the index of the last member of @var{a}.
17032 Increments the value in the variable @var{v} by one. Returns the new value.
17034 @item INC(@var{v},@var{i})
17035 Increments the value in the variable @var{v} by @var{i}. Returns the
17038 @item INCL(@var{m},@var{s})
17039 Adds the element @var{m} to the set @var{s} if it is not already
17040 there. Returns the new set.
17043 Returns the maximum value of the type @var{t}.
17046 Returns the minimum value of the type @var{t}.
17049 Returns boolean TRUE if @var{i} is an odd number.
17052 Returns the ordinal value of its argument. For example, the ordinal
17053 value of a character is its @sc{ascii} value (on machines supporting
17054 the @sc{ascii} character set). The argument @var{x} must be of an
17055 ordered type, which include integral, character and enumerated types.
17057 @item SIZE(@var{x})
17058 Returns the size of its argument. The argument @var{x} can be a
17059 variable or a type.
17061 @item TRUNC(@var{r})
17062 Returns the integral part of @var{r}.
17064 @item TSIZE(@var{x})
17065 Returns the size of its argument. The argument @var{x} can be a
17066 variable or a type.
17068 @item VAL(@var{t},@var{i})
17069 Returns the member of the type @var{t} whose ordinal value is @var{i}.
17073 @emph{Warning:} Sets and their operations are not yet supported, so
17074 @value{GDBN} treats the use of procedures @code{INCL} and @code{EXCL} as
17078 @cindex Modula-2 constants
17080 @subsubsection Constants
17082 @value{GDBN} allows you to express the constants of Modula-2 in the following
17088 Integer constants are simply a sequence of digits. When used in an
17089 expression, a constant is interpreted to be type-compatible with the
17090 rest of the expression. Hexadecimal integers are specified by a
17091 trailing @samp{H}, and octal integers by a trailing @samp{B}.
17094 Floating point constants appear as a sequence of digits, followed by a
17095 decimal point and another sequence of digits. An optional exponent can
17096 then be specified, in the form @samp{E@r{[}+@r{|}-@r{]}@var{nnn}}, where
17097 @samp{@r{[}+@r{|}-@r{]}@var{nnn}} is the desired exponent. All of the
17098 digits of the floating point constant must be valid decimal (base 10)
17102 Character constants consist of a single character enclosed by a pair of
17103 like quotes, either single (@code{'}) or double (@code{"}). They may
17104 also be expressed by their ordinal value (their @sc{ascii} value, usually)
17105 followed by a @samp{C}.
17108 String constants consist of a sequence of characters enclosed by a
17109 pair of like quotes, either single (@code{'}) or double (@code{"}).
17110 Escape sequences in the style of C are also allowed. @xref{C
17111 Constants, ,C and C@t{++} Constants}, for a brief explanation of escape
17115 Enumerated constants consist of an enumerated identifier.
17118 Boolean constants consist of the identifiers @code{TRUE} and
17122 Pointer constants consist of integral values only.
17125 Set constants are not yet supported.
17129 @subsubsection Modula-2 Types
17130 @cindex Modula-2 types
17132 Currently @value{GDBN} can print the following data types in Modula-2
17133 syntax: array types, record types, set types, pointer types, procedure
17134 types, enumerated types, subrange types and base types. You can also
17135 print the contents of variables declared using these type.
17136 This section gives a number of simple source code examples together with
17137 sample @value{GDBN} sessions.
17139 The first example contains the following section of code:
17148 and you can request @value{GDBN} to interrogate the type and value of
17149 @code{r} and @code{s}.
17152 (@value{GDBP}) print s
17154 (@value{GDBP}) ptype s
17156 (@value{GDBP}) print r
17158 (@value{GDBP}) ptype r
17163 Likewise if your source code declares @code{s} as:
17167 s: SET ['A'..'Z'] ;
17171 then you may query the type of @code{s} by:
17174 (@value{GDBP}) ptype s
17175 type = SET ['A'..'Z']
17179 Note that at present you cannot interactively manipulate set
17180 expressions using the debugger.
17182 The following example shows how you might declare an array in Modula-2
17183 and how you can interact with @value{GDBN} to print its type and contents:
17187 s: ARRAY [-10..10] OF CHAR ;
17191 (@value{GDBP}) ptype s
17192 ARRAY [-10..10] OF CHAR
17195 Note that the array handling is not yet complete and although the type
17196 is printed correctly, expression handling still assumes that all
17197 arrays have a lower bound of zero and not @code{-10} as in the example
17200 Here are some more type related Modula-2 examples:
17204 colour = (blue, red, yellow, green) ;
17205 t = [blue..yellow] ;
17213 The @value{GDBN} interaction shows how you can query the data type
17214 and value of a variable.
17217 (@value{GDBP}) print s
17219 (@value{GDBP}) ptype t
17220 type = [blue..yellow]
17224 In this example a Modula-2 array is declared and its contents
17225 displayed. Observe that the contents are written in the same way as
17226 their @code{C} counterparts.
17230 s: ARRAY [1..5] OF CARDINAL ;
17236 (@value{GDBP}) print s
17237 $1 = @{1, 0, 0, 0, 0@}
17238 (@value{GDBP}) ptype s
17239 type = ARRAY [1..5] OF CARDINAL
17242 The Modula-2 language interface to @value{GDBN} also understands
17243 pointer types as shown in this example:
17247 s: POINTER TO ARRAY [1..5] OF CARDINAL ;
17254 and you can request that @value{GDBN} describes the type of @code{s}.
17257 (@value{GDBP}) ptype s
17258 type = POINTER TO ARRAY [1..5] OF CARDINAL
17261 @value{GDBN} handles compound types as we can see in this example.
17262 Here we combine array types, record types, pointer types and subrange
17273 myarray = ARRAY myrange OF CARDINAL ;
17274 myrange = [-2..2] ;
17276 s: POINTER TO ARRAY myrange OF foo ;
17280 and you can ask @value{GDBN} to describe the type of @code{s} as shown
17284 (@value{GDBP}) ptype s
17285 type = POINTER TO ARRAY [-2..2] OF foo = RECORD
17288 f3 : ARRAY [-2..2] OF CARDINAL;
17293 @subsubsection Modula-2 Defaults
17294 @cindex Modula-2 defaults
17296 If type and range checking are set automatically by @value{GDBN}, they
17297 both default to @code{on} whenever the working language changes to
17298 Modula-2. This happens regardless of whether you or @value{GDBN}
17299 selected the working language.
17301 If you allow @value{GDBN} to set the language automatically, then entering
17302 code compiled from a file whose name ends with @file{.mod} sets the
17303 working language to Modula-2. @xref{Automatically, ,Having @value{GDBN}
17304 Infer the Source Language}, for further details.
17307 @subsubsection Deviations from Standard Modula-2
17308 @cindex Modula-2, deviations from
17310 A few changes have been made to make Modula-2 programs easier to debug.
17311 This is done primarily via loosening its type strictness:
17315 Unlike in standard Modula-2, pointer constants can be formed by
17316 integers. This allows you to modify pointer variables during
17317 debugging. (In standard Modula-2, the actual address contained in a
17318 pointer variable is hidden from you; it can only be modified
17319 through direct assignment to another pointer variable or expression that
17320 returned a pointer.)
17323 C escape sequences can be used in strings and characters to represent
17324 non-printable characters. @value{GDBN} prints out strings with these
17325 escape sequences embedded. Single non-printable characters are
17326 printed using the @samp{CHR(@var{nnn})} format.
17329 The assignment operator (@code{:=}) returns the value of its right-hand
17333 All built-in procedures both modify @emph{and} return their argument.
17337 @subsubsection Modula-2 Type and Range Checks
17338 @cindex Modula-2 checks
17341 @emph{Warning:} in this release, @value{GDBN} does not yet perform type or
17344 @c FIXME remove warning when type/range checks added
17346 @value{GDBN} considers two Modula-2 variables type equivalent if:
17350 They are of types that have been declared equivalent via a @code{TYPE
17351 @var{t1} = @var{t2}} statement
17354 They have been declared on the same line. (Note: This is true of the
17355 @sc{gnu} Modula-2 compiler, but it may not be true of other compilers.)
17358 As long as type checking is enabled, any attempt to combine variables
17359 whose types are not equivalent is an error.
17361 Range checking is done on all mathematical operations, assignment, array
17362 index bounds, and all built-in functions and procedures.
17365 @subsubsection The Scope Operators @code{::} and @code{.}
17367 @cindex @code{.}, Modula-2 scope operator
17368 @cindex colon, doubled as scope operator
17370 @vindex colon-colon@r{, in Modula-2}
17371 @c Info cannot handle :: but TeX can.
17374 @vindex ::@r{, in Modula-2}
17377 There are a few subtle differences between the Modula-2 scope operator
17378 (@code{.}) and the @value{GDBN} scope operator (@code{::}). The two have
17383 @var{module} . @var{id}
17384 @var{scope} :: @var{id}
17388 where @var{scope} is the name of a module or a procedure,
17389 @var{module} the name of a module, and @var{id} is any declared
17390 identifier within your program, except another module.
17392 Using the @code{::} operator makes @value{GDBN} search the scope
17393 specified by @var{scope} for the identifier @var{id}. If it is not
17394 found in the specified scope, then @value{GDBN} searches all scopes
17395 enclosing the one specified by @var{scope}.
17397 Using the @code{.} operator makes @value{GDBN} search the current scope for
17398 the identifier specified by @var{id} that was imported from the
17399 definition module specified by @var{module}. With this operator, it is
17400 an error if the identifier @var{id} was not imported from definition
17401 module @var{module}, or if @var{id} is not an identifier in
17405 @subsubsection @value{GDBN} and Modula-2
17407 Some @value{GDBN} commands have little use when debugging Modula-2 programs.
17408 Five subcommands of @code{set print} and @code{show print} apply
17409 specifically to C and C@t{++}: @samp{vtbl}, @samp{demangle},
17410 @samp{asm-demangle}, @samp{object}, and @samp{union}. The first four
17411 apply to C@t{++}, and the last to the C @code{union} type, which has no direct
17412 analogue in Modula-2.
17414 The @code{@@} operator (@pxref{Expressions, ,Expressions}), while available
17415 with any language, is not useful with Modula-2. Its
17416 intent is to aid the debugging of @dfn{dynamic arrays}, which cannot be
17417 created in Modula-2 as they can in C or C@t{++}. However, because an
17418 address can be specified by an integral constant, the construct
17419 @samp{@{@var{type}@}@var{adrexp}} is still useful.
17421 @cindex @code{#} in Modula-2
17422 In @value{GDBN} scripts, the Modula-2 inequality operator @code{#} is
17423 interpreted as the beginning of a comment. Use @code{<>} instead.
17429 The extensions made to @value{GDBN} for Ada only support
17430 output from the @sc{gnu} Ada (GNAT) compiler.
17431 Other Ada compilers are not currently supported, and
17432 attempting to debug executables produced by them is most likely
17436 @cindex expressions in Ada
17438 * Ada Mode Intro:: General remarks on the Ada syntax
17439 and semantics supported by Ada mode
17441 * Omissions from Ada:: Restrictions on the Ada expression syntax.
17442 * Additions to Ada:: Extensions of the Ada expression syntax.
17443 * Overloading support for Ada:: Support for expressions involving overloaded
17445 * Stopping Before Main Program:: Debugging the program during elaboration.
17446 * Ada Exceptions:: Ada Exceptions
17447 * Ada Tasks:: Listing and setting breakpoints in tasks.
17448 * Ada Tasks and Core Files:: Tasking Support when Debugging Core Files
17449 * Ravenscar Profile:: Tasking Support when using the Ravenscar
17451 * Ada Settings:: New settable GDB parameters for Ada.
17452 * Ada Glitches:: Known peculiarities of Ada mode.
17455 @node Ada Mode Intro
17456 @subsubsection Introduction
17457 @cindex Ada mode, general
17459 The Ada mode of @value{GDBN} supports a fairly large subset of Ada expression
17460 syntax, with some extensions.
17461 The philosophy behind the design of this subset is
17465 That @value{GDBN} should provide basic literals and access to operations for
17466 arithmetic, dereferencing, field selection, indexing, and subprogram calls,
17467 leaving more sophisticated computations to subprograms written into the
17468 program (which therefore may be called from @value{GDBN}).
17471 That type safety and strict adherence to Ada language restrictions
17472 are not particularly important to the @value{GDBN} user.
17475 That brevity is important to the @value{GDBN} user.
17478 Thus, for brevity, the debugger acts as if all names declared in
17479 user-written packages are directly visible, even if they are not visible
17480 according to Ada rules, thus making it unnecessary to fully qualify most
17481 names with their packages, regardless of context. Where this causes
17482 ambiguity, @value{GDBN} asks the user's intent.
17484 The debugger will start in Ada mode if it detects an Ada main program.
17485 As for other languages, it will enter Ada mode when stopped in a program that
17486 was translated from an Ada source file.
17488 While in Ada mode, you may use `@t{--}' for comments. This is useful
17489 mostly for documenting command files. The standard @value{GDBN} comment
17490 (@samp{#}) still works at the beginning of a line in Ada mode, but not in the
17491 middle (to allow based literals).
17493 @node Omissions from Ada
17494 @subsubsection Omissions from Ada
17495 @cindex Ada, omissions from
17497 Here are the notable omissions from the subset:
17501 Only a subset of the attributes are supported:
17505 @t{'First}, @t{'Last}, and @t{'Length}
17506 on array objects (not on types and subtypes).
17509 @t{'Min} and @t{'Max}.
17512 @t{'Pos} and @t{'Val}.
17518 @t{'Range} on array objects (not subtypes), but only as the right
17519 operand of the membership (@code{in}) operator.
17522 @t{'Access}, @t{'Unchecked_Access}, and
17523 @t{'Unrestricted_Access} (a GNAT extension).
17531 @code{Characters.Latin_1} are not available and
17532 concatenation is not implemented. Thus, escape characters in strings are
17533 not currently available.
17536 Equality tests (@samp{=} and @samp{/=}) on arrays test for bitwise
17537 equality of representations. They will generally work correctly
17538 for strings and arrays whose elements have integer or enumeration types.
17539 They may not work correctly for arrays whose element
17540 types have user-defined equality, for arrays of real values
17541 (in particular, IEEE-conformant floating point, because of negative
17542 zeroes and NaNs), and for arrays whose elements contain unused bits with
17543 indeterminate values.
17546 The other component-by-component array operations (@code{and}, @code{or},
17547 @code{xor}, @code{not}, and relational tests other than equality)
17548 are not implemented.
17551 @cindex array aggregates (Ada)
17552 @cindex record aggregates (Ada)
17553 @cindex aggregates (Ada)
17554 There is limited support for array and record aggregates. They are
17555 permitted only on the right sides of assignments, as in these examples:
17558 (@value{GDBP}) set An_Array := (1, 2, 3, 4, 5, 6)
17559 (@value{GDBP}) set An_Array := (1, others => 0)
17560 (@value{GDBP}) set An_Array := (0|4 => 1, 1..3 => 2, 5 => 6)
17561 (@value{GDBP}) set A_2D_Array := ((1, 2, 3), (4, 5, 6), (7, 8, 9))
17562 (@value{GDBP}) set A_Record := (1, "Peter", True);
17563 (@value{GDBP}) set A_Record := (Name => "Peter", Id => 1, Alive => True)
17567 discriminant's value by assigning an aggregate has an
17568 undefined effect if that discriminant is used within the record.
17569 However, you can first modify discriminants by directly assigning to
17570 them (which normally would not be allowed in Ada), and then performing an
17571 aggregate assignment. For example, given a variable @code{A_Rec}
17572 declared to have a type such as:
17575 type Rec (Len : Small_Integer := 0) is record
17577 Vals : IntArray (1 .. Len);
17581 you can assign a value with a different size of @code{Vals} with two
17585 (@value{GDBP}) set A_Rec.Len := 4
17586 (@value{GDBP}) set A_Rec := (Id => 42, Vals => (1, 2, 3, 4))
17589 As this example also illustrates, @value{GDBN} is very loose about the usual
17590 rules concerning aggregates. You may leave out some of the
17591 components of an array or record aggregate (such as the @code{Len}
17592 component in the assignment to @code{A_Rec} above); they will retain their
17593 original values upon assignment. You may freely use dynamic values as
17594 indices in component associations. You may even use overlapping or
17595 redundant component associations, although which component values are
17596 assigned in such cases is not defined.
17599 Calls to dispatching subprograms are not implemented.
17602 The overloading algorithm is much more limited (i.e., less selective)
17603 than that of real Ada. It makes only limited use of the context in
17604 which a subexpression appears to resolve its meaning, and it is much
17605 looser in its rules for allowing type matches. As a result, some
17606 function calls will be ambiguous, and the user will be asked to choose
17607 the proper resolution.
17610 The @code{new} operator is not implemented.
17613 Entry calls are not implemented.
17616 Aside from printing, arithmetic operations on the native VAX floating-point
17617 formats are not supported.
17620 It is not possible to slice a packed array.
17623 The names @code{True} and @code{False}, when not part of a qualified name,
17624 are interpreted as if implicitly prefixed by @code{Standard}, regardless of
17626 Should your program
17627 redefine these names in a package or procedure (at best a dubious practice),
17628 you will have to use fully qualified names to access their new definitions.
17631 @node Additions to Ada
17632 @subsubsection Additions to Ada
17633 @cindex Ada, deviations from
17635 As it does for other languages, @value{GDBN} makes certain generic
17636 extensions to Ada (@pxref{Expressions}):
17640 If the expression @var{E} is a variable residing in memory (typically
17641 a local variable or array element) and @var{N} is a positive integer,
17642 then @code{@var{E}@@@var{N}} displays the values of @var{E} and the
17643 @var{N}-1 adjacent variables following it in memory as an array. In
17644 Ada, this operator is generally not necessary, since its prime use is
17645 in displaying parts of an array, and slicing will usually do this in
17646 Ada. However, there are occasional uses when debugging programs in
17647 which certain debugging information has been optimized away.
17650 @code{@var{B}::@var{var}} means ``the variable named @var{var} that
17651 appears in function or file @var{B}.'' When @var{B} is a file name,
17652 you must typically surround it in single quotes.
17655 The expression @code{@{@var{type}@} @var{addr}} means ``the variable of type
17656 @var{type} that appears at address @var{addr}.''
17659 A name starting with @samp{$} is a convenience variable
17660 (@pxref{Convenience Vars}) or a machine register (@pxref{Registers}).
17663 In addition, @value{GDBN} provides a few other shortcuts and outright
17664 additions specific to Ada:
17668 The assignment statement is allowed as an expression, returning
17669 its right-hand operand as its value. Thus, you may enter
17672 (@value{GDBP}) set x := y + 3
17673 (@value{GDBP}) print A(tmp := y + 1)
17677 The semicolon is allowed as an ``operator,'' returning as its value
17678 the value of its right-hand operand.
17679 This allows, for example,
17680 complex conditional breaks:
17683 (@value{GDBP}) break f
17684 (@value{GDBP}) condition 1 (report(i); k += 1; A(k) > 100)
17688 Rather than use catenation and symbolic character names to introduce special
17689 characters into strings, one may instead use a special bracket notation,
17690 which is also used to print strings. A sequence of characters of the form
17691 @samp{["@var{XX}"]} within a string or character literal denotes the
17692 (single) character whose numeric encoding is @var{XX} in hexadecimal. The
17693 sequence of characters @samp{["""]} also denotes a single quotation mark
17694 in strings. For example,
17696 "One line.["0a"]Next line.["0a"]"
17699 contains an ASCII newline character (@code{Ada.Characters.Latin_1.LF})
17703 The subtype used as a prefix for the attributes @t{'Pos}, @t{'Min}, and
17704 @t{'Max} is optional (and is ignored in any case). For example, it is valid
17708 (@value{GDBP}) print 'max(x, y)
17712 When printing arrays, @value{GDBN} uses positional notation when the
17713 array has a lower bound of 1, and uses a modified named notation otherwise.
17714 For example, a one-dimensional array of three integers with a lower bound
17715 of 3 might print as
17722 That is, in contrast to valid Ada, only the first component has a @code{=>}
17726 You may abbreviate attributes in expressions with any unique,
17727 multi-character subsequence of
17728 their names (an exact match gets preference).
17729 For example, you may use @t{a'len}, @t{a'gth}, or @t{a'lh}
17730 in place of @t{a'length}.
17733 @cindex quoting Ada internal identifiers
17734 Since Ada is case-insensitive, the debugger normally maps identifiers you type
17735 to lower case. The GNAT compiler uses upper-case characters for
17736 some of its internal identifiers, which are normally of no interest to users.
17737 For the rare occasions when you actually have to look at them,
17738 enclose them in angle brackets to avoid the lower-case mapping.
17741 (@value{GDBP}) print <JMPBUF_SAVE>[0]
17745 Printing an object of class-wide type or dereferencing an
17746 access-to-class-wide value will display all the components of the object's
17747 specific type (as indicated by its run-time tag). Likewise, component
17748 selection on such a value will operate on the specific type of the
17753 @node Overloading support for Ada
17754 @subsubsection Overloading support for Ada
17755 @cindex overloading, Ada
17757 The debugger supports limited overloading. Given a subprogram call in which
17758 the function symbol has multiple definitions, it will use the number of
17759 actual parameters and some information about their types to attempt to narrow
17760 the set of definitions. It also makes very limited use of context, preferring
17761 procedures to functions in the context of the @code{call} command, and
17762 functions to procedures elsewhere.
17764 If, after narrowing, the set of matching definitions still contains more than
17765 one definition, @value{GDBN} will display a menu to query which one it should
17769 (@value{GDBP}) print f(1)
17770 Multiple matches for f
17772 [1] foo.f (integer) return boolean at foo.adb:23
17773 [2] foo.f (foo.new_integer) return boolean at foo.adb:28
17777 In this case, just select one menu entry either to cancel expression evaluation
17778 (type @kbd{0} and press @key{RET}) or to continue evaluation with a specific
17779 instance (type the corresponding number and press @key{RET}).
17781 Here are a couple of commands to customize @value{GDBN}'s behavior in this
17786 @kindex set ada print-signatures
17787 @item set ada print-signatures
17788 Control whether parameter types and return types are displayed in overloads
17789 selection menus. It is @code{on} by default.
17790 @xref{Overloading support for Ada}.
17792 @kindex show ada print-signatures
17793 @item show ada print-signatures
17794 Show the current setting for displaying parameter types and return types in
17795 overloads selection menu.
17796 @xref{Overloading support for Ada}.
17800 @node Stopping Before Main Program
17801 @subsubsection Stopping at the Very Beginning
17803 @cindex breakpointing Ada elaboration code
17804 It is sometimes necessary to debug the program during elaboration, and
17805 before reaching the main procedure.
17806 As defined in the Ada Reference
17807 Manual, the elaboration code is invoked from a procedure called
17808 @code{adainit}. To run your program up to the beginning of
17809 elaboration, simply use the following two commands:
17810 @code{tbreak adainit} and @code{run}.
17812 @node Ada Exceptions
17813 @subsubsection Ada Exceptions
17815 A command is provided to list all Ada exceptions:
17818 @kindex info exceptions
17819 @item info exceptions
17820 @itemx info exceptions @var{regexp}
17821 The @code{info exceptions} command allows you to list all Ada exceptions
17822 defined within the program being debugged, as well as their addresses.
17823 With a regular expression, @var{regexp}, as argument, only those exceptions
17824 whose names match @var{regexp} are listed.
17827 Below is a small example, showing how the command can be used, first
17828 without argument, and next with a regular expression passed as an
17832 (@value{GDBP}) info exceptions
17833 All defined Ada exceptions:
17834 constraint_error: 0x613da0
17835 program_error: 0x613d20
17836 storage_error: 0x613ce0
17837 tasking_error: 0x613ca0
17838 const.aint_global_e: 0x613b00
17839 (@value{GDBP}) info exceptions const.aint
17840 All Ada exceptions matching regular expression "const.aint":
17841 constraint_error: 0x613da0
17842 const.aint_global_e: 0x613b00
17845 It is also possible to ask @value{GDBN} to stop your program's execution
17846 when an exception is raised. For more details, see @ref{Set Catchpoints}.
17849 @subsubsection Extensions for Ada Tasks
17850 @cindex Ada, tasking
17852 Support for Ada tasks is analogous to that for threads (@pxref{Threads}).
17853 @value{GDBN} provides the following task-related commands:
17858 This command shows a list of current Ada tasks, as in the following example:
17865 (@value{GDBP}) info tasks
17866 ID TID P-ID Pri State Name
17867 1 8088000 0 15 Child Activation Wait main_task
17868 2 80a4000 1 15 Accept Statement b
17869 3 809a800 1 15 Child Activation Wait a
17870 * 4 80ae800 3 15 Runnable c
17875 In this listing, the asterisk before the last task indicates it to be the
17876 task currently being inspected.
17880 Represents @value{GDBN}'s internal task number.
17886 The parent's task ID (@value{GDBN}'s internal task number).
17889 The base priority of the task.
17892 Current state of the task.
17896 The task has been created but has not been activated. It cannot be
17900 The task is not blocked for any reason known to Ada. (It may be waiting
17901 for a mutex, though.) It is conceptually "executing" in normal mode.
17904 The task is terminated, in the sense of ARM 9.3 (5). Any dependents
17905 that were waiting on terminate alternatives have been awakened and have
17906 terminated themselves.
17908 @item Child Activation Wait
17909 The task is waiting for created tasks to complete activation.
17911 @item Accept Statement
17912 The task is waiting on an accept or selective wait statement.
17914 @item Waiting on entry call
17915 The task is waiting on an entry call.
17917 @item Async Select Wait
17918 The task is waiting to start the abortable part of an asynchronous
17922 The task is waiting on a select statement with only a delay
17925 @item Child Termination Wait
17926 The task is sleeping having completed a master within itself, and is
17927 waiting for the tasks dependent on that master to become terminated or
17928 waiting on a terminate Phase.
17930 @item Wait Child in Term Alt
17931 The task is sleeping waiting for tasks on terminate alternatives to
17932 finish terminating.
17934 @item Accepting RV with @var{taskno}
17935 The task is accepting a rendez-vous with the task @var{taskno}.
17939 Name of the task in the program.
17943 @kindex info task @var{taskno}
17944 @item info task @var{taskno}
17945 This command shows detailed informations on the specified task, as in
17946 the following example:
17951 (@value{GDBP}) info tasks
17952 ID TID P-ID Pri State Name
17953 1 8077880 0 15 Child Activation Wait main_task
17954 * 2 807c468 1 15 Runnable task_1
17955 (@value{GDBP}) info task 2
17956 Ada Task: 0x807c468
17960 Parent: 1 ("main_task")
17966 @kindex task@r{ (Ada)}
17967 @cindex current Ada task ID
17968 This command prints the ID and name of the current task.
17974 (@value{GDBP}) info tasks
17975 ID TID P-ID Pri State Name
17976 1 8077870 0 15 Child Activation Wait main_task
17977 * 2 807c458 1 15 Runnable some_task
17978 (@value{GDBP}) task
17979 [Current task is 2 "some_task"]
17982 @item task @var{taskno}
17983 @cindex Ada task switching
17984 This command is like the @code{thread @var{thread-id}}
17985 command (@pxref{Threads}). It switches the context of debugging
17986 from the current task to the given task.
17992 (@value{GDBP}) info tasks
17993 ID TID P-ID Pri State Name
17994 1 8077870 0 15 Child Activation Wait main_task
17995 * 2 807c458 1 15 Runnable some_task
17996 (@value{GDBP}) task 1
17997 [Switching to task 1 "main_task"]
17998 #0 0x8067726 in pthread_cond_wait ()
18000 #0 0x8067726 in pthread_cond_wait ()
18001 #1 0x8056714 in system.os_interface.pthread_cond_wait ()
18002 #2 0x805cb63 in system.task_primitives.operations.sleep ()
18003 #3 0x806153e in system.tasking.stages.activate_tasks ()
18004 #4 0x804aacc in un () at un.adb:5
18007 @item break @var{location} task @var{taskno}
18008 @itemx break @var{location} task @var{taskno} if @dots{}
18009 @cindex breakpoints and tasks, in Ada
18010 @cindex task breakpoints, in Ada
18011 @kindex break @dots{} task @var{taskno}@r{ (Ada)}
18012 These commands are like the @code{break @dots{} thread @dots{}}
18013 command (@pxref{Thread Stops}). The
18014 @var{location} argument specifies source lines, as described
18015 in @ref{Specify Location}.
18017 Use the qualifier @samp{task @var{taskno}} with a breakpoint command
18018 to specify that you only want @value{GDBN} to stop the program when a
18019 particular Ada task reaches this breakpoint. The @var{taskno} is one of the
18020 numeric task identifiers assigned by @value{GDBN}, shown in the first
18021 column of the @samp{info tasks} display.
18023 If you do not specify @samp{task @var{taskno}} when you set a
18024 breakpoint, the breakpoint applies to @emph{all} tasks of your
18027 You can use the @code{task} qualifier on conditional breakpoints as
18028 well; in this case, place @samp{task @var{taskno}} before the
18029 breakpoint condition (before the @code{if}).
18037 (@value{GDBP}) info tasks
18038 ID TID P-ID Pri State Name
18039 1 140022020 0 15 Child Activation Wait main_task
18040 2 140045060 1 15 Accept/Select Wait t2
18041 3 140044840 1 15 Runnable t1
18042 * 4 140056040 1 15 Runnable t3
18043 (@value{GDBP}) b 15 task 2
18044 Breakpoint 5 at 0x120044cb0: file test_task_debug.adb, line 15.
18045 (@value{GDBP}) cont
18050 Breakpoint 5, test_task_debug () at test_task_debug.adb:15
18052 (@value{GDBP}) info tasks
18053 ID TID P-ID Pri State Name
18054 1 140022020 0 15 Child Activation Wait main_task
18055 * 2 140045060 1 15 Runnable t2
18056 3 140044840 1 15 Runnable t1
18057 4 140056040 1 15 Delay Sleep t3
18061 @node Ada Tasks and Core Files
18062 @subsubsection Tasking Support when Debugging Core Files
18063 @cindex Ada tasking and core file debugging
18065 When inspecting a core file, as opposed to debugging a live program,
18066 tasking support may be limited or even unavailable, depending on
18067 the platform being used.
18068 For instance, on x86-linux, the list of tasks is available, but task
18069 switching is not supported.
18071 On certain platforms, the debugger needs to perform some
18072 memory writes in order to provide Ada tasking support. When inspecting
18073 a core file, this means that the core file must be opened with read-write
18074 privileges, using the command @samp{"set write on"} (@pxref{Patching}).
18075 Under these circumstances, you should make a backup copy of the core
18076 file before inspecting it with @value{GDBN}.
18078 @node Ravenscar Profile
18079 @subsubsection Tasking Support when using the Ravenscar Profile
18080 @cindex Ravenscar Profile
18082 The @dfn{Ravenscar Profile} is a subset of the Ada tasking features,
18083 specifically designed for systems with safety-critical real-time
18087 @kindex set ravenscar task-switching on
18088 @cindex task switching with program using Ravenscar Profile
18089 @item set ravenscar task-switching on
18090 Allows task switching when debugging a program that uses the Ravenscar
18091 Profile. This is the default.
18093 @kindex set ravenscar task-switching off
18094 @item set ravenscar task-switching off
18095 Turn off task switching when debugging a program that uses the Ravenscar
18096 Profile. This is mostly intended to disable the code that adds support
18097 for the Ravenscar Profile, in case a bug in either @value{GDBN} or in
18098 the Ravenscar runtime is preventing @value{GDBN} from working properly.
18099 To be effective, this command should be run before the program is started.
18101 @kindex show ravenscar task-switching
18102 @item show ravenscar task-switching
18103 Show whether it is possible to switch from task to task in a program
18104 using the Ravenscar Profile.
18109 @subsubsection Ada Settings
18110 @cindex Ada settings
18113 @kindex set varsize-limit
18114 @item set varsize-limit @var{size}
18115 Prevent @value{GDBN} from attempting to evaluate objects whose size
18116 is above the given limit (@var{size}) when those sizes are computed
18117 from run-time quantities. This is typically the case when the object
18118 has a variable size, such as an array whose bounds are not known at
18119 compile time for example. Setting @var{size} to @code{unlimited}
18120 removes the size limitation. By default, the limit is about 65KB.
18122 The purpose of having such a limit is to prevent @value{GDBN} from
18123 trying to grab enormous chunks of virtual memory when asked to evaluate
18124 a quantity whose bounds have been corrupted or have not yet been fully
18125 initialized. The limit applies to the results of some subexpressions
18126 as well as to complete expressions. For example, an expression denoting
18127 a simple integer component, such as @code{x.y.z}, may fail if the size of
18128 @code{x.y} is variable and exceeds @code{size}. On the other hand,
18129 @value{GDBN} is sometimes clever; the expression @code{A(i)}, where
18130 @code{A} is an array variable with non-constant size, will generally
18131 succeed regardless of the bounds on @code{A}, as long as the component
18132 size is less than @var{size}.
18134 @kindex show varsize-limit
18135 @item show varsize-limit
18136 Show the limit on types whose size is determined by run-time quantities.
18140 @subsubsection Known Peculiarities of Ada Mode
18141 @cindex Ada, problems
18143 Besides the omissions listed previously (@pxref{Omissions from Ada}),
18144 we know of several problems with and limitations of Ada mode in
18146 some of which will be fixed with planned future releases of the debugger
18147 and the GNU Ada compiler.
18151 Static constants that the compiler chooses not to materialize as objects in
18152 storage are invisible to the debugger.
18155 Named parameter associations in function argument lists are ignored (the
18156 argument lists are treated as positional).
18159 Many useful library packages are currently invisible to the debugger.
18162 Fixed-point arithmetic, conversions, input, and output is carried out using
18163 floating-point arithmetic, and may give results that only approximate those on
18167 The GNAT compiler never generates the prefix @code{Standard} for any of
18168 the standard symbols defined by the Ada language. @value{GDBN} knows about
18169 this: it will strip the prefix from names when you use it, and will never
18170 look for a name you have so qualified among local symbols, nor match against
18171 symbols in other packages or subprograms. If you have
18172 defined entities anywhere in your program other than parameters and
18173 local variables whose simple names match names in @code{Standard},
18174 GNAT's lack of qualification here can cause confusion. When this happens,
18175 you can usually resolve the confusion
18176 by qualifying the problematic names with package
18177 @code{Standard} explicitly.
18180 Older versions of the compiler sometimes generate erroneous debugging
18181 information, resulting in the debugger incorrectly printing the value
18182 of affected entities. In some cases, the debugger is able to work
18183 around an issue automatically. In other cases, the debugger is able
18184 to work around the issue, but the work-around has to be specifically
18187 @kindex set ada trust-PAD-over-XVS
18188 @kindex show ada trust-PAD-over-XVS
18191 @item set ada trust-PAD-over-XVS on
18192 Configure GDB to strictly follow the GNAT encoding when computing the
18193 value of Ada entities, particularly when @code{PAD} and @code{PAD___XVS}
18194 types are involved (see @code{ada/exp_dbug.ads} in the GCC sources for
18195 a complete description of the encoding used by the GNAT compiler).
18196 This is the default.
18198 @item set ada trust-PAD-over-XVS off
18199 This is related to the encoding using by the GNAT compiler. If @value{GDBN}
18200 sometimes prints the wrong value for certain entities, changing @code{ada
18201 trust-PAD-over-XVS} to @code{off} activates a work-around which may fix
18202 the issue. It is always safe to set @code{ada trust-PAD-over-XVS} to
18203 @code{off}, but this incurs a slight performance penalty, so it is
18204 recommended to leave this setting to @code{on} unless necessary.
18208 @cindex GNAT descriptive types
18209 @cindex GNAT encoding
18210 Internally, the debugger also relies on the compiler following a number
18211 of conventions known as the @samp{GNAT Encoding}, all documented in
18212 @file{gcc/ada/exp_dbug.ads} in the GCC sources. This encoding describes
18213 how the debugging information should be generated for certain types.
18214 In particular, this convention makes use of @dfn{descriptive types},
18215 which are artificial types generated purely to help the debugger.
18217 These encodings were defined at a time when the debugging information
18218 format used was not powerful enough to describe some of the more complex
18219 types available in Ada. Since DWARF allows us to express nearly all
18220 Ada features, the long-term goal is to slowly replace these descriptive
18221 types by their pure DWARF equivalent. To facilitate that transition,
18222 a new maintenance option is available to force the debugger to ignore
18223 those descriptive types. It allows the user to quickly evaluate how
18224 well @value{GDBN} works without them.
18228 @kindex maint ada set ignore-descriptive-types
18229 @item maintenance ada set ignore-descriptive-types [on|off]
18230 Control whether the debugger should ignore descriptive types.
18231 The default is not to ignore descriptives types (@code{off}).
18233 @kindex maint ada show ignore-descriptive-types
18234 @item maintenance ada show ignore-descriptive-types
18235 Show if descriptive types are ignored by @value{GDBN}.
18239 @node Unsupported Languages
18240 @section Unsupported Languages
18242 @cindex unsupported languages
18243 @cindex minimal language
18244 In addition to the other fully-supported programming languages,
18245 @value{GDBN} also provides a pseudo-language, called @code{minimal}.
18246 It does not represent a real programming language, but provides a set
18247 of capabilities close to what the C or assembly languages provide.
18248 This should allow most simple operations to be performed while debugging
18249 an application that uses a language currently not supported by @value{GDBN}.
18251 If the language is set to @code{auto}, @value{GDBN} will automatically
18252 select this language if the current frame corresponds to an unsupported
18256 @chapter Examining the Symbol Table
18258 The commands described in this chapter allow you to inquire about the
18259 symbols (names of variables, functions and types) defined in your
18260 program. This information is inherent in the text of your program and
18261 does not change as your program executes. @value{GDBN} finds it in your
18262 program's symbol table, in the file indicated when you started @value{GDBN}
18263 (@pxref{File Options, ,Choosing Files}), or by one of the
18264 file-management commands (@pxref{Files, ,Commands to Specify Files}).
18266 @cindex symbol names
18267 @cindex names of symbols
18268 @cindex quoting names
18269 @anchor{quoting names}
18270 Occasionally, you may need to refer to symbols that contain unusual
18271 characters, which @value{GDBN} ordinarily treats as word delimiters. The
18272 most frequent case is in referring to static variables in other
18273 source files (@pxref{Variables,,Program Variables}). File names
18274 are recorded in object files as debugging symbols, but @value{GDBN} would
18275 ordinarily parse a typical file name, like @file{foo.c}, as the three words
18276 @samp{foo} @samp{.} @samp{c}. To allow @value{GDBN} to recognize
18277 @samp{foo.c} as a single symbol, enclose it in single quotes; for example,
18284 looks up the value of @code{x} in the scope of the file @file{foo.c}.
18287 @cindex case-insensitive symbol names
18288 @cindex case sensitivity in symbol names
18289 @kindex set case-sensitive
18290 @item set case-sensitive on
18291 @itemx set case-sensitive off
18292 @itemx set case-sensitive auto
18293 Normally, when @value{GDBN} looks up symbols, it matches their names
18294 with case sensitivity determined by the current source language.
18295 Occasionally, you may wish to control that. The command @code{set
18296 case-sensitive} lets you do that by specifying @code{on} for
18297 case-sensitive matches or @code{off} for case-insensitive ones. If
18298 you specify @code{auto}, case sensitivity is reset to the default
18299 suitable for the source language. The default is case-sensitive
18300 matches for all languages except for Fortran, for which the default is
18301 case-insensitive matches.
18303 @kindex show case-sensitive
18304 @item show case-sensitive
18305 This command shows the current setting of case sensitivity for symbols
18308 @kindex set print type methods
18309 @item set print type methods
18310 @itemx set print type methods on
18311 @itemx set print type methods off
18312 Normally, when @value{GDBN} prints a class, it displays any methods
18313 declared in that class. You can control this behavior either by
18314 passing the appropriate flag to @code{ptype}, or using @command{set
18315 print type methods}. Specifying @code{on} will cause @value{GDBN} to
18316 display the methods; this is the default. Specifying @code{off} will
18317 cause @value{GDBN} to omit the methods.
18319 @kindex show print type methods
18320 @item show print type methods
18321 This command shows the current setting of method display when printing
18324 @kindex set print type nested-type-limit
18325 @item set print type nested-type-limit @var{limit}
18326 @itemx set print type nested-type-limit unlimited
18327 Set the limit of displayed nested types that the type printer will
18328 show. A @var{limit} of @code{unlimited} or @code{-1} will show all
18329 nested definitions. By default, the type printer will not show any nested
18330 types defined in classes.
18332 @kindex show print type nested-type-limit
18333 @item show print type nested-type-limit
18334 This command shows the current display limit of nested types when
18337 @kindex set print type typedefs
18338 @item set print type typedefs
18339 @itemx set print type typedefs on
18340 @itemx set print type typedefs off
18342 Normally, when @value{GDBN} prints a class, it displays any typedefs
18343 defined in that class. You can control this behavior either by
18344 passing the appropriate flag to @code{ptype}, or using @command{set
18345 print type typedefs}. Specifying @code{on} will cause @value{GDBN} to
18346 display the typedef definitions; this is the default. Specifying
18347 @code{off} will cause @value{GDBN} to omit the typedef definitions.
18348 Note that this controls whether the typedef definition itself is
18349 printed, not whether typedef names are substituted when printing other
18352 @kindex show print type typedefs
18353 @item show print type typedefs
18354 This command shows the current setting of typedef display when
18357 @kindex info address
18358 @cindex address of a symbol
18359 @item info address @var{symbol}
18360 Describe where the data for @var{symbol} is stored. For a register
18361 variable, this says which register it is kept in. For a non-register
18362 local variable, this prints the stack-frame offset at which the variable
18365 Note the contrast with @samp{print &@var{symbol}}, which does not work
18366 at all for a register variable, and for a stack local variable prints
18367 the exact address of the current instantiation of the variable.
18369 @kindex info symbol
18370 @cindex symbol from address
18371 @cindex closest symbol and offset for an address
18372 @item info symbol @var{addr}
18373 Print the name of a symbol which is stored at the address @var{addr}.
18374 If no symbol is stored exactly at @var{addr}, @value{GDBN} prints the
18375 nearest symbol and an offset from it:
18378 (@value{GDBP}) info symbol 0x54320
18379 _initialize_vx + 396 in section .text
18383 This is the opposite of the @code{info address} command. You can use
18384 it to find out the name of a variable or a function given its address.
18386 For dynamically linked executables, the name of executable or shared
18387 library containing the symbol is also printed:
18390 (@value{GDBP}) info symbol 0x400225
18391 _start + 5 in section .text of /tmp/a.out
18392 (@value{GDBP}) info symbol 0x2aaaac2811cf
18393 __read_nocancel + 6 in section .text of /usr/lib64/libc.so.6
18398 @item demangle @r{[}-l @var{language}@r{]} @r{[}@var{--}@r{]} @var{name}
18399 Demangle @var{name}.
18400 If @var{language} is provided it is the name of the language to demangle
18401 @var{name} in. Otherwise @var{name} is demangled in the current language.
18403 The @samp{--} option specifies the end of options,
18404 and is useful when @var{name} begins with a dash.
18406 The parameter @code{demangle-style} specifies how to interpret the kind
18407 of mangling used. @xref{Print Settings}.
18410 @item whatis[/@var{flags}] [@var{arg}]
18411 Print the data type of @var{arg}, which can be either an expression
18412 or a name of a data type. With no argument, print the data type of
18413 @code{$}, the last value in the value history.
18415 If @var{arg} is an expression (@pxref{Expressions, ,Expressions}), it
18416 is not actually evaluated, and any side-effecting operations (such as
18417 assignments or function calls) inside it do not take place.
18419 If @var{arg} is a variable or an expression, @code{whatis} prints its
18420 literal type as it is used in the source code. If the type was
18421 defined using a @code{typedef}, @code{whatis} will @emph{not} print
18422 the data type underlying the @code{typedef}. If the type of the
18423 variable or the expression is a compound data type, such as
18424 @code{struct} or @code{class}, @code{whatis} never prints their
18425 fields or methods. It just prints the @code{struct}/@code{class}
18426 name (a.k.a.@: its @dfn{tag}). If you want to see the members of
18427 such a compound data type, use @code{ptype}.
18429 If @var{arg} is a type name that was defined using @code{typedef},
18430 @code{whatis} @dfn{unrolls} only one level of that @code{typedef}.
18431 Unrolling means that @code{whatis} will show the underlying type used
18432 in the @code{typedef} declaration of @var{arg}. However, if that
18433 underlying type is also a @code{typedef}, @code{whatis} will not
18436 For C code, the type names may also have the form @samp{class
18437 @var{class-name}}, @samp{struct @var{struct-tag}}, @samp{union
18438 @var{union-tag}} or @samp{enum @var{enum-tag}}.
18440 @var{flags} can be used to modify how the type is displayed.
18441 Available flags are:
18445 Display in ``raw'' form. Normally, @value{GDBN} substitutes template
18446 parameters and typedefs defined in a class when printing the class'
18447 members. The @code{/r} flag disables this.
18450 Do not print methods defined in the class.
18453 Print methods defined in the class. This is the default, but the flag
18454 exists in case you change the default with @command{set print type methods}.
18457 Do not print typedefs defined in the class. Note that this controls
18458 whether the typedef definition itself is printed, not whether typedef
18459 names are substituted when printing other types.
18462 Print typedefs defined in the class. This is the default, but the flag
18463 exists in case you change the default with @command{set print type typedefs}.
18466 Print the offsets and sizes of fields in a struct, similar to what the
18467 @command{pahole} tool does. This option implies the @code{/tm} flags.
18469 For example, given the following declarations:
18505 Issuing a @kbd{ptype /o struct tuv} command would print:
18508 (@value{GDBP}) ptype /o struct tuv
18509 /* offset | size */ type = struct tuv @{
18510 /* 0 | 4 */ int a1;
18511 /* XXX 4-byte hole */
18512 /* 8 | 8 */ char *a2;
18513 /* 16 | 4 */ int a3;
18515 /* total size (bytes): 24 */
18519 Notice the format of the first column of comments. There, you can
18520 find two parts separated by the @samp{|} character: the @emph{offset},
18521 which indicates where the field is located inside the struct, in
18522 bytes, and the @emph{size} of the field. Another interesting line is
18523 the marker of a @emph{hole} in the struct, indicating that it may be
18524 possible to pack the struct and make it use less space by reorganizing
18527 It is also possible to print offsets inside an union:
18530 (@value{GDBP}) ptype /o union qwe
18531 /* offset | size */ type = union qwe @{
18532 /* 24 */ struct tuv @{
18533 /* 0 | 4 */ int a1;
18534 /* XXX 4-byte hole */
18535 /* 8 | 8 */ char *a2;
18536 /* 16 | 4 */ int a3;
18538 /* total size (bytes): 24 */
18540 /* 40 */ struct xyz @{
18541 /* 0 | 4 */ int f1;
18542 /* 4 | 1 */ char f2;
18543 /* XXX 3-byte hole */
18544 /* 8 | 8 */ void *f3;
18545 /* 16 | 24 */ struct tuv @{
18546 /* 16 | 4 */ int a1;
18547 /* XXX 4-byte hole */
18548 /* 24 | 8 */ char *a2;
18549 /* 32 | 4 */ int a3;
18551 /* total size (bytes): 24 */
18554 /* total size (bytes): 40 */
18557 /* total size (bytes): 40 */
18561 In this case, since @code{struct tuv} and @code{struct xyz} occupy the
18562 same space (because we are dealing with an union), the offset is not
18563 printed for them. However, you can still examine the offset of each
18564 of these structures' fields.
18566 Another useful scenario is printing the offsets of a struct containing
18570 (@value{GDBP}) ptype /o struct tyu
18571 /* offset | size */ type = struct tyu @{
18572 /* 0:31 | 4 */ int a1 : 1;
18573 /* 0:28 | 4 */ int a2 : 3;
18574 /* 0: 5 | 4 */ int a3 : 23;
18575 /* 3: 3 | 1 */ signed char a4 : 2;
18576 /* XXX 3-bit hole */
18577 /* XXX 4-byte hole */
18578 /* 8 | 8 */ int64_t a5;
18579 /* 16: 0 | 4 */ int a6 : 5;
18580 /* 16: 5 | 8 */ int64_t a7 : 3;
18581 "/* XXX 7-byte padding */
18583 /* total size (bytes): 24 */
18587 Note how the offset information is now extended to also include the
18588 first bit of the bitfield.
18592 @item ptype[/@var{flags}] [@var{arg}]
18593 @code{ptype} accepts the same arguments as @code{whatis}, but prints a
18594 detailed description of the type, instead of just the name of the type.
18595 @xref{Expressions, ,Expressions}.
18597 Contrary to @code{whatis}, @code{ptype} always unrolls any
18598 @code{typedef}s in its argument declaration, whether the argument is
18599 a variable, expression, or a data type. This means that @code{ptype}
18600 of a variable or an expression will not print literally its type as
18601 present in the source code---use @code{whatis} for that. @code{typedef}s at
18602 the pointer or reference targets are also unrolled. Only @code{typedef}s of
18603 fields, methods and inner @code{class typedef}s of @code{struct}s,
18604 @code{class}es and @code{union}s are not unrolled even with @code{ptype}.
18606 For example, for this variable declaration:
18609 typedef double real_t;
18610 struct complex @{ real_t real; double imag; @};
18611 typedef struct complex complex_t;
18613 real_t *real_pointer_var;
18617 the two commands give this output:
18621 (@value{GDBP}) whatis var
18623 (@value{GDBP}) ptype var
18624 type = struct complex @{
18628 (@value{GDBP}) whatis complex_t
18629 type = struct complex
18630 (@value{GDBP}) whatis struct complex
18631 type = struct complex
18632 (@value{GDBP}) ptype struct complex
18633 type = struct complex @{
18637 (@value{GDBP}) whatis real_pointer_var
18639 (@value{GDBP}) ptype real_pointer_var
18645 As with @code{whatis}, using @code{ptype} without an argument refers to
18646 the type of @code{$}, the last value in the value history.
18648 @cindex incomplete type
18649 Sometimes, programs use opaque data types or incomplete specifications
18650 of complex data structure. If the debug information included in the
18651 program does not allow @value{GDBN} to display a full declaration of
18652 the data type, it will say @samp{<incomplete type>}. For example,
18653 given these declarations:
18657 struct foo *fooptr;
18661 but no definition for @code{struct foo} itself, @value{GDBN} will say:
18664 (@value{GDBP}) ptype foo
18665 $1 = <incomplete type>
18669 ``Incomplete type'' is C terminology for data types that are not
18670 completely specified.
18672 @cindex unknown type
18673 Othertimes, information about a variable's type is completely absent
18674 from the debug information included in the program. This most often
18675 happens when the program or library where the variable is defined
18676 includes no debug information at all. @value{GDBN} knows the variable
18677 exists from inspecting the linker/loader symbol table (e.g., the ELF
18678 dynamic symbol table), but such symbols do not contain type
18679 information. Inspecting the type of a (global) variable for which
18680 @value{GDBN} has no type information shows:
18683 (@value{GDBP}) ptype var
18684 type = <data variable, no debug info>
18687 @xref{Variables, no debug info variables}, for how to print the values
18691 @item info types [-q] [@var{regexp}]
18692 Print a brief description of all types whose names match the regular
18693 expression @var{regexp} (or all types in your program, if you supply
18694 no argument). Each complete typename is matched as though it were a
18695 complete line; thus, @samp{i type value} gives information on all
18696 types in your program whose names include the string @code{value}, but
18697 @samp{i type ^value$} gives information only on types whose complete
18698 name is @code{value}.
18700 In programs using different languages, @value{GDBN} chooses the syntax
18701 to print the type description according to the
18702 @samp{set language} value: using @samp{set language auto}
18703 (see @ref{Automatically, ,Set Language Automatically}) means to use the
18704 language of the type, other values mean to use
18705 the manually specified language (see @ref{Manually, ,Set Language Manually}).
18707 This command differs from @code{ptype} in two ways: first, like
18708 @code{whatis}, it does not print a detailed description; second, it
18709 lists all source files and line numbers where a type is defined.
18711 The output from @samp{into types} is proceeded with a header line
18712 describing what types are being listed. The optional flag @samp{-q},
18713 which stands for @samp{quiet}, disables printing this header
18716 @kindex info type-printers
18717 @item info type-printers
18718 Versions of @value{GDBN} that ship with Python scripting enabled may
18719 have ``type printers'' available. When using @command{ptype} or
18720 @command{whatis}, these printers are consulted when the name of a type
18721 is needed. @xref{Type Printing API}, for more information on writing
18724 @code{info type-printers} displays all the available type printers.
18726 @kindex enable type-printer
18727 @kindex disable type-printer
18728 @item enable type-printer @var{name}@dots{}
18729 @item disable type-printer @var{name}@dots{}
18730 These commands can be used to enable or disable type printers.
18733 @cindex local variables
18734 @item info scope @var{location}
18735 List all the variables local to a particular scope. This command
18736 accepts a @var{location} argument---a function name, a source line, or
18737 an address preceded by a @samp{*}, and prints all the variables local
18738 to the scope defined by that location. (@xref{Specify Location}, for
18739 details about supported forms of @var{location}.) For example:
18742 (@value{GDBP}) @b{info scope command_line_handler}
18743 Scope for command_line_handler:
18744 Symbol rl is an argument at stack/frame offset 8, length 4.
18745 Symbol linebuffer is in static storage at address 0x150a18, length 4.
18746 Symbol linelength is in static storage at address 0x150a1c, length 4.
18747 Symbol p is a local variable in register $esi, length 4.
18748 Symbol p1 is a local variable in register $ebx, length 4.
18749 Symbol nline is a local variable in register $edx, length 4.
18750 Symbol repeat is a local variable at frame offset -8, length 4.
18754 This command is especially useful for determining what data to collect
18755 during a @dfn{trace experiment}, see @ref{Tracepoint Actions,
18758 @kindex info source
18760 Show information about the current source file---that is, the source file for
18761 the function containing the current point of execution:
18764 the name of the source file, and the directory containing it,
18766 the directory it was compiled in,
18768 its length, in lines,
18770 which programming language it is written in,
18772 if the debug information provides it, the program that compiled the file
18773 (which may include, e.g., the compiler version and command line arguments),
18775 whether the executable includes debugging information for that file, and
18776 if so, what format the information is in (e.g., STABS, Dwarf 2, etc.), and
18778 whether the debugging information includes information about
18779 preprocessor macros.
18783 @kindex info sources
18785 Print the names of all source files in your program for which there is
18786 debugging information, organized into two lists: files whose symbols
18787 have already been read, and files whose symbols will be read when needed.
18789 @item info sources [-dirname | -basename] [--] [@var{regexp}]
18790 Like @samp{info sources}, but only print the names of the files
18791 matching the provided @var{regexp}.
18792 By default, the @var{regexp} is used to match anywhere in the filename.
18793 If @code{-dirname}, only files having a dirname matching @var{regexp} are shown.
18794 If @code{-basename}, only files having a basename matching @var{regexp}
18796 The matching is case-sensitive, except on operating systems that
18797 have case-insensitive filesystem (e.g., MS-Windows).
18799 @kindex info functions
18800 @item info functions [-q] [-n]
18801 Print the names and data types of all defined functions.
18802 Similarly to @samp{info types}, this command groups its output by source
18803 files and annotates each function definition with its source line
18806 In programs using different languages, @value{GDBN} chooses the syntax
18807 to print the function name and type according to the
18808 @samp{set language} value: using @samp{set language auto}
18809 (see @ref{Automatically, ,Set Language Automatically}) means to use the
18810 language of the function, other values mean to use
18811 the manually specified language (see @ref{Manually, ,Set Language Manually}).
18813 The @samp{-n} flag excludes @dfn{non-debugging symbols} from the
18814 results. A non-debugging symbol is a symbol that comes from the
18815 executable's symbol table, not from the debug information (for
18816 example, DWARF) associated with the executable.
18818 The optional flag @samp{-q}, which stands for @samp{quiet}, disables
18819 printing header information and messages explaining why no functions
18822 @item info functions [-q] [-n] [-t @var{type_regexp}] [@var{regexp}]
18823 Like @samp{info functions}, but only print the names and data types
18824 of the functions selected with the provided regexp(s).
18826 If @var{regexp} is provided, print only the functions whose names
18827 match the regular expression @var{regexp}.
18828 Thus, @samp{info fun step} finds all functions whose
18829 names include @code{step}; @samp{info fun ^step} finds those whose names
18830 start with @code{step}. If a function name contains characters that
18831 conflict with the regular expression language (e.g.@:
18832 @samp{operator*()}), they may be quoted with a backslash.
18834 If @var{type_regexp} is provided, print only the functions whose
18835 types, as printed by the @code{whatis} command, match
18836 the regular expression @var{type_regexp}.
18837 If @var{type_regexp} contains space(s), it should be enclosed in
18838 quote characters. If needed, use backslash to escape the meaning
18839 of special characters or quotes.
18840 Thus, @samp{info fun -t '^int ('} finds the functions that return
18841 an integer; @samp{info fun -t '(.*int.*'} finds the functions that
18842 have an argument type containing int; @samp{info fun -t '^int (' ^step}
18843 finds the functions whose names start with @code{step} and that return
18846 If both @var{regexp} and @var{type_regexp} are provided, a function
18847 is printed only if its name matches @var{regexp} and its type matches
18851 @kindex info variables
18852 @item info variables [-q] [-n]
18853 Print the names and data types of all variables that are defined
18854 outside of functions (i.e.@: excluding local variables).
18855 The printed variables are grouped by source files and annotated with
18856 their respective source line numbers.
18858 In programs using different languages, @value{GDBN} chooses the syntax
18859 to print the variable name and type according to the
18860 @samp{set language} value: using @samp{set language auto}
18861 (see @ref{Automatically, ,Set Language Automatically}) means to use the
18862 language of the variable, other values mean to use
18863 the manually specified language (see @ref{Manually, ,Set Language Manually}).
18865 The @samp{-n} flag excludes non-debugging symbols from the results.
18867 The optional flag @samp{-q}, which stands for @samp{quiet}, disables
18868 printing header information and messages explaining why no variables
18871 @item info variables [-q] [-n] [-t @var{type_regexp}] [@var{regexp}]
18872 Like @kbd{info variables}, but only print the variables selected
18873 with the provided regexp(s).
18875 If @var{regexp} is provided, print only the variables whose names
18876 match the regular expression @var{regexp}.
18878 If @var{type_regexp} is provided, print only the variables whose
18879 types, as printed by the @code{whatis} command, match
18880 the regular expression @var{type_regexp}.
18881 If @var{type_regexp} contains space(s), it should be enclosed in
18882 quote characters. If needed, use backslash to escape the meaning
18883 of special characters or quotes.
18885 If both @var{regexp} and @var{type_regexp} are provided, an argument
18886 is printed only if its name matches @var{regexp} and its type matches
18889 @kindex info modules
18891 @item info modules @r{[}-q@r{]} @r{[}@var{regexp}@r{]}
18892 List all Fortran modules in the program, or all modules matching the
18893 optional regular expression @var{regexp}.
18895 The optional flag @samp{-q}, which stands for @samp{quiet}, disables
18896 printing header information and messages explaining why no modules
18899 @kindex info module
18900 @cindex Fortran modules, information about
18901 @cindex functions and variables by Fortran module
18902 @cindex module functions and variables
18903 @item info module functions @r{[}-q@r{]} @r{[}-m @var{module-regexp}@r{]} @r{[}-t @var{type-regexp}@r{]} @r{[}@var{regexp}@r{]}
18904 @itemx info module variables @r{[}-q@r{]} @r{[}-m @var{module-regexp}@r{]} @r{[}-t @var{type-regexp}@r{]} @r{[}@var{regexp}@r{]}
18905 List all functions or variables within all Fortran modules. The set
18906 of functions or variables listed can be limited by providing some or
18907 all of the optional regular expressions. If @var{module-regexp} is
18908 provided, then only Fortran modules matching @var{module-regexp} will
18909 be searched. Only functions or variables whose type matches the
18910 optional regular expression @var{type-regexp} will be listed. And
18911 only functions or variables whose name matches the optional regular
18912 expression @var{regexp} will be listed.
18914 The optional flag @samp{-q}, which stands for @samp{quiet}, disables
18915 printing header information and messages explaining why no functions
18916 or variables have been printed.
18918 @kindex info classes
18919 @cindex Objective-C, classes and selectors
18921 @itemx info classes @var{regexp}
18922 Display all Objective-C classes in your program, or
18923 (with the @var{regexp} argument) all those matching a particular regular
18926 @kindex info selectors
18927 @item info selectors
18928 @itemx info selectors @var{regexp}
18929 Display all Objective-C selectors in your program, or
18930 (with the @var{regexp} argument) all those matching a particular regular
18934 This was never implemented.
18935 @kindex info methods
18937 @itemx info methods @var{regexp}
18938 The @code{info methods} command permits the user to examine all defined
18939 methods within C@t{++} program, or (with the @var{regexp} argument) a
18940 specific set of methods found in the various C@t{++} classes. Many
18941 C@t{++} classes provide a large number of methods. Thus, the output
18942 from the @code{ptype} command can be overwhelming and hard to use. The
18943 @code{info-methods} command filters the methods, printing only those
18944 which match the regular-expression @var{regexp}.
18947 @cindex opaque data types
18948 @kindex set opaque-type-resolution
18949 @item set opaque-type-resolution on
18950 Tell @value{GDBN} to resolve opaque types. An opaque type is a type
18951 declared as a pointer to a @code{struct}, @code{class}, or
18952 @code{union}---for example, @code{struct MyType *}---that is used in one
18953 source file although the full declaration of @code{struct MyType} is in
18954 another source file. The default is on.
18956 A change in the setting of this subcommand will not take effect until
18957 the next time symbols for a file are loaded.
18959 @item set opaque-type-resolution off
18960 Tell @value{GDBN} not to resolve opaque types. In this case, the type
18961 is printed as follows:
18963 @{<no data fields>@}
18966 @kindex show opaque-type-resolution
18967 @item show opaque-type-resolution
18968 Show whether opaque types are resolved or not.
18970 @kindex set print symbol-loading
18971 @cindex print messages when symbols are loaded
18972 @item set print symbol-loading
18973 @itemx set print symbol-loading full
18974 @itemx set print symbol-loading brief
18975 @itemx set print symbol-loading off
18976 The @code{set print symbol-loading} command allows you to control the
18977 printing of messages when @value{GDBN} loads symbol information.
18978 By default a message is printed for the executable and one for each
18979 shared library, and normally this is what you want. However, when
18980 debugging apps with large numbers of shared libraries these messages
18982 When set to @code{brief} a message is printed for each executable,
18983 and when @value{GDBN} loads a collection of shared libraries at once
18984 it will only print one message regardless of the number of shared
18985 libraries. When set to @code{off} no messages are printed.
18987 @kindex show print symbol-loading
18988 @item show print symbol-loading
18989 Show whether messages will be printed when a @value{GDBN} command
18990 entered from the keyboard causes symbol information to be loaded.
18992 @kindex maint print symbols
18993 @cindex symbol dump
18994 @kindex maint print psymbols
18995 @cindex partial symbol dump
18996 @kindex maint print msymbols
18997 @cindex minimal symbol dump
18998 @item maint print symbols @r{[}-pc @var{address}@r{]} @r{[}@var{filename}@r{]}
18999 @itemx maint print symbols @r{[}-objfile @var{objfile}@r{]} @r{[}-source @var{source}@r{]} @r{[}--@r{]} @r{[}@var{filename}@r{]}
19000 @itemx maint print psymbols @r{[}-objfile @var{objfile}@r{]} @r{[}-pc @var{address}@r{]} @r{[}--@r{]} @r{[}@var{filename}@r{]}
19001 @itemx maint print psymbols @r{[}-objfile @var{objfile}@r{]} @r{[}-source @var{source}@r{]} @r{[}--@r{]} @r{[}@var{filename}@r{]}
19002 @itemx maint print msymbols @r{[}-objfile @var{objfile}@r{]} @r{[}--@r{]} @r{[}@var{filename}@r{]}
19003 Write a dump of debugging symbol data into the file @var{filename} or
19004 the terminal if @var{filename} is unspecified.
19005 If @code{-objfile @var{objfile}} is specified, only dump symbols for
19007 If @code{-pc @var{address}} is specified, only dump symbols for the file
19008 with code at that address. Note that @var{address} may be a symbol like
19010 If @code{-source @var{source}} is specified, only dump symbols for that
19013 These commands are used to debug the @value{GDBN} symbol-reading code.
19014 These commands do not modify internal @value{GDBN} state, therefore
19015 @samp{maint print symbols} will only print symbols for already expanded symbol
19017 You can use the command @code{info sources} to find out which files these are.
19018 If you use @samp{maint print psymbols} instead, the dump shows information
19019 about symbols that @value{GDBN} only knows partially---that is, symbols
19020 defined in files that @value{GDBN} has skimmed, but not yet read completely.
19021 Finally, @samp{maint print msymbols} just dumps ``minimal symbols'', e.g.,
19024 @xref{Files, ,Commands to Specify Files}, for a discussion of how
19025 @value{GDBN} reads symbols (in the description of @code{symbol-file}).
19027 @kindex maint info symtabs
19028 @kindex maint info psymtabs
19029 @cindex listing @value{GDBN}'s internal symbol tables
19030 @cindex symbol tables, listing @value{GDBN}'s internal
19031 @cindex full symbol tables, listing @value{GDBN}'s internal
19032 @cindex partial symbol tables, listing @value{GDBN}'s internal
19033 @item maint info symtabs @r{[} @var{regexp} @r{]}
19034 @itemx maint info psymtabs @r{[} @var{regexp} @r{]}
19036 List the @code{struct symtab} or @code{struct partial_symtab}
19037 structures whose names match @var{regexp}. If @var{regexp} is not
19038 given, list them all. The output includes expressions which you can
19039 copy into a @value{GDBN} debugging this one to examine a particular
19040 structure in more detail. For example:
19043 (@value{GDBP}) maint info psymtabs dwarf2read
19044 @{ objfile /home/gnu/build/gdb/gdb
19045 ((struct objfile *) 0x82e69d0)
19046 @{ psymtab /home/gnu/src/gdb/dwarf2read.c
19047 ((struct partial_symtab *) 0x8474b10)
19050 text addresses 0x814d3c8 -- 0x8158074
19051 globals (* (struct partial_symbol **) 0x8507a08 @@ 9)
19052 statics (* (struct partial_symbol **) 0x40e95b78 @@ 2882)
19053 dependencies (none)
19056 (@value{GDBP}) maint info symtabs
19060 We see that there is one partial symbol table whose filename contains
19061 the string @samp{dwarf2read}, belonging to the @samp{gdb} executable;
19062 and we see that @value{GDBN} has not read in any symtabs yet at all.
19063 If we set a breakpoint on a function, that will cause @value{GDBN} to
19064 read the symtab for the compilation unit containing that function:
19067 (@value{GDBP}) break dwarf2_psymtab_to_symtab
19068 Breakpoint 1 at 0x814e5da: file /home/gnu/src/gdb/dwarf2read.c,
19070 (@value{GDBP}) maint info symtabs
19071 @{ objfile /home/gnu/build/gdb/gdb
19072 ((struct objfile *) 0x82e69d0)
19073 @{ symtab /home/gnu/src/gdb/dwarf2read.c
19074 ((struct symtab *) 0x86c1f38)
19077 blockvector ((struct blockvector *) 0x86c1bd0) (primary)
19078 linetable ((struct linetable *) 0x8370fa0)
19079 debugformat DWARF 2
19085 @kindex maint info line-table
19086 @cindex listing @value{GDBN}'s internal line tables
19087 @cindex line tables, listing @value{GDBN}'s internal
19088 @item maint info line-table @r{[} @var{regexp} @r{]}
19090 List the @code{struct linetable} from all @code{struct symtab}
19091 instances whose name matches @var{regexp}. If @var{regexp} is not
19092 given, list the @code{struct linetable} from all @code{struct symtab}.
19094 @kindex maint set symbol-cache-size
19095 @cindex symbol cache size
19096 @item maint set symbol-cache-size @var{size}
19097 Set the size of the symbol cache to @var{size}.
19098 The default size is intended to be good enough for debugging
19099 most applications. This option exists to allow for experimenting
19100 with different sizes.
19102 @kindex maint show symbol-cache-size
19103 @item maint show symbol-cache-size
19104 Show the size of the symbol cache.
19106 @kindex maint print symbol-cache
19107 @cindex symbol cache, printing its contents
19108 @item maint print symbol-cache
19109 Print the contents of the symbol cache.
19110 This is useful when debugging symbol cache issues.
19112 @kindex maint print symbol-cache-statistics
19113 @cindex symbol cache, printing usage statistics
19114 @item maint print symbol-cache-statistics
19115 Print symbol cache usage statistics.
19116 This helps determine how well the cache is being utilized.
19118 @kindex maint flush-symbol-cache
19119 @cindex symbol cache, flushing
19120 @item maint flush-symbol-cache
19121 Flush the contents of the symbol cache, all entries are removed.
19122 This command is useful when debugging the symbol cache.
19123 It is also useful when collecting performance data.
19128 @chapter Altering Execution
19130 Once you think you have found an error in your program, you might want to
19131 find out for certain whether correcting the apparent error would lead to
19132 correct results in the rest of the run. You can find the answer by
19133 experiment, using the @value{GDBN} features for altering execution of the
19136 For example, you can store new values into variables or memory
19137 locations, give your program a signal, restart it at a different
19138 address, or even return prematurely from a function.
19141 * Assignment:: Assignment to variables
19142 * Jumping:: Continuing at a different address
19143 * Signaling:: Giving your program a signal
19144 * Returning:: Returning from a function
19145 * Calling:: Calling your program's functions
19146 * Patching:: Patching your program
19147 * Compiling and Injecting Code:: Compiling and injecting code in @value{GDBN}
19151 @section Assignment to Variables
19154 @cindex setting variables
19155 To alter the value of a variable, evaluate an assignment expression.
19156 @xref{Expressions, ,Expressions}. For example,
19163 stores the value 4 into the variable @code{x}, and then prints the
19164 value of the assignment expression (which is 4).
19165 @xref{Languages, ,Using @value{GDBN} with Different Languages}, for more
19166 information on operators in supported languages.
19168 @kindex set variable
19169 @cindex variables, setting
19170 If you are not interested in seeing the value of the assignment, use the
19171 @code{set} command instead of the @code{print} command. @code{set} is
19172 really the same as @code{print} except that the expression's value is
19173 not printed and is not put in the value history (@pxref{Value History,
19174 ,Value History}). The expression is evaluated only for its effects.
19176 If the beginning of the argument string of the @code{set} command
19177 appears identical to a @code{set} subcommand, use the @code{set
19178 variable} command instead of just @code{set}. This command is identical
19179 to @code{set} except for its lack of subcommands. For example, if your
19180 program has a variable @code{width}, you get an error if you try to set
19181 a new value with just @samp{set width=13}, because @value{GDBN} has the
19182 command @code{set width}:
19185 (@value{GDBP}) whatis width
19187 (@value{GDBP}) p width
19189 (@value{GDBP}) set width=47
19190 Invalid syntax in expression.
19194 The invalid expression, of course, is @samp{=47}. In
19195 order to actually set the program's variable @code{width}, use
19198 (@value{GDBP}) set var width=47
19201 Because the @code{set} command has many subcommands that can conflict
19202 with the names of program variables, it is a good idea to use the
19203 @code{set variable} command instead of just @code{set}. For example, if
19204 your program has a variable @code{g}, you run into problems if you try
19205 to set a new value with just @samp{set g=4}, because @value{GDBN} has
19206 the command @code{set gnutarget}, abbreviated @code{set g}:
19210 (@value{GDBP}) whatis g
19214 (@value{GDBP}) set g=4
19218 The program being debugged has been started already.
19219 Start it from the beginning? (y or n) y
19220 Starting program: /home/smith/cc_progs/a.out
19221 "/home/smith/cc_progs/a.out": can't open to read symbols:
19222 Invalid bfd target.
19223 (@value{GDBP}) show g
19224 The current BFD target is "=4".
19229 The program variable @code{g} did not change, and you silently set the
19230 @code{gnutarget} to an invalid value. In order to set the variable
19234 (@value{GDBP}) set var g=4
19237 @value{GDBN} allows more implicit conversions in assignments than C; you can
19238 freely store an integer value into a pointer variable or vice versa,
19239 and you can convert any structure to any other structure that is the
19240 same length or shorter.
19241 @comment FIXME: how do structs align/pad in these conversions?
19242 @comment /doc@cygnus.com 18dec1990
19244 To store values into arbitrary places in memory, use the @samp{@{@dots{}@}}
19245 construct to generate a value of specified type at a specified address
19246 (@pxref{Expressions, ,Expressions}). For example, @code{@{int@}0x83040} refers
19247 to memory location @code{0x83040} as an integer (which implies a certain size
19248 and representation in memory), and
19251 set @{int@}0x83040 = 4
19255 stores the value 4 into that memory location.
19258 @section Continuing at a Different Address
19260 Ordinarily, when you continue your program, you do so at the place where
19261 it stopped, with the @code{continue} command. You can instead continue at
19262 an address of your own choosing, with the following commands:
19266 @kindex j @r{(@code{jump})}
19267 @item jump @var{location}
19268 @itemx j @var{location}
19269 Resume execution at @var{location}. Execution stops again immediately
19270 if there is a breakpoint there. @xref{Specify Location}, for a description
19271 of the different forms of @var{location}. It is common
19272 practice to use the @code{tbreak} command in conjunction with
19273 @code{jump}. @xref{Set Breaks, ,Setting Breakpoints}.
19275 The @code{jump} command does not change the current stack frame, or
19276 the stack pointer, or the contents of any memory location or any
19277 register other than the program counter. If @var{location} is in
19278 a different function from the one currently executing, the results may
19279 be bizarre if the two functions expect different patterns of arguments or
19280 of local variables. For this reason, the @code{jump} command requests
19281 confirmation if the specified line is not in the function currently
19282 executing. However, even bizarre results are predictable if you are
19283 well acquainted with the machine-language code of your program.
19286 On many systems, you can get much the same effect as the @code{jump}
19287 command by storing a new value into the register @code{$pc}. The
19288 difference is that this does not start your program running; it only
19289 changes the address of where it @emph{will} run when you continue. For
19297 makes the next @code{continue} command or stepping command execute at
19298 address @code{0x485}, rather than at the address where your program stopped.
19299 @xref{Continuing and Stepping, ,Continuing and Stepping}.
19301 The most common occasion to use the @code{jump} command is to back
19302 up---perhaps with more breakpoints set---over a portion of a program
19303 that has already executed, in order to examine its execution in more
19308 @section Giving your Program a Signal
19309 @cindex deliver a signal to a program
19313 @item signal @var{signal}
19314 Resume execution where your program is stopped, but immediately give it the
19315 signal @var{signal}. The @var{signal} can be the name or the number of a
19316 signal. For example, on many systems @code{signal 2} and @code{signal
19317 SIGINT} are both ways of sending an interrupt signal.
19319 Alternatively, if @var{signal} is zero, continue execution without
19320 giving a signal. This is useful when your program stopped on account of
19321 a signal and would ordinarily see the signal when resumed with the
19322 @code{continue} command; @samp{signal 0} causes it to resume without a
19325 @emph{Note:} When resuming a multi-threaded program, @var{signal} is
19326 delivered to the currently selected thread, not the thread that last
19327 reported a stop. This includes the situation where a thread was
19328 stopped due to a signal. So if you want to continue execution
19329 suppressing the signal that stopped a thread, you should select that
19330 same thread before issuing the @samp{signal 0} command. If you issue
19331 the @samp{signal 0} command with another thread as the selected one,
19332 @value{GDBN} detects that and asks for confirmation.
19334 Invoking the @code{signal} command is not the same as invoking the
19335 @code{kill} utility from the shell. Sending a signal with @code{kill}
19336 causes @value{GDBN} to decide what to do with the signal depending on
19337 the signal handling tables (@pxref{Signals}). The @code{signal} command
19338 passes the signal directly to your program.
19340 @code{signal} does not repeat when you press @key{RET} a second time
19341 after executing the command.
19343 @kindex queue-signal
19344 @item queue-signal @var{signal}
19345 Queue @var{signal} to be delivered immediately to the current thread
19346 when execution of the thread resumes. The @var{signal} can be the name or
19347 the number of a signal. For example, on many systems @code{signal 2} and
19348 @code{signal SIGINT} are both ways of sending an interrupt signal.
19349 The handling of the signal must be set to pass the signal to the program,
19350 otherwise @value{GDBN} will report an error.
19351 You can control the handling of signals from @value{GDBN} with the
19352 @code{handle} command (@pxref{Signals}).
19354 Alternatively, if @var{signal} is zero, any currently queued signal
19355 for the current thread is discarded and when execution resumes no signal
19356 will be delivered. This is useful when your program stopped on account
19357 of a signal and would ordinarily see the signal when resumed with the
19358 @code{continue} command.
19360 This command differs from the @code{signal} command in that the signal
19361 is just queued, execution is not resumed. And @code{queue-signal} cannot
19362 be used to pass a signal whose handling state has been set to @code{nopass}
19367 @xref{stepping into signal handlers}, for information on how stepping
19368 commands behave when the thread has a signal queued.
19371 @section Returning from a Function
19374 @cindex returning from a function
19377 @itemx return @var{expression}
19378 You can cancel execution of a function call with the @code{return}
19379 command. If you give an
19380 @var{expression} argument, its value is used as the function's return
19384 When you use @code{return}, @value{GDBN} discards the selected stack frame
19385 (and all frames within it). You can think of this as making the
19386 discarded frame return prematurely. If you wish to specify a value to
19387 be returned, give that value as the argument to @code{return}.
19389 This pops the selected stack frame (@pxref{Selection, ,Selecting a
19390 Frame}), and any other frames inside of it, leaving its caller as the
19391 innermost remaining frame. That frame becomes selected. The
19392 specified value is stored in the registers used for returning values
19395 The @code{return} command does not resume execution; it leaves the
19396 program stopped in the state that would exist if the function had just
19397 returned. In contrast, the @code{finish} command (@pxref{Continuing
19398 and Stepping, ,Continuing and Stepping}) resumes execution until the
19399 selected stack frame returns naturally.
19401 @value{GDBN} needs to know how the @var{expression} argument should be set for
19402 the inferior. The concrete registers assignment depends on the OS ABI and the
19403 type being returned by the selected stack frame. For example it is common for
19404 OS ABI to return floating point values in FPU registers while integer values in
19405 CPU registers. Still some ABIs return even floating point values in CPU
19406 registers. Larger integer widths (such as @code{long long int}) also have
19407 specific placement rules. @value{GDBN} already knows the OS ABI from its
19408 current target so it needs to find out also the type being returned to make the
19409 assignment into the right register(s).
19411 Normally, the selected stack frame has debug info. @value{GDBN} will always
19412 use the debug info instead of the implicit type of @var{expression} when the
19413 debug info is available. For example, if you type @kbd{return -1}, and the
19414 function in the current stack frame is declared to return a @code{long long
19415 int}, @value{GDBN} transparently converts the implicit @code{int} value of -1
19416 into a @code{long long int}:
19419 Breakpoint 1, func () at gdb.base/return-nodebug.c:29
19421 (@value{GDBP}) return -1
19422 Make func return now? (y or n) y
19423 #0 0x004004f6 in main () at gdb.base/return-nodebug.c:43
19424 43 printf ("result=%lld\n", func ());
19428 However, if the selected stack frame does not have a debug info, e.g., if the
19429 function was compiled without debug info, @value{GDBN} has to find out the type
19430 to return from user. Specifying a different type by mistake may set the value
19431 in different inferior registers than the caller code expects. For example,
19432 typing @kbd{return -1} with its implicit type @code{int} would set only a part
19433 of a @code{long long int} result for a debug info less function (on 32-bit
19434 architectures). Therefore the user is required to specify the return type by
19435 an appropriate cast explicitly:
19438 Breakpoint 2, 0x0040050b in func ()
19439 (@value{GDBP}) return -1
19440 Return value type not available for selected stack frame.
19441 Please use an explicit cast of the value to return.
19442 (@value{GDBP}) return (long long int) -1
19443 Make selected stack frame return now? (y or n) y
19444 #0 0x00400526 in main ()
19449 @section Calling Program Functions
19452 @cindex calling functions
19453 @cindex inferior functions, calling
19454 @item print @var{expr}
19455 Evaluate the expression @var{expr} and display the resulting value.
19456 The expression may include calls to functions in the program being
19460 @item call @var{expr}
19461 Evaluate the expression @var{expr} without displaying @code{void}
19464 You can use this variant of the @code{print} command if you want to
19465 execute a function from your program that does not return anything
19466 (a.k.a.@: @dfn{a void function}), but without cluttering the output
19467 with @code{void} returned values that @value{GDBN} will otherwise
19468 print. If the result is not void, it is printed and saved in the
19472 It is possible for the function you call via the @code{print} or
19473 @code{call} command to generate a signal (e.g., if there's a bug in
19474 the function, or if you passed it incorrect arguments). What happens
19475 in that case is controlled by the @code{set unwindonsignal} command.
19477 Similarly, with a C@t{++} program it is possible for the function you
19478 call via the @code{print} or @code{call} command to generate an
19479 exception that is not handled due to the constraints of the dummy
19480 frame. In this case, any exception that is raised in the frame, but has
19481 an out-of-frame exception handler will not be found. GDB builds a
19482 dummy-frame for the inferior function call, and the unwinder cannot
19483 seek for exception handlers outside of this dummy-frame. What happens
19484 in that case is controlled by the
19485 @code{set unwind-on-terminating-exception} command.
19488 @item set unwindonsignal
19489 @kindex set unwindonsignal
19490 @cindex unwind stack in called functions
19491 @cindex call dummy stack unwinding
19492 Set unwinding of the stack if a signal is received while in a function
19493 that @value{GDBN} called in the program being debugged. If set to on,
19494 @value{GDBN} unwinds the stack it created for the call and restores
19495 the context to what it was before the call. If set to off (the
19496 default), @value{GDBN} stops in the frame where the signal was
19499 @item show unwindonsignal
19500 @kindex show unwindonsignal
19501 Show the current setting of stack unwinding in the functions called by
19504 @item set unwind-on-terminating-exception
19505 @kindex set unwind-on-terminating-exception
19506 @cindex unwind stack in called functions with unhandled exceptions
19507 @cindex call dummy stack unwinding on unhandled exception.
19508 Set unwinding of the stack if a C@t{++} exception is raised, but left
19509 unhandled while in a function that @value{GDBN} called in the program being
19510 debugged. If set to on (the default), @value{GDBN} unwinds the stack
19511 it created for the call and restores the context to what it was before
19512 the call. If set to off, @value{GDBN} the exception is delivered to
19513 the default C@t{++} exception handler and the inferior terminated.
19515 @item show unwind-on-terminating-exception
19516 @kindex show unwind-on-terminating-exception
19517 Show the current setting of stack unwinding in the functions called by
19520 @item set may-call-functions
19521 @kindex set may-call-functions
19522 @cindex disabling calling functions in the program
19523 @cindex calling functions in the program, disabling
19524 Set permission to call functions in the program.
19525 This controls whether @value{GDBN} will attempt to call functions in
19526 the program, such as with expressions in the @code{print} command. It
19527 defaults to @code{on}.
19529 To call a function in the program, @value{GDBN} has to temporarily
19530 modify the state of the inferior. This has potentially undesired side
19531 effects. Also, having @value{GDBN} call nested functions is likely to
19532 be erroneous and may even crash the program being debugged. You can
19533 avoid such hazards by forbidding @value{GDBN} from calling functions
19534 in the program being debugged. If calling functions in the program
19535 is forbidden, GDB will throw an error when a command (such as printing
19536 an expression) starts a function call in the program.
19538 @item show may-call-functions
19539 @kindex show may-call-functions
19540 Show permission to call functions in the program.
19544 @subsection Calling functions with no debug info
19546 @cindex no debug info functions
19547 Sometimes, a function you wish to call is missing debug information.
19548 In such case, @value{GDBN} does not know the type of the function,
19549 including the types of the function's parameters. To avoid calling
19550 the inferior function incorrectly, which could result in the called
19551 function functioning erroneously and even crash, @value{GDBN} refuses
19552 to call the function unless you tell it the type of the function.
19554 For prototyped (i.e.@: ANSI/ISO style) functions, there are two ways
19555 to do that. The simplest is to cast the call to the function's
19556 declared return type. For example:
19559 (@value{GDBP}) p getenv ("PATH")
19560 'getenv' has unknown return type; cast the call to its declared return type
19561 (@value{GDBP}) p (char *) getenv ("PATH")
19562 $1 = 0x7fffffffe7ba "/usr/local/bin:/"...
19565 Casting the return type of a no-debug function is equivalent to
19566 casting the function to a pointer to a prototyped function that has a
19567 prototype that matches the types of the passed-in arguments, and
19568 calling that. I.e., the call above is equivalent to:
19571 (@value{GDBP}) p ((char * (*) (const char *)) getenv) ("PATH")
19575 and given this prototyped C or C++ function with float parameters:
19578 float multiply (float v1, float v2) @{ return v1 * v2; @}
19582 these calls are equivalent:
19585 (@value{GDBP}) p (float) multiply (2.0f, 3.0f)
19586 (@value{GDBP}) p ((float (*) (float, float)) multiply) (2.0f, 3.0f)
19589 If the function you wish to call is declared as unprototyped (i.e.@:
19590 old K&R style), you must use the cast-to-function-pointer syntax, so
19591 that @value{GDBN} knows that it needs to apply default argument
19592 promotions (promote float arguments to double). @xref{ABI, float
19593 promotion}. For example, given this unprototyped C function with
19594 float parameters, and no debug info:
19598 multiply_noproto (v1, v2)
19606 you call it like this:
19609 (@value{GDBP}) p ((float (*) ()) multiply_noproto) (2.0f, 3.0f)
19613 @section Patching Programs
19615 @cindex patching binaries
19616 @cindex writing into executables
19617 @cindex writing into corefiles
19619 By default, @value{GDBN} opens the file containing your program's
19620 executable code (or the corefile) read-only. This prevents accidental
19621 alterations to machine code; but it also prevents you from intentionally
19622 patching your program's binary.
19624 If you'd like to be able to patch the binary, you can specify that
19625 explicitly with the @code{set write} command. For example, you might
19626 want to turn on internal debugging flags, or even to make emergency
19632 @itemx set write off
19633 If you specify @samp{set write on}, @value{GDBN} opens executable and
19634 core files for both reading and writing; if you specify @kbd{set write
19635 off} (the default), @value{GDBN} opens them read-only.
19637 If you have already loaded a file, you must load it again (using the
19638 @code{exec-file} or @code{core-file} command) after changing @code{set
19639 write}, for your new setting to take effect.
19643 Display whether executable files and core files are opened for writing
19644 as well as reading.
19647 @node Compiling and Injecting Code
19648 @section Compiling and injecting code in @value{GDBN}
19649 @cindex injecting code
19650 @cindex writing into executables
19651 @cindex compiling code
19653 @value{GDBN} supports on-demand compilation and code injection into
19654 programs running under @value{GDBN}. GCC 5.0 or higher built with
19655 @file{libcc1.so} must be installed for this functionality to be enabled.
19656 This functionality is implemented with the following commands.
19659 @kindex compile code
19660 @item compile code @var{source-code}
19661 @itemx compile code -raw @var{--} @var{source-code}
19662 Compile @var{source-code} with the compiler language found as the current
19663 language in @value{GDBN} (@pxref{Languages}). If compilation and
19664 injection is not supported with the current language specified in
19665 @value{GDBN}, or the compiler does not support this feature, an error
19666 message will be printed. If @var{source-code} compiles and links
19667 successfully, @value{GDBN} will load the object-code emitted,
19668 and execute it within the context of the currently selected inferior.
19669 It is important to note that the compiled code is executed immediately.
19670 After execution, the compiled code is removed from @value{GDBN} and any
19671 new types or variables you have defined will be deleted.
19673 The command allows you to specify @var{source-code} in two ways.
19674 The simplest method is to provide a single line of code to the command.
19678 compile code printf ("hello world\n");
19681 If you specify options on the command line as well as source code, they
19682 may conflict. The @samp{--} delimiter can be used to separate options
19683 from actual source code. E.g.:
19686 compile code -r -- printf ("hello world\n");
19689 Alternatively you can enter source code as multiple lines of text. To
19690 enter this mode, invoke the @samp{compile code} command without any text
19691 following the command. This will start the multiple-line editor and
19692 allow you to type as many lines of source code as required. When you
19693 have completed typing, enter @samp{end} on its own line to exit the
19698 >printf ("hello\n");
19699 >printf ("world\n");
19703 Specifying @samp{-raw}, prohibits @value{GDBN} from wrapping the
19704 provided @var{source-code} in a callable scope. In this case, you must
19705 specify the entry point of the code by defining a function named
19706 @code{_gdb_expr_}. The @samp{-raw} code cannot access variables of the
19707 inferior. Using @samp{-raw} option may be needed for example when
19708 @var{source-code} requires @samp{#include} lines which may conflict with
19709 inferior symbols otherwise.
19711 @kindex compile file
19712 @item compile file @var{filename}
19713 @itemx compile file -raw @var{filename}
19714 Like @code{compile code}, but take the source code from @var{filename}.
19717 compile file /home/user/example.c
19722 @item compile print [[@var{options}] --] @var{expr}
19723 @itemx compile print [[@var{options}] --] /@var{f} @var{expr}
19724 Compile and execute @var{expr} with the compiler language found as the
19725 current language in @value{GDBN} (@pxref{Languages}). By default the
19726 value of @var{expr} is printed in a format appropriate to its data type;
19727 you can choose a different format by specifying @samp{/@var{f}}, where
19728 @var{f} is a letter specifying the format; see @ref{Output Formats,,Output
19729 Formats}. The @code{compile print} command accepts the same options
19730 as the @code{print} command; see @ref{print options}.
19732 @item compile print [[@var{options}] --]
19733 @itemx compile print [[@var{options}] --] /@var{f}
19734 @cindex reprint the last value
19735 Alternatively you can enter the expression (source code producing it) as
19736 multiple lines of text. To enter this mode, invoke the @samp{compile print}
19737 command without any text following the command. This will start the
19738 multiple-line editor.
19742 The process of compiling and injecting the code can be inspected using:
19745 @anchor{set debug compile}
19746 @item set debug compile
19747 @cindex compile command debugging info
19748 Turns on or off display of @value{GDBN} process of compiling and
19749 injecting the code. The default is off.
19751 @item show debug compile
19752 Displays the current state of displaying @value{GDBN} process of
19753 compiling and injecting the code.
19755 @anchor{set debug compile-cplus-types}
19756 @item set debug compile-cplus-types
19757 @cindex compile C@t{++} type conversion
19758 Turns on or off the display of C@t{++} type conversion debugging information.
19759 The default is off.
19761 @item show debug compile-cplus-types
19762 Displays the current state of displaying debugging information for
19763 C@t{++} type conversion.
19766 @subsection Compilation options for the @code{compile} command
19768 @value{GDBN} needs to specify the right compilation options for the code
19769 to be injected, in part to make its ABI compatible with the inferior
19770 and in part to make the injected code compatible with @value{GDBN}'s
19774 The options used, in increasing precedence:
19777 @item target architecture and OS options (@code{gdbarch})
19778 These options depend on target processor type and target operating
19779 system, usually they specify at least 32-bit (@code{-m32}) or 64-bit
19780 (@code{-m64}) compilation option.
19782 @item compilation options recorded in the target
19783 @value{NGCC} (since version 4.7) stores the options used for compilation
19784 into @code{DW_AT_producer} part of DWARF debugging information according
19785 to the @value{NGCC} option @code{-grecord-gcc-switches}. One has to
19786 explicitly specify @code{-g} during inferior compilation otherwise
19787 @value{NGCC} produces no DWARF. This feature is only relevant for
19788 platforms where @code{-g} produces DWARF by default, otherwise one may
19789 try to enforce DWARF by using @code{-gdwarf-4}.
19791 @item compilation options set by @code{set compile-args}
19795 You can override compilation options using the following command:
19798 @item set compile-args
19799 @cindex compile command options override
19800 Set compilation options used for compiling and injecting code with the
19801 @code{compile} commands. These options override any conflicting ones
19802 from the target architecture and/or options stored during inferior
19805 @item show compile-args
19806 Displays the current state of compilation options override.
19807 This does not show all the options actually used during compilation,
19808 use @ref{set debug compile} for that.
19811 @subsection Caveats when using the @code{compile} command
19813 There are a few caveats to keep in mind when using the @code{compile}
19814 command. As the caveats are different per language, the table below
19815 highlights specific issues on a per language basis.
19818 @item C code examples and caveats
19819 When the language in @value{GDBN} is set to @samp{C}, the compiler will
19820 attempt to compile the source code with a @samp{C} compiler. The source
19821 code provided to the @code{compile} command will have much the same
19822 access to variables and types as it normally would if it were part of
19823 the program currently being debugged in @value{GDBN}.
19825 Below is a sample program that forms the basis of the examples that
19826 follow. This program has been compiled and loaded into @value{GDBN},
19827 much like any other normal debugging session.
19830 void function1 (void)
19833 printf ("function 1\n");
19836 void function2 (void)
19851 For the purposes of the examples in this section, the program above has
19852 been compiled, loaded into @value{GDBN}, stopped at the function
19853 @code{main}, and @value{GDBN} is awaiting input from the user.
19855 To access variables and types for any program in @value{GDBN}, the
19856 program must be compiled and packaged with debug information. The
19857 @code{compile} command is not an exception to this rule. Without debug
19858 information, you can still use the @code{compile} command, but you will
19859 be very limited in what variables and types you can access.
19861 So with that in mind, the example above has been compiled with debug
19862 information enabled. The @code{compile} command will have access to
19863 all variables and types (except those that may have been optimized
19864 out). Currently, as @value{GDBN} has stopped the program in the
19865 @code{main} function, the @code{compile} command would have access to
19866 the variable @code{k}. You could invoke the @code{compile} command
19867 and type some source code to set the value of @code{k}. You can also
19868 read it, or do anything with that variable you would normally do in
19869 @code{C}. Be aware that changes to inferior variables in the
19870 @code{compile} command are persistent. In the following example:
19873 compile code k = 3;
19877 the variable @code{k} is now 3. It will retain that value until
19878 something else in the example program changes it, or another
19879 @code{compile} command changes it.
19881 Normal scope and access rules apply to source code compiled and
19882 injected by the @code{compile} command. In the example, the variables
19883 @code{j} and @code{k} are not accessible yet, because the program is
19884 currently stopped in the @code{main} function, where these variables
19885 are not in scope. Therefore, the following command
19888 compile code j = 3;
19892 will result in a compilation error message.
19894 Once the program is continued, execution will bring these variables in
19895 scope, and they will become accessible; then the code you specify via
19896 the @code{compile} command will be able to access them.
19898 You can create variables and types with the @code{compile} command as
19899 part of your source code. Variables and types that are created as part
19900 of the @code{compile} command are not visible to the rest of the program for
19901 the duration of its run. This example is valid:
19904 compile code int ff = 5; printf ("ff is %d\n", ff);
19907 However, if you were to type the following into @value{GDBN} after that
19908 command has completed:
19911 compile code printf ("ff is %d\n'', ff);
19915 a compiler error would be raised as the variable @code{ff} no longer
19916 exists. Object code generated and injected by the @code{compile}
19917 command is removed when its execution ends. Caution is advised
19918 when assigning to program variables values of variables created by the
19919 code submitted to the @code{compile} command. This example is valid:
19922 compile code int ff = 5; k = ff;
19925 The value of the variable @code{ff} is assigned to @code{k}. The variable
19926 @code{k} does not require the existence of @code{ff} to maintain the value
19927 it has been assigned. However, pointers require particular care in
19928 assignment. If the source code compiled with the @code{compile} command
19929 changed the address of a pointer in the example program, perhaps to a
19930 variable created in the @code{compile} command, that pointer would point
19931 to an invalid location when the command exits. The following example
19932 would likely cause issues with your debugged program:
19935 compile code int ff = 5; p = &ff;
19938 In this example, @code{p} would point to @code{ff} when the
19939 @code{compile} command is executing the source code provided to it.
19940 However, as variables in the (example) program persist with their
19941 assigned values, the variable @code{p} would point to an invalid
19942 location when the command exists. A general rule should be followed
19943 in that you should either assign @code{NULL} to any assigned pointers,
19944 or restore a valid location to the pointer before the command exits.
19946 Similar caution must be exercised with any structs, unions, and typedefs
19947 defined in @code{compile} command. Types defined in the @code{compile}
19948 command will no longer be available in the next @code{compile} command.
19949 Therefore, if you cast a variable to a type defined in the
19950 @code{compile} command, care must be taken to ensure that any future
19951 need to resolve the type can be achieved.
19954 (gdb) compile code static struct a @{ int a; @} v = @{ 42 @}; argv = &v;
19955 (gdb) compile code printf ("%d\n", ((struct a *) argv)->a);
19956 gdb command line:1:36: error: dereferencing pointer to incomplete type ‘struct a’
19957 Compilation failed.
19958 (gdb) compile code struct a @{ int a; @}; printf ("%d\n", ((struct a *) argv)->a);
19962 Variables that have been optimized away by the compiler are not
19963 accessible to the code submitted to the @code{compile} command.
19964 Access to those variables will generate a compiler error which @value{GDBN}
19965 will print to the console.
19968 @subsection Compiler search for the @code{compile} command
19970 @value{GDBN} needs to find @value{NGCC} for the inferior being debugged
19971 which may not be obvious for remote targets of different architecture
19972 than where @value{GDBN} is running. Environment variable @code{PATH} on
19973 @value{GDBN} host is searched for @value{NGCC} binary matching the
19974 target architecture and operating system. This search can be overriden
19975 by @code{set compile-gcc} @value{GDBN} command below. @code{PATH} is
19976 taken from shell that executed @value{GDBN}, it is not the value set by
19977 @value{GDBN} command @code{set environment}). @xref{Environment}.
19980 Specifically @code{PATH} is searched for binaries matching regular expression
19981 @code{@var{arch}(-[^-]*)?-@var{os}-gcc} according to the inferior target being
19982 debugged. @var{arch} is processor name --- multiarch is supported, so for
19983 example both @code{i386} and @code{x86_64} targets look for pattern
19984 @code{(x86_64|i.86)} and both @code{s390} and @code{s390x} targets look
19985 for pattern @code{s390x?}. @var{os} is currently supported only for
19986 pattern @code{linux(-gnu)?}.
19988 On Posix hosts the compiler driver @value{GDBN} needs to find also
19989 shared library @file{libcc1.so} from the compiler. It is searched in
19990 default shared library search path (overridable with usual environment
19991 variable @code{LD_LIBRARY_PATH}), unrelated to @code{PATH} or @code{set
19992 compile-gcc} settings. Contrary to it @file{libcc1plugin.so} is found
19993 according to the installation of the found compiler --- as possibly
19994 specified by the @code{set compile-gcc} command.
19997 @item set compile-gcc
19998 @cindex compile command driver filename override
19999 Set compilation command used for compiling and injecting code with the
20000 @code{compile} commands. If this option is not set (it is set to
20001 an empty string), the search described above will occur --- that is the
20004 @item show compile-gcc
20005 Displays the current compile command @value{NGCC} driver filename.
20006 If set, it is the main command @command{gcc}, found usually for example
20007 under name @file{x86_64-linux-gnu-gcc}.
20011 @chapter @value{GDBN} Files
20013 @value{GDBN} needs to know the file name of the program to be debugged,
20014 both in order to read its symbol table and in order to start your
20015 program. To debug a core dump of a previous run, you must also tell
20016 @value{GDBN} the name of the core dump file.
20019 * Files:: Commands to specify files
20020 * File Caching:: Information about @value{GDBN}'s file caching
20021 * Separate Debug Files:: Debugging information in separate files
20022 * MiniDebugInfo:: Debugging information in a special section
20023 * Index Files:: Index files speed up GDB
20024 * Symbol Errors:: Errors reading symbol files
20025 * Data Files:: GDB data files
20029 @section Commands to Specify Files
20031 @cindex symbol table
20032 @cindex core dump file
20034 You may want to specify executable and core dump file names. The usual
20035 way to do this is at start-up time, using the arguments to
20036 @value{GDBN}'s start-up commands (@pxref{Invocation, , Getting In and
20037 Out of @value{GDBN}}).
20039 Occasionally it is necessary to change to a different file during a
20040 @value{GDBN} session. Or you may run @value{GDBN} and forget to
20041 specify a file you want to use. Or you are debugging a remote target
20042 via @code{gdbserver} (@pxref{Server, file, Using the @code{gdbserver}
20043 Program}). In these situations the @value{GDBN} commands to specify
20044 new files are useful.
20047 @cindex executable file
20049 @item file @var{filename}
20050 Use @var{filename} as the program to be debugged. It is read for its
20051 symbols and for the contents of pure memory. It is also the program
20052 executed when you use the @code{run} command. If you do not specify a
20053 directory and the file is not found in the @value{GDBN} working directory,
20054 @value{GDBN} uses the environment variable @code{PATH} as a list of
20055 directories to search, just as the shell does when looking for a program
20056 to run. You can change the value of this variable, for both @value{GDBN}
20057 and your program, using the @code{path} command.
20059 @cindex unlinked object files
20060 @cindex patching object files
20061 You can load unlinked object @file{.o} files into @value{GDBN} using
20062 the @code{file} command. You will not be able to ``run'' an object
20063 file, but you can disassemble functions and inspect variables. Also,
20064 if the underlying BFD functionality supports it, you could use
20065 @kbd{gdb -write} to patch object files using this technique. Note
20066 that @value{GDBN} can neither interpret nor modify relocations in this
20067 case, so branches and some initialized variables will appear to go to
20068 the wrong place. But this feature is still handy from time to time.
20071 @code{file} with no argument makes @value{GDBN} discard any information it
20072 has on both executable file and the symbol table.
20075 @item exec-file @r{[} @var{filename} @r{]}
20076 Specify that the program to be run (but not the symbol table) is found
20077 in @var{filename}. @value{GDBN} searches the environment variable @code{PATH}
20078 if necessary to locate your program. Omitting @var{filename} means to
20079 discard information on the executable file.
20081 @kindex symbol-file
20082 @item symbol-file @r{[} @var{filename} @r{[} -o @var{offset} @r{]]}
20083 Read symbol table information from file @var{filename}. @code{PATH} is
20084 searched when necessary. Use the @code{file} command to get both symbol
20085 table and program to run from the same file.
20087 If an optional @var{offset} is specified, it is added to the start
20088 address of each section in the symbol file. This is useful if the
20089 program is relocated at runtime, such as the Linux kernel with kASLR
20092 @code{symbol-file} with no argument clears out @value{GDBN} information on your
20093 program's symbol table.
20095 The @code{symbol-file} command causes @value{GDBN} to forget the contents of
20096 some breakpoints and auto-display expressions. This is because they may
20097 contain pointers to the internal data recording symbols and data types,
20098 which are part of the old symbol table data being discarded inside
20101 @code{symbol-file} does not repeat if you press @key{RET} again after
20104 When @value{GDBN} is configured for a particular environment, it
20105 understands debugging information in whatever format is the standard
20106 generated for that environment; you may use either a @sc{gnu} compiler, or
20107 other compilers that adhere to the local conventions.
20108 Best results are usually obtained from @sc{gnu} compilers; for example,
20109 using @code{@value{NGCC}} you can generate debugging information for
20112 For most kinds of object files, with the exception of old SVR3 systems
20113 using COFF, the @code{symbol-file} command does not normally read the
20114 symbol table in full right away. Instead, it scans the symbol table
20115 quickly to find which source files and which symbols are present. The
20116 details are read later, one source file at a time, as they are needed.
20118 The purpose of this two-stage reading strategy is to make @value{GDBN}
20119 start up faster. For the most part, it is invisible except for
20120 occasional pauses while the symbol table details for a particular source
20121 file are being read. (The @code{set verbose} command can turn these
20122 pauses into messages if desired. @xref{Messages/Warnings, ,Optional
20123 Warnings and Messages}.)
20125 We have not implemented the two-stage strategy for COFF yet. When the
20126 symbol table is stored in COFF format, @code{symbol-file} reads the
20127 symbol table data in full right away. Note that ``stabs-in-COFF''
20128 still does the two-stage strategy, since the debug info is actually
20132 @cindex reading symbols immediately
20133 @cindex symbols, reading immediately
20134 @item symbol-file @r{[} -readnow @r{]} @var{filename}
20135 @itemx file @r{[} -readnow @r{]} @var{filename}
20136 You can override the @value{GDBN} two-stage strategy for reading symbol
20137 tables by using the @samp{-readnow} option with any of the commands that
20138 load symbol table information, if you want to be sure @value{GDBN} has the
20139 entire symbol table available.
20141 @cindex @code{-readnever}, option for symbol-file command
20142 @cindex never read symbols
20143 @cindex symbols, never read
20144 @item symbol-file @r{[} -readnever @r{]} @var{filename}
20145 @itemx file @r{[} -readnever @r{]} @var{filename}
20146 You can instruct @value{GDBN} to never read the symbolic information
20147 contained in @var{filename} by using the @samp{-readnever} option.
20148 @xref{--readnever}.
20150 @c FIXME: for now no mention of directories, since this seems to be in
20151 @c flux. 13mar1992 status is that in theory GDB would look either in
20152 @c current dir or in same dir as myprog; but issues like competing
20153 @c GDB's, or clutter in system dirs, mean that in practice right now
20154 @c only current dir is used. FFish says maybe a special GDB hierarchy
20155 @c (eg rooted in val of env var GDBSYMS) could exist for mappable symbol
20159 @item core-file @r{[}@var{filename}@r{]}
20161 Specify the whereabouts of a core dump file to be used as the ``contents
20162 of memory''. Traditionally, core files contain only some parts of the
20163 address space of the process that generated them; @value{GDBN} can access the
20164 executable file itself for other parts.
20166 @code{core-file} with no argument specifies that no core file is
20169 Note that the core file is ignored when your program is actually running
20170 under @value{GDBN}. So, if you have been running your program and you
20171 wish to debug a core file instead, you must kill the subprocess in which
20172 the program is running. To do this, use the @code{kill} command
20173 (@pxref{Kill Process, ,Killing the Child Process}).
20175 @kindex add-symbol-file
20176 @cindex dynamic linking
20177 @item add-symbol-file @var{filename} @r{[} -readnow @r{|} -readnever @r{]} @r{[} -o @var{offset} @r{]} @r{[} @var{textaddress} @r{]} @r{[} -s @var{section} @var{address} @dots{} @r{]}
20178 The @code{add-symbol-file} command reads additional symbol table
20179 information from the file @var{filename}. You would use this command
20180 when @var{filename} has been dynamically loaded (by some other means)
20181 into the program that is running. The @var{textaddress} parameter gives
20182 the memory address at which the file's text section has been loaded.
20183 You can additionally specify the base address of other sections using
20184 an arbitrary number of @samp{-s @var{section} @var{address}} pairs.
20185 If a section is omitted, @value{GDBN} will use its default addresses
20186 as found in @var{filename}. Any @var{address} or @var{textaddress}
20187 can be given as an expression.
20189 If an optional @var{offset} is specified, it is added to the start
20190 address of each section, except those for which the address was
20191 specified explicitly.
20193 The symbol table of the file @var{filename} is added to the symbol table
20194 originally read with the @code{symbol-file} command. You can use the
20195 @code{add-symbol-file} command any number of times; the new symbol data
20196 thus read is kept in addition to the old.
20198 Changes can be reverted using the command @code{remove-symbol-file}.
20200 @cindex relocatable object files, reading symbols from
20201 @cindex object files, relocatable, reading symbols from
20202 @cindex reading symbols from relocatable object files
20203 @cindex symbols, reading from relocatable object files
20204 @cindex @file{.o} files, reading symbols from
20205 Although @var{filename} is typically a shared library file, an
20206 executable file, or some other object file which has been fully
20207 relocated for loading into a process, you can also load symbolic
20208 information from relocatable @file{.o} files, as long as:
20212 the file's symbolic information refers only to linker symbols defined in
20213 that file, not to symbols defined by other object files,
20215 every section the file's symbolic information refers to has actually
20216 been loaded into the inferior, as it appears in the file, and
20218 you can determine the address at which every section was loaded, and
20219 provide these to the @code{add-symbol-file} command.
20223 Some embedded operating systems, like Sun Chorus and VxWorks, can load
20224 relocatable files into an already running program; such systems
20225 typically make the requirements above easy to meet. However, it's
20226 important to recognize that many native systems use complex link
20227 procedures (@code{.linkonce} section factoring and C@t{++} constructor table
20228 assembly, for example) that make the requirements difficult to meet. In
20229 general, one cannot assume that using @code{add-symbol-file} to read a
20230 relocatable object file's symbolic information will have the same effect
20231 as linking the relocatable object file into the program in the normal
20234 @code{add-symbol-file} does not repeat if you press @key{RET} after using it.
20236 @kindex remove-symbol-file
20237 @item remove-symbol-file @var{filename}
20238 @item remove-symbol-file -a @var{address}
20239 Remove a symbol file added via the @code{add-symbol-file} command. The
20240 file to remove can be identified by its @var{filename} or by an @var{address}
20241 that lies within the boundaries of this symbol file in memory. Example:
20244 (gdb) add-symbol-file /home/user/gdb/mylib.so 0x7ffff7ff9480
20245 add symbol table from file "/home/user/gdb/mylib.so" at
20246 .text_addr = 0x7ffff7ff9480
20248 Reading symbols from /home/user/gdb/mylib.so...done.
20249 (gdb) remove-symbol-file -a 0x7ffff7ff9480
20250 Remove symbol table from file "/home/user/gdb/mylib.so"? (y or n) y
20255 @code{remove-symbol-file} does not repeat if you press @key{RET} after using it.
20257 @kindex add-symbol-file-from-memory
20258 @cindex @code{syscall DSO}
20259 @cindex load symbols from memory
20260 @item add-symbol-file-from-memory @var{address}
20261 Load symbols from the given @var{address} in a dynamically loaded
20262 object file whose image is mapped directly into the inferior's memory.
20263 For example, the Linux kernel maps a @code{syscall DSO} into each
20264 process's address space; this DSO provides kernel-specific code for
20265 some system calls. The argument can be any expression whose
20266 evaluation yields the address of the file's shared object file header.
20267 For this command to work, you must have used @code{symbol-file} or
20268 @code{exec-file} commands in advance.
20271 @item section @var{section} @var{addr}
20272 The @code{section} command changes the base address of the named
20273 @var{section} of the exec file to @var{addr}. This can be used if the
20274 exec file does not contain section addresses, (such as in the
20275 @code{a.out} format), or when the addresses specified in the file
20276 itself are wrong. Each section must be changed separately. The
20277 @code{info files} command, described below, lists all the sections and
20281 @kindex info target
20284 @code{info files} and @code{info target} are synonymous; both print the
20285 current target (@pxref{Targets, ,Specifying a Debugging Target}),
20286 including the names of the executable and core dump files currently in
20287 use by @value{GDBN}, and the files from which symbols were loaded. The
20288 command @code{help target} lists all possible targets rather than
20291 @kindex maint info sections
20292 @item maint info sections
20293 Another command that can give you extra information about program sections
20294 is @code{maint info sections}. In addition to the section information
20295 displayed by @code{info files}, this command displays the flags and file
20296 offset of each section in the executable and core dump files. In addition,
20297 @code{maint info sections} provides the following command options (which
20298 may be arbitrarily combined):
20302 Display sections for all loaded object files, including shared libraries.
20303 @item @var{sections}
20304 Display info only for named @var{sections}.
20305 @item @var{section-flags}
20306 Display info only for sections for which @var{section-flags} are true.
20307 The section flags that @value{GDBN} currently knows about are:
20310 Section will have space allocated in the process when loaded.
20311 Set for all sections except those containing debug information.
20313 Section will be loaded from the file into the child process memory.
20314 Set for pre-initialized code and data, clear for @code{.bss} sections.
20316 Section needs to be relocated before loading.
20318 Section cannot be modified by the child process.
20320 Section contains executable code only.
20322 Section contains data only (no executable code).
20324 Section will reside in ROM.
20326 Section contains data for constructor/destructor lists.
20328 Section is not empty.
20330 An instruction to the linker to not output the section.
20331 @item COFF_SHARED_LIBRARY
20332 A notification to the linker that the section contains
20333 COFF shared library information.
20335 Section contains common symbols.
20338 @kindex set trust-readonly-sections
20339 @cindex read-only sections
20340 @item set trust-readonly-sections on
20341 Tell @value{GDBN} that readonly sections in your object file
20342 really are read-only (i.e.@: that their contents will not change).
20343 In that case, @value{GDBN} can fetch values from these sections
20344 out of the object file, rather than from the target program.
20345 For some targets (notably embedded ones), this can be a significant
20346 enhancement to debugging performance.
20348 The default is off.
20350 @item set trust-readonly-sections off
20351 Tell @value{GDBN} not to trust readonly sections. This means that
20352 the contents of the section might change while the program is running,
20353 and must therefore be fetched from the target when needed.
20355 @item show trust-readonly-sections
20356 Show the current setting of trusting readonly sections.
20359 All file-specifying commands allow both absolute and relative file names
20360 as arguments. @value{GDBN} always converts the file name to an absolute file
20361 name and remembers it that way.
20363 @cindex shared libraries
20364 @anchor{Shared Libraries}
20365 @value{GDBN} supports @sc{gnu}/Linux, MS-Windows, SunOS,
20366 Darwin/Mach-O, SVr4, IBM RS/6000 AIX, QNX Neutrino, FDPIC (FR-V), and
20367 DSBT (TIC6X) shared libraries.
20369 On MS-Windows @value{GDBN} must be linked with the Expat library to support
20370 shared libraries. @xref{Expat}.
20372 @value{GDBN} automatically loads symbol definitions from shared libraries
20373 when you use the @code{run} command, or when you examine a core file.
20374 (Before you issue the @code{run} command, @value{GDBN} does not understand
20375 references to a function in a shared library, however---unless you are
20376 debugging a core file).
20378 @c FIXME: some @value{GDBN} release may permit some refs to undef
20379 @c FIXME...symbols---eg in a break cmd---assuming they are from a shared
20380 @c FIXME...lib; check this from time to time when updating manual
20382 There are times, however, when you may wish to not automatically load
20383 symbol definitions from shared libraries, such as when they are
20384 particularly large or there are many of them.
20386 To control the automatic loading of shared library symbols, use the
20390 @kindex set auto-solib-add
20391 @item set auto-solib-add @var{mode}
20392 If @var{mode} is @code{on}, symbols from all shared object libraries
20393 will be loaded automatically when the inferior begins execution, you
20394 attach to an independently started inferior, or when the dynamic linker
20395 informs @value{GDBN} that a new library has been loaded. If @var{mode}
20396 is @code{off}, symbols must be loaded manually, using the
20397 @code{sharedlibrary} command. The default value is @code{on}.
20399 @cindex memory used for symbol tables
20400 If your program uses lots of shared libraries with debug info that
20401 takes large amounts of memory, you can decrease the @value{GDBN}
20402 memory footprint by preventing it from automatically loading the
20403 symbols from shared libraries. To that end, type @kbd{set
20404 auto-solib-add off} before running the inferior, then load each
20405 library whose debug symbols you do need with @kbd{sharedlibrary
20406 @var{regexp}}, where @var{regexp} is a regular expression that matches
20407 the libraries whose symbols you want to be loaded.
20409 @kindex show auto-solib-add
20410 @item show auto-solib-add
20411 Display the current autoloading mode.
20414 @cindex load shared library
20415 To explicitly load shared library symbols, use the @code{sharedlibrary}
20419 @kindex info sharedlibrary
20421 @item info share @var{regex}
20422 @itemx info sharedlibrary @var{regex}
20423 Print the names of the shared libraries which are currently loaded
20424 that match @var{regex}. If @var{regex} is omitted then print
20425 all shared libraries that are loaded.
20428 @item info dll @var{regex}
20429 This is an alias of @code{info sharedlibrary}.
20431 @kindex sharedlibrary
20433 @item sharedlibrary @var{regex}
20434 @itemx share @var{regex}
20435 Load shared object library symbols for files matching a
20436 Unix regular expression.
20437 As with files loaded automatically, it only loads shared libraries
20438 required by your program for a core file or after typing @code{run}. If
20439 @var{regex} is omitted all shared libraries required by your program are
20442 @item nosharedlibrary
20443 @kindex nosharedlibrary
20444 @cindex unload symbols from shared libraries
20445 Unload all shared object library symbols. This discards all symbols
20446 that have been loaded from all shared libraries. Symbols from shared
20447 libraries that were loaded by explicit user requests are not
20451 Sometimes you may wish that @value{GDBN} stops and gives you control
20452 when any of shared library events happen. The best way to do this is
20453 to use @code{catch load} and @code{catch unload} (@pxref{Set
20456 @value{GDBN} also supports the the @code{set stop-on-solib-events}
20457 command for this. This command exists for historical reasons. It is
20458 less useful than setting a catchpoint, because it does not allow for
20459 conditions or commands as a catchpoint does.
20462 @item set stop-on-solib-events
20463 @kindex set stop-on-solib-events
20464 This command controls whether @value{GDBN} should give you control
20465 when the dynamic linker notifies it about some shared library event.
