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, if you want
878 to debug a running process:
881 @value{GDBP} @var{program} 1234
885 would attach @value{GDBN} to process @code{1234} (unless you also have a file
886 named @file{1234}; @value{GDBN} does check for a core file first).
888 Taking advantage of the second command-line argument requires a fairly
889 complete operating system; when you use @value{GDBN} as a remote
890 debugger attached to a bare board, there may not be any notion of
891 ``process'', and there is often no way to get a core dump. @value{GDBN}
892 will warn you if it is unable to attach or to read core dumps.
894 You can optionally have @code{@value{GDBP}} pass any arguments after the
895 executable file to the inferior using @code{--args}. This option stops
898 @value{GDBP} --args gcc -O2 -c foo.c
900 This will cause @code{@value{GDBP}} to debug @code{gcc}, and to set
901 @code{gcc}'s command-line arguments (@pxref{Arguments}) to @samp{-O2 -c foo.c}.
903 You can run @code{@value{GDBP}} without printing the front material, which describes
904 @value{GDBN}'s non-warranty, by specifying @code{--silent}
905 (or @code{-q}/@code{--quiet}):
908 @value{GDBP} --silent
912 You can further control how @value{GDBN} starts up by using command-line
913 options. @value{GDBN} itself can remind you of the options available.
923 to display all available options and briefly describe their use
924 (@samp{@value{GDBP} -h} is a shorter equivalent).
926 All options and command line arguments you give are processed
927 in sequential order. The order makes a difference when the
928 @samp{-x} option is used.
932 * File Options:: Choosing files
933 * Mode Options:: Choosing modes
934 * Startup:: What @value{GDBN} does during startup
938 @subsection Choosing Files
940 When @value{GDBN} starts, it reads any arguments other than options as
941 specifying an executable file and core file (or process ID). This is
942 the same as if the arguments were specified by the @samp{-se} and
943 @samp{-c} (or @samp{-p}) options respectively. (@value{GDBN} reads the
944 first argument that does not have an associated option flag as
945 equivalent to the @samp{-se} option followed by that argument; and the
946 second argument that does not have an associated option flag, if any, as
947 equivalent to the @samp{-c}/@samp{-p} option followed by that argument.)
948 If the second argument begins with a decimal digit, @value{GDBN} will
949 first attempt to attach to it as a process, and if that fails, attempt
950 to open it as a corefile. If you have a corefile whose name begins with
951 a digit, you can prevent @value{GDBN} from treating it as a pid by
952 prefixing it with @file{./}, e.g.@: @file{./12345}.
954 If @value{GDBN} has not been configured to included core file support,
955 such as for most embedded targets, then it will complain about a second
956 argument and ignore it.
958 Many options have both long and short forms; both are shown in the
959 following list. @value{GDBN} also recognizes the long forms if you truncate
960 them, so long as enough of the option is present to be unambiguous.
961 (If you prefer, you can flag option arguments with @samp{--} rather
962 than @samp{-}, though we illustrate the more usual convention.)
964 @c NOTE: the @cindex entries here use double dashes ON PURPOSE. This
965 @c way, both those who look for -foo and --foo in the index, will find
969 @item -symbols @var{file}
971 @cindex @code{--symbols}
973 Read symbol table from file @var{file}.
975 @item -exec @var{file}
977 @cindex @code{--exec}
979 Use file @var{file} as the executable file to execute when appropriate,
980 and for examining pure data in conjunction with a core dump.
984 Read symbol table from file @var{file} and use it as the executable
987 @item -core @var{file}
989 @cindex @code{--core}
991 Use file @var{file} as a core dump to examine.
993 @item -pid @var{number}
994 @itemx -p @var{number}
997 Connect to process ID @var{number}, as with the @code{attach} command.
999 @item -command @var{file}
1000 @itemx -x @var{file}
1001 @cindex @code{--command}
1003 Execute commands from file @var{file}. The contents of this file is
1004 evaluated exactly as the @code{source} command would.
1005 @xref{Command Files,, Command files}.
1007 @item -eval-command @var{command}
1008 @itemx -ex @var{command}
1009 @cindex @code{--eval-command}
1011 Execute a single @value{GDBN} command.
1013 This option may be used multiple times to call multiple commands. It may
1014 also be interleaved with @samp{-command} as required.
1017 @value{GDBP} -ex 'target sim' -ex 'load' \
1018 -x setbreakpoints -ex 'run' a.out
1021 @item -init-command @var{file}
1022 @itemx -ix @var{file}
1023 @cindex @code{--init-command}
1025 Execute commands from file @var{file} before loading the inferior (but
1026 after loading gdbinit files).
1029 @item -init-eval-command @var{command}
1030 @itemx -iex @var{command}
1031 @cindex @code{--init-eval-command}
1033 Execute a single @value{GDBN} command before loading the inferior (but
1034 after loading gdbinit files).
1037 @item -directory @var{directory}
1038 @itemx -d @var{directory}
1039 @cindex @code{--directory}
1041 Add @var{directory} to the path to search for source and script files.
1045 @cindex @code{--readnow}
1047 Read each symbol file's entire symbol table immediately, rather than
1048 the default, which is to read it incrementally as it is needed.
1049 This makes startup slower, but makes future operations faster.
1052 @anchor{--readnever}
1053 @cindex @code{--readnever}, command-line option
1054 Do not read each symbol file's symbolic debug information. This makes
1055 startup faster but at the expense of not being able to perform
1056 symbolic debugging. DWARF unwind information is also not read,
1057 meaning backtraces may become incomplete or inaccurate. One use of
1058 this is when a user simply wants to do the following sequence: attach,
1059 dump core, detach. Loading the debugging information in this case is
1060 an unnecessary cause of delay.
1064 @subsection Choosing Modes
1066 You can run @value{GDBN} in various alternative modes---for example, in
1067 batch mode or quiet mode.
1075 Do not execute commands found in any initialization file.
1076 There are three init files, loaded in the following order:
1079 @item @file{system.gdbinit}
1080 This is the system-wide init file.
1081 Its location is specified with the @code{--with-system-gdbinit}
1082 configure option (@pxref{System-wide configuration}).
1083 It is loaded first when @value{GDBN} starts, before command line options
1084 have been processed.
1085 @item @file{~/.gdbinit}
1086 This is the init file in your home directory.
1087 It is loaded next, after @file{system.gdbinit}, and before
1088 command options have been processed.
1089 @item @file{./.gdbinit}
1090 This is the init file in the current directory.
1091 It is loaded last, after command line options other than @code{-x} and
1092 @code{-ex} have been processed. Command line options @code{-x} and
1093 @code{-ex} are processed last, after @file{./.gdbinit} has been loaded.
1096 For further documentation on startup processing, @xref{Startup}.
1097 For documentation on how to write command files,
1098 @xref{Command Files,,Command Files}.
1103 Do not execute commands found in @file{~/.gdbinit}, the init file
1104 in your home directory.
1110 @cindex @code{--quiet}
1111 @cindex @code{--silent}
1113 ``Quiet''. Do not print the introductory and copyright messages. These
1114 messages are also suppressed in batch mode.
1117 @cindex @code{--batch}
1118 Run in batch mode. Exit with status @code{0} after processing all the
1119 command files specified with @samp{-x} (and all commands from
1120 initialization files, if not inhibited with @samp{-n}). Exit with
1121 nonzero status if an error occurs in executing the @value{GDBN} commands
1122 in the command files. Batch mode also disables pagination, sets unlimited
1123 terminal width and height @pxref{Screen Size}, and acts as if @kbd{set confirm
1124 off} were in effect (@pxref{Messages/Warnings}).
1126 Batch mode may be useful for running @value{GDBN} as a filter, for
1127 example to download and run a program on another computer; in order to
1128 make this more useful, the message
1131 Program exited normally.
1135 (which is ordinarily issued whenever a program running under
1136 @value{GDBN} control terminates) is not issued when running in batch
1140 @cindex @code{--batch-silent}
1141 Run in batch mode exactly like @samp{-batch}, but totally silently. All
1142 @value{GDBN} output to @code{stdout} is prevented (@code{stderr} is
1143 unaffected). This is much quieter than @samp{-silent} and would be useless
1144 for an interactive session.
1146 This is particularly useful when using targets that give @samp{Loading section}
1147 messages, for example.
1149 Note that targets that give their output via @value{GDBN}, as opposed to
1150 writing directly to @code{stdout}, will also be made silent.
1152 @item -return-child-result
1153 @cindex @code{--return-child-result}
1154 The return code from @value{GDBN} will be the return code from the child
1155 process (the process being debugged), with the following exceptions:
1159 @value{GDBN} exits abnormally. E.g., due to an incorrect argument or an
1160 internal error. In this case the exit code is the same as it would have been
1161 without @samp{-return-child-result}.
1163 The user quits with an explicit value. E.g., @samp{quit 1}.
1165 The child process never runs, or is not allowed to terminate, in which case
1166 the exit code will be -1.
1169 This option is useful in conjunction with @samp{-batch} or @samp{-batch-silent},
1170 when @value{GDBN} is being used as a remote program loader or simulator
1175 @cindex @code{--nowindows}
1177 ``No windows''. If @value{GDBN} comes with a graphical user interface
1178 (GUI) built in, then this option tells @value{GDBN} to only use the command-line
1179 interface. If no GUI is available, this option has no effect.
1183 @cindex @code{--windows}
1185 If @value{GDBN} includes a GUI, then this option requires it to be
1188 @item -cd @var{directory}
1190 Run @value{GDBN} using @var{directory} as its working directory,
1191 instead of the current directory.
1193 @item -data-directory @var{directory}
1194 @itemx -D @var{directory}
1195 @cindex @code{--data-directory}
1197 Run @value{GDBN} using @var{directory} as its data directory.
1198 The data directory is where @value{GDBN} searches for its
1199 auxiliary files. @xref{Data Files}.
1203 @cindex @code{--fullname}
1205 @sc{gnu} Emacs sets this option when it runs @value{GDBN} as a
1206 subprocess. It tells @value{GDBN} to output the full file name and line
1207 number in a standard, recognizable fashion each time a stack frame is
1208 displayed (which includes each time your program stops). This
1209 recognizable format looks like two @samp{\032} characters, followed by
1210 the file name, line number and character position separated by colons,
1211 and a newline. The Emacs-to-@value{GDBN} interface program uses the two
1212 @samp{\032} characters as a signal to display the source code for the
1215 @item -annotate @var{level}
1216 @cindex @code{--annotate}
1217 This option sets the @dfn{annotation level} inside @value{GDBN}. Its
1218 effect is identical to using @samp{set annotate @var{level}}
1219 (@pxref{Annotations}). The annotation @var{level} controls how much
1220 information @value{GDBN} prints together with its prompt, values of
1221 expressions, source lines, and other types of output. Level 0 is the
1222 normal, level 1 is for use when @value{GDBN} is run as a subprocess of
1223 @sc{gnu} Emacs, level 3 is the maximum annotation suitable for programs
1224 that control @value{GDBN}, and level 2 has been deprecated.
1226 The annotation mechanism has largely been superseded by @sc{gdb/mi}
1230 @cindex @code{--args}
1231 Change interpretation of command line so that arguments following the
1232 executable file are passed as command line arguments to the inferior.
1233 This option stops option processing.
1235 @item -baud @var{bps}
1237 @cindex @code{--baud}
1239 Set the line speed (baud rate or bits per second) of any serial
1240 interface used by @value{GDBN} for remote debugging.
1242 @item -l @var{timeout}
1244 Set the timeout (in seconds) of any communication used by @value{GDBN}
1245 for remote debugging.
1247 @item -tty @var{device}
1248 @itemx -t @var{device}
1249 @cindex @code{--tty}
1251 Run using @var{device} for your program's standard input and output.
1252 @c FIXME: kingdon thinks there is more to -tty. Investigate.
1254 @c resolve the situation of these eventually
1256 @cindex @code{--tui}
1257 Activate the @dfn{Text User Interface} when starting. The Text User
1258 Interface manages several text windows on the terminal, showing
1259 source, assembly, registers and @value{GDBN} command outputs
1260 (@pxref{TUI, ,@value{GDBN} Text User Interface}). Do not use this
1261 option if you run @value{GDBN} from Emacs (@pxref{Emacs, ,
1262 Using @value{GDBN} under @sc{gnu} Emacs}).
1264 @item -interpreter @var{interp}
1265 @cindex @code{--interpreter}
1266 Use the interpreter @var{interp} for interface with the controlling
1267 program or device. This option is meant to be set by programs which
1268 communicate with @value{GDBN} using it as a back end.
1269 @xref{Interpreters, , Command Interpreters}.
1271 @samp{--interpreter=mi} (or @samp{--interpreter=mi3}) causes
1272 @value{GDBN} to use the @dfn{@sc{gdb/mi} interface} version 3 (@pxref{GDB/MI, ,
1273 The @sc{gdb/mi} Interface}) included since @value{GDBN} version 9.1. @sc{gdb/mi}
1274 version 2 (@code{mi2}), included in @value{GDBN} 6.0 and version 1 (@code{mi1}),
1275 included in @value{GDBN} 5.3, are also available. Earlier @sc{gdb/mi}
1276 interfaces are no longer supported.
1279 @cindex @code{--write}
1280 Open the executable and core files for both reading and writing. This
1281 is equivalent to the @samp{set write on} command inside @value{GDBN}
1285 @cindex @code{--statistics}
1286 This option causes @value{GDBN} to print statistics about time and
1287 memory usage after it completes each command and returns to the prompt.
1290 @cindex @code{--version}
1291 This option causes @value{GDBN} to print its version number and
1292 no-warranty blurb, and exit.
1294 @item -configuration
1295 @cindex @code{--configuration}
1296 This option causes @value{GDBN} to print details about its build-time
1297 configuration parameters, and then exit. These details can be
1298 important when reporting @value{GDBN} bugs (@pxref{GDB Bugs}).
1303 @subsection What @value{GDBN} Does During Startup
1304 @cindex @value{GDBN} startup
1306 Here's the description of what @value{GDBN} does during session startup:
1310 Sets up the command interpreter as specified by the command line
1311 (@pxref{Mode Options, interpreter}).
1315 Reads the system-wide @dfn{init file} (if @option{--with-system-gdbinit} was
1316 used when building @value{GDBN}; @pxref{System-wide configuration,
1317 ,System-wide configuration and settings}) and executes all the commands in
1320 @anchor{Home Directory Init File}
1322 Reads the init file (if any) in your home directory@footnote{On
1323 DOS/Windows systems, the home directory is the one pointed to by the
1324 @code{HOME} environment variable.} and executes all the commands in
1327 @anchor{Option -init-eval-command}
1329 Executes commands and command files specified by the @samp{-iex} and
1330 @samp{-ix} options in their specified order. Usually you should use the
1331 @samp{-ex} and @samp{-x} options instead, but this way you can apply
1332 settings before @value{GDBN} init files get executed and before inferior
1336 Processes command line options and operands.
1338 @anchor{Init File in the Current Directory during Startup}
1340 Reads and executes the commands from init file (if any) in the current
1341 working directory as long as @samp{set auto-load local-gdbinit} is set to
1342 @samp{on} (@pxref{Init File in the Current Directory}).
1343 This is only done if the current directory is
1344 different from your home directory. Thus, you can have more than one
1345 init file, one generic in your home directory, and another, specific
1346 to the program you are debugging, in the directory where you invoke
1350 If the command line specified a program to debug, or a process to
1351 attach to, or a core file, @value{GDBN} loads any auto-loaded
1352 scripts provided for the program or for its loaded shared libraries.
1353 @xref{Auto-loading}.
1355 If you wish to disable the auto-loading during startup,
1356 you must do something like the following:
1359 $ gdb -iex "set auto-load python-scripts off" myprogram
1362 Option @samp{-ex} does not work because the auto-loading is then turned
1366 Executes commands and command files specified by the @samp{-ex} and
1367 @samp{-x} options in their specified order. @xref{Command Files}, for
1368 more details about @value{GDBN} command files.
1371 Reads the command history recorded in the @dfn{history file}.
1372 @xref{Command History}, for more details about the command history and the
1373 files where @value{GDBN} records it.
1376 Init files use the same syntax as @dfn{command files} (@pxref{Command
1377 Files}) and are processed by @value{GDBN} in the same way. The init
1378 file in your home directory can set options (such as @samp{set
1379 complaints}) that affect subsequent processing of command line options
1380 and operands. Init files are not executed if you use the @samp{-nx}
1381 option (@pxref{Mode Options, ,Choosing Modes}).
1383 To display the list of init files loaded by gdb at startup, you
1384 can use @kbd{gdb --help}.
1386 @cindex init file name
1387 @cindex @file{.gdbinit}
1388 @cindex @file{gdb.ini}
1389 The @value{GDBN} init files are normally called @file{.gdbinit}.
1390 The DJGPP port of @value{GDBN} uses the name @file{gdb.ini}, due to
1391 the limitations of file names imposed by DOS filesystems. The Windows
1392 port of @value{GDBN} uses the standard name, but if it finds a
1393 @file{gdb.ini} file in your home directory, it warns you about that
1394 and suggests to rename the file to the standard name.
1398 @section Quitting @value{GDBN}
1399 @cindex exiting @value{GDBN}
1400 @cindex leaving @value{GDBN}
1403 @kindex quit @r{[}@var{expression}@r{]}
1404 @kindex q @r{(@code{quit})}
1405 @item quit @r{[}@var{expression}@r{]}
1407 To exit @value{GDBN}, use the @code{quit} command (abbreviated
1408 @code{q}), or type an end-of-file character (usually @kbd{Ctrl-d}). If you
1409 do not supply @var{expression}, @value{GDBN} will terminate normally;
1410 otherwise it will terminate using the result of @var{expression} as the
1415 An interrupt (often @kbd{Ctrl-c}) does not exit from @value{GDBN}, but rather
1416 terminates the action of any @value{GDBN} command that is in progress and
1417 returns to @value{GDBN} command level. It is safe to type the interrupt
1418 character at any time because @value{GDBN} does not allow it to take effect
1419 until a time when it is safe.
1421 If you have been using @value{GDBN} to control an attached process or
1422 device, you can release it with the @code{detach} command
1423 (@pxref{Attach, ,Debugging an Already-running Process}).
1425 @node Shell Commands
1426 @section Shell Commands
1428 If you need to execute occasional shell commands during your
1429 debugging session, there is no need to leave or suspend @value{GDBN}; you can
1430 just use the @code{shell} command.
1435 @cindex shell escape
1436 @item shell @var{command-string}
1437 @itemx !@var{command-string}
1438 Invoke a standard shell to execute @var{command-string}.
1439 Note that no space is needed between @code{!} and @var{command-string}.
1440 If it exists, the environment variable @code{SHELL} determines which
1441 shell to run. Otherwise @value{GDBN} uses the default shell
1442 (@file{/bin/sh} on Unix systems, @file{COMMAND.COM} on MS-DOS, etc.).
1445 The utility @code{make} is often needed in development environments.
1446 You do not have to use the @code{shell} command for this purpose in
1451 @cindex calling make
1452 @item make @var{make-args}
1453 Execute the @code{make} program with the specified
1454 arguments. This is equivalent to @samp{shell make @var{make-args}}.
1460 @cindex send the output of a gdb command to a shell command
1462 @item pipe [@var{command}] | @var{shell_command}
1463 @itemx | [@var{command}] | @var{shell_command}
1464 @itemx pipe -d @var{delim} @var{command} @var{delim} @var{shell_command}
1465 @itemx | -d @var{delim} @var{command} @var{delim} @var{shell_command}
1466 Executes @var{command} and sends its output to @var{shell_command}.
1467 Note that no space is needed around @code{|}.
1468 If no @var{command} is provided, the last command executed is repeated.
1470 In case the @var{command} contains a @code{|}, the option @code{-d @var{delim}}
1471 can be used to specify an alternate delimiter string @var{delim} that separates
1472 the @var{command} from the @var{shell_command}.
1505 (gdb) | -d ! echo this contains a | char\n ! sed -e 's/|/PIPE/'
1506 this contains a PIPE char
1507 (gdb) | -d xxx echo this contains a | char!\n xxx sed -e 's/|/PIPE/'
1508 this contains a PIPE char!
1514 The convenience variables @code{$_shell_exitcode} and @code{$_shell_exitsignal}
1515 can be used to examine the exit status of the last shell command launched
1516 by @code{shell}, @code{make}, @code{pipe} and @code{|}.
1517 @xref{Convenience Vars,, Convenience Variables}.
1519 @node Logging Output
1520 @section Logging Output
1521 @cindex logging @value{GDBN} output
1522 @cindex save @value{GDBN} output to a file
1524 You may want to save the output of @value{GDBN} commands to a file.
1525 There are several commands to control @value{GDBN}'s logging.
1529 @item set logging on
1531 @item set logging off
1533 @cindex logging file name
1534 @item set logging file @var{file}
1535 Change the name of the current logfile. The default logfile is @file{gdb.txt}.
1536 @item set logging overwrite [on|off]
1537 By default, @value{GDBN} will append to the logfile. Set @code{overwrite} if
1538 you want @code{set logging on} to overwrite the logfile instead.
1539 @item set logging redirect [on|off]
1540 By default, @value{GDBN} output will go to both the terminal and the logfile.
1541 Set @code{redirect} if you want output to go only to the log file.
1542 @item set logging debugredirect [on|off]
1543 By default, @value{GDBN} debug output will go to both the terminal and the logfile.
1544 Set @code{debugredirect} if you want debug output to go only to the log file.
1545 @kindex show logging
1547 Show the current values of the logging settings.
1550 You can also redirect the output of a @value{GDBN} command to a
1551 shell command. @xref{pipe}.
1553 @chapter @value{GDBN} Commands
1555 You can abbreviate a @value{GDBN} command to the first few letters of the command
1556 name, if that abbreviation is unambiguous; and you can repeat certain
1557 @value{GDBN} commands by typing just @key{RET}. You can also use the @key{TAB}
1558 key to get @value{GDBN} to fill out the rest of a word in a command (or to
1559 show you the alternatives available, if there is more than one possibility).
1562 * Command Syntax:: How to give commands to @value{GDBN}
1563 * Completion:: Command completion
1564 * Help:: How to ask @value{GDBN} for help
1567 @node Command Syntax
1568 @section Command Syntax
1570 A @value{GDBN} command is a single line of input. There is no limit on
1571 how long it can be. It starts with a command name, which is followed by
1572 arguments whose meaning depends on the command name. For example, the
1573 command @code{step} accepts an argument which is the number of times to
1574 step, as in @samp{step 5}. You can also use the @code{step} command
1575 with no arguments. Some commands do not allow any arguments.
1577 @cindex abbreviation
1578 @value{GDBN} command names may always be truncated if that abbreviation is
1579 unambiguous. Other possible command abbreviations are listed in the
1580 documentation for individual commands. In some cases, even ambiguous
1581 abbreviations are allowed; for example, @code{s} is specially defined as
1582 equivalent to @code{step} even though there are other commands whose
1583 names start with @code{s}. You can test abbreviations by using them as
1584 arguments to the @code{help} command.
1586 @cindex repeating commands
1587 @kindex RET @r{(repeat last command)}
1588 A blank line as input to @value{GDBN} (typing just @key{RET}) means to
1589 repeat the previous command. Certain commands (for example, @code{run})
1590 will not repeat this way; these are commands whose unintentional
1591 repetition might cause trouble and which you are unlikely to want to
1592 repeat. User-defined commands can disable this feature; see
1593 @ref{Define, dont-repeat}.
1595 The @code{list} and @code{x} commands, when you repeat them with
1596 @key{RET}, construct new arguments rather than repeating
1597 exactly as typed. This permits easy scanning of source or memory.
1599 @value{GDBN} can also use @key{RET} in another way: to partition lengthy
1600 output, in a way similar to the common utility @code{more}
1601 (@pxref{Screen Size,,Screen Size}). Since it is easy to press one
1602 @key{RET} too many in this situation, @value{GDBN} disables command
1603 repetition after any command that generates this sort of display.
1605 @kindex # @r{(a comment)}
1607 Any text from a @kbd{#} to the end of the line is a comment; it does
1608 nothing. This is useful mainly in command files (@pxref{Command
1609 Files,,Command Files}).
1611 @cindex repeating command sequences
1612 @kindex Ctrl-o @r{(operate-and-get-next)}
1613 The @kbd{Ctrl-o} binding is useful for repeating a complex sequence of
1614 commands. This command accepts the current line, like @key{RET}, and
1615 then fetches the next line relative to the current line from the history
1619 @section Command Completion
1622 @cindex word completion
1623 @value{GDBN} can fill in the rest of a word in a command for you, if there is
1624 only one possibility; it can also show you what the valid possibilities
1625 are for the next word in a command, at any time. This works for @value{GDBN}
1626 commands, @value{GDBN} subcommands, and the names of symbols in your program.
1628 Press the @key{TAB} key whenever you want @value{GDBN} to fill out the rest
1629 of a word. If there is only one possibility, @value{GDBN} fills in the
1630 word, and waits for you to finish the command (or press @key{RET} to
1631 enter it). For example, if you type
1633 @c FIXME "@key" does not distinguish its argument sufficiently to permit
1634 @c complete accuracy in these examples; space introduced for clarity.
1635 @c If texinfo enhancements make it unnecessary, it would be nice to
1636 @c replace " @key" by "@key" in the following...
1638 (@value{GDBP}) info bre @key{TAB}
1642 @value{GDBN} fills in the rest of the word @samp{breakpoints}, since that is
1643 the only @code{info} subcommand beginning with @samp{bre}:
1646 (@value{GDBP}) info breakpoints
1650 You can either press @key{RET} at this point, to run the @code{info
1651 breakpoints} command, or backspace and enter something else, if
1652 @samp{breakpoints} does not look like the command you expected. (If you
1653 were sure you wanted @code{info breakpoints} in the first place, you
1654 might as well just type @key{RET} immediately after @samp{info bre},
1655 to exploit command abbreviations rather than command completion).
1657 If there is more than one possibility for the next word when you press
1658 @key{TAB}, @value{GDBN} sounds a bell. You can either supply more
1659 characters and try again, or just press @key{TAB} a second time;
1660 @value{GDBN} displays all the possible completions for that word. For
1661 example, you might want to set a breakpoint on a subroutine whose name
1662 begins with @samp{make_}, but when you type @kbd{b make_@key{TAB}} @value{GDBN}
1663 just sounds the bell. Typing @key{TAB} again displays all the
1664 function names in your program that begin with those characters, for
1668 (@value{GDBP}) b make_ @key{TAB}
1669 @exdent @value{GDBN} sounds bell; press @key{TAB} again, to see:
1670 make_a_section_from_file make_environ
1671 make_abs_section make_function_type
1672 make_blockvector make_pointer_type
1673 make_cleanup make_reference_type
1674 make_command make_symbol_completion_list
1675 (@value{GDBP}) b make_
1679 After displaying the available possibilities, @value{GDBN} copies your
1680 partial input (@samp{b make_} in the example) so you can finish the
1683 If you just want to see the list of alternatives in the first place, you
1684 can press @kbd{M-?} rather than pressing @key{TAB} twice. @kbd{M-?}
1685 means @kbd{@key{META} ?}. You can type this either by holding down a
1686 key designated as the @key{META} shift on your keyboard (if there is
1687 one) while typing @kbd{?}, or as @key{ESC} followed by @kbd{?}.
1689 If the number of possible completions is large, @value{GDBN} will
1690 print as much of the list as it has collected, as well as a message
1691 indicating that the list may be truncated.
1694 (@value{GDBP}) b m@key{TAB}@key{TAB}
1696 <... the rest of the possible completions ...>
1697 *** List may be truncated, max-completions reached. ***
1702 This behavior can be controlled with the following commands:
1705 @kindex set max-completions
1706 @item set max-completions @var{limit}
1707 @itemx set max-completions unlimited
1708 Set the maximum number of completion candidates. @value{GDBN} will
1709 stop looking for more completions once it collects this many candidates.
1710 This is useful when completing on things like function names as collecting
1711 all the possible candidates can be time consuming.
1712 The default value is 200. A value of zero disables tab-completion.
1713 Note that setting either no limit or a very large limit can make
1715 @kindex show max-completions
1716 @item show max-completions
1717 Show the maximum number of candidates that @value{GDBN} will collect and show
1721 @cindex quotes in commands
1722 @cindex completion of quoted strings
1723 Sometimes the string you need, while logically a ``word'', may contain
1724 parentheses or other characters that @value{GDBN} normally excludes from
1725 its notion of a word. To permit word completion to work in this
1726 situation, you may enclose words in @code{'} (single quote marks) in
1727 @value{GDBN} commands.
1729 A likely situation where you might need this is in typing an
1730 expression that involves a C@t{++} symbol name with template
1731 parameters. This is because when completing expressions, GDB treats
1732 the @samp{<} character as word delimiter, assuming that it's the
1733 less-than comparison operator (@pxref{C Operators, , C and C@t{++}
1736 For example, when you want to call a C@t{++} template function
1737 interactively using the @code{print} or @code{call} commands, you may
1738 need to distinguish whether you mean the version of @code{name} that
1739 was specialized for @code{int}, @code{name<int>()}, or the version
1740 that was specialized for @code{float}, @code{name<float>()}. To use
1741 the word-completion facilities in this situation, type a single quote
1742 @code{'} at the beginning of the function name. This alerts
1743 @value{GDBN} that it may need to consider more information than usual
1744 when you press @key{TAB} or @kbd{M-?} to request word completion:
1747 (@value{GDBP}) p 'func< @kbd{M-?}
1748 func<int>() func<float>()
1749 (@value{GDBP}) p 'func<
1752 When setting breakpoints however (@pxref{Specify Location}), you don't
1753 usually need to type a quote before the function name, because
1754 @value{GDBN} understands that you want to set a breakpoint on a
1758 (@value{GDBP}) b func< @kbd{M-?}
1759 func<int>() func<float>()
1760 (@value{GDBP}) b func<
1763 This is true even in the case of typing the name of C@t{++} overloaded
1764 functions (multiple definitions of the same function, distinguished by
1765 argument type). For example, when you want to set a breakpoint you
1766 don't need to distinguish whether you mean the version of @code{name}
1767 that takes an @code{int} parameter, @code{name(int)}, or the version
1768 that takes a @code{float} parameter, @code{name(float)}.
1771 (@value{GDBP}) b bubble( @kbd{M-?}
1772 bubble(int) bubble(double)
1773 (@value{GDBP}) b bubble(dou @kbd{M-?}
1777 See @ref{quoting names} for a description of other scenarios that
1780 For more information about overloaded functions, see @ref{C Plus Plus
1781 Expressions, ,C@t{++} Expressions}. You can use the command @code{set
1782 overload-resolution off} to disable overload resolution;
1783 see @ref{Debugging C Plus Plus, ,@value{GDBN} Features for C@t{++}}.
1785 @cindex completion of structure field names
1786 @cindex structure field name completion
1787 @cindex completion of union field names
1788 @cindex union field name completion
1789 When completing in an expression which looks up a field in a
1790 structure, @value{GDBN} also tries@footnote{The completer can be
1791 confused by certain kinds of invalid expressions. Also, it only
1792 examines the static type of the expression, not the dynamic type.} to
1793 limit completions to the field names available in the type of the
1797 (@value{GDBP}) p gdb_stdout.@kbd{M-?}
1798 magic to_fputs to_rewind
1799 to_data to_isatty to_write
1800 to_delete to_put to_write_async_safe
1805 This is because the @code{gdb_stdout} is a variable of the type
1806 @code{struct ui_file} that is defined in @value{GDBN} sources as
1813 ui_file_flush_ftype *to_flush;
1814 ui_file_write_ftype *to_write;
1815 ui_file_write_async_safe_ftype *to_write_async_safe;
1816 ui_file_fputs_ftype *to_fputs;
1817 ui_file_read_ftype *to_read;
1818 ui_file_delete_ftype *to_delete;
1819 ui_file_isatty_ftype *to_isatty;
1820 ui_file_rewind_ftype *to_rewind;
1821 ui_file_put_ftype *to_put;
1828 @section Getting Help
1829 @cindex online documentation
1832 You can always ask @value{GDBN} itself for information on its commands,
1833 using the command @code{help}.
1836 @kindex h @r{(@code{help})}
1839 You can use @code{help} (abbreviated @code{h}) with no arguments to
1840 display a short list of named classes of commands:
1844 List of classes of commands:
1846 aliases -- Aliases of other commands
1847 breakpoints -- Making program stop at certain points
1848 data -- Examining data
1849 files -- Specifying and examining files
1850 internals -- Maintenance commands
1851 obscure -- Obscure features
1852 running -- Running the program
1853 stack -- Examining the stack
1854 status -- Status inquiries
1855 support -- Support facilities
1856 tracepoints -- Tracing of program execution without
1857 stopping the program
1858 user-defined -- User-defined commands
1860 Type "help" followed by a class name for a list of
1861 commands in that class.
1862 Type "help" followed by command name for full
1864 Command name abbreviations are allowed if unambiguous.
1867 @c the above line break eliminates huge line overfull...
1869 @item help @var{class}
1870 Using one of the general help classes as an argument, you can get a
1871 list of the individual commands in that class. For example, here is the
1872 help display for the class @code{status}:
1875 (@value{GDBP}) help status
1880 @c Line break in "show" line falsifies real output, but needed
1881 @c to fit in smallbook page size.
1882 info -- Generic command for showing things
1883 about the program being debugged
1884 show -- Generic command for showing things
1887 Type "help" followed by command name for full
1889 Command name abbreviations are allowed if unambiguous.
1893 @item help @var{command}
1894 With a command name as @code{help} argument, @value{GDBN} displays a
1895 short paragraph on how to use that command.
1898 @item apropos [-v] @var{regexp}
1899 The @code{apropos} command searches through all of the @value{GDBN}
1900 commands, and their documentation, for the regular expression specified in
1901 @var{args}. It prints out all matches found. The optional flag @samp{-v},
1902 which stands for @samp{verbose}, indicates to output the full documentation
1903 of the matching commands and highlight the parts of the documentation
1904 matching @var{regexp}. For example:
1915 alias -- Define a new command that is an alias of an existing command
1916 aliases -- Aliases of other commands
1917 d -- Delete some breakpoints or auto-display expressions
1918 del -- Delete some breakpoints or auto-display expressions
1919 delete -- Delete some breakpoints or auto-display expressions
1927 apropos -v cut.*thread apply
1931 results in the below output, where @samp{cut for 'thread apply}
1932 is highlighted if styling is enabled.
1936 taas -- Apply a command to all threads (ignoring errors
1939 shortcut for 'thread apply all -s COMMAND'
1941 tfaas -- Apply a command to all frames of all threads
1942 (ignoring errors and empty output).
1943 Usage: tfaas COMMAND
1944 shortcut for 'thread apply all -s frame apply all -s COMMAND'
1949 @item complete @var{args}
1950 The @code{complete @var{args}} command lists all the possible completions
1951 for the beginning of a command. Use @var{args} to specify the beginning of the
1952 command you want completed. For example:
1958 @noindent results in:
1969 @noindent This is intended for use by @sc{gnu} Emacs.
1972 In addition to @code{help}, you can use the @value{GDBN} commands @code{info}
1973 and @code{show} to inquire about the state of your program, or the state
1974 of @value{GDBN} itself. Each command supports many topics of inquiry; this
1975 manual introduces each of them in the appropriate context. The listings
1976 under @code{info} and under @code{show} in the Command, Variable, and
1977 Function Index point to all the sub-commands. @xref{Command and Variable
1983 @kindex i @r{(@code{info})}
1985 This command (abbreviated @code{i}) is for describing the state of your
1986 program. For example, you can show the arguments passed to a function
1987 with @code{info args}, list the registers currently in use with @code{info
1988 registers}, or list the breakpoints you have set with @code{info breakpoints}.
1989 You can get a complete list of the @code{info} sub-commands with
1990 @w{@code{help info}}.
1994 You can assign the result of an expression to an environment variable with
1995 @code{set}. For example, you can set the @value{GDBN} prompt to a $-sign with
1996 @code{set prompt $}.
2000 In contrast to @code{info}, @code{show} is for describing the state of
2001 @value{GDBN} itself.
2002 You can change most of the things you can @code{show}, by using the
2003 related command @code{set}; for example, you can control what number
2004 system is used for displays with @code{set radix}, or simply inquire
2005 which is currently in use with @code{show radix}.
2008 To display all the settable parameters and their current
2009 values, you can use @code{show} with no arguments; you may also use
2010 @code{info set}. Both commands produce the same display.
2011 @c FIXME: "info set" violates the rule that "info" is for state of
2012 @c FIXME...program. Ck w/ GNU: "info set" to be called something else,
2013 @c FIXME...or change desc of rule---eg "state of prog and debugging session"?
2017 Here are several miscellaneous @code{show} subcommands, all of which are
2018 exceptional in lacking corresponding @code{set} commands:
2021 @kindex show version
2022 @cindex @value{GDBN} version number
2024 Show what version of @value{GDBN} is running. You should include this
2025 information in @value{GDBN} bug-reports. If multiple versions of
2026 @value{GDBN} are in use at your site, you may need to determine which
2027 version of @value{GDBN} you are running; as @value{GDBN} evolves, new
2028 commands are introduced, and old ones may wither away. Also, many
2029 system vendors ship variant versions of @value{GDBN}, and there are
2030 variant versions of @value{GDBN} in @sc{gnu}/Linux distributions as well.
2031 The version number is the same as the one announced when you start
2034 @kindex show copying
2035 @kindex info copying
2036 @cindex display @value{GDBN} copyright
2039 Display information about permission for copying @value{GDBN}.
2041 @kindex show warranty
2042 @kindex info warranty
2044 @itemx info warranty
2045 Display the @sc{gnu} ``NO WARRANTY'' statement, or a warranty,
2046 if your version of @value{GDBN} comes with one.
2048 @kindex show configuration
2049 @item show configuration
2050 Display detailed information about the way @value{GDBN} was configured
2051 when it was built. This displays the optional arguments passed to the
2052 @file{configure} script and also configuration parameters detected
2053 automatically by @command{configure}. When reporting a @value{GDBN}
2054 bug (@pxref{GDB Bugs}), it is important to include this information in
2060 @chapter Running Programs Under @value{GDBN}
2062 When you run a program under @value{GDBN}, you must first generate
2063 debugging information when you compile it.
2065 You may start @value{GDBN} with its arguments, if any, in an environment
2066 of your choice. If you are doing native debugging, you may redirect
2067 your program's input and output, debug an already running process, or
2068 kill a child process.
2071 * Compilation:: Compiling for debugging
2072 * Starting:: Starting your program
2073 * Arguments:: Your program's arguments
2074 * Environment:: Your program's environment
2076 * Working Directory:: Your program's working directory
2077 * Input/Output:: Your program's input and output
2078 * Attach:: Debugging an already-running process
2079 * Kill Process:: Killing the child process
2081 * Inferiors and Programs:: Debugging multiple inferiors and programs
2082 * Threads:: Debugging programs with multiple threads
2083 * Forks:: Debugging forks
2084 * Checkpoint/Restart:: Setting a @emph{bookmark} to return to later
2088 @section Compiling for Debugging
2090 In order to debug a program effectively, you need to generate
2091 debugging information when you compile it. This debugging information
2092 is stored in the object file; it describes the data type of each
2093 variable or function and the correspondence between source line numbers
2094 and addresses in the executable code.
2096 To request debugging information, specify the @samp{-g} option when you run
2099 Programs that are to be shipped to your customers are compiled with
2100 optimizations, using the @samp{-O} compiler option. However, some
2101 compilers are unable to handle the @samp{-g} and @samp{-O} options
2102 together. Using those compilers, you cannot generate optimized
2103 executables containing debugging information.
2105 @value{NGCC}, the @sc{gnu} C/C@t{++} compiler, supports @samp{-g} with or
2106 without @samp{-O}, making it possible to debug optimized code. We
2107 recommend that you @emph{always} use @samp{-g} whenever you compile a
2108 program. You may think your program is correct, but there is no sense
2109 in pushing your luck. For more information, see @ref{Optimized Code}.
2111 Older versions of the @sc{gnu} C compiler permitted a variant option
2112 @w{@samp{-gg}} for debugging information. @value{GDBN} no longer supports this
2113 format; if your @sc{gnu} C compiler has this option, do not use it.
2115 @value{GDBN} knows about preprocessor macros and can show you their
2116 expansion (@pxref{Macros}). Most compilers do not include information
2117 about preprocessor macros in the debugging information if you specify
2118 the @option{-g} flag alone. Version 3.1 and later of @value{NGCC},
2119 the @sc{gnu} C compiler, provides macro information if you are using
2120 the DWARF debugging format, and specify the option @option{-g3}.
2122 @xref{Debugging Options,,Options for Debugging Your Program or GCC,
2123 gcc, Using the @sc{gnu} Compiler Collection (GCC)}, for more
2124 information on @value{NGCC} options affecting debug information.
2126 You will have the best debugging experience if you use the latest
2127 version of the DWARF debugging format that your compiler supports.
2128 DWARF is currently the most expressive and best supported debugging
2129 format in @value{GDBN}.
2133 @section Starting your Program
2139 @kindex r @r{(@code{run})}
2142 Use the @code{run} command to start your program under @value{GDBN}.
2143 You must first specify the program name with an argument to
2144 @value{GDBN} (@pxref{Invocation, ,Getting In and Out of
2145 @value{GDBN}}), or by using the @code{file} or @code{exec-file}
2146 command (@pxref{Files, ,Commands to Specify Files}).
2150 If you are running your program in an execution environment that
2151 supports processes, @code{run} creates an inferior process and makes
2152 that process run your program. In some environments without processes,
2153 @code{run} jumps to the start of your program. Other targets,
2154 like @samp{remote}, are always running. If you get an error
2155 message like this one:
2158 The "remote" target does not support "run".
2159 Try "help target" or "continue".
2163 then use @code{continue} to run your program. You may need @code{load}
2164 first (@pxref{load}).
2166 The execution of a program is affected by certain information it
2167 receives from its superior. @value{GDBN} provides ways to specify this
2168 information, which you must do @emph{before} starting your program. (You
2169 can change it after starting your program, but such changes only affect
2170 your program the next time you start it.) This information may be
2171 divided into four categories:
2174 @item The @emph{arguments.}
2175 Specify the arguments to give your program as the arguments of the
2176 @code{run} command. If a shell is available on your target, the shell
2177 is used to pass the arguments, so that you may use normal conventions
2178 (such as wildcard expansion or variable substitution) in describing
2180 In Unix systems, you can control which shell is used with the
2181 @code{SHELL} environment variable. If you do not define @code{SHELL},
2182 @value{GDBN} uses the default shell (@file{/bin/sh}). You can disable
2183 use of any shell with the @code{set startup-with-shell} command (see
2186 @item The @emph{environment.}
2187 Your program normally inherits its environment from @value{GDBN}, but you can
2188 use the @value{GDBN} commands @code{set environment} and @code{unset
2189 environment} to change parts of the environment that affect
2190 your program. @xref{Environment, ,Your Program's Environment}.
2192 @item The @emph{working directory.}
2193 You can set your program's working directory with the command
2194 @kbd{set cwd}. If you do not set any working directory with this
2195 command, your program will inherit @value{GDBN}'s working directory if
2196 native debugging, or the remote server's working directory if remote
2197 debugging. @xref{Working Directory, ,Your Program's Working
2200 @item The @emph{standard input and output.}
2201 Your program normally uses the same device for standard input and
2202 standard output as @value{GDBN} is using. You can redirect input and output
2203 in the @code{run} command line, or you can use the @code{tty} command to
2204 set a different device for your program.
2205 @xref{Input/Output, ,Your Program's Input and Output}.
2208 @emph{Warning:} While input and output redirection work, you cannot use
2209 pipes to pass the output of the program you are debugging to another
2210 program; if you attempt this, @value{GDBN} is likely to wind up debugging the
2214 When you issue the @code{run} command, your program begins to execute
2215 immediately. @xref{Stopping, ,Stopping and Continuing}, for discussion
2216 of how to arrange for your program to stop. Once your program has
2217 stopped, you may call functions in your program, using the @code{print}
2218 or @code{call} commands. @xref{Data, ,Examining Data}.
2220 If the modification time of your symbol file has changed since the last
2221 time @value{GDBN} read its symbols, @value{GDBN} discards its symbol
2222 table, and reads it again. When it does this, @value{GDBN} tries to retain
2223 your current breakpoints.
2228 @cindex run to main procedure
2229 The name of the main procedure can vary from language to language.
2230 With C or C@t{++}, the main procedure name is always @code{main}, but
2231 other languages such as Ada do not require a specific name for their
2232 main procedure. The debugger provides a convenient way to start the
2233 execution of the program and to stop at the beginning of the main
2234 procedure, depending on the language used.
2236 The @samp{start} command does the equivalent of setting a temporary
2237 breakpoint at the beginning of the main procedure and then invoking
2238 the @samp{run} command.
2240 @cindex elaboration phase
2241 Some programs contain an @dfn{elaboration} phase where some startup code is
2242 executed before the main procedure is called. This depends on the
2243 languages used to write your program. In C@t{++}, for instance,
2244 constructors for static and global objects are executed before
2245 @code{main} is called. It is therefore possible that the debugger stops
2246 before reaching the main procedure. However, the temporary breakpoint
2247 will remain to halt execution.
2249 Specify the arguments to give to your program as arguments to the
2250 @samp{start} command. These arguments will be given verbatim to the
2251 underlying @samp{run} command. Note that the same arguments will be
2252 reused if no argument is provided during subsequent calls to
2253 @samp{start} or @samp{run}.
2255 It is sometimes necessary to debug the program during elaboration. In
2256 these cases, using the @code{start} command would stop the execution
2257 of your program too late, as the program would have already completed
2258 the elaboration phase. Under these circumstances, either insert
2259 breakpoints in your elaboration code before running your program or
2260 use the @code{starti} command.
2264 @cindex run to first instruction
2265 The @samp{starti} command does the equivalent of setting a temporary
2266 breakpoint at the first instruction of a program's execution and then
2267 invoking the @samp{run} command. For programs containing an
2268 elaboration phase, the @code{starti} command will stop execution at
2269 the start of the elaboration phase.
2271 @anchor{set exec-wrapper}
2272 @kindex set exec-wrapper
2273 @item set exec-wrapper @var{wrapper}
2274 @itemx show exec-wrapper
2275 @itemx unset exec-wrapper
2276 When @samp{exec-wrapper} is set, the specified wrapper is used to
2277 launch programs for debugging. @value{GDBN} starts your program
2278 with a shell command of the form @kbd{exec @var{wrapper}
2279 @var{program}}. Quoting is added to @var{program} and its
2280 arguments, but not to @var{wrapper}, so you should add quotes if
2281 appropriate for your shell. The wrapper runs until it executes
2282 your program, and then @value{GDBN} takes control.
2284 You can use any program that eventually calls @code{execve} with
2285 its arguments as a wrapper. Several standard Unix utilities do
2286 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
2287 with @code{exec "$@@"} will also work.
2289 For example, you can use @code{env} to pass an environment variable to
2290 the debugged program, without setting the variable in your shell's
2294 (@value{GDBP}) set exec-wrapper env 'LD_PRELOAD=libtest.so'
2298 This command is available when debugging locally on most targets, excluding
2299 @sc{djgpp}, Cygwin, MS Windows, and QNX Neutrino.
2301 @kindex set startup-with-shell
2302 @anchor{set startup-with-shell}
2303 @item set startup-with-shell
2304 @itemx set startup-with-shell on
2305 @itemx set startup-with-shell off
2306 @itemx show startup-with-shell
2307 On Unix systems, by default, if a shell is available on your target,
2308 @value{GDBN}) uses it to start your program. Arguments of the
2309 @code{run} command are passed to the shell, which does variable
2310 substitution, expands wildcard characters and performs redirection of
2311 I/O. In some circumstances, it may be useful to disable such use of a
2312 shell, for example, when debugging the shell itself or diagnosing
2313 startup failures such as:
2317 Starting program: ./a.out
2318 During startup program terminated with signal SIGSEGV, Segmentation fault.
2322 which indicates the shell or the wrapper specified with
2323 @samp{exec-wrapper} crashed, not your program. Most often, this is
2324 caused by something odd in your shell's non-interactive mode
2325 initialization file---such as @file{.cshrc} for C-shell,
2326 $@file{.zshenv} for the Z shell, or the file specified in the
2327 @samp{BASH_ENV} environment variable for BASH.
2329 @anchor{set auto-connect-native-target}
2330 @kindex set auto-connect-native-target
2331 @item set auto-connect-native-target
2332 @itemx set auto-connect-native-target on
2333 @itemx set auto-connect-native-target off
2334 @itemx show auto-connect-native-target
2336 By default, if not connected to any target yet (e.g., with
2337 @code{target remote}), the @code{run} command starts your program as a
2338 native process under @value{GDBN}, on your local machine. If you're
2339 sure you don't want to debug programs on your local machine, you can
2340 tell @value{GDBN} to not connect to the native target automatically
2341 with the @code{set auto-connect-native-target off} command.
2343 If @code{on}, which is the default, and if @value{GDBN} is not
2344 connected to a target already, the @code{run} command automaticaly
2345 connects to the native target, if one is available.
2347 If @code{off}, and if @value{GDBN} is not connected to a target
2348 already, the @code{run} command fails with an error:
2352 Don't know how to run. Try "help target".
2355 If @value{GDBN} is already connected to a target, @value{GDBN} always
2356 uses it with the @code{run} command.
2358 In any case, you can explicitly connect to the native target with the
2359 @code{target native} command. For example,
2362 (@value{GDBP}) set auto-connect-native-target off
2364 Don't know how to run. Try "help target".
2365 (@value{GDBP}) target native
2367 Starting program: ./a.out
2368 [Inferior 1 (process 10421) exited normally]
2371 In case you connected explicitly to the @code{native} target,
2372 @value{GDBN} remains connected even if all inferiors exit, ready for
2373 the next @code{run} command. Use the @code{disconnect} command to
2376 Examples of other commands that likewise respect the
2377 @code{auto-connect-native-target} setting: @code{attach}, @code{info
2378 proc}, @code{info os}.
2380 @kindex set disable-randomization
2381 @item set disable-randomization
2382 @itemx set disable-randomization on
2383 This option (enabled by default in @value{GDBN}) will turn off the native
2384 randomization of the virtual address space of the started program. This option
2385 is useful for multiple debugging sessions to make the execution better
2386 reproducible and memory addresses reusable across debugging sessions.
2388 This feature is implemented only on certain targets, including @sc{gnu}/Linux.
2389 On @sc{gnu}/Linux you can get the same behavior using
2392 (@value{GDBP}) set exec-wrapper setarch `uname -m` -R
2395 @item set disable-randomization off
2396 Leave the behavior of the started executable unchanged. Some bugs rear their
2397 ugly heads only when the program is loaded at certain addresses. If your bug
2398 disappears when you run the program under @value{GDBN}, that might be because
2399 @value{GDBN} by default disables the address randomization on platforms, such
2400 as @sc{gnu}/Linux, which do that for stand-alone programs. Use @kbd{set
2401 disable-randomization off} to try to reproduce such elusive bugs.
2403 On targets where it is available, virtual address space randomization
2404 protects the programs against certain kinds of security attacks. In these
2405 cases the attacker needs to know the exact location of a concrete executable
2406 code. Randomizing its location makes it impossible to inject jumps misusing
2407 a code at its expected addresses.
2409 Prelinking shared libraries provides a startup performance advantage but it
2410 makes addresses in these libraries predictable for privileged processes by
2411 having just unprivileged access at the target system. Reading the shared
2412 library binary gives enough information for assembling the malicious code
2413 misusing it. Still even a prelinked shared library can get loaded at a new
2414 random address just requiring the regular relocation process during the
2415 startup. Shared libraries not already prelinked are always loaded at
2416 a randomly chosen address.
2418 Position independent executables (PIE) contain position independent code
2419 similar to the shared libraries and therefore such executables get loaded at
2420 a randomly chosen address upon startup. PIE executables always load even
2421 already prelinked shared libraries at a random address. You can build such
2422 executable using @command{gcc -fPIE -pie}.
2424 Heap (malloc storage), stack and custom mmap areas are always placed randomly
2425 (as long as the randomization is enabled).
2427 @item show disable-randomization
2428 Show the current setting of the explicit disable of the native randomization of
2429 the virtual address space of the started program.
2434 @section Your Program's Arguments
2436 @cindex arguments (to your program)
2437 The arguments to your program can be specified by the arguments of the
2439 They are passed to a shell, which expands wildcard characters and
2440 performs redirection of I/O, and thence to your program. Your
2441 @code{SHELL} environment variable (if it exists) specifies what shell
2442 @value{GDBN} uses. If you do not define @code{SHELL}, @value{GDBN} uses
2443 the default shell (@file{/bin/sh} on Unix).
2445 On non-Unix systems, the program is usually invoked directly by
2446 @value{GDBN}, which emulates I/O redirection via the appropriate system
2447 calls, and the wildcard characters are expanded by the startup code of
2448 the program, not by the shell.
2450 @code{run} with no arguments uses the same arguments used by the previous
2451 @code{run}, or those set by the @code{set args} command.
2456 Specify the arguments to be used the next time your program is run. If
2457 @code{set args} has no arguments, @code{run} executes your program
2458 with no arguments. Once you have run your program with arguments,
2459 using @code{set args} before the next @code{run} is the only way to run
2460 it again without arguments.
2464 Show the arguments to give your program when it is started.
2468 @section Your Program's Environment
2470 @cindex environment (of your program)
2471 The @dfn{environment} consists of a set of environment variables and
2472 their values. Environment variables conventionally record such things as
2473 your user name, your home directory, your terminal type, and your search
2474 path for programs to run. Usually you set up environment variables with
2475 the shell and they are inherited by all the other programs you run. When
2476 debugging, it can be useful to try running your program with a modified
2477 environment without having to start @value{GDBN} over again.
2481 @item path @var{directory}
2482 Add @var{directory} to the front of the @code{PATH} environment variable
2483 (the search path for executables) that will be passed to your program.
2484 The value of @code{PATH} used by @value{GDBN} does not change.
2485 You may specify several directory names, separated by whitespace or by a
2486 system-dependent separator character (@samp{:} on Unix, @samp{;} on
2487 MS-DOS and MS-Windows). If @var{directory} is already in the path, it
2488 is moved to the front, so it is searched sooner.
2490 You can use the string @samp{$cwd} to refer to whatever is the current
2491 working directory at the time @value{GDBN} searches the path. If you
2492 use @samp{.} instead, it refers to the directory where you executed the
2493 @code{path} command. @value{GDBN} replaces @samp{.} in the
2494 @var{directory} argument (with the current path) before adding
2495 @var{directory} to the search path.
2496 @c 'path' is explicitly nonrepeatable, but RMS points out it is silly to
2497 @c document that, since repeating it would be a no-op.
2501 Display the list of search paths for executables (the @code{PATH}
2502 environment variable).
2504 @kindex show environment
2505 @item show environment @r{[}@var{varname}@r{]}
2506 Print the value of environment variable @var{varname} to be given to
2507 your program when it starts. If you do not supply @var{varname},
2508 print the names and values of all environment variables to be given to
2509 your program. You can abbreviate @code{environment} as @code{env}.
2511 @kindex set environment
2512 @anchor{set environment}
2513 @item set environment @var{varname} @r{[}=@var{value}@r{]}
2514 Set environment variable @var{varname} to @var{value}. The value
2515 changes for your program (and the shell @value{GDBN} uses to launch
2516 it), not for @value{GDBN} itself. The @var{value} may be any string; the
2517 values of environment variables are just strings, and any
2518 interpretation is supplied by your program itself. The @var{value}
2519 parameter is optional; if it is eliminated, the variable is set to a
2521 @c "any string" here does not include leading, trailing
2522 @c blanks. Gnu asks: does anyone care?
2524 For example, this command:
2531 tells the debugged program, when subsequently run, that its user is named
2532 @samp{foo}. (The spaces around @samp{=} are used for clarity here; they
2533 are not actually required.)
2535 Note that on Unix systems, @value{GDBN} runs your program via a shell,
2536 which also inherits the environment set with @code{set environment}.
2537 If necessary, you can avoid that by using the @samp{env} program as a
2538 wrapper instead of using @code{set environment}. @xref{set
2539 exec-wrapper}, for an example doing just that.
2541 Environment variables that are set by the user are also transmitted to
2542 @command{gdbserver} to be used when starting the remote inferior.
2543 @pxref{QEnvironmentHexEncoded}.
2545 @kindex unset environment
2546 @anchor{unset environment}
2547 @item unset environment @var{varname}
2548 Remove variable @var{varname} from the environment to be passed to your
2549 program. This is different from @samp{set env @var{varname} =};
2550 @code{unset environment} removes the variable from the environment,
2551 rather than assigning it an empty value.
2553 Environment variables that are unset by the user are also unset on
2554 @command{gdbserver} when starting the remote inferior.
2555 @pxref{QEnvironmentUnset}.
2558 @emph{Warning:} On Unix systems, @value{GDBN} runs your program using
2559 the shell indicated by your @code{SHELL} environment variable if it
2560 exists (or @code{/bin/sh} if not). If your @code{SHELL} variable
2561 names a shell that runs an initialization file when started
2562 non-interactively---such as @file{.cshrc} for C-shell, $@file{.zshenv}
2563 for the Z shell, or the file specified in the @samp{BASH_ENV}
2564 environment variable for BASH---any variables you set in that file
2565 affect your program. You may wish to move setting of environment
2566 variables to files that are only run when you sign on, such as
2567 @file{.login} or @file{.profile}.
2569 @node Working Directory
2570 @section Your Program's Working Directory
2572 @cindex working directory (of your program)
2573 Each time you start your program with @code{run}, the inferior will be
2574 initialized with the current working directory specified by the
2575 @kbd{set cwd} command. If no directory has been specified by this
2576 command, then the inferior will inherit @value{GDBN}'s current working
2577 directory as its working directory if native debugging, or it will
2578 inherit the remote server's current working directory if remote
2583 @cindex change inferior's working directory
2584 @anchor{set cwd command}
2585 @item set cwd @r{[}@var{directory}@r{]}
2586 Set the inferior's working directory to @var{directory}, which will be
2587 @code{glob}-expanded in order to resolve tildes (@file{~}). If no
2588 argument has been specified, the command clears the setting and resets
2589 it to an empty state. This setting has no effect on @value{GDBN}'s
2590 working directory, and it only takes effect the next time you start
2591 the inferior. The @file{~} in @var{directory} is a short for the
2592 @dfn{home directory}, usually pointed to by the @env{HOME} environment
2593 variable. On MS-Windows, if @env{HOME} is not defined, @value{GDBN}
2594 uses the concatenation of @env{HOMEDRIVE} and @env{HOMEPATH} as
2597 You can also change @value{GDBN}'s current working directory by using
2598 the @code{cd} command.
2602 @cindex show inferior's working directory
2604 Show the inferior's working directory. If no directory has been
2605 specified by @kbd{set cwd}, then the default inferior's working
2606 directory is the same as @value{GDBN}'s working directory.
2609 @cindex change @value{GDBN}'s working directory
2611 @item cd @r{[}@var{directory}@r{]}
2612 Set the @value{GDBN} working directory to @var{directory}. If not
2613 given, @var{directory} uses @file{'~'}.
2615 The @value{GDBN} working directory serves as a default for the
2616 commands that specify files for @value{GDBN} to operate on.
2617 @xref{Files, ,Commands to Specify Files}.
2618 @xref{set cwd command}.
2622 Print the @value{GDBN} working directory.
2625 It is generally impossible to find the current working directory of
2626 the process being debugged (since a program can change its directory
2627 during its run). If you work on a system where @value{GDBN} supports
2628 the @code{info proc} command (@pxref{Process Information}), you can
2629 use the @code{info proc} command to find out the
2630 current working directory of the debuggee.
2633 @section Your Program's Input and Output
2638 By default, the program you run under @value{GDBN} does input and output to
2639 the same terminal that @value{GDBN} uses. @value{GDBN} switches the terminal
2640 to its own terminal modes to interact with you, but it records the terminal
2641 modes your program was using and switches back to them when you continue
2642 running your program.
2645 @kindex info terminal
2647 Displays information recorded by @value{GDBN} about the terminal modes your
2651 You can redirect your program's input and/or output using shell
2652 redirection with the @code{run} command. For example,
2659 starts your program, diverting its output to the file @file{outfile}.
2662 @cindex controlling terminal
2663 Another way to specify where your program should do input and output is
2664 with the @code{tty} command. This command accepts a file name as
2665 argument, and causes this file to be the default for future @code{run}
2666 commands. It also resets the controlling terminal for the child
2667 process, for future @code{run} commands. For example,
2674 directs that processes started with subsequent @code{run} commands
2675 default to do input and output on the terminal @file{/dev/ttyb} and have
2676 that as their controlling terminal.
2678 An explicit redirection in @code{run} overrides the @code{tty} command's
2679 effect on the input/output device, but not its effect on the controlling
2682 When you use the @code{tty} command or redirect input in the @code{run}
2683 command, only the input @emph{for your program} is affected. The input
2684 for @value{GDBN} still comes from your terminal. @code{tty} is an alias
2685 for @code{set inferior-tty}.
2687 @cindex inferior tty
2688 @cindex set inferior controlling terminal
2689 You can use the @code{show inferior-tty} command to tell @value{GDBN} to
2690 display the name of the terminal that will be used for future runs of your
2694 @item set inferior-tty [ @var{tty} ]
2695 @kindex set inferior-tty
2696 Set the tty for the program being debugged to @var{tty}. Omitting @var{tty}
2697 restores the default behavior, which is to use the same terminal as
2700 @item show inferior-tty
2701 @kindex show inferior-tty
2702 Show the current tty for the program being debugged.
2706 @section Debugging an Already-running Process
2711 @item attach @var{process-id}
2712 This command attaches to a running process---one that was started
2713 outside @value{GDBN}. (@code{info files} shows your active
2714 targets.) The command takes as argument a process ID. The usual way to
2715 find out the @var{process-id} of a Unix process is with the @code{ps} utility,
2716 or with the @samp{jobs -l} shell command.
2718 @code{attach} does not repeat if you press @key{RET} a second time after
2719 executing the command.
2722 To use @code{attach}, your program must be running in an environment
2723 which supports processes; for example, @code{attach} does not work for
2724 programs on bare-board targets that lack an operating system. You must
2725 also have permission to send the process a signal.
2727 When you use @code{attach}, the debugger finds the program running in
2728 the process first by looking in the current working directory, then (if
2729 the program is not found) by using the source file search path
2730 (@pxref{Source Path, ,Specifying Source Directories}). You can also use
2731 the @code{file} command to load the program. @xref{Files, ,Commands to
2734 The first thing @value{GDBN} does after arranging to debug the specified
2735 process is to stop it. You can examine and modify an attached process
2736 with all the @value{GDBN} commands that are ordinarily available when
2737 you start processes with @code{run}. You can insert breakpoints; you
2738 can step and continue; you can modify storage. If you would rather the
2739 process continue running, you may use the @code{continue} command after
2740 attaching @value{GDBN} to the process.
2745 When you have finished debugging the attached process, you can use the
2746 @code{detach} command to release it from @value{GDBN} control. Detaching
2747 the process continues its execution. After the @code{detach} command,
2748 that process and @value{GDBN} become completely independent once more, and you
2749 are ready to @code{attach} another process or start one with @code{run}.
2750 @code{detach} does not repeat if you press @key{RET} again after
2751 executing the command.
2754 If you exit @value{GDBN} while you have an attached process, you detach
2755 that process. If you use the @code{run} command, you kill that process.
2756 By default, @value{GDBN} asks for confirmation if you try to do either of these
2757 things; you can control whether or not you need to confirm by using the
2758 @code{set confirm} command (@pxref{Messages/Warnings, ,Optional Warnings and
2762 @section Killing the Child Process
2767 Kill the child process in which your program is running under @value{GDBN}.
2770 This command is useful if you wish to debug a core dump instead of a
2771 running process. @value{GDBN} ignores any core dump file while your program
2774 On some operating systems, a program cannot be executed outside @value{GDBN}
2775 while you have breakpoints set on it inside @value{GDBN}. You can use the
2776 @code{kill} command in this situation to permit running your program
2777 outside the debugger.
2779 The @code{kill} command is also useful if you wish to recompile and
2780 relink your program, since on many systems it is impossible to modify an
2781 executable file while it is running in a process. In this case, when you
2782 next type @code{run}, @value{GDBN} notices that the file has changed, and
2783 reads the symbol table again (while trying to preserve your current
2784 breakpoint settings).
2786 @node Inferiors and Programs
2787 @section Debugging Multiple Inferiors and Programs
2789 @value{GDBN} lets you run and debug multiple programs in a single
2790 session. In addition, @value{GDBN} on some systems may let you run
2791 several programs simultaneously (otherwise you have to exit from one
2792 before starting another). In the most general case, you can have
2793 multiple threads of execution in each of multiple processes, launched
2794 from multiple executables.
2797 @value{GDBN} represents the state of each program execution with an
2798 object called an @dfn{inferior}. An inferior typically corresponds to
2799 a process, but is more general and applies also to targets that do not
2800 have processes. Inferiors may be created before a process runs, and
2801 may be retained after a process exits. Inferiors have unique
2802 identifiers that are different from process ids. Usually each
2803 inferior will also have its own distinct address space, although some
2804 embedded targets may have several inferiors running in different parts
2805 of a single address space. Each inferior may in turn have multiple
2806 threads running in it.
2808 To find out what inferiors exist at any moment, use @w{@code{info
2812 @kindex info inferiors [ @var{id}@dots{} ]
2813 @item info inferiors
2814 Print a list of all inferiors currently being managed by @value{GDBN}.
2815 By default all inferiors are printed, but the argument @var{id}@dots{}
2816 -- a space separated list of inferior numbers -- can be used to limit
2817 the display to just the requested inferiors.
2819 @value{GDBN} displays for each inferior (in this order):
2823 the inferior number assigned by @value{GDBN}
2826 the target system's inferior identifier
2829 the name of the executable the inferior is running.
2834 An asterisk @samp{*} preceding the @value{GDBN} inferior number
2835 indicates the current inferior.
2839 @c end table here to get a little more width for example
2842 (@value{GDBP}) info inferiors
2843 Num Description Executable
2844 2 process 2307 hello
2845 * 1 process 3401 goodbye
2848 To switch focus between inferiors, use the @code{inferior} command:
2851 @kindex inferior @var{infno}
2852 @item inferior @var{infno}
2853 Make inferior number @var{infno} the current inferior. The argument
2854 @var{infno} is the inferior number assigned by @value{GDBN}, as shown
2855 in the first field of the @samp{info inferiors} display.
2858 @vindex $_inferior@r{, convenience variable}
2859 The debugger convenience variable @samp{$_inferior} contains the
2860 number of the current inferior. You may find this useful in writing
2861 breakpoint conditional expressions, command scripts, and so forth.
2862 @xref{Convenience Vars,, Convenience Variables}, for general
2863 information on convenience variables.
2865 You can get multiple executables into a debugging session via the
2866 @code{add-inferior} and @w{@code{clone-inferior}} commands. On some
2867 systems @value{GDBN} can add inferiors to the debug session
2868 automatically by following calls to @code{fork} and @code{exec}. To
2869 remove inferiors from the debugging session use the
2870 @w{@code{remove-inferiors}} command.
2873 @kindex add-inferior
2874 @item add-inferior [ -copies @var{n} ] [ -exec @var{executable} ]
2875 Adds @var{n} inferiors to be run using @var{executable} as the
2876 executable; @var{n} defaults to 1. If no executable is specified,
2877 the inferiors begins empty, with no program. You can still assign or
2878 change the program assigned to the inferior at any time by using the
2879 @code{file} command with the executable name as its argument.
2881 @kindex clone-inferior
2882 @item clone-inferior [ -copies @var{n} ] [ @var{infno} ]
2883 Adds @var{n} inferiors ready to execute the same program as inferior
2884 @var{infno}; @var{n} defaults to 1, and @var{infno} defaults to the
2885 number of the current inferior. This is a convenient command when you
2886 want to run another instance of the inferior you are debugging.
2889 (@value{GDBP}) info inferiors
2890 Num Description Executable
2891 * 1 process 29964 helloworld
2892 (@value{GDBP}) clone-inferior
2895 (@value{GDBP}) info inferiors
2896 Num Description Executable
2898 * 1 process 29964 helloworld
2901 You can now simply switch focus to inferior 2 and run it.
2903 @kindex remove-inferiors
2904 @item remove-inferiors @var{infno}@dots{}
2905 Removes the inferior or inferiors @var{infno}@dots{}. It is not
2906 possible to remove an inferior that is running with this command. For
2907 those, use the @code{kill} or @code{detach} command first.
2911 To quit debugging one of the running inferiors that is not the current
2912 inferior, you can either detach from it by using the @w{@code{detach
2913 inferior}} command (allowing it to run independently), or kill it
2914 using the @w{@code{kill inferiors}} command:
2917 @kindex detach inferiors @var{infno}@dots{}
2918 @item detach inferior @var{infno}@dots{}
2919 Detach from the inferior or inferiors identified by @value{GDBN}
2920 inferior number(s) @var{infno}@dots{}. Note that the inferior's entry
2921 still stays on the list of inferiors shown by @code{info inferiors},
2922 but its Description will show @samp{<null>}.
2924 @kindex kill inferiors @var{infno}@dots{}
2925 @item kill inferiors @var{infno}@dots{}
2926 Kill the inferior or inferiors identified by @value{GDBN} inferior
2927 number(s) @var{infno}@dots{}. Note that the inferior's entry still
2928 stays on the list of inferiors shown by @code{info inferiors}, but its
2929 Description will show @samp{<null>}.
2932 After the successful completion of a command such as @code{detach},
2933 @code{detach inferiors}, @code{kill} or @code{kill inferiors}, or after
2934 a normal process exit, the inferior is still valid and listed with
2935 @code{info inferiors}, ready to be restarted.
2938 To be notified when inferiors are started or exit under @value{GDBN}'s
2939 control use @w{@code{set print inferior-events}}:
2942 @kindex set print inferior-events
2943 @cindex print messages on inferior start and exit
2944 @item set print inferior-events
2945 @itemx set print inferior-events on
2946 @itemx set print inferior-events off
2947 The @code{set print inferior-events} command allows you to enable or
2948 disable printing of messages when @value{GDBN} notices that new
2949 inferiors have started or that inferiors have exited or have been
2950 detached. By default, these messages will not be printed.
2952 @kindex show print inferior-events
2953 @item show print inferior-events
2954 Show whether messages will be printed when @value{GDBN} detects that
2955 inferiors have started, exited or have been detached.
2958 Many commands will work the same with multiple programs as with a
2959 single program: e.g., @code{print myglobal} will simply display the
2960 value of @code{myglobal} in the current inferior.
2963 Occasionaly, when debugging @value{GDBN} itself, it may be useful to
2964 get more info about the relationship of inferiors, programs, address
2965 spaces in a debug session. You can do that with the @w{@code{maint
2966 info program-spaces}} command.
2969 @kindex maint info program-spaces
2970 @item maint info program-spaces
2971 Print a list of all program spaces currently being managed by
2974 @value{GDBN} displays for each program space (in this order):
2978 the program space number assigned by @value{GDBN}
2981 the name of the executable loaded into the program space, with e.g.,
2982 the @code{file} command.
2987 An asterisk @samp{*} preceding the @value{GDBN} program space number
2988 indicates the current program space.
2990 In addition, below each program space line, @value{GDBN} prints extra
2991 information that isn't suitable to display in tabular form. For
2992 example, the list of inferiors bound to the program space.
2995 (@value{GDBP}) maint info program-spaces
2999 Bound inferiors: ID 1 (process 21561)
3002 Here we can see that no inferior is running the program @code{hello},
3003 while @code{process 21561} is running the program @code{goodbye}. On
3004 some targets, it is possible that multiple inferiors are bound to the
3005 same program space. The most common example is that of debugging both
3006 the parent and child processes of a @code{vfork} call. For example,
3009 (@value{GDBP}) maint info program-spaces
3012 Bound inferiors: ID 2 (process 18050), ID 1 (process 18045)
3015 Here, both inferior 2 and inferior 1 are running in the same program
3016 space as a result of inferior 1 having executed a @code{vfork} call.
3020 @section Debugging Programs with Multiple Threads
3022 @cindex threads of execution
3023 @cindex multiple threads
3024 @cindex switching threads
3025 In some operating systems, such as GNU/Linux and Solaris, a single program
3026 may have more than one @dfn{thread} of execution. The precise semantics
3027 of threads differ from one operating system to another, but in general
3028 the threads of a single program are akin to multiple processes---except
3029 that they share one address space (that is, they can all examine and
3030 modify the same variables). On the other hand, each thread has its own
3031 registers and execution stack, and perhaps private memory.
3033 @value{GDBN} provides these facilities for debugging multi-thread
3037 @item automatic notification of new threads
3038 @item @samp{thread @var{thread-id}}, a command to switch among threads
3039 @item @samp{info threads}, a command to inquire about existing threads
3040 @item @samp{thread apply [@var{thread-id-list} | all] @var{args}},
3041 a command to apply a command to a list of threads
3042 @item thread-specific breakpoints
3043 @item @samp{set print thread-events}, which controls printing of
3044 messages on thread start and exit.
3045 @item @samp{set libthread-db-search-path @var{path}}, which lets
3046 the user specify which @code{libthread_db} to use if the default choice
3047 isn't compatible with the program.
3050 @cindex focus of debugging
3051 @cindex current thread
3052 The @value{GDBN} thread debugging facility allows you to observe all
3053 threads while your program runs---but whenever @value{GDBN} takes
3054 control, one thread in particular is always the focus of debugging.
3055 This thread is called the @dfn{current thread}. Debugging commands show
3056 program information from the perspective of the current thread.
3058 @cindex @code{New} @var{systag} message
3059 @cindex thread identifier (system)
3060 @c FIXME-implementors!! It would be more helpful if the [New...] message
3061 @c included GDB's numeric thread handle, so you could just go to that
3062 @c thread without first checking `info threads'.
3063 Whenever @value{GDBN} detects a new thread in your program, it displays
3064 the target system's identification for the thread with a message in the
3065 form @samp{[New @var{systag}]}, where @var{systag} is a thread identifier
3066 whose form varies depending on the particular system. For example, on
3067 @sc{gnu}/Linux, you might see
3070 [New Thread 0x41e02940 (LWP 25582)]
3074 when @value{GDBN} notices a new thread. In contrast, on other systems,
3075 the @var{systag} is simply something like @samp{process 368}, with no
3078 @c FIXME!! (1) Does the [New...] message appear even for the very first
3079 @c thread of a program, or does it only appear for the
3080 @c second---i.e.@: when it becomes obvious we have a multithread
3082 @c (2) *Is* there necessarily a first thread always? Or do some
3083 @c multithread systems permit starting a program with multiple
3084 @c threads ab initio?
3086 @anchor{thread numbers}
3087 @cindex thread number, per inferior
3088 @cindex thread identifier (GDB)
3089 For debugging purposes, @value{GDBN} associates its own thread number
3090 ---always a single integer---with each thread of an inferior. This
3091 number is unique between all threads of an inferior, but not unique
3092 between threads of different inferiors.
3094 @cindex qualified thread ID
3095 You can refer to a given thread in an inferior using the qualified
3096 @var{inferior-num}.@var{thread-num} syntax, also known as
3097 @dfn{qualified thread ID}, with @var{inferior-num} being the inferior
3098 number and @var{thread-num} being the thread number of the given
3099 inferior. For example, thread @code{2.3} refers to thread number 3 of
3100 inferior 2. If you omit @var{inferior-num} (e.g., @code{thread 3}),
3101 then @value{GDBN} infers you're referring to a thread of the current
3104 Until you create a second inferior, @value{GDBN} does not show the
3105 @var{inferior-num} part of thread IDs, even though you can always use
3106 the full @var{inferior-num}.@var{thread-num} form to refer to threads
3107 of inferior 1, the initial inferior.
3109 @anchor{thread ID lists}
3110 @cindex thread ID lists
3111 Some commands accept a space-separated @dfn{thread ID list} as
3112 argument. A list element can be:
3116 A thread ID as shown in the first field of the @samp{info threads}
3117 display, with or without an inferior qualifier. E.g., @samp{2.1} or
3121 A range of thread numbers, again with or without an inferior
3122 qualifier, as in @var{inf}.@var{thr1}-@var{thr2} or
3123 @var{thr1}-@var{thr2}. E.g., @samp{1.2-4} or @samp{2-4}.
3126 All threads of an inferior, specified with a star wildcard, with or
3127 without an inferior qualifier, as in @var{inf}.@code{*} (e.g.,
3128 @samp{1.*}) or @code{*}. The former refers to all threads of the
3129 given inferior, and the latter form without an inferior qualifier
3130 refers to all threads of the current inferior.
3134 For example, if the current inferior is 1, and inferior 7 has one
3135 thread with ID 7.1, the thread list @samp{1 2-3 4.5 6.7-9 7.*}
3136 includes threads 1 to 3 of inferior 1, thread 5 of inferior 4, threads
3137 7 to 9 of inferior 6 and all threads of inferior 7. That is, in
3138 expanded qualified form, the same as @samp{1.1 1.2 1.3 4.5 6.7 6.8 6.9
3142 @anchor{global thread numbers}
3143 @cindex global thread number
3144 @cindex global thread identifier (GDB)
3145 In addition to a @emph{per-inferior} number, each thread is also
3146 assigned a unique @emph{global} number, also known as @dfn{global
3147 thread ID}, a single integer. Unlike the thread number component of
3148 the thread ID, no two threads have the same global ID, even when
3149 you're debugging multiple inferiors.
3151 From @value{GDBN}'s perspective, a process always has at least one
3152 thread. In other words, @value{GDBN} assigns a thread number to the
3153 program's ``main thread'' even if the program is not multi-threaded.
3155 @vindex $_thread@r{, convenience variable}
3156 @vindex $_gthread@r{, convenience variable}
3157 The debugger convenience variables @samp{$_thread} and
3158 @samp{$_gthread} contain, respectively, the per-inferior thread number
3159 and the global thread number of the current thread. You may find this
3160 useful in writing breakpoint conditional expressions, command scripts,
3161 and so forth. @xref{Convenience Vars,, Convenience Variables}, for
3162 general information on convenience variables.
3164 If @value{GDBN} detects the program is multi-threaded, it augments the
3165 usual message about stopping at a breakpoint with the ID and name of
3166 the thread that hit the breakpoint.
3169 Thread 2 "client" hit Breakpoint 1, send_message () at client.c:68
3172 Likewise when the program receives a signal:
3175 Thread 1 "main" received signal SIGINT, Interrupt.
3179 @kindex info threads
3180 @item info threads @r{[}@var{thread-id-list}@r{]}
3182 Display information about one or more threads. With no arguments
3183 displays information about all threads. You can specify the list of
3184 threads that you want to display using the thread ID list syntax
3185 (@pxref{thread ID lists}).
3187 @value{GDBN} displays for each thread (in this order):
3191 the per-inferior thread number assigned by @value{GDBN}
3194 the global thread number assigned by @value{GDBN}, if the @samp{-gid}
3195 option was specified
3198 the target system's thread identifier (@var{systag})
3201 the thread's name, if one is known. A thread can either be named by
3202 the user (see @code{thread name}, below), or, in some cases, by the
3206 the current stack frame summary for that thread
3210 An asterisk @samp{*} to the left of the @value{GDBN} thread number
3211 indicates the current thread.
3215 @c end table here to get a little more width for example
3218 (@value{GDBP}) info threads
3220 * 1 process 35 thread 13 main (argc=1, argv=0x7ffffff8)
3221 2 process 35 thread 23 0x34e5 in sigpause ()
3222 3 process 35 thread 27 0x34e5 in sigpause ()
3226 If you're debugging multiple inferiors, @value{GDBN} displays thread
3227 IDs using the qualified @var{inferior-num}.@var{thread-num} format.
3228 Otherwise, only @var{thread-num} is shown.
3230 If you specify the @samp{-gid} option, @value{GDBN} displays a column
3231 indicating each thread's global thread ID:
3234 (@value{GDBP}) info threads
3235 Id GId Target Id Frame
3236 1.1 1 process 35 thread 13 main (argc=1, argv=0x7ffffff8)
3237 1.2 3 process 35 thread 23 0x34e5 in sigpause ()
3238 1.3 4 process 35 thread 27 0x34e5 in sigpause ()
3239 * 2.1 2 process 65 thread 1 main (argc=1, argv=0x7ffffff8)
3242 On Solaris, you can display more information about user threads with a
3243 Solaris-specific command:
3246 @item maint info sol-threads
3247 @kindex maint info sol-threads
3248 @cindex thread info (Solaris)
3249 Display info on Solaris user threads.
3253 @kindex thread @var{thread-id}
3254 @item thread @var{thread-id}
3255 Make thread ID @var{thread-id} the current thread. The command
3256 argument @var{thread-id} is the @value{GDBN} thread ID, as shown in
3257 the first field of the @samp{info threads} display, with or without an
3258 inferior qualifier (e.g., @samp{2.1} or @samp{1}).
3260 @value{GDBN} responds by displaying the system identifier of the
3261 thread you selected, and its current stack frame summary:
3264 (@value{GDBP}) thread 2
3265 [Switching to thread 2 (Thread 0xb7fdab70 (LWP 12747))]
3266 #0 some_function (ignore=0x0) at example.c:8
3267 8 printf ("hello\n");
3271 As with the @samp{[New @dots{}]} message, the form of the text after
3272 @samp{Switching to} depends on your system's conventions for identifying
3275 @kindex thread apply
3276 @cindex apply command to several threads
3277 @item thread apply [@var{thread-id-list} | all [-ascending]] [@var{flag}]@dots{} @var{command}
3278 The @code{thread apply} command allows you to apply the named
3279 @var{command} to one or more threads. Specify the threads that you
3280 want affected using the thread ID list syntax (@pxref{thread ID
3281 lists}), or specify @code{all} to apply to all threads. To apply a
3282 command to all threads in descending order, type @kbd{thread apply all
3283 @var{command}}. To apply a command to all threads in ascending order,
3284 type @kbd{thread apply all -ascending @var{command}}.
3286 The @var{flag} arguments control what output to produce and how to handle
3287 errors raised when applying @var{command} to a thread. @var{flag}
3288 must start with a @code{-} directly followed by one letter in
3289 @code{qcs}. If several flags are provided, they must be given
3290 individually, such as @code{-c -q}.
3292 By default, @value{GDBN} displays some thread information before the
3293 output produced by @var{command}, and an error raised during the
3294 execution of a @var{command} will abort @code{thread apply}. The
3295 following flags can be used to fine-tune this behavior:
3299 The flag @code{-c}, which stands for @samp{continue}, causes any
3300 errors in @var{command} to be displayed, and the execution of
3301 @code{thread apply} then continues.
3303 The flag @code{-s}, which stands for @samp{silent}, causes any errors
3304 or empty output produced by a @var{command} to be silently ignored.
3305 That is, the execution continues, but the thread information and errors
3308 The flag @code{-q} (@samp{quiet}) disables printing the thread
3312 Flags @code{-c} and @code{-s} cannot be used together.
3315 @cindex apply command to all threads (ignoring errors and empty output)
3316 @item taas @var{command}
3317 Shortcut for @code{thread apply all -s @var{command}}.
3318 Applies @var{command} on all threads, ignoring errors and empty output.
3321 @cindex apply a command to all frames of all threads (ignoring errors and empty output)
3322 @item tfaas @var{command}
3323 Shortcut for @code{thread apply all -s frame apply all -s @var{command}}.
3324 Applies @var{command} on all frames of all threads, ignoring errors
3325 and empty output. Note that the flag @code{-s} is specified twice:
3326 The first @code{-s} ensures that @code{thread apply} only shows the thread
3327 information of the threads for which @code{frame apply} produces
3328 some output. The second @code{-s} is needed to ensure that @code{frame
3329 apply} shows the frame information of a frame only if the
3330 @var{command} successfully produced some output.
3332 It can for example be used to print a local variable or a function
3333 argument without knowing the thread or frame where this variable or argument
3336 (@value{GDBP}) tfaas p some_local_var_i_do_not_remember_where_it_is
3341 @cindex name a thread
3342 @item thread name [@var{name}]
3343 This command assigns a name to the current thread. If no argument is
3344 given, any existing user-specified name is removed. The thread name
3345 appears in the @samp{info threads} display.
3347 On some systems, such as @sc{gnu}/Linux, @value{GDBN} is able to
3348 determine the name of the thread as given by the OS. On these
3349 systems, a name specified with @samp{thread name} will override the
3350 system-give name, and removing the user-specified name will cause
3351 @value{GDBN} to once again display the system-specified name.
3354 @cindex search for a thread
3355 @item thread find [@var{regexp}]
3356 Search for and display thread ids whose name or @var{systag}
3357 matches the supplied regular expression.
3359 As well as being the complement to the @samp{thread name} command,
3360 this command also allows you to identify a thread by its target
3361 @var{systag}. For instance, on @sc{gnu}/Linux, the target @var{systag}
3365 (@value{GDBN}) thread find 26688
3366 Thread 4 has target id 'Thread 0x41e02940 (LWP 26688)'
3367 (@value{GDBN}) info thread 4
3369 4 Thread 0x41e02940 (LWP 26688) 0x00000031ca6cd372 in select ()
3372 @kindex set print thread-events
3373 @cindex print messages on thread start and exit
3374 @item set print thread-events
3375 @itemx set print thread-events on
3376 @itemx set print thread-events off
3377 The @code{set print thread-events} command allows you to enable or
3378 disable printing of messages when @value{GDBN} notices that new threads have
3379 started or that threads have exited. By default, these messages will
3380 be printed if detection of these events is supported by the target.
3381 Note that these messages cannot be disabled on all targets.
3383 @kindex show print thread-events
3384 @item show print thread-events
3385 Show whether messages will be printed when @value{GDBN} detects that threads
3386 have started and exited.
3389 @xref{Thread Stops,,Stopping and Starting Multi-thread Programs}, for
3390 more information about how @value{GDBN} behaves when you stop and start
3391 programs with multiple threads.
3393 @xref{Set Watchpoints,,Setting Watchpoints}, for information about
3394 watchpoints in programs with multiple threads.
3396 @anchor{set libthread-db-search-path}
3398 @kindex set libthread-db-search-path
3399 @cindex search path for @code{libthread_db}
3400 @item set libthread-db-search-path @r{[}@var{path}@r{]}
3401 If this variable is set, @var{path} is a colon-separated list of
3402 directories @value{GDBN} will use to search for @code{libthread_db}.
3403 If you omit @var{path}, @samp{libthread-db-search-path} will be reset to
3404 its default value (@code{$sdir:$pdir} on @sc{gnu}/Linux and Solaris systems).
3405 Internally, the default value comes from the @code{LIBTHREAD_DB_SEARCH_PATH}
3408 On @sc{gnu}/Linux and Solaris systems, @value{GDBN} uses a ``helper''
3409 @code{libthread_db} library to obtain information about threads in the
3410 inferior process. @value{GDBN} will use @samp{libthread-db-search-path}
3411 to find @code{libthread_db}. @value{GDBN} also consults first if inferior
3412 specific thread debugging library loading is enabled
3413 by @samp{set auto-load libthread-db} (@pxref{libthread_db.so.1 file}).
3415 A special entry @samp{$sdir} for @samp{libthread-db-search-path}
3416 refers to the default system directories that are
3417 normally searched for loading shared libraries. The @samp{$sdir} entry
3418 is the only kind not needing to be enabled by @samp{set auto-load libthread-db}
3419 (@pxref{libthread_db.so.1 file}).
3421 A special entry @samp{$pdir} for @samp{libthread-db-search-path}
3422 refers to the directory from which @code{libpthread}
3423 was loaded in the inferior process.
3425 For any @code{libthread_db} library @value{GDBN} finds in above directories,
3426 @value{GDBN} attempts to initialize it with the current inferior process.
3427 If this initialization fails (which could happen because of a version
3428 mismatch between @code{libthread_db} and @code{libpthread}), @value{GDBN}
3429 will unload @code{libthread_db}, and continue with the next directory.
3430 If none of @code{libthread_db} libraries initialize successfully,
3431 @value{GDBN} will issue a warning and thread debugging will be disabled.
3433 Setting @code{libthread-db-search-path} is currently implemented
3434 only on some platforms.
3436 @kindex show libthread-db-search-path
3437 @item show libthread-db-search-path
3438 Display current libthread_db search path.
3440 @kindex set debug libthread-db
3441 @kindex show debug libthread-db
3442 @cindex debugging @code{libthread_db}
3443 @item set debug libthread-db
3444 @itemx show debug libthread-db
3445 Turns on or off display of @code{libthread_db}-related events.
3446 Use @code{1} to enable, @code{0} to disable.
3450 @section Debugging Forks
3452 @cindex fork, debugging programs which call
3453 @cindex multiple processes
3454 @cindex processes, multiple
3455 On most systems, @value{GDBN} has no special support for debugging
3456 programs which create additional processes using the @code{fork}
3457 function. When a program forks, @value{GDBN} will continue to debug the
3458 parent process and the child process will run unimpeded. If you have
3459 set a breakpoint in any code which the child then executes, the child
3460 will get a @code{SIGTRAP} signal which (unless it catches the signal)
3461 will cause it to terminate.
3463 However, if you want to debug the child process there is a workaround
3464 which isn't too painful. Put a call to @code{sleep} in the code which
3465 the child process executes after the fork. It may be useful to sleep
3466 only if a certain environment variable is set, or a certain file exists,
3467 so that the delay need not occur when you don't want to run @value{GDBN}
3468 on the child. While the child is sleeping, use the @code{ps} program to
3469 get its process ID. Then tell @value{GDBN} (a new invocation of
3470 @value{GDBN} if you are also debugging the parent process) to attach to
3471 the child process (@pxref{Attach}). From that point on you can debug
3472 the child process just like any other process which you attached to.
3474 On some systems, @value{GDBN} provides support for debugging programs
3475 that create additional processes using the @code{fork} or @code{vfork}
3476 functions. On @sc{gnu}/Linux platforms, this feature is supported
3477 with kernel version 2.5.46 and later.
3479 The fork debugging commands are supported in native mode and when
3480 connected to @code{gdbserver} in either @code{target remote} mode or
3481 @code{target extended-remote} mode.
3483 By default, when a program forks, @value{GDBN} will continue to debug
3484 the parent process and the child process will run unimpeded.
3486 If you want to follow the child process instead of the parent process,
3487 use the command @w{@code{set follow-fork-mode}}.
3490 @kindex set follow-fork-mode
3491 @item set follow-fork-mode @var{mode}
3492 Set the debugger response to a program call of @code{fork} or
3493 @code{vfork}. A call to @code{fork} or @code{vfork} creates a new
3494 process. The @var{mode} argument can be:
3498 The original process is debugged after a fork. The child process runs
3499 unimpeded. This is the default.
3502 The new process is debugged after a fork. The parent process runs
3507 @kindex show follow-fork-mode
3508 @item show follow-fork-mode
3509 Display the current debugger response to a @code{fork} or @code{vfork} call.
3512 @cindex debugging multiple processes
3513 On Linux, if you want to debug both the parent and child processes, use the
3514 command @w{@code{set detach-on-fork}}.
3517 @kindex set detach-on-fork
3518 @item set detach-on-fork @var{mode}
3519 Tells gdb whether to detach one of the processes after a fork, or
3520 retain debugger control over them both.
3524 The child process (or parent process, depending on the value of
3525 @code{follow-fork-mode}) will be detached and allowed to run
3526 independently. This is the default.
3529 Both processes will be held under the control of @value{GDBN}.
3530 One process (child or parent, depending on the value of
3531 @code{follow-fork-mode}) is debugged as usual, while the other
3536 @kindex show detach-on-fork
3537 @item show detach-on-fork
3538 Show whether detach-on-fork mode is on/off.
3541 If you choose to set @samp{detach-on-fork} mode off, then @value{GDBN}
3542 will retain control of all forked processes (including nested forks).
3543 You can list the forked processes under the control of @value{GDBN} by
3544 using the @w{@code{info inferiors}} command, and switch from one fork
3545 to another by using the @code{inferior} command (@pxref{Inferiors and
3546 Programs, ,Debugging Multiple Inferiors and Programs}).
3548 To quit debugging one of the forked processes, you can either detach
3549 from it by using the @w{@code{detach inferiors}} command (allowing it
3550 to run independently), or kill it using the @w{@code{kill inferiors}}
3551 command. @xref{Inferiors and Programs, ,Debugging Multiple Inferiors
3554 If you ask to debug a child process and a @code{vfork} is followed by an
3555 @code{exec}, @value{GDBN} executes the new target up to the first
3556 breakpoint in the new target. If you have a breakpoint set on
3557 @code{main} in your original program, the breakpoint will also be set on
3558 the child process's @code{main}.
3560 On some systems, when a child process is spawned by @code{vfork}, you
3561 cannot debug the child or parent until an @code{exec} call completes.
3563 If you issue a @code{run} command to @value{GDBN} after an @code{exec}
3564 call executes, the new target restarts. To restart the parent
3565 process, use the @code{file} command with the parent executable name
3566 as its argument. By default, after an @code{exec} call executes,
3567 @value{GDBN} discards the symbols of the previous executable image.
3568 You can change this behaviour with the @w{@code{set follow-exec-mode}}
3572 @kindex set follow-exec-mode
3573 @item set follow-exec-mode @var{mode}
3575 Set debugger response to a program call of @code{exec}. An
3576 @code{exec} call replaces the program image of a process.
3578 @code{follow-exec-mode} can be:
3582 @value{GDBN} creates a new inferior and rebinds the process to this
3583 new inferior. The program the process was running before the
3584 @code{exec} call can be restarted afterwards by restarting the
3590 (@value{GDBP}) info inferiors
3592 Id Description Executable
3595 process 12020 is executing new program: prog2
3596 Program exited normally.
3597 (@value{GDBP}) info inferiors
3598 Id Description Executable
3604 @value{GDBN} keeps the process bound to the same inferior. The new
3605 executable image replaces the previous executable loaded in the
3606 inferior. Restarting the inferior after the @code{exec} call, with
3607 e.g., the @code{run} command, restarts the executable the process was
3608 running after the @code{exec} call. This is the default mode.
3613 (@value{GDBP}) info inferiors
3614 Id Description Executable
3617 process 12020 is executing new program: prog2
3618 Program exited normally.
3619 (@value{GDBP}) info inferiors
3620 Id Description Executable
3627 @code{follow-exec-mode} is supported in native mode and
3628 @code{target extended-remote} mode.
3630 You can use the @code{catch} command to make @value{GDBN} stop whenever
3631 a @code{fork}, @code{vfork}, or @code{exec} call is made. @xref{Set
3632 Catchpoints, ,Setting Catchpoints}.
3634 @node Checkpoint/Restart
3635 @section Setting a @emph{Bookmark} to Return to Later
3640 @cindex snapshot of a process
3641 @cindex rewind program state
3643 On certain operating systems@footnote{Currently, only
3644 @sc{gnu}/Linux.}, @value{GDBN} is able to save a @dfn{snapshot} of a
3645 program's state, called a @dfn{checkpoint}, and come back to it
3648 Returning to a checkpoint effectively undoes everything that has
3649 happened in the program since the @code{checkpoint} was saved. This
3650 includes changes in memory, registers, and even (within some limits)
3651 system state. Effectively, it is like going back in time to the
3652 moment when the checkpoint was saved.
3654 Thus, if you're stepping thru a program and you think you're
3655 getting close to the point where things go wrong, you can save
3656 a checkpoint. Then, if you accidentally go too far and miss
3657 the critical statement, instead of having to restart your program
3658 from the beginning, you can just go back to the checkpoint and
3659 start again from there.
3661 This can be especially useful if it takes a lot of time or
3662 steps to reach the point where you think the bug occurs.
3664 To use the @code{checkpoint}/@code{restart} method of debugging:
3669 Save a snapshot of the debugged program's current execution state.
3670 The @code{checkpoint} command takes no arguments, but each checkpoint
3671 is assigned a small integer id, similar to a breakpoint id.
3673 @kindex info checkpoints
3674 @item info checkpoints
3675 List the checkpoints that have been saved in the current debugging
3676 session. For each checkpoint, the following information will be
3683 @item Source line, or label
3686 @kindex restart @var{checkpoint-id}
3687 @item restart @var{checkpoint-id}
3688 Restore the program state that was saved as checkpoint number
3689 @var{checkpoint-id}. All program variables, registers, stack frames
3690 etc.@: will be returned to the values that they had when the checkpoint
3691 was saved. In essence, gdb will ``wind back the clock'' to the point
3692 in time when the checkpoint was saved.
3694 Note that breakpoints, @value{GDBN} variables, command history etc.
3695 are not affected by restoring a checkpoint. In general, a checkpoint
3696 only restores things that reside in the program being debugged, not in
3699 @kindex delete checkpoint @var{checkpoint-id}
3700 @item delete checkpoint @var{checkpoint-id}
3701 Delete the previously-saved checkpoint identified by @var{checkpoint-id}.
3705 Returning to a previously saved checkpoint will restore the user state
3706 of the program being debugged, plus a significant subset of the system
3707 (OS) state, including file pointers. It won't ``un-write'' data from
3708 a file, but it will rewind the file pointer to the previous location,
3709 so that the previously written data can be overwritten. For files
3710 opened in read mode, the pointer will also be restored so that the
3711 previously read data can be read again.
3713 Of course, characters that have been sent to a printer (or other
3714 external device) cannot be ``snatched back'', and characters received
3715 from eg.@: a serial device can be removed from internal program buffers,
3716 but they cannot be ``pushed back'' into the serial pipeline, ready to
3717 be received again. Similarly, the actual contents of files that have
3718 been changed cannot be restored (at this time).
3720 However, within those constraints, you actually can ``rewind'' your
3721 program to a previously saved point in time, and begin debugging it
3722 again --- and you can change the course of events so as to debug a
3723 different execution path this time.
3725 @cindex checkpoints and process id
3726 Finally, there is one bit of internal program state that will be
3727 different when you return to a checkpoint --- the program's process
3728 id. Each checkpoint will have a unique process id (or @var{pid}),
3729 and each will be different from the program's original @var{pid}.
3730 If your program has saved a local copy of its process id, this could
3731 potentially pose a problem.
3733 @subsection A Non-obvious Benefit of Using Checkpoints
3735 On some systems such as @sc{gnu}/Linux, address space randomization
3736 is performed on new processes for security reasons. This makes it
3737 difficult or impossible to set a breakpoint, or watchpoint, on an
3738 absolute address if you have to restart the program, since the
3739 absolute location of a symbol will change from one execution to the
3742 A checkpoint, however, is an @emph{identical} copy of a process.
3743 Therefore if you create a checkpoint at (eg.@:) the start of main,
3744 and simply return to that checkpoint instead of restarting the
3745 process, you can avoid the effects of address randomization and
3746 your symbols will all stay in the same place.
3749 @chapter Stopping and Continuing
3751 The principal purposes of using a debugger are so that you can stop your
3752 program before it terminates; or so that, if your program runs into
3753 trouble, you can investigate and find out why.
3755 Inside @value{GDBN}, your program may stop for any of several reasons,
3756 such as a signal, a breakpoint, or reaching a new line after a
3757 @value{GDBN} command such as @code{step}. You may then examine and
3758 change variables, set new breakpoints or remove old ones, and then
3759 continue execution. Usually, the messages shown by @value{GDBN} provide
3760 ample explanation of the status of your program---but you can also
3761 explicitly request this information at any time.
3764 @kindex info program
3766 Display information about the status of your program: whether it is
3767 running or not, what process it is, and why it stopped.
3771 * Breakpoints:: Breakpoints, watchpoints, and catchpoints
3772 * Continuing and Stepping:: Resuming execution
3773 * Skipping Over Functions and Files::
3774 Skipping over functions and files
3776 * Thread Stops:: Stopping and starting multi-thread programs
3780 @section Breakpoints, Watchpoints, and Catchpoints
3783 A @dfn{breakpoint} makes your program stop whenever a certain point in
3784 the program is reached. For each breakpoint, you can add conditions to
3785 control in finer detail whether your program stops. You can set
3786 breakpoints with the @code{break} command and its variants (@pxref{Set
3787 Breaks, ,Setting Breakpoints}), to specify the place where your program
3788 should stop by line number, function name or exact address in the
3791 On some systems, you can set breakpoints in shared libraries before
3792 the executable is run.
3795 @cindex data breakpoints
3796 @cindex memory tracing
3797 @cindex breakpoint on memory address
3798 @cindex breakpoint on variable modification
3799 A @dfn{watchpoint} is a special breakpoint that stops your program
3800 when the value of an expression changes. The expression may be a value
3801 of a variable, or it could involve values of one or more variables
3802 combined by operators, such as @samp{a + b}. This is sometimes called
3803 @dfn{data breakpoints}. You must use a different command to set
3804 watchpoints (@pxref{Set Watchpoints, ,Setting Watchpoints}), but aside
3805 from that, you can manage a watchpoint like any other breakpoint: you
3806 enable, disable, and delete both breakpoints and watchpoints using the
3809 You can arrange to have values from your program displayed automatically
3810 whenever @value{GDBN} stops at a breakpoint. @xref{Auto Display,,
3814 @cindex breakpoint on events
3815 A @dfn{catchpoint} is another special breakpoint that stops your program
3816 when a certain kind of event occurs, such as the throwing of a C@t{++}
3817 exception or the loading of a library. As with watchpoints, you use a
3818 different command to set a catchpoint (@pxref{Set Catchpoints, ,Setting
3819 Catchpoints}), but aside from that, you can manage a catchpoint like any
3820 other breakpoint. (To stop when your program receives a signal, use the
3821 @code{handle} command; see @ref{Signals, ,Signals}.)
3823 @cindex breakpoint numbers
3824 @cindex numbers for breakpoints
3825 @value{GDBN} assigns a number to each breakpoint, watchpoint, or
3826 catchpoint when you create it; these numbers are successive integers
3827 starting with one. In many of the commands for controlling various
3828 features of breakpoints you use the breakpoint number to say which
3829 breakpoint you want to change. Each breakpoint may be @dfn{enabled} or
3830 @dfn{disabled}; if disabled, it has no effect on your program until you
3833 @cindex breakpoint ranges
3834 @cindex breakpoint lists
3835 @cindex ranges of breakpoints
3836 @cindex lists of breakpoints
3837 Some @value{GDBN} commands accept a space-separated list of breakpoints
3838 on which to operate. A list element can be either a single breakpoint number,
3839 like @samp{5}, or a range of such numbers, like @samp{5-7}.
3840 When a breakpoint list is given to a command, all breakpoints in that list
3844 * Set Breaks:: Setting breakpoints
3845 * Set Watchpoints:: Setting watchpoints
3846 * Set Catchpoints:: Setting catchpoints
3847 * Delete Breaks:: Deleting breakpoints
3848 * Disabling:: Disabling breakpoints
3849 * Conditions:: Break conditions
3850 * Break Commands:: Breakpoint command lists
3851 * Dynamic Printf:: Dynamic printf
3852 * Save Breakpoints:: How to save breakpoints in a file
3853 * Static Probe Points:: Listing static probe points
3854 * Error in Breakpoints:: ``Cannot insert breakpoints''
3855 * Breakpoint-related Warnings:: ``Breakpoint address adjusted...''
3859 @subsection Setting Breakpoints
3861 @c FIXME LMB what does GDB do if no code on line of breakpt?
3862 @c consider in particular declaration with/without initialization.
3864 @c FIXME 2 is there stuff on this already? break at fun start, already init?
3867 @kindex b @r{(@code{break})}
3868 @vindex $bpnum@r{, convenience variable}
3869 @cindex latest breakpoint
3870 Breakpoints are set with the @code{break} command (abbreviated
3871 @code{b}). The debugger convenience variable @samp{$bpnum} records the
3872 number of the breakpoint you've set most recently; see @ref{Convenience
3873 Vars,, Convenience Variables}, for a discussion of what you can do with
3874 convenience variables.
3877 @item break @var{location}
3878 Set a breakpoint at the given @var{location}, which can specify a
3879 function name, a line number, or an address of an instruction.
3880 (@xref{Specify Location}, for a list of all the possible ways to
3881 specify a @var{location}.) The breakpoint will stop your program just
3882 before it executes any of the code in the specified @var{location}.
3884 When using source languages that permit overloading of symbols, such as
3885 C@t{++}, a function name may refer to more than one possible place to break.
3886 @xref{Ambiguous Expressions,,Ambiguous Expressions}, for a discussion of
3889 It is also possible to insert a breakpoint that will stop the program
3890 only if a specific thread (@pxref{Thread-Specific Breakpoints})
3891 or a specific task (@pxref{Ada Tasks}) hits that breakpoint.
3894 When called without any arguments, @code{break} sets a breakpoint at
3895 the next instruction to be executed in the selected stack frame
3896 (@pxref{Stack, ,Examining the Stack}). In any selected frame but the
3897 innermost, this makes your program stop as soon as control
3898 returns to that frame. This is similar to the effect of a
3899 @code{finish} command in the frame inside the selected frame---except
3900 that @code{finish} does not leave an active breakpoint. If you use
3901 @code{break} without an argument in the innermost frame, @value{GDBN} stops
3902 the next time it reaches the current location; this may be useful
3905 @value{GDBN} normally ignores breakpoints when it resumes execution, until at
3906 least one instruction has been executed. If it did not do this, you
3907 would be unable to proceed past a breakpoint without first disabling the
3908 breakpoint. This rule applies whether or not the breakpoint already
3909 existed when your program stopped.
3911 @item break @dots{} if @var{cond}
3912 Set a breakpoint with condition @var{cond}; evaluate the expression
3913 @var{cond} each time the breakpoint is reached, and stop only if the
3914 value is nonzero---that is, if @var{cond} evaluates as true.
3915 @samp{@dots{}} stands for one of the possible arguments described
3916 above (or no argument) specifying where to break. @xref{Conditions,
3917 ,Break Conditions}, for more information on breakpoint conditions.
3920 @item tbreak @var{args}
3921 Set a breakpoint enabled only for one stop. The @var{args} are the
3922 same as for the @code{break} command, and the breakpoint is set in the same
3923 way, but the breakpoint is automatically deleted after the first time your
3924 program stops there. @xref{Disabling, ,Disabling Breakpoints}.
3927 @cindex hardware breakpoints
3928 @item hbreak @var{args}
3929 Set a hardware-assisted breakpoint. The @var{args} are the same as for the
3930 @code{break} command and the breakpoint is set in the same way, but the
3931 breakpoint requires hardware support and some target hardware may not
3932 have this support. The main purpose of this is EPROM/ROM code
3933 debugging, so you can set a breakpoint at an instruction without
3934 changing the instruction. This can be used with the new trap-generation
3935 provided by SPARClite DSU and most x86-based targets. These targets
3936 will generate traps when a program accesses some data or instruction
3937 address that is assigned to the debug registers. However the hardware
3938 breakpoint registers can take a limited number of breakpoints. For
3939 example, on the DSU, only two data breakpoints can be set at a time, and
3940 @value{GDBN} will reject this command if more than two are used. Delete
3941 or disable unused hardware breakpoints before setting new ones
3942 (@pxref{Disabling, ,Disabling Breakpoints}).
3943 @xref{Conditions, ,Break Conditions}.
3944 For remote targets, you can restrict the number of hardware
3945 breakpoints @value{GDBN} will use, see @ref{set remote
3946 hardware-breakpoint-limit}.
3949 @item thbreak @var{args}
3950 Set a hardware-assisted breakpoint enabled only for one stop. The @var{args}
3951 are the same as for the @code{hbreak} command and the breakpoint is set in
3952 the same way. However, like the @code{tbreak} command,
3953 the breakpoint is automatically deleted after the
3954 first time your program stops there. Also, like the @code{hbreak}
3955 command, the breakpoint requires hardware support and some target hardware
3956 may not have this support. @xref{Disabling, ,Disabling Breakpoints}.
3957 See also @ref{Conditions, ,Break Conditions}.
3960 @cindex regular expression
3961 @cindex breakpoints at functions matching a regexp
3962 @cindex set breakpoints in many functions
3963 @item rbreak @var{regex}
3964 Set breakpoints on all functions matching the regular expression
3965 @var{regex}. This command sets an unconditional breakpoint on all
3966 matches, printing a list of all breakpoints it set. Once these
3967 breakpoints are set, they are treated just like the breakpoints set with
3968 the @code{break} command. You can delete them, disable them, or make
3969 them conditional the same way as any other breakpoint.
3971 In programs using different languages, @value{GDBN} chooses the syntax
3972 to print the list of all breakpoints it sets according to the
3973 @samp{set language} value: using @samp{set language auto}
3974 (see @ref{Automatically, ,Set Language Automatically}) means to use the
3975 language of the breakpoint's function, other values mean to use
3976 the manually specified language (see @ref{Manually, ,Set Language Manually}).
3978 The syntax of the regular expression is the standard one used with tools
3979 like @file{grep}. Note that this is different from the syntax used by
3980 shells, so for instance @code{foo*} matches all functions that include
3981 an @code{fo} followed by zero or more @code{o}s. There is an implicit
3982 @code{.*} leading and trailing the regular expression you supply, so to
3983 match only functions that begin with @code{foo}, use @code{^foo}.
3985 @cindex non-member C@t{++} functions, set breakpoint in
3986 When debugging C@t{++} programs, @code{rbreak} is useful for setting
3987 breakpoints on overloaded functions that are not members of any special
3990 @cindex set breakpoints on all functions
3991 The @code{rbreak} command can be used to set breakpoints in
3992 @strong{all} the functions in a program, like this:
3995 (@value{GDBP}) rbreak .
3998 @item rbreak @var{file}:@var{regex}
3999 If @code{rbreak} is called with a filename qualification, it limits
4000 the search for functions matching the given regular expression to the
4001 specified @var{file}. This can be used, for example, to set breakpoints on
4002 every function in a given file:
4005 (@value{GDBP}) rbreak file.c:.
4008 The colon separating the filename qualifier from the regex may
4009 optionally be surrounded by spaces.
4011 @kindex info breakpoints
4012 @cindex @code{$_} and @code{info breakpoints}
4013 @item info breakpoints @r{[}@var{list}@dots{}@r{]}
4014 @itemx info break @r{[}@var{list}@dots{}@r{]}
4015 Print a table of all breakpoints, watchpoints, and catchpoints set and
4016 not deleted. Optional argument @var{n} means print information only
4017 about the specified breakpoint(s) (or watchpoint(s) or catchpoint(s)).
4018 For each breakpoint, following columns are printed:
4021 @item Breakpoint Numbers
4023 Breakpoint, watchpoint, or catchpoint.
4025 Whether the breakpoint is marked to be disabled or deleted when hit.
4026 @item Enabled or Disabled
4027 Enabled breakpoints are marked with @samp{y}. @samp{n} marks breakpoints
4028 that are not enabled.
4030 Where the breakpoint is in your program, as a memory address. For a
4031 pending breakpoint whose address is not yet known, this field will
4032 contain @samp{<PENDING>}. Such breakpoint won't fire until a shared
4033 library that has the symbol or line referred by breakpoint is loaded.
4034 See below for details. A breakpoint with several locations will
4035 have @samp{<MULTIPLE>} in this field---see below for details.
4037 Where the breakpoint is in the source for your program, as a file and
4038 line number. For a pending breakpoint, the original string passed to
4039 the breakpoint command will be listed as it cannot be resolved until
4040 the appropriate shared library is loaded in the future.
4044 If a breakpoint is conditional, there are two evaluation modes: ``host'' and
4045 ``target''. If mode is ``host'', breakpoint condition evaluation is done by
4046 @value{GDBN} on the host's side. If it is ``target'', then the condition
4047 is evaluated by the target. The @code{info break} command shows
4048 the condition on the line following the affected breakpoint, together with
4049 its condition evaluation mode in between parentheses.
4051 Breakpoint commands, if any, are listed after that. A pending breakpoint is
4052 allowed to have a condition specified for it. The condition is not parsed for
4053 validity until a shared library is loaded that allows the pending
4054 breakpoint to resolve to a valid location.
4057 @code{info break} with a breakpoint
4058 number @var{n} as argument lists only that breakpoint. The
4059 convenience variable @code{$_} and the default examining-address for
4060 the @code{x} command are set to the address of the last breakpoint
4061 listed (@pxref{Memory, ,Examining Memory}).
4064 @code{info break} displays a count of the number of times the breakpoint
4065 has been hit. This is especially useful in conjunction with the
4066 @code{ignore} command. You can ignore a large number of breakpoint
4067 hits, look at the breakpoint info to see how many times the breakpoint
4068 was hit, and then run again, ignoring one less than that number. This
4069 will get you quickly to the last hit of that breakpoint.
4072 For a breakpoints with an enable count (xref) greater than 1,
4073 @code{info break} also displays that count.
4077 @value{GDBN} allows you to set any number of breakpoints at the same place in
4078 your program. There is nothing silly or meaningless about this. When
4079 the breakpoints are conditional, this is even useful
4080 (@pxref{Conditions, ,Break Conditions}).
4082 @cindex multiple locations, breakpoints
4083 @cindex breakpoints, multiple locations
4084 It is possible that a breakpoint corresponds to several locations
4085 in your program. Examples of this situation are:
4089 Multiple functions in the program may have the same name.
4092 For a C@t{++} constructor, the @value{NGCC} compiler generates several
4093 instances of the function body, used in different cases.
4096 For a C@t{++} template function, a given line in the function can
4097 correspond to any number of instantiations.
4100 For an inlined function, a given source line can correspond to
4101 several places where that function is inlined.
4104 In all those cases, @value{GDBN} will insert a breakpoint at all
4105 the relevant locations.
4107 A breakpoint with multiple locations is displayed in the breakpoint
4108 table using several rows---one header row, followed by one row for
4109 each breakpoint location. The header row has @samp{<MULTIPLE>} in the
4110 address column. The rows for individual locations contain the actual
4111 addresses for locations, and show the functions to which those
4112 locations belong. The number column for a location is of the form
4113 @var{breakpoint-number}.@var{location-number}.
4118 Num Type Disp Enb Address What
4119 1 breakpoint keep y <MULTIPLE>
4121 breakpoint already hit 1 time
4122 1.1 y 0x080486a2 in void foo<int>() at t.cc:8
4123 1.2 y 0x080486ca in void foo<double>() at t.cc:8
4126 You cannot delete the individual locations from a breakpoint. However,
4127 each location can be individually enabled or disabled by passing
4128 @var{breakpoint-number}.@var{location-number} as argument to the
4129 @code{enable} and @code{disable} commands. It's also possible to
4130 @code{enable} and @code{disable} a range of @var{location-number}
4131 locations using a @var{breakpoint-number} and two @var{location-number}s,
4132 in increasing order, separated by a hyphen, like
4133 @kbd{@var{breakpoint-number}.@var{location-number1}-@var{location-number2}},
4134 in which case @value{GDBN} acts on all the locations in the range (inclusive).
4135 Disabling or enabling the parent breakpoint (@pxref{Disabling}) affects
4136 all of the locations that belong to that breakpoint.
4138 @cindex pending breakpoints
4139 It's quite common to have a breakpoint inside a shared library.
4140 Shared libraries can be loaded and unloaded explicitly,
4141 and possibly repeatedly, as the program is executed. To support
4142 this use case, @value{GDBN} updates breakpoint locations whenever
4143 any shared library is loaded or unloaded. Typically, you would
4144 set a breakpoint in a shared library at the beginning of your
4145 debugging session, when the library is not loaded, and when the
4146 symbols from the library are not available. When you try to set
4147 breakpoint, @value{GDBN} will ask you if you want to set
4148 a so called @dfn{pending breakpoint}---breakpoint whose address
4149 is not yet resolved.
4151 After the program is run, whenever a new shared library is loaded,
4152 @value{GDBN} reevaluates all the breakpoints. When a newly loaded
4153 shared library contains the symbol or line referred to by some
4154 pending breakpoint, that breakpoint is resolved and becomes an
4155 ordinary breakpoint. When a library is unloaded, all breakpoints
4156 that refer to its symbols or source lines become pending again.
4158 This logic works for breakpoints with multiple locations, too. For
4159 example, if you have a breakpoint in a C@t{++} template function, and
4160 a newly loaded shared library has an instantiation of that template,
4161 a new location is added to the list of locations for the breakpoint.
4163 Except for having unresolved address, pending breakpoints do not
4164 differ from regular breakpoints. You can set conditions or commands,
4165 enable and disable them and perform other breakpoint operations.
4167 @value{GDBN} provides some additional commands for controlling what
4168 happens when the @samp{break} command cannot resolve breakpoint
4169 address specification to an address:
4171 @kindex set breakpoint pending
4172 @kindex show breakpoint pending
4174 @item set breakpoint pending auto
4175 This is the default behavior. When @value{GDBN} cannot find the breakpoint
4176 location, it queries you whether a pending breakpoint should be created.
4178 @item set breakpoint pending on
4179 This indicates that an unrecognized breakpoint location should automatically
4180 result in a pending breakpoint being created.
4182 @item set breakpoint pending off
4183 This indicates that pending breakpoints are not to be created. Any
4184 unrecognized breakpoint location results in an error. This setting does
4185 not affect any pending breakpoints previously created.
4187 @item show breakpoint pending
4188 Show the current behavior setting for creating pending breakpoints.
4191 The settings above only affect the @code{break} command and its
4192 variants. Once breakpoint is set, it will be automatically updated
4193 as shared libraries are loaded and unloaded.
4195 @cindex automatic hardware breakpoints
4196 For some targets, @value{GDBN} can automatically decide if hardware or
4197 software breakpoints should be used, depending on whether the
4198 breakpoint address is read-only or read-write. This applies to
4199 breakpoints set with the @code{break} command as well as to internal
4200 breakpoints set by commands like @code{next} and @code{finish}. For
4201 breakpoints set with @code{hbreak}, @value{GDBN} will always use hardware
4204 You can control this automatic behaviour with the following commands:
4206 @kindex set breakpoint auto-hw
4207 @kindex show breakpoint auto-hw
4209 @item set breakpoint auto-hw on
4210 This is the default behavior. When @value{GDBN} sets a breakpoint, it
4211 will try to use the target memory map to decide if software or hardware
4212 breakpoint must be used.
4214 @item set breakpoint auto-hw off
4215 This indicates @value{GDBN} should not automatically select breakpoint
4216 type. If the target provides a memory map, @value{GDBN} will warn when
4217 trying to set software breakpoint at a read-only address.
4220 @value{GDBN} normally implements breakpoints by replacing the program code
4221 at the breakpoint address with a special instruction, which, when
4222 executed, given control to the debugger. By default, the program
4223 code is so modified only when the program is resumed. As soon as
4224 the program stops, @value{GDBN} restores the original instructions. This
4225 behaviour guards against leaving breakpoints inserted in the
4226 target should gdb abrubptly disconnect. However, with slow remote
4227 targets, inserting and removing breakpoint can reduce the performance.
4228 This behavior can be controlled with the following commands::
4230 @kindex set breakpoint always-inserted
4231 @kindex show breakpoint always-inserted
4233 @item set breakpoint always-inserted off
4234 All breakpoints, including newly added by the user, are inserted in
4235 the target only when the target is resumed. All breakpoints are
4236 removed from the target when it stops. This is the default mode.
4238 @item set breakpoint always-inserted on
4239 Causes all breakpoints to be inserted in the target at all times. If
4240 the user adds a new breakpoint, or changes an existing breakpoint, the
4241 breakpoints in the target are updated immediately. A breakpoint is
4242 removed from the target only when breakpoint itself is deleted.
4245 @value{GDBN} handles conditional breakpoints by evaluating these conditions
4246 when a breakpoint breaks. If the condition is true, then the process being
4247 debugged stops, otherwise the process is resumed.
4249 If the target supports evaluating conditions on its end, @value{GDBN} may
4250 download the breakpoint, together with its conditions, to it.
4252 This feature can be controlled via the following commands:
4254 @kindex set breakpoint condition-evaluation
4255 @kindex show breakpoint condition-evaluation
4257 @item set breakpoint condition-evaluation host
4258 This option commands @value{GDBN} to evaluate the breakpoint
4259 conditions on the host's side. Unconditional breakpoints are sent to
4260 the target which in turn receives the triggers and reports them back to GDB
4261 for condition evaluation. This is the standard evaluation mode.
4263 @item set breakpoint condition-evaluation target
4264 This option commands @value{GDBN} to download breakpoint conditions
4265 to the target at the moment of their insertion. The target
4266 is responsible for evaluating the conditional expression and reporting
4267 breakpoint stop events back to @value{GDBN} whenever the condition
4268 is true. Due to limitations of target-side evaluation, some conditions
4269 cannot be evaluated there, e.g., conditions that depend on local data
4270 that is only known to the host. Examples include
4271 conditional expressions involving convenience variables, complex types
4272 that cannot be handled by the agent expression parser and expressions
4273 that are too long to be sent over to the target, specially when the
4274 target is a remote system. In these cases, the conditions will be
4275 evaluated by @value{GDBN}.
4277 @item set breakpoint condition-evaluation auto
4278 This is the default mode. If the target supports evaluating breakpoint
4279 conditions on its end, @value{GDBN} will download breakpoint conditions to
4280 the target (limitations mentioned previously apply). If the target does
4281 not support breakpoint condition evaluation, then @value{GDBN} will fallback
4282 to evaluating all these conditions on the host's side.
4286 @cindex negative breakpoint numbers
4287 @cindex internal @value{GDBN} breakpoints
4288 @value{GDBN} itself sometimes sets breakpoints in your program for
4289 special purposes, such as proper handling of @code{longjmp} (in C
4290 programs). These internal breakpoints are assigned negative numbers,
4291 starting with @code{-1}; @samp{info breakpoints} does not display them.
4292 You can see these breakpoints with the @value{GDBN} maintenance command
4293 @samp{maint info breakpoints} (@pxref{maint info breakpoints}).
4296 @node Set Watchpoints
4297 @subsection Setting Watchpoints
4299 @cindex setting watchpoints
4300 You can use a watchpoint to stop execution whenever the value of an
4301 expression changes, without having to predict a particular place where
4302 this may happen. (This is sometimes called a @dfn{data breakpoint}.)
4303 The expression may be as simple as the value of a single variable, or
4304 as complex as many variables combined by operators. Examples include:
4308 A reference to the value of a single variable.
4311 An address cast to an appropriate data type. For example,
4312 @samp{*(int *)0x12345678} will watch a 4-byte region at the specified
4313 address (assuming an @code{int} occupies 4 bytes).
4316 An arbitrarily complex expression, such as @samp{a*b + c/d}. The
4317 expression can use any operators valid in the program's native
4318 language (@pxref{Languages}).
4321 You can set a watchpoint on an expression even if the expression can
4322 not be evaluated yet. For instance, you can set a watchpoint on
4323 @samp{*global_ptr} before @samp{global_ptr} is initialized.
4324 @value{GDBN} will stop when your program sets @samp{global_ptr} and
4325 the expression produces a valid value. If the expression becomes
4326 valid in some other way than changing a variable (e.g.@: if the memory
4327 pointed to by @samp{*global_ptr} becomes readable as the result of a
4328 @code{malloc} call), @value{GDBN} may not stop until the next time
4329 the expression changes.
4331 @cindex software watchpoints
4332 @cindex hardware watchpoints
4333 Depending on your system, watchpoints may be implemented in software or
4334 hardware. @value{GDBN} does software watchpointing by single-stepping your
4335 program and testing the variable's value each time, which is hundreds of
4336 times slower than normal execution. (But this may still be worth it, to
4337 catch errors where you have no clue what part of your program is the
4340 On some systems, such as most PowerPC or x86-based targets,
4341 @value{GDBN} includes support for hardware watchpoints, which do not
4342 slow down the running of your program.
4346 @item watch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{thread-id}@r{]} @r{[}mask @var{maskvalue}@r{]}
4347 Set a watchpoint for an expression. @value{GDBN} will break when the
4348 expression @var{expr} is written into by the program and its value
4349 changes. The simplest (and the most popular) use of this command is
4350 to watch the value of a single variable:
4353 (@value{GDBP}) watch foo
4356 If the command includes a @code{@r{[}thread @var{thread-id}@r{]}}
4357 argument, @value{GDBN} breaks only when the thread identified by
4358 @var{thread-id} changes the value of @var{expr}. If any other threads
4359 change the value of @var{expr}, @value{GDBN} will not break. Note
4360 that watchpoints restricted to a single thread in this way only work
4361 with Hardware Watchpoints.
4363 Ordinarily a watchpoint respects the scope of variables in @var{expr}
4364 (see below). The @code{-location} argument tells @value{GDBN} to
4365 instead watch the memory referred to by @var{expr}. In this case,
4366 @value{GDBN} will evaluate @var{expr}, take the address of the result,
4367 and watch the memory at that address. The type of the result is used
4368 to determine the size of the watched memory. If the expression's
4369 result does not have an address, then @value{GDBN} will print an
4372 The @code{@r{[}mask @var{maskvalue}@r{]}} argument allows creation
4373 of masked watchpoints, if the current architecture supports this
4374 feature (e.g., PowerPC Embedded architecture, see @ref{PowerPC
4375 Embedded}.) A @dfn{masked watchpoint} specifies a mask in addition
4376 to an address to watch. The mask specifies that some bits of an address
4377 (the bits which are reset in the mask) should be ignored when matching
4378 the address accessed by the inferior against the watchpoint address.
4379 Thus, a masked watchpoint watches many addresses simultaneously---those
4380 addresses whose unmasked bits are identical to the unmasked bits in the
4381 watchpoint address. The @code{mask} argument implies @code{-location}.
4385 (@value{GDBP}) watch foo mask 0xffff00ff
4386 (@value{GDBP}) watch *0xdeadbeef mask 0xffffff00
4390 @item rwatch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{thread-id}@r{]} @r{[}mask @var{maskvalue}@r{]}
4391 Set a watchpoint that will break when the value of @var{expr} is read
4395 @item awatch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{thread-id}@r{]} @r{[}mask @var{maskvalue}@r{]}
4396 Set a watchpoint that will break when @var{expr} is either read from
4397 or written into by the program.
4399 @kindex info watchpoints @r{[}@var{list}@dots{}@r{]}
4400 @item info watchpoints @r{[}@var{list}@dots{}@r{]}
4401 This command prints a list of watchpoints, using the same format as
4402 @code{info break} (@pxref{Set Breaks}).
4405 If you watch for a change in a numerically entered address you need to
4406 dereference it, as the address itself is just a constant number which will
4407 never change. @value{GDBN} refuses to create a watchpoint that watches
4408 a never-changing value:
4411 (@value{GDBP}) watch 0x600850
4412 Cannot watch constant value 0x600850.
4413 (@value{GDBP}) watch *(int *) 0x600850
4414 Watchpoint 1: *(int *) 6293584
4417 @value{GDBN} sets a @dfn{hardware watchpoint} if possible. Hardware
4418 watchpoints execute very quickly, and the debugger reports a change in
4419 value at the exact instruction where the change occurs. If @value{GDBN}
4420 cannot set a hardware watchpoint, it sets a software watchpoint, which
4421 executes more slowly and reports the change in value at the next
4422 @emph{statement}, not the instruction, after the change occurs.
4424 @cindex use only software watchpoints
4425 You can force @value{GDBN} to use only software watchpoints with the
4426 @kbd{set can-use-hw-watchpoints 0} command. With this variable set to
4427 zero, @value{GDBN} will never try to use hardware watchpoints, even if
4428 the underlying system supports them. (Note that hardware-assisted
4429 watchpoints that were set @emph{before} setting
4430 @code{can-use-hw-watchpoints} to zero will still use the hardware
4431 mechanism of watching expression values.)
4434 @item set can-use-hw-watchpoints
4435 @kindex set can-use-hw-watchpoints
4436 Set whether or not to use hardware watchpoints.
4438 @item show can-use-hw-watchpoints
4439 @kindex show can-use-hw-watchpoints
4440 Show the current mode of using hardware watchpoints.
4443 For remote targets, you can restrict the number of hardware
4444 watchpoints @value{GDBN} will use, see @ref{set remote
4445 hardware-breakpoint-limit}.
4447 When you issue the @code{watch} command, @value{GDBN} reports
4450 Hardware watchpoint @var{num}: @var{expr}
4454 if it was able to set a hardware watchpoint.
4456 Currently, the @code{awatch} and @code{rwatch} commands can only set
4457 hardware watchpoints, because accesses to data that don't change the
4458 value of the watched expression cannot be detected without examining
4459 every instruction as it is being executed, and @value{GDBN} does not do
4460 that currently. If @value{GDBN} finds that it is unable to set a
4461 hardware breakpoint with the @code{awatch} or @code{rwatch} command, it
4462 will print a message like this:
4465 Expression cannot be implemented with read/access watchpoint.
4468 Sometimes, @value{GDBN} cannot set a hardware watchpoint because the
4469 data type of the watched expression is wider than what a hardware
4470 watchpoint on the target machine can handle. For example, some systems
4471 can only watch regions that are up to 4 bytes wide; on such systems you
4472 cannot set hardware watchpoints for an expression that yields a
4473 double-precision floating-point number (which is typically 8 bytes
4474 wide). As a work-around, it might be possible to break the large region
4475 into a series of smaller ones and watch them with separate watchpoints.
4477 If you set too many hardware watchpoints, @value{GDBN} might be unable
4478 to insert all of them when you resume the execution of your program.
4479 Since the precise number of active watchpoints is unknown until such
4480 time as the program is about to be resumed, @value{GDBN} might not be
4481 able to warn you about this when you set the watchpoints, and the
4482 warning will be printed only when the program is resumed:
4485 Hardware watchpoint @var{num}: Could not insert watchpoint
4489 If this happens, delete or disable some of the watchpoints.
4491 Watching complex expressions that reference many variables can also
4492 exhaust the resources available for hardware-assisted watchpoints.
4493 That's because @value{GDBN} needs to watch every variable in the
4494 expression with separately allocated resources.
4496 If you call a function interactively using @code{print} or @code{call},
4497 any watchpoints you have set will be inactive until @value{GDBN} reaches another
4498 kind of breakpoint or the call completes.
4500 @value{GDBN} automatically deletes watchpoints that watch local
4501 (automatic) variables, or expressions that involve such variables, when
4502 they go out of scope, that is, when the execution leaves the block in
4503 which these variables were defined. In particular, when the program
4504 being debugged terminates, @emph{all} local variables go out of scope,
4505 and so only watchpoints that watch global variables remain set. If you
4506 rerun the program, you will need to set all such watchpoints again. One
4507 way of doing that would be to set a code breakpoint at the entry to the
4508 @code{main} function and when it breaks, set all the watchpoints.
4510 @cindex watchpoints and threads
4511 @cindex threads and watchpoints
4512 In multi-threaded programs, watchpoints will detect changes to the
4513 watched expression from every thread.
4516 @emph{Warning:} In multi-threaded programs, software watchpoints
4517 have only limited usefulness. If @value{GDBN} creates a software
4518 watchpoint, it can only watch the value of an expression @emph{in a
4519 single thread}. If you are confident that the expression can only
4520 change due to the current thread's activity (and if you are also
4521 confident that no other thread can become current), then you can use
4522 software watchpoints as usual. However, @value{GDBN} may not notice
4523 when a non-current thread's activity changes the expression. (Hardware
4524 watchpoints, in contrast, watch an expression in all threads.)
4527 @xref{set remote hardware-watchpoint-limit}.
4529 @node Set Catchpoints
4530 @subsection Setting Catchpoints
4531 @cindex catchpoints, setting
4532 @cindex exception handlers
4533 @cindex event handling
4535 You can use @dfn{catchpoints} to cause the debugger to stop for certain
4536 kinds of program events, such as C@t{++} exceptions or the loading of a
4537 shared library. Use the @code{catch} command to set a catchpoint.
4541 @item catch @var{event}
4542 Stop when @var{event} occurs. The @var{event} can be any of the following:
4545 @item throw @r{[}@var{regexp}@r{]}
4546 @itemx rethrow @r{[}@var{regexp}@r{]}
4547 @itemx catch @r{[}@var{regexp}@r{]}
4549 @kindex catch rethrow
4551 @cindex stop on C@t{++} exceptions
4552 The throwing, re-throwing, or catching of a C@t{++} exception.
4554 If @var{regexp} is given, then only exceptions whose type matches the
4555 regular expression will be caught.
4557 @vindex $_exception@r{, convenience variable}
4558 The convenience variable @code{$_exception} is available at an
4559 exception-related catchpoint, on some systems. This holds the
4560 exception being thrown.
4562 There are currently some limitations to C@t{++} exception handling in
4567 The support for these commands is system-dependent. Currently, only
4568 systems using the @samp{gnu-v3} C@t{++} ABI (@pxref{ABI}) are
4572 The regular expression feature and the @code{$_exception} convenience
4573 variable rely on the presence of some SDT probes in @code{libstdc++}.
4574 If these probes are not present, then these features cannot be used.
4575 These probes were first available in the GCC 4.8 release, but whether
4576 or not they are available in your GCC also depends on how it was
4580 The @code{$_exception} convenience variable is only valid at the
4581 instruction at which an exception-related catchpoint is set.
4584 When an exception-related catchpoint is hit, @value{GDBN} stops at a
4585 location in the system library which implements runtime exception
4586 support for C@t{++}, usually @code{libstdc++}. You can use @code{up}
4587 (@pxref{Selection}) to get to your code.
4590 If you call a function interactively, @value{GDBN} normally returns
4591 control to you when the function has finished executing. If the call
4592 raises an exception, however, the call may bypass the mechanism that
4593 returns control to you and cause your program either to abort or to
4594 simply continue running until it hits a breakpoint, catches a signal
4595 that @value{GDBN} is listening for, or exits. This is the case even if
4596 you set a catchpoint for the exception; catchpoints on exceptions are
4597 disabled within interactive calls. @xref{Calling}, for information on
4598 controlling this with @code{set unwind-on-terminating-exception}.
4601 You cannot raise an exception interactively.
4604 You cannot install an exception handler interactively.
4607 @item exception @r{[}@var{name}@r{]}
4608 @kindex catch exception
4609 @cindex Ada exception catching
4610 @cindex catch Ada exceptions
4611 An Ada exception being raised. If an exception name is specified
4612 at the end of the command (eg @code{catch exception Program_Error}),
4613 the debugger will stop only when this specific exception is raised.
4614 Otherwise, the debugger stops execution when any Ada exception is raised.
4616 When inserting an exception catchpoint on a user-defined exception whose
4617 name is identical to one of the exceptions defined by the language, the
4618 fully qualified name must be used as the exception name. Otherwise,
4619 @value{GDBN} will assume that it should stop on the pre-defined exception
4620 rather than the user-defined one. For instance, assuming an exception
4621 called @code{Constraint_Error} is defined in package @code{Pck}, then
4622 the command to use to catch such exceptions is @kbd{catch exception
4623 Pck.Constraint_Error}.
4625 @item exception unhandled
4626 @kindex catch exception unhandled
4627 An exception that was raised but is not handled by the program.
4629 @item handlers @r{[}@var{name}@r{]}
4630 @kindex catch handlers
4631 @cindex Ada exception handlers catching
4632 @cindex catch Ada exceptions when handled
4633 An Ada exception being handled. If an exception name is
4634 specified at the end of the command
4635 (eg @kbd{catch handlers Program_Error}), the debugger will stop
4636 only when this specific exception is handled.
4637 Otherwise, the debugger stops execution when any Ada exception is handled.
4639 When inserting a handlers catchpoint on a user-defined
4640 exception whose name is identical to one of the exceptions
4641 defined by the language, the fully qualified name must be used
4642 as the exception name. Otherwise, @value{GDBN} will assume that it
4643 should stop on the pre-defined exception rather than the
4644 user-defined one. For instance, assuming an exception called
4645 @code{Constraint_Error} is defined in package @code{Pck}, then the
4646 command to use to catch such exceptions handling is
4647 @kbd{catch handlers Pck.Constraint_Error}.
4650 @kindex catch assert
4651 A failed Ada assertion.
4655 @cindex break on fork/exec
4656 A call to @code{exec}.
4658 @anchor{catch syscall}
4660 @itemx syscall @r{[}@var{name} @r{|} @var{number} @r{|} @r{group:}@var{groupname} @r{|} @r{g:}@var{groupname}@r{]} @dots{}
4661 @kindex catch syscall
4662 @cindex break on a system call.
4663 A call to or return from a system call, a.k.a.@: @dfn{syscall}. A
4664 syscall is a mechanism for application programs to request a service
4665 from the operating system (OS) or one of the OS system services.
4666 @value{GDBN} can catch some or all of the syscalls issued by the
4667 debuggee, and show the related information for each syscall. If no
4668 argument is specified, calls to and returns from all system calls
4671 @var{name} can be any system call name that is valid for the
4672 underlying OS. Just what syscalls are valid depends on the OS. On
4673 GNU and Unix systems, you can find the full list of valid syscall
4674 names on @file{/usr/include/asm/unistd.h}.
4676 @c For MS-Windows, the syscall names and the corresponding numbers
4677 @c can be found, e.g., on this URL:
4678 @c http://www.metasploit.com/users/opcode/syscalls.html
4679 @c but we don't support Windows syscalls yet.
4681 Normally, @value{GDBN} knows in advance which syscalls are valid for
4682 each OS, so you can use the @value{GDBN} command-line completion
4683 facilities (@pxref{Completion,, command completion}) to list the
4686 You may also specify the system call numerically. A syscall's
4687 number is the value passed to the OS's syscall dispatcher to
4688 identify the requested service. When you specify the syscall by its
4689 name, @value{GDBN} uses its database of syscalls to convert the name
4690 into the corresponding numeric code, but using the number directly
4691 may be useful if @value{GDBN}'s database does not have the complete
4692 list of syscalls on your system (e.g., because @value{GDBN} lags
4693 behind the OS upgrades).
4695 You may specify a group of related syscalls to be caught at once using
4696 the @code{group:} syntax (@code{g:} is a shorter equivalent). For
4697 instance, on some platforms @value{GDBN} allows you to catch all
4698 network related syscalls, by passing the argument @code{group:network}
4699 to @code{catch syscall}. Note that not all syscall groups are
4700 available in every system. You can use the command completion
4701 facilities (@pxref{Completion,, command completion}) to list the
4702 syscall groups available on your environment.
4704 The example below illustrates how this command works if you don't provide
4708 (@value{GDBP}) catch syscall
4709 Catchpoint 1 (syscall)
4711 Starting program: /tmp/catch-syscall
4713 Catchpoint 1 (call to syscall 'close'), \
4714 0xffffe424 in __kernel_vsyscall ()
4718 Catchpoint 1 (returned from syscall 'close'), \
4719 0xffffe424 in __kernel_vsyscall ()
4723 Here is an example of catching a system call by name:
4726 (@value{GDBP}) catch syscall chroot
4727 Catchpoint 1 (syscall 'chroot' [61])
4729 Starting program: /tmp/catch-syscall
4731 Catchpoint 1 (call to syscall 'chroot'), \
4732 0xffffe424 in __kernel_vsyscall ()
4736 Catchpoint 1 (returned from syscall 'chroot'), \
4737 0xffffe424 in __kernel_vsyscall ()
4741 An example of specifying a system call numerically. In the case
4742 below, the syscall number has a corresponding entry in the XML
4743 file, so @value{GDBN} finds its name and prints it:
4746 (@value{GDBP}) catch syscall 252
4747 Catchpoint 1 (syscall(s) 'exit_group')
4749 Starting program: /tmp/catch-syscall
4751 Catchpoint 1 (call to syscall 'exit_group'), \
4752 0xffffe424 in __kernel_vsyscall ()
4756 Program exited normally.
4760 Here is an example of catching a syscall group:
4763 (@value{GDBP}) catch syscall group:process
4764 Catchpoint 1 (syscalls 'exit' [1] 'fork' [2] 'waitpid' [7]
4765 'execve' [11] 'wait4' [114] 'clone' [120] 'vfork' [190]
4766 'exit_group' [252] 'waitid' [284] 'unshare' [310])
4768 Starting program: /tmp/catch-syscall
4770 Catchpoint 1 (call to syscall fork), 0x00007ffff7df4e27 in open64 ()
4771 from /lib64/ld-linux-x86-64.so.2
4777 However, there can be situations when there is no corresponding name
4778 in XML file for that syscall number. In this case, @value{GDBN} prints
4779 a warning message saying that it was not able to find the syscall name,
4780 but the catchpoint will be set anyway. See the example below:
4783 (@value{GDBP}) catch syscall 764
4784 warning: The number '764' does not represent a known syscall.
4785 Catchpoint 2 (syscall 764)
4789 If you configure @value{GDBN} using the @samp{--without-expat} option,
4790 it will not be able to display syscall names. Also, if your
4791 architecture does not have an XML file describing its system calls,
4792 you will not be able to see the syscall names. It is important to
4793 notice that these two features are used for accessing the syscall
4794 name database. In either case, you will see a warning like this:
4797 (@value{GDBP}) catch syscall
4798 warning: Could not open "syscalls/i386-linux.xml"
4799 warning: Could not load the syscall XML file 'syscalls/i386-linux.xml'.
4800 GDB will not be able to display syscall names.
4801 Catchpoint 1 (syscall)
4805 Of course, the file name will change depending on your architecture and system.
4807 Still using the example above, you can also try to catch a syscall by its
4808 number. In this case, you would see something like:
4811 (@value{GDBP}) catch syscall 252
4812 Catchpoint 1 (syscall(s) 252)
4815 Again, in this case @value{GDBN} would not be able to display syscall's names.
4819 A call to @code{fork}.
4823 A call to @code{vfork}.
4825 @item load @r{[}@var{regexp}@r{]}
4826 @itemx unload @r{[}@var{regexp}@r{]}
4828 @kindex catch unload
4829 The loading or unloading of a shared library. If @var{regexp} is
4830 given, then the catchpoint will stop only if the regular expression
4831 matches one of the affected libraries.
4833 @item signal @r{[}@var{signal}@dots{} @r{|} @samp{all}@r{]}
4834 @kindex catch signal
4835 The delivery of a signal.
4837 With no arguments, this catchpoint will catch any signal that is not
4838 used internally by @value{GDBN}, specifically, all signals except
4839 @samp{SIGTRAP} and @samp{SIGINT}.
4841 With the argument @samp{all}, all signals, including those used by
4842 @value{GDBN}, will be caught. This argument cannot be used with other
4845 Otherwise, the arguments are a list of signal names as given to
4846 @code{handle} (@pxref{Signals}). Only signals specified in this list
4849 One reason that @code{catch signal} can be more useful than
4850 @code{handle} is that you can attach commands and conditions to the
4853 When a signal is caught by a catchpoint, the signal's @code{stop} and
4854 @code{print} settings, as specified by @code{handle}, are ignored.
4855 However, whether the signal is still delivered to the inferior depends
4856 on the @code{pass} setting; this can be changed in the catchpoint's
4861 @item tcatch @var{event}
4863 Set a catchpoint that is enabled only for one stop. The catchpoint is
4864 automatically deleted after the first time the event is caught.
4868 Use the @code{info break} command to list the current catchpoints.
4872 @subsection Deleting Breakpoints
4874 @cindex clearing breakpoints, watchpoints, catchpoints
4875 @cindex deleting breakpoints, watchpoints, catchpoints
4876 It is often necessary to eliminate a breakpoint, watchpoint, or
4877 catchpoint once it has done its job and you no longer want your program
4878 to stop there. This is called @dfn{deleting} the breakpoint. A
4879 breakpoint that has been deleted no longer exists; it is forgotten.
4881 With the @code{clear} command you can delete breakpoints according to
4882 where they are in your program. With the @code{delete} command you can
4883 delete individual breakpoints, watchpoints, or catchpoints by specifying
4884 their breakpoint numbers.
4886 It is not necessary to delete a breakpoint to proceed past it. @value{GDBN}
4887 automatically ignores breakpoints on the first instruction to be executed
4888 when you continue execution without changing the execution address.
4893 Delete any breakpoints at the next instruction to be executed in the
4894 selected stack frame (@pxref{Selection, ,Selecting a Frame}). When
4895 the innermost frame is selected, this is a good way to delete a
4896 breakpoint where your program just stopped.
4898 @item clear @var{location}
4899 Delete any breakpoints set at the specified @var{location}.
4900 @xref{Specify Location}, for the various forms of @var{location}; the
4901 most useful ones are listed below:
4904 @item clear @var{function}
4905 @itemx clear @var{filename}:@var{function}
4906 Delete any breakpoints set at entry to the named @var{function}.
4908 @item clear @var{linenum}
4909 @itemx clear @var{filename}:@var{linenum}
4910 Delete any breakpoints set at or within the code of the specified
4911 @var{linenum} of the specified @var{filename}.
4914 @cindex delete breakpoints
4916 @kindex d @r{(@code{delete})}
4917 @item delete @r{[}breakpoints@r{]} @r{[}@var{list}@dots{}@r{]}
4918 Delete the breakpoints, watchpoints, or catchpoints of the breakpoint
4919 list specified as argument. If no argument is specified, delete all
4920 breakpoints (@value{GDBN} asks confirmation, unless you have @code{set
4921 confirm off}). You can abbreviate this command as @code{d}.
4925 @subsection Disabling Breakpoints
4927 @cindex enable/disable a breakpoint
4928 Rather than deleting a breakpoint, watchpoint, or catchpoint, you might
4929 prefer to @dfn{disable} it. This makes the breakpoint inoperative as if
4930 it had been deleted, but remembers the information on the breakpoint so
4931 that you can @dfn{enable} it again later.
4933 You disable and enable breakpoints, watchpoints, and catchpoints with
4934 the @code{enable} and @code{disable} commands, optionally specifying
4935 one or more breakpoint numbers as arguments. Use @code{info break} to
4936 print a list of all breakpoints, watchpoints, and catchpoints if you
4937 do not know which numbers to use.
4939 Disabling and enabling a breakpoint that has multiple locations
4940 affects all of its locations.
4942 A breakpoint, watchpoint, or catchpoint can have any of several
4943 different states of enablement:
4947 Enabled. The breakpoint stops your program. A breakpoint set
4948 with the @code{break} command starts out in this state.
4950 Disabled. The breakpoint has no effect on your program.
4952 Enabled once. The breakpoint stops your program, but then becomes
4955 Enabled for a count. The breakpoint stops your program for the next
4956 N times, then becomes disabled.
4958 Enabled for deletion. The breakpoint stops your program, but
4959 immediately after it does so it is deleted permanently. A breakpoint
4960 set with the @code{tbreak} command starts out in this state.
4963 You can use the following commands to enable or disable breakpoints,
4964 watchpoints, and catchpoints:
4968 @kindex dis @r{(@code{disable})}
4969 @item disable @r{[}breakpoints@r{]} @r{[}@var{list}@dots{}@r{]}
4970 Disable the specified breakpoints---or all breakpoints, if none are
4971 listed. A disabled breakpoint has no effect but is not forgotten. All
4972 options such as ignore-counts, conditions and commands are remembered in
4973 case the breakpoint is enabled again later. You may abbreviate
4974 @code{disable} as @code{dis}.
4977 @item enable @r{[}breakpoints@r{]} @r{[}@var{list}@dots{}@r{]}
4978 Enable the specified breakpoints (or all defined breakpoints). They
4979 become effective once again in stopping your program.
4981 @item enable @r{[}breakpoints@r{]} once @var{list}@dots{}
4982 Enable the specified breakpoints temporarily. @value{GDBN} disables any
4983 of these breakpoints immediately after stopping your program.
4985 @item enable @r{[}breakpoints@r{]} count @var{count} @var{list}@dots{}
4986 Enable the specified breakpoints temporarily. @value{GDBN} records
4987 @var{count} with each of the specified breakpoints, and decrements a
4988 breakpoint's count when it is hit. When any count reaches 0,
4989 @value{GDBN} disables that breakpoint. If a breakpoint has an ignore
4990 count (@pxref{Conditions, ,Break Conditions}), that will be
4991 decremented to 0 before @var{count} is affected.
4993 @item enable @r{[}breakpoints@r{]} delete @var{list}@dots{}
4994 Enable the specified breakpoints to work once, then die. @value{GDBN}
4995 deletes any of these breakpoints as soon as your program stops there.
4996 Breakpoints set by the @code{tbreak} command start out in this state.
4999 @c FIXME: I think the following ``Except for [...] @code{tbreak}'' is
5000 @c confusing: tbreak is also initially enabled.
5001 Except for a breakpoint set with @code{tbreak} (@pxref{Set Breaks,
5002 ,Setting Breakpoints}), breakpoints that you set are initially enabled;
5003 subsequently, they become disabled or enabled only when you use one of
5004 the commands above. (The command @code{until} can set and delete a
5005 breakpoint of its own, but it does not change the state of your other
5006 breakpoints; see @ref{Continuing and Stepping, ,Continuing and
5010 @subsection Break Conditions
5011 @cindex conditional breakpoints
5012 @cindex breakpoint conditions
5014 @c FIXME what is scope of break condition expr? Context where wanted?
5015 @c in particular for a watchpoint?
5016 The simplest sort of breakpoint breaks every time your program reaches a
5017 specified place. You can also specify a @dfn{condition} for a
5018 breakpoint. A condition is just a Boolean expression in your
5019 programming language (@pxref{Expressions, ,Expressions}). A breakpoint with
5020 a condition evaluates the expression each time your program reaches it,
5021 and your program stops only if the condition is @emph{true}.
5023 This is the converse of using assertions for program validation; in that
5024 situation, you want to stop when the assertion is violated---that is,
5025 when the condition is false. In C, if you want to test an assertion expressed
5026 by the condition @var{assert}, you should set the condition
5027 @samp{! @var{assert}} on the appropriate breakpoint.
5029 Conditions are also accepted for watchpoints; you may not need them,
5030 since a watchpoint is inspecting the value of an expression anyhow---but
5031 it might be simpler, say, to just set a watchpoint on a variable name,
5032 and specify a condition that tests whether the new value is an interesting
5035 Break conditions can have side effects, and may even call functions in
5036 your program. This can be useful, for example, to activate functions
5037 that log program progress, or to use your own print functions to
5038 format special data structures. The effects are completely predictable
5039 unless there is another enabled breakpoint at the same address. (In
5040 that case, @value{GDBN} might see the other breakpoint first and stop your
5041 program without checking the condition of this one.) Note that
5042 breakpoint commands are usually more convenient and flexible than break
5044 purpose of performing side effects when a breakpoint is reached
5045 (@pxref{Break Commands, ,Breakpoint Command Lists}).
5047 Breakpoint conditions can also be evaluated on the target's side if
5048 the target supports it. Instead of evaluating the conditions locally,
5049 @value{GDBN} encodes the expression into an agent expression
5050 (@pxref{Agent Expressions}) suitable for execution on the target,
5051 independently of @value{GDBN}. Global variables become raw memory
5052 locations, locals become stack accesses, and so forth.
5054 In this case, @value{GDBN} will only be notified of a breakpoint trigger
5055 when its condition evaluates to true. This mechanism may provide faster
5056 response times depending on the performance characteristics of the target
5057 since it does not need to keep @value{GDBN} informed about
5058 every breakpoint trigger, even those with false conditions.
5060 Break conditions can be specified when a breakpoint is set, by using
5061 @samp{if} in the arguments to the @code{break} command. @xref{Set
5062 Breaks, ,Setting Breakpoints}. They can also be changed at any time
5063 with the @code{condition} command.
5065 You can also use the @code{if} keyword with the @code{watch} command.
5066 The @code{catch} command does not recognize the @code{if} keyword;
5067 @code{condition} is the only way to impose a further condition on a
5072 @item condition @var{bnum} @var{expression}
5073 Specify @var{expression} as the break condition for breakpoint,
5074 watchpoint, or catchpoint number @var{bnum}. After you set a condition,
5075 breakpoint @var{bnum} stops your program only if the value of
5076 @var{expression} is true (nonzero, in C). When you use
5077 @code{condition}, @value{GDBN} checks @var{expression} immediately for
5078 syntactic correctness, and to determine whether symbols in it have
5079 referents in the context of your breakpoint. If @var{expression} uses
5080 symbols not referenced in the context of the breakpoint, @value{GDBN}
5081 prints an error message:
5084 No symbol "foo" in current context.
5089 not actually evaluate @var{expression} at the time the @code{condition}
5090 command (or a command that sets a breakpoint with a condition, like
5091 @code{break if @dots{}}) is given, however. @xref{Expressions, ,Expressions}.
5093 @item condition @var{bnum}
5094 Remove the condition from breakpoint number @var{bnum}. It becomes
5095 an ordinary unconditional breakpoint.
5098 @cindex ignore count (of breakpoint)
5099 A special case of a breakpoint condition is to stop only when the
5100 breakpoint has been reached a certain number of times. This is so
5101 useful that there is a special way to do it, using the @dfn{ignore
5102 count} of the breakpoint. Every breakpoint has an ignore count, which
5103 is an integer. Most of the time, the ignore count is zero, and
5104 therefore has no effect. But if your program reaches a breakpoint whose
5105 ignore count is positive, then instead of stopping, it just decrements
5106 the ignore count by one and continues. As a result, if the ignore count
5107 value is @var{n}, the breakpoint does not stop the next @var{n} times
5108 your program reaches it.
5112 @item ignore @var{bnum} @var{count}
5113 Set the ignore count of breakpoint number @var{bnum} to @var{count}.
5114 The next @var{count} times the breakpoint is reached, your program's
5115 execution does not stop; other than to decrement the ignore count, @value{GDBN}
5118 To make the breakpoint stop the next time it is reached, specify
5121 When you use @code{continue} to resume execution of your program from a
5122 breakpoint, you can specify an ignore count directly as an argument to
5123 @code{continue}, rather than using @code{ignore}. @xref{Continuing and
5124 Stepping,,Continuing and Stepping}.
5126 If a breakpoint has a positive ignore count and a condition, the
5127 condition is not checked. Once the ignore count reaches zero,
5128 @value{GDBN} resumes checking the condition.
5130 You could achieve the effect of the ignore count with a condition such
5131 as @w{@samp{$foo-- <= 0}} using a debugger convenience variable that
5132 is decremented each time. @xref{Convenience Vars, ,Convenience
5136 Ignore counts apply to breakpoints, watchpoints, and catchpoints.
5139 @node Break Commands
5140 @subsection Breakpoint Command Lists
5142 @cindex breakpoint commands
5143 You can give any breakpoint (or watchpoint or catchpoint) a series of
5144 commands to execute when your program stops due to that breakpoint. For
5145 example, you might want to print the values of certain expressions, or
5146 enable other breakpoints.
5150 @kindex end@r{ (breakpoint commands)}
5151 @item commands @r{[}@var{list}@dots{}@r{]}
5152 @itemx @dots{} @var{command-list} @dots{}
5154 Specify a list of commands for the given breakpoints. The commands
5155 themselves appear on the following lines. Type a line containing just
5156 @code{end} to terminate the commands.
5158 To remove all commands from a breakpoint, type @code{commands} and
5159 follow it immediately with @code{end}; that is, give no commands.
5161 With no argument, @code{commands} refers to the last breakpoint,
5162 watchpoint, or catchpoint set (not to the breakpoint most recently
5163 encountered). If the most recent breakpoints were set with a single
5164 command, then the @code{commands} will apply to all the breakpoints
5165 set by that command. This applies to breakpoints set by
5166 @code{rbreak}, and also applies when a single @code{break} command
5167 creates multiple breakpoints (@pxref{Ambiguous Expressions,,Ambiguous
5171 Pressing @key{RET} as a means of repeating the last @value{GDBN} command is
5172 disabled within a @var{command-list}.
5174 You can use breakpoint commands to start your program up again. Simply
5175 use the @code{continue} command, or @code{step}, or any other command
5176 that resumes execution.
5178 Any other commands in the command list, after a command that resumes
5179 execution, are ignored. This is because any time you resume execution
5180 (even with a simple @code{next} or @code{step}), you may encounter
5181 another breakpoint---which could have its own command list, leading to
5182 ambiguities about which list to execute.
5185 If the first command you specify in a command list is @code{silent}, the
5186 usual message about stopping at a breakpoint is not printed. This may
5187 be desirable for breakpoints that are to print a specific message and
5188 then continue. If none of the remaining commands print anything, you
5189 see no sign that the breakpoint was reached. @code{silent} is
5190 meaningful only at the beginning of a breakpoint command list.
5192 The commands @code{echo}, @code{output}, and @code{printf} allow you to
5193 print precisely controlled output, and are often useful in silent
5194 breakpoints. @xref{Output, ,Commands for Controlled Output}.
5196 For example, here is how you could use breakpoint commands to print the
5197 value of @code{x} at entry to @code{foo} whenever @code{x} is positive.
5203 printf "x is %d\n",x
5208 One application for breakpoint commands is to compensate for one bug so
5209 you can test for another. Put a breakpoint just after the erroneous line
5210 of code, give it a condition to detect the case in which something
5211 erroneous has been done, and give it commands to assign correct values
5212 to any variables that need them. End with the @code{continue} command
5213 so that your program does not stop, and start with the @code{silent}
5214 command so that no output is produced. Here is an example:
5225 @node Dynamic Printf
5226 @subsection Dynamic Printf
5228 @cindex dynamic printf
5230 The dynamic printf command @code{dprintf} combines a breakpoint with
5231 formatted printing of your program's data to give you the effect of
5232 inserting @code{printf} calls into your program on-the-fly, without
5233 having to recompile it.
5235 In its most basic form, the output goes to the GDB console. However,
5236 you can set the variable @code{dprintf-style} for alternate handling.
5237 For instance, you can ask to format the output by calling your
5238 program's @code{printf} function. This has the advantage that the
5239 characters go to the program's output device, so they can recorded in
5240 redirects to files and so forth.
5242 If you are doing remote debugging with a stub or agent, you can also
5243 ask to have the printf handled by the remote agent. In addition to
5244 ensuring that the output goes to the remote program's device along
5245 with any other output the program might produce, you can also ask that
5246 the dprintf remain active even after disconnecting from the remote
5247 target. Using the stub/agent is also more efficient, as it can do
5248 everything without needing to communicate with @value{GDBN}.
5252 @item dprintf @var{location},@var{template},@var{expression}[,@var{expression}@dots{}]
5253 Whenever execution reaches @var{location}, print the values of one or
5254 more @var{expressions} under the control of the string @var{template}.
5255 To print several values, separate them with commas.
5257 @item set dprintf-style @var{style}
5258 Set the dprintf output to be handled in one of several different
5259 styles enumerated below. A change of style affects all existing
5260 dynamic printfs immediately. (If you need individual control over the
5261 print commands, simply define normal breakpoints with
5262 explicitly-supplied command lists.)
5266 @kindex dprintf-style gdb
5267 Handle the output using the @value{GDBN} @code{printf} command.
5270 @kindex dprintf-style call
5271 Handle the output by calling a function in your program (normally
5275 @kindex dprintf-style agent
5276 Have the remote debugging agent (such as @code{gdbserver}) handle
5277 the output itself. This style is only available for agents that
5278 support running commands on the target.
5281 @item set dprintf-function @var{function}
5282 Set the function to call if the dprintf style is @code{call}. By
5283 default its value is @code{printf}. You may set it to any expression.
5284 that @value{GDBN} can evaluate to a function, as per the @code{call}
5287 @item set dprintf-channel @var{channel}
5288 Set a ``channel'' for dprintf. If set to a non-empty value,
5289 @value{GDBN} will evaluate it as an expression and pass the result as
5290 a first argument to the @code{dprintf-function}, in the manner of
5291 @code{fprintf} and similar functions. Otherwise, the dprintf format
5292 string will be the first argument, in the manner of @code{printf}.
5294 As an example, if you wanted @code{dprintf} output to go to a logfile
5295 that is a standard I/O stream assigned to the variable @code{mylog},
5296 you could do the following:
5299 (gdb) set dprintf-style call
5300 (gdb) set dprintf-function fprintf
5301 (gdb) set dprintf-channel mylog
5302 (gdb) dprintf 25,"at line 25, glob=%d\n",glob
5303 Dprintf 1 at 0x123456: file main.c, line 25.
5305 1 dprintf keep y 0x00123456 in main at main.c:25
5306 call (void) fprintf (mylog,"at line 25, glob=%d\n",glob)
5311 Note that the @code{info break} displays the dynamic printf commands
5312 as normal breakpoint commands; you can thus easily see the effect of
5313 the variable settings.
5315 @item set disconnected-dprintf on
5316 @itemx set disconnected-dprintf off
5317 @kindex set disconnected-dprintf
5318 Choose whether @code{dprintf} commands should continue to run if
5319 @value{GDBN} has disconnected from the target. This only applies
5320 if the @code{dprintf-style} is @code{agent}.
5322 @item show disconnected-dprintf off
5323 @kindex show disconnected-dprintf
5324 Show the current choice for disconnected @code{dprintf}.
5328 @value{GDBN} does not check the validity of function and channel,
5329 relying on you to supply values that are meaningful for the contexts
5330 in which they are being used. For instance, the function and channel
5331 may be the values of local variables, but if that is the case, then
5332 all enabled dynamic prints must be at locations within the scope of
5333 those locals. If evaluation fails, @value{GDBN} will report an error.
5335 @node Save Breakpoints
5336 @subsection How to save breakpoints to a file
5338 To save breakpoint definitions to a file use the @w{@code{save
5339 breakpoints}} command.
5342 @kindex save breakpoints
5343 @cindex save breakpoints to a file for future sessions
5344 @item save breakpoints [@var{filename}]
5345 This command saves all current breakpoint definitions together with
5346 their commands and ignore counts, into a file @file{@var{filename}}
5347 suitable for use in a later debugging session. This includes all
5348 types of breakpoints (breakpoints, watchpoints, catchpoints,
5349 tracepoints). To read the saved breakpoint definitions, use the
5350 @code{source} command (@pxref{Command Files}). Note that watchpoints
5351 with expressions involving local variables may fail to be recreated
5352 because it may not be possible to access the context where the
5353 watchpoint is valid anymore. Because the saved breakpoint definitions
5354 are simply a sequence of @value{GDBN} commands that recreate the
5355 breakpoints, you can edit the file in your favorite editing program,
5356 and remove the breakpoint definitions you're not interested in, or
5357 that can no longer be recreated.
5360 @node Static Probe Points
5361 @subsection Static Probe Points
5363 @cindex static probe point, SystemTap
5364 @cindex static probe point, DTrace
5365 @value{GDBN} supports @dfn{SDT} probes in the code. @acronym{SDT} stands
5366 for Statically Defined Tracing, and the probes are designed to have a tiny
5367 runtime code and data footprint, and no dynamic relocations.
5369 Currently, the following types of probes are supported on
5370 ELF-compatible systems:
5374 @item @code{SystemTap} (@uref{http://sourceware.org/systemtap/})
5375 @acronym{SDT} probes@footnote{See
5376 @uref{http://sourceware.org/systemtap/wiki/AddingUserSpaceProbingToApps}
5377 for more information on how to add @code{SystemTap} @acronym{SDT}
5378 probes in your applications.}. @code{SystemTap} probes are usable
5379 from assembly, C and C@t{++} languages@footnote{See
5380 @uref{http://sourceware.org/systemtap/wiki/UserSpaceProbeImplementation}
5381 for a good reference on how the @acronym{SDT} probes are implemented.}.
5383 @item @code{DTrace} (@uref{http://oss.oracle.com/projects/DTrace})
5384 @acronym{USDT} probes. @code{DTrace} probes are usable from C and
5388 @cindex semaphores on static probe points
5389 Some @code{SystemTap} probes have an associated semaphore variable;
5390 for instance, this happens automatically if you defined your probe
5391 using a DTrace-style @file{.d} file. If your probe has a semaphore,
5392 @value{GDBN} will automatically enable it when you specify a
5393 breakpoint using the @samp{-probe-stap} notation. But, if you put a
5394 breakpoint at a probe's location by some other method (e.g.,
5395 @code{break file:line}), then @value{GDBN} will not automatically set
5396 the semaphore. @code{DTrace} probes do not support semaphores.
5398 You can examine the available static static probes using @code{info
5399 probes}, with optional arguments:
5403 @item info probes @r{[}@var{type}@r{]} @r{[}@var{provider} @r{[}@var{name} @r{[}@var{objfile}@r{]}@r{]}@r{]}
5404 If given, @var{type} is either @code{stap} for listing
5405 @code{SystemTap} probes or @code{dtrace} for listing @code{DTrace}
5406 probes. If omitted all probes are listed regardless of their types.
5408 If given, @var{provider} is a regular expression used to match against provider
5409 names when selecting which probes to list. If omitted, probes by all
5410 probes from all providers are listed.
5412 If given, @var{name} is a regular expression to match against probe names
5413 when selecting which probes to list. If omitted, probe names are not
5414 considered when deciding whether to display them.
5416 If given, @var{objfile} is a regular expression used to select which
5417 object files (executable or shared libraries) to examine. If not
5418 given, all object files are considered.
5420 @item info probes all
5421 List the available static probes, from all types.
5424 @cindex enabling and disabling probes
5425 Some probe points can be enabled and/or disabled. The effect of
5426 enabling or disabling a probe depends on the type of probe being
5427 handled. Some @code{DTrace} probes can be enabled or
5428 disabled, but @code{SystemTap} probes cannot be disabled.
5430 You can enable (or disable) one or more probes using the following
5431 commands, with optional arguments:
5434 @kindex enable probes
5435 @item enable probes @r{[}@var{provider} @r{[}@var{name} @r{[}@var{objfile}@r{]}@r{]}@r{]}
5436 If given, @var{provider} is a regular expression used to match against
5437 provider names when selecting which probes to enable. If omitted,
5438 all probes from all providers are enabled.
5440 If given, @var{name} is a regular expression to match against probe
5441 names when selecting which probes to enable. If omitted, probe names
5442 are not considered when deciding whether to enable them.
5444 If given, @var{objfile} is a regular expression used to select which
5445 object files (executable or shared libraries) to examine. If not
5446 given, all object files are considered.
5448 @kindex disable probes
5449 @item disable probes @r{[}@var{provider} @r{[}@var{name} @r{[}@var{objfile}@r{]}@r{]}@r{]}
5450 See the @code{enable probes} command above for a description of the
5451 optional arguments accepted by this command.
5454 @vindex $_probe_arg@r{, convenience variable}
5455 A probe may specify up to twelve arguments. These are available at the
5456 point at which the probe is defined---that is, when the current PC is
5457 at the probe's location. The arguments are available using the
5458 convenience variables (@pxref{Convenience Vars})
5459 @code{$_probe_arg0}@dots{}@code{$_probe_arg11}. In @code{SystemTap}
5460 probes each probe argument is an integer of the appropriate size;
5461 types are not preserved. In @code{DTrace} probes types are preserved
5462 provided that they are recognized as such by @value{GDBN}; otherwise
5463 the value of the probe argument will be a long integer. The
5464 convenience variable @code{$_probe_argc} holds the number of arguments
5465 at the current probe point.
5467 These variables are always available, but attempts to access them at
5468 any location other than a probe point will cause @value{GDBN} to give
5472 @c @ifclear BARETARGET
5473 @node Error in Breakpoints
5474 @subsection ``Cannot insert breakpoints''
5476 If you request too many active hardware-assisted breakpoints and
5477 watchpoints, you will see this error message:
5479 @c FIXME: the precise wording of this message may change; the relevant
5480 @c source change is not committed yet (Sep 3, 1999).
5482 Stopped; cannot insert breakpoints.
5483 You may have requested too many hardware breakpoints and watchpoints.
5487 This message is printed when you attempt to resume the program, since
5488 only then @value{GDBN} knows exactly how many hardware breakpoints and
5489 watchpoints it needs to insert.
5491 When this message is printed, you need to disable or remove some of the
5492 hardware-assisted breakpoints and watchpoints, and then continue.
5494 @node Breakpoint-related Warnings
5495 @subsection ``Breakpoint address adjusted...''
5496 @cindex breakpoint address adjusted
5498 Some processor architectures place constraints on the addresses at
5499 which breakpoints may be placed. For architectures thus constrained,
5500 @value{GDBN} will attempt to adjust the breakpoint's address to comply
5501 with the constraints dictated by the architecture.
5503 One example of such an architecture is the Fujitsu FR-V. The FR-V is
5504 a VLIW architecture in which a number of RISC-like instructions may be
5505 bundled together for parallel execution. The FR-V architecture
5506 constrains the location of a breakpoint instruction within such a
5507 bundle to the instruction with the lowest address. @value{GDBN}
5508 honors this constraint by adjusting a breakpoint's address to the
5509 first in the bundle.
5511 It is not uncommon for optimized code to have bundles which contain
5512 instructions from different source statements, thus it may happen that
5513 a breakpoint's address will be adjusted from one source statement to
5514 another. Since this adjustment may significantly alter @value{GDBN}'s
5515 breakpoint related behavior from what the user expects, a warning is
5516 printed when the breakpoint is first set and also when the breakpoint
5519 A warning like the one below is printed when setting a breakpoint
5520 that's been subject to address adjustment:
5523 warning: Breakpoint address adjusted from 0x00010414 to 0x00010410.
5526 Such warnings are printed both for user settable and @value{GDBN}'s
5527 internal breakpoints. If you see one of these warnings, you should
5528 verify that a breakpoint set at the adjusted address will have the
5529 desired affect. If not, the breakpoint in question may be removed and
5530 other breakpoints may be set which will have the desired behavior.
5531 E.g., it may be sufficient to place the breakpoint at a later
5532 instruction. A conditional breakpoint may also be useful in some
5533 cases to prevent the breakpoint from triggering too often.
5535 @value{GDBN} will also issue a warning when stopping at one of these
5536 adjusted breakpoints:
5539 warning: Breakpoint 1 address previously adjusted from 0x00010414
5543 When this warning is encountered, it may be too late to take remedial
5544 action except in cases where the breakpoint is hit earlier or more
5545 frequently than expected.
5547 @node Continuing and Stepping
5548 @section Continuing and Stepping
5552 @cindex resuming execution
5553 @dfn{Continuing} means resuming program execution until your program
5554 completes normally. In contrast, @dfn{stepping} means executing just
5555 one more ``step'' of your program, where ``step'' may mean either one
5556 line of source code, or one machine instruction (depending on what
5557 particular command you use). Either when continuing or when stepping,
5558 your program may stop even sooner, due to a breakpoint or a signal. (If
5559 it stops due to a signal, you may want to use @code{handle}, or use
5560 @samp{signal 0} to resume execution (@pxref{Signals, ,Signals}),
5561 or you may step into the signal's handler (@pxref{stepping and signal
5566 @kindex c @r{(@code{continue})}
5567 @kindex fg @r{(resume foreground execution)}
5568 @item continue @r{[}@var{ignore-count}@r{]}
5569 @itemx c @r{[}@var{ignore-count}@r{]}
5570 @itemx fg @r{[}@var{ignore-count}@r{]}
5571 Resume program execution, at the address where your program last stopped;
5572 any breakpoints set at that address are bypassed. The optional argument
5573 @var{ignore-count} allows you to specify a further number of times to
5574 ignore a breakpoint at this location; its effect is like that of
5575 @code{ignore} (@pxref{Conditions, ,Break Conditions}).
5577 The argument @var{ignore-count} is meaningful only when your program
5578 stopped due to a breakpoint. At other times, the argument to
5579 @code{continue} is ignored.
5581 The synonyms @code{c} and @code{fg} (for @dfn{foreground}, as the
5582 debugged program is deemed to be the foreground program) are provided
5583 purely for convenience, and have exactly the same behavior as
5587 To resume execution at a different place, you can use @code{return}
5588 (@pxref{Returning, ,Returning from a Function}) to go back to the
5589 calling function; or @code{jump} (@pxref{Jumping, ,Continuing at a
5590 Different Address}) to go to an arbitrary location in your program.
5592 A typical technique for using stepping is to set a breakpoint
5593 (@pxref{Breakpoints, ,Breakpoints; Watchpoints; and Catchpoints}) at the
5594 beginning of the function or the section of your program where a problem
5595 is believed to lie, run your program until it stops at that breakpoint,
5596 and then step through the suspect area, examining the variables that are
5597 interesting, until you see the problem happen.
5601 @kindex s @r{(@code{step})}
5603 Continue running your program until control reaches a different source
5604 line, then stop it and return control to @value{GDBN}. This command is
5605 abbreviated @code{s}.
5608 @c "without debugging information" is imprecise; actually "without line
5609 @c numbers in the debugging information". (gcc -g1 has debugging info but
5610 @c not line numbers). But it seems complex to try to make that
5611 @c distinction here.
5612 @emph{Warning:} If you use the @code{step} command while control is
5613 within a function that was compiled without debugging information,
5614 execution proceeds until control reaches a function that does have
5615 debugging information. Likewise, it will not step into a function which
5616 is compiled without debugging information. To step through functions
5617 without debugging information, use the @code{stepi} command, described
5621 The @code{step} command only stops at the first instruction of a source
5622 line. This prevents the multiple stops that could otherwise occur in
5623 @code{switch} statements, @code{for} loops, etc. @code{step} continues
5624 to stop if a function that has debugging information is called within
5625 the line. In other words, @code{step} @emph{steps inside} any functions
5626 called within the line.
5628 Also, the @code{step} command only enters a function if there is line
5629 number information for the function. Otherwise it acts like the
5630 @code{next} command. This avoids problems when using @code{cc -gl}
5631 on @acronym{MIPS} machines. Previously, @code{step} entered subroutines if there
5632 was any debugging information about the routine.
5634 @item step @var{count}
5635 Continue running as in @code{step}, but do so @var{count} times. If a
5636 breakpoint is reached, or a signal not related to stepping occurs before
5637 @var{count} steps, stepping stops right away.
5640 @kindex n @r{(@code{next})}
5641 @item next @r{[}@var{count}@r{]}
5642 Continue to the next source line in the current (innermost) stack frame.
5643 This is similar to @code{step}, but function calls that appear within
5644 the line of code are executed without stopping. Execution stops when
5645 control reaches a different line of code at the original stack level
5646 that was executing when you gave the @code{next} command. This command
5647 is abbreviated @code{n}.
5649 An argument @var{count} is a repeat count, as for @code{step}.
5652 @c FIX ME!! Do we delete this, or is there a way it fits in with
5653 @c the following paragraph? --- Vctoria
5655 @c @code{next} within a function that lacks debugging information acts like
5656 @c @code{step}, but any function calls appearing within the code of the
5657 @c function are executed without stopping.
5659 The @code{next} command only stops at the first instruction of a
5660 source line. This prevents multiple stops that could otherwise occur in
5661 @code{switch} statements, @code{for} loops, etc.
5663 @kindex set step-mode
5665 @cindex functions without line info, and stepping
5666 @cindex stepping into functions with no line info
5667 @itemx set step-mode on
5668 The @code{set step-mode on} command causes the @code{step} command to
5669 stop at the first instruction of a function which contains no debug line
5670 information rather than stepping over it.
5672 This is useful in cases where you may be interested in inspecting the
5673 machine instructions of a function which has no symbolic info and do not
5674 want @value{GDBN} to automatically skip over this function.
5676 @item set step-mode off
5677 Causes the @code{step} command to step over any functions which contains no
5678 debug information. This is the default.
5680 @item show step-mode
5681 Show whether @value{GDBN} will stop in or step over functions without
5682 source line debug information.
5685 @kindex fin @r{(@code{finish})}
5687 Continue running until just after function in the selected stack frame
5688 returns. Print the returned value (if any). This command can be
5689 abbreviated as @code{fin}.
5691 Contrast this with the @code{return} command (@pxref{Returning,
5692 ,Returning from a Function}).
5694 @kindex set print finish
5695 @kindex show print finish
5696 @item set print finish @r{[}on|off@r{]}
5697 @itemx show print finish
5698 By default the @code{finish} command will show the value that is
5699 returned by the function. This can be disabled using @code{set print
5700 finish off}. When disabled, the value is still entered into the value
5701 history (@pxref{Value History}), but not displayed.
5704 @kindex u @r{(@code{until})}
5705 @cindex run until specified location
5708 Continue running until a source line past the current line, in the
5709 current stack frame, is reached. This command is used to avoid single
5710 stepping through a loop more than once. It is like the @code{next}
5711 command, except that when @code{until} encounters a jump, it
5712 automatically continues execution until the program counter is greater
5713 than the address of the jump.
5715 This means that when you reach the end of a loop after single stepping
5716 though it, @code{until} makes your program continue execution until it
5717 exits the loop. In contrast, a @code{next} command at the end of a loop
5718 simply steps back to the beginning of the loop, which forces you to step
5719 through the next iteration.
5721 @code{until} always stops your program if it attempts to exit the current
5724 @code{until} may produce somewhat counterintuitive results if the order
5725 of machine code does not match the order of the source lines. For
5726 example, in the following excerpt from a debugging session, the @code{f}
5727 (@code{frame}) command shows that execution is stopped at line
5728 @code{206}; yet when we use @code{until}, we get to line @code{195}:
5732 #0 main (argc=4, argv=0xf7fffae8) at m4.c:206
5734 (@value{GDBP}) until
5735 195 for ( ; argc > 0; NEXTARG) @{
5738 This happened because, for execution efficiency, the compiler had
5739 generated code for the loop closure test at the end, rather than the
5740 start, of the loop---even though the test in a C @code{for}-loop is
5741 written before the body of the loop. The @code{until} command appeared
5742 to step back to the beginning of the loop when it advanced to this
5743 expression; however, it has not really gone to an earlier
5744 statement---not in terms of the actual machine code.
5746 @code{until} with no argument works by means of single
5747 instruction stepping, and hence is slower than @code{until} with an
5750 @item until @var{location}
5751 @itemx u @var{location}
5752 Continue running your program until either the specified @var{location} is
5753 reached, or the current stack frame returns. The location is any of
5754 the forms described in @ref{Specify Location}.
5755 This form of the command uses temporary breakpoints, and
5756 hence is quicker than @code{until} without an argument. The specified
5757 location is actually reached only if it is in the current frame. This
5758 implies that @code{until} can be used to skip over recursive function
5759 invocations. For instance in the code below, if the current location is
5760 line @code{96}, issuing @code{until 99} will execute the program up to
5761 line @code{99} in the same invocation of factorial, i.e., after the inner
5762 invocations have returned.
5765 94 int factorial (int value)
5767 96 if (value > 1) @{
5768 97 value *= factorial (value - 1);
5775 @kindex advance @var{location}
5776 @item advance @var{location}
5777 Continue running the program up to the given @var{location}. An argument is
5778 required, which should be of one of the forms described in
5779 @ref{Specify Location}.
5780 Execution will also stop upon exit from the current stack
5781 frame. This command is similar to @code{until}, but @code{advance} will
5782 not skip over recursive function calls, and the target location doesn't
5783 have to be in the same frame as the current one.
5787 @kindex si @r{(@code{stepi})}
5789 @itemx stepi @var{arg}
5791 Execute one machine instruction, then stop and return to the debugger.
5793 It is often useful to do @samp{display/i $pc} when stepping by machine
5794 instructions. This makes @value{GDBN} automatically display the next
5795 instruction to be executed, each time your program stops. @xref{Auto
5796 Display,, Automatic Display}.
5798 An argument is a repeat count, as in @code{step}.
5802 @kindex ni @r{(@code{nexti})}
5804 @itemx nexti @var{arg}
5806 Execute one machine instruction, but if it is a function call,
5807 proceed until the function returns.
5809 An argument is a repeat count, as in @code{next}.
5813 @anchor{range stepping}
5814 @cindex range stepping
5815 @cindex target-assisted range stepping
5816 By default, and if available, @value{GDBN} makes use of
5817 target-assisted @dfn{range stepping}. In other words, whenever you
5818 use a stepping command (e.g., @code{step}, @code{next}), @value{GDBN}
5819 tells the target to step the corresponding range of instruction
5820 addresses instead of issuing multiple single-steps. This speeds up
5821 line stepping, particularly for remote targets. Ideally, there should
5822 be no reason you would want to turn range stepping off. However, it's
5823 possible that a bug in the debug info, a bug in the remote stub (for
5824 remote targets), or even a bug in @value{GDBN} could make line
5825 stepping behave incorrectly when target-assisted range stepping is
5826 enabled. You can use the following command to turn off range stepping
5830 @kindex set range-stepping
5831 @kindex show range-stepping
5832 @item set range-stepping
5833 @itemx show range-stepping
5834 Control whether range stepping is enabled.
5836 If @code{on}, and the target supports it, @value{GDBN} tells the
5837 target to step a range of addresses itself, instead of issuing
5838 multiple single-steps. If @code{off}, @value{GDBN} always issues
5839 single-steps, even if range stepping is supported by the target. The
5840 default is @code{on}.
5844 @node Skipping Over Functions and Files
5845 @section Skipping Over Functions and Files
5846 @cindex skipping over functions and files
5848 The program you are debugging may contain some functions which are
5849 uninteresting to debug. The @code{skip} command lets you tell @value{GDBN} to
5850 skip a function, all functions in a file or a particular function in
5851 a particular file when stepping.
5853 For example, consider the following C function:
5864 Suppose you wish to step into the functions @code{foo} and @code{bar}, but you
5865 are not interested in stepping through @code{boring}. If you run @code{step}
5866 at line 103, you'll enter @code{boring()}, but if you run @code{next}, you'll
5867 step over both @code{foo} and @code{boring}!
5869 One solution is to @code{step} into @code{boring} and use the @code{finish}
5870 command to immediately exit it. But this can become tedious if @code{boring}
5871 is called from many places.
5873 A more flexible solution is to execute @kbd{skip boring}. This instructs
5874 @value{GDBN} never to step into @code{boring}. Now when you execute
5875 @code{step} at line 103, you'll step over @code{boring} and directly into
5878 Functions may be skipped by providing either a function name, linespec
5879 (@pxref{Specify Location}), regular expression that matches the function's
5880 name, file name or a @code{glob}-style pattern that matches the file name.
5882 On Posix systems the form of the regular expression is
5883 ``Extended Regular Expressions''. See for example @samp{man 7 regex}
5884 on @sc{gnu}/Linux systems. On non-Posix systems the form of the regular
5885 expression is whatever is provided by the @code{regcomp} function of
5886 the underlying system.
5887 See for example @samp{man 7 glob} on @sc{gnu}/Linux systems for a
5888 description of @code{glob}-style patterns.
5892 @item skip @r{[}@var{options}@r{]}
5893 The basic form of the @code{skip} command takes zero or more options
5894 that specify what to skip.
5895 The @var{options} argument is any useful combination of the following:
5898 @item -file @var{file}
5899 @itemx -fi @var{file}
5900 Functions in @var{file} will be skipped over when stepping.
5902 @item -gfile @var{file-glob-pattern}
5903 @itemx -gfi @var{file-glob-pattern}
5904 @cindex skipping over files via glob-style patterns
5905 Functions in files matching @var{file-glob-pattern} will be skipped
5909 (gdb) skip -gfi utils/*.c
5912 @item -function @var{linespec}
5913 @itemx -fu @var{linespec}
5914 Functions named by @var{linespec} or the function containing the line
5915 named by @var{linespec} will be skipped over when stepping.
5916 @xref{Specify Location}.
5918 @item -rfunction @var{regexp}
5919 @itemx -rfu @var{regexp}
5920 @cindex skipping over functions via regular expressions
5921 Functions whose name matches @var{regexp} will be skipped over when stepping.
5923 This form is useful for complex function names.
5924 For example, there is generally no need to step into C@t{++} @code{std::string}
5925 constructors or destructors. Plus with C@t{++} templates it can be hard to
5926 write out the full name of the function, and often it doesn't matter what
5927 the template arguments are. Specifying the function to be skipped as a
5928 regular expression makes this easier.