20466 The most common event of interest is loading or unloading of a new
20469 @item show stop-on-solib-events
20470 @kindex show stop-on-solib-events
20471 Show whether @value{GDBN} stops and gives you control when shared
20472 library events happen.
20475 Shared libraries are also supported in many cross or remote debugging
20476 configurations. @value{GDBN} needs to have access to the target's libraries;
20477 this can be accomplished either by providing copies of the libraries
20478 on the host system, or by asking @value{GDBN} to automatically retrieve the
20479 libraries from the target. If copies of the target libraries are
20480 provided, they need to be the same as the target libraries, although the
20481 copies on the target can be stripped as long as the copies on the host are
20484 @cindex where to look for shared libraries
20485 For remote debugging, you need to tell @value{GDBN} where the target
20486 libraries are, so that it can load the correct copies---otherwise, it
20487 may try to load the host's libraries. @value{GDBN} has two variables
20488 to specify the search directories for target libraries.
20491 @cindex prefix for executable and shared library file names
20492 @cindex system root, alternate
20493 @kindex set solib-absolute-prefix
20494 @kindex set sysroot
20495 @item set sysroot @var{path}
20496 Use @var{path} as the system root for the program being debugged. Any
20497 absolute shared library paths will be prefixed with @var{path}; many
20498 runtime loaders store the absolute paths to the shared library in the
20499 target program's memory. When starting processes remotely, and when
20500 attaching to already-running processes (local or remote), their
20501 executable filenames will be prefixed with @var{path} if reported to
20502 @value{GDBN} as absolute by the operating system. If you use
20503 @code{set sysroot} to find executables and shared libraries, they need
20504 to be laid out in the same way that they are on the target, with
20505 e.g.@: a @file{/bin}, @file{/lib} and @file{/usr/lib} hierarchy under
20508 If @var{path} starts with the sequence @file{target:} and the target
20509 system is remote then @value{GDBN} will retrieve the target binaries
20510 from the remote system. This is only supported when using a remote
20511 target that supports the @code{remote get} command (@pxref{File
20512 Transfer,,Sending files to a remote system}). The part of @var{path}
20513 following the initial @file{target:} (if present) is used as system
20514 root prefix on the remote file system. If @var{path} starts with the
20515 sequence @file{remote:} this is converted to the sequence
20516 @file{target:} by @code{set sysroot}@footnote{Historically the
20517 functionality to retrieve binaries from the remote system was
20518 provided by prefixing @var{path} with @file{remote:}}. If you want
20519 to specify a local system root using a directory that happens to be
20520 named @file{target:} or @file{remote:}, you need to use some
20521 equivalent variant of the name like @file{./target:}.
20523 For targets with an MS-DOS based filesystem, such as MS-Windows and
20524 SymbianOS, @value{GDBN} tries prefixing a few variants of the target
20525 absolute file name with @var{path}. But first, on Unix hosts,
20526 @value{GDBN} converts all backslash directory separators into forward
20527 slashes, because the backslash is not a directory separator on Unix:
20530 c:\foo\bar.dll @result{} c:/foo/bar.dll
20533 Then, @value{GDBN} attempts prefixing the target file name with
20534 @var{path}, and looks for the resulting file name in the host file
20538 c:/foo/bar.dll @result{} /path/to/sysroot/c:/foo/bar.dll
20541 If that does not find the binary, @value{GDBN} tries removing
20542 the @samp{:} character from the drive spec, both for convenience, and,
20543 for the case of the host file system not supporting file names with
20547 c:/foo/bar.dll @result{} /path/to/sysroot/c/foo/bar.dll
20550 This makes it possible to have a system root that mirrors a target
20551 with more than one drive. E.g., you may want to setup your local
20552 copies of the target system shared libraries like so (note @samp{c} vs
20556 @file{/path/to/sysroot/c/sys/bin/foo.dll}
20557 @file{/path/to/sysroot/c/sys/bin/bar.dll}
20558 @file{/path/to/sysroot/z/sys/bin/bar.dll}
20562 and point the system root at @file{/path/to/sysroot}, so that
20563 @value{GDBN} can find the correct copies of both
20564 @file{c:\sys\bin\foo.dll}, and @file{z:\sys\bin\bar.dll}.
20566 If that still does not find the binary, @value{GDBN} tries
20567 removing the whole drive spec from the target file name:
20570 c:/foo/bar.dll @result{} /path/to/sysroot/foo/bar.dll
20573 This last lookup makes it possible to not care about the drive name,
20574 if you don't want or need to.
20576 The @code{set solib-absolute-prefix} command is an alias for @code{set
20579 @cindex default system root
20580 @cindex @samp{--with-sysroot}
20581 You can set the default system root by using the configure-time
20582 @samp{--with-sysroot} option. If the system root is inside
20583 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
20584 @samp{--exec-prefix}), then the default system root will be updated
20585 automatically if the installed @value{GDBN} is moved to a new
20588 @kindex show sysroot
20590 Display the current executable and shared library prefix.
20592 @kindex set solib-search-path
20593 @item set solib-search-path @var{path}
20594 If this variable is set, @var{path} is a colon-separated list of
20595 directories to search for shared libraries. @samp{solib-search-path}
20596 is used after @samp{sysroot} fails to locate the library, or if the
20597 path to the library is relative instead of absolute. If you want to
20598 use @samp{solib-search-path} instead of @samp{sysroot}, be sure to set
20599 @samp{sysroot} to a nonexistent directory to prevent @value{GDBN} from
20600 finding your host's libraries. @samp{sysroot} is preferred; setting
20601 it to a nonexistent directory may interfere with automatic loading
20602 of shared library symbols.
20604 @kindex show solib-search-path
20605 @item show solib-search-path
20606 Display the current shared library search path.
20608 @cindex DOS file-name semantics of file names.
20609 @kindex set target-file-system-kind (unix|dos-based|auto)
20610 @kindex show target-file-system-kind
20611 @item set target-file-system-kind @var{kind}
20612 Set assumed file system kind for target reported file names.
20614 Shared library file names as reported by the target system may not
20615 make sense as is on the system @value{GDBN} is running on. For
20616 example, when remote debugging a target that has MS-DOS based file
20617 system semantics, from a Unix host, the target may be reporting to
20618 @value{GDBN} a list of loaded shared libraries with file names such as
20619 @file{c:\Windows\kernel32.dll}. On Unix hosts, there's no concept of
20620 drive letters, so the @samp{c:\} prefix is not normally understood as
20621 indicating an absolute file name, and neither is the backslash
20622 normally considered a directory separator character. In that case,
20623 the native file system would interpret this whole absolute file name
20624 as a relative file name with no directory components. This would make
20625 it impossible to point @value{GDBN} at a copy of the remote target's
20626 shared libraries on the host using @code{set sysroot}, and impractical
20627 with @code{set solib-search-path}. Setting
20628 @code{target-file-system-kind} to @code{dos-based} tells @value{GDBN}
20629 to interpret such file names similarly to how the target would, and to
20630 map them to file names valid on @value{GDBN}'s native file system
20631 semantics. The value of @var{kind} can be @code{"auto"}, in addition
20632 to one of the supported file system kinds. In that case, @value{GDBN}
20633 tries to determine the appropriate file system variant based on the
20634 current target's operating system (@pxref{ABI, ,Configuring the
20635 Current ABI}). The supported file system settings are:
20639 Instruct @value{GDBN} to assume the target file system is of Unix
20640 kind. Only file names starting the forward slash (@samp{/}) character
20641 are considered absolute, and the directory separator character is also
20645 Instruct @value{GDBN} to assume the target file system is DOS based.
20646 File names starting with either a forward slash, or a drive letter
20647 followed by a colon (e.g., @samp{c:}), are considered absolute, and
20648 both the slash (@samp{/}) and the backslash (@samp{\\}) characters are
20649 considered directory separators.
20652 Instruct @value{GDBN} to use the file system kind associated with the
20653 target operating system (@pxref{ABI, ,Configuring the Current ABI}).
20654 This is the default.
20658 @cindex file name canonicalization
20659 @cindex base name differences
20660 When processing file names provided by the user, @value{GDBN}
20661 frequently needs to compare them to the file names recorded in the
20662 program's debug info. Normally, @value{GDBN} compares just the
20663 @dfn{base names} of the files as strings, which is reasonably fast
20664 even for very large programs. (The base name of a file is the last
20665 portion of its name, after stripping all the leading directories.)
20666 This shortcut in comparison is based upon the assumption that files
20667 cannot have more than one base name. This is usually true, but
20668 references to files that use symlinks or similar filesystem
20669 facilities violate that assumption. If your program records files
20670 using such facilities, or if you provide file names to @value{GDBN}
20671 using symlinks etc., you can set @code{basenames-may-differ} to
20672 @code{true} to instruct @value{GDBN} to completely canonicalize each
20673 pair of file names it needs to compare. This will make file-name
20674 comparisons accurate, but at a price of a significant slowdown.
20677 @item set basenames-may-differ
20678 @kindex set basenames-may-differ
20679 Set whether a source file may have multiple base names.
20681 @item show basenames-may-differ
20682 @kindex show basenames-may-differ
20683 Show whether a source file may have multiple base names.
20687 @section File Caching
20688 @cindex caching of opened files
20689 @cindex caching of bfd objects
20691 To speed up file loading, and reduce memory usage, @value{GDBN} will
20692 reuse the @code{bfd} objects used to track open files. @xref{Top, ,
20693 BFD, bfd, The Binary File Descriptor Library}. The following commands
20694 allow visibility and control of the caching behavior.
20697 @kindex maint info bfds
20698 @item maint info bfds
20699 This prints information about each @code{bfd} object that is known to
20702 @kindex maint set bfd-sharing
20703 @kindex maint show bfd-sharing
20704 @kindex bfd caching
20705 @item maint set bfd-sharing
20706 @item maint show bfd-sharing
20707 Control whether @code{bfd} objects can be shared. When sharing is
20708 enabled @value{GDBN} reuses already open @code{bfd} objects rather
20709 than reopening the same file. Turning sharing off does not cause
20710 already shared @code{bfd} objects to be unshared, but all future files
20711 that are opened will create a new @code{bfd} object. Similarly,
20712 re-enabling sharing does not cause multiple existing @code{bfd}
20713 objects to be collapsed into a single shared @code{bfd} object.
20715 @kindex set debug bfd-cache @var{level}
20716 @kindex bfd caching
20717 @item set debug bfd-cache @var{level}
20718 Turns on debugging of the bfd cache, setting the level to @var{level}.
20720 @kindex show debug bfd-cache
20721 @kindex bfd caching
20722 @item show debug bfd-cache
20723 Show the current debugging level of the bfd cache.
20726 @node Separate Debug Files
20727 @section Debugging Information in Separate Files
20728 @cindex separate debugging information files
20729 @cindex debugging information in separate files
20730 @cindex @file{.debug} subdirectories
20731 @cindex debugging information directory, global
20732 @cindex global debugging information directories
20733 @cindex build ID, and separate debugging files
20734 @cindex @file{.build-id} directory
20736 @value{GDBN} allows you to put a program's debugging information in a
20737 file separate from the executable itself, in a way that allows
20738 @value{GDBN} to find and load the debugging information automatically.
20739 Since debugging information can be very large---sometimes larger
20740 than the executable code itself---some systems distribute debugging
20741 information for their executables in separate files, which users can
20742 install only when they need to debug a problem.
20744 @value{GDBN} supports two ways of specifying the separate debug info
20749 The executable contains a @dfn{debug link} that specifies the name of
20750 the separate debug info file. The separate debug file's name is
20751 usually @file{@var{executable}.debug}, where @var{executable} is the
20752 name of the corresponding executable file without leading directories
20753 (e.g., @file{ls.debug} for @file{/usr/bin/ls}). In addition, the
20754 debug link specifies a 32-bit @dfn{Cyclic Redundancy Check} (CRC)
20755 checksum for the debug file, which @value{GDBN} uses to validate that
20756 the executable and the debug file came from the same build.
20759 The executable contains a @dfn{build ID}, a unique bit string that is
20760 also present in the corresponding debug info file. (This is supported
20761 only on some operating systems, when using the ELF or PE file formats
20762 for binary files and the @sc{gnu} Binutils.) For more details about
20763 this feature, see the description of the @option{--build-id}
20764 command-line option in @ref{Options, , Command Line Options, ld,
20765 The GNU Linker}. The debug info file's name is not specified
20766 explicitly by the build ID, but can be computed from the build ID, see
20770 Depending on the way the debug info file is specified, @value{GDBN}
20771 uses two different methods of looking for the debug file:
20775 For the ``debug link'' method, @value{GDBN} looks up the named file in
20776 the directory of the executable file, then in a subdirectory of that
20777 directory named @file{.debug}, and finally under each one of the
20778 global debug directories, in a subdirectory whose name is identical to
20779 the leading directories of the executable's absolute file name. (On
20780 MS-Windows/MS-DOS, the drive letter of the executable's leading
20781 directories is converted to a one-letter subdirectory, i.e.@:
20782 @file{d:/usr/bin/} is converted to @file{/d/usr/bin/}, because Windows
20783 filesystems disallow colons in file names.)
20786 For the ``build ID'' method, @value{GDBN} looks in the
20787 @file{.build-id} subdirectory of each one of the global debug directories for
20788 a file named @file{@var{nn}/@var{nnnnnnnn}.debug}, where @var{nn} are the
20789 first 2 hex characters of the build ID bit string, and @var{nnnnnnnn}
20790 are the rest of the bit string. (Real build ID strings are 32 or more
20791 hex characters, not 10.)
20794 So, for example, suppose you ask @value{GDBN} to debug
20795 @file{/usr/bin/ls}, which has a debug link that specifies the
20796 file @file{ls.debug}, and a build ID whose value in hex is
20797 @code{abcdef1234}. If the list of the global debug directories includes
20798 @file{/usr/lib/debug}, then @value{GDBN} will look for the following
20799 debug information files, in the indicated order:
20803 @file{/usr/lib/debug/.build-id/ab/cdef1234.debug}
20805 @file{/usr/bin/ls.debug}
20807 @file{/usr/bin/.debug/ls.debug}
20809 @file{/usr/lib/debug/usr/bin/ls.debug}.
20812 @anchor{debug-file-directory}
20813 Global debugging info directories default to what is set by @value{GDBN}
20814 configure option @option{--with-separate-debug-dir}. During @value{GDBN} run
20815 you can also set the global debugging info directories, and view the list
20816 @value{GDBN} is currently using.
20820 @kindex set debug-file-directory
20821 @item set debug-file-directory @var{directories}
20822 Set the directories which @value{GDBN} searches for separate debugging
20823 information files to @var{directory}. Multiple path components can be set
20824 concatenating them by a path separator.
20826 @kindex show debug-file-directory
20827 @item show debug-file-directory
20828 Show the directories @value{GDBN} searches for separate debugging
20833 @cindex @code{.gnu_debuglink} sections
20834 @cindex debug link sections
20835 A debug link is a special section of the executable file named
20836 @code{.gnu_debuglink}. The section must contain:
20840 A filename, with any leading directory components removed, followed by
20843 zero to three bytes of padding, as needed to reach the next four-byte
20844 boundary within the section, and
20846 a four-byte CRC checksum, stored in the same endianness used for the
20847 executable file itself. The checksum is computed on the debugging
20848 information file's full contents by the function given below, passing
20849 zero as the @var{crc} argument.
20852 Any executable file format can carry a debug link, as long as it can
20853 contain a section named @code{.gnu_debuglink} with the contents
20856 @cindex @code{.note.gnu.build-id} sections
20857 @cindex build ID sections
20858 The build ID is a special section in the executable file (and in other
20859 ELF binary files that @value{GDBN} may consider). This section is
20860 often named @code{.note.gnu.build-id}, but that name is not mandatory.
20861 It contains unique identification for the built files---the ID remains
20862 the same across multiple builds of the same build tree. The default
20863 algorithm SHA1 produces 160 bits (40 hexadecimal characters) of the
20864 content for the build ID string. The same section with an identical
20865 value is present in the original built binary with symbols, in its
20866 stripped variant, and in the separate debugging information file.
20868 The debugging information file itself should be an ordinary
20869 executable, containing a full set of linker symbols, sections, and
20870 debugging information. The sections of the debugging information file
20871 should have the same names, addresses, and sizes as the original file,
20872 but they need not contain any data---much like a @code{.bss} section
20873 in an ordinary executable.
20875 The @sc{gnu} binary utilities (Binutils) package includes the
20876 @samp{objcopy} utility that can produce
20877 the separated executable / debugging information file pairs using the
20878 following commands:
20881 @kbd{objcopy --only-keep-debug foo foo.debug}
20886 These commands remove the debugging
20887 information from the executable file @file{foo} and place it in the file
20888 @file{foo.debug}. You can use the first, second or both methods to link the
20893 The debug link method needs the following additional command to also leave
20894 behind a debug link in @file{foo}:
20897 @kbd{objcopy --add-gnu-debuglink=foo.debug foo}
20900 Ulrich Drepper's @file{elfutils} package, starting with version 0.53, contains
20901 a version of the @code{strip} command such that the command @kbd{strip foo -f
20902 foo.debug} has the same functionality as the two @code{objcopy} commands and
20903 the @code{ln -s} command above, together.
20906 Build ID gets embedded into the main executable using @code{ld --build-id} or
20907 the @value{NGCC} counterpart @code{gcc -Wl,--build-id}. Build ID support plus
20908 compatibility fixes for debug files separation are present in @sc{gnu} binary
20909 utilities (Binutils) package since version 2.18.
20914 @cindex CRC algorithm definition
20915 The CRC used in @code{.gnu_debuglink} is the CRC-32 defined in
20916 IEEE 802.3 using the polynomial:
20918 @c TexInfo requires naked braces for multi-digit exponents for Tex
20919 @c output, but this causes HTML output to barf. HTML has to be set using
20920 @c raw commands. So we end up having to specify this equation in 2
20925 <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>
20926 + <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
20932 @math{x^{32} + x^{26} + x^{23} + x^{22} + x^{16} + x^{12} + x^{11}}
20933 @math{+ x^{10} + x^8 + x^7 + x^5 + x^4 + x^2 + x + 1}
20937 The function is computed byte at a time, taking the least
20938 significant bit of each byte first. The initial pattern
20939 @code{0xffffffff} is used, to ensure leading zeros affect the CRC and
20940 the final result is inverted to ensure trailing zeros also affect the
20943 @emph{Note:} This is the same CRC polynomial as used in handling the
20944 @dfn{Remote Serial Protocol} @code{qCRC} packet (@pxref{qCRC packet}).
20945 However in the case of the Remote Serial Protocol, the CRC is computed
20946 @emph{most} significant bit first, and the result is not inverted, so
20947 trailing zeros have no effect on the CRC value.
20949 To complete the description, we show below the code of the function
20950 which produces the CRC used in @code{.gnu_debuglink}. Inverting the
20951 initially supplied @code{crc} argument means that an initial call to
20952 this function passing in zero will start computing the CRC using
20955 @kindex gnu_debuglink_crc32
20958 gnu_debuglink_crc32 (unsigned long crc,
20959 unsigned char *buf, size_t len)
20961 static const unsigned long crc32_table[256] =
20963 0x00000000, 0x77073096, 0xee0e612c, 0x990951ba, 0x076dc419,
20964 0x706af48f, 0xe963a535, 0x9e6495a3, 0x0edb8832, 0x79dcb8a4,
20965 0xe0d5e91e, 0x97d2d988, 0x09b64c2b, 0x7eb17cbd, 0xe7b82d07,
20966 0x90bf1d91, 0x1db71064, 0x6ab020f2, 0xf3b97148, 0x84be41de,
20967 0x1adad47d, 0x6ddde4eb, 0xf4d4b551, 0x83d385c7, 0x136c9856,
20968 0x646ba8c0, 0xfd62f97a, 0x8a65c9ec, 0x14015c4f, 0x63066cd9,
20969 0xfa0f3d63, 0x8d080df5, 0x3b6e20c8, 0x4c69105e, 0xd56041e4,
20970 0xa2677172, 0x3c03e4d1, 0x4b04d447, 0xd20d85fd, 0xa50ab56b,
20971 0x35b5a8fa, 0x42b2986c, 0xdbbbc9d6, 0xacbcf940, 0x32d86ce3,
20972 0x45df5c75, 0xdcd60dcf, 0xabd13d59, 0x26d930ac, 0x51de003a,
20973 0xc8d75180, 0xbfd06116, 0x21b4f4b5, 0x56b3c423, 0xcfba9599,
20974 0xb8bda50f, 0x2802b89e, 0x5f058808, 0xc60cd9b2, 0xb10be924,
20975 0x2f6f7c87, 0x58684c11, 0xc1611dab, 0xb6662d3d, 0x76dc4190,
20976 0x01db7106, 0x98d220bc, 0xefd5102a, 0x71b18589, 0x06b6b51f,
20977 0x9fbfe4a5, 0xe8b8d433, 0x7807c9a2, 0x0f00f934, 0x9609a88e,
20978 0xe10e9818, 0x7f6a0dbb, 0x086d3d2d, 0x91646c97, 0xe6635c01,
20979 0x6b6b51f4, 0x1c6c6162, 0x856530d8, 0xf262004e, 0x6c0695ed,
20980 0x1b01a57b, 0x8208f4c1, 0xf50fc457, 0x65b0d9c6, 0x12b7e950,
20981 0x8bbeb8ea, 0xfcb9887c, 0x62dd1ddf, 0x15da2d49, 0x8cd37cf3,
20982 0xfbd44c65, 0x4db26158, 0x3ab551ce, 0xa3bc0074, 0xd4bb30e2,
20983 0x4adfa541, 0x3dd895d7, 0xa4d1c46d, 0xd3d6f4fb, 0x4369e96a,
20984 0x346ed9fc, 0xad678846, 0xda60b8d0, 0x44042d73, 0x33031de5,
20985 0xaa0a4c5f, 0xdd0d7cc9, 0x5005713c, 0x270241aa, 0xbe0b1010,
20986 0xc90c2086, 0x5768b525, 0x206f85b3, 0xb966d409, 0xce61e49f,
20987 0x5edef90e, 0x29d9c998, 0xb0d09822, 0xc7d7a8b4, 0x59b33d17,
20988 0x2eb40d81, 0xb7bd5c3b, 0xc0ba6cad, 0xedb88320, 0x9abfb3b6,
20989 0x03b6e20c, 0x74b1d29a, 0xead54739, 0x9dd277af, 0x04db2615,
20990 0x73dc1683, 0xe3630b12, 0x94643b84, 0x0d6d6a3e, 0x7a6a5aa8,
20991 0xe40ecf0b, 0x9309ff9d, 0x0a00ae27, 0x7d079eb1, 0xf00f9344,
20992 0x8708a3d2, 0x1e01f268, 0x6906c2fe, 0xf762575d, 0x806567cb,
20993 0x196c3671, 0x6e6b06e7, 0xfed41b76, 0x89d32be0, 0x10da7a5a,
20994 0x67dd4acc, 0xf9b9df6f, 0x8ebeeff9, 0x17b7be43, 0x60b08ed5,
20995 0xd6d6a3e8, 0xa1d1937e, 0x38d8c2c4, 0x4fdff252, 0xd1bb67f1,
20996 0xa6bc5767, 0x3fb506dd, 0x48b2364b, 0xd80d2bda, 0xaf0a1b4c,
20997 0x36034af6, 0x41047a60, 0xdf60efc3, 0xa867df55, 0x316e8eef,
20998 0x4669be79, 0xcb61b38c, 0xbc66831a, 0x256fd2a0, 0x5268e236,
20999 0xcc0c7795, 0xbb0b4703, 0x220216b9, 0x5505262f, 0xc5ba3bbe,
21000 0xb2bd0b28, 0x2bb45a92, 0x5cb36a04, 0xc2d7ffa7, 0xb5d0cf31,
21001 0x2cd99e8b, 0x5bdeae1d, 0x9b64c2b0, 0xec63f226, 0x756aa39c,
21002 0x026d930a, 0x9c0906a9, 0xeb0e363f, 0x72076785, 0x05005713,
21003 0x95bf4a82, 0xe2b87a14, 0x7bb12bae, 0x0cb61b38, 0x92d28e9b,
21004 0xe5d5be0d, 0x7cdcefb7, 0x0bdbdf21, 0x86d3d2d4, 0xf1d4e242,
21005 0x68ddb3f8, 0x1fda836e, 0x81be16cd, 0xf6b9265b, 0x6fb077e1,
21006 0x18b74777, 0x88085ae6, 0xff0f6a70, 0x66063bca, 0x11010b5c,
21007 0x8f659eff, 0xf862ae69, 0x616bffd3, 0x166ccf45, 0xa00ae278,
21008 0xd70dd2ee, 0x4e048354, 0x3903b3c2, 0xa7672661, 0xd06016f7,
21009 0x4969474d, 0x3e6e77db, 0xaed16a4a, 0xd9d65adc, 0x40df0b66,
21010 0x37d83bf0, 0xa9bcae53, 0xdebb9ec5, 0x47b2cf7f, 0x30b5ffe9,
21011 0xbdbdf21c, 0xcabac28a, 0x53b39330, 0x24b4a3a6, 0xbad03605,
21012 0xcdd70693, 0x54de5729, 0x23d967bf, 0xb3667a2e, 0xc4614ab8,
21013 0x5d681b02, 0x2a6f2b94, 0xb40bbe37, 0xc30c8ea1, 0x5a05df1b,
21016 unsigned char *end;
21018 crc = ~crc & 0xffffffff;
21019 for (end = buf + len; buf < end; ++buf)
21020 crc = crc32_table[(crc ^ *buf) & 0xff] ^ (crc >> 8);
21021 return ~crc & 0xffffffff;
21026 This computation does not apply to the ``build ID'' method.
21028 @node MiniDebugInfo
21029 @section Debugging information in a special section
21030 @cindex separate debug sections
21031 @cindex @samp{.gnu_debugdata} section
21033 Some systems ship pre-built executables and libraries that have a
21034 special @samp{.gnu_debugdata} section. This feature is called
21035 @dfn{MiniDebugInfo}. This section holds an LZMA-compressed object and
21036 is used to supply extra symbols for backtraces.
21038 The intent of this section is to provide extra minimal debugging
21039 information for use in simple backtraces. It is not intended to be a
21040 replacement for full separate debugging information (@pxref{Separate
21041 Debug Files}). The example below shows the intended use; however,
21042 @value{GDBN} does not currently put restrictions on what sort of
21043 debugging information might be included in the section.
21045 @value{GDBN} has support for this extension. If the section exists,
21046 then it is used provided that no other source of debugging information
21047 can be found, and that @value{GDBN} was configured with LZMA support.
21049 This section can be easily created using @command{objcopy} and other
21050 standard utilities:
21053 # Extract the dynamic symbols from the main binary, there is no need
21054 # to also have these in the normal symbol table.
21055 nm -D @var{binary} --format=posix --defined-only \
21056 | awk '@{ print $1 @}' | sort > dynsyms
21058 # Extract all the text (i.e. function) symbols from the debuginfo.
21059 # (Note that we actually also accept "D" symbols, for the benefit
21060 # of platforms like PowerPC64 that use function descriptors.)
21061 nm @var{binary} --format=posix --defined-only \
21062 | awk '@{ if ($2 == "T" || $2 == "t" || $2 == "D") print $1 @}' \
21065 # Keep all the function symbols not already in the dynamic symbol
21067 comm -13 dynsyms funcsyms > keep_symbols
21069 # Separate full debug info into debug binary.
21070 objcopy --only-keep-debug @var{binary} debug
21072 # Copy the full debuginfo, keeping only a minimal set of symbols and
21073 # removing some unnecessary sections.
21074 objcopy -S --remove-section .gdb_index --remove-section .comment \
21075 --keep-symbols=keep_symbols debug mini_debuginfo
21077 # Drop the full debug info from the original binary.
21078 strip --strip-all -R .comment @var{binary}
21080 # Inject the compressed data into the .gnu_debugdata section of the
21083 objcopy --add-section .gnu_debugdata=mini_debuginfo.xz @var{binary}
21087 @section Index Files Speed Up @value{GDBN}
21088 @cindex index files
21089 @cindex @samp{.gdb_index} section
21091 When @value{GDBN} finds a symbol file, it scans the symbols in the
21092 file in order to construct an internal symbol table. This lets most
21093 @value{GDBN} operations work quickly---at the cost of a delay early
21094 on. For large programs, this delay can be quite lengthy, so
21095 @value{GDBN} provides a way to build an index, which speeds up
21098 For convenience, @value{GDBN} comes with a program,
21099 @command{gdb-add-index}, which can be used to add the index to a
21100 symbol file. It takes the symbol file as its only argument:
21103 $ gdb-add-index symfile
21106 @xref{gdb-add-index}.
21108 It is also possible to do the work manually. Here is what
21109 @command{gdb-add-index} does behind the curtains.
21111 The index is stored as a section in the symbol file. @value{GDBN} can
21112 write the index to a file, then you can put it into the symbol file
21113 using @command{objcopy}.
21115 To create an index file, use the @code{save gdb-index} command:
21118 @item save gdb-index [-dwarf-5] @var{directory}
21119 @kindex save gdb-index
21120 Create index files for all symbol files currently known by
21121 @value{GDBN}. For each known @var{symbol-file}, this command by
21122 default creates it produces a single file
21123 @file{@var{symbol-file}.gdb-index}. If you invoke this command with
21124 the @option{-dwarf-5} option, it produces 2 files:
21125 @file{@var{symbol-file}.debug_names} and
21126 @file{@var{symbol-file}.debug_str}. The files are created in the
21127 given @var{directory}.
21130 Once you have created an index file you can merge it into your symbol
21131 file, here named @file{symfile}, using @command{objcopy}:
21134 $ objcopy --add-section .gdb_index=symfile.gdb-index \
21135 --set-section-flags .gdb_index=readonly symfile symfile
21138 Or for @code{-dwarf-5}:
21141 $ objcopy --dump-section .debug_str=symfile.debug_str.new symfile
21142 $ cat symfile.debug_str >>symfile.debug_str.new
21143 $ objcopy --add-section .debug_names=symfile.gdb-index \
21144 --set-section-flags .debug_names=readonly \
21145 --update-section .debug_str=symfile.debug_str.new symfile symfile
21148 @value{GDBN} will normally ignore older versions of @file{.gdb_index}
21149 sections that have been deprecated. Usually they are deprecated because
21150 they are missing a new feature or have performance issues.
21151 To tell @value{GDBN} to use a deprecated index section anyway
21152 specify @code{set use-deprecated-index-sections on}.
21153 The default is @code{off}.
21154 This can speed up startup, but may result in some functionality being lost.
21155 @xref{Index Section Format}.
21157 @emph{Warning:} Setting @code{use-deprecated-index-sections} to @code{on}
21158 must be done before gdb reads the file. The following will not work:
21161 $ gdb -ex "set use-deprecated-index-sections on" <program>
21164 Instead you must do, for example,
21167 $ gdb -iex "set use-deprecated-index-sections on" <program>
21170 There are currently some limitation on indices. They only work when
21171 using DWARF debugging information, not stabs. And, only the
21172 @code{-dwarf-5} index works for programs using Ada.
21174 @subsection Automatic symbol index cache
21176 @cindex automatic symbol index cache
21177 It is possible for @value{GDBN} to automatically save a copy of this index in a
21178 cache on disk and retrieve it from there when loading the same binary in the
21179 future. This feature can be turned on with @kbd{set index-cache on}. The
21180 following commands can be used to tweak the behavior of the index cache.
21184 @kindex set index-cache
21185 @item set index-cache on
21186 @itemx set index-cache off
21187 Enable or disable the use of the symbol index cache.
21189 @item set index-cache directory @var{directory}
21190 @kindex show index-cache
21191 @itemx show index-cache directory
21192 Set/show the directory where index files will be saved.
21194 The default value for this directory depends on the host platform. On
21195 most systems, the index is cached in the @file{gdb} subdirectory of
21196 the directory pointed to by the @env{XDG_CACHE_HOME} environment
21197 variable, if it is defined, else in the @file{.cache/gdb} subdirectory
21198 of your home directory. However, on some systems, the default may
21199 differ according to local convention.
21201 There is no limit on the disk space used by index cache. It is perfectly safe
21202 to delete the content of that directory to free up disk space.
21204 @item show index-cache stats
21205 Print the number of cache hits and misses since the launch of @value{GDBN}.
21209 @node Symbol Errors
21210 @section Errors Reading Symbol Files
21212 While reading a symbol file, @value{GDBN} occasionally encounters problems,
21213 such as symbol types it does not recognize, or known bugs in compiler
21214 output. By default, @value{GDBN} does not notify you of such problems, since
21215 they are relatively common and primarily of interest to people
21216 debugging compilers. If you are interested in seeing information
21217 about ill-constructed symbol tables, you can either ask @value{GDBN} to print
21218 only one message about each such type of problem, no matter how many
21219 times the problem occurs; or you can ask @value{GDBN} to print more messages,
21220 to see how many times the problems occur, with the @code{set
21221 complaints} command (@pxref{Messages/Warnings, ,Optional Warnings and
21224 The messages currently printed, and their meanings, include:
21227 @item inner block not inside outer block in @var{symbol}
21229 The symbol information shows where symbol scopes begin and end
21230 (such as at the start of a function or a block of statements). This
21231 error indicates that an inner scope block is not fully contained
21232 in its outer scope blocks.
21234 @value{GDBN} circumvents the problem by treating the inner block as if it had
21235 the same scope as the outer block. In the error message, @var{symbol}
21236 may be shown as ``@code{(don't know)}'' if the outer block is not a
21239 @item block at @var{address} out of order
21241 The symbol information for symbol scope blocks should occur in
21242 order of increasing addresses. This error indicates that it does not
21245 @value{GDBN} does not circumvent this problem, and has trouble
21246 locating symbols in the source file whose symbols it is reading. (You
21247 can often determine what source file is affected by specifying
21248 @code{set verbose on}. @xref{Messages/Warnings, ,Optional Warnings and
21251 @item bad block start address patched
21253 The symbol information for a symbol scope block has a start address
21254 smaller than the address of the preceding source line. This is known
21255 to occur in the SunOS 4.1.1 (and earlier) C compiler.
21257 @value{GDBN} circumvents the problem by treating the symbol scope block as
21258 starting on the previous source line.
21260 @item bad string table offset in symbol @var{n}
21263 Symbol number @var{n} contains a pointer into the string table which is
21264 larger than the size of the string table.
21266 @value{GDBN} circumvents the problem by considering the symbol to have the
21267 name @code{foo}, which may cause other problems if many symbols end up
21270 @item unknown symbol type @code{0x@var{nn}}
21272 The symbol information contains new data types that @value{GDBN} does
21273 not yet know how to read. @code{0x@var{nn}} is the symbol type of the
21274 uncomprehended information, in hexadecimal.
21276 @value{GDBN} circumvents the error by ignoring this symbol information.
21277 This usually allows you to debug your program, though certain symbols
21278 are not accessible. If you encounter such a problem and feel like
21279 debugging it, you can debug @code{@value{GDBP}} with itself, breakpoint
21280 on @code{complain}, then go up to the function @code{read_dbx_symtab}
21281 and examine @code{*bufp} to see the symbol.
21283 @item stub type has NULL name
21285 @value{GDBN} could not find the full definition for a struct or class.
21287 @item const/volatile indicator missing (ok if using g++ v1.x), got@dots{}
21288 The symbol information for a C@t{++} member function is missing some
21289 information that recent versions of the compiler should have output for
21292 @item info mismatch between compiler and debugger
21294 @value{GDBN} could not parse a type specification output by the compiler.
21299 @section GDB Data Files
21301 @cindex prefix for data files
21302 @value{GDBN} will sometimes read an auxiliary data file. These files
21303 are kept in a directory known as the @dfn{data directory}.
21305 You can set the data directory's name, and view the name @value{GDBN}
21306 is currently using.
21309 @kindex set data-directory
21310 @item set data-directory @var{directory}
21311 Set the directory which @value{GDBN} searches for auxiliary data files
21312 to @var{directory}.
21314 @kindex show data-directory
21315 @item show data-directory
21316 Show the directory @value{GDBN} searches for auxiliary data files.
21319 @cindex default data directory
21320 @cindex @samp{--with-gdb-datadir}
21321 You can set the default data directory by using the configure-time
21322 @samp{--with-gdb-datadir} option. If the data directory is inside
21323 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
21324 @samp{--exec-prefix}), then the default data directory will be updated
21325 automatically if the installed @value{GDBN} is moved to a new
21328 The data directory may also be specified with the
21329 @code{--data-directory} command line option.
21330 @xref{Mode Options}.
21333 @chapter Specifying a Debugging Target
21335 @cindex debugging target
21336 A @dfn{target} is the execution environment occupied by your program.
21338 Often, @value{GDBN} runs in the same host environment as your program;
21339 in that case, the debugging target is specified as a side effect when
21340 you use the @code{file} or @code{core} commands. When you need more
21341 flexibility---for example, running @value{GDBN} on a physically separate
21342 host, or controlling a standalone system over a serial port or a
21343 realtime system over a TCP/IP connection---you can use the @code{target}
21344 command to specify one of the target types configured for @value{GDBN}
21345 (@pxref{Target Commands, ,Commands for Managing Targets}).
21347 @cindex target architecture
21348 It is possible to build @value{GDBN} for several different @dfn{target
21349 architectures}. When @value{GDBN} is built like that, you can choose
21350 one of the available architectures with the @kbd{set architecture}
21354 @kindex set architecture
21355 @kindex show architecture
21356 @item set architecture @var{arch}
21357 This command sets the current target architecture to @var{arch}. The
21358 value of @var{arch} can be @code{"auto"}, in addition to one of the
21359 supported architectures.
21361 @item show architecture
21362 Show the current target architecture.
21364 @item set processor
21366 @kindex set processor
21367 @kindex show processor
21368 These are alias commands for, respectively, @code{set architecture}
21369 and @code{show architecture}.
21373 * Active Targets:: Active targets
21374 * Target Commands:: Commands for managing targets
21375 * Byte Order:: Choosing target byte order
21378 @node Active Targets
21379 @section Active Targets
21381 @cindex stacking targets
21382 @cindex active targets
21383 @cindex multiple targets
21385 There are multiple classes of targets such as: processes, executable files or
21386 recording sessions. Core files belong to the process class, making core file
21387 and process mutually exclusive. Otherwise, @value{GDBN} can work concurrently
21388 on multiple active targets, one in each class. This allows you to (for
21389 example) start a process and inspect its activity, while still having access to
21390 the executable file after the process finishes. Or if you start process
21391 recording (@pxref{Reverse Execution}) and @code{reverse-step} there, you are
21392 presented a virtual layer of the recording target, while the process target
21393 remains stopped at the chronologically last point of the process execution.
21395 Use the @code{core-file} and @code{exec-file} commands to select a new core
21396 file or executable target (@pxref{Files, ,Commands to Specify Files}). To
21397 specify as a target a process that is already running, use the @code{attach}
21398 command (@pxref{Attach, ,Debugging an Already-running Process}).
21400 @node Target Commands
21401 @section Commands for Managing Targets
21404 @item target @var{type} @var{parameters}
21405 Connects the @value{GDBN} host environment to a target machine or
21406 process. A target is typically a protocol for talking to debugging
21407 facilities. You use the argument @var{type} to specify the type or
21408 protocol of the target machine.
21410 Further @var{parameters} are interpreted by the target protocol, but
21411 typically include things like device names or host names to connect
21412 with, process numbers, and baud rates.
21414 The @code{target} command does not repeat if you press @key{RET} again
21415 after executing the command.
21417 @kindex help target
21419 Displays the names of all targets available. To display targets
21420 currently selected, use either @code{info target} or @code{info files}
21421 (@pxref{Files, ,Commands to Specify Files}).
21423 @item help target @var{name}
21424 Describe a particular target, including any parameters necessary to
21427 @kindex set gnutarget
21428 @item set gnutarget @var{args}
21429 @value{GDBN} uses its own library BFD to read your files. @value{GDBN}
21430 knows whether it is reading an @dfn{executable},
21431 a @dfn{core}, or a @dfn{.o} file; however, you can specify the file format
21432 with the @code{set gnutarget} command. Unlike most @code{target} commands,
21433 with @code{gnutarget} the @code{target} refers to a program, not a machine.
21436 @emph{Warning:} To specify a file format with @code{set gnutarget},
21437 you must know the actual BFD name.
21441 @xref{Files, , Commands to Specify Files}.
21443 @kindex show gnutarget
21444 @item show gnutarget
21445 Use the @code{show gnutarget} command to display what file format
21446 @code{gnutarget} is set to read. If you have not set @code{gnutarget},
21447 @value{GDBN} will determine the file format for each file automatically,
21448 and @code{show gnutarget} displays @samp{The current BFD target is "auto"}.
21451 @cindex common targets
21452 Here are some common targets (available, or not, depending on the GDB
21457 @item target exec @var{program}
21458 @cindex executable file target
21459 An executable file. @samp{target exec @var{program}} is the same as
21460 @samp{exec-file @var{program}}.
21462 @item target core @var{filename}
21463 @cindex core dump file target
21464 A core dump file. @samp{target core @var{filename}} is the same as
21465 @samp{core-file @var{filename}}.
21467 @item target remote @var{medium}
21468 @cindex remote target
21469 A remote system connected to @value{GDBN} via a serial line or network
21470 connection. This command tells @value{GDBN} to use its own remote
21471 protocol over @var{medium} for debugging. @xref{Remote Debugging}.
21473 For example, if you have a board connected to @file{/dev/ttya} on the
21474 machine running @value{GDBN}, you could say:
21477 target remote /dev/ttya
21480 @code{target remote} supports the @code{load} command. This is only
21481 useful if you have some other way of getting the stub to the target
21482 system, and you can put it somewhere in memory where it won't get
21483 clobbered by the download.
21485 @item target sim @r{[}@var{simargs}@r{]} @dots{}
21486 @cindex built-in simulator target
21487 Builtin CPU simulator. @value{GDBN} includes simulators for most architectures.
21495 works; however, you cannot assume that a specific memory map, device
21496 drivers, or even basic I/O is available, although some simulators do
21497 provide these. For info about any processor-specific simulator details,
21498 see the appropriate section in @ref{Embedded Processors, ,Embedded
21501 @item target native
21502 @cindex native target
21503 Setup for local/native process debugging. Useful to make the
21504 @code{run} command spawn native processes (likewise @code{attach},
21505 etc.@:) even when @code{set auto-connect-native-target} is @code{off}
21506 (@pxref{set auto-connect-native-target}).
21510 Different targets are available on different configurations of @value{GDBN};
21511 your configuration may have more or fewer targets.
21513 Many remote targets require you to download the executable's code once
21514 you've successfully established a connection. You may wish to control
21515 various aspects of this process.
21520 @kindex set hash@r{, for remote monitors}
21521 @cindex hash mark while downloading
21522 This command controls whether a hash mark @samp{#} is displayed while
21523 downloading a file to the remote monitor. If on, a hash mark is
21524 displayed after each S-record is successfully downloaded to the
21528 @kindex show hash@r{, for remote monitors}
21529 Show the current status of displaying the hash mark.
21531 @item set debug monitor
21532 @kindex set debug monitor
21533 @cindex display remote monitor communications
21534 Enable or disable display of communications messages between
21535 @value{GDBN} and the remote monitor.
21537 @item show debug monitor
21538 @kindex show debug monitor
21539 Show the current status of displaying communications between
21540 @value{GDBN} and the remote monitor.
21545 @kindex load @var{filename} @var{offset}
21546 @item load @var{filename} @var{offset}
21548 Depending on what remote debugging facilities are configured into
21549 @value{GDBN}, the @code{load} command may be available. Where it exists, it
21550 is meant to make @var{filename} (an executable) available for debugging
21551 on the remote system---by downloading, or dynamic linking, for example.
21552 @code{load} also records the @var{filename} symbol table in @value{GDBN}, like
21553 the @code{add-symbol-file} command.
21555 If your @value{GDBN} does not have a @code{load} command, attempting to
21556 execute it gets the error message ``@code{You can't do that when your
21557 target is @dots{}}''
21559 The file is loaded at whatever address is specified in the executable.
21560 For some object file formats, you can specify the load address when you
21561 link the program; for other formats, like a.out, the object file format
21562 specifies a fixed address.
21563 @c FIXME! This would be a good place for an xref to the GNU linker doc.
21565 It is also possible to tell @value{GDBN} to load the executable file at a
21566 specific offset described by the optional argument @var{offset}. When
21567 @var{offset} is provided, @var{filename} must also be provided.
21569 Depending on the remote side capabilities, @value{GDBN} may be able to
21570 load programs into flash memory.
21572 @code{load} does not repeat if you press @key{RET} again after using it.
21577 @kindex flash-erase
21579 @anchor{flash-erase}
21581 Erases all known flash memory regions on the target.
21586 @section Choosing Target Byte Order
21588 @cindex choosing target byte order
21589 @cindex target byte order
21591 Some types of processors, such as the @acronym{MIPS}, PowerPC, and Renesas SH,
21592 offer the ability to run either big-endian or little-endian byte
21593 orders. Usually the executable or symbol will include a bit to
21594 designate the endian-ness, and you will not need to worry about
21595 which to use. However, you may still find it useful to adjust
21596 @value{GDBN}'s idea of processor endian-ness manually.
21600 @item set endian big
21601 Instruct @value{GDBN} to assume the target is big-endian.
21603 @item set endian little
21604 Instruct @value{GDBN} to assume the target is little-endian.
21606 @item set endian auto
21607 Instruct @value{GDBN} to use the byte order associated with the
21611 Display @value{GDBN}'s current idea of the target byte order.
21615 If the @code{set endian auto} mode is in effect and no executable has
21616 been selected, then the endianness used is the last one chosen either
21617 by one of the @code{set endian big} and @code{set endian little}
21618 commands or by inferring from the last executable used. If no
21619 endianness has been previously chosen, then the default for this mode
21620 is inferred from the target @value{GDBN} has been built for, and is
21621 @code{little} if the name of the target CPU has an @code{el} suffix
21622 and @code{big} otherwise.
21624 Note that these commands merely adjust interpretation of symbolic
21625 data on the host, and that they have absolutely no effect on the
21629 @node Remote Debugging
21630 @chapter Debugging Remote Programs
21631 @cindex remote debugging
21633 If you are trying to debug a program running on a machine that cannot run
21634 @value{GDBN} in the usual way, it is often useful to use remote debugging.
21635 For example, you might use remote debugging on an operating system kernel,
21636 or on a small system which does not have a general purpose operating system
21637 powerful enough to run a full-featured debugger.
21639 Some configurations of @value{GDBN} have special serial or TCP/IP interfaces
21640 to make this work with particular debugging targets. In addition,
21641 @value{GDBN} comes with a generic serial protocol (specific to @value{GDBN},
21642 but not specific to any particular target system) which you can use if you
21643 write the remote stubs---the code that runs on the remote system to
21644 communicate with @value{GDBN}.
21646 Other remote targets may be available in your
21647 configuration of @value{GDBN}; use @code{help target} to list them.
21650 * Connecting:: Connecting to a remote target
21651 * File Transfer:: Sending files to a remote system
21652 * Server:: Using the gdbserver program
21653 * Remote Configuration:: Remote configuration
21654 * Remote Stub:: Implementing a remote stub
21658 @section Connecting to a Remote Target
21659 @cindex remote debugging, connecting
21660 @cindex @code{gdbserver}, connecting
21661 @cindex remote debugging, types of connections
21662 @cindex @code{gdbserver}, types of connections
21663 @cindex @code{gdbserver}, @code{target remote} mode
21664 @cindex @code{gdbserver}, @code{target extended-remote} mode
21666 This section describes how to connect to a remote target, including the
21667 types of connections and their differences, how to set up executable and
21668 symbol files on the host and target, and the commands used for
21669 connecting to and disconnecting from the remote target.
21671 @subsection Types of Remote Connections
21673 @value{GDBN} supports two types of remote connections, @code{target remote}
21674 mode and @code{target extended-remote} mode. Note that many remote targets
21675 support only @code{target remote} mode. There are several major
21676 differences between the two types of connections, enumerated here:
21680 @cindex remote debugging, detach and program exit
21681 @item Result of detach or program exit
21682 @strong{With target remote mode:} When the debugged program exits or you
21683 detach from it, @value{GDBN} disconnects from the target. When using
21684 @code{gdbserver}, @code{gdbserver} will exit.
21686 @strong{With target extended-remote mode:} When the debugged program exits or
21687 you detach from it, @value{GDBN} remains connected to the target, even
21688 though no program is running. You can rerun the program, attach to a
21689 running program, or use @code{monitor} commands specific to the target.
21691 When using @code{gdbserver} in this case, it does not exit unless it was
21692 invoked using the @option{--once} option. If the @option{--once} option
21693 was not used, you can ask @code{gdbserver} to exit using the
21694 @code{monitor exit} command (@pxref{Monitor Commands for gdbserver}).
21696 @item Specifying the program to debug
21697 For both connection types you use the @code{file} command to specify the
21698 program on the host system. If you are using @code{gdbserver} there are
21699 some differences in how to specify the location of the program on the
21702 @strong{With target remote mode:} You must either specify the program to debug
21703 on the @code{gdbserver} command line or use the @option{--attach} option
21704 (@pxref{Attaching to a program,,Attaching to a Running Program}).
21706 @cindex @option{--multi}, @code{gdbserver} option
21707 @strong{With target extended-remote mode:} You may specify the program to debug
21708 on the @code{gdbserver} command line, or you can load the program or attach
21709 to it using @value{GDBN} commands after connecting to @code{gdbserver}.
21711 @anchor{--multi Option in Types of Remote Connnections}
21712 You can start @code{gdbserver} without supplying an initial command to run
21713 or process ID to attach. To do this, use the @option{--multi} command line
21714 option. Then you can connect using @code{target extended-remote} and start
21715 the program you want to debug (see below for details on using the
21716 @code{run} command in this scenario). Note that the conditions under which
21717 @code{gdbserver} terminates depend on how @value{GDBN} connects to it
21718 (@code{target remote} or @code{target extended-remote}). The
21719 @option{--multi} option to @code{gdbserver} has no influence on that.
21721 @item The @code{run} command
21722 @strong{With target remote mode:} The @code{run} command is not
21723 supported. Once a connection has been established, you can use all
21724 the usual @value{GDBN} commands to examine and change data. The
21725 remote program is already running, so you can use commands like
21726 @kbd{step} and @kbd{continue}.
21728 @strong{With target extended-remote mode:} The @code{run} command is
21729 supported. The @code{run} command uses the value set by
21730 @code{set remote exec-file} (@pxref{set remote exec-file}) to select
21731 the program to run. Command line arguments are supported, except for
21732 wildcard expansion and I/O redirection (@pxref{Arguments}).
21734 If you specify the program to debug on the command line, then the
21735 @code{run} command is not required to start execution, and you can
21736 resume using commands like @kbd{step} and @kbd{continue} as with
21737 @code{target remote} mode.
21739 @anchor{Attaching in Types of Remote Connections}
21741 @strong{With target remote mode:} The @value{GDBN} command @code{attach} is
21742 not supported. To attach to a running program using @code{gdbserver}, you
21743 must use the @option{--attach} option (@pxref{Running gdbserver}).
21745 @strong{With target extended-remote mode:} To attach to a running program,
21746 you may use the @code{attach} command after the connection has been
21747 established. If you are using @code{gdbserver}, you may also invoke
21748 @code{gdbserver} using the @option{--attach} option
21749 (@pxref{Running gdbserver}).
21753 @anchor{Host and target files}
21754 @subsection Host and Target Files
21755 @cindex remote debugging, symbol files
21756 @cindex symbol files, remote debugging
21758 @value{GDBN}, running on the host, needs access to symbol and debugging
21759 information for your program running on the target. This requires
21760 access to an unstripped copy of your program, and possibly any associated
21761 symbol files. Note that this section applies equally to both @code{target
21762 remote} mode and @code{target extended-remote} mode.
21764 Some remote targets (@pxref{qXfer executable filename read}, and
21765 @pxref{Host I/O Packets}) allow @value{GDBN} to access program files over
21766 the same connection used to communicate with @value{GDBN}. With such a
21767 target, if the remote program is unstripped, the only command you need is
21768 @code{target remote} (or @code{target extended-remote}).
21770 If the remote program is stripped, or the target does not support remote
21771 program file access, start up @value{GDBN} using the name of the local
21772 unstripped copy of your program as the first argument, or use the
21773 @code{file} command. Use @code{set sysroot} to specify the location (on
21774 the host) of target libraries (unless your @value{GDBN} was compiled with
21775 the correct sysroot using @code{--with-sysroot}). Alternatively, you
21776 may use @code{set solib-search-path} to specify how @value{GDBN} locates
21779 The symbol file and target libraries must exactly match the executable
21780 and libraries on the target, with one exception: the files on the host
21781 system should not be stripped, even if the files on the target system
21782 are. Mismatched or missing files will lead to confusing results
21783 during debugging. On @sc{gnu}/Linux targets, mismatched or missing
21784 files may also prevent @code{gdbserver} from debugging multi-threaded
21787 @subsection Remote Connection Commands
21788 @cindex remote connection commands
21789 @value{GDBN} can communicate with the target over a serial line, a
21790 local Unix domain socket, or
21791 over an @acronym{IP} network using @acronym{TCP} or @acronym{UDP}. In
21792 each case, @value{GDBN} uses the same protocol for debugging your
21793 program; only the medium carrying the debugging packets varies. The
21794 @code{target remote} and @code{target extended-remote} commands
21795 establish a connection to the target. Both commands accept the same
21796 arguments, which indicate the medium to use:
21800 @item target remote @var{serial-device}
21801 @itemx target extended-remote @var{serial-device}
21802 @cindex serial line, @code{target remote}
21803 Use @var{serial-device} to communicate with the target. For example,
21804 to use a serial line connected to the device named @file{/dev/ttyb}:
21807 target remote /dev/ttyb
21810 If you're using a serial line, you may want to give @value{GDBN} the
21811 @samp{--baud} option, or use the @code{set serial baud} command
21812 (@pxref{Remote Configuration, set serial baud}) before the
21813 @code{target} command.
21815 @item target remote @var{local-socket}
21816 @itemx target extended-remote @var{local-socket}
21817 @cindex local socket, @code{target remote}
21818 @cindex Unix domain socket
21819 Use @var{local-socket} to communicate with the target. For example,
21820 to use a local Unix domain socket bound to the file system entry @file{/tmp/gdb-socket0}:
21823 target remote /tmp/gdb-socket0
21826 Note that this command has the same form as the command to connect
21827 to a serial line. @value{GDBN} will automatically determine which
21828 kind of file you have specified and will make the appropriate kind
21830 This feature is not available if the host system does not support
21831 Unix domain sockets.
21833 @item target remote @code{@var{host}:@var{port}}
21834 @itemx target remote @code{@var{[host]}:@var{port}}
21835 @itemx target remote @code{tcp:@var{host}:@var{port}}
21836 @itemx target remote @code{tcp:@var{[host]}:@var{port}}
21837 @itemx target remote @code{tcp4:@var{host}:@var{port}}
21838 @itemx target remote @code{tcp6:@var{host}:@var{port}}
21839 @itemx target remote @code{tcp6:@var{[host]}:@var{port}}
21840 @itemx target extended-remote @code{@var{host}:@var{port}}
21841 @itemx target extended-remote @code{@var{[host]}:@var{port}}
21842 @itemx target extended-remote @code{tcp:@var{host}:@var{port}}
21843 @itemx target extended-remote @code{tcp:@var{[host]}:@var{port}}
21844 @itemx target extended-remote @code{tcp4:@var{host}:@var{port}}
21845 @itemx target extended-remote @code{tcp6:@var{host}:@var{port}}
21846 @itemx target extended-remote @code{tcp6:@var{[host]}:@var{port}}
21847 @cindex @acronym{TCP} port, @code{target remote}
21848 Debug using a @acronym{TCP} connection to @var{port} on @var{host}.
21849 The @var{host} may be either a host name, a numeric @acronym{IPv4}
21850 address, or a numeric @acronym{IPv6} address (with or without the
21851 square brackets to separate the address from the port); @var{port}
21852 must be a decimal number. The @var{host} could be the target machine
21853 itself, if it is directly connected to the net, or it might be a
21854 terminal server which in turn has a serial line to the target.
21856 For example, to connect to port 2828 on a terminal server named
21860 target remote manyfarms:2828
21863 To connect to port 2828 on a terminal server whose address is
21864 @code{2001:0db8:85a3:0000:0000:8a2e:0370:7334}, you can either use the
21865 square bracket syntax:
21868 target remote [2001:0db8:85a3:0000:0000:8a2e:0370:7334]:2828
21872 or explicitly specify the @acronym{IPv6} protocol:
21875 target remote tcp6:2001:0db8:85a3:0000:0000:8a2e:0370:7334:2828
21878 This last example may be confusing to the reader, because there is no
21879 visible separation between the hostname and the port number.
21880 Therefore, we recommend the user to provide @acronym{IPv6} addresses
21881 using square brackets for clarity. However, it is important to
21882 mention that for @value{GDBN} there is no ambiguity: the number after
21883 the last colon is considered to be the port number.
21885 If your remote target is actually running on the same machine as your
21886 debugger session (e.g.@: a simulator for your target running on the
21887 same host), you can omit the hostname. For example, to connect to
21888 port 1234 on your local machine:
21891 target remote :1234
21895 Note that the colon is still required here.
21897 @item target remote @code{udp:@var{host}:@var{port}}
21898 @itemx target remote @code{udp:@var{[host]}:@var{port}}
21899 @itemx target remote @code{udp4:@var{host}:@var{port}}
21900 @itemx target remote @code{udp6:@var{[host]}:@var{port}}
21901 @itemx target extended-remote @code{udp:@var{host}:@var{port}}
21902 @itemx target extended-remote @code{udp:@var{host}:@var{port}}
21903 @itemx target extended-remote @code{udp:@var{[host]}:@var{port}}
21904 @itemx target extended-remote @code{udp4:@var{host}:@var{port}}
21905 @itemx target extended-remote @code{udp6:@var{host}:@var{port}}
21906 @itemx target extended-remote @code{udp6:@var{[host]}:@var{port}}
21907 @cindex @acronym{UDP} port, @code{target remote}
21908 Debug using @acronym{UDP} packets to @var{port} on @var{host}. For example, to
21909 connect to @acronym{UDP} port 2828 on a terminal server named @code{manyfarms}:
21912 target remote udp:manyfarms:2828
21915 When using a @acronym{UDP} connection for remote debugging, you should
21916 keep in mind that the `U' stands for ``Unreliable''. @acronym{UDP}
21917 can silently drop packets on busy or unreliable networks, which will
21918 cause havoc with your debugging session.
21920 @item target remote | @var{command}
21921 @itemx target extended-remote | @var{command}
21922 @cindex pipe, @code{target remote} to
21923 Run @var{command} in the background and communicate with it using a
21924 pipe. The @var{command} is a shell command, to be parsed and expanded
21925 by the system's command shell, @code{/bin/sh}; it should expect remote
21926 protocol packets on its standard input, and send replies on its
21927 standard output. You could use this to run a stand-alone simulator
21928 that speaks the remote debugging protocol, to make net connections
21929 using programs like @code{ssh}, or for other similar tricks.
21931 If @var{command} closes its standard output (perhaps by exiting),
21932 @value{GDBN} will try to send it a @code{SIGTERM} signal. (If the
21933 program has already exited, this will have no effect.)
21937 @cindex interrupting remote programs
21938 @cindex remote programs, interrupting
21939 Whenever @value{GDBN} is waiting for the remote program, if you type the
21940 interrupt character (often @kbd{Ctrl-c}), @value{GDBN} attempts to stop the
21941 program. This may or may not succeed, depending in part on the hardware
21942 and the serial drivers the remote system uses. If you type the
21943 interrupt character once again, @value{GDBN} displays this prompt:
21946 Interrupted while waiting for the program.
21947 Give up (and stop debugging it)? (y or n)
21950 In @code{target remote} mode, if you type @kbd{y}, @value{GDBN} abandons
21951 the remote debugging session. (If you decide you want to try again later,
21952 you can use @kbd{target remote} again to connect once more.) If you type
21953 @kbd{n}, @value{GDBN} goes back to waiting.
21955 In @code{target extended-remote} mode, typing @kbd{n} will leave
21956 @value{GDBN} connected to the target.
21959 @kindex detach (remote)
21961 When you have finished debugging the remote program, you can use the
21962 @code{detach} command to release it from @value{GDBN} control.
21963 Detaching from the target normally resumes its execution, but the results
21964 will depend on your particular remote stub. After the @code{detach}
21965 command in @code{target remote} mode, @value{GDBN} is free to connect to
21966 another target. In @code{target extended-remote} mode, @value{GDBN} is
21967 still connected to the target.
21971 The @code{disconnect} command closes the connection to the target, and
21972 the target is generally not resumed. It will wait for @value{GDBN}
21973 (this instance or another one) to connect and continue debugging. After
21974 the @code{disconnect} command, @value{GDBN} is again free to connect to
21977 @cindex send command to remote monitor
21978 @cindex extend @value{GDBN} for remote targets
21979 @cindex add new commands for external monitor
21981 @item monitor @var{cmd}
21982 This command allows you to send arbitrary commands directly to the
21983 remote monitor. Since @value{GDBN} doesn't care about the commands it
21984 sends like this, this command is the way to extend @value{GDBN}---you
21985 can add new commands that only the external monitor will understand
21989 @node File Transfer
21990 @section Sending files to a remote system
21991 @cindex remote target, file transfer
21992 @cindex file transfer
21993 @cindex sending files to remote systems
21995 Some remote targets offer the ability to transfer files over the same
21996 connection used to communicate with @value{GDBN}. This is convenient
21997 for targets accessible through other means, e.g.@: @sc{gnu}/Linux systems
21998 running @code{gdbserver} over a network interface. For other targets,
21999 e.g.@: embedded devices with only a single serial port, this may be
22000 the only way to upload or download files.
22002 Not all remote targets support these commands.
22006 @item remote put @var{hostfile} @var{targetfile}
22007 Copy file @var{hostfile} from the host system (the machine running
22008 @value{GDBN}) to @var{targetfile} on the target system.
22011 @item remote get @var{targetfile} @var{hostfile}
22012 Copy file @var{targetfile} from the target system to @var{hostfile}
22013 on the host system.
22015 @kindex remote delete
22016 @item remote delete @var{targetfile}
22017 Delete @var{targetfile} from the target system.
22022 @section Using the @code{gdbserver} Program
22025 @cindex remote connection without stubs
22026 @code{gdbserver} is a control program for Unix-like systems, which
22027 allows you to connect your program with a remote @value{GDBN} via
22028 @code{target remote} or @code{target extended-remote}---but without
22029 linking in the usual debugging stub.
22031 @code{gdbserver} is not a complete replacement for the debugging stubs,
22032 because it requires essentially the same operating-system facilities
22033 that @value{GDBN} itself does. In fact, a system that can run
22034 @code{gdbserver} to connect to a remote @value{GDBN} could also run
22035 @value{GDBN} locally! @code{gdbserver} is sometimes useful nevertheless,
22036 because it is a much smaller program than @value{GDBN} itself. It is
22037 also easier to port than all of @value{GDBN}, so you may be able to get
22038 started more quickly on a new system by using @code{gdbserver}.
22039 Finally, if you develop code for real-time systems, you may find that
22040 the tradeoffs involved in real-time operation make it more convenient to
22041 do as much development work as possible on another system, for example
22042 by cross-compiling. You can use @code{gdbserver} to make a similar
22043 choice for debugging.
22045 @value{GDBN} and @code{gdbserver} communicate via either a serial line
22046 or a TCP connection, using the standard @value{GDBN} remote serial
22050 @emph{Warning:} @code{gdbserver} does not have any built-in security.
22051 Do not run @code{gdbserver} connected to any public network; a
22052 @value{GDBN} connection to @code{gdbserver} provides access to the
22053 target system with the same privileges as the user running
22057 @anchor{Running gdbserver}
22058 @subsection Running @code{gdbserver}
22059 @cindex arguments, to @code{gdbserver}
22060 @cindex @code{gdbserver}, command-line arguments
22062 Run @code{gdbserver} on the target system. You need a copy of the
22063 program you want to debug, including any libraries it requires.
22064 @code{gdbserver} does not need your program's symbol table, so you can
22065 strip the program if necessary to save space. @value{GDBN} on the host
22066 system does all the symbol handling.
22068 To use the server, you must tell it how to communicate with @value{GDBN};
22069 the name of your program; and the arguments for your program. The usual
22073 target> gdbserver @var{comm} @var{program} [ @var{args} @dots{} ]
22076 @var{comm} is either a device name (to use a serial line), or a TCP
22077 hostname and portnumber, or @code{-} or @code{stdio} to use
22078 stdin/stdout of @code{gdbserver}.
22079 For example, to debug Emacs with the argument
22080 @samp{foo.txt} and communicate with @value{GDBN} over the serial port
22084 target> gdbserver /dev/com1 emacs foo.txt
22087 @code{gdbserver} waits passively for the host @value{GDBN} to communicate
22090 To use a TCP connection instead of a serial line:
22093 target> gdbserver host:2345 emacs foo.txt
22096 The only difference from the previous example is the first argument,
22097 specifying that you are communicating with the host @value{GDBN} via
22098 TCP. The @samp{host:2345} argument means that @code{gdbserver} is to
22099 expect a TCP connection from machine @samp{host} to local TCP port 2345.
22100 (Currently, the @samp{host} part is ignored.) You can choose any number
22101 you want for the port number as long as it does not conflict with any
22102 TCP ports already in use on the target system (for example, @code{23} is
22103 reserved for @code{telnet}).@footnote{If you choose a port number that
22104 conflicts with another service, @code{gdbserver} prints an error message
22105 and exits.} You must use the same port number with the host @value{GDBN}
22106 @code{target remote} command.
22108 The @code{stdio} connection is useful when starting @code{gdbserver}
22112 (gdb) target remote | ssh -T hostname gdbserver - hello
22115 The @samp{-T} option to ssh is provided because we don't need a remote pty,
22116 and we don't want escape-character handling. Ssh does this by default when
22117 a command is provided, the flag is provided to make it explicit.
22118 You could elide it if you want to.
22120 Programs started with stdio-connected gdbserver have @file{/dev/null} for
22121 @code{stdin}, and @code{stdout},@code{stderr} are sent back to gdb for
22122 display through a pipe connected to gdbserver.
22123 Both @code{stdout} and @code{stderr} use the same pipe.
22125 @anchor{Attaching to a program}
22126 @subsubsection Attaching to a Running Program
22127 @cindex attach to a program, @code{gdbserver}
22128 @cindex @option{--attach}, @code{gdbserver} option
22130 On some targets, @code{gdbserver} can also attach to running programs.
22131 This is accomplished via the @code{--attach} argument. The syntax is:
22134 target> gdbserver --attach @var{comm} @var{pid}
22137 @var{pid} is the process ID of a currently running process. It isn't
22138 necessary to point @code{gdbserver} at a binary for the running process.
22140 In @code{target extended-remote} mode, you can also attach using the
22141 @value{GDBN} attach command
22142 (@pxref{Attaching in Types of Remote Connections}).
22145 You can debug processes by name instead of process ID if your target has the
22146 @code{pidof} utility:
22149 target> gdbserver --attach @var{comm} `pidof @var{program}`
22152 In case more than one copy of @var{program} is running, or @var{program}
22153 has multiple threads, most versions of @code{pidof} support the
22154 @code{-s} option to only return the first process ID.
22156 @subsubsection TCP port allocation lifecycle of @code{gdbserver}
22158 This section applies only when @code{gdbserver} is run to listen on a TCP
22161 @code{gdbserver} normally terminates after all of its debugged processes have
22162 terminated in @kbd{target remote} mode. On the other hand, for @kbd{target
22163 extended-remote}, @code{gdbserver} stays running even with no processes left.
22164 @value{GDBN} normally terminates the spawned debugged process on its exit,
22165 which normally also terminates @code{gdbserver} in the @kbd{target remote}
22166 mode. Therefore, when the connection drops unexpectedly, and @value{GDBN}
22167 cannot ask @code{gdbserver} to kill its debugged processes, @code{gdbserver}
22168 stays running even in the @kbd{target remote} mode.
22170 When @code{gdbserver} stays running, @value{GDBN} can connect to it again later.
22171 Such reconnecting is useful for features like @ref{disconnected tracing}. For
22172 completeness, at most one @value{GDBN} can be connected at a time.
22174 @cindex @option{--once}, @code{gdbserver} option
22175 By default, @code{gdbserver} keeps the listening TCP port open, so that
22176 subsequent connections are possible. However, if you start @code{gdbserver}
22177 with the @option{--once} option, it will stop listening for any further
22178 connection attempts after connecting to the first @value{GDBN} session. This
22179 means no further connections to @code{gdbserver} will be possible after the
22180 first one. It also means @code{gdbserver} will terminate after the first
22181 connection with remote @value{GDBN} has closed, even for unexpectedly closed
22182 connections and even in the @kbd{target extended-remote} mode. The
22183 @option{--once} option allows reusing the same port number for connecting to
22184 multiple instances of @code{gdbserver} running on the same host, since each
22185 instance closes its port after the first connection.
22187 @anchor{Other Command-Line Arguments for gdbserver}
22188 @subsubsection Other Command-Line Arguments for @code{gdbserver}
22190 You can use the @option{--multi} option to start @code{gdbserver} without
22191 specifying a program to debug or a process to attach to. Then you can
22192 attach in @code{target extended-remote} mode and run or attach to a
22193 program. For more information,
22194 @pxref{--multi Option in Types of Remote Connnections}.
22196 @cindex @option{--debug}, @code{gdbserver} option
22197 The @option{--debug} option tells @code{gdbserver} to display extra
22198 status information about the debugging process.
22199 @cindex @option{--remote-debug}, @code{gdbserver} option
22200 The @option{--remote-debug} option tells @code{gdbserver} to display
22201 remote protocol debug output.
22202 @cindex @option{--debug-file}, @code{gdbserver} option
22203 @cindex @code{gdbserver}, send all debug output to a single file
22204 The @option{--debug-file=@var{filename}} option tells @code{gdbserver} to
22205 write any debug output to the given @var{filename}. These options are intended
22206 for @code{gdbserver} development and for bug reports to the developers.
22208 @cindex @option{--debug-format}, @code{gdbserver} option
22209 The @option{--debug-format=option1[,option2,...]} option tells
22210 @code{gdbserver} to include additional information in each output.
22211 Possible options are:
22215 Turn off all extra information in debugging output.
22217 Turn on all extra information in debugging output.
22219 Include a timestamp in each line of debugging output.
22222 Options are processed in order. Thus, for example, if @option{none}
22223 appears last then no additional information is added to debugging output.
22225 @cindex @option{--wrapper}, @code{gdbserver} option
22226 The @option{--wrapper} option specifies a wrapper to launch programs
22227 for debugging. The option should be followed by the name of the
22228 wrapper, then any command-line arguments to pass to the wrapper, then
22229 @kbd{--} indicating the end of the wrapper arguments.
22231 @code{gdbserver} runs the specified wrapper program with a combined
22232 command line including the wrapper arguments, then the name of the
22233 program to debug, then any arguments to the program. The wrapper
22234 runs until it executes your program, and then @value{GDBN} gains control.
22236 You can use any program that eventually calls @code{execve} with
22237 its arguments as a wrapper. Several standard Unix utilities do
22238 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
22239 with @code{exec "$@@"} will also work.
22241 For example, you can use @code{env} to pass an environment variable to
22242 the debugged program, without setting the variable in @code{gdbserver}'s
22246 $ gdbserver --wrapper env LD_PRELOAD=libtest.so -- :2222 ./testprog
22249 @cindex @option{--selftest}
22250 The @option{--selftest} option runs the self tests in @code{gdbserver}:
22253 $ gdbserver --selftest
22254 Ran 2 unit tests, 0 failed
22257 These tests are disabled in release.
22258 @subsection Connecting to @code{gdbserver}
22260 The basic procedure for connecting to the remote target is:
22264 Run @value{GDBN} on the host system.
22267 Make sure you have the necessary symbol files
22268 (@pxref{Host and target files}).
22269 Load symbols for your application using the @code{file} command before you
22270 connect. Use @code{set sysroot} to locate target libraries (unless your
22271 @value{GDBN} was compiled with the correct sysroot using
22272 @code{--with-sysroot}).
22275 Connect to your target (@pxref{Connecting,,Connecting to a Remote Target}).
22276 For TCP connections, you must start up @code{gdbserver} prior to using
22277 the @code{target} command. Otherwise you may get an error whose
22278 text depends on the host system, but which usually looks something like
22279 @samp{Connection refused}. Don't use the @code{load}
22280 command in @value{GDBN} when using @code{target remote} mode, since the
22281 program is already on the target.
22285 @anchor{Monitor Commands for gdbserver}
22286 @subsection Monitor Commands for @code{gdbserver}
22287 @cindex monitor commands, for @code{gdbserver}
22289 During a @value{GDBN} session using @code{gdbserver}, you can use the
22290 @code{monitor} command to send special requests to @code{gdbserver}.
22291 Here are the available commands.
22295 List the available monitor commands.
22297 @item monitor set debug 0
22298 @itemx monitor set debug 1
22299 Disable or enable general debugging messages.
22301 @item monitor set remote-debug 0
22302 @itemx monitor set remote-debug 1
22303 Disable or enable specific debugging messages associated with the remote
22304 protocol (@pxref{Remote Protocol}).
22306 @item monitor set debug-file filename
22307 @itemx monitor set debug-file
22308 Send any debug output to the given file, or to stderr.
22310 @item monitor set debug-format option1@r{[},option2,...@r{]}
22311 Specify additional text to add to debugging messages.
22312 Possible options are:
22316 Turn off all extra information in debugging output.
22318 Turn on all extra information in debugging output.
22320 Include a timestamp in each line of debugging output.
22323 Options are processed in order. Thus, for example, if @option{none}
22324 appears last then no additional information is added to debugging output.
22326 @item monitor set libthread-db-search-path [PATH]
22327 @cindex gdbserver, search path for @code{libthread_db}
22328 When this command is issued, @var{path} is a colon-separated list of
22329 directories to search for @code{libthread_db} (@pxref{Threads,,set
22330 libthread-db-search-path}). If you omit @var{path},
22331 @samp{libthread-db-search-path} will be reset to its default value.
22333 The special entry @samp{$pdir} for @samp{libthread-db-search-path} is
22334 not supported in @code{gdbserver}.
22337 Tell gdbserver to exit immediately. This command should be followed by
22338 @code{disconnect} to close the debugging session. @code{gdbserver} will
22339 detach from any attached processes and kill any processes it created.
22340 Use @code{monitor exit} to terminate @code{gdbserver} at the end
22341 of a multi-process mode debug session.
22345 @subsection Tracepoints support in @code{gdbserver}
22346 @cindex tracepoints support in @code{gdbserver}
22348 On some targets, @code{gdbserver} supports tracepoints, fast
22349 tracepoints and static tracepoints.
22351 For fast or static tracepoints to work, a special library called the
22352 @dfn{in-process agent} (IPA), must be loaded in the inferior process.
22353 This library is built and distributed as an integral part of
22354 @code{gdbserver}. In addition, support for static tracepoints
22355 requires building the in-process agent library with static tracepoints
22356 support. At present, the UST (LTTng Userspace Tracer,
22357 @url{http://lttng.org/ust}) tracing engine is supported. This support
22358 is automatically available if UST development headers are found in the
22359 standard include path when @code{gdbserver} is built, or if
22360 @code{gdbserver} was explicitly configured using @option{--with-ust}
22361 to point at such headers. You can explicitly disable the support
22362 using @option{--with-ust=no}.
22364 There are several ways to load the in-process agent in your program:
22367 @item Specifying it as dependency at link time
22369 You can link your program dynamically with the in-process agent
22370 library. On most systems, this is accomplished by adding
22371 @code{-linproctrace} to the link command.
22373 @item Using the system's preloading mechanisms
22375 You can force loading the in-process agent at startup time by using
22376 your system's support for preloading shared libraries. Many Unixes
22377 support the concept of preloading user defined libraries. In most
22378 cases, you do that by specifying @code{LD_PRELOAD=libinproctrace.so}
22379 in the environment. See also the description of @code{gdbserver}'s
22380 @option{--wrapper} command line option.
22382 @item Using @value{GDBN} to force loading the agent at run time
22384 On some systems, you can force the inferior to load a shared library,
22385 by calling a dynamic loader function in the inferior that takes care
22386 of dynamically looking up and loading a shared library. On most Unix
22387 systems, the function is @code{dlopen}. You'll use the @code{call}
22388 command for that. For example:
22391 (@value{GDBP}) call dlopen ("libinproctrace.so", ...)
22394 Note that on most Unix systems, for the @code{dlopen} function to be
22395 available, the program needs to be linked with @code{-ldl}.
22398 On systems that have a userspace dynamic loader, like most Unix
22399 systems, when you connect to @code{gdbserver} using @code{target
22400 remote}, you'll find that the program is stopped at the dynamic
22401 loader's entry point, and no shared library has been loaded in the
22402 program's address space yet, including the in-process agent. In that
22403 case, before being able to use any of the fast or static tracepoints
22404 features, you need to let the loader run and load the shared
22405 libraries. The simplest way to do that is to run the program to the
22406 main procedure. E.g., if debugging a C or C@t{++} program, start
22407 @code{gdbserver} like so:
22410 $ gdbserver :9999 myprogram
22413 Start GDB and connect to @code{gdbserver} like so, and run to main:
22417 (@value{GDBP}) target remote myhost:9999
22418 0x00007f215893ba60 in ?? () from /lib64/ld-linux-x86-64.so.2
22419 (@value{GDBP}) b main
22420 (@value{GDBP}) continue
22423 The in-process tracing agent library should now be loaded into the
22424 process; you can confirm it with the @code{info sharedlibrary}
22425 command, which will list @file{libinproctrace.so} as loaded in the
22426 process. You are now ready to install fast tracepoints, list static
22427 tracepoint markers, probe static tracepoints markers, and start
22430 @node Remote Configuration
22431 @section Remote Configuration
22434 @kindex show remote
22435 This section documents the configuration options available when
22436 debugging remote programs. For the options related to the File I/O
22437 extensions of the remote protocol, see @ref{system,
22438 system-call-allowed}.
22441 @item set remoteaddresssize @var{bits}
22442 @cindex address size for remote targets
22443 @cindex bits in remote address
22444 Set the maximum size of address in a memory packet to the specified
22445 number of bits. @value{GDBN} will mask off the address bits above
22446 that number, when it passes addresses to the remote target. The
22447 default value is the number of bits in the target's address.
22449 @item show remoteaddresssize
22450 Show the current value of remote address size in bits.
22452 @item set serial baud @var{n}
22453 @cindex baud rate for remote targets
22454 Set the baud rate for the remote serial I/O to @var{n} baud. The
22455 value is used to set the speed of the serial port used for debugging
22458 @item show serial baud
22459 Show the current speed of the remote connection.
22461 @item set serial parity @var{parity}
22462 Set the parity for the remote serial I/O. Supported values of @var{parity} are:
22463 @code{even}, @code{none}, and @code{odd}. The default is @code{none}.
22465 @item show serial parity
22466 Show the current parity of the serial port.
22468 @item set remotebreak
22469 @cindex interrupt remote programs
22470 @cindex BREAK signal instead of Ctrl-C
22471 @anchor{set remotebreak}
22472 If set to on, @value{GDBN} sends a @code{BREAK} signal to the remote
22473 when you type @kbd{Ctrl-c} to interrupt the program running
22474 on the remote. If set to off, @value{GDBN} sends the @samp{Ctrl-C}
22475 character instead. The default is off, since most remote systems
22476 expect to see @samp{Ctrl-C} as the interrupt signal.
22478 @item show remotebreak
22479 Show whether @value{GDBN} sends @code{BREAK} or @samp{Ctrl-C} to
22480 interrupt the remote program.
22482 @item set remoteflow on
22483 @itemx set remoteflow off
22484 @kindex set remoteflow
22485 Enable or disable hardware flow control (@code{RTS}/@code{CTS})
22486 on the serial port used to communicate to the remote target.
22488 @item show remoteflow
22489 @kindex show remoteflow
22490 Show the current setting of hardware flow control.
22492 @item set remotelogbase @var{base}
22493 Set the base (a.k.a.@: radix) of logging serial protocol
22494 communications to @var{base}. Supported values of @var{base} are:
22495 @code{ascii}, @code{octal}, and @code{hex}. The default is
22498 @item show remotelogbase
22499 Show the current setting of the radix for logging remote serial
22502 @item set remotelogfile @var{file}
22503 @cindex record serial communications on file
22504 Record remote serial communications on the named @var{file}. The
22505 default is not to record at all.
22507 @item show remotelogfile
22508 Show the current setting of the file name on which to record the
22509 serial communications.
22511 @item set remotetimeout @var{num}
22512 @cindex timeout for serial communications
22513 @cindex remote timeout
22514 Set the timeout limit to wait for the remote target to respond to
22515 @var{num} seconds. The default is 2 seconds.
22517 @item show remotetimeout
22518 Show the current number of seconds to wait for the remote target
22521 @cindex limit hardware breakpoints and watchpoints
22522 @cindex remote target, limit break- and watchpoints
22523 @anchor{set remote hardware-watchpoint-limit}
22524 @anchor{set remote hardware-breakpoint-limit}
22525 @item set remote hardware-watchpoint-limit @var{limit}
22526 @itemx set remote hardware-breakpoint-limit @var{limit}
22527 Restrict @value{GDBN} to using @var{limit} remote hardware watchpoints
22528 or breakpoints. The @var{limit} can be set to 0 to disable hardware
22529 watchpoints or breakpoints, and @code{unlimited} for unlimited
22530 watchpoints or breakpoints.
22532 @item show remote hardware-watchpoint-limit
22533 @itemx show remote hardware-breakpoint-limit
22534 Show the current limit for the number of hardware watchpoints or
22535 breakpoints that @value{GDBN} can use.
22537 @cindex limit hardware watchpoints length
22538 @cindex remote target, limit watchpoints length
22539 @anchor{set remote hardware-watchpoint-length-limit}
22540 @item set remote hardware-watchpoint-length-limit @var{limit}
22541 Restrict @value{GDBN} to using @var{limit} bytes for the maximum
22542 length of a remote hardware watchpoint. A @var{limit} of 0 disables
22543 hardware watchpoints and @code{unlimited} allows watchpoints of any
22546 @item show remote hardware-watchpoint-length-limit
22547 Show the current limit (in bytes) of the maximum length of
22548 a remote hardware watchpoint.
22550 @item set remote exec-file @var{filename}
22551 @itemx show remote exec-file
22552 @anchor{set remote exec-file}
22553 @cindex executable file, for remote target
22554 Select the file used for @code{run} with @code{target
22555 extended-remote}. This should be set to a filename valid on the
22556 target system. If it is not set, the target will use a default
22557 filename (e.g.@: the last program run).
22559 @item set remote interrupt-sequence
22560 @cindex interrupt remote programs
22561 @cindex select Ctrl-C, BREAK or BREAK-g
22562 Allow the user to select one of @samp{Ctrl-C}, a @code{BREAK} or
22563 @samp{BREAK-g} as the
22564 sequence to the remote target in order to interrupt the execution.
22565 @samp{Ctrl-C} is a default. Some system prefers @code{BREAK} which
22566 is high level of serial line for some certain time.
22567 Linux kernel prefers @samp{BREAK-g}, a.k.a Magic SysRq g.
22568 It is @code{BREAK} signal followed by character @code{g}.
22570 @item show interrupt-sequence
22571 Show which of @samp{Ctrl-C}, @code{BREAK} or @code{BREAK-g}
22572 is sent by @value{GDBN} to interrupt the remote program.
22573 @code{BREAK-g} is BREAK signal followed by @code{g} and
22574 also known as Magic SysRq g.
22576 @item set remote interrupt-on-connect
22577 @cindex send interrupt-sequence on start
22578 Specify whether interrupt-sequence is sent to remote target when
22579 @value{GDBN} connects to it. This is mostly needed when you debug
22580 Linux kernel. Linux kernel expects @code{BREAK} followed by @code{g}
22581 which is known as Magic SysRq g in order to connect @value{GDBN}.
22583 @item show interrupt-on-connect
22584 Show whether interrupt-sequence is sent
22585 to remote target when @value{GDBN} connects to it.
22589 @item set tcp auto-retry on
22590 @cindex auto-retry, for remote TCP target
22591 Enable auto-retry for remote TCP connections. This is useful if the remote
22592 debugging agent is launched in parallel with @value{GDBN}; there is a race
22593 condition because the agent may not become ready to accept the connection
22594 before @value{GDBN} attempts to connect. When auto-retry is
22595 enabled, if the initial attempt to connect fails, @value{GDBN} reattempts
22596 to establish the connection using the timeout specified by
22597 @code{set tcp connect-timeout}.
22599 @item set tcp auto-retry off
22600 Do not auto-retry failed TCP connections.
22602 @item show tcp auto-retry
22603 Show the current auto-retry setting.
22605 @item set tcp connect-timeout @var{seconds}
22606 @itemx set tcp connect-timeout unlimited
22607 @cindex connection timeout, for remote TCP target
22608 @cindex timeout, for remote target connection
22609 Set the timeout for establishing a TCP connection to the remote target to
22610 @var{seconds}. The timeout affects both polling to retry failed connections
22611 (enabled by @code{set tcp auto-retry on}) and waiting for connections
22612 that are merely slow to complete, and represents an approximate cumulative
22613 value. If @var{seconds} is @code{unlimited}, there is no timeout and
22614 @value{GDBN} will keep attempting to establish a connection forever,
22615 unless interrupted with @kbd{Ctrl-c}. The default is 15 seconds.
22617 @item show tcp connect-timeout
22618 Show the current connection timeout setting.
22621 @cindex remote packets, enabling and disabling
22622 The @value{GDBN} remote protocol autodetects the packets supported by
22623 your debugging stub. If you need to override the autodetection, you
22624 can use these commands to enable or disable individual packets. Each
22625 packet can be set to @samp{on} (the remote target supports this
22626 packet), @samp{off} (the remote target does not support this packet),
22627 or @samp{auto} (detect remote target support for this packet). They
22628 all default to @samp{auto}. For more information about each packet,
22629 see @ref{Remote Protocol}.
22631 During normal use, you should not have to use any of these commands.
22632 If you do, that may be a bug in your remote debugging stub, or a bug
22633 in @value{GDBN}. You may want to report the problem to the
22634 @value{GDBN} developers.
22636 For each packet @var{name}, the command to enable or disable the
22637 packet is @code{set remote @var{name}-packet}. The available settings
22640 @multitable @columnfractions 0.28 0.32 0.25
22643 @tab Related Features
22645 @item @code{fetch-register}
22647 @tab @code{info registers}
22649 @item @code{set-register}
22653 @item @code{binary-download}
22655 @tab @code{load}, @code{set}
22657 @item @code{read-aux-vector}
22658 @tab @code{qXfer:auxv:read}
22659 @tab @code{info auxv}
22661 @item @code{symbol-lookup}
22662 @tab @code{qSymbol}
22663 @tab Detecting multiple threads
22665 @item @code{attach}
22666 @tab @code{vAttach}
22669 @item @code{verbose-resume}
22671 @tab Stepping or resuming multiple threads
22677 @item @code{software-breakpoint}
22681 @item @code{hardware-breakpoint}
22685 @item @code{write-watchpoint}
22689 @item @code{read-watchpoint}
22693 @item @code{access-watchpoint}
22697 @item @code{pid-to-exec-file}
22698 @tab @code{qXfer:exec-file:read}
22699 @tab @code{attach}, @code{run}
22701 @item @code{target-features}
22702 @tab @code{qXfer:features:read}
22703 @tab @code{set architecture}
22705 @item @code{library-info}
22706 @tab @code{qXfer:libraries:read}
22707 @tab @code{info sharedlibrary}
22709 @item @code{memory-map}
22710 @tab @code{qXfer:memory-map:read}
22711 @tab @code{info mem}
22713 @item @code{read-sdata-object}
22714 @tab @code{qXfer:sdata:read}
22715 @tab @code{print $_sdata}
22717 @item @code{read-siginfo-object}
22718 @tab @code{qXfer:siginfo:read}
22719 @tab @code{print $_siginfo}
22721 @item @code{write-siginfo-object}
22722 @tab @code{qXfer:siginfo:write}
22723 @tab @code{set $_siginfo}
22725 @item @code{threads}
22726 @tab @code{qXfer:threads:read}
22727 @tab @code{info threads}
22729 @item @code{get-thread-local-@*storage-address}
22730 @tab @code{qGetTLSAddr}
22731 @tab Displaying @code{__thread} variables
22733 @item @code{get-thread-information-block-address}
22734 @tab @code{qGetTIBAddr}
22735 @tab Display MS-Windows Thread Information Block.
22737 @item @code{search-memory}
22738 @tab @code{qSearch:memory}
22741 @item @code{supported-packets}
22742 @tab @code{qSupported}
22743 @tab Remote communications parameters
22745 @item @code{catch-syscalls}
22746 @tab @code{QCatchSyscalls}
22747 @tab @code{catch syscall}
22749 @item @code{pass-signals}
22750 @tab @code{QPassSignals}
22751 @tab @code{handle @var{signal}}
22753 @item @code{program-signals}
22754 @tab @code{QProgramSignals}
22755 @tab @code{handle @var{signal}}
22757 @item @code{hostio-close-packet}
22758 @tab @code{vFile:close}
22759 @tab @code{remote get}, @code{remote put}
22761 @item @code{hostio-open-packet}
22762 @tab @code{vFile:open}
22763 @tab @code{remote get}, @code{remote put}
22765 @item @code{hostio-pread-packet}
22766 @tab @code{vFile:pread}
22767 @tab @code{remote get}, @code{remote put}
22769 @item @code{hostio-pwrite-packet}
22770 @tab @code{vFile:pwrite}
22771 @tab @code{remote get}, @code{remote put}
22773 @item @code{hostio-unlink-packet}
22774 @tab @code{vFile:unlink}
22775 @tab @code{remote delete}
22777 @item @code{hostio-readlink-packet}
22778 @tab @code{vFile:readlink}
22781 @item @code{hostio-fstat-packet}
22782 @tab @code{vFile:fstat}
22785 @item @code{hostio-setfs-packet}
22786 @tab @code{vFile:setfs}
22789 @item @code{noack-packet}
22790 @tab @code{QStartNoAckMode}
22791 @tab Packet acknowledgment
22793 @item @code{osdata}
22794 @tab @code{qXfer:osdata:read}
22795 @tab @code{info os}
22797 @item @code{query-attached}
22798 @tab @code{qAttached}
22799 @tab Querying remote process attach state.
22801 @item @code{trace-buffer-size}
22802 @tab @code{QTBuffer:size}
22803 @tab @code{set trace-buffer-size}
22805 @item @code{trace-status}
22806 @tab @code{qTStatus}
22807 @tab @code{tstatus}
22809 @item @code{traceframe-info}
22810 @tab @code{qXfer:traceframe-info:read}
22811 @tab Traceframe info
22813 @item @code{install-in-trace}
22814 @tab @code{InstallInTrace}
22815 @tab Install tracepoint in tracing
22817 @item @code{disable-randomization}
22818 @tab @code{QDisableRandomization}
22819 @tab @code{set disable-randomization}
22821 @item @code{startup-with-shell}
22822 @tab @code{QStartupWithShell}
22823 @tab @code{set startup-with-shell}
22825 @item @code{environment-hex-encoded}
22826 @tab @code{QEnvironmentHexEncoded}
22827 @tab @code{set environment}
22829 @item @code{environment-unset}
22830 @tab @code{QEnvironmentUnset}
22831 @tab @code{unset environment}
22833 @item @code{environment-reset}
22834 @tab @code{QEnvironmentReset}
22835 @tab @code{Reset the inferior environment (i.e., unset user-set variables)}
22837 @item @code{set-working-dir}
22838 @tab @code{QSetWorkingDir}
22839 @tab @code{set cwd}
22841 @item @code{conditional-breakpoints-packet}
22842 @tab @code{Z0 and Z1}
22843 @tab @code{Support for target-side breakpoint condition evaluation}
22845 @item @code{multiprocess-extensions}
22846 @tab @code{multiprocess extensions}
22847 @tab Debug multiple processes and remote process PID awareness
22849 @item @code{swbreak-feature}
22850 @tab @code{swbreak stop reason}
22853 @item @code{hwbreak-feature}
22854 @tab @code{hwbreak stop reason}
22857 @item @code{fork-event-feature}
22858 @tab @code{fork stop reason}
22861 @item @code{vfork-event-feature}
22862 @tab @code{vfork stop reason}
22865 @item @code{exec-event-feature}
22866 @tab @code{exec stop reason}
22869 @item @code{thread-events}
22870 @tab @code{QThreadEvents}
22871 @tab Tracking thread lifetime.
22873 @item @code{no-resumed-stop-reply}
22874 @tab @code{no resumed thread left stop reply}
22875 @tab Tracking thread lifetime.
22880 @section Implementing a Remote Stub
22882 @cindex debugging stub, example
22883 @cindex remote stub, example
22884 @cindex stub example, remote debugging
22885 The stub files provided with @value{GDBN} implement the target side of the
22886 communication protocol, and the @value{GDBN} side is implemented in the
22887 @value{GDBN} source file @file{remote.c}. Normally, you can simply allow
22888 these subroutines to communicate, and ignore the details. (If you're
22889 implementing your own stub file, you can still ignore the details: start
22890 with one of the existing stub files. @file{sparc-stub.c} is the best
22891 organized, and therefore the easiest to read.)
22893 @cindex remote serial debugging, overview
22894 To debug a program running on another machine (the debugging
22895 @dfn{target} machine), you must first arrange for all the usual
22896 prerequisites for the program to run by itself. For example, for a C
22901 A startup routine to set up the C runtime environment; these usually
22902 have a name like @file{crt0}. The startup routine may be supplied by
22903 your hardware supplier, or you may have to write your own.
22906 A C subroutine library to support your program's
22907 subroutine calls, notably managing input and output.
22910 A way of getting your program to the other machine---for example, a
22911 download program. These are often supplied by the hardware
22912 manufacturer, but you may have to write your own from hardware
22916 The next step is to arrange for your program to use a serial port to
22917 communicate with the machine where @value{GDBN} is running (the @dfn{host}
22918 machine). In general terms, the scheme looks like this:
22922 @value{GDBN} already understands how to use this protocol; when everything
22923 else is set up, you can simply use the @samp{target remote} command
22924 (@pxref{Targets,,Specifying a Debugging Target}).
22926 @item On the target,
22927 you must link with your program a few special-purpose subroutines that
22928 implement the @value{GDBN} remote serial protocol. The file containing these
22929 subroutines is called a @dfn{debugging stub}.
22931 On certain remote targets, you can use an auxiliary program
22932 @code{gdbserver} instead of linking a stub into your program.
22933 @xref{Server,,Using the @code{gdbserver} Program}, for details.
22936 The debugging stub is specific to the architecture of the remote
22937 machine; for example, use @file{sparc-stub.c} to debug programs on
22940 @cindex remote serial stub list
22941 These working remote stubs are distributed with @value{GDBN}:
22946 @cindex @file{i386-stub.c}
22949 For Intel 386 and compatible architectures.
22952 @cindex @file{m68k-stub.c}
22953 @cindex Motorola 680x0
22955 For Motorola 680x0 architectures.
22958 @cindex @file{sh-stub.c}
22961 For Renesas SH architectures.
22964 @cindex @file{sparc-stub.c}
22966 For @sc{sparc} architectures.
22968 @item sparcl-stub.c
22969 @cindex @file{sparcl-stub.c}
22972 For Fujitsu @sc{sparclite} architectures.
22976 The @file{README} file in the @value{GDBN} distribution may list other
22977 recently added stubs.
22980 * Stub Contents:: What the stub can do for you
22981 * Bootstrapping:: What you must do for the stub
22982 * Debug Session:: Putting it all together
22985 @node Stub Contents
22986 @subsection What the Stub Can Do for You
22988 @cindex remote serial stub
22989 The debugging stub for your architecture supplies these three
22993 @item set_debug_traps
22994 @findex set_debug_traps
22995 @cindex remote serial stub, initialization
22996 This routine arranges for @code{handle_exception} to run when your
22997 program stops. You must call this subroutine explicitly in your
22998 program's startup code.
23000 @item handle_exception
23001 @findex handle_exception
23002 @cindex remote serial stub, main routine
23003 This is the central workhorse, but your program never calls it
23004 explicitly---the setup code arranges for @code{handle_exception} to
23005 run when a trap is triggered.
23007 @code{handle_exception} takes control when your program stops during
23008 execution (for example, on a breakpoint), and mediates communications
23009 with @value{GDBN} on the host machine. This is where the communications
23010 protocol is implemented; @code{handle_exception} acts as the @value{GDBN}
23011 representative on the target machine. It begins by sending summary
23012 information on the state of your program, then continues to execute,
23013 retrieving and transmitting any information @value{GDBN} needs, until you
23014 execute a @value{GDBN} command that makes your program resume; at that point,
23015 @code{handle_exception} returns control to your own code on the target
23019 @cindex @code{breakpoint} subroutine, remote
23020 Use this auxiliary subroutine to make your program contain a
23021 breakpoint. Depending on the particular situation, this may be the only
23022 way for @value{GDBN} to get control. For instance, if your target
23023 machine has some sort of interrupt button, you won't need to call this;
23024 pressing the interrupt button transfers control to
23025 @code{handle_exception}---in effect, to @value{GDBN}. On some machines,
23026 simply receiving characters on the serial port may also trigger a trap;
23027 again, in that situation, you don't need to call @code{breakpoint} from
23028 your own program---simply running @samp{target remote} from the host
23029 @value{GDBN} session gets control.
23031 Call @code{breakpoint} if none of these is true, or if you simply want
23032 to make certain your program stops at a predetermined point for the
23033 start of your debugging session.
23036 @node Bootstrapping
23037 @subsection What You Must Do for the Stub
23039 @cindex remote stub, support routines
23040 The debugging stubs that come with @value{GDBN} are set up for a particular
23041 chip architecture, but they have no information about the rest of your
23042 debugging target machine.
23044 First of all you need to tell the stub how to communicate with the
23048 @item int getDebugChar()
23049 @findex getDebugChar
23050 Write this subroutine to read a single character from the serial port.
23051 It may be identical to @code{getchar} for your target system; a
23052 different name is used to allow you to distinguish the two if you wish.
23054 @item void putDebugChar(int)
23055 @findex putDebugChar
23056 Write this subroutine to write a single character to the serial port.
23057 It may be identical to @code{putchar} for your target system; a
23058 different name is used to allow you to distinguish the two if you wish.
23061 @cindex control C, and remote debugging
23062 @cindex interrupting remote targets
23063 If you want @value{GDBN} to be able to stop your program while it is
23064 running, you need to use an interrupt-driven serial driver, and arrange
23065 for it to stop when it receives a @code{^C} (@samp{\003}, the control-C
23066 character). That is the character which @value{GDBN} uses to tell the
23067 remote system to stop.
23069 Getting the debugging target to return the proper status to @value{GDBN}
23070 probably requires changes to the standard stub; one quick and dirty way
23071 is to just execute a breakpoint instruction (the ``dirty'' part is that
23072 @value{GDBN} reports a @code{SIGTRAP} instead of a @code{SIGINT}).
23074 Other routines you need to supply are:
23077 @item void exceptionHandler (int @var{exception_number}, void *@var{exception_address})
23078 @findex exceptionHandler
23079 Write this function to install @var{exception_address} in the exception
23080 handling tables. You need to do this because the stub does not have any
23081 way of knowing what the exception handling tables on your target system
23082 are like (for example, the processor's table might be in @sc{rom},
23083 containing entries which point to a table in @sc{ram}).
23084 The @var{exception_number} specifies the exception which should be changed;
23085 its meaning is architecture-dependent (for example, different numbers
23086 might represent divide by zero, misaligned access, etc). When this
23087 exception occurs, control should be transferred directly to
23088 @var{exception_address}, and the processor state (stack, registers,
23089 and so on) should be just as it is when a processor exception occurs. So if
23090 you want to use a jump instruction to reach @var{exception_address}, it
23091 should be a simple jump, not a jump to subroutine.
23093 For the 386, @var{exception_address} should be installed as an interrupt
23094 gate so that interrupts are masked while the handler runs. The gate
23095 should be at privilege level 0 (the most privileged level). The
23096 @sc{sparc} and 68k stubs are able to mask interrupts themselves without
23097 help from @code{exceptionHandler}.
23099 @item void flush_i_cache()
23100 @findex flush_i_cache
23101 On @sc{sparc} and @sc{sparclite} only, write this subroutine to flush the
23102 instruction cache, if any, on your target machine. If there is no
23103 instruction cache, this subroutine may be a no-op.
23105 On target machines that have instruction caches, @value{GDBN} requires this
23106 function to make certain that the state of your program is stable.
23110 You must also make sure this library routine is available:
23113 @item void *memset(void *, int, int)
23115 This is the standard library function @code{memset} that sets an area of
23116 memory to a known value. If you have one of the free versions of
23117 @code{libc.a}, @code{memset} can be found there; otherwise, you must
23118 either obtain it from your hardware manufacturer, or write your own.
23121 If you do not use the GNU C compiler, you may need other standard
23122 library subroutines as well; this varies from one stub to another,
23123 but in general the stubs are likely to use any of the common library
23124 subroutines which @code{@value{NGCC}} generates as inline code.
23127 @node Debug Session
23128 @subsection Putting it All Together
23130 @cindex remote serial debugging summary
23131 In summary, when your program is ready to debug, you must follow these
23136 Make sure you have defined the supporting low-level routines
23137 (@pxref{Bootstrapping,,What You Must Do for the Stub}):
23139 @code{getDebugChar}, @code{putDebugChar},
23140 @code{flush_i_cache}, @code{memset}, @code{exceptionHandler}.
23144 Insert these lines in your program's startup code, before the main
23145 procedure is called:
23152 On some machines, when a breakpoint trap is raised, the hardware
23153 automatically makes the PC point to the instruction after the
23154 breakpoint. If your machine doesn't do that, you may need to adjust
23155 @code{handle_exception} to arrange for it to return to the instruction
23156 after the breakpoint on this first invocation, so that your program
23157 doesn't keep hitting the initial breakpoint instead of making
23161 For the 680x0 stub only, you need to provide a variable called
23162 @code{exceptionHook}. Normally you just use:
23165 void (*exceptionHook)() = 0;
23169 but if before calling @code{set_debug_traps}, you set it to point to a
23170 function in your program, that function is called when
23171 @code{@value{GDBN}} continues after stopping on a trap (for example, bus
23172 error). The function indicated by @code{exceptionHook} is called with
23173 one parameter: an @code{int} which is the exception number.
23176 Compile and link together: your program, the @value{GDBN} debugging stub for
23177 your target architecture, and the supporting subroutines.
23180 Make sure you have a serial connection between your target machine and
23181 the @value{GDBN} host, and identify the serial port on the host.
23184 @c The "remote" target now provides a `load' command, so we should
23185 @c document that. FIXME.
23186 Download your program to your target machine (or get it there by
23187 whatever means the manufacturer provides), and start it.
23190 Start @value{GDBN} on the host, and connect to the target
23191 (@pxref{Connecting,,Connecting to a Remote Target}).
23195 @node Configurations
23196 @chapter Configuration-Specific Information
23198 While nearly all @value{GDBN} commands are available for all native and
23199 cross versions of the debugger, there are some exceptions. This chapter
23200 describes things that are only available in certain configurations.
23202 There are three major categories of configurations: native
23203 configurations, where the host and target are the same, embedded
23204 operating system configurations, which are usually the same for several
23205 different processor architectures, and bare embedded processors, which
23206 are quite different from each other.
23211 * Embedded Processors::
23218 This section describes details specific to particular native
23222 * BSD libkvm Interface:: Debugging BSD kernel memory images
23223 * Process Information:: Process information
23224 * DJGPP Native:: Features specific to the DJGPP port
23225 * Cygwin Native:: Features specific to the Cygwin port
23226 * Hurd Native:: Features specific to @sc{gnu} Hurd
23227 * Darwin:: Features specific to Darwin
23228 * FreeBSD:: Features specific to FreeBSD
23231 @node BSD libkvm Interface
23232 @subsection BSD libkvm Interface
23235 @cindex kernel memory image
23236 @cindex kernel crash dump
23238 BSD-derived systems (FreeBSD/NetBSD/OpenBSD) have a kernel memory
23239 interface that provides a uniform interface for accessing kernel virtual
23240 memory images, including live systems and crash dumps. @value{GDBN}
23241 uses this interface to allow you to debug live kernels and kernel crash
23242 dumps on many native BSD configurations. This is implemented as a
23243 special @code{kvm} debugging target. For debugging a live system, load
23244 the currently running kernel into @value{GDBN} and connect to the
23248 (@value{GDBP}) @b{target kvm}
23251 For debugging crash dumps, provide the file name of the crash dump as an
23255 (@value{GDBP}) @b{target kvm /var/crash/bsd.0}
23258 Once connected to the @code{kvm} target, the following commands are
23264 Set current context from the @dfn{Process Control Block} (PCB) address.
23267 Set current context from proc address. This command isn't available on
23268 modern FreeBSD systems.
23271 @node Process Information
23272 @subsection Process Information
23274 @cindex examine process image
23275 @cindex process info via @file{/proc}
23277 Some operating systems provide interfaces to fetch additional
23278 information about running processes beyond memory and per-thread
23279 register state. If @value{GDBN} is configured for an operating system
23280 with a supported interface, the command @code{info proc} is available
23281 to report information about the process running your program, or about
23282 any process running on your system.
23284 One supported interface is a facility called @samp{/proc} that can be
23285 used to examine the image of a running process using file-system
23286 subroutines. This facility is supported on @sc{gnu}/Linux and Solaris
23289 On FreeBSD systems, system control nodes are used to query process
23292 In addition, some systems may provide additional process information
23293 in core files. Note that a core file may include a subset of the
23294 information available from a live process. Process information is
23295 currently available from cores created on @sc{gnu}/Linux and FreeBSD
23302 @itemx info proc @var{process-id}
23303 Summarize available information about a process. If a
23304 process ID is specified by @var{process-id}, display information about
23305 that process; otherwise display information about the program being
23306 debugged. The summary includes the debugged process ID, the command
23307 line used to invoke it, its current working directory, and its
23308 executable file's absolute file name.
23310 On some systems, @var{process-id} can be of the form
23311 @samp{[@var{pid}]/@var{tid}} which specifies a certain thread ID
23312 within a process. If the optional @var{pid} part is missing, it means
23313 a thread from the process being debugged (the leading @samp{/} still
23314 needs to be present, or else @value{GDBN} will interpret the number as
23315 a process ID rather than a thread ID).
23317 @item info proc cmdline
23318 @cindex info proc cmdline
23319 Show the original command line of the process. This command is
23320 supported on @sc{gnu}/Linux and FreeBSD.
23322 @item info proc cwd
23323 @cindex info proc cwd
23324 Show the current working directory of the process. This command is
23325 supported on @sc{gnu}/Linux and FreeBSD.
23327 @item info proc exe
23328 @cindex info proc exe
23329 Show the name of executable of the process. This command is supported
23330 on @sc{gnu}/Linux and FreeBSD.
23332 @item info proc files
23333 @cindex info proc files
23334 Show the file descriptors open by the process. For each open file
23335 descriptor, @value{GDBN} shows its number, type (file, directory,
23336 character device, socket), file pointer offset, and the name of the
23337 resource open on the descriptor. The resource name can be a file name
23338 (for files, directories, and devices) or a protocol followed by socket
23339 address (for network connections). This command is supported on
23342 This example shows the open file descriptors for a process using a
23343 tty for standard input and output as well as two network sockets:
23346 (gdb) info proc files 22136
23350 FD Type Offset Flags Name
23351 text file - r-------- /usr/bin/ssh
23352 ctty chr - rw------- /dev/pts/20
23353 cwd dir - r-------- /usr/home/john
23354 root dir - r-------- /
23355 0 chr 0x32933a4 rw------- /dev/pts/20
23356 1 chr 0x32933a4 rw------- /dev/pts/20
23357 2 chr 0x32933a4 rw------- /dev/pts/20
23358 3 socket 0x0 rw----n-- tcp4 10.0.1.2:53014 -> 10.0.1.10:22
23359 4 socket 0x0 rw------- unix stream:/tmp/ssh-FIt89oAzOn5f/agent.2456
23362 @item info proc mappings
23363 @cindex memory address space mappings
23364 Report the memory address space ranges accessible in a process. On
23365 Solaris and FreeBSD systems, each memory range includes information on
23366 whether the process has read, write, or execute access rights to each
23367 range. On @sc{gnu}/Linux and FreeBSD systems, each memory range
23368 includes the object file which is mapped to that range.
23370 @item info proc stat
23371 @itemx info proc status
23372 @cindex process detailed status information
23373 Show additional process-related information, including the user ID and
23374 group ID; virtual memory usage; the signals that are pending, blocked,
23375 and ignored; its TTY; its consumption of system and user time; its
23376 stack size; its @samp{nice} value; etc. These commands are supported
23377 on @sc{gnu}/Linux and FreeBSD.
23379 For @sc{gnu}/Linux systems, see the @samp{proc} man page for more
23380 information (type @kbd{man 5 proc} from your shell prompt).
23382 For FreeBSD systems, @code{info proc stat} is an alias for @code{info
23385 @item info proc all
23386 Show all the information about the process described under all of the
23387 above @code{info proc} subcommands.
23390 @comment These sub-options of 'info proc' were not included when
23391 @comment procfs.c was re-written. Keep their descriptions around
23392 @comment against the day when someone finds the time to put them back in.
23393 @kindex info proc times
23394 @item info proc times
23395 Starting time, user CPU time, and system CPU time for your program and
23398 @kindex info proc id
23400 Report on the process IDs related to your program: its own process ID,
23401 the ID of its parent, the process group ID, and the session ID.
23404 @item set procfs-trace
23405 @kindex set procfs-trace
23406 @cindex @code{procfs} API calls
23407 This command enables and disables tracing of @code{procfs} API calls.
23409 @item show procfs-trace
23410 @kindex show procfs-trace
23411 Show the current state of @code{procfs} API call tracing.
23413 @item set procfs-file @var{file}
23414 @kindex set procfs-file
23415 Tell @value{GDBN} to write @code{procfs} API trace to the named
23416 @var{file}. @value{GDBN} appends the trace info to the previous
23417 contents of the file. The default is to display the trace on the
23420 @item show procfs-file
23421 @kindex show procfs-file
23422 Show the file to which @code{procfs} API trace is written.
23424 @item proc-trace-entry
23425 @itemx proc-trace-exit
23426 @itemx proc-untrace-entry
23427 @itemx proc-untrace-exit
23428 @kindex proc-trace-entry
23429 @kindex proc-trace-exit
23430 @kindex proc-untrace-entry
23431 @kindex proc-untrace-exit
23432 These commands enable and disable tracing of entries into and exits
23433 from the @code{syscall} interface.
23436 @kindex info pidlist
23437 @cindex process list, QNX Neutrino
23438 For QNX Neutrino only, this command displays the list of all the
23439 processes and all the threads within each process.
23442 @kindex info meminfo
23443 @cindex mapinfo list, QNX Neutrino
23444 For QNX Neutrino only, this command displays the list of all mapinfos.
23448 @subsection Features for Debugging @sc{djgpp} Programs
23449 @cindex @sc{djgpp} debugging
23450 @cindex native @sc{djgpp} debugging
23451 @cindex MS-DOS-specific commands
23454 @sc{djgpp} is a port of the @sc{gnu} development tools to MS-DOS and
23455 MS-Windows. @sc{djgpp} programs are 32-bit protected-mode programs
23456 that use the @dfn{DPMI} (DOS Protected-Mode Interface) API to run on
23457 top of real-mode DOS systems and their emulations.
23459 @value{GDBN} supports native debugging of @sc{djgpp} programs, and
23460 defines a few commands specific to the @sc{djgpp} port. This
23461 subsection describes those commands.
23466 This is a prefix of @sc{djgpp}-specific commands which print
23467 information about the target system and important OS structures.
23470 @cindex MS-DOS system info
23471 @cindex free memory information (MS-DOS)
23472 @item info dos sysinfo
23473 This command displays assorted information about the underlying
23474 platform: the CPU type and features, the OS version and flavor, the
23475 DPMI version, and the available conventional and DPMI memory.
23480 @cindex segment descriptor tables
23481 @cindex descriptor tables display
23483 @itemx info dos ldt
23484 @itemx info dos idt
23485 These 3 commands display entries from, respectively, Global, Local,
23486 and Interrupt Descriptor Tables (GDT, LDT, and IDT). The descriptor
23487 tables are data structures which store a descriptor for each segment
23488 that is currently in use. The segment's selector is an index into a
23489 descriptor table; the table entry for that index holds the
23490 descriptor's base address and limit, and its attributes and access
23493 A typical @sc{djgpp} program uses 3 segments: a code segment, a data
23494 segment (used for both data and the stack), and a DOS segment (which
23495 allows access to DOS/BIOS data structures and absolute addresses in
23496 conventional memory). However, the DPMI host will usually define
23497 additional segments in order to support the DPMI environment.
23499 @cindex garbled pointers
23500 These commands allow to display entries from the descriptor tables.
23501 Without an argument, all entries from the specified table are
23502 displayed. An argument, which should be an integer expression, means
23503 display a single entry whose index is given by the argument. For
23504 example, here's a convenient way to display information about the
23505 debugged program's data segment:
23508 @exdent @code{(@value{GDBP}) info dos ldt $ds}
23509 @exdent @code{0x13f: base=0x11970000 limit=0x0009ffff 32-Bit Data (Read/Write, Exp-up)}
23513 This comes in handy when you want to see whether a pointer is outside
23514 the data segment's limit (i.e.@: @dfn{garbled}).
23516 @cindex page tables display (MS-DOS)
23518 @itemx info dos pte
23519 These two commands display entries from, respectively, the Page
23520 Directory and the Page Tables. Page Directories and Page Tables are
23521 data structures which control how virtual memory addresses are mapped
23522 into physical addresses. A Page Table includes an entry for every
23523 page of memory that is mapped into the program's address space; there
23524 may be several Page Tables, each one holding up to 4096 entries. A
23525 Page Directory has up to 4096 entries, one each for every Page Table
23526 that is currently in use.
23528 Without an argument, @kbd{info dos pde} displays the entire Page
23529 Directory, and @kbd{info dos pte} displays all the entries in all of
23530 the Page Tables. An argument, an integer expression, given to the
23531 @kbd{info dos pde} command means display only that entry from the Page
23532 Directory table. An argument given to the @kbd{info dos pte} command
23533 means display entries from a single Page Table, the one pointed to by
23534 the specified entry in the Page Directory.
23536 @cindex direct memory access (DMA) on MS-DOS
23537 These commands are useful when your program uses @dfn{DMA} (Direct
23538 Memory Access), which needs physical addresses to program the DMA
23541 These commands are supported only with some DPMI servers.
23543 @cindex physical address from linear address
23544 @item info dos address-pte @var{addr}
23545 This command displays the Page Table entry for a specified linear
23546 address. The argument @var{addr} is a linear address which should
23547 already have the appropriate segment's base address added to it,
23548 because this command accepts addresses which may belong to @emph{any}
23549 segment. For example, here's how to display the Page Table entry for
23550 the page where a variable @code{i} is stored:
23553 @exdent @code{(@value{GDBP}) info dos address-pte __djgpp_base_address + (char *)&i}
23554 @exdent @code{Page Table entry for address 0x11a00d30:}
23555 @exdent @code{Base=0x02698000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0xd30}
23559 This says that @code{i} is stored at offset @code{0xd30} from the page
23560 whose physical base address is @code{0x02698000}, and shows all the
23561 attributes of that page.
23563 Note that you must cast the addresses of variables to a @code{char *},
23564 since otherwise the value of @code{__djgpp_base_address}, the base
23565 address of all variables and functions in a @sc{djgpp} program, will
23566 be added using the rules of C pointer arithmetics: if @code{i} is
23567 declared an @code{int}, @value{GDBN} will add 4 times the value of
23568 @code{__djgpp_base_address} to the address of @code{i}.
23570 Here's another example, it displays the Page Table entry for the
23574 @exdent @code{(@value{GDBP}) info dos address-pte *((unsigned *)&_go32_info_block + 3)}
23575 @exdent @code{Page Table entry for address 0x29110:}
23576 @exdent @code{Base=0x00029000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0x110}
23580 (The @code{+ 3} offset is because the transfer buffer's address is the
23581 3rd member of the @code{_go32_info_block} structure.) The output
23582 clearly shows that this DPMI server maps the addresses in conventional
23583 memory 1:1, i.e.@: the physical (@code{0x00029000} + @code{0x110}) and
23584 linear (@code{0x29110}) addresses are identical.
23586 This command is supported only with some DPMI servers.
23589 @cindex DOS serial data link, remote debugging
23590 In addition to native debugging, the DJGPP port supports remote
23591 debugging via a serial data link. The following commands are specific
23592 to remote serial debugging in the DJGPP port of @value{GDBN}.
23595 @kindex set com1base
23596 @kindex set com1irq
23597 @kindex set com2base
23598 @kindex set com2irq
23599 @kindex set com3base
23600 @kindex set com3irq
23601 @kindex set com4base
23602 @kindex set com4irq
23603 @item set com1base @var{addr}
23604 This command sets the base I/O port address of the @file{COM1} serial
23607 @item set com1irq @var{irq}
23608 This command sets the @dfn{Interrupt Request} (@code{IRQ}) line to use
23609 for the @file{COM1} serial port.
23611 There are similar commands @samp{set com2base}, @samp{set com3irq},
23612 etc.@: for setting the port address and the @code{IRQ} lines for the
23615 @kindex show com1base
23616 @kindex show com1irq
23617 @kindex show com2base
23618 @kindex show com2irq
23619 @kindex show com3base
23620 @kindex show com3irq
23621 @kindex show com4base
23622 @kindex show com4irq
23623 The related commands @samp{show com1base}, @samp{show com1irq} etc.@:
23624 display the current settings of the base address and the @code{IRQ}
23625 lines used by the COM ports.
23628 @kindex info serial
23629 @cindex DOS serial port status
23630 This command prints the status of the 4 DOS serial ports. For each
23631 port, it prints whether it's active or not, its I/O base address and
23632 IRQ number, whether it uses a 16550-style FIFO, its baudrate, and the
23633 counts of various errors encountered so far.
23637 @node Cygwin Native
23638 @subsection Features for Debugging MS Windows PE Executables
23639 @cindex MS Windows debugging
23640 @cindex native Cygwin debugging
23641 @cindex Cygwin-specific commands
23643 @value{GDBN} supports native debugging of MS Windows programs, including
23644 DLLs with and without symbolic debugging information.
23646 @cindex Ctrl-BREAK, MS-Windows
23647 @cindex interrupt debuggee on MS-Windows
23648 MS-Windows programs that call @code{SetConsoleMode} to switch off the
23649 special meaning of the @samp{Ctrl-C} keystroke cannot be interrupted
23650 by typing @kbd{C-c}. For this reason, @value{GDBN} on MS-Windows
23651 supports @kbd{C-@key{BREAK}} as an alternative interrupt key
23652 sequence, which can be used to interrupt the debuggee even if it
23655 There are various additional Cygwin-specific commands, described in
23656 this section. Working with DLLs that have no debugging symbols is
23657 described in @ref{Non-debug DLL Symbols}.
23662 This is a prefix of MS Windows-specific commands which print
23663 information about the target system and important OS structures.
23665 @item info w32 selector
23666 This command displays information returned by
23667 the Win32 API @code{GetThreadSelectorEntry} function.
23668 It takes an optional argument that is evaluated to
23669 a long value to give the information about this given selector.
23670 Without argument, this command displays information
23671 about the six segment registers.
23673 @item info w32 thread-information-block
23674 This command displays thread specific information stored in the
23675 Thread Information Block (readable on the X86 CPU family using @code{$fs}
23676 selector for 32-bit programs and @code{$gs} for 64-bit programs).
23678 @kindex signal-event
23679 @item signal-event @var{id}
23680 This command signals an event with user-provided @var{id}. Used to resume
23681 crashing process when attached to it using MS-Windows JIT debugging (AeDebug).
23683 To use it, create or edit the following keys in
23684 @code{HKLM\SOFTWARE\Microsoft\Windows NT\CurrentVersion\AeDebug} and/or
23685 @code{HKLM\SOFTWARE\Wow6432Node\Microsoft\Windows NT\CurrentVersion\AeDebug}
23686 (for x86_64 versions):
23690 @code{Debugger} (REG_SZ) --- a command to launch the debugger.
23691 Suggested command is: @code{@var{fully-qualified-path-to-gdb.exe} -ex
23692 "attach %ld" -ex "signal-event %ld" -ex "continue"}.
23694 The first @code{%ld} will be replaced by the process ID of the
23695 crashing process, the second @code{%ld} will be replaced by the ID of
23696 the event that blocks the crashing process, waiting for @value{GDBN}
23700 @code{Auto} (REG_SZ) --- either @code{1} or @code{0}. @code{1} will
23701 make the system run debugger specified by the Debugger key
23702 automatically, @code{0} will cause a dialog box with ``OK'' and
23703 ``Cancel'' buttons to appear, which allows the user to either
23704 terminate the crashing process (OK) or debug it (Cancel).
23707 @kindex set cygwin-exceptions
23708 @cindex debugging the Cygwin DLL
23709 @cindex Cygwin DLL, debugging
23710 @item set cygwin-exceptions @var{mode}
23711 If @var{mode} is @code{on}, @value{GDBN} will break on exceptions that
23712 happen inside the Cygwin DLL. If @var{mode} is @code{off},
23713 @value{GDBN} will delay recognition of exceptions, and may ignore some
23714 exceptions which seem to be caused by internal Cygwin DLL
23715 ``bookkeeping''. This option is meant primarily for debugging the
23716 Cygwin DLL itself; the default value is @code{off} to avoid annoying
23717 @value{GDBN} users with false @code{SIGSEGV} signals.
23719 @kindex show cygwin-exceptions
23720 @item show cygwin-exceptions
23721 Displays whether @value{GDBN} will break on exceptions that happen
23722 inside the Cygwin DLL itself.
23724 @kindex set new-console
23725 @item set new-console @var{mode}
23726 If @var{mode} is @code{on} the debuggee will
23727 be started in a new console on next start.
23728 If @var{mode} is @code{off}, the debuggee will
23729 be started in the same console as the debugger.
23731 @kindex show new-console
23732 @item show new-console
23733 Displays whether a new console is used
23734 when the debuggee is started.
23736 @kindex set new-group
23737 @item set new-group @var{mode}
23738 This boolean value controls whether the debuggee should
23739 start a new group or stay in the same group as the debugger.
23740 This affects the way the Windows OS handles
23743 @kindex show new-group
23744 @item show new-group
23745 Displays current value of new-group boolean.
23747 @kindex set debugevents
23748 @item set debugevents
23749 This boolean value adds debug output concerning kernel events related
23750 to the debuggee seen by the debugger. This includes events that
23751 signal thread and process creation and exit, DLL loading and
23752 unloading, console interrupts, and debugging messages produced by the
23753 Windows @code{OutputDebugString} API call.
23755 @kindex set debugexec
23756 @item set debugexec
23757 This boolean value adds debug output concerning execute events
23758 (such as resume thread) seen by the debugger.
23760 @kindex set debugexceptions
23761 @item set debugexceptions
23762 This boolean value adds debug output concerning exceptions in the
23763 debuggee seen by the debugger.
23765 @kindex set debugmemory
23766 @item set debugmemory
23767 This boolean value adds debug output concerning debuggee memory reads
23768 and writes by the debugger.
23772 This boolean values specifies whether the debuggee is called
23773 via a shell or directly (default value is on).
23777 Displays if the debuggee will be started with a shell.
23782 * Non-debug DLL Symbols:: Support for DLLs without debugging symbols
23785 @node Non-debug DLL Symbols
23786 @subsubsection Support for DLLs without Debugging Symbols
23787 @cindex DLLs with no debugging symbols
23788 @cindex Minimal symbols and DLLs
23790 Very often on windows, some of the DLLs that your program relies on do
23791 not include symbolic debugging information (for example,
23792 @file{kernel32.dll}). When @value{GDBN} doesn't recognize any debugging
23793 symbols in a DLL, it relies on the minimal amount of symbolic
23794 information contained in the DLL's export table. This section
23795 describes working with such symbols, known internally to @value{GDBN} as
23796 ``minimal symbols''.
23798 Note that before the debugged program has started execution, no DLLs
23799 will have been loaded. The easiest way around this problem is simply to
23800 start the program --- either by setting a breakpoint or letting the
23801 program run once to completion.
23803 @subsubsection DLL Name Prefixes
23805 In keeping with the naming conventions used by the Microsoft debugging
23806 tools, DLL export symbols are made available with a prefix based on the
23807 DLL name, for instance @code{KERNEL32!CreateFileA}. The plain name is
23808 also entered into the symbol table, so @code{CreateFileA} is often
23809 sufficient. In some cases there will be name clashes within a program
23810 (particularly if the executable itself includes full debugging symbols)
23811 necessitating the use of the fully qualified name when referring to the
23812 contents of the DLL. Use single-quotes around the name to avoid the
23813 exclamation mark (``!'') being interpreted as a language operator.
23815 Note that the internal name of the DLL may be all upper-case, even
23816 though the file name of the DLL is lower-case, or vice-versa. Since
23817 symbols within @value{GDBN} are @emph{case-sensitive} this may cause
23818 some confusion. If in doubt, try the @code{info functions} and
23819 @code{info variables} commands or even @code{maint print msymbols}
23820 (@pxref{Symbols}). Here's an example:
23823 (@value{GDBP}) info function CreateFileA
23824 All functions matching regular expression "CreateFileA":
23826 Non-debugging symbols:
23827 0x77e885f4 CreateFileA
23828 0x77e885f4 KERNEL32!CreateFileA
23832 (@value{GDBP}) info function !
23833 All functions matching regular expression "!":
23835 Non-debugging symbols:
23836 0x6100114c cygwin1!__assert
23837 0x61004034 cygwin1!_dll_crt0@@0
23838 0x61004240 cygwin1!dll_crt0(per_process *)
23842 @subsubsection Working with Minimal Symbols
23844 Symbols extracted from a DLL's export table do not contain very much
23845 type information. All that @value{GDBN} can do is guess whether a symbol
23846 refers to a function or variable depending on the linker section that
23847 contains the symbol. Also note that the actual contents of the memory
23848 contained in a DLL are not available unless the program is running. This
23849 means that you cannot examine the contents of a variable or disassemble
23850 a function within a DLL without a running program.
23852 Variables are generally treated as pointers and dereferenced
23853 automatically. For this reason, it is often necessary to prefix a
23854 variable name with the address-of operator (``&'') and provide explicit
23855 type information in the command. Here's an example of the type of
23859 (@value{GDBP}) print 'cygwin1!__argv'
23860 'cygwin1!__argv' has unknown type; cast it to its declared type
23864 (@value{GDBP}) x 'cygwin1!__argv'
23865 'cygwin1!__argv' has unknown type; cast it to its declared type
23868 And two possible solutions:
23871 (@value{GDBP}) print ((char **)'cygwin1!__argv')[0]
23872 $2 = 0x22fd98 "/cygdrive/c/mydirectory/myprogram"
23876 (@value{GDBP}) x/2x &'cygwin1!__argv'
23877 0x610c0aa8 <cygwin1!__argv>: 0x10021608 0x00000000
23878 (@value{GDBP}) x/x 0x10021608
23879 0x10021608: 0x0022fd98
23880 (@value{GDBP}) x/s 0x0022fd98
23881 0x22fd98: "/cygdrive/c/mydirectory/myprogram"
23884 Setting a break point within a DLL is possible even before the program
23885 starts execution. However, under these circumstances, @value{GDBN} can't
23886 examine the initial instructions of the function in order to skip the
23887 function's frame set-up code. You can work around this by using ``*&''
23888 to set the breakpoint at a raw memory address:
23891 (@value{GDBP}) break *&'python22!PyOS_Readline'
23892 Breakpoint 1 at 0x1e04eff0
23895 The author of these extensions is not entirely convinced that setting a
23896 break point within a shared DLL like @file{kernel32.dll} is completely
23900 @subsection Commands Specific to @sc{gnu} Hurd Systems
23901 @cindex @sc{gnu} Hurd debugging
23903 This subsection describes @value{GDBN} commands specific to the
23904 @sc{gnu} Hurd native debugging.
23909 @kindex set signals@r{, Hurd command}
23910 @kindex set sigs@r{, Hurd command}
23911 This command toggles the state of inferior signal interception by
23912 @value{GDBN}. Mach exceptions, such as breakpoint traps, are not
23913 affected by this command. @code{sigs} is a shorthand alias for
23918 @kindex show signals@r{, Hurd command}
23919 @kindex show sigs@r{, Hurd command}
23920 Show the current state of intercepting inferior's signals.
23922 @item set signal-thread
23923 @itemx set sigthread
23924 @kindex set signal-thread
23925 @kindex set sigthread
23926 This command tells @value{GDBN} which thread is the @code{libc} signal
23927 thread. That thread is run when a signal is delivered to a running
23928 process. @code{set sigthread} is the shorthand alias of @code{set
23931 @item show signal-thread
23932 @itemx show sigthread
23933 @kindex show signal-thread
23934 @kindex show sigthread
23935 These two commands show which thread will run when the inferior is
23936 delivered a signal.
23939 @kindex set stopped@r{, Hurd command}
23940 This commands tells @value{GDBN} that the inferior process is stopped,
23941 as with the @code{SIGSTOP} signal. The stopped process can be
23942 continued by delivering a signal to it.
23945 @kindex show stopped@r{, Hurd command}
23946 This command shows whether @value{GDBN} thinks the debuggee is
23949 @item set exceptions
23950 @kindex set exceptions@r{, Hurd command}
23951 Use this command to turn off trapping of exceptions in the inferior.
23952 When exception trapping is off, neither breakpoints nor
23953 single-stepping will work. To restore the default, set exception
23956 @item show exceptions
23957 @kindex show exceptions@r{, Hurd command}
23958 Show the current state of trapping exceptions in the inferior.
23960 @item set task pause
23961 @kindex set task@r{, Hurd commands}
23962 @cindex task attributes (@sc{gnu} Hurd)
23963 @cindex pause current task (@sc{gnu} Hurd)
23964 This command toggles task suspension when @value{GDBN} has control.
23965 Setting it to on takes effect immediately, and the task is suspended
23966 whenever @value{GDBN} gets control. Setting it to off will take
23967 effect the next time the inferior is continued. If this option is set
23968 to off, you can use @code{set thread default pause on} or @code{set
23969 thread pause on} (see below) to pause individual threads.
23971 @item show task pause
23972 @kindex show task@r{, Hurd commands}
23973 Show the current state of task suspension.
23975 @item set task detach-suspend-count
23976 @cindex task suspend count
23977 @cindex detach from task, @sc{gnu} Hurd
23978 This command sets the suspend count the task will be left with when
23979 @value{GDBN} detaches from it.
23981 @item show task detach-suspend-count
23982 Show the suspend count the task will be left with when detaching.
23984 @item set task exception-port
23985 @itemx set task excp
23986 @cindex task exception port, @sc{gnu} Hurd
23987 This command sets the task exception port to which @value{GDBN} will
23988 forward exceptions. The argument should be the value of the @dfn{send
23989 rights} of the task. @code{set task excp} is a shorthand alias.
23991 @item set noninvasive
23992 @cindex noninvasive task options
23993 This command switches @value{GDBN} to a mode that is the least
23994 invasive as far as interfering with the inferior is concerned. This
23995 is the same as using @code{set task pause}, @code{set exceptions}, and
23996 @code{set signals} to values opposite to the defaults.
23998 @item info send-rights
23999 @itemx info receive-rights
24000 @itemx info port-rights
24001 @itemx info port-sets
24002 @itemx info dead-names
24005 @cindex send rights, @sc{gnu} Hurd
24006 @cindex receive rights, @sc{gnu} Hurd
24007 @cindex port rights, @sc{gnu} Hurd
24008 @cindex port sets, @sc{gnu} Hurd
24009 @cindex dead names, @sc{gnu} Hurd
24010 These commands display information about, respectively, send rights,
24011 receive rights, port rights, port sets, and dead names of a task.
24012 There are also shorthand aliases: @code{info ports} for @code{info
24013 port-rights} and @code{info psets} for @code{info port-sets}.
24015 @item set thread pause
24016 @kindex set thread@r{, Hurd command}
24017 @cindex thread properties, @sc{gnu} Hurd
24018 @cindex pause current thread (@sc{gnu} Hurd)
24019 This command toggles current thread suspension when @value{GDBN} has
24020 control. Setting it to on takes effect immediately, and the current
24021 thread is suspended whenever @value{GDBN} gets control. Setting it to
24022 off will take effect the next time the inferior is continued.
24023 Normally, this command has no effect, since when @value{GDBN} has
24024 control, the whole task is suspended. However, if you used @code{set
24025 task pause off} (see above), this command comes in handy to suspend
24026 only the current thread.
24028 @item show thread pause
24029 @kindex show thread@r{, Hurd command}
24030 This command shows the state of current thread suspension.
24032 @item set thread run
24033 This command sets whether the current thread is allowed to run.
24035 @item show thread run
24036 Show whether the current thread is allowed to run.
24038 @item set thread detach-suspend-count
24039 @cindex thread suspend count, @sc{gnu} Hurd
24040 @cindex detach from thread, @sc{gnu} Hurd
24041 This command sets the suspend count @value{GDBN} will leave on a
24042 thread when detaching. This number is relative to the suspend count
24043 found by @value{GDBN} when it notices the thread; use @code{set thread
24044 takeover-suspend-count} to force it to an absolute value.
24046 @item show thread detach-suspend-count
24047 Show the suspend count @value{GDBN} will leave on the thread when
24050 @item set thread exception-port
24051 @itemx set thread excp
24052 Set the thread exception port to which to forward exceptions. This
24053 overrides the port set by @code{set task exception-port} (see above).
24054 @code{set thread excp} is the shorthand alias.
24056 @item set thread takeover-suspend-count
24057 Normally, @value{GDBN}'s thread suspend counts are relative to the
24058 value @value{GDBN} finds when it notices each thread. This command
24059 changes the suspend counts to be absolute instead.
24061 @item set thread default
24062 @itemx show thread default
24063 @cindex thread default settings, @sc{gnu} Hurd
24064 Each of the above @code{set thread} commands has a @code{set thread
24065 default} counterpart (e.g., @code{set thread default pause}, @code{set
24066 thread default exception-port}, etc.). The @code{thread default}
24067 variety of commands sets the default thread properties for all
24068 threads; you can then change the properties of individual threads with
24069 the non-default commands.
24076 @value{GDBN} provides the following commands specific to the Darwin target:
24079 @item set debug darwin @var{num}
24080 @kindex set debug darwin
24081 When set to a non zero value, enables debugging messages specific to
24082 the Darwin support. Higher values produce more verbose output.
24084 @item show debug darwin
24085 @kindex show debug darwin
24086 Show the current state of Darwin messages.
24088 @item set debug mach-o @var{num}
24089 @kindex set debug mach-o
24090 When set to a non zero value, enables debugging messages while
24091 @value{GDBN} is reading Darwin object files. (@dfn{Mach-O} is the
24092 file format used on Darwin for object and executable files.) Higher
24093 values produce more verbose output. This is a command to diagnose
24094 problems internal to @value{GDBN} and should not be needed in normal
24097 @item show debug mach-o
24098 @kindex show debug mach-o
24099 Show the current state of Mach-O file messages.
24101 @item set mach-exceptions on
24102 @itemx set mach-exceptions off
24103 @kindex set mach-exceptions
24104 On Darwin, faults are first reported as a Mach exception and are then
24105 mapped to a Posix signal. Use this command to turn on trapping of
24106 Mach exceptions in the inferior. This might be sometimes useful to
24107 better understand the cause of a fault. The default is off.
24109 @item show mach-exceptions
24110 @kindex show mach-exceptions
24111 Show the current state of exceptions trapping.
24115 @subsection FreeBSD
24118 When the ABI of a system call is changed in the FreeBSD kernel, this
24119 is implemented by leaving a compatibility system call using the old
24120 ABI at the existing number and allocating a new system call number for
24121 the version using the new ABI. As a convenience, when a system call
24122 is caught by name (@pxref{catch syscall}), compatibility system calls
24125 For example, FreeBSD 12 introduced a new variant of the @code{kevent}
24126 system call and catching the @code{kevent} system call by name catches
24130 (@value{GDBP}) catch syscall kevent
24131 Catchpoint 1 (syscalls 'freebsd11_kevent' [363] 'kevent' [560])
24137 @section Embedded Operating Systems
24139 This section describes configurations involving the debugging of
24140 embedded operating systems that are available for several different
24143 @value{GDBN} includes the ability to debug programs running on
24144 various real-time operating systems.
24146 @node Embedded Processors
24147 @section Embedded Processors
24149 This section goes into details specific to particular embedded
24152 @cindex send command to simulator
24153 Whenever a specific embedded processor has a simulator, @value{GDBN}
24154 allows to send an arbitrary command to the simulator.
24157 @item sim @var{command}
24158 @kindex sim@r{, a command}
24159 Send an arbitrary @var{command} string to the simulator. Consult the
24160 documentation for the specific simulator in use for information about
24161 acceptable commands.
24166 * ARC:: Synopsys ARC
24168 * M68K:: Motorola M68K
24169 * MicroBlaze:: Xilinx MicroBlaze
24170 * MIPS Embedded:: MIPS Embedded
24171 * OpenRISC 1000:: OpenRISC 1000 (or1k)
24172 * PowerPC Embedded:: PowerPC Embedded
24175 * Super-H:: Renesas Super-H
24179 @subsection Synopsys ARC
24180 @cindex Synopsys ARC
24181 @cindex ARC specific commands
24187 @value{GDBN} provides the following ARC-specific commands:
24190 @item set debug arc
24191 @kindex set debug arc
24192 Control the level of ARC specific debug messages. Use 0 for no messages (the
24193 default), 1 for debug messages, and 2 for even more debug messages.
24195 @item show debug arc
24196 @kindex show debug arc
24197 Show the level of ARC specific debugging in operation.
24199 @item maint print arc arc-instruction @var{address}
24200 @kindex maint print arc arc-instruction
24201 Print internal disassembler information about instruction at a given address.
24208 @value{GDBN} provides the following ARM-specific commands:
24211 @item set arm disassembler
24213 This commands selects from a list of disassembly styles. The
24214 @code{"std"} style is the standard style.
24216 @item show arm disassembler
24218 Show the current disassembly style.
24220 @item set arm apcs32
24221 @cindex ARM 32-bit mode
24222 This command toggles ARM operation mode between 32-bit and 26-bit.
24224 @item show arm apcs32
24225 Display the current usage of the ARM 32-bit mode.
24227 @item set arm fpu @var{fputype}
24228 This command sets the ARM floating-point unit (FPU) type. The
24229 argument @var{fputype} can be one of these:
24233 Determine the FPU type by querying the OS ABI.
24235 Software FPU, with mixed-endian doubles on little-endian ARM
24238 GCC-compiled FPA co-processor.
24240 Software FPU with pure-endian doubles.
24246 Show the current type of the FPU.
24249 This command forces @value{GDBN} to use the specified ABI.
24252 Show the currently used ABI.
24254 @item set arm fallback-mode (arm|thumb|auto)
24255 @value{GDBN} uses the symbol table, when available, to determine
24256 whether instructions are ARM or Thumb. This command controls
24257 @value{GDBN}'s default behavior when the symbol table is not
24258 available. The default is @samp{auto}, which causes @value{GDBN} to
24259 use the current execution mode (from the @code{T} bit in the @code{CPSR}
24262 @item show arm fallback-mode
24263 Show the current fallback instruction mode.
24265 @item set arm force-mode (arm|thumb|auto)
24266 This command overrides use of the symbol table to determine whether
24267 instructions are ARM or Thumb. The default is @samp{auto}, which
24268 causes @value{GDBN} to use the symbol table and then the setting
24269 of @samp{set arm fallback-mode}.
24271 @item show arm force-mode
24272 Show the current forced instruction mode.
24274 @item set debug arm
24275 Toggle whether to display ARM-specific debugging messages from the ARM
24276 target support subsystem.
24278 @item show debug arm
24279 Show whether ARM-specific debugging messages are enabled.
24283 @item target sim @r{[}@var{simargs}@r{]} @dots{}
24284 The @value{GDBN} ARM simulator accepts the following optional arguments.
24287 @item --swi-support=@var{type}
24288 Tell the simulator which SWI interfaces to support. The argument
24289 @var{type} may be a comma separated list of the following values.
24290 The default value is @code{all}.
24305 The Motorola m68k configuration includes ColdFire support.
24308 @subsection MicroBlaze
24309 @cindex Xilinx MicroBlaze
24310 @cindex XMD, Xilinx Microprocessor Debugger
24312 The MicroBlaze is a soft-core processor supported on various Xilinx
24313 FPGAs, such as Spartan or Virtex series. Boards with these processors
24314 usually have JTAG ports which connect to a host system running the Xilinx
24315 Embedded Development Kit (EDK) or Software Development Kit (SDK).
24316 This host system is used to download the configuration bitstream to
24317 the target FPGA. The Xilinx Microprocessor Debugger (XMD) program
24318 communicates with the target board using the JTAG interface and
24319 presents a @code{gdbserver} interface to the board. By default
24320 @code{xmd} uses port @code{1234}. (While it is possible to change
24321 this default port, it requires the use of undocumented @code{xmd}
24322 commands. Contact Xilinx support if you need to do this.)
24324 Use these GDB commands to connect to the MicroBlaze target processor.
24327 @item target remote :1234
24328 Use this command to connect to the target if you are running @value{GDBN}
24329 on the same system as @code{xmd}.
24331 @item target remote @var{xmd-host}:1234
24332 Use this command to connect to the target if it is connected to @code{xmd}
24333 running on a different system named @var{xmd-host}.
24336 Use this command to download a program to the MicroBlaze target.
24338 @item set debug microblaze @var{n}
24339 Enable MicroBlaze-specific debugging messages if non-zero.
24341 @item show debug microblaze @var{n}
24342 Show MicroBlaze-specific debugging level.
24345 @node MIPS Embedded
24346 @subsection @acronym{MIPS} Embedded
24349 @value{GDBN} supports these special commands for @acronym{MIPS} targets:
24352 @item set mipsfpu double
24353 @itemx set mipsfpu single
24354 @itemx set mipsfpu none
24355 @itemx set mipsfpu auto
24356 @itemx show mipsfpu
24357 @kindex set mipsfpu
24358 @kindex show mipsfpu
24359 @cindex @acronym{MIPS} remote floating point
24360 @cindex floating point, @acronym{MIPS} remote
24361 If your target board does not support the @acronym{MIPS} floating point
24362 coprocessor, you should use the command @samp{set mipsfpu none} (if you
24363 need this, you may wish to put the command in your @value{GDBN} init
24364 file). This tells @value{GDBN} how to find the return value of
24365 functions which return floating point values. It also allows
24366 @value{GDBN} to avoid saving the floating point registers when calling
24367 functions on the board. If you are using a floating point coprocessor
24368 with only single precision floating point support, as on the @sc{r4650}
24369 processor, use the command @samp{set mipsfpu single}. The default
24370 double precision floating point coprocessor may be selected using
24371 @samp{set mipsfpu double}.
24373 In previous versions the only choices were double precision or no
24374 floating point, so @samp{set mipsfpu on} will select double precision
24375 and @samp{set mipsfpu off} will select no floating point.
24377 As usual, you can inquire about the @code{mipsfpu} variable with
24378 @samp{show mipsfpu}.
24381 @node OpenRISC 1000
24382 @subsection OpenRISC 1000
24383 @cindex OpenRISC 1000
24386 The OpenRISC 1000 provides a free RISC instruction set architecture. It is
24387 mainly provided as a soft-core which can run on Xilinx, Altera and other
24390 @value{GDBN} for OpenRISC supports the below commands when connecting to
24398 Runs the builtin CPU simulator which can run very basic
24399 programs but does not support most hardware functions like MMU.
24400 For more complex use cases the user is advised to run an external
24401 target, and connect using @samp{target remote}.
24403 Example: @code{target sim}
24405 @item set debug or1k
24406 Toggle whether to display OpenRISC-specific debugging messages from the
24407 OpenRISC target support subsystem.
24409 @item show debug or1k
24410 Show whether OpenRISC-specific debugging messages are enabled.
24413 @node PowerPC Embedded
24414 @subsection PowerPC Embedded
24416 @cindex DVC register
24417 @value{GDBN} supports using the DVC (Data Value Compare) register to
24418 implement in hardware simple hardware watchpoint conditions of the form:
24421 (@value{GDBP}) watch @var{ADDRESS|VARIABLE} \
24422 if @var{ADDRESS|VARIABLE} == @var{CONSTANT EXPRESSION}
24425 The DVC register will be automatically used when @value{GDBN} detects
24426 such pattern in a condition expression, and the created watchpoint uses one
24427 debug register (either the @code{exact-watchpoints} option is on and the
24428 variable is scalar, or the variable has a length of one byte). This feature
24429 is available in native @value{GDBN} running on a Linux kernel version 2.6.34
24432 When running on PowerPC embedded processors, @value{GDBN} automatically uses
24433 ranged hardware watchpoints, unless the @code{exact-watchpoints} option is on,
24434 in which case watchpoints using only one debug register are created when
24435 watching variables of scalar types.
24437 You can create an artificial array to watch an arbitrary memory
24438 region using one of the following commands (@pxref{Expressions}):
24441 (@value{GDBP}) watch *((char *) @var{address})@@@var{length}
24442 (@value{GDBP}) watch @{char[@var{length}]@} @var{address}
24445 PowerPC embedded processors support masked watchpoints. See the discussion
24446 about the @code{mask} argument in @ref{Set Watchpoints}.
24448 @cindex ranged breakpoint
24449 PowerPC embedded processors support hardware accelerated
24450 @dfn{ranged breakpoints}. A ranged breakpoint stops execution of
24451 the inferior whenever it executes an instruction at any address within
24452 the range it specifies. To set a ranged breakpoint in @value{GDBN},
24453 use the @code{break-range} command.
24455 @value{GDBN} provides the following PowerPC-specific commands:
24458 @kindex break-range
24459 @item break-range @var{start-location}, @var{end-location}
24460 Set a breakpoint for an address range given by
24461 @var{start-location} and @var{end-location}, which can specify a function name,
24462 a line number, an offset of lines from the current line or from the start
24463 location, or an address of an instruction (see @ref{Specify Location},
24464 for a list of all the possible ways to specify a @var{location}.)
24465 The breakpoint will stop execution of the inferior whenever it
24466 executes an instruction at any address within the specified range,
24467 (including @var{start-location} and @var{end-location}.)
24469 @kindex set powerpc
24470 @item set powerpc soft-float
24471 @itemx show powerpc soft-float
24472 Force @value{GDBN} to use (or not use) a software floating point calling
24473 convention. By default, @value{GDBN} selects the calling convention based
24474 on the selected architecture and the provided executable file.
24476 @item set powerpc vector-abi
24477 @itemx show powerpc vector-abi
24478 Force @value{GDBN} to use the specified calling convention for vector
24479 arguments and return values. The valid options are @samp{auto};
24480 @samp{generic}, to avoid vector registers even if they are present;
24481 @samp{altivec}, to use AltiVec registers; and @samp{spe} to use SPE
24482 registers. By default, @value{GDBN} selects the calling convention
24483 based on the selected architecture and the provided executable file.
24485 @item set powerpc exact-watchpoints
24486 @itemx show powerpc exact-watchpoints
24487 Allow @value{GDBN} to use only one debug register when watching a variable
24488 of scalar type, thus assuming that the variable is accessed through the
24489 address of its first byte.
24494 @subsection Atmel AVR
24497 When configured for debugging the Atmel AVR, @value{GDBN} supports the
24498 following AVR-specific commands:
24501 @item info io_registers
24502 @kindex info io_registers@r{, AVR}
24503 @cindex I/O registers (Atmel AVR)
24504 This command displays information about the AVR I/O registers. For
24505 each register, @value{GDBN} prints its number and value.
24512 When configured for debugging CRIS, @value{GDBN} provides the
24513 following CRIS-specific commands:
24516 @item set cris-version @var{ver}
24517 @cindex CRIS version
24518 Set the current CRIS version to @var{ver}, either @samp{10} or @samp{32}.
24519 The CRIS version affects register names and sizes. This command is useful in
24520 case autodetection of the CRIS version fails.
24522 @item show cris-version
24523 Show the current CRIS version.
24525 @item set cris-dwarf2-cfi
24526 @cindex DWARF-2 CFI and CRIS
24527 Set the usage of DWARF-2 CFI for CRIS debugging. The default is @samp{on}.
24528 Change to @samp{off} when using @code{gcc-cris} whose version is below
24531 @item show cris-dwarf2-cfi
24532 Show the current state of using DWARF-2 CFI.
24534 @item set cris-mode @var{mode}
24536 Set the current CRIS mode to @var{mode}. It should only be changed when
24537 debugging in guru mode, in which case it should be set to
24538 @samp{guru} (the default is @samp{normal}).
24540 @item show cris-mode
24541 Show the current CRIS mode.
24545 @subsection Renesas Super-H
24548 For the Renesas Super-H processor, @value{GDBN} provides these
24552 @item set sh calling-convention @var{convention}
24553 @kindex set sh calling-convention
24554 Set the calling-convention used when calling functions from @value{GDBN}.
24555 Allowed values are @samp{gcc}, which is the default setting, and @samp{renesas}.
24556 With the @samp{gcc} setting, functions are called using the @value{NGCC} calling
24557 convention. If the DWARF-2 information of the called function specifies
24558 that the function follows the Renesas calling convention, the function
24559 is called using the Renesas calling convention. If the calling convention
24560 is set to @samp{renesas}, the Renesas calling convention is always used,
24561 regardless of the DWARF-2 information. This can be used to override the
24562 default of @samp{gcc} if debug information is missing, or the compiler
24563 does not emit the DWARF-2 calling convention entry for a function.
24565 @item show sh calling-convention
24566 @kindex show sh calling-convention
24567 Show the current calling convention setting.
24572 @node Architectures
24573 @section Architectures
24575 This section describes characteristics of architectures that affect
24576 all uses of @value{GDBN} with the architecture, both native and cross.
24583 * HPPA:: HP PA architecture
24591 @subsection AArch64
24592 @cindex AArch64 support
24594 When @value{GDBN} is debugging the AArch64 architecture, it provides the
24595 following special commands:
24598 @item set debug aarch64
24599 @kindex set debug aarch64
24600 This command determines whether AArch64 architecture-specific debugging
24601 messages are to be displayed.
24603 @item show debug aarch64
24604 Show whether AArch64 debugging messages are displayed.
24608 @subsubsection AArch64 SVE.
24609 @cindex AArch64 SVE.
24611 When @value{GDBN} is debugging the AArch64 architecture, if the Scalable Vector
24612 Extension (SVE) is present, then @value{GDBN} will provide the vector registers
24613 @code{$z0} through @code{$z31}, vector predicate registers @code{$p0} through
24614 @code{$p15}, and the @code{$ffr} register. In addition, the pseudo register
24615 @code{$vg} will be provided. This is the vector granule for the current thread
24616 and represents the number of 64-bit chunks in an SVE @code{z} register.
24618 If the vector length changes, then the @code{$vg} register will be updated,
24619 but the lengths of the @code{z} and @code{p} registers will not change. This
24620 is a known limitation of @value{GDBN} and does not affect the execution of the
24623 @subsubsection AArch64 Pointer Authentication.
24624 @cindex AArch64 Pointer Authentication.
24626 When @value{GDBN} is debugging the AArch64 architecture, and the program is
24627 using the v8.3-A feature Pointer Authentication (PAC), then whenever the link
24628 register @code{$lr} is pointing to an PAC function its value will be masked.
24629 When GDB prints a backtrace, any addresses that required unmasking will be
24630 postfixed with the marker [PAC]. When using the MI, this is printed as part
24631 of the @code{addr_flags} field.
24634 @subsection x86 Architecture-specific Issues
24637 @item set struct-convention @var{mode}
24638 @kindex set struct-convention
24639 @cindex struct return convention
24640 @cindex struct/union returned in registers
24641 Set the convention used by the inferior to return @code{struct}s and
24642 @code{union}s from functions to @var{mode}. Possible values of
24643 @var{mode} are @code{"pcc"}, @code{"reg"}, and @code{"default"} (the
24644 default). @code{"default"} or @code{"pcc"} means that @code{struct}s
24645 are returned on the stack, while @code{"reg"} means that a
24646 @code{struct} or a @code{union} whose size is 1, 2, 4, or 8 bytes will
24647 be returned in a register.
24649 @item show struct-convention
24650 @kindex show struct-convention
24651 Show the current setting of the convention to return @code{struct}s
24656 @subsubsection Intel @dfn{Memory Protection Extensions} (MPX).
24657 @cindex Intel Memory Protection Extensions (MPX).
24659 Memory Protection Extension (MPX) adds the bound registers @samp{BND0}
24660 @footnote{The register named with capital letters represent the architecture
24661 registers.} through @samp{BND3}. Bound registers store a pair of 64-bit values
24662 which are the lower bound and upper bound. Bounds are effective addresses or
24663 memory locations. The upper bounds are architecturally represented in 1's
24664 complement form. A bound having lower bound = 0, and upper bound = 0
24665 (1's complement of all bits set) will allow access to the entire address space.
24667 @samp{BND0} through @samp{BND3} are represented in @value{GDBN} as @samp{bnd0raw}
24668 through @samp{bnd3raw}. Pseudo registers @samp{bnd0} through @samp{bnd3}
24669 display the upper bound performing the complement of one operation on the
24670 upper bound value, i.e.@ when upper bound in @samp{bnd0raw} is 0 in the
24671 @value{GDBN} @samp{bnd0} it will be @code{0xfff@dots{}}. In this sense it
24672 can also be noted that the upper bounds are inclusive.
24674 As an example, assume that the register BND0 holds bounds for a pointer having
24675 access allowed for the range between 0x32 and 0x71. The values present on
24676 bnd0raw and bnd registers are presented as follows:
24679 bnd0raw = @{0x32, 0xffffffff8e@}
24680 bnd0 = @{lbound = 0x32, ubound = 0x71@} : size 64
24683 This way the raw value can be accessed via bnd0raw@dots{}bnd3raw. Any
24684 change on bnd0@dots{}bnd3 or bnd0raw@dots{}bnd3raw is reflect on its
24685 counterpart. When the bnd0@dots{}bnd3 registers are displayed via
24686 Python, the display includes the memory size, in bits, accessible to
24689 Bounds can also be stored in bounds tables, which are stored in
24690 application memory. These tables store bounds for pointers by specifying
24691 the bounds pointer's value along with its bounds. Evaluating and changing
24692 bounds located in bound tables is therefore interesting while investigating
24693 bugs on MPX context. @value{GDBN} provides commands for this purpose:
24696 @item show mpx bound @var{pointer}
24697 @kindex show mpx bound
24698 Display bounds of the given @var{pointer}.
24700 @item set mpx bound @var{pointer}, @var{lbound}, @var{ubound}
24701 @kindex set mpx bound
24702 Set the bounds of a pointer in the bound table.
24703 This command takes three parameters: @var{pointer} is the pointers
24704 whose bounds are to be changed, @var{lbound} and @var{ubound} are new values
24705 for lower and upper bounds respectively.
24708 When you call an inferior function on an Intel MPX enabled program,
24709 GDB sets the inferior's bound registers to the init (disabled) state
24710 before calling the function. As a consequence, bounds checks for the
24711 pointer arguments passed to the function will always pass.
24713 This is necessary because when you call an inferior function, the
24714 program is usually in the middle of the execution of other function.
24715 Since at that point bound registers are in an arbitrary state, not
24716 clearing them would lead to random bound violations in the called
24719 You can still examine the influence of the bound registers on the
24720 execution of the called function by stopping the execution of the
24721 called function at its prologue, setting bound registers, and
24722 continuing the execution. For example:
24726 Breakpoint 2 at 0x4009de: file i386-mpx-call.c, line 47.
24727 $ print upper (a, b, c, d, 1)
24728 Breakpoint 2, upper (a=0x0, b=0x6e0000005b, c=0x0, d=0x0, len=48)....
24730 @{lbound = 0x0, ubound = ffffffff@} : size -1
24733 At this last step the value of bnd0 can be changed for investigation of bound
24734 violations caused along the execution of the call. In order to know how to
24735 set the bound registers or bound table for the call consult the ABI.
24740 See the following section.
24743 @subsection @acronym{MIPS}
24745 @cindex stack on Alpha
24746 @cindex stack on @acronym{MIPS}
24747 @cindex Alpha stack
24748 @cindex @acronym{MIPS} stack
24749 Alpha- and @acronym{MIPS}-based computers use an unusual stack frame, which
24750 sometimes requires @value{GDBN} to search backward in the object code to
24751 find the beginning of a function.
24753 @cindex response time, @acronym{MIPS} debugging
24754 To improve response time (especially for embedded applications, where
24755 @value{GDBN} may be restricted to a slow serial line for this search)
24756 you may want to limit the size of this search, using one of these
24760 @cindex @code{heuristic-fence-post} (Alpha, @acronym{MIPS})
24761 @item set heuristic-fence-post @var{limit}
24762 Restrict @value{GDBN} to examining at most @var{limit} bytes in its
24763 search for the beginning of a function. A value of @var{0} (the
24764 default) means there is no limit. However, except for @var{0}, the
24765 larger the limit the more bytes @code{heuristic-fence-post} must search
24766 and therefore the longer it takes to run. You should only need to use
24767 this command when debugging a stripped executable.
24769 @item show heuristic-fence-post
24770 Display the current limit.
24774 These commands are available @emph{only} when @value{GDBN} is configured
24775 for debugging programs on Alpha or @acronym{MIPS} processors.
24777 Several @acronym{MIPS}-specific commands are available when debugging @acronym{MIPS}
24781 @item set mips abi @var{arg}
24782 @kindex set mips abi
24783 @cindex set ABI for @acronym{MIPS}
24784 Tell @value{GDBN} which @acronym{MIPS} ABI is used by the inferior. Possible
24785 values of @var{arg} are:
24789 The default ABI associated with the current binary (this is the
24799 @item show mips abi
24800 @kindex show mips abi
24801 Show the @acronym{MIPS} ABI used by @value{GDBN} to debug the inferior.
24803 @item set mips compression @var{arg}
24804 @kindex set mips compression
24805 @cindex code compression, @acronym{MIPS}
24806 Tell @value{GDBN} which @acronym{MIPS} compressed
24807 @acronym{ISA, Instruction Set Architecture} encoding is used by the
24808 inferior. @value{GDBN} uses this for code disassembly and other
24809 internal interpretation purposes. This setting is only referred to
24810 when no executable has been associated with the debugging session or
24811 the executable does not provide information about the encoding it uses.
24812 Otherwise this setting is automatically updated from information
24813 provided by the executable.
24815 Possible values of @var{arg} are @samp{mips16} and @samp{micromips}.
24816 The default compressed @acronym{ISA} encoding is @samp{mips16}, as
24817 executables containing @acronym{MIPS16} code frequently are not
24818 identified as such.
24820 This setting is ``sticky''; that is, it retains its value across
24821 debugging sessions until reset either explicitly with this command or
24822 implicitly from an executable.
24824 The compiler and/or assembler typically add symbol table annotations to
24825 identify functions compiled for the @acronym{MIPS16} or
24826 @acronym{microMIPS} @acronym{ISA}s. If these function-scope annotations
24827 are present, @value{GDBN} uses them in preference to the global
24828 compressed @acronym{ISA} encoding setting.
24830 @item show mips compression
24831 @kindex show mips compression
24832 Show the @acronym{MIPS} compressed @acronym{ISA} encoding used by
24833 @value{GDBN} to debug the inferior.
24836 @itemx show mipsfpu
24837 @xref{MIPS Embedded, set mipsfpu}.
24839 @item set mips mask-address @var{arg}
24840 @kindex set mips mask-address
24841 @cindex @acronym{MIPS} addresses, masking
24842 This command determines whether the most-significant 32 bits of 64-bit
24843 @acronym{MIPS} addresses are masked off. The argument @var{arg} can be
24844 @samp{on}, @samp{off}, or @samp{auto}. The latter is the default
24845 setting, which lets @value{GDBN} determine the correct value.
24847 @item show mips mask-address
24848 @kindex show mips mask-address
24849 Show whether the upper 32 bits of @acronym{MIPS} addresses are masked off or
24852 @item set remote-mips64-transfers-32bit-regs
24853 @kindex set remote-mips64-transfers-32bit-regs
24854 This command controls compatibility with 64-bit @acronym{MIPS} targets that
24855 transfer data in 32-bit quantities. If you have an old @acronym{MIPS} 64 target
24856 that transfers 32 bits for some registers, like @sc{sr} and @sc{fsr},
24857 and 64 bits for other registers, set this option to @samp{on}.
24859 @item show remote-mips64-transfers-32bit-regs
24860 @kindex show remote-mips64-transfers-32bit-regs
24861 Show the current setting of compatibility with older @acronym{MIPS} 64 targets.
24863 @item set debug mips
24864 @kindex set debug mips
24865 This command turns on and off debugging messages for the @acronym{MIPS}-specific
24866 target code in @value{GDBN}.
24868 @item show debug mips
24869 @kindex show debug mips
24870 Show the current setting of @acronym{MIPS} debugging messages.
24876 @cindex HPPA support
24878 When @value{GDBN} is debugging the HP PA architecture, it provides the
24879 following special commands:
24882 @item set debug hppa
24883 @kindex set debug hppa
24884 This command determines whether HPPA architecture-specific debugging
24885 messages are to be displayed.
24887 @item show debug hppa
24888 Show whether HPPA debugging messages are displayed.
24890 @item maint print unwind @var{address}
24891 @kindex maint print unwind@r{, HPPA}
24892 This command displays the contents of the unwind table entry at the
24893 given @var{address}.
24899 @subsection PowerPC
24900 @cindex PowerPC architecture
24902 When @value{GDBN} is debugging the PowerPC architecture, it provides a set of
24903 pseudo-registers to enable inspection of 128-bit wide Decimal Floating Point
24904 numbers stored in the floating point registers. These values must be stored
24905 in two consecutive registers, always starting at an even register like
24906 @code{f0} or @code{f2}.
24908 The pseudo-registers go from @code{$dl0} through @code{$dl15}, and are formed
24909 by joining the even/odd register pairs @code{f0} and @code{f1} for @code{$dl0},
24910 @code{f2} and @code{f3} for @code{$dl1} and so on.
24912 For POWER7 processors, @value{GDBN} provides a set of pseudo-registers, the 64-bit
24913 wide Extended Floating Point Registers (@samp{f32} through @samp{f63}).
24916 @subsection Nios II
24917 @cindex Nios II architecture
24919 When @value{GDBN} is debugging the Nios II architecture,
24920 it provides the following special commands:
24924 @item set debug nios2
24925 @kindex set debug nios2
24926 This command turns on and off debugging messages for the Nios II
24927 target code in @value{GDBN}.
24929 @item show debug nios2
24930 @kindex show debug nios2
24931 Show the current setting of Nios II debugging messages.
24935 @subsection Sparc64
24936 @cindex Sparc64 support
24937 @cindex Application Data Integrity
24938 @subsubsection ADI Support
24940 The M7 processor supports an Application Data Integrity (ADI) feature that
24941 detects invalid data accesses. When software allocates memory and enables
24942 ADI on the allocated memory, it chooses a 4-bit version number, sets the
24943 version in the upper 4 bits of the 64-bit pointer to that data, and stores
24944 the 4-bit version in every cacheline of that data. Hardware saves the latter
24945 in spare bits in the cache and memory hierarchy. On each load and store,
24946 the processor compares the upper 4 VA (virtual address) bits to the
24947 cacheline's version. If there is a mismatch, the processor generates a
24948 version mismatch trap which can be either precise or disrupting. The trap
24949 is an error condition which the kernel delivers to the process as a SIGSEGV
24952 Note that only 64-bit applications can use ADI and need to be built with
24955 Values of the ADI version tags, which are in granularity of a
24956 cacheline (64 bytes), can be viewed or modified.
24960 @kindex adi examine
24961 @item adi (examine | x) [ / @var{n} ] @var{addr}
24963 The @code{adi examine} command displays the value of one ADI version tag per
24966 @var{n} is a decimal integer specifying the number in bytes; the default
24967 is 1. It specifies how much ADI version information, at the ratio of 1:ADI
24968 block size, to display.
24970 @var{addr} is the address in user address space where you want @value{GDBN}
24971 to begin displaying the ADI version tags.
24973 Below is an example of displaying ADI versions of variable "shmaddr".
24976 (@value{GDBP}) adi x/100 shmaddr
24977 0xfff800010002c000: 0 0
24981 @item adi (assign | a) [ / @var{n} ] @var{addr} = @var{tag}
24983 The @code{adi assign} command is used to assign new ADI version tag
24986 @var{n} is a decimal integer specifying the number in bytes;
24987 the default is 1. It specifies how much ADI version information, at the
24988 ratio of 1:ADI block size, to modify.
24990 @var{addr} is the address in user address space where you want @value{GDBN}
24991 to begin modifying the ADI version tags.
24993 @var{tag} is the new ADI version tag.
24995 For example, do the following to modify then verify ADI versions of
24996 variable "shmaddr":
24999 (@value{GDBP}) adi a/100 shmaddr = 7
25000 (@value{GDBP}) adi x/100 shmaddr
25001 0xfff800010002c000: 7 7
25008 @cindex S12Z support
25010 When @value{GDBN} is debugging the S12Z architecture,
25011 it provides the following special command:
25014 @item maint info bdccsr
25015 @kindex maint info bdccsr@r{, S12Z}
25016 This command displays the current value of the microprocessor's
25021 @node Controlling GDB
25022 @chapter Controlling @value{GDBN}
25024 You can alter the way @value{GDBN} interacts with you by using the
25025 @code{set} command. For commands controlling how @value{GDBN} displays
25026 data, see @ref{Print Settings, ,Print Settings}. Other settings are
25031 * Editing:: Command editing
25032 * Command History:: Command history
25033 * Screen Size:: Screen size
25034 * Output Styling:: Output styling
25035 * Numbers:: Numbers
25036 * ABI:: Configuring the current ABI
25037 * Auto-loading:: Automatically loading associated files
25038 * Messages/Warnings:: Optional warnings and messages
25039 * Debugging Output:: Optional messages about internal happenings
25040 * Other Misc Settings:: Other Miscellaneous Settings
25048 @value{GDBN} indicates its readiness to read a command by printing a string
25049 called the @dfn{prompt}. This string is normally @samp{(@value{GDBP})}. You
25050 can change the prompt string with the @code{set prompt} command. For
25051 instance, when debugging @value{GDBN} with @value{GDBN}, it is useful to change
25052 the prompt in one of the @value{GDBN} sessions so that you can always tell
25053 which one you are talking to.
25055 @emph{Note:} @code{set prompt} does not add a space for you after the
25056 prompt you set. This allows you to set a prompt which ends in a space
25057 or a prompt that does not.
25061 @item set prompt @var{newprompt}
25062 Directs @value{GDBN} to use @var{newprompt} as its prompt string henceforth.
25064 @kindex show prompt
25066 Prints a line of the form: @samp{Gdb's prompt is: @var{your-prompt}}
25069 Versions of @value{GDBN} that ship with Python scripting enabled have
25070 prompt extensions. The commands for interacting with these extensions
25074 @kindex set extended-prompt
25075 @item set extended-prompt @var{prompt}
25076 Set an extended prompt that allows for substitutions.
25077 @xref{gdb.prompt}, for a list of escape sequences that can be used for
25078 substitution. Any escape sequences specified as part of the prompt
25079 string are replaced with the corresponding strings each time the prompt
25085 set extended-prompt Current working directory: \w (gdb)
25088 Note that when an extended-prompt is set, it takes control of the
25089 @var{prompt_hook} hook. @xref{prompt_hook}, for further information.
25091 @kindex show extended-prompt
25092 @item show extended-prompt
25093 Prints the extended prompt. Any escape sequences specified as part of
25094 the prompt string with @code{set extended-prompt}, are replaced with the
25095 corresponding strings each time the prompt is displayed.
25099 @section Command Editing
25101 @cindex command line editing
25103 @value{GDBN} reads its input commands via the @dfn{Readline} interface. This
25104 @sc{gnu} library provides consistent behavior for programs which provide a
25105 command line interface to the user. Advantages are @sc{gnu} Emacs-style
25106 or @dfn{vi}-style inline editing of commands, @code{csh}-like history
25107 substitution, and a storage and recall of command history across
25108 debugging sessions.
25110 You may control the behavior of command line editing in @value{GDBN} with the
25111 command @code{set}.
25114 @kindex set editing
25117 @itemx set editing on
25118 Enable command line editing (enabled by default).
25120 @item set editing off
25121 Disable command line editing.
25123 @kindex show editing
25125 Show whether command line editing is enabled.
25128 @ifset SYSTEM_READLINE
25129 @xref{Command Line Editing, , , rluserman, GNU Readline Library},
25131 @ifclear SYSTEM_READLINE
25132 @xref{Command Line Editing},
25134 for more details about the Readline
25135 interface. Users unfamiliar with @sc{gnu} Emacs or @code{vi} are
25136 encouraged to read that chapter.
25138 @cindex Readline application name
25139 @value{GDBN} sets the Readline application name to @samp{gdb}. This
25140 is useful for conditions in @file{.inputrc}.
25142 @cindex operate-and-get-next
25143 @value{GDBN} defines a bindable Readline command,
25144 @code{operate-and-get-next}. This is bound to @kbd{C-o} by default.
25145 This command accepts the current line for execution and fetches the
25146 next line relative to the current line from the history for editing.
25147 Any argument is ignored.
25149 @node Command History
25150 @section Command History
25151 @cindex command history
25153 @value{GDBN} can keep track of the commands you type during your
25154 debugging sessions, so that you can be certain of precisely what
25155 happened. Use these commands to manage the @value{GDBN} command
25158 @value{GDBN} uses the @sc{gnu} History library, a part of the Readline
25159 package, to provide the history facility.
25160 @ifset SYSTEM_READLINE
25161 @xref{Using History Interactively, , , history, GNU History Library},
25163 @ifclear SYSTEM_READLINE
25164 @xref{Using History Interactively},
25166 for the detailed description of the History library.
25168 To issue a command to @value{GDBN} without affecting certain aspects of
25169 the state which is seen by users, prefix it with @samp{server }
25170 (@pxref{Server Prefix}). This
25171 means that this command will not affect the command history, nor will it
25172 affect @value{GDBN}'s notion of which command to repeat if @key{RET} is
25173 pressed on a line by itself.
25175 @cindex @code{server}, command prefix
25176 The server prefix does not affect the recording of values into the value
25177 history; to print a value without recording it into the value history,
25178 use the @code{output} command instead of the @code{print} command.
25180 Here is the description of @value{GDBN} commands related to command
25184 @cindex history substitution
25185 @cindex history file
25186 @kindex set history filename
25187 @cindex @env{GDBHISTFILE}, environment variable
25188 @item set history filename @var{fname}
25189 Set the name of the @value{GDBN} command history file to @var{fname}.
25190 This is the file where @value{GDBN} reads an initial command history
25191 list, and where it writes the command history from this session when it
25192 exits. You can access this list through history expansion or through
25193 the history command editing characters listed below. This file defaults
25194 to the value of the environment variable @code{GDBHISTFILE}, or to
25195 @file{./.gdb_history} (@file{./_gdb_history} on MS-DOS) if this variable
25198 @cindex save command history
25199 @kindex set history save
25200 @item set history save
25201 @itemx set history save on
25202 Record command history in a file, whose name may be specified with the
25203 @code{set history filename} command. By default, this option is disabled.
25205 @item set history save off
25206 Stop recording command history in a file.
25208 @cindex history size
25209 @kindex set history size
25210 @cindex @env{GDBHISTSIZE}, environment variable
25211 @item set history size @var{size}
25212 @itemx set history size unlimited
25213 Set the number of commands which @value{GDBN} keeps in its history list.
25214 This defaults to the value of the environment variable @env{GDBHISTSIZE}, or
25215 to 256 if this variable is not set. Non-numeric values of @env{GDBHISTSIZE}
25216 are ignored. If @var{size} is @code{unlimited} or if @env{GDBHISTSIZE} is
25217 either a negative number or the empty string, then the number of commands
25218 @value{GDBN} keeps in the history list is unlimited.
25220 @cindex remove duplicate history
25221 @kindex set history remove-duplicates
25222 @item set history remove-duplicates @var{count}
25223 @itemx set history remove-duplicates unlimited
25224 Control the removal of duplicate history entries in the command history list.
25225 If @var{count} is non-zero, @value{GDBN} will look back at the last @var{count}
25226 history entries and remove the first entry that is a duplicate of the current
25227 entry being added to the command history list. If @var{count} is
25228 @code{unlimited} then this lookbehind is unbounded. If @var{count} is 0, then
25229 removal of duplicate history entries is disabled.
25231 Only history entries added during the current session are considered for
25232 removal. This option is set to 0 by default.
25236 History expansion assigns special meaning to the character @kbd{!}.
25237 @ifset SYSTEM_READLINE
25238 @xref{Event Designators, , , history, GNU History Library},
25240 @ifclear SYSTEM_READLINE
25241 @xref{Event Designators},
25245 @cindex history expansion, turn on/off
25246 Since @kbd{!} is also the logical not operator in C, history expansion
25247 is off by default. If you decide to enable history expansion with the
25248 @code{set history expansion on} command, you may sometimes need to
25249 follow @kbd{!} (when it is used as logical not, in an expression) with
25250 a space or a tab to prevent it from being expanded. The readline
25251 history facilities do not attempt substitution on the strings
25252 @kbd{!=} and @kbd{!(}, even when history expansion is enabled.
25254 The commands to control history expansion are:
25257 @item set history expansion on
25258 @itemx set history expansion
25259 @kindex set history expansion
25260 Enable history expansion. History expansion is off by default.
25262 @item set history expansion off
25263 Disable history expansion.
25266 @kindex show history
25268 @itemx show history filename
25269 @itemx show history save
25270 @itemx show history size
25271 @itemx show history expansion
25272 These commands display the state of the @value{GDBN} history parameters.
25273 @code{show history} by itself displays all four states.
25278 @kindex show commands
25279 @cindex show last commands
25280 @cindex display command history
25281 @item show commands
25282 Display the last ten commands in the command history.
25284 @item show commands @var{n}
25285 Print ten commands centered on command number @var{n}.
25287 @item show commands +
25288 Print ten commands just after the commands last printed.
25292 @section Screen Size
25293 @cindex size of screen
25294 @cindex screen size
25297 @cindex pauses in output
25299 Certain commands to @value{GDBN} may produce large amounts of
25300 information output to the screen. To help you read all of it,
25301 @value{GDBN} pauses and asks you for input at the end of each page of
25302 output. Type @key{RET} when you want to see one more page of output,
25303 @kbd{q} to discard the remaining output, or @kbd{c} to continue
25304 without paging for the rest of the current command. Also, the screen
25305 width setting determines when to wrap lines of output. Depending on
25306 what is being printed, @value{GDBN} tries to break the line at a
25307 readable place, rather than simply letting it overflow onto the
25310 Normally @value{GDBN} knows the size of the screen from the terminal
25311 driver software. For example, on Unix @value{GDBN} uses the termcap data base
25312 together with the value of the @code{TERM} environment variable and the
25313 @code{stty rows} and @code{stty cols} settings. If this is not correct,
25314 you can override it with the @code{set height} and @code{set
25321 @kindex show height
25322 @item set height @var{lpp}
25323 @itemx set height unlimited
25325 @itemx set width @var{cpl}
25326 @itemx set width unlimited
25328 These @code{set} commands specify a screen height of @var{lpp} lines and
25329 a screen width of @var{cpl} characters. The associated @code{show}
25330 commands display the current settings.
25332 If you specify a height of either @code{unlimited} or zero lines,
25333 @value{GDBN} does not pause during output no matter how long the
25334 output is. This is useful if output is to a file or to an editor
25337 Likewise, you can specify @samp{set width unlimited} or @samp{set
25338 width 0} to prevent @value{GDBN} from wrapping its output.
25340 @item set pagination on
25341 @itemx set pagination off
25342 @kindex set pagination
25343 Turn the output pagination on or off; the default is on. Turning
25344 pagination off is the alternative to @code{set height unlimited}. Note that
25345 running @value{GDBN} with the @option{--batch} option (@pxref{Mode
25346 Options, -batch}) also automatically disables pagination.
25348 @item show pagination
25349 @kindex show pagination
25350 Show the current pagination mode.
25353 @node Output Styling
25354 @section Output Styling
25360 @value{GDBN} can style its output on a capable terminal. This is
25361 enabled by default on most systems, but disabled by default when in
25362 batch mode (@pxref{Mode Options}). Various style settings are available;
25363 and styles can also be disabled entirely.
25366 @item set style enabled @samp{on|off}
25367 Enable or disable all styling. The default is host-dependent, with
25368 most hosts defaulting to @samp{on}.
25370 @item show style enabled
25371 Show the current state of styling.
25373 @item set style sources @samp{on|off}
25374 Enable or disable source code styling. This affects whether source
25375 code, such as the output of the @code{list} command, is styled. Note
25376 that source styling only works if styling in general is enabled, and
25377 if @value{GDBN} was linked with the GNU Source Highlight library. The
25378 default is @samp{on}.
25380 @item show style sources
25381 Show the current state of source code styling.
25384 Subcommands of @code{set style} control specific forms of styling.
25385 These subcommands all follow the same pattern: each style-able object
25386 can be styled with a foreground color, a background color, and an
25389 For example, the style of file names can be controlled using the
25390 @code{set style filename} group of commands:
25393 @item set style filename background @var{color}
25394 Set the background to @var{color}. Valid colors are @samp{none}
25395 (meaning the terminal's default color), @samp{black}, @samp{red},
25396 @samp{green}, @samp{yellow}, @samp{blue}, @samp{magenta}, @samp{cyan},
25399 @item set style filename foreground @var{color}
25400 Set the foreground to @var{color}. Valid colors are @samp{none}
25401 (meaning the terminal's default color), @samp{black}, @samp{red},
25402 @samp{green}, @samp{yellow}, @samp{blue}, @samp{magenta}, @samp{cyan},
25405 @item set style filename intensity @var{value}
25406 Set the intensity to @var{value}. Valid intensities are @samp{normal}
25407 (the default), @samp{bold}, and @samp{dim}.
25410 The @code{show style} command and its subcommands are styling
25411 a style name in their output using its own style.
25412 So, use @command{show style} to see the complete list of styles,
25413 their characteristics and the visual aspect of each style.
25415 The style-able objects are:
25418 Control the styling of file names. By default, this style's
25419 foreground color is green.
25422 Control the styling of function names. These are managed with the
25423 @code{set style function} family of commands. By default, this
25424 style's foreground color is yellow.
25427 Control the styling of variable names. These are managed with the
25428 @code{set style variable} family of commands. By default, this style's
25429 foreground color is cyan.
25432 Control the styling of addresses. These are managed with the
25433 @code{set style address} family of commands. By default, this style's
25434 foreground color is blue.
25437 Control the styling of titles. These are managed with the
25438 @code{set style title} family of commands. By default, this style's
25439 intensity is bold. Commands are using the title style to improve
25440 the readability of large output. For example, the commands
25441 @command{apropos} and @command{help} are using the title style
25442 for the command names.
25445 Control the styling of highlightings. These are managed with the
25446 @code{set style highlight} family of commands. By default, this style's
25447 foreground color is red. Commands are using the highlight style to draw
25448 the user attention to some specific parts of their output. For example,
25449 the command @command{apropos -v REGEXP} uses the highlight style to
25450 mark the documentation parts matching @var{regexp}.
25456 @cindex number representation
25457 @cindex entering numbers
25459 You can always enter numbers in octal, decimal, or hexadecimal in
25460 @value{GDBN} by the usual conventions: octal numbers begin with
25461 @samp{0}, decimal numbers end with @samp{.}, and hexadecimal numbers
25462 begin with @samp{0x}. Numbers that neither begin with @samp{0} or
25463 @samp{0x}, nor end with a @samp{.} are, by default, entered in base
25464 10; likewise, the default display for numbers---when no particular
25465 format is specified---is base 10. You can change the default base for
25466 both input and output with the commands described below.
25469 @kindex set input-radix
25470 @item set input-radix @var{base}
25471 Set the default base for numeric input. Supported choices
25472 for @var{base} are decimal 8, 10, or 16. The base must itself be
25473 specified either unambiguously or using the current input radix; for
25477 set input-radix 012
25478 set input-radix 10.
25479 set input-radix 0xa
25483 sets the input base to decimal. On the other hand, @samp{set input-radix 10}
25484 leaves the input radix unchanged, no matter what it was, since
25485 @samp{10}, being without any leading or trailing signs of its base, is
25486 interpreted in the current radix. Thus, if the current radix is 16,
25487 @samp{10} is interpreted in hex, i.e.@: as 16 decimal, which doesn't
25490 @kindex set output-radix
25491 @item set output-radix @var{base}
25492 Set the default base for numeric display. Supported choices
25493 for @var{base} are decimal 8, 10, or 16. The base must itself be
25494 specified either unambiguously or using the current input radix.
25496 @kindex show input-radix
25497 @item show input-radix
25498 Display the current default base for numeric input.
25500 @kindex show output-radix
25501 @item show output-radix
25502 Display the current default base for numeric display.
25504 @item set radix @r{[}@var{base}@r{]}
25508 These commands set and show the default base for both input and output
25509 of numbers. @code{set radix} sets the radix of input and output to
25510 the same base; without an argument, it resets the radix back to its
25511 default value of 10.
25516 @section Configuring the Current ABI
25518 @value{GDBN} can determine the @dfn{ABI} (Application Binary Interface) of your
25519 application automatically. However, sometimes you need to override its
25520 conclusions. Use these commands to manage @value{GDBN}'s view of the
25526 @cindex Newlib OS ABI and its influence on the longjmp handling
25528 One @value{GDBN} configuration can debug binaries for multiple operating
25529 system targets, either via remote debugging or native emulation.
25530 @value{GDBN} will autodetect the @dfn{OS ABI} (Operating System ABI) in use,
25531 but you can override its conclusion using the @code{set osabi} command.
25532 One example where this is useful is in debugging of binaries which use
25533 an alternate C library (e.g.@: @sc{uClibc} for @sc{gnu}/Linux) which does
25534 not have the same identifying marks that the standard C library for your
25537 When @value{GDBN} is debugging the AArch64 architecture, it provides a
25538 ``Newlib'' OS ABI. This is useful for handling @code{setjmp} and
25539 @code{longjmp} when debugging binaries that use the @sc{newlib} C library.
25540 The ``Newlib'' OS ABI can be selected by @code{set osabi Newlib}.
25544 Show the OS ABI currently in use.
25547 With no argument, show the list of registered available OS ABI's.
25549 @item set osabi @var{abi}
25550 Set the current OS ABI to @var{abi}.
25553 @cindex float promotion
25555 Generally, the way that an argument of type @code{float} is passed to a
25556 function depends on whether the function is prototyped. For a prototyped
25557 (i.e.@: ANSI/ISO style) function, @code{float} arguments are passed unchanged,
25558 according to the architecture's convention for @code{float}. For unprototyped
25559 (i.e.@: K&R style) functions, @code{float} arguments are first promoted to type
25560 @code{double} and then passed.
25562 Unfortunately, some forms of debug information do not reliably indicate whether
25563 a function is prototyped. If @value{GDBN} calls a function that is not marked
25564 as prototyped, it consults @kbd{set coerce-float-to-double}.
25567 @kindex set coerce-float-to-double
25568 @item set coerce-float-to-double
25569 @itemx set coerce-float-to-double on
25570 Arguments of type @code{float} will be promoted to @code{double} when passed
25571 to an unprototyped function. This is the default setting.
25573 @item set coerce-float-to-double off
25574 Arguments of type @code{float} will be passed directly to unprototyped
25577 @kindex show coerce-float-to-double
25578 @item show coerce-float-to-double
25579 Show the current setting of promoting @code{float} to @code{double}.
25583 @kindex show cp-abi
25584 @value{GDBN} needs to know the ABI used for your program's C@t{++}
25585 objects. The correct C@t{++} ABI depends on which C@t{++} compiler was
25586 used to build your application. @value{GDBN} only fully supports
25587 programs with a single C@t{++} ABI; if your program contains code using
25588 multiple C@t{++} ABI's or if @value{GDBN} can not identify your
25589 program's ABI correctly, you can tell @value{GDBN} which ABI to use.
25590 Currently supported ABI's include ``gnu-v2'', for @code{g++} versions
25591 before 3.0, ``gnu-v3'', for @code{g++} versions 3.0 and later, and
25592 ``hpaCC'' for the HP ANSI C@t{++} compiler. Other C@t{++} compilers may
25593 use the ``gnu-v2'' or ``gnu-v3'' ABI's as well. The default setting is
25598 Show the C@t{++} ABI currently in use.
25601 With no argument, show the list of supported C@t{++} ABI's.
25603 @item set cp-abi @var{abi}
25604 @itemx set cp-abi auto
25605 Set the current C@t{++} ABI to @var{abi}, or return to automatic detection.
25609 @section Automatically loading associated files
25610 @cindex auto-loading
25612 @value{GDBN} sometimes reads files with commands and settings automatically,
25613 without being explicitly told so by the user. We call this feature
25614 @dfn{auto-loading}. While auto-loading is useful for automatically adapting
25615 @value{GDBN} to the needs of your project, it can sometimes produce unexpected
25616 results or introduce security risks (e.g., if the file comes from untrusted
25620 * Init File in the Current Directory:: @samp{set/show/info auto-load local-gdbinit}
25621 * libthread_db.so.1 file:: @samp{set/show/info auto-load libthread-db}
25623 * Auto-loading safe path:: @samp{set/show/info auto-load safe-path}
25624 * Auto-loading verbose mode:: @samp{set/show debug auto-load}
25627 There are various kinds of files @value{GDBN} can automatically load.
25628 In addition to these files, @value{GDBN} supports auto-loading code written
25629 in various extension languages. @xref{Auto-loading extensions}.
25631 Note that loading of these associated files (including the local @file{.gdbinit}
25632 file) requires accordingly configured @code{auto-load safe-path}
25633 (@pxref{Auto-loading safe path}).
25635 For these reasons, @value{GDBN} includes commands and options to let you
25636 control when to auto-load files and which files should be auto-loaded.
25639 @anchor{set auto-load off}
25640 @kindex set auto-load off
25641 @item set auto-load off
25642 Globally disable loading of all auto-loaded files.
25643 You may want to use this command with the @samp{-iex} option
25644 (@pxref{Option -init-eval-command}) such as:
25646 $ @kbd{gdb -iex "set auto-load off" untrusted-executable corefile}
25649 Be aware that system init file (@pxref{System-wide configuration})
25650 and init files from your home directory (@pxref{Home Directory Init File})
25651 still get read (as they come from generally trusted directories).
25652 To prevent @value{GDBN} from auto-loading even those init files, use the
25653 @option{-nx} option (@pxref{Mode Options}), in addition to
25654 @code{set auto-load no}.
25656 @anchor{show auto-load}
25657 @kindex show auto-load
25658 @item show auto-load
25659 Show whether auto-loading of each specific @samp{auto-load} file(s) is enabled
25663 (gdb) show auto-load
25664 gdb-scripts: Auto-loading of canned sequences of commands scripts is on.
25665 libthread-db: Auto-loading of inferior specific libthread_db is on.
25666 local-gdbinit: Auto-loading of .gdbinit script from current directory
25668 python-scripts: Auto-loading of Python scripts is on.
25669 safe-path: List of directories from which it is safe to auto-load files
25670 is $debugdir:$datadir/auto-load.
25671 scripts-directory: List of directories from which to load auto-loaded scripts
25672 is $debugdir:$datadir/auto-load.
25675 @anchor{info auto-load}
25676 @kindex info auto-load
25677 @item info auto-load
25678 Print whether each specific @samp{auto-load} file(s) have been auto-loaded or
25682 (gdb) info auto-load
25685 Yes /home/user/gdb/gdb-gdb.gdb
25686 libthread-db: No auto-loaded libthread-db.
25687 local-gdbinit: Local .gdbinit file "/home/user/gdb/.gdbinit" has been
25691 Yes /home/user/gdb/gdb-gdb.py
25695 These are @value{GDBN} control commands for the auto-loading:
25697 @multitable @columnfractions .5 .5
25698 @item @xref{set auto-load off}.
25699 @tab Disable auto-loading globally.
25700 @item @xref{show auto-load}.
25701 @tab Show setting of all kinds of files.
25702 @item @xref{info auto-load}.
25703 @tab Show state of all kinds of files.
25704 @item @xref{set auto-load gdb-scripts}.
25705 @tab Control for @value{GDBN} command scripts.
25706 @item @xref{show auto-load gdb-scripts}.
25707 @tab Show setting of @value{GDBN} command scripts.
25708 @item @xref{info auto-load gdb-scripts}.
25709 @tab Show state of @value{GDBN} command scripts.
25710 @item @xref{set auto-load python-scripts}.
25711 @tab Control for @value{GDBN} Python scripts.
25712 @item @xref{show auto-load python-scripts}.
25713 @tab Show setting of @value{GDBN} Python scripts.
25714 @item @xref{info auto-load python-scripts}.
25715 @tab Show state of @value{GDBN} Python scripts.
25716 @item @xref{set auto-load guile-scripts}.
25717 @tab Control for @value{GDBN} Guile scripts.
25718 @item @xref{show auto-load guile-scripts}.
25719 @tab Show setting of @value{GDBN} Guile scripts.
25720 @item @xref{info auto-load guile-scripts}.
25721 @tab Show state of @value{GDBN} Guile scripts.
25722 @item @xref{set auto-load scripts-directory}.
25723 @tab Control for @value{GDBN} auto-loaded scripts location.
25724 @item @xref{show auto-load scripts-directory}.
25725 @tab Show @value{GDBN} auto-loaded scripts location.
25726 @item @xref{add-auto-load-scripts-directory}.
25727 @tab Add directory for auto-loaded scripts location list.
25728 @item @xref{set auto-load local-gdbinit}.
25729 @tab Control for init file in the current directory.
25730 @item @xref{show auto-load local-gdbinit}.
25731 @tab Show setting of init file in the current directory.
25732 @item @xref{info auto-load local-gdbinit}.
25733 @tab Show state of init file in the current directory.
25734 @item @xref{set auto-load libthread-db}.
25735 @tab Control for thread debugging library.
25736 @item @xref{show auto-load libthread-db}.
25737 @tab Show setting of thread debugging library.
25738 @item @xref{info auto-load libthread-db}.
25739 @tab Show state of thread debugging library.
25740 @item @xref{set auto-load safe-path}.
25741 @tab Control directories trusted for automatic loading.
25742 @item @xref{show auto-load safe-path}.
25743 @tab Show directories trusted for automatic loading.
25744 @item @xref{add-auto-load-safe-path}.
25745 @tab Add directory trusted for automatic loading.
25748 @node Init File in the Current Directory
25749 @subsection Automatically loading init file in the current directory
25750 @cindex auto-loading init file in the current directory
25752 By default, @value{GDBN} reads and executes the canned sequences of commands
25753 from init file (if any) in the current working directory,
25754 see @ref{Init File in the Current Directory during Startup}.
25756 Note that loading of this local @file{.gdbinit} file also requires accordingly
25757 configured @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
25760 @anchor{set auto-load local-gdbinit}
25761 @kindex set auto-load local-gdbinit
25762 @item set auto-load local-gdbinit [on|off]
25763 Enable or disable the auto-loading of canned sequences of commands
25764 (@pxref{Sequences}) found in init file in the current directory.
25766 @anchor{show auto-load local-gdbinit}
25767 @kindex show auto-load local-gdbinit
25768 @item show auto-load local-gdbinit
25769 Show whether auto-loading of canned sequences of commands from init file in the
25770 current directory is enabled or disabled.
25772 @anchor{info auto-load local-gdbinit}
25773 @kindex info auto-load local-gdbinit
25774 @item info auto-load local-gdbinit
25775 Print whether canned sequences of commands from init file in the
25776 current directory have been auto-loaded.
25779 @node libthread_db.so.1 file
25780 @subsection Automatically loading thread debugging library
25781 @cindex auto-loading libthread_db.so.1
25783 This feature is currently present only on @sc{gnu}/Linux native hosts.
25785 @value{GDBN} reads in some cases thread debugging library from places specific
25786 to the inferior (@pxref{set libthread-db-search-path}).
25788 The special @samp{libthread-db-search-path} entry @samp{$sdir} is processed
25789 without checking this @samp{set auto-load libthread-db} switch as system
25790 libraries have to be trusted in general. In all other cases of
25791 @samp{libthread-db-search-path} entries @value{GDBN} checks first if @samp{set
25792 auto-load libthread-db} is enabled before trying to open such thread debugging
25795 Note that loading of this debugging library also requires accordingly configured
25796 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
25799 @anchor{set auto-load libthread-db}
25800 @kindex set auto-load libthread-db
25801 @item set auto-load libthread-db [on|off]
25802 Enable or disable the auto-loading of inferior specific thread debugging library.
25804 @anchor{show auto-load libthread-db}
25805 @kindex show auto-load libthread-db
25806 @item show auto-load libthread-db
25807 Show whether auto-loading of inferior specific thread debugging library is
25808 enabled or disabled.
25810 @anchor{info auto-load libthread-db}
25811 @kindex info auto-load libthread-db
25812 @item info auto-load libthread-db
25813 Print the list of all loaded inferior specific thread debugging libraries and
25814 for each such library print list of inferior @var{pid}s using it.
25817 @node Auto-loading safe path
25818 @subsection Security restriction for auto-loading
25819 @cindex auto-loading safe-path
25821 As the files of inferior can come from untrusted source (such as submitted by
25822 an application user) @value{GDBN} does not always load any files automatically.
25823 @value{GDBN} provides the @samp{set auto-load safe-path} setting to list
25824 directories trusted for loading files not explicitly requested by user.
25825 Each directory can also be a shell wildcard pattern.
25827 If the path is not set properly you will see a warning and the file will not
25832 Reading symbols from /home/user/gdb/gdb...done.
25833 warning: File "/home/user/gdb/gdb-gdb.gdb" auto-loading has been
25834 declined by your `auto-load safe-path' set
25835 to "$debugdir:$datadir/auto-load".
25836 warning: File "/home/user/gdb/gdb-gdb.py" auto-loading has been
25837 declined by your `auto-load safe-path' set
25838 to "$debugdir:$datadir/auto-load".
25842 To instruct @value{GDBN} to go ahead and use the init files anyway,
25843 invoke @value{GDBN} like this:
25846 $ gdb -q -iex "set auto-load safe-path /home/user/gdb" ./gdb
25849 The list of trusted directories is controlled by the following commands:
25852 @anchor{set auto-load safe-path}
25853 @kindex set auto-load safe-path
25854 @item set auto-load safe-path @r{[}@var{directories}@r{]}
25855 Set the list of directories (and their subdirectories) trusted for automatic
25856 loading and execution of scripts. You can also enter a specific trusted file.
25857 Each directory can also be a shell wildcard pattern; wildcards do not match
25858 directory separator - see @code{FNM_PATHNAME} for system function @code{fnmatch}
25859 (@pxref{Wildcard Matching, fnmatch, , libc, GNU C Library Reference Manual}).
25860 If you omit @var{directories}, @samp{auto-load safe-path} will be reset to
25861 its default value as specified during @value{GDBN} compilation.
25863 The list of directories uses path separator (@samp{:} on GNU and Unix
25864 systems, @samp{;} on MS-Windows and MS-DOS) to separate directories, similarly
25865 to the @env{PATH} environment variable.
25867 @anchor{show auto-load safe-path}
25868 @kindex show auto-load safe-path
25869 @item show auto-load safe-path
25870 Show the list of directories trusted for automatic loading and execution of
25873 @anchor{add-auto-load-safe-path}
25874 @kindex add-auto-load-safe-path
25875 @item add-auto-load-safe-path
25876 Add an entry (or list of entries) to the list of directories trusted for
25877 automatic loading and execution of scripts. Multiple entries may be delimited
25878 by the host platform path separator in use.
25881 This variable defaults to what @code{--with-auto-load-dir} has been configured
25882 to (@pxref{with-auto-load-dir}). @file{$debugdir} and @file{$datadir}
25883 substitution applies the same as for @ref{set auto-load scripts-directory}.
25884 The default @code{set auto-load safe-path} value can be also overriden by
25885 @value{GDBN} configuration option @option{--with-auto-load-safe-path}.
25887 Setting this variable to @file{/} disables this security protection,
25888 corresponding @value{GDBN} configuration option is
25889 @option{--without-auto-load-safe-path}.
25890 This variable is supposed to be set to the system directories writable by the
25891 system superuser only. Users can add their source directories in init files in
25892 their home directories (@pxref{Home Directory Init File}). See also deprecated
25893 init file in the current directory
25894 (@pxref{Init File in the Current Directory during Startup}).
25896 To force @value{GDBN} to load the files it declined to load in the previous
25897 example, you could use one of the following ways:
25900 @item @file{~/.gdbinit}: @samp{add-auto-load-safe-path ~/src/gdb}
25901 Specify this trusted directory (or a file) as additional component of the list.
25902 You have to specify also any existing directories displayed by
25903 by @samp{show auto-load safe-path} (such as @samp{/usr:/bin} in this example).
25905 @item @kbd{gdb -iex "set auto-load safe-path /usr:/bin:~/src/gdb" @dots{}}
25906 Specify this directory as in the previous case but just for a single
25907 @value{GDBN} session.
25909 @item @kbd{gdb -iex "set auto-load safe-path /" @dots{}}
25910 Disable auto-loading safety for a single @value{GDBN} session.
25911 This assumes all the files you debug during this @value{GDBN} session will come
25912 from trusted sources.
25914 @item @kbd{./configure --without-auto-load-safe-path}
25915 During compilation of @value{GDBN} you may disable any auto-loading safety.
25916 This assumes all the files you will ever debug with this @value{GDBN} come from
25920 On the other hand you can also explicitly forbid automatic files loading which
25921 also suppresses any such warning messages:
25924 @item @kbd{gdb -iex "set auto-load no" @dots{}}
25925 You can use @value{GDBN} command-line option for a single @value{GDBN} session.
25927 @item @file{~/.gdbinit}: @samp{set auto-load no}
25928 Disable auto-loading globally for the user
25929 (@pxref{Home Directory Init File}). While it is improbable, you could also
25930 use system init file instead (@pxref{System-wide configuration}).
25933 This setting applies to the file names as entered by user. If no entry matches
25934 @value{GDBN} tries as a last resort to also resolve all the file names into
25935 their canonical form (typically resolving symbolic links) and compare the
25936 entries again. @value{GDBN} already canonicalizes most of the filenames on its
25937 own before starting the comparison so a canonical form of directories is
25938 recommended to be entered.
25940 @node Auto-loading verbose mode
25941 @subsection Displaying files tried for auto-load
25942 @cindex auto-loading verbose mode
25944 For better visibility of all the file locations where you can place scripts to
25945 be auto-loaded with inferior --- or to protect yourself against accidental
25946 execution of untrusted scripts --- @value{GDBN} provides a feature for printing
25947 all the files attempted to be loaded. Both existing and non-existing files may
25950 For example the list of directories from which it is safe to auto-load files
25951 (@pxref{Auto-loading safe path}) applies also to canonicalized filenames which
25952 may not be too obvious while setting it up.
25955 (gdb) set debug auto-load on
25956 (gdb) file ~/src/t/true
25957 auto-load: Loading canned sequences of commands script "/tmp/true-gdb.gdb"
25958 for objfile "/tmp/true".
25959 auto-load: Updating directories of "/usr:/opt".
25960 auto-load: Using directory "/usr".
25961 auto-load: Using directory "/opt".
25962 warning: File "/tmp/true-gdb.gdb" auto-loading has been declined
25963 by your `auto-load safe-path' set to "/usr:/opt".
25967 @anchor{set debug auto-load}
25968 @kindex set debug auto-load
25969 @item set debug auto-load [on|off]
25970 Set whether to print the filenames attempted to be auto-loaded.
25972 @anchor{show debug auto-load}
25973 @kindex show debug auto-load
25974 @item show debug auto-load
25975 Show whether printing of the filenames attempted to be auto-loaded is turned
25979 @node Messages/Warnings
25980 @section Optional Warnings and Messages
25982 @cindex verbose operation
25983 @cindex optional warnings
25984 By default, @value{GDBN} is silent about its inner workings. If you are
25985 running on a slow machine, you may want to use the @code{set verbose}
25986 command. This makes @value{GDBN} tell you when it does a lengthy
25987 internal operation, so you will not think it has crashed.
25989 Currently, the messages controlled by @code{set verbose} are those
25990 which announce that the symbol table for a source file is being read;
25991 see @code{symbol-file} in @ref{Files, ,Commands to Specify Files}.
25994 @kindex set verbose
25995 @item set verbose on
25996 Enables @value{GDBN} output of certain informational messages.
25998 @item set verbose off
25999 Disables @value{GDBN} output of certain informational messages.
26001 @kindex show verbose
26003 Displays whether @code{set verbose} is on or off.
26006 By default, if @value{GDBN} encounters bugs in the symbol table of an
26007 object file, it is silent; but if you are debugging a compiler, you may
26008 find this information useful (@pxref{Symbol Errors, ,Errors Reading
26013 @kindex set complaints
26014 @item set complaints @var{limit}
26015 Permits @value{GDBN} to output @var{limit} complaints about each type of
26016 unusual symbols before becoming silent about the problem. Set
26017 @var{limit} to zero to suppress all complaints; set it to a large number
26018 to prevent complaints from being suppressed.
26020 @kindex show complaints
26021 @item show complaints
26022 Displays how many symbol complaints @value{GDBN} is permitted to produce.
26026 @anchor{confirmation requests}
26027 By default, @value{GDBN} is cautious, and asks what sometimes seems to be a
26028 lot of stupid questions to confirm certain commands. For example, if
26029 you try to run a program which is already running:
26033 The program being debugged has been started already.
26034 Start it from the beginning? (y or n)
26037 If you are willing to unflinchingly face the consequences of your own
26038 commands, you can disable this ``feature'':
26042 @kindex set confirm
26044 @cindex confirmation
26045 @cindex stupid questions
26046 @item set confirm off
26047 Disables confirmation requests. Note that running @value{GDBN} with
26048 the @option{--batch} option (@pxref{Mode Options, -batch}) also
26049 automatically disables confirmation requests.
26051 @item set confirm on
26052 Enables confirmation requests (the default).
26054 @kindex show confirm
26056 Displays state of confirmation requests.
26060 @cindex command tracing
26061 If you need to debug user-defined commands or sourced files you may find it
26062 useful to enable @dfn{command tracing}. In this mode each command will be
26063 printed as it is executed, prefixed with one or more @samp{+} symbols, the
26064 quantity denoting the call depth of each command.
26067 @kindex set trace-commands
26068 @cindex command scripts, debugging
26069 @item set trace-commands on
26070 Enable command tracing.
26071 @item set trace-commands off
26072 Disable command tracing.
26073 @item show trace-commands
26074 Display the current state of command tracing.
26077 @node Debugging Output
26078 @section Optional Messages about Internal Happenings
26079 @cindex optional debugging messages
26081 @value{GDBN} has commands that enable optional debugging messages from
26082 various @value{GDBN} subsystems; normally these commands are of
26083 interest to @value{GDBN} maintainers, or when reporting a bug. This
26084 section documents those commands.
26087 @kindex set exec-done-display
26088 @item set exec-done-display
26089 Turns on or off the notification of asynchronous commands'
26090 completion. When on, @value{GDBN} will print a message when an
26091 asynchronous command finishes its execution. The default is off.
26092 @kindex show exec-done-display
26093 @item show exec-done-display
26094 Displays the current setting of asynchronous command completion
26097 @cindex ARM AArch64
26098 @item set debug aarch64
26099 Turns on or off display of debugging messages related to ARM AArch64.
26100 The default is off.
26102 @item show debug aarch64
26103 Displays the current state of displaying debugging messages related to
26105 @cindex gdbarch debugging info
26106 @cindex architecture debugging info
26107 @item set debug arch
26108 Turns on or off display of gdbarch debugging info. The default is off
26109 @item show debug arch
26110 Displays the current state of displaying gdbarch debugging info.
26111 @item set debug aix-solib
26112 @cindex AIX shared library debugging
26113 Control display of debugging messages from the AIX shared library
26114 support module. The default is off.
26115 @item show debug aix-thread
26116 Show the current state of displaying AIX shared library debugging messages.
26117 @item set debug aix-thread
26118 @cindex AIX threads
26119 Display debugging messages about inner workings of the AIX thread
26121 @item show debug aix-thread
26122 Show the current state of AIX thread debugging info display.
26123 @item set debug check-physname
26125 Check the results of the ``physname'' computation. When reading DWARF
26126 debugging information for C@t{++}, @value{GDBN} attempts to compute
26127 each entity's name. @value{GDBN} can do this computation in two
26128 different ways, depending on exactly what information is present.
26129 When enabled, this setting causes @value{GDBN} to compute the names
26130 both ways and display any discrepancies.
26131 @item show debug check-physname
26132 Show the current state of ``physname'' checking.
26133 @item set debug coff-pe-read
26134 @cindex COFF/PE exported symbols
26135 Control display of debugging messages related to reading of COFF/PE
26136 exported symbols. The default is off.
26137 @item show debug coff-pe-read
26138 Displays the current state of displaying debugging messages related to
26139 reading of COFF/PE exported symbols.
26140 @item set debug dwarf-die
26142 Dump DWARF DIEs after they are read in.
26143 The value is the number of nesting levels to print.
26144 A value of zero turns off the display.
26145 @item show debug dwarf-die
26146 Show the current state of DWARF DIE debugging.
26147 @item set debug dwarf-line
26148 @cindex DWARF Line Tables
26149 Turns on or off display of debugging messages related to reading
26150 DWARF line tables. The default is 0 (off).
26151 A value of 1 provides basic information.
26152 A value greater than 1 provides more verbose information.
26153 @item show debug dwarf-line
26154 Show the current state of DWARF line table debugging.
26155 @item set debug dwarf-read
26156 @cindex DWARF Reading
26157 Turns on or off display of debugging messages related to reading
26158 DWARF debug info. The default is 0 (off).
26159 A value of 1 provides basic information.
26160 A value greater than 1 provides more verbose information.
26161 @item show debug dwarf-read
26162 Show the current state of DWARF reader debugging.
26163 @item set debug displaced
26164 @cindex displaced stepping debugging info
26165 Turns on or off display of @value{GDBN} debugging info for the
26166 displaced stepping support. The default is off.
26167 @item show debug displaced
26168 Displays the current state of displaying @value{GDBN} debugging info
26169 related to displaced stepping.
26170 @item set debug event
26171 @cindex event debugging info
26172 Turns on or off display of @value{GDBN} event debugging info. The
26174 @item show debug event
26175 Displays the current state of displaying @value{GDBN} event debugging
26177 @item set debug expression
26178 @cindex expression debugging info
26179 Turns on or off display of debugging info about @value{GDBN}
26180 expression parsing. The default is off.
26181 @item show debug expression
26182 Displays the current state of displaying debugging info about
26183 @value{GDBN} expression parsing.
26184 @item set debug fbsd-lwp
26185 @cindex FreeBSD LWP debug messages
26186 Turns on or off debugging messages from the FreeBSD LWP debug support.
26187 @item show debug fbsd-lwp
26188 Show the current state of FreeBSD LWP debugging messages.
26189 @item set debug fbsd-nat
26190 @cindex FreeBSD native target debug messages
26191 Turns on or off debugging messages from the FreeBSD native target.
26192 @item show debug fbsd-nat
26193 Show the current state of FreeBSD native target debugging messages.
26194 @item set debug frame
26195 @cindex frame debugging info
26196 Turns on or off display of @value{GDBN} frame debugging info. The
26198 @item show debug frame
26199 Displays the current state of displaying @value{GDBN} frame debugging
26201 @item set debug gnu-nat
26202 @cindex @sc{gnu}/Hurd debug messages
26203 Turn on or off debugging messages from the @sc{gnu}/Hurd debug support.
26204 @item show debug gnu-nat
26205 Show the current state of @sc{gnu}/Hurd debugging messages.
26206 @item set debug infrun
26207 @cindex inferior debugging info
26208 Turns on or off display of @value{GDBN} debugging info for running the inferior.
26209 The default is off. @file{infrun.c} contains GDB's runtime state machine used
26210 for implementing operations such as single-stepping the inferior.
26211 @item show debug infrun
26212 Displays the current state of @value{GDBN} inferior debugging.
26213 @item set debug jit
26214 @cindex just-in-time compilation, debugging messages
26215 Turn on or off debugging messages from JIT debug support.
26216 @item show debug jit
26217 Displays the current state of @value{GDBN} JIT debugging.
26218 @item set debug lin-lwp
26219 @cindex @sc{gnu}/Linux LWP debug messages
26220 @cindex Linux lightweight processes
26221 Turn on or off debugging messages from the Linux LWP debug support.
26222 @item show debug lin-lwp
26223 Show the current state of Linux LWP debugging messages.
26224 @item set debug linux-namespaces
26225 @cindex @sc{gnu}/Linux namespaces debug messages
26226 Turn on or off debugging messages from the Linux namespaces debug support.
26227 @item show debug linux-namespaces
26228 Show the current state of Linux namespaces debugging messages.
26229 @item set debug mach-o
26230 @cindex Mach-O symbols processing
26231 Control display of debugging messages related to Mach-O symbols
26232 processing. The default is off.
26233 @item show debug mach-o
26234 Displays the current state of displaying debugging messages related to
26235 reading of COFF/PE exported symbols.
26236 @item set debug notification
26237 @cindex remote async notification debugging info
26238 Turn on or off debugging messages about remote async notification.
26239 The default is off.
26240 @item show debug notification
26241 Displays the current state of remote async notification debugging messages.
26242 @item set debug observer
26243 @cindex observer debugging info
26244 Turns on or off display of @value{GDBN} observer debugging. This
26245 includes info such as the notification of observable events.
26246 @item show debug observer
26247 Displays the current state of observer debugging.
26248 @item set debug overload
26249 @cindex C@t{++} overload debugging info
26250 Turns on or off display of @value{GDBN} C@t{++} overload debugging
26251 info. This includes info such as ranking of functions, etc. The default
26253 @item show debug overload
26254 Displays the current state of displaying @value{GDBN} C@t{++} overload
26256 @cindex expression parser, debugging info
26257 @cindex debug expression parser
26258 @item set debug parser
26259 Turns on or off the display of expression parser debugging output.
26260 Internally, this sets the @code{yydebug} variable in the expression
26261 parser. @xref{Tracing, , Tracing Your Parser, bison, Bison}, for
26262 details. The default is off.
26263 @item show debug parser
26264 Show the current state of expression parser debugging.
26265 @cindex packets, reporting on stdout
26266 @cindex serial connections, debugging
26267 @cindex debug remote protocol
26268 @cindex remote protocol debugging
26269 @cindex display remote packets
26270 @item set debug remote
26271 Turns on or off display of reports on all packets sent back and forth across
26272 the serial line to the remote machine. The info is printed on the
26273 @value{GDBN} standard output stream. The default is off.
26274 @item show debug remote
26275 Displays the state of display of remote packets.
26277 @item set debug remote-packet-max-chars
26278 Sets the maximum number of characters to display for each remote packet when
26279 @code{set debug remote} is on. This is useful to prevent @value{GDBN} from
26280 displaying lengthy remote packets and polluting the console.
26282 The default value is @code{512}, which means @value{GDBN} will truncate each
26283 remote packet after 512 bytes.
26285 Setting this option to @code{unlimited} will disable truncation and will output
26286 the full length of the remote packets.
26287 @item show debug remote-packet-max-chars
26288 Displays the number of bytes to output for remote packet debugging.
26290 @item set debug separate-debug-file
26291 Turns on or off display of debug output about separate debug file search.
26292 @item show debug separate-debug-file
26293 Displays the state of separate debug file search debug output.
26295 @item set debug serial
26296 Turns on or off display of @value{GDBN} serial debugging info. The
26298 @item show debug serial
26299 Displays the current state of displaying @value{GDBN} serial debugging
26301 @item set debug solib-frv
26302 @cindex FR-V shared-library debugging
26303 Turn on or off debugging messages for FR-V shared-library code.
26304 @item show debug solib-frv
26305 Display the current state of FR-V shared-library code debugging
26307 @item set debug symbol-lookup
26308 @cindex symbol lookup
26309 Turns on or off display of debugging messages related to symbol lookup.
26310 The default is 0 (off).
26311 A value of 1 provides basic information.
26312 A value greater than 1 provides more verbose information.
26313 @item show debug symbol-lookup
26314 Show the current state of symbol lookup debugging messages.
26315 @item set debug symfile
26316 @cindex symbol file functions
26317 Turns on or off display of debugging messages related to symbol file functions.
26318 The default is off. @xref{Files}.
26319 @item show debug symfile
26320 Show the current state of symbol file debugging messages.
26321 @item set debug symtab-create
26322 @cindex symbol table creation
26323 Turns on or off display of debugging messages related to symbol table creation.
26324 The default is 0 (off).
26325 A value of 1 provides basic information.
26326 A value greater than 1 provides more verbose information.
26327 @item show debug symtab-create
26328 Show the current state of symbol table creation debugging.
26329 @item set debug target
26330 @cindex target debugging info
26331 Turns on or off display of @value{GDBN} target debugging info. This info
26332 includes what is going on at the target level of GDB, as it happens. The
26333 default is 0. Set it to 1 to track events, and to 2 to also track the
26334 value of large memory transfers.
26335 @item show debug target
26336 Displays the current state of displaying @value{GDBN} target debugging
26338 @item set debug timestamp
26339 @cindex timestamping debugging info
26340 Turns on or off display of timestamps with @value{GDBN} debugging info.
26341 When enabled, seconds and microseconds are displayed before each debugging
26343 @item show debug timestamp
26344 Displays the current state of displaying timestamps with @value{GDBN}
26346 @item set debug varobj
26347 @cindex variable object debugging info
26348 Turns on or off display of @value{GDBN} variable object debugging
26349 info. The default is off.
26350 @item show debug varobj
26351 Displays the current state of displaying @value{GDBN} variable object
26353 @item set debug xml
26354 @cindex XML parser debugging
26355 Turn on or off debugging messages for built-in XML parsers.
26356 @item show debug xml
26357 Displays the current state of XML debugging messages.
26360 @node Other Misc Settings
26361 @section Other Miscellaneous Settings
26362 @cindex miscellaneous settings
26365 @kindex set interactive-mode
26366 @item set interactive-mode
26367 If @code{on}, forces @value{GDBN} to assume that GDB was started
26368 in a terminal. In practice, this means that @value{GDBN} should wait
26369 for the user to answer queries generated by commands entered at
26370 the command prompt. If @code{off}, forces @value{GDBN} to operate
26371 in the opposite mode, and it uses the default answers to all queries.
26372 If @code{auto} (the default), @value{GDBN} tries to determine whether
26373 its standard input is a terminal, and works in interactive-mode if it
26374 is, non-interactively otherwise.
26376 In the vast majority of cases, the debugger should be able to guess
26377 correctly which mode should be used. But this setting can be useful
26378 in certain specific cases, such as running a MinGW @value{GDBN}
26379 inside a cygwin window.
26381 @kindex show interactive-mode
26382 @item show interactive-mode
26383 Displays whether the debugger is operating in interactive mode or not.
26386 @node Extending GDB
26387 @chapter Extending @value{GDBN}
26388 @cindex extending GDB
26390 @value{GDBN} provides several mechanisms for extension.
26391 @value{GDBN} also provides the ability to automatically load
26392 extensions when it reads a file for debugging. This allows the
26393 user to automatically customize @value{GDBN} for the program
26397 * Sequences:: Canned Sequences of @value{GDBN} Commands
26398 * Python:: Extending @value{GDBN} using Python
26399 * Guile:: Extending @value{GDBN} using Guile
26400 * Auto-loading extensions:: Automatically loading extensions
26401 * Multiple Extension Languages:: Working with multiple extension languages
26402 * Aliases:: Creating new spellings of existing commands
26405 To facilitate the use of extension languages, @value{GDBN} is capable
26406 of evaluating the contents of a file. When doing so, @value{GDBN}
26407 can recognize which extension language is being used by looking at
26408 the filename extension. Files with an unrecognized filename extension
26409 are always treated as a @value{GDBN} Command Files.
26410 @xref{Command Files,, Command files}.
26412 You can control how @value{GDBN} evaluates these files with the following
26416 @kindex set script-extension
26417 @kindex show script-extension
26418 @item set script-extension off
26419 All scripts are always evaluated as @value{GDBN} Command Files.
26421 @item set script-extension soft
26422 The debugger determines the scripting language based on filename
26423 extension. If this scripting language is supported, @value{GDBN}
26424 evaluates the script using that language. Otherwise, it evaluates
26425 the file as a @value{GDBN} Command File.
26427 @item set script-extension strict
26428 The debugger determines the scripting language based on filename
26429 extension, and evaluates the script using that language. If the
26430 language is not supported, then the evaluation fails.
26432 @item show script-extension
26433 Display the current value of the @code{script-extension} option.
26437 @ifset SYSTEM_GDBINIT_DIR
26438 This setting is not used for files in the system-wide gdbinit directory.
26439 Files in that directory must have an extension matching their language,
26440 or have a @file{.gdb} extension to be interpreted as regular @value{GDBN}
26441 commands. @xref{Startup}.
26445 @section Canned Sequences of Commands
26447 Aside from breakpoint commands (@pxref{Break Commands, ,Breakpoint
26448 Command Lists}), @value{GDBN} provides two ways to store sequences of
26449 commands for execution as a unit: user-defined commands and command
26453 * Define:: How to define your own commands
26454 * Hooks:: Hooks for user-defined commands
26455 * Command Files:: How to write scripts of commands to be stored in a file
26456 * Output:: Commands for controlled output
26457 * Auto-loading sequences:: Controlling auto-loaded command files
26461 @subsection User-defined Commands
26463 @cindex user-defined command
26464 @cindex arguments, to user-defined commands
26465 A @dfn{user-defined command} is a sequence of @value{GDBN} commands to
26466 which you assign a new name as a command. This is done with the
26467 @code{define} command. User commands may accept an unlimited number of arguments
26468 separated by whitespace. Arguments are accessed within the user command
26469 via @code{$arg0@dots{}$argN}. A trivial example:
26473 print $arg0 + $arg1 + $arg2
26478 To execute the command use:
26485 This defines the command @code{adder}, which prints the sum of
26486 its three arguments. Note the arguments are text substitutions, so they may
26487 reference variables, use complex expressions, or even perform inferior
26490 @cindex argument count in user-defined commands
26491 @cindex how many arguments (user-defined commands)
26492 In addition, @code{$argc} may be used to find out how many arguments have
26498 print $arg0 + $arg1
26501 print $arg0 + $arg1 + $arg2
26506 Combining with the @code{eval} command (@pxref{eval}) makes it easier
26507 to process a variable number of arguments:
26514 eval "set $sum = $sum + $arg%d", $i
26524 @item define @var{commandname}
26525 Define a command named @var{commandname}. If there is already a command
26526 by that name, you are asked to confirm that you want to redefine it.
26527 The argument @var{commandname} may be a bare command name consisting of letters,
26528 numbers, dashes, and underscores. It may also start with any predefined
26529 prefix command. For example, @samp{define target my-target} creates
26530 a user-defined @samp{target my-target} command.
26532 The definition of the command is made up of other @value{GDBN} command lines,
26533 which are given following the @code{define} command. The end of these
26534 commands is marked by a line containing @code{end}.
26537 @kindex end@r{ (user-defined commands)}
26538 @item document @var{commandname}
26539 Document the user-defined command @var{commandname}, so that it can be
26540 accessed by @code{help}. The command @var{commandname} must already be
26541 defined. This command reads lines of documentation just as @code{define}
26542 reads the lines of the command definition, ending with @code{end}.
26543 After the @code{document} command is finished, @code{help} on command
26544 @var{commandname} displays the documentation you have written.
26546 You may use the @code{document} command again to change the
26547 documentation of a command. Redefining the command with @code{define}
26548 does not change the documentation.
26550 @kindex dont-repeat
26551 @cindex don't repeat command
26553 Used inside a user-defined command, this tells @value{GDBN} that this
26554 command should not be repeated when the user hits @key{RET}
26555 (@pxref{Command Syntax, repeat last command}).
26557 @kindex help user-defined
26558 @item help user-defined
26559 List all user-defined commands and all python commands defined in class
26560 COMMAND_USER. The first line of the documentation or docstring is
26565 @itemx show user @var{commandname}
26566 Display the @value{GDBN} commands used to define @var{commandname} (but
26567 not its documentation). If no @var{commandname} is given, display the
26568 definitions for all user-defined commands.
26569 This does not work for user-defined python commands.
26571 @cindex infinite recursion in user-defined commands
26572 @kindex show max-user-call-depth
26573 @kindex set max-user-call-depth
26574 @item show max-user-call-depth
26575 @itemx set max-user-call-depth
26576 The value of @code{max-user-call-depth} controls how many recursion
26577 levels are allowed in user-defined commands before @value{GDBN} suspects an
26578 infinite recursion and aborts the command.
26579 This does not apply to user-defined python commands.
26582 In addition to the above commands, user-defined commands frequently
26583 use control flow commands, described in @ref{Command Files}.
26585 When user-defined commands are executed, the
26586 commands of the definition are not printed. An error in any command
26587 stops execution of the user-defined command.
26589 If used interactively, commands that would ask for confirmation proceed
26590 without asking when used inside a user-defined command. Many @value{GDBN}
26591 commands that normally print messages to say what they are doing omit the
26592 messages when used in a user-defined command.
26595 @subsection User-defined Command Hooks
26596 @cindex command hooks
26597 @cindex hooks, for commands
26598 @cindex hooks, pre-command
26601 You may define @dfn{hooks}, which are a special kind of user-defined
26602 command. Whenever you run the command @samp{foo}, if the user-defined
26603 command @samp{hook-foo} exists, it is executed (with no arguments)
26604 before that command.
26606 @cindex hooks, post-command
26608 A hook may also be defined which is run after the command you executed.
26609 Whenever you run the command @samp{foo}, if the user-defined command
26610 @samp{hookpost-foo} exists, it is executed (with no arguments) after
26611 that command. Post-execution hooks may exist simultaneously with
26612 pre-execution hooks, for the same command.
26614 It is valid for a hook to call the command which it hooks. If this
26615 occurs, the hook is not re-executed, thereby avoiding infinite recursion.
26617 @c It would be nice if hookpost could be passed a parameter indicating
26618 @c if the command it hooks executed properly or not. FIXME!
26620 @kindex stop@r{, a pseudo-command}
26621 In addition, a pseudo-command, @samp{stop} exists. Defining
26622 (@samp{hook-stop}) makes the associated commands execute every time
26623 execution stops in your program: before breakpoint commands are run,
26624 displays are printed, or the stack frame is printed.
26626 For example, to ignore @code{SIGALRM} signals while
26627 single-stepping, but treat them normally during normal execution,
26632 handle SIGALRM nopass
26636 handle SIGALRM pass
26639 define hook-continue
26640 handle SIGALRM pass
26644 As a further example, to hook at the beginning and end of the @code{echo}
26645 command, and to add extra text to the beginning and end of the message,
26653 define hookpost-echo
26657 (@value{GDBP}) echo Hello World
26658 <<<---Hello World--->>>
26663 You can define a hook for any single-word command in @value{GDBN}, but
26664 not for command aliases; you should define a hook for the basic command
26665 name, e.g.@: @code{backtrace} rather than @code{bt}.
26666 @c FIXME! So how does Joe User discover whether a command is an alias
26668 You can hook a multi-word command by adding @code{hook-} or
26669 @code{hookpost-} to the last word of the command, e.g.@:
26670 @samp{define target hook-remote} to add a hook to @samp{target remote}.
26672 If an error occurs during the execution of your hook, execution of
26673 @value{GDBN} commands stops and @value{GDBN} issues a prompt
26674 (before the command that you actually typed had a chance to run).
26676 If you try to define a hook which does not match any known command, you
26677 get a warning from the @code{define} command.
26679 @node Command Files
26680 @subsection Command Files
26682 @cindex command files
26683 @cindex scripting commands
26684 A command file for @value{GDBN} is a text file made of lines that are
26685 @value{GDBN} commands. Comments (lines starting with @kbd{#}) may
26686 also be included. An empty line in a command file does nothing; it
26687 does not mean to repeat the last command, as it would from the
26690 You can request the execution of a command file with the @code{source}
26691 command. Note that the @code{source} command is also used to evaluate
26692 scripts that are not Command Files. The exact behavior can be configured
26693 using the @code{script-extension} setting.
26694 @xref{Extending GDB,, Extending GDB}.
26698 @cindex execute commands from a file
26699 @item source [-s] [-v] @var{filename}
26700 Execute the command file @var{filename}.
26703 The lines in a command file are generally executed sequentially,
26704 unless the order of execution is changed by one of the
26705 @emph{flow-control commands} described below. The commands are not
26706 printed as they are executed. An error in any command terminates
26707 execution of the command file and control is returned to the console.
26709 @value{GDBN} first searches for @var{filename} in the current directory.
26710 If the file is not found there, and @var{filename} does not specify a
26711 directory, then @value{GDBN} also looks for the file on the source search path
26712 (specified with the @samp{directory} command);
26713 except that @file{$cdir} is not searched because the compilation directory
26714 is not relevant to scripts.
26716 If @code{-s} is specified, then @value{GDBN} searches for @var{filename}
26717 on the search path even if @var{filename} specifies a directory.
26718 The search is done by appending @var{filename} to each element of the
26719 search path. So, for example, if @var{filename} is @file{mylib/myscript}
26720 and the search path contains @file{/home/user} then @value{GDBN} will
26721 look for the script @file{/home/user/mylib/myscript}.
26722 The search is also done if @var{filename} is an absolute path.
26723 For example, if @var{filename} is @file{/tmp/myscript} and
26724 the search path contains @file{/home/user} then @value{GDBN} will
26725 look for the script @file{/home/user/tmp/myscript}.
26726 For DOS-like systems, if @var{filename} contains a drive specification,
26727 it is stripped before concatenation. For example, if @var{filename} is
26728 @file{d:myscript} and the search path contains @file{c:/tmp} then @value{GDBN}
26729 will look for the script @file{c:/tmp/myscript}.
26731 If @code{-v}, for verbose mode, is given then @value{GDBN} displays
26732 each command as it is executed. The option must be given before
26733 @var{filename}, and is interpreted as part of the filename anywhere else.
26735 Commands that would ask for confirmation if used interactively proceed
26736 without asking when used in a command file. Many @value{GDBN} commands that
26737 normally print messages to say what they are doing omit the messages
26738 when called from command files.
26740 @value{GDBN} also accepts command input from standard input. In this
26741 mode, normal output goes to standard output and error output goes to
26742 standard error. Errors in a command file supplied on standard input do
26743 not terminate execution of the command file---execution continues with
26747 gdb < cmds > log 2>&1
26750 (The syntax above will vary depending on the shell used.) This example
26751 will execute commands from the file @file{cmds}. All output and errors
26752 would be directed to @file{log}.
26754 Since commands stored on command files tend to be more general than
26755 commands typed interactively, they frequently need to deal with
26756 complicated situations, such as different or unexpected values of
26757 variables and symbols, changes in how the program being debugged is
26758 built, etc. @value{GDBN} provides a set of flow-control commands to
26759 deal with these complexities. Using these commands, you can write
26760 complex scripts that loop over data structures, execute commands
26761 conditionally, etc.
26768 This command allows to include in your script conditionally executed
26769 commands. The @code{if} command takes a single argument, which is an
26770 expression to evaluate. It is followed by a series of commands that
26771 are executed only if the expression is true (its value is nonzero).
26772 There can then optionally be an @code{else} line, followed by a series
26773 of commands that are only executed if the expression was false. The
26774 end of the list is marked by a line containing @code{end}.
26778 This command allows to write loops. Its syntax is similar to
26779 @code{if}: the command takes a single argument, which is an expression
26780 to evaluate, and must be followed by the commands to execute, one per
26781 line, terminated by an @code{end}. These commands are called the
26782 @dfn{body} of the loop. The commands in the body of @code{while} are
26783 executed repeatedly as long as the expression evaluates to true.
26787 This command exits the @code{while} loop in whose body it is included.
26788 Execution of the script continues after that @code{while}s @code{end}
26791 @kindex loop_continue
26792 @item loop_continue
26793 This command skips the execution of the rest of the body of commands
26794 in the @code{while} loop in whose body it is included. Execution
26795 branches to the beginning of the @code{while} loop, where it evaluates
26796 the controlling expression.
26798 @kindex end@r{ (if/else/while commands)}
26800 Terminate the block of commands that are the body of @code{if},
26801 @code{else}, or @code{while} flow-control commands.
26806 @subsection Commands for Controlled Output
26808 During the execution of a command file or a user-defined command, normal
26809 @value{GDBN} output is suppressed; the only output that appears is what is
26810 explicitly printed by the commands in the definition. This section
26811 describes three commands useful for generating exactly the output you
26816 @item echo @var{text}
26817 @c I do not consider backslash-space a standard C escape sequence
26818 @c because it is not in ANSI.
26819 Print @var{text}. Nonprinting characters can be included in
26820 @var{text} using C escape sequences, such as @samp{\n} to print a
26821 newline. @strong{No newline is printed unless you specify one.}
26822 In addition to the standard C escape sequences, a backslash followed
26823 by a space stands for a space. This is useful for displaying a
26824 string with spaces at the beginning or the end, since leading and
26825 trailing spaces are otherwise trimmed from all arguments.
26826 To print @samp{@w{ }and foo =@w{ }}, use the command
26827 @samp{echo \@w{ }and foo = \@w{ }}.
26829 A backslash at the end of @var{text} can be used, as in C, to continue
26830 the command onto subsequent lines. For example,
26833 echo This is some text\n\
26834 which is continued\n\
26835 onto several lines.\n
26838 produces the same output as
26841 echo This is some text\n
26842 echo which is continued\n
26843 echo onto several lines.\n
26847 @item output @var{expression}
26848 Print the value of @var{expression} and nothing but that value: no
26849 newlines, no @samp{$@var{nn} = }. The value is not entered in the
26850 value history either. @xref{Expressions, ,Expressions}, for more information
26853 @item output/@var{fmt} @var{expression}
26854 Print the value of @var{expression} in format @var{fmt}. You can use
26855 the same formats as for @code{print}. @xref{Output Formats,,Output
26856 Formats}, for more information.
26859 @item printf @var{template}, @var{expressions}@dots{}
26860 Print the values of one or more @var{expressions} under the control of
26861 the string @var{template}. To print several values, make
26862 @var{expressions} be a comma-separated list of individual expressions,
26863 which may be either numbers or pointers. Their values are printed as
26864 specified by @var{template}, exactly as a C program would do by
26865 executing the code below:
26868 printf (@var{template}, @var{expressions}@dots{});
26871 As in @code{C} @code{printf}, ordinary characters in @var{template}
26872 are printed verbatim, while @dfn{conversion specification} introduced
26873 by the @samp{%} character cause subsequent @var{expressions} to be
26874 evaluated, their values converted and formatted according to type and
26875 style information encoded in the conversion specifications, and then
26878 For example, you can print two values in hex like this:
26881 printf "foo, bar-foo = 0x%x, 0x%x\n", foo, bar-foo
26884 @code{printf} supports all the standard @code{C} conversion
26885 specifications, including the flags and modifiers between the @samp{%}
26886 character and the conversion letter, with the following exceptions:
26890 The argument-ordering modifiers, such as @samp{2$}, are not supported.
26893 The modifier @samp{*} is not supported for specifying precision or
26897 The @samp{'} flag (for separation of digits into groups according to
26898 @code{LC_NUMERIC'}) is not supported.
26901 The type modifiers @samp{hh}, @samp{j}, @samp{t}, and @samp{z} are not
26905 The conversion letter @samp{n} (as in @samp{%n}) is not supported.
26908 The conversion letters @samp{a} and @samp{A} are not supported.
26912 Note that the @samp{ll} type modifier is supported only if the
26913 underlying @code{C} implementation used to build @value{GDBN} supports
26914 the @code{long long int} type, and the @samp{L} type modifier is
26915 supported only if @code{long double} type is available.
26917 As in @code{C}, @code{printf} supports simple backslash-escape
26918 sequences, such as @code{\n}, @samp{\t}, @samp{\\}, @samp{\"},
26919 @samp{\a}, and @samp{\f}, that consist of backslash followed by a
26920 single character. Octal and hexadecimal escape sequences are not
26923 Additionally, @code{printf} supports conversion specifications for DFP
26924 (@dfn{Decimal Floating Point}) types using the following length modifiers
26925 together with a floating point specifier.
26930 @samp{H} for printing @code{Decimal32} types.
26933 @samp{D} for printing @code{Decimal64} types.
26936 @samp{DD} for printing @code{Decimal128} types.
26939 If the underlying @code{C} implementation used to build @value{GDBN} has
26940 support for the three length modifiers for DFP types, other modifiers
26941 such as width and precision will also be available for @value{GDBN} to use.
26943 In case there is no such @code{C} support, no additional modifiers will be
26944 available and the value will be printed in the standard way.
26946 Here's an example of printing DFP types using the above conversion letters:
26948 printf "D32: %Hf - D64: %Df - D128: %DDf\n",1.2345df,1.2E10dd,1.2E1dl
26953 @item eval @var{template}, @var{expressions}@dots{}
26954 Convert the values of one or more @var{expressions} under the control of
26955 the string @var{template} to a command line, and call it.
26959 @node Auto-loading sequences
26960 @subsection Controlling auto-loading native @value{GDBN} scripts
26961 @cindex native script auto-loading
26963 When a new object file is read (for example, due to the @code{file}
26964 command, or because the inferior has loaded a shared library),
26965 @value{GDBN} will look for the command file @file{@var{objfile}-gdb.gdb}.
26966 @xref{Auto-loading extensions}.
26968 Auto-loading can be enabled or disabled,
26969 and the list of auto-loaded scripts can be printed.