5931 (gdb) skip -rfu ^std::(allocator|basic_string)<.*>::~?\1 *\(
5934 If you want to skip every templated C@t{++} constructor and destructor
5935 in the @code{std} namespace you can do:
5938 (gdb) skip -rfu ^std::([a-zA-z0-9_]+)<.*>::~?\1 *\(
5942 If no options are specified, the function you're currently debugging
5945 @kindex skip function
5946 @item skip function @r{[}@var{linespec}@r{]}
5947 After running this command, the function named by @var{linespec} or the
5948 function containing the line named by @var{linespec} will be skipped over when
5949 stepping. @xref{Specify Location}.
5951 If you do not specify @var{linespec}, the function you're currently debugging
5954 (If you have a function called @code{file} that you want to skip, use
5955 @kbd{skip function file}.)
5958 @item skip file @r{[}@var{filename}@r{]}
5959 After running this command, any function whose source lives in @var{filename}
5960 will be skipped over when stepping.
5963 (gdb) skip file boring.c
5964 File boring.c will be skipped when stepping.
5967 If you do not specify @var{filename}, functions whose source lives in the file
5968 you're currently debugging will be skipped.
5971 Skips can be listed, deleted, disabled, and enabled, much like breakpoints.
5972 These are the commands for managing your list of skips:
5976 @item info skip @r{[}@var{range}@r{]}
5977 Print details about the specified skip(s). If @var{range} is not specified,
5978 print a table with details about all functions and files marked for skipping.
5979 @code{info skip} prints the following information about each skip:
5983 A number identifying this skip.
5984 @item Enabled or Disabled
5985 Enabled skips are marked with @samp{y}.
5986 Disabled skips are marked with @samp{n}.
5988 If the file name is a @samp{glob} pattern this is @samp{y}.
5989 Otherwise it is @samp{n}.
5991 The name or @samp{glob} pattern of the file to be skipped.
5992 If no file is specified this is @samp{<none>}.
5994 If the function name is a @samp{regular expression} this is @samp{y}.
5995 Otherwise it is @samp{n}.
5997 The name or regular expression of the function to skip.
5998 If no function is specified this is @samp{<none>}.
6002 @item skip delete @r{[}@var{range}@r{]}
6003 Delete the specified skip(s). If @var{range} is not specified, delete all
6007 @item skip enable @r{[}@var{range}@r{]}
6008 Enable the specified skip(s). If @var{range} is not specified, enable all
6011 @kindex skip disable
6012 @item skip disable @r{[}@var{range}@r{]}
6013 Disable the specified skip(s). If @var{range} is not specified, disable all
6016 @kindex set debug skip
6017 @item set debug skip @r{[}on|off@r{]}
6018 Set whether to print the debug output about skipping files and functions.
6020 @kindex show debug skip
6021 @item show debug skip
6022 Show whether the debug output about skipping files and functions is printed.
6030 A signal is an asynchronous event that can happen in a program. The
6031 operating system defines the possible kinds of signals, and gives each
6032 kind a name and a number. For example, in Unix @code{SIGINT} is the
6033 signal a program gets when you type an interrupt character (often @kbd{Ctrl-c});
6034 @code{SIGSEGV} is the signal a program gets from referencing a place in
6035 memory far away from all the areas in use; @code{SIGALRM} occurs when
6036 the alarm clock timer goes off (which happens only if your program has
6037 requested an alarm).
6039 @cindex fatal signals
6040 Some signals, including @code{SIGALRM}, are a normal part of the
6041 functioning of your program. Others, such as @code{SIGSEGV}, indicate
6042 errors; these signals are @dfn{fatal} (they kill your program immediately) if the
6043 program has not specified in advance some other way to handle the signal.
6044 @code{SIGINT} does not indicate an error in your program, but it is normally
6045 fatal so it can carry out the purpose of the interrupt: to kill the program.
6047 @value{GDBN} has the ability to detect any occurrence of a signal in your
6048 program. You can tell @value{GDBN} in advance what to do for each kind of
6051 @cindex handling signals
6052 Normally, @value{GDBN} is set up to let the non-erroneous signals like
6053 @code{SIGALRM} be silently passed to your program
6054 (so as not to interfere with their role in the program's functioning)
6055 but to stop your program immediately whenever an error signal happens.
6056 You can change these settings with the @code{handle} command.
6059 @kindex info signals
6063 Print a table of all the kinds of signals and how @value{GDBN} has been told to
6064 handle each one. You can use this to see the signal numbers of all
6065 the defined types of signals.
6067 @item info signals @var{sig}
6068 Similar, but print information only about the specified signal number.
6070 @code{info handle} is an alias for @code{info signals}.
6072 @item catch signal @r{[}@var{signal}@dots{} @r{|} @samp{all}@r{]}
6073 Set a catchpoint for the indicated signals. @xref{Set Catchpoints},
6074 for details about this command.
6077 @item handle @var{signal} @r{[}@var{keywords}@dots{}@r{]}
6078 Change the way @value{GDBN} handles signal @var{signal}. The @var{signal}
6079 can be the number of a signal or its name (with or without the
6080 @samp{SIG} at the beginning); a list of signal numbers of the form
6081 @samp{@var{low}-@var{high}}; or the word @samp{all}, meaning all the
6082 known signals. Optional arguments @var{keywords}, described below,
6083 say what change to make.
6087 The keywords allowed by the @code{handle} command can be abbreviated.
6088 Their full names are:
6092 @value{GDBN} should not stop your program when this signal happens. It may
6093 still print a message telling you that the signal has come in.
6096 @value{GDBN} should stop your program when this signal happens. This implies
6097 the @code{print} keyword as well.
6100 @value{GDBN} should print a message when this signal happens.
6103 @value{GDBN} should not mention the occurrence of the signal at all. This
6104 implies the @code{nostop} keyword as well.
6108 @value{GDBN} should allow your program to see this signal; your program
6109 can handle the signal, or else it may terminate if the signal is fatal
6110 and not handled. @code{pass} and @code{noignore} are synonyms.
6114 @value{GDBN} should not allow your program to see this signal.
6115 @code{nopass} and @code{ignore} are synonyms.
6119 When a signal stops your program, the signal is not visible to the
6121 continue. Your program sees the signal then, if @code{pass} is in
6122 effect for the signal in question @emph{at that time}. In other words,
6123 after @value{GDBN} reports a signal, you can use the @code{handle}
6124 command with @code{pass} or @code{nopass} to control whether your
6125 program sees that signal when you continue.
6127 The default is set to @code{nostop}, @code{noprint}, @code{pass} for
6128 non-erroneous signals such as @code{SIGALRM}, @code{SIGWINCH} and
6129 @code{SIGCHLD}, and to @code{stop}, @code{print}, @code{pass} for the
6132 You can also use the @code{signal} command to prevent your program from
6133 seeing a signal, or cause it to see a signal it normally would not see,
6134 or to give it any signal at any time. For example, if your program stopped
6135 due to some sort of memory reference error, you might store correct
6136 values into the erroneous variables and continue, hoping to see more
6137 execution; but your program would probably terminate immediately as
6138 a result of the fatal signal once it saw the signal. To prevent this,
6139 you can continue with @samp{signal 0}. @xref{Signaling, ,Giving your
6142 @cindex stepping and signal handlers
6143 @anchor{stepping and signal handlers}
6145 @value{GDBN} optimizes for stepping the mainline code. If a signal
6146 that has @code{handle nostop} and @code{handle pass} set arrives while
6147 a stepping command (e.g., @code{stepi}, @code{step}, @code{next}) is
6148 in progress, @value{GDBN} lets the signal handler run and then resumes
6149 stepping the mainline code once the signal handler returns. In other
6150 words, @value{GDBN} steps over the signal handler. This prevents
6151 signals that you've specified as not interesting (with @code{handle
6152 nostop}) from changing the focus of debugging unexpectedly. Note that
6153 the signal handler itself may still hit a breakpoint, stop for another
6154 signal that has @code{handle stop} in effect, or for any other event
6155 that normally results in stopping the stepping command sooner. Also
6156 note that @value{GDBN} still informs you that the program received a
6157 signal if @code{handle print} is set.
6159 @anchor{stepping into signal handlers}
6161 If you set @code{handle pass} for a signal, and your program sets up a
6162 handler for it, then issuing a stepping command, such as @code{step}
6163 or @code{stepi}, when your program is stopped due to the signal will
6164 step @emph{into} the signal handler (if the target supports that).
6166 Likewise, if you use the @code{queue-signal} command to queue a signal
6167 to be delivered to the current thread when execution of the thread
6168 resumes (@pxref{Signaling, ,Giving your Program a Signal}), then a
6169 stepping command will step into the signal handler.
6171 Here's an example, using @code{stepi} to step to the first instruction
6172 of @code{SIGUSR1}'s handler:
6175 (@value{GDBP}) handle SIGUSR1
6176 Signal Stop Print Pass to program Description
6177 SIGUSR1 Yes Yes Yes User defined signal 1
6181 Program received signal SIGUSR1, User defined signal 1.
6182 main () sigusr1.c:28
6185 sigusr1_handler () at sigusr1.c:9
6189 The same, but using @code{queue-signal} instead of waiting for the
6190 program to receive the signal first:
6195 (@value{GDBP}) queue-signal SIGUSR1
6197 sigusr1_handler () at sigusr1.c:9
6202 @cindex extra signal information
6203 @anchor{extra signal information}
6205 On some targets, @value{GDBN} can inspect extra signal information
6206 associated with the intercepted signal, before it is actually
6207 delivered to the program being debugged. This information is exported
6208 by the convenience variable @code{$_siginfo}, and consists of data
6209 that is passed by the kernel to the signal handler at the time of the
6210 receipt of a signal. The data type of the information itself is
6211 target dependent. You can see the data type using the @code{ptype
6212 $_siginfo} command. On Unix systems, it typically corresponds to the
6213 standard @code{siginfo_t} type, as defined in the @file{signal.h}
6216 Here's an example, on a @sc{gnu}/Linux system, printing the stray
6217 referenced address that raised a segmentation fault.
6221 (@value{GDBP}) continue
6222 Program received signal SIGSEGV, Segmentation fault.
6223 0x0000000000400766 in main ()
6225 (@value{GDBP}) ptype $_siginfo
6232 struct @{...@} _kill;
6233 struct @{...@} _timer;
6235 struct @{...@} _sigchld;
6236 struct @{...@} _sigfault;
6237 struct @{...@} _sigpoll;
6240 (@value{GDBP}) ptype $_siginfo._sifields._sigfault
6244 (@value{GDBP}) p $_siginfo._sifields._sigfault.si_addr
6245 $1 = (void *) 0x7ffff7ff7000
6249 Depending on target support, @code{$_siginfo} may also be writable.
6251 @cindex Intel MPX boundary violations
6252 @cindex boundary violations, Intel MPX
6253 On some targets, a @code{SIGSEGV} can be caused by a boundary
6254 violation, i.e., accessing an address outside of the allowed range.
6255 In those cases @value{GDBN} may displays additional information,
6256 depending on how @value{GDBN} has been told to handle the signal.
6257 With @code{handle stop SIGSEGV}, @value{GDBN} displays the violation
6258 kind: "Upper" or "Lower", the memory address accessed and the
6259 bounds, while with @code{handle nostop SIGSEGV} no additional
6260 information is displayed.
6262 The usual output of a segfault is:
6264 Program received signal SIGSEGV, Segmentation fault
6265 0x0000000000400d7c in upper () at i386-mpx-sigsegv.c:68
6266 68 value = *(p + len);
6269 While a bound violation is presented as:
6271 Program received signal SIGSEGV, Segmentation fault
6272 Upper bound violation while accessing address 0x7fffffffc3b3
6273 Bounds: [lower = 0x7fffffffc390, upper = 0x7fffffffc3a3]
6274 0x0000000000400d7c in upper () at i386-mpx-sigsegv.c:68
6275 68 value = *(p + len);
6279 @section Stopping and Starting Multi-thread Programs
6281 @cindex stopped threads
6282 @cindex threads, stopped
6284 @cindex continuing threads
6285 @cindex threads, continuing
6287 @value{GDBN} supports debugging programs with multiple threads
6288 (@pxref{Threads,, Debugging Programs with Multiple Threads}). There
6289 are two modes of controlling execution of your program within the
6290 debugger. In the default mode, referred to as @dfn{all-stop mode},
6291 when any thread in your program stops (for example, at a breakpoint
6292 or while being stepped), all other threads in the program are also stopped by
6293 @value{GDBN}. On some targets, @value{GDBN} also supports
6294 @dfn{non-stop mode}, in which other threads can continue to run freely while
6295 you examine the stopped thread in the debugger.
6298 * All-Stop Mode:: All threads stop when GDB takes control
6299 * Non-Stop Mode:: Other threads continue to execute
6300 * Background Execution:: Running your program asynchronously
6301 * Thread-Specific Breakpoints:: Controlling breakpoints
6302 * Interrupted System Calls:: GDB may interfere with system calls
6303 * Observer Mode:: GDB does not alter program behavior
6307 @subsection All-Stop Mode
6309 @cindex all-stop mode
6311 In all-stop mode, whenever your program stops under @value{GDBN} for any reason,
6312 @emph{all} threads of execution stop, not just the current thread. This
6313 allows you to examine the overall state of the program, including
6314 switching between threads, without worrying that things may change
6317 Conversely, whenever you restart the program, @emph{all} threads start
6318 executing. @emph{This is true even when single-stepping} with commands
6319 like @code{step} or @code{next}.
6321 In particular, @value{GDBN} cannot single-step all threads in lockstep.
6322 Since thread scheduling is up to your debugging target's operating
6323 system (not controlled by @value{GDBN}), other threads may
6324 execute more than one statement while the current thread completes a
6325 single step. Moreover, in general other threads stop in the middle of a
6326 statement, rather than at a clean statement boundary, when the program
6329 You might even find your program stopped in another thread after
6330 continuing or even single-stepping. This happens whenever some other
6331 thread runs into a breakpoint, a signal, or an exception before the
6332 first thread completes whatever you requested.
6334 @cindex automatic thread selection
6335 @cindex switching threads automatically
6336 @cindex threads, automatic switching
6337 Whenever @value{GDBN} stops your program, due to a breakpoint or a
6338 signal, it automatically selects the thread where that breakpoint or
6339 signal happened. @value{GDBN} alerts you to the context switch with a
6340 message such as @samp{[Switching to Thread @var{n}]} to identify the
6343 On some OSes, you can modify @value{GDBN}'s default behavior by
6344 locking the OS scheduler to allow only a single thread to run.
6347 @item set scheduler-locking @var{mode}
6348 @cindex scheduler locking mode
6349 @cindex lock scheduler
6350 Set the scheduler locking mode. It applies to normal execution,
6351 record mode, and replay mode. If it is @code{off}, then there is no
6352 locking and any thread may run at any time. If @code{on}, then only
6353 the current thread may run when the inferior is resumed. The
6354 @code{step} mode optimizes for single-stepping; it prevents other
6355 threads from preempting the current thread while you are stepping, so
6356 that the focus of debugging does not change unexpectedly. Other
6357 threads never get a chance to run when you step, and they are
6358 completely free to run when you use commands like @samp{continue},
6359 @samp{until}, or @samp{finish}. However, unless another thread hits a
6360 breakpoint during its timeslice, @value{GDBN} does not change the
6361 current thread away from the thread that you are debugging. The
6362 @code{replay} mode behaves like @code{off} in record mode and like
6363 @code{on} in replay mode.
6365 @item show scheduler-locking
6366 Display the current scheduler locking mode.
6369 @cindex resume threads of multiple processes simultaneously
6370 By default, when you issue one of the execution commands such as
6371 @code{continue}, @code{next} or @code{step}, @value{GDBN} allows only
6372 threads of the current inferior to run. For example, if @value{GDBN}
6373 is attached to two inferiors, each with two threads, the
6374 @code{continue} command resumes only the two threads of the current
6375 inferior. This is useful, for example, when you debug a program that
6376 forks and you want to hold the parent stopped (so that, for instance,
6377 it doesn't run to exit), while you debug the child. In other
6378 situations, you may not be interested in inspecting the current state
6379 of any of the processes @value{GDBN} is attached to, and you may want
6380 to resume them all until some breakpoint is hit. In the latter case,
6381 you can instruct @value{GDBN} to allow all threads of all the
6382 inferiors to run with the @w{@code{set schedule-multiple}} command.
6385 @kindex set schedule-multiple
6386 @item set schedule-multiple
6387 Set the mode for allowing threads of multiple processes to be resumed
6388 when an execution command is issued. When @code{on}, all threads of
6389 all processes are allowed to run. When @code{off}, only the threads
6390 of the current process are resumed. The default is @code{off}. The
6391 @code{scheduler-locking} mode takes precedence when set to @code{on},
6392 or while you are stepping and set to @code{step}.
6394 @item show schedule-multiple
6395 Display the current mode for resuming the execution of threads of
6400 @subsection Non-Stop Mode
6402 @cindex non-stop mode
6404 @c This section is really only a place-holder, and needs to be expanded
6405 @c with more details.
6407 For some multi-threaded targets, @value{GDBN} supports an optional
6408 mode of operation in which you can examine stopped program threads in
6409 the debugger while other threads continue to execute freely. This
6410 minimizes intrusion when debugging live systems, such as programs
6411 where some threads have real-time constraints or must continue to
6412 respond to external events. This is referred to as @dfn{non-stop} mode.
6414 In non-stop mode, when a thread stops to report a debugging event,
6415 @emph{only} that thread is stopped; @value{GDBN} does not stop other
6416 threads as well, in contrast to the all-stop mode behavior. Additionally,
6417 execution commands such as @code{continue} and @code{step} apply by default
6418 only to the current thread in non-stop mode, rather than all threads as
6419 in all-stop mode. This allows you to control threads explicitly in
6420 ways that are not possible in all-stop mode --- for example, stepping
6421 one thread while allowing others to run freely, stepping
6422 one thread while holding all others stopped, or stepping several threads
6423 independently and simultaneously.
6425 To enter non-stop mode, use this sequence of commands before you run
6426 or attach to your program:
6429 # If using the CLI, pagination breaks non-stop.
6432 # Finally, turn it on!
6436 You can use these commands to manipulate the non-stop mode setting:
6439 @kindex set non-stop
6440 @item set non-stop on
6441 Enable selection of non-stop mode.
6442 @item set non-stop off
6443 Disable selection of non-stop mode.
6444 @kindex show non-stop
6446 Show the current non-stop enablement setting.
6449 Note these commands only reflect whether non-stop mode is enabled,
6450 not whether the currently-executing program is being run in non-stop mode.
6451 In particular, the @code{set non-stop} preference is only consulted when
6452 @value{GDBN} starts or connects to the target program, and it is generally
6453 not possible to switch modes once debugging has started. Furthermore,
6454 since not all targets support non-stop mode, even when you have enabled
6455 non-stop mode, @value{GDBN} may still fall back to all-stop operation by
6458 In non-stop mode, all execution commands apply only to the current thread
6459 by default. That is, @code{continue} only continues one thread.
6460 To continue all threads, issue @code{continue -a} or @code{c -a}.
6462 You can use @value{GDBN}'s background execution commands
6463 (@pxref{Background Execution}) to run some threads in the background
6464 while you continue to examine or step others from @value{GDBN}.
6465 The MI execution commands (@pxref{GDB/MI Program Execution}) are
6466 always executed asynchronously in non-stop mode.
6468 Suspending execution is done with the @code{interrupt} command when
6469 running in the background, or @kbd{Ctrl-c} during foreground execution.
6470 In all-stop mode, this stops the whole process;
6471 but in non-stop mode the interrupt applies only to the current thread.
6472 To stop the whole program, use @code{interrupt -a}.
6474 Other execution commands do not currently support the @code{-a} option.
6476 In non-stop mode, when a thread stops, @value{GDBN} doesn't automatically make
6477 that thread current, as it does in all-stop mode. This is because the
6478 thread stop notifications are asynchronous with respect to @value{GDBN}'s
6479 command interpreter, and it would be confusing if @value{GDBN} unexpectedly
6480 changed to a different thread just as you entered a command to operate on the
6481 previously current thread.
6483 @node Background Execution
6484 @subsection Background Execution
6486 @cindex foreground execution
6487 @cindex background execution
6488 @cindex asynchronous execution
6489 @cindex execution, foreground, background and asynchronous
6491 @value{GDBN}'s execution commands have two variants: the normal
6492 foreground (synchronous) behavior, and a background
6493 (asynchronous) behavior. In foreground execution, @value{GDBN} waits for
6494 the program to report that some thread has stopped before prompting for
6495 another command. In background execution, @value{GDBN} immediately gives
6496 a command prompt so that you can issue other commands while your program runs.
6498 If the target doesn't support async mode, @value{GDBN} issues an error
6499 message if you attempt to use the background execution commands.
6501 @cindex @code{&}, background execution of commands
6502 To specify background execution, add a @code{&} to the command. For example,
6503 the background form of the @code{continue} command is @code{continue&}, or
6504 just @code{c&}. The execution commands that accept background execution
6510 @xref{Starting, , Starting your Program}.
6514 @xref{Attach, , Debugging an Already-running Process}.
6518 @xref{Continuing and Stepping, step}.
6522 @xref{Continuing and Stepping, stepi}.
6526 @xref{Continuing and Stepping, next}.
6530 @xref{Continuing and Stepping, nexti}.
6534 @xref{Continuing and Stepping, continue}.
6538 @xref{Continuing and Stepping, finish}.
6542 @xref{Continuing and Stepping, until}.
6546 Background execution is especially useful in conjunction with non-stop
6547 mode for debugging programs with multiple threads; see @ref{Non-Stop Mode}.
6548 However, you can also use these commands in the normal all-stop mode with
6549 the restriction that you cannot issue another execution command until the
6550 previous one finishes. Examples of commands that are valid in all-stop
6551 mode while the program is running include @code{help} and @code{info break}.
6553 You can interrupt your program while it is running in the background by
6554 using the @code{interrupt} command.
6561 Suspend execution of the running program. In all-stop mode,
6562 @code{interrupt} stops the whole process, but in non-stop mode, it stops
6563 only the current thread. To stop the whole program in non-stop mode,
6564 use @code{interrupt -a}.
6567 @node Thread-Specific Breakpoints
6568 @subsection Thread-Specific Breakpoints
6570 When your program has multiple threads (@pxref{Threads,, Debugging
6571 Programs with Multiple Threads}), you can choose whether to set
6572 breakpoints on all threads, or on a particular thread.
6575 @cindex breakpoints and threads
6576 @cindex thread breakpoints
6577 @kindex break @dots{} thread @var{thread-id}
6578 @item break @var{location} thread @var{thread-id}
6579 @itemx break @var{location} thread @var{thread-id} if @dots{}
6580 @var{location} specifies source lines; there are several ways of
6581 writing them (@pxref{Specify Location}), but the effect is always to
6582 specify some source line.
6584 Use the qualifier @samp{thread @var{thread-id}} with a breakpoint command
6585 to specify that you only want @value{GDBN} to stop the program when a
6586 particular thread reaches this breakpoint. The @var{thread-id} specifier
6587 is one of the thread identifiers assigned by @value{GDBN}, shown
6588 in the first column of the @samp{info threads} display.
6590 If you do not specify @samp{thread @var{thread-id}} when you set a
6591 breakpoint, the breakpoint applies to @emph{all} threads of your
6594 You can use the @code{thread} qualifier on conditional breakpoints as
6595 well; in this case, place @samp{thread @var{thread-id}} before or
6596 after the breakpoint condition, like this:
6599 (@value{GDBP}) break frik.c:13 thread 28 if bartab > lim
6604 Thread-specific breakpoints are automatically deleted when
6605 @value{GDBN} detects the corresponding thread is no longer in the
6606 thread list. For example:
6610 Thread-specific breakpoint 3 deleted - thread 28 no longer in the thread list.
6613 There are several ways for a thread to disappear, such as a regular
6614 thread exit, but also when you detach from the process with the
6615 @code{detach} command (@pxref{Attach, ,Debugging an Already-running
6616 Process}), or if @value{GDBN} loses the remote connection
6617 (@pxref{Remote Debugging}), etc. Note that with some targets,
6618 @value{GDBN} is only able to detect a thread has exited when the user
6619 explictly asks for the thread list with the @code{info threads}
6622 @node Interrupted System Calls
6623 @subsection Interrupted System Calls
6625 @cindex thread breakpoints and system calls
6626 @cindex system calls and thread breakpoints
6627 @cindex premature return from system calls
6628 There is an unfortunate side effect when using @value{GDBN} to debug
6629 multi-threaded programs. If one thread stops for a
6630 breakpoint, or for some other reason, and another thread is blocked in a
6631 system call, then the system call may return prematurely. This is a
6632 consequence of the interaction between multiple threads and the signals
6633 that @value{GDBN} uses to implement breakpoints and other events that
6636 To handle this problem, your program should check the return value of
6637 each system call and react appropriately. This is good programming
6640 For example, do not write code like this:
6646 The call to @code{sleep} will return early if a different thread stops
6647 at a breakpoint or for some other reason.
6649 Instead, write this:
6654 unslept = sleep (unslept);
6657 A system call is allowed to return early, so the system is still
6658 conforming to its specification. But @value{GDBN} does cause your
6659 multi-threaded program to behave differently than it would without
6662 Also, @value{GDBN} uses internal breakpoints in the thread library to
6663 monitor certain events such as thread creation and thread destruction.
6664 When such an event happens, a system call in another thread may return
6665 prematurely, even though your program does not appear to stop.
6668 @subsection Observer Mode
6670 If you want to build on non-stop mode and observe program behavior
6671 without any chance of disruption by @value{GDBN}, you can set
6672 variables to disable all of the debugger's attempts to modify state,
6673 whether by writing memory, inserting breakpoints, etc. These operate
6674 at a low level, intercepting operations from all commands.
6676 When all of these are set to @code{off}, then @value{GDBN} is said to
6677 be @dfn{observer mode}. As a convenience, the variable
6678 @code{observer} can be set to disable these, plus enable non-stop
6681 Note that @value{GDBN} will not prevent you from making nonsensical
6682 combinations of these settings. For instance, if you have enabled
6683 @code{may-insert-breakpoints} but disabled @code{may-write-memory},
6684 then breakpoints that work by writing trap instructions into the code
6685 stream will still not be able to be placed.
6690 @item set observer on
6691 @itemx set observer off
6692 When set to @code{on}, this disables all the permission variables
6693 below (except for @code{insert-fast-tracepoints}), plus enables
6694 non-stop debugging. Setting this to @code{off} switches back to
6695 normal debugging, though remaining in non-stop mode.
6698 Show whether observer mode is on or off.
6700 @kindex may-write-registers
6701 @item set may-write-registers on
6702 @itemx set may-write-registers off
6703 This controls whether @value{GDBN} will attempt to alter the values of
6704 registers, such as with assignment expressions in @code{print}, or the
6705 @code{jump} command. It defaults to @code{on}.
6707 @item show may-write-registers
6708 Show the current permission to write registers.
6710 @kindex may-write-memory
6711 @item set may-write-memory on
6712 @itemx set may-write-memory off
6713 This controls whether @value{GDBN} will attempt to alter the contents
6714 of memory, such as with assignment expressions in @code{print}. It
6715 defaults to @code{on}.
6717 @item show may-write-memory
6718 Show the current permission to write memory.
6720 @kindex may-insert-breakpoints
6721 @item set may-insert-breakpoints on
6722 @itemx set may-insert-breakpoints off
6723 This controls whether @value{GDBN} will attempt to insert breakpoints.
6724 This affects all breakpoints, including internal breakpoints defined
6725 by @value{GDBN}. It defaults to @code{on}.
6727 @item show may-insert-breakpoints
6728 Show the current permission to insert breakpoints.
6730 @kindex may-insert-tracepoints
6731 @item set may-insert-tracepoints on
6732 @itemx set may-insert-tracepoints off
6733 This controls whether @value{GDBN} will attempt to insert (regular)
6734 tracepoints at the beginning of a tracing experiment. It affects only
6735 non-fast tracepoints, fast tracepoints being under the control of
6736 @code{may-insert-fast-tracepoints}. It defaults to @code{on}.
6738 @item show may-insert-tracepoints
6739 Show the current permission to insert tracepoints.
6741 @kindex may-insert-fast-tracepoints
6742 @item set may-insert-fast-tracepoints on
6743 @itemx set may-insert-fast-tracepoints off
6744 This controls whether @value{GDBN} will attempt to insert fast
6745 tracepoints at the beginning of a tracing experiment. It affects only
6746 fast tracepoints, regular (non-fast) tracepoints being under the
6747 control of @code{may-insert-tracepoints}. It defaults to @code{on}.
6749 @item show may-insert-fast-tracepoints
6750 Show the current permission to insert fast tracepoints.
6752 @kindex may-interrupt
6753 @item set may-interrupt on
6754 @itemx set may-interrupt off
6755 This controls whether @value{GDBN} will attempt to interrupt or stop
6756 program execution. When this variable is @code{off}, the
6757 @code{interrupt} command will have no effect, nor will
6758 @kbd{Ctrl-c}. It defaults to @code{on}.
6760 @item show may-interrupt
6761 Show the current permission to interrupt or stop the program.
6765 @node Reverse Execution
6766 @chapter Running programs backward
6767 @cindex reverse execution
6768 @cindex running programs backward
6770 When you are debugging a program, it is not unusual to realize that
6771 you have gone too far, and some event of interest has already happened.
6772 If the target environment supports it, @value{GDBN} can allow you to
6773 ``rewind'' the program by running it backward.
6775 A target environment that supports reverse execution should be able
6776 to ``undo'' the changes in machine state that have taken place as the
6777 program was executing normally. Variables, registers etc.@: should
6778 revert to their previous values. Obviously this requires a great
6779 deal of sophistication on the part of the target environment; not
6780 all target environments can support reverse execution.
6782 When a program is executed in reverse, the instructions that
6783 have most recently been executed are ``un-executed'', in reverse
6784 order. The program counter runs backward, following the previous
6785 thread of execution in reverse. As each instruction is ``un-executed'',
6786 the values of memory and/or registers that were changed by that
6787 instruction are reverted to their previous states. After executing
6788 a piece of source code in reverse, all side effects of that code
6789 should be ``undone'', and all variables should be returned to their
6790 prior values@footnote{
6791 Note that some side effects are easier to undo than others. For instance,
6792 memory and registers are relatively easy, but device I/O is hard. Some
6793 targets may be able undo things like device I/O, and some may not.
6795 The contract between @value{GDBN} and the reverse executing target
6796 requires only that the target do something reasonable when
6797 @value{GDBN} tells it to execute backwards, and then report the
6798 results back to @value{GDBN}. Whatever the target reports back to
6799 @value{GDBN}, @value{GDBN} will report back to the user. @value{GDBN}
6800 assumes that the memory and registers that the target reports are in a
6801 consistant state, but @value{GDBN} accepts whatever it is given.
6804 On some platforms, @value{GDBN} has built-in support for reverse
6805 execution, activated with the @code{record} or @code{record btrace}
6806 commands. @xref{Process Record and Replay}. Some remote targets,
6807 typically full system emulators, support reverse execution directly
6808 without requiring any special command.
6810 If you are debugging in a target environment that supports
6811 reverse execution, @value{GDBN} provides the following commands.
6814 @kindex reverse-continue
6815 @kindex rc @r{(@code{reverse-continue})}
6816 @item reverse-continue @r{[}@var{ignore-count}@r{]}
6817 @itemx rc @r{[}@var{ignore-count}@r{]}
6818 Beginning at the point where your program last stopped, start executing
6819 in reverse. Reverse execution will stop for breakpoints and synchronous
6820 exceptions (signals), just like normal execution. Behavior of
6821 asynchronous signals depends on the target environment.
6823 @kindex reverse-step
6824 @kindex rs @r{(@code{step})}
6825 @item reverse-step @r{[}@var{count}@r{]}
6826 Run the program backward until control reaches the start of a
6827 different source line; then stop it, and return control to @value{GDBN}.
6829 Like the @code{step} command, @code{reverse-step} will only stop
6830 at the beginning of a source line. It ``un-executes'' the previously
6831 executed source line. If the previous source line included calls to
6832 debuggable functions, @code{reverse-step} will step (backward) into
6833 the called function, stopping at the beginning of the @emph{last}
6834 statement in the called function (typically a return statement).
6836 Also, as with the @code{step} command, if non-debuggable functions are
6837 called, @code{reverse-step} will run thru them backward without stopping.
6839 @kindex reverse-stepi
6840 @kindex rsi @r{(@code{reverse-stepi})}
6841 @item reverse-stepi @r{[}@var{count}@r{]}
6842 Reverse-execute one machine instruction. Note that the instruction
6843 to be reverse-executed is @emph{not} the one pointed to by the program
6844 counter, but the instruction executed prior to that one. For instance,
6845 if the last instruction was a jump, @code{reverse-stepi} will take you
6846 back from the destination of the jump to the jump instruction itself.
6848 @kindex reverse-next
6849 @kindex rn @r{(@code{reverse-next})}
6850 @item reverse-next @r{[}@var{count}@r{]}
6851 Run backward to the beginning of the previous line executed in
6852 the current (innermost) stack frame. If the line contains function
6853 calls, they will be ``un-executed'' without stopping. Starting from
6854 the first line of a function, @code{reverse-next} will take you back
6855 to the caller of that function, @emph{before} the function was called,
6856 just as the normal @code{next} command would take you from the last
6857 line of a function back to its return to its caller
6858 @footnote{Unless the code is too heavily optimized.}.
6860 @kindex reverse-nexti
6861 @kindex rni @r{(@code{reverse-nexti})}
6862 @item reverse-nexti @r{[}@var{count}@r{]}
6863 Like @code{nexti}, @code{reverse-nexti} executes a single instruction
6864 in reverse, except that called functions are ``un-executed'' atomically.
6865 That is, if the previously executed instruction was a return from
6866 another function, @code{reverse-nexti} will continue to execute
6867 in reverse until the call to that function (from the current stack
6870 @kindex reverse-finish
6871 @item reverse-finish
6872 Just as the @code{finish} command takes you to the point where the
6873 current function returns, @code{reverse-finish} takes you to the point
6874 where it was called. Instead of ending up at the end of the current
6875 function invocation, you end up at the beginning.
6877 @kindex set exec-direction
6878 @item set exec-direction
6879 Set the direction of target execution.
6880 @item set exec-direction reverse
6881 @cindex execute forward or backward in time
6882 @value{GDBN} will perform all execution commands in reverse, until the
6883 exec-direction mode is changed to ``forward''. Affected commands include
6884 @code{step, stepi, next, nexti, continue, and finish}. The @code{return}
6885 command cannot be used in reverse mode.
6886 @item set exec-direction forward
6887 @value{GDBN} will perform all execution commands in the normal fashion.
6888 This is the default.
6892 @node Process Record and Replay
6893 @chapter Recording Inferior's Execution and Replaying It
6894 @cindex process record and replay
6895 @cindex recording inferior's execution and replaying it
6897 On some platforms, @value{GDBN} provides a special @dfn{process record
6898 and replay} target that can record a log of the process execution, and
6899 replay it later with both forward and reverse execution commands.
6902 When this target is in use, if the execution log includes the record
6903 for the next instruction, @value{GDBN} will debug in @dfn{replay
6904 mode}. In the replay mode, the inferior does not really execute code
6905 instructions. Instead, all the events that normally happen during
6906 code execution are taken from the execution log. While code is not
6907 really executed in replay mode, the values of registers (including the
6908 program counter register) and the memory of the inferior are still
6909 changed as they normally would. Their contents are taken from the
6913 If the record for the next instruction is not in the execution log,
6914 @value{GDBN} will debug in @dfn{record mode}. In this mode, the
6915 inferior executes normally, and @value{GDBN} records the execution log
6918 The process record and replay target supports reverse execution
6919 (@pxref{Reverse Execution}), even if the platform on which the
6920 inferior runs does not. However, the reverse execution is limited in
6921 this case by the range of the instructions recorded in the execution
6922 log. In other words, reverse execution on platforms that don't
6923 support it directly can only be done in the replay mode.
6925 When debugging in the reverse direction, @value{GDBN} will work in
6926 replay mode as long as the execution log includes the record for the
6927 previous instruction; otherwise, it will work in record mode, if the
6928 platform supports reverse execution, or stop if not.
6930 Currently, process record and replay is supported on ARM, Aarch64,
6931 Moxie, PowerPC, PowerPC64, S/390, and x86 (i386/amd64) running
6932 GNU/Linux. Process record and replay can be used both when native
6933 debugging, and when remote debugging via @code{gdbserver}.
6935 For architecture environments that support process record and replay,
6936 @value{GDBN} provides the following commands:
6939 @kindex target record
6940 @kindex target record-full
6941 @kindex target record-btrace
6944 @kindex record btrace
6945 @kindex record btrace bts
6946 @kindex record btrace pt
6952 @kindex rec btrace bts
6953 @kindex rec btrace pt
6956 @item record @var{method}
6957 This command starts the process record and replay target. The
6958 recording method can be specified as parameter. Without a parameter
6959 the command uses the @code{full} recording method. The following
6960 recording methods are available:
6964 Full record/replay recording using @value{GDBN}'s software record and
6965 replay implementation. This method allows replaying and reverse
6968 @item btrace @var{format}
6969 Hardware-supported instruction recording, supported on Intel
6970 processors. This method does not record data. Further, the data is
6971 collected in a ring buffer so old data will be overwritten when the
6972 buffer is full. It allows limited reverse execution. Variables and
6973 registers are not available during reverse execution. In remote
6974 debugging, recording continues on disconnect. Recorded data can be
6975 inspected after reconnecting. The recording may be stopped using
6978 The recording format can be specified as parameter. Without a parameter
6979 the command chooses the recording format. The following recording
6980 formats are available:
6984 @cindex branch trace store
6985 Use the @dfn{Branch Trace Store} (@acronym{BTS}) recording format. In
6986 this format, the processor stores a from/to record for each executed
6987 branch in the btrace ring buffer.
6990 @cindex Intel Processor Trace
6991 Use the @dfn{Intel Processor Trace} recording format. In this
6992 format, the processor stores the execution trace in a compressed form
6993 that is afterwards decoded by @value{GDBN}.
6995 The trace can be recorded with very low overhead. The compressed
6996 trace format also allows small trace buffers to already contain a big
6997 number of instructions compared to @acronym{BTS}.
6999 Decoding the recorded execution trace, on the other hand, is more
7000 expensive than decoding @acronym{BTS} trace. This is mostly due to the
7001 increased number of instructions to process. You should increase the
7002 buffer-size with care.
7005 Not all recording formats may be available on all processors.
7008 The process record and replay target can only debug a process that is
7009 already running. Therefore, you need first to start the process with
7010 the @kbd{run} or @kbd{start} commands, and then start the recording
7011 with the @kbd{record @var{method}} command.
7013 @cindex displaced stepping, and process record and replay
7014 Displaced stepping (@pxref{Maintenance Commands,, displaced stepping})
7015 will be automatically disabled when process record and replay target
7016 is started. That's because the process record and replay target
7017 doesn't support displaced stepping.
7019 @cindex non-stop mode, and process record and replay
7020 @cindex asynchronous execution, and process record and replay
7021 If the inferior is in the non-stop mode (@pxref{Non-Stop Mode}) or in
7022 the asynchronous execution mode (@pxref{Background Execution}), not
7023 all recording methods are available. The @code{full} recording method
7024 does not support these two modes.
7029 Stop the process record and replay target. When process record and
7030 replay target stops, the entire execution log will be deleted and the
7031 inferior will either be terminated, or will remain in its final state.
7033 When you stop the process record and replay target in record mode (at
7034 the end of the execution log), the inferior will be stopped at the
7035 next instruction that would have been recorded. In other words, if
7036 you record for a while and then stop recording, the inferior process
7037 will be left in the same state as if the recording never happened.
7039 On the other hand, if the process record and replay target is stopped
7040 while in replay mode (that is, not at the end of the execution log,
7041 but at some earlier point), the inferior process will become ``live''
7042 at that earlier state, and it will then be possible to continue the
7043 usual ``live'' debugging of the process from that state.
7045 When the inferior process exits, or @value{GDBN} detaches from it,
7046 process record and replay target will automatically stop itself.
7050 Go to a specific location in the execution log. There are several
7051 ways to specify the location to go to:
7054 @item record goto begin
7055 @itemx record goto start
7056 Go to the beginning of the execution log.
7058 @item record goto end
7059 Go to the end of the execution log.
7061 @item record goto @var{n}
7062 Go to instruction number @var{n} in the execution log.
7066 @item record save @var{filename}
7067 Save the execution log to a file @file{@var{filename}}.
7068 Default filename is @file{gdb_record.@var{process_id}}, where
7069 @var{process_id} is the process ID of the inferior.
7071 This command may not be available for all recording methods.
7073 @kindex record restore
7074 @item record restore @var{filename}
7075 Restore the execution log from a file @file{@var{filename}}.
7076 File must have been created with @code{record save}.
7078 @kindex set record full
7079 @item set record full insn-number-max @var{limit}
7080 @itemx set record full insn-number-max unlimited
7081 Set the limit of instructions to be recorded for the @code{full}
7082 recording method. Default value is 200000.
7084 If @var{limit} is a positive number, then @value{GDBN} will start
7085 deleting instructions from the log once the number of the record
7086 instructions becomes greater than @var{limit}. For every new recorded
7087 instruction, @value{GDBN} will delete the earliest recorded
7088 instruction to keep the number of recorded instructions at the limit.
7089 (Since deleting recorded instructions loses information, @value{GDBN}
7090 lets you control what happens when the limit is reached, by means of
7091 the @code{stop-at-limit} option, described below.)
7093 If @var{limit} is @code{unlimited} or zero, @value{GDBN} will never
7094 delete recorded instructions from the execution log. The number of
7095 recorded instructions is limited only by the available memory.
7097 @kindex show record full
7098 @item show record full insn-number-max
7099 Show the limit of instructions to be recorded with the @code{full}
7102 @item set record full stop-at-limit
7103 Control the behavior of the @code{full} recording method when the
7104 number of recorded instructions reaches the limit. If ON (the
7105 default), @value{GDBN} will stop when the limit is reached for the
7106 first time and ask you whether you want to stop the inferior or
7107 continue running it and recording the execution log. If you decide
7108 to continue recording, each new recorded instruction will cause the
7109 oldest one to be deleted.
7111 If this option is OFF, @value{GDBN} will automatically delete the
7112 oldest record to make room for each new one, without asking.
7114 @item show record full stop-at-limit
7115 Show the current setting of @code{stop-at-limit}.
7117 @item set record full memory-query
7118 Control the behavior when @value{GDBN} is unable to record memory
7119 changes caused by an instruction for the @code{full} recording method.
7120 If ON, @value{GDBN} will query whether to stop the inferior in that
7123 If this option is OFF (the default), @value{GDBN} will automatically
7124 ignore the effect of such instructions on memory. Later, when
7125 @value{GDBN} replays this execution log, it will mark the log of this
7126 instruction as not accessible, and it will not affect the replay
7129 @item show record full memory-query
7130 Show the current setting of @code{memory-query}.
7132 @kindex set record btrace
7133 The @code{btrace} record target does not trace data. As a
7134 convenience, when replaying, @value{GDBN} reads read-only memory off
7135 the live program directly, assuming that the addresses of the
7136 read-only areas don't change. This for example makes it possible to
7137 disassemble code while replaying, but not to print variables.
7138 In some cases, being able to inspect variables might be useful.
7139 You can use the following command for that:
7141 @item set record btrace replay-memory-access
7142 Control the behavior of the @code{btrace} recording method when
7143 accessing memory during replay. If @code{read-only} (the default),
7144 @value{GDBN} will only allow accesses to read-only memory.
7145 If @code{read-write}, @value{GDBN} will allow accesses to read-only
7146 and to read-write memory. Beware that the accessed memory corresponds
7147 to the live target and not necessarily to the current replay
7150 @item set record btrace cpu @var{identifier}
7151 Set the processor to be used for enabling workarounds for processor
7152 errata when decoding the trace.
7154 Processor errata are defects in processor operation, caused by its
7155 design or manufacture. They can cause a trace not to match the
7156 specification. This, in turn, may cause trace decode to fail.
7157 @value{GDBN} can detect erroneous trace packets and correct them, thus
7158 avoiding the decoding failures. These corrections are known as
7159 @dfn{errata workarounds}, and are enabled based on the processor on
7160 which the trace was recorded.
7162 By default, @value{GDBN} attempts to detect the processor
7163 automatically, and apply the necessary workarounds for it. However,
7164 you may need to specify the processor if @value{GDBN} does not yet
7165 support it. This command allows you to do that, and also allows to
7166 disable the workarounds.
7168 The argument @var{identifier} identifies the @sc{cpu} and is of the
7169 form: @code{@var{vendor}:@var{procesor identifier}}. In addition,
7170 there are two special identifiers, @code{none} and @code{auto}
7173 The following vendor identifiers and corresponding processor
7174 identifiers are currently supported:
7176 @multitable @columnfractions .1 .9
7179 @tab @var{family}/@var{model}[/@var{stepping}]
7183 On GNU/Linux systems, the processor @var{family}, @var{model}, and
7184 @var{stepping} can be obtained from @code{/proc/cpuinfo}.
7186 If @var{identifier} is @code{auto}, enable errata workarounds for the
7187 processor on which the trace was recorded. If @var{identifier} is
7188 @code{none}, errata workarounds are disabled.
7190 For example, when using an old @value{GDBN} on a new system, decode
7191 may fail because @value{GDBN} does not support the new processor. It
7192 often suffices to specify an older processor that @value{GDBN}
7197 Active record target: record-btrace
7198 Recording format: Intel Processor Trace.
7200 Failed to configure the Intel Processor Trace decoder: unknown cpu.
7201 (gdb) set record btrace cpu intel:6/158
7203 Active record target: record-btrace
7204 Recording format: Intel Processor Trace.
7206 Recorded 84872 instructions in 3189 functions (0 gaps) for thread 1 (...).
7209 @kindex show record btrace
7210 @item show record btrace replay-memory-access
7211 Show the current setting of @code{replay-memory-access}.
7213 @item show record btrace cpu
7214 Show the processor to be used for enabling trace decode errata
7217 @kindex set record btrace bts
7218 @item set record btrace bts buffer-size @var{size}
7219 @itemx set record btrace bts buffer-size unlimited
7220 Set the requested ring buffer size for branch tracing in @acronym{BTS}
7221 format. Default is 64KB.
7223 If @var{size} is a positive number, then @value{GDBN} will try to
7224 allocate a buffer of at least @var{size} bytes for each new thread
7225 that uses the btrace recording method and the @acronym{BTS} format.
7226 The actually obtained buffer size may differ from the requested
7227 @var{size}. Use the @code{info record} command to see the actual
7228 buffer size for each thread that uses the btrace recording method and
7229 the @acronym{BTS} format.
7231 If @var{limit} is @code{unlimited} or zero, @value{GDBN} will try to
7232 allocate a buffer of 4MB.
7234 Bigger buffers mean longer traces. On the other hand, @value{GDBN} will
7235 also need longer to process the branch trace data before it can be used.
7237 @item show record btrace bts buffer-size @var{size}
7238 Show the current setting of the requested ring buffer size for branch
7239 tracing in @acronym{BTS} format.
7241 @kindex set record btrace pt
7242 @item set record btrace pt buffer-size @var{size}
7243 @itemx set record btrace pt buffer-size unlimited
7244 Set the requested ring buffer size for branch tracing in Intel
7245 Processor Trace format. Default is 16KB.
7247 If @var{size} is a positive number, then @value{GDBN} will try to
7248 allocate a buffer of at least @var{size} bytes for each new thread
7249 that uses the btrace recording method and the Intel Processor Trace
7250 format. The actually obtained buffer size may differ from the
7251 requested @var{size}. Use the @code{info record} command to see the
7252 actual buffer size for each thread.
7254 If @var{limit} is @code{unlimited} or zero, @value{GDBN} will try to
7255 allocate a buffer of 4MB.
7257 Bigger buffers mean longer traces. On the other hand, @value{GDBN} will
7258 also need longer to process the branch trace data before it can be used.
7260 @item show record btrace pt buffer-size @var{size}
7261 Show the current setting of the requested ring buffer size for branch
7262 tracing in Intel Processor Trace format.
7266 Show various statistics about the recording depending on the recording
7271 For the @code{full} recording method, it shows the state of process
7272 record and its in-memory execution log buffer, including:
7276 Whether in record mode or replay mode.
7278 Lowest recorded instruction number (counting from when the current execution log started recording instructions).
7280 Highest recorded instruction number.
7282 Current instruction about to be replayed (if in replay mode).
7284 Number of instructions contained in the execution log.
7286 Maximum number of instructions that may be contained in the execution log.
7290 For the @code{btrace} recording method, it shows:
7296 Number of instructions that have been recorded.
7298 Number of blocks of sequential control-flow formed by the recorded
7301 Whether in record mode or replay mode.
7304 For the @code{bts} recording format, it also shows:
7307 Size of the perf ring buffer.
7310 For the @code{pt} recording format, it also shows:
7313 Size of the perf ring buffer.
7317 @kindex record delete
7320 When record target runs in replay mode (``in the past''), delete the
7321 subsequent execution log and begin to record a new execution log starting
7322 from the current address. This means you will abandon the previously
7323 recorded ``future'' and begin recording a new ``future''.
7325 @kindex record instruction-history
7326 @kindex rec instruction-history
7327 @item record instruction-history
7328 Disassembles instructions from the recorded execution log. By
7329 default, ten instructions are disassembled. This can be changed using
7330 the @code{set record instruction-history-size} command. Instructions
7331 are printed in execution order.
7333 It can also print mixed source+disassembly if you specify the the
7334 @code{/m} or @code{/s} modifier, and print the raw instructions in hex
7335 as well as in symbolic form by specifying the @code{/r} modifier.
7337 The current position marker is printed for the instruction at the
7338 current program counter value. This instruction can appear multiple
7339 times in the trace and the current position marker will be printed
7340 every time. To omit the current position marker, specify the
7343 To better align the printed instructions when the trace contains
7344 instructions from more than one function, the function name may be
7345 omitted by specifying the @code{/f} modifier.
7347 Speculatively executed instructions are prefixed with @samp{?}. This
7348 feature is not available for all recording formats.
7350 There are several ways to specify what part of the execution log to
7354 @item record instruction-history @var{insn}
7355 Disassembles ten instructions starting from instruction number
7358 @item record instruction-history @var{insn}, +/-@var{n}
7359 Disassembles @var{n} instructions around instruction number
7360 @var{insn}. If @var{n} is preceded with @code{+}, disassembles
7361 @var{n} instructions after instruction number @var{insn}. If
7362 @var{n} is preceded with @code{-}, disassembles @var{n}
7363 instructions before instruction number @var{insn}.
7365 @item record instruction-history
7366 Disassembles ten more instructions after the last disassembly.
7368 @item record instruction-history -
7369 Disassembles ten more instructions before the last disassembly.
7371 @item record instruction-history @var{begin}, @var{end}
7372 Disassembles instructions beginning with instruction number
7373 @var{begin} until instruction number @var{end}. The instruction
7374 number @var{end} is included.
7377 This command may not be available for all recording methods.
7380 @item set record instruction-history-size @var{size}
7381 @itemx set record instruction-history-size unlimited
7382 Define how many instructions to disassemble in the @code{record
7383 instruction-history} command. The default value is 10.
7384 A @var{size} of @code{unlimited} means unlimited instructions.
7387 @item show record instruction-history-size
7388 Show how many instructions to disassemble in the @code{record
7389 instruction-history} command.
7391 @kindex record function-call-history
7392 @kindex rec function-call-history
7393 @item record function-call-history
7394 Prints the execution history at function granularity. It prints one
7395 line for each sequence of instructions that belong to the same
7396 function giving the name of that function, the source lines
7397 for this instruction sequence (if the @code{/l} modifier is
7398 specified), and the instructions numbers that form the sequence (if
7399 the @code{/i} modifier is specified). The function names are indented
7400 to reflect the call stack depth if the @code{/c} modifier is
7401 specified. The @code{/l}, @code{/i}, and @code{/c} modifiers can be
7405 (@value{GDBP}) @b{list 1, 10}
7416 (@value{GDBP}) @b{record function-call-history /ilc}
7417 1 bar inst 1,4 at foo.c:6,8
7418 2 foo inst 5,10 at foo.c:2,3
7419 3 bar inst 11,13 at foo.c:9,10
7422 By default, ten lines are printed. This can be changed using the
7423 @code{set record function-call-history-size} command. Functions are
7424 printed in execution order. There are several ways to specify what
7428 @item record function-call-history @var{func}
7429 Prints ten functions starting from function number @var{func}.
7431 @item record function-call-history @var{func}, +/-@var{n}
7432 Prints @var{n} functions around function number @var{func}. If
7433 @var{n} is preceded with @code{+}, prints @var{n} functions after
7434 function number @var{func}. If @var{n} is preceded with @code{-},
7435 prints @var{n} functions before function number @var{func}.
7437 @item record function-call-history
7438 Prints ten more functions after the last ten-line print.
7440 @item record function-call-history -
7441 Prints ten more functions before the last ten-line print.
7443 @item record function-call-history @var{begin}, @var{end}
7444 Prints functions beginning with function number @var{begin} until
7445 function number @var{end}. The function number @var{end} is included.
7448 This command may not be available for all recording methods.
7450 @item set record function-call-history-size @var{size}
7451 @itemx set record function-call-history-size unlimited
7452 Define how many lines to print in the
7453 @code{record function-call-history} command. The default value is 10.
7454 A size of @code{unlimited} means unlimited lines.
7456 @item show record function-call-history-size
7457 Show how many lines to print in the
7458 @code{record function-call-history} command.
7463 @chapter Examining the Stack
7465 When your program has stopped, the first thing you need to know is where it
7466 stopped and how it got there.
7469 Each time your program performs a function call, information about the call
7471 That information includes the location of the call in your program,
7472 the arguments of the call,
7473 and the local variables of the function being called.
7474 The information is saved in a block of data called a @dfn{stack frame}.
7475 The stack frames are allocated in a region of memory called the @dfn{call
7478 When your program stops, the @value{GDBN} commands for examining the
7479 stack allow you to see all of this information.
7481 @cindex selected frame
7482 One of the stack frames is @dfn{selected} by @value{GDBN} and many
7483 @value{GDBN} commands refer implicitly to the selected frame. In
7484 particular, whenever you ask @value{GDBN} for the value of a variable in
7485 your program, the value is found in the selected frame. There are
7486 special @value{GDBN} commands to select whichever frame you are
7487 interested in. @xref{Selection, ,Selecting a Frame}.
7489 When your program stops, @value{GDBN} automatically selects the
7490 currently executing frame and describes it briefly, similar to the
7491 @code{frame} command (@pxref{Frame Info, ,Information about a Frame}).
7494 * Frames:: Stack frames
7495 * Backtrace:: Backtraces
7496 * Selection:: Selecting a frame
7497 * Frame Info:: Information on a frame
7498 * Frame Apply:: Applying a command to several frames
7499 * Frame Filter Management:: Managing frame filters
7504 @section Stack Frames
7506 @cindex frame, definition
7508 The call stack is divided up into contiguous pieces called @dfn{stack
7509 frames}, or @dfn{frames} for short; each frame is the data associated
7510 with one call to one function. The frame contains the arguments given
7511 to the function, the function's local variables, and the address at
7512 which the function is executing.
7514 @cindex initial frame
7515 @cindex outermost frame
7516 @cindex innermost frame
7517 When your program is started, the stack has only one frame, that of the
7518 function @code{main}. This is called the @dfn{initial} frame or the
7519 @dfn{outermost} frame. Each time a function is called, a new frame is
7520 made. Each time a function returns, the frame for that function invocation
7521 is eliminated. If a function is recursive, there can be many frames for
7522 the same function. The frame for the function in which execution is
7523 actually occurring is called the @dfn{innermost} frame. This is the most
7524 recently created of all the stack frames that still exist.
7526 @cindex frame pointer
7527 Inside your program, stack frames are identified by their addresses. A
7528 stack frame consists of many bytes, each of which has its own address; each
7529 kind of computer has a convention for choosing one byte whose
7530 address serves as the address of the frame. Usually this address is kept
7531 in a register called the @dfn{frame pointer register}
7532 (@pxref{Registers, $fp}) while execution is going on in that frame.
7535 @cindex frame number
7536 @value{GDBN} labels each existing stack frame with a @dfn{level}, a
7537 number that is zero for the innermost frame, one for the frame that
7538 called it, and so on upward. These level numbers give you a way of
7539 designating stack frames in @value{GDBN} commands. The terms
7540 @dfn{frame number} and @dfn{frame level} can be used interchangeably to
7541 describe this number.
7543 @c The -fomit-frame-pointer below perennially causes hbox overflow
7544 @c underflow problems.
7545 @cindex frameless execution
7546 Some compilers provide a way to compile functions so that they operate
7547 without stack frames. (For example, the @value{NGCC} option
7549 @samp{-fomit-frame-pointer}
7551 generates functions without a frame.)
7552 This is occasionally done with heavily used library functions to save
7553 the frame setup time. @value{GDBN} has limited facilities for dealing
7554 with these function invocations. If the innermost function invocation
7555 has no stack frame, @value{GDBN} nevertheless regards it as though
7556 it had a separate frame, which is numbered zero as usual, allowing
7557 correct tracing of the function call chain. However, @value{GDBN} has
7558 no provision for frameless functions elsewhere in the stack.
7564 @cindex call stack traces
7565 A backtrace is a summary of how your program got where it is. It shows one
7566 line per frame, for many frames, starting with the currently executing
7567 frame (frame zero), followed by its caller (frame one), and on up the
7570 @anchor{backtrace-command}
7572 @kindex bt @r{(@code{backtrace})}
7573 To print a backtrace of the entire stack, use the @code{backtrace}
7574 command, or its alias @code{bt}. This command will print one line per
7575 frame for frames in the stack. By default, all stack frames are
7576 printed. You can stop the backtrace at any time by typing the system
7577 interrupt character, normally @kbd{Ctrl-c}.
7580 @item backtrace [@var{args}@dots{}]
7581 @itemx bt [@var{args}@dots{}]
7582 Print the backtrace of the entire stack. The optional @var{args} can
7583 be one of the following:
7588 Print only the innermost @var{n} frames, where @var{n} is a positive
7593 Print only the outermost @var{n} frames, where @var{n} is a positive
7597 Print the values of the local variables also. This can be combined
7598 with a number to limit the number of frames shown.
7601 Do not run Python frame filters on this backtrace. @xref{Frame
7602 Filter API}, for more information. Additionally use @ref{disable
7603 frame-filter all} to turn off all frame filters. This is only
7604 relevant when @value{GDBN} has been configured with @code{Python}
7608 A Python frame filter might decide to ``elide'' some frames. Normally
7609 such elided frames are still printed, but they are indented relative
7610 to the filtered frames that cause them to be elided. The @code{hide}
7611 option causes elided frames to not be printed at all.
7617 The names @code{where} and @code{info stack} (abbreviated @code{info s})
7618 are additional aliases for @code{backtrace}.
7620 @cindex multiple threads, backtrace
7621 In a multi-threaded program, @value{GDBN} by default shows the
7622 backtrace only for the current thread. To display the backtrace for
7623 several or all of the threads, use the command @code{thread apply}
7624 (@pxref{Threads, thread apply}). For example, if you type @kbd{thread
7625 apply all backtrace}, @value{GDBN} will display the backtrace for all
7626 the threads; this is handy when you debug a core dump of a
7627 multi-threaded program.
7629 Each line in the backtrace shows the frame number and the function name.
7630 The program counter value is also shown---unless you use @code{set
7631 print address off}. The backtrace also shows the source file name and
7632 line number, as well as the arguments to the function. The program
7633 counter value is omitted if it is at the beginning of the code for that
7636 Here is an example of a backtrace. It was made with the command
7637 @samp{bt 3}, so it shows the innermost three frames.
7641 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
7643 #1 0x6e38 in expand_macro (sym=0x2b600, data=...) at macro.c:242
7644 #2 0x6840 in expand_token (obs=0x0, t=177664, td=0xf7fffb08)
7646 (More stack frames follow...)
7651 The display for frame zero does not begin with a program counter
7652 value, indicating that your program has stopped at the beginning of the
7653 code for line @code{993} of @code{builtin.c}.
7656 The value of parameter @code{data} in frame 1 has been replaced by
7657 @code{@dots{}}. By default, @value{GDBN} prints the value of a parameter
7658 only if it is a scalar (integer, pointer, enumeration, etc). See command
7659 @kbd{set print frame-arguments} in @ref{Print Settings} for more details
7660 on how to configure the way function parameter values are printed.
7662 @cindex optimized out, in backtrace
7663 @cindex function call arguments, optimized out
7664 If your program was compiled with optimizations, some compilers will
7665 optimize away arguments passed to functions if those arguments are
7666 never used after the call. Such optimizations generate code that
7667 passes arguments through registers, but doesn't store those arguments
7668 in the stack frame. @value{GDBN} has no way of displaying such
7669 arguments in stack frames other than the innermost one. Here's what
7670 such a backtrace might look like:
7674 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
7676 #1 0x6e38 in expand_macro (sym=<optimized out>) at macro.c:242
7677 #2 0x6840 in expand_token (obs=0x0, t=<optimized out>, td=0xf7fffb08)
7679 (More stack frames follow...)
7684 The values of arguments that were not saved in their stack frames are
7685 shown as @samp{<optimized out>}.
7687 If you need to display the values of such optimized-out arguments,
7688 either deduce that from other variables whose values depend on the one
7689 you are interested in, or recompile without optimizations.
7691 @cindex backtrace beyond @code{main} function
7692 @cindex program entry point
7693 @cindex startup code, and backtrace
7694 Most programs have a standard user entry point---a place where system
7695 libraries and startup code transition into user code. For C this is
7696 @code{main}@footnote{
7697 Note that embedded programs (the so-called ``free-standing''
7698 environment) are not required to have a @code{main} function as the
7699 entry point. They could even have multiple entry points.}.
7700 When @value{GDBN} finds the entry function in a backtrace
7701 it will terminate the backtrace, to avoid tracing into highly
7702 system-specific (and generally uninteresting) code.
7704 If you need to examine the startup code, or limit the number of levels
7705 in a backtrace, you can change this behavior:
7708 @item set backtrace past-main
7709 @itemx set backtrace past-main on
7710 @kindex set backtrace
7711 Backtraces will continue past the user entry point.
7713 @item set backtrace past-main off
7714 Backtraces will stop when they encounter the user entry point. This is the
7717 @item show backtrace past-main
7718 @kindex show backtrace
7719 Display the current user entry point backtrace policy.
7721 @item set backtrace past-entry
7722 @itemx set backtrace past-entry on
7723 Backtraces will continue past the internal entry point of an application.
7724 This entry point is encoded by the linker when the application is built,
7725 and is likely before the user entry point @code{main} (or equivalent) is called.
7727 @item set backtrace past-entry off
7728 Backtraces will stop when they encounter the internal entry point of an
7729 application. This is the default.
7731 @item show backtrace past-entry
7732 Display the current internal entry point backtrace policy.
7734 @item set backtrace limit @var{n}
7735 @itemx set backtrace limit 0
7736 @itemx set backtrace limit unlimited
7737 @cindex backtrace limit
7738 Limit the backtrace to @var{n} levels. A value of @code{unlimited}
7739 or zero means unlimited levels.
7741 @item show backtrace limit
7742 Display the current limit on backtrace levels.
7745 You can control how file names are displayed.
7748 @item set filename-display
7749 @itemx set filename-display relative
7750 @cindex filename-display
7751 Display file names relative to the compilation directory. This is the default.
7753 @item set filename-display basename
7754 Display only basename of a filename.
7756 @item set filename-display absolute
7757 Display an absolute filename.
7759 @item show filename-display
7760 Show the current way to display filenames.
7764 @section Selecting a Frame
7766 Most commands for examining the stack and other data in your program work on
7767 whichever stack frame is selected at the moment. Here are the commands for
7768 selecting a stack frame; all of them finish by printing a brief description
7769 of the stack frame just selected.
7772 @kindex frame@r{, selecting}
7773 @kindex f @r{(@code{frame})}
7774 @item frame @r{[} @var{frame-selection-spec} @r{]}
7775 @item f @r{[} @var{frame-selection-spec} @r{]}
7776 The @command{frame} command allows different stack frames to be
7777 selected. The @var{frame-selection-spec} can be any of the following:
7782 @item level @var{num}
7783 Select frame level @var{num}. Recall that frame zero is the innermost
7784 (currently executing) frame, frame one is the frame that called the
7785 innermost one, and so on. The highest level frame is usually the one
7788 As this is the most common method of navigating the frame stack, the
7789 string @command{level} can be omitted. For example, the following two
7790 commands are equivalent:
7793 (@value{GDBP}) frame 3
7794 (@value{GDBP}) frame level 3
7797 @kindex frame address
7798 @item address @var{stack-address}
7799 Select the frame with stack address @var{stack-address}. The
7800 @var{stack-address} for a frame can be seen in the output of
7801 @command{info frame}, for example:
7805 Stack level 1, frame at 0x7fffffffda30:
7806 rip = 0x40066d in b (amd64-entry-value.cc:59); saved rip 0x4004c5
7807 tail call frame, caller of frame at 0x7fffffffda30
7808 source language c++.
7809 Arglist at unknown address.
7810 Locals at unknown address, Previous frame's sp is 0x7fffffffda30
7813 The @var{stack-address} for this frame is @code{0x7fffffffda30} as
7814 indicated by the line:
7817 Stack level 1, frame at 0x7fffffffda30:
7820 @kindex frame function
7821 @item function @var{function-name}
7822 Select the stack frame for function @var{function-name}. If there are
7823 multiple stack frames for function @var{function-name} then the inner
7824 most stack frame is selected.
7827 @item view @var{stack-address} @r{[} @var{pc-addr} @r{]}
7828 View a frame that is not part of @value{GDBN}'s backtrace. The frame
7829 viewed has stack address @var{stack-addr}, and optionally, a program
7830 counter address of @var{pc-addr}.
7832 This is useful mainly if the chaining of stack frames has been
7833 damaged by a bug, making it impossible for @value{GDBN} to assign
7834 numbers properly to all frames. In addition, this can be useful
7835 when your program has multiple stacks and switches between them.
7837 When viewing a frame outside the current backtrace using
7838 @command{frame view} then you can always return to the original
7839 stack using one of the previous stack frame selection instructions,
7840 for example @command{frame level 0}.
7846 Move @var{n} frames up the stack; @var{n} defaults to 1. For positive
7847 numbers @var{n}, this advances toward the outermost frame, to higher
7848 frame numbers, to frames that have existed longer.
7851 @kindex do @r{(@code{down})}
7853 Move @var{n} frames down the stack; @var{n} defaults to 1. For
7854 positive numbers @var{n}, this advances toward the innermost frame, to
7855 lower frame numbers, to frames that were created more recently.
7856 You may abbreviate @code{down} as @code{do}.
7859 All of these commands end by printing two lines of output describing the
7860 frame. The first line shows the frame number, the function name, the
7861 arguments, and the source file and line number of execution in that
7862 frame. The second line shows the text of that source line.
7870 #1 0x22f0 in main (argc=1, argv=0xf7fffbf4, env=0xf7fffbfc)
7872 10 read_input_file (argv[i]);
7876 After such a printout, the @code{list} command with no arguments
7877 prints ten lines centered on the point of execution in the frame.
7878 You can also edit the program at the point of execution with your favorite
7879 editing program by typing @code{edit}.
7880 @xref{List, ,Printing Source Lines},
7884 @kindex select-frame
7885 @item select-frame @r{[} @var{frame-selection-spec} @r{]}
7886 The @code{select-frame} command is a variant of @code{frame} that does
7887 not display the new frame after selecting it. This command is
7888 intended primarily for use in @value{GDBN} command scripts, where the
7889 output might be unnecessary and distracting. The
7890 @var{frame-selection-spec} is as for the @command{frame} command
7891 described in @ref{Selection, ,Selecting a Frame}.
7893 @kindex down-silently
7895 @item up-silently @var{n}
7896 @itemx down-silently @var{n}
7897 These two commands are variants of @code{up} and @code{down},
7898 respectively; they differ in that they do their work silently, without
7899 causing display of the new frame. They are intended primarily for use
7900 in @value{GDBN} command scripts, where the output might be unnecessary and
7905 @section Information About a Frame
7907 There are several other commands to print information about the selected
7913 When used without any argument, this command does not change which
7914 frame is selected, but prints a brief description of the currently
7915 selected stack frame. It can be abbreviated @code{f}. With an
7916 argument, this command is used to select a stack frame.
7917 @xref{Selection, ,Selecting a Frame}.
7920 @kindex info f @r{(@code{info frame})}
7923 This command prints a verbose description of the selected stack frame,
7928 the address of the frame
7930 the address of the next frame down (called by this frame)
7932 the address of the next frame up (caller of this frame)
7934 the language in which the source code corresponding to this frame is written
7936 the address of the frame's arguments
7938 the address of the frame's local variables
7940 the program counter saved in it (the address of execution in the caller frame)
7942 which registers were saved in the frame
7945 @noindent The verbose description is useful when
7946 something has gone wrong that has made the stack format fail to fit
7947 the usual conventions.
7949 @item info frame @r{[} @var{frame-selection-spec} @r{]}
7950 @itemx info f @r{[} @var{frame-selection-spec} @r{]}
7951 Print a verbose description of the frame selected by
7952 @var{frame-selection-spec}. The @var{frame-selection-spec} is the
7953 same as for the @command{frame} command (@pxref{Selection, ,Selecting
7954 a Frame}). The selected frame remains unchanged by this command.
7957 @item info args [-q]
7958 Print the arguments of the selected frame, each on a separate line.
7960 The optional flag @samp{-q}, which stands for @samp{quiet}, disables
7961 printing header information and messages explaining why no argument
7964 @item info args [-q] [-t @var{type_regexp}] [@var{regexp}]
7965 Like @kbd{info args}, but only print the arguments selected
7966 with the provided regexp(s).
7968 If @var{regexp} is provided, print only the arguments whose names
7969 match the regular expression @var{regexp}.
7971 If @var{type_regexp} is provided, print only the arguments whose
7972 types, as printed by the @code{whatis} command, match
7973 the regular expression @var{type_regexp}.
7974 If @var{type_regexp} contains space(s), it should be enclosed in
7975 quote characters. If needed, use backslash to escape the meaning
7976 of special characters or quotes.
7978 If both @var{regexp} and @var{type_regexp} are provided, an argument
7979 is printed only if its name matches @var{regexp} and its type matches
7982 @item info locals [-q]
7984 Print the local variables of the selected frame, each on a separate
7985 line. These are all variables (declared either static or automatic)
7986 accessible at the point of execution of the selected frame.
7988 The optional flag @samp{-q}, which stands for @samp{quiet}, disables
7989 printing header information and messages explaining why no local variables
7992 @item info locals [-q] [-t @var{type_regexp}] [@var{regexp}]
7993 Like @kbd{info locals}, but only print the local variables selected
7994 with the provided regexp(s).
7996 If @var{regexp} is provided, print only the local variables whose names
7997 match the regular expression @var{regexp}.
7999 If @var{type_regexp} is provided, print only the local variables whose
8000 types, as printed by the @code{whatis} command, match
8001 the regular expression @var{type_regexp}.
8002 If @var{type_regexp} contains space(s), it should be enclosed in
8003 quote characters. If needed, use backslash to escape the meaning
8004 of special characters or quotes.
8006 If both @var{regexp} and @var{type_regexp} are provided, a local variable
8007 is printed only if its name matches @var{regexp} and its type matches
8010 The command @kbd{info locals -q -t @var{type_regexp}} can usefully be
8011 combined with the commands @kbd{frame apply} and @kbd{thread apply}.
8012 For example, your program might use Resource Acquisition Is
8013 Initialization types (RAII) such as @code{lock_something_t}: each
8014 local variable of type @code{lock_something_t} automatically places a
8015 lock that is destroyed when the variable goes out of scope. You can
8016 then list all acquired locks in your program by doing
8018 thread apply all -s frame apply all -s info locals -q -t lock_something_t
8021 or the equivalent shorter form
8023 tfaas i lo -q -t lock_something_t
8029 @section Applying a Command to Several Frames.
8031 @cindex apply command to several frames
8033 @item frame apply [all | @var{count} | @var{-count} | level @var{level}@dots{}] [@var{flag}]@dots{} @var{command}
8034 The @code{frame apply} command allows you to apply the named
8035 @var{command} to one or more frames.
8039 Specify @code{all} to apply @var{command} to all frames.
8042 Use @var{count} to apply @var{command} to the innermost @var{count}
8043 frames, where @var{count} is a positive number.
8046 Use @var{-count} to apply @var{command} to the outermost @var{count}
8047 frames, where @var{count} is a positive number.
8050 Use @code{level} to apply @var{command} to the set of frames identified
8051 by the @var{level} list. @var{level} is a frame level or a range of frame
8052 levels as @var{level1}-@var{level2}. The frame level is the number shown
8053 in the first field of the @samp{backtrace} command output.
8054 E.g., @samp{2-4 6-8 3} indicates to apply @var{command} for the frames
8055 at levels 2, 3, 4, 6, 7, 8, and then again on frame at level 3.
8061 Note that the frames on which @code{frame apply} applies a command are
8062 also influenced by the @code{set backtrace} settings such as @code{set
8063 backtrace past-main} and @code{set backtrace limit N}. See
8064 @xref{Backtrace,,Backtraces}.
8066 The @var{flag} arguments control what output to produce and how to handle
8067 errors raised when applying @var{command} to a frame. @var{flag}
8068 must start with a @code{-} directly followed by one letter in
8069 @code{qcs}. If several flags are provided, they must be given
8070 individually, such as @code{-c -q}.
8072 By default, @value{GDBN} displays some frame information before the
8073 output produced by @var{command}, and an error raised during the
8074 execution of a @var{command} will abort @code{frame apply}. The
8075 following flags can be used to fine-tune this behavior:
8079 The flag @code{-c}, which stands for @samp{continue}, causes any
8080 errors in @var{command} to be displayed, and the execution of
8081 @code{frame apply} then continues.
8083 The flag @code{-s}, which stands for @samp{silent}, causes any errors
8084 or empty output produced by a @var{command} to be silently ignored.
8085 That is, the execution continues, but the frame information and errors
8088 The flag @code{-q} (@samp{quiet}) disables printing the frame
8092 The following example shows how the flags @code{-c} and @code{-s} are
8093 working when applying the command @code{p j} to all frames, where
8094 variable @code{j} can only be successfully printed in the outermost
8095 @code{#1 main} frame.
8099 (gdb) frame apply all p j
8100 #0 some_function (i=5) at fun.c:4
8101 No symbol "j" in current context.
8102 (gdb) frame apply all -c p j
8103 #0 some_function (i=5) at fun.c:4
8104 No symbol "j" in current context.
8105 #1 0x565555fb in main (argc=1, argv=0xffffd2c4) at fun.c:11
8107 (gdb) frame apply all -s p j
8108 #1 0x565555fb in main (argc=1, argv=0xffffd2c4) at fun.c:11
8114 By default, @samp{frame apply}, prints the frame location
8115 information before the command output:
8119 (gdb) frame apply all p $sp
8120 #0 some_function (i=5) at fun.c:4
8121 $4 = (void *) 0xffffd1e0
8122 #1 0x565555fb in main (argc=1, argv=0xffffd2c4) at fun.c:11
8123 $5 = (void *) 0xffffd1f0
8128 If flag @code{-q} is given, no frame information is printed:
8131 (gdb) frame apply all -q p $sp
8132 $12 = (void *) 0xffffd1e0
8133 $13 = (void *) 0xffffd1f0
8141 @cindex apply a command to all frames (ignoring errors and empty output)
8142 @item faas @var{command}
8143 Shortcut for @code{frame apply all -s @var{command}}.
8144 Applies @var{command} on all frames, ignoring errors and empty output.
8146 It can for example be used to print a local variable or a function
8147 argument without knowing the frame where this variable or argument
8150 (@value{GDBP}) faas p some_local_var_i_do_not_remember_where_it_is
8153 Note that the command @code{tfaas @var{command}} applies @var{command}
8154 on all frames of all threads. See @xref{Threads,,Threads}.
8158 @node Frame Filter Management
8159 @section Management of Frame Filters.
8160 @cindex managing frame filters
8162 Frame filters are Python based utilities to manage and decorate the
8163 output of frames. @xref{Frame Filter API}, for further information.
8165 Managing frame filters is performed by several commands available
8166 within @value{GDBN}, detailed here.
8169 @kindex info frame-filter
8170 @item info frame-filter
8171 Print a list of installed frame filters from all dictionaries, showing
8172 their name, priority and enabled status.
8174 @kindex disable frame-filter
8175 @anchor{disable frame-filter all}
8176 @item disable frame-filter @var{filter-dictionary} @var{filter-name}
8177 Disable a frame filter in the dictionary matching
8178 @var{filter-dictionary} and @var{filter-name}. The
8179 @var{filter-dictionary} may be @code{all}, @code{global},
8180 @code{progspace}, or the name of the object file where the frame filter
8181 dictionary resides. When @code{all} is specified, all frame filters
8182 across all dictionaries are disabled. The @var{filter-name} is the name
8183 of the frame filter and is used when @code{all} is not the option for
8184 @var{filter-dictionary}. A disabled frame-filter is not deleted, it
8185 may be enabled again later.
8187 @kindex enable frame-filter
8188 @item enable frame-filter @var{filter-dictionary} @var{filter-name}
8189 Enable a frame filter in the dictionary matching
8190 @var{filter-dictionary} and @var{filter-name}. The
8191 @var{filter-dictionary} may be @code{all}, @code{global},
8192 @code{progspace} or the name of the object file where the frame filter
8193 dictionary resides. When @code{all} is specified, all frame filters across
8194 all dictionaries are enabled. The @var{filter-name} is the name of the frame
8195 filter and is used when @code{all} is not the option for
8196 @var{filter-dictionary}.
8201 (gdb) info frame-filter
8203 global frame-filters:
8204 Priority Enabled Name
8205 1000 No PrimaryFunctionFilter
8208 progspace /build/test frame-filters:
8209 Priority Enabled Name
8210 100 Yes ProgspaceFilter
8212 objfile /build/test frame-filters:
8213 Priority Enabled Name
8214 999 Yes BuildProgra Filter
8216 (gdb) disable frame-filter /build/test BuildProgramFilter
8217 (gdb) info frame-filter
8219 global frame-filters:
8220 Priority Enabled Name
8221 1000 No PrimaryFunctionFilter
8224 progspace /build/test frame-filters:
8225 Priority Enabled Name
8226 100 Yes ProgspaceFilter
8228 objfile /build/test frame-filters:
8229 Priority Enabled Name
8230 999 No BuildProgramFilter
8232 (gdb) enable frame-filter global PrimaryFunctionFilter
8233 (gdb) info frame-filter
8235 global frame-filters:
8236 Priority Enabled Name
8237 1000 Yes PrimaryFunctionFilter
8240 progspace /build/test frame-filters:
8241 Priority Enabled Name
8242 100 Yes ProgspaceFilter
8244 objfile /build/test frame-filters:
8245 Priority Enabled Name
8246 999 No BuildProgramFilter
8249 @kindex set frame-filter priority
8250 @item set frame-filter priority @var{filter-dictionary} @var{filter-name} @var{priority}
8251 Set the @var{priority} of a frame filter in the dictionary matching
8252 @var{filter-dictionary}, and the frame filter name matching
8253 @var{filter-name}. The @var{filter-dictionary} may be @code{global},
8254 @code{progspace} or the name of the object file where the frame filter
8255 dictionary resides. The @var{priority} is an integer.
8257 @kindex show frame-filter priority
8258 @item show frame-filter priority @var{filter-dictionary} @var{filter-name}
8259 Show the @var{priority} of a frame filter in the dictionary matching
8260 @var{filter-dictionary}, and the frame filter name matching
8261 @var{filter-name}. The @var{filter-dictionary} may be @code{global},
8262 @code{progspace} or the name of the object file where the frame filter
8268 (gdb) info frame-filter
8270 global frame-filters:
8271 Priority Enabled Name
8272 1000 Yes PrimaryFunctionFilter
8275 progspace /build/test frame-filters:
8276 Priority Enabled Name
8277 100 Yes ProgspaceFilter
8279 objfile /build/test frame-filters:
8280 Priority Enabled Name
8281 999 No BuildProgramFilter
8283 (gdb) set frame-filter priority global Reverse 50
8284 (gdb) info frame-filter
8286 global frame-filters:
8287 Priority Enabled Name
8288 1000 Yes PrimaryFunctionFilter
8291 progspace /build/test frame-filters:
8292 Priority Enabled Name
8293 100 Yes ProgspaceFilter
8295 objfile /build/test frame-filters:
8296 Priority Enabled Name
8297 999 No BuildProgramFilter
8302 @chapter Examining Source Files
8304 @value{GDBN} can print parts of your program's source, since the debugging
8305 information recorded in the program tells @value{GDBN} what source files were
8306 used to build it. When your program stops, @value{GDBN} spontaneously prints
8307 the line where it stopped. Likewise, when you select a stack frame
8308 (@pxref{Selection, ,Selecting a Frame}), @value{GDBN} prints the line where
8309 execution in that frame has stopped. You can print other portions of
8310 source files by explicit command.
8312 If you use @value{GDBN} through its @sc{gnu} Emacs interface, you may
8313 prefer to use Emacs facilities to view source; see @ref{Emacs, ,Using
8314 @value{GDBN} under @sc{gnu} Emacs}.
8317 * List:: Printing source lines
8318 * Specify Location:: How to specify code locations
8319 * Edit:: Editing source files
8320 * Search:: Searching source files
8321 * Source Path:: Specifying source directories
8322 * Machine Code:: Source and machine code
8326 @section Printing Source Lines
8329 @kindex l @r{(@code{list})}
8330 To print lines from a source file, use the @code{list} command
8331 (abbreviated @code{l}). By default, ten lines are printed.
8332 There are several ways to specify what part of the file you want to
8333 print; see @ref{Specify Location}, for the full list.
8335 Here are the forms of the @code{list} command most commonly used:
8338 @item list @var{linenum}
8339 Print lines centered around line number @var{linenum} in the
8340 current source file.
8342 @item list @var{function}
8343 Print lines centered around the beginning of function
8347 Print more lines. If the last lines printed were printed with a
8348 @code{list} command, this prints lines following the last lines
8349 printed; however, if the last line printed was a solitary line printed
8350 as part of displaying a stack frame (@pxref{Stack, ,Examining the
8351 Stack}), this prints lines centered around that line.
8354 Print lines just before the lines last printed.
8357 @cindex @code{list}, how many lines to display
8358 By default, @value{GDBN} prints ten source lines with any of these forms of
8359 the @code{list} command. You can change this using @code{set listsize}:
8362 @kindex set listsize
8363 @item set listsize @var{count}
8364 @itemx set listsize unlimited
8365 Make the @code{list} command display @var{count} source lines (unless
8366 the @code{list} argument explicitly specifies some other number).
8367 Setting @var{count} to @code{unlimited} or 0 means there's no limit.
8369 @kindex show listsize
8371 Display the number of lines that @code{list} prints.
8374 Repeating a @code{list} command with @key{RET} discards the argument,
8375 so it is equivalent to typing just @code{list}. This is more useful
8376 than listing the same lines again. An exception is made for an
8377 argument of @samp{-}; that argument is preserved in repetition so that
8378 each repetition moves up in the source file.
8380 In general, the @code{list} command expects you to supply zero, one or two
8381 @dfn{locations}. Locations specify source lines; there are several ways
8382 of writing them (@pxref{Specify Location}), but the effect is always
8383 to specify some source line.
8385 Here is a complete description of the possible arguments for @code{list}:
8388 @item list @var{location}
8389 Print lines centered around the line specified by @var{location}.
8391 @item list @var{first},@var{last}
8392 Print lines from @var{first} to @var{last}. Both arguments are
8393 locations. When a @code{list} command has two locations, and the
8394 source file of the second location is omitted, this refers to
8395 the same source file as the first location.
8397 @item list ,@var{last}
8398 Print lines ending with @var{last}.
8400 @item list @var{first},
8401 Print lines starting with @var{first}.
8404 Print lines just after the lines last printed.
8407 Print lines just before the lines last printed.
8410 As described in the preceding table.
8413 @node Specify Location
8414 @section Specifying a Location
8415 @cindex specifying location
8417 @cindex source location
8420 * Linespec Locations:: Linespec locations
8421 * Explicit Locations:: Explicit locations
8422 * Address Locations:: Address locations
8425 Several @value{GDBN} commands accept arguments that specify a location
8426 of your program's code. Since @value{GDBN} is a source-level
8427 debugger, a location usually specifies some line in the source code.
8428 Locations may be specified using three different formats:
8429 linespec locations, explicit locations, or address locations.
8431 @node Linespec Locations
8432 @subsection Linespec Locations
8433 @cindex linespec locations
8435 A @dfn{linespec} is a colon-separated list of source location parameters such
8436 as file name, function name, etc. Here are all the different ways of
8437 specifying a linespec:
8441 Specifies the line number @var{linenum} of the current source file.
8444 @itemx +@var{offset}
8445 Specifies the line @var{offset} lines before or after the @dfn{current
8446 line}. For the @code{list} command, the current line is the last one
8447 printed; for the breakpoint commands, this is the line at which
8448 execution stopped in the currently selected @dfn{stack frame}
8449 (@pxref{Frames, ,Frames}, for a description of stack frames.) When
8450 used as the second of the two linespecs in a @code{list} command,
8451 this specifies the line @var{offset} lines up or down from the first
8454 @item @var{filename}:@var{linenum}
8455 Specifies the line @var{linenum} in the source file @var{filename}.
8456 If @var{filename} is a relative file name, then it will match any
8457 source file name with the same trailing components. For example, if
8458 @var{filename} is @samp{gcc/expr.c}, then it will match source file
8459 name of @file{/build/trunk/gcc/expr.c}, but not
8460 @file{/build/trunk/libcpp/expr.c} or @file{/build/trunk/gcc/x-expr.c}.
8462 @item @var{function}
8463 Specifies the line that begins the body of the function @var{function}.
8464 For example, in C, this is the line with the open brace.
8466 By default, in C@t{++} and Ada, @var{function} is interpreted as
8467 specifying all functions named @var{function} in all scopes. For
8468 C@t{++}, this means in all namespaces and classes. For Ada, this
8469 means in all packages.
8471 For example, assuming a program with C@t{++} symbols named
8472 @code{A::B::func} and @code{B::func}, both commands @w{@kbd{break
8473 func}} and @w{@kbd{break B::func}} set a breakpoint on both symbols.
8475 Commands that accept a linespec let you override this with the
8476 @code{-qualified} option. For example, @w{@kbd{break -qualified
8477 func}} sets a breakpoint on a free-function named @code{func} ignoring
8478 any C@t{++} class methods and namespace functions called @code{func}.
8480 @xref{Explicit Locations}.
8482 @item @var{function}:@var{label}
8483 Specifies the line where @var{label} appears in @var{function}.
8485 @item @var{filename}:@var{function}
8486 Specifies the line that begins the body of the function @var{function}
8487 in the file @var{filename}. You only need the file name with a
8488 function name to avoid ambiguity when there are identically named
8489 functions in different source files.
8492 Specifies the line at which the label named @var{label} appears
8493 in the function corresponding to the currently selected stack frame.
8494 If there is no current selected stack frame (for instance, if the inferior
8495 is not running), then @value{GDBN} will not search for a label.
8497 @cindex breakpoint at static probe point
8498 @item -pstap|-probe-stap @r{[}@var{objfile}:@r{[}@var{provider}:@r{]}@r{]}@var{name}
8499 The @sc{gnu}/Linux tool @code{SystemTap} provides a way for
8500 applications to embed static probes. @xref{Static Probe Points}, for more
8501 information on finding and using static probes. This form of linespec
8502 specifies the location of such a static probe.
8504 If @var{objfile} is given, only probes coming from that shared library
8505 or executable matching @var{objfile} as a regular expression are considered.
8506 If @var{provider} is given, then only probes from that provider are considered.
8507 If several probes match the spec, @value{GDBN} will insert a breakpoint at
8508 each one of those probes.
8511 @node Explicit Locations
8512 @subsection Explicit Locations
8513 @cindex explicit locations
8515 @dfn{Explicit locations} allow the user to directly specify the source
8516 location's parameters using option-value pairs.
8518 Explicit locations are useful when several functions, labels, or
8519 file names have the same name (base name for files) in the program's
8520 sources. In these cases, explicit locations point to the source
8521 line you meant more accurately and unambiguously. Also, using
8522 explicit locations might be faster in large programs.
8524 For example, the linespec @samp{foo:bar} may refer to a function @code{bar}
8525 defined in the file named @file{foo} or the label @code{bar} in a function
8526 named @code{foo}. @value{GDBN} must search either the file system or
8527 the symbol table to know.
8529 The list of valid explicit location options is summarized in the
8533 @item -source @var{filename}
8534 The value specifies the source file name. To differentiate between
8535 files with the same base name, prepend as many directories as is necessary
8536 to uniquely identify the desired file, e.g., @file{foo/bar/baz.c}. Otherwise
8537 @value{GDBN} will use the first file it finds with the given base
8538 name. This option requires the use of either @code{-function} or @code{-line}.
8540 @item -function @var{function}
8541 The value specifies the name of a function. Operations
8542 on function locations unmodified by other options (such as @code{-label}
8543 or @code{-line}) refer to the line that begins the body of the function.
8544 In C, for example, this is the line with the open brace.
8546 By default, in C@t{++} and Ada, @var{function} is interpreted as
8547 specifying all functions named @var{function} in all scopes. For
8548 C@t{++}, this means in all namespaces and classes. For Ada, this
8549 means in all packages.
8551 For example, assuming a program with C@t{++} symbols named
8552 @code{A::B::func} and @code{B::func}, both commands @w{@kbd{break
8553 -function func}} and @w{@kbd{break -function B::func}} set a
8554 breakpoint on both symbols.
8556 You can use the @kbd{-qualified} flag to override this (see below).
8560 This flag makes @value{GDBN} interpret a function name specified with
8561 @kbd{-function} as a complete fully-qualified name.
8563 For example, assuming a C@t{++} program with symbols named
8564 @code{A::B::func} and @code{B::func}, the @w{@kbd{break -qualified
8565 -function B::func}} command sets a breakpoint on @code{B::func}, only.
8567 (Note: the @kbd{-qualified} option can precede a linespec as well
8568 (@pxref{Linespec Locations}), so the particular example above could be
8569 simplified as @w{@kbd{break -qualified B::func}}.)
8571 @item -label @var{label}
8572 The value specifies the name of a label. When the function
8573 name is not specified, the label is searched in the function of the currently
8574 selected stack frame.
8576 @item -line @var{number}
8577 The value specifies a line offset for the location. The offset may either
8578 be absolute (@code{-line 3}) or relative (@code{-line +3}), depending on
8579 the command. When specified without any other options, the line offset is
8580 relative to the current line.
8583 Explicit location options may be abbreviated by omitting any non-unique
8584 trailing characters from the option name, e.g., @w{@kbd{break -s main.c -li 3}}.
8586 @node Address Locations
8587 @subsection Address Locations
8588 @cindex address locations
8590 @dfn{Address locations} indicate a specific program address. They have
8591 the generalized form *@var{address}.
8593 For line-oriented commands, such as @code{list} and @code{edit}, this
8594 specifies a source line that contains @var{address}. For @code{break} and
8595 other breakpoint-oriented commands, this can be used to set breakpoints in
8596 parts of your program which do not have debugging information or
8599 Here @var{address} may be any expression valid in the current working
8600 language (@pxref{Languages, working language}) that specifies a code
8601 address. In addition, as a convenience, @value{GDBN} extends the
8602 semantics of expressions used in locations to cover several situations
8603 that frequently occur during debugging. Here are the various forms
8607 @item @var{expression}
8608 Any expression valid in the current working language.
8610 @item @var{funcaddr}
8611 An address of a function or procedure derived from its name. In C,
8612 C@t{++}, Objective-C, Fortran, minimal, and assembly, this is
8613 simply the function's name @var{function} (and actually a special case
8614 of a valid expression). In Pascal and Modula-2, this is
8615 @code{&@var{function}}. In Ada, this is @code{@var{function}'Address}
8616 (although the Pascal form also works).
8618 This form specifies the address of the function's first instruction,
8619 before the stack frame and arguments have been set up.
8621 @item '@var{filename}':@var{funcaddr}
8622 Like @var{funcaddr} above, but also specifies the name of the source
8623 file explicitly. This is useful if the name of the function does not
8624 specify the function unambiguously, e.g., if there are several
8625 functions with identical names in different source files.
8629 @section Editing Source Files
8630 @cindex editing source files
8633 @kindex e @r{(@code{edit})}
8634 To edit the lines in a source file, use the @code{edit} command.
8635 The editing program of your choice
8636 is invoked with the current line set to
8637 the active line in the program.
8638 Alternatively, there are several ways to specify what part of the file you
8639 want to print if you want to see other parts of the program:
8642 @item edit @var{location}
8643 Edit the source file specified by @code{location}. Editing starts at
8644 that @var{location}, e.g., at the specified source line of the
8645 specified file. @xref{Specify Location}, for all the possible forms
8646 of the @var{location} argument; here are the forms of the @code{edit}
8647 command most commonly used:
8650 @item edit @var{number}
8651 Edit the current source file with @var{number} as the active line number.
8653 @item edit @var{function}
8654 Edit the file containing @var{function} at the beginning of its definition.
8659 @subsection Choosing your Editor
8660 You can customize @value{GDBN} to use any editor you want
8662 The only restriction is that your editor (say @code{ex}), recognizes the
8663 following command-line syntax:
8665 ex +@var{number} file
8667 The optional numeric value +@var{number} specifies the number of the line in
8668 the file where to start editing.}.
8669 By default, it is @file{@value{EDITOR}}, but you can change this
8670 by setting the environment variable @code{EDITOR} before using
8671 @value{GDBN}. For example, to configure @value{GDBN} to use the
8672 @code{vi} editor, you could use these commands with the @code{sh} shell:
8678 or in the @code{csh} shell,
8680 setenv EDITOR /usr/bin/vi
8685 @section Searching Source Files
8686 @cindex searching source files
8688 There are two commands for searching through the current source file for a
8693 @kindex forward-search
8694 @kindex fo @r{(@code{forward-search})}
8695 @item forward-search @var{regexp}
8696 @itemx search @var{regexp}
8697 The command @samp{forward-search @var{regexp}} checks each line,
8698 starting with the one following the last line listed, for a match for
8699 @var{regexp}. It lists the line that is found. You can use the
8700 synonym @samp{search @var{regexp}} or abbreviate the command name as
8703 @kindex reverse-search
8704 @item reverse-search @var{regexp}
8705 The command @samp{reverse-search @var{regexp}} checks each line, starting
8706 with the one before the last line listed and going backward, for a match
8707 for @var{regexp}. It lists the line that is found. You can abbreviate
8708 this command as @code{rev}.
8712 @section Specifying Source Directories
8715 @cindex directories for source files
8716 Executable programs sometimes do not record the directories of the source
8717 files from which they were compiled, just the names. Even when they do,
8718 the directories could be moved between the compilation and your debugging
8719 session. @value{GDBN} has a list of directories to search for source files;
8720 this is called the @dfn{source path}. Each time @value{GDBN} wants a source file,
8721 it tries all the directories in the list, in the order they are present
8722 in the list, until it finds a file with the desired name.
8724 For example, suppose an executable references the file
8725 @file{/usr/src/foo-1.0/lib/foo.c}, and our source path is
8726 @file{/mnt/cross}. The file is first looked up literally; if this
8727 fails, @file{/mnt/cross/usr/src/foo-1.0/lib/foo.c} is tried; if this
8728 fails, @file{/mnt/cross/foo.c} is opened; if this fails, an error
8729 message is printed. @value{GDBN} does not look up the parts of the
8730 source file name, such as @file{/mnt/cross/src/foo-1.0/lib/foo.c}.
8731 Likewise, the subdirectories of the source path are not searched: if
8732 the source path is @file{/mnt/cross}, and the binary refers to
8733 @file{foo.c}, @value{GDBN} would not find it under
8734 @file{/mnt/cross/usr/src/foo-1.0/lib}.
8736 Plain file names, relative file names with leading directories, file
8737 names containing dots, etc.@: are all treated as described above; for
8738 instance, if the source path is @file{/mnt/cross}, and the source file
8739 is recorded as @file{../lib/foo.c}, @value{GDBN} would first try
8740 @file{../lib/foo.c}, then @file{/mnt/cross/../lib/foo.c}, and after
8741 that---@file{/mnt/cross/foo.c}.
8743 Note that the executable search path is @emph{not} used to locate the
8746 Whenever you reset or rearrange the source path, @value{GDBN} clears out
8747 any information it has cached about where source files are found and where
8748 each line is in the file.
8752 When you start @value{GDBN}, its source path includes only @samp{cdir}
8753 and @samp{cwd}, in that order.
8754 To add other directories, use the @code{directory} command.
8756 The search path is used to find both program source files and @value{GDBN}
8757 script files (read using the @samp{-command} option and @samp{source} command).
8759 In addition to the source path, @value{GDBN} provides a set of commands
8760 that manage a list of source path substitution rules. A @dfn{substitution
8761 rule} specifies how to rewrite source directories stored in the program's
8762 debug information in case the sources were moved to a different
8763 directory between compilation and debugging. A rule is made of
8764 two strings, the first specifying what needs to be rewritten in
8765 the path, and the second specifying how it should be rewritten.
8766 In @ref{set substitute-path}, we name these two parts @var{from} and
8767 @var{to} respectively. @value{GDBN} does a simple string replacement
8768 of @var{from} with @var{to} at the start of the directory part of the
8769 source file name, and uses that result instead of the original file
8770 name to look up the sources.
8772 Using the previous example, suppose the @file{foo-1.0} tree has been
8773 moved from @file{/usr/src} to @file{/mnt/cross}, then you can tell
8774 @value{GDBN} to replace @file{/usr/src} in all source path names with
8775 @file{/mnt/cross}. The first lookup will then be
8776 @file{/mnt/cross/foo-1.0/lib/foo.c} in place of the original location
8777 of @file{/usr/src/foo-1.0/lib/foo.c}. To define a source path
8778 substitution rule, use the @code{set substitute-path} command
8779 (@pxref{set substitute-path}).
8781 To avoid unexpected substitution results, a rule is applied only if the
8782 @var{from} part of the directory name ends at a directory separator.
8783 For instance, a rule substituting @file{/usr/source} into
8784 @file{/mnt/cross} will be applied to @file{/usr/source/foo-1.0} but
8785 not to @file{/usr/sourceware/foo-2.0}. And because the substitution
8786 is applied only at the beginning of the directory name, this rule will
8787 not be applied to @file{/root/usr/source/baz.c} either.
8789 In many cases, you can achieve the same result using the @code{directory}
8790 command. However, @code{set substitute-path} can be more efficient in
8791 the case where the sources are organized in a complex tree with multiple
8792 subdirectories. With the @code{directory} command, you need to add each
8793 subdirectory of your project. If you moved the entire tree while
8794 preserving its internal organization, then @code{set substitute-path}
8795 allows you to direct the debugger to all the sources with one single
8798 @code{set substitute-path} is also more than just a shortcut command.
8799 The source path is only used if the file at the original location no
8800 longer exists. On the other hand, @code{set substitute-path} modifies
8801 the debugger behavior to look at the rewritten location instead. So, if
8802 for any reason a source file that is not relevant to your executable is
8803 located at the original location, a substitution rule is the only
8804 method available to point @value{GDBN} at the new location.
8806 @cindex @samp{--with-relocated-sources}
8807 @cindex default source path substitution
8808 You can configure a default source path substitution rule by
8809 configuring @value{GDBN} with the
8810 @samp{--with-relocated-sources=@var{dir}} option. The @var{dir}
8811 should be the name of a directory under @value{GDBN}'s configured
8812 prefix (set with @samp{--prefix} or @samp{--exec-prefix}), and
8813 directory names in debug information under @var{dir} will be adjusted
8814 automatically if the installed @value{GDBN} is moved to a new
8815 location. This is useful if @value{GDBN}, libraries or executables
8816 with debug information and corresponding source code are being moved
8820 @item directory @var{dirname} @dots{}
8821 @item dir @var{dirname} @dots{}
8822 Add directory @var{dirname} to the front of the source path. Several
8823 directory names may be given to this command, separated by @samp{:}
8824 (@samp{;} on MS-DOS and MS-Windows, where @samp{:} usually appears as
8825 part of absolute file names) or
8826 whitespace. You may specify a directory that is already in the source
8827 path; this moves it forward, so @value{GDBN} searches it sooner.
8831 @vindex $cdir@r{, convenience variable}
8832 @vindex $cwd@r{, convenience variable}
8833 @cindex compilation directory
8834 @cindex current directory
8835 @cindex working directory
8836 @cindex directory, current
8837 @cindex directory, compilation
8838 You can use the string @samp{$cdir} to refer to the compilation
8839 directory (if one is recorded), and @samp{$cwd} to refer to the current
8840 working directory. @samp{$cwd} is not the same as @samp{.}---the former
8841 tracks the current working directory as it changes during your @value{GDBN}
8842 session, while the latter is immediately expanded to the current
8843 directory at the time you add an entry to the source path.
8846 Reset the source path to its default value (@samp{$cdir:$cwd} on Unix systems). This requires confirmation.
8848 @c RET-repeat for @code{directory} is explicitly disabled, but since
8849 @c repeating it would be a no-op we do not say that. (thanks to RMS)
8851 @item set directories @var{path-list}
8852 @kindex set directories
8853 Set the source path to @var{path-list}.
8854 @samp{$cdir:$cwd} are added if missing.
8856 @item show directories
8857 @kindex show directories
8858 Print the source path: show which directories it contains.
8860 @anchor{set substitute-path}
8861 @item set substitute-path @var{from} @var{to}
8862 @kindex set substitute-path
8863 Define a source path substitution rule, and add it at the end of the
8864 current list of existing substitution rules. If a rule with the same
8865 @var{from} was already defined, then the old rule is also deleted.
8867 For example, if the file @file{/foo/bar/baz.c} was moved to
8868 @file{/mnt/cross/baz.c}, then the command
8871 (@value{GDBP}) set substitute-path /foo/bar /mnt/cross
8875 will tell @value{GDBN} to replace @samp{/foo/bar} with
8876 @samp{/mnt/cross}, which will allow @value{GDBN} to find the file
8877 @file{baz.c} even though it was moved.
8879 In the case when more than one substitution rule have been defined,
8880 the rules are evaluated one by one in the order where they have been
8881 defined. The first one matching, if any, is selected to perform
8884 For instance, if we had entered the following commands:
8887 (@value{GDBP}) set substitute-path /usr/src/include /mnt/include
8888 (@value{GDBP}) set substitute-path /usr/src /mnt/src
8892 @value{GDBN} would then rewrite @file{/usr/src/include/defs.h} into
8893 @file{/mnt/include/defs.h} by using the first rule. However, it would
8894 use the second rule to rewrite @file{/usr/src/lib/foo.c} into
8895 @file{/mnt/src/lib/foo.c}.
8898 @item unset substitute-path [path]
8899 @kindex unset substitute-path
8900 If a path is specified, search the current list of substitution rules
8901 for a rule that would rewrite that path. Delete that rule if found.
8902 A warning is emitted by the debugger if no rule could be found.
8904 If no path is specified, then all substitution rules are deleted.
8906 @item show substitute-path [path]
8907 @kindex show substitute-path
8908 If a path is specified, then print the source path substitution rule
8909 which would rewrite that path, if any.
8911 If no path is specified, then print all existing source path substitution
8916 If your source path is cluttered with directories that are no longer of
8917 interest, @value{GDBN} may sometimes cause confusion by finding the wrong
8918 versions of source. You can correct the situation as follows:
8922 Use @code{directory} with no argument to reset the source path to its default value.
8925 Use @code{directory} with suitable arguments to reinstall the
8926 directories you want in the source path. You can add all the
8927 directories in one command.
8931 @section Source and Machine Code
8932 @cindex source line and its code address
8934 You can use the command @code{info line} to map source lines to program
8935 addresses (and vice versa), and the command @code{disassemble} to display
8936 a range of addresses as machine instructions. You can use the command
8937 @code{set disassemble-next-line} to set whether to disassemble next
8938 source line when execution stops. When run under @sc{gnu} Emacs
8939 mode, the @code{info line} command causes the arrow to point to the
8940 line specified. Also, @code{info line} prints addresses in symbolic form as
8946 @itemx info line @var{location}
8947 Print the starting and ending addresses of the compiled code for
8948 source line @var{location}. You can specify source lines in any of
8949 the ways documented in @ref{Specify Location}. With no @var{location}
8950 information about the current source line is printed.
8953 For example, we can use @code{info line} to discover the location of
8954 the object code for the first line of function
8955 @code{m4_changequote}:
8958 (@value{GDBP}) info line m4_changequote
8959 Line 895 of "builtin.c" starts at pc 0x634c <m4_changequote> and \
8960 ends at 0x6350 <m4_changequote+4>.
8964 @cindex code address and its source line
8965 We can also inquire (using @code{*@var{addr}} as the form for
8966 @var{location}) what source line covers a particular address:
8968 (@value{GDBP}) info line *0x63ff
8969 Line 926 of "builtin.c" starts at pc 0x63e4 <m4_changequote+152> and \
8970 ends at 0x6404 <m4_changequote+184>.
8973 @cindex @code{$_} and @code{info line}
8974 @cindex @code{x} command, default address
8975 @kindex x@r{(examine), and} info line
8976 After @code{info line}, the default address for the @code{x} command
8977 is changed to the starting address of the line, so that @samp{x/i} is
8978 sufficient to begin examining the machine code (@pxref{Memory,
8979 ,Examining Memory}). Also, this address is saved as the value of the
8980 convenience variable @code{$_} (@pxref{Convenience Vars, ,Convenience
8983 @cindex info line, repeated calls
8984 After @code{info line}, using @code{info line} again without
8985 specifying a location will display information about the next source
8990 @cindex assembly instructions
8991 @cindex instructions, assembly
8992 @cindex machine instructions
8993 @cindex listing machine instructions
8995 @itemx disassemble /m
8996 @itemx disassemble /s
8997 @itemx disassemble /r
8998 This specialized command dumps a range of memory as machine
8999 instructions. It can also print mixed source+disassembly by specifying
9000 the @code{/m} or @code{/s} modifier and print the raw instructions in hex
9001 as well as in symbolic form by specifying the @code{/r} modifier.
9002 The default memory range is the function surrounding the
9003 program counter of the selected frame. A single argument to this
9004 command is a program counter value; @value{GDBN} dumps the function
9005 surrounding this value. When two arguments are given, they should
9006 be separated by a comma, possibly surrounded by whitespace. The
9007 arguments specify a range of addresses to dump, in one of two forms:
9010 @item @var{start},@var{end}
9011 the addresses from @var{start} (inclusive) to @var{end} (exclusive)
9012 @item @var{start},+@var{length}
9013 the addresses from @var{start} (inclusive) to
9014 @code{@var{start}+@var{length}} (exclusive).
9018 When 2 arguments are specified, the name of the function is also
9019 printed (since there could be several functions in the given range).
9021 The argument(s) can be any expression yielding a numeric value, such as
9022 @samp{0x32c4}, @samp{&main+10} or @samp{$pc - 8}.
9024 If the range of memory being disassembled contains current program counter,
9025 the instruction at that location is shown with a @code{=>} marker.
9028 The following example shows the disassembly of a range of addresses of
9029 HP PA-RISC 2.0 code:
9032 (@value{GDBP}) disas 0x32c4, 0x32e4
9033 Dump of assembler code from 0x32c4 to 0x32e4:
9034 0x32c4 <main+204>: addil 0,dp
9035 0x32c8 <main+208>: ldw 0x22c(sr0,r1),r26
9036 0x32cc <main+212>: ldil 0x3000,r31
9037 0x32d0 <main+216>: ble 0x3f8(sr4,r31)
9038 0x32d4 <main+220>: ldo 0(r31),rp
9039 0x32d8 <main+224>: addil -0x800,dp
9040 0x32dc <main+228>: ldo 0x588(r1),r26
9041 0x32e0 <main+232>: ldil 0x3000,r31
9042 End of assembler dump.
9045 Here is an example showing mixed source+assembly for Intel x86
9046 with @code{/m} or @code{/s}, when the program is stopped just after
9047 function prologue in a non-optimized function with no inline code.
9050 (@value{GDBP}) disas /m main
9051 Dump of assembler code for function main:
9053 0x08048330 <+0>: push %ebp
9054 0x08048331 <+1>: mov %esp,%ebp
9055 0x08048333 <+3>: sub $0x8,%esp
9056 0x08048336 <+6>: and $0xfffffff0,%esp
9057 0x08048339 <+9>: sub $0x10,%esp
9059 6 printf ("Hello.\n");
9060 => 0x0804833c <+12>: movl $0x8048440,(%esp)
9061 0x08048343 <+19>: call 0x8048284 <puts@@plt>
9065 0x08048348 <+24>: mov $0x0,%eax
9066 0x0804834d <+29>: leave
9067 0x0804834e <+30>: ret
9069 End of assembler dump.
9072 The @code{/m} option is deprecated as its output is not useful when
9073 there is either inlined code or re-ordered code.
9074 The @code{/s} option is the preferred choice.
9075 Here is an example for AMD x86-64 showing the difference between
9076 @code{/m} output and @code{/s} output.
9077 This example has one inline function defined in a header file,
9078 and the code is compiled with @samp{-O2} optimization.
9079 Note how the @code{/m} output is missing the disassembly of
9080 several instructions that are present in the @code{/s} output.
9110 (@value{GDBP}) disas /m main
9111 Dump of assembler code for function main:
9115 0x0000000000400400 <+0>: mov 0x200c2e(%rip),%eax # 0x601034 <y>
9116 0x0000000000400417 <+23>: mov %eax,0x200c13(%rip) # 0x601030 <x>
9120 0x000000000040041d <+29>: xor %eax,%eax
9121 0x000000000040041f <+31>: retq
9122 0x0000000000400420 <+32>: add %eax,%eax
9123 0x0000000000400422 <+34>: jmp 0x400417 <main+23>
9125 End of assembler dump.
9126 (@value{GDBP}) disas /s main
9127 Dump of assembler code for function main:
9131 0x0000000000400400 <+0>: mov 0x200c2e(%rip),%eax # 0x601034 <y>
9135 0x0000000000400406 <+6>: test %eax,%eax
9136 0x0000000000400408 <+8>: js 0x400420 <main+32>
9141 0x000000000040040a <+10>: lea 0xa(%rax),%edx
9142 0x000000000040040d <+13>: test %eax,%eax
9143 0x000000000040040f <+15>: mov $0x1,%eax
9144 0x0000000000400414 <+20>: cmovne %edx,%eax
9148 0x0000000000400417 <+23>: mov %eax,0x200c13(%rip) # 0x601030 <x>
9152 0x000000000040041d <+29>: xor %eax,%eax
9153 0x000000000040041f <+31>: retq
9157 0x0000000000400420 <+32>: add %eax,%eax
9158 0x0000000000400422 <+34>: jmp 0x400417 <main+23>
9159 End of assembler dump.
9162 Here is another example showing raw instructions in hex for AMD x86-64,
9165 (gdb) disas /r 0x400281,+10
9166 Dump of assembler code from 0x400281 to 0x40028b:
9167 0x0000000000400281: 38 36 cmp %dh,(%rsi)
9168 0x0000000000400283: 2d 36 34 2e 73 sub $0x732e3436,%eax
9169 0x0000000000400288: 6f outsl %ds:(%rsi),(%dx)
9170 0x0000000000400289: 2e 32 00 xor %cs:(%rax),%al
9171 End of assembler dump.
9174 Addresses cannot be specified as a location (@pxref{Specify Location}).
9175 So, for example, if you want to disassemble function @code{bar}
9176 in file @file{foo.c}, you must type @samp{disassemble 'foo.c'::bar}
9177 and not @samp{disassemble foo.c:bar}.
9179 Some architectures have more than one commonly-used set of instruction
9180 mnemonics or other syntax.
9182 For programs that were dynamically linked and use shared libraries,
9183 instructions that call functions or branch to locations in the shared
9184 libraries might show a seemingly bogus location---it's actually a
9185 location of the relocation table. On some architectures, @value{GDBN}
9186 might be able to resolve these to actual function names.
9189 @kindex set disassembler-options
9190 @cindex disassembler options
9191 @item set disassembler-options @var{option1}[,@var{option2}@dots{}]
9192 This command controls the passing of target specific information to
9193 the disassembler. For a list of valid options, please refer to the
9194 @code{-M}/@code{--disassembler-options} section of the @samp{objdump}
9195 manual and/or the output of @kbd{objdump --help}
9196 (@pxref{objdump,,objdump,binutils,The GNU Binary Utilities}).
9197 The default value is the empty string.
9199 If it is necessary to specify more than one disassembler option, then
9200 multiple options can be placed together into a comma separated list.
9201 Currently this command is only supported on targets ARM, MIPS, PowerPC
9204 @kindex show disassembler-options
9205 @item show disassembler-options
9206 Show the current setting of the disassembler options.
9210 @kindex set disassembly-flavor
9211 @cindex Intel disassembly flavor
9212 @cindex AT&T disassembly flavor
9213 @item set disassembly-flavor @var{instruction-set}
9214 Select the instruction set to use when disassembling the
9215 program via the @code{disassemble} or @code{x/i} commands.
9217 Currently this command is only defined for the Intel x86 family. You
9218 can set @var{instruction-set} to either @code{intel} or @code{att}.
9219 The default is @code{att}, the AT&T flavor used by default by Unix
9220 assemblers for x86-based targets.
9222 @kindex show disassembly-flavor
9223 @item show disassembly-flavor
9224 Show the current setting of the disassembly flavor.
9228 @kindex set disassemble-next-line
9229 @kindex show disassemble-next-line
9230 @item set disassemble-next-line
9231 @itemx show disassemble-next-line
9232 Control whether or not @value{GDBN} will disassemble the next source
9233 line or instruction when execution stops. If ON, @value{GDBN} will
9234 display disassembly of the next source line when execution of the
9235 program being debugged stops. This is @emph{in addition} to
9236 displaying the source line itself, which @value{GDBN} always does if
9237 possible. If the next source line cannot be displayed for some reason
9238 (e.g., if @value{GDBN} cannot find the source file, or there's no line
9239 info in the debug info), @value{GDBN} will display disassembly of the
9240 next @emph{instruction} instead of showing the next source line. If
9241 AUTO, @value{GDBN} will display disassembly of next instruction only
9242 if the source line cannot be displayed. This setting causes
9243 @value{GDBN} to display some feedback when you step through a function
9244 with no line info or whose source file is unavailable. The default is
9245 OFF, which means never display the disassembly of the next line or
9251 @chapter Examining Data
9253 @cindex printing data
9254 @cindex examining data
9257 The usual way to examine data in your program is with the @code{print}
9258 command (abbreviated @code{p}), or its synonym @code{inspect}. It
9259 evaluates and prints the value of an expression of the language your
9260 program is written in (@pxref{Languages, ,Using @value{GDBN} with
9261 Different Languages}). It may also print the expression using a
9262 Python-based pretty-printer (@pxref{Pretty Printing}).
9265 @item print @var{expr}
9266 @itemx print /@var{f} @var{expr}
9267 @var{expr} is an expression (in the source language). By default the
9268 value of @var{expr} is printed in a format appropriate to its data type;
9269 you can choose a different format by specifying @samp{/@var{f}}, where
9270 @var{f} is a letter specifying the format; see @ref{Output Formats,,Output
9274 @itemx print /@var{f}
9275 @cindex reprint the last value
9276 If you omit @var{expr}, @value{GDBN} displays the last value again (from the
9277 @dfn{value history}; @pxref{Value History, ,Value History}). This allows you to
9278 conveniently inspect the same value in an alternative format.
9281 A more low-level way of examining data is with the @code{x} command.
9282 It examines data in memory at a specified address and prints it in a
9283 specified format. @xref{Memory, ,Examining Memory}.
9285 If you are interested in information about types, or about how the
9286 fields of a struct or a class are declared, use the @code{ptype @var{exp}}
9287 command rather than @code{print}. @xref{Symbols, ,Examining the Symbol
9290 @cindex exploring hierarchical data structures
9292 Another way of examining values of expressions and type information is
9293 through the Python extension command @code{explore} (available only if
9294 the @value{GDBN} build is configured with @code{--with-python}). It
9295 offers an interactive way to start at the highest level (or, the most
9296 abstract level) of the data type of an expression (or, the data type
9297 itself) and explore all the way down to leaf scalar values/fields
9298 embedded in the higher level data types.
9301 @item explore @var{arg}
9302 @var{arg} is either an expression (in the source language), or a type
9303 visible in the current context of the program being debugged.
9306 The working of the @code{explore} command can be illustrated with an
9307 example. If a data type @code{struct ComplexStruct} is defined in your
9317 struct ComplexStruct
9319 struct SimpleStruct *ss_p;
9325 followed by variable declarations as
9328 struct SimpleStruct ss = @{ 10, 1.11 @};
9329 struct ComplexStruct cs = @{ &ss, @{ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9 @} @};
9333 then, the value of the variable @code{cs} can be explored using the
9334 @code{explore} command as follows.
9338 The value of `cs' is a struct/class of type `struct ComplexStruct' with
9339 the following fields:
9341 ss_p = <Enter 0 to explore this field of type `struct SimpleStruct *'>
9342 arr = <Enter 1 to explore this field of type `int [10]'>
9344 Enter the field number of choice:
9348 Since the fields of @code{cs} are not scalar values, you are being
9349 prompted to chose the field you want to explore. Let's say you choose
9350 the field @code{ss_p} by entering @code{0}. Then, since this field is a
9351 pointer, you will be asked if it is pointing to a single value. From
9352 the declaration of @code{cs} above, it is indeed pointing to a single
9353 value, hence you enter @code{y}. If you enter @code{n}, then you will
9354 be asked if it were pointing to an array of values, in which case this
9355 field will be explored as if it were an array.
9358 `cs.ss_p' is a pointer to a value of type `struct SimpleStruct'
9359 Continue exploring it as a pointer to a single value [y/n]: y
9360 The value of `*(cs.ss_p)' is a struct/class of type `struct
9361 SimpleStruct' with the following fields:
9363 i = 10 .. (Value of type `int')
9364 d = 1.1100000000000001 .. (Value of type `double')
9366 Press enter to return to parent value:
9370 If the field @code{arr} of @code{cs} was chosen for exploration by
9371 entering @code{1} earlier, then since it is as array, you will be
9372 prompted to enter the index of the element in the array that you want
9376 `cs.arr' is an array of `int'.
9377 Enter the index of the element you want to explore in `cs.arr': 5
9379 `(cs.arr)[5]' is a scalar value of type `int'.
9383 Press enter to return to parent value:
9386 In general, at any stage of exploration, you can go deeper towards the
9387 leaf values by responding to the prompts appropriately, or hit the
9388 return key to return to the enclosing data structure (the @i{higher}
9389 level data structure).
9391 Similar to exploring values, you can use the @code{explore} command to
9392 explore types. Instead of specifying a value (which is typically a
9393 variable name or an expression valid in the current context of the
9394 program being debugged), you specify a type name. If you consider the
9395 same example as above, your can explore the type
9396 @code{struct ComplexStruct} by passing the argument
9397 @code{struct ComplexStruct} to the @code{explore} command.
9400 (gdb) explore struct ComplexStruct
9404 By responding to the prompts appropriately in the subsequent interactive
9405 session, you can explore the type @code{struct ComplexStruct} in a
9406 manner similar to how the value @code{cs} was explored in the above
9409 The @code{explore} command also has two sub-commands,
9410 @code{explore value} and @code{explore type}. The former sub-command is
9411 a way to explicitly specify that value exploration of the argument is
9412 being invoked, while the latter is a way to explicitly specify that type
9413 exploration of the argument is being invoked.
9416 @item explore value @var{expr}
9417 @cindex explore value
9418 This sub-command of @code{explore} explores the value of the
9419 expression @var{expr} (if @var{expr} is an expression valid in the
9420 current context of the program being debugged). The behavior of this
9421 command is identical to that of the behavior of the @code{explore}
9422 command being passed the argument @var{expr}.
9424 @item explore type @var{arg}
9425 @cindex explore type
9426 This sub-command of @code{explore} explores the type of @var{arg} (if
9427 @var{arg} is a type visible in the current context of program being
9428 debugged), or the type of the value/expression @var{arg} (if @var{arg}
9429 is an expression valid in the current context of the program being
9430 debugged). If @var{arg} is a type, then the behavior of this command is
9431 identical to that of the @code{explore} command being passed the
9432 argument @var{arg}. If @var{arg} is an expression, then the behavior of
9433 this command will be identical to that of the @code{explore} command
9434 being passed the type of @var{arg} as the argument.
9438 * Expressions:: Expressions
9439 * Ambiguous Expressions:: Ambiguous Expressions
9440 * Variables:: Program variables
9441 * Arrays:: Artificial arrays
9442 * Output Formats:: Output formats
9443 * Memory:: Examining memory
9444 * Auto Display:: Automatic display
9445 * Print Settings:: Print settings
9446 * Pretty Printing:: Python pretty printing
9447 * Value History:: Value history
9448 * Convenience Vars:: Convenience variables
9449 * Convenience Funs:: Convenience functions
9450 * Registers:: Registers
9451 * Floating Point Hardware:: Floating point hardware
9452 * Vector Unit:: Vector Unit
9453 * OS Information:: Auxiliary data provided by operating system
9454 * Memory Region Attributes:: Memory region attributes
9455 * Dump/Restore Files:: Copy between memory and a file
9456 * Core File Generation:: Cause a program dump its core
9457 * Character Sets:: Debugging programs that use a different
9458 character set than GDB does
9459 * Caching Target Data:: Data caching for targets
9460 * Searching Memory:: Searching memory for a sequence of bytes
9461 * Value Sizes:: Managing memory allocated for values
9465 @section Expressions
9468 @code{print} and many other @value{GDBN} commands accept an expression and
9469 compute its value. Any kind of constant, variable or operator defined
9470 by the programming language you are using is valid in an expression in
9471 @value{GDBN}. This includes conditional expressions, function calls,
9472 casts, and string constants. It also includes preprocessor macros, if
9473 you compiled your program to include this information; see
9476 @cindex arrays in expressions
9477 @value{GDBN} supports array constants in expressions input by
9478 the user. The syntax is @{@var{element}, @var{element}@dots{}@}. For example,
9479 you can use the command @code{print @{1, 2, 3@}} to create an array
9480 of three integers. If you pass an array to a function or assign it
9481 to a program variable, @value{GDBN} copies the array to memory that
9482 is @code{malloc}ed in the target program.
9484 Because C is so widespread, most of the expressions shown in examples in
9485 this manual are in C. @xref{Languages, , Using @value{GDBN} with Different
9486 Languages}, for information on how to use expressions in other
9489 In this section, we discuss operators that you can use in @value{GDBN}
9490 expressions regardless of your programming language.
9492 @cindex casts, in expressions
9493 Casts are supported in all languages, not just in C, because it is so
9494 useful to cast a number into a pointer in order to examine a structure
9495 at that address in memory.
9496 @c FIXME: casts supported---Mod2 true?
9498 @value{GDBN} supports these operators, in addition to those common
9499 to programming languages:
9503 @samp{@@} is a binary operator for treating parts of memory as arrays.
9504 @xref{Arrays, ,Artificial Arrays}, for more information.
9507 @samp{::} allows you to specify a variable in terms of the file or
9508 function where it is defined. @xref{Variables, ,Program Variables}.
9510 @cindex @{@var{type}@}
9511 @cindex type casting memory
9512 @cindex memory, viewing as typed object
9513 @cindex casts, to view memory
9514 @item @{@var{type}@} @var{addr}
9515 Refers to an object of type @var{type} stored at address @var{addr} in
9516 memory. The address @var{addr} may be any expression whose value is
9517 an integer or pointer (but parentheses are required around binary
9518 operators, just as in a cast). This construct is allowed regardless
9519 of what kind of data is normally supposed to reside at @var{addr}.
9522 @node Ambiguous Expressions
9523 @section Ambiguous Expressions
9524 @cindex ambiguous expressions
9526 Expressions can sometimes contain some ambiguous elements. For instance,
9527 some programming languages (notably Ada, C@t{++} and Objective-C) permit
9528 a single function name to be defined several times, for application in
9529 different contexts. This is called @dfn{overloading}. Another example
9530 involving Ada is generics. A @dfn{generic package} is similar to C@t{++}
9531 templates and is typically instantiated several times, resulting in
9532 the same function name being defined in different contexts.
9534 In some cases and depending on the language, it is possible to adjust
9535 the expression to remove the ambiguity. For instance in C@t{++}, you
9536 can specify the signature of the function you want to break on, as in
9537 @kbd{break @var{function}(@var{types})}. In Ada, using the fully
9538 qualified name of your function often makes the expression unambiguous
9541 When an ambiguity that needs to be resolved is detected, the debugger
9542 has the capability to display a menu of numbered choices for each
9543 possibility, and then waits for the selection with the prompt @samp{>}.
9544 The first option is always @samp{[0] cancel}, and typing @kbd{0 @key{RET}}
9545 aborts the current command. If the command in which the expression was
9546 used allows more than one choice to be selected, the next option in the
9547 menu is @samp{[1] all}, and typing @kbd{1 @key{RET}} selects all possible
9550 For example, the following session excerpt shows an attempt to set a
9551 breakpoint at the overloaded symbol @code{String::after}.
9552 We choose three particular definitions of that function name:
9554 @c FIXME! This is likely to change to show arg type lists, at least
9557 (@value{GDBP}) b String::after
9560 [2] file:String.cc; line number:867
9561 [3] file:String.cc; line number:860
9562 [4] file:String.cc; line number:875
9563 [5] file:String.cc; line number:853
9564 [6] file:String.cc; line number:846
9565 [7] file:String.cc; line number:735
9567 Breakpoint 1 at 0xb26c: file String.cc, line 867.
9568 Breakpoint 2 at 0xb344: file String.cc, line 875.
9569 Breakpoint 3 at 0xafcc: file String.cc, line 846.
9570 Multiple breakpoints were set.
9571 Use the "delete" command to delete unwanted
9578 @kindex set multiple-symbols
9579 @item set multiple-symbols @var{mode}
9580 @cindex multiple-symbols menu
9582 This option allows you to adjust the debugger behavior when an expression
9585 By default, @var{mode} is set to @code{all}. If the command with which
9586 the expression is used allows more than one choice, then @value{GDBN}
9587 automatically selects all possible choices. For instance, inserting
9588 a breakpoint on a function using an ambiguous name results in a breakpoint
9589 inserted on each possible match. However, if a unique choice must be made,
9590 then @value{GDBN} uses the menu to help you disambiguate the expression.
9591 For instance, printing the address of an overloaded function will result
9592 in the use of the menu.
9594 When @var{mode} is set to @code{ask}, the debugger always uses the menu
9595 when an ambiguity is detected.
9597 Finally, when @var{mode} is set to @code{cancel}, the debugger reports
9598 an error due to the ambiguity and the command is aborted.
9600 @kindex show multiple-symbols
9601 @item show multiple-symbols
9602 Show the current value of the @code{multiple-symbols} setting.
9606 @section Program Variables
9608 The most common kind of expression to use is the name of a variable
9611 Variables in expressions are understood in the selected stack frame
9612 (@pxref{Selection, ,Selecting a Frame}); they must be either:
9616 global (or file-static)
9623 visible according to the scope rules of the
9624 programming language from the point of execution in that frame
9627 @noindent This means that in the function
9642 you can examine and use the variable @code{a} whenever your program is
9643 executing within the function @code{foo}, but you can only use or
9644 examine the variable @code{b} while your program is executing inside
9645 the block where @code{b} is declared.
9647 @cindex variable name conflict
9648 There is an exception: you can refer to a variable or function whose
9649 scope is a single source file even if the current execution point is not
9650 in this file. But it is possible to have more than one such variable or
9651 function with the same name (in different source files). If that
9652 happens, referring to that name has unpredictable effects. If you wish,
9653 you can specify a static variable in a particular function or file by
9654 using the colon-colon (@code{::}) notation:
9656 @cindex colon-colon, context for variables/functions
9658 @c info cannot cope with a :: index entry, but why deprive hard copy readers?
9659 @cindex @code{::}, context for variables/functions
9662 @var{file}::@var{variable}
9663 @var{function}::@var{variable}
9667 Here @var{file} or @var{function} is the name of the context for the
9668 static @var{variable}. In the case of file names, you can use quotes to
9669 make sure @value{GDBN} parses the file name as a single word---for example,
9670 to print a global value of @code{x} defined in @file{f2.c}:
9673 (@value{GDBP}) p 'f2.c'::x
9676 The @code{::} notation is normally used for referring to
9677 static variables, since you typically disambiguate uses of local variables
9678 in functions by selecting the appropriate frame and using the
9679 simple name of the variable. However, you may also use this notation
9680 to refer to local variables in frames enclosing the selected frame:
9689 process (a); /* Stop here */
9700 For example, if there is a breakpoint at the commented line,
9701 here is what you might see
9702 when the program stops after executing the call @code{bar(0)}:
9707 (@value{GDBP}) p bar::a
9710 #2 0x080483d0 in foo (a=5) at foobar.c:12
9713 (@value{GDBP}) p bar::a
9717 @cindex C@t{++} scope resolution
9718 These uses of @samp{::} are very rarely in conflict with the very
9719 similar use of the same notation in C@t{++}. When they are in
9720 conflict, the C@t{++} meaning takes precedence; however, this can be
9721 overridden by quoting the file or function name with single quotes.
9723 For example, suppose the program is stopped in a method of a class
9724 that has a field named @code{includefile}, and there is also an
9725 include file named @file{includefile} that defines a variable,
9729 (@value{GDBP}) p includefile
9731 (@value{GDBP}) p includefile::some_global
9732 A syntax error in expression, near `'.
9733 (@value{GDBP}) p 'includefile'::some_global
9737 @cindex wrong values
9738 @cindex variable values, wrong
9739 @cindex function entry/exit, wrong values of variables
9740 @cindex optimized code, wrong values of variables
9742 @emph{Warning:} Occasionally, a local variable may appear to have the
9743 wrong value at certain points in a function---just after entry to a new
9744 scope, and just before exit.
9746 You may see this problem when you are stepping by machine instructions.
9747 This is because, on most machines, it takes more than one instruction to
9748 set up a stack frame (including local variable definitions); if you are
9749 stepping by machine instructions, variables may appear to have the wrong
9750 values until the stack frame is completely built. On exit, it usually
9751 also takes more than one machine instruction to destroy a stack frame;
9752 after you begin stepping through that group of instructions, local
9753 variable definitions may be gone.
9755 This may also happen when the compiler does significant optimizations.
9756 To be sure of always seeing accurate values, turn off all optimization
9759 @cindex ``No symbol "foo" in current context''
9760 Another possible effect of compiler optimizations is to optimize
9761 unused variables out of existence, or assign variables to registers (as
9762 opposed to memory addresses). Depending on the support for such cases
9763 offered by the debug info format used by the compiler, @value{GDBN}
9764 might not be able to display values for such local variables. If that
9765 happens, @value{GDBN} will print a message like this:
9768 No symbol "foo" in current context.
9771 To solve such problems, either recompile without optimizations, or use a
9772 different debug info format, if the compiler supports several such
9773 formats. @xref{Compilation}, for more information on choosing compiler
9774 options. @xref{C, ,C and C@t{++}}, for more information about debug
9775 info formats that are best suited to C@t{++} programs.
9777 If you ask to print an object whose contents are unknown to
9778 @value{GDBN}, e.g., because its data type is not completely specified
9779 by the debug information, @value{GDBN} will say @samp{<incomplete
9780 type>}. @xref{Symbols, incomplete type}, for more about this.
9782 @cindex no debug info variables
9783 If you try to examine or use the value of a (global) variable for
9784 which @value{GDBN} has no type information, e.g., because the program
9785 includes no debug information, @value{GDBN} displays an error message.
9786 @xref{Symbols, unknown type}, for more about unknown types. If you
9787 cast the variable to its declared type, @value{GDBN} gets the
9788 variable's value using the cast-to type as the variable's type. For
9789 example, in a C program:
9792 (@value{GDBP}) p var
9793 'var' has unknown type; cast it to its declared type
9794 (@value{GDBP}) p (float) var
9798 If you append @kbd{@@entry} string to a function parameter name you get its
9799 value at the time the function got called. If the value is not available an
9800 error message is printed. Entry values are available only with some compilers.
9801 Entry values are normally also printed at the function parameter list according
9802 to @ref{set print entry-values}.
9805 Breakpoint 1, d (i=30) at gdb.base/entry-value.c:29
9811 (gdb) print i@@entry
9815 Strings are identified as arrays of @code{char} values without specified
9816 signedness. Arrays of either @code{signed char} or @code{unsigned char} get
9817 printed as arrays of 1 byte sized integers. @code{-fsigned-char} or
9818 @code{-funsigned-char} @value{NGCC} options have no effect as @value{GDBN}
9819 defines literal string type @code{"char"} as @code{char} without a sign.
9824 signed char var1[] = "A";
9827 You get during debugging
9832 $2 = @{65 'A', 0 '\0'@}
9836 @section Artificial Arrays
9838 @cindex artificial array
9840 @kindex @@@r{, referencing memory as an array}
9841 It is often useful to print out several successive objects of the
9842 same type in memory; a section of an array, or an array of
9843 dynamically determined size for which only a pointer exists in the
9846 You can do this by referring to a contiguous span of memory as an
9847 @dfn{artificial array}, using the binary operator @samp{@@}. The left
9848 operand of @samp{@@} should be the first element of the desired array
9849 and be an individual object. The right operand should be the desired length
9850 of the array. The result is an array value whose elements are all of
9851 the type of the left argument. The first element is actually the left
9852 argument; the second element comes from bytes of memory immediately
9853 following those that hold the first element, and so on. Here is an
9854 example. If a program says
9857 int *array = (int *) malloc (len * sizeof (int));
9861 you can print the contents of @code{array} with
9867 The left operand of @samp{@@} must reside in memory. Array values made
9868 with @samp{@@} in this way behave just like other arrays in terms of
9869 subscripting, and are coerced to pointers when used in expressions.
9870 Artificial arrays most often appear in expressions via the value history
9871 (@pxref{Value History, ,Value History}), after printing one out.
9873 Another way to create an artificial array is to use a cast.
9874 This re-interprets a value as if it were an array.
9875 The value need not be in memory:
9877 (@value{GDBP}) p/x (short[2])0x12345678
9878 $1 = @{0x1234, 0x5678@}
9881 As a convenience, if you leave the array length out (as in
9882 @samp{(@var{type}[])@var{value}}) @value{GDBN} calculates the size to fill
9883 the value (as @samp{sizeof(@var{value})/sizeof(@var{type})}:
9885 (@value{GDBP}) p/x (short[])0x12345678
9886 $2 = @{0x1234, 0x5678@}
9889 Sometimes the artificial array mechanism is not quite enough; in
9890 moderately complex data structures, the elements of interest may not
9891 actually be adjacent---for example, if you are interested in the values
9892 of pointers in an array. One useful work-around in this situation is
9893 to use a convenience variable (@pxref{Convenience Vars, ,Convenience
9894 Variables}) as a counter in an expression that prints the first
9895 interesting value, and then repeat that expression via @key{RET}. For
9896 instance, suppose you have an array @code{dtab} of pointers to
9897 structures, and you are interested in the values of a field @code{fv}
9898 in each structure. Here is an example of what you might type:
9908 @node Output Formats
9909 @section Output Formats
9911 @cindex formatted output
9912 @cindex output formats
9913 By default, @value{GDBN} prints a value according to its data type. Sometimes
9914 this is not what you want. For example, you might want to print a number
9915 in hex, or a pointer in decimal. Or you might want to view data in memory
9916 at a certain address as a character string or as an instruction. To do
9917 these things, specify an @dfn{output format} when you print a value.
9919 The simplest use of output formats is to say how to print a value
9920 already computed. This is done by starting the arguments of the
9921 @code{print} command with a slash and a format letter. The format
9922 letters supported are:
9926 Regard the bits of the value as an integer, and print the integer in
9930 Print as integer in signed decimal.
9933 Print as integer in unsigned decimal.
9936 Print as integer in octal.
9939 Print as integer in binary. The letter @samp{t} stands for ``two''.
9940 @footnote{@samp{b} cannot be used because these format letters are also
9941 used with the @code{x} command, where @samp{b} stands for ``byte'';
9942 see @ref{Memory,,Examining Memory}.}
9945 @cindex unknown address, locating
9946 @cindex locate address
9947 Print as an address, both absolute in hexadecimal and as an offset from
9948 the nearest preceding symbol. You can use this format used to discover
9949 where (in what function) an unknown address is located:
9952 (@value{GDBP}) p/a 0x54320
9953 $3 = 0x54320 <_initialize_vx+396>
9957 The command @code{info symbol 0x54320} yields similar results.
9958 @xref{Symbols, info symbol}.
9961 Regard as an integer and print it as a character constant. This
9962 prints both the numerical value and its character representation. The
9963 character representation is replaced with the octal escape @samp{\nnn}
9964 for characters outside the 7-bit @sc{ascii} range.
9966 Without this format, @value{GDBN} displays @code{char},
9967 @w{@code{unsigned char}}, and @w{@code{signed char}} data as character
9968 constants. Single-byte members of vectors are displayed as integer
9972 Regard the bits of the value as a floating point number and print
9973 using typical floating point syntax.
9976 @cindex printing strings
9977 @cindex printing byte arrays
9978 Regard as a string, if possible. With this format, pointers to single-byte
9979 data are displayed as null-terminated strings and arrays of single-byte data
9980 are displayed as fixed-length strings. Other values are displayed in their
9983 Without this format, @value{GDBN} displays pointers to and arrays of
9984 @code{char}, @w{@code{unsigned char}}, and @w{@code{signed char}} as
9985 strings. Single-byte members of a vector are displayed as an integer
9989 Like @samp{x} formatting, the value is treated as an integer and
9990 printed as hexadecimal, but leading zeros are printed to pad the value
9991 to the size of the integer type.
9994 @cindex raw printing
9995 Print using the @samp{raw} formatting. By default, @value{GDBN} will
9996 use a Python-based pretty-printer, if one is available (@pxref{Pretty
9997 Printing}). This typically results in a higher-level display of the
9998 value's contents. The @samp{r} format bypasses any Python
9999 pretty-printer which might exist.
10002 For example, to print the program counter in hex (@pxref{Registers}), type
10009 Note that no space is required before the slash; this is because command
10010 names in @value{GDBN} cannot contain a slash.
10012 To reprint the last value in the value history with a different format,
10013 you can use the @code{print} command with just a format and no
10014 expression. For example, @samp{p/x} reprints the last value in hex.
10017 @section Examining Memory
10019 You can use the command @code{x} (for ``examine'') to examine memory in
10020 any of several formats, independently of your program's data types.
10022 @cindex examining memory
10024 @kindex x @r{(examine memory)}
10025 @item x/@var{nfu} @var{addr}
10026 @itemx x @var{addr}
10028 Use the @code{x} command to examine memory.
10031 @var{n}, @var{f}, and @var{u} are all optional parameters that specify how
10032 much memory to display and how to format it; @var{addr} is an
10033 expression giving the address where you want to start displaying memory.
10034 If you use defaults for @var{nfu}, you need not type the slash @samp{/}.
10035 Several commands set convenient defaults for @var{addr}.
10038 @item @var{n}, the repeat count
10039 The repeat count is a decimal integer; the default is 1. It specifies
10040 how much memory (counting by units @var{u}) to display. If a negative
10041 number is specified, memory is examined backward from @var{addr}.
10042 @c This really is **decimal**; unaffected by 'set radix' as of GDB
10045 @item @var{f}, the display format
10046 The display format is one of the formats used by @code{print}
10047 (@samp{x}, @samp{d}, @samp{u}, @samp{o}, @samp{t}, @samp{a}, @samp{c},
10048 @samp{f}, @samp{s}), and in addition @samp{i} (for machine instructions).
10049 The default is @samp{x} (hexadecimal) initially. The default changes
10050 each time you use either @code{x} or @code{print}.
10052 @item @var{u}, the unit size
10053 The unit size is any of
10059 Halfwords (two bytes).
10061 Words (four bytes). This is the initial default.
10063 Giant words (eight bytes).
10066 Each time you specify a unit size with @code{x}, that size becomes the
10067 default unit the next time you use @code{x}. For the @samp{i} format,
10068 the unit size is ignored and is normally not written. For the @samp{s} format,
10069 the unit size defaults to @samp{b}, unless it is explicitly given.
10070 Use @kbd{x /hs} to display 16-bit char strings and @kbd{x /ws} to display
10071 32-bit strings. The next use of @kbd{x /s} will again display 8-bit strings.
10072 Note that the results depend on the programming language of the
10073 current compilation unit. If the language is C, the @samp{s}
10074 modifier will use the UTF-16 encoding while @samp{w} will use
10075 UTF-32. The encoding is set by the programming language and cannot
10078 @item @var{addr}, starting display address
10079 @var{addr} is the address where you want @value{GDBN} to begin displaying
10080 memory. The expression need not have a pointer value (though it may);
10081 it is always interpreted as an integer address of a byte of memory.
10082 @xref{Expressions, ,Expressions}, for more information on expressions. The default for
10083 @var{addr} is usually just after the last address examined---but several
10084 other commands also set the default address: @code{info breakpoints} (to
10085 the address of the last breakpoint listed), @code{info line} (to the
10086 starting address of a line), and @code{print} (if you use it to display
10087 a value from memory).
10090 For example, @samp{x/3uh 0x54320} is a request to display three halfwords
10091 (@code{h}) of memory, formatted as unsigned decimal integers (@samp{u}),
10092 starting at address @code{0x54320}. @samp{x/4xw $sp} prints the four
10093 words (@samp{w}) of memory above the stack pointer (here, @samp{$sp};
10094 @pxref{Registers, ,Registers}) in hexadecimal (@samp{x}).
10096 You can also specify a negative repeat count to examine memory backward
10097 from the given address. For example, @samp{x/-3uh 0x54320} prints three
10098 halfwords (@code{h}) at @code{0x54314}, @code{0x54328}, and @code{0x5431c}.
10100 Since the letters indicating unit sizes are all distinct from the
10101 letters specifying output formats, you do not have to remember whether
10102 unit size or format comes first; either order works. The output
10103 specifications @samp{4xw} and @samp{4wx} mean exactly the same thing.
10104 (However, the count @var{n} must come first; @samp{wx4} does not work.)
10106 Even though the unit size @var{u} is ignored for the formats @samp{s}
10107 and @samp{i}, you might still want to use a count @var{n}; for example,
10108 @samp{3i} specifies that you want to see three machine instructions,
10109 including any operands. For convenience, especially when used with
10110 the @code{display} command, the @samp{i} format also prints branch delay
10111 slot instructions, if any, beyond the count specified, which immediately
10112 follow the last instruction that is within the count. The command
10113 @code{disassemble} gives an alternative way of inspecting machine
10114 instructions; see @ref{Machine Code,,Source and Machine Code}.
10116 If a negative repeat count is specified for the formats @samp{s} or @samp{i},
10117 the command displays null-terminated strings or instructions before the given
10118 address as many as the absolute value of the given number. For the @samp{i}
10119 format, we use line number information in the debug info to accurately locate
10120 instruction boundaries while disassembling backward. If line info is not
10121 available, the command stops examining memory with an error message.
10123 All the defaults for the arguments to @code{x} are designed to make it
10124 easy to continue scanning memory with minimal specifications each time
10125 you use @code{x}. For example, after you have inspected three machine
10126 instructions with @samp{x/3i @var{addr}}, you can inspect the next seven
10127 with just @samp{x/7}. If you use @key{RET} to repeat the @code{x} command,
10128 the repeat count @var{n} is used again; the other arguments default as
10129 for successive uses of @code{x}.
10131 When examining machine instructions, the instruction at current program
10132 counter is shown with a @code{=>} marker. For example:
10135 (@value{GDBP}) x/5i $pc-6
10136 0x804837f <main+11>: mov %esp,%ebp
10137 0x8048381 <main+13>: push %ecx
10138 0x8048382 <main+14>: sub $0x4,%esp
10139 => 0x8048385 <main+17>: movl $0x8048460,(%esp)
10140 0x804838c <main+24>: call 0x80482d4 <puts@@plt>
10143 @cindex @code{$_}, @code{$__}, and value history
10144 The addresses and contents printed by the @code{x} command are not saved
10145 in the value history because there is often too much of them and they
10146 would get in the way. Instead, @value{GDBN} makes these values available for
10147 subsequent use in expressions as values of the convenience variables
10148 @code{$_} and @code{$__}. After an @code{x} command, the last address
10149 examined is available for use in expressions in the convenience variable
10150 @code{$_}. The contents of that address, as examined, are available in
10151 the convenience variable @code{$__}.
10153 If the @code{x} command has a repeat count, the address and contents saved
10154 are from the last memory unit printed; this is not the same as the last
10155 address printed if several units were printed on the last line of output.
10157 @anchor{addressable memory unit}
10158 @cindex addressable memory unit
10159 Most targets have an addressable memory unit size of 8 bits. This means
10160 that to each memory address are associated 8 bits of data. Some
10161 targets, however, have other addressable memory unit sizes.
10162 Within @value{GDBN} and this document, the term
10163 @dfn{addressable memory unit} (or @dfn{memory unit} for short) is used
10164 when explicitly referring to a chunk of data of that size. The word
10165 @dfn{byte} is used to refer to a chunk of data of 8 bits, regardless of
10166 the addressable memory unit size of the target. For most systems,
10167 addressable memory unit is a synonym of byte.
10169 @cindex remote memory comparison
10170 @cindex target memory comparison
10171 @cindex verify remote memory image
10172 @cindex verify target memory image
10173 When you are debugging a program running on a remote target machine
10174 (@pxref{Remote Debugging}), you may wish to verify the program's image
10175 in the remote machine's memory against the executable file you
10176 downloaded to the target. Or, on any target, you may want to check
10177 whether the program has corrupted its own read-only sections. The
10178 @code{compare-sections} command is provided for such situations.
10181 @kindex compare-sections
10182 @item compare-sections @r{[}@var{section-name}@r{|}@code{-r}@r{]}
10183 Compare the data of a loadable section @var{section-name} in the
10184 executable file of the program being debugged with the same section in
10185 the target machine's memory, and report any mismatches. With no
10186 arguments, compares all loadable sections. With an argument of
10187 @code{-r}, compares all loadable read-only sections.
10189 Note: for remote targets, this command can be accelerated if the
10190 target supports computing the CRC checksum of a block of memory
10191 (@pxref{qCRC packet}).
10195 @section Automatic Display
10196 @cindex automatic display
10197 @cindex display of expressions
10199 If you find that you want to print the value of an expression frequently
10200 (to see how it changes), you might want to add it to the @dfn{automatic
10201 display list} so that @value{GDBN} prints its value each time your program stops.
10202 Each expression added to the list is given a number to identify it;
10203 to remove an expression from the list, you specify that number.
10204 The automatic display looks like this:
10208 3: bar[5] = (struct hack *) 0x3804
10212 This display shows item numbers, expressions and their current values. As with
10213 displays you request manually using @code{x} or @code{print}, you can
10214 specify the output format you prefer; in fact, @code{display} decides
10215 whether to use @code{print} or @code{x} depending your format
10216 specification---it uses @code{x} if you specify either the @samp{i}
10217 or @samp{s} format, or a unit size; otherwise it uses @code{print}.
10221 @item display @var{expr}
10222 Add the expression @var{expr} to the list of expressions to display
10223 each time your program stops. @xref{Expressions, ,Expressions}.
10225 @code{display} does not repeat if you press @key{RET} again after using it.
10227 @item display/@var{fmt} @var{expr}
10228 For @var{fmt} specifying only a display format and not a size or
10229 count, add the expression @var{expr} to the auto-display list but
10230 arrange to display it each time in the specified format @var{fmt}.
10231 @xref{Output Formats,,Output Formats}.
10233 @item display/@var{fmt} @var{addr}
10234 For @var{fmt} @samp{i} or @samp{s}, or including a unit-size or a
10235 number of units, add the expression @var{addr} as a memory address to
10236 be examined each time your program stops. Examining means in effect
10237 doing @samp{x/@var{fmt} @var{addr}}. @xref{Memory, ,Examining Memory}.
10240 For example, @samp{display/i $pc} can be helpful, to see the machine
10241 instruction about to be executed each time execution stops (@samp{$pc}
10242 is a common name for the program counter; @pxref{Registers, ,Registers}).
10245 @kindex delete display
10247 @item undisplay @var{dnums}@dots{}
10248 @itemx delete display @var{dnums}@dots{}
10249 Remove items from the list of expressions to display. Specify the
10250 numbers of the displays that you want affected with the command
10251 argument @var{dnums}. It can be a single display number, one of the
10252 numbers shown in the first field of the @samp{info display} display;
10253 or it could be a range of display numbers, as in @code{2-4}.
10255 @code{undisplay} does not repeat if you press @key{RET} after using it.
10256 (Otherwise you would just get the error @samp{No display number @dots{}}.)
10258 @kindex disable display
10259 @item disable display @var{dnums}@dots{}
10260 Disable the display of item numbers @var{dnums}. A disabled display
10261 item is not printed automatically, but is not forgotten. It may be
10262 enabled again later. Specify the numbers of the displays that you
10263 want affected with the command argument @var{dnums}. It can be a
10264 single display number, one of the numbers shown in the first field of
10265 the @samp{info display} display; or it could be a range of display
10266 numbers, as in @code{2-4}.
10268 @kindex enable display
10269 @item enable display @var{dnums}@dots{}
10270 Enable display of item numbers @var{dnums}. It becomes effective once
10271 again in auto display of its expression, until you specify otherwise.
10272 Specify the numbers of the displays that you want affected with the
10273 command argument @var{dnums}. It can be a single display number, one
10274 of the numbers shown in the first field of the @samp{info display}
10275 display; or it could be a range of display numbers, as in @code{2-4}.
10278 Display the current values of the expressions on the list, just as is
10279 done when your program stops.
10281 @kindex info display
10283 Print the list of expressions previously set up to display
10284 automatically, each one with its item number, but without showing the
10285 values. This includes disabled expressions, which are marked as such.
10286 It also includes expressions which would not be displayed right now
10287 because they refer to automatic variables not currently available.
10290 @cindex display disabled out of scope
10291 If a display expression refers to local variables, then it does not make
10292 sense outside the lexical context for which it was set up. Such an
10293 expression is disabled when execution enters a context where one of its
10294 variables is not defined. For example, if you give the command
10295 @code{display last_char} while inside a function with an argument
10296 @code{last_char}, @value{GDBN} displays this argument while your program
10297 continues to stop inside that function. When it stops elsewhere---where
10298 there is no variable @code{last_char}---the display is disabled
10299 automatically. The next time your program stops where @code{last_char}
10300 is meaningful, you can enable the display expression once again.
10302 @node Print Settings
10303 @section Print Settings
10305 @cindex format options
10306 @cindex print settings
10307 @value{GDBN} provides the following ways to control how arrays, structures,
10308 and symbols are printed.
10311 These settings are useful for debugging programs in any language:
10315 @item set print address
10316 @itemx set print address on
10317 @cindex print/don't print memory addresses
10318 @value{GDBN} prints memory addresses showing the location of stack
10319 traces, structure values, pointer values, breakpoints, and so forth,
10320 even when it also displays the contents of those addresses. The default
10321 is @code{on}. For example, this is what a stack frame display looks like with
10322 @code{set print address on}:
10327 #0 set_quotes (lq=0x34c78 "<<", rq=0x34c88 ">>")
10329 530 if (lquote != def_lquote)
10333 @item set print address off
10334 Do not print addresses when displaying their contents. For example,
10335 this is the same stack frame displayed with @code{set print address off}:
10339 (@value{GDBP}) set print addr off
10341 #0 set_quotes (lq="<<", rq=">>") at input.c:530
10342 530 if (lquote != def_lquote)
10346 You can use @samp{set print address off} to eliminate all machine
10347 dependent displays from the @value{GDBN} interface. For example, with
10348 @code{print address off}, you should get the same text for backtraces on
10349 all machines---whether or not they involve pointer arguments.
10352 @item show print address
10353 Show whether or not addresses are to be printed.
10356 When @value{GDBN} prints a symbolic address, it normally prints the
10357 closest earlier symbol plus an offset. If that symbol does not uniquely
10358 identify the address (for example, it is a name whose scope is a single
10359 source file), you may need to clarify. One way to do this is with
10360 @code{info line}, for example @samp{info line *0x4537}. Alternately,
10361 you can set @value{GDBN} to print the source file and line number when
10362 it prints a symbolic address:
10365 @item set print symbol-filename on
10366 @cindex source file and line of a symbol
10367 @cindex symbol, source file and line
10368 Tell @value{GDBN} to print the source file name and line number of a
10369 symbol in the symbolic form of an address.
10371 @item set print symbol-filename off
10372 Do not print source file name and line number of a symbol. This is the
10375 @item show print symbol-filename
10376 Show whether or not @value{GDBN} will print the source file name and
10377 line number of a symbol in the symbolic form of an address.
10380 Another situation where it is helpful to show symbol filenames and line
10381 numbers is when disassembling code; @value{GDBN} shows you the line
10382 number and source file that corresponds to each instruction.
10384 Also, you may wish to see the symbolic form only if the address being
10385 printed is reasonably close to the closest earlier symbol:
10388 @item set print max-symbolic-offset @var{max-offset}
10389 @itemx set print max-symbolic-offset unlimited
10390 @cindex maximum value for offset of closest symbol
10391 Tell @value{GDBN} to only display the symbolic form of an address if the
10392 offset between the closest earlier symbol and the address is less than
10393 @var{max-offset}. The default is @code{unlimited}, which tells @value{GDBN}
10394 to always print the symbolic form of an address if any symbol precedes
10395 it. Zero is equivalent to @code{unlimited}.
10397 @item show print max-symbolic-offset
10398 Ask how large the maximum offset is that @value{GDBN} prints in a
10402 @cindex wild pointer, interpreting
10403 @cindex pointer, finding referent
10404 If you have a pointer and you are not sure where it points, try
10405 @samp{set print symbol-filename on}. Then you can determine the name
10406 and source file location of the variable where it points, using
10407 @samp{p/a @var{pointer}}. This interprets the address in symbolic form.
10408 For example, here @value{GDBN} shows that a variable @code{ptt} points
10409 at another variable @code{t}, defined in @file{hi2.c}:
10412 (@value{GDBP}) set print symbol-filename on
10413 (@value{GDBP}) p/a ptt
10414 $4 = 0xe008 <t in hi2.c>
10418 @emph{Warning:} For pointers that point to a local variable, @samp{p/a}
10419 does not show the symbol name and filename of the referent, even with
10420 the appropriate @code{set print} options turned on.
10423 You can also enable @samp{/a}-like formatting all the time using
10424 @samp{set print symbol on}:
10427 @item set print symbol on
10428 Tell @value{GDBN} to print the symbol corresponding to an address, if
10431 @item set print symbol off
10432 Tell @value{GDBN} not to print the symbol corresponding to an
10433 address. In this mode, @value{GDBN} will still print the symbol
10434 corresponding to pointers to functions. This is the default.
10436 @item show print symbol
10437 Show whether @value{GDBN} will display the symbol corresponding to an
10441 Other settings control how different kinds of objects are printed:
10444 @item set print array
10445 @itemx set print array on
10446 @cindex pretty print arrays
10447 Pretty print arrays. This format is more convenient to read,
10448 but uses more space. The default is off.
10450 @item set print array off
10451 Return to compressed format for arrays.
10453 @item show print array
10454 Show whether compressed or pretty format is selected for displaying
10457 @cindex print array indexes
10458 @item set print array-indexes
10459 @itemx set print array-indexes on
10460 Print the index of each element when displaying arrays. May be more
10461 convenient to locate a given element in the array or quickly find the
10462 index of a given element in that printed array. The default is off.
10464 @item set print array-indexes off
10465 Stop printing element indexes when displaying arrays.
10467 @item show print array-indexes
10468 Show whether the index of each element is printed when displaying
10471 @item set print elements @var{number-of-elements}
10472 @itemx set print elements unlimited
10473 @cindex number of array elements to print
10474 @cindex limit on number of printed array elements
10475 Set a limit on how many elements of an array @value{GDBN} will print.
10476 If @value{GDBN} is printing a large array, it stops printing after it has
10477 printed the number of elements set by the @code{set print elements} command.
10478 This limit also applies to the display of strings.
10479 When @value{GDBN} starts, this limit is set to 200.
10480 Setting @var{number-of-elements} to @code{unlimited} or zero means
10481 that the number of elements to print is unlimited.
10483 @item show print elements
10484 Display the number of elements of a large array that @value{GDBN} will print.
10485 If the number is 0, then the printing is unlimited.
10487 @item set print frame-arguments @var{value}
10488 @kindex set print frame-arguments
10489 @cindex printing frame argument values
10490 @cindex print all frame argument values
10491 @cindex print frame argument values for scalars only
10492 @cindex do not print frame argument values
10493 This command allows to control how the values of arguments are printed
10494 when the debugger prints a frame (@pxref{Frames}). The possible
10499 The values of all arguments are printed.
10502 Print the value of an argument only if it is a scalar. The value of more
10503 complex arguments such as arrays, structures, unions, etc, is replaced
10504 by @code{@dots{}}. This is the default. Here is an example where
10505 only scalar arguments are shown:
10508 #1 0x08048361 in call_me (i=3, s=@dots{}, ss=0xbf8d508c, u=@dots{}, e=green)
10513 None of the argument values are printed. Instead, the value of each argument
10514 is replaced by @code{@dots{}}. In this case, the example above now becomes:
10517 #1 0x08048361 in call_me (i=@dots{}, s=@dots{}, ss=@dots{}, u=@dots{}, e=@dots{})
10522 By default, only scalar arguments are printed. This command can be used
10523 to configure the debugger to print the value of all arguments, regardless
10524 of their type. However, it is often advantageous to not print the value
10525 of more complex parameters. For instance, it reduces the amount of
10526 information printed in each frame, making the backtrace more readable.
10527 Also, it improves performance when displaying Ada frames, because
10528 the computation of large arguments can sometimes be CPU-intensive,
10529 especially in large applications. Setting @code{print frame-arguments}
10530 to @code{scalars} (the default) or @code{none} avoids this computation,
10531 thus speeding up the display of each Ada frame.
10533 @item show print frame-arguments
10534 Show how the value of arguments should be displayed when printing a frame.
10536 @item set print raw frame-arguments on
10537 Print frame arguments in raw, non pretty-printed, form.
10539 @item set print raw frame-arguments off
10540 Print frame arguments in pretty-printed form, if there is a pretty-printer
10541 for the value (@pxref{Pretty Printing}),
10542 otherwise print the value in raw form.
10543 This is the default.
10545 @item show print raw frame-arguments
10546 Show whether to print frame arguments in raw form.
10548 @anchor{set print entry-values}
10549 @item set print entry-values @var{value}
10550 @kindex set print entry-values
10551 Set printing of frame argument values at function entry. In some cases
10552 @value{GDBN} can determine the value of function argument which was passed by
10553 the function caller, even if the value was modified inside the called function
10554 and therefore is different. With optimized code, the current value could be
10555 unavailable, but the entry value may still be known.
10557 The default value is @code{default} (see below for its description). Older
10558 @value{GDBN} behaved as with the setting @code{no}. Compilers not supporting
10559 this feature will behave in the @code{default} setting the same way as with the
10562 This functionality is currently supported only by DWARF 2 debugging format and
10563 the compiler has to produce @samp{DW_TAG_call_site} tags. With
10564 @value{NGCC}, you need to specify @option{-O -g} during compilation, to get
10567 The @var{value} parameter can be one of the following:
10571 Print only actual parameter values, never print values from function entry
10575 #0 different (val=6)
10576 #0 lost (val=<optimized out>)
10578 #0 invalid (val=<optimized out>)
10582 Print only parameter values from function entry point. The actual parameter
10583 values are never printed.
10585 #0 equal (val@@entry=5)
10586 #0 different (val@@entry=5)
10587 #0 lost (val@@entry=5)
10588 #0 born (val@@entry=<optimized out>)
10589 #0 invalid (val@@entry=<optimized out>)
10593 Print only parameter values from function entry point. If value from function
10594 entry point is not known while the actual value is known, print the actual
10595 value for such parameter.
10597 #0 equal (val@@entry=5)
10598 #0 different (val@@entry=5)
10599 #0 lost (val@@entry=5)
10601 #0 invalid (val@@entry=<optimized out>)
10605 Print actual parameter values. If actual parameter value is not known while
10606 value from function entry point is known, print the entry point value for such
10610 #0 different (val=6)
10611 #0 lost (val@@entry=5)
10613 #0 invalid (val=<optimized out>)
10617 Always print both the actual parameter value and its value from function entry
10618 point, even if values of one or both are not available due to compiler
10621 #0 equal (val=5, val@@entry=5)
10622 #0 different (val=6, val@@entry=5)
10623 #0 lost (val=<optimized out>, val@@entry=5)
10624 #0 born (val=10, val@@entry=<optimized out>)
10625 #0 invalid (val=<optimized out>, val@@entry=<optimized out>)
10629 Print the actual parameter value if it is known and also its value from
10630 function entry point if it is known. If neither is known, print for the actual
10631 value @code{<optimized out>}. If not in MI mode (@pxref{GDB/MI}) and if both
10632 values are known and identical, print the shortened
10633 @code{param=param@@entry=VALUE} notation.
10635 #0 equal (val=val@@entry=5)
10636 #0 different (val=6, val@@entry=5)
10637 #0 lost (val@@entry=5)
10639 #0 invalid (val=<optimized out>)
10643 Always print the actual parameter value. Print also its value from function
10644 entry point, but only if it is known. If not in MI mode (@pxref{GDB/MI}) and
10645 if both values are known and identical, print the shortened
10646 @code{param=param@@entry=VALUE} notation.
10648 #0 equal (val=val@@entry=5)
10649 #0 different (val=6, val@@entry=5)
10650 #0 lost (val=<optimized out>, val@@entry=5)
10652 #0 invalid (val=<optimized out>)
10656 For analysis messages on possible failures of frame argument values at function
10657 entry resolution see @ref{set debug entry-values}.
10659 @item show print entry-values
10660 Show the method being used for printing of frame argument values at function
10663 @item set print repeats @var{number-of-repeats}
10664 @itemx set print repeats unlimited
10665 @cindex repeated array elements
10666 Set the threshold for suppressing display of repeated array
10667 elements. When the number of consecutive identical elements of an
10668 array exceeds the threshold, @value{GDBN} prints the string
10669 @code{"<repeats @var{n} times>"}, where @var{n} is the number of
10670 identical repetitions, instead of displaying the identical elements
10671 themselves. Setting the threshold to @code{unlimited} or zero will
10672 cause all elements to be individually printed. The default threshold
10675 @item show print repeats
10676 Display the current threshold for printing repeated identical
10679 @item set print max-depth @var{depth}
10680 @item set print max-depth unlimited
10681 @cindex printing nested structures
10682 Set the threshold after which nested structures are replaced with
10683 ellipsis, this can make visualising deeply nested structures easier.
10685 For example, given this C code
10688 typedef struct s1 @{ int a; @} s1;
10689 typedef struct s2 @{ s1 b; @} s2;
10690 typedef struct s3 @{ s2 c; @} s3;
10691 typedef struct s4 @{ s3 d; @} s4;
10693 s4 var = @{ @{ @{ @{ 3 @} @} @} @};
10696 The following table shows how different values of @var{depth} will
10697 effect how @code{var} is printed by @value{GDBN}:
10699 @multitable @columnfractions .3 .7
10700 @headitem @var{depth} setting @tab Result of @samp{p var}
10702 @tab @code{$1 = @{d = @{c = @{b = @{a = 3@}@}@}@}}
10704 @tab @code{$1 = @{...@}}
10706 @tab @code{$1 = @{d = @{...@}@}}
10708 @tab @code{$1 = @{d = @{c = @{...@}@}@}}
10710 @tab @code{$1 = @{d = @{c = @{b = @{...@}@}@}@}}
10712 @tab @code{$1 = @{d = @{c = @{b = @{a = 3@}@}@}@}}
10715 To see the contents of structures that have been hidden the user can
10716 either increase the print max-depth, or they can print the elements of
10717 the structure that are visible, for example
10720 (gdb) set print max-depth 2
10722 $1 = @{d = @{c = @{...@}@}@}
10724 $2 = @{c = @{b = @{...@}@}@}
10726 $3 = @{b = @{a = 3@}@}
10729 The pattern used to replace nested structures varies based on
10730 language, for most languages @code{@{...@}} is used, but Fortran uses
10733 @item show print max-depth
10734 Display the current threshold after which nested structures are
10735 replaces with ellipsis.
10737 @item set print null-stop
10738 @cindex @sc{null} elements in arrays
10739 Cause @value{GDBN} to stop printing the characters of an array when the first
10740 @sc{null} is encountered. This is useful when large arrays actually
10741 contain only short strings.
10742 The default is off.
10744 @item show print null-stop
10745 Show whether @value{GDBN} stops printing an array on the first
10746 @sc{null} character.
10748 @item set print pretty on
10749 @cindex print structures in indented form
10750 @cindex indentation in structure display
10751 Cause @value{GDBN} to print structures in an indented format with one member
10752 per line, like this:
10767 @item set print pretty off
10768 Cause @value{GDBN} to print structures in a compact format, like this:
10772 $1 = @{next = 0x0, flags = @{sweet = 1, sour = 1@}, \
10773 meat = 0x54 "Pork"@}
10778 This is the default format.
10780 @item show print pretty
10781 Show which format @value{GDBN} is using to print structures.
10783 @item set print sevenbit-strings on
10784 @cindex eight-bit characters in strings
10785 @cindex octal escapes in strings
10786 Print using only seven-bit characters; if this option is set,
10787 @value{GDBN} displays any eight-bit characters (in strings or
10788 character values) using the notation @code{\}@var{nnn}. This setting is
10789 best if you are working in English (@sc{ascii}) and you use the
10790 high-order bit of characters as a marker or ``meta'' bit.
10792 @item set print sevenbit-strings off
10793 Print full eight-bit characters. This allows the use of more
10794 international character sets, and is the default.
10796 @item show print sevenbit-strings
10797 Show whether or not @value{GDBN} is printing only seven-bit characters.
10799 @item set print union on
10800 @cindex unions in structures, printing
10801 Tell @value{GDBN} to print unions which are contained in structures
10802 and other unions. This is the default setting.
10804 @item set print union off
10805 Tell @value{GDBN} not to print unions which are contained in
10806 structures and other unions. @value{GDBN} will print @code{"@{...@}"}
10809 @item show print union
10810 Ask @value{GDBN} whether or not it will print unions which are contained in
10811 structures and other unions.
10813 For example, given the declarations
10816 typedef enum @{Tree, Bug@} Species;
10817 typedef enum @{Big_tree, Acorn, Seedling@} Tree_forms;
10818 typedef enum @{Caterpillar, Cocoon, Butterfly@}
10829 struct thing foo = @{Tree, @{Acorn@}@};
10833 with @code{set print union on} in effect @samp{p foo} would print
10836 $1 = @{it = Tree, form = @{tree = Acorn, bug = Cocoon@}@}
10840 and with @code{set print union off} in effect it would print
10843 $1 = @{it = Tree, form = @{...@}@}
10847 @code{set print union} affects programs written in C-like languages
10853 These settings are of interest when debugging C@t{++} programs:
10856 @cindex demangling C@t{++} names
10857 @item set print demangle
10858 @itemx set print demangle on
10859 Print C@t{++} names in their source form rather than in the encoded
10860 (``mangled'') form passed to the assembler and linker for type-safe
10861 linkage. The default is on.
10863 @item show print demangle
10864 Show whether C@t{++} names are printed in mangled or demangled form.
10866 @item set print asm-demangle
10867 @itemx set print asm-demangle on
10868 Print C@t{++} names in their source form rather than their mangled form, even
10869 in assembler code printouts such as instruction disassemblies.
10870 The default is off.
10872 @item show print asm-demangle
10873 Show whether C@t{++} names in assembly listings are printed in mangled
10876 @cindex C@t{++} symbol decoding style
10877 @cindex symbol decoding style, C@t{++}
10878 @kindex set demangle-style
10879 @item set demangle-style @var{style}
10880 Choose among several encoding schemes used by different compilers to represent
10881 C@t{++} names. If you omit @var{style}, you will see a list of possible
10882 formats. The default value is @var{auto}, which lets @value{GDBN} choose a
10883 decoding style by inspecting your program.
10885 @item show demangle-style
10886 Display the encoding style currently in use for decoding C@t{++} symbols.
10888 @item set print object
10889 @itemx set print object on
10890 @cindex derived type of an object, printing
10891 @cindex display derived types
10892 When displaying a pointer to an object, identify the @emph{actual}
10893 (derived) type of the object rather than the @emph{declared} type, using
10894 the virtual function table. Note that the virtual function table is
10895 required---this feature can only work for objects that have run-time
10896 type identification; a single virtual method in the object's declared
10897 type is sufficient. Note that this setting is also taken into account when
10898 working with variable objects via MI (@pxref{GDB/MI}).
10900 @item set print object off
10901 Display only the declared type of objects, without reference to the
10902 virtual function table. This is the default setting.
10904 @item show print object
10905 Show whether actual, or declared, object types are displayed.
10907 @item set print static-members
10908 @itemx set print static-members on
10909 @cindex static members of C@t{++} objects
10910 Print static members when displaying a C@t{++} object. The default is on.
10912 @item set print static-members off
10913 Do not print static members when displaying a C@t{++} object.
10915 @item show print static-members
10916 Show whether C@t{++} static members are printed or not.
10918 @item set print pascal_static-members
10919 @itemx set print pascal_static-members on
10920 @cindex static members of Pascal objects
10921 @cindex Pascal objects, static members display
10922 Print static members when displaying a Pascal object. The default is on.
10924 @item set print pascal_static-members off
10925 Do not print static members when displaying a Pascal object.
10927 @item show print pascal_static-members
10928 Show whether Pascal static members are printed or not.
10930 @c These don't work with HP ANSI C++ yet.
10931 @item set print vtbl
10932 @itemx set print vtbl on
10933 @cindex pretty print C@t{++} virtual function tables
10934 @cindex virtual functions (C@t{++}) display
10935 @cindex VTBL display
10936 Pretty print C@t{++} virtual function tables. The default is off.
10937 (The @code{vtbl} commands do not work on programs compiled with the HP
10938 ANSI C@t{++} compiler (@code{aCC}).)
10940 @item set print vtbl off
10941 Do not pretty print C@t{++} virtual function tables.
10943 @item show print vtbl
10944 Show whether C@t{++} virtual function tables are pretty printed, or not.
10947 @node Pretty Printing
10948 @section Pretty Printing
10950 @value{GDBN} provides a mechanism to allow pretty-printing of values using
10951 Python code. It greatly simplifies the display of complex objects. This
10952 mechanism works for both MI and the CLI.
10955 * Pretty-Printer Introduction:: Introduction to pretty-printers
10956 * Pretty-Printer Example:: An example pretty-printer
10957 * Pretty-Printer Commands:: Pretty-printer commands
10960 @node Pretty-Printer Introduction
10961 @subsection Pretty-Printer Introduction
10963 When @value{GDBN} prints a value, it first sees if there is a pretty-printer
10964 registered for the value. If there is then @value{GDBN} invokes the
10965 pretty-printer to print the value. Otherwise the value is printed normally.
10967 Pretty-printers are normally named. This makes them easy to manage.
10968 The @samp{info pretty-printer} command will list all the installed
10969 pretty-printers with their names.
10970 If a pretty-printer can handle multiple data types, then its
10971 @dfn{subprinters} are the printers for the individual data types.
10972 Each such subprinter has its own name.
10973 The format of the name is @var{printer-name};@var{subprinter-name}.
10975 Pretty-printers are installed by @dfn{registering} them with @value{GDBN}.
10976 Typically they are automatically loaded and registered when the corresponding
10977 debug information is loaded, thus making them available without having to
10978 do anything special.
10980 There are three places where a pretty-printer can be registered.
10984 Pretty-printers registered globally are available when debugging
10988 Pretty-printers registered with a program space are available only
10989 when debugging that program.
10990 @xref{Progspaces In Python}, for more details on program spaces in Python.
10993 Pretty-printers registered with an objfile are loaded and unloaded
10994 with the corresponding objfile (e.g., shared library).
10995 @xref{Objfiles In Python}, for more details on objfiles in Python.
10998 @xref{Selecting Pretty-Printers}, for further information on how
10999 pretty-printers are selected,
11001 @xref{Writing a Pretty-Printer}, for implementing pretty printers
11004 @node Pretty-Printer Example
11005 @subsection Pretty-Printer Example
11007 Here is how a C@t{++} @code{std::string} looks without a pretty-printer:
11010 (@value{GDBP}) print s
11012 static npos = 4294967295,
11014 <std::allocator<char>> = @{
11015 <__gnu_cxx::new_allocator<char>> = @{
11016 <No data fields>@}, <No data fields>
11018 members of std::basic_string<char, std::char_traits<char>,
11019 std::allocator<char> >::_Alloc_hider:
11020 _M_p = 0x804a014 "abcd"
11025 With a pretty-printer for @code{std::string} only the contents are printed:
11028 (@value{GDBP}) print s
11032 @node Pretty-Printer Commands
11033 @subsection Pretty-Printer Commands
11034 @cindex pretty-printer commands
11037 @kindex info pretty-printer
11038 @item info pretty-printer [@var{object-regexp} [@var{name-regexp}]]
11039 Print the list of installed pretty-printers.
11040 This includes disabled pretty-printers, which are marked as such.
11042 @var{object-regexp} is a regular expression matching the objects
11043 whose pretty-printers to list.
11044 Objects can be @code{global}, the program space's file
11045 (@pxref{Progspaces In Python}),
11046 and the object files within that program space (@pxref{Objfiles In Python}).
11047 @xref{Selecting Pretty-Printers}, for details on how @value{GDBN}
11048 looks up a printer from these three objects.
11050 @var{name-regexp} is a regular expression matching the name of the printers
11053 @kindex disable pretty-printer
11054 @item disable pretty-printer [@var{object-regexp} [@var{name-regexp}]]
11055 Disable pretty-printers matching @var{object-regexp} and @var{name-regexp}.
11056 A disabled pretty-printer is not forgotten, it may be enabled again later.
11058 @kindex enable pretty-printer
11059 @item enable pretty-printer [@var{object-regexp} [@var{name-regexp}]]
11060 Enable pretty-printers matching @var{object-regexp} and @var{name-regexp}.
11065 Suppose we have three pretty-printers installed: one from library1.so
11066 named @code{foo} that prints objects of type @code{foo}, and
11067 another from library2.so named @code{bar} that prints two types of objects,
11068 @code{bar1} and @code{bar2}.
11071 (gdb) info pretty-printer
11078 (gdb) info pretty-printer library2
11083 (gdb) disable pretty-printer library1
11085 2 of 3 printers enabled
11086 (gdb) info pretty-printer
11093 (gdb) disable pretty-printer library2 bar;bar1
11095 1 of 3 printers enabled
11096 (gdb) info pretty-printer library2
11103 (gdb) disable pretty-printer library2 bar
11105 0 of 3 printers enabled
11106 (gdb) info pretty-printer library2
11115 Note that for @code{bar} the entire printer can be disabled,
11116 as can each individual subprinter.
11118 @node Value History
11119 @section Value History
11121 @cindex value history
11122 @cindex history of values printed by @value{GDBN}
11123 Values printed by the @code{print} command are saved in the @value{GDBN}
11124 @dfn{value history}. This allows you to refer to them in other expressions.
11125 Values are kept until the symbol table is re-read or discarded
11126 (for example with the @code{file} or @code{symbol-file} commands).
11127 When the symbol table changes, the value history is discarded,
11128 since the values may contain pointers back to the types defined in the
11133 @cindex history number
11134 The values printed are given @dfn{history numbers} by which you can
11135 refer to them. These are successive integers starting with one.
11136 @code{print} shows you the history number assigned to a value by
11137 printing @samp{$@var{num} = } before the value; here @var{num} is the
11140 To refer to any previous value, use @samp{$} followed by the value's
11141 history number. The way @code{print} labels its output is designed to
11142 remind you of this. Just @code{$} refers to the most recent value in
11143 the history, and @code{$$} refers to the value before that.
11144 @code{$$@var{n}} refers to the @var{n}th value from the end; @code{$$2}
11145 is the value just prior to @code{$$}, @code{$$1} is equivalent to
11146 @code{$$}, and @code{$$0} is equivalent to @code{$}.
11148 For example, suppose you have just printed a pointer to a structure and
11149 want to see the contents of the structure. It suffices to type
11155 If you have a chain of structures where the component @code{next} points
11156 to the next one, you can print the contents of the next one with this:
11163 You can print successive links in the chain by repeating this
11164 command---which you can do by just typing @key{RET}.
11166 Note that the history records values, not expressions. If the value of
11167 @code{x} is 4 and you type these commands:
11175 then the value recorded in the value history by the @code{print} command
11176 remains 4 even though the value of @code{x} has changed.
11179 @kindex show values
11181 Print the last ten values in the value history, with their item numbers.
11182 This is like @samp{p@ $$9} repeated ten times, except that @code{show
11183 values} does not change the history.
11185 @item show values @var{n}
11186 Print ten history values centered on history item number @var{n}.
11188 @item show values +
11189 Print ten history values just after the values last printed. If no more
11190 values are available, @code{show values +} produces no display.
11193 Pressing @key{RET} to repeat @code{show values @var{n}} has exactly the
11194 same effect as @samp{show values +}.
11196 @node Convenience Vars
11197 @section Convenience Variables
11199 @cindex convenience variables
11200 @cindex user-defined variables
11201 @value{GDBN} provides @dfn{convenience variables} that you can use within
11202 @value{GDBN} to hold on to a value and refer to it later. These variables
11203 exist entirely within @value{GDBN}; they are not part of your program, and
11204 setting a convenience variable has no direct effect on further execution
11205 of your program. That is why you can use them freely.
11207 Convenience variables are prefixed with @samp{$}. Any name preceded by
11208 @samp{$} can be used for a convenience variable, unless it is one of
11209 the predefined machine-specific register names (@pxref{Registers, ,Registers}).
11210 (Value history references, in contrast, are @emph{numbers} preceded
11211 by @samp{$}. @xref{Value History, ,Value History}.)
11213 You can save a value in a convenience variable with an assignment
11214 expression, just as you would set a variable in your program.
11218 set $foo = *object_ptr
11222 would save in @code{$foo} the value contained in the object pointed to by
11225 Using a convenience variable for the first time creates it, but its
11226 value is @code{void} until you assign a new value. You can alter the
11227 value with another assignment at any time.
11229 Convenience variables have no fixed types. You can assign a convenience
11230 variable any type of value, including structures and arrays, even if
11231 that variable already has a value of a different type. The convenience
11232 variable, when used as an expression, has the type of its current value.
11235 @kindex show convenience
11236 @cindex show all user variables and functions
11237 @item show convenience
11238 Print a list of convenience variables used so far, and their values,
11239 as well as a list of the convenience functions.
11240 Abbreviated @code{show conv}.
11242 @kindex init-if-undefined
11243 @cindex convenience variables, initializing
11244 @item init-if-undefined $@var{variable} = @var{expression}
11245 Set a convenience variable if it has not already been set. This is useful
11246 for user-defined commands that keep some state. It is similar, in concept,
11247 to using local static variables with initializers in C (except that
11248 convenience variables are global). It can also be used to allow users to
11249 override default values used in a command script.
11251 If the variable is already defined then the expression is not evaluated so
11252 any side-effects do not occur.
11255 One of the ways to use a convenience variable is as a counter to be
11256 incremented or a pointer to be advanced. For example, to print
11257 a field from successive elements of an array of structures:
11261 print bar[$i++]->contents
11265 Repeat that command by typing @key{RET}.
11267 Some convenience variables are created automatically by @value{GDBN} and given
11268 values likely to be useful.
11271 @vindex $_@r{, convenience variable}
11273 The variable @code{$_} is automatically set by the @code{x} command to
11274 the last address examined (@pxref{Memory, ,Examining Memory}). Other
11275 commands which provide a default address for @code{x} to examine also
11276 set @code{$_} to that address; these commands include @code{info line}
11277 and @code{info breakpoint}. The type of @code{$_} is @code{void *}
11278 except when set by the @code{x} command, in which case it is a pointer
11279 to the type of @code{$__}.
11281 @vindex $__@r{, convenience variable}
11283 The variable @code{$__} is automatically set by the @code{x} command
11284 to the value found in the last address examined. Its type is chosen
11285 to match the format in which the data was printed.
11288 @vindex $_exitcode@r{, convenience variable}
11289 When the program being debugged terminates normally, @value{GDBN}
11290 automatically sets this variable to the exit code of the program, and
11291 resets @code{$_exitsignal} to @code{void}.
11294 @vindex $_exitsignal@r{, convenience variable}
11295 When the program being debugged dies due to an uncaught signal,
11296 @value{GDBN} automatically sets this variable to that signal's number,
11297 and resets @code{$_exitcode} to @code{void}.
11299 To distinguish between whether the program being debugged has exited
11300 (i.e., @code{$_exitcode} is not @code{void}) or signalled (i.e.,
11301 @code{$_exitsignal} is not @code{void}), the convenience function
11302 @code{$_isvoid} can be used (@pxref{Convenience Funs,, Convenience
11303 Functions}). For example, considering the following source code:
11306 #include <signal.h>
11309 main (int argc, char *argv[])
11316 A valid way of telling whether the program being debugged has exited
11317 or signalled would be:
11320 (@value{GDBP}) define has_exited_or_signalled
11321 Type commands for definition of ``has_exited_or_signalled''.
11322 End with a line saying just ``end''.
11323 >if $_isvoid ($_exitsignal)
11324 >echo The program has exited\n
11326 >echo The program has signalled\n
11332 Program terminated with signal SIGALRM, Alarm clock.
11333 The program no longer exists.
11334 (@value{GDBP}) has_exited_or_signalled
11335 The program has signalled
11338 As can be seen, @value{GDBN} correctly informs that the program being
11339 debugged has signalled, since it calls @code{raise} and raises a
11340 @code{SIGALRM} signal. If the program being debugged had not called
11341 @code{raise}, then @value{GDBN} would report a normal exit:
11344 (@value{GDBP}) has_exited_or_signalled
11345 The program has exited
11349 The variable @code{$_exception} is set to the exception object being
11350 thrown at an exception-related catchpoint. @xref{Set Catchpoints}.
11353 @itemx $_probe_arg0@dots{}$_probe_arg11
11354 Arguments to a static probe. @xref{Static Probe Points}.
11357 @vindex $_sdata@r{, inspect, convenience variable}
11358 The variable @code{$_sdata} contains extra collected static tracepoint
11359 data. @xref{Tracepoint Actions,,Tracepoint Action Lists}. Note that
11360 @code{$_sdata} could be empty, if not inspecting a trace buffer, or
11361 if extra static tracepoint data has not been collected.
11364 @vindex $_siginfo@r{, convenience variable}
11365 The variable @code{$_siginfo} contains extra signal information
11366 (@pxref{extra signal information}). Note that @code{$_siginfo}
11367 could be empty, if the application has not yet received any signals.