26972 @anchor{set auto-load gdb-scripts}
26973 @kindex set auto-load gdb-scripts
26974 @item set auto-load gdb-scripts [on|off]
26975 Enable or disable the auto-loading of canned sequences of commands scripts.
26977 @anchor{show auto-load gdb-scripts}
26978 @kindex show auto-load gdb-scripts
26979 @item show auto-load gdb-scripts
26980 Show whether auto-loading of canned sequences of commands scripts is enabled or
26983 @anchor{info auto-load gdb-scripts}
26984 @kindex info auto-load gdb-scripts
26985 @cindex print list of auto-loaded canned sequences of commands scripts
26986 @item info auto-load gdb-scripts [@var{regexp}]
26987 Print the list of all canned sequences of commands scripts that @value{GDBN}
26991 If @var{regexp} is supplied only canned sequences of commands scripts with
26992 matching names are printed.
26994 @c Python docs live in a separate file.
26995 @include python.texi
26997 @c Guile docs live in a separate file.
26998 @include guile.texi
27000 @node Auto-loading extensions
27001 @section Auto-loading extensions
27002 @cindex auto-loading extensions
27004 @value{GDBN} provides two mechanisms for automatically loading extensions
27005 when a new object file is read (for example, due to the @code{file}
27006 command, or because the inferior has loaded a shared library):
27007 @file{@var{objfile}-gdb.@var{ext}} and the @code{.debug_gdb_scripts}
27008 section of modern file formats like ELF.
27011 * objfile-gdb.ext file: objfile-gdbdotext file. The @file{@var{objfile}-gdb.@var{ext}} file
27012 * .debug_gdb_scripts section: dotdebug_gdb_scripts section. The @code{.debug_gdb_scripts} section
27013 * Which flavor to choose?::
27016 The auto-loading feature is useful for supplying application-specific
27017 debugging commands and features.
27019 Auto-loading can be enabled or disabled,
27020 and the list of auto-loaded scripts can be printed.
27021 See the @samp{auto-loading} section of each extension language
27022 for more information.
27023 For @value{GDBN} command files see @ref{Auto-loading sequences}.
27024 For Python files see @ref{Python Auto-loading}.
27026 Note that loading of this script file also requires accordingly configured
27027 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
27029 @node objfile-gdbdotext file
27030 @subsection The @file{@var{objfile}-gdb.@var{ext}} file
27031 @cindex @file{@var{objfile}-gdb.gdb}
27032 @cindex @file{@var{objfile}-gdb.py}
27033 @cindex @file{@var{objfile}-gdb.scm}
27035 When a new object file is read, @value{GDBN} looks for a file named
27036 @file{@var{objfile}-gdb.@var{ext}} (we call it @var{script-name} below),
27037 where @var{objfile} is the object file's name and
27038 where @var{ext} is the file extension for the extension language:
27041 @item @file{@var{objfile}-gdb.gdb}
27042 GDB's own command language
27043 @item @file{@var{objfile}-gdb.py}
27045 @item @file{@var{objfile}-gdb.scm}
27049 @var{script-name} is formed by ensuring that the file name of @var{objfile}
27050 is absolute, following all symlinks, and resolving @code{.} and @code{..}
27051 components, and appending the @file{-gdb.@var{ext}} suffix.
27052 If this file exists and is readable, @value{GDBN} will evaluate it as a
27053 script in the specified extension language.
27055 If this file does not exist, then @value{GDBN} will look for
27056 @var{script-name} file in all of the directories as specified below.
27058 Note that loading of these files requires an accordingly configured
27059 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
27061 For object files using @file{.exe} suffix @value{GDBN} tries to load first the
27062 scripts normally according to its @file{.exe} filename. But if no scripts are
27063 found @value{GDBN} also tries script filenames matching the object file without
27064 its @file{.exe} suffix. This @file{.exe} stripping is case insensitive and it
27065 is attempted on any platform. This makes the script filenames compatible
27066 between Unix and MS-Windows hosts.
27069 @anchor{set auto-load scripts-directory}
27070 @kindex set auto-load scripts-directory
27071 @item set auto-load scripts-directory @r{[}@var{directories}@r{]}
27072 Control @value{GDBN} auto-loaded scripts location. Multiple directory entries
27073 may be delimited by the host platform path separator in use
27074 (@samp{:} on Unix, @samp{;} on MS-Windows and MS-DOS).
27076 Each entry here needs to be covered also by the security setting
27077 @code{set auto-load safe-path} (@pxref{set auto-load safe-path}).
27079 @anchor{with-auto-load-dir}
27080 This variable defaults to @file{$debugdir:$datadir/auto-load}. The default
27081 @code{set auto-load safe-path} value can be also overriden by @value{GDBN}
27082 configuration option @option{--with-auto-load-dir}.
27084 Any reference to @file{$debugdir} will get replaced by
27085 @var{debug-file-directory} value (@pxref{Separate Debug Files}) and any
27086 reference to @file{$datadir} will get replaced by @var{data-directory} which is
27087 determined at @value{GDBN} startup (@pxref{Data Files}). @file{$debugdir} and
27088 @file{$datadir} must be placed as a directory component --- either alone or
27089 delimited by @file{/} or @file{\} directory separators, depending on the host
27092 The list of directories uses path separator (@samp{:} on GNU and Unix
27093 systems, @samp{;} on MS-Windows and MS-DOS) to separate directories, similarly
27094 to the @env{PATH} environment variable.
27096 @anchor{show auto-load scripts-directory}
27097 @kindex show auto-load scripts-directory
27098 @item show auto-load scripts-directory
27099 Show @value{GDBN} auto-loaded scripts location.
27101 @anchor{add-auto-load-scripts-directory}
27102 @kindex add-auto-load-scripts-directory
27103 @item add-auto-load-scripts-directory @r{[}@var{directories}@dots{}@r{]}
27104 Add an entry (or list of entries) to the list of auto-loaded scripts locations.
27105 Multiple entries may be delimited by the host platform path separator in use.
27108 @value{GDBN} does not track which files it has already auto-loaded this way.
27109 @value{GDBN} will load the associated script every time the corresponding
27110 @var{objfile} is opened.
27111 So your @file{-gdb.@var{ext}} file should be careful to avoid errors if it
27112 is evaluated more than once.
27114 @node dotdebug_gdb_scripts section
27115 @subsection The @code{.debug_gdb_scripts} section
27116 @cindex @code{.debug_gdb_scripts} section
27118 For systems using file formats like ELF and COFF,
27119 when @value{GDBN} loads a new object file
27120 it will look for a special section named @code{.debug_gdb_scripts}.
27121 If this section exists, its contents is a list of null-terminated entries
27122 specifying scripts to load. Each entry begins with a non-null prefix byte that
27123 specifies the kind of entry, typically the extension language and whether the
27124 script is in a file or inlined in @code{.debug_gdb_scripts}.
27126 The following entries are supported:
27129 @item SECTION_SCRIPT_ID_PYTHON_FILE = 1
27130 @item SECTION_SCRIPT_ID_SCHEME_FILE = 3
27131 @item SECTION_SCRIPT_ID_PYTHON_TEXT = 4
27132 @item SECTION_SCRIPT_ID_SCHEME_TEXT = 6
27135 @subsubsection Script File Entries
27137 If the entry specifies a file, @value{GDBN} will look for the file first
27138 in the current directory and then along the source search path
27139 (@pxref{Source Path, ,Specifying Source Directories}),
27140 except that @file{$cdir} is not searched, since the compilation
27141 directory is not relevant to scripts.
27143 File entries can be placed in section @code{.debug_gdb_scripts} with,
27144 for example, this GCC macro for Python scripts.
27147 /* Note: The "MS" section flags are to remove duplicates. */
27148 #define DEFINE_GDB_PY_SCRIPT(script_name) \
27150 .pushsection \".debug_gdb_scripts\", \"MS\",@@progbits,1\n\
27151 .byte 1 /* Python */\n\
27152 .asciz \"" script_name "\"\n\
27158 For Guile scripts, replace @code{.byte 1} with @code{.byte 3}.
27159 Then one can reference the macro in a header or source file like this:
27162 DEFINE_GDB_PY_SCRIPT ("my-app-scripts.py")
27165 The script name may include directories if desired.
27167 Note that loading of this script file also requires accordingly configured
27168 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
27170 If the macro invocation is put in a header, any application or library
27171 using this header will get a reference to the specified script,
27172 and with the use of @code{"MS"} attributes on the section, the linker
27173 will remove duplicates.
27175 @subsubsection Script Text Entries
27177 Script text entries allow to put the executable script in the entry
27178 itself instead of loading it from a file.
27179 The first line of the entry, everything after the prefix byte and up to
27180 the first newline (@code{0xa}) character, is the script name, and must not
27181 contain any kind of space character, e.g., spaces or tabs.
27182 The rest of the entry, up to the trailing null byte, is the script to
27183 execute in the specified language. The name needs to be unique among
27184 all script names, as @value{GDBN} executes each script only once based
27187 Here is an example from file @file{py-section-script.c} in the @value{GDBN}
27191 #include "symcat.h"
27192 #include "gdb/section-scripts.h"
27194 ".pushsection \".debug_gdb_scripts\", \"MS\",@@progbits,1\n"
27195 ".byte " XSTRING (SECTION_SCRIPT_ID_PYTHON_TEXT) "\n"
27196 ".ascii \"gdb.inlined-script\\n\"\n"
27197 ".ascii \"class test_cmd (gdb.Command):\\n\"\n"
27198 ".ascii \" def __init__ (self):\\n\"\n"
27199 ".ascii \" super (test_cmd, self).__init__ ("
27200 "\\\"test-cmd\\\", gdb.COMMAND_OBSCURE)\\n\"\n"
27201 ".ascii \" def invoke (self, arg, from_tty):\\n\"\n"
27202 ".ascii \" print (\\\"test-cmd output, arg = %s\\\" % arg)\\n\"\n"
27203 ".ascii \"test_cmd ()\\n\"\n"
27209 Loading of inlined scripts requires a properly configured
27210 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
27211 The path to specify in @code{auto-load safe-path} is the path of the file
27212 containing the @code{.debug_gdb_scripts} section.
27214 @node Which flavor to choose?
27215 @subsection Which flavor to choose?
27217 Given the multiple ways of auto-loading extensions, it might not always
27218 be clear which one to choose. This section provides some guidance.
27221 Benefits of the @file{-gdb.@var{ext}} way:
27225 Can be used with file formats that don't support multiple sections.
27228 Ease of finding scripts for public libraries.
27230 Scripts specified in the @code{.debug_gdb_scripts} section are searched for
27231 in the source search path.
27232 For publicly installed libraries, e.g., @file{libstdc++}, there typically
27233 isn't a source directory in which to find the script.
27236 Doesn't require source code additions.
27240 Benefits of the @code{.debug_gdb_scripts} way:
27244 Works with static linking.
27246 Scripts for libraries done the @file{-gdb.@var{ext}} way require an objfile to
27247 trigger their loading. When an application is statically linked the only
27248 objfile available is the executable, and it is cumbersome to attach all the
27249 scripts from all the input libraries to the executable's
27250 @file{-gdb.@var{ext}} script.
27253 Works with classes that are entirely inlined.
27255 Some classes can be entirely inlined, and thus there may not be an associated
27256 shared library to attach a @file{-gdb.@var{ext}} script to.
27259 Scripts needn't be copied out of the source tree.
27261 In some circumstances, apps can be built out of large collections of internal
27262 libraries, and the build infrastructure necessary to install the
27263 @file{-gdb.@var{ext}} scripts in a place where @value{GDBN} can find them is
27264 cumbersome. It may be easier to specify the scripts in the
27265 @code{.debug_gdb_scripts} section as relative paths, and add a path to the
27266 top of the source tree to the source search path.
27269 @node Multiple Extension Languages
27270 @section Multiple Extension Languages
27272 The Guile and Python extension languages do not share any state,
27273 and generally do not interfere with each other.
27274 There are some things to be aware of, however.
27276 @subsection Python comes first
27278 Python was @value{GDBN}'s first extension language, and to avoid breaking
27279 existing behaviour Python comes first. This is generally solved by the
27280 ``first one wins'' principle. @value{GDBN} maintains a list of enabled
27281 extension languages, and when it makes a call to an extension language,
27282 (say to pretty-print a value), it tries each in turn until an extension
27283 language indicates it has performed the request (e.g., has returned the
27284 pretty-printed form of a value).
27285 This extends to errors while performing such requests: If an error happens
27286 while, for example, trying to pretty-print an object then the error is
27287 reported and any following extension languages are not tried.
27290 @section Creating new spellings of existing commands
27291 @cindex aliases for commands
27293 It is often useful to define alternate spellings of existing commands.
27294 For example, if a new @value{GDBN} command defined in Python has
27295 a long name to type, it is handy to have an abbreviated version of it
27296 that involves less typing.
27298 @value{GDBN} itself uses aliases. For example @samp{s} is an alias
27299 of the @samp{step} command even though it is otherwise an ambiguous
27300 abbreviation of other commands like @samp{set} and @samp{show}.
27302 Aliases are also used to provide shortened or more common versions
27303 of multi-word commands. For example, @value{GDBN} provides the
27304 @samp{tty} alias of the @samp{set inferior-tty} command.
27306 You can define a new alias with the @samp{alias} command.
27311 @item alias [-a] [--] @var{ALIAS} = @var{COMMAND}
27315 @var{ALIAS} specifies the name of the new alias.
27316 Each word of @var{ALIAS} must consist of letters, numbers, dashes and
27319 @var{COMMAND} specifies the name of an existing command
27320 that is being aliased.
27322 The @samp{-a} option specifies that the new alias is an abbreviation
27323 of the command. Abbreviations are not shown in command
27324 lists displayed by the @samp{help} command.
27326 The @samp{--} option specifies the end of options,
27327 and is useful when @var{ALIAS} begins with a dash.
27329 Here is a simple example showing how to make an abbreviation
27330 of a command so that there is less to type.
27331 Suppose you were tired of typing @samp{disas}, the current
27332 shortest unambiguous abbreviation of the @samp{disassemble} command
27333 and you wanted an even shorter version named @samp{di}.
27334 The following will accomplish this.
27337 (gdb) alias -a di = disas
27340 Note that aliases are different from user-defined commands.
27341 With a user-defined command, you also need to write documentation
27342 for it with the @samp{document} command.
27343 An alias automatically picks up the documentation of the existing command.
27345 Here is an example where we make @samp{elms} an abbreviation of
27346 @samp{elements} in the @samp{set print elements} command.
27347 This is to show that you can make an abbreviation of any part
27351 (gdb) alias -a set print elms = set print elements
27352 (gdb) alias -a show print elms = show print elements
27353 (gdb) set p elms 20
27355 Limit on string chars or array elements to print is 200.
27358 Note that if you are defining an alias of a @samp{set} command,
27359 and you want to have an alias for the corresponding @samp{show}
27360 command, then you need to define the latter separately.
27362 Unambiguously abbreviated commands are allowed in @var{COMMAND} and
27363 @var{ALIAS}, just as they are normally.
27366 (gdb) alias -a set pr elms = set p ele
27369 Finally, here is an example showing the creation of a one word
27370 alias for a more complex command.
27371 This creates alias @samp{spe} of the command @samp{set print elements}.
27374 (gdb) alias spe = set print elements
27379 @chapter Command Interpreters
27380 @cindex command interpreters
27382 @value{GDBN} supports multiple command interpreters, and some command
27383 infrastructure to allow users or user interface writers to switch
27384 between interpreters or run commands in other interpreters.
27386 @value{GDBN} currently supports two command interpreters, the console
27387 interpreter (sometimes called the command-line interpreter or @sc{cli})
27388 and the machine interface interpreter (or @sc{gdb/mi}). This manual
27389 describes both of these interfaces in great detail.
27391 By default, @value{GDBN} will start with the console interpreter.
27392 However, the user may choose to start @value{GDBN} with another
27393 interpreter by specifying the @option{-i} or @option{--interpreter}
27394 startup options. Defined interpreters include:
27398 @cindex console interpreter
27399 The traditional console or command-line interpreter. This is the most often
27400 used interpreter with @value{GDBN}. With no interpreter specified at runtime,
27401 @value{GDBN} will use this interpreter.
27404 @cindex mi interpreter
27405 The newest @sc{gdb/mi} interface (currently @code{mi3}). Used primarily
27406 by programs wishing to use @value{GDBN} as a backend for a debugger GUI
27407 or an IDE. For more information, see @ref{GDB/MI, ,The @sc{gdb/mi}
27411 @cindex mi3 interpreter
27412 The @sc{gdb/mi} interface introduced in @value{GDBN} 9.1.
27415 @cindex mi2 interpreter
27416 The @sc{gdb/mi} interface introduced in @value{GDBN} 6.0.
27419 @cindex mi1 interpreter
27420 The @sc{gdb/mi} interface introduced in @value{GDBN} 5.1.
27424 @cindex invoke another interpreter
27426 @kindex interpreter-exec
27427 You may execute commands in any interpreter from the current
27428 interpreter using the appropriate command. If you are running the
27429 console interpreter, simply use the @code{interpreter-exec} command:
27432 interpreter-exec mi "-data-list-register-names"
27435 @sc{gdb/mi} has a similar command, although it is only available in versions of
27436 @value{GDBN} which support @sc{gdb/mi} version 2 (or greater).
27438 Note that @code{interpreter-exec} only changes the interpreter for the
27439 duration of the specified command. It does not change the interpreter
27442 @cindex start a new independent interpreter
27444 Although you may only choose a single interpreter at startup, it is
27445 possible to run an independent interpreter on a specified input/output
27446 device (usually a tty).
27448 For example, consider a debugger GUI or IDE that wants to provide a
27449 @value{GDBN} console view. It may do so by embedding a terminal
27450 emulator widget in its GUI, starting @value{GDBN} in the traditional
27451 command-line mode with stdin/stdout/stderr redirected to that
27452 terminal, and then creating an MI interpreter running on a specified
27453 input/output device. The console interpreter created by @value{GDBN}
27454 at startup handles commands the user types in the terminal widget,
27455 while the GUI controls and synchronizes state with @value{GDBN} using
27456 the separate MI interpreter.
27458 To start a new secondary @dfn{user interface} running MI, use the
27459 @code{new-ui} command:
27462 @cindex new user interface
27464 new-ui @var{interpreter} @var{tty}
27467 The @var{interpreter} parameter specifies the interpreter to run.
27468 This accepts the same values as the @code{interpreter-exec} command.
27469 For example, @samp{console}, @samp{mi}, @samp{mi2}, etc. The
27470 @var{tty} parameter specifies the name of the bidirectional file the
27471 interpreter uses for input/output, usually the name of a
27472 pseudoterminal slave on Unix systems. For example:
27475 (@value{GDBP}) new-ui mi /dev/pts/9
27479 runs an MI interpreter on @file{/dev/pts/9}.
27482 @chapter @value{GDBN} Text User Interface
27484 @cindex Text User Interface
27487 * TUI Overview:: TUI overview
27488 * TUI Keys:: TUI key bindings
27489 * TUI Single Key Mode:: TUI single key mode
27490 * TUI Commands:: TUI-specific commands
27491 * TUI Configuration:: TUI configuration variables
27494 The @value{GDBN} Text User Interface (TUI) is a terminal
27495 interface which uses the @code{curses} library to show the source
27496 file, the assembly output, the program registers and @value{GDBN}
27497 commands in separate text windows. The TUI mode is supported only
27498 on platforms where a suitable version of the @code{curses} library
27501 The TUI mode is enabled by default when you invoke @value{GDBN} as
27502 @samp{@value{GDBP} -tui}.
27503 You can also switch in and out of TUI mode while @value{GDBN} runs by
27504 using various TUI commands and key bindings, such as @command{tui
27505 enable} or @kbd{C-x C-a}. @xref{TUI Commands, ,TUI Commands}, and
27506 @ref{TUI Keys, ,TUI Key Bindings}.
27509 @section TUI Overview
27511 In TUI mode, @value{GDBN} can display several text windows:
27515 This window is the @value{GDBN} command window with the @value{GDBN}
27516 prompt and the @value{GDBN} output. The @value{GDBN} input is still
27517 managed using readline.
27520 The source window shows the source file of the program. The current
27521 line and active breakpoints are displayed in this window.
27524 The assembly window shows the disassembly output of the program.
27527 This window shows the processor registers. Registers are highlighted
27528 when their values change.
27531 The source and assembly windows show the current program position
27532 by highlighting the current line and marking it with a @samp{>} marker.
27533 Breakpoints are indicated with two markers. The first marker
27534 indicates the breakpoint type:
27538 Breakpoint which was hit at least once.
27541 Breakpoint which was never hit.
27544 Hardware breakpoint which was hit at least once.
27547 Hardware breakpoint which was never hit.
27550 The second marker indicates whether the breakpoint is enabled or not:
27554 Breakpoint is enabled.
27557 Breakpoint is disabled.
27560 The source, assembly and register windows are updated when the current
27561 thread changes, when the frame changes, or when the program counter
27564 These windows are not all visible at the same time. The command
27565 window is always visible. The others can be arranged in several
27576 source and assembly,
27579 source and registers, or
27582 assembly and registers.
27585 A status line above the command window shows the following information:
27589 Indicates the current @value{GDBN} target.
27590 (@pxref{Targets, ,Specifying a Debugging Target}).
27593 Gives the current process or thread number.
27594 When no process is being debugged, this field is set to @code{No process}.
27597 Gives the current function name for the selected frame.
27598 The name is demangled if demangling is turned on (@pxref{Print Settings}).
27599 When there is no symbol corresponding to the current program counter,
27600 the string @code{??} is displayed.
27603 Indicates the current line number for the selected frame.
27604 When the current line number is not known, the string @code{??} is displayed.
27607 Indicates the current program counter address.
27611 @section TUI Key Bindings
27612 @cindex TUI key bindings
27614 The TUI installs several key bindings in the readline keymaps
27615 @ifset SYSTEM_READLINE
27616 (@pxref{Command Line Editing, , , rluserman, GNU Readline Library}).
27618 @ifclear SYSTEM_READLINE
27619 (@pxref{Command Line Editing}).
27621 The following key bindings are installed for both TUI mode and the
27622 @value{GDBN} standard mode.
27631 Enter or leave the TUI mode. When leaving the TUI mode,
27632 the curses window management stops and @value{GDBN} operates using
27633 its standard mode, writing on the terminal directly. When reentering
27634 the TUI mode, control is given back to the curses windows.
27635 The screen is then refreshed.
27637 This key binding uses the bindable Readline function
27638 @code{tui-switch-mode}.
27642 Use a TUI layout with only one window. The layout will
27643 either be @samp{source} or @samp{assembly}. When the TUI mode
27644 is not active, it will switch to the TUI mode.
27646 Think of this key binding as the Emacs @kbd{C-x 1} binding.
27648 This key binding uses the bindable Readline function
27649 @code{tui-delete-other-windows}.
27653 Use a TUI layout with at least two windows. When the current
27654 layout already has two windows, the next layout with two windows is used.
27655 When a new layout is chosen, one window will always be common to the
27656 previous layout and the new one.
27658 Think of it as the Emacs @kbd{C-x 2} binding.
27660 This key binding uses the bindable Readline function
27661 @code{tui-change-windows}.
27665 Change the active window. The TUI associates several key bindings
27666 (like scrolling and arrow keys) with the active window. This command
27667 gives the focus to the next TUI window.
27669 Think of it as the Emacs @kbd{C-x o} binding.
27671 This key binding uses the bindable Readline function
27672 @code{tui-other-window}.
27676 Switch in and out of the TUI SingleKey mode that binds single
27677 keys to @value{GDBN} commands (@pxref{TUI Single Key Mode}).
27679 This key binding uses the bindable Readline function
27680 @code{next-keymap}.
27683 The following key bindings only work in the TUI mode:
27688 Scroll the active window one page up.
27692 Scroll the active window one page down.
27696 Scroll the active window one line up.
27700 Scroll the active window one line down.
27704 Scroll the active window one column left.
27708 Scroll the active window one column right.
27712 Refresh the screen.
27715 Because the arrow keys scroll the active window in the TUI mode, they
27716 are not available for their normal use by readline unless the command
27717 window has the focus. When another window is active, you must use
27718 other readline key bindings such as @kbd{C-p}, @kbd{C-n}, @kbd{C-b}
27719 and @kbd{C-f} to control the command window.
27721 @node TUI Single Key Mode
27722 @section TUI Single Key Mode
27723 @cindex TUI single key mode
27725 The TUI also provides a @dfn{SingleKey} mode, which binds several
27726 frequently used @value{GDBN} commands to single keys. Type @kbd{C-x s} to
27727 switch into this mode, where the following key bindings are used:
27730 @kindex c @r{(SingleKey TUI key)}
27734 @kindex d @r{(SingleKey TUI key)}
27738 @kindex f @r{(SingleKey TUI key)}
27742 @kindex n @r{(SingleKey TUI key)}
27746 @kindex o @r{(SingleKey TUI key)}
27748 nexti. The shortcut letter @samp{o} stands for ``step Over''.
27750 @kindex q @r{(SingleKey TUI key)}
27752 exit the SingleKey mode.
27754 @kindex r @r{(SingleKey TUI key)}
27758 @kindex s @r{(SingleKey TUI key)}
27762 @kindex i @r{(SingleKey TUI key)}
27764 stepi. The shortcut letter @samp{i} stands for ``step Into''.
27766 @kindex u @r{(SingleKey TUI key)}
27770 @kindex v @r{(SingleKey TUI key)}
27774 @kindex w @r{(SingleKey TUI key)}
27779 Other keys temporarily switch to the @value{GDBN} command prompt.
27780 The key that was pressed is inserted in the editing buffer so that
27781 it is possible to type most @value{GDBN} commands without interaction
27782 with the TUI SingleKey mode. Once the command is entered the TUI
27783 SingleKey mode is restored. The only way to permanently leave
27784 this mode is by typing @kbd{q} or @kbd{C-x s}.
27786 @cindex SingleKey keymap name
27787 If @value{GDBN} was built with Readline 8.0 or later, the TUI
27788 SingleKey keymap will be named @samp{SingleKey}. This can be used in
27789 @file{.inputrc} to add additional bindings to this keymap.
27792 @section TUI-specific Commands
27793 @cindex TUI commands
27795 The TUI has specific commands to control the text windows.
27796 These commands are always available, even when @value{GDBN} is not in
27797 the TUI mode. When @value{GDBN} is in the standard mode, most
27798 of these commands will automatically switch to the TUI mode.
27800 Note that if @value{GDBN}'s @code{stdout} is not connected to a
27801 terminal, or @value{GDBN} has been started with the machine interface
27802 interpreter (@pxref{GDB/MI, ,The @sc{gdb/mi} Interface}), most of
27803 these commands will fail with an error, because it would not be
27804 possible or desirable to enable curses window management.
27809 Activate TUI mode. The last active TUI window layout will be used if
27810 TUI mode has previously been used in the current debugging session,
27811 otherwise a default layout is used.
27814 @kindex tui disable
27815 Disable TUI mode, returning to the console interpreter.
27819 List and give the size of all displayed windows.
27821 @item layout @var{name}
27823 Changes which TUI windows are displayed. In each layout the command
27824 window is always displayed, the @var{name} parameter controls which
27825 additional windows are displayed, and can be any of the following:
27829 Display the next layout.
27832 Display the previous layout.
27835 Display the source and command windows.
27838 Display the assembly and command windows.
27841 Display the source, assembly, and command windows.
27844 When in @code{src} layout display the register, source, and command
27845 windows. When in @code{asm} or @code{split} layout display the
27846 register, assembler, and command windows.
27849 @item focus @var{name}
27851 Changes which TUI window is currently active for scrolling. The
27852 @var{name} parameter can be any of the following:
27856 Make the next window active for scrolling.
27859 Make the previous window active for scrolling.
27862 Make the source window active for scrolling.
27865 Make the assembly window active for scrolling.
27868 Make the register window active for scrolling.
27871 Make the command window active for scrolling.
27876 Refresh the screen. This is similar to typing @kbd{C-L}.
27878 @item tui reg @var{group}
27880 Changes the register group displayed in the tui register window to
27881 @var{group}. If the register window is not currently displayed this
27882 command will cause the register window to be displayed. The list of
27883 register groups, as well as their order is target specific. The
27884 following groups are available on most targets:
27887 Repeatedly selecting this group will cause the display to cycle
27888 through all of the available register groups.
27891 Repeatedly selecting this group will cause the display to cycle
27892 through all of the available register groups in the reverse order to
27896 Display the general registers.
27898 Display the floating point registers.
27900 Display the system registers.
27902 Display the vector registers.
27904 Display all registers.
27909 Update the source window and the current execution point.
27911 @item winheight @var{name} +@var{count}
27912 @itemx winheight @var{name} -@var{count}
27914 Change the height of the window @var{name} by @var{count}
27915 lines. Positive counts increase the height, while negative counts
27916 decrease it. The @var{name} parameter can be one of @code{src} (the
27917 source window), @code{cmd} (the command window), @code{asm} (the
27918 disassembly window), or @code{regs} (the register display window).
27921 @node TUI Configuration
27922 @section TUI Configuration Variables
27923 @cindex TUI configuration variables
27925 Several configuration variables control the appearance of TUI windows.
27928 @item set tui border-kind @var{kind}
27929 @kindex set tui border-kind
27930 Select the border appearance for the source, assembly and register windows.
27931 The possible values are the following:
27934 Use a space character to draw the border.
27937 Use @sc{ascii} characters @samp{+}, @samp{-} and @samp{|} to draw the border.
27940 Use the Alternate Character Set to draw the border. The border is
27941 drawn using character line graphics if the terminal supports them.
27944 @item set tui border-mode @var{mode}
27945 @kindex set tui border-mode
27946 @itemx set tui active-border-mode @var{mode}
27947 @kindex set tui active-border-mode
27948 Select the display attributes for the borders of the inactive windows
27949 or the active window. The @var{mode} can be one of the following:
27952 Use normal attributes to display the border.
27958 Use reverse video mode.
27961 Use half bright mode.
27963 @item half-standout
27964 Use half bright and standout mode.
27967 Use extra bright or bold mode.
27969 @item bold-standout
27970 Use extra bright or bold and standout mode.
27973 @item set tui tab-width @var{nchars}
27974 @kindex set tui tab-width
27976 Set the width of tab stops to be @var{nchars} characters. This
27977 setting affects the display of TAB characters in the source and
27982 @chapter Using @value{GDBN} under @sc{gnu} Emacs
27985 @cindex @sc{gnu} Emacs
27986 A special interface allows you to use @sc{gnu} Emacs to view (and
27987 edit) the source files for the program you are debugging with
27990 To use this interface, use the command @kbd{M-x gdb} in Emacs. Give the
27991 executable file you want to debug as an argument. This command starts
27992 @value{GDBN} as a subprocess of Emacs, with input and output through a newly
27993 created Emacs buffer.
27994 @c (Do not use the @code{-tui} option to run @value{GDBN} from Emacs.)
27996 Running @value{GDBN} under Emacs can be just like running @value{GDBN} normally except for two
28001 All ``terminal'' input and output goes through an Emacs buffer, called
28004 This applies both to @value{GDBN} commands and their output, and to the input
28005 and output done by the program you are debugging.
28007 This is useful because it means that you can copy the text of previous
28008 commands and input them again; you can even use parts of the output
28011 All the facilities of Emacs' Shell mode are available for interacting
28012 with your program. In particular, you can send signals the usual
28013 way---for example, @kbd{C-c C-c} for an interrupt, @kbd{C-c C-z} for a
28017 @value{GDBN} displays source code through Emacs.
28019 Each time @value{GDBN} displays a stack frame, Emacs automatically finds the
28020 source file for that frame and puts an arrow (@samp{=>}) at the
28021 left margin of the current line. Emacs uses a separate buffer for
28022 source display, and splits the screen to show both your @value{GDBN} session
28025 Explicit @value{GDBN} @code{list} or search commands still produce output as
28026 usual, but you probably have no reason to use them from Emacs.
28029 We call this @dfn{text command mode}. Emacs 22.1, and later, also uses
28030 a graphical mode, enabled by default, which provides further buffers
28031 that can control the execution and describe the state of your program.
28032 @xref{GDB Graphical Interface,,, Emacs, The @sc{gnu} Emacs Manual}.
28034 If you specify an absolute file name when prompted for the @kbd{M-x
28035 gdb} argument, then Emacs sets your current working directory to where
28036 your program resides. If you only specify the file name, then Emacs
28037 sets your current working directory to the directory associated
28038 with the previous buffer. In this case, @value{GDBN} may find your
28039 program by searching your environment's @code{PATH} variable, but on
28040 some operating systems it might not find the source. So, although the
28041 @value{GDBN} input and output session proceeds normally, the auxiliary
28042 buffer does not display the current source and line of execution.
28044 The initial working directory of @value{GDBN} is printed on the top
28045 line of the GUD buffer and this serves as a default for the commands
28046 that specify files for @value{GDBN} to operate on. @xref{Files,
28047 ,Commands to Specify Files}.
28049 By default, @kbd{M-x gdb} calls the program called @file{gdb}. If you
28050 need to call @value{GDBN} by a different name (for example, if you
28051 keep several configurations around, with different names) you can
28052 customize the Emacs variable @code{gud-gdb-command-name} to run the
28055 In the GUD buffer, you can use these special Emacs commands in
28056 addition to the standard Shell mode commands:
28060 Describe the features of Emacs' GUD Mode.
28063 Execute to another source line, like the @value{GDBN} @code{step} command; also
28064 update the display window to show the current file and location.
28067 Execute to next source line in this function, skipping all function
28068 calls, like the @value{GDBN} @code{next} command. Then update the display window
28069 to show the current file and location.
28072 Execute one instruction, like the @value{GDBN} @code{stepi} command; update
28073 display window accordingly.
28076 Execute until exit from the selected stack frame, like the @value{GDBN}
28077 @code{finish} command.
28080 Continue execution of your program, like the @value{GDBN} @code{continue}
28084 Go up the number of frames indicated by the numeric argument
28085 (@pxref{Arguments, , Numeric Arguments, Emacs, The @sc{gnu} Emacs Manual}),
28086 like the @value{GDBN} @code{up} command.
28089 Go down the number of frames indicated by the numeric argument, like the
28090 @value{GDBN} @code{down} command.
28093 In any source file, the Emacs command @kbd{C-x @key{SPC}} (@code{gud-break})
28094 tells @value{GDBN} to set a breakpoint on the source line point is on.
28096 In text command mode, if you type @kbd{M-x speedbar}, Emacs displays a
28097 separate frame which shows a backtrace when the GUD buffer is current.
28098 Move point to any frame in the stack and type @key{RET} to make it
28099 become the current frame and display the associated source in the
28100 source buffer. Alternatively, click @kbd{Mouse-2} to make the
28101 selected frame become the current one. In graphical mode, the
28102 speedbar displays watch expressions.
28104 If you accidentally delete the source-display buffer, an easy way to get
28105 it back is to type the command @code{f} in the @value{GDBN} buffer, to
28106 request a frame display; when you run under Emacs, this recreates
28107 the source buffer if necessary to show you the context of the current
28110 The source files displayed in Emacs are in ordinary Emacs buffers
28111 which are visiting the source files in the usual way. You can edit
28112 the files with these buffers if you wish; but keep in mind that @value{GDBN}
28113 communicates with Emacs in terms of line numbers. If you add or
28114 delete lines from the text, the line numbers that @value{GDBN} knows cease
28115 to correspond properly with the code.
28117 A more detailed description of Emacs' interaction with @value{GDBN} is
28118 given in the Emacs manual (@pxref{Debuggers,,, Emacs, The @sc{gnu}
28122 @chapter The @sc{gdb/mi} Interface
28124 @unnumberedsec Function and Purpose
28126 @cindex @sc{gdb/mi}, its purpose
28127 @sc{gdb/mi} is a line based machine oriented text interface to
28128 @value{GDBN} and is activated by specifying using the
28129 @option{--interpreter} command line option (@pxref{Mode Options}). It
28130 is specifically intended to support the development of systems which
28131 use the debugger as just one small component of a larger system.
28133 This chapter is a specification of the @sc{gdb/mi} interface. It is written
28134 in the form of a reference manual.
28136 Note that @sc{gdb/mi} is still under construction, so some of the
28137 features described below are incomplete and subject to change
28138 (@pxref{GDB/MI Development and Front Ends, , @sc{gdb/mi} Development and Front Ends}).
28140 @unnumberedsec Notation and Terminology
28142 @cindex notational conventions, for @sc{gdb/mi}
28143 This chapter uses the following notation:
28147 @code{|} separates two alternatives.
28150 @code{[ @var{something} ]} indicates that @var{something} is optional:
28151 it may or may not be given.
28154 @code{( @var{group} )*} means that @var{group} inside the parentheses
28155 may repeat zero or more times.
28158 @code{( @var{group} )+} means that @var{group} inside the parentheses
28159 may repeat one or more times.
28162 @code{"@var{string}"} means a literal @var{string}.
28166 @heading Dependencies
28170 * GDB/MI General Design::
28171 * GDB/MI Command Syntax::
28172 * GDB/MI Compatibility with CLI::
28173 * GDB/MI Development and Front Ends::
28174 * GDB/MI Output Records::
28175 * GDB/MI Simple Examples::
28176 * GDB/MI Command Description Format::
28177 * GDB/MI Breakpoint Commands::
28178 * GDB/MI Catchpoint Commands::
28179 * GDB/MI Program Context::
28180 * GDB/MI Thread Commands::
28181 * GDB/MI Ada Tasking Commands::
28182 * GDB/MI Program Execution::
28183 * GDB/MI Stack Manipulation::
28184 * GDB/MI Variable Objects::
28185 * GDB/MI Data Manipulation::
28186 * GDB/MI Tracepoint Commands::
28187 * GDB/MI Symbol Query::
28188 * GDB/MI File Commands::
28190 * GDB/MI Kod Commands::
28191 * GDB/MI Memory Overlay Commands::
28192 * GDB/MI Signal Handling Commands::
28194 * GDB/MI Target Manipulation::
28195 * GDB/MI File Transfer Commands::
28196 * GDB/MI Ada Exceptions Commands::
28197 * GDB/MI Support Commands::
28198 * GDB/MI Miscellaneous Commands::
28201 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28202 @node GDB/MI General Design
28203 @section @sc{gdb/mi} General Design
28204 @cindex GDB/MI General Design
28206 Interaction of a @sc{GDB/MI} frontend with @value{GDBN} involves three
28207 parts---commands sent to @value{GDBN}, responses to those commands
28208 and notifications. Each command results in exactly one response,
28209 indicating either successful completion of the command, or an error.
28210 For the commands that do not resume the target, the response contains the
28211 requested information. For the commands that resume the target, the
28212 response only indicates whether the target was successfully resumed.
28213 Notifications is the mechanism for reporting changes in the state of the
28214 target, or in @value{GDBN} state, that cannot conveniently be associated with
28215 a command and reported as part of that command response.
28217 The important examples of notifications are:
28221 Exec notifications. These are used to report changes in
28222 target state---when a target is resumed, or stopped. It would not
28223 be feasible to include this information in response of resuming
28224 commands, because one resume commands can result in multiple events in
28225 different threads. Also, quite some time may pass before any event
28226 happens in the target, while a frontend needs to know whether the resuming
28227 command itself was successfully executed.
28230 Console output, and status notifications. Console output
28231 notifications are used to report output of CLI commands, as well as
28232 diagnostics for other commands. Status notifications are used to
28233 report the progress of a long-running operation. Naturally, including
28234 this information in command response would mean no output is produced
28235 until the command is finished, which is undesirable.
28238 General notifications. Commands may have various side effects on
28239 the @value{GDBN} or target state beyond their official purpose. For example,
28240 a command may change the selected thread. Although such changes can
28241 be included in command response, using notification allows for more
28242 orthogonal frontend design.
28246 There's no guarantee that whenever an MI command reports an error,
28247 @value{GDBN} or the target are in any specific state, and especially,
28248 the state is not reverted to the state before the MI command was
28249 processed. Therefore, whenever an MI command results in an error,
28250 we recommend that the frontend refreshes all the information shown in
28251 the user interface.
28255 * Context management::
28256 * Asynchronous and non-stop modes::
28260 @node Context management
28261 @subsection Context management
28263 @subsubsection Threads and Frames
28265 In most cases when @value{GDBN} accesses the target, this access is
28266 done in context of a specific thread and frame (@pxref{Frames}).
28267 Often, even when accessing global data, the target requires that a thread
28268 be specified. The CLI interface maintains the selected thread and frame,
28269 and supplies them to target on each command. This is convenient,
28270 because a command line user would not want to specify that information
28271 explicitly on each command, and because user interacts with
28272 @value{GDBN} via a single terminal, so no confusion is possible as
28273 to what thread and frame are the current ones.
28275 In the case of MI, the concept of selected thread and frame is less
28276 useful. First, a frontend can easily remember this information
28277 itself. Second, a graphical frontend can have more than one window,
28278 each one used for debugging a different thread, and the frontend might
28279 want to access additional threads for internal purposes. This
28280 increases the risk that by relying on implicitly selected thread, the
28281 frontend may be operating on a wrong one. Therefore, each MI command
28282 should explicitly specify which thread and frame to operate on. To
28283 make it possible, each MI command accepts the @samp{--thread} and
28284 @samp{--frame} options, the value to each is @value{GDBN} global
28285 identifier for thread and frame to operate on.
28287 Usually, each top-level window in a frontend allows the user to select
28288 a thread and a frame, and remembers the user selection for further
28289 operations. However, in some cases @value{GDBN} may suggest that the
28290 current thread or frame be changed. For example, when stopping on a
28291 breakpoint it is reasonable to switch to the thread where breakpoint is
28292 hit. For another example, if the user issues the CLI @samp{thread} or
28293 @samp{frame} commands via the frontend, it is desirable to change the
28294 frontend's selection to the one specified by user. @value{GDBN}
28295 communicates the suggestion to change current thread and frame using the
28296 @samp{=thread-selected} notification.
28298 Note that historically, MI shares the selected thread with CLI, so
28299 frontends used the @code{-thread-select} to execute commands in the
28300 right context. However, getting this to work right is cumbersome. The
28301 simplest way is for frontend to emit @code{-thread-select} command
28302 before every command. This doubles the number of commands that need
28303 to be sent. The alternative approach is to suppress @code{-thread-select}
28304 if the selected thread in @value{GDBN} is supposed to be identical to the
28305 thread the frontend wants to operate on. However, getting this
28306 optimization right can be tricky. In particular, if the frontend
28307 sends several commands to @value{GDBN}, and one of the commands changes the
28308 selected thread, then the behaviour of subsequent commands will
28309 change. So, a frontend should either wait for response from such
28310 problematic commands, or explicitly add @code{-thread-select} for
28311 all subsequent commands. No frontend is known to do this exactly
28312 right, so it is suggested to just always pass the @samp{--thread} and
28313 @samp{--frame} options.
28315 @subsubsection Language
28317 The execution of several commands depends on which language is selected.
28318 By default, the current language (@pxref{show language}) is used.
28319 But for commands known to be language-sensitive, it is recommended
28320 to use the @samp{--language} option. This option takes one argument,
28321 which is the name of the language to use while executing the command.
28325 -data-evaluate-expression --language c "sizeof (void*)"
28330 The valid language names are the same names accepted by the
28331 @samp{set language} command (@pxref{Manually}), excluding @samp{auto},
28332 @samp{local} or @samp{unknown}.
28334 @node Asynchronous and non-stop modes
28335 @subsection Asynchronous command execution and non-stop mode
28337 On some targets, @value{GDBN} is capable of processing MI commands
28338 even while the target is running. This is called @dfn{asynchronous
28339 command execution} (@pxref{Background Execution}). The frontend may
28340 specify a preference for asynchronous execution using the
28341 @code{-gdb-set mi-async 1} command, which should be emitted before
28342 either running the executable or attaching to the target. After the
28343 frontend has started the executable or attached to the target, it can
28344 find if asynchronous execution is enabled using the
28345 @code{-list-target-features} command.
28348 @item -gdb-set mi-async on
28349 @item -gdb-set mi-async off
28350 Set whether MI is in asynchronous mode.
28352 When @code{off}, which is the default, MI execution commands (e.g.,
28353 @code{-exec-continue}) are foreground commands, and @value{GDBN} waits
28354 for the program to stop before processing further commands.
28356 When @code{on}, MI execution commands are background execution
28357 commands (e.g., @code{-exec-continue} becomes the equivalent of the
28358 @code{c&} CLI command), and so @value{GDBN} is capable of processing
28359 MI commands even while the target is running.
28361 @item -gdb-show mi-async
28362 Show whether MI asynchronous mode is enabled.
28365 Note: In @value{GDBN} version 7.7 and earlier, this option was called
28366 @code{target-async} instead of @code{mi-async}, and it had the effect
28367 of both putting MI in asynchronous mode and making CLI background
28368 commands possible. CLI background commands are now always possible
28369 ``out of the box'' if the target supports them. The old spelling is
28370 kept as a deprecated alias for backwards compatibility.
28372 Even if @value{GDBN} can accept a command while target is running,
28373 many commands that access the target do not work when the target is
28374 running. Therefore, asynchronous command execution is most useful
28375 when combined with non-stop mode (@pxref{Non-Stop Mode}). Then,
28376 it is possible to examine the state of one thread, while other threads
28379 When a given thread is running, MI commands that try to access the
28380 target in the context of that thread may not work, or may work only on
28381 some targets. In particular, commands that try to operate on thread's
28382 stack will not work, on any target. Commands that read memory, or
28383 modify breakpoints, may work or not work, depending on the target. Note
28384 that even commands that operate on global state, such as @code{print},
28385 @code{set}, and breakpoint commands, still access the target in the
28386 context of a specific thread, so frontend should try to find a
28387 stopped thread and perform the operation on that thread (using the
28388 @samp{--thread} option).
28390 Which commands will work in the context of a running thread is
28391 highly target dependent. However, the two commands
28392 @code{-exec-interrupt}, to stop a thread, and @code{-thread-info},
28393 to find the state of a thread, will always work.
28395 @node Thread groups
28396 @subsection Thread groups
28397 @value{GDBN} may be used to debug several processes at the same time.
28398 On some platforms, @value{GDBN} may support debugging of several
28399 hardware systems, each one having several cores with several different
28400 processes running on each core. This section describes the MI
28401 mechanism to support such debugging scenarios.
28403 The key observation is that regardless of the structure of the
28404 target, MI can have a global list of threads, because most commands that
28405 accept the @samp{--thread} option do not need to know what process that
28406 thread belongs to. Therefore, it is not necessary to introduce
28407 neither additional @samp{--process} option, nor an notion of the
28408 current process in the MI interface. The only strictly new feature
28409 that is required is the ability to find how the threads are grouped
28412 To allow the user to discover such grouping, and to support arbitrary
28413 hierarchy of machines/cores/processes, MI introduces the concept of a
28414 @dfn{thread group}. Thread group is a collection of threads and other
28415 thread groups. A thread group always has a string identifier, a type,
28416 and may have additional attributes specific to the type. A new
28417 command, @code{-list-thread-groups}, returns the list of top-level
28418 thread groups, which correspond to processes that @value{GDBN} is
28419 debugging at the moment. By passing an identifier of a thread group
28420 to the @code{-list-thread-groups} command, it is possible to obtain
28421 the members of specific thread group.
28423 To allow the user to easily discover processes, and other objects, he
28424 wishes to debug, a concept of @dfn{available thread group} is
28425 introduced. Available thread group is an thread group that
28426 @value{GDBN} is not debugging, but that can be attached to, using the
28427 @code{-target-attach} command. The list of available top-level thread
28428 groups can be obtained using @samp{-list-thread-groups --available}.
28429 In general, the content of a thread group may be only retrieved only
28430 after attaching to that thread group.
28432 Thread groups are related to inferiors (@pxref{Inferiors and
28433 Programs}). Each inferior corresponds to a thread group of a special
28434 type @samp{process}, and some additional operations are permitted on
28435 such thread groups.
28437 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28438 @node GDB/MI Command Syntax
28439 @section @sc{gdb/mi} Command Syntax
28442 * GDB/MI Input Syntax::
28443 * GDB/MI Output Syntax::
28446 @node GDB/MI Input Syntax
28447 @subsection @sc{gdb/mi} Input Syntax
28449 @cindex input syntax for @sc{gdb/mi}
28450 @cindex @sc{gdb/mi}, input syntax
28452 @item @var{command} @expansion{}
28453 @code{@var{cli-command} | @var{mi-command}}
28455 @item @var{cli-command} @expansion{}
28456 @code{[ @var{token} ] @var{cli-command} @var{nl}}, where
28457 @var{cli-command} is any existing @value{GDBN} CLI command.
28459 @item @var{mi-command} @expansion{}
28460 @code{[ @var{token} ] "-" @var{operation} ( " " @var{option} )*
28461 @code{[} " --" @code{]} ( " " @var{parameter} )* @var{nl}}
28463 @item @var{token} @expansion{}
28464 "any sequence of digits"
28466 @item @var{option} @expansion{}
28467 @code{"-" @var{parameter} [ " " @var{parameter} ]}
28469 @item @var{parameter} @expansion{}
28470 @code{@var{non-blank-sequence} | @var{c-string}}
28472 @item @var{operation} @expansion{}
28473 @emph{any of the operations described in this chapter}
28475 @item @var{non-blank-sequence} @expansion{}
28476 @emph{anything, provided it doesn't contain special characters such as
28477 "-", @var{nl}, """ and of course " "}
28479 @item @var{c-string} @expansion{}
28480 @code{""" @var{seven-bit-iso-c-string-content} """}
28482 @item @var{nl} @expansion{}
28491 The CLI commands are still handled by the @sc{mi} interpreter; their
28492 output is described below.
28495 The @code{@var{token}}, when present, is passed back when the command
28499 Some @sc{mi} commands accept optional arguments as part of the parameter
28500 list. Each option is identified by a leading @samp{-} (dash) and may be
28501 followed by an optional argument parameter. Options occur first in the
28502 parameter list and can be delimited from normal parameters using
28503 @samp{--} (this is useful when some parameters begin with a dash).
28510 We want easy access to the existing CLI syntax (for debugging).
28513 We want it to be easy to spot a @sc{mi} operation.
28516 @node GDB/MI Output Syntax
28517 @subsection @sc{gdb/mi} Output Syntax
28519 @cindex output syntax of @sc{gdb/mi}
28520 @cindex @sc{gdb/mi}, output syntax
28521 The output from @sc{gdb/mi} consists of zero or more out-of-band records
28522 followed, optionally, by a single result record. This result record
28523 is for the most recent command. The sequence of output records is
28524 terminated by @samp{(gdb)}.
28526 If an input command was prefixed with a @code{@var{token}} then the
28527 corresponding output for that command will also be prefixed by that same
28531 @item @var{output} @expansion{}
28532 @code{( @var{out-of-band-record} )* [ @var{result-record} ] "(gdb)" @var{nl}}
28534 @item @var{result-record} @expansion{}
28535 @code{ [ @var{token} ] "^" @var{result-class} ( "," @var{result} )* @var{nl}}
28537 @item @var{out-of-band-record} @expansion{}
28538 @code{@var{async-record} | @var{stream-record}}
28540 @item @var{async-record} @expansion{}
28541 @code{@var{exec-async-output} | @var{status-async-output} | @var{notify-async-output}}
28543 @item @var{exec-async-output} @expansion{}
28544 @code{[ @var{token} ] "*" @var{async-output nl}}
28546 @item @var{status-async-output} @expansion{}
28547 @code{[ @var{token} ] "+" @var{async-output nl}}
28549 @item @var{notify-async-output} @expansion{}
28550 @code{[ @var{token} ] "=" @var{async-output nl}}
28552 @item @var{async-output} @expansion{}
28553 @code{@var{async-class} ( "," @var{result} )*}
28555 @item @var{result-class} @expansion{}
28556 @code{"done" | "running" | "connected" | "error" | "exit"}
28558 @item @var{async-class} @expansion{}
28559 @code{"stopped" | @var{others}} (where @var{others} will be added
28560 depending on the needs---this is still in development).
28562 @item @var{result} @expansion{}
28563 @code{ @var{variable} "=" @var{value}}
28565 @item @var{variable} @expansion{}
28566 @code{ @var{string} }
28568 @item @var{value} @expansion{}
28569 @code{ @var{const} | @var{tuple} | @var{list} }
28571 @item @var{const} @expansion{}
28572 @code{@var{c-string}}
28574 @item @var{tuple} @expansion{}
28575 @code{ "@{@}" | "@{" @var{result} ( "," @var{result} )* "@}" }
28577 @item @var{list} @expansion{}
28578 @code{ "[]" | "[" @var{value} ( "," @var{value} )* "]" | "["
28579 @var{result} ( "," @var{result} )* "]" }
28581 @item @var{stream-record} @expansion{}
28582 @code{@var{console-stream-output} | @var{target-stream-output} | @var{log-stream-output}}
28584 @item @var{console-stream-output} @expansion{}
28585 @code{"~" @var{c-string nl}}
28587 @item @var{target-stream-output} @expansion{}
28588 @code{"@@" @var{c-string nl}}
28590 @item @var{log-stream-output} @expansion{}
28591 @code{"&" @var{c-string nl}}
28593 @item @var{nl} @expansion{}
28596 @item @var{token} @expansion{}
28597 @emph{any sequence of digits}.
28605 All output sequences end in a single line containing a period.
28608 The @code{@var{token}} is from the corresponding request. Note that
28609 for all async output, while the token is allowed by the grammar and
28610 may be output by future versions of @value{GDBN} for select async
28611 output messages, it is generally omitted. Frontends should treat
28612 all async output as reporting general changes in the state of the
28613 target and there should be no need to associate async output to any
28617 @cindex status output in @sc{gdb/mi}
28618 @var{status-async-output} contains on-going status information about the
28619 progress of a slow operation. It can be discarded. All status output is
28620 prefixed by @samp{+}.
28623 @cindex async output in @sc{gdb/mi}
28624 @var{exec-async-output} contains asynchronous state change on the target
28625 (stopped, started, disappeared). All async output is prefixed by
28629 @cindex notify output in @sc{gdb/mi}
28630 @var{notify-async-output} contains supplementary information that the
28631 client should handle (e.g., a new breakpoint information). All notify
28632 output is prefixed by @samp{=}.
28635 @cindex console output in @sc{gdb/mi}
28636 @var{console-stream-output} is output that should be displayed as is in the
28637 console. It is the textual response to a CLI command. All the console
28638 output is prefixed by @samp{~}.
28641 @cindex target output in @sc{gdb/mi}
28642 @var{target-stream-output} is the output produced by the target program.
28643 All the target output is prefixed by @samp{@@}.
28646 @cindex log output in @sc{gdb/mi}
28647 @var{log-stream-output} is output text coming from @value{GDBN}'s internals, for
28648 instance messages that should be displayed as part of an error log. All
28649 the log output is prefixed by @samp{&}.
28652 @cindex list output in @sc{gdb/mi}
28653 New @sc{gdb/mi} commands should only output @var{lists} containing
28659 @xref{GDB/MI Stream Records, , @sc{gdb/mi} Stream Records}, for more
28660 details about the various output records.
28662 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28663 @node GDB/MI Compatibility with CLI
28664 @section @sc{gdb/mi} Compatibility with CLI
28666 @cindex compatibility, @sc{gdb/mi} and CLI
28667 @cindex @sc{gdb/mi}, compatibility with CLI
28669 For the developers convenience CLI commands can be entered directly,
28670 but there may be some unexpected behaviour. For example, commands
28671 that query the user will behave as if the user replied yes, breakpoint
28672 command lists are not executed and some CLI commands, such as
28673 @code{if}, @code{when} and @code{define}, prompt for further input with
28674 @samp{>}, which is not valid MI output.
28676 This feature may be removed at some stage in the future and it is
28677 recommended that front ends use the @code{-interpreter-exec} command
28678 (@pxref{-interpreter-exec}).
28680 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28681 @node GDB/MI Development and Front Ends
28682 @section @sc{gdb/mi} Development and Front Ends
28683 @cindex @sc{gdb/mi} development
28685 The application which takes the MI output and presents the state of the
28686 program being debugged to the user is called a @dfn{front end}.
28688 Since @sc{gdb/mi} is used by a variety of front ends to @value{GDBN}, changes
28689 to the MI interface may break existing usage. This section describes how the
28690 protocol changes and how to request previous version of the protocol when it
28693 Some changes in MI need not break a carefully designed front end, and
28694 for these the MI version will remain unchanged. The following is a
28695 list of changes that may occur within one level, so front ends should
28696 parse MI output in a way that can handle them:
28700 New MI commands may be added.
28703 New fields may be added to the output of any MI command.
28706 The range of values for fields with specified values, e.g.,
28707 @code{in_scope} (@pxref{-var-update}) may be extended.
28709 @c The format of field's content e.g type prefix, may change so parse it
28710 @c at your own risk. Yes, in general?
28712 @c The order of fields may change? Shouldn't really matter but it might
28713 @c resolve inconsistencies.
28716 If the changes are likely to break front ends, the MI version level
28717 will be increased by one. The new versions of the MI protocol are not compatible
28718 with the old versions. Old versions of MI remain available, allowing front ends
28719 to keep using them until they are modified to use the latest MI version.
28721 Since @code{--interpreter=mi} always points to the latest MI version, it is
28722 recommended that front ends request a specific version of MI when launching
28723 @value{GDBN} (e.g. @code{--interpreter=mi2}) to make sure they get an
28724 interpreter with the MI version they expect.
28726 The following table gives a summary of the the released versions of the MI
28727 interface: the version number, the version of GDB in which it first appeared
28728 and the breaking changes compared to the previous version.
28730 @multitable @columnfractions .05 .05 .9
28731 @headitem MI version @tab GDB version @tab Breaking changes
28748 The @code{-environment-pwd}, @code{-environment-directory} and
28749 @code{-environment-path} commands now returns values using the MI output
28750 syntax, rather than CLI output syntax.
28753 @code{-var-list-children}'s @code{children} result field is now a list, rather
28757 @code{-var-update}'s @code{changelist} result field is now a list, rather than
28769 The output of information about multi-location breakpoints has changed in the
28770 responses to the @code{-break-insert} and @code{-break-info} commands, as well
28771 as in the @code{=breakpoint-created} and @code{=breakpoint-modified} events.
28772 The multiple locations are now placed in a @code{locations} field, whose value
28778 If your front end cannot yet migrate to a more recent version of the
28779 MI protocol, you can nevertheless selectively enable specific features
28780 available in those recent MI versions, using the following commands:
28784 @item -fix-multi-location-breakpoint-output
28785 Use the output for multi-location breakpoints which was introduced by
28786 MI 3, even when using MI versions 2 or 1. This command has no
28787 effect when using MI version 3 or later.
28791 The best way to avoid unexpected changes in MI that might break your front
28792 end is to make your project known to @value{GDBN} developers and
28793 follow development on @email{gdb@@sourceware.org} and
28794 @email{gdb-patches@@sourceware.org}.
28795 @cindex mailing lists
28797 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28798 @node GDB/MI Output Records
28799 @section @sc{gdb/mi} Output Records
28802 * GDB/MI Result Records::
28803 * GDB/MI Stream Records::
28804 * GDB/MI Async Records::
28805 * GDB/MI Breakpoint Information::
28806 * GDB/MI Frame Information::
28807 * GDB/MI Thread Information::
28808 * GDB/MI Ada Exception Information::
28811 @node GDB/MI Result Records
28812 @subsection @sc{gdb/mi} Result Records
28814 @cindex result records in @sc{gdb/mi}
28815 @cindex @sc{gdb/mi}, result records
28816 In addition to a number of out-of-band notifications, the response to a
28817 @sc{gdb/mi} command includes one of the following result indications:
28821 @item "^done" [ "," @var{results} ]
28822 The synchronous operation was successful, @code{@var{results}} are the return
28827 This result record is equivalent to @samp{^done}. Historically, it
28828 was output instead of @samp{^done} if the command has resumed the
28829 target. This behaviour is maintained for backward compatibility, but
28830 all frontends should treat @samp{^done} and @samp{^running}
28831 identically and rely on the @samp{*running} output record to determine
28832 which threads are resumed.
28836 @value{GDBN} has connected to a remote target.
28838 @item "^error" "," "msg=" @var{c-string} [ "," "code=" @var{c-string} ]
28840 The operation failed. The @code{msg=@var{c-string}} variable contains
28841 the corresponding error message.
28843 If present, the @code{code=@var{c-string}} variable provides an error
28844 code on which consumers can rely on to detect the corresponding
28845 error condition. At present, only one error code is defined:
28848 @item "undefined-command"
28849 Indicates that the command causing the error does not exist.
28854 @value{GDBN} has terminated.
28858 @node GDB/MI Stream Records
28859 @subsection @sc{gdb/mi} Stream Records
28861 @cindex @sc{gdb/mi}, stream records
28862 @cindex stream records in @sc{gdb/mi}
28863 @value{GDBN} internally maintains a number of output streams: the console, the
28864 target, and the log. The output intended for each of these streams is
28865 funneled through the @sc{gdb/mi} interface using @dfn{stream records}.
28867 Each stream record begins with a unique @dfn{prefix character} which
28868 identifies its stream (@pxref{GDB/MI Output Syntax, , @sc{gdb/mi} Output
28869 Syntax}). In addition to the prefix, each stream record contains a
28870 @code{@var{string-output}}. This is either raw text (with an implicit new
28871 line) or a quoted C string (which does not contain an implicit newline).
28874 @item "~" @var{string-output}
28875 The console output stream contains text that should be displayed in the
28876 CLI console window. It contains the textual responses to CLI commands.
28878 @item "@@" @var{string-output}
28879 The target output stream contains any textual output from the running
28880 target. This is only present when GDB's event loop is truly
28881 asynchronous, which is currently only the case for remote targets.
28883 @item "&" @var{string-output}
28884 The log stream contains debugging messages being produced by @value{GDBN}'s
28888 @node GDB/MI Async Records
28889 @subsection @sc{gdb/mi} Async Records
28891 @cindex async records in @sc{gdb/mi}
28892 @cindex @sc{gdb/mi}, async records
28893 @dfn{Async} records are used to notify the @sc{gdb/mi} client of
28894 additional changes that have occurred. Those changes can either be a
28895 consequence of @sc{gdb/mi} commands (e.g., a breakpoint modified) or a result of
28896 target activity (e.g., target stopped).
28898 The following is the list of possible async records:
28902 @item *running,thread-id="@var{thread}"
28903 The target is now running. The @var{thread} field can be the global
28904 thread ID of the the thread that is now running, and it can be
28905 @samp{all} if all threads are running. The frontend should assume
28906 that no interaction with a running thread is possible after this
28907 notification is produced. The frontend should not assume that this
28908 notification is output only once for any command. @value{GDBN} may
28909 emit this notification several times, either for different threads,
28910 because it cannot resume all threads together, or even for a single
28911 thread, if the thread must be stepped though some code before letting
28914 @item *stopped,reason="@var{reason}",thread-id="@var{id}",stopped-threads="@var{stopped}",core="@var{core}"
28915 The target has stopped. The @var{reason} field can have one of the
28919 @item breakpoint-hit
28920 A breakpoint was reached.
28921 @item watchpoint-trigger
28922 A watchpoint was triggered.
28923 @item read-watchpoint-trigger
28924 A read watchpoint was triggered.
28925 @item access-watchpoint-trigger
28926 An access watchpoint was triggered.
28927 @item function-finished
28928 An -exec-finish or similar CLI command was accomplished.
28929 @item location-reached
28930 An -exec-until or similar CLI command was accomplished.
28931 @item watchpoint-scope
28932 A watchpoint has gone out of scope.
28933 @item end-stepping-range
28934 An -exec-next, -exec-next-instruction, -exec-step, -exec-step-instruction or
28935 similar CLI command was accomplished.
28936 @item exited-signalled
28937 The inferior exited because of a signal.
28939 The inferior exited.
28940 @item exited-normally
28941 The inferior exited normally.
28942 @item signal-received
28943 A signal was received by the inferior.
28945 The inferior has stopped due to a library being loaded or unloaded.
28946 This can happen when @code{stop-on-solib-events} (@pxref{Files}) is
28947 set or when a @code{catch load} or @code{catch unload} catchpoint is
28948 in use (@pxref{Set Catchpoints}).
28950 The inferior has forked. This is reported when @code{catch fork}
28951 (@pxref{Set Catchpoints}) has been used.
28953 The inferior has vforked. This is reported in when @code{catch vfork}
28954 (@pxref{Set Catchpoints}) has been used.
28955 @item syscall-entry
28956 The inferior entered a system call. This is reported when @code{catch
28957 syscall} (@pxref{Set Catchpoints}) has been used.
28958 @item syscall-return
28959 The inferior returned from a system call. This is reported when
28960 @code{catch syscall} (@pxref{Set Catchpoints}) has been used.
28962 The inferior called @code{exec}. This is reported when @code{catch exec}
28963 (@pxref{Set Catchpoints}) has been used.
28966 The @var{id} field identifies the global thread ID of the thread
28967 that directly caused the stop -- for example by hitting a breakpoint.
28968 Depending on whether all-stop
28969 mode is in effect (@pxref{All-Stop Mode}), @value{GDBN} may either
28970 stop all threads, or only the thread that directly triggered the stop.
28971 If all threads are stopped, the @var{stopped} field will have the
28972 value of @code{"all"}. Otherwise, the value of the @var{stopped}
28973 field will be a list of thread identifiers. Presently, this list will
28974 always include a single thread, but frontend should be prepared to see
28975 several threads in the list. The @var{core} field reports the
28976 processor core on which the stop event has happened. This field may be absent
28977 if such information is not available.
28979 @item =thread-group-added,id="@var{id}"
28980 @itemx =thread-group-removed,id="@var{id}"
28981 A thread group was either added or removed. The @var{id} field
28982 contains the @value{GDBN} identifier of the thread group. When a thread
28983 group is added, it generally might not be associated with a running
28984 process. When a thread group is removed, its id becomes invalid and
28985 cannot be used in any way.
28987 @item =thread-group-started,id="@var{id}",pid="@var{pid}"
28988 A thread group became associated with a running program,
28989 either because the program was just started or the thread group
28990 was attached to a program. The @var{id} field contains the
28991 @value{GDBN} identifier of the thread group. The @var{pid} field
28992 contains process identifier, specific to the operating system.
28994 @item =thread-group-exited,id="@var{id}"[,exit-code="@var{code}"]
28995 A thread group is no longer associated with a running program,
28996 either because the program has exited, or because it was detached
28997 from. The @var{id} field contains the @value{GDBN} identifier of the
28998 thread group. The @var{code} field is the exit code of the inferior; it exists
28999 only when the inferior exited with some code.
29001 @item =thread-created,id="@var{id}",group-id="@var{gid}"
29002 @itemx =thread-exited,id="@var{id}",group-id="@var{gid}"
29003 A thread either was created, or has exited. The @var{id} field
29004 contains the global @value{GDBN} identifier of the thread. The @var{gid}
29005 field identifies the thread group this thread belongs to.
29007 @item =thread-selected,id="@var{id}"[,frame="@var{frame}"]
29008 Informs that the selected thread or frame were changed. This notification
29009 is not emitted as result of the @code{-thread-select} or
29010 @code{-stack-select-frame} commands, but is emitted whenever an MI command
29011 that is not documented to change the selected thread and frame actually
29012 changes them. In particular, invoking, directly or indirectly
29013 (via user-defined command), the CLI @code{thread} or @code{frame} commands,
29014 will generate this notification. Changing the thread or frame from another
29015 user interface (see @ref{Interpreters}) will also generate this notification.
29017 The @var{frame} field is only present if the newly selected thread is
29018 stopped. See @ref{GDB/MI Frame Information} for the format of its value.
29020 We suggest that in response to this notification, front ends
29021 highlight the selected thread and cause subsequent commands to apply to
29024 @item =library-loaded,...
29025 Reports that a new library file was loaded by the program. This
29026 notification has 5 fields---@var{id}, @var{target-name},
29027 @var{host-name}, @var{symbols-loaded} and @var{ranges}. The @var{id} field is an
29028 opaque identifier of the library. For remote debugging case,
29029 @var{target-name} and @var{host-name} fields give the name of the
29030 library file on the target, and on the host respectively. For native
29031 debugging, both those fields have the same value. The
29032 @var{symbols-loaded} field is emitted only for backward compatibility
29033 and should not be relied on to convey any useful information. The
29034 @var{thread-group} field, if present, specifies the id of the thread
29035 group in whose context the library was loaded. If the field is
29036 absent, it means the library was loaded in the context of all present
29037 thread groups. The @var{ranges} field specifies the ranges of addresses belonging
29040 @item =library-unloaded,...
29041 Reports that a library was unloaded by the program. This notification
29042 has 3 fields---@var{id}, @var{target-name} and @var{host-name} with
29043 the same meaning as for the @code{=library-loaded} notification.
29044 The @var{thread-group} field, if present, specifies the id of the
29045 thread group in whose context the library was unloaded. If the field is
29046 absent, it means the library was unloaded in the context of all present
29049 @item =traceframe-changed,num=@var{tfnum},tracepoint=@var{tpnum}
29050 @itemx =traceframe-changed,end
29051 Reports that the trace frame was changed and its new number is
29052 @var{tfnum}. The number of the tracepoint associated with this trace
29053 frame is @var{tpnum}.
29055 @item =tsv-created,name=@var{name},initial=@var{initial}
29056 Reports that the new trace state variable @var{name} is created with
29057 initial value @var{initial}.
29059 @item =tsv-deleted,name=@var{name}
29060 @itemx =tsv-deleted
29061 Reports that the trace state variable @var{name} is deleted or all
29062 trace state variables are deleted.
29064 @item =tsv-modified,name=@var{name},initial=@var{initial}[,current=@var{current}]
29065 Reports that the trace state variable @var{name} is modified with
29066 the initial value @var{initial}. The current value @var{current} of
29067 trace state variable is optional and is reported if the current
29068 value of trace state variable is known.
29070 @item =breakpoint-created,bkpt=@{...@}
29071 @itemx =breakpoint-modified,bkpt=@{...@}
29072 @itemx =breakpoint-deleted,id=@var{number}
29073 Reports that a breakpoint was created, modified, or deleted,
29074 respectively. Only user-visible breakpoints are reported to the MI
29077 The @var{bkpt} argument is of the same form as returned by the various
29078 breakpoint commands; @xref{GDB/MI Breakpoint Commands}. The
29079 @var{number} is the ordinal number of the breakpoint.
29081 Note that if a breakpoint is emitted in the result record of a
29082 command, then it will not also be emitted in an async record.
29084 @item =record-started,thread-group="@var{id}",method="@var{method}"[,format="@var{format}"]
29085 @itemx =record-stopped,thread-group="@var{id}"
29086 Execution log recording was either started or stopped on an
29087 inferior. The @var{id} is the @value{GDBN} identifier of the thread
29088 group corresponding to the affected inferior.
29090 The @var{method} field indicates the method used to record execution. If the
29091 method in use supports multiple recording formats, @var{format} will be present
29092 and contain the currently used format. @xref{Process Record and Replay},
29093 for existing method and format values.
29095 @item =cmd-param-changed,param=@var{param},value=@var{value}
29096 Reports that a parameter of the command @code{set @var{param}} is
29097 changed to @var{value}. In the multi-word @code{set} command,
29098 the @var{param} is the whole parameter list to @code{set} command.
29099 For example, In command @code{set check type on}, @var{param}
29100 is @code{check type} and @var{value} is @code{on}.
29102 @item =memory-changed,thread-group=@var{id},addr=@var{addr},len=@var{len}[,type="code"]
29103 Reports that bytes from @var{addr} to @var{data} + @var{len} were
29104 written in an inferior. The @var{id} is the identifier of the
29105 thread group corresponding to the affected inferior. The optional
29106 @code{type="code"} part is reported if the memory written to holds
29110 @node GDB/MI Breakpoint Information
29111 @subsection @sc{gdb/mi} Breakpoint Information
29113 When @value{GDBN} reports information about a breakpoint, a
29114 tracepoint, a watchpoint, or a catchpoint, it uses a tuple with the
29119 The breakpoint number.
29122 The type of the breakpoint. For ordinary breakpoints this will be
29123 @samp{breakpoint}, but many values are possible.
29126 If the type of the breakpoint is @samp{catchpoint}, then this
29127 indicates the exact type of catchpoint.
29130 This is the breakpoint disposition---either @samp{del}, meaning that
29131 the breakpoint will be deleted at the next stop, or @samp{keep},
29132 meaning that the breakpoint will not be deleted.
29135 This indicates whether the breakpoint is enabled, in which case the
29136 value is @samp{y}, or disabled, in which case the value is @samp{n}.
29137 Note that this is not the same as the field @code{enable}.
29140 The address of the breakpoint. This may be a hexidecimal number,
29141 giving the address; or the string @samp{<PENDING>}, for a pending
29142 breakpoint; or the string @samp{<MULTIPLE>}, for a breakpoint with
29143 multiple locations. This field will not be present if no address can
29144 be determined. For example, a watchpoint does not have an address.
29147 Optional field containing any flags related to the address. These flags are
29148 architecture-dependent; see @ref{Architectures} for their meaning for a
29152 If known, the function in which the breakpoint appears.
29153 If not known, this field is not present.
29156 The name of the source file which contains this function, if known.
29157 If not known, this field is not present.
29160 The full file name of the source file which contains this function, if
29161 known. If not known, this field is not present.
29164 The line number at which this breakpoint appears, if known.
29165 If not known, this field is not present.
29168 If the source file is not known, this field may be provided. If
29169 provided, this holds the address of the breakpoint, possibly followed
29173 If this breakpoint is pending, this field is present and holds the
29174 text used to set the breakpoint, as entered by the user.
29177 Where this breakpoint's condition is evaluated, either @samp{host} or
29181 If this is a thread-specific breakpoint, then this identifies the
29182 thread in which the breakpoint can trigger.
29185 If this breakpoint is restricted to a particular Ada task, then this
29186 field will hold the task identifier.
29189 If the breakpoint is conditional, this is the condition expression.
29192 The ignore count of the breakpoint.
29195 The enable count of the breakpoint.
29197 @item traceframe-usage
29200 @item static-tracepoint-marker-string-id
29201 For a static tracepoint, the name of the static tracepoint marker.
29204 For a masked watchpoint, this is the mask.
29207 A tracepoint's pass count.
29209 @item original-location
29210 The location of the breakpoint as originally specified by the user.
29211 This field is optional.
29214 The number of times the breakpoint has been hit.
29217 This field is only given for tracepoints. This is either @samp{y},
29218 meaning that the tracepoint is installed, or @samp{n}, meaning that it
29222 Some extra data, the exact contents of which are type-dependent.
29225 This field is present if the breakpoint has multiple locations. It is also
29226 exceptionally present if the breakpoint is enabled and has a single, disabled
29229 The value is a list of locations. The format of a location is described below.
29233 A location in a multi-location breakpoint is represented as a tuple with the
29239 The location number as a dotted pair, like @samp{1.2}. The first digit is the
29240 number of the parent breakpoint. The second digit is the number of the
29241 location within that breakpoint.
29244 This indicates whether the location is enabled, in which case the
29245 value is @samp{y}, or disabled, in which case the value is @samp{n}.
29246 Note that this is not the same as the field @code{enable}.
29249 The address of this location as an hexidecimal number.
29252 Optional field containing any flags related to the address. These flags are
29253 architecture-dependent; see @ref{Architectures} for their meaning for a
29257 If known, the function in which the location appears.
29258 If not known, this field is not present.
29261 The name of the source file which contains this location, if known.
29262 If not known, this field is not present.
29265 The full file name of the source file which contains this location, if
29266 known. If not known, this field is not present.
29269 The line number at which this location appears, if known.
29270 If not known, this field is not present.
29272 @item thread-groups
29273 The thread groups this location is in.
29277 For example, here is what the output of @code{-break-insert}
29278 (@pxref{GDB/MI Breakpoint Commands}) might be:
29281 -> -break-insert main
29282 <- ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
29283 enabled="y",addr="0x08048564",func="main",file="myprog.c",
29284 fullname="/home/nickrob/myprog.c",line="68",thread-groups=["i1"],
29289 @node GDB/MI Frame Information
29290 @subsection @sc{gdb/mi} Frame Information
29292 Response from many MI commands includes an information about stack
29293 frame. This information is a tuple that may have the following
29298 The level of the stack frame. The innermost frame has the level of
29299 zero. This field is always present.
29302 The name of the function corresponding to the frame. This field may
29303 be absent if @value{GDBN} is unable to determine the function name.
29306 The code address for the frame. This field is always present.
29309 Optional field containing any flags related to the address. These flags are
29310 architecture-dependent; see @ref{Architectures} for their meaning for a
29314 The name of the source files that correspond to the frame's code
29315 address. This field may be absent.
29318 The source line corresponding to the frames' code address. This field
29322 The name of the binary file (either executable or shared library) the
29323 corresponds to the frame's code address. This field may be absent.
29327 @node GDB/MI Thread Information
29328 @subsection @sc{gdb/mi} Thread Information
29330 Whenever @value{GDBN} has to report an information about a thread, it
29331 uses a tuple with the following fields. The fields are always present unless
29336 The global numeric id assigned to the thread by @value{GDBN}.
29339 The target-specific string identifying the thread.
29342 Additional information about the thread provided by the target.
29343 It is supposed to be human-readable and not interpreted by the
29344 frontend. This field is optional.
29347 The name of the thread. If the user specified a name using the
29348 @code{thread name} command, then this name is given. Otherwise, if
29349 @value{GDBN} can extract the thread name from the target, then that
29350 name is given. If @value{GDBN} cannot find the thread name, then this
29354 The execution state of the thread, either @samp{stopped} or @samp{running},
29355 depending on whether the thread is presently running.
29358 The stack frame currently executing in the thread. This field is only present
29359 if the thread is stopped. Its format is documented in
29360 @ref{GDB/MI Frame Information}.
29363 The value of this field is an integer number of the processor core the
29364 thread was last seen on. This field is optional.
29367 @node GDB/MI Ada Exception Information
29368 @subsection @sc{gdb/mi} Ada Exception Information
29370 Whenever a @code{*stopped} record is emitted because the program
29371 stopped after hitting an exception catchpoint (@pxref{Set Catchpoints}),
29372 @value{GDBN} provides the name of the exception that was raised via
29373 the @code{exception-name} field. Also, for exceptions that were raised
29374 with an exception message, @value{GDBN} provides that message via
29375 the @code{exception-message} field.
29377 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
29378 @node GDB/MI Simple Examples
29379 @section Simple Examples of @sc{gdb/mi} Interaction
29380 @cindex @sc{gdb/mi}, simple examples
29382 This subsection presents several simple examples of interaction using
29383 the @sc{gdb/mi} interface. In these examples, @samp{->} means that the
29384 following line is passed to @sc{gdb/mi} as input, while @samp{<-} means
29385 the output received from @sc{gdb/mi}.
29387 Note the line breaks shown in the examples are here only for
29388 readability, they don't appear in the real output.
29390 @subheading Setting a Breakpoint
29392 Setting a breakpoint generates synchronous output which contains detailed
29393 information of the breakpoint.
29396 -> -break-insert main
29397 <- ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
29398 enabled="y",addr="0x08048564",func="main",file="myprog.c",
29399 fullname="/home/nickrob/myprog.c",line="68",thread-groups=["i1"],
29404 @subheading Program Execution
29406 Program execution generates asynchronous records and MI gives the
29407 reason that execution stopped.
29413 <- *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
29414 frame=@{addr="0x08048564",func="main",
29415 args=[@{name="argc",value="1"@},@{name="argv",value="0xbfc4d4d4"@}],
29416 file="myprog.c",fullname="/home/nickrob/myprog.c",line="68",
29417 arch="i386:x86_64"@}
29422 <- *stopped,reason="exited-normally"
29426 @subheading Quitting @value{GDBN}
29428 Quitting @value{GDBN} just prints the result class @samp{^exit}.
29436 Please note that @samp{^exit} is printed immediately, but it might
29437 take some time for @value{GDBN} to actually exit. During that time, @value{GDBN}
29438 performs necessary cleanups, including killing programs being debugged
29439 or disconnecting from debug hardware, so the frontend should wait till
29440 @value{GDBN} exits and should only forcibly kill @value{GDBN} if it
29441 fails to exit in reasonable time.
29443 @subheading A Bad Command
29445 Here's what happens if you pass a non-existent command:
29449 <- ^error,msg="Undefined MI command: rubbish"
29454 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
29455 @node GDB/MI Command Description Format
29456 @section @sc{gdb/mi} Command Description Format
29458 The remaining sections describe blocks of commands. Each block of
29459 commands is laid out in a fashion similar to this section.
29461 @subheading Motivation
29463 The motivation for this collection of commands.
29465 @subheading Introduction
29467 A brief introduction to this collection of commands as a whole.
29469 @subheading Commands
29471 For each command in the block, the following is described:
29473 @subsubheading Synopsis
29476 -command @var{args}@dots{}
29479 @subsubheading Result
29481 @subsubheading @value{GDBN} Command
29483 The corresponding @value{GDBN} CLI command(s), if any.
29485 @subsubheading Example
29487 Example(s) formatted for readability. Some of the described commands have
29488 not been implemented yet and these are labeled N.A.@: (not available).
29491 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
29492 @node GDB/MI Breakpoint Commands
29493 @section @sc{gdb/mi} Breakpoint Commands
29495 @cindex breakpoint commands for @sc{gdb/mi}
29496 @cindex @sc{gdb/mi}, breakpoint commands
29497 This section documents @sc{gdb/mi} commands for manipulating
29500 @subheading The @code{-break-after} Command
29501 @findex -break-after
29503 @subsubheading Synopsis
29506 -break-after @var{number} @var{count}
29509 The breakpoint number @var{number} is not in effect until it has been
29510 hit @var{count} times. To see how this is reflected in the output of
29511 the @samp{-break-list} command, see the description of the
29512 @samp{-break-list} command below.
29514 @subsubheading @value{GDBN} Command
29516 The corresponding @value{GDBN} command is @samp{ignore}.
29518 @subsubheading Example
29523 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
29524 enabled="y",addr="0x000100d0",func="main",file="hello.c",
29525 fullname="/home/foo/hello.c",line="5",thread-groups=["i1"],
29533 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
29534 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29535 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29536 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29537 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29538 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29539 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29540 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
29541 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
29542 line="5",thread-groups=["i1"],times="0",ignore="3"@}]@}
29547 @subheading The @code{-break-catch} Command
29548 @findex -break-catch
29551 @subheading The @code{-break-commands} Command
29552 @findex -break-commands
29554 @subsubheading Synopsis
29557 -break-commands @var{number} [ @var{command1} ... @var{commandN} ]
29560 Specifies the CLI commands that should be executed when breakpoint
29561 @var{number} is hit. The parameters @var{command1} to @var{commandN}
29562 are the commands. If no command is specified, any previously-set
29563 commands are cleared. @xref{Break Commands}. Typical use of this
29564 functionality is tracing a program, that is, printing of values of
29565 some variables whenever breakpoint is hit and then continuing.
29567 @subsubheading @value{GDBN} Command
29569 The corresponding @value{GDBN} command is @samp{commands}.
29571 @subsubheading Example
29576 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
29577 enabled="y",addr="0x000100d0",func="main",file="hello.c",
29578 fullname="/home/foo/hello.c",line="5",thread-groups=["i1"],
29581 -break-commands 1 "print v" "continue"
29586 @subheading The @code{-break-condition} Command
29587 @findex -break-condition
29589 @subsubheading Synopsis
29592 -break-condition @var{number} @var{expr}
29595 Breakpoint @var{number} will stop the program only if the condition in
29596 @var{expr} is true. The condition becomes part of the
29597 @samp{-break-list} output (see the description of the @samp{-break-list}
29600 @subsubheading @value{GDBN} Command
29602 The corresponding @value{GDBN} command is @samp{condition}.
29604 @subsubheading Example
29608 -break-condition 1 1
29612 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
29613 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29614 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29615 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29616 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29617 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29618 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29619 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
29620 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
29621 line="5",cond="1",thread-groups=["i1"],times="0",ignore="3"@}]@}
29625 @subheading The @code{-break-delete} Command
29626 @findex -break-delete
29628 @subsubheading Synopsis
29631 -break-delete ( @var{breakpoint} )+
29634 Delete the breakpoint(s) whose number(s) are specified in the argument
29635 list. This is obviously reflected in the breakpoint list.
29637 @subsubheading @value{GDBN} Command
29639 The corresponding @value{GDBN} command is @samp{delete}.
29641 @subsubheading Example
29649 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
29650 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29651 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29652 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29653 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29654 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29655 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29660 @subheading The @code{-break-disable} Command
29661 @findex -break-disable
29663 @subsubheading Synopsis
29666 -break-disable ( @var{breakpoint} )+
29669 Disable the named @var{breakpoint}(s). The field @samp{enabled} in the
29670 break list is now set to @samp{n} for the named @var{breakpoint}(s).
29672 @subsubheading @value{GDBN} Command
29674 The corresponding @value{GDBN} command is @samp{disable}.
29676 @subsubheading Example
29684 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
29685 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29686 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29687 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29688 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29689 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29690 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29691 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="n",
29692 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
29693 line="5",thread-groups=["i1"],times="0"@}]@}
29697 @subheading The @code{-break-enable} Command
29698 @findex -break-enable
29700 @subsubheading Synopsis
29703 -break-enable ( @var{breakpoint} )+
29706 Enable (previously disabled) @var{breakpoint}(s).
29708 @subsubheading @value{GDBN} Command
29710 The corresponding @value{GDBN} command is @samp{enable}.
29712 @subsubheading Example
29720 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
29721 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29722 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29723 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29724 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29725 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29726 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29727 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
29728 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
29729 line="5",thread-groups=["i1"],times="0"@}]@}
29733 @subheading The @code{-break-info} Command
29734 @findex -break-info
29736 @subsubheading Synopsis
29739 -break-info @var{breakpoint}
29743 Get information about a single breakpoint.
29745 The result is a table of breakpoints. @xref{GDB/MI Breakpoint
29746 Information}, for details on the format of each breakpoint in the
29749 @subsubheading @value{GDBN} Command
29751 The corresponding @value{GDBN} command is @samp{info break @var{breakpoint}}.
29753 @subsubheading Example
29756 @subheading The @code{-break-insert} Command
29757 @findex -break-insert
29758 @anchor{-break-insert}
29760 @subsubheading Synopsis
29763 -break-insert [ -t ] [ -h ] [ -f ] [ -d ] [ -a ]
29764 [ -c @var{condition} ] [ -i @var{ignore-count} ]
29765 [ -p @var{thread-id} ] [ @var{location} ]
29769 If specified, @var{location}, can be one of:
29772 @item linespec location
29773 A linespec location. @xref{Linespec Locations}.
29775 @item explicit location
29776 An explicit location. @sc{gdb/mi} explicit locations are
29777 analogous to the CLI's explicit locations using the option names
29778 listed below. @xref{Explicit Locations}.
29781 @item --source @var{filename}
29782 The source file name of the location. This option requires the use
29783 of either @samp{--function} or @samp{--line}.
29785 @item --function @var{function}
29786 The name of a function or method.
29788 @item --label @var{label}
29789 The name of a label.
29791 @item --line @var{lineoffset}
29792 An absolute or relative line offset from the start of the location.
29795 @item address location
29796 An address location, *@var{address}. @xref{Address Locations}.
29800 The possible optional parameters of this command are:
29804 Insert a temporary breakpoint.
29806 Insert a hardware breakpoint.
29808 If @var{location} cannot be parsed (for example if it
29809 refers to unknown files or functions), create a pending
29810 breakpoint. Without this flag, @value{GDBN} will report
29811 an error, and won't create a breakpoint, if @var{location}
29814 Create a disabled breakpoint.
29816 Create a tracepoint. @xref{Tracepoints}. When this parameter
29817 is used together with @samp{-h}, a fast tracepoint is created.
29818 @item -c @var{condition}
29819 Make the breakpoint conditional on @var{condition}.
29820 @item -i @var{ignore-count}
29821 Initialize the @var{ignore-count}.
29822 @item -p @var{thread-id}
29823 Restrict the breakpoint to the thread with the specified global
29827 @subsubheading Result
29829 @xref{GDB/MI Breakpoint Information}, for details on the format of the
29830 resulting breakpoint.
29832 Note: this format is open to change.
29833 @c An out-of-band breakpoint instead of part of the result?
29835 @subsubheading @value{GDBN} Command
29837 The corresponding @value{GDBN} commands are @samp{break}, @samp{tbreak},
29838 @samp{hbreak}, and @samp{thbreak}. @c and @samp{rbreak}.
29840 @subsubheading Example
29845 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",
29846 fullname="/home/foo/recursive2.c,line="4",thread-groups=["i1"],
29849 -break-insert -t foo
29850 ^done,bkpt=@{number="2",addr="0x00010774",file="recursive2.c",
29851 fullname="/home/foo/recursive2.c,line="11",thread-groups=["i1"],
29855 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
29856 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29857 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29858 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29859 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29860 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29861 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29862 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
29863 addr="0x0001072c", func="main",file="recursive2.c",
29864 fullname="/home/foo/recursive2.c,"line="4",thread-groups=["i1"],
29866 bkpt=@{number="2",type="breakpoint",disp="del",enabled="y",
29867 addr="0x00010774",func="foo",file="recursive2.c",
29868 fullname="/home/foo/recursive2.c",line="11",thread-groups=["i1"],
29871 @c -break-insert -r foo.*
29872 @c ~int foo(int, int);
29873 @c ^done,bkpt=@{number="3",addr="0x00010774",file="recursive2.c,
29874 @c "fullname="/home/foo/recursive2.c",line="11",thread-groups=["i1"],
29879 @subheading The @code{-dprintf-insert} Command
29880 @findex -dprintf-insert
29882 @subsubheading Synopsis
29885 -dprintf-insert [ -t ] [ -f ] [ -d ]
29886 [ -c @var{condition} ] [ -i @var{ignore-count} ]
29887 [ -p @var{thread-id} ] [ @var{location} ] [ @var{format} ]
29892 If supplied, @var{location} may be specified the same way as for
29893 the @code{-break-insert} command. @xref{-break-insert}.
29895 The possible optional parameters of this command are:
29899 Insert a temporary breakpoint.
29901 If @var{location} cannot be parsed (for example, if it
29902 refers to unknown files or functions), create a pending
29903 breakpoint. Without this flag, @value{GDBN} will report
29904 an error, and won't create a breakpoint, if @var{location}
29907 Create a disabled breakpoint.
29908 @item -c @var{condition}
29909 Make the breakpoint conditional on @var{condition}.
29910 @item -i @var{ignore-count}
29911 Set the ignore count of the breakpoint (@pxref{Conditions, ignore count})
29912 to @var{ignore-count}.
29913 @item -p @var{thread-id}
29914 Restrict the breakpoint to the thread with the specified global
29918 @subsubheading Result
29920 @xref{GDB/MI Breakpoint Information}, for details on the format of the
29921 resulting breakpoint.
29923 @c An out-of-band breakpoint instead of part of the result?
29925 @subsubheading @value{GDBN} Command
29927 The corresponding @value{GDBN} command is @samp{dprintf}.
29929 @subsubheading Example
29933 4-dprintf-insert foo "At foo entry\n"
29934 4^done,bkpt=@{number="1",type="dprintf",disp="keep",enabled="y",
29935 addr="0x000000000040061b",func="foo",file="mi-dprintf.c",
29936 fullname="mi-dprintf.c",line="25",thread-groups=["i1"],
29937 times="0",script=@{"printf \"At foo entry\\n\"","continue"@},
29938 original-location="foo"@}
29940 5-dprintf-insert 26 "arg=%d, g=%d\n" arg g
29941 5^done,bkpt=@{number="2",type="dprintf",disp="keep",enabled="y",
29942 addr="0x000000000040062a",func="foo",file="mi-dprintf.c",
29943 fullname="mi-dprintf.c",line="26",thread-groups=["i1"],
29944 times="0",script=@{"printf \"arg=%d, g=%d\\n\", arg, g","continue"@},
29945 original-location="mi-dprintf.c:26"@}
29949 @subheading The @code{-break-list} Command
29950 @findex -break-list
29952 @subsubheading Synopsis
29958 Displays the list of inserted breakpoints, showing the following fields:
29962 number of the breakpoint
29964 type of the breakpoint: @samp{breakpoint} or @samp{watchpoint}
29966 should the breakpoint be deleted or disabled when it is hit: @samp{keep}
29969 is the breakpoint enabled or no: @samp{y} or @samp{n}
29971 memory location at which the breakpoint is set
29973 logical location of the breakpoint, expressed by function name, file
29975 @item Thread-groups
29976 list of thread groups to which this breakpoint applies
29978 number of times the breakpoint has been hit
29981 If there are no breakpoints or watchpoints, the @code{BreakpointTable}
29982 @code{body} field is an empty list.
29984 @subsubheading @value{GDBN} Command
29986 The corresponding @value{GDBN} command is @samp{info break}.
29988 @subsubheading Example
29993 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
29994 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29995 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29996 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29997 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29998 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29999 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
30000 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
30001 addr="0x000100d0",func="main",file="hello.c",line="5",thread-groups=["i1"],
30003 bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
30004 addr="0x00010114",func="foo",file="hello.c",fullname="/home/foo/hello.c",
30005 line="13",thread-groups=["i1"],times="0"@}]@}
30009 Here's an example of the result when there are no breakpoints:
30014 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
30015 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
30016 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
30017 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
30018 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
30019 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
30020 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
30025 @subheading The @code{-break-passcount} Command
30026 @findex -break-passcount
30028 @subsubheading Synopsis
30031 -break-passcount @var{tracepoint-number} @var{passcount}
30034 Set the passcount for tracepoint @var{tracepoint-number} to
30035 @var{passcount}. If the breakpoint referred to by @var{tracepoint-number}
30036 is not a tracepoint, error is emitted. This corresponds to CLI
30037 command @samp{passcount}.
30039 @subheading The @code{-break-watch} Command
30040 @findex -break-watch
30042 @subsubheading Synopsis
30045 -break-watch [ -a | -r ]
30048 Create a watchpoint. With the @samp{-a} option it will create an
30049 @dfn{access} watchpoint, i.e., a watchpoint that triggers either on a
30050 read from or on a write to the memory location. With the @samp{-r}
30051 option, the watchpoint created is a @dfn{read} watchpoint, i.e., it will
30052 trigger only when the memory location is accessed for reading. Without
30053 either of the options, the watchpoint created is a regular watchpoint,
30054 i.e., it will trigger when the memory location is accessed for writing.
30055 @xref{Set Watchpoints, , Setting Watchpoints}.
30057 Note that @samp{-break-list} will report a single list of watchpoints and
30058 breakpoints inserted.
30060 @subsubheading @value{GDBN} Command
30062 The corresponding @value{GDBN} commands are @samp{watch}, @samp{awatch}, and
30065 @subsubheading Example
30067 Setting a watchpoint on a variable in the @code{main} function:
30072 ^done,wpt=@{number="2",exp="x"@}
30077 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="x"@},
30078 value=@{old="-268439212",new="55"@},
30079 frame=@{func="main",args=[],file="recursive2.c",
30080 fullname="/home/foo/bar/recursive2.c",line="5",arch="i386:x86_64"@}
30084 Setting a watchpoint on a variable local to a function. @value{GDBN} will stop
30085 the program execution twice: first for the variable changing value, then
30086 for the watchpoint going out of scope.
30091 ^done,wpt=@{number="5",exp="C"@}
30096 *stopped,reason="watchpoint-trigger",
30097 wpt=@{number="5",exp="C"@},value=@{old="-276895068",new="3"@},
30098 frame=@{func="callee4",args=[],
30099 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
30100 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13",
30101 arch="i386:x86_64"@}
30106 *stopped,reason="watchpoint-scope",wpnum="5",
30107 frame=@{func="callee3",args=[@{name="strarg",
30108 value="0x11940 \"A string argument.\""@}],
30109 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
30110 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18",
30111 arch="i386:x86_64"@}
30115 Listing breakpoints and watchpoints, at different points in the program
30116 execution. Note that once the watchpoint goes out of scope, it is
30122 ^done,wpt=@{number="2",exp="C"@}
30125 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
30126 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
30127 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
30128 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
30129 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
30130 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
30131 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
30132 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
30133 addr="0x00010734",func="callee4",
30134 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
30135 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c"line="8",thread-groups=["i1"],
30137 bkpt=@{number="2",type="watchpoint",disp="keep",
30138 enabled="y",addr="",what="C",thread-groups=["i1"],times="0"@}]@}
30143 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="C"@},
30144 value=@{old="-276895068",new="3"@},
30145 frame=@{func="callee4",args=[],
30146 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
30147 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13",
30148 arch="i386:x86_64"@}
30151 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
30152 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
30153 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
30154 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
30155 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
30156 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
30157 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
30158 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
30159 addr="0x00010734",func="callee4",
30160 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
30161 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",thread-groups=["i1"],
30163 bkpt=@{number="2",type="watchpoint",disp="keep",
30164 enabled="y",addr="",what="C",thread-groups=["i1"],times="-5"@}]@}
30168 ^done,reason="watchpoint-scope",wpnum="2",
30169 frame=@{func="callee3",args=[@{name="strarg",
30170 value="0x11940 \"A string argument.\""@}],
30171 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
30172 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18",
30173 arch="i386:x86_64"@}
30176 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
30177 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
30178 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
30179 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
30180 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
30181 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
30182 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
30183 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
30184 addr="0x00010734",func="callee4",
30185 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
30186 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",
30187 thread-groups=["i1"],times="1"@}]@}
30192 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30193 @node GDB/MI Catchpoint Commands
30194 @section @sc{gdb/mi} Catchpoint Commands
30196 This section documents @sc{gdb/mi} commands for manipulating
30200 * Shared Library GDB/MI Catchpoint Commands::
30201 * Ada Exception GDB/MI Catchpoint Commands::
30202 * C++ Exception GDB/MI Catchpoint Commands::
30205 @node Shared Library GDB/MI Catchpoint Commands
30206 @subsection Shared Library @sc{gdb/mi} Catchpoints
30208 @subheading The @code{-catch-load} Command
30209 @findex -catch-load
30211 @subsubheading Synopsis
30214 -catch-load [ -t ] [ -d ] @var{regexp}
30217 Add a catchpoint for library load events. If the @samp{-t} option is used,
30218 the catchpoint is a temporary one (@pxref{Set Breaks, ,Setting
30219 Breakpoints}). If the @samp{-d} option is used, the catchpoint is created
30220 in a disabled state. The @samp{regexp} argument is a regular
30221 expression used to match the name of the loaded library.
30224 @subsubheading @value{GDBN} Command
30226 The corresponding @value{GDBN} command is @samp{catch load}.
30228 @subsubheading Example
30231 -catch-load -t foo.so
30232 ^done,bkpt=@{number="1",type="catchpoint",disp="del",enabled="y",
30233 what="load of library matching foo.so",catch-type="load",times="0"@}
30238 @subheading The @code{-catch-unload} Command
30239 @findex -catch-unload
30241 @subsubheading Synopsis
30244 -catch-unload [ -t ] [ -d ] @var{regexp}
30247 Add a catchpoint for library unload events. If the @samp{-t} option is
30248 used, the catchpoint is a temporary one (@pxref{Set Breaks, ,Setting
30249 Breakpoints}). If the @samp{-d} option is used, the catchpoint is
30250 created in a disabled state. The @samp{regexp} argument is a regular
30251 expression used to match the name of the unloaded library.
30253 @subsubheading @value{GDBN} Command
30255 The corresponding @value{GDBN} command is @samp{catch unload}.
30257 @subsubheading Example
30260 -catch-unload -d bar.so
30261 ^done,bkpt=@{number="2",type="catchpoint",disp="keep",enabled="n",
30262 what="load of library matching bar.so",catch-type="unload",times="0"@}
30266 @node Ada Exception GDB/MI Catchpoint Commands
30267 @subsection Ada Exception @sc{gdb/mi} Catchpoints
30269 The following @sc{gdb/mi} commands can be used to create catchpoints
30270 that stop the execution when Ada exceptions are being raised.
30272 @subheading The @code{-catch-assert} Command
30273 @findex -catch-assert
30275 @subsubheading Synopsis
30278 -catch-assert [ -c @var{condition}] [ -d ] [ -t ]
30281 Add a catchpoint for failed Ada assertions.
30283 The possible optional parameters for this command are:
30286 @item -c @var{condition}
30287 Make the catchpoint conditional on @var{condition}.
30289 Create a disabled catchpoint.
30291 Create a temporary catchpoint.
30294 @subsubheading @value{GDBN} Command
30296 The corresponding @value{GDBN} command is @samp{catch assert}.
30298 @subsubheading Example
30302 ^done,bkptno="5",bkpt=@{number="5",type="breakpoint",disp="keep",
30303 enabled="y",addr="0x0000000000404888",what="failed Ada assertions",
30304 thread-groups=["i1"],times="0",
30305 original-location="__gnat_debug_raise_assert_failure"@}
30309 @subheading The @code{-catch-exception} Command
30310 @findex -catch-exception
30312 @subsubheading Synopsis
30315 -catch-exception [ -c @var{condition}] [ -d ] [ -e @var{exception-name} ]
30319 Add a catchpoint stopping when Ada exceptions are raised.
30320 By default, the command stops the program when any Ada exception
30321 gets raised. But it is also possible, by using some of the
30322 optional parameters described below, to create more selective
30325 The possible optional parameters for this command are:
30328 @item -c @var{condition}
30329 Make the catchpoint conditional on @var{condition}.
30331 Create a disabled catchpoint.
30332 @item -e @var{exception-name}
30333 Only stop when @var{exception-name} is raised. This option cannot
30334 be used combined with @samp{-u}.
30336 Create a temporary catchpoint.
30338 Stop only when an unhandled exception gets raised. This option
30339 cannot be used combined with @samp{-e}.
30342 @subsubheading @value{GDBN} Command
30344 The corresponding @value{GDBN} commands are @samp{catch exception}
30345 and @samp{catch exception unhandled}.
30347 @subsubheading Example
30350 -catch-exception -e Program_Error
30351 ^done,bkptno="4",bkpt=@{number="4",type="breakpoint",disp="keep",
30352 enabled="y",addr="0x0000000000404874",
30353 what="`Program_Error' Ada exception", thread-groups=["i1"],
30354 times="0",original-location="__gnat_debug_raise_exception"@}
30358 @subheading The @code{-catch-handlers} Command
30359 @findex -catch-handlers
30361 @subsubheading Synopsis
30364 -catch-handlers [ -c @var{condition}] [ -d ] [ -e @var{exception-name} ]
30368 Add a catchpoint stopping when Ada exceptions are handled.
30369 By default, the command stops the program when any Ada exception
30370 gets handled. But it is also possible, by using some of the
30371 optional parameters described below, to create more selective
30374 The possible optional parameters for this command are:
30377 @item -c @var{condition}
30378 Make the catchpoint conditional on @var{condition}.
30380 Create a disabled catchpoint.
30381 @item -e @var{exception-name}
30382 Only stop when @var{exception-name} is handled.
30384 Create a temporary catchpoint.
30387 @subsubheading @value{GDBN} Command
30389 The corresponding @value{GDBN} command is @samp{catch handlers}.
30391 @subsubheading Example
30394 -catch-handlers -e Constraint_Error
30395 ^done,bkptno="4",bkpt=@{number="4",type="breakpoint",disp="keep",
30396 enabled="y",addr="0x0000000000402f68",
30397 what="`Constraint_Error' Ada exception handlers",thread-groups=["i1"],
30398 times="0",original-location="__gnat_begin_handler"@}
30402 @node C++ Exception GDB/MI Catchpoint Commands
30403 @subsection C@t{++} Exception @sc{gdb/mi} Catchpoints
30405 The following @sc{gdb/mi} commands can be used to create catchpoints
30406 that stop the execution when C@t{++} exceptions are being throw, rethrown,
30409 @subheading The @code{-catch-throw} Command
30410 @findex -catch-throw
30412 @subsubheading Synopsis
30415 -catch-throw [ -t ] [ -r @var{regexp}]
30418 Stop when the debuggee throws a C@t{++} exception. If @var{regexp} is
30419 given, then only exceptions whose type matches the regular expression
30422 If @samp{-t} is given, then the catchpoint is enabled only for one
30423 stop, the catchpoint is automatically deleted after stopping once for
30426 @subsubheading @value{GDBN} Command
30428 The corresponding @value{GDBN} commands are @samp{catch throw}
30429 and @samp{tcatch throw} (@pxref{Set Catchpoints}).
30431 @subsubheading Example
30434 -catch-throw -r exception_type
30435 ^done,bkpt=@{number="1",type="catchpoint",disp="keep",enabled="y",
30436 what="exception throw",catch-type="throw",
30437 thread-groups=["i1"],
30438 regexp="exception_type",times="0"@}
30444 ~"Catchpoint 1 (exception thrown), 0x00007ffff7ae00ed
30445 in __cxa_throw () from /lib64/libstdc++.so.6\n"
30446 *stopped,bkptno="1",reason="breakpoint-hit",disp="keep",
30447 frame=@{addr="0x00007ffff7ae00ed",func="__cxa_throw",
30448 args=[],from="/lib64/libstdc++.so.6",arch="i386:x86-64"@},
30449 thread-id="1",stopped-threads="all",core="6"
30453 @subheading The @code{-catch-rethrow} Command
30454 @findex -catch-rethrow
30456 @subsubheading Synopsis
30459 -catch-rethrow [ -t ] [ -r @var{regexp}]
30462 Stop when a C@t{++} exception is re-thrown. If @var{regexp} is given,
30463 then only exceptions whose type matches the regular expression will be
30466 If @samp{-t} is given, then the catchpoint is enabled only for one
30467 stop, the catchpoint is automatically deleted after the first event is
30470 @subsubheading @value{GDBN} Command
30472 The corresponding @value{GDBN} commands are @samp{catch rethrow}
30473 and @samp{tcatch rethrow} (@pxref{Set Catchpoints}).
30475 @subsubheading Example
30478 -catch-rethrow -r exception_type
30479 ^done,bkpt=@{number="1",type="catchpoint",disp="keep",enabled="y",
30480 what="exception rethrow",catch-type="rethrow",
30481 thread-groups=["i1"],
30482 regexp="exception_type",times="0"@}
30488 ~"Catchpoint 1 (exception rethrown), 0x00007ffff7ae00ed
30489 in __cxa_rethrow () from /lib64/libstdc++.so.6\n"
30490 *stopped,bkptno="1",reason="breakpoint-hit",disp="keep",
30491 frame=@{addr="0x00007ffff7ae00ed",func="__cxa_rethrow",
30492 args=[],from="/lib64/libstdc++.so.6",arch="i386:x86-64"@},
30493 thread-id="1",stopped-threads="all",core="6"
30497 @subheading The @code{-catch-catch} Command
30498 @findex -catch-catch
30500 @subsubheading Synopsis
30503 -catch-catch [ -t ] [ -r @var{regexp}]
30506 Stop when the debuggee catches a C@t{++} exception. If @var{regexp}
30507 is given, then only exceptions whose type matches the regular
30508 expression will be caught.
30510 If @samp{-t} is given, then the catchpoint is enabled only for one
30511 stop, the catchpoint is automatically deleted after the first event is
30514 @subsubheading @value{GDBN} Command
30516 The corresponding @value{GDBN} commands are @samp{catch catch}
30517 and @samp{tcatch catch} (@pxref{Set Catchpoints}).
30519 @subsubheading Example
30522 -catch-catch -r exception_type
30523 ^done,bkpt=@{number="1",type="catchpoint",disp="keep",enabled="y",
30524 what="exception catch",catch-type="catch",
30525 thread-groups=["i1"],
30526 regexp="exception_type",times="0"@}
30532 ~"Catchpoint 1 (exception caught), 0x00007ffff7ae00ed
30533 in __cxa_begin_catch () from /lib64/libstdc++.so.6\n"
30534 *stopped,bkptno="1",reason="breakpoint-hit",disp="keep",
30535 frame=@{addr="0x00007ffff7ae00ed",func="__cxa_begin_catch",
30536 args=[],from="/lib64/libstdc++.so.6",arch="i386:x86-64"@},
30537 thread-id="1",stopped-threads="all",core="6"
30541 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30542 @node GDB/MI Program Context
30543 @section @sc{gdb/mi} Program Context
30545 @subheading The @code{-exec-arguments} Command
30546 @findex -exec-arguments
30549 @subsubheading Synopsis
30552 -exec-arguments @var{args}
30555 Set the inferior program arguments, to be used in the next
30558 @subsubheading @value{GDBN} Command
30560 The corresponding @value{GDBN} command is @samp{set args}.
30562 @subsubheading Example
30566 -exec-arguments -v word
30573 @subheading The @code{-exec-show-arguments} Command
30574 @findex -exec-show-arguments
30576 @subsubheading Synopsis
30579 -exec-show-arguments
30582 Print the arguments of the program.
30584 @subsubheading @value{GDBN} Command
30586 The corresponding @value{GDBN} command is @samp{show args}.
30588 @subsubheading Example
30593 @subheading The @code{-environment-cd} Command
30594 @findex -environment-cd
30596 @subsubheading Synopsis
30599 -environment-cd @var{pathdir}
30602 Set @value{GDBN}'s working directory.
30604 @subsubheading @value{GDBN} Command
30606 The corresponding @value{GDBN} command is @samp{cd}.
30608 @subsubheading Example
30612 -environment-cd /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
30618 @subheading The @code{-environment-directory} Command
30619 @findex -environment-directory
30621 @subsubheading Synopsis
30624 -environment-directory [ -r ] [ @var{pathdir} ]+
30627 Add directories @var{pathdir} to beginning of search path for source files.
30628 If the @samp{-r} option is used, the search path is reset to the default
30629 search path. If directories @var{pathdir} are supplied in addition to the
30630 @samp{-r} option, the search path is first reset and then addition
30632 Multiple directories may be specified, separated by blanks. Specifying
30633 multiple directories in a single command
30634 results in the directories added to the beginning of the
30635 search path in the same order they were presented in the command.
30636 If blanks are needed as
30637 part of a directory name, double-quotes should be used around
30638 the name. In the command output, the path will show up separated
30639 by the system directory-separator character. The directory-separator
30640 character must not be used
30641 in any directory name.
30642 If no directories are specified, the current search path is displayed.
30644 @subsubheading @value{GDBN} Command
30646 The corresponding @value{GDBN} command is @samp{dir}.
30648 @subsubheading Example
30652 -environment-directory /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
30653 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
30655 -environment-directory ""
30656 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
30658 -environment-directory -r /home/jjohnstn/src/gdb /usr/src
30659 ^done,source-path="/home/jjohnstn/src/gdb:/usr/src:$cdir:$cwd"
30661 -environment-directory -r
30662 ^done,source-path="$cdir:$cwd"
30667 @subheading The @code{-environment-path} Command
30668 @findex -environment-path
30670 @subsubheading Synopsis
30673 -environment-path [ -r ] [ @var{pathdir} ]+
30676 Add directories @var{pathdir} to beginning of search path for object files.
30677 If the @samp{-r} option is used, the search path is reset to the original
30678 search path that existed at gdb start-up. If directories @var{pathdir} are
30679 supplied in addition to the
30680 @samp{-r} option, the search path is first reset and then addition
30682 Multiple directories may be specified, separated by blanks. Specifying
30683 multiple directories in a single command
30684 results in the directories added to the beginning of the
30685 search path in the same order they were presented in the command.
30686 If blanks are needed as
30687 part of a directory name, double-quotes should be used around
30688 the name. In the command output, the path will show up separated
30689 by the system directory-separator character. The directory-separator
30690 character must not be used
30691 in any directory name.
30692 If no directories are specified, the current path is displayed.
30695 @subsubheading @value{GDBN} Command
30697 The corresponding @value{GDBN} command is @samp{path}.
30699 @subsubheading Example
30704 ^done,path="/usr/bin"
30706 -environment-path /kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb /bin
30707 ^done,path="/kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb:/bin:/usr/bin"
30709 -environment-path -r /usr/local/bin
30710 ^done,path="/usr/local/bin:/usr/bin"
30715 @subheading The @code{-environment-pwd} Command
30716 @findex -environment-pwd
30718 @subsubheading Synopsis
30724 Show the current working directory.
30726 @subsubheading @value{GDBN} Command
30728 The corresponding @value{GDBN} command is @samp{pwd}.
30730 @subsubheading Example
30735 ^done,cwd="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb"
30739 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30740 @node GDB/MI Thread Commands
30741 @section @sc{gdb/mi} Thread Commands
30744 @subheading The @code{-thread-info} Command
30745 @findex -thread-info
30747 @subsubheading Synopsis
30750 -thread-info [ @var{thread-id} ]
30753 Reports information about either a specific thread, if the
30754 @var{thread-id} parameter is present, or about all threads.
30755 @var{thread-id} is the thread's global thread ID. When printing
30756 information about all threads, also reports the global ID of the
30759 @subsubheading @value{GDBN} Command
30761 The @samp{info thread} command prints the same information
30764 @subsubheading Result
30766 The result contains the following attributes:
30770 A list of threads. The format of the elements of the list is described in
30771 @ref{GDB/MI Thread Information}.
30773 @item current-thread-id
30774 The global id of the currently selected thread. This field is omitted if there
30775 is no selected thread (for example, when the selected inferior is not running,
30776 and therefore has no threads) or if a @var{thread-id} argument was passed to
30781 @subsubheading Example
30786 @{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
30787 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",
30788 args=[]@},state="running"@},
30789 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
30790 frame=@{level="0",addr="0x0804891f",func="foo",
30791 args=[@{name="i",value="10"@}],
30792 file="/tmp/a.c",fullname="/tmp/a.c",line="158",arch="i386:x86_64"@},
30793 state="running"@}],
30794 current-thread-id="1"
30798 @subheading The @code{-thread-list-ids} Command
30799 @findex -thread-list-ids
30801 @subsubheading Synopsis
30807 Produces a list of the currently known global @value{GDBN} thread ids.
30808 At the end of the list it also prints the total number of such
30811 This command is retained for historical reasons, the
30812 @code{-thread-info} command should be used instead.
30814 @subsubheading @value{GDBN} Command
30816 Part of @samp{info threads} supplies the same information.
30818 @subsubheading Example
30823 ^done,thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
30824 current-thread-id="1",number-of-threads="3"
30829 @subheading The @code{-thread-select} Command
30830 @findex -thread-select
30832 @subsubheading Synopsis
30835 -thread-select @var{thread-id}
30838 Make thread with global thread number @var{thread-id} the current
30839 thread. It prints the number of the new current thread, and the
30840 topmost frame for that thread.
30842 This command is deprecated in favor of explicitly using the
30843 @samp{--thread} option to each command.
30845 @subsubheading @value{GDBN} Command
30847 The corresponding @value{GDBN} command is @samp{thread}.
30849 @subsubheading Example
30856 *stopped,reason="end-stepping-range",thread-id="2",line="187",
30857 file="../../../devo/gdb/testsuite/gdb.threads/linux-dp.c"
30861 thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
30862 number-of-threads="3"
30865 ^done,new-thread-id="3",
30866 frame=@{level="0",func="vprintf",
30867 args=[@{name="format",value="0x8048e9c \"%*s%c %d %c\\n\""@},
30868 @{name="arg",value="0x2"@}],file="vprintf.c",line="31",arch="i386:x86_64"@}
30872 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30873 @node GDB/MI Ada Tasking Commands
30874 @section @sc{gdb/mi} Ada Tasking Commands
30876 @subheading The @code{-ada-task-info} Command
30877 @findex -ada-task-info
30879 @subsubheading Synopsis
30882 -ada-task-info [ @var{task-id} ]
30885 Reports information about either a specific Ada task, if the
30886 @var{task-id} parameter is present, or about all Ada tasks.
30888 @subsubheading @value{GDBN} Command
30890 The @samp{info tasks} command prints the same information
30891 about all Ada tasks (@pxref{Ada Tasks}).
30893 @subsubheading Result
30895 The result is a table of Ada tasks. The following columns are
30896 defined for each Ada task:
30900 This field exists only for the current thread. It has the value @samp{*}.
30903 The identifier that @value{GDBN} uses to refer to the Ada task.
30906 The identifier that the target uses to refer to the Ada task.
30909 The global thread identifier of the thread corresponding to the Ada
30912 This field should always exist, as Ada tasks are always implemented
30913 on top of a thread. But if @value{GDBN} cannot find this corresponding
30914 thread for any reason, the field is omitted.
30917 This field exists only when the task was created by another task.
30918 In this case, it provides the ID of the parent task.
30921 The base priority of the task.
30924 The current state of the task. For a detailed description of the
30925 possible states, see @ref{Ada Tasks}.