11368 For example, it will be empty before you execute the @code{run} command.
11371 @vindex $_tlb@r{, convenience variable}
11372 The variable @code{$_tlb} is automatically set when debugging
11373 applications running on MS-Windows in native mode or connected to
11374 gdbserver that supports the @code{qGetTIBAddr} request.
11375 @xref{General Query Packets}.
11376 This variable contains the address of the thread information block.
11379 The number of the current inferior. @xref{Inferiors and
11380 Programs, ,Debugging Multiple Inferiors and Programs}.
11383 The thread number of the current thread. @xref{thread numbers}.
11386 The global number of the current thread. @xref{global thread numbers}.
11390 @vindex $_gdb_major@r{, convenience variable}
11391 @vindex $_gdb_minor@r{, convenience variable}
11392 The major and minor version numbers of the running @value{GDBN}.
11393 Development snapshots and pretest versions have their minor version
11394 incremented by one; thus, @value{GDBN} pretest 9.11.90 will produce
11395 the value 12 for @code{$_gdb_minor}. These variables allow you to
11396 write scripts that work with different versions of @value{GDBN}
11397 without errors caused by features unavailable in some of those
11400 @item $_shell_exitcode
11401 @itemx $_shell_exitsignal
11402 @vindex $_shell_exitcode@r{, convenience variable}
11403 @vindex $_shell_exitsignal@r{, convenience variable}
11404 @cindex shell command, exit code
11405 @cindex shell command, exit signal
11406 @cindex exit status of shell commands
11407 @value{GDBN} commands such as @code{shell} and @code{|} are launching
11408 shell commands. When a launched command terminates, @value{GDBN}
11409 automatically maintains the variables @code{$_shell_exitcode}
11410 and @code{$_shell_exitsignal} according to the exit status of the last
11411 launched command. These variables are set and used similarly to
11412 the variables @code{$_exitcode} and @code{$_exitsignal}.
11416 @node Convenience Funs
11417 @section Convenience Functions
11419 @cindex convenience functions
11420 @value{GDBN} also supplies some @dfn{convenience functions}. These
11421 have a syntax similar to convenience variables. A convenience
11422 function can be used in an expression just like an ordinary function;
11423 however, a convenience function is implemented internally to
11426 These functions do not require @value{GDBN} to be configured with
11427 @code{Python} support, which means that they are always available.
11431 @item $_isvoid (@var{expr})
11432 @findex $_isvoid@r{, convenience function}
11433 Return one if the expression @var{expr} is @code{void}. Otherwise it
11436 A @code{void} expression is an expression where the type of the result
11437 is @code{void}. For example, you can examine a convenience variable
11438 (see @ref{Convenience Vars,, Convenience Variables}) to check whether
11442 (@value{GDBP}) print $_exitcode
11444 (@value{GDBP}) print $_isvoid ($_exitcode)
11447 Starting program: ./a.out
11448 [Inferior 1 (process 29572) exited normally]
11449 (@value{GDBP}) print $_exitcode
11451 (@value{GDBP}) print $_isvoid ($_exitcode)
11455 In the example above, we used @code{$_isvoid} to check whether
11456 @code{$_exitcode} is @code{void} before and after the execution of the
11457 program being debugged. Before the execution there is no exit code to
11458 be examined, therefore @code{$_exitcode} is @code{void}. After the
11459 execution the program being debugged returned zero, therefore
11460 @code{$_exitcode} is zero, which means that it is not @code{void}
11463 The @code{void} expression can also be a call of a function from the
11464 program being debugged. For example, given the following function:
11473 The result of calling it inside @value{GDBN} is @code{void}:
11476 (@value{GDBP}) print foo ()
11478 (@value{GDBP}) print $_isvoid (foo ())
11480 (@value{GDBP}) set $v = foo ()
11481 (@value{GDBP}) print $v
11483 (@value{GDBP}) print $_isvoid ($v)
11489 These functions require @value{GDBN} to be configured with
11490 @code{Python} support.
11494 @item $_memeq(@var{buf1}, @var{buf2}, @var{length})
11495 @findex $_memeq@r{, convenience function}
11496 Returns one if the @var{length} bytes at the addresses given by
11497 @var{buf1} and @var{buf2} are equal.
11498 Otherwise it returns zero.
11500 @item $_regex(@var{str}, @var{regex})
11501 @findex $_regex@r{, convenience function}
11502 Returns one if the string @var{str} matches the regular expression
11503 @var{regex}. Otherwise it returns zero.
11504 The syntax of the regular expression is that specified by @code{Python}'s
11505 regular expression support.
11507 @item $_streq(@var{str1}, @var{str2})
11508 @findex $_streq@r{, convenience function}
11509 Returns one if the strings @var{str1} and @var{str2} are equal.
11510 Otherwise it returns zero.
11512 @item $_strlen(@var{str})
11513 @findex $_strlen@r{, convenience function}
11514 Returns the length of string @var{str}.
11516 @item $_caller_is(@var{name}@r{[}, @var{number_of_frames}@r{]})
11517 @findex $_caller_is@r{, convenience function}
11518 Returns one if the calling function's name is equal to @var{name}.
11519 Otherwise it returns zero.
11521 If the optional argument @var{number_of_frames} is provided,
11522 it is the number of frames up in the stack to look.
11530 at testsuite/gdb.python/py-caller-is.c:21
11531 #1 0x00000000004005a0 in middle_func ()
11532 at testsuite/gdb.python/py-caller-is.c:27
11533 #2 0x00000000004005ab in top_func ()
11534 at testsuite/gdb.python/py-caller-is.c:33
11535 #3 0x00000000004005b6 in main ()
11536 at testsuite/gdb.python/py-caller-is.c:39
11537 (gdb) print $_caller_is ("middle_func")
11539 (gdb) print $_caller_is ("top_func", 2)
11543 @item $_caller_matches(@var{regexp}@r{[}, @var{number_of_frames}@r{]})
11544 @findex $_caller_matches@r{, convenience function}
11545 Returns one if the calling function's name matches the regular expression
11546 @var{regexp}. Otherwise it returns zero.
11548 If the optional argument @var{number_of_frames} is provided,
11549 it is the number of frames up in the stack to look.
11552 @item $_any_caller_is(@var{name}@r{[}, @var{number_of_frames}@r{]})
11553 @findex $_any_caller_is@r{, convenience function}
11554 Returns one if any calling function's name is equal to @var{name}.
11555 Otherwise it returns zero.
11557 If the optional argument @var{number_of_frames} is provided,
11558 it is the number of frames up in the stack to look.
11561 This function differs from @code{$_caller_is} in that this function
11562 checks all stack frames from the immediate caller to the frame specified
11563 by @var{number_of_frames}, whereas @code{$_caller_is} only checks the
11564 frame specified by @var{number_of_frames}.
11566 @item $_any_caller_matches(@var{regexp}@r{[}, @var{number_of_frames}@r{]})
11567 @findex $_any_caller_matches@r{, convenience function}
11568 Returns one if any calling function's name matches the regular expression
11569 @var{regexp}. Otherwise it returns zero.
11571 If the optional argument @var{number_of_frames} is provided,
11572 it is the number of frames up in the stack to look.
11575 This function differs from @code{$_caller_matches} in that this function
11576 checks all stack frames from the immediate caller to the frame specified
11577 by @var{number_of_frames}, whereas @code{$_caller_matches} only checks the
11578 frame specified by @var{number_of_frames}.
11580 @item $_as_string(@var{value})
11581 @findex $_as_string@r{, convenience function}
11582 Return the string representation of @var{value}.
11584 This function is useful to obtain the textual label (enumerator) of an
11585 enumeration value. For example, assuming the variable @var{node} is of
11586 an enumerated type:
11589 (gdb) printf "Visiting node of type %s\n", $_as_string(node)
11590 Visiting node of type NODE_INTEGER
11593 @item $_cimag(@var{value})
11594 @itemx $_creal(@var{value})
11595 @findex $_cimag@r{, convenience function}
11596 @findex $_creal@r{, convenience function}
11597 Return the imaginary (@code{$_cimag}) or real (@code{$_creal}) part of
11598 the complex number @var{value}.
11600 The type of the imaginary or real part depends on the type of the
11601 complex number, e.g., using @code{$_cimag} on a @code{float complex}
11602 will return an imaginary part of type @code{float}.
11606 @value{GDBN} provides the ability to list and get help on
11607 convenience functions.
11610 @item help function
11611 @kindex help function
11612 @cindex show all convenience functions
11613 Print a list of all convenience functions.
11620 You can refer to machine register contents, in expressions, as variables
11621 with names starting with @samp{$}. The names of registers are different
11622 for each machine; use @code{info registers} to see the names used on
11626 @kindex info registers
11627 @item info registers
11628 Print the names and values of all registers except floating-point
11629 and vector registers (in the selected stack frame).
11631 @kindex info all-registers
11632 @cindex floating point registers
11633 @item info all-registers
11634 Print the names and values of all registers, including floating-point
11635 and vector registers (in the selected stack frame).
11637 @item info registers @var{reggroup} @dots{}
11638 Print the name and value of the registers in each of the specified
11639 @var{reggroup}s. The @var{reggoup} can be any of those returned by
11640 @code{maint print reggroups} (@pxref{Maintenance Commands}).
11642 @item info registers @var{regname} @dots{}
11643 Print the @dfn{relativized} value of each specified register @var{regname}.
11644 As discussed in detail below, register values are normally relative to
11645 the selected stack frame. The @var{regname} may be any register name valid on
11646 the machine you are using, with or without the initial @samp{$}.
11649 @anchor{standard registers}
11650 @cindex stack pointer register
11651 @cindex program counter register
11652 @cindex process status register
11653 @cindex frame pointer register
11654 @cindex standard registers
11655 @value{GDBN} has four ``standard'' register names that are available (in
11656 expressions) on most machines---whenever they do not conflict with an
11657 architecture's canonical mnemonics for registers. The register names
11658 @code{$pc} and @code{$sp} are used for the program counter register and
11659 the stack pointer. @code{$fp} is used for a register that contains a
11660 pointer to the current stack frame, and @code{$ps} is used for a
11661 register that contains the processor status. For example,
11662 you could print the program counter in hex with
11669 or print the instruction to be executed next with
11676 or add four to the stack pointer@footnote{This is a way of removing
11677 one word from the stack, on machines where stacks grow downward in
11678 memory (most machines, nowadays). This assumes that the innermost
11679 stack frame is selected; setting @code{$sp} is not allowed when other
11680 stack frames are selected. To pop entire frames off the stack,
11681 regardless of machine architecture, use @code{return};
11682 see @ref{Returning, ,Returning from a Function}.} with
11688 Whenever possible, these four standard register names are available on
11689 your machine even though the machine has different canonical mnemonics,
11690 so long as there is no conflict. The @code{info registers} command
11691 shows the canonical names. For example, on the SPARC, @code{info
11692 registers} displays the processor status register as @code{$psr} but you
11693 can also refer to it as @code{$ps}; and on x86-based machines @code{$ps}
11694 is an alias for the @sc{eflags} register.
11696 @value{GDBN} always considers the contents of an ordinary register as an
11697 integer when the register is examined in this way. Some machines have
11698 special registers which can hold nothing but floating point; these
11699 registers are considered to have floating point values. There is no way
11700 to refer to the contents of an ordinary register as floating point value
11701 (although you can @emph{print} it as a floating point value with
11702 @samp{print/f $@var{regname}}).
11704 Some registers have distinct ``raw'' and ``virtual'' data formats. This
11705 means that the data format in which the register contents are saved by
11706 the operating system is not the same one that your program normally
11707 sees. For example, the registers of the 68881 floating point
11708 coprocessor are always saved in ``extended'' (raw) format, but all C
11709 programs expect to work with ``double'' (virtual) format. In such
11710 cases, @value{GDBN} normally works with the virtual format only (the format
11711 that makes sense for your program), but the @code{info registers} command
11712 prints the data in both formats.
11714 @cindex SSE registers (x86)
11715 @cindex MMX registers (x86)
11716 Some machines have special registers whose contents can be interpreted
11717 in several different ways. For example, modern x86-based machines
11718 have SSE and MMX registers that can hold several values packed
11719 together in several different formats. @value{GDBN} refers to such
11720 registers in @code{struct} notation:
11723 (@value{GDBP}) print $xmm1
11725 v4_float = @{0, 3.43859137e-038, 1.54142831e-044, 1.821688e-044@},
11726 v2_double = @{9.92129282474342e-303, 2.7585945287983262e-313@},
11727 v16_int8 = "\000\000\000\000\3706;\001\v\000\000\000\r\000\000",
11728 v8_int16 = @{0, 0, 14072, 315, 11, 0, 13, 0@},
11729 v4_int32 = @{0, 20657912, 11, 13@},
11730 v2_int64 = @{88725056443645952, 55834574859@},
11731 uint128 = 0x0000000d0000000b013b36f800000000
11736 To set values of such registers, you need to tell @value{GDBN} which
11737 view of the register you wish to change, as if you were assigning
11738 value to a @code{struct} member:
11741 (@value{GDBP}) set $xmm1.uint128 = 0x000000000000000000000000FFFFFFFF
11744 Normally, register values are relative to the selected stack frame
11745 (@pxref{Selection, ,Selecting a Frame}). This means that you get the
11746 value that the register would contain if all stack frames farther in
11747 were exited and their saved registers restored. In order to see the
11748 true contents of hardware registers, you must select the innermost
11749 frame (with @samp{frame 0}).
11751 @cindex caller-saved registers
11752 @cindex call-clobbered registers
11753 @cindex volatile registers
11754 @cindex <not saved> values
11755 Usually ABIs reserve some registers as not needed to be saved by the
11756 callee (a.k.a.: ``caller-saved'', ``call-clobbered'' or ``volatile''
11757 registers). It may therefore not be possible for @value{GDBN} to know
11758 the value a register had before the call (in other words, in the outer
11759 frame), if the register value has since been changed by the callee.
11760 @value{GDBN} tries to deduce where the inner frame saved
11761 (``callee-saved'') registers, from the debug info, unwind info, or the
11762 machine code generated by your compiler. If some register is not
11763 saved, and @value{GDBN} knows the register is ``caller-saved'' (via
11764 its own knowledge of the ABI, or because the debug/unwind info
11765 explicitly says the register's value is undefined), @value{GDBN}
11766 displays @w{@samp{<not saved>}} as the register's value. With targets
11767 that @value{GDBN} has no knowledge of the register saving convention,
11768 if a register was not saved by the callee, then its value and location
11769 in the outer frame are assumed to be the same of the inner frame.
11770 This is usually harmless, because if the register is call-clobbered,
11771 the caller either does not care what is in the register after the
11772 call, or has code to restore the value that it does care about. Note,
11773 however, that if you change such a register in the outer frame, you
11774 may also be affecting the inner frame. Also, the more ``outer'' the
11775 frame is you're looking at, the more likely a call-clobbered
11776 register's value is to be wrong, in the sense that it doesn't actually
11777 represent the value the register had just before the call.
11779 @node Floating Point Hardware
11780 @section Floating Point Hardware
11781 @cindex floating point
11783 Depending on the configuration, @value{GDBN} may be able to give
11784 you more information about the status of the floating point hardware.
11789 Display hardware-dependent information about the floating
11790 point unit. The exact contents and layout vary depending on the
11791 floating point chip. Currently, @samp{info float} is supported on
11792 the ARM and x86 machines.
11796 @section Vector Unit
11797 @cindex vector unit
11799 Depending on the configuration, @value{GDBN} may be able to give you
11800 more information about the status of the vector unit.
11803 @kindex info vector
11805 Display information about the vector unit. The exact contents and
11806 layout vary depending on the hardware.
11809 @node OS Information
11810 @section Operating System Auxiliary Information
11811 @cindex OS information
11813 @value{GDBN} provides interfaces to useful OS facilities that can help
11814 you debug your program.
11816 @cindex auxiliary vector
11817 @cindex vector, auxiliary
11818 Some operating systems supply an @dfn{auxiliary vector} to programs at
11819 startup. This is akin to the arguments and environment that you
11820 specify for a program, but contains a system-dependent variety of
11821 binary values that tell system libraries important details about the
11822 hardware, operating system, and process. Each value's purpose is
11823 identified by an integer tag; the meanings are well-known but system-specific.
11824 Depending on the configuration and operating system facilities,
11825 @value{GDBN} may be able to show you this information. For remote
11826 targets, this functionality may further depend on the remote stub's
11827 support of the @samp{qXfer:auxv:read} packet, see
11828 @ref{qXfer auxiliary vector read}.
11833 Display the auxiliary vector of the inferior, which can be either a
11834 live process or a core dump file. @value{GDBN} prints each tag value
11835 numerically, and also shows names and text descriptions for recognized
11836 tags. Some values in the vector are numbers, some bit masks, and some
11837 pointers to strings or other data. @value{GDBN} displays each value in the
11838 most appropriate form for a recognized tag, and in hexadecimal for
11839 an unrecognized tag.
11842 On some targets, @value{GDBN} can access operating system-specific
11843 information and show it to you. The types of information available
11844 will differ depending on the type of operating system running on the
11845 target. The mechanism used to fetch the data is described in
11846 @ref{Operating System Information}. For remote targets, this
11847 functionality depends on the remote stub's support of the
11848 @samp{qXfer:osdata:read} packet, see @ref{qXfer osdata read}.
11852 @item info os @var{infotype}
11854 Display OS information of the requested type.
11856 On @sc{gnu}/Linux, the following values of @var{infotype} are valid:
11858 @anchor{linux info os infotypes}
11860 @kindex info os cpus
11862 Display the list of all CPUs/cores. For each CPU/core, @value{GDBN} prints
11863 the available fields from /proc/cpuinfo. For each supported architecture
11864 different fields are available. Two common entries are processor which gives
11865 CPU number and bogomips; a system constant that is calculated during
11866 kernel initialization.
11868 @kindex info os files
11870 Display the list of open file descriptors on the target. For each
11871 file descriptor, @value{GDBN} prints the identifier of the process
11872 owning the descriptor, the command of the owning process, the value
11873 of the descriptor, and the target of the descriptor.
11875 @kindex info os modules
11877 Display the list of all loaded kernel modules on the target. For each
11878 module, @value{GDBN} prints the module name, the size of the module in
11879 bytes, the number of times the module is used, the dependencies of the
11880 module, the status of the module, and the address of the loaded module
11883 @kindex info os msg
11885 Display the list of all System V message queues on the target. For each
11886 message queue, @value{GDBN} prints the message queue key, the message
11887 queue identifier, the access permissions, the current number of bytes
11888 on the queue, the current number of messages on the queue, the processes
11889 that last sent and received a message on the queue, the user and group
11890 of the owner and creator of the message queue, the times at which a
11891 message was last sent and received on the queue, and the time at which
11892 the message queue was last changed.
11894 @kindex info os processes
11896 Display the list of processes on the target. For each process,
11897 @value{GDBN} prints the process identifier, the name of the user, the
11898 command corresponding to the process, and the list of processor cores
11899 that the process is currently running on. (To understand what these
11900 properties mean, for this and the following info types, please consult
11901 the general @sc{gnu}/Linux documentation.)
11903 @kindex info os procgroups
11905 Display the list of process groups on the target. For each process,
11906 @value{GDBN} prints the identifier of the process group that it belongs
11907 to, the command corresponding to the process group leader, the process
11908 identifier, and the command line of the process. The list is sorted
11909 first by the process group identifier, then by the process identifier,
11910 so that processes belonging to the same process group are grouped together
11911 and the process group leader is listed first.
11913 @kindex info os semaphores
11915 Display the list of all System V semaphore sets on the target. For each
11916 semaphore set, @value{GDBN} prints the semaphore set key, the semaphore
11917 set identifier, the access permissions, the number of semaphores in the
11918 set, the user and group of the owner and creator of the semaphore set,
11919 and the times at which the semaphore set was operated upon and changed.
11921 @kindex info os shm
11923 Display the list of all System V shared-memory regions on the target.
11924 For each shared-memory region, @value{GDBN} prints the region key,
11925 the shared-memory identifier, the access permissions, the size of the
11926 region, the process that created the region, the process that last
11927 attached to or detached from the region, the current number of live
11928 attaches to the region, and the times at which the region was last
11929 attached to, detach from, and changed.
11931 @kindex info os sockets
11933 Display the list of Internet-domain sockets on the target. For each
11934 socket, @value{GDBN} prints the address and port of the local and
11935 remote endpoints, the current state of the connection, the creator of
11936 the socket, the IP address family of the socket, and the type of the
11939 @kindex info os threads
11941 Display the list of threads running on the target. For each thread,
11942 @value{GDBN} prints the identifier of the process that the thread
11943 belongs to, the command of the process, the thread identifier, and the
11944 processor core that it is currently running on. The main thread of a
11945 process is not listed.
11949 If @var{infotype} is omitted, then list the possible values for
11950 @var{infotype} and the kind of OS information available for each
11951 @var{infotype}. If the target does not return a list of possible
11952 types, this command will report an error.
11955 @node Memory Region Attributes
11956 @section Memory Region Attributes
11957 @cindex memory region attributes
11959 @dfn{Memory region attributes} allow you to describe special handling
11960 required by regions of your target's memory. @value{GDBN} uses
11961 attributes to determine whether to allow certain types of memory
11962 accesses; whether to use specific width accesses; and whether to cache
11963 target memory. By default the description of memory regions is
11964 fetched from the target (if the current target supports this), but the
11965 user can override the fetched regions.
11967 Defined memory regions can be individually enabled and disabled. When a
11968 memory region is disabled, @value{GDBN} uses the default attributes when
11969 accessing memory in that region. Similarly, if no memory regions have
11970 been defined, @value{GDBN} uses the default attributes when accessing
11973 When a memory region is defined, it is given a number to identify it;
11974 to enable, disable, or remove a memory region, you specify that number.
11978 @item mem @var{lower} @var{upper} @var{attributes}@dots{}
11979 Define a memory region bounded by @var{lower} and @var{upper} with
11980 attributes @var{attributes}@dots{}, and add it to the list of regions
11981 monitored by @value{GDBN}. Note that @var{upper} == 0 is a special
11982 case: it is treated as the target's maximum memory address.
11983 (0xffff on 16 bit targets, 0xffffffff on 32 bit targets, etc.)
11986 Discard any user changes to the memory regions and use target-supplied
11987 regions, if available, or no regions if the target does not support.
11990 @item delete mem @var{nums}@dots{}
11991 Remove memory regions @var{nums}@dots{} from the list of regions
11992 monitored by @value{GDBN}.
11994 @kindex disable mem
11995 @item disable mem @var{nums}@dots{}
11996 Disable monitoring of memory regions @var{nums}@dots{}.
11997 A disabled memory region is not forgotten.
11998 It may be enabled again later.
12001 @item enable mem @var{nums}@dots{}
12002 Enable monitoring of memory regions @var{nums}@dots{}.
12006 Print a table of all defined memory regions, with the following columns
12010 @item Memory Region Number
12011 @item Enabled or Disabled.
12012 Enabled memory regions are marked with @samp{y}.
12013 Disabled memory regions are marked with @samp{n}.
12016 The address defining the inclusive lower bound of the memory region.
12019 The address defining the exclusive upper bound of the memory region.
12022 The list of attributes set for this memory region.
12027 @subsection Attributes
12029 @subsubsection Memory Access Mode
12030 The access mode attributes set whether @value{GDBN} may make read or
12031 write accesses to a memory region.
12033 While these attributes prevent @value{GDBN} from performing invalid
12034 memory accesses, they do nothing to prevent the target system, I/O DMA,
12035 etc.@: from accessing memory.
12039 Memory is read only.
12041 Memory is write only.
12043 Memory is read/write. This is the default.
12046 @subsubsection Memory Access Size
12047 The access size attribute tells @value{GDBN} to use specific sized
12048 accesses in the memory region. Often memory mapped device registers
12049 require specific sized accesses. If no access size attribute is
12050 specified, @value{GDBN} may use accesses of any size.
12054 Use 8 bit memory accesses.
12056 Use 16 bit memory accesses.
12058 Use 32 bit memory accesses.
12060 Use 64 bit memory accesses.
12063 @c @subsubsection Hardware/Software Breakpoints
12064 @c The hardware/software breakpoint attributes set whether @value{GDBN}
12065 @c will use hardware or software breakpoints for the internal breakpoints
12066 @c used by the step, next, finish, until, etc. commands.
12070 @c Always use hardware breakpoints
12071 @c @item swbreak (default)
12074 @subsubsection Data Cache
12075 The data cache attributes set whether @value{GDBN} will cache target
12076 memory. While this generally improves performance by reducing debug
12077 protocol overhead, it can lead to incorrect results because @value{GDBN}
12078 does not know about volatile variables or memory mapped device
12083 Enable @value{GDBN} to cache target memory.
12085 Disable @value{GDBN} from caching target memory. This is the default.
12088 @subsection Memory Access Checking
12089 @value{GDBN} can be instructed to refuse accesses to memory that is
12090 not explicitly described. This can be useful if accessing such
12091 regions has undesired effects for a specific target, or to provide
12092 better error checking. The following commands control this behaviour.
12095 @kindex set mem inaccessible-by-default
12096 @item set mem inaccessible-by-default [on|off]
12097 If @code{on} is specified, make @value{GDBN} treat memory not
12098 explicitly described by the memory ranges as non-existent and refuse accesses
12099 to such memory. The checks are only performed if there's at least one
12100 memory range defined. If @code{off} is specified, make @value{GDBN}
12101 treat the memory not explicitly described by the memory ranges as RAM.
12102 The default value is @code{on}.
12103 @kindex show mem inaccessible-by-default
12104 @item show mem inaccessible-by-default
12105 Show the current handling of accesses to unknown memory.
12109 @c @subsubsection Memory Write Verification
12110 @c The memory write verification attributes set whether @value{GDBN}
12111 @c will re-reads data after each write to verify the write was successful.
12115 @c @item noverify (default)
12118 @node Dump/Restore Files
12119 @section Copy Between Memory and a File
12120 @cindex dump/restore files
12121 @cindex append data to a file
12122 @cindex dump data to a file
12123 @cindex restore data from a file
12125 You can use the commands @code{dump}, @code{append}, and
12126 @code{restore} to copy data between target memory and a file. The
12127 @code{dump} and @code{append} commands write data to a file, and the
12128 @code{restore} command reads data from a file back into the inferior's
12129 memory. Files may be in binary, Motorola S-record, Intel hex,
12130 Tektronix Hex, or Verilog Hex format; however, @value{GDBN} can only
12131 append to binary files, and cannot read from Verilog Hex files.
12136 @item dump @r{[}@var{format}@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
12137 @itemx dump @r{[}@var{format}@r{]} value @var{filename} @var{expr}
12138 Dump the contents of memory from @var{start_addr} to @var{end_addr},
12139 or the value of @var{expr}, to @var{filename} in the given format.
12141 The @var{format} parameter may be any one of:
12148 Motorola S-record format.
12150 Tektronix Hex format.
12152 Verilog Hex format.
12155 @value{GDBN} uses the same definitions of these formats as the
12156 @sc{gnu} binary utilities, like @samp{objdump} and @samp{objcopy}. If
12157 @var{format} is omitted, @value{GDBN} dumps the data in raw binary
12161 @item append @r{[}binary@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
12162 @itemx append @r{[}binary@r{]} value @var{filename} @var{expr}
12163 Append the contents of memory from @var{start_addr} to @var{end_addr},
12164 or the value of @var{expr}, to the file @var{filename}, in raw binary form.
12165 (@value{GDBN} can only append data to files in raw binary form.)
12168 @item restore @var{filename} @r{[}binary@r{]} @var{bias} @var{start} @var{end}
12169 Restore the contents of file @var{filename} into memory. The
12170 @code{restore} command can automatically recognize any known @sc{bfd}
12171 file format, except for raw binary. To restore a raw binary file you
12172 must specify the optional keyword @code{binary} after the filename.
12174 If @var{bias} is non-zero, its value will be added to the addresses
12175 contained in the file. Binary files always start at address zero, so
12176 they will be restored at address @var{bias}. Other bfd files have
12177 a built-in location; they will be restored at offset @var{bias}
12178 from that location.
12180 If @var{start} and/or @var{end} are non-zero, then only data between
12181 file offset @var{start} and file offset @var{end} will be restored.
12182 These offsets are relative to the addresses in the file, before
12183 the @var{bias} argument is applied.
12187 @node Core File Generation
12188 @section How to Produce a Core File from Your Program
12189 @cindex dump core from inferior
12191 A @dfn{core file} or @dfn{core dump} is a file that records the memory
12192 image of a running process and its process status (register values
12193 etc.). Its primary use is post-mortem debugging of a program that
12194 crashed while it ran outside a debugger. A program that crashes
12195 automatically produces a core file, unless this feature is disabled by
12196 the user. @xref{Files}, for information on invoking @value{GDBN} in
12197 the post-mortem debugging mode.
12199 Occasionally, you may wish to produce a core file of the program you
12200 are debugging in order to preserve a snapshot of its state.
12201 @value{GDBN} has a special command for that.
12205 @kindex generate-core-file
12206 @item generate-core-file [@var{file}]
12207 @itemx gcore [@var{file}]
12208 Produce a core dump of the inferior process. The optional argument
12209 @var{file} specifies the file name where to put the core dump. If not
12210 specified, the file name defaults to @file{core.@var{pid}}, where
12211 @var{pid} is the inferior process ID.
12213 Note that this command is implemented only for some systems (as of
12214 this writing, @sc{gnu}/Linux, FreeBSD, Solaris, and S390).
12216 On @sc{gnu}/Linux, this command can take into account the value of the
12217 file @file{/proc/@var{pid}/coredump_filter} when generating the core
12218 dump (@pxref{set use-coredump-filter}), and by default honors the
12219 @code{VM_DONTDUMP} flag for mappings where it is present in the file
12220 @file{/proc/@var{pid}/smaps} (@pxref{set dump-excluded-mappings}).
12222 @kindex set use-coredump-filter
12223 @anchor{set use-coredump-filter}
12224 @item set use-coredump-filter on
12225 @itemx set use-coredump-filter off
12226 Enable or disable the use of the file
12227 @file{/proc/@var{pid}/coredump_filter} when generating core dump
12228 files. This file is used by the Linux kernel to decide what types of
12229 memory mappings will be dumped or ignored when generating a core dump
12230 file. @var{pid} is the process ID of a currently running process.
12232 To make use of this feature, you have to write in the
12233 @file{/proc/@var{pid}/coredump_filter} file a value, in hexadecimal,
12234 which is a bit mask representing the memory mapping types. If a bit
12235 is set in the bit mask, then the memory mappings of the corresponding
12236 types will be dumped; otherwise, they will be ignored. This
12237 configuration is inherited by child processes. For more information
12238 about the bits that can be set in the
12239 @file{/proc/@var{pid}/coredump_filter} file, please refer to the
12240 manpage of @code{core(5)}.
12242 By default, this option is @code{on}. If this option is turned
12243 @code{off}, @value{GDBN} does not read the @file{coredump_filter} file
12244 and instead uses the same default value as the Linux kernel in order
12245 to decide which pages will be dumped in the core dump file. This
12246 value is currently @code{0x33}, which means that bits @code{0}
12247 (anonymous private mappings), @code{1} (anonymous shared mappings),
12248 @code{4} (ELF headers) and @code{5} (private huge pages) are active.
12249 This will cause these memory mappings to be dumped automatically.
12251 @kindex set dump-excluded-mappings
12252 @anchor{set dump-excluded-mappings}
12253 @item set dump-excluded-mappings on
12254 @itemx set dump-excluded-mappings off
12255 If @code{on} is specified, @value{GDBN} will dump memory mappings
12256 marked with the @code{VM_DONTDUMP} flag. This flag is represented in
12257 the file @file{/proc/@var{pid}/smaps} with the acronym @code{dd}.
12259 The default value is @code{off}.
12262 @node Character Sets
12263 @section Character Sets
12264 @cindex character sets
12266 @cindex translating between character sets
12267 @cindex host character set
12268 @cindex target character set
12270 If the program you are debugging uses a different character set to
12271 represent characters and strings than the one @value{GDBN} uses itself,
12272 @value{GDBN} can automatically translate between the character sets for
12273 you. The character set @value{GDBN} uses we call the @dfn{host
12274 character set}; the one the inferior program uses we call the
12275 @dfn{target character set}.
12277 For example, if you are running @value{GDBN} on a @sc{gnu}/Linux system, which
12278 uses the ISO Latin 1 character set, but you are using @value{GDBN}'s
12279 remote protocol (@pxref{Remote Debugging}) to debug a program
12280 running on an IBM mainframe, which uses the @sc{ebcdic} character set,
12281 then the host character set is Latin-1, and the target character set is
12282 @sc{ebcdic}. If you give @value{GDBN} the command @code{set
12283 target-charset EBCDIC-US}, then @value{GDBN} translates between
12284 @sc{ebcdic} and Latin 1 as you print character or string values, or use
12285 character and string literals in expressions.
12287 @value{GDBN} has no way to automatically recognize which character set
12288 the inferior program uses; you must tell it, using the @code{set
12289 target-charset} command, described below.
12291 Here are the commands for controlling @value{GDBN}'s character set
12295 @item set target-charset @var{charset}
12296 @kindex set target-charset
12297 Set the current target character set to @var{charset}. To display the
12298 list of supported target character sets, type
12299 @kbd{@w{set target-charset @key{TAB}@key{TAB}}}.
12301 @item set host-charset @var{charset}
12302 @kindex set host-charset
12303 Set the current host character set to @var{charset}.
12305 By default, @value{GDBN} uses a host character set appropriate to the
12306 system it is running on; you can override that default using the
12307 @code{set host-charset} command. On some systems, @value{GDBN} cannot
12308 automatically determine the appropriate host character set. In this
12309 case, @value{GDBN} uses @samp{UTF-8}.
12311 @value{GDBN} can only use certain character sets as its host character
12312 set. If you type @kbd{@w{set host-charset @key{TAB}@key{TAB}}},
12313 @value{GDBN} will list the host character sets it supports.
12315 @item set charset @var{charset}
12316 @kindex set charset
12317 Set the current host and target character sets to @var{charset}. As
12318 above, if you type @kbd{@w{set charset @key{TAB}@key{TAB}}},
12319 @value{GDBN} will list the names of the character sets that can be used
12320 for both host and target.
12323 @kindex show charset
12324 Show the names of the current host and target character sets.
12326 @item show host-charset
12327 @kindex show host-charset
12328 Show the name of the current host character set.
12330 @item show target-charset
12331 @kindex show target-charset
12332 Show the name of the current target character set.
12334 @item set target-wide-charset @var{charset}
12335 @kindex set target-wide-charset
12336 Set the current target's wide character set to @var{charset}. This is
12337 the character set used by the target's @code{wchar_t} type. To
12338 display the list of supported wide character sets, type
12339 @kbd{@w{set target-wide-charset @key{TAB}@key{TAB}}}.
12341 @item show target-wide-charset
12342 @kindex show target-wide-charset
12343 Show the name of the current target's wide character set.
12346 Here is an example of @value{GDBN}'s character set support in action.
12347 Assume that the following source code has been placed in the file
12348 @file{charset-test.c}:
12354 = @{72, 101, 108, 108, 111, 44, 32, 119,
12355 111, 114, 108, 100, 33, 10, 0@};
12356 char ibm1047_hello[]
12357 = @{200, 133, 147, 147, 150, 107, 64, 166,
12358 150, 153, 147, 132, 90, 37, 0@};
12362 printf ("Hello, world!\n");
12366 In this program, @code{ascii_hello} and @code{ibm1047_hello} are arrays
12367 containing the string @samp{Hello, world!} followed by a newline,
12368 encoded in the @sc{ascii} and @sc{ibm1047} character sets.
12370 We compile the program, and invoke the debugger on it:
12373 $ gcc -g charset-test.c -o charset-test
12374 $ gdb -nw charset-test
12375 GNU gdb 2001-12-19-cvs
12376 Copyright 2001 Free Software Foundation, Inc.
12381 We can use the @code{show charset} command to see what character sets
12382 @value{GDBN} is currently using to interpret and display characters and
12386 (@value{GDBP}) show charset
12387 The current host and target character set is `ISO-8859-1'.
12391 For the sake of printing this manual, let's use @sc{ascii} as our
12392 initial character set:
12394 (@value{GDBP}) set charset ASCII
12395 (@value{GDBP}) show charset
12396 The current host and target character set is `ASCII'.
12400 Let's assume that @sc{ascii} is indeed the correct character set for our
12401 host system --- in other words, let's assume that if @value{GDBN} prints
12402 characters using the @sc{ascii} character set, our terminal will display
12403 them properly. Since our current target character set is also
12404 @sc{ascii}, the contents of @code{ascii_hello} print legibly:
12407 (@value{GDBP}) print ascii_hello
12408 $1 = 0x401698 "Hello, world!\n"
12409 (@value{GDBP}) print ascii_hello[0]
12414 @value{GDBN} uses the target character set for character and string
12415 literals you use in expressions:
12418 (@value{GDBP}) print '+'
12423 The @sc{ascii} character set uses the number 43 to encode the @samp{+}
12426 @value{GDBN} relies on the user to tell it which character set the
12427 target program uses. If we print @code{ibm1047_hello} while our target
12428 character set is still @sc{ascii}, we get jibberish:
12431 (@value{GDBP}) print ibm1047_hello
12432 $4 = 0x4016a8 "\310\205\223\223\226k@@\246\226\231\223\204Z%"
12433 (@value{GDBP}) print ibm1047_hello[0]
12438 If we invoke the @code{set target-charset} followed by @key{TAB}@key{TAB},
12439 @value{GDBN} tells us the character sets it supports:
12442 (@value{GDBP}) set target-charset
12443 ASCII EBCDIC-US IBM1047 ISO-8859-1
12444 (@value{GDBP}) set target-charset
12447 We can select @sc{ibm1047} as our target character set, and examine the
12448 program's strings again. Now the @sc{ascii} string is wrong, but
12449 @value{GDBN} translates the contents of @code{ibm1047_hello} from the
12450 target character set, @sc{ibm1047}, to the host character set,
12451 @sc{ascii}, and they display correctly:
12454 (@value{GDBP}) set target-charset IBM1047
12455 (@value{GDBP}) show charset
12456 The current host character set is `ASCII'.
12457 The current target character set is `IBM1047'.
12458 (@value{GDBP}) print ascii_hello
12459 $6 = 0x401698 "\110\145%%?\054\040\167?\162%\144\041\012"
12460 (@value{GDBP}) print ascii_hello[0]
12462 (@value{GDBP}) print ibm1047_hello
12463 $8 = 0x4016a8 "Hello, world!\n"
12464 (@value{GDBP}) print ibm1047_hello[0]
12469 As above, @value{GDBN} uses the target character set for character and
12470 string literals you use in expressions:
12473 (@value{GDBP}) print '+'
12478 The @sc{ibm1047} character set uses the number 78 to encode the @samp{+}
12481 @node Caching Target Data
12482 @section Caching Data of Targets
12483 @cindex caching data of targets
12485 @value{GDBN} caches data exchanged between the debugger and a target.
12486 Each cache is associated with the address space of the inferior.
12487 @xref{Inferiors and Programs}, about inferior and address space.
12488 Such caching generally improves performance in remote debugging
12489 (@pxref{Remote Debugging}), because it reduces the overhead of the
12490 remote protocol by bundling memory reads and writes into large chunks.
12491 Unfortunately, simply caching everything would lead to incorrect results,
12492 since @value{GDBN} does not necessarily know anything about volatile
12493 values, memory-mapped I/O addresses, etc. Furthermore, in non-stop mode
12494 (@pxref{Non-Stop Mode}) memory can be changed @emph{while} a gdb command
12496 Therefore, by default, @value{GDBN} only caches data
12497 known to be on the stack@footnote{In non-stop mode, it is moderately
12498 rare for a running thread to modify the stack of a stopped thread
12499 in a way that would interfere with a backtrace, and caching of
12500 stack reads provides a significant speed up of remote backtraces.} or
12501 in the code segment.
12502 Other regions of memory can be explicitly marked as
12503 cacheable; @pxref{Memory Region Attributes}.
12506 @kindex set remotecache
12507 @item set remotecache on
12508 @itemx set remotecache off
12509 This option no longer does anything; it exists for compatibility
12512 @kindex show remotecache
12513 @item show remotecache
12514 Show the current state of the obsolete remotecache flag.
12516 @kindex set stack-cache
12517 @item set stack-cache on
12518 @itemx set stack-cache off
12519 Enable or disable caching of stack accesses. When @code{on}, use
12520 caching. By default, this option is @code{on}.
12522 @kindex show stack-cache
12523 @item show stack-cache
12524 Show the current state of data caching for memory accesses.
12526 @kindex set code-cache
12527 @item set code-cache on
12528 @itemx set code-cache off
12529 Enable or disable caching of code segment accesses. When @code{on},
12530 use caching. By default, this option is @code{on}. This improves
12531 performance of disassembly in remote debugging.
12533 @kindex show code-cache
12534 @item show code-cache
12535 Show the current state of target memory cache for code segment
12538 @kindex info dcache
12539 @item info dcache @r{[}line@r{]}
12540 Print the information about the performance of data cache of the
12541 current inferior's address space. The information displayed
12542 includes the dcache width and depth, and for each cache line, its
12543 number, address, and how many times it was referenced. This
12544 command is useful for debugging the data cache operation.
12546 If a line number is specified, the contents of that line will be
12549 @item set dcache size @var{size}
12550 @cindex dcache size
12551 @kindex set dcache size
12552 Set maximum number of entries in dcache (dcache depth above).
12554 @item set dcache line-size @var{line-size}
12555 @cindex dcache line-size
12556 @kindex set dcache line-size
12557 Set number of bytes each dcache entry caches (dcache width above).
12558 Must be a power of 2.
12560 @item show dcache size
12561 @kindex show dcache size
12562 Show maximum number of dcache entries. @xref{Caching Target Data, info dcache}.
12564 @item show dcache line-size
12565 @kindex show dcache line-size
12566 Show default size of dcache lines.
12570 @node Searching Memory
12571 @section Search Memory
12572 @cindex searching memory
12574 Memory can be searched for a particular sequence of bytes with the
12575 @code{find} command.
12579 @item find @r{[}/@var{sn}@r{]} @var{start_addr}, +@var{len}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
12580 @itemx find @r{[}/@var{sn}@r{]} @var{start_addr}, @var{end_addr}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
12581 Search memory for the sequence of bytes specified by @var{val1}, @var{val2},
12582 etc. The search begins at address @var{start_addr} and continues for either
12583 @var{len} bytes or through to @var{end_addr} inclusive.
12586 @var{s} and @var{n} are optional parameters.
12587 They may be specified in either order, apart or together.
12590 @item @var{s}, search query size
12591 The size of each search query value.
12597 halfwords (two bytes)
12601 giant words (eight bytes)
12604 All values are interpreted in the current language.
12605 This means, for example, that if the current source language is C/C@t{++}
12606 then searching for the string ``hello'' includes the trailing '\0'.
12607 The null terminator can be removed from searching by using casts,
12608 e.g.: @samp{@{char[5]@}"hello"}.
12610 If the value size is not specified, it is taken from the
12611 value's type in the current language.
12612 This is useful when one wants to specify the search
12613 pattern as a mixture of types.
12614 Note that this means, for example, that in the case of C-like languages
12615 a search for an untyped 0x42 will search for @samp{(int) 0x42}
12616 which is typically four bytes.
12618 @item @var{n}, maximum number of finds
12619 The maximum number of matches to print. The default is to print all finds.
12622 You can use strings as search values. Quote them with double-quotes
12624 The string value is copied into the search pattern byte by byte,
12625 regardless of the endianness of the target and the size specification.
12627 The address of each match found is printed as well as a count of the
12628 number of matches found.
12630 The address of the last value found is stored in convenience variable
12632 A count of the number of matches is stored in @samp{$numfound}.
12634 For example, if stopped at the @code{printf} in this function:
12640 static char hello[] = "hello-hello";
12641 static struct @{ char c; short s; int i; @}
12642 __attribute__ ((packed)) mixed
12643 = @{ 'c', 0x1234, 0x87654321 @};
12644 printf ("%s\n", hello);
12649 you get during debugging:
12652 (gdb) find &hello[0], +sizeof(hello), "hello"
12653 0x804956d <hello.1620+6>
12655 (gdb) find &hello[0], +sizeof(hello), 'h', 'e', 'l', 'l', 'o'
12656 0x8049567 <hello.1620>
12657 0x804956d <hello.1620+6>
12659 (gdb) find &hello[0], +sizeof(hello), @{char[5]@}"hello"
12660 0x8049567 <hello.1620>
12661 0x804956d <hello.1620+6>
12663 (gdb) find /b1 &hello[0], +sizeof(hello), 'h', 0x65, 'l'
12664 0x8049567 <hello.1620>
12666 (gdb) find &mixed, +sizeof(mixed), (char) 'c', (short) 0x1234, (int) 0x87654321
12667 0x8049560 <mixed.1625>
12669 (gdb) print $numfound
12672 $2 = (void *) 0x8049560
12676 @section Value Sizes
12678 Whenever @value{GDBN} prints a value memory will be allocated within
12679 @value{GDBN} to hold the contents of the value. It is possible in
12680 some languages with dynamic typing systems, that an invalid program
12681 may indicate a value that is incorrectly large, this in turn may cause
12682 @value{GDBN} to try and allocate an overly large ammount of memory.
12685 @kindex set max-value-size
12686 @item set max-value-size @var{bytes}
12687 @itemx set max-value-size unlimited
12688 Set the maximum size of memory that @value{GDBN} will allocate for the
12689 contents of a value to @var{bytes}, trying to display a value that
12690 requires more memory than that will result in an error.
12692 Setting this variable does not effect values that have already been
12693 allocated within @value{GDBN}, only future allocations.
12695 There's a minimum size that @code{max-value-size} can be set to in
12696 order that @value{GDBN} can still operate correctly, this minimum is
12697 currently 16 bytes.
12699 The limit applies to the results of some subexpressions as well as to
12700 complete expressions. For example, an expression denoting a simple
12701 integer component, such as @code{x.y.z}, may fail if the size of
12702 @var{x.y} is dynamic and exceeds @var{bytes}. On the other hand,
12703 @value{GDBN} is sometimes clever; the expression @code{A[i]}, where
12704 @var{A} is an array variable with non-constant size, will generally
12705 succeed regardless of the bounds on @var{A}, as long as the component
12706 size is less than @var{bytes}.
12708 The default value of @code{max-value-size} is currently 64k.
12710 @kindex show max-value-size
12711 @item show max-value-size
12712 Show the maximum size of memory, in bytes, that @value{GDBN} will
12713 allocate for the contents of a value.
12716 @node Optimized Code
12717 @chapter Debugging Optimized Code
12718 @cindex optimized code, debugging
12719 @cindex debugging optimized code
12721 Almost all compilers support optimization. With optimization
12722 disabled, the compiler generates assembly code that corresponds
12723 directly to your source code, in a simplistic way. As the compiler
12724 applies more powerful optimizations, the generated assembly code
12725 diverges from your original source code. With help from debugging
12726 information generated by the compiler, @value{GDBN} can map from
12727 the running program back to constructs from your original source.
12729 @value{GDBN} is more accurate with optimization disabled. If you
12730 can recompile without optimization, it is easier to follow the
12731 progress of your program during debugging. But, there are many cases
12732 where you may need to debug an optimized version.
12734 When you debug a program compiled with @samp{-g -O}, remember that the
12735 optimizer has rearranged your code; the debugger shows you what is
12736 really there. Do not be too surprised when the execution path does not
12737 exactly match your source file! An extreme example: if you define a
12738 variable, but never use it, @value{GDBN} never sees that
12739 variable---because the compiler optimizes it out of existence.
12741 Some things do not work as well with @samp{-g -O} as with just
12742 @samp{-g}, particularly on machines with instruction scheduling. If in
12743 doubt, recompile with @samp{-g} alone, and if this fixes the problem,
12744 please report it to us as a bug (including a test case!).
12745 @xref{Variables}, for more information about debugging optimized code.
12748 * Inline Functions:: How @value{GDBN} presents inlining
12749 * Tail Call Frames:: @value{GDBN} analysis of jumps to functions
12752 @node Inline Functions
12753 @section Inline Functions
12754 @cindex inline functions, debugging
12756 @dfn{Inlining} is an optimization that inserts a copy of the function
12757 body directly at each call site, instead of jumping to a shared
12758 routine. @value{GDBN} displays inlined functions just like
12759 non-inlined functions. They appear in backtraces. You can view their
12760 arguments and local variables, step into them with @code{step}, skip
12761 them with @code{next}, and escape from them with @code{finish}.
12762 You can check whether a function was inlined by using the
12763 @code{info frame} command.
12765 For @value{GDBN} to support inlined functions, the compiler must
12766 record information about inlining in the debug information ---
12767 @value{NGCC} using the @sc{dwarf 2} format does this, and several
12768 other compilers do also. @value{GDBN} only supports inlined functions
12769 when using @sc{dwarf 2}. Versions of @value{NGCC} before 4.1
12770 do not emit two required attributes (@samp{DW_AT_call_file} and
12771 @samp{DW_AT_call_line}); @value{GDBN} does not display inlined
12772 function calls with earlier versions of @value{NGCC}. It instead
12773 displays the arguments and local variables of inlined functions as
12774 local variables in the caller.
12776 The body of an inlined function is directly included at its call site;
12777 unlike a non-inlined function, there are no instructions devoted to
12778 the call. @value{GDBN} still pretends that the call site and the
12779 start of the inlined function are different instructions. Stepping to
12780 the call site shows the call site, and then stepping again shows
12781 the first line of the inlined function, even though no additional
12782 instructions are executed.
12784 This makes source-level debugging much clearer; you can see both the
12785 context of the call and then the effect of the call. Only stepping by
12786 a single instruction using @code{stepi} or @code{nexti} does not do
12787 this; single instruction steps always show the inlined body.
12789 There are some ways that @value{GDBN} does not pretend that inlined
12790 function calls are the same as normal calls:
12794 Setting breakpoints at the call site of an inlined function may not
12795 work, because the call site does not contain any code. @value{GDBN}
12796 may incorrectly move the breakpoint to the next line of the enclosing
12797 function, after the call. This limitation will be removed in a future
12798 version of @value{GDBN}; until then, set a breakpoint on an earlier line
12799 or inside the inlined function instead.
12802 @value{GDBN} cannot locate the return value of inlined calls after
12803 using the @code{finish} command. This is a limitation of compiler-generated
12804 debugging information; after @code{finish}, you can step to the next line
12805 and print a variable where your program stored the return value.
12809 @node Tail Call Frames
12810 @section Tail Call Frames
12811 @cindex tail call frames, debugging
12813 Function @code{B} can call function @code{C} in its very last statement. In
12814 unoptimized compilation the call of @code{C} is immediately followed by return
12815 instruction at the end of @code{B} code. Optimizing compiler may replace the
12816 call and return in function @code{B} into one jump to function @code{C}
12817 instead. Such use of a jump instruction is called @dfn{tail call}.
12819 During execution of function @code{C}, there will be no indication in the
12820 function call stack frames that it was tail-called from @code{B}. If function
12821 @code{A} regularly calls function @code{B} which tail-calls function @code{C},
12822 then @value{GDBN} will see @code{A} as the caller of @code{C}. However, in
12823 some cases @value{GDBN} can determine that @code{C} was tail-called from
12824 @code{B}, and it will then create fictitious call frame for that, with the
12825 return address set up as if @code{B} called @code{C} normally.
12827 This functionality is currently supported only by DWARF 2 debugging format and
12828 the compiler has to produce @samp{DW_TAG_call_site} tags. With
12829 @value{NGCC}, you need to specify @option{-O -g} during compilation, to get
12832 @kbd{info frame} command (@pxref{Frame Info}) will indicate the tail call frame
12833 kind by text @code{tail call frame} such as in this sample @value{GDBN} output:
12837 0x40066b <b(int, double)+11>: jmp 0x400640 <c(int, double)>
12839 Stack level 1, frame at 0x7fffffffda30:
12840 rip = 0x40066d in b (amd64-entry-value.cc:59); saved rip 0x4004c5
12841 tail call frame, caller of frame at 0x7fffffffda30
12842 source language c++.
12843 Arglist at unknown address.
12844 Locals at unknown address, Previous frame's sp is 0x7fffffffda30
12847 The detection of all the possible code path executions can find them ambiguous.
12848 There is no execution history stored (possible @ref{Reverse Execution} is never
12849 used for this purpose) and the last known caller could have reached the known
12850 callee by multiple different jump sequences. In such case @value{GDBN} still
12851 tries to show at least all the unambiguous top tail callers and all the
12852 unambiguous bottom tail calees, if any.
12855 @anchor{set debug entry-values}
12856 @item set debug entry-values
12857 @kindex set debug entry-values
12858 When set to on, enables printing of analysis messages for both frame argument
12859 values at function entry and tail calls. It will show all the possible valid
12860 tail calls code paths it has considered. It will also print the intersection
12861 of them with the final unambiguous (possibly partial or even empty) code path
12864 @item show debug entry-values
12865 @kindex show debug entry-values
12866 Show the current state of analysis messages printing for both frame argument
12867 values at function entry and tail calls.
12870 The analysis messages for tail calls can for example show why the virtual tail
12871 call frame for function @code{c} has not been recognized (due to the indirect
12872 reference by variable @code{x}):
12875 static void __attribute__((noinline, noclone)) c (void);
12876 void (*x) (void) = c;
12877 static void __attribute__((noinline, noclone)) a (void) @{ x++; @}
12878 static void __attribute__((noinline, noclone)) c (void) @{ a (); @}
12879 int main (void) @{ x (); return 0; @}
12881 Breakpoint 1, DW_OP_entry_value resolving cannot find
12882 DW_TAG_call_site 0x40039a in main
12884 3 static void __attribute__((noinline, noclone)) a (void) @{ x++; @}
12887 #1 0x000000000040039a in main () at t.c:5
12890 Another possibility is an ambiguous virtual tail call frames resolution:
12894 static void __attribute__((noinline, noclone)) f (void) @{ i++; @}
12895 static void __attribute__((noinline, noclone)) e (void) @{ f (); @}
12896 static void __attribute__((noinline, noclone)) d (void) @{ f (); @}
12897 static void __attribute__((noinline, noclone)) c (void) @{ d (); @}
12898 static void __attribute__((noinline, noclone)) b (void)
12899 @{ if (i) c (); else e (); @}
12900 static void __attribute__((noinline, noclone)) a (void) @{ b (); @}
12901 int main (void) @{ a (); return 0; @}
12903 tailcall: initial: 0x4004d2(a) 0x4004ce(b) 0x4004b2(c) 0x4004a2(d)
12904 tailcall: compare: 0x4004d2(a) 0x4004cc(b) 0x400492(e)
12905 tailcall: reduced: 0x4004d2(a) |
12908 #1 0x00000000004004d2 in a () at t.c:8
12909 #2 0x0000000000400395 in main () at t.c:9
12912 @set CALLSEQ1A @code{main@value{ARROW}a@value{ARROW}b@value{ARROW}c@value{ARROW}d@value{ARROW}f}
12913 @set CALLSEQ2A @code{main@value{ARROW}a@value{ARROW}b@value{ARROW}e@value{ARROW}f}
12915 @c Convert CALLSEQ#A to CALLSEQ#B depending on HAVE_MAKEINFO_CLICK.
12916 @ifset HAVE_MAKEINFO_CLICK
12917 @set ARROW @click{}
12918 @set CALLSEQ1B @clicksequence{@value{CALLSEQ1A}}
12919 @set CALLSEQ2B @clicksequence{@value{CALLSEQ2A}}
12921 @ifclear HAVE_MAKEINFO_CLICK
12923 @set CALLSEQ1B @value{CALLSEQ1A}
12924 @set CALLSEQ2B @value{CALLSEQ2A}
12927 Frames #0 and #2 are real, #1 is a virtual tail call frame.
12928 The code can have possible execution paths @value{CALLSEQ1B} or
12929 @value{CALLSEQ2B}, @value{GDBN} cannot find which one from the inferior state.
12931 @code{initial:} state shows some random possible calling sequence @value{GDBN}
12932 has found. It then finds another possible calling sequcen - that one is
12933 prefixed by @code{compare:}. The non-ambiguous intersection of these two is
12934 printed as the @code{reduced:} calling sequence. That one could have many
12935 futher @code{compare:} and @code{reduced:} statements as long as there remain
12936 any non-ambiguous sequence entries.
12938 For the frame of function @code{b} in both cases there are different possible
12939 @code{$pc} values (@code{0x4004cc} or @code{0x4004ce}), therefore this frame is
12940 also ambigous. The only non-ambiguous frame is the one for function @code{a},
12941 therefore this one is displayed to the user while the ambiguous frames are
12944 There can be also reasons why printing of frame argument values at function
12949 static void __attribute__((noinline, noclone)) c (int i) @{ v++; @}
12950 static void __attribute__((noinline, noclone)) a (int i);
12951 static void __attribute__((noinline, noclone)) b (int i) @{ a (i); @}
12952 static void __attribute__((noinline, noclone)) a (int i)
12953 @{ if (i) b (i - 1); else c (0); @}
12954 int main (void) @{ a (5); return 0; @}
12957 #0 c (i=i@@entry=0) at t.c:2
12958 #1 0x0000000000400428 in a (DW_OP_entry_value resolving has found
12959 function "a" at 0x400420 can call itself via tail calls
12960 i=<optimized out>) at t.c:6
12961 #2 0x000000000040036e in main () at t.c:7
12964 @value{GDBN} cannot find out from the inferior state if and how many times did
12965 function @code{a} call itself (via function @code{b}) as these calls would be
12966 tail calls. Such tail calls would modify thue @code{i} variable, therefore
12967 @value{GDBN} cannot be sure the value it knows would be right - @value{GDBN}
12968 prints @code{<optimized out>} instead.
12971 @chapter C Preprocessor Macros
12973 Some languages, such as C and C@t{++}, provide a way to define and invoke
12974 ``preprocessor macros'' which expand into strings of tokens.
12975 @value{GDBN} can evaluate expressions containing macro invocations, show
12976 the result of macro expansion, and show a macro's definition, including
12977 where it was defined.
12979 You may need to compile your program specially to provide @value{GDBN}
12980 with information about preprocessor macros. Most compilers do not
12981 include macros in their debugging information, even when you compile
12982 with the @option{-g} flag. @xref{Compilation}.
12984 A program may define a macro at one point, remove that definition later,
12985 and then provide a different definition after that. Thus, at different
12986 points in the program, a macro may have different definitions, or have
12987 no definition at all. If there is a current stack frame, @value{GDBN}
12988 uses the macros in scope at that frame's source code line. Otherwise,
12989 @value{GDBN} uses the macros in scope at the current listing location;
12992 Whenever @value{GDBN} evaluates an expression, it always expands any
12993 macro invocations present in the expression. @value{GDBN} also provides
12994 the following commands for working with macros explicitly.
12998 @kindex macro expand
12999 @cindex macro expansion, showing the results of preprocessor
13000 @cindex preprocessor macro expansion, showing the results of
13001 @cindex expanding preprocessor macros
13002 @item macro expand @var{expression}
13003 @itemx macro exp @var{expression}
13004 Show the results of expanding all preprocessor macro invocations in
13005 @var{expression}. Since @value{GDBN} simply expands macros, but does
13006 not parse the result, @var{expression} need not be a valid expression;
13007 it can be any string of tokens.
13010 @item macro expand-once @var{expression}
13011 @itemx macro exp1 @var{expression}
13012 @cindex expand macro once
13013 @i{(This command is not yet implemented.)} Show the results of
13014 expanding those preprocessor macro invocations that appear explicitly in
13015 @var{expression}. Macro invocations appearing in that expansion are
13016 left unchanged. This command allows you to see the effect of a
13017 particular macro more clearly, without being confused by further
13018 expansions. Since @value{GDBN} simply expands macros, but does not
13019 parse the result, @var{expression} need not be a valid expression; it
13020 can be any string of tokens.
13023 @cindex macro definition, showing
13024 @cindex definition of a macro, showing
13025 @cindex macros, from debug info
13026 @item info macro [-a|-all] [--] @var{macro}
13027 Show the current definition or all definitions of the named @var{macro},
13028 and describe the source location or compiler command-line where that
13029 definition was established. The optional double dash is to signify the end of
13030 argument processing and the beginning of @var{macro} for non C-like macros where
13031 the macro may begin with a hyphen.
13033 @kindex info macros
13034 @item info macros @var{location}
13035 Show all macro definitions that are in effect at the location specified
13036 by @var{location}, and describe the source location or compiler
13037 command-line where those definitions were established.
13039 @kindex macro define
13040 @cindex user-defined macros
13041 @cindex defining macros interactively
13042 @cindex macros, user-defined
13043 @item macro define @var{macro} @var{replacement-list}
13044 @itemx macro define @var{macro}(@var{arglist}) @var{replacement-list}
13045 Introduce a definition for a preprocessor macro named @var{macro},
13046 invocations of which are replaced by the tokens given in
13047 @var{replacement-list}. The first form of this command defines an
13048 ``object-like'' macro, which takes no arguments; the second form
13049 defines a ``function-like'' macro, which takes the arguments given in
13052 A definition introduced by this command is in scope in every
13053 expression evaluated in @value{GDBN}, until it is removed with the
13054 @code{macro undef} command, described below. The definition overrides
13055 all definitions for @var{macro} present in the program being debugged,
13056 as well as any previous user-supplied definition.
13058 @kindex macro undef
13059 @item macro undef @var{macro}
13060 Remove any user-supplied definition for the macro named @var{macro}.
13061 This command only affects definitions provided with the @code{macro
13062 define} command, described above; it cannot remove definitions present
13063 in the program being debugged.
13067 List all the macros defined using the @code{macro define} command.
13070 @cindex macros, example of debugging with
13071 Here is a transcript showing the above commands in action. First, we
13072 show our source files:
13077 #include "sample.h"
13080 #define ADD(x) (M + x)
13085 printf ("Hello, world!\n");
13087 printf ("We're so creative.\n");
13089 printf ("Goodbye, world!\n");
13096 Now, we compile the program using the @sc{gnu} C compiler,
13097 @value{NGCC}. We pass the @option{-gdwarf-2}@footnote{This is the
13098 minimum. Recent versions of @value{NGCC} support @option{-gdwarf-3}
13099 and @option{-gdwarf-4}; we recommend always choosing the most recent
13100 version of DWARF.} @emph{and} @option{-g3} flags to ensure the compiler
13101 includes information about preprocessor macros in the debugging
13105 $ gcc -gdwarf-2 -g3 sample.c -o sample
13109 Now, we start @value{GDBN} on our sample program:
13113 GNU gdb 2002-05-06-cvs
13114 Copyright 2002 Free Software Foundation, Inc.
13115 GDB is free software, @dots{}
13119 We can expand macros and examine their definitions, even when the
13120 program is not running. @value{GDBN} uses the current listing position
13121 to decide which macro definitions are in scope:
13124 (@value{GDBP}) list main
13127 5 #define ADD(x) (M + x)
13132 10 printf ("Hello, world!\n");
13134 12 printf ("We're so creative.\n");
13135 (@value{GDBP}) info macro ADD
13136 Defined at /home/jimb/gdb/macros/play/sample.c:5
13137 #define ADD(x) (M + x)
13138 (@value{GDBP}) info macro Q
13139 Defined at /home/jimb/gdb/macros/play/sample.h:1
13140 included at /home/jimb/gdb/macros/play/sample.c:2
13142 (@value{GDBP}) macro expand ADD(1)
13143 expands to: (42 + 1)
13144 (@value{GDBP}) macro expand-once ADD(1)
13145 expands to: once (M + 1)
13149 In the example above, note that @code{macro expand-once} expands only
13150 the macro invocation explicit in the original text --- the invocation of
13151 @code{ADD} --- but does not expand the invocation of the macro @code{M},
13152 which was introduced by @code{ADD}.
13154 Once the program is running, @value{GDBN} uses the macro definitions in
13155 force at the source line of the current stack frame:
13158 (@value{GDBP}) break main
13159 Breakpoint 1 at 0x8048370: file sample.c, line 10.
13161 Starting program: /home/jimb/gdb/macros/play/sample
13163 Breakpoint 1, main () at sample.c:10
13164 10 printf ("Hello, world!\n");
13168 At line 10, the definition of the macro @code{N} at line 9 is in force:
13171 (@value{GDBP}) info macro N
13172 Defined at /home/jimb/gdb/macros/play/sample.c:9
13174 (@value{GDBP}) macro expand N Q M
13175 expands to: 28 < 42
13176 (@value{GDBP}) print N Q M
13181 As we step over directives that remove @code{N}'s definition, and then
13182 give it a new definition, @value{GDBN} finds the definition (or lack
13183 thereof) in force at each point:
13186 (@value{GDBP}) next
13188 12 printf ("We're so creative.\n");
13189 (@value{GDBP}) info macro N
13190 The symbol `N' has no definition as a C/C++ preprocessor macro
13191 at /home/jimb/gdb/macros/play/sample.c:12
13192 (@value{GDBP}) next
13194 14 printf ("Goodbye, world!\n");
13195 (@value{GDBP}) info macro N
13196 Defined at /home/jimb/gdb/macros/play/sample.c:13
13198 (@value{GDBP}) macro expand N Q M
13199 expands to: 1729 < 42
13200 (@value{GDBP}) print N Q M
13205 In addition to source files, macros can be defined on the compilation command
13206 line using the @option{-D@var{name}=@var{value}} syntax. For macros defined in
13207 such a way, @value{GDBN} displays the location of their definition as line zero
13208 of the source file submitted to the compiler.
13211 (@value{GDBP}) info macro __STDC__
13212 Defined at /home/jimb/gdb/macros/play/sample.c:0
13219 @chapter Tracepoints
13220 @c This chapter is based on the documentation written by Michael
13221 @c Snyder, David Taylor, Jim Blandy, and Elena Zannoni.
13223 @cindex tracepoints
13224 In some applications, it is not feasible for the debugger to interrupt
13225 the program's execution long enough for the developer to learn
13226 anything helpful about its behavior. If the program's correctness
13227 depends on its real-time behavior, delays introduced by a debugger
13228 might cause the program to change its behavior drastically, or perhaps
13229 fail, even when the code itself is correct. It is useful to be able
13230 to observe the program's behavior without interrupting it.
13232 Using @value{GDBN}'s @code{trace} and @code{collect} commands, you can
13233 specify locations in the program, called @dfn{tracepoints}, and
13234 arbitrary expressions to evaluate when those tracepoints are reached.
13235 Later, using the @code{tfind} command, you can examine the values
13236 those expressions had when the program hit the tracepoints. The
13237 expressions may also denote objects in memory---structures or arrays,
13238 for example---whose values @value{GDBN} should record; while visiting
13239 a particular tracepoint, you may inspect those objects as if they were
13240 in memory at that moment. However, because @value{GDBN} records these
13241 values without interacting with you, it can do so quickly and
13242 unobtrusively, hopefully not disturbing the program's behavior.
13244 The tracepoint facility is currently available only for remote
13245 targets. @xref{Targets}. In addition, your remote target must know
13246 how to collect trace data. This functionality is implemented in the
13247 remote stub; however, none of the stubs distributed with @value{GDBN}
13248 support tracepoints as of this writing. The format of the remote
13249 packets used to implement tracepoints are described in @ref{Tracepoint
13252 It is also possible to get trace data from a file, in a manner reminiscent
13253 of corefiles; you specify the filename, and use @code{tfind} to search
13254 through the file. @xref{Trace Files}, for more details.
13256 This chapter describes the tracepoint commands and features.
13259 * Set Tracepoints::
13260 * Analyze Collected Data::
13261 * Tracepoint Variables::
13265 @node Set Tracepoints
13266 @section Commands to Set Tracepoints
13268 Before running such a @dfn{trace experiment}, an arbitrary number of
13269 tracepoints can be set. A tracepoint is actually a special type of
13270 breakpoint (@pxref{Set Breaks}), so you can manipulate it using
13271 standard breakpoint commands. For instance, as with breakpoints,
13272 tracepoint numbers are successive integers starting from one, and many
13273 of the commands associated with tracepoints take the tracepoint number
13274 as their argument, to identify which tracepoint to work on.
13276 For each tracepoint, you can specify, in advance, some arbitrary set
13277 of data that you want the target to collect in the trace buffer when
13278 it hits that tracepoint. The collected data can include registers,
13279 local variables, or global data. Later, you can use @value{GDBN}
13280 commands to examine the values these data had at the time the
13281 tracepoint was hit.
13283 Tracepoints do not support every breakpoint feature. Ignore counts on
13284 tracepoints have no effect, and tracepoints cannot run @value{GDBN}
13285 commands when they are hit. Tracepoints may not be thread-specific
13288 @cindex fast tracepoints
13289 Some targets may support @dfn{fast tracepoints}, which are inserted in
13290 a different way (such as with a jump instead of a trap), that is
13291 faster but possibly restricted in where they may be installed.
13293 @cindex static tracepoints
13294 @cindex markers, static tracepoints
13295 @cindex probing markers, static tracepoints
13296 Regular and fast tracepoints are dynamic tracing facilities, meaning
13297 that they can be used to insert tracepoints at (almost) any location
13298 in the target. Some targets may also support controlling @dfn{static
13299 tracepoints} from @value{GDBN}. With static tracing, a set of
13300 instrumentation points, also known as @dfn{markers}, are embedded in
13301 the target program, and can be activated or deactivated by name or
13302 address. These are usually placed at locations which facilitate
13303 investigating what the target is actually doing. @value{GDBN}'s
13304 support for static tracing includes being able to list instrumentation
13305 points, and attach them with @value{GDBN} defined high level
13306 tracepoints that expose the whole range of convenience of
13307 @value{GDBN}'s tracepoints support. Namely, support for collecting
13308 registers values and values of global or local (to the instrumentation
13309 point) variables; tracepoint conditions and trace state variables.
13310 The act of installing a @value{GDBN} static tracepoint on an
13311 instrumentation point, or marker, is referred to as @dfn{probing} a
13312 static tracepoint marker.
13314 @code{gdbserver} supports tracepoints on some target systems.
13315 @xref{Server,,Tracepoints support in @code{gdbserver}}.
13317 This section describes commands to set tracepoints and associated
13318 conditions and actions.
13321 * Create and Delete Tracepoints::
13322 * Enable and Disable Tracepoints::
13323 * Tracepoint Passcounts::
13324 * Tracepoint Conditions::
13325 * Trace State Variables::
13326 * Tracepoint Actions::
13327 * Listing Tracepoints::
13328 * Listing Static Tracepoint Markers::
13329 * Starting and Stopping Trace Experiments::
13330 * Tracepoint Restrictions::
13333 @node Create and Delete Tracepoints
13334 @subsection Create and Delete Tracepoints
13337 @cindex set tracepoint
13339 @item trace @var{location}
13340 The @code{trace} command is very similar to the @code{break} command.
13341 Its argument @var{location} can be any valid location.
13342 @xref{Specify Location}. The @code{trace} command defines a tracepoint,
13343 which is a point in the target program where the debugger will briefly stop,
13344 collect some data, and then allow the program to continue. Setting a tracepoint
13345 or changing its actions takes effect immediately if the remote stub
13346 supports the @samp{InstallInTrace} feature (@pxref{install tracepoint
13348 If remote stub doesn't support the @samp{InstallInTrace} feature, all
13349 these changes don't take effect until the next @code{tstart}
13350 command, and once a trace experiment is running, further changes will
13351 not have any effect until the next trace experiment starts. In addition,
13352 @value{GDBN} supports @dfn{pending tracepoints}---tracepoints whose
13353 address is not yet resolved. (This is similar to pending breakpoints.)
13354 Pending tracepoints are not downloaded to the target and not installed
13355 until they are resolved. The resolution of pending tracepoints requires
13356 @value{GDBN} support---when debugging with the remote target, and
13357 @value{GDBN} disconnects from the remote stub (@pxref{disconnected
13358 tracing}), pending tracepoints can not be resolved (and downloaded to
13359 the remote stub) while @value{GDBN} is disconnected.
13361 Here are some examples of using the @code{trace} command:
13364 (@value{GDBP}) @b{trace foo.c:121} // a source file and line number
13366 (@value{GDBP}) @b{trace +2} // 2 lines forward
13368 (@value{GDBP}) @b{trace my_function} // first source line of function
13370 (@value{GDBP}) @b{trace *my_function} // EXACT start address of function
13372 (@value{GDBP}) @b{trace *0x2117c4} // an address
13376 You can abbreviate @code{trace} as @code{tr}.
13378 @item trace @var{location} if @var{cond}
13379 Set a tracepoint with condition @var{cond}; evaluate the expression
13380 @var{cond} each time the tracepoint is reached, and collect data only
13381 if the value is nonzero---that is, if @var{cond} evaluates as true.
13382 @xref{Tracepoint Conditions, ,Tracepoint Conditions}, for more
13383 information on tracepoint conditions.
13385 @item ftrace @var{location} [ if @var{cond} ]
13386 @cindex set fast tracepoint
13387 @cindex fast tracepoints, setting
13389 The @code{ftrace} command sets a fast tracepoint. For targets that
13390 support them, fast tracepoints will use a more efficient but possibly
13391 less general technique to trigger data collection, such as a jump
13392 instruction instead of a trap, or some sort of hardware support. It
13393 may not be possible to create a fast tracepoint at the desired
13394 location, in which case the command will exit with an explanatory
13397 @value{GDBN} handles arguments to @code{ftrace} exactly as for
13400 On 32-bit x86-architecture systems, fast tracepoints normally need to
13401 be placed at an instruction that is 5 bytes or longer, but can be
13402 placed at 4-byte instructions if the low 64K of memory of the target
13403 program is available to install trampolines. Some Unix-type systems,
13404 such as @sc{gnu}/Linux, exclude low addresses from the program's
13405 address space; but for instance with the Linux kernel it is possible
13406 to let @value{GDBN} use this area by doing a @command{sysctl} command
13407 to set the @code{mmap_min_addr} kernel parameter, as in
13410 sudo sysctl -w vm.mmap_min_addr=32768
13414 which sets the low address to 32K, which leaves plenty of room for
13415 trampolines. The minimum address should be set to a page boundary.
13417 @item strace @var{location} [ if @var{cond} ]
13418 @cindex set static tracepoint
13419 @cindex static tracepoints, setting
13420 @cindex probe static tracepoint marker
13422 The @code{strace} command sets a static tracepoint. For targets that
13423 support it, setting a static tracepoint probes a static
13424 instrumentation point, or marker, found at @var{location}. It may not
13425 be possible to set a static tracepoint at the desired location, in
13426 which case the command will exit with an explanatory message.
13428 @value{GDBN} handles arguments to @code{strace} exactly as for
13429 @code{trace}, with the addition that the user can also specify
13430 @code{-m @var{marker}} as @var{location}. This probes the marker
13431 identified by the @var{marker} string identifier. This identifier
13432 depends on the static tracepoint backend library your program is
13433 using. You can find all the marker identifiers in the @samp{ID} field
13434 of the @code{info static-tracepoint-markers} command output.
13435 @xref{Listing Static Tracepoint Markers,,Listing Static Tracepoint
13436 Markers}. For example, in the following small program using the UST
13442 trace_mark(ust, bar33, "str %s", "FOOBAZ");
13447 the marker id is composed of joining the first two arguments to the
13448 @code{trace_mark} call with a slash, which translates to:
13451 (@value{GDBP}) info static-tracepoint-markers
13452 Cnt Enb ID Address What
13453 1 n ust/bar33 0x0000000000400ddc in main at stexample.c:22
13459 so you may probe the marker above with:
13462 (@value{GDBP}) strace -m ust/bar33
13465 Static tracepoints accept an extra collect action --- @code{collect
13466 $_sdata}. This collects arbitrary user data passed in the probe point
13467 call to the tracing library. In the UST example above, you'll see
13468 that the third argument to @code{trace_mark} is a printf-like format
13469 string. The user data is then the result of running that formating
13470 string against the following arguments. Note that @code{info
13471 static-tracepoint-markers} command output lists that format string in
13472 the @samp{Data:} field.
13474 You can inspect this data when analyzing the trace buffer, by printing
13475 the $_sdata variable like any other variable available to
13476 @value{GDBN}. @xref{Tracepoint Actions,,Tracepoint Action Lists}.
13479 @cindex last tracepoint number
13480 @cindex recent tracepoint number
13481 @cindex tracepoint number
13482 The convenience variable @code{$tpnum} records the tracepoint number
13483 of the most recently set tracepoint.
13485 @kindex delete tracepoint
13486 @cindex tracepoint deletion
13487 @item delete tracepoint @r{[}@var{num}@r{]}
13488 Permanently delete one or more tracepoints. With no argument, the
13489 default is to delete all tracepoints. Note that the regular
13490 @code{delete} command can remove tracepoints also.
13495 (@value{GDBP}) @b{delete trace 1 2 3} // remove three tracepoints
13497 (@value{GDBP}) @b{delete trace} // remove all tracepoints
13501 You can abbreviate this command as @code{del tr}.
13504 @node Enable and Disable Tracepoints
13505 @subsection Enable and Disable Tracepoints
13507 These commands are deprecated; they are equivalent to plain @code{disable} and @code{enable}.
13510 @kindex disable tracepoint
13511 @item disable tracepoint @r{[}@var{num}@r{]}
13512 Disable tracepoint @var{num}, or all tracepoints if no argument
13513 @var{num} is given. A disabled tracepoint will have no effect during
13514 a trace experiment, but it is not forgotten. You can re-enable
13515 a disabled tracepoint using the @code{enable tracepoint} command.
13516 If the command is issued during a trace experiment and the debug target
13517 has support for disabling tracepoints during a trace experiment, then the
13518 change will be effective immediately. Otherwise, it will be applied to the
13519 next trace experiment.
13521 @kindex enable tracepoint
13522 @item enable tracepoint @r{[}@var{num}@r{]}
13523 Enable tracepoint @var{num}, or all tracepoints. If this command is
13524 issued during a trace experiment and the debug target supports enabling
13525 tracepoints during a trace experiment, then the enabled tracepoints will
13526 become effective immediately. Otherwise, they will become effective the
13527 next time a trace experiment is run.
13530 @node Tracepoint Passcounts
13531 @subsection Tracepoint Passcounts
13535 @cindex tracepoint pass count
13536 @item passcount @r{[}@var{n} @r{[}@var{num}@r{]]}
13537 Set the @dfn{passcount} of a tracepoint. The passcount is a way to
13538 automatically stop a trace experiment. If a tracepoint's passcount is
13539 @var{n}, then the trace experiment will be automatically stopped on
13540 the @var{n}'th time that tracepoint is hit. If the tracepoint number
13541 @var{num} is not specified, the @code{passcount} command sets the
13542 passcount of the most recently defined tracepoint. If no passcount is
13543 given, the trace experiment will run until stopped explicitly by the
13549 (@value{GDBP}) @b{passcount 5 2} // Stop on the 5th execution of
13550 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// tracepoint 2}
13552 (@value{GDBP}) @b{passcount 12} // Stop on the 12th execution of the
13553 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// most recently defined tracepoint.}
13554 (@value{GDBP}) @b{trace foo}
13555 (@value{GDBP}) @b{pass 3}
13556 (@value{GDBP}) @b{trace bar}
13557 (@value{GDBP}) @b{pass 2}
13558 (@value{GDBP}) @b{trace baz}
13559 (@value{GDBP}) @b{pass 1} // Stop tracing when foo has been
13560 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// executed 3 times OR when bar has}
13561 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// been executed 2 times}
13562 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// OR when baz has been executed 1 time.}
13566 @node Tracepoint Conditions
13567 @subsection Tracepoint Conditions
13568 @cindex conditional tracepoints
13569 @cindex tracepoint conditions
13571 The simplest sort of tracepoint collects data every time your program
13572 reaches a specified place. You can also specify a @dfn{condition} for
13573 a tracepoint. A condition is just a Boolean expression in your
13574 programming language (@pxref{Expressions, ,Expressions}). A
13575 tracepoint with a condition evaluates the expression each time your
13576 program reaches it, and data collection happens only if the condition
13579 Tracepoint conditions can be specified when a tracepoint is set, by
13580 using @samp{if} in the arguments to the @code{trace} command.
13581 @xref{Create and Delete Tracepoints, ,Setting Tracepoints}. They can
13582 also be set or changed at any time with the @code{condition} command,
13583 just as with breakpoints.
13585 Unlike breakpoint conditions, @value{GDBN} does not actually evaluate
13586 the conditional expression itself. Instead, @value{GDBN} encodes the
13587 expression into an agent expression (@pxref{Agent Expressions})
13588 suitable for execution on the target, independently of @value{GDBN}.
13589 Global variables become raw memory locations, locals become stack
13590 accesses, and so forth.
13592 For instance, suppose you have a function that is usually called
13593 frequently, but should not be called after an error has occurred. You
13594 could use the following tracepoint command to collect data about calls
13595 of that function that happen while the error code is propagating
13596 through the program; an unconditional tracepoint could end up
13597 collecting thousands of useless trace frames that you would have to
13601 (@value{GDBP}) @kbd{trace normal_operation if errcode > 0}
13604 @node Trace State Variables
13605 @subsection Trace State Variables
13606 @cindex trace state variables
13608 A @dfn{trace state variable} is a special type of variable that is
13609 created and managed by target-side code. The syntax is the same as
13610 that for GDB's convenience variables (a string prefixed with ``$''),
13611 but they are stored on the target. They must be created explicitly,
13612 using a @code{tvariable} command. They are always 64-bit signed
13615 Trace state variables are remembered by @value{GDBN}, and downloaded
13616 to the target along with tracepoint information when the trace
13617 experiment starts. There are no intrinsic limits on the number of
13618 trace state variables, beyond memory limitations of the target.
13620 @cindex convenience variables, and trace state variables
13621 Although trace state variables are managed by the target, you can use
13622 them in print commands and expressions as if they were convenience
13623 variables; @value{GDBN} will get the current value from the target
13624 while the trace experiment is running. Trace state variables share
13625 the same namespace as other ``$'' variables, which means that you
13626 cannot have trace state variables with names like @code{$23} or
13627 @code{$pc}, nor can you have a trace state variable and a convenience
13628 variable with the same name.
13632 @item tvariable $@var{name} [ = @var{expression} ]
13634 The @code{tvariable} command creates a new trace state variable named
13635 @code{$@var{name}}, and optionally gives it an initial value of
13636 @var{expression}. The @var{expression} is evaluated when this command is
13637 entered; the result will be converted to an integer if possible,
13638 otherwise @value{GDBN} will report an error. A subsequent
13639 @code{tvariable} command specifying the same name does not create a
13640 variable, but instead assigns the supplied initial value to the
13641 existing variable of that name, overwriting any previous initial
13642 value. The default initial value is 0.
13644 @item info tvariables
13645 @kindex info tvariables
13646 List all the trace state variables along with their initial values.
13647 Their current values may also be displayed, if the trace experiment is
13650 @item delete tvariable @r{[} $@var{name} @dots{} @r{]}
13651 @kindex delete tvariable
13652 Delete the given trace state variables, or all of them if no arguments
13657 @node Tracepoint Actions
13658 @subsection Tracepoint Action Lists
13662 @cindex tracepoint actions
13663 @item actions @r{[}@var{num}@r{]}
13664 This command will prompt for a list of actions to be taken when the
13665 tracepoint is hit. If the tracepoint number @var{num} is not
13666 specified, this command sets the actions for the one that was most
13667 recently defined (so that you can define a tracepoint and then say
13668 @code{actions} without bothering about its number). You specify the
13669 actions themselves on the following lines, one action at a time, and
13670 terminate the actions list with a line containing just @code{end}. So
13671 far, the only defined actions are @code{collect}, @code{teval}, and
13672 @code{while-stepping}.
13674 @code{actions} is actually equivalent to @code{commands} (@pxref{Break
13675 Commands, ,Breakpoint Command Lists}), except that only the defined
13676 actions are allowed; any other @value{GDBN} command is rejected.
13678 @cindex remove actions from a tracepoint
13679 To remove all actions from a tracepoint, type @samp{actions @var{num}}
13680 and follow it immediately with @samp{end}.
13683 (@value{GDBP}) @b{collect @var{data}} // collect some data
13685 (@value{GDBP}) @b{while-stepping 5} // single-step 5 times, collect data
13687 (@value{GDBP}) @b{end} // signals the end of actions.
13690 In the following example, the action list begins with @code{collect}
13691 commands indicating the things to be collected when the tracepoint is
13692 hit. Then, in order to single-step and collect additional data
13693 following the tracepoint, a @code{while-stepping} command is used,
13694 followed by the list of things to be collected after each step in a
13695 sequence of single steps. The @code{while-stepping} command is
13696 terminated by its own separate @code{end} command. Lastly, the action
13697 list is terminated by an @code{end} command.
13700 (@value{GDBP}) @b{trace foo}
13701 (@value{GDBP}) @b{actions}
13702 Enter actions for tracepoint 1, one per line:
13705 > while-stepping 12
13706 > collect $pc, arr[i]
13711 @kindex collect @r{(tracepoints)}
13712 @item collect@r{[}/@var{mods}@r{]} @var{expr1}, @var{expr2}, @dots{}
13713 Collect values of the given expressions when the tracepoint is hit.
13714 This command accepts a comma-separated list of any valid expressions.
13715 In addition to global, static, or local variables, the following
13716 special arguments are supported:
13720 Collect all registers.
13723 Collect all function arguments.
13726 Collect all local variables.
13729 Collect the return address. This is helpful if you want to see more
13732 @emph{Note:} The return address location can not always be reliably
13733 determined up front, and the wrong address / registers may end up
13734 collected instead. On some architectures the reliability is higher
13735 for tracepoints at function entry, while on others it's the opposite.
13736 When this happens, backtracing will stop because the return address is
13737 found unavailable (unless another collect rule happened to match it).
13740 Collects the number of arguments from the static probe at which the
13741 tracepoint is located.
13742 @xref{Static Probe Points}.
13744 @item $_probe_arg@var{n}
13745 @var{n} is an integer between 0 and 11. Collects the @var{n}th argument
13746 from the static probe at which the tracepoint is located.
13747 @xref{Static Probe Points}.
13750 @vindex $_sdata@r{, collect}
13751 Collect static tracepoint marker specific data. Only available for
13752 static tracepoints. @xref{Tracepoint Actions,,Tracepoint Action
13753 Lists}. On the UST static tracepoints library backend, an
13754 instrumentation point resembles a @code{printf} function call. The
13755 tracing library is able to collect user specified data formatted to a
13756 character string using the format provided by the programmer that
13757 instrumented the program. Other backends have similar mechanisms.
13758 Here's an example of a UST marker call:
13761 const char master_name[] = "$your_name";
13762 trace_mark(channel1, marker1, "hello %s", master_name)
13765 In this case, collecting @code{$_sdata} collects the string
13766 @samp{hello $yourname}. When analyzing the trace buffer, you can
13767 inspect @samp{$_sdata} like any other variable available to
13771 You can give several consecutive @code{collect} commands, each one
13772 with a single argument, or one @code{collect} command with several
13773 arguments separated by commas; the effect is the same.
13775 The optional @var{mods} changes the usual handling of the arguments.
13776 @code{s} requests that pointers to chars be handled as strings, in
13777 particular collecting the contents of the memory being pointed at, up
13778 to the first zero. The upper bound is by default the value of the
13779 @code{print elements} variable; if @code{s} is followed by a decimal
13780 number, that is the upper bound instead. So for instance
13781 @samp{collect/s25 mystr} collects as many as 25 characters at
13784 The command @code{info scope} (@pxref{Symbols, info scope}) is
13785 particularly useful for figuring out what data to collect.
13787 @kindex teval @r{(tracepoints)}
13788 @item teval @var{expr1}, @var{expr2}, @dots{}
13789 Evaluate the given expressions when the tracepoint is hit. This
13790 command accepts a comma-separated list of expressions. The results
13791 are discarded, so this is mainly useful for assigning values to trace
13792 state variables (@pxref{Trace State Variables}) without adding those
13793 values to the trace buffer, as would be the case if the @code{collect}
13796 @kindex while-stepping @r{(tracepoints)}
13797 @item while-stepping @var{n}
13798 Perform @var{n} single-step instruction traces after the tracepoint,
13799 collecting new data after each step. The @code{while-stepping}
13800 command is followed by the list of what to collect while stepping
13801 (followed by its own @code{end} command):
13804 > while-stepping 12
13805 > collect $regs, myglobal
13811 Note that @code{$pc} is not automatically collected by
13812 @code{while-stepping}; you need to explicitly collect that register if
13813 you need it. You may abbreviate @code{while-stepping} as @code{ws} or
13816 @item set default-collect @var{expr1}, @var{expr2}, @dots{}
13817 @kindex set default-collect
13818 @cindex default collection action
13819 This variable is a list of expressions to collect at each tracepoint
13820 hit. It is effectively an additional @code{collect} action prepended
13821 to every tracepoint action list. The expressions are parsed
13822 individually for each tracepoint, so for instance a variable named
13823 @code{xyz} may be interpreted as a global for one tracepoint, and a
13824 local for another, as appropriate to the tracepoint's location.
13826 @item show default-collect
13827 @kindex show default-collect
13828 Show the list of expressions that are collected by default at each
13833 @node Listing Tracepoints
13834 @subsection Listing Tracepoints
13837 @kindex info tracepoints @r{[}@var{n}@dots{}@r{]}
13838 @kindex info tp @r{[}@var{n}@dots{}@r{]}
13839 @cindex information about tracepoints
13840 @item info tracepoints @r{[}@var{num}@dots{}@r{]}
13841 Display information about the tracepoint @var{num}. If you don't
13842 specify a tracepoint number, displays information about all the
13843 tracepoints defined so far. The format is similar to that used for
13844 @code{info breakpoints}; in fact, @code{info tracepoints} is the same
13845 command, simply restricting itself to tracepoints.
13847 A tracepoint's listing may include additional information specific to
13852 its passcount as given by the @code{passcount @var{n}} command
13855 the state about installed on target of each location
13859 (@value{GDBP}) @b{info trace}
13860 Num Type Disp Enb Address What
13861 1 tracepoint keep y 0x0804ab57 in foo() at main.cxx:7
13863 collect globfoo, $regs
13868 2 tracepoint keep y <MULTIPLE>
13870 2.1 y 0x0804859c in func4 at change-loc.h:35
13871 installed on target
13872 2.2 y 0xb7ffc480 in func4 at change-loc.h:35
13873 installed on target
13874 2.3 y <PENDING> set_tracepoint
13875 3 tracepoint keep y 0x080485b1 in foo at change-loc.c:29
13876 not installed on target
13881 This command can be abbreviated @code{info tp}.
13884 @node Listing Static Tracepoint Markers
13885 @subsection Listing Static Tracepoint Markers
13888 @kindex info static-tracepoint-markers
13889 @cindex information about static tracepoint markers
13890 @item info static-tracepoint-markers
13891 Display information about all static tracepoint markers defined in the
13894 For each marker, the following columns are printed:
13898 An incrementing counter, output to help readability. This is not a
13901 The marker ID, as reported by the target.
13902 @item Enabled or Disabled
13903 Probed markers are tagged with @samp{y}. @samp{n} identifies marks
13904 that are not enabled.
13906 Where the marker is in your program, as a memory address.
13908 Where the marker is in the source for your program, as a file and line
13909 number. If the debug information included in the program does not
13910 allow @value{GDBN} to locate the source of the marker, this column
13911 will be left blank.
13915 In addition, the following information may be printed for each marker:
13919 User data passed to the tracing library by the marker call. In the
13920 UST backend, this is the format string passed as argument to the
13922 @item Static tracepoints probing the marker
13923 The list of static tracepoints attached to the marker.
13927 (@value{GDBP}) info static-tracepoint-markers
13928 Cnt ID Enb Address What
13929 1 ust/bar2 y 0x0000000000400e1a in main at stexample.c:25
13930 Data: number1 %d number2 %d
13931 Probed by static tracepoints: #2
13932 2 ust/bar33 n 0x0000000000400c87 in main at stexample.c:24
13938 @node Starting and Stopping Trace Experiments
13939 @subsection Starting and Stopping Trace Experiments
13942 @kindex tstart [ @var{notes} ]
13943 @cindex start a new trace experiment
13944 @cindex collected data discarded
13946 This command starts the trace experiment, and begins collecting data.
13947 It has the side effect of discarding all the data collected in the
13948 trace buffer during the previous trace experiment. If any arguments
13949 are supplied, they are taken as a note and stored with the trace
13950 experiment's state. The notes may be arbitrary text, and are
13951 especially useful with disconnected tracing in a multi-user context;
13952 the notes can explain what the trace is doing, supply user contact
13953 information, and so forth.
13955 @kindex tstop [ @var{notes} ]
13956 @cindex stop a running trace experiment
13958 This command stops the trace experiment. If any arguments are
13959 supplied, they are recorded with the experiment as a note. This is
13960 useful if you are stopping a trace started by someone else, for
13961 instance if the trace is interfering with the system's behavior and
13962 needs to be stopped quickly.
13964 @strong{Note}: a trace experiment and data collection may stop
13965 automatically if any tracepoint's passcount is reached
13966 (@pxref{Tracepoint Passcounts}), or if the trace buffer becomes full.
13969 @cindex status of trace data collection
13970 @cindex trace experiment, status of
13972 This command displays the status of the current trace data
13976 Here is an example of the commands we described so far:
13979 (@value{GDBP}) @b{trace gdb_c_test}
13980 (@value{GDBP}) @b{actions}
13981 Enter actions for tracepoint #1, one per line.
13982 > collect $regs,$locals,$args
13983 > while-stepping 11
13987 (@value{GDBP}) @b{tstart}
13988 [time passes @dots{}]
13989 (@value{GDBP}) @b{tstop}
13992 @anchor{disconnected tracing}
13993 @cindex disconnected tracing
13994 You can choose to continue running the trace experiment even if
13995 @value{GDBN} disconnects from the target, voluntarily or
13996 involuntarily. For commands such as @code{detach}, the debugger will
13997 ask what you want to do with the trace. But for unexpected
13998 terminations (@value{GDBN} crash, network outage), it would be
13999 unfortunate to lose hard-won trace data, so the variable
14000 @code{disconnected-tracing} lets you decide whether the trace should
14001 continue running without @value{GDBN}.
14004 @item set disconnected-tracing on
14005 @itemx set disconnected-tracing off
14006 @kindex set disconnected-tracing
14007 Choose whether a tracing run should continue to run if @value{GDBN}
14008 has disconnected from the target. Note that @code{detach} or
14009 @code{quit} will ask you directly what to do about a running trace no
14010 matter what this variable's setting, so the variable is mainly useful
14011 for handling unexpected situations, such as loss of the network.
14013 @item show disconnected-tracing
14014 @kindex show disconnected-tracing
14015 Show the current choice for disconnected tracing.
14019 When you reconnect to the target, the trace experiment may or may not
14020 still be running; it might have filled the trace buffer in the
14021 meantime, or stopped for one of the other reasons. If it is running,
14022 it will continue after reconnection.
14024 Upon reconnection, the target will upload information about the
14025 tracepoints in effect. @value{GDBN} will then compare that
14026 information to the set of tracepoints currently defined, and attempt
14027 to match them up, allowing for the possibility that the numbers may
14028 have changed due to creation and deletion in the meantime. If one of
14029 the target's tracepoints does not match any in @value{GDBN}, the
14030 debugger will create a new tracepoint, so that you have a number with
14031 which to specify that tracepoint. This matching-up process is
14032 necessarily heuristic, and it may result in useless tracepoints being
14033 created; you may simply delete them if they are of no use.
14035 @cindex circular trace buffer
14036 If your target agent supports a @dfn{circular trace buffer}, then you
14037 can run a trace experiment indefinitely without filling the trace
14038 buffer; when space runs out, the agent deletes already-collected trace
14039 frames, oldest first, until there is enough room to continue
14040 collecting. This is especially useful if your tracepoints are being
14041 hit too often, and your trace gets terminated prematurely because the
14042 buffer is full. To ask for a circular trace buffer, simply set
14043 @samp{circular-trace-buffer} to on. You can set this at any time,
14044 including during tracing; if the agent can do it, it will change
14045 buffer handling on the fly, otherwise it will not take effect until
14049 @item set circular-trace-buffer on
14050 @itemx set circular-trace-buffer off
14051 @kindex set circular-trace-buffer
14052 Choose whether a tracing run should use a linear or circular buffer
14053 for trace data. A linear buffer will not lose any trace data, but may
14054 fill up prematurely, while a circular buffer will discard old trace
14055 data, but it will have always room for the latest tracepoint hits.
14057 @item show circular-trace-buffer
14058 @kindex show circular-trace-buffer
14059 Show the current choice for the trace buffer. Note that this may not
14060 match the agent's current buffer handling, nor is it guaranteed to
14061 match the setting that might have been in effect during a past run,
14062 for instance if you are looking at frames from a trace file.
14067 @item set trace-buffer-size @var{n}
14068 @itemx set trace-buffer-size unlimited
14069 @kindex set trace-buffer-size
14070 Request that the target use a trace buffer of @var{n} bytes. Not all
14071 targets will honor the request; they may have a compiled-in size for
14072 the trace buffer, or some other limitation. Set to a value of
14073 @code{unlimited} or @code{-1} to let the target use whatever size it
14074 likes. This is also the default.
14076 @item show trace-buffer-size
14077 @kindex show trace-buffer-size
14078 Show the current requested size for the trace buffer. Note that this
14079 will only match the actual size if the target supports size-setting,
14080 and was able to handle the requested size. For instance, if the
14081 target can only change buffer size between runs, this variable will
14082 not reflect the change until the next run starts. Use @code{tstatus}
14083 to get a report of the actual buffer size.
14087 @item set trace-user @var{text}
14088 @kindex set trace-user
14090 @item show trace-user
14091 @kindex show trace-user
14093 @item set trace-notes @var{text}
14094 @kindex set trace-notes
14095 Set the trace run's notes.
14097 @item show trace-notes
14098 @kindex show trace-notes
14099 Show the trace run's notes.
14101 @item set trace-stop-notes @var{text}
14102 @kindex set trace-stop-notes
14103 Set the trace run's stop notes. The handling of the note is as for
14104 @code{tstop} arguments; the set command is convenient way to fix a
14105 stop note that is mistaken or incomplete.
14107 @item show trace-stop-notes
14108 @kindex show trace-stop-notes
14109 Show the trace run's stop notes.
14113 @node Tracepoint Restrictions
14114 @subsection Tracepoint Restrictions
14116 @cindex tracepoint restrictions
14117 There are a number of restrictions on the use of tracepoints. As
14118 described above, tracepoint data gathering occurs on the target
14119 without interaction from @value{GDBN}. Thus the full capabilities of
14120 the debugger are not available during data gathering, and then at data
14121 examination time, you will be limited by only having what was
14122 collected. The following items describe some common problems, but it
14123 is not exhaustive, and you may run into additional difficulties not
14129 Tracepoint expressions are intended to gather objects (lvalues). Thus
14130 the full flexibility of GDB's expression evaluator is not available.
14131 You cannot call functions, cast objects to aggregate types, access
14132 convenience variables or modify values (except by assignment to trace
14133 state variables). Some language features may implicitly call
14134 functions (for instance Objective-C fields with accessors), and therefore
14135 cannot be collected either.
14138 Collection of local variables, either individually or in bulk with
14139 @code{$locals} or @code{$args}, during @code{while-stepping} may
14140 behave erratically. The stepping action may enter a new scope (for
14141 instance by stepping into a function), or the location of the variable
14142 may change (for instance it is loaded into a register). The
14143 tracepoint data recorded uses the location information for the
14144 variables that is correct for the tracepoint location. When the
14145 tracepoint is created, it is not possible, in general, to determine
14146 where the steps of a @code{while-stepping} sequence will advance the
14147 program---particularly if a conditional branch is stepped.
14150 Collection of an incompletely-initialized or partially-destroyed object
14151 may result in something that @value{GDBN} cannot display, or displays
14152 in a misleading way.
14155 When @value{GDBN} displays a pointer to character it automatically
14156 dereferences the pointer to also display characters of the string
14157 being pointed to. However, collecting the pointer during tracing does
14158 not automatically collect the string. You need to explicitly
14159 dereference the pointer and provide size information if you want to
14160 collect not only the pointer, but the memory pointed to. For example,
14161 @code{*ptr@@50} can be used to collect the 50 element array pointed to
14165 It is not possible to collect a complete stack backtrace at a
14166 tracepoint. Instead, you may collect the registers and a few hundred
14167 bytes from the stack pointer with something like @code{*(unsigned char *)$esp@@300}
14168 (adjust to use the name of the actual stack pointer register on your
14169 target architecture, and the amount of stack you wish to capture).
14170 Then the @code{backtrace} command will show a partial backtrace when
14171 using a trace frame. The number of stack frames that can be examined
14172 depends on the sizes of the frames in the collected stack. Note that
14173 if you ask for a block so large that it goes past the bottom of the
14174 stack, the target agent may report an error trying to read from an
14178 If you do not collect registers at a tracepoint, @value{GDBN} can
14179 infer that the value of @code{$pc} must be the same as the address of
14180 the tracepoint and use that when you are looking at a trace frame
14181 for that tracepoint. However, this cannot work if the tracepoint has
14182 multiple locations (for instance if it was set in a function that was
14183 inlined), or if it has a @code{while-stepping} loop. In those cases
14184 @value{GDBN} will warn you that it can't infer @code{$pc}, and default
14189 @node Analyze Collected Data
14190 @section Using the Collected Data
14192 After the tracepoint experiment ends, you use @value{GDBN} commands
14193 for examining the trace data. The basic idea is that each tracepoint
14194 collects a trace @dfn{snapshot} every time it is hit and another
14195 snapshot every time it single-steps. All these snapshots are
14196 consecutively numbered from zero and go into a buffer, and you can
14197 examine them later. The way you examine them is to @dfn{focus} on a
14198 specific trace snapshot. When the remote stub is focused on a trace
14199 snapshot, it will respond to all @value{GDBN} requests for memory and
14200 registers by reading from the buffer which belongs to that snapshot,
14201 rather than from @emph{real} memory or registers of the program being
14202 debugged. This means that @strong{all} @value{GDBN} commands
14203 (@code{print}, @code{info registers}, @code{backtrace}, etc.) will
14204 behave as if we were currently debugging the program state as it was
14205 when the tracepoint occurred. Any requests for data that are not in
14206 the buffer will fail.
14209 * tfind:: How to select a trace snapshot
14210 * tdump:: How to display all data for a snapshot
14211 * save tracepoints:: How to save tracepoints for a future run
14215 @subsection @code{tfind @var{n}}
14218 @cindex select trace snapshot
14219 @cindex find trace snapshot
14220 The basic command for selecting a trace snapshot from the buffer is
14221 @code{tfind @var{n}}, which finds trace snapshot number @var{n},
14222 counting from zero. If no argument @var{n} is given, the next
14223 snapshot is selected.
14225 Here are the various forms of using the @code{tfind} command.
14229 Find the first snapshot in the buffer. This is a synonym for
14230 @code{tfind 0} (since 0 is the number of the first snapshot).
14233 Stop debugging trace snapshots, resume @emph{live} debugging.
14236 Same as @samp{tfind none}.
14239 No argument means find the next trace snapshot or find the first
14240 one if no trace snapshot is selected.
14243 Find the previous trace snapshot before the current one. This permits
14244 retracing earlier steps.
14246 @item tfind tracepoint @var{num}
14247 Find the next snapshot associated with tracepoint @var{num}. Search
14248 proceeds forward from the last examined trace snapshot. If no
14249 argument @var{num} is given, it means find the next snapshot collected
14250 for the same tracepoint as the current snapshot.
14252 @item tfind pc @var{addr}
14253 Find the next snapshot associated with the value @var{addr} of the
14254 program counter. Search proceeds forward from the last examined trace
14255 snapshot. If no argument @var{addr} is given, it means find the next
14256 snapshot with the same value of PC as the current snapshot.
14258 @item tfind outside @var{addr1}, @var{addr2}
14259 Find the next snapshot whose PC is outside the given range of
14260 addresses (exclusive).
14262 @item tfind range @var{addr1}, @var{addr2}
14263 Find the next snapshot whose PC is between @var{addr1} and
14264 @var{addr2} (inclusive).
14266 @item tfind line @r{[}@var{file}:@r{]}@var{n}
14267 Find the next snapshot associated with the source line @var{n}. If
14268 the optional argument @var{file} is given, refer to line @var{n} in
14269 that source file. Search proceeds forward from the last examined
14270 trace snapshot. If no argument @var{n} is given, it means find the
14271 next line other than the one currently being examined; thus saying
14272 @code{tfind line} repeatedly can appear to have the same effect as
14273 stepping from line to line in a @emph{live} debugging session.
14276 The default arguments for the @code{tfind} commands are specifically
14277 designed to make it easy to scan through the trace buffer. For
14278 instance, @code{tfind} with no argument selects the next trace
14279 snapshot, and @code{tfind -} with no argument selects the previous
14280 trace snapshot. So, by giving one @code{tfind} command, and then
14281 simply hitting @key{RET} repeatedly you can examine all the trace
14282 snapshots in order. Or, by saying @code{tfind -} and then hitting
14283 @key{RET} repeatedly you can examine the snapshots in reverse order.
14284 The @code{tfind line} command with no argument selects the snapshot
14285 for the next source line executed. The @code{tfind pc} command with
14286 no argument selects the next snapshot with the same program counter
14287 (PC) as the current frame. The @code{tfind tracepoint} command with
14288 no argument selects the next trace snapshot collected by the same
14289 tracepoint as the current one.
14291 In addition to letting you scan through the trace buffer manually,
14292 these commands make it easy to construct @value{GDBN} scripts that
14293 scan through the trace buffer and print out whatever collected data
14294 you are interested in. Thus, if we want to examine the PC, FP, and SP
14295 registers from each trace frame in the buffer, we can say this:
14298 (@value{GDBP}) @b{tfind start}
14299 (@value{GDBP}) @b{while ($trace_frame != -1)}
14300 > printf "Frame %d, PC = %08X, SP = %08X, FP = %08X\n", \
14301 $trace_frame, $pc, $sp, $fp
14305 Frame 0, PC = 0020DC64, SP = 0030BF3C, FP = 0030BF44
14306 Frame 1, PC = 0020DC6C, SP = 0030BF38, FP = 0030BF44
14307 Frame 2, PC = 0020DC70, SP = 0030BF34, FP = 0030BF44
14308 Frame 3, PC = 0020DC74, SP = 0030BF30, FP = 0030BF44
14309 Frame 4, PC = 0020DC78, SP = 0030BF2C, FP = 0030BF44
14310 Frame 5, PC = 0020DC7C, SP = 0030BF28, FP = 0030BF44
14311 Frame 6, PC = 0020DC80, SP = 0030BF24, FP = 0030BF44
14312 Frame 7, PC = 0020DC84, SP = 0030BF20, FP = 0030BF44
14313 Frame 8, PC = 0020DC88, SP = 0030BF1C, FP = 0030BF44
14314 Frame 9, PC = 0020DC8E, SP = 0030BF18, FP = 0030BF44
14315 Frame 10, PC = 00203F6C, SP = 0030BE3C, FP = 0030BF14
14318 Or, if we want to examine the variable @code{X} at each source line in
14322 (@value{GDBP}) @b{tfind start}
14323 (@value{GDBP}) @b{while ($trace_frame != -1)}
14324 > printf "Frame %d, X == %d\n", $trace_frame, X
14334 @subsection @code{tdump}
14336 @cindex dump all data collected at tracepoint
14337 @cindex tracepoint data, display
14339 This command takes no arguments. It prints all the data collected at
14340 the current trace snapshot.
14343 (@value{GDBP}) @b{trace 444}
14344 (@value{GDBP}) @b{actions}
14345 Enter actions for tracepoint #2, one per line:
14346 > collect $regs, $locals, $args, gdb_long_test
14349 (@value{GDBP}) @b{tstart}
14351 (@value{GDBP}) @b{tfind line 444}
14352 #0 gdb_test (p1=0x11, p2=0x22, p3=0x33, p4=0x44, p5=0x55, p6=0x66)
14354 444 printp( "%s: arguments = 0x%X 0x%X 0x%X 0x%X 0x%X 0x%X\n", )
14356 (@value{GDBP}) @b{tdump}
14357 Data collected at tracepoint 2, trace frame 1:
14358 d0 0xc4aa0085 -995491707
14362 d4 0x71aea3d 119204413
14365 d7 0x380035 3670069
14366 a0 0x19e24a 1696330
14367 a1 0x3000668 50333288
14369 a3 0x322000 3284992
14370 a4 0x3000698 50333336
14371 a5 0x1ad3cc 1758156
14372 fp 0x30bf3c 0x30bf3c
14373 sp 0x30bf34 0x30bf34
14375 pc 0x20b2c8 0x20b2c8
14379 p = 0x20e5b4 "gdb-test"
14386 gdb_long_test = 17 '\021'
14391 @code{tdump} works by scanning the tracepoint's current collection
14392 actions and printing the value of each expression listed. So
14393 @code{tdump} can fail, if after a run, you change the tracepoint's
14394 actions to mention variables that were not collected during the run.
14396 Also, for tracepoints with @code{while-stepping} loops, @code{tdump}
14397 uses the collected value of @code{$pc} to distinguish between trace
14398 frames that were collected at the tracepoint hit, and frames that were
14399 collected while stepping. This allows it to correctly choose whether
14400 to display the basic list of collections, or the collections from the
14401 body of the while-stepping loop. However, if @code{$pc} was not collected,
14402 then @code{tdump} will always attempt to dump using the basic collection
14403 list, and may fail if a while-stepping frame does not include all the
14404 same data that is collected at the tracepoint hit.
14405 @c This is getting pretty arcane, example would be good.
14407 @node save tracepoints
14408 @subsection @code{save tracepoints @var{filename}}
14409 @kindex save tracepoints
14410 @kindex save-tracepoints
14411 @cindex save tracepoints for future sessions
14413 This command saves all current tracepoint definitions together with
14414 their actions and passcounts, into a file @file{@var{filename}}
14415 suitable for use in a later debugging session. To read the saved
14416 tracepoint definitions, use the @code{source} command (@pxref{Command
14417 Files}). The @w{@code{save-tracepoints}} command is a deprecated
14418 alias for @w{@code{save tracepoints}}
14420 @node Tracepoint Variables
14421 @section Convenience Variables for Tracepoints
14422 @cindex tracepoint variables
14423 @cindex convenience variables for tracepoints
14426 @vindex $trace_frame
14427 @item (int) $trace_frame
14428 The current trace snapshot (a.k.a.@: @dfn{frame}) number, or -1 if no
14429 snapshot is selected.
14431 @vindex $tracepoint
14432 @item (int) $tracepoint
14433 The tracepoint for the current trace snapshot.
14435 @vindex $trace_line
14436 @item (int) $trace_line
14437 The line number for the current trace snapshot.
14439 @vindex $trace_file
14440 @item (char []) $trace_file
14441 The source file for the current trace snapshot.
14443 @vindex $trace_func
14444 @item (char []) $trace_func
14445 The name of the function containing @code{$tracepoint}.
14448 Note: @code{$trace_file} is not suitable for use in @code{printf},
14449 use @code{output} instead.
14451 Here's a simple example of using these convenience variables for
14452 stepping through all the trace snapshots and printing some of their
14453 data. Note that these are not the same as trace state variables,
14454 which are managed by the target.
14457 (@value{GDBP}) @b{tfind start}
14459 (@value{GDBP}) @b{while $trace_frame != -1}
14460 > output $trace_file
14461 > printf ", line %d (tracepoint #%d)\n", $trace_line, $tracepoint
14467 @section Using Trace Files
14468 @cindex trace files
14470 In some situations, the target running a trace experiment may no
14471 longer be available; perhaps it crashed, or the hardware was needed
14472 for a different activity. To handle these cases, you can arrange to
14473 dump the trace data into a file, and later use that file as a source
14474 of trace data, via the @code{target tfile} command.
14479 @item tsave [ -r ] @var{filename}
14480 @itemx tsave [-ctf] @var{dirname}
14481 Save the trace data to @var{filename}. By default, this command
14482 assumes that @var{filename} refers to the host filesystem, so if
14483 necessary @value{GDBN} will copy raw trace data up from the target and
14484 then save it. If the target supports it, you can also supply the
14485 optional argument @code{-r} (``remote'') to direct the target to save
14486 the data directly into @var{filename} in its own filesystem, which may be
14487 more efficient if the trace buffer is very large. (Note, however, that
14488 @code{target tfile} can only read from files accessible to the host.)
14489 By default, this command will save trace frame in tfile format.
14490 You can supply the optional argument @code{-ctf} to save data in CTF
14491 format. The @dfn{Common Trace Format} (CTF) is proposed as a trace format
14492 that can be shared by multiple debugging and tracing tools. Please go to
14493 @indicateurl{http://www.efficios.com/ctf} to get more information.
14495 @kindex target tfile
14499 @item target tfile @var{filename}
14500 @itemx target ctf @var{dirname}
14501 Use the file named @var{filename} or directory named @var{dirname} as
14502 a source of trace data. Commands that examine data work as they do with
14503 a live target, but it is not possible to run any new trace experiments.
14504 @code{tstatus} will report the state of the trace run at the moment
14505 the data was saved, as well as the current trace frame you are examining.
14506 Both @var{filename} and @var{dirname} must be on a filesystem accessible to
14510 (@value{GDBP}) target ctf ctf.ctf
14511 (@value{GDBP}) tfind
14512 Found trace frame 0, tracepoint 2
14513 39 ++a; /* set tracepoint 1 here */
14514 (@value{GDBP}) tdump
14515 Data collected at tracepoint 2, trace frame 0:
14519 c = @{"123", "456", "789", "123", "456", "789"@}
14520 d = @{@{@{a = 1, b = 2@}, @{a = 3, b = 4@}@}, @{@{a = 5, b = 6@}, @{a = 7, b = 8@}@}@}
14528 @chapter Debugging Programs That Use Overlays
14531 If your program is too large to fit completely in your target system's
14532 memory, you can sometimes use @dfn{overlays} to work around this
14533 problem. @value{GDBN} provides some support for debugging programs that
14537 * How Overlays Work:: A general explanation of overlays.
14538 * Overlay Commands:: Managing overlays in @value{GDBN}.
14539 * Automatic Overlay Debugging:: @value{GDBN} can find out which overlays are
14540 mapped by asking the inferior.
14541 * Overlay Sample Program:: A sample program using overlays.
14544 @node How Overlays Work
14545 @section How Overlays Work
14546 @cindex mapped overlays
14547 @cindex unmapped overlays
14548 @cindex load address, overlay's
14549 @cindex mapped address
14550 @cindex overlay area
14552 Suppose you have a computer whose instruction address space is only 64
14553 kilobytes long, but which has much more memory which can be accessed by
14554 other means: special instructions, segment registers, or memory
14555 management hardware, for example. Suppose further that you want to
14556 adapt a program which is larger than 64 kilobytes to run on this system.
14558 One solution is to identify modules of your program which are relatively
14559 independent, and need not call each other directly; call these modules
14560 @dfn{overlays}. Separate the overlays from the main program, and place
14561 their machine code in the larger memory. Place your main program in
14562 instruction memory, but leave at least enough space there to hold the
14563 largest overlay as well.
14565 Now, to call a function located in an overlay, you must first copy that
14566 overlay's machine code from the large memory into the space set aside
14567 for it in the instruction memory, and then jump to its entry point
14570 @c NB: In the below the mapped area's size is greater or equal to the
14571 @c size of all overlays. This is intentional to remind the developer
14572 @c that overlays don't necessarily need to be the same size.
14576 Data Instruction Larger
14577 Address Space Address Space Address Space
14578 +-----------+ +-----------+ +-----------+
14580 +-----------+ +-----------+ +-----------+<-- overlay 1
14581 | program | | main | .----| overlay 1 | load address
14582 | variables | | program | | +-----------+
14583 | and heap | | | | | |
14584 +-----------+ | | | +-----------+<-- overlay 2
14585 | | +-----------+ | | | load address
14586 +-----------+ | | | .-| overlay 2 |
14588 mapped --->+-----------+ | | +-----------+
14589 address | | | | | |
14590 | overlay | <-' | | |
14591 | area | <---' +-----------+<-- overlay 3
14592 | | <---. | | load address
14593 +-----------+ `--| overlay 3 |
14600 @anchor{A code overlay}A code overlay
14604 The diagram (@pxref{A code overlay}) shows a system with separate data
14605 and instruction address spaces. To map an overlay, the program copies
14606 its code from the larger address space to the instruction address space.
14607 Since the overlays shown here all use the same mapped address, only one
14608 may be mapped at a time. For a system with a single address space for
14609 data and instructions, the diagram would be similar, except that the
14610 program variables and heap would share an address space with the main
14611 program and the overlay area.
14613 An overlay loaded into instruction memory and ready for use is called a
14614 @dfn{mapped} overlay; its @dfn{mapped address} is its address in the
14615 instruction memory. An overlay not present (or only partially present)
14616 in instruction memory is called @dfn{unmapped}; its @dfn{load address}
14617 is its address in the larger memory. The mapped address is also called
14618 the @dfn{virtual memory address}, or @dfn{VMA}; the load address is also
14619 called the @dfn{load memory address}, or @dfn{LMA}.
14621 Unfortunately, overlays are not a completely transparent way to adapt a
14622 program to limited instruction memory. They introduce a new set of
14623 global constraints you must keep in mind as you design your program:
14628 Before calling or returning to a function in an overlay, your program
14629 must make sure that overlay is actually mapped. Otherwise, the call or
14630 return will transfer control to the right address, but in the wrong
14631 overlay, and your program will probably crash.
14634 If the process of mapping an overlay is expensive on your system, you
14635 will need to choose your overlays carefully to minimize their effect on
14636 your program's performance.
14639 The executable file you load onto your system must contain each
14640 overlay's instructions, appearing at the overlay's load address, not its
14641 mapped address. However, each overlay's instructions must be relocated
14642 and its symbols defined as if the overlay were at its mapped address.
14643 You can use GNU linker scripts to specify different load and relocation
14644 addresses for pieces of your program; see @ref{Overlay Description,,,
14645 ld.info, Using ld: the GNU linker}.
14648 The procedure for loading executable files onto your system must be able
14649 to load their contents into the larger address space as well as the
14650 instruction and data spaces.
14654 The overlay system described above is rather simple, and could be
14655 improved in many ways:
14660 If your system has suitable bank switch registers or memory management
14661 hardware, you could use those facilities to make an overlay's load area
14662 contents simply appear at their mapped address in instruction space.
14663 This would probably be faster than copying the overlay to its mapped
14664 area in the usual way.
14667 If your overlays are small enough, you could set aside more than one
14668 overlay area, and have more than one overlay mapped at a time.
14671 You can use overlays to manage data, as well as instructions. In
14672 general, data overlays are even less transparent to your design than
14673 code overlays: whereas code overlays only require care when you call or
14674 return to functions, data overlays require care every time you access
14675 the data. Also, if you change the contents of a data overlay, you
14676 must copy its contents back out to its load address before you can copy a
14677 different data overlay into the same mapped area.
14682 @node Overlay Commands
14683 @section Overlay Commands
14685 To use @value{GDBN}'s overlay support, each overlay in your program must
14686 correspond to a separate section of the executable file. The section's
14687 virtual memory address and load memory address must be the overlay's
14688 mapped and load addresses. Identifying overlays with sections allows
14689 @value{GDBN} to determine the appropriate address of a function or
14690 variable, depending on whether the overlay is mapped or not.
14692 @value{GDBN}'s overlay commands all start with the word @code{overlay};
14693 you can abbreviate this as @code{ov} or @code{ovly}. The commands are:
14698 Disable @value{GDBN}'s overlay support. When overlay support is
14699 disabled, @value{GDBN} assumes that all functions and variables are
14700 always present at their mapped addresses. By default, @value{GDBN}'s
14701 overlay support is disabled.
14703 @item overlay manual
14704 @cindex manual overlay debugging
14705 Enable @dfn{manual} overlay debugging. In this mode, @value{GDBN}
14706 relies on you to tell it which overlays are mapped, and which are not,
14707 using the @code{overlay map-overlay} and @code{overlay unmap-overlay}
14708 commands described below.
14710 @item overlay map-overlay @var{overlay}
14711 @itemx overlay map @var{overlay}
14712 @cindex map an overlay
14713 Tell @value{GDBN} that @var{overlay} is now mapped; @var{overlay} must
14714 be the name of the object file section containing the overlay. When an
14715 overlay is mapped, @value{GDBN} assumes it can find the overlay's
14716 functions and variables at their mapped addresses. @value{GDBN} assumes
14717 that any other overlays whose mapped ranges overlap that of
14718 @var{overlay} are now unmapped.
14720 @item overlay unmap-overlay @var{overlay}
14721 @itemx overlay unmap @var{overlay}
14722 @cindex unmap an overlay
14723 Tell @value{GDBN} that @var{overlay} is no longer mapped; @var{overlay}
14724 must be the name of the object file section containing the overlay.
14725 When an overlay is unmapped, @value{GDBN} assumes it can find the
14726 overlay's functions and variables at their load addresses.
14729 Enable @dfn{automatic} overlay debugging. In this mode, @value{GDBN}
14730 consults a data structure the overlay manager maintains in the inferior
14731 to see which overlays are mapped. For details, see @ref{Automatic
14732 Overlay Debugging}.
14734 @item overlay load-target
14735 @itemx overlay load
14736 @cindex reloading the overlay table
14737 Re-read the overlay table from the inferior. Normally, @value{GDBN}
14738 re-reads the table @value{GDBN} automatically each time the inferior
14739 stops, so this command should only be necessary if you have changed the
14740 overlay mapping yourself using @value{GDBN}. This command is only
14741 useful when using automatic overlay debugging.
14743 @item overlay list-overlays
14744 @itemx overlay list
14745 @cindex listing mapped overlays
14746 Display a list of the overlays currently mapped, along with their mapped
14747 addresses, load addresses, and sizes.
14751 Normally, when @value{GDBN} prints a code address, it includes the name
14752 of the function the address falls in:
14755 (@value{GDBP}) print main
14756 $3 = @{int ()@} 0x11a0 <main>
14759 When overlay debugging is enabled, @value{GDBN} recognizes code in
14760 unmapped overlays, and prints the names of unmapped functions with
14761 asterisks around them. For example, if @code{foo} is a function in an
14762 unmapped overlay, @value{GDBN} prints it this way:
14765 (@value{GDBP}) overlay list
14766 No sections are mapped.
14767 (@value{GDBP}) print foo
14768 $5 = @{int (int)@} 0x100000 <*foo*>
14771 When @code{foo}'s overlay is mapped, @value{GDBN} prints the function's
14775 (@value{GDBP}) overlay list
14776 Section .ov.foo.text, loaded at 0x100000 - 0x100034,
14777 mapped at 0x1016 - 0x104a
14778 (@value{GDBP}) print foo
14779 $6 = @{int (int)@} 0x1016 <foo>
14782 When overlay debugging is enabled, @value{GDBN} can find the correct
14783 address for functions and variables in an overlay, whether or not the
14784 overlay is mapped. This allows most @value{GDBN} commands, like
14785 @code{break} and @code{disassemble}, to work normally, even on unmapped
14786 code. However, @value{GDBN}'s breakpoint support has some limitations:
14790 @cindex breakpoints in overlays
14791 @cindex overlays, setting breakpoints in
14792 You can set breakpoints in functions in unmapped overlays, as long as
14793 @value{GDBN} can write to the overlay at its load address.
14795 @value{GDBN} can not set hardware or simulator-based breakpoints in
14796 unmapped overlays. However, if you set a breakpoint at the end of your
14797 overlay manager (and tell @value{GDBN} which overlays are now mapped, if
14798 you are using manual overlay management), @value{GDBN} will re-set its
14799 breakpoints properly.
14803 @node Automatic Overlay Debugging
14804 @section Automatic Overlay Debugging
14805 @cindex automatic overlay debugging
14807 @value{GDBN} can automatically track which overlays are mapped and which
14808 are not, given some simple co-operation from the overlay manager in the
14809 inferior. If you enable automatic overlay debugging with the
14810 @code{overlay auto} command (@pxref{Overlay Commands}), @value{GDBN}
14811 looks in the inferior's memory for certain variables describing the
14812 current state of the overlays.
14814 Here are the variables your overlay manager must define to support
14815 @value{GDBN}'s automatic overlay debugging:
14819 @item @code{_ovly_table}:
14820 This variable must be an array of the following structures:
14825 /* The overlay's mapped address. */
14828 /* The size of the overlay, in bytes. */
14829 unsigned long size;
14831 /* The overlay's load address. */
14834 /* Non-zero if the overlay is currently mapped;
14836 unsigned long mapped;
14840 @item @code{_novlys}:
14841 This variable must be a four-byte signed integer, holding the total
14842 number of elements in @code{_ovly_table}.
14846 To decide whether a particular overlay is mapped or not, @value{GDBN}
14847 looks for an entry in @w{@code{_ovly_table}} whose @code{vma} and
14848 @code{lma} members equal the VMA and LMA of the overlay's section in the
14849 executable file. When @value{GDBN} finds a matching entry, it consults
14850 the entry's @code{mapped} member to determine whether the overlay is
14853 In addition, your overlay manager may define a function called
14854 @code{_ovly_debug_event}. If this function is defined, @value{GDBN}
14855 will silently set a breakpoint there. If the overlay manager then
14856 calls this function whenever it has changed the overlay table, this
14857 will enable @value{GDBN} to accurately keep track of which overlays
14858 are in program memory, and update any breakpoints that may be set
14859 in overlays. This will allow breakpoints to work even if the
14860 overlays are kept in ROM or other non-writable memory while they
14861 are not being executed.
14863 @node Overlay Sample Program
14864 @section Overlay Sample Program
14865 @cindex overlay example program
14867 When linking a program which uses overlays, you must place the overlays
14868 at their load addresses, while relocating them to run at their mapped
14869 addresses. To do this, you must write a linker script (@pxref{Overlay
14870 Description,,, ld.info, Using ld: the GNU linker}). Unfortunately,
14871 since linker scripts are specific to a particular host system, target
14872 architecture, and target memory layout, this manual cannot provide
14873 portable sample code demonstrating @value{GDBN}'s overlay support.
14875 However, the @value{GDBN} source distribution does contain an overlaid
14876 program, with linker scripts for a few systems, as part of its test
14877 suite. The program consists of the following files from
14878 @file{gdb/testsuite/gdb.base}:
14882 The main program file.
14884 A simple overlay manager, used by @file{overlays.c}.
14889 Overlay modules, loaded and used by @file{overlays.c}.
14892 Linker scripts for linking the test program on the @code{d10v-elf}
14893 and @code{m32r-elf} targets.
14896 You can build the test program using the @code{d10v-elf} GCC
14897 cross-compiler like this:
14900 $ d10v-elf-gcc -g -c overlays.c
14901 $ d10v-elf-gcc -g -c ovlymgr.c
14902 $ d10v-elf-gcc -g -c foo.c
14903 $ d10v-elf-gcc -g -c bar.c
14904 $ d10v-elf-gcc -g -c baz.c
14905 $ d10v-elf-gcc -g -c grbx.c
14906 $ d10v-elf-gcc -g overlays.o ovlymgr.o foo.o bar.o \
14907 baz.o grbx.o -Wl,-Td10v.ld -o overlays
14910 The build process is identical for any other architecture, except that
14911 you must substitute the appropriate compiler and linker script for the
14912 target system for @code{d10v-elf-gcc} and @code{d10v.ld}.
14916 @chapter Using @value{GDBN} with Different Languages
14919 Although programming languages generally have common aspects, they are
14920 rarely expressed in the same manner. For instance, in ANSI C,
14921 dereferencing a pointer @code{p} is accomplished by @code{*p}, but in
14922 Modula-2, it is accomplished by @code{p^}. Values can also be
14923 represented (and displayed) differently. Hex numbers in C appear as
14924 @samp{0x1ae}, while in Modula-2 they appear as @samp{1AEH}.
14926 @cindex working language
14927 Language-specific information is built into @value{GDBN} for some languages,
14928 allowing you to express operations like the above in your program's
14929 native language, and allowing @value{GDBN} to output values in a manner
14930 consistent with the syntax of your program's native language. The
14931 language you use to build expressions is called the @dfn{working
14935 * Setting:: Switching between source languages
14936 * Show:: Displaying the language
14937 * Checks:: Type and range checks
14938 * Supported Languages:: Supported languages
14939 * Unsupported Languages:: Unsupported languages
14943 @section Switching Between Source Languages
14945 There are two ways to control the working language---either have @value{GDBN}
14946 set it automatically, or select it manually yourself. You can use the
14947 @code{set language} command for either purpose. On startup, @value{GDBN}
14948 defaults to setting the language automatically. The working language is
14949 used to determine how expressions you type are interpreted, how values
14952 In addition to the working language, every source file that
14953 @value{GDBN} knows about has its own working language. For some object
14954 file formats, the compiler might indicate which language a particular
14955 source file is in. However, most of the time @value{GDBN} infers the
14956 language from the name of the file. The language of a source file
14957 controls whether C@t{++} names are demangled---this way @code{backtrace} can
14958 show each frame appropriately for its own language. There is no way to
14959 set the language of a source file from within @value{GDBN}, but you can
14960 set the language associated with a filename extension. @xref{Show, ,
14961 Displaying the Language}.
14963 This is most commonly a problem when you use a program, such
14964 as @code{cfront} or @code{f2c}, that generates C but is written in
14965 another language. In that case, make the
14966 program use @code{#line} directives in its C output; that way
14967 @value{GDBN} will know the correct language of the source code of the original
14968 program, and will display that source code, not the generated C code.
14971 * Filenames:: Filename extensions and languages.
14972 * Manually:: Setting the working language manually
14973 * Automatically:: Having @value{GDBN} infer the source language
14977 @subsection List of Filename Extensions and Languages
14979 If a source file name ends in one of the following extensions, then
14980 @value{GDBN} infers that its language is the one indicated.
14998 C@t{++} source file
15004 Objective-C source file
15008 Fortran source file
15011 Modula-2 source file
15015 Assembler source file. This actually behaves almost like C, but
15016 @value{GDBN} does not skip over function prologues when stepping.
15019 In addition, you may set the language associated with a filename
15020 extension. @xref{Show, , Displaying the Language}.
15023 @subsection Setting the Working Language
15025 If you allow @value{GDBN} to set the language automatically,
15026 expressions are interpreted the same way in your debugging session and
15029 @kindex set language
15030 If you wish, you may set the language manually. To do this, issue the
15031 command @samp{set language @var{lang}}, where @var{lang} is the name of
15032 a language, such as
15033 @code{c} or @code{modula-2}.
15034 For a list of the supported languages, type @samp{set language}.
15036 Setting the language manually prevents @value{GDBN} from updating the working
15037 language automatically. This can lead to confusion if you try
15038 to debug a program when the working language is not the same as the
15039 source language, when an expression is acceptable to both
15040 languages---but means different things. For instance, if the current
15041 source file were written in C, and @value{GDBN} was parsing Modula-2, a
15049 might not have the effect you intended. In C, this means to add
15050 @code{b} and @code{c} and place the result in @code{a}. The result
15051 printed would be the value of @code{a}. In Modula-2, this means to compare
15052 @code{a} to the result of @code{b+c}, yielding a @code{BOOLEAN} value.
15054 @node Automatically
15055 @subsection Having @value{GDBN} Infer the Source Language
15057 To have @value{GDBN} set the working language automatically, use
15058 @samp{set language local} or @samp{set language auto}. @value{GDBN}
15059 then infers the working language. That is, when your program stops in a
15060 frame (usually by encountering a breakpoint), @value{GDBN} sets the
15061 working language to the language recorded for the function in that
15062 frame. If the language for a frame is unknown (that is, if the function
15063 or block corresponding to the frame was defined in a source file that
15064 does not have a recognized extension), the current working language is
15065 not changed, and @value{GDBN} issues a warning.
15067 This may not seem necessary for most programs, which are written
15068 entirely in one source language. However, program modules and libraries
15069 written in one source language can be used by a main program written in
15070 a different source language. Using @samp{set language auto} in this
15071 case frees you from having to set the working language manually.
15074 @section Displaying the Language
15076 The following commands help you find out which language is the
15077 working language, and also what language source files were written in.
15080 @item show language
15081 @anchor{show language}
15082 @kindex show language
15083 Display the current working language. This is the
15084 language you can use with commands such as @code{print} to
15085 build and compute expressions that may involve variables in your program.
15088 @kindex info frame@r{, show the source language}
15089 Display the source language for this frame. This language becomes the
15090 working language if you use an identifier from this frame.
15091 @xref{Frame Info, ,Information about a Frame}, to identify the other
15092 information listed here.
15095 @kindex info source@r{, show the source language}
15096 Display the source language of this source file.
15097 @xref{Symbols, ,Examining the Symbol Table}, to identify the other
15098 information listed here.
15101 In unusual circumstances, you may have source files with extensions
15102 not in the standard list. You can then set the extension associated
15103 with a language explicitly:
15106 @item set extension-language @var{ext} @var{language}
15107 @kindex set extension-language
15108 Tell @value{GDBN} that source files with extension @var{ext} are to be
15109 assumed as written in the source language @var{language}.
15111 @item info extensions
15112 @kindex info extensions
15113 List all the filename extensions and the associated languages.
15117 @section Type and Range Checking
15119 Some languages are designed to guard you against making seemingly common
15120 errors through a series of compile- and run-time checks. These include
15121 checking the type of arguments to functions and operators and making
15122 sure mathematical overflows are caught at run time. Checks such as
15123 these help to ensure a program's correctness once it has been compiled
15124 by eliminating type mismatches and providing active checks for range
15125 errors when your program is running.
15127 By default @value{GDBN} checks for these errors according to the
15128 rules of the current source language. Although @value{GDBN} does not check
15129 the statements in your program, it can check expressions entered directly
15130 into @value{GDBN} for evaluation via the @code{print} command, for example.
15133 * Type Checking:: An overview of type checking
15134 * Range Checking:: An overview of range checking
15137 @cindex type checking
15138 @cindex checks, type
15139 @node Type Checking
15140 @subsection An Overview of Type Checking
15142 Some languages, such as C and C@t{++}, are strongly typed, meaning that the
15143 arguments to operators and functions have to be of the correct type,
15144 otherwise an error occurs. These checks prevent type mismatch
15145 errors from ever causing any run-time problems. For example,
15148 int klass::my_method(char *b) @{ return b ? 1 : 2; @}
15150 (@value{GDBP}) print obj.my_method (0)
15153 (@value{GDBP}) print obj.my_method (0x1234)
15154 Cannot resolve method klass::my_method to any overloaded instance
15157 The second example fails because in C@t{++} the integer constant
15158 @samp{0x1234} is not type-compatible with the pointer parameter type.
15160 For the expressions you use in @value{GDBN} commands, you can tell
15161 @value{GDBN} to not enforce strict type checking or
15162 to treat any mismatches as errors and abandon the expression;
15163 When type checking is disabled, @value{GDBN} successfully evaluates
15164 expressions like the second example above.
15166 Even if type checking is off, there may be other reasons
15167 related to type that prevent @value{GDBN} from evaluating an expression.
15168 For instance, @value{GDBN} does not know how to add an @code{int} and
15169 a @code{struct foo}. These particular type errors have nothing to do
15170 with the language in use and usually arise from expressions which make
15171 little sense to evaluate anyway.
15173 @value{GDBN} provides some additional commands for controlling type checking:
15175 @kindex set check type
15176 @kindex show check type
15178 @item set check type on
15179 @itemx set check type off
15180 Set strict type checking on or off. If any type mismatches occur in
15181 evaluating an expression while type checking is on, @value{GDBN} prints a
15182 message and aborts evaluation of the expression.
15184 @item show check type
15185 Show the current setting of type checking and whether @value{GDBN}
15186 is enforcing strict type checking rules.
15189 @cindex range checking
15190 @cindex checks, range
15191 @node Range Checking
15192 @subsection An Overview of Range Checking
15194 In some languages (such as Modula-2), it is an error to exceed the
15195 bounds of a type; this is enforced with run-time checks. Such range
15196 checking is meant to ensure program correctness by making sure
15197 computations do not overflow, or indices on an array element access do
15198 not exceed the bounds of the array.
15200 For expressions you use in @value{GDBN} commands, you can tell
15201 @value{GDBN} to treat range errors in one of three ways: ignore them,
15202 always treat them as errors and abandon the expression, or issue
15203 warnings but evaluate the expression anyway.
15205 A range error can result from numerical overflow, from exceeding an
15206 array index bound, or when you type a constant that is not a member
15207 of any type. Some languages, however, do not treat overflows as an
15208 error. In many implementations of C, mathematical overflow causes the
15209 result to ``wrap around'' to lower values---for example, if @var{m} is
15210 the largest integer value, and @var{s} is the smallest, then
15213 @var{m} + 1 @result{} @var{s}
15216 This, too, is specific to individual languages, and in some cases
15217 specific to individual compilers or machines. @xref{Supported Languages, ,
15218 Supported Languages}, for further details on specific languages.
15220 @value{GDBN} provides some additional commands for controlling the range checker:
15222 @kindex set check range
15223 @kindex show check range
15225 @item set check range auto
15226 Set range checking on or off based on the current working language.
15227 @xref{Supported Languages, ,Supported Languages}, for the default settings for
15230 @item set check range on
15231 @itemx set check range off
15232 Set range checking on or off, overriding the default setting for the
15233 current working language. A warning is issued if the setting does not
15234 match the language default. If a range error occurs and range checking is on,
15235 then a message is printed and evaluation of the expression is aborted.
15237 @item set check range warn
15238 Output messages when the @value{GDBN} range checker detects a range error,
15239 but attempt to evaluate the expression anyway. Evaluating the
15240 expression may still be impossible for other reasons, such as accessing
15241 memory that the process does not own (a typical example from many Unix
15245 Show the current setting of the range checker, and whether or not it is
15246 being set automatically by @value{GDBN}.
15249 @node Supported Languages
15250 @section Supported Languages
15252 @value{GDBN} supports C, C@t{++}, D, Go, Objective-C, Fortran,
15253 OpenCL C, Pascal, Rust, assembly, Modula-2, and Ada.
15254 @c This is false ...
15255 Some @value{GDBN} features may be used in expressions regardless of the
15256 language you use: the @value{GDBN} @code{@@} and @code{::} operators,
15257 and the @samp{@{type@}addr} construct (@pxref{Expressions,
15258 ,Expressions}) can be used with the constructs of any supported
15261 The following sections detail to what degree each source language is
15262 supported by @value{GDBN}. These sections are not meant to be language
15263 tutorials or references, but serve only as a reference guide to what the
15264 @value{GDBN} expression parser accepts, and what input and output
15265 formats should look like for different languages. There are many good
15266 books written on each of these languages; please look to these for a
15267 language reference or tutorial.
15270 * C:: C and C@t{++}
15273 * Objective-C:: Objective-C
15274 * OpenCL C:: OpenCL C
15275 * Fortran:: Fortran
15278 * Modula-2:: Modula-2
15283 @subsection C and C@t{++}
15285 @cindex C and C@t{++}
15286 @cindex expressions in C or C@t{++}
15288 Since C and C@t{++} are so closely related, many features of @value{GDBN} apply
15289 to both languages. Whenever this is the case, we discuss those languages
15293 @cindex @code{g++}, @sc{gnu} C@t{++} compiler
15294 @cindex @sc{gnu} C@t{++}
15295 The C@t{++} debugging facilities are jointly implemented by the C@t{++}
15296 compiler and @value{GDBN}. Therefore, to debug your C@t{++} code
15297 effectively, you must compile your C@t{++} programs with a supported
15298 C@t{++} compiler, such as @sc{gnu} @code{g++}, or the HP ANSI C@t{++}
15299 compiler (@code{aCC}).
15302 * C Operators:: C and C@t{++} operators
15303 * C Constants:: C and C@t{++} constants
15304 * C Plus Plus Expressions:: C@t{++} expressions
15305 * C Defaults:: Default settings for C and C@t{++}
15306 * C Checks:: C and C@t{++} type and range checks
15307 * Debugging C:: @value{GDBN} and C
15308 * Debugging C Plus Plus:: @value{GDBN} features for C@t{++}
15309 * Decimal Floating Point:: Numbers in Decimal Floating Point format
15313 @subsubsection C and C@t{++} Operators
15315 @cindex C and C@t{++} operators
15317 Operators must be defined on values of specific types. For instance,
15318 @code{+} is defined on numbers, but not on structures. Operators are
15319 often defined on groups of types.
15321 For the purposes of C and C@t{++}, the following definitions hold:
15326 @emph{Integral types} include @code{int} with any of its storage-class
15327 specifiers; @code{char}; @code{enum}; and, for C@t{++}, @code{bool}.
15330 @emph{Floating-point types} include @code{float}, @code{double}, and
15331 @code{long double} (if supported by the target platform).
15334 @emph{Pointer types} include all types defined as @code{(@var{type} *)}.
15337 @emph{Scalar types} include all of the above.
15342 The following operators are supported. They are listed here
15343 in order of increasing precedence:
15347 The comma or sequencing operator. Expressions in a comma-separated list
15348 are evaluated from left to right, with the result of the entire
15349 expression being the last expression evaluated.
15352 Assignment. The value of an assignment expression is the value
15353 assigned. Defined on scalar types.
15356 Used in an expression of the form @w{@code{@var{a} @var{op}= @var{b}}},
15357 and translated to @w{@code{@var{a} = @var{a op b}}}.
15358 @w{@code{@var{op}=}} and @code{=} have the same precedence. The operator
15359 @var{op} is any one of the operators @code{|}, @code{^}, @code{&},
15360 @code{<<}, @code{>>}, @code{+}, @code{-}, @code{*}, @code{/}, @code{%}.
15363 The ternary operator. @code{@var{a} ? @var{b} : @var{c}} can be thought
15364 of as: if @var{a} then @var{b} else @var{c}. The argument @var{a}
15365 should be of an integral type.
15368 Logical @sc{or}. Defined on integral types.
15371 Logical @sc{and}. Defined on integral types.
15374 Bitwise @sc{or}. Defined on integral types.
15377 Bitwise exclusive-@sc{or}. Defined on integral types.
15380 Bitwise @sc{and}. Defined on integral types.
15383 Equality and inequality. Defined on scalar types. The value of these
15384 expressions is 0 for false and non-zero for true.
15386 @item <@r{, }>@r{, }<=@r{, }>=
15387 Less than, greater than, less than or equal, greater than or equal.
15388 Defined on scalar types. The value of these expressions is 0 for false
15389 and non-zero for true.
15392 left shift, and right shift. Defined on integral types.
15395 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
15398 Addition and subtraction. Defined on integral types, floating-point types and
15401 @item *@r{, }/@r{, }%
15402 Multiplication, division, and modulus. Multiplication and division are
15403 defined on integral and floating-point types. Modulus is defined on
15407 Increment and decrement. When appearing before a variable, the
15408 operation is performed before the variable is used in an expression;
15409 when appearing after it, the variable's value is used before the
15410 operation takes place.
15413 Pointer dereferencing. Defined on pointer types. Same precedence as
15417 Address operator. Defined on variables. Same precedence as @code{++}.
15419 For debugging C@t{++}, @value{GDBN} implements a use of @samp{&} beyond what is
15420 allowed in the C@t{++} language itself: you can use @samp{&(&@var{ref})}
15421 to examine the address
15422 where a C@t{++} reference variable (declared with @samp{&@var{ref}}) is
15426 Negative. Defined on integral and floating-point types. Same
15427 precedence as @code{++}.
15430 Logical negation. Defined on integral types. Same precedence as
15434 Bitwise complement operator. Defined on integral types. Same precedence as
15439 Structure member, and pointer-to-structure member. For convenience,
15440 @value{GDBN} regards the two as equivalent, choosing whether to dereference a
15441 pointer based on the stored type information.
15442 Defined on @code{struct} and @code{union} data.
15445 Dereferences of pointers to members.
15448 Array indexing. @code{@var{a}[@var{i}]} is defined as
15449 @code{*(@var{a}+@var{i})}. Same precedence as @code{->}.
15452 Function parameter list. Same precedence as @code{->}.
15455 C@t{++} scope resolution operator. Defined on @code{struct}, @code{union},
15456 and @code{class} types.
15459 Doubled colons also represent the @value{GDBN} scope operator
15460 (@pxref{Expressions, ,Expressions}). Same precedence as @code{::},
15464 If an operator is redefined in the user code, @value{GDBN} usually
15465 attempts to invoke the redefined version instead of using the operator's
15466 predefined meaning.
15469 @subsubsection C and C@t{++} Constants
15471 @cindex C and C@t{++} constants
15473 @value{GDBN} allows you to express the constants of C and C@t{++} in the
15478 Integer constants are a sequence of digits. Octal constants are
15479 specified by a leading @samp{0} (i.e.@: zero), and hexadecimal constants
15480 by a leading @samp{0x} or @samp{0X}. Constants may also end with a letter
15481 @samp{l}, specifying that the constant should be treated as a
15485 Floating point constants are a sequence of digits, followed by a decimal
15486 point, followed by a sequence of digits, and optionally followed by an
15487 exponent. An exponent is of the form:
15488 @samp{@w{e@r{[[}+@r{]|}-@r{]}@var{nnn}}}, where @var{nnn} is another
15489 sequence of digits. The @samp{+} is optional for positive exponents.
15490 A floating-point constant may also end with a letter @samp{f} or
15491 @samp{F}, specifying that the constant should be treated as being of
15492 the @code{float} (as opposed to the default @code{double}) type; or with
15493 a letter @samp{l} or @samp{L}, which specifies a @code{long double}
15497 Enumerated constants consist of enumerated identifiers, or their
15498 integral equivalents.
15501 Character constants are a single character surrounded by single quotes
15502 (@code{'}), or a number---the ordinal value of the corresponding character
15503 (usually its @sc{ascii} value). Within quotes, the single character may
15504 be represented by a letter or by @dfn{escape sequences}, which are of
15505 the form @samp{\@var{nnn}}, where @var{nnn} is the octal representation
15506 of the character's ordinal value; or of the form @samp{\@var{x}}, where
15507 @samp{@var{x}} is a predefined special character---for example,
15508 @samp{\n} for newline.
15510 Wide character constants can be written by prefixing a character
15511 constant with @samp{L}, as in C. For example, @samp{L'x'} is the wide
15512 form of @samp{x}. The target wide character set is used when
15513 computing the value of this constant (@pxref{Character Sets}).
15516 String constants are a sequence of character constants surrounded by
15517 double quotes (@code{"}). Any valid character constant (as described
15518 above) may appear. Double quotes within the string must be preceded by
15519 a backslash, so for instance @samp{"a\"b'c"} is a string of five
15522 Wide string constants can be written by prefixing a string constant
15523 with @samp{L}, as in C. The target wide character set is used when
15524 computing the value of this constant (@pxref{Character Sets}).
15527 Pointer constants are an integral value. You can also write pointers
15528 to constants using the C operator @samp{&}.
15531 Array constants are comma-separated lists surrounded by braces @samp{@{}
15532 and @samp{@}}; for example, @samp{@{1,2,3@}} is a three-element array of
15533 integers, @samp{@{@{1,2@}, @{3,4@}, @{5,6@}@}} is a three-by-two array,
15534 and @samp{@{&"hi", &"there", &"fred"@}} is a three-element array of pointers.
15537 @node C Plus Plus Expressions
15538 @subsubsection C@t{++} Expressions
15540 @cindex expressions in C@t{++}
15541 @value{GDBN} expression handling can interpret most C@t{++} expressions.
15543 @cindex debugging C@t{++} programs
15544 @cindex C@t{++} compilers
15545 @cindex debug formats and C@t{++}
15546 @cindex @value{NGCC} and C@t{++}
15548 @emph{Warning:} @value{GDBN} can only debug C@t{++} code if you use
15549 the proper compiler and the proper debug format. Currently,
15550 @value{GDBN} works best when debugging C@t{++} code that is compiled
15551 with the most recent version of @value{NGCC} possible. The DWARF
15552 debugging format is preferred; @value{NGCC} defaults to this on most
15553 popular platforms. Other compilers and/or debug formats are likely to
15554 work badly or not at all when using @value{GDBN} to debug C@t{++}
15555 code. @xref{Compilation}.
15560 @cindex member functions
15562 Member function calls are allowed; you can use expressions like
15565 count = aml->GetOriginal(x, y)
15568 @vindex this@r{, inside C@t{++} member functions}
15569 @cindex namespace in C@t{++}
15571 While a member function is active (in the selected stack frame), your
15572 expressions have the same namespace available as the member function;
15573 that is, @value{GDBN} allows implicit references to the class instance
15574 pointer @code{this} following the same rules as C@t{++}. @code{using}
15575 declarations in the current scope are also respected by @value{GDBN}.
15577 @cindex call overloaded functions
15578 @cindex overloaded functions, calling
15579 @cindex type conversions in C@t{++}
15581 You can call overloaded functions; @value{GDBN} resolves the function
15582 call to the right definition, with some restrictions. @value{GDBN} does not
15583 perform overload resolution involving user-defined type conversions,
15584 calls to constructors, or instantiations of templates that do not exist
15585 in the program. It also cannot handle ellipsis argument lists or
15588 It does perform integral conversions and promotions, floating-point
15589 promotions, arithmetic conversions, pointer conversions, conversions of
15590 class objects to base classes, and standard conversions such as those of
15591 functions or arrays to pointers; it requires an exact match on the
15592 number of function arguments.
15594 Overload resolution is always performed, unless you have specified
15595 @code{set overload-resolution off}. @xref{Debugging C Plus Plus,
15596 ,@value{GDBN} Features for C@t{++}}.
15598 You must specify @code{set overload-resolution off} in order to use an
15599 explicit function signature to call an overloaded function, as in
15601 p 'foo(char,int)'('x', 13)
15604 The @value{GDBN} command-completion facility can simplify this;
15605 see @ref{Completion, ,Command Completion}.
15607 @cindex reference declarations
15609 @value{GDBN} understands variables declared as C@t{++} lvalue or rvalue
15610 references; you can use them in expressions just as you do in C@t{++}
15611 source---they are automatically dereferenced.
15613 In the parameter list shown when @value{GDBN} displays a frame, the values of
15614 reference variables are not displayed (unlike other variables); this
15615 avoids clutter, since references are often used for large structures.
15616 The @emph{address} of a reference variable is always shown, unless
15617 you have specified @samp{set print address off}.
15620 @value{GDBN} supports the C@t{++} name resolution operator @code{::}---your
15621 expressions can use it just as expressions in your program do. Since
15622 one scope may be defined in another, you can use @code{::} repeatedly if
15623 necessary, for example in an expression like
15624 @samp{@var{scope1}::@var{scope2}::@var{name}}. @value{GDBN} also allows
15625 resolving name scope by reference to source files, in both C and C@t{++}
15626 debugging (@pxref{Variables, ,Program Variables}).
15629 @value{GDBN} performs argument-dependent lookup, following the C@t{++}
15634 @subsubsection C and C@t{++} Defaults
15636 @cindex C and C@t{++} defaults
15638 If you allow @value{GDBN} to set range checking automatically, it
15639 defaults to @code{off} whenever the working language changes to
15640 C or C@t{++}. This happens regardless of whether you or @value{GDBN}
15641 selects the working language.
15643 If you allow @value{GDBN} to set the language automatically, it
15644 recognizes source files whose names end with @file{.c}, @file{.C}, or
15645 @file{.cc}, etc, and when @value{GDBN} enters code compiled from one of
15646 these files, it sets the working language to C or C@t{++}.
15647 @xref{Automatically, ,Having @value{GDBN} Infer the Source Language},
15648 for further details.
15651 @subsubsection C and C@t{++} Type and Range Checks
15653 @cindex C and C@t{++} checks
15655 By default, when @value{GDBN} parses C or C@t{++} expressions, strict type
15656 checking is used. However, if you turn type checking off, @value{GDBN}
15657 will allow certain non-standard conversions, such as promoting integer
15658 constants to pointers.
15660 Range checking, if turned on, is done on mathematical operations. Array
15661 indices are not checked, since they are often used to index a pointer
15662 that is not itself an array.
15665 @subsubsection @value{GDBN} and C
15667 The @code{set print union} and @code{show print union} commands apply to
15668 the @code{union} type. When set to @samp{on}, any @code{union} that is
15669 inside a @code{struct} or @code{class} is also printed. Otherwise, it
15670 appears as @samp{@{...@}}.
15672 The @code{@@} operator aids in the debugging of dynamic arrays, formed
15673 with pointers and a memory allocation function. @xref{Expressions,
15676 @node Debugging C Plus Plus
15677 @subsubsection @value{GDBN} Features for C@t{++}
15679 @cindex commands for C@t{++}
15681 Some @value{GDBN} commands are particularly useful with C@t{++}, and some are
15682 designed specifically for use with C@t{++}. Here is a summary:
15685 @cindex break in overloaded functions
15686 @item @r{breakpoint menus}
15687 When you want a breakpoint in a function whose name is overloaded,
15688 @value{GDBN} has the capability to display a menu of possible breakpoint
15689 locations to help you specify which function definition you want.
15690 @xref{Ambiguous Expressions,,Ambiguous Expressions}.
15692 @cindex overloading in C@t{++}
15693 @item rbreak @var{regex}
15694 Setting breakpoints using regular expressions is helpful for setting
15695 breakpoints on overloaded functions that are not members of any special
15697 @xref{Set Breaks, ,Setting Breakpoints}.
15699 @cindex C@t{++} exception handling
15701 @itemx catch rethrow
15703 Debug C@t{++} exception handling using these commands. @xref{Set
15704 Catchpoints, , Setting Catchpoints}.
15706 @cindex inheritance
15707 @item ptype @var{typename}
15708 Print inheritance relationships as well as other information for type
15710 @xref{Symbols, ,Examining the Symbol Table}.
15712 @item info vtbl @var{expression}.
15713 The @code{info vtbl} command can be used to display the virtual
15714 method tables of the object computed by @var{expression}. This shows
15715 one entry per virtual table; there may be multiple virtual tables when
15716 multiple inheritance is in use.
15718 @cindex C@t{++} demangling
15719 @item demangle @var{name}
15720 Demangle @var{name}.
15721 @xref{Symbols}, for a more complete description of the @code{demangle} command.
15723 @cindex C@t{++} symbol display
15724 @item set print demangle
15725 @itemx show print demangle
15726 @itemx set print asm-demangle
15727 @itemx show print asm-demangle
15728 Control whether C@t{++} symbols display in their source form, both when
15729 displaying code as C@t{++} source and when displaying disassemblies.
15730 @xref{Print Settings, ,Print Settings}.
15732 @item set print object
15733 @itemx show print object
15734 Choose whether to print derived (actual) or declared types of objects.
15735 @xref{Print Settings, ,Print Settings}.
15737 @item set print vtbl
15738 @itemx show print vtbl
15739 Control the format for printing virtual function tables.
15740 @xref{Print Settings, ,Print Settings}.
15741 (The @code{vtbl} commands do not work on programs compiled with the HP
15742 ANSI C@t{++} compiler (@code{aCC}).)
15744 @kindex set overload-resolution
15745 @cindex overloaded functions, overload resolution
15746 @item set overload-resolution on
15747 Enable overload resolution for C@t{++} expression evaluation. The default
15748 is on. For overloaded functions, @value{GDBN} evaluates the arguments
15749 and searches for a function whose signature matches the argument types,
15750 using the standard C@t{++} conversion rules (see @ref{C Plus Plus
15751 Expressions, ,C@t{++} Expressions}, for details).
15752 If it cannot find a match, it emits a message.
15754 @item set overload-resolution off
15755 Disable overload resolution for C@t{++} expression evaluation. For
15756 overloaded functions that are not class member functions, @value{GDBN}
15757 chooses the first function of the specified name that it finds in the
15758 symbol table, whether or not its arguments are of the correct type. For
15759 overloaded functions that are class member functions, @value{GDBN}
15760 searches for a function whose signature @emph{exactly} matches the
15763 @kindex show overload-resolution
15764 @item show overload-resolution
15765 Show the current setting of overload resolution.
15767 @item @r{Overloaded symbol names}
15768 You can specify a particular definition of an overloaded symbol, using
15769 the same notation that is used to declare such symbols in C@t{++}: type
15770 @code{@var{symbol}(@var{types})} rather than just @var{symbol}. You can
15771 also use the @value{GDBN} command-line word completion facilities to list the
15772 available choices, or to finish the type list for you.
15773 @xref{Completion,, Command Completion}, for details on how to do this.
15775 @item @r{Breakpoints in functions with ABI tags}
15777 The GNU C@t{++} compiler introduced the notion of ABI ``tags'', which
15778 correspond to changes in the ABI of a type, function, or variable that
15779 would not otherwise be reflected in a mangled name. See
15780 @url{https://developers.redhat.com/blog/2015/02/05/gcc5-and-the-c11-abi/}
15783 The ABI tags are visible in C@t{++} demangled names. For example, a
15784 function that returns a std::string:
15787 std::string function(int);
15791 when compiled for the C++11 ABI is marked with the @code{cxx11} ABI
15792 tag, and @value{GDBN} displays the symbol like this:
15795 function[abi:cxx11](int)
15798 You can set a breakpoint on such functions simply as if they had no
15802 (gdb) b function(int)
15803 Breakpoint 2 at 0x40060d: file main.cc, line 10.
15804 (gdb) info breakpoints
15805 Num Type Disp Enb Address What
15806 1 breakpoint keep y 0x0040060d in function[abi:cxx11](int)
15810 On the rare occasion you need to disambiguate between different ABI
15811 tags, you can do so by simply including the ABI tag in the function
15815 (@value{GDBP}) b ambiguous[abi:other_tag](int)
15819 @node Decimal Floating Point
15820 @subsubsection Decimal Floating Point format
15821 @cindex decimal floating point format
15823 @value{GDBN} can examine, set and perform computations with numbers in
15824 decimal floating point format, which in the C language correspond to the
15825 @code{_Decimal32}, @code{_Decimal64} and @code{_Decimal128} types as
15826 specified by the extension to support decimal floating-point arithmetic.
15828 There are two encodings in use, depending on the architecture: BID (Binary
15829 Integer Decimal) for x86 and x86-64, and DPD (Densely Packed Decimal) for
15830 PowerPC and S/390. @value{GDBN} will use the appropriate encoding for the
15833 Because of a limitation in @file{libdecnumber}, the library used by @value{GDBN}
15834 to manipulate decimal floating point numbers, it is not possible to convert
15835 (using a cast, for example) integers wider than 32-bit to decimal float.
15837 In addition, in order to imitate @value{GDBN}'s behaviour with binary floating
15838 point computations, error checking in decimal float operations ignores
15839 underflow, overflow and divide by zero exceptions.
15841 In the PowerPC architecture, @value{GDBN} provides a set of pseudo-registers
15842 to inspect @code{_Decimal128} values stored in floating point registers.
15843 See @ref{PowerPC,,PowerPC} for more details.
15849 @value{GDBN} can be used to debug programs written in D and compiled with
15850 GDC, LDC or DMD compilers. Currently @value{GDBN} supports only one D
15851 specific feature --- dynamic arrays.
15856 @cindex Go (programming language)
15857 @value{GDBN} can be used to debug programs written in Go and compiled with
15858 @file{gccgo} or @file{6g} compilers.
15860 Here is a summary of the Go-specific features and restrictions:
15863 @cindex current Go package
15864 @item The current Go package
15865 The name of the current package does not need to be specified when
15866 specifying global variables and functions.
15868 For example, given the program:
15872 var myglob = "Shall we?"
15878 When stopped inside @code{main} either of these work:
15882 (gdb) p main.myglob
15885 @cindex builtin Go types
15886 @item Builtin Go types
15887 The @code{string} type is recognized by @value{GDBN} and is printed
15890 @cindex builtin Go functions
15891 @item Builtin Go functions
15892 The @value{GDBN} expression parser recognizes the @code{unsafe.Sizeof}
15893 function and handles it internally.
15895 @cindex restrictions on Go expressions
15896 @item Restrictions on Go expressions
15897 All Go operators are supported except @code{&^}.
15898 The Go @code{_} ``blank identifier'' is not supported.
15899 Automatic dereferencing of pointers is not supported.
15903 @subsection Objective-C
15905 @cindex Objective-C
15906 This section provides information about some commands and command
15907 options that are useful for debugging Objective-C code. See also
15908 @ref{Symbols, info classes}, and @ref{Symbols, info selectors}, for a
15909 few more commands specific to Objective-C support.
15912 * Method Names in Commands::
15913 * The Print Command with Objective-C::
15916 @node Method Names in Commands
15917 @subsubsection Method Names in Commands
15919 The following commands have been extended to accept Objective-C method
15920 names as line specifications:
15922 @kindex clear@r{, and Objective-C}
15923 @kindex break@r{, and Objective-C}
15924 @kindex info line@r{, and Objective-C}
15925 @kindex jump@r{, and Objective-C}
15926 @kindex list@r{, and Objective-C}
15930 @item @code{info line}
15935 A fully qualified Objective-C method name is specified as
15938 -[@var{Class} @var{methodName}]
15941 where the minus sign is used to indicate an instance method and a
15942 plus sign (not shown) is used to indicate a class method. The class
15943 name @var{Class} and method name @var{methodName} are enclosed in
15944 brackets, similar to the way messages are specified in Objective-C
15945 source code. For example, to set a breakpoint at the @code{create}
15946 instance method of class @code{Fruit} in the program currently being
15950 break -[Fruit create]
15953 To list ten program lines around the @code{initialize} class method,
15957 list +[NSText initialize]
15960 In the current version of @value{GDBN}, the plus or minus sign is
15961 required. In future versions of @value{GDBN}, the plus or minus
15962 sign will be optional, but you can use it to narrow the search. It
15963 is also possible to specify just a method name:
15969 You must specify the complete method name, including any colons. If
15970 your program's source files contain more than one @code{create} method,
15971 you'll be presented with a numbered list of classes that implement that
15972 method. Indicate your choice by number, or type @samp{0} to exit if
15975 As another example, to clear a breakpoint established at the
15976 @code{makeKeyAndOrderFront:} method of the @code{NSWindow} class, enter:
15979 clear -[NSWindow makeKeyAndOrderFront:]
15982 @node The Print Command with Objective-C
15983 @subsubsection The Print Command With Objective-C
15984 @cindex Objective-C, print objects
15985 @kindex print-object
15986 @kindex po @r{(@code{print-object})}
15988 The print command has also been extended to accept methods. For example:
15991 print -[@var{object} hash]
15994 @cindex print an Objective-C object description
15995 @cindex @code{_NSPrintForDebugger}, and printing Objective-C objects
15997 will tell @value{GDBN} to send the @code{hash} message to @var{object}
15998 and print the result. Also, an additional command has been added,
15999 @code{print-object} or @code{po} for short, which is meant to print
16000 the description of an object. However, this command may only work
16001 with certain Objective-C libraries that have a particular hook
16002 function, @code{_NSPrintForDebugger}, defined.
16005 @subsection OpenCL C
16008 This section provides information about @value{GDBN}s OpenCL C support.
16011 * OpenCL C Datatypes::
16012 * OpenCL C Expressions::
16013 * OpenCL C Operators::
16016 @node OpenCL C Datatypes
16017 @subsubsection OpenCL C Datatypes
16019 @cindex OpenCL C Datatypes
16020 @value{GDBN} supports the builtin scalar and vector datatypes specified
16021 by OpenCL 1.1. In addition the half- and double-precision floating point
16022 data types of the @code{cl_khr_fp16} and @code{cl_khr_fp64} OpenCL
16023 extensions are also known to @value{GDBN}.
16025 @node OpenCL C Expressions
16026 @subsubsection OpenCL C Expressions
16028 @cindex OpenCL C Expressions
16029 @value{GDBN} supports accesses to vector components including the access as
16030 lvalue where possible. Since OpenCL C is based on C99 most C expressions
16031 supported by @value{GDBN} can be used as well.
16033 @node OpenCL C Operators
16034 @subsubsection OpenCL C Operators
16036 @cindex OpenCL C Operators
16037 @value{GDBN} supports the operators specified by OpenCL 1.1 for scalar and
16041 @subsection Fortran
16042 @cindex Fortran-specific support in @value{GDBN}
16044 @value{GDBN} can be used to debug programs written in Fortran, but it
16045 currently supports only the features of Fortran 77 language.
16047 @cindex trailing underscore, in Fortran symbols
16048 Some Fortran compilers (@sc{gnu} Fortran 77 and Fortran 95 compilers
16049 among them) append an underscore to the names of variables and
16050 functions. When you debug programs compiled by those compilers, you
16051 will need to refer to variables and functions with a trailing
16055 * Fortran Operators:: Fortran operators and expressions
16056 * Fortran Defaults:: Default settings for Fortran
16057 * Special Fortran Commands:: Special @value{GDBN} commands for Fortran
16060 @node Fortran Operators
16061 @subsubsection Fortran Operators and Expressions
16063 @cindex Fortran operators and expressions
16065 Operators must be defined on values of specific types. For instance,
16066 @code{+} is defined on numbers, but not on characters or other non-
16067 arithmetic types. Operators are often defined on groups of types.
16071 The exponentiation operator. It raises the first operand to the power
16075 The range operator. Normally used in the form of array(low:high) to
16076 represent a section of array.
16079 The access component operator. Normally used to access elements in derived
16080 types. Also suitable for unions. As unions aren't part of regular Fortran,
16081 this can only happen when accessing a register that uses a gdbarch-defined
16085 @node Fortran Defaults
16086 @subsubsection Fortran Defaults
16088 @cindex Fortran Defaults
16090 Fortran symbols are usually case-insensitive, so @value{GDBN} by
16091 default uses case-insensitive matches for Fortran symbols. You can
16092 change that with the @samp{set case-insensitive} command, see
16093 @ref{Symbols}, for the details.
16095 @node Special Fortran Commands
16096 @subsubsection Special Fortran Commands
16098 @cindex Special Fortran commands
16100 @value{GDBN} has some commands to support Fortran-specific features,
16101 such as displaying common blocks.
16104 @cindex @code{COMMON} blocks, Fortran
16105 @kindex info common
16106 @item info common @r{[}@var{common-name}@r{]}
16107 This command prints the values contained in the Fortran @code{COMMON}
16108 block whose name is @var{common-name}. With no argument, the names of
16109 all @code{COMMON} blocks visible at the current program location are
16116 @cindex Pascal support in @value{GDBN}, limitations
16117 Debugging Pascal programs which use sets, subranges, file variables, or
16118 nested functions does not currently work. @value{GDBN} does not support
16119 entering expressions, printing values, or similar features using Pascal
16122 The Pascal-specific command @code{set print pascal_static-members}
16123 controls whether static members of Pascal objects are displayed.
16124 @xref{Print Settings, pascal_static-members}.
16129 @value{GDBN} supports the @url{https://www.rust-lang.org/, Rust
16130 Programming Language}. Type- and value-printing, and expression
16131 parsing, are reasonably complete. However, there are a few
16132 peculiarities and holes to be aware of.
16136 Linespecs (@pxref{Specify Location}) are never relative to the current
16137 crate. Instead, they act as if there were a global namespace of
16138 crates, somewhat similar to the way @code{extern crate} behaves.
16140 That is, if @value{GDBN} is stopped at a breakpoint in a function in
16141 crate @samp{A}, module @samp{B}, then @code{break B::f} will attempt
16142 to set a breakpoint in a function named @samp{f} in a crate named
16145 As a consequence of this approach, linespecs also cannot refer to
16146 items using @samp{self::} or @samp{super::}.
16149 Because @value{GDBN} implements Rust name-lookup semantics in
16150 expressions, it will sometimes prepend the current crate to a name.
16151 For example, if @value{GDBN} is stopped at a breakpoint in the crate
16152 @samp{K}, then @code{print ::x::y} will try to find the symbol
16155 However, since it is useful to be able to refer to other crates when
16156 debugging, @value{GDBN} provides the @code{extern} extension to
16157 circumvent this. To use the extension, just put @code{extern} before
16158 a path expression to refer to the otherwise unavailable ``global''
16161 In the above example, if you wanted to refer to the symbol @samp{y} in
16162 the crate @samp{x}, you would use @code{print extern x::y}.
16165 The Rust expression evaluator does not support ``statement-like''
16166 expressions such as @code{if} or @code{match}, or lambda expressions.
16169 Tuple expressions are not implemented.
16172 The Rust expression evaluator does not currently implement the
16173 @code{Drop} trait. Objects that may be created by the evaluator will
16174 never be destroyed.
16177 @value{GDBN} does not implement type inference for generics. In order
16178 to call generic functions or otherwise refer to generic items, you
16179 will have to specify the type parameters manually.
16182 @value{GDBN} currently uses the C@t{++} demangler for Rust. In most
16183 cases this does not cause any problems. However, in an expression
16184 context, completing a generic function name will give syntactically
16185 invalid results. This happens because Rust requires the @samp{::}
16186 operator between the function name and its generic arguments. For
16187 example, @value{GDBN} might provide a completion like
16188 @code{crate::f<u32>}, where the parser would require
16189 @code{crate::f::<u32>}.
16192 As of this writing, the Rust compiler (version 1.8) has a few holes in
16193 the debugging information it generates. These holes prevent certain
16194 features from being implemented by @value{GDBN}:
16198 Method calls cannot be made via traits.
16201 Operator overloading is not implemented.
16204 When debugging in a monomorphized function, you cannot use the generic
16208 The type @code{Self} is not available.
16211 @code{use} statements are not available, so some names may not be
16212 available in the crate.
16217 @subsection Modula-2
16219 @cindex Modula-2, @value{GDBN} support
16221 The extensions made to @value{GDBN} to support Modula-2 only support
16222 output from the @sc{gnu} Modula-2 compiler (which is currently being
16223 developed). Other Modula-2 compilers are not currently supported, and
16224 attempting to debug executables produced by them is most likely
16225 to give an error as @value{GDBN} reads in the executable's symbol
16228 @cindex expressions in Modula-2
16230 * M2 Operators:: Built-in operators
16231 * Built-In Func/Proc:: Built-in functions and procedures
16232 * M2 Constants:: Modula-2 constants
16233 * M2 Types:: Modula-2 types
16234 * M2 Defaults:: Default settings for Modula-2
16235 * Deviations:: Deviations from standard Modula-2
16236 * M2 Checks:: Modula-2 type and range checks
16237 * M2 Scope:: The scope operators @code{::} and @code{.}
16238 * GDB/M2:: @value{GDBN} and Modula-2
16242 @subsubsection Operators
16243 @cindex Modula-2 operators
16245 Operators must be defined on values of specific types. For instance,
16246 @code{+} is defined on numbers, but not on structures. Operators are
16247 often defined on groups of types. For the purposes of Modula-2, the
16248 following definitions hold:
16253 @emph{Integral types} consist of @code{INTEGER}, @code{CARDINAL}, and
16257 @emph{Character types} consist of @code{CHAR} and its subranges.
16260 @emph{Floating-point types} consist of @code{REAL}.
16263 @emph{Pointer types} consist of anything declared as @code{POINTER TO
16267 @emph{Scalar types} consist of all of the above.
16270 @emph{Set types} consist of @code{SET} and @code{BITSET} types.
16273 @emph{Boolean types} consist of @code{BOOLEAN}.
16277 The following operators are supported, and appear in order of
16278 increasing precedence:
16282 Function argument or array index separator.
16285 Assignment. The value of @var{var} @code{:=} @var{value} is
16289 Less than, greater than on integral, floating-point, or enumerated
16293 Less than or equal to, greater than or equal to
16294 on integral, floating-point and enumerated types, or set inclusion on
16295 set types. Same precedence as @code{<}.
16297 @item =@r{, }<>@r{, }#
16298 Equality and two ways of expressing inequality, valid on scalar types.
16299 Same precedence as @code{<}. In @value{GDBN} scripts, only @code{<>} is
16300 available for inequality, since @code{#} conflicts with the script
16304 Set membership. Defined on set types and the types of their members.
16305 Same precedence as @code{<}.
16308 Boolean disjunction. Defined on boolean types.
16311 Boolean conjunction. Defined on boolean types.
16314 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
16317 Addition and subtraction on integral and floating-point types, or union
16318 and difference on set types.
16321 Multiplication on integral and floating-point types, or set intersection
16325 Division on floating-point types, or symmetric set difference on set
16326 types. Same precedence as @code{*}.
16329 Integer division and remainder. Defined on integral types. Same
16330 precedence as @code{*}.
16333 Negative. Defined on @code{INTEGER} and @code{REAL} data.
16336 Pointer dereferencing. Defined on pointer types.
16339 Boolean negation. Defined on boolean types. Same precedence as
16343 @code{RECORD} field selector. Defined on @code{RECORD} data. Same
16344 precedence as @code{^}.
16347 Array indexing. Defined on @code{ARRAY} data. Same precedence as @code{^}.
16350 Procedure argument list. Defined on @code{PROCEDURE} objects. Same precedence
16354 @value{GDBN} and Modula-2 scope operators.
16358 @emph{Warning:} Set expressions and their operations are not yet supported, so @value{GDBN}
16359 treats the use of the operator @code{IN}, or the use of operators
16360 @code{+}, @code{-}, @code{*}, @code{/}, @code{=}, , @code{<>}, @code{#},
16361 @code{<=}, and @code{>=} on sets as an error.
16365 @node Built-In Func/Proc
16366 @subsubsection Built-in Functions and Procedures
16367 @cindex Modula-2 built-ins
16369 Modula-2 also makes available several built-in procedures and functions.
16370 In describing these, the following metavariables are used:
16375 represents an @code{ARRAY} variable.
16378 represents a @code{CHAR} constant or variable.
16381 represents a variable or constant of integral type.
16384 represents an identifier that belongs to a set. Generally used in the
16385 same function with the metavariable @var{s}. The type of @var{s} should
16386 be @code{SET OF @var{mtype}} (where @var{mtype} is the type of @var{m}).
16389 represents a variable or constant of integral or floating-point type.
16392 represents a variable or constant of floating-point type.
16398 represents a variable.
16401 represents a variable or constant of one of many types. See the
16402 explanation of the function for details.
16405 All Modula-2 built-in procedures also return a result, described below.
16409 Returns the absolute value of @var{n}.
16412 If @var{c} is a lower case letter, it returns its upper case
16413 equivalent, otherwise it returns its argument.
16416 Returns the character whose ordinal value is @var{i}.
16419 Decrements the value in the variable @var{v} by one. Returns the new value.
16421 @item DEC(@var{v},@var{i})
16422 Decrements the value in the variable @var{v} by @var{i}. Returns the
16425 @item EXCL(@var{m},@var{s})
16426 Removes the element @var{m} from the set @var{s}. Returns the new
16429 @item FLOAT(@var{i})
16430 Returns the floating point equivalent of the integer @var{i}.
16432 @item HIGH(@var{a})
16433 Returns the index of the last member of @var{a}.
16436 Increments the value in the variable @var{v} by one. Returns the new value.
16438 @item INC(@var{v},@var{i})
16439 Increments the value in the variable @var{v} by @var{i}. Returns the
16442 @item INCL(@var{m},@var{s})
16443 Adds the element @var{m} to the set @var{s} if it is not already
16444 there. Returns the new set.
16447 Returns the maximum value of the type @var{t}.
16450 Returns the minimum value of the type @var{t}.
16453 Returns boolean TRUE if @var{i} is an odd number.
16456 Returns the ordinal value of its argument. For example, the ordinal
16457 value of a character is its @sc{ascii} value (on machines supporting
16458 the @sc{ascii} character set). The argument @var{x} must be of an
16459 ordered type, which include integral, character and enumerated types.
16461 @item SIZE(@var{x})
16462 Returns the size of its argument. The argument @var{x} can be a
16463 variable or a type.
16465 @item TRUNC(@var{r})
16466 Returns the integral part of @var{r}.
16468 @item TSIZE(@var{x})
16469 Returns the size of its argument. The argument @var{x} can be a
16470 variable or a type.
16472 @item VAL(@var{t},@var{i})
16473 Returns the member of the type @var{t} whose ordinal value is @var{i}.
16477 @emph{Warning:} Sets and their operations are not yet supported, so
16478 @value{GDBN} treats the use of procedures @code{INCL} and @code{EXCL} as
16482 @cindex Modula-2 constants
16484 @subsubsection Constants
16486 @value{GDBN} allows you to express the constants of Modula-2 in the following
16492 Integer constants are simply a sequence of digits. When used in an
16493 expression, a constant is interpreted to be type-compatible with the
16494 rest of the expression. Hexadecimal integers are specified by a
16495 trailing @samp{H}, and octal integers by a trailing @samp{B}.
16498 Floating point constants appear as a sequence of digits, followed by a
16499 decimal point and another sequence of digits. An optional exponent can
16500 then be specified, in the form @samp{E@r{[}+@r{|}-@r{]}@var{nnn}}, where
16501 @samp{@r{[}+@r{|}-@r{]}@var{nnn}} is the desired exponent. All of the
16502 digits of the floating point constant must be valid decimal (base 10)
16506 Character constants consist of a single character enclosed by a pair of
16507 like quotes, either single (@code{'}) or double (@code{"}). They may
16508 also be expressed by their ordinal value (their @sc{ascii} value, usually)
16509 followed by a @samp{C}.
16512 String constants consist of a sequence of characters enclosed by a
16513 pair of like quotes, either single (@code{'}) or double (@code{"}).
16514 Escape sequences in the style of C are also allowed. @xref{C
16515 Constants, ,C and C@t{++} Constants}, for a brief explanation of escape
16519 Enumerated constants consist of an enumerated identifier.
16522 Boolean constants consist of the identifiers @code{TRUE} and
16526 Pointer constants consist of integral values only.
16529 Set constants are not yet supported.
16533 @subsubsection Modula-2 Types
16534 @cindex Modula-2 types
16536 Currently @value{GDBN} can print the following data types in Modula-2
16537 syntax: array types, record types, set types, pointer types, procedure
16538 types, enumerated types, subrange types and base types. You can also
16539 print the contents of variables declared using these type.
16540 This section gives a number of simple source code examples together with
16541 sample @value{GDBN} sessions.
16543 The first example contains the following section of code:
16552 and you can request @value{GDBN} to interrogate the type and value of
16553 @code{r} and @code{s}.
16556 (@value{GDBP}) print s
16558 (@value{GDBP}) ptype s
16560 (@value{GDBP}) print r
16562 (@value{GDBP}) ptype r
16567 Likewise if your source code declares @code{s} as:
16571 s: SET ['A'..'Z'] ;
16575 then you may query the type of @code{s} by:
16578 (@value{GDBP}) ptype s
16579 type = SET ['A'..'Z']
16583 Note that at present you cannot interactively manipulate set
16584 expressions using the debugger.
16586 The following example shows how you might declare an array in Modula-2
16587 and how you can interact with @value{GDBN} to print its type and contents:
16591 s: ARRAY [-10..10] OF CHAR ;
16595 (@value{GDBP}) ptype s
16596 ARRAY [-10..10] OF CHAR
16599 Note that the array handling is not yet complete and although the type
16600 is printed correctly, expression handling still assumes that all
16601 arrays have a lower bound of zero and not @code{-10} as in the example
16604 Here are some more type related Modula-2 examples:
16608 colour = (blue, red, yellow, green) ;
16609 t = [blue..yellow] ;
16617 The @value{GDBN} interaction shows how you can query the data type
16618 and value of a variable.
16621 (@value{GDBP}) print s
16623 (@value{GDBP}) ptype t
16624 type = [blue..yellow]
16628 In this example a Modula-2 array is declared and its contents
16629 displayed. Observe that the contents are written in the same way as
16630 their @code{C} counterparts.
16634 s: ARRAY [1..5] OF CARDINAL ;
16640 (@value{GDBP}) print s
16641 $1 = @{1, 0, 0, 0, 0@}
16642 (@value{GDBP}) ptype s
16643 type = ARRAY [1..5] OF CARDINAL
16646 The Modula-2 language interface to @value{GDBN} also understands
16647 pointer types as shown in this example:
16651 s: POINTER TO ARRAY [1..5] OF CARDINAL ;
16658 and you can request that @value{GDBN} describes the type of @code{s}.
16661 (@value{GDBP}) ptype s
16662 type = POINTER TO ARRAY [1..5] OF CARDINAL
16665 @value{GDBN} handles compound types as we can see in this example.
16666 Here we combine array types, record types, pointer types and subrange
16677 myarray = ARRAY myrange OF CARDINAL ;
16678 myrange = [-2..2] ;
16680 s: POINTER TO ARRAY myrange OF foo ;
16684 and you can ask @value{GDBN} to describe the type of @code{s} as shown
16688 (@value{GDBP}) ptype s
16689 type = POINTER TO ARRAY [-2..2] OF foo = RECORD
16692 f3 : ARRAY [-2..2] OF CARDINAL;
16697 @subsubsection Modula-2 Defaults
16698 @cindex Modula-2 defaults
16700 If type and range checking are set automatically by @value{GDBN}, they
16701 both default to @code{on} whenever the working language changes to
16702 Modula-2. This happens regardless of whether you or @value{GDBN}
16703 selected the working language.
16705 If you allow @value{GDBN} to set the language automatically, then entering
16706 code compiled from a file whose name ends with @file{.mod} sets the
16707 working language to Modula-2. @xref{Automatically, ,Having @value{GDBN}
16708 Infer the Source Language}, for further details.
16711 @subsubsection Deviations from Standard Modula-2
16712 @cindex Modula-2, deviations from
16714 A few changes have been made to make Modula-2 programs easier to debug.
16715 This is done primarily via loosening its type strictness:
16719 Unlike in standard Modula-2, pointer constants can be formed by
16720 integers. This allows you to modify pointer variables during
16721 debugging. (In standard Modula-2, the actual address contained in a
16722 pointer variable is hidden from you; it can only be modified
16723 through direct assignment to another pointer variable or expression that
16724 returned a pointer.)
16727 C escape sequences can be used in strings and characters to represent
16728 non-printable characters. @value{GDBN} prints out strings with these
16729 escape sequences embedded. Single non-printable characters are
16730 printed using the @samp{CHR(@var{nnn})} format.
16733 The assignment operator (@code{:=}) returns the value of its right-hand
16737 All built-in procedures both modify @emph{and} return their argument.
16741 @subsubsection Modula-2 Type and Range Checks
16742 @cindex Modula-2 checks
16745 @emph{Warning:} in this release, @value{GDBN} does not yet perform type or
16748 @c FIXME remove warning when type/range checks added
16750 @value{GDBN} considers two Modula-2 variables type equivalent if:
16754 They are of types that have been declared equivalent via a @code{TYPE
16755 @var{t1} = @var{t2}} statement
16758 They have been declared on the same line. (Note: This is true of the
16759 @sc{gnu} Modula-2 compiler, but it may not be true of other compilers.)
16762 As long as type checking is enabled, any attempt to combine variables
16763 whose types are not equivalent is an error.
16765 Range checking is done on all mathematical operations, assignment, array
16766 index bounds, and all built-in functions and procedures.
16769 @subsubsection The Scope Operators @code{::} and @code{.}
16771 @cindex @code{.}, Modula-2 scope operator
16772 @cindex colon, doubled as scope operator
16774 @vindex colon-colon@r{, in Modula-2}
16775 @c Info cannot handle :: but TeX can.
16778 @vindex ::@r{, in Modula-2}
16781 There are a few subtle differences between the Modula-2 scope operator
16782 (@code{.}) and the @value{GDBN} scope operator (@code{::}). The two have
16787 @var{module} . @var{id}
16788 @var{scope} :: @var{id}
16792 where @var{scope} is the name of a module or a procedure,
16793 @var{module} the name of a module, and @var{id} is any declared
16794 identifier within your program, except another module.
16796 Using the @code{::} operator makes @value{GDBN} search the scope
16797 specified by @var{scope} for the identifier @var{id}. If it is not
16798 found in the specified scope, then @value{GDBN} searches all scopes
16799 enclosing the one specified by @var{scope}.
16801 Using the @code{.} operator makes @value{GDBN} search the current scope for
16802 the identifier specified by @var{id} that was imported from the
16803 definition module specified by @var{module}. With this operator, it is
16804 an error if the identifier @var{id} was not imported from definition
16805 module @var{module}, or if @var{id} is not an identifier in
16809 @subsubsection @value{GDBN} and Modula-2
16811 Some @value{GDBN} commands have little use when debugging Modula-2 programs.
16812 Five subcommands of @code{set print} and @code{show print} apply
16813 specifically to C and C@t{++}: @samp{vtbl}, @samp{demangle},
16814 @samp{asm-demangle}, @samp{object}, and @samp{union}. The first four
16815 apply to C@t{++}, and the last to the C @code{union} type, which has no direct
16816 analogue in Modula-2.
16818 The @code{@@} operator (@pxref{Expressions, ,Expressions}), while available
16819 with any language, is not useful with Modula-2. Its
16820 intent is to aid the debugging of @dfn{dynamic arrays}, which cannot be
16821 created in Modula-2 as they can in C or C@t{++}. However, because an
16822 address can be specified by an integral constant, the construct
16823 @samp{@{@var{type}@}@var{adrexp}} is still useful.
16825 @cindex @code{#} in Modula-2
16826 In @value{GDBN} scripts, the Modula-2 inequality operator @code{#} is
16827 interpreted as the beginning of a comment. Use @code{<>} instead.
16833 The extensions made to @value{GDBN} for Ada only support
16834 output from the @sc{gnu} Ada (GNAT) compiler.
16835 Other Ada compilers are not currently supported, and
16836 attempting to debug executables produced by them is most likely
16840 @cindex expressions in Ada
16842 * Ada Mode Intro:: General remarks on the Ada syntax
16843 and semantics supported by Ada mode
16845 * Omissions from Ada:: Restrictions on the Ada expression syntax.
16846 * Additions to Ada:: Extensions of the Ada expression syntax.
16847 * Overloading support for Ada:: Support for expressions involving overloaded
16849 * Stopping Before Main Program:: Debugging the program during elaboration.
16850 * Ada Exceptions:: Ada Exceptions
16851 * Ada Tasks:: Listing and setting breakpoints in tasks.
16852 * Ada Tasks and Core Files:: Tasking Support when Debugging Core Files
16853 * Ravenscar Profile:: Tasking Support when using the Ravenscar
16855 * Ada Settings:: New settable GDB parameters for Ada.
16856 * Ada Glitches:: Known peculiarities of Ada mode.
16859 @node Ada Mode Intro
16860 @subsubsection Introduction
16861 @cindex Ada mode, general
16863 The Ada mode of @value{GDBN} supports a fairly large subset of Ada expression
16864 syntax, with some extensions.
16865 The philosophy behind the design of this subset is
16869 That @value{GDBN} should provide basic literals and access to operations for
16870 arithmetic, dereferencing, field selection, indexing, and subprogram calls,
16871 leaving more sophisticated computations to subprograms written into the
16872 program (which therefore may be called from @value{GDBN}).
16875 That type safety and strict adherence to Ada language restrictions
16876 are not particularly important to the @value{GDBN} user.
16879 That brevity is important to the @value{GDBN} user.
16882 Thus, for brevity, the debugger acts as if all names declared in
16883 user-written packages are directly visible, even if they are not visible
16884 according to Ada rules, thus making it unnecessary to fully qualify most
16885 names with their packages, regardless of context. Where this causes
16886 ambiguity, @value{GDBN} asks the user's intent.
16888 The debugger will start in Ada mode if it detects an Ada main program.
16889 As for other languages, it will enter Ada mode when stopped in a program that
16890 was translated from an Ada source file.
16892 While in Ada mode, you may use `@t{--}' for comments. This is useful
16893 mostly for documenting command files. The standard @value{GDBN} comment
16894 (@samp{#}) still works at the beginning of a line in Ada mode, but not in the
16895 middle (to allow based literals).
16897 @node Omissions from Ada
16898 @subsubsection Omissions from Ada
16899 @cindex Ada, omissions from
16901 Here are the notable omissions from the subset:
16905 Only a subset of the attributes are supported:
16909 @t{'First}, @t{'Last}, and @t{'Length}
16910 on array objects (not on types and subtypes).
16913 @t{'Min} and @t{'Max}.
16916 @t{'Pos} and @t{'Val}.
16922 @t{'Range} on array objects (not subtypes), but only as the right
16923 operand of the membership (@code{in}) operator.
16926 @t{'Access}, @t{'Unchecked_Access}, and
16927 @t{'Unrestricted_Access} (a GNAT extension).
16935 @code{Characters.Latin_1} are not available and
16936 concatenation is not implemented. Thus, escape characters in strings are
16937 not currently available.
16940 Equality tests (@samp{=} and @samp{/=}) on arrays test for bitwise
16941 equality of representations. They will generally work correctly
16942 for strings and arrays whose elements have integer or enumeration types.
16943 They may not work correctly for arrays whose element
16944 types have user-defined equality, for arrays of real values
16945 (in particular, IEEE-conformant floating point, because of negative
16946 zeroes and NaNs), and for arrays whose elements contain unused bits with
16947 indeterminate values.
16950 The other component-by-component array operations (@code{and}, @code{or},
16951 @code{xor}, @code{not}, and relational tests other than equality)
16952 are not implemented.
16955 @cindex array aggregates (Ada)
16956 @cindex record aggregates (Ada)
16957 @cindex aggregates (Ada)
16958 There is limited support for array and record aggregates. They are
16959 permitted only on the right sides of assignments, as in these examples:
16962 (@value{GDBP}) set An_Array := (1, 2, 3, 4, 5, 6)
16963 (@value{GDBP}) set An_Array := (1, others => 0)
16964 (@value{GDBP}) set An_Array := (0|4 => 1, 1..3 => 2, 5 => 6)
16965 (@value{GDBP}) set A_2D_Array := ((1, 2, 3), (4, 5, 6), (7, 8, 9))
16966 (@value{GDBP}) set A_Record := (1, "Peter", True);
16967 (@value{GDBP}) set A_Record := (Name => "Peter", Id => 1, Alive => True)
16971 discriminant's value by assigning an aggregate has an
16972 undefined effect if that discriminant is used within the record.
16973 However, you can first modify discriminants by directly assigning to
16974 them (which normally would not be allowed in Ada), and then performing an
16975 aggregate assignment. For example, given a variable @code{A_Rec}
16976 declared to have a type such as:
16979 type Rec (Len : Small_Integer := 0) is record
16981 Vals : IntArray (1 .. Len);
16985 you can assign a value with a different size of @code{Vals} with two
16989 (@value{GDBP}) set A_Rec.Len := 4
16990 (@value{GDBP}) set A_Rec := (Id => 42, Vals => (1, 2, 3, 4))
16993 As this example also illustrates, @value{GDBN} is very loose about the usual
16994 rules concerning aggregates. You may leave out some of the
16995 components of an array or record aggregate (such as the @code{Len}
16996 component in the assignment to @code{A_Rec} above); they will retain their
16997 original values upon assignment. You may freely use dynamic values as
16998 indices in component associations. You may even use overlapping or
16999 redundant component associations, although which component values are
17000 assigned in such cases is not defined.
17003 Calls to dispatching subprograms are not implemented.
17006 The overloading algorithm is much more limited (i.e., less selective)
17007 than that of real Ada. It makes only limited use of the context in
17008 which a subexpression appears to resolve its meaning, and it is much
17009 looser in its rules for allowing type matches. As a result, some
17010 function calls will be ambiguous, and the user will be asked to choose
17011 the proper resolution.
17014 The @code{new} operator is not implemented.
17017 Entry calls are not implemented.
17020 Aside from printing, arithmetic operations on the native VAX floating-point
17021 formats are not supported.
17024 It is not possible to slice a packed array.
17027 The names @code{True} and @code{False}, when not part of a qualified name,
17028 are interpreted as if implicitly prefixed by @code{Standard}, regardless of
17030 Should your program
17031 redefine these names in a package or procedure (at best a dubious practice),
17032 you will have to use fully qualified names to access their new definitions.
17035 @node Additions to Ada
17036 @subsubsection Additions to Ada
17037 @cindex Ada, deviations from
17039 As it does for other languages, @value{GDBN} makes certain generic
17040 extensions to Ada (@pxref{Expressions}):
17044 If the expression @var{E} is a variable residing in memory (typically
17045 a local variable or array element) and @var{N} is a positive integer,
17046 then @code{@var{E}@@@var{N}} displays the values of @var{E} and the
17047 @var{N}-1 adjacent variables following it in memory as an array. In
17048 Ada, this operator is generally not necessary, since its prime use is
17049 in displaying parts of an array, and slicing will usually do this in
17050 Ada. However, there are occasional uses when debugging programs in
17051 which certain debugging information has been optimized away.
17054 @code{@var{B}::@var{var}} means ``the variable named @var{var} that
17055 appears in function or file @var{B}.'' When @var{B} is a file name,
17056 you must typically surround it in single quotes.
17059 The expression @code{@{@var{type}@} @var{addr}} means ``the variable of type
17060 @var{type} that appears at address @var{addr}.''
17063 A name starting with @samp{$} is a convenience variable
17064 (@pxref{Convenience Vars}) or a machine register (@pxref{Registers}).
17067 In addition, @value{GDBN} provides a few other shortcuts and outright
17068 additions specific to Ada:
17072 The assignment statement is allowed as an expression, returning
17073 its right-hand operand as its value. Thus, you may enter
17076 (@value{GDBP}) set x := y + 3
17077 (@value{GDBP}) print A(tmp := y + 1)
17081 The semicolon is allowed as an ``operator,'' returning as its value
17082 the value of its right-hand operand.
17083 This allows, for example,
17084 complex conditional breaks:
17087 (@value{GDBP}) break f
17088 (@value{GDBP}) condition 1 (report(i); k += 1; A(k) > 100)
17092 Rather than use catenation and symbolic character names to introduce special
17093 characters into strings, one may instead use a special bracket notation,
17094 which is also used to print strings. A sequence of characters of the form
17095 @samp{["@var{XX}"]} within a string or character literal denotes the
17096 (single) character whose numeric encoding is @var{XX} in hexadecimal. The
17097 sequence of characters @samp{["""]} also denotes a single quotation mark
17098 in strings. For example,
17100 "One line.["0a"]Next line.["0a"]"
17103 contains an ASCII newline character (@code{Ada.Characters.Latin_1.LF})
17107 The subtype used as a prefix for the attributes @t{'Pos}, @t{'Min}, and
17108 @t{'Max} is optional (and is ignored in any case). For example, it is valid
17112 (@value{GDBP}) print 'max(x, y)
17116 When printing arrays, @value{GDBN} uses positional notation when the
17117 array has a lower bound of 1, and uses a modified named notation otherwise.
17118 For example, a one-dimensional array of three integers with a lower bound
17119 of 3 might print as
17126 That is, in contrast to valid Ada, only the first component has a @code{=>}
17130 You may abbreviate attributes in expressions with any unique,
17131 multi-character subsequence of
17132 their names (an exact match gets preference).
17133 For example, you may use @t{a'len}, @t{a'gth}, or @t{a'lh}
17134 in place of @t{a'length}.
17137 @cindex quoting Ada internal identifiers
17138 Since Ada is case-insensitive, the debugger normally maps identifiers you type
17139 to lower case. The GNAT compiler uses upper-case characters for
17140 some of its internal identifiers, which are normally of no interest to users.
17141 For the rare occasions when you actually have to look at them,
17142 enclose them in angle brackets to avoid the lower-case mapping.
17145 (@value{GDBP}) print <JMPBUF_SAVE>[0]
17149 Printing an object of class-wide type or dereferencing an
17150 access-to-class-wide value will display all the components of the object's
17151 specific type (as indicated by its run-time tag). Likewise, component
17152 selection on such a value will operate on the specific type of the
17157 @node Overloading support for Ada
17158 @subsubsection Overloading support for Ada
17159 @cindex overloading, Ada
17161 The debugger supports limited overloading. Given a subprogram call in which
17162 the function symbol has multiple definitions, it will use the number of
17163 actual parameters and some information about their types to attempt to narrow
17164 the set of definitions. It also makes very limited use of context, preferring
17165 procedures to functions in the context of the @code{call} command, and
17166 functions to procedures elsewhere.
17168 If, after narrowing, the set of matching definitions still contains more than
17169 one definition, @value{GDBN} will display a menu to query which one it should
17173 (@value{GDBP}) print f(1)
17174 Multiple matches for f
17176 [1] foo.f (integer) return boolean at foo.adb:23
17177 [2] foo.f (foo.new_integer) return boolean at foo.adb:28
17181 In this case, just select one menu entry either to cancel expression evaluation
17182 (type @kbd{0} and press @key{RET}) or to continue evaluation with a specific
17183 instance (type the corresponding number and press @key{RET}).
17185 Here are a couple of commands to customize @value{GDBN}'s behavior in this
17190 @kindex set ada print-signatures
17191 @item set ada print-signatures
17192 Control whether parameter types and return types are displayed in overloads
17193 selection menus. It is @code{on} by default.
17194 @xref{Overloading support for Ada}.
17196 @kindex show ada print-signatures
17197 @item show ada print-signatures
17198 Show the current setting for displaying parameter types and return types in
17199 overloads selection menu.
17200 @xref{Overloading support for Ada}.
17204 @node Stopping Before Main Program
17205 @subsubsection Stopping at the Very Beginning
17207 @cindex breakpointing Ada elaboration code
17208 It is sometimes necessary to debug the program during elaboration, and
17209 before reaching the main procedure.
17210 As defined in the Ada Reference
17211 Manual, the elaboration code is invoked from a procedure called
17212 @code{adainit}. To run your program up to the beginning of
17213 elaboration, simply use the following two commands:
17214 @code{tbreak adainit} and @code{run}.
17216 @node Ada Exceptions
17217 @subsubsection Ada Exceptions
17219 A command is provided to list all Ada exceptions:
17222 @kindex info exceptions
17223 @item info exceptions
17224 @itemx info exceptions @var{regexp}
17225 The @code{info exceptions} command allows you to list all Ada exceptions
17226 defined within the program being debugged, as well as their addresses.
17227 With a regular expression, @var{regexp}, as argument, only those exceptions
17228 whose names match @var{regexp} are listed.
17231 Below is a small example, showing how the command can be used, first
17232 without argument, and next with a regular expression passed as an
17236 (@value{GDBP}) info exceptions
17237 All defined Ada exceptions:
17238 constraint_error: 0x613da0
17239 program_error: 0x613d20
17240 storage_error: 0x613ce0
17241 tasking_error: 0x613ca0
17242 const.aint_global_e: 0x613b00
17243 (@value{GDBP}) info exceptions const.aint
17244 All Ada exceptions matching regular expression "const.aint":
17245 constraint_error: 0x613da0
17246 const.aint_global_e: 0x613b00
17249 It is also possible to ask @value{GDBN} to stop your program's execution
17250 when an exception is raised. For more details, see @ref{Set Catchpoints}.
17253 @subsubsection Extensions for Ada Tasks
17254 @cindex Ada, tasking
17256 Support for Ada tasks is analogous to that for threads (@pxref{Threads}).
17257 @value{GDBN} provides the following task-related commands:
17262 This command shows a list of current Ada tasks, as in the following example:
17269 (@value{GDBP}) info tasks
17270 ID TID P-ID Pri State Name
17271 1 8088000 0 15 Child Activation Wait main_task
17272 2 80a4000 1 15 Accept Statement b
17273 3 809a800 1 15 Child Activation Wait a
17274 * 4 80ae800 3 15 Runnable c
17279 In this listing, the asterisk before the last task indicates it to be the
17280 task currently being inspected.
17284 Represents @value{GDBN}'s internal task number.
17290 The parent's task ID (@value{GDBN}'s internal task number).
17293 The base priority of the task.
17296 Current state of the task.
17300 The task has been created but has not been activated. It cannot be
17304 The task is not blocked for any reason known to Ada. (It may be waiting
17305 for a mutex, though.) It is conceptually "executing" in normal mode.
17308 The task is terminated, in the sense of ARM 9.3 (5). Any dependents
17309 that were waiting on terminate alternatives have been awakened and have
17310 terminated themselves.
17312 @item Child Activation Wait
17313 The task is waiting for created tasks to complete activation.
17315 @item Accept Statement
17316 The task is waiting on an accept or selective wait statement.
17318 @item Waiting on entry call
17319 The task is waiting on an entry call.
17321 @item Async Select Wait
17322 The task is waiting to start the abortable part of an asynchronous
17326 The task is waiting on a select statement with only a delay
17329 @item Child Termination Wait
17330 The task is sleeping having completed a master within itself, and is
17331 waiting for the tasks dependent on that master to become terminated or
17332 waiting on a terminate Phase.
17334 @item Wait Child in Term Alt
17335 The task is sleeping waiting for tasks on terminate alternatives to
17336 finish terminating.
17338 @item Accepting RV with @var{taskno}
17339 The task is accepting a rendez-vous with the task @var{taskno}.
17343 Name of the task in the program.
17347 @kindex info task @var{taskno}
17348 @item info task @var{taskno}
17349 This command shows detailled informations on the specified task, as in
17350 the following example:
17355 (@value{GDBP}) info tasks
17356 ID TID P-ID Pri State Name
17357 1 8077880 0 15 Child Activation Wait main_task
17358 * 2 807c468 1 15 Runnable task_1
17359 (@value{GDBP}) info task 2
17360 Ada Task: 0x807c468
17364 Parent: 1 (main_task)
17370 @kindex task@r{ (Ada)}
17371 @cindex current Ada task ID
17372 This command prints the ID of the current task.
17378 (@value{GDBP}) info tasks
17379 ID TID P-ID Pri State Name
17380 1 8077870 0 15 Child Activation Wait main_task
17381 * 2 807c458 1 15 Runnable t
17382 (@value{GDBP}) task
17383 [Current task is 2]
17386 @item task @var{taskno}
17387 @cindex Ada task switching
17388 This command is like the @code{thread @var{thread-id}}
17389 command (@pxref{Threads}). It switches the context of debugging
17390 from the current task to the given task.
17396 (@value{GDBP}) info tasks
17397 ID TID P-ID Pri State Name
17398 1 8077870 0 15 Child Activation Wait main_task
17399 * 2 807c458 1 15 Runnable t
17400 (@value{GDBP}) task 1
17401 [Switching to task 1]
17402 #0 0x8067726 in pthread_cond_wait ()
17404 #0 0x8067726 in pthread_cond_wait ()
17405 #1 0x8056714 in system.os_interface.pthread_cond_wait ()
17406 #2 0x805cb63 in system.task_primitives.operations.sleep ()
17407 #3 0x806153e in system.tasking.stages.activate_tasks ()
17408 #4 0x804aacc in un () at un.adb:5
17411 @item break @var{location} task @var{taskno}
17412 @itemx break @var{location} task @var{taskno} if @dots{}
17413 @cindex breakpoints and tasks, in Ada
17414 @cindex task breakpoints, in Ada
17415 @kindex break @dots{} task @var{taskno}@r{ (Ada)}
17416 These commands are like the @code{break @dots{} thread @dots{}}
17417 command (@pxref{Thread Stops}). The
17418 @var{location} argument specifies source lines, as described
17419 in @ref{Specify Location}.
17421 Use the qualifier @samp{task @var{taskno}} with a breakpoint command
17422 to specify that you only want @value{GDBN} to stop the program when a
17423 particular Ada task reaches this breakpoint. The @var{taskno} is one of the
17424 numeric task identifiers assigned by @value{GDBN}, shown in the first
17425 column of the @samp{info tasks} display.
17427 If you do not specify @samp{task @var{taskno}} when you set a
17428 breakpoint, the breakpoint applies to @emph{all} tasks of your
17431 You can use the @code{task} qualifier on conditional breakpoints as
17432 well; in this case, place @samp{task @var{taskno}} before the
17433 breakpoint condition (before the @code{if}).
17441 (@value{GDBP}) info tasks
17442 ID TID P-ID Pri State Name
17443 1 140022020 0 15 Child Activation Wait main_task
17444 2 140045060 1 15 Accept/Select Wait t2
17445 3 140044840 1 15 Runnable t1
17446 * 4 140056040 1 15 Runnable t3
17447 (@value{GDBP}) b 15 task 2
17448 Breakpoint 5 at 0x120044cb0: file test_task_debug.adb, line 15.
17449 (@value{GDBP}) cont
17454 Breakpoint 5, test_task_debug () at test_task_debug.adb:15
17456 (@value{GDBP}) info tasks
17457 ID TID P-ID Pri State Name
17458 1 140022020 0 15 Child Activation Wait main_task
17459 * 2 140045060 1 15 Runnable t2
17460 3 140044840 1 15 Runnable t1
17461 4 140056040 1 15 Delay Sleep t3
17465 @node Ada Tasks and Core Files
17466 @subsubsection Tasking Support when Debugging Core Files
17467 @cindex Ada tasking and core file debugging
17469 When inspecting a core file, as opposed to debugging a live program,
17470 tasking support may be limited or even unavailable, depending on
17471 the platform being used.
17472 For instance, on x86-linux, the list of tasks is available, but task
17473 switching is not supported.
17475 On certain platforms, the debugger needs to perform some
17476 memory writes in order to provide Ada tasking support. When inspecting
17477 a core file, this means that the core file must be opened with read-write
17478 privileges, using the command @samp{"set write on"} (@pxref{Patching}).
17479 Under these circumstances, you should make a backup copy of the core
17480 file before inspecting it with @value{GDBN}.
17482 @node Ravenscar Profile
17483 @subsubsection Tasking Support when using the Ravenscar Profile
17484 @cindex Ravenscar Profile
17486 The @dfn{Ravenscar Profile} is a subset of the Ada tasking features,
17487 specifically designed for systems with safety-critical real-time
17491 @kindex set ravenscar task-switching on
17492 @cindex task switching with program using Ravenscar Profile
17493 @item set ravenscar task-switching on
17494 Allows task switching when debugging a program that uses the Ravenscar
17495 Profile. This is the default.
17497 @kindex set ravenscar task-switching off
17498 @item set ravenscar task-switching off
17499 Turn off task switching when debugging a program that uses the Ravenscar
17500 Profile. This is mostly intended to disable the code that adds support
17501 for the Ravenscar Profile, in case a bug in either @value{GDBN} or in
17502 the Ravenscar runtime is preventing @value{GDBN} from working properly.
17503 To be effective, this command should be run before the program is started.
17505 @kindex show ravenscar task-switching
17506 @item show ravenscar task-switching
17507 Show whether it is possible to switch from task to task in a program
17508 using the Ravenscar Profile.
17513 @subsubsection Ada Settings
17514 @cindex Ada settings
17517 @kindex set varsize-limit
17518 @item set varsize-limit @var{size}
17519 Prevent @value{GDBN} from attempting to evaluate objects whose size
17520 is above the given limit (@var{size}) when those sizes are computed
17521 from run-time quantities. This is typically the case when the object
17522 has a variable size, such as an array whose bounds are not known at
17523 compile time for example. Setting @var{size} to @code{unlimited}
17524 removes the size limitation. By default, the limit is about 65KB.
17526 The purpose of having such a limit is to prevent @value{GDBN} from
17527 trying to grab enormous chunks of virtual memory when asked to evaluate
17528 a quantity whose bounds have been corrupted or have not yet been fully
17529 initialized. The limit applies to the results of some subexpressions
17530 as well as to complete expressions. For example, an expression denoting
17531 a simple integer component, such as @code{x.y.z}, may fail if the size of
17532 @code{x.y} is variable and exceeds @code{size}. On the other hand,
17533 @value{GDBN} is sometimes clever; the expression @code{A(i)}, where
17534 @code{A} is an array variable with non-constant size, will generally
17535 succeed regardless of the bounds on @code{A}, as long as the component
17536 size is less than @var{size}.
17538 @kindex show varsize-limit
17539 @item show varsize-limit
17540 Show the limit on types whose size is determined by run-time quantities.
17544 @subsubsection Known Peculiarities of Ada Mode
17545 @cindex Ada, problems
17547 Besides the omissions listed previously (@pxref{Omissions from Ada}),
17548 we know of several problems with and limitations of Ada mode in
17550 some of which will be fixed with planned future releases of the debugger
17551 and the GNU Ada compiler.
17555 Static constants that the compiler chooses not to materialize as objects in
17556 storage are invisible to the debugger.
17559 Named parameter associations in function argument lists are ignored (the
17560 argument lists are treated as positional).
17563 Many useful library packages are currently invisible to the debugger.
17566 Fixed-point arithmetic, conversions, input, and output is carried out using
17567 floating-point arithmetic, and may give results that only approximate those on
17571 The GNAT compiler never generates the prefix @code{Standard} for any of
17572 the standard symbols defined by the Ada language. @value{GDBN} knows about
17573 this: it will strip the prefix from names when you use it, and will never
17574 look for a name you have so qualified among local symbols, nor match against
17575 symbols in other packages or subprograms. If you have
17576 defined entities anywhere in your program other than parameters and
17577 local variables whose simple names match names in @code{Standard},
17578 GNAT's lack of qualification here can cause confusion. When this happens,
17579 you can usually resolve the confusion
17580 by qualifying the problematic names with package
17581 @code{Standard} explicitly.
17584 Older versions of the compiler sometimes generate erroneous debugging
17585 information, resulting in the debugger incorrectly printing the value
17586 of affected entities. In some cases, the debugger is able to work
17587 around an issue automatically. In other cases, the debugger is able
17588 to work around the issue, but the work-around has to be specifically
17591 @kindex set ada trust-PAD-over-XVS
17592 @kindex show ada trust-PAD-over-XVS
17595 @item set ada trust-PAD-over-XVS on
17596 Configure GDB to strictly follow the GNAT encoding when computing the
17597 value of Ada entities, particularly when @code{PAD} and @code{PAD___XVS}
17598 types are involved (see @code{ada/exp_dbug.ads} in the GCC sources for
17599 a complete description of the encoding used by the GNAT compiler).
17600 This is the default.
17602 @item set ada trust-PAD-over-XVS off
17603 This is related to the encoding using by the GNAT compiler. If @value{GDBN}
17604 sometimes prints the wrong value for certain entities, changing @code{ada
17605 trust-PAD-over-XVS} to @code{off} activates a work-around which may fix
17606 the issue. It is always safe to set @code{ada trust-PAD-over-XVS} to
17607 @code{off}, but this incurs a slight performance penalty, so it is
17608 recommended to leave this setting to @code{on} unless necessary.
17612 @cindex GNAT descriptive types
17613 @cindex GNAT encoding
17614 Internally, the debugger also relies on the compiler following a number
17615 of conventions known as the @samp{GNAT Encoding}, all documented in
17616 @file{gcc/ada/exp_dbug.ads} in the GCC sources. This encoding describes
17617 how the debugging information should be generated for certain types.
17618 In particular, this convention makes use of @dfn{descriptive types},
17619 which are artificial types generated purely to help the debugger.
17621 These encodings were defined at a time when the debugging information
17622 format used was not powerful enough to describe some of the more complex
17623 types available in Ada. Since DWARF allows us to express nearly all
17624 Ada features, the long-term goal is to slowly replace these descriptive
17625 types by their pure DWARF equivalent. To facilitate that transition,
17626 a new maintenance option is available to force the debugger to ignore
17627 those descriptive types. It allows the user to quickly evaluate how
17628 well @value{GDBN} works without them.
17632 @kindex maint ada set ignore-descriptive-types
17633 @item maintenance ada set ignore-descriptive-types [on|off]
17634 Control whether the debugger should ignore descriptive types.
17635 The default is not to ignore descriptives types (@code{off}).
17637 @kindex maint ada show ignore-descriptive-types
17638 @item maintenance ada show ignore-descriptive-types
17639 Show if descriptive types are ignored by @value{GDBN}.
17643 @node Unsupported Languages
17644 @section Unsupported Languages
17646 @cindex unsupported languages
17647 @cindex minimal language
17648 In addition to the other fully-supported programming languages,
17649 @value{GDBN} also provides a pseudo-language, called @code{minimal}.
17650 It does not represent a real programming language, but provides a set
17651 of capabilities close to what the C or assembly languages provide.
17652 This should allow most simple operations to be performed while debugging
17653 an application that uses a language currently not supported by @value{GDBN}.
17655 If the language is set to @code{auto}, @value{GDBN} will automatically
17656 select this language if the current frame corresponds to an unsupported
17660 @chapter Examining the Symbol Table
17662 The commands described in this chapter allow you to inquire about the
17663 symbols (names of variables, functions and types) defined in your
17664 program. This information is inherent in the text of your program and
17665 does not change as your program executes. @value{GDBN} finds it in your
17666 program's symbol table, in the file indicated when you started @value{GDBN}
17667 (@pxref{File Options, ,Choosing Files}), or by one of the
17668 file-management commands (@pxref{Files, ,Commands to Specify Files}).
17670 @cindex symbol names
17671 @cindex names of symbols
17672 @cindex quoting names
17673 @anchor{quoting names}
17674 Occasionally, you may need to refer to symbols that contain unusual
17675 characters, which @value{GDBN} ordinarily treats as word delimiters. The
17676 most frequent case is in referring to static variables in other
17677 source files (@pxref{Variables,,Program Variables}). File names
17678 are recorded in object files as debugging symbols, but @value{GDBN} would
17679 ordinarily parse a typical file name, like @file{foo.c}, as the three words
17680 @samp{foo} @samp{.} @samp{c}. To allow @value{GDBN} to recognize
17681 @samp{foo.c} as a single symbol, enclose it in single quotes; for example,
17688 looks up the value of @code{x} in the scope of the file @file{foo.c}.
17691 @cindex case-insensitive symbol names
17692 @cindex case sensitivity in symbol names
17693 @kindex set case-sensitive
17694 @item set case-sensitive on
17695 @itemx set case-sensitive off
17696 @itemx set case-sensitive auto
17697 Normally, when @value{GDBN} looks up symbols, it matches their names
17698 with case sensitivity determined by the current source language.
17699 Occasionally, you may wish to control that. The command @code{set
17700 case-sensitive} lets you do that by specifying @code{on} for
17701 case-sensitive matches or @code{off} for case-insensitive ones. If
17702 you specify @code{auto}, case sensitivity is reset to the default
17703 suitable for the source language. The default is case-sensitive
17704 matches for all languages except for Fortran, for which the default is
17705 case-insensitive matches.
17707 @kindex show case-sensitive
17708 @item show case-sensitive
17709 This command shows the current setting of case sensitivity for symbols
17712 @kindex set print type methods
17713 @item set print type methods
17714 @itemx set print type methods on
17715 @itemx set print type methods off
17716 Normally, when @value{GDBN} prints a class, it displays any methods
17717 declared in that class. You can control this behavior either by
17718 passing the appropriate flag to @code{ptype}, or using @command{set
17719 print type methods}. Specifying @code{on} will cause @value{GDBN} to
17720 display the methods; this is the default. Specifying @code{off} will
17721 cause @value{GDBN} to omit the methods.
17723 @kindex show print type methods
17724 @item show print type methods
17725 This command shows the current setting of method display when printing
17728 @kindex set print type nested-type-limit
17729 @item set print type nested-type-limit @var{limit}
17730 @itemx set print type nested-type-limit unlimited
17731 Set the limit of displayed nested types that the type printer will
17732 show. A @var{limit} of @code{unlimited} or @code{-1} will show all
17733 nested definitions. By default, the type printer will not show any nested
17734 types defined in classes.
17736 @kindex show print type nested-type-limit
17737 @item show print type nested-type-limit
17738 This command shows the current display limit of nested types when
17741 @kindex set print type typedefs
17742 @item set print type typedefs
17743 @itemx set print type typedefs on
17744 @itemx set print type typedefs off
17746 Normally, when @value{GDBN} prints a class, it displays any typedefs
17747 defined in that class. You can control this behavior either by
17748 passing the appropriate flag to @code{ptype}, or using @command{set
17749 print type typedefs}. Specifying @code{on} will cause @value{GDBN} to
17750 display the typedef definitions; this is the default. Specifying
17751 @code{off} will cause @value{GDBN} to omit the typedef definitions.
17752 Note that this controls whether the typedef definition itself is
17753 printed, not whether typedef names are substituted when printing other
17756 @kindex show print type typedefs
17757 @item show print type typedefs
17758 This command shows the current setting of typedef display when
17761 @kindex info address
17762 @cindex address of a symbol
17763 @item info address @var{symbol}
17764 Describe where the data for @var{symbol} is stored. For a register
17765 variable, this says which register it is kept in. For a non-register
17766 local variable, this prints the stack-frame offset at which the variable
17769 Note the contrast with @samp{print &@var{symbol}}, which does not work
17770 at all for a register variable, and for a stack local variable prints
17771 the exact address of the current instantiation of the variable.
17773 @kindex info symbol
17774 @cindex symbol from address
17775 @cindex closest symbol and offset for an address
17776 @item info symbol @var{addr}
17777 Print the name of a symbol which is stored at the address @var{addr}.
17778 If no symbol is stored exactly at @var{addr}, @value{GDBN} prints the
17779 nearest symbol and an offset from it:
17782 (@value{GDBP}) info symbol 0x54320
17783 _initialize_vx + 396 in section .text
17787 This is the opposite of the @code{info address} command. You can use
17788 it to find out the name of a variable or a function given its address.
17790 For dynamically linked executables, the name of executable or shared
17791 library containing the symbol is also printed:
17794 (@value{GDBP}) info symbol 0x400225
17795 _start + 5 in section .text of /tmp/a.out
17796 (@value{GDBP}) info symbol 0x2aaaac2811cf
17797 __read_nocancel + 6 in section .text of /usr/lib64/libc.so.6
17802 @item demangle @r{[}-l @var{language}@r{]} @r{[}@var{--}@r{]} @var{name}
17803 Demangle @var{name}.
17804 If @var{language} is provided it is the name of the language to demangle
17805 @var{name} in. Otherwise @var{name} is demangled in the current language.
17807 The @samp{--} option specifies the end of options,
17808 and is useful when @var{name} begins with a dash.
17810 The parameter @code{demangle-style} specifies how to interpret the kind
17811 of mangling used. @xref{Print Settings}.
17814 @item whatis[/@var{flags}] [@var{arg}]
17815 Print the data type of @var{arg}, which can be either an expression
17816 or a name of a data type. With no argument, print the data type of
17817 @code{$}, the last value in the value history.
17819 If @var{arg} is an expression (@pxref{Expressions, ,Expressions}), it
17820 is not actually evaluated, and any side-effecting operations (such as
17821 assignments or function calls) inside it do not take place.
17823 If @var{arg} is a variable or an expression, @code{whatis} prints its
17824 literal type as it is used in the source code. If the type was
17825 defined using a @code{typedef}, @code{whatis} will @emph{not} print
17826 the data type underlying the @code{typedef}. If the type of the
17827 variable or the expression is a compound data type, such as
17828 @code{struct} or @code{class}, @code{whatis} never prints their
17829 fields or methods. It just prints the @code{struct}/@code{class}
17830 name (a.k.a.@: its @dfn{tag}). If you want to see the members of
17831 such a compound data type, use @code{ptype}.
17833 If @var{arg} is a type name that was defined using @code{typedef},
17834 @code{whatis} @dfn{unrolls} only one level of that @code{typedef}.
17835 Unrolling means that @code{whatis} will show the underlying type used
17836 in the @code{typedef} declaration of @var{arg}. However, if that
17837 underlying type is also a @code{typedef}, @code{whatis} will not
17840 For C code, the type names may also have the form @samp{class
17841 @var{class-name}}, @samp{struct @var{struct-tag}}, @samp{union
17842 @var{union-tag}} or @samp{enum @var{enum-tag}}.
17844 @var{flags} can be used to modify how the type is displayed.
17845 Available flags are:
17849 Display in ``raw'' form. Normally, @value{GDBN} substitutes template
17850 parameters and typedefs defined in a class when printing the class'
17851 members. The @code{/r} flag disables this.
17854 Do not print methods defined in the class.
17857 Print methods defined in the class. This is the default, but the flag
17858 exists in case you change the default with @command{set print type methods}.
17861 Do not print typedefs defined in the class. Note that this controls
17862 whether the typedef definition itself is printed, not whether typedef
17863 names are substituted when printing other types.
17866 Print typedefs defined in the class. This is the default, but the flag
17867 exists in case you change the default with @command{set print type typedefs}.
17870 Print the offsets and sizes of fields in a struct, similar to what the
17871 @command{pahole} tool does. This option implies the @code{/tm} flags.
17873 For example, given the following declarations:
17909 Issuing a @kbd{ptype /o struct tuv} command would print:
17912 (@value{GDBP}) ptype /o struct tuv
17913 /* offset | size */ type = struct tuv @{
17914 /* 0 | 4 */ int a1;
17915 /* XXX 4-byte hole */
17916 /* 8 | 8 */ char *a2;
17917 /* 16 | 4 */ int a3;
17919 /* total size (bytes): 24 */
17923 Notice the format of the first column of comments. There, you can
17924 find two parts separated by the @samp{|} character: the @emph{offset},
17925 which indicates where the field is located inside the struct, in
17926 bytes, and the @emph{size} of the field. Another interesting line is
17927 the marker of a @emph{hole} in the struct, indicating that it may be
17928 possible to pack the struct and make it use less space by reorganizing
17931 It is also possible to print offsets inside an union:
17934 (@value{GDBP}) ptype /o union qwe
17935 /* offset | size */ type = union qwe @{
17936 /* 24 */ struct tuv @{
17937 /* 0 | 4 */ int a1;
17938 /* XXX 4-byte hole */
17939 /* 8 | 8 */ char *a2;
17940 /* 16 | 4 */ int a3;
17942 /* total size (bytes): 24 */
17944 /* 40 */ struct xyz @{
17945 /* 0 | 4 */ int f1;
17946 /* 4 | 1 */ char f2;
17947 /* XXX 3-byte hole */
17948 /* 8 | 8 */ void *f3;
17949 /* 16 | 24 */ struct tuv @{
17950 /* 16 | 4 */ int a1;
17951 /* XXX 4-byte hole */
17952 /* 24 | 8 */ char *a2;
17953 /* 32 | 4 */ int a3;
17955 /* total size (bytes): 24 */
17958 /* total size (bytes): 40 */
17961 /* total size (bytes): 40 */
17965 In this case, since @code{struct tuv} and @code{struct xyz} occupy the
17966 same space (because we are dealing with an union), the offset is not
17967 printed for them. However, you can still examine the offset of each
17968 of these structures' fields.
17970 Another useful scenario is printing the offsets of a struct containing
17974 (@value{GDBP}) ptype /o struct tyu
17975 /* offset | size */ type = struct tyu @{
17976 /* 0:31 | 4 */ int a1 : 1;
17977 /* 0:28 | 4 */ int a2 : 3;
17978 /* 0: 5 | 4 */ int a3 : 23;
17979 /* 3: 3 | 1 */ signed char a4 : 2;
17980 /* XXX 3-bit hole */
17981 /* XXX 4-byte hole */
17982 /* 8 | 8 */ int64_t a5;
17983 /* 16: 0 | 4 */ int a6 : 5;
17984 /* 16: 5 | 8 */ int64_t a7 : 3;
17985 "/* XXX 7-byte padding */
17987 /* total size (bytes): 24 */
17991 Note how the offset information is now extended to also include the
17992 first bit of the bitfield.
17996 @item ptype[/@var{flags}] [@var{arg}]
17997 @code{ptype} accepts the same arguments as @code{whatis}, but prints a
17998 detailed description of the type, instead of just the name of the type.
17999 @xref{Expressions, ,Expressions}.
18001 Contrary to @code{whatis}, @code{ptype} always unrolls any
18002 @code{typedef}s in its argument declaration, whether the argument is
18003 a variable, expression, or a data type. This means that @code{ptype}
18004 of a variable or an expression will not print literally its type as
18005 present in the source code---use @code{whatis} for that. @code{typedef}s at
18006 the pointer or reference targets are also unrolled. Only @code{typedef}s of
18007 fields, methods and inner @code{class typedef}s of @code{struct}s,
18008 @code{class}es and @code{union}s are not unrolled even with @code{ptype}.
18010 For example, for this variable declaration:
18013 typedef double real_t;
18014 struct complex @{ real_t real; double imag; @};
18015 typedef struct complex complex_t;
18017 real_t *real_pointer_var;
18021 the two commands give this output:
18025 (@value{GDBP}) whatis var
18027 (@value{GDBP}) ptype var
18028 type = struct complex @{
18032 (@value{GDBP}) whatis complex_t
18033 type = struct complex
18034 (@value{GDBP}) whatis struct complex
18035 type = struct complex
18036 (@value{GDBP}) ptype struct complex
18037 type = struct complex @{
18041 (@value{GDBP}) whatis real_pointer_var
18043 (@value{GDBP}) ptype real_pointer_var
18049 As with @code{whatis}, using @code{ptype} without an argument refers to
18050 the type of @code{$}, the last value in the value history.
18052 @cindex incomplete type
18053 Sometimes, programs use opaque data types or incomplete specifications
18054 of complex data structure. If the debug information included in the
18055 program does not allow @value{GDBN} to display a full declaration of
18056 the data type, it will say @samp{<incomplete type>}. For example,
18057 given these declarations:
18061 struct foo *fooptr;
18065 but no definition for @code{struct foo} itself, @value{GDBN} will say:
18068 (@value{GDBP}) ptype foo
18069 $1 = <incomplete type>
18073 ``Incomplete type'' is C terminology for data types that are not
18074 completely specified.
18076 @cindex unknown type
18077 Othertimes, information about a variable's type is completely absent
18078 from the debug information included in the program. This most often
18079 happens when the program or library where the variable is defined
18080 includes no debug information at all. @value{GDBN} knows the variable
18081 exists from inspecting the linker/loader symbol table (e.g., the ELF
18082 dynamic symbol table), but such symbols do not contain type
18083 information. Inspecting the type of a (global) variable for which
18084 @value{GDBN} has no type information shows:
18087 (@value{GDBP}) ptype var
18088 type = <data variable, no debug info>
18091 @xref{Variables, no debug info variables}, for how to print the values
18095 @item info types @var{regexp}
18097 Print a brief description of all types whose names match the regular
18098 expression @var{regexp} (or all types in your program, if you supply
18099 no argument). Each complete typename is matched as though it were a
18100 complete line; thus, @samp{i type value} gives information on all
18101 types in your program whose names include the string @code{value}, but
18102 @samp{i type ^value$} gives information only on types whose complete
18103 name is @code{value}.
18105 In programs using different languages, @value{GDBN} chooses the syntax
18106 to print the type description according to the
18107 @samp{set language} value: using @samp{set language auto}
18108 (see @ref{Automatically, ,Set Language Automatically}) means to use the
18109 language of the type, other values mean to use
18110 the manually specified language (see @ref{Manually, ,Set Language Manually}).
18112 This command differs from @code{ptype} in two ways: first, like
18113 @code{whatis}, it does not print a detailed description; second, it
18114 lists all source files and line numbers where a type is defined.
18116 @kindex info type-printers
18117 @item info type-printers
18118 Versions of @value{GDBN} that ship with Python scripting enabled may
18119 have ``type printers'' available. When using @command{ptype} or
18120 @command{whatis}, these printers are consulted when the name of a type
18121 is needed. @xref{Type Printing API}, for more information on writing
18124 @code{info type-printers} displays all the available type printers.
18126 @kindex enable type-printer
18127 @kindex disable type-printer
18128 @item enable type-printer @var{name}@dots{}
18129 @item disable type-printer @var{name}@dots{}
18130 These commands can be used to enable or disable type printers.
18133 @cindex local variables
18134 @item info scope @var{location}
18135 List all the variables local to a particular scope. This command
18136 accepts a @var{location} argument---a function name, a source line, or
18137 an address preceded by a @samp{*}, and prints all the variables local
18138 to the scope defined by that location. (@xref{Specify Location}, for
18139 details about supported forms of @var{location}.) For example:
18142 (@value{GDBP}) @b{info scope command_line_handler}
18143 Scope for command_line_handler:
18144 Symbol rl is an argument at stack/frame offset 8, length 4.
18145 Symbol linebuffer is in static storage at address 0x150a18, length 4.
18146 Symbol linelength is in static storage at address 0x150a1c, length 4.
18147 Symbol p is a local variable in register $esi, length 4.
18148 Symbol p1 is a local variable in register $ebx, length 4.
18149 Symbol nline is a local variable in register $edx, length 4.
18150 Symbol repeat is a local variable at frame offset -8, length 4.
18154 This command is especially useful for determining what data to collect
18155 during a @dfn{trace experiment}, see @ref{Tracepoint Actions,
18158 @kindex info source
18160 Show information about the current source file---that is, the source file for
18161 the function containing the current point of execution:
18164 the name of the source file, and the directory containing it,
18166 the directory it was compiled in,
18168 its length, in lines,
18170 which programming language it is written in,
18172 if the debug information provides it, the program that compiled the file
18173 (which may include, e.g., the compiler version and command line arguments),
18175 whether the executable includes debugging information for that file, and
18176 if so, what format the information is in (e.g., STABS, Dwarf 2, etc.), and
18178 whether the debugging information includes information about
18179 preprocessor macros.
18183 @kindex info sources
18185 Print the names of all source files in your program for which there is
18186 debugging information, organized into two lists: files whose symbols
18187 have already been read, and files whose symbols will be read when needed.
18189 @kindex info functions
18190 @item info functions [-q]
18191 Print the names and data types of all defined functions.
18192 Similarly to @samp{info types}, this command groups its output by source
18193 files and annotates each function definition with its source line
18196 In programs using different languages, @value{GDBN} chooses the syntax
18197 to print the function name and type according to the
18198 @samp{set language} value: using @samp{set language auto}
18199 (see @ref{Automatically, ,Set Language Automatically}) means to use the
18200 language of the function, other values mean to use
18201 the manually specified language (see @ref{Manually, ,Set Language Manually}).
18203 The optional flag @samp{-q}, which stands for @samp{quiet}, disables
18204 printing header information and messages explaining why no functions
18207 @item info functions [-q] [-t @var{type_regexp}] [@var{regexp}]
18208 Like @samp{info functions}, but only print the names and data types
18209 of the functions selected with the provided regexp(s).
18211 If @var{regexp} is provided, print only the functions whose names
18212 match the regular expression @var{regexp}.
18213 Thus, @samp{info fun step} finds all functions whose
18214 names include @code{step}; @samp{info fun ^step} finds those whose names
18215 start with @code{step}. If a function name contains characters that
18216 conflict with the regular expression language (e.g.@:
18217 @samp{operator*()}), they may be quoted with a backslash.
18219 If @var{type_regexp} is provided, print only the functions whose
18220 types, as printed by the @code{whatis} command, match
18221 the regular expression @var{type_regexp}.
18222 If @var{type_regexp} contains space(s), it should be enclosed in
18223 quote characters. If needed, use backslash to escape the meaning
18224 of special characters or quotes.
18225 Thus, @samp{info fun -t '^int ('} finds the functions that return
18226 an integer; @samp{info fun -t '(.*int.*'} finds the functions that
18227 have an argument type containing int; @samp{info fun -t '^int (' ^step}
18228 finds the functions whose names start with @code{step} and that return
18231 If both @var{regexp} and @var{type_regexp} are provided, a function
18232 is printed only if its name matches @var{regexp} and its type matches
18236 @kindex info variables
18237 @item info variables [-q]
18238 Print the names and data types of all variables that are defined
18239 outside of functions (i.e.@: excluding local variables).
18240 The printed variables are grouped by source files and annotated with
18241 their respective source line numbers.
18243 In programs using different languages, @value{GDBN} chooses the syntax
18244 to print the variable name and type according to the
18245 @samp{set language} value: using @samp{set language auto}
18246 (see @ref{Automatically, ,Set Language Automatically}) means to use the
18247 language of the variable, other values mean to use
18248 the manually specified language (see @ref{Manually, ,Set Language Manually}).
18250 The optional flag @samp{-q}, which stands for @samp{quiet}, disables
18251 printing header information and messages explaining why no variables
18254 @item info variables [-q] [-t @var{type_regexp}] [@var{regexp}]
18255 Like @kbd{info variables}, but only print the variables selected
18256 with the provided regexp(s).
18258 If @var{regexp} is provided, print only the variables whose names
18259 match the regular expression @var{regexp}.
18261 If @var{type_regexp} is provided, print only the variables whose
18262 types, as printed by the @code{whatis} command, match
18263 the regular expression @var{type_regexp}.
18264 If @var{type_regexp} contains space(s), it should be enclosed in
18265 quote characters. If needed, use backslash to escape the meaning
18266 of special characters or quotes.
18268 If both @var{regexp} and @var{type_regexp} are provided, an argument
18269 is printed only if its name matches @var{regexp} and its type matches
18272 @kindex info classes
18273 @cindex Objective-C, classes and selectors
18275 @itemx info classes @var{regexp}
18276 Display all Objective-C classes in your program, or
18277 (with the @var{regexp} argument) all those matching a particular regular
18280 @kindex info selectors
18281 @item info selectors
18282 @itemx info selectors @var{regexp}
18283 Display all Objective-C selectors in your program, or
18284 (with the @var{regexp} argument) all those matching a particular regular
18288 This was never implemented.
18289 @kindex info methods
18291 @itemx info methods @var{regexp}
18292 The @code{info methods} command permits the user to examine all defined
18293 methods within C@t{++} program, or (with the @var{regexp} argument) a
18294 specific set of methods found in the various C@t{++} classes. Many
18295 C@t{++} classes provide a large number of methods. Thus, the output
18296 from the @code{ptype} command can be overwhelming and hard to use. The
18297 @code{info-methods} command filters the methods, printing only those
18298 which match the regular-expression @var{regexp}.
18301 @cindex opaque data types
18302 @kindex set opaque-type-resolution
18303 @item set opaque-type-resolution on
18304 Tell @value{GDBN} to resolve opaque types. An opaque type is a type
18305 declared as a pointer to a @code{struct}, @code{class}, or
18306 @code{union}---for example, @code{struct MyType *}---that is used in one
18307 source file although the full declaration of @code{struct MyType} is in
18308 another source file. The default is on.
18310 A change in the setting of this subcommand will not take effect until
18311 the next time symbols for a file are loaded.
18313 @item set opaque-type-resolution off
18314 Tell @value{GDBN} not to resolve opaque types. In this case, the type
18315 is printed as follows:
18317 @{<no data fields>@}
18320 @kindex show opaque-type-resolution
18321 @item show opaque-type-resolution
18322 Show whether opaque types are resolved or not.
18324 @kindex set print symbol-loading
18325 @cindex print messages when symbols are loaded
18326 @item set print symbol-loading
18327 @itemx set print symbol-loading full
18328 @itemx set print symbol-loading brief
18329 @itemx set print symbol-loading off
18330 The @code{set print symbol-loading} command allows you to control the
18331 printing of messages when @value{GDBN} loads symbol information.
18332 By default a message is printed for the executable and one for each
18333 shared library, and normally this is what you want. However, when
18334 debugging apps with large numbers of shared libraries these messages
18336 When set to @code{brief} a message is printed for each executable,
18337 and when @value{GDBN} loads a collection of shared libraries at once
18338 it will only print one message regardless of the number of shared
18339 libraries. When set to @code{off} no messages are printed.
18341 @kindex show print symbol-loading
18342 @item show print symbol-loading
18343 Show whether messages will be printed when a @value{GDBN} command
18344 entered from the keyboard causes symbol information to be loaded.
18346 @kindex maint print symbols
18347 @cindex symbol dump
18348 @kindex maint print psymbols
18349 @cindex partial symbol dump
18350 @kindex maint print msymbols
18351 @cindex minimal symbol dump
18352 @item maint print symbols @r{[}-pc @var{address}@r{]} @r{[}@var{filename}@r{]}
18353 @itemx maint print symbols @r{[}-objfile @var{objfile}@r{]} @r{[}-source @var{source}@r{]} @r{[}--@r{]} @r{[}@var{filename}@r{]}
18354 @itemx maint print psymbols @r{[}-objfile @var{objfile}@r{]} @r{[}-pc @var{address}@r{]} @r{[}--@r{]} @r{[}@var{filename}@r{]}
18355 @itemx maint print psymbols @r{[}-objfile @var{objfile}@r{]} @r{[}-source @var{source}@r{]} @r{[}--@r{]} @r{[}@var{filename}@r{]}
18356 @itemx maint print msymbols @r{[}-objfile @var{objfile}@r{]} @r{[}--@r{]} @r{[}@var{filename}@r{]}
18357 Write a dump of debugging symbol data into the file @var{filename} or
18358 the terminal if @var{filename} is unspecified.
18359 If @code{-objfile @var{objfile}} is specified, only dump symbols for
18361 If @code{-pc @var{address}} is specified, only dump symbols for the file
18362 with code at that address. Note that @var{address} may be a symbol like
18364 If @code{-source @var{source}} is specified, only dump symbols for that
18367 These commands are used to debug the @value{GDBN} symbol-reading code.
18368 These commands do not modify internal @value{GDBN} state, therefore
18369 @samp{maint print symbols} will only print symbols for already expanded symbol
18371 You can use the command @code{info sources} to find out which files these are.
18372 If you use @samp{maint print psymbols} instead, the dump shows information
18373 about symbols that @value{GDBN} only knows partially---that is, symbols
18374 defined in files that @value{GDBN} has skimmed, but not yet read completely.
18375 Finally, @samp{maint print msymbols} just dumps ``minimal symbols'', e.g.,
18378 @xref{Files, ,Commands to Specify Files}, for a discussion of how
18379 @value{GDBN} reads symbols (in the description of @code{symbol-file}).
18381 @kindex maint info symtabs
18382 @kindex maint info psymtabs
18383 @cindex listing @value{GDBN}'s internal symbol tables
18384 @cindex symbol tables, listing @value{GDBN}'s internal
18385 @cindex full symbol tables, listing @value{GDBN}'s internal
18386 @cindex partial symbol tables, listing @value{GDBN}'s internal
18387 @item maint info symtabs @r{[} @var{regexp} @r{]}
18388 @itemx maint info psymtabs @r{[} @var{regexp} @r{]}
18390 List the @code{struct symtab} or @code{struct partial_symtab}
18391 structures whose names match @var{regexp}. If @var{regexp} is not
18392 given, list them all. The output includes expressions which you can
18393 copy into a @value{GDBN} debugging this one to examine a particular
18394 structure in more detail. For example:
18397 (@value{GDBP}) maint info psymtabs dwarf2read
18398 @{ objfile /home/gnu/build/gdb/gdb
18399 ((struct objfile *) 0x82e69d0)
18400 @{ psymtab /home/gnu/src/gdb/dwarf2read.c
18401 ((struct partial_symtab *) 0x8474b10)
18404 text addresses 0x814d3c8 -- 0x8158074
18405 globals (* (struct partial_symbol **) 0x8507a08 @@ 9)
18406 statics (* (struct partial_symbol **) 0x40e95b78 @@ 2882)
18407 dependencies (none)
18410 (@value{GDBP}) maint info symtabs
18414 We see that there is one partial symbol table whose filename contains
18415 the string @samp{dwarf2read}, belonging to the @samp{gdb} executable;
18416 and we see that @value{GDBN} has not read in any symtabs yet at all.
18417 If we set a breakpoint on a function, that will cause @value{GDBN} to
18418 read the symtab for the compilation unit containing that function:
18421 (@value{GDBP}) break dwarf2_psymtab_to_symtab
18422 Breakpoint 1 at 0x814e5da: file /home/gnu/src/gdb/dwarf2read.c,
18424 (@value{GDBP}) maint info symtabs
18425 @{ objfile /home/gnu/build/gdb/gdb
18426 ((struct objfile *) 0x82e69d0)
18427 @{ symtab /home/gnu/src/gdb/dwarf2read.c
18428 ((struct symtab *) 0x86c1f38)
18431 blockvector ((struct blockvector *) 0x86c1bd0) (primary)
18432 linetable ((struct linetable *) 0x8370fa0)
18433 debugformat DWARF 2
18439 @kindex maint info line-table
18440 @cindex listing @value{GDBN}'s internal line tables
18441 @cindex line tables, listing @value{GDBN}'s internal
18442 @item maint info line-table @r{[} @var{regexp} @r{]}
18444 List the @code{struct linetable} from all @code{struct symtab}
18445 instances whose name matches @var{regexp}. If @var{regexp} is not
18446 given, list the @code{struct linetable} from all @code{struct symtab}.
18448 @kindex maint set symbol-cache-size
18449 @cindex symbol cache size
18450 @item maint set symbol-cache-size @var{size}
18451 Set the size of the symbol cache to @var{size}.
18452 The default size is intended to be good enough for debugging
18453 most applications. This option exists to allow for experimenting
18454 with different sizes.
18456 @kindex maint show symbol-cache-size
18457 @item maint show symbol-cache-size
18458 Show the size of the symbol cache.
18460 @kindex maint print symbol-cache
18461 @cindex symbol cache, printing its contents
18462 @item maint print symbol-cache
18463 Print the contents of the symbol cache.
18464 This is useful when debugging symbol cache issues.
18466 @kindex maint print symbol-cache-statistics
18467 @cindex symbol cache, printing usage statistics
18468 @item maint print symbol-cache-statistics
18469 Print symbol cache usage statistics.
18470 This helps determine how well the cache is being utilized.
18472 @kindex maint flush-symbol-cache
18473 @cindex symbol cache, flushing
18474 @item maint flush-symbol-cache
18475 Flush the contents of the symbol cache, all entries are removed.
18476 This command is useful when debugging the symbol cache.
18477 It is also useful when collecting performance data.
18482 @chapter Altering Execution
18484 Once you think you have found an error in your program, you might want to
18485 find out for certain whether correcting the apparent error would lead to
18486 correct results in the rest of the run. You can find the answer by
18487 experiment, using the @value{GDBN} features for altering execution of the
18490 For example, you can store new values into variables or memory
18491 locations, give your program a signal, restart it at a different
18492 address, or even return prematurely from a function.
18495 * Assignment:: Assignment to variables
18496 * Jumping:: Continuing at a different address
18497 * Signaling:: Giving your program a signal
18498 * Returning:: Returning from a function
18499 * Calling:: Calling your program's functions
18500 * Patching:: Patching your program
18501 * Compiling and Injecting Code:: Compiling and injecting code in @value{GDBN}
18505 @section Assignment to Variables
18508 @cindex setting variables
18509 To alter the value of a variable, evaluate an assignment expression.
18510 @xref{Expressions, ,Expressions}. For example,
18517 stores the value 4 into the variable @code{x}, and then prints the
18518 value of the assignment expression (which is 4).
18519 @xref{Languages, ,Using @value{GDBN} with Different Languages}, for more
18520 information on operators in supported languages.
18522 @kindex set variable
18523 @cindex variables, setting
18524 If you are not interested in seeing the value of the assignment, use the
18525 @code{set} command instead of the @code{print} command. @code{set} is
18526 really the same as @code{print} except that the expression's value is
18527 not printed and is not put in the value history (@pxref{Value History,
18528 ,Value History}). The expression is evaluated only for its effects.
18530 If the beginning of the argument string of the @code{set} command
18531 appears identical to a @code{set} subcommand, use the @code{set
18532 variable} command instead of just @code{set}. This command is identical
18533 to @code{set} except for its lack of subcommands. For example, if your
18534 program has a variable @code{width}, you get an error if you try to set
18535 a new value with just @samp{set width=13}, because @value{GDBN} has the
18536 command @code{set width}:
18539 (@value{GDBP}) whatis width
18541 (@value{GDBP}) p width
18543 (@value{GDBP}) set width=47
18544 Invalid syntax in expression.
18548 The invalid expression, of course, is @samp{=47}. In
18549 order to actually set the program's variable @code{width}, use
18552 (@value{GDBP}) set var width=47
18555 Because the @code{set} command has many subcommands that can conflict
18556 with the names of program variables, it is a good idea to use the
18557 @code{set variable} command instead of just @code{set}. For example, if
18558 your program has a variable @code{g}, you run into problems if you try
18559 to set a new value with just @samp{set g=4}, because @value{GDBN} has
18560 the command @code{set gnutarget}, abbreviated @code{set g}:
18564 (@value{GDBP}) whatis g
18568 (@value{GDBP}) set g=4
18572 The program being debugged has been started already.
18573 Start it from the beginning? (y or n) y
18574 Starting program: /home/smith/cc_progs/a.out
18575 "/home/smith/cc_progs/a.out": can't open to read symbols:
18576 Invalid bfd target.
18577 (@value{GDBP}) show g
18578 The current BFD target is "=4".
18583 The program variable @code{g} did not change, and you silently set the
18584 @code{gnutarget} to an invalid value. In order to set the variable
18588 (@value{GDBP}) set var g=4
18591 @value{GDBN} allows more implicit conversions in assignments than C; you can
18592 freely store an integer value into a pointer variable or vice versa,
18593 and you can convert any structure to any other structure that is the
18594 same length or shorter.
18595 @comment FIXME: how do structs align/pad in these conversions?
18596 @comment /doc@cygnus.com 18dec1990
18598 To store values into arbitrary places in memory, use the @samp{@{@dots{}@}}
18599 construct to generate a value of specified type at a specified address
18600 (@pxref{Expressions, ,Expressions}). For example, @code{@{int@}0x83040} refers
18601 to memory location @code{0x83040} as an integer (which implies a certain size
18602 and representation in memory), and
18605 set @{int@}0x83040 = 4
18609 stores the value 4 into that memory location.
18612 @section Continuing at a Different Address
18614 Ordinarily, when you continue your program, you do so at the place where
18615 it stopped, with the @code{continue} command. You can instead continue at
18616 an address of your own choosing, with the following commands:
18620 @kindex j @r{(@code{jump})}
18621 @item jump @var{location}
18622 @itemx j @var{location}
18623 Resume execution at @var{location}. Execution stops again immediately
18624 if there is a breakpoint there. @xref{Specify Location}, for a description
18625 of the different forms of @var{location}. It is common
18626 practice to use the @code{tbreak} command in conjunction with
18627 @code{jump}. @xref{Set Breaks, ,Setting Breakpoints}.
18629 The @code{jump} command does not change the current stack frame, or
18630 the stack pointer, or the contents of any memory location or any
18631 register other than the program counter. If @var{location} is in
18632 a different function from the one currently executing, the results may
18633 be bizarre if the two functions expect different patterns of arguments or
18634 of local variables. For this reason, the @code{jump} command requests
18635 confirmation if the specified line is not in the function currently
18636 executing. However, even bizarre results are predictable if you are
18637 well acquainted with the machine-language code of your program.
18640 On many systems, you can get much the same effect as the @code{jump}
18641 command by storing a new value into the register @code{$pc}. The
18642 difference is that this does not start your program running; it only
18643 changes the address of where it @emph{will} run when you continue. For
18651 makes the next @code{continue} command or stepping command execute at
18652 address @code{0x485}, rather than at the address where your program stopped.
18653 @xref{Continuing and Stepping, ,Continuing and Stepping}.
18655 The most common occasion to use the @code{jump} command is to back
18656 up---perhaps with more breakpoints set---over a portion of a program
18657 that has already executed, in order to examine its execution in more
18662 @section Giving your Program a Signal
18663 @cindex deliver a signal to a program
18667 @item signal @var{signal}
18668 Resume execution where your program is stopped, but immediately give it the
18669 signal @var{signal}. The @var{signal} can be the name or the number of a
18670 signal. For example, on many systems @code{signal 2} and @code{signal
18671 SIGINT} are both ways of sending an interrupt signal.
18673 Alternatively, if @var{signal} is zero, continue execution without
18674 giving a signal. This is useful when your program stopped on account of
18675 a signal and would ordinarily see the signal when resumed with the
18676 @code{continue} command; @samp{signal 0} causes it to resume without a
18679 @emph{Note:} When resuming a multi-threaded program, @var{signal} is
18680 delivered to the currently selected thread, not the thread that last
18681 reported a stop. This includes the situation where a thread was
18682 stopped due to a signal. So if you want to continue execution
18683 suppressing the signal that stopped a thread, you should select that
18684 same thread before issuing the @samp{signal 0} command. If you issue
18685 the @samp{signal 0} command with another thread as the selected one,
18686 @value{GDBN} detects that and asks for confirmation.
18688 Invoking the @code{signal} command is not the same as invoking the
18689 @code{kill} utility from the shell. Sending a signal with @code{kill}
18690 causes @value{GDBN} to decide what to do with the signal depending on
18691 the signal handling tables (@pxref{Signals}). The @code{signal} command
18692 passes the signal directly to your program.
18694 @code{signal} does not repeat when you press @key{RET} a second time
18695 after executing the command.
18697 @kindex queue-signal
18698 @item queue-signal @var{signal}
18699 Queue @var{signal} to be delivered immediately to the current thread
18700 when execution of the thread resumes. The @var{signal} can be the name or
18701 the number of a signal. For example, on many systems @code{signal 2} and
18702 @code{signal SIGINT} are both ways of sending an interrupt signal.
18703 The handling of the signal must be set to pass the signal to the program,
18704 otherwise @value{GDBN} will report an error.
18705 You can control the handling of signals from @value{GDBN} with the
18706 @code{handle} command (@pxref{Signals}).
18708 Alternatively, if @var{signal} is zero, any currently queued signal
18709 for the current thread is discarded and when execution resumes no signal
18710 will be delivered. This is useful when your program stopped on account
18711 of a signal and would ordinarily see the signal when resumed with the
18712 @code{continue} command.
18714 This command differs from the @code{signal} command in that the signal
18715 is just queued, execution is not resumed. And @code{queue-signal} cannot
18716 be used to pass a signal whose handling state has been set to @code{nopass}
18721 @xref{stepping into signal handlers}, for information on how stepping
18722 commands behave when the thread has a signal queued.
18725 @section Returning from a Function
18728 @cindex returning from a function
18731 @itemx return @var{expression}
18732 You can cancel execution of a function call with the @code{return}
18733 command. If you give an
18734 @var{expression} argument, its value is used as the function's return
18738 When you use @code{return}, @value{GDBN} discards the selected stack frame
18739 (and all frames within it). You can think of this as making the
18740 discarded frame return prematurely. If you wish to specify a value to
18741 be returned, give that value as the argument to @code{return}.
18743 This pops the selected stack frame (@pxref{Selection, ,Selecting a
18744 Frame}), and any other frames inside of it, leaving its caller as the
18745 innermost remaining frame. That frame becomes selected. The
18746 specified value is stored in the registers used for returning values
18749 The @code{return} command does not resume execution; it leaves the
18750 program stopped in the state that would exist if the function had just
18751 returned. In contrast, the @code{finish} command (@pxref{Continuing
18752 and Stepping, ,Continuing and Stepping}) resumes execution until the
18753 selected stack frame returns naturally.
18755 @value{GDBN} needs to know how the @var{expression} argument should be set for
18756 the inferior. The concrete registers assignment depends on the OS ABI and the
18757 type being returned by the selected stack frame. For example it is common for
18758 OS ABI to return floating point values in FPU registers while integer values in
18759 CPU registers. Still some ABIs return even floating point values in CPU
18760 registers. Larger integer widths (such as @code{long long int}) also have
18761 specific placement rules. @value{GDBN} already knows the OS ABI from its
18762 current target so it needs to find out also the type being returned to make the
18763 assignment into the right register(s).
18765 Normally, the selected stack frame has debug info. @value{GDBN} will always
18766 use the debug info instead of the implicit type of @var{expression} when the
18767 debug info is available. For example, if you type @kbd{return -1}, and the
18768 function in the current stack frame is declared to return a @code{long long
18769 int}, @value{GDBN} transparently converts the implicit @code{int} value of -1
18770 into a @code{long long int}:
18773 Breakpoint 1, func () at gdb.base/return-nodebug.c:29
18775 (@value{GDBP}) return -1
18776 Make func return now? (y or n) y
18777 #0 0x004004f6 in main () at gdb.base/return-nodebug.c:43
18778 43 printf ("result=%lld\n", func ());
18782 However, if the selected stack frame does not have a debug info, e.g., if the
18783 function was compiled without debug info, @value{GDBN} has to find out the type
18784 to return from user. Specifying a different type by mistake may set the value
18785 in different inferior registers than the caller code expects. For example,
18786 typing @kbd{return -1} with its implicit type @code{int} would set only a part
18787 of a @code{long long int} result for a debug info less function (on 32-bit
18788 architectures). Therefore the user is required to specify the return type by
18789 an appropriate cast explicitly:
18792 Breakpoint 2, 0x0040050b in func ()
18793 (@value{GDBP}) return -1
18794 Return value type not available for selected stack frame.
18795 Please use an explicit cast of the value to return.
18796 (@value{GDBP}) return (long long int) -1
18797 Make selected stack frame return now? (y or n) y
18798 #0 0x00400526 in main ()
18803 @section Calling Program Functions
18806 @cindex calling functions
18807 @cindex inferior functions, calling
18808 @item print @var{expr}
18809 Evaluate the expression @var{expr} and display the resulting value.
18810 The expression may include calls to functions in the program being
18814 @item call @var{expr}
18815 Evaluate the expression @var{expr} without displaying @code{void}
18818 You can use this variant of the @code{print} command if you want to
18819 execute a function from your program that does not return anything
18820 (a.k.a.@: @dfn{a void function}), but without cluttering the output
18821 with @code{void} returned values that @value{GDBN} will otherwise
18822 print. If the result is not void, it is printed and saved in the
18826 It is possible for the function you call via the @code{print} or
18827 @code{call} command to generate a signal (e.g., if there's a bug in
18828 the function, or if you passed it incorrect arguments). What happens
18829 in that case is controlled by the @code{set unwindonsignal} command.
18831 Similarly, with a C@t{++} program it is possible for the function you
18832 call via the @code{print} or @code{call} command to generate an
18833 exception that is not handled due to the constraints of the dummy
18834 frame. In this case, any exception that is raised in the frame, but has
18835 an out-of-frame exception handler will not be found. GDB builds a
18836 dummy-frame for the inferior function call, and the unwinder cannot
18837 seek for exception handlers outside of this dummy-frame. What happens
18838 in that case is controlled by the
18839 @code{set unwind-on-terminating-exception} command.
18842 @item set unwindonsignal
18843 @kindex set unwindonsignal
18844 @cindex unwind stack in called functions
18845 @cindex call dummy stack unwinding
18846 Set unwinding of the stack if a signal is received while in a function
18847 that @value{GDBN} called in the program being debugged. If set to on,
18848 @value{GDBN} unwinds the stack it created for the call and restores
18849 the context to what it was before the call. If set to off (the
18850 default), @value{GDBN} stops in the frame where the signal was
18853 @item show unwindonsignal
18854 @kindex show unwindonsignal
18855 Show the current setting of stack unwinding in the functions called by
18858 @item set unwind-on-terminating-exception
18859 @kindex set unwind-on-terminating-exception
18860 @cindex unwind stack in called functions with unhandled exceptions
18861 @cindex call dummy stack unwinding on unhandled exception.
18862 Set unwinding of the stack if a C@t{++} exception is raised, but left
18863 unhandled while in a function that @value{GDBN} called in the program being
18864 debugged. If set to on (the default), @value{GDBN} unwinds the stack
18865 it created for the call and restores the context to what it was before
18866 the call. If set to off, @value{GDBN} the exception is delivered to
18867 the default C@t{++} exception handler and the inferior terminated.
18869 @item show unwind-on-terminating-exception
18870 @kindex show unwind-on-terminating-exception
18871 Show the current setting of stack unwinding in the functions called by
18874 @item set may-call-functions
18875 @kindex set may-call-functions
18876 @cindex disabling calling functions in the program
18877 @cindex calling functions in the program, disabling
18878 Set permission to call functions in the program.
18879 This controls whether @value{GDBN} will attempt to call functions in
18880 the program, such as with expressions in the @code{print} command. It
18881 defaults to @code{on}.
18883 To call a function in the program, @value{GDBN} has to temporarily
18884 modify the state of the inferior. This has potentially undesired side
18885 effects. Also, having @value{GDBN} call nested functions is likely to
18886 be erroneous and may even crash the program being debugged. You can
18887 avoid such hazards by forbidding @value{GDBN} from calling functions
18888 in the program being debugged. If calling functions in the program
18889 is forbidden, GDB will throw an error when a command (such as printing
18890 an expression) starts a function call in the program.
18892 @item show may-call-functions
18893 @kindex show may-call-functions
18894 Show permission to call functions in the program.
18898 @subsection Calling functions with no debug info
18900 @cindex no debug info functions
18901 Sometimes, a function you wish to call is missing debug information.
18902 In such case, @value{GDBN} does not know the type of the function,
18903 including the types of the function's parameters. To avoid calling
18904 the inferior function incorrectly, which could result in the called
18905 function functioning erroneously and even crash, @value{GDBN} refuses
18906 to call the function unless you tell it the type of the function.
18908 For prototyped (i.e.@: ANSI/ISO style) functions, there are two ways
18909 to do that. The simplest is to cast the call to the function's
18910 declared return type. For example:
18913 (@value{GDBP}) p getenv ("PATH")
18914 'getenv' has unknown return type; cast the call to its declared return type
18915 (@value{GDBP}) p (char *) getenv ("PATH")
18916 $1 = 0x7fffffffe7ba "/usr/local/bin:/"...
18919 Casting the return type of a no-debug function is equivalent to
18920 casting the function to a pointer to a prototyped function that has a
18921 prototype that matches the types of the passed-in arguments, and
18922 calling that. I.e., the call above is equivalent to:
18925 (@value{GDBP}) p ((char * (*) (const char *)) getenv) ("PATH")
18929 and given this prototyped C or C++ function with float parameters:
18932 float multiply (float v1, float v2) @{ return v1 * v2; @}
18936 these calls are equivalent:
18939 (@value{GDBP}) p (float) multiply (2.0f, 3.0f)
18940 (@value{GDBP}) p ((float (*) (float, float)) multiply) (2.0f, 3.0f)
18943 If the function you wish to call is declared as unprototyped (i.e.@:
18944 old K&R style), you must use the cast-to-function-pointer syntax, so
18945 that @value{GDBN} knows that it needs to apply default argument
18946 promotions (promote float arguments to double). @xref{ABI, float
18947 promotion}. For example, given this unprototyped C function with
18948 float parameters, and no debug info:
18952 multiply_noproto (v1, v2)
18960 you call it like this:
18963 (@value{GDBP}) p ((float (*) ()) multiply_noproto) (2.0f, 3.0f)
18967 @section Patching Programs
18969 @cindex patching binaries
18970 @cindex writing into executables
18971 @cindex writing into corefiles
18973 By default, @value{GDBN} opens the file containing your program's
18974 executable code (or the corefile) read-only. This prevents accidental
18975 alterations to machine code; but it also prevents you from intentionally
18976 patching your program's binary.
18978 If you'd like to be able to patch the binary, you can specify that
18979 explicitly with the @code{set write} command. For example, you might
18980 want to turn on internal debugging flags, or even to make emergency
18986 @itemx set write off
18987 If you specify @samp{set write on}, @value{GDBN} opens executable and
18988 core files for both reading and writing; if you specify @kbd{set write
18989 off} (the default), @value{GDBN} opens them read-only.
18991 If you have already loaded a file, you must load it again (using the
18992 @code{exec-file} or @code{core-file} command) after changing @code{set
18993 write}, for your new setting to take effect.
18997 Display whether executable files and core files are opened for writing
18998 as well as reading.
19001 @node Compiling and Injecting Code
19002 @section Compiling and injecting code in @value{GDBN}
19003 @cindex injecting code
19004 @cindex writing into executables
19005 @cindex compiling code
19007 @value{GDBN} supports on-demand compilation and code injection into
19008 programs running under @value{GDBN}. GCC 5.0 or higher built with
19009 @file{libcc1.so} must be installed for this functionality to be enabled.
19010 This functionality is implemented with the following commands.
19013 @kindex compile code
19014 @item compile code @var{source-code}
19015 @itemx compile code -raw @var{--} @var{source-code}
19016 Compile @var{source-code} with the compiler language found as the current
19017 language in @value{GDBN} (@pxref{Languages}). If compilation and
19018 injection is not supported with the current language specified in
19019 @value{GDBN}, or the compiler does not support this feature, an error
19020 message will be printed. If @var{source-code} compiles and links
19021 successfully, @value{GDBN} will load the object-code emitted,
19022 and execute it within the context of the currently selected inferior.
19023 It is important to note that the compiled code is executed immediately.
19024 After execution, the compiled code is removed from @value{GDBN} and any
19025 new types or variables you have defined will be deleted.
19027 The command allows you to specify @var{source-code} in two ways.
19028 The simplest method is to provide a single line of code to the command.
19032 compile code printf ("hello world\n");
19035 If you specify options on the command line as well as source code, they
19036 may conflict. The @samp{--} delimiter can be used to separate options
19037 from actual source code. E.g.:
19040 compile code -r -- printf ("hello world\n");
19043 Alternatively you can enter source code as multiple lines of text. To
19044 enter this mode, invoke the @samp{compile code} command without any text
19045 following the command. This will start the multiple-line editor and
19046 allow you to type as many lines of source code as required. When you
19047 have completed typing, enter @samp{end} on its own line to exit the
19052 >printf ("hello\n");
19053 >printf ("world\n");
19057 Specifying @samp{-raw}, prohibits @value{GDBN} from wrapping the
19058 provided @var{source-code} in a callable scope. In this case, you must
19059 specify the entry point of the code by defining a function named
19060 @code{_gdb_expr_}. The @samp{-raw} code cannot access variables of the
19061 inferior. Using @samp{-raw} option may be needed for example when
19062 @var{source-code} requires @samp{#include} lines which may conflict with
19063 inferior symbols otherwise.
19065 @kindex compile file
19066 @item compile file @var{filename}
19067 @itemx compile file -raw @var{filename}
19068 Like @code{compile code}, but take the source code from @var{filename}.
19071 compile file /home/user/example.c
19076 @item compile print @var{expr}
19077 @itemx compile print /@var{f} @var{expr}
19078 Compile and execute @var{expr} with the compiler language found as the
19079 current language in @value{GDBN} (@pxref{Languages}). By default the
19080 value of @var{expr} is printed in a format appropriate to its data type;
19081 you can choose a different format by specifying @samp{/@var{f}}, where
19082 @var{f} is a letter specifying the format; see @ref{Output Formats,,Output
19085 @item compile print
19086 @itemx compile print /@var{f}
19087 @cindex reprint the last value
19088 Alternatively you can enter the expression (source code producing it) as
19089 multiple lines of text. To enter this mode, invoke the @samp{compile print}
19090 command without any text following the command. This will start the
19091 multiple-line editor.
19095 The process of compiling and injecting the code can be inspected using:
19098 @anchor{set debug compile}
19099 @item set debug compile
19100 @cindex compile command debugging info
19101 Turns on or off display of @value{GDBN} process of compiling and
19102 injecting the code. The default is off.
19104 @item show debug compile
19105 Displays the current state of displaying @value{GDBN} process of
19106 compiling and injecting the code.
19108 @anchor{set debug compile-cplus-types}
19109 @item set debug compile-cplus-types
19110 @cindex compile C@t{++} type conversion
19111 Turns on or off the display of C@t{++} type conversion debugging information.
19112 The default is off.
19114 @item show debug compile-cplus-types
19115 Displays the current state of displaying debugging information for
19116 C@t{++} type conversion.
19119 @subsection Compilation options for the @code{compile} command
19121 @value{GDBN} needs to specify the right compilation options for the code
19122 to be injected, in part to make its ABI compatible with the inferior
19123 and in part to make the injected code compatible with @value{GDBN}'s
19127 The options used, in increasing precedence:
19130 @item target architecture and OS options (@code{gdbarch})
19131 These options depend on target processor type and target operating
19132 system, usually they specify at least 32-bit (@code{-m32}) or 64-bit
19133 (@code{-m64}) compilation option.
19135 @item compilation options recorded in the target
19136 @value{NGCC} (since version 4.7) stores the options used for compilation
19137 into @code{DW_AT_producer} part of DWARF debugging information according
19138 to the @value{NGCC} option @code{-grecord-gcc-switches}. One has to
19139 explicitly specify @code{-g} during inferior compilation otherwise
19140 @value{NGCC} produces no DWARF. This feature is only relevant for
19141 platforms where @code{-g} produces DWARF by default, otherwise one may
19142 try to enforce DWARF by using @code{-gdwarf-4}.
19144 @item compilation options set by @code{set compile-args}
19148 You can override compilation options using the following command:
19151 @item set compile-args
19152 @cindex compile command options override
19153 Set compilation options used for compiling and injecting code with the
19154 @code{compile} commands. These options override any conflicting ones
19155 from the target architecture and/or options stored during inferior
19158 @item show compile-args
19159 Displays the current state of compilation options override.
19160 This does not show all the options actually used during compilation,
19161 use @ref{set debug compile} for that.
19164 @subsection Caveats when using the @code{compile} command
19166 There are a few caveats to keep in mind when using the @code{compile}
19167 command. As the caveats are different per language, the table below
19168 highlights specific issues on a per language basis.
19171 @item C code examples and caveats
19172 When the language in @value{GDBN} is set to @samp{C}, the compiler will
19173 attempt to compile the source code with a @samp{C} compiler. The source
19174 code provided to the @code{compile} command will have much the same
19175 access to variables and types as it normally would if it were part of
19176 the program currently being debugged in @value{GDBN}.
19178 Below is a sample program that forms the basis of the examples that
19179 follow. This program has been compiled and loaded into @value{GDBN},
19180 much like any other normal debugging session.
19183 void function1 (void)
19186 printf ("function 1\n");
19189 void function2 (void)
19204 For the purposes of the examples in this section, the program above has
19205 been compiled, loaded into @value{GDBN}, stopped at the function
19206 @code{main}, and @value{GDBN} is awaiting input from the user.
19208 To access variables and types for any program in @value{GDBN}, the
19209 program must be compiled and packaged with debug information. The
19210 @code{compile} command is not an exception to this rule. Without debug
19211 information, you can still use the @code{compile} command, but you will
19212 be very limited in what variables and types you can access.
19214 So with that in mind, the example above has been compiled with debug
19215 information enabled. The @code{compile} command will have access to
19216 all variables and types (except those that may have been optimized
19217 out). Currently, as @value{GDBN} has stopped the program in the
19218 @code{main} function, the @code{compile} command would have access to
19219 the variable @code{k}. You could invoke the @code{compile} command
19220 and type some source code to set the value of @code{k}. You can also
19221 read it, or do anything with that variable you would normally do in
19222 @code{C}. Be aware that changes to inferior variables in the
19223 @code{compile} command are persistent. In the following example:
19226 compile code k = 3;
19230 the variable @code{k} is now 3. It will retain that value until
19231 something else in the example program changes it, or another
19232 @code{compile} command changes it.
19234 Normal scope and access rules apply to source code compiled and
19235 injected by the @code{compile} command. In the example, the variables
19236 @code{j} and @code{k} are not accessible yet, because the program is
19237 currently stopped in the @code{main} function, where these variables
19238 are not in scope. Therefore, the following command
19241 compile code j = 3;
19245 will result in a compilation error message.
19247 Once the program is continued, execution will bring these variables in
19248 scope, and they will become accessible; then the code you specify via
19249 the @code{compile} command will be able to access them.
19251 You can create variables and types with the @code{compile} command as
19252 part of your source code. Variables and types that are created as part
19253 of the @code{compile} command are not visible to the rest of the program for
19254 the duration of its run. This example is valid:
19257 compile code int ff = 5; printf ("ff is %d\n", ff);
19260 However, if you were to type the following into @value{GDBN} after that
19261 command has completed:
19264 compile code printf ("ff is %d\n'', ff);
19268 a compiler error would be raised as the variable @code{ff} no longer
19269 exists. Object code generated and injected by the @code{compile}
19270 command is removed when its execution ends. Caution is advised
19271 when assigning to program variables values of variables created by the
19272 code submitted to the @code{compile} command. This example is valid:
19275 compile code int ff = 5; k = ff;
19278 The value of the variable @code{ff} is assigned to @code{k}. The variable
19279 @code{k} does not require the existence of @code{ff} to maintain the value
19280 it has been assigned. However, pointers require particular care in
19281 assignment. If the source code compiled with the @code{compile} command
19282 changed the address of a pointer in the example program, perhaps to a
19283 variable created in the @code{compile} command, that pointer would point
19284 to an invalid location when the command exits. The following example
19285 would likely cause issues with your debugged program:
19288 compile code int ff = 5; p = &ff;
19291 In this example, @code{p} would point to @code{ff} when the
19292 @code{compile} command is executing the source code provided to it.
19293 However, as variables in the (example) program persist with their
19294 assigned values, the variable @code{p} would point to an invalid
19295 location when the command exists. A general rule should be followed
19296 in that you should either assign @code{NULL} to any assigned pointers,
19297 or restore a valid location to the pointer before the command exits.
19299 Similar caution must be exercised with any structs, unions, and typedefs
19300 defined in @code{compile} command. Types defined in the @code{compile}
19301 command will no longer be available in the next @code{compile} command.
19302 Therefore, if you cast a variable to a type defined in the
19303 @code{compile} command, care must be taken to ensure that any future
19304 need to resolve the type can be achieved.
19307 (gdb) compile code static struct a @{ int a; @} v = @{ 42 @}; argv = &v;
19308 (gdb) compile code printf ("%d\n", ((struct a *) argv)->a);
19309 gdb command line:1:36: error: dereferencing pointer to incomplete type ‘struct a’
19310 Compilation failed.
19311 (gdb) compile code struct a @{ int a; @}; printf ("%d\n", ((struct a *) argv)->a);
19315 Variables that have been optimized away by the compiler are not
19316 accessible to the code submitted to the @code{compile} command.
19317 Access to those variables will generate a compiler error which @value{GDBN}
19318 will print to the console.
19321 @subsection Compiler search for the @code{compile} command
19323 @value{GDBN} needs to find @value{NGCC} for the inferior being debugged
19324 which may not be obvious for remote targets of different architecture
19325 than where @value{GDBN} is running. Environment variable @code{PATH} on
19326 @value{GDBN} host is searched for @value{NGCC} binary matching the
19327 target architecture and operating system. This search can be overriden
19328 by @code{set compile-gcc} @value{GDBN} command below. @code{PATH} is
19329 taken from shell that executed @value{GDBN}, it is not the value set by
19330 @value{GDBN} command @code{set environment}). @xref{Environment}.
19333 Specifically @code{PATH} is searched for binaries matching regular expression
19334 @code{@var{arch}(-[^-]*)?-@var{os}-gcc} according to the inferior target being
19335 debugged. @var{arch} is processor name --- multiarch is supported, so for
19336 example both @code{i386} and @code{x86_64} targets look for pattern
19337 @code{(x86_64|i.86)} and both @code{s390} and @code{s390x} targets look
19338 for pattern @code{s390x?}. @var{os} is currently supported only for
19339 pattern @code{linux(-gnu)?}.
19341 On Posix hosts the compiler driver @value{GDBN} needs to find also
19342 shared library @file{libcc1.so} from the compiler. It is searched in
19343 default shared library search path (overridable with usual environment
19344 variable @code{LD_LIBRARY_PATH}), unrelated to @code{PATH} or @code{set
19345 compile-gcc} settings. Contrary to it @file{libcc1plugin.so} is found
19346 according to the installation of the found compiler --- as possibly
19347 specified by the @code{set compile-gcc} command.
19350 @item set compile-gcc
19351 @cindex compile command driver filename override
19352 Set compilation command used for compiling and injecting code with the
19353 @code{compile} commands. If this option is not set (it is set to
19354 an empty string), the search described above will occur --- that is the
19357 @item show compile-gcc
19358 Displays the current compile command @value{NGCC} driver filename.
19359 If set, it is the main command @command{gcc}, found usually for example
19360 under name @file{x86_64-linux-gnu-gcc}.
19364 @chapter @value{GDBN} Files
19366 @value{GDBN} needs to know the file name of the program to be debugged,
19367 both in order to read its symbol table and in order to start your
19368 program. To debug a core dump of a previous run, you must also tell
19369 @value{GDBN} the name of the core dump file.
19372 * Files:: Commands to specify files
19373 * File Caching:: Information about @value{GDBN}'s file caching
19374 * Separate Debug Files:: Debugging information in separate files
19375 * MiniDebugInfo:: Debugging information in a special section
19376 * Index Files:: Index files speed up GDB
19377 * Symbol Errors:: Errors reading symbol files
19378 * Data Files:: GDB data files
19382 @section Commands to Specify Files
19384 @cindex symbol table
19385 @cindex core dump file
19387 You may want to specify executable and core dump file names. The usual
19388 way to do this is at start-up time, using the arguments to
19389 @value{GDBN}'s start-up commands (@pxref{Invocation, , Getting In and
19390 Out of @value{GDBN}}).
19392 Occasionally it is necessary to change to a different file during a
19393 @value{GDBN} session. Or you may run @value{GDBN} and forget to
19394 specify a file you want to use. Or you are debugging a remote target
19395 via @code{gdbserver} (@pxref{Server, file, Using the @code{gdbserver}
19396 Program}). In these situations the @value{GDBN} commands to specify
19397 new files are useful.
19400 @cindex executable file
19402 @item file @var{filename}
19403 Use @var{filename} as the program to be debugged. It is read for its
19404 symbols and for the contents of pure memory. It is also the program
19405 executed when you use the @code{run} command. If you do not specify a
19406 directory and the file is not found in the @value{GDBN} working directory,
19407 @value{GDBN} uses the environment variable @code{PATH} as a list of
19408 directories to search, just as the shell does when looking for a program
19409 to run. You can change the value of this variable, for both @value{GDBN}
19410 and your program, using the @code{path} command.
19412 @cindex unlinked object files
19413 @cindex patching object files
19414 You can load unlinked object @file{.o} files into @value{GDBN} using
19415 the @code{file} command. You will not be able to ``run'' an object
19416 file, but you can disassemble functions and inspect variables. Also,
19417 if the underlying BFD functionality supports it, you could use
19418 @kbd{gdb -write} to patch object files using this technique. Note
19419 that @value{GDBN} can neither interpret nor modify relocations in this
19420 case, so branches and some initialized variables will appear to go to
19421 the wrong place. But this feature is still handy from time to time.
19424 @code{file} with no argument makes @value{GDBN} discard any information it
19425 has on both executable file and the symbol table.
19428 @item exec-file @r{[} @var{filename} @r{]}
19429 Specify that the program to be run (but not the symbol table) is found
19430 in @var{filename}. @value{GDBN} searches the environment variable @code{PATH}
19431 if necessary to locate your program. Omitting @var{filename} means to
19432 discard information on the executable file.
19434 @kindex symbol-file
19435 @item symbol-file @r{[} @var{filename} @r{[} -o @var{offset} @r{]]}
19436 Read symbol table information from file @var{filename}. @code{PATH} is
19437 searched when necessary. Use the @code{file} command to get both symbol
19438 table and program to run from the same file.
19440 If an optional @var{offset} is specified, it is added to the start
19441 address of each section in the symbol file. This is useful if the
19442 program is relocated at runtime, such as the Linux kernel with kASLR
19445 @code{symbol-file} with no argument clears out @value{GDBN} information on your
19446 program's symbol table.
19448 The @code{symbol-file} command causes @value{GDBN} to forget the contents of
19449 some breakpoints and auto-display expressions. This is because they may
19450 contain pointers to the internal data recording symbols and data types,
19451 which are part of the old symbol table data being discarded inside
19454 @code{symbol-file} does not repeat if you press @key{RET} again after
19457 When @value{GDBN} is configured for a particular environment, it
19458 understands debugging information in whatever format is the standard
19459 generated for that environment; you may use either a @sc{gnu} compiler, or
19460 other compilers that adhere to the local conventions.
19461 Best results are usually obtained from @sc{gnu} compilers; for example,
19462 using @code{@value{NGCC}} you can generate debugging information for
19465 For most kinds of object files, with the exception of old SVR3 systems
19466 using COFF, the @code{symbol-file} command does not normally read the
19467 symbol table in full right away. Instead, it scans the symbol table
19468 quickly to find which source files and which symbols are present. The
19469 details are read later, one source file at a time, as they are needed.
19471 The purpose of this two-stage reading strategy is to make @value{GDBN}
19472 start up faster. For the most part, it is invisible except for
19473 occasional pauses while the symbol table details for a particular source
19474 file are being read. (The @code{set verbose} command can turn these
19475 pauses into messages if desired. @xref{Messages/Warnings, ,Optional
19476 Warnings and Messages}.)
19478 We have not implemented the two-stage strategy for COFF yet. When the
19479 symbol table is stored in COFF format, @code{symbol-file} reads the
19480 symbol table data in full right away. Note that ``stabs-in-COFF''
19481 still does the two-stage strategy, since the debug info is actually
19485 @cindex reading symbols immediately
19486 @cindex symbols, reading immediately
19487 @item symbol-file @r{[} -readnow @r{]} @var{filename}
19488 @itemx file @r{[} -readnow @r{]} @var{filename}
19489 You can override the @value{GDBN} two-stage strategy for reading symbol
19490 tables by using the @samp{-readnow} option with any of the commands that
19491 load symbol table information, if you want to be sure @value{GDBN} has the
19492 entire symbol table available.
19494 @cindex @code{-readnever}, option for symbol-file command
19495 @cindex never read symbols
19496 @cindex symbols, never read
19497 @item symbol-file @r{[} -readnever @r{]} @var{filename}
19498 @itemx file @r{[} -readnever @r{]} @var{filename}
19499 You can instruct @value{GDBN} to never read the symbolic information
19500 contained in @var{filename} by using the @samp{-readnever} option.
19501 @xref{--readnever}.
19503 @c FIXME: for now no mention of directories, since this seems to be in
19504 @c flux. 13mar1992 status is that in theory GDB would look either in
19505 @c current dir or in same dir as myprog; but issues like competing
19506 @c GDB's, or clutter in system dirs, mean that in practice right now
19507 @c only current dir is used. FFish says maybe a special GDB hierarchy
19508 @c (eg rooted in val of env var GDBSYMS) could exist for mappable symbol
19512 @item core-file @r{[}@var{filename}@r{]}
19514 Specify the whereabouts of a core dump file to be used as the ``contents
19515 of memory''. Traditionally, core files contain only some parts of the
19516 address space of the process that generated them; @value{GDBN} can access the
19517 executable file itself for other parts.
19519 @code{core-file} with no argument specifies that no core file is
19522 Note that the core file is ignored when your program is actually running
19523 under @value{GDBN}. So, if you have been running your program and you
19524 wish to debug a core file instead, you must kill the subprocess in which
19525 the program is running. To do this, use the @code{kill} command
19526 (@pxref{Kill Process, ,Killing the Child Process}).
19528 @kindex add-symbol-file
19529 @cindex dynamic linking
19530 @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{]}
19531 The @code{add-symbol-file} command reads additional symbol table
19532 information from the file @var{filename}. You would use this command
19533 when @var{filename} has been dynamically loaded (by some other means)
19534 into the program that is running. The @var{textaddress} parameter gives
19535 the memory address at which the file's text section has been loaded.
19536 You can additionally specify the base address of other sections using
19537 an arbitrary number of @samp{-s @var{section} @var{address}} pairs.
19538 If a section is omitted, @value{GDBN} will use its default addresses
19539 as found in @var{filename}. Any @var{address} or @var{textaddress}
19540 can be given as an expression.
19542 If an optional @var{offset} is specified, it is added to the start
19543 address of each section, except those for which the address was
19544 specified explicitly.
19546 The symbol table of the file @var{filename} is added to the symbol table
19547 originally read with the @code{symbol-file} command. You can use the
19548 @code{add-symbol-file} command any number of times; the new symbol data
19549 thus read is kept in addition to the old.
19551 Changes can be reverted using the command @code{remove-symbol-file}.
19553 @cindex relocatable object files, reading symbols from
19554 @cindex object files, relocatable, reading symbols from
19555 @cindex reading symbols from relocatable object files
19556 @cindex symbols, reading from relocatable object files
19557 @cindex @file{.o} files, reading symbols from
19558 Although @var{filename} is typically a shared library file, an
19559 executable file, or some other object file which has been fully
19560 relocated for loading into a process, you can also load symbolic
19561 information from relocatable @file{.o} files, as long as:
19565 the file's symbolic information refers only to linker symbols defined in
19566 that file, not to symbols defined by other object files,
19568 every section the file's symbolic information refers to has actually
19569 been loaded into the inferior, as it appears in the file, and
19571 you can determine the address at which every section was loaded, and
19572 provide these to the @code{add-symbol-file} command.
19576 Some embedded operating systems, like Sun Chorus and VxWorks, can load
19577 relocatable files into an already running program; such systems
19578 typically make the requirements above easy to meet. However, it's
19579 important to recognize that many native systems use complex link
19580 procedures (@code{.linkonce} section factoring and C@t{++} constructor table
19581 assembly, for example) that make the requirements difficult to meet. In
19582 general, one cannot assume that using @code{add-symbol-file} to read a
19583 relocatable object file's symbolic information will have the same effect
19584 as linking the relocatable object file into the program in the normal
19587 @code{add-symbol-file} does not repeat if you press @key{RET} after using it.
19589 @kindex remove-symbol-file
19590 @item remove-symbol-file @var{filename}
19591 @item remove-symbol-file -a @var{address}
19592 Remove a symbol file added via the @code{add-symbol-file} command. The
19593 file to remove can be identified by its @var{filename} or by an @var{address}
19594 that lies within the boundaries of this symbol file in memory. Example:
19597 (gdb) add-symbol-file /home/user/gdb/mylib.so 0x7ffff7ff9480
19598 add symbol table from file "/home/user/gdb/mylib.so" at
19599 .text_addr = 0x7ffff7ff9480
19601 Reading symbols from /home/user/gdb/mylib.so...done.
19602 (gdb) remove-symbol-file -a 0x7ffff7ff9480
19603 Remove symbol table from file "/home/user/gdb/mylib.so"? (y or n) y
19608 @code{remove-symbol-file} does not repeat if you press @key{RET} after using it.
19610 @kindex add-symbol-file-from-memory
19611 @cindex @code{syscall DSO}
19612 @cindex load symbols from memory
19613 @item add-symbol-file-from-memory @var{address}
19614 Load symbols from the given @var{address} in a dynamically loaded
19615 object file whose image is mapped directly into the inferior's memory.
19616 For example, the Linux kernel maps a @code{syscall DSO} into each
19617 process's address space; this DSO provides kernel-specific code for
19618 some system calls. The argument can be any expression whose
19619 evaluation yields the address of the file's shared object file header.
19620 For this command to work, you must have used @code{symbol-file} or
19621 @code{exec-file} commands in advance.
19624 @item section @var{section} @var{addr}
19625 The @code{section} command changes the base address of the named
19626 @var{section} of the exec file to @var{addr}. This can be used if the
19627 exec file does not contain section addresses, (such as in the
19628 @code{a.out} format), or when the addresses specified in the file
19629 itself are wrong. Each section must be changed separately. The
19630 @code{info files} command, described below, lists all the sections and
19634 @kindex info target
19637 @code{info files} and @code{info target} are synonymous; both print the
19638 current target (@pxref{Targets, ,Specifying a Debugging Target}),
19639 including the names of the executable and core dump files currently in
19640 use by @value{GDBN}, and the files from which symbols were loaded. The
19641 command @code{help target} lists all possible targets rather than
19644 @kindex maint info sections
19645 @item maint info sections
19646 Another command that can give you extra information about program sections
19647 is @code{maint info sections}. In addition to the section information
19648 displayed by @code{info files}, this command displays the flags and file
19649 offset of each section in the executable and core dump files. In addition,
19650 @code{maint info sections} provides the following command options (which
19651 may be arbitrarily combined):
19655 Display sections for all loaded object files, including shared libraries.
19656 @item @var{sections}
19657 Display info only for named @var{sections}.
19658 @item @var{section-flags}
19659 Display info only for sections for which @var{section-flags} are true.
19660 The section flags that @value{GDBN} currently knows about are:
19663 Section will have space allocated in the process when loaded.
19664 Set for all sections except those containing debug information.
19666 Section will be loaded from the file into the child process memory.
19667 Set for pre-initialized code and data, clear for @code{.bss} sections.
19669 Section needs to be relocated before loading.
19671 Section cannot be modified by the child process.
19673 Section contains executable code only.
19675 Section contains data only (no executable code).
19677 Section will reside in ROM.
19679 Section contains data for constructor/destructor lists.
19681 Section is not empty.
19683 An instruction to the linker to not output the section.
19684 @item COFF_SHARED_LIBRARY
19685 A notification to the linker that the section contains
19686 COFF shared library information.
19688 Section contains common symbols.
19691 @kindex set trust-readonly-sections
19692 @cindex read-only sections
19693 @item set trust-readonly-sections on
19694 Tell @value{GDBN} that readonly sections in your object file
19695 really are read-only (i.e.@: that their contents will not change).
19696 In that case, @value{GDBN} can fetch values from these sections
19697 out of the object file, rather than from the target program.
19698 For some targets (notably embedded ones), this can be a significant
19699 enhancement to debugging performance.
19701 The default is off.
19703 @item set trust-readonly-sections off
19704 Tell @value{GDBN} not to trust readonly sections. This means that
19705 the contents of the section might change while the program is running,
19706 and must therefore be fetched from the target when needed.
19708 @item show trust-readonly-sections
19709 Show the current setting of trusting readonly sections.
19712 All file-specifying commands allow both absolute and relative file names
19713 as arguments. @value{GDBN} always converts the file name to an absolute file
19714 name and remembers it that way.
19716 @cindex shared libraries
19717 @anchor{Shared Libraries}
19718 @value{GDBN} supports @sc{gnu}/Linux, MS-Windows, SunOS,
19719 Darwin/Mach-O, SVr4, IBM RS/6000 AIX, QNX Neutrino, FDPIC (FR-V), and
19720 DSBT (TIC6X) shared libraries.
19722 On MS-Windows @value{GDBN} must be linked with the Expat library to support
19723 shared libraries. @xref{Expat}.
19725 @value{GDBN} automatically loads symbol definitions from shared libraries
19726 when you use the @code{run} command, or when you examine a core file.
19727 (Before you issue the @code{run} command, @value{GDBN} does not understand
19728 references to a function in a shared library, however---unless you are
19729 debugging a core file).
19731 @c FIXME: some @value{GDBN} release may permit some refs to undef
19732 @c FIXME...symbols---eg in a break cmd---assuming they are from a shared
19733 @c FIXME...lib; check this from time to time when updating manual
19735 There are times, however, when you may wish to not automatically load
19736 symbol definitions from shared libraries, such as when they are
19737 particularly large or there are many of them.
19739 To control the automatic loading of shared library symbols, use the
19743 @kindex set auto-solib-add
19744 @item set auto-solib-add @var{mode}
19745 If @var{mode} is @code{on}, symbols from all shared object libraries
19746 will be loaded automatically when the inferior begins execution, you
19747 attach to an independently started inferior, or when the dynamic linker
19748 informs @value{GDBN} that a new library has been loaded. If @var{mode}
19749 is @code{off}, symbols must be loaded manually, using the
19750 @code{sharedlibrary} command. The default value is @code{on}.
19752 @cindex memory used for symbol tables
19753 If your program uses lots of shared libraries with debug info that
19754 takes large amounts of memory, you can decrease the @value{GDBN}
19755 memory footprint by preventing it from automatically loading the
19756 symbols from shared libraries. To that end, type @kbd{set
19757 auto-solib-add off} before running the inferior, then load each
19758 library whose debug symbols you do need with @kbd{sharedlibrary
19759 @var{regexp}}, where @var{regexp} is a regular expression that matches
19760 the libraries whose symbols you want to be loaded.
19762 @kindex show auto-solib-add
19763 @item show auto-solib-add
19764 Display the current autoloading mode.
19767 @cindex load shared library
19768 To explicitly load shared library symbols, use the @code{sharedlibrary}
19772 @kindex info sharedlibrary
19774 @item info share @var{regex}
19775 @itemx info sharedlibrary @var{regex}
19776 Print the names of the shared libraries which are currently loaded
19777 that match @var{regex}. If @var{regex} is omitted then print
19778 all shared libraries that are loaded.
19781 @item info dll @var{regex}
19782 This is an alias of @code{info sharedlibrary}.
19784 @kindex sharedlibrary
19786 @item sharedlibrary @var{regex}
19787 @itemx share @var{regex}
19788 Load shared object library symbols for files matching a
19789 Unix regular expression.
19790 As with files loaded automatically, it only loads shared libraries
19791 required by your program for a core file or after typing @code{run}. If
19792 @var{regex} is omitted all shared libraries required by your program are
19795 @item nosharedlibrary
19796 @kindex nosharedlibrary
19797 @cindex unload symbols from shared libraries
19798 Unload all shared object library symbols. This discards all symbols
19799 that have been loaded from all shared libraries. Symbols from shared
19800 libraries that were loaded by explicit user requests are not
19804 Sometimes you may wish that @value{GDBN} stops and gives you control
19805 when any of shared library events happen. The best way to do this is
19806 to use @code{catch load} and @code{catch unload} (@pxref{Set
19809 @value{GDBN} also supports the the @code{set stop-on-solib-events}
19810 command for this. This command exists for historical reasons. It is
19811 less useful than setting a catchpoint, because it does not allow for
19812 conditions or commands as a catchpoint does.
19815 @item set stop-on-solib-events
19816 @kindex set stop-on-solib-events
19817 This command controls whether @value{GDBN} should give you control
19818 when the dynamic linker notifies it about some shared library event.
19819 The most common event of interest is loading or unloading of a new
19822 @item show stop-on-solib-events
19823 @kindex show stop-on-solib-events
19824 Show whether @value{GDBN} stops and gives you control when shared
19825 library events happen.
19828 Shared libraries are also supported in many cross or remote debugging
19829 configurations. @value{GDBN} needs to have access to the target's libraries;
19830 this can be accomplished either by providing copies of the libraries
19831 on the host system, or by asking @value{GDBN} to automatically retrieve the
19832 libraries from the target. If copies of the target libraries are
19833 provided, they need to be the same as the target libraries, although the
19834 copies on the target can be stripped as long as the copies on the host are
19837 @cindex where to look for shared libraries
19838 For remote debugging, you need to tell @value{GDBN} where the target
19839 libraries are, so that it can load the correct copies---otherwise, it
19840 may try to load the host's libraries. @value{GDBN} has two variables
19841 to specify the search directories for target libraries.
19844 @cindex prefix for executable and shared library file names
19845 @cindex system root, alternate
19846 @kindex set solib-absolute-prefix
19847 @kindex set sysroot
19848 @item set sysroot @var{path}
19849 Use @var{path} as the system root for the program being debugged. Any
19850 absolute shared library paths will be prefixed with @var{path}; many
19851 runtime loaders store the absolute paths to the shared library in the
19852 target program's memory. When starting processes remotely, and when
19853 attaching to already-running processes (local or remote), their
19854 executable filenames will be prefixed with @var{path} if reported to
19855 @value{GDBN} as absolute by the operating system. If you use
19856 @code{set sysroot} to find executables and shared libraries, they need
19857 to be laid out in the same way that they are on the target, with
19858 e.g.@: a @file{/bin}, @file{/lib} and @file{/usr/lib} hierarchy under
19861 If @var{path} starts with the sequence @file{target:} and the target
19862 system is remote then @value{GDBN} will retrieve the target binaries
19863 from the remote system. This is only supported when using a remote
19864 target that supports the @code{remote get} command (@pxref{File
19865 Transfer,,Sending files to a remote system}). The part of @var{path}
19866 following the initial @file{target:} (if present) is used as system
19867 root prefix on the remote file system. If @var{path} starts with the
19868 sequence @file{remote:} this is converted to the sequence
19869 @file{target:} by @code{set sysroot}@footnote{Historically the
19870 functionality to retrieve binaries from the remote system was
19871 provided by prefixing @var{path} with @file{remote:}}. If you want
19872 to specify a local system root using a directory that happens to be
19873 named @file{target:} or @file{remote:}, you need to use some
19874 equivalent variant of the name like @file{./target:}.
19876 For targets with an MS-DOS based filesystem, such as MS-Windows and
19877 SymbianOS, @value{GDBN} tries prefixing a few variants of the target
19878 absolute file name with @var{path}. But first, on Unix hosts,
19879 @value{GDBN} converts all backslash directory separators into forward
19880 slashes, because the backslash is not a directory separator on Unix:
19883 c:\foo\bar.dll @result{} c:/foo/bar.dll
19886 Then, @value{GDBN} attempts prefixing the target file name with
19887 @var{path}, and looks for the resulting file name in the host file
19891 c:/foo/bar.dll @result{} /path/to/sysroot/c:/foo/bar.dll
19894 If that does not find the binary, @value{GDBN} tries removing
19895 the @samp{:} character from the drive spec, both for convenience, and,
19896 for the case of the host file system not supporting file names with
19900 c:/foo/bar.dll @result{} /path/to/sysroot/c/foo/bar.dll
19903 This makes it possible to have a system root that mirrors a target
19904 with more than one drive. E.g., you may want to setup your local
19905 copies of the target system shared libraries like so (note @samp{c} vs
19909 @file{/path/to/sysroot/c/sys/bin/foo.dll}
19910 @file{/path/to/sysroot/c/sys/bin/bar.dll}
19911 @file{/path/to/sysroot/z/sys/bin/bar.dll}
19915 and point the system root at @file{/path/to/sysroot}, so that
19916 @value{GDBN} can find the correct copies of both
19917 @file{c:\sys\bin\foo.dll}, and @file{z:\sys\bin\bar.dll}.
19919 If that still does not find the binary, @value{GDBN} tries
19920 removing the whole drive spec from the target file name:
19923 c:/foo/bar.dll @result{} /path/to/sysroot/foo/bar.dll
19926 This last lookup makes it possible to not care about the drive name,
19927 if you don't want or need to.
19929 The @code{set solib-absolute-prefix} command is an alias for @code{set
19932 @cindex default system root
19933 @cindex @samp{--with-sysroot}
19934 You can set the default system root by using the configure-time
19935 @samp{--with-sysroot} option. If the system root is inside
19936 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
19937 @samp{--exec-prefix}), then the default system root will be updated
19938 automatically if the installed @value{GDBN} is moved to a new
19941 @kindex show sysroot
19943 Display the current executable and shared library prefix.
19945 @kindex set solib-search-path
19946 @item set solib-search-path @var{path}
19947 If this variable is set, @var{path} is a colon-separated list of
19948 directories to search for shared libraries. @samp{solib-search-path}
19949 is used after @samp{sysroot} fails to locate the library, or if the
19950 path to the library is relative instead of absolute. If you want to
19951 use @samp{solib-search-path} instead of @samp{sysroot}, be sure to set
19952 @samp{sysroot} to a nonexistent directory to prevent @value{GDBN} from
19953 finding your host's libraries. @samp{sysroot} is preferred; setting
19954 it to a nonexistent directory may interfere with automatic loading
19955 of shared library symbols.
19957 @kindex show solib-search-path
19958 @item show solib-search-path
19959 Display the current shared library search path.
19961 @cindex DOS file-name semantics of file names.
19962 @kindex set target-file-system-kind (unix|dos-based|auto)
19963 @kindex show target-file-system-kind
19964 @item set target-file-system-kind @var{kind}
19965 Set assumed file system kind for target reported file names.
19967 Shared library file names as reported by the target system may not
19968 make sense as is on the system @value{GDBN} is running on. For
19969 example, when remote debugging a target that has MS-DOS based file
19970 system semantics, from a Unix host, the target may be reporting to
19971 @value{GDBN} a list of loaded shared libraries with file names such as
19972 @file{c:\Windows\kernel32.dll}. On Unix hosts, there's no concept of
19973 drive letters, so the @samp{c:\} prefix is not normally understood as
19974 indicating an absolute file name, and neither is the backslash
19975 normally considered a directory separator character. In that case,
19976 the native file system would interpret this whole absolute file name
19977 as a relative file name with no directory components. This would make
19978 it impossible to point @value{GDBN} at a copy of the remote target's
19979 shared libraries on the host using @code{set sysroot}, and impractical
19980 with @code{set solib-search-path}. Setting
19981 @code{target-file-system-kind} to @code{dos-based} tells @value{GDBN}
19982 to interpret such file names similarly to how the target would, and to
19983 map them to file names valid on @value{GDBN}'s native file system
19984 semantics. The value of @var{kind} can be @code{"auto"}, in addition
19985 to one of the supported file system kinds. In that case, @value{GDBN}
19986 tries to determine the appropriate file system variant based on the
19987 current target's operating system (@pxref{ABI, ,Configuring the
19988 Current ABI}). The supported file system settings are:
19992 Instruct @value{GDBN} to assume the target file system is of Unix
19993 kind. Only file names starting the forward slash (@samp{/}) character
19994 are considered absolute, and the directory separator character is also
19998 Instruct @value{GDBN} to assume the target file system is DOS based.
19999 File names starting with either a forward slash, or a drive letter
20000 followed by a colon (e.g., @samp{c:}), are considered absolute, and
20001 both the slash (@samp{/}) and the backslash (@samp{\\}) characters are
20002 considered directory separators.
20005 Instruct @value{GDBN} to use the file system kind associated with the
20006 target operating system (@pxref{ABI, ,Configuring the Current ABI}).
20007 This is the default.
20011 @cindex file name canonicalization
20012 @cindex base name differences
20013 When processing file names provided by the user, @value{GDBN}
20014 frequently needs to compare them to the file names recorded in the
20015 program's debug info. Normally, @value{GDBN} compares just the
20016 @dfn{base names} of the files as strings, which is reasonably fast
20017 even for very large programs. (The base name of a file is the last
20018 portion of its name, after stripping all the leading directories.)
20019 This shortcut in comparison is based upon the assumption that files
20020 cannot have more than one base name. This is usually true, but
20021 references to files that use symlinks or similar filesystem
20022 facilities violate that assumption. If your program records files
20023 using such facilities, or if you provide file names to @value{GDBN}
20024 using symlinks etc., you can set @code{basenames-may-differ} to
20025 @code{true} to instruct @value{GDBN} to completely canonicalize each
20026 pair of file names it needs to compare. This will make file-name
20027 comparisons accurate, but at a price of a significant slowdown.
20030 @item set basenames-may-differ
20031 @kindex set basenames-may-differ
20032 Set whether a source file may have multiple base names.
20034 @item show basenames-may-differ
20035 @kindex show basenames-may-differ
20036 Show whether a source file may have multiple base names.
20040 @section File Caching
20041 @cindex caching of opened files
20042 @cindex caching of bfd objects
20044 To speed up file loading, and reduce memory usage, @value{GDBN} will
20045 reuse the @code{bfd} objects used to track open files. @xref{Top, ,
20046 BFD, bfd, The Binary File Descriptor Library}. The following commands
20047 allow visibility and control of the caching behavior.
20050 @kindex maint info bfds
20051 @item maint info bfds
20052 This prints information about each @code{bfd} object that is known to
20055 @kindex maint set bfd-sharing
20056 @kindex maint show bfd-sharing
20057 @kindex bfd caching
20058 @item maint set bfd-sharing
20059 @item maint show bfd-sharing
20060 Control whether @code{bfd} objects can be shared. When sharing is
20061 enabled @value{GDBN} reuses already open @code{bfd} objects rather
20062 than reopening the same file. Turning sharing off does not cause
20063 already shared @code{bfd} objects to be unshared, but all future files
20064 that are opened will create a new @code{bfd} object. Similarly,
20065 re-enabling sharing does not cause multiple existing @code{bfd}
20066 objects to be collapsed into a single shared @code{bfd} object.
20068 @kindex set debug bfd-cache @var{level}
20069 @kindex bfd caching
20070 @item set debug bfd-cache @var{level}
20071 Turns on debugging of the bfd cache, setting the level to @var{level}.
20073 @kindex show debug bfd-cache
20074 @kindex bfd caching
20075 @item show debug bfd-cache
20076 Show the current debugging level of the bfd cache.
20079 @node Separate Debug Files
20080 @section Debugging Information in Separate Files
20081 @cindex separate debugging information files
20082 @cindex debugging information in separate files
20083 @cindex @file{.debug} subdirectories
20084 @cindex debugging information directory, global
20085 @cindex global debugging information directories
20086 @cindex build ID, and separate debugging files
20087 @cindex @file{.build-id} directory
20089 @value{GDBN} allows you to put a program's debugging information in a
20090 file separate from the executable itself, in a way that allows
20091 @value{GDBN} to find and load the debugging information automatically.
20092 Since debugging information can be very large---sometimes larger
20093 than the executable code itself---some systems distribute debugging
20094 information for their executables in separate files, which users can
20095 install only when they need to debug a problem.
20097 @value{GDBN} supports two ways of specifying the separate debug info
20102 The executable contains a @dfn{debug link} that specifies the name of
20103 the separate debug info file. The separate debug file's name is
20104 usually @file{@var{executable}.debug}, where @var{executable} is the
20105 name of the corresponding executable file without leading directories
20106 (e.g., @file{ls.debug} for @file{/usr/bin/ls}). In addition, the
20107 debug link specifies a 32-bit @dfn{Cyclic Redundancy Check} (CRC)
20108 checksum for the debug file, which @value{GDBN} uses to validate that
20109 the executable and the debug file came from the same build.
20112 The executable contains a @dfn{build ID}, a unique bit string that is
20113 also present in the corresponding debug info file. (This is supported
20114 only on some operating systems, when using the ELF or PE file formats
20115 for binary files and the @sc{gnu} Binutils.) For more details about
20116 this feature, see the description of the @option{--build-id}
20117 command-line option in @ref{Options, , Command Line Options, ld,
20118 The GNU Linker}. The debug info file's name is not specified
20119 explicitly by the build ID, but can be computed from the build ID, see
20123 Depending on the way the debug info file is specified, @value{GDBN}
20124 uses two different methods of looking for the debug file:
20128 For the ``debug link'' method, @value{GDBN} looks up the named file in
20129 the directory of the executable file, then in a subdirectory of that
20130 directory named @file{.debug}, and finally under each one of the
20131 global debug directories, in a subdirectory whose name is identical to
20132 the leading directories of the executable's absolute file name. (On
20133 MS-Windows/MS-DOS, the drive letter of the executable's leading
20134 directories is converted to a one-letter subdirectory, i.e.@:
20135 @file{d:/usr/bin/} is converted to @file{/d/usr/bin/}, because Windows
20136 filesystems disallow colons in file names.)
20139 For the ``build ID'' method, @value{GDBN} looks in the
20140 @file{.build-id} subdirectory of each one of the global debug directories for
20141 a file named @file{@var{nn}/@var{nnnnnnnn}.debug}, where @var{nn} are the
20142 first 2 hex characters of the build ID bit string, and @var{nnnnnnnn}
20143 are the rest of the bit string. (Real build ID strings are 32 or more
20144 hex characters, not 10.)
20147 So, for example, suppose you ask @value{GDBN} to debug
20148 @file{/usr/bin/ls}, which has a debug link that specifies the
20149 file @file{ls.debug}, and a build ID whose value in hex is
20150 @code{abcdef1234}. If the list of the global debug directories includes
20151 @file{/usr/lib/debug}, then @value{GDBN} will look for the following
20152 debug information files, in the indicated order:
20156 @file{/usr/lib/debug/.build-id/ab/cdef1234.debug}
20158 @file{/usr/bin/ls.debug}
20160 @file{/usr/bin/.debug/ls.debug}
20162 @file{/usr/lib/debug/usr/bin/ls.debug}.
20165 @anchor{debug-file-directory}
20166 Global debugging info directories default to what is set by @value{GDBN}
20167 configure option @option{--with-separate-debug-dir}. During @value{GDBN} run
20168 you can also set the global debugging info directories, and view the list
20169 @value{GDBN} is currently using.
20173 @kindex set debug-file-directory
20174 @item set debug-file-directory @var{directories}
20175 Set the directories which @value{GDBN} searches for separate debugging
20176 information files to @var{directory}. Multiple path components can be set
20177 concatenating them by a path separator.
20179 @kindex show debug-file-directory
20180 @item show debug-file-directory
20181 Show the directories @value{GDBN} searches for separate debugging
20186 @cindex @code{.gnu_debuglink} sections
20187 @cindex debug link sections
20188 A debug link is a special section of the executable file named
20189 @code{.gnu_debuglink}. The section must contain:
20193 A filename, with any leading directory components removed, followed by
20196 zero to three bytes of padding, as needed to reach the next four-byte
20197 boundary within the section, and
20199 a four-byte CRC checksum, stored in the same endianness used for the
20200 executable file itself. The checksum is computed on the debugging
20201 information file's full contents by the function given below, passing
20202 zero as the @var{crc} argument.
20205 Any executable file format can carry a debug link, as long as it can
20206 contain a section named @code{.gnu_debuglink} with the contents
20209 @cindex @code{.note.gnu.build-id} sections
20210 @cindex build ID sections
20211 The build ID is a special section in the executable file (and in other
20212 ELF binary files that @value{GDBN} may consider). This section is
20213 often named @code{.note.gnu.build-id}, but that name is not mandatory.
20214 It contains unique identification for the built files---the ID remains
20215 the same across multiple builds of the same build tree. The default
20216 algorithm SHA1 produces 160 bits (40 hexadecimal characters) of the
20217 content for the build ID string. The same section with an identical
20218 value is present in the original built binary with symbols, in its
20219 stripped variant, and in the separate debugging information file.
20221 The debugging information file itself should be an ordinary
20222 executable, containing a full set of linker symbols, sections, and
20223 debugging information. The sections of the debugging information file
20224 should have the same names, addresses, and sizes as the original file,
20225 but they need not contain any data---much like a @code{.bss} section
20226 in an ordinary executable.
20228 The @sc{gnu} binary utilities (Binutils) package includes the
20229 @samp{objcopy} utility that can produce
20230 the separated executable / debugging information file pairs using the
20231 following commands:
20234 @kbd{objcopy --only-keep-debug foo foo.debug}
20239 These commands remove the debugging
20240 information from the executable file @file{foo} and place it in the file
20241 @file{foo.debug}. You can use the first, second or both methods to link the
20246 The debug link method needs the following additional command to also leave
20247 behind a debug link in @file{foo}:
20250 @kbd{objcopy --add-gnu-debuglink=foo.debug foo}
20253 Ulrich Drepper's @file{elfutils} package, starting with version 0.53, contains
20254 a version of the @code{strip} command such that the command @kbd{strip foo -f
20255 foo.debug} has the same functionality as the two @code{objcopy} commands and
20256 the @code{ln -s} command above, together.
20259 Build ID gets embedded into the main executable using @code{ld --build-id} or
20260 the @value{NGCC} counterpart @code{gcc -Wl,--build-id}. Build ID support plus
20261 compatibility fixes for debug files separation are present in @sc{gnu} binary
20262 utilities (Binutils) package since version 2.18.
20267 @cindex CRC algorithm definition
20268 The CRC used in @code{.gnu_debuglink} is the CRC-32 defined in
20269 IEEE 802.3 using the polynomial:
20271 @c TexInfo requires naked braces for multi-digit exponents for Tex
20272 @c output, but this causes HTML output to barf. HTML has to be set using
20273 @c raw commands. So we end up having to specify this equation in 2
20278 <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>
20279 + <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
20285 @math{x^{32} + x^{26} + x^{23} + x^{22} + x^{16} + x^{12} + x^{11}}
20286 @math{+ x^{10} + x^8 + x^7 + x^5 + x^4 + x^2 + x + 1}
20290 The function is computed byte at a time, taking the least
20291 significant bit of each byte first. The initial pattern
20292 @code{0xffffffff} is used, to ensure leading zeros affect the CRC and
20293 the final result is inverted to ensure trailing zeros also affect the
20296 @emph{Note:} This is the same CRC polynomial as used in handling the
20297 @dfn{Remote Serial Protocol} @code{qCRC} packet (@pxref{qCRC packet}).
20298 However in the case of the Remote Serial Protocol, the CRC is computed
20299 @emph{most} significant bit first, and the result is not inverted, so
20300 trailing zeros have no effect on the CRC value.
20302 To complete the description, we show below the code of the function
20303 which produces the CRC used in @code{.gnu_debuglink}. Inverting the
20304 initially supplied @code{crc} argument means that an initial call to
20305 this function passing in zero will start computing the CRC using
20308 @kindex gnu_debuglink_crc32
20311 gnu_debuglink_crc32 (unsigned long crc,
20312 unsigned char *buf, size_t len)
20314 static const unsigned long crc32_table[256] =
20316 0x00000000, 0x77073096, 0xee0e612c, 0x990951ba, 0x076dc419,
20317 0x706af48f, 0xe963a535, 0x9e6495a3, 0x0edb8832, 0x79dcb8a4,
20318 0xe0d5e91e, 0x97d2d988, 0x09b64c2b, 0x7eb17cbd, 0xe7b82d07,
20319 0x90bf1d91, 0x1db71064, 0x6ab020f2, 0xf3b97148, 0x84be41de,
20320 0x1adad47d, 0x6ddde4eb, 0xf4d4b551, 0x83d385c7, 0x136c9856,
20321 0x646ba8c0, 0xfd62f97a, 0x8a65c9ec, 0x14015c4f, 0x63066cd9,
20322 0xfa0f3d63, 0x8d080df5, 0x3b6e20c8, 0x4c69105e, 0xd56041e4,
20323 0xa2677172, 0x3c03e4d1, 0x4b04d447, 0xd20d85fd, 0xa50ab56b,
20324 0x35b5a8fa, 0x42b2986c, 0xdbbbc9d6, 0xacbcf940, 0x32d86ce3,
20325 0x45df5c75, 0xdcd60dcf, 0xabd13d59, 0x26d930ac, 0x51de003a,
20326 0xc8d75180, 0xbfd06116, 0x21b4f4b5, 0x56b3c423, 0xcfba9599,
20327 0xb8bda50f, 0x2802b89e, 0x5f058808, 0xc60cd9b2, 0xb10be924,
20328 0x2f6f7c87, 0x58684c11, 0xc1611dab, 0xb6662d3d, 0x76dc4190,
20329 0x01db7106, 0x98d220bc, 0xefd5102a, 0x71b18589, 0x06b6b51f,
20330 0x9fbfe4a5, 0xe8b8d433, 0x7807c9a2, 0x0f00f934, 0x9609a88e,
20331 0xe10e9818, 0x7f6a0dbb, 0x086d3d2d, 0x91646c97, 0xe6635c01,
20332 0x6b6b51f4, 0x1c6c6162, 0x856530d8, 0xf262004e, 0x6c0695ed,
20333 0x1b01a57b, 0x8208f4c1, 0xf50fc457, 0x65b0d9c6, 0x12b7e950,
20334 0x8bbeb8ea, 0xfcb9887c, 0x62dd1ddf, 0x15da2d49, 0x8cd37cf3,
20335 0xfbd44c65, 0x4db26158, 0x3ab551ce, 0xa3bc0074, 0xd4bb30e2,
20336 0x4adfa541, 0x3dd895d7, 0xa4d1c46d, 0xd3d6f4fb, 0x4369e96a,
20337 0x346ed9fc, 0xad678846, 0xda60b8d0, 0x44042d73, 0x33031de5,
20338 0xaa0a4c5f, 0xdd0d7cc9, 0x5005713c, 0x270241aa, 0xbe0b1010,
20339 0xc90c2086, 0x5768b525, 0x206f85b3, 0xb966d409, 0xce61e49f,
20340 0x5edef90e, 0x29d9c998, 0xb0d09822, 0xc7d7a8b4, 0x59b33d17,
20341 0x2eb40d81, 0xb7bd5c3b, 0xc0ba6cad, 0xedb88320, 0x9abfb3b6,
20342 0x03b6e20c, 0x74b1d29a, 0xead54739, 0x9dd277af, 0x04db2615,
20343 0x73dc1683, 0xe3630b12, 0x94643b84, 0x0d6d6a3e, 0x7a6a5aa8,
20344 0xe40ecf0b, 0x9309ff9d, 0x0a00ae27, 0x7d079eb1, 0xf00f9344,
20345 0x8708a3d2, 0x1e01f268, 0x6906c2fe, 0xf762575d, 0x806567cb,
20346 0x196c3671, 0x6e6b06e7, 0xfed41b76, 0x89d32be0, 0x10da7a5a,
20347 0x67dd4acc, 0xf9b9df6f, 0x8ebeeff9, 0x17b7be43, 0x60b08ed5,
20348 0xd6d6a3e8, 0xa1d1937e, 0x38d8c2c4, 0x4fdff252, 0xd1bb67f1,
20349 0xa6bc5767, 0x3fb506dd, 0x48b2364b, 0xd80d2bda, 0xaf0a1b4c,
20350 0x36034af6, 0x41047a60, 0xdf60efc3, 0xa867df55, 0x316e8eef,
20351 0x4669be79, 0xcb61b38c, 0xbc66831a, 0x256fd2a0, 0x5268e236,
20352 0xcc0c7795, 0xbb0b4703, 0x220216b9, 0x5505262f, 0xc5ba3bbe,
20353 0xb2bd0b28, 0x2bb45a92, 0x5cb36a04, 0xc2d7ffa7, 0xb5d0cf31,
20354 0x2cd99e8b, 0x5bdeae1d, 0x9b64c2b0, 0xec63f226, 0x756aa39c,
20355 0x026d930a, 0x9c0906a9, 0xeb0e363f, 0x72076785, 0x05005713,
20356 0x95bf4a82, 0xe2b87a14, 0x7bb12bae, 0x0cb61b38, 0x92d28e9b,
20357 0xe5d5be0d, 0x7cdcefb7, 0x0bdbdf21, 0x86d3d2d4, 0xf1d4e242,
20358 0x68ddb3f8, 0x1fda836e, 0x81be16cd, 0xf6b9265b, 0x6fb077e1,
20359 0x18b74777, 0x88085ae6, 0xff0f6a70, 0x66063bca, 0x11010b5c,
20360 0x8f659eff, 0xf862ae69, 0x616bffd3, 0x166ccf45, 0xa00ae278,
20361 0xd70dd2ee, 0x4e048354, 0x3903b3c2, 0xa7672661, 0xd06016f7,
20362 0x4969474d, 0x3e6e77db, 0xaed16a4a, 0xd9d65adc, 0x40df0b66,
20363 0x37d83bf0, 0xa9bcae53, 0xdebb9ec5, 0x47b2cf7f, 0x30b5ffe9,
20364 0xbdbdf21c, 0xcabac28a, 0x53b39330, 0x24b4a3a6, 0xbad03605,
20365 0xcdd70693, 0x54de5729, 0x23d967bf, 0xb3667a2e, 0xc4614ab8,
20366 0x5d681b02, 0x2a6f2b94, 0xb40bbe37, 0xc30c8ea1, 0x5a05df1b,
20369 unsigned char *end;
20371 crc = ~crc & 0xffffffff;
20372 for (end = buf + len; buf < end; ++buf)
20373 crc = crc32_table[(crc ^ *buf) & 0xff] ^ (crc >> 8);
20374 return ~crc & 0xffffffff;
20379 This computation does not apply to the ``build ID'' method.
20381 @node MiniDebugInfo
20382 @section Debugging information in a special section
20383 @cindex separate debug sections
20384 @cindex @samp{.gnu_debugdata} section
20386 Some systems ship pre-built executables and libraries that have a
20387 special @samp{.gnu_debugdata} section. This feature is called
20388 @dfn{MiniDebugInfo}. This section holds an LZMA-compressed object and
20389 is used to supply extra symbols for backtraces.
20391 The intent of this section is to provide extra minimal debugging
20392 information for use in simple backtraces. It is not intended to be a
20393 replacement for full separate debugging information (@pxref{Separate
20394 Debug Files}). The example below shows the intended use; however,
20395 @value{GDBN} does not currently put restrictions on what sort of
20396 debugging information might be included in the section.
20398 @value{GDBN} has support for this extension. If the section exists,
20399 then it is used provided that no other source of debugging information
20400 can be found, and that @value{GDBN} was configured with LZMA support.
20402 This section can be easily created using @command{objcopy} and other
20403 standard utilities:
20406 # Extract the dynamic symbols from the main binary, there is no need
20407 # to also have these in the normal symbol table.
20408 nm -D @var{binary} --format=posix --defined-only \
20409 | awk '@{ print $1 @}' | sort > dynsyms
20411 # Extract all the text (i.e. function) symbols from the debuginfo.
20412 # (Note that we actually also accept "D" symbols, for the benefit
20413 # of platforms like PowerPC64 that use function descriptors.)
20414 nm @var{binary} --format=posix --defined-only \
20415 | awk '@{ if ($2 == "T" || $2 == "t" || $2 == "D") print $1 @}' \
20418 # Keep all the function symbols not already in the dynamic symbol
20420 comm -13 dynsyms funcsyms > keep_symbols
20422 # Separate full debug info into debug binary.
20423 objcopy --only-keep-debug @var{binary} debug
20425 # Copy the full debuginfo, keeping only a minimal set of symbols and
20426 # removing some unnecessary sections.
20427 objcopy -S --remove-section .gdb_index --remove-section .comment \
20428 --keep-symbols=keep_symbols debug mini_debuginfo
20430 # Drop the full debug info from the original binary.
20431 strip --strip-all -R .comment @var{binary}
20433 # Inject the compressed data into the .gnu_debugdata section of the
20436 objcopy --add-section .gnu_debugdata=mini_debuginfo.xz @var{binary}
20440 @section Index Files Speed Up @value{GDBN}
20441 @cindex index files
20442 @cindex @samp{.gdb_index} section
20444 When @value{GDBN} finds a symbol file, it scans the symbols in the
20445 file in order to construct an internal symbol table. This lets most
20446 @value{GDBN} operations work quickly---at the cost of a delay early
20447 on. For large programs, this delay can be quite lengthy, so
20448 @value{GDBN} provides a way to build an index, which speeds up
20451 For convenience, @value{GDBN} comes with a program,
20452 @command{gdb-add-index}, which can be used to add the index to a
20453 symbol file. It takes the symbol file as its only argument:
20456 $ gdb-add-index symfile
20459 @xref{gdb-add-index}.
20461 It is also possible to do the work manually. Here is what
20462 @command{gdb-add-index} does behind the curtains.
20464 The index is stored as a section in the symbol file. @value{GDBN} can
20465 write the index to a file, then you can put it into the symbol file
20466 using @command{objcopy}.
20468 To create an index file, use the @code{save gdb-index} command:
20471 @item save gdb-index [-dwarf-5] @var{directory}
20472 @kindex save gdb-index
20473 Create index files for all symbol files currently known by
20474 @value{GDBN}. For each known @var{symbol-file}, this command by
20475 default creates it produces a single file
20476 @file{@var{symbol-file}.gdb-index}. If you invoke this command with
20477 the @option{-dwarf-5} option, it produces 2 files:
20478 @file{@var{symbol-file}.debug_names} and
20479 @file{@var{symbol-file}.debug_str}. The files are created in the
20480 given @var{directory}.
20483 Once you have created an index file you can merge it into your symbol
20484 file, here named @file{symfile}, using @command{objcopy}:
20487 $ objcopy --add-section .gdb_index=symfile.gdb-index \
20488 --set-section-flags .gdb_index=readonly symfile symfile
20491 Or for @code{-dwarf-5}:
20494 $ objcopy --dump-section .debug_str=symfile.debug_str.new symfile
20495 $ cat symfile.debug_str >>symfile.debug_str.new
20496 $ objcopy --add-section .debug_names=symfile.gdb-index \
20497 --set-section-flags .debug_names=readonly \
20498 --update-section .debug_str=symfile.debug_str.new symfile symfile
20501 @value{GDBN} will normally ignore older versions of @file{.gdb_index}
20502 sections that have been deprecated. Usually they are deprecated because
20503 they are missing a new feature or have performance issues.
20504 To tell @value{GDBN} to use a deprecated index section anyway
20505 specify @code{set use-deprecated-index-sections on}.
20506 The default is @code{off}.
20507 This can speed up startup, but may result in some functionality being lost.
20508 @xref{Index Section Format}.
20510 @emph{Warning:} Setting @code{use-deprecated-index-sections} to @code{on}
20511 must be done before gdb reads the file. The following will not work:
20514 $ gdb -ex "set use-deprecated-index-sections on" <program>
20517 Instead you must do, for example,
20520 $ gdb -iex "set use-deprecated-index-sections on" <program>
20523 There are currently some limitation on indices. They only work when
20524 for DWARF debugging information, not stabs. And, they do not
20525 currently work for programs using Ada.
20527 @subsection Automatic symbol index cache
20529 @cindex automatic symbol index cache
20530 It is possible for @value{GDBN} to automatically save a copy of this index in a
20531 cache on disk and retrieve it from there when loading the same binary in the
20532 future. This feature can be turned on with @kbd{set index-cache on}. The
20533 following commands can be used to tweak the behavior of the index cache.
20537 @kindex set index-cache
20538 @item set index-cache on
20539 @itemx set index-cache off
20540 Enable or disable the use of the symbol index cache.
20542 @item set index-cache directory @var{directory}
20543 @kindex show index-cache
20544 @itemx show index-cache directory
20545 Set/show the directory where index files will be saved.
20547 The default value for this directory depends on the host platform. On
20548 most systems, the index is cached in the @file{gdb} subdirectory of
20549 the directory pointed to by the @env{XDG_CACHE_HOME} environment
20550 variable, if it is defined, else in the @file{.cache/gdb} subdirectory
20551 of your home directory. However, on some systems, the default may
20552 differ according to local convention.
20554 There is no limit on the disk space used by index cache. It is perfectly safe
20555 to delete the content of that directory to free up disk space.
20557 @item show index-cache stats
20558 Print the number of cache hits and misses since the launch of @value{GDBN}.
20562 @node Symbol Errors
20563 @section Errors Reading Symbol Files
20565 While reading a symbol file, @value{GDBN} occasionally encounters problems,
20566 such as symbol types it does not recognize, or known bugs in compiler
20567 output. By default, @value{GDBN} does not notify you of such problems, since
20568 they are relatively common and primarily of interest to people
20569 debugging compilers. If you are interested in seeing information
20570 about ill-constructed symbol tables, you can either ask @value{GDBN} to print
20571 only one message about each such type of problem, no matter how many
20572 times the problem occurs; or you can ask @value{GDBN} to print more messages,
20573 to see how many times the problems occur, with the @code{set
20574 complaints} command (@pxref{Messages/Warnings, ,Optional Warnings and
20577 The messages currently printed, and their meanings, include:
20580 @item inner block not inside outer block in @var{symbol}
20582 The symbol information shows where symbol scopes begin and end
20583 (such as at the start of a function or a block of statements). This
20584 error indicates that an inner scope block is not fully contained
20585 in its outer scope blocks.
20587 @value{GDBN} circumvents the problem by treating the inner block as if it had
20588 the same scope as the outer block. In the error message, @var{symbol}
20589 may be shown as ``@code{(don't know)}'' if the outer block is not a
20592 @item block at @var{address} out of order
20594 The symbol information for symbol scope blocks should occur in
20595 order of increasing addresses. This error indicates that it does not
20598 @value{GDBN} does not circumvent this problem, and has trouble
20599 locating symbols in the source file whose symbols it is reading. (You
20600 can often determine what source file is affected by specifying
20601 @code{set verbose on}. @xref{Messages/Warnings, ,Optional Warnings and
20604 @item bad block start address patched
20606 The symbol information for a symbol scope block has a start address
20607 smaller than the address of the preceding source line. This is known
20608 to occur in the SunOS 4.1.1 (and earlier) C compiler.
20610 @value{GDBN} circumvents the problem by treating the symbol scope block as
20611 starting on the previous source line.
20613 @item bad string table offset in symbol @var{n}
20616 Symbol number @var{n} contains a pointer into the string table which is
20617 larger than the size of the string table.
20619 @value{GDBN} circumvents the problem by considering the symbol to have the
20620 name @code{foo}, which may cause other problems if many symbols end up
20623 @item unknown symbol type @code{0x@var{nn}}
20625 The symbol information contains new data types that @value{GDBN} does
20626 not yet know how to read. @code{0x@var{nn}} is the symbol type of the
20627 uncomprehended information, in hexadecimal.
20629 @value{GDBN} circumvents the error by ignoring this symbol information.
20630 This usually allows you to debug your program, though certain symbols
20631 are not accessible. If you encounter such a problem and feel like
20632 debugging it, you can debug @code{@value{GDBP}} with itself, breakpoint
20633 on @code{complain}, then go up to the function @code{read_dbx_symtab}
20634 and examine @code{*bufp} to see the symbol.
20636 @item stub type has NULL name
20638 @value{GDBN} could not find the full definition for a struct or class.
20640 @item const/volatile indicator missing (ok if using g++ v1.x), got@dots{}
20641 The symbol information for a C@t{++} member function is missing some
20642 information that recent versions of the compiler should have output for
20645 @item info mismatch between compiler and debugger
20647 @value{GDBN} could not parse a type specification output by the compiler.
20652 @section GDB Data Files
20654 @cindex prefix for data files
20655 @value{GDBN} will sometimes read an auxiliary data file. These files
20656 are kept in a directory known as the @dfn{data directory}.
20658 You can set the data directory's name, and view the name @value{GDBN}
20659 is currently using.
20662 @kindex set data-directory
20663 @item set data-directory @var{directory}
20664 Set the directory which @value{GDBN} searches for auxiliary data files
20665 to @var{directory}.
20667 @kindex show data-directory
20668 @item show data-directory
20669 Show the directory @value{GDBN} searches for auxiliary data files.
20672 @cindex default data directory
20673 @cindex @samp{--with-gdb-datadir}
20674 You can set the default data directory by using the configure-time
20675 @samp{--with-gdb-datadir} option. If the data directory is inside
20676 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
20677 @samp{--exec-prefix}), then the default data directory will be updated
20678 automatically if the installed @value{GDBN} is moved to a new
20681 The data directory may also be specified with the
20682 @code{--data-directory} command line option.
20683 @xref{Mode Options}.
20686 @chapter Specifying a Debugging Target
20688 @cindex debugging target
20689 A @dfn{target} is the execution environment occupied by your program.
20691 Often, @value{GDBN} runs in the same host environment as your program;
20692 in that case, the debugging target is specified as a side effect when
20693 you use the @code{file} or @code{core} commands. When you need more
20694 flexibility---for example, running @value{GDBN} on a physically separate
20695 host, or controlling a standalone system over a serial port or a
20696 realtime system over a TCP/IP connection---you can use the @code{target}
20697 command to specify one of the target types configured for @value{GDBN}
20698 (@pxref{Target Commands, ,Commands for Managing Targets}).
20700 @cindex target architecture
20701 It is possible to build @value{GDBN} for several different @dfn{target
20702 architectures}. When @value{GDBN} is built like that, you can choose
20703 one of the available architectures with the @kbd{set architecture}
20707 @kindex set architecture
20708 @kindex show architecture
20709 @item set architecture @var{arch}
20710 This command sets the current target architecture to @var{arch}. The
20711 value of @var{arch} can be @code{"auto"}, in addition to one of the
20712 supported architectures.
20714 @item show architecture
20715 Show the current target architecture.
20717 @item set processor
20719 @kindex set processor
20720 @kindex show processor
20721 These are alias commands for, respectively, @code{set architecture}
20722 and @code{show architecture}.
20726 * Active Targets:: Active targets
20727 * Target Commands:: Commands for managing targets
20728 * Byte Order:: Choosing target byte order
20731 @node Active Targets
20732 @section Active Targets
20734 @cindex stacking targets
20735 @cindex active targets
20736 @cindex multiple targets
20738 There are multiple classes of targets such as: processes, executable files or
20739 recording sessions. Core files belong to the process class, making core file
20740 and process mutually exclusive. Otherwise, @value{GDBN} can work concurrently
20741 on multiple active targets, one in each class. This allows you to (for
20742 example) start a process and inspect its activity, while still having access to
20743 the executable file after the process finishes. Or if you start process
20744 recording (@pxref{Reverse Execution}) and @code{reverse-step} there, you are
20745 presented a virtual layer of the recording target, while the process target
20746 remains stopped at the chronologically last point of the process execution.
20748 Use the @code{core-file} and @code{exec-file} commands to select a new core
20749 file or executable target (@pxref{Files, ,Commands to Specify Files}). To
20750 specify as a target a process that is already running, use the @code{attach}
20751 command (@pxref{Attach, ,Debugging an Already-running Process}).
20753 @node Target Commands
20754 @section Commands for Managing Targets
20757 @item target @var{type} @var{parameters}
20758 Connects the @value{GDBN} host environment to a target machine or
20759 process. A target is typically a protocol for talking to debugging
20760 facilities. You use the argument @var{type} to specify the type or
20761 protocol of the target machine.
20763 Further @var{parameters} are interpreted by the target protocol, but
20764 typically include things like device names or host names to connect
20765 with, process numbers, and baud rates.
20767 The @code{target} command does not repeat if you press @key{RET} again
20768 after executing the command.
20770 @kindex help target
20772 Displays the names of all targets available. To display targets
20773 currently selected, use either @code{info target} or @code{info files}
20774 (@pxref{Files, ,Commands to Specify Files}).
20776 @item help target @var{name}
20777 Describe a particular target, including any parameters necessary to
20780 @kindex set gnutarget
20781 @item set gnutarget @var{args}
20782 @value{GDBN} uses its own library BFD to read your files. @value{GDBN}
20783 knows whether it is reading an @dfn{executable},
20784 a @dfn{core}, or a @dfn{.o} file; however, you can specify the file format
20785 with the @code{set gnutarget} command. Unlike most @code{target} commands,
20786 with @code{gnutarget} the @code{target} refers to a program, not a machine.
20789 @emph{Warning:} To specify a file format with @code{set gnutarget},
20790 you must know the actual BFD name.
20794 @xref{Files, , Commands to Specify Files}.
20796 @kindex show gnutarget
20797 @item show gnutarget
20798 Use the @code{show gnutarget} command to display what file format
20799 @code{gnutarget} is set to read. If you have not set @code{gnutarget},
20800 @value{GDBN} will determine the file format for each file automatically,
20801 and @code{show gnutarget} displays @samp{The current BFD target is "auto"}.
20804 @cindex common targets
20805 Here are some common targets (available, or not, depending on the GDB
20810 @item target exec @var{program}
20811 @cindex executable file target
20812 An executable file. @samp{target exec @var{program}} is the same as
20813 @samp{exec-file @var{program}}.
20815 @item target core @var{filename}
20816 @cindex core dump file target
20817 A core dump file. @samp{target core @var{filename}} is the same as
20818 @samp{core-file @var{filename}}.
20820 @item target remote @var{medium}
20821 @cindex remote target
20822 A remote system connected to @value{GDBN} via a serial line or network
20823 connection. This command tells @value{GDBN} to use its own remote
20824 protocol over @var{medium} for debugging. @xref{Remote Debugging}.
20826 For example, if you have a board connected to @file{/dev/ttya} on the
20827 machine running @value{GDBN}, you could say:
20830 target remote /dev/ttya
20833 @code{target remote} supports the @code{load} command. This is only
20834 useful if you have some other way of getting the stub to the target
20835 system, and you can put it somewhere in memory where it won't get
20836 clobbered by the download.
20838 @item target sim @r{[}@var{simargs}@r{]} @dots{}
20839 @cindex built-in simulator target
20840 Builtin CPU simulator. @value{GDBN} includes simulators for most architectures.
20848 works; however, you cannot assume that a specific memory map, device
20849 drivers, or even basic I/O is available, although some simulators do
20850 provide these. For info about any processor-specific simulator details,
20851 see the appropriate section in @ref{Embedded Processors, ,Embedded
20854 @item target native
20855 @cindex native target
20856 Setup for local/native process debugging. Useful to make the
20857 @code{run} command spawn native processes (likewise @code{attach},
20858 etc.@:) even when @code{set auto-connect-native-target} is @code{off}
20859 (@pxref{set auto-connect-native-target}).
20863 Different targets are available on different configurations of @value{GDBN};
20864 your configuration may have more or fewer targets.
20866 Many remote targets require you to download the executable's code once
20867 you've successfully established a connection. You may wish to control
20868 various aspects of this process.
20873 @kindex set hash@r{, for remote monitors}
20874 @cindex hash mark while downloading
20875 This command controls whether a hash mark @samp{#} is displayed while
20876 downloading a file to the remote monitor. If on, a hash mark is
20877 displayed after each S-record is successfully downloaded to the
20881 @kindex show hash@r{, for remote monitors}
20882 Show the current status of displaying the hash mark.
20884 @item set debug monitor
20885 @kindex set debug monitor
20886 @cindex display remote monitor communications
20887 Enable or disable display of communications messages between
20888 @value{GDBN} and the remote monitor.
20890 @item show debug monitor
20891 @kindex show debug monitor
20892 Show the current status of displaying communications between
20893 @value{GDBN} and the remote monitor.
20898 @kindex load @var{filename} @var{offset}
20899 @item load @var{filename} @var{offset}
20901 Depending on what remote debugging facilities are configured into
20902 @value{GDBN}, the @code{load} command may be available. Where it exists, it
20903 is meant to make @var{filename} (an executable) available for debugging
20904 on the remote system---by downloading, or dynamic linking, for example.
20905 @code{load} also records the @var{filename} symbol table in @value{GDBN}, like
20906 the @code{add-symbol-file} command.
20908 If your @value{GDBN} does not have a @code{load} command, attempting to
20909 execute it gets the error message ``@code{You can't do that when your
20910 target is @dots{}}''
20912 The file is loaded at whatever address is specified in the executable.
20913 For some object file formats, you can specify the load address when you
20914 link the program; for other formats, like a.out, the object file format
20915 specifies a fixed address.
20916 @c FIXME! This would be a good place for an xref to the GNU linker doc.
20918 It is also possible to tell @value{GDBN} to load the executable file at a
20919 specific offset described by the optional argument @var{offset}. When
20920 @var{offset} is provided, @var{filename} must also be provided.
20922 Depending on the remote side capabilities, @value{GDBN} may be able to
20923 load programs into flash memory.
20925 @code{load} does not repeat if you press @key{RET} again after using it.
20930 @kindex flash-erase
20932 @anchor{flash-erase}
20934 Erases all known flash memory regions on the target.
20939 @section Choosing Target Byte Order
20941 @cindex choosing target byte order
20942 @cindex target byte order
20944 Some types of processors, such as the @acronym{MIPS}, PowerPC, and Renesas SH,
20945 offer the ability to run either big-endian or little-endian byte
20946 orders. Usually the executable or symbol will include a bit to
20947 designate the endian-ness, and you will not need to worry about
20948 which to use. However, you may still find it useful to adjust
20949 @value{GDBN}'s idea of processor endian-ness manually.
20953 @item set endian big
20954 Instruct @value{GDBN} to assume the target is big-endian.
20956 @item set endian little
20957 Instruct @value{GDBN} to assume the target is little-endian.
20959 @item set endian auto
20960 Instruct @value{GDBN} to use the byte order associated with the
20964 Display @value{GDBN}'s current idea of the target byte order.
20968 If the @code{set endian auto} mode is in effect and no executable has
20969 been selected, then the endianness used is the last one chosen either
20970 by one of the @code{set endian big} and @code{set endian little}
20971 commands or by inferring from the last executable used. If no
20972 endianness has been previously chosen, then the default for this mode
20973 is inferred from the target @value{GDBN} has been built for, and is
20974 @code{little} if the name of the target CPU has an @code{el} suffix
20975 and @code{big} otherwise.
20977 Note that these commands merely adjust interpretation of symbolic
20978 data on the host, and that they have absolutely no effect on the
20982 @node Remote Debugging
20983 @chapter Debugging Remote Programs
20984 @cindex remote debugging
20986 If you are trying to debug a program running on a machine that cannot run
20987 @value{GDBN} in the usual way, it is often useful to use remote debugging.
20988 For example, you might use remote debugging on an operating system kernel,
20989 or on a small system which does not have a general purpose operating system
20990 powerful enough to run a full-featured debugger.
20992 Some configurations of @value{GDBN} have special serial or TCP/IP interfaces
20993 to make this work with particular debugging targets. In addition,
20994 @value{GDBN} comes with a generic serial protocol (specific to @value{GDBN},
20995 but not specific to any particular target system) which you can use if you
20996 write the remote stubs---the code that runs on the remote system to
20997 communicate with @value{GDBN}.
20999 Other remote targets may be available in your
21000 configuration of @value{GDBN}; use @code{help target} to list them.
21003 * Connecting:: Connecting to a remote target
21004 * File Transfer:: Sending files to a remote system
21005 * Server:: Using the gdbserver program
21006 * Remote Configuration:: Remote configuration
21007 * Remote Stub:: Implementing a remote stub
21011 @section Connecting to a Remote Target
21012 @cindex remote debugging, connecting
21013 @cindex @code{gdbserver}, connecting
21014 @cindex remote debugging, types of connections
21015 @cindex @code{gdbserver}, types of connections
21016 @cindex @code{gdbserver}, @code{target remote} mode
21017 @cindex @code{gdbserver}, @code{target extended-remote} mode
21019 This section describes how to connect to a remote target, including the
21020 types of connections and their differences, how to set up executable and
21021 symbol files on the host and target, and the commands used for
21022 connecting to and disconnecting from the remote target.
21024 @subsection Types of Remote Connections
21026 @value{GDBN} supports two types of remote connections, @code{target remote}
21027 mode and @code{target extended-remote} mode. Note that many remote targets
21028 support only @code{target remote} mode. There are several major
21029 differences between the two types of connections, enumerated here:
21033 @cindex remote debugging, detach and program exit
21034 @item Result of detach or program exit
21035 @strong{With target remote mode:} When the debugged program exits or you
21036 detach from it, @value{GDBN} disconnects from the target. When using
21037 @code{gdbserver}, @code{gdbserver} will exit.
21039 @strong{With target extended-remote mode:} When the debugged program exits or
21040 you detach from it, @value{GDBN} remains connected to the target, even
21041 though no program is running. You can rerun the program, attach to a
21042 running program, or use @code{monitor} commands specific to the target.
21044 When using @code{gdbserver} in this case, it does not exit unless it was
21045 invoked using the @option{--once} option. If the @option{--once} option
21046 was not used, you can ask @code{gdbserver} to exit using the
21047 @code{monitor exit} command (@pxref{Monitor Commands for gdbserver}).
21049 @item Specifying the program to debug
21050 For both connection types you use the @code{file} command to specify the
21051 program on the host system. If you are using @code{gdbserver} there are
21052 some differences in how to specify the location of the program on the
21055 @strong{With target remote mode:} You must either specify the program to debug
21056 on the @code{gdbserver} command line or use the @option{--attach} option
21057 (@pxref{Attaching to a program,,Attaching to a Running Program}).
21059 @cindex @option{--multi}, @code{gdbserver} option
21060 @strong{With target extended-remote mode:} You may specify the program to debug
21061 on the @code{gdbserver} command line, or you can load the program or attach
21062 to it using @value{GDBN} commands after connecting to @code{gdbserver}.
21064 @anchor{--multi Option in Types of Remote Connnections}
21065 You can start @code{gdbserver} without supplying an initial command to run
21066 or process ID to attach. To do this, use the @option{--multi} command line
21067 option. Then you can connect using @code{target extended-remote} and start
21068 the program you want to debug (see below for details on using the
21069 @code{run} command in this scenario). Note that the conditions under which
21070 @code{gdbserver} terminates depend on how @value{GDBN} connects to it
21071 (@code{target remote} or @code{target extended-remote}). The
21072 @option{--multi} option to @code{gdbserver} has no influence on that.
21074 @item The @code{run} command
21075 @strong{With target remote mode:} The @code{run} command is not
21076 supported. Once a connection has been established, you can use all
21077 the usual @value{GDBN} commands to examine and change data. The
21078 remote program is already running, so you can use commands like
21079 @kbd{step} and @kbd{continue}.
21081 @strong{With target extended-remote mode:} The @code{run} command is
21082 supported. The @code{run} command uses the value set by
21083 @code{set remote exec-file} (@pxref{set remote exec-file}) to select
21084 the program to run. Command line arguments are supported, except for
21085 wildcard expansion and I/O redirection (@pxref{Arguments}).
21087 If you specify the program to debug on the command line, then the
21088 @code{run} command is not required to start execution, and you can
21089 resume using commands like @kbd{step} and @kbd{continue} as with
21090 @code{target remote} mode.
21092 @anchor{Attaching in Types of Remote Connections}
21094 @strong{With target remote mode:} The @value{GDBN} command @code{attach} is
21095 not supported. To attach to a running program using @code{gdbserver}, you
21096 must use the @option{--attach} option (@pxref{Running gdbserver}).
21098 @strong{With target extended-remote mode:} To attach to a running program,
21099 you may use the @code{attach} command after the connection has been
21100 established. If you are using @code{gdbserver}, you may also invoke
21101 @code{gdbserver} using the @option{--attach} option
21102 (@pxref{Running gdbserver}).
21106 @anchor{Host and target files}
21107 @subsection Host and Target Files
21108 @cindex remote debugging, symbol files
21109 @cindex symbol files, remote debugging
21111 @value{GDBN}, running on the host, needs access to symbol and debugging
21112 information for your program running on the target. This requires
21113 access to an unstripped copy of your program, and possibly any associated
21114 symbol files. Note that this section applies equally to both @code{target
21115 remote} mode and @code{target extended-remote} mode.
21117 Some remote targets (@pxref{qXfer executable filename read}, and
21118 @pxref{Host I/O Packets}) allow @value{GDBN} to access program files over
21119 the same connection used to communicate with @value{GDBN}. With such a
21120 target, if the remote program is unstripped, the only command you need is
21121 @code{target remote} (or @code{target extended-remote}).
21123 If the remote program is stripped, or the target does not support remote
21124 program file access, start up @value{GDBN} using the name of the local
21125 unstripped copy of your program as the first argument, or use the
21126 @code{file} command. Use @code{set sysroot} to specify the location (on
21127 the host) of target libraries (unless your @value{GDBN} was compiled with
21128 the correct sysroot using @code{--with-sysroot}). Alternatively, you
21129 may use @code{set solib-search-path} to specify how @value{GDBN} locates
21132 The symbol file and target libraries must exactly match the executable
21133 and libraries on the target, with one exception: the files on the host
21134 system should not be stripped, even if the files on the target system
21135 are. Mismatched or missing files will lead to confusing results
21136 during debugging. On @sc{gnu}/Linux targets, mismatched or missing
21137 files may also prevent @code{gdbserver} from debugging multi-threaded
21140 @subsection Remote Connection Commands
21141 @cindex remote connection commands
21142 @value{GDBN} can communicate with the target over a serial line, a
21143 local Unix domain socket, or
21144 over an @acronym{IP} network using @acronym{TCP} or @acronym{UDP}. In
21145 each case, @value{GDBN} uses the same protocol for debugging your
21146 program; only the medium carrying the debugging packets varies. The
21147 @code{target remote} and @code{target extended-remote} commands
21148 establish a connection to the target. Both commands accept the same
21149 arguments, which indicate the medium to use:
21153 @item target remote @var{serial-device}
21154 @itemx target extended-remote @var{serial-device}
21155 @cindex serial line, @code{target remote}
21156 Use @var{serial-device} to communicate with the target. For example,
21157 to use a serial line connected to the device named @file{/dev/ttyb}:
21160 target remote /dev/ttyb
21163 If you're using a serial line, you may want to give @value{GDBN} the
21164 @samp{--baud} option, or use the @code{set serial baud} command
21165 (@pxref{Remote Configuration, set serial baud}) before the
21166 @code{target} command.
21168 @item target remote @var{local-socket}
21169 @itemx target extended-remote @var{local-socket}
21170 @cindex local socket, @code{target remote}
21171 @cindex Unix domain socket
21172 Use @var{local-socket} to communicate with the target. For example,
21173 to use a local Unix domain socket bound to the file system entry @file{/tmp/gdb-socket0}:
21176 target remote /tmp/gdb-socket0
21179 Note that this command has the same form as the command to connect
21180 to a serial line. @value{GDBN} will automatically determine which
21181 kind of file you have specified and will make the appropriate kind
21183 This feature is not available if the host system does not support
21184 Unix domain sockets.
21186 @item target remote @code{@var{host}:@var{port}}
21187 @itemx target remote @code{@var{[host]}:@var{port}}
21188 @itemx target remote @code{tcp:@var{host}:@var{port}}
21189 @itemx target remote @code{tcp:@var{[host]}:@var{port}}
21190 @itemx target remote @code{tcp4:@var{host}:@var{port}}
21191 @itemx target remote @code{tcp6:@var{host}:@var{port}}
21192 @itemx target remote @code{tcp6:@var{[host]}:@var{port}}
21193 @itemx target extended-remote @code{@var{host}:@var{port}}
21194 @itemx target extended-remote @code{@var{[host]}:@var{port}}
21195 @itemx target extended-remote @code{tcp:@var{host}:@var{port}}
21196 @itemx target extended-remote @code{tcp:@var{[host]}:@var{port}}
21197 @itemx target extended-remote @code{tcp4:@var{host}:@var{port}}
21198 @itemx target extended-remote @code{tcp6:@var{host}:@var{port}}
21199 @itemx target extended-remote @code{tcp6:@var{[host]}:@var{port}}
21200 @cindex @acronym{TCP} port, @code{target remote}
21201 Debug using a @acronym{TCP} connection to @var{port} on @var{host}.
21202 The @var{host} may be either a host name, a numeric @acronym{IPv4}
21203 address, or a numeric @acronym{IPv6} address (with or without the
21204 square brackets to separate the address from the port); @var{port}
21205 must be a decimal number. The @var{host} could be the target machine
21206 itself, if it is directly connected to the net, or it might be a
21207 terminal server which in turn has a serial line to the target.
21209 For example, to connect to port 2828 on a terminal server named
21213 target remote manyfarms:2828
21216 To connect to port 2828 on a terminal server whose address is
21217 @code{2001:0db8:85a3:0000:0000:8a2e:0370:7334}, you can either use the
21218 square bracket syntax:
21221 target remote [2001:0db8:85a3:0000:0000:8a2e:0370:7334]:2828
21225 or explicitly specify the @acronym{IPv6} protocol:
21228 target remote tcp6:2001:0db8:85a3:0000:0000:8a2e:0370:7334:2828
21231 This last example may be confusing to the reader, because there is no
21232 visible separation between the hostname and the port number.
21233 Therefore, we recommend the user to provide @acronym{IPv6} addresses
21234 using square brackets for clarity. However, it is important to
21235 mention that for @value{GDBN} there is no ambiguity: the number after
21236 the last colon is considered to be the port number.
21238 If your remote target is actually running on the same machine as your
21239 debugger session (e.g.@: a simulator for your target running on the
21240 same host), you can omit the hostname. For example, to connect to
21241 port 1234 on your local machine:
21244 target remote :1234
21248 Note that the colon is still required here.
21250 @item target remote @code{udp:@var{host}:@var{port}}
21251 @itemx target remote @code{udp:@var{[host]}:@var{port}}
21252 @itemx target remote @code{udp4:@var{host}:@var{port}}
21253 @itemx target remote @code{udp6:@var{[host]}:@var{port}}
21254 @itemx target extended-remote @code{udp:@var{host}:@var{port}}
21255 @itemx target extended-remote @code{udp:@var{host}:@var{port}}
21256 @itemx target extended-remote @code{udp:@var{[host]}:@var{port}}
21257 @itemx target extended-remote @code{udp4:@var{host}:@var{port}}
21258 @itemx target extended-remote @code{udp6:@var{host}:@var{port}}
21259 @itemx target extended-remote @code{udp6:@var{[host]}:@var{port}}
21260 @cindex @acronym{UDP} port, @code{target remote}
21261 Debug using @acronym{UDP} packets to @var{port} on @var{host}. For example, to
21262 connect to @acronym{UDP} port 2828 on a terminal server named @code{manyfarms}:
21265 target remote udp:manyfarms:2828
21268 When using a @acronym{UDP} connection for remote debugging, you should
21269 keep in mind that the `U' stands for ``Unreliable''. @acronym{UDP}
21270 can silently drop packets on busy or unreliable networks, which will
21271 cause havoc with your debugging session.
21273 @item target remote | @var{command}
21274 @itemx target extended-remote | @var{command}
21275 @cindex pipe, @code{target remote} to
21276 Run @var{command} in the background and communicate with it using a
21277 pipe. The @var{command} is a shell command, to be parsed and expanded
21278 by the system's command shell, @code{/bin/sh}; it should expect remote
21279 protocol packets on its standard input, and send replies on its
21280 standard output. You could use this to run a stand-alone simulator
21281 that speaks the remote debugging protocol, to make net connections
21282 using programs like @code{ssh}, or for other similar tricks.
21284 If @var{command} closes its standard output (perhaps by exiting),
21285 @value{GDBN} will try to send it a @code{SIGTERM} signal. (If the
21286 program has already exited, this will have no effect.)
21290 @cindex interrupting remote programs
21291 @cindex remote programs, interrupting
21292 Whenever @value{GDBN} is waiting for the remote program, if you type the
21293 interrupt character (often @kbd{Ctrl-c}), @value{GDBN} attempts to stop the
21294 program. This may or may not succeed, depending in part on the hardware
21295 and the serial drivers the remote system uses. If you type the
21296 interrupt character once again, @value{GDBN} displays this prompt:
21299 Interrupted while waiting for the program.
21300 Give up (and stop debugging it)? (y or n)
21303 In @code{target remote} mode, if you type @kbd{y}, @value{GDBN} abandons
21304 the remote debugging session. (If you decide you want to try again later,
21305 you can use @kbd{target remote} again to connect once more.) If you type
21306 @kbd{n}, @value{GDBN} goes back to waiting.
21308 In @code{target extended-remote} mode, typing @kbd{n} will leave
21309 @value{GDBN} connected to the target.
21312 @kindex detach (remote)
21314 When you have finished debugging the remote program, you can use the
21315 @code{detach} command to release it from @value{GDBN} control.
21316 Detaching from the target normally resumes its execution, but the results
21317 will depend on your particular remote stub. After the @code{detach}
21318 command in @code{target remote} mode, @value{GDBN} is free to connect to
21319 another target. In @code{target extended-remote} mode, @value{GDBN} is
21320 still connected to the target.
21324 The @code{disconnect} command closes the connection to the target, and
21325 the target is generally not resumed. It will wait for @value{GDBN}
21326 (this instance or another one) to connect and continue debugging. After
21327 the @code{disconnect} command, @value{GDBN} is again free to connect to
21330 @cindex send command to remote monitor
21331 @cindex extend @value{GDBN} for remote targets
21332 @cindex add new commands for external monitor
21334 @item monitor @var{cmd}
21335 This command allows you to send arbitrary commands directly to the
21336 remote monitor. Since @value{GDBN} doesn't care about the commands it
21337 sends like this, this command is the way to extend @value{GDBN}---you
21338 can add new commands that only the external monitor will understand
21342 @node File Transfer
21343 @section Sending files to a remote system
21344 @cindex remote target, file transfer
21345 @cindex file transfer
21346 @cindex sending files to remote systems
21348 Some remote targets offer the ability to transfer files over the same
21349 connection used to communicate with @value{GDBN}. This is convenient
21350 for targets accessible through other means, e.g.@: @sc{gnu}/Linux systems
21351 running @code{gdbserver} over a network interface. For other targets,
21352 e.g.@: embedded devices with only a single serial port, this may be
21353 the only way to upload or download files.
21355 Not all remote targets support these commands.
21359 @item remote put @var{hostfile} @var{targetfile}
21360 Copy file @var{hostfile} from the host system (the machine running
21361 @value{GDBN}) to @var{targetfile} on the target system.
21364 @item remote get @var{targetfile} @var{hostfile}
21365 Copy file @var{targetfile} from the target system to @var{hostfile}
21366 on the host system.
21368 @kindex remote delete
21369 @item remote delete @var{targetfile}
21370 Delete @var{targetfile} from the target system.
21375 @section Using the @code{gdbserver} Program
21378 @cindex remote connection without stubs
21379 @code{gdbserver} is a control program for Unix-like systems, which
21380 allows you to connect your program with a remote @value{GDBN} via
21381 @code{target remote} or @code{target extended-remote}---but without
21382 linking in the usual debugging stub.
21384 @code{gdbserver} is not a complete replacement for the debugging stubs,
21385 because it requires essentially the same operating-system facilities
21386 that @value{GDBN} itself does. In fact, a system that can run
21387 @code{gdbserver} to connect to a remote @value{GDBN} could also run
21388 @value{GDBN} locally! @code{gdbserver} is sometimes useful nevertheless,
21389 because it is a much smaller program than @value{GDBN} itself. It is
21390 also easier to port than all of @value{GDBN}, so you may be able to get
21391 started more quickly on a new system by using @code{gdbserver}.
21392 Finally, if you develop code for real-time systems, you may find that
21393 the tradeoffs involved in real-time operation make it more convenient to
21394 do as much development work as possible on another system, for example
21395 by cross-compiling. You can use @code{gdbserver} to make a similar
21396 choice for debugging.
21398 @value{GDBN} and @code{gdbserver} communicate via either a serial line
21399 or a TCP connection, using the standard @value{GDBN} remote serial
21403 @emph{Warning:} @code{gdbserver} does not have any built-in security.
21404 Do not run @code{gdbserver} connected to any public network; a
21405 @value{GDBN} connection to @code{gdbserver} provides access to the
21406 target system with the same privileges as the user running
21410 @anchor{Running gdbserver}
21411 @subsection Running @code{gdbserver}
21412 @cindex arguments, to @code{gdbserver}
21413 @cindex @code{gdbserver}, command-line arguments
21415 Run @code{gdbserver} on the target system. You need a copy of the
21416 program you want to debug, including any libraries it requires.
21417 @code{gdbserver} does not need your program's symbol table, so you can
21418 strip the program if necessary to save space. @value{GDBN} on the host
21419 system does all the symbol handling.
21421 To use the server, you must tell it how to communicate with @value{GDBN};
21422 the name of your program; and the arguments for your program. The usual
21426 target> gdbserver @var{comm} @var{program} [ @var{args} @dots{} ]
21429 @var{comm} is either a device name (to use a serial line), or a TCP
21430 hostname and portnumber, or @code{-} or @code{stdio} to use
21431 stdin/stdout of @code{gdbserver}.
21432 For example, to debug Emacs with the argument
21433 @samp{foo.txt} and communicate with @value{GDBN} over the serial port
21437 target> gdbserver /dev/com1 emacs foo.txt
21440 @code{gdbserver} waits passively for the host @value{GDBN} to communicate
21443 To use a TCP connection instead of a serial line:
21446 target> gdbserver host:2345 emacs foo.txt
21449 The only difference from the previous example is the first argument,
21450 specifying that you are communicating with the host @value{GDBN} via
21451 TCP. The @samp{host:2345} argument means that @code{gdbserver} is to
21452 expect a TCP connection from machine @samp{host} to local TCP port 2345.
21453 (Currently, the @samp{host} part is ignored.) You can choose any number
21454 you want for the port number as long as it does not conflict with any
21455 TCP ports already in use on the target system (for example, @code{23} is
21456 reserved for @code{telnet}).@footnote{If you choose a port number that
21457 conflicts with another service, @code{gdbserver} prints an error message
21458 and exits.} You must use the same port number with the host @value{GDBN}
21459 @code{target remote} command.
21461 The @code{stdio} connection is useful when starting @code{gdbserver}
21465 (gdb) target remote | ssh -T hostname gdbserver - hello
21468 The @samp{-T} option to ssh is provided because we don't need a remote pty,
21469 and we don't want escape-character handling. Ssh does this by default when
21470 a command is provided, the flag is provided to make it explicit.
21471 You could elide it if you want to.
21473 Programs started with stdio-connected gdbserver have @file{/dev/null} for
21474 @code{stdin}, and @code{stdout},@code{stderr} are sent back to gdb for
21475 display through a pipe connected to gdbserver.
21476 Both @code{stdout} and @code{stderr} use the same pipe.
21478 @anchor{Attaching to a program}
21479 @subsubsection Attaching to a Running Program
21480 @cindex attach to a program, @code{gdbserver}
21481 @cindex @option{--attach}, @code{gdbserver} option
21483 On some targets, @code{gdbserver} can also attach to running programs.
21484 This is accomplished via the @code{--attach} argument. The syntax is:
21487 target> gdbserver --attach @var{comm} @var{pid}
21490 @var{pid} is the process ID of a currently running process. It isn't
21491 necessary to point @code{gdbserver} at a binary for the running process.
21493 In @code{target extended-remote} mode, you can also attach using the
21494 @value{GDBN} attach command
21495 (@pxref{Attaching in Types of Remote Connections}).
21498 You can debug processes by name instead of process ID if your target has the
21499 @code{pidof} utility:
21502 target> gdbserver --attach @var{comm} `pidof @var{program}`
21505 In case more than one copy of @var{program} is running, or @var{program}
21506 has multiple threads, most versions of @code{pidof} support the
21507 @code{-s} option to only return the first process ID.
21509 @subsubsection TCP port allocation lifecycle of @code{gdbserver}
21511 This section applies only when @code{gdbserver} is run to listen on a TCP
21514 @code{gdbserver} normally terminates after all of its debugged processes have
21515 terminated in @kbd{target remote} mode. On the other hand, for @kbd{target
21516 extended-remote}, @code{gdbserver} stays running even with no processes left.
21517 @value{GDBN} normally terminates the spawned debugged process on its exit,
21518 which normally also terminates @code{gdbserver} in the @kbd{target remote}
21519 mode. Therefore, when the connection drops unexpectedly, and @value{GDBN}
21520 cannot ask @code{gdbserver} to kill its debugged processes, @code{gdbserver}
21521 stays running even in the @kbd{target remote} mode.
21523 When @code{gdbserver} stays running, @value{GDBN} can connect to it again later.
21524 Such reconnecting is useful for features like @ref{disconnected tracing}. For
21525 completeness, at most one @value{GDBN} can be connected at a time.
21527 @cindex @option{--once}, @code{gdbserver} option
21528 By default, @code{gdbserver} keeps the listening TCP port open, so that
21529 subsequent connections are possible. However, if you start @code{gdbserver}
21530 with the @option{--once} option, it will stop listening for any further
21531 connection attempts after connecting to the first @value{GDBN} session. This
21532 means no further connections to @code{gdbserver} will be possible after the
21533 first one. It also means @code{gdbserver} will terminate after the first
21534 connection with remote @value{GDBN} has closed, even for unexpectedly closed
21535 connections and even in the @kbd{target extended-remote} mode. The
21536 @option{--once} option allows reusing the same port number for connecting to
21537 multiple instances of @code{gdbserver} running on the same host, since each
21538 instance closes its port after the first connection.
21540 @anchor{Other Command-Line Arguments for gdbserver}
21541 @subsubsection Other Command-Line Arguments for @code{gdbserver}
21543 You can use the @option{--multi} option to start @code{gdbserver} without
21544 specifying a program to debug or a process to attach to. Then you can
21545 attach in @code{target extended-remote} mode and run or attach to a
21546 program. For more information,
21547 @pxref{--multi Option in Types of Remote Connnections}.
21549 @cindex @option{--debug}, @code{gdbserver} option
21550 The @option{--debug} option tells @code{gdbserver} to display extra
21551 status information about the debugging process.
21552 @cindex @option{--remote-debug}, @code{gdbserver} option
21553 The @option{--remote-debug} option tells @code{gdbserver} to display
21554 remote protocol debug output.
21555 @cindex @option{--debug-file}, @code{gdbserver} option
21556 @cindex @code{gdbserver}, send all debug output to a single file
21557 The @option{--debug-file=@var{filename}} option tells @code{gdbserver} to
21558 write any debug output to the given @var{filename}. These options are intended
21559 for @code{gdbserver} development and for bug reports to the developers.
21561 @cindex @option{--debug-format}, @code{gdbserver} option
21562 The @option{--debug-format=option1[,option2,...]} option tells
21563 @code{gdbserver} to include additional information in each output.
21564 Possible options are:
21568 Turn off all extra information in debugging output.
21570 Turn on all extra information in debugging output.
21572 Include a timestamp in each line of debugging output.
21575 Options are processed in order. Thus, for example, if @option{none}
21576 appears last then no additional information is added to debugging output.
21578 @cindex @option{--wrapper}, @code{gdbserver} option
21579 The @option{--wrapper} option specifies a wrapper to launch programs
21580 for debugging. The option should be followed by the name of the
21581 wrapper, then any command-line arguments to pass to the wrapper, then
21582 @kbd{--} indicating the end of the wrapper arguments.
21584 @code{gdbserver} runs the specified wrapper program with a combined
21585 command line including the wrapper arguments, then the name of the
21586 program to debug, then any arguments to the program. The wrapper
21587 runs until it executes your program, and then @value{GDBN} gains control.
21589 You can use any program that eventually calls @code{execve} with
21590 its arguments as a wrapper. Several standard Unix utilities do
21591 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
21592 with @code{exec "$@@"} will also work.
21594 For example, you can use @code{env} to pass an environment variable to
21595 the debugged program, without setting the variable in @code{gdbserver}'s
21599 $ gdbserver --wrapper env LD_PRELOAD=libtest.so -- :2222 ./testprog
21602 @cindex @option{--selftest}
21603 The @option{--selftest} option runs the self tests in @code{gdbserver}:
21606 $ gdbserver --selftest
21607 Ran 2 unit tests, 0 failed
21610 These tests are disabled in release.
21611 @subsection Connecting to @code{gdbserver}
21613 The basic procedure for connecting to the remote target is:
21617 Run @value{GDBN} on the host system.
21620 Make sure you have the necessary symbol files
21621 (@pxref{Host and target files}).
21622 Load symbols for your application using the @code{file} command before you
21623 connect. Use @code{set sysroot} to locate target libraries (unless your
21624 @value{GDBN} was compiled with the correct sysroot using
21625 @code{--with-sysroot}).
21628 Connect to your target (@pxref{Connecting,,Connecting to a Remote Target}).
21629 For TCP connections, you must start up @code{gdbserver} prior to using
21630 the @code{target} command. Otherwise you may get an error whose
21631 text depends on the host system, but which usually looks something like
21632 @samp{Connection refused}. Don't use the @code{load}
21633 command in @value{GDBN} when using @code{target remote} mode, since the
21634 program is already on the target.
21638 @anchor{Monitor Commands for gdbserver}
21639 @subsection Monitor Commands for @code{gdbserver}
21640 @cindex monitor commands, for @code{gdbserver}
21642 During a @value{GDBN} session using @code{gdbserver}, you can use the
21643 @code{monitor} command to send special requests to @code{gdbserver}.
21644 Here are the available commands.
21648 List the available monitor commands.
21650 @item monitor set debug 0
21651 @itemx monitor set debug 1
21652 Disable or enable general debugging messages.
21654 @item monitor set remote-debug 0
21655 @itemx monitor set remote-debug 1
21656 Disable or enable specific debugging messages associated with the remote
21657 protocol (@pxref{Remote Protocol}).
21659 @item monitor set debug-file filename
21660 @itemx monitor set debug-file
21661 Send any debug output to the given file, or to stderr.
21663 @item monitor set debug-format option1@r{[},option2,...@r{]}
21664 Specify additional text to add to debugging messages.
21665 Possible options are:
21669 Turn off all extra information in debugging output.
21671 Turn on all extra information in debugging output.
21673 Include a timestamp in each line of debugging output.
21676 Options are processed in order. Thus, for example, if @option{none}
21677 appears last then no additional information is added to debugging output.
21679 @item monitor set libthread-db-search-path [PATH]
21680 @cindex gdbserver, search path for @code{libthread_db}
21681 When this command is issued, @var{path} is a colon-separated list of
21682 directories to search for @code{libthread_db} (@pxref{Threads,,set
21683 libthread-db-search-path}). If you omit @var{path},
21684 @samp{libthread-db-search-path} will be reset to its default value.
21686 The special entry @samp{$pdir} for @samp{libthread-db-search-path} is
21687 not supported in @code{gdbserver}.
21690 Tell gdbserver to exit immediately. This command should be followed by
21691 @code{disconnect} to close the debugging session. @code{gdbserver} will
21692 detach from any attached processes and kill any processes it created.
21693 Use @code{monitor exit} to terminate @code{gdbserver} at the end
21694 of a multi-process mode debug session.
21698 @subsection Tracepoints support in @code{gdbserver}
21699 @cindex tracepoints support in @code{gdbserver}
21701 On some targets, @code{gdbserver} supports tracepoints, fast
21702 tracepoints and static tracepoints.
21704 For fast or static tracepoints to work, a special library called the
21705 @dfn{in-process agent} (IPA), must be loaded in the inferior process.
21706 This library is built and distributed as an integral part of
21707 @code{gdbserver}. In addition, support for static tracepoints
21708 requires building the in-process agent library with static tracepoints
21709 support. At present, the UST (LTTng Userspace Tracer,
21710 @url{http://lttng.org/ust}) tracing engine is supported. This support
21711 is automatically available if UST development headers are found in the
21712 standard include path when @code{gdbserver} is built, or if
21713 @code{gdbserver} was explicitly configured using @option{--with-ust}
21714 to point at such headers. You can explicitly disable the support
21715 using @option{--with-ust=no}.
21717 There are several ways to load the in-process agent in your program:
21720 @item Specifying it as dependency at link time
21722 You can link your program dynamically with the in-process agent
21723 library. On most systems, this is accomplished by adding
21724 @code{-linproctrace} to the link command.
21726 @item Using the system's preloading mechanisms
21728 You can force loading the in-process agent at startup time by using
21729 your system's support for preloading shared libraries. Many Unixes
21730 support the concept of preloading user defined libraries. In most
21731 cases, you do that by specifying @code{LD_PRELOAD=libinproctrace.so}
21732 in the environment. See also the description of @code{gdbserver}'s
21733 @option{--wrapper} command line option.
21735 @item Using @value{GDBN} to force loading the agent at run time
21737 On some systems, you can force the inferior to load a shared library,
21738 by calling a dynamic loader function in the inferior that takes care
21739 of dynamically looking up and loading a shared library. On most Unix
21740 systems, the function is @code{dlopen}. You'll use the @code{call}
21741 command for that. For example:
21744 (@value{GDBP}) call dlopen ("libinproctrace.so", ...)
21747 Note that on most Unix systems, for the @code{dlopen} function to be
21748 available, the program needs to be linked with @code{-ldl}.
21751 On systems that have a userspace dynamic loader, like most Unix
21752 systems, when you connect to @code{gdbserver} using @code{target
21753 remote}, you'll find that the program is stopped at the dynamic
21754 loader's entry point, and no shared library has been loaded in the
21755 program's address space yet, including the in-process agent. In that
21756 case, before being able to use any of the fast or static tracepoints
21757 features, you need to let the loader run and load the shared
21758 libraries. The simplest way to do that is to run the program to the
21759 main procedure. E.g., if debugging a C or C@t{++} program, start
21760 @code{gdbserver} like so:
21763 $ gdbserver :9999 myprogram
21766 Start GDB and connect to @code{gdbserver} like so, and run to main:
21770 (@value{GDBP}) target remote myhost:9999
21771 0x00007f215893ba60 in ?? () from /lib64/ld-linux-x86-64.so.2
21772 (@value{GDBP}) b main
21773 (@value{GDBP}) continue
21776 The in-process tracing agent library should now be loaded into the
21777 process; you can confirm it with the @code{info sharedlibrary}
21778 command, which will list @file{libinproctrace.so} as loaded in the
21779 process. You are now ready to install fast tracepoints, list static
21780 tracepoint markers, probe static tracepoints markers, and start
21783 @node Remote Configuration
21784 @section Remote Configuration
21787 @kindex show remote
21788 This section documents the configuration options available when
21789 debugging remote programs. For the options related to the File I/O
21790 extensions of the remote protocol, see @ref{system,
21791 system-call-allowed}.