30928 The name of the task.
30932 @subsubheading Example
30936 ^done,tasks=@{nr_rows="3",nr_cols="8",
30937 hdr=[@{width="1",alignment="-1",col_name="current",colhdr=""@},
30938 @{width="3",alignment="1",col_name="id",colhdr="ID"@},
30939 @{width="9",alignment="1",col_name="task-id",colhdr="TID"@},
30940 @{width="4",alignment="1",col_name="thread-id",colhdr=""@},
30941 @{width="4",alignment="1",col_name="parent-id",colhdr="P-ID"@},
30942 @{width="3",alignment="1",col_name="priority",colhdr="Pri"@},
30943 @{width="22",alignment="-1",col_name="state",colhdr="State"@},
30944 @{width="1",alignment="2",col_name="name",colhdr="Name"@}],
30945 body=[@{current="*",id="1",task-id=" 644010",thread-id="1",priority="48",
30946 state="Child Termination Wait",name="main_task"@}]@}
30950 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30951 @node GDB/MI Program Execution
30952 @section @sc{gdb/mi} Program Execution
30954 These are the asynchronous commands which generate the out-of-band
30955 record @samp{*stopped}. Currently @value{GDBN} only really executes
30956 asynchronously with remote targets and this interaction is mimicked in
30959 @subheading The @code{-exec-continue} Command
30960 @findex -exec-continue
30962 @subsubheading Synopsis
30965 -exec-continue [--reverse] [--all|--thread-group N]
30968 Resumes the execution of the inferior program, which will continue
30969 to execute until it reaches a debugger stop event. If the
30970 @samp{--reverse} option is specified, execution resumes in reverse until
30971 it reaches a stop event. Stop events may include
30974 breakpoints or watchpoints
30976 signals or exceptions
30978 the end of the process (or its beginning under @samp{--reverse})
30980 the end or beginning of a replay log if one is being used.
30982 In all-stop mode (@pxref{All-Stop
30983 Mode}), may resume only one thread, or all threads, depending on the
30984 value of the @samp{scheduler-locking} variable. If @samp{--all} is
30985 specified, all threads (in all inferiors) will be resumed. The @samp{--all} option is
30986 ignored in all-stop mode. If the @samp{--thread-group} options is
30987 specified, then all threads in that thread group are resumed.
30989 @subsubheading @value{GDBN} Command
30991 The corresponding @value{GDBN} corresponding is @samp{continue}.
30993 @subsubheading Example
31000 *stopped,reason="breakpoint-hit",disp="keep",bkptno="2",frame=@{
31001 func="foo",args=[],file="hello.c",fullname="/home/foo/bar/hello.c",
31002 line="13",arch="i386:x86_64"@}
31007 @subheading The @code{-exec-finish} Command
31008 @findex -exec-finish
31010 @subsubheading Synopsis
31013 -exec-finish [--reverse]
31016 Resumes the execution of the inferior program until the current
31017 function is exited. Displays the results returned by the function.
31018 If the @samp{--reverse} option is specified, resumes the reverse
31019 execution of the inferior program until the point where current
31020 function was called.
31022 @subsubheading @value{GDBN} Command
31024 The corresponding @value{GDBN} command is @samp{finish}.
31026 @subsubheading Example
31028 Function returning @code{void}.
31035 *stopped,reason="function-finished",frame=@{func="main",args=[],
31036 file="hello.c",fullname="/home/foo/bar/hello.c",line="7",arch="i386:x86_64"@}
31040 Function returning other than @code{void}. The name of the internal
31041 @value{GDBN} variable storing the result is printed, together with the
31048 *stopped,reason="function-finished",frame=@{addr="0x000107b0",func="foo",
31049 args=[@{name="a",value="1"],@{name="b",value="9"@}@},
31050 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31051 arch="i386:x86_64"@},
31052 gdb-result-var="$1",return-value="0"
31057 @subheading The @code{-exec-interrupt} Command
31058 @findex -exec-interrupt
31060 @subsubheading Synopsis
31063 -exec-interrupt [--all|--thread-group N]
31066 Interrupts the background execution of the target. Note how the token
31067 associated with the stop message is the one for the execution command
31068 that has been interrupted. The token for the interrupt itself only
31069 appears in the @samp{^done} output. If the user is trying to
31070 interrupt a non-running program, an error message will be printed.
31072 Note that when asynchronous execution is enabled, this command is
31073 asynchronous just like other execution commands. That is, first the
31074 @samp{^done} response will be printed, and the target stop will be
31075 reported after that using the @samp{*stopped} notification.
31077 In non-stop mode, only the context thread is interrupted by default.
31078 All threads (in all inferiors) will be interrupted if the
31079 @samp{--all} option is specified. If the @samp{--thread-group}
31080 option is specified, all threads in that group will be interrupted.
31082 @subsubheading @value{GDBN} Command
31084 The corresponding @value{GDBN} command is @samp{interrupt}.
31086 @subsubheading Example
31097 111*stopped,signal-name="SIGINT",signal-meaning="Interrupt",
31098 frame=@{addr="0x00010140",func="foo",args=[],file="try.c",
31099 fullname="/home/foo/bar/try.c",line="13",arch="i386:x86_64"@}
31104 ^error,msg="mi_cmd_exec_interrupt: Inferior not executing."
31108 @subheading The @code{-exec-jump} Command
31111 @subsubheading Synopsis
31114 -exec-jump @var{location}
31117 Resumes execution of the inferior program at the location specified by
31118 parameter. @xref{Specify Location}, for a description of the
31119 different forms of @var{location}.
31121 @subsubheading @value{GDBN} Command
31123 The corresponding @value{GDBN} command is @samp{jump}.
31125 @subsubheading Example
31128 -exec-jump foo.c:10
31129 *running,thread-id="all"
31134 @subheading The @code{-exec-next} Command
31137 @subsubheading Synopsis
31140 -exec-next [--reverse]
31143 Resumes execution of the inferior program, stopping when the beginning
31144 of the next source line is reached.
31146 If the @samp{--reverse} option is specified, resumes reverse execution
31147 of the inferior program, stopping at the beginning of the previous
31148 source line. If you issue this command on the first line of a
31149 function, it will take you back to the caller of that function, to the
31150 source line where the function was called.
31153 @subsubheading @value{GDBN} Command
31155 The corresponding @value{GDBN} command is @samp{next}.
31157 @subsubheading Example
31163 *stopped,reason="end-stepping-range",line="8",file="hello.c"
31168 @subheading The @code{-exec-next-instruction} Command
31169 @findex -exec-next-instruction
31171 @subsubheading Synopsis
31174 -exec-next-instruction [--reverse]
31177 Executes one machine instruction. If the instruction is a function
31178 call, continues until the function returns. If the program stops at an
31179 instruction in the middle of a source line, the address will be
31182 If the @samp{--reverse} option is specified, resumes reverse execution
31183 of the inferior program, stopping at the previous instruction. If the
31184 previously executed instruction was a return from another function,
31185 it will continue to execute in reverse until the call to that function
31186 (from the current stack frame) is reached.
31188 @subsubheading @value{GDBN} Command
31190 The corresponding @value{GDBN} command is @samp{nexti}.
31192 @subsubheading Example
31196 -exec-next-instruction
31200 *stopped,reason="end-stepping-range",
31201 addr="0x000100d4",line="5",file="hello.c"
31206 @subheading The @code{-exec-return} Command
31207 @findex -exec-return
31209 @subsubheading Synopsis
31215 Makes current function return immediately. Doesn't execute the inferior.
31216 Displays the new current frame.
31218 @subsubheading @value{GDBN} Command
31220 The corresponding @value{GDBN} command is @samp{return}.
31222 @subsubheading Example
31226 200-break-insert callee4
31227 200^done,bkpt=@{number="1",addr="0x00010734",
31228 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",line="8"@}
31233 000*stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
31234 frame=@{func="callee4",args=[],
31235 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31236 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8",
31237 arch="i386:x86_64"@}
31243 111^done,frame=@{level="0",func="callee3",
31244 args=[@{name="strarg",
31245 value="0x11940 \"A string argument.\""@}],
31246 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31247 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18",
31248 arch="i386:x86_64"@}
31253 @subheading The @code{-exec-run} Command
31256 @subsubheading Synopsis
31259 -exec-run [ --all | --thread-group N ] [ --start ]
31262 Starts execution of the inferior from the beginning. The inferior
31263 executes until either a breakpoint is encountered or the program
31264 exits. In the latter case the output will include an exit code, if
31265 the program has exited exceptionally.
31267 When neither the @samp{--all} nor the @samp{--thread-group} option
31268 is specified, the current inferior is started. If the
31269 @samp{--thread-group} option is specified, it should refer to a thread
31270 group of type @samp{process}, and that thread group will be started.
31271 If the @samp{--all} option is specified, then all inferiors will be started.
31273 Using the @samp{--start} option instructs the debugger to stop
31274 the execution at the start of the inferior's main subprogram,
31275 following the same behavior as the @code{start} command
31276 (@pxref{Starting}).
31278 @subsubheading @value{GDBN} Command
31280 The corresponding @value{GDBN} command is @samp{run}.
31282 @subsubheading Examples
31287 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",line="4"@}
31292 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
31293 frame=@{func="main",args=[],file="recursive2.c",
31294 fullname="/home/foo/bar/recursive2.c",line="4",arch="i386:x86_64"@}
31299 Program exited normally:
31307 *stopped,reason="exited-normally"
31312 Program exited exceptionally:
31320 *stopped,reason="exited",exit-code="01"
31324 Another way the program can terminate is if it receives a signal such as
31325 @code{SIGINT}. In this case, @sc{gdb/mi} displays this:
31329 *stopped,reason="exited-signalled",signal-name="SIGINT",
31330 signal-meaning="Interrupt"
31334 @c @subheading -exec-signal
31337 @subheading The @code{-exec-step} Command
31340 @subsubheading Synopsis
31343 -exec-step [--reverse]
31346 Resumes execution of the inferior program, stopping when the beginning
31347 of the next source line is reached, if the next source line is not a
31348 function call. If it is, stop at the first instruction of the called
31349 function. If the @samp{--reverse} option is specified, resumes reverse
31350 execution of the inferior program, stopping at the beginning of the
31351 previously executed source line.
31353 @subsubheading @value{GDBN} Command
31355 The corresponding @value{GDBN} command is @samp{step}.
31357 @subsubheading Example
31359 Stepping into a function:
31365 *stopped,reason="end-stepping-range",
31366 frame=@{func="foo",args=[@{name="a",value="10"@},
31367 @{name="b",value="0"@}],file="recursive2.c",
31368 fullname="/home/foo/bar/recursive2.c",line="11",arch="i386:x86_64"@}
31378 *stopped,reason="end-stepping-range",line="14",file="recursive2.c"
31383 @subheading The @code{-exec-step-instruction} Command
31384 @findex -exec-step-instruction
31386 @subsubheading Synopsis
31389 -exec-step-instruction [--reverse]
31392 Resumes the inferior which executes one machine instruction. If the
31393 @samp{--reverse} option is specified, resumes reverse execution of the
31394 inferior program, stopping at the previously executed instruction.
31395 The output, once @value{GDBN} has stopped, will vary depending on
31396 whether we have stopped in the middle of a source line or not. In the
31397 former case, the address at which the program stopped will be printed
31400 @subsubheading @value{GDBN} Command
31402 The corresponding @value{GDBN} command is @samp{stepi}.
31404 @subsubheading Example
31408 -exec-step-instruction
31412 *stopped,reason="end-stepping-range",
31413 frame=@{func="foo",args=[],file="try.c",
31414 fullname="/home/foo/bar/try.c",line="10",arch="i386:x86_64"@}
31416 -exec-step-instruction
31420 *stopped,reason="end-stepping-range",
31421 frame=@{addr="0x000100f4",func="foo",args=[],file="try.c",
31422 fullname="/home/foo/bar/try.c",line="10",arch="i386:x86_64"@}
31427 @subheading The @code{-exec-until} Command
31428 @findex -exec-until
31430 @subsubheading Synopsis
31433 -exec-until [ @var{location} ]
31436 Executes the inferior until the @var{location} specified in the
31437 argument is reached. If there is no argument, the inferior executes
31438 until a source line greater than the current one is reached. The
31439 reason for stopping in this case will be @samp{location-reached}.
31441 @subsubheading @value{GDBN} Command
31443 The corresponding @value{GDBN} command is @samp{until}.
31445 @subsubheading Example
31449 -exec-until recursive2.c:6
31453 *stopped,reason="location-reached",frame=@{func="main",args=[],
31454 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="6",
31455 arch="i386:x86_64"@}
31460 @subheading -file-clear
31461 Is this going away????
31464 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31465 @node GDB/MI Stack Manipulation
31466 @section @sc{gdb/mi} Stack Manipulation Commands
31468 @subheading The @code{-enable-frame-filters} Command
31469 @findex -enable-frame-filters
31472 -enable-frame-filters
31475 @value{GDBN} allows Python-based frame filters to affect the output of
31476 the MI commands relating to stack traces. As there is no way to
31477 implement this in a fully backward-compatible way, a front end must
31478 request that this functionality be enabled.
31480 Once enabled, this feature cannot be disabled.
31482 Note that if Python support has not been compiled into @value{GDBN},
31483 this command will still succeed (and do nothing).
31485 @subheading The @code{-stack-info-frame} Command
31486 @findex -stack-info-frame
31488 @subsubheading Synopsis
31494 Get info on the selected frame.
31496 @subsubheading @value{GDBN} Command
31498 The corresponding @value{GDBN} command is @samp{info frame} or @samp{frame}
31499 (without arguments).
31501 @subsubheading Example
31506 ^done,frame=@{level="1",addr="0x0001076c",func="callee3",
31507 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31508 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17",
31509 arch="i386:x86_64"@}
31513 @subheading The @code{-stack-info-depth} Command
31514 @findex -stack-info-depth
31516 @subsubheading Synopsis
31519 -stack-info-depth [ @var{max-depth} ]
31522 Return the depth of the stack. If the integer argument @var{max-depth}
31523 is specified, do not count beyond @var{max-depth} frames.
31525 @subsubheading @value{GDBN} Command
31527 There's no equivalent @value{GDBN} command.
31529 @subsubheading Example
31531 For a stack with frame levels 0 through 11:
31538 -stack-info-depth 4
31541 -stack-info-depth 12
31544 -stack-info-depth 11
31547 -stack-info-depth 13
31552 @anchor{-stack-list-arguments}
31553 @subheading The @code{-stack-list-arguments} Command
31554 @findex -stack-list-arguments
31556 @subsubheading Synopsis
31559 -stack-list-arguments [ --no-frame-filters ] [ --skip-unavailable ] @var{print-values}
31560 [ @var{low-frame} @var{high-frame} ]
31563 Display a list of the arguments for the frames between @var{low-frame}
31564 and @var{high-frame} (inclusive). If @var{low-frame} and
31565 @var{high-frame} are not provided, list the arguments for the whole
31566 call stack. If the two arguments are equal, show the single frame
31567 at the corresponding level. It is an error if @var{low-frame} is
31568 larger than the actual number of frames. On the other hand,
31569 @var{high-frame} may be larger than the actual number of frames, in
31570 which case only existing frames will be returned.
31572 If @var{print-values} is 0 or @code{--no-values}, print only the names of
31573 the variables; if it is 1 or @code{--all-values}, print also their
31574 values; and if it is 2 or @code{--simple-values}, print the name,
31575 type and value for simple data types, and the name and type for arrays,
31576 structures and unions. If the option @code{--no-frame-filters} is
31577 supplied, then Python frame filters will not be executed.
31579 If the @code{--skip-unavailable} option is specified, arguments that
31580 are not available are not listed. Partially available arguments
31581 are still displayed, however.
31583 Use of this command to obtain arguments in a single frame is
31584 deprecated in favor of the @samp{-stack-list-variables} command.
31586 @subsubheading @value{GDBN} Command
31588 @value{GDBN} does not have an equivalent command. @code{gdbtk} has a
31589 @samp{gdb_get_args} command which partially overlaps with the
31590 functionality of @samp{-stack-list-arguments}.
31592 @subsubheading Example
31599 frame=@{level="0",addr="0x00010734",func="callee4",
31600 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31601 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8",
31602 arch="i386:x86_64"@},
31603 frame=@{level="1",addr="0x0001076c",func="callee3",
31604 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31605 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17",
31606 arch="i386:x86_64"@},
31607 frame=@{level="2",addr="0x0001078c",func="callee2",
31608 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31609 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="22",
31610 arch="i386:x86_64"@},
31611 frame=@{level="3",addr="0x000107b4",func="callee1",
31612 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31613 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="27",
31614 arch="i386:x86_64"@},
31615 frame=@{level="4",addr="0x000107e0",func="main",
31616 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31617 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="32",
31618 arch="i386:x86_64"@}]
31620 -stack-list-arguments 0
31623 frame=@{level="0",args=[]@},
31624 frame=@{level="1",args=[name="strarg"]@},
31625 frame=@{level="2",args=[name="intarg",name="strarg"]@},
31626 frame=@{level="3",args=[name="intarg",name="strarg",name="fltarg"]@},
31627 frame=@{level="4",args=[]@}]
31629 -stack-list-arguments 1
31632 frame=@{level="0",args=[]@},
31634 args=[@{name="strarg",value="0x11940 \"A string argument.\""@}]@},
31635 frame=@{level="2",args=[
31636 @{name="intarg",value="2"@},
31637 @{name="strarg",value="0x11940 \"A string argument.\""@}]@},
31638 @{frame=@{level="3",args=[
31639 @{name="intarg",value="2"@},
31640 @{name="strarg",value="0x11940 \"A string argument.\""@},
31641 @{name="fltarg",value="3.5"@}]@},
31642 frame=@{level="4",args=[]@}]
31644 -stack-list-arguments 0 2 2
31645 ^done,stack-args=[frame=@{level="2",args=[name="intarg",name="strarg"]@}]
31647 -stack-list-arguments 1 2 2
31648 ^done,stack-args=[frame=@{level="2",
31649 args=[@{name="intarg",value="2"@},
31650 @{name="strarg",value="0x11940 \"A string argument.\""@}]@}]
31654 @c @subheading -stack-list-exception-handlers
31657 @anchor{-stack-list-frames}
31658 @subheading The @code{-stack-list-frames} Command
31659 @findex -stack-list-frames
31661 @subsubheading Synopsis
31664 -stack-list-frames [ --no-frame-filters @var{low-frame} @var{high-frame} ]
31667 List the frames currently on the stack. For each frame it displays the
31672 The frame number, 0 being the topmost frame, i.e., the innermost function.
31674 The @code{$pc} value for that frame.
31678 File name of the source file where the function lives.
31679 @item @var{fullname}
31680 The full file name of the source file where the function lives.
31682 Line number corresponding to the @code{$pc}.
31684 The shared library where this function is defined. This is only given
31685 if the frame's function is not known.
31687 Frame's architecture.
31690 If invoked without arguments, this command prints a backtrace for the
31691 whole stack. If given two integer arguments, it shows the frames whose
31692 levels are between the two arguments (inclusive). If the two arguments
31693 are equal, it shows the single frame at the corresponding level. It is
31694 an error if @var{low-frame} is larger than the actual number of
31695 frames. On the other hand, @var{high-frame} may be larger than the
31696 actual number of frames, in which case only existing frames will be
31697 returned. If the option @code{--no-frame-filters} is supplied, then
31698 Python frame filters will not be executed.
31700 @subsubheading @value{GDBN} Command
31702 The corresponding @value{GDBN} commands are @samp{backtrace} and @samp{where}.
31704 @subsubheading Example
31706 Full stack backtrace:
31712 [frame=@{level="0",addr="0x0001076c",func="foo",
31713 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="11",
31714 arch="i386:x86_64"@},
31715 frame=@{level="1",addr="0x000107a4",func="foo",
31716 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31717 arch="i386:x86_64"@},
31718 frame=@{level="2",addr="0x000107a4",func="foo",
31719 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31720 arch="i386:x86_64"@},
31721 frame=@{level="3",addr="0x000107a4",func="foo",
31722 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31723 arch="i386:x86_64"@},
31724 frame=@{level="4",addr="0x000107a4",func="foo",
31725 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31726 arch="i386:x86_64"@},
31727 frame=@{level="5",addr="0x000107a4",func="foo",
31728 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31729 arch="i386:x86_64"@},
31730 frame=@{level="6",addr="0x000107a4",func="foo",
31731 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31732 arch="i386:x86_64"@},
31733 frame=@{level="7",addr="0x000107a4",func="foo",
31734 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31735 arch="i386:x86_64"@},
31736 frame=@{level="8",addr="0x000107a4",func="foo",
31737 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31738 arch="i386:x86_64"@},
31739 frame=@{level="9",addr="0x000107a4",func="foo",
31740 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31741 arch="i386:x86_64"@},
31742 frame=@{level="10",addr="0x000107a4",func="foo",
31743 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31744 arch="i386:x86_64"@},
31745 frame=@{level="11",addr="0x00010738",func="main",
31746 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="4",
31747 arch="i386:x86_64"@}]
31751 Show frames between @var{low_frame} and @var{high_frame}:
31755 -stack-list-frames 3 5
31757 [frame=@{level="3",addr="0x000107a4",func="foo",
31758 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31759 arch="i386:x86_64"@},
31760 frame=@{level="4",addr="0x000107a4",func="foo",
31761 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31762 arch="i386:x86_64"@},
31763 frame=@{level="5",addr="0x000107a4",func="foo",
31764 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31765 arch="i386:x86_64"@}]
31769 Show a single frame:
31773 -stack-list-frames 3 3
31775 [frame=@{level="3",addr="0x000107a4",func="foo",
31776 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31777 arch="i386:x86_64"@}]
31782 @subheading The @code{-stack-list-locals} Command
31783 @findex -stack-list-locals
31784 @anchor{-stack-list-locals}
31786 @subsubheading Synopsis
31789 -stack-list-locals [ --no-frame-filters ] [ --skip-unavailable ] @var{print-values}
31792 Display the local variable names for the selected frame. If
31793 @var{print-values} is 0 or @code{--no-values}, print only the names of
31794 the variables; if it is 1 or @code{--all-values}, print also their
31795 values; and if it is 2 or @code{--simple-values}, print the name,
31796 type and value for simple data types, and the name and type for arrays,
31797 structures and unions. In this last case, a frontend can immediately
31798 display the value of simple data types and create variable objects for
31799 other data types when the user wishes to explore their values in
31800 more detail. If the option @code{--no-frame-filters} is supplied, then
31801 Python frame filters will not be executed.
31803 If the @code{--skip-unavailable} option is specified, local variables
31804 that are not available are not listed. Partially available local
31805 variables are still displayed, however.
31807 This command is deprecated in favor of the
31808 @samp{-stack-list-variables} command.
31810 @subsubheading @value{GDBN} Command
31812 @samp{info locals} in @value{GDBN}, @samp{gdb_get_locals} in @code{gdbtk}.
31814 @subsubheading Example
31818 -stack-list-locals 0
31819 ^done,locals=[name="A",name="B",name="C"]
31821 -stack-list-locals --all-values
31822 ^done,locals=[@{name="A",value="1"@},@{name="B",value="2"@},
31823 @{name="C",value="@{1, 2, 3@}"@}]
31824 -stack-list-locals --simple-values
31825 ^done,locals=[@{name="A",type="int",value="1"@},
31826 @{name="B",type="int",value="2"@},@{name="C",type="int [3]"@}]
31830 @anchor{-stack-list-variables}
31831 @subheading The @code{-stack-list-variables} Command
31832 @findex -stack-list-variables
31834 @subsubheading Synopsis
31837 -stack-list-variables [ --no-frame-filters ] [ --skip-unavailable ] @var{print-values}
31840 Display the names of local variables and function arguments for the selected frame. If
31841 @var{print-values} is 0 or @code{--no-values}, print only the names of
31842 the variables; if it is 1 or @code{--all-values}, print also their
31843 values; and if it is 2 or @code{--simple-values}, print the name,
31844 type and value for simple data types, and the name and type for arrays,
31845 structures and unions. If the option @code{--no-frame-filters} is
31846 supplied, then Python frame filters will not be executed.
31848 If the @code{--skip-unavailable} option is specified, local variables
31849 and arguments that are not available are not listed. Partially
31850 available arguments and local variables are still displayed, however.
31852 @subsubheading Example
31856 -stack-list-variables --thread 1 --frame 0 --all-values
31857 ^done,variables=[@{name="x",value="11"@},@{name="s",value="@{a = 1, b = 2@}"@}]
31862 @subheading The @code{-stack-select-frame} Command
31863 @findex -stack-select-frame
31865 @subsubheading Synopsis
31868 -stack-select-frame @var{framenum}
31871 Change the selected frame. Select a different frame @var{framenum} on
31874 This command in deprecated in favor of passing the @samp{--frame}
31875 option to every command.
31877 @subsubheading @value{GDBN} Command
31879 The corresponding @value{GDBN} commands are @samp{frame}, @samp{up},
31880 @samp{down}, @samp{select-frame}, @samp{up-silent}, and @samp{down-silent}.
31882 @subsubheading Example
31886 -stack-select-frame 2
31891 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31892 @node GDB/MI Variable Objects
31893 @section @sc{gdb/mi} Variable Objects
31897 @subheading Motivation for Variable Objects in @sc{gdb/mi}
31899 For the implementation of a variable debugger window (locals, watched
31900 expressions, etc.), we are proposing the adaptation of the existing code
31901 used by @code{Insight}.
31903 The two main reasons for that are:
31907 It has been proven in practice (it is already on its second generation).
31910 It will shorten development time (needless to say how important it is
31914 The original interface was designed to be used by Tcl code, so it was
31915 slightly changed so it could be used through @sc{gdb/mi}. This section
31916 describes the @sc{gdb/mi} operations that will be available and gives some
31917 hints about their use.
31919 @emph{Note}: In addition to the set of operations described here, we
31920 expect the @sc{gui} implementation of a variable window to require, at
31921 least, the following operations:
31924 @item @code{-gdb-show} @code{output-radix}
31925 @item @code{-stack-list-arguments}
31926 @item @code{-stack-list-locals}
31927 @item @code{-stack-select-frame}
31932 @subheading Introduction to Variable Objects
31934 @cindex variable objects in @sc{gdb/mi}
31936 Variable objects are "object-oriented" MI interface for examining and
31937 changing values of expressions. Unlike some other MI interfaces that
31938 work with expressions, variable objects are specifically designed for
31939 simple and efficient presentation in the frontend. A variable object
31940 is identified by string name. When a variable object is created, the
31941 frontend specifies the expression for that variable object. The
31942 expression can be a simple variable, or it can be an arbitrary complex
31943 expression, and can even involve CPU registers. After creating a
31944 variable object, the frontend can invoke other variable object
31945 operations---for example to obtain or change the value of a variable
31946 object, or to change display format.
31948 Variable objects have hierarchical tree structure. Any variable object
31949 that corresponds to a composite type, such as structure in C, has
31950 a number of child variable objects, for example corresponding to each
31951 element of a structure. A child variable object can itself have
31952 children, recursively. Recursion ends when we reach
31953 leaf variable objects, which always have built-in types. Child variable
31954 objects are created only by explicit request, so if a frontend
31955 is not interested in the children of a particular variable object, no
31956 child will be created.
31958 For a leaf variable object it is possible to obtain its value as a
31959 string, or set the value from a string. String value can be also
31960 obtained for a non-leaf variable object, but it's generally a string
31961 that only indicates the type of the object, and does not list its
31962 contents. Assignment to a non-leaf variable object is not allowed.
31964 A frontend does not need to read the values of all variable objects each time
31965 the program stops. Instead, MI provides an update command that lists all
31966 variable objects whose values has changed since the last update
31967 operation. This considerably reduces the amount of data that must
31968 be transferred to the frontend. As noted above, children variable
31969 objects are created on demand, and only leaf variable objects have a
31970 real value. As result, gdb will read target memory only for leaf
31971 variables that frontend has created.
31973 The automatic update is not always desirable. For example, a frontend
31974 might want to keep a value of some expression for future reference,
31975 and never update it. For another example, fetching memory is
31976 relatively slow for embedded targets, so a frontend might want
31977 to disable automatic update for the variables that are either not
31978 visible on the screen, or ``closed''. This is possible using so
31979 called ``frozen variable objects''. Such variable objects are never
31980 implicitly updated.
31982 Variable objects can be either @dfn{fixed} or @dfn{floating}. For the
31983 fixed variable object, the expression is parsed when the variable
31984 object is created, including associating identifiers to specific
31985 variables. The meaning of expression never changes. For a floating
31986 variable object the values of variables whose names appear in the
31987 expressions are re-evaluated every time in the context of the current
31988 frame. Consider this example:
31993 struct work_state state;
32000 If a fixed variable object for the @code{state} variable is created in
32001 this function, and we enter the recursive call, the variable
32002 object will report the value of @code{state} in the top-level
32003 @code{do_work} invocation. On the other hand, a floating variable
32004 object will report the value of @code{state} in the current frame.
32006 If an expression specified when creating a fixed variable object
32007 refers to a local variable, the variable object becomes bound to the
32008 thread and frame in which the variable object is created. When such
32009 variable object is updated, @value{GDBN} makes sure that the
32010 thread/frame combination the variable object is bound to still exists,
32011 and re-evaluates the variable object in context of that thread/frame.
32013 The following is the complete set of @sc{gdb/mi} operations defined to
32014 access this functionality:
32016 @multitable @columnfractions .4 .6
32017 @item @strong{Operation}
32018 @tab @strong{Description}
32020 @item @code{-enable-pretty-printing}
32021 @tab enable Python-based pretty-printing
32022 @item @code{-var-create}
32023 @tab create a variable object
32024 @item @code{-var-delete}
32025 @tab delete the variable object and/or its children
32026 @item @code{-var-set-format}
32027 @tab set the display format of this variable
32028 @item @code{-var-show-format}
32029 @tab show the display format of this variable
32030 @item @code{-var-info-num-children}
32031 @tab tells how many children this object has
32032 @item @code{-var-list-children}
32033 @tab return a list of the object's children
32034 @item @code{-var-info-type}
32035 @tab show the type of this variable object
32036 @item @code{-var-info-expression}
32037 @tab print parent-relative expression that this variable object represents
32038 @item @code{-var-info-path-expression}
32039 @tab print full expression that this variable object represents
32040 @item @code{-var-show-attributes}
32041 @tab is this variable editable? does it exist here?
32042 @item @code{-var-evaluate-expression}
32043 @tab get the value of this variable
32044 @item @code{-var-assign}
32045 @tab set the value of this variable
32046 @item @code{-var-update}
32047 @tab update the variable and its children
32048 @item @code{-var-set-frozen}
32049 @tab set frozenness attribute
32050 @item @code{-var-set-update-range}
32051 @tab set range of children to display on update
32054 In the next subsection we describe each operation in detail and suggest
32055 how it can be used.
32057 @subheading Description And Use of Operations on Variable Objects
32059 @subheading The @code{-enable-pretty-printing} Command
32060 @findex -enable-pretty-printing
32063 -enable-pretty-printing
32066 @value{GDBN} allows Python-based visualizers to affect the output of the
32067 MI variable object commands. However, because there was no way to
32068 implement this in a fully backward-compatible way, a front end must
32069 request that this functionality be enabled.
32071 Once enabled, this feature cannot be disabled.
32073 Note that if Python support has not been compiled into @value{GDBN},
32074 this command will still succeed (and do nothing).
32076 This feature is currently (as of @value{GDBN} 7.0) experimental, and
32077 may work differently in future versions of @value{GDBN}.
32079 @subheading The @code{-var-create} Command
32080 @findex -var-create
32082 @subsubheading Synopsis
32085 -var-create @{@var{name} | "-"@}
32086 @{@var{frame-addr} | "*" | "@@"@} @var{expression}
32089 This operation creates a variable object, which allows the monitoring of
32090 a variable, the result of an expression, a memory cell or a CPU
32093 The @var{name} parameter is the string by which the object can be
32094 referenced. It must be unique. If @samp{-} is specified, the varobj
32095 system will generate a string ``varNNNNNN'' automatically. It will be
32096 unique provided that one does not specify @var{name} of that format.
32097 The command fails if a duplicate name is found.
32099 The frame under which the expression should be evaluated can be
32100 specified by @var{frame-addr}. A @samp{*} indicates that the current
32101 frame should be used. A @samp{@@} indicates that a floating variable
32102 object must be created.
32104 @var{expression} is any expression valid on the current language set (must not
32105 begin with a @samp{*}), or one of the following:
32109 @samp{*@var{addr}}, where @var{addr} is the address of a memory cell
32112 @samp{*@var{addr}-@var{addr}} --- a memory address range (TBD)
32115 @samp{$@var{regname}} --- a CPU register name
32118 @cindex dynamic varobj
32119 A varobj's contents may be provided by a Python-based pretty-printer. In this
32120 case the varobj is known as a @dfn{dynamic varobj}. Dynamic varobjs
32121 have slightly different semantics in some cases. If the
32122 @code{-enable-pretty-printing} command is not sent, then @value{GDBN}
32123 will never create a dynamic varobj. This ensures backward
32124 compatibility for existing clients.
32126 @subsubheading Result
32128 This operation returns attributes of the newly-created varobj. These
32133 The name of the varobj.
32136 The number of children of the varobj. This number is not necessarily
32137 reliable for a dynamic varobj. Instead, you must examine the
32138 @samp{has_more} attribute.
32141 The varobj's scalar value. For a varobj whose type is some sort of
32142 aggregate (e.g., a @code{struct}), or for a dynamic varobj, this value
32143 will not be interesting.
32146 The varobj's type. This is a string representation of the type, as
32147 would be printed by the @value{GDBN} CLI. If @samp{print object}
32148 (@pxref{Print Settings, set print object}) is set to @code{on}, the
32149 @emph{actual} (derived) type of the object is shown rather than the
32150 @emph{declared} one.
32153 If a variable object is bound to a specific thread, then this is the
32154 thread's global identifier.
32157 For a dynamic varobj, this indicates whether there appear to be any
32158 children available. For a non-dynamic varobj, this will be 0.
32161 This attribute will be present and have the value @samp{1} if the
32162 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
32163 then this attribute will not be present.
32166 A dynamic varobj can supply a display hint to the front end. The
32167 value comes directly from the Python pretty-printer object's
32168 @code{display_hint} method. @xref{Pretty Printing API}.
32171 Typical output will look like this:
32174 name="@var{name}",numchild="@var{N}",type="@var{type}",thread-id="@var{M}",
32175 has_more="@var{has_more}"
32179 @subheading The @code{-var-delete} Command
32180 @findex -var-delete
32182 @subsubheading Synopsis
32185 -var-delete [ -c ] @var{name}
32188 Deletes a previously created variable object and all of its children.
32189 With the @samp{-c} option, just deletes the children.
32191 Returns an error if the object @var{name} is not found.
32194 @subheading The @code{-var-set-format} Command
32195 @findex -var-set-format
32197 @subsubheading Synopsis
32200 -var-set-format @var{name} @var{format-spec}
32203 Sets the output format for the value of the object @var{name} to be
32206 @anchor{-var-set-format}
32207 The syntax for the @var{format-spec} is as follows:
32210 @var{format-spec} @expansion{}
32211 @{binary | decimal | hexadecimal | octal | natural | zero-hexadecimal@}
32214 The natural format is the default format choosen automatically
32215 based on the variable type (like decimal for an @code{int}, hex
32216 for pointers, etc.).
32218 The zero-hexadecimal format has a representation similar to hexadecimal
32219 but with padding zeroes to the left of the value. For example, a 32-bit
32220 hexadecimal value of 0x1234 would be represented as 0x00001234 in the
32221 zero-hexadecimal format.
32223 For a variable with children, the format is set only on the
32224 variable itself, and the children are not affected.
32226 @subheading The @code{-var-show-format} Command
32227 @findex -var-show-format
32229 @subsubheading Synopsis
32232 -var-show-format @var{name}
32235 Returns the format used to display the value of the object @var{name}.
32238 @var{format} @expansion{}
32243 @subheading The @code{-var-info-num-children} Command
32244 @findex -var-info-num-children
32246 @subsubheading Synopsis
32249 -var-info-num-children @var{name}
32252 Returns the number of children of a variable object @var{name}:
32258 Note that this number is not completely reliable for a dynamic varobj.
32259 It will return the current number of children, but more children may
32263 @subheading The @code{-var-list-children} Command
32264 @findex -var-list-children
32266 @subsubheading Synopsis
32269 -var-list-children [@var{print-values}] @var{name} [@var{from} @var{to}]
32271 @anchor{-var-list-children}
32273 Return a list of the children of the specified variable object and
32274 create variable objects for them, if they do not already exist. With
32275 a single argument or if @var{print-values} has a value of 0 or
32276 @code{--no-values}, print only the names of the variables; if
32277 @var{print-values} is 1 or @code{--all-values}, also print their
32278 values; and if it is 2 or @code{--simple-values} print the name and
32279 value for simple data types and just the name for arrays, structures
32282 @var{from} and @var{to}, if specified, indicate the range of children
32283 to report. If @var{from} or @var{to} is less than zero, the range is
32284 reset and all children will be reported. Otherwise, children starting
32285 at @var{from} (zero-based) and up to and excluding @var{to} will be
32288 If a child range is requested, it will only affect the current call to
32289 @code{-var-list-children}, but not future calls to @code{-var-update}.
32290 For this, you must instead use @code{-var-set-update-range}. The
32291 intent of this approach is to enable a front end to implement any
32292 update approach it likes; for example, scrolling a view may cause the
32293 front end to request more children with @code{-var-list-children}, and
32294 then the front end could call @code{-var-set-update-range} with a
32295 different range to ensure that future updates are restricted to just
32298 For each child the following results are returned:
32303 Name of the variable object created for this child.
32306 The expression to be shown to the user by the front end to designate this child.
32307 For example this may be the name of a structure member.
32309 For a dynamic varobj, this value cannot be used to form an
32310 expression. There is no way to do this at all with a dynamic varobj.
32312 For C/C@t{++} structures there are several pseudo children returned to
32313 designate access qualifiers. For these pseudo children @var{exp} is
32314 @samp{public}, @samp{private}, or @samp{protected}. In this case the
32315 type and value are not present.
32317 A dynamic varobj will not report the access qualifying
32318 pseudo-children, regardless of the language. This information is not
32319 available at all with a dynamic varobj.
32322 Number of children this child has. For a dynamic varobj, this will be
32326 The type of the child. If @samp{print object}
32327 (@pxref{Print Settings, set print object}) is set to @code{on}, the
32328 @emph{actual} (derived) type of the object is shown rather than the
32329 @emph{declared} one.
32332 If values were requested, this is the value.
32335 If this variable object is associated with a thread, this is the
32336 thread's global thread id. Otherwise this result is not present.
32339 If the variable object is frozen, this variable will be present with a value of 1.
32342 A dynamic varobj can supply a display hint to the front end. The
32343 value comes directly from the Python pretty-printer object's
32344 @code{display_hint} method. @xref{Pretty Printing API}.
32347 This attribute will be present and have the value @samp{1} if the
32348 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
32349 then this attribute will not be present.
32353 The result may have its own attributes:
32357 A dynamic varobj can supply a display hint to the front end. The
32358 value comes directly from the Python pretty-printer object's
32359 @code{display_hint} method. @xref{Pretty Printing API}.
32362 This is an integer attribute which is nonzero if there are children
32363 remaining after the end of the selected range.
32366 @subsubheading Example
32370 -var-list-children n
32371 ^done,numchild=@var{n},children=[child=@{name=@var{name},exp=@var{exp},
32372 numchild=@var{n},type=@var{type}@},@r{(repeats N times)}]
32374 -var-list-children --all-values n
32375 ^done,numchild=@var{n},children=[child=@{name=@var{name},exp=@var{exp},
32376 numchild=@var{n},value=@var{value},type=@var{type}@},@r{(repeats N times)}]
32380 @subheading The @code{-var-info-type} Command
32381 @findex -var-info-type
32383 @subsubheading Synopsis
32386 -var-info-type @var{name}
32389 Returns the type of the specified variable @var{name}. The type is
32390 returned as a string in the same format as it is output by the
32394 type=@var{typename}
32398 @subheading The @code{-var-info-expression} Command
32399 @findex -var-info-expression
32401 @subsubheading Synopsis
32404 -var-info-expression @var{name}
32407 Returns a string that is suitable for presenting this
32408 variable object in user interface. The string is generally
32409 not valid expression in the current language, and cannot be evaluated.
32411 For example, if @code{a} is an array, and variable object
32412 @code{A} was created for @code{a}, then we'll get this output:
32415 (gdb) -var-info-expression A.1
32416 ^done,lang="C",exp="1"
32420 Here, the value of @code{lang} is the language name, which can be
32421 found in @ref{Supported Languages}.
32423 Note that the output of the @code{-var-list-children} command also
32424 includes those expressions, so the @code{-var-info-expression} command
32427 @subheading The @code{-var-info-path-expression} Command
32428 @findex -var-info-path-expression
32430 @subsubheading Synopsis
32433 -var-info-path-expression @var{name}
32436 Returns an expression that can be evaluated in the current
32437 context and will yield the same value that a variable object has.
32438 Compare this with the @code{-var-info-expression} command, which
32439 result can be used only for UI presentation. Typical use of
32440 the @code{-var-info-path-expression} command is creating a
32441 watchpoint from a variable object.
32443 This command is currently not valid for children of a dynamic varobj,
32444 and will give an error when invoked on one.
32446 For example, suppose @code{C} is a C@t{++} class, derived from class
32447 @code{Base}, and that the @code{Base} class has a member called
32448 @code{m_size}. Assume a variable @code{c} is has the type of
32449 @code{C} and a variable object @code{C} was created for variable
32450 @code{c}. Then, we'll get this output:
32452 (gdb) -var-info-path-expression C.Base.public.m_size
32453 ^done,path_expr=((Base)c).m_size)
32456 @subheading The @code{-var-show-attributes} Command
32457 @findex -var-show-attributes
32459 @subsubheading Synopsis
32462 -var-show-attributes @var{name}
32465 List attributes of the specified variable object @var{name}:
32468 status=@var{attr} [ ( ,@var{attr} )* ]
32472 where @var{attr} is @code{@{ @{ editable | noneditable @} | TBD @}}.
32474 @subheading The @code{-var-evaluate-expression} Command
32475 @findex -var-evaluate-expression
32477 @subsubheading Synopsis
32480 -var-evaluate-expression [-f @var{format-spec}] @var{name}
32483 Evaluates the expression that is represented by the specified variable
32484 object and returns its value as a string. The format of the string
32485 can be specified with the @samp{-f} option. The possible values of
32486 this option are the same as for @code{-var-set-format}
32487 (@pxref{-var-set-format}). If the @samp{-f} option is not specified,
32488 the current display format will be used. The current display format
32489 can be changed using the @code{-var-set-format} command.
32495 Note that one must invoke @code{-var-list-children} for a variable
32496 before the value of a child variable can be evaluated.
32498 @subheading The @code{-var-assign} Command
32499 @findex -var-assign
32501 @subsubheading Synopsis
32504 -var-assign @var{name} @var{expression}
32507 Assigns the value of @var{expression} to the variable object specified
32508 by @var{name}. The object must be @samp{editable}. If the variable's
32509 value is altered by the assign, the variable will show up in any
32510 subsequent @code{-var-update} list.
32512 @subsubheading Example
32520 ^done,changelist=[@{name="var1",in_scope="true",type_changed="false"@}]
32524 @subheading The @code{-var-update} Command
32525 @findex -var-update
32527 @subsubheading Synopsis
32530 -var-update [@var{print-values}] @{@var{name} | "*"@}
32533 Reevaluate the expressions corresponding to the variable object
32534 @var{name} and all its direct and indirect children, and return the
32535 list of variable objects whose values have changed; @var{name} must
32536 be a root variable object. Here, ``changed'' means that the result of
32537 @code{-var-evaluate-expression} before and after the
32538 @code{-var-update} is different. If @samp{*} is used as the variable
32539 object names, all existing variable objects are updated, except
32540 for frozen ones (@pxref{-var-set-frozen}). The option
32541 @var{print-values} determines whether both names and values, or just
32542 names are printed. The possible values of this option are the same
32543 as for @code{-var-list-children} (@pxref{-var-list-children}). It is
32544 recommended to use the @samp{--all-values} option, to reduce the
32545 number of MI commands needed on each program stop.
32547 With the @samp{*} parameter, if a variable object is bound to a
32548 currently running thread, it will not be updated, without any
32551 If @code{-var-set-update-range} was previously used on a varobj, then
32552 only the selected range of children will be reported.
32554 @code{-var-update} reports all the changed varobjs in a tuple named
32557 Each item in the change list is itself a tuple holding:
32561 The name of the varobj.
32564 If values were requested for this update, then this field will be
32565 present and will hold the value of the varobj.
32568 @anchor{-var-update}
32569 This field is a string which may take one of three values:
32573 The variable object's current value is valid.
32576 The variable object does not currently hold a valid value but it may
32577 hold one in the future if its associated expression comes back into
32581 The variable object no longer holds a valid value.
32582 This can occur when the executable file being debugged has changed,
32583 either through recompilation or by using the @value{GDBN} @code{file}
32584 command. The front end should normally choose to delete these variable
32588 In the future new values may be added to this list so the front should
32589 be prepared for this possibility. @xref{GDB/MI Development and Front Ends, ,@sc{GDB/MI} Development and Front Ends}.
32592 This is only present if the varobj is still valid. If the type
32593 changed, then this will be the string @samp{true}; otherwise it will
32596 When a varobj's type changes, its children are also likely to have
32597 become incorrect. Therefore, the varobj's children are automatically
32598 deleted when this attribute is @samp{true}. Also, the varobj's update
32599 range, when set using the @code{-var-set-update-range} command, is
32603 If the varobj's type changed, then this field will be present and will
32606 @item new_num_children
32607 For a dynamic varobj, if the number of children changed, or if the
32608 type changed, this will be the new number of children.
32610 The @samp{numchild} field in other varobj responses is generally not
32611 valid for a dynamic varobj -- it will show the number of children that
32612 @value{GDBN} knows about, but because dynamic varobjs lazily
32613 instantiate their children, this will not reflect the number of
32614 children which may be available.
32616 The @samp{new_num_children} attribute only reports changes to the
32617 number of children known by @value{GDBN}. This is the only way to
32618 detect whether an update has removed children (which necessarily can
32619 only happen at the end of the update range).
32622 The display hint, if any.
32625 This is an integer value, which will be 1 if there are more children
32626 available outside the varobj's update range.
32629 This attribute will be present and have the value @samp{1} if the
32630 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
32631 then this attribute will not be present.
32634 If new children were added to a dynamic varobj within the selected
32635 update range (as set by @code{-var-set-update-range}), then they will
32636 be listed in this attribute.
32639 @subsubheading Example
32646 -var-update --all-values var1
32647 ^done,changelist=[@{name="var1",value="3",in_scope="true",
32648 type_changed="false"@}]
32652 @subheading The @code{-var-set-frozen} Command
32653 @findex -var-set-frozen
32654 @anchor{-var-set-frozen}
32656 @subsubheading Synopsis
32659 -var-set-frozen @var{name} @var{flag}
32662 Set the frozenness flag on the variable object @var{name}. The
32663 @var{flag} parameter should be either @samp{1} to make the variable
32664 frozen or @samp{0} to make it unfrozen. If a variable object is
32665 frozen, then neither itself, nor any of its children, are
32666 implicitly updated by @code{-var-update} of
32667 a parent variable or by @code{-var-update *}. Only
32668 @code{-var-update} of the variable itself will update its value and
32669 values of its children. After a variable object is unfrozen, it is
32670 implicitly updated by all subsequent @code{-var-update} operations.
32671 Unfreezing a variable does not update it, only subsequent
32672 @code{-var-update} does.
32674 @subsubheading Example
32678 -var-set-frozen V 1
32683 @subheading The @code{-var-set-update-range} command
32684 @findex -var-set-update-range
32685 @anchor{-var-set-update-range}
32687 @subsubheading Synopsis
32690 -var-set-update-range @var{name} @var{from} @var{to}
32693 Set the range of children to be returned by future invocations of
32694 @code{-var-update}.
32696 @var{from} and @var{to} indicate the range of children to report. If
32697 @var{from} or @var{to} is less than zero, the range is reset and all
32698 children will be reported. Otherwise, children starting at @var{from}
32699 (zero-based) and up to and excluding @var{to} will be reported.
32701 @subsubheading Example
32705 -var-set-update-range V 1 2
32709 @subheading The @code{-var-set-visualizer} command
32710 @findex -var-set-visualizer
32711 @anchor{-var-set-visualizer}
32713 @subsubheading Synopsis
32716 -var-set-visualizer @var{name} @var{visualizer}
32719 Set a visualizer for the variable object @var{name}.
32721 @var{visualizer} is the visualizer to use. The special value
32722 @samp{None} means to disable any visualizer in use.
32724 If not @samp{None}, @var{visualizer} must be a Python expression.
32725 This expression must evaluate to a callable object which accepts a
32726 single argument. @value{GDBN} will call this object with the value of
32727 the varobj @var{name} as an argument (this is done so that the same
32728 Python pretty-printing code can be used for both the CLI and MI).
32729 When called, this object must return an object which conforms to the
32730 pretty-printing interface (@pxref{Pretty Printing API}).
32732 The pre-defined function @code{gdb.default_visualizer} may be used to
32733 select a visualizer by following the built-in process
32734 (@pxref{Selecting Pretty-Printers}). This is done automatically when
32735 a varobj is created, and so ordinarily is not needed.
32737 This feature is only available if Python support is enabled. The MI
32738 command @code{-list-features} (@pxref{GDB/MI Support Commands})
32739 can be used to check this.
32741 @subsubheading Example
32743 Resetting the visualizer:
32747 -var-set-visualizer V None
32751 Reselecting the default (type-based) visualizer:
32755 -var-set-visualizer V gdb.default_visualizer
32759 Suppose @code{SomeClass} is a visualizer class. A lambda expression
32760 can be used to instantiate this class for a varobj:
32764 -var-set-visualizer V "lambda val: SomeClass()"
32768 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
32769 @node GDB/MI Data Manipulation
32770 @section @sc{gdb/mi} Data Manipulation
32772 @cindex data manipulation, in @sc{gdb/mi}
32773 @cindex @sc{gdb/mi}, data manipulation
32774 This section describes the @sc{gdb/mi} commands that manipulate data:
32775 examine memory and registers, evaluate expressions, etc.
32777 For details about what an addressable memory unit is,
32778 @pxref{addressable memory unit}.
32780 @c REMOVED FROM THE INTERFACE.
32781 @c @subheading -data-assign
32782 @c Change the value of a program variable. Plenty of side effects.
32783 @c @subsubheading GDB Command
32785 @c @subsubheading Example
32788 @subheading The @code{-data-disassemble} Command
32789 @findex -data-disassemble
32791 @subsubheading Synopsis
32795 [ -s @var{start-addr} -e @var{end-addr} ]
32796 | [ -a @var{addr} ]
32797 | [ -f @var{filename} -l @var{linenum} [ -n @var{lines} ] ]
32805 @item @var{start-addr}
32806 is the beginning address (or @code{$pc})
32807 @item @var{end-addr}
32810 is an address anywhere within (or the name of) the function to
32811 disassemble. If an address is specified, the whole function
32812 surrounding that address will be disassembled. If a name is
32813 specified, the whole function with that name will be disassembled.
32814 @item @var{filename}
32815 is the name of the file to disassemble
32816 @item @var{linenum}
32817 is the line number to disassemble around
32819 is the number of disassembly lines to be produced. If it is -1,
32820 the whole function will be disassembled, in case no @var{end-addr} is
32821 specified. If @var{end-addr} is specified as a non-zero value, and
32822 @var{lines} is lower than the number of disassembly lines between
32823 @var{start-addr} and @var{end-addr}, only @var{lines} lines are
32824 displayed; if @var{lines} is higher than the number of lines between
32825 @var{start-addr} and @var{end-addr}, only the lines up to @var{end-addr}
32830 @item 0 disassembly only
32831 @item 1 mixed source and disassembly (deprecated)
32832 @item 2 disassembly with raw opcodes
32833 @item 3 mixed source and disassembly with raw opcodes (deprecated)
32834 @item 4 mixed source and disassembly
32835 @item 5 mixed source and disassembly with raw opcodes
32838 Modes 1 and 3 are deprecated. The output is ``source centric''
32839 which hasn't proved useful in practice.
32840 @xref{Machine Code}, for a discussion of the difference between
32841 @code{/m} and @code{/s} output of the @code{disassemble} command.
32844 @subsubheading Result
32846 The result of the @code{-data-disassemble} command will be a list named
32847 @samp{asm_insns}, the contents of this list depend on the @var{mode}
32848 used with the @code{-data-disassemble} command.
32850 For modes 0 and 2 the @samp{asm_insns} list contains tuples with the
32855 The address at which this instruction was disassembled.
32858 The name of the function this instruction is within.
32861 The decimal offset in bytes from the start of @samp{func-name}.
32864 The text disassembly for this @samp{address}.
32867 This field is only present for modes 2, 3 and 5. This contains the raw opcode
32868 bytes for the @samp{inst} field.
32872 For modes 1, 3, 4 and 5 the @samp{asm_insns} list contains tuples named
32873 @samp{src_and_asm_line}, each of which has the following fields:
32877 The line number within @samp{file}.
32880 The file name from the compilation unit. This might be an absolute
32881 file name or a relative file name depending on the compile command
32885 Absolute file name of @samp{file}. It is converted to a canonical form
32886 using the source file search path
32887 (@pxref{Source Path, ,Specifying Source Directories})
32888 and after resolving all the symbolic links.
32890 If the source file is not found this field will contain the path as
32891 present in the debug information.
32893 @item line_asm_insn
32894 This is a list of tuples containing the disassembly for @samp{line} in
32895 @samp{file}. The fields of each tuple are the same as for
32896 @code{-data-disassemble} in @var{mode} 0 and 2, so @samp{address},
32897 @samp{func-name}, @samp{offset}, @samp{inst}, and optionally
32902 Note that whatever included in the @samp{inst} field, is not
32903 manipulated directly by @sc{gdb/mi}, i.e., it is not possible to
32906 @subsubheading @value{GDBN} Command
32908 The corresponding @value{GDBN} command is @samp{disassemble}.
32910 @subsubheading Example
32912 Disassemble from the current value of @code{$pc} to @code{$pc + 20}:
32916 -data-disassemble -s $pc -e "$pc + 20" -- 0
32919 @{address="0x000107c0",func-name="main",offset="4",
32920 inst="mov 2, %o0"@},
32921 @{address="0x000107c4",func-name="main",offset="8",
32922 inst="sethi %hi(0x11800), %o2"@},
32923 @{address="0x000107c8",func-name="main",offset="12",
32924 inst="or %o2, 0x140, %o1\t! 0x11940 <_lib_version+8>"@},
32925 @{address="0x000107cc",func-name="main",offset="16",
32926 inst="sethi %hi(0x11800), %o2"@},
32927 @{address="0x000107d0",func-name="main",offset="20",
32928 inst="or %o2, 0x168, %o4\t! 0x11968 <_lib_version+48>"@}]
32932 Disassemble the whole @code{main} function. Line 32 is part of
32936 -data-disassemble -f basics.c -l 32 -- 0
32938 @{address="0x000107bc",func-name="main",offset="0",
32939 inst="save %sp, -112, %sp"@},
32940 @{address="0x000107c0",func-name="main",offset="4",
32941 inst="mov 2, %o0"@},
32942 @{address="0x000107c4",func-name="main",offset="8",
32943 inst="sethi %hi(0x11800), %o2"@},
32945 @{address="0x0001081c",func-name="main",offset="96",inst="ret "@},
32946 @{address="0x00010820",func-name="main",offset="100",inst="restore "@}]
32950 Disassemble 3 instructions from the start of @code{main}:
32954 -data-disassemble -f basics.c -l 32 -n 3 -- 0
32956 @{address="0x000107bc",func-name="main",offset="0",
32957 inst="save %sp, -112, %sp"@},
32958 @{address="0x000107c0",func-name="main",offset="4",
32959 inst="mov 2, %o0"@},
32960 @{address="0x000107c4",func-name="main",offset="8",
32961 inst="sethi %hi(0x11800), %o2"@}]
32965 Disassemble 3 instructions from the start of @code{main} in mixed mode:
32969 -data-disassemble -f basics.c -l 32 -n 3 -- 1
32971 src_and_asm_line=@{line="31",
32972 file="../../../src/gdb/testsuite/gdb.mi/basics.c",
32973 fullname="/absolute/path/to/src/gdb/testsuite/gdb.mi/basics.c",
32974 line_asm_insn=[@{address="0x000107bc",
32975 func-name="main",offset="0",inst="save %sp, -112, %sp"@}]@},
32976 src_and_asm_line=@{line="32",
32977 file="../../../src/gdb/testsuite/gdb.mi/basics.c",
32978 fullname="/absolute/path/to/src/gdb/testsuite/gdb.mi/basics.c",
32979 line_asm_insn=[@{address="0x000107c0",
32980 func-name="main",offset="4",inst="mov 2, %o0"@},
32981 @{address="0x000107c4",func-name="main",offset="8",
32982 inst="sethi %hi(0x11800), %o2"@}]@}]
32987 @subheading The @code{-data-evaluate-expression} Command
32988 @findex -data-evaluate-expression
32990 @subsubheading Synopsis
32993 -data-evaluate-expression @var{expr}
32996 Evaluate @var{expr} as an expression. The expression could contain an
32997 inferior function call. The function call will execute synchronously.
32998 If the expression contains spaces, it must be enclosed in double quotes.
33000 @subsubheading @value{GDBN} Command
33002 The corresponding @value{GDBN} commands are @samp{print}, @samp{output}, and
33003 @samp{call}. In @code{gdbtk} only, there's a corresponding
33004 @samp{gdb_eval} command.
33006 @subsubheading Example
33008 In the following example, the numbers that precede the commands are the
33009 @dfn{tokens} described in @ref{GDB/MI Command Syntax, ,@sc{gdb/mi}
33010 Command Syntax}. Notice how @sc{gdb/mi} returns the same tokens in its
33014 211-data-evaluate-expression A
33017 311-data-evaluate-expression &A
33018 311^done,value="0xefffeb7c"
33020 411-data-evaluate-expression A+3
33023 511-data-evaluate-expression "A + 3"
33029 @subheading The @code{-data-list-changed-registers} Command
33030 @findex -data-list-changed-registers
33032 @subsubheading Synopsis
33035 -data-list-changed-registers
33038 Display a list of the registers that have changed.
33040 @subsubheading @value{GDBN} Command
33042 @value{GDBN} doesn't have a direct analog for this command; @code{gdbtk}
33043 has the corresponding command @samp{gdb_changed_register_list}.
33045 @subsubheading Example
33047 On a PPC MBX board:
33055 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",frame=@{
33056 func="main",args=[],file="try.c",fullname="/home/foo/bar/try.c",
33057 line="5",arch="powerpc"@}
33059 -data-list-changed-registers
33060 ^done,changed-registers=["0","1","2","4","5","6","7","8","9",
33061 "10","11","13","14","15","16","17","18","19","20","21","22","23",
33062 "24","25","26","27","28","30","31","64","65","66","67","69"]
33067 @subheading The @code{-data-list-register-names} Command
33068 @findex -data-list-register-names
33070 @subsubheading Synopsis
33073 -data-list-register-names [ ( @var{regno} )+ ]
33076 Show a list of register names for the current target. If no arguments
33077 are given, it shows a list of the names of all the registers. If
33078 integer numbers are given as arguments, it will print a list of the
33079 names of the registers corresponding to the arguments. To ensure
33080 consistency between a register name and its number, the output list may
33081 include empty register names.
33083 @subsubheading @value{GDBN} Command
33085 @value{GDBN} does not have a command which corresponds to
33086 @samp{-data-list-register-names}. In @code{gdbtk} there is a
33087 corresponding command @samp{gdb_regnames}.
33089 @subsubheading Example
33091 For the PPC MBX board:
33094 -data-list-register-names
33095 ^done,register-names=["r0","r1","r2","r3","r4","r5","r6","r7",
33096 "r8","r9","r10","r11","r12","r13","r14","r15","r16","r17","r18",
33097 "r19","r20","r21","r22","r23","r24","r25","r26","r27","r28","r29",
33098 "r30","r31","f0","f1","f2","f3","f4","f5","f6","f7","f8","f9",
33099 "f10","f11","f12","f13","f14","f15","f16","f17","f18","f19","f20",
33100 "f21","f22","f23","f24","f25","f26","f27","f28","f29","f30","f31",
33101 "", "pc","ps","cr","lr","ctr","xer"]
33103 -data-list-register-names 1 2 3
33104 ^done,register-names=["r1","r2","r3"]
33108 @subheading The @code{-data-list-register-values} Command
33109 @findex -data-list-register-values
33111 @subsubheading Synopsis
33114 -data-list-register-values
33115 [ @code{--skip-unavailable} ] @var{fmt} [ ( @var{regno} )*]
33118 Display the registers' contents. The format according to which the
33119 registers' contents are to be returned is given by @var{fmt}, followed
33120 by an optional list of numbers specifying the registers to display. A
33121 missing list of numbers indicates that the contents of all the
33122 registers must be returned. The @code{--skip-unavailable} option
33123 indicates that only the available registers are to be returned.
33125 Allowed formats for @var{fmt} are:
33142 @subsubheading @value{GDBN} Command
33144 The corresponding @value{GDBN} commands are @samp{info reg}, @samp{info
33145 all-reg}, and (in @code{gdbtk}) @samp{gdb_fetch_registers}.
33147 @subsubheading Example
33149 For a PPC MBX board (note: line breaks are for readability only, they
33150 don't appear in the actual output):
33154 -data-list-register-values r 64 65
33155 ^done,register-values=[@{number="64",value="0xfe00a300"@},
33156 @{number="65",value="0x00029002"@}]
33158 -data-list-register-values x
33159 ^done,register-values=[@{number="0",value="0xfe0043c8"@},
33160 @{number="1",value="0x3fff88"@},@{number="2",value="0xfffffffe"@},
33161 @{number="3",value="0x0"@},@{number="4",value="0xa"@},
33162 @{number="5",value="0x3fff68"@},@{number="6",value="0x3fff58"@},
33163 @{number="7",value="0xfe011e98"@},@{number="8",value="0x2"@},
33164 @{number="9",value="0xfa202820"@},@{number="10",value="0xfa202808"@},
33165 @{number="11",value="0x1"@},@{number="12",value="0x0"@},
33166 @{number="13",value="0x4544"@},@{number="14",value="0xffdfffff"@},
33167 @{number="15",value="0xffffffff"@},@{number="16",value="0xfffffeff"@},
33168 @{number="17",value="0xefffffed"@},@{number="18",value="0xfffffffe"@},
33169 @{number="19",value="0xffffffff"@},@{number="20",value="0xffffffff"@},
33170 @{number="21",value="0xffffffff"@},@{number="22",value="0xfffffff7"@},
33171 @{number="23",value="0xffffffff"@},@{number="24",value="0xffffffff"@},
33172 @{number="25",value="0xffffffff"@},@{number="26",value="0xfffffffb"@},
33173 @{number="27",value="0xffffffff"@},@{number="28",value="0xf7bfffff"@},
33174 @{number="29",value="0x0"@},@{number="30",value="0xfe010000"@},
33175 @{number="31",value="0x0"@},@{number="32",value="0x0"@},
33176 @{number="33",value="0x0"@},@{number="34",value="0x0"@},
33177 @{number="35",value="0x0"@},@{number="36",value="0x0"@},
33178 @{number="37",value="0x0"@},@{number="38",value="0x0"@},
33179 @{number="39",value="0x0"@},@{number="40",value="0x0"@},
33180 @{number="41",value="0x0"@},@{number="42",value="0x0"@},
33181 @{number="43",value="0x0"@},@{number="44",value="0x0"@},
33182 @{number="45",value="0x0"@},@{number="46",value="0x0"@},
33183 @{number="47",value="0x0"@},@{number="48",value="0x0"@},
33184 @{number="49",value="0x0"@},@{number="50",value="0x0"@},
33185 @{number="51",value="0x0"@},@{number="52",value="0x0"@},
33186 @{number="53",value="0x0"@},@{number="54",value="0x0"@},
33187 @{number="55",value="0x0"@},@{number="56",value="0x0"@},
33188 @{number="57",value="0x0"@},@{number="58",value="0x0"@},
33189 @{number="59",value="0x0"@},@{number="60",value="0x0"@},
33190 @{number="61",value="0x0"@},@{number="62",value="0x0"@},
33191 @{number="63",value="0x0"@},@{number="64",value="0xfe00a300"@},
33192 @{number="65",value="0x29002"@},@{number="66",value="0x202f04b5"@},
33193 @{number="67",value="0xfe0043b0"@},@{number="68",value="0xfe00b3e4"@},
33194 @{number="69",value="0x20002b03"@}]
33199 @subheading The @code{-data-read-memory} Command
33200 @findex -data-read-memory
33202 This command is deprecated, use @code{-data-read-memory-bytes} instead.
33204 @subsubheading Synopsis
33207 -data-read-memory [ -o @var{byte-offset} ]
33208 @var{address} @var{word-format} @var{word-size}
33209 @var{nr-rows} @var{nr-cols} [ @var{aschar} ]
33216 @item @var{address}
33217 An expression specifying the address of the first memory word to be
33218 read. Complex expressions containing embedded white space should be
33219 quoted using the C convention.
33221 @item @var{word-format}
33222 The format to be used to print the memory words. The notation is the
33223 same as for @value{GDBN}'s @code{print} command (@pxref{Output Formats,
33226 @item @var{word-size}
33227 The size of each memory word in bytes.
33229 @item @var{nr-rows}
33230 The number of rows in the output table.
33232 @item @var{nr-cols}
33233 The number of columns in the output table.
33236 If present, indicates that each row should include an @sc{ascii} dump. The
33237 value of @var{aschar} is used as a padding character when a byte is not a
33238 member of the printable @sc{ascii} character set (printable @sc{ascii}
33239 characters are those whose code is between 32 and 126, inclusively).
33241 @item @var{byte-offset}
33242 An offset to add to the @var{address} before fetching memory.
33245 This command displays memory contents as a table of @var{nr-rows} by
33246 @var{nr-cols} words, each word being @var{word-size} bytes. In total,
33247 @code{@var{nr-rows} * @var{nr-cols} * @var{word-size}} bytes are read
33248 (returned as @samp{total-bytes}). Should less than the requested number
33249 of bytes be returned by the target, the missing words are identified
33250 using @samp{N/A}. The number of bytes read from the target is returned
33251 in @samp{nr-bytes} and the starting address used to read memory in
33254 The address of the next/previous row or page is available in
33255 @samp{next-row} and @samp{prev-row}, @samp{next-page} and
33258 @subsubheading @value{GDBN} Command
33260 The corresponding @value{GDBN} command is @samp{x}. @code{gdbtk} has
33261 @samp{gdb_get_mem} memory read command.
33263 @subsubheading Example
33265 Read six bytes of memory starting at @code{bytes+6} but then offset by
33266 @code{-6} bytes. Format as three rows of two columns. One byte per
33267 word. Display each word in hex.
33271 9-data-read-memory -o -6 -- bytes+6 x 1 3 2
33272 9^done,addr="0x00001390",nr-bytes="6",total-bytes="6",
33273 next-row="0x00001396",prev-row="0x0000138e",next-page="0x00001396",
33274 prev-page="0x0000138a",memory=[
33275 @{addr="0x00001390",data=["0x00","0x01"]@},
33276 @{addr="0x00001392",data=["0x02","0x03"]@},
33277 @{addr="0x00001394",data=["0x04","0x05"]@}]
33281 Read two bytes of memory starting at address @code{shorts + 64} and
33282 display as a single word formatted in decimal.
33286 5-data-read-memory shorts+64 d 2 1 1
33287 5^done,addr="0x00001510",nr-bytes="2",total-bytes="2",
33288 next-row="0x00001512",prev-row="0x0000150e",
33289 next-page="0x00001512",prev-page="0x0000150e",memory=[
33290 @{addr="0x00001510",data=["128"]@}]
33294 Read thirty two bytes of memory starting at @code{bytes+16} and format
33295 as eight rows of four columns. Include a string encoding with @samp{x}
33296 used as the non-printable character.
33300 4-data-read-memory bytes+16 x 1 8 4 x
33301 4^done,addr="0x000013a0",nr-bytes="32",total-bytes="32",
33302 next-row="0x000013c0",prev-row="0x0000139c",
33303 next-page="0x000013c0",prev-page="0x00001380",memory=[
33304 @{addr="0x000013a0",data=["0x10","0x11","0x12","0x13"],ascii="xxxx"@},
33305 @{addr="0x000013a4",data=["0x14","0x15","0x16","0x17"],ascii="xxxx"@},
33306 @{addr="0x000013a8",data=["0x18","0x19","0x1a","0x1b"],ascii="xxxx"@},
33307 @{addr="0x000013ac",data=["0x1c","0x1d","0x1e","0x1f"],ascii="xxxx"@},
33308 @{addr="0x000013b0",data=["0x20","0x21","0x22","0x23"],ascii=" !\"#"@},
33309 @{addr="0x000013b4",data=["0x24","0x25","0x26","0x27"],ascii="$%&'"@},
33310 @{addr="0x000013b8",data=["0x28","0x29","0x2a","0x2b"],ascii="()*+"@},
33311 @{addr="0x000013bc",data=["0x2c","0x2d","0x2e","0x2f"],ascii=",-./"@}]
33315 @subheading The @code{-data-read-memory-bytes} Command
33316 @findex -data-read-memory-bytes
33318 @subsubheading Synopsis
33321 -data-read-memory-bytes [ -o @var{offset} ]
33322 @var{address} @var{count}
33329 @item @var{address}
33330 An expression specifying the address of the first addressable memory unit
33331 to be read. Complex expressions containing embedded white space should be
33332 quoted using the C convention.
33335 The number of addressable memory units to read. This should be an integer
33339 The offset relative to @var{address} at which to start reading. This
33340 should be an integer literal. This option is provided so that a frontend
33341 is not required to first evaluate address and then perform address
33342 arithmetics itself.
33346 This command attempts to read all accessible memory regions in the
33347 specified range. First, all regions marked as unreadable in the memory
33348 map (if one is defined) will be skipped. @xref{Memory Region
33349 Attributes}. Second, @value{GDBN} will attempt to read the remaining
33350 regions. For each one, if reading full region results in an errors,
33351 @value{GDBN} will try to read a subset of the region.
33353 In general, every single memory unit in the region may be readable or not,
33354 and the only way to read every readable unit is to try a read at
33355 every address, which is not practical. Therefore, @value{GDBN} will
33356 attempt to read all accessible memory units at either beginning or the end
33357 of the region, using a binary division scheme. This heuristic works
33358 well for reading across a memory map boundary. Note that if a region
33359 has a readable range that is neither at the beginning or the end,
33360 @value{GDBN} will not read it.
33362 The result record (@pxref{GDB/MI Result Records}) that is output of
33363 the command includes a field named @samp{memory} whose content is a
33364 list of tuples. Each tuple represent a successfully read memory block
33365 and has the following fields:
33369 The start address of the memory block, as hexadecimal literal.
33372 The end address of the memory block, as hexadecimal literal.
33375 The offset of the memory block, as hexadecimal literal, relative to
33376 the start address passed to @code{-data-read-memory-bytes}.
33379 The contents of the memory block, in hex.
33385 @subsubheading @value{GDBN} Command
33387 The corresponding @value{GDBN} command is @samp{x}.
33389 @subsubheading Example
33393 -data-read-memory-bytes &a 10
33394 ^done,memory=[@{begin="0xbffff154",offset="0x00000000",
33396 contents="01000000020000000300"@}]
33401 @subheading The @code{-data-write-memory-bytes} Command
33402 @findex -data-write-memory-bytes
33404 @subsubheading Synopsis
33407 -data-write-memory-bytes @var{address} @var{contents}
33408 -data-write-memory-bytes @var{address} @var{contents} @r{[}@var{count}@r{]}
33415 @item @var{address}
33416 An expression specifying the address of the first addressable memory unit
33417 to be written. Complex expressions containing embedded white space should
33418 be quoted using the C convention.
33420 @item @var{contents}
33421 The hex-encoded data to write. It is an error if @var{contents} does
33422 not represent an integral number of addressable memory units.
33425 Optional argument indicating the number of addressable memory units to be
33426 written. If @var{count} is greater than @var{contents}' length,
33427 @value{GDBN} will repeatedly write @var{contents} until it fills
33428 @var{count} memory units.
33432 @subsubheading @value{GDBN} Command
33434 There's no corresponding @value{GDBN} command.
33436 @subsubheading Example
33440 -data-write-memory-bytes &a "aabbccdd"
33447 -data-write-memory-bytes &a "aabbccdd" 16e
33452 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
33453 @node GDB/MI Tracepoint Commands
33454 @section @sc{gdb/mi} Tracepoint Commands
33456 The commands defined in this section implement MI support for
33457 tracepoints. For detailed introduction, see @ref{Tracepoints}.
33459 @subheading The @code{-trace-find} Command
33460 @findex -trace-find
33462 @subsubheading Synopsis
33465 -trace-find @var{mode} [@var{parameters}@dots{}]
33468 Find a trace frame using criteria defined by @var{mode} and
33469 @var{parameters}. The following table lists permissible
33470 modes and their parameters. For details of operation, see @ref{tfind}.
33475 No parameters are required. Stops examining trace frames.
33478 An integer is required as parameter. Selects tracepoint frame with
33481 @item tracepoint-number
33482 An integer is required as parameter. Finds next
33483 trace frame that corresponds to tracepoint with the specified number.
33486 An address is required as parameter. Finds
33487 next trace frame that corresponds to any tracepoint at the specified
33490 @item pc-inside-range
33491 Two addresses are required as parameters. Finds next trace
33492 frame that corresponds to a tracepoint at an address inside the
33493 specified range. Both bounds are considered to be inside the range.
33495 @item pc-outside-range
33496 Two addresses are required as parameters. Finds
33497 next trace frame that corresponds to a tracepoint at an address outside
33498 the specified range. Both bounds are considered to be inside the range.
33501 Line specification is required as parameter. @xref{Specify Location}.
33502 Finds next trace frame that corresponds to a tracepoint at
33503 the specified location.
33507 If @samp{none} was passed as @var{mode}, the response does not
33508 have fields. Otherwise, the response may have the following fields:
33512 This field has either @samp{0} or @samp{1} as the value, depending
33513 on whether a matching tracepoint was found.
33516 The index of the found traceframe. This field is present iff
33517 the @samp{found} field has value of @samp{1}.
33520 The index of the found tracepoint. This field is present iff
33521 the @samp{found} field has value of @samp{1}.
33524 The information about the frame corresponding to the found trace
33525 frame. This field is present only if a trace frame was found.
33526 @xref{GDB/MI Frame Information}, for description of this field.
33530 @subsubheading @value{GDBN} Command
33532 The corresponding @value{GDBN} command is @samp{tfind}.
33534 @subheading -trace-define-variable
33535 @findex -trace-define-variable
33537 @subsubheading Synopsis
33540 -trace-define-variable @var{name} [ @var{value} ]
33543 Create trace variable @var{name} if it does not exist. If
33544 @var{value} is specified, sets the initial value of the specified
33545 trace variable to that value. Note that the @var{name} should start
33546 with the @samp{$} character.
33548 @subsubheading @value{GDBN} Command
33550 The corresponding @value{GDBN} command is @samp{tvariable}.
33552 @subheading The @code{-trace-frame-collected} Command
33553 @findex -trace-frame-collected
33555 @subsubheading Synopsis
33558 -trace-frame-collected
33559 [--var-print-values @var{var_pval}]
33560 [--comp-print-values @var{comp_pval}]
33561 [--registers-format @var{regformat}]
33562 [--memory-contents]
33565 This command returns the set of collected objects, register names,
33566 trace state variable names, memory ranges and computed expressions
33567 that have been collected at a particular trace frame. The optional
33568 parameters to the command affect the output format in different ways.
33569 See the output description table below for more details.
33571 The reported names can be used in the normal manner to create
33572 varobjs and inspect the objects themselves. The items returned by
33573 this command are categorized so that it is clear which is a variable,
33574 which is a register, which is a trace state variable, which is a
33575 memory range and which is a computed expression.
33577 For instance, if the actions were
33579 collect myVar, myArray[myIndex], myObj.field, myPtr->field, myCount + 2
33580 collect *(int*)0xaf02bef0@@40
33584 the object collected in its entirety would be @code{myVar}. The
33585 object @code{myArray} would be partially collected, because only the
33586 element at index @code{myIndex} would be collected. The remaining
33587 objects would be computed expressions.
33589 An example output would be:
33593 -trace-frame-collected
33595 explicit-variables=[@{name="myVar",value="1"@}],
33596 computed-expressions=[@{name="myArray[myIndex]",value="0"@},
33597 @{name="myObj.field",value="0"@},
33598 @{name="myPtr->field",value="1"@},
33599 @{name="myCount + 2",value="3"@},
33600 @{name="$tvar1 + 1",value="43970027"@}],
33601 registers=[@{number="0",value="0x7fe2c6e79ec8"@},
33602 @{number="1",value="0x0"@},
33603 @{number="2",value="0x4"@},
33605 @{number="125",value="0x0"@}],
33606 tvars=[@{name="$tvar1",current="43970026"@}],
33607 memory=[@{address="0x0000000000602264",length="4"@},
33608 @{address="0x0000000000615bc0",length="4"@}]
33615 @item explicit-variables
33616 The set of objects that have been collected in their entirety (as
33617 opposed to collecting just a few elements of an array or a few struct
33618 members). For each object, its name and value are printed.
33619 The @code{--var-print-values} option affects how or whether the value
33620 field is output. If @var{var_pval} is 0, then print only the names;
33621 if it is 1, print also their values; and if it is 2, print the name,
33622 type and value for simple data types, and the name and type for
33623 arrays, structures and unions.
33625 @item computed-expressions
33626 The set of computed expressions that have been collected at the
33627 current trace frame. The @code{--comp-print-values} option affects
33628 this set like the @code{--var-print-values} option affects the
33629 @code{explicit-variables} set. See above.
33632 The registers that have been collected at the current trace frame.
33633 For each register collected, the name and current value are returned.
33634 The value is formatted according to the @code{--registers-format}
33635 option. See the @command{-data-list-register-values} command for a
33636 list of the allowed formats. The default is @samp{x}.
33639 The trace state variables that have been collected at the current
33640 trace frame. For each trace state variable collected, the name and
33641 current value are returned.
33644 The set of memory ranges that have been collected at the current trace
33645 frame. Its content is a list of tuples. Each tuple represents a
33646 collected memory range and has the following fields:
33650 The start address of the memory range, as hexadecimal literal.
33653 The length of the memory range, as decimal literal.
33656 The contents of the memory block, in hex. This field is only present
33657 if the @code{--memory-contents} option is specified.
33663 @subsubheading @value{GDBN} Command
33665 There is no corresponding @value{GDBN} command.
33667 @subsubheading Example
33669 @subheading -trace-list-variables
33670 @findex -trace-list-variables
33672 @subsubheading Synopsis
33675 -trace-list-variables
33678 Return a table of all defined trace variables. Each element of the
33679 table has the following fields:
33683 The name of the trace variable. This field is always present.
33686 The initial value. This is a 64-bit signed integer. This
33687 field is always present.
33690 The value the trace variable has at the moment. This is a 64-bit
33691 signed integer. This field is absent iff current value is
33692 not defined, for example if the trace was never run, or is
33697 @subsubheading @value{GDBN} Command
33699 The corresponding @value{GDBN} command is @samp{tvariables}.
33701 @subsubheading Example
33705 -trace-list-variables
33706 ^done,trace-variables=@{nr_rows="1",nr_cols="3",
33707 hdr=[@{width="15",alignment="-1",col_name="name",colhdr="Name"@},
33708 @{width="11",alignment="-1",col_name="initial",colhdr="Initial"@},
33709 @{width="11",alignment="-1",col_name="current",colhdr="Current"@}],
33710 body=[variable=@{name="$trace_timestamp",initial="0"@}
33711 variable=@{name="$foo",initial="10",current="15"@}]@}
33715 @subheading -trace-save
33716 @findex -trace-save
33718 @subsubheading Synopsis
33721 -trace-save [ -r ] [ -ctf ] @var{filename}
33724 Saves the collected trace data to @var{filename}. Without the
33725 @samp{-r} option, the data is downloaded from the target and saved
33726 in a local file. With the @samp{-r} option the target is asked
33727 to perform the save.
33729 By default, this command will save the trace in the tfile format. You can
33730 supply the optional @samp{-ctf} argument to save it the CTF format. See
33731 @ref{Trace Files} for more information about CTF.
33733 @subsubheading @value{GDBN} Command
33735 The corresponding @value{GDBN} command is @samp{tsave}.
33738 @subheading -trace-start
33739 @findex -trace-start
33741 @subsubheading Synopsis
33747 Starts a tracing experiment. The result of this command does not
33750 @subsubheading @value{GDBN} Command
33752 The corresponding @value{GDBN} command is @samp{tstart}.
33754 @subheading -trace-status
33755 @findex -trace-status
33757 @subsubheading Synopsis
33763 Obtains the status of a tracing experiment. The result may include
33764 the following fields:
33769 May have a value of either @samp{0}, when no tracing operations are
33770 supported, @samp{1}, when all tracing operations are supported, or
33771 @samp{file} when examining trace file. In the latter case, examining
33772 of trace frame is possible but new tracing experiement cannot be
33773 started. This field is always present.
33776 May have a value of either @samp{0} or @samp{1} depending on whether
33777 tracing experiement is in progress on target. This field is present
33778 if @samp{supported} field is not @samp{0}.
33781 Report the reason why the tracing was stopped last time. This field
33782 may be absent iff tracing was never stopped on target yet. The
33783 value of @samp{request} means the tracing was stopped as result of
33784 the @code{-trace-stop} command. The value of @samp{overflow} means
33785 the tracing buffer is full. The value of @samp{disconnection} means
33786 tracing was automatically stopped when @value{GDBN} has disconnected.
33787 The value of @samp{passcount} means tracing was stopped when a
33788 tracepoint was passed a maximal number of times for that tracepoint.
33789 This field is present if @samp{supported} field is not @samp{0}.
33791 @item stopping-tracepoint
33792 The number of tracepoint whose passcount as exceeded. This field is
33793 present iff the @samp{stop-reason} field has the value of
33797 @itemx frames-created
33798 The @samp{frames} field is a count of the total number of trace frames
33799 in the trace buffer, while @samp{frames-created} is the total created
33800 during the run, including ones that were discarded, such as when a
33801 circular trace buffer filled up. Both fields are optional.
33805 These fields tell the current size of the tracing buffer and the
33806 remaining space. These fields are optional.
33809 The value of the circular trace buffer flag. @code{1} means that the
33810 trace buffer is circular and old trace frames will be discarded if
33811 necessary to make room, @code{0} means that the trace buffer is linear
33815 The value of the disconnected tracing flag. @code{1} means that
33816 tracing will continue after @value{GDBN} disconnects, @code{0} means
33817 that the trace run will stop.
33820 The filename of the trace file being examined. This field is
33821 optional, and only present when examining a trace file.
33825 @subsubheading @value{GDBN} Command
33827 The corresponding @value{GDBN} command is @samp{tstatus}.
33829 @subheading -trace-stop
33830 @findex -trace-stop
33832 @subsubheading Synopsis
33838 Stops a tracing experiment. The result of this command has the same
33839 fields as @code{-trace-status}, except that the @samp{supported} and
33840 @samp{running} fields are not output.
33842 @subsubheading @value{GDBN} Command
33844 The corresponding @value{GDBN} command is @samp{tstop}.
33847 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
33848 @node GDB/MI Symbol Query
33849 @section @sc{gdb/mi} Symbol Query Commands
33853 @subheading The @code{-symbol-info-address} Command
33854 @findex -symbol-info-address
33856 @subsubheading Synopsis
33859 -symbol-info-address @var{symbol}
33862 Describe where @var{symbol} is stored.
33864 @subsubheading @value{GDBN} Command
33866 The corresponding @value{GDBN} command is @samp{info address}.
33868 @subsubheading Example
33872 @subheading The @code{-symbol-info-file} Command
33873 @findex -symbol-info-file
33875 @subsubheading Synopsis
33881 Show the file for the symbol.
33883 @subsubheading @value{GDBN} Command
33885 There's no equivalent @value{GDBN} command. @code{gdbtk} has
33886 @samp{gdb_find_file}.
33888 @subsubheading Example
33892 @subheading The @code{-symbol-info-function} Command
33893 @findex -symbol-info-function
33895 @subsubheading Synopsis
33898 -symbol-info-function
33901 Show which function the symbol lives in.
33903 @subsubheading @value{GDBN} Command
33905 @samp{gdb_get_function} in @code{gdbtk}.
33907 @subsubheading Example
33911 @subheading The @code{-symbol-info-line} Command
33912 @findex -symbol-info-line
33914 @subsubheading Synopsis
33920 Show the core addresses of the code for a source line.
33922 @subsubheading @value{GDBN} Command
33924 The corresponding @value{GDBN} command is @samp{info line}.
33925 @code{gdbtk} has the @samp{gdb_get_line} and @samp{gdb_get_file} commands.
33927 @subsubheading Example
33931 @subheading The @code{-symbol-info-symbol} Command
33932 @findex -symbol-info-symbol
33934 @subsubheading Synopsis
33937 -symbol-info-symbol @var{addr}
33940 Describe what symbol is at location @var{addr}.
33942 @subsubheading @value{GDBN} Command
33944 The corresponding @value{GDBN} command is @samp{info symbol}.
33946 @subsubheading Example
33950 @subheading The @code{-symbol-list-functions} Command
33951 @findex -symbol-list-functions
33953 @subsubheading Synopsis
33956 -symbol-list-functions
33959 List the functions in the executable.
33961 @subsubheading @value{GDBN} Command
33963 @samp{info functions} in @value{GDBN}, @samp{gdb_listfunc} and
33964 @samp{gdb_search} in @code{gdbtk}.
33966 @subsubheading Example
33971 @subheading The @code{-symbol-list-lines} Command
33972 @findex -symbol-list-lines
33974 @subsubheading Synopsis
33977 -symbol-list-lines @var{filename}
33980 Print the list of lines that contain code and their associated program
33981 addresses for the given source filename. The entries are sorted in
33982 ascending PC order.
33984 @subsubheading @value{GDBN} Command
33986 There is no corresponding @value{GDBN} command.
33988 @subsubheading Example
33991 -symbol-list-lines basics.c
33992 ^done,lines=[@{pc="0x08048554",line="7"@},@{pc="0x0804855a",line="8"@}]
33998 @subheading The @code{-symbol-list-types} Command
33999 @findex -symbol-list-types
34001 @subsubheading Synopsis
34007 List all the type names.
34009 @subsubheading @value{GDBN} Command
34011 The corresponding commands are @samp{info types} in @value{GDBN},
34012 @samp{gdb_search} in @code{gdbtk}.
34014 @subsubheading Example
34018 @subheading The @code{-symbol-list-variables} Command
34019 @findex -symbol-list-variables
34021 @subsubheading Synopsis
34024 -symbol-list-variables
34027 List all the global and static variable names.
34029 @subsubheading @value{GDBN} Command
34031 @samp{info variables} in @value{GDBN}, @samp{gdb_search} in @code{gdbtk}.
34033 @subsubheading Example
34037 @subheading The @code{-symbol-locate} Command
34038 @findex -symbol-locate
34040 @subsubheading Synopsis
34046 @subsubheading @value{GDBN} Command
34048 @samp{gdb_loc} in @code{gdbtk}.
34050 @subsubheading Example
34054 @subheading The @code{-symbol-type} Command
34055 @findex -symbol-type
34057 @subsubheading Synopsis
34060 -symbol-type @var{variable}
34063 Show type of @var{variable}.
34065 @subsubheading @value{GDBN} Command
34067 The corresponding @value{GDBN} command is @samp{ptype}, @code{gdbtk} has
34068 @samp{gdb_obj_variable}.
34070 @subsubheading Example
34075 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
34076 @node GDB/MI File Commands
34077 @section @sc{gdb/mi} File Commands
34079 This section describes the GDB/MI commands to specify executable file names
34080 and to read in and obtain symbol table information.
34082 @subheading The @code{-file-exec-and-symbols} Command
34083 @findex -file-exec-and-symbols
34085 @subsubheading Synopsis
34088 -file-exec-and-symbols @var{file}
34091 Specify the executable file to be debugged. This file is the one from
34092 which the symbol table is also read. If no file is specified, the
34093 command clears the executable and symbol information. If breakpoints
34094 are set when using this command with no arguments, @value{GDBN} will produce
34095 error messages. Otherwise, no output is produced, except a completion
34098 @subsubheading @value{GDBN} Command
34100 The corresponding @value{GDBN} command is @samp{file}.
34102 @subsubheading Example
34106 -file-exec-and-symbols /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
34112 @subheading The @code{-file-exec-file} Command
34113 @findex -file-exec-file
34115 @subsubheading Synopsis
34118 -file-exec-file @var{file}
34121 Specify the executable file to be debugged. Unlike
34122 @samp{-file-exec-and-symbols}, the symbol table is @emph{not} read
34123 from this file. If used without argument, @value{GDBN} clears the information
34124 about the executable file. No output is produced, except a completion
34127 @subsubheading @value{GDBN} Command
34129 The corresponding @value{GDBN} command is @samp{exec-file}.
34131 @subsubheading Example
34135 -file-exec-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
34142 @subheading The @code{-file-list-exec-sections} Command
34143 @findex -file-list-exec-sections
34145 @subsubheading Synopsis
34148 -file-list-exec-sections
34151 List the sections of the current executable file.
34153 @subsubheading @value{GDBN} Command
34155 The @value{GDBN} command @samp{info file} shows, among the rest, the same
34156 information as this command. @code{gdbtk} has a corresponding command
34157 @samp{gdb_load_info}.
34159 @subsubheading Example
34164 @subheading The @code{-file-list-exec-source-file} Command
34165 @findex -file-list-exec-source-file
34167 @subsubheading Synopsis
34170 -file-list-exec-source-file
34173 List the line number, the current source file, and the absolute path
34174 to the current source file for the current executable. The macro
34175 information field has a value of @samp{1} or @samp{0} depending on
34176 whether or not the file includes preprocessor macro information.
34178 @subsubheading @value{GDBN} Command
34180 The @value{GDBN} equivalent is @samp{info source}
34182 @subsubheading Example
34186 123-file-list-exec-source-file
34187 123^done,line="1",file="foo.c",fullname="/home/bar/foo.c,macro-info="1"
34192 @subheading The @code{-file-list-exec-source-files} Command
34193 @findex -file-list-exec-source-files
34195 @subsubheading Synopsis
34198 -file-list-exec-source-files
34201 List the source files for the current executable.
34203 It will always output both the filename and fullname (absolute file
34204 name) of a source file.
34206 @subsubheading @value{GDBN} Command
34208 The @value{GDBN} equivalent is @samp{info sources}.
34209 @code{gdbtk} has an analogous command @samp{gdb_listfiles}.
34211 @subsubheading Example
34214 -file-list-exec-source-files
34216 @{file=foo.c,fullname=/home/foo.c@},
34217 @{file=/home/bar.c,fullname=/home/bar.c@},
34218 @{file=gdb_could_not_find_fullpath.c@}]
34222 @subheading The @code{-file-list-shared-libraries} Command
34223 @findex -file-list-shared-libraries
34225 @subsubheading Synopsis
34228 -file-list-shared-libraries [ @var{regexp} ]
34231 List the shared libraries in the program.
34232 With a regular expression @var{regexp}, only those libraries whose
34233 names match @var{regexp} are listed.
34235 @subsubheading @value{GDBN} Command
34237 The corresponding @value{GDBN} command is @samp{info shared}. The fields
34238 have a similar meaning to the @code{=library-loaded} notification.
34239 The @code{ranges} field specifies the multiple segments belonging to this
34240 library. Each range has the following fields:
34244 The address defining the inclusive lower bound of the segment.
34246 The address defining the exclusive upper bound of the segment.
34249 @subsubheading Example
34252 -file-list-exec-source-files
34253 ^done,shared-libraries=[
34254 @{id="/lib/libfoo.so",target-name="/lib/libfoo.so",host-name="/lib/libfoo.so",symbols-loaded="1",thread-group="i1",ranges=[@{from="0x72815989",to="0x728162c0"@}]@},
34255 @{id="/lib/libbar.so",target-name="/lib/libbar.so",host-name="/lib/libbar.so",symbols-loaded="1",thread-group="i1",ranges=[@{from="0x76ee48c0",to="0x76ee9160"@}]@}]
34261 @subheading The @code{-file-list-symbol-files} Command
34262 @findex -file-list-symbol-files
34264 @subsubheading Synopsis
34267 -file-list-symbol-files
34272 @subsubheading @value{GDBN} Command
34274 The corresponding @value{GDBN} command is @samp{info file} (part of it).
34276 @subsubheading Example
34281 @subheading The @code{-file-symbol-file} Command
34282 @findex -file-symbol-file
34284 @subsubheading Synopsis
34287 -file-symbol-file @var{file}
34290 Read symbol table info from the specified @var{file} argument. When
34291 used without arguments, clears @value{GDBN}'s symbol table info. No output is
34292 produced, except for a completion notification.
34294 @subsubheading @value{GDBN} Command
34296 The corresponding @value{GDBN} command is @samp{symbol-file}.
34298 @subsubheading Example
34302 -file-symbol-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
34308 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
34309 @node GDB/MI Memory Overlay Commands
34310 @section @sc{gdb/mi} Memory Overlay Commands
34312 The memory overlay commands are not implemented.
34314 @c @subheading -overlay-auto
34316 @c @subheading -overlay-list-mapping-state
34318 @c @subheading -overlay-list-overlays
34320 @c @subheading -overlay-map
34322 @c @subheading -overlay-off
34324 @c @subheading -overlay-on
34326 @c @subheading -overlay-unmap
34328 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
34329 @node GDB/MI Signal Handling Commands
34330 @section @sc{gdb/mi} Signal Handling Commands
34332 Signal handling commands are not implemented.
34334 @c @subheading -signal-handle
34336 @c @subheading -signal-list-handle-actions
34338 @c @subheading -signal-list-signal-types
34342 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
34343 @node GDB/MI Target Manipulation
34344 @section @sc{gdb/mi} Target Manipulation Commands
34347 @subheading The @code{-target-attach} Command
34348 @findex -target-attach
34350 @subsubheading Synopsis
34353 -target-attach @var{pid} | @var{gid} | @var{file}
34356 Attach to a process @var{pid} or a file @var{file} outside of
34357 @value{GDBN}, or a thread group @var{gid}. If attaching to a thread
34358 group, the id previously returned by
34359 @samp{-list-thread-groups --available} must be used.
34361 @subsubheading @value{GDBN} Command
34363 The corresponding @value{GDBN} command is @samp{attach}.
34365 @subsubheading Example
34369 =thread-created,id="1"
34370 *stopped,thread-id="1",frame=@{addr="0xb7f7e410",func="bar",args=[]@}
34376 @subheading The @code{-target-compare-sections} Command
34377 @findex -target-compare-sections
34379 @subsubheading Synopsis
34382 -target-compare-sections [ @var{section} ]
34385 Compare data of section @var{section} on target to the exec file.
34386 Without the argument, all sections are compared.
34388 @subsubheading @value{GDBN} Command
34390 The @value{GDBN} equivalent is @samp{compare-sections}.
34392 @subsubheading Example
34397 @subheading The @code{-target-detach} Command
34398 @findex -target-detach
34400 @subsubheading Synopsis
34403 -target-detach [ @var{pid} | @var{gid} ]
34406 Detach from the remote target which normally resumes its execution.
34407 If either @var{pid} or @var{gid} is specified, detaches from either
34408 the specified process, or specified thread group. There's no output.
34410 @subsubheading @value{GDBN} Command
34412 The corresponding @value{GDBN} command is @samp{detach}.
34414 @subsubheading Example
34424 @subheading The @code{-target-disconnect} Command
34425 @findex -target-disconnect
34427 @subsubheading Synopsis
34433 Disconnect from the remote target. There's no output and the target is
34434 generally not resumed.
34436 @subsubheading @value{GDBN} Command
34438 The corresponding @value{GDBN} command is @samp{disconnect}.
34440 @subsubheading Example
34450 @subheading The @code{-target-download} Command
34451 @findex -target-download
34453 @subsubheading Synopsis
34459 Loads the executable onto the remote target.
34460 It prints out an update message every half second, which includes the fields:
34464 The name of the section.
34466 The size of what has been sent so far for that section.
34468 The size of the section.
34470 The total size of what was sent so far (the current and the previous sections).
34472 The size of the overall executable to download.
34476 Each message is sent as status record (@pxref{GDB/MI Output Syntax, ,
34477 @sc{gdb/mi} Output Syntax}).
34479 In addition, it prints the name and size of the sections, as they are
34480 downloaded. These messages include the following fields:
34484 The name of the section.
34486 The size of the section.
34488 The size of the overall executable to download.
34492 At the end, a summary is printed.
34494 @subsubheading @value{GDBN} Command
34496 The corresponding @value{GDBN} command is @samp{load}.
34498 @subsubheading Example
34500 Note: each status message appears on a single line. Here the messages
34501 have been broken down so that they can fit onto a page.
34506 +download,@{section=".text",section-size="6668",total-size="9880"@}
34507 +download,@{section=".text",section-sent="512",section-size="6668",
34508 total-sent="512",total-size="9880"@}
34509 +download,@{section=".text",section-sent="1024",section-size="6668",
34510 total-sent="1024",total-size="9880"@}
34511 +download,@{section=".text",section-sent="1536",section-size="6668",
34512 total-sent="1536",total-size="9880"@}
34513 +download,@{section=".text",section-sent="2048",section-size="6668",
34514 total-sent="2048",total-size="9880"@}
34515 +download,@{section=".text",section-sent="2560",section-size="6668",
34516 total-sent="2560",total-size="9880"@}
34517 +download,@{section=".text",section-sent="3072",section-size="6668",
34518 total-sent="3072",total-size="9880"@}
34519 +download,@{section=".text",section-sent="3584",section-size="6668",
34520 total-sent="3584",total-size="9880"@}
34521 +download,@{section=".text",section-sent="4096",section-size="6668",
34522 total-sent="4096",total-size="9880"@}
34523 +download,@{section=".text",section-sent="4608",section-size="6668",
34524 total-sent="4608",total-size="9880"@}
34525 +download,@{section=".text",section-sent="5120",section-size="6668",
34526 total-sent="5120",total-size="9880"@}
34527 +download,@{section=".text",section-sent="5632",section-size="6668",
34528 total-sent="5632",total-size="9880"@}
34529 +download,@{section=".text",section-sent="6144",section-size="6668",
34530 total-sent="6144",total-size="9880"@}
34531 +download,@{section=".text",section-sent="6656",section-size="6668",
34532 total-sent="6656",total-size="9880"@}
34533 +download,@{section=".init",section-size="28",total-size="9880"@}
34534 +download,@{section=".fini",section-size="28",total-size="9880"@}
34535 +download,@{section=".data",section-size="3156",total-size="9880"@}
34536 +download,@{section=".data",section-sent="512",section-size="3156",
34537 total-sent="7236",total-size="9880"@}
34538 +download,@{section=".data",section-sent="1024",section-size="3156",
34539 total-sent="7748",total-size="9880"@}
34540 +download,@{section=".data",section-sent="1536",section-size="3156",
34541 total-sent="8260",total-size="9880"@}
34542 +download,@{section=".data",section-sent="2048",section-size="3156",
34543 total-sent="8772",total-size="9880"@}
34544 +download,@{section=".data",section-sent="2560",section-size="3156",
34545 total-sent="9284",total-size="9880"@}
34546 +download,@{section=".data",section-sent="3072",section-size="3156",
34547 total-sent="9796",total-size="9880"@}
34548 ^done,address="0x10004",load-size="9880",transfer-rate="6586",
34555 @subheading The @code{-target-exec-status} Command
34556 @findex -target-exec-status
34558 @subsubheading Synopsis
34561 -target-exec-status
34564 Provide information on the state of the target (whether it is running or
34565 not, for instance).
34567 @subsubheading @value{GDBN} Command
34569 There's no equivalent @value{GDBN} command.
34571 @subsubheading Example
34575 @subheading The @code{-target-list-available-targets} Command
34576 @findex -target-list-available-targets
34578 @subsubheading Synopsis
34581 -target-list-available-targets
34584 List the possible targets to connect to.
34586 @subsubheading @value{GDBN} Command
34588 The corresponding @value{GDBN} command is @samp{help target}.
34590 @subsubheading Example
34594 @subheading The @code{-target-list-current-targets} Command
34595 @findex -target-list-current-targets
34597 @subsubheading Synopsis
34600 -target-list-current-targets
34603 Describe the current target.
34605 @subsubheading @value{GDBN} Command
34607 The corresponding information is printed by @samp{info file} (among
34610 @subsubheading Example
34614 @subheading The @code{-target-list-parameters} Command
34615 @findex -target-list-parameters
34617 @subsubheading Synopsis
34620 -target-list-parameters
34626 @subsubheading @value{GDBN} Command
34630 @subsubheading Example
34633 @subheading The @code{-target-flash-erase} Command
34634 @findex -target-flash-erase
34636 @subsubheading Synopsis
34639 -target-flash-erase
34642 Erases all known flash memory regions on the target.
34644 The corresponding @value{GDBN} command is @samp{flash-erase}.
34646 The output is a list of flash regions that have been erased, with starting
34647 addresses and memory region sizes.
34651 -target-flash-erase
34652 ^done,erased-regions=@{address="0x0",size="0x40000"@}
34656 @subheading The @code{-target-select} Command
34657 @findex -target-select
34659 @subsubheading Synopsis
34662 -target-select @var{type} @var{parameters @dots{}}
34665 Connect @value{GDBN} to the remote target. This command takes two args:
34669 The type of target, for instance @samp{remote}, etc.
34670 @item @var{parameters}
34671 Device names, host names and the like. @xref{Target Commands, ,
34672 Commands for Managing Targets}, for more details.
34675 The output is a connection notification, followed by the address at
34676 which the target program is, in the following form:
34679 ^connected,addr="@var{address}",func="@var{function name}",
34680 args=[@var{arg list}]
34683 @subsubheading @value{GDBN} Command
34685 The corresponding @value{GDBN} command is @samp{target}.
34687 @subsubheading Example
34691 -target-select remote /dev/ttya
34692 ^connected,addr="0xfe00a300",func="??",args=[]
34696 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
34697 @node GDB/MI File Transfer Commands
34698 @section @sc{gdb/mi} File Transfer Commands
34701 @subheading The @code{-target-file-put} Command
34702 @findex -target-file-put
34704 @subsubheading Synopsis
34707 -target-file-put @var{hostfile} @var{targetfile}
34710 Copy file @var{hostfile} from the host system (the machine running
34711 @value{GDBN}) to @var{targetfile} on the target system.
34713 @subsubheading @value{GDBN} Command
34715 The corresponding @value{GDBN} command is @samp{remote put}.
34717 @subsubheading Example
34721 -target-file-put localfile remotefile
34727 @subheading The @code{-target-file-get} Command
34728 @findex -target-file-get
34730 @subsubheading Synopsis
34733 -target-file-get @var{targetfile} @var{hostfile}
34736 Copy file @var{targetfile} from the target system to @var{hostfile}
34737 on the host system.
34739 @subsubheading @value{GDBN} Command
34741 The corresponding @value{GDBN} command is @samp{remote get}.
34743 @subsubheading Example
34747 -target-file-get remotefile localfile
34753 @subheading The @code{-target-file-delete} Command
34754 @findex -target-file-delete
34756 @subsubheading Synopsis
34759 -target-file-delete @var{targetfile}
34762 Delete @var{targetfile} from the target system.
34764 @subsubheading @value{GDBN} Command
34766 The corresponding @value{GDBN} command is @samp{remote delete}.
34768 @subsubheading Example
34772 -target-file-delete remotefile
34778 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
34779 @node GDB/MI Ada Exceptions Commands
34780 @section Ada Exceptions @sc{gdb/mi} Commands
34782 @subheading The @code{-info-ada-exceptions} Command
34783 @findex -info-ada-exceptions
34785 @subsubheading Synopsis
34788 -info-ada-exceptions [ @var{regexp}]
34791 List all Ada exceptions defined within the program being debugged.
34792 With a regular expression @var{regexp}, only those exceptions whose
34793 names match @var{regexp} are listed.
34795 @subsubheading @value{GDBN} Command
34797 The corresponding @value{GDBN} command is @samp{info exceptions}.
34799 @subsubheading Result
34801 The result is a table of Ada exceptions. The following columns are
34802 defined for each exception:
34806 The name of the exception.
34809 The address of the exception.
34813 @subsubheading Example
34816 -info-ada-exceptions aint
34817 ^done,ada-exceptions=@{nr_rows="2",nr_cols="2",
34818 hdr=[@{width="1",alignment="-1",col_name="name",colhdr="Name"@},
34819 @{width="1",alignment="-1",col_name="address",colhdr="Address"@}],
34820 body=[@{name="constraint_error",address="0x0000000000613da0"@},
34821 @{name="const.aint_global_e",address="0x0000000000613b00"@}]@}
34824 @subheading Catching Ada Exceptions
34826 The commands describing how to ask @value{GDBN} to stop when a program
34827 raises an exception are described at @ref{Ada Exception GDB/MI
34828 Catchpoint Commands}.
34831 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
34832 @node GDB/MI Support Commands
34833 @section @sc{gdb/mi} Support Commands
34835 Since new commands and features get regularly added to @sc{gdb/mi},
34836 some commands are available to help front-ends query the debugger
34837 about support for these capabilities. Similarly, it is also possible
34838 to query @value{GDBN} about target support of certain features.
34840 @subheading The @code{-info-gdb-mi-command} Command
34841 @cindex @code{-info-gdb-mi-command}
34842 @findex -info-gdb-mi-command
34844 @subsubheading Synopsis
34847 -info-gdb-mi-command @var{cmd_name}
34850 Query support for the @sc{gdb/mi} command named @var{cmd_name}.
34852 Note that the dash (@code{-}) starting all @sc{gdb/mi} commands
34853 is technically not part of the command name (@pxref{GDB/MI Input
34854 Syntax}), and thus should be omitted in @var{cmd_name}. However,
34855 for ease of use, this command also accepts the form with the leading
34858 @subsubheading @value{GDBN} Command
34860 There is no corresponding @value{GDBN} command.
34862 @subsubheading Result
34864 The result is a tuple. There is currently only one field:
34868 This field is equal to @code{"true"} if the @sc{gdb/mi} command exists,
34869 @code{"false"} otherwise.
34873 @subsubheading Example
34875 Here is an example where the @sc{gdb/mi} command does not exist:
34878 -info-gdb-mi-command unsupported-command
34879 ^done,command=@{exists="false"@}
34883 And here is an example where the @sc{gdb/mi} command is known
34887 -info-gdb-mi-command symbol-list-lines
34888 ^done,command=@{exists="true"@}
34891 @subheading The @code{-list-features} Command
34892 @findex -list-features
34893 @cindex supported @sc{gdb/mi} features, list
34895 Returns a list of particular features of the MI protocol that
34896 this version of gdb implements. A feature can be a command,
34897 or a new field in an output of some command, or even an
34898 important bugfix. While a frontend can sometimes detect presence
34899 of a feature at runtime, it is easier to perform detection at debugger
34902 The command returns a list of strings, with each string naming an
34903 available feature. Each returned string is just a name, it does not
34904 have any internal structure. The list of possible feature names
34910 (gdb) -list-features
34911 ^done,result=["feature1","feature2"]
34914 The current list of features is:
34917 @item frozen-varobjs
34918 Indicates support for the @code{-var-set-frozen} command, as well
34919 as possible presence of the @code{frozen} field in the output
34920 of @code{-varobj-create}.
34921 @item pending-breakpoints
34922 Indicates support for the @option{-f} option to the @code{-break-insert}
34925 Indicates Python scripting support, Python-based
34926 pretty-printing commands, and possible presence of the
34927 @samp{display_hint} field in the output of @code{-var-list-children}
34929 Indicates support for the @code{-thread-info} command.
34930 @item data-read-memory-bytes
34931 Indicates support for the @code{-data-read-memory-bytes} and the
34932 @code{-data-write-memory-bytes} commands.
34933 @item breakpoint-notifications
34934 Indicates that changes to breakpoints and breakpoints created via the
34935 CLI will be announced via async records.
34936 @item ada-task-info
34937 Indicates support for the @code{-ada-task-info} command.
34938 @item language-option
34939 Indicates that all @sc{gdb/mi} commands accept the @option{--language}
34940 option (@pxref{Context management}).
34941 @item info-gdb-mi-command
34942 Indicates support for the @code{-info-gdb-mi-command} command.
34943 @item undefined-command-error-code
34944 Indicates support for the "undefined-command" error code in error result
34945 records, produced when trying to execute an undefined @sc{gdb/mi} command
34946 (@pxref{GDB/MI Result Records}).
34947 @item exec-run-start-option
34948 Indicates that the @code{-exec-run} command supports the @option{--start}
34949 option (@pxref{GDB/MI Program Execution}).
34950 @item data-disassemble-a-option
34951 Indicates that the @code{-data-disassemble} command supports the @option{-a}
34952 option (@pxref{GDB/MI Data Manipulation}).
34955 @subheading The @code{-list-target-features} Command
34956 @findex -list-target-features
34958 Returns a list of particular features that are supported by the
34959 target. Those features affect the permitted MI commands, but
34960 unlike the features reported by the @code{-list-features} command, the
34961 features depend on which target GDB is using at the moment. Whenever
34962 a target can change, due to commands such as @code{-target-select},
34963 @code{-target-attach} or @code{-exec-run}, the list of target features
34964 may change, and the frontend should obtain it again.
34968 (gdb) -list-target-features
34969 ^done,result=["async"]
34972 The current list of features is:
34976 Indicates that the target is capable of asynchronous command
34977 execution, which means that @value{GDBN} will accept further commands
34978 while the target is running.
34981 Indicates that the target is capable of reverse execution.
34982 @xref{Reverse Execution}, for more information.
34986 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
34987 @node GDB/MI Miscellaneous Commands
34988 @section Miscellaneous @sc{gdb/mi} Commands
34990 @c @subheading -gdb-complete
34992 @subheading The @code{-gdb-exit} Command
34995 @subsubheading Synopsis
35001 Exit @value{GDBN} immediately.
35003 @subsubheading @value{GDBN} Command
35005 Approximately corresponds to @samp{quit}.
35007 @subsubheading Example
35017 @subheading The @code{-exec-abort} Command
35018 @findex -exec-abort
35020 @subsubheading Synopsis
35026 Kill the inferior running program.
35028 @subsubheading @value{GDBN} Command
35030 The corresponding @value{GDBN} command is @samp{kill}.
35032 @subsubheading Example
35037 @subheading The @code{-gdb-set} Command
35040 @subsubheading Synopsis
35046 Set an internal @value{GDBN} variable.
35047 @c IS THIS A DOLLAR VARIABLE? OR SOMETHING LIKE ANNOTATE ?????
35049 @subsubheading @value{GDBN} Command
35051 The corresponding @value{GDBN} command is @samp{set}.
35053 @subsubheading Example
35063 @subheading The @code{-gdb-show} Command
35066 @subsubheading Synopsis
35072 Show the current value of a @value{GDBN} variable.
35074 @subsubheading @value{GDBN} Command
35076 The corresponding @value{GDBN} command is @samp{show}.
35078 @subsubheading Example
35087 @c @subheading -gdb-source
35090 @subheading The @code{-gdb-version} Command
35091 @findex -gdb-version
35093 @subsubheading Synopsis
35099 Show version information for @value{GDBN}. Used mostly in testing.
35101 @subsubheading @value{GDBN} Command
35103 The @value{GDBN} equivalent is @samp{show version}. @value{GDBN} by
35104 default shows this information when you start an interactive session.
35106 @subsubheading Example
35108 @c This example modifies the actual output from GDB to avoid overfull
35114 ~Copyright 2000 Free Software Foundation, Inc.
35115 ~GDB is free software, covered by the GNU General Public License, and
35116 ~you are welcome to change it and/or distribute copies of it under
35117 ~ certain conditions.
35118 ~Type "show copying" to see the conditions.
35119 ~There is absolutely no warranty for GDB. Type "show warranty" for
35121 ~This GDB was configured as
35122 "--host=sparc-sun-solaris2.5.1 --target=ppc-eabi".
35127 @subheading The @code{-list-thread-groups} Command
35128 @findex -list-thread-groups
35130 @subheading Synopsis
35133 -list-thread-groups [ --available ] [ --recurse 1 ] [ @var{group} ... ]
35136 Lists thread groups (@pxref{Thread groups}). When a single thread
35137 group is passed as the argument, lists the children of that group.
35138 When several thread group are passed, lists information about those
35139 thread groups. Without any parameters, lists information about all
35140 top-level thread groups.
35142 Normally, thread groups that are being debugged are reported.
35143 With the @samp{--available} option, @value{GDBN} reports thread groups
35144 available on the target.
35146 The output of this command may have either a @samp{threads} result or
35147 a @samp{groups} result. The @samp{thread} result has a list of tuples
35148 as value, with each tuple describing a thread (@pxref{GDB/MI Thread
35149 Information}). The @samp{groups} result has a list of tuples as value,
35150 each tuple describing a thread group. If top-level groups are
35151 requested (that is, no parameter is passed), or when several groups
35152 are passed, the output always has a @samp{groups} result. The format
35153 of the @samp{group} result is described below.
35155 To reduce the number of roundtrips it's possible to list thread groups
35156 together with their children, by passing the @samp{--recurse} option
35157 and the recursion depth. Presently, only recursion depth of 1 is
35158 permitted. If this option is present, then every reported thread group
35159 will also include its children, either as @samp{group} or
35160 @samp{threads} field.
35162 In general, any combination of option and parameters is permitted, with
35163 the following caveats:
35167 When a single thread group is passed, the output will typically
35168 be the @samp{threads} result. Because threads may not contain
35169 anything, the @samp{recurse} option will be ignored.
35172 When the @samp{--available} option is passed, limited information may
35173 be available. In particular, the list of threads of a process might
35174 be inaccessible. Further, specifying specific thread groups might
35175 not give any performance advantage over listing all thread groups.
35176 The frontend should assume that @samp{-list-thread-groups --available}
35177 is always an expensive operation and cache the results.
35181 The @samp{groups} result is a list of tuples, where each tuple may
35182 have the following fields:
35186 Identifier of the thread group. This field is always present.
35187 The identifier is an opaque string; frontends should not try to
35188 convert it to an integer, even though it might look like one.
35191 The type of the thread group. At present, only @samp{process} is a
35195 The target-specific process identifier. This field is only present
35196 for thread groups of type @samp{process} and only if the process exists.
35199 The exit code of this group's last exited thread, formatted in octal.
35200 This field is only present for thread groups of type @samp{process} and
35201 only if the process is not running.
35204 The number of children this thread group has. This field may be
35205 absent for an available thread group.
35208 This field has a list of tuples as value, each tuple describing a
35209 thread. It may be present if the @samp{--recurse} option is
35210 specified, and it's actually possible to obtain the threads.
35213 This field is a list of integers, each identifying a core that one
35214 thread of the group is running on. This field may be absent if
35215 such information is not available.
35218 The name of the executable file that corresponds to this thread group.
35219 The field is only present for thread groups of type @samp{process},
35220 and only if there is a corresponding executable file.
35224 @subheading Example
35228 -list-thread-groups
35229 ^done,groups=[@{id="17",type="process",pid="yyy",num_children="2"@}]
35230 -list-thread-groups 17
35231 ^done,threads=[@{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
35232 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",args=[]@},state="running"@},
35233 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
35234 frame=@{level="0",addr="0x0804891f",func="foo",args=[@{name="i",value="10"@}],
35235 file="/tmp/a.c",fullname="/tmp/a.c",line="158",arch="i386:x86_64"@},state="running"@}]]
35236 -list-thread-groups --available
35237 ^done,groups=[@{id="17",type="process",pid="yyy",num_children="2",cores=[1,2]@}]
35238 -list-thread-groups --available --recurse 1
35239 ^done,groups=[@{id="17", types="process",pid="yyy",num_children="2",cores=[1,2],
35240 threads=[@{id="1",target-id="Thread 0xb7e14b90",cores=[1]@},
35241 @{id="2",target-id="Thread 0xb7e14b90",cores=[2]@}]@},..]
35242 -list-thread-groups --available --recurse 1 17 18
35243 ^done,groups=[@{id="17", types="process",pid="yyy",num_children="2",cores=[1,2],
35244 threads=[@{id="1",target-id="Thread 0xb7e14b90",cores=[1]@},
35245 @{id="2",target-id="Thread 0xb7e14b90",cores=[2]@}]@},...]
35248 @subheading The @code{-info-os} Command
35251 @subsubheading Synopsis
35254 -info-os [ @var{type} ]
35257 If no argument is supplied, the command returns a table of available
35258 operating-system-specific information types. If one of these types is
35259 supplied as an argument @var{type}, then the command returns a table
35260 of data of that type.