21794 @item set remoteaddresssize @var{bits}
21795 @cindex address size for remote targets
21796 @cindex bits in remote address
21797 Set the maximum size of address in a memory packet to the specified
21798 number of bits. @value{GDBN} will mask off the address bits above
21799 that number, when it passes addresses to the remote target. The
21800 default value is the number of bits in the target's address.
21802 @item show remoteaddresssize
21803 Show the current value of remote address size in bits.
21805 @item set serial baud @var{n}
21806 @cindex baud rate for remote targets
21807 Set the baud rate for the remote serial I/O to @var{n} baud. The
21808 value is used to set the speed of the serial port used for debugging
21811 @item show serial baud
21812 Show the current speed of the remote connection.
21814 @item set serial parity @var{parity}
21815 Set the parity for the remote serial I/O. Supported values of @var{parity} are:
21816 @code{even}, @code{none}, and @code{odd}. The default is @code{none}.
21818 @item show serial parity
21819 Show the current parity of the serial port.
21821 @item set remotebreak
21822 @cindex interrupt remote programs
21823 @cindex BREAK signal instead of Ctrl-C
21824 @anchor{set remotebreak}
21825 If set to on, @value{GDBN} sends a @code{BREAK} signal to the remote
21826 when you type @kbd{Ctrl-c} to interrupt the program running
21827 on the remote. If set to off, @value{GDBN} sends the @samp{Ctrl-C}
21828 character instead. The default is off, since most remote systems
21829 expect to see @samp{Ctrl-C} as the interrupt signal.
21831 @item show remotebreak
21832 Show whether @value{GDBN} sends @code{BREAK} or @samp{Ctrl-C} to
21833 interrupt the remote program.
21835 @item set remoteflow on
21836 @itemx set remoteflow off
21837 @kindex set remoteflow
21838 Enable or disable hardware flow control (@code{RTS}/@code{CTS})
21839 on the serial port used to communicate to the remote target.
21841 @item show remoteflow
21842 @kindex show remoteflow
21843 Show the current setting of hardware flow control.
21845 @item set remotelogbase @var{base}
21846 Set the base (a.k.a.@: radix) of logging serial protocol
21847 communications to @var{base}. Supported values of @var{base} are:
21848 @code{ascii}, @code{octal}, and @code{hex}. The default is
21851 @item show remotelogbase
21852 Show the current setting of the radix for logging remote serial
21855 @item set remotelogfile @var{file}
21856 @cindex record serial communications on file
21857 Record remote serial communications on the named @var{file}. The
21858 default is not to record at all.
21860 @item show remotelogfile
21861 Show the current setting of the file name on which to record the
21862 serial communications.
21864 @item set remotetimeout @var{num}
21865 @cindex timeout for serial communications
21866 @cindex remote timeout
21867 Set the timeout limit to wait for the remote target to respond to
21868 @var{num} seconds. The default is 2 seconds.
21870 @item show remotetimeout
21871 Show the current number of seconds to wait for the remote target
21874 @cindex limit hardware breakpoints and watchpoints
21875 @cindex remote target, limit break- and watchpoints
21876 @anchor{set remote hardware-watchpoint-limit}
21877 @anchor{set remote hardware-breakpoint-limit}
21878 @item set remote hardware-watchpoint-limit @var{limit}
21879 @itemx set remote hardware-breakpoint-limit @var{limit}
21880 Restrict @value{GDBN} to using @var{limit} remote hardware watchpoints
21881 or breakpoints. The @var{limit} can be set to 0 to disable hardware
21882 watchpoints or breakpoints, and @code{unlimited} for unlimited
21883 watchpoints or breakpoints.
21885 @item show remote hardware-watchpoint-limit
21886 @itemx show remote hardware-breakpoint-limit
21887 Show the current limit for the number of hardware watchpoints or
21888 breakpoints that @value{GDBN} can use.
21890 @cindex limit hardware watchpoints length
21891 @cindex remote target, limit watchpoints length
21892 @anchor{set remote hardware-watchpoint-length-limit}
21893 @item set remote hardware-watchpoint-length-limit @var{limit}
21894 Restrict @value{GDBN} to using @var{limit} bytes for the maximum
21895 length of a remote hardware watchpoint. A @var{limit} of 0 disables
21896 hardware watchpoints and @code{unlimited} allows watchpoints of any
21899 @item show remote hardware-watchpoint-length-limit
21900 Show the current limit (in bytes) of the maximum length of
21901 a remote hardware watchpoint.
21903 @item set remote exec-file @var{filename}
21904 @itemx show remote exec-file
21905 @anchor{set remote exec-file}
21906 @cindex executable file, for remote target
21907 Select the file used for @code{run} with @code{target
21908 extended-remote}. This should be set to a filename valid on the
21909 target system. If it is not set, the target will use a default
21910 filename (e.g.@: the last program run).
21912 @item set remote interrupt-sequence
21913 @cindex interrupt remote programs
21914 @cindex select Ctrl-C, BREAK or BREAK-g
21915 Allow the user to select one of @samp{Ctrl-C}, a @code{BREAK} or
21916 @samp{BREAK-g} as the
21917 sequence to the remote target in order to interrupt the execution.
21918 @samp{Ctrl-C} is a default. Some system prefers @code{BREAK} which
21919 is high level of serial line for some certain time.
21920 Linux kernel prefers @samp{BREAK-g}, a.k.a Magic SysRq g.
21921 It is @code{BREAK} signal followed by character @code{g}.
21923 @item show interrupt-sequence
21924 Show which of @samp{Ctrl-C}, @code{BREAK} or @code{BREAK-g}
21925 is sent by @value{GDBN} to interrupt the remote program.
21926 @code{BREAK-g} is BREAK signal followed by @code{g} and
21927 also known as Magic SysRq g.
21929 @item set remote interrupt-on-connect
21930 @cindex send interrupt-sequence on start
21931 Specify whether interrupt-sequence is sent to remote target when
21932 @value{GDBN} connects to it. This is mostly needed when you debug
21933 Linux kernel. Linux kernel expects @code{BREAK} followed by @code{g}
21934 which is known as Magic SysRq g in order to connect @value{GDBN}.
21936 @item show interrupt-on-connect
21937 Show whether interrupt-sequence is sent
21938 to remote target when @value{GDBN} connects to it.
21942 @item set tcp auto-retry on
21943 @cindex auto-retry, for remote TCP target
21944 Enable auto-retry for remote TCP connections. This is useful if the remote
21945 debugging agent is launched in parallel with @value{GDBN}; there is a race
21946 condition because the agent may not become ready to accept the connection
21947 before @value{GDBN} attempts to connect. When auto-retry is
21948 enabled, if the initial attempt to connect fails, @value{GDBN} reattempts
21949 to establish the connection using the timeout specified by
21950 @code{set tcp connect-timeout}.
21952 @item set tcp auto-retry off
21953 Do not auto-retry failed TCP connections.
21955 @item show tcp auto-retry
21956 Show the current auto-retry setting.
21958 @item set tcp connect-timeout @var{seconds}
21959 @itemx set tcp connect-timeout unlimited
21960 @cindex connection timeout, for remote TCP target
21961 @cindex timeout, for remote target connection
21962 Set the timeout for establishing a TCP connection to the remote target to
21963 @var{seconds}. The timeout affects both polling to retry failed connections
21964 (enabled by @code{set tcp auto-retry on}) and waiting for connections
21965 that are merely slow to complete, and represents an approximate cumulative
21966 value. If @var{seconds} is @code{unlimited}, there is no timeout and
21967 @value{GDBN} will keep attempting to establish a connection forever,
21968 unless interrupted with @kbd{Ctrl-c}. The default is 15 seconds.
21970 @item show tcp connect-timeout
21971 Show the current connection timeout setting.
21974 @cindex remote packets, enabling and disabling
21975 The @value{GDBN} remote protocol autodetects the packets supported by
21976 your debugging stub. If you need to override the autodetection, you
21977 can use these commands to enable or disable individual packets. Each
21978 packet can be set to @samp{on} (the remote target supports this
21979 packet), @samp{off} (the remote target does not support this packet),
21980 or @samp{auto} (detect remote target support for this packet). They
21981 all default to @samp{auto}. For more information about each packet,
21982 see @ref{Remote Protocol}.
21984 During normal use, you should not have to use any of these commands.
21985 If you do, that may be a bug in your remote debugging stub, or a bug
21986 in @value{GDBN}. You may want to report the problem to the
21987 @value{GDBN} developers.
21989 For each packet @var{name}, the command to enable or disable the
21990 packet is @code{set remote @var{name}-packet}. The available settings
21993 @multitable @columnfractions 0.28 0.32 0.25
21996 @tab Related Features
21998 @item @code{fetch-register}
22000 @tab @code{info registers}
22002 @item @code{set-register}
22006 @item @code{binary-download}
22008 @tab @code{load}, @code{set}
22010 @item @code{read-aux-vector}
22011 @tab @code{qXfer:auxv:read}
22012 @tab @code{info auxv}
22014 @item @code{symbol-lookup}
22015 @tab @code{qSymbol}
22016 @tab Detecting multiple threads
22018 @item @code{attach}
22019 @tab @code{vAttach}
22022 @item @code{verbose-resume}
22024 @tab Stepping or resuming multiple threads
22030 @item @code{software-breakpoint}
22034 @item @code{hardware-breakpoint}
22038 @item @code{write-watchpoint}
22042 @item @code{read-watchpoint}
22046 @item @code{access-watchpoint}
22050 @item @code{pid-to-exec-file}
22051 @tab @code{qXfer:exec-file:read}
22052 @tab @code{attach}, @code{run}
22054 @item @code{target-features}
22055 @tab @code{qXfer:features:read}
22056 @tab @code{set architecture}
22058 @item @code{library-info}
22059 @tab @code{qXfer:libraries:read}
22060 @tab @code{info sharedlibrary}
22062 @item @code{memory-map}
22063 @tab @code{qXfer:memory-map:read}
22064 @tab @code{info mem}
22066 @item @code{read-sdata-object}
22067 @tab @code{qXfer:sdata:read}
22068 @tab @code{print $_sdata}
22070 @item @code{read-spu-object}
22071 @tab @code{qXfer:spu:read}
22072 @tab @code{info spu}
22074 @item @code{write-spu-object}
22075 @tab @code{qXfer:spu:write}
22076 @tab @code{info spu}
22078 @item @code{read-siginfo-object}
22079 @tab @code{qXfer:siginfo:read}
22080 @tab @code{print $_siginfo}
22082 @item @code{write-siginfo-object}
22083 @tab @code{qXfer:siginfo:write}
22084 @tab @code{set $_siginfo}
22086 @item @code{threads}
22087 @tab @code{qXfer:threads:read}
22088 @tab @code{info threads}
22090 @item @code{get-thread-local-@*storage-address}
22091 @tab @code{qGetTLSAddr}
22092 @tab Displaying @code{__thread} variables
22094 @item @code{get-thread-information-block-address}
22095 @tab @code{qGetTIBAddr}
22096 @tab Display MS-Windows Thread Information Block.
22098 @item @code{search-memory}
22099 @tab @code{qSearch:memory}
22102 @item @code{supported-packets}
22103 @tab @code{qSupported}
22104 @tab Remote communications parameters
22106 @item @code{catch-syscalls}
22107 @tab @code{QCatchSyscalls}
22108 @tab @code{catch syscall}
22110 @item @code{pass-signals}
22111 @tab @code{QPassSignals}
22112 @tab @code{handle @var{signal}}
22114 @item @code{program-signals}
22115 @tab @code{QProgramSignals}
22116 @tab @code{handle @var{signal}}
22118 @item @code{hostio-close-packet}
22119 @tab @code{vFile:close}
22120 @tab @code{remote get}, @code{remote put}
22122 @item @code{hostio-open-packet}
22123 @tab @code{vFile:open}
22124 @tab @code{remote get}, @code{remote put}
22126 @item @code{hostio-pread-packet}
22127 @tab @code{vFile:pread}
22128 @tab @code{remote get}, @code{remote put}
22130 @item @code{hostio-pwrite-packet}
22131 @tab @code{vFile:pwrite}
22132 @tab @code{remote get}, @code{remote put}
22134 @item @code{hostio-unlink-packet}
22135 @tab @code{vFile:unlink}
22136 @tab @code{remote delete}
22138 @item @code{hostio-readlink-packet}
22139 @tab @code{vFile:readlink}
22142 @item @code{hostio-fstat-packet}
22143 @tab @code{vFile:fstat}
22146 @item @code{hostio-setfs-packet}
22147 @tab @code{vFile:setfs}
22150 @item @code{noack-packet}
22151 @tab @code{QStartNoAckMode}
22152 @tab Packet acknowledgment
22154 @item @code{osdata}
22155 @tab @code{qXfer:osdata:read}
22156 @tab @code{info os}
22158 @item @code{query-attached}
22159 @tab @code{qAttached}
22160 @tab Querying remote process attach state.
22162 @item @code{trace-buffer-size}
22163 @tab @code{QTBuffer:size}
22164 @tab @code{set trace-buffer-size}
22166 @item @code{trace-status}
22167 @tab @code{qTStatus}
22168 @tab @code{tstatus}
22170 @item @code{traceframe-info}
22171 @tab @code{qXfer:traceframe-info:read}
22172 @tab Traceframe info
22174 @item @code{install-in-trace}
22175 @tab @code{InstallInTrace}
22176 @tab Install tracepoint in tracing
22178 @item @code{disable-randomization}
22179 @tab @code{QDisableRandomization}
22180 @tab @code{set disable-randomization}
22182 @item @code{startup-with-shell}
22183 @tab @code{QStartupWithShell}
22184 @tab @code{set startup-with-shell}
22186 @item @code{environment-hex-encoded}
22187 @tab @code{QEnvironmentHexEncoded}
22188 @tab @code{set environment}
22190 @item @code{environment-unset}
22191 @tab @code{QEnvironmentUnset}
22192 @tab @code{unset environment}
22194 @item @code{environment-reset}
22195 @tab @code{QEnvironmentReset}
22196 @tab @code{Reset the inferior environment (i.e., unset user-set variables)}
22198 @item @code{set-working-dir}
22199 @tab @code{QSetWorkingDir}
22200 @tab @code{set cwd}
22202 @item @code{conditional-breakpoints-packet}
22203 @tab @code{Z0 and Z1}
22204 @tab @code{Support for target-side breakpoint condition evaluation}
22206 @item @code{multiprocess-extensions}
22207 @tab @code{multiprocess extensions}
22208 @tab Debug multiple processes and remote process PID awareness
22210 @item @code{swbreak-feature}
22211 @tab @code{swbreak stop reason}
22214 @item @code{hwbreak-feature}
22215 @tab @code{hwbreak stop reason}
22218 @item @code{fork-event-feature}
22219 @tab @code{fork stop reason}
22222 @item @code{vfork-event-feature}
22223 @tab @code{vfork stop reason}
22226 @item @code{exec-event-feature}
22227 @tab @code{exec stop reason}
22230 @item @code{thread-events}
22231 @tab @code{QThreadEvents}
22232 @tab Tracking thread lifetime.
22234 @item @code{no-resumed-stop-reply}
22235 @tab @code{no resumed thread left stop reply}
22236 @tab Tracking thread lifetime.
22241 @section Implementing a Remote Stub
22243 @cindex debugging stub, example
22244 @cindex remote stub, example
22245 @cindex stub example, remote debugging
22246 The stub files provided with @value{GDBN} implement the target side of the
22247 communication protocol, and the @value{GDBN} side is implemented in the
22248 @value{GDBN} source file @file{remote.c}. Normally, you can simply allow
22249 these subroutines to communicate, and ignore the details. (If you're
22250 implementing your own stub file, you can still ignore the details: start
22251 with one of the existing stub files. @file{sparc-stub.c} is the best
22252 organized, and therefore the easiest to read.)
22254 @cindex remote serial debugging, overview
22255 To debug a program running on another machine (the debugging
22256 @dfn{target} machine), you must first arrange for all the usual
22257 prerequisites for the program to run by itself. For example, for a C
22262 A startup routine to set up the C runtime environment; these usually
22263 have a name like @file{crt0}. The startup routine may be supplied by
22264 your hardware supplier, or you may have to write your own.
22267 A C subroutine library to support your program's
22268 subroutine calls, notably managing input and output.
22271 A way of getting your program to the other machine---for example, a
22272 download program. These are often supplied by the hardware
22273 manufacturer, but you may have to write your own from hardware
22277 The next step is to arrange for your program to use a serial port to
22278 communicate with the machine where @value{GDBN} is running (the @dfn{host}
22279 machine). In general terms, the scheme looks like this:
22283 @value{GDBN} already understands how to use this protocol; when everything
22284 else is set up, you can simply use the @samp{target remote} command
22285 (@pxref{Targets,,Specifying a Debugging Target}).
22287 @item On the target,
22288 you must link with your program a few special-purpose subroutines that
22289 implement the @value{GDBN} remote serial protocol. The file containing these
22290 subroutines is called a @dfn{debugging stub}.
22292 On certain remote targets, you can use an auxiliary program
22293 @code{gdbserver} instead of linking a stub into your program.
22294 @xref{Server,,Using the @code{gdbserver} Program}, for details.
22297 The debugging stub is specific to the architecture of the remote
22298 machine; for example, use @file{sparc-stub.c} to debug programs on
22301 @cindex remote serial stub list
22302 These working remote stubs are distributed with @value{GDBN}:
22307 @cindex @file{i386-stub.c}
22310 For Intel 386 and compatible architectures.
22313 @cindex @file{m68k-stub.c}
22314 @cindex Motorola 680x0
22316 For Motorola 680x0 architectures.
22319 @cindex @file{sh-stub.c}
22322 For Renesas SH architectures.
22325 @cindex @file{sparc-stub.c}
22327 For @sc{sparc} architectures.
22329 @item sparcl-stub.c
22330 @cindex @file{sparcl-stub.c}
22333 For Fujitsu @sc{sparclite} architectures.
22337 The @file{README} file in the @value{GDBN} distribution may list other
22338 recently added stubs.
22341 * Stub Contents:: What the stub can do for you
22342 * Bootstrapping:: What you must do for the stub
22343 * Debug Session:: Putting it all together
22346 @node Stub Contents
22347 @subsection What the Stub Can Do for You
22349 @cindex remote serial stub
22350 The debugging stub for your architecture supplies these three
22354 @item set_debug_traps
22355 @findex set_debug_traps
22356 @cindex remote serial stub, initialization
22357 This routine arranges for @code{handle_exception} to run when your
22358 program stops. You must call this subroutine explicitly in your
22359 program's startup code.
22361 @item handle_exception
22362 @findex handle_exception
22363 @cindex remote serial stub, main routine
22364 This is the central workhorse, but your program never calls it
22365 explicitly---the setup code arranges for @code{handle_exception} to
22366 run when a trap is triggered.
22368 @code{handle_exception} takes control when your program stops during
22369 execution (for example, on a breakpoint), and mediates communications
22370 with @value{GDBN} on the host machine. This is where the communications
22371 protocol is implemented; @code{handle_exception} acts as the @value{GDBN}
22372 representative on the target machine. It begins by sending summary
22373 information on the state of your program, then continues to execute,
22374 retrieving and transmitting any information @value{GDBN} needs, until you
22375 execute a @value{GDBN} command that makes your program resume; at that point,
22376 @code{handle_exception} returns control to your own code on the target
22380 @cindex @code{breakpoint} subroutine, remote
22381 Use this auxiliary subroutine to make your program contain a
22382 breakpoint. Depending on the particular situation, this may be the only
22383 way for @value{GDBN} to get control. For instance, if your target
22384 machine has some sort of interrupt button, you won't need to call this;
22385 pressing the interrupt button transfers control to
22386 @code{handle_exception}---in effect, to @value{GDBN}. On some machines,
22387 simply receiving characters on the serial port may also trigger a trap;
22388 again, in that situation, you don't need to call @code{breakpoint} from
22389 your own program---simply running @samp{target remote} from the host
22390 @value{GDBN} session gets control.
22392 Call @code{breakpoint} if none of these is true, or if you simply want
22393 to make certain your program stops at a predetermined point for the
22394 start of your debugging session.
22397 @node Bootstrapping
22398 @subsection What You Must Do for the Stub
22400 @cindex remote stub, support routines
22401 The debugging stubs that come with @value{GDBN} are set up for a particular
22402 chip architecture, but they have no information about the rest of your
22403 debugging target machine.
22405 First of all you need to tell the stub how to communicate with the
22409 @item int getDebugChar()
22410 @findex getDebugChar
22411 Write this subroutine to read a single character from the serial port.
22412 It may be identical to @code{getchar} for your target system; a
22413 different name is used to allow you to distinguish the two if you wish.
22415 @item void putDebugChar(int)
22416 @findex putDebugChar
22417 Write this subroutine to write a single character to the serial port.
22418 It may be identical to @code{putchar} for your target system; a
22419 different name is used to allow you to distinguish the two if you wish.
22422 @cindex control C, and remote debugging
22423 @cindex interrupting remote targets
22424 If you want @value{GDBN} to be able to stop your program while it is
22425 running, you need to use an interrupt-driven serial driver, and arrange
22426 for it to stop when it receives a @code{^C} (@samp{\003}, the control-C
22427 character). That is the character which @value{GDBN} uses to tell the
22428 remote system to stop.
22430 Getting the debugging target to return the proper status to @value{GDBN}
22431 probably requires changes to the standard stub; one quick and dirty way
22432 is to just execute a breakpoint instruction (the ``dirty'' part is that
22433 @value{GDBN} reports a @code{SIGTRAP} instead of a @code{SIGINT}).
22435 Other routines you need to supply are:
22438 @item void exceptionHandler (int @var{exception_number}, void *@var{exception_address})
22439 @findex exceptionHandler
22440 Write this function to install @var{exception_address} in the exception
22441 handling tables. You need to do this because the stub does not have any
22442 way of knowing what the exception handling tables on your target system
22443 are like (for example, the processor's table might be in @sc{rom},
22444 containing entries which point to a table in @sc{ram}).
22445 The @var{exception_number} specifies the exception which should be changed;
22446 its meaning is architecture-dependent (for example, different numbers
22447 might represent divide by zero, misaligned access, etc). When this
22448 exception occurs, control should be transferred directly to
22449 @var{exception_address}, and the processor state (stack, registers,
22450 and so on) should be just as it is when a processor exception occurs. So if
22451 you want to use a jump instruction to reach @var{exception_address}, it
22452 should be a simple jump, not a jump to subroutine.
22454 For the 386, @var{exception_address} should be installed as an interrupt
22455 gate so that interrupts are masked while the handler runs. The gate
22456 should be at privilege level 0 (the most privileged level). The
22457 @sc{sparc} and 68k stubs are able to mask interrupts themselves without
22458 help from @code{exceptionHandler}.
22460 @item void flush_i_cache()
22461 @findex flush_i_cache
22462 On @sc{sparc} and @sc{sparclite} only, write this subroutine to flush the
22463 instruction cache, if any, on your target machine. If there is no
22464 instruction cache, this subroutine may be a no-op.
22466 On target machines that have instruction caches, @value{GDBN} requires this
22467 function to make certain that the state of your program is stable.
22471 You must also make sure this library routine is available:
22474 @item void *memset(void *, int, int)
22476 This is the standard library function @code{memset} that sets an area of
22477 memory to a known value. If you have one of the free versions of
22478 @code{libc.a}, @code{memset} can be found there; otherwise, you must
22479 either obtain it from your hardware manufacturer, or write your own.
22482 If you do not use the GNU C compiler, you may need other standard
22483 library subroutines as well; this varies from one stub to another,
22484 but in general the stubs are likely to use any of the common library
22485 subroutines which @code{@value{NGCC}} generates as inline code.
22488 @node Debug Session
22489 @subsection Putting it All Together
22491 @cindex remote serial debugging summary
22492 In summary, when your program is ready to debug, you must follow these
22497 Make sure you have defined the supporting low-level routines
22498 (@pxref{Bootstrapping,,What You Must Do for the Stub}):
22500 @code{getDebugChar}, @code{putDebugChar},
22501 @code{flush_i_cache}, @code{memset}, @code{exceptionHandler}.
22505 Insert these lines in your program's startup code, before the main
22506 procedure is called:
22513 On some machines, when a breakpoint trap is raised, the hardware
22514 automatically makes the PC point to the instruction after the
22515 breakpoint. If your machine doesn't do that, you may need to adjust
22516 @code{handle_exception} to arrange for it to return to the instruction
22517 after the breakpoint on this first invocation, so that your program
22518 doesn't keep hitting the initial breakpoint instead of making
22522 For the 680x0 stub only, you need to provide a variable called
22523 @code{exceptionHook}. Normally you just use:
22526 void (*exceptionHook)() = 0;
22530 but if before calling @code{set_debug_traps}, you set it to point to a
22531 function in your program, that function is called when
22532 @code{@value{GDBN}} continues after stopping on a trap (for example, bus
22533 error). The function indicated by @code{exceptionHook} is called with
22534 one parameter: an @code{int} which is the exception number.
22537 Compile and link together: your program, the @value{GDBN} debugging stub for
22538 your target architecture, and the supporting subroutines.
22541 Make sure you have a serial connection between your target machine and
22542 the @value{GDBN} host, and identify the serial port on the host.
22545 @c The "remote" target now provides a `load' command, so we should
22546 @c document that. FIXME.
22547 Download your program to your target machine (or get it there by
22548 whatever means the manufacturer provides), and start it.
22551 Start @value{GDBN} on the host, and connect to the target
22552 (@pxref{Connecting,,Connecting to a Remote Target}).
22556 @node Configurations
22557 @chapter Configuration-Specific Information
22559 While nearly all @value{GDBN} commands are available for all native and
22560 cross versions of the debugger, there are some exceptions. This chapter
22561 describes things that are only available in certain configurations.
22563 There are three major categories of configurations: native
22564 configurations, where the host and target are the same, embedded
22565 operating system configurations, which are usually the same for several
22566 different processor architectures, and bare embedded processors, which
22567 are quite different from each other.
22572 * Embedded Processors::
22579 This section describes details specific to particular native
22583 * BSD libkvm Interface:: Debugging BSD kernel memory images
22584 * Process Information:: Process information
22585 * DJGPP Native:: Features specific to the DJGPP port
22586 * Cygwin Native:: Features specific to the Cygwin port
22587 * Hurd Native:: Features specific to @sc{gnu} Hurd
22588 * Darwin:: Features specific to Darwin
22589 * FreeBSD:: Features specific to FreeBSD
22592 @node BSD libkvm Interface
22593 @subsection BSD libkvm Interface
22596 @cindex kernel memory image
22597 @cindex kernel crash dump
22599 BSD-derived systems (FreeBSD/NetBSD/OpenBSD) have a kernel memory
22600 interface that provides a uniform interface for accessing kernel virtual
22601 memory images, including live systems and crash dumps. @value{GDBN}
22602 uses this interface to allow you to debug live kernels and kernel crash
22603 dumps on many native BSD configurations. This is implemented as a
22604 special @code{kvm} debugging target. For debugging a live system, load
22605 the currently running kernel into @value{GDBN} and connect to the
22609 (@value{GDBP}) @b{target kvm}
22612 For debugging crash dumps, provide the file name of the crash dump as an
22616 (@value{GDBP}) @b{target kvm /var/crash/bsd.0}
22619 Once connected to the @code{kvm} target, the following commands are
22625 Set current context from the @dfn{Process Control Block} (PCB) address.
22628 Set current context from proc address. This command isn't available on
22629 modern FreeBSD systems.
22632 @node Process Information
22633 @subsection Process Information
22635 @cindex examine process image
22636 @cindex process info via @file{/proc}
22638 Some operating systems provide interfaces to fetch additional
22639 information about running processes beyond memory and per-thread
22640 register state. If @value{GDBN} is configured for an operating system
22641 with a supported interface, the command @code{info proc} is available
22642 to report information about the process running your program, or about
22643 any process running on your system.
22645 One supported interface is a facility called @samp{/proc} that can be
22646 used to examine the image of a running process using file-system
22647 subroutines. This facility is supported on @sc{gnu}/Linux and Solaris
22650 On FreeBSD systems, system control nodes are used to query process
22653 In addition, some systems may provide additional process information
22654 in core files. Note that a core file may include a subset of the
22655 information available from a live process. Process information is
22656 currently avaiable from cores created on @sc{gnu}/Linux and FreeBSD
22663 @itemx info proc @var{process-id}
22664 Summarize available information about a process. If a
22665 process ID is specified by @var{process-id}, display information about
22666 that process; otherwise display information about the program being
22667 debugged. The summary includes the debugged process ID, the command
22668 line used to invoke it, its current working directory, and its
22669 executable file's absolute file name.
22671 On some systems, @var{process-id} can be of the form
22672 @samp{[@var{pid}]/@var{tid}} which specifies a certain thread ID
22673 within a process. If the optional @var{pid} part is missing, it means
22674 a thread from the process being debugged (the leading @samp{/} still
22675 needs to be present, or else @value{GDBN} will interpret the number as
22676 a process ID rather than a thread ID).
22678 @item info proc cmdline
22679 @cindex info proc cmdline
22680 Show the original command line of the process. This command is
22681 supported on @sc{gnu}/Linux and FreeBSD.
22683 @item info proc cwd
22684 @cindex info proc cwd
22685 Show the current working directory of the process. This command is
22686 supported on @sc{gnu}/Linux and FreeBSD.
22688 @item info proc exe
22689 @cindex info proc exe
22690 Show the name of executable of the process. This command is supported
22691 on @sc{gnu}/Linux and FreeBSD.
22693 @item info proc files
22694 @cindex info proc files
22695 Show the file descriptors open by the process. For each open file
22696 descriptor, @value{GDBN} shows its number, type (file, directory,
22697 character device, socket), file pointer offset, and the name of the
22698 resource open on the descriptor. The resource name can be a file name
22699 (for files, directories, and devices) or a protocol followed by socket
22700 address (for network connections). This command is supported on
22703 This example shows the open file descriptors for a process using a
22704 tty for standard input and output as well as two network sockets:
22707 (gdb) info proc files 22136
22711 FD Type Offset Flags Name
22712 text file - r-------- /usr/bin/ssh
22713 ctty chr - rw------- /dev/pts/20
22714 cwd dir - r-------- /usr/home/john
22715 root dir - r-------- /
22716 0 chr 0x32933a4 rw------- /dev/pts/20
22717 1 chr 0x32933a4 rw------- /dev/pts/20
22718 2 chr 0x32933a4 rw------- /dev/pts/20
22719 3 socket 0x0 rw----n-- tcp4 10.0.1.2:53014 -> 10.0.1.10:22
22720 4 socket 0x0 rw------- unix stream:/tmp/ssh-FIt89oAzOn5f/agent.2456
22723 @item info proc mappings
22724 @cindex memory address space mappings
22725 Report the memory address space ranges accessible in a process. On
22726 Solaris and FreeBSD systems, each memory range includes information on
22727 whether the process has read, write, or execute access rights to each
22728 range. On @sc{gnu}/Linux and FreeBSD systems, each memory range
22729 includes the object file which is mapped to that range.
22731 @item info proc stat
22732 @itemx info proc status
22733 @cindex process detailed status information
22734 Show additional process-related information, including the user ID and
22735 group ID; virtual memory usage; the signals that are pending, blocked,
22736 and ignored; its TTY; its consumption of system and user time; its
22737 stack size; its @samp{nice} value; etc. These commands are supported
22738 on @sc{gnu}/Linux and FreeBSD.
22740 For @sc{gnu}/Linux systems, see the @samp{proc} man page for more
22741 information (type @kbd{man 5 proc} from your shell prompt).
22743 For FreeBSD systems, @code{info proc stat} is an alias for @code{info
22746 @item info proc all
22747 Show all the information about the process described under all of the
22748 above @code{info proc} subcommands.
22751 @comment These sub-options of 'info proc' were not included when
22752 @comment procfs.c was re-written. Keep their descriptions around
22753 @comment against the day when someone finds the time to put them back in.
22754 @kindex info proc times
22755 @item info proc times
22756 Starting time, user CPU time, and system CPU time for your program and
22759 @kindex info proc id
22761 Report on the process IDs related to your program: its own process ID,
22762 the ID of its parent, the process group ID, and the session ID.
22765 @item set procfs-trace
22766 @kindex set procfs-trace
22767 @cindex @code{procfs} API calls
22768 This command enables and disables tracing of @code{procfs} API calls.
22770 @item show procfs-trace
22771 @kindex show procfs-trace
22772 Show the current state of @code{procfs} API call tracing.
22774 @item set procfs-file @var{file}
22775 @kindex set procfs-file
22776 Tell @value{GDBN} to write @code{procfs} API trace to the named
22777 @var{file}. @value{GDBN} appends the trace info to the previous
22778 contents of the file. The default is to display the trace on the
22781 @item show procfs-file
22782 @kindex show procfs-file
22783 Show the file to which @code{procfs} API trace is written.
22785 @item proc-trace-entry
22786 @itemx proc-trace-exit
22787 @itemx proc-untrace-entry
22788 @itemx proc-untrace-exit
22789 @kindex proc-trace-entry
22790 @kindex proc-trace-exit
22791 @kindex proc-untrace-entry
22792 @kindex proc-untrace-exit
22793 These commands enable and disable tracing of entries into and exits
22794 from the @code{syscall} interface.
22797 @kindex info pidlist
22798 @cindex process list, QNX Neutrino
22799 For QNX Neutrino only, this command displays the list of all the
22800 processes and all the threads within each process.
22803 @kindex info meminfo
22804 @cindex mapinfo list, QNX Neutrino
22805 For QNX Neutrino only, this command displays the list of all mapinfos.
22809 @subsection Features for Debugging @sc{djgpp} Programs
22810 @cindex @sc{djgpp} debugging
22811 @cindex native @sc{djgpp} debugging
22812 @cindex MS-DOS-specific commands
22815 @sc{djgpp} is a port of the @sc{gnu} development tools to MS-DOS and
22816 MS-Windows. @sc{djgpp} programs are 32-bit protected-mode programs
22817 that use the @dfn{DPMI} (DOS Protected-Mode Interface) API to run on
22818 top of real-mode DOS systems and their emulations.
22820 @value{GDBN} supports native debugging of @sc{djgpp} programs, and
22821 defines a few commands specific to the @sc{djgpp} port. This
22822 subsection describes those commands.
22827 This is a prefix of @sc{djgpp}-specific commands which print
22828 information about the target system and important OS structures.
22831 @cindex MS-DOS system info
22832 @cindex free memory information (MS-DOS)
22833 @item info dos sysinfo
22834 This command displays assorted information about the underlying
22835 platform: the CPU type and features, the OS version and flavor, the
22836 DPMI version, and the available conventional and DPMI memory.
22841 @cindex segment descriptor tables
22842 @cindex descriptor tables display
22844 @itemx info dos ldt
22845 @itemx info dos idt
22846 These 3 commands display entries from, respectively, Global, Local,
22847 and Interrupt Descriptor Tables (GDT, LDT, and IDT). The descriptor
22848 tables are data structures which store a descriptor for each segment
22849 that is currently in use. The segment's selector is an index into a
22850 descriptor table; the table entry for that index holds the
22851 descriptor's base address and limit, and its attributes and access
22854 A typical @sc{djgpp} program uses 3 segments: a code segment, a data
22855 segment (used for both data and the stack), and a DOS segment (which
22856 allows access to DOS/BIOS data structures and absolute addresses in
22857 conventional memory). However, the DPMI host will usually define
22858 additional segments in order to support the DPMI environment.
22860 @cindex garbled pointers
22861 These commands allow to display entries from the descriptor tables.
22862 Without an argument, all entries from the specified table are
22863 displayed. An argument, which should be an integer expression, means
22864 display a single entry whose index is given by the argument. For
22865 example, here's a convenient way to display information about the
22866 debugged program's data segment:
22869 @exdent @code{(@value{GDBP}) info dos ldt $ds}
22870 @exdent @code{0x13f: base=0x11970000 limit=0x0009ffff 32-Bit Data (Read/Write, Exp-up)}
22874 This comes in handy when you want to see whether a pointer is outside
22875 the data segment's limit (i.e.@: @dfn{garbled}).
22877 @cindex page tables display (MS-DOS)
22879 @itemx info dos pte
22880 These two commands display entries from, respectively, the Page
22881 Directory and the Page Tables. Page Directories and Page Tables are
22882 data structures which control how virtual memory addresses are mapped
22883 into physical addresses. A Page Table includes an entry for every
22884 page of memory that is mapped into the program's address space; there
22885 may be several Page Tables, each one holding up to 4096 entries. A
22886 Page Directory has up to 4096 entries, one each for every Page Table
22887 that is currently in use.
22889 Without an argument, @kbd{info dos pde} displays the entire Page
22890 Directory, and @kbd{info dos pte} displays all the entries in all of
22891 the Page Tables. An argument, an integer expression, given to the
22892 @kbd{info dos pde} command means display only that entry from the Page
22893 Directory table. An argument given to the @kbd{info dos pte} command
22894 means display entries from a single Page Table, the one pointed to by
22895 the specified entry in the Page Directory.
22897 @cindex direct memory access (DMA) on MS-DOS
22898 These commands are useful when your program uses @dfn{DMA} (Direct
22899 Memory Access), which needs physical addresses to program the DMA
22902 These commands are supported only with some DPMI servers.
22904 @cindex physical address from linear address
22905 @item info dos address-pte @var{addr}
22906 This command displays the Page Table entry for a specified linear
22907 address. The argument @var{addr} is a linear address which should
22908 already have the appropriate segment's base address added to it,
22909 because this command accepts addresses which may belong to @emph{any}
22910 segment. For example, here's how to display the Page Table entry for
22911 the page where a variable @code{i} is stored:
22914 @exdent @code{(@value{GDBP}) info dos address-pte __djgpp_base_address + (char *)&i}
22915 @exdent @code{Page Table entry for address 0x11a00d30:}
22916 @exdent @code{Base=0x02698000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0xd30}
22920 This says that @code{i} is stored at offset @code{0xd30} from the page
22921 whose physical base address is @code{0x02698000}, and shows all the
22922 attributes of that page.
22924 Note that you must cast the addresses of variables to a @code{char *},
22925 since otherwise the value of @code{__djgpp_base_address}, the base
22926 address of all variables and functions in a @sc{djgpp} program, will
22927 be added using the rules of C pointer arithmetics: if @code{i} is
22928 declared an @code{int}, @value{GDBN} will add 4 times the value of
22929 @code{__djgpp_base_address} to the address of @code{i}.
22931 Here's another example, it displays the Page Table entry for the
22935 @exdent @code{(@value{GDBP}) info dos address-pte *((unsigned *)&_go32_info_block + 3)}
22936 @exdent @code{Page Table entry for address 0x29110:}
22937 @exdent @code{Base=0x00029000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0x110}
22941 (The @code{+ 3} offset is because the transfer buffer's address is the
22942 3rd member of the @code{_go32_info_block} structure.) The output
22943 clearly shows that this DPMI server maps the addresses in conventional
22944 memory 1:1, i.e.@: the physical (@code{0x00029000} + @code{0x110}) and
22945 linear (@code{0x29110}) addresses are identical.
22947 This command is supported only with some DPMI servers.
22950 @cindex DOS serial data link, remote debugging
22951 In addition to native debugging, the DJGPP port supports remote
22952 debugging via a serial data link. The following commands are specific
22953 to remote serial debugging in the DJGPP port of @value{GDBN}.
22956 @kindex set com1base
22957 @kindex set com1irq
22958 @kindex set com2base
22959 @kindex set com2irq
22960 @kindex set com3base
22961 @kindex set com3irq
22962 @kindex set com4base
22963 @kindex set com4irq
22964 @item set com1base @var{addr}
22965 This command sets the base I/O port address of the @file{COM1} serial
22968 @item set com1irq @var{irq}
22969 This command sets the @dfn{Interrupt Request} (@code{IRQ}) line to use
22970 for the @file{COM1} serial port.
22972 There are similar commands @samp{set com2base}, @samp{set com3irq},
22973 etc.@: for setting the port address and the @code{IRQ} lines for the
22976 @kindex show com1base
22977 @kindex show com1irq
22978 @kindex show com2base
22979 @kindex show com2irq
22980 @kindex show com3base
22981 @kindex show com3irq
22982 @kindex show com4base
22983 @kindex show com4irq
22984 The related commands @samp{show com1base}, @samp{show com1irq} etc.@:
22985 display the current settings of the base address and the @code{IRQ}
22986 lines used by the COM ports.
22989 @kindex info serial
22990 @cindex DOS serial port status
22991 This command prints the status of the 4 DOS serial ports. For each
22992 port, it prints whether it's active or not, its I/O base address and
22993 IRQ number, whether it uses a 16550-style FIFO, its baudrate, and the
22994 counts of various errors encountered so far.
22998 @node Cygwin Native
22999 @subsection Features for Debugging MS Windows PE Executables
23000 @cindex MS Windows debugging
23001 @cindex native Cygwin debugging
23002 @cindex Cygwin-specific commands
23004 @value{GDBN} supports native debugging of MS Windows programs, including
23005 DLLs with and without symbolic debugging information.
23007 @cindex Ctrl-BREAK, MS-Windows
23008 @cindex interrupt debuggee on MS-Windows
23009 MS-Windows programs that call @code{SetConsoleMode} to switch off the
23010 special meaning of the @samp{Ctrl-C} keystroke cannot be interrupted
23011 by typing @kbd{C-c}. For this reason, @value{GDBN} on MS-Windows
23012 supports @kbd{C-@key{BREAK}} as an alternative interrupt key
23013 sequence, which can be used to interrupt the debuggee even if it
23016 There are various additional Cygwin-specific commands, described in
23017 this section. Working with DLLs that have no debugging symbols is
23018 described in @ref{Non-debug DLL Symbols}.
23023 This is a prefix of MS Windows-specific commands which print
23024 information about the target system and important OS structures.
23026 @item info w32 selector
23027 This command displays information returned by
23028 the Win32 API @code{GetThreadSelectorEntry} function.
23029 It takes an optional argument that is evaluated to
23030 a long value to give the information about this given selector.
23031 Without argument, this command displays information
23032 about the six segment registers.
23034 @item info w32 thread-information-block
23035 This command displays thread specific information stored in the
23036 Thread Information Block (readable on the X86 CPU family using @code{$fs}
23037 selector for 32-bit programs and @code{$gs} for 64-bit programs).
23039 @kindex signal-event
23040 @item signal-event @var{id}
23041 This command signals an event with user-provided @var{id}. Used to resume
23042 crashing process when attached to it using MS-Windows JIT debugging (AeDebug).
23044 To use it, create or edit the following keys in
23045 @code{HKLM\SOFTWARE\Microsoft\Windows NT\CurrentVersion\AeDebug} and/or
23046 @code{HKLM\SOFTWARE\Wow6432Node\Microsoft\Windows NT\CurrentVersion\AeDebug}
23047 (for x86_64 versions):
23051 @code{Debugger} (REG_SZ) --- a command to launch the debugger.
23052 Suggested command is: @code{@var{fully-qualified-path-to-gdb.exe} -ex
23053 "attach %ld" -ex "signal-event %ld" -ex "continue"}.
23055 The first @code{%ld} will be replaced by the process ID of the
23056 crashing process, the second @code{%ld} will be replaced by the ID of
23057 the event that blocks the crashing process, waiting for @value{GDBN}
23061 @code{Auto} (REG_SZ) --- either @code{1} or @code{0}. @code{1} will
23062 make the system run debugger specified by the Debugger key
23063 automatically, @code{0} will cause a dialog box with ``OK'' and
23064 ``Cancel'' buttons to appear, which allows the user to either
23065 terminate the crashing process (OK) or debug it (Cancel).
23068 @kindex set cygwin-exceptions
23069 @cindex debugging the Cygwin DLL
23070 @cindex Cygwin DLL, debugging
23071 @item set cygwin-exceptions @var{mode}
23072 If @var{mode} is @code{on}, @value{GDBN} will break on exceptions that
23073 happen inside the Cygwin DLL. If @var{mode} is @code{off},
23074 @value{GDBN} will delay recognition of exceptions, and may ignore some
23075 exceptions which seem to be caused by internal Cygwin DLL
23076 ``bookkeeping''. This option is meant primarily for debugging the
23077 Cygwin DLL itself; the default value is @code{off} to avoid annoying
23078 @value{GDBN} users with false @code{SIGSEGV} signals.
23080 @kindex show cygwin-exceptions
23081 @item show cygwin-exceptions
23082 Displays whether @value{GDBN} will break on exceptions that happen
23083 inside the Cygwin DLL itself.
23085 @kindex set new-console
23086 @item set new-console @var{mode}
23087 If @var{mode} is @code{on} the debuggee will
23088 be started in a new console on next start.
23089 If @var{mode} is @code{off}, the debuggee will
23090 be started in the same console as the debugger.
23092 @kindex show new-console
23093 @item show new-console
23094 Displays whether a new console is used
23095 when the debuggee is started.
23097 @kindex set new-group
23098 @item set new-group @var{mode}
23099 This boolean value controls whether the debuggee should
23100 start a new group or stay in the same group as the debugger.
23101 This affects the way the Windows OS handles
23104 @kindex show new-group
23105 @item show new-group
23106 Displays current value of new-group boolean.
23108 @kindex set debugevents
23109 @item set debugevents
23110 This boolean value adds debug output concerning kernel events related
23111 to the debuggee seen by the debugger. This includes events that
23112 signal thread and process creation and exit, DLL loading and
23113 unloading, console interrupts, and debugging messages produced by the
23114 Windows @code{OutputDebugString} API call.
23116 @kindex set debugexec
23117 @item set debugexec
23118 This boolean value adds debug output concerning execute events
23119 (such as resume thread) seen by the debugger.
23121 @kindex set debugexceptions
23122 @item set debugexceptions
23123 This boolean value adds debug output concerning exceptions in the
23124 debuggee seen by the debugger.
23126 @kindex set debugmemory
23127 @item set debugmemory
23128 This boolean value adds debug output concerning debuggee memory reads
23129 and writes by the debugger.
23133 This boolean values specifies whether the debuggee is called
23134 via a shell or directly (default value is on).
23138 Displays if the debuggee will be started with a shell.
23143 * Non-debug DLL Symbols:: Support for DLLs without debugging symbols
23146 @node Non-debug DLL Symbols
23147 @subsubsection Support for DLLs without Debugging Symbols
23148 @cindex DLLs with no debugging symbols
23149 @cindex Minimal symbols and DLLs
23151 Very often on windows, some of the DLLs that your program relies on do
23152 not include symbolic debugging information (for example,
23153 @file{kernel32.dll}). When @value{GDBN} doesn't recognize any debugging
23154 symbols in a DLL, it relies on the minimal amount of symbolic
23155 information contained in the DLL's export table. This section
23156 describes working with such symbols, known internally to @value{GDBN} as
23157 ``minimal symbols''.
23159 Note that before the debugged program has started execution, no DLLs
23160 will have been loaded. The easiest way around this problem is simply to
23161 start the program --- either by setting a breakpoint or letting the
23162 program run once to completion.
23164 @subsubsection DLL Name Prefixes
23166 In keeping with the naming conventions used by the Microsoft debugging
23167 tools, DLL export symbols are made available with a prefix based on the
23168 DLL name, for instance @code{KERNEL32!CreateFileA}. The plain name is
23169 also entered into the symbol table, so @code{CreateFileA} is often
23170 sufficient. In some cases there will be name clashes within a program
23171 (particularly if the executable itself includes full debugging symbols)
23172 necessitating the use of the fully qualified name when referring to the
23173 contents of the DLL. Use single-quotes around the name to avoid the
23174 exclamation mark (``!'') being interpreted as a language operator.
23176 Note that the internal name of the DLL may be all upper-case, even
23177 though the file name of the DLL is lower-case, or vice-versa. Since
23178 symbols within @value{GDBN} are @emph{case-sensitive} this may cause
23179 some confusion. If in doubt, try the @code{info functions} and
23180 @code{info variables} commands or even @code{maint print msymbols}
23181 (@pxref{Symbols}). Here's an example:
23184 (@value{GDBP}) info function CreateFileA
23185 All functions matching regular expression "CreateFileA":
23187 Non-debugging symbols:
23188 0x77e885f4 CreateFileA
23189 0x77e885f4 KERNEL32!CreateFileA
23193 (@value{GDBP}) info function !
23194 All functions matching regular expression "!":
23196 Non-debugging symbols:
23197 0x6100114c cygwin1!__assert
23198 0x61004034 cygwin1!_dll_crt0@@0
23199 0x61004240 cygwin1!dll_crt0(per_process *)
23203 @subsubsection Working with Minimal Symbols
23205 Symbols extracted from a DLL's export table do not contain very much
23206 type information. All that @value{GDBN} can do is guess whether a symbol
23207 refers to a function or variable depending on the linker section that
23208 contains the symbol. Also note that the actual contents of the memory
23209 contained in a DLL are not available unless the program is running. This
23210 means that you cannot examine the contents of a variable or disassemble
23211 a function within a DLL without a running program.
23213 Variables are generally treated as pointers and dereferenced
23214 automatically. For this reason, it is often necessary to prefix a
23215 variable name with the address-of operator (``&'') and provide explicit
23216 type information in the command. Here's an example of the type of
23220 (@value{GDBP}) print 'cygwin1!__argv'
23221 'cygwin1!__argv' has unknown type; cast it to its declared type
23225 (@value{GDBP}) x 'cygwin1!__argv'
23226 'cygwin1!__argv' has unknown type; cast it to its declared type
23229 And two possible solutions:
23232 (@value{GDBP}) print ((char **)'cygwin1!__argv')[0]
23233 $2 = 0x22fd98 "/cygdrive/c/mydirectory/myprogram"
23237 (@value{GDBP}) x/2x &'cygwin1!__argv'
23238 0x610c0aa8 <cygwin1!__argv>: 0x10021608 0x00000000
23239 (@value{GDBP}) x/x 0x10021608
23240 0x10021608: 0x0022fd98
23241 (@value{GDBP}) x/s 0x0022fd98
23242 0x22fd98: "/cygdrive/c/mydirectory/myprogram"
23245 Setting a break point within a DLL is possible even before the program
23246 starts execution. However, under these circumstances, @value{GDBN} can't
23247 examine the initial instructions of the function in order to skip the
23248 function's frame set-up code. You can work around this by using ``*&''
23249 to set the breakpoint at a raw memory address:
23252 (@value{GDBP}) break *&'python22!PyOS_Readline'
23253 Breakpoint 1 at 0x1e04eff0
23256 The author of these extensions is not entirely convinced that setting a
23257 break point within a shared DLL like @file{kernel32.dll} is completely
23261 @subsection Commands Specific to @sc{gnu} Hurd Systems
23262 @cindex @sc{gnu} Hurd debugging
23264 This subsection describes @value{GDBN} commands specific to the
23265 @sc{gnu} Hurd native debugging.
23270 @kindex set signals@r{, Hurd command}
23271 @kindex set sigs@r{, Hurd command}
23272 This command toggles the state of inferior signal interception by
23273 @value{GDBN}. Mach exceptions, such as breakpoint traps, are not
23274 affected by this command. @code{sigs} is a shorthand alias for
23279 @kindex show signals@r{, Hurd command}
23280 @kindex show sigs@r{, Hurd command}
23281 Show the current state of intercepting inferior's signals.
23283 @item set signal-thread
23284 @itemx set sigthread
23285 @kindex set signal-thread
23286 @kindex set sigthread
23287 This command tells @value{GDBN} which thread is the @code{libc} signal
23288 thread. That thread is run when a signal is delivered to a running
23289 process. @code{set sigthread} is the shorthand alias of @code{set
23292 @item show signal-thread
23293 @itemx show sigthread
23294 @kindex show signal-thread
23295 @kindex show sigthread
23296 These two commands show which thread will run when the inferior is
23297 delivered a signal.
23300 @kindex set stopped@r{, Hurd command}
23301 This commands tells @value{GDBN} that the inferior process is stopped,
23302 as with the @code{SIGSTOP} signal. The stopped process can be
23303 continued by delivering a signal to it.
23306 @kindex show stopped@r{, Hurd command}
23307 This command shows whether @value{GDBN} thinks the debuggee is
23310 @item set exceptions
23311 @kindex set exceptions@r{, Hurd command}
23312 Use this command to turn off trapping of exceptions in the inferior.
23313 When exception trapping is off, neither breakpoints nor
23314 single-stepping will work. To restore the default, set exception
23317 @item show exceptions
23318 @kindex show exceptions@r{, Hurd command}
23319 Show the current state of trapping exceptions in the inferior.
23321 @item set task pause
23322 @kindex set task@r{, Hurd commands}
23323 @cindex task attributes (@sc{gnu} Hurd)
23324 @cindex pause current task (@sc{gnu} Hurd)
23325 This command toggles task suspension when @value{GDBN} has control.
23326 Setting it to on takes effect immediately, and the task is suspended
23327 whenever @value{GDBN} gets control. Setting it to off will take
23328 effect the next time the inferior is continued. If this option is set
23329 to off, you can use @code{set thread default pause on} or @code{set
23330 thread pause on} (see below) to pause individual threads.
23332 @item show task pause
23333 @kindex show task@r{, Hurd commands}
23334 Show the current state of task suspension.
23336 @item set task detach-suspend-count
23337 @cindex task suspend count
23338 @cindex detach from task, @sc{gnu} Hurd
23339 This command sets the suspend count the task will be left with when
23340 @value{GDBN} detaches from it.
23342 @item show task detach-suspend-count
23343 Show the suspend count the task will be left with when detaching.
23345 @item set task exception-port
23346 @itemx set task excp
23347 @cindex task exception port, @sc{gnu} Hurd
23348 This command sets the task exception port to which @value{GDBN} will
23349 forward exceptions. The argument should be the value of the @dfn{send
23350 rights} of the task. @code{set task excp} is a shorthand alias.
23352 @item set noninvasive
23353 @cindex noninvasive task options
23354 This command switches @value{GDBN} to a mode that is the least
23355 invasive as far as interfering with the inferior is concerned. This
23356 is the same as using @code{set task pause}, @code{set exceptions}, and
23357 @code{set signals} to values opposite to the defaults.
23359 @item info send-rights
23360 @itemx info receive-rights
23361 @itemx info port-rights
23362 @itemx info port-sets
23363 @itemx info dead-names
23366 @cindex send rights, @sc{gnu} Hurd
23367 @cindex receive rights, @sc{gnu} Hurd
23368 @cindex port rights, @sc{gnu} Hurd
23369 @cindex port sets, @sc{gnu} Hurd
23370 @cindex dead names, @sc{gnu} Hurd
23371 These commands display information about, respectively, send rights,
23372 receive rights, port rights, port sets, and dead names of a task.
23373 There are also shorthand aliases: @code{info ports} for @code{info
23374 port-rights} and @code{info psets} for @code{info port-sets}.
23376 @item set thread pause
23377 @kindex set thread@r{, Hurd command}
23378 @cindex thread properties, @sc{gnu} Hurd
23379 @cindex pause current thread (@sc{gnu} Hurd)
23380 This command toggles current thread suspension when @value{GDBN} has
23381 control. Setting it to on takes effect immediately, and the current
23382 thread is suspended whenever @value{GDBN} gets control. Setting it to
23383 off will take effect the next time the inferior is continued.
23384 Normally, this command has no effect, since when @value{GDBN} has
23385 control, the whole task is suspended. However, if you used @code{set
23386 task pause off} (see above), this command comes in handy to suspend
23387 only the current thread.
23389 @item show thread pause
23390 @kindex show thread@r{, Hurd command}
23391 This command shows the state of current thread suspension.
23393 @item set thread run
23394 This command sets whether the current thread is allowed to run.
23396 @item show thread run
23397 Show whether the current thread is allowed to run.
23399 @item set thread detach-suspend-count
23400 @cindex thread suspend count, @sc{gnu} Hurd
23401 @cindex detach from thread, @sc{gnu} Hurd
23402 This command sets the suspend count @value{GDBN} will leave on a
23403 thread when detaching. This number is relative to the suspend count
23404 found by @value{GDBN} when it notices the thread; use @code{set thread
23405 takeover-suspend-count} to force it to an absolute value.
23407 @item show thread detach-suspend-count
23408 Show the suspend count @value{GDBN} will leave on the thread when
23411 @item set thread exception-port
23412 @itemx set thread excp
23413 Set the thread exception port to which to forward exceptions. This
23414 overrides the port set by @code{set task exception-port} (see above).
23415 @code{set thread excp} is the shorthand alias.
23417 @item set thread takeover-suspend-count
23418 Normally, @value{GDBN}'s thread suspend counts are relative to the
23419 value @value{GDBN} finds when it notices each thread. This command
23420 changes the suspend counts to be absolute instead.
23422 @item set thread default
23423 @itemx show thread default
23424 @cindex thread default settings, @sc{gnu} Hurd
23425 Each of the above @code{set thread} commands has a @code{set thread
23426 default} counterpart (e.g., @code{set thread default pause}, @code{set
23427 thread default exception-port}, etc.). The @code{thread default}
23428 variety of commands sets the default thread properties for all
23429 threads; you can then change the properties of individual threads with
23430 the non-default commands.
23437 @value{GDBN} provides the following commands specific to the Darwin target:
23440 @item set debug darwin @var{num}
23441 @kindex set debug darwin
23442 When set to a non zero value, enables debugging messages specific to
23443 the Darwin support. Higher values produce more verbose output.
23445 @item show debug darwin
23446 @kindex show debug darwin
23447 Show the current state of Darwin messages.
23449 @item set debug mach-o @var{num}
23450 @kindex set debug mach-o
23451 When set to a non zero value, enables debugging messages while
23452 @value{GDBN} is reading Darwin object files. (@dfn{Mach-O} is the
23453 file format used on Darwin for object and executable files.) Higher
23454 values produce more verbose output. This is a command to diagnose
23455 problems internal to @value{GDBN} and should not be needed in normal
23458 @item show debug mach-o
23459 @kindex show debug mach-o
23460 Show the current state of Mach-O file messages.
23462 @item set mach-exceptions on
23463 @itemx set mach-exceptions off
23464 @kindex set mach-exceptions
23465 On Darwin, faults are first reported as a Mach exception and are then
23466 mapped to a Posix signal. Use this command to turn on trapping of
23467 Mach exceptions in the inferior. This might be sometimes useful to
23468 better understand the cause of a fault. The default is off.
23470 @item show mach-exceptions
23471 @kindex show mach-exceptions
23472 Show the current state of exceptions trapping.
23476 @subsection FreeBSD
23479 When the ABI of a system call is changed in the FreeBSD kernel, this
23480 is implemented by leaving a compatibility system call using the old
23481 ABI at the existing number and allocating a new system call number for
23482 the version using the new ABI. As a convenience, when a system call
23483 is caught by name (@pxref{catch syscall}), compatibility system calls
23486 For example, FreeBSD 12 introduced a new variant of the @code{kevent}
23487 system call and catching the @code{kevent} system call by name catches
23491 (@value{GDBP}) catch syscall kevent
23492 Catchpoint 1 (syscalls 'freebsd11_kevent' [363] 'kevent' [560])
23498 @section Embedded Operating Systems
23500 This section describes configurations involving the debugging of
23501 embedded operating systems that are available for several different
23504 @value{GDBN} includes the ability to debug programs running on
23505 various real-time operating systems.
23507 @node Embedded Processors
23508 @section Embedded Processors
23510 This section goes into details specific to particular embedded
23513 @cindex send command to simulator
23514 Whenever a specific embedded processor has a simulator, @value{GDBN}
23515 allows to send an arbitrary command to the simulator.
23518 @item sim @var{command}
23519 @kindex sim@r{, a command}
23520 Send an arbitrary @var{command} string to the simulator. Consult the
23521 documentation for the specific simulator in use for information about
23522 acceptable commands.
23527 * ARC:: Synopsys ARC
23529 * M68K:: Motorola M68K
23530 * MicroBlaze:: Xilinx MicroBlaze
23531 * MIPS Embedded:: MIPS Embedded
23532 * OpenRISC 1000:: OpenRISC 1000 (or1k)
23533 * PowerPC Embedded:: PowerPC Embedded
23536 * Super-H:: Renesas Super-H
23540 @subsection Synopsys ARC
23541 @cindex Synopsys ARC
23542 @cindex ARC specific commands
23548 @value{GDBN} provides the following ARC-specific commands:
23551 @item set debug arc
23552 @kindex set debug arc
23553 Control the level of ARC specific debug messages. Use 0 for no messages (the
23554 default), 1 for debug messages, and 2 for even more debug messages.
23556 @item show debug arc
23557 @kindex show debug arc
23558 Show the level of ARC specific debugging in operation.
23560 @item maint print arc arc-instruction @var{address}
23561 @kindex maint print arc arc-instruction
23562 Print internal disassembler information about instruction at a given address.
23569 @value{GDBN} provides the following ARM-specific commands:
23572 @item set arm disassembler
23574 This commands selects from a list of disassembly styles. The
23575 @code{"std"} style is the standard style.
23577 @item show arm disassembler
23579 Show the current disassembly style.
23581 @item set arm apcs32
23582 @cindex ARM 32-bit mode
23583 This command toggles ARM operation mode between 32-bit and 26-bit.
23585 @item show arm apcs32
23586 Display the current usage of the ARM 32-bit mode.
23588 @item set arm fpu @var{fputype}
23589 This command sets the ARM floating-point unit (FPU) type. The
23590 argument @var{fputype} can be one of these:
23594 Determine the FPU type by querying the OS ABI.
23596 Software FPU, with mixed-endian doubles on little-endian ARM
23599 GCC-compiled FPA co-processor.
23601 Software FPU with pure-endian doubles.
23607 Show the current type of the FPU.
23610 This command forces @value{GDBN} to use the specified ABI.
23613 Show the currently used ABI.
23615 @item set arm fallback-mode (arm|thumb|auto)
23616 @value{GDBN} uses the symbol table, when available, to determine
23617 whether instructions are ARM or Thumb. This command controls
23618 @value{GDBN}'s default behavior when the symbol table is not
23619 available. The default is @samp{auto}, which causes @value{GDBN} to
23620 use the current execution mode (from the @code{T} bit in the @code{CPSR}
23623 @item show arm fallback-mode
23624 Show the current fallback instruction mode.
23626 @item set arm force-mode (arm|thumb|auto)
23627 This command overrides use of the symbol table to determine whether
23628 instructions are ARM or Thumb. The default is @samp{auto}, which
23629 causes @value{GDBN} to use the symbol table and then the setting
23630 of @samp{set arm fallback-mode}.
23632 @item show arm force-mode
23633 Show the current forced instruction mode.
23635 @item set debug arm
23636 Toggle whether to display ARM-specific debugging messages from the ARM
23637 target support subsystem.
23639 @item show debug arm
23640 Show whether ARM-specific debugging messages are enabled.
23644 @item target sim @r{[}@var{simargs}@r{]} @dots{}
23645 The @value{GDBN} ARM simulator accepts the following optional arguments.
23648 @item --swi-support=@var{type}
23649 Tell the simulator which SWI interfaces to support. The argument
23650 @var{type} may be a comma separated list of the following values.
23651 The default value is @code{all}.
23666 The Motorola m68k configuration includes ColdFire support.
23669 @subsection MicroBlaze
23670 @cindex Xilinx MicroBlaze
23671 @cindex XMD, Xilinx Microprocessor Debugger
23673 The MicroBlaze is a soft-core processor supported on various Xilinx
23674 FPGAs, such as Spartan or Virtex series. Boards with these processors
23675 usually have JTAG ports which connect to a host system running the Xilinx
23676 Embedded Development Kit (EDK) or Software Development Kit (SDK).
23677 This host system is used to download the configuration bitstream to
23678 the target FPGA. The Xilinx Microprocessor Debugger (XMD) program
23679 communicates with the target board using the JTAG interface and
23680 presents a @code{gdbserver} interface to the board. By default
23681 @code{xmd} uses port @code{1234}. (While it is possible to change
23682 this default port, it requires the use of undocumented @code{xmd}
23683 commands. Contact Xilinx support if you need to do this.)
23685 Use these GDB commands to connect to the MicroBlaze target processor.
23688 @item target remote :1234
23689 Use this command to connect to the target if you are running @value{GDBN}
23690 on the same system as @code{xmd}.
23692 @item target remote @var{xmd-host}:1234
23693 Use this command to connect to the target if it is connected to @code{xmd}
23694 running on a different system named @var{xmd-host}.
23697 Use this command to download a program to the MicroBlaze target.
23699 @item set debug microblaze @var{n}
23700 Enable MicroBlaze-specific debugging messages if non-zero.
23702 @item show debug microblaze @var{n}
23703 Show MicroBlaze-specific debugging level.
23706 @node MIPS Embedded
23707 @subsection @acronym{MIPS} Embedded
23710 @value{GDBN} supports these special commands for @acronym{MIPS} targets:
23713 @item set mipsfpu double
23714 @itemx set mipsfpu single
23715 @itemx set mipsfpu none
23716 @itemx set mipsfpu auto
23717 @itemx show mipsfpu
23718 @kindex set mipsfpu
23719 @kindex show mipsfpu
23720 @cindex @acronym{MIPS} remote floating point
23721 @cindex floating point, @acronym{MIPS} remote
23722 If your target board does not support the @acronym{MIPS} floating point
23723 coprocessor, you should use the command @samp{set mipsfpu none} (if you
23724 need this, you may wish to put the command in your @value{GDBN} init
23725 file). This tells @value{GDBN} how to find the return value of
23726 functions which return floating point values. It also allows
23727 @value{GDBN} to avoid saving the floating point registers when calling
23728 functions on the board. If you are using a floating point coprocessor
23729 with only single precision floating point support, as on the @sc{r4650}
23730 processor, use the command @samp{set mipsfpu single}. The default
23731 double precision floating point coprocessor may be selected using
23732 @samp{set mipsfpu double}.
23734 In previous versions the only choices were double precision or no
23735 floating point, so @samp{set mipsfpu on} will select double precision
23736 and @samp{set mipsfpu off} will select no floating point.
23738 As usual, you can inquire about the @code{mipsfpu} variable with
23739 @samp{show mipsfpu}.
23742 @node OpenRISC 1000
23743 @subsection OpenRISC 1000
23744 @cindex OpenRISC 1000
23747 The OpenRISC 1000 provides a free RISC instruction set architecture. It is
23748 mainly provided as a soft-core which can run on Xilinx, Altera and other
23751 @value{GDBN} for OpenRISC supports the below commands when connecting to
23759 Runs the builtin CPU simulator which can run very basic
23760 programs but does not support most hardware functions like MMU.
23761 For more complex use cases the user is advised to run an external
23762 target, and connect using @samp{target remote}.
23764 Example: @code{target sim}
23766 @item set debug or1k
23767 Toggle whether to display OpenRISC-specific debugging messages from the
23768 OpenRISC target support subsystem.
23770 @item show debug or1k
23771 Show whether OpenRISC-specific debugging messages are enabled.
23774 @node PowerPC Embedded
23775 @subsection PowerPC Embedded
23777 @cindex DVC register
23778 @value{GDBN} supports using the DVC (Data Value Compare) register to
23779 implement in hardware simple hardware watchpoint conditions of the form:
23782 (@value{GDBP}) watch @var{ADDRESS|VARIABLE} \
23783 if @var{ADDRESS|VARIABLE} == @var{CONSTANT EXPRESSION}
23786 The DVC register will be automatically used when @value{GDBN} detects
23787 such pattern in a condition expression, and the created watchpoint uses one
23788 debug register (either the @code{exact-watchpoints} option is on and the
23789 variable is scalar, or the variable has a length of one byte). This feature
23790 is available in native @value{GDBN} running on a Linux kernel version 2.6.34
23793 When running on PowerPC embedded processors, @value{GDBN} automatically uses
23794 ranged hardware watchpoints, unless the @code{exact-watchpoints} option is on,
23795 in which case watchpoints using only one debug register are created when
23796 watching variables of scalar types.
23798 You can create an artificial array to watch an arbitrary memory
23799 region using one of the following commands (@pxref{Expressions}):
23802 (@value{GDBP}) watch *((char *) @var{address})@@@var{length}
23803 (@value{GDBP}) watch @{char[@var{length}]@} @var{address}
23806 PowerPC embedded processors support masked watchpoints. See the discussion
23807 about the @code{mask} argument in @ref{Set Watchpoints}.
23809 @cindex ranged breakpoint
23810 PowerPC embedded processors support hardware accelerated
23811 @dfn{ranged breakpoints}. A ranged breakpoint stops execution of
23812 the inferior whenever it executes an instruction at any address within
23813 the range it specifies. To set a ranged breakpoint in @value{GDBN},
23814 use the @code{break-range} command.
23816 @value{GDBN} provides the following PowerPC-specific commands:
23819 @kindex break-range
23820 @item break-range @var{start-location}, @var{end-location}
23821 Set a breakpoint for an address range given by
23822 @var{start-location} and @var{end-location}, which can specify a function name,
23823 a line number, an offset of lines from the current line or from the start
23824 location, or an address of an instruction (see @ref{Specify Location},
23825 for a list of all the possible ways to specify a @var{location}.)
23826 The breakpoint will stop execution of the inferior whenever it
23827 executes an instruction at any address within the specified range,
23828 (including @var{start-location} and @var{end-location}.)
23830 @kindex set powerpc
23831 @item set powerpc soft-float
23832 @itemx show powerpc soft-float
23833 Force @value{GDBN} to use (or not use) a software floating point calling
23834 convention. By default, @value{GDBN} selects the calling convention based
23835 on the selected architecture and the provided executable file.
23837 @item set powerpc vector-abi
23838 @itemx show powerpc vector-abi
23839 Force @value{GDBN} to use the specified calling convention for vector
23840 arguments and return values. The valid options are @samp{auto};
23841 @samp{generic}, to avoid vector registers even if they are present;
23842 @samp{altivec}, to use AltiVec registers; and @samp{spe} to use SPE
23843 registers. By default, @value{GDBN} selects the calling convention
23844 based on the selected architecture and the provided executable file.
23846 @item set powerpc exact-watchpoints
23847 @itemx show powerpc exact-watchpoints
23848 Allow @value{GDBN} to use only one debug register when watching a variable
23849 of scalar type, thus assuming that the variable is accessed through the
23850 address of its first byte.
23855 @subsection Atmel AVR
23858 When configured for debugging the Atmel AVR, @value{GDBN} supports the
23859 following AVR-specific commands:
23862 @item info io_registers
23863 @kindex info io_registers@r{, AVR}
23864 @cindex I/O registers (Atmel AVR)
23865 This command displays information about the AVR I/O registers. For
23866 each register, @value{GDBN} prints its number and value.
23873 When configured for debugging CRIS, @value{GDBN} provides the
23874 following CRIS-specific commands:
23877 @item set cris-version @var{ver}
23878 @cindex CRIS version
23879 Set the current CRIS version to @var{ver}, either @samp{10} or @samp{32}.
23880 The CRIS version affects register names and sizes. This command is useful in
23881 case autodetection of the CRIS version fails.
23883 @item show cris-version
23884 Show the current CRIS version.
23886 @item set cris-dwarf2-cfi
23887 @cindex DWARF-2 CFI and CRIS
23888 Set the usage of DWARF-2 CFI for CRIS debugging. The default is @samp{on}.
23889 Change to @samp{off} when using @code{gcc-cris} whose version is below
23892 @item show cris-dwarf2-cfi
23893 Show the current state of using DWARF-2 CFI.
23895 @item set cris-mode @var{mode}
23897 Set the current CRIS mode to @var{mode}. It should only be changed when
23898 debugging in guru mode, in which case it should be set to
23899 @samp{guru} (the default is @samp{normal}).
23901 @item show cris-mode
23902 Show the current CRIS mode.
23906 @subsection Renesas Super-H
23909 For the Renesas Super-H processor, @value{GDBN} provides these
23913 @item set sh calling-convention @var{convention}
23914 @kindex set sh calling-convention
23915 Set the calling-convention used when calling functions from @value{GDBN}.
23916 Allowed values are @samp{gcc}, which is the default setting, and @samp{renesas}.
23917 With the @samp{gcc} setting, functions are called using the @value{NGCC} calling
23918 convention. If the DWARF-2 information of the called function specifies
23919 that the function follows the Renesas calling convention, the function
23920 is called using the Renesas calling convention. If the calling convention
23921 is set to @samp{renesas}, the Renesas calling convention is always used,
23922 regardless of the DWARF-2 information. This can be used to override the
23923 default of @samp{gcc} if debug information is missing, or the compiler
23924 does not emit the DWARF-2 calling convention entry for a function.
23926 @item show sh calling-convention
23927 @kindex show sh calling-convention
23928 Show the current calling convention setting.
23933 @node Architectures
23934 @section Architectures
23936 This section describes characteristics of architectures that affect
23937 all uses of @value{GDBN} with the architecture, both native and cross.
23944 * HPPA:: HP PA architecture
23945 * SPU:: Cell Broadband Engine SPU architecture
23953 @subsection AArch64
23954 @cindex AArch64 support
23956 When @value{GDBN} is debugging the AArch64 architecture, it provides the
23957 following special commands:
23960 @item set debug aarch64
23961 @kindex set debug aarch64
23962 This command determines whether AArch64 architecture-specific debugging
23963 messages are to be displayed.
23965 @item show debug aarch64
23966 Show whether AArch64 debugging messages are displayed.
23970 @subsubsection AArch64 SVE.
23971 @cindex AArch64 SVE.
23973 When @value{GDBN} is debugging the AArch64 architecture, if the Scalable Vector
23974 Extension (SVE) is present, then @value{GDBN} will provide the vector registers
23975 @code{$z0} through @code{$z31}, vector predicate registers @code{$p0} through
23976 @code{$p15}, and the @code{$ffr} register. In addition, the pseudo register
23977 @code{$vg} will be provided. This is the vector granule for the current thread
23978 and represents the number of 64-bit chunks in an SVE @code{z} register.
23980 If the vector length changes, then the @code{$vg} register will be updated,
23981 but the lengths of the @code{z} and @code{p} registers will not change. This
23982 is a known limitation of @value{GDBN} and does not affect the execution of the
23987 @subsection x86 Architecture-specific Issues
23990 @item set struct-convention @var{mode}
23991 @kindex set struct-convention
23992 @cindex struct return convention
23993 @cindex struct/union returned in registers
23994 Set the convention used by the inferior to return @code{struct}s and
23995 @code{union}s from functions to @var{mode}. Possible values of
23996 @var{mode} are @code{"pcc"}, @code{"reg"}, and @code{"default"} (the
23997 default). @code{"default"} or @code{"pcc"} means that @code{struct}s
23998 are returned on the stack, while @code{"reg"} means that a
23999 @code{struct} or a @code{union} whose size is 1, 2, 4, or 8 bytes will
24000 be returned in a register.
24002 @item show struct-convention
24003 @kindex show struct-convention
24004 Show the current setting of the convention to return @code{struct}s
24009 @subsubsection Intel @dfn{Memory Protection Extensions} (MPX).
24010 @cindex Intel Memory Protection Extensions (MPX).
24012 Memory Protection Extension (MPX) adds the bound registers @samp{BND0}
24013 @footnote{The register named with capital letters represent the architecture
24014 registers.} through @samp{BND3}. Bound registers store a pair of 64-bit values
24015 which are the lower bound and upper bound. Bounds are effective addresses or
24016 memory locations. The upper bounds are architecturally represented in 1's
24017 complement form. A bound having lower bound = 0, and upper bound = 0
24018 (1's complement of all bits set) will allow access to the entire address space.
24020 @samp{BND0} through @samp{BND3} are represented in @value{GDBN} as @samp{bnd0raw}
24021 through @samp{bnd3raw}. Pseudo registers @samp{bnd0} through @samp{bnd3}
24022 display the upper bound performing the complement of one operation on the
24023 upper bound value, i.e.@ when upper bound in @samp{bnd0raw} is 0 in the
24024 @value{GDBN} @samp{bnd0} it will be @code{0xfff@dots{}}. In this sense it
24025 can also be noted that the upper bounds are inclusive.
24027 As an example, assume that the register BND0 holds bounds for a pointer having
24028 access allowed for the range between 0x32 and 0x71. The values present on
24029 bnd0raw and bnd registers are presented as follows:
24032 bnd0raw = @{0x32, 0xffffffff8e@}
24033 bnd0 = @{lbound = 0x32, ubound = 0x71@} : size 64
24036 This way the raw value can be accessed via bnd0raw@dots{}bnd3raw. Any
24037 change on bnd0@dots{}bnd3 or bnd0raw@dots{}bnd3raw is reflect on its
24038 counterpart. When the bnd0@dots{}bnd3 registers are displayed via
24039 Python, the display includes the memory size, in bits, accessible to
24042 Bounds can also be stored in bounds tables, which are stored in
24043 application memory. These tables store bounds for pointers by specifying
24044 the bounds pointer's value along with its bounds. Evaluating and changing
24045 bounds located in bound tables is therefore interesting while investigating
24046 bugs on MPX context. @value{GDBN} provides commands for this purpose:
24049 @item show mpx bound @var{pointer}
24050 @kindex show mpx bound
24051 Display bounds of the given @var{pointer}.
24053 @item set mpx bound @var{pointer}, @var{lbound}, @var{ubound}
24054 @kindex set mpx bound
24055 Set the bounds of a pointer in the bound table.
24056 This command takes three parameters: @var{pointer} is the pointers
24057 whose bounds are to be changed, @var{lbound} and @var{ubound} are new values
24058 for lower and upper bounds respectively.
24061 When you call an inferior function on an Intel MPX enabled program,
24062 GDB sets the inferior's bound registers to the init (disabled) state
24063 before calling the function. As a consequence, bounds checks for the
24064 pointer arguments passed to the function will always pass.
24066 This is necessary because when you call an inferior function, the
24067 program is usually in the middle of the execution of other function.
24068 Since at that point bound registers are in an arbitrary state, not
24069 clearing them would lead to random bound violations in the called
24072 You can still examine the influence of the bound registers on the
24073 execution of the called function by stopping the execution of the
24074 called function at its prologue, setting bound registers, and
24075 continuing the execution. For example:
24079 Breakpoint 2 at 0x4009de: file i386-mpx-call.c, line 47.
24080 $ print upper (a, b, c, d, 1)
24081 Breakpoint 2, upper (a=0x0, b=0x6e0000005b, c=0x0, d=0x0, len=48)....
24083 @{lbound = 0x0, ubound = ffffffff@} : size -1
24086 At this last step the value of bnd0 can be changed for investigation of bound
24087 violations caused along the execution of the call. In order to know how to
24088 set the bound registers or bound table for the call consult the ABI.
24093 See the following section.
24096 @subsection @acronym{MIPS}
24098 @cindex stack on Alpha
24099 @cindex stack on @acronym{MIPS}
24100 @cindex Alpha stack
24101 @cindex @acronym{MIPS} stack
24102 Alpha- and @acronym{MIPS}-based computers use an unusual stack frame, which
24103 sometimes requires @value{GDBN} to search backward in the object code to
24104 find the beginning of a function.
24106 @cindex response time, @acronym{MIPS} debugging
24107 To improve response time (especially for embedded applications, where
24108 @value{GDBN} may be restricted to a slow serial line for this search)
24109 you may want to limit the size of this search, using one of these
24113 @cindex @code{heuristic-fence-post} (Alpha, @acronym{MIPS})
24114 @item set heuristic-fence-post @var{limit}
24115 Restrict @value{GDBN} to examining at most @var{limit} bytes in its
24116 search for the beginning of a function. A value of @var{0} (the
24117 default) means there is no limit. However, except for @var{0}, the
24118 larger the limit the more bytes @code{heuristic-fence-post} must search
24119 and therefore the longer it takes to run. You should only need to use
24120 this command when debugging a stripped executable.
24122 @item show heuristic-fence-post
24123 Display the current limit.
24127 These commands are available @emph{only} when @value{GDBN} is configured
24128 for debugging programs on Alpha or @acronym{MIPS} processors.
24130 Several @acronym{MIPS}-specific commands are available when debugging @acronym{MIPS}
24134 @item set mips abi @var{arg}
24135 @kindex set mips abi
24136 @cindex set ABI for @acronym{MIPS}
24137 Tell @value{GDBN} which @acronym{MIPS} ABI is used by the inferior. Possible
24138 values of @var{arg} are:
24142 The default ABI associated with the current binary (this is the
24152 @item show mips abi
24153 @kindex show mips abi
24154 Show the @acronym{MIPS} ABI used by @value{GDBN} to debug the inferior.
24156 @item set mips compression @var{arg}
24157 @kindex set mips compression
24158 @cindex code compression, @acronym{MIPS}
24159 Tell @value{GDBN} which @acronym{MIPS} compressed
24160 @acronym{ISA, Instruction Set Architecture} encoding is used by the
24161 inferior. @value{GDBN} uses this for code disassembly and other
24162 internal interpretation purposes. This setting is only referred to
24163 when no executable has been associated with the debugging session or
24164 the executable does not provide information about the encoding it uses.
24165 Otherwise this setting is automatically updated from information
24166 provided by the executable.
24168 Possible values of @var{arg} are @samp{mips16} and @samp{micromips}.
24169 The default compressed @acronym{ISA} encoding is @samp{mips16}, as
24170 executables containing @acronym{MIPS16} code frequently are not
24171 identified as such.
24173 This setting is ``sticky''; that is, it retains its value across
24174 debugging sessions until reset either explicitly with this command or
24175 implicitly from an executable.
24177 The compiler and/or assembler typically add symbol table annotations to
24178 identify functions compiled for the @acronym{MIPS16} or
24179 @acronym{microMIPS} @acronym{ISA}s. If these function-scope annotations
24180 are present, @value{GDBN} uses them in preference to the global
24181 compressed @acronym{ISA} encoding setting.
24183 @item show mips compression
24184 @kindex show mips compression
24185 Show the @acronym{MIPS} compressed @acronym{ISA} encoding used by
24186 @value{GDBN} to debug the inferior.
24189 @itemx show mipsfpu
24190 @xref{MIPS Embedded, set mipsfpu}.
24192 @item set mips mask-address @var{arg}
24193 @kindex set mips mask-address
24194 @cindex @acronym{MIPS} addresses, masking
24195 This command determines whether the most-significant 32 bits of 64-bit
24196 @acronym{MIPS} addresses are masked off. The argument @var{arg} can be
24197 @samp{on}, @samp{off}, or @samp{auto}. The latter is the default
24198 setting, which lets @value{GDBN} determine the correct value.
24200 @item show mips mask-address
24201 @kindex show mips mask-address
24202 Show whether the upper 32 bits of @acronym{MIPS} addresses are masked off or
24205 @item set remote-mips64-transfers-32bit-regs
24206 @kindex set remote-mips64-transfers-32bit-regs
24207 This command controls compatibility with 64-bit @acronym{MIPS} targets that
24208 transfer data in 32-bit quantities. If you have an old @acronym{MIPS} 64 target
24209 that transfers 32 bits for some registers, like @sc{sr} and @sc{fsr},
24210 and 64 bits for other registers, set this option to @samp{on}.
24212 @item show remote-mips64-transfers-32bit-regs
24213 @kindex show remote-mips64-transfers-32bit-regs
24214 Show the current setting of compatibility with older @acronym{MIPS} 64 targets.
24216 @item set debug mips
24217 @kindex set debug mips
24218 This command turns on and off debugging messages for the @acronym{MIPS}-specific
24219 target code in @value{GDBN}.
24221 @item show debug mips
24222 @kindex show debug mips
24223 Show the current setting of @acronym{MIPS} debugging messages.
24229 @cindex HPPA support
24231 When @value{GDBN} is debugging the HP PA architecture, it provides the
24232 following special commands:
24235 @item set debug hppa
24236 @kindex set debug hppa
24237 This command determines whether HPPA architecture-specific debugging
24238 messages are to be displayed.
24240 @item show debug hppa
24241 Show whether HPPA debugging messages are displayed.
24243 @item maint print unwind @var{address}
24244 @kindex maint print unwind@r{, HPPA}
24245 This command displays the contents of the unwind table entry at the
24246 given @var{address}.
24252 @subsection Cell Broadband Engine SPU architecture
24253 @cindex Cell Broadband Engine
24256 When @value{GDBN} is debugging the Cell Broadband Engine SPU architecture,
24257 it provides the following special commands:
24260 @item info spu event
24262 Display SPU event facility status. Shows current event mask
24263 and pending event status.
24265 @item info spu signal
24266 Display SPU signal notification facility status. Shows pending
24267 signal-control word and signal notification mode of both signal
24268 notification channels.
24270 @item info spu mailbox
24271 Display SPU mailbox facility status. Shows all pending entries,
24272 in order of processing, in each of the SPU Write Outbound,
24273 SPU Write Outbound Interrupt, and SPU Read Inbound mailboxes.
24276 Display MFC DMA status. Shows all pending commands in the MFC
24277 DMA queue. For each entry, opcode, tag, class IDs, effective
24278 and local store addresses and transfer size are shown.
24280 @item info spu proxydma
24281 Display MFC Proxy-DMA status. Shows all pending commands in the MFC
24282 Proxy-DMA queue. For each entry, opcode, tag, class IDs, effective
24283 and local store addresses and transfer size are shown.
24287 When @value{GDBN} is debugging a combined PowerPC/SPU application
24288 on the Cell Broadband Engine, it provides in addition the following
24292 @item set spu stop-on-load @var{arg}
24294 Set whether to stop for new SPE threads. When set to @code{on}, @value{GDBN}
24295 will give control to the user when a new SPE thread enters its @code{main}
24296 function. The default is @code{off}.
24298 @item show spu stop-on-load
24300 Show whether to stop for new SPE threads.
24302 @item set spu auto-flush-cache @var{arg}
24303 Set whether to automatically flush the software-managed cache. When set to
24304 @code{on}, @value{GDBN} will automatically cause the SPE software-managed
24305 cache to be flushed whenever SPE execution stops. This provides a consistent
24306 view of PowerPC memory that is accessed via the cache. If an application
24307 does not use the software-managed cache, this option has no effect.
24309 @item show spu auto-flush-cache
24310 Show whether to automatically flush the software-managed cache.
24315 @subsection PowerPC
24316 @cindex PowerPC architecture
24318 When @value{GDBN} is debugging the PowerPC architecture, it provides a set of
24319 pseudo-registers to enable inspection of 128-bit wide Decimal Floating Point
24320 numbers stored in the floating point registers. These values must be stored
24321 in two consecutive registers, always starting at an even register like
24322 @code{f0} or @code{f2}.
24324 The pseudo-registers go from @code{$dl0} through @code{$dl15}, and are formed
24325 by joining the even/odd register pairs @code{f0} and @code{f1} for @code{$dl0},
24326 @code{f2} and @code{f3} for @code{$dl1} and so on.
24328 For POWER7 processors, @value{GDBN} provides a set of pseudo-registers, the 64-bit
24329 wide Extended Floating Point Registers (@samp{f32} through @samp{f63}).
24332 @subsection Nios II
24333 @cindex Nios II architecture
24335 When @value{GDBN} is debugging the Nios II architecture,
24336 it provides the following special commands:
24340 @item set debug nios2
24341 @kindex set debug nios2
24342 This command turns on and off debugging messages for the Nios II
24343 target code in @value{GDBN}.
24345 @item show debug nios2
24346 @kindex show debug nios2
24347 Show the current setting of Nios II debugging messages.
24351 @subsection Sparc64
24352 @cindex Sparc64 support
24353 @cindex Application Data Integrity
24354 @subsubsection ADI Support
24356 The M7 processor supports an Application Data Integrity (ADI) feature that
24357 detects invalid data accesses. When software allocates memory and enables
24358 ADI on the allocated memory, it chooses a 4-bit version number, sets the
24359 version in the upper 4 bits of the 64-bit pointer to that data, and stores
24360 the 4-bit version in every cacheline of that data. Hardware saves the latter
24361 in spare bits in the cache and memory hierarchy. On each load and store,
24362 the processor compares the upper 4 VA (virtual address) bits to the
24363 cacheline's version. If there is a mismatch, the processor generates a
24364 version mismatch trap which can be either precise or disrupting. The trap
24365 is an error condition which the kernel delivers to the process as a SIGSEGV
24368 Note that only 64-bit applications can use ADI and need to be built with
24371 Values of the ADI version tags, which are in granularity of a
24372 cacheline (64 bytes), can be viewed or modified.
24376 @kindex adi examine
24377 @item adi (examine | x) [ / @var{n} ] @var{addr}
24379 The @code{adi examine} command displays the value of one ADI version tag per
24382 @var{n} is a decimal integer specifying the number in bytes; the default
24383 is 1. It specifies how much ADI version information, at the ratio of 1:ADI
24384 block size, to display.
24386 @var{addr} is the address in user address space where you want @value{GDBN}
24387 to begin displaying the ADI version tags.
24389 Below is an example of displaying ADI versions of variable "shmaddr".
24392 (@value{GDBP}) adi x/100 shmaddr
24393 0xfff800010002c000: 0 0
24397 @item adi (assign | a) [ / @var{n} ] @var{addr} = @var{tag}
24399 The @code{adi assign} command is used to assign new ADI version tag
24402 @var{n} is a decimal integer specifying the number in bytes;
24403 the default is 1. It specifies how much ADI version information, at the
24404 ratio of 1:ADI block size, to modify.
24406 @var{addr} is the address in user address space where you want @value{GDBN}
24407 to begin modifying the ADI version tags.
24409 @var{tag} is the new ADI version tag.
24411 For example, do the following to modify then verify ADI versions of
24412 variable "shmaddr":
24415 (@value{GDBP}) adi a/100 shmaddr = 7
24416 (@value{GDBP}) adi x/100 shmaddr
24417 0xfff800010002c000: 7 7
24424 @cindex S12Z support
24426 When @value{GDBN} is debugging the S12Z architecture,
24427 it provides the following special command:
24430 @item maint info bdccsr
24431 @kindex maint info bdccsr@r{, S12Z}
24432 This command displays the current value of the microprocessor's
24437 @node Controlling GDB
24438 @chapter Controlling @value{GDBN}
24440 You can alter the way @value{GDBN} interacts with you by using the
24441 @code{set} command. For commands controlling how @value{GDBN} displays
24442 data, see @ref{Print Settings, ,Print Settings}. Other settings are
24447 * Editing:: Command editing
24448 * Command History:: Command history
24449 * Screen Size:: Screen size
24450 * Output Styling:: Output styling
24451 * Numbers:: Numbers
24452 * ABI:: Configuring the current ABI
24453 * Auto-loading:: Automatically loading associated files
24454 * Messages/Warnings:: Optional warnings and messages
24455 * Debugging Output:: Optional messages about internal happenings
24456 * Other Misc Settings:: Other Miscellaneous Settings
24464 @value{GDBN} indicates its readiness to read a command by printing a string
24465 called the @dfn{prompt}. This string is normally @samp{(@value{GDBP})}. You
24466 can change the prompt string with the @code{set prompt} command. For
24467 instance, when debugging @value{GDBN} with @value{GDBN}, it is useful to change
24468 the prompt in one of the @value{GDBN} sessions so that you can always tell
24469 which one you are talking to.
24471 @emph{Note:} @code{set prompt} does not add a space for you after the
24472 prompt you set. This allows you to set a prompt which ends in a space
24473 or a prompt that does not.
24477 @item set prompt @var{newprompt}
24478 Directs @value{GDBN} to use @var{newprompt} as its prompt string henceforth.
24480 @kindex show prompt
24482 Prints a line of the form: @samp{Gdb's prompt is: @var{your-prompt}}
24485 Versions of @value{GDBN} that ship with Python scripting enabled have
24486 prompt extensions. The commands for interacting with these extensions
24490 @kindex set extended-prompt
24491 @item set extended-prompt @var{prompt}
24492 Set an extended prompt that allows for substitutions.
24493 @xref{gdb.prompt}, for a list of escape sequences that can be used for
24494 substitution. Any escape sequences specified as part of the prompt
24495 string are replaced with the corresponding strings each time the prompt
24501 set extended-prompt Current working directory: \w (gdb)
24504 Note that when an extended-prompt is set, it takes control of the
24505 @var{prompt_hook} hook. @xref{prompt_hook}, for further information.
24507 @kindex show extended-prompt
24508 @item show extended-prompt
24509 Prints the extended prompt. Any escape sequences specified as part of
24510 the prompt string with @code{set extended-prompt}, are replaced with the
24511 corresponding strings each time the prompt is displayed.
24515 @section Command Editing
24517 @cindex command line editing
24519 @value{GDBN} reads its input commands via the @dfn{Readline} interface. This
24520 @sc{gnu} library provides consistent behavior for programs which provide a
24521 command line interface to the user. Advantages are @sc{gnu} Emacs-style
24522 or @dfn{vi}-style inline editing of commands, @code{csh}-like history
24523 substitution, and a storage and recall of command history across
24524 debugging sessions.
24526 You may control the behavior of command line editing in @value{GDBN} with the
24527 command @code{set}.
24530 @kindex set editing
24533 @itemx set editing on
24534 Enable command line editing (enabled by default).
24536 @item set editing off
24537 Disable command line editing.
24539 @kindex show editing
24541 Show whether command line editing is enabled.
24544 @ifset SYSTEM_READLINE
24545 @xref{Command Line Editing, , , rluserman, GNU Readline Library},
24547 @ifclear SYSTEM_READLINE
24548 @xref{Command Line Editing},
24550 for more details about the Readline
24551 interface. Users unfamiliar with @sc{gnu} Emacs or @code{vi} are
24552 encouraged to read that chapter.
24554 @node Command History
24555 @section Command History
24556 @cindex command history
24558 @value{GDBN} can keep track of the commands you type during your
24559 debugging sessions, so that you can be certain of precisely what
24560 happened. Use these commands to manage the @value{GDBN} command
24563 @value{GDBN} uses the @sc{gnu} History library, a part of the Readline
24564 package, to provide the history facility.
24565 @ifset SYSTEM_READLINE
24566 @xref{Using History Interactively, , , history, GNU History Library},
24568 @ifclear SYSTEM_READLINE
24569 @xref{Using History Interactively},
24571 for the detailed description of the History library.
24573 To issue a command to @value{GDBN} without affecting certain aspects of
24574 the state which is seen by users, prefix it with @samp{server }
24575 (@pxref{Server Prefix}). This
24576 means that this command will not affect the command history, nor will it
24577 affect @value{GDBN}'s notion of which command to repeat if @key{RET} is
24578 pressed on a line by itself.
24580 @cindex @code{server}, command prefix
24581 The server prefix does not affect the recording of values into the value
24582 history; to print a value without recording it into the value history,
24583 use the @code{output} command instead of the @code{print} command.
24585 Here is the description of @value{GDBN} commands related to command
24589 @cindex history substitution
24590 @cindex history file
24591 @kindex set history filename
24592 @cindex @env{GDBHISTFILE}, environment variable
24593 @item set history filename @var{fname}
24594 Set the name of the @value{GDBN} command history file to @var{fname}.
24595 This is the file where @value{GDBN} reads an initial command history
24596 list, and where it writes the command history from this session when it
24597 exits. You can access this list through history expansion or through
24598 the history command editing characters listed below. This file defaults
24599 to the value of the environment variable @code{GDBHISTFILE}, or to
24600 @file{./.gdb_history} (@file{./_gdb_history} on MS-DOS) if this variable
24603 @cindex save command history
24604 @kindex set history save
24605 @item set history save
24606 @itemx set history save on
24607 Record command history in a file, whose name may be specified with the
24608 @code{set history filename} command. By default, this option is disabled.
24610 @item set history save off
24611 Stop recording command history in a file.
24613 @cindex history size
24614 @kindex set history size
24615 @cindex @env{GDBHISTSIZE}, environment variable
24616 @item set history size @var{size}
24617 @itemx set history size unlimited
24618 Set the number of commands which @value{GDBN} keeps in its history list.
24619 This defaults to the value of the environment variable @env{GDBHISTSIZE}, or
24620 to 256 if this variable is not set. Non-numeric values of @env{GDBHISTSIZE}
24621 are ignored. If @var{size} is @code{unlimited} or if @env{GDBHISTSIZE} is
24622 either a negative number or the empty string, then the number of commands
24623 @value{GDBN} keeps in the history list is unlimited.
24625 @cindex remove duplicate history
24626 @kindex set history remove-duplicates
24627 @item set history remove-duplicates @var{count}
24628 @itemx set history remove-duplicates unlimited
24629 Control the removal of duplicate history entries in the command history list.
24630 If @var{count} is non-zero, @value{GDBN} will look back at the last @var{count}
24631 history entries and remove the first entry that is a duplicate of the current
24632 entry being added to the command history list. If @var{count} is
24633 @code{unlimited} then this lookbehind is unbounded. If @var{count} is 0, then
24634 removal of duplicate history entries is disabled.
24636 Only history entries added during the current session are considered for
24637 removal. This option is set to 0 by default.
24641 History expansion assigns special meaning to the character @kbd{!}.
24642 @ifset SYSTEM_READLINE
24643 @xref{Event Designators, , , history, GNU History Library},
24645 @ifclear SYSTEM_READLINE
24646 @xref{Event Designators},
24650 @cindex history expansion, turn on/off
24651 Since @kbd{!} is also the logical not operator in C, history expansion
24652 is off by default. If you decide to enable history expansion with the
24653 @code{set history expansion on} command, you may sometimes need to
24654 follow @kbd{!} (when it is used as logical not, in an expression) with
24655 a space or a tab to prevent it from being expanded. The readline
24656 history facilities do not attempt substitution on the strings
24657 @kbd{!=} and @kbd{!(}, even when history expansion is enabled.
24659 The commands to control history expansion are:
24662 @item set history expansion on
24663 @itemx set history expansion
24664 @kindex set history expansion
24665 Enable history expansion. History expansion is off by default.
24667 @item set history expansion off
24668 Disable history expansion.
24671 @kindex show history
24673 @itemx show history filename
24674 @itemx show history save
24675 @itemx show history size
24676 @itemx show history expansion
24677 These commands display the state of the @value{GDBN} history parameters.
24678 @code{show history} by itself displays all four states.
24683 @kindex show commands
24684 @cindex show last commands
24685 @cindex display command history
24686 @item show commands
24687 Display the last ten commands in the command history.
24689 @item show commands @var{n}
24690 Print ten commands centered on command number @var{n}.
24692 @item show commands +
24693 Print ten commands just after the commands last printed.
24697 @section Screen Size
24698 @cindex size of screen
24699 @cindex screen size
24702 @cindex pauses in output
24704 Certain commands to @value{GDBN} may produce large amounts of
24705 information output to the screen. To help you read all of it,
24706 @value{GDBN} pauses and asks you for input at the end of each page of
24707 output. Type @key{RET} when you want to see one more page of output,
24708 @kbd{q} to discard the remaining output, or @kbd{c} to continue
24709 without paging for the rest of the current command. Also, the screen
24710 width setting determines when to wrap lines of output. Depending on
24711 what is being printed, @value{GDBN} tries to break the line at a
24712 readable place, rather than simply letting it overflow onto the
24715 Normally @value{GDBN} knows the size of the screen from the terminal
24716 driver software. For example, on Unix @value{GDBN} uses the termcap data base
24717 together with the value of the @code{TERM} environment variable and the
24718 @code{stty rows} and @code{stty cols} settings. If this is not correct,
24719 you can override it with the @code{set height} and @code{set
24726 @kindex show height
24727 @item set height @var{lpp}
24728 @itemx set height unlimited
24730 @itemx set width @var{cpl}
24731 @itemx set width unlimited
24733 These @code{set} commands specify a screen height of @var{lpp} lines and
24734 a screen width of @var{cpl} characters. The associated @code{show}
24735 commands display the current settings.
24737 If you specify a height of either @code{unlimited} or zero lines,
24738 @value{GDBN} does not pause during output no matter how long the
24739 output is. This is useful if output is to a file or to an editor
24742 Likewise, you can specify @samp{set width unlimited} or @samp{set
24743 width 0} to prevent @value{GDBN} from wrapping its output.
24745 @item set pagination on
24746 @itemx set pagination off
24747 @kindex set pagination
24748 Turn the output pagination on or off; the default is on. Turning
24749 pagination off is the alternative to @code{set height unlimited}. Note that
24750 running @value{GDBN} with the @option{--batch} option (@pxref{Mode
24751 Options, -batch}) also automatically disables pagination.
24753 @item show pagination
24754 @kindex show pagination
24755 Show the current pagination mode.
24758 @node Output Styling
24759 @section Output Styling
24765 @value{GDBN} can style its output on a capable terminal. This is
24766 enabled by default on most systems, but disabled by default when in
24767 batch mode (@pxref{Mode Options}). Various style settings are available;
24768 and styles can also be disabled entirely.
24771 @item set style enabled @samp{on|off}
24772 Enable or disable all styling. The default is host-dependent, with
24773 most hosts defaulting to @samp{on}.
24775 @item show style enabled
24776 Show the current state of styling.
24778 @item set style sources @samp{on|off}
24779 Enable or disable source code styling. This affects whether source
24780 code, such as the output of the @code{list} command, is styled. Note
24781 that source styling only works if styling in general is enabled, and
24782 if @value{GDBN} was linked with the GNU Source Highlight library. The
24783 default is @samp{on}.
24785 @item show style sources
24786 Show the current state of source code styling.
24789 Subcommands of @code{set style} control specific forms of styling.
24790 These subcommands all follow the same pattern: each style-able object
24791 can be styled with a foreground color, a background color, and an
24794 For example, the style of file names can be controlled using the
24795 @code{set style filename} group of commands:
24798 @item set style filename background @var{color}
24799 Set the background to @var{color}. Valid colors are @samp{none}
24800 (meaning the terminal's default color), @samp{black}, @samp{red},
24801 @samp{green}, @samp{yellow}, @samp{blue}, @samp{magenta}, @samp{cyan},
24804 @item set style filename foreground @var{color}
24805 Set the foreground to @var{color}. Valid colors are @samp{none}
24806 (meaning the terminal's default color), @samp{black}, @samp{red},
24807 @samp{green}, @samp{yellow}, @samp{blue}, @samp{magenta}, @samp{cyan},
24810 @item set style filename intensity @var{value}
24811 Set the intensity to @var{value}. Valid intensities are @samp{normal}
24812 (the default), @samp{bold}, and @samp{dim}.
24815 The @code{show style} command and its subcommands are styling
24816 a style name in their output using its own style.
24817 So, use @command{show style} to see the complete list of styles,
24818 their characteristics and the visual aspect of each style.
24820 The style-able objects are:
24823 Control the styling of file names. By default, this style's
24824 foreground color is green.
24827 Control the styling of function names. These are managed with the
24828 @code{set style function} family of commands. By default, this
24829 style's foreground color is yellow.
24832 Control the styling of variable names. These are managed with the
24833 @code{set style variable} family of commands. By default, this style's
24834 foreground color is cyan.
24837 Control the styling of addresses. These are managed with the
24838 @code{set style address} family of commands. By default, this style's
24839 foreground color is blue.
24842 Control the styling of titles. These are managed with the
24843 @code{set style title} family of commands. By default, this style's
24844 intensity is bold. Commands are using the title style to improve
24845 the readibility of large output. For example, the commands
24846 @command{apropos} and @command{help} are using the title style
24847 for the command names.
24850 Control the styling of highlightings. These are managed with the
24851 @code{set style highlight} family of commands. By default, this style's
24852 foreground color is red. Commands are using the highlight style to draw
24853 the user attention to some specific parts of their output. For example,
24854 the command @command{apropos -v REGEXP} uses the highlight style to
24855 mark the documentation parts matching @var{regexp}.
24861 @cindex number representation
24862 @cindex entering numbers
24864 You can always enter numbers in octal, decimal, or hexadecimal in
24865 @value{GDBN} by the usual conventions: octal numbers begin with
24866 @samp{0}, decimal numbers end with @samp{.}, and hexadecimal numbers
24867 begin with @samp{0x}. Numbers that neither begin with @samp{0} or
24868 @samp{0x}, nor end with a @samp{.} are, by default, entered in base
24869 10; likewise, the default display for numbers---when no particular
24870 format is specified---is base 10. You can change the default base for
24871 both input and output with the commands described below.
24874 @kindex set input-radix
24875 @item set input-radix @var{base}
24876 Set the default base for numeric input. Supported choices
24877 for @var{base} are decimal 8, 10, or 16. The base must itself be
24878 specified either unambiguously or using the current input radix; for
24882 set input-radix 012
24883 set input-radix 10.
24884 set input-radix 0xa
24888 sets the input base to decimal. On the other hand, @samp{set input-radix 10}
24889 leaves the input radix unchanged, no matter what it was, since
24890 @samp{10}, being without any leading or trailing signs of its base, is
24891 interpreted in the current radix. Thus, if the current radix is 16,
24892 @samp{10} is interpreted in hex, i.e.@: as 16 decimal, which doesn't
24895 @kindex set output-radix
24896 @item set output-radix @var{base}
24897 Set the default base for numeric display. Supported choices
24898 for @var{base} are decimal 8, 10, or 16. The base must itself be
24899 specified either unambiguously or using the current input radix.
24901 @kindex show input-radix
24902 @item show input-radix
24903 Display the current default base for numeric input.
24905 @kindex show output-radix
24906 @item show output-radix
24907 Display the current default base for numeric display.
24909 @item set radix @r{[}@var{base}@r{]}
24913 These commands set and show the default base for both input and output
24914 of numbers. @code{set radix} sets the radix of input and output to
24915 the same base; without an argument, it resets the radix back to its
24916 default value of 10.
24921 @section Configuring the Current ABI
24923 @value{GDBN} can determine the @dfn{ABI} (Application Binary Interface) of your
24924 application automatically. However, sometimes you need to override its
24925 conclusions. Use these commands to manage @value{GDBN}'s view of the
24931 @cindex Newlib OS ABI and its influence on the longjmp handling
24933 One @value{GDBN} configuration can debug binaries for multiple operating
24934 system targets, either via remote debugging or native emulation.
24935 @value{GDBN} will autodetect the @dfn{OS ABI} (Operating System ABI) in use,
24936 but you can override its conclusion using the @code{set osabi} command.
24937 One example where this is useful is in debugging of binaries which use
24938 an alternate C library (e.g.@: @sc{uClibc} for @sc{gnu}/Linux) which does
24939 not have the same identifying marks that the standard C library for your
24942 When @value{GDBN} is debugging the AArch64 architecture, it provides a
24943 ``Newlib'' OS ABI. This is useful for handling @code{setjmp} and
24944 @code{longjmp} when debugging binaries that use the @sc{newlib} C library.
24945 The ``Newlib'' OS ABI can be selected by @code{set osabi Newlib}.
24949 Show the OS ABI currently in use.
24952 With no argument, show the list of registered available OS ABI's.
24954 @item set osabi @var{abi}
24955 Set the current OS ABI to @var{abi}.
24958 @cindex float promotion
24960 Generally, the way that an argument of type @code{float} is passed to a
24961 function depends on whether the function is prototyped. For a prototyped
24962 (i.e.@: ANSI/ISO style) function, @code{float} arguments are passed unchanged,
24963 according to the architecture's convention for @code{float}. For unprototyped
24964 (i.e.@: K&R style) functions, @code{float} arguments are first promoted to type
24965 @code{double} and then passed.
24967 Unfortunately, some forms of debug information do not reliably indicate whether
24968 a function is prototyped. If @value{GDBN} calls a function that is not marked
24969 as prototyped, it consults @kbd{set coerce-float-to-double}.
24972 @kindex set coerce-float-to-double
24973 @item set coerce-float-to-double
24974 @itemx set coerce-float-to-double on
24975 Arguments of type @code{float} will be promoted to @code{double} when passed
24976 to an unprototyped function. This is the default setting.
24978 @item set coerce-float-to-double off
24979 Arguments of type @code{float} will be passed directly to unprototyped
24982 @kindex show coerce-float-to-double
24983 @item show coerce-float-to-double
24984 Show the current setting of promoting @code{float} to @code{double}.
24988 @kindex show cp-abi
24989 @value{GDBN} needs to know the ABI used for your program's C@t{++}
24990 objects. The correct C@t{++} ABI depends on which C@t{++} compiler was
24991 used to build your application. @value{GDBN} only fully supports
24992 programs with a single C@t{++} ABI; if your program contains code using
24993 multiple C@t{++} ABI's or if @value{GDBN} can not identify your
24994 program's ABI correctly, you can tell @value{GDBN} which ABI to use.
24995 Currently supported ABI's include ``gnu-v2'', for @code{g++} versions
24996 before 3.0, ``gnu-v3'', for @code{g++} versions 3.0 and later, and
24997 ``hpaCC'' for the HP ANSI C@t{++} compiler. Other C@t{++} compilers may
24998 use the ``gnu-v2'' or ``gnu-v3'' ABI's as well. The default setting is
25003 Show the C@t{++} ABI currently in use.
25006 With no argument, show the list of supported C@t{++} ABI's.
25008 @item set cp-abi @var{abi}
25009 @itemx set cp-abi auto
25010 Set the current C@t{++} ABI to @var{abi}, or return to automatic detection.
25014 @section Automatically loading associated files
25015 @cindex auto-loading
25017 @value{GDBN} sometimes reads files with commands and settings automatically,
25018 without being explicitly told so by the user. We call this feature
25019 @dfn{auto-loading}. While auto-loading is useful for automatically adapting
25020 @value{GDBN} to the needs of your project, it can sometimes produce unexpected
25021 results or introduce security risks (e.g., if the file comes from untrusted
25025 * Init File in the Current Directory:: @samp{set/show/info auto-load local-gdbinit}
25026 * libthread_db.so.1 file:: @samp{set/show/info auto-load libthread-db}
25028 * Auto-loading safe path:: @samp{set/show/info auto-load safe-path}
25029 * Auto-loading verbose mode:: @samp{set/show debug auto-load}
25032 There are various kinds of files @value{GDBN} can automatically load.
25033 In addition to these files, @value{GDBN} supports auto-loading code written
25034 in various extension languages. @xref{Auto-loading extensions}.
25036 Note that loading of these associated files (including the local @file{.gdbinit}
25037 file) requires accordingly configured @code{auto-load safe-path}
25038 (@pxref{Auto-loading safe path}).
25040 For these reasons, @value{GDBN} includes commands and options to let you
25041 control when to auto-load files and which files should be auto-loaded.
25044 @anchor{set auto-load off}
25045 @kindex set auto-load off
25046 @item set auto-load off
25047 Globally disable loading of all auto-loaded files.
25048 You may want to use this command with the @samp{-iex} option
25049 (@pxref{Option -init-eval-command}) such as:
25051 $ @kbd{gdb -iex "set auto-load off" untrusted-executable corefile}
25054 Be aware that system init file (@pxref{System-wide configuration})
25055 and init files from your home directory (@pxref{Home Directory Init File})
25056 still get read (as they come from generally trusted directories).
25057 To prevent @value{GDBN} from auto-loading even those init files, use the
25058 @option{-nx} option (@pxref{Mode Options}), in addition to
25059 @code{set auto-load no}.
25061 @anchor{show auto-load}
25062 @kindex show auto-load
25063 @item show auto-load
25064 Show whether auto-loading of each specific @samp{auto-load} file(s) is enabled
25068 (gdb) show auto-load
25069 gdb-scripts: Auto-loading of canned sequences of commands scripts is on.
25070 libthread-db: Auto-loading of inferior specific libthread_db is on.
25071 local-gdbinit: Auto-loading of .gdbinit script from current directory
25073 python-scripts: Auto-loading of Python scripts is on.
25074 safe-path: List of directories from which it is safe to auto-load files
25075 is $debugdir:$datadir/auto-load.
25076 scripts-directory: List of directories from which to load auto-loaded scripts
25077 is $debugdir:$datadir/auto-load.
25080 @anchor{info auto-load}
25081 @kindex info auto-load
25082 @item info auto-load
25083 Print whether each specific @samp{auto-load} file(s) have been auto-loaded or
25087 (gdb) info auto-load
25090 Yes /home/user/gdb/gdb-gdb.gdb
25091 libthread-db: No auto-loaded libthread-db.
25092 local-gdbinit: Local .gdbinit file "/home/user/gdb/.gdbinit" has been
25096 Yes /home/user/gdb/gdb-gdb.py
25100 These are @value{GDBN} control commands for the auto-loading:
25102 @multitable @columnfractions .5 .5
25103 @item @xref{set auto-load off}.
25104 @tab Disable auto-loading globally.
25105 @item @xref{show auto-load}.
25106 @tab Show setting of all kinds of files.
25107 @item @xref{info auto-load}.
25108 @tab Show state of all kinds of files.
25109 @item @xref{set auto-load gdb-scripts}.
25110 @tab Control for @value{GDBN} command scripts.
25111 @item @xref{show auto-load gdb-scripts}.
25112 @tab Show setting of @value{GDBN} command scripts.
25113 @item @xref{info auto-load gdb-scripts}.
25114 @tab Show state of @value{GDBN} command scripts.
25115 @item @xref{set auto-load python-scripts}.
25116 @tab Control for @value{GDBN} Python scripts.
25117 @item @xref{show auto-load python-scripts}.
25118 @tab Show setting of @value{GDBN} Python scripts.
25119 @item @xref{info auto-load python-scripts}.
25120 @tab Show state of @value{GDBN} Python scripts.
25121 @item @xref{set auto-load guile-scripts}.
25122 @tab Control for @value{GDBN} Guile scripts.
25123 @item @xref{show auto-load guile-scripts}.
25124 @tab Show setting of @value{GDBN} Guile scripts.
25125 @item @xref{info auto-load guile-scripts}.
25126 @tab Show state of @value{GDBN} Guile scripts.
25127 @item @xref{set auto-load scripts-directory}.
25128 @tab Control for @value{GDBN} auto-loaded scripts location.
25129 @item @xref{show auto-load scripts-directory}.
25130 @tab Show @value{GDBN} auto-loaded scripts location.
25131 @item @xref{add-auto-load-scripts-directory}.
25132 @tab Add directory for auto-loaded scripts location list.
25133 @item @xref{set auto-load local-gdbinit}.
25134 @tab Control for init file in the current directory.
25135 @item @xref{show auto-load local-gdbinit}.
25136 @tab Show setting of init file in the current directory.
25137 @item @xref{info auto-load local-gdbinit}.
25138 @tab Show state of init file in the current directory.
25139 @item @xref{set auto-load libthread-db}.
25140 @tab Control for thread debugging library.
25141 @item @xref{show auto-load libthread-db}.
25142 @tab Show setting of thread debugging library.
25143 @item @xref{info auto-load libthread-db}.
25144 @tab Show state of thread debugging library.
25145 @item @xref{set auto-load safe-path}.
25146 @tab Control directories trusted for automatic loading.
25147 @item @xref{show auto-load safe-path}.
25148 @tab Show directories trusted for automatic loading.
25149 @item @xref{add-auto-load-safe-path}.
25150 @tab Add directory trusted for automatic loading.
25153 @node Init File in the Current Directory
25154 @subsection Automatically loading init file in the current directory
25155 @cindex auto-loading init file in the current directory
25157 By default, @value{GDBN} reads and executes the canned sequences of commands
25158 from init file (if any) in the current working directory,
25159 see @ref{Init File in the Current Directory during Startup}.
25161 Note that loading of this local @file{.gdbinit} file also requires accordingly
25162 configured @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
25165 @anchor{set auto-load local-gdbinit}
25166 @kindex set auto-load local-gdbinit
25167 @item set auto-load local-gdbinit [on|off]
25168 Enable or disable the auto-loading of canned sequences of commands
25169 (@pxref{Sequences}) found in init file in the current directory.
25171 @anchor{show auto-load local-gdbinit}
25172 @kindex show auto-load local-gdbinit
25173 @item show auto-load local-gdbinit
25174 Show whether auto-loading of canned sequences of commands from init file in the
25175 current directory is enabled or disabled.
25177 @anchor{info auto-load local-gdbinit}
25178 @kindex info auto-load local-gdbinit
25179 @item info auto-load local-gdbinit
25180 Print whether canned sequences of commands from init file in the
25181 current directory have been auto-loaded.
25184 @node libthread_db.so.1 file
25185 @subsection Automatically loading thread debugging library
25186 @cindex auto-loading libthread_db.so.1
25188 This feature is currently present only on @sc{gnu}/Linux native hosts.
25190 @value{GDBN} reads in some cases thread debugging library from places specific
25191 to the inferior (@pxref{set libthread-db-search-path}).
25193 The special @samp{libthread-db-search-path} entry @samp{$sdir} is processed
25194 without checking this @samp{set auto-load libthread-db} switch as system
25195 libraries have to be trusted in general. In all other cases of
25196 @samp{libthread-db-search-path} entries @value{GDBN} checks first if @samp{set
25197 auto-load libthread-db} is enabled before trying to open such thread debugging
25200 Note that loading of this debugging library also requires accordingly configured
25201 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
25204 @anchor{set auto-load libthread-db}
25205 @kindex set auto-load libthread-db
25206 @item set auto-load libthread-db [on|off]
25207 Enable or disable the auto-loading of inferior specific thread debugging library.
25209 @anchor{show auto-load libthread-db}
25210 @kindex show auto-load libthread-db
25211 @item show auto-load libthread-db
25212 Show whether auto-loading of inferior specific thread debugging library is
25213 enabled or disabled.
25215 @anchor{info auto-load libthread-db}
25216 @kindex info auto-load libthread-db
25217 @item info auto-load libthread-db
25218 Print the list of all loaded inferior specific thread debugging libraries and
25219 for each such library print list of inferior @var{pid}s using it.
25222 @node Auto-loading safe path
25223 @subsection Security restriction for auto-loading
25224 @cindex auto-loading safe-path
25226 As the files of inferior can come from untrusted source (such as submitted by
25227 an application user) @value{GDBN} does not always load any files automatically.
25228 @value{GDBN} provides the @samp{set auto-load safe-path} setting to list
25229 directories trusted for loading files not explicitly requested by user.
25230 Each directory can also be a shell wildcard pattern.
25232 If the path is not set properly you will see a warning and the file will not
25237 Reading symbols from /home/user/gdb/gdb...done.
25238 warning: File "/home/user/gdb/gdb-gdb.gdb" auto-loading has been
25239 declined by your `auto-load safe-path' set
25240 to "$debugdir:$datadir/auto-load".
25241 warning: File "/home/user/gdb/gdb-gdb.py" auto-loading has been
25242 declined by your `auto-load safe-path' set
25243 to "$debugdir:$datadir/auto-load".
25247 To instruct @value{GDBN} to go ahead and use the init files anyway,
25248 invoke @value{GDBN} like this:
25251 $ gdb -q -iex "set auto-load safe-path /home/user/gdb" ./gdb
25254 The list of trusted directories is controlled by the following commands:
25257 @anchor{set auto-load safe-path}
25258 @kindex set auto-load safe-path
25259 @item set auto-load safe-path @r{[}@var{directories}@r{]}
25260 Set the list of directories (and their subdirectories) trusted for automatic
25261 loading and execution of scripts. You can also enter a specific trusted file.
25262 Each directory can also be a shell wildcard pattern; wildcards do not match
25263 directory separator - see @code{FNM_PATHNAME} for system function @code{fnmatch}
25264 (@pxref{Wildcard Matching, fnmatch, , libc, GNU C Library Reference Manual}).
25265 If you omit @var{directories}, @samp{auto-load safe-path} will be reset to
25266 its default value as specified during @value{GDBN} compilation.
25268 The list of directories uses path separator (@samp{:} on GNU and Unix
25269 systems, @samp{;} on MS-Windows and MS-DOS) to separate directories, similarly
25270 to the @env{PATH} environment variable.
25272 @anchor{show auto-load safe-path}
25273 @kindex show auto-load safe-path
25274 @item show auto-load safe-path
25275 Show the list of directories trusted for automatic loading and execution of
25278 @anchor{add-auto-load-safe-path}
25279 @kindex add-auto-load-safe-path
25280 @item add-auto-load-safe-path
25281 Add an entry (or list of entries) to the list of directories trusted for
25282 automatic loading and execution of scripts. Multiple entries may be delimited
25283 by the host platform path separator in use.
25286 This variable defaults to what @code{--with-auto-load-dir} has been configured
25287 to (@pxref{with-auto-load-dir}). @file{$debugdir} and @file{$datadir}
25288 substitution applies the same as for @ref{set auto-load scripts-directory}.
25289 The default @code{set auto-load safe-path} value can be also overriden by
25290 @value{GDBN} configuration option @option{--with-auto-load-safe-path}.
25292 Setting this variable to @file{/} disables this security protection,
25293 corresponding @value{GDBN} configuration option is
25294 @option{--without-auto-load-safe-path}.
25295 This variable is supposed to be set to the system directories writable by the
25296 system superuser only. Users can add their source directories in init files in
25297 their home directories (@pxref{Home Directory Init File}). See also deprecated
25298 init file in the current directory
25299 (@pxref{Init File in the Current Directory during Startup}).
25301 To force @value{GDBN} to load the files it declined to load in the previous
25302 example, you could use one of the following ways:
25305 @item @file{~/.gdbinit}: @samp{add-auto-load-safe-path ~/src/gdb}
25306 Specify this trusted directory (or a file) as additional component of the list.
25307 You have to specify also any existing directories displayed by
25308 by @samp{show auto-load safe-path} (such as @samp{/usr:/bin} in this example).
25310 @item @kbd{gdb -iex "set auto-load safe-path /usr:/bin:~/src/gdb" @dots{}}
25311 Specify this directory as in the previous case but just for a single
25312 @value{GDBN} session.
25314 @item @kbd{gdb -iex "set auto-load safe-path /" @dots{}}
25315 Disable auto-loading safety for a single @value{GDBN} session.
25316 This assumes all the files you debug during this @value{GDBN} session will come
25317 from trusted sources.
25319 @item @kbd{./configure --without-auto-load-safe-path}
25320 During compilation of @value{GDBN} you may disable any auto-loading safety.
25321 This assumes all the files you will ever debug with this @value{GDBN} come from
25325 On the other hand you can also explicitly forbid automatic files loading which
25326 also suppresses any such warning messages:
25329 @item @kbd{gdb -iex "set auto-load no" @dots{}}
25330 You can use @value{GDBN} command-line option for a single @value{GDBN} session.
25332 @item @file{~/.gdbinit}: @samp{set auto-load no}
25333 Disable auto-loading globally for the user
25334 (@pxref{Home Directory Init File}). While it is improbable, you could also
25335 use system init file instead (@pxref{System-wide configuration}).
25338 This setting applies to the file names as entered by user. If no entry matches
25339 @value{GDBN} tries as a last resort to also resolve all the file names into
25340 their canonical form (typically resolving symbolic links) and compare the
25341 entries again. @value{GDBN} already canonicalizes most of the filenames on its
25342 own before starting the comparison so a canonical form of directories is
25343 recommended to be entered.
25345 @node Auto-loading verbose mode
25346 @subsection Displaying files tried for auto-load
25347 @cindex auto-loading verbose mode
25349 For better visibility of all the file locations where you can place scripts to
25350 be auto-loaded with inferior --- or to protect yourself against accidental
25351 execution of untrusted scripts --- @value{GDBN} provides a feature for printing
25352 all the files attempted to be loaded. Both existing and non-existing files may
25355 For example the list of directories from which it is safe to auto-load files
25356 (@pxref{Auto-loading safe path}) applies also to canonicalized filenames which
25357 may not be too obvious while setting it up.
25360 (gdb) set debug auto-load on
25361 (gdb) file ~/src/t/true
25362 auto-load: Loading canned sequences of commands script "/tmp/true-gdb.gdb"
25363 for objfile "/tmp/true".
25364 auto-load: Updating directories of "/usr:/opt".
25365 auto-load: Using directory "/usr".
25366 auto-load: Using directory "/opt".
25367 warning: File "/tmp/true-gdb.gdb" auto-loading has been declined
25368 by your `auto-load safe-path' set to "/usr:/opt".
25372 @anchor{set debug auto-load}
25373 @kindex set debug auto-load
25374 @item set debug auto-load [on|off]
25375 Set whether to print the filenames attempted to be auto-loaded.
25377 @anchor{show debug auto-load}
25378 @kindex show debug auto-load
25379 @item show debug auto-load
25380 Show whether printing of the filenames attempted to be auto-loaded is turned
25384 @node Messages/Warnings
25385 @section Optional Warnings and Messages
25387 @cindex verbose operation
25388 @cindex optional warnings
25389 By default, @value{GDBN} is silent about its inner workings. If you are
25390 running on a slow machine, you may want to use the @code{set verbose}
25391 command. This makes @value{GDBN} tell you when it does a lengthy
25392 internal operation, so you will not think it has crashed.
25394 Currently, the messages controlled by @code{set verbose} are those
25395 which announce that the symbol table for a source file is being read;
25396 see @code{symbol-file} in @ref{Files, ,Commands to Specify Files}.
25399 @kindex set verbose
25400 @item set verbose on
25401 Enables @value{GDBN} output of certain informational messages.
25403 @item set verbose off
25404 Disables @value{GDBN} output of certain informational messages.
25406 @kindex show verbose
25408 Displays whether @code{set verbose} is on or off.
25411 By default, if @value{GDBN} encounters bugs in the symbol table of an
25412 object file, it is silent; but if you are debugging a compiler, you may
25413 find this information useful (@pxref{Symbol Errors, ,Errors Reading
25418 @kindex set complaints
25419 @item set complaints @var{limit}
25420 Permits @value{GDBN} to output @var{limit} complaints about each type of
25421 unusual symbols before becoming silent about the problem. Set
25422 @var{limit} to zero to suppress all complaints; set it to a large number
25423 to prevent complaints from being suppressed.
25425 @kindex show complaints
25426 @item show complaints
25427 Displays how many symbol complaints @value{GDBN} is permitted to produce.
25431 @anchor{confirmation requests}
25432 By default, @value{GDBN} is cautious, and asks what sometimes seems to be a
25433 lot of stupid questions to confirm certain commands. For example, if
25434 you try to run a program which is already running:
25438 The program being debugged has been started already.
25439 Start it from the beginning? (y or n)
25442 If you are willing to unflinchingly face the consequences of your own
25443 commands, you can disable this ``feature'':
25447 @kindex set confirm
25449 @cindex confirmation
25450 @cindex stupid questions
25451 @item set confirm off
25452 Disables confirmation requests. Note that running @value{GDBN} with
25453 the @option{--batch} option (@pxref{Mode Options, -batch}) also
25454 automatically disables confirmation requests.
25456 @item set confirm on
25457 Enables confirmation requests (the default).
25459 @kindex show confirm
25461 Displays state of confirmation requests.
25465 @cindex command tracing
25466 If you need to debug user-defined commands or sourced files you may find it
25467 useful to enable @dfn{command tracing}. In this mode each command will be
25468 printed as it is executed, prefixed with one or more @samp{+} symbols, the
25469 quantity denoting the call depth of each command.
25472 @kindex set trace-commands
25473 @cindex command scripts, debugging
25474 @item set trace-commands on
25475 Enable command tracing.
25476 @item set trace-commands off
25477 Disable command tracing.
25478 @item show trace-commands
25479 Display the current state of command tracing.
25482 @node Debugging Output
25483 @section Optional Messages about Internal Happenings
25484 @cindex optional debugging messages
25486 @value{GDBN} has commands that enable optional debugging messages from
25487 various @value{GDBN} subsystems; normally these commands are of
25488 interest to @value{GDBN} maintainers, or when reporting a bug. This
25489 section documents those commands.
25492 @kindex set exec-done-display
25493 @item set exec-done-display
25494 Turns on or off the notification of asynchronous commands'
25495 completion. When on, @value{GDBN} will print a message when an
25496 asynchronous command finishes its execution. The default is off.
25497 @kindex show exec-done-display
25498 @item show exec-done-display
25499 Displays the current setting of asynchronous command completion
25502 @cindex ARM AArch64
25503 @item set debug aarch64
25504 Turns on or off display of debugging messages related to ARM AArch64.
25505 The default is off.
25507 @item show debug aarch64
25508 Displays the current state of displaying debugging messages related to
25510 @cindex gdbarch debugging info
25511 @cindex architecture debugging info
25512 @item set debug arch
25513 Turns on or off display of gdbarch debugging info. The default is off
25514 @item show debug arch
25515 Displays the current state of displaying gdbarch debugging info.
25516 @item set debug aix-solib
25517 @cindex AIX shared library debugging
25518 Control display of debugging messages from the AIX shared library
25519 support module. The default is off.
25520 @item show debug aix-thread
25521 Show the current state of displaying AIX shared library debugging messages.
25522 @item set debug aix-thread
25523 @cindex AIX threads
25524 Display debugging messages about inner workings of the AIX thread
25526 @item show debug aix-thread
25527 Show the current state of AIX thread debugging info display.
25528 @item set debug check-physname
25530 Check the results of the ``physname'' computation. When reading DWARF
25531 debugging information for C@t{++}, @value{GDBN} attempts to compute
25532 each entity's name. @value{GDBN} can do this computation in two
25533 different ways, depending on exactly what information is present.
25534 When enabled, this setting causes @value{GDBN} to compute the names
25535 both ways and display any discrepancies.
25536 @item show debug check-physname
25537 Show the current state of ``physname'' checking.
25538 @item set debug coff-pe-read
25539 @cindex COFF/PE exported symbols
25540 Control display of debugging messages related to reading of COFF/PE
25541 exported symbols. The default is off.
25542 @item show debug coff-pe-read
25543 Displays the current state of displaying debugging messages related to
25544 reading of COFF/PE exported symbols.
25545 @item set debug dwarf-die
25547 Dump DWARF DIEs after they are read in.
25548 The value is the number of nesting levels to print.
25549 A value of zero turns off the display.
25550 @item show debug dwarf-die
25551 Show the current state of DWARF DIE debugging.
25552 @item set debug dwarf-line
25553 @cindex DWARF Line Tables
25554 Turns on or off display of debugging messages related to reading
25555 DWARF line tables. The default is 0 (off).
25556 A value of 1 provides basic information.
25557 A value greater than 1 provides more verbose information.
25558 @item show debug dwarf-line
25559 Show the current state of DWARF line table debugging.
25560 @item set debug dwarf-read
25561 @cindex DWARF Reading
25562 Turns on or off display of debugging messages related to reading
25563 DWARF debug info. The default is 0 (off).
25564 A value of 1 provides basic information.
25565 A value greater than 1 provides more verbose information.
25566 @item show debug dwarf-read
25567 Show the current state of DWARF reader debugging.
25568 @item set debug displaced
25569 @cindex displaced stepping debugging info
25570 Turns on or off display of @value{GDBN} debugging info for the
25571 displaced stepping support. The default is off.
25572 @item show debug displaced
25573 Displays the current state of displaying @value{GDBN} debugging info
25574 related to displaced stepping.
25575 @item set debug event
25576 @cindex event debugging info
25577 Turns on or off display of @value{GDBN} event debugging info. The
25579 @item show debug event
25580 Displays the current state of displaying @value{GDBN} event debugging
25582 @item set debug expression
25583 @cindex expression debugging info
25584 Turns on or off display of debugging info about @value{GDBN}
25585 expression parsing. The default is off.
25586 @item show debug expression
25587 Displays the current state of displaying debugging info about
25588 @value{GDBN} expression parsing.
25589 @item set debug fbsd-lwp
25590 @cindex FreeBSD LWP debug messages
25591 Turns on or off debugging messages from the FreeBSD LWP debug support.
25592 @item show debug fbsd-lwp
25593 Show the current state of FreeBSD LWP debugging messages.
25594 @item set debug fbsd-nat
25595 @cindex FreeBSD native target debug messages
25596 Turns on or off debugging messages from the FreeBSD native target.
25597 @item show debug fbsd-nat
25598 Show the current state of FreeBSD native target debugging messages.
25599 @item set debug frame
25600 @cindex frame debugging info
25601 Turns on or off display of @value{GDBN} frame debugging info. The
25603 @item show debug frame
25604 Displays the current state of displaying @value{GDBN} frame debugging
25606 @item set debug gnu-nat
25607 @cindex @sc{gnu}/Hurd debug messages
25608 Turn on or off debugging messages from the @sc{gnu}/Hurd debug support.
25609 @item show debug gnu-nat
25610 Show the current state of @sc{gnu}/Hurd debugging messages.
25611 @item set debug infrun
25612 @cindex inferior debugging info
25613 Turns on or off display of @value{GDBN} debugging info for running the inferior.
25614 The default is off. @file{infrun.c} contains GDB's runtime state machine used
25615 for implementing operations such as single-stepping the inferior.
25616 @item show debug infrun
25617 Displays the current state of @value{GDBN} inferior debugging.
25618 @item set debug jit
25619 @cindex just-in-time compilation, debugging messages
25620 Turn on or off debugging messages from JIT debug support.
25621 @item show debug jit
25622 Displays the current state of @value{GDBN} JIT debugging.
25623 @item set debug lin-lwp
25624 @cindex @sc{gnu}/Linux LWP debug messages
25625 @cindex Linux lightweight processes
25626 Turn on or off debugging messages from the Linux LWP debug support.
25627 @item show debug lin-lwp
25628 Show the current state of Linux LWP debugging messages.
25629 @item set debug linux-namespaces
25630 @cindex @sc{gnu}/Linux namespaces debug messages
25631 Turn on or off debugging messages from the Linux namespaces debug support.
25632 @item show debug linux-namespaces
25633 Show the current state of Linux namespaces debugging messages.
25634 @item set debug mach-o
25635 @cindex Mach-O symbols processing
25636 Control display of debugging messages related to Mach-O symbols
25637 processing. The default is off.
25638 @item show debug mach-o
25639 Displays the current state of displaying debugging messages related to
25640 reading of COFF/PE exported symbols.
25641 @item set debug notification
25642 @cindex remote async notification debugging info
25643 Turn on or off debugging messages about remote async notification.
25644 The default is off.
25645 @item show debug notification
25646 Displays the current state of remote async notification debugging messages.
25647 @item set debug observer
25648 @cindex observer debugging info
25649 Turns on or off display of @value{GDBN} observer debugging. This
25650 includes info such as the notification of observable events.
25651 @item show debug observer
25652 Displays the current state of observer debugging.
25653 @item set debug overload
25654 @cindex C@t{++} overload debugging info
25655 Turns on or off display of @value{GDBN} C@t{++} overload debugging
25656 info. This includes info such as ranking of functions, etc. The default
25658 @item show debug overload
25659 Displays the current state of displaying @value{GDBN} C@t{++} overload
25661 @cindex expression parser, debugging info
25662 @cindex debug expression parser
25663 @item set debug parser
25664 Turns on or off the display of expression parser debugging output.
25665 Internally, this sets the @code{yydebug} variable in the expression
25666 parser. @xref{Tracing, , Tracing Your Parser, bison, Bison}, for
25667 details. The default is off.
25668 @item show debug parser
25669 Show the current state of expression parser debugging.
25670 @cindex packets, reporting on stdout
25671 @cindex serial connections, debugging
25672 @cindex debug remote protocol
25673 @cindex remote protocol debugging
25674 @cindex display remote packets
25675 @item set debug remote
25676 Turns on or off display of reports on all packets sent back and forth across
25677 the serial line to the remote machine. The info is printed on the
25678 @value{GDBN} standard output stream. The default is off.
25679 @item show debug remote
25680 Displays the state of display of remote packets.
25682 @item set debug separate-debug-file
25683 Turns on or off display of debug output about separate debug file search.
25684 @item show debug separate-debug-file
25685 Displays the state of separate debug file search debug output.
25687 @item set debug serial
25688 Turns on or off display of @value{GDBN} serial debugging info. The
25690 @item show debug serial
25691 Displays the current state of displaying @value{GDBN} serial debugging
25693 @item set debug solib-frv
25694 @cindex FR-V shared-library debugging
25695 Turn on or off debugging messages for FR-V shared-library code.
25696 @item show debug solib-frv
25697 Display the current state of FR-V shared-library code debugging
25699 @item set debug symbol-lookup
25700 @cindex symbol lookup
25701 Turns on or off display of debugging messages related to symbol lookup.
25702 The default is 0 (off).
25703 A value of 1 provides basic information.
25704 A value greater than 1 provides more verbose information.
25705 @item show debug symbol-lookup
25706 Show the current state of symbol lookup debugging messages.
25707 @item set debug symfile
25708 @cindex symbol file functions
25709 Turns on or off display of debugging messages related to symbol file functions.
25710 The default is off. @xref{Files}.
25711 @item show debug symfile
25712 Show the current state of symbol file debugging messages.
25713 @item set debug symtab-create
25714 @cindex symbol table creation
25715 Turns on or off display of debugging messages related to symbol table creation.
25716 The default is 0 (off).
25717 A value of 1 provides basic information.
25718 A value greater than 1 provides more verbose information.
25719 @item show debug symtab-create
25720 Show the current state of symbol table creation debugging.
25721 @item set debug target
25722 @cindex target debugging info
25723 Turns on or off display of @value{GDBN} target debugging info. This info
25724 includes what is going on at the target level of GDB, as it happens. The
25725 default is 0. Set it to 1 to track events, and to 2 to also track the
25726 value of large memory transfers.
25727 @item show debug target
25728 Displays the current state of displaying @value{GDBN} target debugging
25730 @item set debug timestamp
25731 @cindex timestampping debugging info
25732 Turns on or off display of timestamps with @value{GDBN} debugging info.
25733 When enabled, seconds and microseconds are displayed before each debugging
25735 @item show debug timestamp
25736 Displays the current state of displaying timestamps with @value{GDBN}
25738 @item set debug varobj
25739 @cindex variable object debugging info
25740 Turns on or off display of @value{GDBN} variable object debugging
25741 info. The default is off.
25742 @item show debug varobj
25743 Displays the current state of displaying @value{GDBN} variable object
25745 @item set debug xml
25746 @cindex XML parser debugging
25747 Turn on or off debugging messages for built-in XML parsers.
25748 @item show debug xml
25749 Displays the current state of XML debugging messages.
25752 @node Other Misc Settings
25753 @section Other Miscellaneous Settings
25754 @cindex miscellaneous settings
25757 @kindex set interactive-mode
25758 @item set interactive-mode
25759 If @code{on}, forces @value{GDBN} to assume that GDB was started
25760 in a terminal. In practice, this means that @value{GDBN} should wait
25761 for the user to answer queries generated by commands entered at
25762 the command prompt. If @code{off}, forces @value{GDBN} to operate
25763 in the opposite mode, and it uses the default answers to all queries.
25764 If @code{auto} (the default), @value{GDBN} tries to determine whether
25765 its standard input is a terminal, and works in interactive-mode if it
25766 is, non-interactively otherwise.
25768 In the vast majority of cases, the debugger should be able to guess
25769 correctly which mode should be used. But this setting can be useful
25770 in certain specific cases, such as running a MinGW @value{GDBN}
25771 inside a cygwin window.
25773 @kindex show interactive-mode
25774 @item show interactive-mode
25775 Displays whether the debugger is operating in interactive mode or not.
25778 @node Extending GDB
25779 @chapter Extending @value{GDBN}
25780 @cindex extending GDB
25782 @value{GDBN} provides several mechanisms for extension.
25783 @value{GDBN} also provides the ability to automatically load
25784 extensions when it reads a file for debugging. This allows the
25785 user to automatically customize @value{GDBN} for the program
25789 * Sequences:: Canned Sequences of @value{GDBN} Commands
25790 * Python:: Extending @value{GDBN} using Python
25791 * Guile:: Extending @value{GDBN} using Guile
25792 * Auto-loading extensions:: Automatically loading extensions
25793 * Multiple Extension Languages:: Working with multiple extension languages
25794 * Aliases:: Creating new spellings of existing commands
25797 To facilitate the use of extension languages, @value{GDBN} is capable
25798 of evaluating the contents of a file. When doing so, @value{GDBN}
25799 can recognize which extension language is being used by looking at
25800 the filename extension. Files with an unrecognized filename extension
25801 are always treated as a @value{GDBN} Command Files.
25802 @xref{Command Files,, Command files}.
25804 You can control how @value{GDBN} evaluates these files with the following
25808 @kindex set script-extension
25809 @kindex show script-extension
25810 @item set script-extension off
25811 All scripts are always evaluated as @value{GDBN} Command Files.
25813 @item set script-extension soft
25814 The debugger determines the scripting language based on filename
25815 extension. If this scripting language is supported, @value{GDBN}
25816 evaluates the script using that language. Otherwise, it evaluates
25817 the file as a @value{GDBN} Command File.
25819 @item set script-extension strict
25820 The debugger determines the scripting language based on filename
25821 extension, and evaluates the script using that language. If the
25822 language is not supported, then the evaluation fails.
25824 @item show script-extension
25825 Display the current value of the @code{script-extension} option.
25830 @section Canned Sequences of Commands
25832 Aside from breakpoint commands (@pxref{Break Commands, ,Breakpoint
25833 Command Lists}), @value{GDBN} provides two ways to store sequences of
25834 commands for execution as a unit: user-defined commands and command
25838 * Define:: How to define your own commands
25839 * Hooks:: Hooks for user-defined commands
25840 * Command Files:: How to write scripts of commands to be stored in a file
25841 * Output:: Commands for controlled output
25842 * Auto-loading sequences:: Controlling auto-loaded command files
25846 @subsection User-defined Commands
25848 @cindex user-defined command
25849 @cindex arguments, to user-defined commands
25850 A @dfn{user-defined command} is a sequence of @value{GDBN} commands to
25851 which you assign a new name as a command. This is done with the
25852 @code{define} command. User commands may accept an unlimited number of arguments
25853 separated by whitespace. Arguments are accessed within the user command
25854 via @code{$arg0@dots{}$argN}. A trivial example:
25858 print $arg0 + $arg1 + $arg2
25863 To execute the command use:
25870 This defines the command @code{adder}, which prints the sum of
25871 its three arguments. Note the arguments are text substitutions, so they may
25872 reference variables, use complex expressions, or even perform inferior
25875 @cindex argument count in user-defined commands
25876 @cindex how many arguments (user-defined commands)
25877 In addition, @code{$argc} may be used to find out how many arguments have
25883 print $arg0 + $arg1
25886 print $arg0 + $arg1 + $arg2
25891 Combining with the @code{eval} command (@pxref{eval}) makes it easier
25892 to process a variable number of arguments:
25899 eval "set $sum = $sum + $arg%d", $i
25909 @item define @var{commandname}
25910 Define a command named @var{commandname}. If there is already a command
25911 by that name, you are asked to confirm that you want to redefine it.
25912 The argument @var{commandname} may be a bare command name consisting of letters,
25913 numbers, dashes, and underscores. It may also start with any predefined
25914 prefix command. For example, @samp{define target my-target} creates
25915 a user-defined @samp{target my-target} command.
25917 The definition of the command is made up of other @value{GDBN} command lines,
25918 which are given following the @code{define} command. The end of these
25919 commands is marked by a line containing @code{end}.
25922 @kindex end@r{ (user-defined commands)}
25923 @item document @var{commandname}
25924 Document the user-defined command @var{commandname}, so that it can be
25925 accessed by @code{help}. The command @var{commandname} must already be
25926 defined. This command reads lines of documentation just as @code{define}
25927 reads the lines of the command definition, ending with @code{end}.
25928 After the @code{document} command is finished, @code{help} on command
25929 @var{commandname} displays the documentation you have written.
25931 You may use the @code{document} command again to change the
25932 documentation of a command. Redefining the command with @code{define}
25933 does not change the documentation.
25935 @kindex dont-repeat
25936 @cindex don't repeat command
25938 Used inside a user-defined command, this tells @value{GDBN} that this
25939 command should not be repeated when the user hits @key{RET}
25940 (@pxref{Command Syntax, repeat last command}).
25942 @kindex help user-defined
25943 @item help user-defined
25944 List all user-defined commands and all python commands defined in class
25945 COMAND_USER. The first line of the documentation or docstring is
25950 @itemx show user @var{commandname}
25951 Display the @value{GDBN} commands used to define @var{commandname} (but
25952 not its documentation). If no @var{commandname} is given, display the
25953 definitions for all user-defined commands.
25954 This does not work for user-defined python commands.
25956 @cindex infinite recursion in user-defined commands
25957 @kindex show max-user-call-depth
25958 @kindex set max-user-call-depth
25959 @item show max-user-call-depth
25960 @itemx set max-user-call-depth
25961 The value of @code{max-user-call-depth} controls how many recursion
25962 levels are allowed in user-defined commands before @value{GDBN} suspects an
25963 infinite recursion and aborts the command.
25964 This does not apply to user-defined python commands.
25967 In addition to the above commands, user-defined commands frequently
25968 use control flow commands, described in @ref{Command Files}.
25970 When user-defined commands are executed, the
25971 commands of the definition are not printed. An error in any command
25972 stops execution of the user-defined command.
25974 If used interactively, commands that would ask for confirmation proceed
25975 without asking when used inside a user-defined command. Many @value{GDBN}
25976 commands that normally print messages to say what they are doing omit the
25977 messages when used in a user-defined command.
25980 @subsection User-defined Command Hooks
25981 @cindex command hooks
25982 @cindex hooks, for commands
25983 @cindex hooks, pre-command
25986 You may define @dfn{hooks}, which are a special kind of user-defined
25987 command. Whenever you run the command @samp{foo}, if the user-defined
25988 command @samp{hook-foo} exists, it is executed (with no arguments)
25989 before that command.
25991 @cindex hooks, post-command
25993 A hook may also be defined which is run after the command you executed.
25994 Whenever you run the command @samp{foo}, if the user-defined command
25995 @samp{hookpost-foo} exists, it is executed (with no arguments) after
25996 that command. Post-execution hooks may exist simultaneously with
25997 pre-execution hooks, for the same command.
25999 It is valid for a hook to call the command which it hooks. If this
26000 occurs, the hook is not re-executed, thereby avoiding infinite recursion.
26002 @c It would be nice if hookpost could be passed a parameter indicating
26003 @c if the command it hooks executed properly or not. FIXME!
26005 @kindex stop@r{, a pseudo-command}
26006 In addition, a pseudo-command, @samp{stop} exists. Defining
26007 (@samp{hook-stop}) makes the associated commands execute every time
26008 execution stops in your program: before breakpoint commands are run,
26009 displays are printed, or the stack frame is printed.
26011 For example, to ignore @code{SIGALRM} signals while
26012 single-stepping, but treat them normally during normal execution,
26017 handle SIGALRM nopass
26021 handle SIGALRM pass
26024 define hook-continue
26025 handle SIGALRM pass
26029 As a further example, to hook at the beginning and end of the @code{echo}
26030 command, and to add extra text to the beginning and end of the message,
26038 define hookpost-echo
26042 (@value{GDBP}) echo Hello World
26043 <<<---Hello World--->>>
26048 You can define a hook for any single-word command in @value{GDBN}, but
26049 not for command aliases; you should define a hook for the basic command
26050 name, e.g.@: @code{backtrace} rather than @code{bt}.
26051 @c FIXME! So how does Joe User discover whether a command is an alias
26053 You can hook a multi-word command by adding @code{hook-} or
26054 @code{hookpost-} to the last word of the command, e.g.@:
26055 @samp{define target hook-remote} to add a hook to @samp{target remote}.
26057 If an error occurs during the execution of your hook, execution of
26058 @value{GDBN} commands stops and @value{GDBN} issues a prompt
26059 (before the command that you actually typed had a chance to run).
26061 If you try to define a hook which does not match any known command, you
26062 get a warning from the @code{define} command.
26064 @node Command Files
26065 @subsection Command Files
26067 @cindex command files
26068 @cindex scripting commands
26069 A command file for @value{GDBN} is a text file made of lines that are
26070 @value{GDBN} commands. Comments (lines starting with @kbd{#}) may
26071 also be included. An empty line in a command file does nothing; it
26072 does not mean to repeat the last command, as it would from the
26075 You can request the execution of a command file with the @code{source}
26076 command. Note that the @code{source} command is also used to evaluate
26077 scripts that are not Command Files. The exact behavior can be configured
26078 using the @code{script-extension} setting.
26079 @xref{Extending GDB,, Extending GDB}.
26083 @cindex execute commands from a file
26084 @item source [-s] [-v] @var{filename}
26085 Execute the command file @var{filename}.
26088 The lines in a command file are generally executed sequentially,
26089 unless the order of execution is changed by one of the
26090 @emph{flow-control commands} described below. The commands are not
26091 printed as they are executed. An error in any command terminates
26092 execution of the command file and control is returned to the console.
26094 @value{GDBN} first searches for @var{filename} in the current directory.
26095 If the file is not found there, and @var{filename} does not specify a
26096 directory, then @value{GDBN} also looks for the file on the source search path
26097 (specified with the @samp{directory} command);
26098 except that @file{$cdir} is not searched because the compilation directory
26099 is not relevant to scripts.
26101 If @code{-s} is specified, then @value{GDBN} searches for @var{filename}
26102 on the search path even if @var{filename} specifies a directory.
26103 The search is done by appending @var{filename} to each element of the
26104 search path. So, for example, if @var{filename} is @file{mylib/myscript}
26105 and the search path contains @file{/home/user} then @value{GDBN} will
26106 look for the script @file{/home/user/mylib/myscript}.
26107 The search is also done if @var{filename} is an absolute path.
26108 For example, if @var{filename} is @file{/tmp/myscript} and
26109 the search path contains @file{/home/user} then @value{GDBN} will
26110 look for the script @file{/home/user/tmp/myscript}.
26111 For DOS-like systems, if @var{filename} contains a drive specification,
26112 it is stripped before concatenation. For example, if @var{filename} is
26113 @file{d:myscript} and the search path contains @file{c:/tmp} then @value{GDBN}
26114 will look for the script @file{c:/tmp/myscript}.
26116 If @code{-v}, for verbose mode, is given then @value{GDBN} displays
26117 each command as it is executed. The option must be given before
26118 @var{filename}, and is interpreted as part of the filename anywhere else.
26120 Commands that would ask for confirmation if used interactively proceed
26121 without asking when used in a command file. Many @value{GDBN} commands that
26122 normally print messages to say what they are doing omit the messages
26123 when called from command files.
26125 @value{GDBN} also accepts command input from standard input. In this
26126 mode, normal output goes to standard output and error output goes to
26127 standard error. Errors in a command file supplied on standard input do
26128 not terminate execution of the command file---execution continues with
26132 gdb < cmds > log 2>&1
26135 (The syntax above will vary depending on the shell used.) This example
26136 will execute commands from the file @file{cmds}. All output and errors
26137 would be directed to @file{log}.
26139 Since commands stored on command files tend to be more general than
26140 commands typed interactively, they frequently need to deal with
26141 complicated situations, such as different or unexpected values of
26142 variables and symbols, changes in how the program being debugged is
26143 built, etc. @value{GDBN} provides a set of flow-control commands to
26144 deal with these complexities. Using these commands, you can write
26145 complex scripts that loop over data structures, execute commands
26146 conditionally, etc.
26153 This command allows to include in your script conditionally executed
26154 commands. The @code{if} command takes a single argument, which is an
26155 expression to evaluate. It is followed by a series of commands that
26156 are executed only if the expression is true (its value is nonzero).
26157 There can then optionally be an @code{else} line, followed by a series
26158 of commands that are only executed if the expression was false. The
26159 end of the list is marked by a line containing @code{end}.
26163 This command allows to write loops. Its syntax is similar to
26164 @code{if}: the command takes a single argument, which is an expression
26165 to evaluate, and must be followed by the commands to execute, one per
26166 line, terminated by an @code{end}. These commands are called the
26167 @dfn{body} of the loop. The commands in the body of @code{while} are
26168 executed repeatedly as long as the expression evaluates to true.
26172 This command exits the @code{while} loop in whose body it is included.
26173 Execution of the script continues after that @code{while}s @code{end}
26176 @kindex loop_continue
26177 @item loop_continue
26178 This command skips the execution of the rest of the body of commands
26179 in the @code{while} loop in whose body it is included. Execution
26180 branches to the beginning of the @code{while} loop, where it evaluates
26181 the controlling expression.
26183 @kindex end@r{ (if/else/while commands)}
26185 Terminate the block of commands that are the body of @code{if},
26186 @code{else}, or @code{while} flow-control commands.
26191 @subsection Commands for Controlled Output
26193 During the execution of a command file or a user-defined command, normal
26194 @value{GDBN} output is suppressed; the only output that appears is what is
26195 explicitly printed by the commands in the definition. This section
26196 describes three commands useful for generating exactly the output you
26201 @item echo @var{text}
26202 @c I do not consider backslash-space a standard C escape sequence
26203 @c because it is not in ANSI.
26204 Print @var{text}. Nonprinting characters can be included in
26205 @var{text} using C escape sequences, such as @samp{\n} to print a
26206 newline. @strong{No newline is printed unless you specify one.}
26207 In addition to the standard C escape sequences, a backslash followed
26208 by a space stands for a space. This is useful for displaying a
26209 string with spaces at the beginning or the end, since leading and
26210 trailing spaces are otherwise trimmed from all arguments.
26211 To print @samp{@w{ }and foo =@w{ }}, use the command
26212 @samp{echo \@w{ }and foo = \@w{ }}.
26214 A backslash at the end of @var{text} can be used, as in C, to continue
26215 the command onto subsequent lines. For example,
26218 echo This is some text\n\
26219 which is continued\n\
26220 onto several lines.\n
26223 produces the same output as
26226 echo This is some text\n
26227 echo which is continued\n
26228 echo onto several lines.\n
26232 @item output @var{expression}
26233 Print the value of @var{expression} and nothing but that value: no
26234 newlines, no @samp{$@var{nn} = }. The value is not entered in the
26235 value history either. @xref{Expressions, ,Expressions}, for more information
26238 @item output/@var{fmt} @var{expression}
26239 Print the value of @var{expression} in format @var{fmt}. You can use
26240 the same formats as for @code{print}. @xref{Output Formats,,Output
26241 Formats}, for more information.
26244 @item printf @var{template}, @var{expressions}@dots{}
26245 Print the values of one or more @var{expressions} under the control of
26246 the string @var{template}. To print several values, make
26247 @var{expressions} be a comma-separated list of individual expressions,
26248 which may be either numbers or pointers. Their values are printed as
26249 specified by @var{template}, exactly as a C program would do by
26250 executing the code below:
26253 printf (@var{template}, @var{expressions}@dots{});
26256 As in @code{C} @code{printf}, ordinary characters in @var{template}
26257 are printed verbatim, while @dfn{conversion specification} introduced
26258 by the @samp{%} character cause subsequent @var{expressions} to be
26259 evaluated, their values converted and formatted according to type and
26260 style information encoded in the conversion specifications, and then
26263 For example, you can print two values in hex like this:
26266 printf "foo, bar-foo = 0x%x, 0x%x\n", foo, bar-foo
26269 @code{printf} supports all the standard @code{C} conversion
26270 specifications, including the flags and modifiers between the @samp{%}
26271 character and the conversion letter, with the following exceptions:
26275 The argument-ordering modifiers, such as @samp{2$}, are not supported.
26278 The modifier @samp{*} is not supported for specifying precision or
26282 The @samp{'} flag (for separation of digits into groups according to
26283 @code{LC_NUMERIC'}) is not supported.
26286 The type modifiers @samp{hh}, @samp{j}, @samp{t}, and @samp{z} are not
26290 The conversion letter @samp{n} (as in @samp{%n}) is not supported.
26293 The conversion letters @samp{a} and @samp{A} are not supported.
26297 Note that the @samp{ll} type modifier is supported only if the
26298 underlying @code{C} implementation used to build @value{GDBN} supports
26299 the @code{long long int} type, and the @samp{L} type modifier is
26300 supported only if @code{long double} type is available.
26302 As in @code{C}, @code{printf} supports simple backslash-escape
26303 sequences, such as @code{\n}, @samp{\t}, @samp{\\}, @samp{\"},
26304 @samp{\a}, and @samp{\f}, that consist of backslash followed by a
26305 single character. Octal and hexadecimal escape sequences are not
26308 Additionally, @code{printf} supports conversion specifications for DFP
26309 (@dfn{Decimal Floating Point}) types using the following length modifiers
26310 together with a floating point specifier.
26315 @samp{H} for printing @code{Decimal32} types.
26318 @samp{D} for printing @code{Decimal64} types.
26321 @samp{DD} for printing @code{Decimal128} types.
26324 If the underlying @code{C} implementation used to build @value{GDBN} has
26325 support for the three length modifiers for DFP types, other modifiers
26326 such as width and precision will also be available for @value{GDBN} to use.
26328 In case there is no such @code{C} support, no additional modifiers will be
26329 available and the value will be printed in the standard way.
26331 Here's an example of printing DFP types using the above conversion letters:
26333 printf "D32: %Hf - D64: %Df - D128: %DDf\n",1.2345df,1.2E10dd,1.2E1dl
26338 @item eval @var{template}, @var{expressions}@dots{}
26339 Convert the values of one or more @var{expressions} under the control of
26340 the string @var{template} to a command line, and call it.
26344 @node Auto-loading sequences
26345 @subsection Controlling auto-loading native @value{GDBN} scripts
26346 @cindex native script auto-loading
26348 When a new object file is read (for example, due to the @code{file}
26349 command, or because the inferior has loaded a shared library),
26350 @value{GDBN} will look for the command file @file{@var{objfile}-gdb.gdb}.
26351 @xref{Auto-loading extensions}.
26353 Auto-loading can be enabled or disabled,
26354 and the list of auto-loaded scripts can be printed.
26357 @anchor{set auto-load gdb-scripts}
26358 @kindex set auto-load gdb-scripts
26359 @item set auto-load gdb-scripts [on|off]
26360 Enable or disable the auto-loading of canned sequences of commands scripts.
26362 @anchor{show auto-load gdb-scripts}
26363 @kindex show auto-load gdb-scripts
26364 @item show auto-load gdb-scripts
26365 Show whether auto-loading of canned sequences of commands scripts is enabled or
26368 @anchor{info auto-load gdb-scripts}
26369 @kindex info auto-load gdb-scripts
26370 @cindex print list of auto-loaded canned sequences of commands scripts
26371 @item info auto-load gdb-scripts [@var{regexp}]
26372 Print the list of all canned sequences of commands scripts that @value{GDBN}
26376 If @var{regexp} is supplied only canned sequences of commands scripts with
26377 matching names are printed.
26379 @c Python docs live in a separate file.
26380 @include python.texi
26382 @c Guile docs live in a separate file.
26383 @include guile.texi
26385 @node Auto-loading extensions
26386 @section Auto-loading extensions
26387 @cindex auto-loading extensions
26389 @value{GDBN} provides two mechanisms for automatically loading extensions
26390 when a new object file is read (for example, due to the @code{file}
26391 command, or because the inferior has loaded a shared library):
26392 @file{@var{objfile}-gdb.@var{ext}} and the @code{.debug_gdb_scripts}
26393 section of modern file formats like ELF.
26396 * objfile-gdb.ext file: objfile-gdbdotext file. The @file{@var{objfile}-gdb.@var{ext}} file
26397 * .debug_gdb_scripts section: dotdebug_gdb_scripts section. The @code{.debug_gdb_scripts} section
26398 * Which flavor to choose?::
26401 The auto-loading feature is useful for supplying application-specific
26402 debugging commands and features.
26404 Auto-loading can be enabled or disabled,
26405 and the list of auto-loaded scripts can be printed.
26406 See the @samp{auto-loading} section of each extension language
26407 for more information.
26408 For @value{GDBN} command files see @ref{Auto-loading sequences}.
26409 For Python files see @ref{Python Auto-loading}.
26411 Note that loading of this script file also requires accordingly configured
26412 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
26414 @node objfile-gdbdotext file
26415 @subsection The @file{@var{objfile}-gdb.@var{ext}} file
26416 @cindex @file{@var{objfile}-gdb.gdb}
26417 @cindex @file{@var{objfile}-gdb.py}
26418 @cindex @file{@var{objfile}-gdb.scm}
26420 When a new object file is read, @value{GDBN} looks for a file named
26421 @file{@var{objfile}-gdb.@var{ext}} (we call it @var{script-name} below),
26422 where @var{objfile} is the object file's name and
26423 where @var{ext} is the file extension for the extension language:
26426 @item @file{@var{objfile}-gdb.gdb}
26427 GDB's own command language
26428 @item @file{@var{objfile}-gdb.py}
26430 @item @file{@var{objfile}-gdb.scm}
26434 @var{script-name} is formed by ensuring that the file name of @var{objfile}
26435 is absolute, following all symlinks, and resolving @code{.} and @code{..}
26436 components, and appending the @file{-gdb.@var{ext}} suffix.
26437 If this file exists and is readable, @value{GDBN} will evaluate it as a
26438 script in the specified extension language.
26440 If this file does not exist, then @value{GDBN} will look for
26441 @var{script-name} file in all of the directories as specified below.
26443 Note that loading of these files requires an accordingly configured
26444 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
26446 For object files using @file{.exe} suffix @value{GDBN} tries to load first the
26447 scripts normally according to its @file{.exe} filename. But if no scripts are
26448 found @value{GDBN} also tries script filenames matching the object file without
26449 its @file{.exe} suffix. This @file{.exe} stripping is case insensitive and it
26450 is attempted on any platform. This makes the script filenames compatible
26451 between Unix and MS-Windows hosts.
26454 @anchor{set auto-load scripts-directory}
26455 @kindex set auto-load scripts-directory
26456 @item set auto-load scripts-directory @r{[}@var{directories}@r{]}
26457 Control @value{GDBN} auto-loaded scripts location. Multiple directory entries
26458 may be delimited by the host platform path separator in use
26459 (@samp{:} on Unix, @samp{;} on MS-Windows and MS-DOS).
26461 Each entry here needs to be covered also by the security setting
26462 @code{set auto-load safe-path} (@pxref{set auto-load safe-path}).
26464 @anchor{with-auto-load-dir}
26465 This variable defaults to @file{$debugdir:$datadir/auto-load}. The default
26466 @code{set auto-load safe-path} value can be also overriden by @value{GDBN}
26467 configuration option @option{--with-auto-load-dir}.
26469 Any reference to @file{$debugdir} will get replaced by
26470 @var{debug-file-directory} value (@pxref{Separate Debug Files}) and any
26471 reference to @file{$datadir} will get replaced by @var{data-directory} which is
26472 determined at @value{GDBN} startup (@pxref{Data Files}). @file{$debugdir} and
26473 @file{$datadir} must be placed as a directory component --- either alone or
26474 delimited by @file{/} or @file{\} directory separators, depending on the host
26477 The list of directories uses path separator (@samp{:} on GNU and Unix
26478 systems, @samp{;} on MS-Windows and MS-DOS) to separate directories, similarly
26479 to the @env{PATH} environment variable.
26481 @anchor{show auto-load scripts-directory}
26482 @kindex show auto-load scripts-directory
26483 @item show auto-load scripts-directory
26484 Show @value{GDBN} auto-loaded scripts location.
26486 @anchor{add-auto-load-scripts-directory}
26487 @kindex add-auto-load-scripts-directory
26488 @item add-auto-load-scripts-directory @r{[}@var{directories}@dots{}@r{]}
26489 Add an entry (or list of entries) to the list of auto-loaded scripts locations.
26490 Multiple entries may be delimited by the host platform path separator in use.
26493 @value{GDBN} does not track which files it has already auto-loaded this way.
26494 @value{GDBN} will load the associated script every time the corresponding
26495 @var{objfile} is opened.
26496 So your @file{-gdb.@var{ext}} file should be careful to avoid errors if it
26497 is evaluated more than once.
26499 @node dotdebug_gdb_scripts section
26500 @subsection The @code{.debug_gdb_scripts} section
26501 @cindex @code{.debug_gdb_scripts} section
26503 For systems using file formats like ELF and COFF,
26504 when @value{GDBN} loads a new object file
26505 it will look for a special section named @code{.debug_gdb_scripts}.
26506 If this section exists, its contents is a list of null-terminated entries
26507 specifying scripts to load. Each entry begins with a non-null prefix byte that
26508 specifies the kind of entry, typically the extension language and whether the
26509 script is in a file or inlined in @code{.debug_gdb_scripts}.
26511 The following entries are supported:
26514 @item SECTION_SCRIPT_ID_PYTHON_FILE = 1
26515 @item SECTION_SCRIPT_ID_SCHEME_FILE = 3
26516 @item SECTION_SCRIPT_ID_PYTHON_TEXT = 4
26517 @item SECTION_SCRIPT_ID_SCHEME_TEXT = 6
26520 @subsubsection Script File Entries
26522 If the entry specifies a file, @value{GDBN} will look for the file first
26523 in the current directory and then along the source search path
26524 (@pxref{Source Path, ,Specifying Source Directories}),
26525 except that @file{$cdir} is not searched, since the compilation
26526 directory is not relevant to scripts.
26528 File entries can be placed in section @code{.debug_gdb_scripts} with,
26529 for example, this GCC macro for Python scripts.
26532 /* Note: The "MS" section flags are to remove duplicates. */
26533 #define DEFINE_GDB_PY_SCRIPT(script_name) \
26535 .pushsection \".debug_gdb_scripts\", \"MS\",@@progbits,1\n\
26536 .byte 1 /* Python */\n\
26537 .asciz \"" script_name "\"\n\
26543 For Guile scripts, replace @code{.byte 1} with @code{.byte 3}.
26544 Then one can reference the macro in a header or source file like this:
26547 DEFINE_GDB_PY_SCRIPT ("my-app-scripts.py")
26550 The script name may include directories if desired.
26552 Note that loading of this script file also requires accordingly configured
26553 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
26555 If the macro invocation is put in a header, any application or library
26556 using this header will get a reference to the specified script,
26557 and with the use of @code{"MS"} attributes on the section, the linker
26558 will remove duplicates.
26560 @subsubsection Script Text Entries
26562 Script text entries allow to put the executable script in the entry
26563 itself instead of loading it from a file.
26564 The first line of the entry, everything after the prefix byte and up to
26565 the first newline (@code{0xa}) character, is the script name, and must not
26566 contain any kind of space character, e.g., spaces or tabs.
26567 The rest of the entry, up to the trailing null byte, is the script to
26568 execute in the specified language. The name needs to be unique among
26569 all script names, as @value{GDBN} executes each script only once based
26572 Here is an example from file @file{py-section-script.c} in the @value{GDBN}
26576 #include "symcat.h"
26577 #include "gdb/section-scripts.h"
26579 ".pushsection \".debug_gdb_scripts\", \"MS\",@@progbits,1\n"
26580 ".byte " XSTRING (SECTION_SCRIPT_ID_PYTHON_TEXT) "\n"
26581 ".ascii \"gdb.inlined-script\\n\"\n"
26582 ".ascii \"class test_cmd (gdb.Command):\\n\"\n"
26583 ".ascii \" def __init__ (self):\\n\"\n"
26584 ".ascii \" super (test_cmd, self).__init__ ("
26585 "\\\"test-cmd\\\", gdb.COMMAND_OBSCURE)\\n\"\n"
26586 ".ascii \" def invoke (self, arg, from_tty):\\n\"\n"
26587 ".ascii \" print (\\\"test-cmd output, arg = %s\\\" % arg)\\n\"\n"
26588 ".ascii \"test_cmd ()\\n\"\n"
26594 Loading of inlined scripts requires a properly configured
26595 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
26596 The path to specify in @code{auto-load safe-path} is the path of the file
26597 containing the @code{.debug_gdb_scripts} section.
26599 @node Which flavor to choose?
26600 @subsection Which flavor to choose?
26602 Given the multiple ways of auto-loading extensions, it might not always
26603 be clear which one to choose. This section provides some guidance.
26606 Benefits of the @file{-gdb.@var{ext}} way:
26610 Can be used with file formats that don't support multiple sections.
26613 Ease of finding scripts for public libraries.
26615 Scripts specified in the @code{.debug_gdb_scripts} section are searched for
26616 in the source search path.
26617 For publicly installed libraries, e.g., @file{libstdc++}, there typically
26618 isn't a source directory in which to find the script.
26621 Doesn't require source code additions.
26625 Benefits of the @code{.debug_gdb_scripts} way:
26629 Works with static linking.
26631 Scripts for libraries done the @file{-gdb.@var{ext}} way require an objfile to
26632 trigger their loading. When an application is statically linked the only
26633 objfile available is the executable, and it is cumbersome to attach all the
26634 scripts from all the input libraries to the executable's
26635 @file{-gdb.@var{ext}} script.
26638 Works with classes that are entirely inlined.
26640 Some classes can be entirely inlined, and thus there may not be an associated
26641 shared library to attach a @file{-gdb.@var{ext}} script to.
26644 Scripts needn't be copied out of the source tree.
26646 In some circumstances, apps can be built out of large collections of internal
26647 libraries, and the build infrastructure necessary to install the
26648 @file{-gdb.@var{ext}} scripts in a place where @value{GDBN} can find them is
26649 cumbersome. It may be easier to specify the scripts in the
26650 @code{.debug_gdb_scripts} section as relative paths, and add a path to the
26651 top of the source tree to the source search path.
26654 @node Multiple Extension Languages
26655 @section Multiple Extension Languages
26657 The Guile and Python extension languages do not share any state,
26658 and generally do not interfere with each other.
26659 There are some things to be aware of, however.
26661 @subsection Python comes first
26663 Python was @value{GDBN}'s first extension language, and to avoid breaking
26664 existing behaviour Python comes first. This is generally solved by the
26665 ``first one wins'' principle. @value{GDBN} maintains a list of enabled
26666 extension languages, and when it makes a call to an extension language,
26667 (say to pretty-print a value), it tries each in turn until an extension
26668 language indicates it has performed the request (e.g., has returned the
26669 pretty-printed form of a value).
26670 This extends to errors while performing such requests: If an error happens
26671 while, for example, trying to pretty-print an object then the error is
26672 reported and any following extension languages are not tried.
26675 @section Creating new spellings of existing commands
26676 @cindex aliases for commands
26678 It is often useful to define alternate spellings of existing commands.
26679 For example, if a new @value{GDBN} command defined in Python has
26680 a long name to type, it is handy to have an abbreviated version of it
26681 that involves less typing.
26683 @value{GDBN} itself uses aliases. For example @samp{s} is an alias
26684 of the @samp{step} command even though it is otherwise an ambiguous
26685 abbreviation of other commands like @samp{set} and @samp{show}.
26687 Aliases are also used to provide shortened or more common versions
26688 of multi-word commands. For example, @value{GDBN} provides the
26689 @samp{tty} alias of the @samp{set inferior-tty} command.
26691 You can define a new alias with the @samp{alias} command.
26696 @item alias [-a] [--] @var{ALIAS} = @var{COMMAND}
26700 @var{ALIAS} specifies the name of the new alias.
26701 Each word of @var{ALIAS} must consist of letters, numbers, dashes and
26704 @var{COMMAND} specifies the name of an existing command
26705 that is being aliased.
26707 The @samp{-a} option specifies that the new alias is an abbreviation
26708 of the command. Abbreviations are not shown in command
26709 lists displayed by the @samp{help} command.
26711 The @samp{--} option specifies the end of options,
26712 and is useful when @var{ALIAS} begins with a dash.
26714 Here is a simple example showing how to make an abbreviation
26715 of a command so that there is less to type.
26716 Suppose you were tired of typing @samp{disas}, the current
26717 shortest unambiguous abbreviation of the @samp{disassemble} command
26718 and you wanted an even shorter version named @samp{di}.
26719 The following will accomplish this.
26722 (gdb) alias -a di = disas
26725 Note that aliases are different from user-defined commands.
26726 With a user-defined command, you also need to write documentation
26727 for it with the @samp{document} command.
26728 An alias automatically picks up the documentation of the existing command.
26730 Here is an example where we make @samp{elms} an abbreviation of
26731 @samp{elements} in the @samp{set print elements} command.
26732 This is to show that you can make an abbreviation of any part
26736 (gdb) alias -a set print elms = set print elements
26737 (gdb) alias -a show print elms = show print elements
26738 (gdb) set p elms 20
26740 Limit on string chars or array elements to print is 200.
26743 Note that if you are defining an alias of a @samp{set} command,
26744 and you want to have an alias for the corresponding @samp{show}
26745 command, then you need to define the latter separately.
26747 Unambiguously abbreviated commands are allowed in @var{COMMAND} and
26748 @var{ALIAS}, just as they are normally.
26751 (gdb) alias -a set pr elms = set p ele
26754 Finally, here is an example showing the creation of a one word
26755 alias for a more complex command.
26756 This creates alias @samp{spe} of the command @samp{set print elements}.
26759 (gdb) alias spe = set print elements
26764 @chapter Command Interpreters
26765 @cindex command interpreters
26767 @value{GDBN} supports multiple command interpreters, and some command
26768 infrastructure to allow users or user interface writers to switch
26769 between interpreters or run commands in other interpreters.
26771 @value{GDBN} currently supports two command interpreters, the console
26772 interpreter (sometimes called the command-line interpreter or @sc{cli})
26773 and the machine interface interpreter (or @sc{gdb/mi}). This manual
26774 describes both of these interfaces in great detail.
26776 By default, @value{GDBN} will start with the console interpreter.
26777 However, the user may choose to start @value{GDBN} with another
26778 interpreter by specifying the @option{-i} or @option{--interpreter}
26779 startup options. Defined interpreters include:
26783 @cindex console interpreter
26784 The traditional console or command-line interpreter. This is the most often
26785 used interpreter with @value{GDBN}. With no interpreter specified at runtime,
26786 @value{GDBN} will use this interpreter.
26789 @cindex mi interpreter
26790 The newest @sc{gdb/mi} interface (currently @code{mi3}). Used primarily
26791 by programs wishing to use @value{GDBN} as a backend for a debugger GUI
26792 or an IDE. For more information, see @ref{GDB/MI, ,The @sc{gdb/mi}
26796 @cindex mi3 interpreter
26797 The @sc{gdb/mi} interface introduced in @value{GDBN} 9.1.
26800 @cindex mi2 interpreter
26801 The @sc{gdb/mi} interface introduced in @value{GDBN} 6.0.
26804 @cindex mi1 interpreter
26805 The @sc{gdb/mi} interface introduced in @value{GDBN} 5.1.
26809 @cindex invoke another interpreter
26811 @kindex interpreter-exec
26812 You may execute commands in any interpreter from the current
26813 interpreter using the appropriate command. If you are running the
26814 console interpreter, simply use the @code{interpreter-exec} command:
26817 interpreter-exec mi "-data-list-register-names"
26820 @sc{gdb/mi} has a similar command, although it is only available in versions of
26821 @value{GDBN} which support @sc{gdb/mi} version 2 (or greater).
26823 Note that @code{interpreter-exec} only changes the interpreter for the
26824 duration of the specified command. It does not change the interpreter
26827 @cindex start a new independent interpreter
26829 Although you may only choose a single interpreter at startup, it is
26830 possible to run an independent interpreter on a specified input/output
26831 device (usually a tty).
26833 For example, consider a debugger GUI or IDE that wants to provide a
26834 @value{GDBN} console view. It may do so by embedding a terminal
26835 emulator widget in its GUI, starting @value{GDBN} in the traditional
26836 command-line mode with stdin/stdout/stderr redirected to that
26837 terminal, and then creating an MI interpreter running on a specified
26838 input/output device. The console interpreter created by @value{GDBN}
26839 at startup handles commands the user types in the terminal widget,
26840 while the GUI controls and synchronizes state with @value{GDBN} using
26841 the separate MI interpreter.
26843 To start a new secondary @dfn{user interface} running MI, use the
26844 @code{new-ui} command:
26847 @cindex new user interface
26849 new-ui @var{interpreter} @var{tty}
26852 The @var{interpreter} parameter specifies the interpreter to run.
26853 This accepts the same values as the @code{interpreter-exec} command.
26854 For example, @samp{console}, @samp{mi}, @samp{mi2}, etc. The
26855 @var{tty} parameter specifies the name of the bidirectional file the
26856 interpreter uses for input/output, usually the name of a
26857 pseudoterminal slave on Unix systems. For example:
26860 (@value{GDBP}) new-ui mi /dev/pts/9
26864 runs an MI interpreter on @file{/dev/pts/9}.
26867 @chapter @value{GDBN} Text User Interface
26869 @cindex Text User Interface
26872 * TUI Overview:: TUI overview
26873 * TUI Keys:: TUI key bindings
26874 * TUI Single Key Mode:: TUI single key mode
26875 * TUI Commands:: TUI-specific commands
26876 * TUI Configuration:: TUI configuration variables
26879 The @value{GDBN} Text User Interface (TUI) is a terminal
26880 interface which uses the @code{curses} library to show the source
26881 file, the assembly output, the program registers and @value{GDBN}
26882 commands in separate text windows. The TUI mode is supported only
26883 on platforms where a suitable version of the @code{curses} library
26886 The TUI mode is enabled by default when you invoke @value{GDBN} as
26887 @samp{@value{GDBP} -tui}.
26888 You can also switch in and out of TUI mode while @value{GDBN} runs by
26889 using various TUI commands and key bindings, such as @command{tui
26890 enable} or @kbd{C-x C-a}. @xref{TUI Commands, ,TUI Commands}, and
26891 @ref{TUI Keys, ,TUI Key Bindings}.
26894 @section TUI Overview
26896 In TUI mode, @value{GDBN} can display several text windows:
26900 This window is the @value{GDBN} command window with the @value{GDBN}
26901 prompt and the @value{GDBN} output. The @value{GDBN} input is still
26902 managed using readline.
26905 The source window shows the source file of the program. The current
26906 line and active breakpoints are displayed in this window.
26909 The assembly window shows the disassembly output of the program.
26912 This window shows the processor registers. Registers are highlighted
26913 when their values change.
26916 The source and assembly windows show the current program position
26917 by highlighting the current line and marking it with a @samp{>} marker.
26918 Breakpoints are indicated with two markers. The first marker
26919 indicates the breakpoint type:
26923 Breakpoint which was hit at least once.
26926 Breakpoint which was never hit.
26929 Hardware breakpoint which was hit at least once.
26932 Hardware breakpoint which was never hit.
26935 The second marker indicates whether the breakpoint is enabled or not:
26939 Breakpoint is enabled.
26942 Breakpoint is disabled.
26945 The source, assembly and register windows are updated when the current
26946 thread changes, when the frame changes, or when the program counter
26949 These windows are not all visible at the same time. The command
26950 window is always visible. The others can be arranged in several
26961 source and assembly,
26964 source and registers, or
26967 assembly and registers.
26970 A status line above the command window shows the following information:
26974 Indicates the current @value{GDBN} target.
26975 (@pxref{Targets, ,Specifying a Debugging Target}).
26978 Gives the current process or thread number.
26979 When no process is being debugged, this field is set to @code{No process}.
26982 Gives the current function name for the selected frame.
26983 The name is demangled if demangling is turned on (@pxref{Print Settings}).
26984 When there is no symbol corresponding to the current program counter,
26985 the string @code{??} is displayed.
26988 Indicates the current line number for the selected frame.
26989 When the current line number is not known, the string @code{??} is displayed.
26992 Indicates the current program counter address.
26996 @section TUI Key Bindings
26997 @cindex TUI key bindings
26999 The TUI installs several key bindings in the readline keymaps
27000 @ifset SYSTEM_READLINE
27001 (@pxref{Command Line Editing, , , rluserman, GNU Readline Library}).
27003 @ifclear SYSTEM_READLINE
27004 (@pxref{Command Line Editing}).
27006 The following key bindings are installed for both TUI mode and the
27007 @value{GDBN} standard mode.
27016 Enter or leave the TUI mode. When leaving the TUI mode,
27017 the curses window management stops and @value{GDBN} operates using
27018 its standard mode, writing on the terminal directly. When reentering
27019 the TUI mode, control is given back to the curses windows.
27020 The screen is then refreshed.
27024 Use a TUI layout with only one window. The layout will
27025 either be @samp{source} or @samp{assembly}. When the TUI mode
27026 is not active, it will switch to the TUI mode.
27028 Think of this key binding as the Emacs @kbd{C-x 1} binding.
27032 Use a TUI layout with at least two windows. When the current
27033 layout already has two windows, the next layout with two windows is used.
27034 When a new layout is chosen, one window will always be common to the
27035 previous layout and the new one.
27037 Think of it as the Emacs @kbd{C-x 2} binding.
27041 Change the active window. The TUI associates several key bindings
27042 (like scrolling and arrow keys) with the active window. This command
27043 gives the focus to the next TUI window.
27045 Think of it as the Emacs @kbd{C-x o} binding.
27049 Switch in and out of the TUI SingleKey mode that binds single
27050 keys to @value{GDBN} commands (@pxref{TUI Single Key Mode}).
27053 The following key bindings only work in the TUI mode:
27058 Scroll the active window one page up.
27062 Scroll the active window one page down.
27066 Scroll the active window one line up.
27070 Scroll the active window one line down.
27074 Scroll the active window one column left.
27078 Scroll the active window one column right.
27082 Refresh the screen.
27085 Because the arrow keys scroll the active window in the TUI mode, they
27086 are not available for their normal use by readline unless the command
27087 window has the focus. When another window is active, you must use
27088 other readline key bindings such as @kbd{C-p}, @kbd{C-n}, @kbd{C-b}
27089 and @kbd{C-f} to control the command window.
27091 @node TUI Single Key Mode
27092 @section TUI Single Key Mode
27093 @cindex TUI single key mode
27095 The TUI also provides a @dfn{SingleKey} mode, which binds several
27096 frequently used @value{GDBN} commands to single keys. Type @kbd{C-x s} to
27097 switch into this mode, where the following key bindings are used:
27100 @kindex c @r{(SingleKey TUI key)}
27104 @kindex d @r{(SingleKey TUI key)}
27108 @kindex f @r{(SingleKey TUI key)}
27112 @kindex n @r{(SingleKey TUI key)}
27116 @kindex o @r{(SingleKey TUI key)}
27118 nexti. The shortcut letter @samp{o} stands for ``step Over''.
27120 @kindex q @r{(SingleKey TUI key)}
27122 exit the SingleKey mode.
27124 @kindex r @r{(SingleKey TUI key)}
27128 @kindex s @r{(SingleKey TUI key)}
27132 @kindex i @r{(SingleKey TUI key)}
27134 stepi. The shortcut letter @samp{i} stands for ``step Into''.
27136 @kindex u @r{(SingleKey TUI key)}
27140 @kindex v @r{(SingleKey TUI key)}
27144 @kindex w @r{(SingleKey TUI key)}
27149 Other keys temporarily switch to the @value{GDBN} command prompt.
27150 The key that was pressed is inserted in the editing buffer so that
27151 it is possible to type most @value{GDBN} commands without interaction
27152 with the TUI SingleKey mode. Once the command is entered the TUI
27153 SingleKey mode is restored. The only way to permanently leave
27154 this mode is by typing @kbd{q} or @kbd{C-x s}.
27158 @section TUI-specific Commands
27159 @cindex TUI commands
27161 The TUI has specific commands to control the text windows.
27162 These commands are always available, even when @value{GDBN} is not in
27163 the TUI mode. When @value{GDBN} is in the standard mode, most
27164 of these commands will automatically switch to the TUI mode.
27166 Note that if @value{GDBN}'s @code{stdout} is not connected to a
27167 terminal, or @value{GDBN} has been started with the machine interface
27168 interpreter (@pxref{GDB/MI, ,The @sc{gdb/mi} Interface}), most of
27169 these commands will fail with an error, because it would not be
27170 possible or desirable to enable curses window management.
27175 Activate TUI mode. The last active TUI window layout will be used if
27176 TUI mode has prevsiouly been used in the current debugging session,
27177 otherwise a default layout is used.
27180 @kindex tui disable
27181 Disable TUI mode, returning to the console interpreter.
27185 List and give the size of all displayed windows.
27187 @item layout @var{name}
27189 Changes which TUI windows are displayed. In each layout the command
27190 window is always displayed, the @var{name} parameter controls which
27191 additional windows are displayed, and can be any of the following:
27195 Display the next layout.
27198 Display the previous layout.
27201 Display the source and command windows.
27204 Display the assembly and command windows.
27207 Display the source, assembly, and command windows.
27210 When in @code{src} layout display the register, source, and command
27211 windows. When in @code{asm} or @code{split} layout display the
27212 register, assembler, and command windows.
27215 @item focus @var{name}
27217 Changes which TUI window is currently active for scrolling. The
27218 @var{name} parameter can be any of the following:
27222 Make the next window active for scrolling.
27225 Make the previous window active for scrolling.
27228 Make the source window active for scrolling.
27231 Make the assembly window active for scrolling.
27234 Make the register window active for scrolling.
27237 Make the command window active for scrolling.
27242 Refresh the screen. This is similar to typing @kbd{C-L}.
27244 @item tui reg @var{group}
27246 Changes the register group displayed in the tui register window to
27247 @var{group}. If the register window is not currently displayed this
27248 command will cause the register window to be displayed. The list of
27249 register groups, as well as their order is target specific. The
27250 following groups are available on most targets:
27253 Repeatedly selecting this group will cause the display to cycle
27254 through all of the available register groups.
27257 Repeatedly selecting this group will cause the display to cycle
27258 through all of the available register groups in the reverse order to
27262 Display the general registers.
27264 Display the floating point registers.
27266 Display the system registers.
27268 Display the vector registers.
27270 Display all registers.
27275 Update the source window and the current execution point.
27277 @item winheight @var{name} +@var{count}
27278 @itemx winheight @var{name} -@var{count}
27280 Change the height of the window @var{name} by @var{count}
27281 lines. Positive counts increase the height, while negative counts
27282 decrease it. The @var{name} parameter can be one of @code{src} (the
27283 source window), @code{cmd} (the command window), @code{asm} (the
27284 disassembly window), or @code{regs} (the register display window).
27287 @node TUI Configuration
27288 @section TUI Configuration Variables
27289 @cindex TUI configuration variables
27291 Several configuration variables control the appearance of TUI windows.
27294 @item set tui border-kind @var{kind}
27295 @kindex set tui border-kind
27296 Select the border appearance for the source, assembly and register windows.
27297 The possible values are the following:
27300 Use a space character to draw the border.
27303 Use @sc{ascii} characters @samp{+}, @samp{-} and @samp{|} to draw the border.
27306 Use the Alternate Character Set to draw the border. The border is
27307 drawn using character line graphics if the terminal supports them.
27310 @item set tui border-mode @var{mode}
27311 @kindex set tui border-mode
27312 @itemx set tui active-border-mode @var{mode}
27313 @kindex set tui active-border-mode
27314 Select the display attributes for the borders of the inactive windows
27315 or the active window. The @var{mode} can be one of the following:
27318 Use normal attributes to display the border.
27324 Use reverse video mode.
27327 Use half bright mode.
27329 @item half-standout
27330 Use half bright and standout mode.
27333 Use extra bright or bold mode.
27335 @item bold-standout
27336 Use extra bright or bold and standout mode.
27339 @item set tui tab-width @var{nchars}
27340 @kindex set tui tab-width
27342 Set the width of tab stops to be @var{nchars} characters. This
27343 setting affects the display of TAB characters in the source and
27348 @chapter Using @value{GDBN} under @sc{gnu} Emacs
27351 @cindex @sc{gnu} Emacs
27352 A special interface allows you to use @sc{gnu} Emacs to view (and
27353 edit) the source files for the program you are debugging with
27356 To use this interface, use the command @kbd{M-x gdb} in Emacs. Give the
27357 executable file you want to debug as an argument. This command starts
27358 @value{GDBN} as a subprocess of Emacs, with input and output through a newly
27359 created Emacs buffer.
27360 @c (Do not use the @code{-tui} option to run @value{GDBN} from Emacs.)
27362 Running @value{GDBN} under Emacs can be just like running @value{GDBN} normally except for two
27367 All ``terminal'' input and output goes through an Emacs buffer, called
27370 This applies both to @value{GDBN} commands and their output, and to the input
27371 and output done by the program you are debugging.
27373 This is useful because it means that you can copy the text of previous
27374 commands and input them again; you can even use parts of the output
27377 All the facilities of Emacs' Shell mode are available for interacting
27378 with your program. In particular, you can send signals the usual
27379 way---for example, @kbd{C-c C-c} for an interrupt, @kbd{C-c C-z} for a
27383 @value{GDBN} displays source code through Emacs.
27385 Each time @value{GDBN} displays a stack frame, Emacs automatically finds the
27386 source file for that frame and puts an arrow (@samp{=>}) at the
27387 left margin of the current line. Emacs uses a separate buffer for
27388 source display, and splits the screen to show both your @value{GDBN} session
27391 Explicit @value{GDBN} @code{list} or search commands still produce output as
27392 usual, but you probably have no reason to use them from Emacs.
27395 We call this @dfn{text command mode}. Emacs 22.1, and later, also uses
27396 a graphical mode, enabled by default, which provides further buffers
27397 that can control the execution and describe the state of your program.
27398 @xref{GDB Graphical Interface,,, Emacs, The @sc{gnu} Emacs Manual}.
27400 If you specify an absolute file name when prompted for the @kbd{M-x
27401 gdb} argument, then Emacs sets your current working directory to where
27402 your program resides. If you only specify the file name, then Emacs
27403 sets your current working directory to the directory associated
27404 with the previous buffer. In this case, @value{GDBN} may find your
27405 program by searching your environment's @code{PATH} variable, but on
27406 some operating systems it might not find the source. So, although the
27407 @value{GDBN} input and output session proceeds normally, the auxiliary
27408 buffer does not display the current source and line of execution.
27410 The initial working directory of @value{GDBN} is printed on the top
27411 line of the GUD buffer and this serves as a default for the commands
27412 that specify files for @value{GDBN} to operate on. @xref{Files,
27413 ,Commands to Specify Files}.
27415 By default, @kbd{M-x gdb} calls the program called @file{gdb}. If you
27416 need to call @value{GDBN} by a different name (for example, if you
27417 keep several configurations around, with different names) you can
27418 customize the Emacs variable @code{gud-gdb-command-name} to run the
27421 In the GUD buffer, you can use these special Emacs commands in
27422 addition to the standard Shell mode commands:
27426 Describe the features of Emacs' GUD Mode.
27429 Execute to another source line, like the @value{GDBN} @code{step} command; also
27430 update the display window to show the current file and location.
27433 Execute to next source line in this function, skipping all function
27434 calls, like the @value{GDBN} @code{next} command. Then update the display window
27435 to show the current file and location.
27438 Execute one instruction, like the @value{GDBN} @code{stepi} command; update
27439 display window accordingly.
27442 Execute until exit from the selected stack frame, like the @value{GDBN}
27443 @code{finish} command.
27446 Continue execution of your program, like the @value{GDBN} @code{continue}
27450 Go up the number of frames indicated by the numeric argument
27451 (@pxref{Arguments, , Numeric Arguments, Emacs, The @sc{gnu} Emacs Manual}),
27452 like the @value{GDBN} @code{up} command.
27455 Go down the number of frames indicated by the numeric argument, like the
27456 @value{GDBN} @code{down} command.
27459 In any source file, the Emacs command @kbd{C-x @key{SPC}} (@code{gud-break})
27460 tells @value{GDBN} to set a breakpoint on the source line point is on.
27462 In text command mode, if you type @kbd{M-x speedbar}, Emacs displays a
27463 separate frame which shows a backtrace when the GUD buffer is current.
27464 Move point to any frame in the stack and type @key{RET} to make it
27465 become the current frame and display the associated source in the
27466 source buffer. Alternatively, click @kbd{Mouse-2} to make the
27467 selected frame become the current one. In graphical mode, the
27468 speedbar displays watch expressions.
27470 If you accidentally delete the source-display buffer, an easy way to get
27471 it back is to type the command @code{f} in the @value{GDBN} buffer, to
27472 request a frame display; when you run under Emacs, this recreates
27473 the source buffer if necessary to show you the context of the current
27476 The source files displayed in Emacs are in ordinary Emacs buffers
27477 which are visiting the source files in the usual way. You can edit
27478 the files with these buffers if you wish; but keep in mind that @value{GDBN}
27479 communicates with Emacs in terms of line numbers. If you add or
27480 delete lines from the text, the line numbers that @value{GDBN} knows cease
27481 to correspond properly with the code.
27483 A more detailed description of Emacs' interaction with @value{GDBN} is
27484 given in the Emacs manual (@pxref{Debuggers,,, Emacs, The @sc{gnu}
27488 @chapter The @sc{gdb/mi} Interface
27490 @unnumberedsec Function and Purpose
27492 @cindex @sc{gdb/mi}, its purpose
27493 @sc{gdb/mi} is a line based machine oriented text interface to
27494 @value{GDBN} and is activated by specifying using the
27495 @option{--interpreter} command line option (@pxref{Mode Options}). It
27496 is specifically intended to support the development of systems which
27497 use the debugger as just one small component of a larger system.
27499 This chapter is a specification of the @sc{gdb/mi} interface. It is written
27500 in the form of a reference manual.
27502 Note that @sc{gdb/mi} is still under construction, so some of the
27503 features described below are incomplete and subject to change
27504 (@pxref{GDB/MI Development and Front Ends, , @sc{gdb/mi} Development and Front Ends}).
27506 @unnumberedsec Notation and Terminology
27508 @cindex notational conventions, for @sc{gdb/mi}
27509 This chapter uses the following notation:
27513 @code{|} separates two alternatives.
27516 @code{[ @var{something} ]} indicates that @var{something} is optional:
27517 it may or may not be given.
27520 @code{( @var{group} )*} means that @var{group} inside the parentheses
27521 may repeat zero or more times.
27524 @code{( @var{group} )+} means that @var{group} inside the parentheses
27525 may repeat one or more times.
27528 @code{"@var{string}"} means a literal @var{string}.
27532 @heading Dependencies
27536 * GDB/MI General Design::
27537 * GDB/MI Command Syntax::
27538 * GDB/MI Compatibility with CLI::
27539 * GDB/MI Development and Front Ends::
27540 * GDB/MI Output Records::
27541 * GDB/MI Simple Examples::
27542 * GDB/MI Command Description Format::
27543 * GDB/MI Breakpoint Commands::
27544 * GDB/MI Catchpoint Commands::
27545 * GDB/MI Program Context::
27546 * GDB/MI Thread Commands::
27547 * GDB/MI Ada Tasking Commands::
27548 * GDB/MI Program Execution::
27549 * GDB/MI Stack Manipulation::
27550 * GDB/MI Variable Objects::
27551 * GDB/MI Data Manipulation::
27552 * GDB/MI Tracepoint Commands::
27553 * GDB/MI Symbol Query::
27554 * GDB/MI File Commands::
27556 * GDB/MI Kod Commands::
27557 * GDB/MI Memory Overlay Commands::
27558 * GDB/MI Signal Handling Commands::
27560 * GDB/MI Target Manipulation::
27561 * GDB/MI File Transfer Commands::
27562 * GDB/MI Ada Exceptions Commands::
27563 * GDB/MI Support Commands::
27564 * GDB/MI Miscellaneous Commands::
27567 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
27568 @node GDB/MI General Design
27569 @section @sc{gdb/mi} General Design
27570 @cindex GDB/MI General Design
27572 Interaction of a @sc{GDB/MI} frontend with @value{GDBN} involves three
27573 parts---commands sent to @value{GDBN}, responses to those commands
27574 and notifications. Each command results in exactly one response,
27575 indicating either successful completion of the command, or an error.
27576 For the commands that do not resume the target, the response contains the
27577 requested information. For the commands that resume the target, the
27578 response only indicates whether the target was successfully resumed.
27579 Notifications is the mechanism for reporting changes in the state of the
27580 target, or in @value{GDBN} state, that cannot conveniently be associated with
27581 a command and reported as part of that command response.
27583 The important examples of notifications are:
27587 Exec notifications. These are used to report changes in
27588 target state---when a target is resumed, or stopped. It would not
27589 be feasible to include this information in response of resuming
27590 commands, because one resume commands can result in multiple events in
27591 different threads. Also, quite some time may pass before any event
27592 happens in the target, while a frontend needs to know whether the resuming
27593 command itself was successfully executed.
27596 Console output, and status notifications. Console output
27597 notifications are used to report output of CLI commands, as well as
27598 diagnostics for other commands. Status notifications are used to
27599 report the progress of a long-running operation. Naturally, including
27600 this information in command response would mean no output is produced
27601 until the command is finished, which is undesirable.
27604 General notifications. Commands may have various side effects on
27605 the @value{GDBN} or target state beyond their official purpose. For example,
27606 a command may change the selected thread. Although such changes can
27607 be included in command response, using notification allows for more
27608 orthogonal frontend design.
27612 There's no guarantee that whenever an MI command reports an error,
27613 @value{GDBN} or the target are in any specific state, and especially,
27614 the state is not reverted to the state before the MI command was
27615 processed. Therefore, whenever an MI command results in an error,
27616 we recommend that the frontend refreshes all the information shown in
27617 the user interface.
27621 * Context management::
27622 * Asynchronous and non-stop modes::
27626 @node Context management
27627 @subsection Context management
27629 @subsubsection Threads and Frames
27631 In most cases when @value{GDBN} accesses the target, this access is
27632 done in context of a specific thread and frame (@pxref{Frames}).
27633 Often, even when accessing global data, the target requires that a thread
27634 be specified. The CLI interface maintains the selected thread and frame,
27635 and supplies them to target on each command. This is convenient,
27636 because a command line user would not want to specify that information
27637 explicitly on each command, and because user interacts with
27638 @value{GDBN} via a single terminal, so no confusion is possible as
27639 to what thread and frame are the current ones.
27641 In the case of MI, the concept of selected thread and frame is less
27642 useful. First, a frontend can easily remember this information
27643 itself. Second, a graphical frontend can have more than one window,
27644 each one used for debugging a different thread, and the frontend might
27645 want to access additional threads for internal purposes. This
27646 increases the risk that by relying on implicitly selected thread, the
27647 frontend may be operating on a wrong one. Therefore, each MI command
27648 should explicitly specify which thread and frame to operate on. To
27649 make it possible, each MI command accepts the @samp{--thread} and
27650 @samp{--frame} options, the value to each is @value{GDBN} global
27651 identifier for thread and frame to operate on.
27653 Usually, each top-level window in a frontend allows the user to select
27654 a thread and a frame, and remembers the user selection for further
27655 operations. However, in some cases @value{GDBN} may suggest that the
27656 current thread or frame be changed. For example, when stopping on a
27657 breakpoint it is reasonable to switch to the thread where breakpoint is
27658 hit. For another example, if the user issues the CLI @samp{thread} or
27659 @samp{frame} commands via the frontend, it is desirable to change the
27660 frontend's selection to the one specified by user. @value{GDBN}
27661 communicates the suggestion to change current thread and frame using the
27662 @samp{=thread-selected} notification.
27664 Note that historically, MI shares the selected thread with CLI, so
27665 frontends used the @code{-thread-select} to execute commands in the
27666 right context. However, getting this to work right is cumbersome. The
27667 simplest way is for frontend to emit @code{-thread-select} command
27668 before every command. This doubles the number of commands that need
27669 to be sent. The alternative approach is to suppress @code{-thread-select}
27670 if the selected thread in @value{GDBN} is supposed to be identical to the
27671 thread the frontend wants to operate on. However, getting this
27672 optimization right can be tricky. In particular, if the frontend
27673 sends several commands to @value{GDBN}, and one of the commands changes the
27674 selected thread, then the behaviour of subsequent commands will
27675 change. So, a frontend should either wait for response from such
27676 problematic commands, or explicitly add @code{-thread-select} for
27677 all subsequent commands. No frontend is known to do this exactly
27678 right, so it is suggested to just always pass the @samp{--thread} and
27679 @samp{--frame} options.
27681 @subsubsection Language
27683 The execution of several commands depends on which language is selected.
27684 By default, the current language (@pxref{show language}) is used.
27685 But for commands known to be language-sensitive, it is recommended
27686 to use the @samp{--language} option. This option takes one argument,
27687 which is the name of the language to use while executing the command.
27691 -data-evaluate-expression --language c "sizeof (void*)"
27696 The valid language names are the same names accepted by the
27697 @samp{set language} command (@pxref{Manually}), excluding @samp{auto},
27698 @samp{local} or @samp{unknown}.
27700 @node Asynchronous and non-stop modes
27701 @subsection Asynchronous command execution and non-stop mode
27703 On some targets, @value{GDBN} is capable of processing MI commands
27704 even while the target is running. This is called @dfn{asynchronous
27705 command execution} (@pxref{Background Execution}). The frontend may
27706 specify a preferrence for asynchronous execution using the
27707 @code{-gdb-set mi-async 1} command, which should be emitted before
27708 either running the executable or attaching to the target. After the
27709 frontend has started the executable or attached to the target, it can
27710 find if asynchronous execution is enabled using the
27711 @code{-list-target-features} command.
27714 @item -gdb-set mi-async on
27715 @item -gdb-set mi-async off
27716 Set whether MI is in asynchronous mode.
27718 When @code{off}, which is the default, MI execution commands (e.g.,
27719 @code{-exec-continue}) are foreground commands, and @value{GDBN} waits
27720 for the program to stop before processing further commands.
27722 When @code{on}, MI execution commands are background execution
27723 commands (e.g., @code{-exec-continue} becomes the equivalent of the
27724 @code{c&} CLI command), and so @value{GDBN} is capable of processing
27725 MI commands even while the target is running.
27727 @item -gdb-show mi-async
27728 Show whether MI asynchronous mode is enabled.
27731 Note: In @value{GDBN} version 7.7 and earlier, this option was called
27732 @code{target-async} instead of @code{mi-async}, and it had the effect
27733 of both putting MI in asynchronous mode and making CLI background
27734 commands possible. CLI background commands are now always possible
27735 ``out of the box'' if the target supports them. The old spelling is
27736 kept as a deprecated alias for backwards compatibility.
27738 Even if @value{GDBN} can accept a command while target is running,
27739 many commands that access the target do not work when the target is
27740 running. Therefore, asynchronous command execution is most useful
27741 when combined with non-stop mode (@pxref{Non-Stop Mode}). Then,
27742 it is possible to examine the state of one thread, while other threads
27745 When a given thread is running, MI commands that try to access the
27746 target in the context of that thread may not work, or may work only on
27747 some targets. In particular, commands that try to operate on thread's
27748 stack will not work, on any target. Commands that read memory, or
27749 modify breakpoints, may work or not work, depending on the target. Note
27750 that even commands that operate on global state, such as @code{print},
27751 @code{set}, and breakpoint commands, still access the target in the
27752 context of a specific thread, so frontend should try to find a
27753 stopped thread and perform the operation on that thread (using the
27754 @samp{--thread} option).
27756 Which commands will work in the context of a running thread is
27757 highly target dependent. However, the two commands
27758 @code{-exec-interrupt}, to stop a thread, and @code{-thread-info},
27759 to find the state of a thread, will always work.
27761 @node Thread groups
27762 @subsection Thread groups
27763 @value{GDBN} may be used to debug several processes at the same time.
27764 On some platfroms, @value{GDBN} may support debugging of several
27765 hardware systems, each one having several cores with several different
27766 processes running on each core. This section describes the MI
27767 mechanism to support such debugging scenarios.
27769 The key observation is that regardless of the structure of the
27770 target, MI can have a global list of threads, because most commands that
27771 accept the @samp{--thread} option do not need to know what process that
27772 thread belongs to. Therefore, it is not necessary to introduce
27773 neither additional @samp{--process} option, nor an notion of the
27774 current process in the MI interface. The only strictly new feature
27775 that is required is the ability to find how the threads are grouped
27778 To allow the user to discover such grouping, and to support arbitrary
27779 hierarchy of machines/cores/processes, MI introduces the concept of a
27780 @dfn{thread group}. Thread group is a collection of threads and other
27781 thread groups. A thread group always has a string identifier, a type,
27782 and may have additional attributes specific to the type. A new
27783 command, @code{-list-thread-groups}, returns the list of top-level
27784 thread groups, which correspond to processes that @value{GDBN} is
27785 debugging at the moment. By passing an identifier of a thread group
27786 to the @code{-list-thread-groups} command, it is possible to obtain
27787 the members of specific thread group.
27789 To allow the user to easily discover processes, and other objects, he
27790 wishes to debug, a concept of @dfn{available thread group} is
27791 introduced. Available thread group is an thread group that
27792 @value{GDBN} is not debugging, but that can be attached to, using the
27793 @code{-target-attach} command. The list of available top-level thread
27794 groups can be obtained using @samp{-list-thread-groups --available}.
27795 In general, the content of a thread group may be only retrieved only
27796 after attaching to that thread group.
27798 Thread groups are related to inferiors (@pxref{Inferiors and
27799 Programs}). Each inferior corresponds to a thread group of a special
27800 type @samp{process}, and some additional operations are permitted on
27801 such thread groups.
27803 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
27804 @node GDB/MI Command Syntax
27805 @section @sc{gdb/mi} Command Syntax
27808 * GDB/MI Input Syntax::
27809 * GDB/MI Output Syntax::
27812 @node GDB/MI Input Syntax
27813 @subsection @sc{gdb/mi} Input Syntax
27815 @cindex input syntax for @sc{gdb/mi}
27816 @cindex @sc{gdb/mi}, input syntax
27818 @item @var{command} @expansion{}
27819 @code{@var{cli-command} | @var{mi-command}}
27821 @item @var{cli-command} @expansion{}
27822 @code{[ @var{token} ] @var{cli-command} @var{nl}}, where
27823 @var{cli-command} is any existing @value{GDBN} CLI command.
27825 @item @var{mi-command} @expansion{}
27826 @code{[ @var{token} ] "-" @var{operation} ( " " @var{option} )*
27827 @code{[} " --" @code{]} ( " " @var{parameter} )* @var{nl}}
27829 @item @var{token} @expansion{}
27830 "any sequence of digits"
27832 @item @var{option} @expansion{}
27833 @code{"-" @var{parameter} [ " " @var{parameter} ]}
27835 @item @var{parameter} @expansion{}
27836 @code{@var{non-blank-sequence} | @var{c-string}}
27838 @item @var{operation} @expansion{}
27839 @emph{any of the operations described in this chapter}
27841 @item @var{non-blank-sequence} @expansion{}
27842 @emph{anything, provided it doesn't contain special characters such as
27843 "-", @var{nl}, """ and of course " "}
27845 @item @var{c-string} @expansion{}
27846 @code{""" @var{seven-bit-iso-c-string-content} """}
27848 @item @var{nl} @expansion{}
27857 The CLI commands are still handled by the @sc{mi} interpreter; their
27858 output is described below.
27861 The @code{@var{token}}, when present, is passed back when the command
27865 Some @sc{mi} commands accept optional arguments as part of the parameter
27866 list. Each option is identified by a leading @samp{-} (dash) and may be
27867 followed by an optional argument parameter. Options occur first in the
27868 parameter list and can be delimited from normal parameters using
27869 @samp{--} (this is useful when some parameters begin with a dash).
27876 We want easy access to the existing CLI syntax (for debugging).
27879 We want it to be easy to spot a @sc{mi} operation.
27882 @node GDB/MI Output Syntax
27883 @subsection @sc{gdb/mi} Output Syntax
27885 @cindex output syntax of @sc{gdb/mi}
27886 @cindex @sc{gdb/mi}, output syntax
27887 The output from @sc{gdb/mi} consists of zero or more out-of-band records
27888 followed, optionally, by a single result record. This result record
27889 is for the most recent command. The sequence of output records is
27890 terminated by @samp{(gdb)}.
27892 If an input command was prefixed with a @code{@var{token}} then the
27893 corresponding output for that command will also be prefixed by that same
27897 @item @var{output} @expansion{}
27898 @code{( @var{out-of-band-record} )* [ @var{result-record} ] "(gdb)" @var{nl}}
27900 @item @var{result-record} @expansion{}
27901 @code{ [ @var{token} ] "^" @var{result-class} ( "," @var{result} )* @var{nl}}
27903 @item @var{out-of-band-record} @expansion{}
27904 @code{@var{async-record} | @var{stream-record}}
27906 @item @var{async-record} @expansion{}
27907 @code{@var{exec-async-output} | @var{status-async-output} | @var{notify-async-output}}
27909 @item @var{exec-async-output} @expansion{}
27910 @code{[ @var{token} ] "*" @var{async-output nl}}
27912 @item @var{status-async-output} @expansion{}
27913 @code{[ @var{token} ] "+" @var{async-output nl}}
27915 @item @var{notify-async-output} @expansion{}
27916 @code{[ @var{token} ] "=" @var{async-output nl}}
27918 @item @var{async-output} @expansion{}
27919 @code{@var{async-class} ( "," @var{result} )*}
27921 @item @var{result-class} @expansion{}
27922 @code{"done" | "running" | "connected" | "error" | "exit"}
27924 @item @var{async-class} @expansion{}
27925 @code{"stopped" | @var{others}} (where @var{others} will be added
27926 depending on the needs---this is still in development).
27928 @item @var{result} @expansion{}
27929 @code{ @var{variable} "=" @var{value}}
27931 @item @var{variable} @expansion{}
27932 @code{ @var{string} }
27934 @item @var{value} @expansion{}
27935 @code{ @var{const} | @var{tuple} | @var{list} }
27937 @item @var{const} @expansion{}
27938 @code{@var{c-string}}
27940 @item @var{tuple} @expansion{}
27941 @code{ "@{@}" | "@{" @var{result} ( "," @var{result} )* "@}" }
27943 @item @var{list} @expansion{}
27944 @code{ "[]" | "[" @var{value} ( "," @var{value} )* "]" | "["
27945 @var{result} ( "," @var{result} )* "]" }
27947 @item @var{stream-record} @expansion{}
27948 @code{@var{console-stream-output} | @var{target-stream-output} | @var{log-stream-output}}
27950 @item @var{console-stream-output} @expansion{}
27951 @code{"~" @var{c-string nl}}
27953 @item @var{target-stream-output} @expansion{}
27954 @code{"@@" @var{c-string nl}}
27956 @item @var{log-stream-output} @expansion{}
27957 @code{"&" @var{c-string nl}}
27959 @item @var{nl} @expansion{}
27962 @item @var{token} @expansion{}
27963 @emph{any sequence of digits}.
27971 All output sequences end in a single line containing a period.
27974 The @code{@var{token}} is from the corresponding request. Note that
27975 for all async output, while the token is allowed by the grammar and
27976 may be output by future versions of @value{GDBN} for select async
27977 output messages, it is generally omitted. Frontends should treat
27978 all async output as reporting general changes in the state of the
27979 target and there should be no need to associate async output to any
27983 @cindex status output in @sc{gdb/mi}
27984 @var{status-async-output} contains on-going status information about the
27985 progress of a slow operation. It can be discarded. All status output is
27986 prefixed by @samp{+}.
27989 @cindex async output in @sc{gdb/mi}
27990 @var{exec-async-output} contains asynchronous state change on the target
27991 (stopped, started, disappeared). All async output is prefixed by
27995 @cindex notify output in @sc{gdb/mi}
27996 @var{notify-async-output} contains supplementary information that the
27997 client should handle (e.g., a new breakpoint information). All notify
27998 output is prefixed by @samp{=}.
28001 @cindex console output in @sc{gdb/mi}
28002 @var{console-stream-output} is output that should be displayed as is in the
28003 console. It is the textual response to a CLI command. All the console
28004 output is prefixed by @samp{~}.
28007 @cindex target output in @sc{gdb/mi}
28008 @var{target-stream-output} is the output produced by the target program.
28009 All the target output is prefixed by @samp{@@}.
28012 @cindex log output in @sc{gdb/mi}
28013 @var{log-stream-output} is output text coming from @value{GDBN}'s internals, for
28014 instance messages that should be displayed as part of an error log. All
28015 the log output is prefixed by @samp{&}.
28018 @cindex list output in @sc{gdb/mi}
28019 New @sc{gdb/mi} commands should only output @var{lists} containing
28025 @xref{GDB/MI Stream Records, , @sc{gdb/mi} Stream Records}, for more
28026 details about the various output records.
28028 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28029 @node GDB/MI Compatibility with CLI
28030 @section @sc{gdb/mi} Compatibility with CLI
28032 @cindex compatibility, @sc{gdb/mi} and CLI
28033 @cindex @sc{gdb/mi}, compatibility with CLI
28035 For the developers convenience CLI commands can be entered directly,
28036 but there may be some unexpected behaviour. For example, commands
28037 that query the user will behave as if the user replied yes, breakpoint
28038 command lists are not executed and some CLI commands, such as
28039 @code{if}, @code{when} and @code{define}, prompt for further input with
28040 @samp{>}, which is not valid MI output.
28042 This feature may be removed at some stage in the future and it is
28043 recommended that front ends use the @code{-interpreter-exec} command
28044 (@pxref{-interpreter-exec}).
28046 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28047 @node GDB/MI Development and Front Ends
28048 @section @sc{gdb/mi} Development and Front Ends
28049 @cindex @sc{gdb/mi} development
28051 The application which takes the MI output and presents the state of the
28052 program being debugged to the user is called a @dfn{front end}.
28054 Since @sc{gdb/mi} is used by a variety of front ends to @value{GDBN}, changes
28055 to the MI interface may break existing usage. This section describes how the
28056 protocol changes and how to request previous version of the protocol when it
28059 Some changes in MI need not break a carefully designed front end, and
28060 for these the MI version will remain unchanged. The following is a
28061 list of changes that may occur within one level, so front ends should
28062 parse MI output in a way that can handle them:
28066 New MI commands may be added.
28069 New fields may be added to the output of any MI command.
28072 The range of values for fields with specified values, e.g.,
28073 @code{in_scope} (@pxref{-var-update}) may be extended.
28075 @c The format of field's content e.g type prefix, may change so parse it
28076 @c at your own risk. Yes, in general?
28078 @c The order of fields may change? Shouldn't really matter but it might
28079 @c resolve inconsistencies.
28082 If the changes are likely to break front ends, the MI version level
28083 will be increased by one. The new versions of the MI protocol are not compatible
28084 with the old versions. Old versions of MI remain available, allowing front ends
28085 to keep using them until they are modified to use the latest MI version.
28087 Since @code{--interpreter=mi} always points to the latest MI version, it is
28088 recommended that front ends request a specific version of MI when launching
28089 @value{GDBN} (e.g. @code{--interpreter=mi2}) to make sure they get an
28090 interpreter with the MI version they expect.
28092 The following table gives a summary of the the released versions of the MI
28093 interface: the version number, the version of GDB in which it first appeared
28094 and the breaking changes compared to the previous version.
28096 @multitable @columnfractions .05 .05 .9
28097 @headitem MI version @tab GDB version @tab Breaking changes
28114 The @code{-environment-pwd}, @code{-environment-directory} and
28115 @code{-environment-path} commands now returns values using the MI output
28116 syntax, rather than CLI output syntax.
28119 @code{-var-list-children}'s @code{children} result field is now a list, rather
28123 @code{-var-update}'s @code{changelist} result field is now a list, rather than
28135 The output of information about multi-location breakpoints has changed in the
28136 responses to the @code{-break-insert} and @code{-break-info} commands, as well
28137 as in the @code{=breakpoint-created} and @code{=breakpoint-modified} events.
28138 The multiple locations are now placed in a @code{locations} field, whose value
28144 If your front end cannot yet migrate to a more recent version of the
28145 MI protocol, you can nevertheless selectively enable specific features
28146 available in those recent MI versions, using the following commands:
28150 @item -fix-multi-location-breakpoint-output
28151 Use the output for multi-location breakpoints which was introduced by
28152 MI 3, even when using MI versions 2 or 1. This command has no
28153 effect when using MI version 3 or later.
28157 The best way to avoid unexpected changes in MI that might break your front
28158 end is to make your project known to @value{GDBN} developers and
28159 follow development on @email{gdb@@sourceware.org} and
28160 @email{gdb-patches@@sourceware.org}.
28161 @cindex mailing lists
28163 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28164 @node GDB/MI Output Records
28165 @section @sc{gdb/mi} Output Records
28168 * GDB/MI Result Records::
28169 * GDB/MI Stream Records::
28170 * GDB/MI Async Records::
28171 * GDB/MI Breakpoint Information::
28172 * GDB/MI Frame Information::
28173 * GDB/MI Thread Information::
28174 * GDB/MI Ada Exception Information::
28177 @node GDB/MI Result Records
28178 @subsection @sc{gdb/mi} Result Records
28180 @cindex result records in @sc{gdb/mi}
28181 @cindex @sc{gdb/mi}, result records
28182 In addition to a number of out-of-band notifications, the response to a
28183 @sc{gdb/mi} command includes one of the following result indications:
28187 @item "^done" [ "," @var{results} ]
28188 The synchronous operation was successful, @code{@var{results}} are the return
28193 This result record is equivalent to @samp{^done}. Historically, it
28194 was output instead of @samp{^done} if the command has resumed the
28195 target. This behaviour is maintained for backward compatibility, but
28196 all frontends should treat @samp{^done} and @samp{^running}
28197 identically and rely on the @samp{*running} output record to determine
28198 which threads are resumed.
28202 @value{GDBN} has connected to a remote target.
28204 @item "^error" "," "msg=" @var{c-string} [ "," "code=" @var{c-string} ]
28206 The operation failed. The @code{msg=@var{c-string}} variable contains
28207 the corresponding error message.
28209 If present, the @code{code=@var{c-string}} variable provides an error
28210 code on which consumers can rely on to detect the corresponding
28211 error condition. At present, only one error code is defined:
28214 @item "undefined-command"
28215 Indicates that the command causing the error does not exist.
28220 @value{GDBN} has terminated.
28224 @node GDB/MI Stream Records
28225 @subsection @sc{gdb/mi} Stream Records
28227 @cindex @sc{gdb/mi}, stream records
28228 @cindex stream records in @sc{gdb/mi}
28229 @value{GDBN} internally maintains a number of output streams: the console, the
28230 target, and the log. The output intended for each of these streams is
28231 funneled through the @sc{gdb/mi} interface using @dfn{stream records}.
28233 Each stream record begins with a unique @dfn{prefix character} which
28234 identifies its stream (@pxref{GDB/MI Output Syntax, , @sc{gdb/mi} Output
28235 Syntax}). In addition to the prefix, each stream record contains a
28236 @code{@var{string-output}}. This is either raw text (with an implicit new
28237 line) or a quoted C string (which does not contain an implicit newline).
28240 @item "~" @var{string-output}
28241 The console output stream contains text that should be displayed in the
28242 CLI console window. It contains the textual responses to CLI commands.
28244 @item "@@" @var{string-output}
28245 The target output stream contains any textual output from the running
28246 target. This is only present when GDB's event loop is truly
28247 asynchronous, which is currently only the case for remote targets.
28249 @item "&" @var{string-output}
28250 The log stream contains debugging messages being produced by @value{GDBN}'s
28254 @node GDB/MI Async Records
28255 @subsection @sc{gdb/mi} Async Records
28257 @cindex async records in @sc{gdb/mi}
28258 @cindex @sc{gdb/mi}, async records
28259 @dfn{Async} records are used to notify the @sc{gdb/mi} client of
28260 additional changes that have occurred. Those changes can either be a
28261 consequence of @sc{gdb/mi} commands (e.g., a breakpoint modified) or a result of
28262 target activity (e.g., target stopped).
28264 The following is the list of possible async records:
28268 @item *running,thread-id="@var{thread}"
28269 The target is now running. The @var{thread} field can be the global
28270 thread ID of the the thread that is now running, and it can be
28271 @samp{all} if all threads are running. The frontend should assume
28272 that no interaction with a running thread is possible after this
28273 notification is produced. The frontend should not assume that this
28274 notification is output only once for any command. @value{GDBN} may
28275 emit this notification several times, either for different threads,
28276 because it cannot resume all threads together, or even for a single
28277 thread, if the thread must be stepped though some code before letting
28280 @item *stopped,reason="@var{reason}",thread-id="@var{id}",stopped-threads="@var{stopped}",core="@var{core}"
28281 The target has stopped. The @var{reason} field can have one of the
28285 @item breakpoint-hit
28286 A breakpoint was reached.
28287 @item watchpoint-trigger
28288 A watchpoint was triggered.
28289 @item read-watchpoint-trigger
28290 A read watchpoint was triggered.
28291 @item access-watchpoint-trigger
28292 An access watchpoint was triggered.
28293 @item function-finished
28294 An -exec-finish or similar CLI command was accomplished.
28295 @item location-reached
28296 An -exec-until or similar CLI command was accomplished.
28297 @item watchpoint-scope
28298 A watchpoint has gone out of scope.
28299 @item end-stepping-range
28300 An -exec-next, -exec-next-instruction, -exec-step, -exec-step-instruction or
28301 similar CLI command was accomplished.
28302 @item exited-signalled
28303 The inferior exited because of a signal.
28305 The inferior exited.
28306 @item exited-normally
28307 The inferior exited normally.
28308 @item signal-received
28309 A signal was received by the inferior.
28311 The inferior has stopped due to a library being loaded or unloaded.
28312 This can happen when @code{stop-on-solib-events} (@pxref{Files}) is
28313 set or when a @code{catch load} or @code{catch unload} catchpoint is
28314 in use (@pxref{Set Catchpoints}).
28316 The inferior has forked. This is reported when @code{catch fork}
28317 (@pxref{Set Catchpoints}) has been used.
28319 The inferior has vforked. This is reported in when @code{catch vfork}
28320 (@pxref{Set Catchpoints}) has been used.
28321 @item syscall-entry
28322 The inferior entered a system call. This is reported when @code{catch
28323 syscall} (@pxref{Set Catchpoints}) has been used.
28324 @item syscall-return
28325 The inferior returned from a system call. This is reported when
28326 @code{catch syscall} (@pxref{Set Catchpoints}) has been used.
28328 The inferior called @code{exec}. This is reported when @code{catch exec}
28329 (@pxref{Set Catchpoints}) has been used.
28332 The @var{id} field identifies the global thread ID of the thread
28333 that directly caused the stop -- for example by hitting a breakpoint.
28334 Depending on whether all-stop
28335 mode is in effect (@pxref{All-Stop Mode}), @value{GDBN} may either
28336 stop all threads, or only the thread that directly triggered the stop.
28337 If all threads are stopped, the @var{stopped} field will have the
28338 value of @code{"all"}. Otherwise, the value of the @var{stopped}
28339 field will be a list of thread identifiers. Presently, this list will
28340 always include a single thread, but frontend should be prepared to see
28341 several threads in the list. The @var{core} field reports the
28342 processor core on which the stop event has happened. This field may be absent
28343 if such information is not available.
28345 @item =thread-group-added,id="@var{id}"
28346 @itemx =thread-group-removed,id="@var{id}"
28347 A thread group was either added or removed. The @var{id} field
28348 contains the @value{GDBN} identifier of the thread group. When a thread
28349 group is added, it generally might not be associated with a running
28350 process. When a thread group is removed, its id becomes invalid and
28351 cannot be used in any way.
28353 @item =thread-group-started,id="@var{id}",pid="@var{pid}"
28354 A thread group became associated with a running program,
28355 either because the program was just started or the thread group
28356 was attached to a program. The @var{id} field contains the
28357 @value{GDBN} identifier of the thread group. The @var{pid} field
28358 contains process identifier, specific to the operating system.
28360 @item =thread-group-exited,id="@var{id}"[,exit-code="@var{code}"]
28361 A thread group is no longer associated with a running program,
28362 either because the program has exited, or because it was detached
28363 from. The @var{id} field contains the @value{GDBN} identifier of the
28364 thread group. The @var{code} field is the exit code of the inferior; it exists
28365 only when the inferior exited with some code.
28367 @item =thread-created,id="@var{id}",group-id="@var{gid}"
28368 @itemx =thread-exited,id="@var{id}",group-id="@var{gid}"
28369 A thread either was created, or has exited. The @var{id} field
28370 contains the global @value{GDBN} identifier of the thread. The @var{gid}
28371 field identifies the thread group this thread belongs to.
28373 @item =thread-selected,id="@var{id}"[,frame="@var{frame}"]
28374 Informs that the selected thread or frame were changed. This notification
28375 is not emitted as result of the @code{-thread-select} or
28376 @code{-stack-select-frame} commands, but is emitted whenever an MI command
28377 that is not documented to change the selected thread and frame actually
28378 changes them. In particular, invoking, directly or indirectly
28379 (via user-defined command), the CLI @code{thread} or @code{frame} commands,
28380 will generate this notification. Changing the thread or frame from another
28381 user interface (see @ref{Interpreters}) will also generate this notification.
28383 The @var{frame} field is only present if the newly selected thread is
28384 stopped. See @ref{GDB/MI Frame Information} for the format of its value.
28386 We suggest that in response to this notification, front ends
28387 highlight the selected thread and cause subsequent commands to apply to
28390 @item =library-loaded,...
28391 Reports that a new library file was loaded by the program. This
28392 notification has 5 fields---@var{id}, @var{target-name},
28393 @var{host-name}, @var{symbols-loaded} and @var{ranges}. The @var{id} field is an
28394 opaque identifier of the library. For remote debugging case,
28395 @var{target-name} and @var{host-name} fields give the name of the
28396 library file on the target, and on the host respectively. For native
28397 debugging, both those fields have the same value. The
28398 @var{symbols-loaded} field is emitted only for backward compatibility
28399 and should not be relied on to convey any useful information. The
28400 @var{thread-group} field, if present, specifies the id of the thread
28401 group in whose context the library was loaded. If the field is
28402 absent, it means the library was loaded in the context of all present
28403 thread groups. The @var{ranges} field specifies the ranges of addresses belonging
28406 @item =library-unloaded,...
28407 Reports that a library was unloaded by the program. This notification
28408 has 3 fields---@var{id}, @var{target-name} and @var{host-name} with
28409 the same meaning as for the @code{=library-loaded} notification.
28410 The @var{thread-group} field, if present, specifies the id of the
28411 thread group in whose context the library was unloaded. If the field is
28412 absent, it means the library was unloaded in the context of all present
28415 @item =traceframe-changed,num=@var{tfnum},tracepoint=@var{tpnum}
28416 @itemx =traceframe-changed,end
28417 Reports that the trace frame was changed and its new number is
28418 @var{tfnum}. The number of the tracepoint associated with this trace
28419 frame is @var{tpnum}.
28421 @item =tsv-created,name=@var{name},initial=@var{initial}
28422 Reports that the new trace state variable @var{name} is created with
28423 initial value @var{initial}.
28425 @item =tsv-deleted,name=@var{name}
28426 @itemx =tsv-deleted
28427 Reports that the trace state variable @var{name} is deleted or all
28428 trace state variables are deleted.
28430 @item =tsv-modified,name=@var{name},initial=@var{initial}[,current=@var{current}]
28431 Reports that the trace state variable @var{name} is modified with
28432 the initial value @var{initial}. The current value @var{current} of
28433 trace state variable is optional and is reported if the current
28434 value of trace state variable is known.
28436 @item =breakpoint-created,bkpt=@{...@}
28437 @itemx =breakpoint-modified,bkpt=@{...@}
28438 @itemx =breakpoint-deleted,id=@var{number}
28439 Reports that a breakpoint was created, modified, or deleted,
28440 respectively. Only user-visible breakpoints are reported to the MI
28443 The @var{bkpt} argument is of the same form as returned by the various
28444 breakpoint commands; @xref{GDB/MI Breakpoint Commands}. The
28445 @var{number} is the ordinal number of the breakpoint.
28447 Note that if a breakpoint is emitted in the result record of a
28448 command, then it will not also be emitted in an async record.
28450 @item =record-started,thread-group="@var{id}",method="@var{method}"[,format="@var{format}"]
28451 @itemx =record-stopped,thread-group="@var{id}"
28452 Execution log recording was either started or stopped on an
28453 inferior. The @var{id} is the @value{GDBN} identifier of the thread
28454 group corresponding to the affected inferior.
28456 The @var{method} field indicates the method used to record execution. If the
28457 method in use supports multiple recording formats, @var{format} will be present
28458 and contain the currently used format. @xref{Process Record and Replay},
28459 for existing method and format values.
28461 @item =cmd-param-changed,param=@var{param},value=@var{value}
28462 Reports that a parameter of the command @code{set @var{param}} is
28463 changed to @var{value}. In the multi-word @code{set} command,
28464 the @var{param} is the whole parameter list to @code{set} command.
28465 For example, In command @code{set check type on}, @var{param}
28466 is @code{check type} and @var{value} is @code{on}.
28468 @item =memory-changed,thread-group=@var{id},addr=@var{addr},len=@var{len}[,type="code"]
28469 Reports that bytes from @var{addr} to @var{data} + @var{len} were
28470 written in an inferior. The @var{id} is the identifier of the
28471 thread group corresponding to the affected inferior. The optional
28472 @code{type="code"} part is reported if the memory written to holds
28476 @node GDB/MI Breakpoint Information
28477 @subsection @sc{gdb/mi} Breakpoint Information
28479 When @value{GDBN} reports information about a breakpoint, a
28480 tracepoint, a watchpoint, or a catchpoint, it uses a tuple with the
28485 The breakpoint number.
28488 The type of the breakpoint. For ordinary breakpoints this will be
28489 @samp{breakpoint}, but many values are possible.
28492 If the type of the breakpoint is @samp{catchpoint}, then this
28493 indicates the exact type of catchpoint.
28496 This is the breakpoint disposition---either @samp{del}, meaning that
28497 the breakpoint will be deleted at the next stop, or @samp{keep},
28498 meaning that the breakpoint will not be deleted.
28501 This indicates whether the breakpoint is enabled, in which case the
28502 value is @samp{y}, or disabled, in which case the value is @samp{n}.
28503 Note that this is not the same as the field @code{enable}.
28506 The address of the breakpoint. This may be a hexidecimal number,
28507 giving the address; or the string @samp{<PENDING>}, for a pending
28508 breakpoint; or the string @samp{<MULTIPLE>}, for a breakpoint with
28509 multiple locations. This field will not be present if no address can
28510 be determined. For example, a watchpoint does not have an address.
28513 If known, the function in which the breakpoint appears.
28514 If not known, this field is not present.
28517 The name of the source file which contains this function, if known.
28518 If not known, this field is not present.
28521 The full file name of the source file which contains this function, if
28522 known. If not known, this field is not present.
28525 The line number at which this breakpoint appears, if known.
28526 If not known, this field is not present.
28529 If the source file is not known, this field may be provided. If
28530 provided, this holds the address of the breakpoint, possibly followed
28534 If this breakpoint is pending, this field is present and holds the
28535 text used to set the breakpoint, as entered by the user.
28538 Where this breakpoint's condition is evaluated, either @samp{host} or
28542 If this is a thread-specific breakpoint, then this identifies the
28543 thread in which the breakpoint can trigger.
28546 If this breakpoint is restricted to a particular Ada task, then this
28547 field will hold the task identifier.
28550 If the breakpoint is conditional, this is the condition expression.
28553 The ignore count of the breakpoint.
28556 The enable count of the breakpoint.
28558 @item traceframe-usage
28561 @item static-tracepoint-marker-string-id
28562 For a static tracepoint, the name of the static tracepoint marker.
28565 For a masked watchpoint, this is the mask.
28568 A tracepoint's pass count.
28570 @item original-location
28571 The location of the breakpoint as originally specified by the user.
28572 This field is optional.
28575 The number of times the breakpoint has been hit.
28578 This field is only given for tracepoints. This is either @samp{y},
28579 meaning that the tracepoint is installed, or @samp{n}, meaning that it
28583 Some extra data, the exact contents of which are type-dependent.
28586 This field is present if the breakpoint has multiple locations. It is also
28587 exceptionally present if the breakpoint is enabled and has a single, disabled
28590 The value is a list of locations. The format of a location is decribed below.
28594 A location in a multi-location breakpoint is represented as a tuple with the
28600 The location number as a dotted pair, like @samp{1.2}. The first digit is the
28601 number of the parent breakpoint. The second digit is the number of the
28602 location within that breakpoint.
28605 This indicates whether the location is enabled, in which case the
28606 value is @samp{y}, or disabled, in which case the value is @samp{n}.
28607 Note that this is not the same as the field @code{enable}.
28610 The address of this location as an hexidecimal number.
28613 If known, the function in which the location appears.
28614 If not known, this field is not present.
28617 The name of the source file which contains this location, if known.
28618 If not known, this field is not present.
28621 The full file name of the source file which contains this location, if
28622 known. If not known, this field is not present.
28625 The line number at which this location appears, if known.
28626 If not known, this field is not present.
28628 @item thread-groups
28629 The thread groups this location is in.
28633 For example, here is what the output of @code{-break-insert}
28634 (@pxref{GDB/MI Breakpoint Commands}) might be:
28637 -> -break-insert main
28638 <- ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
28639 enabled="y",addr="0x08048564",func="main",file="myprog.c",
28640 fullname="/home/nickrob/myprog.c",line="68",thread-groups=["i1"],
28645 @node GDB/MI Frame Information
28646 @subsection @sc{gdb/mi} Frame Information
28648 Response from many MI commands includes an information about stack
28649 frame. This information is a tuple that may have the following
28654 The level of the stack frame. The innermost frame has the level of
28655 zero. This field is always present.
28658 The name of the function corresponding to the frame. This field may
28659 be absent if @value{GDBN} is unable to determine the function name.
28662 The code address for the frame. This field is always present.
28665 The name of the source files that correspond to the frame's code
28666 address. This field may be absent.
28669 The source line corresponding to the frames' code address. This field
28673 The name of the binary file (either executable or shared library) the
28674 corresponds to the frame's code address. This field may be absent.
28678 @node GDB/MI Thread Information
28679 @subsection @sc{gdb/mi} Thread Information
28681 Whenever @value{GDBN} has to report an information about a thread, it
28682 uses a tuple with the following fields. The fields are always present unless
28687 The global numeric id assigned to the thread by @value{GDBN}.
28690 The target-specific string identifying the thread.
28693 Additional information about the thread provided by the target.
28694 It is supposed to be human-readable and not interpreted by the
28695 frontend. This field is optional.
28698 The name of the thread. If the user specified a name using the
28699 @code{thread name} command, then this name is given. Otherwise, if
28700 @value{GDBN} can extract the thread name from the target, then that
28701 name is given. If @value{GDBN} cannot find the thread name, then this
28705 The execution state of the thread, either @samp{stopped} or @samp{running},
28706 depending on whether the thread is presently running.
28709 The stack frame currently executing in the thread. This field is only present
28710 if the thread is stopped. Its format is documented in
28711 @ref{GDB/MI Frame Information}.
28714 The value of this field is an integer number of the processor core the
28715 thread was last seen on. This field is optional.
28718 @node GDB/MI Ada Exception Information
28719 @subsection @sc{gdb/mi} Ada Exception Information
28721 Whenever a @code{*stopped} record is emitted because the program
28722 stopped after hitting an exception catchpoint (@pxref{Set Catchpoints}),
28723 @value{GDBN} provides the name of the exception that was raised via
28724 the @code{exception-name} field. Also, for exceptions that were raised
28725 with an exception message, @value{GDBN} provides that message via
28726 the @code{exception-message} field.
28728 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28729 @node GDB/MI Simple Examples
28730 @section Simple Examples of @sc{gdb/mi} Interaction
28731 @cindex @sc{gdb/mi}, simple examples
28733 This subsection presents several simple examples of interaction using
28734 the @sc{gdb/mi} interface. In these examples, @samp{->} means that the
28735 following line is passed to @sc{gdb/mi} as input, while @samp{<-} means
28736 the output received from @sc{gdb/mi}.
28738 Note the line breaks shown in the examples are here only for
28739 readability, they don't appear in the real output.
28741 @subheading Setting a Breakpoint
28743 Setting a breakpoint generates synchronous output which contains detailed
28744 information of the breakpoint.
28747 -> -break-insert main
28748 <- ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
28749 enabled="y",addr="0x08048564",func="main",file="myprog.c",
28750 fullname="/home/nickrob/myprog.c",line="68",thread-groups=["i1"],
28755 @subheading Program Execution
28757 Program execution generates asynchronous records and MI gives the
28758 reason that execution stopped.
28764 <- *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
28765 frame=@{addr="0x08048564",func="main",
28766 args=[@{name="argc",value="1"@},@{name="argv",value="0xbfc4d4d4"@}],
28767 file="myprog.c",fullname="/home/nickrob/myprog.c",line="68",
28768 arch="i386:x86_64"@}
28773 <- *stopped,reason="exited-normally"
28777 @subheading Quitting @value{GDBN}
28779 Quitting @value{GDBN} just prints the result class @samp{^exit}.
28787 Please note that @samp{^exit} is printed immediately, but it might
28788 take some time for @value{GDBN} to actually exit. During that time, @value{GDBN}
28789 performs necessary cleanups, including killing programs being debugged
28790 or disconnecting from debug hardware, so the frontend should wait till
28791 @value{GDBN} exits and should only forcibly kill @value{GDBN} if it
28792 fails to exit in reasonable time.
28794 @subheading A Bad Command
28796 Here's what happens if you pass a non-existent command:
28800 <- ^error,msg="Undefined MI command: rubbish"
28805 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28806 @node GDB/MI Command Description Format
28807 @section @sc{gdb/mi} Command Description Format
28809 The remaining sections describe blocks of commands. Each block of
28810 commands is laid out in a fashion similar to this section.
28812 @subheading Motivation
28814 The motivation for this collection of commands.
28816 @subheading Introduction
28818 A brief introduction to this collection of commands as a whole.
28820 @subheading Commands
28822 For each command in the block, the following is described:
28824 @subsubheading Synopsis
28827 -command @var{args}@dots{}
28830 @subsubheading Result
28832 @subsubheading @value{GDBN} Command
28834 The corresponding @value{GDBN} CLI command(s), if any.
28836 @subsubheading Example
28838 Example(s) formatted for readability. Some of the described commands have
28839 not been implemented yet and these are labeled N.A.@: (not available).
28842 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28843 @node GDB/MI Breakpoint Commands
28844 @section @sc{gdb/mi} Breakpoint Commands
28846 @cindex breakpoint commands for @sc{gdb/mi}
28847 @cindex @sc{gdb/mi}, breakpoint commands
28848 This section documents @sc{gdb/mi} commands for manipulating
28851 @subheading The @code{-break-after} Command
28852 @findex -break-after
28854 @subsubheading Synopsis
28857 -break-after @var{number} @var{count}
28860 The breakpoint number @var{number} is not in effect until it has been
28861 hit @var{count} times. To see how this is reflected in the output of
28862 the @samp{-break-list} command, see the description of the
28863 @samp{-break-list} command below.
28865 @subsubheading @value{GDBN} Command
28867 The corresponding @value{GDBN} command is @samp{ignore}.
28869 @subsubheading Example
28874 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
28875 enabled="y",addr="0x000100d0",func="main",file="hello.c",
28876 fullname="/home/foo/hello.c",line="5",thread-groups=["i1"],
28884 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
28885 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
28886 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
28887 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
28888 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
28889 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
28890 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
28891 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
28892 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
28893 line="5",thread-groups=["i1"],times="0",ignore="3"@}]@}
28898 @subheading The @code{-break-catch} Command
28899 @findex -break-catch
28902 @subheading The @code{-break-commands} Command
28903 @findex -break-commands
28905 @subsubheading Synopsis
28908 -break-commands @var{number} [ @var{command1} ... @var{commandN} ]
28911 Specifies the CLI commands that should be executed when breakpoint
28912 @var{number} is hit. The parameters @var{command1} to @var{commandN}
28913 are the commands. If no command is specified, any previously-set
28914 commands are cleared. @xref{Break Commands}. Typical use of this
28915 functionality is tracing a program, that is, printing of values of
28916 some variables whenever breakpoint is hit and then continuing.
28918 @subsubheading @value{GDBN} Command
28920 The corresponding @value{GDBN} command is @samp{commands}.
28922 @subsubheading Example
28927 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
28928 enabled="y",addr="0x000100d0",func="main",file="hello.c",
28929 fullname="/home/foo/hello.c",line="5",thread-groups=["i1"],
28932 -break-commands 1 "print v" "continue"
28937 @subheading The @code{-break-condition} Command
28938 @findex -break-condition
28940 @subsubheading Synopsis
28943 -break-condition @var{number} @var{expr}
28946 Breakpoint @var{number} will stop the program only if the condition in
28947 @var{expr} is true. The condition becomes part of the
28948 @samp{-break-list} output (see the description of the @samp{-break-list}
28951 @subsubheading @value{GDBN} Command
28953 The corresponding @value{GDBN} command is @samp{condition}.
28955 @subsubheading Example
28959 -break-condition 1 1
28963 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
28964 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
28965 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
28966 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
28967 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
28968 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
28969 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
28970 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
28971 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
28972 line="5",cond="1",thread-groups=["i1"],times="0",ignore="3"@}]@}
28976 @subheading The @code{-break-delete} Command
28977 @findex -break-delete
28979 @subsubheading Synopsis
28982 -break-delete ( @var{breakpoint} )+
28985 Delete the breakpoint(s) whose number(s) are specified in the argument
28986 list. This is obviously reflected in the breakpoint list.
28988 @subsubheading @value{GDBN} Command
28990 The corresponding @value{GDBN} command is @samp{delete}.
28992 @subsubheading Example
29000 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
29001 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29002 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29003 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29004 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29005 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29006 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29011 @subheading The @code{-break-disable} Command
29012 @findex -break-disable
29014 @subsubheading Synopsis
29017 -break-disable ( @var{breakpoint} )+
29020 Disable the named @var{breakpoint}(s). The field @samp{enabled} in the
29021 break list is now set to @samp{n} for the named @var{breakpoint}(s).
29023 @subsubheading @value{GDBN} Command
29025 The corresponding @value{GDBN} command is @samp{disable}.
29027 @subsubheading Example
29035 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
29036 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29037 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29038 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29039 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29040 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29041 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29042 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="n",
29043 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
29044 line="5",thread-groups=["i1"],times="0"@}]@}
29048 @subheading The @code{-break-enable} Command
29049 @findex -break-enable
29051 @subsubheading Synopsis
29054 -break-enable ( @var{breakpoint} )+
29057 Enable (previously disabled) @var{breakpoint}(s).
29059 @subsubheading @value{GDBN} Command
29061 The corresponding @value{GDBN} command is @samp{enable}.
29063 @subsubheading Example
29071 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
29072 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29073 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29074 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29075 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29076 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29077 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29078 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
29079 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
29080 line="5",thread-groups=["i1"],times="0"@}]@}
29084 @subheading The @code{-break-info} Command
29085 @findex -break-info
29087 @subsubheading Synopsis
29090 -break-info @var{breakpoint}
29094 Get information about a single breakpoint.
29096 The result is a table of breakpoints. @xref{GDB/MI Breakpoint
29097 Information}, for details on the format of each breakpoint in the
29100 @subsubheading @value{GDBN} Command
29102 The corresponding @value{GDBN} command is @samp{info break @var{breakpoint}}.
29104 @subsubheading Example
29107 @subheading The @code{-break-insert} Command
29108 @findex -break-insert
29109 @anchor{-break-insert}
29111 @subsubheading Synopsis
29114 -break-insert [ -t ] [ -h ] [ -f ] [ -d ] [ -a ]
29115 [ -c @var{condition} ] [ -i @var{ignore-count} ]
29116 [ -p @var{thread-id} ] [ @var{location} ]
29120 If specified, @var{location}, can be one of:
29123 @item linespec location
29124 A linespec location. @xref{Linespec Locations}.
29126 @item explicit location
29127 An explicit location. @sc{gdb/mi} explicit locations are
29128 analogous to the CLI's explicit locations using the option names
29129 listed below. @xref{Explicit Locations}.
29132 @item --source @var{filename}
29133 The source file name of the location. This option requires the use
29134 of either @samp{--function} or @samp{--line}.
29136 @item --function @var{function}
29137 The name of a function or method.
29139 @item --label @var{label}
29140 The name of a label.
29142 @item --line @var{lineoffset}
29143 An absolute or relative line offset from the start of the location.
29146 @item address location
29147 An address location, *@var{address}. @xref{Address Locations}.
29151 The possible optional parameters of this command are:
29155 Insert a temporary breakpoint.
29157 Insert a hardware breakpoint.
29159 If @var{location} cannot be parsed (for example if it
29160 refers to unknown files or functions), create a pending
29161 breakpoint. Without this flag, @value{GDBN} will report
29162 an error, and won't create a breakpoint, if @var{location}
29165 Create a disabled breakpoint.
29167 Create a tracepoint. @xref{Tracepoints}. When this parameter
29168 is used together with @samp{-h}, a fast tracepoint is created.
29169 @item -c @var{condition}
29170 Make the breakpoint conditional on @var{condition}.
29171 @item -i @var{ignore-count}
29172 Initialize the @var{ignore-count}.
29173 @item -p @var{thread-id}
29174 Restrict the breakpoint to the thread with the specified global
29178 @subsubheading Result
29180 @xref{GDB/MI Breakpoint Information}, for details on the format of the
29181 resulting breakpoint.
29183 Note: this format is open to change.
29184 @c An out-of-band breakpoint instead of part of the result?
29186 @subsubheading @value{GDBN} Command
29188 The corresponding @value{GDBN} commands are @samp{break}, @samp{tbreak},
29189 @samp{hbreak}, and @samp{thbreak}. @c and @samp{rbreak}.
29191 @subsubheading Example
29196 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",
29197 fullname="/home/foo/recursive2.c,line="4",thread-groups=["i1"],
29200 -break-insert -t foo
29201 ^done,bkpt=@{number="2",addr="0x00010774",file="recursive2.c",
29202 fullname="/home/foo/recursive2.c,line="11",thread-groups=["i1"],
29206 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
29207 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29208 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29209 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29210 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29211 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29212 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29213 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
29214 addr="0x0001072c", func="main",file="recursive2.c",
29215 fullname="/home/foo/recursive2.c,"line="4",thread-groups=["i1"],
29217 bkpt=@{number="2",type="breakpoint",disp="del",enabled="y",
29218 addr="0x00010774",func="foo",file="recursive2.c",
29219 fullname="/home/foo/recursive2.c",line="11",thread-groups=["i1"],
29222 @c -break-insert -r foo.*
29223 @c ~int foo(int, int);
29224 @c ^done,bkpt=@{number="3",addr="0x00010774",file="recursive2.c,
29225 @c "fullname="/home/foo/recursive2.c",line="11",thread-groups=["i1"],
29230 @subheading The @code{-dprintf-insert} Command
29231 @findex -dprintf-insert
29233 @subsubheading Synopsis
29236 -dprintf-insert [ -t ] [ -f ] [ -d ]
29237 [ -c @var{condition} ] [ -i @var{ignore-count} ]
29238 [ -p @var{thread-id} ] [ @var{location} ] [ @var{format} ]
29243 If supplied, @var{location} may be specified the same way as for
29244 the @code{-break-insert} command. @xref{-break-insert}.
29246 The possible optional parameters of this command are:
29250 Insert a temporary breakpoint.
29252 If @var{location} cannot be parsed (for example, if it
29253 refers to unknown files or functions), create a pending
29254 breakpoint. Without this flag, @value{GDBN} will report
29255 an error, and won't create a breakpoint, if @var{location}
29258 Create a disabled breakpoint.
29259 @item -c @var{condition}
29260 Make the breakpoint conditional on @var{condition}.
29261 @item -i @var{ignore-count}
29262 Set the ignore count of the breakpoint (@pxref{Conditions, ignore count})
29263 to @var{ignore-count}.
29264 @item -p @var{thread-id}
29265 Restrict the breakpoint to the thread with the specified global
29269 @subsubheading Result
29271 @xref{GDB/MI Breakpoint Information}, for details on the format of the
29272 resulting breakpoint.
29274 @c An out-of-band breakpoint instead of part of the result?
29276 @subsubheading @value{GDBN} Command
29278 The corresponding @value{GDBN} command is @samp{dprintf}.
29280 @subsubheading Example
29284 4-dprintf-insert foo "At foo entry\n"
29285 4^done,bkpt=@{number="1",type="dprintf",disp="keep",enabled="y",
29286 addr="0x000000000040061b",func="foo",file="mi-dprintf.c",
29287 fullname="mi-dprintf.c",line="25",thread-groups=["i1"],
29288 times="0",script=@{"printf \"At foo entry\\n\"","continue"@},
29289 original-location="foo"@}
29291 5-dprintf-insert 26 "arg=%d, g=%d\n" arg g
29292 5^done,bkpt=@{number="2",type="dprintf",disp="keep",enabled="y",
29293 addr="0x000000000040062a",func="foo",file="mi-dprintf.c",
29294 fullname="mi-dprintf.c",line="26",thread-groups=["i1"],
29295 times="0",script=@{"printf \"arg=%d, g=%d\\n\", arg, g","continue"@},
29296 original-location="mi-dprintf.c:26"@}
29300 @subheading The @code{-break-list} Command
29301 @findex -break-list
29303 @subsubheading Synopsis
29309 Displays the list of inserted breakpoints, showing the following fields:
29313 number of the breakpoint
29315 type of the breakpoint: @samp{breakpoint} or @samp{watchpoint}
29317 should the breakpoint be deleted or disabled when it is hit: @samp{keep}
29320 is the breakpoint enabled or no: @samp{y} or @samp{n}
29322 memory location at which the breakpoint is set
29324 logical location of the breakpoint, expressed by function name, file
29326 @item Thread-groups
29327 list of thread groups to which this breakpoint applies
29329 number of times the breakpoint has been hit
29332 If there are no breakpoints or watchpoints, the @code{BreakpointTable}
29333 @code{body} field is an empty list.
29335 @subsubheading @value{GDBN} Command
29337 The corresponding @value{GDBN} command is @samp{info break}.
29339 @subsubheading Example
29344 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
29345 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29346 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29347 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29348 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29349 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29350 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29351 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
29352 addr="0x000100d0",func="main",file="hello.c",line="5",thread-groups=["i1"],
29354 bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
29355 addr="0x00010114",func="foo",file="hello.c",fullname="/home/foo/hello.c",
29356 line="13",thread-groups=["i1"],times="0"@}]@}
29360 Here's an example of the result when there are no breakpoints:
29365 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
29366 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29367 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29368 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29369 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29370 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29371 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29376 @subheading The @code{-break-passcount} Command
29377 @findex -break-passcount
29379 @subsubheading Synopsis
29382 -break-passcount @var{tracepoint-number} @var{passcount}
29385 Set the passcount for tracepoint @var{tracepoint-number} to
29386 @var{passcount}. If the breakpoint referred to by @var{tracepoint-number}
29387 is not a tracepoint, error is emitted. This corresponds to CLI
29388 command @samp{passcount}.
29390 @subheading The @code{-break-watch} Command
29391 @findex -break-watch
29393 @subsubheading Synopsis
29396 -break-watch [ -a | -r ]
29399 Create a watchpoint. With the @samp{-a} option it will create an
29400 @dfn{access} watchpoint, i.e., a watchpoint that triggers either on a
29401 read from or on a write to the memory location. With the @samp{-r}
29402 option, the watchpoint created is a @dfn{read} watchpoint, i.e., it will
29403 trigger only when the memory location is accessed for reading. Without
29404 either of the options, the watchpoint created is a regular watchpoint,
29405 i.e., it will trigger when the memory location is accessed for writing.
29406 @xref{Set Watchpoints, , Setting Watchpoints}.
29408 Note that @samp{-break-list} will report a single list of watchpoints and
29409 breakpoints inserted.
29411 @subsubheading @value{GDBN} Command
29413 The corresponding @value{GDBN} commands are @samp{watch}, @samp{awatch}, and
29416 @subsubheading Example
29418 Setting a watchpoint on a variable in the @code{main} function:
29423 ^done,wpt=@{number="2",exp="x"@}
29428 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="x"@},
29429 value=@{old="-268439212",new="55"@},
29430 frame=@{func="main",args=[],file="recursive2.c",
29431 fullname="/home/foo/bar/recursive2.c",line="5",arch="i386:x86_64"@}
29435 Setting a watchpoint on a variable local to a function. @value{GDBN} will stop
29436 the program execution twice: first for the variable changing value, then
29437 for the watchpoint going out of scope.
29442 ^done,wpt=@{number="5",exp="C"@}
29447 *stopped,reason="watchpoint-trigger",
29448 wpt=@{number="5",exp="C"@},value=@{old="-276895068",new="3"@},
29449 frame=@{func="callee4",args=[],
29450 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29451 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13",
29452 arch="i386:x86_64"@}
29457 *stopped,reason="watchpoint-scope",wpnum="5",
29458 frame=@{func="callee3",args=[@{name="strarg",
29459 value="0x11940 \"A string argument.\""@}],
29460 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29461 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18",
29462 arch="i386:x86_64"@}
29466 Listing breakpoints and watchpoints, at different points in the program
29467 execution. Note that once the watchpoint goes out of scope, it is
29473 ^done,wpt=@{number="2",exp="C"@}
29476 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
29477 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29478 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29479 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29480 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29481 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29482 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29483 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
29484 addr="0x00010734",func="callee4",
29485 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29486 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c"line="8",thread-groups=["i1"],
29488 bkpt=@{number="2",type="watchpoint",disp="keep",
29489 enabled="y",addr="",what="C",thread-groups=["i1"],times="0"@}]@}
29494 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="C"@},
29495 value=@{old="-276895068",new="3"@},
29496 frame=@{func="callee4",args=[],
29497 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29498 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13",
29499 arch="i386:x86_64"@}
29502 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
29503 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29504 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29505 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29506 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29507 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29508 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29509 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
29510 addr="0x00010734",func="callee4",
29511 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29512 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",thread-groups=["i1"],
29514 bkpt=@{number="2",type="watchpoint",disp="keep",
29515 enabled="y",addr="",what="C",thread-groups=["i1"],times="-5"@}]@}
29519 ^done,reason="watchpoint-scope",wpnum="2",
29520 frame=@{func="callee3",args=[@{name="strarg",
29521 value="0x11940 \"A string argument.\""@}],
29522 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29523 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18",
29524 arch="i386:x86_64"@}
29527 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
29528 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29529 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29530 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29531 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29532 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29533 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29534 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
29535 addr="0x00010734",func="callee4",
29536 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29537 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",
29538 thread-groups=["i1"],times="1"@}]@}
29543 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
29544 @node GDB/MI Catchpoint Commands
29545 @section @sc{gdb/mi} Catchpoint Commands
29547 This section documents @sc{gdb/mi} commands for manipulating
29551 * Shared Library GDB/MI Catchpoint Commands::
29552 * Ada Exception GDB/MI Catchpoint Commands::
29555 @node Shared Library GDB/MI Catchpoint Commands
29556 @subsection Shared Library @sc{gdb/mi} Catchpoints
29558 @subheading The @code{-catch-load} Command
29559 @findex -catch-load
29561 @subsubheading Synopsis
29564 -catch-load [ -t ] [ -d ] @var{regexp}
29567 Add a catchpoint for library load events. If the @samp{-t} option is used,
29568 the catchpoint is a temporary one (@pxref{Set Breaks, ,Setting
29569 Breakpoints}). If the @samp{-d} option is used, the catchpoint is created
29570 in a disabled state. The @samp{regexp} argument is a regular
29571 expression used to match the name of the loaded library.
29574 @subsubheading @value{GDBN} Command
29576 The corresponding @value{GDBN} command is @samp{catch load}.
29578 @subsubheading Example
29581 -catch-load -t foo.so
29582 ^done,bkpt=@{number="1",type="catchpoint",disp="del",enabled="y",
29583 what="load of library matching foo.so",catch-type="load",times="0"@}
29588 @subheading The @code{-catch-unload} Command
29589 @findex -catch-unload
29591 @subsubheading Synopsis
29594 -catch-unload [ -t ] [ -d ] @var{regexp}
29597 Add a catchpoint for library unload events. If the @samp{-t} option is
29598 used, the catchpoint is a temporary one (@pxref{Set Breaks, ,Setting
29599 Breakpoints}). If the @samp{-d} option is used, the catchpoint is
29600 created in a disabled state. The @samp{regexp} argument is a regular
29601 expression used to match the name of the unloaded library.
29603 @subsubheading @value{GDBN} Command
29605 The corresponding @value{GDBN} command is @samp{catch unload}.
29607 @subsubheading Example
29610 -catch-unload -d bar.so
29611 ^done,bkpt=@{number="2",type="catchpoint",disp="keep",enabled="n",
29612 what="load of library matching bar.so",catch-type="unload",times="0"@}
29616 @node Ada Exception GDB/MI Catchpoint Commands
29617 @subsection Ada Exception @sc{gdb/mi} Catchpoints
29619 The following @sc{gdb/mi} commands can be used to create catchpoints
29620 that stop the execution when Ada exceptions are being raised.
29622 @subheading The @code{-catch-assert} Command
29623 @findex -catch-assert
29625 @subsubheading Synopsis
29628 -catch-assert [ -c @var{condition}] [ -d ] [ -t ]
29631 Add a catchpoint for failed Ada assertions.
29633 The possible optional parameters for this command are:
29636 @item -c @var{condition}
29637 Make the catchpoint conditional on @var{condition}.
29639 Create a disabled catchpoint.
29641 Create a temporary catchpoint.
29644 @subsubheading @value{GDBN} Command
29646 The corresponding @value{GDBN} command is @samp{catch assert}.
29648 @subsubheading Example
29652 ^done,bkptno="5",bkpt=@{number="5",type="breakpoint",disp="keep",
29653 enabled="y",addr="0x0000000000404888",what="failed Ada assertions",
29654 thread-groups=["i1"],times="0",
29655 original-location="__gnat_debug_raise_assert_failure"@}
29659 @subheading The @code{-catch-exception} Command
29660 @findex -catch-exception
29662 @subsubheading Synopsis
29665 -catch-exception [ -c @var{condition}] [ -d ] [ -e @var{exception-name} ]
29669 Add a catchpoint stopping when Ada exceptions are raised.
29670 By default, the command stops the program when any Ada exception
29671 gets raised. But it is also possible, by using some of the
29672 optional parameters described below, to create more selective
29675 The possible optional parameters for this command are:
29678 @item -c @var{condition}
29679 Make the catchpoint conditional on @var{condition}.
29681 Create a disabled catchpoint.
29682 @item -e @var{exception-name}
29683 Only stop when @var{exception-name} is raised. This option cannot
29684 be used combined with @samp{-u}.
29686 Create a temporary catchpoint.
29688 Stop only when an unhandled exception gets raised. This option
29689 cannot be used combined with @samp{-e}.
29692 @subsubheading @value{GDBN} Command
29694 The corresponding @value{GDBN} commands are @samp{catch exception}
29695 and @samp{catch exception unhandled}.
29697 @subsubheading Example
29700 -catch-exception -e Program_Error
29701 ^done,bkptno="4",bkpt=@{number="4",type="breakpoint",disp="keep",
29702 enabled="y",addr="0x0000000000404874",
29703 what="`Program_Error' Ada exception", thread-groups=["i1"],
29704 times="0",original-location="__gnat_debug_raise_exception"@}
29708 @subheading The @code{-catch-handlers} Command
29709 @findex -catch-handlers
29711 @subsubheading Synopsis
29714 -catch-handlers [ -c @var{condition}] [ -d ] [ -e @var{exception-name} ]
29718 Add a catchpoint stopping when Ada exceptions are handled.
29719 By default, the command stops the program when any Ada exception
29720 gets handled. But it is also possible, by using some of the
29721 optional parameters described below, to create more selective
29724 The possible optional parameters for this command are:
29727 @item -c @var{condition}
29728 Make the catchpoint conditional on @var{condition}.
29730 Create a disabled catchpoint.
29731 @item -e @var{exception-name}
29732 Only stop when @var{exception-name} is handled.
29734 Create a temporary catchpoint.
29737 @subsubheading @value{GDBN} Command
29739 The corresponding @value{GDBN} command is @samp{catch handlers}.
29741 @subsubheading Example
29744 -catch-handlers -e Constraint_Error
29745 ^done,bkptno="4",bkpt=@{number="4",type="breakpoint",disp="keep",
29746 enabled="y",addr="0x0000000000402f68",
29747 what="`Constraint_Error' Ada exception handlers",thread-groups=["i1"],
29748 times="0",original-location="__gnat_begin_handler"@}
29752 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
29753 @node GDB/MI Program Context
29754 @section @sc{gdb/mi} Program Context
29756 @subheading The @code{-exec-arguments} Command
29757 @findex -exec-arguments
29760 @subsubheading Synopsis
29763 -exec-arguments @var{args}
29766 Set the inferior program arguments, to be used in the next
29769 @subsubheading @value{GDBN} Command
29771 The corresponding @value{GDBN} command is @samp{set args}.
29773 @subsubheading Example
29777 -exec-arguments -v word
29784 @subheading The @code{-exec-show-arguments} Command
29785 @findex -exec-show-arguments
29787 @subsubheading Synopsis
29790 -exec-show-arguments
29793 Print the arguments of the program.
29795 @subsubheading @value{GDBN} Command
29797 The corresponding @value{GDBN} command is @samp{show args}.
29799 @subsubheading Example
29804 @subheading The @code{-environment-cd} Command
29805 @findex -environment-cd
29807 @subsubheading Synopsis
29810 -environment-cd @var{pathdir}
29813 Set @value{GDBN}'s working directory.
29815 @subsubheading @value{GDBN} Command
29817 The corresponding @value{GDBN} command is @samp{cd}.
29819 @subsubheading Example
29823 -environment-cd /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
29829 @subheading The @code{-environment-directory} Command
29830 @findex -environment-directory
29832 @subsubheading Synopsis
29835 -environment-directory [ -r ] [ @var{pathdir} ]+
29838 Add directories @var{pathdir} to beginning of search path for source files.
29839 If the @samp{-r} option is used, the search path is reset to the default
29840 search path. If directories @var{pathdir} are supplied in addition to the
29841 @samp{-r} option, the search path is first reset and then addition
29843 Multiple directories may be specified, separated by blanks. Specifying
29844 multiple directories in a single command
29845 results in the directories added to the beginning of the
29846 search path in the same order they were presented in the command.
29847 If blanks are needed as
29848 part of a directory name, double-quotes should be used around
29849 the name. In the command output, the path will show up separated
29850 by the system directory-separator character. The directory-separator
29851 character must not be used
29852 in any directory name.
29853 If no directories are specified, the current search path is displayed.
29855 @subsubheading @value{GDBN} Command
29857 The corresponding @value{GDBN} command is @samp{dir}.
29859 @subsubheading Example
29863 -environment-directory /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
29864 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
29866 -environment-directory ""
29867 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
29869 -environment-directory -r /home/jjohnstn/src/gdb /usr/src
29870 ^done,source-path="/home/jjohnstn/src/gdb:/usr/src:$cdir:$cwd"
29872 -environment-directory -r
29873 ^done,source-path="$cdir:$cwd"
29878 @subheading The @code{-environment-path} Command
29879 @findex -environment-path
29881 @subsubheading Synopsis
29884 -environment-path [ -r ] [ @var{pathdir} ]+
29887 Add directories @var{pathdir} to beginning of search path for object files.
29888 If the @samp{-r} option is used, the search path is reset to the original
29889 search path that existed at gdb start-up. If directories @var{pathdir} are
29890 supplied in addition to the
29891 @samp{-r} option, the search path is first reset and then addition
29893 Multiple directories may be specified, separated by blanks. Specifying
29894 multiple directories in a single command
29895 results in the directories added to the beginning of the
29896 search path in the same order they were presented in the command.
29897 If blanks are needed as
29898 part of a directory name, double-quotes should be used around
29899 the name. In the command output, the path will show up separated
29900 by the system directory-separator character. The directory-separator
29901 character must not be used
29902 in any directory name.
29903 If no directories are specified, the current path is displayed.
29906 @subsubheading @value{GDBN} Command
29908 The corresponding @value{GDBN} command is @samp{path}.
29910 @subsubheading Example
29915 ^done,path="/usr/bin"
29917 -environment-path /kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb /bin
29918 ^done,path="/kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb:/bin:/usr/bin"
29920 -environment-path -r /usr/local/bin
29921 ^done,path="/usr/local/bin:/usr/bin"
29926 @subheading The @code{-environment-pwd} Command
29927 @findex -environment-pwd
29929 @subsubheading Synopsis
29935 Show the current working directory.
29937 @subsubheading @value{GDBN} Command
29939 The corresponding @value{GDBN} command is @samp{pwd}.
29941 @subsubheading Example
29946 ^done,cwd="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb"
29950 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
29951 @node GDB/MI Thread Commands
29952 @section @sc{gdb/mi} Thread Commands
29955 @subheading The @code{-thread-info} Command
29956 @findex -thread-info
29958 @subsubheading Synopsis
29961 -thread-info [ @var{thread-id} ]
29964 Reports information about either a specific thread, if the
29965 @var{thread-id} parameter is present, or about all threads.
29966 @var{thread-id} is the thread's global thread ID. When printing
29967 information about all threads, also reports the global ID of the
29970 @subsubheading @value{GDBN} Command
29972 The @samp{info thread} command prints the same information
29975 @subsubheading Result
29977 The result contains the following attributes:
29981 A list of threads. The format of the elements of the list is described in
29982 @ref{GDB/MI Thread Information}.
29984 @item current-thread-id
29985 The global id of the currently selected thread. This field is omitted if there
29986 is no selected thread (for example, when the selected inferior is not running,
29987 and therefore has no threads) or if a @var{thread-id} argument was passed to
29992 @subsubheading Example
29997 @{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
29998 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",
29999 args=[]@},state="running"@},
30000 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
30001 frame=@{level="0",addr="0x0804891f",func="foo",
30002 args=[@{name="i",value="10"@}],
30003 file="/tmp/a.c",fullname="/tmp/a.c",line="158",arch="i386:x86_64"@},
30004 state="running"@}],
30005 current-thread-id="1"
30009 @subheading The @code{-thread-list-ids} Command
30010 @findex -thread-list-ids
30012 @subsubheading Synopsis
30018 Produces a list of the currently known global @value{GDBN} thread ids.
30019 At the end of the list it also prints the total number of such
30022 This command is retained for historical reasons, the
30023 @code{-thread-info} command should be used instead.
30025 @subsubheading @value{GDBN} Command
30027 Part of @samp{info threads} supplies the same information.
30029 @subsubheading Example
30034 ^done,thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
30035 current-thread-id="1",number-of-threads="3"
30040 @subheading The @code{-thread-select} Command
30041 @findex -thread-select
30043 @subsubheading Synopsis
30046 -thread-select @var{thread-id}
30049 Make thread with global thread number @var{thread-id} the current
30050 thread. It prints the number of the new current thread, and the
30051 topmost frame for that thread.
30053 This command is deprecated in favor of explicitly using the
30054 @samp{--thread} option to each command.
30056 @subsubheading @value{GDBN} Command
30058 The corresponding @value{GDBN} command is @samp{thread}.
30060 @subsubheading Example
30067 *stopped,reason="end-stepping-range",thread-id="2",line="187",
30068 file="../../../devo/gdb/testsuite/gdb.threads/linux-dp.c"
30072 thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
30073 number-of-threads="3"
30076 ^done,new-thread-id="3",
30077 frame=@{level="0",func="vprintf",
30078 args=[@{name="format",value="0x8048e9c \"%*s%c %d %c\\n\""@},
30079 @{name="arg",value="0x2"@}],file="vprintf.c",line="31",arch="i386:x86_64"@}
30083 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30084 @node GDB/MI Ada Tasking Commands
30085 @section @sc{gdb/mi} Ada Tasking Commands
30087 @subheading The @code{-ada-task-info} Command
30088 @findex -ada-task-info
30090 @subsubheading Synopsis
30093 -ada-task-info [ @var{task-id} ]
30096 Reports information about either a specific Ada task, if the
30097 @var{task-id} parameter is present, or about all Ada tasks.
30099 @subsubheading @value{GDBN} Command
30101 The @samp{info tasks} command prints the same information
30102 about all Ada tasks (@pxref{Ada Tasks}).
30104 @subsubheading Result
30106 The result is a table of Ada tasks. The following columns are
30107 defined for each Ada task:
30111 This field exists only for the current thread. It has the value @samp{*}.
30114 The identifier that @value{GDBN} uses to refer to the Ada task.
30117 The identifier that the target uses to refer to the Ada task.
30120 The global thread identifier of the thread corresponding to the Ada
30123 This field should always exist, as Ada tasks are always implemented
30124 on top of a thread. But if @value{GDBN} cannot find this corresponding
30125 thread for any reason, the field is omitted.
30128 This field exists only when the task was created by another task.
30129 In this case, it provides the ID of the parent task.
30132 The base priority of the task.
30135 The current state of the task. For a detailed description of the
30136 possible states, see @ref{Ada Tasks}.
30139 The name of the task.
30143 @subsubheading Example
30147 ^done,tasks=@{nr_rows="3",nr_cols="8",
30148 hdr=[@{width="1",alignment="-1",col_name="current",colhdr=""@},
30149 @{width="3",alignment="1",col_name="id",colhdr="ID"@},
30150 @{width="9",alignment="1",col_name="task-id",colhdr="TID"@},
30151 @{width="4",alignment="1",col_name="thread-id",colhdr=""@},
30152 @{width="4",alignment="1",col_name="parent-id",colhdr="P-ID"@},
30153 @{width="3",alignment="1",col_name="priority",colhdr="Pri"@},
30154 @{width="22",alignment="-1",col_name="state",colhdr="State"@},
30155 @{width="1",alignment="2",col_name="name",colhdr="Name"@}],
30156 body=[@{current="*",id="1",task-id=" 644010",thread-id="1",priority="48",
30157 state="Child Termination Wait",name="main_task"@}]@}
30161 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30162 @node GDB/MI Program Execution
30163 @section @sc{gdb/mi} Program Execution
30165 These are the asynchronous commands which generate the out-of-band
30166 record @samp{*stopped}. Currently @value{GDBN} only really executes
30167 asynchronously with remote targets and this interaction is mimicked in
30170 @subheading The @code{-exec-continue} Command
30171 @findex -exec-continue
30173 @subsubheading Synopsis
30176 -exec-continue [--reverse] [--all|--thread-group N]
30179 Resumes the execution of the inferior program, which will continue
30180 to execute until it reaches a debugger stop event. If the
30181 @samp{--reverse} option is specified, execution resumes in reverse until
30182 it reaches a stop event. Stop events may include
30185 breakpoints or watchpoints
30187 signals or exceptions
30189 the end of the process (or its beginning under @samp{--reverse})
30191 the end or beginning of a replay log if one is being used.
30193 In all-stop mode (@pxref{All-Stop
30194 Mode}), may resume only one thread, or all threads, depending on the
30195 value of the @samp{scheduler-locking} variable. If @samp{--all} is
30196 specified, all threads (in all inferiors) will be resumed. The @samp{--all} option is
30197 ignored in all-stop mode. If the @samp{--thread-group} options is
30198 specified, then all threads in that thread group are resumed.
30200 @subsubheading @value{GDBN} Command
30202 The corresponding @value{GDBN} corresponding is @samp{continue}.
30204 @subsubheading Example
30211 *stopped,reason="breakpoint-hit",disp="keep",bkptno="2",frame=@{
30212 func="foo",args=[],file="hello.c",fullname="/home/foo/bar/hello.c",
30213 line="13",arch="i386:x86_64"@}
30218 @subheading The @code{-exec-finish} Command
30219 @findex -exec-finish
30221 @subsubheading Synopsis
30224 -exec-finish [--reverse]
30227 Resumes the execution of the inferior program until the current
30228 function is exited. Displays the results returned by the function.
30229 If the @samp{--reverse} option is specified, resumes the reverse
30230 execution of the inferior program until the point where current
30231 function was called.
30233 @subsubheading @value{GDBN} Command
30235 The corresponding @value{GDBN} command is @samp{finish}.
30237 @subsubheading Example
30239 Function returning @code{void}.
30246 *stopped,reason="function-finished",frame=@{func="main",args=[],
30247 file="hello.c",fullname="/home/foo/bar/hello.c",line="7",arch="i386:x86_64"@}
30251 Function returning other than @code{void}. The name of the internal
30252 @value{GDBN} variable storing the result is printed, together with the
30259 *stopped,reason="function-finished",frame=@{addr="0x000107b0",func="foo",
30260 args=[@{name="a",value="1"],@{name="b",value="9"@}@},
30261 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
30262 arch="i386:x86_64"@},
30263 gdb-result-var="$1",return-value="0"
30268 @subheading The @code{-exec-interrupt} Command
30269 @findex -exec-interrupt
30271 @subsubheading Synopsis
30274 -exec-interrupt [--all|--thread-group N]
30277 Interrupts the background execution of the target. Note how the token
30278 associated with the stop message is the one for the execution command
30279 that has been interrupted. The token for the interrupt itself only
30280 appears in the @samp{^done} output. If the user is trying to
30281 interrupt a non-running program, an error message will be printed.
30283 Note that when asynchronous execution is enabled, this command is
30284 asynchronous just like other execution commands. That is, first the
30285 @samp{^done} response will be printed, and the target stop will be
30286 reported after that using the @samp{*stopped} notification.
30288 In non-stop mode, only the context thread is interrupted by default.
30289 All threads (in all inferiors) will be interrupted if the
30290 @samp{--all} option is specified. If the @samp{--thread-group}
30291 option is specified, all threads in that group will be interrupted.
30293 @subsubheading @value{GDBN} Command
30295 The corresponding @value{GDBN} command is @samp{interrupt}.
30297 @subsubheading Example
30308 111*stopped,signal-name="SIGINT",signal-meaning="Interrupt",
30309 frame=@{addr="0x00010140",func="foo",args=[],file="try.c",
30310 fullname="/home/foo/bar/try.c",line="13",arch="i386:x86_64"@}
30315 ^error,msg="mi_cmd_exec_interrupt: Inferior not executing."
30319 @subheading The @code{-exec-jump} Command
30322 @subsubheading Synopsis
30325 -exec-jump @var{location}
30328 Resumes execution of the inferior program at the location specified by
30329 parameter. @xref{Specify Location}, for a description of the
30330 different forms of @var{location}.
30332 @subsubheading @value{GDBN} Command
30334 The corresponding @value{GDBN} command is @samp{jump}.
30336 @subsubheading Example
30339 -exec-jump foo.c:10
30340 *running,thread-id="all"
30345 @subheading The @code{-exec-next} Command
30348 @subsubheading Synopsis
30351 -exec-next [--reverse]
30354 Resumes execution of the inferior program, stopping when the beginning
30355 of the next source line is reached.
30357 If the @samp{--reverse} option is specified, resumes reverse execution
30358 of the inferior program, stopping at the beginning of the previous
30359 source line. If you issue this command on the first line of a
30360 function, it will take you back to the caller of that function, to the
30361 source line where the function was called.
30364 @subsubheading @value{GDBN} Command
30366 The corresponding @value{GDBN} command is @samp{next}.
30368 @subsubheading Example
30374 *stopped,reason="end-stepping-range",line="8",file="hello.c"
30379 @subheading The @code{-exec-next-instruction} Command
30380 @findex -exec-next-instruction
30382 @subsubheading Synopsis
30385 -exec-next-instruction [--reverse]
30388 Executes one machine instruction. If the instruction is a function
30389 call, continues until the function returns. If the program stops at an
30390 instruction in the middle of a source line, the address will be
30393 If the @samp{--reverse} option is specified, resumes reverse execution
30394 of the inferior program, stopping at the previous instruction. If the
30395 previously executed instruction was a return from another function,
30396 it will continue to execute in reverse until the call to that function
30397 (from the current stack frame) is reached.
30399 @subsubheading @value{GDBN} Command
30401 The corresponding @value{GDBN} command is @samp{nexti}.
30403 @subsubheading Example
30407 -exec-next-instruction
30411 *stopped,reason="end-stepping-range",
30412 addr="0x000100d4",line="5",file="hello.c"
30417 @subheading The @code{-exec-return} Command
30418 @findex -exec-return
30420 @subsubheading Synopsis
30426 Makes current function return immediately. Doesn't execute the inferior.
30427 Displays the new current frame.
30429 @subsubheading @value{GDBN} Command
30431 The corresponding @value{GDBN} command is @samp{return}.
30433 @subsubheading Example
30437 200-break-insert callee4
30438 200^done,bkpt=@{number="1",addr="0x00010734",
30439 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",line="8"@}
30444 000*stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
30445 frame=@{func="callee4",args=[],
30446 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
30447 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8",
30448 arch="i386:x86_64"@}
30454 111^done,frame=@{level="0",func="callee3",
30455 args=[@{name="strarg",
30456 value="0x11940 \"A string argument.\""@}],
30457 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
30458 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18",
30459 arch="i386:x86_64"@}
30464 @subheading The @code{-exec-run} Command
30467 @subsubheading Synopsis
30470 -exec-run [ --all | --thread-group N ] [ --start ]
30473 Starts execution of the inferior from the beginning. The inferior
30474 executes until either a breakpoint is encountered or the program
30475 exits. In the latter case the output will include an exit code, if
30476 the program has exited exceptionally.
30478 When neither the @samp{--all} nor the @samp{--thread-group} option
30479 is specified, the current inferior is started. If the
30480 @samp{--thread-group} option is specified, it should refer to a thread
30481 group of type @samp{process}, and that thread group will be started.
30482 If the @samp{--all} option is specified, then all inferiors will be started.
30484 Using the @samp{--start} option instructs the debugger to stop
30485 the execution at the start of the inferior's main subprogram,
30486 following the same behavior as the @code{start} command
30487 (@pxref{Starting}).
30489 @subsubheading @value{GDBN} Command
30491 The corresponding @value{GDBN} command is @samp{run}.
30493 @subsubheading Examples
30498 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",line="4"@}
30503 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
30504 frame=@{func="main",args=[],file="recursive2.c",
30505 fullname="/home/foo/bar/recursive2.c",line="4",arch="i386:x86_64"@}
30510 Program exited normally:
30518 *stopped,reason="exited-normally"
30523 Program exited exceptionally:
30531 *stopped,reason="exited",exit-code="01"
30535 Another way the program can terminate is if it receives a signal such as
30536 @code{SIGINT}. In this case, @sc{gdb/mi} displays this:
30540 *stopped,reason="exited-signalled",signal-name="SIGINT",
30541 signal-meaning="Interrupt"
30545 @c @subheading -exec-signal
30548 @subheading The @code{-exec-step} Command
30551 @subsubheading Synopsis
30554 -exec-step [--reverse]
30557 Resumes execution of the inferior program, stopping when the beginning
30558 of the next source line is reached, if the next source line is not a
30559 function call. If it is, stop at the first instruction of the called
30560 function. If the @samp{--reverse} option is specified, resumes reverse
30561 execution of the inferior program, stopping at the beginning of the
30562 previously executed source line.
30564 @subsubheading @value{GDBN} Command
30566 The corresponding @value{GDBN} command is @samp{step}.
30568 @subsubheading Example
30570 Stepping into a function:
30576 *stopped,reason="end-stepping-range",
30577 frame=@{func="foo",args=[@{name="a",value="10"@},
30578 @{name="b",value="0"@}],file="recursive2.c",
30579 fullname="/home/foo/bar/recursive2.c",line="11",arch="i386:x86_64"@}
30589 *stopped,reason="end-stepping-range",line="14",file="recursive2.c"
30594 @subheading The @code{-exec-step-instruction} Command
30595 @findex -exec-step-instruction
30597 @subsubheading Synopsis
30600 -exec-step-instruction [--reverse]
30603 Resumes the inferior which executes one machine instruction. If the
30604 @samp{--reverse} option is specified, resumes reverse execution of the
30605 inferior program, stopping at the previously executed instruction.
30606 The output, once @value{GDBN} has stopped, will vary depending on
30607 whether we have stopped in the middle of a source line or not. In the
30608 former case, the address at which the program stopped will be printed
30611 @subsubheading @value{GDBN} Command
30613 The corresponding @value{GDBN} command is @samp{stepi}.
30615 @subsubheading Example
30619 -exec-step-instruction
30623 *stopped,reason="end-stepping-range",
30624 frame=@{func="foo",args=[],file="try.c",
30625 fullname="/home/foo/bar/try.c",line="10",arch="i386:x86_64"@}
30627 -exec-step-instruction
30631 *stopped,reason="end-stepping-range",
30632 frame=@{addr="0x000100f4",func="foo",args=[],file="try.c",
30633 fullname="/home/foo/bar/try.c",line="10",arch="i386:x86_64"@}
30638 @subheading The @code{-exec-until} Command
30639 @findex -exec-until
30641 @subsubheading Synopsis
30644 -exec-until [ @var{location} ]
30647 Executes the inferior until the @var{location} specified in the
30648 argument is reached. If there is no argument, the inferior executes
30649 until a source line greater than the current one is reached. The
30650 reason for stopping in this case will be @samp{location-reached}.
30652 @subsubheading @value{GDBN} Command
30654 The corresponding @value{GDBN} command is @samp{until}.
30656 @subsubheading Example
30660 -exec-until recursive2.c:6
30664 *stopped,reason="location-reached",frame=@{func="main",args=[],
30665 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="6",
30666 arch="i386:x86_64"@}
30671 @subheading -file-clear
30672 Is this going away????
30675 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30676 @node GDB/MI Stack Manipulation
30677 @section @sc{gdb/mi} Stack Manipulation Commands
30679 @subheading The @code{-enable-frame-filters} Command
30680 @findex -enable-frame-filters
30683 -enable-frame-filters
30686 @value{GDBN} allows Python-based frame filters to affect the output of
30687 the MI commands relating to stack traces. As there is no way to
30688 implement this in a fully backward-compatible way, a front end must
30689 request that this functionality be enabled.
30691 Once enabled, this feature cannot be disabled.
30693 Note that if Python support has not been compiled into @value{GDBN},
30694 this command will still succeed (and do nothing).
30696 @subheading The @code{-stack-info-frame} Command
30697 @findex -stack-info-frame
30699 @subsubheading Synopsis
30705 Get info on the selected frame.
30707 @subsubheading @value{GDBN} Command
30709 The corresponding @value{GDBN} command is @samp{info frame} or @samp{frame}
30710 (without arguments).
30712 @subsubheading Example
30717 ^done,frame=@{level="1",addr="0x0001076c",func="callee3",
30718 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
30719 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17",
30720 arch="i386:x86_64"@}
30724 @subheading The @code{-stack-info-depth} Command
30725 @findex -stack-info-depth
30727 @subsubheading Synopsis
30730 -stack-info-depth [ @var{max-depth} ]
30733 Return the depth of the stack. If the integer argument @var{max-depth}
30734 is specified, do not count beyond @var{max-depth} frames.
30736 @subsubheading @value{GDBN} Command
30738 There's no equivalent @value{GDBN} command.
30740 @subsubheading Example
30742 For a stack with frame levels 0 through 11:
30749 -stack-info-depth 4
30752 -stack-info-depth 12
30755 -stack-info-depth 11
30758 -stack-info-depth 13
30763 @anchor{-stack-list-arguments}
30764 @subheading The @code{-stack-list-arguments} Command
30765 @findex -stack-list-arguments
30767 @subsubheading Synopsis
30770 -stack-list-arguments [ --no-frame-filters ] [ --skip-unavailable ] @var{print-values}
30771 [ @var{low-frame} @var{high-frame} ]
30774 Display a list of the arguments for the frames between @var{low-frame}
30775 and @var{high-frame} (inclusive). If @var{low-frame} and
30776 @var{high-frame} are not provided, list the arguments for the whole
30777 call stack. If the two arguments are equal, show the single frame
30778 at the corresponding level. It is an error if @var{low-frame} is
30779 larger than the actual number of frames. On the other hand,
30780 @var{high-frame} may be larger than the actual number of frames, in
30781 which case only existing frames will be returned.
30783 If @var{print-values} is 0 or @code{--no-values}, print only the names of
30784 the variables; if it is 1 or @code{--all-values}, print also their
30785 values; and if it is 2 or @code{--simple-values}, print the name,
30786 type and value for simple data types, and the name and type for arrays,
30787 structures and unions. If the option @code{--no-frame-filters} is
30788 supplied, then Python frame filters will not be executed.
30790 If the @code{--skip-unavailable} option is specified, arguments that
30791 are not available are not listed. Partially available arguments
30792 are still displayed, however.
30794 Use of this command to obtain arguments in a single frame is
30795 deprecated in favor of the @samp{-stack-list-variables} command.
30797 @subsubheading @value{GDBN} Command
30799 @value{GDBN} does not have an equivalent command. @code{gdbtk} has a
30800 @samp{gdb_get_args} command which partially overlaps with the
30801 functionality of @samp{-stack-list-arguments}.
30803 @subsubheading Example
30810 frame=@{level="0",addr="0x00010734",func="callee4",
30811 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
30812 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8",
30813 arch="i386:x86_64"@},
30814 frame=@{level="1",addr="0x0001076c",func="callee3",
30815 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
30816 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17",
30817 arch="i386:x86_64"@},
30818 frame=@{level="2",addr="0x0001078c",func="callee2",
30819 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
30820 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="22",
30821 arch="i386:x86_64"@},
30822 frame=@{level="3",addr="0x000107b4",func="callee1",
30823 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
30824 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="27",
30825 arch="i386:x86_64"@},
30826 frame=@{level="4",addr="0x000107e0",func="main",
30827 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
30828 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="32",
30829 arch="i386:x86_64"@}]
30831 -stack-list-arguments 0
30834 frame=@{level="0",args=[]@},
30835 frame=@{level="1",args=[name="strarg"]@},
30836 frame=@{level="2",args=[name="intarg",name="strarg"]@},
30837 frame=@{level="3",args=[name="intarg",name="strarg",name="fltarg"]@},
30838 frame=@{level="4",args=[]@}]
30840 -stack-list-arguments 1
30843 frame=@{level="0",args=[]@},
30845 args=[@{name="strarg",value="0x11940 \"A string argument.\""@}]@},
30846 frame=@{level="2",args=[
30847 @{name="intarg",value="2"@},
30848 @{name="strarg",value="0x11940 \"A string argument.\""@}]@},
30849 @{frame=@{level="3",args=[
30850 @{name="intarg",value="2"@},
30851 @{name="strarg",value="0x11940 \"A string argument.\""@},
30852 @{name="fltarg",value="3.5"@}]@},
30853 frame=@{level="4",args=[]@}]
30855 -stack-list-arguments 0 2 2
30856 ^done,stack-args=[frame=@{level="2",args=[name="intarg",name="strarg"]@}]
30858 -stack-list-arguments 1 2 2
30859 ^done,stack-args=[frame=@{level="2",
30860 args=[@{name="intarg",value="2"@},
30861 @{name="strarg",value="0x11940 \"A string argument.\""@}]@}]
30865 @c @subheading -stack-list-exception-handlers
30868 @anchor{-stack-list-frames}
30869 @subheading The @code{-stack-list-frames} Command
30870 @findex -stack-list-frames
30872 @subsubheading Synopsis
30875 -stack-list-frames [ --no-frame-filters @var{low-frame} @var{high-frame} ]
30878 List the frames currently on the stack. For each frame it displays the
30883 The frame number, 0 being the topmost frame, i.e., the innermost function.
30885 The @code{$pc} value for that frame.
30889 File name of the source file where the function lives.
30890 @item @var{fullname}
30891 The full file name of the source file where the function lives.
30893 Line number corresponding to the @code{$pc}.
30895 The shared library where this function is defined. This is only given
30896 if the frame's function is not known.
30898 Frame's architecture.
30901 If invoked without arguments, this command prints a backtrace for the
30902 whole stack. If given two integer arguments, it shows the frames whose
30903 levels are between the two arguments (inclusive). If the two arguments
30904 are equal, it shows the single frame at the corresponding level. It is
30905 an error if @var{low-frame} is larger than the actual number of
30906 frames. On the other hand, @var{high-frame} may be larger than the
30907 actual number of frames, in which case only existing frames will be
30908 returned. If the option @code{--no-frame-filters} is supplied, then
30909 Python frame filters will not be executed.
30911 @subsubheading @value{GDBN} Command
30913 The corresponding @value{GDBN} commands are @samp{backtrace} and @samp{where}.
30915 @subsubheading Example
30917 Full stack backtrace:
30923 [frame=@{level="0",addr="0x0001076c",func="foo",
30924 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="11",
30925 arch="i386:x86_64"@},
30926 frame=@{level="1",addr="0x000107a4",func="foo",
30927 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
30928 arch="i386:x86_64"@},
30929 frame=@{level="2",addr="0x000107a4",func="foo",
30930 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
30931 arch="i386:x86_64"@},
30932 frame=@{level="3",addr="0x000107a4",func="foo",
30933 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
30934 arch="i386:x86_64"@},
30935 frame=@{level="4",addr="0x000107a4",func="foo",
30936 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
30937 arch="i386:x86_64"@},
30938 frame=@{level="5",addr="0x000107a4",func="foo",
30939 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
30940 arch="i386:x86_64"@},
30941 frame=@{level="6",addr="0x000107a4",func="foo",
30942 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
30943 arch="i386:x86_64"@},
30944 frame=@{level="7",addr="0x000107a4",func="foo",
30945 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
30946 arch="i386:x86_64"@},
30947 frame=@{level="8",addr="0x000107a4",func="foo",
30948 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
30949 arch="i386:x86_64"@},
30950 frame=@{level="9",addr="0x000107a4",func="foo",
30951 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
30952 arch="i386:x86_64"@},
30953 frame=@{level="10",addr="0x000107a4",func="foo",
30954 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
30955 arch="i386:x86_64"@},
30956 frame=@{level="11",addr="0x00010738",func="main",
30957 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="4",
30958 arch="i386:x86_64"@}]
30962 Show frames between @var{low_frame} and @var{high_frame}:
30966 -stack-list-frames 3 5
30968 [frame=@{level="3",addr="0x000107a4",func="foo",
30969 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
30970 arch="i386:x86_64"@},
30971 frame=@{level="4",addr="0x000107a4",func="foo",
30972 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
30973 arch="i386:x86_64"@},
30974 frame=@{level="5",addr="0x000107a4",func="foo",
30975 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
30976 arch="i386:x86_64"@}]
30980 Show a single frame:
30984 -stack-list-frames 3 3
30986 [frame=@{level="3",addr="0x000107a4",func="foo",
30987 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
30988 arch="i386:x86_64"@}]
30993 @subheading The @code{-stack-list-locals} Command
30994 @findex -stack-list-locals
30995 @anchor{-stack-list-locals}
30997 @subsubheading Synopsis
31000 -stack-list-locals [ --no-frame-filters ] [ --skip-unavailable ] @var{print-values}
31003 Display the local variable names for the selected frame. If
31004 @var{print-values} is 0 or @code{--no-values}, print only the names of
31005 the variables; if it is 1 or @code{--all-values}, print also their
31006 values; and if it is 2 or @code{--simple-values}, print the name,
31007 type and value for simple data types, and the name and type for arrays,
31008 structures and unions. In this last case, a frontend can immediately
31009 display the value of simple data types and create variable objects for
31010 other data types when the user wishes to explore their values in
31011 more detail. If the option @code{--no-frame-filters} is supplied, then
31012 Python frame filters will not be executed.
31014 If the @code{--skip-unavailable} option is specified, local variables
31015 that are not available are not listed. Partially available local
31016 variables are still displayed, however.
31018 This command is deprecated in favor of the
31019 @samp{-stack-list-variables} command.
31021 @subsubheading @value{GDBN} Command
31023 @samp{info locals} in @value{GDBN}, @samp{gdb_get_locals} in @code{gdbtk}.
31025 @subsubheading Example
31029 -stack-list-locals 0
31030 ^done,locals=[name="A",name="B",name="C"]
31032 -stack-list-locals --all-values
31033 ^done,locals=[@{name="A",value="1"@},@{name="B",value="2"@},
31034 @{name="C",value="@{1, 2, 3@}"@}]
31035 -stack-list-locals --simple-values
31036 ^done,locals=[@{name="A",type="int",value="1"@},
31037 @{name="B",type="int",value="2"@},@{name="C",type="int [3]"@}]
31041 @anchor{-stack-list-variables}
31042 @subheading The @code{-stack-list-variables} Command
31043 @findex -stack-list-variables
31045 @subsubheading Synopsis
31048 -stack-list-variables [ --no-frame-filters ] [ --skip-unavailable ] @var{print-values}
31051 Display the names of local variables and function arguments for the selected frame. If
31052 @var{print-values} is 0 or @code{--no-values}, print only the names of
31053 the variables; if it is 1 or @code{--all-values}, print also their
31054 values; and if it is 2 or @code{--simple-values}, print the name,
31055 type and value for simple data types, and the name and type for arrays,
31056 structures and unions. If the option @code{--no-frame-filters} is
31057 supplied, then Python frame filters will not be executed.
31059 If the @code{--skip-unavailable} option is specified, local variables
31060 and arguments that are not available are not listed. Partially
31061 available arguments and local variables are still displayed, however.
31063 @subsubheading Example
31067 -stack-list-variables --thread 1 --frame 0 --all-values
31068 ^done,variables=[@{name="x",value="11"@},@{name="s",value="@{a = 1, b = 2@}"@}]
31073 @subheading The @code{-stack-select-frame} Command
31074 @findex -stack-select-frame
31076 @subsubheading Synopsis
31079 -stack-select-frame @var{framenum}
31082 Change the selected frame. Select a different frame @var{framenum} on
31085 This command in deprecated in favor of passing the @samp{--frame}
31086 option to every command.
31088 @subsubheading @value{GDBN} Command
31090 The corresponding @value{GDBN} commands are @samp{frame}, @samp{up},
31091 @samp{down}, @samp{select-frame}, @samp{up-silent}, and @samp{down-silent}.
31093 @subsubheading Example
31097 -stack-select-frame 2
31102 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31103 @node GDB/MI Variable Objects
31104 @section @sc{gdb/mi} Variable Objects
31108 @subheading Motivation for Variable Objects in @sc{gdb/mi}
31110 For the implementation of a variable debugger window (locals, watched
31111 expressions, etc.), we are proposing the adaptation of the existing code
31112 used by @code{Insight}.
31114 The two main reasons for that are:
31118 It has been proven in practice (it is already on its second generation).
31121 It will shorten development time (needless to say how important it is
31125 The original interface was designed to be used by Tcl code, so it was
31126 slightly changed so it could be used through @sc{gdb/mi}. This section
31127 describes the @sc{gdb/mi} operations that will be available and gives some
31128 hints about their use.
31130 @emph{Note}: In addition to the set of operations described here, we
31131 expect the @sc{gui} implementation of a variable window to require, at
31132 least, the following operations:
31135 @item @code{-gdb-show} @code{output-radix}
31136 @item @code{-stack-list-arguments}
31137 @item @code{-stack-list-locals}
31138 @item @code{-stack-select-frame}
31143 @subheading Introduction to Variable Objects
31145 @cindex variable objects in @sc{gdb/mi}
31147 Variable objects are "object-oriented" MI interface for examining and
31148 changing values of expressions. Unlike some other MI interfaces that
31149 work with expressions, variable objects are specifically designed for
31150 simple and efficient presentation in the frontend. A variable object
31151 is identified by string name. When a variable object is created, the
31152 frontend specifies the expression for that variable object. The
31153 expression can be a simple variable, or it can be an arbitrary complex
31154 expression, and can even involve CPU registers. After creating a
31155 variable object, the frontend can invoke other variable object
31156 operations---for example to obtain or change the value of a variable
31157 object, or to change display format.
31159 Variable objects have hierarchical tree structure. Any variable object
31160 that corresponds to a composite type, such as structure in C, has
31161 a number of child variable objects, for example corresponding to each
31162 element of a structure. A child variable object can itself have
31163 children, recursively. Recursion ends when we reach
31164 leaf variable objects, which always have built-in types. Child variable
31165 objects are created only by explicit request, so if a frontend
31166 is not interested in the children of a particular variable object, no
31167 child will be created.
31169 For a leaf variable object it is possible to obtain its value as a
31170 string, or set the value from a string. String value can be also
31171 obtained for a non-leaf variable object, but it's generally a string
31172 that only indicates the type of the object, and does not list its
31173 contents. Assignment to a non-leaf variable object is not allowed.
31175 A frontend does not need to read the values of all variable objects each time
31176 the program stops. Instead, MI provides an update command that lists all
31177 variable objects whose values has changed since the last update
31178 operation. This considerably reduces the amount of data that must
31179 be transferred to the frontend. As noted above, children variable
31180 objects are created on demand, and only leaf variable objects have a
31181 real value. As result, gdb will read target memory only for leaf
31182 variables that frontend has created.
31184 The automatic update is not always desirable. For example, a frontend
31185 might want to keep a value of some expression for future reference,
31186 and never update it. For another example, fetching memory is
31187 relatively slow for embedded targets, so a frontend might want
31188 to disable automatic update for the variables that are either not
31189 visible on the screen, or ``closed''. This is possible using so
31190 called ``frozen variable objects''. Such variable objects are never
31191 implicitly updated.
31193 Variable objects can be either @dfn{fixed} or @dfn{floating}. For the
31194 fixed variable object, the expression is parsed when the variable
31195 object is created, including associating identifiers to specific
31196 variables. The meaning of expression never changes. For a floating
31197 variable object the values of variables whose names appear in the
31198 expressions are re-evaluated every time in the context of the current
31199 frame. Consider this example:
31204 struct work_state state;
31211 If a fixed variable object for the @code{state} variable is created in
31212 this function, and we enter the recursive call, the variable
31213 object will report the value of @code{state} in the top-level
31214 @code{do_work} invocation. On the other hand, a floating variable
31215 object will report the value of @code{state} in the current frame.
31217 If an expression specified when creating a fixed variable object
31218 refers to a local variable, the variable object becomes bound to the
31219 thread and frame in which the variable object is created. When such
31220 variable object is updated, @value{GDBN} makes sure that the
31221 thread/frame combination the variable object is bound to still exists,
31222 and re-evaluates the variable object in context of that thread/frame.
31224 The following is the complete set of @sc{gdb/mi} operations defined to
31225 access this functionality:
31227 @multitable @columnfractions .4 .6
31228 @item @strong{Operation}
31229 @tab @strong{Description}
31231 @item @code{-enable-pretty-printing}
31232 @tab enable Python-based pretty-printing
31233 @item @code{-var-create}
31234 @tab create a variable object
31235 @item @code{-var-delete}
31236 @tab delete the variable object and/or its children
31237 @item @code{-var-set-format}
31238 @tab set the display format of this variable
31239 @item @code{-var-show-format}
31240 @tab show the display format of this variable
31241 @item @code{-var-info-num-children}
31242 @tab tells how many children this object has
31243 @item @code{-var-list-children}
31244 @tab return a list of the object's children
31245 @item @code{-var-info-type}
31246 @tab show the type of this variable object
31247 @item @code{-var-info-expression}
31248 @tab print parent-relative expression that this variable object represents
31249 @item @code{-var-info-path-expression}
31250 @tab print full expression that this variable object represents
31251 @item @code{-var-show-attributes}
31252 @tab is this variable editable? does it exist here?
31253 @item @code{-var-evaluate-expression}
31254 @tab get the value of this variable
31255 @item @code{-var-assign}
31256 @tab set the value of this variable
31257 @item @code{-var-update}
31258 @tab update the variable and its children
31259 @item @code{-var-set-frozen}
31260 @tab set frozeness attribute
31261 @item @code{-var-set-update-range}
31262 @tab set range of children to display on update
31265 In the next subsection we describe each operation in detail and suggest
31266 how it can be used.
31268 @subheading Description And Use of Operations on Variable Objects
31270 @subheading The @code{-enable-pretty-printing} Command
31271 @findex -enable-pretty-printing
31274 -enable-pretty-printing
31277 @value{GDBN} allows Python-based visualizers to affect the output of the
31278 MI variable object commands. However, because there was no way to
31279 implement this in a fully backward-compatible way, a front end must
31280 request that this functionality be enabled.
31282 Once enabled, this feature cannot be disabled.
31284 Note that if Python support has not been compiled into @value{GDBN},
31285 this command will still succeed (and do nothing).
31287 This feature is currently (as of @value{GDBN} 7.0) experimental, and
31288 may work differently in future versions of @value{GDBN}.
31290 @subheading The @code{-var-create} Command
31291 @findex -var-create
31293 @subsubheading Synopsis
31296 -var-create @{@var{name} | "-"@}
31297 @{@var{frame-addr} | "*" | "@@"@} @var{expression}
31300 This operation creates a variable object, which allows the monitoring of
31301 a variable, the result of an expression, a memory cell or a CPU
31304 The @var{name} parameter is the string by which the object can be
31305 referenced. It must be unique. If @samp{-} is specified, the varobj
31306 system will generate a string ``varNNNNNN'' automatically. It will be
31307 unique provided that one does not specify @var{name} of that format.
31308 The command fails if a duplicate name is found.
31310 The frame under which the expression should be evaluated can be
31311 specified by @var{frame-addr}. A @samp{*} indicates that the current
31312 frame should be used. A @samp{@@} indicates that a floating variable
31313 object must be created.
31315 @var{expression} is any expression valid on the current language set (must not
31316 begin with a @samp{*}), or one of the following:
31320 @samp{*@var{addr}}, where @var{addr} is the address of a memory cell
31323 @samp{*@var{addr}-@var{addr}} --- a memory address range (TBD)
31326 @samp{$@var{regname}} --- a CPU register name
31329 @cindex dynamic varobj
31330 A varobj's contents may be provided by a Python-based pretty-printer. In this
31331 case the varobj is known as a @dfn{dynamic varobj}. Dynamic varobjs
31332 have slightly different semantics in some cases. If the
31333 @code{-enable-pretty-printing} command is not sent, then @value{GDBN}
31334 will never create a dynamic varobj. This ensures backward
31335 compatibility for existing clients.
31337 @subsubheading Result
31339 This operation returns attributes of the newly-created varobj. These
31344 The name of the varobj.
31347 The number of children of the varobj. This number is not necessarily
31348 reliable for a dynamic varobj. Instead, you must examine the
31349 @samp{has_more} attribute.
31352 The varobj's scalar value. For a varobj whose type is some sort of
31353 aggregate (e.g., a @code{struct}), or for a dynamic varobj, this value
31354 will not be interesting.
31357 The varobj's type. This is a string representation of the type, as
31358 would be printed by the @value{GDBN} CLI. If @samp{print object}
31359 (@pxref{Print Settings, set print object}) is set to @code{on}, the
31360 @emph{actual} (derived) type of the object is shown rather than the
31361 @emph{declared} one.
31364 If a variable object is bound to a specific thread, then this is the
31365 thread's global identifier.
31368 For a dynamic varobj, this indicates whether there appear to be any
31369 children available. For a non-dynamic varobj, this will be 0.
31372 This attribute will be present and have the value @samp{1} if the
31373 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
31374 then this attribute will not be present.
31377 A dynamic varobj can supply a display hint to the front end. The
31378 value comes directly from the Python pretty-printer object's
31379 @code{display_hint} method. @xref{Pretty Printing API}.
31382 Typical output will look like this:
31385 name="@var{name}",numchild="@var{N}",type="@var{type}",thread-id="@var{M}",
31386 has_more="@var{has_more}"
31390 @subheading The @code{-var-delete} Command
31391 @findex -var-delete
31393 @subsubheading Synopsis
31396 -var-delete [ -c ] @var{name}
31399 Deletes a previously created variable object and all of its children.
31400 With the @samp{-c} option, just deletes the children.
31402 Returns an error if the object @var{name} is not found.
31405 @subheading The @code{-var-set-format} Command
31406 @findex -var-set-format
31408 @subsubheading Synopsis
31411 -var-set-format @var{name} @var{format-spec}
31414 Sets the output format for the value of the object @var{name} to be
31417 @anchor{-var-set-format}
31418 The syntax for the @var{format-spec} is as follows:
31421 @var{format-spec} @expansion{}
31422 @{binary | decimal | hexadecimal | octal | natural | zero-hexadecimal@}
31425 The natural format is the default format choosen automatically
31426 based on the variable type (like decimal for an @code{int}, hex
31427 for pointers, etc.).
31429 The zero-hexadecimal format has a representation similar to hexadecimal
31430 but with padding zeroes to the left of the value. For example, a 32-bit
31431 hexadecimal value of 0x1234 would be represented as 0x00001234 in the
31432 zero-hexadecimal format.
31434 For a variable with children, the format is set only on the
31435 variable itself, and the children are not affected.
31437 @subheading The @code{-var-show-format} Command
31438 @findex -var-show-format
31440 @subsubheading Synopsis
31443 -var-show-format @var{name}
31446 Returns the format used to display the value of the object @var{name}.
31449 @var{format} @expansion{}
31454 @subheading The @code{-var-info-num-children} Command
31455 @findex -var-info-num-children
31457 @subsubheading Synopsis
31460 -var-info-num-children @var{name}
31463 Returns the number of children of a variable object @var{name}:
31469 Note that this number is not completely reliable for a dynamic varobj.
31470 It will return the current number of children, but more children may
31474 @subheading The @code{-var-list-children} Command
31475 @findex -var-list-children
31477 @subsubheading Synopsis
31480 -var-list-children [@var{print-values}] @var{name} [@var{from} @var{to}]
31482 @anchor{-var-list-children}
31484 Return a list of the children of the specified variable object and
31485 create variable objects for them, if they do not already exist. With
31486 a single argument or if @var{print-values} has a value of 0 or
31487 @code{--no-values}, print only the names of the variables; if
31488 @var{print-values} is 1 or @code{--all-values}, also print their
31489 values; and if it is 2 or @code{--simple-values} print the name and
31490 value for simple data types and just the name for arrays, structures
31493 @var{from} and @var{to}, if specified, indicate the range of children
31494 to report. If @var{from} or @var{to} is less than zero, the range is
31495 reset and all children will be reported. Otherwise, children starting
31496 at @var{from} (zero-based) and up to and excluding @var{to} will be
31499 If a child range is requested, it will only affect the current call to
31500 @code{-var-list-children}, but not future calls to @code{-var-update}.
31501 For this, you must instead use @code{-var-set-update-range}. The
31502 intent of this approach is to enable a front end to implement any
31503 update approach it likes; for example, scrolling a view may cause the
31504 front end to request more children with @code{-var-list-children}, and
31505 then the front end could call @code{-var-set-update-range} with a
31506 different range to ensure that future updates are restricted to just
31509 For each child the following results are returned:
31514 Name of the variable object created for this child.
31517 The expression to be shown to the user by the front end to designate this child.
31518 For example this may be the name of a structure member.
31520 For a dynamic varobj, this value cannot be used to form an
31521 expression. There is no way to do this at all with a dynamic varobj.
31523 For C/C@t{++} structures there are several pseudo children returned to
31524 designate access qualifiers. For these pseudo children @var{exp} is
31525 @samp{public}, @samp{private}, or @samp{protected}. In this case the
31526 type and value are not present.
31528 A dynamic varobj will not report the access qualifying
31529 pseudo-children, regardless of the language. This information is not
31530 available at all with a dynamic varobj.
31533 Number of children this child has. For a dynamic varobj, this will be
31537 The type of the child. If @samp{print object}
31538 (@pxref{Print Settings, set print object}) is set to @code{on}, the
31539 @emph{actual} (derived) type of the object is shown rather than the
31540 @emph{declared} one.
31543 If values were requested, this is the value.
31546 If this variable object is associated with a thread, this is the
31547 thread's global thread id. Otherwise this result is not present.
31550 If the variable object is frozen, this variable will be present with a value of 1.
31553 A dynamic varobj can supply a display hint to the front end. The
31554 value comes directly from the Python pretty-printer object's
31555 @code{display_hint} method. @xref{Pretty Printing API}.
31558 This attribute will be present and have the value @samp{1} if the
31559 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
31560 then this attribute will not be present.
31564 The result may have its own attributes:
31568 A dynamic varobj can supply a display hint to the front end. The
31569 value comes directly from the Python pretty-printer object's
31570 @code{display_hint} method. @xref{Pretty Printing API}.
31573 This is an integer attribute which is nonzero if there are children
31574 remaining after the end of the selected range.
31577 @subsubheading Example
31581 -var-list-children n
31582 ^done,numchild=@var{n},children=[child=@{name=@var{name},exp=@var{exp},
31583 numchild=@var{n},type=@var{type}@},@r{(repeats N times)}]
31585 -var-list-children --all-values n
31586 ^done,numchild=@var{n},children=[child=@{name=@var{name},exp=@var{exp},
31587 numchild=@var{n},value=@var{value},type=@var{type}@},@r{(repeats N times)}]
31591 @subheading The @code{-var-info-type} Command
31592 @findex -var-info-type
31594 @subsubheading Synopsis
31597 -var-info-type @var{name}
31600 Returns the type of the specified variable @var{name}. The type is
31601 returned as a string in the same format as it is output by the
31605 type=@var{typename}
31609 @subheading The @code{-var-info-expression} Command
31610 @findex -var-info-expression
31612 @subsubheading Synopsis
31615 -var-info-expression @var{name}
31618 Returns a string that is suitable for presenting this
31619 variable object in user interface. The string is generally
31620 not valid expression in the current language, and cannot be evaluated.
31622 For example, if @code{a} is an array, and variable object
31623 @code{A} was created for @code{a}, then we'll get this output:
31626 (gdb) -var-info-expression A.1
31627 ^done,lang="C",exp="1"
31631 Here, the value of @code{lang} is the language name, which can be
31632 found in @ref{Supported Languages}.
31634 Note that the output of the @code{-var-list-children} command also
31635 includes those expressions, so the @code{-var-info-expression} command
31638 @subheading The @code{-var-info-path-expression} Command
31639 @findex -var-info-path-expression
31641 @subsubheading Synopsis
31644 -var-info-path-expression @var{name}
31647 Returns an expression that can be evaluated in the current
31648 context and will yield the same value that a variable object has.
31649 Compare this with the @code{-var-info-expression} command, which
31650 result can be used only for UI presentation. Typical use of
31651 the @code{-var-info-path-expression} command is creating a
31652 watchpoint from a variable object.
31654 This command is currently not valid for children of a dynamic varobj,
31655 and will give an error when invoked on one.
31657 For example, suppose @code{C} is a C@t{++} class, derived from class
31658 @code{Base}, and that the @code{Base} class has a member called
31659 @code{m_size}. Assume a variable @code{c} is has the type of
31660 @code{C} and a variable object @code{C} was created for variable
31661 @code{c}. Then, we'll get this output:
31663 (gdb) -var-info-path-expression C.Base.public.m_size
31664 ^done,path_expr=((Base)c).m_size)
31667 @subheading The @code{-var-show-attributes} Command
31668 @findex -var-show-attributes
31670 @subsubheading Synopsis
31673 -var-show-attributes @var{name}
31676 List attributes of the specified variable object @var{name}:
31679 status=@var{attr} [ ( ,@var{attr} )* ]
31683 where @var{attr} is @code{@{ @{ editable | noneditable @} | TBD @}}.
31685 @subheading The @code{-var-evaluate-expression} Command
31686 @findex -var-evaluate-expression
31688 @subsubheading Synopsis
31691 -var-evaluate-expression [-f @var{format-spec}] @var{name}
31694 Evaluates the expression that is represented by the specified variable
31695 object and returns its value as a string. The format of the string
31696 can be specified with the @samp{-f} option. The possible values of
31697 this option are the same as for @code{-var-set-format}
31698 (@pxref{-var-set-format}). If the @samp{-f} option is not specified,
31699 the current display format will be used. The current display format
31700 can be changed using the @code{-var-set-format} command.
31706 Note that one must invoke @code{-var-list-children} for a variable
31707 before the value of a child variable can be evaluated.
31709 @subheading The @code{-var-assign} Command
31710 @findex -var-assign
31712 @subsubheading Synopsis
31715 -var-assign @var{name} @var{expression}
31718 Assigns the value of @var{expression} to the variable object specified
31719 by @var{name}. The object must be @samp{editable}. If the variable's
31720 value is altered by the assign, the variable will show up in any
31721 subsequent @code{-var-update} list.
31723 @subsubheading Example
31731 ^done,changelist=[@{name="var1",in_scope="true",type_changed="false"@}]
31735 @subheading The @code{-var-update} Command
31736 @findex -var-update
31738 @subsubheading Synopsis
31741 -var-update [@var{print-values}] @{@var{name} | "*"@}
31744 Reevaluate the expressions corresponding to the variable object
31745 @var{name} and all its direct and indirect children, and return the
31746 list of variable objects whose values have changed; @var{name} must
31747 be a root variable object. Here, ``changed'' means that the result of
31748 @code{-var-evaluate-expression} before and after the
31749 @code{-var-update} is different. If @samp{*} is used as the variable
31750 object names, all existing variable objects are updated, except
31751 for frozen ones (@pxref{-var-set-frozen}). The option
31752 @var{print-values} determines whether both names and values, or just
31753 names are printed. The possible values of this option are the same
31754 as for @code{-var-list-children} (@pxref{-var-list-children}). It is
31755 recommended to use the @samp{--all-values} option, to reduce the
31756 number of MI commands needed on each program stop.
31758 With the @samp{*} parameter, if a variable object is bound to a
31759 currently running thread, it will not be updated, without any
31762 If @code{-var-set-update-range} was previously used on a varobj, then
31763 only the selected range of children will be reported.
31765 @code{-var-update} reports all the changed varobjs in a tuple named
31768 Each item in the change list is itself a tuple holding:
31772 The name of the varobj.
31775 If values were requested for this update, then this field will be
31776 present and will hold the value of the varobj.
31779 @anchor{-var-update}
31780 This field is a string which may take one of three values:
31784 The variable object's current value is valid.
31787 The variable object does not currently hold a valid value but it may
31788 hold one in the future if its associated expression comes back into
31792 The variable object no longer holds a valid value.
31793 This can occur when the executable file being debugged has changed,
31794 either through recompilation or by using the @value{GDBN} @code{file}
31795 command. The front end should normally choose to delete these variable
31799 In the future new values may be added to this list so the front should
31800 be prepared for this possibility. @xref{GDB/MI Development and Front Ends, ,@sc{GDB/MI} Development and Front Ends}.
31803 This is only present if the varobj is still valid. If the type
31804 changed, then this will be the string @samp{true}; otherwise it will
31807 When a varobj's type changes, its children are also likely to have
31808 become incorrect. Therefore, the varobj's children are automatically
31809 deleted when this attribute is @samp{true}. Also, the varobj's update
31810 range, when set using the @code{-var-set-update-range} command, is
31814 If the varobj's type changed, then this field will be present and will
31817 @item new_num_children
31818 For a dynamic varobj, if the number of children changed, or if the
31819 type changed, this will be the new number of children.
31821 The @samp{numchild} field in other varobj responses is generally not
31822 valid for a dynamic varobj -- it will show the number of children that
31823 @value{GDBN} knows about, but because dynamic varobjs lazily
31824 instantiate their children, this will not reflect the number of
31825 children which may be available.
31827 The @samp{new_num_children} attribute only reports changes to the
31828 number of children known by @value{GDBN}. This is the only way to
31829 detect whether an update has removed children (which necessarily can
31830 only happen at the end of the update range).
31833 The display hint, if any.
31836 This is an integer value, which will be 1 if there are more children
31837 available outside the varobj's update range.
31840 This attribute will be present and have the value @samp{1} if the
31841 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
31842 then this attribute will not be present.
31845 If new children were added to a dynamic varobj within the selected
31846 update range (as set by @code{-var-set-update-range}), then they will
31847 be listed in this attribute.
31850 @subsubheading Example
31857 -var-update --all-values var1
31858 ^done,changelist=[@{name="var1",value="3",in_scope="true",
31859 type_changed="false"@}]
31863 @subheading The @code{-var-set-frozen} Command
31864 @findex -var-set-frozen
31865 @anchor{-var-set-frozen}
31867 @subsubheading Synopsis
31870 -var-set-frozen @var{name} @var{flag}
31873 Set the frozenness flag on the variable object @var{name}. The
31874 @var{flag} parameter should be either @samp{1} to make the variable
31875 frozen or @samp{0} to make it unfrozen. If a variable object is
31876 frozen, then neither itself, nor any of its children, are
31877 implicitly updated by @code{-var-update} of
31878 a parent variable or by @code{-var-update *}. Only
31879 @code{-var-update} of the variable itself will update its value and
31880 values of its children. After a variable object is unfrozen, it is
31881 implicitly updated by all subsequent @code{-var-update} operations.
31882 Unfreezing a variable does not update it, only subsequent
31883 @code{-var-update} does.
31885 @subsubheading Example
31889 -var-set-frozen V 1
31894 @subheading The @code{-var-set-update-range} command
31895 @findex -var-set-update-range
31896 @anchor{-var-set-update-range}
31898 @subsubheading Synopsis
31901 -var-set-update-range @var{name} @var{from} @var{to}
31904 Set the range of children to be returned by future invocations of
31905 @code{-var-update}.
31907 @var{from} and @var{to} indicate the range of children to report. If
31908 @var{from} or @var{to} is less than zero, the range is reset and all
31909 children will be reported. Otherwise, children starting at @var{from}
31910 (zero-based) and up to and excluding @var{to} will be reported.
31912 @subsubheading Example
31916 -var-set-update-range V 1 2
31920 @subheading The @code{-var-set-visualizer} command
31921 @findex -var-set-visualizer
31922 @anchor{-var-set-visualizer}
31924 @subsubheading Synopsis
31927 -var-set-visualizer @var{name} @var{visualizer}
31930 Set a visualizer for the variable object @var{name}.
31932 @var{visualizer} is the visualizer to use. The special value
31933 @samp{None} means to disable any visualizer in use.
31935 If not @samp{None}, @var{visualizer} must be a Python expression.
31936 This expression must evaluate to a callable object which accepts a
31937 single argument. @value{GDBN} will call this object with the value of
31938 the varobj @var{name} as an argument (this is done so that the same
31939 Python pretty-printing code can be used for both the CLI and MI).
31940 When called, this object must return an object which conforms to the
31941 pretty-printing interface (@pxref{Pretty Printing API}).
31943 The pre-defined function @code{gdb.default_visualizer} may be used to
31944 select a visualizer by following the built-in process
31945 (@pxref{Selecting Pretty-Printers}). This is done automatically when
31946 a varobj is created, and so ordinarily is not needed.
31948 This feature is only available if Python support is enabled. The MI
31949 command @code{-list-features} (@pxref{GDB/MI Support Commands})
31950 can be used to check this.
31952 @subsubheading Example
31954 Resetting the visualizer:
31958 -var-set-visualizer V None
31962 Reselecting the default (type-based) visualizer:
31966 -var-set-visualizer V gdb.default_visualizer
31970 Suppose @code{SomeClass} is a visualizer class. A lambda expression
31971 can be used to instantiate this class for a varobj:
31975 -var-set-visualizer V "lambda val: SomeClass()"
31979 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31980 @node GDB/MI Data Manipulation
31981 @section @sc{gdb/mi} Data Manipulation
31983 @cindex data manipulation, in @sc{gdb/mi}
31984 @cindex @sc{gdb/mi}, data manipulation
31985 This section describes the @sc{gdb/mi} commands that manipulate data:
31986 examine memory and registers, evaluate expressions, etc.
31988 For details about what an addressable memory unit is,
31989 @pxref{addressable memory unit}.
31991 @c REMOVED FROM THE INTERFACE.
31992 @c @subheading -data-assign
31993 @c Change the value of a program variable. Plenty of side effects.
31994 @c @subsubheading GDB Command
31996 @c @subsubheading Example
31999 @subheading The @code{-data-disassemble} Command
32000 @findex -data-disassemble
32002 @subsubheading Synopsis
32006 [ -s @var{start-addr} -e @var{end-addr} ]
32007 | [ -a @var{addr} ]
32008 | [ -f @var{filename} -l @var{linenum} [ -n @var{lines} ] ]
32016 @item @var{start-addr}
32017 is the beginning address (or @code{$pc})
32018 @item @var{end-addr}
32021 is an address anywhere within (or the name of) the function to
32022 disassemble. If an address is specified, the whole function
32023 surrounding that address will be disassembled. If a name is
32024 specified, the whole function with that name will be disassembled.
32025 @item @var{filename}
32026 is the name of the file to disassemble
32027 @item @var{linenum}
32028 is the line number to disassemble around
32030 is the number of disassembly lines to be produced. If it is -1,
32031 the whole function will be disassembled, in case no @var{end-addr} is
32032 specified. If @var{end-addr} is specified as a non-zero value, and
32033 @var{lines} is lower than the number of disassembly lines between
32034 @var{start-addr} and @var{end-addr}, only @var{lines} lines are
32035 displayed; if @var{lines} is higher than the number of lines between
32036 @var{start-addr} and @var{end-addr}, only the lines up to @var{end-addr}
32041 @item 0 disassembly only
32042 @item 1 mixed source and disassembly (deprecated)
32043 @item 2 disassembly with raw opcodes
32044 @item 3 mixed source and disassembly with raw opcodes (deprecated)
32045 @item 4 mixed source and disassembly
32046 @item 5 mixed source and disassembly with raw opcodes
32049 Modes 1 and 3 are deprecated. The output is ``source centric''
32050 which hasn't proved useful in practice.
32051 @xref{Machine Code}, for a discussion of the difference between
32052 @code{/m} and @code{/s} output of the @code{disassemble} command.
32055 @subsubheading Result
32057 The result of the @code{-data-disassemble} command will be a list named
32058 @samp{asm_insns}, the contents of this list depend on the @var{mode}
32059 used with the @code{-data-disassemble} command.
32061 For modes 0 and 2 the @samp{asm_insns} list contains tuples with the
32066 The address at which this instruction was disassembled.
32069 The name of the function this instruction is within.
32072 The decimal offset in bytes from the start of @samp{func-name}.
32075 The text disassembly for this @samp{address}.
32078 This field is only present for modes 2, 3 and 5. This contains the raw opcode
32079 bytes for the @samp{inst} field.
32083 For modes 1, 3, 4 and 5 the @samp{asm_insns} list contains tuples named
32084 @samp{src_and_asm_line}, each of which has the following fields:
32088 The line number within @samp{file}.
32091 The file name from the compilation unit. This might be an absolute
32092 file name or a relative file name depending on the compile command
32096 Absolute file name of @samp{file}. It is converted to a canonical form
32097 using the source file search path
32098 (@pxref{Source Path, ,Specifying Source Directories})
32099 and after resolving all the symbolic links.
32101 If the source file is not found this field will contain the path as
32102 present in the debug information.
32104 @item line_asm_insn
32105 This is a list of tuples containing the disassembly for @samp{line} in
32106 @samp{file}. The fields of each tuple are the same as for
32107 @code{-data-disassemble} in @var{mode} 0 and 2, so @samp{address},
32108 @samp{func-name}, @samp{offset}, @samp{inst}, and optionally
32113 Note that whatever included in the @samp{inst} field, is not
32114 manipulated directly by @sc{gdb/mi}, i.e., it is not possible to
32117 @subsubheading @value{GDBN} Command
32119 The corresponding @value{GDBN} command is @samp{disassemble}.
32121 @subsubheading Example
32123 Disassemble from the current value of @code{$pc} to @code{$pc + 20}:
32127 -data-disassemble -s $pc -e "$pc + 20" -- 0
32130 @{address="0x000107c0",func-name="main",offset="4",
32131 inst="mov 2, %o0"@},
32132 @{address="0x000107c4",func-name="main",offset="8",
32133 inst="sethi %hi(0x11800), %o2"@},
32134 @{address="0x000107c8",func-name="main",offset="12",
32135 inst="or %o2, 0x140, %o1\t! 0x11940 <_lib_version+8>"@},
32136 @{address="0x000107cc",func-name="main",offset="16",
32137 inst="sethi %hi(0x11800), %o2"@},
32138 @{address="0x000107d0",func-name="main",offset="20",
32139 inst="or %o2, 0x168, %o4\t! 0x11968 <_lib_version+48>"@}]
32143 Disassemble the whole @code{main} function. Line 32 is part of
32147 -data-disassemble -f basics.c -l 32 -- 0
32149 @{address="0x000107bc",func-name="main",offset="0",
32150 inst="save %sp, -112, %sp"@},
32151 @{address="0x000107c0",func-name="main",offset="4",
32152 inst="mov 2, %o0"@},
32153 @{address="0x000107c4",func-name="main",offset="8",
32154 inst="sethi %hi(0x11800), %o2"@},
32156 @{address="0x0001081c",func-name="main",offset="96",inst="ret "@},
32157 @{address="0x00010820",func-name="main",offset="100",inst="restore "@}]
32161 Disassemble 3 instructions from the start of @code{main}:
32165 -data-disassemble -f basics.c -l 32 -n 3 -- 0
32167 @{address="0x000107bc",func-name="main",offset="0",
32168 inst="save %sp, -112, %sp"@},
32169 @{address="0x000107c0",func-name="main",offset="4",
32170 inst="mov 2, %o0"@},
32171 @{address="0x000107c4",func-name="main",offset="8",
32172 inst="sethi %hi(0x11800), %o2"@}]
32176 Disassemble 3 instructions from the start of @code{main} in mixed mode:
32180 -data-disassemble -f basics.c -l 32 -n 3 -- 1
32182 src_and_asm_line=@{line="31",
32183 file="../../../src/gdb/testsuite/gdb.mi/basics.c",
32184 fullname="/absolute/path/to/src/gdb/testsuite/gdb.mi/basics.c",
32185 line_asm_insn=[@{address="0x000107bc",
32186 func-name="main",offset="0",inst="save %sp, -112, %sp"@}]@},
32187 src_and_asm_line=@{line="32",
32188 file="../../../src/gdb/testsuite/gdb.mi/basics.c",
32189 fullname="/absolute/path/to/src/gdb/testsuite/gdb.mi/basics.c",
32190 line_asm_insn=[@{address="0x000107c0",
32191 func-name="main",offset="4",inst="mov 2, %o0"@},
32192 @{address="0x000107c4",func-name="main",offset="8",
32193 inst="sethi %hi(0x11800), %o2"@}]@}]
32198 @subheading The @code{-data-evaluate-expression} Command
32199 @findex -data-evaluate-expression
32201 @subsubheading Synopsis
32204 -data-evaluate-expression @var{expr}
32207 Evaluate @var{expr} as an expression. The expression could contain an
32208 inferior function call. The function call will execute synchronously.
32209 If the expression contains spaces, it must be enclosed in double quotes.
32211 @subsubheading @value{GDBN} Command
32213 The corresponding @value{GDBN} commands are @samp{print}, @samp{output}, and
32214 @samp{call}. In @code{gdbtk} only, there's a corresponding
32215 @samp{gdb_eval} command.
32217 @subsubheading Example
32219 In the following example, the numbers that precede the commands are the
32220 @dfn{tokens} described in @ref{GDB/MI Command Syntax, ,@sc{gdb/mi}
32221 Command Syntax}. Notice how @sc{gdb/mi} returns the same tokens in its
32225 211-data-evaluate-expression A
32228 311-data-evaluate-expression &A
32229 311^done,value="0xefffeb7c"
32231 411-data-evaluate-expression A+3
32234 511-data-evaluate-expression "A + 3"
32240 @subheading The @code{-data-list-changed-registers} Command
32241 @findex -data-list-changed-registers
32243 @subsubheading Synopsis
32246 -data-list-changed-registers
32249 Display a list of the registers that have changed.
32251 @subsubheading @value{GDBN} Command
32253 @value{GDBN} doesn't have a direct analog for this command; @code{gdbtk}
32254 has the corresponding command @samp{gdb_changed_register_list}.
32256 @subsubheading Example
32258 On a PPC MBX board:
32266 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",frame=@{
32267 func="main",args=[],file="try.c",fullname="/home/foo/bar/try.c",
32268 line="5",arch="powerpc"@}
32270 -data-list-changed-registers
32271 ^done,changed-registers=["0","1","2","4","5","6","7","8","9",
32272 "10","11","13","14","15","16","17","18","19","20","21","22","23",
32273 "24","25","26","27","28","30","31","64","65","66","67","69"]
32278 @subheading The @code{-data-list-register-names} Command
32279 @findex -data-list-register-names
32281 @subsubheading Synopsis
32284 -data-list-register-names [ ( @var{regno} )+ ]
32287 Show a list of register names for the current target. If no arguments
32288 are given, it shows a list of the names of all the registers. If
32289 integer numbers are given as arguments, it will print a list of the
32290 names of the registers corresponding to the arguments. To ensure
32291 consistency between a register name and its number, the output list may
32292 include empty register names.
32294 @subsubheading @value{GDBN} Command
32296 @value{GDBN} does not have a command which corresponds to
32297 @samp{-data-list-register-names}. In @code{gdbtk} there is a
32298 corresponding command @samp{gdb_regnames}.
32300 @subsubheading Example
32302 For the PPC MBX board:
32305 -data-list-register-names
32306 ^done,register-names=["r0","r1","r2","r3","r4","r5","r6","r7",
32307 "r8","r9","r10","r11","r12","r13","r14","r15","r16","r17","r18",
32308 "r19","r20","r21","r22","r23","r24","r25","r26","r27","r28","r29",
32309 "r30","r31","f0","f1","f2","f3","f4","f5","f6","f7","f8","f9",
32310 "f10","f11","f12","f13","f14","f15","f16","f17","f18","f19","f20",
32311 "f21","f22","f23","f24","f25","f26","f27","f28","f29","f30","f31",
32312 "", "pc","ps","cr","lr","ctr","xer"]
32314 -data-list-register-names 1 2 3
32315 ^done,register-names=["r1","r2","r3"]
32319 @subheading The @code{-data-list-register-values} Command
32320 @findex -data-list-register-values
32322 @subsubheading Synopsis
32325 -data-list-register-values
32326 [ @code{--skip-unavailable} ] @var{fmt} [ ( @var{regno} )*]
32329 Display the registers' contents. The format according to which the
32330 registers' contents are to be returned is given by @var{fmt}, followed
32331 by an optional list of numbers specifying the registers to display. A
32332 missing list of numbers indicates that the contents of all the
32333 registers must be returned. The @code{--skip-unavailable} option
32334 indicates that only the available registers are to be returned.
32336 Allowed formats for @var{fmt} are:
32353 @subsubheading @value{GDBN} Command
32355 The corresponding @value{GDBN} commands are @samp{info reg}, @samp{info
32356 all-reg}, and (in @code{gdbtk}) @samp{gdb_fetch_registers}.
32358 @subsubheading Example
32360 For a PPC MBX board (note: line breaks are for readability only, they
32361 don't appear in the actual output):
32365 -data-list-register-values r 64 65
32366 ^done,register-values=[@{number="64",value="0xfe00a300"@},
32367 @{number="65",value="0x00029002"@}]
32369 -data-list-register-values x
32370 ^done,register-values=[@{number="0",value="0xfe0043c8"@},
32371 @{number="1",value="0x3fff88"@},@{number="2",value="0xfffffffe"@},
32372 @{number="3",value="0x0"@},@{number="4",value="0xa"@},
32373 @{number="5",value="0x3fff68"@},@{number="6",value="0x3fff58"@},
32374 @{number="7",value="0xfe011e98"@},@{number="8",value="0x2"@},
32375 @{number="9",value="0xfa202820"@},@{number="10",value="0xfa202808"@},
32376 @{number="11",value="0x1"@},@{number="12",value="0x0"@},
32377 @{number="13",value="0x4544"@},@{number="14",value="0xffdfffff"@},
32378 @{number="15",value="0xffffffff"@},@{number="16",value="0xfffffeff"@},
32379 @{number="17",value="0xefffffed"@},@{number="18",value="0xfffffffe"@},
32380 @{number="19",value="0xffffffff"@},@{number="20",value="0xffffffff"@},
32381 @{number="21",value="0xffffffff"@},@{number="22",value="0xfffffff7"@},
32382 @{number="23",value="0xffffffff"@},@{number="24",value="0xffffffff"@},
32383 @{number="25",value="0xffffffff"@},@{number="26",value="0xfffffffb"@},
32384 @{number="27",value="0xffffffff"@},@{number="28",value="0xf7bfffff"@},
32385 @{number="29",value="0x0"@},@{number="30",value="0xfe010000"@},
32386 @{number="31",value="0x0"@},@{number="32",value="0x0"@},
32387 @{number="33",value="0x0"@},@{number="34",value="0x0"@},
32388 @{number="35",value="0x0"@},@{number="36",value="0x0"@},
32389 @{number="37",value="0x0"@},@{number="38",value="0x0"@},
32390 @{number="39",value="0x0"@},@{number="40",value="0x0"@},
32391 @{number="41",value="0x0"@},@{number="42",value="0x0"@},
32392 @{number="43",value="0x0"@},@{number="44",value="0x0"@},
32393 @{number="45",value="0x0"@},@{number="46",value="0x0"@},
32394 @{number="47",value="0x0"@},@{number="48",value="0x0"@},
32395 @{number="49",value="0x0"@},@{number="50",value="0x0"@},
32396 @{number="51",value="0x0"@},@{number="52",value="0x0"@},
32397 @{number="53",value="0x0"@},@{number="54",value="0x0"@},
32398 @{number="55",value="0x0"@},@{number="56",value="0x0"@},
32399 @{number="57",value="0x0"@},@{number="58",value="0x0"@},
32400 @{number="59",value="0x0"@},@{number="60",value="0x0"@},
32401 @{number="61",value="0x0"@},@{number="62",value="0x0"@},
32402 @{number="63",value="0x0"@},@{number="64",value="0xfe00a300"@},
32403 @{number="65",value="0x29002"@},@{number="66",value="0x202f04b5"@},
32404 @{number="67",value="0xfe0043b0"@},@{number="68",value="0xfe00b3e4"@},
32405 @{number="69",value="0x20002b03"@}]
32410 @subheading The @code{-data-read-memory} Command
32411 @findex -data-read-memory
32413 This command is deprecated, use @code{-data-read-memory-bytes} instead.
32415 @subsubheading Synopsis
32418 -data-read-memory [ -o @var{byte-offset} ]
32419 @var{address} @var{word-format} @var{word-size}
32420 @var{nr-rows} @var{nr-cols} [ @var{aschar} ]
32427 @item @var{address}
32428 An expression specifying the address of the first memory word to be
32429 read. Complex expressions containing embedded white space should be
32430 quoted using the C convention.
32432 @item @var{word-format}
32433 The format to be used to print the memory words. The notation is the
32434 same as for @value{GDBN}'s @code{print} command (@pxref{Output Formats,
32437 @item @var{word-size}
32438 The size of each memory word in bytes.
32440 @item @var{nr-rows}
32441 The number of rows in the output table.
32443 @item @var{nr-cols}
32444 The number of columns in the output table.
32447 If present, indicates that each row should include an @sc{ascii} dump. The
32448 value of @var{aschar} is used as a padding character when a byte is not a
32449 member of the printable @sc{ascii} character set (printable @sc{ascii}
32450 characters are those whose code is between 32 and 126, inclusively).
32452 @item @var{byte-offset}
32453 An offset to add to the @var{address} before fetching memory.
32456 This command displays memory contents as a table of @var{nr-rows} by
32457 @var{nr-cols} words, each word being @var{word-size} bytes. In total,
32458 @code{@var{nr-rows} * @var{nr-cols} * @var{word-size}} bytes are read
32459 (returned as @samp{total-bytes}). Should less than the requested number
32460 of bytes be returned by the target, the missing words are identified
32461 using @samp{N/A}. The number of bytes read from the target is returned
32462 in @samp{nr-bytes} and the starting address used to read memory in
32465 The address of the next/previous row or page is available in
32466 @samp{next-row} and @samp{prev-row}, @samp{next-page} and
32469 @subsubheading @value{GDBN} Command
32471 The corresponding @value{GDBN} command is @samp{x}. @code{gdbtk} has
32472 @samp{gdb_get_mem} memory read command.
32474 @subsubheading Example
32476 Read six bytes of memory starting at @code{bytes+6} but then offset by
32477 @code{-6} bytes. Format as three rows of two columns. One byte per
32478 word. Display each word in hex.
32482 9-data-read-memory -o -6 -- bytes+6 x 1 3 2
32483 9^done,addr="0x00001390",nr-bytes="6",total-bytes="6",
32484 next-row="0x00001396",prev-row="0x0000138e",next-page="0x00001396",
32485 prev-page="0x0000138a",memory=[
32486 @{addr="0x00001390",data=["0x00","0x01"]@},
32487 @{addr="0x00001392",data=["0x02","0x03"]@},
32488 @{addr="0x00001394",data=["0x04","0x05"]@}]
32492 Read two bytes of memory starting at address @code{shorts + 64} and
32493 display as a single word formatted in decimal.
32497 5-data-read-memory shorts+64 d 2 1 1
32498 5^done,addr="0x00001510",nr-bytes="2",total-bytes="2",
32499 next-row="0x00001512",prev-row="0x0000150e",
32500 next-page="0x00001512",prev-page="0x0000150e",memory=[
32501 @{addr="0x00001510",data=["128"]@}]
32505 Read thirty two bytes of memory starting at @code{bytes+16} and format
32506 as eight rows of four columns. Include a string encoding with @samp{x}
32507 used as the non-printable character.
32511 4-data-read-memory bytes+16 x 1 8 4 x
32512 4^done,addr="0x000013a0",nr-bytes="32",total-bytes="32",
32513 next-row="0x000013c0",prev-row="0x0000139c",
32514 next-page="0x000013c0",prev-page="0x00001380",memory=[
32515 @{addr="0x000013a0",data=["0x10","0x11","0x12","0x13"],ascii="xxxx"@},
32516 @{addr="0x000013a4",data=["0x14","0x15","0x16","0x17"],ascii="xxxx"@},
32517 @{addr="0x000013a8",data=["0x18","0x19","0x1a","0x1b"],ascii="xxxx"@},
32518 @{addr="0x000013ac",data=["0x1c","0x1d","0x1e","0x1f"],ascii="xxxx"@},
32519 @{addr="0x000013b0",data=["0x20","0x21","0x22","0x23"],ascii=" !\"#"@},
32520 @{addr="0x000013b4",data=["0x24","0x25","0x26","0x27"],ascii="$%&'"@},
32521 @{addr="0x000013b8",data=["0x28","0x29","0x2a","0x2b"],ascii="()*+"@},
32522 @{addr="0x000013bc",data=["0x2c","0x2d","0x2e","0x2f"],ascii=",-./"@}]
32526 @subheading The @code{-data-read-memory-bytes} Command
32527 @findex -data-read-memory-bytes
32529 @subsubheading Synopsis
32532 -data-read-memory-bytes [ -o @var{offset} ]
32533 @var{address} @var{count}
32540 @item @var{address}
32541 An expression specifying the address of the first addressable memory unit
32542 to be read. Complex expressions containing embedded white space should be
32543 quoted using the C convention.
32546 The number of addressable memory units to read. This should be an integer
32550 The offset relative to @var{address} at which to start reading. This
32551 should be an integer literal. This option is provided so that a frontend
32552 is not required to first evaluate address and then perform address
32553 arithmetics itself.
32557 This command attempts to read all accessible memory regions in the
32558 specified range. First, all regions marked as unreadable in the memory
32559 map (if one is defined) will be skipped. @xref{Memory Region
32560 Attributes}. Second, @value{GDBN} will attempt to read the remaining
32561 regions. For each one, if reading full region results in an errors,
32562 @value{GDBN} will try to read a subset of the region.
32564 In general, every single memory unit in the region may be readable or not,
32565 and the only way to read every readable unit is to try a read at
32566 every address, which is not practical. Therefore, @value{GDBN} will
32567 attempt to read all accessible memory units at either beginning or the end
32568 of the region, using a binary division scheme. This heuristic works
32569 well for reading accross a memory map boundary. Note that if a region
32570 has a readable range that is neither at the beginning or the end,
32571 @value{GDBN} will not read it.
32573 The result record (@pxref{GDB/MI Result Records}) that is output of
32574 the command includes a field named @samp{memory} whose content is a
32575 list of tuples. Each tuple represent a successfully read memory block
32576 and has the following fields:
32580 The start address of the memory block, as hexadecimal literal.
32583 The end address of the memory block, as hexadecimal literal.
32586 The offset of the memory block, as hexadecimal literal, relative to
32587 the start address passed to @code{-data-read-memory-bytes}.
32590 The contents of the memory block, in hex.
32596 @subsubheading @value{GDBN} Command
32598 The corresponding @value{GDBN} command is @samp{x}.
32600 @subsubheading Example
32604 -data-read-memory-bytes &a 10
32605 ^done,memory=[@{begin="0xbffff154",offset="0x00000000",
32607 contents="01000000020000000300"@}]
32612 @subheading The @code{-data-write-memory-bytes} Command
32613 @findex -data-write-memory-bytes
32615 @subsubheading Synopsis
32618 -data-write-memory-bytes @var{address} @var{contents}
32619 -data-write-memory-bytes @var{address} @var{contents} @r{[}@var{count}@r{]}
32626 @item @var{address}
32627 An expression specifying the address of the first addressable memory unit
32628 to be written. Complex expressions containing embedded white space should
32629 be quoted using the C convention.
32631 @item @var{contents}
32632 The hex-encoded data to write. It is an error if @var{contents} does
32633 not represent an integral number of addressable memory units.
32636 Optional argument indicating the number of addressable memory units to be
32637 written. If @var{count} is greater than @var{contents}' length,
32638 @value{GDBN} will repeatedly write @var{contents} until it fills
32639 @var{count} memory units.
32643 @subsubheading @value{GDBN} Command
32645 There's no corresponding @value{GDBN} command.
32647 @subsubheading Example
32651 -data-write-memory-bytes &a "aabbccdd"
32658 -data-write-memory-bytes &a "aabbccdd" 16e
32663 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
32664 @node GDB/MI Tracepoint Commands
32665 @section @sc{gdb/mi} Tracepoint Commands
32667 The commands defined in this section implement MI support for
32668 tracepoints. For detailed introduction, see @ref{Tracepoints}.
32670 @subheading The @code{-trace-find} Command
32671 @findex -trace-find
32673 @subsubheading Synopsis
32676 -trace-find @var{mode} [@var{parameters}@dots{}]
32679 Find a trace frame using criteria defined by @var{mode} and
32680 @var{parameters}. The following table lists permissible
32681 modes and their parameters. For details of operation, see @ref{tfind}.
32686 No parameters are required. Stops examining trace frames.
32689 An integer is required as parameter. Selects tracepoint frame with
32692 @item tracepoint-number
32693 An integer is required as parameter. Finds next
32694 trace frame that corresponds to tracepoint with the specified number.
32697 An address is required as parameter. Finds
32698 next trace frame that corresponds to any tracepoint at the specified
32701 @item pc-inside-range
32702 Two addresses are required as parameters. Finds next trace
32703 frame that corresponds to a tracepoint at an address inside the
32704 specified range. Both bounds are considered to be inside the range.
32706 @item pc-outside-range
32707 Two addresses are required as parameters. Finds
32708 next trace frame that corresponds to a tracepoint at an address outside
32709 the specified range. Both bounds are considered to be inside the range.
32712 Line specification is required as parameter. @xref{Specify Location}.
32713 Finds next trace frame that corresponds to a tracepoint at
32714 the specified location.
32718 If @samp{none} was passed as @var{mode}, the response does not
32719 have fields. Otherwise, the response may have the following fields:
32723 This field has either @samp{0} or @samp{1} as the value, depending
32724 on whether a matching tracepoint was found.
32727 The index of the found traceframe. This field is present iff
32728 the @samp{found} field has value of @samp{1}.
32731 The index of the found tracepoint. This field is present iff
32732 the @samp{found} field has value of @samp{1}.
32735 The information about the frame corresponding to the found trace
32736 frame. This field is present only if a trace frame was found.
32737 @xref{GDB/MI Frame Information}, for description of this field.
32741 @subsubheading @value{GDBN} Command
32743 The corresponding @value{GDBN} command is @samp{tfind}.
32745 @subheading -trace-define-variable
32746 @findex -trace-define-variable
32748 @subsubheading Synopsis
32751 -trace-define-variable @var{name} [ @var{value} ]
32754 Create trace variable @var{name} if it does not exist. If
32755 @var{value} is specified, sets the initial value of the specified
32756 trace variable to that value. Note that the @var{name} should start
32757 with the @samp{$} character.
32759 @subsubheading @value{GDBN} Command
32761 The corresponding @value{GDBN} command is @samp{tvariable}.
32763 @subheading The @code{-trace-frame-collected} Command
32764 @findex -trace-frame-collected
32766 @subsubheading Synopsis
32769 -trace-frame-collected
32770 [--var-print-values @var{var_pval}]
32771 [--comp-print-values @var{comp_pval}]
32772 [--registers-format @var{regformat}]
32773 [--memory-contents]
32776 This command returns the set of collected objects, register names,
32777 trace state variable names, memory ranges and computed expressions
32778 that have been collected at a particular trace frame. The optional
32779 parameters to the command affect the output format in different ways.
32780 See the output description table below for more details.
32782 The reported names can be used in the normal manner to create
32783 varobjs and inspect the objects themselves. The items returned by
32784 this command are categorized so that it is clear which is a variable,
32785 which is a register, which is a trace state variable, which is a
32786 memory range and which is a computed expression.
32788 For instance, if the actions were
32790 collect myVar, myArray[myIndex], myObj.field, myPtr->field, myCount + 2
32791 collect *(int*)0xaf02bef0@@40
32795 the object collected in its entirety would be @code{myVar}. The
32796 object @code{myArray} would be partially collected, because only the
32797 element at index @code{myIndex} would be collected. The remaining
32798 objects would be computed expressions.
32800 An example output would be:
32804 -trace-frame-collected
32806 explicit-variables=[@{name="myVar",value="1"@}],
32807 computed-expressions=[@{name="myArray[myIndex]",value="0"@},
32808 @{name="myObj.field",value="0"@},
32809 @{name="myPtr->field",value="1"@},
32810 @{name="myCount + 2",value="3"@},
32811 @{name="$tvar1 + 1",value="43970027"@}],
32812 registers=[@{number="0",value="0x7fe2c6e79ec8"@},
32813 @{number="1",value="0x0"@},
32814 @{number="2",value="0x4"@},
32816 @{number="125",value="0x0"@}],
32817 tvars=[@{name="$tvar1",current="43970026"@}],
32818 memory=[@{address="0x0000000000602264",length="4"@},
32819 @{address="0x0000000000615bc0",length="4"@}]
32826 @item explicit-variables
32827 The set of objects that have been collected in their entirety (as
32828 opposed to collecting just a few elements of an array or a few struct
32829 members). For each object, its name and value are printed.
32830 The @code{--var-print-values} option affects how or whether the value
32831 field is output. If @var{var_pval} is 0, then print only the names;
32832 if it is 1, print also their values; and if it is 2, print the name,
32833 type and value for simple data types, and the name and type for
32834 arrays, structures and unions.
32836 @item computed-expressions
32837 The set of computed expressions that have been collected at the
32838 current trace frame. The @code{--comp-print-values} option affects
32839 this set like the @code{--var-print-values} option affects the
32840 @code{explicit-variables} set. See above.
32843 The registers that have been collected at the current trace frame.
32844 For each register collected, the name and current value are returned.
32845 The value is formatted according to the @code{--registers-format}
32846 option. See the @command{-data-list-register-values} command for a
32847 list of the allowed formats. The default is @samp{x}.
32850 The trace state variables that have been collected at the current
32851 trace frame. For each trace state variable collected, the name and
32852 current value are returned.
32855 The set of memory ranges that have been collected at the current trace
32856 frame. Its content is a list of tuples. Each tuple represents a
32857 collected memory range and has the following fields:
32861 The start address of the memory range, as hexadecimal literal.
32864 The length of the memory range, as decimal literal.
32867 The contents of the memory block, in hex. This field is only present
32868 if the @code{--memory-contents} option is specified.
32874 @subsubheading @value{GDBN} Command
32876 There is no corresponding @value{GDBN} command.
32878 @subsubheading Example
32880 @subheading -trace-list-variables
32881 @findex -trace-list-variables
32883 @subsubheading Synopsis
32886 -trace-list-variables
32889 Return a table of all defined trace variables. Each element of the
32890 table has the following fields:
32894 The name of the trace variable. This field is always present.
32897 The initial value. This is a 64-bit signed integer. This
32898 field is always present.
32901 The value the trace variable has at the moment. This is a 64-bit
32902 signed integer. This field is absent iff current value is
32903 not defined, for example if the trace was never run, or is
32908 @subsubheading @value{GDBN} Command
32910 The corresponding @value{GDBN} command is @samp{tvariables}.
32912 @subsubheading Example
32916 -trace-list-variables
32917 ^done,trace-variables=@{nr_rows="1",nr_cols="3",
32918 hdr=[@{width="15",alignment="-1",col_name="name",colhdr="Name"@},
32919 @{width="11",alignment="-1",col_name="initial",colhdr="Initial"@},
32920 @{width="11",alignment="-1",col_name="current",colhdr="Current"@}],
32921 body=[variable=@{name="$trace_timestamp",initial="0"@}
32922 variable=@{name="$foo",initial="10",current="15"@}]@}
32926 @subheading -trace-save
32927 @findex -trace-save
32929 @subsubheading Synopsis
32932 -trace-save [ -r ] [ -ctf ] @var{filename}
32935 Saves the collected trace data to @var{filename}. Without the
32936 @samp{-r} option, the data is downloaded from the target and saved
32937 in a local file. With the @samp{-r} option the target is asked
32938 to perform the save.
32940 By default, this command will save the trace in the tfile format. You can
32941 supply the optional @samp{-ctf} argument to save it the CTF format. See
32942 @ref{Trace Files} for more information about CTF.
32944 @subsubheading @value{GDBN} Command
32946 The corresponding @value{GDBN} command is @samp{tsave}.
32949 @subheading -trace-start
32950 @findex -trace-start
32952 @subsubheading Synopsis
32958 Starts a tracing experiment. The result of this command does not
32961 @subsubheading @value{GDBN} Command
32963 The corresponding @value{GDBN} command is @samp{tstart}.
32965 @subheading -trace-status
32966 @findex -trace-status
32968 @subsubheading Synopsis
32974 Obtains the status of a tracing experiment. The result may include
32975 the following fields:
32980 May have a value of either @samp{0}, when no tracing operations are
32981 supported, @samp{1}, when all tracing operations are supported, or
32982 @samp{file} when examining trace file. In the latter case, examining
32983 of trace frame is possible but new tracing experiement cannot be
32984 started. This field is always present.
32987 May have a value of either @samp{0} or @samp{1} depending on whether
32988 tracing experiement is in progress on target. This field is present
32989 if @samp{supported} field is not @samp{0}.
32992 Report the reason why the tracing was stopped last time. This field
32993 may be absent iff tracing was never stopped on target yet. The
32994 value of @samp{request} means the tracing was stopped as result of
32995 the @code{-trace-stop} command. The value of @samp{overflow} means
32996 the tracing buffer is full. The value of @samp{disconnection} means
32997 tracing was automatically stopped when @value{GDBN} has disconnected.
32998 The value of @samp{passcount} means tracing was stopped when a
32999 tracepoint was passed a maximal number of times for that tracepoint.
33000 This field is present if @samp{supported} field is not @samp{0}.
33002 @item stopping-tracepoint
33003 The number of tracepoint whose passcount as exceeded. This field is
33004 present iff the @samp{stop-reason} field has the value of
33008 @itemx frames-created
33009 The @samp{frames} field is a count of the total number of trace frames
33010 in the trace buffer, while @samp{frames-created} is the total created
33011 during the run, including ones that were discarded, such as when a
33012 circular trace buffer filled up. Both fields are optional.
33016 These fields tell the current size of the tracing buffer and the
33017 remaining space. These fields are optional.
33020 The value of the circular trace buffer flag. @code{1} means that the
33021 trace buffer is circular and old trace frames will be discarded if
33022 necessary to make room, @code{0} means that the trace buffer is linear
33026 The value of the disconnected tracing flag. @code{1} means that
33027 tracing will continue after @value{GDBN} disconnects, @code{0} means
33028 that the trace run will stop.
33031 The filename of the trace file being examined. This field is
33032 optional, and only present when examining a trace file.
33036 @subsubheading @value{GDBN} Command
33038 The corresponding @value{GDBN} command is @samp{tstatus}.
33040 @subheading -trace-stop
33041 @findex -trace-stop
33043 @subsubheading Synopsis
33049 Stops a tracing experiment. The result of this command has the same
33050 fields as @code{-trace-status}, except that the @samp{supported} and
33051 @samp{running} fields are not output.
33053 @subsubheading @value{GDBN} Command
33055 The corresponding @value{GDBN} command is @samp{tstop}.
33058 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
33059 @node GDB/MI Symbol Query
33060 @section @sc{gdb/mi} Symbol Query Commands
33064 @subheading The @code{-symbol-info-address} Command
33065 @findex -symbol-info-address
33067 @subsubheading Synopsis
33070 -symbol-info-address @var{symbol}
33073 Describe where @var{symbol} is stored.
33075 @subsubheading @value{GDBN} Command
33077 The corresponding @value{GDBN} command is @samp{info address}.
33079 @subsubheading Example
33083 @subheading The @code{-symbol-info-file} Command
33084 @findex -symbol-info-file
33086 @subsubheading Synopsis
33092 Show the file for the symbol.
33094 @subsubheading @value{GDBN} Command
33096 There's no equivalent @value{GDBN} command. @code{gdbtk} has
33097 @samp{gdb_find_file}.
33099 @subsubheading Example
33103 @subheading The @code{-symbol-info-function} Command
33104 @findex -symbol-info-function
33106 @subsubheading Synopsis
33109 -symbol-info-function
33112 Show which function the symbol lives in.
33114 @subsubheading @value{GDBN} Command
33116 @samp{gdb_get_function} in @code{gdbtk}.
33118 @subsubheading Example
33122 @subheading The @code{-symbol-info-line} Command
33123 @findex -symbol-info-line
33125 @subsubheading Synopsis
33131 Show the core addresses of the code for a source line.
33133 @subsubheading @value{GDBN} Command
33135 The corresponding @value{GDBN} command is @samp{info line}.
33136 @code{gdbtk} has the @samp{gdb_get_line} and @samp{gdb_get_file} commands.
33138 @subsubheading Example
33142 @subheading The @code{-symbol-info-symbol} Command
33143 @findex -symbol-info-symbol
33145 @subsubheading Synopsis
33148 -symbol-info-symbol @var{addr}
33151 Describe what symbol is at location @var{addr}.
33153 @subsubheading @value{GDBN} Command
33155 The corresponding @value{GDBN} command is @samp{info symbol}.
33157 @subsubheading Example
33161 @subheading The @code{-symbol-list-functions} Command
33162 @findex -symbol-list-functions
33164 @subsubheading Synopsis
33167 -symbol-list-functions
33170 List the functions in the executable.
33172 @subsubheading @value{GDBN} Command
33174 @samp{info functions} in @value{GDBN}, @samp{gdb_listfunc} and
33175 @samp{gdb_search} in @code{gdbtk}.
33177 @subsubheading Example
33182 @subheading The @code{-symbol-list-lines} Command
33183 @findex -symbol-list-lines
33185 @subsubheading Synopsis
33188 -symbol-list-lines @var{filename}
33191 Print the list of lines that contain code and their associated program
33192 addresses for the given source filename. The entries are sorted in
33193 ascending PC order.
33195 @subsubheading @value{GDBN} Command
33197 There is no corresponding @value{GDBN} command.
33199 @subsubheading Example
33202 -symbol-list-lines basics.c
33203 ^done,lines=[@{pc="0x08048554",line="7"@},@{pc="0x0804855a",line="8"@}]
33209 @subheading The @code{-symbol-list-types} Command
33210 @findex -symbol-list-types
33212 @subsubheading Synopsis
33218 List all the type names.
33220 @subsubheading @value{GDBN} Command
33222 The corresponding commands are @samp{info types} in @value{GDBN},
33223 @samp{gdb_search} in @code{gdbtk}.
33225 @subsubheading Example
33229 @subheading The @code{-symbol-list-variables} Command
33230 @findex -symbol-list-variables
33232 @subsubheading Synopsis
33235 -symbol-list-variables
33238 List all the global and static variable names.
33240 @subsubheading @value{GDBN} Command
33242 @samp{info variables} in @value{GDBN}, @samp{gdb_search} in @code{gdbtk}.
33244 @subsubheading Example
33248 @subheading The @code{-symbol-locate} Command
33249 @findex -symbol-locate
33251 @subsubheading Synopsis
33257 @subsubheading @value{GDBN} Command
33259 @samp{gdb_loc} in @code{gdbtk}.
33261 @subsubheading Example
33265 @subheading The @code{-symbol-type} Command
33266 @findex -symbol-type
33268 @subsubheading Synopsis
33271 -symbol-type @var{variable}
33274 Show type of @var{variable}.
33276 @subsubheading @value{GDBN} Command
33278 The corresponding @value{GDBN} command is @samp{ptype}, @code{gdbtk} has
33279 @samp{gdb_obj_variable}.
33281 @subsubheading Example
33286 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
33287 @node GDB/MI File Commands
33288 @section @sc{gdb/mi} File Commands
33290 This section describes the GDB/MI commands to specify executable file names
33291 and to read in and obtain symbol table information.
33293 @subheading The @code{-file-exec-and-symbols} Command
33294 @findex -file-exec-and-symbols
33296 @subsubheading Synopsis
33299 -file-exec-and-symbols @var{file}
33302 Specify the executable file to be debugged. This file is the one from
33303 which the symbol table is also read. If no file is specified, the
33304 command clears the executable and symbol information. If breakpoints
33305 are set when using this command with no arguments, @value{GDBN} will produce
33306 error messages. Otherwise, no output is produced, except a completion
33309 @subsubheading @value{GDBN} Command
33311 The corresponding @value{GDBN} command is @samp{file}.
33313 @subsubheading Example
33317 -file-exec-and-symbols /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
33323 @subheading The @code{-file-exec-file} Command
33324 @findex -file-exec-file
33326 @subsubheading Synopsis
33329 -file-exec-file @var{file}
33332 Specify the executable file to be debugged. Unlike
33333 @samp{-file-exec-and-symbols}, the symbol table is @emph{not} read
33334 from this file. If used without argument, @value{GDBN} clears the information
33335 about the executable file. No output is produced, except a completion
33338 @subsubheading @value{GDBN} Command
33340 The corresponding @value{GDBN} command is @samp{exec-file}.
33342 @subsubheading Example
33346 -file-exec-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
33353 @subheading The @code{-file-list-exec-sections} Command
33354 @findex -file-list-exec-sections
33356 @subsubheading Synopsis
33359 -file-list-exec-sections
33362 List the sections of the current executable file.
33364 @subsubheading @value{GDBN} Command
33366 The @value{GDBN} command @samp{info file} shows, among the rest, the same
33367 information as this command. @code{gdbtk} has a corresponding command
33368 @samp{gdb_load_info}.
33370 @subsubheading Example
33375 @subheading The @code{-file-list-exec-source-file} Command
33376 @findex -file-list-exec-source-file
33378 @subsubheading Synopsis
33381 -file-list-exec-source-file
33384 List the line number, the current source file, and the absolute path
33385 to the current source file for the current executable. The macro
33386 information field has a value of @samp{1} or @samp{0} depending on
33387 whether or not the file includes preprocessor macro information.
33389 @subsubheading @value{GDBN} Command
33391 The @value{GDBN} equivalent is @samp{info source}
33393 @subsubheading Example
33397 123-file-list-exec-source-file
33398 123^done,line="1",file="foo.c",fullname="/home/bar/foo.c,macro-info="1"
33403 @subheading The @code{-file-list-exec-source-files} Command
33404 @findex -file-list-exec-source-files
33406 @subsubheading Synopsis
33409 -file-list-exec-source-files
33412 List the source files for the current executable.
33414 It will always output both the filename and fullname (absolute file
33415 name) of a source file.
33417 @subsubheading @value{GDBN} Command
33419 The @value{GDBN} equivalent is @samp{info sources}.
33420 @code{gdbtk} has an analogous command @samp{gdb_listfiles}.
33422 @subsubheading Example
33425 -file-list-exec-source-files
33427 @{file=foo.c,fullname=/home/foo.c@},
33428 @{file=/home/bar.c,fullname=/home/bar.c@},
33429 @{file=gdb_could_not_find_fullpath.c@}]
33433 @subheading The @code{-file-list-shared-libraries} Command
33434 @findex -file-list-shared-libraries
33436 @subsubheading Synopsis
33439 -file-list-shared-libraries [ @var{regexp} ]
33442 List the shared libraries in the program.
33443 With a regular expression @var{regexp}, only those libraries whose
33444 names match @var{regexp} are listed.
33446 @subsubheading @value{GDBN} Command
33448 The corresponding @value{GDBN} command is @samp{info shared}. The fields
33449 have a similar meaning to the @code{=library-loaded} notification.
33450 The @code{ranges} field specifies the multiple segments belonging to this
33451 library. Each range has the following fields:
33455 The address defining the inclusive lower bound of the segment.
33457 The address defining the exclusive upper bound of the segment.
33460 @subsubheading Example
33463 -file-list-exec-source-files
33464 ^done,shared-libraries=[
33465 @{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"@}]@},
33466 @{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"@}]@}]
33472 @subheading The @code{-file-list-symbol-files} Command
33473 @findex -file-list-symbol-files
33475 @subsubheading Synopsis
33478 -file-list-symbol-files
33483 @subsubheading @value{GDBN} Command
33485 The corresponding @value{GDBN} command is @samp{info file} (part of it).
33487 @subsubheading Example
33492 @subheading The @code{-file-symbol-file} Command
33493 @findex -file-symbol-file
33495 @subsubheading Synopsis
33498 -file-symbol-file @var{file}
33501 Read symbol table info from the specified @var{file} argument. When
33502 used without arguments, clears @value{GDBN}'s symbol table info. No output is
33503 produced, except for a completion notification.
33505 @subsubheading @value{GDBN} Command
33507 The corresponding @value{GDBN} command is @samp{symbol-file}.
33509 @subsubheading Example
33513 -file-symbol-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
33519 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
33520 @node GDB/MI Memory Overlay Commands
33521 @section @sc{gdb/mi} Memory Overlay Commands
33523 The memory overlay commands are not implemented.
33525 @c @subheading -overlay-auto
33527 @c @subheading -overlay-list-mapping-state
33529 @c @subheading -overlay-list-overlays
33531 @c @subheading -overlay-map
33533 @c @subheading -overlay-off
33535 @c @subheading -overlay-on
33537 @c @subheading -overlay-unmap
33539 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
33540 @node GDB/MI Signal Handling Commands
33541 @section @sc{gdb/mi} Signal Handling Commands
33543 Signal handling commands are not implemented.
33545 @c @subheading -signal-handle
33547 @c @subheading -signal-list-handle-actions
33549 @c @subheading -signal-list-signal-types
33553 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
33554 @node GDB/MI Target Manipulation
33555 @section @sc{gdb/mi} Target Manipulation Commands
33558 @subheading The @code{-target-attach} Command
33559 @findex -target-attach
33561 @subsubheading Synopsis
33564 -target-attach @var{pid} | @var{gid} | @var{file}
33567 Attach to a process @var{pid} or a file @var{file} outside of
33568 @value{GDBN}, or a thread group @var{gid}. If attaching to a thread
33569 group, the id previously returned by
33570 @samp{-list-thread-groups --available} must be used.
33572 @subsubheading @value{GDBN} Command
33574 The corresponding @value{GDBN} command is @samp{attach}.
33576 @subsubheading Example
33580 =thread-created,id="1"
33581 *stopped,thread-id="1",frame=@{addr="0xb7f7e410",func="bar",args=[]@}
33587 @subheading The @code{-target-compare-sections} Command
33588 @findex -target-compare-sections
33590 @subsubheading Synopsis
33593 -target-compare-sections [ @var{section} ]
33596 Compare data of section @var{section} on target to the exec file.
33597 Without the argument, all sections are compared.
33599 @subsubheading @value{GDBN} Command
33601 The @value{GDBN} equivalent is @samp{compare-sections}.
33603 @subsubheading Example
33608 @subheading The @code{-target-detach} Command
33609 @findex -target-detach
33611 @subsubheading Synopsis
33614 -target-detach [ @var{pid} | @var{gid} ]
33617 Detach from the remote target which normally resumes its execution.
33618 If either @var{pid} or @var{gid} is specified, detaches from either
33619 the specified process, or specified thread group. There's no output.
33621 @subsubheading @value{GDBN} Command
33623 The corresponding @value{GDBN} command is @samp{detach}.
33625 @subsubheading Example
33635 @subheading The @code{-target-disconnect} Command
33636 @findex -target-disconnect
33638 @subsubheading Synopsis
33644 Disconnect from the remote target. There's no output and the target is
33645 generally not resumed.
33647 @subsubheading @value{GDBN} Command
33649 The corresponding @value{GDBN} command is @samp{disconnect}.
33651 @subsubheading Example
33661 @subheading The @code{-target-download} Command
33662 @findex -target-download
33664 @subsubheading Synopsis
33670 Loads the executable onto the remote target.
33671 It prints out an update message every half second, which includes the fields:
33675 The name of the section.
33677 The size of what has been sent so far for that section.
33679 The size of the section.
33681 The total size of what was sent so far (the current and the previous sections).
33683 The size of the overall executable to download.
33687 Each message is sent as status record (@pxref{GDB/MI Output Syntax, ,
33688 @sc{gdb/mi} Output Syntax}).
33690 In addition, it prints the name and size of the sections, as they are
33691 downloaded. These messages include the following fields:
33695 The name of the section.
33697 The size of the section.
33699 The size of the overall executable to download.
33703 At the end, a summary is printed.
33705 @subsubheading @value{GDBN} Command
33707 The corresponding @value{GDBN} command is @samp{load}.
33709 @subsubheading Example
33711 Note: each status message appears on a single line. Here the messages
33712 have been broken down so that they can fit onto a page.
33717 +download,@{section=".text",section-size="6668",total-size="9880"@}
33718 +download,@{section=".text",section-sent="512",section-size="6668",
33719 total-sent="512",total-size="9880"@}
33720 +download,@{section=".text",section-sent="1024",section-size="6668",
33721 total-sent="1024",total-size="9880"@}
33722 +download,@{section=".text",section-sent="1536",section-size="6668",
33723 total-sent="1536",total-size="9880"@}
33724 +download,@{section=".text",section-sent="2048",section-size="6668",
33725 total-sent="2048",total-size="9880"@}
33726 +download,@{section=".text",section-sent="2560",section-size="6668",
33727 total-sent="2560",total-size="9880"@}
33728 +download,@{section=".text",section-sent="3072",section-size="6668",
33729 total-sent="3072",total-size="9880"@}
33730 +download,@{section=".text",section-sent="3584",section-size="6668",
33731 total-sent="3584",total-size="9880"@}
33732 +download,@{section=".text",section-sent="4096",section-size="6668",
33733 total-sent="4096",total-size="9880"@}
33734 +download,@{section=".text",section-sent="4608",section-size="6668",
33735 total-sent="4608",total-size="9880"@}
33736 +download,@{section=".text",section-sent="5120",section-size="6668",
33737 total-sent="5120",total-size="9880"@}
33738 +download,@{section=".text",section-sent="5632",section-size="6668",
33739 total-sent="5632",total-size="9880"@}
33740 +download,@{section=".text",section-sent="6144",section-size="6668",
33741 total-sent="6144",total-size="9880"@}
33742 +download,@{section=".text",section-sent="6656",section-size="6668",
33743 total-sent="6656",total-size="9880"@}
33744 +download,@{section=".init",section-size="28",total-size="9880"@}
33745 +download,@{section=".fini",section-size="28",total-size="9880"@}
33746 +download,@{section=".data",section-size="3156",total-size="9880"@}
33747 +download,@{section=".data",section-sent="512",section-size="3156",
33748 total-sent="7236",total-size="9880"@}
33749 +download,@{section=".data",section-sent="1024",section-size="3156",
33750 total-sent="7748",total-size="9880"@}
33751 +download,@{section=".data",section-sent="1536",section-size="3156",
33752 total-sent="8260",total-size="9880"@}
33753 +download,@{section=".data",section-sent="2048",section-size="3156",
33754 total-sent="8772",total-size="9880"@}
33755 +download,@{section=".data",section-sent="2560",section-size="3156",
33756 total-sent="9284",total-size="9880"@}
33757 +download,@{section=".data",section-sent="3072",section-size="3156",
33758 total-sent="9796",total-size="9880"@}
33759 ^done,address="0x10004",load-size="9880",transfer-rate="6586",
33766 @subheading The @code{-target-exec-status} Command
33767 @findex -target-exec-status
33769 @subsubheading Synopsis
33772 -target-exec-status
33775 Provide information on the state of the target (whether it is running or
33776 not, for instance).