35262 The types of information available depend on the target operating
35265 @subsubheading @value{GDBN} Command
35267 The corresponding @value{GDBN} command is @samp{info os}.
35269 @subsubheading Example
35271 When run on a @sc{gnu}/Linux system, the output will look something
35277 ^done,OSDataTable=@{nr_rows="10",nr_cols="3",
35278 hdr=[@{width="10",alignment="-1",col_name="col0",colhdr="Type"@},
35279 @{width="10",alignment="-1",col_name="col1",colhdr="Description"@},
35280 @{width="10",alignment="-1",col_name="col2",colhdr="Title"@}],
35281 body=[item=@{col0="cpus",col1="Listing of all cpus/cores on the system",
35283 item=@{col0="files",col1="Listing of all file descriptors",
35284 col2="File descriptors"@},
35285 item=@{col0="modules",col1="Listing of all loaded kernel modules",
35286 col2="Kernel modules"@},
35287 item=@{col0="msg",col1="Listing of all message queues",
35288 col2="Message queues"@},
35289 item=@{col0="processes",col1="Listing of all processes",
35290 col2="Processes"@},
35291 item=@{col0="procgroups",col1="Listing of all process groups",
35292 col2="Process groups"@},
35293 item=@{col0="semaphores",col1="Listing of all semaphores",
35294 col2="Semaphores"@},
35295 item=@{col0="shm",col1="Listing of all shared-memory regions",
35296 col2="Shared-memory regions"@},
35297 item=@{col0="sockets",col1="Listing of all internet-domain sockets",
35299 item=@{col0="threads",col1="Listing of all threads",
35303 ^done,OSDataTable=@{nr_rows="190",nr_cols="4",
35304 hdr=[@{width="10",alignment="-1",col_name="col0",colhdr="pid"@},
35305 @{width="10",alignment="-1",col_name="col1",colhdr="user"@},
35306 @{width="10",alignment="-1",col_name="col2",colhdr="command"@},
35307 @{width="10",alignment="-1",col_name="col3",colhdr="cores"@}],
35308 body=[item=@{col0="1",col1="root",col2="/sbin/init",col3="0"@},
35309 item=@{col0="2",col1="root",col2="[kthreadd]",col3="1"@},
35310 item=@{col0="3",col1="root",col2="[ksoftirqd/0]",col3="0"@},
35312 item=@{col0="26446",col1="stan",col2="bash",col3="0"@},
35313 item=@{col0="28152",col1="stan",col2="bash",col3="1"@}]@}
35317 (Note that the MI output here includes a @code{"Title"} column that
35318 does not appear in command-line @code{info os}; this column is useful
35319 for MI clients that want to enumerate the types of data, such as in a
35320 popup menu, but is needless clutter on the command line, and
35321 @code{info os} omits it.)
35323 @subheading The @code{-add-inferior} Command
35324 @findex -add-inferior
35326 @subheading Synopsis
35332 Creates a new inferior (@pxref{Inferiors and Programs}). The created
35333 inferior is not associated with any executable. Such association may
35334 be established with the @samp{-file-exec-and-symbols} command
35335 (@pxref{GDB/MI File Commands}). The command response has a single
35336 field, @samp{inferior}, whose value is the identifier of the
35337 thread group corresponding to the new inferior.
35339 @subheading Example
35344 ^done,inferior="i3"
35347 @subheading The @code{-interpreter-exec} Command
35348 @findex -interpreter-exec
35350 @subheading Synopsis
35353 -interpreter-exec @var{interpreter} @var{command}
35355 @anchor{-interpreter-exec}
35357 Execute the specified @var{command} in the given @var{interpreter}.
35359 @subheading @value{GDBN} Command
35361 The corresponding @value{GDBN} command is @samp{interpreter-exec}.
35363 @subheading Example
35367 -interpreter-exec console "break main"
35368 &"During symbol reading, couldn't parse type; debugger out of date?.\n"
35369 &"During symbol reading, bad structure-type format.\n"
35370 ~"Breakpoint 1 at 0x8074fc6: file ../../src/gdb/main.c, line 743.\n"
35375 @subheading The @code{-inferior-tty-set} Command
35376 @findex -inferior-tty-set
35378 @subheading Synopsis
35381 -inferior-tty-set /dev/pts/1
35384 Set terminal for future runs of the program being debugged.
35386 @subheading @value{GDBN} Command
35388 The corresponding @value{GDBN} command is @samp{set inferior-tty} /dev/pts/1.
35390 @subheading Example
35394 -inferior-tty-set /dev/pts/1
35399 @subheading The @code{-inferior-tty-show} Command
35400 @findex -inferior-tty-show
35402 @subheading Synopsis
35408 Show terminal for future runs of program being debugged.
35410 @subheading @value{GDBN} Command
35412 The corresponding @value{GDBN} command is @samp{show inferior-tty}.
35414 @subheading Example
35418 -inferior-tty-set /dev/pts/1
35422 ^done,inferior_tty_terminal="/dev/pts/1"
35426 @subheading The @code{-enable-timings} Command
35427 @findex -enable-timings
35429 @subheading Synopsis
35432 -enable-timings [yes | no]
35435 Toggle the printing of the wallclock, user and system times for an MI
35436 command as a field in its output. This command is to help frontend
35437 developers optimize the performance of their code. No argument is
35438 equivalent to @samp{yes}.
35440 @subheading @value{GDBN} Command
35444 @subheading Example
35452 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
35453 addr="0x080484ed",func="main",file="myprog.c",
35454 fullname="/home/nickrob/myprog.c",line="73",thread-groups=["i1"],
35456 time=@{wallclock="0.05185",user="0.00800",system="0.00000"@}
35464 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
35465 frame=@{addr="0x080484ed",func="main",args=[@{name="argc",value="1"@},
35466 @{name="argv",value="0xbfb60364"@}],file="myprog.c",
35467 fullname="/home/nickrob/myprog.c",line="73",arch="i386:x86_64"@}
35471 @subheading The @code{-complete} Command
35474 @subheading Synopsis
35477 -complete @var{command}
35480 Show a list of completions for partially typed CLI @var{command}.
35482 This command is intended for @sc{gdb/mi} frontends that cannot use two separate
35483 CLI and MI channels --- for example: because of lack of PTYs like on Windows or
35484 because @value{GDBN} is used remotely via a SSH connection.
35488 The result consists of two or three fields:
35492 This field contains the completed @var{command}. If @var{command}
35493 has no known completions, this field is omitted.
35496 This field contains a (possibly empty) array of matches. It is always present.
35498 @item max_completions_reached
35499 This field contains @code{1} if number of known completions is above
35500 @code{max-completions} limit (@pxref{Completion}), otherwise it contains
35501 @code{0}. It is always present.
35505 @subheading @value{GDBN} Command
35507 The corresponding @value{GDBN} command is @samp{complete}.
35509 @subheading Example
35514 ^done,completion="break",
35515 matches=["break","break-range"],
35516 max_completions_reached="0"
35519 ^done,completion="b ma",
35520 matches=["b madvise","b main"],max_completions_reached="0"
35522 -complete "b push_b"
35523 ^done,completion="b push_back(",
35525 "b A::push_back(void*)",
35526 "b std::string::push_back(char)",
35527 "b std::vector<int, std::allocator<int> >::push_back(int&&)"],
35528 max_completions_reached="0"
35530 -complete "nonexist"
35531 ^done,matches=[],max_completions_reached="0"
35537 @chapter @value{GDBN} Annotations
35539 This chapter describes annotations in @value{GDBN}. Annotations were
35540 designed to interface @value{GDBN} to graphical user interfaces or other
35541 similar programs which want to interact with @value{GDBN} at a
35542 relatively high level.
35544 The annotation mechanism has largely been superseded by @sc{gdb/mi}
35548 This is Edition @value{EDITION}, @value{DATE}.
35552 * Annotations Overview:: What annotations are; the general syntax.
35553 * Server Prefix:: Issuing a command without affecting user state.
35554 * Prompting:: Annotations marking @value{GDBN}'s need for input.
35555 * Errors:: Annotations for error messages.
35556 * Invalidation:: Some annotations describe things now invalid.
35557 * Annotations for Running::
35558 Whether the program is running, how it stopped, etc.
35559 * Source Annotations:: Annotations describing source code.
35562 @node Annotations Overview
35563 @section What is an Annotation?
35564 @cindex annotations
35566 Annotations start with a newline character, two @samp{control-z}
35567 characters, and the name of the annotation. If there is no additional
35568 information associated with this annotation, the name of the annotation
35569 is followed immediately by a newline. If there is additional
35570 information, the name of the annotation is followed by a space, the
35571 additional information, and a newline. The additional information
35572 cannot contain newline characters.
35574 Any output not beginning with a newline and two @samp{control-z}
35575 characters denotes literal output from @value{GDBN}. Currently there is
35576 no need for @value{GDBN} to output a newline followed by two
35577 @samp{control-z} characters, but if there was such a need, the
35578 annotations could be extended with an @samp{escape} annotation which
35579 means those three characters as output.
35581 The annotation @var{level}, which is specified using the
35582 @option{--annotate} command line option (@pxref{Mode Options}), controls
35583 how much information @value{GDBN} prints together with its prompt,
35584 values of expressions, source lines, and other types of output. Level 0
35585 is for no annotations, level 1 is for use when @value{GDBN} is run as a
35586 subprocess of @sc{gnu} Emacs, level 3 is the maximum annotation suitable
35587 for programs that control @value{GDBN}, and level 2 annotations have
35588 been made obsolete (@pxref{Limitations, , Limitations of the Annotation
35589 Interface, annotate, GDB's Obsolete Annotations}).
35592 @kindex set annotate
35593 @item set annotate @var{level}
35594 The @value{GDBN} command @code{set annotate} sets the level of
35595 annotations to the specified @var{level}.
35597 @item show annotate
35598 @kindex show annotate
35599 Show the current annotation level.
35602 This chapter describes level 3 annotations.
35604 A simple example of starting up @value{GDBN} with annotations is:
35607 $ @kbd{gdb --annotate=3}
35609 Copyright 2003 Free Software Foundation, Inc.
35610 GDB is free software, covered by the GNU General Public License,
35611 and you are welcome to change it and/or distribute copies of it
35612 under certain conditions.
35613 Type "show copying" to see the conditions.
35614 There is absolutely no warranty for GDB. Type "show warranty"
35616 This GDB was configured as "i386-pc-linux-gnu"
35627 Here @samp{quit} is input to @value{GDBN}; the rest is output from
35628 @value{GDBN}. The three lines beginning @samp{^Z^Z} (where @samp{^Z}
35629 denotes a @samp{control-z} character) are annotations; the rest is
35630 output from @value{GDBN}.
35632 @node Server Prefix
35633 @section The Server Prefix
35634 @cindex server prefix
35636 If you prefix a command with @samp{server } then it will not affect
35637 the command history, nor will it affect @value{GDBN}'s notion of which
35638 command to repeat if @key{RET} is pressed on a line by itself. This
35639 means that commands can be run behind a user's back by a front-end in
35640 a transparent manner.
35642 The @code{server } prefix does not affect the recording of values into
35643 the value history; to print a value without recording it into the
35644 value history, use the @code{output} command instead of the
35645 @code{print} command.
35647 Using this prefix also disables confirmation requests
35648 (@pxref{confirmation requests}).
35651 @section Annotation for @value{GDBN} Input
35653 @cindex annotations for prompts
35654 When @value{GDBN} prompts for input, it annotates this fact so it is possible
35655 to know when to send output, when the output from a given command is
35658 Different kinds of input each have a different @dfn{input type}. Each
35659 input type has three annotations: a @code{pre-} annotation, which
35660 denotes the beginning of any prompt which is being output, a plain
35661 annotation, which denotes the end of the prompt, and then a @code{post-}
35662 annotation which denotes the end of any echo which may (or may not) be
35663 associated with the input. For example, the @code{prompt} input type
35664 features the following annotations:
35672 The input types are
35675 @findex pre-prompt annotation
35676 @findex prompt annotation
35677 @findex post-prompt annotation
35679 When @value{GDBN} is prompting for a command (the main @value{GDBN} prompt).
35681 @findex pre-commands annotation
35682 @findex commands annotation
35683 @findex post-commands annotation
35685 When @value{GDBN} prompts for a set of commands, like in the @code{commands}
35686 command. The annotations are repeated for each command which is input.
35688 @findex pre-overload-choice annotation
35689 @findex overload-choice annotation
35690 @findex post-overload-choice annotation
35691 @item overload-choice
35692 When @value{GDBN} wants the user to select between various overloaded functions.
35694 @findex pre-query annotation
35695 @findex query annotation
35696 @findex post-query annotation
35698 When @value{GDBN} wants the user to confirm a potentially dangerous operation.
35700 @findex pre-prompt-for-continue annotation
35701 @findex prompt-for-continue annotation
35702 @findex post-prompt-for-continue annotation
35703 @item prompt-for-continue
35704 When @value{GDBN} is asking the user to press return to continue. Note: Don't
35705 expect this to work well; instead use @code{set height 0} to disable
35706 prompting. This is because the counting of lines is buggy in the
35707 presence of annotations.
35712 @cindex annotations for errors, warnings and interrupts
35714 @findex quit annotation
35719 This annotation occurs right before @value{GDBN} responds to an interrupt.
35721 @findex error annotation
35726 This annotation occurs right before @value{GDBN} responds to an error.
35728 Quit and error annotations indicate that any annotations which @value{GDBN} was
35729 in the middle of may end abruptly. For example, if a
35730 @code{value-history-begin} annotation is followed by a @code{error}, one
35731 cannot expect to receive the matching @code{value-history-end}. One
35732 cannot expect not to receive it either, however; an error annotation
35733 does not necessarily mean that @value{GDBN} is immediately returning all the way
35736 @findex error-begin annotation
35737 A quit or error annotation may be preceded by
35743 Any output between that and the quit or error annotation is the error
35746 Warning messages are not yet annotated.
35747 @c If we want to change that, need to fix warning(), type_error(),
35748 @c range_error(), and possibly other places.
35751 @section Invalidation Notices
35753 @cindex annotations for invalidation messages
35754 The following annotations say that certain pieces of state may have
35758 @findex frames-invalid annotation
35759 @item ^Z^Zframes-invalid
35761 The frames (for example, output from the @code{backtrace} command) may
35764 @findex breakpoints-invalid annotation
35765 @item ^Z^Zbreakpoints-invalid
35767 The breakpoints may have changed. For example, the user just added or
35768 deleted a breakpoint.
35771 @node Annotations for Running
35772 @section Running the Program
35773 @cindex annotations for running programs
35775 @findex starting annotation
35776 @findex stopping annotation
35777 When the program starts executing due to a @value{GDBN} command such as
35778 @code{step} or @code{continue},
35784 is output. When the program stops,
35790 is output. Before the @code{stopped} annotation, a variety of
35791 annotations describe how the program stopped.
35794 @findex exited annotation
35795 @item ^Z^Zexited @var{exit-status}
35796 The program exited, and @var{exit-status} is the exit status (zero for
35797 successful exit, otherwise nonzero).
35799 @findex signalled annotation
35800 @findex signal-name annotation
35801 @findex signal-name-end annotation
35802 @findex signal-string annotation
35803 @findex signal-string-end annotation
35804 @item ^Z^Zsignalled
35805 The program exited with a signal. After the @code{^Z^Zsignalled}, the
35806 annotation continues:
35812 ^Z^Zsignal-name-end
35816 ^Z^Zsignal-string-end
35821 where @var{name} is the name of the signal, such as @code{SIGILL} or
35822 @code{SIGSEGV}, and @var{string} is the explanation of the signal, such
35823 as @code{Illegal Instruction} or @code{Segmentation fault}. The arguments
35824 @var{intro-text}, @var{middle-text}, and @var{end-text} are for the
35825 user's benefit and have no particular format.
35827 @findex signal annotation
35829 The syntax of this annotation is just like @code{signalled}, but @value{GDBN} is
35830 just saying that the program received the signal, not that it was
35831 terminated with it.
35833 @findex breakpoint annotation
35834 @item ^Z^Zbreakpoint @var{number}
35835 The program hit breakpoint number @var{number}.
35837 @findex watchpoint annotation
35838 @item ^Z^Zwatchpoint @var{number}
35839 The program hit watchpoint number @var{number}.
35842 @node Source Annotations
35843 @section Displaying Source
35844 @cindex annotations for source display
35846 @findex source annotation
35847 The following annotation is used instead of displaying source code:
35850 ^Z^Zsource @var{filename}:@var{line}:@var{character}:@var{middle}:@var{addr}
35853 where @var{filename} is an absolute file name indicating which source
35854 file, @var{line} is the line number within that file (where 1 is the
35855 first line in the file), @var{character} is the character position
35856 within the file (where 0 is the first character in the file) (for most
35857 debug formats this will necessarily point to the beginning of a line),
35858 @var{middle} is @samp{middle} if @var{addr} is in the middle of the
35859 line, or @samp{beg} if @var{addr} is at the beginning of the line, and
35860 @var{addr} is the address in the target program associated with the
35861 source which is being displayed. The @var{addr} is in the form @samp{0x}
35862 followed by one or more lowercase hex digits (note that this does not
35863 depend on the language).
35865 @node JIT Interface
35866 @chapter JIT Compilation Interface
35867 @cindex just-in-time compilation
35868 @cindex JIT compilation interface
35870 This chapter documents @value{GDBN}'s @dfn{just-in-time} (JIT) compilation
35871 interface. A JIT compiler is a program or library that generates native
35872 executable code at runtime and executes it, usually in order to achieve good
35873 performance while maintaining platform independence.
35875 Programs that use JIT compilation are normally difficult to debug because
35876 portions of their code are generated at runtime, instead of being loaded from
35877 object files, which is where @value{GDBN} normally finds the program's symbols
35878 and debug information. In order to debug programs that use JIT compilation,
35879 @value{GDBN} has an interface that allows the program to register in-memory
35880 symbol files with @value{GDBN} at runtime.
35882 If you are using @value{GDBN} to debug a program that uses this interface, then
35883 it should work transparently so long as you have not stripped the binary. If
35884 you are developing a JIT compiler, then the interface is documented in the rest
35885 of this chapter. At this time, the only known client of this interface is the
35888 Broadly speaking, the JIT interface mirrors the dynamic loader interface. The
35889 JIT compiler communicates with @value{GDBN} by writing data into a global
35890 variable and calling a function at a well-known symbol. When @value{GDBN}
35891 attaches, it reads a linked list of symbol files from the global variable to
35892 find existing code, and puts a breakpoint in the function so that it can find
35893 out about additional code.
35896 * Declarations:: Relevant C struct declarations
35897 * Registering Code:: Steps to register code
35898 * Unregistering Code:: Steps to unregister code
35899 * Custom Debug Info:: Emit debug information in a custom format
35903 @section JIT Declarations
35905 These are the relevant struct declarations that a C program should include to
35906 implement the interface:
35916 struct jit_code_entry
35918 struct jit_code_entry *next_entry;
35919 struct jit_code_entry *prev_entry;
35920 const char *symfile_addr;
35921 uint64_t symfile_size;
35924 struct jit_descriptor
35927 /* This type should be jit_actions_t, but we use uint32_t
35928 to be explicit about the bitwidth. */
35929 uint32_t action_flag;
35930 struct jit_code_entry *relevant_entry;
35931 struct jit_code_entry *first_entry;
35934 /* GDB puts a breakpoint in this function. */
35935 void __attribute__((noinline)) __jit_debug_register_code() @{ @};
35937 /* Make sure to specify the version statically, because the
35938 debugger may check the version before we can set it. */
35939 struct jit_descriptor __jit_debug_descriptor = @{ 1, 0, 0, 0 @};
35942 If the JIT is multi-threaded, then it is important that the JIT synchronize any
35943 modifications to this global data properly, which can easily be done by putting
35944 a global mutex around modifications to these structures.
35946 @node Registering Code
35947 @section Registering Code
35949 To register code with @value{GDBN}, the JIT should follow this protocol:
35953 Generate an object file in memory with symbols and other desired debug
35954 information. The file must include the virtual addresses of the sections.
35957 Create a code entry for the file, which gives the start and size of the symbol
35961 Add it to the linked list in the JIT descriptor.
35964 Point the relevant_entry field of the descriptor at the entry.
35967 Set @code{action_flag} to @code{JIT_REGISTER} and call
35968 @code{__jit_debug_register_code}.
35971 When @value{GDBN} is attached and the breakpoint fires, @value{GDBN} uses the
35972 @code{relevant_entry} pointer so it doesn't have to walk the list looking for
35973 new code. However, the linked list must still be maintained in order to allow
35974 @value{GDBN} to attach to a running process and still find the symbol files.
35976 @node Unregistering Code
35977 @section Unregistering Code
35979 If code is freed, then the JIT should use the following protocol:
35983 Remove the code entry corresponding to the code from the linked list.
35986 Point the @code{relevant_entry} field of the descriptor at the code entry.
35989 Set @code{action_flag} to @code{JIT_UNREGISTER} and call
35990 @code{__jit_debug_register_code}.
35993 If the JIT frees or recompiles code without unregistering it, then @value{GDBN}
35994 and the JIT will leak the memory used for the associated symbol files.
35996 @node Custom Debug Info
35997 @section Custom Debug Info
35998 @cindex custom JIT debug info
35999 @cindex JIT debug info reader
36001 Generating debug information in platform-native file formats (like ELF
36002 or COFF) may be an overkill for JIT compilers; especially if all the
36003 debug info is used for is displaying a meaningful backtrace. The
36004 issue can be resolved by having the JIT writers decide on a debug info
36005 format and also provide a reader that parses the debug info generated
36006 by the JIT compiler. This section gives a brief overview on writing
36007 such a parser. More specific details can be found in the source file
36008 @file{gdb/jit-reader.in}, which is also installed as a header at
36009 @file{@var{includedir}/gdb/jit-reader.h} for easy inclusion.
36011 The reader is implemented as a shared object (so this functionality is
36012 not available on platforms which don't allow loading shared objects at
36013 runtime). Two @value{GDBN} commands, @code{jit-reader-load} and
36014 @code{jit-reader-unload} are provided, to be used to load and unload
36015 the readers from a preconfigured directory. Once loaded, the shared
36016 object is used the parse the debug information emitted by the JIT
36020 * Using JIT Debug Info Readers:: How to use supplied readers correctly
36021 * Writing JIT Debug Info Readers:: Creating a debug-info reader
36024 @node Using JIT Debug Info Readers
36025 @subsection Using JIT Debug Info Readers
36026 @kindex jit-reader-load
36027 @kindex jit-reader-unload
36029 Readers can be loaded and unloaded using the @code{jit-reader-load}
36030 and @code{jit-reader-unload} commands.
36033 @item jit-reader-load @var{reader}
36034 Load the JIT reader named @var{reader}, which is a shared
36035 object specified as either an absolute or a relative file name. In
36036 the latter case, @value{GDBN} will try to load the reader from a
36037 pre-configured directory, usually @file{@var{libdir}/gdb/} on a UNIX
36038 system (here @var{libdir} is the system library directory, often
36039 @file{/usr/local/lib}).
36041 Only one reader can be active at a time; trying to load a second
36042 reader when one is already loaded will result in @value{GDBN}
36043 reporting an error. A new JIT reader can be loaded by first unloading
36044 the current one using @code{jit-reader-unload} and then invoking
36045 @code{jit-reader-load}.
36047 @item jit-reader-unload
36048 Unload the currently loaded JIT reader.
36052 @node Writing JIT Debug Info Readers
36053 @subsection Writing JIT Debug Info Readers
36054 @cindex writing JIT debug info readers
36056 As mentioned, a reader is essentially a shared object conforming to a
36057 certain ABI. This ABI is described in @file{jit-reader.h}.
36059 @file{jit-reader.h} defines the structures, macros and functions
36060 required to write a reader. It is installed (along with
36061 @value{GDBN}), in @file{@var{includedir}/gdb} where @var{includedir} is
36062 the system include directory.
36064 Readers need to be released under a GPL compatible license. A reader
36065 can be declared as released under such a license by placing the macro
36066 @code{GDB_DECLARE_GPL_COMPATIBLE_READER} in a source file.
36068 The entry point for readers is the symbol @code{gdb_init_reader},
36069 which is expected to be a function with the prototype
36071 @findex gdb_init_reader
36073 extern struct gdb_reader_funcs *gdb_init_reader (void);
36076 @cindex @code{struct gdb_reader_funcs}
36078 @code{struct gdb_reader_funcs} contains a set of pointers to callback
36079 functions. These functions are executed to read the debug info
36080 generated by the JIT compiler (@code{read}), to unwind stack frames
36081 (@code{unwind}) and to create canonical frame IDs
36082 (@code{get_Frame_id}). It also has a callback that is called when the
36083 reader is being unloaded (@code{destroy}). The struct looks like this
36086 struct gdb_reader_funcs
36088 /* Must be set to GDB_READER_INTERFACE_VERSION. */
36089 int reader_version;
36091 /* For use by the reader. */
36094 gdb_read_debug_info *read;
36095 gdb_unwind_frame *unwind;
36096 gdb_get_frame_id *get_frame_id;
36097 gdb_destroy_reader *destroy;
36101 @cindex @code{struct gdb_symbol_callbacks}
36102 @cindex @code{struct gdb_unwind_callbacks}
36104 The callbacks are provided with another set of callbacks by
36105 @value{GDBN} to do their job. For @code{read}, these callbacks are
36106 passed in a @code{struct gdb_symbol_callbacks} and for @code{unwind}
36107 and @code{get_frame_id}, in a @code{struct gdb_unwind_callbacks}.
36108 @code{struct gdb_symbol_callbacks} has callbacks to create new object
36109 files and new symbol tables inside those object files. @code{struct
36110 gdb_unwind_callbacks} has callbacks to read registers off the current
36111 frame and to write out the values of the registers in the previous
36112 frame. Both have a callback (@code{target_read}) to read bytes off the
36113 target's address space.
36115 @node In-Process Agent
36116 @chapter In-Process Agent
36117 @cindex debugging agent
36118 The traditional debugging model is conceptually low-speed, but works fine,
36119 because most bugs can be reproduced in debugging-mode execution. However,
36120 as multi-core or many-core processors are becoming mainstream, and
36121 multi-threaded programs become more and more popular, there should be more
36122 and more bugs that only manifest themselves at normal-mode execution, for
36123 example, thread races, because debugger's interference with the program's
36124 timing may conceal the bugs. On the other hand, in some applications,
36125 it is not feasible for the debugger to interrupt the program's execution
36126 long enough for the developer to learn anything helpful about its behavior.
36127 If the program's correctness depends on its real-time behavior, delays
36128 introduced by a debugger might cause the program to fail, even when the
36129 code itself is correct. It is useful to be able to observe the program's
36130 behavior without interrupting it.
36132 Therefore, traditional debugging model is too intrusive to reproduce
36133 some bugs. In order to reduce the interference with the program, we can
36134 reduce the number of operations performed by debugger. The
36135 @dfn{In-Process Agent}, a shared library, is running within the same
36136 process with inferior, and is able to perform some debugging operations
36137 itself. As a result, debugger is only involved when necessary, and
36138 performance of debugging can be improved accordingly. Note that
36139 interference with program can be reduced but can't be removed completely,
36140 because the in-process agent will still stop or slow down the program.
36142 The in-process agent can interpret and execute Agent Expressions
36143 (@pxref{Agent Expressions}) during performing debugging operations. The
36144 agent expressions can be used for different purposes, such as collecting
36145 data in tracepoints, and condition evaluation in breakpoints.
36147 @anchor{Control Agent}
36148 You can control whether the in-process agent is used as an aid for
36149 debugging with the following commands:
36152 @kindex set agent on
36154 Causes the in-process agent to perform some operations on behalf of the
36155 debugger. Just which operations requested by the user will be done
36156 by the in-process agent depends on the its capabilities. For example,
36157 if you request to evaluate breakpoint conditions in the in-process agent,
36158 and the in-process agent has such capability as well, then breakpoint
36159 conditions will be evaluated in the in-process agent.
36161 @kindex set agent off
36162 @item set agent off
36163 Disables execution of debugging operations by the in-process agent. All
36164 of the operations will be performed by @value{GDBN}.
36168 Display the current setting of execution of debugging operations by
36169 the in-process agent.
36173 * In-Process Agent Protocol::
36176 @node In-Process Agent Protocol
36177 @section In-Process Agent Protocol
36178 @cindex in-process agent protocol
36180 The in-process agent is able to communicate with both @value{GDBN} and
36181 GDBserver (@pxref{In-Process Agent}). This section documents the protocol
36182 used for communications between @value{GDBN} or GDBserver and the IPA.
36183 In general, @value{GDBN} or GDBserver sends commands
36184 (@pxref{IPA Protocol Commands}) and data to in-process agent, and then
36185 in-process agent replies back with the return result of the command, or
36186 some other information. The data sent to in-process agent is composed
36187 of primitive data types, such as 4-byte or 8-byte type, and composite
36188 types, which are called objects (@pxref{IPA Protocol Objects}).
36191 * IPA Protocol Objects::
36192 * IPA Protocol Commands::
36195 @node IPA Protocol Objects
36196 @subsection IPA Protocol Objects
36197 @cindex ipa protocol objects
36199 The commands sent to and results received from agent may contain some
36200 complex data types called @dfn{objects}.
36202 The in-process agent is running on the same machine with @value{GDBN}
36203 or GDBserver, so it doesn't have to handle as much differences between
36204 two ends as remote protocol (@pxref{Remote Protocol}) tries to handle.
36205 However, there are still some differences of two ends in two processes:
36209 word size. On some 64-bit machines, @value{GDBN} or GDBserver can be
36210 compiled as a 64-bit executable, while in-process agent is a 32-bit one.
36212 ABI. Some machines may have multiple types of ABI, @value{GDBN} or
36213 GDBserver is compiled with one, and in-process agent is compiled with
36217 Here are the IPA Protocol Objects:
36221 agent expression object. It represents an agent expression
36222 (@pxref{Agent Expressions}).
36223 @anchor{agent expression object}
36225 tracepoint action object. It represents a tracepoint action
36226 (@pxref{Tracepoint Actions,,Tracepoint Action Lists}) to collect registers,
36227 memory, static trace data and to evaluate expression.
36228 @anchor{tracepoint action object}
36230 tracepoint object. It represents a tracepoint (@pxref{Tracepoints}).
36231 @anchor{tracepoint object}
36235 The following table describes important attributes of each IPA protocol
36238 @multitable @columnfractions .30 .20 .50
36239 @headitem Name @tab Size @tab Description
36240 @item @emph{agent expression object} @tab @tab
36241 @item length @tab 4 @tab length of bytes code
36242 @item byte code @tab @var{length} @tab contents of byte code
36243 @item @emph{tracepoint action for collecting memory} @tab @tab
36244 @item 'M' @tab 1 @tab type of tracepoint action
36245 @item addr @tab 8 @tab if @var{basereg} is @samp{-1}, @var{addr} is the
36246 address of the lowest byte to collect, otherwise @var{addr} is the offset
36247 of @var{basereg} for memory collecting.
36248 @item len @tab 8 @tab length of memory for collecting
36249 @item basereg @tab 4 @tab the register number containing the starting
36250 memory address for collecting.
36251 @item @emph{tracepoint action for collecting registers} @tab @tab
36252 @item 'R' @tab 1 @tab type of tracepoint action
36253 @item @emph{tracepoint action for collecting static trace data} @tab @tab
36254 @item 'L' @tab 1 @tab type of tracepoint action
36255 @item @emph{tracepoint action for expression evaluation} @tab @tab
36256 @item 'X' @tab 1 @tab type of tracepoint action
36257 @item agent expression @tab length of @tab @ref{agent expression object}
36258 @item @emph{tracepoint object} @tab @tab
36259 @item number @tab 4 @tab number of tracepoint
36260 @item address @tab 8 @tab address of tracepoint inserted on
36261 @item type @tab 4 @tab type of tracepoint
36262 @item enabled @tab 1 @tab enable or disable of tracepoint
36263 @item step_count @tab 8 @tab step
36264 @item pass_count @tab 8 @tab pass
36265 @item numactions @tab 4 @tab number of tracepoint actions
36266 @item hit count @tab 8 @tab hit count
36267 @item trace frame usage @tab 8 @tab trace frame usage
36268 @item compiled_cond @tab 8 @tab compiled condition
36269 @item orig_size @tab 8 @tab orig size
36270 @item condition @tab 4 if condition is NULL otherwise length of
36271 @ref{agent expression object}
36272 @tab zero if condition is NULL, otherwise is
36273 @ref{agent expression object}
36274 @item actions @tab variable
36275 @tab numactions number of @ref{tracepoint action object}
36278 @node IPA Protocol Commands
36279 @subsection IPA Protocol Commands
36280 @cindex ipa protocol commands
36282 The spaces in each command are delimiters to ease reading this commands
36283 specification. They don't exist in real commands.
36287 @item FastTrace:@var{tracepoint_object} @var{gdb_jump_pad_head}
36288 Installs a new fast tracepoint described by @var{tracepoint_object}
36289 (@pxref{tracepoint object}). The @var{gdb_jump_pad_head}, 8-byte long, is the
36290 head of @dfn{jumppad}, which is used to jump to data collection routine
36295 @item OK @var{target_address} @var{gdb_jump_pad_head} @var{fjump_size} @var{fjump}
36296 @var{target_address} is address of tracepoint in the inferior.
36297 The @var{gdb_jump_pad_head} is updated head of jumppad. Both of
36298 @var{target_address} and @var{gdb_jump_pad_head} are 8-byte long.
36299 The @var{fjump} contains a sequence of instructions jump to jumppad entry.
36300 The @var{fjump_size}, 4-byte long, is the size of @var{fjump}.
36307 Closes the in-process agent. This command is sent when @value{GDBN} or GDBserver
36308 is about to kill inferiors.
36316 @item probe_marker_at:@var{address}
36317 Asks in-process agent to probe the marker at @var{address}.
36324 @item unprobe_marker_at:@var{address}
36325 Asks in-process agent to unprobe the marker at @var{address}.
36329 @chapter Reporting Bugs in @value{GDBN}
36330 @cindex bugs in @value{GDBN}
36331 @cindex reporting bugs in @value{GDBN}
36333 Your bug reports play an essential role in making @value{GDBN} reliable.
36335 Reporting a bug may help you by bringing a solution to your problem, or it
36336 may not. But in any case the principal function of a bug report is to help
36337 the entire community by making the next version of @value{GDBN} work better. Bug
36338 reports are your contribution to the maintenance of @value{GDBN}.
36340 In order for a bug report to serve its purpose, you must include the
36341 information that enables us to fix the bug.
36344 * Bug Criteria:: Have you found a bug?
36345 * Bug Reporting:: How to report bugs
36349 @section Have You Found a Bug?
36350 @cindex bug criteria
36352 If you are not sure whether you have found a bug, here are some guidelines:
36355 @cindex fatal signal
36356 @cindex debugger crash
36357 @cindex crash of debugger
36359 If the debugger gets a fatal signal, for any input whatever, that is a
36360 @value{GDBN} bug. Reliable debuggers never crash.
36362 @cindex error on valid input
36364 If @value{GDBN} produces an error message for valid input, that is a
36365 bug. (Note that if you're cross debugging, the problem may also be
36366 somewhere in the connection to the target.)
36368 @cindex invalid input
36370 If @value{GDBN} does not produce an error message for invalid input,
36371 that is a bug. However, you should note that your idea of
36372 ``invalid input'' might be our idea of ``an extension'' or ``support
36373 for traditional practice''.
36376 If you are an experienced user of debugging tools, your suggestions
36377 for improvement of @value{GDBN} are welcome in any case.
36380 @node Bug Reporting
36381 @section How to Report Bugs
36382 @cindex bug reports
36383 @cindex @value{GDBN} bugs, reporting
36385 A number of companies and individuals offer support for @sc{gnu} products.
36386 If you obtained @value{GDBN} from a support organization, we recommend you
36387 contact that organization first.
36389 You can find contact information for many support companies and
36390 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
36392 @c should add a web page ref...
36395 @ifset BUGURL_DEFAULT
36396 In any event, we also recommend that you submit bug reports for
36397 @value{GDBN}. The preferred method is to submit them directly using
36398 @uref{http://www.gnu.org/software/gdb/bugs/, @value{GDBN}'s Bugs web
36399 page}. Alternatively, the @email{bug-gdb@@gnu.org, e-mail gateway} can
36402 @strong{Do not send bug reports to @samp{info-gdb}, or to
36403 @samp{help-gdb}, or to any newsgroups.} Most users of @value{GDBN} do
36404 not want to receive bug reports. Those that do have arranged to receive
36407 The mailing list @samp{bug-gdb} has a newsgroup @samp{gnu.gdb.bug} which
36408 serves as a repeater. The mailing list and the newsgroup carry exactly
36409 the same messages. Often people think of posting bug reports to the
36410 newsgroup instead of mailing them. This appears to work, but it has one
36411 problem which can be crucial: a newsgroup posting often lacks a mail
36412 path back to the sender. Thus, if we need to ask for more information,
36413 we may be unable to reach you. For this reason, it is better to send
36414 bug reports to the mailing list.
36416 @ifclear BUGURL_DEFAULT
36417 In any event, we also recommend that you submit bug reports for
36418 @value{GDBN} to @value{BUGURL}.
36422 The fundamental principle of reporting bugs usefully is this:
36423 @strong{report all the facts}. If you are not sure whether to state a
36424 fact or leave it out, state it!
36426 Often people omit facts because they think they know what causes the
36427 problem and assume that some details do not matter. Thus, you might
36428 assume that the name of the variable you use in an example does not matter.
36429 Well, probably it does not, but one cannot be sure. Perhaps the bug is a
36430 stray memory reference which happens to fetch from the location where that
36431 name is stored in memory; perhaps, if the name were different, the contents
36432 of that location would fool the debugger into doing the right thing despite
36433 the bug. Play it safe and give a specific, complete example. That is the
36434 easiest thing for you to do, and the most helpful.
36436 Keep in mind that the purpose of a bug report is to enable us to fix the
36437 bug. It may be that the bug has been reported previously, but neither
36438 you nor we can know that unless your bug report is complete and
36441 Sometimes people give a few sketchy facts and ask, ``Does this ring a
36442 bell?'' Those bug reports are useless, and we urge everyone to
36443 @emph{refuse to respond to them} except to chide the sender to report
36446 To enable us to fix the bug, you should include all these things:
36450 The version of @value{GDBN}. @value{GDBN} announces it if you start
36451 with no arguments; you can also print it at any time using @code{show
36454 Without this, we will not know whether there is any point in looking for
36455 the bug in the current version of @value{GDBN}.
36458 The type of machine you are using, and the operating system name and
36462 The details of the @value{GDBN} build-time configuration.
36463 @value{GDBN} shows these details if you invoke it with the
36464 @option{--configuration} command-line option, or if you type
36465 @code{show configuration} at @value{GDBN}'s prompt.
36468 What compiler (and its version) was used to compile @value{GDBN}---e.g.@:
36469 ``@value{GCC}--2.8.1''.
36472 What compiler (and its version) was used to compile the program you are
36473 debugging---e.g.@: ``@value{GCC}--2.8.1'', or ``HP92453-01 A.10.32.03 HP
36474 C Compiler''. For @value{NGCC}, you can say @kbd{@value{GCC} --version}
36475 to get this information; for other compilers, see the documentation for
36479 The command arguments you gave the compiler to compile your example and
36480 observe the bug. For example, did you use @samp{-O}? To guarantee
36481 you will not omit something important, list them all. A copy of the
36482 Makefile (or the output from make) is sufficient.
36484 If we were to try to guess the arguments, we would probably guess wrong
36485 and then we might not encounter the bug.
36488 A complete input script, and all necessary source files, that will
36492 A description of what behavior you observe that you believe is
36493 incorrect. For example, ``It gets a fatal signal.''
36495 Of course, if the bug is that @value{GDBN} gets a fatal signal, then we
36496 will certainly notice it. But if the bug is incorrect output, we might
36497 not notice unless it is glaringly wrong. You might as well not give us
36498 a chance to make a mistake.
36500 Even if the problem you experience is a fatal signal, you should still
36501 say so explicitly. Suppose something strange is going on, such as, your
36502 copy of @value{GDBN} is out of synch, or you have encountered a bug in
36503 the C library on your system. (This has happened!) Your copy might
36504 crash and ours would not. If you told us to expect a crash, then when
36505 ours fails to crash, we would know that the bug was not happening for
36506 us. If you had not told us to expect a crash, then we would not be able
36507 to draw any conclusion from our observations.
36510 @cindex recording a session script
36511 To collect all this information, you can use a session recording program
36512 such as @command{script}, which is available on many Unix systems.
36513 Just run your @value{GDBN} session inside @command{script} and then
36514 include the @file{typescript} file with your bug report.
36516 Another way to record a @value{GDBN} session is to run @value{GDBN}
36517 inside Emacs and then save the entire buffer to a file.
36520 If you wish to suggest changes to the @value{GDBN} source, send us context
36521 diffs. If you even discuss something in the @value{GDBN} source, refer to
36522 it by context, not by line number.
36524 The line numbers in our development sources will not match those in your
36525 sources. Your line numbers would convey no useful information to us.
36529 Here are some things that are not necessary:
36533 A description of the envelope of the bug.
36535 Often people who encounter a bug spend a lot of time investigating
36536 which changes to the input file will make the bug go away and which
36537 changes will not affect it.
36539 This is often time consuming and not very useful, because the way we
36540 will find the bug is by running a single example under the debugger
36541 with breakpoints, not by pure deduction from a series of examples.
36542 We recommend that you save your time for something else.
36544 Of course, if you can find a simpler example to report @emph{instead}
36545 of the original one, that is a convenience for us. Errors in the
36546 output will be easier to spot, running under the debugger will take
36547 less time, and so on.
36549 However, simplification is not vital; if you do not want to do this,
36550 report the bug anyway and send us the entire test case you used.
36553 A patch for the bug.
36555 A patch for the bug does help us if it is a good one. But do not omit
36556 the necessary information, such as the test case, on the assumption that
36557 a patch is all we need. We might see problems with your patch and decide
36558 to fix the problem another way, or we might not understand it at all.
36560 Sometimes with a program as complicated as @value{GDBN} it is very hard to
36561 construct an example that will make the program follow a certain path
36562 through the code. If you do not send us the example, we will not be able
36563 to construct one, so we will not be able to verify that the bug is fixed.
36565 And if we cannot understand what bug you are trying to fix, or why your
36566 patch should be an improvement, we will not install it. A test case will
36567 help us to understand.
36570 A guess about what the bug is or what it depends on.
36572 Such guesses are usually wrong. Even we cannot guess right about such
36573 things without first using the debugger to find the facts.
36576 @c The readline documentation is distributed with the readline code
36577 @c and consists of the two following files:
36580 @c Use -I with makeinfo to point to the appropriate directory,
36581 @c environment var TEXINPUTS with TeX.
36582 @ifclear SYSTEM_READLINE
36583 @include rluser.texi
36584 @include hsuser.texi
36588 @appendix In Memoriam
36590 The @value{GDBN} project mourns the loss of the following long-time
36595 Fred was a long-standing contributor to @value{GDBN} (1991-2006), and
36596 to Free Software in general. Outside of @value{GDBN}, he was known in
36597 the Amiga world for his series of Fish Disks, and the GeekGadget project.
36599 @item Michael Snyder
36600 Michael was one of the Global Maintainers of the @value{GDBN} project,
36601 with contributions recorded as early as 1996, until 2011. In addition
36602 to his day to day participation, he was a large driving force behind
36603 adding Reverse Debugging to @value{GDBN}.
36606 Beyond their technical contributions to the project, they were also
36607 enjoyable members of the Free Software Community. We will miss them.
36609 @node Formatting Documentation
36610 @appendix Formatting Documentation
36612 @cindex @value{GDBN} reference card
36613 @cindex reference card
36614 The @value{GDBN} 4 release includes an already-formatted reference card, ready
36615 for printing with PostScript or Ghostscript, in the @file{gdb}
36616 subdirectory of the main source directory@footnote{In
36617 @file{gdb-@value{GDBVN}/gdb/refcard.ps} of the version @value{GDBVN}
36618 release.}. If you can use PostScript or Ghostscript with your printer,
36619 you can print the reference card immediately with @file{refcard.ps}.
36621 The release also includes the source for the reference card. You
36622 can format it, using @TeX{}, by typing:
36628 The @value{GDBN} reference card is designed to print in @dfn{landscape}
36629 mode on US ``letter'' size paper;
36630 that is, on a sheet 11 inches wide by 8.5 inches
36631 high. You will need to specify this form of printing as an option to
36632 your @sc{dvi} output program.
36634 @cindex documentation
36636 All the documentation for @value{GDBN} comes as part of the machine-readable
36637 distribution. The documentation is written in Texinfo format, which is
36638 a documentation system that uses a single source file to produce both
36639 on-line information and a printed manual. You can use one of the Info
36640 formatting commands to create the on-line version of the documentation
36641 and @TeX{} (or @code{texi2roff}) to typeset the printed version.
36643 @value{GDBN} includes an already formatted copy of the on-line Info
36644 version of this manual in the @file{gdb} subdirectory. The main Info
36645 file is @file{gdb-@value{GDBVN}/gdb/gdb.info}, and it refers to
36646 subordinate files matching @samp{gdb.info*} in the same directory. If
36647 necessary, you can print out these files, or read them with any editor;
36648 but they are easier to read using the @code{info} subsystem in @sc{gnu}
36649 Emacs or the standalone @code{info} program, available as part of the
36650 @sc{gnu} Texinfo distribution.
36652 If you want to format these Info files yourself, you need one of the
36653 Info formatting programs, such as @code{texinfo-format-buffer} or
36656 If you have @code{makeinfo} installed, and are in the top level
36657 @value{GDBN} source directory (@file{gdb-@value{GDBVN}}, in the case of
36658 version @value{GDBVN}), you can make the Info file by typing:
36665 If you want to typeset and print copies of this manual, you need @TeX{},
36666 a program to print its @sc{dvi} output files, and @file{texinfo.tex}, the
36667 Texinfo definitions file.
36669 @TeX{} is a typesetting program; it does not print files directly, but
36670 produces output files called @sc{dvi} files. To print a typeset
36671 document, you need a program to print @sc{dvi} files. If your system
36672 has @TeX{} installed, chances are it has such a program. The precise
36673 command to use depends on your system; @kbd{lpr -d} is common; another
36674 (for PostScript devices) is @kbd{dvips}. The @sc{dvi} print command may
36675 require a file name without any extension or a @samp{.dvi} extension.
36677 @TeX{} also requires a macro definitions file called
36678 @file{texinfo.tex}. This file tells @TeX{} how to typeset a document
36679 written in Texinfo format. On its own, @TeX{} cannot either read or
36680 typeset a Texinfo file. @file{texinfo.tex} is distributed with GDB
36681 and is located in the @file{gdb-@var{version-number}/texinfo}
36684 If you have @TeX{} and a @sc{dvi} printer program installed, you can
36685 typeset and print this manual. First switch to the @file{gdb}
36686 subdirectory of the main source directory (for example, to
36687 @file{gdb-@value{GDBVN}/gdb}) and type:
36693 Then give @file{gdb.dvi} to your @sc{dvi} printing program.
36695 @node Installing GDB
36696 @appendix Installing @value{GDBN}
36697 @cindex installation
36700 * Requirements:: Requirements for building @value{GDBN}
36701 * Running Configure:: Invoking the @value{GDBN} @file{configure} script
36702 * Separate Objdir:: Compiling @value{GDBN} in another directory
36703 * Config Names:: Specifying names for hosts and targets
36704 * Configure Options:: Summary of options for configure
36705 * System-wide configuration:: Having a system-wide init file
36709 @section Requirements for Building @value{GDBN}
36710 @cindex building @value{GDBN}, requirements for
36712 Building @value{GDBN} requires various tools and packages to be available.
36713 Other packages will be used only if they are found.
36715 @heading Tools/Packages Necessary for Building @value{GDBN}
36717 @item C@t{++}11 compiler
36718 @value{GDBN} is written in C@t{++}11. It should be buildable with any
36719 recent C@t{++}11 compiler, e.g.@: GCC.
36722 @value{GDBN}'s build system relies on features only found in the GNU
36723 make program. Other variants of @code{make} will not work.
36726 @heading Tools/Packages Optional for Building @value{GDBN}
36730 @value{GDBN} can use the Expat XML parsing library. This library may be
36731 included with your operating system distribution; if it is not, you
36732 can get the latest version from @url{http://expat.sourceforge.net}.
36733 The @file{configure} script will search for this library in several
36734 standard locations; if it is installed in an unusual path, you can
36735 use the @option{--with-libexpat-prefix} option to specify its location.
36741 Remote protocol memory maps (@pxref{Memory Map Format})
36743 Target descriptions (@pxref{Target Descriptions})
36745 Remote shared library lists (@xref{Library List Format},
36746 or alternatively @pxref{Library List Format for SVR4 Targets})
36748 MS-Windows shared libraries (@pxref{Shared Libraries})
36750 Traceframe info (@pxref{Traceframe Info Format})
36752 Branch trace (@pxref{Branch Trace Format},
36753 @pxref{Branch Trace Configuration Format})
36757 @value{GDBN} can be scripted using GNU Guile. @xref{Guile}. By
36758 default, @value{GDBN} will be compiled if the Guile libraries are
36759 installed and are found by @file{configure}. You can use the
36760 @code{--with-guile} option to request Guile, and pass either the Guile
36761 version number or the file name of the relevant @code{pkg-config}
36762 program to choose a particular version of Guile.
36765 @value{GDBN}'s features related to character sets (@pxref{Character
36766 Sets}) require a functioning @code{iconv} implementation. If you are
36767 on a GNU system, then this is provided by the GNU C Library. Some
36768 other systems also provide a working @code{iconv}.
36770 If @value{GDBN} is using the @code{iconv} program which is installed
36771 in a non-standard place, you will need to tell @value{GDBN} where to
36772 find it. This is done with @option{--with-iconv-bin} which specifies
36773 the directory that contains the @code{iconv} program. This program is
36774 run in order to make a list of the available character sets.
36776 On systems without @code{iconv}, you can install GNU Libiconv. If
36777 Libiconv is installed in a standard place, @value{GDBN} will
36778 automatically use it if it is needed. If you have previously
36779 installed Libiconv in a non-standard place, you can use the
36780 @option{--with-libiconv-prefix} option to @file{configure}.
36782 @value{GDBN}'s top-level @file{configure} and @file{Makefile} will
36783 arrange to build Libiconv if a directory named @file{libiconv} appears
36784 in the top-most source directory. If Libiconv is built this way, and
36785 if the operating system does not provide a suitable @code{iconv}
36786 implementation, then the just-built library will automatically be used
36787 by @value{GDBN}. One easy way to set this up is to download GNU
36788 Libiconv, unpack it inside the top-level directory of the @value{GDBN}
36789 source tree, and then rename the directory holding the Libiconv source
36790 code to @samp{libiconv}.
36793 @value{GDBN} can support debugging sections that are compressed with
36794 the LZMA library. @xref{MiniDebugInfo}. If this library is not
36795 included with your operating system, you can find it in the xz package
36796 at @url{http://tukaani.org/xz/}. If the LZMA library is available in
36797 the usual place, then the @file{configure} script will use it
36798 automatically. If it is installed in an unusual path, you can use the
36799 @option{--with-lzma-prefix} option to specify its location.
36803 @value{GDBN} can use the GNU MPFR multiple-precision floating-point
36804 library. This library may be included with your operating system
36805 distribution; if it is not, you can get the latest version from
36806 @url{http://www.mpfr.org}. The @file{configure} script will search
36807 for this library in several standard locations; if it is installed
36808 in an unusual path, you can use the @option{--with-libmpfr-prefix}
36809 option to specify its location.
36811 GNU MPFR is used to emulate target floating-point arithmetic during
36812 expression evaluation when the target uses different floating-point
36813 formats than the host. If GNU MPFR it is not available, @value{GDBN}
36814 will fall back to using host floating-point arithmetic.
36817 @value{GDBN} can be scripted using Python language. @xref{Python}.
36818 By default, @value{GDBN} will be compiled if the Python libraries are
36819 installed and are found by @file{configure}. You can use the
36820 @code{--with-python} option to request Python, and pass either the
36821 file name of the relevant @code{python} executable, or the name of the
36822 directory in which Python is installed, to choose a particular
36823 installation of Python.
36826 @cindex compressed debug sections
36827 @value{GDBN} will use the @samp{zlib} library, if available, to read
36828 compressed debug sections. Some linkers, such as GNU gold, are capable
36829 of producing binaries with compressed debug sections. If @value{GDBN}
36830 is compiled with @samp{zlib}, it will be able to read the debug
36831 information in such binaries.
36833 The @samp{zlib} library is likely included with your operating system
36834 distribution; if it is not, you can get the latest version from
36835 @url{http://zlib.net}.
36838 @node Running Configure
36839 @section Invoking the @value{GDBN} @file{configure} Script
36840 @cindex configuring @value{GDBN}
36841 @value{GDBN} comes with a @file{configure} script that automates the process
36842 of preparing @value{GDBN} for installation; you can then use @code{make} to
36843 build the @code{gdb} program.
36845 @c irrelevant in info file; it's as current as the code it lives with.
36846 @footnote{If you have a more recent version of @value{GDBN} than @value{GDBVN},
36847 look at the @file{README} file in the sources; we may have improved the
36848 installation procedures since publishing this manual.}
36851 The @value{GDBN} distribution includes all the source code you need for
36852 @value{GDBN} in a single directory, whose name is usually composed by
36853 appending the version number to @samp{gdb}.
36855 For example, the @value{GDBN} version @value{GDBVN} distribution is in the
36856 @file{gdb-@value{GDBVN}} directory. That directory contains:
36859 @item gdb-@value{GDBVN}/configure @r{(and supporting files)}
36860 script for configuring @value{GDBN} and all its supporting libraries
36862 @item gdb-@value{GDBVN}/gdb
36863 the source specific to @value{GDBN} itself
36865 @item gdb-@value{GDBVN}/bfd
36866 source for the Binary File Descriptor library
36868 @item gdb-@value{GDBVN}/include
36869 @sc{gnu} include files
36871 @item gdb-@value{GDBVN}/libiberty
36872 source for the @samp{-liberty} free software library
36874 @item gdb-@value{GDBVN}/opcodes
36875 source for the library of opcode tables and disassemblers
36877 @item gdb-@value{GDBVN}/readline
36878 source for the @sc{gnu} command-line interface
36881 There may be other subdirectories as well.
36883 The simplest way to configure and build @value{GDBN} is to run @file{configure}
36884 from the @file{gdb-@var{version-number}} source directory, which in
36885 this example is the @file{gdb-@value{GDBVN}} directory.
36887 First switch to the @file{gdb-@var{version-number}} source directory
36888 if you are not already in it; then run @file{configure}. Pass the
36889 identifier for the platform on which @value{GDBN} will run as an
36895 cd gdb-@value{GDBVN}
36900 Running @samp{configure} and then running @code{make} builds the
36901 included supporting libraries, then @code{gdb} itself. The configured
36902 source files, and the binaries, are left in the corresponding source
36906 @file{configure} is a Bourne-shell (@code{/bin/sh}) script; if your
36907 system does not recognize this automatically when you run a different
36908 shell, you may need to run @code{sh} on it explicitly:
36914 You should run the @file{configure} script from the top directory in the
36915 source tree, the @file{gdb-@var{version-number}} directory. If you run
36916 @file{configure} from one of the subdirectories, you will configure only
36917 that subdirectory. That is usually not what you want. In particular,
36918 if you run the first @file{configure} from the @file{gdb} subdirectory
36919 of the @file{gdb-@var{version-number}} directory, you will omit the
36920 configuration of @file{bfd}, @file{readline}, and other sibling
36921 directories of the @file{gdb} subdirectory. This leads to build errors
36922 about missing include files such as @file{bfd/bfd.h}.
36924 You can install @code{@value{GDBN}} anywhere. The best way to do this
36925 is to pass the @code{--prefix} option to @code{configure}, and then
36926 install it with @code{make install}.
36928 @node Separate Objdir
36929 @section Compiling @value{GDBN} in Another Directory
36931 If you want to run @value{GDBN} versions for several host or target machines,
36932 you need a different @code{gdb} compiled for each combination of
36933 host and target. @file{configure} is designed to make this easy by
36934 allowing you to generate each configuration in a separate subdirectory,
36935 rather than in the source directory. If your @code{make} program
36936 handles the @samp{VPATH} feature (@sc{gnu} @code{make} does), running
36937 @code{make} in each of these directories builds the @code{gdb}
36938 program specified there.
36940 To build @code{gdb} in a separate directory, run @file{configure}
36941 with the @samp{--srcdir} option to specify where to find the source.
36942 (You also need to specify a path to find @file{configure}
36943 itself from your working directory. If the path to @file{configure}
36944 would be the same as the argument to @samp{--srcdir}, you can leave out
36945 the @samp{--srcdir} option; it is assumed.)
36947 For example, with version @value{GDBVN}, you can build @value{GDBN} in a
36948 separate directory for a Sun 4 like this:
36952 cd gdb-@value{GDBVN}
36955 ../gdb-@value{GDBVN}/configure
36960 When @file{configure} builds a configuration using a remote source
36961 directory, it creates a tree for the binaries with the same structure
36962 (and using the same names) as the tree under the source directory. In
36963 the example, you'd find the Sun 4 library @file{libiberty.a} in the
36964 directory @file{gdb-sun4/libiberty}, and @value{GDBN} itself in
36965 @file{gdb-sun4/gdb}.
36967 Make sure that your path to the @file{configure} script has just one
36968 instance of @file{gdb} in it. If your path to @file{configure} looks
36969 like @file{../gdb-@value{GDBVN}/gdb/configure}, you are configuring only
36970 one subdirectory of @value{GDBN}, not the whole package. This leads to
36971 build errors about missing include files such as @file{bfd/bfd.h}.
36973 One popular reason to build several @value{GDBN} configurations in separate
36974 directories is to configure @value{GDBN} for cross-compiling (where
36975 @value{GDBN} runs on one machine---the @dfn{host}---while debugging
36976 programs that run on another machine---the @dfn{target}).
36977 You specify a cross-debugging target by
36978 giving the @samp{--target=@var{target}} option to @file{configure}.
36980 When you run @code{make} to build a program or library, you must run
36981 it in a configured directory---whatever directory you were in when you
36982 called @file{configure} (or one of its subdirectories).
36984 The @code{Makefile} that @file{configure} generates in each source
36985 directory also runs recursively. If you type @code{make} in a source
36986 directory such as @file{gdb-@value{GDBVN}} (or in a separate configured
36987 directory configured with @samp{--srcdir=@var{dirname}/gdb-@value{GDBVN}}), you
36988 will build all the required libraries, and then build GDB.
36990 When you have multiple hosts or targets configured in separate
36991 directories, you can run @code{make} on them in parallel (for example,
36992 if they are NFS-mounted on each of the hosts); they will not interfere
36996 @section Specifying Names for Hosts and Targets
36998 The specifications used for hosts and targets in the @file{configure}
36999 script are based on a three-part naming scheme, but some short predefined
37000 aliases are also supported. The full naming scheme encodes three pieces
37001 of information in the following pattern:
37004 @var{architecture}-@var{vendor}-@var{os}
37007 For example, you can use the alias @code{sun4} as a @var{host} argument,
37008 or as the value for @var{target} in a @code{--target=@var{target}}
37009 option. The equivalent full name is @samp{sparc-sun-sunos4}.
37011 The @file{configure} script accompanying @value{GDBN} does not provide
37012 any query facility to list all supported host and target names or
37013 aliases. @file{configure} calls the Bourne shell script
37014 @code{config.sub} to map abbreviations to full names; you can read the
37015 script, if you wish, or you can use it to test your guesses on
37016 abbreviations---for example:
37019 % sh config.sub i386-linux
37021 % sh config.sub alpha-linux
37022 alpha-unknown-linux-gnu
37023 % sh config.sub hp9k700
37025 % sh config.sub sun4
37026 sparc-sun-sunos4.1.1
37027 % sh config.sub sun3
37028 m68k-sun-sunos4.1.1
37029 % sh config.sub i986v
37030 Invalid configuration `i986v': machine `i986v' not recognized
37034 @code{config.sub} is also distributed in the @value{GDBN} source
37035 directory (@file{gdb-@value{GDBVN}}, for version @value{GDBVN}).
37037 @node Configure Options
37038 @section @file{configure} Options
37040 Here is a summary of the @file{configure} options and arguments that
37041 are most often useful for building @value{GDBN}. @file{configure}
37042 also has several other options not listed here. @inforef{Running
37043 configure scripts,,autoconf.info}, for a full
37044 explanation of @file{configure}.
37047 configure @r{[}--help@r{]}
37048 @r{[}--prefix=@var{dir}@r{]}
37049 @r{[}--exec-prefix=@var{dir}@r{]}
37050 @r{[}--srcdir=@var{dirname}@r{]}
37051 @r{[}--target=@var{target}@r{]}
37055 You may introduce options with a single @samp{-} rather than
37056 @samp{--} if you prefer; but you may abbreviate option names if you use
37061 Display a quick summary of how to invoke @file{configure}.
37063 @item --prefix=@var{dir}
37064 Configure the source to install programs and files under directory
37067 @item --exec-prefix=@var{dir}
37068 Configure the source to install programs under directory
37071 @c avoid splitting the warning from the explanation:
37073 @item --srcdir=@var{dirname}
37074 Use this option to make configurations in directories separate from the
37075 @value{GDBN} source directories. Among other things, you can use this to
37076 build (or maintain) several configurations simultaneously, in separate
37077 directories. @file{configure} writes configuration-specific files in
37078 the current directory, but arranges for them to use the source in the
37079 directory @var{dirname}. @file{configure} creates directories under
37080 the working directory in parallel to the source directories below
37083 @item --target=@var{target}
37084 Configure @value{GDBN} for cross-debugging programs running on the specified
37085 @var{target}. Without this option, @value{GDBN} is configured to debug
37086 programs that run on the same machine (@var{host}) as @value{GDBN} itself.
37088 There is no convenient way to generate a list of all available
37089 targets. Also see the @code{--enable-targets} option, below.
37092 There are many other options that are specific to @value{GDBN}. This
37093 lists just the most common ones; there are some very specialized
37094 options not described here.
37097 @item --enable-targets=@r{[}@var{target}@r{]}@dots{}
37098 @itemx --enable-targets=all
37099 Configure @value{GDBN} for cross-debugging programs running on the
37100 specified list of targets. The special value @samp{all} configures
37101 @value{GDBN} for debugging programs running on any target it supports.
37103 @item --with-gdb-datadir=@var{path}
37104 Set the @value{GDBN}-specific data directory. @value{GDBN} will look
37105 here for certain supporting files or scripts. This defaults to the
37106 @file{gdb} subdirectory of @samp{datadir} (which can be set using
37109 @item --with-relocated-sources=@var{dir}
37110 Sets up the default source path substitution rule so that directory
37111 names recorded in debug information will be automatically adjusted for
37112 any directory under @var{dir}. @var{dir} should be a subdirectory of
37113 @value{GDBN}'s configured prefix, the one mentioned in the
37114 @code{--prefix} or @code{--exec-prefix} options to configure. This
37115 option is useful if GDB is supposed to be moved to a different place
37118 @item --enable-64-bit-bfd
37119 Enable 64-bit support in BFD on 32-bit hosts.
37121 @item --disable-gdbmi
37122 Build @value{GDBN} without the GDB/MI machine interface
37126 Build @value{GDBN} with the text-mode full-screen user interface
37127 (TUI). Requires a curses library (ncurses and cursesX are also
37130 @item --with-curses
37131 Use the curses library instead of the termcap library, for text-mode
37132 terminal operations.
37134 @item --with-libunwind-ia64
37135 Use the libunwind library for unwinding function call stack on ia64
37136 target platforms. See http://www.nongnu.org/libunwind/index.html for
37139 @item --with-system-readline
37140 Use the readline library installed on the host, rather than the
37141 library supplied as part of @value{GDBN}. Readline 7 or newer is
37142 required; this is enforced by the build system.
37144 @item --with-system-zlib
37145 Use the zlib library installed on the host, rather than the library
37146 supplied as part of @value{GDBN}.
37149 Build @value{GDBN} with Expat, a library for XML parsing. (Done by
37150 default if libexpat is installed and found at configure time.) This
37151 library is used to read XML files supplied with @value{GDBN}. If it
37152 is unavailable, some features, such as remote protocol memory maps,
37153 target descriptions, and shared library lists, that are based on XML
37154 files, will not be available in @value{GDBN}. If your host does not
37155 have libexpat installed, you can get the latest version from
37156 `http://expat.sourceforge.net'.
37158 @item --with-libiconv-prefix@r{[}=@var{dir}@r{]}
37160 Build @value{GDBN} with GNU libiconv, a character set encoding
37161 conversion library. This is not done by default, as on GNU systems
37162 the @code{iconv} that is built in to the C library is sufficient. If
37163 your host does not have a working @code{iconv}, you can get the latest
37164 version of GNU iconv from `https://www.gnu.org/software/libiconv/'.
37166 @value{GDBN}'s build system also supports building GNU libiconv as
37167 part of the overall build. @xref{Requirements}.
37170 Build @value{GDBN} with LZMA, a compression library. (Done by default
37171 if liblzma is installed and found at configure time.) LZMA is used by
37172 @value{GDBN}'s "mini debuginfo" feature, which is only useful on
37173 platforms using the ELF object file format. If your host does not
37174 have liblzma installed, you can get the latest version from
37175 `https://tukaani.org/xz/'.
37178 Build @value{GDBN} with GNU MPFR, a library for multiple-precision
37179 floating-point computation with correct rounding. (Done by default if
37180 GNU MPFR is installed and found at configure time.) This library is
37181 used to emulate target floating-point arithmetic during expression
37182 evaluation when the target uses different floating-point formats than
37183 the host. If GNU MPFR is not available, @value{GDBN} will fall back
37184 to using host floating-point arithmetic. If your host does not have
37185 GNU MPFR installed, you can get the latest version from
37186 `http://www.mpfr.org'.
37188 @item --with-python@r{[}=@var{python}@r{]}
37189 Build @value{GDBN} with Python scripting support. (Done by default if
37190 libpython is present and found at configure time.) Python makes
37191 @value{GDBN} scripting much more powerful than the restricted CLI
37192 scripting language. If your host does not have Python installed, you
37193 can find it on `http://www.python.org/download/'. The oldest version
37194 of Python supported by GDB is 2.6. The optional argument @var{python}
37195 is used to find the Python headers and libraries. It can be either
37196 the name of a Python executable, or the name of the directory in which
37197 Python is installed.
37199 @item --with-guile[=GUILE]'
37200 Build @value{GDBN} with GNU Guile scripting support. (Done by default
37201 if libguile is present and found at configure time.) If your host
37202 does not have Guile installed, you can find it at
37203 `https://www.gnu.org/software/guile/'. The optional argument GUILE
37204 can be a version number, which will cause @code{configure} to try to
37205 use that version of Guile; or the file name of a @code{pkg-config}
37206 executable, which will be queried to find the information needed to
37207 compile and link against Guile.
37209 @item --without-included-regex
37210 Don't use the regex library included with @value{GDBN} (as part of the
37211 libiberty library). This is the default on hosts with version 2 of
37214 @item --with-sysroot=@var{dir}
37215 Use @var{dir} as the default system root directory for libraries whose
37216 file names begin with @file{/lib}' or @file{/usr/lib'}. (The value of
37217 @var{dir} can be modified at run time by using the @command{set
37218 sysroot} command.) If @var{dir} is under the @value{GDBN} configured
37219 prefix (set with @code{--prefix} or @code{--exec-prefix options}, the
37220 default system root will be automatically adjusted if and when
37221 @value{GDBN} is moved to a different location.
37223 @item --with-system-gdbinit=@var{file}
37224 Configure @value{GDBN} to automatically load a system-wide init file.
37225 @var{file} should be an absolute file name. If @var{file} is in a
37226 directory under the configured prefix, and @value{GDBN} is moved to
37227 another location after being built, the location of the system-wide
37228 init file will be adjusted accordingly.
37230 @item --with-system-gdbinit-dir=@var{directory}
37231 Configure @value{GDBN} to automatically load init files from a
37232 system-wide directory. @var{directory} should be an absolute directory
37233 name. If @var{directory} is in a directory under the configured
37234 prefix, and @value{GDBN} is moved to another location after being
37235 built, the location of the system-wide init directory will be
37236 adjusted accordingly.
37238 @item --enable-build-warnings
37239 When building the @value{GDBN} sources, ask the compiler to warn about
37240 any code which looks even vaguely suspicious. It passes many
37241 different warning flags, depending on the exact version of the
37242 compiler you are using.
37244 @item --enable-werror
37245 Treat compiler warnings as werrors. It adds the @code{-Werror} flag
37246 to the compiler, which will fail the compilation if the compiler
37247 outputs any warning messages.
37249 @item --enable-ubsan
37250 Enable the GCC undefined behavior sanitizer. This is disabled by
37251 default, but passing @code{--enable-ubsan=yes} or
37252 @code{--enable-ubsan=auto} to @code{configure} will enable it. The
37253 undefined behavior sanitizer checks for C@t{++} undefined behavior.
37254 It has a performance cost, so if you are looking at @value{GDBN}'s
37255 performance, you should disable it. The undefined behavior sanitizer
37256 was first introduced in GCC 4.9.
37259 @node System-wide configuration
37260 @section System-wide configuration and settings
37261 @cindex system-wide init file
37263 @value{GDBN} can be configured to have a system-wide init file and a
37264 system-wide init file directory; this file and files in that directory
37265 (if they have a recognized file extension) will be read and executed at
37266 startup (@pxref{Startup, , What @value{GDBN} does during startup}).
37268 Here are the corresponding configure options:
37271 @item --with-system-gdbinit=@var{file}
37272 Specify that the default location of the system-wide init file is
37274 @item --with-system-gdbinit-dir=@var{directory}
37275 Specify that the default location of the system-wide init file directory
37276 is @var{directory}.
37279 If @value{GDBN} has been configured with the option @option{--prefix=$prefix},
37280 they may be subject to relocation. Two possible cases:
37284 If the default location of this init file/directory contains @file{$prefix},
37285 it will be subject to relocation. Suppose that the configure options
37286 are @option{--prefix=$prefix --with-system-gdbinit=$prefix/etc/gdbinit};
37287 if @value{GDBN} is moved from @file{$prefix} to @file{$install}, the system
37288 init file is looked for as @file{$install/etc/gdbinit} instead of
37289 @file{$prefix/etc/gdbinit}.
37292 By contrast, if the default location does not contain the prefix,
37293 it will not be relocated. E.g.@: if @value{GDBN} has been configured with
37294 @option{--prefix=/usr/local --with-system-gdbinit=/usr/share/gdb/gdbinit},
37295 then @value{GDBN} will always look for @file{/usr/share/gdb/gdbinit},
37296 wherever @value{GDBN} is installed.
37299 If the configured location of the system-wide init file (as given by the
37300 @option{--with-system-gdbinit} option at configure time) is in the
37301 data-directory (as specified by @option{--with-gdb-datadir} at configure
37302 time) or in one of its subdirectories, then @value{GDBN} will look for the
37303 system-wide init file in the directory specified by the
37304 @option{--data-directory} command-line option.
37305 Note that the system-wide init file is only read once, during @value{GDBN}
37306 initialization. If the data-directory is changed after @value{GDBN} has
37307 started with the @code{set data-directory} command, the file will not be
37310 This applies similarly to the system-wide directory specified in
37311 @option{--with-system-gdbinit-dir}.
37313 Any supported scripting language can be used for these init files, as long
37314 as the file extension matches the scripting language. To be interpreted
37315 as regular @value{GDBN} commands, the files needs to have a @file{.gdb}
37319 * System-wide Configuration Scripts:: Installed System-wide Configuration Scripts
37322 @node System-wide Configuration Scripts
37323 @subsection Installed System-wide Configuration Scripts
37324 @cindex system-wide configuration scripts
37326 The @file{system-gdbinit} directory, located inside the data-directory
37327 (as specified by @option{--with-gdb-datadir} at configure time) contains
37328 a number of scripts which can be used as system-wide init files. To
37329 automatically source those scripts at startup, @value{GDBN} should be
37330 configured with @option{--with-system-gdbinit}. Otherwise, any user
37331 should be able to source them by hand as needed.
37333 The following scripts are currently available:
37336 @item @file{elinos.py}
37338 @cindex ELinOS system-wide configuration script
37339 This script is useful when debugging a program on an ELinOS target.
37340 It takes advantage of the environment variables defined in a standard
37341 ELinOS environment in order to determine the location of the system
37342 shared libraries, and then sets the @samp{solib-absolute-prefix}
37343 and @samp{solib-search-path} variables appropriately.
37345 @item @file{wrs-linux.py}
37346 @pindex wrs-linux.py
37347 @cindex Wind River Linux system-wide configuration script
37348 This script is useful when debugging a program on a target running
37349 Wind River Linux. It expects the @env{ENV_PREFIX} to be set to
37350 the host-side sysroot used by the target system.
37354 @node Maintenance Commands
37355 @appendix Maintenance Commands
37356 @cindex maintenance commands
37357 @cindex internal commands
37359 In addition to commands intended for @value{GDBN} users, @value{GDBN}
37360 includes a number of commands intended for @value{GDBN} developers,
37361 that are not documented elsewhere in this manual. These commands are
37362 provided here for reference. (For commands that turn on debugging
37363 messages, see @ref{Debugging Output}.)
37366 @kindex maint agent
37367 @kindex maint agent-eval
37368 @item maint agent @r{[}-at @var{location}@r{,}@r{]} @var{expression}
37369 @itemx maint agent-eval @r{[}-at @var{location}@r{,}@r{]} @var{expression}
37370 Translate the given @var{expression} into remote agent bytecodes.
37371 This command is useful for debugging the Agent Expression mechanism
37372 (@pxref{Agent Expressions}). The @samp{agent} version produces an
37373 expression useful for data collection, such as by tracepoints, while
37374 @samp{maint agent-eval} produces an expression that evaluates directly
37375 to a result. For instance, a collection expression for @code{globa +
37376 globb} will include bytecodes to record four bytes of memory at each
37377 of the addresses of @code{globa} and @code{globb}, while discarding
37378 the result of the addition, while an evaluation expression will do the
37379 addition and return the sum.
37380 If @code{-at} is given, generate remote agent bytecode for @var{location}.
37381 If not, generate remote agent bytecode for current frame PC address.
37383 @kindex maint agent-printf
37384 @item maint agent-printf @var{format},@var{expr},...
37385 Translate the given format string and list of argument expressions
37386 into remote agent bytecodes and display them as a disassembled list.
37387 This command is useful for debugging the agent version of dynamic
37388 printf (@pxref{Dynamic Printf}).
37390 @kindex maint info breakpoints
37391 @item @anchor{maint info breakpoints}maint info breakpoints
37392 Using the same format as @samp{info breakpoints}, display both the
37393 breakpoints you've set explicitly, and those @value{GDBN} is using for
37394 internal purposes. Internal breakpoints are shown with negative
37395 breakpoint numbers. The type column identifies what kind of breakpoint
37400 Normal, explicitly set breakpoint.
37403 Normal, explicitly set watchpoint.
37406 Internal breakpoint, used to handle correctly stepping through
37407 @code{longjmp} calls.
37409 @item longjmp resume
37410 Internal breakpoint at the target of a @code{longjmp}.
37413 Temporary internal breakpoint used by the @value{GDBN} @code{until} command.
37416 Temporary internal breakpoint used by the @value{GDBN} @code{finish} command.
37419 Shared library events.
37423 @kindex maint info btrace
37424 @item maint info btrace
37425 Pint information about raw branch tracing data.
37427 @kindex maint btrace packet-history
37428 @item maint btrace packet-history
37429 Print the raw branch trace packets that are used to compute the
37430 execution history for the @samp{record btrace} command. Both the
37431 information and the format in which it is printed depend on the btrace
37436 For the BTS recording format, print a list of blocks of sequential
37437 code. For each block, the following information is printed:
37441 Newer blocks have higher numbers. The oldest block has number zero.
37442 @item Lowest @samp{PC}
37443 @item Highest @samp{PC}
37447 For the Intel Processor Trace recording format, print a list of
37448 Intel Processor Trace packets. For each packet, the following
37449 information is printed:
37452 @item Packet number
37453 Newer packets have higher numbers. The oldest packet has number zero.
37455 The packet's offset in the trace stream.
37456 @item Packet opcode and payload
37460 @kindex maint btrace clear-packet-history
37461 @item maint btrace clear-packet-history
37462 Discards the cached packet history printed by the @samp{maint btrace
37463 packet-history} command. The history will be computed again when
37466 @kindex maint btrace clear
37467 @item maint btrace clear
37468 Discard the branch trace data. The data will be fetched anew and the
37469 branch trace will be recomputed when needed.
37471 This implicitly truncates the branch trace to a single branch trace
37472 buffer. When updating branch trace incrementally, the branch trace
37473 available to @value{GDBN} may be bigger than a single branch trace
37476 @kindex maint set btrace pt skip-pad
37477 @item maint set btrace pt skip-pad
37478 @kindex maint show btrace pt skip-pad
37479 @item maint show btrace pt skip-pad
37480 Control whether @value{GDBN} will skip PAD packets when computing the
37483 @kindex set displaced-stepping
37484 @kindex show displaced-stepping
37485 @cindex displaced stepping support
37486 @cindex out-of-line single-stepping
37487 @item set displaced-stepping
37488 @itemx show displaced-stepping
37489 Control whether or not @value{GDBN} will do @dfn{displaced stepping}
37490 if the target supports it. Displaced stepping is a way to single-step
37491 over breakpoints without removing them from the inferior, by executing
37492 an out-of-line copy of the instruction that was originally at the
37493 breakpoint location. It is also known as out-of-line single-stepping.
37496 @item set displaced-stepping on
37497 If the target architecture supports it, @value{GDBN} will use
37498 displaced stepping to step over breakpoints.
37500 @item set displaced-stepping off
37501 @value{GDBN} will not use displaced stepping to step over breakpoints,
37502 even if such is supported by the target architecture.
37504 @cindex non-stop mode, and @samp{set displaced-stepping}
37505 @item set displaced-stepping auto
37506 This is the default mode. @value{GDBN} will use displaced stepping
37507 only if non-stop mode is active (@pxref{Non-Stop Mode}) and the target
37508 architecture supports displaced stepping.
37511 @kindex maint check-psymtabs
37512 @item maint check-psymtabs
37513 Check the consistency of currently expanded psymtabs versus symtabs.
37514 Use this to check, for example, whether a symbol is in one but not the other.
37516 @kindex maint check-symtabs
37517 @item maint check-symtabs
37518 Check the consistency of currently expanded symtabs.
37520 @kindex maint expand-symtabs
37521 @item maint expand-symtabs [@var{regexp}]
37522 Expand symbol tables.
37523 If @var{regexp} is specified, only expand symbol tables for file
37524 names matching @var{regexp}.
37526 @kindex maint set catch-demangler-crashes
37527 @kindex maint show catch-demangler-crashes
37528 @cindex demangler crashes
37529 @item maint set catch-demangler-crashes [on|off]
37530 @itemx maint show catch-demangler-crashes
37531 Control whether @value{GDBN} should attempt to catch crashes in the
37532 symbol name demangler. The default is to attempt to catch crashes.
37533 If enabled, the first time a crash is caught, a core file is created,
37534 the offending symbol is displayed and the user is presented with the
37535 option to terminate the current session.
37537 @kindex maint cplus first_component
37538 @item maint cplus first_component @var{name}
37539 Print the first C@t{++} class/namespace component of @var{name}.
37541 @kindex maint cplus namespace
37542 @item maint cplus namespace
37543 Print the list of possible C@t{++} namespaces.
37545 @kindex maint deprecate
37546 @kindex maint undeprecate
37547 @cindex deprecated commands
37548 @item maint deprecate @var{command} @r{[}@var{replacement}@r{]}
37549 @itemx maint undeprecate @var{command}
37550 Deprecate or undeprecate the named @var{command}. Deprecated commands
37551 cause @value{GDBN} to issue a warning when you use them. The optional
37552 argument @var{replacement} says which newer command should be used in
37553 favor of the deprecated one; if it is given, @value{GDBN} will mention
37554 the replacement as part of the warning.
37556 @kindex maint dump-me
37557 @item maint dump-me
37558 @cindex @code{SIGQUIT} signal, dump core of @value{GDBN}
37559 Cause a fatal signal in the debugger and force it to dump its core.
37560 This is supported only on systems which support aborting a program
37561 with the @code{SIGQUIT} signal.
37563 @kindex maint internal-error
37564 @kindex maint internal-warning
37565 @kindex maint demangler-warning
37566 @cindex demangler crashes
37567 @item maint internal-error @r{[}@var{message-text}@r{]}
37568 @itemx maint internal-warning @r{[}@var{message-text}@r{]}
37569 @itemx maint demangler-warning @r{[}@var{message-text}@r{]}
37571 Cause @value{GDBN} to call the internal function @code{internal_error},
37572 @code{internal_warning} or @code{demangler_warning} and hence behave
37573 as though an internal problem has been detected. In addition to
37574 reporting the internal problem, these functions give the user the
37575 opportunity to either quit @value{GDBN} or (for @code{internal_error}
37576 and @code{internal_warning}) create a core file of the current
37577 @value{GDBN} session.
37579 These commands take an optional parameter @var{message-text} that is
37580 used as the text of the error or warning message.
37582 Here's an example of using @code{internal-error}:
37585 (@value{GDBP}) @kbd{maint internal-error testing, 1, 2}
37586 @dots{}/maint.c:121: internal-error: testing, 1, 2
37587 A problem internal to GDB has been detected. Further
37588 debugging may prove unreliable.
37589 Quit this debugging session? (y or n) @kbd{n}
37590 Create a core file? (y or n) @kbd{n}
37594 @cindex @value{GDBN} internal error
37595 @cindex internal errors, control of @value{GDBN} behavior
37596 @cindex demangler crashes
37598 @kindex maint set internal-error
37599 @kindex maint show internal-error
37600 @kindex maint set internal-warning
37601 @kindex maint show internal-warning
37602 @kindex maint set demangler-warning
37603 @kindex maint show demangler-warning
37604 @item maint set internal-error @var{action} [ask|yes|no]
37605 @itemx maint show internal-error @var{action}
37606 @itemx maint set internal-warning @var{action} [ask|yes|no]
37607 @itemx maint show internal-warning @var{action}
37608 @itemx maint set demangler-warning @var{action} [ask|yes|no]
37609 @itemx maint show demangler-warning @var{action}
37610 When @value{GDBN} reports an internal problem (error or warning) it
37611 gives the user the opportunity to both quit @value{GDBN} and create a
37612 core file of the current @value{GDBN} session. These commands let you
37613 override the default behaviour for each particular @var{action},
37614 described in the table below.
37618 You can specify that @value{GDBN} should always (yes) or never (no)
37619 quit. The default is to ask the user what to do.
37622 You can specify that @value{GDBN} should always (yes) or never (no)
37623 create a core file. The default is to ask the user what to do. Note
37624 that there is no @code{corefile} option for @code{demangler-warning}:
37625 demangler warnings always create a core file and this cannot be
37629 @kindex maint packet
37630 @item maint packet @var{text}
37631 If @value{GDBN} is talking to an inferior via the serial protocol,
37632 then this command sends the string @var{text} to the inferior, and
37633 displays the response packet. @value{GDBN} supplies the initial
37634 @samp{$} character, the terminating @samp{#} character, and the
37637 @kindex maint print architecture
37638 @item maint print architecture @r{[}@var{file}@r{]}
37639 Print the entire architecture configuration. The optional argument
37640 @var{file} names the file where the output goes.
37642 @kindex maint print c-tdesc @r{[}@var{file}@r{]}
37643 @item maint print c-tdesc
37644 Print the target description (@pxref{Target Descriptions}) as
37645 a C source file. By default, the target description is for the current
37646 target, but if the optional argument @var{file} is provided, that file
37647 is used to produce the description. The @var{file} should be an XML
37648 document, of the form described in @ref{Target Description Format}.
37649 The created source file is built into @value{GDBN} when @value{GDBN} is
37650 built again. This command is used by developers after they add or
37651 modify XML target descriptions.
37653 @kindex maint check xml-descriptions
37654 @item maint check xml-descriptions @var{dir}
37655 Check that the target descriptions dynamically created by @value{GDBN}
37656 equal the descriptions created from XML files found in @var{dir}.
37658 @anchor{maint check libthread-db}
37659 @kindex maint check libthread-db
37660 @item maint check libthread-db
37661 Run integrity checks on the current inferior's thread debugging
37662 library. This exercises all @code{libthread_db} functionality used by
37663 @value{GDBN} on GNU/Linux systems, and by extension also exercises the
37664 @code{proc_service} functions provided by @value{GDBN} that
37665 @code{libthread_db} uses. Note that parts of the test may be skipped
37666 on some platforms when debugging core files.
37668 @kindex maint print dummy-frames
37669 @item maint print dummy-frames
37670 Prints the contents of @value{GDBN}'s internal dummy-frame stack.
37673 (@value{GDBP}) @kbd{b add}
37675 (@value{GDBP}) @kbd{print add(2,3)}
37676 Breakpoint 2, add (a=2, b=3) at @dots{}
37678 The program being debugged stopped while in a function called from GDB.
37680 (@value{GDBP}) @kbd{maint print dummy-frames}
37681 0xa8206d8: id=@{stack=0xbfffe734,code=0xbfffe73f,!special@}, ptid=process 9353
37685 Takes an optional file parameter.
37687 @kindex maint print registers
37688 @kindex maint print raw-registers
37689 @kindex maint print cooked-registers
37690 @kindex maint print register-groups
37691 @kindex maint print remote-registers
37692 @item maint print registers @r{[}@var{file}@r{]}
37693 @itemx maint print raw-registers @r{[}@var{file}@r{]}
37694 @itemx maint print cooked-registers @r{[}@var{file}@r{]}
37695 @itemx maint print register-groups @r{[}@var{file}@r{]}
37696 @itemx maint print remote-registers @r{[}@var{file}@r{]}
37697 Print @value{GDBN}'s internal register data structures.
37699 The command @code{maint print raw-registers} includes the contents of
37700 the raw register cache; the command @code{maint print
37701 cooked-registers} includes the (cooked) value of all registers,
37702 including registers which aren't available on the target nor visible
37703 to user; the command @code{maint print register-groups} includes the
37704 groups that each register is a member of; and the command @code{maint
37705 print remote-registers} includes the remote target's register numbers
37706 and offsets in the `G' packets.
37708 These commands take an optional parameter, a file name to which to
37709 write the information.
37711 @kindex maint print reggroups
37712 @item maint print reggroups @r{[}@var{file}@r{]}
37713 Print @value{GDBN}'s internal register group data structures. The
37714 optional argument @var{file} tells to what file to write the
37717 The register groups info looks like this:
37720 (@value{GDBP}) @kbd{maint print reggroups}
37733 This command forces @value{GDBN} to flush its internal register cache.
37735 @kindex maint print objfiles
37736 @cindex info for known object files
37737 @item maint print objfiles @r{[}@var{regexp}@r{]}
37738 Print a dump of all known object files.
37739 If @var{regexp} is specified, only print object files whose names
37740 match @var{regexp}. For each object file, this command prints its name,
37741 address in memory, and all of its psymtabs and symtabs.
37743 @kindex maint print user-registers
37744 @cindex user registers
37745 @item maint print user-registers
37746 List all currently available @dfn{user registers}. User registers
37747 typically provide alternate names for actual hardware registers. They
37748 include the four ``standard'' registers @code{$fp}, @code{$pc},
37749 @code{$sp}, and @code{$ps}. @xref{standard registers}. User
37750 registers can be used in expressions in the same way as the canonical
37751 register names, but only the latter are listed by the @code{info
37752 registers} and @code{maint print registers} commands.
37754 @kindex maint print section-scripts
37755 @cindex info for known .debug_gdb_scripts-loaded scripts
37756 @item maint print section-scripts [@var{regexp}]
37757 Print a dump of scripts specified in the @code{.debug_gdb_section} section.
37758 If @var{regexp} is specified, only print scripts loaded by object files
37759 matching @var{regexp}.
37760 For each script, this command prints its name as specified in the objfile,
37761 and the full path if known.
37762 @xref{dotdebug_gdb_scripts section}.
37764 @kindex maint print statistics
37765 @cindex bcache statistics
37766 @item maint print statistics
37767 This command prints, for each object file in the program, various data
37768 about that object file followed by the byte cache (@dfn{bcache})
37769 statistics for the object file. The objfile data includes the number
37770 of minimal, partial, full, and stabs symbols, the number of types
37771 defined by the objfile, the number of as yet unexpanded psym tables,
37772 the number of line tables and string tables, and the amount of memory
37773 used by the various tables. The bcache statistics include the counts,
37774 sizes, and counts of duplicates of all and unique objects, max,
37775 average, and median entry size, total memory used and its overhead and
37776 savings, and various measures of the hash table size and chain
37779 @kindex maint print target-stack
37780 @cindex target stack description
37781 @item maint print target-stack
37782 A @dfn{target} is an interface between the debugger and a particular
37783 kind of file or process. Targets can be stacked in @dfn{strata},
37784 so that more than one target can potentially respond to a request.
37785 In particular, memory accesses will walk down the stack of targets
37786 until they find a target that is interested in handling that particular
37789 This command prints a short description of each layer that was pushed on
37790 the @dfn{target stack}, starting from the top layer down to the bottom one.
37792 @kindex maint print type
37793 @cindex type chain of a data type
37794 @item maint print type @var{expr}
37795 Print the type chain for a type specified by @var{expr}. The argument
37796 can be either a type name or a symbol. If it is a symbol, the type of
37797 that symbol is described. The type chain produced by this command is
37798 a recursive definition of the data type as stored in @value{GDBN}'s
37799 data structures, including its flags and contained types.
37801 @kindex maint selftest
37803 @item maint selftest @r{[}@var{filter}@r{]}
37804 Run any self tests that were compiled in to @value{GDBN}. This will
37805 print a message showing how many tests were run, and how many failed.
37806 If a @var{filter} is passed, only the tests with @var{filter} in their
37809 @kindex maint info selftests
37811 @item maint info selftests
37812 List the selftests compiled in to @value{GDBN}.
37814 @kindex maint set dwarf always-disassemble
37815 @kindex maint show dwarf always-disassemble
37816 @item maint set dwarf always-disassemble
37817 @item maint show dwarf always-disassemble
37818 Control the behavior of @code{info address} when using DWARF debugging
37821 The default is @code{off}, which means that @value{GDBN} should try to
37822 describe a variable's location in an easily readable format. When
37823 @code{on}, @value{GDBN} will instead display the DWARF location
37824 expression in an assembly-like format. Note that some locations are
37825 too complex for @value{GDBN} to describe simply; in this case you will
37826 always see the disassembly form.
37828 Here is an example of the resulting disassembly:
37831 (gdb) info addr argc
37832 Symbol "argc" is a complex DWARF expression:
37836 For more information on these expressions, see
37837 @uref{http://www.dwarfstd.org/, the DWARF standard}.
37839 @kindex maint set dwarf max-cache-age
37840 @kindex maint show dwarf max-cache-age
37841 @item maint set dwarf max-cache-age
37842 @itemx maint show dwarf max-cache-age
37843 Control the DWARF compilation unit cache.
37845 @cindex DWARF compilation units cache
37846 In object files with inter-compilation-unit references, such as those
37847 produced by the GCC option @samp{-feliminate-dwarf2-dups}, the DWARF
37848 reader needs to frequently refer to previously read compilation units.
37849 This setting controls how long a compilation unit will remain in the
37850 cache if it is not referenced. A higher limit means that cached
37851 compilation units will be stored in memory longer, and more total
37852 memory will be used. Setting it to zero disables caching, which will
37853 slow down @value{GDBN} startup, but reduce memory consumption.
37855 @kindex maint set dwarf unwinders
37856 @kindex maint show dwarf unwinders
37857 @item maint set dwarf unwinders
37858 @itemx maint show dwarf unwinders
37859 Control use of the DWARF frame unwinders.
37861 @cindex DWARF frame unwinders
37862 Many targets that support DWARF debugging use @value{GDBN}'s DWARF
37863 frame unwinders to build the backtrace. Many of these targets will
37864 also have a second mechanism for building the backtrace for use in
37865 cases where DWARF information is not available, this second mechanism
37866 is often an analysis of a function's prologue.
37868 In order to extend testing coverage of the second level stack
37869 unwinding mechanisms it is helpful to be able to disable the DWARF
37870 stack unwinders, this can be done with this switch.
37872 In normal use of @value{GDBN} disabling the DWARF unwinders is not
37873 advisable, there are cases that are better handled through DWARF than
37874 prologue analysis, and the debug experience is likely to be better
37875 with the DWARF frame unwinders enabled.
37877 If DWARF frame unwinders are not supported for a particular target
37878 architecture, then enabling this flag does not cause them to be used.
37880 @kindex maint set worker-threads
37881 @kindex maint show worker-threads
37882 @item maint set worker-threads
37883 @item maint show worker-threads
37884 Control the number of worker threads that may be used by @value{GDBN}.
37885 On capable hosts, @value{GDBN} may use multiple threads to speed up
37886 certain CPU-intensive operations, such as demangling symbol names.
37887 While the number of threads used by @value{GDBN} may vary, this
37888 command can be used to set an upper bound on this number. The default
37889 is @code{unlimited}, which lets @value{GDBN} choose a reasonable
37890 number. Note that this only controls worker threads started by
37891 @value{GDBN} itself; libraries used by @value{GDBN} may start threads
37894 @kindex maint set profile
37895 @kindex maint show profile
37896 @cindex profiling GDB
37897 @item maint set profile
37898 @itemx maint show profile
37899 Control profiling of @value{GDBN}.
37901 Profiling will be disabled until you use the @samp{maint set profile}
37902 command to enable it. When you enable profiling, the system will begin
37903 collecting timing and execution count data; when you disable profiling or
37904 exit @value{GDBN}, the results will be written to a log file. Remember that
37905 if you use profiling, @value{GDBN} will overwrite the profiling log file
37906 (often called @file{gmon.out}). If you have a record of important profiling
37907 data in a @file{gmon.out} file, be sure to move it to a safe location.
37909 Configuring with @samp{--enable-profiling} arranges for @value{GDBN} to be
37910 compiled with the @samp{-pg} compiler option.
37912 @kindex maint set show-debug-regs
37913 @kindex maint show show-debug-regs
37914 @cindex hardware debug registers
37915 @item maint set show-debug-regs
37916 @itemx maint show show-debug-regs
37917 Control whether to show variables that mirror the hardware debug
37918 registers. Use @code{on} to enable, @code{off} to disable. If
37919 enabled, the debug registers values are shown when @value{GDBN} inserts or
37920 removes a hardware breakpoint or watchpoint, and when the inferior
37921 triggers a hardware-assisted breakpoint or watchpoint.
37923 @kindex maint set show-all-tib
37924 @kindex maint show show-all-tib
37925 @item maint set show-all-tib
37926 @itemx maint show show-all-tib
37927 Control whether to show all non zero areas within a 1k block starting
37928 at thread local base, when using the @samp{info w32 thread-information-block}
37931 @kindex maint set target-async
37932 @kindex maint show target-async
37933 @item maint set target-async
37934 @itemx maint show target-async
37935 This controls whether @value{GDBN} targets operate in synchronous or
37936 asynchronous mode (@pxref{Background Execution}). Normally the
37937 default is asynchronous, if it is available; but this can be changed
37938 to more easily debug problems occurring only in synchronous mode.
37940 @kindex maint set target-non-stop @var{mode} [on|off|auto]
37941 @kindex maint show target-non-stop
37942 @item maint set target-non-stop
37943 @itemx maint show target-non-stop
37945 This controls whether @value{GDBN} targets always operate in non-stop
37946 mode even if @code{set non-stop} is @code{off} (@pxref{Non-Stop
37947 Mode}). The default is @code{auto}, meaning non-stop mode is enabled
37948 if supported by the target.
37951 @item maint set target-non-stop auto
37952 This is the default mode. @value{GDBN} controls the target in
37953 non-stop mode if the target supports it.
37955 @item maint set target-non-stop on
37956 @value{GDBN} controls the target in non-stop mode even if the target
37957 does not indicate support.
37959 @item maint set target-non-stop off
37960 @value{GDBN} does not control the target in non-stop mode even if the
37961 target supports it.
37964 @kindex maint set tui-resize-message
37965 @kindex maint show tui-resize-message
37966 @item maint set tui-resize-message
37967 @item maint show tui-resize-message
37968 Control whether @value{GDBN} displays a message each time the terminal
37969 is resized when in TUI mode. The default is @code{off}, which means
37970 that @value{GDBN} is silent during resizes. When @code{on},
37971 @value{GDBN} will display a message after a resize is completed; the
37972 message will include a number indicating how many times the terminal
37973 has been resized. This setting is intended for use by the test suite,
37974 where it would otherwise be difficult to determine when a resize and
37975 refresh has been completed.
37977 @kindex maint set per-command
37978 @kindex maint show per-command
37979 @item maint set per-command
37980 @itemx maint show per-command
37981 @cindex resources used by commands
37983 @value{GDBN} can display the resources used by each command.
37984 This is useful in debugging performance problems.
37987 @item maint set per-command space [on|off]
37988 @itemx maint show per-command space
37989 Enable or disable the printing of the memory used by GDB for each command.
37990 If enabled, @value{GDBN} will display how much memory each command
37991 took, following the command's own output.
37992 This can also be requested by invoking @value{GDBN} with the
37993 @option{--statistics} command-line switch (@pxref{Mode Options}).
37995 @item maint set per-command time [on|off]
37996 @itemx maint show per-command time
37997 Enable or disable the printing of the execution time of @value{GDBN}
37999 If enabled, @value{GDBN} will display how much time it
38000 took to execute each command, following the command's own output.
38001 Both CPU time and wallclock time are printed.
38002 Printing both is useful when trying to determine whether the cost is
38003 CPU or, e.g., disk/network latency.
38004 Note that the CPU time printed is for @value{GDBN} only, it does not include
38005 the execution time of the inferior because there's no mechanism currently
38006 to compute how much time was spent by @value{GDBN} and how much time was
38007 spent by the program been debugged.
38008 This can also be requested by invoking @value{GDBN} with the
38009 @option{--statistics} command-line switch (@pxref{Mode Options}).
38011 @item maint set per-command symtab [on|off]
38012 @itemx maint show per-command symtab
38013 Enable or disable the printing of basic symbol table statistics
38015 If enabled, @value{GDBN} will display the following information:
38019 number of symbol tables
38021 number of primary symbol tables
38023 number of blocks in the blockvector
38027 @kindex maint set check-libthread-db
38028 @kindex maint show check-libthread-db
38029 @item maint set check-libthread-db [on|off]
38030 @itemx maint show check-libthread-db
38031 Control whether @value{GDBN} should run integrity checks on inferior
38032 specific thread debugging libraries as they are loaded. The default
38033 is not to perform such checks. If any check fails @value{GDBN} will
38034 unload the library and continue searching for a suitable candidate as
38035 described in @ref{set libthread-db-search-path}. For more information
38036 about the tests, see @ref{maint check libthread-db}.
38038 @kindex maint space
38039 @cindex memory used by commands
38040 @item maint space @var{value}
38041 An alias for @code{maint set per-command space}.
38042 A non-zero value enables it, zero disables it.
38045 @cindex time of command execution
38046 @item maint time @var{value}
38047 An alias for @code{maint set per-command time}.
38048 A non-zero value enables it, zero disables it.
38050 @kindex maint translate-address
38051 @item maint translate-address @r{[}@var{section}@r{]} @var{addr}
38052 Find the symbol stored at the location specified by the address
38053 @var{addr} and an optional section name @var{section}. If found,
38054 @value{GDBN} prints the name of the closest symbol and an offset from
38055 the symbol's location to the specified address. This is similar to
38056 the @code{info address} command (@pxref{Symbols}), except that this
38057 command also allows to find symbols in other sections.
38059 If section was not specified, the section in which the symbol was found
38060 is also printed. For dynamically linked executables, the name of
38061 executable or shared library containing the symbol is printed as well.
38063 @kindex maint test-options
38064 @item maint test-options require-delimiter
38065 @itemx maint test-options unknown-is-error
38066 @itemx maint test-options unknown-is-operand
38067 These commands are used by the testsuite to validate the command
38068 options framework. The @code{require-delimiter} variant requires a
38069 double-dash delimiter to indicate end of options. The
38070 @code{unknown-is-error} and @code{unknown-is-operand} do not. The
38071 @code{unknown-is-error} variant throws an error on unknown option,
38072 while @code{unknown-is-operand} treats unknown options as the start of
38073 the command's operands. When run, the commands output the result of
38074 the processed options. When completed, the commands store the
38075 internal result of completion in a variable exposed by the @code{maint
38076 show test-options-completion-result} command.
38078 @kindex maint show test-options-completion-result
38079 @item maint show test-options-completion-result
38080 Shows the result of completing the @code{maint test-options}
38081 subcommands. This is used by the testsuite to validate completion
38082 support in the command options framework.
38084 @kindex maint set test-settings
38085 @kindex maint show test-settings
38086 @item maint set test-settings @var{kind}
38087 @itemx maint show test-settings @var{kind}
38088 These are representative commands for each @var{kind} of setting type
38089 @value{GDBN} supports. They are used by the testsuite for exercising
38090 the settings infrastructure.
38093 @item maint with @var{setting} [@var{value}] [-- @var{command}]
38094 Like the @code{with} command, but works with @code{maintenance set}
38095 variables. This is used by the testsuite to exercise the @code{with}
38096 command's infrastructure.
38100 The following command is useful for non-interactive invocations of
38101 @value{GDBN}, such as in the test suite.
38104 @item set watchdog @var{nsec}
38105 @kindex set watchdog
38106 @cindex watchdog timer
38107 @cindex timeout for commands
38108 Set the maximum number of seconds @value{GDBN} will wait for the
38109 target operation to finish. If this time expires, @value{GDBN}
38110 reports and error and the command is aborted.
38112 @item show watchdog
38113 Show the current setting of the target wait timeout.
38116 @node Remote Protocol
38117 @appendix @value{GDBN} Remote Serial Protocol
38122 * Stop Reply Packets::
38123 * General Query Packets::
38124 * Architecture-Specific Protocol Details::
38125 * Tracepoint Packets::
38126 * Host I/O Packets::
38128 * Notification Packets::
38129 * Remote Non-Stop::
38130 * Packet Acknowledgment::
38132 * File-I/O Remote Protocol Extension::
38133 * Library List Format::
38134 * Library List Format for SVR4 Targets::
38135 * Memory Map Format::
38136 * Thread List Format::
38137 * Traceframe Info Format::
38138 * Branch Trace Format::
38139 * Branch Trace Configuration Format::
38145 There may be occasions when you need to know something about the
38146 protocol---for example, if there is only one serial port to your target
38147 machine, you might want your program to do something special if it
38148 recognizes a packet meant for @value{GDBN}.
38150 In the examples below, @samp{->} and @samp{<-} are used to indicate
38151 transmitted and received data, respectively.
38153 @cindex protocol, @value{GDBN} remote serial
38154 @cindex serial protocol, @value{GDBN} remote
38155 @cindex remote serial protocol
38156 All @value{GDBN} commands and responses (other than acknowledgments
38157 and notifications, see @ref{Notification Packets}) are sent as a
38158 @var{packet}. A @var{packet} is introduced with the character
38159 @samp{$}, the actual @var{packet-data}, and the terminating character
38160 @samp{#} followed by a two-digit @var{checksum}:
38163 @code{$}@var{packet-data}@code{#}@var{checksum}
38167 @cindex checksum, for @value{GDBN} remote
38169 The two-digit @var{checksum} is computed as the modulo 256 sum of all
38170 characters between the leading @samp{$} and the trailing @samp{#} (an
38171 eight bit unsigned checksum).
38173 Implementors should note that prior to @value{GDBN} 5.0 the protocol
38174 specification also included an optional two-digit @var{sequence-id}:
38177 @code{$}@var{sequence-id}@code{:}@var{packet-data}@code{#}@var{checksum}
38180 @cindex sequence-id, for @value{GDBN} remote
38182 That @var{sequence-id} was appended to the acknowledgment. @value{GDBN}
38183 has never output @var{sequence-id}s. Stubs that handle packets added
38184 since @value{GDBN} 5.0 must not accept @var{sequence-id}.
38186 When either the host or the target machine receives a packet, the first
38187 response expected is an acknowledgment: either @samp{+} (to indicate
38188 the package was received correctly) or @samp{-} (to request
38192 -> @code{$}@var{packet-data}@code{#}@var{checksum}
38197 The @samp{+}/@samp{-} acknowledgments can be disabled
38198 once a connection is established.
38199 @xref{Packet Acknowledgment}, for details.
38201 The host (@value{GDBN}) sends @var{command}s, and the target (the
38202 debugging stub incorporated in your program) sends a @var{response}. In
38203 the case of step and continue @var{command}s, the response is only sent
38204 when the operation has completed, and the target has again stopped all
38205 threads in all attached processes. This is the default all-stop mode
38206 behavior, but the remote protocol also supports @value{GDBN}'s non-stop
38207 execution mode; see @ref{Remote Non-Stop}, for details.
38209 @var{packet-data} consists of a sequence of characters with the
38210 exception of @samp{#} and @samp{$} (see @samp{X} packet for additional
38213 @cindex remote protocol, field separator
38214 Fields within the packet should be separated using @samp{,} @samp{;} or
38215 @samp{:}. Except where otherwise noted all numbers are represented in
38216 @sc{hex} with leading zeros suppressed.
38218 Implementors should note that prior to @value{GDBN} 5.0, the character
38219 @samp{:} could not appear as the third character in a packet (as it
38220 would potentially conflict with the @var{sequence-id}).
38222 @cindex remote protocol, binary data
38223 @anchor{Binary Data}
38224 Binary data in most packets is encoded either as two hexadecimal
38225 digits per byte of binary data. This allowed the traditional remote
38226 protocol to work over connections which were only seven-bit clean.
38227 Some packets designed more recently assume an eight-bit clean
38228 connection, and use a more efficient encoding to send and receive
38231 The binary data representation uses @code{7d} (@sc{ascii} @samp{@}})
38232 as an escape character. Any escaped byte is transmitted as the escape
38233 character followed by the original character XORed with @code{0x20}.
38234 For example, the byte @code{0x7d} would be transmitted as the two
38235 bytes @code{0x7d 0x5d}. The bytes @code{0x23} (@sc{ascii} @samp{#}),
38236 @code{0x24} (@sc{ascii} @samp{$}), and @code{0x7d} (@sc{ascii}
38237 @samp{@}}) must always be escaped. Responses sent by the stub
38238 must also escape @code{0x2a} (@sc{ascii} @samp{*}), so that it
38239 is not interpreted as the start of a run-length encoded sequence
38242 Response @var{data} can be run-length encoded to save space.
38243 Run-length encoding replaces runs of identical characters with one
38244 instance of the repeated character, followed by a @samp{*} and a
38245 repeat count. The repeat count is itself sent encoded, to avoid
38246 binary characters in @var{data}: a value of @var{n} is sent as
38247 @code{@var{n}+29}. For a repeat count greater or equal to 3, this
38248 produces a printable @sc{ascii} character, e.g.@: a space (@sc{ascii}
38249 code 32) for a repeat count of 3. (This is because run-length
38250 encoding starts to win for counts 3 or more.) Thus, for example,
38251 @samp{0* } is a run-length encoding of ``0000'': the space character
38252 after @samp{*} means repeat the leading @code{0} @w{@code{32 - 29 =
38255 The printable characters @samp{#} and @samp{$} or with a numeric value
38256 greater than 126 must not be used. Runs of six repeats (@samp{#}) or
38257 seven repeats (@samp{$}) can be expanded using a repeat count of only
38258 five (@samp{"}). For example, @samp{00000000} can be encoded as
38261 The error response returned for some packets includes a two character
38262 error number. That number is not well defined.
38264 @cindex empty response, for unsupported packets
38265 For any @var{command} not supported by the stub, an empty response
38266 (@samp{$#00}) should be returned. That way it is possible to extend the
38267 protocol. A newer @value{GDBN} can tell if a packet is supported based
38270 At a minimum, a stub is required to support the @samp{g} and @samp{G}
38271 commands for register access, and the @samp{m} and @samp{M} commands
38272 for memory access. Stubs that only control single-threaded targets
38273 can implement run control with the @samp{c} (continue), and @samp{s}
38274 (step) commands. Stubs that support multi-threading targets should
38275 support the @samp{vCont} command. All other commands are optional.
38280 The following table provides a complete list of all currently defined
38281 @var{command}s and their corresponding response @var{data}.
38282 @xref{File-I/O Remote Protocol Extension}, for details about the File
38283 I/O extension of the remote protocol.
38285 Each packet's description has a template showing the packet's overall
38286 syntax, followed by an explanation of the packet's meaning. We
38287 include spaces in some of the templates for clarity; these are not
38288 part of the packet's syntax. No @value{GDBN} packet uses spaces to
38289 separate its components. For example, a template like @samp{foo
38290 @var{bar} @var{baz}} describes a packet beginning with the three ASCII
38291 bytes @samp{foo}, followed by a @var{bar}, followed directly by a
38292 @var{baz}. @value{GDBN} does not transmit a space character between the
38293 @samp{foo} and the @var{bar}, or between the @var{bar} and the
38296 @cindex @var{thread-id}, in remote protocol
38297 @anchor{thread-id syntax}
38298 Several packets and replies include a @var{thread-id} field to identify
38299 a thread. Normally these are positive numbers with a target-specific
38300 interpretation, formatted as big-endian hex strings. A @var{thread-id}
38301 can also be a literal @samp{-1} to indicate all threads, or @samp{0} to
38304 In addition, the remote protocol supports a multiprocess feature in
38305 which the @var{thread-id} syntax is extended to optionally include both
38306 process and thread ID fields, as @samp{p@var{pid}.@var{tid}}.
38307 The @var{pid} (process) and @var{tid} (thread) components each have the
38308 format described above: a positive number with target-specific
38309 interpretation formatted as a big-endian hex string, literal @samp{-1}
38310 to indicate all processes or threads (respectively), or @samp{0} to
38311 indicate an arbitrary process or thread. Specifying just a process, as
38312 @samp{p@var{pid}}, is equivalent to @samp{p@var{pid}.-1}. It is an
38313 error to specify all processes but a specific thread, such as
38314 @samp{p-1.@var{tid}}. Note that the @samp{p} prefix is @emph{not} used
38315 for those packets and replies explicitly documented to include a process
38316 ID, rather than a @var{thread-id}.
38318 The multiprocess @var{thread-id} syntax extensions are only used if both
38319 @value{GDBN} and the stub report support for the @samp{multiprocess}
38320 feature using @samp{qSupported}. @xref{multiprocess extensions}, for
38323 Note that all packet forms beginning with an upper- or lower-case
38324 letter, other than those described here, are reserved for future use.
38326 Here are the packet descriptions.
38331 @cindex @samp{!} packet
38332 @anchor{extended mode}
38333 Enable extended mode. In extended mode, the remote server is made
38334 persistent. The @samp{R} packet is used to restart the program being
38340 The remote target both supports and has enabled extended mode.
38344 @cindex @samp{?} packet
38346 Indicate the reason the target halted. The reply is the same as for
38347 step and continue. This packet has a special interpretation when the
38348 target is in non-stop mode; see @ref{Remote Non-Stop}.
38351 @xref{Stop Reply Packets}, for the reply specifications.
38353 @item A @var{arglen},@var{argnum},@var{arg},@dots{}
38354 @cindex @samp{A} packet
38355 Initialized @code{argv[]} array passed into program. @var{arglen}
38356 specifies the number of bytes in the hex encoded byte stream
38357 @var{arg}. See @code{gdbserver} for more details.
38362 The arguments were set.
38368 @cindex @samp{b} packet
38369 (Don't use this packet; its behavior is not well-defined.)
38370 Change the serial line speed to @var{baud}.
38372 JTC: @emph{When does the transport layer state change? When it's
38373 received, or after the ACK is transmitted. In either case, there are
38374 problems if the command or the acknowledgment packet is dropped.}
38376 Stan: @emph{If people really wanted to add something like this, and get
38377 it working for the first time, they ought to modify ser-unix.c to send
38378 some kind of out-of-band message to a specially-setup stub and have the
38379 switch happen "in between" packets, so that from remote protocol's point
38380 of view, nothing actually happened.}
38382 @item B @var{addr},@var{mode}
38383 @cindex @samp{B} packet
38384 Set (@var{mode} is @samp{S}) or clear (@var{mode} is @samp{C}) a
38385 breakpoint at @var{addr}.
38387 Don't use this packet. Use the @samp{Z} and @samp{z} packets instead
38388 (@pxref{insert breakpoint or watchpoint packet}).
38390 @cindex @samp{bc} packet
38393 Backward continue. Execute the target system in reverse. No parameter.
38394 @xref{Reverse Execution}, for more information.
38397 @xref{Stop Reply Packets}, for the reply specifications.
38399 @cindex @samp{bs} packet
38402 Backward single step. Execute one instruction in reverse. No parameter.
38403 @xref{Reverse Execution}, for more information.
38406 @xref{Stop Reply Packets}, for the reply specifications.
38408 @item c @r{[}@var{addr}@r{]}
38409 @cindex @samp{c} packet
38410 Continue at @var{addr}, which is the address to resume. If @var{addr}
38411 is omitted, resume at current address.
38413 This packet is deprecated for multi-threading support. @xref{vCont
38417 @xref{Stop Reply Packets}, for the reply specifications.
38419 @item C @var{sig}@r{[};@var{addr}@r{]}
38420 @cindex @samp{C} packet
38421 Continue with signal @var{sig} (hex signal number). If
38422 @samp{;@var{addr}} is omitted, resume at same address.
38424 This packet is deprecated for multi-threading support. @xref{vCont
38428 @xref{Stop Reply Packets}, for the reply specifications.
38431 @cindex @samp{d} packet
38434 Don't use this packet; instead, define a general set packet
38435 (@pxref{General Query Packets}).
38439 @cindex @samp{D} packet
38440 The first form of the packet is used to detach @value{GDBN} from the
38441 remote system. It is sent to the remote target
38442 before @value{GDBN} disconnects via the @code{detach} command.
38444 The second form, including a process ID, is used when multiprocess
38445 protocol extensions are enabled (@pxref{multiprocess extensions}), to
38446 detach only a specific process. The @var{pid} is specified as a
38447 big-endian hex string.
38457 @item F @var{RC},@var{EE},@var{CF};@var{XX}
38458 @cindex @samp{F} packet
38459 A reply from @value{GDBN} to an @samp{F} packet sent by the target.
38460 This is part of the File-I/O protocol extension. @xref{File-I/O
38461 Remote Protocol Extension}, for the specification.
38464 @anchor{read registers packet}
38465 @cindex @samp{g} packet
38466 Read general registers.
38470 @item @var{XX@dots{}}
38471 Each byte of register data is described by two hex digits. The bytes
38472 with the register are transmitted in target byte order. The size of
38473 each register and their position within the @samp{g} packet are
38474 determined by the @value{GDBN} internal gdbarch functions
38475 @code{DEPRECATED_REGISTER_RAW_SIZE} and @code{gdbarch_register_name}.
38477 When reading registers from a trace frame (@pxref{Analyze Collected
38478 Data,,Using the Collected Data}), the stub may also return a string of
38479 literal @samp{x}'s in place of the register data digits, to indicate
38480 that the corresponding register has not been collected, thus its value
38481 is unavailable. For example, for an architecture with 4 registers of
38482 4 bytes each, the following reply indicates to @value{GDBN} that
38483 registers 0 and 2 have not been collected, while registers 1 and 3
38484 have been collected, and both have zero value:
38488 <- @code{xxxxxxxx00000000xxxxxxxx00000000}
38495 @item G @var{XX@dots{}}
38496 @cindex @samp{G} packet
38497 Write general registers. @xref{read registers packet}, for a
38498 description of the @var{XX@dots{}} data.
38508 @item H @var{op} @var{thread-id}
38509 @cindex @samp{H} packet
38510 Set thread for subsequent operations (@samp{m}, @samp{M}, @samp{g},
38511 @samp{G}, et.al.). Depending on the operation to be performed, @var{op}
38512 should be @samp{c} for step and continue operations (note that this
38513 is deprecated, supporting the @samp{vCont} command is a better
38514 option), and @samp{g} for other operations. The thread designator
38515 @var{thread-id} has the format and interpretation described in
38516 @ref{thread-id syntax}.
38527 @c 'H': How restrictive (or permissive) is the thread model. If a
38528 @c thread is selected and stopped, are other threads allowed
38529 @c to continue to execute? As I mentioned above, I think the
38530 @c semantics of each command when a thread is selected must be
38531 @c described. For example:
38533 @c 'g': If the stub supports threads and a specific thread is
38534 @c selected, returns the register block from that thread;
38535 @c otherwise returns current registers.
38537 @c 'G' If the stub supports threads and a specific thread is
38538 @c selected, sets the registers of the register block of
38539 @c that thread; otherwise sets current registers.
38541 @item i @r{[}@var{addr}@r{[},@var{nnn}@r{]]}
38542 @anchor{cycle step packet}
38543 @cindex @samp{i} packet
38544 Step the remote target by a single clock cycle. If @samp{,@var{nnn}} is
38545 present, cycle step @var{nnn} cycles. If @var{addr} is present, cycle
38546 step starting at that address.
38549 @cindex @samp{I} packet
38550 Signal, then cycle step. @xref{step with signal packet}. @xref{cycle
38554 @cindex @samp{k} packet
38557 The exact effect of this packet is not specified.
38559 For a bare-metal target, it may power cycle or reset the target
38560 system. For that reason, the @samp{k} packet has no reply.
38562 For a single-process target, it may kill that process if possible.
38564 A multiple-process target may choose to kill just one process, or all
38565 that are under @value{GDBN}'s control. For more precise control, use
38566 the vKill packet (@pxref{vKill packet}).
38568 If the target system immediately closes the connection in response to
38569 @samp{k}, @value{GDBN} does not consider the lack of packet
38570 acknowledgment to be an error, and assumes the kill was successful.
38572 If connected using @kbd{target extended-remote}, and the target does
38573 not close the connection in response to a kill request, @value{GDBN}
38574 probes the target state as if a new connection was opened
38575 (@pxref{? packet}).
38577 @item m @var{addr},@var{length}
38578 @cindex @samp{m} packet
38579 Read @var{length} addressable memory units starting at address @var{addr}
38580 (@pxref{addressable memory unit}). Note that @var{addr} may not be aligned to
38581 any particular boundary.
38583 The stub need not use any particular size or alignment when gathering
38584 data from memory for the response; even if @var{addr} is word-aligned
38585 and @var{length} is a multiple of the word size, the stub is free to
38586 use byte accesses, or not. For this reason, this packet may not be
38587 suitable for accessing memory-mapped I/O devices.
38588 @cindex alignment of remote memory accesses
38589 @cindex size of remote memory accesses
38590 @cindex memory, alignment and size of remote accesses
38594 @item @var{XX@dots{}}
38595 Memory contents; each byte is transmitted as a two-digit hexadecimal number.
38596 The reply may contain fewer addressable memory units than requested if the
38597 server was able to read only part of the region of memory.
38602 @item M @var{addr},@var{length}:@var{XX@dots{}}
38603 @cindex @samp{M} packet
38604 Write @var{length} addressable memory units starting at address @var{addr}
38605 (@pxref{addressable memory unit}). The data is given by @var{XX@dots{}}; each
38606 byte is transmitted as a two-digit hexadecimal number.
38613 for an error (this includes the case where only part of the data was
38618 @cindex @samp{p} packet
38619 Read the value of register @var{n}; @var{n} is in hex.
38620 @xref{read registers packet}, for a description of how the returned
38621 register value is encoded.
38625 @item @var{XX@dots{}}
38626 the register's value
38630 Indicating an unrecognized @var{query}.
38633 @item P @var{n@dots{}}=@var{r@dots{}}
38634 @anchor{write register packet}
38635 @cindex @samp{P} packet
38636 Write register @var{n@dots{}} with value @var{r@dots{}}. The register
38637 number @var{n} is in hexadecimal, and @var{r@dots{}} contains two hex
38638 digits for each byte in the register (target byte order).
38648 @item q @var{name} @var{params}@dots{}
38649 @itemx Q @var{name} @var{params}@dots{}
38650 @cindex @samp{q} packet
38651 @cindex @samp{Q} packet
38652 General query (@samp{q}) and set (@samp{Q}). These packets are
38653 described fully in @ref{General Query Packets}.
38656 @cindex @samp{r} packet
38657 Reset the entire system.
38659 Don't use this packet; use the @samp{R} packet instead.
38662 @cindex @samp{R} packet
38663 Restart the program being debugged. The @var{XX}, while needed, is ignored.
38664 This packet is only available in extended mode (@pxref{extended mode}).
38666 The @samp{R} packet has no reply.
38668 @item s @r{[}@var{addr}@r{]}
38669 @cindex @samp{s} packet
38670 Single step, resuming at @var{addr}. If
38671 @var{addr} is omitted, resume at same address.
38673 This packet is deprecated for multi-threading support. @xref{vCont
38677 @xref{Stop Reply Packets}, for the reply specifications.
38679 @item S @var{sig}@r{[};@var{addr}@r{]}
38680 @anchor{step with signal packet}
38681 @cindex @samp{S} packet
38682 Step with signal. This is analogous to the @samp{C} packet, but
38683 requests a single-step, rather than a normal resumption of execution.
38685 This packet is deprecated for multi-threading support. @xref{vCont
38689 @xref{Stop Reply Packets}, for the reply specifications.
38691 @item t @var{addr}:@var{PP},@var{MM}
38692 @cindex @samp{t} packet
38693 Search backwards starting at address @var{addr} for a match with pattern
38694 @var{PP} and mask @var{MM}, both of which are are 4 byte long.
38695 There must be at least 3 digits in @var{addr}.
38697 @item T @var{thread-id}
38698 @cindex @samp{T} packet
38699 Find out if the thread @var{thread-id} is alive. @xref{thread-id syntax}.
38704 thread is still alive
38710 Packets starting with @samp{v} are identified by a multi-letter name,
38711 up to the first @samp{;} or @samp{?} (or the end of the packet).
38713 @item vAttach;@var{pid}
38714 @cindex @samp{vAttach} packet
38715 Attach to a new process with the specified process ID @var{pid}.
38716 The process ID is a
38717 hexadecimal integer identifying the process. In all-stop mode, all
38718 threads in the attached process are stopped; in non-stop mode, it may be
38719 attached without being stopped if that is supported by the target.
38721 @c In non-stop mode, on a successful vAttach, the stub should set the
38722 @c current thread to a thread of the newly-attached process. After
38723 @c attaching, GDB queries for the attached process's thread ID with qC.
38724 @c Also note that, from a user perspective, whether or not the
38725 @c target is stopped on attach in non-stop mode depends on whether you
38726 @c use the foreground or background version of the attach command, not
38727 @c on what vAttach does; GDB does the right thing with respect to either
38728 @c stopping or restarting threads.
38730 This packet is only available in extended mode (@pxref{extended mode}).
38736 @item @r{Any stop packet}
38737 for success in all-stop mode (@pxref{Stop Reply Packets})
38739 for success in non-stop mode (@pxref{Remote Non-Stop})
38742 @item vCont@r{[};@var{action}@r{[}:@var{thread-id}@r{]]}@dots{}
38743 @cindex @samp{vCont} packet
38744 @anchor{vCont packet}
38745 Resume the inferior, specifying different actions for each thread.
38747 For each inferior thread, the leftmost action with a matching
38748 @var{thread-id} is applied. Threads that don't match any action
38749 remain in their current state. Thread IDs are specified using the
38750 syntax described in @ref{thread-id syntax}. If multiprocess
38751 extensions (@pxref{multiprocess extensions}) are supported, actions
38752 can be specified to match all threads in a process by using the
38753 @samp{p@var{pid}.-1} form of the @var{thread-id}. An action with no
38754 @var{thread-id} matches all threads. Specifying no actions is an
38757 Currently supported actions are:
38763 Continue with signal @var{sig}. The signal @var{sig} should be two hex digits.
38767 Step with signal @var{sig}. The signal @var{sig} should be two hex digits.
38770 @item r @var{start},@var{end}
38771 Step once, and then keep stepping as long as the thread stops at
38772 addresses between @var{start} (inclusive) and @var{end} (exclusive).
38773 The remote stub reports a stop reply when either the thread goes out
38774 of the range or is stopped due to an unrelated reason, such as hitting
38775 a breakpoint. @xref{range stepping}.
38777 If the range is empty (@var{start} == @var{end}), then the action
38778 becomes equivalent to the @samp{s} action. In other words,
38779 single-step once, and report the stop (even if the stepped instruction
38780 jumps to @var{start}).
38782 (A stop reply may be sent at any point even if the PC is still within
38783 the stepping range; for example, it is valid to implement this packet
38784 in a degenerate way as a single instruction step operation.)
38788 The optional argument @var{addr} normally associated with the
38789 @samp{c}, @samp{C}, @samp{s}, and @samp{S} packets is
38790 not supported in @samp{vCont}.
38792 The @samp{t} action is only relevant in non-stop mode
38793 (@pxref{Remote Non-Stop}) and may be ignored by the stub otherwise.
38794 A stop reply should be generated for any affected thread not already stopped.
38795 When a thread is stopped by means of a @samp{t} action,
38796 the corresponding stop reply should indicate that the thread has stopped with
38797 signal @samp{0}, regardless of whether the target uses some other signal
38798 as an implementation detail.
38800 The server must ignore @samp{c}, @samp{C}, @samp{s}, @samp{S}, and
38801 @samp{r} actions for threads that are already running. Conversely,
38802 the server must ignore @samp{t} actions for threads that are already
38805 @emph{Note:} In non-stop mode, a thread is considered running until
38806 @value{GDBN} acknowledges an asynchronous stop notification for it with
38807 the @samp{vStopped} packet (@pxref{Remote Non-Stop}).
38809 The stub must support @samp{vCont} if it reports support for
38810 multiprocess extensions (@pxref{multiprocess extensions}).
38813 @xref{Stop Reply Packets}, for the reply specifications.
38816 @cindex @samp{vCont?} packet
38817 Request a list of actions supported by the @samp{vCont} packet.
38821 @item vCont@r{[};@var{action}@dots{}@r{]}
38822 The @samp{vCont} packet is supported. Each @var{action} is a supported
38823 command in the @samp{vCont} packet.
38825 The @samp{vCont} packet is not supported.
38828 @anchor{vCtrlC packet}
38830 @cindex @samp{vCtrlC} packet
38831 Interrupt remote target as if a control-C was pressed on the remote
38832 terminal. This is the equivalent to reacting to the @code{^C}
38833 (@samp{\003}, the control-C character) character in all-stop mode
38834 while the target is running, except this works in non-stop mode.
38835 @xref{interrupting remote targets}, for more info on the all-stop
38846 @item vFile:@var{operation}:@var{parameter}@dots{}
38847 @cindex @samp{vFile} packet
38848 Perform a file operation on the target system. For details,
38849 see @ref{Host I/O Packets}.
38851 @item vFlashErase:@var{addr},@var{length}
38852 @cindex @samp{vFlashErase} packet
38853 Direct the stub to erase @var{length} bytes of flash starting at
38854 @var{addr}. The region may enclose any number of flash blocks, but
38855 its start and end must fall on block boundaries, as indicated by the
38856 flash block size appearing in the memory map (@pxref{Memory Map
38857 Format}). @value{GDBN} groups flash memory programming operations
38858 together, and sends a @samp{vFlashDone} request after each group; the
38859 stub is allowed to delay erase operation until the @samp{vFlashDone}
38860 packet is received.
38870 @item vFlashWrite:@var{addr}:@var{XX@dots{}}
38871 @cindex @samp{vFlashWrite} packet
38872 Direct the stub to write data to flash address @var{addr}. The data
38873 is passed in binary form using the same encoding as for the @samp{X}
38874 packet (@pxref{Binary Data}). The memory ranges specified by
38875 @samp{vFlashWrite} packets preceding a @samp{vFlashDone} packet must
38876 not overlap, and must appear in order of increasing addresses
38877 (although @samp{vFlashErase} packets for higher addresses may already
38878 have been received; the ordering is guaranteed only between
38879 @samp{vFlashWrite} packets). If a packet writes to an address that was
38880 neither erased by a preceding @samp{vFlashErase} packet nor by some other
38881 target-specific method, the results are unpredictable.
38889 for vFlashWrite addressing non-flash memory
38895 @cindex @samp{vFlashDone} packet
38896 Indicate to the stub that flash programming operation is finished.
38897 The stub is permitted to delay or batch the effects of a group of
38898 @samp{vFlashErase} and @samp{vFlashWrite} packets until a
38899 @samp{vFlashDone} packet is received. The contents of the affected
38900 regions of flash memory are unpredictable until the @samp{vFlashDone}
38901 request is completed.
38903 @item vKill;@var{pid}
38904 @cindex @samp{vKill} packet
38905 @anchor{vKill packet}
38906 Kill the process with the specified process ID @var{pid}, which is a
38907 hexadecimal integer identifying the process. This packet is used in
38908 preference to @samp{k} when multiprocess protocol extensions are
38909 supported; see @ref{multiprocess extensions}.
38919 @item vMustReplyEmpty
38920 @cindex @samp{vMustReplyEmpty} packet
38921 The correct reply to an unknown @samp{v} packet is to return the empty
38922 string, however, some older versions of @command{gdbserver} would
38923 incorrectly return @samp{OK} for unknown @samp{v} packets.
38925 The @samp{vMustReplyEmpty} is used as a feature test to check how
38926 @command{gdbserver} handles unknown packets, it is important that this
38927 packet be handled in the same way as other unknown @samp{v} packets.
38928 If this packet is handled differently to other unknown @samp{v}
38929 packets then it is possible that @value{GDBN} may run into problems in
38930 other areas, specifically around use of @samp{vFile:setfs:}.
38932 @item vRun;@var{filename}@r{[};@var{argument}@r{]}@dots{}
38933 @cindex @samp{vRun} packet
38934 Run the program @var{filename}, passing it each @var{argument} on its
38935 command line. The file and arguments are hex-encoded strings. If
38936 @var{filename} is an empty string, the stub may use a default program
38937 (e.g.@: the last program run). The program is created in the stopped
38940 @c FIXME: What about non-stop mode?
38942 This packet is only available in extended mode (@pxref{extended mode}).
38948 @item @r{Any stop packet}
38949 for success (@pxref{Stop Reply Packets})
38953 @cindex @samp{vStopped} packet
38954 @xref{Notification Packets}.
38956 @item X @var{addr},@var{length}:@var{XX@dots{}}
38958 @cindex @samp{X} packet
38959 Write data to memory, where the data is transmitted in binary.
38960 Memory is specified by its address @var{addr} and number of addressable memory
38961 units @var{length} (@pxref{addressable memory unit});
38962 @samp{@var{XX}@dots{}} is binary data (@pxref{Binary Data}).
38972 @item z @var{type},@var{addr},@var{kind}
38973 @itemx Z @var{type},@var{addr},@var{kind}
38974 @anchor{insert breakpoint or watchpoint packet}
38975 @cindex @samp{z} packet
38976 @cindex @samp{Z} packets
38977 Insert (@samp{Z}) or remove (@samp{z}) a @var{type} breakpoint or
38978 watchpoint starting at address @var{address} of kind @var{kind}.
38980 Each breakpoint and watchpoint packet @var{type} is documented
38983 @emph{Implementation notes: A remote target shall return an empty string
38984 for an unrecognized breakpoint or watchpoint packet @var{type}. A
38985 remote target shall support either both or neither of a given
38986 @samp{Z@var{type}@dots{}} and @samp{z@var{type}@dots{}} packet pair. To
38987 avoid potential problems with duplicate packets, the operations should
38988 be implemented in an idempotent way.}
38990 @item z0,@var{addr},@var{kind}
38991 @itemx Z0,@var{addr},@var{kind}@r{[};@var{cond_list}@dots{}@r{]}@r{[};cmds:@var{persist},@var{cmd_list}@dots{}@r{]}
38992 @cindex @samp{z0} packet
38993 @cindex @samp{Z0} packet
38994 Insert (@samp{Z0}) or remove (@samp{z0}) a software breakpoint at address
38995 @var{addr} of type @var{kind}.
38997 A software breakpoint is implemented by replacing the instruction at
38998 @var{addr} with a software breakpoint or trap instruction. The
38999 @var{kind} is target-specific and typically indicates the size of the
39000 breakpoint in bytes that should be inserted. E.g., the @sc{arm} and
39001 @sc{mips} can insert either a 2 or 4 byte breakpoint. Some
39002 architectures have additional meanings for @var{kind}
39003 (@pxref{Architecture-Specific Protocol Details}); if no
39004 architecture-specific value is being used, it should be @samp{0}.
39005 @var{kind} is hex-encoded. @var{cond_list} is an optional list of
39006 conditional expressions in bytecode form that should be evaluated on
39007 the target's side. These are the conditions that should be taken into
39008 consideration when deciding if the breakpoint trigger should be
39009 reported back to @value{GDBN}.
39011 See also the @samp{swbreak} stop reason (@pxref{swbreak stop reason})
39012 for how to best report a software breakpoint event to @value{GDBN}.
39014 The @var{cond_list} parameter is comprised of a series of expressions,
39015 concatenated without separators. Each expression has the following form:
39019 @item X @var{len},@var{expr}
39020 @var{len} is the length of the bytecode expression and @var{expr} is the
39021 actual conditional expression in bytecode form.
39025 The optional @var{cmd_list} parameter introduces commands that may be
39026 run on the target, rather than being reported back to @value{GDBN}.
39027 The parameter starts with a numeric flag @var{persist}; if the flag is
39028 nonzero, then the breakpoint may remain active and the commands
39029 continue to be run even when @value{GDBN} disconnects from the target.
39030 Following this flag is a series of expressions concatenated with no
39031 separators. Each expression has the following form:
39035 @item X @var{len},@var{expr}
39036 @var{len} is the length of the bytecode expression and @var{expr} is the
39037 actual commands expression in bytecode form.
39041 @emph{Implementation note: It is possible for a target to copy or move
39042 code that contains software breakpoints (e.g., when implementing
39043 overlays). The behavior of this packet, in the presence of such a
39044 target, is not defined.}
39056 @item z1,@var{addr},@var{kind}
39057 @itemx Z1,@var{addr},@var{kind}@r{[};@var{cond_list}@dots{}@r{]}@r{[};cmds:@var{persist},@var{cmd_list}@dots{}@r{]}
39058 @cindex @samp{z1} packet
39059 @cindex @samp{Z1} packet
39060 Insert (@samp{Z1}) or remove (@samp{z1}) a hardware breakpoint at
39061 address @var{addr}.
39063 A hardware breakpoint is implemented using a mechanism that is not
39064 dependent on being able to modify the target's memory. The
39065 @var{kind}, @var{cond_list}, and @var{cmd_list} arguments have the
39066 same meaning as in @samp{Z0} packets.
39068 @emph{Implementation note: A hardware breakpoint is not affected by code
39081 @item z2,@var{addr},@var{kind}
39082 @itemx Z2,@var{addr},@var{kind}
39083 @cindex @samp{z2} packet
39084 @cindex @samp{Z2} packet
39085 Insert (@samp{Z2}) or remove (@samp{z2}) a write watchpoint at @var{addr}.
39086 The number of bytes to watch is specified by @var{kind}.
39098 @item z3,@var{addr},@var{kind}
39099 @itemx Z3,@var{addr},@var{kind}
39100 @cindex @samp{z3} packet
39101 @cindex @samp{Z3} packet
39102 Insert (@samp{Z3}) or remove (@samp{z3}) a read watchpoint at @var{addr}.
39103 The number of bytes to watch is specified by @var{kind}.
39115 @item z4,@var{addr},@var{kind}
39116 @itemx Z4,@var{addr},@var{kind}
39117 @cindex @samp{z4} packet
39118 @cindex @samp{Z4} packet
39119 Insert (@samp{Z4}) or remove (@samp{z4}) an access watchpoint at @var{addr}.
39120 The number of bytes to watch is specified by @var{kind}.
39134 @node Stop Reply Packets
39135 @section Stop Reply Packets
39136 @cindex stop reply packets
39138 The @samp{C}, @samp{c}, @samp{S}, @samp{s}, @samp{vCont},
39139 @samp{vAttach}, @samp{vRun}, @samp{vStopped}, and @samp{?} packets can
39140 receive any of the below as a reply. Except for @samp{?}
39141 and @samp{vStopped}, that reply is only returned
39142 when the target halts. In the below the exact meaning of @dfn{signal
39143 number} is defined by the header @file{include/gdb/signals.h} in the
39144 @value{GDBN} source code.
39146 In non-stop mode, the server will simply reply @samp{OK} to commands
39147 such as @samp{vCont}; any stop will be the subject of a future
39148 notification. @xref{Remote Non-Stop}.
39150 As in the description of request packets, we include spaces in the
39151 reply templates for clarity; these are not part of the reply packet's
39152 syntax. No @value{GDBN} stop reply packet uses spaces to separate its
39158 The program received signal number @var{AA} (a two-digit hexadecimal
39159 number). This is equivalent to a @samp{T} response with no
39160 @var{n}:@var{r} pairs.
39162 @item T @var{AA} @var{n1}:@var{r1};@var{n2}:@var{r2};@dots{}
39163 @cindex @samp{T} packet reply
39164 The program received signal number @var{AA} (a two-digit hexadecimal
39165 number). This is equivalent to an @samp{S} response, except that the
39166 @samp{@var{n}:@var{r}} pairs can carry values of important registers
39167 and other information directly in the stop reply packet, reducing
39168 round-trip latency. Single-step and breakpoint traps are reported
39169 this way. Each @samp{@var{n}:@var{r}} pair is interpreted as follows:
39173 If @var{n} is a hexadecimal number, it is a register number, and the
39174 corresponding @var{r} gives that register's value. The data @var{r} is a
39175 series of bytes in target byte order, with each byte given by a
39176 two-digit hex number.
39179 If @var{n} is @samp{thread}, then @var{r} is the @var{thread-id} of
39180 the stopped thread, as specified in @ref{thread-id syntax}.
39183 If @var{n} is @samp{core}, then @var{r} is the hexadecimal number of
39184 the core on which the stop event was detected.
39187 If @var{n} is a recognized @dfn{stop reason}, it describes a more
39188 specific event that stopped the target. The currently defined stop
39189 reasons are listed below. The @var{aa} should be @samp{05}, the trap
39190 signal. At most one stop reason should be present.
39193 Otherwise, @value{GDBN} should ignore this @samp{@var{n}:@var{r}} pair
39194 and go on to the next; this allows us to extend the protocol in the
39198 The currently defined stop reasons are:
39204 The packet indicates a watchpoint hit, and @var{r} is the data address, in
39207 @item syscall_entry
39208 @itemx syscall_return
39209 The packet indicates a syscall entry or return, and @var{r} is the
39210 syscall number, in hex.
39212 @cindex shared library events, remote reply
39214 The packet indicates that the loaded libraries have changed.
39215 @value{GDBN} should use @samp{qXfer:libraries:read} to fetch a new
39216 list of loaded libraries. The @var{r} part is ignored.
39218 @cindex replay log events, remote reply
39220 The packet indicates that the target cannot continue replaying
39221 logged execution events, because it has reached the end (or the
39222 beginning when executing backward) of the log. The value of @var{r}
39223 will be either @samp{begin} or @samp{end}. @xref{Reverse Execution},
39224 for more information.
39227 @anchor{swbreak stop reason}
39228 The packet indicates a software breakpoint instruction was executed,
39229 irrespective of whether it was @value{GDBN} that planted the
39230 breakpoint or the breakpoint is hardcoded in the program. The @var{r}
39231 part must be left empty.
39233 On some architectures, such as x86, at the architecture level, when a
39234 breakpoint instruction executes the program counter points at the
39235 breakpoint address plus an offset. On such targets, the stub is
39236 responsible for adjusting the PC to point back at the breakpoint
39239 This packet should not be sent by default; older @value{GDBN} versions
39240 did not support it. @value{GDBN} requests it, by supplying an
39241 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
39242 remote stub must also supply the appropriate @samp{qSupported} feature
39243 indicating support.
39245 This packet is required for correct non-stop mode operation.
39248 The packet indicates the target stopped for a hardware breakpoint.
39249 The @var{r} part must be left empty.
39251 The same remarks about @samp{qSupported} and non-stop mode above
39254 @cindex fork events, remote reply
39256 The packet indicates that @code{fork} was called, and @var{r}
39257 is the thread ID of the new child process. Refer to
39258 @ref{thread-id syntax} for the format of the @var{thread-id}
39259 field. This packet is only applicable to targets that support
39262 This packet should not be sent by default; older @value{GDBN} versions
39263 did not support it. @value{GDBN} requests it, by supplying an
39264 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
39265 remote stub must also supply the appropriate @samp{qSupported} feature
39266 indicating support.
39268 @cindex vfork events, remote reply
39270 The packet indicates that @code{vfork} was called, and @var{r}
39271 is the thread ID of the new child process. Refer to
39272 @ref{thread-id syntax} for the format of the @var{thread-id}
39273 field. This packet is only applicable to targets that support
39276 This packet should not be sent by default; older @value{GDBN} versions
39277 did not support it. @value{GDBN} requests it, by supplying an
39278 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
39279 remote stub must also supply the appropriate @samp{qSupported} feature
39280 indicating support.
39282 @cindex vforkdone events, remote reply
39284 The packet indicates that a child process created by a vfork
39285 has either called @code{exec} or terminated, so that the
39286 address spaces of the parent and child process are no longer
39287 shared. The @var{r} part is ignored. This packet is only
39288 applicable to targets that support vforkdone events.
39290 This packet should not be sent by default; older @value{GDBN} versions
39291 did not support it. @value{GDBN} requests it, by supplying an
39292 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
39293 remote stub must also supply the appropriate @samp{qSupported} feature
39294 indicating support.
39296 @cindex exec events, remote reply
39298 The packet indicates that @code{execve} was called, and @var{r}
39299 is the absolute pathname of the file that was executed, in hex.
39300 This packet is only applicable to targets that support exec events.
39302 This packet should not be sent by default; older @value{GDBN} versions
39303 did not support it. @value{GDBN} requests it, by supplying an
39304 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
39305 remote stub must also supply the appropriate @samp{qSupported} feature
39306 indicating support.
39308 @cindex thread create event, remote reply
39309 @anchor{thread create event}
39311 The packet indicates that the thread was just created. The new thread
39312 is stopped until @value{GDBN} sets it running with a resumption packet
39313 (@pxref{vCont packet}). This packet should not be sent by default;
39314 @value{GDBN} requests it with the @ref{QThreadEvents} packet. See
39315 also the @samp{w} (@pxref{thread exit event}) remote reply below. The
39316 @var{r} part is ignored.
39321 @itemx W @var{AA} ; process:@var{pid}
39322 The process exited, and @var{AA} is the exit status. This is only
39323 applicable to certain targets.
39325 The second form of the response, including the process ID of the
39326 exited process, can be used only when @value{GDBN} has reported
39327 support for multiprocess protocol extensions; see @ref{multiprocess
39328 extensions}. Both @var{AA} and @var{pid} are formatted as big-endian
39332 @itemx X @var{AA} ; process:@var{pid}
39333 The process terminated with signal @var{AA}.
39335 The second form of the response, including the process ID of the
39336 terminated process, can be used only when @value{GDBN} has reported
39337 support for multiprocess protocol extensions; see @ref{multiprocess
39338 extensions}. Both @var{AA} and @var{pid} are formatted as big-endian
39341 @anchor{thread exit event}
39342 @cindex thread exit event, remote reply
39343 @item w @var{AA} ; @var{tid}
39345 The thread exited, and @var{AA} is the exit status. This response
39346 should not be sent by default; @value{GDBN} requests it with the
39347 @ref{QThreadEvents} packet. See also @ref{thread create event} above.
39348 @var{AA} is formatted as a big-endian hex string.
39351 There are no resumed threads left in the target. In other words, even
39352 though the process is alive, the last resumed thread has exited. For
39353 example, say the target process has two threads: thread 1 and thread
39354 2. The client leaves thread 1 stopped, and resumes thread 2, which
39355 subsequently exits. At this point, even though the process is still
39356 alive, and thus no @samp{W} stop reply is sent, no thread is actually
39357 executing either. The @samp{N} stop reply thus informs the client
39358 that it can stop waiting for stop replies. This packet should not be
39359 sent by default; older @value{GDBN} versions did not support it.
39360 @value{GDBN} requests it, by supplying an appropriate
39361 @samp{qSupported} feature (@pxref{qSupported}). The remote stub must
39362 also supply the appropriate @samp{qSupported} feature indicating
39365 @item O @var{XX}@dots{}
39366 @samp{@var{XX}@dots{}} is hex encoding of @sc{ascii} data, to be
39367 written as the program's console output. This can happen at any time
39368 while the program is running and the debugger should continue to wait
39369 for @samp{W}, @samp{T}, etc. This reply is not permitted in non-stop mode.
39371 @item F @var{call-id},@var{parameter}@dots{}
39372 @var{call-id} is the identifier which says which host system call should
39373 be called. This is just the name of the function. Translation into the
39374 correct system call is only applicable as it's defined in @value{GDBN}.
39375 @xref{File-I/O Remote Protocol Extension}, for a list of implemented
39378 @samp{@var{parameter}@dots{}} is a list of parameters as defined for
39379 this very system call.
39381 The target replies with this packet when it expects @value{GDBN} to
39382 call a host system call on behalf of the target. @value{GDBN} replies
39383 with an appropriate @samp{F} packet and keeps up waiting for the next
39384 reply packet from the target. The latest @samp{C}, @samp{c}, @samp{S}
39385 or @samp{s} action is expected to be continued. @xref{File-I/O Remote
39386 Protocol Extension}, for more details.
39390 @node General Query Packets
39391 @section General Query Packets
39392 @cindex remote query requests
39394 Packets starting with @samp{q} are @dfn{general query packets};
39395 packets starting with @samp{Q} are @dfn{general set packets}. General
39396 query and set packets are a semi-unified form for retrieving and
39397 sending information to and from the stub.
39399 The initial letter of a query or set packet is followed by a name
39400 indicating what sort of thing the packet applies to. For example,
39401 @value{GDBN} may use a @samp{qSymbol} packet to exchange symbol
39402 definitions with the stub. These packet names follow some
39407 The name must not contain commas, colons or semicolons.
39409 Most @value{GDBN} query and set packets have a leading upper case
39412 The names of custom vendor packets should use a company prefix, in
39413 lower case, followed by a period. For example, packets designed at
39414 the Acme Corporation might begin with @samp{qacme.foo} (for querying
39415 foos) or @samp{Qacme.bar} (for setting bars).
39418 The name of a query or set packet should be separated from any
39419 parameters by a @samp{:}; the parameters themselves should be
39420 separated by @samp{,} or @samp{;}. Stubs must be careful to match the
39421 full packet name, and check for a separator or the end of the packet,
39422 in case two packet names share a common prefix. New packets should not begin
39423 with @samp{qC}, @samp{qP}, or @samp{qL}@footnote{The @samp{qP} and @samp{qL}
39424 packets predate these conventions, and have arguments without any terminator
39425 for the packet name; we suspect they are in widespread use in places that
39426 are difficult to upgrade. The @samp{qC} packet has no arguments, but some
39427 existing stubs (e.g.@: RedBoot) are known to not check for the end of the
39430 Like the descriptions of the other packets, each description here
39431 has a template showing the packet's overall syntax, followed by an
39432 explanation of the packet's meaning. We include spaces in some of the
39433 templates for clarity; these are not part of the packet's syntax. No
39434 @value{GDBN} packet uses spaces to separate its components.
39436 Here are the currently defined query and set packets:
39442 Turn on or off the agent as a helper to perform some debugging operations
39443 delegated from @value{GDBN} (@pxref{Control Agent}).
39445 @item QAllow:@var{op}:@var{val}@dots{}
39446 @cindex @samp{QAllow} packet
39447 Specify which operations @value{GDBN} expects to request of the
39448 target, as a semicolon-separated list of operation name and value
39449 pairs. Possible values for @var{op} include @samp{WriteReg},
39450 @samp{WriteMem}, @samp{InsertBreak}, @samp{InsertTrace},
39451 @samp{InsertFastTrace}, and @samp{Stop}. @var{val} is either 0,
39452 indicating that @value{GDBN} will not request the operation, or 1,
39453 indicating that it may. (The target can then use this to set up its
39454 own internals optimally, for instance if the debugger never expects to
39455 insert breakpoints, it may not need to install its own trap handler.)
39458 @cindex current thread, remote request
39459 @cindex @samp{qC} packet
39460 Return the current thread ID.
39464 @item QC @var{thread-id}
39465 Where @var{thread-id} is a thread ID as documented in
39466 @ref{thread-id syntax}.
39467 @item @r{(anything else)}
39468 Any other reply implies the old thread ID.
39471 @item qCRC:@var{addr},@var{length}
39472 @cindex CRC of memory block, remote request
39473 @cindex @samp{qCRC} packet
39474 @anchor{qCRC packet}
39475 Compute the CRC checksum of a block of memory using CRC-32 defined in
39476 IEEE 802.3. The CRC is computed byte at a time, taking the most
39477 significant bit of each byte first. The initial pattern code
39478 @code{0xffffffff} is used to ensure leading zeros affect the CRC.
39480 @emph{Note:} This is the same CRC used in validating separate debug
39481 files (@pxref{Separate Debug Files, , Debugging Information in Separate
39482 Files}). However the algorithm is slightly different. When validating
39483 separate debug files, the CRC is computed taking the @emph{least}
39484 significant bit of each byte first, and the final result is inverted to
39485 detect trailing zeros.
39490 An error (such as memory fault)
39491 @item C @var{crc32}
39492 The specified memory region's checksum is @var{crc32}.
39495 @item QDisableRandomization:@var{value}
39496 @cindex disable address space randomization, remote request
39497 @cindex @samp{QDisableRandomization} packet
39498 Some target operating systems will randomize the virtual address space
39499 of the inferior process as a security feature, but provide a feature
39500 to disable such randomization, e.g.@: to allow for a more deterministic
39501 debugging experience. On such systems, this packet with a @var{value}
39502 of 1 directs the target to disable address space randomization for
39503 processes subsequently started via @samp{vRun} packets, while a packet
39504 with a @var{value} of 0 tells the target to enable address space
39507 This packet is only available in extended mode (@pxref{extended mode}).
39512 The request succeeded.
39515 An error occurred. The error number @var{nn} is given as hex digits.
39518 An empty reply indicates that @samp{QDisableRandomization} is not supported
39522 This packet is not probed by default; the remote stub must request it,
39523 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
39524 This should only be done on targets that actually support disabling
39525 address space randomization.
39527 @item QStartupWithShell:@var{value}
39528 @cindex startup with shell, remote request
39529 @cindex @samp{QStartupWithShell} packet
39530 On UNIX-like targets, it is possible to start the inferior using a
39531 shell program. This is the default behavior on both @value{GDBN} and
39532 @command{gdbserver} (@pxref{set startup-with-shell}). This packet is
39533 used to inform @command{gdbserver} whether it should start the
39534 inferior using a shell or not.
39536 If @var{value} is @samp{0}, @command{gdbserver} will not use a shell
39537 to start the inferior. If @var{value} is @samp{1},
39538 @command{gdbserver} will use a shell to start the inferior. All other
39539 values are considered an error.
39541 This packet is only available in extended mode (@pxref{extended
39547 The request succeeded.
39550 An error occurred. The error number @var{nn} is given as hex digits.
39553 This packet is not probed by default; the remote stub must request it,
39554 by supplying an appropriate @samp{qSupported} response
39555 (@pxref{qSupported}). This should only be done on targets that
39556 actually support starting the inferior using a shell.
39558 Use of this packet is controlled by the @code{set startup-with-shell}
39559 command; @pxref{set startup-with-shell}.
39561 @item QEnvironmentHexEncoded:@var{hex-value}
39562 @anchor{QEnvironmentHexEncoded}
39563 @cindex set environment variable, remote request
39564 @cindex @samp{QEnvironmentHexEncoded} packet
39565 On UNIX-like targets, it is possible to set environment variables that
39566 will be passed to the inferior during the startup process. This
39567 packet is used to inform @command{gdbserver} of an environment
39568 variable that has been defined by the user on @value{GDBN} (@pxref{set
39571 The packet is composed by @var{hex-value}, an hex encoded
39572 representation of the @var{name=value} format representing an
39573 environment variable. The name of the environment variable is
39574 represented by @var{name}, and the value to be assigned to the
39575 environment variable is represented by @var{value}. If the variable
39576 has no value (i.e., the value is @code{null}), then @var{value} will
39579 This packet is only available in extended mode (@pxref{extended
39585 The request succeeded.
39588 This packet is not probed by default; the remote stub must request it,
39589 by supplying an appropriate @samp{qSupported} response
39590 (@pxref{qSupported}). This should only be done on targets that
39591 actually support passing environment variables to the starting
39594 This packet is related to the @code{set environment} command;
39595 @pxref{set environment}.
39597 @item QEnvironmentUnset:@var{hex-value}
39598 @anchor{QEnvironmentUnset}
39599 @cindex unset environment variable, remote request
39600 @cindex @samp{QEnvironmentUnset} packet
39601 On UNIX-like targets, it is possible to unset environment variables
39602 before starting the inferior in the remote target. This packet is
39603 used to inform @command{gdbserver} of an environment variable that has
39604 been unset by the user on @value{GDBN} (@pxref{unset environment}).
39606 The packet is composed by @var{hex-value}, an hex encoded
39607 representation of the name of the environment variable to be unset.
39609 This packet is only available in extended mode (@pxref{extended
39615 The request succeeded.
39618 This packet is not probed by default; the remote stub must request it,
39619 by supplying an appropriate @samp{qSupported} response
39620 (@pxref{qSupported}). This should only be done on targets that
39621 actually support passing environment variables to the starting
39624 This packet is related to the @code{unset environment} command;
39625 @pxref{unset environment}.
39627 @item QEnvironmentReset
39628 @anchor{QEnvironmentReset}
39629 @cindex reset environment, remote request
39630 @cindex @samp{QEnvironmentReset} packet
39631 On UNIX-like targets, this packet is used to reset the state of
39632 environment variables in the remote target before starting the
39633 inferior. In this context, reset means unsetting all environment
39634 variables that were previously set by the user (i.e., were not
39635 initially present in the environment). It is sent to
39636 @command{gdbserver} before the @samp{QEnvironmentHexEncoded}
39637 (@pxref{QEnvironmentHexEncoded}) and the @samp{QEnvironmentUnset}
39638 (@pxref{QEnvironmentUnset}) packets.
39640 This packet is only available in extended mode (@pxref{extended
39646 The request succeeded.
39649 This packet is not probed by default; the remote stub must request it,
39650 by supplying an appropriate @samp{qSupported} response
39651 (@pxref{qSupported}). This should only be done on targets that
39652 actually support passing environment variables to the starting
39655 @item QSetWorkingDir:@r{[}@var{directory}@r{]}
39656 @anchor{QSetWorkingDir packet}
39657 @cindex set working directory, remote request
39658 @cindex @samp{QSetWorkingDir} packet
39659 This packet is used to inform the remote server of the intended
39660 current working directory for programs that are going to be executed.
39662 The packet is composed by @var{directory}, an hex encoded
39663 representation of the directory that the remote inferior will use as
39664 its current working directory. If @var{directory} is an empty string,
39665 the remote server should reset the inferior's current working
39666 directory to its original, empty value.
39668 This packet is only available in extended mode (@pxref{extended
39674 The request succeeded.
39678 @itemx qsThreadInfo
39679 @cindex list active threads, remote request
39680 @cindex @samp{qfThreadInfo} packet
39681 @cindex @samp{qsThreadInfo} packet
39682 Obtain a list of all active thread IDs from the target (OS). Since there
39683 may be too many active threads to fit into one reply packet, this query
39684 works iteratively: it may require more than one query/reply sequence to
39685 obtain the entire list of threads. The first query of the sequence will
39686 be the @samp{qfThreadInfo} query; subsequent queries in the
39687 sequence will be the @samp{qsThreadInfo} query.
39689 NOTE: This packet replaces the @samp{qL} query (see below).
39693 @item m @var{thread-id}
39695 @item m @var{thread-id},@var{thread-id}@dots{}
39696 a comma-separated list of thread IDs
39698 (lower case letter @samp{L}) denotes end of list.
39701 In response to each query, the target will reply with a list of one or
39702 more thread IDs, separated by commas.
39703 @value{GDBN} will respond to each reply with a request for more thread
39704 ids (using the @samp{qs} form of the query), until the target responds
39705 with @samp{l} (lower-case ell, for @dfn{last}).
39706 Refer to @ref{thread-id syntax}, for the format of the @var{thread-id}
39709 @emph{Note: @value{GDBN} will send the @code{qfThreadInfo} query during the
39710 initial connection with the remote target, and the very first thread ID
39711 mentioned in the reply will be stopped by @value{GDBN} in a subsequent
39712 message. Therefore, the stub should ensure that the first thread ID in
39713 the @code{qfThreadInfo} reply is suitable for being stopped by @value{GDBN}.}
39715 @item qGetTLSAddr:@var{thread-id},@var{offset},@var{lm}
39716 @cindex get thread-local storage address, remote request
39717 @cindex @samp{qGetTLSAddr} packet
39718 Fetch the address associated with thread local storage specified
39719 by @var{thread-id}, @var{offset}, and @var{lm}.
39721 @var{thread-id} is the thread ID associated with the
39722 thread for which to fetch the TLS address. @xref{thread-id syntax}.
39724 @var{offset} is the (big endian, hex encoded) offset associated with the
39725 thread local variable. (This offset is obtained from the debug
39726 information associated with the variable.)
39728 @var{lm} is the (big endian, hex encoded) OS/ABI-specific encoding of the
39729 load module associated with the thread local storage. For example,
39730 a @sc{gnu}/Linux system will pass the link map address of the shared
39731 object associated with the thread local storage under consideration.
39732 Other operating environments may choose to represent the load module
39733 differently, so the precise meaning of this parameter will vary.
39737 @item @var{XX}@dots{}
39738 Hex encoded (big endian) bytes representing the address of the thread
39739 local storage requested.
39742 An error occurred. The error number @var{nn} is given as hex digits.
39745 An empty reply indicates that @samp{qGetTLSAddr} is not supported by the stub.
39748 @item qGetTIBAddr:@var{thread-id}
39749 @cindex get thread information block address
39750 @cindex @samp{qGetTIBAddr} packet
39751 Fetch address of the Windows OS specific Thread Information Block.
39753 @var{thread-id} is the thread ID associated with the thread.
39757 @item @var{XX}@dots{}
39758 Hex encoded (big endian) bytes representing the linear address of the
39759 thread information block.
39762 An error occured. This means that either the thread was not found, or the
39763 address could not be retrieved.
39766 An empty reply indicates that @samp{qGetTIBAddr} is not supported by the stub.
39769 @item qL @var{startflag} @var{threadcount} @var{nextthread}
39770 Obtain thread information from RTOS. Where: @var{startflag} (one hex
39771 digit) is one to indicate the first query and zero to indicate a
39772 subsequent query; @var{threadcount} (two hex digits) is the maximum
39773 number of threads the response packet can contain; and @var{nextthread}
39774 (eight hex digits), for subsequent queries (@var{startflag} is zero), is
39775 returned in the response as @var{argthread}.
39777 Don't use this packet; use the @samp{qfThreadInfo} query instead (see above).
39781 @item qM @var{count} @var{done} @var{argthread} @var{thread}@dots{}
39782 Where: @var{count} (two hex digits) is the number of threads being
39783 returned; @var{done} (one hex digit) is zero to indicate more threads
39784 and one indicates no further threads; @var{argthreadid} (eight hex
39785 digits) is @var{nextthread} from the request packet; @var{thread}@dots{}
39786 is a sequence of thread IDs, @var{threadid} (eight hex
39787 digits), from the target. See @code{remote.c:parse_threadlist_response()}.
39791 @cindex section offsets, remote request
39792 @cindex @samp{qOffsets} packet
39793 Get section offsets that the target used when relocating the downloaded
39798 @item Text=@var{xxx};Data=@var{yyy}@r{[};Bss=@var{zzz}@r{]}
39799 Relocate the @code{Text} section by @var{xxx} from its original address.
39800 Relocate the @code{Data} section by @var{yyy} from its original address.
39801 If the object file format provides segment information (e.g.@: @sc{elf}
39802 @samp{PT_LOAD} program headers), @value{GDBN} will relocate entire
39803 segments by the supplied offsets.
39805 @emph{Note: while a @code{Bss} offset may be included in the response,
39806 @value{GDBN} ignores this and instead applies the @code{Data} offset
39807 to the @code{Bss} section.}
39809 @item TextSeg=@var{xxx}@r{[};DataSeg=@var{yyy}@r{]}
39810 Relocate the first segment of the object file, which conventionally
39811 contains program code, to a starting address of @var{xxx}. If
39812 @samp{DataSeg} is specified, relocate the second segment, which
39813 conventionally contains modifiable data, to a starting address of
39814 @var{yyy}. @value{GDBN} will report an error if the object file
39815 does not contain segment information, or does not contain at least
39816 as many segments as mentioned in the reply. Extra segments are
39817 kept at fixed offsets relative to the last relocated segment.
39820 @item qP @var{mode} @var{thread-id}
39821 @cindex thread information, remote request
39822 @cindex @samp{qP} packet
39823 Returns information on @var{thread-id}. Where: @var{mode} is a hex
39824 encoded 32 bit mode; @var{thread-id} is a thread ID
39825 (@pxref{thread-id syntax}).
39827 Don't use this packet; use the @samp{qThreadExtraInfo} query instead
39830 Reply: see @code{remote.c:remote_unpack_thread_info_response()}.
39834 @cindex non-stop mode, remote request
39835 @cindex @samp{QNonStop} packet
39837 Enter non-stop (@samp{QNonStop:1}) or all-stop (@samp{QNonStop:0}) mode.
39838 @xref{Remote Non-Stop}, for more information.
39843 The request succeeded.
39846 An error occurred. The error number @var{nn} is given as hex digits.
39849 An empty reply indicates that @samp{QNonStop} is not supported by
39853 This packet is not probed by default; the remote stub must request it,
39854 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
39855 Use of this packet is controlled by the @code{set non-stop} command;
39856 @pxref{Non-Stop Mode}.
39858 @item QCatchSyscalls:1 @r{[};@var{sysno}@r{]}@dots{}
39859 @itemx QCatchSyscalls:0
39860 @cindex catch syscalls from inferior, remote request
39861 @cindex @samp{QCatchSyscalls} packet
39862 @anchor{QCatchSyscalls}
39863 Enable (@samp{QCatchSyscalls:1}) or disable (@samp{QCatchSyscalls:0})
39864 catching syscalls from the inferior process.
39866 For @samp{QCatchSyscalls:1}, each listed syscall @var{sysno} (encoded
39867 in hex) should be reported to @value{GDBN}. If no syscall @var{sysno}
39868 is listed, every system call should be reported.
39870 Note that if a syscall not in the list is reported, @value{GDBN} will
39871 still filter the event according to its own list from all corresponding
39872 @code{catch syscall} commands. However, it is more efficient to only
39873 report the requested syscalls.
39875 Multiple @samp{QCatchSyscalls:1} packets do not combine; any earlier
39876 @samp{QCatchSyscalls:1} list is completely replaced by the new list.
39878 If the inferior process execs, the state of @samp{QCatchSyscalls} is
39879 kept for the new process too. On targets where exec may affect syscall
39880 numbers, for example with exec between 32 and 64-bit processes, the
39881 client should send a new packet with the new syscall list.
39886 The request succeeded.
39889 An error occurred. @var{nn} are hex digits.
39892 An empty reply indicates that @samp{QCatchSyscalls} is not supported by
39896 Use of this packet is controlled by the @code{set remote catch-syscalls}
39897 command (@pxref{Remote Configuration, set remote catch-syscalls}).
39898 This packet is not probed by default; the remote stub must request it,
39899 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
39901 @item QPassSignals: @var{signal} @r{[};@var{signal}@r{]}@dots{}
39902 @cindex pass signals to inferior, remote request
39903 @cindex @samp{QPassSignals} packet
39904 @anchor{QPassSignals}
39905 Each listed @var{signal} should be passed directly to the inferior process.
39906 Signals are numbered identically to continue packets and stop replies
39907 (@pxref{Stop Reply Packets}). Each @var{signal} list item should be
39908 strictly greater than the previous item. These signals do not need to stop
39909 the inferior, or be reported to @value{GDBN}. All other signals should be
39910 reported to @value{GDBN}. Multiple @samp{QPassSignals} packets do not
39911 combine; any earlier @samp{QPassSignals} list is completely replaced by the
39912 new list. This packet improves performance when using @samp{handle
39913 @var{signal} nostop noprint pass}.
39918 The request succeeded.
39921 An error occurred. The error number @var{nn} is given as hex digits.
39924 An empty reply indicates that @samp{QPassSignals} is not supported by
39928 Use of this packet is controlled by the @code{set remote pass-signals}
39929 command (@pxref{Remote Configuration, set remote pass-signals}).
39930 This packet is not probed by default; the remote stub must request it,
39931 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
39933 @item QProgramSignals: @var{signal} @r{[};@var{signal}@r{]}@dots{}
39934 @cindex signals the inferior may see, remote request
39935 @cindex @samp{QProgramSignals} packet
39936 @anchor{QProgramSignals}
39937 Each listed @var{signal} may be delivered to the inferior process.
39938 Others should be silently discarded.
39940 In some cases, the remote stub may need to decide whether to deliver a
39941 signal to the program or not without @value{GDBN} involvement. One
39942 example of that is while detaching --- the program's threads may have
39943 stopped for signals that haven't yet had a chance of being reported to
39944 @value{GDBN}, and so the remote stub can use the signal list specified
39945 by this packet to know whether to deliver or ignore those pending
39948 This does not influence whether to deliver a signal as requested by a
39949 resumption packet (@pxref{vCont packet}).
39951 Signals are numbered identically to continue packets and stop replies
39952 (@pxref{Stop Reply Packets}). Each @var{signal} list item should be
39953 strictly greater than the previous item. Multiple
39954 @samp{QProgramSignals} packets do not combine; any earlier
39955 @samp{QProgramSignals} list is completely replaced by the new list.
39960 The request succeeded.
39963 An error occurred. The error number @var{nn} is given as hex digits.
39966 An empty reply indicates that @samp{QProgramSignals} is not supported
39970 Use of this packet is controlled by the @code{set remote program-signals}
39971 command (@pxref{Remote Configuration, set remote program-signals}).
39972 This packet is not probed by default; the remote stub must request it,
39973 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
39975 @anchor{QThreadEvents}
39976 @item QThreadEvents:1
39977 @itemx QThreadEvents:0
39978 @cindex thread create/exit events, remote request
39979 @cindex @samp{QThreadEvents} packet
39981 Enable (@samp{QThreadEvents:1}) or disable (@samp{QThreadEvents:0})
39982 reporting of thread create and exit events. @xref{thread create
39983 event}, for the reply specifications. For example, this is used in
39984 non-stop mode when @value{GDBN} stops a set of threads and
39985 synchronously waits for the their corresponding stop replies. Without
39986 exit events, if one of the threads exits, @value{GDBN} would hang
39987 forever not knowing that it should no longer expect a stop for that
39988 same thread. @value{GDBN} does not enable this feature unless the
39989 stub reports that it supports it by including @samp{QThreadEvents+} in
39990 its @samp{qSupported} reply.
39995 The request succeeded.
39998 An error occurred. The error number @var{nn} is given as hex digits.
40001 An empty reply indicates that @samp{QThreadEvents} is not supported by
40005 Use of this packet is controlled by the @code{set remote thread-events}
40006 command (@pxref{Remote Configuration, set remote thread-events}).
40008 @item qRcmd,@var{command}
40009 @cindex execute remote command, remote request
40010 @cindex @samp{qRcmd} packet
40011 @var{command} (hex encoded) is passed to the local interpreter for
40012 execution. Invalid commands should be reported using the output
40013 string. Before the final result packet, the target may also respond
40014 with a number of intermediate @samp{O@var{output}} console output
40015 packets. @emph{Implementors should note that providing access to a
40016 stubs's interpreter may have security implications}.
40021 A command response with no output.
40023 A command response with the hex encoded output string @var{OUTPUT}.
40025 Indicate a badly formed request.
40027 An empty reply indicates that @samp{qRcmd} is not recognized.
40030 (Note that the @code{qRcmd} packet's name is separated from the
40031 command by a @samp{,}, not a @samp{:}, contrary to the naming
40032 conventions above. Please don't use this packet as a model for new
40035 @item qSearch:memory:@var{address};@var{length};@var{search-pattern}
40036 @cindex searching memory, in remote debugging
40038 @cindex @samp{qSearch:memory} packet
40040 @cindex @samp{qSearch memory} packet
40041 @anchor{qSearch memory}
40042 Search @var{length} bytes at @var{address} for @var{search-pattern}.
40043 Both @var{address} and @var{length} are encoded in hex;
40044 @var{search-pattern} is a sequence of bytes, also hex encoded.
40049 The pattern was not found.
40051 The pattern was found at @var{address}.
40053 A badly formed request or an error was encountered while searching memory.
40055 An empty reply indicates that @samp{qSearch:memory} is not recognized.
40058 @item QStartNoAckMode
40059 @cindex @samp{QStartNoAckMode} packet
40060 @anchor{QStartNoAckMode}
40061 Request that the remote stub disable the normal @samp{+}/@samp{-}
40062 protocol acknowledgments (@pxref{Packet Acknowledgment}).
40067 The stub has switched to no-acknowledgment mode.
40068 @value{GDBN} acknowledges this response,
40069 but neither the stub nor @value{GDBN} shall send or expect further
40070 @samp{+}/@samp{-} acknowledgments in the current connection.
40072 An empty reply indicates that the stub does not support no-acknowledgment mode.
40075 @item qSupported @r{[}:@var{gdbfeature} @r{[};@var{gdbfeature}@r{]}@dots{} @r{]}
40076 @cindex supported packets, remote query
40077 @cindex features of the remote protocol
40078 @cindex @samp{qSupported} packet
40079 @anchor{qSupported}
40080 Tell the remote stub about features supported by @value{GDBN}, and
40081 query the stub for features it supports. This packet allows
40082 @value{GDBN} and the remote stub to take advantage of each others'
40083 features. @samp{qSupported} also consolidates multiple feature probes
40084 at startup, to improve @value{GDBN} performance---a single larger
40085 packet performs better than multiple smaller probe packets on
40086 high-latency links. Some features may enable behavior which must not
40087 be on by default, e.g.@: because it would confuse older clients or
40088 stubs. Other features may describe packets which could be
40089 automatically probed for, but are not. These features must be
40090 reported before @value{GDBN} will use them. This ``default
40091 unsupported'' behavior is not appropriate for all packets, but it
40092 helps to keep the initial connection time under control with new
40093 versions of @value{GDBN} which support increasing numbers of packets.
40097 @item @var{stubfeature} @r{[};@var{stubfeature}@r{]}@dots{}
40098 The stub supports or does not support each returned @var{stubfeature},
40099 depending on the form of each @var{stubfeature} (see below for the
40102 An empty reply indicates that @samp{qSupported} is not recognized,
40103 or that no features needed to be reported to @value{GDBN}.
40106 The allowed forms for each feature (either a @var{gdbfeature} in the
40107 @samp{qSupported} packet, or a @var{stubfeature} in the response)
40111 @item @var{name}=@var{value}
40112 The remote protocol feature @var{name} is supported, and associated
40113 with the specified @var{value}. The format of @var{value} depends
40114 on the feature, but it must not include a semicolon.
40116 The remote protocol feature @var{name} is supported, and does not
40117 need an associated value.
40119 The remote protocol feature @var{name} is not supported.
40121 The remote protocol feature @var{name} may be supported, and
40122 @value{GDBN} should auto-detect support in some other way when it is
40123 needed. This form will not be used for @var{gdbfeature} notifications,
40124 but may be used for @var{stubfeature} responses.
40127 Whenever the stub receives a @samp{qSupported} request, the
40128 supplied set of @value{GDBN} features should override any previous
40129 request. This allows @value{GDBN} to put the stub in a known
40130 state, even if the stub had previously been communicating with
40131 a different version of @value{GDBN}.
40133 The following values of @var{gdbfeature} (for the packet sent by @value{GDBN})
40138 This feature indicates whether @value{GDBN} supports multiprocess
40139 extensions to the remote protocol. @value{GDBN} does not use such
40140 extensions unless the stub also reports that it supports them by
40141 including @samp{multiprocess+} in its @samp{qSupported} reply.
40142 @xref{multiprocess extensions}, for details.
40145 This feature indicates that @value{GDBN} supports the XML target
40146 description. If the stub sees @samp{xmlRegisters=} with target
40147 specific strings separated by a comma, it will report register
40151 This feature indicates whether @value{GDBN} supports the
40152 @samp{qRelocInsn} packet (@pxref{Tracepoint Packets,,Relocate
40153 instruction reply packet}).
40156 This feature indicates whether @value{GDBN} supports the swbreak stop
40157 reason in stop replies. @xref{swbreak stop reason}, for details.
40160 This feature indicates whether @value{GDBN} supports the hwbreak stop
40161 reason in stop replies. @xref{swbreak stop reason}, for details.
40164 This feature indicates whether @value{GDBN} supports fork event
40165 extensions to the remote protocol. @value{GDBN} does not use such
40166 extensions unless the stub also reports that it supports them by
40167 including @samp{fork-events+} in its @samp{qSupported} reply.
40170 This feature indicates whether @value{GDBN} supports vfork event
40171 extensions to the remote protocol. @value{GDBN} does not use such
40172 extensions unless the stub also reports that it supports them by
40173 including @samp{vfork-events+} in its @samp{qSupported} reply.
40176 This feature indicates whether @value{GDBN} supports exec event
40177 extensions to the remote protocol. @value{GDBN} does not use such
40178 extensions unless the stub also reports that it supports them by
40179 including @samp{exec-events+} in its @samp{qSupported} reply.
40181 @item vContSupported
40182 This feature indicates whether @value{GDBN} wants to know the
40183 supported actions in the reply to @samp{vCont?} packet.
40186 Stubs should ignore any unknown values for
40187 @var{gdbfeature}. Any @value{GDBN} which sends a @samp{qSupported}
40188 packet supports receiving packets of unlimited length (earlier
40189 versions of @value{GDBN} may reject overly long responses). Additional values
40190 for @var{gdbfeature} may be defined in the future to let the stub take
40191 advantage of new features in @value{GDBN}, e.g.@: incompatible
40192 improvements in the remote protocol---the @samp{multiprocess} feature is
40193 an example of such a feature. The stub's reply should be independent
40194 of the @var{gdbfeature} entries sent by @value{GDBN}; first @value{GDBN}
40195 describes all the features it supports, and then the stub replies with
40196 all the features it supports.
40198 Similarly, @value{GDBN} will silently ignore unrecognized stub feature
40199 responses, as long as each response uses one of the standard forms.
40201 Some features are flags. A stub which supports a flag feature
40202 should respond with a @samp{+} form response. Other features
40203 require values, and the stub should respond with an @samp{=}
40206 Each feature has a default value, which @value{GDBN} will use if
40207 @samp{qSupported} is not available or if the feature is not mentioned
40208 in the @samp{qSupported} response. The default values are fixed; a
40209 stub is free to omit any feature responses that match the defaults.
40211 Not all features can be probed, but for those which can, the probing
40212 mechanism is useful: in some cases, a stub's internal
40213 architecture may not allow the protocol layer to know some information
40214 about the underlying target in advance. This is especially common in
40215 stubs which may be configured for multiple targets.
40217 These are the currently defined stub features and their properties:
40219 @multitable @columnfractions 0.35 0.2 0.12 0.2
40220 @c NOTE: The first row should be @headitem, but we do not yet require
40221 @c a new enough version of Texinfo (4.7) to use @headitem.
40223 @tab Value Required
40227 @item @samp{PacketSize}
40232 @item @samp{qXfer:auxv:read}
40237 @item @samp{qXfer:btrace:read}
40242 @item @samp{qXfer:btrace-conf:read}
40247 @item @samp{qXfer:exec-file:read}
40252 @item @samp{qXfer:features:read}
40257 @item @samp{qXfer:libraries:read}
40262 @item @samp{qXfer:libraries-svr4:read}
40267 @item @samp{augmented-libraries-svr4-read}
40272 @item @samp{qXfer:memory-map:read}
40277 @item @samp{qXfer:sdata:read}
40282 @item @samp{qXfer:siginfo:read}
40287 @item @samp{qXfer:siginfo:write}
40292 @item @samp{qXfer:threads:read}
40297 @item @samp{qXfer:traceframe-info:read}
40302 @item @samp{qXfer:uib:read}
40307 @item @samp{qXfer:fdpic:read}
40312 @item @samp{Qbtrace:off}
40317 @item @samp{Qbtrace:bts}
40322 @item @samp{Qbtrace:pt}
40327 @item @samp{Qbtrace-conf:bts:size}
40332 @item @samp{Qbtrace-conf:pt:size}
40337 @item @samp{QNonStop}
40342 @item @samp{QCatchSyscalls}
40347 @item @samp{QPassSignals}
40352 @item @samp{QStartNoAckMode}
40357 @item @samp{multiprocess}
40362 @item @samp{ConditionalBreakpoints}
40367 @item @samp{ConditionalTracepoints}
40372 @item @samp{ReverseContinue}
40377 @item @samp{ReverseStep}
40382 @item @samp{TracepointSource}
40387 @item @samp{QAgent}
40392 @item @samp{QAllow}
40397 @item @samp{QDisableRandomization}
40402 @item @samp{EnableDisableTracepoints}
40407 @item @samp{QTBuffer:size}
40412 @item @samp{tracenz}
40417 @item @samp{BreakpointCommands}
40422 @item @samp{swbreak}
40427 @item @samp{hwbreak}
40432 @item @samp{fork-events}
40437 @item @samp{vfork-events}
40442 @item @samp{exec-events}
40447 @item @samp{QThreadEvents}
40452 @item @samp{no-resumed}
40459 These are the currently defined stub features, in more detail:
40462 @cindex packet size, remote protocol
40463 @item PacketSize=@var{bytes}
40464 The remote stub can accept packets up to at least @var{bytes} in
40465 length. @value{GDBN} will send packets up to this size for bulk
40466 transfers, and will never send larger packets. This is a limit on the
40467 data characters in the packet, including the frame and checksum.
40468 There is no trailing NUL byte in a remote protocol packet; if the stub
40469 stores packets in a NUL-terminated format, it should allow an extra
40470 byte in its buffer for the NUL. If this stub feature is not supported,
40471 @value{GDBN} guesses based on the size of the @samp{g} packet response.
40473 @item qXfer:auxv:read
40474 The remote stub understands the @samp{qXfer:auxv:read} packet
40475 (@pxref{qXfer auxiliary vector read}).
40477 @item qXfer:btrace:read
40478 The remote stub understands the @samp{qXfer:btrace:read}
40479 packet (@pxref{qXfer btrace read}).
40481 @item qXfer:btrace-conf:read
40482 The remote stub understands the @samp{qXfer:btrace-conf:read}
40483 packet (@pxref{qXfer btrace-conf read}).
40485 @item qXfer:exec-file:read
40486 The remote stub understands the @samp{qXfer:exec-file:read} packet
40487 (@pxref{qXfer executable filename read}).
40489 @item qXfer:features:read
40490 The remote stub understands the @samp{qXfer:features:read} packet
40491 (@pxref{qXfer target description read}).
40493 @item qXfer:libraries:read
40494 The remote stub understands the @samp{qXfer:libraries:read} packet
40495 (@pxref{qXfer library list read}).
40497 @item qXfer:libraries-svr4:read
40498 The remote stub understands the @samp{qXfer:libraries-svr4:read} packet
40499 (@pxref{qXfer svr4 library list read}).
40501 @item augmented-libraries-svr4-read
40502 The remote stub understands the augmented form of the
40503 @samp{qXfer:libraries-svr4:read} packet
40504 (@pxref{qXfer svr4 library list read}).
40506 @item qXfer:memory-map:read
40507 The remote stub understands the @samp{qXfer:memory-map:read} packet
40508 (@pxref{qXfer memory map read}).
40510 @item qXfer:sdata:read
40511 The remote stub understands the @samp{qXfer:sdata:read} packet
40512 (@pxref{qXfer sdata read}).
40514 @item qXfer:siginfo:read
40515 The remote stub understands the @samp{qXfer:siginfo:read} packet
40516 (@pxref{qXfer siginfo read}).
40518 @item qXfer:siginfo:write
40519 The remote stub understands the @samp{qXfer:siginfo:write} packet
40520 (@pxref{qXfer siginfo write}).
40522 @item qXfer:threads:read
40523 The remote stub understands the @samp{qXfer:threads:read} packet
40524 (@pxref{qXfer threads read}).
40526 @item qXfer:traceframe-info:read
40527 The remote stub understands the @samp{qXfer:traceframe-info:read}
40528 packet (@pxref{qXfer traceframe info read}).
40530 @item qXfer:uib:read
40531 The remote stub understands the @samp{qXfer:uib:read}
40532 packet (@pxref{qXfer unwind info block}).
40534 @item qXfer:fdpic:read
40535 The remote stub understands the @samp{qXfer:fdpic:read}
40536 packet (@pxref{qXfer fdpic loadmap read}).
40539 The remote stub understands the @samp{QNonStop} packet
40540 (@pxref{QNonStop}).
40542 @item QCatchSyscalls
40543 The remote stub understands the @samp{QCatchSyscalls} packet
40544 (@pxref{QCatchSyscalls}).
40547 The remote stub understands the @samp{QPassSignals} packet
40548 (@pxref{QPassSignals}).
40550 @item QStartNoAckMode
40551 The remote stub understands the @samp{QStartNoAckMode} packet and
40552 prefers to operate in no-acknowledgment mode. @xref{Packet Acknowledgment}.
40555 @anchor{multiprocess extensions}
40556 @cindex multiprocess extensions, in remote protocol
40557 The remote stub understands the multiprocess extensions to the remote
40558 protocol syntax. The multiprocess extensions affect the syntax of
40559 thread IDs in both packets and replies (@pxref{thread-id syntax}), and
40560 add process IDs to the @samp{D} packet and @samp{W} and @samp{X}
40561 replies. Note that reporting this feature indicates support for the
40562 syntactic extensions only, not that the stub necessarily supports
40563 debugging of more than one process at a time. The stub must not use
40564 multiprocess extensions in packet replies unless @value{GDBN} has also
40565 indicated it supports them in its @samp{qSupported} request.
40567 @item qXfer:osdata:read
40568 The remote stub understands the @samp{qXfer:osdata:read} packet
40569 ((@pxref{qXfer osdata read}).
40571 @item ConditionalBreakpoints
40572 The target accepts and implements evaluation of conditional expressions
40573 defined for breakpoints. The target will only report breakpoint triggers
40574 when such conditions are true (@pxref{Conditions, ,Break Conditions}).
40576 @item ConditionalTracepoints
40577 The remote stub accepts and implements conditional expressions defined
40578 for tracepoints (@pxref{Tracepoint Conditions}).
40580 @item ReverseContinue
40581 The remote stub accepts and implements the reverse continue packet
40585 The remote stub accepts and implements the reverse step packet
40588 @item TracepointSource
40589 The remote stub understands the @samp{QTDPsrc} packet that supplies
40590 the source form of tracepoint definitions.
40593 The remote stub understands the @samp{QAgent} packet.
40596 The remote stub understands the @samp{QAllow} packet.
40598 @item QDisableRandomization
40599 The remote stub understands the @samp{QDisableRandomization} packet.
40601 @item StaticTracepoint
40602 @cindex static tracepoints, in remote protocol
40603 The remote stub supports static tracepoints.
40605 @item InstallInTrace
40606 @anchor{install tracepoint in tracing}
40607 The remote stub supports installing tracepoint in tracing.
40609 @item EnableDisableTracepoints
40610 The remote stub supports the @samp{QTEnable} (@pxref{QTEnable}) and
40611 @samp{QTDisable} (@pxref{QTDisable}) packets that allow tracepoints
40612 to be enabled and disabled while a trace experiment is running.
40614 @item QTBuffer:size
40615 The remote stub supports the @samp{QTBuffer:size} (@pxref{QTBuffer-size})
40616 packet that allows to change the size of the trace buffer.
40619 @cindex string tracing, in remote protocol
40620 The remote stub supports the @samp{tracenz} bytecode for collecting strings.
40621 See @ref{Bytecode Descriptions} for details about the bytecode.
40623 @item BreakpointCommands
40624 @cindex breakpoint commands, in remote protocol
40625 The remote stub supports running a breakpoint's command list itself,
40626 rather than reporting the hit to @value{GDBN}.
40629 The remote stub understands the @samp{Qbtrace:off} packet.
40632 The remote stub understands the @samp{Qbtrace:bts} packet.
40635 The remote stub understands the @samp{Qbtrace:pt} packet.
40637 @item Qbtrace-conf:bts:size
40638 The remote stub understands the @samp{Qbtrace-conf:bts:size} packet.
40640 @item Qbtrace-conf:pt:size
40641 The remote stub understands the @samp{Qbtrace-conf:pt:size} packet.
40644 The remote stub reports the @samp{swbreak} stop reason for memory
40648 The remote stub reports the @samp{hwbreak} stop reason for hardware
40652 The remote stub reports the @samp{fork} stop reason for fork events.
40655 The remote stub reports the @samp{vfork} stop reason for vfork events
40656 and vforkdone events.
40659 The remote stub reports the @samp{exec} stop reason for exec events.
40661 @item vContSupported
40662 The remote stub reports the supported actions in the reply to
40663 @samp{vCont?} packet.
40665 @item QThreadEvents
40666 The remote stub understands the @samp{QThreadEvents} packet.
40669 The remote stub reports the @samp{N} stop reply.
40674 @cindex symbol lookup, remote request
40675 @cindex @samp{qSymbol} packet
40676 Notify the target that @value{GDBN} is prepared to serve symbol lookup
40677 requests. Accept requests from the target for the values of symbols.
40682 The target does not need to look up any (more) symbols.
40683 @item qSymbol:@var{sym_name}
40684 The target requests the value of symbol @var{sym_name} (hex encoded).
40685 @value{GDBN} may provide the value by using the
40686 @samp{qSymbol:@var{sym_value}:@var{sym_name}} message, described
40690 @item qSymbol:@var{sym_value}:@var{sym_name}
40691 Set the value of @var{sym_name} to @var{sym_value}.
40693 @var{sym_name} (hex encoded) is the name of a symbol whose value the
40694 target has previously requested.
40696 @var{sym_value} (hex) is the value for symbol @var{sym_name}. If
40697 @value{GDBN} cannot supply a value for @var{sym_name}, then this field
40703 The target does not need to look up any (more) symbols.
40704 @item qSymbol:@var{sym_name}
40705 The target requests the value of a new symbol @var{sym_name} (hex
40706 encoded). @value{GDBN} will continue to supply the values of symbols
40707 (if available), until the target ceases to request them.
40712 @itemx QTDisconnected
40719 @itemx qTMinFTPILen
40721 @xref{Tracepoint Packets}.
40723 @item qThreadExtraInfo,@var{thread-id}
40724 @cindex thread attributes info, remote request
40725 @cindex @samp{qThreadExtraInfo} packet
40726 Obtain from the target OS a printable string description of thread
40727 attributes for the thread @var{thread-id}; see @ref{thread-id syntax},
40728 for the forms of @var{thread-id}. This
40729 string may contain anything that the target OS thinks is interesting
40730 for @value{GDBN} to tell the user about the thread. The string is
40731 displayed in @value{GDBN}'s @code{info threads} display. Some
40732 examples of possible thread extra info strings are @samp{Runnable}, or
40733 @samp{Blocked on Mutex}.
40737 @item @var{XX}@dots{}
40738 Where @samp{@var{XX}@dots{}} is a hex encoding of @sc{ascii} data,
40739 comprising the printable string containing the extra information about
40740 the thread's attributes.
40743 (Note that the @code{qThreadExtraInfo} packet's name is separated from
40744 the command by a @samp{,}, not a @samp{:}, contrary to the naming
40745 conventions above. Please don't use this packet as a model for new
40764 @xref{Tracepoint Packets}.
40766 @item qXfer:@var{object}:read:@var{annex}:@var{offset},@var{length}
40767 @cindex read special object, remote request
40768 @cindex @samp{qXfer} packet
40769 @anchor{qXfer read}
40770 Read uninterpreted bytes from the target's special data area
40771 identified by the keyword @var{object}. Request @var{length} bytes
40772 starting at @var{offset} bytes into the data. The content and
40773 encoding of @var{annex} is specific to @var{object}; it can supply
40774 additional details about what data to access.
40779 Data @var{data} (@pxref{Binary Data}) has been read from the
40780 target. There may be more data at a higher address (although
40781 it is permitted to return @samp{m} even for the last valid
40782 block of data, as long as at least one byte of data was read).
40783 It is possible for @var{data} to have fewer bytes than the @var{length} in the
40787 Data @var{data} (@pxref{Binary Data}) has been read from the target.
40788 There is no more data to be read. It is possible for @var{data} to
40789 have fewer bytes than the @var{length} in the request.
40792 The @var{offset} in the request is at the end of the data.
40793 There is no more data to be read.
40796 The request was malformed, or @var{annex} was invalid.
40799 The offset was invalid, or there was an error encountered reading the data.
40800 The @var{nn} part is a hex-encoded @code{errno} value.
40803 An empty reply indicates the @var{object} string was not recognized by
40804 the stub, or that the object does not support reading.
40807 Here are the specific requests of this form defined so far. All the
40808 @samp{qXfer:@var{object}:read:@dots{}} requests use the same reply
40809 formats, listed above.
40812 @item qXfer:auxv:read::@var{offset},@var{length}
40813 @anchor{qXfer auxiliary vector read}
40814 Access the target's @dfn{auxiliary vector}. @xref{OS Information,
40815 auxiliary vector}. Note @var{annex} must be empty.
40817 This packet is not probed by default; the remote stub must request it,
40818 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40820 @item qXfer:btrace:read:@var{annex}:@var{offset},@var{length}
40821 @anchor{qXfer btrace read}
40823 Return a description of the current branch trace.
40824 @xref{Branch Trace Format}. The annex part of the generic @samp{qXfer}
40825 packet may have one of the following values:
40829 Returns all available branch trace.
40832 Returns all available branch trace if the branch trace changed since
40833 the last read request.
40836 Returns the new branch trace since the last read request. Adds a new
40837 block to the end of the trace that begins at zero and ends at the source
40838 location of the first branch in the trace buffer. This extra block is
40839 used to stitch traces together.
40841 If the trace buffer overflowed, returns an error indicating the overflow.
40844 This packet is not probed by default; the remote stub must request it
40845 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40847 @item qXfer:btrace-conf:read::@var{offset},@var{length}
40848 @anchor{qXfer btrace-conf read}
40850 Return a description of the current branch trace configuration.
40851 @xref{Branch Trace Configuration Format}.
40853 This packet is not probed by default; the remote stub must request it
40854 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40856 @item qXfer:exec-file:read:@var{annex}:@var{offset},@var{length}
40857 @anchor{qXfer executable filename read}
40858 Return the full absolute name of the file that was executed to create
40859 a process running on the remote system. The annex specifies the
40860 numeric process ID of the process to query, encoded as a hexadecimal
40861 number. If the annex part is empty the remote stub should return the
40862 filename corresponding to the currently executing process.
40864 This packet is not probed by default; the remote stub must request it,
40865 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40867 @item qXfer:features:read:@var{annex}:@var{offset},@var{length}
40868 @anchor{qXfer target description read}
40869 Access the @dfn{target description}. @xref{Target Descriptions}. The
40870 annex specifies which XML document to access. The main description is
40871 always loaded from the @samp{target.xml} annex.
40873 This packet is not probed by default; the remote stub must request it,
40874 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40876 @item qXfer:libraries:read:@var{annex}:@var{offset},@var{length}
40877 @anchor{qXfer library list read}
40878 Access the target's list of loaded libraries. @xref{Library List Format}.
40879 The annex part of the generic @samp{qXfer} packet must be empty
40880 (@pxref{qXfer read}).
40882 Targets which maintain a list of libraries in the program's memory do
40883 not need to implement this packet; it is designed for platforms where
40884 the operating system manages the list of loaded libraries.
40886 This packet is not probed by default; the remote stub must request it,
40887 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40889 @item qXfer:libraries-svr4:read:@var{annex}:@var{offset},@var{length}
40890 @anchor{qXfer svr4 library list read}
40891 Access the target's list of loaded libraries when the target is an SVR4
40892 platform. @xref{Library List Format for SVR4 Targets}. The annex part
40893 of the generic @samp{qXfer} packet must be empty unless the remote
40894 stub indicated it supports the augmented form of this packet
40895 by supplying an appropriate @samp{qSupported} response
40896 (@pxref{qXfer read}, @ref{qSupported}).
40898 This packet is optional for better performance on SVR4 targets.
40899 @value{GDBN} uses memory read packets to read the SVR4 library list otherwise.
40901 This packet is not probed by default; the remote stub must request it,
40902 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40904 If the remote stub indicates it supports the augmented form of this
40905 packet then the annex part of the generic @samp{qXfer} packet may
40906 contain a semicolon-separated list of @samp{@var{name}=@var{value}}
40907 arguments. The currently supported arguments are:
40910 @item start=@var{address}
40911 A hexadecimal number specifying the address of the @samp{struct
40912 link_map} to start reading the library list from. If unset or zero
40913 then the first @samp{struct link_map} in the library list will be
40914 chosen as the starting point.
40916 @item prev=@var{address}
40917 A hexadecimal number specifying the address of the @samp{struct
40918 link_map} immediately preceding the @samp{struct link_map}
40919 specified by the @samp{start} argument. If unset or zero then
40920 the remote stub will expect that no @samp{struct link_map}
40921 exists prior to the starting point.
40925 Arguments that are not understood by the remote stub will be silently
40928 @item qXfer:memory-map:read::@var{offset},@var{length}
40929 @anchor{qXfer memory map read}
40930 Access the target's @dfn{memory-map}. @xref{Memory Map Format}. The
40931 annex part of the generic @samp{qXfer} packet must be empty
40932 (@pxref{qXfer read}).
40934 This packet is not probed by default; the remote stub must request it,
40935 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40937 @item qXfer:sdata:read::@var{offset},@var{length}
40938 @anchor{qXfer sdata read}
40940 Read contents of the extra collected static tracepoint marker
40941 information. The annex part of the generic @samp{qXfer} packet must
40942 be empty (@pxref{qXfer read}). @xref{Tracepoint Actions,,Tracepoint
40945 This packet is not probed by default; the remote stub must request it,
40946 by supplying an appropriate @samp{qSupported} response
40947 (@pxref{qSupported}).
40949 @item qXfer:siginfo:read::@var{offset},@var{length}
40950 @anchor{qXfer siginfo read}
40951 Read contents of the extra signal information on the target
40952 system. The annex part of the generic @samp{qXfer} packet must be
40953 empty (@pxref{qXfer read}).
40955 This packet is not probed by default; the remote stub must request it,
40956 by supplying an appropriate @samp{qSupported} response
40957 (@pxref{qSupported}).
40959 @item qXfer:threads:read::@var{offset},@var{length}
40960 @anchor{qXfer threads read}
40961 Access the list of threads on target. @xref{Thread List Format}. The
40962 annex part of the generic @samp{qXfer} packet must be empty
40963 (@pxref{qXfer read}).
40965 This packet is not probed by default; the remote stub must request it,
40966 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40968 @item qXfer:traceframe-info:read::@var{offset},@var{length}
40969 @anchor{qXfer traceframe info read}
40971 Return a description of the current traceframe's contents.
40972 @xref{Traceframe Info Format}. The annex part of the generic
40973 @samp{qXfer} packet must be empty (@pxref{qXfer read}).
40975 This packet is not probed by default; the remote stub must request it,
40976 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40978 @item qXfer:uib:read:@var{pc}:@var{offset},@var{length}
40979 @anchor{qXfer unwind info block}
40981 Return the unwind information block for @var{pc}. This packet is used
40982 on OpenVMS/ia64 to ask the kernel unwind information.
40984 This packet is not probed by default.
40986 @item qXfer:fdpic:read:@var{annex}:@var{offset},@var{length}
40987 @anchor{qXfer fdpic loadmap read}
40988 Read contents of @code{loadmap}s on the target system. The
40989 annex, either @samp{exec} or @samp{interp}, specifies which @code{loadmap},
40990 executable @code{loadmap} or interpreter @code{loadmap} to read.
40992 This packet is not probed by default; the remote stub must request it,
40993 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40995 @item qXfer:osdata:read::@var{offset},@var{length}
40996 @anchor{qXfer osdata read}
40997 Access the target's @dfn{operating system information}.
40998 @xref{Operating System Information}.
41002 @item qXfer:@var{object}:write:@var{annex}:@var{offset}:@var{data}@dots{}
41003 @cindex write data into object, remote request
41004 @anchor{qXfer write}
41005 Write uninterpreted bytes into the target's special data area
41006 identified by the keyword @var{object}, starting at @var{offset} bytes
41007 into the data. The binary-encoded data (@pxref{Binary Data}) to be
41008 written is given by @var{data}@dots{}. The content and encoding of @var{annex}
41009 is specific to @var{object}; it can supply additional details about what data
41015 @var{nn} (hex encoded) is the number of bytes written.
41016 This may be fewer bytes than supplied in the request.
41019 The request was malformed, or @var{annex} was invalid.
41022 The offset was invalid, or there was an error encountered writing the data.
41023 The @var{nn} part is a hex-encoded @code{errno} value.
41026 An empty reply indicates the @var{object} string was not
41027 recognized by the stub, or that the object does not support writing.
41030 Here are the specific requests of this form defined so far. All the
41031 @samp{qXfer:@var{object}:write:@dots{}} requests use the same reply
41032 formats, listed above.
41035 @item qXfer:siginfo:write::@var{offset}:@var{data}@dots{}
41036 @anchor{qXfer siginfo write}
41037 Write @var{data} to the extra signal information on the target system.
41038 The annex part of the generic @samp{qXfer} packet must be
41039 empty (@pxref{qXfer write}).
41041 This packet is not probed by default; the remote stub must request it,
41042 by supplying an appropriate @samp{qSupported} response
41043 (@pxref{qSupported}).
41046 @item qXfer:@var{object}:@var{operation}:@dots{}
41047 Requests of this form may be added in the future. When a stub does
41048 not recognize the @var{object} keyword, or its support for
41049 @var{object} does not recognize the @var{operation} keyword, the stub
41050 must respond with an empty packet.
41052 @item qAttached:@var{pid}
41053 @cindex query attached, remote request
41054 @cindex @samp{qAttached} packet
41055 Return an indication of whether the remote server attached to an
41056 existing process or created a new process. When the multiprocess
41057 protocol extensions are supported (@pxref{multiprocess extensions}),
41058 @var{pid} is an integer in hexadecimal format identifying the target
41059 process. Otherwise, @value{GDBN} will omit the @var{pid} field and
41060 the query packet will be simplified as @samp{qAttached}.
41062 This query is used, for example, to know whether the remote process
41063 should be detached or killed when a @value{GDBN} session is ended with
41064 the @code{quit} command.
41069 The remote server attached to an existing process.
41071 The remote server created a new process.
41073 A badly formed request or an error was encountered.
41077 Enable branch tracing for the current thread using Branch Trace Store.
41082 Branch tracing has been enabled.
41084 A badly formed request or an error was encountered.
41088 Enable branch tracing for the current thread using Intel Processor Trace.
41093 Branch tracing has been enabled.
41095 A badly formed request or an error was encountered.
41099 Disable branch tracing for the current thread.
41104 Branch tracing has been disabled.
41106 A badly formed request or an error was encountered.
41109 @item Qbtrace-conf:bts:size=@var{value}
41110 Set the requested ring buffer size for new threads that use the
41111 btrace recording method in bts format.
41116 The ring buffer size has been set.
41118 A badly formed request or an error was encountered.
41121 @item Qbtrace-conf:pt:size=@var{value}
41122 Set the requested ring buffer size for new threads that use the
41123 btrace recording method in pt format.
41128 The ring buffer size has been set.
41130 A badly formed request or an error was encountered.
41135 @node Architecture-Specific Protocol Details
41136 @section Architecture-Specific Protocol Details
41138 This section describes how the remote protocol is applied to specific
41139 target architectures. Also see @ref{Standard Target Features}, for
41140 details of XML target descriptions for each architecture.
41143 * ARM-Specific Protocol Details::
41144 * MIPS-Specific Protocol Details::
41147 @node ARM-Specific Protocol Details
41148 @subsection @acronym{ARM}-specific Protocol Details
41151 * ARM Breakpoint Kinds::
41154 @node ARM Breakpoint Kinds
41155 @subsubsection @acronym{ARM} Breakpoint Kinds
41156 @cindex breakpoint kinds, @acronym{ARM}
41158 These breakpoint kinds are defined for the @samp{Z0} and @samp{Z1} packets.
41163 16-bit Thumb mode breakpoint.
41166 32-bit Thumb mode (Thumb-2) breakpoint.
41169 32-bit @acronym{ARM} mode breakpoint.
41173 @node MIPS-Specific Protocol Details
41174 @subsection @acronym{MIPS}-specific Protocol Details
41177 * MIPS Register packet Format::
41178 * MIPS Breakpoint Kinds::
41181 @node MIPS Register packet Format
41182 @subsubsection @acronym{MIPS} Register Packet Format
41183 @cindex register packet format, @acronym{MIPS}
41185 The following @code{g}/@code{G} packets have previously been defined.
41186 In the below, some thirty-two bit registers are transferred as
41187 sixty-four bits. Those registers should be zero/sign extended (which?)
41188 to fill the space allocated. Register bytes are transferred in target
41189 byte order. The two nibbles within a register byte are transferred
41190 most-significant -- least-significant.
41195 All registers are transferred as thirty-two bit quantities in the order:
41196 32 general-purpose; sr; lo; hi; bad; cause; pc; 32 floating-point
41197 registers; fsr; fir; fp.
41200 All registers are transferred as sixty-four bit quantities (including
41201 thirty-two bit registers such as @code{sr}). The ordering is the same
41206 @node MIPS Breakpoint Kinds
41207 @subsubsection @acronym{MIPS} Breakpoint Kinds
41208 @cindex breakpoint kinds, @acronym{MIPS}
41210 These breakpoint kinds are defined for the @samp{Z0} and @samp{Z1} packets.
41215 16-bit @acronym{MIPS16} mode breakpoint.
41218 16-bit @acronym{microMIPS} mode breakpoint.
41221 32-bit standard @acronym{MIPS} mode breakpoint.
41224 32-bit @acronym{microMIPS} mode breakpoint.
41228 @node Tracepoint Packets
41229 @section Tracepoint Packets
41230 @cindex tracepoint packets
41231 @cindex packets, tracepoint
41233 Here we describe the packets @value{GDBN} uses to implement
41234 tracepoints (@pxref{Tracepoints}).
41238 @item QTDP:@var{n}:@var{addr}:@var{ena}:@var{step}:@var{pass}[:F@var{flen}][:X@var{len},@var{bytes}]@r{[}-@r{]}
41239 @cindex @samp{QTDP} packet
41240 Create a new tracepoint, number @var{n}, at @var{addr}. If @var{ena}
41241 is @samp{E}, then the tracepoint is enabled; if it is @samp{D}, then
41242 the tracepoint is disabled. The @var{step} gives the tracepoint's step
41243 count, and @var{pass} gives its pass count. If an @samp{F} is present,
41244 then the tracepoint is to be a fast tracepoint, and the @var{flen} is
41245 the number of bytes that the target should copy elsewhere to make room
41246 for the tracepoint. If an @samp{X} is present, it introduces a
41247 tracepoint condition, which consists of a hexadecimal length, followed
41248 by a comma and hex-encoded bytes, in a manner similar to action
41249 encodings as described below. If the trailing @samp{-} is present,
41250 further @samp{QTDP} packets will follow to specify this tracepoint's
41256 The packet was understood and carried out.
41258 @xref{Tracepoint Packets,,Relocate instruction reply packet}.
41260 The packet was not recognized.
41263 @item QTDP:-@var{n}:@var{addr}:@r{[}S@r{]}@var{action}@dots{}@r{[}-@r{]}
41264 Define actions to be taken when a tracepoint is hit. The @var{n} and
41265 @var{addr} must be the same as in the initial @samp{QTDP} packet for
41266 this tracepoint. This packet may only be sent immediately after
41267 another @samp{QTDP} packet that ended with a @samp{-}. If the
41268 trailing @samp{-} is present, further @samp{QTDP} packets will follow,
41269 specifying more actions for this tracepoint.
41271 In the series of action packets for a given tracepoint, at most one
41272 can have an @samp{S} before its first @var{action}. If such a packet
41273 is sent, it and the following packets define ``while-stepping''
41274 actions. Any prior packets define ordinary actions --- that is, those
41275 taken when the tracepoint is first hit. If no action packet has an
41276 @samp{S}, then all the packets in the series specify ordinary
41277 tracepoint actions.
41279 The @samp{@var{action}@dots{}} portion of the packet is a series of
41280 actions, concatenated without separators. Each action has one of the
41286 Collect the registers whose bits are set in @var{mask},
41287 a hexadecimal number whose @var{i}'th bit is set if register number
41288 @var{i} should be collected. (The least significant bit is numbered
41289 zero.) Note that @var{mask} may be any number of digits long; it may
41290 not fit in a 32-bit word.
41292 @item M @var{basereg},@var{offset},@var{len}
41293 Collect @var{len} bytes of memory starting at the address in register
41294 number @var{basereg}, plus @var{offset}. If @var{basereg} is
41295 @samp{-1}, then the range has a fixed address: @var{offset} is the
41296 address of the lowest byte to collect. The @var{basereg},
41297 @var{offset}, and @var{len} parameters are all unsigned hexadecimal
41298 values (the @samp{-1} value for @var{basereg} is a special case).
41300 @item X @var{len},@var{expr}
41301 Evaluate @var{expr}, whose length is @var{len}, and collect memory as
41302 it directs. The agent expression @var{expr} is as described in
41303 @ref{Agent Expressions}. Each byte of the expression is encoded as a
41304 two-digit hex number in the packet; @var{len} is the number of bytes
41305 in the expression (and thus one-half the number of hex digits in the
41310 Any number of actions may be packed together in a single @samp{QTDP}
41311 packet, as long as the packet does not exceed the maximum packet
41312 length (400 bytes, for many stubs). There may be only one @samp{R}
41313 action per tracepoint, and it must precede any @samp{M} or @samp{X}
41314 actions. Any registers referred to by @samp{M} and @samp{X} actions
41315 must be collected by a preceding @samp{R} action. (The
41316 ``while-stepping'' actions are treated as if they were attached to a
41317 separate tracepoint, as far as these restrictions are concerned.)
41322 The packet was understood and carried out.
41324 @xref{Tracepoint Packets,,Relocate instruction reply packet}.
41326 The packet was not recognized.
41329 @item QTDPsrc:@var{n}:@var{addr}:@var{type}:@var{start}:@var{slen}:@var{bytes}
41330 @cindex @samp{QTDPsrc} packet
41331 Specify a source string of tracepoint @var{n} at address @var{addr}.
41332 This is useful to get accurate reproduction of the tracepoints
41333 originally downloaded at the beginning of the trace run. The @var{type}
41334 is the name of the tracepoint part, such as @samp{cond} for the
41335 tracepoint's conditional expression (see below for a list of types), while
41336 @var{bytes} is the string, encoded in hexadecimal.
41338 @var{start} is the offset of the @var{bytes} within the overall source
41339 string, while @var{slen} is the total length of the source string.
41340 This is intended for handling source strings that are longer than will
41341 fit in a single packet.
41342 @c Add detailed example when this info is moved into a dedicated
41343 @c tracepoint descriptions section.
41345 The available string types are @samp{at} for the location,
41346 @samp{cond} for the conditional, and @samp{cmd} for an action command.
41347 @value{GDBN} sends a separate packet for each command in the action
41348 list, in the same order in which the commands are stored in the list.
41350 The target does not need to do anything with source strings except
41351 report them back as part of the replies to the @samp{qTfP}/@samp{qTsP}
41354 Although this packet is optional, and @value{GDBN} will only send it
41355 if the target replies with @samp{TracepointSource} @xref{General
41356 Query Packets}, it makes both disconnected tracing and trace files
41357 much easier to use. Otherwise the user must be careful that the
41358 tracepoints in effect while looking at trace frames are identical to
41359 the ones in effect during the trace run; even a small discrepancy
41360 could cause @samp{tdump} not to work, or a particular trace frame not
41363 @item QTDV:@var{n}:@var{value}:@var{builtin}:@var{name}
41364 @cindex define trace state variable, remote request
41365 @cindex @samp{QTDV} packet
41366 Create a new trace state variable, number @var{n}, with an initial
41367 value of @var{value}, which is a 64-bit signed integer. Both @var{n}
41368 and @var{value} are encoded as hexadecimal values. @value{GDBN} has
41369 the option of not using this packet for initial values of zero; the
41370 target should simply create the trace state variables as they are
41371 mentioned in expressions. The value @var{builtin} should be 1 (one)
41372 if the trace state variable is builtin and 0 (zero) if it is not builtin.
41373 @value{GDBN} only sets @var{builtin} to 1 if a previous @samp{qTfV} or
41374 @samp{qTsV} packet had it set. The contents of @var{name} is the
41375 hex-encoded name (without the leading @samp{$}) of the trace state
41378 @item QTFrame:@var{n}
41379 @cindex @samp{QTFrame} packet
41380 Select the @var{n}'th tracepoint frame from the buffer, and use the
41381 register and memory contents recorded there to answer subsequent
41382 request packets from @value{GDBN}.
41384 A successful reply from the stub indicates that the stub has found the
41385 requested frame. The response is a series of parts, concatenated
41386 without separators, describing the frame we selected. Each part has
41387 one of the following forms:
41391 The selected frame is number @var{n} in the trace frame buffer;
41392 @var{f} is a hexadecimal number. If @var{f} is @samp{-1}, then there
41393 was no frame matching the criteria in the request packet.
41396 The selected trace frame records a hit of tracepoint number @var{t};
41397 @var{t} is a hexadecimal number.
41401 @item QTFrame:pc:@var{addr}
41402 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
41403 currently selected frame whose PC is @var{addr};
41404 @var{addr} is a hexadecimal number.
41406 @item QTFrame:tdp:@var{t}
41407 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
41408 currently selected frame that is a hit of tracepoint @var{t}; @var{t}
41409 is a hexadecimal number.
41411 @item QTFrame:range:@var{start}:@var{end}
41412 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
41413 currently selected frame whose PC is between @var{start} (inclusive)
41414 and @var{end} (inclusive); @var{start} and @var{end} are hexadecimal
41417 @item QTFrame:outside:@var{start}:@var{end}
41418 Like @samp{QTFrame:range:@var{start}:@var{end}}, but select the first
41419 frame @emph{outside} the given range of addresses (exclusive).
41422 @cindex @samp{qTMinFTPILen} packet
41423 This packet requests the minimum length of instruction at which a fast
41424 tracepoint (@pxref{Set Tracepoints}) may be placed. For instance, on
41425 the 32-bit x86 architecture, it is possible to use a 4-byte jump, but
41426 it depends on the target system being able to create trampolines in
41427 the first 64K of memory, which might or might not be possible for that
41428 system. So the reply to this packet will be 4 if it is able to
41435 The minimum instruction length is currently unknown.
41437 The minimum instruction length is @var{length}, where @var{length}
41438 is a hexadecimal number greater or equal to 1. A reply
41439 of 1 means that a fast tracepoint may be placed on any instruction
41440 regardless of size.
41442 An error has occurred.
41444 An empty reply indicates that the request is not supported by the stub.
41448 @cindex @samp{QTStart} packet
41449 Begin the tracepoint experiment. Begin collecting data from
41450 tracepoint hits in the trace frame buffer. This packet supports the
41451 @samp{qRelocInsn} reply (@pxref{Tracepoint Packets,,Relocate
41452 instruction reply packet}).
41455 @cindex @samp{QTStop} packet
41456 End the tracepoint experiment. Stop collecting trace frames.
41458 @item QTEnable:@var{n}:@var{addr}
41460 @cindex @samp{QTEnable} packet
41461 Enable tracepoint @var{n} at address @var{addr} in a started tracepoint
41462 experiment. If the tracepoint was previously disabled, then collection
41463 of data from it will resume.
41465 @item QTDisable:@var{n}:@var{addr}
41467 @cindex @samp{QTDisable} packet
41468 Disable tracepoint @var{n} at address @var{addr} in a started tracepoint
41469 experiment. No more data will be collected from the tracepoint unless
41470 @samp{QTEnable:@var{n}:@var{addr}} is subsequently issued.
41473 @cindex @samp{QTinit} packet
41474 Clear the table of tracepoints, and empty the trace frame buffer.
41476 @item QTro:@var{start1},@var{end1}:@var{start2},@var{end2}:@dots{}
41477 @cindex @samp{QTro} packet
41478 Establish the given ranges of memory as ``transparent''. The stub
41479 will answer requests for these ranges from memory's current contents,
41480 if they were not collected as part of the tracepoint hit.
41482 @value{GDBN} uses this to mark read-only regions of memory, like those
41483 containing program code. Since these areas never change, they should
41484 still have the same contents they did when the tracepoint was hit, so
41485 there's no reason for the stub to refuse to provide their contents.
41487 @item QTDisconnected:@var{value}
41488 @cindex @samp{QTDisconnected} packet
41489 Set the choice to what to do with the tracing run when @value{GDBN}
41490 disconnects from the target. A @var{value} of 1 directs the target to
41491 continue the tracing run, while 0 tells the target to stop tracing if
41492 @value{GDBN} is no longer in the picture.
41495 @cindex @samp{qTStatus} packet
41496 Ask the stub if there is a trace experiment running right now.
41498 The reply has the form:
41502 @item T@var{running}@r{[};@var{field}@r{]}@dots{}
41503 @var{running} is a single digit @code{1} if the trace is presently
41504 running, or @code{0} if not. It is followed by semicolon-separated
41505 optional fields that an agent may use to report additional status.
41509 If the trace is not running, the agent may report any of several
41510 explanations as one of the optional fields:
41515 No trace has been run yet.
41517 @item tstop[:@var{text}]:0
41518 The trace was stopped by a user-originated stop command. The optional
41519 @var{text} field is a user-supplied string supplied as part of the
41520 stop command (for instance, an explanation of why the trace was
41521 stopped manually). It is hex-encoded.
41524 The trace stopped because the trace buffer filled up.
41526 @item tdisconnected:0
41527 The trace stopped because @value{GDBN} disconnected from the target.
41529 @item tpasscount:@var{tpnum}
41530 The trace stopped because tracepoint @var{tpnum} exceeded its pass count.
41532 @item terror:@var{text}:@var{tpnum}
41533 The trace stopped because tracepoint @var{tpnum} had an error. The
41534 string @var{text} is available to describe the nature of the error
41535 (for instance, a divide by zero in the condition expression); it
41539 The trace stopped for some other reason.
41543 Additional optional fields supply statistical and other information.
41544 Although not required, they are extremely useful for users monitoring
41545 the progress of a trace run. If a trace has stopped, and these
41546 numbers are reported, they must reflect the state of the just-stopped
41551 @item tframes:@var{n}
41552 The number of trace frames in the buffer.
41554 @item tcreated:@var{n}
41555 The total number of trace frames created during the run. This may
41556 be larger than the trace frame count, if the buffer is circular.
41558 @item tsize:@var{n}
41559 The total size of the trace buffer, in bytes.
41561 @item tfree:@var{n}
41562 The number of bytes still unused in the buffer.
41564 @item circular:@var{n}
41565 The value of the circular trace buffer flag. @code{1} means that the
41566 trace buffer is circular and old trace frames will be discarded if
41567 necessary to make room, @code{0} means that the trace buffer is linear
41570 @item disconn:@var{n}
41571 The value of the disconnected tracing flag. @code{1} means that
41572 tracing will continue after @value{GDBN} disconnects, @code{0} means
41573 that the trace run will stop.
41577 @item qTP:@var{tp}:@var{addr}
41578 @cindex tracepoint status, remote request
41579 @cindex @samp{qTP} packet
41580 Ask the stub for the current state of tracepoint number @var{tp} at
41581 address @var{addr}.
41585 @item V@var{hits}:@var{usage}
41586 The tracepoint has been hit @var{hits} times so far during the trace
41587 run, and accounts for @var{usage} in the trace buffer. Note that
41588 @code{while-stepping} steps are not counted as separate hits, but the
41589 steps' space consumption is added into the usage number.
41593 @item qTV:@var{var}
41594 @cindex trace state variable value, remote request
41595 @cindex @samp{qTV} packet
41596 Ask the stub for the value of the trace state variable number @var{var}.
41601 The value of the variable is @var{value}. This will be the current
41602 value of the variable if the user is examining a running target, or a
41603 saved value if the variable was collected in the trace frame that the
41604 user is looking at. Note that multiple requests may result in
41605 different reply values, such as when requesting values while the
41606 program is running.
41609 The value of the variable is unknown. This would occur, for example,
41610 if the user is examining a trace frame in which the requested variable
41615 @cindex @samp{qTfP} packet
41617 @cindex @samp{qTsP} packet
41618 These packets request data about tracepoints that are being used by
41619 the target. @value{GDBN} sends @code{qTfP} to get the first piece
41620 of data, and multiple @code{qTsP} to get additional pieces. Replies
41621 to these packets generally take the form of the @code{QTDP} packets
41622 that define tracepoints. (FIXME add detailed syntax)
41625 @cindex @samp{qTfV} packet
41627 @cindex @samp{qTsV} packet
41628 These packets request data about trace state variables that are on the
41629 target. @value{GDBN} sends @code{qTfV} to get the first vari of data,
41630 and multiple @code{qTsV} to get additional variables. Replies to
41631 these packets follow the syntax of the @code{QTDV} packets that define
41632 trace state variables.
41638 @cindex @samp{qTfSTM} packet
41639 @cindex @samp{qTsSTM} packet
41640 These packets request data about static tracepoint markers that exist
41641 in the target program. @value{GDBN} sends @code{qTfSTM} to get the
41642 first piece of data, and multiple @code{qTsSTM} to get additional
41643 pieces. Replies to these packets take the following form:
41647 @item m @var{address}:@var{id}:@var{extra}
41649 @item m @var{address}:@var{id}:@var{extra},@var{address}:@var{id}:@var{extra}@dots{}
41650 a comma-separated list of markers
41652 (lower case letter @samp{L}) denotes end of list.
41654 An error occurred. The error number @var{nn} is given as hex digits.
41656 An empty reply indicates that the request is not supported by the
41660 The @var{address} is encoded in hex;
41661 @var{id} and @var{extra} are strings encoded in hex.
41663 In response to each query, the target will reply with a list of one or
41664 more markers, separated by commas. @value{GDBN} will respond to each
41665 reply with a request for more markers (using the @samp{qs} form of the
41666 query), until the target responds with @samp{l} (lower-case ell, for
41669 @item qTSTMat:@var{address}
41671 @cindex @samp{qTSTMat} packet
41672 This packets requests data about static tracepoint markers in the
41673 target program at @var{address}. Replies to this packet follow the
41674 syntax of the @samp{qTfSTM} and @code{qTsSTM} packets that list static
41675 tracepoint markers.
41677 @item QTSave:@var{filename}
41678 @cindex @samp{QTSave} packet
41679 This packet directs the target to save trace data to the file name
41680 @var{filename} in the target's filesystem. The @var{filename} is encoded
41681 as a hex string; the interpretation of the file name (relative vs
41682 absolute, wild cards, etc) is up to the target.
41684 @item qTBuffer:@var{offset},@var{len}
41685 @cindex @samp{qTBuffer} packet
41686 Return up to @var{len} bytes of the current contents of trace buffer,
41687 starting at @var{offset}. The trace buffer is treated as if it were
41688 a contiguous collection of traceframes, as per the trace file format.
41689 The reply consists as many hex-encoded bytes as the target can deliver
41690 in a packet; it is not an error to return fewer than were asked for.
41691 A reply consisting of just @code{l} indicates that no bytes are
41694 @item QTBuffer:circular:@var{value}
41695 This packet directs the target to use a circular trace buffer if
41696 @var{value} is 1, or a linear buffer if the value is 0.
41698 @item QTBuffer:size:@var{size}
41699 @anchor{QTBuffer-size}
41700 @cindex @samp{QTBuffer size} packet
41701 This packet directs the target to make the trace buffer be of size
41702 @var{size} if possible. A value of @code{-1} tells the target to
41703 use whatever size it prefers.
41705 @item QTNotes:@r{[}@var{type}:@var{text}@r{]}@r{[};@var{type}:@var{text}@r{]}@dots{}
41706 @cindex @samp{QTNotes} packet
41707 This packet adds optional textual notes to the trace run. Allowable
41708 types include @code{user}, @code{notes}, and @code{tstop}, the
41709 @var{text} fields are arbitrary strings, hex-encoded.
41713 @subsection Relocate instruction reply packet
41714 When installing fast tracepoints in memory, the target may need to
41715 relocate the instruction currently at the tracepoint address to a
41716 different address in memory. For most instructions, a simple copy is
41717 enough, but, for example, call instructions that implicitly push the
41718 return address on the stack, and relative branches or other
41719 PC-relative instructions require offset adjustment, so that the effect
41720 of executing the instruction at a different address is the same as if
41721 it had executed in the original location.
41723 In response to several of the tracepoint packets, the target may also
41724 respond with a number of intermediate @samp{qRelocInsn} request
41725 packets before the final result packet, to have @value{GDBN} handle
41726 this relocation operation. If a packet supports this mechanism, its
41727 documentation will explicitly say so. See for example the above
41728 descriptions for the @samp{QTStart} and @samp{QTDP} packets. The
41729 format of the request is:
41732 @item qRelocInsn:@var{from};@var{to}
41734 This requests @value{GDBN} to copy instruction at address @var{from}
41735 to address @var{to}, possibly adjusted so that executing the
41736 instruction at @var{to} has the same effect as executing it at
41737 @var{from}. @value{GDBN} writes the adjusted instruction to target
41738 memory starting at @var{to}.
41743 @item qRelocInsn:@var{adjusted_size}
41744 Informs the stub the relocation is complete. The @var{adjusted_size} is
41745 the length in bytes of resulting relocated instruction sequence.
41747 A badly formed request was detected, or an error was encountered while
41748 relocating the instruction.
41751 @node Host I/O Packets
41752 @section Host I/O Packets
41753 @cindex Host I/O, remote protocol
41754 @cindex file transfer, remote protocol
41756 The @dfn{Host I/O} packets allow @value{GDBN} to perform I/O
41757 operations on the far side of a remote link. For example, Host I/O is
41758 used to upload and download files to a remote target with its own
41759 filesystem. Host I/O uses the same constant values and data structure
41760 layout as the target-initiated File-I/O protocol. However, the
41761 Host I/O packets are structured differently. The target-initiated
41762 protocol relies on target memory to store parameters and buffers.
41763 Host I/O requests are initiated by @value{GDBN}, and the
41764 target's memory is not involved. @xref{File-I/O Remote Protocol
41765 Extension}, for more details on the target-initiated protocol.
41767 The Host I/O request packets all encode a single operation along with
41768 its arguments. They have this format:
41772 @item vFile:@var{operation}: @var{parameter}@dots{}
41773 @var{operation} is the name of the particular request; the target
41774 should compare the entire packet name up to the second colon when checking
41775 for a supported operation. The format of @var{parameter} depends on
41776 the operation. Numbers are always passed in hexadecimal. Negative
41777 numbers have an explicit minus sign (i.e.@: two's complement is not
41778 used). Strings (e.g.@: filenames) are encoded as a series of
41779 hexadecimal bytes. The last argument to a system call may be a
41780 buffer of escaped binary data (@pxref{Binary Data}).
41784 The valid responses to Host I/O packets are:
41788 @item F @var{result} [, @var{errno}] [; @var{attachment}]
41789 @var{result} is the integer value returned by this operation, usually
41790 non-negative for success and -1 for errors. If an error has occured,
41791 @var{errno} will be included in the result specifying a
41792 value defined by the File-I/O protocol (@pxref{Errno Values}). For
41793 operations which return data, @var{attachment} supplies the data as a
41794 binary buffer. Binary buffers in response packets are escaped in the
41795 normal way (@pxref{Binary Data}). See the individual packet
41796 documentation for the interpretation of @var{result} and
41800 An empty response indicates that this operation is not recognized.
41804 These are the supported Host I/O operations:
41807 @item vFile:open: @var{filename}, @var{flags}, @var{mode}
41808 Open a file at @var{filename} and return a file descriptor for it, or
41809 return -1 if an error occurs. The @var{filename} is a string,
41810 @var{flags} is an integer indicating a mask of open flags
41811 (@pxref{Open Flags}), and @var{mode} is an integer indicating a mask
41812 of mode bits to use if the file is created (@pxref{mode_t Values}).
41813 @xref{open}, for details of the open flags and mode values.
41815 @item vFile:close: @var{fd}
41816 Close the open file corresponding to @var{fd} and return 0, or
41817 -1 if an error occurs.
41819 @item vFile:pread: @var{fd}, @var{count}, @var{offset}
41820 Read data from the open file corresponding to @var{fd}. Up to
41821 @var{count} bytes will be read from the file, starting at @var{offset}
41822 relative to the start of the file. The target may read fewer bytes;
41823 common reasons include packet size limits and an end-of-file
41824 condition. The number of bytes read is returned. Zero should only be
41825 returned for a successful read at the end of the file, or if
41826 @var{count} was zero.
41828 The data read should be returned as a binary attachment on success.
41829 If zero bytes were read, the response should include an empty binary
41830 attachment (i.e.@: a trailing semicolon). The return value is the
41831 number of target bytes read; the binary attachment may be longer if
41832 some characters were escaped.
41834 @item vFile:pwrite: @var{fd}, @var{offset}, @var{data}
41835 Write @var{data} (a binary buffer) to the open file corresponding
41836 to @var{fd}. Start the write at @var{offset} from the start of the
41837 file. Unlike many @code{write} system calls, there is no
41838 separate @var{count} argument; the length of @var{data} in the
41839 packet is used. @samp{vFile:pwrite} returns the number of bytes written,
41840 which may be shorter than the length of @var{data}, or -1 if an
41843 @item vFile:fstat: @var{fd}
41844 Get information about the open file corresponding to @var{fd}.
41845 On success the information is returned as a binary attachment
41846 and the return value is the size of this attachment in bytes.
41847 If an error occurs the return value is -1. The format of the
41848 returned binary attachment is as described in @ref{struct stat}.
41850 @item vFile:unlink: @var{filename}
41851 Delete the file at @var{filename} on the target. Return 0,
41852 or -1 if an error occurs. The @var{filename} is a string.
41854 @item vFile:readlink: @var{filename}
41855 Read value of symbolic link @var{filename} on the target. Return
41856 the number of bytes read, or -1 if an error occurs.
41858 The data read should be returned as a binary attachment on success.
41859 If zero bytes were read, the response should include an empty binary
41860 attachment (i.e.@: a trailing semicolon). The return value is the
41861 number of target bytes read; the binary attachment may be longer if
41862 some characters were escaped.
41864 @item vFile:setfs: @var{pid}
41865 Select the filesystem on which @code{vFile} operations with
41866 @var{filename} arguments will operate. This is required for
41867 @value{GDBN} to be able to access files on remote targets where
41868 the remote stub does not share a common filesystem with the
41871 If @var{pid} is nonzero, select the filesystem as seen by process
41872 @var{pid}. If @var{pid} is zero, select the filesystem as seen by
41873 the remote stub. Return 0 on success, or -1 if an error occurs.
41874 If @code{vFile:setfs:} indicates success, the selected filesystem
41875 remains selected until the next successful @code{vFile:setfs:}
41881 @section Interrupts
41882 @cindex interrupts (remote protocol)
41883 @anchor{interrupting remote targets}
41885 In all-stop mode, when a program on the remote target is running,
41886 @value{GDBN} may attempt to interrupt it by sending a @samp{Ctrl-C},
41887 @code{BREAK} or a @code{BREAK} followed by @code{g}, control of which
41888 is specified via @value{GDBN}'s @samp{interrupt-sequence}.
41890 The precise meaning of @code{BREAK} is defined by the transport
41891 mechanism and may, in fact, be undefined. @value{GDBN} does not
41892 currently define a @code{BREAK} mechanism for any of the network
41893 interfaces except for TCP, in which case @value{GDBN} sends the
41894 @code{telnet} BREAK sequence.
41896 @samp{Ctrl-C}, on the other hand, is defined and implemented for all
41897 transport mechanisms. It is represented by sending the single byte
41898 @code{0x03} without any of the usual packet overhead described in
41899 the Overview section (@pxref{Overview}). When a @code{0x03} byte is
41900 transmitted as part of a packet, it is considered to be packet data
41901 and does @emph{not} represent an interrupt. E.g., an @samp{X} packet
41902 (@pxref{X packet}), used for binary downloads, may include an unescaped
41903 @code{0x03} as part of its packet.
41905 @code{BREAK} followed by @code{g} is also known as Magic SysRq g.
41906 When Linux kernel receives this sequence from serial port,
41907 it stops execution and connects to gdb.
41909 In non-stop mode, because packet resumptions are asynchronous
41910 (@pxref{vCont packet}), @value{GDBN} is always free to send a remote
41911 command to the remote stub, even when the target is running. For that
41912 reason, @value{GDBN} instead sends a regular packet (@pxref{vCtrlC
41913 packet}) with the usual packet framing instead of the single byte
41916 Stubs are not required to recognize these interrupt mechanisms and the
41917 precise meaning associated with receipt of the interrupt is
41918 implementation defined. If the target supports debugging of multiple
41919 threads and/or processes, it should attempt to interrupt all
41920 currently-executing threads and processes.
41921 If the stub is successful at interrupting the
41922 running program, it should send one of the stop
41923 reply packets (@pxref{Stop Reply Packets}) to @value{GDBN} as a result
41924 of successfully stopping the program in all-stop mode, and a stop reply
41925 for each stopped thread in non-stop mode.
41926 Interrupts received while the
41927 program is stopped are queued and the program will be interrupted when
41928 it is resumed next time.
41930 @node Notification Packets
41931 @section Notification Packets
41932 @cindex notification packets
41933 @cindex packets, notification
41935 The @value{GDBN} remote serial protocol includes @dfn{notifications},
41936 packets that require no acknowledgment. Both the GDB and the stub
41937 may send notifications (although the only notifications defined at
41938 present are sent by the stub). Notifications carry information
41939 without incurring the round-trip latency of an acknowledgment, and so
41940 are useful for low-impact communications where occasional packet loss
41943 A notification packet has the form @samp{% @var{data} #
41944 @var{checksum}}, where @var{data} is the content of the notification,
41945 and @var{checksum} is a checksum of @var{data}, computed and formatted
41946 as for ordinary @value{GDBN} packets. A notification's @var{data}
41947 never contains @samp{$}, @samp{%} or @samp{#} characters. Upon
41948 receiving a notification, the recipient sends no @samp{+} or @samp{-}
41949 to acknowledge the notification's receipt or to report its corruption.
41951 Every notification's @var{data} begins with a name, which contains no
41952 colon characters, followed by a colon character.
41954 Recipients should silently ignore corrupted notifications and
41955 notifications they do not understand. Recipients should restart
41956 timeout periods on receipt of a well-formed notification, whether or
41957 not they understand it.
41959 Senders should only send the notifications described here when this
41960 protocol description specifies that they are permitted. In the
41961 future, we may extend the protocol to permit existing notifications in
41962 new contexts; this rule helps older senders avoid confusing newer
41965 (Older versions of @value{GDBN} ignore bytes received until they see
41966 the @samp{$} byte that begins an ordinary packet, so new stubs may
41967 transmit notifications without fear of confusing older clients. There
41968 are no notifications defined for @value{GDBN} to send at the moment, but we
41969 assume that most older stubs would ignore them, as well.)
41971 Each notification is comprised of three parts:
41973 @item @var{name}:@var{event}
41974 The notification packet is sent by the side that initiates the
41975 exchange (currently, only the stub does that), with @var{event}
41976 carrying the specific information about the notification, and
41977 @var{name} specifying the name of the notification.
41979 The acknowledge sent by the other side, usually @value{GDBN}, to
41980 acknowledge the exchange and request the event.
41983 The purpose of an asynchronous notification mechanism is to report to
41984 @value{GDBN} that something interesting happened in the remote stub.
41986 The remote stub may send notification @var{name}:@var{event}
41987 at any time, but @value{GDBN} acknowledges the notification when
41988 appropriate. The notification event is pending before @value{GDBN}
41989 acknowledges. Only one notification at a time may be pending; if
41990 additional events occur before @value{GDBN} has acknowledged the
41991 previous notification, they must be queued by the stub for later
41992 synchronous transmission in response to @var{ack} packets from
41993 @value{GDBN}. Because the notification mechanism is unreliable,
41994 the stub is permitted to resend a notification if it believes
41995 @value{GDBN} may not have received it.
41997 Specifically, notifications may appear when @value{GDBN} is not
41998 otherwise reading input from the stub, or when @value{GDBN} is
41999 expecting to read a normal synchronous response or a
42000 @samp{+}/@samp{-} acknowledgment to a packet it has sent.
42001 Notification packets are distinct from any other communication from
42002 the stub so there is no ambiguity.
42004 After receiving a notification, @value{GDBN} shall acknowledge it by
42005 sending a @var{ack} packet as a regular, synchronous request to the
42006 stub. Such acknowledgment is not required to happen immediately, as
42007 @value{GDBN} is permitted to send other, unrelated packets to the
42008 stub first, which the stub should process normally.
42010 Upon receiving a @var{ack} packet, if the stub has other queued
42011 events to report to @value{GDBN}, it shall respond by sending a
42012 normal @var{event}. @value{GDBN} shall then send another @var{ack}
42013 packet to solicit further responses; again, it is permitted to send
42014 other, unrelated packets as well which the stub should process
42017 If the stub receives a @var{ack} packet and there are no additional
42018 @var{event} to report, the stub shall return an @samp{OK} response.
42019 At this point, @value{GDBN} has finished processing a notification
42020 and the stub has completed sending any queued events. @value{GDBN}
42021 won't accept any new notifications until the final @samp{OK} is
42022 received . If further notification events occur, the stub shall send
42023 a new notification, @value{GDBN} shall accept the notification, and
42024 the process shall be repeated.
42026 The process of asynchronous notification can be illustrated by the
42029 <- @code{%Stop:T0505:98e7ffbf;04:4ce6ffbf;08:b1b6e54c;thread:p7526.7526;core:0;}
42032 <- @code{T0505:68f37db7;04:40f37db7;08:63850408;thread:p7526.7528;core:0;}
42034 <- @code{T0505:68e3fdb6;04:40e3fdb6;08:63850408;thread:p7526.7529;core:0;}
42039 The following notifications are defined:
42040 @multitable @columnfractions 0.12 0.12 0.38 0.38
42049 @tab @var{reply}. The @var{reply} has the form of a stop reply, as
42050 described in @ref{Stop Reply Packets}. Refer to @ref{Remote Non-Stop},
42051 for information on how these notifications are acknowledged by
42053 @tab Report an asynchronous stop event in non-stop mode.