33778 @subsubheading @value{GDBN} Command
33780 There's no equivalent @value{GDBN} command.
33782 @subsubheading Example
33786 @subheading The @code{-target-list-available-targets} Command
33787 @findex -target-list-available-targets
33789 @subsubheading Synopsis
33792 -target-list-available-targets
33795 List the possible targets to connect to.
33797 @subsubheading @value{GDBN} Command
33799 The corresponding @value{GDBN} command is @samp{help target}.
33801 @subsubheading Example
33805 @subheading The @code{-target-list-current-targets} Command
33806 @findex -target-list-current-targets
33808 @subsubheading Synopsis
33811 -target-list-current-targets
33814 Describe the current target.
33816 @subsubheading @value{GDBN} Command
33818 The corresponding information is printed by @samp{info file} (among
33821 @subsubheading Example
33825 @subheading The @code{-target-list-parameters} Command
33826 @findex -target-list-parameters
33828 @subsubheading Synopsis
33831 -target-list-parameters
33837 @subsubheading @value{GDBN} Command
33841 @subsubheading Example
33844 @subheading The @code{-target-flash-erase} Command
33845 @findex -target-flash-erase
33847 @subsubheading Synopsis
33850 -target-flash-erase
33853 Erases all known flash memory regions on the target.
33855 The corresponding @value{GDBN} command is @samp{flash-erase}.
33857 The output is a list of flash regions that have been erased, with starting
33858 addresses and memory region sizes.
33862 -target-flash-erase
33863 ^done,erased-regions=@{address="0x0",size="0x40000"@}
33867 @subheading The @code{-target-select} Command
33868 @findex -target-select
33870 @subsubheading Synopsis
33873 -target-select @var{type} @var{parameters @dots{}}
33876 Connect @value{GDBN} to the remote target. This command takes two args:
33880 The type of target, for instance @samp{remote}, etc.
33881 @item @var{parameters}
33882 Device names, host names and the like. @xref{Target Commands, ,
33883 Commands for Managing Targets}, for more details.
33886 The output is a connection notification, followed by the address at
33887 which the target program is, in the following form:
33890 ^connected,addr="@var{address}",func="@var{function name}",
33891 args=[@var{arg list}]
33894 @subsubheading @value{GDBN} Command
33896 The corresponding @value{GDBN} command is @samp{target}.
33898 @subsubheading Example
33902 -target-select remote /dev/ttya
33903 ^connected,addr="0xfe00a300",func="??",args=[]
33907 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
33908 @node GDB/MI File Transfer Commands
33909 @section @sc{gdb/mi} File Transfer Commands
33912 @subheading The @code{-target-file-put} Command
33913 @findex -target-file-put
33915 @subsubheading Synopsis
33918 -target-file-put @var{hostfile} @var{targetfile}
33921 Copy file @var{hostfile} from the host system (the machine running
33922 @value{GDBN}) to @var{targetfile} on the target system.
33924 @subsubheading @value{GDBN} Command
33926 The corresponding @value{GDBN} command is @samp{remote put}.
33928 @subsubheading Example
33932 -target-file-put localfile remotefile
33938 @subheading The @code{-target-file-get} Command
33939 @findex -target-file-get
33941 @subsubheading Synopsis
33944 -target-file-get @var{targetfile} @var{hostfile}
33947 Copy file @var{targetfile} from the target system to @var{hostfile}
33948 on the host system.
33950 @subsubheading @value{GDBN} Command
33952 The corresponding @value{GDBN} command is @samp{remote get}.
33954 @subsubheading Example
33958 -target-file-get remotefile localfile
33964 @subheading The @code{-target-file-delete} Command
33965 @findex -target-file-delete
33967 @subsubheading Synopsis
33970 -target-file-delete @var{targetfile}
33973 Delete @var{targetfile} from the target system.
33975 @subsubheading @value{GDBN} Command
33977 The corresponding @value{GDBN} command is @samp{remote delete}.
33979 @subsubheading Example
33983 -target-file-delete remotefile
33989 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
33990 @node GDB/MI Ada Exceptions Commands
33991 @section Ada Exceptions @sc{gdb/mi} Commands
33993 @subheading The @code{-info-ada-exceptions} Command
33994 @findex -info-ada-exceptions
33996 @subsubheading Synopsis
33999 -info-ada-exceptions [ @var{regexp}]
34002 List all Ada exceptions defined within the program being debugged.
34003 With a regular expression @var{regexp}, only those exceptions whose
34004 names match @var{regexp} are listed.
34006 @subsubheading @value{GDBN} Command
34008 The corresponding @value{GDBN} command is @samp{info exceptions}.
34010 @subsubheading Result
34012 The result is a table of Ada exceptions. The following columns are
34013 defined for each exception:
34017 The name of the exception.
34020 The address of the exception.
34024 @subsubheading Example
34027 -info-ada-exceptions aint
34028 ^done,ada-exceptions=@{nr_rows="2",nr_cols="2",
34029 hdr=[@{width="1",alignment="-1",col_name="name",colhdr="Name"@},
34030 @{width="1",alignment="-1",col_name="address",colhdr="Address"@}],
34031 body=[@{name="constraint_error",address="0x0000000000613da0"@},
34032 @{name="const.aint_global_e",address="0x0000000000613b00"@}]@}
34035 @subheading Catching Ada Exceptions
34037 The commands describing how to ask @value{GDBN} to stop when a program
34038 raises an exception are described at @ref{Ada Exception GDB/MI
34039 Catchpoint Commands}.
34042 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
34043 @node GDB/MI Support Commands
34044 @section @sc{gdb/mi} Support Commands
34046 Since new commands and features get regularly added to @sc{gdb/mi},
34047 some commands are available to help front-ends query the debugger
34048 about support for these capabilities. Similarly, it is also possible
34049 to query @value{GDBN} about target support of certain features.
34051 @subheading The @code{-info-gdb-mi-command} Command
34052 @cindex @code{-info-gdb-mi-command}
34053 @findex -info-gdb-mi-command
34055 @subsubheading Synopsis
34058 -info-gdb-mi-command @var{cmd_name}
34061 Query support for the @sc{gdb/mi} command named @var{cmd_name}.
34063 Note that the dash (@code{-}) starting all @sc{gdb/mi} commands
34064 is technically not part of the command name (@pxref{GDB/MI Input
34065 Syntax}), and thus should be omitted in @var{cmd_name}. However,
34066 for ease of use, this command also accepts the form with the leading
34069 @subsubheading @value{GDBN} Command
34071 There is no corresponding @value{GDBN} command.
34073 @subsubheading Result
34075 The result is a tuple. There is currently only one field:
34079 This field is equal to @code{"true"} if the @sc{gdb/mi} command exists,
34080 @code{"false"} otherwise.
34084 @subsubheading Example
34086 Here is an example where the @sc{gdb/mi} command does not exist:
34089 -info-gdb-mi-command unsupported-command
34090 ^done,command=@{exists="false"@}
34094 And here is an example where the @sc{gdb/mi} command is known
34098 -info-gdb-mi-command symbol-list-lines
34099 ^done,command=@{exists="true"@}
34102 @subheading The @code{-list-features} Command
34103 @findex -list-features
34104 @cindex supported @sc{gdb/mi} features, list
34106 Returns a list of particular features of the MI protocol that
34107 this version of gdb implements. A feature can be a command,
34108 or a new field in an output of some command, or even an
34109 important bugfix. While a frontend can sometimes detect presence
34110 of a feature at runtime, it is easier to perform detection at debugger
34113 The command returns a list of strings, with each string naming an
34114 available feature. Each returned string is just a name, it does not
34115 have any internal structure. The list of possible feature names
34121 (gdb) -list-features
34122 ^done,result=["feature1","feature2"]
34125 The current list of features is:
34128 @item frozen-varobjs
34129 Indicates support for the @code{-var-set-frozen} command, as well
34130 as possible presense of the @code{frozen} field in the output
34131 of @code{-varobj-create}.
34132 @item pending-breakpoints
34133 Indicates support for the @option{-f} option to the @code{-break-insert}
34136 Indicates Python scripting support, Python-based
34137 pretty-printing commands, and possible presence of the
34138 @samp{display_hint} field in the output of @code{-var-list-children}
34140 Indicates support for the @code{-thread-info} command.
34141 @item data-read-memory-bytes
34142 Indicates support for the @code{-data-read-memory-bytes} and the
34143 @code{-data-write-memory-bytes} commands.
34144 @item breakpoint-notifications
34145 Indicates that changes to breakpoints and breakpoints created via the
34146 CLI will be announced via async records.
34147 @item ada-task-info
34148 Indicates support for the @code{-ada-task-info} command.
34149 @item language-option
34150 Indicates that all @sc{gdb/mi} commands accept the @option{--language}
34151 option (@pxref{Context management}).
34152 @item info-gdb-mi-command
34153 Indicates support for the @code{-info-gdb-mi-command} command.
34154 @item undefined-command-error-code
34155 Indicates support for the "undefined-command" error code in error result
34156 records, produced when trying to execute an undefined @sc{gdb/mi} command
34157 (@pxref{GDB/MI Result Records}).
34158 @item exec-run-start-option
34159 Indicates that the @code{-exec-run} command supports the @option{--start}
34160 option (@pxref{GDB/MI Program Execution}).
34161 @item data-disassemble-a-option
34162 Indicates that the @code{-data-disassemble} command supports the @option{-a}
34163 option (@pxref{GDB/MI Data Manipulation}).
34166 @subheading The @code{-list-target-features} Command
34167 @findex -list-target-features
34169 Returns a list of particular features that are supported by the
34170 target. Those features affect the permitted MI commands, but
34171 unlike the features reported by the @code{-list-features} command, the
34172 features depend on which target GDB is using at the moment. Whenever
34173 a target can change, due to commands such as @code{-target-select},
34174 @code{-target-attach} or @code{-exec-run}, the list of target features
34175 may change, and the frontend should obtain it again.
34179 (gdb) -list-target-features
34180 ^done,result=["async"]
34183 The current list of features is:
34187 Indicates that the target is capable of asynchronous command
34188 execution, which means that @value{GDBN} will accept further commands
34189 while the target is running.
34192 Indicates that the target is capable of reverse execution.
34193 @xref{Reverse Execution}, for more information.
34197 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
34198 @node GDB/MI Miscellaneous Commands
34199 @section Miscellaneous @sc{gdb/mi} Commands
34201 @c @subheading -gdb-complete
34203 @subheading The @code{-gdb-exit} Command
34206 @subsubheading Synopsis
34212 Exit @value{GDBN} immediately.
34214 @subsubheading @value{GDBN} Command
34216 Approximately corresponds to @samp{quit}.
34218 @subsubheading Example
34228 @subheading The @code{-exec-abort} Command
34229 @findex -exec-abort
34231 @subsubheading Synopsis
34237 Kill the inferior running program.
34239 @subsubheading @value{GDBN} Command
34241 The corresponding @value{GDBN} command is @samp{kill}.
34243 @subsubheading Example
34248 @subheading The @code{-gdb-set} Command
34251 @subsubheading Synopsis
34257 Set an internal @value{GDBN} variable.
34258 @c IS THIS A DOLLAR VARIABLE? OR SOMETHING LIKE ANNOTATE ?????
34260 @subsubheading @value{GDBN} Command
34262 The corresponding @value{GDBN} command is @samp{set}.
34264 @subsubheading Example
34274 @subheading The @code{-gdb-show} Command
34277 @subsubheading Synopsis
34283 Show the current value of a @value{GDBN} variable.
34285 @subsubheading @value{GDBN} Command
34287 The corresponding @value{GDBN} command is @samp{show}.
34289 @subsubheading Example
34298 @c @subheading -gdb-source
34301 @subheading The @code{-gdb-version} Command
34302 @findex -gdb-version
34304 @subsubheading Synopsis
34310 Show version information for @value{GDBN}. Used mostly in testing.
34312 @subsubheading @value{GDBN} Command
34314 The @value{GDBN} equivalent is @samp{show version}. @value{GDBN} by
34315 default shows this information when you start an interactive session.
34317 @subsubheading Example
34319 @c This example modifies the actual output from GDB to avoid overfull
34325 ~Copyright 2000 Free Software Foundation, Inc.
34326 ~GDB is free software, covered by the GNU General Public License, and
34327 ~you are welcome to change it and/or distribute copies of it under
34328 ~ certain conditions.
34329 ~Type "show copying" to see the conditions.
34330 ~There is absolutely no warranty for GDB. Type "show warranty" for
34332 ~This GDB was configured as
34333 "--host=sparc-sun-solaris2.5.1 --target=ppc-eabi".
34338 @subheading The @code{-list-thread-groups} Command
34339 @findex -list-thread-groups
34341 @subheading Synopsis
34344 -list-thread-groups [ --available ] [ --recurse 1 ] [ @var{group} ... ]
34347 Lists thread groups (@pxref{Thread groups}). When a single thread
34348 group is passed as the argument, lists the children of that group.
34349 When several thread group are passed, lists information about those
34350 thread groups. Without any parameters, lists information about all
34351 top-level thread groups.
34353 Normally, thread groups that are being debugged are reported.
34354 With the @samp{--available} option, @value{GDBN} reports thread groups
34355 available on the target.
34357 The output of this command may have either a @samp{threads} result or
34358 a @samp{groups} result. The @samp{thread} result has a list of tuples
34359 as value, with each tuple describing a thread (@pxref{GDB/MI Thread
34360 Information}). The @samp{groups} result has a list of tuples as value,
34361 each tuple describing a thread group. If top-level groups are
34362 requested (that is, no parameter is passed), or when several groups
34363 are passed, the output always has a @samp{groups} result. The format
34364 of the @samp{group} result is described below.
34366 To reduce the number of roundtrips it's possible to list thread groups
34367 together with their children, by passing the @samp{--recurse} option
34368 and the recursion depth. Presently, only recursion depth of 1 is
34369 permitted. If this option is present, then every reported thread group
34370 will also include its children, either as @samp{group} or
34371 @samp{threads} field.
34373 In general, any combination of option and parameters is permitted, with
34374 the following caveats:
34378 When a single thread group is passed, the output will typically
34379 be the @samp{threads} result. Because threads may not contain
34380 anything, the @samp{recurse} option will be ignored.
34383 When the @samp{--available} option is passed, limited information may
34384 be available. In particular, the list of threads of a process might
34385 be inaccessible. Further, specifying specific thread groups might
34386 not give any performance advantage over listing all thread groups.
34387 The frontend should assume that @samp{-list-thread-groups --available}
34388 is always an expensive operation and cache the results.
34392 The @samp{groups} result is a list of tuples, where each tuple may
34393 have the following fields:
34397 Identifier of the thread group. This field is always present.
34398 The identifier is an opaque string; frontends should not try to
34399 convert it to an integer, even though it might look like one.
34402 The type of the thread group. At present, only @samp{process} is a
34406 The target-specific process identifier. This field is only present
34407 for thread groups of type @samp{process} and only if the process exists.
34410 The exit code of this group's last exited thread, formatted in octal.
34411 This field is only present for thread groups of type @samp{process} and
34412 only if the process is not running.
34415 The number of children this thread group has. This field may be
34416 absent for an available thread group.
34419 This field has a list of tuples as value, each tuple describing a
34420 thread. It may be present if the @samp{--recurse} option is
34421 specified, and it's actually possible to obtain the threads.
34424 This field is a list of integers, each identifying a core that one
34425 thread of the group is running on. This field may be absent if
34426 such information is not available.
34429 The name of the executable file that corresponds to this thread group.
34430 The field is only present for thread groups of type @samp{process},
34431 and only if there is a corresponding executable file.
34435 @subheading Example
34439 -list-thread-groups
34440 ^done,groups=[@{id="17",type="process",pid="yyy",num_children="2"@}]
34441 -list-thread-groups 17
34442 ^done,threads=[@{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
34443 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",args=[]@},state="running"@},
34444 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
34445 frame=@{level="0",addr="0x0804891f",func="foo",args=[@{name="i",value="10"@}],
34446 file="/tmp/a.c",fullname="/tmp/a.c",line="158",arch="i386:x86_64"@},state="running"@}]]
34447 -list-thread-groups --available
34448 ^done,groups=[@{id="17",type="process",pid="yyy",num_children="2",cores=[1,2]@}]
34449 -list-thread-groups --available --recurse 1
34450 ^done,groups=[@{id="17", types="process",pid="yyy",num_children="2",cores=[1,2],
34451 threads=[@{id="1",target-id="Thread 0xb7e14b90",cores=[1]@},
34452 @{id="2",target-id="Thread 0xb7e14b90",cores=[2]@}]@},..]
34453 -list-thread-groups --available --recurse 1 17 18
34454 ^done,groups=[@{id="17", types="process",pid="yyy",num_children="2",cores=[1,2],
34455 threads=[@{id="1",target-id="Thread 0xb7e14b90",cores=[1]@},
34456 @{id="2",target-id="Thread 0xb7e14b90",cores=[2]@}]@},...]
34459 @subheading The @code{-info-os} Command
34462 @subsubheading Synopsis
34465 -info-os [ @var{type} ]
34468 If no argument is supplied, the command returns a table of available
34469 operating-system-specific information types. If one of these types is
34470 supplied as an argument @var{type}, then the command returns a table
34471 of data of that type.
34473 The types of information available depend on the target operating
34476 @subsubheading @value{GDBN} Command
34478 The corresponding @value{GDBN} command is @samp{info os}.
34480 @subsubheading Example
34482 When run on a @sc{gnu}/Linux system, the output will look something
34488 ^done,OSDataTable=@{nr_rows="10",nr_cols="3",
34489 hdr=[@{width="10",alignment="-1",col_name="col0",colhdr="Type"@},
34490 @{width="10",alignment="-1",col_name="col1",colhdr="Description"@},
34491 @{width="10",alignment="-1",col_name="col2",colhdr="Title"@}],
34492 body=[item=@{col0="cpus",col1="Listing of all cpus/cores on the system",
34494 item=@{col0="files",col1="Listing of all file descriptors",
34495 col2="File descriptors"@},
34496 item=@{col0="modules",col1="Listing of all loaded kernel modules",
34497 col2="Kernel modules"@},
34498 item=@{col0="msg",col1="Listing of all message queues",
34499 col2="Message queues"@},
34500 item=@{col0="processes",col1="Listing of all processes",
34501 col2="Processes"@},
34502 item=@{col0="procgroups",col1="Listing of all process groups",
34503 col2="Process groups"@},
34504 item=@{col0="semaphores",col1="Listing of all semaphores",
34505 col2="Semaphores"@},
34506 item=@{col0="shm",col1="Listing of all shared-memory regions",
34507 col2="Shared-memory regions"@},
34508 item=@{col0="sockets",col1="Listing of all internet-domain sockets",
34510 item=@{col0="threads",col1="Listing of all threads",
34514 ^done,OSDataTable=@{nr_rows="190",nr_cols="4",
34515 hdr=[@{width="10",alignment="-1",col_name="col0",colhdr="pid"@},
34516 @{width="10",alignment="-1",col_name="col1",colhdr="user"@},
34517 @{width="10",alignment="-1",col_name="col2",colhdr="command"@},
34518 @{width="10",alignment="-1",col_name="col3",colhdr="cores"@}],
34519 body=[item=@{col0="1",col1="root",col2="/sbin/init",col3="0"@},
34520 item=@{col0="2",col1="root",col2="[kthreadd]",col3="1"@},
34521 item=@{col0="3",col1="root",col2="[ksoftirqd/0]",col3="0"@},
34523 item=@{col0="26446",col1="stan",col2="bash",col3="0"@},
34524 item=@{col0="28152",col1="stan",col2="bash",col3="1"@}]@}
34528 (Note that the MI output here includes a @code{"Title"} column that
34529 does not appear in command-line @code{info os}; this column is useful
34530 for MI clients that want to enumerate the types of data, such as in a
34531 popup menu, but is needless clutter on the command line, and
34532 @code{info os} omits it.)
34534 @subheading The @code{-add-inferior} Command
34535 @findex -add-inferior
34537 @subheading Synopsis
34543 Creates a new inferior (@pxref{Inferiors and Programs}). The created
34544 inferior is not associated with any executable. Such association may
34545 be established with the @samp{-file-exec-and-symbols} command
34546 (@pxref{GDB/MI File Commands}). The command response has a single
34547 field, @samp{inferior}, whose value is the identifier of the
34548 thread group corresponding to the new inferior.
34550 @subheading Example
34555 ^done,inferior="i3"
34558 @subheading The @code{-interpreter-exec} Command
34559 @findex -interpreter-exec
34561 @subheading Synopsis
34564 -interpreter-exec @var{interpreter} @var{command}
34566 @anchor{-interpreter-exec}
34568 Execute the specified @var{command} in the given @var{interpreter}.
34570 @subheading @value{GDBN} Command
34572 The corresponding @value{GDBN} command is @samp{interpreter-exec}.
34574 @subheading Example
34578 -interpreter-exec console "break main"
34579 &"During symbol reading, couldn't parse type; debugger out of date?.\n"
34580 &"During symbol reading, bad structure-type format.\n"
34581 ~"Breakpoint 1 at 0x8074fc6: file ../../src/gdb/main.c, line 743.\n"
34586 @subheading The @code{-inferior-tty-set} Command
34587 @findex -inferior-tty-set
34589 @subheading Synopsis
34592 -inferior-tty-set /dev/pts/1
34595 Set terminal for future runs of the program being debugged.
34597 @subheading @value{GDBN} Command
34599 The corresponding @value{GDBN} command is @samp{set inferior-tty} /dev/pts/1.
34601 @subheading Example
34605 -inferior-tty-set /dev/pts/1
34610 @subheading The @code{-inferior-tty-show} Command
34611 @findex -inferior-tty-show
34613 @subheading Synopsis
34619 Show terminal for future runs of program being debugged.
34621 @subheading @value{GDBN} Command
34623 The corresponding @value{GDBN} command is @samp{show inferior-tty}.
34625 @subheading Example
34629 -inferior-tty-set /dev/pts/1
34633 ^done,inferior_tty_terminal="/dev/pts/1"
34637 @subheading The @code{-enable-timings} Command
34638 @findex -enable-timings
34640 @subheading Synopsis
34643 -enable-timings [yes | no]
34646 Toggle the printing of the wallclock, user and system times for an MI
34647 command as a field in its output. This command is to help frontend
34648 developers optimize the performance of their code. No argument is
34649 equivalent to @samp{yes}.
34651 @subheading @value{GDBN} Command
34655 @subheading Example
34663 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
34664 addr="0x080484ed",func="main",file="myprog.c",
34665 fullname="/home/nickrob/myprog.c",line="73",thread-groups=["i1"],
34667 time=@{wallclock="0.05185",user="0.00800",system="0.00000"@}
34675 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
34676 frame=@{addr="0x080484ed",func="main",args=[@{name="argc",value="1"@},
34677 @{name="argv",value="0xbfb60364"@}],file="myprog.c",
34678 fullname="/home/nickrob/myprog.c",line="73",arch="i386:x86_64"@}
34682 @subheading The @code{-complete} Command
34685 @subheading Synopsis
34688 -complete @var{command}
34691 Show a list of completions for partially typed CLI @var{command}.
34693 This command is intended for @sc{gdb/mi} frontends that cannot use two separate
34694 CLI and MI channels --- for example: because of lack of PTYs like on Windows or
34695 because @value{GDBN} is used remotely via a SSH connection.
34699 The result consists of two or three fields:
34703 This field contains the completed @var{command}. If @var{command}
34704 has no known completions, this field is omitted.
34707 This field contains a (possibly empty) array of matches. It is always present.
34709 @item max_completions_reached
34710 This field contains @code{1} if number of known completions is above
34711 @code{max-completions} limit (@pxref{Completion}), otherwise it contains
34712 @code{0}. It is always present.
34716 @subheading @value{GDBN} Command
34718 The corresponding @value{GDBN} command is @samp{complete}.
34720 @subheading Example
34725 ^done,completion="break",
34726 matches=["break","break-range"],
34727 max_completions_reached="0"
34730 ^done,completion="b ma",
34731 matches=["b madvise","b main"],max_completions_reached="0"
34733 -complete "b push_b"
34734 ^done,completion="b push_back(",
34736 "b A::push_back(void*)",
34737 "b std::string::push_back(char)",
34738 "b std::vector<int, std::allocator<int> >::push_back(int&&)"],
34739 max_completions_reached="0"
34741 -complete "nonexist"
34742 ^done,matches=[],max_completions_reached="0"
34748 @chapter @value{GDBN} Annotations
34750 This chapter describes annotations in @value{GDBN}. Annotations were
34751 designed to interface @value{GDBN} to graphical user interfaces or other
34752 similar programs which want to interact with @value{GDBN} at a
34753 relatively high level.
34755 The annotation mechanism has largely been superseded by @sc{gdb/mi}
34759 This is Edition @value{EDITION}, @value{DATE}.
34763 * Annotations Overview:: What annotations are; the general syntax.
34764 * Server Prefix:: Issuing a command without affecting user state.
34765 * Prompting:: Annotations marking @value{GDBN}'s need for input.
34766 * Errors:: Annotations for error messages.
34767 * Invalidation:: Some annotations describe things now invalid.
34768 * Annotations for Running::
34769 Whether the program is running, how it stopped, etc.
34770 * Source Annotations:: Annotations describing source code.
34773 @node Annotations Overview
34774 @section What is an Annotation?
34775 @cindex annotations
34777 Annotations start with a newline character, two @samp{control-z}
34778 characters, and the name of the annotation. If there is no additional
34779 information associated with this annotation, the name of the annotation
34780 is followed immediately by a newline. If there is additional
34781 information, the name of the annotation is followed by a space, the
34782 additional information, and a newline. The additional information
34783 cannot contain newline characters.
34785 Any output not beginning with a newline and two @samp{control-z}
34786 characters denotes literal output from @value{GDBN}. Currently there is
34787 no need for @value{GDBN} to output a newline followed by two
34788 @samp{control-z} characters, but if there was such a need, the
34789 annotations could be extended with an @samp{escape} annotation which
34790 means those three characters as output.
34792 The annotation @var{level}, which is specified using the
34793 @option{--annotate} command line option (@pxref{Mode Options}), controls
34794 how much information @value{GDBN} prints together with its prompt,
34795 values of expressions, source lines, and other types of output. Level 0
34796 is for no annotations, level 1 is for use when @value{GDBN} is run as a
34797 subprocess of @sc{gnu} Emacs, level 3 is the maximum annotation suitable
34798 for programs that control @value{GDBN}, and level 2 annotations have
34799 been made obsolete (@pxref{Limitations, , Limitations of the Annotation
34800 Interface, annotate, GDB's Obsolete Annotations}).
34803 @kindex set annotate
34804 @item set annotate @var{level}
34805 The @value{GDBN} command @code{set annotate} sets the level of
34806 annotations to the specified @var{level}.
34808 @item show annotate
34809 @kindex show annotate
34810 Show the current annotation level.
34813 This chapter describes level 3 annotations.
34815 A simple example of starting up @value{GDBN} with annotations is:
34818 $ @kbd{gdb --annotate=3}
34820 Copyright 2003 Free Software Foundation, Inc.
34821 GDB is free software, covered by the GNU General Public License,
34822 and you are welcome to change it and/or distribute copies of it
34823 under certain conditions.
34824 Type "show copying" to see the conditions.
34825 There is absolutely no warranty for GDB. Type "show warranty"
34827 This GDB was configured as "i386-pc-linux-gnu"
34838 Here @samp{quit} is input to @value{GDBN}; the rest is output from
34839 @value{GDBN}. The three lines beginning @samp{^Z^Z} (where @samp{^Z}
34840 denotes a @samp{control-z} character) are annotations; the rest is
34841 output from @value{GDBN}.
34843 @node Server Prefix
34844 @section The Server Prefix
34845 @cindex server prefix
34847 If you prefix a command with @samp{server } then it will not affect
34848 the command history, nor will it affect @value{GDBN}'s notion of which
34849 command to repeat if @key{RET} is pressed on a line by itself. This
34850 means that commands can be run behind a user's back by a front-end in
34851 a transparent manner.
34853 The @code{server } prefix does not affect the recording of values into
34854 the value history; to print a value without recording it into the
34855 value history, use the @code{output} command instead of the
34856 @code{print} command.
34858 Using this prefix also disables confirmation requests
34859 (@pxref{confirmation requests}).
34862 @section Annotation for @value{GDBN} Input
34864 @cindex annotations for prompts
34865 When @value{GDBN} prompts for input, it annotates this fact so it is possible
34866 to know when to send output, when the output from a given command is
34869 Different kinds of input each have a different @dfn{input type}. Each
34870 input type has three annotations: a @code{pre-} annotation, which
34871 denotes the beginning of any prompt which is being output, a plain
34872 annotation, which denotes the end of the prompt, and then a @code{post-}
34873 annotation which denotes the end of any echo which may (or may not) be
34874 associated with the input. For example, the @code{prompt} input type
34875 features the following annotations:
34883 The input types are
34886 @findex pre-prompt annotation
34887 @findex prompt annotation
34888 @findex post-prompt annotation
34890 When @value{GDBN} is prompting for a command (the main @value{GDBN} prompt).
34892 @findex pre-commands annotation
34893 @findex commands annotation
34894 @findex post-commands annotation
34896 When @value{GDBN} prompts for a set of commands, like in the @code{commands}
34897 command. The annotations are repeated for each command which is input.
34899 @findex pre-overload-choice annotation
34900 @findex overload-choice annotation
34901 @findex post-overload-choice annotation
34902 @item overload-choice
34903 When @value{GDBN} wants the user to select between various overloaded functions.
34905 @findex pre-query annotation
34906 @findex query annotation
34907 @findex post-query annotation
34909 When @value{GDBN} wants the user to confirm a potentially dangerous operation.
34911 @findex pre-prompt-for-continue annotation
34912 @findex prompt-for-continue annotation
34913 @findex post-prompt-for-continue annotation
34914 @item prompt-for-continue
34915 When @value{GDBN} is asking the user to press return to continue. Note: Don't
34916 expect this to work well; instead use @code{set height 0} to disable
34917 prompting. This is because the counting of lines is buggy in the
34918 presence of annotations.
34923 @cindex annotations for errors, warnings and interrupts
34925 @findex quit annotation
34930 This annotation occurs right before @value{GDBN} responds to an interrupt.
34932 @findex error annotation
34937 This annotation occurs right before @value{GDBN} responds to an error.
34939 Quit and error annotations indicate that any annotations which @value{GDBN} was
34940 in the middle of may end abruptly. For example, if a
34941 @code{value-history-begin} annotation is followed by a @code{error}, one
34942 cannot expect to receive the matching @code{value-history-end}. One
34943 cannot expect not to receive it either, however; an error annotation
34944 does not necessarily mean that @value{GDBN} is immediately returning all the way
34947 @findex error-begin annotation
34948 A quit or error annotation may be preceded by
34954 Any output between that and the quit or error annotation is the error
34957 Warning messages are not yet annotated.
34958 @c If we want to change that, need to fix warning(), type_error(),
34959 @c range_error(), and possibly other places.
34962 @section Invalidation Notices
34964 @cindex annotations for invalidation messages
34965 The following annotations say that certain pieces of state may have
34969 @findex frames-invalid annotation
34970 @item ^Z^Zframes-invalid
34972 The frames (for example, output from the @code{backtrace} command) may
34975 @findex breakpoints-invalid annotation
34976 @item ^Z^Zbreakpoints-invalid
34978 The breakpoints may have changed. For example, the user just added or
34979 deleted a breakpoint.
34982 @node Annotations for Running
34983 @section Running the Program
34984 @cindex annotations for running programs
34986 @findex starting annotation
34987 @findex stopping annotation
34988 When the program starts executing due to a @value{GDBN} command such as
34989 @code{step} or @code{continue},
34995 is output. When the program stops,
35001 is output. Before the @code{stopped} annotation, a variety of
35002 annotations describe how the program stopped.
35005 @findex exited annotation
35006 @item ^Z^Zexited @var{exit-status}
35007 The program exited, and @var{exit-status} is the exit status (zero for
35008 successful exit, otherwise nonzero).
35010 @findex signalled annotation
35011 @findex signal-name annotation
35012 @findex signal-name-end annotation
35013 @findex signal-string annotation
35014 @findex signal-string-end annotation
35015 @item ^Z^Zsignalled
35016 The program exited with a signal. After the @code{^Z^Zsignalled}, the
35017 annotation continues:
35023 ^Z^Zsignal-name-end
35027 ^Z^Zsignal-string-end
35032 where @var{name} is the name of the signal, such as @code{SIGILL} or
35033 @code{SIGSEGV}, and @var{string} is the explanation of the signal, such
35034 as @code{Illegal Instruction} or @code{Segmentation fault}. The arguments
35035 @var{intro-text}, @var{middle-text}, and @var{end-text} are for the
35036 user's benefit and have no particular format.
35038 @findex signal annotation
35040 The syntax of this annotation is just like @code{signalled}, but @value{GDBN} is
35041 just saying that the program received the signal, not that it was
35042 terminated with it.
35044 @findex breakpoint annotation
35045 @item ^Z^Zbreakpoint @var{number}
35046 The program hit breakpoint number @var{number}.
35048 @findex watchpoint annotation
35049 @item ^Z^Zwatchpoint @var{number}
35050 The program hit watchpoint number @var{number}.
35053 @node Source Annotations
35054 @section Displaying Source
35055 @cindex annotations for source display
35057 @findex source annotation
35058 The following annotation is used instead of displaying source code:
35061 ^Z^Zsource @var{filename}:@var{line}:@var{character}:@var{middle}:@var{addr}
35064 where @var{filename} is an absolute file name indicating which source
35065 file, @var{line} is the line number within that file (where 1 is the
35066 first line in the file), @var{character} is the character position
35067 within the file (where 0 is the first character in the file) (for most
35068 debug formats this will necessarily point to the beginning of a line),
35069 @var{middle} is @samp{middle} if @var{addr} is in the middle of the
35070 line, or @samp{beg} if @var{addr} is at the beginning of the line, and
35071 @var{addr} is the address in the target program associated with the
35072 source which is being displayed. The @var{addr} is in the form @samp{0x}
35073 followed by one or more lowercase hex digits (note that this does not
35074 depend on the language).
35076 @node JIT Interface
35077 @chapter JIT Compilation Interface
35078 @cindex just-in-time compilation
35079 @cindex JIT compilation interface
35081 This chapter documents @value{GDBN}'s @dfn{just-in-time} (JIT) compilation
35082 interface. A JIT compiler is a program or library that generates native
35083 executable code at runtime and executes it, usually in order to achieve good
35084 performance while maintaining platform independence.
35086 Programs that use JIT compilation are normally difficult to debug because
35087 portions of their code are generated at runtime, instead of being loaded from
35088 object files, which is where @value{GDBN} normally finds the program's symbols
35089 and debug information. In order to debug programs that use JIT compilation,
35090 @value{GDBN} has an interface that allows the program to register in-memory
35091 symbol files with @value{GDBN} at runtime.
35093 If you are using @value{GDBN} to debug a program that uses this interface, then
35094 it should work transparently so long as you have not stripped the binary. If
35095 you are developing a JIT compiler, then the interface is documented in the rest
35096 of this chapter. At this time, the only known client of this interface is the
35099 Broadly speaking, the JIT interface mirrors the dynamic loader interface. The
35100 JIT compiler communicates with @value{GDBN} by writing data into a global
35101 variable and calling a fuction at a well-known symbol. When @value{GDBN}
35102 attaches, it reads a linked list of symbol files from the global variable to
35103 find existing code, and puts a breakpoint in the function so that it can find
35104 out about additional code.
35107 * Declarations:: Relevant C struct declarations
35108 * Registering Code:: Steps to register code
35109 * Unregistering Code:: Steps to unregister code
35110 * Custom Debug Info:: Emit debug information in a custom format
35114 @section JIT Declarations
35116 These are the relevant struct declarations that a C program should include to
35117 implement the interface:
35127 struct jit_code_entry
35129 struct jit_code_entry *next_entry;
35130 struct jit_code_entry *prev_entry;
35131 const char *symfile_addr;
35132 uint64_t symfile_size;
35135 struct jit_descriptor
35138 /* This type should be jit_actions_t, but we use uint32_t
35139 to be explicit about the bitwidth. */
35140 uint32_t action_flag;
35141 struct jit_code_entry *relevant_entry;
35142 struct jit_code_entry *first_entry;
35145 /* GDB puts a breakpoint in this function. */
35146 void __attribute__((noinline)) __jit_debug_register_code() @{ @};
35148 /* Make sure to specify the version statically, because the
35149 debugger may check the version before we can set it. */
35150 struct jit_descriptor __jit_debug_descriptor = @{ 1, 0, 0, 0 @};
35153 If the JIT is multi-threaded, then it is important that the JIT synchronize any
35154 modifications to this global data properly, which can easily be done by putting
35155 a global mutex around modifications to these structures.
35157 @node Registering Code
35158 @section Registering Code
35160 To register code with @value{GDBN}, the JIT should follow this protocol:
35164 Generate an object file in memory with symbols and other desired debug
35165 information. The file must include the virtual addresses of the sections.
35168 Create a code entry for the file, which gives the start and size of the symbol
35172 Add it to the linked list in the JIT descriptor.
35175 Point the relevant_entry field of the descriptor at the entry.
35178 Set @code{action_flag} to @code{JIT_REGISTER} and call
35179 @code{__jit_debug_register_code}.
35182 When @value{GDBN} is attached and the breakpoint fires, @value{GDBN} uses the
35183 @code{relevant_entry} pointer so it doesn't have to walk the list looking for
35184 new code. However, the linked list must still be maintained in order to allow
35185 @value{GDBN} to attach to a running process and still find the symbol files.
35187 @node Unregistering Code
35188 @section Unregistering Code
35190 If code is freed, then the JIT should use the following protocol:
35194 Remove the code entry corresponding to the code from the linked list.
35197 Point the @code{relevant_entry} field of the descriptor at the code entry.
35200 Set @code{action_flag} to @code{JIT_UNREGISTER} and call
35201 @code{__jit_debug_register_code}.
35204 If the JIT frees or recompiles code without unregistering it, then @value{GDBN}
35205 and the JIT will leak the memory used for the associated symbol files.
35207 @node Custom Debug Info
35208 @section Custom Debug Info
35209 @cindex custom JIT debug info
35210 @cindex JIT debug info reader
35212 Generating debug information in platform-native file formats (like ELF
35213 or COFF) may be an overkill for JIT compilers; especially if all the
35214 debug info is used for is displaying a meaningful backtrace. The
35215 issue can be resolved by having the JIT writers decide on a debug info
35216 format and also provide a reader that parses the debug info generated
35217 by the JIT compiler. This section gives a brief overview on writing
35218 such a parser. More specific details can be found in the source file
35219 @file{gdb/jit-reader.in}, which is also installed as a header at
35220 @file{@var{includedir}/gdb/jit-reader.h} for easy inclusion.
35222 The reader is implemented as a shared object (so this functionality is
35223 not available on platforms which don't allow loading shared objects at
35224 runtime). Two @value{GDBN} commands, @code{jit-reader-load} and
35225 @code{jit-reader-unload} are provided, to be used to load and unload
35226 the readers from a preconfigured directory. Once loaded, the shared
35227 object is used the parse the debug information emitted by the JIT
35231 * Using JIT Debug Info Readers:: How to use supplied readers correctly
35232 * Writing JIT Debug Info Readers:: Creating a debug-info reader
35235 @node Using JIT Debug Info Readers
35236 @subsection Using JIT Debug Info Readers
35237 @kindex jit-reader-load
35238 @kindex jit-reader-unload
35240 Readers can be loaded and unloaded using the @code{jit-reader-load}
35241 and @code{jit-reader-unload} commands.
35244 @item jit-reader-load @var{reader}
35245 Load the JIT reader named @var{reader}, which is a shared
35246 object specified as either an absolute or a relative file name. In
35247 the latter case, @value{GDBN} will try to load the reader from a
35248 pre-configured directory, usually @file{@var{libdir}/gdb/} on a UNIX
35249 system (here @var{libdir} is the system library directory, often
35250 @file{/usr/local/lib}).
35252 Only one reader can be active at a time; trying to load a second
35253 reader when one is already loaded will result in @value{GDBN}
35254 reporting an error. A new JIT reader can be loaded by first unloading
35255 the current one using @code{jit-reader-unload} and then invoking
35256 @code{jit-reader-load}.
35258 @item jit-reader-unload
35259 Unload the currently loaded JIT reader.
35263 @node Writing JIT Debug Info Readers
35264 @subsection Writing JIT Debug Info Readers
35265 @cindex writing JIT debug info readers
35267 As mentioned, a reader is essentially a shared object conforming to a
35268 certain ABI. This ABI is described in @file{jit-reader.h}.
35270 @file{jit-reader.h} defines the structures, macros and functions
35271 required to write a reader. It is installed (along with
35272 @value{GDBN}), in @file{@var{includedir}/gdb} where @var{includedir} is
35273 the system include directory.
35275 Readers need to be released under a GPL compatible license. A reader
35276 can be declared as released under such a license by placing the macro
35277 @code{GDB_DECLARE_GPL_COMPATIBLE_READER} in a source file.
35279 The entry point for readers is the symbol @code{gdb_init_reader},
35280 which is expected to be a function with the prototype
35282 @findex gdb_init_reader
35284 extern struct gdb_reader_funcs *gdb_init_reader (void);
35287 @cindex @code{struct gdb_reader_funcs}
35289 @code{struct gdb_reader_funcs} contains a set of pointers to callback
35290 functions. These functions are executed to read the debug info
35291 generated by the JIT compiler (@code{read}), to unwind stack frames
35292 (@code{unwind}) and to create canonical frame IDs
35293 (@code{get_Frame_id}). It also has a callback that is called when the
35294 reader is being unloaded (@code{destroy}). The struct looks like this
35297 struct gdb_reader_funcs
35299 /* Must be set to GDB_READER_INTERFACE_VERSION. */
35300 int reader_version;
35302 /* For use by the reader. */
35305 gdb_read_debug_info *read;
35306 gdb_unwind_frame *unwind;
35307 gdb_get_frame_id *get_frame_id;
35308 gdb_destroy_reader *destroy;
35312 @cindex @code{struct gdb_symbol_callbacks}
35313 @cindex @code{struct gdb_unwind_callbacks}
35315 The callbacks are provided with another set of callbacks by
35316 @value{GDBN} to do their job. For @code{read}, these callbacks are
35317 passed in a @code{struct gdb_symbol_callbacks} and for @code{unwind}
35318 and @code{get_frame_id}, in a @code{struct gdb_unwind_callbacks}.
35319 @code{struct gdb_symbol_callbacks} has callbacks to create new object
35320 files and new symbol tables inside those object files. @code{struct
35321 gdb_unwind_callbacks} has callbacks to read registers off the current
35322 frame and to write out the values of the registers in the previous
35323 frame. Both have a callback (@code{target_read}) to read bytes off the
35324 target's address space.
35326 @node In-Process Agent
35327 @chapter In-Process Agent
35328 @cindex debugging agent
35329 The traditional debugging model is conceptually low-speed, but works fine,
35330 because most bugs can be reproduced in debugging-mode execution. However,
35331 as multi-core or many-core processors are becoming mainstream, and
35332 multi-threaded programs become more and more popular, there should be more
35333 and more bugs that only manifest themselves at normal-mode execution, for
35334 example, thread races, because debugger's interference with the program's
35335 timing may conceal the bugs. On the other hand, in some applications,
35336 it is not feasible for the debugger to interrupt the program's execution
35337 long enough for the developer to learn anything helpful about its behavior.
35338 If the program's correctness depends on its real-time behavior, delays
35339 introduced by a debugger might cause the program to fail, even when the
35340 code itself is correct. It is useful to be able to observe the program's
35341 behavior without interrupting it.
35343 Therefore, traditional debugging model is too intrusive to reproduce
35344 some bugs. In order to reduce the interference with the program, we can
35345 reduce the number of operations performed by debugger. The
35346 @dfn{In-Process Agent}, a shared library, is running within the same
35347 process with inferior, and is able to perform some debugging operations
35348 itself. As a result, debugger is only involved when necessary, and
35349 performance of debugging can be improved accordingly. Note that
35350 interference with program can be reduced but can't be removed completely,
35351 because the in-process agent will still stop or slow down the program.
35353 The in-process agent can interpret and execute Agent Expressions
35354 (@pxref{Agent Expressions}) during performing debugging operations. The
35355 agent expressions can be used for different purposes, such as collecting
35356 data in tracepoints, and condition evaluation in breakpoints.
35358 @anchor{Control Agent}
35359 You can control whether the in-process agent is used as an aid for
35360 debugging with the following commands:
35363 @kindex set agent on
35365 Causes the in-process agent to perform some operations on behalf of the
35366 debugger. Just which operations requested by the user will be done
35367 by the in-process agent depends on the its capabilities. For example,
35368 if you request to evaluate breakpoint conditions in the in-process agent,
35369 and the in-process agent has such capability as well, then breakpoint
35370 conditions will be evaluated in the in-process agent.
35372 @kindex set agent off
35373 @item set agent off
35374 Disables execution of debugging operations by the in-process agent. All
35375 of the operations will be performed by @value{GDBN}.
35379 Display the current setting of execution of debugging operations by
35380 the in-process agent.
35384 * In-Process Agent Protocol::
35387 @node In-Process Agent Protocol
35388 @section In-Process Agent Protocol
35389 @cindex in-process agent protocol
35391 The in-process agent is able to communicate with both @value{GDBN} and
35392 GDBserver (@pxref{In-Process Agent}). This section documents the protocol
35393 used for communications between @value{GDBN} or GDBserver and the IPA.
35394 In general, @value{GDBN} or GDBserver sends commands
35395 (@pxref{IPA Protocol Commands}) and data to in-process agent, and then
35396 in-process agent replies back with the return result of the command, or
35397 some other information. The data sent to in-process agent is composed
35398 of primitive data types, such as 4-byte or 8-byte type, and composite
35399 types, which are called objects (@pxref{IPA Protocol Objects}).
35402 * IPA Protocol Objects::
35403 * IPA Protocol Commands::
35406 @node IPA Protocol Objects
35407 @subsection IPA Protocol Objects
35408 @cindex ipa protocol objects
35410 The commands sent to and results received from agent may contain some
35411 complex data types called @dfn{objects}.
35413 The in-process agent is running on the same machine with @value{GDBN}
35414 or GDBserver, so it doesn't have to handle as much differences between
35415 two ends as remote protocol (@pxref{Remote Protocol}) tries to handle.
35416 However, there are still some differences of two ends in two processes:
35420 word size. On some 64-bit machines, @value{GDBN} or GDBserver can be
35421 compiled as a 64-bit executable, while in-process agent is a 32-bit one.
35423 ABI. Some machines may have multiple types of ABI, @value{GDBN} or
35424 GDBserver is compiled with one, and in-process agent is compiled with
35428 Here are the IPA Protocol Objects:
35432 agent expression object. It represents an agent expression
35433 (@pxref{Agent Expressions}).
35434 @anchor{agent expression object}
35436 tracepoint action object. It represents a tracepoint action
35437 (@pxref{Tracepoint Actions,,Tracepoint Action Lists}) to collect registers,
35438 memory, static trace data and to evaluate expression.
35439 @anchor{tracepoint action object}
35441 tracepoint object. It represents a tracepoint (@pxref{Tracepoints}).
35442 @anchor{tracepoint object}
35446 The following table describes important attributes of each IPA protocol
35449 @multitable @columnfractions .30 .20 .50
35450 @headitem Name @tab Size @tab Description
35451 @item @emph{agent expression object} @tab @tab
35452 @item length @tab 4 @tab length of bytes code
35453 @item byte code @tab @var{length} @tab contents of byte code
35454 @item @emph{tracepoint action for collecting memory} @tab @tab
35455 @item 'M' @tab 1 @tab type of tracepoint action
35456 @item addr @tab 8 @tab if @var{basereg} is @samp{-1}, @var{addr} is the
35457 address of the lowest byte to collect, otherwise @var{addr} is the offset
35458 of @var{basereg} for memory collecting.
35459 @item len @tab 8 @tab length of memory for collecting
35460 @item basereg @tab 4 @tab the register number containing the starting
35461 memory address for collecting.
35462 @item @emph{tracepoint action for collecting registers} @tab @tab
35463 @item 'R' @tab 1 @tab type of tracepoint action
35464 @item @emph{tracepoint action for collecting static trace data} @tab @tab
35465 @item 'L' @tab 1 @tab type of tracepoint action
35466 @item @emph{tracepoint action for expression evaluation} @tab @tab
35467 @item 'X' @tab 1 @tab type of tracepoint action
35468 @item agent expression @tab length of @tab @ref{agent expression object}
35469 @item @emph{tracepoint object} @tab @tab
35470 @item number @tab 4 @tab number of tracepoint
35471 @item address @tab 8 @tab address of tracepoint inserted on
35472 @item type @tab 4 @tab type of tracepoint
35473 @item enabled @tab 1 @tab enable or disable of tracepoint
35474 @item step_count @tab 8 @tab step
35475 @item pass_count @tab 8 @tab pass
35476 @item numactions @tab 4 @tab number of tracepoint actions
35477 @item hit count @tab 8 @tab hit count
35478 @item trace frame usage @tab 8 @tab trace frame usage
35479 @item compiled_cond @tab 8 @tab compiled condition
35480 @item orig_size @tab 8 @tab orig size
35481 @item condition @tab 4 if condition is NULL otherwise length of
35482 @ref{agent expression object}
35483 @tab zero if condition is NULL, otherwise is
35484 @ref{agent expression object}
35485 @item actions @tab variable
35486 @tab numactions number of @ref{tracepoint action object}
35489 @node IPA Protocol Commands
35490 @subsection IPA Protocol Commands
35491 @cindex ipa protocol commands
35493 The spaces in each command are delimiters to ease reading this commands
35494 specification. They don't exist in real commands.
35498 @item FastTrace:@var{tracepoint_object} @var{gdb_jump_pad_head}
35499 Installs a new fast tracepoint described by @var{tracepoint_object}
35500 (@pxref{tracepoint object}). The @var{gdb_jump_pad_head}, 8-byte long, is the
35501 head of @dfn{jumppad}, which is used to jump to data collection routine
35506 @item OK @var{target_address} @var{gdb_jump_pad_head} @var{fjump_size} @var{fjump}
35507 @var{target_address} is address of tracepoint in the inferior.
35508 The @var{gdb_jump_pad_head} is updated head of jumppad. Both of
35509 @var{target_address} and @var{gdb_jump_pad_head} are 8-byte long.
35510 The @var{fjump} contains a sequence of instructions jump to jumppad entry.
35511 The @var{fjump_size}, 4-byte long, is the size of @var{fjump}.
35518 Closes the in-process agent. This command is sent when @value{GDBN} or GDBserver
35519 is about to kill inferiors.
35527 @item probe_marker_at:@var{address}
35528 Asks in-process agent to probe the marker at @var{address}.
35535 @item unprobe_marker_at:@var{address}
35536 Asks in-process agent to unprobe the marker at @var{address}.
35540 @chapter Reporting Bugs in @value{GDBN}
35541 @cindex bugs in @value{GDBN}
35542 @cindex reporting bugs in @value{GDBN}
35544 Your bug reports play an essential role in making @value{GDBN} reliable.
35546 Reporting a bug may help you by bringing a solution to your problem, or it
35547 may not. But in any case the principal function of a bug report is to help
35548 the entire community by making the next version of @value{GDBN} work better. Bug
35549 reports are your contribution to the maintenance of @value{GDBN}.
35551 In order for a bug report to serve its purpose, you must include the
35552 information that enables us to fix the bug.
35555 * Bug Criteria:: Have you found a bug?
35556 * Bug Reporting:: How to report bugs
35560 @section Have You Found a Bug?
35561 @cindex bug criteria
35563 If you are not sure whether you have found a bug, here are some guidelines:
35566 @cindex fatal signal
35567 @cindex debugger crash
35568 @cindex crash of debugger
35570 If the debugger gets a fatal signal, for any input whatever, that is a
35571 @value{GDBN} bug. Reliable debuggers never crash.
35573 @cindex error on valid input
35575 If @value{GDBN} produces an error message for valid input, that is a
35576 bug. (Note that if you're cross debugging, the problem may also be
35577 somewhere in the connection to the target.)
35579 @cindex invalid input
35581 If @value{GDBN} does not produce an error message for invalid input,
35582 that is a bug. However, you should note that your idea of
35583 ``invalid input'' might be our idea of ``an extension'' or ``support
35584 for traditional practice''.
35587 If you are an experienced user of debugging tools, your suggestions
35588 for improvement of @value{GDBN} are welcome in any case.
35591 @node Bug Reporting
35592 @section How to Report Bugs
35593 @cindex bug reports
35594 @cindex @value{GDBN} bugs, reporting
35596 A number of companies and individuals offer support for @sc{gnu} products.
35597 If you obtained @value{GDBN} from a support organization, we recommend you
35598 contact that organization first.
35600 You can find contact information for many support companies and
35601 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
35603 @c should add a web page ref...
35606 @ifset BUGURL_DEFAULT
35607 In any event, we also recommend that you submit bug reports for
35608 @value{GDBN}. The preferred method is to submit them directly using
35609 @uref{http://www.gnu.org/software/gdb/bugs/, @value{GDBN}'s Bugs web
35610 page}. Alternatively, the @email{bug-gdb@@gnu.org, e-mail gateway} can
35613 @strong{Do not send bug reports to @samp{info-gdb}, or to
35614 @samp{help-gdb}, or to any newsgroups.} Most users of @value{GDBN} do
35615 not want to receive bug reports. Those that do have arranged to receive
35618 The mailing list @samp{bug-gdb} has a newsgroup @samp{gnu.gdb.bug} which
35619 serves as a repeater. The mailing list and the newsgroup carry exactly
35620 the same messages. Often people think of posting bug reports to the
35621 newsgroup instead of mailing them. This appears to work, but it has one
35622 problem which can be crucial: a newsgroup posting often lacks a mail
35623 path back to the sender. Thus, if we need to ask for more information,
35624 we may be unable to reach you. For this reason, it is better to send
35625 bug reports to the mailing list.
35627 @ifclear BUGURL_DEFAULT
35628 In any event, we also recommend that you submit bug reports for
35629 @value{GDBN} to @value{BUGURL}.
35633 The fundamental principle of reporting bugs usefully is this:
35634 @strong{report all the facts}. If you are not sure whether to state a
35635 fact or leave it out, state it!
35637 Often people omit facts because they think they know what causes the
35638 problem and assume that some details do not matter. Thus, you might
35639 assume that the name of the variable you use in an example does not matter.
35640 Well, probably it does not, but one cannot be sure. Perhaps the bug is a
35641 stray memory reference which happens to fetch from the location where that
35642 name is stored in memory; perhaps, if the name were different, the contents
35643 of that location would fool the debugger into doing the right thing despite
35644 the bug. Play it safe and give a specific, complete example. That is the
35645 easiest thing for you to do, and the most helpful.
35647 Keep in mind that the purpose of a bug report is to enable us to fix the
35648 bug. It may be that the bug has been reported previously, but neither
35649 you nor we can know that unless your bug report is complete and
35652 Sometimes people give a few sketchy facts and ask, ``Does this ring a
35653 bell?'' Those bug reports are useless, and we urge everyone to
35654 @emph{refuse to respond to them} except to chide the sender to report
35657 To enable us to fix the bug, you should include all these things:
35661 The version of @value{GDBN}. @value{GDBN} announces it if you start
35662 with no arguments; you can also print it at any time using @code{show
35665 Without this, we will not know whether there is any point in looking for
35666 the bug in the current version of @value{GDBN}.
35669 The type of machine you are using, and the operating system name and
35673 The details of the @value{GDBN} build-time configuration.
35674 @value{GDBN} shows these details if you invoke it with the
35675 @option{--configuration} command-line option, or if you type
35676 @code{show configuration} at @value{GDBN}'s prompt.
35679 What compiler (and its version) was used to compile @value{GDBN}---e.g.@:
35680 ``@value{GCC}--2.8.1''.
35683 What compiler (and its version) was used to compile the program you are
35684 debugging---e.g.@: ``@value{GCC}--2.8.1'', or ``HP92453-01 A.10.32.03 HP
35685 C Compiler''. For @value{NGCC}, you can say @kbd{@value{GCC} --version}
35686 to get this information; for other compilers, see the documentation for
35690 The command arguments you gave the compiler to compile your example and
35691 observe the bug. For example, did you use @samp{-O}? To guarantee
35692 you will not omit something important, list them all. A copy of the
35693 Makefile (or the output from make) is sufficient.
35695 If we were to try to guess the arguments, we would probably guess wrong
35696 and then we might not encounter the bug.
35699 A complete input script, and all necessary source files, that will
35703 A description of what behavior you observe that you believe is
35704 incorrect. For example, ``It gets a fatal signal.''
35706 Of course, if the bug is that @value{GDBN} gets a fatal signal, then we
35707 will certainly notice it. But if the bug is incorrect output, we might
35708 not notice unless it is glaringly wrong. You might as well not give us
35709 a chance to make a mistake.
35711 Even if the problem you experience is a fatal signal, you should still
35712 say so explicitly. Suppose something strange is going on, such as, your
35713 copy of @value{GDBN} is out of synch, or you have encountered a bug in
35714 the C library on your system. (This has happened!) Your copy might
35715 crash and ours would not. If you told us to expect a crash, then when
35716 ours fails to crash, we would know that the bug was not happening for
35717 us. If you had not told us to expect a crash, then we would not be able
35718 to draw any conclusion from our observations.
35721 @cindex recording a session script
35722 To collect all this information, you can use a session recording program
35723 such as @command{script}, which is available on many Unix systems.
35724 Just run your @value{GDBN} session inside @command{script} and then
35725 include the @file{typescript} file with your bug report.
35727 Another way to record a @value{GDBN} session is to run @value{GDBN}
35728 inside Emacs and then save the entire buffer to a file.
35731 If you wish to suggest changes to the @value{GDBN} source, send us context
35732 diffs. If you even discuss something in the @value{GDBN} source, refer to
35733 it by context, not by line number.
35735 The line numbers in our development sources will not match those in your
35736 sources. Your line numbers would convey no useful information to us.
35740 Here are some things that are not necessary:
35744 A description of the envelope of the bug.
35746 Often people who encounter a bug spend a lot of time investigating
35747 which changes to the input file will make the bug go away and which
35748 changes will not affect it.
35750 This is often time consuming and not very useful, because the way we
35751 will find the bug is by running a single example under the debugger
35752 with breakpoints, not by pure deduction from a series of examples.
35753 We recommend that you save your time for something else.
35755 Of course, if you can find a simpler example to report @emph{instead}
35756 of the original one, that is a convenience for us. Errors in the
35757 output will be easier to spot, running under the debugger will take
35758 less time, and so on.
35760 However, simplification is not vital; if you do not want to do this,
35761 report the bug anyway and send us the entire test case you used.
35764 A patch for the bug.
35766 A patch for the bug does help us if it is a good one. But do not omit
35767 the necessary information, such as the test case, on the assumption that
35768 a patch is all we need. We might see problems with your patch and decide
35769 to fix the problem another way, or we might not understand it at all.
35771 Sometimes with a program as complicated as @value{GDBN} it is very hard to
35772 construct an example that will make the program follow a certain path
35773 through the code. If you do not send us the example, we will not be able
35774 to construct one, so we will not be able to verify that the bug is fixed.
35776 And if we cannot understand what bug you are trying to fix, or why your
35777 patch should be an improvement, we will not install it. A test case will
35778 help us to understand.
35781 A guess about what the bug is or what it depends on.
35783 Such guesses are usually wrong. Even we cannot guess right about such
35784 things without first using the debugger to find the facts.
35787 @c The readline documentation is distributed with the readline code
35788 @c and consists of the two following files:
35791 @c Use -I with makeinfo to point to the appropriate directory,
35792 @c environment var TEXINPUTS with TeX.
35793 @ifclear SYSTEM_READLINE
35794 @include rluser.texi
35795 @include hsuser.texi
35799 @appendix In Memoriam
35801 The @value{GDBN} project mourns the loss of the following long-time
35806 Fred was a long-standing contributor to @value{GDBN} (1991-2006), and
35807 to Free Software in general. Outside of @value{GDBN}, he was known in
35808 the Amiga world for his series of Fish Disks, and the GeekGadget project.
35810 @item Michael Snyder
35811 Michael was one of the Global Maintainers of the @value{GDBN} project,
35812 with contributions recorded as early as 1996, until 2011. In addition
35813 to his day to day participation, he was a large driving force behind
35814 adding Reverse Debugging to @value{GDBN}.
35817 Beyond their technical contributions to the project, they were also
35818 enjoyable members of the Free Software Community. We will miss them.
35820 @node Formatting Documentation
35821 @appendix Formatting Documentation
35823 @cindex @value{GDBN} reference card
35824 @cindex reference card
35825 The @value{GDBN} 4 release includes an already-formatted reference card, ready
35826 for printing with PostScript or Ghostscript, in the @file{gdb}
35827 subdirectory of the main source directory@footnote{In
35828 @file{gdb-@value{GDBVN}/gdb/refcard.ps} of the version @value{GDBVN}
35829 release.}. If you can use PostScript or Ghostscript with your printer,
35830 you can print the reference card immediately with @file{refcard.ps}.
35832 The release also includes the source for the reference card. You
35833 can format it, using @TeX{}, by typing:
35839 The @value{GDBN} reference card is designed to print in @dfn{landscape}
35840 mode on US ``letter'' size paper;
35841 that is, on a sheet 11 inches wide by 8.5 inches
35842 high. You will need to specify this form of printing as an option to
35843 your @sc{dvi} output program.
35845 @cindex documentation
35847 All the documentation for @value{GDBN} comes as part of the machine-readable
35848 distribution. The documentation is written in Texinfo format, which is
35849 a documentation system that uses a single source file to produce both
35850 on-line information and a printed manual. You can use one of the Info
35851 formatting commands to create the on-line version of the documentation
35852 and @TeX{} (or @code{texi2roff}) to typeset the printed version.
35854 @value{GDBN} includes an already formatted copy of the on-line Info
35855 version of this manual in the @file{gdb} subdirectory. The main Info
35856 file is @file{gdb-@value{GDBVN}/gdb/gdb.info}, and it refers to
35857 subordinate files matching @samp{gdb.info*} in the same directory. If
35858 necessary, you can print out these files, or read them with any editor;
35859 but they are easier to read using the @code{info} subsystem in @sc{gnu}
35860 Emacs or the standalone @code{info} program, available as part of the
35861 @sc{gnu} Texinfo distribution.
35863 If you want to format these Info files yourself, you need one of the
35864 Info formatting programs, such as @code{texinfo-format-buffer} or
35867 If you have @code{makeinfo} installed, and are in the top level
35868 @value{GDBN} source directory (@file{gdb-@value{GDBVN}}, in the case of
35869 version @value{GDBVN}), you can make the Info file by typing:
35876 If you want to typeset and print copies of this manual, you need @TeX{},
35877 a program to print its @sc{dvi} output files, and @file{texinfo.tex}, the
35878 Texinfo definitions file.
35880 @TeX{} is a typesetting program; it does not print files directly, but
35881 produces output files called @sc{dvi} files. To print a typeset
35882 document, you need a program to print @sc{dvi} files. If your system
35883 has @TeX{} installed, chances are it has such a program. The precise
35884 command to use depends on your system; @kbd{lpr -d} is common; another
35885 (for PostScript devices) is @kbd{dvips}. The @sc{dvi} print command may
35886 require a file name without any extension or a @samp{.dvi} extension.
35888 @TeX{} also requires a macro definitions file called
35889 @file{texinfo.tex}. This file tells @TeX{} how to typeset a document
35890 written in Texinfo format. On its own, @TeX{} cannot either read or
35891 typeset a Texinfo file. @file{texinfo.tex} is distributed with GDB
35892 and is located in the @file{gdb-@var{version-number}/texinfo}
35895 If you have @TeX{} and a @sc{dvi} printer program installed, you can
35896 typeset and print this manual. First switch to the @file{gdb}
35897 subdirectory of the main source directory (for example, to
35898 @file{gdb-@value{GDBVN}/gdb}) and type:
35904 Then give @file{gdb.dvi} to your @sc{dvi} printing program.
35906 @node Installing GDB
35907 @appendix Installing @value{GDBN}
35908 @cindex installation
35911 * Requirements:: Requirements for building @value{GDBN}
35912 * Running Configure:: Invoking the @value{GDBN} @file{configure} script
35913 * Separate Objdir:: Compiling @value{GDBN} in another directory
35914 * Config Names:: Specifying names for hosts and targets
35915 * Configure Options:: Summary of options for configure
35916 * System-wide configuration:: Having a system-wide init file
35920 @section Requirements for Building @value{GDBN}
35921 @cindex building @value{GDBN}, requirements for
35923 Building @value{GDBN} requires various tools and packages to be available.
35924 Other packages will be used only if they are found.
35926 @heading Tools/Packages Necessary for Building @value{GDBN}
35928 @item C@t{++}11 compiler
35929 @value{GDBN} is written in C@t{++}11. It should be buildable with any
35930 recent C@t{++}11 compiler, e.g.@: GCC.
35933 @value{GDBN}'s build system relies on features only found in the GNU
35934 make program. Other variants of @code{make} will not work.
35937 @heading Tools/Packages Optional for Building @value{GDBN}
35941 @value{GDBN} can use the Expat XML parsing library. This library may be
35942 included with your operating system distribution; if it is not, you
35943 can get the latest version from @url{http://expat.sourceforge.net}.
35944 The @file{configure} script will search for this library in several
35945 standard locations; if it is installed in an unusual path, you can
35946 use the @option{--with-libexpat-prefix} option to specify its location.
35952 Remote protocol memory maps (@pxref{Memory Map Format})
35954 Target descriptions (@pxref{Target Descriptions})
35956 Remote shared library lists (@xref{Library List Format},
35957 or alternatively @pxref{Library List Format for SVR4 Targets})
35959 MS-Windows shared libraries (@pxref{Shared Libraries})
35961 Traceframe info (@pxref{Traceframe Info Format})
35963 Branch trace (@pxref{Branch Trace Format},
35964 @pxref{Branch Trace Configuration Format})
35968 @value{GDBN} can be scripted using GNU Guile. @xref{Guile}. By
35969 default, @value{GDBN} will be compiled if the Guile libraries are
35970 installed and are found by @file{configure}. You can use the
35971 @code{--with-guile} option to request Guile, and pass either the Guile
35972 version number or the file name of the relevant @code{pkg-config}
35973 program to choose a particular version of Guile.
35976 @value{GDBN}'s features related to character sets (@pxref{Character
35977 Sets}) require a functioning @code{iconv} implementation. If you are
35978 on a GNU system, then this is provided by the GNU C Library. Some
35979 other systems also provide a working @code{iconv}.
35981 If @value{GDBN} is using the @code{iconv} program which is installed
35982 in a non-standard place, you will need to tell @value{GDBN} where to
35983 find it. This is done with @option{--with-iconv-bin} which specifies
35984 the directory that contains the @code{iconv} program. This program is
35985 run in order to make a list of the available character sets.
35987 On systems without @code{iconv}, you can install GNU Libiconv. If
35988 Libiconv is installed in a standard place, @value{GDBN} will
35989 automatically use it if it is needed. If you have previously
35990 installed Libiconv in a non-standard place, you can use the
35991 @option{--with-libiconv-prefix} option to @file{configure}.
35993 @value{GDBN}'s top-level @file{configure} and @file{Makefile} will
35994 arrange to build Libiconv if a directory named @file{libiconv} appears
35995 in the top-most source directory. If Libiconv is built this way, and
35996 if the operating system does not provide a suitable @code{iconv}
35997 implementation, then the just-built library will automatically be used
35998 by @value{GDBN}. One easy way to set this up is to download GNU
35999 Libiconv, unpack it inside the top-level directory of the @value{GDBN}
36000 source tree, and then rename the directory holding the Libiconv source
36001 code to @samp{libiconv}.
36004 @value{GDBN} can support debugging sections that are compressed with
36005 the LZMA library. @xref{MiniDebugInfo}. If this library is not
36006 included with your operating system, you can find it in the xz package
36007 at @url{http://tukaani.org/xz/}. If the LZMA library is available in
36008 the usual place, then the @file{configure} script will use it
36009 automatically. If it is installed in an unusual path, you can use the
36010 @option{--with-lzma-prefix} option to specify its location.
36014 @value{GDBN} can use the GNU MPFR multiple-precision floating-point
36015 library. This library may be included with your operating system
36016 distribution; if it is not, you can get the latest version from
36017 @url{http://www.mpfr.org}. The @file{configure} script will search
36018 for this library in several standard locations; if it is installed
36019 in an unusual path, you can use the @option{--with-libmpfr-prefix}
36020 option to specify its location.
36022 GNU MPFR is used to emulate target floating-point arithmetic during
36023 expression evaluation when the target uses different floating-point
36024 formats than the host. If GNU MPFR it is not available, @value{GDBN}
36025 will fall back to using host floating-point arithmetic.
36028 @value{GDBN} can be scripted using Python language. @xref{Python}.
36029 By default, @value{GDBN} will be compiled if the Python libraries are
36030 installed and are found by @file{configure}. You can use the
36031 @code{--with-python} option to request Python, and pass either the
36032 file name of the relevant @code{python} executable, or the name of the
36033 directory in which Python is installed, to choose a particular
36034 installation of Python.
36037 @cindex compressed debug sections
36038 @value{GDBN} will use the @samp{zlib} library, if available, to read
36039 compressed debug sections. Some linkers, such as GNU gold, are capable
36040 of producing binaries with compressed debug sections. If @value{GDBN}
36041 is compiled with @samp{zlib}, it will be able to read the debug
36042 information in such binaries.
36044 The @samp{zlib} library is likely included with your operating system
36045 distribution; if it is not, you can get the latest version from
36046 @url{http://zlib.net}.
36049 @node Running Configure
36050 @section Invoking the @value{GDBN} @file{configure} Script
36051 @cindex configuring @value{GDBN}
36052 @value{GDBN} comes with a @file{configure} script that automates the process
36053 of preparing @value{GDBN} for installation; you can then use @code{make} to
36054 build the @code{gdb} program.
36056 @c irrelevant in info file; it's as current as the code it lives with.
36057 @footnote{If you have a more recent version of @value{GDBN} than @value{GDBVN},
36058 look at the @file{README} file in the sources; we may have improved the
36059 installation procedures since publishing this manual.}
36062 The @value{GDBN} distribution includes all the source code you need for
36063 @value{GDBN} in a single directory, whose name is usually composed by
36064 appending the version number to @samp{gdb}.
36066 For example, the @value{GDBN} version @value{GDBVN} distribution is in the
36067 @file{gdb-@value{GDBVN}} directory. That directory contains:
36070 @item gdb-@value{GDBVN}/configure @r{(and supporting files)}
36071 script for configuring @value{GDBN} and all its supporting libraries
36073 @item gdb-@value{GDBVN}/gdb
36074 the source specific to @value{GDBN} itself
36076 @item gdb-@value{GDBVN}/bfd
36077 source for the Binary File Descriptor library
36079 @item gdb-@value{GDBVN}/include
36080 @sc{gnu} include files
36082 @item gdb-@value{GDBVN}/libiberty
36083 source for the @samp{-liberty} free software library
36085 @item gdb-@value{GDBVN}/opcodes
36086 source for the library of opcode tables and disassemblers
36088 @item gdb-@value{GDBVN}/readline
36089 source for the @sc{gnu} command-line interface
36092 There may be other subdirectories as well.
36094 The simplest way to configure and build @value{GDBN} is to run @file{configure}
36095 from the @file{gdb-@var{version-number}} source directory, which in
36096 this example is the @file{gdb-@value{GDBVN}} directory.
36098 First switch to the @file{gdb-@var{version-number}} source directory
36099 if you are not already in it; then run @file{configure}. Pass the
36100 identifier for the platform on which @value{GDBN} will run as an
36106 cd gdb-@value{GDBVN}
36111 Running @samp{configure} and then running @code{make} builds the
36112 included supporting libraries, then @code{gdb} itself. The configured
36113 source files, and the binaries, are left in the corresponding source
36117 @file{configure} is a Bourne-shell (@code{/bin/sh}) script; if your
36118 system does not recognize this automatically when you run a different
36119 shell, you may need to run @code{sh} on it explicitly:
36125 You should run the @file{configure} script from the top directory in the
36126 source tree, the @file{gdb-@var{version-number}} directory. If you run
36127 @file{configure} from one of the subdirectories, you will configure only
36128 that subdirectory. That is usually not what you want. In particular,
36129 if you run the first @file{configure} from the @file{gdb} subdirectory
36130 of the @file{gdb-@var{version-number}} directory, you will omit the
36131 configuration of @file{bfd}, @file{readline}, and other sibling
36132 directories of the @file{gdb} subdirectory. This leads to build errors
36133 about missing include files such as @file{bfd/bfd.h}.
36135 You can install @code{@value{GDBN}} anywhere. The best way to do this
36136 is to pass the @code{--prefix} option to @code{configure}, and then
36137 install it with @code{make install}.
36139 @node Separate Objdir
36140 @section Compiling @value{GDBN} in Another Directory
36142 If you want to run @value{GDBN} versions for several host or target machines,
36143 you need a different @code{gdb} compiled for each combination of
36144 host and target. @file{configure} is designed to make this easy by
36145 allowing you to generate each configuration in a separate subdirectory,
36146 rather than in the source directory. If your @code{make} program
36147 handles the @samp{VPATH} feature (@sc{gnu} @code{make} does), running
36148 @code{make} in each of these directories builds the @code{gdb}
36149 program specified there.
36151 To build @code{gdb} in a separate directory, run @file{configure}
36152 with the @samp{--srcdir} option to specify where to find the source.
36153 (You also need to specify a path to find @file{configure}
36154 itself from your working directory. If the path to @file{configure}
36155 would be the same as the argument to @samp{--srcdir}, you can leave out
36156 the @samp{--srcdir} option; it is assumed.)
36158 For example, with version @value{GDBVN}, you can build @value{GDBN} in a
36159 separate directory for a Sun 4 like this:
36163 cd gdb-@value{GDBVN}
36166 ../gdb-@value{GDBVN}/configure
36171 When @file{configure} builds a configuration using a remote source
36172 directory, it creates a tree for the binaries with the same structure
36173 (and using the same names) as the tree under the source directory. In
36174 the example, you'd find the Sun 4 library @file{libiberty.a} in the
36175 directory @file{gdb-sun4/libiberty}, and @value{GDBN} itself in
36176 @file{gdb-sun4/gdb}.
36178 Make sure that your path to the @file{configure} script has just one
36179 instance of @file{gdb} in it. If your path to @file{configure} looks
36180 like @file{../gdb-@value{GDBVN}/gdb/configure}, you are configuring only
36181 one subdirectory of @value{GDBN}, not the whole package. This leads to
36182 build errors about missing include files such as @file{bfd/bfd.h}.
36184 One popular reason to build several @value{GDBN} configurations in separate
36185 directories is to configure @value{GDBN} for cross-compiling (where
36186 @value{GDBN} runs on one machine---the @dfn{host}---while debugging
36187 programs that run on another machine---the @dfn{target}).
36188 You specify a cross-debugging target by
36189 giving the @samp{--target=@var{target}} option to @file{configure}.
36191 When you run @code{make} to build a program or library, you must run
36192 it in a configured directory---whatever directory you were in when you
36193 called @file{configure} (or one of its subdirectories).
36195 The @code{Makefile} that @file{configure} generates in each source
36196 directory also runs recursively. If you type @code{make} in a source
36197 directory such as @file{gdb-@value{GDBVN}} (or in a separate configured
36198 directory configured with @samp{--srcdir=@var{dirname}/gdb-@value{GDBVN}}), you
36199 will build all the required libraries, and then build GDB.
36201 When you have multiple hosts or targets configured in separate
36202 directories, you can run @code{make} on them in parallel (for example,
36203 if they are NFS-mounted on each of the hosts); they will not interfere
36207 @section Specifying Names for Hosts and Targets
36209 The specifications used for hosts and targets in the @file{configure}
36210 script are based on a three-part naming scheme, but some short predefined
36211 aliases are also supported. The full naming scheme encodes three pieces
36212 of information in the following pattern:
36215 @var{architecture}-@var{vendor}-@var{os}
36218 For example, you can use the alias @code{sun4} as a @var{host} argument,
36219 or as the value for @var{target} in a @code{--target=@var{target}}
36220 option. The equivalent full name is @samp{sparc-sun-sunos4}.
36222 The @file{configure} script accompanying @value{GDBN} does not provide
36223 any query facility to list all supported host and target names or
36224 aliases. @file{configure} calls the Bourne shell script
36225 @code{config.sub} to map abbreviations to full names; you can read the
36226 script, if you wish, or you can use it to test your guesses on
36227 abbreviations---for example:
36230 % sh config.sub i386-linux
36232 % sh config.sub alpha-linux
36233 alpha-unknown-linux-gnu
36234 % sh config.sub hp9k700
36236 % sh config.sub sun4
36237 sparc-sun-sunos4.1.1
36238 % sh config.sub sun3
36239 m68k-sun-sunos4.1.1
36240 % sh config.sub i986v
36241 Invalid configuration `i986v': machine `i986v' not recognized
36245 @code{config.sub} is also distributed in the @value{GDBN} source
36246 directory (@file{gdb-@value{GDBVN}}, for version @value{GDBVN}).
36248 @node Configure Options
36249 @section @file{configure} Options
36251 Here is a summary of the @file{configure} options and arguments that
36252 are most often useful for building @value{GDBN}. @file{configure}
36253 also has several other options not listed here. @inforef{Running
36254 configure scripts,,autoconf.info}, for a full
36255 explanation of @file{configure}.
36258 configure @r{[}--help@r{]}
36259 @r{[}--prefix=@var{dir}@r{]}
36260 @r{[}--exec-prefix=@var{dir}@r{]}
36261 @r{[}--srcdir=@var{dirname}@r{]}
36262 @r{[}--target=@var{target}@r{]}
36266 You may introduce options with a single @samp{-} rather than
36267 @samp{--} if you prefer; but you may abbreviate option names if you use
36272 Display a quick summary of how to invoke @file{configure}.
36274 @item --prefix=@var{dir}
36275 Configure the source to install programs and files under directory
36278 @item --exec-prefix=@var{dir}
36279 Configure the source to install programs under directory
36282 @c avoid splitting the warning from the explanation:
36284 @item --srcdir=@var{dirname}
36285 Use this option to make configurations in directories separate from the
36286 @value{GDBN} source directories. Among other things, you can use this to
36287 build (or maintain) several configurations simultaneously, in separate
36288 directories. @file{configure} writes configuration-specific files in
36289 the current directory, but arranges for them to use the source in the
36290 directory @var{dirname}. @file{configure} creates directories under
36291 the working directory in parallel to the source directories below
36294 @item --target=@var{target}
36295 Configure @value{GDBN} for cross-debugging programs running on the specified
36296 @var{target}. Without this option, @value{GDBN} is configured to debug
36297 programs that run on the same machine (@var{host}) as @value{GDBN} itself.
36299 There is no convenient way to generate a list of all available
36300 targets. Also see the @code{--enable-targets} option, below.
36303 There are many other options that are specific to @value{GDBN}. This
36304 lists just the most common ones; there are some very specialized
36305 options not described here.
36308 @item --enable-targets=@r{[}@var{target}@r{]}@dots{}
36309 @itemx --enable-targets=all
36310 Configure @value{GDBN} for cross-debugging programs running on the
36311 specified list of targets. The special value @samp{all} configures
36312 @value{GDBN} for debugging programs running on any target it supports.
36314 @item --with-gdb-datadir=@var{path}
36315 Set the @value{GDBN}-specific data directory. @value{GDBN} will look
36316 here for certain supporting files or scripts. This defaults to the
36317 @file{gdb} subdirectory of @samp{datadi} (which can be set using
36320 @item --with-relocated-sources=@var{dir}
36321 Sets up the default source path substitution rule so that directory
36322 names recorded in debug information will be automatically adjusted for
36323 any directory under @var{dir}. @var{dir} should be a subdirectory of
36324 @value{GDBN}'s configured prefix, the one mentioned in the
36325 @code{--prefix} or @code{--exec-prefix} options to configure. This
36326 option is useful if GDB is supposed to be moved to a different place
36329 @item --enable-64-bit-bfd
36330 Enable 64-bit support in BFD on 32-bit hosts.
36332 @item --disable-gdbmi
36333 Build @value{GDBN} without the GDB/MI machine interface
36337 Build @value{GDBN} with the text-mode full-screen user interface
36338 (TUI). Requires a curses library (ncurses and cursesX are also
36341 @item --with-curses
36342 Use the curses library instead of the termcap library, for text-mode
36343 terminal operations.
36345 @item --with-libunwind-ia64
36346 Use the libunwind library for unwinding function call stack on ia64
36347 target platforms. See http://www.nongnu.org/libunwind/index.html for
36350 @item --with-system-readline
36351 Use the readline library installed on the host, rather than the
36352 library supplied as part of @value{GDBN}.
36354 @item --with-system-zlib
36355 Use the zlib library installed on the host, rather than the library
36356 supplied as part of @value{GDBN}.
36359 Build @value{GDBN} with Expat, a library for XML parsing. (Done by
36360 default if libexpat is installed and found at configure time.) This
36361 library is used to read XML files supplied with @value{GDBN}. If it
36362 is unavailable, some features, such as remote protocol memory maps,
36363 target descriptions, and shared library lists, that are based on XML
36364 files, will not be available in @value{GDBN}. If your host does not
36365 have libexpat installed, you can get the latest version from
36366 `http://expat.sourceforge.net'.
36368 @item --with-libiconv-prefix@r{[}=@var{dir}@r{]}
36370 Build @value{GDBN} with GNU libiconv, a character set encoding
36371 conversion library. This is not done by default, as on GNU systems
36372 the @code{iconv} that is built in to the C library is sufficient. If
36373 your host does not have a working @code{iconv}, you can get the latest
36374 version of GNU iconv from `https://www.gnu.org/software/libiconv/'.
36376 @value{GDBN}'s build system also supports building GNU libiconv as
36377 part of the overall build. @xref{Requirements}.
36380 Build @value{GDBN} with LZMA, a compression library. (Done by default
36381 if liblzma is installed and found at configure time.) LZMA is used by
36382 @value{GDBN}'s "mini debuginfo" feature, which is only useful on
36383 platforms using the ELF object file format. If your host does not
36384 have liblzma installed, you can get the latest version from
36385 `https://tukaani.org/xz/'.
36388 Build @value{GDBN} with GNU MPFR, a library for multiple-precision
36389 floating-point computation with correct rounding. (Done by default if
36390 GNU MPFR is installed and found at configure time.) This library is
36391 used to emulate target floating-point arithmetic during expression
36392 evaluation when the target uses different floating-point formats than
36393 the host. If GNU MPFR is not available, @value{GDBN} will fall back
36394 to using host floating-point arithmetic. If your host does not have
36395 GNU MPFR installed, you can get the latest version from
36396 `http://www.mpfr.org'.
36398 @item --with-python@r{[}=@var{python}@r{]}
36399 Build @value{GDBN} with Python scripting support. (Done by default if
36400 libpython is present and found at configure time.) Python makes
36401 @value{GDBN} scripting much more powerful than the restricted CLI
36402 scripting language. If your host does not have Python installed, you
36403 can find it on `http://www.python.org/download/'. The oldest version
36404 of Python supported by GDB is 2.6. The optional argument @var{python}
36405 is used to find the Python headers and libraries. It can be either
36406 the name of a Python executable, or the name of the directory in which
36407 Python is installed.
36409 @item --with-guile[=GUILE]'
36410 Build @value{GDBN} with GNU Guile scripting support. (Done by default
36411 if libguile is present and found at configure time.) If your host
36412 does not have Guile installed, you can find it at
36413 `https://www.gnu.org/software/guile/'. The optional argument GUILE
36414 can be a version number, which will cause @code{configure} to try to
36415 use that version of Guile; or the file name of a @code{pkg-config}
36416 executable, which will be queried to find the information needed to
36417 compile and link against Guile.
36419 @item --without-included-regex
36420 Don't use the regex library included with @value{GDBN} (as part of the
36421 libiberty library). This is the default on hosts with version 2 of
36424 @item --with-sysroot=@var{dir}
36425 Use @var{dir} as the default system root directory for libraries whose
36426 file names begin with @file{/lib}' or @file{/usr/lib'}. (The value of
36427 @var{dir} can be modified at run time by using the @command{set
36428 sysroot} command.) If @var{dir} is under the @value{GDBN} configured
36429 prefix (set with @code{--prefix} or @code{--exec-prefix options}, the
36430 default system root will be automatically adjusted if and when
36431 @value{GDBN} is moved to a different location.
36433 @item --with-system-gdbinit=@var{file}
36434 Configure @value{GDBN} to automatically load a system-wide init file.
36435 @var{file} should be an absolute file name. If @var{file} is in a
36436 directory under the configured prefix, and @value{GDBN} is moved to
36437 another location after being built, the location of the system-wide
36438 init file will be adjusted accordingly.
36440 @item --enable-build-warnings
36441 When building the @value{GDBN} sources, ask the compiler to warn about
36442 any code which looks even vaguely suspicious. It passes many
36443 different warning flags, depending on the exact version of the
36444 compiler you are using.
36446 @item --enable-werror
36447 Treat compiler warnings as werrors. It adds the @code{-Werror} flag
36448 to the compiler, which will fail the compilation if the compiler
36449 outputs any warning messages.
36451 @item --enable-ubsan
36452 Enable the GCC undefined behavior sanitizer. This is disabled by
36453 default, but passing @code{--enable-ubsan=yes} or
36454 @code{--enable-ubsan=auto} to @code{configure} will enable it. The
36455 undefined behavior sanitizer checks for C@t{++} undefined behavior.
36456 It has a performance cost, so if you are looking at @value{GDBN}'s
36457 performance, you should disable it. The undefined behavior sanitizer
36458 was first introduced in GCC 4.9.
36461 @node System-wide configuration
36462 @section System-wide configuration and settings
36463 @cindex system-wide init file
36465 @value{GDBN} can be configured to have a system-wide init file;
36466 this file will be read and executed at startup (@pxref{Startup, , What
36467 @value{GDBN} does during startup}).
36469 Here is the corresponding configure option:
36472 @item --with-system-gdbinit=@var{file}
36473 Specify that the default location of the system-wide init file is
36477 If @value{GDBN} has been configured with the option @option{--prefix=$prefix},
36478 it may be subject to relocation. Two possible cases:
36482 If the default location of this init file contains @file{$prefix},
36483 it will be subject to relocation. Suppose that the configure options
36484 are @option{--prefix=$prefix --with-system-gdbinit=$prefix/etc/gdbinit};
36485 if @value{GDBN} is moved from @file{$prefix} to @file{$install}, the system
36486 init file is looked for as @file{$install/etc/gdbinit} instead of
36487 @file{$prefix/etc/gdbinit}.
36490 By contrast, if the default location does not contain the prefix,
36491 it will not be relocated. E.g.@: if @value{GDBN} has been configured with
36492 @option{--prefix=/usr/local --with-system-gdbinit=/usr/share/gdb/gdbinit},
36493 then @value{GDBN} will always look for @file{/usr/share/gdb/gdbinit},
36494 wherever @value{GDBN} is installed.
36497 If the configured location of the system-wide init file (as given by the
36498 @option{--with-system-gdbinit} option at configure time) is in the
36499 data-directory (as specified by @option{--with-gdb-datadir} at configure
36500 time) or in one of its subdirectories, then @value{GDBN} will look for the
36501 system-wide init file in the directory specified by the
36502 @option{--data-directory} command-line option.
36503 Note that the system-wide init file is only read once, during @value{GDBN}
36504 initialization. If the data-directory is changed after @value{GDBN} has
36505 started with the @code{set data-directory} command, the file will not be
36509 * System-wide Configuration Scripts:: Installed System-wide Configuration Scripts
36512 @node System-wide Configuration Scripts
36513 @subsection Installed System-wide Configuration Scripts
36514 @cindex system-wide configuration scripts
36516 The @file{system-gdbinit} directory, located inside the data-directory
36517 (as specified by @option{--with-gdb-datadir} at configure time) contains
36518 a number of scripts which can be used as system-wide init files. To
36519 automatically source those scripts at startup, @value{GDBN} should be
36520 configured with @option{--with-system-gdbinit}. Otherwise, any user
36521 should be able to source them by hand as needed.
36523 The following scripts are currently available:
36526 @item @file{elinos.py}
36528 @cindex ELinOS system-wide configuration script
36529 This script is useful when debugging a program on an ELinOS target.
36530 It takes advantage of the environment variables defined in a standard
36531 ELinOS environment in order to determine the location of the system
36532 shared libraries, and then sets the @samp{solib-absolute-prefix}
36533 and @samp{solib-search-path} variables appropriately.
36535 @item @file{wrs-linux.py}
36536 @pindex wrs-linux.py
36537 @cindex Wind River Linux system-wide configuration script
36538 This script is useful when debugging a program on a target running
36539 Wind River Linux. It expects the @env{ENV_PREFIX} to be set to
36540 the host-side sysroot used by the target system.
36544 @node Maintenance Commands
36545 @appendix Maintenance Commands
36546 @cindex maintenance commands
36547 @cindex internal commands
36549 In addition to commands intended for @value{GDBN} users, @value{GDBN}
36550 includes a number of commands intended for @value{GDBN} developers,
36551 that are not documented elsewhere in this manual. These commands are
36552 provided here for reference. (For commands that turn on debugging
36553 messages, see @ref{Debugging Output}.)
36556 @kindex maint agent
36557 @kindex maint agent-eval
36558 @item maint agent @r{[}-at @var{location}@r{,}@r{]} @var{expression}
36559 @itemx maint agent-eval @r{[}-at @var{location}@r{,}@r{]} @var{expression}
36560 Translate the given @var{expression} into remote agent bytecodes.
36561 This command is useful for debugging the Agent Expression mechanism
36562 (@pxref{Agent Expressions}). The @samp{agent} version produces an
36563 expression useful for data collection, such as by tracepoints, while
36564 @samp{maint agent-eval} produces an expression that evaluates directly
36565 to a result. For instance, a collection expression for @code{globa +
36566 globb} will include bytecodes to record four bytes of memory at each
36567 of the addresses of @code{globa} and @code{globb}, while discarding
36568 the result of the addition, while an evaluation expression will do the
36569 addition and return the sum.
36570 If @code{-at} is given, generate remote agent bytecode for @var{location}.
36571 If not, generate remote agent bytecode for current frame PC address.
36573 @kindex maint agent-printf
36574 @item maint agent-printf @var{format},@var{expr},...
36575 Translate the given format string and list of argument expressions
36576 into remote agent bytecodes and display them as a disassembled list.
36577 This command is useful for debugging the agent version of dynamic
36578 printf (@pxref{Dynamic Printf}).
36580 @kindex maint info breakpoints
36581 @item @anchor{maint info breakpoints}maint info breakpoints
36582 Using the same format as @samp{info breakpoints}, display both the
36583 breakpoints you've set explicitly, and those @value{GDBN} is using for
36584 internal purposes. Internal breakpoints are shown with negative
36585 breakpoint numbers. The type column identifies what kind of breakpoint
36590 Normal, explicitly set breakpoint.
36593 Normal, explicitly set watchpoint.
36596 Internal breakpoint, used to handle correctly stepping through
36597 @code{longjmp} calls.
36599 @item longjmp resume
36600 Internal breakpoint at the target of a @code{longjmp}.
36603 Temporary internal breakpoint used by the @value{GDBN} @code{until} command.
36606 Temporary internal breakpoint used by the @value{GDBN} @code{finish} command.
36609 Shared library events.
36613 @kindex maint info btrace
36614 @item maint info btrace
36615 Pint information about raw branch tracing data.
36617 @kindex maint btrace packet-history
36618 @item maint btrace packet-history
36619 Print the raw branch trace packets that are used to compute the
36620 execution history for the @samp{record btrace} command. Both the
36621 information and the format in which it is printed depend on the btrace
36626 For the BTS recording format, print a list of blocks of sequential
36627 code. For each block, the following information is printed:
36631 Newer blocks have higher numbers. The oldest block has number zero.
36632 @item Lowest @samp{PC}
36633 @item Highest @samp{PC}
36637 For the Intel Processor Trace recording format, print a list of
36638 Intel Processor Trace packets. For each packet, the following
36639 information is printed:
36642 @item Packet number
36643 Newer packets have higher numbers. The oldest packet has number zero.
36645 The packet's offset in the trace stream.
36646 @item Packet opcode and payload
36650 @kindex maint btrace clear-packet-history
36651 @item maint btrace clear-packet-history
36652 Discards the cached packet history printed by the @samp{maint btrace
36653 packet-history} command. The history will be computed again when
36656 @kindex maint btrace clear
36657 @item maint btrace clear
36658 Discard the branch trace data. The data will be fetched anew and the
36659 branch trace will be recomputed when needed.
36661 This implicitly truncates the branch trace to a single branch trace
36662 buffer. When updating branch trace incrementally, the branch trace
36663 available to @value{GDBN} may be bigger than a single branch trace
36666 @kindex maint set btrace pt skip-pad
36667 @item maint set btrace pt skip-pad
36668 @kindex maint show btrace pt skip-pad
36669 @item maint show btrace pt skip-pad
36670 Control whether @value{GDBN} will skip PAD packets when computing the
36673 @kindex set displaced-stepping
36674 @kindex show displaced-stepping
36675 @cindex displaced stepping support
36676 @cindex out-of-line single-stepping
36677 @item set displaced-stepping
36678 @itemx show displaced-stepping
36679 Control whether or not @value{GDBN} will do @dfn{displaced stepping}
36680 if the target supports it. Displaced stepping is a way to single-step
36681 over breakpoints without removing them from the inferior, by executing
36682 an out-of-line copy of the instruction that was originally at the
36683 breakpoint location. It is also known as out-of-line single-stepping.
36686 @item set displaced-stepping on
36687 If the target architecture supports it, @value{GDBN} will use
36688 displaced stepping to step over breakpoints.
36690 @item set displaced-stepping off
36691 @value{GDBN} will not use displaced stepping to step over breakpoints,
36692 even if such is supported by the target architecture.
36694 @cindex non-stop mode, and @samp{set displaced-stepping}
36695 @item set displaced-stepping auto
36696 This is the default mode. @value{GDBN} will use displaced stepping
36697 only if non-stop mode is active (@pxref{Non-Stop Mode}) and the target
36698 architecture supports displaced stepping.
36701 @kindex maint check-psymtabs
36702 @item maint check-psymtabs
36703 Check the consistency of currently expanded psymtabs versus symtabs.
36704 Use this to check, for example, whether a symbol is in one but not the other.
36706 @kindex maint check-symtabs
36707 @item maint check-symtabs
36708 Check the consistency of currently expanded symtabs.
36710 @kindex maint expand-symtabs
36711 @item maint expand-symtabs [@var{regexp}]
36712 Expand symbol tables.
36713 If @var{regexp} is specified, only expand symbol tables for file
36714 names matching @var{regexp}.
36716 @kindex maint set catch-demangler-crashes
36717 @kindex maint show catch-demangler-crashes
36718 @cindex demangler crashes
36719 @item maint set catch-demangler-crashes [on|off]
36720 @itemx maint show catch-demangler-crashes
36721 Control whether @value{GDBN} should attempt to catch crashes in the
36722 symbol name demangler. The default is to attempt to catch crashes.
36723 If enabled, the first time a crash is caught, a core file is created,
36724 the offending symbol is displayed and the user is presented with the
36725 option to terminate the current session.
36727 @kindex maint cplus first_component
36728 @item maint cplus first_component @var{name}
36729 Print the first C@t{++} class/namespace component of @var{name}.
36731 @kindex maint cplus namespace
36732 @item maint cplus namespace
36733 Print the list of possible C@t{++} namespaces.
36735 @kindex maint deprecate
36736 @kindex maint undeprecate
36737 @cindex deprecated commands
36738 @item maint deprecate @var{command} @r{[}@var{replacement}@r{]}
36739 @itemx maint undeprecate @var{command}
36740 Deprecate or undeprecate the named @var{command}. Deprecated commands
36741 cause @value{GDBN} to issue a warning when you use them. The optional
36742 argument @var{replacement} says which newer command should be used in
36743 favor of the deprecated one; if it is given, @value{GDBN} will mention
36744 the replacement as part of the warning.
36746 @kindex maint dump-me
36747 @item maint dump-me
36748 @cindex @code{SIGQUIT} signal, dump core of @value{GDBN}
36749 Cause a fatal signal in the debugger and force it to dump its core.
36750 This is supported only on systems which support aborting a program
36751 with the @code{SIGQUIT} signal.
36753 @kindex maint internal-error
36754 @kindex maint internal-warning
36755 @kindex maint demangler-warning
36756 @cindex demangler crashes
36757 @item maint internal-error @r{[}@var{message-text}@r{]}
36758 @itemx maint internal-warning @r{[}@var{message-text}@r{]}
36759 @itemx maint demangler-warning @r{[}@var{message-text}@r{]}
36761 Cause @value{GDBN} to call the internal function @code{internal_error},
36762 @code{internal_warning} or @code{demangler_warning} and hence behave
36763 as though an internal problem has been detected. In addition to
36764 reporting the internal problem, these functions give the user the
36765 opportunity to either quit @value{GDBN} or (for @code{internal_error}
36766 and @code{internal_warning}) create a core file of the current
36767 @value{GDBN} session.
36769 These commands take an optional parameter @var{message-text} that is
36770 used as the text of the error or warning message.
36772 Here's an example of using @code{internal-error}:
36775 (@value{GDBP}) @kbd{maint internal-error testing, 1, 2}
36776 @dots{}/maint.c:121: internal-error: testing, 1, 2
36777 A problem internal to GDB has been detected. Further
36778 debugging may prove unreliable.
36779 Quit this debugging session? (y or n) @kbd{n}
36780 Create a core file? (y or n) @kbd{n}
36784 @cindex @value{GDBN} internal error
36785 @cindex internal errors, control of @value{GDBN} behavior
36786 @cindex demangler crashes
36788 @kindex maint set internal-error
36789 @kindex maint show internal-error
36790 @kindex maint set internal-warning
36791 @kindex maint show internal-warning
36792 @kindex maint set demangler-warning
36793 @kindex maint show demangler-warning
36794 @item maint set internal-error @var{action} [ask|yes|no]
36795 @itemx maint show internal-error @var{action}
36796 @itemx maint set internal-warning @var{action} [ask|yes|no]
36797 @itemx maint show internal-warning @var{action}
36798 @itemx maint set demangler-warning @var{action} [ask|yes|no]
36799 @itemx maint show demangler-warning @var{action}
36800 When @value{GDBN} reports an internal problem (error or warning) it
36801 gives the user the opportunity to both quit @value{GDBN} and create a
36802 core file of the current @value{GDBN} session. These commands let you
36803 override the default behaviour for each particular @var{action},
36804 described in the table below.
36808 You can specify that @value{GDBN} should always (yes) or never (no)
36809 quit. The default is to ask the user what to do.
36812 You can specify that @value{GDBN} should always (yes) or never (no)
36813 create a core file. The default is to ask the user what to do. Note
36814 that there is no @code{corefile} option for @code{demangler-warning}:
36815 demangler warnings always create a core file and this cannot be
36819 @kindex maint packet
36820 @item maint packet @var{text}
36821 If @value{GDBN} is talking to an inferior via the serial protocol,
36822 then this command sends the string @var{text} to the inferior, and
36823 displays the response packet. @value{GDBN} supplies the initial
36824 @samp{$} character, the terminating @samp{#} character, and the
36827 @kindex maint print architecture
36828 @item maint print architecture @r{[}@var{file}@r{]}
36829 Print the entire architecture configuration. The optional argument
36830 @var{file} names the file where the output goes.
36832 @kindex maint print c-tdesc @r{[}@var{file}@r{]}
36833 @item maint print c-tdesc
36834 Print the target description (@pxref{Target Descriptions}) as
36835 a C source file. By default, the target description is for the current
36836 target, but if the optional argument @var{file} is provided, that file
36837 is used to produce the description. The @var{file} should be an XML
36838 document, of the form described in @ref{Target Description Format}.
36839 The created source file is built into @value{GDBN} when @value{GDBN} is
36840 built again. This command is used by developers after they add or
36841 modify XML target descriptions.
36843 @kindex maint check xml-descriptions
36844 @item maint check xml-descriptions @var{dir}
36845 Check that the target descriptions dynamically created by @value{GDBN}
36846 equal the descriptions created from XML files found in @var{dir}.
36848 @anchor{maint check libthread-db}
36849 @kindex maint check libthread-db
36850 @item maint check libthread-db
36851 Run integrity checks on the current inferior's thread debugging
36852 library. This exercises all @code{libthread_db} functionality used by
36853 @value{GDBN} on GNU/Linux systems, and by extension also exercises the
36854 @code{proc_service} functions provided by @value{GDBN} that
36855 @code{libthread_db} uses. Note that parts of the test may be skipped
36856 on some platforms when debugging core files.
36858 @kindex maint print dummy-frames
36859 @item maint print dummy-frames
36860 Prints the contents of @value{GDBN}'s internal dummy-frame stack.
36863 (@value{GDBP}) @kbd{b add}
36865 (@value{GDBP}) @kbd{print add(2,3)}
36866 Breakpoint 2, add (a=2, b=3) at @dots{}
36868 The program being debugged stopped while in a function called from GDB.
36870 (@value{GDBP}) @kbd{maint print dummy-frames}
36871 0xa8206d8: id=@{stack=0xbfffe734,code=0xbfffe73f,!special@}, ptid=process 9353
36875 Takes an optional file parameter.
36877 @kindex maint print registers
36878 @kindex maint print raw-registers
36879 @kindex maint print cooked-registers
36880 @kindex maint print register-groups
36881 @kindex maint print remote-registers
36882 @item maint print registers @r{[}@var{file}@r{]}
36883 @itemx maint print raw-registers @r{[}@var{file}@r{]}
36884 @itemx maint print cooked-registers @r{[}@var{file}@r{]}
36885 @itemx maint print register-groups @r{[}@var{file}@r{]}
36886 @itemx maint print remote-registers @r{[}@var{file}@r{]}
36887 Print @value{GDBN}'s internal register data structures.
36889 The command @code{maint print raw-registers} includes the contents of
36890 the raw register cache; the command @code{maint print
36891 cooked-registers} includes the (cooked) value of all registers,
36892 including registers which aren't available on the target nor visible
36893 to user; the command @code{maint print register-groups} includes the
36894 groups that each register is a member of; and the command @code{maint
36895 print remote-registers} includes the remote target's register numbers
36896 and offsets in the `G' packets.
36898 These commands take an optional parameter, a file name to which to
36899 write the information.
36901 @kindex maint print reggroups
36902 @item maint print reggroups @r{[}@var{file}@r{]}
36903 Print @value{GDBN}'s internal register group data structures. The
36904 optional argument @var{file} tells to what file to write the
36907 The register groups info looks like this:
36910 (@value{GDBP}) @kbd{maint print reggroups}
36923 This command forces @value{GDBN} to flush its internal register cache.
36925 @kindex maint print objfiles
36926 @cindex info for known object files
36927 @item maint print objfiles @r{[}@var{regexp}@r{]}
36928 Print a dump of all known object files.
36929 If @var{regexp} is specified, only print object files whose names
36930 match @var{regexp}. For each object file, this command prints its name,
36931 address in memory, and all of its psymtabs and symtabs.
36933 @kindex maint print user-registers
36934 @cindex user registers
36935 @item maint print user-registers
36936 List all currently available @dfn{user registers}. User registers
36937 typically provide alternate names for actual hardware registers. They
36938 include the four ``standard'' registers @code{$fp}, @code{$pc},
36939 @code{$sp}, and @code{$ps}. @xref{standard registers}. User
36940 registers can be used in expressions in the same way as the canonical
36941 register names, but only the latter are listed by the @code{info
36942 registers} and @code{maint print registers} commands.
36944 @kindex maint print section-scripts
36945 @cindex info for known .debug_gdb_scripts-loaded scripts
36946 @item maint print section-scripts [@var{regexp}]
36947 Print a dump of scripts specified in the @code{.debug_gdb_section} section.
36948 If @var{regexp} is specified, only print scripts loaded by object files
36949 matching @var{regexp}.
36950 For each script, this command prints its name as specified in the objfile,
36951 and the full path if known.
36952 @xref{dotdebug_gdb_scripts section}.
36954 @kindex maint print statistics
36955 @cindex bcache statistics
36956 @item maint print statistics
36957 This command prints, for each object file in the program, various data
36958 about that object file followed by the byte cache (@dfn{bcache})
36959 statistics for the object file. The objfile data includes the number
36960 of minimal, partial, full, and stabs symbols, the number of types
36961 defined by the objfile, the number of as yet unexpanded psym tables,
36962 the number of line tables and string tables, and the amount of memory
36963 used by the various tables. The bcache statistics include the counts,
36964 sizes, and counts of duplicates of all and unique objects, max,
36965 average, and median entry size, total memory used and its overhead and
36966 savings, and various measures of the hash table size and chain
36969 @kindex maint print target-stack
36970 @cindex target stack description
36971 @item maint print target-stack
36972 A @dfn{target} is an interface between the debugger and a particular
36973 kind of file or process. Targets can be stacked in @dfn{strata},
36974 so that more than one target can potentially respond to a request.
36975 In particular, memory accesses will walk down the stack of targets
36976 until they find a target that is interested in handling that particular
36979 This command prints a short description of each layer that was pushed on
36980 the @dfn{target stack}, starting from the top layer down to the bottom one.
36982 @kindex maint print type
36983 @cindex type chain of a data type
36984 @item maint print type @var{expr}
36985 Print the type chain for a type specified by @var{expr}. The argument
36986 can be either a type name or a symbol. If it is a symbol, the type of
36987 that symbol is described. The type chain produced by this command is
36988 a recursive definition of the data type as stored in @value{GDBN}'s
36989 data structures, including its flags and contained types.
36991 @kindex maint selftest
36993 @item maint selftest @r{[}@var{filter}@r{]}
36994 Run any self tests that were compiled in to @value{GDBN}. This will
36995 print a message showing how many tests were run, and how many failed.
36996 If a @var{filter} is passed, only the tests with @var{filter} in their
36999 @kindex maint info selftests
37001 @item maint info selftests
37002 List the selftests compiled in to @value{GDBN}.
37004 @kindex maint set dwarf always-disassemble
37005 @kindex maint show dwarf always-disassemble
37006 @item maint set dwarf always-disassemble
37007 @item maint show dwarf always-disassemble
37008 Control the behavior of @code{info address} when using DWARF debugging
37011 The default is @code{off}, which means that @value{GDBN} should try to
37012 describe a variable's location in an easily readable format. When
37013 @code{on}, @value{GDBN} will instead display the DWARF location
37014 expression in an assembly-like format. Note that some locations are
37015 too complex for @value{GDBN} to describe simply; in this case you will
37016 always see the disassembly form.
37018 Here is an example of the resulting disassembly:
37021 (gdb) info addr argc
37022 Symbol "argc" is a complex DWARF expression:
37026 For more information on these expressions, see
37027 @uref{http://www.dwarfstd.org/, the DWARF standard}.
37029 @kindex maint set dwarf max-cache-age
37030 @kindex maint show dwarf max-cache-age
37031 @item maint set dwarf max-cache-age
37032 @itemx maint show dwarf max-cache-age
37033 Control the DWARF compilation unit cache.
37035 @cindex DWARF compilation units cache
37036 In object files with inter-compilation-unit references, such as those
37037 produced by the GCC option @samp{-feliminate-dwarf2-dups}, the DWARF
37038 reader needs to frequently refer to previously read compilation units.
37039 This setting controls how long a compilation unit will remain in the
37040 cache if it is not referenced. A higher limit means that cached
37041 compilation units will be stored in memory longer, and more total
37042 memory will be used. Setting it to zero disables caching, which will
37043 slow down @value{GDBN} startup, but reduce memory consumption.
37045 @kindex maint set dwarf unwinders
37046 @kindex maint show dwarf unwinders
37047 @item maint set dwarf unwinders
37048 @itemx maint show dwarf unwinders
37049 Control use of the DWARF frame unwinders.
37051 @cindex DWARF frame unwinders
37052 Many targets that support DWARF debugging use @value{GDBN}'s DWARF
37053 frame unwinders to build the backtrace. Many of these targets will
37054 also have a second mechanism for building the backtrace for use in
37055 cases where DWARF information is not available, this second mechanism
37056 is often an analysis of a function's prologue.
37058 In order to extend testing coverage of the second level stack
37059 unwinding mechanisms it is helpful to be able to disable the DWARF
37060 stack unwinders, this can be done with this switch.
37062 In normal use of @value{GDBN} disabling the DWARF unwinders is not
37063 advisable, there are cases that are better handled through DWARF than
37064 prologue analysis, and the debug experience is likely to be better
37065 with the DWARF frame unwinders enabled.
37067 If DWARF frame unwinders are not supported for a particular target
37068 architecture, then enabling this flag does not cause them to be used.
37069 @kindex maint set profile
37070 @kindex maint show profile
37071 @cindex profiling GDB
37072 @item maint set profile
37073 @itemx maint show profile
37074 Control profiling of @value{GDBN}.
37076 Profiling will be disabled until you use the @samp{maint set profile}
37077 command to enable it. When you enable profiling, the system will begin
37078 collecting timing and execution count data; when you disable profiling or
37079 exit @value{GDBN}, the results will be written to a log file. Remember that
37080 if you use profiling, @value{GDBN} will overwrite the profiling log file
37081 (often called @file{gmon.out}). If you have a record of important profiling
37082 data in a @file{gmon.out} file, be sure to move it to a safe location.
37084 Configuring with @samp{--enable-profiling} arranges for @value{GDBN} to be
37085 compiled with the @samp{-pg} compiler option.
37087 @kindex maint set show-debug-regs
37088 @kindex maint show show-debug-regs
37089 @cindex hardware debug registers
37090 @item maint set show-debug-regs
37091 @itemx maint show show-debug-regs
37092 Control whether to show variables that mirror the hardware debug
37093 registers. Use @code{on} to enable, @code{off} to disable. If
37094 enabled, the debug registers values are shown when @value{GDBN} inserts or
37095 removes a hardware breakpoint or watchpoint, and when the inferior
37096 triggers a hardware-assisted breakpoint or watchpoint.
37098 @kindex maint set show-all-tib
37099 @kindex maint show show-all-tib
37100 @item maint set show-all-tib
37101 @itemx maint show show-all-tib
37102 Control whether to show all non zero areas within a 1k block starting
37103 at thread local base, when using the @samp{info w32 thread-information-block}
37106 @kindex maint set target-async
37107 @kindex maint show target-async
37108 @item maint set target-async
37109 @itemx maint show target-async
37110 This controls whether @value{GDBN} targets operate in synchronous or
37111 asynchronous mode (@pxref{Background Execution}). Normally the
37112 default is asynchronous, if it is available; but this can be changed
37113 to more easily debug problems occurring only in synchronous mode.
37115 @kindex maint set target-non-stop @var{mode} [on|off|auto]
37116 @kindex maint show target-non-stop
37117 @item maint set target-non-stop
37118 @itemx maint show target-non-stop
37120 This controls whether @value{GDBN} targets always operate in non-stop
37121 mode even if @code{set non-stop} is @code{off} (@pxref{Non-Stop
37122 Mode}). The default is @code{auto}, meaning non-stop mode is enabled
37123 if supported by the target.
37126 @item maint set target-non-stop auto
37127 This is the default mode. @value{GDBN} controls the target in
37128 non-stop mode if the target supports it.
37130 @item maint set target-non-stop on
37131 @value{GDBN} controls the target in non-stop mode even if the target
37132 does not indicate support.
37134 @item maint set target-non-stop off
37135 @value{GDBN} does not control the target in non-stop mode even if the
37136 target supports it.
37139 @kindex maint set per-command
37140 @kindex maint show per-command
37141 @item maint set per-command
37142 @itemx maint show per-command
37143 @cindex resources used by commands
37145 @value{GDBN} can display the resources used by each command.
37146 This is useful in debugging performance problems.
37149 @item maint set per-command space [on|off]
37150 @itemx maint show per-command space
37151 Enable or disable the printing of the memory used by GDB for each command.
37152 If enabled, @value{GDBN} will display how much memory each command
37153 took, following the command's own output.
37154 This can also be requested by invoking @value{GDBN} with the
37155 @option{--statistics} command-line switch (@pxref{Mode Options}).
37157 @item maint set per-command time [on|off]
37158 @itemx maint show per-command time
37159 Enable or disable the printing of the execution time of @value{GDBN}
37161 If enabled, @value{GDBN} will display how much time it
37162 took to execute each command, following the command's own output.
37163 Both CPU time and wallclock time are printed.
37164 Printing both is useful when trying to determine whether the cost is
37165 CPU or, e.g., disk/network latency.
37166 Note that the CPU time printed is for @value{GDBN} only, it does not include
37167 the execution time of the inferior because there's no mechanism currently
37168 to compute how much time was spent by @value{GDBN} and how much time was
37169 spent by the program been debugged.
37170 This can also be requested by invoking @value{GDBN} with the
37171 @option{--statistics} command-line switch (@pxref{Mode Options}).
37173 @item maint set per-command symtab [on|off]
37174 @itemx maint show per-command symtab
37175 Enable or disable the printing of basic symbol table statistics
37177 If enabled, @value{GDBN} will display the following information:
37181 number of symbol tables
37183 number of primary symbol tables
37185 number of blocks in the blockvector
37189 @kindex maint set check-libthread-db
37190 @kindex maint show check-libthread-db
37191 @item maint set check-libthread-db [on|off]
37192 @itemx maint show check-libthread-db
37193 Control whether @value{GDBN} should run integrity checks on inferior
37194 specific thread debugging libraries as they are loaded. The default
37195 is not to perform such checks. If any check fails @value{GDBN} will
37196 unload the library and continue searching for a suitable candidate as
37197 described in @ref{set libthread-db-search-path}. For more information
37198 about the tests, see @ref{maint check libthread-db}.
37200 @kindex maint space
37201 @cindex memory used by commands
37202 @item maint space @var{value}
37203 An alias for @code{maint set per-command space}.
37204 A non-zero value enables it, zero disables it.
37207 @cindex time of command execution
37208 @item maint time @var{value}
37209 An alias for @code{maint set per-command time}.
37210 A non-zero value enables it, zero disables it.
37212 @kindex maint translate-address
37213 @item maint translate-address @r{[}@var{section}@r{]} @var{addr}
37214 Find the symbol stored at the location specified by the address
37215 @var{addr} and an optional section name @var{section}. If found,
37216 @value{GDBN} prints the name of the closest symbol and an offset from
37217 the symbol's location to the specified address. This is similar to
37218 the @code{info address} command (@pxref{Symbols}), except that this
37219 command also allows to find symbols in other sections.
37221 If section was not specified, the section in which the symbol was found
37222 is also printed. For dynamically linked executables, the name of
37223 executable or shared library containing the symbol is printed as well.
37225 @kindex maint test-settings
37226 @item maint test-settings set @var{kind}
37227 @itemx maint test-settings show @var{kind}
37228 These are representative commands for each @var{kind} of setting type
37229 @value{GDBN} supports. They are used by the testsuite for exercising
37230 the settings infrastructure.
37233 The following command is useful for non-interactive invocations of
37234 @value{GDBN}, such as in the test suite.
37237 @item set watchdog @var{nsec}
37238 @kindex set watchdog
37239 @cindex watchdog timer
37240 @cindex timeout for commands
37241 Set the maximum number of seconds @value{GDBN} will wait for the
37242 target operation to finish. If this time expires, @value{GDBN}
37243 reports and error and the command is aborted.
37245 @item show watchdog
37246 Show the current setting of the target wait timeout.
37249 @node Remote Protocol
37250 @appendix @value{GDBN} Remote Serial Protocol
37255 * Stop Reply Packets::
37256 * General Query Packets::
37257 * Architecture-Specific Protocol Details::
37258 * Tracepoint Packets::
37259 * Host I/O Packets::
37261 * Notification Packets::
37262 * Remote Non-Stop::
37263 * Packet Acknowledgment::
37265 * File-I/O Remote Protocol Extension::
37266 * Library List Format::
37267 * Library List Format for SVR4 Targets::
37268 * Memory Map Format::
37269 * Thread List Format::
37270 * Traceframe Info Format::
37271 * Branch Trace Format::
37272 * Branch Trace Configuration Format::
37278 There may be occasions when you need to know something about the
37279 protocol---for example, if there is only one serial port to your target
37280 machine, you might want your program to do something special if it
37281 recognizes a packet meant for @value{GDBN}.
37283 In the examples below, @samp{->} and @samp{<-} are used to indicate
37284 transmitted and received data, respectively.
37286 @cindex protocol, @value{GDBN} remote serial
37287 @cindex serial protocol, @value{GDBN} remote
37288 @cindex remote serial protocol
37289 All @value{GDBN} commands and responses (other than acknowledgments
37290 and notifications, see @ref{Notification Packets}) are sent as a
37291 @var{packet}. A @var{packet} is introduced with the character
37292 @samp{$}, the actual @var{packet-data}, and the terminating character
37293 @samp{#} followed by a two-digit @var{checksum}:
37296 @code{$}@var{packet-data}@code{#}@var{checksum}
37300 @cindex checksum, for @value{GDBN} remote
37302 The two-digit @var{checksum} is computed as the modulo 256 sum of all
37303 characters between the leading @samp{$} and the trailing @samp{#} (an
37304 eight bit unsigned checksum).
37306 Implementors should note that prior to @value{GDBN} 5.0 the protocol
37307 specification also included an optional two-digit @var{sequence-id}:
37310 @code{$}@var{sequence-id}@code{:}@var{packet-data}@code{#}@var{checksum}
37313 @cindex sequence-id, for @value{GDBN} remote
37315 That @var{sequence-id} was appended to the acknowledgment. @value{GDBN}
37316 has never output @var{sequence-id}s. Stubs that handle packets added
37317 since @value{GDBN} 5.0 must not accept @var{sequence-id}.
37319 When either the host or the target machine receives a packet, the first
37320 response expected is an acknowledgment: either @samp{+} (to indicate
37321 the package was received correctly) or @samp{-} (to request
37325 -> @code{$}@var{packet-data}@code{#}@var{checksum}
37330 The @samp{+}/@samp{-} acknowledgments can be disabled
37331 once a connection is established.
37332 @xref{Packet Acknowledgment}, for details.
37334 The host (@value{GDBN}) sends @var{command}s, and the target (the
37335 debugging stub incorporated in your program) sends a @var{response}. In
37336 the case of step and continue @var{command}s, the response is only sent
37337 when the operation has completed, and the target has again stopped all
37338 threads in all attached processes. This is the default all-stop mode
37339 behavior, but the remote protocol also supports @value{GDBN}'s non-stop
37340 execution mode; see @ref{Remote Non-Stop}, for details.
37342 @var{packet-data} consists of a sequence of characters with the
37343 exception of @samp{#} and @samp{$} (see @samp{X} packet for additional
37346 @cindex remote protocol, field separator
37347 Fields within the packet should be separated using @samp{,} @samp{;} or
37348 @samp{:}. Except where otherwise noted all numbers are represented in
37349 @sc{hex} with leading zeros suppressed.
37351 Implementors should note that prior to @value{GDBN} 5.0, the character
37352 @samp{:} could not appear as the third character in a packet (as it
37353 would potentially conflict with the @var{sequence-id}).
37355 @cindex remote protocol, binary data
37356 @anchor{Binary Data}
37357 Binary data in most packets is encoded either as two hexadecimal
37358 digits per byte of binary data. This allowed the traditional remote
37359 protocol to work over connections which were only seven-bit clean.
37360 Some packets designed more recently assume an eight-bit clean
37361 connection, and use a more efficient encoding to send and receive
37364 The binary data representation uses @code{7d} (@sc{ascii} @samp{@}})
37365 as an escape character. Any escaped byte is transmitted as the escape
37366 character followed by the original character XORed with @code{0x20}.
37367 For example, the byte @code{0x7d} would be transmitted as the two
37368 bytes @code{0x7d 0x5d}. The bytes @code{0x23} (@sc{ascii} @samp{#}),
37369 @code{0x24} (@sc{ascii} @samp{$}), and @code{0x7d} (@sc{ascii}
37370 @samp{@}}) must always be escaped. Responses sent by the stub
37371 must also escape @code{0x2a} (@sc{ascii} @samp{*}), so that it
37372 is not interpreted as the start of a run-length encoded sequence
37375 Response @var{data} can be run-length encoded to save space.
37376 Run-length encoding replaces runs of identical characters with one
37377 instance of the repeated character, followed by a @samp{*} and a
37378 repeat count. The repeat count is itself sent encoded, to avoid
37379 binary characters in @var{data}: a value of @var{n} is sent as
37380 @code{@var{n}+29}. For a repeat count greater or equal to 3, this
37381 produces a printable @sc{ascii} character, e.g.@: a space (@sc{ascii}
37382 code 32) for a repeat count of 3. (This is because run-length
37383 encoding starts to win for counts 3 or more.) Thus, for example,
37384 @samp{0* } is a run-length encoding of ``0000'': the space character
37385 after @samp{*} means repeat the leading @code{0} @w{@code{32 - 29 =
37388 The printable characters @samp{#} and @samp{$} or with a numeric value
37389 greater than 126 must not be used. Runs of six repeats (@samp{#}) or
37390 seven repeats (@samp{$}) can be expanded using a repeat count of only
37391 five (@samp{"}). For example, @samp{00000000} can be encoded as
37394 The error response returned for some packets includes a two character
37395 error number. That number is not well defined.
37397 @cindex empty response, for unsupported packets
37398 For any @var{command} not supported by the stub, an empty response
37399 (@samp{$#00}) should be returned. That way it is possible to extend the
37400 protocol. A newer @value{GDBN} can tell if a packet is supported based
37403 At a minimum, a stub is required to support the @samp{g} and @samp{G}
37404 commands for register access, and the @samp{m} and @samp{M} commands
37405 for memory access. Stubs that only control single-threaded targets
37406 can implement run control with the @samp{c} (continue), and @samp{s}
37407 (step) commands. Stubs that support multi-threading targets should
37408 support the @samp{vCont} command. All other commands are optional.
37413 The following table provides a complete list of all currently defined
37414 @var{command}s and their corresponding response @var{data}.
37415 @xref{File-I/O Remote Protocol Extension}, for details about the File
37416 I/O extension of the remote protocol.
37418 Each packet's description has a template showing the packet's overall
37419 syntax, followed by an explanation of the packet's meaning. We
37420 include spaces in some of the templates for clarity; these are not
37421 part of the packet's syntax. No @value{GDBN} packet uses spaces to
37422 separate its components. For example, a template like @samp{foo
37423 @var{bar} @var{baz}} describes a packet beginning with the three ASCII
37424 bytes @samp{foo}, followed by a @var{bar}, followed directly by a
37425 @var{baz}. @value{GDBN} does not transmit a space character between the
37426 @samp{foo} and the @var{bar}, or between the @var{bar} and the
37429 @cindex @var{thread-id}, in remote protocol
37430 @anchor{thread-id syntax}
37431 Several packets and replies include a @var{thread-id} field to identify
37432 a thread. Normally these are positive numbers with a target-specific
37433 interpretation, formatted as big-endian hex strings. A @var{thread-id}
37434 can also be a literal @samp{-1} to indicate all threads, or @samp{0} to
37437 In addition, the remote protocol supports a multiprocess feature in
37438 which the @var{thread-id} syntax is extended to optionally include both
37439 process and thread ID fields, as @samp{p@var{pid}.@var{tid}}.
37440 The @var{pid} (process) and @var{tid} (thread) components each have the
37441 format described above: a positive number with target-specific
37442 interpretation formatted as a big-endian hex string, literal @samp{-1}
37443 to indicate all processes or threads (respectively), or @samp{0} to
37444 indicate an arbitrary process or thread. Specifying just a process, as
37445 @samp{p@var{pid}}, is equivalent to @samp{p@var{pid}.-1}. It is an
37446 error to specify all processes but a specific thread, such as
37447 @samp{p-1.@var{tid}}. Note that the @samp{p} prefix is @emph{not} used
37448 for those packets and replies explicitly documented to include a process
37449 ID, rather than a @var{thread-id}.
37451 The multiprocess @var{thread-id} syntax extensions are only used if both
37452 @value{GDBN} and the stub report support for the @samp{multiprocess}
37453 feature using @samp{qSupported}. @xref{multiprocess extensions}, for
37456 Note that all packet forms beginning with an upper- or lower-case
37457 letter, other than those described here, are reserved for future use.
37459 Here are the packet descriptions.
37464 @cindex @samp{!} packet
37465 @anchor{extended mode}
37466 Enable extended mode. In extended mode, the remote server is made
37467 persistent. The @samp{R} packet is used to restart the program being
37473 The remote target both supports and has enabled extended mode.
37477 @cindex @samp{?} packet
37479 Indicate the reason the target halted. The reply is the same as for
37480 step and continue. This packet has a special interpretation when the
37481 target is in non-stop mode; see @ref{Remote Non-Stop}.
37484 @xref{Stop Reply Packets}, for the reply specifications.
37486 @item A @var{arglen},@var{argnum},@var{arg},@dots{}
37487 @cindex @samp{A} packet
37488 Initialized @code{argv[]} array passed into program. @var{arglen}
37489 specifies the number of bytes in the hex encoded byte stream
37490 @var{arg}. See @code{gdbserver} for more details.
37495 The arguments were set.
37501 @cindex @samp{b} packet
37502 (Don't use this packet; its behavior is not well-defined.)
37503 Change the serial line speed to @var{baud}.
37505 JTC: @emph{When does the transport layer state change? When it's
37506 received, or after the ACK is transmitted. In either case, there are
37507 problems if the command or the acknowledgment packet is dropped.}
37509 Stan: @emph{If people really wanted to add something like this, and get
37510 it working for the first time, they ought to modify ser-unix.c to send
37511 some kind of out-of-band message to a specially-setup stub and have the
37512 switch happen "in between" packets, so that from remote protocol's point
37513 of view, nothing actually happened.}
37515 @item B @var{addr},@var{mode}
37516 @cindex @samp{B} packet
37517 Set (@var{mode} is @samp{S}) or clear (@var{mode} is @samp{C}) a
37518 breakpoint at @var{addr}.
37520 Don't use this packet. Use the @samp{Z} and @samp{z} packets instead
37521 (@pxref{insert breakpoint or watchpoint packet}).
37523 @cindex @samp{bc} packet
37526 Backward continue. Execute the target system in reverse. No parameter.
37527 @xref{Reverse Execution}, for more information.
37530 @xref{Stop Reply Packets}, for the reply specifications.
37532 @cindex @samp{bs} packet
37535 Backward single step. Execute one instruction in reverse. No parameter.
37536 @xref{Reverse Execution}, for more information.
37539 @xref{Stop Reply Packets}, for the reply specifications.
37541 @item c @r{[}@var{addr}@r{]}
37542 @cindex @samp{c} packet
37543 Continue at @var{addr}, which is the address to resume. If @var{addr}
37544 is omitted, resume at current address.
37546 This packet is deprecated for multi-threading support. @xref{vCont
37550 @xref{Stop Reply Packets}, for the reply specifications.
37552 @item C @var{sig}@r{[};@var{addr}@r{]}
37553 @cindex @samp{C} packet
37554 Continue with signal @var{sig} (hex signal number). If
37555 @samp{;@var{addr}} is omitted, resume at same address.
37557 This packet is deprecated for multi-threading support. @xref{vCont
37561 @xref{Stop Reply Packets}, for the reply specifications.
37564 @cindex @samp{d} packet
37567 Don't use this packet; instead, define a general set packet
37568 (@pxref{General Query Packets}).
37572 @cindex @samp{D} packet
37573 The first form of the packet is used to detach @value{GDBN} from the
37574 remote system. It is sent to the remote target
37575 before @value{GDBN} disconnects via the @code{detach} command.
37577 The second form, including a process ID, is used when multiprocess
37578 protocol extensions are enabled (@pxref{multiprocess extensions}), to
37579 detach only a specific process. The @var{pid} is specified as a
37580 big-endian hex string.
37590 @item F @var{RC},@var{EE},@var{CF};@var{XX}
37591 @cindex @samp{F} packet
37592 A reply from @value{GDBN} to an @samp{F} packet sent by the target.
37593 This is part of the File-I/O protocol extension. @xref{File-I/O
37594 Remote Protocol Extension}, for the specification.
37597 @anchor{read registers packet}
37598 @cindex @samp{g} packet
37599 Read general registers.
37603 @item @var{XX@dots{}}
37604 Each byte of register data is described by two hex digits. The bytes
37605 with the register are transmitted in target byte order. The size of
37606 each register and their position within the @samp{g} packet are
37607 determined by the @value{GDBN} internal gdbarch functions
37608 @code{DEPRECATED_REGISTER_RAW_SIZE} and @code{gdbarch_register_name}.
37610 When reading registers from a trace frame (@pxref{Analyze Collected
37611 Data,,Using the Collected Data}), the stub may also return a string of
37612 literal @samp{x}'s in place of the register data digits, to indicate
37613 that the corresponding register has not been collected, thus its value
37614 is unavailable. For example, for an architecture with 4 registers of
37615 4 bytes each, the following reply indicates to @value{GDBN} that
37616 registers 0 and 2 have not been collected, while registers 1 and 3
37617 have been collected, and both have zero value:
37621 <- @code{xxxxxxxx00000000xxxxxxxx00000000}
37628 @item G @var{XX@dots{}}
37629 @cindex @samp{G} packet
37630 Write general registers. @xref{read registers packet}, for a
37631 description of the @var{XX@dots{}} data.
37641 @item H @var{op} @var{thread-id}
37642 @cindex @samp{H} packet
37643 Set thread for subsequent operations (@samp{m}, @samp{M}, @samp{g},
37644 @samp{G}, et.al.). Depending on the operation to be performed, @var{op}
37645 should be @samp{c} for step and continue operations (note that this
37646 is deprecated, supporting the @samp{vCont} command is a better
37647 option), and @samp{g} for other operations. The thread designator
37648 @var{thread-id} has the format and interpretation described in
37649 @ref{thread-id syntax}.
37660 @c 'H': How restrictive (or permissive) is the thread model. If a
37661 @c thread is selected and stopped, are other threads allowed
37662 @c to continue to execute? As I mentioned above, I think the
37663 @c semantics of each command when a thread is selected must be
37664 @c described. For example:
37666 @c 'g': If the stub supports threads and a specific thread is
37667 @c selected, returns the register block from that thread;
37668 @c otherwise returns current registers.
37670 @c 'G' If the stub supports threads and a specific thread is
37671 @c selected, sets the registers of the register block of
37672 @c that thread; otherwise sets current registers.
37674 @item i @r{[}@var{addr}@r{[},@var{nnn}@r{]]}
37675 @anchor{cycle step packet}
37676 @cindex @samp{i} packet
37677 Step the remote target by a single clock cycle. If @samp{,@var{nnn}} is
37678 present, cycle step @var{nnn} cycles. If @var{addr} is present, cycle
37679 step starting at that address.
37682 @cindex @samp{I} packet
37683 Signal, then cycle step. @xref{step with signal packet}. @xref{cycle
37687 @cindex @samp{k} packet
37690 The exact effect of this packet is not specified.
37692 For a bare-metal target, it may power cycle or reset the target
37693 system. For that reason, the @samp{k} packet has no reply.
37695 For a single-process target, it may kill that process if possible.
37697 A multiple-process target may choose to kill just one process, or all
37698 that are under @value{GDBN}'s control. For more precise control, use
37699 the vKill packet (@pxref{vKill packet}).
37701 If the target system immediately closes the connection in response to
37702 @samp{k}, @value{GDBN} does not consider the lack of packet
37703 acknowledgment to be an error, and assumes the kill was successful.
37705 If connected using @kbd{target extended-remote}, and the target does
37706 not close the connection in response to a kill request, @value{GDBN}
37707 probes the target state as if a new connection was opened
37708 (@pxref{? packet}).
37710 @item m @var{addr},@var{length}
37711 @cindex @samp{m} packet
37712 Read @var{length} addressable memory units starting at address @var{addr}
37713 (@pxref{addressable memory unit}). Note that @var{addr} may not be aligned to
37714 any particular boundary.
37716 The stub need not use any particular size or alignment when gathering
37717 data from memory for the response; even if @var{addr} is word-aligned
37718 and @var{length} is a multiple of the word size, the stub is free to
37719 use byte accesses, or not. For this reason, this packet may not be
37720 suitable for accessing memory-mapped I/O devices.
37721 @cindex alignment of remote memory accesses
37722 @cindex size of remote memory accesses
37723 @cindex memory, alignment and size of remote accesses
37727 @item @var{XX@dots{}}
37728 Memory contents; each byte is transmitted as a two-digit hexadecimal number.
37729 The reply may contain fewer addressable memory units than requested if the
37730 server was able to read only part of the region of memory.
37735 @item M @var{addr},@var{length}:@var{XX@dots{}}
37736 @cindex @samp{M} packet
37737 Write @var{length} addressable memory units starting at address @var{addr}
37738 (@pxref{addressable memory unit}). The data is given by @var{XX@dots{}}; each
37739 byte is transmitted as a two-digit hexadecimal number.
37746 for an error (this includes the case where only part of the data was
37751 @cindex @samp{p} packet
37752 Read the value of register @var{n}; @var{n} is in hex.
37753 @xref{read registers packet}, for a description of how the returned
37754 register value is encoded.
37758 @item @var{XX@dots{}}
37759 the register's value
37763 Indicating an unrecognized @var{query}.
37766 @item P @var{n@dots{}}=@var{r@dots{}}
37767 @anchor{write register packet}
37768 @cindex @samp{P} packet
37769 Write register @var{n@dots{}} with value @var{r@dots{}}. The register
37770 number @var{n} is in hexadecimal, and @var{r@dots{}} contains two hex
37771 digits for each byte in the register (target byte order).
37781 @item q @var{name} @var{params}@dots{}
37782 @itemx Q @var{name} @var{params}@dots{}
37783 @cindex @samp{q} packet
37784 @cindex @samp{Q} packet
37785 General query (@samp{q}) and set (@samp{Q}). These packets are
37786 described fully in @ref{General Query Packets}.
37789 @cindex @samp{r} packet
37790 Reset the entire system.
37792 Don't use this packet; use the @samp{R} packet instead.
37795 @cindex @samp{R} packet
37796 Restart the program being debugged. The @var{XX}, while needed, is ignored.
37797 This packet is only available in extended mode (@pxref{extended mode}).
37799 The @samp{R} packet has no reply.
37801 @item s @r{[}@var{addr}@r{]}
37802 @cindex @samp{s} packet
37803 Single step, resuming at @var{addr}. If
37804 @var{addr} is omitted, resume at same address.
37806 This packet is deprecated for multi-threading support. @xref{vCont
37810 @xref{Stop Reply Packets}, for the reply specifications.
37812 @item S @var{sig}@r{[};@var{addr}@r{]}
37813 @anchor{step with signal packet}
37814 @cindex @samp{S} packet
37815 Step with signal. This is analogous to the @samp{C} packet, but
37816 requests a single-step, rather than a normal resumption of execution.
37818 This packet is deprecated for multi-threading support. @xref{vCont
37822 @xref{Stop Reply Packets}, for the reply specifications.
37824 @item t @var{addr}:@var{PP},@var{MM}
37825 @cindex @samp{t} packet
37826 Search backwards starting at address @var{addr} for a match with pattern
37827 @var{PP} and mask @var{MM}, both of which are are 4 byte long.
37828 There must be at least 3 digits in @var{addr}.
37830 @item T @var{thread-id}
37831 @cindex @samp{T} packet
37832 Find out if the thread @var{thread-id} is alive. @xref{thread-id syntax}.
37837 thread is still alive
37843 Packets starting with @samp{v} are identified by a multi-letter name,
37844 up to the first @samp{;} or @samp{?} (or the end of the packet).
37846 @item vAttach;@var{pid}
37847 @cindex @samp{vAttach} packet
37848 Attach to a new process with the specified process ID @var{pid}.
37849 The process ID is a
37850 hexadecimal integer identifying the process. In all-stop mode, all
37851 threads in the attached process are stopped; in non-stop mode, it may be
37852 attached without being stopped if that is supported by the target.
37854 @c In non-stop mode, on a successful vAttach, the stub should set the
37855 @c current thread to a thread of the newly-attached process. After
37856 @c attaching, GDB queries for the attached process's thread ID with qC.
37857 @c Also note that, from a user perspective, whether or not the
37858 @c target is stopped on attach in non-stop mode depends on whether you
37859 @c use the foreground or background version of the attach command, not
37860 @c on what vAttach does; GDB does the right thing with respect to either
37861 @c stopping or restarting threads.
37863 This packet is only available in extended mode (@pxref{extended mode}).
37869 @item @r{Any stop packet}
37870 for success in all-stop mode (@pxref{Stop Reply Packets})
37872 for success in non-stop mode (@pxref{Remote Non-Stop})
37875 @item vCont@r{[};@var{action}@r{[}:@var{thread-id}@r{]]}@dots{}
37876 @cindex @samp{vCont} packet
37877 @anchor{vCont packet}
37878 Resume the inferior, specifying different actions for each thread.
37880 For each inferior thread, the leftmost action with a matching
37881 @var{thread-id} is applied. Threads that don't match any action
37882 remain in their current state. Thread IDs are specified using the
37883 syntax described in @ref{thread-id syntax}. If multiprocess
37884 extensions (@pxref{multiprocess extensions}) are supported, actions
37885 can be specified to match all threads in a process by using the
37886 @samp{p@var{pid}.-1} form of the @var{thread-id}. An action with no
37887 @var{thread-id} matches all threads. Specifying no actions is an
37890 Currently supported actions are:
37896 Continue with signal @var{sig}. The signal @var{sig} should be two hex digits.
37900 Step with signal @var{sig}. The signal @var{sig} should be two hex digits.
37903 @item r @var{start},@var{end}
37904 Step once, and then keep stepping as long as the thread stops at
37905 addresses between @var{start} (inclusive) and @var{end} (exclusive).
37906 The remote stub reports a stop reply when either the thread goes out
37907 of the range or is stopped due to an unrelated reason, such as hitting
37908 a breakpoint. @xref{range stepping}.
37910 If the range is empty (@var{start} == @var{end}), then the action
37911 becomes equivalent to the @samp{s} action. In other words,
37912 single-step once, and report the stop (even if the stepped instruction
37913 jumps to @var{start}).
37915 (A stop reply may be sent at any point even if the PC is still within
37916 the stepping range; for example, it is valid to implement this packet
37917 in a degenerate way as a single instruction step operation.)
37921 The optional argument @var{addr} normally associated with the
37922 @samp{c}, @samp{C}, @samp{s}, and @samp{S} packets is
37923 not supported in @samp{vCont}.
37925 The @samp{t} action is only relevant in non-stop mode
37926 (@pxref{Remote Non-Stop}) and may be ignored by the stub otherwise.
37927 A stop reply should be generated for any affected thread not already stopped.
37928 When a thread is stopped by means of a @samp{t} action,
37929 the corresponding stop reply should indicate that the thread has stopped with
37930 signal @samp{0}, regardless of whether the target uses some other signal
37931 as an implementation detail.
37933 The server must ignore @samp{c}, @samp{C}, @samp{s}, @samp{S}, and
37934 @samp{r} actions for threads that are already running. Conversely,
37935 the server must ignore @samp{t} actions for threads that are already
37938 @emph{Note:} In non-stop mode, a thread is considered running until
37939 @value{GDBN} acknowleges an asynchronous stop notification for it with
37940 the @samp{vStopped} packet (@pxref{Remote Non-Stop}).
37942 The stub must support @samp{vCont} if it reports support for
37943 multiprocess extensions (@pxref{multiprocess extensions}).
37946 @xref{Stop Reply Packets}, for the reply specifications.
37949 @cindex @samp{vCont?} packet
37950 Request a list of actions supported by the @samp{vCont} packet.
37954 @item vCont@r{[};@var{action}@dots{}@r{]}
37955 The @samp{vCont} packet is supported. Each @var{action} is a supported
37956 command in the @samp{vCont} packet.
37958 The @samp{vCont} packet is not supported.
37961 @anchor{vCtrlC packet}
37963 @cindex @samp{vCtrlC} packet
37964 Interrupt remote target as if a control-C was pressed on the remote
37965 terminal. This is the equivalent to reacting to the @code{^C}
37966 (@samp{\003}, the control-C character) character in all-stop mode
37967 while the target is running, except this works in non-stop mode.
37968 @xref{interrupting remote targets}, for more info on the all-stop
37979 @item vFile:@var{operation}:@var{parameter}@dots{}
37980 @cindex @samp{vFile} packet
37981 Perform a file operation on the target system. For details,
37982 see @ref{Host I/O Packets}.
37984 @item vFlashErase:@var{addr},@var{length}
37985 @cindex @samp{vFlashErase} packet
37986 Direct the stub to erase @var{length} bytes of flash starting at
37987 @var{addr}. The region may enclose any number of flash blocks, but
37988 its start and end must fall on block boundaries, as indicated by the
37989 flash block size appearing in the memory map (@pxref{Memory Map
37990 Format}). @value{GDBN} groups flash memory programming operations
37991 together, and sends a @samp{vFlashDone} request after each group; the
37992 stub is allowed to delay erase operation until the @samp{vFlashDone}
37993 packet is received.
38003 @item vFlashWrite:@var{addr}:@var{XX@dots{}}
38004 @cindex @samp{vFlashWrite} packet
38005 Direct the stub to write data to flash address @var{addr}. The data
38006 is passed in binary form using the same encoding as for the @samp{X}
38007 packet (@pxref{Binary Data}). The memory ranges specified by
38008 @samp{vFlashWrite} packets preceding a @samp{vFlashDone} packet must
38009 not overlap, and must appear in order of increasing addresses
38010 (although @samp{vFlashErase} packets for higher addresses may already
38011 have been received; the ordering is guaranteed only between
38012 @samp{vFlashWrite} packets). If a packet writes to an address that was
38013 neither erased by a preceding @samp{vFlashErase} packet nor by some other
38014 target-specific method, the results are unpredictable.
38022 for vFlashWrite addressing non-flash memory
38028 @cindex @samp{vFlashDone} packet
38029 Indicate to the stub that flash programming operation is finished.
38030 The stub is permitted to delay or batch the effects of a group of
38031 @samp{vFlashErase} and @samp{vFlashWrite} packets until a
38032 @samp{vFlashDone} packet is received. The contents of the affected
38033 regions of flash memory are unpredictable until the @samp{vFlashDone}
38034 request is completed.
38036 @item vKill;@var{pid}
38037 @cindex @samp{vKill} packet
38038 @anchor{vKill packet}
38039 Kill the process with the specified process ID @var{pid}, which is a
38040 hexadecimal integer identifying the process. This packet is used in
38041 preference to @samp{k} when multiprocess protocol extensions are
38042 supported; see @ref{multiprocess extensions}.
38052 @item vMustReplyEmpty
38053 @cindex @samp{vMustReplyEmpty} packet
38054 The correct reply to an unknown @samp{v} packet is to return the empty
38055 string, however, some older versions of @command{gdbserver} would
38056 incorrectly return @samp{OK} for unknown @samp{v} packets.
38058 The @samp{vMustReplyEmpty} is used as a feature test to check how
38059 @command{gdbserver} handles unknown packets, it is important that this
38060 packet be handled in the same way as other unknown @samp{v} packets.
38061 If this packet is handled differently to other unknown @samp{v}
38062 packets then it is possile that @value{GDBN} may run into problems in
38063 other areas, specifically around use of @samp{vFile:setfs:}.
38065 @item vRun;@var{filename}@r{[};@var{argument}@r{]}@dots{}
38066 @cindex @samp{vRun} packet
38067 Run the program @var{filename}, passing it each @var{argument} on its
38068 command line. The file and arguments are hex-encoded strings. If
38069 @var{filename} is an empty string, the stub may use a default program
38070 (e.g.@: the last program run). The program is created in the stopped
38073 @c FIXME: What about non-stop mode?
38075 This packet is only available in extended mode (@pxref{extended mode}).
38081 @item @r{Any stop packet}
38082 for success (@pxref{Stop Reply Packets})
38086 @cindex @samp{vStopped} packet
38087 @xref{Notification Packets}.
38089 @item X @var{addr},@var{length}:@var{XX@dots{}}
38091 @cindex @samp{X} packet
38092 Write data to memory, where the data is transmitted in binary.
38093 Memory is specified by its address @var{addr} and number of addressable memory
38094 units @var{length} (@pxref{addressable memory unit});
38095 @samp{@var{XX}@dots{}} is binary data (@pxref{Binary Data}).
38105 @item z @var{type},@var{addr},@var{kind}
38106 @itemx Z @var{type},@var{addr},@var{kind}
38107 @anchor{insert breakpoint or watchpoint packet}
38108 @cindex @samp{z} packet
38109 @cindex @samp{Z} packets
38110 Insert (@samp{Z}) or remove (@samp{z}) a @var{type} breakpoint or
38111 watchpoint starting at address @var{address} of kind @var{kind}.
38113 Each breakpoint and watchpoint packet @var{type} is documented
38116 @emph{Implementation notes: A remote target shall return an empty string
38117 for an unrecognized breakpoint or watchpoint packet @var{type}. A
38118 remote target shall support either both or neither of a given
38119 @samp{Z@var{type}@dots{}} and @samp{z@var{type}@dots{}} packet pair. To
38120 avoid potential problems with duplicate packets, the operations should
38121 be implemented in an idempotent way.}
38123 @item z0,@var{addr},@var{kind}
38124 @itemx Z0,@var{addr},@var{kind}@r{[};@var{cond_list}@dots{}@r{]}@r{[};cmds:@var{persist},@var{cmd_list}@dots{}@r{]}
38125 @cindex @samp{z0} packet
38126 @cindex @samp{Z0} packet
38127 Insert (@samp{Z0}) or remove (@samp{z0}) a software breakpoint at address
38128 @var{addr} of type @var{kind}.
38130 A software breakpoint is implemented by replacing the instruction at
38131 @var{addr} with a software breakpoint or trap instruction. The
38132 @var{kind} is target-specific and typically indicates the size of the
38133 breakpoint in bytes that should be inserted. E.g., the @sc{arm} and
38134 @sc{mips} can insert either a 2 or 4 byte breakpoint. Some
38135 architectures have additional meanings for @var{kind}
38136 (@pxref{Architecture-Specific Protocol Details}); if no
38137 architecture-specific value is being used, it should be @samp{0}.
38138 @var{kind} is hex-encoded. @var{cond_list} is an optional list of
38139 conditional expressions in bytecode form that should be evaluated on
38140 the target's side. These are the conditions that should be taken into
38141 consideration when deciding if the breakpoint trigger should be
38142 reported back to @value{GDBN}.
38144 See also the @samp{swbreak} stop reason (@pxref{swbreak stop reason})
38145 for how to best report a software breakpoint event to @value{GDBN}.
38147 The @var{cond_list} parameter is comprised of a series of expressions,
38148 concatenated without separators. Each expression has the following form:
38152 @item X @var{len},@var{expr}
38153 @var{len} is the length of the bytecode expression and @var{expr} is the
38154 actual conditional expression in bytecode form.
38158 The optional @var{cmd_list} parameter introduces commands that may be
38159 run on the target, rather than being reported back to @value{GDBN}.
38160 The parameter starts with a numeric flag @var{persist}; if the flag is
38161 nonzero, then the breakpoint may remain active and the commands
38162 continue to be run even when @value{GDBN} disconnects from the target.
38163 Following this flag is a series of expressions concatenated with no
38164 separators. Each expression has the following form:
38168 @item X @var{len},@var{expr}
38169 @var{len} is the length of the bytecode expression and @var{expr} is the
38170 actual commands expression in bytecode form.
38174 @emph{Implementation note: It is possible for a target to copy or move
38175 code that contains software breakpoints (e.g., when implementing
38176 overlays). The behavior of this packet, in the presence of such a
38177 target, is not defined.}
38189 @item z1,@var{addr},@var{kind}
38190 @itemx Z1,@var{addr},@var{kind}@r{[};@var{cond_list}@dots{}@r{]}@r{[};cmds:@var{persist},@var{cmd_list}@dots{}@r{]}
38191 @cindex @samp{z1} packet
38192 @cindex @samp{Z1} packet
38193 Insert (@samp{Z1}) or remove (@samp{z1}) a hardware breakpoint at
38194 address @var{addr}.
38196 A hardware breakpoint is implemented using a mechanism that is not
38197 dependent on being able to modify the target's memory. The
38198 @var{kind}, @var{cond_list}, and @var{cmd_list} arguments have the
38199 same meaning as in @samp{Z0} packets.
38201 @emph{Implementation note: A hardware breakpoint is not affected by code
38214 @item z2,@var{addr},@var{kind}
38215 @itemx Z2,@var{addr},@var{kind}
38216 @cindex @samp{z2} packet
38217 @cindex @samp{Z2} packet
38218 Insert (@samp{Z2}) or remove (@samp{z2}) a write watchpoint at @var{addr}.
38219 The number of bytes to watch is specified by @var{kind}.
38231 @item z3,@var{addr},@var{kind}
38232 @itemx Z3,@var{addr},@var{kind}
38233 @cindex @samp{z3} packet
38234 @cindex @samp{Z3} packet
38235 Insert (@samp{Z3}) or remove (@samp{z3}) a read watchpoint at @var{addr}.
38236 The number of bytes to watch is specified by @var{kind}.
38248 @item z4,@var{addr},@var{kind}
38249 @itemx Z4,@var{addr},@var{kind}
38250 @cindex @samp{z4} packet
38251 @cindex @samp{Z4} packet
38252 Insert (@samp{Z4}) or remove (@samp{z4}) an access watchpoint at @var{addr}.
38253 The number of bytes to watch is specified by @var{kind}.
38267 @node Stop Reply Packets
38268 @section Stop Reply Packets
38269 @cindex stop reply packets
38271 The @samp{C}, @samp{c}, @samp{S}, @samp{s}, @samp{vCont},
38272 @samp{vAttach}, @samp{vRun}, @samp{vStopped}, and @samp{?} packets can
38273 receive any of the below as a reply. Except for @samp{?}
38274 and @samp{vStopped}, that reply is only returned
38275 when the target halts. In the below the exact meaning of @dfn{signal
38276 number} is defined by the header @file{include/gdb/signals.h} in the
38277 @value{GDBN} source code.
38279 In non-stop mode, the server will simply reply @samp{OK} to commands
38280 such as @samp{vCont}; any stop will be the subject of a future
38281 notification. @xref{Remote Non-Stop}.
38283 As in the description of request packets, we include spaces in the
38284 reply templates for clarity; these are not part of the reply packet's
38285 syntax. No @value{GDBN} stop reply packet uses spaces to separate its
38291 The program received signal number @var{AA} (a two-digit hexadecimal
38292 number). This is equivalent to a @samp{T} response with no
38293 @var{n}:@var{r} pairs.
38295 @item T @var{AA} @var{n1}:@var{r1};@var{n2}:@var{r2};@dots{}
38296 @cindex @samp{T} packet reply
38297 The program received signal number @var{AA} (a two-digit hexadecimal
38298 number). This is equivalent to an @samp{S} response, except that the
38299 @samp{@var{n}:@var{r}} pairs can carry values of important registers
38300 and other information directly in the stop reply packet, reducing
38301 round-trip latency. Single-step and breakpoint traps are reported
38302 this way. Each @samp{@var{n}:@var{r}} pair is interpreted as follows:
38306 If @var{n} is a hexadecimal number, it is a register number, and the
38307 corresponding @var{r} gives that register's value. The data @var{r} is a
38308 series of bytes in target byte order, with each byte given by a
38309 two-digit hex number.
38312 If @var{n} is @samp{thread}, then @var{r} is the @var{thread-id} of
38313 the stopped thread, as specified in @ref{thread-id syntax}.
38316 If @var{n} is @samp{core}, then @var{r} is the hexadecimal number of
38317 the core on which the stop event was detected.
38320 If @var{n} is a recognized @dfn{stop reason}, it describes a more
38321 specific event that stopped the target. The currently defined stop
38322 reasons are listed below. The @var{aa} should be @samp{05}, the trap
38323 signal. At most one stop reason should be present.
38326 Otherwise, @value{GDBN} should ignore this @samp{@var{n}:@var{r}} pair
38327 and go on to the next; this allows us to extend the protocol in the
38331 The currently defined stop reasons are:
38337 The packet indicates a watchpoint hit, and @var{r} is the data address, in
38340 @item syscall_entry
38341 @itemx syscall_return
38342 The packet indicates a syscall entry or return, and @var{r} is the
38343 syscall number, in hex.
38345 @cindex shared library events, remote reply
38347 The packet indicates that the loaded libraries have changed.
38348 @value{GDBN} should use @samp{qXfer:libraries:read} to fetch a new
38349 list of loaded libraries. The @var{r} part is ignored.
38351 @cindex replay log events, remote reply
38353 The packet indicates that the target cannot continue replaying
38354 logged execution events, because it has reached the end (or the
38355 beginning when executing backward) of the log. The value of @var{r}
38356 will be either @samp{begin} or @samp{end}. @xref{Reverse Execution},
38357 for more information.
38360 @anchor{swbreak stop reason}
38361 The packet indicates a software breakpoint instruction was executed,
38362 irrespective of whether it was @value{GDBN} that planted the
38363 breakpoint or the breakpoint is hardcoded in the program. The @var{r}
38364 part must be left empty.
38366 On some architectures, such as x86, at the architecture level, when a
38367 breakpoint instruction executes the program counter points at the
38368 breakpoint address plus an offset. On such targets, the stub is
38369 responsible for adjusting the PC to point back at the breakpoint
38372 This packet should not be sent by default; older @value{GDBN} versions
38373 did not support it. @value{GDBN} requests it, by supplying an
38374 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
38375 remote stub must also supply the appropriate @samp{qSupported} feature
38376 indicating support.
38378 This packet is required for correct non-stop mode operation.
38381 The packet indicates the target stopped for a hardware breakpoint.
38382 The @var{r} part must be left empty.
38384 The same remarks about @samp{qSupported} and non-stop mode above
38387 @cindex fork events, remote reply
38389 The packet indicates that @code{fork} was called, and @var{r}
38390 is the thread ID of the new child process. Refer to
38391 @ref{thread-id syntax} for the format of the @var{thread-id}
38392 field. This packet is only applicable to targets that support
38395 This packet should not be sent by default; older @value{GDBN} versions
38396 did not support it. @value{GDBN} requests it, by supplying an
38397 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
38398 remote stub must also supply the appropriate @samp{qSupported} feature
38399 indicating support.
38401 @cindex vfork events, remote reply
38403 The packet indicates that @code{vfork} was called, and @var{r}
38404 is the thread ID of the new child process. Refer to
38405 @ref{thread-id syntax} for the format of the @var{thread-id}
38406 field. This packet is only applicable to targets that support
38409 This packet should not be sent by default; older @value{GDBN} versions
38410 did not support it. @value{GDBN} requests it, by supplying an
38411 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
38412 remote stub must also supply the appropriate @samp{qSupported} feature
38413 indicating support.
38415 @cindex vforkdone events, remote reply
38417 The packet indicates that a child process created by a vfork
38418 has either called @code{exec} or terminated, so that the
38419 address spaces of the parent and child process are no longer
38420 shared. The @var{r} part is ignored. This packet is only
38421 applicable to targets that support vforkdone events.
38423 This packet should not be sent by default; older @value{GDBN} versions
38424 did not support it. @value{GDBN} requests it, by supplying an
38425 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
38426 remote stub must also supply the appropriate @samp{qSupported} feature
38427 indicating support.
38429 @cindex exec events, remote reply
38431 The packet indicates that @code{execve} was called, and @var{r}
38432 is the absolute pathname of the file that was executed, in hex.
38433 This packet is only applicable to targets that support exec events.
38435 This packet should not be sent by default; older @value{GDBN} versions
38436 did not support it. @value{GDBN} requests it, by supplying an
38437 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
38438 remote stub must also supply the appropriate @samp{qSupported} feature
38439 indicating support.
38441 @cindex thread create event, remote reply
38442 @anchor{thread create event}
38444 The packet indicates that the thread was just created. The new thread
38445 is stopped until @value{GDBN} sets it running with a resumption packet
38446 (@pxref{vCont packet}). This packet should not be sent by default;
38447 @value{GDBN} requests it with the @ref{QThreadEvents} packet. See
38448 also the @samp{w} (@pxref{thread exit event}) remote reply below. The
38449 @var{r} part is ignored.
38454 @itemx W @var{AA} ; process:@var{pid}
38455 The process exited, and @var{AA} is the exit status. This is only
38456 applicable to certain targets.
38458 The second form of the response, including the process ID of the
38459 exited process, can be used only when @value{GDBN} has reported
38460 support for multiprocess protocol extensions; see @ref{multiprocess
38461 extensions}. Both @var{AA} and @var{pid} are formatted as big-endian
38465 @itemx X @var{AA} ; process:@var{pid}
38466 The process terminated with signal @var{AA}.
38468 The second form of the response, including the process ID of the
38469 terminated process, can be used only when @value{GDBN} has reported
38470 support for multiprocess protocol extensions; see @ref{multiprocess
38471 extensions}. Both @var{AA} and @var{pid} are formatted as big-endian
38474 @anchor{thread exit event}
38475 @cindex thread exit event, remote reply
38476 @item w @var{AA} ; @var{tid}
38478 The thread exited, and @var{AA} is the exit status. This response
38479 should not be sent by default; @value{GDBN} requests it with the
38480 @ref{QThreadEvents} packet. See also @ref{thread create event} above.
38481 @var{AA} is formatted as a big-endian hex string.
38484 There are no resumed threads left in the target. In other words, even
38485 though the process is alive, the last resumed thread has exited. For
38486 example, say the target process has two threads: thread 1 and thread
38487 2. The client leaves thread 1 stopped, and resumes thread 2, which
38488 subsequently exits. At this point, even though the process is still
38489 alive, and thus no @samp{W} stop reply is sent, no thread is actually
38490 executing either. The @samp{N} stop reply thus informs the client
38491 that it can stop waiting for stop replies. This packet should not be
38492 sent by default; older @value{GDBN} versions did not support it.
38493 @value{GDBN} requests it, by supplying an appropriate
38494 @samp{qSupported} feature (@pxref{qSupported}). The remote stub must
38495 also supply the appropriate @samp{qSupported} feature indicating
38498 @item O @var{XX}@dots{}
38499 @samp{@var{XX}@dots{}} is hex encoding of @sc{ascii} data, to be
38500 written as the program's console output. This can happen at any time
38501 while the program is running and the debugger should continue to wait
38502 for @samp{W}, @samp{T}, etc. This reply is not permitted in non-stop mode.
38504 @item F @var{call-id},@var{parameter}@dots{}
38505 @var{call-id} is the identifier which says which host system call should
38506 be called. This is just the name of the function. Translation into the
38507 correct system call is only applicable as it's defined in @value{GDBN}.
38508 @xref{File-I/O Remote Protocol Extension}, for a list of implemented
38511 @samp{@var{parameter}@dots{}} is a list of parameters as defined for
38512 this very system call.
38514 The target replies with this packet when it expects @value{GDBN} to
38515 call a host system call on behalf of the target. @value{GDBN} replies
38516 with an appropriate @samp{F} packet and keeps up waiting for the next
38517 reply packet from the target. The latest @samp{C}, @samp{c}, @samp{S}
38518 or @samp{s} action is expected to be continued. @xref{File-I/O Remote
38519 Protocol Extension}, for more details.
38523 @node General Query Packets
38524 @section General Query Packets
38525 @cindex remote query requests
38527 Packets starting with @samp{q} are @dfn{general query packets};
38528 packets starting with @samp{Q} are @dfn{general set packets}. General
38529 query and set packets are a semi-unified form for retrieving and
38530 sending information to and from the stub.
38532 The initial letter of a query or set packet is followed by a name
38533 indicating what sort of thing the packet applies to. For example,
38534 @value{GDBN} may use a @samp{qSymbol} packet to exchange symbol
38535 definitions with the stub. These packet names follow some
38540 The name must not contain commas, colons or semicolons.
38542 Most @value{GDBN} query and set packets have a leading upper case
38545 The names of custom vendor packets should use a company prefix, in
38546 lower case, followed by a period. For example, packets designed at
38547 the Acme Corporation might begin with @samp{qacme.foo} (for querying
38548 foos) or @samp{Qacme.bar} (for setting bars).
38551 The name of a query or set packet should be separated from any
38552 parameters by a @samp{:}; the parameters themselves should be
38553 separated by @samp{,} or @samp{;}. Stubs must be careful to match the
38554 full packet name, and check for a separator or the end of the packet,
38555 in case two packet names share a common prefix. New packets should not begin
38556 with @samp{qC}, @samp{qP}, or @samp{qL}@footnote{The @samp{qP} and @samp{qL}
38557 packets predate these conventions, and have arguments without any terminator
38558 for the packet name; we suspect they are in widespread use in places that
38559 are difficult to upgrade. The @samp{qC} packet has no arguments, but some
38560 existing stubs (e.g.@: RedBoot) are known to not check for the end of the
38563 Like the descriptions of the other packets, each description here
38564 has a template showing the packet's overall syntax, followed by an
38565 explanation of the packet's meaning. We include spaces in some of the
38566 templates for clarity; these are not part of the packet's syntax. No
38567 @value{GDBN} packet uses spaces to separate its components.
38569 Here are the currently defined query and set packets:
38575 Turn on or off the agent as a helper to perform some debugging operations
38576 delegated from @value{GDBN} (@pxref{Control Agent}).
38578 @item QAllow:@var{op}:@var{val}@dots{}
38579 @cindex @samp{QAllow} packet
38580 Specify which operations @value{GDBN} expects to request of the
38581 target, as a semicolon-separated list of operation name and value
38582 pairs. Possible values for @var{op} include @samp{WriteReg},
38583 @samp{WriteMem}, @samp{InsertBreak}, @samp{InsertTrace},
38584 @samp{InsertFastTrace}, and @samp{Stop}. @var{val} is either 0,
38585 indicating that @value{GDBN} will not request the operation, or 1,
38586 indicating that it may. (The target can then use this to set up its
38587 own internals optimally, for instance if the debugger never expects to
38588 insert breakpoints, it may not need to install its own trap handler.)
38591 @cindex current thread, remote request
38592 @cindex @samp{qC} packet
38593 Return the current thread ID.
38597 @item QC @var{thread-id}
38598 Where @var{thread-id} is a thread ID as documented in
38599 @ref{thread-id syntax}.
38600 @item @r{(anything else)}
38601 Any other reply implies the old thread ID.
38604 @item qCRC:@var{addr},@var{length}
38605 @cindex CRC of memory block, remote request
38606 @cindex @samp{qCRC} packet
38607 @anchor{qCRC packet}
38608 Compute the CRC checksum of a block of memory using CRC-32 defined in
38609 IEEE 802.3. The CRC is computed byte at a time, taking the most
38610 significant bit of each byte first. The initial pattern code
38611 @code{0xffffffff} is used to ensure leading zeros affect the CRC.
38613 @emph{Note:} This is the same CRC used in validating separate debug
38614 files (@pxref{Separate Debug Files, , Debugging Information in Separate
38615 Files}). However the algorithm is slightly different. When validating
38616 separate debug files, the CRC is computed taking the @emph{least}
38617 significant bit of each byte first, and the final result is inverted to
38618 detect trailing zeros.
38623 An error (such as memory fault)
38624 @item C @var{crc32}
38625 The specified memory region's checksum is @var{crc32}.
38628 @item QDisableRandomization:@var{value}
38629 @cindex disable address space randomization, remote request
38630 @cindex @samp{QDisableRandomization} packet
38631 Some target operating systems will randomize the virtual address space
38632 of the inferior process as a security feature, but provide a feature
38633 to disable such randomization, e.g.@: to allow for a more deterministic
38634 debugging experience. On such systems, this packet with a @var{value}
38635 of 1 directs the target to disable address space randomization for
38636 processes subsequently started via @samp{vRun} packets, while a packet
38637 with a @var{value} of 0 tells the target to enable address space
38640 This packet is only available in extended mode (@pxref{extended mode}).
38645 The request succeeded.
38648 An error occurred. The error number @var{nn} is given as hex digits.
38651 An empty reply indicates that @samp{QDisableRandomization} is not supported
38655 This packet is not probed by default; the remote stub must request it,
38656 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
38657 This should only be done on targets that actually support disabling
38658 address space randomization.
38660 @item QStartupWithShell:@var{value}
38661 @cindex startup with shell, remote request
38662 @cindex @samp{QStartupWithShell} packet
38663 On UNIX-like targets, it is possible to start the inferior using a
38664 shell program. This is the default behavior on both @value{GDBN} and
38665 @command{gdbserver} (@pxref{set startup-with-shell}). This packet is
38666 used to inform @command{gdbserver} whether it should start the
38667 inferior using a shell or not.
38669 If @var{value} is @samp{0}, @command{gdbserver} will not use a shell
38670 to start the inferior. If @var{value} is @samp{1},
38671 @command{gdbserver} will use a shell to start the inferior. All other
38672 values are considered an error.
38674 This packet is only available in extended mode (@pxref{extended
38680 The request succeeded.
38683 An error occurred. The error number @var{nn} is given as hex digits.
38686 This packet is not probed by default; the remote stub must request it,
38687 by supplying an appropriate @samp{qSupported} response
38688 (@pxref{qSupported}). This should only be done on targets that
38689 actually support starting the inferior using a shell.
38691 Use of this packet is controlled by the @code{set startup-with-shell}
38692 command; @pxref{set startup-with-shell}.
38694 @item QEnvironmentHexEncoded:@var{hex-value}
38695 @anchor{QEnvironmentHexEncoded}
38696 @cindex set environment variable, remote request
38697 @cindex @samp{QEnvironmentHexEncoded} packet
38698 On UNIX-like targets, it is possible to set environment variables that
38699 will be passed to the inferior during the startup process. This
38700 packet is used to inform @command{gdbserver} of an environment
38701 variable that has been defined by the user on @value{GDBN} (@pxref{set
38704 The packet is composed by @var{hex-value}, an hex encoded
38705 representation of the @var{name=value} format representing an
38706 environment variable. The name of the environment variable is
38707 represented by @var{name}, and the value to be assigned to the
38708 environment variable is represented by @var{value}. If the variable
38709 has no value (i.e., the value is @code{null}), then @var{value} will
38712 This packet is only available in extended mode (@pxref{extended
38718 The request succeeded.
38721 This packet is not probed by default; the remote stub must request it,
38722 by supplying an appropriate @samp{qSupported} response
38723 (@pxref{qSupported}). This should only be done on targets that
38724 actually support passing environment variables to the starting
38727 This packet is related to the @code{set environment} command;
38728 @pxref{set environment}.
38730 @item QEnvironmentUnset:@var{hex-value}
38731 @anchor{QEnvironmentUnset}
38732 @cindex unset environment variable, remote request
38733 @cindex @samp{QEnvironmentUnset} packet
38734 On UNIX-like targets, it is possible to unset environment variables
38735 before starting the inferior in the remote target. This packet is
38736 used to inform @command{gdbserver} of an environment variable that has
38737 been unset by the user on @value{GDBN} (@pxref{unset environment}).
38739 The packet is composed by @var{hex-value}, an hex encoded
38740 representation of the name of the environment variable to be unset.
38742 This packet is only available in extended mode (@pxref{extended
38748 The request succeeded.
38751 This packet is not probed by default; the remote stub must request it,
38752 by supplying an appropriate @samp{qSupported} response
38753 (@pxref{qSupported}). This should only be done on targets that
38754 actually support passing environment variables to the starting
38757 This packet is related to the @code{unset environment} command;
38758 @pxref{unset environment}.
38760 @item QEnvironmentReset
38761 @anchor{QEnvironmentReset}
38762 @cindex reset environment, remote request
38763 @cindex @samp{QEnvironmentReset} packet
38764 On UNIX-like targets, this packet is used to reset the state of
38765 environment variables in the remote target before starting the
38766 inferior. In this context, reset means unsetting all environment
38767 variables that were previously set by the user (i.e., were not
38768 initially present in the environment). It is sent to
38769 @command{gdbserver} before the @samp{QEnvironmentHexEncoded}
38770 (@pxref{QEnvironmentHexEncoded}) and the @samp{QEnvironmentUnset}
38771 (@pxref{QEnvironmentUnset}) packets.
38773 This packet is only available in extended mode (@pxref{extended
38779 The request succeeded.
38782 This packet is not probed by default; the remote stub must request it,
38783 by supplying an appropriate @samp{qSupported} response
38784 (@pxref{qSupported}). This should only be done on targets that
38785 actually support passing environment variables to the starting
38788 @item QSetWorkingDir:@r{[}@var{directory}@r{]}
38789 @anchor{QSetWorkingDir packet}
38790 @cindex set working directory, remote request
38791 @cindex @samp{QSetWorkingDir} packet
38792 This packet is used to inform the remote server of the intended
38793 current working directory for programs that are going to be executed.
38795 The packet is composed by @var{directory}, an hex encoded
38796 representation of the directory that the remote inferior will use as
38797 its current working directory. If @var{directory} is an empty string,
38798 the remote server should reset the inferior's current working
38799 directory to its original, empty value.
38801 This packet is only available in extended mode (@pxref{extended
38807 The request succeeded.
38811 @itemx qsThreadInfo
38812 @cindex list active threads, remote request
38813 @cindex @samp{qfThreadInfo} packet
38814 @cindex @samp{qsThreadInfo} packet
38815 Obtain a list of all active thread IDs from the target (OS). Since there
38816 may be too many active threads to fit into one reply packet, this query
38817 works iteratively: it may require more than one query/reply sequence to
38818 obtain the entire list of threads. The first query of the sequence will
38819 be the @samp{qfThreadInfo} query; subsequent queries in the
38820 sequence will be the @samp{qsThreadInfo} query.
38822 NOTE: This packet replaces the @samp{qL} query (see below).
38826 @item m @var{thread-id}
38828 @item m @var{thread-id},@var{thread-id}@dots{}
38829 a comma-separated list of thread IDs
38831 (lower case letter @samp{L}) denotes end of list.
38834 In response to each query, the target will reply with a list of one or
38835 more thread IDs, separated by commas.
38836 @value{GDBN} will respond to each reply with a request for more thread
38837 ids (using the @samp{qs} form of the query), until the target responds
38838 with @samp{l} (lower-case ell, for @dfn{last}).
38839 Refer to @ref{thread-id syntax}, for the format of the @var{thread-id}
38842 @emph{Note: @value{GDBN} will send the @code{qfThreadInfo} query during the
38843 initial connection with the remote target, and the very first thread ID
38844 mentioned in the reply will be stopped by @value{GDBN} in a subsequent
38845 message. Therefore, the stub should ensure that the first thread ID in
38846 the @code{qfThreadInfo} reply is suitable for being stopped by @value{GDBN}.}
38848 @item qGetTLSAddr:@var{thread-id},@var{offset},@var{lm}
38849 @cindex get thread-local storage address, remote request
38850 @cindex @samp{qGetTLSAddr} packet
38851 Fetch the address associated with thread local storage specified
38852 by @var{thread-id}, @var{offset}, and @var{lm}.
38854 @var{thread-id} is the thread ID associated with the
38855 thread for which to fetch the TLS address. @xref{thread-id syntax}.
38857 @var{offset} is the (big endian, hex encoded) offset associated with the
38858 thread local variable. (This offset is obtained from the debug
38859 information associated with the variable.)
38861 @var{lm} is the (big endian, hex encoded) OS/ABI-specific encoding of the
38862 load module associated with the thread local storage. For example,
38863 a @sc{gnu}/Linux system will pass the link map address of the shared
38864 object associated with the thread local storage under consideration.
38865 Other operating environments may choose to represent the load module
38866 differently, so the precise meaning of this parameter will vary.
38870 @item @var{XX}@dots{}
38871 Hex encoded (big endian) bytes representing the address of the thread
38872 local storage requested.
38875 An error occurred. The error number @var{nn} is given as hex digits.
38878 An empty reply indicates that @samp{qGetTLSAddr} is not supported by the stub.
38881 @item qGetTIBAddr:@var{thread-id}
38882 @cindex get thread information block address
38883 @cindex @samp{qGetTIBAddr} packet
38884 Fetch address of the Windows OS specific Thread Information Block.
38886 @var{thread-id} is the thread ID associated with the thread.
38890 @item @var{XX}@dots{}
38891 Hex encoded (big endian) bytes representing the linear address of the
38892 thread information block.
38895 An error occured. This means that either the thread was not found, or the
38896 address could not be retrieved.
38899 An empty reply indicates that @samp{qGetTIBAddr} is not supported by the stub.
38902 @item qL @var{startflag} @var{threadcount} @var{nextthread}
38903 Obtain thread information from RTOS. Where: @var{startflag} (one hex
38904 digit) is one to indicate the first query and zero to indicate a
38905 subsequent query; @var{threadcount} (two hex digits) is the maximum
38906 number of threads the response packet can contain; and @var{nextthread}
38907 (eight hex digits), for subsequent queries (@var{startflag} is zero), is
38908 returned in the response as @var{argthread}.
38910 Don't use this packet; use the @samp{qfThreadInfo} query instead (see above).
38914 @item qM @var{count} @var{done} @var{argthread} @var{thread}@dots{}
38915 Where: @var{count} (two hex digits) is the number of threads being
38916 returned; @var{done} (one hex digit) is zero to indicate more threads
38917 and one indicates no further threads; @var{argthreadid} (eight hex
38918 digits) is @var{nextthread} from the request packet; @var{thread}@dots{}
38919 is a sequence of thread IDs, @var{threadid} (eight hex
38920 digits), from the target. See @code{remote.c:parse_threadlist_response()}.
38924 @cindex section offsets, remote request
38925 @cindex @samp{qOffsets} packet
38926 Get section offsets that the target used when relocating the downloaded
38931 @item Text=@var{xxx};Data=@var{yyy}@r{[};Bss=@var{zzz}@r{]}
38932 Relocate the @code{Text} section by @var{xxx} from its original address.
38933 Relocate the @code{Data} section by @var{yyy} from its original address.
38934 If the object file format provides segment information (e.g.@: @sc{elf}
38935 @samp{PT_LOAD} program headers), @value{GDBN} will relocate entire
38936 segments by the supplied offsets.
38938 @emph{Note: while a @code{Bss} offset may be included in the response,
38939 @value{GDBN} ignores this and instead applies the @code{Data} offset
38940 to the @code{Bss} section.}
38942 @item TextSeg=@var{xxx}@r{[};DataSeg=@var{yyy}@r{]}
38943 Relocate the first segment of the object file, which conventionally
38944 contains program code, to a starting address of @var{xxx}. If
38945 @samp{DataSeg} is specified, relocate the second segment, which
38946 conventionally contains modifiable data, to a starting address of
38947 @var{yyy}. @value{GDBN} will report an error if the object file
38948 does not contain segment information, or does not contain at least
38949 as many segments as mentioned in the reply. Extra segments are
38950 kept at fixed offsets relative to the last relocated segment.
38953 @item qP @var{mode} @var{thread-id}
38954 @cindex thread information, remote request
38955 @cindex @samp{qP} packet
38956 Returns information on @var{thread-id}. Where: @var{mode} is a hex
38957 encoded 32 bit mode; @var{thread-id} is a thread ID
38958 (@pxref{thread-id syntax}).
38960 Don't use this packet; use the @samp{qThreadExtraInfo} query instead
38963 Reply: see @code{remote.c:remote_unpack_thread_info_response()}.
38967 @cindex non-stop mode, remote request
38968 @cindex @samp{QNonStop} packet
38970 Enter non-stop (@samp{QNonStop:1}) or all-stop (@samp{QNonStop:0}) mode.
38971 @xref{Remote Non-Stop}, for more information.
38976 The request succeeded.
38979 An error occurred. The error number @var{nn} is given as hex digits.
38982 An empty reply indicates that @samp{QNonStop} is not supported by
38986 This packet is not probed by default; the remote stub must request it,
38987 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
38988 Use of this packet is controlled by the @code{set non-stop} command;
38989 @pxref{Non-Stop Mode}.
38991 @item QCatchSyscalls:1 @r{[};@var{sysno}@r{]}@dots{}
38992 @itemx QCatchSyscalls:0
38993 @cindex catch syscalls from inferior, remote request
38994 @cindex @samp{QCatchSyscalls} packet
38995 @anchor{QCatchSyscalls}
38996 Enable (@samp{QCatchSyscalls:1}) or disable (@samp{QCatchSyscalls:0})
38997 catching syscalls from the inferior process.
38999 For @samp{QCatchSyscalls:1}, each listed syscall @var{sysno} (encoded
39000 in hex) should be reported to @value{GDBN}. If no syscall @var{sysno}
39001 is listed, every system call should be reported.
39003 Note that if a syscall not in the list is reported, @value{GDBN} will
39004 still filter the event according to its own list from all corresponding
39005 @code{catch syscall} commands. However, it is more efficient to only
39006 report the requested syscalls.
39008 Multiple @samp{QCatchSyscalls:1} packets do not combine; any earlier
39009 @samp{QCatchSyscalls:1} list is completely replaced by the new list.
39011 If the inferior process execs, the state of @samp{QCatchSyscalls} is
39012 kept for the new process too. On targets where exec may affect syscall
39013 numbers, for example with exec between 32 and 64-bit processes, the
39014 client should send a new packet with the new syscall list.
39019 The request succeeded.
39022 An error occurred. @var{nn} are hex digits.
39025 An empty reply indicates that @samp{QCatchSyscalls} is not supported by
39029 Use of this packet is controlled by the @code{set remote catch-syscalls}
39030 command (@pxref{Remote Configuration, set remote catch-syscalls}).
39031 This packet is not probed by default; the remote stub must request it,
39032 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
39034 @item QPassSignals: @var{signal} @r{[};@var{signal}@r{]}@dots{}
39035 @cindex pass signals to inferior, remote request
39036 @cindex @samp{QPassSignals} packet
39037 @anchor{QPassSignals}
39038 Each listed @var{signal} should be passed directly to the inferior process.
39039 Signals are numbered identically to continue packets and stop replies
39040 (@pxref{Stop Reply Packets}). Each @var{signal} list item should be
39041 strictly greater than the previous item. These signals do not need to stop
39042 the inferior, or be reported to @value{GDBN}. All other signals should be
39043 reported to @value{GDBN}. Multiple @samp{QPassSignals} packets do not
39044 combine; any earlier @samp{QPassSignals} list is completely replaced by the
39045 new list. This packet improves performance when using @samp{handle
39046 @var{signal} nostop noprint pass}.
39051 The request succeeded.
39054 An error occurred. The error number @var{nn} is given as hex digits.
39057 An empty reply indicates that @samp{QPassSignals} is not supported by
39061 Use of this packet is controlled by the @code{set remote pass-signals}
39062 command (@pxref{Remote Configuration, set remote pass-signals}).
39063 This packet is not probed by default; the remote stub must request it,
39064 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
39066 @item QProgramSignals: @var{signal} @r{[};@var{signal}@r{]}@dots{}
39067 @cindex signals the inferior may see, remote request
39068 @cindex @samp{QProgramSignals} packet
39069 @anchor{QProgramSignals}
39070 Each listed @var{signal} may be delivered to the inferior process.
39071 Others should be silently discarded.
39073 In some cases, the remote stub may need to decide whether to deliver a
39074 signal to the program or not without @value{GDBN} involvement. One
39075 example of that is while detaching --- the program's threads may have
39076 stopped for signals that haven't yet had a chance of being reported to
39077 @value{GDBN}, and so the remote stub can use the signal list specified
39078 by this packet to know whether to deliver or ignore those pending
39081 This does not influence whether to deliver a signal as requested by a
39082 resumption packet (@pxref{vCont packet}).
39084 Signals are numbered identically to continue packets and stop replies
39085 (@pxref{Stop Reply Packets}). Each @var{signal} list item should be
39086 strictly greater than the previous item. Multiple
39087 @samp{QProgramSignals} packets do not combine; any earlier
39088 @samp{QProgramSignals} list is completely replaced by the new list.
39093 The request succeeded.
39096 An error occurred. The error number @var{nn} is given as hex digits.
39099 An empty reply indicates that @samp{QProgramSignals} is not supported
39103 Use of this packet is controlled by the @code{set remote program-signals}
39104 command (@pxref{Remote Configuration, set remote program-signals}).
39105 This packet is not probed by default; the remote stub must request it,
39106 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
39108 @anchor{QThreadEvents}
39109 @item QThreadEvents:1
39110 @itemx QThreadEvents:0
39111 @cindex thread create/exit events, remote request
39112 @cindex @samp{QThreadEvents} packet
39114 Enable (@samp{QThreadEvents:1}) or disable (@samp{QThreadEvents:0})
39115 reporting of thread create and exit events. @xref{thread create
39116 event}, for the reply specifications. For example, this is used in
39117 non-stop mode when @value{GDBN} stops a set of threads and
39118 synchronously waits for the their corresponding stop replies. Without
39119 exit events, if one of the threads exits, @value{GDBN} would hang
39120 forever not knowing that it should no longer expect a stop for that
39121 same thread. @value{GDBN} does not enable this feature unless the
39122 stub reports that it supports it by including @samp{QThreadEvents+} in
39123 its @samp{qSupported} reply.
39128 The request succeeded.
39131 An error occurred. The error number @var{nn} is given as hex digits.
39134 An empty reply indicates that @samp{QThreadEvents} is not supported by
39138 Use of this packet is controlled by the @code{set remote thread-events}
39139 command (@pxref{Remote Configuration, set remote thread-events}).
39141 @item qRcmd,@var{command}
39142 @cindex execute remote command, remote request
39143 @cindex @samp{qRcmd} packet
39144 @var{command} (hex encoded) is passed to the local interpreter for
39145 execution. Invalid commands should be reported using the output
39146 string. Before the final result packet, the target may also respond
39147 with a number of intermediate @samp{O@var{output}} console output
39148 packets. @emph{Implementors should note that providing access to a
39149 stubs's interpreter may have security implications}.
39154 A command response with no output.
39156 A command response with the hex encoded output string @var{OUTPUT}.
39158 Indicate a badly formed request.
39160 An empty reply indicates that @samp{qRcmd} is not recognized.
39163 (Note that the @code{qRcmd} packet's name is separated from the
39164 command by a @samp{,}, not a @samp{:}, contrary to the naming
39165 conventions above. Please don't use this packet as a model for new
39168 @item qSearch:memory:@var{address};@var{length};@var{search-pattern}
39169 @cindex searching memory, in remote debugging
39171 @cindex @samp{qSearch:memory} packet
39173 @cindex @samp{qSearch memory} packet
39174 @anchor{qSearch memory}
39175 Search @var{length} bytes at @var{address} for @var{search-pattern}.
39176 Both @var{address} and @var{length} are encoded in hex;
39177 @var{search-pattern} is a sequence of bytes, also hex encoded.
39182 The pattern was not found.
39184 The pattern was found at @var{address}.
39186 A badly formed request or an error was encountered while searching memory.
39188 An empty reply indicates that @samp{qSearch:memory} is not recognized.
39191 @item QStartNoAckMode
39192 @cindex @samp{QStartNoAckMode} packet
39193 @anchor{QStartNoAckMode}
39194 Request that the remote stub disable the normal @samp{+}/@samp{-}
39195 protocol acknowledgments (@pxref{Packet Acknowledgment}).