42057 @node Remote Non-Stop
42058 @section Remote Protocol Support for Non-Stop Mode
42060 @value{GDBN}'s remote protocol supports non-stop debugging of
42061 multi-threaded programs, as described in @ref{Non-Stop Mode}. If the stub
42062 supports non-stop mode, it should report that to @value{GDBN} by including
42063 @samp{QNonStop+} in its @samp{qSupported} response (@pxref{qSupported}).
42065 @value{GDBN} typically sends a @samp{QNonStop} packet only when
42066 establishing a new connection with the stub. Entering non-stop mode
42067 does not alter the state of any currently-running threads, but targets
42068 must stop all threads in any already-attached processes when entering
42069 all-stop mode. @value{GDBN} uses the @samp{?} packet as necessary to
42070 probe the target state after a mode change.
42072 In non-stop mode, when an attached process encounters an event that
42073 would otherwise be reported with a stop reply, it uses the
42074 asynchronous notification mechanism (@pxref{Notification Packets}) to
42075 inform @value{GDBN}. In contrast to all-stop mode, where all threads
42076 in all processes are stopped when a stop reply is sent, in non-stop
42077 mode only the thread reporting the stop event is stopped. That is,
42078 when reporting a @samp{S} or @samp{T} response to indicate completion
42079 of a step operation, hitting a breakpoint, or a fault, only the
42080 affected thread is stopped; any other still-running threads continue
42081 to run. When reporting a @samp{W} or @samp{X} response, all running
42082 threads belonging to other attached processes continue to run.
42084 In non-stop mode, the target shall respond to the @samp{?} packet as
42085 follows. First, any incomplete stop reply notification/@samp{vStopped}
42086 sequence in progress is abandoned. The target must begin a new
42087 sequence reporting stop events for all stopped threads, whether or not
42088 it has previously reported those events to @value{GDBN}. The first
42089 stop reply is sent as a synchronous reply to the @samp{?} packet, and
42090 subsequent stop replies are sent as responses to @samp{vStopped} packets
42091 using the mechanism described above. The target must not send
42092 asynchronous stop reply notifications until the sequence is complete.
42093 If all threads are running when the target receives the @samp{?} packet,
42094 or if the target is not attached to any process, it shall respond
42097 If the stub supports non-stop mode, it should also support the
42098 @samp{swbreak} stop reason if software breakpoints are supported, and
42099 the @samp{hwbreak} stop reason if hardware breakpoints are supported
42100 (@pxref{swbreak stop reason}). This is because given the asynchronous
42101 nature of non-stop mode, between the time a thread hits a breakpoint
42102 and the time the event is finally processed by @value{GDBN}, the
42103 breakpoint may have already been removed from the target. Due to
42104 this, @value{GDBN} needs to be able to tell whether a trap stop was
42105 caused by a delayed breakpoint event, which should be ignored, as
42106 opposed to a random trap signal, which should be reported to the user.
42107 Note the @samp{swbreak} feature implies that the target is responsible
42108 for adjusting the PC when a software breakpoint triggers, if
42109 necessary, such as on the x86 architecture.
42111 @node Packet Acknowledgment
42112 @section Packet Acknowledgment
42114 @cindex acknowledgment, for @value{GDBN} remote
42115 @cindex packet acknowledgment, for @value{GDBN} remote
42116 By default, when either the host or the target machine receives a packet,
42117 the first response expected is an acknowledgment: either @samp{+} (to indicate
42118 the package was received correctly) or @samp{-} (to request retransmission).
42119 This mechanism allows the @value{GDBN} remote protocol to operate over
42120 unreliable transport mechanisms, such as a serial line.
42122 In cases where the transport mechanism is itself reliable (such as a pipe or
42123 TCP connection), the @samp{+}/@samp{-} acknowledgments are redundant.
42124 It may be desirable to disable them in that case to reduce communication
42125 overhead, or for other reasons. This can be accomplished by means of the
42126 @samp{QStartNoAckMode} packet; @pxref{QStartNoAckMode}.
42128 When in no-acknowledgment mode, neither the stub nor @value{GDBN} shall send or
42129 expect @samp{+}/@samp{-} protocol acknowledgments. The packet
42130 and response format still includes the normal checksum, as described in
42131 @ref{Overview}, but the checksum may be ignored by the receiver.
42133 If the stub supports @samp{QStartNoAckMode} and prefers to operate in
42134 no-acknowledgment mode, it should report that to @value{GDBN}
42135 by including @samp{QStartNoAckMode+} in its response to @samp{qSupported};
42136 @pxref{qSupported}.
42137 If @value{GDBN} also supports @samp{QStartNoAckMode} and it has not been
42138 disabled via the @code{set remote noack-packet off} command
42139 (@pxref{Remote Configuration}),
42140 @value{GDBN} may then send a @samp{QStartNoAckMode} packet to the stub.
42141 Only then may the stub actually turn off packet acknowledgments.
42142 @value{GDBN} sends a final @samp{+} acknowledgment of the stub's @samp{OK}
42143 response, which can be safely ignored by the stub.
42145 Note that @code{set remote noack-packet} command only affects negotiation
42146 between @value{GDBN} and the stub when subsequent connections are made;
42147 it does not affect the protocol acknowledgment state for any current
42149 Since @samp{+}/@samp{-} acknowledgments are enabled by default when a
42150 new connection is established,
42151 there is also no protocol request to re-enable the acknowledgments
42152 for the current connection, once disabled.
42157 Example sequence of a target being re-started. Notice how the restart
42158 does not get any direct output:
42163 @emph{target restarts}
42166 <- @code{T001:1234123412341234}
42170 Example sequence of a target being stepped by a single instruction:
42173 -> @code{G1445@dots{}}
42178 <- @code{T001:1234123412341234}
42182 <- @code{1455@dots{}}
42186 @node File-I/O Remote Protocol Extension
42187 @section File-I/O Remote Protocol Extension
42188 @cindex File-I/O remote protocol extension
42191 * File-I/O Overview::
42192 * Protocol Basics::
42193 * The F Request Packet::
42194 * The F Reply Packet::
42195 * The Ctrl-C Message::
42197 * List of Supported Calls::
42198 * Protocol-specific Representation of Datatypes::
42200 * File-I/O Examples::
42203 @node File-I/O Overview
42204 @subsection File-I/O Overview
42205 @cindex file-i/o overview
42207 The @dfn{File I/O remote protocol extension} (short: File-I/O) allows the
42208 target to use the host's file system and console I/O to perform various
42209 system calls. System calls on the target system are translated into a
42210 remote protocol packet to the host system, which then performs the needed
42211 actions and returns a response packet to the target system.
42212 This simulates file system operations even on targets that lack file systems.
42214 The protocol is defined to be independent of both the host and target systems.
42215 It uses its own internal representation of datatypes and values. Both
42216 @value{GDBN} and the target's @value{GDBN} stub are responsible for
42217 translating the system-dependent value representations into the internal
42218 protocol representations when data is transmitted.
42220 The communication is synchronous. A system call is possible only when
42221 @value{GDBN} is waiting for a response from the @samp{C}, @samp{c}, @samp{S}
42222 or @samp{s} packets. While @value{GDBN} handles the request for a system call,
42223 the target is stopped to allow deterministic access to the target's
42224 memory. Therefore File-I/O is not interruptible by target signals. On
42225 the other hand, it is possible to interrupt File-I/O by a user interrupt
42226 (@samp{Ctrl-C}) within @value{GDBN}.
42228 The target's request to perform a host system call does not finish
42229 the latest @samp{C}, @samp{c}, @samp{S} or @samp{s} action. That means,
42230 after finishing the system call, the target returns to continuing the
42231 previous activity (continue, step). No additional continue or step
42232 request from @value{GDBN} is required.
42235 (@value{GDBP}) continue
42236 <- target requests 'system call X'
42237 target is stopped, @value{GDBN} executes system call
42238 -> @value{GDBN} returns result
42239 ... target continues, @value{GDBN} returns to wait for the target
42240 <- target hits breakpoint and sends a Txx packet
42243 The protocol only supports I/O on the console and to regular files on
42244 the host file system. Character or block special devices, pipes,
42245 named pipes, sockets or any other communication method on the host
42246 system are not supported by this protocol.
42248 File I/O is not supported in non-stop mode.
42250 @node Protocol Basics
42251 @subsection Protocol Basics
42252 @cindex protocol basics, file-i/o
42254 The File-I/O protocol uses the @code{F} packet as the request as well
42255 as reply packet. Since a File-I/O system call can only occur when
42256 @value{GDBN} is waiting for a response from the continuing or stepping target,
42257 the File-I/O request is a reply that @value{GDBN} has to expect as a result
42258 of a previous @samp{C}, @samp{c}, @samp{S} or @samp{s} packet.
42259 This @code{F} packet contains all information needed to allow @value{GDBN}
42260 to call the appropriate host system call:
42264 A unique identifier for the requested system call.
42267 All parameters to the system call. Pointers are given as addresses
42268 in the target memory address space. Pointers to strings are given as
42269 pointer/length pair. Numerical values are given as they are.
42270 Numerical control flags are given in a protocol-specific representation.
42274 At this point, @value{GDBN} has to perform the following actions.
42278 If the parameters include pointer values to data needed as input to a
42279 system call, @value{GDBN} requests this data from the target with a
42280 standard @code{m} packet request. This additional communication has to be
42281 expected by the target implementation and is handled as any other @code{m}
42285 @value{GDBN} translates all value from protocol representation to host
42286 representation as needed. Datatypes are coerced into the host types.
42289 @value{GDBN} calls the system call.
42292 It then coerces datatypes back to protocol representation.
42295 If the system call is expected to return data in buffer space specified
42296 by pointer parameters to the call, the data is transmitted to the
42297 target using a @code{M} or @code{X} packet. This packet has to be expected
42298 by the target implementation and is handled as any other @code{M} or @code{X}
42303 Eventually @value{GDBN} replies with another @code{F} packet which contains all
42304 necessary information for the target to continue. This at least contains
42311 @code{errno}, if has been changed by the system call.
42318 After having done the needed type and value coercion, the target continues
42319 the latest continue or step action.
42321 @node The F Request Packet
42322 @subsection The @code{F} Request Packet
42323 @cindex file-i/o request packet
42324 @cindex @code{F} request packet
42326 The @code{F} request packet has the following format:
42329 @item F@var{call-id},@var{parameter@dots{}}
42331 @var{call-id} is the identifier to indicate the host system call to be called.
42332 This is just the name of the function.
42334 @var{parameter@dots{}} are the parameters to the system call.
42335 Parameters are hexadecimal integer values, either the actual values in case
42336 of scalar datatypes, pointers to target buffer space in case of compound
42337 datatypes and unspecified memory areas, or pointer/length pairs in case
42338 of string parameters. These are appended to the @var{call-id} as a
42339 comma-delimited list. All values are transmitted in ASCII
42340 string representation, pointer/length pairs separated by a slash.
42346 @node The F Reply Packet
42347 @subsection The @code{F} Reply Packet
42348 @cindex file-i/o reply packet
42349 @cindex @code{F} reply packet
42351 The @code{F} reply packet has the following format:
42355 @item F@var{retcode},@var{errno},@var{Ctrl-C flag};@var{call-specific attachment}
42357 @var{retcode} is the return code of the system call as hexadecimal value.
42359 @var{errno} is the @code{errno} set by the call, in protocol-specific
42361 This parameter can be omitted if the call was successful.
42363 @var{Ctrl-C flag} is only sent if the user requested a break. In this
42364 case, @var{errno} must be sent as well, even if the call was successful.
42365 The @var{Ctrl-C flag} itself consists of the character @samp{C}:
42372 or, if the call was interrupted before the host call has been performed:
42379 assuming 4 is the protocol-specific representation of @code{EINTR}.
42384 @node The Ctrl-C Message
42385 @subsection The @samp{Ctrl-C} Message
42386 @cindex ctrl-c message, in file-i/o protocol
42388 If the @samp{Ctrl-C} flag is set in the @value{GDBN}
42389 reply packet (@pxref{The F Reply Packet}),
42390 the target should behave as if it had
42391 gotten a break message. The meaning for the target is ``system call
42392 interrupted by @code{SIGINT}''. Consequentially, the target should actually stop
42393 (as with a break message) and return to @value{GDBN} with a @code{T02}
42396 It's important for the target to know in which
42397 state the system call was interrupted. There are two possible cases:
42401 The system call hasn't been performed on the host yet.
42404 The system call on the host has been finished.
42408 These two states can be distinguished by the target by the value of the
42409 returned @code{errno}. If it's the protocol representation of @code{EINTR}, the system
42410 call hasn't been performed. This is equivalent to the @code{EINTR} handling
42411 on POSIX systems. In any other case, the target may presume that the
42412 system call has been finished --- successfully or not --- and should behave
42413 as if the break message arrived right after the system call.
42415 @value{GDBN} must behave reliably. If the system call has not been called
42416 yet, @value{GDBN} may send the @code{F} reply immediately, setting @code{EINTR} as
42417 @code{errno} in the packet. If the system call on the host has been finished
42418 before the user requests a break, the full action must be finished by
42419 @value{GDBN}. This requires sending @code{M} or @code{X} packets as necessary.
42420 The @code{F} packet may only be sent when either nothing has happened
42421 or the full action has been completed.
42424 @subsection Console I/O
42425 @cindex console i/o as part of file-i/o
42427 By default and if not explicitly closed by the target system, the file
42428 descriptors 0, 1 and 2 are connected to the @value{GDBN} console. Output
42429 on the @value{GDBN} console is handled as any other file output operation
42430 (@code{write(1, @dots{})} or @code{write(2, @dots{})}). Console input is handled
42431 by @value{GDBN} so that after the target read request from file descriptor
42432 0 all following typing is buffered until either one of the following
42437 The user types @kbd{Ctrl-c}. The behaviour is as explained above, and the
42439 system call is treated as finished.
42442 The user presses @key{RET}. This is treated as end of input with a trailing
42446 The user types @kbd{Ctrl-d}. This is treated as end of input. No trailing
42447 character (neither newline nor @samp{Ctrl-D}) is appended to the input.
42451 If the user has typed more characters than fit in the buffer given to
42452 the @code{read} call, the trailing characters are buffered in @value{GDBN} until
42453 either another @code{read(0, @dots{})} is requested by the target, or debugging
42454 is stopped at the user's request.
42457 @node List of Supported Calls
42458 @subsection List of Supported Calls
42459 @cindex list of supported file-i/o calls
42476 @unnumberedsubsubsec open
42477 @cindex open, file-i/o system call
42482 int open(const char *pathname, int flags);
42483 int open(const char *pathname, int flags, mode_t mode);
42487 @samp{Fopen,@var{pathptr}/@var{len},@var{flags},@var{mode}}
42490 @var{flags} is the bitwise @code{OR} of the following values:
42494 If the file does not exist it will be created. The host
42495 rules apply as far as file ownership and time stamps
42499 When used with @code{O_CREAT}, if the file already exists it is
42500 an error and open() fails.
42503 If the file already exists and the open mode allows
42504 writing (@code{O_RDWR} or @code{O_WRONLY} is given) it will be
42505 truncated to zero length.
42508 The file is opened in append mode.
42511 The file is opened for reading only.
42514 The file is opened for writing only.
42517 The file is opened for reading and writing.
42521 Other bits are silently ignored.
42525 @var{mode} is the bitwise @code{OR} of the following values:
42529 User has read permission.
42532 User has write permission.
42535 Group has read permission.
42538 Group has write permission.
42541 Others have read permission.
42544 Others have write permission.
42548 Other bits are silently ignored.
42551 @item Return value:
42552 @code{open} returns the new file descriptor or -1 if an error
42559 @var{pathname} already exists and @code{O_CREAT} and @code{O_EXCL} were used.
42562 @var{pathname} refers to a directory.
42565 The requested access is not allowed.
42568 @var{pathname} was too long.
42571 A directory component in @var{pathname} does not exist.
42574 @var{pathname} refers to a device, pipe, named pipe or socket.
42577 @var{pathname} refers to a file on a read-only filesystem and
42578 write access was requested.
42581 @var{pathname} is an invalid pointer value.
42584 No space on device to create the file.
42587 The process already has the maximum number of files open.
42590 The limit on the total number of files open on the system
42594 The call was interrupted by the user.
42600 @unnumberedsubsubsec close
42601 @cindex close, file-i/o system call
42610 @samp{Fclose,@var{fd}}
42612 @item Return value:
42613 @code{close} returns zero on success, or -1 if an error occurred.
42619 @var{fd} isn't a valid open file descriptor.
42622 The call was interrupted by the user.
42628 @unnumberedsubsubsec read
42629 @cindex read, file-i/o system call
42634 int read(int fd, void *buf, unsigned int count);
42638 @samp{Fread,@var{fd},@var{bufptr},@var{count}}
42640 @item Return value:
42641 On success, the number of bytes read is returned.
42642 Zero indicates end of file. If count is zero, read
42643 returns zero as well. On error, -1 is returned.
42649 @var{fd} is not a valid file descriptor or is not open for
42653 @var{bufptr} is an invalid pointer value.
42656 The call was interrupted by the user.
42662 @unnumberedsubsubsec write
42663 @cindex write, file-i/o system call
42668 int write(int fd, const void *buf, unsigned int count);
42672 @samp{Fwrite,@var{fd},@var{bufptr},@var{count}}
42674 @item Return value:
42675 On success, the number of bytes written are returned.
42676 Zero indicates nothing was written. On error, -1
42683 @var{fd} is not a valid file descriptor or is not open for
42687 @var{bufptr} is an invalid pointer value.
42690 An attempt was made to write a file that exceeds the
42691 host-specific maximum file size allowed.
42694 No space on device to write the data.
42697 The call was interrupted by the user.
42703 @unnumberedsubsubsec lseek
42704 @cindex lseek, file-i/o system call
42709 long lseek (int fd, long offset, int flag);
42713 @samp{Flseek,@var{fd},@var{offset},@var{flag}}
42715 @var{flag} is one of:
42719 The offset is set to @var{offset} bytes.
42722 The offset is set to its current location plus @var{offset}
42726 The offset is set to the size of the file plus @var{offset}
42730 @item Return value:
42731 On success, the resulting unsigned offset in bytes from
42732 the beginning of the file is returned. Otherwise, a
42733 value of -1 is returned.
42739 @var{fd} is not a valid open file descriptor.
42742 @var{fd} is associated with the @value{GDBN} console.
42745 @var{flag} is not a proper value.
42748 The call was interrupted by the user.
42754 @unnumberedsubsubsec rename
42755 @cindex rename, file-i/o system call
42760 int rename(const char *oldpath, const char *newpath);
42764 @samp{Frename,@var{oldpathptr}/@var{len},@var{newpathptr}/@var{len}}
42766 @item Return value:
42767 On success, zero is returned. On error, -1 is returned.
42773 @var{newpath} is an existing directory, but @var{oldpath} is not a
42777 @var{newpath} is a non-empty directory.
42780 @var{oldpath} or @var{newpath} is a directory that is in use by some
42784 An attempt was made to make a directory a subdirectory
42788 A component used as a directory in @var{oldpath} or new
42789 path is not a directory. Or @var{oldpath} is a directory
42790 and @var{newpath} exists but is not a directory.
42793 @var{oldpathptr} or @var{newpathptr} are invalid pointer values.
42796 No access to the file or the path of the file.
42800 @var{oldpath} or @var{newpath} was too long.
42803 A directory component in @var{oldpath} or @var{newpath} does not exist.
42806 The file is on a read-only filesystem.
42809 The device containing the file has no room for the new
42813 The call was interrupted by the user.
42819 @unnumberedsubsubsec unlink
42820 @cindex unlink, file-i/o system call
42825 int unlink(const char *pathname);
42829 @samp{Funlink,@var{pathnameptr}/@var{len}}
42831 @item Return value:
42832 On success, zero is returned. On error, -1 is returned.
42838 No access to the file or the path of the file.
42841 The system does not allow unlinking of directories.
42844 The file @var{pathname} cannot be unlinked because it's
42845 being used by another process.
42848 @var{pathnameptr} is an invalid pointer value.
42851 @var{pathname} was too long.
42854 A directory component in @var{pathname} does not exist.
42857 A component of the path is not a directory.
42860 The file is on a read-only filesystem.
42863 The call was interrupted by the user.
42869 @unnumberedsubsubsec stat/fstat
42870 @cindex fstat, file-i/o system call
42871 @cindex stat, file-i/o system call
42876 int stat(const char *pathname, struct stat *buf);
42877 int fstat(int fd, struct stat *buf);
42881 @samp{Fstat,@var{pathnameptr}/@var{len},@var{bufptr}}@*
42882 @samp{Ffstat,@var{fd},@var{bufptr}}
42884 @item Return value:
42885 On success, zero is returned. On error, -1 is returned.
42891 @var{fd} is not a valid open file.
42894 A directory component in @var{pathname} does not exist or the
42895 path is an empty string.
42898 A component of the path is not a directory.
42901 @var{pathnameptr} is an invalid pointer value.
42904 No access to the file or the path of the file.
42907 @var{pathname} was too long.
42910 The call was interrupted by the user.
42916 @unnumberedsubsubsec gettimeofday
42917 @cindex gettimeofday, file-i/o system call
42922 int gettimeofday(struct timeval *tv, void *tz);
42926 @samp{Fgettimeofday,@var{tvptr},@var{tzptr}}
42928 @item Return value:
42929 On success, 0 is returned, -1 otherwise.
42935 @var{tz} is a non-NULL pointer.
42938 @var{tvptr} and/or @var{tzptr} is an invalid pointer value.
42944 @unnumberedsubsubsec isatty
42945 @cindex isatty, file-i/o system call
42950 int isatty(int fd);
42954 @samp{Fisatty,@var{fd}}
42956 @item Return value:
42957 Returns 1 if @var{fd} refers to the @value{GDBN} console, 0 otherwise.
42963 The call was interrupted by the user.
42968 Note that the @code{isatty} call is treated as a special case: it returns
42969 1 to the target if the file descriptor is attached
42970 to the @value{GDBN} console, 0 otherwise. Implementing through system calls
42971 would require implementing @code{ioctl} and would be more complex than
42976 @unnumberedsubsubsec system
42977 @cindex system, file-i/o system call
42982 int system(const char *command);
42986 @samp{Fsystem,@var{commandptr}/@var{len}}
42988 @item Return value:
42989 If @var{len} is zero, the return value indicates whether a shell is
42990 available. A zero return value indicates a shell is not available.
42991 For non-zero @var{len}, the value returned is -1 on error and the
42992 return status of the command otherwise. Only the exit status of the
42993 command is returned, which is extracted from the host's @code{system}
42994 return value by calling @code{WEXITSTATUS(retval)}. In case
42995 @file{/bin/sh} could not be executed, 127 is returned.
43001 The call was interrupted by the user.
43006 @value{GDBN} takes over the full task of calling the necessary host calls
43007 to perform the @code{system} call. The return value of @code{system} on
43008 the host is simplified before it's returned
43009 to the target. Any termination signal information from the child process
43010 is discarded, and the return value consists
43011 entirely of the exit status of the called command.
43013 Due to security concerns, the @code{system} call is by default refused
43014 by @value{GDBN}. The user has to allow this call explicitly with the
43015 @code{set remote system-call-allowed 1} command.
43018 @item set remote system-call-allowed
43019 @kindex set remote system-call-allowed
43020 Control whether to allow the @code{system} calls in the File I/O
43021 protocol for the remote target. The default is zero (disabled).
43023 @item show remote system-call-allowed
43024 @kindex show remote system-call-allowed
43025 Show whether the @code{system} calls are allowed in the File I/O
43029 @node Protocol-specific Representation of Datatypes
43030 @subsection Protocol-specific Representation of Datatypes
43031 @cindex protocol-specific representation of datatypes, in file-i/o protocol
43034 * Integral Datatypes::
43036 * Memory Transfer::
43041 @node Integral Datatypes
43042 @unnumberedsubsubsec Integral Datatypes
43043 @cindex integral datatypes, in file-i/o protocol
43045 The integral datatypes used in the system calls are @code{int},
43046 @code{unsigned int}, @code{long}, @code{unsigned long},
43047 @code{mode_t}, and @code{time_t}.
43049 @code{int}, @code{unsigned int}, @code{mode_t} and @code{time_t} are
43050 implemented as 32 bit values in this protocol.
43052 @code{long} and @code{unsigned long} are implemented as 64 bit types.
43054 @xref{Limits}, for corresponding MIN and MAX values (similar to those
43055 in @file{limits.h}) to allow range checking on host and target.
43057 @code{time_t} datatypes are defined as seconds since the Epoch.
43059 All integral datatypes transferred as part of a memory read or write of a
43060 structured datatype e.g.@: a @code{struct stat} have to be given in big endian
43063 @node Pointer Values
43064 @unnumberedsubsubsec Pointer Values
43065 @cindex pointer values, in file-i/o protocol
43067 Pointers to target data are transmitted as they are. An exception
43068 is made for pointers to buffers for which the length isn't
43069 transmitted as part of the function call, namely strings. Strings
43070 are transmitted as a pointer/length pair, both as hex values, e.g.@:
43077 which is a pointer to data of length 18 bytes at position 0x1aaf.
43078 The length is defined as the full string length in bytes, including
43079 the trailing null byte. For example, the string @code{"hello world"}
43080 at address 0x123456 is transmitted as
43086 @node Memory Transfer
43087 @unnumberedsubsubsec Memory Transfer
43088 @cindex memory transfer, in file-i/o protocol
43090 Structured data which is transferred using a memory read or write (for
43091 example, a @code{struct stat}) is expected to be in a protocol-specific format
43092 with all scalar multibyte datatypes being big endian. Translation to
43093 this representation needs to be done both by the target before the @code{F}
43094 packet is sent, and by @value{GDBN} before
43095 it transfers memory to the target. Transferred pointers to structured
43096 data should point to the already-coerced data at any time.
43100 @unnumberedsubsubsec struct stat
43101 @cindex struct stat, in file-i/o protocol
43103 The buffer of type @code{struct stat} used by the target and @value{GDBN}
43104 is defined as follows:
43108 unsigned int st_dev; /* device */
43109 unsigned int st_ino; /* inode */
43110 mode_t st_mode; /* protection */
43111 unsigned int st_nlink; /* number of hard links */
43112 unsigned int st_uid; /* user ID of owner */
43113 unsigned int st_gid; /* group ID of owner */
43114 unsigned int st_rdev; /* device type (if inode device) */
43115 unsigned long st_size; /* total size, in bytes */
43116 unsigned long st_blksize; /* blocksize for filesystem I/O */
43117 unsigned long st_blocks; /* number of blocks allocated */
43118 time_t st_atime; /* time of last access */
43119 time_t st_mtime; /* time of last modification */
43120 time_t st_ctime; /* time of last change */
43124 The integral datatypes conform to the definitions given in the
43125 appropriate section (see @ref{Integral Datatypes}, for details) so this
43126 structure is of size 64 bytes.
43128 The values of several fields have a restricted meaning and/or
43134 A value of 0 represents a file, 1 the console.
43137 No valid meaning for the target. Transmitted unchanged.
43140 Valid mode bits are described in @ref{Constants}. Any other
43141 bits have currently no meaning for the target.
43146 No valid meaning for the target. Transmitted unchanged.
43151 These values have a host and file system dependent
43152 accuracy. Especially on Windows hosts, the file system may not
43153 support exact timing values.
43156 The target gets a @code{struct stat} of the above representation and is
43157 responsible for coercing it to the target representation before
43160 Note that due to size differences between the host, target, and protocol
43161 representations of @code{struct stat} members, these members could eventually
43162 get truncated on the target.
43164 @node struct timeval
43165 @unnumberedsubsubsec struct timeval
43166 @cindex struct timeval, in file-i/o protocol
43168 The buffer of type @code{struct timeval} used by the File-I/O protocol
43169 is defined as follows:
43173 time_t tv_sec; /* second */
43174 long tv_usec; /* microsecond */
43178 The integral datatypes conform to the definitions given in the
43179 appropriate section (see @ref{Integral Datatypes}, for details) so this
43180 structure is of size 8 bytes.
43183 @subsection Constants
43184 @cindex constants, in file-i/o protocol
43186 The following values are used for the constants inside of the
43187 protocol. @value{GDBN} and target are responsible for translating these
43188 values before and after the call as needed.
43199 @unnumberedsubsubsec Open Flags
43200 @cindex open flags, in file-i/o protocol
43202 All values are given in hexadecimal representation.
43214 @node mode_t Values
43215 @unnumberedsubsubsec mode_t Values
43216 @cindex mode_t values, in file-i/o protocol
43218 All values are given in octal representation.
43235 @unnumberedsubsubsec Errno Values
43236 @cindex errno values, in file-i/o protocol
43238 All values are given in decimal representation.
43263 @code{EUNKNOWN} is used as a fallback error value if a host system returns
43264 any error value not in the list of supported error numbers.
43267 @unnumberedsubsubsec Lseek Flags
43268 @cindex lseek flags, in file-i/o protocol
43277 @unnumberedsubsubsec Limits
43278 @cindex limits, in file-i/o protocol
43280 All values are given in decimal representation.
43283 INT_MIN -2147483648
43285 UINT_MAX 4294967295
43286 LONG_MIN -9223372036854775808
43287 LONG_MAX 9223372036854775807
43288 ULONG_MAX 18446744073709551615
43291 @node File-I/O Examples
43292 @subsection File-I/O Examples
43293 @cindex file-i/o examples
43295 Example sequence of a write call, file descriptor 3, buffer is at target
43296 address 0x1234, 6 bytes should be written:
43299 <- @code{Fwrite,3,1234,6}
43300 @emph{request memory read from target}
43303 @emph{return "6 bytes written"}
43307 Example sequence of a read call, file descriptor 3, buffer is at target
43308 address 0x1234, 6 bytes should be read:
43311 <- @code{Fread,3,1234,6}
43312 @emph{request memory write to target}
43313 -> @code{X1234,6:XXXXXX}
43314 @emph{return "6 bytes read"}
43318 Example sequence of a read call, call fails on the host due to invalid
43319 file descriptor (@code{EBADF}):
43322 <- @code{Fread,3,1234,6}
43326 Example sequence of a read call, user presses @kbd{Ctrl-c} before syscall on
43330 <- @code{Fread,3,1234,6}
43335 Example sequence of a read call, user presses @kbd{Ctrl-c} after syscall on
43339 <- @code{Fread,3,1234,6}
43340 -> @code{X1234,6:XXXXXX}
43344 @node Library List Format
43345 @section Library List Format
43346 @cindex library list format, remote protocol
43348 On some platforms, a dynamic loader (e.g.@: @file{ld.so}) runs in the
43349 same process as your application to manage libraries. In this case,
43350 @value{GDBN} can use the loader's symbol table and normal memory
43351 operations to maintain a list of shared libraries. On other
43352 platforms, the operating system manages loaded libraries.
43353 @value{GDBN} can not retrieve the list of currently loaded libraries
43354 through memory operations, so it uses the @samp{qXfer:libraries:read}
43355 packet (@pxref{qXfer library list read}) instead. The remote stub
43356 queries the target's operating system and reports which libraries
43359 The @samp{qXfer:libraries:read} packet returns an XML document which
43360 lists loaded libraries and their offsets. Each library has an
43361 associated name and one or more segment or section base addresses,
43362 which report where the library was loaded in memory.
43364 For the common case of libraries that are fully linked binaries, the
43365 library should have a list of segments. If the target supports
43366 dynamic linking of a relocatable object file, its library XML element
43367 should instead include a list of allocated sections. The segment or
43368 section bases are start addresses, not relocation offsets; they do not
43369 depend on the library's link-time base addresses.
43371 @value{GDBN} must be linked with the Expat library to support XML
43372 library lists. @xref{Expat}.
43374 A simple memory map, with one loaded library relocated by a single
43375 offset, looks like this:
43379 <library name="/lib/libc.so.6">
43380 <segment address="0x10000000"/>
43385 Another simple memory map, with one loaded library with three
43386 allocated sections (.text, .data, .bss), looks like this:
43390 <library name="sharedlib.o">
43391 <section address="0x10000000"/>
43392 <section address="0x20000000"/>
43393 <section address="0x30000000"/>
43398 The format of a library list is described by this DTD:
43401 <!-- library-list: Root element with versioning -->
43402 <!ELEMENT library-list (library)*>
43403 <!ATTLIST library-list version CDATA #FIXED "1.0">
43404 <!ELEMENT library (segment*, section*)>
43405 <!ATTLIST library name CDATA #REQUIRED>
43406 <!ELEMENT segment EMPTY>
43407 <!ATTLIST segment address CDATA #REQUIRED>
43408 <!ELEMENT section EMPTY>
43409 <!ATTLIST section address CDATA #REQUIRED>
43412 In addition, segments and section descriptors cannot be mixed within a
43413 single library element, and you must supply at least one segment or
43414 section for each library.
43416 @node Library List Format for SVR4 Targets
43417 @section Library List Format for SVR4 Targets
43418 @cindex library list format, remote protocol
43420 On SVR4 platforms @value{GDBN} can use the symbol table of a dynamic loader
43421 (e.g.@: @file{ld.so}) and normal memory operations to maintain a list of
43422 shared libraries. Still a special library list provided by this packet is
43423 more efficient for the @value{GDBN} remote protocol.
43425 The @samp{qXfer:libraries-svr4:read} packet returns an XML document which lists
43426 loaded libraries and their SVR4 linker parameters. For each library on SVR4
43427 target, the following parameters are reported:
43431 @code{name}, the absolute file name from the @code{l_name} field of
43432 @code{struct link_map}.
43434 @code{lm} with address of @code{struct link_map} used for TLS
43435 (Thread Local Storage) access.
43437 @code{l_addr}, the displacement as read from the field @code{l_addr} of
43438 @code{struct link_map}. For prelinked libraries this is not an absolute
43439 memory address. It is a displacement of absolute memory address against
43440 address the file was prelinked to during the library load.
43442 @code{l_ld}, which is memory address of the @code{PT_DYNAMIC} segment
43445 Additionally the single @code{main-lm} attribute specifies address of
43446 @code{struct link_map} used for the main executable. This parameter is used
43447 for TLS access and its presence is optional.
43449 @value{GDBN} must be linked with the Expat library to support XML
43450 SVR4 library lists. @xref{Expat}.
43452 A simple memory map, with two loaded libraries (which do not use prelink),
43456 <library-list-svr4 version="1.0" main-lm="0xe4f8f8">
43457 <library name="/lib/ld-linux.so.2" lm="0xe4f51c" l_addr="0xe2d000"
43459 <library name="/lib/libc.so.6" lm="0xe4fbe8" l_addr="0x154000"
43461 </library-list-svr>
43464 The format of an SVR4 library list is described by this DTD:
43467 <!-- library-list-svr4: Root element with versioning -->
43468 <!ELEMENT library-list-svr4 (library)*>
43469 <!ATTLIST library-list-svr4 version CDATA #FIXED "1.0">
43470 <!ATTLIST library-list-svr4 main-lm CDATA #IMPLIED>
43471 <!ELEMENT library EMPTY>
43472 <!ATTLIST library name CDATA #REQUIRED>
43473 <!ATTLIST library lm CDATA #REQUIRED>
43474 <!ATTLIST library l_addr CDATA #REQUIRED>
43475 <!ATTLIST library l_ld CDATA #REQUIRED>
43478 @node Memory Map Format
43479 @section Memory Map Format
43480 @cindex memory map format
43482 To be able to write into flash memory, @value{GDBN} needs to obtain a
43483 memory map from the target. This section describes the format of the
43486 The memory map is obtained using the @samp{qXfer:memory-map:read}
43487 (@pxref{qXfer memory map read}) packet and is an XML document that
43488 lists memory regions.
43490 @value{GDBN} must be linked with the Expat library to support XML
43491 memory maps. @xref{Expat}.
43493 The top-level structure of the document is shown below:
43496 <?xml version="1.0"?>
43497 <!DOCTYPE memory-map
43498 PUBLIC "+//IDN gnu.org//DTD GDB Memory Map V1.0//EN"
43499 "http://sourceware.org/gdb/gdb-memory-map.dtd">
43505 Each region can be either:
43510 A region of RAM starting at @var{addr} and extending for @var{length}
43514 <memory type="ram" start="@var{addr}" length="@var{length}"/>
43519 A region of read-only memory:
43522 <memory type="rom" start="@var{addr}" length="@var{length}"/>
43527 A region of flash memory, with erasure blocks @var{blocksize}
43531 <memory type="flash" start="@var{addr}" length="@var{length}">
43532 <property name="blocksize">@var{blocksize}</property>
43538 Regions must not overlap. @value{GDBN} assumes that areas of memory not covered
43539 by the memory map are RAM, and uses the ordinary @samp{M} and @samp{X}
43540 packets to write to addresses in such ranges.
43542 The formal DTD for memory map format is given below:
43545 <!-- ................................................... -->
43546 <!-- Memory Map XML DTD ................................ -->
43547 <!-- File: memory-map.dtd .............................. -->
43548 <!-- .................................... .............. -->
43549 <!-- memory-map.dtd -->
43550 <!-- memory-map: Root element with versioning -->
43551 <!ELEMENT memory-map (memory)*>
43552 <!ATTLIST memory-map version CDATA #FIXED "1.0.0">
43553 <!ELEMENT memory (property)*>
43554 <!-- memory: Specifies a memory region,
43555 and its type, or device. -->
43556 <!ATTLIST memory type (ram|rom|flash) #REQUIRED
43557 start CDATA #REQUIRED
43558 length CDATA #REQUIRED>
43559 <!-- property: Generic attribute tag -->
43560 <!ELEMENT property (#PCDATA | property)*>
43561 <!ATTLIST property name (blocksize) #REQUIRED>
43564 @node Thread List Format
43565 @section Thread List Format
43566 @cindex thread list format
43568 To efficiently update the list of threads and their attributes,
43569 @value{GDBN} issues the @samp{qXfer:threads:read} packet
43570 (@pxref{qXfer threads read}) and obtains the XML document with
43571 the following structure:
43574 <?xml version="1.0"?>
43576 <thread id="id" core="0" name="name">
43577 ... description ...
43582 Each @samp{thread} element must have the @samp{id} attribute that
43583 identifies the thread (@pxref{thread-id syntax}). The
43584 @samp{core} attribute, if present, specifies which processor core
43585 the thread was last executing on. The @samp{name} attribute, if
43586 present, specifies the human-readable name of the thread. The content
43587 of the of @samp{thread} element is interpreted as human-readable
43588 auxiliary information. The @samp{handle} attribute, if present,
43589 is a hex encoded representation of the thread handle.
43592 @node Traceframe Info Format
43593 @section Traceframe Info Format
43594 @cindex traceframe info format
43596 To be able to know which objects in the inferior can be examined when
43597 inspecting a tracepoint hit, @value{GDBN} needs to obtain the list of
43598 memory ranges, registers and trace state variables that have been
43599 collected in a traceframe.
43601 This list is obtained using the @samp{qXfer:traceframe-info:read}
43602 (@pxref{qXfer traceframe info read}) packet and is an XML document.
43604 @value{GDBN} must be linked with the Expat library to support XML
43605 traceframe info discovery. @xref{Expat}.
43607 The top-level structure of the document is shown below:
43610 <?xml version="1.0"?>
43611 <!DOCTYPE traceframe-info
43612 PUBLIC "+//IDN gnu.org//DTD GDB Memory Map V1.0//EN"
43613 "http://sourceware.org/gdb/gdb-traceframe-info.dtd">
43619 Each traceframe block can be either:
43624 A region of collected memory starting at @var{addr} and extending for
43625 @var{length} bytes from there:
43628 <memory start="@var{addr}" length="@var{length}"/>
43632 A block indicating trace state variable numbered @var{number} has been
43636 <tvar id="@var{number}"/>
43641 The formal DTD for the traceframe info format is given below:
43644 <!ELEMENT traceframe-info (memory | tvar)* >
43645 <!ATTLIST traceframe-info version CDATA #FIXED "1.0">
43647 <!ELEMENT memory EMPTY>
43648 <!ATTLIST memory start CDATA #REQUIRED
43649 length CDATA #REQUIRED>
43651 <!ATTLIST tvar id CDATA #REQUIRED>
43654 @node Branch Trace Format
43655 @section Branch Trace Format
43656 @cindex branch trace format
43658 In order to display the branch trace of an inferior thread,
43659 @value{GDBN} needs to obtain the list of branches. This list is
43660 represented as list of sequential code blocks that are connected via
43661 branches. The code in each block has been executed sequentially.
43663 This list is obtained using the @samp{qXfer:btrace:read}
43664 (@pxref{qXfer btrace read}) packet and is an XML document.
43666 @value{GDBN} must be linked with the Expat library to support XML
43667 traceframe info discovery. @xref{Expat}.
43669 The top-level structure of the document is shown below:
43672 <?xml version="1.0"?>
43674 PUBLIC "+//IDN gnu.org//DTD GDB Branch Trace V1.0//EN"
43675 "http://sourceware.org/gdb/gdb-btrace.dtd">
43684 A block of sequentially executed instructions starting at @var{begin}
43685 and ending at @var{end}:
43688 <block begin="@var{begin}" end="@var{end}"/>
43693 The formal DTD for the branch trace format is given below:
43696 <!ELEMENT btrace (block* | pt) >
43697 <!ATTLIST btrace version CDATA #FIXED "1.0">
43699 <!ELEMENT block EMPTY>
43700 <!ATTLIST block begin CDATA #REQUIRED
43701 end CDATA #REQUIRED>
43703 <!ELEMENT pt (pt-config?, raw?)>
43705 <!ELEMENT pt-config (cpu?)>
43707 <!ELEMENT cpu EMPTY>
43708 <!ATTLIST cpu vendor CDATA #REQUIRED
43709 family CDATA #REQUIRED
43710 model CDATA #REQUIRED
43711 stepping CDATA #REQUIRED>
43713 <!ELEMENT raw (#PCDATA)>
43716 @node Branch Trace Configuration Format
43717 @section Branch Trace Configuration Format
43718 @cindex branch trace configuration format
43720 For each inferior thread, @value{GDBN} can obtain the branch trace
43721 configuration using the @samp{qXfer:btrace-conf:read}
43722 (@pxref{qXfer btrace-conf read}) packet.
43724 The configuration describes the branch trace format and configuration
43725 settings for that format. The following information is described:
43729 This thread uses the @dfn{Branch Trace Store} (@acronym{BTS}) format.
43732 The size of the @acronym{BTS} ring buffer in bytes.
43735 This thread uses the @dfn{Intel Processor Trace} (@acronym{Intel
43739 The size of the @acronym{Intel PT} ring buffer in bytes.
43743 @value{GDBN} must be linked with the Expat library to support XML
43744 branch trace configuration discovery. @xref{Expat}.
43746 The formal DTD for the branch trace configuration format is given below:
43749 <!ELEMENT btrace-conf (bts?, pt?)>
43750 <!ATTLIST btrace-conf version CDATA #FIXED "1.0">
43752 <!ELEMENT bts EMPTY>
43753 <!ATTLIST bts size CDATA #IMPLIED>
43755 <!ELEMENT pt EMPTY>
43756 <!ATTLIST pt size CDATA #IMPLIED>
43759 @include agentexpr.texi
43761 @node Target Descriptions
43762 @appendix Target Descriptions
43763 @cindex target descriptions
43765 One of the challenges of using @value{GDBN} to debug embedded systems
43766 is that there are so many minor variants of each processor
43767 architecture in use. It is common practice for vendors to start with
43768 a standard processor core --- ARM, PowerPC, or @acronym{MIPS}, for example ---
43769 and then make changes to adapt it to a particular market niche. Some
43770 architectures have hundreds of variants, available from dozens of
43771 vendors. This leads to a number of problems:
43775 With so many different customized processors, it is difficult for
43776 the @value{GDBN} maintainers to keep up with the changes.
43778 Since individual variants may have short lifetimes or limited
43779 audiences, it may not be worthwhile to carry information about every
43780 variant in the @value{GDBN} source tree.
43782 When @value{GDBN} does support the architecture of the embedded system
43783 at hand, the task of finding the correct architecture name to give the
43784 @command{set architecture} command can be error-prone.
43787 To address these problems, the @value{GDBN} remote protocol allows a
43788 target system to not only identify itself to @value{GDBN}, but to
43789 actually describe its own features. This lets @value{GDBN} support
43790 processor variants it has never seen before --- to the extent that the
43791 descriptions are accurate, and that @value{GDBN} understands them.
43793 @value{GDBN} must be linked with the Expat library to support XML
43794 target descriptions. @xref{Expat}.
43797 * Retrieving Descriptions:: How descriptions are fetched from a target.
43798 * Target Description Format:: The contents of a target description.
43799 * Predefined Target Types:: Standard types available for target
43801 * Enum Target Types:: How to define enum target types.
43802 * Standard Target Features:: Features @value{GDBN} knows about.
43805 @node Retrieving Descriptions
43806 @section Retrieving Descriptions
43808 Target descriptions can be read from the target automatically, or
43809 specified by the user manually. The default behavior is to read the
43810 description from the target. @value{GDBN} retrieves it via the remote
43811 protocol using @samp{qXfer} requests (@pxref{General Query Packets,
43812 qXfer}). The @var{annex} in the @samp{qXfer} packet will be
43813 @samp{target.xml}. The contents of the @samp{target.xml} annex are an
43814 XML document, of the form described in @ref{Target Description
43817 Alternatively, you can specify a file to read for the target description.
43818 If a file is set, the target will not be queried. The commands to
43819 specify a file are:
43822 @cindex set tdesc filename
43823 @item set tdesc filename @var{path}
43824 Read the target description from @var{path}.
43826 @cindex unset tdesc filename
43827 @item unset tdesc filename
43828 Do not read the XML target description from a file. @value{GDBN}
43829 will use the description supplied by the current target.
43831 @cindex show tdesc filename
43832 @item show tdesc filename
43833 Show the filename to read for a target description, if any.
43837 @node Target Description Format
43838 @section Target Description Format
43839 @cindex target descriptions, XML format
43841 A target description annex is an @uref{http://www.w3.org/XML/, XML}
43842 document which complies with the Document Type Definition provided in
43843 the @value{GDBN} sources in @file{gdb/features/gdb-target.dtd}. This
43844 means you can use generally available tools like @command{xmllint} to
43845 check that your feature descriptions are well-formed and valid.
43846 However, to help people unfamiliar with XML write descriptions for
43847 their targets, we also describe the grammar here.
43849 Target descriptions can identify the architecture of the remote target
43850 and (for some architectures) provide information about custom register
43851 sets. They can also identify the OS ABI of the remote target.
43852 @value{GDBN} can use this information to autoconfigure for your
43853 target, or to warn you if you connect to an unsupported target.
43855 Here is a simple target description:
43858 <target version="1.0">
43859 <architecture>i386:x86-64</architecture>
43864 This minimal description only says that the target uses
43865 the x86-64 architecture.
43867 A target description has the following overall form, with [ ] marking
43868 optional elements and @dots{} marking repeatable elements. The elements
43869 are explained further below.
43872 <?xml version="1.0"?>
43873 <!DOCTYPE target SYSTEM "gdb-target.dtd">
43874 <target version="1.0">
43875 @r{[}@var{architecture}@r{]}
43876 @r{[}@var{osabi}@r{]}
43877 @r{[}@var{compatible}@r{]}
43878 @r{[}@var{feature}@dots{}@r{]}
43883 The description is generally insensitive to whitespace and line
43884 breaks, under the usual common-sense rules. The XML version
43885 declaration and document type declaration can generally be omitted
43886 (@value{GDBN} does not require them), but specifying them may be
43887 useful for XML validation tools. The @samp{version} attribute for
43888 @samp{<target>} may also be omitted, but we recommend
43889 including it; if future versions of @value{GDBN} use an incompatible
43890 revision of @file{gdb-target.dtd}, they will detect and report
43891 the version mismatch.
43893 @subsection Inclusion
43894 @cindex target descriptions, inclusion
43897 @cindex <xi:include>
43900 It can sometimes be valuable to split a target description up into
43901 several different annexes, either for organizational purposes, or to
43902 share files between different possible target descriptions. You can
43903 divide a description into multiple files by replacing any element of
43904 the target description with an inclusion directive of the form:
43907 <xi:include href="@var{document}"/>
43911 When @value{GDBN} encounters an element of this form, it will retrieve
43912 the named XML @var{document}, and replace the inclusion directive with
43913 the contents of that document. If the current description was read
43914 using @samp{qXfer}, then so will be the included document;
43915 @var{document} will be interpreted as the name of an annex. If the
43916 current description was read from a file, @value{GDBN} will look for
43917 @var{document} as a file in the same directory where it found the
43918 original description.
43920 @subsection Architecture
43921 @cindex <architecture>
43923 An @samp{<architecture>} element has this form:
43926 <architecture>@var{arch}</architecture>
43929 @var{arch} is one of the architectures from the set accepted by
43930 @code{set architecture} (@pxref{Targets, ,Specifying a Debugging Target}).
43933 @cindex @code{<osabi>}
43935 This optional field was introduced in @value{GDBN} version 7.0.
43936 Previous versions of @value{GDBN} ignore it.
43938 An @samp{<osabi>} element has this form:
43941 <osabi>@var{abi-name}</osabi>
43944 @var{abi-name} is an OS ABI name from the same selection accepted by
43945 @w{@code{set osabi}} (@pxref{ABI, ,Configuring the Current ABI}).
43947 @subsection Compatible Architecture
43948 @cindex @code{<compatible>}
43950 This optional field was introduced in @value{GDBN} version 7.0.
43951 Previous versions of @value{GDBN} ignore it.
43953 A @samp{<compatible>} element has this form:
43956 <compatible>@var{arch}</compatible>
43959 @var{arch} is one of the architectures from the set accepted by
43960 @code{set architecture} (@pxref{Targets, ,Specifying a Debugging Target}).
43962 A @samp{<compatible>} element is used to specify that the target
43963 is able to run binaries in some other than the main target architecture
43964 given by the @samp{<architecture>} element. For example, on the
43965 Cell Broadband Engine, the main architecture is @code{powerpc:common}
43966 or @code{powerpc:common64}, but the system is able to run binaries
43967 in the @code{spu} architecture as well. The way to describe this
43968 capability with @samp{<compatible>} is as follows:
43971 <architecture>powerpc:common</architecture>
43972 <compatible>spu</compatible>
43975 @subsection Features
43978 Each @samp{<feature>} describes some logical portion of the target
43979 system. Features are currently used to describe available CPU
43980 registers and the types of their contents. A @samp{<feature>} element
43984 <feature name="@var{name}">
43985 @r{[}@var{type}@dots{}@r{]}
43991 Each feature's name should be unique within the description. The name
43992 of a feature does not matter unless @value{GDBN} has some special
43993 knowledge of the contents of that feature; if it does, the feature
43994 should have its standard name. @xref{Standard Target Features}.
43998 Any register's value is a collection of bits which @value{GDBN} must
43999 interpret. The default interpretation is a two's complement integer,
44000 but other types can be requested by name in the register description.
44001 Some predefined types are provided by @value{GDBN} (@pxref{Predefined
44002 Target Types}), and the description can define additional composite
44005 Each type element must have an @samp{id} attribute, which gives
44006 a unique (within the containing @samp{<feature>}) name to the type.
44007 Types must be defined before they are used.
44010 Some targets offer vector registers, which can be treated as arrays
44011 of scalar elements. These types are written as @samp{<vector>} elements,
44012 specifying the array element type, @var{type}, and the number of elements,
44016 <vector id="@var{id}" type="@var{type}" count="@var{count}"/>
44020 If a register's value is usefully viewed in multiple ways, define it
44021 with a union type containing the useful representations. The
44022 @samp{<union>} element contains one or more @samp{<field>} elements,
44023 each of which has a @var{name} and a @var{type}:
44026 <union id="@var{id}">
44027 <field name="@var{name}" type="@var{type}"/>
44034 If a register's value is composed from several separate values, define
44035 it with either a structure type or a flags type.
44036 A flags type may only contain bitfields.
44037 A structure type may either contain only bitfields or contain no bitfields.
44038 If the value contains only bitfields, its total size in bytes must be
44041 Non-bitfield values have a @var{name} and @var{type}.
44044 <struct id="@var{id}">
44045 <field name="@var{name}" type="@var{type}"/>
44050 Both @var{name} and @var{type} values are required.
44051 No implicit padding is added.
44053 Bitfield values have a @var{name}, @var{start}, @var{end} and @var{type}.
44056 <struct id="@var{id}" size="@var{size}">
44057 <field name="@var{name}" start="@var{start}" end="@var{end}" type="@var{type}"/>
44063 <flags id="@var{id}" size="@var{size}">
44064 <field name="@var{name}" start="@var{start}" end="@var{end}" type="@var{type}"/>
44069 The @var{name} value is required.
44070 Bitfield values may be named with the empty string, @samp{""},
44071 in which case the field is ``filler'' and its value is not printed.
44072 Not all bits need to be specified, so ``filler'' fields are optional.
44074 The @var{start} and @var{end} values are required, and @var{type}
44076 The field's @var{start} must be less than or equal to its @var{end},
44077 and zero represents the least significant bit.
44079 The default value of @var{type} is @code{bool} for single bit fields,
44080 and an unsigned integer otherwise.
44082 Which to choose? Structures or flags?
44084 Registers defined with @samp{flags} have these advantages over
44085 defining them with @samp{struct}:
44089 Arithmetic may be performed on them as if they were integers.
44091 They are printed in a more readable fashion.
44094 Registers defined with @samp{struct} have one advantage over
44095 defining them with @samp{flags}:
44099 One can fetch individual fields like in @samp{C}.
44102 (gdb) print $my_struct_reg.field3
44108 @subsection Registers
44111 Each register is represented as an element with this form:
44114 <reg name="@var{name}"
44115 bitsize="@var{size}"
44116 @r{[}regnum="@var{num}"@r{]}
44117 @r{[}save-restore="@var{save-restore}"@r{]}
44118 @r{[}type="@var{type}"@r{]}
44119 @r{[}group="@var{group}"@r{]}/>
44123 The components are as follows:
44128 The register's name; it must be unique within the target description.
44131 The register's size, in bits.
44134 The register's number. If omitted, a register's number is one greater
44135 than that of the previous register (either in the current feature or in
44136 a preceding feature); the first register in the target description
44137 defaults to zero. This register number is used to read or write
44138 the register; e.g.@: it is used in the remote @code{p} and @code{P}
44139 packets, and registers appear in the @code{g} and @code{G} packets
44140 in order of increasing register number.
44143 Whether the register should be preserved across inferior function
44144 calls; this must be either @code{yes} or @code{no}. The default is
44145 @code{yes}, which is appropriate for most registers except for
44146 some system control registers; this is not related to the target's
44150 The type of the register. It may be a predefined type, a type
44151 defined in the current feature, or one of the special types @code{int}
44152 and @code{float}. @code{int} is an integer type of the correct size
44153 for @var{bitsize}, and @code{float} is a floating point type (in the
44154 architecture's normal floating point format) of the correct size for
44155 @var{bitsize}. The default is @code{int}.
44158 The register group to which this register belongs. It can be one of the
44159 standard register groups @code{general}, @code{float}, @code{vector} or an
44160 arbitrary string. Group names should be limited to alphanumeric characters.
44161 If a group name is made up of multiple words the words may be separated by
44162 hyphens; e.g.@: @code{special-group} or @code{ultra-special-group}. If no
44163 @var{group} is specified, @value{GDBN} will not display the register in
44164 @code{info registers}.
44168 @node Predefined Target Types
44169 @section Predefined Target Types
44170 @cindex target descriptions, predefined types
44172 Type definitions in the self-description can build up composite types
44173 from basic building blocks, but can not define fundamental types. Instead,
44174 standard identifiers are provided by @value{GDBN} for the fundamental
44175 types. The currently supported types are:
44180 Boolean type, occupying a single bit.
44188 Signed integer types holding the specified number of bits.
44196 Unsigned integer types holding the specified number of bits.
44200 Pointers to unspecified code and data. The program counter and
44201 any dedicated return address register may be marked as code
44202 pointers; printing a code pointer converts it into a symbolic
44203 address. The stack pointer and any dedicated address registers
44204 may be marked as data pointers.
44207 Single precision IEEE floating point.
44210 Double precision IEEE floating point.
44213 The 12-byte extended precision format used by ARM FPA registers.
44216 The 10-byte extended precision format used by x87 registers.
44219 32bit @sc{eflags} register used by x86.
44222 32bit @sc{mxcsr} register used by x86.
44226 @node Enum Target Types
44227 @section Enum Target Types
44228 @cindex target descriptions, enum types
44230 Enum target types are useful in @samp{struct} and @samp{flags}
44231 register descriptions. @xref{Target Description Format}.
44233 Enum types have a name, size and a list of name/value pairs.
44236 <enum id="@var{id}" size="@var{size}">
44237 <evalue name="@var{name}" value="@var{value}"/>
44242 Enums must be defined before they are used.
44245 <enum id="levels_type" size="4">
44246 <evalue name="low" value="0"/>
44247 <evalue name="high" value="1"/>
44249 <flags id="flags_type" size="4">
44250 <field name="X" start="0"/>
44251 <field name="LEVEL" start="1" end="1" type="levels_type"/>
44253 <reg name="flags" bitsize="32" type="flags_type"/>
44256 Given that description, a value of 3 for the @samp{flags} register
44257 would be printed as:
44260 (gdb) info register flags
44261 flags 0x3 [ X LEVEL=high ]
44264 @node Standard Target Features
44265 @section Standard Target Features
44266 @cindex target descriptions, standard features
44268 A target description must contain either no registers or all the
44269 target's registers. If the description contains no registers, then
44270 @value{GDBN} will assume a default register layout, selected based on
44271 the architecture. If the description contains any registers, the
44272 default layout will not be used; the standard registers must be
44273 described in the target description, in such a way that @value{GDBN}
44274 can recognize them.
44276 This is accomplished by giving specific names to feature elements
44277 which contain standard registers. @value{GDBN} will look for features
44278 with those names and verify that they contain the expected registers;
44279 if any known feature is missing required registers, or if any required
44280 feature is missing, @value{GDBN} will reject the target
44281 description. You can add additional registers to any of the
44282 standard features --- @value{GDBN} will display them just as if
44283 they were added to an unrecognized feature.
44285 This section lists the known features and their expected contents.
44286 Sample XML documents for these features are included in the
44287 @value{GDBN} source tree, in the directory @file{gdb/features}.
44289 Names recognized by @value{GDBN} should include the name of the
44290 company or organization which selected the name, and the overall
44291 architecture to which the feature applies; so e.g.@: the feature
44292 containing ARM core registers is named @samp{org.gnu.gdb.arm.core}.
44294 The names of registers are not case sensitive for the purpose
44295 of recognizing standard features, but @value{GDBN} will only display
44296 registers using the capitalization used in the description.
44299 * AArch64 Features::
44303 * MicroBlaze Features::
44307 * Nios II Features::
44308 * OpenRISC 1000 Features::
44309 * PowerPC Features::
44310 * RISC-V Features::
44312 * S/390 and System z Features::
44318 @node AArch64 Features
44319 @subsection AArch64 Features
44320 @cindex target descriptions, AArch64 features
44322 The @samp{org.gnu.gdb.aarch64.core} feature is required for AArch64
44323 targets. It should contain registers @samp{x0} through @samp{x30},
44324 @samp{sp}, @samp{pc}, and @samp{cpsr}.
44326 The @samp{org.gnu.gdb.aarch64.fpu} feature is optional. If present,
44327 it should contain registers @samp{v0} through @samp{v31}, @samp{fpsr},
44330 The @samp{org.gnu.gdb.aarch64.sve} feature is optional. If present,
44331 it should contain registers @samp{z0} through @samp{z31}, @samp{p0}
44332 through @samp{p15}, @samp{ffr} and @samp{vg}.
44334 The @samp{org.gnu.gdb.aarch64.pauth} feature is optional. If present,
44335 it should contain registers @samp{pauth_dmask} and @samp{pauth_cmask}.
44338 @subsection ARC Features
44339 @cindex target descriptions, ARC Features
44341 ARC processors are highly configurable, so even core registers and their number
44342 are not completely predetermined. In addition flags and PC registers which are
44343 important to @value{GDBN} are not ``core'' registers in ARC. It is required
44344 that one of the core registers features is present.
44345 @samp{org.gnu.gdb.arc.aux-minimal} feature is mandatory.
44347 The @samp{org.gnu.gdb.arc.core.v2} feature is required for ARC EM and ARC HS
44348 targets with a normal register file. It should contain registers @samp{r0}
44349 through @samp{r25}, @samp{gp}, @samp{fp}, @samp{sp}, @samp{r30}, @samp{blink},
44350 @samp{lp_count} and @samp{pcl}. This feature may contain register @samp{ilink}
44351 and any of extension core registers @samp{r32} through @samp{r59/acch}.
44352 @samp{ilink} and extension core registers are not available to read/write, when
44353 debugging GNU/Linux applications, thus @samp{ilink} is made optional.
44355 The @samp{org.gnu.gdb.arc.core-reduced.v2} feature is required for ARC EM and
44356 ARC HS targets with a reduced register file. It should contain registers
44357 @samp{r0} through @samp{r3}, @samp{r10} through @samp{r15}, @samp{gp},
44358 @samp{fp}, @samp{sp}, @samp{r30}, @samp{blink}, @samp{lp_count} and @samp{pcl}.
44359 This feature may contain register @samp{ilink} and any of extension core
44360 registers @samp{r32} through @samp{r59/acch}.
44362 The @samp{org.gnu.gdb.arc.core.arcompact} feature is required for ARCompact
44363 targets with a normal register file. It should contain registers @samp{r0}
44364 through @samp{r25}, @samp{gp}, @samp{fp}, @samp{sp}, @samp{r30}, @samp{blink},
44365 @samp{lp_count} and @samp{pcl}. This feature may contain registers
44366 @samp{ilink1}, @samp{ilink2} and any of extension core registers @samp{r32}
44367 through @samp{r59/acch}. @samp{ilink1} and @samp{ilink2} and extension core
44368 registers are not available when debugging GNU/Linux applications. The only
44369 difference with @samp{org.gnu.gdb.arc.core.v2} feature is in the names of
44370 @samp{ilink1} and @samp{ilink2} registers and that @samp{r30} is mandatory in
44371 ARC v2, but @samp{ilink2} is optional on ARCompact.
44373 The @samp{org.gnu.gdb.arc.aux-minimal} feature is required for all ARC
44374 targets. It should contain registers @samp{pc} and @samp{status32}.
44377 @subsection ARM Features
44378 @cindex target descriptions, ARM features
44380 The @samp{org.gnu.gdb.arm.core} feature is required for non-M-profile
44382 It should contain registers @samp{r0} through @samp{r13}, @samp{sp},
44383 @samp{lr}, @samp{pc}, and @samp{cpsr}.
44385 For M-profile targets (e.g. Cortex-M3), the @samp{org.gnu.gdb.arm.core}
44386 feature is replaced by @samp{org.gnu.gdb.arm.m-profile}. It should contain
44387 registers @samp{r0} through @samp{r13}, @samp{sp}, @samp{lr}, @samp{pc},
44390 The @samp{org.gnu.gdb.arm.fpa} feature is optional. If present, it
44391 should contain registers @samp{f0} through @samp{f7} and @samp{fps}.
44393 The @samp{org.gnu.gdb.xscale.iwmmxt} feature is optional. If present,
44394 it should contain at least registers @samp{wR0} through @samp{wR15} and
44395 @samp{wCGR0} through @samp{wCGR3}. The @samp{wCID}, @samp{wCon},
44396 @samp{wCSSF}, and @samp{wCASF} registers are optional.
44398 The @samp{org.gnu.gdb.arm.vfp} feature is optional. If present, it
44399 should contain at least registers @samp{d0} through @samp{d15}. If
44400 they are present, @samp{d16} through @samp{d31} should also be included.
44401 @value{GDBN} will synthesize the single-precision registers from
44402 halves of the double-precision registers.
44404 The @samp{org.gnu.gdb.arm.neon} feature is optional. It does not
44405 need to contain registers; it instructs @value{GDBN} to display the
44406 VFP double-precision registers as vectors and to synthesize the
44407 quad-precision registers from pairs of double-precision registers.
44408 If this feature is present, @samp{org.gnu.gdb.arm.vfp} must also
44409 be present and include 32 double-precision registers.
44411 @node i386 Features
44412 @subsection i386 Features
44413 @cindex target descriptions, i386 features
44415 The @samp{org.gnu.gdb.i386.core} feature is required for i386/amd64
44416 targets. It should describe the following registers:
44420 @samp{eax} through @samp{edi} plus @samp{eip} for i386
44422 @samp{rax} through @samp{r15} plus @samp{rip} for amd64
44424 @samp{eflags}, @samp{cs}, @samp{ss}, @samp{ds}, @samp{es},
44425 @samp{fs}, @samp{gs}
44427 @samp{st0} through @samp{st7}
44429 @samp{fctrl}, @samp{fstat}, @samp{ftag}, @samp{fiseg}, @samp{fioff},
44430 @samp{foseg}, @samp{fooff} and @samp{fop}
44433 The register sets may be different, depending on the target.
44435 The @samp{org.gnu.gdb.i386.sse} feature is optional. It should
44436 describe registers:
44440 @samp{xmm0} through @samp{xmm7} for i386
44442 @samp{xmm0} through @samp{xmm15} for amd64
44447 The @samp{org.gnu.gdb.i386.avx} feature is optional and requires the
44448 @samp{org.gnu.gdb.i386.sse} feature. It should
44449 describe the upper 128 bits of @sc{ymm} registers:
44453 @samp{ymm0h} through @samp{ymm7h} for i386
44455 @samp{ymm0h} through @samp{ymm15h} for amd64
44458 The @samp{org.gnu.gdb.i386.mpx} is an optional feature representing Intel
44459 Memory Protection Extension (MPX). It should describe the following registers:
44463 @samp{bnd0raw} through @samp{bnd3raw} for i386 and amd64.
44465 @samp{bndcfgu} and @samp{bndstatus} for i386 and amd64.
44468 The @samp{org.gnu.gdb.i386.linux} feature is optional. It should
44469 describe a single register, @samp{orig_eax}.
44471 The @samp{org.gnu.gdb.i386.segments} feature is optional. It should
44472 describe two system registers: @samp{fs_base} and @samp{gs_base}.
44474 The @samp{org.gnu.gdb.i386.avx512} feature is optional and requires the
44475 @samp{org.gnu.gdb.i386.avx} feature. It should
44476 describe additional @sc{xmm} registers:
44480 @samp{xmm16h} through @samp{xmm31h}, only valid for amd64.
44483 It should describe the upper 128 bits of additional @sc{ymm} registers:
44487 @samp{ymm16h} through @samp{ymm31h}, only valid for amd64.
44491 describe the upper 256 bits of @sc{zmm} registers:
44495 @samp{zmm0h} through @samp{zmm7h} for i386.
44497 @samp{zmm0h} through @samp{zmm15h} for amd64.
44501 describe the additional @sc{zmm} registers:
44505 @samp{zmm16h} through @samp{zmm31h}, only valid for amd64.
44508 The @samp{org.gnu.gdb.i386.pkeys} feature is optional. It should
44509 describe a single register, @samp{pkru}. It is a 32-bit register
44510 valid for i386 and amd64.
44512 @node MicroBlaze Features
44513 @subsection MicroBlaze Features
44514 @cindex target descriptions, MicroBlaze features
44516 The @samp{org.gnu.gdb.microblaze.core} feature is required for MicroBlaze
44517 targets. It should contain registers @samp{r0} through @samp{r31},
44518 @samp{rpc}, @samp{rmsr}, @samp{rear}, @samp{resr}, @samp{rfsr}, @samp{rbtr},
44519 @samp{rpvr}, @samp{rpvr1} through @samp{rpvr11}, @samp{redr}, @samp{rpid},
44520 @samp{rzpr}, @samp{rtlbx}, @samp{rtlbsx}, @samp{rtlblo}, and @samp{rtlbhi}.
44522 The @samp{org.gnu.gdb.microblaze.stack-protect} feature is optional.
44523 If present, it should contain registers @samp{rshr} and @samp{rslr}
44525 @node MIPS Features
44526 @subsection @acronym{MIPS} Features
44527 @cindex target descriptions, @acronym{MIPS} features
44529 The @samp{org.gnu.gdb.mips.cpu} feature is required for @acronym{MIPS} targets.
44530 It should contain registers @samp{r0} through @samp{r31}, @samp{lo},
44531 @samp{hi}, and @samp{pc}. They may be 32-bit or 64-bit depending
44534 The @samp{org.gnu.gdb.mips.cp0} feature is also required. It should
44535 contain at least the @samp{status}, @samp{badvaddr}, and @samp{cause}
44536 registers. They may be 32-bit or 64-bit depending on the target.
44538 The @samp{org.gnu.gdb.mips.fpu} feature is currently required, though
44539 it may be optional in a future version of @value{GDBN}. It should
44540 contain registers @samp{f0} through @samp{f31}, @samp{fcsr}, and
44541 @samp{fir}. They may be 32-bit or 64-bit depending on the target.
44543 The @samp{org.gnu.gdb.mips.dsp} feature is optional. It should
44544 contain registers @samp{hi1} through @samp{hi3}, @samp{lo1} through
44545 @samp{lo3}, and @samp{dspctl}. The @samp{dspctl} register should
44546 be 32-bit and the rest may be 32-bit or 64-bit depending on the target.
44548 The @samp{org.gnu.gdb.mips.linux} feature is optional. It should
44549 contain a single register, @samp{restart}, which is used by the
44550 Linux kernel to control restartable syscalls.
44552 @node M68K Features
44553 @subsection M68K Features
44554 @cindex target descriptions, M68K features
44557 @item @samp{org.gnu.gdb.m68k.core}
44558 @itemx @samp{org.gnu.gdb.coldfire.core}
44559 @itemx @samp{org.gnu.gdb.fido.core}
44560 One of those features must be always present.
44561 The feature that is present determines which flavor of m68k is
44562 used. The feature that is present should contain registers
44563 @samp{d0} through @samp{d7}, @samp{a0} through @samp{a5}, @samp{fp},
44564 @samp{sp}, @samp{ps} and @samp{pc}.
44566 @item @samp{org.gnu.gdb.coldfire.fp}
44567 This feature is optional. If present, it should contain registers
44568 @samp{fp0} through @samp{fp7}, @samp{fpcontrol}, @samp{fpstatus} and
44572 @node NDS32 Features
44573 @subsection NDS32 Features
44574 @cindex target descriptions, NDS32 features
44576 The @samp{org.gnu.gdb.nds32.core} feature is required for NDS32
44577 targets. It should contain at least registers @samp{r0} through
44578 @samp{r10}, @samp{r15}, @samp{fp}, @samp{gp}, @samp{lp}, @samp{sp},
44581 The @samp{org.gnu.gdb.nds32.fpu} feature is optional. If present,
44582 it should contain 64-bit double-precision floating-point registers
44583 @samp{fd0} through @emph{fdN}, which should be @samp{fd3}, @samp{fd7},
44584 @samp{fd15}, or @samp{fd31} based on the FPU configuration implemented.
44586 @emph{Note:} The first sixteen 64-bit double-precision floating-point
44587 registers are overlapped with the thirty-two 32-bit single-precision
44588 floating-point registers. The 32-bit single-precision registers, if
44589 not being listed explicitly, will be synthesized from halves of the
44590 overlapping 64-bit double-precision registers. Listing 32-bit
44591 single-precision registers explicitly is deprecated, and the
44592 support to it could be totally removed some day.
44594 @node Nios II Features
44595 @subsection Nios II Features
44596 @cindex target descriptions, Nios II features
44598 The @samp{org.gnu.gdb.nios2.cpu} feature is required for Nios II
44599 targets. It should contain the 32 core registers (@samp{zero},
44600 @samp{at}, @samp{r2} through @samp{r23}, @samp{et} through @samp{ra}),
44601 @samp{pc}, and the 16 control registers (@samp{status} through
44604 @node OpenRISC 1000 Features
44605 @subsection Openrisc 1000 Features
44606 @cindex target descriptions, OpenRISC 1000 features
44608 The @samp{org.gnu.gdb.or1k.group0} feature is required for OpenRISC 1000
44609 targets. It should contain the 32 general purpose registers (@samp{r0}
44610 through @samp{r31}), @samp{ppc}, @samp{npc} and @samp{sr}.
44612 @node PowerPC Features
44613 @subsection PowerPC Features
44614 @cindex target descriptions, PowerPC features
44616 The @samp{org.gnu.gdb.power.core} feature is required for PowerPC
44617 targets. It should contain registers @samp{r0} through @samp{r31},
44618 @samp{pc}, @samp{msr}, @samp{cr}, @samp{lr}, @samp{ctr}, and
44619 @samp{xer}. They may be 32-bit or 64-bit depending on the target.
44621 The @samp{org.gnu.gdb.power.fpu} feature is optional. It should
44622 contain registers @samp{f0} through @samp{f31} and @samp{fpscr}.
44624 The @samp{org.gnu.gdb.power.altivec} feature is optional. It should
44625 contain registers @samp{vr0} through @samp{vr31}, @samp{vscr}, and
44626 @samp{vrsave}. @value{GDBN} will define pseudo-registers @samp{v0}
44627 through @samp{v31} as aliases for the corresponding @samp{vrX}
44630 The @samp{org.gnu.gdb.power.vsx} feature is optional. It should
44631 contain registers @samp{vs0h} through @samp{vs31h}. @value{GDBN} will
44632 combine these registers with the floating point registers (@samp{f0}
44633 through @samp{f31}) and the altivec registers (@samp{vr0} through
44634 @samp{vr31}) to present the 128-bit wide registers @samp{vs0} through
44635 @samp{vs63}, the set of vector-scalar registers for POWER7.
44636 Therefore, this feature requires both @samp{org.gnu.gdb.power.fpu} and
44637 @samp{org.gnu.gdb.power.altivec}.
44639 The @samp{org.gnu.gdb.power.spe} feature is optional. It should
44640 contain registers @samp{ev0h} through @samp{ev31h}, @samp{acc}, and
44641 @samp{spefscr}. SPE targets should provide 32-bit registers in
44642 @samp{org.gnu.gdb.power.core} and provide the upper halves in
44643 @samp{ev0h} through @samp{ev31h}. @value{GDBN} will combine
44644 these to present registers @samp{ev0} through @samp{ev31} to the
44647 The @samp{org.gnu.gdb.power.ppr} feature is optional. It should
44648 contain the 64-bit register @samp{ppr}.
44650 The @samp{org.gnu.gdb.power.dscr} feature is optional. It should
44651 contain the 64-bit register @samp{dscr}.
44653 The @samp{org.gnu.gdb.power.tar} feature is optional. It should
44654 contain the 64-bit register @samp{tar}.
44656 The @samp{org.gnu.gdb.power.ebb} feature is optional. It should
44657 contain registers @samp{bescr}, @samp{ebbhr} and @samp{ebbrr}, all
44660 The @samp{org.gnu.gdb.power.linux.pmu} feature is optional. It should
44661 contain registers @samp{mmcr0}, @samp{mmcr2}, @samp{siar}, @samp{sdar}
44662 and @samp{sier}, all 64-bit wide. This is the subset of the isa 2.07
44663 server PMU registers provided by @sc{gnu}/Linux.
44665 The @samp{org.gnu.gdb.power.htm.spr} feature is optional. It should
44666 contain registers @samp{tfhar}, @samp{texasr} and @samp{tfiar}, all
44669 The @samp{org.gnu.gdb.power.htm.core} feature is optional. It should
44670 contain the checkpointed general-purpose registers @samp{cr0} through
44671 @samp{cr31}, as well as the checkpointed registers @samp{clr} and
44672 @samp{cctr}. These registers may all be either 32-bit or 64-bit
44673 depending on the target. It should also contain the checkpointed
44674 registers @samp{ccr} and @samp{cxer}, which should both be 32-bit
44677 The @samp{org.gnu.gdb.power.htm.fpu} feature is optional. It should
44678 contain the checkpointed 64-bit floating-point registers @samp{cf0}
44679 through @samp{cf31}, as well as the checkpointed 64-bit register
44682 The @samp{org.gnu.gdb.power.htm.altivec} feature is optional. It
44683 should contain the checkpointed altivec registers @samp{cvr0} through
44684 @samp{cvr31}, all 128-bit wide. It should also contain the
44685 checkpointed registers @samp{cvscr} and @samp{cvrsave}, both 32-bit
44688 The @samp{org.gnu.gdb.power.htm.vsx} feature is optional. It should
44689 contain registers @samp{cvs0h} through @samp{cvs31h}. @value{GDBN}
44690 will combine these registers with the checkpointed floating point
44691 registers (@samp{cf0} through @samp{cf31}) and the checkpointed
44692 altivec registers (@samp{cvr0} through @samp{cvr31}) to present the
44693 128-bit wide checkpointed vector-scalar registers @samp{cvs0} through
44694 @samp{cvs63}. Therefore, this feature requires both
44695 @samp{org.gnu.gdb.power.htm.altivec} and
44696 @samp{org.gnu.gdb.power.htm.fpu}.
44698 The @samp{org.gnu.gdb.power.htm.ppr} feature is optional. It should
44699 contain the 64-bit checkpointed register @samp{cppr}.
44701 The @samp{org.gnu.gdb.power.htm.dscr} feature is optional. It should
44702 contain the 64-bit checkpointed register @samp{cdscr}.
44704 The @samp{org.gnu.gdb.power.htm.tar} feature is optional. It should
44705 contain the 64-bit checkpointed register @samp{ctar}.
44708 @node RISC-V Features
44709 @subsection RISC-V Features
44710 @cindex target descriptions, RISC-V Features
44712 The @samp{org.gnu.gdb.riscv.cpu} feature is required for RISC-V
44713 targets. It should contain the registers @samp{x0} through
44714 @samp{x31}, and @samp{pc}. Either the architectural names (@samp{x0},
44715 @samp{x1}, etc) can be used, or the ABI names (@samp{zero}, @samp{ra},
44718 The @samp{org.gnu.gdb.riscv.fpu} feature is optional. If present, it
44719 should contain registers @samp{f0} through @samp{f31}, @samp{fflags},
44720 @samp{frm}, and @samp{fcsr}. As with the cpu feature, either the
44721 architectural register names, or the ABI names can be used.
44723 The @samp{org.gnu.gdb.riscv.virtual} feature is optional. If present,
44724 it should contain registers that are not backed by real registers on
44725 the target, but are instead virtual, where the register value is
44726 derived from other target state. In many ways these are like
44727 @value{GDBN}s pseudo-registers, except implemented by the target.
44728 Currently the only register expected in this set is the one byte
44729 @samp{priv} register that contains the target's privilege level in the
44730 least significant two bits.
44732 The @samp{org.gnu.gdb.riscv.csr} feature is optional. If present, it
44733 should contain all of the target's standard CSRs. Standard CSRs are
44734 those defined in the RISC-V specification documents. There is some
44735 overlap between this feature and the fpu feature; the @samp{fflags},
44736 @samp{frm}, and @samp{fcsr} registers could be in either feature. The
44737 expectation is that these registers will be in the fpu feature if the
44738 target has floating point hardware, but can be moved into the csr
44739 feature if the target has the floating point control registers, but no
44740 other floating point hardware.
44743 @subsection RX Features
44744 @cindex target descriptions, RX Features
44746 The @samp{org.gnu.gdb.rx.core} feature is required for RX
44747 targets. It should contain the registers @samp{r0} through
44748 @samp{r15}, @samp{usp}, @samp{isp}, @samp{psw}, @samp{pc}, @samp{intb},
44749 @samp{bpsw}, @samp{bpc}, @samp{fintv}, @samp{fpsw}, and @samp{acc}.
44751 @node S/390 and System z Features
44752 @subsection S/390 and System z Features
44753 @cindex target descriptions, S/390 features
44754 @cindex target descriptions, System z features
44756 The @samp{org.gnu.gdb.s390.core} feature is required for S/390 and
44757 System z targets. It should contain the PSW and the 16 general
44758 registers. In particular, System z targets should provide the 64-bit
44759 registers @samp{pswm}, @samp{pswa}, and @samp{r0} through @samp{r15}.
44760 S/390 targets should provide the 32-bit versions of these registers.
44761 A System z target that runs in 31-bit addressing mode should provide
44762 32-bit versions of @samp{pswm} and @samp{pswa}, as well as the general
44763 register's upper halves @samp{r0h} through @samp{r15h}, and their
44764 lower halves @samp{r0l} through @samp{r15l}.
44766 The @samp{org.gnu.gdb.s390.fpr} feature is required. It should
44767 contain the 64-bit registers @samp{f0} through @samp{f15}, and
44770 The @samp{org.gnu.gdb.s390.acr} feature is required. It should
44771 contain the 32-bit registers @samp{acr0} through @samp{acr15}.
44773 The @samp{org.gnu.gdb.s390.linux} feature is optional. It should
44774 contain the register @samp{orig_r2}, which is 64-bit wide on System z
44775 targets and 32-bit otherwise. In addition, the feature may contain
44776 the @samp{last_break} register, whose width depends on the addressing
44777 mode, as well as the @samp{system_call} register, which is always
44780 The @samp{org.gnu.gdb.s390.tdb} feature is optional. It should
44781 contain the 64-bit registers @samp{tdb0}, @samp{tac}, @samp{tct},
44782 @samp{atia}, and @samp{tr0} through @samp{tr15}.
44784 The @samp{org.gnu.gdb.s390.vx} feature is optional. It should contain
44785 64-bit wide registers @samp{v0l} through @samp{v15l}, which will be
44786 combined by @value{GDBN} with the floating point registers @samp{f0}
44787 through @samp{f15} to present the 128-bit wide vector registers
44788 @samp{v0} through @samp{v15}. In addition, this feature should
44789 contain the 128-bit wide vector registers @samp{v16} through
44792 The @samp{org.gnu.gdb.s390.gs} feature is optional. It should contain
44793 the 64-bit wide guarded-storage-control registers @samp{gsd},
44794 @samp{gssm}, and @samp{gsepla}.
44796 The @samp{org.gnu.gdb.s390.gsbc} feature is optional. It should contain
44797 the 64-bit wide guarded-storage broadcast control registers
44798 @samp{bc_gsd}, @samp{bc_gssm}, and @samp{bc_gsepla}.
44800 @node Sparc Features
44801 @subsection Sparc Features
44802 @cindex target descriptions, sparc32 features
44803 @cindex target descriptions, sparc64 features
44804 The @samp{org.gnu.gdb.sparc.cpu} feature is required for sparc32/sparc64
44805 targets. It should describe the following registers:
44809 @samp{g0} through @samp{g7}
44811 @samp{o0} through @samp{o7}
44813 @samp{l0} through @samp{l7}
44815 @samp{i0} through @samp{i7}
44818 They may be 32-bit or 64-bit depending on the target.
44820 Also the @samp{org.gnu.gdb.sparc.fpu} feature is required for sparc32/sparc64
44821 targets. It should describe the following registers:
44825 @samp{f0} through @samp{f31}
44827 @samp{f32} through @samp{f62} for sparc64
44830 The @samp{org.gnu.gdb.sparc.cp0} feature is required for sparc32/sparc64
44831 targets. It should describe the following registers:
44835 @samp{y}, @samp{psr}, @samp{wim}, @samp{tbr}, @samp{pc}, @samp{npc},
44836 @samp{fsr}, and @samp{csr} for sparc32
44838 @samp{pc}, @samp{npc}, @samp{state}, @samp{fsr}, @samp{fprs}, and @samp{y}
44842 @node TIC6x Features
44843 @subsection TMS320C6x Features
44844 @cindex target descriptions, TIC6x features
44845 @cindex target descriptions, TMS320C6x features
44846 The @samp{org.gnu.gdb.tic6x.core} feature is required for TMS320C6x
44847 targets. It should contain registers @samp{A0} through @samp{A15},
44848 registers @samp{B0} through @samp{B15}, @samp{CSR} and @samp{PC}.
44850 The @samp{org.gnu.gdb.tic6x.gp} feature is optional. It should
44851 contain registers @samp{A16} through @samp{A31} and @samp{B16}
44852 through @samp{B31}.
44854 The @samp{org.gnu.gdb.tic6x.c6xp} feature is optional. It should
44855 contain registers @samp{TSR}, @samp{ILC} and @samp{RILC}.
44857 @node Operating System Information
44858 @appendix Operating System Information
44859 @cindex operating system information
44865 Users of @value{GDBN} often wish to obtain information about the state of
44866 the operating system running on the target---for example the list of
44867 processes, or the list of open files. This section describes the
44868 mechanism that makes it possible. This mechanism is similar to the
44869 target features mechanism (@pxref{Target Descriptions}), but focuses
44870 on a different aspect of target.
44872 Operating system information is retrieved from the target via the
44873 remote protocol, using @samp{qXfer} requests (@pxref{qXfer osdata
44874 read}). The object name in the request should be @samp{osdata}, and
44875 the @var{annex} identifies the data to be fetched.
44878 @appendixsection Process list
44879 @cindex operating system information, process list
44881 When requesting the process list, the @var{annex} field in the
44882 @samp{qXfer} request should be @samp{processes}. The returned data is
44883 an XML document. The formal syntax of this document is defined in
44884 @file{gdb/features/osdata.dtd}.
44886 An example document is:
44889 <?xml version="1.0"?>
44890 <!DOCTYPE target SYSTEM "osdata.dtd">
44891 <osdata type="processes">
44893 <column name="pid">1</column>
44894 <column name="user">root</column>
44895 <column name="command">/sbin/init</column>
44896 <column name="cores">1,2,3</column>
44901 Each item should include a column whose name is @samp{pid}. The value
44902 of that column should identify the process on the target. The
44903 @samp{user} and @samp{command} columns are optional, and will be
44904 displayed by @value{GDBN}. The @samp{cores} column, if present,
44905 should contain a comma-separated list of cores that this process
44906 is running on. Target may provide additional columns,
44907 which @value{GDBN} currently ignores.
44909 @node Trace File Format
44910 @appendix Trace File Format
44911 @cindex trace file format
44913 The trace file comes in three parts: a header, a textual description
44914 section, and a trace frame section with binary data.
44916 The header has the form @code{\x7fTRACE0\n}. The first byte is
44917 @code{0x7f} so as to indicate that the file contains binary data,
44918 while the @code{0} is a version number that may have different values
44921 The description section consists of multiple lines of @sc{ascii} text
44922 separated by newline characters (@code{0xa}). The lines may include a
44923 variety of optional descriptive or context-setting information, such
44924 as tracepoint definitions or register set size. @value{GDBN} will
44925 ignore any line that it does not recognize. An empty line marks the end
44930 Specifies the size of a register block in bytes. This is equal to the
44931 size of a @code{g} packet payload in the remote protocol. @var{size}
44932 is an ascii decimal number. There should be only one such line in
44933 a single trace file.
44935 @item status @var{status}
44936 Trace status. @var{status} has the same format as a @code{qTStatus}
44937 remote packet reply. There should be only one such line in a single trace
44940 @item tp @var{payload}
44941 Tracepoint definition. The @var{payload} has the same format as
44942 @code{qTfP}/@code{qTsP} remote packet reply payload. A single tracepoint
44943 may take multiple lines of definition, corresponding to the multiple
44946 @item tsv @var{payload}
44947 Trace state variable definition. The @var{payload} has the same format as
44948 @code{qTfV}/@code{qTsV} remote packet reply payload. A single variable
44949 may take multiple lines of definition, corresponding to the multiple
44952 @item tdesc @var{payload}
44953 Target description in XML format. The @var{payload} is a single line of
44954 the XML file. All such lines should be concatenated together to get
44955 the original XML file. This file is in the same format as @code{qXfer}
44956 @code{features} payload, and corresponds to the main @code{target.xml}
44957 file. Includes are not allowed.
44961 The trace frame section consists of a number of consecutive frames.
44962 Each frame begins with a two-byte tracepoint number, followed by a
44963 four-byte size giving the amount of data in the frame. The data in
44964 the frame consists of a number of blocks, each introduced by a
44965 character indicating its type (at least register, memory, and trace
44966 state variable). The data in this section is raw binary, not a
44967 hexadecimal or other encoding; its endianness matches the target's
44970 @c FIXME bi-arch may require endianness/arch info in description section
44973 @item R @var{bytes}
44974 Register block. The number and ordering of bytes matches that of a
44975 @code{g} packet in the remote protocol. Note that these are the
44976 actual bytes, in target order, not a hexadecimal encoding.
44978 @item M @var{address} @var{length} @var{bytes}...
44979 Memory block. This is a contiguous block of memory, at the 8-byte
44980 address @var{address}, with a 2-byte length @var{length}, followed by
44981 @var{length} bytes.
44983 @item V @var{number} @var{value}
44984 Trace state variable block. This records the 8-byte signed value
44985 @var{value} of trace state variable numbered @var{number}.
44989 Future enhancements of the trace file format may include additional types
44992 @node Index Section Format
44993 @appendix @code{.gdb_index} section format
44994 @cindex .gdb_index section format
44995 @cindex index section format
44997 This section documents the index section that is created by @code{save
44998 gdb-index} (@pxref{Index Files}). The index section is
44999 DWARF-specific; some knowledge of DWARF is assumed in this
45002 The mapped index file format is designed to be directly
45003 @code{mmap}able on any architecture. In most cases, a datum is
45004 represented using a little-endian 32-bit integer value, called an
45005 @code{offset_type}. Big endian machines must byte-swap the values
45006 before using them. Exceptions to this rule are noted. The data is
45007 laid out such that alignment is always respected.
45009 A mapped index consists of several areas, laid out in order.
45013 The file header. This is a sequence of values, of @code{offset_type}
45014 unless otherwise noted:
45018 The version number, currently 8. Versions 1, 2 and 3 are obsolete.
45019 Version 4 uses a different hashing function from versions 5 and 6.
45020 Version 6 includes symbols for inlined functions, whereas versions 4
45021 and 5 do not. Version 7 adds attributes to the CU indices in the
45022 symbol table. Version 8 specifies that symbols from DWARF type units
45023 (@samp{DW_TAG_type_unit}) refer to the type unit's symbol table and not the
45024 compilation unit (@samp{DW_TAG_comp_unit}) using the type.
45026 @value{GDBN} will only read version 4, 5, or 6 indices
45027 by specifying @code{set use-deprecated-index-sections on}.
45028 GDB has a workaround for potentially broken version 7 indices so it is
45029 currently not flagged as deprecated.
45032 The offset, from the start of the file, of the CU list.
45035 The offset, from the start of the file, of the types CU list. Note
45036 that this area can be empty, in which case this offset will be equal
45037 to the next offset.
45040 The offset, from the start of the file, of the address area.
45043 The offset, from the start of the file, of the symbol table.
45046 The offset, from the start of the file, of the constant pool.
45050 The CU list. This is a sequence of pairs of 64-bit little-endian
45051 values, sorted by the CU offset. The first element in each pair is
45052 the offset of a CU in the @code{.debug_info} section. The second
45053 element in each pair is the length of that CU. References to a CU
45054 elsewhere in the map are done using a CU index, which is just the
45055 0-based index into this table. Note that if there are type CUs, then
45056 conceptually CUs and type CUs form a single list for the purposes of
45060 The types CU list. This is a sequence of triplets of 64-bit
45061 little-endian values. In a triplet, the first value is the CU offset,
45062 the second value is the type offset in the CU, and the third value is
45063 the type signature. The types CU list is not sorted.
45066 The address area. The address area consists of a sequence of address
45067 entries. Each address entry has three elements:
45071 The low address. This is a 64-bit little-endian value.
45074 The high address. This is a 64-bit little-endian value. Like
45075 @code{DW_AT_high_pc}, the value is one byte beyond the end.
45078 The CU index. This is an @code{offset_type} value.
45082 The symbol table. This is an open-addressed hash table. The size of
45083 the hash table is always a power of 2.
45085 Each slot in the hash table consists of a pair of @code{offset_type}
45086 values. The first value is the offset of the symbol's name in the
45087 constant pool. The second value is the offset of the CU vector in the
45090 If both values are 0, then this slot in the hash table is empty. This
45091 is ok because while 0 is a valid constant pool index, it cannot be a
45092 valid index for both a string and a CU vector.
45094 The hash value for a table entry is computed by applying an
45095 iterative hash function to the symbol's name. Starting with an
45096 initial value of @code{r = 0}, each (unsigned) character @samp{c} in
45097 the string is incorporated into the hash using the formula depending on the
45102 The formula is @code{r = r * 67 + c - 113}.
45104 @item Versions 5 to 7
45105 The formula is @code{r = r * 67 + tolower (c) - 113}.
45108 The terminating @samp{\0} is not incorporated into the hash.
45110 The step size used in the hash table is computed via
45111 @code{((hash * 17) & (size - 1)) | 1}, where @samp{hash} is the hash
45112 value, and @samp{size} is the size of the hash table. The step size
45113 is used to find the next candidate slot when handling a hash
45116 The names of C@t{++} symbols in the hash table are canonicalized. We
45117 don't currently have a simple description of the canonicalization
45118 algorithm; if you intend to create new index sections, you must read
45122 The constant pool. This is simply a bunch of bytes. It is organized
45123 so that alignment is correct: CU vectors are stored first, followed by
45126 A CU vector in the constant pool is a sequence of @code{offset_type}
45127 values. The first value is the number of CU indices in the vector.
45128 Each subsequent value is the index and symbol attributes of a CU in
45129 the CU list. This element in the hash table is used to indicate which
45130 CUs define the symbol and how the symbol is used.
45131 See below for the format of each CU index+attributes entry.
45133 A string in the constant pool is zero-terminated.
45136 Attributes were added to CU index values in @code{.gdb_index} version 7.
45137 If a symbol has multiple uses within a CU then there is one
45138 CU index+attributes value for each use.
45140 The format of each CU index+attributes entry is as follows
45146 This is the index of the CU in the CU list.
45148 These bits are reserved for future purposes and must be zero.
45150 The kind of the symbol in the CU.
45154 This value is reserved and should not be used.
45155 By reserving zero the full @code{offset_type} value is backwards compatible
45156 with previous versions of the index.
45158 The symbol is a type.
45160 The symbol is a variable or an enum value.
45162 The symbol is a function.
45164 Any other kind of symbol.
45166 These values are reserved.
45170 This bit is zero if the value is global and one if it is static.
45172 The determination of whether a symbol is global or static is complicated.
45173 The authorative reference is the file @file{dwarf2read.c} in
45174 @value{GDBN} sources.
45178 This pseudo-code describes the computation of a symbol's kind and
45179 global/static attributes in the index.
45182 is_external = get_attribute (die, DW_AT_external);
45183 language = get_attribute (cu_die, DW_AT_language);
45186 case DW_TAG_typedef:
45187 case DW_TAG_base_type:
45188 case DW_TAG_subrange_type:
45192 case DW_TAG_enumerator:
45194 is_static = language != CPLUS;
45196 case DW_TAG_subprogram:
45198 is_static = ! (is_external || language == ADA);
45200 case DW_TAG_constant:
45202 is_static = ! is_external;
45204 case DW_TAG_variable:
45206 is_static = ! is_external;
45208 case DW_TAG_namespace:
45212 case DW_TAG_class_type:
45213 case DW_TAG_interface_type:
45214 case DW_TAG_structure_type:
45215 case DW_TAG_union_type:
45216 case DW_TAG_enumeration_type:
45218 is_static = language != CPLUS;
45226 @appendix Manual pages
45230 * gdb man:: The GNU Debugger man page
45231 * gdbserver man:: Remote Server for the GNU Debugger man page
45232 * gcore man:: Generate a core file of a running program
45233 * gdbinit man:: gdbinit scripts
45234 * gdb-add-index man:: Add index files to speed up GDB
45240 @c man title gdb The GNU Debugger
45242 @c man begin SYNOPSIS gdb
45243 gdb [@option{-help}] [@option{-nh}] [@option{-nx}] [@option{-q}]
45244 [@option{-batch}] [@option{-cd=}@var{dir}] [@option{-f}]
45245 [@option{-b}@w{ }@var{bps}]
45246 [@option{-tty=}@var{dev}] [@option{-s} @var{symfile}]
45247 [@option{-e}@w{ }@var{prog}] [@option{-se}@w{ }@var{prog}]
45248 [@option{-c}@w{ }@var{core}] [@option{-p}@w{ }@var{procID}]
45249 [@option{-x}@w{ }@var{cmds}] [@option{-d}@w{ }@var{dir}]
45250 [@var{prog}|@var{prog} @var{procID}|@var{prog} @var{core}]
45253 @c man begin DESCRIPTION gdb
45254 The purpose of a debugger such as @value{GDBN} is to allow you to see what is
45255 going on ``inside'' another program while it executes -- or what another
45256 program was doing at the moment it crashed.
45258 @value{GDBN} can do four main kinds of things (plus other things in support of
45259 these) to help you catch bugs in the act:
45263 Start your program, specifying anything that might affect its behavior.
45266 Make your program stop on specified conditions.
45269 Examine what has happened, when your program has stopped.
45272 Change things in your program, so you can experiment with correcting the
45273 effects of one bug and go on to learn about another.
45276 You can use @value{GDBN} to debug programs written in C, C@t{++}, Fortran and
45279 @value{GDBN} is invoked with the shell command @code{gdb}. Once started, it reads
45280 commands from the terminal until you tell it to exit with the @value{GDBN}
45281 command @code{quit}. You can get online help from @value{GDBN} itself
45282 by using the command @code{help}.
45284 You can run @code{gdb} with no arguments or options; but the most
45285 usual way to start @value{GDBN} is with one argument or two, specifying an
45286 executable program as the argument:
45292 You can also start with both an executable program and a core file specified:
45298 You can, instead, specify a process ID as a second argument or use option
45299 @code{-p}, if you want to debug a running process:
45307 would attach @value{GDBN} to process @code{1234}. With option @option{-p} you
45308 can omit the @var{program} filename.
45310 Here are some of the most frequently needed @value{GDBN} commands:
45312 @c pod2man highlights the right hand side of the @item lines.
45314 @item break [@var{file}:]@var{function}
45315 Set a breakpoint at @var{function} (in @var{file}).
45317 @item run [@var{arglist}]
45318 Start your program (with @var{arglist}, if specified).
45321 Backtrace: display the program stack.
45323 @item print @var{expr}
45324 Display the value of an expression.
45327 Continue running your program (after stopping, e.g. at a breakpoint).
45330 Execute next program line (after stopping); step @emph{over} any
45331 function calls in the line.
45333 @item edit [@var{file}:]@var{function}
45334 look at the program line where it is presently stopped.
45336 @item list [@var{file}:]@var{function}
45337 type the text of the program in the vicinity of where it is presently stopped.
45340 Execute next program line (after stopping); step @emph{into} any
45341 function calls in the line.
45343 @item help [@var{name}]
45344 Show information about @value{GDBN} command @var{name}, or general information
45345 about using @value{GDBN}.
45348 Exit from @value{GDBN}.
45352 For full details on @value{GDBN},
45353 see @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
45354 by Richard M. Stallman and Roland H. Pesch. The same text is available online
45355 as the @code{gdb} entry in the @code{info} program.
45359 @c man begin OPTIONS gdb
45360 Any arguments other than options specify an executable
45361 file and core file (or process ID); that is, the first argument
45362 encountered with no
45363 associated option flag is equivalent to a @option{-se} option, and the second,
45364 if any, is equivalent to a @option{-c} option if it's the name of a file.
45366 both long and short forms; both are shown here. The long forms are also
45367 recognized if you truncate them, so long as enough of the option is
45368 present to be unambiguous. (If you prefer, you can flag option
45369 arguments with @option{+} rather than @option{-}, though we illustrate the
45370 more usual convention.)
45372 All the options and command line arguments you give are processed
45373 in sequential order. The order makes a difference when the @option{-x}
45379 List all options, with brief explanations.
45381 @item -symbols=@var{file}
45382 @itemx -s @var{file}
45383 Read symbol table from file @var{file}.
45386 Enable writing into executable and core files.
45388 @item -exec=@var{file}
45389 @itemx -e @var{file}
45390 Use file @var{file} as the executable file to execute when
45391 appropriate, and for examining pure data in conjunction with a core
45394 @item -se=@var{file}
45395 Read symbol table from file @var{file} and use it as the executable
45398 @item -core=@var{file}
45399 @itemx -c @var{file}
45400 Use file @var{file} as a core dump to examine.
45402 @item -command=@var{file}
45403 @itemx -x @var{file}
45404 Execute @value{GDBN} commands from file @var{file}.
45406 @item -ex @var{command}
45407 Execute given @value{GDBN} @var{command}.
45409 @item -directory=@var{directory}
45410 @itemx -d @var{directory}
45411 Add @var{directory} to the path to search for source files.
45414 Do not execute commands from @file{~/.gdbinit}.
45418 Do not execute commands from any @file{.gdbinit} initialization files.
45422 ``Quiet''. Do not print the introductory and copyright messages. These
45423 messages are also suppressed in batch mode.
45426 Run in batch mode. Exit with status @code{0} after processing all the command
45427 files specified with @option{-x} (and @file{.gdbinit}, if not inhibited).
45428 Exit with nonzero status if an error occurs in executing the @value{GDBN}
45429 commands in the command files.
45431 Batch mode may be useful for running @value{GDBN} as a filter, for example to
45432 download and run a program on another computer; in order to make this
45433 more useful, the message
45436 Program exited normally.
45440 (which is ordinarily issued whenever a program running under @value{GDBN} control
45441 terminates) is not issued when running in batch mode.
45443 @item -cd=@var{directory}
45444 Run @value{GDBN} using @var{directory} as its working directory,
45445 instead of the current directory.
45449 Emacs sets this option when it runs @value{GDBN} as a subprocess. It tells
45450 @value{GDBN} to output the full file name and line number in a standard,
45451 recognizable fashion each time a stack frame is displayed (which
45452 includes each time the program stops). This recognizable format looks
45453 like two @samp{\032} characters, followed by the file name, line number
45454 and character position separated by colons, and a newline. The
45455 Emacs-to-@value{GDBN} interface program uses the two @samp{\032}
45456 characters as a signal to display the source code for the frame.
45459 Set the line speed (baud rate or bits per second) of any serial
45460 interface used by @value{GDBN} for remote debugging.
45462 @item -tty=@var{device}
45463 Run using @var{device} for your program's standard input and output.
45467 @c man begin SEEALSO gdb
45469 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
45470 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
45471 documentation are properly installed at your site, the command
45478 should give you access to the complete manual.
45480 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
45481 Richard M. Stallman and Roland H. Pesch, July 1991.
45485 @node gdbserver man
45486 @heading gdbserver man
45488 @c man title gdbserver Remote Server for the GNU Debugger
45490 @c man begin SYNOPSIS gdbserver
45491 gdbserver @var{comm} @var{prog} [@var{args}@dots{}]
45493 gdbserver --attach @var{comm} @var{pid}
45495 gdbserver --multi @var{comm}
45499 @c man begin DESCRIPTION gdbserver
45500 @command{gdbserver} is a program that allows you to run @value{GDBN} on a different machine
45501 than the one which is running the program being debugged.
45504 @subheading Usage (server (target) side)
45507 Usage (server (target) side):
45510 First, you need to have a copy of the program you want to debug put onto
45511 the target system. The program can be stripped to save space if needed, as
45512 @command{gdbserver} doesn't care about symbols. All symbol handling is taken care of by
45513 the @value{GDBN} running on the host system.
45515 To use the server, you log on to the target system, and run the @command{gdbserver}
45516 program. You must tell it (a) how to communicate with @value{GDBN}, (b) the name of
45517 your program, and (c) its arguments. The general syntax is:
45520 target> gdbserver @var{comm} @var{program} [@var{args} ...]
45523 For example, using a serial port, you might say:
45527 @c @file would wrap it as F</dev/com1>.
45528 target> gdbserver /dev/com1 emacs foo.txt
45531 target> gdbserver @file{/dev/com1} emacs foo.txt
45535 This tells @command{gdbserver} to debug emacs with an argument of foo.txt, and
45536 to communicate with @value{GDBN} via @file{/dev/com1}. @command{gdbserver} now
45537 waits patiently for the host @value{GDBN} to communicate with it.
45539 To use a TCP connection, you could say:
45542 target> gdbserver host:2345 emacs foo.txt
45545 This says pretty much the same thing as the last example, except that we are
45546 going to communicate with the @code{host} @value{GDBN} via TCP. The @code{host:2345} argument means
45547 that we are expecting to see a TCP connection from @code{host} to local TCP port
45548 2345. (Currently, the @code{host} part is ignored.) You can choose any number you
45549 want for the port number as long as it does not conflict with any existing TCP
45550 ports on the target system. This same port number must be used in the host
45551 @value{GDBN}s @code{target remote} command, which will be described shortly. Note that if
45552 you chose a port number that conflicts with another service, @command{gdbserver} will
45553 print an error message and exit.
45555 @command{gdbserver} can also attach to running programs.
45556 This is accomplished via the @option{--attach} argument. The syntax is:
45559 target> gdbserver --attach @var{comm} @var{pid}
45562 @var{pid} is the process ID of a currently running process. It isn't
45563 necessary to point @command{gdbserver} at a binary for the running process.
45565 To start @code{gdbserver} without supplying an initial command to run
45566 or process ID to attach, use the @option{--multi} command line option.
45567 In such case you should connect using @kbd{target extended-remote} to start
45568 the program you want to debug.
45571 target> gdbserver --multi @var{comm}
45575 @subheading Usage (host side)
45581 You need an unstripped copy of the target program on your host system, since
45582 @value{GDBN} needs to examine its symbol tables and such. Start up @value{GDBN} as you normally
45583 would, with the target program as the first argument. (You may need to use the
45584 @option{--baud} option if the serial line is running at anything except 9600 baud.)
45585 That is @code{gdb TARGET-PROG}, or @code{gdb --baud BAUD TARGET-PROG}. After that, the only
45586 new command you need to know about is @code{target remote}
45587 (or @code{target extended-remote}). Its argument is either
45588 a device name (usually a serial device, like @file{/dev/ttyb}), or a @code{HOST:PORT}
45589 descriptor. For example:
45593 @c @file would wrap it as F</dev/ttyb>.
45594 (gdb) target remote /dev/ttyb
45597 (gdb) target remote @file{/dev/ttyb}
45602 communicates with the server via serial line @file{/dev/ttyb}, and:
45605 (gdb) target remote the-target:2345
45609 communicates via a TCP connection to port 2345 on host `the-target', where
45610 you previously started up @command{gdbserver} with the same port number. Note that for
45611 TCP connections, you must start up @command{gdbserver} prior to using the `target remote'
45612 command, otherwise you may get an error that looks something like
45613 `Connection refused'.
45615 @command{gdbserver} can also debug multiple inferiors at once,
45618 the @value{GDBN} manual in node @code{Inferiors and Programs}
45619 -- shell command @code{info -f gdb -n 'Inferiors and Programs'}.
45622 @ref{Inferiors and Programs}.
45624 In such case use the @code{extended-remote} @value{GDBN} command variant:
45627 (gdb) target extended-remote the-target:2345
45630 The @command{gdbserver} option @option{--multi} may or may not be used in such
45634 @c man begin OPTIONS gdbserver
45635 There are three different modes for invoking @command{gdbserver}:
45640 Debug a specific program specified by its program name:
45643 gdbserver @var{comm} @var{prog} [@var{args}@dots{}]
45646 The @var{comm} parameter specifies how should the server communicate
45647 with @value{GDBN}; it is either a device name (to use a serial line),
45648 a TCP port number (@code{:1234}), or @code{-} or @code{stdio} to use
45649 stdin/stdout of @code{gdbserver}. Specify the name of the program to
45650 debug in @var{prog}. Any remaining arguments will be passed to the
45651 program verbatim. When the program exits, @value{GDBN} will close the
45652 connection, and @code{gdbserver} will exit.
45655 Debug a specific program by specifying the process ID of a running
45659 gdbserver --attach @var{comm} @var{pid}
45662 The @var{comm} parameter is as described above. Supply the process ID
45663 of a running program in @var{pid}; @value{GDBN} will do everything
45664 else. Like with the previous mode, when the process @var{pid} exits,
45665 @value{GDBN} will close the connection, and @code{gdbserver} will exit.
45668 Multi-process mode -- debug more than one program/process:
45671 gdbserver --multi @var{comm}
45674 In this mode, @value{GDBN} can instruct @command{gdbserver} which
45675 command(s) to run. Unlike the other 2 modes, @value{GDBN} will not
45676 close the connection when a process being debugged exits, so you can
45677 debug several processes in the same session.
45680 In each of the modes you may specify these options:
45685 List all options, with brief explanations.
45688 This option causes @command{gdbserver} to print its version number and exit.
45691 @command{gdbserver} will attach to a running program. The syntax is:
45694 target> gdbserver --attach @var{comm} @var{pid}
45697 @var{pid} is the process ID of a currently running process. It isn't
45698 necessary to point @command{gdbserver} at a binary for the running process.
45701 To start @code{gdbserver} without supplying an initial command to run
45702 or process ID to attach, use this command line option.
45703 Then you can connect using @kbd{target extended-remote} and start
45704 the program you want to debug. The syntax is:
45707 target> gdbserver --multi @var{comm}
45711 Instruct @code{gdbserver} to display extra status information about the debugging
45713 This option is intended for @code{gdbserver} development and for bug reports to
45716 @item --remote-debug
45717 Instruct @code{gdbserver} to display remote protocol debug output.
45718 This option is intended for @code{gdbserver} development and for bug reports to
45721 @item --debug-file=@var{filename}
45722 Instruct @code{gdbserver} to send any debug output to the given @var{filename}.
45723 This option is intended for @code{gdbserver} development and for bug reports to
45726 @item --debug-format=option1@r{[},option2,...@r{]}
45727 Instruct @code{gdbserver} to include extra information in each line
45728 of debugging output.
45729 @xref{Other Command-Line Arguments for gdbserver}.
45732 Specify a wrapper to launch programs
45733 for debugging. The option should be followed by the name of the
45734 wrapper, then any command-line arguments to pass to the wrapper, then
45735 @kbd{--} indicating the end of the wrapper arguments.
45738 By default, @command{gdbserver} keeps the listening TCP port open, so that
45739 additional connections are possible. However, if you start @code{gdbserver}
45740 with the @option{--once} option, it will stop listening for any further
45741 connection attempts after connecting to the first @value{GDBN} session.
45743 @c --disable-packet is not documented for users.
45745 @c --disable-randomization and --no-disable-randomization are superseded by
45746 @c QDisableRandomization.
45751 @c man begin SEEALSO gdbserver
45753 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
45754 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
45755 documentation are properly installed at your site, the command
45761 should give you access to the complete manual.
45763 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
45764 Richard M. Stallman and Roland H. Pesch, July 1991.
45771 @c man title gcore Generate a core file of a running program
45774 @c man begin SYNOPSIS gcore
45775 gcore [-a] [-o @var{prefix}] @var{pid1} [@var{pid2}...@var{pidN}]
45779 @c man begin DESCRIPTION gcore
45780 Generate core dumps of one or more running programs with process IDs
45781 @var{pid1}, @var{pid2}, etc. A core file produced by @command{gcore}
45782 is equivalent to one produced by the kernel when the process crashes
45783 (and when @kbd{ulimit -c} was used to set up an appropriate core dump
45784 limit). However, unlike after a crash, after @command{gcore} finishes
45785 its job the program remains running without any change.
45788 @c man begin OPTIONS gcore
45791 Dump all memory mappings. The actual effect of this option depends on
45792 the Operating System. On @sc{gnu}/Linux, it will disable
45793 @code{use-coredump-filter} (@pxref{set use-coredump-filter}) and
45794 enable @code{dump-excluded-mappings} (@pxref{set
45795 dump-excluded-mappings}).
45797 @item -o @var{prefix}
45798 The optional argument @var{prefix} specifies the prefix to be used
45799 when composing the file names of the core dumps. The file name is
45800 composed as @file{@var{prefix}.@var{pid}}, where @var{pid} is the
45801 process ID of the running program being analyzed by @command{gcore}.
45802 If not specified, @var{prefix} defaults to @var{gcore}.
45806 @c man begin SEEALSO gcore
45808 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
45809 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
45810 documentation are properly installed at your site, the command
45817 should give you access to the complete manual.
45819 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
45820 Richard M. Stallman and Roland H. Pesch, July 1991.
45827 @c man title gdbinit GDB initialization scripts
45830 @c man begin SYNOPSIS gdbinit
45831 @ifset SYSTEM_GDBINIT
45832 @value{SYSTEM_GDBINIT}
45835 @ifset SYSTEM_GDBINIT_DIR
45836 @value{SYSTEM_GDBINIT_DIR}/*
45845 @c man begin DESCRIPTION gdbinit
45846 These files contain @value{GDBN} commands to automatically execute during
45847 @value{GDBN} startup. The lines of contents are canned sequences of commands,
45850 the @value{GDBN} manual in node @code{Sequences}
45851 -- shell command @code{info -f gdb -n Sequences}.
45857 Please read more in
45859 the @value{GDBN} manual in node @code{Startup}
45860 -- shell command @code{info -f gdb -n Startup}.
45867 @ifset SYSTEM_GDBINIT
45868 @item @value{SYSTEM_GDBINIT}
45870 @ifclear SYSTEM_GDBINIT
45871 @item (not enabled with @code{--with-system-gdbinit} during compilation)
45873 System-wide initialization file. It is executed unless user specified
45874 @value{GDBN} option @code{-nx} or @code{-n}.
45877 the @value{GDBN} manual in node @code{System-wide configuration}
45878 -- shell command @code{info -f gdb -n 'System-wide configuration'}.
45880 @ifset SYSTEM_GDBINIT_DIR
45881 @item @value{SYSTEM_GDBINIT_DIR}
45883 @ifclear SYSTEM_GDBINIT_DIR
45884 @item (not enabled with @code{--with-system-gdbinit-dir} during compilation)
45886 System-wide initialization directory. All files in this directory are
45887 executed on startup unless user specified @value{GDBN} option @code{-nx} or
45888 @code{-n}, as long as they have a recognized file extension.
45891 the @value{GDBN} manual in node @code{System-wide configuration}
45892 -- shell command @code{info -f gdb -n 'System-wide configuration'}.
45895 @ref{System-wide configuration}.
45899 User initialization file. It is executed unless user specified
45900 @value{GDBN} options @code{-nx}, @code{-n} or @code{-nh}.
45903 Initialization file for current directory. It may need to be enabled with
45904 @value{GDBN} security command @code{set auto-load local-gdbinit}.
45907 the @value{GDBN} manual in node @code{Init File in the Current Directory}
45908 -- shell command @code{info -f gdb -n 'Init File in the Current Directory'}.
45911 @ref{Init File in the Current Directory}.
45916 @c man begin SEEALSO gdbinit
45918 gdb(1), @code{info -f gdb -n Startup}
45920 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
45921 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
45922 documentation are properly installed at your site, the command
45928 should give you access to the complete manual.
45930 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
45931 Richard M. Stallman and Roland H. Pesch, July 1991.
45935 @node gdb-add-index man
45936 @heading gdb-add-index
45937 @pindex gdb-add-index
45938 @anchor{gdb-add-index}
45940 @c man title gdb-add-index Add index files to speed up GDB
45942 @c man begin SYNOPSIS gdb-add-index
45943 gdb-add-index @var{filename}
45946 @c man begin DESCRIPTION gdb-add-index
45947 When @value{GDBN} finds a symbol file, it scans the symbols in the
45948 file in order to construct an internal symbol table. This lets most
45949 @value{GDBN} operations work quickly--at the cost of a delay early on.
45950 For large programs, this delay can be quite lengthy, so @value{GDBN}
45951 provides a way to build an index, which speeds up startup.
45953 To determine whether a file contains such an index, use the command
45954 @kbd{readelf -S filename}: the index is stored in a section named
45955 @code{.gdb_index}. The index file can only be produced on systems
45956 which use ELF binaries and DWARF debug information (i.e., sections
45957 named @code{.debug_*}).
45959 @command{gdb-add-index} uses @value{GDBN} and @command{objdump} found
45960 in the @env{PATH} environment variable. If you want to use different
45961 versions of these programs, you can specify them through the
45962 @env{GDB} and @env{OBJDUMP} environment variables.
45966 the @value{GDBN} manual in node @code{Index Files}
45967 -- shell command @kbd{info -f gdb -n "Index Files"}.
45974 @c man begin SEEALSO gdb-add-index
45976 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
45977 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
45978 documentation are properly installed at your site, the command
45984 should give you access to the complete manual.
45986 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
45987 Richard M. Stallman and Roland H. Pesch, July 1991.
45993 @node GNU Free Documentation License
45994 @appendix GNU Free Documentation License
45997 @node Concept Index
45998 @unnumbered Concept Index
46002 @node Command and Variable Index
46003 @unnumbered Command, Variable, and Function Index
46008 % I think something like @@colophon should be in texinfo. In the
46010 \long\def\colophon{\hbox to0pt{}\vfill
46011 \centerline{The body of this manual is set in}
46012 \centerline{\fontname\tenrm,}
46013 \centerline{with headings in {\bf\fontname\tenbf}}
46014 \centerline{and examples in {\tt\fontname\tentt}.}
46015 \centerline{{\it\fontname\tenit\/},}
46016 \centerline{{\bf\fontname\tenbf}, and}
46017 \centerline{{\sl\fontname\tensl\/}}
46018 \centerline{are used for emphasis.}\vfill}
46020 % Blame: doc@@cygnus.com, 1991.