39200 The stub has switched to no-acknowledgment mode.
39201 @value{GDBN} acknowledges this reponse,
39202 but neither the stub nor @value{GDBN} shall send or expect further
39203 @samp{+}/@samp{-} acknowledgments in the current connection.
39205 An empty reply indicates that the stub does not support no-acknowledgment mode.
39208 @item qSupported @r{[}:@var{gdbfeature} @r{[};@var{gdbfeature}@r{]}@dots{} @r{]}
39209 @cindex supported packets, remote query
39210 @cindex features of the remote protocol
39211 @cindex @samp{qSupported} packet
39212 @anchor{qSupported}
39213 Tell the remote stub about features supported by @value{GDBN}, and
39214 query the stub for features it supports. This packet allows
39215 @value{GDBN} and the remote stub to take advantage of each others'
39216 features. @samp{qSupported} also consolidates multiple feature probes
39217 at startup, to improve @value{GDBN} performance---a single larger
39218 packet performs better than multiple smaller probe packets on
39219 high-latency links. Some features may enable behavior which must not
39220 be on by default, e.g.@: because it would confuse older clients or
39221 stubs. Other features may describe packets which could be
39222 automatically probed for, but are not. These features must be
39223 reported before @value{GDBN} will use them. This ``default
39224 unsupported'' behavior is not appropriate for all packets, but it
39225 helps to keep the initial connection time under control with new
39226 versions of @value{GDBN} which support increasing numbers of packets.
39230 @item @var{stubfeature} @r{[};@var{stubfeature}@r{]}@dots{}
39231 The stub supports or does not support each returned @var{stubfeature},
39232 depending on the form of each @var{stubfeature} (see below for the
39235 An empty reply indicates that @samp{qSupported} is not recognized,
39236 or that no features needed to be reported to @value{GDBN}.
39239 The allowed forms for each feature (either a @var{gdbfeature} in the
39240 @samp{qSupported} packet, or a @var{stubfeature} in the response)
39244 @item @var{name}=@var{value}
39245 The remote protocol feature @var{name} is supported, and associated
39246 with the specified @var{value}. The format of @var{value} depends
39247 on the feature, but it must not include a semicolon.
39249 The remote protocol feature @var{name} is supported, and does not
39250 need an associated value.
39252 The remote protocol feature @var{name} is not supported.
39254 The remote protocol feature @var{name} may be supported, and
39255 @value{GDBN} should auto-detect support in some other way when it is
39256 needed. This form will not be used for @var{gdbfeature} notifications,
39257 but may be used for @var{stubfeature} responses.
39260 Whenever the stub receives a @samp{qSupported} request, the
39261 supplied set of @value{GDBN} features should override any previous
39262 request. This allows @value{GDBN} to put the stub in a known
39263 state, even if the stub had previously been communicating with
39264 a different version of @value{GDBN}.
39266 The following values of @var{gdbfeature} (for the packet sent by @value{GDBN})
39271 This feature indicates whether @value{GDBN} supports multiprocess
39272 extensions to the remote protocol. @value{GDBN} does not use such
39273 extensions unless the stub also reports that it supports them by
39274 including @samp{multiprocess+} in its @samp{qSupported} reply.
39275 @xref{multiprocess extensions}, for details.
39278 This feature indicates that @value{GDBN} supports the XML target
39279 description. If the stub sees @samp{xmlRegisters=} with target
39280 specific strings separated by a comma, it will report register
39284 This feature indicates whether @value{GDBN} supports the
39285 @samp{qRelocInsn} packet (@pxref{Tracepoint Packets,,Relocate
39286 instruction reply packet}).
39289 This feature indicates whether @value{GDBN} supports the swbreak stop
39290 reason in stop replies. @xref{swbreak stop reason}, for details.
39293 This feature indicates whether @value{GDBN} supports the hwbreak stop
39294 reason in stop replies. @xref{swbreak stop reason}, for details.
39297 This feature indicates whether @value{GDBN} supports fork event
39298 extensions to the remote protocol. @value{GDBN} does not use such
39299 extensions unless the stub also reports that it supports them by
39300 including @samp{fork-events+} in its @samp{qSupported} reply.
39303 This feature indicates whether @value{GDBN} supports vfork event
39304 extensions to the remote protocol. @value{GDBN} does not use such
39305 extensions unless the stub also reports that it supports them by
39306 including @samp{vfork-events+} in its @samp{qSupported} reply.
39309 This feature indicates whether @value{GDBN} supports exec event
39310 extensions to the remote protocol. @value{GDBN} does not use such
39311 extensions unless the stub also reports that it supports them by
39312 including @samp{exec-events+} in its @samp{qSupported} reply.
39314 @item vContSupported
39315 This feature indicates whether @value{GDBN} wants to know the
39316 supported actions in the reply to @samp{vCont?} packet.
39319 Stubs should ignore any unknown values for
39320 @var{gdbfeature}. Any @value{GDBN} which sends a @samp{qSupported}
39321 packet supports receiving packets of unlimited length (earlier
39322 versions of @value{GDBN} may reject overly long responses). Additional values
39323 for @var{gdbfeature} may be defined in the future to let the stub take
39324 advantage of new features in @value{GDBN}, e.g.@: incompatible
39325 improvements in the remote protocol---the @samp{multiprocess} feature is
39326 an example of such a feature. The stub's reply should be independent
39327 of the @var{gdbfeature} entries sent by @value{GDBN}; first @value{GDBN}
39328 describes all the features it supports, and then the stub replies with
39329 all the features it supports.
39331 Similarly, @value{GDBN} will silently ignore unrecognized stub feature
39332 responses, as long as each response uses one of the standard forms.
39334 Some features are flags. A stub which supports a flag feature
39335 should respond with a @samp{+} form response. Other features
39336 require values, and the stub should respond with an @samp{=}
39339 Each feature has a default value, which @value{GDBN} will use if
39340 @samp{qSupported} is not available or if the feature is not mentioned
39341 in the @samp{qSupported} response. The default values are fixed; a
39342 stub is free to omit any feature responses that match the defaults.
39344 Not all features can be probed, but for those which can, the probing
39345 mechanism is useful: in some cases, a stub's internal
39346 architecture may not allow the protocol layer to know some information
39347 about the underlying target in advance. This is especially common in
39348 stubs which may be configured for multiple targets.
39350 These are the currently defined stub features and their properties:
39352 @multitable @columnfractions 0.35 0.2 0.12 0.2
39353 @c NOTE: The first row should be @headitem, but we do not yet require
39354 @c a new enough version of Texinfo (4.7) to use @headitem.
39356 @tab Value Required
39360 @item @samp{PacketSize}
39365 @item @samp{qXfer:auxv:read}
39370 @item @samp{qXfer:btrace:read}
39375 @item @samp{qXfer:btrace-conf:read}
39380 @item @samp{qXfer:exec-file:read}
39385 @item @samp{qXfer:features:read}
39390 @item @samp{qXfer:libraries:read}
39395 @item @samp{qXfer:libraries-svr4:read}
39400 @item @samp{augmented-libraries-svr4-read}
39405 @item @samp{qXfer:memory-map:read}
39410 @item @samp{qXfer:sdata:read}
39415 @item @samp{qXfer:spu:read}
39420 @item @samp{qXfer:spu:write}
39425 @item @samp{qXfer:siginfo:read}
39430 @item @samp{qXfer:siginfo:write}
39435 @item @samp{qXfer:threads:read}
39440 @item @samp{qXfer:traceframe-info:read}
39445 @item @samp{qXfer:uib:read}
39450 @item @samp{qXfer:fdpic:read}
39455 @item @samp{Qbtrace:off}
39460 @item @samp{Qbtrace:bts}
39465 @item @samp{Qbtrace:pt}
39470 @item @samp{Qbtrace-conf:bts:size}
39475 @item @samp{Qbtrace-conf:pt:size}
39480 @item @samp{QNonStop}
39485 @item @samp{QCatchSyscalls}
39490 @item @samp{QPassSignals}
39495 @item @samp{QStartNoAckMode}
39500 @item @samp{multiprocess}
39505 @item @samp{ConditionalBreakpoints}
39510 @item @samp{ConditionalTracepoints}
39515 @item @samp{ReverseContinue}
39520 @item @samp{ReverseStep}
39525 @item @samp{TracepointSource}
39530 @item @samp{QAgent}
39535 @item @samp{QAllow}
39540 @item @samp{QDisableRandomization}
39545 @item @samp{EnableDisableTracepoints}
39550 @item @samp{QTBuffer:size}
39555 @item @samp{tracenz}
39560 @item @samp{BreakpointCommands}
39565 @item @samp{swbreak}
39570 @item @samp{hwbreak}
39575 @item @samp{fork-events}
39580 @item @samp{vfork-events}
39585 @item @samp{exec-events}
39590 @item @samp{QThreadEvents}
39595 @item @samp{no-resumed}
39602 These are the currently defined stub features, in more detail:
39605 @cindex packet size, remote protocol
39606 @item PacketSize=@var{bytes}
39607 The remote stub can accept packets up to at least @var{bytes} in
39608 length. @value{GDBN} will send packets up to this size for bulk
39609 transfers, and will never send larger packets. This is a limit on the
39610 data characters in the packet, including the frame and checksum.
39611 There is no trailing NUL byte in a remote protocol packet; if the stub
39612 stores packets in a NUL-terminated format, it should allow an extra
39613 byte in its buffer for the NUL. If this stub feature is not supported,
39614 @value{GDBN} guesses based on the size of the @samp{g} packet response.
39616 @item qXfer:auxv:read
39617 The remote stub understands the @samp{qXfer:auxv:read} packet
39618 (@pxref{qXfer auxiliary vector read}).
39620 @item qXfer:btrace:read
39621 The remote stub understands the @samp{qXfer:btrace:read}
39622 packet (@pxref{qXfer btrace read}).
39624 @item qXfer:btrace-conf:read
39625 The remote stub understands the @samp{qXfer:btrace-conf:read}
39626 packet (@pxref{qXfer btrace-conf read}).
39628 @item qXfer:exec-file:read
39629 The remote stub understands the @samp{qXfer:exec-file:read} packet
39630 (@pxref{qXfer executable filename read}).
39632 @item qXfer:features:read
39633 The remote stub understands the @samp{qXfer:features:read} packet
39634 (@pxref{qXfer target description read}).
39636 @item qXfer:libraries:read
39637 The remote stub understands the @samp{qXfer:libraries:read} packet
39638 (@pxref{qXfer library list read}).
39640 @item qXfer:libraries-svr4:read
39641 The remote stub understands the @samp{qXfer:libraries-svr4:read} packet
39642 (@pxref{qXfer svr4 library list read}).
39644 @item augmented-libraries-svr4-read
39645 The remote stub understands the augmented form of the
39646 @samp{qXfer:libraries-svr4:read} packet
39647 (@pxref{qXfer svr4 library list read}).
39649 @item qXfer:memory-map:read
39650 The remote stub understands the @samp{qXfer:memory-map:read} packet
39651 (@pxref{qXfer memory map read}).
39653 @item qXfer:sdata:read
39654 The remote stub understands the @samp{qXfer:sdata:read} packet
39655 (@pxref{qXfer sdata read}).
39657 @item qXfer:spu:read
39658 The remote stub understands the @samp{qXfer:spu:read} packet
39659 (@pxref{qXfer spu read}).
39661 @item qXfer:spu:write
39662 The remote stub understands the @samp{qXfer:spu:write} packet
39663 (@pxref{qXfer spu write}).
39665 @item qXfer:siginfo:read
39666 The remote stub understands the @samp{qXfer:siginfo:read} packet
39667 (@pxref{qXfer siginfo read}).
39669 @item qXfer:siginfo:write
39670 The remote stub understands the @samp{qXfer:siginfo:write} packet
39671 (@pxref{qXfer siginfo write}).
39673 @item qXfer:threads:read
39674 The remote stub understands the @samp{qXfer:threads:read} packet
39675 (@pxref{qXfer threads read}).
39677 @item qXfer:traceframe-info:read
39678 The remote stub understands the @samp{qXfer:traceframe-info:read}
39679 packet (@pxref{qXfer traceframe info read}).
39681 @item qXfer:uib:read
39682 The remote stub understands the @samp{qXfer:uib:read}
39683 packet (@pxref{qXfer unwind info block}).
39685 @item qXfer:fdpic:read
39686 The remote stub understands the @samp{qXfer:fdpic:read}
39687 packet (@pxref{qXfer fdpic loadmap read}).
39690 The remote stub understands the @samp{QNonStop} packet
39691 (@pxref{QNonStop}).
39693 @item QCatchSyscalls
39694 The remote stub understands the @samp{QCatchSyscalls} packet
39695 (@pxref{QCatchSyscalls}).
39698 The remote stub understands the @samp{QPassSignals} packet
39699 (@pxref{QPassSignals}).
39701 @item QStartNoAckMode
39702 The remote stub understands the @samp{QStartNoAckMode} packet and
39703 prefers to operate in no-acknowledgment mode. @xref{Packet Acknowledgment}.
39706 @anchor{multiprocess extensions}
39707 @cindex multiprocess extensions, in remote protocol
39708 The remote stub understands the multiprocess extensions to the remote
39709 protocol syntax. The multiprocess extensions affect the syntax of
39710 thread IDs in both packets and replies (@pxref{thread-id syntax}), and
39711 add process IDs to the @samp{D} packet and @samp{W} and @samp{X}
39712 replies. Note that reporting this feature indicates support for the
39713 syntactic extensions only, not that the stub necessarily supports
39714 debugging of more than one process at a time. The stub must not use
39715 multiprocess extensions in packet replies unless @value{GDBN} has also
39716 indicated it supports them in its @samp{qSupported} request.
39718 @item qXfer:osdata:read
39719 The remote stub understands the @samp{qXfer:osdata:read} packet
39720 ((@pxref{qXfer osdata read}).
39722 @item ConditionalBreakpoints
39723 The target accepts and implements evaluation of conditional expressions
39724 defined for breakpoints. The target will only report breakpoint triggers
39725 when such conditions are true (@pxref{Conditions, ,Break Conditions}).
39727 @item ConditionalTracepoints
39728 The remote stub accepts and implements conditional expressions defined
39729 for tracepoints (@pxref{Tracepoint Conditions}).
39731 @item ReverseContinue
39732 The remote stub accepts and implements the reverse continue packet
39736 The remote stub accepts and implements the reverse step packet
39739 @item TracepointSource
39740 The remote stub understands the @samp{QTDPsrc} packet that supplies
39741 the source form of tracepoint definitions.
39744 The remote stub understands the @samp{QAgent} packet.
39747 The remote stub understands the @samp{QAllow} packet.
39749 @item QDisableRandomization
39750 The remote stub understands the @samp{QDisableRandomization} packet.
39752 @item StaticTracepoint
39753 @cindex static tracepoints, in remote protocol
39754 The remote stub supports static tracepoints.
39756 @item InstallInTrace
39757 @anchor{install tracepoint in tracing}
39758 The remote stub supports installing tracepoint in tracing.
39760 @item EnableDisableTracepoints
39761 The remote stub supports the @samp{QTEnable} (@pxref{QTEnable}) and
39762 @samp{QTDisable} (@pxref{QTDisable}) packets that allow tracepoints
39763 to be enabled and disabled while a trace experiment is running.
39765 @item QTBuffer:size
39766 The remote stub supports the @samp{QTBuffer:size} (@pxref{QTBuffer-size})
39767 packet that allows to change the size of the trace buffer.
39770 @cindex string tracing, in remote protocol
39771 The remote stub supports the @samp{tracenz} bytecode for collecting strings.
39772 See @ref{Bytecode Descriptions} for details about the bytecode.
39774 @item BreakpointCommands
39775 @cindex breakpoint commands, in remote protocol
39776 The remote stub supports running a breakpoint's command list itself,
39777 rather than reporting the hit to @value{GDBN}.
39780 The remote stub understands the @samp{Qbtrace:off} packet.
39783 The remote stub understands the @samp{Qbtrace:bts} packet.
39786 The remote stub understands the @samp{Qbtrace:pt} packet.
39788 @item Qbtrace-conf:bts:size
39789 The remote stub understands the @samp{Qbtrace-conf:bts:size} packet.
39791 @item Qbtrace-conf:pt:size
39792 The remote stub understands the @samp{Qbtrace-conf:pt:size} packet.
39795 The remote stub reports the @samp{swbreak} stop reason for memory
39799 The remote stub reports the @samp{hwbreak} stop reason for hardware
39803 The remote stub reports the @samp{fork} stop reason for fork events.
39806 The remote stub reports the @samp{vfork} stop reason for vfork events
39807 and vforkdone events.
39810 The remote stub reports the @samp{exec} stop reason for exec events.
39812 @item vContSupported
39813 The remote stub reports the supported actions in the reply to
39814 @samp{vCont?} packet.
39816 @item QThreadEvents
39817 The remote stub understands the @samp{QThreadEvents} packet.
39820 The remote stub reports the @samp{N} stop reply.
39825 @cindex symbol lookup, remote request
39826 @cindex @samp{qSymbol} packet
39827 Notify the target that @value{GDBN} is prepared to serve symbol lookup
39828 requests. Accept requests from the target for the values of symbols.
39833 The target does not need to look up any (more) symbols.
39834 @item qSymbol:@var{sym_name}
39835 The target requests the value of symbol @var{sym_name} (hex encoded).
39836 @value{GDBN} may provide the value by using the
39837 @samp{qSymbol:@var{sym_value}:@var{sym_name}} message, described
39841 @item qSymbol:@var{sym_value}:@var{sym_name}
39842 Set the value of @var{sym_name} to @var{sym_value}.
39844 @var{sym_name} (hex encoded) is the name of a symbol whose value the
39845 target has previously requested.
39847 @var{sym_value} (hex) is the value for symbol @var{sym_name}. If
39848 @value{GDBN} cannot supply a value for @var{sym_name}, then this field
39854 The target does not need to look up any (more) symbols.
39855 @item qSymbol:@var{sym_name}
39856 The target requests the value of a new symbol @var{sym_name} (hex
39857 encoded). @value{GDBN} will continue to supply the values of symbols
39858 (if available), until the target ceases to request them.
39863 @itemx QTDisconnected
39870 @itemx qTMinFTPILen
39872 @xref{Tracepoint Packets}.
39874 @item qThreadExtraInfo,@var{thread-id}
39875 @cindex thread attributes info, remote request
39876 @cindex @samp{qThreadExtraInfo} packet
39877 Obtain from the target OS a printable string description of thread
39878 attributes for the thread @var{thread-id}; see @ref{thread-id syntax},
39879 for the forms of @var{thread-id}. This
39880 string may contain anything that the target OS thinks is interesting
39881 for @value{GDBN} to tell the user about the thread. The string is
39882 displayed in @value{GDBN}'s @code{info threads} display. Some
39883 examples of possible thread extra info strings are @samp{Runnable}, or
39884 @samp{Blocked on Mutex}.
39888 @item @var{XX}@dots{}
39889 Where @samp{@var{XX}@dots{}} is a hex encoding of @sc{ascii} data,
39890 comprising the printable string containing the extra information about
39891 the thread's attributes.
39894 (Note that the @code{qThreadExtraInfo} packet's name is separated from
39895 the command by a @samp{,}, not a @samp{:}, contrary to the naming
39896 conventions above. Please don't use this packet as a model for new
39915 @xref{Tracepoint Packets}.
39917 @item qXfer:@var{object}:read:@var{annex}:@var{offset},@var{length}
39918 @cindex read special object, remote request
39919 @cindex @samp{qXfer} packet
39920 @anchor{qXfer read}
39921 Read uninterpreted bytes from the target's special data area
39922 identified by the keyword @var{object}. Request @var{length} bytes
39923 starting at @var{offset} bytes into the data. The content and
39924 encoding of @var{annex} is specific to @var{object}; it can supply
39925 additional details about what data to access.
39930 Data @var{data} (@pxref{Binary Data}) has been read from the
39931 target. There may be more data at a higher address (although
39932 it is permitted to return @samp{m} even for the last valid
39933 block of data, as long as at least one byte of data was read).
39934 It is possible for @var{data} to have fewer bytes than the @var{length} in the
39938 Data @var{data} (@pxref{Binary Data}) has been read from the target.
39939 There is no more data to be read. It is possible for @var{data} to
39940 have fewer bytes than the @var{length} in the request.
39943 The @var{offset} in the request is at the end of the data.
39944 There is no more data to be read.
39947 The request was malformed, or @var{annex} was invalid.
39950 The offset was invalid, or there was an error encountered reading the data.
39951 The @var{nn} part is a hex-encoded @code{errno} value.
39954 An empty reply indicates the @var{object} string was not recognized by
39955 the stub, or that the object does not support reading.
39958 Here are the specific requests of this form defined so far. All the
39959 @samp{qXfer:@var{object}:read:@dots{}} requests use the same reply
39960 formats, listed above.
39963 @item qXfer:auxv:read::@var{offset},@var{length}
39964 @anchor{qXfer auxiliary vector read}
39965 Access the target's @dfn{auxiliary vector}. @xref{OS Information,
39966 auxiliary vector}. Note @var{annex} must be empty.
39968 This packet is not probed by default; the remote stub must request it,
39969 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
39971 @item qXfer:btrace:read:@var{annex}:@var{offset},@var{length}
39972 @anchor{qXfer btrace read}
39974 Return a description of the current branch trace.
39975 @xref{Branch Trace Format}. The annex part of the generic @samp{qXfer}
39976 packet may have one of the following values:
39980 Returns all available branch trace.
39983 Returns all available branch trace if the branch trace changed since
39984 the last read request.
39987 Returns the new branch trace since the last read request. Adds a new
39988 block to the end of the trace that begins at zero and ends at the source
39989 location of the first branch in the trace buffer. This extra block is
39990 used to stitch traces together.
39992 If the trace buffer overflowed, returns an error indicating the overflow.
39995 This packet is not probed by default; the remote stub must request it
39996 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
39998 @item qXfer:btrace-conf:read::@var{offset},@var{length}
39999 @anchor{qXfer btrace-conf read}
40001 Return a description of the current branch trace configuration.
40002 @xref{Branch Trace Configuration Format}.
40004 This packet is not probed by default; the remote stub must request it
40005 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40007 @item qXfer:exec-file:read:@var{annex}:@var{offset},@var{length}
40008 @anchor{qXfer executable filename read}
40009 Return the full absolute name of the file that was executed to create
40010 a process running on the remote system. The annex specifies the
40011 numeric process ID of the process to query, encoded as a hexadecimal
40012 number. If the annex part is empty the remote stub should return the
40013 filename corresponding to the currently executing process.
40015 This packet is not probed by default; the remote stub must request it,
40016 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40018 @item qXfer:features:read:@var{annex}:@var{offset},@var{length}
40019 @anchor{qXfer target description read}
40020 Access the @dfn{target description}. @xref{Target Descriptions}. The
40021 annex specifies which XML document to access. The main description is
40022 always loaded from the @samp{target.xml} annex.
40024 This packet is not probed by default; the remote stub must request it,
40025 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40027 @item qXfer:libraries:read:@var{annex}:@var{offset},@var{length}
40028 @anchor{qXfer library list read}
40029 Access the target's list of loaded libraries. @xref{Library List Format}.
40030 The annex part of the generic @samp{qXfer} packet must be empty
40031 (@pxref{qXfer read}).
40033 Targets which maintain a list of libraries in the program's memory do
40034 not need to implement this packet; it is designed for platforms where
40035 the operating system manages the list of loaded libraries.
40037 This packet is not probed by default; the remote stub must request it,
40038 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40040 @item qXfer:libraries-svr4:read:@var{annex}:@var{offset},@var{length}
40041 @anchor{qXfer svr4 library list read}
40042 Access the target's list of loaded libraries when the target is an SVR4
40043 platform. @xref{Library List Format for SVR4 Targets}. The annex part
40044 of the generic @samp{qXfer} packet must be empty unless the remote
40045 stub indicated it supports the augmented form of this packet
40046 by supplying an appropriate @samp{qSupported} response
40047 (@pxref{qXfer read}, @ref{qSupported}).
40049 This packet is optional for better performance on SVR4 targets.
40050 @value{GDBN} uses memory read packets to read the SVR4 library list otherwise.
40052 This packet is not probed by default; the remote stub must request it,
40053 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40055 If the remote stub indicates it supports the augmented form of this
40056 packet then the annex part of the generic @samp{qXfer} packet may
40057 contain a semicolon-separated list of @samp{@var{name}=@var{value}}
40058 arguments. The currently supported arguments are:
40061 @item start=@var{address}
40062 A hexadecimal number specifying the address of the @samp{struct
40063 link_map} to start reading the library list from. If unset or zero
40064 then the first @samp{struct link_map} in the library list will be
40065 chosen as the starting point.
40067 @item prev=@var{address}
40068 A hexadecimal number specifying the address of the @samp{struct
40069 link_map} immediately preceding the @samp{struct link_map}
40070 specified by the @samp{start} argument. If unset or zero then
40071 the remote stub will expect that no @samp{struct link_map}
40072 exists prior to the starting point.
40076 Arguments that are not understood by the remote stub will be silently
40079 @item qXfer:memory-map:read::@var{offset},@var{length}
40080 @anchor{qXfer memory map read}
40081 Access the target's @dfn{memory-map}. @xref{Memory Map Format}. The
40082 annex part of the generic @samp{qXfer} packet must be empty
40083 (@pxref{qXfer read}).
40085 This packet is not probed by default; the remote stub must request it,
40086 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40088 @item qXfer:sdata:read::@var{offset},@var{length}
40089 @anchor{qXfer sdata read}
40091 Read contents of the extra collected static tracepoint marker
40092 information. The annex part of the generic @samp{qXfer} packet must
40093 be empty (@pxref{qXfer read}). @xref{Tracepoint Actions,,Tracepoint
40096 This packet is not probed by default; the remote stub must request it,
40097 by supplying an appropriate @samp{qSupported} response
40098 (@pxref{qSupported}).
40100 @item qXfer:siginfo:read::@var{offset},@var{length}
40101 @anchor{qXfer siginfo read}
40102 Read contents of the extra signal information on the target
40103 system. The annex part of the generic @samp{qXfer} packet must be
40104 empty (@pxref{qXfer read}).
40106 This packet is not probed by default; the remote stub must request it,
40107 by supplying an appropriate @samp{qSupported} response
40108 (@pxref{qSupported}).
40110 @item qXfer:spu:read:@var{annex}:@var{offset},@var{length}
40111 @anchor{qXfer spu read}
40112 Read contents of an @code{spufs} file on the target system. The
40113 annex specifies which file to read; it must be of the form
40114 @file{@var{id}/@var{name}}, where @var{id} specifies an SPU context ID
40115 in the target process, and @var{name} identifes the @code{spufs} file
40116 in that context to be accessed.
40118 This packet is not probed by default; the remote stub must request it,
40119 by supplying an appropriate @samp{qSupported} response
40120 (@pxref{qSupported}).
40122 @item qXfer:threads:read::@var{offset},@var{length}
40123 @anchor{qXfer threads read}
40124 Access the list of threads on target. @xref{Thread List Format}. The
40125 annex part of the generic @samp{qXfer} packet must be empty
40126 (@pxref{qXfer read}).
40128 This packet is not probed by default; the remote stub must request it,
40129 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40131 @item qXfer:traceframe-info:read::@var{offset},@var{length}
40132 @anchor{qXfer traceframe info read}
40134 Return a description of the current traceframe's contents.
40135 @xref{Traceframe Info Format}. The annex part of the generic
40136 @samp{qXfer} packet must be empty (@pxref{qXfer read}).
40138 This packet is not probed by default; the remote stub must request it,
40139 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40141 @item qXfer:uib:read:@var{pc}:@var{offset},@var{length}
40142 @anchor{qXfer unwind info block}
40144 Return the unwind information block for @var{pc}. This packet is used
40145 on OpenVMS/ia64 to ask the kernel unwind information.
40147 This packet is not probed by default.
40149 @item qXfer:fdpic:read:@var{annex}:@var{offset},@var{length}
40150 @anchor{qXfer fdpic loadmap read}
40151 Read contents of @code{loadmap}s on the target system. The
40152 annex, either @samp{exec} or @samp{interp}, specifies which @code{loadmap},
40153 executable @code{loadmap} or interpreter @code{loadmap} to read.
40155 This packet is not probed by default; the remote stub must request it,
40156 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40158 @item qXfer:osdata:read::@var{offset},@var{length}
40159 @anchor{qXfer osdata read}
40160 Access the target's @dfn{operating system information}.
40161 @xref{Operating System Information}.
40165 @item qXfer:@var{object}:write:@var{annex}:@var{offset}:@var{data}@dots{}
40166 @cindex write data into object, remote request
40167 @anchor{qXfer write}
40168 Write uninterpreted bytes into the target's special data area
40169 identified by the keyword @var{object}, starting at @var{offset} bytes
40170 into the data. The binary-encoded data (@pxref{Binary Data}) to be
40171 written is given by @var{data}@dots{}. The content and encoding of @var{annex}
40172 is specific to @var{object}; it can supply additional details about what data
40178 @var{nn} (hex encoded) is the number of bytes written.
40179 This may be fewer bytes than supplied in the request.
40182 The request was malformed, or @var{annex} was invalid.
40185 The offset was invalid, or there was an error encountered writing the data.
40186 The @var{nn} part is a hex-encoded @code{errno} value.
40189 An empty reply indicates the @var{object} string was not
40190 recognized by the stub, or that the object does not support writing.
40193 Here are the specific requests of this form defined so far. All the
40194 @samp{qXfer:@var{object}:write:@dots{}} requests use the same reply
40195 formats, listed above.
40198 @item qXfer:siginfo:write::@var{offset}:@var{data}@dots{}
40199 @anchor{qXfer siginfo write}
40200 Write @var{data} to the extra signal information on the target system.
40201 The annex part of the generic @samp{qXfer} packet must be
40202 empty (@pxref{qXfer write}).
40204 This packet is not probed by default; the remote stub must request it,
40205 by supplying an appropriate @samp{qSupported} response
40206 (@pxref{qSupported}).
40208 @item qXfer:spu:write:@var{annex}:@var{offset}:@var{data}@dots{}
40209 @anchor{qXfer spu write}
40210 Write @var{data} to an @code{spufs} file on the target system. The
40211 annex specifies which file to write; it must be of the form
40212 @file{@var{id}/@var{name}}, where @var{id} specifies an SPU context ID
40213 in the target process, and @var{name} identifes the @code{spufs} file
40214 in that context to be accessed.
40216 This packet is not probed by default; the remote stub must request it,
40217 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40220 @item qXfer:@var{object}:@var{operation}:@dots{}
40221 Requests of this form may be added in the future. When a stub does
40222 not recognize the @var{object} keyword, or its support for
40223 @var{object} does not recognize the @var{operation} keyword, the stub
40224 must respond with an empty packet.
40226 @item qAttached:@var{pid}
40227 @cindex query attached, remote request
40228 @cindex @samp{qAttached} packet
40229 Return an indication of whether the remote server attached to an
40230 existing process or created a new process. When the multiprocess
40231 protocol extensions are supported (@pxref{multiprocess extensions}),
40232 @var{pid} is an integer in hexadecimal format identifying the target
40233 process. Otherwise, @value{GDBN} will omit the @var{pid} field and
40234 the query packet will be simplified as @samp{qAttached}.
40236 This query is used, for example, to know whether the remote process
40237 should be detached or killed when a @value{GDBN} session is ended with
40238 the @code{quit} command.
40243 The remote server attached to an existing process.
40245 The remote server created a new process.
40247 A badly formed request or an error was encountered.
40251 Enable branch tracing for the current thread using Branch Trace Store.
40256 Branch tracing has been enabled.
40258 A badly formed request or an error was encountered.
40262 Enable branch tracing for the current thread using Intel Processor Trace.
40267 Branch tracing has been enabled.
40269 A badly formed request or an error was encountered.
40273 Disable branch tracing for the current thread.
40278 Branch tracing has been disabled.
40280 A badly formed request or an error was encountered.
40283 @item Qbtrace-conf:bts:size=@var{value}
40284 Set the requested ring buffer size for new threads that use the
40285 btrace recording method in bts format.
40290 The ring buffer size has been set.
40292 A badly formed request or an error was encountered.
40295 @item Qbtrace-conf:pt:size=@var{value}
40296 Set the requested ring buffer size for new threads that use the
40297 btrace recording method in pt format.
40302 The ring buffer size has been set.
40304 A badly formed request or an error was encountered.
40309 @node Architecture-Specific Protocol Details
40310 @section Architecture-Specific Protocol Details
40312 This section describes how the remote protocol is applied to specific
40313 target architectures. Also see @ref{Standard Target Features}, for
40314 details of XML target descriptions for each architecture.
40317 * ARM-Specific Protocol Details::
40318 * MIPS-Specific Protocol Details::
40321 @node ARM-Specific Protocol Details
40322 @subsection @acronym{ARM}-specific Protocol Details
40325 * ARM Breakpoint Kinds::
40328 @node ARM Breakpoint Kinds
40329 @subsubsection @acronym{ARM} Breakpoint Kinds
40330 @cindex breakpoint kinds, @acronym{ARM}
40332 These breakpoint kinds are defined for the @samp{Z0} and @samp{Z1} packets.
40337 16-bit Thumb mode breakpoint.
40340 32-bit Thumb mode (Thumb-2) breakpoint.
40343 32-bit @acronym{ARM} mode breakpoint.
40347 @node MIPS-Specific Protocol Details
40348 @subsection @acronym{MIPS}-specific Protocol Details
40351 * MIPS Register packet Format::
40352 * MIPS Breakpoint Kinds::
40355 @node MIPS Register packet Format
40356 @subsubsection @acronym{MIPS} Register Packet Format
40357 @cindex register packet format, @acronym{MIPS}
40359 The following @code{g}/@code{G} packets have previously been defined.
40360 In the below, some thirty-two bit registers are transferred as
40361 sixty-four bits. Those registers should be zero/sign extended (which?)
40362 to fill the space allocated. Register bytes are transferred in target
40363 byte order. The two nibbles within a register byte are transferred
40364 most-significant -- least-significant.
40369 All registers are transferred as thirty-two bit quantities in the order:
40370 32 general-purpose; sr; lo; hi; bad; cause; pc; 32 floating-point
40371 registers; fsr; fir; fp.
40374 All registers are transferred as sixty-four bit quantities (including
40375 thirty-two bit registers such as @code{sr}). The ordering is the same
40380 @node MIPS Breakpoint Kinds
40381 @subsubsection @acronym{MIPS} Breakpoint Kinds
40382 @cindex breakpoint kinds, @acronym{MIPS}
40384 These breakpoint kinds are defined for the @samp{Z0} and @samp{Z1} packets.
40389 16-bit @acronym{MIPS16} mode breakpoint.
40392 16-bit @acronym{microMIPS} mode breakpoint.
40395 32-bit standard @acronym{MIPS} mode breakpoint.
40398 32-bit @acronym{microMIPS} mode breakpoint.
40402 @node Tracepoint Packets
40403 @section Tracepoint Packets
40404 @cindex tracepoint packets
40405 @cindex packets, tracepoint
40407 Here we describe the packets @value{GDBN} uses to implement
40408 tracepoints (@pxref{Tracepoints}).
40412 @item QTDP:@var{n}:@var{addr}:@var{ena}:@var{step}:@var{pass}[:F@var{flen}][:X@var{len},@var{bytes}]@r{[}-@r{]}
40413 @cindex @samp{QTDP} packet
40414 Create a new tracepoint, number @var{n}, at @var{addr}. If @var{ena}
40415 is @samp{E}, then the tracepoint is enabled; if it is @samp{D}, then
40416 the tracepoint is disabled. The @var{step} gives the tracepoint's step
40417 count, and @var{pass} gives its pass count. If an @samp{F} is present,
40418 then the tracepoint is to be a fast tracepoint, and the @var{flen} is
40419 the number of bytes that the target should copy elsewhere to make room
40420 for the tracepoint. If an @samp{X} is present, it introduces a
40421 tracepoint condition, which consists of a hexadecimal length, followed
40422 by a comma and hex-encoded bytes, in a manner similar to action
40423 encodings as described below. If the trailing @samp{-} is present,
40424 further @samp{QTDP} packets will follow to specify this tracepoint's
40430 The packet was understood and carried out.
40432 @xref{Tracepoint Packets,,Relocate instruction reply packet}.
40434 The packet was not recognized.
40437 @item QTDP:-@var{n}:@var{addr}:@r{[}S@r{]}@var{action}@dots{}@r{[}-@r{]}
40438 Define actions to be taken when a tracepoint is hit. The @var{n} and
40439 @var{addr} must be the same as in the initial @samp{QTDP} packet for
40440 this tracepoint. This packet may only be sent immediately after
40441 another @samp{QTDP} packet that ended with a @samp{-}. If the
40442 trailing @samp{-} is present, further @samp{QTDP} packets will follow,
40443 specifying more actions for this tracepoint.
40445 In the series of action packets for a given tracepoint, at most one
40446 can have an @samp{S} before its first @var{action}. If such a packet
40447 is sent, it and the following packets define ``while-stepping''
40448 actions. Any prior packets define ordinary actions --- that is, those
40449 taken when the tracepoint is first hit. If no action packet has an
40450 @samp{S}, then all the packets in the series specify ordinary
40451 tracepoint actions.
40453 The @samp{@var{action}@dots{}} portion of the packet is a series of
40454 actions, concatenated without separators. Each action has one of the
40460 Collect the registers whose bits are set in @var{mask},
40461 a hexadecimal number whose @var{i}'th bit is set if register number
40462 @var{i} should be collected. (The least significant bit is numbered
40463 zero.) Note that @var{mask} may be any number of digits long; it may
40464 not fit in a 32-bit word.
40466 @item M @var{basereg},@var{offset},@var{len}
40467 Collect @var{len} bytes of memory starting at the address in register
40468 number @var{basereg}, plus @var{offset}. If @var{basereg} is
40469 @samp{-1}, then the range has a fixed address: @var{offset} is the
40470 address of the lowest byte to collect. The @var{basereg},
40471 @var{offset}, and @var{len} parameters are all unsigned hexadecimal
40472 values (the @samp{-1} value for @var{basereg} is a special case).
40474 @item X @var{len},@var{expr}
40475 Evaluate @var{expr}, whose length is @var{len}, and collect memory as
40476 it directs. The agent expression @var{expr} is as described in
40477 @ref{Agent Expressions}. Each byte of the expression is encoded as a
40478 two-digit hex number in the packet; @var{len} is the number of bytes
40479 in the expression (and thus one-half the number of hex digits in the
40484 Any number of actions may be packed together in a single @samp{QTDP}
40485 packet, as long as the packet does not exceed the maximum packet
40486 length (400 bytes, for many stubs). There may be only one @samp{R}
40487 action per tracepoint, and it must precede any @samp{M} or @samp{X}
40488 actions. Any registers referred to by @samp{M} and @samp{X} actions
40489 must be collected by a preceding @samp{R} action. (The
40490 ``while-stepping'' actions are treated as if they were attached to a
40491 separate tracepoint, as far as these restrictions are concerned.)
40496 The packet was understood and carried out.
40498 @xref{Tracepoint Packets,,Relocate instruction reply packet}.
40500 The packet was not recognized.
40503 @item QTDPsrc:@var{n}:@var{addr}:@var{type}:@var{start}:@var{slen}:@var{bytes}
40504 @cindex @samp{QTDPsrc} packet
40505 Specify a source string of tracepoint @var{n} at address @var{addr}.
40506 This is useful to get accurate reproduction of the tracepoints
40507 originally downloaded at the beginning of the trace run. The @var{type}
40508 is the name of the tracepoint part, such as @samp{cond} for the
40509 tracepoint's conditional expression (see below for a list of types), while
40510 @var{bytes} is the string, encoded in hexadecimal.
40512 @var{start} is the offset of the @var{bytes} within the overall source
40513 string, while @var{slen} is the total length of the source string.
40514 This is intended for handling source strings that are longer than will
40515 fit in a single packet.
40516 @c Add detailed example when this info is moved into a dedicated
40517 @c tracepoint descriptions section.
40519 The available string types are @samp{at} for the location,
40520 @samp{cond} for the conditional, and @samp{cmd} for an action command.
40521 @value{GDBN} sends a separate packet for each command in the action
40522 list, in the same order in which the commands are stored in the list.
40524 The target does not need to do anything with source strings except
40525 report them back as part of the replies to the @samp{qTfP}/@samp{qTsP}
40528 Although this packet is optional, and @value{GDBN} will only send it
40529 if the target replies with @samp{TracepointSource} @xref{General
40530 Query Packets}, it makes both disconnected tracing and trace files
40531 much easier to use. Otherwise the user must be careful that the
40532 tracepoints in effect while looking at trace frames are identical to
40533 the ones in effect during the trace run; even a small discrepancy
40534 could cause @samp{tdump} not to work, or a particular trace frame not
40537 @item QTDV:@var{n}:@var{value}:@var{builtin}:@var{name}
40538 @cindex define trace state variable, remote request
40539 @cindex @samp{QTDV} packet
40540 Create a new trace state variable, number @var{n}, with an initial
40541 value of @var{value}, which is a 64-bit signed integer. Both @var{n}
40542 and @var{value} are encoded as hexadecimal values. @value{GDBN} has
40543 the option of not using this packet for initial values of zero; the
40544 target should simply create the trace state variables as they are
40545 mentioned in expressions. The value @var{builtin} should be 1 (one)
40546 if the trace state variable is builtin and 0 (zero) if it is not builtin.
40547 @value{GDBN} only sets @var{builtin} to 1 if a previous @samp{qTfV} or
40548 @samp{qTsV} packet had it set. The contents of @var{name} is the
40549 hex-encoded name (without the leading @samp{$}) of the trace state
40552 @item QTFrame:@var{n}
40553 @cindex @samp{QTFrame} packet
40554 Select the @var{n}'th tracepoint frame from the buffer, and use the
40555 register and memory contents recorded there to answer subsequent
40556 request packets from @value{GDBN}.
40558 A successful reply from the stub indicates that the stub has found the
40559 requested frame. The response is a series of parts, concatenated
40560 without separators, describing the frame we selected. Each part has
40561 one of the following forms:
40565 The selected frame is number @var{n} in the trace frame buffer;
40566 @var{f} is a hexadecimal number. If @var{f} is @samp{-1}, then there
40567 was no frame matching the criteria in the request packet.
40570 The selected trace frame records a hit of tracepoint number @var{t};
40571 @var{t} is a hexadecimal number.
40575 @item QTFrame:pc:@var{addr}
40576 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
40577 currently selected frame whose PC is @var{addr};
40578 @var{addr} is a hexadecimal number.
40580 @item QTFrame:tdp:@var{t}
40581 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
40582 currently selected frame that is a hit of tracepoint @var{t}; @var{t}
40583 is a hexadecimal number.
40585 @item QTFrame:range:@var{start}:@var{end}
40586 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
40587 currently selected frame whose PC is between @var{start} (inclusive)
40588 and @var{end} (inclusive); @var{start} and @var{end} are hexadecimal
40591 @item QTFrame:outside:@var{start}:@var{end}
40592 Like @samp{QTFrame:range:@var{start}:@var{end}}, but select the first
40593 frame @emph{outside} the given range of addresses (exclusive).
40596 @cindex @samp{qTMinFTPILen} packet
40597 This packet requests the minimum length of instruction at which a fast
40598 tracepoint (@pxref{Set Tracepoints}) may be placed. For instance, on
40599 the 32-bit x86 architecture, it is possible to use a 4-byte jump, but
40600 it depends on the target system being able to create trampolines in
40601 the first 64K of memory, which might or might not be possible for that
40602 system. So the reply to this packet will be 4 if it is able to
40609 The minimum instruction length is currently unknown.
40611 The minimum instruction length is @var{length}, where @var{length}
40612 is a hexadecimal number greater or equal to 1. A reply
40613 of 1 means that a fast tracepoint may be placed on any instruction
40614 regardless of size.
40616 An error has occurred.
40618 An empty reply indicates that the request is not supported by the stub.
40622 @cindex @samp{QTStart} packet
40623 Begin the tracepoint experiment. Begin collecting data from
40624 tracepoint hits in the trace frame buffer. This packet supports the
40625 @samp{qRelocInsn} reply (@pxref{Tracepoint Packets,,Relocate
40626 instruction reply packet}).
40629 @cindex @samp{QTStop} packet
40630 End the tracepoint experiment. Stop collecting trace frames.
40632 @item QTEnable:@var{n}:@var{addr}
40634 @cindex @samp{QTEnable} packet
40635 Enable tracepoint @var{n} at address @var{addr} in a started tracepoint
40636 experiment. If the tracepoint was previously disabled, then collection
40637 of data from it will resume.
40639 @item QTDisable:@var{n}:@var{addr}
40641 @cindex @samp{QTDisable} packet
40642 Disable tracepoint @var{n} at address @var{addr} in a started tracepoint
40643 experiment. No more data will be collected from the tracepoint unless
40644 @samp{QTEnable:@var{n}:@var{addr}} is subsequently issued.
40647 @cindex @samp{QTinit} packet
40648 Clear the table of tracepoints, and empty the trace frame buffer.
40650 @item QTro:@var{start1},@var{end1}:@var{start2},@var{end2}:@dots{}
40651 @cindex @samp{QTro} packet
40652 Establish the given ranges of memory as ``transparent''. The stub
40653 will answer requests for these ranges from memory's current contents,
40654 if they were not collected as part of the tracepoint hit.
40656 @value{GDBN} uses this to mark read-only regions of memory, like those
40657 containing program code. Since these areas never change, they should
40658 still have the same contents they did when the tracepoint was hit, so
40659 there's no reason for the stub to refuse to provide their contents.
40661 @item QTDisconnected:@var{value}
40662 @cindex @samp{QTDisconnected} packet
40663 Set the choice to what to do with the tracing run when @value{GDBN}
40664 disconnects from the target. A @var{value} of 1 directs the target to
40665 continue the tracing run, while 0 tells the target to stop tracing if
40666 @value{GDBN} is no longer in the picture.
40669 @cindex @samp{qTStatus} packet
40670 Ask the stub if there is a trace experiment running right now.
40672 The reply has the form:
40676 @item T@var{running}@r{[};@var{field}@r{]}@dots{}
40677 @var{running} is a single digit @code{1} if the trace is presently
40678 running, or @code{0} if not. It is followed by semicolon-separated
40679 optional fields that an agent may use to report additional status.
40683 If the trace is not running, the agent may report any of several
40684 explanations as one of the optional fields:
40689 No trace has been run yet.
40691 @item tstop[:@var{text}]:0
40692 The trace was stopped by a user-originated stop command. The optional
40693 @var{text} field is a user-supplied string supplied as part of the
40694 stop command (for instance, an explanation of why the trace was
40695 stopped manually). It is hex-encoded.
40698 The trace stopped because the trace buffer filled up.
40700 @item tdisconnected:0
40701 The trace stopped because @value{GDBN} disconnected from the target.
40703 @item tpasscount:@var{tpnum}
40704 The trace stopped because tracepoint @var{tpnum} exceeded its pass count.
40706 @item terror:@var{text}:@var{tpnum}
40707 The trace stopped because tracepoint @var{tpnum} had an error. The
40708 string @var{text} is available to describe the nature of the error
40709 (for instance, a divide by zero in the condition expression); it
40713 The trace stopped for some other reason.
40717 Additional optional fields supply statistical and other information.
40718 Although not required, they are extremely useful for users monitoring
40719 the progress of a trace run. If a trace has stopped, and these
40720 numbers are reported, they must reflect the state of the just-stopped
40725 @item tframes:@var{n}
40726 The number of trace frames in the buffer.
40728 @item tcreated:@var{n}
40729 The total number of trace frames created during the run. This may
40730 be larger than the trace frame count, if the buffer is circular.
40732 @item tsize:@var{n}
40733 The total size of the trace buffer, in bytes.
40735 @item tfree:@var{n}
40736 The number of bytes still unused in the buffer.
40738 @item circular:@var{n}
40739 The value of the circular trace buffer flag. @code{1} means that the
40740 trace buffer is circular and old trace frames will be discarded if
40741 necessary to make room, @code{0} means that the trace buffer is linear
40744 @item disconn:@var{n}
40745 The value of the disconnected tracing flag. @code{1} means that
40746 tracing will continue after @value{GDBN} disconnects, @code{0} means
40747 that the trace run will stop.
40751 @item qTP:@var{tp}:@var{addr}
40752 @cindex tracepoint status, remote request
40753 @cindex @samp{qTP} packet
40754 Ask the stub for the current state of tracepoint number @var{tp} at
40755 address @var{addr}.
40759 @item V@var{hits}:@var{usage}
40760 The tracepoint has been hit @var{hits} times so far during the trace
40761 run, and accounts for @var{usage} in the trace buffer. Note that
40762 @code{while-stepping} steps are not counted as separate hits, but the
40763 steps' space consumption is added into the usage number.
40767 @item qTV:@var{var}
40768 @cindex trace state variable value, remote request
40769 @cindex @samp{qTV} packet
40770 Ask the stub for the value of the trace state variable number @var{var}.
40775 The value of the variable is @var{value}. This will be the current
40776 value of the variable if the user is examining a running target, or a
40777 saved value if the variable was collected in the trace frame that the
40778 user is looking at. Note that multiple requests may result in
40779 different reply values, such as when requesting values while the
40780 program is running.
40783 The value of the variable is unknown. This would occur, for example,
40784 if the user is examining a trace frame in which the requested variable
40789 @cindex @samp{qTfP} packet
40791 @cindex @samp{qTsP} packet
40792 These packets request data about tracepoints that are being used by
40793 the target. @value{GDBN} sends @code{qTfP} to get the first piece
40794 of data, and multiple @code{qTsP} to get additional pieces. Replies
40795 to these packets generally take the form of the @code{QTDP} packets
40796 that define tracepoints. (FIXME add detailed syntax)
40799 @cindex @samp{qTfV} packet
40801 @cindex @samp{qTsV} packet
40802 These packets request data about trace state variables that are on the
40803 target. @value{GDBN} sends @code{qTfV} to get the first vari of data,
40804 and multiple @code{qTsV} to get additional variables. Replies to
40805 these packets follow the syntax of the @code{QTDV} packets that define
40806 trace state variables.
40812 @cindex @samp{qTfSTM} packet
40813 @cindex @samp{qTsSTM} packet
40814 These packets request data about static tracepoint markers that exist
40815 in the target program. @value{GDBN} sends @code{qTfSTM} to get the
40816 first piece of data, and multiple @code{qTsSTM} to get additional
40817 pieces. Replies to these packets take the following form:
40821 @item m @var{address}:@var{id}:@var{extra}
40823 @item m @var{address}:@var{id}:@var{extra},@var{address}:@var{id}:@var{extra}@dots{}
40824 a comma-separated list of markers
40826 (lower case letter @samp{L}) denotes end of list.
40828 An error occurred. The error number @var{nn} is given as hex digits.
40830 An empty reply indicates that the request is not supported by the
40834 The @var{address} is encoded in hex;
40835 @var{id} and @var{extra} are strings encoded in hex.
40837 In response to each query, the target will reply with a list of one or
40838 more markers, separated by commas. @value{GDBN} will respond to each
40839 reply with a request for more markers (using the @samp{qs} form of the
40840 query), until the target responds with @samp{l} (lower-case ell, for
40843 @item qTSTMat:@var{address}
40845 @cindex @samp{qTSTMat} packet
40846 This packets requests data about static tracepoint markers in the
40847 target program at @var{address}. Replies to this packet follow the
40848 syntax of the @samp{qTfSTM} and @code{qTsSTM} packets that list static
40849 tracepoint markers.
40851 @item QTSave:@var{filename}
40852 @cindex @samp{QTSave} packet
40853 This packet directs the target to save trace data to the file name
40854 @var{filename} in the target's filesystem. The @var{filename} is encoded
40855 as a hex string; the interpretation of the file name (relative vs
40856 absolute, wild cards, etc) is up to the target.
40858 @item qTBuffer:@var{offset},@var{len}
40859 @cindex @samp{qTBuffer} packet
40860 Return up to @var{len} bytes of the current contents of trace buffer,
40861 starting at @var{offset}. The trace buffer is treated as if it were
40862 a contiguous collection of traceframes, as per the trace file format.
40863 The reply consists as many hex-encoded bytes as the target can deliver
40864 in a packet; it is not an error to return fewer than were asked for.
40865 A reply consisting of just @code{l} indicates that no bytes are
40868 @item QTBuffer:circular:@var{value}
40869 This packet directs the target to use a circular trace buffer if
40870 @var{value} is 1, or a linear buffer if the value is 0.
40872 @item QTBuffer:size:@var{size}
40873 @anchor{QTBuffer-size}
40874 @cindex @samp{QTBuffer size} packet
40875 This packet directs the target to make the trace buffer be of size
40876 @var{size} if possible. A value of @code{-1} tells the target to
40877 use whatever size it prefers.
40879 @item QTNotes:@r{[}@var{type}:@var{text}@r{]}@r{[};@var{type}:@var{text}@r{]}@dots{}
40880 @cindex @samp{QTNotes} packet
40881 This packet adds optional textual notes to the trace run. Allowable
40882 types include @code{user}, @code{notes}, and @code{tstop}, the
40883 @var{text} fields are arbitrary strings, hex-encoded.
40887 @subsection Relocate instruction reply packet
40888 When installing fast tracepoints in memory, the target may need to
40889 relocate the instruction currently at the tracepoint address to a
40890 different address in memory. For most instructions, a simple copy is
40891 enough, but, for example, call instructions that implicitly push the
40892 return address on the stack, and relative branches or other
40893 PC-relative instructions require offset adjustment, so that the effect
40894 of executing the instruction at a different address is the same as if
40895 it had executed in the original location.
40897 In response to several of the tracepoint packets, the target may also
40898 respond with a number of intermediate @samp{qRelocInsn} request
40899 packets before the final result packet, to have @value{GDBN} handle
40900 this relocation operation. If a packet supports this mechanism, its
40901 documentation will explicitly say so. See for example the above
40902 descriptions for the @samp{QTStart} and @samp{QTDP} packets. The
40903 format of the request is:
40906 @item qRelocInsn:@var{from};@var{to}
40908 This requests @value{GDBN} to copy instruction at address @var{from}
40909 to address @var{to}, possibly adjusted so that executing the
40910 instruction at @var{to} has the same effect as executing it at
40911 @var{from}. @value{GDBN} writes the adjusted instruction to target
40912 memory starting at @var{to}.
40917 @item qRelocInsn:@var{adjusted_size}
40918 Informs the stub the relocation is complete. The @var{adjusted_size} is
40919 the length in bytes of resulting relocated instruction sequence.
40921 A badly formed request was detected, or an error was encountered while
40922 relocating the instruction.
40925 @node Host I/O Packets
40926 @section Host I/O Packets
40927 @cindex Host I/O, remote protocol
40928 @cindex file transfer, remote protocol
40930 The @dfn{Host I/O} packets allow @value{GDBN} to perform I/O
40931 operations on the far side of a remote link. For example, Host I/O is
40932 used to upload and download files to a remote target with its own
40933 filesystem. Host I/O uses the same constant values and data structure
40934 layout as the target-initiated File-I/O protocol. However, the
40935 Host I/O packets are structured differently. The target-initiated
40936 protocol relies on target memory to store parameters and buffers.
40937 Host I/O requests are initiated by @value{GDBN}, and the
40938 target's memory is not involved. @xref{File-I/O Remote Protocol
40939 Extension}, for more details on the target-initiated protocol.
40941 The Host I/O request packets all encode a single operation along with
40942 its arguments. They have this format:
40946 @item vFile:@var{operation}: @var{parameter}@dots{}
40947 @var{operation} is the name of the particular request; the target
40948 should compare the entire packet name up to the second colon when checking
40949 for a supported operation. The format of @var{parameter} depends on
40950 the operation. Numbers are always passed in hexadecimal. Negative
40951 numbers have an explicit minus sign (i.e.@: two's complement is not
40952 used). Strings (e.g.@: filenames) are encoded as a series of
40953 hexadecimal bytes. The last argument to a system call may be a
40954 buffer of escaped binary data (@pxref{Binary Data}).
40958 The valid responses to Host I/O packets are:
40962 @item F @var{result} [, @var{errno}] [; @var{attachment}]
40963 @var{result} is the integer value returned by this operation, usually
40964 non-negative for success and -1 for errors. If an error has occured,
40965 @var{errno} will be included in the result specifying a
40966 value defined by the File-I/O protocol (@pxref{Errno Values}). For
40967 operations which return data, @var{attachment} supplies the data as a
40968 binary buffer. Binary buffers in response packets are escaped in the
40969 normal way (@pxref{Binary Data}). See the individual packet
40970 documentation for the interpretation of @var{result} and
40974 An empty response indicates that this operation is not recognized.
40978 These are the supported Host I/O operations:
40981 @item vFile:open: @var{filename}, @var{flags}, @var{mode}
40982 Open a file at @var{filename} and return a file descriptor for it, or
40983 return -1 if an error occurs. The @var{filename} is a string,
40984 @var{flags} is an integer indicating a mask of open flags
40985 (@pxref{Open Flags}), and @var{mode} is an integer indicating a mask
40986 of mode bits to use if the file is created (@pxref{mode_t Values}).
40987 @xref{open}, for details of the open flags and mode values.
40989 @item vFile:close: @var{fd}
40990 Close the open file corresponding to @var{fd} and return 0, or
40991 -1 if an error occurs.
40993 @item vFile:pread: @var{fd}, @var{count}, @var{offset}
40994 Read data from the open file corresponding to @var{fd}. Up to
40995 @var{count} bytes will be read from the file, starting at @var{offset}
40996 relative to the start of the file. The target may read fewer bytes;
40997 common reasons include packet size limits and an end-of-file
40998 condition. The number of bytes read is returned. Zero should only be
40999 returned for a successful read at the end of the file, or if
41000 @var{count} was zero.
41002 The data read should be returned as a binary attachment on success.
41003 If zero bytes were read, the response should include an empty binary
41004 attachment (i.e.@: a trailing semicolon). The return value is the
41005 number of target bytes read; the binary attachment may be longer if
41006 some characters were escaped.
41008 @item vFile:pwrite: @var{fd}, @var{offset}, @var{data}
41009 Write @var{data} (a binary buffer) to the open file corresponding
41010 to @var{fd}. Start the write at @var{offset} from the start of the
41011 file. Unlike many @code{write} system calls, there is no
41012 separate @var{count} argument; the length of @var{data} in the
41013 packet is used. @samp{vFile:write} returns the number of bytes written,
41014 which may be shorter than the length of @var{data}, or -1 if an
41017 @item vFile:fstat: @var{fd}
41018 Get information about the open file corresponding to @var{fd}.
41019 On success the information is returned as a binary attachment
41020 and the return value is the size of this attachment in bytes.
41021 If an error occurs the return value is -1. The format of the
41022 returned binary attachment is as described in @ref{struct stat}.
41024 @item vFile:unlink: @var{filename}
41025 Delete the file at @var{filename} on the target. Return 0,
41026 or -1 if an error occurs. The @var{filename} is a string.
41028 @item vFile:readlink: @var{filename}
41029 Read value of symbolic link @var{filename} on the target. Return
41030 the number of bytes read, or -1 if an error occurs.
41032 The data read should be returned as a binary attachment on success.
41033 If zero bytes were read, the response should include an empty binary
41034 attachment (i.e.@: a trailing semicolon). The return value is the
41035 number of target bytes read; the binary attachment may be longer if
41036 some characters were escaped.
41038 @item vFile:setfs: @var{pid}
41039 Select the filesystem on which @code{vFile} operations with
41040 @var{filename} arguments will operate. This is required for
41041 @value{GDBN} to be able to access files on remote targets where
41042 the remote stub does not share a common filesystem with the
41045 If @var{pid} is nonzero, select the filesystem as seen by process
41046 @var{pid}. If @var{pid} is zero, select the filesystem as seen by
41047 the remote stub. Return 0 on success, or -1 if an error occurs.
41048 If @code{vFile:setfs:} indicates success, the selected filesystem
41049 remains selected until the next successful @code{vFile:setfs:}
41055 @section Interrupts
41056 @cindex interrupts (remote protocol)
41057 @anchor{interrupting remote targets}
41059 In all-stop mode, when a program on the remote target is running,
41060 @value{GDBN} may attempt to interrupt it by sending a @samp{Ctrl-C},
41061 @code{BREAK} or a @code{BREAK} followed by @code{g}, control of which
41062 is specified via @value{GDBN}'s @samp{interrupt-sequence}.
41064 The precise meaning of @code{BREAK} is defined by the transport
41065 mechanism and may, in fact, be undefined. @value{GDBN} does not
41066 currently define a @code{BREAK} mechanism for any of the network
41067 interfaces except for TCP, in which case @value{GDBN} sends the
41068 @code{telnet} BREAK sequence.
41070 @samp{Ctrl-C}, on the other hand, is defined and implemented for all
41071 transport mechanisms. It is represented by sending the single byte
41072 @code{0x03} without any of the usual packet overhead described in
41073 the Overview section (@pxref{Overview}). When a @code{0x03} byte is
41074 transmitted as part of a packet, it is considered to be packet data
41075 and does @emph{not} represent an interrupt. E.g., an @samp{X} packet
41076 (@pxref{X packet}), used for binary downloads, may include an unescaped
41077 @code{0x03} as part of its packet.
41079 @code{BREAK} followed by @code{g} is also known as Magic SysRq g.
41080 When Linux kernel receives this sequence from serial port,
41081 it stops execution and connects to gdb.
41083 In non-stop mode, because packet resumptions are asynchronous
41084 (@pxref{vCont packet}), @value{GDBN} is always free to send a remote
41085 command to the remote stub, even when the target is running. For that
41086 reason, @value{GDBN} instead sends a regular packet (@pxref{vCtrlC
41087 packet}) with the usual packet framing instead of the single byte
41090 Stubs are not required to recognize these interrupt mechanisms and the
41091 precise meaning associated with receipt of the interrupt is
41092 implementation defined. If the target supports debugging of multiple
41093 threads and/or processes, it should attempt to interrupt all
41094 currently-executing threads and processes.
41095 If the stub is successful at interrupting the
41096 running program, it should send one of the stop
41097 reply packets (@pxref{Stop Reply Packets}) to @value{GDBN} as a result
41098 of successfully stopping the program in all-stop mode, and a stop reply
41099 for each stopped thread in non-stop mode.
41100 Interrupts received while the
41101 program is stopped are queued and the program will be interrupted when
41102 it is resumed next time.
41104 @node Notification Packets
41105 @section Notification Packets
41106 @cindex notification packets
41107 @cindex packets, notification
41109 The @value{GDBN} remote serial protocol includes @dfn{notifications},
41110 packets that require no acknowledgment. Both the GDB and the stub
41111 may send notifications (although the only notifications defined at
41112 present are sent by the stub). Notifications carry information
41113 without incurring the round-trip latency of an acknowledgment, and so
41114 are useful for low-impact communications where occasional packet loss
41117 A notification packet has the form @samp{% @var{data} #
41118 @var{checksum}}, where @var{data} is the content of the notification,
41119 and @var{checksum} is a checksum of @var{data}, computed and formatted
41120 as for ordinary @value{GDBN} packets. A notification's @var{data}
41121 never contains @samp{$}, @samp{%} or @samp{#} characters. Upon
41122 receiving a notification, the recipient sends no @samp{+} or @samp{-}
41123 to acknowledge the notification's receipt or to report its corruption.
41125 Every notification's @var{data} begins with a name, which contains no
41126 colon characters, followed by a colon character.
41128 Recipients should silently ignore corrupted notifications and
41129 notifications they do not understand. Recipients should restart
41130 timeout periods on receipt of a well-formed notification, whether or
41131 not they understand it.
41133 Senders should only send the notifications described here when this
41134 protocol description specifies that they are permitted. In the
41135 future, we may extend the protocol to permit existing notifications in
41136 new contexts; this rule helps older senders avoid confusing newer
41139 (Older versions of @value{GDBN} ignore bytes received until they see
41140 the @samp{$} byte that begins an ordinary packet, so new stubs may
41141 transmit notifications without fear of confusing older clients. There
41142 are no notifications defined for @value{GDBN} to send at the moment, but we
41143 assume that most older stubs would ignore them, as well.)
41145 Each notification is comprised of three parts:
41147 @item @var{name}:@var{event}
41148 The notification packet is sent by the side that initiates the
41149 exchange (currently, only the stub does that), with @var{event}
41150 carrying the specific information about the notification, and
41151 @var{name} specifying the name of the notification.
41153 The acknowledge sent by the other side, usually @value{GDBN}, to
41154 acknowledge the exchange and request the event.
41157 The purpose of an asynchronous notification mechanism is to report to
41158 @value{GDBN} that something interesting happened in the remote stub.
41160 The remote stub may send notification @var{name}:@var{event}
41161 at any time, but @value{GDBN} acknowledges the notification when
41162 appropriate. The notification event is pending before @value{GDBN}
41163 acknowledges. Only one notification at a time may be pending; if
41164 additional events occur before @value{GDBN} has acknowledged the
41165 previous notification, they must be queued by the stub for later
41166 synchronous transmission in response to @var{ack} packets from
41167 @value{GDBN}. Because the notification mechanism is unreliable,
41168 the stub is permitted to resend a notification if it believes
41169 @value{GDBN} may not have received it.
41171 Specifically, notifications may appear when @value{GDBN} is not
41172 otherwise reading input from the stub, or when @value{GDBN} is
41173 expecting to read a normal synchronous response or a
41174 @samp{+}/@samp{-} acknowledgment to a packet it has sent.
41175 Notification packets are distinct from any other communication from
41176 the stub so there is no ambiguity.
41178 After receiving a notification, @value{GDBN} shall acknowledge it by
41179 sending a @var{ack} packet as a regular, synchronous request to the
41180 stub. Such acknowledgment is not required to happen immediately, as
41181 @value{GDBN} is permitted to send other, unrelated packets to the
41182 stub first, which the stub should process normally.
41184 Upon receiving a @var{ack} packet, if the stub has other queued
41185 events to report to @value{GDBN}, it shall respond by sending a
41186 normal @var{event}. @value{GDBN} shall then send another @var{ack}
41187 packet to solicit further responses; again, it is permitted to send
41188 other, unrelated packets as well which the stub should process
41191 If the stub receives a @var{ack} packet and there are no additional
41192 @var{event} to report, the stub shall return an @samp{OK} response.
41193 At this point, @value{GDBN} has finished processing a notification
41194 and the stub has completed sending any queued events. @value{GDBN}
41195 won't accept any new notifications until the final @samp{OK} is
41196 received . If further notification events occur, the stub shall send
41197 a new notification, @value{GDBN} shall accept the notification, and
41198 the process shall be repeated.
41200 The process of asynchronous notification can be illustrated by the
41203 <- @code{%Stop:T0505:98e7ffbf;04:4ce6ffbf;08:b1b6e54c;thread:p7526.7526;core:0;}
41206 <- @code{T0505:68f37db7;04:40f37db7;08:63850408;thread:p7526.7528;core:0;}
41208 <- @code{T0505:68e3fdb6;04:40e3fdb6;08:63850408;thread:p7526.7529;core:0;}
41213 The following notifications are defined:
41214 @multitable @columnfractions 0.12 0.12 0.38 0.38
41223 @tab @var{reply}. The @var{reply} has the form of a stop reply, as
41224 described in @ref{Stop Reply Packets}. Refer to @ref{Remote Non-Stop},
41225 for information on how these notifications are acknowledged by
41227 @tab Report an asynchronous stop event in non-stop mode.
41231 @node Remote Non-Stop
41232 @section Remote Protocol Support for Non-Stop Mode
41234 @value{GDBN}'s remote protocol supports non-stop debugging of
41235 multi-threaded programs, as described in @ref{Non-Stop Mode}. If the stub
41236 supports non-stop mode, it should report that to @value{GDBN} by including
41237 @samp{QNonStop+} in its @samp{qSupported} response (@pxref{qSupported}).
41239 @value{GDBN} typically sends a @samp{QNonStop} packet only when
41240 establishing a new connection with the stub. Entering non-stop mode
41241 does not alter the state of any currently-running threads, but targets
41242 must stop all threads in any already-attached processes when entering
41243 all-stop mode. @value{GDBN} uses the @samp{?} packet as necessary to
41244 probe the target state after a mode change.
41246 In non-stop mode, when an attached process encounters an event that
41247 would otherwise be reported with a stop reply, it uses the
41248 asynchronous notification mechanism (@pxref{Notification Packets}) to
41249 inform @value{GDBN}. In contrast to all-stop mode, where all threads
41250 in all processes are stopped when a stop reply is sent, in non-stop
41251 mode only the thread reporting the stop event is stopped. That is,
41252 when reporting a @samp{S} or @samp{T} response to indicate completion
41253 of a step operation, hitting a breakpoint, or a fault, only the
41254 affected thread is stopped; any other still-running threads continue
41255 to run. When reporting a @samp{W} or @samp{X} response, all running
41256 threads belonging to other attached processes continue to run.
41258 In non-stop mode, the target shall respond to the @samp{?} packet as
41259 follows. First, any incomplete stop reply notification/@samp{vStopped}
41260 sequence in progress is abandoned. The target must begin a new
41261 sequence reporting stop events for all stopped threads, whether or not
41262 it has previously reported those events to @value{GDBN}. The first
41263 stop reply is sent as a synchronous reply to the @samp{?} packet, and
41264 subsequent stop replies are sent as responses to @samp{vStopped} packets
41265 using the mechanism described above. The target must not send
41266 asynchronous stop reply notifications until the sequence is complete.
41267 If all threads are running when the target receives the @samp{?} packet,
41268 or if the target is not attached to any process, it shall respond
41271 If the stub supports non-stop mode, it should also support the
41272 @samp{swbreak} stop reason if software breakpoints are supported, and
41273 the @samp{hwbreak} stop reason if hardware breakpoints are supported
41274 (@pxref{swbreak stop reason}). This is because given the asynchronous
41275 nature of non-stop mode, between the time a thread hits a breakpoint
41276 and the time the event is finally processed by @value{GDBN}, the
41277 breakpoint may have already been removed from the target. Due to
41278 this, @value{GDBN} needs to be able to tell whether a trap stop was
41279 caused by a delayed breakpoint event, which should be ignored, as
41280 opposed to a random trap signal, which should be reported to the user.
41281 Note the @samp{swbreak} feature implies that the target is responsible
41282 for adjusting the PC when a software breakpoint triggers, if
41283 necessary, such as on the x86 architecture.
41285 @node Packet Acknowledgment
41286 @section Packet Acknowledgment
41288 @cindex acknowledgment, for @value{GDBN} remote
41289 @cindex packet acknowledgment, for @value{GDBN} remote
41290 By default, when either the host or the target machine receives a packet,
41291 the first response expected is an acknowledgment: either @samp{+} (to indicate
41292 the package was received correctly) or @samp{-} (to request retransmission).
41293 This mechanism allows the @value{GDBN} remote protocol to operate over
41294 unreliable transport mechanisms, such as a serial line.
41296 In cases where the transport mechanism is itself reliable (such as a pipe or
41297 TCP connection), the @samp{+}/@samp{-} acknowledgments are redundant.
41298 It may be desirable to disable them in that case to reduce communication
41299 overhead, or for other reasons. This can be accomplished by means of the
41300 @samp{QStartNoAckMode} packet; @pxref{QStartNoAckMode}.
41302 When in no-acknowledgment mode, neither the stub nor @value{GDBN} shall send or
41303 expect @samp{+}/@samp{-} protocol acknowledgments. The packet
41304 and response format still includes the normal checksum, as described in
41305 @ref{Overview}, but the checksum may be ignored by the receiver.
41307 If the stub supports @samp{QStartNoAckMode} and prefers to operate in
41308 no-acknowledgment mode, it should report that to @value{GDBN}
41309 by including @samp{QStartNoAckMode+} in its response to @samp{qSupported};
41310 @pxref{qSupported}.
41311 If @value{GDBN} also supports @samp{QStartNoAckMode} and it has not been
41312 disabled via the @code{set remote noack-packet off} command
41313 (@pxref{Remote Configuration}),
41314 @value{GDBN} may then send a @samp{QStartNoAckMode} packet to the stub.
41315 Only then may the stub actually turn off packet acknowledgments.
41316 @value{GDBN} sends a final @samp{+} acknowledgment of the stub's @samp{OK}
41317 response, which can be safely ignored by the stub.
41319 Note that @code{set remote noack-packet} command only affects negotiation
41320 between @value{GDBN} and the stub when subsequent connections are made;
41321 it does not affect the protocol acknowledgment state for any current
41323 Since @samp{+}/@samp{-} acknowledgments are enabled by default when a
41324 new connection is established,
41325 there is also no protocol request to re-enable the acknowledgments
41326 for the current connection, once disabled.
41331 Example sequence of a target being re-started. Notice how the restart
41332 does not get any direct output:
41337 @emph{target restarts}
41340 <- @code{T001:1234123412341234}
41344 Example sequence of a target being stepped by a single instruction:
41347 -> @code{G1445@dots{}}
41352 <- @code{T001:1234123412341234}
41356 <- @code{1455@dots{}}
41360 @node File-I/O Remote Protocol Extension
41361 @section File-I/O Remote Protocol Extension
41362 @cindex File-I/O remote protocol extension
41365 * File-I/O Overview::
41366 * Protocol Basics::
41367 * The F Request Packet::
41368 * The F Reply Packet::
41369 * The Ctrl-C Message::
41371 * List of Supported Calls::
41372 * Protocol-specific Representation of Datatypes::
41374 * File-I/O Examples::
41377 @node File-I/O Overview
41378 @subsection File-I/O Overview
41379 @cindex file-i/o overview
41381 The @dfn{File I/O remote protocol extension} (short: File-I/O) allows the
41382 target to use the host's file system and console I/O to perform various
41383 system calls. System calls on the target system are translated into a
41384 remote protocol packet to the host system, which then performs the needed
41385 actions and returns a response packet to the target system.
41386 This simulates file system operations even on targets that lack file systems.
41388 The protocol is defined to be independent of both the host and target systems.
41389 It uses its own internal representation of datatypes and values. Both
41390 @value{GDBN} and the target's @value{GDBN} stub are responsible for
41391 translating the system-dependent value representations into the internal
41392 protocol representations when data is transmitted.
41394 The communication is synchronous. A system call is possible only when
41395 @value{GDBN} is waiting for a response from the @samp{C}, @samp{c}, @samp{S}
41396 or @samp{s} packets. While @value{GDBN} handles the request for a system call,
41397 the target is stopped to allow deterministic access to the target's
41398 memory. Therefore File-I/O is not interruptible by target signals. On
41399 the other hand, it is possible to interrupt File-I/O by a user interrupt
41400 (@samp{Ctrl-C}) within @value{GDBN}.
41402 The target's request to perform a host system call does not finish
41403 the latest @samp{C}, @samp{c}, @samp{S} or @samp{s} action. That means,
41404 after finishing the system call, the target returns to continuing the
41405 previous activity (continue, step). No additional continue or step
41406 request from @value{GDBN} is required.
41409 (@value{GDBP}) continue
41410 <- target requests 'system call X'
41411 target is stopped, @value{GDBN} executes system call
41412 -> @value{GDBN} returns result
41413 ... target continues, @value{GDBN} returns to wait for the target
41414 <- target hits breakpoint and sends a Txx packet
41417 The protocol only supports I/O on the console and to regular files on
41418 the host file system. Character or block special devices, pipes,
41419 named pipes, sockets or any other communication method on the host
41420 system are not supported by this protocol.
41422 File I/O is not supported in non-stop mode.
41424 @node Protocol Basics
41425 @subsection Protocol Basics
41426 @cindex protocol basics, file-i/o
41428 The File-I/O protocol uses the @code{F} packet as the request as well
41429 as reply packet. Since a File-I/O system call can only occur when
41430 @value{GDBN} is waiting for a response from the continuing or stepping target,
41431 the File-I/O request is a reply that @value{GDBN} has to expect as a result
41432 of a previous @samp{C}, @samp{c}, @samp{S} or @samp{s} packet.
41433 This @code{F} packet contains all information needed to allow @value{GDBN}
41434 to call the appropriate host system call:
41438 A unique identifier for the requested system call.
41441 All parameters to the system call. Pointers are given as addresses
41442 in the target memory address space. Pointers to strings are given as
41443 pointer/length pair. Numerical values are given as they are.
41444 Numerical control flags are given in a protocol-specific representation.
41448 At this point, @value{GDBN} has to perform the following actions.
41452 If the parameters include pointer values to data needed as input to a
41453 system call, @value{GDBN} requests this data from the target with a
41454 standard @code{m} packet request. This additional communication has to be
41455 expected by the target implementation and is handled as any other @code{m}
41459 @value{GDBN} translates all value from protocol representation to host
41460 representation as needed. Datatypes are coerced into the host types.
41463 @value{GDBN} calls the system call.
41466 It then coerces datatypes back to protocol representation.
41469 If the system call is expected to return data in buffer space specified
41470 by pointer parameters to the call, the data is transmitted to the
41471 target using a @code{M} or @code{X} packet. This packet has to be expected
41472 by the target implementation and is handled as any other @code{M} or @code{X}
41477 Eventually @value{GDBN} replies with another @code{F} packet which contains all
41478 necessary information for the target to continue. This at least contains
41485 @code{errno}, if has been changed by the system call.
41492 After having done the needed type and value coercion, the target continues
41493 the latest continue or step action.
41495 @node The F Request Packet
41496 @subsection The @code{F} Request Packet
41497 @cindex file-i/o request packet
41498 @cindex @code{F} request packet
41500 The @code{F} request packet has the following format:
41503 @item F@var{call-id},@var{parameter@dots{}}
41505 @var{call-id} is the identifier to indicate the host system call to be called.
41506 This is just the name of the function.
41508 @var{parameter@dots{}} are the parameters to the system call.
41509 Parameters are hexadecimal integer values, either the actual values in case
41510 of scalar datatypes, pointers to target buffer space in case of compound
41511 datatypes and unspecified memory areas, or pointer/length pairs in case
41512 of string parameters. These are appended to the @var{call-id} as a
41513 comma-delimited list. All values are transmitted in ASCII
41514 string representation, pointer/length pairs separated by a slash.
41520 @node The F Reply Packet
41521 @subsection The @code{F} Reply Packet
41522 @cindex file-i/o reply packet
41523 @cindex @code{F} reply packet
41525 The @code{F} reply packet has the following format:
41529 @item F@var{retcode},@var{errno},@var{Ctrl-C flag};@var{call-specific attachment}
41531 @var{retcode} is the return code of the system call as hexadecimal value.
41533 @var{errno} is the @code{errno} set by the call, in protocol-specific
41535 This parameter can be omitted if the call was successful.
41537 @var{Ctrl-C flag} is only sent if the user requested a break. In this
41538 case, @var{errno} must be sent as well, even if the call was successful.
41539 The @var{Ctrl-C flag} itself consists of the character @samp{C}:
41546 or, if the call was interrupted before the host call has been performed:
41553 assuming 4 is the protocol-specific representation of @code{EINTR}.
41558 @node The Ctrl-C Message
41559 @subsection The @samp{Ctrl-C} Message
41560 @cindex ctrl-c message, in file-i/o protocol
41562 If the @samp{Ctrl-C} flag is set in the @value{GDBN}
41563 reply packet (@pxref{The F Reply Packet}),
41564 the target should behave as if it had
41565 gotten a break message. The meaning for the target is ``system call
41566 interrupted by @code{SIGINT}''. Consequentially, the target should actually stop
41567 (as with a break message) and return to @value{GDBN} with a @code{T02}
41570 It's important for the target to know in which
41571 state the system call was interrupted. There are two possible cases:
41575 The system call hasn't been performed on the host yet.
41578 The system call on the host has been finished.
41582 These two states can be distinguished by the target by the value of the
41583 returned @code{errno}. If it's the protocol representation of @code{EINTR}, the system
41584 call hasn't been performed. This is equivalent to the @code{EINTR} handling
41585 on POSIX systems. In any other case, the target may presume that the
41586 system call has been finished --- successfully or not --- and should behave
41587 as if the break message arrived right after the system call.
41589 @value{GDBN} must behave reliably. If the system call has not been called
41590 yet, @value{GDBN} may send the @code{F} reply immediately, setting @code{EINTR} as
41591 @code{errno} in the packet. If the system call on the host has been finished
41592 before the user requests a break, the full action must be finished by
41593 @value{GDBN}. This requires sending @code{M} or @code{X} packets as necessary.
41594 The @code{F} packet may only be sent when either nothing has happened
41595 or the full action has been completed.
41598 @subsection Console I/O
41599 @cindex console i/o as part of file-i/o
41601 By default and if not explicitly closed by the target system, the file
41602 descriptors 0, 1 and 2 are connected to the @value{GDBN} console. Output
41603 on the @value{GDBN} console is handled as any other file output operation
41604 (@code{write(1, @dots{})} or @code{write(2, @dots{})}). Console input is handled
41605 by @value{GDBN} so that after the target read request from file descriptor
41606 0 all following typing is buffered until either one of the following
41611 The user types @kbd{Ctrl-c}. The behaviour is as explained above, and the
41613 system call is treated as finished.
41616 The user presses @key{RET}. This is treated as end of input with a trailing
41620 The user types @kbd{Ctrl-d}. This is treated as end of input. No trailing
41621 character (neither newline nor @samp{Ctrl-D}) is appended to the input.
41625 If the user has typed more characters than fit in the buffer given to
41626 the @code{read} call, the trailing characters are buffered in @value{GDBN} until
41627 either another @code{read(0, @dots{})} is requested by the target, or debugging
41628 is stopped at the user's request.
41631 @node List of Supported Calls
41632 @subsection List of Supported Calls
41633 @cindex list of supported file-i/o calls
41650 @unnumberedsubsubsec open
41651 @cindex open, file-i/o system call
41656 int open(const char *pathname, int flags);
41657 int open(const char *pathname, int flags, mode_t mode);
41661 @samp{Fopen,@var{pathptr}/@var{len},@var{flags},@var{mode}}
41664 @var{flags} is the bitwise @code{OR} of the following values:
41668 If the file does not exist it will be created. The host
41669 rules apply as far as file ownership and time stamps
41673 When used with @code{O_CREAT}, if the file already exists it is
41674 an error and open() fails.
41677 If the file already exists and the open mode allows
41678 writing (@code{O_RDWR} or @code{O_WRONLY} is given) it will be
41679 truncated to zero length.
41682 The file is opened in append mode.
41685 The file is opened for reading only.
41688 The file is opened for writing only.
41691 The file is opened for reading and writing.
41695 Other bits are silently ignored.
41699 @var{mode} is the bitwise @code{OR} of the following values:
41703 User has read permission.
41706 User has write permission.
41709 Group has read permission.
41712 Group has write permission.
41715 Others have read permission.
41718 Others have write permission.
41722 Other bits are silently ignored.
41725 @item Return value:
41726 @code{open} returns the new file descriptor or -1 if an error
41733 @var{pathname} already exists and @code{O_CREAT} and @code{O_EXCL} were used.
41736 @var{pathname} refers to a directory.
41739 The requested access is not allowed.
41742 @var{pathname} was too long.
41745 A directory component in @var{pathname} does not exist.
41748 @var{pathname} refers to a device, pipe, named pipe or socket.
41751 @var{pathname} refers to a file on a read-only filesystem and
41752 write access was requested.
41755 @var{pathname} is an invalid pointer value.
41758 No space on device to create the file.
41761 The process already has the maximum number of files open.
41764 The limit on the total number of files open on the system
41768 The call was interrupted by the user.
41774 @unnumberedsubsubsec close
41775 @cindex close, file-i/o system call
41784 @samp{Fclose,@var{fd}}
41786 @item Return value:
41787 @code{close} returns zero on success, or -1 if an error occurred.
41793 @var{fd} isn't a valid open file descriptor.
41796 The call was interrupted by the user.
41802 @unnumberedsubsubsec read
41803 @cindex read, file-i/o system call
41808 int read(int fd, void *buf, unsigned int count);
41812 @samp{Fread,@var{fd},@var{bufptr},@var{count}}
41814 @item Return value:
41815 On success, the number of bytes read is returned.
41816 Zero indicates end of file. If count is zero, read
41817 returns zero as well. On error, -1 is returned.
41823 @var{fd} is not a valid file descriptor or is not open for
41827 @var{bufptr} is an invalid pointer value.
41830 The call was interrupted by the user.
41836 @unnumberedsubsubsec write
41837 @cindex write, file-i/o system call
41842 int write(int fd, const void *buf, unsigned int count);
41846 @samp{Fwrite,@var{fd},@var{bufptr},@var{count}}
41848 @item Return value:
41849 On success, the number of bytes written are returned.
41850 Zero indicates nothing was written. On error, -1
41857 @var{fd} is not a valid file descriptor or is not open for
41861 @var{bufptr} is an invalid pointer value.
41864 An attempt was made to write a file that exceeds the
41865 host-specific maximum file size allowed.
41868 No space on device to write the data.
41871 The call was interrupted by the user.
41877 @unnumberedsubsubsec lseek
41878 @cindex lseek, file-i/o system call
41883 long lseek (int fd, long offset, int flag);
41887 @samp{Flseek,@var{fd},@var{offset},@var{flag}}
41889 @var{flag} is one of:
41893 The offset is set to @var{offset} bytes.
41896 The offset is set to its current location plus @var{offset}
41900 The offset is set to the size of the file plus @var{offset}
41904 @item Return value:
41905 On success, the resulting unsigned offset in bytes from
41906 the beginning of the file is returned. Otherwise, a
41907 value of -1 is returned.
41913 @var{fd} is not a valid open file descriptor.
41916 @var{fd} is associated with the @value{GDBN} console.
41919 @var{flag} is not a proper value.
41922 The call was interrupted by the user.
41928 @unnumberedsubsubsec rename
41929 @cindex rename, file-i/o system call
41934 int rename(const char *oldpath, const char *newpath);
41938 @samp{Frename,@var{oldpathptr}/@var{len},@var{newpathptr}/@var{len}}
41940 @item Return value:
41941 On success, zero is returned. On error, -1 is returned.
41947 @var{newpath} is an existing directory, but @var{oldpath} is not a
41951 @var{newpath} is a non-empty directory.
41954 @var{oldpath} or @var{newpath} is a directory that is in use by some
41958 An attempt was made to make a directory a subdirectory
41962 A component used as a directory in @var{oldpath} or new
41963 path is not a directory. Or @var{oldpath} is a directory
41964 and @var{newpath} exists but is not a directory.
41967 @var{oldpathptr} or @var{newpathptr} are invalid pointer values.
41970 No access to the file or the path of the file.
41974 @var{oldpath} or @var{newpath} was too long.
41977 A directory component in @var{oldpath} or @var{newpath} does not exist.
41980 The file is on a read-only filesystem.
41983 The device containing the file has no room for the new
41987 The call was interrupted by the user.
41993 @unnumberedsubsubsec unlink
41994 @cindex unlink, file-i/o system call
41999 int unlink(const char *pathname);
42003 @samp{Funlink,@var{pathnameptr}/@var{len}}
42005 @item Return value:
42006 On success, zero is returned. On error, -1 is returned.
42012 No access to the file or the path of the file.
42015 The system does not allow unlinking of directories.
42018 The file @var{pathname} cannot be unlinked because it's
42019 being used by another process.
42022 @var{pathnameptr} is an invalid pointer value.
42025 @var{pathname} was too long.
42028 A directory component in @var{pathname} does not exist.
42031 A component of the path is not a directory.
42034 The file is on a read-only filesystem.
42037 The call was interrupted by the user.
42043 @unnumberedsubsubsec stat/fstat
42044 @cindex fstat, file-i/o system call
42045 @cindex stat, file-i/o system call
42050 int stat(const char *pathname, struct stat *buf);
42051 int fstat(int fd, struct stat *buf);
42055 @samp{Fstat,@var{pathnameptr}/@var{len},@var{bufptr}}@*
42056 @samp{Ffstat,@var{fd},@var{bufptr}}
42058 @item Return value:
42059 On success, zero is returned. On error, -1 is returned.
42065 @var{fd} is not a valid open file.
42068 A directory component in @var{pathname} does not exist or the
42069 path is an empty string.
42072 A component of the path is not a directory.
42075 @var{pathnameptr} is an invalid pointer value.
42078 No access to the file or the path of the file.
42081 @var{pathname} was too long.
42084 The call was interrupted by the user.
42090 @unnumberedsubsubsec gettimeofday
42091 @cindex gettimeofday, file-i/o system call
42096 int gettimeofday(struct timeval *tv, void *tz);
42100 @samp{Fgettimeofday,@var{tvptr},@var{tzptr}}
42102 @item Return value:
42103 On success, 0 is returned, -1 otherwise.
42109 @var{tz} is a non-NULL pointer.
42112 @var{tvptr} and/or @var{tzptr} is an invalid pointer value.
42118 @unnumberedsubsubsec isatty
42119 @cindex isatty, file-i/o system call
42124 int isatty(int fd);
42128 @samp{Fisatty,@var{fd}}
42130 @item Return value:
42131 Returns 1 if @var{fd} refers to the @value{GDBN} console, 0 otherwise.
42137 The call was interrupted by the user.
42142 Note that the @code{isatty} call is treated as a special case: it returns
42143 1 to the target if the file descriptor is attached
42144 to the @value{GDBN} console, 0 otherwise. Implementing through system calls
42145 would require implementing @code{ioctl} and would be more complex than
42150 @unnumberedsubsubsec system
42151 @cindex system, file-i/o system call
42156 int system(const char *command);
42160 @samp{Fsystem,@var{commandptr}/@var{len}}
42162 @item Return value:
42163 If @var{len} is zero, the return value indicates whether a shell is
42164 available. A zero return value indicates a shell is not available.
42165 For non-zero @var{len}, the value returned is -1 on error and the
42166 return status of the command otherwise. Only the exit status of the
42167 command is returned, which is extracted from the host's @code{system}
42168 return value by calling @code{WEXITSTATUS(retval)}. In case
42169 @file{/bin/sh} could not be executed, 127 is returned.
42175 The call was interrupted by the user.
42180 @value{GDBN} takes over the full task of calling the necessary host calls
42181 to perform the @code{system} call. The return value of @code{system} on
42182 the host is simplified before it's returned
42183 to the target. Any termination signal information from the child process
42184 is discarded, and the return value consists
42185 entirely of the exit status of the called command.
42187 Due to security concerns, the @code{system} call is by default refused
42188 by @value{GDBN}. The user has to allow this call explicitly with the
42189 @code{set remote system-call-allowed 1} command.
42192 @item set remote system-call-allowed
42193 @kindex set remote system-call-allowed
42194 Control whether to allow the @code{system} calls in the File I/O
42195 protocol for the remote target. The default is zero (disabled).
42197 @item show remote system-call-allowed
42198 @kindex show remote system-call-allowed
42199 Show whether the @code{system} calls are allowed in the File I/O
42203 @node Protocol-specific Representation of Datatypes
42204 @subsection Protocol-specific Representation of Datatypes
42205 @cindex protocol-specific representation of datatypes, in file-i/o protocol
42208 * Integral Datatypes::
42210 * Memory Transfer::
42215 @node Integral Datatypes
42216 @unnumberedsubsubsec Integral Datatypes
42217 @cindex integral datatypes, in file-i/o protocol
42219 The integral datatypes used in the system calls are @code{int},
42220 @code{unsigned int}, @code{long}, @code{unsigned long},
42221 @code{mode_t}, and @code{time_t}.
42223 @code{int}, @code{unsigned int}, @code{mode_t} and @code{time_t} are
42224 implemented as 32 bit values in this protocol.
42226 @code{long} and @code{unsigned long} are implemented as 64 bit types.
42228 @xref{Limits}, for corresponding MIN and MAX values (similar to those
42229 in @file{limits.h}) to allow range checking on host and target.
42231 @code{time_t} datatypes are defined as seconds since the Epoch.
42233 All integral datatypes transferred as part of a memory read or write of a
42234 structured datatype e.g.@: a @code{struct stat} have to be given in big endian
42237 @node Pointer Values
42238 @unnumberedsubsubsec Pointer Values
42239 @cindex pointer values, in file-i/o protocol
42241 Pointers to target data are transmitted as they are. An exception
42242 is made for pointers to buffers for which the length isn't
42243 transmitted as part of the function call, namely strings. Strings
42244 are transmitted as a pointer/length pair, both as hex values, e.g.@:
42251 which is a pointer to data of length 18 bytes at position 0x1aaf.
42252 The length is defined as the full string length in bytes, including
42253 the trailing null byte. For example, the string @code{"hello world"}
42254 at address 0x123456 is transmitted as
42260 @node Memory Transfer
42261 @unnumberedsubsubsec Memory Transfer
42262 @cindex memory transfer, in file-i/o protocol
42264 Structured data which is transferred using a memory read or write (for
42265 example, a @code{struct stat}) is expected to be in a protocol-specific format
42266 with all scalar multibyte datatypes being big endian. Translation to
42267 this representation needs to be done both by the target before the @code{F}
42268 packet is sent, and by @value{GDBN} before
42269 it transfers memory to the target. Transferred pointers to structured
42270 data should point to the already-coerced data at any time.
42274 @unnumberedsubsubsec struct stat
42275 @cindex struct stat, in file-i/o protocol
42277 The buffer of type @code{struct stat} used by the target and @value{GDBN}
42278 is defined as follows:
42282 unsigned int st_dev; /* device */
42283 unsigned int st_ino; /* inode */
42284 mode_t st_mode; /* protection */
42285 unsigned int st_nlink; /* number of hard links */
42286 unsigned int st_uid; /* user ID of owner */
42287 unsigned int st_gid; /* group ID of owner */
42288 unsigned int st_rdev; /* device type (if inode device) */
42289 unsigned long st_size; /* total size, in bytes */
42290 unsigned long st_blksize; /* blocksize for filesystem I/O */
42291 unsigned long st_blocks; /* number of blocks allocated */
42292 time_t st_atime; /* time of last access */
42293 time_t st_mtime; /* time of last modification */
42294 time_t st_ctime; /* time of last change */
42298 The integral datatypes conform to the definitions given in the
42299 appropriate section (see @ref{Integral Datatypes}, for details) so this
42300 structure is of size 64 bytes.
42302 The values of several fields have a restricted meaning and/or
42308 A value of 0 represents a file, 1 the console.
42311 No valid meaning for the target. Transmitted unchanged.
42314 Valid mode bits are described in @ref{Constants}. Any other
42315 bits have currently no meaning for the target.
42320 No valid meaning for the target. Transmitted unchanged.
42325 These values have a host and file system dependent
42326 accuracy. Especially on Windows hosts, the file system may not
42327 support exact timing values.
42330 The target gets a @code{struct stat} of the above representation and is
42331 responsible for coercing it to the target representation before
42334 Note that due to size differences between the host, target, and protocol
42335 representations of @code{struct stat} members, these members could eventually
42336 get truncated on the target.
42338 @node struct timeval
42339 @unnumberedsubsubsec struct timeval
42340 @cindex struct timeval, in file-i/o protocol
42342 The buffer of type @code{struct timeval} used by the File-I/O protocol
42343 is defined as follows:
42347 time_t tv_sec; /* second */
42348 long tv_usec; /* microsecond */
42352 The integral datatypes conform to the definitions given in the
42353 appropriate section (see @ref{Integral Datatypes}, for details) so this
42354 structure is of size 8 bytes.
42357 @subsection Constants
42358 @cindex constants, in file-i/o protocol
42360 The following values are used for the constants inside of the
42361 protocol. @value{GDBN} and target are responsible for translating these
42362 values before and after the call as needed.
42373 @unnumberedsubsubsec Open Flags
42374 @cindex open flags, in file-i/o protocol
42376 All values are given in hexadecimal representation.
42388 @node mode_t Values
42389 @unnumberedsubsubsec mode_t Values
42390 @cindex mode_t values, in file-i/o protocol
42392 All values are given in octal representation.
42409 @unnumberedsubsubsec Errno Values
42410 @cindex errno values, in file-i/o protocol
42412 All values are given in decimal representation.
42437 @code{EUNKNOWN} is used as a fallback error value if a host system returns
42438 any error value not in the list of supported error numbers.
42441 @unnumberedsubsubsec Lseek Flags
42442 @cindex lseek flags, in file-i/o protocol
42451 @unnumberedsubsubsec Limits
42452 @cindex limits, in file-i/o protocol
42454 All values are given in decimal representation.
42457 INT_MIN -2147483648
42459 UINT_MAX 4294967295
42460 LONG_MIN -9223372036854775808
42461 LONG_MAX 9223372036854775807
42462 ULONG_MAX 18446744073709551615
42465 @node File-I/O Examples
42466 @subsection File-I/O Examples
42467 @cindex file-i/o examples
42469 Example sequence of a write call, file descriptor 3, buffer is at target
42470 address 0x1234, 6 bytes should be written:
42473 <- @code{Fwrite,3,1234,6}
42474 @emph{request memory read from target}
42477 @emph{return "6 bytes written"}
42481 Example sequence of a read call, file descriptor 3, buffer is at target
42482 address 0x1234, 6 bytes should be read:
42485 <- @code{Fread,3,1234,6}
42486 @emph{request memory write to target}
42487 -> @code{X1234,6:XXXXXX}
42488 @emph{return "6 bytes read"}
42492 Example sequence of a read call, call fails on the host due to invalid
42493 file descriptor (@code{EBADF}):
42496 <- @code{Fread,3,1234,6}
42500 Example sequence of a read call, user presses @kbd{Ctrl-c} before syscall on
42504 <- @code{Fread,3,1234,6}
42509 Example sequence of a read call, user presses @kbd{Ctrl-c} after syscall on
42513 <- @code{Fread,3,1234,6}
42514 -> @code{X1234,6:XXXXXX}
42518 @node Library List Format
42519 @section Library List Format
42520 @cindex library list format, remote protocol
42522 On some platforms, a dynamic loader (e.g.@: @file{ld.so}) runs in the
42523 same process as your application to manage libraries. In this case,
42524 @value{GDBN} can use the loader's symbol table and normal memory
42525 operations to maintain a list of shared libraries. On other
42526 platforms, the operating system manages loaded libraries.
42527 @value{GDBN} can not retrieve the list of currently loaded libraries
42528 through memory operations, so it uses the @samp{qXfer:libraries:read}
42529 packet (@pxref{qXfer library list read}) instead. The remote stub
42530 queries the target's operating system and reports which libraries
42533 The @samp{qXfer:libraries:read} packet returns an XML document which
42534 lists loaded libraries and their offsets. Each library has an
42535 associated name and one or more segment or section base addresses,
42536 which report where the library was loaded in memory.
42538 For the common case of libraries that are fully linked binaries, the
42539 library should have a list of segments. If the target supports
42540 dynamic linking of a relocatable object file, its library XML element
42541 should instead include a list of allocated sections. The segment or
42542 section bases are start addresses, not relocation offsets; they do not
42543 depend on the library's link-time base addresses.
42545 @value{GDBN} must be linked with the Expat library to support XML
42546 library lists. @xref{Expat}.
42548 A simple memory map, with one loaded library relocated by a single
42549 offset, looks like this:
42553 <library name="/lib/libc.so.6">
42554 <segment address="0x10000000"/>
42559 Another simple memory map, with one loaded library with three
42560 allocated sections (.text, .data, .bss), looks like this:
42564 <library name="sharedlib.o">
42565 <section address="0x10000000"/>
42566 <section address="0x20000000"/>
42567 <section address="0x30000000"/>
42572 The format of a library list is described by this DTD:
42575 <!-- library-list: Root element with versioning -->
42576 <!ELEMENT library-list (library)*>
42577 <!ATTLIST library-list version CDATA #FIXED "1.0">
42578 <!ELEMENT library (segment*, section*)>
42579 <!ATTLIST library name CDATA #REQUIRED>
42580 <!ELEMENT segment EMPTY>
42581 <!ATTLIST segment address CDATA #REQUIRED>
42582 <!ELEMENT section EMPTY>
42583 <!ATTLIST section address CDATA #REQUIRED>
42586 In addition, segments and section descriptors cannot be mixed within a
42587 single library element, and you must supply at least one segment or
42588 section for each library.
42590 @node Library List Format for SVR4 Targets
42591 @section Library List Format for SVR4 Targets
42592 @cindex library list format, remote protocol
42594 On SVR4 platforms @value{GDBN} can use the symbol table of a dynamic loader
42595 (e.g.@: @file{ld.so}) and normal memory operations to maintain a list of
42596 shared libraries. Still a special library list provided by this packet is
42597 more efficient for the @value{GDBN} remote protocol.
42599 The @samp{qXfer:libraries-svr4:read} packet returns an XML document which lists
42600 loaded libraries and their SVR4 linker parameters. For each library on SVR4
42601 target, the following parameters are reported:
42605 @code{name}, the absolute file name from the @code{l_name} field of
42606 @code{struct link_map}.
42608 @code{lm} with address of @code{struct link_map} used for TLS
42609 (Thread Local Storage) access.
42611 @code{l_addr}, the displacement as read from the field @code{l_addr} of
42612 @code{struct link_map}. For prelinked libraries this is not an absolute
42613 memory address. It is a displacement of absolute memory address against
42614 address the file was prelinked to during the library load.
42616 @code{l_ld}, which is memory address of the @code{PT_DYNAMIC} segment
42619 Additionally the single @code{main-lm} attribute specifies address of
42620 @code{struct link_map} used for the main executable. This parameter is used
42621 for TLS access and its presence is optional.
42623 @value{GDBN} must be linked with the Expat library to support XML
42624 SVR4 library lists. @xref{Expat}.
42626 A simple memory map, with two loaded libraries (which do not use prelink),
42630 <library-list-svr4 version="1.0" main-lm="0xe4f8f8">
42631 <library name="/lib/ld-linux.so.2" lm="0xe4f51c" l_addr="0xe2d000"
42633 <library name="/lib/libc.so.6" lm="0xe4fbe8" l_addr="0x154000"
42635 </library-list-svr>
42638 The format of an SVR4 library list is described by this DTD:
42641 <!-- library-list-svr4: Root element with versioning -->
42642 <!ELEMENT library-list-svr4 (library)*>
42643 <!ATTLIST library-list-svr4 version CDATA #FIXED "1.0">
42644 <!ATTLIST library-list-svr4 main-lm CDATA #IMPLIED>
42645 <!ELEMENT library EMPTY>
42646 <!ATTLIST library name CDATA #REQUIRED>
42647 <!ATTLIST library lm CDATA #REQUIRED>
42648 <!ATTLIST library l_addr CDATA #REQUIRED>
42649 <!ATTLIST library l_ld CDATA #REQUIRED>
42652 @node Memory Map Format
42653 @section Memory Map Format
42654 @cindex memory map format
42656 To be able to write into flash memory, @value{GDBN} needs to obtain a
42657 memory map from the target. This section describes the format of the
42660 The memory map is obtained using the @samp{qXfer:memory-map:read}
42661 (@pxref{qXfer memory map read}) packet and is an XML document that
42662 lists memory regions.
42664 @value{GDBN} must be linked with the Expat library to support XML
42665 memory maps. @xref{Expat}.
42667 The top-level structure of the document is shown below:
42670 <?xml version="1.0"?>
42671 <!DOCTYPE memory-map
42672 PUBLIC "+//IDN gnu.org//DTD GDB Memory Map V1.0//EN"
42673 "http://sourceware.org/gdb/gdb-memory-map.dtd">
42679 Each region can be either:
42684 A region of RAM starting at @var{addr} and extending for @var{length}
42688 <memory type="ram" start="@var{addr}" length="@var{length}"/>
42693 A region of read-only memory:
42696 <memory type="rom" start="@var{addr}" length="@var{length}"/>
42701 A region of flash memory, with erasure blocks @var{blocksize}
42705 <memory type="flash" start="@var{addr}" length="@var{length}">
42706 <property name="blocksize">@var{blocksize}</property>
42712 Regions must not overlap. @value{GDBN} assumes that areas of memory not covered
42713 by the memory map are RAM, and uses the ordinary @samp{M} and @samp{X}
42714 packets to write to addresses in such ranges.
42716 The formal DTD for memory map format is given below:
42719 <!-- ................................................... -->
42720 <!-- Memory Map XML DTD ................................ -->
42721 <!-- File: memory-map.dtd .............................. -->
42722 <!-- .................................... .............. -->
42723 <!-- memory-map.dtd -->
42724 <!-- memory-map: Root element with versioning -->
42725 <!ELEMENT memory-map (memory)*>
42726 <!ATTLIST memory-map version CDATA #FIXED "1.0.0">
42727 <!ELEMENT memory (property)*>
42728 <!-- memory: Specifies a memory region,
42729 and its type, or device. -->
42730 <!ATTLIST memory type (ram|rom|flash) #REQUIRED
42731 start CDATA #REQUIRED
42732 length CDATA #REQUIRED>
42733 <!-- property: Generic attribute tag -->
42734 <!ELEMENT property (#PCDATA | property)*>
42735 <!ATTLIST property name (blocksize) #REQUIRED>
42738 @node Thread List Format
42739 @section Thread List Format
42740 @cindex thread list format
42742 To efficiently update the list of threads and their attributes,
42743 @value{GDBN} issues the @samp{qXfer:threads:read} packet
42744 (@pxref{qXfer threads read}) and obtains the XML document with
42745 the following structure:
42748 <?xml version="1.0"?>
42750 <thread id="id" core="0" name="name">
42751 ... description ...
42756 Each @samp{thread} element must have the @samp{id} attribute that
42757 identifies the thread (@pxref{thread-id syntax}). The
42758 @samp{core} attribute, if present, specifies which processor core
42759 the thread was last executing on. The @samp{name} attribute, if
42760 present, specifies the human-readable name of the thread. The content
42761 of the of @samp{thread} element is interpreted as human-readable
42762 auxiliary information. The @samp{handle} attribute, if present,
42763 is a hex encoded representation of the thread handle.
42766 @node Traceframe Info Format
42767 @section Traceframe Info Format
42768 @cindex traceframe info format
42770 To be able to know which objects in the inferior can be examined when
42771 inspecting a tracepoint hit, @value{GDBN} needs to obtain the list of
42772 memory ranges, registers and trace state variables that have been
42773 collected in a traceframe.
42775 This list is obtained using the @samp{qXfer:traceframe-info:read}
42776 (@pxref{qXfer traceframe info read}) packet and is an XML document.
42778 @value{GDBN} must be linked with the Expat library to support XML
42779 traceframe info discovery. @xref{Expat}.
42781 The top-level structure of the document is shown below:
42784 <?xml version="1.0"?>
42785 <!DOCTYPE traceframe-info
42786 PUBLIC "+//IDN gnu.org//DTD GDB Memory Map V1.0//EN"
42787 "http://sourceware.org/gdb/gdb-traceframe-info.dtd">
42793 Each traceframe block can be either:
42798 A region of collected memory starting at @var{addr} and extending for
42799 @var{length} bytes from there:
42802 <memory start="@var{addr}" length="@var{length}"/>
42806 A block indicating trace state variable numbered @var{number} has been
42810 <tvar id="@var{number}"/>
42815 The formal DTD for the traceframe info format is given below:
42818 <!ELEMENT traceframe-info (memory | tvar)* >
42819 <!ATTLIST traceframe-info version CDATA #FIXED "1.0">
42821 <!ELEMENT memory EMPTY>
42822 <!ATTLIST memory start CDATA #REQUIRED
42823 length CDATA #REQUIRED>
42825 <!ATTLIST tvar id CDATA #REQUIRED>
42828 @node Branch Trace Format
42829 @section Branch Trace Format
42830 @cindex branch trace format
42832 In order to display the branch trace of an inferior thread,
42833 @value{GDBN} needs to obtain the list of branches. This list is
42834 represented as list of sequential code blocks that are connected via
42835 branches. The code in each block has been executed sequentially.
42837 This list is obtained using the @samp{qXfer:btrace:read}
42838 (@pxref{qXfer btrace read}) packet and is an XML document.
42840 @value{GDBN} must be linked with the Expat library to support XML
42841 traceframe info discovery. @xref{Expat}.
42843 The top-level structure of the document is shown below:
42846 <?xml version="1.0"?>
42848 PUBLIC "+//IDN gnu.org//DTD GDB Branch Trace V1.0//EN"
42849 "http://sourceware.org/gdb/gdb-btrace.dtd">
42858 A block of sequentially executed instructions starting at @var{begin}
42859 and ending at @var{end}:
42862 <block begin="@var{begin}" end="@var{end}"/>
42867 The formal DTD for the branch trace format is given below:
42870 <!ELEMENT btrace (block* | pt) >
42871 <!ATTLIST btrace version CDATA #FIXED "1.0">
42873 <!ELEMENT block EMPTY>
42874 <!ATTLIST block begin CDATA #REQUIRED
42875 end CDATA #REQUIRED>
42877 <!ELEMENT pt (pt-config?, raw?)>
42879 <!ELEMENT pt-config (cpu?)>
42881 <!ELEMENT cpu EMPTY>
42882 <!ATTLIST cpu vendor CDATA #REQUIRED
42883 family CDATA #REQUIRED
42884 model CDATA #REQUIRED
42885 stepping CDATA #REQUIRED>
42887 <!ELEMENT raw (#PCDATA)>
42890 @node Branch Trace Configuration Format
42891 @section Branch Trace Configuration Format
42892 @cindex branch trace configuration format
42894 For each inferior thread, @value{GDBN} can obtain the branch trace
42895 configuration using the @samp{qXfer:btrace-conf:read}
42896 (@pxref{qXfer btrace-conf read}) packet.
42898 The configuration describes the branch trace format and configuration
42899 settings for that format. The following information is described:
42903 This thread uses the @dfn{Branch Trace Store} (@acronym{BTS}) format.
42906 The size of the @acronym{BTS} ring buffer in bytes.
42909 This thread uses the @dfn{Intel Processor Trace} (@acronym{Intel
42913 The size of the @acronym{Intel PT} ring buffer in bytes.
42917 @value{GDBN} must be linked with the Expat library to support XML
42918 branch trace configuration discovery. @xref{Expat}.
42920 The formal DTD for the branch trace configuration format is given below:
42923 <!ELEMENT btrace-conf (bts?, pt?)>
42924 <!ATTLIST btrace-conf version CDATA #FIXED "1.0">
42926 <!ELEMENT bts EMPTY>
42927 <!ATTLIST bts size CDATA #IMPLIED>
42929 <!ELEMENT pt EMPTY>
42930 <!ATTLIST pt size CDATA #IMPLIED>
42933 @include agentexpr.texi
42935 @node Target Descriptions
42936 @appendix Target Descriptions
42937 @cindex target descriptions
42939 One of the challenges of using @value{GDBN} to debug embedded systems
42940 is that there are so many minor variants of each processor
42941 architecture in use. It is common practice for vendors to start with
42942 a standard processor core --- ARM, PowerPC, or @acronym{MIPS}, for example ---
42943 and then make changes to adapt it to a particular market niche. Some
42944 architectures have hundreds of variants, available from dozens of
42945 vendors. This leads to a number of problems:
42949 With so many different customized processors, it is difficult for
42950 the @value{GDBN} maintainers to keep up with the changes.
42952 Since individual variants may have short lifetimes or limited
42953 audiences, it may not be worthwhile to carry information about every
42954 variant in the @value{GDBN} source tree.
42956 When @value{GDBN} does support the architecture of the embedded system
42957 at hand, the task of finding the correct architecture name to give the
42958 @command{set architecture} command can be error-prone.
42961 To address these problems, the @value{GDBN} remote protocol allows a
42962 target system to not only identify itself to @value{GDBN}, but to
42963 actually describe its own features. This lets @value{GDBN} support
42964 processor variants it has never seen before --- to the extent that the
42965 descriptions are accurate, and that @value{GDBN} understands them.
42967 @value{GDBN} must be linked with the Expat library to support XML
42968 target descriptions. @xref{Expat}.
42971 * Retrieving Descriptions:: How descriptions are fetched from a target.
42972 * Target Description Format:: The contents of a target description.
42973 * Predefined Target Types:: Standard types available for target
42975 * Enum Target Types:: How to define enum target types.
42976 * Standard Target Features:: Features @value{GDBN} knows about.
42979 @node Retrieving Descriptions
42980 @section Retrieving Descriptions
42982 Target descriptions can be read from the target automatically, or
42983 specified by the user manually. The default behavior is to read the
42984 description from the target. @value{GDBN} retrieves it via the remote
42985 protocol using @samp{qXfer} requests (@pxref{General Query Packets,
42986 qXfer}). The @var{annex} in the @samp{qXfer} packet will be
42987 @samp{target.xml}. The contents of the @samp{target.xml} annex are an
42988 XML document, of the form described in @ref{Target Description
42991 Alternatively, you can specify a file to read for the target description.
42992 If a file is set, the target will not be queried. The commands to
42993 specify a file are:
42996 @cindex set tdesc filename
42997 @item set tdesc filename @var{path}
42998 Read the target description from @var{path}.
43000 @cindex unset tdesc filename
43001 @item unset tdesc filename
43002 Do not read the XML target description from a file. @value{GDBN}
43003 will use the description supplied by the current target.
43005 @cindex show tdesc filename
43006 @item show tdesc filename
43007 Show the filename to read for a target description, if any.
43011 @node Target Description Format
43012 @section Target Description Format
43013 @cindex target descriptions, XML format
43015 A target description annex is an @uref{http://www.w3.org/XML/, XML}
43016 document which complies with the Document Type Definition provided in
43017 the @value{GDBN} sources in @file{gdb/features/gdb-target.dtd}. This
43018 means you can use generally available tools like @command{xmllint} to
43019 check that your feature descriptions are well-formed and valid.
43020 However, to help people unfamiliar with XML write descriptions for
43021 their targets, we also describe the grammar here.
43023 Target descriptions can identify the architecture of the remote target
43024 and (for some architectures) provide information about custom register
43025 sets. They can also identify the OS ABI of the remote target.
43026 @value{GDBN} can use this information to autoconfigure for your
43027 target, or to warn you if you connect to an unsupported target.
43029 Here is a simple target description:
43032 <target version="1.0">
43033 <architecture>i386:x86-64</architecture>
43038 This minimal description only says that the target uses
43039 the x86-64 architecture.
43041 A target description has the following overall form, with [ ] marking
43042 optional elements and @dots{} marking repeatable elements. The elements
43043 are explained further below.
43046 <?xml version="1.0"?>
43047 <!DOCTYPE target SYSTEM "gdb-target.dtd">
43048 <target version="1.0">
43049 @r{[}@var{architecture}@r{]}
43050 @r{[}@var{osabi}@r{]}
43051 @r{[}@var{compatible}@r{]}
43052 @r{[}@var{feature}@dots{}@r{]}
43057 The description is generally insensitive to whitespace and line
43058 breaks, under the usual common-sense rules. The XML version
43059 declaration and document type declaration can generally be omitted
43060 (@value{GDBN} does not require them), but specifying them may be
43061 useful for XML validation tools. The @samp{version} attribute for
43062 @samp{<target>} may also be omitted, but we recommend
43063 including it; if future versions of @value{GDBN} use an incompatible
43064 revision of @file{gdb-target.dtd}, they will detect and report
43065 the version mismatch.
43067 @subsection Inclusion
43068 @cindex target descriptions, inclusion
43071 @cindex <xi:include>
43074 It can sometimes be valuable to split a target description up into
43075 several different annexes, either for organizational purposes, or to
43076 share files between different possible target descriptions. You can
43077 divide a description into multiple files by replacing any element of
43078 the target description with an inclusion directive of the form:
43081 <xi:include href="@var{document}"/>
43085 When @value{GDBN} encounters an element of this form, it will retrieve
43086 the named XML @var{document}, and replace the inclusion directive with
43087 the contents of that document. If the current description was read
43088 using @samp{qXfer}, then so will be the included document;
43089 @var{document} will be interpreted as the name of an annex. If the
43090 current description was read from a file, @value{GDBN} will look for
43091 @var{document} as a file in the same directory where it found the
43092 original description.
43094 @subsection Architecture
43095 @cindex <architecture>
43097 An @samp{<architecture>} element has this form:
43100 <architecture>@var{arch}</architecture>
43103 @var{arch} is one of the architectures from the set accepted by
43104 @code{set architecture} (@pxref{Targets, ,Specifying a Debugging Target}).
43107 @cindex @code{<osabi>}
43109 This optional field was introduced in @value{GDBN} version 7.0.
43110 Previous versions of @value{GDBN} ignore it.
43112 An @samp{<osabi>} element has this form:
43115 <osabi>@var{abi-name}</osabi>
43118 @var{abi-name} is an OS ABI name from the same selection accepted by
43119 @w{@code{set osabi}} (@pxref{ABI, ,Configuring the Current ABI}).
43121 @subsection Compatible Architecture
43122 @cindex @code{<compatible>}
43124 This optional field was introduced in @value{GDBN} version 7.0.
43125 Previous versions of @value{GDBN} ignore it.
43127 A @samp{<compatible>} element has this form:
43130 <compatible>@var{arch}</compatible>
43133 @var{arch} is one of the architectures from the set accepted by
43134 @code{set architecture} (@pxref{Targets, ,Specifying a Debugging Target}).
43136 A @samp{<compatible>} element is used to specify that the target
43137 is able to run binaries in some other than the main target architecture
43138 given by the @samp{<architecture>} element. For example, on the
43139 Cell Broadband Engine, the main architecture is @code{powerpc:common}
43140 or @code{powerpc:common64}, but the system is able to run binaries
43141 in the @code{spu} architecture as well. The way to describe this
43142 capability with @samp{<compatible>} is as follows:
43145 <architecture>powerpc:common</architecture>
43146 <compatible>spu</compatible>
43149 @subsection Features
43152 Each @samp{<feature>} describes some logical portion of the target
43153 system. Features are currently used to describe available CPU
43154 registers and the types of their contents. A @samp{<feature>} element
43158 <feature name="@var{name}">
43159 @r{[}@var{type}@dots{}@r{]}
43165 Each feature's name should be unique within the description. The name
43166 of a feature does not matter unless @value{GDBN} has some special
43167 knowledge of the contents of that feature; if it does, the feature
43168 should have its standard name. @xref{Standard Target Features}.
43172 Any register's value is a collection of bits which @value{GDBN} must
43173 interpret. The default interpretation is a two's complement integer,
43174 but other types can be requested by name in the register description.
43175 Some predefined types are provided by @value{GDBN} (@pxref{Predefined
43176 Target Types}), and the description can define additional composite
43179 Each type element must have an @samp{id} attribute, which gives
43180 a unique (within the containing @samp{<feature>}) name to the type.
43181 Types must be defined before they are used.
43184 Some targets offer vector registers, which can be treated as arrays
43185 of scalar elements. These types are written as @samp{<vector>} elements,
43186 specifying the array element type, @var{type}, and the number of elements,
43190 <vector id="@var{id}" type="@var{type}" count="@var{count}"/>
43194 If a register's value is usefully viewed in multiple ways, define it
43195 with a union type containing the useful representations. The
43196 @samp{<union>} element contains one or more @samp{<field>} elements,
43197 each of which has a @var{name} and a @var{type}:
43200 <union id="@var{id}">
43201 <field name="@var{name}" type="@var{type}"/>
43208 If a register's value is composed from several separate values, define
43209 it with either a structure type or a flags type.
43210 A flags type may only contain bitfields.
43211 A structure type may either contain only bitfields or contain no bitfields.
43212 If the value contains only bitfields, its total size in bytes must be
43215 Non-bitfield values have a @var{name} and @var{type}.
43218 <struct id="@var{id}">
43219 <field name="@var{name}" type="@var{type}"/>
43224 Both @var{name} and @var{type} values are required.
43225 No implicit padding is added.
43227 Bitfield values have a @var{name}, @var{start}, @var{end} and @var{type}.
43230 <struct id="@var{id}" size="@var{size}">
43231 <field name="@var{name}" start="@var{start}" end="@var{end}" type="@var{type}"/>
43237 <flags id="@var{id}" size="@var{size}">
43238 <field name="@var{name}" start="@var{start}" end="@var{end}" type="@var{type}"/>
43243 The @var{name} value is required.
43244 Bitfield values may be named with the empty string, @samp{""},
43245 in which case the field is ``filler'' and its value is not printed.
43246 Not all bits need to be specified, so ``filler'' fields are optional.
43248 The @var{start} and @var{end} values are required, and @var{type}
43250 The field's @var{start} must be less than or equal to its @var{end},
43251 and zero represents the least significant bit.
43253 The default value of @var{type} is @code{bool} for single bit fields,
43254 and an unsigned integer otherwise.
43256 Which to choose? Structures or flags?
43258 Registers defined with @samp{flags} have these advantages over
43259 defining them with @samp{struct}:
43263 Arithmetic may be performed on them as if they were integers.
43265 They are printed in a more readable fashion.
43268 Registers defined with @samp{struct} have one advantage over
43269 defining them with @samp{flags}:
43273 One can fetch individual fields like in @samp{C}.
43276 (gdb) print $my_struct_reg.field3
43282 @subsection Registers
43285 Each register is represented as an element with this form:
43288 <reg name="@var{name}"
43289 bitsize="@var{size}"
43290 @r{[}regnum="@var{num}"@r{]}
43291 @r{[}save-restore="@var{save-restore}"@r{]}
43292 @r{[}type="@var{type}"@r{]}
43293 @r{[}group="@var{group}"@r{]}/>
43297 The components are as follows:
43302 The register's name; it must be unique within the target description.
43305 The register's size, in bits.
43308 The register's number. If omitted, a register's number is one greater
43309 than that of the previous register (either in the current feature or in
43310 a preceding feature); the first register in the target description
43311 defaults to zero. This register number is used to read or write
43312 the register; e.g.@: it is used in the remote @code{p} and @code{P}
43313 packets, and registers appear in the @code{g} and @code{G} packets
43314 in order of increasing register number.
43317 Whether the register should be preserved across inferior function
43318 calls; this must be either @code{yes} or @code{no}. The default is
43319 @code{yes}, which is appropriate for most registers except for
43320 some system control registers; this is not related to the target's
43324 The type of the register. It may be a predefined type, a type
43325 defined in the current feature, or one of the special types @code{int}
43326 and @code{float}. @code{int} is an integer type of the correct size
43327 for @var{bitsize}, and @code{float} is a floating point type (in the
43328 architecture's normal floating point format) of the correct size for
43329 @var{bitsize}. The default is @code{int}.
43332 The register group to which this register belongs. It can be one of the
43333 standard register groups @code{general}, @code{float}, @code{vector} or an
43334 arbitrary string. Group names should be limited to alphanumeric characters.
43335 If a group name is made up of multiple words the words may be separated by
43336 hyphens; e.g.@: @code{special-group} or @code{ultra-special-group}. If no
43337 @var{group} is specified, @value{GDBN} will not display the register in
43338 @code{info registers}.
43342 @node Predefined Target Types
43343 @section Predefined Target Types
43344 @cindex target descriptions, predefined types
43346 Type definitions in the self-description can build up composite types
43347 from basic building blocks, but can not define fundamental types. Instead,
43348 standard identifiers are provided by @value{GDBN} for the fundamental
43349 types. The currently supported types are:
43354 Boolean type, occupying a single bit.
43362 Signed integer types holding the specified number of bits.
43370 Unsigned integer types holding the specified number of bits.
43374 Pointers to unspecified code and data. The program counter and
43375 any dedicated return address register may be marked as code
43376 pointers; printing a code pointer converts it into a symbolic
43377 address. The stack pointer and any dedicated address registers
43378 may be marked as data pointers.
43381 Single precision IEEE floating point.
43384 Double precision IEEE floating point.
43387 The 12-byte extended precision format used by ARM FPA registers.
43390 The 10-byte extended precision format used by x87 registers.
43393 32bit @sc{eflags} register used by x86.
43396 32bit @sc{mxcsr} register used by x86.
43400 @node Enum Target Types
43401 @section Enum Target Types
43402 @cindex target descriptions, enum types
43404 Enum target types are useful in @samp{struct} and @samp{flags}
43405 register descriptions. @xref{Target Description Format}.
43407 Enum types have a name, size and a list of name/value pairs.
43410 <enum id="@var{id}" size="@var{size}">
43411 <evalue name="@var{name}" value="@var{value}"/>
43416 Enums must be defined before they are used.
43419 <enum id="levels_type" size="4">
43420 <evalue name="low" value="0"/>
43421 <evalue name="high" value="1"/>
43423 <flags id="flags_type" size="4">
43424 <field name="X" start="0"/>
43425 <field name="LEVEL" start="1" end="1" type="levels_type"/>
43427 <reg name="flags" bitsize="32" type="flags_type"/>
43430 Given that description, a value of 3 for the @samp{flags} register
43431 would be printed as:
43434 (gdb) info register flags
43435 flags 0x3 [ X LEVEL=high ]
43438 @node Standard Target Features
43439 @section Standard Target Features
43440 @cindex target descriptions, standard features
43442 A target description must contain either no registers or all the
43443 target's registers. If the description contains no registers, then
43444 @value{GDBN} will assume a default register layout, selected based on
43445 the architecture. If the description contains any registers, the
43446 default layout will not be used; the standard registers must be
43447 described in the target description, in such a way that @value{GDBN}
43448 can recognize them.
43450 This is accomplished by giving specific names to feature elements
43451 which contain standard registers. @value{GDBN} will look for features
43452 with those names and verify that they contain the expected registers;
43453 if any known feature is missing required registers, or if any required
43454 feature is missing, @value{GDBN} will reject the target
43455 description. You can add additional registers to any of the
43456 standard features --- @value{GDBN} will display them just as if
43457 they were added to an unrecognized feature.
43459 This section lists the known features and their expected contents.
43460 Sample XML documents for these features are included in the
43461 @value{GDBN} source tree, in the directory @file{gdb/features}.
43463 Names recognized by @value{GDBN} should include the name of the
43464 company or organization which selected the name, and the overall
43465 architecture to which the feature applies; so e.g.@: the feature
43466 containing ARM core registers is named @samp{org.gnu.gdb.arm.core}.
43468 The names of registers are not case sensitive for the purpose
43469 of recognizing standard features, but @value{GDBN} will only display
43470 registers using the capitalization used in the description.
43473 * AArch64 Features::
43477 * MicroBlaze Features::
43481 * Nios II Features::
43482 * OpenRISC 1000 Features::
43483 * PowerPC Features::
43484 * RISC-V Features::
43485 * S/390 and System z Features::
43491 @node AArch64 Features
43492 @subsection AArch64 Features
43493 @cindex target descriptions, AArch64 features
43495 The @samp{org.gnu.gdb.aarch64.core} feature is required for AArch64
43496 targets. It should contain registers @samp{x0} through @samp{x30},
43497 @samp{sp}, @samp{pc}, and @samp{cpsr}.
43499 The @samp{org.gnu.gdb.aarch64.fpu} feature is optional. If present,
43500 it should contain registers @samp{v0} through @samp{v31}, @samp{fpsr},
43503 The @samp{org.gnu.gdb.aarch64.sve} feature is optional. If present,
43504 it should contain registers @samp{z0} through @samp{z31}, @samp{p0}
43505 through @samp{p15}, @samp{ffr} and @samp{vg}.
43507 The @samp{org.gnu.gdb.aarch64.pauth} feature is optional. If present,
43508 it should contain registers @samp{pauth_dmask} and @samp{pauth_cmask}.
43511 @subsection ARC Features
43512 @cindex target descriptions, ARC Features
43514 ARC processors are highly configurable, so even core registers and their number
43515 are not completely predetermined. In addition flags and PC registers which are
43516 important to @value{GDBN} are not ``core'' registers in ARC. It is required
43517 that one of the core registers features is present.
43518 @samp{org.gnu.gdb.arc.aux-minimal} feature is mandatory.
43520 The @samp{org.gnu.gdb.arc.core.v2} feature is required for ARC EM and ARC HS
43521 targets with a normal register file. It should contain registers @samp{r0}
43522 through @samp{r25}, @samp{gp}, @samp{fp}, @samp{sp}, @samp{r30}, @samp{blink},
43523 @samp{lp_count} and @samp{pcl}. This feature may contain register @samp{ilink}
43524 and any of extension core registers @samp{r32} through @samp{r59/acch}.
43525 @samp{ilink} and extension core registers are not available to read/write, when
43526 debugging GNU/Linux applications, thus @samp{ilink} is made optional.
43528 The @samp{org.gnu.gdb.arc.core-reduced.v2} feature is required for ARC EM and
43529 ARC HS targets with a reduced register file. It should contain registers
43530 @samp{r0} through @samp{r3}, @samp{r10} through @samp{r15}, @samp{gp},
43531 @samp{fp}, @samp{sp}, @samp{r30}, @samp{blink}, @samp{lp_count} and @samp{pcl}.
43532 This feature may contain register @samp{ilink} and any of extension core
43533 registers @samp{r32} through @samp{r59/acch}.
43535 The @samp{org.gnu.gdb.arc.core.arcompact} feature is required for ARCompact
43536 targets with a normal register file. It should contain registers @samp{r0}
43537 through @samp{r25}, @samp{gp}, @samp{fp}, @samp{sp}, @samp{r30}, @samp{blink},
43538 @samp{lp_count} and @samp{pcl}. This feature may contain registers
43539 @samp{ilink1}, @samp{ilink2} and any of extension core registers @samp{r32}
43540 through @samp{r59/acch}. @samp{ilink1} and @samp{ilink2} and extension core
43541 registers are not available when debugging GNU/Linux applications. The only
43542 difference with @samp{org.gnu.gdb.arc.core.v2} feature is in the names of
43543 @samp{ilink1} and @samp{ilink2} registers and that @samp{r30} is mandatory in
43544 ARC v2, but @samp{ilink2} is optional on ARCompact.
43546 The @samp{org.gnu.gdb.arc.aux-minimal} feature is required for all ARC
43547 targets. It should contain registers @samp{pc} and @samp{status32}.
43550 @subsection ARM Features
43551 @cindex target descriptions, ARM features
43553 The @samp{org.gnu.gdb.arm.core} feature is required for non-M-profile
43555 It should contain registers @samp{r0} through @samp{r13}, @samp{sp},
43556 @samp{lr}, @samp{pc}, and @samp{cpsr}.
43558 For M-profile targets (e.g. Cortex-M3), the @samp{org.gnu.gdb.arm.core}
43559 feature is replaced by @samp{org.gnu.gdb.arm.m-profile}. It should contain
43560 registers @samp{r0} through @samp{r13}, @samp{sp}, @samp{lr}, @samp{pc},
43563 The @samp{org.gnu.gdb.arm.fpa} feature is optional. If present, it
43564 should contain registers @samp{f0} through @samp{f7} and @samp{fps}.
43566 The @samp{org.gnu.gdb.xscale.iwmmxt} feature is optional. If present,
43567 it should contain at least registers @samp{wR0} through @samp{wR15} and
43568 @samp{wCGR0} through @samp{wCGR3}. The @samp{wCID}, @samp{wCon},
43569 @samp{wCSSF}, and @samp{wCASF} registers are optional.
43571 The @samp{org.gnu.gdb.arm.vfp} feature is optional. If present, it
43572 should contain at least registers @samp{d0} through @samp{d15}. If
43573 they are present, @samp{d16} through @samp{d31} should also be included.
43574 @value{GDBN} will synthesize the single-precision registers from
43575 halves of the double-precision registers.
43577 The @samp{org.gnu.gdb.arm.neon} feature is optional. It does not
43578 need to contain registers; it instructs @value{GDBN} to display the
43579 VFP double-precision registers as vectors and to synthesize the
43580 quad-precision registers from pairs of double-precision registers.
43581 If this feature is present, @samp{org.gnu.gdb.arm.vfp} must also
43582 be present and include 32 double-precision registers.
43584 @node i386 Features
43585 @subsection i386 Features
43586 @cindex target descriptions, i386 features
43588 The @samp{org.gnu.gdb.i386.core} feature is required for i386/amd64
43589 targets. It should describe the following registers:
43593 @samp{eax} through @samp{edi} plus @samp{eip} for i386
43595 @samp{rax} through @samp{r15} plus @samp{rip} for amd64
43597 @samp{eflags}, @samp{cs}, @samp{ss}, @samp{ds}, @samp{es},
43598 @samp{fs}, @samp{gs}
43600 @samp{st0} through @samp{st7}
43602 @samp{fctrl}, @samp{fstat}, @samp{ftag}, @samp{fiseg}, @samp{fioff},
43603 @samp{foseg}, @samp{fooff} and @samp{fop}
43606 The register sets may be different, depending on the target.
43608 The @samp{org.gnu.gdb.i386.sse} feature is optional. It should
43609 describe registers:
43613 @samp{xmm0} through @samp{xmm7} for i386
43615 @samp{xmm0} through @samp{xmm15} for amd64
43620 The @samp{org.gnu.gdb.i386.avx} feature is optional and requires the
43621 @samp{org.gnu.gdb.i386.sse} feature. It should
43622 describe the upper 128 bits of @sc{ymm} registers:
43626 @samp{ymm0h} through @samp{ymm7h} for i386
43628 @samp{ymm0h} through @samp{ymm15h} for amd64
43631 The @samp{org.gnu.gdb.i386.mpx} is an optional feature representing Intel
43632 Memory Protection Extension (MPX). It should describe the following registers:
43636 @samp{bnd0raw} through @samp{bnd3raw} for i386 and amd64.
43638 @samp{bndcfgu} and @samp{bndstatus} for i386 and amd64.
43641 The @samp{org.gnu.gdb.i386.linux} feature is optional. It should
43642 describe a single register, @samp{orig_eax}.
43644 The @samp{org.gnu.gdb.i386.segments} feature is optional. It should
43645 describe two system registers: @samp{fs_base} and @samp{gs_base}.
43647 The @samp{org.gnu.gdb.i386.avx512} feature is optional and requires the
43648 @samp{org.gnu.gdb.i386.avx} feature. It should
43649 describe additional @sc{xmm} registers:
43653 @samp{xmm16h} through @samp{xmm31h}, only valid for amd64.
43656 It should describe the upper 128 bits of additional @sc{ymm} registers:
43660 @samp{ymm16h} through @samp{ymm31h}, only valid for amd64.
43664 describe the upper 256 bits of @sc{zmm} registers:
43668 @samp{zmm0h} through @samp{zmm7h} for i386.
43670 @samp{zmm0h} through @samp{zmm15h} for amd64.
43674 describe the additional @sc{zmm} registers:
43678 @samp{zmm16h} through @samp{zmm31h}, only valid for amd64.
43681 The @samp{org.gnu.gdb.i386.pkeys} feature is optional. It should
43682 describe a single register, @samp{pkru}. It is a 32-bit register
43683 valid for i386 and amd64.
43685 @node MicroBlaze Features
43686 @subsection MicroBlaze Features
43687 @cindex target descriptions, MicroBlaze features
43689 The @samp{org.gnu.gdb.microblaze.core} feature is required for MicroBlaze
43690 targets. It should contain registers @samp{r0} through @samp{r31},
43691 @samp{rpc}, @samp{rmsr}, @samp{rear}, @samp{resr}, @samp{rfsr}, @samp{rbtr},
43692 @samp{rpvr}, @samp{rpvr1} through @samp{rpvr11}, @samp{redr}, @samp{rpid},
43693 @samp{rzpr}, @samp{rtlbx}, @samp{rtlbsx}, @samp{rtlblo}, and @samp{rtlbhi}.
43695 The @samp{org.gnu.gdb.microblaze.stack-protect} feature is optional.
43696 If present, it should contain registers @samp{rshr} and @samp{rslr}
43698 @node MIPS Features
43699 @subsection @acronym{MIPS} Features
43700 @cindex target descriptions, @acronym{MIPS} features
43702 The @samp{org.gnu.gdb.mips.cpu} feature is required for @acronym{MIPS} targets.
43703 It should contain registers @samp{r0} through @samp{r31}, @samp{lo},
43704 @samp{hi}, and @samp{pc}. They may be 32-bit or 64-bit depending
43707 The @samp{org.gnu.gdb.mips.cp0} feature is also required. It should
43708 contain at least the @samp{status}, @samp{badvaddr}, and @samp{cause}
43709 registers. They may be 32-bit or 64-bit depending on the target.
43711 The @samp{org.gnu.gdb.mips.fpu} feature is currently required, though
43712 it may be optional in a future version of @value{GDBN}. It should
43713 contain registers @samp{f0} through @samp{f31}, @samp{fcsr}, and
43714 @samp{fir}. They may be 32-bit or 64-bit depending on the target.
43716 The @samp{org.gnu.gdb.mips.dsp} feature is optional. It should
43717 contain registers @samp{hi1} through @samp{hi3}, @samp{lo1} through
43718 @samp{lo3}, and @samp{dspctl}. The @samp{dspctl} register should
43719 be 32-bit and the rest may be 32-bit or 64-bit depending on the target.
43721 The @samp{org.gnu.gdb.mips.linux} feature is optional. It should
43722 contain a single register, @samp{restart}, which is used by the
43723 Linux kernel to control restartable syscalls.
43725 @node M68K Features
43726 @subsection M68K Features
43727 @cindex target descriptions, M68K features
43730 @item @samp{org.gnu.gdb.m68k.core}
43731 @itemx @samp{org.gnu.gdb.coldfire.core}
43732 @itemx @samp{org.gnu.gdb.fido.core}
43733 One of those features must be always present.
43734 The feature that is present determines which flavor of m68k is
43735 used. The feature that is present should contain registers
43736 @samp{d0} through @samp{d7}, @samp{a0} through @samp{a5}, @samp{fp},
43737 @samp{sp}, @samp{ps} and @samp{pc}.
43739 @item @samp{org.gnu.gdb.coldfire.fp}
43740 This feature is optional. If present, it should contain registers
43741 @samp{fp0} through @samp{fp7}, @samp{fpcontrol}, @samp{fpstatus} and
43745 @node NDS32 Features
43746 @subsection NDS32 Features
43747 @cindex target descriptions, NDS32 features
43749 The @samp{org.gnu.gdb.nds32.core} feature is required for NDS32
43750 targets. It should contain at least registers @samp{r0} through
43751 @samp{r10}, @samp{r15}, @samp{fp}, @samp{gp}, @samp{lp}, @samp{sp},
43754 The @samp{org.gnu.gdb.nds32.fpu} feature is optional. If present,
43755 it should contain 64-bit double-precision floating-point registers
43756 @samp{fd0} through @emph{fdN}, which should be @samp{fd3}, @samp{fd7},
43757 @samp{fd15}, or @samp{fd31} based on the FPU configuration implemented.
43759 @emph{Note:} The first sixteen 64-bit double-precision floating-point
43760 registers are overlapped with the thirty-two 32-bit single-precision
43761 floating-point registers. The 32-bit single-precision registers, if
43762 not being listed explicitly, will be synthesized from halves of the
43763 overlapping 64-bit double-precision registers. Listing 32-bit
43764 single-precision registers explicitly is deprecated, and the
43765 support to it could be totally removed some day.
43767 @node Nios II Features
43768 @subsection Nios II Features
43769 @cindex target descriptions, Nios II features
43771 The @samp{org.gnu.gdb.nios2.cpu} feature is required for Nios II
43772 targets. It should contain the 32 core registers (@samp{zero},
43773 @samp{at}, @samp{r2} through @samp{r23}, @samp{et} through @samp{ra}),
43774 @samp{pc}, and the 16 control registers (@samp{status} through
43777 @node OpenRISC 1000 Features
43778 @subsection Openrisc 1000 Features
43779 @cindex target descriptions, OpenRISC 1000 features
43781 The @samp{org.gnu.gdb.or1k.group0} feature is required for OpenRISC 1000
43782 targets. It should contain the 32 general purpose registers (@samp{r0}
43783 through @samp{r31}), @samp{ppc}, @samp{npc} and @samp{sr}.
43785 @node PowerPC Features
43786 @subsection PowerPC Features
43787 @cindex target descriptions, PowerPC features
43789 The @samp{org.gnu.gdb.power.core} feature is required for PowerPC
43790 targets. It should contain registers @samp{r0} through @samp{r31},
43791 @samp{pc}, @samp{msr}, @samp{cr}, @samp{lr}, @samp{ctr}, and
43792 @samp{xer}. They may be 32-bit or 64-bit depending on the target.
43794 The @samp{org.gnu.gdb.power.fpu} feature is optional. It should
43795 contain registers @samp{f0} through @samp{f31} and @samp{fpscr}.
43797 The @samp{org.gnu.gdb.power.altivec} feature is optional. It should
43798 contain registers @samp{vr0} through @samp{vr31}, @samp{vscr}, and
43799 @samp{vrsave}. @value{GDBN} will define pseudo-registers @samp{v0}
43800 through @samp{v31} as aliases for the corresponding @samp{vrX}
43803 The @samp{org.gnu.gdb.power.vsx} feature is optional. It should
43804 contain registers @samp{vs0h} through @samp{vs31h}. @value{GDBN} will
43805 combine these registers with the floating point registers (@samp{f0}
43806 through @samp{f31}) and the altivec registers (@samp{vr0} through
43807 @samp{vr31}) to present the 128-bit wide registers @samp{vs0} through
43808 @samp{vs63}, the set of vector-scalar registers for POWER7.
43809 Therefore, this feature requires both @samp{org.gnu.gdb.power.fpu} and
43810 @samp{org.gnu.gdb.power.altivec}.
43812 The @samp{org.gnu.gdb.power.spe} feature is optional. It should
43813 contain registers @samp{ev0h} through @samp{ev31h}, @samp{acc}, and
43814 @samp{spefscr}. SPE targets should provide 32-bit registers in
43815 @samp{org.gnu.gdb.power.core} and provide the upper halves in
43816 @samp{ev0h} through @samp{ev31h}. @value{GDBN} will combine
43817 these to present registers @samp{ev0} through @samp{ev31} to the
43820 The @samp{org.gnu.gdb.power.ppr} feature is optional. It should
43821 contain the 64-bit register @samp{ppr}.
43823 The @samp{org.gnu.gdb.power.dscr} feature is optional. It should
43824 contain the 64-bit register @samp{dscr}.
43826 The @samp{org.gnu.gdb.power.tar} feature is optional. It should
43827 contain the 64-bit register @samp{tar}.
43829 The @samp{org.gnu.gdb.power.ebb} feature is optional. It should
43830 contain registers @samp{bescr}, @samp{ebbhr} and @samp{ebbrr}, all
43833 The @samp{org.gnu.gdb.power.linux.pmu} feature is optional. It should
43834 contain registers @samp{mmcr0}, @samp{mmcr2}, @samp{siar}, @samp{sdar}
43835 and @samp{sier}, all 64-bit wide. This is the subset of the isa 2.07
43836 server PMU registers provided by @sc{gnu}/Linux.
43838 The @samp{org.gnu.gdb.power.htm.spr} feature is optional. It should
43839 contain registers @samp{tfhar}, @samp{texasr} and @samp{tfiar}, all
43842 The @samp{org.gnu.gdb.power.htm.core} feature is optional. It should
43843 contain the checkpointed general-purpose registers @samp{cr0} through
43844 @samp{cr31}, as well as the checkpointed registers @samp{clr} and
43845 @samp{cctr}. These registers may all be either 32-bit or 64-bit
43846 depending on the target. It should also contain the checkpointed
43847 registers @samp{ccr} and @samp{cxer}, which should both be 32-bit
43850 The @samp{org.gnu.gdb.power.htm.fpu} feature is optional. It should
43851 contain the checkpointed 64-bit floating-point registers @samp{cf0}
43852 through @samp{cf31}, as well as the checkpointed 64-bit register
43855 The @samp{org.gnu.gdb.power.htm.altivec} feature is optional. It
43856 should contain the checkpointed altivec registers @samp{cvr0} through
43857 @samp{cvr31}, all 128-bit wide. It should also contain the
43858 checkpointed registers @samp{cvscr} and @samp{cvrsave}, both 32-bit
43861 The @samp{org.gnu.gdb.power.htm.vsx} feature is optional. It should
43862 contain registers @samp{cvs0h} through @samp{cvs31h}. @value{GDBN}
43863 will combine these registers with the checkpointed floating point
43864 registers (@samp{cf0} through @samp{cf31}) and the checkpointed
43865 altivec registers (@samp{cvr0} through @samp{cvr31}) to present the
43866 128-bit wide checkpointed vector-scalar registers @samp{cvs0} through
43867 @samp{cvs63}. Therefore, this feature requires both
43868 @samp{org.gnu.gdb.power.htm.altivec} and
43869 @samp{org.gnu.gdb.power.htm.fpu}.
43871 The @samp{org.gnu.gdb.power.htm.ppr} feature is optional. It should
43872 contain the 64-bit checkpointed register @samp{cppr}.
43874 The @samp{org.gnu.gdb.power.htm.dscr} feature is optional. It should
43875 contain the 64-bit checkpointed register @samp{cdscr}.
43877 The @samp{org.gnu.gdb.power.htm.tar} feature is optional. It should
43878 contain the 64-bit checkpointed register @samp{ctar}.
43881 @node RISC-V Features
43882 @subsection RISC-V Features
43883 @cindex target descriptions, RISC-V Features
43885 The @samp{org.gnu.gdb.riscv.cpu} feature is required for RISC-V
43886 targets. It should contain the registers @samp{x0} through
43887 @samp{x31}, and @samp{pc}. Either the architectural names (@samp{x0},
43888 @samp{x1}, etc) can be used, or the ABI names (@samp{zero}, @samp{ra},
43891 The @samp{org.gnu.gdb.riscv.fpu} feature is optional. If present, it
43892 should contain registers @samp{f0} through @samp{f31}, @samp{fflags},
43893 @samp{frm}, and @samp{fcsr}. As with the cpu feature, either the
43894 architectural register names, or the ABI names can be used.
43896 The @samp{org.gnu.gdb.riscv.virtual} feature is optional. If present,
43897 it should contain registers that are not backed by real registers on
43898 the target, but are instead virtual, where the register value is
43899 derived from other target state. In many ways these are like
43900 @value{GDBN}s pseudo-registers, except implemented by the target.
43901 Currently the only register expected in this set is the one byte
43902 @samp{priv} register that contains the target's privilege level in the
43903 least significant two bits.
43905 The @samp{org.gnu.gdb.riscv.csr} feature is optional. If present, it
43906 should contain all of the target's standard CSRs. Standard CSRs are
43907 those defined in the RISC-V specification documents. There is some
43908 overlap between this feature and the fpu feature; the @samp{fflags},
43909 @samp{frm}, and @samp{fcsr} registers could be in either feature. The
43910 expectation is that these registers will be in the fpu feature if the
43911 target has floating point hardware, but can be moved into the csr
43912 feature if the target has the floating point control registers, but no
43913 other floating point hardware.
43915 @node S/390 and System z Features
43916 @subsection S/390 and System z Features
43917 @cindex target descriptions, S/390 features
43918 @cindex target descriptions, System z features
43920 The @samp{org.gnu.gdb.s390.core} feature is required for S/390 and
43921 System z targets. It should contain the PSW and the 16 general
43922 registers. In particular, System z targets should provide the 64-bit
43923 registers @samp{pswm}, @samp{pswa}, and @samp{r0} through @samp{r15}.
43924 S/390 targets should provide the 32-bit versions of these registers.
43925 A System z target that runs in 31-bit addressing mode should provide
43926 32-bit versions of @samp{pswm} and @samp{pswa}, as well as the general
43927 register's upper halves @samp{r0h} through @samp{r15h}, and their
43928 lower halves @samp{r0l} through @samp{r15l}.
43930 The @samp{org.gnu.gdb.s390.fpr} feature is required. It should
43931 contain the 64-bit registers @samp{f0} through @samp{f15}, and
43934 The @samp{org.gnu.gdb.s390.acr} feature is required. It should
43935 contain the 32-bit registers @samp{acr0} through @samp{acr15}.
43937 The @samp{org.gnu.gdb.s390.linux} feature is optional. It should
43938 contain the register @samp{orig_r2}, which is 64-bit wide on System z
43939 targets and 32-bit otherwise. In addition, the feature may contain
43940 the @samp{last_break} register, whose width depends on the addressing
43941 mode, as well as the @samp{system_call} register, which is always
43944 The @samp{org.gnu.gdb.s390.tdb} feature is optional. It should
43945 contain the 64-bit registers @samp{tdb0}, @samp{tac}, @samp{tct},
43946 @samp{atia}, and @samp{tr0} through @samp{tr15}.
43948 The @samp{org.gnu.gdb.s390.vx} feature is optional. It should contain
43949 64-bit wide registers @samp{v0l} through @samp{v15l}, which will be
43950 combined by @value{GDBN} with the floating point registers @samp{f0}
43951 through @samp{f15} to present the 128-bit wide vector registers
43952 @samp{v0} through @samp{v15}. In addition, this feature should
43953 contain the 128-bit wide vector registers @samp{v16} through
43956 The @samp{org.gnu.gdb.s390.gs} feature is optional. It should contain
43957 the 64-bit wide guarded-storage-control registers @samp{gsd},
43958 @samp{gssm}, and @samp{gsepla}.
43960 The @samp{org.gnu.gdb.s390.gsbc} feature is optional. It should contain
43961 the 64-bit wide guarded-storage broadcast control registers
43962 @samp{bc_gsd}, @samp{bc_gssm}, and @samp{bc_gsepla}.
43964 @node Sparc Features
43965 @subsection Sparc Features
43966 @cindex target descriptions, sparc32 features
43967 @cindex target descriptions, sparc64 features
43968 The @samp{org.gnu.gdb.sparc.cpu} feature is required for sparc32/sparc64
43969 targets. It should describe the following registers:
43973 @samp{g0} through @samp{g7}
43975 @samp{o0} through @samp{o7}
43977 @samp{l0} through @samp{l7}
43979 @samp{i0} through @samp{i7}
43982 They may be 32-bit or 64-bit depending on the target.
43984 Also the @samp{org.gnu.gdb.sparc.fpu} feature is required for sparc32/sparc64
43985 targets. It should describe the following registers:
43989 @samp{f0} through @samp{f31}
43991 @samp{f32} through @samp{f62} for sparc64
43994 The @samp{org.gnu.gdb.sparc.cp0} feature is required for sparc32/sparc64
43995 targets. It should describe the following registers:
43999 @samp{y}, @samp{psr}, @samp{wim}, @samp{tbr}, @samp{pc}, @samp{npc},
44000 @samp{fsr}, and @samp{csr} for sparc32
44002 @samp{pc}, @samp{npc}, @samp{state}, @samp{fsr}, @samp{fprs}, and @samp{y}
44006 @node TIC6x Features
44007 @subsection TMS320C6x Features
44008 @cindex target descriptions, TIC6x features
44009 @cindex target descriptions, TMS320C6x features
44010 The @samp{org.gnu.gdb.tic6x.core} feature is required for TMS320C6x
44011 targets. It should contain registers @samp{A0} through @samp{A15},
44012 registers @samp{B0} through @samp{B15}, @samp{CSR} and @samp{PC}.
44014 The @samp{org.gnu.gdb.tic6x.gp} feature is optional. It should
44015 contain registers @samp{A16} through @samp{A31} and @samp{B16}
44016 through @samp{B31}.
44018 The @samp{org.gnu.gdb.tic6x.c6xp} feature is optional. It should
44019 contain registers @samp{TSR}, @samp{ILC} and @samp{RILC}.
44021 @node Operating System Information
44022 @appendix Operating System Information
44023 @cindex operating system information
44029 Users of @value{GDBN} often wish to obtain information about the state of
44030 the operating system running on the target---for example the list of
44031 processes, or the list of open files. This section describes the
44032 mechanism that makes it possible. This mechanism is similar to the
44033 target features mechanism (@pxref{Target Descriptions}), but focuses
44034 on a different aspect of target.
44036 Operating system information is retrived from the target via the
44037 remote protocol, using @samp{qXfer} requests (@pxref{qXfer osdata
44038 read}). The object name in the request should be @samp{osdata}, and
44039 the @var{annex} identifies the data to be fetched.
44042 @appendixsection Process list
44043 @cindex operating system information, process list
44045 When requesting the process list, the @var{annex} field in the
44046 @samp{qXfer} request should be @samp{processes}. The returned data is
44047 an XML document. The formal syntax of this document is defined in
44048 @file{gdb/features/osdata.dtd}.
44050 An example document is:
44053 <?xml version="1.0"?>
44054 <!DOCTYPE target SYSTEM "osdata.dtd">
44055 <osdata type="processes">
44057 <column name="pid">1</column>
44058 <column name="user">root</column>
44059 <column name="command">/sbin/init</column>
44060 <column name="cores">1,2,3</column>
44065 Each item should include a column whose name is @samp{pid}. The value
44066 of that column should identify the process on the target. The
44067 @samp{user} and @samp{command} columns are optional, and will be
44068 displayed by @value{GDBN}. The @samp{cores} column, if present,
44069 should contain a comma-separated list of cores that this process
44070 is running on. Target may provide additional columns,
44071 which @value{GDBN} currently ignores.
44073 @node Trace File Format
44074 @appendix Trace File Format
44075 @cindex trace file format
44077 The trace file comes in three parts: a header, a textual description
44078 section, and a trace frame section with binary data.
44080 The header has the form @code{\x7fTRACE0\n}. The first byte is
44081 @code{0x7f} so as to indicate that the file contains binary data,
44082 while the @code{0} is a version number that may have different values
44085 The description section consists of multiple lines of @sc{ascii} text
44086 separated by newline characters (@code{0xa}). The lines may include a
44087 variety of optional descriptive or context-setting information, such
44088 as tracepoint definitions or register set size. @value{GDBN} will
44089 ignore any line that it does not recognize. An empty line marks the end
44094 Specifies the size of a register block in bytes. This is equal to the
44095 size of a @code{g} packet payload in the remote protocol. @var{size}
44096 is an ascii decimal number. There should be only one such line in
44097 a single trace file.
44099 @item status @var{status}
44100 Trace status. @var{status} has the same format as a @code{qTStatus}
44101 remote packet reply. There should be only one such line in a single trace
44104 @item tp @var{payload}
44105 Tracepoint definition. The @var{payload} has the same format as
44106 @code{qTfP}/@code{qTsP} remote packet reply payload. A single tracepoint
44107 may take multiple lines of definition, corresponding to the multiple
44110 @item tsv @var{payload}
44111 Trace state variable definition. The @var{payload} has the same format as
44112 @code{qTfV}/@code{qTsV} remote packet reply payload. A single variable
44113 may take multiple lines of definition, corresponding to the multiple
44116 @item tdesc @var{payload}
44117 Target description in XML format. The @var{payload} is a single line of
44118 the XML file. All such lines should be concatenated together to get
44119 the original XML file. This file is in the same format as @code{qXfer}
44120 @code{features} payload, and corresponds to the main @code{target.xml}
44121 file. Includes are not allowed.
44125 The trace frame section consists of a number of consecutive frames.
44126 Each frame begins with a two-byte tracepoint number, followed by a
44127 four-byte size giving the amount of data in the frame. The data in
44128 the frame consists of a number of blocks, each introduced by a
44129 character indicating its type (at least register, memory, and trace
44130 state variable). The data in this section is raw binary, not a
44131 hexadecimal or other encoding; its endianness matches the target's
44134 @c FIXME bi-arch may require endianness/arch info in description section
44137 @item R @var{bytes}
44138 Register block. The number and ordering of bytes matches that of a
44139 @code{g} packet in the remote protocol. Note that these are the
44140 actual bytes, in target order, not a hexadecimal encoding.
44142 @item M @var{address} @var{length} @var{bytes}...
44143 Memory block. This is a contiguous block of memory, at the 8-byte
44144 address @var{address}, with a 2-byte length @var{length}, followed by
44145 @var{length} bytes.
44147 @item V @var{number} @var{value}
44148 Trace state variable block. This records the 8-byte signed value
44149 @var{value} of trace state variable numbered @var{number}.
44153 Future enhancements of the trace file format may include additional types
44156 @node Index Section Format
44157 @appendix @code{.gdb_index} section format
44158 @cindex .gdb_index section format
44159 @cindex index section format
44161 This section documents the index section that is created by @code{save
44162 gdb-index} (@pxref{Index Files}). The index section is
44163 DWARF-specific; some knowledge of DWARF is assumed in this
44166 The mapped index file format is designed to be directly
44167 @code{mmap}able on any architecture. In most cases, a datum is
44168 represented using a little-endian 32-bit integer value, called an
44169 @code{offset_type}. Big endian machines must byte-swap the values
44170 before using them. Exceptions to this rule are noted. The data is
44171 laid out such that alignment is always respected.
44173 A mapped index consists of several areas, laid out in order.
44177 The file header. This is a sequence of values, of @code{offset_type}
44178 unless otherwise noted:
44182 The version number, currently 8. Versions 1, 2 and 3 are obsolete.
44183 Version 4 uses a different hashing function from versions 5 and 6.
44184 Version 6 includes symbols for inlined functions, whereas versions 4
44185 and 5 do not. Version 7 adds attributes to the CU indices in the
44186 symbol table. Version 8 specifies that symbols from DWARF type units
44187 (@samp{DW_TAG_type_unit}) refer to the type unit's symbol table and not the
44188 compilation unit (@samp{DW_TAG_comp_unit}) using the type.
44190 @value{GDBN} will only read version 4, 5, or 6 indices
44191 by specifying @code{set use-deprecated-index-sections on}.
44192 GDB has a workaround for potentially broken version 7 indices so it is
44193 currently not flagged as deprecated.
44196 The offset, from the start of the file, of the CU list.
44199 The offset, from the start of the file, of the types CU list. Note
44200 that this area can be empty, in which case this offset will be equal
44201 to the next offset.
44204 The offset, from the start of the file, of the address area.
44207 The offset, from the start of the file, of the symbol table.
44210 The offset, from the start of the file, of the constant pool.
44214 The CU list. This is a sequence of pairs of 64-bit little-endian
44215 values, sorted by the CU offset. The first element in each pair is
44216 the offset of a CU in the @code{.debug_info} section. The second
44217 element in each pair is the length of that CU. References to a CU
44218 elsewhere in the map are done using a CU index, which is just the
44219 0-based index into this table. Note that if there are type CUs, then
44220 conceptually CUs and type CUs form a single list for the purposes of
44224 The types CU list. This is a sequence of triplets of 64-bit
44225 little-endian values. In a triplet, the first value is the CU offset,
44226 the second value is the type offset in the CU, and the third value is
44227 the type signature. The types CU list is not sorted.
44230 The address area. The address area consists of a sequence of address
44231 entries. Each address entry has three elements:
44235 The low address. This is a 64-bit little-endian value.
44238 The high address. This is a 64-bit little-endian value. Like
44239 @code{DW_AT_high_pc}, the value is one byte beyond the end.
44242 The CU index. This is an @code{offset_type} value.
44246 The symbol table. This is an open-addressed hash table. The size of
44247 the hash table is always a power of 2.
44249 Each slot in the hash table consists of a pair of @code{offset_type}
44250 values. The first value is the offset of the symbol's name in the
44251 constant pool. The second value is the offset of the CU vector in the
44254 If both values are 0, then this slot in the hash table is empty. This
44255 is ok because while 0 is a valid constant pool index, it cannot be a
44256 valid index for both a string and a CU vector.
44258 The hash value for a table entry is computed by applying an
44259 iterative hash function to the symbol's name. Starting with an
44260 initial value of @code{r = 0}, each (unsigned) character @samp{c} in
44261 the string is incorporated into the hash using the formula depending on the
44266 The formula is @code{r = r * 67 + c - 113}.
44268 @item Versions 5 to 7
44269 The formula is @code{r = r * 67 + tolower (c) - 113}.
44272 The terminating @samp{\0} is not incorporated into the hash.
44274 The step size used in the hash table is computed via
44275 @code{((hash * 17) & (size - 1)) | 1}, where @samp{hash} is the hash
44276 value, and @samp{size} is the size of the hash table. The step size
44277 is used to find the next candidate slot when handling a hash
44280 The names of C@t{++} symbols in the hash table are canonicalized. We
44281 don't currently have a simple description of the canonicalization
44282 algorithm; if you intend to create new index sections, you must read
44286 The constant pool. This is simply a bunch of bytes. It is organized
44287 so that alignment is correct: CU vectors are stored first, followed by
44290 A CU vector in the constant pool is a sequence of @code{offset_type}
44291 values. The first value is the number of CU indices in the vector.
44292 Each subsequent value is the index and symbol attributes of a CU in
44293 the CU list. This element in the hash table is used to indicate which
44294 CUs define the symbol and how the symbol is used.
44295 See below for the format of each CU index+attributes entry.
44297 A string in the constant pool is zero-terminated.
44300 Attributes were added to CU index values in @code{.gdb_index} version 7.
44301 If a symbol has multiple uses within a CU then there is one
44302 CU index+attributes value for each use.
44304 The format of each CU index+attributes entry is as follows
44310 This is the index of the CU in the CU list.
44312 These bits are reserved for future purposes and must be zero.
44314 The kind of the symbol in the CU.
44318 This value is reserved and should not be used.
44319 By reserving zero the full @code{offset_type} value is backwards compatible
44320 with previous versions of the index.
44322 The symbol is a type.
44324 The symbol is a variable or an enum value.
44326 The symbol is a function.
44328 Any other kind of symbol.
44330 These values are reserved.
44334 This bit is zero if the value is global and one if it is static.
44336 The determination of whether a symbol is global or static is complicated.
44337 The authorative reference is the file @file{dwarf2read.c} in
44338 @value{GDBN} sources.
44342 This pseudo-code describes the computation of a symbol's kind and
44343 global/static attributes in the index.
44346 is_external = get_attribute (die, DW_AT_external);
44347 language = get_attribute (cu_die, DW_AT_language);
44350 case DW_TAG_typedef:
44351 case DW_TAG_base_type:
44352 case DW_TAG_subrange_type:
44356 case DW_TAG_enumerator:
44358 is_static = language != CPLUS;
44360 case DW_TAG_subprogram:
44362 is_static = ! (is_external || language == ADA);
44364 case DW_TAG_constant:
44366 is_static = ! is_external;
44368 case DW_TAG_variable:
44370 is_static = ! is_external;
44372 case DW_TAG_namespace:
44376 case DW_TAG_class_type:
44377 case DW_TAG_interface_type:
44378 case DW_TAG_structure_type:
44379 case DW_TAG_union_type:
44380 case DW_TAG_enumeration_type:
44382 is_static = language != CPLUS;
44390 @appendix Manual pages
44394 * gdb man:: The GNU Debugger man page
44395 * gdbserver man:: Remote Server for the GNU Debugger man page
44396 * gcore man:: Generate a core file of a running program
44397 * gdbinit man:: gdbinit scripts
44398 * gdb-add-index man:: Add index files to speed up GDB
44404 @c man title gdb The GNU Debugger
44406 @c man begin SYNOPSIS gdb
44407 gdb [@option{-help}] [@option{-nh}] [@option{-nx}] [@option{-q}]
44408 [@option{-batch}] [@option{-cd=}@var{dir}] [@option{-f}]
44409 [@option{-b}@w{ }@var{bps}]
44410 [@option{-tty=}@var{dev}] [@option{-s} @var{symfile}]
44411 [@option{-e}@w{ }@var{prog}] [@option{-se}@w{ }@var{prog}]
44412 [@option{-c}@w{ }@var{core}] [@option{-p}@w{ }@var{procID}]
44413 [@option{-x}@w{ }@var{cmds}] [@option{-d}@w{ }@var{dir}]
44414 [@var{prog}|@var{prog} @var{procID}|@var{prog} @var{core}]
44417 @c man begin DESCRIPTION gdb
44418 The purpose of a debugger such as @value{GDBN} is to allow you to see what is
44419 going on ``inside'' another program while it executes -- or what another
44420 program was doing at the moment it crashed.
44422 @value{GDBN} can do four main kinds of things (plus other things in support of
44423 these) to help you catch bugs in the act:
44427 Start your program, specifying anything that might affect its behavior.
44430 Make your program stop on specified conditions.
44433 Examine what has happened, when your program has stopped.
44436 Change things in your program, so you can experiment with correcting the
44437 effects of one bug and go on to learn about another.
44440 You can use @value{GDBN} to debug programs written in C, C@t{++}, Fortran and
44443 @value{GDBN} is invoked with the shell command @code{gdb}. Once started, it reads
44444 commands from the terminal until you tell it to exit with the @value{GDBN}
44445 command @code{quit}. You can get online help from @value{GDBN} itself
44446 by using the command @code{help}.
44448 You can run @code{gdb} with no arguments or options; but the most
44449 usual way to start @value{GDBN} is with one argument or two, specifying an
44450 executable program as the argument:
44456 You can also start with both an executable program and a core file specified:
44462 You can, instead, specify a process ID as a second argument, if you want
44463 to debug a running process:
44471 would attach @value{GDBN} to process @code{1234} (unless you also have a file
44472 named @file{1234}; @value{GDBN} does check for a core file first).
44473 With option @option{-p} you can omit the @var{program} filename.
44475 Here are some of the most frequently needed @value{GDBN} commands:
44477 @c pod2man highlights the right hand side of the @item lines.
44479 @item break [@var{file}:]@var{function}
44480 Set a breakpoint at @var{function} (in @var{file}).
44482 @item run [@var{arglist}]
44483 Start your program (with @var{arglist}, if specified).
44486 Backtrace: display the program stack.
44488 @item print @var{expr}
44489 Display the value of an expression.
44492 Continue running your program (after stopping, e.g. at a breakpoint).
44495 Execute next program line (after stopping); step @emph{over} any
44496 function calls in the line.
44498 @item edit [@var{file}:]@var{function}
44499 look at the program line where it is presently stopped.
44501 @item list [@var{file}:]@var{function}
44502 type the text of the program in the vicinity of where it is presently stopped.
44505 Execute next program line (after stopping); step @emph{into} any
44506 function calls in the line.
44508 @item help [@var{name}]
44509 Show information about @value{GDBN} command @var{name}, or general information
44510 about using @value{GDBN}.
44513 Exit from @value{GDBN}.
44517 For full details on @value{GDBN},
44518 see @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
44519 by Richard M. Stallman and Roland H. Pesch. The same text is available online
44520 as the @code{gdb} entry in the @code{info} program.
44524 @c man begin OPTIONS gdb
44525 Any arguments other than options specify an executable
44526 file and core file (or process ID); that is, the first argument
44527 encountered with no
44528 associated option flag is equivalent to a @option{-se} option, and the second,
44529 if any, is equivalent to a @option{-c} option if it's the name of a file.
44531 both long and short forms; both are shown here. The long forms are also
44532 recognized if you truncate them, so long as enough of the option is
44533 present to be unambiguous. (If you prefer, you can flag option
44534 arguments with @option{+} rather than @option{-}, though we illustrate the
44535 more usual convention.)
44537 All the options and command line arguments you give are processed
44538 in sequential order. The order makes a difference when the @option{-x}
44544 List all options, with brief explanations.
44546 @item -symbols=@var{file}
44547 @itemx -s @var{file}
44548 Read symbol table from file @var{file}.
44551 Enable writing into executable and core files.
44553 @item -exec=@var{file}
44554 @itemx -e @var{file}
44555 Use file @var{file} as the executable file to execute when
44556 appropriate, and for examining pure data in conjunction with a core
44559 @item -se=@var{file}
44560 Read symbol table from file @var{file} and use it as the executable
44563 @item -core=@var{file}
44564 @itemx -c @var{file}
44565 Use file @var{file} as a core dump to examine.
44567 @item -command=@var{file}
44568 @itemx -x @var{file}
44569 Execute @value{GDBN} commands from file @var{file}.
44571 @item -ex @var{command}
44572 Execute given @value{GDBN} @var{command}.
44574 @item -directory=@var{directory}
44575 @itemx -d @var{directory}
44576 Add @var{directory} to the path to search for source files.
44579 Do not execute commands from @file{~/.gdbinit}.
44583 Do not execute commands from any @file{.gdbinit} initialization files.
44587 ``Quiet''. Do not print the introductory and copyright messages. These
44588 messages are also suppressed in batch mode.
44591 Run in batch mode. Exit with status @code{0} after processing all the command
44592 files specified with @option{-x} (and @file{.gdbinit}, if not inhibited).
44593 Exit with nonzero status if an error occurs in executing the @value{GDBN}
44594 commands in the command files.
44596 Batch mode may be useful for running @value{GDBN} as a filter, for example to
44597 download and run a program on another computer; in order to make this
44598 more useful, the message
44601 Program exited normally.
44605 (which is ordinarily issued whenever a program running under @value{GDBN} control
44606 terminates) is not issued when running in batch mode.
44608 @item -cd=@var{directory}
44609 Run @value{GDBN} using @var{directory} as its working directory,
44610 instead of the current directory.
44614 Emacs sets this option when it runs @value{GDBN} as a subprocess. It tells
44615 @value{GDBN} to output the full file name and line number in a standard,
44616 recognizable fashion each time a stack frame is displayed (which
44617 includes each time the program stops). This recognizable format looks
44618 like two @samp{\032} characters, followed by the file name, line number
44619 and character position separated by colons, and a newline. The
44620 Emacs-to-@value{GDBN} interface program uses the two @samp{\032}
44621 characters as a signal to display the source code for the frame.
44624 Set the line speed (baud rate or bits per second) of any serial
44625 interface used by @value{GDBN} for remote debugging.
44627 @item -tty=@var{device}
44628 Run using @var{device} for your program's standard input and output.
44632 @c man begin SEEALSO gdb
44634 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
44635 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
44636 documentation are properly installed at your site, the command
44643 should give you access to the complete manual.
44645 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
44646 Richard M. Stallman and Roland H. Pesch, July 1991.
44650 @node gdbserver man
44651 @heading gdbserver man
44653 @c man title gdbserver Remote Server for the GNU Debugger
44655 @c man begin SYNOPSIS gdbserver
44656 gdbserver @var{comm} @var{prog} [@var{args}@dots{}]
44658 gdbserver --attach @var{comm} @var{pid}
44660 gdbserver --multi @var{comm}
44664 @c man begin DESCRIPTION gdbserver
44665 @command{gdbserver} is a program that allows you to run @value{GDBN} on a different machine
44666 than the one which is running the program being debugged.
44669 @subheading Usage (server (target) side)
44672 Usage (server (target) side):
44675 First, you need to have a copy of the program you want to debug put onto
44676 the target system. The program can be stripped to save space if needed, as
44677 @command{gdbserver} doesn't care about symbols. All symbol handling is taken care of by
44678 the @value{GDBN} running on the host system.
44680 To use the server, you log on to the target system, and run the @command{gdbserver}
44681 program. You must tell it (a) how to communicate with @value{GDBN}, (b) the name of
44682 your program, and (c) its arguments. The general syntax is:
44685 target> gdbserver @var{comm} @var{program} [@var{args} ...]
44688 For example, using a serial port, you might say:
44692 @c @file would wrap it as F</dev/com1>.
44693 target> gdbserver /dev/com1 emacs foo.txt
44696 target> gdbserver @file{/dev/com1} emacs foo.txt
44700 This tells @command{gdbserver} to debug emacs with an argument of foo.txt, and
44701 to communicate with @value{GDBN} via @file{/dev/com1}. @command{gdbserver} now
44702 waits patiently for the host @value{GDBN} to communicate with it.
44704 To use a TCP connection, you could say:
44707 target> gdbserver host:2345 emacs foo.txt
44710 This says pretty much the same thing as the last example, except that we are
44711 going to communicate with the @code{host} @value{GDBN} via TCP. The @code{host:2345} argument means
44712 that we are expecting to see a TCP connection from @code{host} to local TCP port
44713 2345. (Currently, the @code{host} part is ignored.) You can choose any number you
44714 want for the port number as long as it does not conflict with any existing TCP
44715 ports on the target system. This same port number must be used in the host
44716 @value{GDBN}s @code{target remote} command, which will be described shortly. Note that if
44717 you chose a port number that conflicts with another service, @command{gdbserver} will
44718 print an error message and exit.
44720 @command{gdbserver} can also attach to running programs.
44721 This is accomplished via the @option{--attach} argument. The syntax is:
44724 target> gdbserver --attach @var{comm} @var{pid}
44727 @var{pid} is the process ID of a currently running process. It isn't
44728 necessary to point @command{gdbserver} at a binary for the running process.
44730 To start @code{gdbserver} without supplying an initial command to run
44731 or process ID to attach, use the @option{--multi} command line option.
44732 In such case you should connect using @kbd{target extended-remote} to start
44733 the program you want to debug.
44736 target> gdbserver --multi @var{comm}
44740 @subheading Usage (host side)
44746 You need an unstripped copy of the target program on your host system, since
44747 @value{GDBN} needs to examine its symbol tables and such. Start up @value{GDBN} as you normally
44748 would, with the target program as the first argument. (You may need to use the
44749 @option{--baud} option if the serial line is running at anything except 9600 baud.)
44750 That is @code{gdb TARGET-PROG}, or @code{gdb --baud BAUD TARGET-PROG}. After that, the only
44751 new command you need to know about is @code{target remote}
44752 (or @code{target extended-remote}). Its argument is either
44753 a device name (usually a serial device, like @file{/dev/ttyb}), or a @code{HOST:PORT}
44754 descriptor. For example:
44758 @c @file would wrap it as F</dev/ttyb>.
44759 (gdb) target remote /dev/ttyb
44762 (gdb) target remote @file{/dev/ttyb}
44767 communicates with the server via serial line @file{/dev/ttyb}, and:
44770 (gdb) target remote the-target:2345
44774 communicates via a TCP connection to port 2345 on host `the-target', where
44775 you previously started up @command{gdbserver} with the same port number. Note that for
44776 TCP connections, you must start up @command{gdbserver} prior to using the `target remote'
44777 command, otherwise you may get an error that looks something like
44778 `Connection refused'.
44780 @command{gdbserver} can also debug multiple inferiors at once,
44783 the @value{GDBN} manual in node @code{Inferiors and Programs}
44784 -- shell command @code{info -f gdb -n 'Inferiors and Programs'}.
44787 @ref{Inferiors and Programs}.
44789 In such case use the @code{extended-remote} @value{GDBN} command variant:
44792 (gdb) target extended-remote the-target:2345
44795 The @command{gdbserver} option @option{--multi} may or may not be used in such
44799 @c man begin OPTIONS gdbserver
44800 There are three different modes for invoking @command{gdbserver}:
44805 Debug a specific program specified by its program name:
44808 gdbserver @var{comm} @var{prog} [@var{args}@dots{}]
44811 The @var{comm} parameter specifies how should the server communicate
44812 with @value{GDBN}; it is either a device name (to use a serial line),
44813 a TCP port number (@code{:1234}), or @code{-} or @code{stdio} to use
44814 stdin/stdout of @code{gdbserver}. Specify the name of the program to
44815 debug in @var{prog}. Any remaining arguments will be passed to the
44816 program verbatim. When the program exits, @value{GDBN} will close the
44817 connection, and @code{gdbserver} will exit.
44820 Debug a specific program by specifying the process ID of a running
44824 gdbserver --attach @var{comm} @var{pid}
44827 The @var{comm} parameter is as described above. Supply the process ID
44828 of a running program in @var{pid}; @value{GDBN} will do everything
44829 else. Like with the previous mode, when the process @var{pid} exits,
44830 @value{GDBN} will close the connection, and @code{gdbserver} will exit.
44833 Multi-process mode -- debug more than one program/process:
44836 gdbserver --multi @var{comm}
44839 In this mode, @value{GDBN} can instruct @command{gdbserver} which
44840 command(s) to run. Unlike the other 2 modes, @value{GDBN} will not
44841 close the connection when a process being debugged exits, so you can
44842 debug several processes in the same session.
44845 In each of the modes you may specify these options:
44850 List all options, with brief explanations.
44853 This option causes @command{gdbserver} to print its version number and exit.
44856 @command{gdbserver} will attach to a running program. The syntax is:
44859 target> gdbserver --attach @var{comm} @var{pid}
44862 @var{pid} is the process ID of a currently running process. It isn't
44863 necessary to point @command{gdbserver} at a binary for the running process.
44866 To start @code{gdbserver} without supplying an initial command to run
44867 or process ID to attach, use this command line option.
44868 Then you can connect using @kbd{target extended-remote} and start
44869 the program you want to debug. The syntax is:
44872 target> gdbserver --multi @var{comm}
44876 Instruct @code{gdbserver} to display extra status information about the debugging
44878 This option is intended for @code{gdbserver} development and for bug reports to
44881 @item --remote-debug
44882 Instruct @code{gdbserver} to display remote protocol debug output.
44883 This option is intended for @code{gdbserver} development and for bug reports to
44886 @item --debug-file=@var{filename}
44887 Instruct @code{gdbserver} to send any debug output to the given @var{filename}.
44888 This option is intended for @code{gdbserver} development and for bug reports to
44891 @item --debug-format=option1@r{[},option2,...@r{]}
44892 Instruct @code{gdbserver} to include extra information in each line
44893 of debugging output.
44894 @xref{Other Command-Line Arguments for gdbserver}.
44897 Specify a wrapper to launch programs
44898 for debugging. The option should be followed by the name of the
44899 wrapper, then any command-line arguments to pass to the wrapper, then
44900 @kbd{--} indicating the end of the wrapper arguments.
44903 By default, @command{gdbserver} keeps the listening TCP port open, so that
44904 additional connections are possible. However, if you start @code{gdbserver}
44905 with the @option{--once} option, it will stop listening for any further
44906 connection attempts after connecting to the first @value{GDBN} session.
44908 @c --disable-packet is not documented for users.
44910 @c --disable-randomization and --no-disable-randomization are superseded by
44911 @c QDisableRandomization.
44916 @c man begin SEEALSO gdbserver
44918 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
44919 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
44920 documentation are properly installed at your site, the command
44926 should give you access to the complete manual.
44928 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
44929 Richard M. Stallman and Roland H. Pesch, July 1991.
44936 @c man title gcore Generate a core file of a running program
44939 @c man begin SYNOPSIS gcore
44940 gcore [-a] [-o @var{prefix}] @var{pid1} [@var{pid2}...@var{pidN}]
44944 @c man begin DESCRIPTION gcore
44945 Generate core dumps of one or more running programs with process IDs
44946 @var{pid1}, @var{pid2}, etc. A core file produced by @command{gcore}
44947 is equivalent to one produced by the kernel when the process crashes
44948 (and when @kbd{ulimit -c} was used to set up an appropriate core dump
44949 limit). However, unlike after a crash, after @command{gcore} finishes
44950 its job the program remains running without any change.
44953 @c man begin OPTIONS gcore
44956 Dump all memory mappings. The actual effect of this option depends on
44957 the Operating System. On @sc{gnu}/Linux, it will disable
44958 @code{use-coredump-filter} (@pxref{set use-coredump-filter}) and
44959 enable @code{dump-excluded-mappings} (@pxref{set
44960 dump-excluded-mappings}).
44962 @item -o @var{prefix}
44963 The optional argument @var{prefix} specifies the prefix to be used
44964 when composing the file names of the core dumps. The file name is
44965 composed as @file{@var{prefix}.@var{pid}}, where @var{pid} is the
44966 process ID of the running program being analyzed by @command{gcore}.
44967 If not specified, @var{prefix} defaults to @var{gcore}.
44971 @c man begin SEEALSO gcore
44973 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
44974 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
44975 documentation are properly installed at your site, the command
44982 should give you access to the complete manual.
44984 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
44985 Richard M. Stallman and Roland H. Pesch, July 1991.
44992 @c man title gdbinit GDB initialization scripts
44995 @c man begin SYNOPSIS gdbinit
44996 @ifset SYSTEM_GDBINIT
44997 @value{SYSTEM_GDBINIT}
45006 @c man begin DESCRIPTION gdbinit
45007 These files contain @value{GDBN} commands to automatically execute during
45008 @value{GDBN} startup. The lines of contents are canned sequences of commands,
45011 the @value{GDBN} manual in node @code{Sequences}
45012 -- shell command @code{info -f gdb -n Sequences}.
45018 Please read more in
45020 the @value{GDBN} manual in node @code{Startup}
45021 -- shell command @code{info -f gdb -n Startup}.
45028 @ifset SYSTEM_GDBINIT
45029 @item @value{SYSTEM_GDBINIT}
45031 @ifclear SYSTEM_GDBINIT
45032 @item (not enabled with @code{--with-system-gdbinit} during compilation)
45034 System-wide initialization file. It is executed unless user specified
45035 @value{GDBN} option @code{-nx} or @code{-n}.
45038 the @value{GDBN} manual in node @code{System-wide configuration}
45039 -- shell command @code{info -f gdb -n 'System-wide configuration'}.
45042 @ref{System-wide configuration}.
45046 User initialization file. It is executed unless user specified
45047 @value{GDBN} options @code{-nx}, @code{-n} or @code{-nh}.
45050 Initialization file for current directory. It may need to be enabled with
45051 @value{GDBN} security command @code{set auto-load local-gdbinit}.
45054 the @value{GDBN} manual in node @code{Init File in the Current Directory}
45055 -- shell command @code{info -f gdb -n 'Init File in the Current Directory'}.
45058 @ref{Init File in the Current Directory}.
45063 @c man begin SEEALSO gdbinit
45065 gdb(1), @code{info -f gdb -n Startup}
45067 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
45068 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
45069 documentation are properly installed at your site, the command
45075 should give you access to the complete manual.
45077 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
45078 Richard M. Stallman and Roland H. Pesch, July 1991.
45082 @node gdb-add-index man
45083 @heading gdb-add-index
45084 @pindex gdb-add-index
45085 @anchor{gdb-add-index}
45087 @c man title gdb-add-index Add index files to speed up GDB
45089 @c man begin SYNOPSIS gdb-add-index
45090 gdb-add-index @var{filename}
45093 @c man begin DESCRIPTION gdb-add-index
45094 When @value{GDBN} finds a symbol file, it scans the symbols in the
45095 file in order to construct an internal symbol table. This lets most
45096 @value{GDBN} operations work quickly--at the cost of a delay early on.
45097 For large programs, this delay can be quite lengthy, so @value{GDBN}
45098 provides a way to build an index, which speeds up startup.
45100 To determine whether a file contains such an index, use the command
45101 @kbd{readelf -S filename}: the index is stored in a section named
45102 @code{.gdb_index}. The index file can only be produced on systems
45103 which use ELF binaries and DWARF debug information (i.e., sections
45104 named @code{.debug_*}).
45106 @command{gdb-add-index} uses @value{GDBN} and @command{objdump} found
45107 in the @env{PATH} environment variable. If you want to use different
45108 versions of these programs, you can specify them through the
45109 @env{GDB} and @env{OBJDUMP} environment variables.
45113 the @value{GDBN} manual in node @code{Index Files}
45114 -- shell command @kbd{info -f gdb -n "Index Files"}.
45121 @c man begin SEEALSO gdb-add-index
45123 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
45124 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
45125 documentation are properly installed at your site, the command
45131 should give you access to the complete manual.
45133 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
45134 Richard M. Stallman and Roland H. Pesch, July 1991.
45140 @node GNU Free Documentation License
45141 @appendix GNU Free Documentation License
45144 @node Concept Index
45145 @unnumbered Concept Index
45149 @node Command and Variable Index
45150 @unnumbered Command, Variable, and Function Index
45155 % I think something like @@colophon should be in texinfo. In the
45157 \long\def\colophon{\hbox to0pt{}\vfill
45158 \centerline{The body of this manual is set in}
45159 \centerline{\fontname\tenrm,}
45160 \centerline{with headings in {\bf\fontname\tenbf}}
45161 \centerline{and examples in {\tt\fontname\tentt}.}
45162 \centerline{{\it\fontname\tenit\/},}
45163 \centerline{{\bf\fontname\tenbf}, and}
45164 \centerline{{\sl\fontname\tensl\/}}
45165 \centerline{are used for emphasis.}\vfill}
45167 % Blame: doc@@cygnus.com, 1991.